vendor/llvm/llvm-trunk-r238337

1414 files changed, 28504 insertions, 12742 deletions

diff --git a/test/Transforms/ADCE/2002-05-23-ZeroArgPHITest.ll b/test/Transforms/ADCE/2002-05-23-ZeroArgPHITest.ll
index 9407b5a68d69..8d1beec8ed35 100644
--- a/test/Transforms/ADCE/2002-05-23-ZeroArgPHITest.ll
+++ b/test/Transforms/ADCE/2002-05-23-ZeroArgPHITest.ll
@@ -15,14 +15,14 @@ bb0:
         br label %bb1
 
 bb1:            ; preds = %bb0
-        %reg107 = load %node_t** %nodelist.upgrd.1              ; <%node_t*> [#uses=2]
+        %reg107 = load %node_t*, %node_t** %nodelist.upgrd.1              ; <%node_t*> [#uses=2]
         %cond211 = icmp eq %node_t* %reg107, null               ; <i1> [#uses=1]
         br i1 %cond211, label %bb3, label %bb2
 
 bb2:            ; preds = %bb2, %bb1
         %reg109 = phi %node_t* [ %reg110, %bb2 ], [ %reg107, %bb1 ]             ; <%node_t*> [#uses=1]
-        %reg212 = getelementptr %node_t* %reg109, i64 0, i32 1          ; <%node_t**> [#uses=1]
-        %reg110 = load %node_t** %reg212                ; <%node_t*> [#uses=2]
+        %reg212 = getelementptr %node_t, %node_t* %reg109, i64 0, i32 1          ; <%node_t**> [#uses=1]
+        %reg110 = load %node_t*, %node_t** %reg212                ; <%node_t*> [#uses=2]
         %cond213 = icmp ne %node_t* %reg110, null               ; <i1> [#uses=1]
         br i1 %cond213, label %bb2, label %bb3
 
diff --git a/test/Transforms/ADCE/2002-05-28-Crash.ll b/test/Transforms/ADCE/2002-05-28-Crash.ll
index 9bbbd055ff96..d88580afad0a 100644
--- a/test/Transforms/ADCE/2002-05-28-Crash.ll
+++ b/test/Transforms/ADCE/2002-05-28-Crash.ll
@@ -15,7 +15,7 @@
 
 define i32 @rx_bitset_empty(i32 %size, i32* %set) {
 bb1:
-        %reg110 = load i32* %set                ; <i32> [#uses=2]
+        %reg110 = load i32, i32* %set                ; <i32> [#uses=2]
         store i32 1, i32* %set
         %cast112 = sext i32 %size to i64                ; <i64> [#uses=1]
         %reg113 = add i64 %cast112, 31          ; <i64> [#uses=1]
@@ -25,8 +25,8 @@ bb1:
         %reg114-idxcast = trunc i64 %reg114 to i32              ; <i32> [#uses=1]
         %reg114-idxcast-offset = add i32 %reg114-idxcast, 1073741823            ; <i32> [#uses=1]
         %reg114-idxcast-offset.upgrd.1 = zext i32 %reg114-idxcast-offset to i64         ; <i64> [#uses=1]
-        %reg124 = getelementptr i32* %set, i64 %reg114-idxcast-offset.upgrd.1           ; <i32*> [#uses=1]
-        %reg125 = load i32* %reg124             ; <i32> [#uses=1]
+        %reg124 = getelementptr i32, i32* %set, i64 %reg114-idxcast-offset.upgrd.1           ; <i32*> [#uses=1]
+        %reg125 = load i32, i32* %reg124             ; <i32> [#uses=1]
         %cond232 = icmp ne i32 %reg125, 0               ; <i1> [#uses=1]
         br i1 %cond232, label %bb3, label %bb2
 
@@ -38,8 +38,8 @@ bb2:            ; preds = %bb2, %bb1
         %reg130-idxcast = bitcast i32 %reg130 to i32            ; <i32> [#uses=1]
         %reg130-idxcast-offset = add i32 %reg130-idxcast, 1073741823            ; <i32> [#uses=1]
         %reg130-idxcast-offset.upgrd.2 = zext i32 %reg130-idxcast-offset to i64         ; <i64> [#uses=1]
-        %reg118 = getelementptr i32* %set, i64 %reg130-idxcast-offset.upgrd.2           ; <i32*> [#uses=1]
-        %reg119 = load i32* %reg118             ; <i32> [#uses=1]
+        %reg118 = getelementptr i32, i32* %set, i64 %reg130-idxcast-offset.upgrd.2           ; <i32*> [#uses=1]
+        %reg119 = load i32, i32* %reg118             ; <i32> [#uses=1]
         %cond233 = icmp eq i32 %reg119, 0               ; <i1> [#uses=1]
         br i1 %cond233, label %bb2, label %bb3
 
diff --git a/test/Transforms/ADCE/2002-07-17-AssertionFailure.ll b/test/Transforms/ADCE/2002-07-17-AssertionFailure.ll
index 8f8dadf7332f..ff8bdb3fd3b3 100644
--- a/test/Transforms/ADCE/2002-07-17-AssertionFailure.ll
+++ b/test/Transforms/ADCE/2002-07-17-AssertionFailure.ll
@@ -7,7 +7,7 @@
 
 define void @foo(i8* %reg5481) {
         %cast611 = bitcast i8* %reg5481 to i8**         ; <i8**> [#uses=1]
-        %reg162 = load i8** %cast611            ; <i8*> [#uses=1]
+        %reg162 = load i8*, i8** %cast611            ; <i8*> [#uses=1]
         ptrtoint i8* %reg162 to i32             ; <i32>:1 [#uses=0]
         ret void
 }
diff --git a/test/Transforms/ADCE/2002-07-17-PHIAssertion.ll b/test/Transforms/ADCE/2002-07-17-PHIAssertion.ll
index 2f0df670d6ac..1bf79e8ec6c7 100644
--- a/test/Transforms/ADCE/2002-07-17-PHIAssertion.ll
+++ b/test/Transforms/ADCE/2002-07-17-PHIAssertion.ll
@@ -17,7 +17,7 @@ bb3:            ; preds = %bb2
         br label %UnifiedExitNode
 
 bb4:            ; preds = %bb2
-        %reg117 = load i32* @hufts              ; <i32> [#uses=2]
+        %reg117 = load i32, i32* @hufts              ; <i32> [#uses=2]
         %cond241 = icmp ule i32 %reg117, %reg128                ; <i1> [#uses=1]
         br i1 %cond241, label %bb6, label %bb5
 
@@ -29,12 +29,12 @@ bb6:            ; preds = %bb5, %bb4
         br i1 false, label %bb2, label %bb7
 
 bb7:            ; preds = %bb6
-        %reg126 = load i32* @bk         ; <i32> [#uses=1]
+        %reg126 = load i32, i32* @bk         ; <i32> [#uses=1]
         %cond247 = icmp ule i32 %reg126, 7              ; <i1> [#uses=1]
         br i1 %cond247, label %bb9, label %bb8
 
 bb8:            ; preds = %bb8, %bb7
-        %reg119 = load i32* @bk         ; <i32> [#uses=1]
+        %reg119 = load i32, i32* @bk         ; <i32> [#uses=1]
         %cond256 = icmp ugt i32 %reg119, 7              ; <i1> [#uses=1]
         br i1 %cond256, label %bb8, label %bb9
 
diff --git a/test/Transforms/ADCE/2003-06-11-InvalidCFG.ll b/test/Transforms/ADCE/2003-06-11-InvalidCFG.ll
index 5206b243e372..7c7e238f9d99 100644
--- a/test/Transforms/ADCE/2003-06-11-InvalidCFG.ll
+++ b/test/Transforms/ADCE/2003-06-11-InvalidCFG.ll
@@ -16,7 +16,7 @@ then.66:                ; preds = %shortcirc_done.12
         br label %endif.42
 
 endif.65:               ; preds = %endif.42
-        %tmp.2846 = load i32** @G               ; <i32*> [#uses=1]
+        %tmp.2846 = load i32*, i32** @G               ; <i32*> [#uses=1]
         br i1 false, label %shortcirc_next.12, label %shortcirc_done.12
 
 shortcirc_next.12:              ; preds = %endif.65
diff --git a/test/Transforms/ADCE/2003-06-24-BadSuccessor.ll b/test/Transforms/ADCE/2003-06-24-BadSuccessor.ll
index eb3ef1e7913f..707e14aad0dd 100644
--- a/test/Transforms/ADCE/2003-06-24-BadSuccessor.ll
+++ b/test/Transforms/ADCE/2003-06-24-BadSuccessor.ll
@@ -48,8 +48,8 @@ shortcirc_next.4:		; preds = %then.44
 	br i1 false, label %no_exit.2, label %loopexit.2
 
 no_exit.2:		; preds = %shortcirc_next.4
-	%tmp.897 = getelementptr i32* %SubArrays.10, i64 0		; <i32*> [#uses=1]
-	%tmp.899 = load i32* %tmp.897		; <i32> [#uses=1]
+	%tmp.897 = getelementptr i32, i32* %SubArrays.10, i64 0		; <i32*> [#uses=1]
+	%tmp.899 = load i32, i32* %tmp.897		; <i32> [#uses=1]
 	store i32 %tmp.899, i32* null
 	ret i32 0
 
@@ -79,7 +79,7 @@ shortcirc_next.8:		; preds = %shortcirc_next.7
 
 then.53:		; preds = %shortcirc_next.7, %label.17
 	%SubArrays.8 = phi i32* [ %SubArrays.10, %shortcirc_next.7 ], [ %SubArrays.10, %label.17 ]		; <i32*> [#uses=1]
-	%tmp.1023 = load i32* null		; <i32> [#uses=1]
+	%tmp.1023 = load i32, i32* null		; <i32> [#uses=1]
 	switch i32 %tmp.1023, label %loopentry.1 [
 	]
 
diff --git a/test/Transforms/ADCE/2003-06-24-BasicFunctionality.ll b/test/Transforms/ADCE/2003-06-24-BasicFunctionality.ll
index 82fa5b2a40e1..f0de43164365 100644
--- a/test/Transforms/ADCE/2003-06-24-BasicFunctionality.ll
+++ b/test/Transforms/ADCE/2003-06-24-BasicFunctionality.ll
@@ -2,20 +2,20 @@
 
 define void @dead_test8(i32* %data.1, i32 %idx.1) {
 entry:
-        %tmp.1 = load i32* %data.1              ; <i32> [#uses=2]
+        %tmp.1 = load i32, i32* %data.1              ; <i32> [#uses=2]
         %tmp.41 = icmp sgt i32 %tmp.1, 0                ; <i1> [#uses=1]
         br i1 %tmp.41, label %no_exit.preheader, label %return
 
 no_exit.preheader:              ; preds = %entry
-        %tmp.11 = getelementptr i32* %data.1, i64 1             ; <i32*> [#uses=1]
+        %tmp.11 = getelementptr i32, i32* %data.1, i64 1             ; <i32*> [#uses=1]
         %tmp.22-idxcast = sext i32 %idx.1 to i64                ; <i64> [#uses=1]
-        %tmp.28 = getelementptr i32* %data.1, i64 %tmp.22-idxcast               ; <i32*> [#uses=1]
+        %tmp.28 = getelementptr i32, i32* %data.1, i64 %tmp.22-idxcast               ; <i32*> [#uses=1]
         br label %no_exit
 
 no_exit:                ; preds = %endif, %no_exit.preheader
         %k.1 = phi i32 [ %k.0, %endif ], [ 0, %no_exit.preheader ]              ; <i32> [#uses=3]
         %i.0 = phi i32 [ %inc.1, %endif ], [ 0, %no_exit.preheader ]            ; <i32> [#uses=1]
-        %tmp.12 = load i32* %tmp.11             ; <i32> [#uses=1]
+        %tmp.12 = load i32, i32* %tmp.11             ; <i32> [#uses=1]
         %tmp.14 = sub i32 0, %tmp.12            ; <i32> [#uses=1]
         %tmp.161 = icmp ne i32 %k.1, %tmp.14            ; <i1> [#uses=1]
         br i1 %tmp.161, label %then, label %else
diff --git a/test/Transforms/ADCE/basictest1.ll b/test/Transforms/ADCE/basictest1.ll
index bbb88783deb2..4d0d386384b9 100644
--- a/test/Transforms/ADCE/basictest1.ll
+++ b/test/Transforms/ADCE/basictest1.ll
@@ -22,12 +22,12 @@ declare i32 @fwrite(i8*, i32, i32, %FILE*)
 declare void @perror(i8*)
 
 define i32 @spec_getc(i32 %fd) {
-	%reg109 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg109 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond266 = icmp sle i32 %reg109, 4		; <i1> [#uses=1]
 	br i1 %cond266, label %bb3, label %bb2
 
 bb2:		; preds = %0
-	%cast273 = getelementptr [17 x i8]* @.LC12, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast273 = getelementptr [17 x i8], [17 x i8]* @.LC12, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb3
 
 bb3:		; preds = %bb2, %0
@@ -35,8 +35,8 @@ bb3:		; preds = %bb2, %0
 	br i1 %cond267, label %bb5, label %bb4
 
 bb4:		; preds = %bb3
-	%reg111 = getelementptr [20 x %FILE]* @__iob, i64 0, i64 1, i32 3		; <i8*> [#uses=1]
-	%cast274 = getelementptr [34 x i8]* @.LC9, i64 0, i64 0		; <i8*> [#uses=0]
+	%reg111 = getelementptr [20 x %FILE], [20 x %FILE]* @__iob, i64 0, i64 1, i32 3		; <i8*> [#uses=1]
+	%cast274 = getelementptr [34 x i8], [34 x i8]* @.LC9, i64 0, i64 0		; <i8*> [#uses=0]
 	%cast282 = bitcast i8* %reg111 to %FILE*		; <%FILE*> [#uses=0]
 	call void @exit( i32 1 )
 	br label %UnifiedExitNode
@@ -44,21 +44,21 @@ bb4:		; preds = %bb3
 bb5:		; preds = %bb3
 	%reg107-idxcast1 = sext i32 %fd to i64		; <i64> [#uses=2]
 	%reg107-idxcast2 = sext i32 %fd to i64		; <i64> [#uses=1]
-	%reg1311 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast2		; <%spec_fd_t*> [#uses=1]
-	%idx1 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
-	%reg1321 = load i32* %idx1		; <i32> [#uses=3]
-	%idx2 = getelementptr %spec_fd_t* %reg1311, i64 0, i32 1		; <i32*> [#uses=1]
-	%reg1331 = load i32* %idx2		; <i32> [#uses=1]
+	%reg1311 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast2		; <%spec_fd_t*> [#uses=1]
+	%idx1 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
+	%reg1321 = load i32, i32* %idx1		; <i32> [#uses=3]
+	%idx2 = getelementptr %spec_fd_t, %spec_fd_t* %reg1311, i64 0, i32 1		; <i32*> [#uses=1]
+	%reg1331 = load i32, i32* %idx2		; <i32> [#uses=1]
 	%cond270 = icmp slt i32 %reg1321, %reg1331		; <i1> [#uses=1]
 	br i1 %cond270, label %bb9, label %bb6
 
 bb6:		; preds = %bb5
-	%reg134 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg134 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond271 = icmp sle i32 %reg134, 4		; <i1> [#uses=1]
 	br i1 %cond271, label %bb8, label %bb7
 
 bb7:		; preds = %bb6
-	%cast277 = getelementptr [4 x i8]* @.LC10, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast277 = getelementptr [4 x i8], [4 x i8]* @.LC10, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb8
 
 bb8:		; preds = %bb7, %bb6
@@ -66,21 +66,21 @@ bb8:		; preds = %bb7, %bb6
 
 bb9:		; preds = %bb5
 	%reg107-idxcast3 = sext i32 %fd to i64		; <i64> [#uses=1]
-	%idx3 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast3, i32 3		; <i8**> [#uses=1]
-	%reg1601 = load i8** %idx3		; <i8*> [#uses=1]
+	%idx3 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast3, i32 3		; <i8**> [#uses=1]
+	%reg1601 = load i8*, i8** %idx3		; <i8*> [#uses=1]
 	%reg132-idxcast1 = sext i32 %reg1321 to i64		; <i64> [#uses=1]
-	%idx4 = getelementptr i8* %reg1601, i64 %reg132-idxcast1		; <i8*> [#uses=1]
-	%reg1621 = load i8* %idx4		; <i8> [#uses=2]
+	%idx4 = getelementptr i8, i8* %reg1601, i64 %reg132-idxcast1		; <i8*> [#uses=1]
+	%reg1621 = load i8, i8* %idx4		; <i8> [#uses=2]
 	%cast108 = zext i8 %reg1621 to i64		; <i64> [#uses=0]
 	%reg157 = add i32 %reg1321, 1		; <i32> [#uses=1]
-	%idx5 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
+	%idx5 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
 	store i32 %reg157, i32* %idx5
-	%reg163 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg163 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond272 = icmp sle i32 %reg163, 4		; <i1> [#uses=1]
 	br i1 %cond272, label %bb11, label %bb10
 
 bb10:		; preds = %bb9
-	%cast279 = getelementptr [4 x i8]* @.LC11, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast279 = getelementptr [4 x i8], [4 x i8]* @.LC11, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb11
 
 bb11:		; preds = %bb10, %bb9
diff --git a/test/Transforms/ADCE/basictest2.ll b/test/Transforms/ADCE/basictest2.ll
index a17795f14246..26b2e85cccba 100644
--- a/test/Transforms/ADCE/basictest2.ll
+++ b/test/Transforms/ADCE/basictest2.ll
@@ -22,12 +22,12 @@ declare i32 @fwrite(i8*, i32, i32, %FILE*)
 declare void @perror(i8*)
 
 define i32 @spec_getc(i32 %fd) {
-	%reg109 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg109 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond266 = icmp sle i32 %reg109, 4		; <i1> [#uses=1]
 	br i1 %cond266, label %bb3, label %bb2
 
 bb2:		; preds = %0
-	%cast273 = getelementptr [17 x i8]* @.LC12, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast273 = getelementptr [17 x i8], [17 x i8]* @.LC12, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb3
 
 bb3:		; preds = %bb2, %0
@@ -35,8 +35,8 @@ bb3:		; preds = %bb2, %0
 	br label %bb5
 
 bb4:		; No predecessors!
-	%reg111 = getelementptr [20 x %FILE]* @__iob, i64 0, i64 1, i32 3		; <i8*> [#uses=1]
-	%cast274 = getelementptr [34 x i8]* @.LC9, i64 0, i64 0		; <i8*> [#uses=0]
+	%reg111 = getelementptr [20 x %FILE], [20 x %FILE]* @__iob, i64 0, i64 1, i32 3		; <i8*> [#uses=1]
+	%cast274 = getelementptr [34 x i8], [34 x i8]* @.LC9, i64 0, i64 0		; <i8*> [#uses=0]
 	%cast282 = bitcast i8* %reg111 to %FILE*		; <%FILE*> [#uses=0]
 	call void @exit( i32 1 )
 	br label %UnifiedExitNode
@@ -44,21 +44,21 @@ bb4:		; No predecessors!
 bb5:		; preds = %bb3
 	%reg107-idxcast1 = sext i32 %fd to i64		; <i64> [#uses=2]
 	%reg107-idxcast2 = sext i32 %fd to i64		; <i64> [#uses=1]
-	%reg1311 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast2		; <%spec_fd_t*> [#uses=1]
-	%idx1 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
-	%reg1321 = load i32* %idx1		; <i32> [#uses=3]
-	%idx2 = getelementptr %spec_fd_t* %reg1311, i64 0, i32 1		; <i32*> [#uses=1]
-	%reg1331 = load i32* %idx2		; <i32> [#uses=1]
+	%reg1311 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast2		; <%spec_fd_t*> [#uses=1]
+	%idx1 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
+	%reg1321 = load i32, i32* %idx1		; <i32> [#uses=3]
+	%idx2 = getelementptr %spec_fd_t, %spec_fd_t* %reg1311, i64 0, i32 1		; <i32*> [#uses=1]
+	%reg1331 = load i32, i32* %idx2		; <i32> [#uses=1]
 	%cond270 = icmp slt i32 %reg1321, %reg1331		; <i1> [#uses=1]
 	br i1 %cond270, label %bb9, label %bb6
 
 bb6:		; preds = %bb5
-	%reg134 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg134 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond271 = icmp sle i32 %reg134, 4		; <i1> [#uses=1]
 	br i1 %cond271, label %bb8, label %bb7
 
 bb7:		; preds = %bb6
-	%cast277 = getelementptr [4 x i8]* @.LC10, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast277 = getelementptr [4 x i8], [4 x i8]* @.LC10, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb8
 
 bb8:		; preds = %bb7, %bb6
@@ -66,21 +66,21 @@ bb8:		; preds = %bb7, %bb6
 
 bb9:		; preds = %bb5
 	%reg107-idxcast3 = sext i32 %fd to i64		; <i64> [#uses=1]
-	%idx3 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast3, i32 3		; <i8**> [#uses=1]
-	%reg1601 = load i8** %idx3		; <i8*> [#uses=1]
+	%idx3 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast3, i32 3		; <i8**> [#uses=1]
+	%reg1601 = load i8*, i8** %idx3		; <i8*> [#uses=1]
 	%reg132-idxcast1 = sext i32 %reg1321 to i64		; <i64> [#uses=1]
-	%idx4 = getelementptr i8* %reg1601, i64 %reg132-idxcast1		; <i8*> [#uses=1]
-	%reg1621 = load i8* %idx4		; <i8> [#uses=2]
+	%idx4 = getelementptr i8, i8* %reg1601, i64 %reg132-idxcast1		; <i8*> [#uses=1]
+	%reg1621 = load i8, i8* %idx4		; <i8> [#uses=2]
 	%cast108 = zext i8 %reg1621 to i64		; <i64> [#uses=0]
 	%reg157 = add i32 %reg1321, 1		; <i32> [#uses=1]
-	%idx5 = getelementptr [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
+	%idx5 = getelementptr [3 x %spec_fd_t], [3 x %spec_fd_t]* @spec_fd, i64 0, i64 %reg107-idxcast1, i32 2		; <i32*> [#uses=1]
 	store i32 %reg157, i32* %idx5
-	%reg163 = load i32* @dbglvl		; <i32> [#uses=1]
+	%reg163 = load i32, i32* @dbglvl		; <i32> [#uses=1]
 	%cond272 = icmp sle i32 %reg163, 4		; <i1> [#uses=1]
 	br i1 %cond272, label %bb11, label %bb10
 
 bb10:		; preds = %bb9
-	%cast279 = getelementptr [4 x i8]* @.LC11, i64 0, i64 0		; <i8*> [#uses=0]
+	%cast279 = getelementptr [4 x i8], [4 x i8]* @.LC11, i64 0, i64 0		; <i8*> [#uses=0]
 	br label %bb11
 
 bb11:		; preds = %bb10, %bb9
diff --git a/test/Transforms/AddDiscriminators/basic.ll b/test/Transforms/AddDiscriminators/basic.ll
index 7c8b3d3a7ce8..cabf707fe5d6 100644
--- a/test/Transforms/AddDiscriminators/basic.ll
+++ b/test/Transforms/AddDiscriminators/basic.ll
@@ -16,22 +16,23 @@ entry:
   %i.addr = alloca i32, align 4
   %x = alloca i32, align 4
   store i32 %i, i32* %i.addr, align 4
-  %0 = load i32* %i.addr, align 4, !dbg !10
+  %0 = load i32, i32* %i.addr, align 4, !dbg !10
   %cmp = icmp slt i32 %0, 10, !dbg !10
   br i1 %cmp, label %if.then, label %if.end, !dbg !10
 
 if.then:                                          ; preds = %entry
-  %1 = load i32* %i.addr, align 4, !dbg !10
-; CHECK:  %1 = load i32* %i.addr, align 4, !dbg !12
+  %1 = load i32, i32* %i.addr, align 4, !dbg !10
+; CHECK:  %1 = load i32, i32* %i.addr, align 4, !dbg ![[THEN:[0-9]+]]
 
   store i32 %1, i32* %x, align 4, !dbg !10
-; CHECK:  store i32 %1, i32* %x, align 4, !dbg !12
+; CHECK:  store i32 %1, i32* %x, align 4, !dbg ![[THEN]]
 
   br label %if.end, !dbg !10
-; CHECK:   br label %if.end, !dbg !12
+; CHECK:   br label %if.end, !dbg ![[THEN]]
 
 if.end:                                           ; preds = %if.then, %entry
   ret void, !dbg !12
+; CHECK:   ret void, !dbg ![[END:[0-9]+]]
 }
 
 attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
@@ -40,20 +41,22 @@ attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointe
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\0012\00clang version 3.5 \000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [basic.c] [DW_LANG_C99]
-!1 = !{!"basic.c", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.5 ", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "basic.c", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\006\00256\000\001", !1, !5, !6, null, void (i32)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [basic.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32)* @foo, variables: !2)
+!5 = !DIFile(filename: "basic.c", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 4}
-!8 = !{i32 1, !"Debug Info Version", i32 2}
+!8 = !{i32 1, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5 "}
-!10 = !MDLocation(line: 3, scope: !11)
-!11 = !{!"0xb\003\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [basic.c]
-!12 = !MDLocation(line: 4, scope: !4)
+!10 = !DILocation(line: 3, scope: !11)
+!11 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !4)
+!12 = !DILocation(line: 4, scope: !4)
 
-; CHECK: !12 = !MDLocation(line: 3, scope: !13)
-; CHECK: !13 = !{!"0xb\001", !1, !11} ; [ DW_TAG_lexical_block ] [./basic.c]
-; CHECK: !14 = !MDLocation(line: 4, scope: !4)
+; CHECK: ![[FOO:[0-9]+]] = !DISubprogram(name: "foo"
+; CHECK: ![[BLOCK:[0-9]+]] = distinct !DILexicalBlock(scope: ![[FOO]],{{.*}} line: 3)
+; CHECK: ![[THEN]] = !DILocation(line: 3, scope: ![[BLOCKFILE:[0-9]+]])
+; CHECK: ![[BLOCKFILE]] = !DILexicalBlockFile(scope: ![[BLOCK]],{{.*}} discriminator: 1)
+; CHECK: ![[END]] = !DILocation(line: 4, scope: ![[FOO]])
diff --git a/test/Transforms/AddDiscriminators/first-only.ll b/test/Transforms/AddDiscriminators/first-only.ll
index 153cfc8a44cc..7f1ea2b15cf6 100644
--- a/test/Transforms/AddDiscriminators/first-only.ll
+++ b/test/Transforms/AddDiscriminators/first-only.ll
@@ -19,29 +19,29 @@ entry:
   %x = alloca i32, align 4
   %y = alloca i32, align 4
   store i32 %i, i32* %i.addr, align 4
-  %0 = load i32* %i.addr, align 4, !dbg !10
+  %0 = load i32, i32* %i.addr, align 4, !dbg !10
   %cmp = icmp slt i32 %0, 10, !dbg !10
   br i1 %cmp, label %if.then, label %if.end, !dbg !10
 
 if.then:                                          ; preds = %entry
-  %1 = load i32* %i.addr, align 4, !dbg !12
+  %1 = load i32, i32* %i.addr, align 4, !dbg !12
   store i32 %1, i32* %x, align 4, !dbg !12
 
-  %2 = load i32* %i.addr, align 4, !dbg !14
-; CHECK:  %2 = load i32* %i.addr, align 4, !dbg !15
+  %2 = load i32, i32* %i.addr, align 4, !dbg !14
+; CHECK:  %2 = load i32, i32* %i.addr, align 4, !dbg ![[THEN:[0-9]+]]
 
   %sub = sub nsw i32 0, %2, !dbg !14
-; CHECK:  %sub = sub nsw i32 0, %2, !dbg !15
+; CHECK:  %sub = sub nsw i32 0, %2, !dbg ![[THEN]]
 
   store i32 %sub, i32* %y, align 4, !dbg !14
-; CHECK:  store i32 %sub, i32* %y, align 4, !dbg !15
+; CHECK:  store i32 %sub, i32* %y, align 4, !dbg ![[THEN]]
 
   br label %if.end, !dbg !15
-; CHECK:  br label %if.end, !dbg !16
+; CHECK:  br label %if.end, !dbg ![[BR:[0-9]+]]
 
 if.end:                                           ; preds = %if.then, %entry
   ret void, !dbg !16
-; CHECK:  ret void, !dbg !17
+; CHECK:  ret void, !dbg ![[END:[0-9]+]]
 }
 
 attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
@@ -50,33 +50,34 @@ attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointe
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\0012\00clang version 3.5 (trunk 199750) (llvm/trunk 199751)\000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [first-only.c] [DW_LANG_C99]
-!1 = !{!"first-only.c", !"."}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.5 (trunk 199750) (llvm/trunk 199751)", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "first-only.c", directory: ".")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\006\00256\000\001", !1, !5, !6, null, void (i32)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [first-only.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32)* @foo, variables: !2)
+!5 = !DIFile(filename: "first-only.c", directory: ".")
+!6 = !DISubroutineType(types: !{null})
 !7 = !{i32 2, !"Dwarf Version", i32 4}
-!8 = !{i32 1, !"Debug Info Version", i32 2}
+!8 = !{i32 1, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5 (trunk 199750) (llvm/trunk 199751)"}
-!10 = !MDLocation(line: 3, scope: !11)
+!10 = !DILocation(line: 3, scope: !11)
 
-!11 = !{!"0xb\003\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [first-only.c]
-; CHECK: !11 = !{!"0xb\003\000\000", !1, !4}
+!11 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !4)
+; CHECK: ![[FOO:[0-9]+]] = !DISubprogram(name: "foo"
+; CHECK: ![[BLOCK1:[0-9]+]] = distinct !DILexicalBlock(scope: ![[FOO]],{{.*}} line: 3)
 
-!12 = !MDLocation(line: 3, scope: !13)
+!12 = !DILocation(line: 3, scope: !13)
 
-!13 = !{!"0xb\003\000\001", !1, !11} ; [ DW_TAG_lexical_block ] [first-only.c]
-; CHECK: !13 = !{!"0xb\001", !1, !14} ; [ DW_TAG_lexical_block ] [./first-only.c]
+!13 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !11)
+; CHECK: !DILexicalBlockFile(scope: ![[BLOCK2:[0-9]+]],{{.*}} discriminator: 1)
 
-!14 = !MDLocation(line: 4, scope: !13)
-; CHECK: !14 = !{!"0xb\003\000\001", !1, !11}
+!14 = !DILocation(line: 4, scope: !13)
+; CHECK: ![[BLOCK2]] = distinct !DILexicalBlock(scope: ![[BLOCK1]],{{.*}} line: 3)
 
-!15 = !MDLocation(line: 5, scope: !13)
-; CHECK: !15 = !MDLocation(line: 4, scope: !14)
+!15 = !DILocation(line: 5, scope: !13)
+; CHECK: ![[THEN]] = !DILocation(line: 4, scope: ![[BLOCK2]])
 
-!16 = !MDLocation(line: 6, scope: !4)
-; CHECK: !16 = !MDLocation(line: 5, scope: !14)
-; CHECK: !17 = !MDLocation(line: 6, scope: !4)
+!16 = !DILocation(line: 6, scope: !4)
+; CHECK: ![[BR]] = !DILocation(line: 5, scope: ![[BLOCK2]])
+; CHECK: ![[END]] = !DILocation(line: 6, scope: ![[FOO]])
 
diff --git a/test/Transforms/AddDiscriminators/multiple.ll b/test/Transforms/AddDiscriminators/multiple.ll
index 5e552a87bbb2..621a7117571c 100644
--- a/test/Transforms/AddDiscriminators/multiple.ll
+++ b/test/Transforms/AddDiscriminators/multiple.ll
@@ -15,29 +15,29 @@ entry:
   %i.addr = alloca i32, align 4
   %x = alloca i32, align 4
   store i32 %i, i32* %i.addr, align 4
-  %0 = load i32* %i.addr, align 4, !dbg !10
+  %0 = load i32, i32* %i.addr, align 4, !dbg !10
   %cmp = icmp slt i32 %0, 10, !dbg !10
   br i1 %cmp, label %if.then, label %if.else, !dbg !10
 
 if.then:                                          ; preds = %entry
-  %1 = load i32* %i.addr, align 4, !dbg !10
-; CHECK:  %1 = load i32* %i.addr, align 4, !dbg !12
+  %1 = load i32, i32* %i.addr, align 4, !dbg !10
+; CHECK:  %1 = load i32, i32* %i.addr, align 4, !dbg ![[THEN:[0-9]+]]
 
   store i32 %1, i32* %x, align 4, !dbg !10
-; CHECK:  store i32 %1, i32* %x, align 4, !dbg !12
+; CHECK:  store i32 %1, i32* %x, align 4, !dbg ![[THEN]]
 
   br label %if.end, !dbg !10
-; CHECK:  br label %if.end, !dbg !12
+; CHECK:  br label %if.end, !dbg ![[THEN]]
 
 if.else:                                          ; preds = %entry
-  %2 = load i32* %i.addr, align 4, !dbg !10
-; CHECK:  %2 = load i32* %i.addr, align 4, !dbg !14
+  %2 = load i32, i32* %i.addr, align 4, !dbg !10
+; CHECK:  %2 = load i32, i32* %i.addr, align 4, !dbg ![[ELSE:[0-9]+]]
 
   %sub = sub nsw i32 0, %2, !dbg !10
-; CHECK:  %sub = sub nsw i32 0, %2, !dbg !14
+; CHECK:  %sub = sub nsw i32 0, %2, !dbg ![[ELSE]]
 
   store i32 %sub, i32* %x, align 4, !dbg !10
-; CHECK:  store i32 %sub, i32* %x, align 4, !dbg !14
+; CHECK:  store i32 %sub, i32* %x, align 4, !dbg ![[ELSE]]
 
   br label %if.end
 
@@ -51,21 +51,22 @@ attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointe
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\0012\00clang version 3.5 (trunk 199750) (llvm/trunk 199751)\000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [multiple.c] [DW_LANG_C99]
-!1 = !{!"multiple.c", !"."}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.5 (trunk 199750) (llvm/trunk 199751)", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "multiple.c", directory: ".")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\006\00256\000\001", !1, !5, !6, null, void (i32)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [multiple.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32)* @foo, variables: !2)
+!5 = !DIFile(filename: "multiple.c", directory: ".")
+!6 = !DISubroutineType(types: !{null, !13})
+!13 = !DIBasicType(encoding: DW_ATE_signed, name: "int", size: 32, align: 32)
 !7 = !{i32 2, !"Dwarf Version", i32 4}
-!8 = !{i32 1, !"Debug Info Version", i32 2}
+!8 = !{i32 1, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5 (trunk 199750) (llvm/trunk 199751)"}
-!10 = !MDLocation(line: 3, scope: !11)
-!11 = !{!"0xb\003\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [multiple.c]
-!12 = !MDLocation(line: 4, scope: !4)
+!10 = !DILocation(line: 3, scope: !11)
+!11 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !4)
+!12 = !DILocation(line: 4, scope: !4)
 
-; CHECK: !12 = !MDLocation(line: 3, scope: !13)
-; CHECK: !13 = !{!"0xb\001", !1, !11} ; [ DW_TAG_lexical_block ] [./multiple.c]
-; CHECK: !14 = !MDLocation(line: 3, scope: !15)
-; CHECK: !15 = !{!"0xb\002", !1, !11} ; [ DW_TAG_lexical_block ] [./multiple.c]
+; CHECK: ![[THEN]] = !DILocation(line: 3, scope: ![[THENBLOCK:[0-9]+]])
+; CHECK: ![[THENBLOCK]] = !DILexicalBlockFile(scope: ![[SCOPE:[0-9]+]],{{.*}} discriminator: 1)
+; CHECK: ![[ELSE]] = !DILocation(line: 3, scope: ![[ELSEBLOCK:[0-9]+]])
+; CHECK: ![[ELSEBLOCK]] = !DILexicalBlockFile(scope: ![[SCOPE]],{{.*}} discriminator: 2)
diff --git a/test/Transforms/AddDiscriminators/no-discriminators.ll b/test/Transforms/AddDiscriminators/no-discriminators.ll
index dd7faf0a71e3..895967e73b4d 100644
--- a/test/Transforms/AddDiscriminators/no-discriminators.ll
+++ b/test/Transforms/AddDiscriminators/no-discriminators.ll
@@ -17,13 +17,13 @@ entry:
   %retval = alloca i32, align 4
   %i.addr = alloca i64, align 8
   store i64 %i, i64* %i.addr, align 8
-  call void @llvm.dbg.declare(metadata i64* %i.addr, metadata !13, metadata !{}), !dbg !14
-  %0 = load i64* %i.addr, align 8, !dbg !15
-; CHECK:  %0 = load i64* %i.addr, align 8, !dbg !15
+  call void @llvm.dbg.declare(metadata i64* %i.addr, metadata !13, metadata !DIExpression()), !dbg !14
+  %0 = load i64, i64* %i.addr, align 8, !dbg !15
+; CHECK:  %0 = load i64, i64* %i.addr, align 8, !dbg ![[ENTRY:[0-9]+]]
   %cmp = icmp slt i64 %0, 5, !dbg !15
-; CHECK:  %cmp = icmp slt i64 %0, 5, !dbg !15
+; CHECK:  %cmp = icmp slt i64 %0, 5, !dbg ![[ENTRY:[0-9]+]]
   br i1 %cmp, label %if.then, label %if.else, !dbg !15
-; CHECK:  br i1 %cmp, label %if.then, label %if.else, !dbg !15
+; CHECK:  br i1 %cmp, label %if.then, label %if.else, !dbg ![[ENTRY:[0-9]+]]
 
 if.then:                                          ; preds = %entry
   store i32 2, i32* %retval, !dbg !15
@@ -34,7 +34,7 @@ if.else:                                          ; preds = %entry
   br label %return, !dbg !15
 
 return:                                           ; preds = %if.else, %if.then
-  %1 = load i32* %retval, !dbg !17
+  %1 = load i32, i32* %retval, !dbg !17
   ret i32 %1, !dbg !17
 }
 
@@ -48,24 +48,25 @@ attributes #1 = { nounwind readnone }
 !llvm.module.flags = !{!10, !11}
 !llvm.ident = !{!12}
 
-!0 = !{!"0x11\0012\00clang version 3.5.0 \000\00\000\00\001", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./no-discriminators] [DW_LANG_C99]
-!1 = !{!"no-discriminators", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.5.0 ", isOptimized: false, emissionKind: 1, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "no-discriminators", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\006\00256\000\001", !1, !5, !6, null, i32 (i64)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./no-discriminators]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: i32 (i64)* @foo, variables: !2)
+; CHECK: ![[FOO:[0-9]+]] = !DISubprogram(name: "foo"
+!5 = !DIFile(filename: "no-discriminators", directory: ".")
+!6 = !DISubroutineType(types: !7)
 !7 = !{!8, !9}
-!8 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!9 = !{!"0x24\00long int\000\0064\0064\000\000\005", null, null} ; [ DW_TAG_base_type ] [long int] [line 0, size 64, align 64, offset 0, enc DW_ATE_signed]
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!9 = !DIBasicType(tag: DW_TAG_base_type, name: "long int", size: 64, align: 64, encoding: DW_ATE_signed)
 !10 = !{i32 2, !"Dwarf Version", i32 2}
-; CHECK: !10 = !{i32 2, !"Dwarf Version", i32 2}
-!11 = !{i32 1, !"Debug Info Version", i32 2}
+; CHECK: !{i32 2, !"Dwarf Version", i32 2}
+!11 = !{i32 1, !"Debug Info Version", i32 3}
 !12 = !{!"clang version 3.5.0 "}
-!13 = !{!"0x101\00i\0016777217\000", !4, !5, !9} ; [ DW_TAG_arg_variable ] [i] [line 1]
-!14 = !MDLocation(line: 1, scope: !4)
-!15 = !MDLocation(line: 2, scope: !16)
-; CHECK: !15 = !MDLocation(line: 2, scope: !16)
-!16 = !{!"0xb\002\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [./no-discriminators]
-; CHECK: !16 = !{!"0xb\002\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [./no-discriminators]
-!17 = !MDLocation(line: 3, scope: !4)
+!13 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i", line: 1, arg: 1, scope: !4, file: !5, type: !9)
+!14 = !DILocation(line: 1, scope: !4)
+!15 = !DILocation(line: 2, scope: !16)
+; CHECK: ![[ENTRY]] = !DILocation(line: 2, scope: ![[BLOCK:[0-9]+]])
+!16 = distinct !DILexicalBlock(line: 2, column: 0, file: !1, scope: !4)
+; CHECK: ![[BLOCK]] = distinct !DILexicalBlock(scope: ![[FOO]],{{.*}} line: 2)
+!17 = !DILocation(line: 3, scope: !4)
diff --git a/test/Transforms/AlignmentFromAssumptions/simple.ll b/test/Transforms/AlignmentFromAssumptions/simple.ll
index 884c8bad02f5..851e6dc3ccc6 100644
--- a/test/Transforms/AlignmentFromAssumptions/simple.ll
+++ b/test/Transforms/AlignmentFromAssumptions/simple.ll
@@ -7,11 +7,11 @@ entry:
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -22,12 +22,12 @@ entry:
   %maskedptr = and i64 %offsetptr, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds i32* %a, i64 2
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 2
+  %0 = load i32, i32* %arrayidx, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo2
-; CHECK: load i32* {{[^,]+}}, align 16
+; CHECK: load i32, i32* {{[^,]+}}, align 16
 ; CHECK: ret i32
 }
 
@@ -38,12 +38,12 @@ entry:
   %maskedptr = and i64 %offsetptr, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds i32* %a, i64 -1
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 -1
+  %0 = load i32, i32* %arrayidx, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo2a
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -53,11 +53,11 @@ entry:
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @goo
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -72,8 +72,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 8
   %1 = trunc i64 %indvars.iv.next to i32
@@ -85,7 +85,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @hoo
-; CHECK: load i32* %arrayidx, align 32
+; CHECK: load i32, i32* %arrayidx, align 32
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -100,8 +100,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 4, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 8
   %1 = trunc i64 %indvars.iv.next to i32
@@ -113,7 +113,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @joo
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -128,8 +128,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 4
   %1 = trunc i64 %indvars.iv.next to i32
@@ -141,7 +141,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @koo
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -156,8 +156,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ -4, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 4
   %1 = trunc i64 %indvars.iv.next to i32
@@ -169,7 +169,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @koo2
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
diff --git a/test/Transforms/AlignmentFromAssumptions/simple32.ll b/test/Transforms/AlignmentFromAssumptions/simple32.ll
index 166e7ef38936..2edc2e95f41a 100644
--- a/test/Transforms/AlignmentFromAssumptions/simple32.ll
+++ b/test/Transforms/AlignmentFromAssumptions/simple32.ll
@@ -7,11 +7,11 @@ entry:
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -22,12 +22,12 @@ entry:
   %maskedptr = and i64 %offsetptr, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds i32* %a, i64 2
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 2
+  %0 = load i32, i32* %arrayidx, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo2
-; CHECK: load i32* {{[^,]+}}, align 16
+; CHECK: load i32, i32* {{[^,]+}}, align 16
 ; CHECK: ret i32
 }
 
@@ -38,12 +38,12 @@ entry:
   %maskedptr = and i64 %offsetptr, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds i32* %a, i64 -1
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 -1
+  %0 = load i32, i32* %arrayidx, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo2a
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -53,11 +53,11 @@ entry:
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @goo
-; CHECK: load i32* {{[^,]+}}, align 32
+; CHECK: load i32, i32* {{[^,]+}}, align 32
 ; CHECK: ret i32
 }
 
@@ -72,8 +72,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 8
   %1 = trunc i64 %indvars.iv.next to i32
@@ -85,7 +85,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @hoo
-; CHECK: load i32* %arrayidx, align 32
+; CHECK: load i32, i32* %arrayidx, align 32
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -100,8 +100,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 4, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 8
   %1 = trunc i64 %indvars.iv.next to i32
@@ -113,7 +113,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @joo
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -128,8 +128,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 4
   %1 = trunc i64 %indvars.iv.next to i32
@@ -141,7 +141,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @koo
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
@@ -156,8 +156,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ -4, %entry ], [ %indvars.iv.next, %for.body ]
   %r.06 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.06
   %indvars.iv.next = add i64 %indvars.iv, 4
   %1 = trunc i64 %indvars.iv.next to i32
@@ -169,7 +169,7 @@ for.end:                                          ; preds = %for.body
   ret i32 %add.lcssa
 
 ; CHECK-LABEL: @koo2
-; CHECK: load i32* %arrayidx, align 16
+; CHECK: load i32, i32* %arrayidx, align 16
 ; CHECK: ret i32 %add.lcssa
 }
 
diff --git a/test/Transforms/AlignmentFromAssumptions/start-unk.ll b/test/Transforms/AlignmentFromAssumptions/start-unk.ll
index b7fe24947cb5..99533cf6ccb9 100644
--- a/test/Transforms/AlignmentFromAssumptions/start-unk.ll
+++ b/test/Transforms/AlignmentFromAssumptions/start-unk.ll
@@ -79,13 +79,13 @@ if.then126:                                       ; preds = %if.end123
   %maskcond.i.i187 = icmp eq i64 %maskedptr.i.i186, 0
   tail call void @llvm.assume(i1 %maskcond.i.i187) #0
   %ret.0..sroa_cast.i.i188 = bitcast %type1* undef to i32*
-  %ret.0.copyload.i.i189 = load i32* %ret.0..sroa_cast.i.i188, align 2
+  %ret.0.copyload.i.i189 = load i32, i32* %ret.0..sroa_cast.i.i188, align 2
 
 ; CHECK: load {{.*}} align 2
 
   %0 = tail call i32 @llvm.bswap.i32(i32 %ret.0.copyload.i.i189) #0
   %conv131 = zext i32 %0 to i64
-  %add.ptr132 = getelementptr inbounds i8* undef, i64 %conv131
+  %add.ptr132 = getelementptr inbounds i8, i8* undef, i64 %conv131
   %1 = bitcast i8* %add.ptr132 to %type1*
   br i1 undef, label %if.end150, label %if.end.i173
 
@@ -113,11 +113,11 @@ for.body137:                                      ; preds = %test1b.exit, %for.b
   br i1 undef, label %for.body137.if.end146_crit_edge, label %if.then140
 
 for.body137.if.end146_crit_edge:                  ; preds = %for.body137
-  %incdec.ptr = getelementptr inbounds %type1* %ShndxTable.0309, i64 1
+  %incdec.ptr = getelementptr inbounds %type1, %type1* %ShndxTable.0309, i64 1
   br i1 undef, label %cond.false.i70, label %cond.end.i
 
 if.then140:                                       ; preds = %for.body137
-  %ret.0.copyload.i.i102 = load i32* %ret.0..sroa_cast.i.i106, align 2
+  %ret.0.copyload.i.i102 = load i32, i32* %ret.0..sroa_cast.i.i106, align 2
 
 ; CHECK: load {{.*}} align 2
 
diff --git a/test/Transforms/ArgumentPromotion/2008-02-01-ReturnAttrs.ll b/test/Transforms/ArgumentPromotion/2008-02-01-ReturnAttrs.ll
index 1226b98a998e..c988774da8a7 100644
--- a/test/Transforms/ArgumentPromotion/2008-02-01-ReturnAttrs.ll
+++ b/test/Transforms/ArgumentPromotion/2008-02-01-ReturnAttrs.ll
@@ -3,7 +3,7 @@
 ; CHECK: define internal i32 @deref(i32 %x.val) #0 {
 define internal i32 @deref(i32* %x) nounwind {
 entry:
-  %tmp2 = load i32* %x, align 4
+  %tmp2 = load i32, i32* %x, align 4
   ret i32 %tmp2
 }
 
diff --git a/test/Transforms/ArgumentPromotion/2008-07-02-array-indexing.ll b/test/Transforms/ArgumentPromotion/2008-07-02-array-indexing.ll
index 210eb97bc1dc..267a6c045974 100644
--- a/test/Transforms/ArgumentPromotion/2008-07-02-array-indexing.ll
+++ b/test/Transforms/ArgumentPromotion/2008-07-02-array-indexing.ll
@@ -7,14 +7,14 @@
 define internal i32 @callee(i1 %C, i32* %A) {
 entry:
         ; Unconditonally load the element at %A
-        %A.0 = load i32* %A
+        %A.0 = load i32, i32* %A
         br i1 %C, label %T, label %F
 T:
         ret i32 %A.0
 F:
         ; Load the element at offset two from %A. This should not be promoted!
-        %A.2 = getelementptr i32* %A, i32 2
-        %R = load i32* %A.2
+        %A.2 = getelementptr i32, i32* %A, i32 2
+        %R = load i32, i32* %A.2
         ret i32 %R
 }
 
diff --git a/test/Transforms/ArgumentPromotion/aggregate-promote.ll b/test/Transforms/ArgumentPromotion/aggregate-promote.ll
index 12de51172739..3f521bace7f3 100644
--- a/test/Transforms/ArgumentPromotion/aggregate-promote.ll
+++ b/test/Transforms/ArgumentPromotion/aggregate-promote.ll
@@ -9,10 +9,10 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
     i32 25 }            ; <%QuadTy*> [#uses=1]
 
 define internal i32 @test(%QuadTy* %P) {
-        %A = getelementptr %QuadTy* %P, i64 0, i32 3            ; <i32*> [#uses=1]
-        %B = getelementptr %QuadTy* %P, i64 0, i32 2            ; <i32*> [#uses=1]
-        %a = load i32* %A               ; <i32> [#uses=1]
-        %b = load i32* %B               ; <i32> [#uses=1]
+        %A = getelementptr %QuadTy, %QuadTy* %P, i64 0, i32 3            ; <i32*> [#uses=1]
+        %B = getelementptr %QuadTy, %QuadTy* %P, i64 0, i32 2            ; <i32*> [#uses=1]
+        %a = load i32, i32* %A               ; <i32> [#uses=1]
+        %b = load i32, i32* %B               ; <i32> [#uses=1]
         %V = add i32 %a, %b             ; <i32> [#uses=1]
         ret i32 %V
 }
diff --git a/test/Transforms/ArgumentPromotion/attrs.ll b/test/Transforms/ArgumentPromotion/attrs.ll
index 49c075029951..46128f93c240 100644
--- a/test/Transforms/ArgumentPromotion/attrs.ll
+++ b/test/Transforms/ArgumentPromotion/attrs.ll
@@ -4,8 +4,8 @@
 
 define internal void @f(%struct.ss* byval  %b, i32* byval %X, i32 %i) nounwind  {
 entry:
-	%tmp = getelementptr %struct.ss* %b, i32 0, i32 0
-	%tmp1 = load i32* %tmp, align 4
+	%tmp = getelementptr %struct.ss, %struct.ss* %b, i32 0, i32 0
+	%tmp1 = load i32, i32* %tmp, align 4
 	%tmp2 = add i32 %tmp1, 1	
 	store i32 %tmp2, i32* %tmp, align 4
 
@@ -16,9 +16,9 @@ entry:
 define i32 @test(i32* %X) {
 entry:
 	%S = alloca %struct.ss		; <%struct.ss*> [#uses=4]
-	%tmp1 = getelementptr %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp1, align 8
-	%tmp4 = getelementptr %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
+	%tmp4 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
 	store i64 2, i64* %tmp4, align 4
 	call void @f( %struct.ss* byval %S, i32* byval %X, i32 zeroext 0) 
 	ret i32 0
diff --git a/test/Transforms/ArgumentPromotion/basictest.ll b/test/Transforms/ArgumentPromotion/basictest.ll
index 8f78b98437bc..89888bbc98a7 100644
--- a/test/Transforms/ArgumentPromotion/basictest.ll
+++ b/test/Transforms/ArgumentPromotion/basictest.ll
@@ -3,8 +3,8 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 
 define internal i32 @test(i32* %X, i32* %Y) {
 ; CHECK-LABEL: define internal i32 @test(i32 %X.val, i32 %Y.val)
-  %A = load i32* %X
-  %B = load i32* %Y
+  %A = load i32, i32* %X
+  %B = load i32, i32* %Y
   %C = add i32 %A, %B
   ret i32 %C
 }
diff --git a/test/Transforms/ArgumentPromotion/byval-2.ll b/test/Transforms/ArgumentPromotion/byval-2.ll
index b412f5ef0856..6c0288f5f989 100644
--- a/test/Transforms/ArgumentPromotion/byval-2.ll
+++ b/test/Transforms/ArgumentPromotion/byval-2.ll
@@ -8,8 +8,8 @@
 define internal void @f(%struct.ss* byval  %b, i32* byval %X) nounwind  {
 ; CHECK-LABEL: define internal void @f(i32 %b.0, i64 %b.1, i32* byval %X)
 entry:
-  %tmp = getelementptr %struct.ss* %b, i32 0, i32 0
-  %tmp1 = load i32* %tmp, align 4
+  %tmp = getelementptr %struct.ss, %struct.ss* %b, i32 0, i32 0
+  %tmp1 = load i32, i32* %tmp, align 4
   %tmp2 = add i32 %tmp1, 1
   store i32 %tmp2, i32* %tmp, align 4
 
@@ -21,9 +21,9 @@ define i32 @test(i32* %X) {
 ; CHECK-LABEL: define i32 @test
 entry:
   %S = alloca %struct.ss
-  %tmp1 = getelementptr %struct.ss* %S, i32 0, i32 0
+  %tmp1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0
   store i32 1, i32* %tmp1, align 8
-  %tmp4 = getelementptr %struct.ss* %S, i32 0, i32 1
+  %tmp4 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1
   store i64 2, i64* %tmp4, align 4
   call void @f( %struct.ss* byval %S, i32* byval %X)
 ; CHECK: call void @f(i32 %{{.*}}, i64 %{{.*}}, i32* byval %{{.*}})
diff --git a/test/Transforms/ArgumentPromotion/byval.ll b/test/Transforms/ArgumentPromotion/byval.ll
index 27305e92068d..b091b09a3597 100644
--- a/test/Transforms/ArgumentPromotion/byval.ll
+++ b/test/Transforms/ArgumentPromotion/byval.ll
@@ -7,8 +7,8 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 define internal void @f(%struct.ss* byval  %b) nounwind  {
 ; CHECK-LABEL: define internal void @f(i32 %b.0, i64 %b.1)
 entry:
-  %tmp = getelementptr %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
-  %tmp1 = load i32* %tmp, align 4		; <i32> [#uses=1]
+  %tmp = getelementptr %struct.ss, %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
+  %tmp1 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
   %tmp2 = add i32 %tmp1, 1		; <i32> [#uses=1]
   store i32 %tmp2, i32* %tmp, align 4
   ret void
@@ -18,9 +18,9 @@ define i32 @main() nounwind  {
 ; CHECK-LABEL: define i32 @main
 entry:
   %S = alloca %struct.ss		; <%struct.ss*> [#uses=4]
-  %tmp1 = getelementptr %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
+  %tmp1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
   store i32 1, i32* %tmp1, align 8
-  %tmp4 = getelementptr %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
+  %tmp4 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
   store i64 2, i64* %tmp4, align 4
   call void @f( %struct.ss* byval  %S ) nounwind 
 ; CHECK: call void @f(i32 %{{.*}}, i64 %{{.*}})
diff --git a/test/Transforms/ArgumentPromotion/chained.ll b/test/Transforms/ArgumentPromotion/chained.ll
index c9a453899d7a..6ba2e8d48694 100644
--- a/test/Transforms/ArgumentPromotion/chained.ll
+++ b/test/Transforms/ArgumentPromotion/chained.ll
@@ -5,8 +5,8 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 @G2 = constant i32* @G1         ; <i32**> [#uses=1]
 
 define internal i32 @test(i32** %X) {
-        %Y = load i32** %X              ; <i32*> [#uses=1]
-        %X.upgrd.1 = load i32* %Y               ; <i32> [#uses=1]
+        %Y = load i32*, i32** %X              ; <i32*> [#uses=1]
+        %X.upgrd.1 = load i32, i32* %Y               ; <i32> [#uses=1]
         ret i32 %X.upgrd.1
 }
 
diff --git a/test/Transforms/ArgumentPromotion/control-flow.ll b/test/Transforms/ArgumentPromotion/control-flow.ll
index e4a61da45cf3..cdff36eb83c0 100644
--- a/test/Transforms/ArgumentPromotion/control-flow.ll
+++ b/test/Transforms/ArgumentPromotion/control-flow.ll
@@ -8,7 +8,7 @@ T:              ; preds = %0
         ret i32 17
 
 F:              ; preds = %0
-        %X = load i32* %P               ; <i32> [#uses=1]
+        %X = load i32, i32* %P               ; <i32> [#uses=1]
         ret i32 %X
 }
 
diff --git a/test/Transforms/ArgumentPromotion/control-flow2.ll b/test/Transforms/ArgumentPromotion/control-flow2.ll
index 2543218bafe8..7413f46a860f 100644
--- a/test/Transforms/ArgumentPromotion/control-flow2.ll
+++ b/test/Transforms/ArgumentPromotion/control-flow2.ll
@@ -1,5 +1,6 @@
-; RUN: opt < %s -argpromotion -S | \
-; RUN:   grep "load i32\* %A"
+; RUN: opt < %s -argpromotion -S | FileCheck %s
+
+; CHECK: load i32, i32* %A
 target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128"
 
 define internal i32 @callee(i1 %C, i32* %P) {
@@ -9,7 +10,7 @@ T:              ; preds = %0
         ret i32 17
 
 F:              ; preds = %0
-        %X = load i32* %P               ; <i32> [#uses=1]
+        %X = load i32, i32* %P               ; <i32> [#uses=1]
         ret i32 %X
 }
 
diff --git a/test/Transforms/ArgumentPromotion/crash.ll b/test/Transforms/ArgumentPromotion/crash.ll
index 5e1a0370dbb1..dbd343ae920e 100644
--- a/test/Transforms/ArgumentPromotion/crash.ll
+++ b/test/Transforms/ArgumentPromotion/crash.ll
@@ -50,10 +50,10 @@ init:
 
 define internal i32 @"clay_assign(Chain, Chain)"(%0* %c, %0* %d) {
 init:
-  %0 = getelementptr %0* %d, i32 0, i32 0
-  %1 = load %0** %0
-  %2 = getelementptr %0* %c, i32 0, i32 0
-  %3 = load %0** %2
+  %0 = getelementptr %0, %0* %d, i32 0, i32 0
+  %1 = load %0*, %0** %0
+  %2 = getelementptr %0, %0* %c, i32 0, i32 0
+  %3 = load %0*, %0** %2
   %4 = call i32 @"clay_assign(Chain, Chain)"(%0* %3, %0* %1)
   ret i32 0
 }
diff --git a/test/Transforms/ArgumentPromotion/dbg.ll b/test/Transforms/ArgumentPromotion/dbg.ll
index 65cf3678bd99..17a34cb62239 100644
--- a/test/Transforms/ArgumentPromotion/dbg.ll
+++ b/test/Transforms/ArgumentPromotion/dbg.ll
@@ -1,12 +1,12 @@
 ; RUN: opt < %s -argpromotion -S | FileCheck %s
 ; CHECK: call void @test(i32 %
-; CHECK: void (i32)* @test, {{.*}} ; [ DW_TAG_subprogram ] {{.*}} [test]
+; CHECK: !DISubprogram(name: "test",{{.*}} function: void (i32)* @test
 
 declare void @sink(i32)
 
 define internal void @test(i32** %X) {
-  %1 = load i32** %X, align 8
-  %2 = load i32* %1, align 8
+  %1 = load i32*, i32** %X, align 8
+  %2 = load i32, i32* %1, align 8
   call void @sink(i32 %2)
   ret void
 }
@@ -19,8 +19,9 @@ define void @caller(i32** %Y) {
 !llvm.module.flags = !{!0}
 !llvm.dbg.cu = !{!3}
 
-!0 = !{i32 2, !"Debug Info Version", i32 2}
-!1 = !MDLocation(line: 8, scope: !2)
-!2 = !{!"0x2e\00test\00test\00\003\001\001\000\006\00256\000\003", null, null, null, null, void (i32**)* @test, null, null, null} ; [ DW_TAG_subprogram ]
-!3 = !{!"0x11\004\00clang version 3.5.0 \000\00\000\00\002", null, null, null, !4, null, null} ; [ DW_TAG_compile_unit ] [/usr/local/google/home/blaikie/dev/scratch/pr20038/reduce/<stdin>] [DW_LANG_C_plus_plus]
+!0 = !{i32 2, !"Debug Info Version", i32 3}
+!1 = !DILocation(line: 8, scope: !2)
+!2 = !DISubprogram(name: "test", line: 3, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 3, scope: null, function: void (i32**)* @test)
+!3 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0 ", isOptimized: false, emissionKind: 2, file: !5, subprograms: !4)
 !4 = !{!2}
+!5 = !DIFile(filename: "test.c", directory: "")
diff --git a/test/Transforms/ArgumentPromotion/fp80.ll b/test/Transforms/ArgumentPromotion/fp80.ll
index a770d6013d6b..84ef603de82c 100644
--- a/test/Transforms/ArgumentPromotion/fp80.ll
+++ b/test/Transforms/ArgumentPromotion/fp80.ll
@@ -24,22 +24,22 @@ entry:
 define internal i8 @UseLongDoubleUnsafely(%union.u* byval align 16 %arg) {
 entry:
   %bitcast = bitcast %union.u* %arg to %struct.s*
-  %gep = getelementptr inbounds %struct.s* %bitcast, i64 0, i32 2
-  %result = load i8* %gep
+  %gep = getelementptr inbounds %struct.s, %struct.s* %bitcast, i64 0, i32 2
+  %result = load i8, i8* %gep
   ret i8 %result
 }
 
 ; CHECK: internal x86_fp80 @UseLongDoubleSafely(x86_fp80 {{%.*}}) {
 define internal x86_fp80 @UseLongDoubleSafely(%union.u* byval align 16 %arg) {
-  %gep = getelementptr inbounds %union.u* %arg, i64 0, i32 0
-  %fp80 = load x86_fp80* %gep
+  %gep = getelementptr inbounds %union.u, %union.u* %arg, i64 0, i32 0
+  %fp80 = load x86_fp80, x86_fp80* %gep
   ret x86_fp80 %fp80
 }
 
 ; CHECK: define internal i64 @AccessPaddingOfStruct(%struct.Foo* byval %a) {
 define internal i64 @AccessPaddingOfStruct(%struct.Foo* byval %a) {
   %p = bitcast %struct.Foo* %a to i64*
-  %v = load i64* %p
+  %v = load i64, i64* %p
   ret i64 %v
 }
 
@@ -53,6 +53,6 @@ loop:
   %phi = phi %struct.Foo* [ null, %entry ], [ %gep, %loop ]
   %0   = phi %struct.Foo* [ %a, %entry ],   [ %0, %loop ]
   store %struct.Foo* %phi, %struct.Foo** %a_ptr
-  %gep = getelementptr %struct.Foo* %a, i64 0
+  %gep = getelementptr %struct.Foo, %struct.Foo* %a, i64 0
   br label %loop
 }
diff --git a/test/Transforms/ArgumentPromotion/inalloca.ll b/test/Transforms/ArgumentPromotion/inalloca.ll
index 089a78f6b319..80bd6fdbc406 100644
--- a/test/Transforms/ArgumentPromotion/inalloca.ll
+++ b/test/Transforms/ArgumentPromotion/inalloca.ll
@@ -7,10 +7,10 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 ; Argpromote + scalarrepl should change this to passing the two integers by value.
 define internal i32 @f(%struct.ss* inalloca  %s) {
 entry:
-  %f0 = getelementptr %struct.ss* %s, i32 0, i32 0
-  %f1 = getelementptr %struct.ss* %s, i32 0, i32 1
-  %a = load i32* %f0, align 4
-  %b = load i32* %f1, align 4
+  %f0 = getelementptr %struct.ss, %struct.ss* %s, i32 0, i32 0
+  %f1 = getelementptr %struct.ss, %struct.ss* %s, i32 0, i32 1
+  %a = load i32, i32* %f0, align 4
+  %b = load i32, i32* %f1, align 4
   %r = add i32 %a, %b
   ret i32 %r
 }
@@ -21,8 +21,8 @@ entry:
 define i32 @main() {
 entry:
   %S = alloca inalloca %struct.ss
-  %f0 = getelementptr %struct.ss* %S, i32 0, i32 0
-  %f1 = getelementptr %struct.ss* %S, i32 0, i32 1
+  %f0 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0
+  %f1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1
   store i32 1, i32* %f0, align 4
   store i32 2, i32* %f1, align 4
   %r = call i32 @f(%struct.ss* inalloca %S)
diff --git a/test/Transforms/ArgumentPromotion/reserve-tbaa.ll b/test/Transforms/ArgumentPromotion/reserve-tbaa.ll
index db9d70d337a1..3c8ed79eeb29 100644
--- a/test/Transforms/ArgumentPromotion/reserve-tbaa.ll
+++ b/test/Transforms/ArgumentPromotion/reserve-tbaa.ll
@@ -14,9 +14,9 @@
 
 define internal fastcc void @fn(i32* nocapture readonly %p1, i64* nocapture readonly %p2) {
 entry:
-  %0 = load i64* %p2, align 8, !tbaa !1
+  %0 = load i64, i64* %p2, align 8, !tbaa !1
   %conv = trunc i64 %0 to i32
-  %1 = load i32* %p1, align 4, !tbaa !5
+  %1 = load i32, i32* %p1, align 4, !tbaa !5
   %conv1 = trunc i32 %1 to i8
   store i8 %conv1, i8* @d, align 1, !tbaa !7
   ret void
@@ -26,11 +26,11 @@ define i32 @main() {
 entry:
 ; CHECK-LABEL: main
 ; CHECK: store i32 1, i32* %{{.*}}, align 4, !tbaa ![[I32:[0-9]+]]
-; CHECK: %g.val = load i32* @g, align 4, !tbaa ![[I32]]
-; CHECK: %c.val = load i64* @c, align 8, !tbaa ![[LONG:[0-9]+]]
-  %0 = load i32*** @e, align 8, !tbaa !8
+; CHECK: %g.val = load i32, i32* @g, align 4, !tbaa ![[I32]]
+; CHECK: %c.val = load i64, i64* @c, align 8, !tbaa ![[LONG:[0-9]+]]
+  %0 = load i32**, i32*** @e, align 8, !tbaa !8
   store i32* @g, i32** %0, align 8, !tbaa !8
-  %1 = load i32** @a, align 8, !tbaa !8
+  %1 = load i32*, i32** @a, align 8, !tbaa !8
   store i32 1, i32* %1, align 4, !tbaa !5
   call fastcc void @fn(i32* @g, i64* @c)
 
diff --git a/test/Transforms/ArgumentPromotion/variadic.ll b/test/Transforms/ArgumentPromotion/variadic.ll
index 0ae52b3bbbd1..0e03882d3b20 100644
--- a/test/Transforms/ArgumentPromotion/variadic.ll
+++ b/test/Transforms/ArgumentPromotion/variadic.ll
@@ -15,7 +15,7 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define i32 @main(i32 %argc, i8** nocapture readnone %argv) #0 {
 entry:
-  tail call void (i8*, i8*, i8*, i8*, i8*, ...)* @callee_t0f(i8* undef, i8* undef, i8* undef, i8* undef, i8* undef, %struct.tt0* byval align 8 @t45)
+  tail call void (i8*, i8*, i8*, i8*, i8*, ...) @callee_t0f(i8* undef, i8* undef, i8* undef, i8* undef, i8* undef, %struct.tt0* byval align 8 @t45)
   ret i32 0
 }
 
diff --git a/test/Transforms/BBVectorize/X86/loop1.ll b/test/Transforms/BBVectorize/X86/loop1.ll
index 401808441c26..c3c30454ce33 100644
--- a/test/Transforms/BBVectorize/X86/loop1.ll
+++ b/test/Transforms/BBVectorize/X86/loop1.ll
@@ -12,10 +12,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double* %in1, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx2 = getelementptr inbounds double* %in2, i64 %indvars.iv
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx = getelementptr inbounds double, double* %in1, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %in2, i64 %indvars.iv
+  %1 = load double, double* %arrayidx2, align 8
   %mul = fmul double %0, %0
   %mul3 = fmul double %0, %1
   %add = fadd double %mul, %mul3
@@ -28,7 +28,7 @@ for.body:                                         ; preds = %for.body, %entry
   %add10 = fadd double %add9, %0
   %mul11 = fmul double %mul8, %add10
   %add12 = fadd double %add7, %mul11
-  %arrayidx14 = getelementptr inbounds double* %out, i64 %indvars.iv
+  %arrayidx14 = getelementptr inbounds double, double* %out, i64 %indvars.iv
   store double %add12, double* %arrayidx14, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/BBVectorize/X86/pr15289.ll b/test/Transforms/BBVectorize/X86/pr15289.ll
index 42bd0ff14032..a383a260fafd 100644
--- a/test/Transforms/BBVectorize/X86/pr15289.ll
+++ b/test/Transforms/BBVectorize/X86/pr15289.ll
@@ -44,43 +44,43 @@ entry:
   %12 = fsub double undef, %7
   %13 = fmul double %3, %12
   %14 = fmul double %3, undef
-  %15 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 0
+  %15 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 0
   store double %13, double* %15, align 8
-  %16 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 1
+  %16 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 0, i32 1
   %17 = fmul double undef, %8
   %18 = fmul double %17, undef
   %19 = fmul double undef, %18
   %20 = fadd double undef, undef
   %21 = fmul double %3, %19
   %22 = fsub double -0.000000e+00, %21
-  %23 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 0
+  %23 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 0
   store double %22, double* %23, align 8
-  %24 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 1
+  %24 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 1, i32 1
   %25 = fmul double undef, 0x3FE42F601A8C6794
   %26 = fmul double undef, 2.000000e+00
   %27 = fsub double %26, %0
   %28 = fmul double %6, undef
   %29 = fsub double undef, %28
-  %30 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 0
+  %30 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 0
   store double undef, double* %30, align 8
-  %31 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 1
+  %31 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 2, i32 1
   %32 = fmul double undef, %17
   %33 = fmul double undef, %17
   %34 = fmul double undef, %32
   %35 = fmul double undef, %33
   %36 = fsub double undef, %35
   %37 = fmul double %3, %34
-  %38 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 0
+  %38 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 0
   store double %37, double* %38, align 8
-  %39 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 1
+  %39 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 3, i32 1
   %40 = fmul double undef, %8
   %41 = fmul double undef, %40
   %42 = fmul double undef, %41
   %43 = fsub double undef, %42
   %44 = fmul double %3, %43
-  %45 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 0
+  %45 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 0
   store double %13, double* %45, align 8
-  %46 = getelementptr inbounds [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 1
+  %46 = getelementptr inbounds [5 x { double, double }], [5 x { double, double }]* %c2ten, i64 0, i64 4, i32 1
   %47 = fsub double -0.000000e+00, %14
   store double %47, double* %16, align 8
   store double undef, double* %24, align 8
diff --git a/test/Transforms/BBVectorize/X86/sh-rec.ll b/test/Transforms/BBVectorize/X86/sh-rec.ll
index ad75fc96d9f2..2cb9dbded224 100644
--- a/test/Transforms/BBVectorize/X86/sh-rec.ll
+++ b/test/Transforms/BBVectorize/X86/sh-rec.ll
@@ -8,39 +8,39 @@ entry:
   br i1 undef, label %return, label %if.end10
 
 if.end10:                                         ; preds = %entry
-  %incdec.ptr = getelementptr inbounds i8* %call, i64 undef
+  %incdec.ptr = getelementptr inbounds i8, i8* %call, i64 undef
   %call17 = call i32 @ptou() nounwind
-  %incdec.ptr26.1 = getelementptr inbounds i8* %incdec.ptr, i64 -2
+  %incdec.ptr26.1 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -2
   store i8 undef, i8* %incdec.ptr26.1, align 1
   %div27.1 = udiv i32 %call17, 100
   %rem.2 = urem i32 %div27.1, 10
   %add2230.2 = or i32 %rem.2, 48
   %conv25.2 = trunc i32 %add2230.2 to i8
-  %incdec.ptr26.2 = getelementptr inbounds i8* %incdec.ptr, i64 -3
+  %incdec.ptr26.2 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -3
   store i8 %conv25.2, i8* %incdec.ptr26.2, align 1
-  %incdec.ptr26.3 = getelementptr inbounds i8* %incdec.ptr, i64 -4
+  %incdec.ptr26.3 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -4
   store i8 undef, i8* %incdec.ptr26.3, align 1
   %div27.3 = udiv i32 %call17, 10000
   %rem.4 = urem i32 %div27.3, 10
   %add2230.4 = or i32 %rem.4, 48
   %conv25.4 = trunc i32 %add2230.4 to i8
-  %incdec.ptr26.4 = getelementptr inbounds i8* %incdec.ptr, i64 -5
+  %incdec.ptr26.4 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -5
   store i8 %conv25.4, i8* %incdec.ptr26.4, align 1
   %div27.4 = udiv i32 %call17, 100000
   %rem.5 = urem i32 %div27.4, 10
   %add2230.5 = or i32 %rem.5, 48
   %conv25.5 = trunc i32 %add2230.5 to i8
-  %incdec.ptr26.5 = getelementptr inbounds i8* %incdec.ptr, i64 -6
+  %incdec.ptr26.5 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -6
   store i8 %conv25.5, i8* %incdec.ptr26.5, align 1
-  %incdec.ptr26.6 = getelementptr inbounds i8* %incdec.ptr, i64 -7
+  %incdec.ptr26.6 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -7
   store i8 0, i8* %incdec.ptr26.6, align 1
-  %incdec.ptr26.7 = getelementptr inbounds i8* %incdec.ptr, i64 -8
+  %incdec.ptr26.7 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -8
   store i8 undef, i8* %incdec.ptr26.7, align 1
   %div27.7 = udiv i32 %call17, 100000000
   %rem.8 = urem i32 %div27.7, 10
   %add2230.8 = or i32 %rem.8, 48
   %conv25.8 = trunc i32 %add2230.8 to i8
-  %incdec.ptr26.8 = getelementptr inbounds i8* %incdec.ptr, i64 -9
+  %incdec.ptr26.8 = getelementptr inbounds i8, i8* %incdec.ptr, i64 -9
   store i8 %conv25.8, i8* %incdec.ptr26.8, align 1
   unreachable
 
diff --git a/test/Transforms/BBVectorize/X86/sh-rec2.ll b/test/Transforms/BBVectorize/X86/sh-rec2.ll
index d65ac1cc12f1..d7a004c21384 100644
--- a/test/Transforms/BBVectorize/X86/sh-rec2.ll
+++ b/test/Transforms/BBVectorize/X86/sh-rec2.ll
@@ -7,72 +7,72 @@ target triple = "x86_64-unknown-linux-gnu"
 define void @gsm_encode(%struct.gsm_state.2.8.14.15.16.17.19.22.23.25.26.28.29.31.32.33.35.36.37.38.40.41.42.44.45.47.48.50.52.53.54.56.57.58.59.60.61.62.63.66.73.83.84.89.90.91.92.93.94.95.96.99.100.101.102.103.104.106.107.114.116.121.122.129.130.135.136.137.138.139.140.141.142.143.144.147.148.149.158.159.160.161.164.165.166.167.168.169.172.179.181.182.183.188.195.200.201.202.203.204.205.208.209.210.212.213.214.215.222.223.225.226.230.231.232.233.234.235.236.237.238.239.240.241.242.243.244.352* %s, i16* %source, i8* %c) nounwind uwtable {
 entry:
   %xmc = alloca [52 x i16], align 16
-  %arraydecay5 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 0
+  %arraydecay5 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 0
   call void @Gsm_Coder(%struct.gsm_state.2.8.14.15.16.17.19.22.23.25.26.28.29.31.32.33.35.36.37.38.40.41.42.44.45.47.48.50.52.53.54.56.57.58.59.60.61.62.63.66.73.83.84.89.90.91.92.93.94.95.96.99.100.101.102.103.104.106.107.114.116.121.122.129.130.135.136.137.138.139.140.141.142.143.144.147.148.149.158.159.160.161.164.165.166.167.168.169.172.179.181.182.183.188.195.200.201.202.203.204.205.208.209.210.212.213.214.215.222.223.225.226.230.231.232.233.234.235.236.237.238.239.240.241.242.243.244.352* %s, i16* %source, i16* undef, i16* null, i16* undef, i16* undef, i16* undef, i16* %arraydecay5) nounwind
-  %incdec.ptr136 = getelementptr inbounds i8* %c, i64 10
-  %incdec.ptr157 = getelementptr inbounds i8* %c, i64 11
+  %incdec.ptr136 = getelementptr inbounds i8, i8* %c, i64 10
+  %incdec.ptr157 = getelementptr inbounds i8, i8* %c, i64 11
   store i8 0, i8* %incdec.ptr136, align 1
-  %arrayidx162 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 11
-  %0 = load i16* %arrayidx162, align 2
+  %arrayidx162 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 11
+  %0 = load i16, i16* %arrayidx162, align 2
   %conv1631 = trunc i16 %0 to i8
   %and164 = shl i8 %conv1631, 3
   %shl165 = and i8 %and164, 56
-  %incdec.ptr172 = getelementptr inbounds i8* %c, i64 12
+  %incdec.ptr172 = getelementptr inbounds i8, i8* %c, i64 12
   store i8 %shl165, i8* %incdec.ptr157, align 1
-  %1 = load i16* inttoptr (i64 2 to i16*), align 2
+  %1 = load i16, i16* inttoptr (i64 2 to i16*), align 2
   %conv1742 = trunc i16 %1 to i8
   %and175 = shl i8 %conv1742, 1
-  %incdec.ptr183 = getelementptr inbounds i8* %c, i64 13
+  %incdec.ptr183 = getelementptr inbounds i8, i8* %c, i64 13
   store i8 %and175, i8* %incdec.ptr172, align 1
-  %incdec.ptr199 = getelementptr inbounds i8* %c, i64 14
+  %incdec.ptr199 = getelementptr inbounds i8, i8* %c, i64 14
   store i8 0, i8* %incdec.ptr183, align 1
-  %arrayidx214 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 15
-  %incdec.ptr220 = getelementptr inbounds i8* %c, i64 15
+  %arrayidx214 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 15
+  %incdec.ptr220 = getelementptr inbounds i8, i8* %c, i64 15
   store i8 0, i8* %incdec.ptr199, align 1
-  %2 = load i16* %arrayidx214, align 2
+  %2 = load i16, i16* %arrayidx214, align 2
   %conv2223 = trunc i16 %2 to i8
   %and223 = shl i8 %conv2223, 6
-  %incdec.ptr235 = getelementptr inbounds i8* %c, i64 16
+  %incdec.ptr235 = getelementptr inbounds i8, i8* %c, i64 16
   store i8 %and223, i8* %incdec.ptr220, align 1
-  %arrayidx240 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 19
-  %3 = load i16* %arrayidx240, align 2
+  %arrayidx240 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 19
+  %3 = load i16, i16* %arrayidx240, align 2
   %conv2414 = trunc i16 %3 to i8
   %and242 = shl i8 %conv2414, 2
   %shl243 = and i8 %and242, 28
-  %incdec.ptr251 = getelementptr inbounds i8* %c, i64 17
+  %incdec.ptr251 = getelementptr inbounds i8, i8* %c, i64 17
   store i8 %shl243, i8* %incdec.ptr235, align 1
-  %incdec.ptr272 = getelementptr inbounds i8* %c, i64 18
+  %incdec.ptr272 = getelementptr inbounds i8, i8* %c, i64 18
   store i8 0, i8* %incdec.ptr251, align 1
-  %arrayidx282 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 25
-  %4 = load i16* %arrayidx282, align 2
+  %arrayidx282 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 25
+  %4 = load i16, i16* %arrayidx282, align 2
   %conv2835 = trunc i16 %4 to i8
   %and284 = and i8 %conv2835, 7
-  %incdec.ptr287 = getelementptr inbounds i8* %c, i64 19
+  %incdec.ptr287 = getelementptr inbounds i8, i8* %c, i64 19
   store i8 %and284, i8* %incdec.ptr272, align 1
-  %incdec.ptr298 = getelementptr inbounds i8* %c, i64 20
+  %incdec.ptr298 = getelementptr inbounds i8, i8* %c, i64 20
   store i8 0, i8* %incdec.ptr287, align 1
-  %incdec.ptr314 = getelementptr inbounds i8* %c, i64 21
+  %incdec.ptr314 = getelementptr inbounds i8, i8* %c, i64 21
   store i8 0, i8* %incdec.ptr298, align 1
-  %arrayidx319 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 26
-  %5 = load i16* %arrayidx319, align 4
+  %arrayidx319 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 26
+  %5 = load i16, i16* %arrayidx319, align 4
   %conv3206 = trunc i16 %5 to i8
   %and321 = shl i8 %conv3206, 4
   %shl322 = and i8 %and321, 112
-  %incdec.ptr335 = getelementptr inbounds i8* %c, i64 22
+  %incdec.ptr335 = getelementptr inbounds i8, i8* %c, i64 22
   store i8 %shl322, i8* %incdec.ptr314, align 1
-  %arrayidx340 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 29
-  %6 = load i16* %arrayidx340, align 2
+  %arrayidx340 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 29
+  %6 = load i16, i16* %arrayidx340, align 2
   %conv3417 = trunc i16 %6 to i8
   %and342 = shl i8 %conv3417, 3
   %shl343 = and i8 %and342, 56
-  %incdec.ptr350 = getelementptr inbounds i8* %c, i64 23
+  %incdec.ptr350 = getelementptr inbounds i8, i8* %c, i64 23
   store i8 %shl343, i8* %incdec.ptr335, align 1
-  %incdec.ptr366 = getelementptr inbounds i8* %c, i64 24
+  %incdec.ptr366 = getelementptr inbounds i8, i8* %c, i64 24
   store i8 0, i8* %incdec.ptr350, align 1
-  %arrayidx381 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 36
-  %incdec.ptr387 = getelementptr inbounds i8* %c, i64 25
+  %arrayidx381 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 36
+  %incdec.ptr387 = getelementptr inbounds i8, i8* %c, i64 25
   store i8 0, i8* %incdec.ptr366, align 1
-  %7 = load i16* %arrayidx381, align 8
+  %7 = load i16, i16* %arrayidx381, align 8
   %conv3898 = trunc i16 %7 to i8
   %and390 = shl i8 %conv3898, 6
   store i8 %and390, i8* %incdec.ptr387, align 1
diff --git a/test/Transforms/BBVectorize/X86/sh-rec3.ll b/test/Transforms/BBVectorize/X86/sh-rec3.ll
index ad880ed8895a..2096deb08a90 100644
--- a/test/Transforms/BBVectorize/X86/sh-rec3.ll
+++ b/test/Transforms/BBVectorize/X86/sh-rec3.ll
@@ -7,153 +7,153 @@ target triple = "x86_64-unknown-linux-gnu"
 define void @gsm_encode(%struct.gsm_state.2.8.39.44.45.55.56.57.58.59.62.63.64.65.74.75.76.77.80.87.92.93.94.95.96.97.110.111.112.113.114.128.130.135.136.137.138.139.140.141.142.143.144.145.148.149.150.151.152.169.170.177.178.179.184.185.186.187.188.201.208.209.219.220.221.223.224.225.230.231.232.233.235.236.237.238.245.246.248.249.272.274.279.280.281.282.283.286.293.298.299.314.315.316.317.318.319.320.321.322.323.324.325.326.327.328.329.330.331.332.333.334.335.336.337.338.339.340.341.342.343.344.345.346.347.348.349.350.351.352.353.565* %s, i16* %source, i8* %c) nounwind uwtable {
 entry:
   %LARc28 = alloca [2 x i64], align 16
-  %LARc28.sub = getelementptr inbounds [2 x i64]* %LARc28, i64 0, i64 0
+  %LARc28.sub = getelementptr inbounds [2 x i64], [2 x i64]* %LARc28, i64 0, i64 0
   %tmpcast = bitcast [2 x i64]* %LARc28 to [8 x i16]*
   %Nc = alloca [4 x i16], align 2
   %Mc = alloca [4 x i16], align 2
   %bc = alloca [4 x i16], align 2
   %xmc = alloca [52 x i16], align 16
   %arraydecay = bitcast [2 x i64]* %LARc28 to i16*
-  %arraydecay1 = getelementptr inbounds [4 x i16]* %Nc, i64 0, i64 0
-  %arraydecay2 = getelementptr inbounds [4 x i16]* %bc, i64 0, i64 0
-  %arraydecay3 = getelementptr inbounds [4 x i16]* %Mc, i64 0, i64 0
-  %arraydecay5 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 0
+  %arraydecay1 = getelementptr inbounds [4 x i16], [4 x i16]* %Nc, i64 0, i64 0
+  %arraydecay2 = getelementptr inbounds [4 x i16], [4 x i16]* %bc, i64 0, i64 0
+  %arraydecay3 = getelementptr inbounds [4 x i16], [4 x i16]* %Mc, i64 0, i64 0
+  %arraydecay5 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 0
   call void @Gsm_Coder(%struct.gsm_state.2.8.39.44.45.55.56.57.58.59.62.63.64.65.74.75.76.77.80.87.92.93.94.95.96.97.110.111.112.113.114.128.130.135.136.137.138.139.140.141.142.143.144.145.148.149.150.151.152.169.170.177.178.179.184.185.186.187.188.201.208.209.219.220.221.223.224.225.230.231.232.233.235.236.237.238.245.246.248.249.272.274.279.280.281.282.283.286.293.298.299.314.315.316.317.318.319.320.321.322.323.324.325.326.327.328.329.330.331.332.333.334.335.336.337.338.339.340.341.342.343.344.345.346.347.348.349.350.351.352.353.565* %s, i16* %source, i16* %arraydecay, i16* %arraydecay1, i16* %arraydecay2, i16* %arraydecay3, i16* undef, i16* %arraydecay5) nounwind
-  %0 = load i64* %LARc28.sub, align 16
+  %0 = load i64, i64* %LARc28.sub, align 16
   %1 = trunc i64 %0 to i32
   %conv1 = lshr i32 %1, 2
   %and = and i32 %conv1, 15
   %or = or i32 %and, 208
   %conv6 = trunc i32 %or to i8
-  %incdec.ptr = getelementptr inbounds i8* %c, i64 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %c, i64 1
   store i8 %conv6, i8* %c, align 1
   %conv84 = trunc i64 %0 to i8
   %and9 = shl i8 %conv84, 6
-  %incdec.ptr15 = getelementptr inbounds i8* %c, i64 2
+  %incdec.ptr15 = getelementptr inbounds i8, i8* %c, i64 2
   store i8 %and9, i8* %incdec.ptr, align 1
   %2 = lshr i64 %0, 50
   %shr226.tr = trunc i64 %2 to i8
   %conv25 = and i8 %shr226.tr, 7
-  %incdec.ptr26 = getelementptr inbounds i8* %c, i64 3
+  %incdec.ptr26 = getelementptr inbounds i8, i8* %c, i64 3
   store i8 %conv25, i8* %incdec.ptr15, align 1
-  %incdec.ptr42 = getelementptr inbounds i8* %c, i64 4
+  %incdec.ptr42 = getelementptr inbounds i8, i8* %c, i64 4
   store i8 0, i8* %incdec.ptr26, align 1
-  %arrayidx52 = getelementptr inbounds [8 x i16]* %tmpcast, i64 0, i64 7
-  %3 = load i16* %arrayidx52, align 2
+  %arrayidx52 = getelementptr inbounds [8 x i16], [8 x i16]* %tmpcast, i64 0, i64 7
+  %3 = load i16, i16* %arrayidx52, align 2
   %conv537 = trunc i16 %3 to i8
   %and54 = and i8 %conv537, 7
-  %incdec.ptr57 = getelementptr inbounds i8* %c, i64 5
+  %incdec.ptr57 = getelementptr inbounds i8, i8* %c, i64 5
   store i8 %and54, i8* %incdec.ptr42, align 1
-  %incdec.ptr68 = getelementptr inbounds i8* %c, i64 6
+  %incdec.ptr68 = getelementptr inbounds i8, i8* %c, i64 6
   store i8 0, i8* %incdec.ptr57, align 1
-  %4 = load i16* %arraydecay3, align 2
+  %4 = load i16, i16* %arraydecay3, align 2
   %conv748 = trunc i16 %4 to i8
   %and75 = shl i8 %conv748, 5
   %shl76 = and i8 %and75, 96
-  %incdec.ptr84 = getelementptr inbounds i8* %c, i64 7
+  %incdec.ptr84 = getelementptr inbounds i8, i8* %c, i64 7
   store i8 %shl76, i8* %incdec.ptr68, align 1
-  %arrayidx94 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 1
-  %5 = load i16* %arrayidx94, align 2
+  %arrayidx94 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 1
+  %5 = load i16, i16* %arrayidx94, align 2
   %conv959 = trunc i16 %5 to i8
   %and96 = shl i8 %conv959, 1
   %shl97 = and i8 %and96, 14
   %or103 = or i8 %shl97, 1
-  %incdec.ptr105 = getelementptr inbounds i8* %c, i64 8
+  %incdec.ptr105 = getelementptr inbounds i8, i8* %c, i64 8
   store i8 %or103, i8* %incdec.ptr84, align 1
-  %arrayidx115 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 4
+  %arrayidx115 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 4
   %6 = bitcast i16* %arrayidx115 to i32*
-  %7 = load i32* %6, align 8
+  %7 = load i32, i32* %6, align 8
   %conv11610 = trunc i32 %7 to i8
   %and117 = and i8 %conv11610, 7
-  %incdec.ptr120 = getelementptr inbounds i8* %c, i64 9
+  %incdec.ptr120 = getelementptr inbounds i8, i8* %c, i64 9
   store i8 %and117, i8* %incdec.ptr105, align 1
   %8 = lshr i32 %7, 16
   %and12330 = shl nuw nsw i32 %8, 5
   %and123 = trunc i32 %and12330 to i8
-  %incdec.ptr136 = getelementptr inbounds i8* %c, i64 10
+  %incdec.ptr136 = getelementptr inbounds i8, i8* %c, i64 10
   store i8 %and123, i8* %incdec.ptr120, align 1
-  %incdec.ptr157 = getelementptr inbounds i8* %c, i64 11
+  %incdec.ptr157 = getelementptr inbounds i8, i8* %c, i64 11
   store i8 0, i8* %incdec.ptr136, align 1
-  %incdec.ptr172 = getelementptr inbounds i8* %c, i64 12
+  %incdec.ptr172 = getelementptr inbounds i8, i8* %c, i64 12
   store i8 0, i8* %incdec.ptr157, align 1
-  %arrayidx173 = getelementptr inbounds [4 x i16]* %Nc, i64 0, i64 1
-  %9 = load i16* %arrayidx173, align 2
+  %arrayidx173 = getelementptr inbounds [4 x i16], [4 x i16]* %Nc, i64 0, i64 1
+  %9 = load i16, i16* %arrayidx173, align 2
   %conv17412 = zext i16 %9 to i32
   %and175 = shl nuw nsw i32 %conv17412, 1
-  %arrayidx177 = getelementptr inbounds [4 x i16]* %bc, i64 0, i64 1
-  %10 = load i16* %arrayidx177, align 2
+  %arrayidx177 = getelementptr inbounds [4 x i16], [4 x i16]* %bc, i64 0, i64 1
+  %10 = load i16, i16* %arrayidx177, align 2
   %conv17826 = zext i16 %10 to i32
   %shr17913 = lshr i32 %conv17826, 1
   %and180 = and i32 %shr17913, 1
   %or181 = or i32 %and175, %and180
   %conv182 = trunc i32 %or181 to i8
-  %incdec.ptr183 = getelementptr inbounds i8* %c, i64 13
+  %incdec.ptr183 = getelementptr inbounds i8, i8* %c, i64 13
   store i8 %conv182, i8* %incdec.ptr172, align 1
-  %arrayidx188 = getelementptr inbounds [4 x i16]* %Mc, i64 0, i64 1
-  %11 = load i16* %arrayidx188, align 2
+  %arrayidx188 = getelementptr inbounds [4 x i16], [4 x i16]* %Mc, i64 0, i64 1
+  %11 = load i16, i16* %arrayidx188, align 2
   %conv18914 = trunc i16 %11 to i8
   %and190 = shl i8 %conv18914, 5
   %shl191 = and i8 %and190, 96
-  %incdec.ptr199 = getelementptr inbounds i8* %c, i64 14
+  %incdec.ptr199 = getelementptr inbounds i8, i8* %c, i64 14
   store i8 %shl191, i8* %incdec.ptr183, align 1
-  %arrayidx209 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 14
-  %12 = load i16* %arrayidx209, align 4
+  %arrayidx209 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 14
+  %12 = load i16, i16* %arrayidx209, align 4
   %conv21015 = trunc i16 %12 to i8
   %and211 = shl i8 %conv21015, 1
   %shl212 = and i8 %and211, 14
   %or218 = or i8 %shl212, 1
-  %incdec.ptr220 = getelementptr inbounds i8* %c, i64 15
+  %incdec.ptr220 = getelementptr inbounds i8, i8* %c, i64 15
   store i8 %or218, i8* %incdec.ptr199, align 1
-  %arrayidx225 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 16
+  %arrayidx225 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 16
   %13 = bitcast i16* %arrayidx225 to i64*
-  %14 = load i64* %13, align 16
+  %14 = load i64, i64* %13, align 16
   %conv22616 = trunc i64 %14 to i8
   %and227 = shl i8 %conv22616, 3
   %shl228 = and i8 %and227, 56
-  %incdec.ptr235 = getelementptr inbounds i8* %c, i64 16
+  %incdec.ptr235 = getelementptr inbounds i8, i8* %c, i64 16
   store i8 %shl228, i8* %incdec.ptr220, align 1
   %15 = lshr i64 %14, 32
   %and23832 = shl nuw nsw i64 %15, 5
   %and238 = trunc i64 %and23832 to i8
-  %incdec.ptr251 = getelementptr inbounds i8* %c, i64 17
+  %incdec.ptr251 = getelementptr inbounds i8, i8* %c, i64 17
   store i8 %and238, i8* %incdec.ptr235, align 1
-  %arrayidx266 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 23
-  %incdec.ptr272 = getelementptr inbounds i8* %c, i64 18
+  %arrayidx266 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 23
+  %incdec.ptr272 = getelementptr inbounds i8, i8* %c, i64 18
   store i8 0, i8* %incdec.ptr251, align 1
-  %16 = load i16* %arrayidx266, align 2
+  %16 = load i16, i16* %arrayidx266, align 2
   %conv27418 = trunc i16 %16 to i8
   %and275 = shl i8 %conv27418, 6
-  %incdec.ptr287 = getelementptr inbounds i8* %c, i64 19
+  %incdec.ptr287 = getelementptr inbounds i8, i8* %c, i64 19
   store i8 %and275, i8* %incdec.ptr272, align 1
-  %arrayidx288 = getelementptr inbounds [4 x i16]* %Nc, i64 0, i64 2
-  %17 = load i16* %arrayidx288, align 2
+  %arrayidx288 = getelementptr inbounds [4 x i16], [4 x i16]* %Nc, i64 0, i64 2
+  %17 = load i16, i16* %arrayidx288, align 2
   %conv28919 = zext i16 %17 to i32
   %and290 = shl nuw nsw i32 %conv28919, 1
-  %arrayidx292 = getelementptr inbounds [4 x i16]* %bc, i64 0, i64 2
-  %18 = load i16* %arrayidx292, align 2
+  %arrayidx292 = getelementptr inbounds [4 x i16], [4 x i16]* %bc, i64 0, i64 2
+  %18 = load i16, i16* %arrayidx292, align 2
   %conv29327 = zext i16 %18 to i32
   %shr29420 = lshr i32 %conv29327, 1
   %and295 = and i32 %shr29420, 1
   %or296 = or i32 %and290, %and295
   %conv297 = trunc i32 %or296 to i8
-  %incdec.ptr298 = getelementptr inbounds i8* %c, i64 20
+  %incdec.ptr298 = getelementptr inbounds i8, i8* %c, i64 20
   store i8 %conv297, i8* %incdec.ptr287, align 1
   %conv30021 = trunc i16 %18 to i8
   %and301 = shl i8 %conv30021, 7
-  %incdec.ptr314 = getelementptr inbounds i8* %c, i64 21
+  %incdec.ptr314 = getelementptr inbounds i8, i8* %c, i64 21
   store i8 %and301, i8* %incdec.ptr298, align 1
-  %incdec.ptr335 = getelementptr inbounds i8* %c, i64 22
+  %incdec.ptr335 = getelementptr inbounds i8, i8* %c, i64 22
   store i8 0, i8* %incdec.ptr314, align 1
-  %arrayidx340 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 29
-  %19 = load i16* %arrayidx340, align 2
+  %arrayidx340 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 29
+  %19 = load i16, i16* %arrayidx340, align 2
   %conv34122 = trunc i16 %19 to i8
   %and342 = shl i8 %conv34122, 3
   %shl343 = and i8 %and342, 56
-  %incdec.ptr350 = getelementptr inbounds i8* %c, i64 23
+  %incdec.ptr350 = getelementptr inbounds i8, i8* %c, i64 23
   store i8 %shl343, i8* %incdec.ptr335, align 1
-  %arrayidx355 = getelementptr inbounds [52 x i16]* %xmc, i64 0, i64 32
+  %arrayidx355 = getelementptr inbounds [52 x i16], [52 x i16]* %xmc, i64 0, i64 32
   %20 = bitcast i16* %arrayidx355 to i32*
-  %21 = load i32* %20, align 16
+  %21 = load i32, i32* %20, align 16
   %conv35623 = shl i32 %21, 2
   %shl358 = and i32 %conv35623, 28
   %22 = lshr i32 %21, 17
diff --git a/test/Transforms/BBVectorize/X86/simple-ldstr.ll b/test/Transforms/BBVectorize/X86/simple-ldstr.ll
index 1abbc34b68ec..2c05f30d0818 100644
--- a/test/Transforms/BBVectorize/X86/simple-ldstr.ll
+++ b/test/Transforms/BBVectorize/X86/simple-ldstr.ll
@@ -4,23 +4,23 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; Simple 3-pair chain with loads and stores
 define void @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test1(
 ; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
 ; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
-; CHECK: %i0 = load <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
 ; CHECK: %0 = bitcast double* %c to <2 x double>*
 ; CHECK: store <2 x double> %mul, <2 x double>* %0, align 8
diff --git a/test/Transforms/BBVectorize/X86/wr-aliases.ll b/test/Transforms/BBVectorize/X86/wr-aliases.ll
index 34b1d4e9cae3..56448c0e5471 100644
--- a/test/Transforms/BBVectorize/X86/wr-aliases.ll
+++ b/test/Transforms/BBVectorize/X86/wr-aliases.ll
@@ -27,27 +27,27 @@ arrayctor.cont.ret.exitStub:                      ; preds = %arrayctor.cont
 ; CHECK: <2 x double>
 ; CHECK: @_ZL12printQBezier7QBezier
 ; CHECK: store double %mul8.i, double* %x3.i, align 16
-; CHECK: load double* %x3.i, align 16
+; CHECK: load double, double* %x3.i, align 16
 ; CHECK: ret
 
 arrayctor.cont:                                   ; preds = %newFuncRoot
-  %ref.tmp.sroa.0.0.idx = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 0
+  %ref.tmp.sroa.0.0.idx = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 0
   store double 1.000000e+01, double* %ref.tmp.sroa.0.0.idx, align 16
-  %ref.tmp.sroa.2.0.idx1 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 1
+  %ref.tmp.sroa.2.0.idx1 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 1
   store double 2.000000e+01, double* %ref.tmp.sroa.2.0.idx1, align 8
-  %ref.tmp.sroa.3.0.idx2 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 2
+  %ref.tmp.sroa.3.0.idx2 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 2
   store double 3.000000e+01, double* %ref.tmp.sroa.3.0.idx2, align 16
-  %ref.tmp.sroa.4.0.idx3 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 3
+  %ref.tmp.sroa.4.0.idx3 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 3
   store double 4.000000e+01, double* %ref.tmp.sroa.4.0.idx3, align 8
-  %ref.tmp.sroa.5.0.idx4 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 4
+  %ref.tmp.sroa.5.0.idx4 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 4
   store double 5.000000e+01, double* %ref.tmp.sroa.5.0.idx4, align 16
-  %ref.tmp.sroa.6.0.idx5 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 5
+  %ref.tmp.sroa.6.0.idx5 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 5
   store double 6.000000e+01, double* %ref.tmp.sroa.6.0.idx5, align 8
-  %ref.tmp.sroa.7.0.idx6 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 6
+  %ref.tmp.sroa.7.0.idx6 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 6
   store double 7.000000e+01, double* %ref.tmp.sroa.7.0.idx6, align 16
-  %ref.tmp.sroa.8.0.idx7 = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 7
+  %ref.tmp.sroa.8.0.idx7 = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 7
   store double 8.000000e+01, double* %ref.tmp.sroa.8.0.idx7, align 8
-  %add.ptr = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1
+  %add.ptr = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1
   %v0 = bitcast %class.QBezier.15* %agg.tmp.i to i8*
   call void @llvm.lifetime.start(i64 64, i8* %v0)
   %v1 = bitcast %class.QBezier.15* %agg.tmp55.i to i8*
@@ -57,77 +57,77 @@ arrayctor.cont:                                   ; preds = %newFuncRoot
   %v3 = bitcast [10 x %class.QBezier.15]* %beziers to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %v0, i8* %v3, i64 64, i32 8, i1 false)
   call fastcc void @_ZL12printQBezier7QBezier(%class.QBezier.15* byval align 8 %agg.tmp.i)
-  %x2.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 2
-  %v4 = load double* %x2.i, align 16
-  %x3.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 4
-  %v5 = load double* %x3.i, align 16
+  %x2.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 2
+  %v4 = load double, double* %x2.i, align 16
+  %x3.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 4
+  %v5 = load double, double* %x3.i, align 16
   %add.i = fadd double %v4, %v5
   %mul.i = fmul double 5.000000e-01, %add.i
-  %x1.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 0
-  %v6 = load double* %x1.i, align 16
+  %x1.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 0
+  %v6 = load double, double* %x1.i, align 16
   %add3.i = fadd double %v4, %v6
   %mul4.i = fmul double 5.000000e-01, %add3.i
-  %x25.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 2
+  %x25.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 2
   store double %mul4.i, double* %x25.i, align 16
-  %v7 = load double* %x3.i, align 16
-  %x4.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 6
-  %v8 = load double* %x4.i, align 16
+  %v7 = load double, double* %x3.i, align 16
+  %x4.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 6
+  %v8 = load double, double* %x4.i, align 16
   %add7.i = fadd double %v7, %v8
   %mul8.i = fmul double 5.000000e-01, %add7.i
   store double %mul8.i, double* %x3.i, align 16
-  %v9 = load double* %x1.i, align 16
-  %x111.i = getelementptr inbounds %class.QBezier.15* %add.ptr, i64 0, i32 0
+  %v9 = load double, double* %x1.i, align 16
+  %x111.i = getelementptr inbounds %class.QBezier.15, %class.QBezier.15* %add.ptr, i64 0, i32 0
   store double %v9, double* %x111.i, align 16
-  %v10 = load double* %x25.i, align 16
+  %v10 = load double, double* %x25.i, align 16
   %add15.i = fadd double %mul.i, %v10
   %mul16.i = fmul double 5.000000e-01, %add15.i
-  %x317.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 4
+  %x317.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 4
   store double %mul16.i, double* %x317.i, align 16
-  %v11 = load double* %x3.i, align 16
+  %v11 = load double, double* %x3.i, align 16
   %add19.i = fadd double %mul.i, %v11
   %mul20.i = fmul double 5.000000e-01, %add19.i
   store double %mul20.i, double* %x2.i, align 16
-  %v12 = load double* %x317.i, align 16
+  %v12 = load double, double* %x317.i, align 16
   %add24.i = fadd double %v12, %mul20.i
   %mul25.i = fmul double 5.000000e-01, %add24.i
   store double %mul25.i, double* %x1.i, align 16
-  %x427.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 6
+  %x427.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 6
   store double %mul25.i, double* %x427.i, align 16
-  %y2.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 3
-  %v13 = load double* %y2.i, align 8
-  %y3.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 5
-  %v14 = load double* %y3.i, align 8
+  %y2.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 3
+  %v13 = load double, double* %y2.i, align 8
+  %y3.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 5
+  %v14 = load double, double* %y3.i, align 8
   %add28.i = fadd double %v13, %v14
   %div.i = fmul double 5.000000e-01, %add28.i
-  %y1.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 1
-  %v15 = load double* %y1.i, align 8
+  %y1.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 1
+  %v15 = load double, double* %y1.i, align 8
   %add30.i = fadd double %v13, %v15
   %mul31.i = fmul double 5.000000e-01, %add30.i
-  %y232.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 3
+  %y232.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 3
   store double %mul31.i, double* %y232.i, align 8
-  %v16 = load double* %y3.i, align 8
-  %y4.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 7
-  %v17 = load double* %y4.i, align 8
+  %v16 = load double, double* %y3.i, align 8
+  %y4.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 0, i32 7
+  %v17 = load double, double* %y4.i, align 8
   %add34.i = fadd double %v16, %v17
   %mul35.i = fmul double 5.000000e-01, %add34.i
   store double %mul35.i, double* %y3.i, align 8
-  %v18 = load double* %y1.i, align 8
-  %y138.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 1
+  %v18 = load double, double* %y1.i, align 8
+  %y138.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 1
   store double %v18, double* %y138.i, align 8
-  %v19 = load double* %y232.i, align 8
+  %v19 = load double, double* %y232.i, align 8
   %add42.i = fadd double %div.i, %v19
   %mul43.i = fmul double 5.000000e-01, %add42.i
-  %y344.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 5
+  %y344.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 5
   store double %mul43.i, double* %y344.i, align 8
-  %v20 = load double* %y3.i, align 8
+  %v20 = load double, double* %y3.i, align 8
   %add46.i = fadd double %div.i, %v20
   %mul47.i = fmul double 5.000000e-01, %add46.i
   store double %mul47.i, double* %y2.i, align 8
-  %v21 = load double* %y344.i, align 8
+  %v21 = load double, double* %y344.i, align 8
   %add51.i = fadd double %v21, %mul47.i
   %mul52.i = fmul double 5.000000e-01, %add51.i
   store double %mul52.i, double* %y1.i, align 8
-  %y454.i = getelementptr inbounds [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 7
+  %y454.i = getelementptr inbounds [10 x %class.QBezier.15], [10 x %class.QBezier.15]* %beziers, i64 0, i64 1, i32 7
   store double %mul52.i, double* %y454.i, align 8
   %v22 = bitcast %class.QBezier.15* %add.ptr to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %v1, i8* %v22, i64 64, i32 8, i1 false)
diff --git a/test/Transforms/BBVectorize/func-alias.ll b/test/Transforms/BBVectorize/func-alias.ll
index 9d0cc07c1593..ab72ec0e1991 100644
--- a/test/Transforms/BBVectorize/func-alias.ll
+++ b/test/Transforms/BBVectorize/func-alias.ll
@@ -47,99 +47,99 @@ codeRepl80.exitStub:                              ; preds = %"<bb 34>"
   ret i1 false
 
 "<bb 34>":                                        ; preds = %newFuncRoot
-  %tmp128 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
-  %tmp129 = getelementptr inbounds %struct.__st_parameter_common* %tmp128, i32 0, i32 2
-  store i8* getelementptr inbounds ([11 x i8]* @.cst4, i64 0, i64 0), i8** %tmp129, align 8
-  %tmp130 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
-  %tmp131 = getelementptr inbounds %struct.__st_parameter_common* %tmp130, i32 0, i32 3
+  %tmp128 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
+  %tmp129 = getelementptr inbounds %struct.__st_parameter_common, %struct.__st_parameter_common* %tmp128, i32 0, i32 2
+  store i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.cst4, i64 0, i64 0), i8** %tmp129, align 8
+  %tmp130 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
+  %tmp131 = getelementptr inbounds %struct.__st_parameter_common, %struct.__st_parameter_common* %tmp130, i32 0, i32 3
   store i32 31495, i32* %tmp131, align 4
-  %tmp132 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 5
-  store i8* getelementptr inbounds ([214 x i8]* @.cst823, i64 0, i64 0), i8** %tmp132, align 8
-  %tmp133 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 6
+  %tmp132 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 5
+  store i8* getelementptr inbounds ([214 x i8], [214 x i8]* @.cst823, i64 0, i64 0), i8** %tmp132, align 8
+  %tmp133 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 6
   store i32 214, i32* %tmp133, align 4
-  %tmp134 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
-  %tmp135 = getelementptr inbounds %struct.__st_parameter_common* %tmp134, i32 0, i32 0
+  %tmp134 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
+  %tmp135 = getelementptr inbounds %struct.__st_parameter_common, %struct.__st_parameter_common* %tmp134, i32 0, i32 0
   store i32 4096, i32* %tmp135, align 4
-  %iounit.8748_288 = load i32* @__main1_MOD_iounit, align 4
-  %tmp136 = getelementptr inbounds %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
-  %tmp137 = getelementptr inbounds %struct.__st_parameter_common* %tmp136, i32 0, i32 1
+  %iounit.8748_288 = load i32, i32* @__main1_MOD_iounit, align 4
+  %tmp136 = getelementptr inbounds %struct.__st_parameter_dt, %struct.__st_parameter_dt* %memtmp3, i32 0, i32 0
+  %tmp137 = getelementptr inbounds %struct.__st_parameter_common, %struct.__st_parameter_common* %tmp136, i32 0, i32 1
   store i32 %iounit.8748_288, i32* %tmp137, align 4
   call void @_gfortran_st_write(%struct.__st_parameter_dt* %memtmp3) nounwind
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_integer_write to void (%struct.__st_parameter_dt*, i32*, i32)*)(%struct.__st_parameter_dt* %memtmp3, i32* @j.4580, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_integer_write
-  %D.75807_289 = load i8** getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 0), align 8
-  %j.8758_290 = load i32* @j.4580, align 4
+  %D.75807_289 = load i8*, i8** getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 0), align 8
+  %j.8758_290 = load i32, i32* @j.4580, align 4
   %D.75760_291 = sext i32 %j.8758_290 to i64
-  %iave.8736_292 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_292 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_293 = sext i32 %iave.8736_292 to i64
-  %D.75808_294 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75808_294 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75809_295 = mul nsw i64 %D.75620_293, %D.75808_294
-  %igrp.8737_296 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_296 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_297 = sext i32 %igrp.8737_296 to i64
-  %D.75810_298 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75810_298 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75811_299 = mul nsw i64 %D.75635_297, %D.75810_298
   %D.75812_300 = add nsw i64 %D.75809_295, %D.75811_299
   %D.75813_301 = add nsw i64 %D.75760_291, %D.75812_300
-  %ityp.8750_302 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_302 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_303 = sext i32 %ityp.8750_302 to i64
-  %D.75814_304 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75814_304 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75815_305 = mul nsw i64 %D.75704_303, %D.75814_304
   %D.75816_306 = add nsw i64 %D.75813_301, %D.75815_305
-  %D.75817_307 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 1), align 8
+  %D.75817_307 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 1), align 8
   %D.75818_308 = add nsw i64 %D.75816_306, %D.75817_307
   %tmp138 = bitcast i8* %D.75807_289 to [0 x float]*
   %tmp139 = bitcast [0 x float]* %tmp138 to float*
-  %D.75819_309 = getelementptr inbounds float* %tmp139, i64 %D.75818_308
+  %D.75819_309 = getelementptr inbounds float, float* %tmp139, i64 %D.75818_308
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_real_write to void (%struct.__st_parameter_dt*, float*, i32)*)(%struct.__st_parameter_dt* %memtmp3, float* %D.75819_309, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_real_write
-  %D.75820_310 = load i8** getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 0), align 8
-  %j.8758_311 = load i32* @j.4580, align 4
+  %D.75820_310 = load i8*, i8** getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 0), align 8
+  %j.8758_311 = load i32, i32* @j.4580, align 4
   %D.75760_312 = sext i32 %j.8758_311 to i64
-  %iave.8736_313 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_313 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_314 = sext i32 %iave.8736_313 to i64
-  %D.75821_315 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75821_315 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75822_316 = mul nsw i64 %D.75620_314, %D.75821_315
-  %igrp.8737_317 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_317 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_318 = sext i32 %igrp.8737_317 to i64
-  %D.75823_319 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75823_319 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75824_320 = mul nsw i64 %D.75635_318, %D.75823_319
   %D.75825_321 = add nsw i64 %D.75822_316, %D.75824_320
   %D.75826_322 = add nsw i64 %D.75760_312, %D.75825_321
-  %ityp.8750_323 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_323 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_324 = sext i32 %ityp.8750_323 to i64
-  %D.75827_325 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75827_325 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75828_326 = mul nsw i64 %D.75704_324, %D.75827_325
   %D.75829_327 = add nsw i64 %D.75826_322, %D.75828_326
-  %D.75830_328 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 1), align 8
+  %D.75830_328 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 1), align 8
   %D.75831_329 = add nsw i64 %D.75829_327, %D.75830_328
   %tmp140 = bitcast i8* %D.75820_310 to [0 x [1 x i8]]*
   %tmp141 = bitcast [0 x [1 x i8]]* %tmp140 to [1 x i8]*
-  %D.75832_330 = getelementptr inbounds [1 x i8]* %tmp141, i64 %D.75831_329
+  %D.75832_330 = getelementptr inbounds [1 x i8], [1 x i8]* %tmp141, i64 %D.75831_329
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_character_write to void (%struct.__st_parameter_dt*, [1 x i8]*, i32)*)(%struct.__st_parameter_dt* %memtmp3, [1 x i8]* %D.75832_330, i32 1) nounwind
 ; CHECK: @_gfortran_transfer_character_write
-  %D.75833_331 = load i8** getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 0), align 8
-  %j.8758_332 = load i32* @j.4580, align 4
+  %D.75833_331 = load i8*, i8** getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 0), align 8
+  %j.8758_332 = load i32, i32* @j.4580, align 4
   %D.75760_333 = sext i32 %j.8758_332 to i64
-  %iave.8736_334 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_334 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_335 = sext i32 %iave.8736_334 to i64
-  %D.75834_336 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75834_336 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75835_337 = mul nsw i64 %D.75620_335, %D.75834_336
-  %igrp.8737_338 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_338 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_339 = sext i32 %igrp.8737_338 to i64
-  %D.75836_340 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75836_340 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75837_341 = mul nsw i64 %D.75635_339, %D.75836_340
   %D.75838_342 = add nsw i64 %D.75835_337, %D.75837_341
   %D.75839_343 = add nsw i64 %D.75760_333, %D.75838_342
-  %ityp.8750_344 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_344 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_345 = sext i32 %ityp.8750_344 to i64
-  %D.75840_346 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75840_346 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75841_347 = mul nsw i64 %D.75704_345, %D.75840_346
   %D.75842_348 = add nsw i64 %D.75839_343, %D.75841_347
-  %D.75843_349 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 1), align 8
+  %D.75843_349 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 1), align 8
   %D.75844_350 = add nsw i64 %D.75842_348, %D.75843_349
   %tmp142 = bitcast i8* %D.75833_331 to [0 x i32]*
   %tmp143 = bitcast [0 x i32]* %tmp142 to i32*
-  %D.75845_351 = getelementptr inbounds i32* %tmp143, i64 %D.75844_350
+  %D.75845_351 = getelementptr inbounds i32, i32* %tmp143, i64 %D.75844_350
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_integer_write to void (%struct.__st_parameter_dt*, i32*, i32)*)(%struct.__st_parameter_dt* %memtmp3, i32* %D.75845_351, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_integer_write
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_real_write to void (%struct.__st_parameter_dt*, float*, i32)*)(%struct.__st_parameter_dt* %memtmp3, float* @xr1.4592, i32 4) nounwind
@@ -150,79 +150,79 @@ codeRepl80.exitStub:                              ; preds = %"<bb 34>"
 ; CHECK: @_gfortran_transfer_character_write
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_integer_write to void (%struct.__st_parameter_dt*, i32*, i32)*)(%struct.__st_parameter_dt* %memtmp3, i32* @j1.4581, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_integer_write
-  %D.75807_352 = load i8** getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 0), align 8
-  %j1.8760_353 = load i32* @j1.4581, align 4
+  %D.75807_352 = load i8*, i8** getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 0), align 8
+  %j1.8760_353 = load i32, i32* @j1.4581, align 4
   %D.75773_354 = sext i32 %j1.8760_353 to i64
-  %iave.8736_355 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_355 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_356 = sext i32 %iave.8736_355 to i64
-  %D.75808_357 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75808_357 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75809_358 = mul nsw i64 %D.75620_356, %D.75808_357
-  %igrp.8737_359 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_359 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_360 = sext i32 %igrp.8737_359 to i64
-  %D.75810_361 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75810_361 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75811_362 = mul nsw i64 %D.75635_360, %D.75810_361
   %D.75812_363 = add nsw i64 %D.75809_358, %D.75811_362
   %D.75846_364 = add nsw i64 %D.75773_354, %D.75812_363
-  %ityp.8750_365 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_365 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_366 = sext i32 %ityp.8750_365 to i64
-  %D.75814_367 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75814_367 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75815_368 = mul nsw i64 %D.75704_366, %D.75814_367
   %D.75847_369 = add nsw i64 %D.75846_364, %D.75815_368
-  %D.75817_370 = load i64* getelementptr inbounds (%"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 1), align 8
+  %D.75817_370 = load i64, i64* getelementptr inbounds (%"struct.array4_real(kind=4)", %"struct.array4_real(kind=4)"* @__main1_MOD_rmxval, i64 0, i32 1), align 8
   %D.75848_371 = add nsw i64 %D.75847_369, %D.75817_370
   %tmp144 = bitcast i8* %D.75807_352 to [0 x float]*
   %tmp145 = bitcast [0 x float]* %tmp144 to float*
-  %D.75849_372 = getelementptr inbounds float* %tmp145, i64 %D.75848_371
+  %D.75849_372 = getelementptr inbounds float, float* %tmp145, i64 %D.75848_371
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_real_write to void (%struct.__st_parameter_dt*, float*, i32)*)(%struct.__st_parameter_dt* %memtmp3, float* %D.75849_372, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_real_write
-  %D.75820_373 = load i8** getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 0), align 8
-  %j1.8760_374 = load i32* @j1.4581, align 4
+  %D.75820_373 = load i8*, i8** getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 0), align 8
+  %j1.8760_374 = load i32, i32* @j1.4581, align 4
   %D.75773_375 = sext i32 %j1.8760_374 to i64
-  %iave.8736_376 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_376 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_377 = sext i32 %iave.8736_376 to i64
-  %D.75821_378 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75821_378 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75822_379 = mul nsw i64 %D.75620_377, %D.75821_378
-  %igrp.8737_380 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_380 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_381 = sext i32 %igrp.8737_380 to i64
-  %D.75823_382 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75823_382 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75824_383 = mul nsw i64 %D.75635_381, %D.75823_382
   %D.75825_384 = add nsw i64 %D.75822_379, %D.75824_383
   %D.75850_385 = add nsw i64 %D.75773_375, %D.75825_384
-  %ityp.8750_386 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_386 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_387 = sext i32 %ityp.8750_386 to i64
-  %D.75827_388 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75827_388 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75828_389 = mul nsw i64 %D.75704_387, %D.75827_388
   %D.75851_390 = add nsw i64 %D.75850_385, %D.75828_389
-  %D.75830_391 = load i64* getelementptr inbounds (%struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 1), align 8
+  %D.75830_391 = load i64, i64* getelementptr inbounds (%struct.array4_unknown, %struct.array4_unknown* @__main1_MOD_mclmsg, i64 0, i32 1), align 8
   %D.75852_392 = add nsw i64 %D.75851_390, %D.75830_391
   %tmp146 = bitcast i8* %D.75820_373 to [0 x [1 x i8]]*
   %tmp147 = bitcast [0 x [1 x i8]]* %tmp146 to [1 x i8]*
-  %D.75853_393 = getelementptr inbounds [1 x i8]* %tmp147, i64 %D.75852_392
+  %D.75853_393 = getelementptr inbounds [1 x i8], [1 x i8]* %tmp147, i64 %D.75852_392
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_character_write to void (%struct.__st_parameter_dt*, [1 x i8]*, i32)*)(%struct.__st_parameter_dt* %memtmp3, [1 x i8]* %D.75853_393, i32 1) nounwind
 ; CHECK: @_gfortran_transfer_character_write
-  %D.75833_394 = load i8** getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 0), align 8
-  %j1.8760_395 = load i32* @j1.4581, align 4
+  %D.75833_394 = load i8*, i8** getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 0), align 8
+  %j1.8760_395 = load i32, i32* @j1.4581, align 4
   %D.75773_396 = sext i32 %j1.8760_395 to i64
-  %iave.8736_397 = load i32* @__main1_MOD_iave, align 4
+  %iave.8736_397 = load i32, i32* @__main1_MOD_iave, align 4
   %D.75620_398 = sext i32 %iave.8736_397 to i64
-  %D.75834_399 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 2, i32 0), align 8
+  %D.75834_399 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 2, i32 0), align 8
   %D.75835_400 = mul nsw i64 %D.75620_398, %D.75834_399
-  %igrp.8737_401 = load i32* @__main1_MOD_igrp, align 4
+  %igrp.8737_401 = load i32, i32* @__main1_MOD_igrp, align 4
   %D.75635_402 = sext i32 %igrp.8737_401 to i64
-  %D.75836_403 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 1, i32 0), align 8
+  %D.75836_403 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 1, i32 0), align 8
   %D.75837_404 = mul nsw i64 %D.75635_402, %D.75836_403
   %D.75838_405 = add nsw i64 %D.75835_400, %D.75837_404
   %D.75854_406 = add nsw i64 %D.75773_396, %D.75838_405
-  %ityp.8750_407 = load i32* @__main1_MOD_ityp, align 4
+  %ityp.8750_407 = load i32, i32* @__main1_MOD_ityp, align 4
   %D.75704_408 = sext i32 %ityp.8750_407 to i64
-  %D.75840_409 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 3, i32 0), align 8
+  %D.75840_409 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 3, i64 3, i32 0), align 8
   %D.75841_410 = mul nsw i64 %D.75704_408, %D.75840_409
   %D.75855_411 = add nsw i64 %D.75854_406, %D.75841_410
-  %D.75843_412 = load i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 1), align 8
+  %D.75843_412 = load i64, i64* getelementptr inbounds (%"struct.array4_integer(kind=4).73", %"struct.array4_integer(kind=4).73"* @__main1_MOD_mxdate, i64 0, i32 1), align 8
   %D.75856_413 = add nsw i64 %D.75855_411, %D.75843_412
   %tmp148 = bitcast i8* %D.75833_394 to [0 x i32]*
   %tmp149 = bitcast [0 x i32]* %tmp148 to i32*
-  %D.75857_414 = getelementptr inbounds i32* %tmp149, i64 %D.75856_413
+  %D.75857_414 = getelementptr inbounds i32, i32* %tmp149, i64 %D.75856_413
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_integer_write to void (%struct.__st_parameter_dt*, i32*, i32)*)(%struct.__st_parameter_dt* %memtmp3, i32* %D.75857_414, i32 4) nounwind
 ; CHECK: @_gfortran_transfer_integer_write
   call void bitcast (void (%struct.__st_parameter_dt*, i8*, i32)* @_gfortran_transfer_real_write to void (%struct.__st_parameter_dt*, float*, i32)*)(%struct.__st_parameter_dt* %memtmp3, float* @xr2.4593, i32 4) nounwind
@@ -233,9 +233,9 @@ codeRepl80.exitStub:                              ; preds = %"<bb 34>"
 ; CHECK: @_gfortran_transfer_character_write
   call void @_gfortran_st_write_done(%struct.__st_parameter_dt* %memtmp3) nounwind
 ; CHECK: @_gfortran_st_write_done
-  %j.8758_415 = load i32* @j.4580, align 4
+  %j.8758_415 = load i32, i32* @j.4580, align 4
   %D.4634_416 = icmp eq i32 %j.8758_415, %D.4627_188.reload
-  %j.8758_417 = load i32* @j.4580, align 4
+  %j.8758_417 = load i32, i32* @j.4580, align 4
   %j.8770_418 = add nsw i32 %j.8758_417, 1
   store i32 %j.8770_418, i32* @j.4580, align 4
   %tmp150 = icmp ne i1 %D.4634_416, false
diff --git a/test/Transforms/BBVectorize/ld1.ll b/test/Transforms/BBVectorize/ld1.ll
index 9c79eef05f7b..368c38aa5ce7 100644
--- a/test/Transforms/BBVectorize/ld1.ll
+++ b/test/Transforms/BBVectorize/ld1.ll
@@ -3,18 +3,18 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 define double @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %i2 = load double* %c, align 8
+  %i2 = load double, double* %c, align 8
   %add = fadd double %mul, %i2
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
-  %arrayidx6 = getelementptr inbounds double* %c, i64 1
-  %i5 = load double* %arrayidx6, align 8
+  %arrayidx6 = getelementptr inbounds double, double* %c, i64 1
+  %i5 = load double, double* %arrayidx6, align 8
   %add7 = fadd double %mul5, %i5
   %mul9 = fmul double %add, %i1
   %add11 = fadd double %mul9, %i2
@@ -26,10 +26,10 @@ entry:
 ; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
 ; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
 ; CHECK: %i2.v.i0 = bitcast double* %c to <2 x double>*
-; CHECK: %i0 = load <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
-; CHECK: %i2 = load <2 x double>* %i2.v.i0, align 8
+; CHECK: %i2 = load <2 x double>, <2 x double>* %i2.v.i0, align 8
 ; CHECK: %add = fadd <2 x double> %mul, %i2
 ; CHECK: %mul9 = fmul <2 x double> %add, %i1
 ; CHECK: %add11 = fadd <2 x double> %mul9, %i2
diff --git a/test/Transforms/BBVectorize/loop1.ll b/test/Transforms/BBVectorize/loop1.ll
index ca361703adcb..70a5def42228 100644
--- a/test/Transforms/BBVectorize/loop1.ll
+++ b/test/Transforms/BBVectorize/loop1.ll
@@ -12,10 +12,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double* %in1, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx2 = getelementptr inbounds double* %in2, i64 %indvars.iv
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx = getelementptr inbounds double, double* %in1, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %in2, i64 %indvars.iv
+  %1 = load double, double* %arrayidx2, align 8
   %mul = fmul double %0, %0
   %mul3 = fmul double %0, %1
   %add = fadd double %mul, %mul3
@@ -28,17 +28,17 @@ for.body:                                         ; preds = %for.body, %entry
   %add10 = fadd double %add9, %0
   %mul11 = fmul double %mul8, %add10
   %add12 = fadd double %add7, %mul11
-  %arrayidx14 = getelementptr inbounds double* %out, i64 %indvars.iv
+  %arrayidx14 = getelementptr inbounds double, double* %out, i64 %indvars.iv
   store double %add12, double* %arrayidx14, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, 10
   br i1 %exitcond, label %for.end, label %for.body
 ; CHECK: %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-; CHECK: %arrayidx = getelementptr inbounds double* %in1, i64 %indvars.iv
-; CHECK: %0 = load double* %arrayidx, align 8
-; CHECK: %arrayidx2 = getelementptr inbounds double* %in2, i64 %indvars.iv
-; CHECK: %1 = load double* %arrayidx2, align 8
+; CHECK: %arrayidx = getelementptr inbounds double, double* %in1, i64 %indvars.iv
+; CHECK: %0 = load double, double* %arrayidx, align 8
+; CHECK: %arrayidx2 = getelementptr inbounds double, double* %in2, i64 %indvars.iv
+; CHECK: %1 = load double, double* %arrayidx2, align 8
 ; CHECK: %mul = fmul double %0, %0
 ; CHECK: %mul3 = fmul double %0, %1
 ; CHECK: %add = fadd double %mul, %mul3
@@ -55,20 +55,20 @@ for.body:                                         ; preds = %for.body, %entry
 ; CHECK: %mul6.v.r2 = extractelement <2 x double> %mul6, i32 1
 ; CHECK: %add7 = fadd double %add, %mul6.v.r1
 ; CHECK: %add12 = fadd double %add7, %mul6.v.r2
-; CHECK: %arrayidx14 = getelementptr inbounds double* %out, i64 %indvars.iv
+; CHECK: %arrayidx14 = getelementptr inbounds double, double* %out, i64 %indvars.iv
 ; CHECK: store double %add12, double* %arrayidx14, align 8
 ; CHECK: %indvars.iv.next = add i64 %indvars.iv, 1
 ; CHECK: %lftr.wideiv = trunc i64 %indvars.iv.next to i32
 ; CHECK: %exitcond = icmp eq i32 %lftr.wideiv, 10
 ; CHECK: br i1 %exitcond, label %for.end, label %for.body
 ; CHECK-UNRL: %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next.1, %for.body ]
-; CHECK-UNRL: %arrayidx = getelementptr inbounds double* %in1, i64 %indvars.iv
+; CHECK-UNRL: %arrayidx = getelementptr inbounds double, double* %in1, i64 %indvars.iv
 ; CHECK-UNRL: %0 = bitcast double* %arrayidx to <2 x double>*
-; CHECK-UNRL: %arrayidx2 = getelementptr inbounds double* %in2, i64 %indvars.iv
+; CHECK-UNRL: %arrayidx2 = getelementptr inbounds double, double* %in2, i64 %indvars.iv
 ; CHECK-UNRL: %1 = bitcast double* %arrayidx2 to <2 x double>*
-; CHECK-UNRL: %arrayidx14 = getelementptr inbounds double* %out, i64 %indvars.iv
-; CHECK-UNRL: %2 = load <2 x double>* %0, align 8
-; CHECK-UNRL: %3 = load <2 x double>* %1, align 8
+; CHECK-UNRL: %arrayidx14 = getelementptr inbounds double, double* %out, i64 %indvars.iv
+; CHECK-UNRL: %2 = load <2 x double>, <2 x double>* %0, align 8
+; CHECK-UNRL: %3 = load <2 x double>, <2 x double>* %1, align 8
 ; CHECK-UNRL: %mul = fmul <2 x double> %2, %2
 ; CHECK-UNRL: %mul3 = fmul <2 x double> %2, %3
 ; CHECK-UNRL: %add = fadd <2 x double> %mul, %mul3
diff --git a/test/Transforms/BBVectorize/mem-op-depth.ll b/test/Transforms/BBVectorize/mem-op-depth.ll
index c31d4521183f..732043b7f8ec 100644
--- a/test/Transforms/BBVectorize/mem-op-depth.ll
+++ b/test/Transforms/BBVectorize/mem-op-depth.ll
@@ -7,15 +7,15 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define i32 @test1() nounwind {
 ; CHECK-LABEL: @test1(
-  %V1 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 0), align 16
-  %V2 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 1), align 4
-  %V3= load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 2), align 8
-  %V4 = load float* getelementptr inbounds ([1024 x float]* @A, i64 0, i64 3), align 4
-; CHECK:   %V1 = load <4 x float>* bitcast ([1024 x float]* @A to <4 x float>*), align 16
-  store float %V1, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 0), align 16
-  store float %V2, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 1), align 4
-  store float %V3, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 2), align 8
-  store float %V4, float* getelementptr inbounds ([1024 x float]* @B, i64 0, i64 3), align 4
+  %V1 = load float, float* getelementptr inbounds ([1024 x float], [1024 x float]* @A, i64 0, i64 0), align 16
+  %V2 = load float, float* getelementptr inbounds ([1024 x float], [1024 x float]* @A, i64 0, i64 1), align 4
+  %V3= load float, float* getelementptr inbounds ([1024 x float], [1024 x float]* @A, i64 0, i64 2), align 8
+  %V4 = load float, float* getelementptr inbounds ([1024 x float], [1024 x float]* @A, i64 0, i64 3), align 4
+; CHECK:   %V1 = load <4 x float>, <4 x float>* bitcast ([1024 x float]* @A to <4 x float>*), align 16
+  store float %V1, float* getelementptr inbounds ([1024 x float], [1024 x float]* @B, i64 0, i64 0), align 16
+  store float %V2, float* getelementptr inbounds ([1024 x float], [1024 x float]* @B, i64 0, i64 1), align 4
+  store float %V3, float* getelementptr inbounds ([1024 x float], [1024 x float]* @B, i64 0, i64 2), align 8
+  store float %V4, float* getelementptr inbounds ([1024 x float], [1024 x float]* @B, i64 0, i64 3), align 4
 ; CHECK-NEXT: store <4 x float> %V1, <4 x float>* bitcast ([1024 x float]* @B to <4 x float>*), align 16
   ret i32 0
 ; CHECK-NEXT: ret i32 0
diff --git a/test/Transforms/BBVectorize/metadata.ll b/test/Transforms/BBVectorize/metadata.ll
index 874fbb87967c..f5580a888616 100644
--- a/test/Transforms/BBVectorize/metadata.ll
+++ b/test/Transforms/BBVectorize/metadata.ll
@@ -4,16 +4,16 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; Simple 3-pair chain with loads and stores (with fpmath)
 define void @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1, !fpmath !2
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4, !fpmath !3
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test1(
@@ -24,16 +24,16 @@ entry:
 ; Simple 3-pair chain with loads and stores (ints with range)
 define void @test2(i64* %a, i64* %b, i64* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load i64* %a, align 8, !range !0
-  %i1 = load i64* %b, align 8
+  %i0 = load i64, i64* %a, align 8, !range !0
+  %i1 = load i64, i64* %b, align 8
   %mul = mul i64 %i0, %i1
-  %arrayidx3 = getelementptr inbounds i64* %a, i64 1
-  %i3 = load i64* %arrayidx3, align 8, !range !1
-  %arrayidx4 = getelementptr inbounds i64* %b, i64 1
-  %i4 = load i64* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds i64, i64* %a, i64 1
+  %i3 = load i64, i64* %arrayidx3, align 8, !range !1
+  %arrayidx4 = getelementptr inbounds i64, i64* %b, i64 1
+  %i4 = load i64, i64* %arrayidx4, align 8
   %mul5 = mul i64 %i3, %i4
   store i64 %mul, i64* %c, align 8
-  %arrayidx5 = getelementptr inbounds i64* %c, i64 1
+  %arrayidx5 = getelementptr inbounds i64, i64* %c, i64 1
   store i64 %mul5, i64* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test2(
diff --git a/test/Transforms/BBVectorize/no-ldstr-conn.ll b/test/Transforms/BBVectorize/no-ldstr-conn.ll
index bcc5ce7531bd..a84cd6585602 100644
--- a/test/Transforms/BBVectorize/no-ldstr-conn.ll
+++ b/test/Transforms/BBVectorize/no-ldstr-conn.ll
@@ -7,10 +7,10 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define i64 @test2(i64 %a) nounwind uwtable readonly {
 entry:
   %a1 = inttoptr i64 %a to i64*
-  %a2 = getelementptr i64* %a1, i64 1
-  %a3 = getelementptr i64* %a1, i64 2
-  %v2 = load i64* %a2, align 8
-  %v3 = load i64* %a3, align 8
+  %a2 = getelementptr i64, i64* %a1, i64 1
+  %a3 = getelementptr i64, i64* %a1, i64 2
+  %v2 = load i64, i64* %a2, align 8
+  %v3 = load i64, i64* %a3, align 8
   %v2a = add i64 %v2, 5
   %v3a = add i64 %v3, 7
   store i64 %v2a, i64* %a2, align 8
@@ -18,6 +18,6 @@ entry:
   %r = add i64 %v2, %v3
   ret i64 %r
 ; CHECK-LABEL: @test2(
-; CHECK-NOT: getelementptr <2 x i64*>
+; CHECK-NOT: getelementptr i64, <2 x i64*>
 }
 
diff --git a/test/Transforms/BBVectorize/simple-ldstr-ptrs.ll b/test/Transforms/BBVectorize/simple-ldstr-ptrs.ll
index 4d2298c1a11d..fcc0236bae9d 100644
--- a/test/Transforms/BBVectorize/simple-ldstr-ptrs.ll
+++ b/test/Transforms/BBVectorize/simple-ldstr-ptrs.ll
@@ -8,37 +8,37 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; Simple 3-pair chain also with loads and stores (using ptrs and gep)
 define double @test1(i64* %a, i64* %b, i64* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load i64* %a, align 8
-  %i1 = load i64* %b, align 8
+  %i0 = load i64, i64* %a, align 8
+  %i1 = load i64, i64* %b, align 8
   %mul = mul i64 %i0, %i1
-  %arrayidx3 = getelementptr inbounds i64* %a, i64 1
-  %i3 = load i64* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds i64* %b, i64 1
-  %i4 = load i64* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds i64, i64* %a, i64 1
+  %i3 = load i64, i64* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds i64, i64* %b, i64 1
+  %i4 = load i64, i64* %arrayidx4, align 8
   %mul5 = mul i64 %i3, %i4
   %ptr = inttoptr i64 %mul to double*
   %ptr5 = inttoptr i64 %mul5 to double*
-  %aptr = getelementptr inbounds double* %ptr, i64 2
-  %aptr5 = getelementptr inbounds double* %ptr5, i64 3
-  %av = load double* %aptr, align 16
-  %av5 = load double* %aptr5, align 16
+  %aptr = getelementptr inbounds double, double* %ptr, i64 2
+  %aptr5 = getelementptr inbounds double, double* %ptr5, i64 3
+  %av = load double, double* %aptr, align 16
+  %av5 = load double, double* %aptr5, align 16
   %r = fmul double %av, %av5
   store i64 %mul, i64* %c, align 8
-  %arrayidx5 = getelementptr inbounds i64* %c, i64 1
+  %arrayidx5 = getelementptr inbounds i64, i64* %c, i64 1
   store i64 %mul5, i64* %arrayidx5, align 8
   ret double %r
 ; CHECK-LABEL: @test1(
 ; CHECK: %i0.v.i0 = bitcast i64* %a to <2 x i64>*
 ; CHECK: %i1.v.i0 = bitcast i64* %b to <2 x i64>*
-; CHECK: %i0 = load <2 x i64>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x i64>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x i64>, <2 x i64>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x i64>, <2 x i64>* %i1.v.i0, align 8
 ; CHECK: %mul = mul <2 x i64> %i0, %i1
 ; CHECK: %ptr = inttoptr <2 x i64> %mul to <2 x double*>
-; CHECK: %aptr = getelementptr inbounds <2 x double*> %ptr, <2 x i64> <i64 2, i64 3>
+; CHECK: %aptr = getelementptr inbounds double, <2 x double*> %ptr, <2 x i64> <i64 2, i64 3>
 ; CHECK: %aptr.v.r1 = extractelement <2 x double*> %aptr, i32 0
 ; CHECK: %aptr.v.r2 = extractelement <2 x double*> %aptr, i32 1
-; CHECK: %av = load double* %aptr.v.r1, align 16
-; CHECK: %av5 = load double* %aptr.v.r2, align 16
+; CHECK: %av = load double, double* %aptr.v.r1, align 16
+; CHECK: %av5 = load double, double* %aptr.v.r2, align 16
 ; CHECK: %r = fmul double %av, %av5
 ; CHECK: %0 = bitcast i64* %c to <2 x i64>*
 ; CHECK: store <2 x i64> %mul, <2 x i64>* %0, align 8
@@ -50,31 +50,31 @@ entry:
 ; Simple 3-pair chain with loads and stores (using ptrs and gep)
 define void @test2(i64** %a, i64** %b, i64** %c) nounwind uwtable readonly {
 entry:
-  %i0 = load i64** %a, align 8
-  %i1 = load i64** %b, align 8
-  %arrayidx3 = getelementptr inbounds i64** %a, i64 1
-  %i3 = load i64** %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds i64** %b, i64 1
-  %i4 = load i64** %arrayidx4, align 8
-  %o1 = load i64* %i1, align 8
-  %o4 = load i64* %i4, align 8
-  %ptr0 = getelementptr inbounds i64* %i0, i64 %o1
-  %ptr3 = getelementptr inbounds i64* %i3, i64 %o4
+  %i0 = load i64*, i64** %a, align 8
+  %i1 = load i64*, i64** %b, align 8
+  %arrayidx3 = getelementptr inbounds i64*, i64** %a, i64 1
+  %i3 = load i64*, i64** %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds i64*, i64** %b, i64 1
+  %i4 = load i64*, i64** %arrayidx4, align 8
+  %o1 = load i64, i64* %i1, align 8
+  %o4 = load i64, i64* %i4, align 8
+  %ptr0 = getelementptr inbounds i64, i64* %i0, i64 %o1
+  %ptr3 = getelementptr inbounds i64, i64* %i3, i64 %o4
   store i64* %ptr0, i64** %c, align 8
-  %arrayidx5 = getelementptr inbounds i64** %c, i64 1
+  %arrayidx5 = getelementptr inbounds i64*, i64** %c, i64 1
   store i64* %ptr3, i64** %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test2(
 ; CHECK: %i0.v.i0 = bitcast i64** %a to <2 x i64*>*
-; CHECK: %i1 = load i64** %b, align 8
-; CHECK: %i0 = load <2 x i64*>* %i0.v.i0, align 8
-; CHECK: %arrayidx4 = getelementptr inbounds i64** %b, i64 1
-; CHECK: %i4 = load i64** %arrayidx4, align 8
-; CHECK: %o1 = load i64* %i1, align 8
-; CHECK: %o4 = load i64* %i4, align 8
+; CHECK: %i1 = load i64*, i64** %b, align 8
+; CHECK: %i0 = load <2 x i64*>, <2 x i64*>* %i0.v.i0, align 8
+; CHECK: %arrayidx4 = getelementptr inbounds i64*, i64** %b, i64 1
+; CHECK: %i4 = load i64*, i64** %arrayidx4, align 8
+; CHECK: %o1 = load i64, i64* %i1, align 8
+; CHECK: %o4 = load i64, i64* %i4, align 8
 ; CHECK: %ptr0.v.i1.1 = insertelement <2 x i64> undef, i64 %o1, i32 0
 ; CHECK: %ptr0.v.i1.2 = insertelement <2 x i64> %ptr0.v.i1.1, i64 %o4, i32 1
-; CHECK: %ptr0 = getelementptr inbounds <2 x i64*> %i0, <2 x i64> %ptr0.v.i1.2
+; CHECK: %ptr0 = getelementptr inbounds i64, <2 x i64*> %i0, <2 x i64> %ptr0.v.i1.2
 ; CHECK: %0 = bitcast i64** %c to <2 x i64*>*
 ; CHECK: store <2 x i64*> %ptr0, <2 x i64*>* %0, align 8
 ; CHECK: ret void
@@ -86,42 +86,42 @@ entry:
 ; using pointer vectors.
 define void @test3(<2 x i64*>* %a, <2 x i64*>* %b, <2 x i64*>* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load <2 x i64*>* %a, align 8
-  %i1 = load <2 x i64*>* %b, align 8
-  %arrayidx3 = getelementptr inbounds <2 x i64*>* %a, i64 1
-  %i3 = load <2 x i64*>* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds <2 x i64*>* %b, i64 1
-  %i4 = load <2 x i64*>* %arrayidx4, align 8
+  %i0 = load <2 x i64*>, <2 x i64*>* %a, align 8
+  %i1 = load <2 x i64*>, <2 x i64*>* %b, align 8
+  %arrayidx3 = getelementptr inbounds <2 x i64*>, <2 x i64*>* %a, i64 1
+  %i3 = load <2 x i64*>, <2 x i64*>* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds <2 x i64*>, <2 x i64*>* %b, i64 1
+  %i4 = load <2 x i64*>, <2 x i64*>* %arrayidx4, align 8
   %j1 = extractelement <2 x i64*> %i1, i32 0
   %j4 = extractelement <2 x i64*> %i4, i32 0
-  %o1 = load i64* %j1, align 8
-  %o4 = load i64* %j4, align 8
+  %o1 = load i64, i64* %j1, align 8
+  %o4 = load i64, i64* %j4, align 8
   %j0 = extractelement <2 x i64*> %i0, i32 0
   %j3 = extractelement <2 x i64*> %i3, i32 0
-  %ptr0 = getelementptr inbounds i64* %j0, i64 %o1
-  %ptr3 = getelementptr inbounds i64* %j3, i64 %o4
+  %ptr0 = getelementptr inbounds i64, i64* %j0, i64 %o1
+  %ptr3 = getelementptr inbounds i64, i64* %j3, i64 %o4
   %qtr0 = insertelement <2 x i64*> undef, i64* %ptr0, i32 0
   %rtr0 = insertelement <2 x i64*> %qtr0, i64* %ptr0, i32 1
   %qtr3 = insertelement <2 x i64*> undef, i64* %ptr3, i32 0
   %rtr3 = insertelement <2 x i64*> %qtr3, i64* %ptr3, i32 1
   store <2 x i64*> %rtr0, <2 x i64*>* %c, align 8
-  %arrayidx5 = getelementptr inbounds <2 x i64*>* %c, i64 1
+  %arrayidx5 = getelementptr inbounds <2 x i64*>, <2 x i64*>* %c, i64 1
   store <2 x i64*> %rtr3, <2 x i64*>* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test3(
 ; CHECK: %i0.v.i0 = bitcast <2 x i64*>* %a to <4 x i64*>*
-; CHECK: %i1 = load <2 x i64*>* %b, align 8
-; CHECK: %i0 = load <4 x i64*>* %i0.v.i0, align 8
-; CHECK: %arrayidx4 = getelementptr inbounds <2 x i64*>* %b, i64 1
-; CHECK: %i4 = load <2 x i64*>* %arrayidx4, align 8
+; CHECK: %i1 = load <2 x i64*>, <2 x i64*>* %b, align 8
+; CHECK: %i0 = load <4 x i64*>, <4 x i64*>* %i0.v.i0, align 8
+; CHECK: %arrayidx4 = getelementptr inbounds <2 x i64*>, <2 x i64*>* %b, i64 1
+; CHECK: %i4 = load <2 x i64*>, <2 x i64*>* %arrayidx4, align 8
 ; CHECK: %j1 = extractelement <2 x i64*> %i1, i32 0
 ; CHECK: %j4 = extractelement <2 x i64*> %i4, i32 0
-; CHECK: %o1 = load i64* %j1, align 8
-; CHECK: %o4 = load i64* %j4, align 8
+; CHECK: %o1 = load i64, i64* %j1, align 8
+; CHECK: %o4 = load i64, i64* %j4, align 8
 ; CHECK: %ptr0.v.i1.1 = insertelement <2 x i64> undef, i64 %o1, i32 0
 ; CHECK: %ptr0.v.i1.2 = insertelement <2 x i64> %ptr0.v.i1.1, i64 %o4, i32 1
 ; CHECK: %ptr0.v.i0 = shufflevector <4 x i64*> %i0, <4 x i64*> undef, <2 x i32> <i32 0, i32 2>
-; CHECK: %ptr0 = getelementptr inbounds <2 x i64*> %ptr0.v.i0, <2 x i64> %ptr0.v.i1.2
+; CHECK: %ptr0 = getelementptr inbounds i64, <2 x i64*> %ptr0.v.i0, <2 x i64> %ptr0.v.i1.2
 ; CHECK: %rtr0 = shufflevector <2 x i64*> %ptr0, <2 x i64*> undef, <2 x i32> zeroinitializer
 ; CHECK: %rtr3 = shufflevector <2 x i64*> %ptr0, <2 x i64*> undef, <2 x i32> <i32 1, i32 1>
 ; CHECK: %0 = bitcast <2 x i64*>* %c to <4 x i64*>*
diff --git a/test/Transforms/BBVectorize/simple-ldstr.ll b/test/Transforms/BBVectorize/simple-ldstr.ll
index 558f8b38d794..56c1a06b42ea 100644
--- a/test/Transforms/BBVectorize/simple-ldstr.ll
+++ b/test/Transforms/BBVectorize/simple-ldstr.ll
@@ -5,23 +5,23 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; Simple 3-pair chain with loads and stores
 define void @test1(double* %a, double* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test1(
 ; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
 ; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
-; CHECK: %i0 = load <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
 ; CHECK: %0 = bitcast double* %c to <2 x double>*
 ; CHECK: store <2 x double> %mul, <2 x double>* %0, align 8
@@ -33,28 +33,28 @@ entry:
 ; Simple chain with extending loads and stores
 define void @test2(float* %a, float* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0f = load float* %a, align 4
+  %i0f = load float, float* %a, align 4
   %i0 = fpext float %i0f to double
-  %i1f = load float* %b, align 4
+  %i1f = load float, float* %b, align 4
   %i1 = fpext float %i1f to double
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds float* %a, i64 1
-  %i3f = load float* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %a, i64 1
+  %i3f = load float, float* %arrayidx3, align 4
   %i3 = fpext float %i3f to double
-  %arrayidx4 = getelementptr inbounds float* %b, i64 1
-  %i4f = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %b, i64 1
+  %i4f = load float, float* %arrayidx4, align 4
   %i4 = fpext float %i4f to double
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 ; CHECK-LABEL: @test2(
 ; CHECK: %i0f.v.i0 = bitcast float* %a to <2 x float>*
 ; CHECK: %i1f.v.i0 = bitcast float* %b to <2 x float>*
-; CHECK: %i0f = load <2 x float>* %i0f.v.i0, align 4
+; CHECK: %i0f = load <2 x float>, <2 x float>* %i0f.v.i0, align 4
 ; CHECK: %i0 = fpext <2 x float> %i0f to <2 x double>
-; CHECK: %i1f = load <2 x float>* %i1f.v.i0, align 4
+; CHECK: %i1f = load <2 x float>, <2 x float>* %i1f.v.i0, align 4
 ; CHECK: %i1 = fpext <2 x float> %i1f to <2 x double>
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
 ; CHECK: %0 = bitcast double* %c to <2 x double>*
@@ -67,37 +67,37 @@ entry:
 ; Simple chain with loads and truncating stores
 define void @test3(double* %a, double* %b, float* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %mulf = fptrunc double %mul to float
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %mul5f = fptrunc double %mul5 to float
   store float %mulf, float* %c, align 8
-  %arrayidx5 = getelementptr inbounds float* %c, i64 1
+  %arrayidx5 = getelementptr inbounds float, float* %c, i64 1
   store float %mul5f, float* %arrayidx5, align 4
   ret void
 ; CHECK-LABEL: @test3(
 ; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
 ; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
-; CHECK: %i0 = load <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
 ; CHECK: %mulf = fptrunc <2 x double> %mul to <2 x float>
 ; CHECK: %0 = bitcast float* %c to <2 x float>*
 ; CHECK: store <2 x float> %mulf, <2 x float>* %0, align 8
 ; CHECK: ret void
 ; CHECK-AO-LABEL: @test3(
-; CHECK-AO: %i0 = load double* %a, align 8
-; CHECK-AO: %i1 = load double* %b, align 8
-; CHECK-AO: %arrayidx3 = getelementptr inbounds double* %a, i64 1
-; CHECK-AO: %i3 = load double* %arrayidx3, align 8
-; CHECK-AO: %arrayidx4 = getelementptr inbounds double* %b, i64 1
-; CHECK-AO: %i4 = load double* %arrayidx4, align 8
+; CHECK-AO: %i0 = load double, double* %a, align 8
+; CHECK-AO: %i1 = load double, double* %b, align 8
+; CHECK-AO: %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+; CHECK-AO: %i3 = load double, double* %arrayidx3, align 8
+; CHECK-AO: %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+; CHECK-AO: %i4 = load double, double* %arrayidx4, align 8
 ; CHECK-AO: %mul.v.i1.1 = insertelement <2 x double> undef, double %i1, i32 0
 ; CHECK-AO: %mul.v.i1.2 = insertelement <2 x double> %mul.v.i1.1, double %i4, i32 1
 ; CHECK-AO: %mul.v.i0.1 = insertelement <2 x double> undef, double %i0, i32 0
@@ -119,16 +119,16 @@ if.then1:
   br label %if.then
 
 if.then:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   br label %if.end
 
@@ -143,23 +143,23 @@ if.end:
 ; Simple 3-pair chain with loads and stores
 define void @test5(double* %a, double* %b, double* %c) nounwind uwtable readonly {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   store double %mul, double* %c, align 4
   ret void
 ; CHECK-LABEL: @test5(
 ; CHECK: %i0.v.i0 = bitcast double* %a to <2 x double>*
 ; CHECK: %i1.v.i0 = bitcast double* %b to <2 x double>*
-; CHECK: %i0 = load <2 x double>* %i0.v.i0, align 8
-; CHECK: %i1 = load <2 x double>* %i1.v.i0, align 8
+; CHECK: %i0 = load <2 x double>, <2 x double>* %i0.v.i0, align 8
+; CHECK: %i1 = load <2 x double>, <2 x double>* %i1.v.i0, align 8
 ; CHECK: %mul = fmul <2 x double> %i0, %i1
 ; CHECK: %0 = bitcast double* %c to <2 x double>*
 ; CHECK: store <2 x double> %mul, <2 x double>* %0, align 4
diff --git a/test/Transforms/BBVectorize/simple3.ll b/test/Transforms/BBVectorize/simple3.ll
index 6edf7f07ac1d..da7f94149414 100644
--- a/test/Transforms/BBVectorize/simple3.ll
+++ b/test/Transforms/BBVectorize/simple3.ll
@@ -4,12 +4,12 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; Basic depth-3 chain
 define double @test1(double %A1, double %A2, double %A3, double %B1, double %B2, double %B3) {
 ; CHECK-LABEL: @test1(
-; CHECK: %X1.v.i1.11 = insertelement <3 x double> undef, double %B1, i32 0
-; CHECK: %X1.v.i1.22 = insertelement <3 x double> %X1.v.i1.11, double %B2, i32 1
-; CHECK: %X1.v.i1 = insertelement <3 x double> %X1.v.i1.22, double %B3, i32 2
-; CHECK: %X1.v.i0.13 = insertelement <3 x double> undef, double %A1, i32 0
-; CHECK: %X1.v.i0.24 = insertelement <3 x double> %X1.v.i0.13, double %A2, i32 1
-; CHECK: %X1.v.i0 = insertelement <3 x double> %X1.v.i0.24, double %A3, i32 2
+; CHECK: %X1.v.i1.1.1 = insertelement <3 x double> undef, double %B1, i32 0
+; CHECK: %X1.v.i1.2.2 = insertelement <3 x double> %X1.v.i1.1.1, double %B2, i32 1
+; CHECK: %X1.v.i1 = insertelement <3 x double> %X1.v.i1.2.2, double %B3, i32 2
+; CHECK: %X1.v.i0.1.3 = insertelement <3 x double> undef, double %A1, i32 0
+; CHECK: %X1.v.i0.2.4 = insertelement <3 x double> %X1.v.i0.1.3, double %A2, i32 1
+; CHECK: %X1.v.i0 = insertelement <3 x double> %X1.v.i0.2.4, double %A3, i32 2
 	%X1 = fsub double %A1, %B1
 	%X2 = fsub double %A2, %B2
 	%X3 = fsub double %A3, %B3
@@ -24,11 +24,11 @@ define double @test1(double %A1, double %A2, double %A3, double %B1, double %B2,
 ; CHECK: %Z1 = fadd <3 x double> %Y1, %X1.v.i1
         %R1 = fmul double %Z1, %Z2
 	%R  = fmul double %R1, %Z3
-; CHECK: %Z1.v.r210 = extractelement <3 x double> %Z1, i32 2
+; CHECK: %Z1.v.r2.10 = extractelement <3 x double> %Z1, i32 2
 ; CHECK: %Z1.v.r1 = extractelement <3 x double> %Z1, i32 0
 ; CHECK: %Z1.v.r2 = extractelement <3 x double> %Z1, i32 1
 ; CHECK: %R1 = fmul double %Z1.v.r1, %Z1.v.r2
-; CHECK: %R = fmul double %R1, %Z1.v.r210
+; CHECK: %R = fmul double %R1, %Z1.v.r2.10
 	ret double %R
 ; CHECK: ret double %R
 }
diff --git a/test/Transforms/BDCE/basic.ll b/test/Transforms/BDCE/basic.ll
new file mode 100644
index 000000000000..6e748c69a16e
--- /dev/null
+++ b/test/Transforms/BDCE/basic.ll
@@ -0,0 +1,348 @@
+; RUN: opt -S -bdce -instsimplify < %s | FileCheck %s
+; RUN: opt -S -instsimplify < %s | FileCheck %s -check-prefix=CHECK-IO
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+; Function Attrs: nounwind readnone
+define signext i32 @bar(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 4
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 8
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %shr = ashr i32 %or15, 4
+  ret i32 %shr
+
+; CHECK-LABEL: @bar
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+
+; Check that instsimplify is not doing this all on its own.
+; CHECK-IO-LABEL: @bar
+; CHECK-IO: tail call signext i32 @foo(i32 signext 5)
+; CHECK-IO: tail call signext i32 @foo(i32 signext 3)
+; CHECK-IO: tail call signext i32 @foo(i32 signext 2)
+; CHECK-IO: tail call signext i32 @foo(i32 signext 1)
+; CHECK-IO: tail call signext i32 @foo(i32 signext 0)
+; CHECK-IO: tail call signext i32 @foo(i32 signext 4)
+; CHECK-IO: ret i32
+}
+
+; Function Attrs: nounwind readnone
+declare signext i32 @foo(i32 signext) #0
+
+; Function Attrs: nounwind readnone
+define signext i32 @far(i32 signext %x) #1 {
+entry:
+  %call = tail call signext i32 @goo(i32 signext 5) #1
+  %and = and i32 %call, 4
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @goo(i32 signext 3) #1
+  %and2 = and i32 %call1, 8
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @goo(i32 signext 2) #1
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @goo(i32 signext 1) #1
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @goo(i32 signext 0) #1
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @goo(i32 signext 4) #1
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %shr = ashr i32 %or15, 4
+  ret i32 %shr
+
+; CHECK-LABEL: @far
+; Calls to foo(5) and foo(3) are still there, but their results are not used.
+; CHECK: tail call signext i32 @goo(i32 signext 5)
+; CHECK-NEXT: tail call signext i32 @goo(i32 signext 3)
+; CHECK-NEXT: tail call signext i32 @goo(i32 signext 2)
+; CHECK: tail call signext i32 @goo(i32 signext 1)
+; CHECK: tail call signext i32 @goo(i32 signext 0)
+; CHECK: tail call signext i32 @goo(i32 signext 4)
+; CHECK: ret i32
+
+; Check that instsimplify is not doing this all on its own.
+; CHECK-IO-LABEL: @far
+; CHECK-IO: tail call signext i32 @goo(i32 signext 5)
+; CHECK-IO: tail call signext i32 @goo(i32 signext 3)
+; CHECK-IO: tail call signext i32 @goo(i32 signext 2)
+; CHECK-IO: tail call signext i32 @goo(i32 signext 1)
+; CHECK-IO: tail call signext i32 @goo(i32 signext 0)
+; CHECK-IO: tail call signext i32 @goo(i32 signext 4)
+; CHECK-IO: ret i32
+}
+
+declare signext i32 @goo(i32 signext) #1
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar1(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %bs = tail call i32 @llvm.bswap.i32(i32 %or15) #0
+  %shr = ashr i32 %bs, 4
+  ret i32 %shr
+
+; CHECK-LABEL: @tar1
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+declare i32 @llvm.bswap.i32(i32) #0
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar2(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %shl = shl i32 %or15, 10
+  ret i32 %shl
+
+; CHECK-LABEL: @tar2
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar3(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %add = add i32 %or15, 5
+  %shl = shl i32 %add, 10
+  ret i32 %shl
+
+; CHECK-LABEL: @tar3
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar4(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %sub = sub i32 %or15, 5
+  %shl = shl i32 %sub, 10
+  ret i32 %shl
+
+; CHECK-LABEL: @tar4
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar5(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %xor = xor i32 %or15, 5
+  %shl = shl i32 %xor, 10
+  ret i32 %shl
+
+; CHECK-LABEL: @tar5
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+define signext i32 @tar7(i32 signext %x, i1 %b) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %v = select i1 %b, i32 %or15, i32 5
+  %shl = shl i32 %v, 10
+  ret i32 %shl
+
+; CHECK-LABEL: @tar7
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i32
+}
+
+; Function Attrs: nounwind readnone
+define signext i16 @tar8(i32 signext %x) #0 {
+entry:
+  %call = tail call signext i32 @foo(i32 signext 5) #0
+  %and = and i32 %call, 33554432
+  %or = or i32 %and, %x
+  %call1 = tail call signext i32 @foo(i32 signext 3) #0
+  %and2 = and i32 %call1, 67108864
+  %or3 = or i32 %or, %and2
+  %call4 = tail call signext i32 @foo(i32 signext 2) #0
+  %and5 = and i32 %call4, 16
+  %or6 = or i32 %or3, %and5
+  %call7 = tail call signext i32 @foo(i32 signext 1) #0
+  %and8 = and i32 %call7, 32
+  %or9 = or i32 %or6, %and8
+  %call10 = tail call signext i32 @foo(i32 signext 0) #0
+  %and11 = and i32 %call10, 64
+  %or12 = or i32 %or9, %and11
+  %call13 = tail call signext i32 @foo(i32 signext 4) #0
+  %and14 = and i32 %call13, 128
+  %or15 = or i32 %or12, %and14
+  %tr = trunc i32 %or15 to i16
+  ret i16 %tr
+
+; CHECK-LABEL: @tar8
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 5)
+; CHECK-NOT: tail call signext i32 @foo(i32 signext 3)
+; CHECK: tail call signext i32 @foo(i32 signext 2)
+; CHECK: tail call signext i32 @foo(i32 signext 1)
+; CHECK: tail call signext i32 @foo(i32 signext 0)
+; CHECK: tail call signext i32 @foo(i32 signext 4)
+; CHECK: ret i16
+}
+
+attributes #0 = { nounwind readnone }
+attributes #1 = { nounwind }
+
diff --git a/test/Transforms/BDCE/dce-pure.ll b/test/Transforms/BDCE/dce-pure.ll
new file mode 100644
index 000000000000..6a432fcc42d7
--- /dev/null
+++ b/test/Transforms/BDCE/dce-pure.ll
@@ -0,0 +1,33 @@
+; RUN: opt -bdce -S < %s | FileCheck %s
+
+declare i32 @strlen(i8*) readonly nounwind
+
+define void @test1() {
+  call i32 @strlen( i8* null )
+  ret void
+
+; CHECK-LABEL: @test1
+; CHECK-NOT: call
+; CHECK: ret void
+}
+
+define i32 @test2() {
+  ; invoke of pure function should not be deleted!
+  invoke i32 @strlen( i8* null ) readnone
+                  to label %Cont unwind label %Other
+
+Cont:           ; preds = %0
+  ret i32 0
+
+Other:          ; preds = %0
+   %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+            cleanup
+  ret i32 1
+
+; CHECK-LABEL: @test2
+; CHECK: invoke
+; CHECK: ret i32 1
+}
+
+declare i32 @__gxx_personality_v0(...)
+
diff --git a/test/Transforms/BDCE/order.ll b/test/Transforms/BDCE/order.ll
new file mode 100644
index 000000000000..301f4476ab5b
--- /dev/null
+++ b/test/Transforms/BDCE/order.ll
@@ -0,0 +1,37 @@
+; RUN: opt -bdce -S < %s | FileCheck %s
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+declare i32 @__gxx_personality_v0(...)
+
+define fastcc void @_ZN11__sanitizerL12TestRegistryEPNS_14ThreadRegistryEb() #0 {
+entry:
+  br i1 undef, label %if.else, label %entry.if.end_crit_edge
+
+if.else:
+  ret void
+
+invoke.cont70:
+  store i32 %call71, i32* undef, align 4
+  br label %if.else
+
+; CHECK-LABEL: @_ZN11__sanitizerL12TestRegistryEPNS_14ThreadRegistryEb
+; CHECK: store i32 %call71
+
+lpad65.loopexit.split-lp.loopexit.split-lp:
+  br label %if.else
+
+lpad65.loopexit.split-lp.loopexit.split-lp.loopexit:
+  %lpad.loopexit1121 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
+          cleanup
+  br label %lpad65.loopexit.split-lp.loopexit.split-lp
+
+entry.if.end_crit_edge:
+  %call71 = invoke i32 @_ZN11__sanitizer14ThreadRegistry12CreateThreadEmbjPv()
+          to label %invoke.cont70 unwind label %lpad65.loopexit.split-lp.loopexit.split-lp.loopexit
+}
+
+declare i32 @_ZN11__sanitizer14ThreadRegistry12CreateThreadEmbjPv()
+
+attributes #0 = { uwtable }
+
diff --git a/test/Transforms/CodeExtractor/2004-03-14-DominanceProblem.ll b/test/Transforms/CodeExtractor/2004-03-14-DominanceProblem.ll
index a6ee63ec45ae..2f9c0c7371c5 100644
--- a/test/Transforms/CodeExtractor/2004-03-14-DominanceProblem.ll
+++ b/test/Transforms/CodeExtractor/2004-03-14-DominanceProblem.ll
@@ -13,7 +13,7 @@ loopentry.1:            ; preds = %no_exit.1, %loopentry.1.preheader
         br i1 false, label %no_exit.1, label %loopexit.0.loopexit1
 
 no_exit.1:              ; preds = %loopentry.1
-        %tmp.53 = load i32* null                ; <i32> [#uses=1]
+        %tmp.53 = load i32, i32* null                ; <i32> [#uses=1]
         br i1 false, label %shortcirc_next.2, label %loopentry.1
 
 shortcirc_next.2:               ; preds = %no_exit.1
diff --git a/test/Transforms/CodeGenPrepare/R600/lit.local.cfg b/test/Transforms/CodeGenPrepare/R600/lit.local.cfg
new file mode 100644
index 000000000000..4086e8d681c3
--- /dev/null
+++ b/test/Transforms/CodeGenPrepare/R600/lit.local.cfg
@@ -0,0 +1,3 @@
+if not 'R600' in config.root.targets:
+    config.unsupported = True
+
diff --git a/test/Transforms/CodeGenPrepare/R600/no-sink-addrspacecast.ll b/test/Transforms/CodeGenPrepare/R600/no-sink-addrspacecast.ll
new file mode 100644
index 000000000000..f6f898fae21b
--- /dev/null
+++ b/test/Transforms/CodeGenPrepare/R600/no-sink-addrspacecast.ll
@@ -0,0 +1,49 @@
+; RUN: opt -S -codegenprepare -mtriple=amdgcn-unknown-unknown < %s | FileCheck -check-prefix=ASC -check-prefix=COMMON %s
+
+; COMMON-LABEL: @test_sink_ptrtoint_asc(
+; ASC: addrspacecast
+; ASC-NOT: ptrtoint
+; ASC-NOT: inttoptr
+
+define void @test_sink_ptrtoint_asc(float addrspace(1)* nocapture %arg, float addrspace(1)* nocapture readonly %arg1, float addrspace(3)* %arg2) #0 {
+bb:
+  %tmp = getelementptr inbounds float, float addrspace(3)* %arg2, i32 16
+  %tmp2 = tail call i32 @llvm.r600.read.tidig.x() #1
+  %tmp3 = sext i32 %tmp2 to i64
+  %tmp4 = getelementptr inbounds float, float addrspace(1)* %arg1, i64 %tmp3
+  %tmp5 = load float, float addrspace(1)* %tmp4, align 4
+  %tmp6 = addrspacecast float addrspace(3)* %tmp to float addrspace(4)*
+  %tmp7 = fcmp olt float %tmp5, 8.388608e+06
+  br i1 %tmp7, label %bb8, label %bb14
+
+bb8:                                              ; preds = %bb
+  %tmp9 = tail call float @llvm.fma.f32(float %tmp5, float 0x3FE45F3060000000, float 5.000000e-01) #1
+  %tmp10 = fmul float %tmp9, 0x3E74442D00000000
+  %tmp11 = fsub float -0.000000e+00, %tmp10
+  %tmp12 = tail call float @llvm.fma.f32(float %tmp9, float 0x3E74442D00000000, float %tmp11) #1
+  store float %tmp12, float addrspace(4)* %tmp6, align 4
+  %tmp13 = fsub float -0.000000e+00, %tmp12
+  br label %bb15
+
+bb14:                                             ; preds = %bb
+  store float 2.000000e+00, float addrspace(4)* %tmp6, align 4
+  br label %bb15
+
+bb15:                                             ; preds = %bb14, %bb8
+  %tmp16 = phi float [ 0.000000e+00, %bb14 ], [ %tmp13, %bb8 ]
+  %tmp17 = fsub float -0.000000e+00, %tmp16
+  %tmp18 = tail call float @llvm.fma.f32(float 1.000000e+00, float 0x3FF0AAAAA0000000, float %tmp17) #1
+  %tmp19 = fsub float 2.187500e-01, %tmp18
+  %tmp20 = fsub float 7.187500e-01, %tmp19
+  %tmp21 = fcmp ogt float %tmp5, 1.600000e+01
+  %tmp22 = select i1 %tmp21, float 0x7FF8000000000000, float %tmp20
+  %tmp23 = getelementptr inbounds float, float addrspace(1)* %arg, i64 %tmp3
+  store float %tmp22, float addrspace(1)* %tmp23, align 4
+  ret void
+}
+
+declare float @llvm.fma.f32(float, float, float) #1
+declare i32 @llvm.r600.read.tidig.x() #1
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readnone }
diff --git a/test/Transforms/CodeGenPrepare/X86/extend-sink-hoist.ll b/test/Transforms/CodeGenPrepare/X86/extend-sink-hoist.ll
index 430b99299d82..519e1ee2ce6f 100644
--- a/test/Transforms/CodeGenPrepare/X86/extend-sink-hoist.ll
+++ b/test/Transforms/CodeGenPrepare/X86/extend-sink-hoist.ll
@@ -9,7 +9,7 @@ define i128 @sink(i64* %mem1, i64* %mem2) {
 ; CHECK-LABEL: block1:
 ; CHECK-NEXT: load
 block1:
-  %l1 = load i64* %mem1
+  %l1 = load i64, i64* %mem1
   %s1 = sext i64 %l1 to i128
   br label %block2
 
@@ -18,7 +18,7 @@ block1:
 ; CHECK-NEXT: load
 ; CHECK-NEXT: sext
 block2:
-  %l2 = load i64* %mem2
+  %l2 = load i64, i64* %mem2
   %s2 = sext i64 %l2 to i128
   %res = mul i128 %s1, %s2
   ret i128 %res
@@ -31,7 +31,7 @@ define i64 @hoist(i32* %mem1, i32* %mem2) {
 ; CHECK-NEXT: load
 ; CHECK-NEXT: sext
 block1:
-  %l1 = load i32* %mem1
+  %l1 = load i32, i32* %mem1
   br label %block2
 
 ; CHECK-LABEL: block2:
@@ -39,7 +39,7 @@ block1:
 ; CHECK-NEXT: sext
 block2:
   %s1 = sext i32 %l1 to i64
-  %l2 = load i32* %mem2
+  %l2 = load i32, i32* %mem2
   %s2 = sext i32 %l2 to i64
   %res = mul i64 %s1, %s2
   ret i64 %res
diff --git a/test/Transforms/CodeGenPrepare/X86/memset_chk-simplify-nobuiltin.ll b/test/Transforms/CodeGenPrepare/X86/memset_chk-simplify-nobuiltin.ll
new file mode 100644
index 000000000000..1e12c0192a95
--- /dev/null
+++ b/test/Transforms/CodeGenPrepare/X86/memset_chk-simplify-nobuiltin.ll
@@ -0,0 +1,18 @@
+; RUN: opt -S -disable-simplify-libcalls -codegenprepare < %s | FileCheck %s
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+; This is a workaround for PR23093: when building with -mkernel/-fno-builtin,
+; we still generate fortified library calls.
+
+; Check that we ignore two things:
+; - attribute nobuiltin
+; - TLI::has (always returns false thanks to -disable-simplify-libcalls)
+
+; CHECK-NOT: _chk
+; CHECK: call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %len, i32 1, i1 false)
+define void @test_nobuiltin(i8* %dst, i64 %len) {
+  call i8* @__memset_chk(i8* %dst, i32 0, i64 %len, i64 -1) nobuiltin
+  ret void
+}
+
+declare i8* @__memset_chk(i8*, i32, i64, i64)
diff --git a/test/Transforms/CodeGenPrepare/X86/sink-addrspacecast.ll b/test/Transforms/CodeGenPrepare/X86/sink-addrspacecast.ll
index a985c36707a9..c9f49b5d4f86 100644
--- a/test/Transforms/CodeGenPrepare/X86/sink-addrspacecast.ll
+++ b/test/Transforms/CodeGenPrepare/X86/sink-addrspacecast.ll
@@ -8,12 +8,12 @@ target triple = "x86_64-unknown-linux-gnu"
 ; CHECK: add i64 %sunkaddr, 40
 define void @load_cast_gep(i1 %cond, i64* %base) {
 entry:
-  %addr = getelementptr inbounds i64* %base, i64 5
+  %addr = getelementptr inbounds i64, i64* %base, i64 5
   %casted = addrspacecast i64* %addr to i32 addrspace(1)*
   br i1 %cond, label %if.then, label %fallthrough
 
 if.then:
-  %v = load i32 addrspace(1)* %casted, align 4
+  %v = load i32, i32 addrspace(1)* %casted, align 4
   br label %fallthrough
 
 fallthrough:
@@ -25,7 +25,7 @@ fallthrough:
 define void @store_gep_cast(i1 %cond, i64* %base) {
 entry:
   %casted = addrspacecast i64* %base to i32 addrspace(1)*
-  %addr = getelementptr inbounds i32 addrspace(1)* %casted, i64 5
+  %addr = getelementptr inbounds i32, i32 addrspace(1)* %casted, i64 5
   br i1 %cond, label %if.then, label %fallthrough
 
 if.then:
diff --git a/test/Transforms/CodeGenPrepare/overflow-intrinsics.ll b/test/Transforms/CodeGenPrepare/overflow-intrinsics.ll
new file mode 100644
index 000000000000..1e413191e064
--- /dev/null
+++ b/test/Transforms/CodeGenPrepare/overflow-intrinsics.ll
@@ -0,0 +1,74 @@
+; RUN: opt -codegenprepare -S < %s | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+target triple = "x86_64-apple-darwin10.0.0"
+
+; CHECK-LABEL: @test1(
+; CHECK: llvm.uadd.with.overflow
+; CHECK: ret i64
+define i64 @test1(i64 %a, i64 %b) nounwind ssp {
+entry:
+  %add = add i64 %b, %a
+  %cmp = icmp ult i64 %add, %a
+  %Q = select i1 %cmp, i64 %b, i64 42
+  ret i64 %Q
+}
+
+; CHECK-LABEL: @test2(
+; CHECK: llvm.uadd.with.overflow
+; CHECK: ret i64
+define i64 @test2(i64 %a, i64 %b) nounwind ssp {
+entry:
+  %add = add i64 %b, %a
+  %cmp = icmp ult i64 %add, %b
+  %Q = select i1 %cmp, i64 %b, i64 42
+  ret i64 %Q
+}
+
+; CHECK-LABEL: @test3(
+; CHECK: llvm.uadd.with.overflow
+; CHECK: ret i64
+define i64 @test3(i64 %a, i64 %b) nounwind ssp {
+entry:
+  %add = add i64 %b, %a
+  %cmp = icmp ugt i64 %b, %add
+  %Q = select i1 %cmp, i64 %b, i64 42
+  ret i64 %Q
+}
+
+; CHECK-LABEL: @test4(
+; CHECK: llvm.uadd.with.overflow
+; CHECK: extractvalue
+; CHECK: extractvalue
+; CHECK: select
+define i64 @test4(i64 %a, i64 %b, i1 %c) nounwind ssp {
+entry:
+  %add = add i64 %b, %a
+  %cmp = icmp ugt i64 %b, %add
+  br i1 %c, label %next, label %exit
+
+ next:
+  %Q = select i1 %cmp, i64 %b, i64 42
+  ret i64 %Q
+
+ exit:
+  ret i64 0
+}
+
+; CHECK-LABEL: @test5(
+; CHECK-NOT: llvm.uadd.with.overflow
+; CHECK: next
+define i64 @test5(i64 %a, i64 %b, i64* %ptr, i1 %c) nounwind ssp {
+entry:
+  %add = add i64 %b, %a
+  store i64 %add, i64* %ptr
+  %cmp = icmp ugt i64 %b, %add
+  br i1 %c, label %next, label %exit
+
+ next:
+  %Q = select i1 %cmp, i64 %b, i64 42
+  ret i64 %Q
+
+ exit:
+  ret i64 0
+}
diff --git a/test/Transforms/CodeGenPrepare/statepoint-relocate.ll b/test/Transforms/CodeGenPrepare/statepoint-relocate.ll
new file mode 100644
index 000000000000..b6898b373113
--- /dev/null
+++ b/test/Transforms/CodeGenPrepare/statepoint-relocate.ll
@@ -0,0 +1,88 @@
+; RUN: opt -codegenprepare -S < %s | FileCheck %s
+
+target datalayout = "e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-pc-linux-gnu"
+
+declare zeroext i1 @return_i1()
+
+define i32 @test_sor_basic(i32* %base) gc "statepoint-example" {
+; CHECK: getelementptr i32, i32* %base, i32 15
+; CHECK: getelementptr i32, i32* %base-new, i32 15
+entry:
+       %ptr = getelementptr i32, i32* %base, i32 15
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32* %base, i32* %ptr)
+       %base-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok,  i32 7, i32 7)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+define i32 @test_sor_two_derived(i32* %base) gc "statepoint-example" {
+; CHECK: getelementptr i32, i32* %base, i32 15
+; CHECK: getelementptr i32, i32* %base, i32 12
+; CHECK: getelementptr i32, i32* %base-new, i32 12
+; CHECK: getelementptr i32, i32* %base-new, i32 15
+entry:
+       %ptr = getelementptr i32, i32* %base, i32 15
+       %ptr2 = getelementptr i32, i32* %base, i32 12
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32* %base, i32* %ptr, i32* %ptr2)
+       %base-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok,  i32 7, i32 7)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %ptr2-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 9)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+define i32 @test_sor_ooo(i32* %base) gc "statepoint-example" {
+; CHECK: getelementptr i32, i32* %base, i32 15
+; CHECK: getelementptr i32, i32* %base-new, i32 15
+entry:
+       %ptr = getelementptr i32, i32* %base, i32 15
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32* %base, i32* %ptr)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %base-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok,  i32 7, i32 7)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+define i32 @test_sor_gep_smallint([3 x i32]* %base) gc "statepoint-example" {
+; CHECK: getelementptr [3 x i32], [3 x i32]* %base, i32 0, i32 2
+; CHECK: getelementptr [3 x i32], [3 x i32]* %base-new, i32 0, i32 2
+entry:
+       %ptr = getelementptr [3 x i32], [3 x i32]* %base, i32 0, i32 2
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, [3 x i32]* %base, i32* %ptr)
+       %base-new = call [3 x i32]* @llvm.experimental.gc.relocate.p0a3i32(i32 %tok,  i32 7, i32 7)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+define i32 @test_sor_gep_largeint([3 x i32]* %base) gc "statepoint-example" {
+; CHECK: getelementptr [3 x i32], [3 x i32]* %base, i32 0, i32 21
+; CHECK-NOT: getelementptr [3 x i32], [3 x i32]* %base-new, i32 0, i32 21
+entry:
+       %ptr = getelementptr [3 x i32], [3 x i32]* %base, i32 0, i32 21
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, [3 x i32]* %base, i32* %ptr)
+       %base-new = call [3 x i32]* @llvm.experimental.gc.relocate.p0a3i32(i32 %tok,  i32 7, i32 7)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+define i32 @test_sor_noop(i32* %base) gc "statepoint-example" {
+; CHECK: getelementptr i32, i32* %base, i32 15
+; CHECK: call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+; CHECK: call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 9)
+entry:
+       %ptr = getelementptr i32, i32* %base, i32 15
+       %ptr2 = getelementptr i32, i32* %base, i32 12
+       %tok = call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32* %base, i32* %ptr, i32* %ptr2)
+       %ptr-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 8)
+       %ptr2-new = call i32* @llvm.experimental.gc.relocate.p0i32(i32 %tok, i32 7, i32 9)
+       %ret = load i32, i32* %ptr-new
+       ret i32 %ret
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_i1f(i64, i32, i1 ()*, i32, i32, ...)
+declare i32* @llvm.experimental.gc.relocate.p0i32(i32, i32, i32)
+declare [3 x i32]* @llvm.experimental.gc.relocate.p0a3i32(i32, i32, i32)
diff --git a/test/Transforms/ConstProp/2005-01-28-SetCCGEP.ll b/test/Transforms/ConstProp/2005-01-28-SetCCGEP.ll
index d68cb26da348..af7e8be4fcd8 100644
--- a/test/Transforms/ConstProp/2005-01-28-SetCCGEP.ll
+++ b/test/Transforms/ConstProp/2005-01-28-SetCCGEP.ll
@@ -4,7 +4,7 @@
 @b = external global [2 x {  }]         ; <[2 x {  }]*> [#uses=2]
 
 define i1 @f() {
-        %tmp.2 = icmp eq {  }* getelementptr ([2 x {  }]* @b, i32 0, i32 0), getelementptr ([2 x {  }]* @b, i32 0, i32 1)                ; <i1> [#uses=1]
+        %tmp.2 = icmp eq {  }* getelementptr ([2 x {  }], [2 x {  }]* @b, i32 0, i32 0), getelementptr ([2 x {  }], [2 x {  }]* @b, i32 0, i32 1)                ; <i1> [#uses=1]
         ret i1 %tmp.2
 }
 
diff --git a/test/Transforms/ConstProp/2009-09-01-GEP-Crash.ll b/test/Transforms/ConstProp/2009-09-01-GEP-Crash.ll
index fc7ff905ecb0..e93a2c066bd2 100644
--- a/test/Transforms/ConstProp/2009-09-01-GEP-Crash.ll
+++ b/test/Transforms/ConstProp/2009-09-01-GEP-Crash.ll
@@ -18,7 +18,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @init_hrtimers_cpu(i32 %cpu) nounwind noredzone section ".cpuinit.text" {
 entry:
-  %tmp3 = getelementptr %struct.hrtimer_cpu_base* bitcast (%1* @per_cpu__hrtimer_bases to %struct.hrtimer_cpu_base*), i32 0, i32 0 ; <%0*> [#uses=1]
+  %tmp3 = getelementptr %struct.hrtimer_cpu_base, %struct.hrtimer_cpu_base* bitcast (%1* @per_cpu__hrtimer_bases to %struct.hrtimer_cpu_base*), i32 0, i32 0 ; <%0*> [#uses=1]
   %tmp5 = bitcast %0* %tmp3 to i8*                ; <i8*> [#uses=0]
   unreachable
 }
diff --git a/test/Transforms/ConstProp/InsertElement.ll b/test/Transforms/ConstProp/InsertElement.ll
new file mode 100644
index 000000000000..011ad3f57f46
--- /dev/null
+++ b/test/Transforms/ConstProp/InsertElement.ll
@@ -0,0 +1,32 @@
+; RUN: opt < %s -constprop -S | FileCheck %s
+
+; CHECK-LABEL: @test1
+define i32 @test1() {
+  %A = bitcast i32 2139171423 to float
+  %B = insertelement <1 x float> undef, float %A, i32 0
+  %C = extractelement <1 x float> %B, i32 0
+  %D = bitcast float %C to i32
+  ret i32 %D
+; CHECK: ret i32 2139171423
+}
+
+; CHECK-LABEL: @insertelement
+define <4 x i64> @insertelement() {
+  %vec1 = insertelement <4 x i64> undef, i64 -1, i32 0
+  %vec2 = insertelement <4 x i64> %vec1, i64 -2, i32 1
+  %vec3 = insertelement <4 x i64> %vec2, i64 -3, i32 2
+  %vec4 = insertelement <4 x i64> %vec3, i64 -4, i32 3
+  ; CHECK: ret <4 x i64> <i64 -1, i64 -2, i64 -3, i64 -4>
+  ret <4 x i64> %vec4
+}
+
+; CHECK-LABEL: @insertelement_undef
+define <4 x i64> @insertelement_undef() {
+  %vec1 = insertelement <4 x i64> undef, i64 -1, i32 0
+  %vec2 = insertelement <4 x i64> %vec1, i64 -2, i32 1
+  %vec3 = insertelement <4 x i64> %vec2, i64 -3, i32 2
+  %vec4 = insertelement <4 x i64> %vec3, i64 -4, i32 3
+  %vec5 = insertelement <4 x i64> %vec3, i64 -5, i32 4
+  ; CHECK: ret <4 x i64> undef
+  ret <4 x i64> %vec5
+}
diff --git a/test/Transforms/ConstProp/convert-from-fp16.ll b/test/Transforms/ConstProp/convert-from-fp16.ll
new file mode 100644
index 000000000000..bb9032313179
--- /dev/null
+++ b/test/Transforms/ConstProp/convert-from-fp16.ll
@@ -0,0 +1,97 @@
+; RUN: opt -constprop -S < %s | FileCheck %s
+
+; Verify that we don't crash with an assertion failure when constant folding
+; a call to intrinsic 'convert.from.fp16' if the return type is not 'float'.
+
+define float @fold_from_fp16_to_fp32() {
+; CHECK-LABEL: @fold_from_fp16_to_fp32
+; CHECK: ret float 0.000000e+00
+entry:
+  %0 = call float @llvm.convert.from.fp16.f32(i16 0)
+  ret float %0
+}
+
+define double @fold_from_fp16_to_fp64() {
+; CHECK-LABEL: @fold_from_fp16_to_fp64
+; CHECK: ret double 0.000000e+00
+entry:
+  %0 = call double @llvm.convert.from.fp16.f64(i16 0)
+  ret double %0
+}
+
+define x86_fp80 @fold_from_fp16_to_fp80() {
+; CHECK-LABEL: @fold_from_fp16_to_fp80
+; CHECK: ret x86_fp80 0xK00000000000000000000
+entry:
+  %0 = call x86_fp80 @llvm.convert.from.fp16.f80(i16 0)
+  ret x86_fp80 %0
+}
+
+define fp128 @fold_from_fp16_to_fp128() {
+; CHECK-LABEL: @fold_from_fp16_to_fp128
+; CHECK: ret fp128 0xL00000000000000000000000000000000
+entry:
+  %0 = call fp128 @llvm.convert.from.fp16.f128(i16 0)
+  ret fp128 %0
+}
+
+define ppc_fp128 @fold_from_fp16_to_ppcfp128() {
+; CHECK-LABEL: @fold_from_fp16_to_ppcfp128
+; CHECK: ret ppc_fp128 0xM00000000000000000000000000000000
+entry:
+  %0 = call ppc_fp128 @llvm.convert.from.fp16.ppcf128(i16 0)
+  ret ppc_fp128 %0
+}
+
+define float @fold_from_fp16_to_fp32_b() {
+; CHECK-LABEL: @fold_from_fp16_to_fp32_b
+; CHECK: ret float 4.000000e+00
+entry:
+  %0 = call i16 @llvm.convert.to.fp16.f64(double 4.0)
+  %1 = call float @llvm.convert.from.fp16.f32(i16 %0)
+  ret float %1
+}
+
+define double @fold_from_fp16_to_fp64_b() {
+; CHECK-LABEL: @fold_from_fp16_to_fp64_b
+; CHECK: ret double 4.000000e+00
+entry:
+  %0 = call i16 @llvm.convert.to.fp16.f64(double 4.0)
+  %1 = call double @llvm.convert.from.fp16.f64(i16 %0)
+  ret double %1
+}
+
+define x86_fp80 @fold_from_fp16_to_fp80_b() {
+; CHECK-LABEL: @fold_from_fp16_to_fp80_b
+; CHECK: ret x86_fp80 0xK40018000000000000000
+entry:
+  %0 = call i16 @llvm.convert.to.fp16.f64(double 4.0)
+  %1 = call x86_fp80 @llvm.convert.from.fp16.f80(i16 %0)
+  ret x86_fp80 %1
+}
+
+define fp128 @fold_from_fp16_to_fp128_b() {
+; CHECK-LABEL: @fold_from_fp16_to_fp128_b
+; CHECK: ret fp128 0xL00000000000000004001000000000000
+entry:
+  %0 = call i16 @llvm.convert.to.fp16.f64(double 4.0)
+  %1 = call fp128 @llvm.convert.from.fp16.f128(i16 %0)
+  ret fp128 %1
+}
+
+define ppc_fp128 @fold_from_fp16_to_ppcfp128_b() {
+; CHECK-LABEL: @fold_from_fp16_to_ppcfp128_b
+; CHECK: ret ppc_fp128 0xM40100000000000000000000000000000
+entry:
+  %0 = call i16 @llvm.convert.to.fp16.f64(double 4.0)
+  %1 = call ppc_fp128 @llvm.convert.from.fp16.ppcf128(i16 %0)
+  ret ppc_fp128 %1
+}
+
+
+declare i16 @llvm.convert.to.fp16.f64(double)
+declare float @llvm.convert.from.fp16.f32(i16)
+declare double @llvm.convert.from.fp16.f64(i16)
+declare x86_fp80 @llvm.convert.from.fp16.f80(i16)
+declare fp128 @llvm.convert.from.fp16.f128(i16)
+declare ppc_fp128 @llvm.convert.from.fp16.ppcf128(i16)
diff --git a/test/Transforms/ConstProp/div-zero.ll b/test/Transforms/ConstProp/div-zero.ll
index a2c59d3c002a..f4049a9615a8 100644
--- a/test/Transforms/ConstProp/div-zero.ll
+++ b/test/Transforms/ConstProp/div-zero.ll
@@ -5,7 +5,7 @@ declare void @ext()
 define i32 @foo(i32 %ptr) {
 entry:
         %zero = sub i32 %ptr, %ptr              ; <i32> [#uses=1]
-        %div_zero = sdiv i32 %zero, ptrtoint (i32* getelementptr (i32* null,
+        %div_zero = sdiv i32 %zero, ptrtoint (i32* getelementptr (i32, i32* null,
 i32 1) to i32)             ; <i32> [#uses=1]
         ret i32 %div_zero
 }
diff --git a/test/Transforms/ConstProp/insertvalue.ll b/test/Transforms/ConstProp/insertvalue.ll
index 0d288b3841df..dce2b728b93b 100644
--- a/test/Transforms/ConstProp/insertvalue.ll
+++ b/test/Transforms/ConstProp/insertvalue.ll
@@ -65,3 +65,12 @@ define [3 x %struct] @undef-test3() {
 ; CHECK: ret [3 x %struct] [%struct undef, %struct { i32 0, [4 x i8] undef }, %struct undef]
 }
 
+define i32 @test-float-Nan() {
+  %A = bitcast i32 2139171423 to float
+  %B = insertvalue [1 x float] undef, float %A, 0
+  %C = extractvalue [1 x float] %B, 0
+  %D = bitcast float %C to i32
+  ret i32 %D
+; CHECK: @test-float-Nan
+; CHECK: ret i32 2139171423
+}
diff --git a/test/Transforms/ConstProp/loads.ll b/test/Transforms/ConstProp/loads.ll
index 5a23dad87f3a..5426ad0f8adb 100644
--- a/test/Transforms/ConstProp/loads.ll
+++ b/test/Transforms/ConstProp/loads.ll
@@ -9,7 +9,7 @@
 
 ; Simple load
 define i32 @test1() {
-  %r = load i32* getelementptr ({{i32,i8},i32}* @g1, i32 0, i32 0, i32 0)
+  %r = load i32, i32* getelementptr ({{i32,i8},i32}, {{i32,i8},i32}* @g1, i32 0, i32 0, i32 0)
   ret i32 %r
 
 ; 0xDEADBEEF
@@ -24,7 +24,7 @@ define i32 @test1() {
 ; PR3152
 ; Load of first 16 bits of 32-bit value.
 define i16 @test2() {
-  %r = load i16* bitcast(i32* getelementptr ({{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*)
+  %r = load i16, i16* bitcast(i32* getelementptr ({{i32,i8},i32}, {{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*)
   ret i16 %r
 
 ; 0xBEEF
@@ -37,7 +37,7 @@ define i16 @test2() {
 }
 
 define i16 @test2_addrspacecast() {
-  %r = load i16 addrspace(1)* addrspacecast(i32* getelementptr ({{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16 addrspace(1)*)
+  %r = load i16, i16 addrspace(1)* addrspacecast(i32* getelementptr ({{i32,i8},i32}, {{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16 addrspace(1)*)
   ret i16 %r
 
 ; 0xBEEF
@@ -51,7 +51,7 @@ define i16 @test2_addrspacecast() {
 
 ; Load of second 16 bits of 32-bit value.
 define i16 @test3() {
-  %r = load i16* getelementptr(i16* bitcast(i32* getelementptr ({{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*), i32 1)
+  %r = load i16, i16* getelementptr(i16, i16* bitcast(i32* getelementptr ({{i32,i8},i32}, {{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*), i32 1)
   ret i16 %r
 
 ; 0xDEAD
@@ -65,7 +65,7 @@ define i16 @test3() {
 
 ; Load of 8 bit field + tail padding.
 define i16 @test4() {
-  %r = load i16* getelementptr(i16* bitcast(i32* getelementptr ({{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*), i32 2)
+  %r = load i16, i16* getelementptr(i16, i16* bitcast(i32* getelementptr ({{i32,i8},i32}, {{i32,i8},i32}* @g1, i32 0, i32 0, i32 0) to i16*), i32 2)
   ret i16 %r
 
 ; 0x00BA
@@ -79,7 +79,7 @@ define i16 @test4() {
 
 ; Load of double bits.
 define i64 @test6() {
-  %r = load i64* bitcast(double* @g2 to i64*)
+  %r = load i64, i64* bitcast(double* @g2 to i64*)
   ret i64 %r
 
 ; 0x3FF_0000000000000
@@ -93,7 +93,7 @@ define i64 @test6() {
 
 ; Load of double bits.
 define i16 @test7() {
-  %r = load i16* bitcast(double* @g2 to i16*)
+  %r = load i16, i16* bitcast(double* @g2 to i16*)
   ret i16 %r
 
 ; 0x0000
@@ -107,7 +107,7 @@ define i16 @test7() {
 
 ; Double load.
 define double @test8() {
-  %r = load double* bitcast({{i32,i8},i32}* @g1 to double*)
+  %r = load double, double* bitcast({{i32,i8},i32}* @g1 to double*)
   ret double %r
 
 ; LE-LABEL: @test8(
@@ -120,7 +120,7 @@ define double @test8() {
 
 ; i128 load.
 define i128 @test9() {
-  %r = load i128* bitcast({i64, i64}* @g3 to i128*)
+  %r = load i128, i128* bitcast({i64, i64}* @g3 to i128*)
   ret i128 %r
 
 ; 0x00000000_06B1BFF8_00000000_0000007B
@@ -134,7 +134,7 @@ define i128 @test9() {
 
 ; vector load.
 define <2 x i64> @test10() {
-  %r = load <2 x i64>* bitcast({i64, i64}* @g3 to <2 x i64>*)
+  %r = load <2 x i64>, <2 x i64>* bitcast({i64, i64}* @g3 to <2 x i64>*)
   ret <2 x i64> %r
 
 ; LE-LABEL: @test10(
@@ -151,7 +151,7 @@ define <2 x i64> @test10() {
 
 define i16 @test11() nounwind {
 entry:
-  %a = load i16* bitcast ({ i8, i8 }* @g4 to i16*)
+  %a = load i16, i16* bitcast ({ i8, i8 }* @g4 to i16*)
   ret i16 %a
 
 ; 0x08A1
@@ -168,7 +168,7 @@ entry:
 @test12g = private constant [6 x i8] c"a\00b\00\00\00"
 
 define i16 @test12() {
-  %a = load i16* getelementptr inbounds ([3 x i16]* bitcast ([6 x i8]* @test12g to [3 x i16]*), i32 0, i64 1)
+  %a = load i16, i16* getelementptr inbounds ([3 x i16], [3 x i16]* bitcast ([6 x i8]* @test12g to [3 x i16]*), i32 0, i64 1)
   ret i16 %a
 
 ; 0x0062
@@ -184,7 +184,7 @@ define i16 @test12() {
 ; PR5978
 @g5 = constant i8 4
 define i1 @test13() {
-  %A = load i1* bitcast (i8* @g5 to i1*)
+  %A = load i1, i1* bitcast (i8* @g5 to i1*)
   ret i1 %A
 
 ; LE-LABEL: @test13(
@@ -197,7 +197,7 @@ define i1 @test13() {
 @g6 = constant [2 x i8*] [i8* inttoptr (i64 1 to i8*), i8* inttoptr (i64 2 to i8*)]
 define i64 @test14() nounwind {
 entry:
-  %tmp = load i64* bitcast ([2 x i8*]* @g6 to i64*)
+  %tmp = load i64, i64* bitcast ([2 x i8*]* @g6 to i64*)
   ret i64 %tmp
 
 ; LE-LABEL: @test14(
@@ -211,7 +211,7 @@ entry:
 @g6_as1 = constant [2 x i8 addrspace(1)*] [i8 addrspace(1)* inttoptr (i16 1 to i8 addrspace(1)*), i8 addrspace(1)* inttoptr (i16 2 to i8 addrspace(1)*)]
 define i16 @test14_as1() nounwind {
 entry:
-  %tmp = load i16* bitcast ([2 x i8 addrspace(1)*]* @g6_as1 to i16*)
+  %tmp = load i16, i16* bitcast ([2 x i8 addrspace(1)*]* @g6_as1 to i16*)
   ret i16 %tmp
 
 ; LE: @test14_as1
@@ -223,7 +223,7 @@ entry:
 
 define i64 @test15() nounwind {
 entry:
-  %tmp = load i64* bitcast (i8** getelementptr inbounds ([2 x i8*]* @g6, i32 0, i64 1) to i64*)
+  %tmp = load i64, i64* bitcast (i8** getelementptr inbounds ([2 x i8*], [2 x i8*]* @g6, i32 0, i64 1) to i64*)
   ret i64 %tmp
 
 ; LE-LABEL: @test15(
@@ -235,7 +235,7 @@ entry:
 
 @gv7 = constant [4 x i8*] [i8* null, i8* inttoptr (i64 -14 to i8*), i8* null, i8* null]
 define i64 @test16.1() {
-  %v = load i64* bitcast ([4 x i8*]* @gv7 to i64*), align 8
+  %v = load i64, i64* bitcast ([4 x i8*]* @gv7 to i64*), align 8
   ret i64 %v
 
 ; LE-LABEL: @test16.1(
@@ -246,7 +246,7 @@ define i64 @test16.1() {
 }
 
 define i64 @test16.2() {
-  %v = load i64* bitcast (i8** getelementptr inbounds ([4 x i8*]* @gv7, i64 0, i64 1) to i64*), align 8
+  %v = load i64, i64* bitcast (i8** getelementptr inbounds ([4 x i8*], [4 x i8*]* @gv7, i64 0, i64 1) to i64*), align 8
   ret i64 %v
 
 ; LE-LABEL: @test16.2(
@@ -257,7 +257,7 @@ define i64 @test16.2() {
 }
 
 define i64 @test16.3() {
-  %v = load i64* bitcast (i8** getelementptr inbounds ([4 x i8*]* @gv7, i64 0, i64 2) to i64*), align 8
+  %v = load i64, i64* bitcast (i8** getelementptr inbounds ([4 x i8*], [4 x i8*]* @gv7, i64 0, i64 2) to i64*), align 8
   ret i64 %v
 
 ; LE-LABEL: @test16.3(
diff --git a/test/Transforms/ConstantHoisting/AArch64/const-addr.ll b/test/Transforms/ConstantHoisting/AArch64/const-addr.ll
index 89d596055c44..4c36d20dc6b0 100644
--- a/test/Transforms/ConstantHoisting/AArch64/const-addr.ll
+++ b/test/Transforms/ConstantHoisting/AArch64/const-addr.ll
@@ -6,17 +6,17 @@ define i32 @test1() nounwind {
 ; CHECK-LABEL: test1
 ; CHECK: %const = bitcast i64 68141056 to i64
 ; CHECK: %1 = inttoptr i64 %const to %T*
-; CHECK: %o1 = getelementptr %T* %1, i32 0, i32 1
-; CHECK: %o2 = getelementptr %T* %1, i32 0, i32 2
-; CHECK: %o3 = getelementptr %T* %1, i32 0, i32 3
+; CHECK: %o1 = getelementptr %T, %T* %1, i32 0, i32 1
+; CHECK: %o2 = getelementptr %T, %T* %1, i32 0, i32 2
+; CHECK: %o3 = getelementptr %T, %T* %1, i32 0, i32 3
   %at = inttoptr i64 68141056 to %T*
-  %o1 = getelementptr %T* %at, i32 0, i32 1
-  %t1 = load i32* %o1
-  %o2 = getelementptr %T* %at, i32 0, i32 2
-  %t2 = load i32* %o2
+  %o1 = getelementptr %T, %T* %at, i32 0, i32 1
+  %t1 = load i32, i32* %o1
+  %o2 = getelementptr %T, %T* %at, i32 0, i32 2
+  %t2 = load i32, i32* %o2
   %a1 = add i32 %t1, %t2
-  %o3 = getelementptr %T* %at, i32 0, i32 3
-  %t3 = load i32* %o3
+  %o3 = getelementptr %T, %T* %at, i32 0, i32 3
+  %t3 = load i32, i32* %o3
   %a2 = add i32 %a1, %t3
   ret i32 %a2
 }
diff --git a/test/Transforms/ConstantHoisting/PowerPC/const-base-addr.ll b/test/Transforms/ConstantHoisting/PowerPC/const-base-addr.ll
index b4337eeda653..69b13cf9c9a4 100644
--- a/test/Transforms/ConstantHoisting/PowerPC/const-base-addr.ll
+++ b/test/Transforms/ConstantHoisting/PowerPC/const-base-addr.ll
@@ -9,13 +9,13 @@ define i32 @test1() nounwind {
 ; CHECK-LABEL:  @test1
 ; CHECK:        %const = bitcast i32 12345678 to i32
 ; CHECK:        %1 = inttoptr i32 %const to %T*
-; CHECK:        %addr1 = getelementptr %T* %1, i32 0, i32 1
-  %addr1 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 1
-  %tmp1 = load i32* %addr1
-  %addr2 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 2
-  %tmp2 = load i32* %addr2
-  %addr3 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 3
-  %tmp3 = load i32* %addr3
+; CHECK:        %addr1 = getelementptr %T, %T* %1, i32 0, i32 1
+  %addr1 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 1
+  %tmp1 = load i32, i32* %addr1
+  %addr2 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 2
+  %tmp2 = load i32, i32* %addr2
+  %addr3 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 3
+  %tmp3 = load i32, i32* %addr3
   %tmp4 = add i32 %tmp1, %tmp2
   %tmp5 = add i32 %tmp3, %tmp4
   ret i32 %tmp5
diff --git a/test/Transforms/ConstantHoisting/PowerPC/masks.ll b/test/Transforms/ConstantHoisting/PowerPC/masks.ll
index d55318201136..4cc504f11bd9 100644
--- a/test/Transforms/ConstantHoisting/PowerPC/masks.ll
+++ b/test/Transforms/ConstantHoisting/PowerPC/masks.ll
@@ -19,7 +19,7 @@ if.then152:
 if.end167:
 ; CHECK: and i32 {{.*}}, 32768
   %shl161 = shl nuw nsw i32 %conv121, 15
-  %0 = load i8* undef, align 1
+  %0 = load i8, i8* undef, align 1
   %conv169 = zext i8 %0 to i32
   %shl170 = shl nuw nsw i32 %conv169, 7
   %shl161.masked = and i32 %shl161, 32768
@@ -49,7 +49,7 @@ if.then152:
 if.end167:
 ; CHECK: add i32 {{.*}}, -32758
   %shl161 = shl nuw nsw i32 %conv121, 15
-  %0 = load i8* undef, align 1
+  %0 = load i8, i8* undef, align 1
   %conv169 = zext i8 %0 to i32
   %shl170 = shl nuw nsw i32 %conv169, 7
   %shl161.masked = and i32 %shl161, 32773
diff --git a/test/Transforms/ConstantHoisting/X86/cast-inst.ll b/test/Transforms/ConstantHoisting/X86/cast-inst.ll
index f490f4a37236..bd1e196b6445 100644
--- a/test/Transforms/ConstantHoisting/X86/cast-inst.ll
+++ b/test/Transforms/ConstantHoisting/X86/cast-inst.ll
@@ -9,19 +9,19 @@ define i32 @cast_inst_test() {
 ; CHECK-LABEL:  @cast_inst_test
 ; CHECK:        %const = bitcast i64 4646526064 to i64
 ; CHECK:        %1 = inttoptr i64 %const to i32*
-; CHECK:        %v0 = load i32* %1, align 16
+; CHECK:        %v0 = load i32, i32* %1, align 16
 ; CHECK:        %const_mat = add i64 %const, 16
 ; CHECK-NEXT:   %2 = inttoptr i64 %const_mat to i32*
-; CHECK-NEXT:   %v1 = load i32* %2, align 16
+; CHECK-NEXT:   %v1 = load i32, i32* %2, align 16
 ; CHECK:        %const_mat1 = add i64 %const, 32
 ; CHECK-NEXT:   %3 = inttoptr i64 %const_mat1 to i32*
-; CHECK-NEXT:   %v2 = load i32* %3, align 16
+; CHECK-NEXT:   %v2 = load i32, i32* %3, align 16
   %a0 = inttoptr i64 4646526064 to i32*
-  %v0 = load i32* %a0, align 16
+  %v0 = load i32, i32* %a0, align 16
   %a1 = inttoptr i64 4646526080 to i32*
-  %v1 = load i32* %a1, align 16
+  %v1 = load i32, i32* %a1, align 16
   %a2 = inttoptr i64 4646526096 to i32*
-  %v2 = load i32* %a2, align 16
+  %v2 = load i32, i32* %a2, align 16
   %r0 = add i32 %v0, %v1
   %r1 = add i32 %r0, %v2
   ret i32 %r1
diff --git a/test/Transforms/ConstantHoisting/X86/const-base-addr.ll b/test/Transforms/ConstantHoisting/X86/const-base-addr.ll
index 01e6cdfd2a5b..db5dfdd1538a 100644
--- a/test/Transforms/ConstantHoisting/X86/const-base-addr.ll
+++ b/test/Transforms/ConstantHoisting/X86/const-base-addr.ll
@@ -10,13 +10,13 @@ define i32 @test1() nounwind {
 ; CHECK-LABEL:  @test1
 ; CHECK:        %const = bitcast i32 12345678 to i32
 ; CHECK:        %1 = inttoptr i32 %const to %T*
-; CHECK:        %addr1 = getelementptr %T* %1, i32 0, i32 1
-  %addr1 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 1
-  %tmp1 = load i32* %addr1
-  %addr2 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 2
-  %tmp2 = load i32* %addr2
-  %addr3 = getelementptr %T* inttoptr (i32 12345678 to %T*), i32 0, i32 3
-  %tmp3 = load i32* %addr3
+; CHECK:        %addr1 = getelementptr %T, %T* %1, i32 0, i32 1
+  %addr1 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 1
+  %tmp1 = load i32, i32* %addr1
+  %addr2 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 2
+  %tmp2 = load i32, i32* %addr2
+  %addr3 = getelementptr %T, %T* inttoptr (i32 12345678 to %T*), i32 0, i32 3
+  %tmp3 = load i32, i32* %addr3
   %tmp4 = add i32 %tmp1, %tmp2
   %tmp5 = add i32 %tmp3, %tmp4
   ret i32 %tmp5
diff --git a/test/Transforms/ConstantHoisting/X86/delete-dead-cast-inst.ll b/test/Transforms/ConstantHoisting/X86/delete-dead-cast-inst.ll
index d35238692078..5df4c1a376e5 100644
--- a/test/Transforms/ConstantHoisting/X86/delete-dead-cast-inst.ll
+++ b/test/Transforms/ConstantHoisting/X86/delete-dead-cast-inst.ll
@@ -10,13 +10,13 @@ define i32 @test1() nounwind {
 ; CHECK:        %const = bitcast i32 12345678 to i32
 ; CHECK-NOT:    %base = inttoptr i32 12345678 to %T*
 ; CHECK-NEXT:   %1 = inttoptr i32 %const to %T*
-; CHECK-NEXT:   %addr1 = getelementptr %T* %1, i32 0, i32 1
-; CHECK-NEXT:   %addr2 = getelementptr %T* %1, i32 0, i32 2
-; CHECK-NEXT:   %addr3 = getelementptr %T* %1, i32 0, i32 3
+; CHECK-NEXT:   %addr1 = getelementptr %T, %T* %1, i32 0, i32 1
+; CHECK-NEXT:   %addr2 = getelementptr %T, %T* %1, i32 0, i32 2
+; CHECK-NEXT:   %addr3 = getelementptr %T, %T* %1, i32 0, i32 3
   %base = inttoptr i32 12345678 to %T*
-  %addr1 = getelementptr %T* %base, i32 0, i32 1
-  %addr2 = getelementptr %T* %base, i32 0, i32 2
-  %addr3 = getelementptr %T* %base, i32 0, i32 3
+  %addr1 = getelementptr %T, %T* %base, i32 0, i32 1
+  %addr2 = getelementptr %T, %T* %base, i32 0, i32 2
+  %addr3 = getelementptr %T, %T* %base, i32 0, i32 3
   ret i32 12345678
 }
 
diff --git a/test/Transforms/ConstantHoisting/X86/stackmap.ll b/test/Transforms/ConstantHoisting/X86/stackmap.ll
index 9df44177820e..b9aee6b38fb0 100644
--- a/test/Transforms/ConstantHoisting/X86/stackmap.ll
+++ b/test/Transforms/ConstantHoisting/X86/stackmap.ll
@@ -7,10 +7,10 @@ target triple = "x86_64-apple-macosx10.9.0"
 define i128 @test1(i128 %a) {
 ; CHECK-LABEL:  @test1
 ; CHECK:        %const = bitcast i128 134646182756734033220 to i128
-; CHECK:        tail call void (i64, i32, ...)* @llvm.experimental.stackmap(i64 1, i32 24, i128 %const)
+; CHECK:        tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 1, i32 24, i128 %const)
 entry:
   %0 = add i128 %a, 134646182756734033220
-  tail call void (i64, i32, ...)* @llvm.experimental.stackmap(i64 1, i32 24, i128 134646182756734033220)
+  tail call void (i64, i32, ...) @llvm.experimental.stackmap(i64 1, i32 24, i128 134646182756734033220)
   ret i128 %0
 }
 
diff --git a/test/Transforms/ConstantMerge/2002-09-23-CPR-Update.ll b/test/Transforms/ConstantMerge/2002-09-23-CPR-Update.ll
index b7b05cf1c370..2e83503af815 100644
--- a/test/Transforms/ConstantMerge/2002-09-23-CPR-Update.ll
+++ b/test/Transforms/ConstantMerge/2002-09-23-CPR-Update.ll
@@ -6,8 +6,8 @@
 declare i32 @test(i32*)
 
 define void @foo() {
-        call i32 @test( i32* getelementptr ({ i32 }* @foo.upgrd.1, i64 0, i32 0) )              ; <i32>:1 [#uses=0]
-        call i32 @test( i32* getelementptr ({ i32 }* @bar, i64 0, i32 0) )              ; <i32>:2 [#uses=0]
+        call i32 @test( i32* getelementptr ({ i32 }, { i32 }* @foo.upgrd.1, i64 0, i32 0) )              ; <i32>:1 [#uses=0]
+        call i32 @test( i32* getelementptr ({ i32 }, { i32 }* @bar, i64 0, i32 0) )              ; <i32>:2 [#uses=0]
         ret void
 }
 
diff --git a/test/Transforms/ConstantMerge/merge-both.ll b/test/Transforms/ConstantMerge/merge-both.ll
index 316267648f1f..11b0621d42d7 100644
--- a/test/Transforms/ConstantMerge/merge-both.ll
+++ b/test/Transforms/ConstantMerge/merge-both.ll
@@ -1,7 +1,6 @@
 ; RUN: opt -constmerge -S < %s | FileCheck %s
 ; Test that in one run var3 is merged into var2 and var1 into var4.
-; Test that we merge @var5 and @var6 into one with the higher alignment, and
-; don't merge var7/var8 into var5/var6.
+; Test that we merge @var5 and @var6 into one with the higher alignment
 
 declare void @zed(%struct.foobar*, %struct.foobar*)
 
@@ -23,8 +22,8 @@ declare void @helper([16 x i8]*)
 @var7 = internal constant [16 x i8] c"foo1bar2foo3bar\00"
 @var8 = private unnamed_addr constant [16 x i8] c"foo1bar2foo3bar\00"
 
-; CHECK-NEXT: @var6 = private constant [16 x i8] c"foo1bar2foo3bar\00", align 16
-; CHECK-NEXT: @var8 = private constant [16 x i8] c"foo1bar2foo3bar\00"
+; CHECK-NEXT: @var7 = internal constant [16 x i8] c"foo1bar2foo3bar\00"
+; CHECK-NEXT: @var8 = private constant [16 x i8] c"foo1bar2foo3bar\00", align 16
 
 @var4a = alias %struct.foobar* @var4
 @llvm.used = appending global [1 x %struct.foobar*] [%struct.foobar* @var4a], section "llvm.metadata"
diff --git a/test/Transforms/CorrelatedValuePropagation/basic.ll b/test/Transforms/CorrelatedValuePropagation/basic.ll
index 9a2264793c50..9d1253a1888e 100644
--- a/test/Transforms/CorrelatedValuePropagation/basic.ll
+++ b/test/Transforms/CorrelatedValuePropagation/basic.ll
@@ -52,8 +52,8 @@ bb:             ; preds = %entry
         ret i8 0
 
 bb2:            ; preds = %entry
-; CHECK: %should_be_const = load i8* @gv
-        %should_be_const = load i8* %a
+; CHECK: %should_be_const = load i8, i8* @gv
+        %should_be_const = load i8, i8* %a
         ret i8 %should_be_const
 }
 
diff --git a/test/Transforms/CorrelatedValuePropagation/non-null.ll b/test/Transforms/CorrelatedValuePropagation/non-null.ll
index b14abd83694a..6bb8bb07c45f 100644
--- a/test/Transforms/CorrelatedValuePropagation/non-null.ll
+++ b/test/Transforms/CorrelatedValuePropagation/non-null.ll
@@ -2,7 +2,7 @@
 
 define void @test1(i8* %ptr) {
 ; CHECK: test1
-  %A = load i8* %ptr
+  %A = load i8, i8* %ptr
   br label %bb
 bb:
   icmp ne i8* %ptr, null
diff --git a/test/Transforms/CorrelatedValuePropagation/select.ll b/test/Transforms/CorrelatedValuePropagation/select.ll
new file mode 100644
index 000000000000..5501438f690f
--- /dev/null
+++ b/test/Transforms/CorrelatedValuePropagation/select.ll
@@ -0,0 +1,53 @@
+; RUN: opt < %s -correlated-propagation -S | FileCheck %s
+
+; CHECK-LABEL: @simple(
+define i8 @simple(i1) {
+entry:
+  %s = select i1 %0, i8 0, i8 1
+  br i1 %0, label %then, label %else
+
+then:
+; CHECK: ret i8 0
+  %a = phi i8 [ %s, %entry ]
+  ret i8 %a
+
+else:
+; CHECK: ret i8 1
+  %b = phi i8 [ %s, %entry ]
+  ret i8 %b
+}
+
+; CHECK-LABEL: @loop(
+define void @loop(i32) {
+entry:
+  br label %loop
+
+loop:
+  %idx = phi i32 [ %0, %entry ], [ %sel, %loop ]
+; CHECK: %idx = phi i32 [ %0, %entry ], [ %2, %loop ]
+  %1 = icmp eq i32 %idx, 0
+  %2 = add i32 %idx, -1
+  %sel = select i1 %1, i32 0, i32 %2
+  br i1 %1, label %out, label %loop
+
+out:
+  ret void
+}
+
+; CHECK-LABEL: @not_correlated(
+define i8 @not_correlated(i1, i1) {
+entry:
+  %s = select i1 %0, i8 0, i8 1
+  br i1 %1, label %then, label %else
+
+then:
+; CHECK: ret i8 %s
+  %a = phi i8 [ %s, %entry ]
+  ret i8 %a
+
+else:
+; CHECK: ret i8 %s
+  %b = phi i8 [ %s, %entry ]
+  ret i8 %b
+}
+
diff --git a/test/Transforms/DeadArgElim/2007-10-18-VarargsReturn.ll b/test/Transforms/DeadArgElim/2007-10-18-VarargsReturn.ll
index cdd893faba26..9bbc275ae7d7 100644
--- a/test/Transforms/DeadArgElim/2007-10-18-VarargsReturn.ll
+++ b/test/Transforms/DeadArgElim/2007-10-18-VarargsReturn.ll
@@ -6,7 +6,7 @@ define internal i32 @test(i32 %A, ...) {
 }
 
 define i32 @foo() {
-	%A = call i32(i32, ...)* @test(i32 1)
+	%A = call i32(i32, ...) @test(i32 1)
 	ret i32 %A
 }
 
diff --git a/test/Transforms/DeadArgElim/2007-12-20-ParamAttrs.ll b/test/Transforms/DeadArgElim/2007-12-20-ParamAttrs.ll
index f049265ce4ea..ab378e9df1ee 100644
--- a/test/Transforms/DeadArgElim/2007-12-20-ParamAttrs.ll
+++ b/test/Transforms/DeadArgElim/2007-12-20-ParamAttrs.ll
@@ -13,7 +13,7 @@ define internal zeroext i8 @foo(i8* inreg %p, i8 signext %y, ... )  nounwind {
 
 define i32 @bar() {
 ; CHECK: call void @foo(i8 signext 1) [[NUW]]
-  %A = call zeroext i8(i8*, i8, ...)* @foo(i8* inreg null, i8 signext 1, %struct* byval null ) nounwind
+  %A = call zeroext i8(i8*, i8, ...) @foo(i8* inreg null, i8 signext 1, %struct* byval null ) nounwind
   ret i32 0
 }
 
diff --git a/test/Transforms/DeadArgElim/2008-01-16-VarargsParamAttrs.ll b/test/Transforms/DeadArgElim/2008-01-16-VarargsParamAttrs.ll
index 93282f7f8f2b..48e43961e7a1 100644
--- a/test/Transforms/DeadArgElim/2008-01-16-VarargsParamAttrs.ll
+++ b/test/Transforms/DeadArgElim/2008-01-16-VarargsParamAttrs.ll
@@ -22,8 +22,8 @@ declare void @llvm.va_end(i8*) nounwind
 define i32 @main() {
 entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%tmp = getelementptr [4 x %struct.point]* @pts, i32 0, i32 0		; <%struct.point*> [#uses=1]
-	%tmp1 = call i32 (i32, ...)* @va1( i32 1, %struct.point* byval  %tmp ) nounwind 		; <i32> [#uses=0]
+	%tmp = getelementptr [4 x %struct.point], [4 x %struct.point]* @pts, i32 0, i32 0		; <%struct.point*> [#uses=1]
+	%tmp1 = call i32 (i32, ...) @va1( i32 1, %struct.point* byval  %tmp ) nounwind 		; <i32> [#uses=0]
 	call void @exit( i32 0 ) noreturn nounwind 
 	unreachable
 }
diff --git a/test/Transforms/DeadArgElim/2010-04-30-DbgInfo.ll b/test/Transforms/DeadArgElim/2010-04-30-DbgInfo.ll
index dd283aebc1d4..5ff05f0d6858 100644
--- a/test/Transforms/DeadArgElim/2010-04-30-DbgInfo.ll
+++ b/test/Transforms/DeadArgElim/2010-04-30-DbgInfo.ll
@@ -4,10 +4,10 @@
 
 define i8* @vfs_addname(i8* %name, i32 %len, i32 %hash, i32 %flags) nounwind ssp {
 entry:
-  call void @llvm.dbg.value(metadata i8* %name, i64 0, metadata !0, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %len, i64 0, metadata !10, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %hash, i64 0, metadata !11, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %flags, i64 0, metadata !12, metadata !{})
+  call void @llvm.dbg.value(metadata i8* %name, i64 0, metadata !0, metadata !DIExpression()), !dbg !DILocation(scope: !1)
+  call void @llvm.dbg.value(metadata i32 %len, i64 0, metadata !10, metadata !DIExpression()), !dbg !DILocation(scope: !1)
+  call void @llvm.dbg.value(metadata i32 %hash, i64 0, metadata !11, metadata !DIExpression()), !dbg !DILocation(scope: !1)
+  call void @llvm.dbg.value(metadata i32 %flags, i64 0, metadata !12, metadata !DIExpression()), !dbg !DILocation(scope: !1)
 ; CHECK:  call fastcc i8* @add_name_internal(i8* %name, i32 %hash) [[NUW:#[0-9]+]], !dbg !{{[0-9]+}}
   %0 = call fastcc i8* @add_name_internal(i8* %name, i32 %len, i32 %hash, i8 zeroext 0, i32 %flags) nounwind, !dbg !13 ; <i8*> [#uses=1]
   ret i8* %0, !dbg !13
@@ -17,11 +17,11 @@ declare void @llvm.dbg.declare(metadata, metadata, metadata) nounwind readnone
 
 define internal fastcc i8* @add_name_internal(i8* %name, i32 %len, i32 %hash, i8 zeroext %extra, i32 %flags) noinline nounwind ssp {
 entry:
-  call void @llvm.dbg.value(metadata i8* %name, i64 0, metadata !15, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %len, i64 0, metadata !20, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %hash, i64 0, metadata !21, metadata !{})
-  call void @llvm.dbg.value(metadata i8 %extra, i64 0, metadata !22, metadata !{})
-  call void @llvm.dbg.value(metadata i32 %flags, i64 0, metadata !23, metadata !{})
+  call void @llvm.dbg.value(metadata i8* %name, i64 0, metadata !15, metadata !DIExpression()), !dbg !DILocation(scope: !16)
+  call void @llvm.dbg.value(metadata i32 %len, i64 0, metadata !20, metadata !DIExpression()), !dbg !DILocation(scope: !16)
+  call void @llvm.dbg.value(metadata i32 %hash, i64 0, metadata !21, metadata !DIExpression()), !dbg !DILocation(scope: !16)
+  call void @llvm.dbg.value(metadata i8 %extra, i64 0, metadata !22, metadata !DIExpression()), !dbg !DILocation(scope: !16)
+  call void @llvm.dbg.value(metadata i32 %flags, i64 0, metadata !23, metadata !DIExpression()), !dbg !DILocation(scope: !16)
   %0 = icmp eq i32 %hash, 0, !dbg !24             ; <i1> [#uses=1]
   br i1 %0, label %bb, label %bb1, !dbg !24
 
@@ -32,7 +32,7 @@ bb1:                                              ; preds = %entry
   br label %bb2, !dbg !27
 
 bb2:                                              ; preds = %bb1, %bb
-  %.0 = phi i8* [ getelementptr inbounds ([1 x i8]* @.str, i64 0, i64 0), %bb ], [ %name, %bb1 ] ; <i8*> [#uses=1]
+  %.0 = phi i8* [ getelementptr inbounds ([1 x i8], [1 x i8]* @.str, i64 0, i64 0), %bb ], [ %name, %bb1 ] ; <i8*> [#uses=1]
   ret i8* %.0, !dbg !27
 }
 
@@ -45,34 +45,34 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 
 !llvm.dbg.cu = !{!3}
 !llvm.module.flags = !{!30}
-!0 = !{!"0x101\00name\008\000", !1, !2, !6} ; [ DW_TAG_arg_variable ]
-!1 = !{!"0x2e\00vfs_addname\00vfs_addname\00vfs_addname\0012\000\001\000\006\000\000\000", !28, !2, !4, null, null, null, null, null} ; [ DW_TAG_subprogram ]
-!2 = !{!"0x29", !28} ; [ DW_TAG_file_type ]
-!3 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build 9999)\001\00\000\00\000", !28, !29, !29, null, null, null} ; [ DW_TAG_compile_unit ]
-!4 = !{!"0x15\00\000\000\000\000\000\000", !28, !2, null, !5, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "name", line: 8, arg: 0, scope: !1, file: !2, type: !6)
+!1 = !DISubprogram(name: "vfs_addname", linkageName: "vfs_addname", line: 12, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, file: !28, scope: !2, type: !4)
+!2 = !DIFile(filename: "tail.c", directory: "/Users/echeng/LLVM/radars/r7927803/")
+!3 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build 9999)", isOptimized: true, emissionKind: 0, file: !28, enums: !29, retainedTypes: !29)
+!4 = !DISubroutineType(types: !5)
 !5 = !{!6, !6, !9, !9, !9}
-!6 = !{!"0xf\00\000\0064\0064\000\000", !28, !2, !7} ; [ DW_TAG_pointer_type ]
-!7 = !{!"0x26\00\000\008\008\000\000", !28, !2, !8} ; [ DW_TAG_const_type ]
-!8 = !{!"0x24\00char\000\008\008\000\000\006", !28, !2} ; [ DW_TAG_base_type ]
-!9 = !{!"0x24\00unsigned int\000\0032\0032\000\000\007", !28, !2} ; [ DW_TAG_base_type ]
-!10 = !{!"0x101\00len\009\000", !1, !2, !9} ; [ DW_TAG_arg_variable ]
-!11 = !{!"0x101\00hash\0010\000", !1, !2, !9} ; [ DW_TAG_arg_variable ]
-!12 = !{!"0x101\00flags\0011\000", !1, !2, !9} ; [ DW_TAG_arg_variable ]
-!13 = !MDLocation(line: 13, scope: !14)
-!14 = !{!"0xb\0012\000\000", !28, !1} ; [ DW_TAG_lexical_block ]
-!15 = !{!"0x101\00name\0017\000", !16, !2, !6} ; [ DW_TAG_arg_variable ]
-!16 = !{!"0x2e\00add_name_internal\00add_name_internal\00add_name_internal\0022\001\001\000\006\000\000\000", !28, !2, !17, null, null, null, null, null} ; [ DW_TAG_subprogram ]
-!17 = !{!"0x15\00\000\000\000\000\000\000", !28, !2, null, !18, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!6 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, file: !28, scope: !2, baseType: !7)
+!7 = !DIDerivedType(tag: DW_TAG_const_type, size: 8, align: 8, file: !28, scope: !2, baseType: !8)
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "char", size: 8, align: 8, encoding: DW_ATE_signed_char)
+!9 = !DIBasicType(tag: DW_TAG_base_type, name: "unsigned int", size: 32, align: 32, encoding: DW_ATE_unsigned)
+!10 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "len", line: 9, arg: 0, scope: !1, file: !2, type: !9)
+!11 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "hash", line: 10, arg: 0, scope: !1, file: !2, type: !9)
+!12 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "flags", line: 11, arg: 0, scope: !1, file: !2, type: !9)
+!13 = !DILocation(line: 13, scope: !14)
+!14 = distinct !DILexicalBlock(line: 12, column: 0, file: !28, scope: !1)
+!15 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "name", line: 17, arg: 0, scope: !16, file: !2, type: !6)
+!16 = !DISubprogram(name: "add_name_internal", linkageName: "add_name_internal", line: 22, isLocal: true, isDefinition: true, virtualIndex: 6, isOptimized: false, file: !28, scope: !2, type: !17)
+!17 = !DISubroutineType(types: !18)
 !18 = !{!6, !6, !9, !9, !19, !9}
-!19 = !{!"0x24\00unsigned char\000\008\008\000\000\008", !28, !2} ; [ DW_TAG_base_type ]
-!20 = !{!"0x101\00len\0018\000", !16, !2, !9} ; [ DW_TAG_arg_variable ]
-!21 = !{!"0x101\00hash\0019\000", !16, !2, !9} ; [ DW_TAG_arg_variable ]
-!22 = !{!"0x101\00extra\0020\000", !16, !2, !19} ; [ DW_TAG_arg_variable ]
-!23 = !{!"0x101\00flags\0021\000", !16, !2, !9} ; [ DW_TAG_arg_variable ]
-!24 = !MDLocation(line: 23, scope: !25)
-!25 = !{!"0xb\0022\000\000", !28, !16} ; [ DW_TAG_lexical_block ]
-!26 = !MDLocation(line: 24, scope: !25)
-!27 = !MDLocation(line: 26, scope: !25)
-!28 = !{!"tail.c", !"/Users/echeng/LLVM/radars/r7927803/"}
-!29 = !{i32 0}
-!30 = !{i32 1, !"Debug Info Version", i32 2}
+!19 = !DIBasicType(tag: DW_TAG_base_type, name: "unsigned char", size: 8, align: 8, encoding: DW_ATE_unsigned_char)
+!20 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "len", line: 18, arg: 0, scope: !16, file: !2, type: !9)
+!21 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "hash", line: 19, arg: 0, scope: !16, file: !2, type: !9)
+!22 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "extra", line: 20, arg: 0, scope: !16, file: !2, type: !19)
+!23 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "flags", line: 21, arg: 0, scope: !16, file: !2, type: !9)
+!24 = !DILocation(line: 23, scope: !25)
+!25 = distinct !DILexicalBlock(line: 22, column: 0, file: !28, scope: !16)
+!26 = !DILocation(line: 24, scope: !25)
+!27 = !DILocation(line: 26, scope: !25)
+!28 = !DIFile(filename: "tail.c", directory: "/Users/echeng/LLVM/radars/r7927803/")
+!29 = !{}
+!30 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/DeadArgElim/2013-05-17-VarargsAndBlockAddress.ll b/test/Transforms/DeadArgElim/2013-05-17-VarargsAndBlockAddress.ll
index 23216031b58a..7552a12142b5 100644
--- a/test/Transforms/DeadArgElim/2013-05-17-VarargsAndBlockAddress.ll
+++ b/test/Transforms/DeadArgElim/2013-05-17-VarargsAndBlockAddress.ll
@@ -19,7 +19,7 @@ l2:
 ; CHECK: define internal i32 @varargs_func(i8* %addr) {
 
 define i32 @caller(i8* %addr) {
-  %r = call i32 (i8*, ...)* @varargs_func(i8* %addr)
+  %r = call i32 (i8*, ...) @varargs_func(i8* %addr)
   ret i32 %r
 }
 ; CHECK: %r = call i32 @varargs_func(i8* %addr)
diff --git a/test/Transforms/DeadArgElim/aggregates.ll b/test/Transforms/DeadArgElim/aggregates.ll
new file mode 100644
index 000000000000..68d253425587
--- /dev/null
+++ b/test/Transforms/DeadArgElim/aggregates.ll
@@ -0,0 +1,162 @@
+; RUN: opt -S -deadargelim %s | FileCheck %s
+
+; Case 0: the basic example: an entire aggregate use is returned, but it's
+; actually only used in ways we can eliminate. We gain benefit from analysing
+; the "use" and applying its results to all sub-values.
+
+; CHECK-LABEL: define internal void @agguse_dead()
+
+define internal { i32, i32 } @agguse_dead() {
+  ret { i32, i32 } { i32 0, i32 1 }
+}
+
+define internal { i32, i32 } @test_agguse_dead() {
+  %val = call { i32, i32 } @agguse_dead()
+  ret { i32, i32 } %val
+}
+
+
+
+; Case 1: an opaque use of the aggregate exists (in this case dead). Otherwise
+; only one value is used, so function can be simplified.
+
+; CHECK-LABEL: define internal i32 @rets_independent_if_agguse_dead()
+; CHECK: [[RET:%.*]] = extractvalue { i32, i32 } { i32 0, i32 1 }, 1
+; CHECK: ret i32 [[RET]]
+
+define internal { i32, i32 } @rets_independent_if_agguse_dead() {
+  ret { i32, i32 } { i32 0, i32 1 }
+}
+
+define internal { i32, i32 } @test_rets_independent_if_agguse_dead(i1 %tst) {
+  %val = call { i32, i32 } @rets_independent_if_agguse_dead()
+  br i1 %tst, label %use_1, label %use_aggregate
+
+use_1:
+  ; This use can be classified as applying only to ret 1.
+  %val0 = extractvalue { i32, i32 } %val, 1
+  call void @callee(i32 %val0)
+  ret { i32, i32 } undef
+
+use_aggregate:
+  ; This use is assumed to apply to both 0 and 1.
+  ret { i32, i32 } %val
+}
+
+; Case 2: an opaque use of the aggregate exists (in this case *live*). Other
+; uses shouldn't matter.
+
+; CHECK-LABEL: define internal { i32, i32 } @rets_live_agguse()
+; CHECK: ret { i32, i32 } { i32 0, i32 1 }
+
+define internal { i32, i32 } @rets_live_agguse() {
+  ret { i32, i32} { i32 0, i32 1 }
+}
+
+define { i32, i32 } @test_rets_live_aggues(i1 %tst) {
+  %val = call { i32, i32 } @rets_live_agguse()
+  br i1 %tst, label %use_1, label %use_aggregate
+
+use_1:
+  ; This use can be classified as applying only to ret 1.
+  %val0 = extractvalue { i32, i32 } %val, 1
+  call void @callee(i32 %val0)
+  ret { i32, i32 } undef
+
+use_aggregate:
+  ; This use is assumed to apply to both 0 and 1.
+  ret { i32, i32 } %val
+}
+
+declare void @callee(i32)
+
+; Case 3: the insertvalue meant %in was live if ret-slot-1 was, but we were only
+; tracking multiple ret-slots for struct types. So %in was eliminated
+; incorrectly.
+
+; CHECK-LABEL: define internal [2 x i32] @array_rets_have_multiple_slots(i32 %in)
+
+define internal [2 x i32] @array_rets_have_multiple_slots(i32 %in) {
+  %ret = insertvalue [2 x i32] undef, i32 %in, 1
+  ret [2 x i32] %ret
+}
+
+define [2 x i32] @test_array_rets_have_multiple_slots() {
+  %res = call [2 x i32] @array_rets_have_multiple_slots(i32 42)
+  ret [2 x i32] %res
+}
+
+; Case 4: we can remove some retvals from the array. It's nice to produce an
+; array again having done so (rather than converting it to a struct).
+
+; CHECK-LABEL: define internal [2 x i32] @can_shrink_arrays()
+; CHECK: [[VAL0:%.*]] = extractvalue [3 x i32] [i32 42, i32 43, i32 44], 0
+; CHECK: [[RESTMP:%.*]] = insertvalue [2 x i32] undef, i32 [[VAL0]], 0
+; CHECK: [[VAL2:%.*]] = extractvalue [3 x i32] [i32 42, i32 43, i32 44], 2
+; CHECK: [[RES:%.*]] = insertvalue [2 x i32] [[RESTMP]], i32 [[VAL2]], 1
+; CHECK: ret [2 x i32] [[RES]]
+
+; CHECK-LABEL: define void @test_can_shrink_arrays()
+
+define internal [3 x i32] @can_shrink_arrays() {
+  ret [3 x i32] [i32 42, i32 43, i32 44]
+}
+
+define void @test_can_shrink_arrays() {
+  %res = call [3 x i32] @can_shrink_arrays()
+
+  %res.0 = extractvalue [3 x i32] %res, 0
+  call void @callee(i32 %res.0)
+
+  %res.2 = extractvalue [3 x i32] %res, 2
+  call void @callee(i32 %res.2)
+
+  ret void
+}
+
+; Case 5: %in gets passed directly to the return. It should mark be marked as
+; used if *any* of the return values are, not just if value 0 is.
+
+; CHECK-LABEL: define internal i32 @ret_applies_to_all({ i32, i32 } %in)
+; CHECK: [[RET:%.*]] = extractvalue { i32, i32 } %in, 1
+; CHECK: ret i32 [[RET]]
+
+define internal {i32, i32} @ret_applies_to_all({i32, i32} %in) {
+  ret {i32, i32} %in
+}
+
+define i32 @test_ret_applies_to_all() {
+  %val = call {i32, i32} @ret_applies_to_all({i32, i32} {i32 42, i32 43})
+  %ret = extractvalue {i32, i32} %val, 1
+  ret i32 %ret
+}
+
+; Case 6: When considering @mid, the return instruciton has sub-value 0
+; unconditionally live, but 1 only conditionally live. Since at that level we're
+; applying the results to the whole of %res, this means %res is live and cannot
+; be reduced. There is scope for further optimisation here (though not visible
+; in this test-case).
+
+; CHECK-LABEL: define internal { i8*, i32 } @inner()
+
+define internal {i8*, i32} @mid() {
+  %res = call {i8*, i32} @inner()
+  %intval = extractvalue {i8*, i32} %res, 1
+  %tst = icmp eq i32 %intval, 42
+  br i1 %tst, label %true, label %true
+
+true:
+  ret {i8*, i32} %res
+}
+
+define internal {i8*, i32} @inner() {
+  ret {i8*, i32} {i8* null, i32 42}
+}
+
+define internal i8 @outer() {
+  %res = call {i8*, i32} @mid()
+  %resptr = extractvalue {i8*, i32} %res, 0
+
+  %val = load i8, i8* %resptr
+  ret i8 %val
+}
\ No newline at end of file
diff --git a/test/Transforms/DeadArgElim/dbginfo.ll b/test/Transforms/DeadArgElim/dbginfo.ll
index 5bbf821ef213..efafd9bbc9e2 100644
--- a/test/Transforms/DeadArgElim/dbginfo.ll
+++ b/test/Transforms/DeadArgElim/dbginfo.ll
@@ -14,7 +14,7 @@
 ; the function->debug info mapping on update to ensure it's accurate when used
 ; again for the next removal.
 
-; CHECK: void ()* @_ZL2f1iz, {{.*}} ; [ DW_TAG_subprogram ] {{.*}} [f1]
+; CHECK: !DISubprogram(name: "f1",{{.*}} function: void ()* @_ZL2f1iz
 
 ; Check that debug info metadata for subprograms stores pointers to
 ; updated LLVM functions.
@@ -22,7 +22,7 @@
 ; Function Attrs: uwtable
 define void @_Z2f2v() #0 {
 entry:
-  call void (i32, ...)* @_ZL2f1iz(i32 1), !dbg !15
+  call void (i32, ...) @_ZL2f1iz(i32 1), !dbg !15
   ret void, !dbg !16
 }
 
@@ -47,24 +47,24 @@ attributes #2 = { nounwind readnone }
 !llvm.module.flags = !{!12, !13}
 !llvm.ident = !{!14}
 
-!0 = !{!"0x11\004\00clang version 3.6.0 \000\00\000\00\001", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [/tmp/dbginfo/dbg.cpp] [DW_LANG_C_plus_plus]
-!1 = !{!"dbg.cpp", !"/tmp/dbginfo"}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.6.0 ", isOptimized: false, emissionKind: 1, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "dbg.cpp", directory: "/tmp/dbginfo")
 !2 = !{}
 !3 = !{!4, !8}
-!4 = !{!"0x2e\00f2\00f2\00_Z2f2v\004\000\001\000\000\00256\000\004", !1, !5, !6, null, void ()* @_Z2f2v, null, null, !2} ; [ DW_TAG_subprogram ] [line 4] [def] [f2]
-!5 = !{!"0x29", !1}    ; [ DW_TAG_file_type ] [/tmp/dbginfo/dbg.cpp]
-!6 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "f2", linkageName: "_Z2f2v", line: 4, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 4, file: !1, scope: !5, type: !6, function: void ()* @_Z2f2v, variables: !2)
+!5 = !DIFile(filename: "dbg.cpp", directory: "/tmp/dbginfo")
+!6 = !DISubroutineType(types: !7)
 !7 = !{null}
-!8 = !{!"0x2e\00f1\00f1\00_ZL2f1iz\001\001\001\000\000\00256\000\001", !1, !5, !9, null, void (i32, ...)* @_ZL2f1iz, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [local] [def] [f1]
-!9 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !10, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!8 = !DISubprogram(name: "f1", linkageName: "_ZL2f1iz", line: 1, isLocal: true, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !9, function: void (i32, ...)* @_ZL2f1iz, variables: !2)
+!9 = !DISubroutineType(types: !10)
 !10 = !{null, !11, null}
-!11 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !12 = !{i32 2, !"Dwarf Version", i32 4}
-!13 = !{i32 2, !"Debug Info Version", i32 2}
+!13 = !{i32 2, !"Debug Info Version", i32 3}
 !14 = !{!"clang version 3.6.0 "}
-!15 = !MDLocation(line: 5, column: 3, scope: !4)
-!16 = !MDLocation(line: 6, column: 1, scope: !4)
-!17 = !{!"0x101\00\0016777217\000", !8, !5, !11} ; [ DW_TAG_arg_variable ] [line 1]
-!18 = !{!"0x102"}               ; [ DW_TAG_expression ]
-!19 = !MDLocation(line: 1, column: 19, scope: !8)
-!20 = !MDLocation(line: 2, column: 1, scope: !8)
+!15 = !DILocation(line: 5, column: 3, scope: !4)
+!16 = !DILocation(line: 6, column: 1, scope: !4)
+!17 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "", line: 1, arg: 1, scope: !8, file: !5, type: !11)
+!18 = !DIExpression()
+!19 = !DILocation(line: 1, column: 19, scope: !8)
+!20 = !DILocation(line: 2, column: 1, scope: !8)
diff --git a/test/Transforms/DeadArgElim/dead_vaargs.ll b/test/Transforms/DeadArgElim/dead_vaargs.ll
index c8189c66b48c..375415948af1 100644
--- a/test/Transforms/DeadArgElim/dead_vaargs.ll
+++ b/test/Transforms/DeadArgElim/dead_vaargs.ll
@@ -1,25 +1,25 @@
 ; RUN: opt < %s -deadargelim -S | FileCheck %s
 
 define i32 @bar(i32 %A) {
-  call void (i32, ...)* @thunk(i32 %A, i64 47, double 1.000000e+00)
-  %a = call i32 (i32, ...)* @has_vastart(i32 %A, i64 47, double 1.000000e+00)
-  %b = call i32 (i32, ...)* @no_vastart( i32 %A, i32 %A, i32 %A, i32 %A, i64 47, double 1.000000e+00 )
+  call void (i32, ...) @thunk(i32 %A, i64 47, double 1.000000e+00)
+  %a = call i32 (i32, ...) @has_vastart(i32 %A, i64 47, double 1.000000e+00)
+  %b = call i32 (i32, ...) @no_vastart( i32 %A, i32 %A, i32 %A, i32 %A, i64 47, double 1.000000e+00 )
   %c = add i32 %a, %b
   ret i32 %c
 }
 ; CHECK-LABEL: define i32 @bar
-; CHECK: call void (i32, ...)* @thunk(i32 %A, i64 47, double 1.000000e+00)
-; CHECK: call i32 (i32, ...)* @has_vastart(i32 %A, i64 47, double 1.000000e+00)
+; CHECK: call void (i32, ...) @thunk(i32 %A, i64 47, double 1.000000e+00)
+; CHECK: call i32 (i32, ...) @has_vastart(i32 %A, i64 47, double 1.000000e+00)
 ; CHECK: call i32 @no_vastart(i32 %A)
 
 declare void @thunk_target(i32 %X, ...)
 
 define internal void @thunk(i32 %X, ...) {
-  musttail call void(i32, ...)* @thunk_target(i32 %X, ...)
+  musttail call void(i32, ...) @thunk_target(i32 %X, ...)
   ret void
 }
 ; CHECK-LABEL: define internal void @thunk(i32 %X, ...)
-; CHECK: musttail call void (i32, ...)* @thunk_target(i32 %X, ...)
+; CHECK: musttail call void (i32, ...) @thunk_target(i32 %X, ...)
 
 define internal i32 @has_vastart(i32 %X, ...) {
   %valist = alloca i8
diff --git a/test/Transforms/DeadArgElim/deadexternal.ll b/test/Transforms/DeadArgElim/deadexternal.ll
index 665d7dbf4937..21cbc37a3f00 100644
--- a/test/Transforms/DeadArgElim/deadexternal.ll
+++ b/test/Transforms/DeadArgElim/deadexternal.ll
@@ -31,9 +31,9 @@ define void @h() {
 entry:
   %i = alloca i32, align 4
   store volatile i32 10, i32* %i, align 4
-; CHECK: %tmp = load volatile i32* %i, align 4
+; CHECK: %tmp = load volatile i32, i32* %i, align 4
 ; CHECK-NEXT: call void @f(i32 undef)
-  %tmp = load volatile i32* %i, align 4
+  %tmp = load volatile i32, i32* %i, align 4
   call void @f(i32 %tmp)
   ret void
 }
diff --git a/test/Transforms/DeadArgElim/deadretval2.ll b/test/Transforms/DeadArgElim/deadretval2.ll
index dcdc36e319a3..b0d2428fbdc0 100644
--- a/test/Transforms/DeadArgElim/deadretval2.ll
+++ b/test/Transforms/DeadArgElim/deadretval2.ll
@@ -20,7 +20,7 @@ define void @test3(i32 %X) {
 }
 
 define internal i32 @foo() {
-        %DEAD = load i32* @P            ; <i32> [#uses=1]
+        %DEAD = load i32, i32* @P            ; <i32> [#uses=1]
         ret i32 %DEAD
 }
 
diff --git a/test/Transforms/DeadArgElim/keepalive.ll b/test/Transforms/DeadArgElim/keepalive.ll
index 16569db4d38f..d8a09933dcf9 100644
--- a/test/Transforms/DeadArgElim/keepalive.ll
+++ b/test/Transforms/DeadArgElim/keepalive.ll
@@ -31,7 +31,7 @@ define void @caller() {
 ; We can't remove 'this' here, as that would put argmem in ecx instead of
 ; memory.
 define internal x86_thiscallcc i32 @unused_this(i32* %this, i32* inalloca %argmem) {
-	%v = load i32* %argmem
+	%v = load i32, i32* %argmem
 	ret i32 %v
 }
 ; CHECK-LABEL: define internal x86_thiscallcc i32 @unused_this(i32* %this, i32* inalloca %argmem)
diff --git a/test/Transforms/DeadArgElim/variadic_safety.ll b/test/Transforms/DeadArgElim/variadic_safety.ll
index 15f57bcfdcb8..2dac2f9180ed 100644
--- a/test/Transforms/DeadArgElim/variadic_safety.ll
+++ b/test/Transforms/DeadArgElim/variadic_safety.ll
@@ -17,9 +17,9 @@ define internal i32 @va_func(i32 %a, i32 %b, ...) {
 define i32 @call_va(i32 %in) {
   %stacked = alloca i32
   store i32 42, i32* %stacked
-  %res = call i32(i32, i32, ...)* @va_func(i32 %in, i32 %in, [6 x i32] undef, i32* byval %stacked)
+  %res = call i32(i32, i32, ...) @va_func(i32 %in, i32 %in, [6 x i32] undef, i32* byval %stacked)
   ret i32 %res
-; CHECK: call i32 (i32, i32, ...)* @va_func(i32 undef, i32 %in, [6 x i32] undef, i32* byval %stacked)
+; CHECK: call i32 (i32, i32, ...) @va_func(i32 undef, i32 %in, [6 x i32] undef, i32* byval %stacked)
 }
 
 define internal i32 @va_deadret_func(i32 %a, i32 %b, ...) {
@@ -32,7 +32,7 @@ define internal i32 @va_deadret_func(i32 %a, i32 %b, ...) {
 define void @call_deadret(i32 %in) {
   %stacked = alloca i32
   store i32 42, i32* %stacked
-  call i32 (i32, i32, ...)* @va_deadret_func(i32 undef, i32 %in, [6 x i32] undef, i32* byval %stacked)
+  call i32 (i32, i32, ...) @va_deadret_func(i32 undef, i32 %in, [6 x i32] undef, i32* byval %stacked)
   ret void
-; CHECK: call void (i32, i32, ...)* @va_deadret_func(i32 undef, i32 undef, [6 x i32] undef, i32* byval %stacked)
+; CHECK: call void (i32, i32, ...) @va_deadret_func(i32 undef, i32 undef, [6 x i32] undef, i32* byval %stacked)
 }
diff --git a/test/Transforms/DeadStoreElimination/2011-03-25-DSEMiscompile.ll b/test/Transforms/DeadStoreElimination/2011-03-25-DSEMiscompile.ll
index 39d53584dc37..7746ccee5218 100644
--- a/test/Transforms/DeadStoreElimination/2011-03-25-DSEMiscompile.ll
+++ b/test/Transforms/DeadStoreElimination/2011-03-25-DSEMiscompile.ll
@@ -11,10 +11,10 @@ define ghccc void @Func1(i32* noalias %Arg1, i32* noalias %Arg2, i32* %Arg3, i32
 entry:
   store i32 add (i32 ptrtoint ([0 x i32]* @A to i32), i32 1), i32* %Arg2
 ; CHECK: store i32 add (i32 ptrtoint ([0 x i32]* @A to i32), i32 1), i32* %Arg2
-  %ln2gz = getelementptr i32* %Arg1, i32 14
+  %ln2gz = getelementptr i32, i32* %Arg1, i32 14
   %ln2gA = bitcast i32* %ln2gz to double*
-  %ln2gB = load double* %ln2gA
-  %ln2gD = getelementptr i32* %Arg2, i32 -3
+  %ln2gB = load double, double* %ln2gA
+  %ln2gD = getelementptr i32, i32* %Arg2, i32 -3
   %ln2gE = bitcast i32* %ln2gD to double*
   store double %ln2gB, double* %ln2gE
 ; CHECK: store double %ln2gB, double* %ln2gE
diff --git a/test/Transforms/DeadStoreElimination/2011-09-06-EndOfFunction.ll b/test/Transforms/DeadStoreElimination/2011-09-06-EndOfFunction.ll
index 95253f6570b5..7e46d28a9c47 100644
--- a/test/Transforms/DeadStoreElimination/2011-09-06-EndOfFunction.ll
+++ b/test/Transforms/DeadStoreElimination/2011-09-06-EndOfFunction.ll
@@ -10,11 +10,11 @@ define void @_Z3foov(%"class.std::auto_ptr"* noalias nocapture sret %agg.result)
 _ZNSt8auto_ptrIiED1Ev.exit:
   %temp.lvalue = alloca %"class.std::auto_ptr", align 8
   call void @_Z3barv(%"class.std::auto_ptr"* sret %temp.lvalue)
-  %_M_ptr.i.i = getelementptr inbounds %"class.std::auto_ptr"* %temp.lvalue, i64 0, i32 0
-  %tmp.i.i = load i32** %_M_ptr.i.i, align 8
+  %_M_ptr.i.i = getelementptr inbounds %"class.std::auto_ptr", %"class.std::auto_ptr"* %temp.lvalue, i64 0, i32 0
+  %tmp.i.i = load i32*, i32** %_M_ptr.i.i, align 8
 ; CHECK-NOT: store i32* null
   store i32* null, i32** %_M_ptr.i.i, align 8
-  %_M_ptr.i.i4 = getelementptr inbounds %"class.std::auto_ptr"* %agg.result, i64 0, i32 0
+  %_M_ptr.i.i4 = getelementptr inbounds %"class.std::auto_ptr", %"class.std::auto_ptr"* %agg.result, i64 0, i32 0
   store i32* %tmp.i.i, i32** %_M_ptr.i.i4, align 8
 ; CHECK: ret void
   ret void
diff --git a/test/Transforms/DeadStoreElimination/2011-09-06-MemCpy.ll b/test/Transforms/DeadStoreElimination/2011-09-06-MemCpy.ll
index 22b87864c063..d30e9a2e6c1f 100644
--- a/test/Transforms/DeadStoreElimination/2011-09-06-MemCpy.ll
+++ b/test/Transforms/DeadStoreElimination/2011-09-06-MemCpy.ll
@@ -68,13 +68,13 @@ declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32,
 define void @_ZSt9iter_swapIPSt4pairIPN4llvm10BasicBlockEjES5_EvT_T0_(%struct.pair.162* %__a, %struct.pair.162* %__b) nounwind uwtable inlinehint {
 entry:
   %memtmp = alloca %struct.pair.162, align 8
-  %0 = getelementptr inbounds %struct.pair.162* %memtmp, i64 0, i32 0
-  %1 = getelementptr inbounds %struct.pair.162* %__a, i64 0, i32 0
-  %2 = load %struct.BasicBlock** %1, align 8
+  %0 = getelementptr inbounds %struct.pair.162, %struct.pair.162* %memtmp, i64 0, i32 0
+  %1 = getelementptr inbounds %struct.pair.162, %struct.pair.162* %__a, i64 0, i32 0
+  %2 = load %struct.BasicBlock*, %struct.BasicBlock** %1, align 8
   store %struct.BasicBlock* %2, %struct.BasicBlock** %0, align 8
-  %3 = getelementptr inbounds %struct.pair.162* %memtmp, i64 0, i32 1
-  %4 = getelementptr inbounds %struct.pair.162* %__a, i64 0, i32 1
-  %5 = load i32* %4, align 4
+  %3 = getelementptr inbounds %struct.pair.162, %struct.pair.162* %memtmp, i64 0, i32 1
+  %4 = getelementptr inbounds %struct.pair.162, %struct.pair.162* %__a, i64 0, i32 1
+  %5 = load i32, i32* %4, align 4
   store i32 %5, i32* %3, align 8
   %6 = bitcast %struct.pair.162* %__a to i8*
   %7 = bitcast %struct.pair.162* %__b to i8*
diff --git a/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll b/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll
index 968d608c67ed..de7a4ccd867c 100644
--- a/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll
+++ b/test/Transforms/DeadStoreElimination/OverwriteStoreEnd.ll
@@ -10,11 +10,11 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define void @write24to28(i32* nocapture %p) nounwind uwtable ssp {
 ; CHECK-LABEL: @write24to28(
 entry:
-  %arrayidx0 = getelementptr inbounds i32* %p, i64 1
+  %arrayidx0 = getelementptr inbounds i32, i32* %p, i64 1
   %p3 = bitcast i32* %arrayidx0 to i8*
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 24, i32 4, i1 false)
   call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 28, i32 4, i1 false)
-  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  %arrayidx1 = getelementptr inbounds i32, i32* %p, i64 7
   store i32 1, i32* %arrayidx1, align 4
   ret void
 }
@@ -25,7 +25,7 @@ entry:
   %p3 = bitcast i32* %p to i8*
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 28, i32 4, i1 false)
   call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 4, i1 false)
-  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  %arrayidx1 = getelementptr inbounds i32, i32* %p, i64 7
   store i32 1, i32* %arrayidx1, align 4
   ret void
 }
@@ -36,7 +36,7 @@ entry:
   %p3 = bitcast i32* %p to i8*
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 16, i1 false)
   call void @llvm.memset.p0i8.i64(i8* %p3, i8 0, i64 32, i32 16, i1 false)
-  %arrayidx1 = getelementptr inbounds i32* %p, i64 7
+  %arrayidx1 = getelementptr inbounds i32, i32* %p, i64 7
   store i32 1, i32* %arrayidx1, align 4
   ret void
 }
@@ -47,7 +47,7 @@ entry:
   %0 = bitcast %struct.vec2plusi* %p to i8*
 ; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2plusi* @glob2 to i8*), i64 32, i32 16, i1 false)
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2plusi* @glob2 to i8*), i64 36, i32 16, i1 false)
-  %c = getelementptr inbounds %struct.vec2plusi* %p, i64 0, i32 2
+  %c = getelementptr inbounds %struct.vec2plusi, %struct.vec2plusi* %p, i64 0, i32 2
   store i32 1, i32* %c, align 4
   ret void
 }
@@ -58,7 +58,7 @@ entry:
   %0 = bitcast %struct.vec2* %p to i8*
 ; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 16, i32 16, i1 false)
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
-  %c = getelementptr inbounds %struct.vec2* %p, i64 0, i32 1
+  %c = getelementptr inbounds %struct.vec2, %struct.vec2* %p, i64 0, i32 1
   store <4 x i32> <i32 1, i32 2, i32 3, i32 4>, <4 x i32>* %c, align 4
   ret void
 }
@@ -69,7 +69,7 @@ entry:
   %0 = bitcast %struct.vec2* %p to i8*
 ; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* bitcast (%struct.vec2* @glob1 to i8*), i64 32, i32 16, i1 false)
-  %arrayidx1 = getelementptr inbounds %struct.vec2* %p, i64 0, i32 0, i64 7
+  %arrayidx1 = getelementptr inbounds %struct.vec2, %struct.vec2* %p, i64 0, i32 0, i64 7
   store i32 1, i32* %arrayidx1, align 4
   ret void
 }
@@ -84,12 +84,12 @@ declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 define void @cpu_lwp_fork(%struct.trapframe* %md_regs, i64 %pcb_rsp0) nounwind uwtable noinline ssp {
 entry:
   %0 = inttoptr i64 %pcb_rsp0 to %struct.trapframe*
-  %add.ptr = getelementptr inbounds %struct.trapframe* %0, i64 -1
+  %add.ptr = getelementptr inbounds %struct.trapframe, %struct.trapframe* %0, i64 -1
   %1 = bitcast %struct.trapframe* %add.ptr to i8*
   %2 = bitcast %struct.trapframe* %md_regs to i8*
 ; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 24, i32 1, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %1, i8* %2, i64 24, i32 1, i1 false)
-  %tf_trapno = getelementptr inbounds %struct.trapframe* %0, i64 -1, i32 1
+  %tf_trapno = getelementptr inbounds %struct.trapframe, %struct.trapframe* %0, i64 -1, i32 1
   store i64 3, i64* %tf_trapno, align 8
   ret void
 }
diff --git a/test/Transforms/DeadStoreElimination/PartialStore.ll b/test/Transforms/DeadStoreElimination/PartialStore.ll
index 80c2bfae846e..d85b4de0ab98 100644
--- a/test/Transforms/DeadStoreElimination/PartialStore.ll
+++ b/test/Transforms/DeadStoreElimination/PartialStore.ll
@@ -30,8 +30,8 @@ define i32 @test3(double %__x) {
   %__u = alloca { [3 x i32] }
   %tmp.1 = bitcast { [3 x i32] }* %__u to double*
   store double %__x, double* %tmp.1
-  %tmp.4 = getelementptr { [3 x i32] }* %__u, i32 0, i32 0, i32 1
-  %tmp.5 = load i32* %tmp.4
+  %tmp.4 = getelementptr { [3 x i32] }, { [3 x i32] }* %__u, i32 0, i32 0, i32 1
+  %tmp.5 = load i32, i32* %tmp.4
   %tmp.6 = icmp slt i32 %tmp.5, 0
   %tmp.7 = zext i1 %tmp.6 to i32
   ret i32 %tmp.7
@@ -44,7 +44,7 @@ define void @test4(i8* %P) {
 ; CHECK-NEXT: store double
 
   store i8 19, i8* %P  ;; dead
-  %A = getelementptr i8* %P, i32 3
+  %A = getelementptr i8, i8* %P, i32 3
 
   store i8 42, i8* %A  ;; dead
 
@@ -58,7 +58,7 @@ declare void @test5a(i32*)
 define void @test5(i32 %i) nounwind ssp {
   %A = alloca i32
   %B = bitcast i32* %A to i8*
-  %C = getelementptr i8* %B, i32 %i
+  %C = getelementptr i8, i8* %B, i32 %i
   store i8 10, i8* %C        ;; Dead store to variable index.
   store i32 20, i32* %A
 
@@ -74,7 +74,7 @@ declare void @test5a_as1(i32*)
 define void @test5_addrspacecast(i32 %i) nounwind ssp {
   %A = alloca i32
   %B = addrspacecast i32* %A to i8 addrspace(1)*
-  %C = getelementptr i8 addrspace(1)* %B, i32 %i
+  %C = getelementptr i8, i8 addrspace(1)* %B, i32 %i
   store i8 10, i8 addrspace(1)* %C        ;; Dead store to variable index.
   store i32 20, i32* %A
 
diff --git a/test/Transforms/DeadStoreElimination/atomic.ll b/test/Transforms/DeadStoreElimination/atomic.ll
index af303fa983ee..79beee803e4e 100644
--- a/test/Transforms/DeadStoreElimination/atomic.ll
+++ b/test/Transforms/DeadStoreElimination/atomic.ll
@@ -23,28 +23,6 @@ define void @test1() {
   ret void
 }
 
-; DSE across seq_cst load (allowed)
-define i32 @test2() {
-; CHECK-LABEL: test2
-; CHECK-NOT: store i32 0
-; CHECK: store i32 1
-  store i32 0, i32* @x
-  %x = load atomic i32* @y seq_cst, align 4
-  store i32 1, i32* @x
-  ret i32 %x
-}
-
-; DSE across seq_cst store (allowed)
-define void @test3() {
-; CHECK-LABEL: test3
-; CHECK-NOT: store i32 0
-; CHECK: store atomic i32 2
-  store i32 0, i32* @x
-  store atomic i32 2, i32* @y seq_cst, align 4
-  store i32 1, i32* @x
-  ret void
-}
-
 ; DSE remove unordered store (allowed)
 define void @test4() {
 ; CHECK-LABEL: test4
@@ -69,7 +47,7 @@ define void @test6() {
 ; CHECK-LABEL: test6
 ; CHECK-NOT: store
 ; CHECK: ret void
-  %x = load atomic i32* @x unordered, align 4
+  %x = load atomic i32, i32* @x unordered, align 4
   store atomic i32 %x, i32* @x unordered, align 4
   ret void
 }
@@ -93,7 +71,7 @@ define i32 @test8() {
   %a = alloca i32
   call void @randomop(i32* %a)
   store i32 0, i32* %a, align 4
-  %x = load atomic i32* @x seq_cst, align 4
+  %x = load atomic i32, i32* @x seq_cst, align 4
   ret i32 %x
 }
 
@@ -103,7 +81,7 @@ define i32 @test9() {
 ; CHECK-NOT: store i32 0
 ; CHECK: store i32 1
   store i32 0, i32* @x
-  %x = load atomic i32* @y monotonic, align 4
+  %x = load atomic i32, i32* @y monotonic, align 4
   store i32 1, i32* @x
   ret i32 %x
 }
@@ -125,7 +103,7 @@ define i32 @test11() {
 ; CHECK: store atomic i32 0
 ; CHECK: store atomic i32 1
   store atomic i32 0, i32* @x monotonic, align 4
-  %x = load atomic i32* @y monotonic, align 4
+  %x = load atomic i32, i32* @y monotonic, align 4
   store atomic i32 1, i32* @x monotonic, align 4
   ret i32 %x
 }
@@ -141,30 +119,6 @@ define void @test12() {
   ret void
 }
 
-; DSE is allowed across a pair of an atomic read and then write.
-define i32 @test13() {
-; CHECK-LABEL: test13
-; CHECK-NOT: store i32 0
-; CHECK: store i32 1
-  store i32 0, i32* @x
-  %x = load atomic i32* @y seq_cst, align 4
-  store atomic i32 %x, i32* @y seq_cst, align 4
-  store i32 1, i32* @x
-  ret i32 %x
-}
-
-; Same if it is acquire-release instead of seq_cst/seq_cst
-define i32 @test14() {
-; CHECK-LABEL: test14
-; CHECK-NOT: store i32 0
-; CHECK: store i32 1
-  store i32 0, i32* @x
-  %x = load atomic i32* @y acquire, align 4
-  store atomic i32 %x, i32* @y release, align 4
-  store i32 1, i32* @x
-  ret i32 %x
-}
-
 ; But DSE is not allowed across a release-acquire pair.
 define i32 @test15() {
 ; CHECK-LABEL: test15
@@ -172,7 +126,7 @@ define i32 @test15() {
 ; CHECK: store i32 1
   store i32 0, i32* @x
   store atomic i32 0, i32* @y release, align 4
-  %x = load atomic i32* @y acquire, align 4
+  %x = load atomic i32, i32* @y acquire, align 4
   store i32 1, i32* @x
   ret i32 %x
 }
diff --git a/test/Transforms/DeadStoreElimination/const-pointers.ll b/test/Transforms/DeadStoreElimination/const-pointers.ll
index 3e772d79c70a..e4403edbed20 100644
--- a/test/Transforms/DeadStoreElimination/const-pointers.ll
+++ b/test/Transforms/DeadStoreElimination/const-pointers.ll
@@ -6,10 +6,10 @@ target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 @g = global i32 42
 
 define void @test1(%t* noalias %pp) {
-  %p = getelementptr inbounds %t* %pp, i32 0, i32 0
+  %p = getelementptr inbounds %t, %t* %pp, i32 0, i32 0
 
   store i32 1, i32* %p; <-- This is dead
-  %x = load i32* inttoptr (i32 12345 to i32*)
+  %x = load i32, i32* inttoptr (i32 12345 to i32*)
   store i32 %x, i32* %p
   ret void
 ; CHECK-LABEL: define void @test1(
@@ -30,7 +30,7 @@ define void @test3() {
 
 define void @test4(i32* %p) {
   store i32 1, i32* %p
-  %x = load i32* @g; <-- %p and @g could alias
+  %x = load i32, i32* @g; <-- %p and @g could alias
   store i32 %x, i32* %p
   ret void
 ; CHECK-LABEL: define void @test4(
diff --git a/test/Transforms/DeadStoreElimination/crash.ll b/test/Transforms/DeadStoreElimination/crash.ll
index 148695fcedbd..78cb842e60ed 100644
--- a/test/Transforms/DeadStoreElimination/crash.ll
+++ b/test/Transforms/DeadStoreElimination/crash.ll
@@ -28,14 +28,14 @@ bb:                                               ; preds = %bb, %entry
 define fastcc i32 @test2() nounwind ssp {
 bb14:                                             ; preds = %bb4
   %0 = bitcast i8* undef to i8**                  ; <i8**> [#uses=1]
-  %1 = getelementptr inbounds i8** %0, i64 undef  ; <i8**> [#uses=1]
+  %1 = getelementptr inbounds i8*, i8** %0, i64 undef  ; <i8**> [#uses=1]
   %2 = bitcast i8** %1 to i16*                    ; <i16*> [#uses=2]
-  %3 = getelementptr inbounds i16* %2, i64 undef  ; <i16*> [#uses=1]
+  %3 = getelementptr inbounds i16, i16* %2, i64 undef  ; <i16*> [#uses=1]
   %4 = bitcast i16* %3 to i8*                     ; <i8*> [#uses=1]
-  %5 = getelementptr inbounds i8* %4, i64 undef   ; <i8*> [#uses=1]
-  %6 = getelementptr inbounds i16* %2, i64 undef  ; <i16*> [#uses=1]
+  %5 = getelementptr inbounds i8, i8* %4, i64 undef   ; <i8*> [#uses=1]
+  %6 = getelementptr inbounds i16, i16* %2, i64 undef  ; <i16*> [#uses=1]
   store i16 undef, i16* %6, align 2
-  %7 = getelementptr inbounds i8* %5, i64 undef   ; <i8*> [#uses=1]
+  %7 = getelementptr inbounds i8, i8* %5, i64 undef   ; <i8*> [#uses=1]
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %7, i8* undef, i64 undef, i32 1, i1 false)
   unreachable
 }
@@ -49,8 +49,8 @@ entry:
   ret i32 0
   
 dead:
-  %P2 = getelementptr i32 *%P2, i32 52
-  %Q2 = getelementptr i32 *%Q2, i32 52
+  %P2 = getelementptr i32, i32 *%P2, i32 52
+  %Q2 = getelementptr i32, i32 *%Q2, i32 52
   store i32 4, i32* %P2
   store i32 4, i32* %Q2
   br label %dead
@@ -65,9 +65,9 @@ dead:
 define void @test4(%struct.f393a00_2__windmill* %a, %struct.f393a00_2__windmill* %b) {
 entry:
 	%t = alloca %struct.f393a00_2__windmill		; <%struct.f393a00_2__windmill*> [#uses=1]
-	%0 = getelementptr %struct.f393a00_2__windmill* %t, i32 0, i32 0, i32 0		; <%struct.ada__tags__dispatch_table**> [#uses=1]
-	%1 = load %struct.ada__tags__dispatch_table** null, align 4		; <%struct.ada__tags__dispatch_table*> [#uses=1]
-	%2 = load %struct.ada__tags__dispatch_table** %0, align 8		; <%struct.ada__tags__dispatch_table*> [#uses=1]
+	%0 = getelementptr %struct.f393a00_2__windmill, %struct.f393a00_2__windmill* %t, i32 0, i32 0, i32 0		; <%struct.ada__tags__dispatch_table**> [#uses=1]
+	%1 = load %struct.ada__tags__dispatch_table*, %struct.ada__tags__dispatch_table** null, align 4		; <%struct.ada__tags__dispatch_table*> [#uses=1]
+	%2 = load %struct.ada__tags__dispatch_table*, %struct.ada__tags__dispatch_table** %0, align 8		; <%struct.ada__tags__dispatch_table*> [#uses=1]
 	store %struct.ada__tags__dispatch_table* %2, %struct.ada__tags__dispatch_table** null, align 4
 	store %struct.ada__tags__dispatch_table* %1, %struct.ada__tags__dispatch_table** null, align 4
 	ret void
diff --git a/test/Transforms/DeadStoreElimination/cs-cs-aliasing.ll b/test/Transforms/DeadStoreElimination/cs-cs-aliasing.ll
index 8953f9ce0834..30bb96fa2b4a 100644
--- a/test/Transforms/DeadStoreElimination/cs-cs-aliasing.ll
+++ b/test/Transforms/DeadStoreElimination/cs-cs-aliasing.ll
@@ -22,18 +22,18 @@ define weak_odr zeroext i1 @test() #2 align 2 {
 bb:
   %tmp = alloca %class.basic_string, align 8
   %tmp1 = alloca %class.basic_string, align 8
-  %tmp3 = getelementptr inbounds %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 2
+  %tmp3 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 2
   %tmp4 = bitcast %union.anon* %tmp3 to i8*
-  %tmp5 = getelementptr inbounds %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 0, i32 0
-  %tmp6 = getelementptr inbounds %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 1
-  %tmp7 = getelementptr inbounds i8* %tmp4, i64 1
+  %tmp5 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 0, i32 0
+  %tmp6 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp, i64 0, i32 0, i32 0, i32 1
+  %tmp7 = getelementptr inbounds i8, i8* %tmp4, i64 1
   %tmp8 = bitcast %class.basic_string* %tmp to i8*
   %tmp9 = bitcast i64 0 to i64
-  %tmp10 = getelementptr inbounds %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 2
+  %tmp10 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 2
   %tmp11 = bitcast %union.anon* %tmp10 to i8*
-  %tmp12 = getelementptr inbounds %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 0, i32 0
-  %tmp13 = getelementptr inbounds %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 1
-  %tmp14 = getelementptr inbounds i8* %tmp11, i64 1
+  %tmp12 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 0, i32 0
+  %tmp13 = getelementptr inbounds %class.basic_string, %class.basic_string* %tmp1, i64 0, i32 0, i32 0, i32 1
+  %tmp14 = getelementptr inbounds i8, i8* %tmp11, i64 1
   %tmp15 = bitcast %class.basic_string* %tmp1 to i8*
   br label %_ZN12basic_stringIcSt11char_traitsIcESaIcEEC2EPKcRKS2_.exit
 
diff --git a/test/Transforms/DeadStoreElimination/dominate.ll b/test/Transforms/DeadStoreElimination/dominate.ll
index 38cf1a066dae..638992bae729 100644
--- a/test/Transforms/DeadStoreElimination/dominate.ll
+++ b/test/Transforms/DeadStoreElimination/dominate.ll
@@ -5,7 +5,7 @@ declare void @bar()
 define void @foo() {
 bb1:
   %memtmp3.i = alloca [21 x i8], align 1
-  %0 = getelementptr inbounds [21 x i8]* %memtmp3.i, i64 0, i64 0
+  %0 = getelementptr inbounds [21 x i8], [21 x i8]* %memtmp3.i, i64 0, i64 0
   br label %bb3
 
 bb2:
diff --git a/test/Transforms/DeadStoreElimination/free.ll b/test/Transforms/DeadStoreElimination/free.ll
index 1d273d67a501..6b69ec860202 100644
--- a/test/Transforms/DeadStoreElimination/free.ll
+++ b/test/Transforms/DeadStoreElimination/free.ll
@@ -10,7 +10,7 @@ declare noalias i8* @malloc(i64)
 ; CHECK-NEXT: @free
 ; CHECK-NEXT: ret void
 define void @test(i32* %Q, i32* %P) {
-        %DEAD = load i32* %Q            ; <i32> [#uses=1]
+        %DEAD = load i32, i32* %Q            ; <i32> [#uses=1]
         store i32 %DEAD, i32* %P
         %1 = bitcast i32* %P to i8*
         tail call void @free(i8* %1)
@@ -22,7 +22,7 @@ define void @test(i32* %Q, i32* %P) {
 ; CHECK-NEXT: @free
 ; CHECK-NEXT: ret void
 define void @test2({i32, i32}* %P) {
-	%Q = getelementptr {i32, i32} *%P, i32 0, i32 1
+	%Q = getelementptr {i32, i32}, {i32, i32} *%P, i32 0, i32 1
 	store i32 4, i32* %Q
         %1 = bitcast {i32, i32}* %P to i8*
         tail call void @free(i8* %1)
@@ -35,7 +35,7 @@ define void @test2({i32, i32}* %P) {
 define void @test3() {
   %m = call i8* @malloc(i64 24)
   store i8 0, i8* %m
-  %m1 = getelementptr i8* %m, i64 1
+  %m1 = getelementptr i8, i8* %m, i64 1
   store i8 1, i8* %m1
   call void @free(i8* %m)
   ret void
diff --git a/test/Transforms/DeadStoreElimination/inst-limits.ll b/test/Transforms/DeadStoreElimination/inst-limits.ll
index 3ef560723387..54e41c8b413b 100644
--- a/test/Transforms/DeadStoreElimination/inst-limits.ll
+++ b/test/Transforms/DeadStoreElimination/inst-limits.ll
@@ -118,7 +118,7 @@ entry:
 
   ; Insert a meaningless dbg.value intrinsic; it should have no
   ; effect on the working of DSE in any way.
-  call void @llvm.dbg.value(metadata i32* undef, i64 0, metadata !10, metadata !{})
+  call void @llvm.dbg.value(metadata i32* undef, i64 0, metadata !10, metadata !DIExpression()), !dbg !DILocation(scope: !4)
 
   ; CHECK:  store i32 -1, i32* @x, align 4
   store i32 -1, i32* @x, align 4
@@ -245,18 +245,18 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata)
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!11, !13}
 
-!0 = !{!"0x11\004\00clang version 3.4\001\00\000\00\000", !1, !2, !2, !3, !9, !2} ; [ DW_TAG_compile_unit ] [/home/tmp/test.c] [DW_LANG_C99]
-!1 = !{!"test.c", !"/home/tmp"}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.4", isOptimized: true, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "test.c", directory: "/home/tmp")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00test_within_limit\00test_within_limit\00\003\000\001\000\006\00256\000\004", !1, !5, !6, null, i32 ()* @test_within_limit, null, null, !2} ; [ DW_TAG_subprogram ] [line 3] [def] [scope 4] [test]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [/home/tmp/test.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "test_within_limit", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 4, file: !1, scope: !5, type: !6, function: i32 ()* @test_within_limit, variables: !2)
+!5 = !DIFile(filename: "test.c", directory: "/home/tmp")
+!6 = !DISubroutineType(types: !7)
 !7 = !{!8}
-!8 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !9 = !{!10}
-!10 = !{!"0x34\00x\00x\00\001\000\001", null, !5, !8, i32* @x, null} ; [ DW_TAG_variable ] [x] [line 1] [def]
+!10 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "x", scope: !4, type: !8)
 !11 = !{i32 2, !"Dwarf Version", i32 4}
 !12 = !{i32* undef}
 
-!13 = !{i32 1, !"Debug Info Version", i32 2}
+!13 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/DeadStoreElimination/libcalls.ll b/test/Transforms/DeadStoreElimination/libcalls.ll
index 65396945bf65..8afa148a58b7 100644
--- a/test/Transforms/DeadStoreElimination/libcalls.ll
+++ b/test/Transforms/DeadStoreElimination/libcalls.ll
@@ -4,7 +4,7 @@ declare i8* @strcpy(i8* %dest, i8* %src) nounwind
 define void @test1(i8* %src) {
 ; CHECK-LABEL: @test1(
   %B = alloca [16 x i8]
-  %dest = getelementptr inbounds [16 x i8]* %B, i64 0, i64 0
+  %dest = getelementptr inbounds [16 x i8], [16 x i8]* %B, i64 0, i64 0
 ; CHECK-NOT: @strcpy
   %call = call i8* @strcpy(i8* %dest, i8* %src)
 ; CHECK: ret void
@@ -15,7 +15,7 @@ declare i8* @strncpy(i8* %dest, i8* %src, i32 %n) nounwind
 define void @test2(i8* %src) {
 ; CHECK-LABEL: @test2(
   %B = alloca [16 x i8]
-  %dest = getelementptr inbounds [16 x i8]* %B, i64 0, i64 0
+  %dest = getelementptr inbounds [16 x i8], [16 x i8]* %B, i64 0, i64 0
 ; CHECK-NOT: @strncpy
   %call = call i8* @strncpy(i8* %dest, i8* %src, i32 12)
 ; CHECK: ret void
@@ -26,7 +26,7 @@ declare i8* @strcat(i8* %dest, i8* %src) nounwind
 define void @test3(i8* %src) {
 ; CHECK-LABEL: @test3(
   %B = alloca [16 x i8]
-  %dest = getelementptr inbounds [16 x i8]* %B, i64 0, i64 0
+  %dest = getelementptr inbounds [16 x i8], [16 x i8]* %B, i64 0, i64 0
 ; CHECK-NOT: @strcat
   %call = call i8* @strcat(i8* %dest, i8* %src)
 ; CHECK: ret void
@@ -37,7 +37,7 @@ declare i8* @strncat(i8* %dest, i8* %src, i32 %n) nounwind
 define void @test4(i8* %src) {
 ; CHECK-LABEL: @test4(
   %B = alloca [16 x i8]
-  %dest = getelementptr inbounds [16 x i8]* %B, i64 0, i64 0
+  %dest = getelementptr inbounds [16 x i8], [16 x i8]* %B, i64 0, i64 0
 ; CHECK-NOT: @strncat
   %call = call i8* @strncat(i8* %dest, i8* %src, i32 12)
 ; CHECK: ret void
@@ -47,10 +47,10 @@ define void @test4(i8* %src) {
 define void @test5(i8* nocapture %src) {
 ; CHECK-LABEL: @test5(
   %dest = alloca [100 x i8], align 16
-  %arraydecay = getelementptr inbounds [100 x i8]* %dest, i64 0, i64 0
+  %arraydecay = getelementptr inbounds [100 x i8], [100 x i8]* %dest, i64 0, i64 0
   %call = call i8* @strcpy(i8* %arraydecay, i8* %src)
 ; CHECK: %call = call i8* @strcpy
-  %arrayidx = getelementptr inbounds i8* %call, i64 10
+  %arrayidx = getelementptr inbounds i8, i8* %call, i64 10
   store i8 97, i8* %arrayidx, align 1
   ret void
 }
@@ -59,7 +59,7 @@ declare void @user(i8* %p)
 define void @test6(i8* %src) {
 ; CHECK-LABEL: @test6(
   %B = alloca [16 x i8]
-  %dest = getelementptr inbounds [16 x i8]* %B, i64 0, i64 0
+  %dest = getelementptr inbounds [16 x i8], [16 x i8]* %B, i64 0, i64 0
 ; CHECK: @strcpy
   %call = call i8* @strcpy(i8* %dest, i8* %src)
 ; CHECK: @user
diff --git a/test/Transforms/DeadStoreElimination/lifetime.ll b/test/Transforms/DeadStoreElimination/lifetime.ll
index 7fe7fbfa7478..305c916dc02b 100644
--- a/test/Transforms/DeadStoreElimination/lifetime.ll
+++ b/test/Transforms/DeadStoreElimination/lifetime.ll
@@ -23,7 +23,7 @@ define void @test1() {
 
 define void @test2(i32* %P) {
 ; CHECK: test2
-  %Q = getelementptr i32* %P, i32 1
+  %Q = getelementptr i32, i32* %P, i32 1
   %R = bitcast i32* %Q to i8*
   call void @llvm.lifetime.start(i64 4, i8* %R)
 ; CHECK: lifetime.start
diff --git a/test/Transforms/DeadStoreElimination/no-targetdata.ll b/test/Transforms/DeadStoreElimination/no-targetdata.ll
index 25395330bf37..f9262ed54b3f 100644
--- a/test/Transforms/DeadStoreElimination/no-targetdata.ll
+++ b/test/Transforms/DeadStoreElimination/no-targetdata.ll
@@ -12,7 +12,7 @@ entry:
 ; CHECK: tail call void @llvm.memcpy.p0i8.p0i8.i64
 ; CHECK: ret void
 
-  %arrayidx = getelementptr i8* %buf, i64 18
+  %arrayidx = getelementptr i8, i8* %buf, i64 18
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %arrayidx, i8* %buf, i64 18, i32 1, i1 false)
   store i8 1, i8* %arrayidx, align 1
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %buf, i8* %arrayidx, i64 18, i32 1, i1 false)
diff --git a/test/Transforms/DeadStoreElimination/pr11390.ll b/test/Transforms/DeadStoreElimination/pr11390.ll
index f63aa1eb8aae..faf3b8b47a92 100644
--- a/test/Transforms/DeadStoreElimination/pr11390.ll
+++ b/test/Transforms/DeadStoreElimination/pr11390.ll
@@ -18,12 +18,12 @@ entry:
 
 if.end:                                           ; preds = %entry
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %call4, i8* %name, i64 %call, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds i8* %call4, i64 %call
+  %arrayidx = getelementptr inbounds i8, i8* %call4, i64 %call
   store i8 46, i8* %arrayidx, align 1
 ; CHECK: store i8 46
-  %add.ptr5 = getelementptr inbounds i8* %call4, i64 %add
+  %add.ptr5 = getelementptr inbounds i8, i8* %call4, i64 %add
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %add.ptr5, i8* %domain, i64 %call1, i32 1, i1 false)
-  %arrayidx8 = getelementptr inbounds i8* %call4, i64 %add2
+  %arrayidx8 = getelementptr inbounds i8, i8* %call4, i64 %add2
   store i8 0, i8* %arrayidx8, align 1
   br label %return
 
diff --git a/test/Transforms/DeadStoreElimination/simple.ll b/test/Transforms/DeadStoreElimination/simple.ll
index 1e813852b04c..2ffe0539098e 100644
--- a/test/Transforms/DeadStoreElimination/simple.ll
+++ b/test/Transforms/DeadStoreElimination/simple.ll
@@ -3,10 +3,10 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 
 declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
-declare i8* @llvm.init.trampoline(i8*, i8*, i8*)
+declare void @llvm.init.trampoline(i8*, i8*, i8*)
 
 define void @test1(i32* %Q, i32* %P) {
-        %DEAD = load i32* %Q
+        %DEAD = load i32, i32* %Q
         store i32 %DEAD, i32* %P
         store i32 0, i32* %P
         ret void
@@ -31,17 +31,17 @@ define void @test2(i32 *%p, i32 *%q) {
 
 define i32 @test3(i32* %g_addr) nounwind {
 ; CHECK-LABEL: @test3(
-; CHECK: load i32* %g_addr
-  %g_value = load i32* %g_addr, align 4
+; CHECK: load i32, i32* %g_addr
+  %g_value = load i32, i32* %g_addr, align 4
   store i32 -1, i32* @g, align 4
   store i32 %g_value, i32* %g_addr, align 4
-  %tmp3 = load i32* @g, align 4
+  %tmp3 = load i32, i32* @g, align 4
   ret i32 %tmp3
 }
 
 
 define void @test4(i32* %Q) {
-        %a = load i32* %Q
+        %a = load i32, i32* %Q
         store volatile i32 %a, i32* %Q
         ret void
 ; CHECK-LABEL: @test4(
@@ -51,7 +51,7 @@ define void @test4(i32* %Q) {
 }
 
 define void @test5(i32* %Q) {
-        %a = load volatile i32* %Q
+        %a = load volatile i32, i32* %Q
         store i32 %a, i32* %Q
         ret void
 ; CHECK-LABEL: @test5(
@@ -87,7 +87,7 @@ define i32 @test8() {
         store i32 1234567, i32* %V
         %V2 = bitcast i32* %V to i8*
         store i8 0, i8* %V2
-        %X = load i32* %V
+        %X = load i32, i32* %V
         ret i32 %X
         
 ; CHECK-LABEL: @test8(
@@ -98,7 +98,7 @@ define i32 @test8() {
 ; Test for byval handling.
 %struct.x = type { i32, i32, i32, i32 }
 define void @test9(%struct.x* byval  %a) nounwind  {
-	%tmp2 = getelementptr %struct.x* %a, i32 0, i32 0
+	%tmp2 = getelementptr %struct.x, %struct.x* %a, i32 0, i32 0
 	store i32 1, i32* %tmp2, align 4
 	ret void
 ; CHECK-LABEL: @test9(
@@ -107,7 +107,7 @@ define void @test9(%struct.x* byval  %a) nounwind  {
 
 ; Test for inalloca handling.
 define void @test9_2(%struct.x* inalloca  %a) nounwind  {
-	%tmp2 = getelementptr %struct.x* %a, i32 0, i32 0
+	%tmp2 = getelementptr %struct.x, %struct.x* %a, i32 0, i32 0
 	store i32 1, i32* %tmp2, align 4
 	ret void
 ; CHECK-LABEL: @test9_2(
@@ -131,8 +131,8 @@ define void @test11() {
 ; CHECK-LABEL: @test11(
 	%storage = alloca [10 x i8], align 16		; <[10 x i8]*> [#uses=1]
 ; CHECK-NOT: alloca
-	%cast = getelementptr [10 x i8]* %storage, i32 0, i32 0		; <i8*> [#uses=1]
-	%tramp = call i8* @llvm.init.trampoline( i8* %cast, i8* bitcast (void ()* @test11f to i8*), i8* null )		; <i8*> [#uses=1]
+	%cast = getelementptr [10 x i8], [10 x i8]* %storage, i32 0, i32 0		; <i8*> [#uses=1]
+	call void @llvm.init.trampoline( i8* %cast, i8* bitcast (void ()* @test11f to i8*), i8* null )		; <i8*> [#uses=1]
 ; CHECK-NOT: trampoline
 	ret void
 ; CHECK: ret void
@@ -141,10 +141,10 @@ define void @test11() {
 
 ; PR2599 - load -> store to same address.
 define void @test12({ i32, i32 }* %x) nounwind  {
-	%tmp4 = getelementptr { i32, i32 }* %x, i32 0, i32 0
-	%tmp5 = load i32* %tmp4, align 4
-	%tmp7 = getelementptr { i32, i32 }* %x, i32 0, i32 1
-	%tmp8 = load i32* %tmp7, align 4
+	%tmp4 = getelementptr { i32, i32 }, { i32, i32 }* %x, i32 0, i32 0
+	%tmp5 = load i32, i32* %tmp4, align 4
+	%tmp7 = getelementptr { i32, i32 }, { i32, i32 }* %x, i32 0, i32 1
+	%tmp8 = load i32, i32* %tmp7, align 4
 	%tmp17 = sub i32 0, %tmp8
 	store i32 %tmp5, i32* %tmp4, align 4
 	store i32 %tmp17, i32* %tmp7, align 4
@@ -160,7 +160,7 @@ declare void @test13f()
 define i32* @test13() {
         %p = tail call i8* @malloc(i32 4)
         %P = bitcast i8* %p to i32*
-        %DEAD = load i32* %P
+        %DEAD = load i32, i32* %P
         %DEAD2 = add i32 %DEAD, 1
         store i32 %DEAD2, i32* %P
         call void @test13f( )
@@ -176,7 +176,7 @@ define i32 addrspace(1)* @test13_addrspacecast() {
   %p = tail call i8* @malloc(i32 4)
   %p.bc = bitcast i8* %p to i32*
   %P = addrspacecast i32* %p.bc to i32 addrspace(1)*
-  %DEAD = load i32 addrspace(1)* %P
+  %DEAD = load i32, i32 addrspace(1)* %P
   %DEAD2 = add i32 %DEAD, 1
   store i32 %DEAD2, i32 addrspace(1)* %P
   call void @test13f( )
@@ -195,7 +195,7 @@ declare noalias i8* @calloc(i32, i32)
 
 define void @test14(i32* %Q) {
         %P = alloca i32
-        %DEAD = load i32* %Q
+        %DEAD = load i32, i32* %Q
         store i32 %DEAD, i32* %P
         ret void
 
@@ -267,7 +267,7 @@ declare void @test19f({i32}* byval align 4 %P)
 
 define void @test19({i32} * nocapture byval align 4 %arg5) nounwind ssp {
 bb:
-  %tmp7 = getelementptr inbounds {i32}* %arg5, i32 0, i32 0
+  %tmp7 = getelementptr inbounds {i32}, {i32}* %arg5, i32 0, i32 0
   store i32 912, i32* %tmp7
   call void @test19f({i32}* byval align 4 %arg5)
   ret void
@@ -310,9 +310,9 @@ define void @test22(i1 %i, i32 %k, i32 %m) nounwind {
 declare noalias i8* @strdup(i8* nocapture) nounwind
 define noalias i8* @test23() nounwind uwtable ssp {
   %x = alloca [2 x i8], align 1
-  %arrayidx = getelementptr inbounds [2 x i8]* %x, i64 0, i64 0
+  %arrayidx = getelementptr inbounds [2 x i8], [2 x i8]* %x, i64 0, i64 0
   store i8 97, i8* %arrayidx, align 1
-  %arrayidx1 = getelementptr inbounds [2 x i8]* %x, i64 0, i64 1
+  %arrayidx1 = getelementptr inbounds [2 x i8], [2 x i8]* %x, i64 0, i64 1
   store i8 0, i8* %arrayidx1, align 1
   %call = call i8* @strdup(i8* %arrayidx) nounwind
   ret i8* %call
@@ -326,13 +326,13 @@ define noalias i8* @test23() nounwind uwtable ssp {
 ; CHECK: store i32 %c
 ; CHECK: ret void
 define void @test24([2 x i32]* %a, i32 %b, i32 %c) nounwind {
-  %1 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 0
+  %1 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 0, i64 0
   store i32 0, i32* %1, align 4
-  %2 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 1
+  %2 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 0, i64 1
   store i32 0, i32* %2, align 4
-  %3 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 0
+  %3 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 0, i64 0
   store i32 %b, i32* %3, align 4
-  %4 = getelementptr inbounds [2 x i32]* %a, i64 0, i64 1
+  %4 = getelementptr inbounds [2 x i32], [2 x i32]* %a, i64 0, i64 1
   store i32 %c, i32* %4, align 4
   ret void
 }
@@ -343,8 +343,8 @@ define void @test24([2 x i32]* %a, i32 %b, i32 %c) nounwind {
 ; CHECK: store i8 0
 ; CHECK: store i8 %tmp
 define i8* @test25(i8* %p) nounwind {
-  %p.4 = getelementptr i8* %p, i64 4
-  %tmp = load i8* %p.4, align 1
+  %p.4 = getelementptr i8, i8* %p, i64 4
+  %tmp = load i8, i8* %p.4, align 1
   store i8 0, i8* %p.4, align 1
   %q = call i8* @strdup(i8* %p) nounwind optsize
   store i8 %tmp, i8* %p.4, align 1
diff --git a/test/Transforms/EarlyCSE/AArch64/intrinsics.ll b/test/Transforms/EarlyCSE/AArch64/intrinsics.ll
new file mode 100644
index 000000000000..d166ff1e1092
--- /dev/null
+++ b/test/Transforms/EarlyCSE/AArch64/intrinsics.ll
@@ -0,0 +1,232 @@
+; RUN: opt < %s -S -mtriple=aarch64-none-linux-gnu -mattr=+neon -early-cse | FileCheck %s
+; RUN: opt < %s -S -mtriple=aarch64-none-linux-gnu -mattr=+neon -passes=early-cse | FileCheck %s
+
+define <4 x i32> @test_cse(i32* %a, [2 x <4 x i32>] %s.coerce, i32 %n) {
+entry:
+; Check that @llvm.aarch64.neon.ld2 is optimized away by Early CSE.
+; CHECK-LABEL: @test_cse
+; CHECK-NOT: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %1 = bitcast <4 x i32> %s.coerce.fca.0.extract to <16 x i8>
+  %2 = bitcast <4 x i32> %s.coerce.fca.1.extract to <16 x i8>
+  %3 = bitcast <16 x i8> %1 to <4 x i32>
+  %4 = bitcast <16 x i8> %2 to <4 x i32>
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0)
+  %5 = bitcast i32* %a to i8*
+  %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %5)
+  %vld2.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 0
+  %vld2.fca.1.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 1
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld2.fca.0.extract, <4 x i32> %vld2.fca.0.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+define <4 x i32> @test_cse2(i32* %a, [2 x <4 x i32>] %s.coerce, i32 %n) {
+entry:
+; Check that the first @llvm.aarch64.neon.st2 is optimized away by Early CSE.
+; CHECK-LABEL: @test_cse2
+; CHECK-NOT: call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %3, i8* %0)
+; CHECK: call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0)
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %1 = bitcast <4 x i32> %s.coerce.fca.0.extract to <16 x i8>
+  %2 = bitcast <4 x i32> %s.coerce.fca.1.extract to <16 x i8>
+  %3 = bitcast <16 x i8> %1 to <4 x i32>
+  %4 = bitcast <16 x i8> %2 to <4 x i32>
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %3, i8* %0)
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0)
+  %5 = bitcast i32* %a to i8*
+  %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %5)
+  %vld2.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 0
+  %vld2.fca.1.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 1
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld2.fca.0.extract, <4 x i32> %vld2.fca.0.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+define <4 x i32> @test_cse3(i32* %a, [2 x <4 x i32>] %s.coerce, i32 %n) #0 {
+entry:
+; Check that the first @llvm.aarch64.neon.ld2 is optimized away by Early CSE.
+; CHECK-LABEL: @test_cse3
+; CHECK: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8
+; CHECK-NOT: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %0)
+  %vld2.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 0
+  %vld2.fca.1.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 1
+  %1 = bitcast i32* %a to i8*
+  %vld22 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %1)
+  %vld22.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld22, 0
+  %vld22.fca.1.extract = extractvalue { <4 x i32>, <4 x i32> } %vld22, 1
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld2.fca.0.extract, <4 x i32> %vld22.fca.0.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+
+define <4 x i32> @test_nocse(i32* %a, i32* %b, [2 x <4 x i32>] %s.coerce, i32 %n) {
+entry:
+; Check that the store prevents @llvm.aarch64.neon.ld2 from being optimized
+; away by Early CSE.
+; CHECK-LABEL: @test_nocse
+; CHECK: call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %1 = bitcast <4 x i32> %s.coerce.fca.0.extract to <16 x i8>
+  %2 = bitcast <4 x i32> %s.coerce.fca.1.extract to <16 x i8>
+  %3 = bitcast <16 x i8> %1 to <4 x i32>
+  %4 = bitcast <16 x i8> %2 to <4 x i32>
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0)
+  store i32 0, i32* %b, align 4
+  %5 = bitcast i32* %a to i8*
+  %vld2 = call { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8* %5)
+  %vld2.fca.0.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 0
+  %vld2.fca.1.extract = extractvalue { <4 x i32>, <4 x i32> } %vld2, 1
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld2.fca.0.extract, <4 x i32> %vld2.fca.0.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+define <4 x i32> @test_nocse2(i32* %a, [2 x <4 x i32>] %s.coerce, i32 %n) {
+entry:
+; Check that @llvm.aarch64.neon.ld3 is not optimized away by Early CSE due
+; to mismatch between st2 and ld3.
+; CHECK-LABEL: @test_nocse2
+; CHECK: call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0i8
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %1 = bitcast <4 x i32> %s.coerce.fca.0.extract to <16 x i8>
+  %2 = bitcast <4 x i32> %s.coerce.fca.1.extract to <16 x i8>
+  %3 = bitcast <16 x i8> %1 to <4 x i32>
+  %4 = bitcast <16 x i8> %2 to <4 x i32>
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %4, i8* %0)
+  %5 = bitcast i32* %a to i8*
+  %vld3 = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0i8(i8* %5)
+  %vld3.fca.0.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld3, 0
+  %vld3.fca.2.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld3, 2
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld3.fca.0.extract, <4 x i32> %vld3.fca.2.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+define <4 x i32> @test_nocse3(i32* %a, [2 x <4 x i32>] %s.coerce, i32 %n) {
+entry:
+; Check that @llvm.aarch64.neon.st3 is not optimized away by Early CSE due to
+; mismatch between st2 and st3.
+; CHECK-LABEL: @test_nocse3
+; CHECK: call void @llvm.aarch64.neon.st3.v4i32.p0i8
+; CHECK: call void @llvm.aarch64.neon.st2.v4i32.p0i8
+  %s.coerce.fca.0.extract = extractvalue [2 x <4 x i32>] %s.coerce, 0
+  %s.coerce.fca.1.extract = extractvalue [2 x <4 x i32>] %s.coerce, 1
+  br label %for.cond
+
+for.cond:                                         ; preds = %for.body, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %res.0 = phi <4 x i32> [ undef, %entry ], [ %call, %for.body ]
+  %cmp = icmp slt i32 %i.0, %n
+  br i1 %cmp, label %for.body, label %for.end
+
+for.body:                                         ; preds = %for.cond
+  %0 = bitcast i32* %a to i8*
+  %1 = bitcast <4 x i32> %s.coerce.fca.0.extract to <16 x i8>
+  %2 = bitcast <4 x i32> %s.coerce.fca.1.extract to <16 x i8>
+  %3 = bitcast <16 x i8> %1 to <4 x i32>
+  %4 = bitcast <16 x i8> %2 to <4 x i32>
+  call void @llvm.aarch64.neon.st3.v4i32.p0i8(<4 x i32> %4, <4 x i32> %3, <4 x i32> %3, i8* %0)
+  call void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32> %3, <4 x i32> %3, i8* %0)
+  %5 = bitcast i32* %a to i8*
+  %vld3 = call { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0i8(i8* %5)
+  %vld3.fca.0.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld3, 0
+  %vld3.fca.1.extract = extractvalue { <4 x i32>, <4 x i32>, <4 x i32> } %vld3, 1
+  %call = call <4 x i32> @vaddq_s32(<4 x i32> %vld3.fca.0.extract, <4 x i32> %vld3.fca.0.extract)
+  %inc = add nsw i32 %i.0, 1
+  br label %for.cond
+
+for.end:                                          ; preds = %for.cond
+  ret <4 x i32> %res.0
+}
+
+; Function Attrs: nounwind
+declare void @llvm.aarch64.neon.st2.v4i32.p0i8(<4 x i32>, <4 x i32>, i8* nocapture)
+
+; Function Attrs: nounwind
+declare void @llvm.aarch64.neon.st3.v4i32.p0i8(<4 x i32>, <4 x i32>, <4 x i32>, i8* nocapture)
+
+; Function Attrs: nounwind readonly
+declare { <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld2.v4i32.p0i8(i8*)
+
+; Function Attrs: nounwind readonly
+declare { <4 x i32>, <4 x i32>, <4 x i32> } @llvm.aarch64.neon.ld3.v4i32.p0i8(i8*)
+
+define internal fastcc <4 x i32> @vaddq_s32(<4 x i32> %__p0, <4 x i32> %__p1) {
+entry:
+  %add = add <4 x i32> %__p0, %__p1
+  ret <4 x i32> %add
+}
diff --git a/test/Transforms/EarlyCSE/AArch64/lit.local.cfg b/test/Transforms/EarlyCSE/AArch64/lit.local.cfg
new file mode 100644
index 000000000000..6642d2870680
--- /dev/null
+++ b/test/Transforms/EarlyCSE/AArch64/lit.local.cfg
@@ -0,0 +1,5 @@
+config.suffixes = ['.ll']
+
+targets = set(config.root.targets_to_build.split())
+if not 'AArch64' in targets:
+    config.unsupported = True
diff --git a/test/Transforms/EarlyCSE/basic.ll b/test/Transforms/EarlyCSE/basic.ll
index 3ec8831def18..43b5e6098f6a 100644
--- a/test/Transforms/EarlyCSE/basic.ll
+++ b/test/Transforms/EarlyCSE/basic.ll
@@ -1,4 +1,5 @@
 ; RUN: opt < %s -S -early-cse | FileCheck %s
+; RUN: opt < %s -S -passes=early-cse | FileCheck %s
 
 declare void @llvm.assume(i1) nounwind
 
@@ -36,8 +37,8 @@ define void @test1(i8 %V, i32 *%P) {
 ;; Simple load value numbering.
 ; CHECK-LABEL: @test2(
 define i32 @test2(i32 *%P) {
-  %V1 = load i32* %P
-  %V2 = load i32* %P
+  %V1 = load i32, i32* %P
+  %V2 = load i32, i32* %P
   %Diff = sub i32 %V1, %V2
   ret i32 %Diff
   ; CHECK: ret i32 0
@@ -45,9 +46,9 @@ define i32 @test2(i32 *%P) {
 
 ; CHECK-LABEL: @test2a(
 define i32 @test2a(i32 *%P, i1 %b) {
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   tail call void @llvm.assume(i1 %b)
-  %V2 = load i32* %P
+  %V2 = load i32, i32* %P
   %Diff = sub i32 %V1, %V2
   ret i32 %Diff
   ; CHECK: ret i32 0
@@ -56,13 +57,13 @@ define i32 @test2a(i32 *%P, i1 %b) {
 ;; Cross block load value numbering.
 ; CHECK-LABEL: @test3(
 define i32 @test3(i32 *%P, i1 %Cond) {
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   br i1 %Cond, label %T, label %F
 T:
   store i32 4, i32* %P
   ret i32 42
 F:
-  %V2 = load i32* %P
+  %V2 = load i32, i32* %P
   %Diff = sub i32 %V1, %V2
   ret i32 %Diff
   ; CHECK: F:
@@ -71,14 +72,14 @@ F:
 
 ; CHECK-LABEL: @test3a(
 define i32 @test3a(i32 *%P, i1 %Cond, i1 %b) {
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   br i1 %Cond, label %T, label %F
 T:
   store i32 4, i32* %P
   ret i32 42
 F:
   tail call void @llvm.assume(i1 %b)
-  %V2 = load i32* %P
+  %V2 = load i32, i32* %P
   %Diff = sub i32 %V1, %V2
   ret i32 %Diff
   ; CHECK: F:
@@ -88,7 +89,7 @@ F:
 ;; Cross block load value numbering stops when stores happen.
 ; CHECK-LABEL: @test4(
 define i32 @test4(i32 *%P, i1 %Cond) {
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   br i1 %Cond, label %T, label %F
 T:
   ret i32 42
@@ -96,7 +97,7 @@ F:
   ; Clobbers V1
   store i32 42, i32* %P
   
-  %V2 = load i32* %P
+  %V2 = load i32, i32* %P
   %Diff = sub i32 %V1, %V2
   ret i32 %Diff
   ; CHECK: F:
@@ -119,7 +120,7 @@ define i32 @test5(i32 *%P) {
 ; CHECK-LABEL: @test6(
 define i32 @test6(i32 *%P) {
   store i32 42, i32* %P
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   ret i32 %V1
   ; CHECK: ret i32 42
 }
@@ -128,7 +129,7 @@ define i32 @test6(i32 *%P) {
 define i32 @test6a(i32 *%P, i1 %b) {
   store i32 42, i32* %P
   tail call void @llvm.assume(i1 %b)
-  %V1 = load i32* %P
+  %V1 = load i32, i32* %P
   ret i32 %V1
   ; CHECK: ret i32 42
 }
@@ -194,11 +195,11 @@ define void @test11(i32 *%P) {
 
 ; CHECK-LABEL: @test12(
 define i32 @test12(i1 %B, i32* %P1, i32* %P2) {
-  %load0 = load i32* %P1
-  %1 = load atomic i32* %P2 seq_cst, align 4
-  %load1 = load i32* %P1
+  %load0 = load i32, i32* %P1
+  %1 = load atomic i32, i32* %P2 seq_cst, align 4
+  %load1 = load i32, i32* %P1
   %sel = select i1 %B, i32 %load0, i32 %load1
   ret i32 %sel
-  ; CHECK: load i32* %P1
-  ; CHECK: load i32* %P1
+  ; CHECK: load i32, i32* %P1
+  ; CHECK: load i32, i32* %P1
 }
diff --git a/test/Transforms/EarlyCSE/conditional.ll b/test/Transforms/EarlyCSE/conditional.ll
new file mode 100644
index 000000000000..5b5a74f112b5
--- /dev/null
+++ b/test/Transforms/EarlyCSE/conditional.ll
@@ -0,0 +1,107 @@
+; RUN: opt -early-cse -S < %s | FileCheck %s
+
+; Can we CSE a known condition to a constant?
+define i1 @test(i8* %p) {
+; CHECK-LABEL: @test
+entry:
+  %cnd1 = icmp eq i8* %p, null
+  br i1 %cnd1, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i1 true
+  %cnd2 = icmp eq i8* %p, null
+  ret i1 %cnd2
+
+untaken:
+; CHECK-LABEL: untaken:
+; CHECK-NEXT: ret i1 false
+  %cnd3 = icmp eq i8* %p, null
+  ret i1 %cnd3
+}
+
+; We can CSE the condition, but we *don't* know it's value after the merge
+define i1 @test_neg1(i8* %p) {
+; CHECK-LABEL: @test_neg1
+entry:
+  %cnd1 = icmp eq i8* %p, null
+  br i1 %cnd1, label %taken, label %untaken
+
+taken:
+  br label %merge
+
+untaken:
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: ret i1 %cnd1
+  %cnd3 = icmp eq i8* %p, null
+  ret i1 %cnd3
+}
+
+; Check specifically for a case where we have a unique predecessor, but
+; not a single predecessor.  We can not know the value of the condition here.
+define i1 @test_neg2(i8* %p) {
+; CHECK-LABEL: @test_neg2
+entry:
+  %cnd1 = icmp eq i8* %p, null
+  br i1 %cnd1, label %merge, label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: ret i1 %cnd1
+  %cnd3 = icmp eq i8* %p, null
+  ret i1 %cnd3
+}
+
+; Replace a use rather than CSE
+define i1 @test2(i8* %p) {
+; CHECK-LABEL: @test2
+entry:
+  %cnd = icmp eq i8* %p, null
+  br i1 %cnd, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i1 true
+  ret i1 %cnd
+
+untaken:
+; CHECK-LABEL: untaken:
+; CHECK-NEXT: ret i1 false
+  ret i1 %cnd
+}
+
+; Not legal to replace use given it's not dominated by edge
+define i1 @test2_neg1(i8* %p) {
+; CHECK-LABEL: @test2_neg1
+entry:
+  %cnd1 = icmp eq i8* %p, null
+  br i1 %cnd1, label %taken, label %untaken
+
+taken:
+  br label %merge
+
+untaken:
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: ret i1 %cnd1
+  ret i1 %cnd1
+}
+
+; Another single predecessor test, but for dominated use
+define i1 @test2_neg2(i8* %p) {
+; CHECK-LABEL: @test2_neg2
+entry:
+  %cnd1 = icmp eq i8* %p, null
+  br i1 %cnd1, label %merge, label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: ret i1 %cnd1
+  ret i1 %cnd1
+}
+
diff --git a/test/Transforms/EarlyCSE/edge.ll b/test/Transforms/EarlyCSE/edge.ll
new file mode 100644
index 000000000000..267db679967b
--- /dev/null
+++ b/test/Transforms/EarlyCSE/edge.ll
@@ -0,0 +1,173 @@
+; RUN: opt -early-cse -S < %s | FileCheck %s
+; Same as GVN/edge.ll, but updated to reflect EarlyCSE's less powerful
+; implementation.  EarlyCSE currently doesn't exploit equality comparisons
+; against constants.
+
+define i32 @f1(i32 %x) {
+  ; CHECK-LABEL: define i32 @f1(
+bb0:
+  %cmp = icmp eq i32 %x, 0
+  br i1 %cmp, label %bb2, label %bb1
+bb1:
+  br label %bb2
+bb2:
+  %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+  %foo = add i32 %cond, %x
+  ret i32 %foo
+  ; CHECK: bb2:
+  ; CHECK: %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+}
+
+define i32 @f2(i32 %x) {
+  ; CHECK-LABEL: define i32 @f2(
+bb0:
+  %cmp = icmp ne i32 %x, 0
+  br i1 %cmp, label %bb1, label %bb2
+bb1:
+  br label %bb2
+bb2:
+  %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+  %foo = add i32 %cond, %x
+  ret i32 %foo
+  ; CHECK: bb2:
+  ; CHECK: %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+}
+
+define i32 @f3(i32 %x) {
+  ; CHECK-LABEL: define i32 @f3(
+bb0:
+  switch i32 %x, label %bb1 [ i32 0, label %bb2]
+bb1:
+  br label %bb2
+bb2:
+  %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+  %foo = add i32 %cond, %x
+  ret i32 %foo
+  ; CHECK: bb2:
+  ; CHECK: %cond = phi i32 [ %x, %bb0 ], [ 0, %bb1 ]
+}
+
+declare void @g(i1)
+define void @f4(i8 * %x)  {
+; CHECK-LABEL: define void @f4(
+bb0:
+  %y = icmp eq i8* null, %x
+  br i1 %y, label %bb2, label %bb1
+bb1:
+  br label %bb2
+bb2:
+  %zed = icmp eq i8* null, %x
+  call void @g(i1 %zed)
+; CHECK: call void @g(i1 %y)
+  ret void
+}
+
+define double @fcmp_oeq_not_zero(double %x, double %y) {
+entry:
+  %cmp = fcmp oeq double %y, 2.0
+  br i1 %cmp, label %if, label %return
+
+if:
+  %div = fdiv double %x, %y
+  br label %return
+
+return:
+  %retval = phi double [ %div, %if ], [ %x, %entry ]
+  ret double %retval
+
+; CHECK-LABEL: define double @fcmp_oeq_not_zero(
+; CHECK: %div = fdiv double %x, %y
+}
+
+define double @fcmp_une_not_zero(double %x, double %y) {
+entry:
+  %cmp = fcmp une double %y, 2.0
+  br i1 %cmp, label %return, label %else
+
+else:
+  %div = fdiv double %x, %y
+  br label %return
+
+return:
+  %retval = phi double [ %div, %else ], [ %x, %entry ]
+  ret double %retval
+
+; CHECK-LABEL: define double @fcmp_une_not_zero(
+; CHECK: %div = fdiv double %x, %y
+}
+
+; PR22376 - We can't propagate zero constants because -0.0 
+; compares equal to 0.0. If %y is -0.0 in this test case,
+; we would produce the wrong sign on the infinity return value.
+define double @fcmp_oeq_zero(double %x, double %y) {
+entry:
+  %cmp = fcmp oeq double %y, 0.0
+  br i1 %cmp, label %if, label %return
+
+if:
+  %div = fdiv double %x, %y
+  br label %return
+
+return:
+  %retval = phi double [ %div, %if ], [ %x, %entry ]
+  ret double %retval
+
+; CHECK-LABEL: define double @fcmp_oeq_zero(
+; CHECK: %div = fdiv double %x, %y
+}
+
+define double @fcmp_une_zero(double %x, double %y) {
+entry:
+  %cmp = fcmp une double %y, -0.0
+  br i1 %cmp, label %return, label %else
+
+else:
+  %div = fdiv double %x, %y
+  br label %return
+
+return:
+  %retval = phi double [ %div, %else ], [ %x, %entry ]
+  ret double %retval
+
+; CHECK-LABEL: define double @fcmp_une_zero(
+; CHECK: %div = fdiv double %x, %y
+}
+
+; We also cannot propagate a value if it's not a constant.
+; This is because the value could be 0.0 or -0.0.
+
+define double @fcmp_oeq_maybe_zero(double %x, double %y, double %z1, double %z2) {
+entry:
+ %z = fadd double %z1, %z2
+ %cmp = fcmp oeq double %y, %z
+ br i1 %cmp, label %if, label %return
+
+if:
+ %div = fdiv double %x, %z
+ br label %return
+
+return:
+ %retval = phi double [ %div, %if ], [ %x, %entry ]
+ ret double %retval
+
+; CHECK-LABEL: define double @fcmp_oeq_maybe_zero(
+; CHECK: %div = fdiv double %x, %z
+}
+
+define double @fcmp_une_maybe_zero(double %x, double %y, double %z1, double %z2) {
+entry:
+ %z = fadd double %z1, %z2
+ %cmp = fcmp une double %y, %z
+ br i1 %cmp, label %return, label %else
+
+else:
+ %div = fdiv double %x, %z
+ br label %return
+
+return:
+ %retval = phi double [ %div, %else ], [ %x, %entry ]
+ ret double %retval
+
+; CHECK-LABEL: define double @fcmp_une_maybe_zero(
+; CHECK: %div = fdiv double %x, %z
+}
diff --git a/test/Transforms/EarlyCSE/read-reg.ll b/test/Transforms/EarlyCSE/read-reg.ll
new file mode 100644
index 000000000000..83a299912c77
--- /dev/null
+++ b/test/Transforms/EarlyCSE/read-reg.ll
@@ -0,0 +1,34 @@
+; RUN: opt -S -early-cse < %s | FileCheck %s
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+; Function Attrs: nounwind
+define i64 @f(i64 %x) #0 {
+entry:
+  %0 = call i64 @llvm.read_register.i64(metadata !0)
+  call void bitcast (void (...)* @foo to void ()*)()
+  %1 = call i64 @llvm.read_register.i64(metadata !0)
+  %add = add nsw i64 %0, %1
+  ret i64 %add
+}
+
+; CHECK-LABEL: @f
+; CHECK: call i64 @llvm.read_register.i64
+; CHECK: call i64 @llvm.read_register.i64
+
+; Function Attrs: nounwind readnone
+declare i64 @llvm.read_register.i64(metadata) #1
+
+; Function Attrs: nounwind
+declare void @llvm.write_register.i64(metadata, i64) #2
+
+declare void @foo(...)
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readnone }
+attributes #2 = { nounwind }
+
+!llvm.named.register.r1 = !{!0}
+
+!0 = !{!"r1"}
+
diff --git a/test/Transforms/Float2Int/basic.ll b/test/Transforms/Float2Int/basic.ll
new file mode 100644
index 000000000000..f4d946914cd4
--- /dev/null
+++ b/test/Transforms/Float2Int/basic.ll
@@ -0,0 +1,256 @@
+; RUN: opt < %s -float2int -S | FileCheck %s
+
+;
+; Positive tests
+;
+
+; CHECK-LABEL: @simple1
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = add i32 %1, 1
+; CHECK:  %3 = trunc i32 %2 to i16
+; CHECK:  ret i16 %3
+define i16 @simple1(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fadd float %1, 1.0
+  %3 = fptoui float %2 to i16
+  ret i16 %3
+}
+
+; CHECK-LABEL: @simple2
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = sub i32 %1, 1
+; CHECK:  %3 = trunc i32 %2 to i8
+; CHECK:  ret i8 %3
+define i8 @simple2(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fsub float %1, 1.0
+  %3 = fptoui float %2 to i8
+  ret i8 %3
+}
+
+; CHECK-LABEL: @simple3
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = sub i32 %1, 1
+; CHECK:  ret i32 %2
+define i32 @simple3(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fsub float %1, 1.0
+  %3 = fptoui float %2 to i32
+  ret i32 %3
+}
+
+; CHECK-LABEL: @cmp
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = zext i8 %b to i32
+; CHECK:  %3 = icmp slt i32 %1, %2
+; CHECK:  ret i1 %3
+define i1 @cmp(i8 %a, i8 %b) {
+  %1 = uitofp i8 %a to float
+  %2 = uitofp i8 %b to float
+  %3 = fcmp ult float %1, %2
+  ret i1 %3
+}
+
+; CHECK-LABEL: @simple4
+; CHECK:  %1 = zext i32 %a to i64
+; CHECK:  %2 = add i64 %1, 1
+; CHECK:  %3 = trunc i64 %2 to i32
+; CHECK:  ret i32 %3
+define i32 @simple4(i32 %a) {
+  %1 = uitofp i32 %a to double
+  %2 = fadd double %1, 1.0
+  %3 = fptoui double %2 to i32
+  ret i32 %3
+}
+
+; CHECK-LABEL: @simple5
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = zext i8 %b to i32
+; CHECK:  %3 = add i32 %1, 1
+; CHECK:  %4 = mul i32 %3, %2
+; CHECK:  ret i32 %4
+define i32 @simple5(i8 %a, i8 %b) {
+  %1 = uitofp i8 %a to float
+  %2 = uitofp i8 %b to float
+  %3 = fadd float %1, 1.0
+  %4 = fmul float %3, %2
+  %5 = fptoui float %4 to i32
+  ret i32 %5
+}
+
+; The two chains don't interact - failure of one shouldn't
+; cause failure of the other.
+
+; CHECK-LABEL: @multi1
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = zext i8 %b to i32
+; CHECK:  %fc = uitofp i8 %c to float
+; CHECK:  %x1 = add i32 %1, %2
+; CHECK:  %z = fadd float %fc, %d
+; CHECK:  %w = fptoui float %z to i32
+; CHECK:  %r = add i32 %x1, %w
+; CHECK:  ret i32 %r
+define i32 @multi1(i8 %a, i8 %b, i8 %c, float %d) {
+  %fa = uitofp i8 %a to float
+  %fb = uitofp i8 %b to float
+  %fc = uitofp i8 %c to float
+  %x = fadd float %fa, %fb
+  %y = fptoui float %x to i32
+  %z = fadd float %fc, %d
+  %w = fptoui float %z to i32
+  %r = add i32 %y, %w
+  ret i32 %r
+}
+
+; CHECK-LABEL: @simple_negzero
+; CHECK:  %1 = zext i8 %a to i32
+; CHECK:  %2 = add i32 %1, 0
+; CHECK:  %3 = trunc i32 %2 to i16
+; CHECK:  ret i16 %3
+define i16 @simple_negzero(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fadd fast float %1, -0.0
+  %3 = fptoui float %2 to i16
+  ret i16 %3
+}
+
+; CHECK-LABEL: @simple_negative
+; CHECK: %1 = sext i8 %call to i32
+; CHECK: %mul1 = mul i32 %1, -3
+; CHECK: %2 = trunc i32 %mul1 to i8
+; CHECK: %conv3 = sext i8 %2 to i32
+; CHECK: ret i32 %conv3
+define i32 @simple_negative(i8 %call) {
+  %conv1 = sitofp i8 %call to float
+  %mul = fmul float %conv1, -3.000000e+00
+  %conv2 = fptosi float %mul to i8
+  %conv3 = sext i8 %conv2 to i32
+  ret i32 %conv3
+}
+
+;
+; Negative tests
+;
+
+; CHECK-LABEL: @neg_multi1
+; CHECK:  %fa = uitofp i8 %a to float
+; CHECK:  %fc = uitofp i8 %c to float
+; CHECK:  %x = fadd float %fa, %fc
+; CHECK:  %y = fptoui float %x to i32
+; CHECK:  %z = fadd float %fc, %d
+; CHECK:  %w = fptoui float %z to i32
+; CHECK:  %r = add i32 %y, %w
+; CHECK:  ret i32 %r
+; The two chains intersect, which means because one fails, no
+; transform can occur.
+define i32 @neg_multi1(i8 %a, i8 %b, i8 %c, float %d) {
+  %fa = uitofp i8 %a to float
+  %fc = uitofp i8 %c to float
+  %x = fadd float %fa, %fc
+  %y = fptoui float %x to i32
+  %z = fadd float %fc, %d
+  %w = fptoui float %z to i32
+  %r = add i32 %y, %w
+  ret i32 %r
+}
+
+; CHECK-LABEL: @neg_muld
+; CHECK:  %fa = uitofp i32 %a to double
+; CHECK:  %fb = uitofp i32 %b to double
+; CHECK:  %mul = fmul double %fa, %fb
+; CHECK:  %r = fptoui double %mul to i64
+; CHECK:  ret i64 %r
+; The i32 * i32 = i64, which has 64 bits, which is greater than the 52 bits
+; that can be exactly represented in a double.
+define i64 @neg_muld(i32 %a, i32 %b) {
+  %fa = uitofp i32 %a to double
+  %fb = uitofp i32 %b to double
+  %mul = fmul double %fa, %fb
+  %r = fptoui double %mul to i64
+  ret i64 %r
+}
+
+; CHECK-LABEL: @neg_mulf
+; CHECK:  %fa = uitofp i16 %a to float
+; CHECK:  %fb = uitofp i16 %b to float
+; CHECK:  %mul = fmul float %fa, %fb
+; CHECK:  %r = fptoui float %mul to i32
+; CHECK:  ret i32 %r
+; The i16 * i16 = i32, which can't be represented in a float, but can in a
+; double. This should fail, as the written code uses floats, not doubles so
+; the original result may be inaccurate.
+define i32 @neg_mulf(i16 %a, i16 %b) {
+  %fa = uitofp i16 %a to float
+  %fb = uitofp i16 %b to float
+  %mul = fmul float %fa, %fb
+  %r = fptoui float %mul to i32
+  ret i32 %r
+}
+
+; CHECK-LABEL: @neg_cmp
+; CHECK:  %1 = uitofp i8 %a to float
+; CHECK:  %2 = uitofp i8 %b to float
+; CHECK:  %3 = fcmp false float %1, %2
+; CHECK:  ret i1 %3
+; "false" doesn't have an icmp equivalent.
+define i1 @neg_cmp(i8 %a, i8 %b) {
+  %1 = uitofp i8 %a to float
+  %2 = uitofp i8 %b to float
+  %3 = fcmp false float %1, %2
+  ret i1 %3
+}
+
+; CHECK-LABEL: @neg_div
+; CHECK:  %1 = uitofp i8 %a to float
+; CHECK:  %2 = fdiv float %1, 1.0
+; CHECK:  %3 = fptoui float %2 to i16
+; CHECK:  ret i16 %3
+; Division isn't a supported operator.
+define i16 @neg_div(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fdiv float %1, 1.0
+  %3 = fptoui float %2 to i16
+  ret i16 %3
+}
+
+; CHECK-LABEL: @neg_remainder
+; CHECK:  %1 = uitofp i8 %a to float
+; CHECK:  %2 = fadd float %1, 1.2
+; CHECK:  %3 = fptoui float %2 to i16
+; CHECK:  ret i16 %3
+; 1.2 is not an integer.
+define i16 @neg_remainder(i8 %a) {
+  %1 = uitofp i8 %a to float
+  %2 = fadd float %1, 1.25
+  %3 = fptoui float %2 to i16
+  ret i16 %3
+}
+
+; CHECK-LABEL: @neg_toolarge
+; CHECK:  %1 = uitofp i80 %a to fp128
+; CHECK:  %2 = fadd fp128 %1, %1
+; CHECK:  %3 = fptoui fp128 %2 to i80
+; CHECK:  ret i80 %3
+; i80 > i64, which is the largest bitwidth handleable by default.
+define i80 @neg_toolarge(i80 %a) {
+  %1 = uitofp i80 %a to fp128
+  %2 = fadd fp128 %1, %1
+  %3 = fptoui fp128 %2 to i80
+  ret i80 %3
+}
+
+; CHECK-LABEL: @neg_calluser
+; CHECK: sitofp
+; CHECK: fcmp
+; The sequence %1..%3 cannot be converted because %4 uses %2.
+define i32 @neg_calluser(i32 %value) {
+  %1 = sitofp i32 %value to double
+  %2 = fadd double %1, 1.0
+  %3 = fcmp olt double %2, 0.000000e+00
+  %4 = tail call double @g(double %2)
+  %5 = fptosi double %4 to i32
+  %6 = zext i1 %3 to i32
+  %7 = add i32 %6, %5
+  ret i32 %7
+}
+declare double @g(double)
diff --git a/test/Transforms/Float2Int/float2int-optnone.ll b/test/Transforms/Float2Int/float2int-optnone.ll
new file mode 100644
index 000000000000..c1eeea7a82e5
--- /dev/null
+++ b/test/Transforms/Float2Int/float2int-optnone.ll
@@ -0,0 +1,17 @@
+; RUN: opt < %s -float2int -S | FileCheck %s
+;
+; Verify that pass float2int is not run on optnone functions.
+
+define i16 @simple1(i8 %a) #0 {
+; CHECK-LABEL: @simple1
+; CHECK:  %1 = uitofp i8 %a to float
+; CHECK-NEXT:  %2 = fadd float %1, 1.0
+; CHECK-NEXT:  %3 = fptoui float %2 to i16
+; CHECK-NEXT:  ret i16 %3
+  %1 = uitofp i8 %a to float
+  %2 = fadd float %1, 1.0
+  %3 = fptoui float %2 to i16
+  ret i16 %3
+}
+
+attributes #0 = { noinline optnone }
diff --git a/test/Transforms/Float2Int/toolarge.ll b/test/Transforms/Float2Int/toolarge.ll
new file mode 100644
index 000000000000..b5d778157685
--- /dev/null
+++ b/test/Transforms/Float2Int/toolarge.ll
@@ -0,0 +1,16 @@
+; RUN: opt < %s -float2int -float2int-max-integer-bw=256 -S | FileCheck %s
+
+; CHECK-LABEL: @neg_toolarge
+; CHECK:  %1 = uitofp i80 %a to fp128
+; CHECK:  %2 = fadd fp128 %1, %1
+; CHECK:  %3 = fptoui fp128 %2 to i80
+; CHECK:  ret i80 %3
+; fp128 has a 112-bit mantissa, which can hold an i80. But we only support
+; up to i64, so it should fail (even though the max integer bitwidth is 256).
+define i80 @neg_toolarge(i80 %a) {
+  %1 = uitofp i80 %a to fp128
+  %2 = fadd fp128 %1, %1
+  %3 = fptoui fp128 %2 to i80
+  ret i80 %3
+}
+
diff --git a/test/Transforms/FunctionAttrs/2008-09-03-ReadNone.ll b/test/Transforms/FunctionAttrs/2008-09-03-ReadNone.ll
index 36a765873487..ca05d63743b5 100644
--- a/test/Transforms/FunctionAttrs/2008-09-03-ReadNone.ll
+++ b/test/Transforms/FunctionAttrs/2008-09-03-ReadNone.ll
@@ -17,7 +17,7 @@ define i32 @g() readonly {
 
 ; CHECK: define i32 @h() #0
 define i32 @h() readnone {
-	%tmp = load i32* @x		; <i32> [#uses=1]
+	%tmp = load i32, i32* @x		; <i32> [#uses=1]
 	ret i32 %tmp
 }
 
diff --git a/test/Transforms/FunctionAttrs/2008-09-13-VolatileRead.ll b/test/Transforms/FunctionAttrs/2008-09-13-VolatileRead.ll
index b7e4d1f87261..fef872c794ae 100644
--- a/test/Transforms/FunctionAttrs/2008-09-13-VolatileRead.ll
+++ b/test/Transforms/FunctionAttrs/2008-09-13-VolatileRead.ll
@@ -4,6 +4,6 @@
 @g = global i32 0		; <i32*> [#uses=1]
 
 define i32 @f() {
-	%t = load volatile i32* @g		; <i32> [#uses=1]
+	%t = load volatile i32, i32* @g		; <i32> [#uses=1]
 	ret i32 %t
 }
diff --git a/test/Transforms/FunctionAttrs/2008-12-29-Constant.ll b/test/Transforms/FunctionAttrs/2008-12-29-Constant.ll
index 9655da45c646..e3a8f0161b86 100644
--- a/test/Transforms/FunctionAttrs/2008-12-29-Constant.ll
+++ b/test/Transforms/FunctionAttrs/2008-12-29-Constant.ll
@@ -3,6 +3,6 @@
 @s = external constant i8		; <i8*> [#uses=1]
 
 define i8 @f() {
-	%tmp = load i8* @s		; <i8> [#uses=1]
+	%tmp = load i8, i8* @s		; <i8> [#uses=1]
 	ret i8 %tmp
 }
diff --git a/test/Transforms/FunctionAttrs/2009-01-02-LocalStores.ll b/test/Transforms/FunctionAttrs/2009-01-02-LocalStores.ll
index 0cf1cb7c638d..ec1db0957283 100644
--- a/test/Transforms/FunctionAttrs/2009-01-02-LocalStores.ll
+++ b/test/Transforms/FunctionAttrs/2009-01-02-LocalStores.ll
@@ -2,7 +2,7 @@
 
 ; CHECK: define i32* @a(i32** nocapture readonly %p)
 define i32* @a(i32** %p) {
-	%tmp = load i32** %p
+	%tmp = load i32*, i32** %p
 	ret i32* %tmp
 }
 
diff --git a/test/Transforms/FunctionAttrs/2010-10-30-volatile.ll b/test/Transforms/FunctionAttrs/2010-10-30-volatile.ll
index 93991d21a1fa..1a64a8393804 100644
--- a/test/Transforms/FunctionAttrs/2010-10-30-volatile.ll
+++ b/test/Transforms/FunctionAttrs/2010-10-30-volatile.ll
@@ -5,6 +5,6 @@
 
 define void @foo() {
 ; CHECK: void @foo() {
-  %tmp = load volatile i32* @g
+  %tmp = load volatile i32, i32* @g
   ret void
 }
diff --git a/test/Transforms/FunctionAttrs/atomic.ll b/test/Transforms/FunctionAttrs/atomic.ll
index d5a8db7d53b9..bb867011cc2a 100644
--- a/test/Transforms/FunctionAttrs/atomic.ll
+++ b/test/Transforms/FunctionAttrs/atomic.ll
@@ -7,7 +7,7 @@ define i32 @test1(i32 %x) uwtable ssp {
 entry:
   %x.addr = alloca i32, align 4
   store atomic i32 %x, i32* %x.addr seq_cst, align 4
-  %r = load atomic i32* %x.addr seq_cst, align 4
+  %r = load atomic i32, i32* %x.addr seq_cst, align 4
   ret i32 %r
 }
 
@@ -15,7 +15,7 @@ entry:
 define i32 @test2(i32* %x) uwtable ssp {
 ; CHECK: define i32 @test2(i32* nocapture readonly %x) #1 {
 entry:
-  %r = load atomic i32* %x seq_cst, align 4
+  %r = load atomic i32, i32* %x seq_cst, align 4
   ret i32 %r
 }
 
diff --git a/test/Transforms/FunctionAttrs/nocapture.ll b/test/Transforms/FunctionAttrs/nocapture.ll
index d3842c8415ca..55a3dc4aa35b 100644
--- a/test/Transforms/FunctionAttrs/nocapture.ll
+++ b/test/Transforms/FunctionAttrs/nocapture.ll
@@ -39,8 +39,8 @@ define i1 @c5(i32* %q, i32 %bitno) {
 	%tmp2 = lshr i32 %tmp, %bitno
 	%bit = and i32 %tmp2, 1
         ; subtle escape mechanism follows
-	%lookup = getelementptr [2 x i1]* @lookup_table, i32 0, i32 %bit
-	%val = load i1* %lookup
+	%lookup = getelementptr [2 x i1], [2 x i1]* @lookup_table, i32 0, i32 %bit
+	%val = load i1, i1* %lookup
 	ret i1 %val
 }
 
@@ -64,14 +64,14 @@ define i1* @lookup_bit(i32* %q, i32 %bitno) readnone nounwind {
 	%tmp = ptrtoint i32* %q to i32
 	%tmp2 = lshr i32 %tmp, %bitno
 	%bit = and i32 %tmp2, 1
-	%lookup = getelementptr [2 x i1]* @lookup_table, i32 0, i32 %bit
+	%lookup = getelementptr [2 x i1], [2 x i1]* @lookup_table, i32 0, i32 %bit
 	ret i1* %lookup
 }
 
 ; CHECK: define i1 @c7(i32* readonly %q, i32 %bitno)
 define i1 @c7(i32* %q, i32 %bitno) {
 	%ptr = call i1* @lookup_bit(i32* %q, i32 %bitno)
-	%val = load i1* %ptr
+	%val = load i1, i1* %ptr
 	ret i1 %val
 }
 
@@ -85,7 +85,7 @@ l:
 	%y = phi i32* [ %q, %e ]
 	%tmp = bitcast i32* %x to i32*		; <i32*> [#uses=2]
 	%tmp2 = select i1 %b, i32* %tmp, i32* %y
-	%val = load i32* %tmp2		; <i32> [#uses=1]
+	%val = load i32, i32* %tmp2		; <i32> [#uses=1]
 	store i32 0, i32* %tmp
 	store i32* %y, i32** @g
 	ret i32 %val
@@ -100,7 +100,7 @@ l:
 	%y = phi i32* [ %q, %e ]
 	%tmp = addrspacecast i32 addrspace(1)* %x to i32*		; <i32*> [#uses=2]
 	%tmp2 = select i1 %b, i32* %tmp, i32* %y
-	%val = load i32* %tmp2		; <i32> [#uses=1]
+	%val = load i32, i32* %tmp2		; <i32> [#uses=1]
 	store i32 0, i32* %tmp
 	store i32* %y, i32** @g
 	ret i32 %val
@@ -188,7 +188,7 @@ declare void @test6_1(i8* %x6_1, i8* nocapture %y6_1, ...)
 
 ; CHECK: define void @test6_2(i8* %x6_2, i8* nocapture %y6_2, i8* %z6_2)
 define void @test6_2(i8* %x6_2, i8* %y6_2, i8* %z6_2) {
-  call void (i8*, i8*, ...)* @test6_1(i8* %x6_2, i8* %y6_2, i8* %z6_2)
+  call void (i8*, i8*, ...) @test6_1(i8* %x6_2, i8* %y6_2, i8* %z6_2)
   store i32* null, i32** @g
   ret void
 }
diff --git a/test/Transforms/FunctionAttrs/optnone-simple.ll b/test/Transforms/FunctionAttrs/optnone-simple.ll
index 9d0f8e3710a6..beaa588da503 100644
--- a/test/Transforms/FunctionAttrs/optnone-simple.ll
+++ b/test/Transforms/FunctionAttrs/optnone-simple.ll
@@ -15,8 +15,8 @@ entry:
   %b.addr = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
   store i32 %b, i32* %b.addr, align 4
-  %0 = load i32* %a.addr, align 4
-  %1 = load i32* %b.addr, align 4
+  %0 = load i32, i32* %a.addr, align 4
+  %1 = load i32, i32* %b.addr, align 4
   %add = add nsw i32 %0, %1
   ret i32 %add
 }
@@ -33,8 +33,8 @@ entry:
   %b.addr = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
   store i32 %b, i32* %b.addr, align 4
-  %0 = load i32* %a.addr, align 4
-  %1 = load i32* %b.addr, align 4
+  %0 = load i32, i32* %a.addr, align 4
+  %1 = load i32, i32* %b.addr, align 4
   %add = add nsw i32 %0, %1
   ret i32 %add
 }
@@ -57,8 +57,8 @@ entry:
   %b.addr = alloca float, align 4
   store float %a, float* %a.addr, align 4
   store float %b, float* %b.addr, align 4
-  %0 = load float* %a.addr, align 4
-  %1 = load float* %b.addr, align 4
+  %0 = load float, float* %a.addr, align 4
+  %1 = load float, float* %b.addr, align 4
   %sub = fsub float %0, %1
   ret float %sub
 }
@@ -75,8 +75,8 @@ entry:
   %b.addr = alloca float, align 4
   store float %a, float* %a.addr, align 4
   store float %b, float* %b.addr, align 4
-  %0 = load float* %a.addr, align 4
-  %1 = load float* %b.addr, align 4
+  %0 = load float, float* %a.addr, align 4
+  %1 = load float, float* %b.addr, align 4
   %sub = fsub float %0, %1
   ret float %sub
 }
@@ -100,8 +100,8 @@ entry:
   %b.addr = alloca <4 x float>, align 16
   store <4 x float> %a, <4 x float>* %a.addr, align 16
   store <4 x float> %b, <4 x float>* %b.addr, align 16
-  %0 = load <4 x float>* %a.addr, align 16
-  %1 = load <4 x float>* %b.addr, align 16
+  %0 = load <4 x float>, <4 x float>* %a.addr, align 16
+  %1 = load <4 x float>, <4 x float>* %b.addr, align 16
   %mul = fmul <4 x float> %0, %1
   ret <4 x float> %mul
 }
@@ -118,8 +118,8 @@ entry:
   %b.addr = alloca <4 x float>, align 16
   store <4 x float> %a, <4 x float>* %a.addr, align 16
   store <4 x float> %b, <4 x float>* %b.addr, align 16
-  %0 = load <4 x float>* %a.addr, align 16
-  %1 = load <4 x float>* %b.addr, align 16
+  %0 = load <4 x float>, <4 x float>* %a.addr, align 16
+  %1 = load <4 x float>, <4 x float>* %b.addr, align 16
   %mul = fmul <4 x float> %0, %1
   ret <4 x float> %mul
 }
diff --git a/test/Transforms/FunctionAttrs/readattrs.ll b/test/Transforms/FunctionAttrs/readattrs.ll
index b4e904cf9b04..7f22e6f2a2c5 100644
--- a/test/Transforms/FunctionAttrs/readattrs.ll
+++ b/test/Transforms/FunctionAttrs/readattrs.ll
@@ -5,7 +5,7 @@ declare void @test1_1(i8* %x1_1, i8* readonly %y1_1, ...)
 
 ; CHECK: define void @test1_2(i8* %x1_2, i8* readonly %y1_2, i8* %z1_2)
 define void @test1_2(i8* %x1_2, i8* %y1_2, i8* %z1_2) {
-  call void (i8*, i8*, ...)* @test1_1(i8* %x1_2, i8* %y1_2, i8* %z1_2)
+  call void (i8*, i8*, ...) @test1_1(i8* %x1_2, i8* %y1_2, i8* %z1_2)
   store i32 0, i32* @x
   ret void
 }
diff --git a/test/Transforms/FunctionAttrs/readnone.ll b/test/Transforms/FunctionAttrs/readnone.ll
new file mode 100644
index 000000000000..eddcdd2641d5
--- /dev/null
+++ b/test/Transforms/FunctionAttrs/readnone.ll
@@ -0,0 +1,13 @@
+; RUN: opt < %s -functionattrs -S | FileCheck %s
+
+; CHECK: define void @bar(i8* nocapture readnone)
+define void @bar(i8* readonly) {
+  call void @foo(i8* %0)
+    ret void
+}
+
+; CHECK: define void @foo(i8* nocapture readnone)
+define void @foo(i8* readonly) {
+  call void @bar(i8* %0)
+  ret void
+}
diff --git a/test/Transforms/GCOVProfiling/function-numbering.ll b/test/Transforms/GCOVProfiling/function-numbering.ll
index 487f4ca01a6b..5a704e4d047b 100644
--- a/test/Transforms/GCOVProfiling/function-numbering.ll
+++ b/test/Transforms/GCOVProfiling/function-numbering.ll
@@ -6,7 +6,7 @@
 ; RUN: cat %s %t1 > %t2
 
 ; RUN: opt -insert-gcov-profiling -S < %t2 | FileCheck --check-prefix GCDA %s
-; RUN: llvm-cov -n -dump %T/function-numbering.gcno 2>&1 | FileCheck --check-prefix GCNO %s
+; RUN: llvm-cov gcov -n -dump %T/function-numbering.gcno 2>&1 | FileCheck --check-prefix GCNO %s
 
 target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-apple-macosx10.10.0"
@@ -15,8 +15,8 @@ target triple = "x86_64-apple-macosx10.10.0"
 ; GCDA-NOT: @{{[0-9]+}} = private unnamed_addr constant .* c"bar\00"
 ; GCDA: @[[BAZ:[0-9]+]] = private unnamed_addr constant [4 x i8] c"baz\00"
 ; GCDA: define internal void @__llvm_gcov_writeout()
-; GCDA: call void @llvm_gcda_emit_function(i32 0, i8* getelementptr inbounds ([4 x i8]* @[[FOO]]
-; GCDA: call void @llvm_gcda_emit_function(i32 1, i8* getelementptr inbounds ([4 x i8]* @[[BAZ]]
+; GCDA: call void @llvm_gcda_emit_function(i32 0, i8* getelementptr inbounds ([4 x i8], [4 x i8]* @[[FOO]]
+; GCDA: call void @llvm_gcda_emit_function(i32 1, i8* getelementptr inbounds ([4 x i8], [4 x i8]* @[[BAZ]]
 
 ; GCNO: == foo (0) @
 ; GCNO-NOT: == bar ({{[0-9]+}}) @
@@ -40,17 +40,17 @@ define void @baz() {
 !llvm.module.flags = !{!9, !10}
 !llvm.ident = !{!11}
 
-!0 = !{!"0x11\0012\00clang version 3.6.0 \000\00\000\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [function-numbering.ll] [DW_LANG_C99]
-!1 = !{!".../llvm/test/Transforms/GCOVProfiling/function-numbering.ll", !""}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.6.0 ", isOptimized: false, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: ".../llvm/test/Transforms/GCOVProfiling/function-numbering.ll", directory: "")
 !2 = !{}
 !3 = !{!4, !7, !8}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\000\000\000\001", !1, !5, !6, null, void ()* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}    ; [ DW_TAG_file_type ] [/Users/bogner/build/llvm-debug//tmp/foo.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!7 = !{!"0x2e\00bar\00bar\00\002\000\001\000\000\000\000\002", !1, !5, !6, null, void ()* @bar, null, null, !2} ; [ DW_TAG_subprogram ] [line 2] [def] [bar]
-!8 = !{!"0x2e\00baz\00baz\00\003\000\001\000\000\000\000\003", !1, !5, !6, null, void ()* @baz, null, null, !2} ; [ DW_TAG_subprogram ] [line 3] [def] [baz]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: void ()* @foo, variables: !2)
+!5 = !DIFile(filename: ".../llvm/test/Transforms/GCOVProfiling/function-numbering.ll", directory: "")
+!6 = !DISubroutineType(types: !2)
+!7 = !DISubprogram(name: "bar", line: 2, isLocal: false, isDefinition: true, isOptimized: false, scopeLine: 2, file: !1, scope: !5, type: !6, function: void ()* @bar, variables: !2)
+!8 = !DISubprogram(name: "baz", line: 3, isLocal: false, isDefinition: true, isOptimized: false, scopeLine: 3, file: !1, scope: !5, type: !6, function: void ()* @baz, variables: !2)
 !9 = !{i32 2, !"Dwarf Version", i32 2}
-!10 = !{i32 2, !"Debug Info Version", i32 2}
+!10 = !{i32 2, !"Debug Info Version", i32 3}
 !11 = !{!"clang version 3.6.0 "}
-!12 = !MDLocation(line: 1, column: 13, scope: !4)
-!13 = !MDLocation(line: 3, column: 13, scope: !8)
+!12 = !DILocation(line: 1, column: 13, scope: !4)
+!13 = !DILocation(line: 3, column: 13, scope: !8)
diff --git a/test/Transforms/GCOVProfiling/global-ctor.ll b/test/Transforms/GCOVProfiling/global-ctor.ll
index 9a9b7cecce7a..29c46d6c2107 100644
--- a/test/Transforms/GCOVProfiling/global-ctor.ll
+++ b/test/Transforms/GCOVProfiling/global-ctor.ll
@@ -38,19 +38,19 @@ attributes #1 = { "less-precise-fpmad"="false" "no-frame-pointer-elim"="false" "
 !llvm.gcov = !{!16}
 !llvm.ident = !{!12}
 
-!0 = !{!"0x11\004\00clang version 3.5.0 (trunk 210217)\000\00\000\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [/home/nlewycky/<stdin>] [DW_LANG_C_plus_plus]
-!1 = !{!"<stdin>", !"/home/nlewycky"}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0 (trunk 210217)", isOptimized: false, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "<stdin>", directory: "/home/nlewycky")
 !2 = !{}
 !3 = !{!4, !8}
-!4 = !{!"0x2e\00__cxx_global_var_init\00__cxx_global_var_init\00\002\001\001\000\006\00256\000\002", !5, !6, !7, null, void ()* @__cxx_global_var_init, null, null, !2} ; [ DW_TAG_subprogram ] [line 2] [local] [def] [__cxx_global_var_init]
-!5 = !{!"global-ctor.ll", !"/home/nlewycky"}
-!6 = !{!"0x29", !5}          ; [ DW_TAG_file_type ] [/home/nlewycky/global-ctor.ll]
-!7 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!8 = !{!"0x2e\00\00\00_GLOBAL__sub_I_global-ctor.ll\000\001\001\000\006\0064\000\000", !1, !9, !7, null, void ()* @_GLOBAL__sub_I_global-ctor.ll, null, null, !2} ; [ DW_TAG_subprogram ] [line 0] [local] [def]
-!9 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [/home/nlewycky/<stdin>]
+!4 = !DISubprogram(name: "__cxx_global_var_init", line: 2, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 2, file: !5, scope: !6, type: !7, function: void ()* @__cxx_global_var_init, variables: !2)
+!5 = !DIFile(filename: "global-ctor.ll", directory: "/home/nlewycky")
+!6 = !DIFile(filename: "global-ctor.ll", directory: "/home/nlewycky")
+!7 = !DISubroutineType(types: !2)
+!8 = !DISubprogram(name: "", linkageName: "_GLOBAL__sub_I_global-ctor.ll", isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagArtificial, isOptimized: false, file: !1, scope: !9, type: !7, function: void ()* @_GLOBAL__sub_I_global-ctor.ll, variables: !2)
+!9 = !DIFile(filename: "<stdin>", directory: "/home/nlewycky")
 !10 = !{i32 2, !"Dwarf Version", i32 4}
-!11 = !{i32 2, !"Debug Info Version", i32 2}
+!11 = !{i32 2, !"Debug Info Version", i32 3}
 !12 = !{!"clang version 3.5.0 (trunk 210217)"}
-!13 = !MDLocation(line: 2, scope: !4)
-!14 = !MDLocation(line: 0, scope: !15)
-!15 = !{!"0xb\000", !5, !8} ; [ DW_TAG_lexical_block ] [/home/nlewycky/global-ctor.ll]
+!13 = !DILocation(line: 2, scope: !4)
+!14 = !DILocation(line: 0, scope: !15)
+!15 = !DILexicalBlockFile(discriminator: 0, file: !5, scope: !8)
diff --git a/test/Transforms/GCOVProfiling/linezero.ll b/test/Transforms/GCOVProfiling/linezero.ll
index cf0fcd2cc820..9e172b752d78 100644
--- a/test/Transforms/GCOVProfiling/linezero.ll
+++ b/test/Transforms/GCOVProfiling/linezero.ll
@@ -1,6 +1,5 @@
-; RUN: sed -e 's|PATTERN|%/T|g' < %s > %t1
-; RUN: opt -insert-gcov-profiling -disable-output < %t1
-; RUN: rm %T/linezero.gcno %t1
+; RUN: sed -e 's|PATTERN|%/T|g' %s | opt -insert-gcov-profiling -disable-output
+; RUN: rm %T/linezero.gcno
 
 ; This is a crash test.
 
@@ -19,38 +18,38 @@ entry:
   %__begin = alloca i8*, align 8
   %__end = alloca i8*, align 8
   %spec = alloca i8, align 1
-  call void @llvm.dbg.declare(metadata %struct.vector** %__range, metadata !27, metadata !{}), !dbg !30
+  call void @llvm.dbg.declare(metadata %struct.vector** %__range, metadata !27, metadata !DIExpression()), !dbg !30
   br label %0
 
 ; <label>:0                                       ; preds = %entry
   call void @_Z13TagFieldSpecsv(), !dbg !31
   store %struct.vector* %ref.tmp, %struct.vector** %__range, align 8, !dbg !31
-  call void @llvm.dbg.declare(metadata i8** %__begin, metadata !32, metadata !{}), !dbg !30
-  %1 = load %struct.vector** %__range, align 8, !dbg !31
+  call void @llvm.dbg.declare(metadata i8** %__begin, metadata !32, metadata !DIExpression()), !dbg !30
+  %1 = load %struct.vector*, %struct.vector** %__range, align 8, !dbg !31
   %call = call i8* @_ZN6vector5beginEv(%struct.vector* %1), !dbg !31
   store i8* %call, i8** %__begin, align 8, !dbg !31
-  call void @llvm.dbg.declare(metadata i8** %__end, metadata !33, metadata !{}), !dbg !30
-  %2 = load %struct.vector** %__range, align 8, !dbg !31
+  call void @llvm.dbg.declare(metadata i8** %__end, metadata !33, metadata !DIExpression()), !dbg !30
+  %2 = load %struct.vector*, %struct.vector** %__range, align 8, !dbg !31
   %call1 = call i8* @_ZN6vector3endEv(%struct.vector* %2), !dbg !31
   store i8* %call1, i8** %__end, align 8, !dbg !31
   br label %for.cond, !dbg !31
 
 for.cond:                                         ; preds = %for.inc, %0
-  %3 = load i8** %__begin, align 8, !dbg !34
-  %4 = load i8** %__end, align 8, !dbg !34
+  %3 = load i8*, i8** %__begin, align 8, !dbg !34
+  %4 = load i8*, i8** %__end, align 8, !dbg !34
   %cmp = icmp ne i8* %3, %4, !dbg !34
   br i1 %cmp, label %for.body, label %for.end, !dbg !34
 
 for.body:                                         ; preds = %for.cond
-  call void @llvm.dbg.declare(metadata i8* %spec, metadata !37, metadata !{}), !dbg !31
-  %5 = load i8** %__begin, align 8, !dbg !38
-  %6 = load i8* %5, align 1, !dbg !38
+  call void @llvm.dbg.declare(metadata i8* %spec, metadata !37, metadata !DIExpression()), !dbg !31
+  %5 = load i8*, i8** %__begin, align 8, !dbg !38
+  %6 = load i8, i8* %5, align 1, !dbg !38
   store i8 %6, i8* %spec, align 1, !dbg !38
   br label %for.inc, !dbg !38
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i8** %__begin, align 8, !dbg !40
-  %incdec.ptr = getelementptr inbounds i8* %7, i32 1, !dbg !40
+  %7 = load i8*, i8** %__begin, align 8, !dbg !40
+  %incdec.ptr = getelementptr inbounds i8, i8* %7, i32 1, !dbg !40
   store i8* %incdec.ptr, i8** %__begin, align 8, !dbg !40
   br label %for.cond, !dbg !40
 
@@ -59,7 +58,7 @@ for.end:                                          ; preds = %for.cond
   unreachable, !dbg !42
 
 return:                                           ; No predecessors!
-  %8 = load i32* %retval, !dbg !44
+  %8 = load i32, i32* %retval, !dbg !44
   ret i32 %8, !dbg !44
 }
 
@@ -94,49 +93,49 @@ attributes #3 = { noreturn nounwind }
 !llvm.gcov = !{!25}
 !llvm.ident = !{!26}
 
-!0 = !{!"0x11\004\00clang version 3.5.0 (trunk 209871)\000\00\000\00\001", !1, !2, !3, !14, !2, !2} ; [ DW_TAG_compile_unit ] [<stdin>] [DW_LANG_C_plus_plus]
-!1 = !{!"<stdin>", !"PATTERN"}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0 (trunk 209871)", isOptimized: false, emissionKind: 1, file: !1, enums: !2, retainedTypes: !3, subprograms: !14, globals: !2, imports: !2)
+!1 = !DIFile(filename: "<stdin>", directory: "PATTERN")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x13\00vector\0021\008\008\000\000\000", !5, null, null, !6, null, null, !"_ZTS6vector"} ; [ DW_TAG_structure_type ] [vector] [line 21, size 8, align 8, offset 0] [def] [from ]
-!5 = !{!"linezero.cc", !"PATTERN"}
+!4 = !DICompositeType(tag: DW_TAG_structure_type, name: "vector", line: 21, size: 8, align: 8, file: !5, elements: !6, identifier: "_ZTS6vector")
+!5 = !DIFile(filename: "linezero.cc", directory: "PATTERN")
 !6 = !{!7, !13}
-!7 = !{!"0x2e\00begin\00begin\00_ZN6vector5beginEv\0025\000\000\000\006\00256\000\0025", !5, !"_ZTS6vector", !8, null, null, null, i32 0, null} ; [ DW_TAG_subprogram ] [line 25] [begin]
-!8 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !9, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!7 = !DISubprogram(name: "begin", linkageName: "_ZN6vector5beginEv", line: 25, isLocal: false, isDefinition: false, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 25, file: !5, scope: !"_ZTS6vector", type: !8)
+!8 = !DISubroutineType(types: !9)
 !9 = !{!10, !12}
-!10 = !{!"0xf\00\000\0064\0064\000\000", null, null, !11} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from char]
-!11 = !{!"0x24\00char\000\008\008\000\000\006", null, null} ; [ DW_TAG_base_type ] [char] [line 0, size 8, align 8, offset 0, enc DW_ATE_signed_char]
-!12 = !{!"0xf\00\000\0064\0064\000\001088", null, null, !"_ZTS6vector"} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [artificial] [from _ZTS6vector]
-!13 = !{!"0x2e\00end\00end\00_ZN6vector3endEv\0026\000\000\000\006\00256\000\0026", !5, !"_ZTS6vector", !8, null, null, null, i32 0, null} ; [ DW_TAG_subprogram ] [line 26] [end]
+!10 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !11)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "char", size: 8, align: 8, encoding: DW_ATE_signed_char)
+!12 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, flags: DIFlagArtificial | DIFlagObjectPointer, baseType: !"_ZTS6vector")
+!13 = !DISubprogram(name: "end", linkageName: "_ZN6vector3endEv", line: 26, isLocal: false, isDefinition: false, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 26, file: !5, scope: !"_ZTS6vector", type: !8)
 !14 = !{!15, !20}
-!15 = !{!"0x2e\00test\00test\00_Z4testv\0050\000\001\000\006\00256\000\0050", !5, !16, !17, null, i32 ()* @_Z4testv, null, null, !2} ; [ DW_TAG_subprogram ] [line 50] [def] [test]
-!16 = !{!"0x29", !5}         ; [ DW_TAG_file_type ] [./linezero.cc]
-!17 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !18, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!15 = !DISubprogram(name: "test", linkageName: "_Z4testv", line: 50, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 50, file: !5, scope: !16, type: !17, function: i32 ()* @_Z4testv, variables: !2)
+!16 = !DIFile(filename: "linezero.cc", directory: "PATTERN")
+!17 = !DISubroutineType(types: !18)
 !18 = !{!19}
-!19 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!20 = !{!"0x2e\00f1\00f1\00_Z2f1v\0054\000\001\000\006\00256\000\0054", !5, !16, !21, null, void ()* @_Z2f1v, null, null, !2} ; [ DW_TAG_subprogram ] [line 54] [def] [f1]
-!21 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !22, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!19 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!20 = !DISubprogram(name: "f1", linkageName: "_Z2f1v", line: 54, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 54, file: !5, scope: !16, type: !21, function: void ()* @_Z2f1v, variables: !2)
+!21 = !DISubroutineType(types: !22)
 !22 = !{null}
 !23 = !{i32 2, !"Dwarf Version", i32 4}
-!24 = !{i32 2, !"Debug Info Version", i32 2}
+!24 = !{i32 2, !"Debug Info Version", i32 3}
 !25 = !{!"PATTERN/linezero.o", !0}
 !26 = !{!"clang version 3.5.0 (trunk 209871)"}
-!27 = !{!"0x100\00__range\000\0064", !28, null, !29} ; [ DW_TAG_auto_variable ] [__range] [line 0]
-!28 = !{!"0xb\0051\000\000", !5, !15} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!29 = !{!"0x42\00\000\000\000\000\000", null, null, !"_ZTS6vector"} ; [ DW_TAG_rvalue_reference_type ] [line 0, size 0, align 0, offset 0] [from _ZTS6vector]
-!30 = !MDLocation(line: 0, scope: !28)
-!31 = !MDLocation(line: 51, scope: !28)
-!32 = !{!"0x100\00__begin\000\0064", !28, null, !10} ; [ DW_TAG_auto_variable ] [__begin] [line 0]
-!33 = !{!"0x100\00__end\000\0064", !28, null, !10} ; [ DW_TAG_auto_variable ] [__end] [line 0]
-!34 = !MDLocation(line: 51, scope: !35)
-!35 = !{!"0xb\0051\000\005", !5, !36} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!36 = !{!"0xb\0051\000\001", !5, !28} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!37 = !{!"0x100\00spec\0051\000", !28, !16, !11} ; [ DW_TAG_auto_variable ] [spec] [line 51]
-!38 = !MDLocation(line: 51, scope: !39)
-!39 = !{!"0xb\0051\000\002", !5, !28} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!40 = !MDLocation(line: 51, scope: !41)
-!41 = !{!"0xb\0051\000\004", !5, !28} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!42 = !MDLocation(line: 51, scope: !43)
-!43 = !{!"0xb\0051\000\003", !5, !28} ; [ DW_TAG_lexical_block ] [./linezero.cc]
-!44 = !MDLocation(line: 52, scope: !15)
-!45 = !MDLocation(line: 54, scope: !20)
+!27 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "__range", flags: DIFlagArtificial, scope: !28, type: !29)
+!28 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !15)
+!29 = !DIDerivedType(tag: DW_TAG_rvalue_reference_type, baseType: !"_ZTS6vector")
+!30 = !DILocation(line: 0, scope: !28)
+!31 = !DILocation(line: 51, scope: !28)
+!32 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "__begin", flags: DIFlagArtificial, scope: !28, type: !10)
+!33 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "__end", flags: DIFlagArtificial, scope: !28, type: !10)
+!34 = !DILocation(line: 51, scope: !35)
+!35 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !36)
+!36 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !28)
+!37 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "spec", line: 51, scope: !28, file: !16, type: !11)
+!38 = !DILocation(line: 51, scope: !39)
+!39 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !28)
+!40 = !DILocation(line: 51, scope: !41)
+!41 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !28)
+!42 = !DILocation(line: 51, scope: !43)
+!43 = distinct !DILexicalBlock(line: 51, column: 0, file: !5, scope: !28)
+!44 = !DILocation(line: 52, scope: !15)
+!45 = !DILocation(line: 54, scope: !20)
diff --git a/test/Transforms/GCOVProfiling/linkagename.ll b/test/Transforms/GCOVProfiling/linkagename.ll
index c30d4a61ced4..7a4119802c9a 100644
--- a/test/Transforms/GCOVProfiling/linkagename.ll
+++ b/test/Transforms/GCOVProfiling/linkagename.ll
@@ -13,15 +13,15 @@ entry:
 !llvm.module.flags = !{!10}
 !llvm.gcov = !{!9}
 
-!0 = !{!"0x11\004\00clang version 3.3 (trunk 177323)\000\00\000\00\000", !2, !3, !3, !4, !3,  !3} ; [ DW_TAG_compile_unit ] [/home/nlewycky/hello.cc] [DW_LANG_C_plus_plus]
-!1 = !{!"0x29", !2}          ; [ DW_TAG_file_type ] [/home/nlewycky/hello.cc]
-!2 = !{!"hello.cc", !"/home/nlewycky"}
-!3 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.3 (trunk 177323)", isOptimized: false, emissionKind: 0, file: !2, enums: !3, retainedTypes: !3, subprograms: !4, globals: !3, imports:  !3)
+!1 = !DIFile(filename: "hello.cc", directory: "/home/nlewycky")
+!2 = !DIFile(filename: "hello.cc", directory: "/home/nlewycky")
+!3 = !{}
 !4 = !{!5}
-!5 = !{!"0x2e\00foo\00foo\00_Z3foov\001\000\001\000\006\00256\000\001", !1, !1, !6, null, void ()* @_Z3foov, null, null, !3} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, i32 0} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!5 = !DISubprogram(name: "foo", linkageName: "_Z3foov", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !1, type: !6, function: void ()* @_Z3foov, variables: !3)
+!6 = !DISubroutineType(types: !7)
 !7 = !{null}
-!8 = !MDLocation(line: 1, scope: !5)
+!8 = !DILocation(line: 1, scope: !5)
 
 
-!10 = !{i32 1, !"Debug Info Version", i32 2}
+!10 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/GCOVProfiling/return-block.ll b/test/Transforms/GCOVProfiling/return-block.ll
index fce374a0a18b..38b5b75e3c2d 100644
--- a/test/Transforms/GCOVProfiling/return-block.ll
+++ b/test/Transforms/GCOVProfiling/return-block.ll
@@ -18,13 +18,13 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define void @test() #0 {
 entry:
-  tail call void (...)* @f() #2, !dbg !14
-  %0 = load i32* @A, align 4, !dbg !15
+  tail call void (...) @f() #2, !dbg !14
+  %0 = load i32, i32* @A, align 4, !dbg !15
   %tobool = icmp eq i32 %0, 0, !dbg !15
   br i1 %tobool, label %if.end, label %if.then, !dbg !15
 
 if.then:                                          ; preds = %entry
-  tail call void (...)* @g() #2, !dbg !16
+  tail call void (...) @g() #2, !dbg !16
   br label %if.end, !dbg !16
 
 if.end:                                           ; preds = %entry, %if.then
@@ -44,25 +44,25 @@ attributes #2 = { nounwind }
 !llvm.module.flags = !{!11, !12}
 !llvm.ident = !{!13}
 
-!0 = !{!"0x11\0012\00clang version 3.6.0 (trunk 223182)\001\00\000\00\001", !1, !2, !2, !3, !8, !2} ; [ DW_TAG_compile_unit ] [return-block.c] [DW_LANG_C99]
-!1 = !{!".../llvm/test/Transforms/GCOVProfiling/return-block.ll", !""}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.6.0 (trunk 223182)", isOptimized: true, emissionKind: 1, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !8, imports: !2)
+!1 = !DIFile(filename: ".../llvm/test/Transforms/GCOVProfiling/return-block.ll", directory: "")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00test\00test\00\005\000\001\000\000\000\001\005", !1, !5, !6, null, void ()* @test, null, null, !2} ; [ DW_TAG_subprogram ] [line 5] [def] [test]
-!5 = !{!"0x29", !1}    ; [ DW_TAG_file_type ] [return-block.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "test", line: 5, isLocal: false, isDefinition: true, isOptimized: true, scopeLine: 5, file: !1, scope: !5, type: !6, function: void ()* @test, variables: !2)
+!5 = !DIFile(filename: ".../llvm/test/Transforms/GCOVProfiling/return-block.ll", directory: "")
+!6 = !DISubroutineType(types: !7)
 !7 = !{null}
 !8 = !{!9}
-!9 = !{!"0x34\00A\00A\00\003\000\001", null, !5, !10, i32* @A, null} ; [ DW_TAG_variable ] [A] [line 3] [def]
-!10 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
+!9 = !DIGlobalVariable(name: "A", line: 3, isLocal: false, isDefinition: true, scope: null, file: !5, type: !10, variable: i32* @A)
+!10 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !11 = !{i32 2, !"Dwarf Version", i32 4}
-!12 = !{i32 2, !"Debug Info Version", i32 2}
+!12 = !{i32 2, !"Debug Info Version", i32 3}
 !13 = !{!"clang version 3.6.0 (trunk 223182)"}
-!14 = !MDLocation(line: 6, column: 3, scope: !4)
-!15 = !MDLocation(line: 7, column: 7, scope: !4)
-!16 = !MDLocation(line: 8, column: 5, scope: !17)
-!17 = !{!"0xb\007\007\000", !1, !4} ; [ DW_TAG_lexical_block ] [return-block.c]
-!18 = !MDLocation(line: 9, column: 1, scope: !4)
+!14 = !DILocation(line: 6, column: 3, scope: !4)
+!15 = !DILocation(line: 7, column: 7, scope: !4)
+!16 = !DILocation(line: 8, column: 5, scope: !17)
+!17 = distinct !DILexicalBlock(line: 7, column: 7, file: !1, scope: !4)
+!18 = !DILocation(line: 9, column: 1, scope: !4)
 
 ; There should be no destination edges for the exit block.
 ; CHECK: Block : 1 Counter : 0
diff --git a/test/Transforms/GCOVProfiling/version.ll b/test/Transforms/GCOVProfiling/version.ll
index 9436bd657da3..487e72123cb0 100644
--- a/test/Transforms/GCOVProfiling/version.ll
+++ b/test/Transforms/GCOVProfiling/version.ll
@@ -16,15 +16,15 @@ define void @test() {
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!12}
 
-!0 = !{!"0x11\004\00clang version 3.3 (trunk 176994)\000\00\000\00\000", !11, !3, !3, !4, !3, null} ; [ DW_TAG_compile_unit ] [./version] [DW_LANG_C_plus_plus]
-!2 = !{!"0x29", !11} ; [ DW_TAG_file_type ]
-!3 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.3 (trunk 176994)", isOptimized: false, emissionKind: 0, file: !11, enums: !3, retainedTypes: !3, subprograms: !4, globals: !3)
+!2 = !DIFile(filename: "version", directory: "/usr/local/google/home/nlewycky")
+!3 = !{}
 !4 = !{!5}
-!5 = !{!"0x2e\00test\00test\00\001\000\001\000\006\00256\000\001", !10, !6, !7, null, void ()* @test, null, null, !3} ; [ DW_TAG_subprogram ] [line 1] [def] [test]
-!6 = !{!"0x29", !10} ; [ DW_TAG_file_type ]
-!7 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !3, i32 0} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!8 = !MDLocation(line: 1, scope: !5)
+!5 = !DISubprogram(name: "test", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !10, scope: !6, type: !7, function: void ()* @test, variables: !3)
+!6 = !DIFile(filename: "<stdin>", directory: ".")
+!7 = !DISubroutineType(types: !{null})
+!8 = !DILocation(line: 1, scope: !5)
 ;; !9 is added through the echo line at the top.
-!10 = !{!"<stdin>", !"."}
-!11 = !{!"version", !"/usr/local/google/home/nlewycky"}
-!12 = !{i32 1, !"Debug Info Version", i32 2}
+!10 = !DIFile(filename: "<stdin>", directory: ".")
+!11 = !DIFile(filename: "version", directory: "/usr/local/google/home/nlewycky")
+!12 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/GVN/2007-07-25-DominatedLoop.ll b/test/Transforms/GVN/2007-07-25-DominatedLoop.ll
index ad580ce1677d..10d1e229b12a 100644
--- a/test/Transforms/GVN/2007-07-25-DominatedLoop.ll
+++ b/test/Transforms/GVN/2007-07-25-DominatedLoop.ll
@@ -71,11 +71,11 @@ cond_true23.i:		; preds = %Perl_safefree.exit68
 	ret void
 
 cond_next150:		; preds = %Perl_safefree.exit68
-	%tmp16092 = load i32* @PL_sv_count, align 4		; <i32> [#uses=0]
+	%tmp16092 = load i32, i32* @PL_sv_count, align 4		; <i32> [#uses=0]
 	br label %cond_next165
 
 bb157:		; preds = %cond_next165
-	%tmp158 = load i32* @PL_sv_count, align 4		; <i32> [#uses=0]
+	%tmp158 = load i32, i32* @PL_sv_count, align 4		; <i32> [#uses=0]
 	br label %cond_next165
 
 cond_next165:		; preds = %bb157, %cond_next150
diff --git a/test/Transforms/GVN/2007-07-25-InfiniteLoop.ll b/test/Transforms/GVN/2007-07-25-InfiniteLoop.ll
index 7e9c982de51c..0ffb34c39b4f 100644
--- a/test/Transforms/GVN/2007-07-25-InfiniteLoop.ll
+++ b/test/Transforms/GVN/2007-07-25-InfiniteLoop.ll
@@ -9,6 +9,6 @@ entry:
 	br label %bb
 
 bb:		; preds = %bb, %entry
-	%tmp10 = load %struct.INT2** @blkshifts, align 4		; <%struct.INT2*> [#uses=0]
+	%tmp10 = load %struct.INT2*, %struct.INT2** @blkshifts, align 4		; <%struct.INT2*> [#uses=0]
 	br label %bb
 }
diff --git a/test/Transforms/GVN/2007-07-25-Loop.ll b/test/Transforms/GVN/2007-07-25-Loop.ll
index 6a9f58e02f65..54c0d982de4d 100644
--- a/test/Transforms/GVN/2007-07-25-Loop.ll
+++ b/test/Transforms/GVN/2007-07-25-Loop.ll
@@ -10,6 +10,6 @@ bb278:		; preds = %bb278, %entry
 	br i1 false, label %bb278, label %bb344
 
 bb344:		; preds = %bb278, %entry
-	%tmp38758 = load i16* null, align 2		; <i16> [#uses=0]
+	%tmp38758 = load i16, i16* null, align 2		; <i16> [#uses=0]
 	ret void
 }
diff --git a/test/Transforms/GVN/2007-07-25-NestedLoop.ll b/test/Transforms/GVN/2007-07-25-NestedLoop.ll
index c6d7750d6272..8f2c182faa89 100644
--- a/test/Transforms/GVN/2007-07-25-NestedLoop.ll
+++ b/test/Transforms/GVN/2007-07-25-NestedLoop.ll
@@ -10,27 +10,27 @@ bb534.preheader:		; preds = %entry
 	ret %struct.TypHeader* null
 
 bb556.preheader:		; preds = %entry
-	%tmp56119 = getelementptr %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp56220 = load i32* %tmp56119		; <i32> [#uses=0]
+	%tmp56119 = getelementptr %struct.TypHeader, %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp56220 = load i32, i32* %tmp56119		; <i32> [#uses=0]
 	br i1 false, label %bb.nph23, label %bb675.preheader
 
 bb.nph23:		; preds = %bb556.preheader
 	ret %struct.TypHeader* null
 
 bb656:		; preds = %bb675.outer, %bb656
-	%tmp678 = load i32* %tmp677		; <i32> [#uses=0]
+	%tmp678 = load i32, i32* %tmp677		; <i32> [#uses=0]
 	br i1 false, label %bb684, label %bb656
 
 bb684:		; preds = %bb675.outer, %bb656
 	br i1 false, label %bb924.preheader, label %bb675.outer
 
 bb675.outer:		; preds = %bb675.preheader, %bb684
-	%tmp67812 = load i32* %tmp67711		; <i32> [#uses=0]
+	%tmp67812 = load i32, i32* %tmp67711		; <i32> [#uses=0]
 	br i1 false, label %bb684, label %bb656
 
 bb675.preheader:		; preds = %bb556.preheader
-	%tmp67711 = getelementptr %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp677 = getelementptr %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp67711 = getelementptr %struct.TypHeader, %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp677 = getelementptr %struct.TypHeader, %struct.TypHeader* %hdR, i32 0, i32 0		; <i32*> [#uses=1]
 	br label %bb675.outer
 
 bb924.preheader:		; preds = %bb684
diff --git a/test/Transforms/GVN/2007-07-25-SinglePredecessor.ll b/test/Transforms/GVN/2007-07-25-SinglePredecessor.ll
index ecff657ed389..d7e6c20324f1 100644
--- a/test/Transforms/GVN/2007-07-25-SinglePredecessor.ll
+++ b/test/Transforms/GVN/2007-07-25-SinglePredecessor.ll
@@ -12,7 +12,7 @@
 
 define i32 @_ZNK21mrZEllipticalCylinder10viewingHitERK6ggRay3dddR18mrViewingHitRecordR16ggMaterialRecord(%struct.mrXEllipticalCylinder* %this, %struct.ggBox3* %ray, double %unnamed_arg, double %tmin, double %tmax, %struct.mrViewingHitRecord* %VHR, %struct.ggMaterialRecord* %unnamed_arg2) {
 entry:
-	%tmp80.i = getelementptr %struct.mrViewingHitRecord* %VHR, i32 0, i32 1, i32 0, i32 0		; <double*> [#uses=1]
+	%tmp80.i = getelementptr %struct.mrViewingHitRecord, %struct.mrViewingHitRecord* %VHR, i32 0, i32 1, i32 0, i32 0		; <double*> [#uses=1]
 	store double 0.000000e+00, double* %tmp80.i
 	br i1 false, label %return, label %cond_next.i
 
@@ -20,8 +20,8 @@ cond_next.i:		; preds = %entry
 	br i1 false, label %return, label %cond_true
 
 cond_true:		; preds = %cond_next.i
-	%tmp3.i8 = getelementptr %struct.mrViewingHitRecord* %VHR, i32 0, i32 1, i32 0, i32 0		; <double*> [#uses=1]
-	%tmp46 = load double* %tmp3.i8		; <double> [#uses=0]
+	%tmp3.i8 = getelementptr %struct.mrViewingHitRecord, %struct.mrViewingHitRecord* %VHR, i32 0, i32 1, i32 0, i32 0		; <double*> [#uses=1]
+	%tmp46 = load double, double* %tmp3.i8		; <double> [#uses=0]
 	ret i32 1
 
 return:		; preds = %cond_next.i, %entry
diff --git a/test/Transforms/GVN/2007-07-26-InterlockingLoops.ll b/test/Transforms/GVN/2007-07-26-InterlockingLoops.ll
index 5a15f0e43aaf..98e0024adf5b 100644
--- a/test/Transforms/GVN/2007-07-26-InterlockingLoops.ll
+++ b/test/Transforms/GVN/2007-07-26-InterlockingLoops.ll
@@ -5,7 +5,7 @@
 define i32 @NextRootMove(i32 %wtm, i32 %x, i32 %y, i32 %z) {
 entry:
         %A = alloca i32*
-	%tmp17618 = load i32** getelementptr ([65 x i32*]* @last, i32 0, i32 1), align 4
+	%tmp17618 = load i32*, i32** getelementptr ([65 x i32*], [65 x i32*]* @last, i32 0, i32 1), align 4
         store i32* %tmp17618, i32** %A
 ; CHECK: entry:
 ; CHECK-NEXT: alloca i32
@@ -19,7 +19,7 @@ cond_true116:
 	br i1 %cmp, label %cond_true128, label %cond_true145
 
 cond_true128:
-	%tmp17625 = load i32** getelementptr ([65 x i32*]* @last, i32 0, i32 1), align 4
+	%tmp17625 = load i32*, i32** getelementptr ([65 x i32*], [65 x i32*]* @last, i32 0, i32 1), align 4
         store i32* %tmp17625, i32** %A
    %cmp1 = icmp eq i32 %x, %z
 	br i1 %cmp1 , label %bb98.backedge, label %return.loopexit
@@ -28,7 +28,7 @@ bb98.backedge:
 	br label %cond_true116
 
 cond_true145:
-	%tmp17631 = load i32** getelementptr ([65 x i32*]* @last, i32 0, i32 1), align 4
+	%tmp17631 = load i32*, i32** getelementptr ([65 x i32*], [65 x i32*]* @last, i32 0, i32 1), align 4
         store i32* %tmp17631, i32** %A
 	br i1 false, label %bb98.backedge, label %return.loopexit
 
diff --git a/test/Transforms/GVN/2007-07-26-NonRedundant.ll b/test/Transforms/GVN/2007-07-26-NonRedundant.ll
index 7579e8aff08b..211830a39af3 100644
--- a/test/Transforms/GVN/2007-07-26-NonRedundant.ll
+++ b/test/Transforms/GVN/2007-07-26-NonRedundant.ll
@@ -11,6 +11,6 @@ cond_next:		; preds = %entry
 	br label %bb19
 
 bb19:		; preds = %cond_next, %entry
-	%tmp29 = load i32* @bsLive, align 4		; <i32> [#uses=0]
+	%tmp29 = load i32, i32* @bsLive, align 4		; <i32> [#uses=0]
 	ret i32 0
 }
diff --git a/test/Transforms/GVN/2007-07-26-PhiErasure.ll b/test/Transforms/GVN/2007-07-26-PhiErasure.ll
index d898ab8e2fa2..82af521ab3d6 100644
--- a/test/Transforms/GVN/2007-07-26-PhiErasure.ll
+++ b/test/Transforms/GVN/2007-07-26-PhiErasure.ll
@@ -9,7 +9,7 @@
 
 define i32 @reload(%struct.rtx_def* %first, i32 %global, %struct.FILE* %dumpfile) {
 cond_next2835.1:		; preds = %cond_next2861
-	%tmp2922 = load i32* @n_spills, align 4		; <i32> [#uses=0]
+	%tmp2922 = load i32, i32* @n_spills, align 4		; <i32> [#uses=0]
 	br label %bb2928
 
 bb2928:		; preds = %cond_next2835.1, %cond_next2943
@@ -22,7 +22,7 @@ cond_next2943:		; preds = %cond_true2935, %bb2928
 	br i1 false, label %bb2982.preheader, label %bb2928
 
 bb2982.preheader:		; preds = %cond_next2943
-	%tmp298316 = load i32* @n_spills, align 4		; <i32> [#uses=0]
+	%tmp298316 = load i32, i32* @n_spills, align 4		; <i32> [#uses=0]
 	ret i32 %tmp298316
 
 }
diff --git a/test/Transforms/GVN/2007-07-30-PredIDom.ll b/test/Transforms/GVN/2007-07-30-PredIDom.ll
index 5cb6bb3ecffb..3a7eec7d1424 100644
--- a/test/Transforms/GVN/2007-07-30-PredIDom.ll
+++ b/test/Transforms/GVN/2007-07-30-PredIDom.ll
@@ -269,6 +269,6 @@ bb970:		; preds = %entry
 	br label %return
 
 return:		; preds = %0, %cond_next967, %cond_next922, %cond_next879, %cond_next807, %cond_next630, %cond_next415, %cond_next267, %cond_next191, %bb
-	%retval980 = load i8** null		; <i8*> [#uses=1]
+	%retval980 = load i8*, i8** null		; <i8*> [#uses=1]
 	ret i8* %retval980
 }
diff --git a/test/Transforms/GVN/2007-07-31-NoDomInherit.ll b/test/Transforms/GVN/2007-07-31-NoDomInherit.ll
index 5018a0747aff..b2e4c64ee770 100644
--- a/test/Transforms/GVN/2007-07-31-NoDomInherit.ll
+++ b/test/Transforms/GVN/2007-07-31-NoDomInherit.ll
@@ -149,27 +149,27 @@ entry:
 	br label %bb91
 
 bb:		; preds = %cond_next97
-	%tmp1 = load i32* @numi		; <i32> [#uses=1]
-	%tmp2 = getelementptr [44 x i8]* @.str43, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp3 = call i32 (i8*, ...)* @printf( i8* %tmp2, i32 %tmp1 )		; <i32> [#uses=0]
+	%tmp1 = load i32, i32* @numi		; <i32> [#uses=1]
+	%tmp2 = getelementptr [44 x i8], [44 x i8]* @.str43, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp3 = call i32 (i8*, ...) @printf( i8* %tmp2, i32 %tmp1 )		; <i32> [#uses=0]
 	store i32 0, i32* %i
 	br label %bb13
 
 bb4:		; preds = %bb13
-	%tmp5 = load i32* %i		; <i32> [#uses=1]
-	%tmp6 = load i32* %i		; <i32> [#uses=1]
-	%tmp7 = getelementptr [17 x i32]* @trialx, i32 0, i32 %tmp6		; <i32*> [#uses=1]
-	%tmp8 = load i32* %tmp7		; <i32> [#uses=1]
+	%tmp5 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp6 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp7 = getelementptr [17 x i32], [17 x i32]* @trialx, i32 0, i32 %tmp6		; <i32*> [#uses=1]
+	%tmp8 = load i32, i32* %tmp7		; <i32> [#uses=1]
 	%tmp9 = call i32 @userfun( i32 %tmp8 )		; <i32> [#uses=1]
-	%tmp10 = getelementptr [17 x i32]* @correct_result, i32 0, i32 %tmp5		; <i32*> [#uses=1]
+	%tmp10 = getelementptr [17 x i32], [17 x i32]* @correct_result, i32 0, i32 %tmp5		; <i32*> [#uses=1]
 	store i32 %tmp9, i32* %tmp10
-	%tmp11 = load i32* %i		; <i32> [#uses=1]
+	%tmp11 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp12 = add i32 %tmp11, 1		; <i32> [#uses=1]
 	store i32 %tmp12, i32* %i
 	br label %bb13
 
 bb13:		; preds = %bb4, %bb
-	%tmp14 = load i32* %i		; <i32> [#uses=1]
+	%tmp14 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp15 = icmp sle i32 %tmp14, 16		; <i1> [#uses=1]
 	%tmp1516 = zext i1 %tmp15 to i32		; <i32> [#uses=1]
 	%toBool = icmp ne i32 %tmp1516, 0		; <i1> [#uses=1]
@@ -180,47 +180,47 @@ bb17:		; preds = %bb13
 	br label %bb49
 
 bb18:		; preds = %bb49
-	%tmp19 = load i32* %i		; <i32> [#uses=1]
-	%tmp20 = getelementptr [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp19		; <{ i32, [3 x i32] }*> [#uses=1]
-	%tmp21 = getelementptr { i32, [3 x i32] }* %tmp20, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp19 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp20 = getelementptr [5 x { i32, [3 x i32] }], [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp19		; <{ i32, [3 x i32] }*> [#uses=1]
+	%tmp21 = getelementptr { i32, [3 x i32] }, { i32, [3 x i32] }* %tmp20, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp21
-	%tmp22 = load i32* %i		; <i32> [#uses=1]
-	%tmp23 = getelementptr [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
-	%tmp24 = getelementptr %struct.anon* %tmp23, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
-	%tmp25 = getelementptr [3 x i32]* %tmp24, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp26 = load i32* %tmp25		; <i32> [#uses=1]
-	%tmp27 = getelementptr [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp22		; <{ i32, [3 x i32] }*> [#uses=1]
-	%tmp28 = getelementptr { i32, [3 x i32] }* %tmp27, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
-	%tmp29 = getelementptr [3 x i32]* %tmp28, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp22 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp23 = getelementptr [13 x %struct.anon], [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
+	%tmp24 = getelementptr %struct.anon, %struct.anon* %tmp23, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
+	%tmp25 = getelementptr [3 x i32], [3 x i32]* %tmp24, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp26 = load i32, i32* %tmp25		; <i32> [#uses=1]
+	%tmp27 = getelementptr [5 x { i32, [3 x i32] }], [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp22		; <{ i32, [3 x i32] }*> [#uses=1]
+	%tmp28 = getelementptr { i32, [3 x i32] }, { i32, [3 x i32] }* %tmp27, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
+	%tmp29 = getelementptr [3 x i32], [3 x i32]* %tmp28, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 %tmp26, i32* %tmp29
-	%tmp30 = load i32* %i		; <i32> [#uses=1]
-	%tmp31 = getelementptr [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
-	%tmp32 = getelementptr %struct.anon* %tmp31, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
-	%tmp33 = getelementptr [3 x i32]* %tmp32, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp34 = load i32* %tmp33		; <i32> [#uses=1]
-	%tmp35 = getelementptr [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp30		; <{ i32, [3 x i32] }*> [#uses=1]
-	%tmp36 = getelementptr { i32, [3 x i32] }* %tmp35, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
-	%tmp37 = getelementptr [3 x i32]* %tmp36, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp30 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp31 = getelementptr [13 x %struct.anon], [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
+	%tmp32 = getelementptr %struct.anon, %struct.anon* %tmp31, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
+	%tmp33 = getelementptr [3 x i32], [3 x i32]* %tmp32, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp34 = load i32, i32* %tmp33		; <i32> [#uses=1]
+	%tmp35 = getelementptr [5 x { i32, [3 x i32] }], [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp30		; <{ i32, [3 x i32] }*> [#uses=1]
+	%tmp36 = getelementptr { i32, [3 x i32] }, { i32, [3 x i32] }* %tmp35, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
+	%tmp37 = getelementptr [3 x i32], [3 x i32]* %tmp36, i32 0, i32 1		; <i32*> [#uses=1]
 	store i32 %tmp34, i32* %tmp37
-	%tmp38 = load i32* %i		; <i32> [#uses=1]
-	%tmp39 = getelementptr [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
-	%tmp40 = getelementptr %struct.anon* %tmp39, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
-	%tmp41 = getelementptr [3 x i32]* %tmp40, i32 0, i32 2		; <i32*> [#uses=1]
-	%tmp42 = load i32* %tmp41		; <i32> [#uses=1]
-	%tmp43 = getelementptr [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp38		; <{ i32, [3 x i32] }*> [#uses=1]
-	%tmp44 = getelementptr { i32, [3 x i32] }* %tmp43, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
-	%tmp45 = getelementptr [3 x i32]* %tmp44, i32 0, i32 2		; <i32*> [#uses=1]
+	%tmp38 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp39 = getelementptr [13 x %struct.anon], [13 x %struct.anon]* @isa, i32 0, i32 0		; <%struct.anon*> [#uses=1]
+	%tmp40 = getelementptr %struct.anon, %struct.anon* %tmp39, i32 0, i32 3		; <[3 x i32]*> [#uses=1]
+	%tmp41 = getelementptr [3 x i32], [3 x i32]* %tmp40, i32 0, i32 2		; <i32*> [#uses=1]
+	%tmp42 = load i32, i32* %tmp41		; <i32> [#uses=1]
+	%tmp43 = getelementptr [5 x { i32, [3 x i32] }], [5 x { i32, [3 x i32] }]* @pgm, i32 0, i32 %tmp38		; <{ i32, [3 x i32] }*> [#uses=1]
+	%tmp44 = getelementptr { i32, [3 x i32] }, { i32, [3 x i32] }* %tmp43, i32 0, i32 1		; <[3 x i32]*> [#uses=1]
+	%tmp45 = getelementptr [3 x i32], [3 x i32]* %tmp44, i32 0, i32 2		; <i32*> [#uses=1]
 	store i32 %tmp42, i32* %tmp45
-	%tmp46 = load i32* %i		; <i32> [#uses=1]
+	%tmp46 = load i32, i32* %i		; <i32> [#uses=1]
 	call void @fix_operands( i32 %tmp46 )
-	%tmp47 = load i32* %i		; <i32> [#uses=1]
+	%tmp47 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp48 = add i32 %tmp47, 1		; <i32> [#uses=1]
 	store i32 %tmp48, i32* %i
 	br label %bb49
 
 bb49:		; preds = %bb18, %bb17
-	%tmp50 = load i32* @numi		; <i32> [#uses=1]
-	%tmp51 = load i32* %i		; <i32> [#uses=1]
+	%tmp50 = load i32, i32* @numi		; <i32> [#uses=1]
+	%tmp51 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp52 = icmp slt i32 %tmp51, %tmp50		; <i1> [#uses=1]
 	%tmp5253 = zext i1 %tmp52 to i32		; <i32> [#uses=1]
 	%toBool54 = icmp ne i32 %tmp5253, 0		; <i1> [#uses=1]
@@ -229,10 +229,10 @@ bb49:		; preds = %bb18, %bb17
 bb55:		; preds = %bb49
 	%tmp56 = call i32 @search( )		; <i32> [#uses=1]
 	store i32 %tmp56, i32* %num_sol
-	%tmp57 = getelementptr [21 x i8]* @.str44, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp58 = load i32* %num_sol		; <i32> [#uses=1]
-	%tmp59 = call i32 (i8*, ...)* @printf( i8* %tmp57, i32 %tmp58 )		; <i32> [#uses=0]
-	%tmp60 = load i32* @counters		; <i32> [#uses=1]
+	%tmp57 = getelementptr [21 x i8], [21 x i8]* @.str44, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp58 = load i32, i32* %num_sol		; <i32> [#uses=1]
+	%tmp59 = call i32 (i8*, ...) @printf( i8* %tmp57, i32 %tmp58 )		; <i32> [#uses=0]
+	%tmp60 = load i32, i32* @counters		; <i32> [#uses=1]
 	%tmp61 = icmp ne i32 %tmp60, 0		; <i1> [#uses=1]
 	%tmp6162 = zext i1 %tmp61 to i32		; <i32> [#uses=1]
 	%toBool63 = icmp ne i32 %tmp6162, 0		; <i1> [#uses=1]
@@ -240,50 +240,50 @@ bb55:		; preds = %bb49
 
 cond_true:		; preds = %bb55
 	store i32 0, i32* %total
-	%tmp64 = getelementptr [12 x i8]* @.str45, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp65 = call i32 (i8*, ...)* @printf( i8* %tmp64 )		; <i32> [#uses=0]
+	%tmp64 = getelementptr [12 x i8], [12 x i8]* @.str45, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp65 = call i32 (i8*, ...) @printf( i8* %tmp64 )		; <i32> [#uses=0]
 	store i32 0, i32* %i
 	br label %bb79
 
 bb66:		; preds = %bb79
-	%tmp67 = load i32* %i		; <i32> [#uses=1]
-	%tmp68 = getelementptr [5 x i32]* @counter, i32 0, i32 %tmp67		; <i32*> [#uses=1]
-	%tmp69 = load i32* %tmp68		; <i32> [#uses=1]
-	%tmp70 = getelementptr [5 x i8]* @.str46, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp71 = call i32 (i8*, ...)* @printf( i8* %tmp70, i32 %tmp69 )		; <i32> [#uses=0]
-	%tmp72 = load i32* %i		; <i32> [#uses=1]
-	%tmp73 = getelementptr [5 x i32]* @counter, i32 0, i32 %tmp72		; <i32*> [#uses=1]
-	%tmp74 = load i32* %tmp73		; <i32> [#uses=1]
-	%tmp75 = load i32* %total		; <i32> [#uses=1]
+	%tmp67 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp68 = getelementptr [5 x i32], [5 x i32]* @counter, i32 0, i32 %tmp67		; <i32*> [#uses=1]
+	%tmp69 = load i32, i32* %tmp68		; <i32> [#uses=1]
+	%tmp70 = getelementptr [5 x i8], [5 x i8]* @.str46, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp71 = call i32 (i8*, ...) @printf( i8* %tmp70, i32 %tmp69 )		; <i32> [#uses=0]
+	%tmp72 = load i32, i32* %i		; <i32> [#uses=1]
+	%tmp73 = getelementptr [5 x i32], [5 x i32]* @counter, i32 0, i32 %tmp72		; <i32*> [#uses=1]
+	%tmp74 = load i32, i32* %tmp73		; <i32> [#uses=1]
+	%tmp75 = load i32, i32* %total		; <i32> [#uses=1]
 	%tmp76 = add i32 %tmp74, %tmp75		; <i32> [#uses=1]
 	store i32 %tmp76, i32* %total
-	%tmp77 = load i32* %i		; <i32> [#uses=1]
+	%tmp77 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp78 = add i32 %tmp77, 1		; <i32> [#uses=1]
 	store i32 %tmp78, i32* %i
 	br label %bb79
 
 bb79:		; preds = %bb66, %cond_true
-	%tmp80 = load i32* @numi		; <i32> [#uses=1]
-	%tmp81 = load i32* %i		; <i32> [#uses=1]
+	%tmp80 = load i32, i32* @numi		; <i32> [#uses=1]
+	%tmp81 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp82 = icmp slt i32 %tmp81, %tmp80		; <i1> [#uses=1]
 	%tmp8283 = zext i1 %tmp82 to i32		; <i32> [#uses=1]
 	%toBool84 = icmp ne i32 %tmp8283, 0		; <i1> [#uses=1]
 	br i1 %toBool84, label %bb66, label %bb85
 
 bb85:		; preds = %bb79
-	%tmp86 = getelementptr [12 x i8]* @.str47, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp87 = load i32* %total		; <i32> [#uses=1]
-	%tmp88 = call i32 (i8*, ...)* @printf( i8* %tmp86, i32 %tmp87 )		; <i32> [#uses=0]
+	%tmp86 = getelementptr [12 x i8], [12 x i8]* @.str47, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp87 = load i32, i32* %total		; <i32> [#uses=1]
+	%tmp88 = call i32 (i8*, ...) @printf( i8* %tmp86, i32 %tmp87 )		; <i32> [#uses=0]
 	br label %cond_next
 
 cond_next:		; preds = %bb85, %bb55
-	%tmp89 = load i32* @numi		; <i32> [#uses=1]
+	%tmp89 = load i32, i32* @numi		; <i32> [#uses=1]
 	%tmp90 = add i32 %tmp89, 1		; <i32> [#uses=1]
 	store i32 %tmp90, i32* @numi
 	br label %bb91
 
 bb91:		; preds = %cond_next, %entry
-	%tmp92 = load i32* @numi		; <i32> [#uses=1]
+	%tmp92 = load i32, i32* @numi		; <i32> [#uses=1]
 	%tmp93 = icmp sgt i32 %tmp92, 5		; <i1> [#uses=1]
 	%tmp9394 = zext i1 %tmp93 to i32		; <i32> [#uses=1]
 	%toBool95 = icmp ne i32 %tmp9394, 0		; <i1> [#uses=1]
@@ -293,7 +293,7 @@ cond_true96:		; preds = %bb91
 	br label %bb102
 
 cond_next97:		; preds = %bb91
-	%tmp98 = load i32* %num_sol		; <i32> [#uses=1]
+	%tmp98 = load i32, i32* %num_sol		; <i32> [#uses=1]
 	%tmp99 = icmp eq i32 %tmp98, 0		; <i1> [#uses=1]
 	%tmp99100 = zext i1 %tmp99 to i32		; <i32> [#uses=1]
 	%toBool101 = icmp ne i32 %tmp99100, 0		; <i1> [#uses=1]
@@ -301,12 +301,12 @@ cond_next97:		; preds = %bb91
 
 bb102:		; preds = %cond_next97, %cond_true96
 	store i32 0, i32* %tmp
-	%tmp103 = load i32* %tmp		; <i32> [#uses=1]
+	%tmp103 = load i32, i32* %tmp		; <i32> [#uses=1]
 	store i32 %tmp103, i32* %retval
 	br label %return
 
 return:		; preds = %bb102
-	%retval104 = load i32* %retval		; <i32> [#uses=1]
+	%retval104 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval104
 }
 
diff --git a/test/Transforms/GVN/2007-07-31-RedundantPhi.ll b/test/Transforms/GVN/2007-07-31-RedundantPhi.ll
index 13419d19e1d5..b285560e6a99 100644
--- a/test/Transforms/GVN/2007-07-31-RedundantPhi.ll
+++ b/test/Transforms/GVN/2007-07-31-RedundantPhi.ll
@@ -17,6 +17,6 @@ cond_false470:		; preds = %cond_next449
 	br label %cond_next698
 
 cond_next698:		; preds = %cond_true492
-	%tmp701 = load i16* @img_width, align 2		; <i16> [#uses=0]
+	%tmp701 = load i16, i16* @img_width, align 2		; <i16> [#uses=0]
 	ret i32 0
 }
diff --git a/test/Transforms/GVN/2008-02-12-UndefLoad.ll b/test/Transforms/GVN/2008-02-12-UndefLoad.ll
index de2aa614ff5d..a1aed867ec63 100644
--- a/test/Transforms/GVN/2008-02-12-UndefLoad.ll
+++ b/test/Transforms/GVN/2008-02-12-UndefLoad.ll
@@ -6,13 +6,13 @@
 define i32 @a() {
 entry:
         %c = alloca %struct.anon                ; <%struct.anon*> [#uses=2]
-        %tmp = getelementptr %struct.anon* %c, i32 0, i32 0             ; <i32*> [#uses=1]
-        %tmp1 = getelementptr i32* %tmp, i32 1          ; <i32*> [#uses=2]
-        %tmp2 = load i32* %tmp1, align 4                ; <i32> [#uses=1]
+        %tmp = getelementptr %struct.anon, %struct.anon* %c, i32 0, i32 0             ; <i32*> [#uses=1]
+        %tmp1 = getelementptr i32, i32* %tmp, i32 1          ; <i32*> [#uses=2]
+        %tmp2 = load i32, i32* %tmp1, align 4                ; <i32> [#uses=1]
         %tmp3 = or i32 %tmp2, 11                ; <i32> [#uses=1]
         %tmp4 = and i32 %tmp3, -21              ; <i32> [#uses=1]
         store i32 %tmp4, i32* %tmp1, align 4
-        %call = call i32 (...)* @x( %struct.anon* %c )          ; <i32> [#uses=0]
+        %call = call i32 (...) @x( %struct.anon* %c )          ; <i32> [#uses=0]
         ret i32 undef
 }
 
diff --git a/test/Transforms/GVN/2008-02-13-NewPHI.ll b/test/Transforms/GVN/2008-02-13-NewPHI.ll
index 80b519d14e9a..638939b15343 100644
--- a/test/Transforms/GVN/2008-02-13-NewPHI.ll
+++ b/test/Transforms/GVN/2008-02-13-NewPHI.ll
@@ -8,7 +8,7 @@ entry:
 	br label %bb33
 
 bb:		; preds = %bb33
-	%tmp27 = load float** %sx_addr, align 4		; <float*> [#uses=1]
+	%tmp27 = load float*, float** %sx_addr, align 4		; <float*> [#uses=1]
 	store float 0.000000e+00, float* %tmp27, align 4
 	store float* null, float** %sx_addr, align 4
 	br label %bb33
@@ -17,6 +17,6 @@ bb33:		; preds = %bb, %entry
 	br i1 false, label %bb, label %return
 
 return:		; preds = %bb33
-	%retval59 = load i32* null, align 4		; <i32> [#uses=1]
+	%retval59 = load i32, i32* null, align 4		; <i32> [#uses=1]
 	ret i32 %retval59
 }
diff --git a/test/Transforms/GVN/2008-07-02-Unreachable.ll b/test/Transforms/GVN/2008-07-02-Unreachable.ll
index ce83fa4e4be9..d9932644da3e 100644
--- a/test/Transforms/GVN/2008-07-02-Unreachable.ll
+++ b/test/Transforms/GVN/2008-07-02-Unreachable.ll
@@ -13,7 +13,7 @@ ifthen:		; preds = %entry
 	br label %ifend
 
 ifelse:		; preds = %entry
-	%tmp3 = load i8* @g_3		; <i8> [#uses=0]
+	%tmp3 = load i8, i8* @g_3		; <i8> [#uses=0]
         store i8 %tmp3, i8* %A
 	br label %afterfor
 
@@ -27,7 +27,7 @@ forinc:		; preds = %forbody
 	br label %forcond
 
 afterfor:		; preds = %forcond, %forcond.thread
-	%tmp10 = load i8* @g_3		; <i8> [#uses=0]
+	%tmp10 = load i8, i8* @g_3		; <i8> [#uses=0]
 	ret i8 %tmp10
 
 ifend:		; preds = %afterfor, %ifthen
diff --git a/test/Transforms/GVN/2008-12-09-SelfRemove.ll b/test/Transforms/GVN/2008-12-09-SelfRemove.ll
index c6833e373d06..d8ab1bae48e9 100644
--- a/test/Transforms/GVN/2008-12-09-SelfRemove.ll
+++ b/test/Transforms/GVN/2008-12-09-SelfRemove.ll
@@ -10,12 +10,12 @@ target triple = "i386-apple-darwin9.5"
 
 define void @d_print_mod_list(%struct.d_print_info* %dpi, %struct.d_print_mod* %mods, i32 %suffix) nounwind {
 entry:
-	%0 = getelementptr %struct.d_print_info* %dpi, i32 0, i32 1		; <i8**> [#uses=1]
+	%0 = getelementptr %struct.d_print_info, %struct.d_print_info* %dpi, i32 0, i32 1		; <i8**> [#uses=1]
 	br i1 false, label %return, label %bb
 
 bb:		; preds = %entry
-	%1 = load i8** %0, align 4		; <i8*> [#uses=0]
-	%2 = getelementptr %struct.d_print_info* %dpi, i32 0, i32 1		; <i8**> [#uses=0]
+	%1 = load i8*, i8** %0, align 4		; <i8*> [#uses=0]
+	%2 = getelementptr %struct.d_print_info, %struct.d_print_info* %dpi, i32 0, i32 1		; <i8**> [#uses=0]
 	br label %bb21
 
 bb21:		; preds = %bb21, %bb
diff --git a/test/Transforms/GVN/2008-12-12-RLE-Crash.ll b/test/Transforms/GVN/2008-12-12-RLE-Crash.ll
index da67ee776696..dabf7fa7aeb2 100644
--- a/test/Transforms/GVN/2008-12-12-RLE-Crash.ll
+++ b/test/Transforms/GVN/2008-12-12-RLE-Crash.ll
@@ -7,8 +7,8 @@ entry:
 	br label %bb84
 
 bb41:		; preds = %bb82
-	%tmp = load i8* %opt.0, align 1		; <i8> [#uses=0]
-	%tmp1 = getelementptr i8* %opt.0, i32 1		; <i8*> [#uses=2]
+	%tmp = load i8, i8* %opt.0, align 1		; <i8> [#uses=0]
+	%tmp1 = getelementptr i8, i8* %opt.0, i32 1		; <i8*> [#uses=2]
 	switch i32 0, label %bb81 [
 		i32 102, label %bb82
 		i32 110, label %bb79
@@ -26,10 +26,10 @@ bb81:		; preds = %bb41
 
 bb82:		; preds = %bb84, %bb79, %bb41
 	%opt.0 = phi i8* [ %tmp3, %bb84 ], [ %tmp1, %bb79 ], [ %tmp1, %bb41 ]		; <i8*> [#uses=3]
-	%tmp2 = load i8* %opt.0, align 1		; <i8> [#uses=0]
+	%tmp2 = load i8, i8* %opt.0, align 1		; <i8> [#uses=0]
 	br i1 false, label %bb84, label %bb41
 
 bb84:		; preds = %bb82, %entry
-	%tmp3 = getelementptr i8* null, i32 1		; <i8*> [#uses=1]
+	%tmp3 = getelementptr i8, i8* null, i32 1		; <i8*> [#uses=1]
 	br label %bb82
 }
diff --git a/test/Transforms/GVN/2008-12-14-rle-reanalyze.ll b/test/Transforms/GVN/2008-12-14-rle-reanalyze.ll
index 41f76c8167e5..207a2519f433 100644
--- a/test/Transforms/GVN/2008-12-14-rle-reanalyze.ll
+++ b/test/Transforms/GVN/2008-12-14-rle-reanalyze.ll
@@ -11,8 +11,8 @@ bb22:		; preds = %bb23, %bb22, %entry
 	br i1 false, label %bb23, label %bb22
 
 bb23:		; preds = %bb23, %bb22
-	%sortv.233 = phi i32* [ getelementptr ([256 x i32]* @sort_value, i32 0, i32 0), %bb22 ], [ %sortv.2, %bb23 ]		; <i32*> [#uses=1]
-	%0 = load i32* %sortv.233, align 4		; <i32> [#uses=0]
-	%sortv.2 = getelementptr [256 x i32]* @sort_value, i32 0, i32 0		; <i32*> [#uses=1]
+	%sortv.233 = phi i32* [ getelementptr ([256 x i32], [256 x i32]* @sort_value, i32 0, i32 0), %bb22 ], [ %sortv.2, %bb23 ]		; <i32*> [#uses=1]
+	%0 = load i32, i32* %sortv.233, align 4		; <i32> [#uses=0]
+	%sortv.2 = getelementptr [256 x i32], [256 x i32]* @sort_value, i32 0, i32 0		; <i32*> [#uses=1]
 	br i1 false, label %bb23, label %bb22
 }
diff --git a/test/Transforms/GVN/2008-12-15-CacheVisited.ll b/test/Transforms/GVN/2008-12-15-CacheVisited.ll
index 0a63f3f4626c..73adacd3cfaf 100644
--- a/test/Transforms/GVN/2008-12-15-CacheVisited.ll
+++ b/test/Transforms/GVN/2008-12-15-CacheVisited.ll
@@ -10,12 +10,12 @@ bb202:		; preds = %entry
 	unreachable
 
 bb203:		; preds = %entry
-	%tmp = getelementptr i32* %decl, i32 1		; <i32*> [#uses=1]
-	%tmp1 = load i32* %tmp, align 4		; <i32> [#uses=0]
+	%tmp = getelementptr i32, i32* %decl, i32 1		; <i32*> [#uses=1]
+	%tmp1 = load i32, i32* %tmp, align 4		; <i32> [#uses=0]
 	br i1 false, label %bb207, label %bb204
 
 bb204:		; preds = %bb203
-	%tmp2 = getelementptr i32* %decl, i32 1		; <i32*> [#uses=1]
+	%tmp2 = getelementptr i32, i32* %decl, i32 1		; <i32*> [#uses=1]
 	br label %bb208
 
 bb207:		; preds = %bb203
@@ -23,6 +23,6 @@ bb207:		; preds = %bb203
 
 bb208:		; preds = %bb207, %bb204
 	%iftmp.1374.0.in = phi i32* [ null, %bb207 ], [ %tmp2, %bb204 ]		; <i32*> [#uses=1]
-	%iftmp.1374.0 = load i32* %iftmp.1374.0.in		; <i32> [#uses=0]
+	%iftmp.1374.0 = load i32, i32* %iftmp.1374.0.in		; <i32> [#uses=0]
 	unreachable
 }
diff --git a/test/Transforms/GVN/2009-01-21-SortInvalidation.ll b/test/Transforms/GVN/2009-01-21-SortInvalidation.ll
index 36775936c8b7..6144697bb5b6 100644
--- a/test/Transforms/GVN/2009-01-21-SortInvalidation.ll
+++ b/test/Transforms/GVN/2009-01-21-SortInvalidation.ll
@@ -37,7 +37,7 @@ bb550:		; preds = %bb543
 	br i1 false, label %bb554, label %bb552
 
 bb552:		; preds = %bb550
-	%0 = load i8* %d.0, align 8		; <i8> [#uses=0]
+	%0 = load i8, i8* %d.0, align 8		; <i8> [#uses=0]
 	br label %bb554
 
 bb554:		; preds = %bb552, %bb550, %bb549
diff --git a/test/Transforms/GVN/2009-01-22-SortInvalidation.ll b/test/Transforms/GVN/2009-01-22-SortInvalidation.ll
index 95690a5a2571..89b058a235b4 100644
--- a/test/Transforms/GVN/2009-01-22-SortInvalidation.ll
+++ b/test/Transforms/GVN/2009-01-22-SortInvalidation.ll
@@ -79,20 +79,20 @@ bb54:		; preds = %entry
 	br label %bb69.loopexit
 
 bb59:		; preds = %bb63.preheader
-	%0 = load %struct..4sPragmaType** %3, align 4		; <%struct..4sPragmaType*> [#uses=0]
+	%0 = load %struct..4sPragmaType*, %struct..4sPragmaType** %3, align 4		; <%struct..4sPragmaType*> [#uses=0]
 	br label %bb65
 
 bb65:		; preds = %bb63.preheader, %bb59
-	%1 = load %struct..4sPragmaType** %4, align 4		; <%struct..4sPragmaType*> [#uses=0]
+	%1 = load %struct..4sPragmaType*, %struct..4sPragmaType** %4, align 4		; <%struct..4sPragmaType*> [#uses=0]
 	br i1 false, label %bb67, label %bb63.preheader
 
 bb67:		; preds = %bb65
-	%2 = getelementptr %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=0]
+	%2 = getelementptr %struct.IdList, %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=0]
 	unreachable
 
 bb69.loopexit:		; preds = %bb54, %entry
-	%3 = getelementptr %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=1]
-	%4 = getelementptr %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=1]
+	%3 = getelementptr %struct.IdList, %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=1]
+	%4 = getelementptr %struct.IdList, %struct.IdList* %pColumn, i32 0, i32 0		; <%struct..4sPragmaType**> [#uses=1]
 	br label %bb63.preheader
 
 bb63.preheader:		; preds = %bb69.loopexit, %bb65
diff --git a/test/Transforms/GVN/2009-02-17-LoadPRECrash.ll b/test/Transforms/GVN/2009-02-17-LoadPRECrash.ll
index c2d57a105fc9..808f28c674c5 100644
--- a/test/Transforms/GVN/2009-02-17-LoadPRECrash.ll
+++ b/test/Transforms/GVN/2009-02-17-LoadPRECrash.ll
@@ -24,11 +24,11 @@ entry:
 	%addr = alloca %struct.rtx_def*		; <%struct.rtx_def**> [#uses=5]
 	%iftmp.1532 = alloca %struct.rtx_def*		; <%struct.rtx_def**> [#uses=3]
 	store %struct.rtx_def* %orig, %struct.rtx_def** null
-	%0 = load %struct.rtx_def** null, align 4		; <%struct.rtx_def*> [#uses=0]
+	%0 = load %struct.rtx_def*, %struct.rtx_def** null, align 4		; <%struct.rtx_def*> [#uses=0]
 	br i1 false, label %bb96, label %bb59
 
 bb59:		; preds = %entry
-	%1 = load %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
+	%1 = load %struct.rtx_def*, %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
 	%2 = call i32 @local_symbolic_operand(%struct.rtx_def* %1, i32 0) nounwind		; <i32> [#uses=0]
 	br i1 false, label %bb96, label %bb63
 
@@ -54,7 +54,7 @@ bb76:		; preds = %bb75, %bb74
 	br i1 false, label %bb77, label %bb84
 
 bb77:		; preds = %bb76
-	%3 = getelementptr [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 0		; <%struct.cgraph_rtl_info*> [#uses=0]
+	%3 = getelementptr [1 x %struct.cgraph_rtl_info], [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 0		; <%struct.cgraph_rtl_info*> [#uses=0]
 	unreachable
 
 bb84:		; preds = %bb76
@@ -89,22 +89,22 @@ bb94:		; preds = %bb93, %bb92
 	unreachable
 
 bb96:		; preds = %bb59, %entry
-	%5 = load %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
-	%6 = getelementptr %struct.rtx_def* %5, i32 0, i32 0		; <i16*> [#uses=1]
-	%7 = load i16* %6, align 2		; <i16> [#uses=0]
+	%5 = load %struct.rtx_def*, %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
+	%6 = getelementptr %struct.rtx_def, %struct.rtx_def* %5, i32 0, i32 0		; <i16*> [#uses=1]
+	%7 = load i16, i16* %6, align 2		; <i16> [#uses=0]
 	br i1 false, label %bb147, label %bb97
 
 bb97:		; preds = %bb96
-	%8 = load %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=0]
+	%8 = load %struct.rtx_def*, %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=0]
 	br i1 false, label %bb147, label %bb99
 
 bb99:		; preds = %bb97
 	unreachable
 
 bb147:		; preds = %bb97, %bb96
-	%9 = load %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
-	%10 = getelementptr %struct.rtx_def* %9, i32 0, i32 0		; <i16*> [#uses=1]
-	%11 = load i16* %10, align 2		; <i16> [#uses=0]
+	%9 = load %struct.rtx_def*, %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=1]
+	%10 = getelementptr %struct.rtx_def, %struct.rtx_def* %9, i32 0, i32 0		; <i16*> [#uses=1]
+	%11 = load i16, i16* %10, align 2		; <i16> [#uses=0]
 	br i1 false, label %bb164, label %bb148
 
 bb148:		; preds = %bb147
@@ -120,7 +120,7 @@ bb152:		; preds = %bb149
 	br label %bb164
 
 bb164:		; preds = %bb152, %bb148, %bb147
-	%12 = getelementptr [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 1		; <%struct.cgraph_rtl_info*> [#uses=0]
+	%12 = getelementptr [1 x %struct.cgraph_rtl_info], [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 1		; <%struct.cgraph_rtl_info*> [#uses=0]
 	br i1 false, label %bb165, label %bb166
 
 bb165:		; preds = %bb164
@@ -163,12 +163,12 @@ bb180:		; preds = %bb179, %bb178
 	br label %bb181
 
 bb181:		; preds = %bb180, %bb170
-	%13 = call %struct.rtvec_def* (i32, ...)* @gen_rtvec(i32 1, %struct.rtx_def* null) nounwind		; <%struct.rtvec_def*> [#uses=0]
+	%13 = call %struct.rtvec_def* (i32, ...) @gen_rtvec(i32 1, %struct.rtx_def* null) nounwind		; <%struct.rtvec_def*> [#uses=0]
 	unreachable
 
 bb211:		; preds = %bb168, %bb167
-	%14 = load %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=0]
-	%15 = getelementptr [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 0		; <%struct.cgraph_rtl_info*> [#uses=0]
+	%14 = load %struct.rtx_def*, %struct.rtx_def** %addr, align 4		; <%struct.rtx_def*> [#uses=0]
+	%15 = getelementptr [1 x %struct.cgraph_rtl_info], [1 x %struct.cgraph_rtl_info]* null, i32 0, i32 0		; <%struct.cgraph_rtl_info*> [#uses=0]
 	store %struct.rtx_def* null, %struct.rtx_def** null, align 4
 	br i1 false, label %bb212, label %bb213
 
@@ -182,12 +182,12 @@ bb213:		; preds = %bb211
 
 bb214:		; preds = %bb213, %bb212
 	%16 = bitcast %struct.block_symbol* null to [1 x %struct.cgraph_rtl_info]*		; <[1 x %struct.cgraph_rtl_info]*> [#uses=1]
-	%17 = getelementptr [1 x %struct.cgraph_rtl_info]* %16, i32 0, i32 1		; <%struct.cgraph_rtl_info*> [#uses=0]
-	%18 = load %struct.rtx_def** %iftmp.1532, align 4		; <%struct.rtx_def*> [#uses=0]
-	%19 = getelementptr %struct.rtx_def* null, i32 0, i32 3		; <%struct.u*> [#uses=1]
-	%20 = getelementptr %struct.u* %19, i32 0, i32 0		; <%struct.block_symbol*> [#uses=1]
+	%17 = getelementptr [1 x %struct.cgraph_rtl_info], [1 x %struct.cgraph_rtl_info]* %16, i32 0, i32 1		; <%struct.cgraph_rtl_info*> [#uses=0]
+	%18 = load %struct.rtx_def*, %struct.rtx_def** %iftmp.1532, align 4		; <%struct.rtx_def*> [#uses=0]
+	%19 = getelementptr %struct.rtx_def, %struct.rtx_def* null, i32 0, i32 3		; <%struct.u*> [#uses=1]
+	%20 = getelementptr %struct.u, %struct.u* %19, i32 0, i32 0		; <%struct.block_symbol*> [#uses=1]
 	%21 = bitcast %struct.block_symbol* %20 to [1 x i64]*		; <[1 x i64]*> [#uses=1]
-	%22 = getelementptr [1 x i64]* %21, i32 0, i32 0		; <i64*> [#uses=0]
+	%22 = getelementptr [1 x i64], [1 x i64]* %21, i32 0, i32 0		; <i64*> [#uses=0]
 	%23 = call %struct.rtx_def* @plus_constant(%struct.rtx_def* null, i64 0) nounwind		; <%struct.rtx_def*> [#uses=0]
 	unreachable
 }
diff --git a/test/Transforms/GVN/2009-06-17-InvalidPRE.ll b/test/Transforms/GVN/2009-06-17-InvalidPRE.ll
index 6ac6072a9273..bf0a234da4ac 100644
--- a/test/Transforms/GVN/2009-06-17-InvalidPRE.ll
+++ b/test/Transforms/GVN/2009-06-17-InvalidPRE.ll
@@ -18,20 +18,20 @@ bb4.preheader:    ; preds = %entry
 
 bb2:    ; preds = %bb1
   %2 = sub i32 %len.0, %13   ; <i32> [#uses=1]
-  %3 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 2    ; <i32*> [#uses=1]
+  %3 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 2    ; <i32*> [#uses=1]
   store i32 0, i32* %3, align 4
-  %4 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 0    ; <%struct.mbuf**> [#uses=1]
-  %5 = load %struct.mbuf** %4, align 4    ; <%struct.mbuf*> [#uses=1]
+  %4 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 0    ; <%struct.mbuf**> [#uses=1]
+  %5 = load %struct.mbuf*, %struct.mbuf** %4, align 4    ; <%struct.mbuf*> [#uses=1]
   br label %bb4.outer
 
 bb4.outer:    ; preds = %bb4.preheader, %bb2
   %m.0.ph = phi %struct.mbuf* [ %5, %bb2 ], [ %mp, %bb4.preheader ]   ; <%struct.mbuf*> [#uses=7]
   %len.0.ph = phi i32 [ %2, %bb2 ], [ %req_len, %bb4.preheader ]    ; <i32> [#uses=1]
   %6 = icmp ne %struct.mbuf* %m.0.ph, null    ; <i1> [#uses=1]
-  %7 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 2    ; <i32*> [#uses=1]
-  %8 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 2   ; <i32*> [#uses=1]
-  %9 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 3   ; <i8**> [#uses=1]
-  %10 = getelementptr %struct.mbuf* %m.0.ph, i32 0, i32 3   ; <i8**> [#uses=1]
+  %7 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 2    ; <i32*> [#uses=1]
+  %8 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 2   ; <i32*> [#uses=1]
+  %9 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 3   ; <i8**> [#uses=1]
+  %10 = getelementptr %struct.mbuf, %struct.mbuf* %m.0.ph, i32 0, i32 3   ; <i8**> [#uses=1]
   br label %bb4
 
 bb4:    ; preds = %bb4.outer, %bb3
@@ -41,21 +41,21 @@ bb4:    ; preds = %bb4.outer, %bb3
   br i1 %12, label %bb1, label %bb7
 
 bb1:    ; preds = %bb4
-  %13 = load i32* %7, align 4    ; <i32> [#uses=3]
+  %13 = load i32, i32* %7, align 4    ; <i32> [#uses=3]
   %14 = icmp sgt i32 %13, %len.0    ; <i1> [#uses=1]
   br i1 %14, label %bb3, label %bb2
 
 bb3:    ; preds = %bb1
   %15 = sub i32 %13, %len.0    ; <i32> [#uses=1]
   store i32 %15, i32* %8, align 4
-  %16 = load i8** %9, align 4    ; <i8*> [#uses=1]
-  %17 = getelementptr i8* %16, i32 %len.0   ; <i8*> [#uses=1]
+  %16 = load i8*, i8** %9, align 4    ; <i8*> [#uses=1]
+  %17 = getelementptr i8, i8* %16, i32 %len.0   ; <i8*> [#uses=1]
   store i8* %17, i8** %10, align 4
   br label %bb4
 
 bb7:    ; preds = %bb4
-  %18 = getelementptr %struct.mbuf* %mp, i32 0, i32 5   ; <i16*> [#uses=1]
-  %19 = load i16* %18, align 2    ; <i16> [#uses=1]
+  %18 = getelementptr %struct.mbuf, %struct.mbuf* %mp, i32 0, i32 5   ; <i16*> [#uses=1]
+  %19 = load i16, i16* %18, align 2    ; <i16> [#uses=1]
   %20 = zext i16 %19 to i32   ; <i32> [#uses=1]
   %21 = and i32 %20, 2    ; <i32> [#uses=1]
   %22 = icmp eq i32 %21, 0    ; <i1> [#uses=1]
@@ -63,7 +63,7 @@ bb7:    ; preds = %bb4
 
 bb8:    ; preds = %bb7
   %23 = sub i32 %req_len, %len.0    ; <i32> [#uses=1]
-  %24 = getelementptr %struct.mbuf* %mp, i32 0, i32 6   ; <i32*> [#uses=1]
+  %24 = getelementptr %struct.mbuf, %struct.mbuf* %mp, i32 0, i32 6   ; <i32*> [#uses=1]
   store i32 %23, i32* %24, align 4
   ret void
 
diff --git a/test/Transforms/GVN/2009-07-13-MemDepSortFail.ll b/test/Transforms/GVN/2009-07-13-MemDepSortFail.ll
index f079108b9bda..0ed52370f80d 100644
--- a/test/Transforms/GVN/2009-07-13-MemDepSortFail.ll
+++ b/test/Transforms/GVN/2009-07-13-MemDepSortFail.ll
@@ -22,11 +22,11 @@ bb2:		; preds = %bb
 	br label %bb62
 
 bb9:		; preds = %bb
-	%0 = load i8* %sp.1, align 1		; <i8> [#uses=0]
+	%0 = load i8, i8* %sp.1, align 1		; <i8> [#uses=0]
 	br label %bb62
 
 bb51:		; preds = %bb
-	%1 = load i8* %sp.1, align 1		; <i8> [#uses=0]
+	%1 = load i8, i8* %sp.1, align 1		; <i8> [#uses=0]
 	ret i8* null
 
 bb62:		; preds = %bb9, %bb2, %bb
diff --git a/test/Transforms/GVN/2009-11-12-MemDepMallocBitCast.ll b/test/Transforms/GVN/2009-11-12-MemDepMallocBitCast.ll
index b433297bba6f..a12fbddbda5a 100644
--- a/test/Transforms/GVN/2009-11-12-MemDepMallocBitCast.ll
+++ b/test/Transforms/GVN/2009-11-12-MemDepMallocBitCast.ll
@@ -3,12 +3,12 @@
 ; RUN: opt < %s -gvn -S | FileCheck %s
 
 define i64 @test() {
-  %1 = tail call i8* @malloc(i64 mul (i64 4, i64 ptrtoint (i64* getelementptr (i64* null, i64 1) to i64))) ; <i8*> [#uses=2]
+  %1 = tail call i8* @malloc(i64 mul (i64 4, i64 ptrtoint (i64* getelementptr (i64, i64* null, i64 1) to i64))) ; <i8*> [#uses=2]
   store i8 42, i8* %1
   %X = bitcast i8* %1 to i64*                     ; <i64*> [#uses=1]
-  %Y = load i64* %X                               ; <i64> [#uses=1]
+  %Y = load i64, i64* %X                               ; <i64> [#uses=1]
   ret i64 %Y
-; CHECK: %Y = load i64* %X
+; CHECK: %Y = load i64, i64* %X
 ; CHECK: ret i64 %Y
 }
 
diff --git a/test/Transforms/GVN/2010-03-31-RedundantPHIs.ll b/test/Transforms/GVN/2010-03-31-RedundantPHIs.ll
index d6e1c6b76d0b..9d9ad5424075 100644
--- a/test/Transforms/GVN/2010-03-31-RedundantPHIs.ll
+++ b/test/Transforms/GVN/2010-03-31-RedundantPHIs.ll
@@ -24,8 +24,8 @@ bb8:                                              ; preds = %bb12
   br i1 undef, label %bb9, label %bb10
 
 bb9:                                              ; preds = %bb8
-  %0 = load i8** undef, align 4                   ; <i8*> [#uses=0]
-  %1 = load i8** undef, align 4                   ; <i8*> [#uses=0]
+  %0 = load i8*, i8** undef, align 4                   ; <i8*> [#uses=0]
+  %1 = load i8*, i8** undef, align 4                   ; <i8*> [#uses=0]
   br label %bb11
 
 bb10:                                             ; preds = %bb8
diff --git a/test/Transforms/GVN/2010-05-08-OneBit.ll b/test/Transforms/GVN/2010-05-08-OneBit.ll
index 480ce8ba0ae1..0e3fa4b8a490 100644
--- a/test/Transforms/GVN/2010-05-08-OneBit.ll
+++ b/test/Transforms/GVN/2010-05-08-OneBit.ll
@@ -6,7 +6,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define i32 @main(i32 %argc, i8** nocapture %argv) {
 entry:
-  %0 = getelementptr inbounds i8* undef, i64 5    ; <i8*> [#uses=1]
+  %0 = getelementptr inbounds i8, i8* undef, i64 5    ; <i8*> [#uses=1]
   %1 = bitcast i8* %0 to i32*                     ; <i32*> [#uses=1]
   store i32 undef, i32* %1, align 1
   br i1 undef, label %k121.i.i, label %l117.i.i
@@ -29,9 +29,9 @@ l129.i.i:                                         ; preds = %k121.i.i
   unreachable
 
 k133.i.i:                                         ; preds = %k121.i.i
-  %2 = getelementptr i8* undef, i64 5             ; <i8*> [#uses=1]
+  %2 = getelementptr i8, i8* undef, i64 5             ; <i8*> [#uses=1]
   %3 = bitcast i8* %2 to i1*                      ; <i1*> [#uses=1]
-  %4 = load i1* %3                                ; <i1> [#uses=1]
+  %4 = load i1, i1* %3                                ; <i1> [#uses=1]
   br i1 %4, label %k151.i.i, label %l147.i.i
 
 l147.i.i:                                         ; preds = %k133.i.i
diff --git a/test/Transforms/GVN/2011-04-27-phioperands.ll b/test/Transforms/GVN/2011-04-27-phioperands.ll
index 42c46500c483..e964120a8b05 100644
--- a/test/Transforms/GVN/2011-04-27-phioperands.ll
+++ b/test/Transforms/GVN/2011-04-27-phioperands.ll
@@ -27,7 +27,7 @@ doemit.exit76.i:
   br label %"<bb 53>.i"
 
 "<L98>.i":
-  store i8* getelementptr inbounds ([10 x i8]* @nuls, i64 0, i64 0), i8** undef, align 8
+  store i8* getelementptr inbounds ([10 x i8], [10 x i8]* @nuls, i64 0, i64 0), i8** undef, align 8
   br label %"<bb 53>.i"
 
 "<L99>.i":
@@ -50,7 +50,7 @@ doemit.exit76.i:
 
 "<bb 53>.i":
   %wascaret_2.i = phi i32 [ 0, %"<L39>.i" ], [ 0, %"<L29>.i" ], [ 0, %"<L28>.i" ], [ 0, %"<bb 35>.i" ], [ 0, %"<L99>.i" ], [ 0, %"<L98>.i" ], [ 0, %doemit.exit76.i ], [ 1, %doemit.exit51.i ], [ 0, %"<L24>.i" ]
-  %D.5496_84.i = load i8** undef, align 8
+  %D.5496_84.i = load i8*, i8** undef, align 8
   br i1 undef, label %"<bb 54>.i", label %"<bb 5>"
 
 "<bb 54>.i":
diff --git a/test/Transforms/GVN/2011-06-01-NonLocalMemdepMiscompile.ll b/test/Transforms/GVN/2011-06-01-NonLocalMemdepMiscompile.ll
index 298f27401aaf..0769575759ba 100644
--- a/test/Transforms/GVN/2011-06-01-NonLocalMemdepMiscompile.ll
+++ b/test/Transforms/GVN/2011-06-01-NonLocalMemdepMiscompile.ll
@@ -19,10 +19,10 @@ bb1:
   br i1 undef, label %bb3, label %bb15
 
 ; CHECK: bb1:
-; CHECK: [[TMP:%.*]] = phi i8* [ getelementptr (i8* null, i64 undef), %bb10 ], [ null, %bb ]
+; CHECK: [[TMP:%.*]] = phi i8* [ getelementptr (i8, i8* null, i64 undef), %bb10 ], [ null, %bb ]
 
 ; CHECK: bb1.bb15_crit_edge:
-; CHECK: %tmp17.pre = load i8* [[TMP]], align 1
+; CHECK: %tmp17.pre = load i8, i8* [[TMP]], align 1
 
 bb3:
   call void @isalnum()
@@ -32,22 +32,22 @@ bb5:
   br i1 undef, label %bb10, label %bb6
 
 bb6:
-  %tmp7 = load i8** %tmp, align 8
-  %tmp8 = load i8* %tmp7, align 1
+  %tmp7 = load i8*, i8** %tmp, align 8
+  %tmp8 = load i8, i8* %tmp7, align 1
   %tmp9 = zext i8 %tmp8 to i64
   br i1 undef, label %bb15, label %bb10
 
 bb10:
-  %tmp11 = load i8** %tmp, align 8
-  %tmp12 = load i8* %tmp11, align 1
+  %tmp11 = load i8*, i8** %tmp, align 8
+  %tmp12 = load i8, i8* %tmp11, align 1
   %tmp13 = zext i8 %tmp12 to i64
-  %tmp14 = getelementptr inbounds i8* null, i64 undef
+  %tmp14 = getelementptr inbounds i8, i8* null, i64 undef
   store i8* %tmp14, i8** %tmp, align 8
   br label %bb1
 
 bb15:
-  %tmp16 = load i8** %tmp, align 8
-  %tmp17 = load i8* %tmp16, align 1
+  %tmp16 = load i8*, i8** %tmp, align 8
+  %tmp17 = load i8, i8* %tmp16, align 1
   %tmp18 = icmp eq i8 %tmp17, 0
   br label %bb19
 
diff --git a/test/Transforms/GVN/MemdepMiscompile.ll b/test/Transforms/GVN/MemdepMiscompile.ll
index d42016961575..0652304b3d7d 100644
--- a/test/Transforms/GVN/MemdepMiscompile.ll
+++ b/test/Transforms/GVN/MemdepMiscompile.ll
@@ -13,14 +13,14 @@ entry:
 ; CHECK: call void @RunInMode
 ; CHECK: br i1 %tobool, label %while.cond.backedge, label %if.then
 ; CHECK: while.cond.backedge:
-; CHECK: load i32* %shouldExit
+; CHECK: load i32, i32* %shouldExit
 ; CHECK: br i1 %cmp, label %while.body
   %shouldExit = alloca i32, align 4
   %tasksIdle = alloca i32, align 4
   store i32 0, i32* %shouldExit, align 4
   store i32 0, i32* %tasksIdle, align 4
   call void @CTestInitialize(i32* %tasksIdle) nounwind
-  %0 = load i32* %shouldExit, align 4
+  %0 = load i32, i32* %shouldExit, align 4
   %cmp1 = icmp eq i32 %0, 0
   br i1 %cmp1, label %while.body.lr.ph, label %while.end
 
@@ -29,7 +29,7 @@ while.body.lr.ph:
 
 while.body:
   call void @RunInMode(i32 100) nounwind
-  %1 = load i32* %tasksIdle, align 4
+  %1 = load i32, i32* %tasksIdle, align 4
   %tobool = icmp eq i32 %1, 0
   br i1 %tobool, label %while.cond.backedge, label %if.then
 
@@ -39,7 +39,7 @@ if.then:
   br label %while.cond.backedge
 
 while.cond.backedge:
-  %2 = load i32* %shouldExit, align 4
+  %2 = load i32, i32* %shouldExit, align 4
   %cmp = icmp eq i32 %2, 0
   br i1 %cmp, label %while.body, label %while.cond.while.end_crit_edge
 
diff --git a/test/Transforms/GVN/atomic.ll b/test/Transforms/GVN/atomic.ll
index 8c13d209c53a..11b54f39756b 100644
--- a/test/Transforms/GVN/atomic.ll
+++ b/test/Transforms/GVN/atomic.ll
@@ -11,21 +11,9 @@ define i32 @test1() nounwind uwtable ssp {
 ; CHECK-LABEL: test1
 ; CHECK: add i32 %x, %x
 entry:
-  %x = load i32* @y
+  %x = load i32, i32* @y
   store atomic i32 %x, i32* @x unordered, align 4
-  %y = load i32* @y
-  %z = add i32 %x, %y
-  ret i32 %z
-}
-
-; GVN across seq_cst store (allowed)
-define i32 @test2() nounwind uwtable ssp {
-; CHECK-LABEL: test2
-; CHECK: add i32 %x, %x
-entry:
-  %x = load i32* @y
-  store atomic i32 %x, i32* @x seq_cst, align 4
-  %y = load i32* @y
+  %y = load i32, i32* @y
   %z = add i32 %x, %y
   ret i32 %z
 }
@@ -35,35 +23,21 @@ define i32 @test3() nounwind uwtable ssp {
 ; CHECK-LABEL: test3
 ; CHECK: add i32 %x, %x
 entry:
-  %x = load i32* @y
-  %y = load atomic i32* @x unordered, align 4
-  %z = load i32* @y
+  %x = load i32, i32* @y
+  %y = load atomic i32, i32* @x unordered, align 4
+  %z = load i32, i32* @y
   %a = add i32 %x, %z
   %b = add i32 %y, %a
   ret i32 %b
 }
 
-; GVN across acquire load (allowed as the original load was not atomic)
-define i32 @test4() nounwind uwtable ssp {
-; CHECK-LABEL: test4
-; CHECK: load atomic i32* @x
-; CHECK-NOT: load i32* @y
-entry:
-  %x = load i32* @y
-  %y = load atomic i32* @x seq_cst, align 4
-  %x2 = load i32* @y
-  %x3 = add i32 %x, %x2
-  %y2 = add i32 %y, %x3
-  ret i32 %y2
-}
-
 ; GVN load to unordered load (allowed)
 define i32 @test5() nounwind uwtable ssp {
 ; CHECK-LABEL: test5
 ; CHECK: add i32 %x, %x
 entry:
-  %x = load atomic i32* @x unordered, align 4
-  %y = load i32* @x
+  %x = load atomic i32, i32* @x unordered, align 4
+  %y = load i32, i32* @x
   %z = add i32 %x, %y
   ret i32 %z
 }
@@ -71,10 +45,10 @@ entry:
 ; GVN unordered load to load (unordered load must not be removed)
 define i32 @test6() nounwind uwtable ssp {
 ; CHECK-LABEL: test6
-; CHECK: load atomic i32* @x unordered
+; CHECK: load atomic i32, i32* @x unordered
 entry:
-  %x = load i32* @x
-  %x2 = load atomic i32* @x unordered, align 4
+  %x = load i32, i32* @x
+  %x2 = load atomic i32, i32* @x unordered, align 4
   %x3 = add i32 %x, %x2
   ret i32 %x3
 }
@@ -84,23 +58,10 @@ define i32 @test7() nounwind uwtable ssp {
 ; CHECK-LABEL: test7
 ; CHECK: add i32 %x, %y
 entry:
-  %x = load i32* @y
+  %x = load i32, i32* @y
   store atomic i32 %x, i32* @x release, align 4
-  %w = load atomic i32* @x acquire, align 4
-  %y = load i32* @y
-  %z = add i32 %x, %y
-  ret i32 %z
-}
-
-; GVN across acquire-release pair (allowed)
-define i32 @test8() nounwind uwtable ssp {
-; CHECK-LABEL: test8
-; CHECK: add i32 %x, %x
-entry:
-  %x = load i32* @y
-  %w = load atomic i32* @x acquire, align 4
-  store atomic i32 %x, i32* @x release, align 4
-  %y = load i32* @y
+  %w = load atomic i32, i32* @x acquire, align 4
+  %y = load i32, i32* @y
   %z = add i32 %x, %y
   ret i32 %z
 }
@@ -110,9 +71,9 @@ define i32 @test9() nounwind uwtable ssp {
 ; CHECK-LABEL: test9
 ; CHECK: add i32 %x, %x
 entry:
-  %x = load i32* @y
+  %x = load i32, i32* @y
   store atomic i32 %x, i32* @x monotonic, align 4
-  %y = load i32* @y
+  %y = load i32, i32* @y
   %z = add i32 %x, %y
   ret i32 %z
 }
@@ -122,10 +83,27 @@ define i32 @test10() nounwind uwtable ssp {
 ; CHECK-LABEL: test10
 ; CHECK: add i32 %x, %y
 entry:
-  %x = load atomic i32* @y unordered, align 4
-  %clobber = load atomic i32* @x monotonic, align 4
-  %y = load atomic i32* @y monotonic, align 4
+  %x = load atomic i32, i32* @y unordered, align 4
+  %clobber = load atomic i32, i32* @x monotonic, align 4
+  %y = load atomic i32, i32* @y monotonic, align 4
   %z = add i32 %x, %y
   ret i32 %z
 }
 
+define i32 @PR22708(i1 %flag) {
+; CHECK-LABEL: PR22708
+entry:
+  br i1 %flag, label %if.then, label %if.end
+
+if.then:
+  store i32 43, i32* @y, align 4
+; CHECK: store i32 43, i32* @y, align 4
+  br label %if.end
+
+if.end:
+  load atomic i32, i32* @x acquire, align 4
+  %load = load i32, i32* @y, align 4
+; CHECK: load atomic i32, i32* @x acquire, align 4
+; CHECK: load i32, i32* @y, align 4
+  ret i32 %load
+}
diff --git a/test/Transforms/GVN/basic-undef-test.ll b/test/Transforms/GVN/basic-undef-test.ll
new file mode 100644
index 000000000000..4f9409567806
--- /dev/null
+++ b/test/Transforms/GVN/basic-undef-test.ll
@@ -0,0 +1,15 @@
+; RUN: opt -basicaa -gvn -S < %s | FileCheck %s
+; ModuleID = 'test3.ll'
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+define i32 @main(i32 *%foo)  {
+entry:
+; CHECK: load i32, i32* %foo, align 4
+  %0 = load i32, i32* %foo, align 4
+  store i32 5, i32* undef, align 4
+; CHECK-NOT: load i32, i32* %foo, align 4
+  %1 = load i32, i32* %foo, align 4
+; CHECK: add i32 %0, %0
+  %2 = add i32 %0, %1
+  ret i32 %2
+}
diff --git a/test/Transforms/GVN/calloc-load-removal.ll b/test/Transforms/GVN/calloc-load-removal.ll
index 2dde5b7b4146..a51f71fd77a1 100644
--- a/test/Transforms/GVN/calloc-load-removal.ll
+++ b/test/Transforms/GVN/calloc-load-removal.ll
@@ -9,11 +9,11 @@ define i32 @test1() {
   %1 = tail call noalias i8* @calloc(i64 1, i64 4)
   %2 = bitcast i8* %1 to i32*
   ; This load is trivially constant zero
-  %3 = load i32* %2, align 4
+  %3 = load i32, i32* %2, align 4
   ret i32 %3
 
 ; CHECK-LABEL: @test1(
-; CHECK-NOT: %3 = load i32* %2, align 4
+; CHECK-NOT: %3 = load i32, i32* %2, align 4
 ; CHECK: ret i32 0
 
 ; CHECK_NO_LIBCALLS-LABEL: @test1(
diff --git a/test/Transforms/GVN/calls-readonly.ll b/test/Transforms/GVN/calls-readonly.ll
index a4777401b190..ba8ad661eabf 100644
--- a/test/Transforms/GVN/calls-readonly.ll
+++ b/test/Transforms/GVN/calls-readonly.ll
@@ -20,7 +20,7 @@ bb1:                                              ; preds = %bb, %entry
   %4 = tail call i32 @strlen(i8* %P)              ; <i32> [#uses=1]
   %5 = add i32 %x_addr.0, %0                      ; <i32> [#uses=1]
   %.sum = sub i32 %5, %4                          ; <i32> [#uses=1]
-  %6 = getelementptr i8* %3, i32 %.sum            ; <i8*> [#uses=1]
+  %6 = getelementptr i8, i8* %3, i32 %.sum            ; <i8*> [#uses=1]
   ret i8* %6
 }
 
diff --git a/test/Transforms/GVN/cond_br.ll b/test/Transforms/GVN/cond_br.ll
index 918e7d41f12c..aeb1a6e6bf70 100644
--- a/test/Transforms/GVN/cond_br.ll
+++ b/test/Transforms/GVN/cond_br.ll
@@ -5,11 +5,11 @@
 ; Function Attrs: nounwind ssp uwtable
 define void @foo(i32 %x) {
 ; CHECK: @foo(i32 %x)
-; CHECK: %.pre = load i32* @y
+; CHECK: %.pre = load i32, i32* @y
 ; CHECK: call void @bar(i32 %.pre)
 
   %t = sub i32 %x, %x
-  %.pre = load i32* @y, align 4
+  %.pre = load i32, i32* @y, align 4
   %cmp = icmp sgt i32 %t, 2
   br i1 %cmp, label %if.then, label %entry.if.end_crit_edge
 
@@ -29,11 +29,11 @@ if.end:                                           ; preds = %entry.if.end_crit_e
 
 define void @foo2(i32 %x) {
 ; CHECK: @foo2(i32 %x)
-; CHECK: %.pre = load i32* @y
+; CHECK: %.pre = load i32, i32* @y
 ; CHECK: tail call void @bar(i32 %.pre)
 entry:
   %t = sub i32 %x, %x
-  %.pre = load i32* @y, align 4
+  %.pre = load i32, i32* @y, align 4
   %cmp = icmp sgt i32 %t, 2
   br i1 %cmp, label %if.then, label %if.else
 
diff --git a/test/Transforms/GVN/cond_br2.ll b/test/Transforms/GVN/cond_br2.ll
index a7ca2190563b..02154a783912 100644
--- a/test/Transforms/GVN/cond_br2.ll
+++ b/test/Transforms/GVN/cond_br2.ll
@@ -19,19 +19,19 @@ entry:
   %sv = alloca %"class.llvm::SmallVector", align 16
   %0 = bitcast %"class.llvm::SmallVector"* %sv to i8*
   call void @llvm.lifetime.start(i64 64, i8* %0) #1
-  %BeginX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 0
-  %FirstEl.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 3
+  %BeginX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 0
+  %FirstEl.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 3
   %1 = bitcast %"union.llvm::SmallVectorBase::U"* %FirstEl.i.i.i.i.i.i to i8*
   store i8* %1, i8** %BeginX.i.i.i.i.i.i, align 16, !tbaa !4
-  %EndX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 1
+  %EndX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 1
   store i8* %1, i8** %EndX.i.i.i.i.i.i, align 8, !tbaa !4
-  %CapacityX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 2
-  %add.ptr.i.i.i.i2.i.i = getelementptr inbounds %"union.llvm::SmallVectorBase::U"* %FirstEl.i.i.i.i.i.i, i64 2
+  %CapacityX.i.i.i.i.i.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 2
+  %add.ptr.i.i.i.i2.i.i = getelementptr inbounds %"union.llvm::SmallVectorBase::U", %"union.llvm::SmallVectorBase::U"* %FirstEl.i.i.i.i.i.i, i64 2
   %add.ptr.i.i.i.i.i.i = bitcast %"union.llvm::SmallVectorBase::U"* %add.ptr.i.i.i.i2.i.i to i8*
   store i8* %add.ptr.i.i.i.i.i.i, i8** %CapacityX.i.i.i.i.i.i, align 16, !tbaa !4
-  %EndX.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 1
-  %2 = load i8** %EndX.i, align 8, !tbaa !4
-  %CapacityX.i = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 2
+  %EndX.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 1
+  %2 = load i8*, i8** %EndX.i, align 8, !tbaa !4
+  %CapacityX.i = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0, i32 2
   %cmp.i = icmp ult i8* %2, %add.ptr.i.i.i.i.i.i
   br i1 %cmp.i, label %Retry.i, label %if.end.i
 
@@ -46,23 +46,23 @@ new.notnull.i:                                    ; preds = %Retry.i
   br label %invoke.cont
 
 if.end.i:                                         ; preds = %entry
-  %5 = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0
+  %5 = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0
   invoke void @_ZN4llvm15SmallVectorBase8grow_podEmm(%"class.llvm::SmallVectorBase"* %5, i64 0, i64 4)
           to label %.noexc unwind label %lpad
 
 .noexc:                                           ; preds = %if.end.i
-  %.pre.i = load i8** %EndX.i, align 8, !tbaa !4
+  %.pre.i = load i8*, i8** %EndX.i, align 8, !tbaa !4
   br label %Retry.i
 
 invoke.cont:                                      ; preds = %new.notnull.i, %Retry.i
-  %add.ptr.i = getelementptr inbounds i8* %3, i64 4
+  %add.ptr.i = getelementptr inbounds i8, i8* %3, i64 4
   store i8* %add.ptr.i, i8** %EndX.i, align 8, !tbaa !4
-  %6 = load i8** %CapacityX.i, align 16, !tbaa !4
+  %6 = load i8*, i8** %CapacityX.i, align 16, !tbaa !4
   %cmp.i8 = icmp ult i8* %add.ptr.i, %6
   br i1 %cmp.i8, label %new.notnull.i11, label %if.end.i14
 
 Retry.i10:                                        ; preds = %if.end.i14
-  %.pre.i13 = load i8** %EndX.i, align 8, !tbaa !4
+  %.pre.i13 = load i8*, i8** %EndX.i, align 8, !tbaa !4
   %new.isnull.i9 = icmp eq i8* %.pre.i13, null
   br i1 %new.isnull.i9, label %invoke.cont2, label %new.notnull.i11
 
@@ -73,19 +73,19 @@ new.notnull.i11:                                  ; preds = %invoke.cont, %Retry
   br label %invoke.cont2
 
 if.end.i14:                                       ; preds = %invoke.cont
-  %9 = getelementptr inbounds %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0
+  %9 = getelementptr inbounds %"class.llvm::SmallVector", %"class.llvm::SmallVector"* %sv, i64 0, i32 0, i32 0, i32 0, i32 0
   invoke void @_ZN4llvm15SmallVectorBase8grow_podEmm(%"class.llvm::SmallVectorBase"* %9, i64 0, i64 4)
           to label %Retry.i10 unwind label %lpad
 
 invoke.cont2:                                     ; preds = %new.notnull.i11, %Retry.i10
   %10 = phi i8* [ null, %Retry.i10 ], [ %7, %new.notnull.i11 ]
-  %add.ptr.i12 = getelementptr inbounds i8* %10, i64 4
+  %add.ptr.i12 = getelementptr inbounds i8, i8* %10, i64 4
   store i8* %add.ptr.i12, i8** %EndX.i, align 8, !tbaa !4
   invoke void @_Z1gRN4llvm11SmallVectorIiLj8EEE(%"class.llvm::SmallVector"* %sv)
           to label %invoke.cont3 unwind label %lpad
 
 invoke.cont3:                                     ; preds = %invoke.cont2
-  %11 = load i8** %BeginX.i.i.i.i.i.i, align 16, !tbaa !4
+  %11 = load i8*, i8** %BeginX.i.i.i.i.i.i, align 16, !tbaa !4
   %cmp.i.i.i.i19 = icmp eq i8* %11, %1
   br i1 %cmp.i.i.i.i19, label %_ZN4llvm11SmallVectorIiLj8EED1Ev.exit21, label %if.then.i.i.i20
 
@@ -100,7 +100,7 @@ _ZN4llvm11SmallVectorIiLj8EED1Ev.exit21:          ; preds = %invoke.cont3, %if.t
 lpad:                                             ; preds = %if.end.i14, %if.end.i, %invoke.cont2
   %12 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
           cleanup
-  %13 = load i8** %BeginX.i.i.i.i.i.i, align 16, !tbaa !4
+  %13 = load i8*, i8** %BeginX.i.i.i.i.i.i, align 16, !tbaa !4
   %cmp.i.i.i.i = icmp eq i8* %13, %1
   br i1 %cmp.i.i.i.i, label %eh.resume, label %if.then.i.i.i
 
diff --git a/test/Transforms/GVN/condprop.ll b/test/Transforms/GVN/condprop.ll
index 845f88e15895..6aa3cb862509 100644
--- a/test/Transforms/GVN/condprop.ll
+++ b/test/Transforms/GVN/condprop.ll
@@ -5,7 +5,7 @@
 ; CHECK-LABEL: @test1(
 define i32 @test1() nounwind {
 entry:
-	%0 = load i32* @a, align 4
+	%0 = load i32, i32* @a, align 4
 	%1 = icmp eq i32 %0, 4
 	br i1 %1, label %bb, label %bb1
 
@@ -13,7 +13,7 @@ bb:		; preds = %entry
 	br label %bb8
 
 bb1:		; preds = %entry
-	%2 = load i32* @a, align 4
+	%2 = load i32, i32* @a, align 4
 	%3 = icmp eq i32 %2, 5
 	br i1 %3, label %bb2, label %bb3
 
@@ -21,29 +21,29 @@ bb2:		; preds = %bb1
 	br label %bb8
 
 bb3:		; preds = %bb1
-	%4 = load i32* @a, align 4
+	%4 = load i32, i32* @a, align 4
 	%5 = icmp eq i32 %4, 4
 ; CHECK: br i1 false, label %bb4, label %bb5
 	br i1 %5, label %bb4, label %bb5
 
 bb4:		; preds = %bb3
-	%6 = load i32* @a, align 4
+	%6 = load i32, i32* @a, align 4
 	%7 = add i32 %6, 5
 	br label %bb8
 
 bb5:		; preds = %bb3
-	%8 = load i32* @a, align 4
+	%8 = load i32, i32* @a, align 4
 	%9 = icmp eq i32 %8, 5
 ; CHECK: br i1 false, label %bb6, label %bb7
 	br i1 %9, label %bb6, label %bb7
 
 bb6:		; preds = %bb5
-	%10 = load i32* @a, align 4
+	%10 = load i32, i32* @a, align 4
 	%11 = add i32 %10, 4
 	br label %bb8
 
 bb7:		; preds = %bb5
-	%12 = load i32* @a, align 4
+	%12 = load i32, i32* @a, align 4
 	br label %bb8
 
 bb8:		; preds = %bb7, %bb6, %bb4, %bb2, %bb
diff --git a/test/Transforms/GVN/crash-no-aa.ll b/test/Transforms/GVN/crash-no-aa.ll
index 9ad63a7350c2..f076a8d81ace 100644
--- a/test/Transforms/GVN/crash-no-aa.ll
+++ b/test/Transforms/GVN/crash-no-aa.ll
@@ -5,11 +5,11 @@ target triple = "x86_64-unknown-freebsd8.0"
 
 ; PR5744
 define i32 @test1({i16, i32} *%P) {
-  %P2 = getelementptr {i16, i32} *%P, i32 0, i32 0
+  %P2 = getelementptr {i16, i32}, {i16, i32} *%P, i32 0, i32 0
   store i16 42, i16* %P2
 
-  %P3 = getelementptr {i16, i32} *%P, i32 0, i32 1
-  %V = load i32* %P3
+  %P3 = getelementptr {i16, i32}, {i16, i32} *%P, i32 0, i32 1
+  %V = load i32, i32* %P3
   ret i32 %V
 }
 
diff --git a/test/Transforms/GVN/crash.ll b/test/Transforms/GVN/crash.ll
index 9fb612fcae13..2abb4194f159 100644
--- a/test/Transforms/GVN/crash.ll
+++ b/test/Transforms/GVN/crash.ll
@@ -22,7 +22,7 @@ while.body:                                       ; preds = %while.body.backedge
 
 lor.lhs.false:                                    ; preds = %while.body
   %tmp20 = bitcast i32* %o.addr.0 to i32*         ; <i32*> [#uses=1]
-  %tmp22 = load i32* %tmp20                       ; <i32> [#uses=0]
+  %tmp22 = load i32, i32* %tmp20                       ; <i32> [#uses=0]
   br i1 undef, label %land.lhs.true24, label %if.end31
 
 land.lhs.true24:                                  ; preds = %lor.lhs.false
@@ -34,11 +34,11 @@ if.end31:                                         ; preds = %land.lhs.true24, %l
 
 if.end41:                                         ; preds = %if.end31
   %tmp43 = bitcast i32* %o.addr.0 to i32*         ; <i32*> [#uses=1]
-  %tmp45 = load i32* %tmp43                       ; <i32> [#uses=0]
+  %tmp45 = load i32, i32* %tmp43                       ; <i32> [#uses=0]
   br i1 undef, label %if.then50, label %if.else
 
 if.then50:                                        ; preds = %if.end41
-  %tmp53 = load i32** undef                       ; <i32*> [#uses=1]
+  %tmp53 = load i32*, i32** undef                       ; <i32*> [#uses=1]
   br label %while.body.backedge
 
 if.else:                                          ; preds = %if.end41
@@ -74,16 +74,16 @@ entry:
   br label %bb69.i
 
 bb69.i:                                           ; preds = %bb57.i.preheader
-  %tmp4 = getelementptr inbounds [4 x %struct.attribute_spec*]* @attribute_tables, i32 0, i32 undef ; <%struct.attribute_spec**> [#uses=1]
-  %tmp3 = load %struct.attribute_spec** %tmp4, align 4 ; <%struct.attribute_spec*> [#uses=1]
+  %tmp4 = getelementptr inbounds [4 x %struct.attribute_spec*], [4 x %struct.attribute_spec*]* @attribute_tables, i32 0, i32 undef ; <%struct.attribute_spec**> [#uses=1]
+  %tmp3 = load %struct.attribute_spec*, %struct.attribute_spec** %tmp4, align 4 ; <%struct.attribute_spec*> [#uses=1]
   br label %bb65.i
 
 bb65.i:                                           ; preds = %bb65.i.preheader, %bb64.i
   %storemerge6.i = phi i32 [ 1, %bb64.i ], [ 0, %bb69.i ] ; <i32> [#uses=3]
-  %scevgep14 = getelementptr inbounds %struct.attribute_spec* %tmp3, i32 %storemerge6.i, i32 0 ; <i8**> [#uses=1]
-  %tmp2 = load i8** %scevgep14, align 4           ; <i8*> [#uses=0]
-  %tmp = load %struct.attribute_spec** %tmp4, align 4 ; <%struct.attribute_spec*> [#uses=1]
-  %scevgep1516 = getelementptr inbounds %struct.attribute_spec* %tmp, i32 %storemerge6.i, i32 0 ; <i8**> [#uses=0]
+  %scevgep14 = getelementptr inbounds %struct.attribute_spec, %struct.attribute_spec* %tmp3, i32 %storemerge6.i, i32 0 ; <i8**> [#uses=1]
+  %tmp2 = load i8*, i8** %scevgep14, align 4           ; <i8*> [#uses=0]
+  %tmp = load %struct.attribute_spec*, %struct.attribute_spec** %tmp4, align 4 ; <%struct.attribute_spec*> [#uses=1]
+  %scevgep1516 = getelementptr inbounds %struct.attribute_spec, %struct.attribute_spec* %tmp, i32 %storemerge6.i, i32 0 ; <i8**> [#uses=0]
   unreachable
 
 bb64.i:                                           ; Unreachable
@@ -101,7 +101,7 @@ bb66.i:                                           ; Unreachable
 
 define i32* @test3() {
 do.end17.i:
-  %tmp18.i = load i7** undef
+  %tmp18.i = load i7*, i7** undef
   %tmp1 = bitcast i7* %tmp18.i to i8*
   br i1 undef, label %do.body36.i, label %if.then21.i
 
@@ -110,12 +110,12 @@ if.then21.i:
   ret i32* undef
 
 do.body36.i:
-  %ivar38.i = load i64* @g 
+  %ivar38.i = load i64, i64* @g 
   %tmp3 = bitcast i7* %tmp18.i to i8*
   %add.ptr39.sum.i = add i64 %ivar38.i, 8
-  %tmp40.i = getelementptr inbounds i8* %tmp3, i64 %add.ptr39.sum.i
+  %tmp40.i = getelementptr inbounds i8, i8* %tmp3, i64 %add.ptr39.sum.i
   %tmp4 = bitcast i8* %tmp40.i to i64*
-  %tmp41.i = load i64* %tmp4
+  %tmp41.i = load i64, i64* %tmp4
   br i1 undef, label %if.then48.i, label %do.body57.i
 
 if.then48.i:
@@ -123,13 +123,13 @@ if.then48.i:
   br label %do.body57.i
 
 do.body57.i:
-  %tmp58.i = load i7** undef
-  %ivar59.i = load i64* @g
+  %tmp58.i = load i7*, i7** undef
+  %ivar59.i = load i64, i64* @g
   %tmp5 = bitcast i7* %tmp58.i to i8*
   %add.ptr65.sum.i = add i64 %ivar59.i, 8
-  %tmp66.i = getelementptr inbounds i8* %tmp5, i64 %add.ptr65.sum.i
+  %tmp66.i = getelementptr inbounds i8, i8* %tmp5, i64 %add.ptr65.sum.i
   %tmp6 = bitcast i8* %tmp66.i to i64*
-  %tmp67.i = load i64* %tmp6
+  %tmp67.i = load i64, i64* %tmp6
   ret i32* undef
 }
 
@@ -142,10 +142,10 @@ entry:
   ret i32 0
   
 dead:
-  %P2 = getelementptr i32 *%P2, i32 52
-  %Q2 = getelementptr i32 *%Q2, i32 52
+  %P2 = getelementptr i32, i32 *%P2, i32 52
+  %Q2 = getelementptr i32, i32 *%Q2, i32 52
   store i32 4, i32* %P2
-  %A = load i32* %Q2
+  %A = load i32, i32* %Q2
   br i1 true, label %dead, label %dead2
   
 dead2:
@@ -156,10 +156,10 @@ dead2:
 ; PR9841
 define fastcc i8 @test5(i8* %P) nounwind {
 entry:
-  %0 = load i8* %P, align 2
+  %0 = load i8, i8* %P, align 2
 
-  %Q = getelementptr i8* %P, i32 1
-  %1 = load i8* %Q, align 1
+  %Q = getelementptr i8, i8* %P, i32 1
+  %1 = load i8, i8* %Q, align 1
   ret i8 %1
 }
 
@@ -178,16 +178,16 @@ entry:
 ;; Unreachable code.
 
 unreachable.bb:
-  %gep.val = getelementptr inbounds %struct.type* %gep.val, i64 1
+  %gep.val = getelementptr inbounds %struct.type, %struct.type* %gep.val, i64 1
   br i1 undef, label %u2.bb, label %u1.bb
 
 u1.bb:
-  %tmp1 = getelementptr inbounds %struct.type* %gep.val, i64 0, i32 0
+  %tmp1 = getelementptr inbounds %struct.type, %struct.type* %gep.val, i64 0, i32 0
   store i64 -1, i64* %tmp1, align 8
   br label %unreachable.bb
 
 u2.bb:
-  %0 = load i32* undef, align 4
+  %0 = load i32, i32* undef, align 4
   %conv.i.i.i.i.i = zext i32 %0 to i64
   br label %u2.bb
 
diff --git a/test/Transforms/GVN/invariant-load.ll b/test/Transforms/GVN/invariant-load.ll
index 2f9f2a6100fb..982da8cfe486 100644
--- a/test/Transforms/GVN/invariant-load.ll
+++ b/test/Transforms/GVN/invariant-load.ll
@@ -3,13 +3,13 @@
 
 define i32 @test1(i32* nocapture %p, i8* nocapture %q) {
 ; CHECK-LABEL: test1
-; CHECK: %x = load i32* %p, align 4, !invariant.load !0
+; CHECK: %x = load i32, i32* %p, align 4, !invariant.load !0
 ; CHECK-NOT: %y = load
 entry:
-  %x = load i32* %p, align 4, !invariant.load !0
+  %x = load i32, i32* %p, align 4, !invariant.load !0
   %conv = trunc i32 %x to i8
   store i8 %conv, i8* %q, align 1
-  %y = load i32* %p, align 4, !invariant.load !0
+  %y = load i32, i32* %p, align 4, !invariant.load !0
   %add = add i32 %y, 1
   ret i32 %add
 }
@@ -19,13 +19,101 @@ define i32 @test2(i32* nocapture %p, i8* nocapture %q) {
 ; CHECK-NOT: !invariant.load
 ; CHECK-NOT: %y = load
 entry:
-  %x = load i32* %p, align 4
+  %x = load i32, i32* %p, align 4
   %conv = trunc i32 %x to i8
   store i8 %conv, i8* %q, align 1
-  %y = load i32* %p, align 4, !invariant.load !0
+  %y = load i32, i32* %p, align 4, !invariant.load !0
   %add = add i32 %y, 1
   ret i32 %add
 }
 
+; With the invariant.load metadata, what would otherwise
+; be a case for PRE becomes a full redundancy.
+define i32 @test3(i1 %cnd, i32* %p, i32* %q) {
+; CHECK-LABEL: test3
+; CHECK-NOT: load
+entry:
+  %v1 = load i32, i32* %p
+  br i1 %cnd, label %bb1, label %bb2
+
+bb1:
+  store i32 5, i32* %q
+  br label %bb2
+
+bb2:
+  %v2 = load i32, i32* %p, !invariant.load !0
+  %res = sub i32 %v1, %v2
+  ret i32 %res
+}
+
+; This test is here to document a case which doesn't optimize
+; as well as it could.  
+define i32 @test4(i1 %cnd, i32* %p, i32* %q) {
+; CHECK-LABEL: test4
+; %v2 is redundant, but GVN currently doesn't catch that
+entry:
+  %v1 = load i32, i32* %p, !invariant.load !0
+  br i1 %cnd, label %bb1, label %bb2
+
+bb1:
+  store i32 5, i32* %q
+  br label %bb2
+
+bb2:
+  %v2 = load i32, i32* %p
+  %res = sub i32 %v1, %v2
+  ret i32 %res
+}
+
+; Checks that we return the mustalias store as a def
+; so that it contributes to value forwarding.  Note
+; that we could and should remove the store too.
+define i32 @test5(i1 %cnd, i32* %p) {
+; CHECK-LABEL: test5
+; CHECK-LABEL: entry:
+; CHECK-NEXT: store i32 5, i32* %p
+; CHECK-NEXT: ret i32 5
+entry:
+  %v1 = load i32, i32* %p, !invariant.load !0
+  store i32 5, i32* %p ;; must alias store, want to exploit
+  %v2 = load i32, i32* %p, !invariant.load !0
+  ret i32 %v2
+}
+
+
+declare void @foo()
+
+; Clobbering (mayalias) stores, even in function calls, can be ignored
+define i32 @test6(i1 %cnd, i32* %p) {
+; CHECK-LABEL: test6
+; CHECK-LABEL: entry:
+; CHECK-NEXT: @foo
+; CHECK-NEXT: ret i32 0
+entry:
+  %v1 = load i32, i32* %p, !invariant.load !0
+  call void @foo()
+  %v2 = load i32, i32* %p, !invariant.load !0
+  %res = sub i32 %v1, %v2
+  ret i32 %res
+}
+
+declare noalias i32* @bar(...) 
+
+; Same as previous, but a function with a noalias result (since they're handled
+; differently in MDA)
+define i32 @test7(i1 %cnd, i32* %p) {
+; CHECK-LABEL: test7
+; CHECK-LABEL: entry:
+; CHECK-NEXT: @bar
+; CHECK-NEXT: ret i32 0
+entry:
+  %v1 = load i32, i32* %p, !invariant.load !0
+  call i32* (...) @bar(i32* %p)
+  %v2 = load i32, i32* %p, !invariant.load !0
+  %res = sub i32 %v1, %v2
+  ret i32 %res
+}
+
+
 !0 = !{ }
 
diff --git a/test/Transforms/GVN/lifetime-simple.ll b/test/Transforms/GVN/lifetime-simple.ll
index 02f7bcc9e857..d03b62c8158a 100644
--- a/test/Transforms/GVN/lifetime-simple.ll
+++ b/test/Transforms/GVN/lifetime-simple.ll
@@ -9,10 +9,10 @@ define i8 @test(i8* %P) nounwind {
 ; CHECK: lifetime.end
 entry:
   call void @llvm.lifetime.start(i64 32, i8* %P)
-  %0 = load i8* %P
+  %0 = load i8, i8* %P
   store i8 1, i8* %P
   call void @llvm.lifetime.end(i64 32, i8* %P)
-  %1 = load i8* %P
+  %1 = load i8, i8* %P
   ret i8 %1
 }
 
diff --git a/test/Transforms/GVN/load-constant-mem.ll b/test/Transforms/GVN/load-constant-mem.ll
index a7dacea6b5da..f870485630fa 100644
--- a/test/Transforms/GVN/load-constant-mem.ll
+++ b/test/Transforms/GVN/load-constant-mem.ll
@@ -4,10 +4,10 @@
 
 define i32 @test(i8* %p, i32 %i) nounwind {
 entry:
-	%P = getelementptr [4 x i32]* @G, i32 0, i32 %i
-	%A = load i32* %P
+	%P = getelementptr [4 x i32], [4 x i32]* @G, i32 0, i32 %i
+	%A = load i32, i32* %P
 	store i8 4, i8* %p
-	%B = load i32* %P
+	%B = load i32, i32* %P
 	%C = sub i32 %A, %B
 	ret i32 %C
 }
diff --git a/test/Transforms/GVN/load-from-unreachable-predecessor.ll b/test/Transforms/GVN/load-from-unreachable-predecessor.ll
index b676d958aa4a..29ea14d62b40 100644
--- a/test/Transforms/GVN/load-from-unreachable-predecessor.ll
+++ b/test/Transforms/GVN/load-from-unreachable-predecessor.ll
@@ -8,13 +8,13 @@ define i32 @f(i32** %f) {
 ; Load should be removed, since it's ignored.
 ; CHECK-NEXT: br label
 bb0:
-  %bar = load i32** %f
+  %bar = load i32*, i32** %f
   br label %bb2
 bb1:
-  %zed = load i32** %f
+  %zed = load i32*, i32** %f
   br i1 false, label %bb1, label %bb2
 bb2:
   %foo = phi i32* [ null, %bb0 ], [ %zed, %bb1 ]
-  %storemerge = load i32* %foo
+  %storemerge = load i32, i32* %foo
   ret i32 %storemerge
 }
diff --git a/test/Transforms/GVN/load-pre-align.ll b/test/Transforms/GVN/load-pre-align.ll
index 4816af2f441b..1198cafaeed9 100644
--- a/test/Transforms/GVN/load-pre-align.ll
+++ b/test/Transforms/GVN/load-pre-align.ll
@@ -25,7 +25,7 @@ for.cond.for.end_crit_edge:
   br label %for.end
 
 for.body:
-  %tmp3 = load i32* @p, align 8
+  %tmp3 = load i32, i32* @p, align 8
   %dec = add i32 %tmp3, -1
   store i32 %dec, i32* @p
   %cmp6 = icmp slt i32 %dec, 0
@@ -39,6 +39,6 @@ for.inc:
   br label %for.cond
 
 for.end:
-  %tmp9 = load i32* @p, align 8
+  %tmp9 = load i32, i32* @p, align 8
   ret i32 %tmp9
 }
diff --git a/test/Transforms/GVN/load-pre-licm.ll b/test/Transforms/GVN/load-pre-licm.ll
index 63541ad181d1..d14b01caf779 100644
--- a/test/Transforms/GVN/load-pre-licm.ll
+++ b/test/Transforms/GVN/load-pre-licm.ll
@@ -16,11 +16,11 @@ entry:
 while.body5:
   %indvar = phi i32 [ 0, %entry ], [ %tmp6, %if.end ]
   %tmp5 = add i32 %indvar, 2
-  %arrayidx9 = getelementptr [5001 x i32]* @sortlist, i32 0, i32 %tmp5
+  %arrayidx9 = getelementptr [5001 x i32], [5001 x i32]* @sortlist, i32 0, i32 %tmp5
   %tmp6 = add i32 %indvar, 1
-  %arrayidx = getelementptr [5001 x i32]* @sortlist, i32 0, i32 %tmp6
-  %tmp7 = load i32* %arrayidx, align 4
-  %tmp10 = load i32* %arrayidx9, align 4
+  %arrayidx = getelementptr [5001 x i32], [5001 x i32]* @sortlist, i32 0, i32 %tmp6
+  %tmp7 = load i32, i32* %arrayidx, align 4
+  %tmp10 = load i32, i32* %arrayidx9, align 4
   %cmp11 = icmp sgt i32 %tmp7, %tmp10
   br i1 %cmp11, label %if.then, label %if.end
 
diff --git a/test/Transforms/GVN/load-pre-nonlocal.ll b/test/Transforms/GVN/load-pre-nonlocal.ll
index ae508b96cf59..c75e54db6287 100644
--- a/test/Transforms/GVN/load-pre-nonlocal.ll
+++ b/test/Transforms/GVN/load-pre-nonlocal.ll
@@ -12,8 +12,8 @@ target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 
 ; CHECK-LABEL: define i32 @volatile_load
 ; CHECK: for.body:
-; CHECK: %2 = load i32*
-; CHECK: %3 = load volatile i32*
+; CHECK: %2 = load i32, i32*
+; CHECK: %3 = load volatile i32, i32*
 ; CHECK: for.cond.for.end_crit_edge:
 
 define i32 @volatile_load(i32 %n) {
@@ -22,21 +22,21 @@ entry:
   br i1 %cmp6, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load i32** @a2, align 8, !tbaa !1
-  %1 = load i32** @a, align 8, !tbaa !1
+  %0 = load i32*, i32** @a2, align 8, !tbaa !1
+  %1 = load i32*, i32** @a, align 8, !tbaa !1
   br label %for.body
 
 for.body:
   %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
   %s.09 = phi i32 [ 0, %for.body.lr.ph ], [ %add, %for.body ]
   %p.08 = phi i32* [ %0, %for.body.lr.ph ], [ %incdec.ptr, %for.body ]
-  %2 = load i32* %p.08, align 4, !tbaa !5
-  %arrayidx = getelementptr inbounds i32* %1, i64 %indvars.iv
+  %2 = load i32, i32* %p.08, align 4, !tbaa !5
+  %arrayidx = getelementptr inbounds i32, i32* %1, i64 %indvars.iv
   store i32 %2, i32* %arrayidx, align 4, !tbaa !5
-  %3 = load volatile i32* %p.08, align 4, !tbaa !5
+  %3 = load volatile i32, i32* %p.08, align 4, !tbaa !5
   %add = add nsw i32 %3, %s.09
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
-  %incdec.ptr = getelementptr inbounds i32* %p.08, i64 1
+  %incdec.ptr = getelementptr inbounds i32, i32* %p.08, i64 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp ne i32 %lftr.wideiv, %n
   br i1 %exitcond, label %for.body, label %for.cond.for.end_crit_edge
@@ -54,7 +54,7 @@ for.end:
 
 ; CHECK-LABEL: define i32 @overaligned_load
 ; CHECK: if.end:
-; CHECK-NOT: %1 = load i32*
+; CHECK-NOT: %1 = load i32, i32*
 
 define i32 @overaligned_load(i32 %a, i32* nocapture %b) {
 entry:
@@ -62,19 +62,19 @@ entry:
   br i1 %cmp, label %if.then, label %if.else
 
 if.then:
-  %0 = load i32* getelementptr inbounds (%struct.S1* @s1, i64 0, i32 0), align 8, !tbaa !5
+  %0 = load i32, i32* getelementptr inbounds (%struct.S1, %struct.S1* @s1, i64 0, i32 0), align 8, !tbaa !5
   br label %if.end
 
 if.else:
-  %arrayidx = getelementptr inbounds i32* %b, i64 2
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 2
   store i32 10, i32* %arrayidx, align 4, !tbaa !5
   br label %if.end
 
 if.end:
   %i.0 = phi i32 [ %0, %if.then ], [ 0, %if.else ]
-  %p.0 = phi i32* [ getelementptr inbounds (%struct.S1* @s1, i64 0, i32 0), %if.then ], [ %b, %if.else ]
-  %add.ptr = getelementptr inbounds i32* %p.0, i64 1
-  %1 = load i32* %add.ptr, align 4, !tbaa !5
+  %p.0 = phi i32* [ getelementptr inbounds (%struct.S1, %struct.S1* @s1, i64 0, i32 0), %if.then ], [ %b, %if.else ]
+  %add.ptr = getelementptr inbounds i32, i32* %p.0, i64 1
+  %1 = load i32, i32* %add.ptr, align 4, !tbaa !5
   %add1 = add nsw i32 %1, %i.0
   ret i32 %add1
 }
diff --git a/test/Transforms/GVN/lpre-call-wrap-2.ll b/test/Transforms/GVN/lpre-call-wrap-2.ll
index 35e3534a9c89..5dc779ef40d1 100644
--- a/test/Transforms/GVN/lpre-call-wrap-2.ll
+++ b/test/Transforms/GVN/lpre-call-wrap-2.ll
@@ -16,10 +16,10 @@ target triple = "i386-apple-darwin7"
 define void @bi_windup(i8* %outbuf, i8 zeroext %bi_buf) nounwind {
 entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%0 = load i32* @outcnt, align 4		; <i32> [#uses=1]
-	%1 = getelementptr i8* %outbuf, i32 %0		; <i8*> [#uses=1]
+	%0 = load i32, i32* @outcnt, align 4		; <i32> [#uses=1]
+	%1 = getelementptr i8, i8* %outbuf, i32 %0		; <i8*> [#uses=1]
 	store i8 %bi_buf, i8* %1, align 1
-	%2 = load i32* @outcnt, align 4		; <i32> [#uses=1]
+	%2 = load i32, i32* @outcnt, align 4		; <i32> [#uses=1]
 	%3 = icmp eq i32 %2, 16384		; <i1> [#uses=1]
 	br i1 %3, label %bb, label %bb1
 
@@ -31,8 +31,8 @@ bb1:		; preds = %bb, %entry
 ; CHECK: bb1:
 ; CHECK-NEXT: phi
 ; CHECK-NEXT: getelementptr
-	%4 = load i32* @outcnt, align 4		; <i32> [#uses=1]
-	%5 = getelementptr i8* %outbuf, i32 %4		; <i8*> [#uses=1]
+	%4 = load i32, i32* @outcnt, align 4		; <i32> [#uses=1]
+	%5 = getelementptr i8, i8* %outbuf, i32 %4		; <i8*> [#uses=1]
 	store i8 %bi_buf, i8* %5, align 1
 	ret void
 }
diff --git a/test/Transforms/GVN/lpre-call-wrap.ll b/test/Transforms/GVN/lpre-call-wrap.ll
index 0646f3fe0aad..274830549394 100644
--- a/test/Transforms/GVN/lpre-call-wrap.ll
+++ b/test/Transforms/GVN/lpre-call-wrap.ll
@@ -19,20 +19,20 @@ target triple = "i386-apple-darwin7"
 
 define void @_Z12testfunctionR1A(%struct.A* %iter) {
 entry:
-	%0 = getelementptr %struct.A* %iter, i32 0, i32 0		; <i32*> [#uses=3]
-	%1 = load i32* %0, align 4		; <i32> [#uses=2]
+	%0 = getelementptr %struct.A, %struct.A* %iter, i32 0, i32 0		; <i32*> [#uses=3]
+	%1 = load i32, i32* %0, align 4		; <i32> [#uses=2]
 	%2 = icmp eq i32 %1, 0		; <i1> [#uses=1]
 	br i1 %2, label %return, label %bb.nph
 
 bb.nph:		; preds = %entry
-	%3 = getelementptr %struct.A* %iter, i32 0, i32 1		; <i32*> [#uses=1]
+	%3 = getelementptr %struct.A, %struct.A* %iter, i32 0, i32 1		; <i32*> [#uses=1]
 	br label %bb
 
 bb:		; preds = %bb3.backedge, %bb.nph
 	%.rle = phi i32 [ %1, %bb.nph ], [ %7, %bb3.backedge ]		; <i32> [#uses=1]
 	%4 = add i32 %.rle, 1		; <i32> [#uses=2]
 	store i32 %4, i32* %0, align 4
-	%5 = load i32* %3, align 4		; <i32> [#uses=1]
+	%5 = load i32, i32* %3, align 4		; <i32> [#uses=1]
 	%6 = icmp eq i32 %4, %5		; <i1> [#uses=1]
 	br i1 %6, label %bb1, label %bb3.backedge
 
@@ -44,7 +44,7 @@ bb3.backedge:		; preds = %bb, %bb1
 ; CHECK: bb3.backedge:
 ; CHECK-NEXT: phi
 ; CHECK-NEXT: icmp
-	%7 = load i32* %0, align 4		; <i32> [#uses=2]
+	%7 = load i32, i32* %0, align 4		; <i32> [#uses=2]
 	%8 = icmp eq i32 %7, 0		; <i1> [#uses=1]
 	br i1 %8, label %return, label %bb
 
diff --git a/test/Transforms/GVN/malloc-load-removal.ll b/test/Transforms/GVN/malloc-load-removal.ll
index d2d2fd77afec..1d7a2ddc4c2d 100644
--- a/test/Transforms/GVN/malloc-load-removal.ll
+++ b/test/Transforms/GVN/malloc-load-removal.ll
@@ -10,7 +10,7 @@ declare i8* @malloc(i64) nounwind
 define noalias i8* @test1() nounwind uwtable ssp {
 entry:
   %call = tail call i8* @malloc(i64 100) nounwind
-  %0 = load i8* %call, align 1
+  %0 = load i8, i8* %call, align 1
   %tobool = icmp eq i8 %0, 0
   br i1 %tobool, label %if.end, label %if.then
 
@@ -35,7 +35,7 @@ declare i8* @_Znwm(i64) nounwind
 define noalias i8* @test2() nounwind uwtable ssp {
 entry:
   %call = tail call i8* @_Znwm(i64 100) nounwind
-  %0 = load i8* %call, align 1
+  %0 = load i8, i8* %call, align 1
   %tobool = icmp eq i8 %0, 0
   br i1 %tobool, label %if.end, label %if.then
 
diff --git a/test/Transforms/GVN/noalias.ll b/test/Transforms/GVN/noalias.ll
index 6c310fad7ec6..cfff096b6952 100644
--- a/test/Transforms/GVN/noalias.ll
+++ b/test/Transforms/GVN/noalias.ll
@@ -2,21 +2,21 @@
 
 define i32 @test1(i32* %p, i32* %q) {
 ; CHECK-LABEL: @test1(i32* %p, i32* %q)
-; CHECK: load i32* %p
+; CHECK: load i32, i32* %p
 ; CHECK-NOT: noalias
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !noalias !0
-  %b = load i32* %p
+  %a = load i32, i32* %p, !noalias !0
+  %b = load i32, i32* %p
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test2(i32* %p, i32* %q) {
 ; CHECK-LABEL: @test2(i32* %p, i32* %q)
-; CHECK: load i32* %p, !alias.scope !0
+; CHECK: load i32, i32* %p, !alias.scope !0
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !alias.scope !0
-  %b = load i32* %p, !alias.scope !0
+  %a = load i32, i32* %p, !alias.scope !0
+  %b = load i32, i32* %p, !alias.scope !0
   %c = add i32 %a, %b
   ret i32 %c
 }
@@ -27,10 +27,10 @@ define i32 @test2(i32* %p, i32* %q) {
 ; throw in between.
 define i32 @test3(i32* %p, i32* %q) {
 ; CHECK-LABEL: @test3(i32* %p, i32* %q)
-; CHECK: load i32* %p, !alias.scope !1
+; CHECK: load i32, i32* %p, !alias.scope !1
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !alias.scope !1
-  %b = load i32* %p, !alias.scope !2
+  %a = load i32, i32* %p, !alias.scope !1
+  %b = load i32, i32* %p, !alias.scope !2
   %c = add i32 %a, %b
   ret i32 %c
 }
diff --git a/test/Transforms/GVN/non-local-offset.ll b/test/Transforms/GVN/non-local-offset.ll
index 0b9edcb8e430..2373ef5888b9 100644
--- a/test/Transforms/GVN/non-local-offset.ll
+++ b/test/Transforms/GVN/non-local-offset.ll
@@ -13,12 +13,12 @@ target datalayout = "e-p:64:64:64"
 define void @yes(i1 %c, i32* %p, i32* %q) nounwind {
 entry:
   store i32 0, i32* %p
-  %p1 = getelementptr inbounds i32* %p, i64 1
+  %p1 = getelementptr inbounds i32, i32* %p, i64 1
   store i32 1, i32* %p1
   br i1 %c, label %if.else, label %if.then
 
 if.then:
-  %t = load i32* %p
+  %t = load i32, i32* %p
   store i32 %t, i32* %q
   ret void
 
@@ -35,25 +35,25 @@ if.else:
 ; CHECK-NEXT: store i32 0, i32* %q
 ; CHECK-NEXT: ret void
 ; CHECK: if.else:
-; CHECK: load i64* %pc
+; CHECK: load i64, i64* %pc
 ; CHECK: store i64
 
 define void @watch_out_for_size_change(i1 %c, i32* %p, i32* %q) nounwind {
 entry:
   store i32 0, i32* %p
-  %p1 = getelementptr inbounds i32* %p, i64 1
+  %p1 = getelementptr inbounds i32, i32* %p, i64 1
   store i32 1, i32* %p1
   br i1 %c, label %if.else, label %if.then
 
 if.then:
-  %t = load i32* %p
+  %t = load i32, i32* %p
   store i32 %t, i32* %q
   ret void
 
 if.else:
   %pc = bitcast i32* %p to i64*
   %qc = bitcast i32* %q to i64*
-  %t64 = load i64* %pc
+  %t64 = load i64, i64* %pc
   store i64 %t64, i64* %qc
   ret void
 }
diff --git a/test/Transforms/GVN/nonescaping-malloc.ll b/test/Transforms/GVN/nonescaping-malloc.ll
index c2eeed56ffc1..f83b317fade4 100644
--- a/test/Transforms/GVN/nonescaping-malloc.ll
+++ b/test/Transforms/GVN/nonescaping-malloc.ll
@@ -22,10 +22,10 @@ target triple = "x86_64-apple-darwin10.0"
 define %"struct.llvm::StringMapEntry<void*>"* @_Z3fooRN4llvm9StringMapIPvNS_15MallocAllocatorEEEPKc(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %X, i8* %P) ssp {
 entry:
   %tmp = alloca %"struct.llvm::StringRef", align 8
-  %tmp.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 0
+  %tmp.i = getelementptr inbounds %"struct.llvm::StringRef", %"struct.llvm::StringRef"* %tmp, i64 0, i32 0
   store i8* %P, i8** %tmp.i, align 8
   %tmp1.i = call i64 @strlen(i8* %P) nounwind readonly
-  %tmp2.i = getelementptr inbounds %"struct.llvm::StringRef"* %tmp, i64 0, i32 1
+  %tmp2.i = getelementptr inbounds %"struct.llvm::StringRef", %"struct.llvm::StringRef"* %tmp, i64 0, i32 1
   store i64 %tmp1.i, i64* %tmp2.i, align 8
   %tmp1 = call %"struct.llvm::StringMapEntry<void*>"* @_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueERKNS_9StringRefE(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %X, %"struct.llvm::StringRef"* %tmp) ssp
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp1
@@ -40,16 +40,16 @@ declare i32 @_ZN4llvm13StringMapImpl15LookupBucketForENS_9StringRefE(%"struct.ll
 define linkonce_odr %"struct.llvm::StringMapEntry<void*>"* @_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueERKNS_9StringRefE(%"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, %"struct.llvm::StringRef"* nocapture %Key) ssp align 2 {
 entry:
   %elt = bitcast %"struct.llvm::StringRef"* %Key to i64*
-  %val = load i64* %elt
-  %tmp = getelementptr inbounds %"struct.llvm::StringRef"* %Key, i64 0, i32 1
-  %val2 = load i64* %tmp
-  %tmp2.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0
+  %val = load i64, i64* %elt
+  %tmp = getelementptr inbounds %"struct.llvm::StringRef", %"struct.llvm::StringRef"* %Key, i64 0, i32 1
+  %val2 = load i64, i64* %tmp
+  %tmp2.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>", %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0
   %tmp3.i = tail call i32 @_ZN4llvm13StringMapImpl15LookupBucketForENS_9StringRefE(%"struct.llvm::StringMapImpl"* %tmp2.i, i64 %val, i64 %val2)
-  %tmp4.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 0
-  %tmp5.i = load %"struct.llvm::StringMapImpl::ItemBucket"** %tmp4.i, align 8
+  %tmp4.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>", %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 0
+  %tmp5.i = load %"struct.llvm::StringMapImpl::ItemBucket"*, %"struct.llvm::StringMapImpl::ItemBucket"** %tmp4.i, align 8
   %tmp6.i = zext i32 %tmp3.i to i64
-  %tmp7.i = getelementptr inbounds %"struct.llvm::StringMapImpl::ItemBucket"* %tmp5.i, i64 %tmp6.i, i32 1
-  %tmp8.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
+  %tmp7.i = getelementptr inbounds %"struct.llvm::StringMapImpl::ItemBucket", %"struct.llvm::StringMapImpl::ItemBucket"* %tmp5.i, i64 %tmp6.i, i32 1
+  %tmp8.i = load %"struct.llvm::StringMapEntryBase"*, %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
   %tmp9.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, null
   %tmp13.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp8.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*)
   %or.cond.i = or i1 %tmp9.i, %tmp13.i
@@ -57,7 +57,7 @@ entry:
 
 bb4.i:                                            ; preds = %entry
   %tmp41.i = inttoptr i64 %val to i8*
-  %tmp4.i35.i = getelementptr inbounds i8* %tmp41.i, i64 %val2
+  %tmp4.i35.i = getelementptr inbounds i8, i8* %tmp41.i, i64 %val2
   %tmp.i.i = ptrtoint i8* %tmp4.i35.i to i64
   %tmp1.i.i = trunc i64 %tmp.i.i to i32
   %tmp3.i.i = trunc i64 %val to i32
@@ -72,22 +72,22 @@ bb4.i:                                            ; preds = %entry
 bb.i.i:                                           ; preds = %bb4.i
   %tmp.i.i.i.i = bitcast i8* %tmp.i20.i.i to i32*
   store i32 %tmp4.i.i, i32* %tmp.i.i.i.i, align 4
-  %tmp1.i19.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8
+  %tmp1.i19.i.i = getelementptr inbounds i8, i8* %tmp.i20.i.i, i64 8
   %0 = bitcast i8* %tmp1.i19.i.i to i8**
   store i8* null, i8** %0, align 8
   br label %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i
 
 _ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i: ; preds = %bb.i.i, %bb4.i
-  %tmp.i18.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 16
+  %tmp.i18.i.i = getelementptr inbounds i8, i8* %tmp.i20.i.i, i64 16
   %tmp15.i.i = zext i32 %tmp4.i.i to i64
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp.i18.i.i, i8* %tmp41.i, i64 %tmp15.i.i, i32 1, i1 false)
   %tmp.i18.sum.i.i = add i64 %tmp15.i.i, 16
-  %tmp17.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 %tmp.i18.sum.i.i
+  %tmp17.i.i = getelementptr inbounds i8, i8* %tmp.i20.i.i, i64 %tmp.i18.sum.i.i
   store i8 0, i8* %tmp17.i.i, align 1
-  %tmp.i.i.i = getelementptr inbounds i8* %tmp.i20.i.i, i64 8
+  %tmp.i.i.i = getelementptr inbounds i8, i8* %tmp.i20.i.i, i64 8
   %1 = bitcast i8* %tmp.i.i.i to i8**
   store i8* null, i8** %1, align 8
-  %tmp22.i = load %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
+  %tmp22.i = load %"struct.llvm::StringMapEntryBase"*, %"struct.llvm::StringMapEntryBase"** %tmp7.i, align 8
   %tmp24.i = icmp eq %"struct.llvm::StringMapEntryBase"* %tmp22.i, inttoptr (i64 -1 to %"struct.llvm::StringMapEntryBase"*)
   br i1 %tmp24.i, label %bb9.i, label %_ZN4llvm9StringMapIPvNS_15MallocAllocatorEE16GetOrCreateValueIS1_EERNS_14StringMapEntryIS1_EENS_9StringRefET_.exit
 
@@ -96,8 +96,8 @@ bb6.i:                                            ; preds = %entry
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp16.i
 
 bb9.i:                                            ; preds = %_ZN4llvm14StringMapEntryIPvE6CreateINS_15MallocAllocatorES1_EEPS2_PKcS7_RT_T0_.exit.i
-  %tmp25.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 3
-  %tmp26.i = load i32* %tmp25.i, align 8
+  %tmp25.i = getelementptr inbounds %"struct.llvm::StringMap<void*,llvm::MallocAllocator>", %"struct.llvm::StringMap<void*,llvm::MallocAllocator>"* %this, i64 0, i32 0, i32 3
+  %tmp26.i = load i32, i32* %tmp25.i, align 8
   %tmp27.i = add i32 %tmp26.i, -1
   store i32 %tmp27.i, i32* %tmp25.i, align 8
   ret %"struct.llvm::StringMapEntry<void*>"* %tmp10.i.i
diff --git a/test/Transforms/GVN/null-aliases-nothing.ll b/test/Transforms/GVN/null-aliases-nothing.ll
index 37bf09d7f3ff..0b7c5eb5bf95 100644
--- a/test/Transforms/GVN/null-aliases-nothing.ll
+++ b/test/Transforms/GVN/null-aliases-nothing.ll
@@ -4,12 +4,12 @@
 declare void @test1f(i8*)
 
 define void @test1(%t* noalias %stuff ) {
-    %p = getelementptr inbounds %t* %stuff, i32 0, i32 0
-    %before = load i32* %p
+    %p = getelementptr inbounds %t, %t* %stuff, i32 0, i32 0
+    %before = load i32, i32* %p
 
     call void @test1f(i8* null)
 
-    %after = load i32* %p ; <--- This should be a dead load
+    %after = load i32, i32* %p ; <--- This should be a dead load
     %sum = add i32 %before, %after
 
     store i32 %sum, i32* %p
diff --git a/test/Transforms/GVN/phi-translate-partial-alias.ll b/test/Transforms/GVN/phi-translate-partial-alias.ll
index 47bec41fb163..f1cf53e45dec 100644
--- a/test/Transforms/GVN/phi-translate-partial-alias.ll
+++ b/test/Transforms/GVN/phi-translate-partial-alias.ll
@@ -8,19 +8,19 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 ; CHECK: define void @test0(i8* %begin)
 ; CHECK: loop:
-; CHECK:   %l0 = load i8* %phi
+; CHECK:   %l0 = load i8, i8* %phi
 ; CHECK:   call void @bar(i8 %l0)
-; CHECK:   %l1 = load i8* %phi
+; CHECK:   %l1 = load i8, i8* %phi
 define void @test0(i8* %begin) {
 entry:
   br label %loop
 
 loop:
   %phi = phi i8* [ %begin, %entry ], [ %next, %loop ]
-  %l0 = load i8* %phi
+  %l0 = load i8, i8* %phi
   call void @bar(i8 %l0)
-  %l1 = load i8* %phi
-  %next = getelementptr inbounds i8* %phi, i8 %l1
+  %l1 = load i8, i8* %phi
+  %next = getelementptr inbounds i8, i8* %phi, i8 %l1
   br label %loop
 }
 
diff --git a/test/Transforms/GVN/phi-translate.ll b/test/Transforms/GVN/phi-translate.ll
index 50d6178c5cc2..6068b05aadf5 100644
--- a/test/Transforms/GVN/phi-translate.ll
+++ b/test/Transforms/GVN/phi-translate.ll
@@ -4,7 +4,7 @@ target datalayout = "e-p:64:64:64"
 
 ; CHECK-LABEL: @foo(
 ; CHECK: entry.end_crit_edge:
-; CHECK:   %n.pre = load i32* %q.phi.trans.insert
+; CHECK:   %n.pre = load i32, i32* %q.phi.trans.insert
 ; CHECK: then:
 ; CHECK:   store i32 %z
 ; CHECK: end:
@@ -19,13 +19,13 @@ entry:
 
 then:
   %i = sext i32 %x to i64
-  %p = getelementptr [100 x i32]* @G, i64 0, i64 %i
+  %p = getelementptr [100 x i32], [100 x i32]* @G, i64 0, i64 %i
   store i32 %z, i32* %p
   br label %end
 
 end:
   %j = sext i32 %x to i64
-  %q = getelementptr [100 x i32]* @G, i64 0, i64 %j
-  %n = load i32* %q
+  %q = getelementptr [100 x i32], [100 x i32]* @G, i64 0, i64 %j
+  %n = load i32, i32* %q
   ret i32 %n
 }
diff --git a/test/Transforms/GVN/pr10820.ll b/test/Transforms/GVN/pr10820.ll
index 12c1e70fc387..c6a9a931df02 100644
--- a/test/Transforms/GVN/pr10820.ll
+++ b/test/Transforms/GVN/pr10820.ll
@@ -11,7 +11,7 @@ entry:
 ; CHECK: store i32
   store i32 402662078, i32* bitcast (i31* @g to i32*), align 8
 ; CHECK-NOT: load i31
-  %0 = load i31* @g, align 8
+  %0 = load i31, i31* @g, align 8
 ; CHECK: store i31
   store i31 %0, i31* undef, align 1
   unreachable
diff --git a/test/Transforms/GVN/pr14166.ll b/test/Transforms/GVN/pr14166.ll
index 4d6820506879..eafe418dbdc6 100644
--- a/test/Transforms/GVN/pr14166.ll
+++ b/test/Transforms/GVN/pr14166.ll
@@ -4,16 +4,16 @@ target triple = "i386-pc-linux-gnu"
 define <2 x i32> @test1() {
   %v1 = alloca <2 x i32>
   call void @anything(<2 x i32>* %v1)
-  %v2 = load <2 x i32>* %v1
+  %v2 = load <2 x i32>, <2 x i32>* %v1
   %v3 = inttoptr <2 x i32> %v2 to <2 x i8*>
   %v4 = bitcast <2 x i32>* %v1 to <2 x i8*>*
   store <2 x i8*> %v3, <2 x i8*>* %v4
-  %v5 = load <2 x i32>* %v1
+  %v5 = load <2 x i32>, <2 x i32>* %v1
   ret <2 x i32> %v5
 ; CHECK-LABEL: @test1(
 ; CHECK: %v1 = alloca <2 x i32>
 ; CHECK: call void @anything(<2 x i32>* %v1)
-; CHECK: %v2 = load <2 x i32>* %v1
+; CHECK: %v2 = load <2 x i32>, <2 x i32>* %v1
 ; CHECK: %v3 = inttoptr <2 x i32> %v2 to <2 x i8*>
 ; CHECK: %v4 = bitcast <2 x i32>* %v1 to <2 x i8*>*
 ; CHECK: store <2 x i8*> %v3, <2 x i8*>* %v4
diff --git a/test/Transforms/GVN/pr17732.ll b/test/Transforms/GVN/pr17732.ll
index 606a195b8538..6c40ccf45019 100644
--- a/test/Transforms/GVN/pr17732.ll
+++ b/test/Transforms/GVN/pr17732.ll
@@ -14,11 +14,11 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define i32 @main() {
 entry:
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* getelementptr inbounds (%struct.with_array* @array_with_zeroinit, i64 0, i32 0, i64 0), i8* getelementptr inbounds ({ [2 x i8], i32, i8, [3 x i8] }* @main.obj_with_array, i64 0, i32 0, i64 0), i64 12, i32 4, i1 false)
-  %0 = load i8* getelementptr inbounds (%struct.with_array* @array_with_zeroinit, i64 0, i32 2), align 4
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* getelementptr inbounds (%struct.with_array, %struct.with_array* @array_with_zeroinit, i64 0, i32 0, i64 0), i8* getelementptr inbounds ({ [2 x i8], i32, i8, [3 x i8] }, { [2 x i8], i32, i8, [3 x i8] }* @main.obj_with_array, i64 0, i32 0, i64 0), i64 12, i32 4, i1 false)
+  %0 = load i8, i8* getelementptr inbounds (%struct.with_array, %struct.with_array* @array_with_zeroinit, i64 0, i32 2), align 4
 
-  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* getelementptr inbounds (%struct.with_vector* @vector_with_zeroinit, i64 0, i32 0, i64 0), i8* getelementptr inbounds ({ <2 x i8>, i32, i8, [3 x i8] }* @main.obj_with_vector, i64 0, i32 0, i64 0), i64 12, i32 4, i1 false)
-  %1 = load i8* getelementptr inbounds (%struct.with_vector* @vector_with_zeroinit, i64 0, i32 2), align 4
+  tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* getelementptr inbounds (%struct.with_vector, %struct.with_vector* @vector_with_zeroinit, i64 0, i32 0, i64 0), i8* getelementptr inbounds ({ <2 x i8>, i32, i8, [3 x i8] }, { <2 x i8>, i32, i8, [3 x i8] }* @main.obj_with_vector, i64 0, i32 0, i64 0), i64 12, i32 4, i1 false)
+  %1 = load i8, i8* getelementptr inbounds (%struct.with_vector, %struct.with_vector* @vector_with_zeroinit, i64 0, i32 2), align 4
   %conv0 = sext i8 %0 to i32
   %conv1 = sext i8 %1 to i32
   %and = and i32 %conv0, %conv1
diff --git a/test/Transforms/GVN/pr17852.ll b/test/Transforms/GVN/pr17852.ll
index e95ff7f5c106..9a8a709ba6a5 100644
--- a/test/Transforms/GVN/pr17852.ll
+++ b/test/Transforms/GVN/pr17852.ll
@@ -5,60 +5,60 @@ define void @fn1(%struct.S0* byval align 8 %p1) {
   br label %for.cond
 for.cond:                                         ; preds = %1, %0
   br label %for.end
-  %f2 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 2
-  %f9 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 7
+  %f2 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 2
+  %f9 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 7
   br label %for.cond
 for.end:                                          ; preds = %for.cond
   br i1 true, label %if.else, label %if.then
 if.then:                                          ; preds = %for.end
-  %f22 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 2
-  %f7 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 5
-  %tmp7 = load i32* %f7, align 8
+  %f22 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 2
+  %f7 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 5
+  %tmp7 = load i32, i32* %f7, align 8
   br label %if.end40
 if.else:                                          ; preds = %for.end
   br i1 false, label %for.cond18, label %if.then6
 if.then6:                                         ; preds = %if.else
-  %f3 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 2
+  %f3 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 2
   %tmp10 = bitcast %struct.S0* %p1 to i16*
-  %f5 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 3
+  %f5 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 3
   %tmp11 = bitcast [2 x i8]* %f5 to i16*
-  %bf.load13 = load i16* %tmp11, align 8
+  %bf.load13 = load i16, i16* %tmp11, align 8
   br label %if.end36
 for.cond18:                                       ; preds = %if.else
   call void @fn4()
   br i1 true, label %if.end, label %if.end36
 if.end:                                           ; preds = %for.cond18
-  %f321 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 2
-  %f925 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 7
-  %f526 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 3
+  %f321 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 2
+  %f925 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 7
+  %f526 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 3
   %tmp15 = bitcast [2 x i8]* %f526 to i16*
-  %bf.load27 = load i16* %tmp15, align 8
+  %bf.load27 = load i16, i16* %tmp15, align 8
   %tmp16 = bitcast %struct.S0* %p1 to i16*
   br label %if.end36
 if.end36:                                         ; preds = %if.end, %for.cond18, %if.then6
-  %f537 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 3
+  %f537 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 3
   %tmp17 = bitcast [2 x i8]* %f537 to i16*
-  %bf.load38 = load i16* %tmp17, align 8
+  %bf.load38 = load i16, i16* %tmp17, align 8
   %bf.clear39 = and i16 %bf.load38, -16384
   br label %if.end40
 if.end40:                                         ; preds = %if.end36, %if.then
-  %f6 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 4
-  %tmp18 = load i32* %f6, align 4
+  %f6 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 4
+  %tmp18 = load i32, i32* %f6, align 4
   call void @fn2(i32 %tmp18)
-  %f8 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 6
-  %tmp19 = load i32* %f8, align 4
+  %f8 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 6
+  %tmp19 = load i32, i32* %f8, align 4
   %tobool41 = icmp eq i32 %tmp19, 0
   br i1 true, label %if.end50, label %if.then42
 if.then42:                                        ; preds = %if.end40
   %tmp20 = bitcast %struct.S0* %p1 to i16*
-  %f547 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 3
+  %f547 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 3
   %tmp21 = bitcast [2 x i8]* %f547 to i16*
-  %bf.load48 = load i16* %tmp21, align 8
+  %bf.load48 = load i16, i16* %tmp21, align 8
   br label %if.end50
 if.end50:                                         ; preds = %if.then42, %if.end40
-  %f551 = getelementptr inbounds %struct.S0* %p1, i64 0, i32 3
+  %f551 = getelementptr inbounds %struct.S0, %struct.S0* %p1, i64 0, i32 3
   %tmp22 = bitcast [2 x i8]* %f551 to i16*
-  %bf.load52 = load i16* %tmp22, align 8
+  %bf.load52 = load i16, i16* %tmp22, align 8
   %bf.clear53 = and i16 %bf.load52, -16384
   ret void
 }
diff --git a/test/Transforms/GVN/pre-basic-add.ll b/test/Transforms/GVN/pre-basic-add.ll
index 4bde05c3387f..fa4e2e3abded 100644
--- a/test/Transforms/GVN/pre-basic-add.ll
+++ b/test/Transforms/GVN/pre-basic-add.ll
@@ -5,8 +5,8 @@
 
 define i32 @test() nounwind {
 entry:
-	%0 = load i32* @H, align 4		; <i32> [#uses=2]
-	%1 = call i32 (...)* @foo() nounwind		; <i32> [#uses=1]
+	%0 = load i32, i32* @H, align 4		; <i32> [#uses=2]
+	%1 = call i32 (...) @foo() nounwind		; <i32> [#uses=1]
 	%2 = icmp ne i32 %1, 0		; <i1> [#uses=1]
 	br i1 %2, label %bb, label %bb1
 
diff --git a/test/Transforms/GVN/pre-compare.ll b/test/Transforms/GVN/pre-compare.ll
index 18d0c2e1085c..a77684a9831e 100644
--- a/test/Transforms/GVN/pre-compare.ll
+++ b/test/Transforms/GVN/pre-compare.ll
@@ -43,7 +43,7 @@ entry:
 
 if.then:                                          ; preds = %entry
   %cmp1 = icmp eq i32 %x, 2
-  %cond = select i1 %cmp1, i8* getelementptr inbounds ([2 x i8]* @.str, i64 0, i64 0), i8* getelementptr inbounds ([2 x i8]* @.str1, i64 0, i64 0)
+  %cond = select i1 %cmp1, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str, i64 0, i64 0), i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str1, i64 0, i64 0)
   %call = tail call i32 @puts(i8* %cond) nounwind
   br label %for.cond.preheader
 
@@ -52,11 +52,11 @@ for.cond.preheader:                               ; preds = %entry, %if.then
   br label %for.cond
 
 for.cond:                                         ; preds = %for.cond.backedge, %for.cond.preheader
-  %call2 = tail call i32 @puts(i8* getelementptr inbounds ([7 x i8]* @.str2, i64 0, i64 0)) nounwind
+  %call2 = tail call i32 @puts(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str2, i64 0, i64 0)) nounwind
   br i1 %cmp3, label %for.cond.backedge, label %if.end5
 
 if.end5:                                          ; preds = %for.cond
-  %call6 = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([12 x i8]* @.str3, i64 0, i64 0), i32 %x) nounwind
+  %call6 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str3, i64 0, i64 0), i32 %x) nounwind
   br label %for.cond.backedge
 
 for.cond.backedge:                                ; preds = %if.end5, %for.cond
diff --git a/test/Transforms/GVN/pre-gep-load.ll b/test/Transforms/GVN/pre-gep-load.ll
index 3ee3a37b81ad..291af359a7a1 100644
--- a/test/Transforms/GVN/pre-gep-load.ll
+++ b/test/Transforms/GVN/pre-gep-load.ll
@@ -13,10 +13,10 @@ entry:
 
 sw.bb:                                            ; preds = %entry, %entry
   %idxprom = sext i32 %i to i64
-  %arrayidx = getelementptr inbounds double** %p, i64 0
-  %0 = load double** %arrayidx, align 8
-  %arrayidx1 = getelementptr inbounds double* %0, i64 %idxprom
-  %1 = load double* %arrayidx1, align 8
+  %arrayidx = getelementptr inbounds double*, double** %p, i64 0
+  %0 = load double*, double** %arrayidx, align 8
+  %arrayidx1 = getelementptr inbounds double, double* %0, i64 %idxprom
+  %1 = load double, double* %arrayidx1, align 8
   %sub = fsub double %1, 1.000000e+00
   %cmp = fcmp olt double %sub, 0.000000e+00
   br i1 %cmp, label %if.then, label %if.end
@@ -29,10 +29,10 @@ if.end:                                           ; preds = %sw.bb
 
 sw.bb2:                                           ; preds = %if.end, %entry
   %idxprom3 = sext i32 %i to i64
-  %arrayidx4 = getelementptr inbounds double** %p, i64 0
-  %2 = load double** %arrayidx4, align 8
-  %arrayidx5 = getelementptr inbounds double* %2, i64 %idxprom3
-  %3 = load double* %arrayidx5, align 8
+  %arrayidx4 = getelementptr inbounds double*, double** %p, i64 0
+  %2 = load double*, double** %arrayidx4, align 8
+  %arrayidx5 = getelementptr inbounds double, double* %2, i64 %idxprom3
+  %3 = load double, double* %arrayidx5, align 8
 ; CHECK: sw.bb2:
 ; CHECK-NEXT-NOT: sext
 ; CHECK-NEXT: phi double [
diff --git a/test/Transforms/GVN/pre-load.ll b/test/Transforms/GVN/pre-load.ll
index 9842886fe3ab..24221d540f22 100644
--- a/test/Transforms/GVN/pre-load.ll
+++ b/test/Transforms/GVN/pre-load.ll
@@ -9,14 +9,14 @@ block1:
 block2:
  br label %block4
 ; CHECK: block2:
-; CHECK-NEXT: load i32* %p
+; CHECK-NEXT: load i32, i32* %p
 
 block3:
   store i32 0, i32* %p
   br label %block4
 
 block4:
-  %PRE = load i32* %p
+  %PRE = load i32, i32* %p
   ret i32 %PRE
 ; CHECK: block4:
 ; CHECK-NEXT: phi i32
@@ -32,7 +32,7 @@ block1:
 block2:
  br label %block4
 ; CHECK: block2:
-; CHECK-NEXT: load i32* %q
+; CHECK-NEXT: load i32, i32* %q
 
 block3:
   store i32 0, i32* %p
@@ -40,7 +40,7 @@ block3:
 
 block4:
   %P2 = phi i32* [%p, %block3], [%q, %block2]
-  %PRE = load i32* %P2
+  %PRE = load i32, i32* %P2
   ret i32 %PRE
 ; CHECK: block4:
 ; CHECK-NEXT: phi i32 [
@@ -52,24 +52,24 @@ block4:
 define i32 @test3(i32* %p, i32* %q, i32** %Hack, i1 %C) {
 ; CHECK-LABEL: @test3(
 block1:
-  %B = getelementptr i32* %q, i32 1
+  %B = getelementptr i32, i32* %q, i32 1
   store i32* %B, i32** %Hack
 	br i1 %C, label %block2, label %block3
 
 block2:
  br label %block4
 ; CHECK: block2:
-; CHECK-NEXT: load i32* %B
+; CHECK-NEXT: load i32, i32* %B
 
 block3:
-  %A = getelementptr i32* %p, i32 1
+  %A = getelementptr i32, i32* %p, i32 1
   store i32 0, i32* %A
   br label %block4
 
 block4:
   %P2 = phi i32* [%p, %block3], [%q, %block2]
-  %P3 = getelementptr i32* %P2, i32 1
-  %PRE = load i32* %P3
+  %P3 = getelementptr i32, i32* %P2, i32 1
+  %PRE = load i32, i32* %P3
   ret i32 %PRE
 ; CHECK: block4:
 ; CHECK-NEXT: phi i32 [
@@ -87,21 +87,21 @@ block1:
 block2:
  br label %block4
 ; CHECK: block2:
-; CHECK:   load i32*
+; CHECK:   load i32, i32*
 ; CHECK:   br label %block4
 
 block3:
-  %B = getelementptr i32* %q, i32 1
+  %B = getelementptr i32, i32* %q, i32 1
   store i32* %B, i32** %Hack
 
-  %A = getelementptr i32* %p, i32 1
+  %A = getelementptr i32, i32* %p, i32 1
   store i32 0, i32* %A
   br label %block4
 
 block4:
   %P2 = phi i32* [%p, %block3], [%q, %block2]
-  %P3 = getelementptr i32* %P2, i32 1
-  %PRE = load i32* %P3
+  %P3 = getelementptr i32, i32* %P2, i32 1
+  %PRE = load i32, i32* %P3
   ret i32 %PRE
 ; CHECK: block4:
 ; CHECK-NEXT: phi i32 [
@@ -127,16 +127,16 @@ bb.nph:
   br label %bb
 
 ; CHECK: bb.nph:
-; CHECK: load double*
+; CHECK: load double, double*
 ; CHECK: br label %bb
 
 bb:             
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp6, %bb ]
   %tmp6 = add i64 %indvar, 1                    
-  %scevgep = getelementptr double* %G, i64 %tmp6
-  %scevgep7 = getelementptr double* %G, i64 %indvar
-  %2 = load double* %scevgep7, align 8
-  %3 = load double* %scevgep, align 8 
+  %scevgep = getelementptr double, double* %G, i64 %tmp6
+  %scevgep7 = getelementptr double, double* %G, i64 %indvar
+  %2 = load double, double* %scevgep7, align 8
+  %3 = load double, double* %scevgep, align 8 
   %4 = fadd double %2, %3             
   store double %4, double* %scevgep7, align 8
   %exitcond = icmp eq i64 %tmp6, %tmp 
@@ -144,8 +144,8 @@ bb:
 
 ; Should only be one load in the loop.
 ; CHECK: bb:
-; CHECK: load double*
-; CHECK-NOT: load double*
+; CHECK: load double, double*
+; CHECK-NOT: load double, double*
 ; CHECK: br i1 %exitcond
 
 return:                               
@@ -170,16 +170,16 @@ bb.nph:
   br label %bb
 
 ; CHECK: bb.nph:
-; CHECK: load double*
+; CHECK: load double, double*
 ; CHECK: br label %bb
 
 bb:             
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp6, %bb ]
   %tmp6 = add i64 %indvar, 1                    
-  %scevgep = getelementptr double* %G, i64 %tmp6
-  %scevgep7 = getelementptr double* %G, i64 %indvar
-  %2 = load double* %scevgep7, align 8
-  %3 = load double* %scevgep, align 8 
+  %scevgep = getelementptr double, double* %G, i64 %tmp6
+  %scevgep7 = getelementptr double, double* %G, i64 %indvar
+  %2 = load double, double* %scevgep7, align 8
+  %3 = load double, double* %scevgep, align 8 
   %4 = fadd double %2, %3             
   store double %4, double* %scevgep, align 8
   %exitcond = icmp eq i64 %tmp6, %tmp 
@@ -187,8 +187,8 @@ bb:
 
 ; Should only be one load in the loop.
 ; CHECK: bb:
-; CHECK: load double*
-; CHECK-NOT: load double*
+; CHECK: load double, double*
+; CHECK-NOT: load double, double*
 ; CHECK: br i1 %exitcond
 
 return:                               
@@ -205,7 +205,7 @@ return:
 ; This requires phi translation of the adds.
 define void @test7(i32 %N, double* nocapture %G) nounwind ssp {
 entry:
-  %0 = getelementptr inbounds double* %G, i64 1   
+  %0 = getelementptr inbounds double, double* %G, i64 1   
   store double 1.000000e+00, double* %0, align 8
   %1 = add i32 %N, -1                             
   %2 = icmp sgt i32 %1, 1                         
@@ -219,11 +219,11 @@ bb.nph:
 bb:                                               
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp9, %bb ] 
   %tmp8 = add i64 %indvar, 2                      
-  %scevgep = getelementptr double* %G, i64 %tmp8  
+  %scevgep = getelementptr double, double* %G, i64 %tmp8  
   %tmp9 = add i64 %indvar, 1                      
-  %scevgep10 = getelementptr double* %G, i64 %tmp9 
-  %3 = load double* %scevgep10, align 8           
-  %4 = load double* %scevgep, align 8             
+  %scevgep10 = getelementptr double, double* %G, i64 %tmp9 
+  %3 = load double, double* %scevgep10, align 8           
+  %4 = load double, double* %scevgep, align 8             
   %5 = fadd double %3, %4                         
   store double %5, double* %scevgep, align 8
   %exitcond = icmp eq i64 %tmp9, %tmp7            
@@ -231,8 +231,8 @@ bb:
 
 ; Should only be one load in the loop.
 ; CHECK: bb:
-; CHECK: load double*
-; CHECK-NOT: load double*
+; CHECK: load double, double*
+; CHECK-NOT: load double, double*
 ; CHECK: br i1 %exitcond
 
 return:                                           
@@ -249,18 +249,18 @@ block1:
 block2:
  br label %block4
 ; CHECK: block2:
-; CHECK:   load i32*
+; CHECK:   load i32, i32*
 ; CHECK:   br label %block4
 
 block3:
-  %A = getelementptr i32* %p, i32 1
+  %A = getelementptr i32, i32* %p, i32 1
   store i32 0, i32* %A
   br label %block4
 
 block4:
   %P2 = phi i32* [%p, %block3], [%q, %block2]
-  %P3 = getelementptr i32* %P2, i32 1
-  %PRE = load i32* %P3
+  %P3 = getelementptr i32, i32* %P2, i32 1
+  %PRE = load i32, i32* %P3
   ret i32 %PRE
 ; CHECK: block4:
 ; CHECK-NEXT: phi i32 [
@@ -288,17 +288,17 @@ bb.nph:
   br label %bb
 
 ; CHECK: bb.nph:
-; CHECK:   load double*
+; CHECK:   load double, double*
 ; CHECK:   br label %bb
 
 bb:                                               
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp9, %bb ] 
   %tmp8 = add i64 %indvar, 2                      
-  %scevgep = getelementptr double* %G, i64 %tmp8  
+  %scevgep = getelementptr double, double* %G, i64 %tmp8  
   %tmp9 = add i64 %indvar, 1                      
-  %scevgep10 = getelementptr double* %G, i64 %tmp9 
-  %3 = load double* %scevgep10, align 8           
-  %4 = load double* %scevgep, align 8             
+  %scevgep10 = getelementptr double, double* %G, i64 %tmp9 
+  %3 = load double, double* %scevgep10, align 8           
+  %4 = load double, double* %scevgep, align 8             
   %5 = fadd double %3, %4                         
   store double %5, double* %scevgep, align 8
   %exitcond = icmp eq i64 %tmp9, %tmp7            
@@ -306,8 +306,8 @@ bb:
 
 ; Should only be one load in the loop.
 ; CHECK: bb:
-; CHECK: load double*
-; CHECK-NOT: load double*
+; CHECK: load double, double*
+; CHECK-NOT: load double, double*
 ; CHECK: br i1 %exitcond
 
 return:                                           
@@ -332,22 +332,22 @@ bb.nph:
   %tmp8 = add i64 %tmp, -1
   br label %bb
 ; CHECK: bb.nph:
-; CHECK:   load double*
-; CHECK:   load double*
+; CHECK:   load double, double*
+; CHECK:   load double, double*
 ; CHECK:   br label %bb
 
 
 bb:
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp11, %bb ]
-  %scevgep = getelementptr double* %G, i64 %indvar
+  %scevgep = getelementptr double, double* %G, i64 %indvar
   %tmp9 = add i64 %indvar, 2
-  %scevgep10 = getelementptr double* %G, i64 %tmp9
+  %scevgep10 = getelementptr double, double* %G, i64 %tmp9
   %tmp11 = add i64 %indvar, 1
-  %scevgep12 = getelementptr double* %G, i64 %tmp11
-  %2 = load double* %scevgep12, align 8
-  %3 = load double* %scevgep10, align 8
+  %scevgep12 = getelementptr double, double* %G, i64 %tmp11
+  %2 = load double, double* %scevgep12, align 8
+  %3 = load double, double* %scevgep10, align 8
   %4 = fadd double %2, %3
-  %5 = load double* %scevgep, align 8
+  %5 = load double, double* %scevgep, align 8
   %6 = fadd double %4, %5
   store double %6, double* %scevgep12, align 8
   %exitcond = icmp eq i64 %tmp11, %tmp8
@@ -355,8 +355,8 @@ bb:
 
 ; Should only be one load in the loop.
 ; CHECK: bb:
-; CHECK: load double*
-; CHECK-NOT: load double*
+; CHECK: load double, double*
+; CHECK-NOT: load double, double*
 ; CHECK: br i1 %exitcond
 
 return:
@@ -372,7 +372,7 @@ block1:
 block2:
  %cond = icmp sgt i32 %N, 1
  br i1 %cond, label %block4, label %block5
-; CHECK: load i32* %p
+; CHECK: load i32, i32* %p
 ; CHECK-NEXT: br label %block4
 
 block3:
@@ -380,7 +380,7 @@ block3:
   br label %block4
 
 block4:
-  %PRE = load i32* %p
+  %PRE = load i32, i32* %p
   br label %block5
 
 block5:
diff --git a/test/Transforms/GVN/pre-no-cost-phi.ll b/test/Transforms/GVN/pre-no-cost-phi.ll
new file mode 100644
index 000000000000..4c5afa12ddc1
--- /dev/null
+++ b/test/Transforms/GVN/pre-no-cost-phi.ll
@@ -0,0 +1,31 @@
+; RUN: opt < %s -gvn -S | FileCheck %s
+; This testcase tests insertion of no-cost phis.  That is,
+; when the value is already available in every predecessor,
+; and we just need to insert a phi node to merge the available values.
+
+@c = global i32 0, align 4
+@d = global i32 0, align 4
+
+
+define i32 @mai(i32 %foo, i32 %a, i32 %b) {
+  %1 = icmp ne i32 %foo, 0
+  br i1 %1, label %bb1, label %bb2
+
+bb1:
+  %2 = add nsw i32 %a, %b
+  store i32 %2, i32* @c, align 4
+  br label %mergeblock
+
+bb2:
+  %3 = add nsw i32 %a, %b
+  store i32 %3, i32* @d, align 4
+  br label %mergeblock
+
+mergeblock:
+; CHECK: pre-phi = phi i32 [ %3, %bb2 ], [ %2, %bb1 ]
+; CHECK-NEXT: ret i32 %.pre-phi
+  %4 = add nsw i32 %a, %b
+  ret i32 %4
+}
+
+
diff --git a/test/Transforms/GVN/pre-single-pred.ll b/test/Transforms/GVN/pre-single-pred.ll
index f1f5c71a93ab..0df45cf5c1c0 100644
--- a/test/Transforms/GVN/pre-single-pred.ll
+++ b/test/Transforms/GVN/pre-single-pred.ll
@@ -23,9 +23,9 @@ for.cond.for.end_crit_edge:		; preds = %for.cond
 	br label %for.end
 
 ; CHECK: for.body:
-; CHECK-NEXT: %tmp3 = load i32* @p
+; CHECK-NEXT: %tmp3 = load i32, i32* @p
 for.body:		; preds = %for.cond
-	%tmp3 = load i32* @p		; <i32> [#uses=1]
+	%tmp3 = load i32, i32* @p		; <i32> [#uses=1]
 	%dec = add i32 %tmp3, -1		; <i32> [#uses=2]
 	store i32 %dec, i32* @p
 	%cmp6 = icmp slt i32 %dec, 0		; <i1> [#uses=1]
@@ -40,6 +40,6 @@ for.inc:		; preds = %for.body
 	br label %for.cond
 
 for.end:		; preds = %for.body.for.end_crit_edge, %for.cond.for.end_crit_edge
-	%tmp9 = load i32* @p		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* @p		; <i32> [#uses=1]
 	ret i32 %tmp9
 }
diff --git a/test/Transforms/GVN/preserve-tbaa.ll b/test/Transforms/GVN/preserve-tbaa.ll
index 587d463eef27..19467eeff404 100644
--- a/test/Transforms/GVN/preserve-tbaa.ll
+++ b/test/Transforms/GVN/preserve-tbaa.ll
@@ -5,7 +5,7 @@ target datalayout = "e-p:64:64:64"
 ; GVN should preserve the TBAA tag on loads when doing PRE.
 
 ; CHECK-LABEL: @test(
-; CHECK: %tmp33.pre = load i16* %P, align 2, !tbaa !0
+; CHECK: %tmp33.pre = load i16, i16* %P, align 2, !tbaa !0
 ; CHECK: br label %for.body
 define void @test(i16 *%P, i16* %Q) nounwind {
 entry:
@@ -15,7 +15,7 @@ bb.nph:                                           ; preds = %entry
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %bb.nph
-  %tmp33 = load i16* %P, align 2, !tbaa !0
+  %tmp33 = load i16, i16* %P, align 2, !tbaa !0
   store i16 %tmp33, i16* %Q
 
   store i16 0, i16* %P, align 2, !tbaa !0
diff --git a/test/Transforms/GVN/range.ll b/test/Transforms/GVN/range.ll
index 37202321a6dc..297c6aac88dd 100644
--- a/test/Transforms/GVN/range.ll
+++ b/test/Transforms/GVN/range.ll
@@ -2,82 +2,82 @@
 
 define i32 @test1(i32* %p) {
 ; CHECK: @test1(i32* %p)
-; CHECK: %a = load i32* %p, !range !0
+; CHECK: %a = load i32, i32* %p, !range !0
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !0
-  %b = load i32* %p, !range !0
+  %a = load i32, i32* %p, !range !0
+  %b = load i32, i32* %p, !range !0
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test2(i32* %p) {
 ; CHECK: @test2(i32* %p)
-; CHECK: %a = load i32* %p
+; CHECK: %a = load i32, i32* %p
 ; CHECK-NOT: range
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !0
-  %b = load i32* %p
+  %a = load i32, i32* %p, !range !0
+  %b = load i32, i32* %p
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test3(i32* %p) {
 ; CHECK: @test3(i32* %p)
-; CHECK: %a = load i32* %p, !range ![[DISJOINT_RANGE:[0-9]+]]
+; CHECK: %a = load i32, i32* %p, !range ![[DISJOINT_RANGE:[0-9]+]]
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !0
-  %b = load i32* %p, !range !1
+  %a = load i32, i32* %p, !range !0
+  %b = load i32, i32* %p, !range !1
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test4(i32* %p) {
 ; CHECK: @test4(i32* %p)
-; CHECK: %a = load i32* %p, !range ![[MERGED_RANGE:[0-9]+]]
+; CHECK: %a = load i32, i32* %p, !range ![[MERGED_RANGE:[0-9]+]]
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !0
-  %b = load i32* %p, !range !2
+  %a = load i32, i32* %p, !range !0
+  %b = load i32, i32* %p, !range !2
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test5(i32* %p) {
 ; CHECK: @test5(i32* %p)
-; CHECK: %a = load i32* %p, !range ![[MERGED_SIGNED_RANGE:[0-9]+]]
+; CHECK: %a = load i32, i32* %p, !range ![[MERGED_SIGNED_RANGE:[0-9]+]]
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !3
-  %b = load i32* %p, !range !4
+  %a = load i32, i32* %p, !range !3
+  %b = load i32, i32* %p, !range !4
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test6(i32* %p) {
 ; CHECK: @test6(i32* %p)
-; CHECK: %a = load i32* %p, !range ![[MERGED_TEST6:[0-9]+]]
+; CHECK: %a = load i32, i32* %p, !range ![[MERGED_TEST6:[0-9]+]]
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !5
-  %b = load i32* %p, !range !6
+  %a = load i32, i32* %p, !range !5
+  %b = load i32, i32* %p, !range !6
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test7(i32* %p) {
 ; CHECK: @test7(i32* %p)
-; CHECK: %a = load i32* %p, !range ![[MERGED_TEST7:[0-9]+]]
+; CHECK: %a = load i32, i32* %p, !range ![[MERGED_TEST7:[0-9]+]]
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !7
-  %b = load i32* %p, !range !8
+  %a = load i32, i32* %p, !range !7
+  %b = load i32, i32* %p, !range !8
   %c = add i32 %a, %b
   ret i32 %c
 }
 
 define i32 @test8(i32* %p) {
 ; CHECK: @test8(i32* %p)
-; CHECK: %a = load i32* %p
+; CHECK: %a = load i32, i32* %p
 ; CHECK-NOT: range
 ; CHECK: %c = add i32 %a, %a
-  %a = load i32* %p, !range !9
-  %b = load i32* %p, !range !10
+  %a = load i32, i32* %p, !range !9
+  %b = load i32, i32* %p, !range !10
   %c = add i32 %a, %b
   ret i32 %c
 }
diff --git a/test/Transforms/GVN/readattrs.ll b/test/Transforms/GVN/readattrs.ll
index ba624a71f9b6..fb36d07c86e6 100644
--- a/test/Transforms/GVN/readattrs.ll
+++ b/test/Transforms/GVN/readattrs.ll
@@ -9,7 +9,7 @@ define i8 @test() {
   %a = alloca i8
   store i8 1, i8* %a
   call void @use(i8* %a)
-  %b = load i8* %a
+  %b = load i8, i8* %a
   ret i8 %b
 ; CHECK-LABEL: define i8 @test(
 ; CHECK: call void @use(i8* %a)
diff --git a/test/Transforms/GVN/rle-must-alias.ll b/test/Transforms/GVN/rle-must-alias.ll
index e7dc9c423fcb..bf8b139a2e86 100644
--- a/test/Transforms/GVN/rle-must-alias.ll
+++ b/test/Transforms/GVN/rle-must-alias.ll
@@ -11,28 +11,28 @@ target triple = "i386-apple-darwin7"
 
 define i32 @test(i32 %i) nounwind {
 entry:
-	%0 = tail call i32 (...)* @foo() nounwind		; <i32> [#uses=1]
+	%0 = tail call i32 (...) @foo() nounwind		; <i32> [#uses=1]
 	%1 = icmp eq i32 %0, 0		; <i1> [#uses=1]
 	br i1 %1, label %bb1, label %bb
 
 bb:		; preds = %entry
-	%2 = tail call i32 (...)* @bar() nounwind		; <i32> [#uses=0]
-	%3 = getelementptr [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
-	%4 = load i32* %3, align 4		; <i32> [#uses=1]
+	%2 = tail call i32 (...) @bar() nounwind		; <i32> [#uses=0]
+	%3 = getelementptr [100 x i32], [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
+	%4 = load i32, i32* %3, align 4		; <i32> [#uses=1]
 	store i32 %4, i32* @G, align 4
 	br label %bb3
 
 bb1:		; preds = %entry
-	%5 = tail call i32 (...)* @baz() nounwind		; <i32> [#uses=0]
-	%6 = getelementptr [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
-	%7 = load i32* %6, align 4		; <i32> [#uses=2]
+	%5 = tail call i32 (...) @baz() nounwind		; <i32> [#uses=0]
+	%6 = getelementptr [100 x i32], [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
+	%7 = load i32, i32* %6, align 4		; <i32> [#uses=2]
 	store i32 %7, i32* @G, align 4
 	%8 = icmp eq i32 %7, 0		; <i1> [#uses=1]
 	br i1 %8, label %bb3, label %bb4
 
 bb3:		; preds = %bb1, %bb
-	%9 = getelementptr [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
-	%DEAD = load i32* %9, align 4		; <i32> [#uses=1]
+	%9 = getelementptr [100 x i32], [100 x i32]* @H, i32 0, i32 %i		; <i32*> [#uses=1]
+	%DEAD = load i32, i32* %9, align 4		; <i32> [#uses=1]
 	ret i32 %DEAD
 
 bb4:		; preds = %bb1
diff --git a/test/Transforms/GVN/rle-no-phi-translate.ll b/test/Transforms/GVN/rle-no-phi-translate.ll
index 96dbf481899c..c1fd20115379 100644
--- a/test/Transforms/GVN/rle-no-phi-translate.ll
+++ b/test/Transforms/GVN/rle-no-phi-translate.ll
@@ -19,7 +19,7 @@ bb:		; preds = %entry
 
 bb2:		; preds = %bb1, %bb
 	%c_addr.0 = phi i32* [ %b, %entry ], [ %c, %bb ]		; <i32*> [#uses=1]
-	%cv = load i32* %c_addr.0, align 4		; <i32> [#uses=1]
+	%cv = load i32, i32* %c_addr.0, align 4		; <i32> [#uses=1]
 	ret i32 %cv
 ; CHECK: bb2:
 ; CHECK-NOT: load i32
diff --git a/test/Transforms/GVN/rle-nonlocal.ll b/test/Transforms/GVN/rle-nonlocal.ll
index 8229aaa14247..7975462313c0 100644
--- a/test/Transforms/GVN/rle-nonlocal.ll
+++ b/test/Transforms/GVN/rle-nonlocal.ll
@@ -6,20 +6,20 @@ block1:
 	br i1 %cmp , label %block2, label %block3
 
 block2:
- %a = load i32** %p
+ %a = load i32*, i32** %p
  br label %block4
 
 block3:
-  %b = load i32** %p
+  %b = load i32*, i32** %p
   br label %block4
 
 block4:
 ; CHECK-NOT: %existingPHI = phi
 ; CHECK: %DEAD = phi
   %existingPHI = phi i32* [ %a, %block2 ], [ %b, %block3 ] 
-  %DEAD = load i32** %p
-  %c = load i32* %DEAD
-  %d = load i32* %existingPHI
+  %DEAD = load i32*, i32** %p
+  %c = load i32, i32* %DEAD
+  %d = load i32, i32* %existingPHI
   %e = add i32 %c, %d
   ret i32 %e
 }
diff --git a/test/Transforms/GVN/rle-phi-translate.ll b/test/Transforms/GVN/rle-phi-translate.ll
index 1ce7e0b93797..7402e1a1fc00 100644
--- a/test/Transforms/GVN/rle-phi-translate.ll
+++ b/test/Transforms/GVN/rle-phi-translate.ll
@@ -11,13 +11,13 @@ entry:
 	br i1 %t1, label %bb, label %bb1
 
 bb:
-	%t2 = load i32* %c, align 4
+	%t2 = load i32, i32* %c, align 4
 	%t3 = add i32 %t2, 1
 	store i32 %t3, i32* %g, align 4
 	br label %bb2
 
 bb1:		; preds = %entry
-	%t5 = load i32* %b, align 4
+	%t5 = load i32, i32* %b, align 4
 	%t6 = add i32 %t5, 1
 	store i32 %t6, i32* %g, align 4
 	br label %bb2
@@ -25,8 +25,8 @@ bb1:		; preds = %entry
 bb2:		; preds = %bb1, %bb
 	%c_addr.0 = phi i32* [ %g, %bb1 ], [ %c, %bb ]
 	%b_addr.0 = phi i32* [ %b, %bb1 ], [ %g, %bb ]
-	%cv = load i32* %c_addr.0, align 4
-	%bv = load i32* %b_addr.0, align 4
+	%cv = load i32, i32* %c_addr.0, align 4
+	%bv = load i32, i32* %b_addr.0, align 4
 ; CHECK: %bv = phi i32
 ; CHECK: %cv = phi i32
 ; CHECK-NOT: load
@@ -53,7 +53,7 @@ bb1:
 bb2:
   %d = phi i32* [ %c, %bb1 ], [ %b, %bb ]
   %d1 = bitcast i32* %d to i8*
-  %dv = load i8* %d1
+  %dv = load i8, i8* %d1
 ; CHECK: %dv = phi i8 [ 92, %bb1 ], [ 4, %bb ]
 ; CHECK-NOT: load
 ; CHECK: ret i8 %dv
@@ -66,20 +66,20 @@ entry:
   br i1 %cond, label %bb, label %bb1
 
 bb:
-  %b1 = getelementptr i32* %b, i32 17
+  %b1 = getelementptr i32, i32* %b, i32 17
   store i32 4, i32* %b1
   br label %bb2
 
 bb1:
-  %c1 = getelementptr i32* %c, i32 7
+  %c1 = getelementptr i32, i32* %c, i32 7
   store i32 82, i32* %c1
   br label %bb2
 
 bb2:
   %d = phi i32* [ %c, %bb1 ], [ %b, %bb ]
   %i = phi i32 [ 7, %bb1 ], [ 17, %bb ]
-  %d1 = getelementptr i32* %d, i32 %i
-  %dv = load i32* %d1
+  %d1 = getelementptr i32, i32* %d, i32 %i
+  %dv = load i32, i32* %d1
 ; CHECK: %dv = phi i32 [ 82, %bb1 ], [ 4, %bb ]
 ; CHECK-NOT: load
 ; CHECK: ret i32 %dv
@@ -97,15 +97,15 @@ bb:
   br label %bb2
 
 bb1:
-  %c1 = getelementptr i32* %c, i32 7
+  %c1 = getelementptr i32, i32* %c, i32 7
   store i32 82, i32* %c1
   br label %bb2
 
 bb2:
   %d = phi i32* [ %c, %bb1 ], [ %b, %bb ]
   %i = phi i32 [ 7, %bb1 ], [ 0, %bb ]
-  %d1 = getelementptr i32* %d, i32 %i
-  %dv = load i32* %d1
+  %d1 = getelementptr i32, i32* %d, i32 %i
+  %dv = load i32, i32* %d1
 ; CHECK: %dv = phi i32 [ 82, %bb1 ], [ 4, %bb ]
 ; CHECK-NOT: load
 ; CHECK: ret i32 %dv
@@ -127,11 +127,11 @@ bb.nph:
 
 for.body:
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp, %for.body ]
-  %arrayidx6 = getelementptr double* %G, i64 %indvar
+  %arrayidx6 = getelementptr double, double* %G, i64 %indvar
   %tmp = add i64 %indvar, 1
-  %arrayidx = getelementptr double* %G, i64 %tmp
-  %tmp3 = load double* %arrayidx
-  %tmp7 = load double* %arrayidx6
+  %arrayidx = getelementptr double, double* %G, i64 %tmp
+  %tmp3 = load double, double* %arrayidx
+  %tmp7 = load double, double* %arrayidx6
   %add = fadd double %tmp3, %tmp7
   store double %add, double* %arrayidx
   %exitcond = icmp eq i64 %tmp, 999
diff --git a/test/Transforms/GVN/rle-semidominated.ll b/test/Transforms/GVN/rle-semidominated.ll
index 923cd03ecdb7..f80d040c93eb 100644
--- a/test/Transforms/GVN/rle-semidominated.ll
+++ b/test/Transforms/GVN/rle-semidominated.ll
@@ -2,7 +2,7 @@
 
 define i32 @main(i32* %p, i32 %x, i32 %y) {
 block1:
-  %z = load i32* %p
+  %z = load i32, i32* %p
   %cmp = icmp eq i32 %x, %y
 	br i1 %cmp, label %block2, label %block3
 
@@ -15,6 +15,6 @@ block3:
   br label %block4
 
 block4:
-  %DEAD = load i32* %p
+  %DEAD = load i32, i32* %p
   ret i32 %DEAD
 }
diff --git a/test/Transforms/GVN/rle.ll b/test/Transforms/GVN/rle.ll
index 6aac93e75a78..3f42135b4092 100644
--- a/test/Transforms/GVN/rle.ll
+++ b/test/Transforms/GVN/rle.ll
@@ -5,7 +5,7 @@
 define i32 @test0(i32 %V, i32* %P) {
   store i32 %V, i32* %P
 
-  %A = load i32* %P
+  %A = load i32, i32* %P
   ret i32 %A
 ; CHECK-LABEL: @test0(
 ; CHECK: ret i32 %V
@@ -20,7 +20,7 @@ define i32 @test0(i32 %V, i32* %P) {
 define i8 @crash0({i32, i32} %A, {i32, i32}* %P) {
   store {i32, i32} %A, {i32, i32}* %P
   %X = bitcast {i32, i32}* %P to i8*
-  %Y = load i8* %X
+  %Y = load i8, i8* %X
   ret i8 %Y
 }
 
@@ -28,7 +28,7 @@ define i8 @crash0({i32, i32} %A, {i32, i32}* %P) {
 declare void @helper()
 define void @crash1() {
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* undef, i8* undef, i64 undef, i32 1, i1 false) nounwind
-  %tmp = load i8* bitcast (void ()* @helper to i8*)
+  %tmp = load i8, i8* bitcast (void ()* @helper to i8*)
   %x = icmp eq i8 %tmp, 15
   ret void
 }
@@ -45,7 +45,7 @@ define float @coerce_mustalias1(i32 %V, i32* %P) {
    
   %P2 = bitcast i32* %P to float*
 
-  %A = load float* %P2
+  %A = load float, float* %P2
   ret float %A
 ; CHECK-LABEL: @coerce_mustalias1(
 ; CHECK-NOT: load
@@ -58,7 +58,7 @@ define float @coerce_mustalias2(i32* %V, i32** %P) {
    
   %P2 = bitcast i32** %P to float*
 
-  %A = load float* %P2
+  %A = load float, float* %P2
   ret float %A
 ; CHECK-LABEL: @coerce_mustalias2(
 ; CHECK-NOT: load
@@ -71,7 +71,7 @@ define i32* @coerce_mustalias3(float %V, float* %P) {
    
   %P2 = bitcast float* %P to i32**
 
-  %A = load i32** %P2
+  %A = load i32*, i32** %P2
   ret i32* %A
 ; CHECK-LABEL: @coerce_mustalias3(
 ; CHECK-NOT: load
@@ -80,10 +80,10 @@ define i32* @coerce_mustalias3(float %V, float* %P) {
 
 ;; i32 -> f32 load forwarding.
 define float @coerce_mustalias4(i32* %P, i1 %cond) {
-  %A = load i32* %P
+  %A = load i32, i32* %P
   
   %P2 = bitcast i32* %P to float*
-  %B = load float* %P2
+  %B = load float, float* %P2
   br i1 %cond, label %T, label %F
 T:
   ret float %B
@@ -93,7 +93,7 @@ F:
   ret float %X
 
 ; CHECK-LABEL: @coerce_mustalias4(
-; CHECK: %A = load i32* %P
+; CHECK: %A = load i32, i32* %P
 ; CHECK-NOT: load
 ; CHECK: ret float
 ; CHECK: F:
@@ -105,7 +105,7 @@ define i8 @coerce_mustalias5(i32 %V, i32* %P) {
    
   %P2 = bitcast i32* %P to i8*
 
-  %A = load i8* %P2
+  %A = load i8, i8* %P2
   ret i8 %A
 ; CHECK-LABEL: @coerce_mustalias5(
 ; CHECK-NOT: load
@@ -118,7 +118,7 @@ define float @coerce_mustalias6(i64 %V, i64* %P) {
    
   %P2 = bitcast i64* %P to float*
 
-  %A = load float* %P2
+  %A = load float, float* %P2
   ret float %A
 ; CHECK-LABEL: @coerce_mustalias6(
 ; CHECK-NOT: load
@@ -131,7 +131,7 @@ define i8* @coerce_mustalias7(i64 %V, i64* %P) {
    
   %P2 = bitcast i64* %P to i8**
 
-  %A = load i8** %P2
+  %A = load i8*, i8** %P2
   ret i8* %A
 ; CHECK-LABEL: @coerce_mustalias7(
 ; CHECK-NOT: load
@@ -143,8 +143,8 @@ define signext i16 @memset_to_i16_local(i16* %A) nounwind ssp {
 entry:
   %conv = bitcast i16* %A to i8* 
   tail call void @llvm.memset.p0i8.i64(i8* %conv, i8 1, i64 200, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds i16* %A, i64 42
-  %tmp2 = load i16* %arrayidx
+  %arrayidx = getelementptr inbounds i16, i16* %A, i64 42
+  %tmp2 = load i16, i16* %arrayidx
   ret i16 %tmp2
 ; CHECK-LABEL: @memset_to_i16_local(
 ; CHECK-NOT: load
@@ -156,8 +156,8 @@ define float @memset_to_float_local(float* %A, i8 %Val) nounwind ssp {
 entry:
   %conv = bitcast float* %A to i8*                ; <i8*> [#uses=1]
   tail call void @llvm.memset.p0i8.i64(i8* %conv, i8 %Val, i64 400, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds float* %A, i64 42 ; <float*> [#uses=1]
-  %tmp2 = load float* %arrayidx                   ; <float> [#uses=1]
+  %arrayidx = getelementptr inbounds float, float* %A, i64 42 ; <float*> [#uses=1]
+  %tmp2 = load float, float* %arrayidx                   ; <float> [#uses=1]
   ret float %tmp2
 ; CHECK-LABEL: @memset_to_float_local(
 ; CHECK-NOT: load
@@ -183,8 +183,8 @@ F:
   br label %Cont
 
 Cont:
-  %P2 = getelementptr i16* %P, i32 4
-  %A = load i16* %P2
+  %P2 = getelementptr i16, i16* %P, i32 4
+  %A = load i16, i16* %P2
   ret i16 %A
 
 ; CHECK-LABEL: @memset_to_i16_nonlocal0(
@@ -202,8 +202,8 @@ define float @memcpy_to_float_local(float* %A) nounwind ssp {
 entry:
   %conv = bitcast float* %A to i8*                ; <i8*> [#uses=1]
   tail call void @llvm.memcpy.p0i8.p0i8.i64(i8* %conv, i8* bitcast ({i32, float, i32 }* @GCst to i8*), i64 12, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds float* %A, i64 1 ; <float*> [#uses=1]
-  %tmp2 = load float* %arrayidx                   ; <float> [#uses=1]
+  %arrayidx = getelementptr inbounds float, float* %A, i64 1 ; <float*> [#uses=1]
+  %tmp2 = load float, float* %arrayidx                   ; <float> [#uses=1]
   ret float %tmp2
 ; CHECK-LABEL: @memcpy_to_float_local(
 ; CHECK-NOT: load
@@ -215,8 +215,8 @@ define float @memcpy_to_float_local_as1(float* %A) nounwind ssp {
 entry:
   %conv = bitcast float* %A to i8*                ; <i8*> [#uses=1]
   tail call void @llvm.memcpy.p0i8.p1i8.i64(i8* %conv, i8 addrspace(1)* bitcast ({i32, float, i32 } addrspace(1)* @GCst_as1 to i8 addrspace(1)*), i64 12, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds float* %A, i64 1 ; <float*> [#uses=1]
-  %tmp2 = load float* %arrayidx                   ; <float> [#uses=1]
+  %arrayidx = getelementptr inbounds float, float* %A, i64 1 ; <float*> [#uses=1]
+  %tmp2 = load float, float* %arrayidx                   ; <float> [#uses=1]
   ret float %tmp2
 ; CHECK-LABEL: @memcpy_to_float_local_as1(
 ; CHECK-NOT: load
@@ -237,7 +237,7 @@ F:
   br label %Cont
 
 Cont:
-  %A = load i8* %P3
+  %A = load i8, i8* %P3
   ret i8 %A
 
 ; CHECK-LABEL: @coerce_mustalias_nonlocal0(
@@ -263,7 +263,7 @@ F:
 
 Cont:
   %P3 = bitcast i32* %P to i8*
-  %A = load i8* %P3
+  %A = load i8, i8* %P3
   ret i8 %A
 
 ; CHECK-LABEL: @coerce_mustalias_nonlocal1(
@@ -286,12 +286,12 @@ F:
   br label %Cont
 
 Cont:
-  %A = load i8* %P3
+  %A = load i8, i8* %P3
   ret i8 %A
 
 ; CHECK-LABEL: @coerce_mustalias_pre0(
 ; CHECK: F:
-; CHECK:   load i8* %P3
+; CHECK:   load i8, i8* %P3
 ; CHECK: Cont:
 ; CHECK:   %A = phi i8 [
 ; CHECK-NOT: load
@@ -309,9 +309,9 @@ define i8 @coerce_offset0(i32 %V, i32* %P) {
   store i32 %V, i32* %P
    
   %P2 = bitcast i32* %P to i8*
-  %P3 = getelementptr i8* %P2, i32 2
+  %P3 = getelementptr i8, i8* %P2, i32 2
 
-  %A = load i8* %P3
+  %A = load i8, i8* %P3
   ret i8 %A
 ; CHECK-LABEL: @coerce_offset0(
 ; CHECK-NOT: load
@@ -322,9 +322,9 @@ define i8 @coerce_offset0_addrspacecast(i32 %V, i32* %P) {
   store i32 %V, i32* %P
 
   %P2 = addrspacecast i32* %P to i8 addrspace(1)*
-  %P3 = getelementptr i8 addrspace(1)* %P2, i32 2
+  %P3 = getelementptr i8, i8 addrspace(1)* %P2, i32 2
 
-  %A = load i8 addrspace(1)* %P3
+  %A = load i8, i8 addrspace(1)* %P3
   ret i8 %A
 ; CHECK-LABEL: @coerce_offset0_addrspacecast(
 ; CHECK-NOT: load
@@ -335,7 +335,7 @@ define i8 @coerce_offset0_addrspacecast(i32 %V, i32* %P) {
 define i8 @coerce_offset_nonlocal0(i32* %P, i1 %cond) {
   %P2 = bitcast i32* %P to float*
   %P3 = bitcast i32* %P to i8*
-  %P4 = getelementptr i8* %P3, i32 2
+  %P4 = getelementptr i8, i8* %P3, i32 2
   br i1 %cond, label %T, label %F
 T:
   store i32 57005, i32* %P
@@ -346,7 +346,7 @@ F:
   br label %Cont
 
 Cont:
-  %A = load i8* %P4
+  %A = load i8, i8* %P4
   ret i8 %A
 
 ; CHECK-LABEL: @coerce_offset_nonlocal0(
@@ -360,7 +360,7 @@ Cont:
 ;; non-local i32 -> i8 partial redundancy load forwarding.
 define i8 @coerce_offset_pre0(i32* %P, i1 %cond) {
   %P3 = bitcast i32* %P to i8*
-  %P4 = getelementptr i8* %P3, i32 2
+  %P4 = getelementptr i8, i8* %P3, i32 2
   br i1 %cond, label %T, label %F
 T:
   store i32 42, i32* %P
@@ -370,12 +370,12 @@ F:
   br label %Cont
 
 Cont:
-  %A = load i8* %P4
+  %A = load i8, i8* %P4
   ret i8 %A
 
 ; CHECK-LABEL: @coerce_offset_pre0(
 ; CHECK: F:
-; CHECK:   load i8* %P4
+; CHECK:   load i8, i8* %P4
 ; CHECK: Cont:
 ; CHECK:   %A = phi i8 [
 ; CHECK-NOT: load
@@ -386,28 +386,28 @@ define i32 @chained_load(i32** %p, i32 %x, i32 %y) {
 block1:
   %A = alloca i32*
 
-  %z = load i32** %p
+  %z = load i32*, i32** %p
   store i32* %z, i32** %A
   %cmp = icmp eq i32 %x, %y
   br i1 %cmp, label %block2, label %block3
 
 block2:
- %a = load i32** %p
+ %a = load i32*, i32** %p
  br label %block4
 
 block3:
-  %b = load i32** %p
+  %b = load i32*, i32** %p
   br label %block4
 
 block4:
-  %c = load i32** %p
-  %d = load i32* %c
+  %c = load i32*, i32** %p
+  %d = load i32, i32* %c
   ret i32 %d
   
 ; CHECK-LABEL: @chained_load(
-; CHECK: %z = load i32** %p
+; CHECK: %z = load i32*, i32** %p
 ; CHECK-NOT: load
-; CHECK: %d = load i32* %z
+; CHECK: %d = load i32, i32* %z
 ; CHECK-NEXT: ret i32 %d
 }
 
@@ -427,13 +427,13 @@ F1:
   br i1 %cond2, label %T1, label %TY
   
 T1:
-  %P2 = getelementptr i32* %P, i32 %A
-  %x = load i32* %P2
+  %P2 = getelementptr i32, i32* %P, i32 %A
+  %x = load i32, i32* %P2
   %cond = call i1 @cond2()
   br i1 %cond, label %TX, label %F
   
 F:
-  %P3 = getelementptr i32* %P, i32 2
+  %P3 = getelementptr i32, i32* %P, i32 2
   store i32 17, i32* %P3
   
   store i32 42, i32* %P2  ; Provides "P[A]".
@@ -464,7 +464,7 @@ block2:
  br label %block4
 
 block3:
-  %p2 = getelementptr i32* %p, i32 43
+  %p2 = getelementptr i32, i32* %p, i32 43
   store i32 97, i32* %p2
   br label %block4
 
@@ -481,11 +481,11 @@ block5:
   br i1 %cmpxy, label %block6, label %exit
   
 block6:
-  %C = getelementptr i32* %p, i32 %B
+  %C = getelementptr i32, i32* %p, i32 %B
   br i1 %cmpxy, label %block7, label %exit
   
 block7:
-  %D = load i32* %C
+  %D = load i32, i32* %C
   ret i32 %D
   
 ; CHECK: block7:
@@ -498,17 +498,17 @@ exit:
 define i8 @phi_trans4(i8* %p) {
 ; CHECK-LABEL: @phi_trans4(
 entry:
-  %X3 = getelementptr i8* %p, i32 192
+  %X3 = getelementptr i8, i8* %p, i32 192
   store i8 192, i8* %X3
   
-  %X = getelementptr i8* %p, i32 4
-  %Y = load i8* %X
+  %X = getelementptr i8, i8* %p, i32 4
+  %Y = load i8, i8* %X
   br label %loop
 
 loop:
   %i = phi i32 [4, %entry], [192, %loop]
-  %X2 = getelementptr i8* %p, i32 %i
-  %Y2 = load i8* %X2
+  %X2 = getelementptr i8, i8* %p, i32 %i
+  %Y2 = load i8, i8* %X2
   
 ; CHECK: loop:
 ; CHECK-NEXT: %Y2 = phi i8 [ %Y, %entry ], [ 0, %loop ]
@@ -529,29 +529,29 @@ define i8 @phi_trans5(i8* %p) {
 ; CHECK-LABEL: @phi_trans5(
 entry:
   
-  %X4 = getelementptr i8* %p, i32 2
+  %X4 = getelementptr i8, i8* %p, i32 2
   store i8 19, i8* %X4
   
-  %X = getelementptr i8* %p, i32 4
-  %Y = load i8* %X
+  %X = getelementptr i8, i8* %p, i32 4
+  %Y = load i8, i8* %X
   br label %loop
 
 loop:
   %i = phi i32 [4, %entry], [3, %cont]
-  %X2 = getelementptr i8* %p, i32 %i
-  %Y2 = load i8* %X2  ; Ensure this load is not being incorrectly replaced.
+  %X2 = getelementptr i8, i8* %p, i32 %i
+  %Y2 = load i8, i8* %X2  ; Ensure this load is not being incorrectly replaced.
   %cond = call i1 @cond2()
   br i1 %cond, label %cont, label %out
 
 cont:
-  %Z = getelementptr i8* %X2, i32 -1
+  %Z = getelementptr i8, i8* %X2, i32 -1
   %Z2 = bitcast i8 *%Z to i32*
   store i32 50462976, i32* %Z2  ;; (1 << 8) | (2 << 16) | (3 << 24)
 
 
 ; CHECK: store i32
-; CHECK-NEXT: getelementptr i8* %p, i32 3
-; CHECK-NEXT: load i8*
+; CHECK-NEXT: getelementptr i8, i8* %p, i32 3
+; CHECK-NEXT: load i8, i8*
   br label %loop
   
 out:
@@ -566,8 +566,8 @@ entry:
   %x = alloca [256 x i32], align 4                ; <[256 x i32]*> [#uses=2]
   %tmp = bitcast [256 x i32]* %x to i8*           ; <i8*> [#uses=1]
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 1024, i32 4, i1 false)
-  %arraydecay = getelementptr inbounds [256 x i32]* %x, i32 0, i32 0 ; <i32*>
-  %tmp1 = load i32* %arraydecay                   ; <i32> [#uses=1]
+  %arraydecay = getelementptr inbounds [256 x i32], [256 x i32]* %x, i32 0, i32 0 ; <i32*>
+  %tmp1 = load i32, i32* %arraydecay                   ; <i32> [#uses=1]
   ret i32 %tmp1
 ; CHECK-LABEL: @memset_to_load(
 ; CHECK: ret i32 0
@@ -581,15 +581,15 @@ entry:
 define i32 @load_load_partial_alias(i8* %P) nounwind ssp {
 entry:
   %0 = bitcast i8* %P to i32*
-  %tmp2 = load i32* %0
-  %add.ptr = getelementptr inbounds i8* %P, i64 1
-  %tmp5 = load i8* %add.ptr
+  %tmp2 = load i32, i32* %0
+  %add.ptr = getelementptr inbounds i8, i8* %P, i64 1
+  %tmp5 = load i8, i8* %add.ptr
   %conv = zext i8 %tmp5 to i32
   %add = add nsw i32 %tmp2, %conv
   ret i32 %add
 
 ; TEMPORARILYDISABLED-LABEL: @load_load_partial_alias(
-; TEMPORARILYDISABLED: load i32*
+; TEMPORARILYDISABLED: load i32, i32*
 ; TEMPORARILYDISABLED-NOT: load
 ; TEMPORARILYDISABLED: lshr i32 {{.*}}, 8
 ; TEMPORARILYDISABLED-NOT: load
@@ -603,13 +603,13 @@ entry:
 define i32 @load_load_partial_alias_cross_block(i8* %P) nounwind ssp {
 entry:
   %xx = bitcast i8* %P to i32*
-  %x1 = load i32* %xx, align 4
+  %x1 = load i32, i32* %xx, align 4
   %cmp = icmp eq i32 %x1, 127
   br i1 %cmp, label %land.lhs.true, label %if.end
 
 land.lhs.true:                                    ; preds = %entry
-  %arrayidx4 = getelementptr inbounds i8* %P, i64 1
-  %tmp5 = load i8* %arrayidx4, align 1
+  %arrayidx4 = getelementptr inbounds i8, i8* %P, i64 1
+  %tmp5 = load i8, i8* %arrayidx4, align 1
   %conv6 = zext i8 %tmp5 to i32
   ret i32 %conv6
 
@@ -632,39 +632,39 @@ if.end:
 
 define i32 @test_widening1(i8* %P) nounwind ssp noredzone {
 entry:
-  %tmp = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 1), align 4
+  %tmp = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 1), align 4
   %conv = zext i8 %tmp to i32
-  %tmp1 = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 2), align 1
+  %tmp1 = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 2), align 1
   %conv2 = zext i8 %tmp1 to i32
   %add = add nsw i32 %conv, %conv2
   ret i32 %add
 ; CHECK-LABEL: @test_widening1(
 ; CHECK-NOT: load
-; CHECK: load i16*
+; CHECK: load i16, i16*
 ; CHECK-NOT: load
 ; CHECK: ret i32
 }
 
 define i32 @test_widening2() nounwind ssp noredzone {
 entry:
-  %tmp = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 1), align 4
+  %tmp = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 1), align 4
   %conv = zext i8 %tmp to i32
-  %tmp1 = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 2), align 1
+  %tmp1 = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 2), align 1
   %conv2 = zext i8 %tmp1 to i32
   %add = add nsw i32 %conv, %conv2
 
-  %tmp2 = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 3), align 2
+  %tmp2 = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 3), align 2
   %conv3 = zext i8 %tmp2 to i32
   %add2 = add nsw i32 %add, %conv3
 
-  %tmp3 = load i8* getelementptr inbounds (%widening1* @f, i64 0, i32 4), align 1
+  %tmp3 = load i8, i8* getelementptr inbounds (%widening1, %widening1* @f, i64 0, i32 4), align 1
   %conv4 = zext i8 %tmp3 to i32
   %add3 = add nsw i32 %add2, %conv3
 
   ret i32 %add3
 ; CHECK-LABEL: @test_widening2(
 ; CHECK-NOT: load
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NOT: load
 ; CHECK: ret i32
 }
@@ -691,9 +691,9 @@ declare void @use3(i8***, i8**)
 ; PR8908
 define void @test_escape1() nounwind {
   %x = alloca i8**, align 8
-  store i8** getelementptr inbounds ([5 x i8*]* @_ZTV1X, i64 0, i64 2), i8*** %x, align 8
+  store i8** getelementptr inbounds ([5 x i8*], [5 x i8*]* @_ZTV1X, i64 0, i64 2), i8*** %x, align 8
   call void @use() nounwind
-  %DEAD = load i8*** %x, align 8
+  %DEAD = load i8**, i8*** %x, align 8
   call void @use3(i8*** %x, i8** %DEAD) nounwind
   ret void
 ; CHECK: test_escape1
diff --git a/test/Transforms/GVN/tbaa.ll b/test/Transforms/GVN/tbaa.ll
index 1b3494ca4be3..b5a717b0984a 100644
--- a/test/Transforms/GVN/tbaa.ll
+++ b/test/Transforms/GVN/tbaa.ll
@@ -1,4 +1,4 @@
-; RUN: opt -basicaa -gvn -S < %s | FileCheck %s
+; RUN: opt -tbaa -basicaa -gvn -S < %s | FileCheck %s
 
 define i32 @test1(i8* %p, i8* %q) {
 ; CHECK: @test1(i8* %p, i8* %q)
@@ -72,6 +72,37 @@ define i32 @test7(i8* %p, i8* %q) {
   ret i32 %c
 }
 
+
+
+define i32 @test8(i32* %p, i32* %q) {
+; CHECK-LABEL: test8
+; CHECK-NEXT: store i32 15, i32* %p
+; CHECK-NEXT: ret i32 0
+; Since we know the location is invariant, we can forward the
+; load across the potentially aliasing store.
+
+  %a = load i32, i32* %q, !tbaa !10
+  store i32 15, i32* %p
+  %b = load i32, i32* %q, !tbaa !10
+  %c = sub i32 %a, %b
+  ret i32 %c
+}
+define i32 @test9(i32* %p, i32* %q) {
+; CHECK-LABEL: test9
+; CHECK-NEXT: call void @clobber()
+; CHECK-NEXT: ret i32 0
+; Since we know the location is invariant, we can forward the
+; load across the potentially aliasing store (within the call).
+
+  %a = load i32, i32* %q, !tbaa !10
+  call void @clobber()
+  %b = load i32, i32* %q, !tbaa !10
+  %c = sub i32 %a, %b
+  ret i32 %c
+}
+
+
+declare void @clobber()
 declare i32 @foo(i8*) readonly
 
 ; CHECK: [[TAGC]] = !{[[TYPEC:!.*]], [[TYPEC]], i64 0}
@@ -89,3 +120,9 @@ declare i32 @foo(i8*) readonly
 !6 = !{!"A", !2}
 !7 = !{!"B", !6}
 !8 = !{!"another root", null}
+
+
+;; A TBAA structure who's only point is to have a constant location
+!9 = !{!"yet another root"}
+!10 = !{!"node", !9, i64 1}
+
diff --git a/test/Transforms/GVN/volatile.ll b/test/Transforms/GVN/volatile.ll
new file mode 100644
index 000000000000..b31058db4ea8
--- /dev/null
+++ b/test/Transforms/GVN/volatile.ll
@@ -0,0 +1,157 @@
+; Tests that check our handling of volatile instructions encountered
+; when scanning for dependencies
+; RUN: opt -basicaa -gvn -S < %s | FileCheck %s
+
+; Check that we can bypass a volatile load when searching
+; for dependencies of a non-volatile load
+define i32 @test1(i32* nocapture %p, i32* nocapture %q) {
+; CHECK-LABEL: test1
+; CHECK:      %0 = load volatile i32, i32* %q
+; CHECK-NEXT: ret i32 0
+entry:
+  %x = load i32, i32* %p
+  load volatile i32, i32* %q
+  %y = load i32, i32* %p
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+; We can not value forward if the query instruction is 
+; volatile, this would be (in effect) removing the volatile load
+define i32 @test2(i32* nocapture %p, i32* nocapture %q) {
+; CHECK-LABEL: test2
+; CHECK:      %x = load i32, i32* %p
+; CHECK-NEXT: %y = load volatile i32, i32* %p
+; CHECK-NEXT: %add = sub i32 %y, %x
+entry:
+  %x = load i32, i32* %p
+  %y = load volatile i32, i32* %p
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+; If the query instruction is itself volatile, we *cannot*
+; reorder it even if p and q are noalias
+define i32 @test3(i32* noalias nocapture %p, i32* noalias nocapture %q) {
+; CHECK-LABEL: test3
+; CHECK:      %x = load i32, i32* %p
+; CHECK-NEXT: %0 = load volatile i32, i32* %q
+; CHECK-NEXT: %y = load volatile i32, i32* %p
+entry:
+  %x = load i32, i32* %p
+  load volatile i32, i32* %q
+  %y = load volatile i32, i32* %p
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+; If an encountered instruction is both volatile and ordered, 
+; we need to use the strictest ordering of either.  In this 
+; case, the ordering prevents forwarding.
+define i32 @test4(i32* noalias nocapture %p, i32* noalias nocapture %q) {
+; CHECK-LABEL: test4
+; CHECK:      %x = load i32, i32* %p
+; CHECK-NEXT: %0 = load atomic volatile i32, i32* %q seq_cst 
+; CHECK-NEXT: %y = load atomic i32, i32* %p seq_cst
+entry:
+  %x = load i32, i32* %p
+  load atomic volatile i32, i32* %q seq_cst, align 4
+  %y = load atomic i32, i32* %p seq_cst, align 4
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+; Value forwarding from a volatile load is perfectly legal
+define i32 @test5(i32* nocapture %p, i32* nocapture %q) {
+; CHECK-LABEL: test5
+; CHECK:      %x = load volatile i32, i32* %p
+; CHECK-NEXT: ret i32 0
+entry:
+  %x = load volatile i32, i32* %p
+  %y = load i32, i32* %p
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+; Does cross block redundancy elimination work with volatiles?
+define i32 @test6(i32* noalias nocapture %p, i32* noalias nocapture %q) {
+; CHECK-LABEL: test6
+; CHECK:      %y1 = load i32, i32* %p
+; CHECK-LABEL: header
+; CHECK:      %x = load volatile i32, i32* %q
+; CHECK-NEXT: %add = sub i32 %y1, %x
+entry:
+  %y1 = load i32, i32* %p
+  call void @use(i32 %y1)
+  br label %header
+header:
+  %x = load volatile i32, i32* %q
+  %y = load i32, i32* %p
+  %add = sub i32 %y, %x
+  %cnd = icmp eq i32 %add, 0
+  br i1 %cnd, label %exit, label %header
+exit:
+  ret i32 %add
+}
+
+; Does cross block PRE work with volatiles?
+define i32 @test7(i1 %c, i32* noalias nocapture %p, i32* noalias nocapture %q) {
+; CHECK-LABEL: test7
+; CHECK-LABEL: entry.header_crit_edge:
+; CHECK:       %y.pre = load i32, i32* %p
+; CHECK-LABEL: skip:
+; CHECK:       %y1 = load i32, i32* %p
+; CHECK-LABEL: header:
+; CHECK:      %y = phi i32
+; CHECK-NEXT: %x = load volatile i32, i32* %q
+; CHECK-NEXT: %add = sub i32 %y, %x
+entry:
+  br i1 %c, label %header, label %skip
+skip:
+  %y1 = load i32, i32* %p
+  call void @use(i32 %y1)
+  br label %header
+header:
+  %x = load volatile i32, i32* %q
+  %y = load i32, i32* %p
+  %add = sub i32 %y, %x
+  %cnd = icmp eq i32 %add, 0
+  br i1 %cnd, label %exit, label %header
+exit:
+  ret i32 %add
+}
+
+; Another volatile PRE case - two paths through a loop
+; load in preheader, one path read only, one not
+define i32 @test8(i1 %b, i1 %c, i32* noalias %p, i32* noalias %q) {
+; CHECK-LABEL: test8
+; CHECK-LABEL: entry
+; CHECK:       %y1 = load i32, i32* %p
+; CHECK-LABEL: header:
+; CHECK:      %y = phi i32
+; CHECK-NEXT: %x = load volatile i32, i32* %q
+; CHECK-NOT:  load
+; CHECK-LABEL: skip.header_crit_edge:
+; CHECK:       %y.pre = load i32, i32* %p
+entry:
+  %y1 = load i32, i32* %p
+  call void @use(i32 %y1)
+  br label %header
+header:
+  %x = load volatile i32, i32* %q
+  %y = load i32, i32* %p
+  call void @use(i32 %y)
+  br i1 %b, label %skip, label %header
+skip:
+  ; escaping the arguments is explicitly required since we marked 
+  ; them noalias
+  call void @clobber(i32* %p, i32* %q)
+  br i1 %c, label %header, label %exit
+exit:
+  %add = sub i32 %y, %x
+  ret i32 %add
+}
+
+declare void @use(i32) readonly
+declare void @clobber(i32* %p, i32* %q)
+
diff --git a/test/Transforms/GlobalDCE/2002-08-17-FunctionDGE.ll b/test/Transforms/GlobalDCE/2002-08-17-FunctionDGE.ll
index 766c227460e6..49a87d902305 100644
--- a/test/Transforms/GlobalDCE/2002-08-17-FunctionDGE.ll
+++ b/test/Transforms/GlobalDCE/2002-08-17-FunctionDGE.ll
@@ -11,7 +11,7 @@
 
  ;; dies when %b dies.
 define internal i32 @foo() {
-        %ret = load i32* @foo.upgrd.1           ; <i32> [#uses=1]
+        %ret = load i32, i32* @foo.upgrd.1           ; <i32> [#uses=1]
         ret i32 %ret
 }
 
diff --git a/test/Transforms/GlobalDCE/2002-08-17-WorkListTest.ll b/test/Transforms/GlobalDCE/2002-08-17-WorkListTest.ll
index 42fcb1e004a6..58d3f4c7d840 100644
--- a/test/Transforms/GlobalDCE/2002-08-17-WorkListTest.ll
+++ b/test/Transforms/GlobalDCE/2002-08-17-WorkListTest.ll
@@ -5,8 +5,8 @@
 
 @t0 = internal global [4 x i8] c"foo\00"                ; <[4 x i8]*> [#uses=1]
 @t1 = internal global [4 x i8] c"bar\00"                ; <[4 x i8]*> [#uses=1]
-@s1 = internal global [1 x i8*] [ i8* getelementptr ([4 x i8]* @t0, i32 0, i32 0) ]             ; <[1 x i8*]*> [#uses=0]
-@s2 = internal global [1 x i8*] [ i8* getelementptr ([4 x i8]* @t1, i64 0, i64 0) ]             ; <[1 x i8*]*> [#uses=0]
+@s1 = internal global [1 x i8*] [ i8* getelementptr ([4 x i8], [4 x i8]* @t0, i32 0, i32 0) ]             ; <[1 x i8*]*> [#uses=0]
+@s2 = internal global [1 x i8*] [ i8* getelementptr ([4 x i8], [4 x i8]* @t1, i64 0, i64 0) ]             ; <[1 x i8*]*> [#uses=0]
 @b = internal global i32* @a            ; <i32**> [#uses=0]
 @a = internal global i32 7              ; <i32*> [#uses=1]
 
diff --git a/test/Transforms/GlobalDCE/2002-09-12-Redeletion.ll b/test/Transforms/GlobalDCE/2002-09-12-Redeletion.ll
index 6221fa3a62fe..afa2629f8f6b 100644
--- a/test/Transforms/GlobalDCE/2002-09-12-Redeletion.ll
+++ b/test/Transforms/GlobalDCE/2002-09-12-Redeletion.ll
@@ -5,7 +5,7 @@
 @bar = internal global [2 x { i32*, i32 }] [ { i32*, i32 } { i32* @foo.upgrd.1, i32 7 }, { i32*, i32 } { i32* @foo.upgrd.1, i32 1 } ]            ; <[2 x { i32*, i32 }]*> [#uses=0]
 
 define internal i32 @foo() {
-        %ret = load i32* @foo.upgrd.1           ; <i32> [#uses=1]
+        %ret = load i32, i32* @foo.upgrd.1           ; <i32> [#uses=1]
         ret i32 %ret
 }
 
diff --git a/test/Transforms/GlobalDCE/2003-07-01-SelfReference.ll b/test/Transforms/GlobalDCE/2003-07-01-SelfReference.ll
index 738ec43aaeb1..781c57e4ad47 100644
--- a/test/Transforms/GlobalDCE/2003-07-01-SelfReference.ll
+++ b/test/Transforms/GlobalDCE/2003-07-01-SelfReference.ll
@@ -4,7 +4,7 @@
 declare i1 ()* @getfunc()
 
 define internal i1 @testfunc() {
-        %F = call i1 ()* ()* @getfunc( )                ; <i1 ()*> [#uses=1]
+        %F = call i1 ()* () @getfunc( )                ; <i1 ()*> [#uses=1]
         %c = icmp eq i1 ()* %F, @testfunc               ; <i1> [#uses=1]
         ret i1 %c
 }
diff --git a/test/Transforms/GlobalDCE/complex-constantexpr.ll b/test/Transforms/GlobalDCE/complex-constantexpr.ll
index 4bf1aeee709b..b4eed02eff23 100644
--- a/test/Transforms/GlobalDCE/complex-constantexpr.ll
+++ b/test/Transforms/GlobalDCE/complex-constantexpr.ll
@@ -19,7 +19,7 @@ bb:
   br label %bb2
 
 bb1:                                              ; preds = %bb11
-  %tmp = load i32* @global5, align 4
+  %tmp = load i32, i32* @global5, align 4
   br label %bb2
 
 bb2:                                              ; preds = %bb1, %bb
@@ -34,12 +34,12 @@ bb6:                                              ; preds = %bb2
   br label %bb8
 
 bb8:                                              ; preds = %bb6, %bb2
-  %tmp9 = load i32* @global7, align 4
+  %tmp9 = load i32, i32* @global7, align 4
   %tmp10 = icmp eq i32 %tmp9, 0
   br i1 %tmp10, label %bb11, label %bb15
 
 bb11:                                             ; preds = %bb8
-  %tmp12 = load i32* @global6, align 4
+  %tmp12 = load i32, i32* @global6, align 4
   %tmp13 = add nsw i32 %tmp12, 1
   store i32 %tmp13, i32* @global6, align 4
   %tmp14 = icmp slt i32 %tmp13, 42
@@ -69,9 +69,9 @@ bb:
 
 define void @blam() {
 bb:
-  store i32 ptrtoint (i32* @global to i32), i32* getelementptr inbounds (%struct.ham* @global8, i64 0, i32 0), align 4
+  store i32 ptrtoint (i32* @global to i32), i32* getelementptr inbounds (%struct.ham, %struct.ham* @global8, i64 0, i32 0), align 4
   store i32 0, i32* @global9, align 4
-  %tmp = load i32* getelementptr inbounds (%struct.ham* @global8, i64 0, i32 0), align 4
+  %tmp = load i32, i32* getelementptr inbounds (%struct.ham, %struct.ham* @global8, i64 0, i32 0), align 4
   br label %bb1
 
 bb1:                                              ; preds = %bb1, %bb
@@ -89,7 +89,7 @@ bb1:                                              ; preds = %bb1, %bb
   br i1 %tmp12, label %bb13, label %bb1
 
 bb13:                                             ; preds = %bb1
-  store i32 %tmp10, i32* getelementptr inbounds (%struct.ham* @global8, i64 0, i32 0), align 4
+  store i32 %tmp10, i32* getelementptr inbounds (%struct.ham, %struct.ham* @global8, i64 0, i32 0), align 4
   store i32 0, i32* @global10, align 4
   store i32 %tmp6, i32* @global11, align 4
   store i32 20, i32* @global9, align 4
diff --git a/test/Transforms/GlobalDCE/global_ctors_integration.ll b/test/Transforms/GlobalDCE/global_ctors_integration.ll
index 5e6cc79f103c..f7f702a980d7 100644
--- a/test/Transforms/GlobalDCE/global_ctors_integration.ll
+++ b/test/Transforms/GlobalDCE/global_ctors_integration.ll
@@ -12,7 +12,7 @@
 @llvm.global_ctors = appending global [1 x { i32, void ()* }] [{ i32, void ()* } { i32 65535, void ()* @_GLOBAL__I_a }]
 
 define internal void @__cxx_global_var_init() section "__TEXT,__StaticInit,regular,pure_instructions" {
-  %1 = load i32* @_ZN3Bar18LINKER_INITIALIZEDE, align 4
+  %1 = load i32, i32* @_ZN3Bar18LINKER_INITIALIZEDE, align 4
   call void @_ZN3FooC1E17LinkerInitialized(%class.Foo* @foo, i32 %1)
   ret void
 }
@@ -23,8 +23,8 @@ define linkonce_odr void @_ZN3FooC1E17LinkerInitialized(%class.Foo* %this, i32)
   %3 = alloca i32, align 4
   store %class.Foo* %this, %class.Foo** %2, align 8
   store i32 %0, i32* %3, align 4
-  %4 = load %class.Foo** %2
-  %5 = load i32* %3, align 4
+  %4 = load %class.Foo*, %class.Foo** %2
+  %5 = load i32, i32* %3, align 4
   call void @_ZN3FooC2E17LinkerInitialized(%class.Foo* %4, i32 %5)
   ret void
 }
@@ -35,7 +35,7 @@ define linkonce_odr void @_ZN3FooC2E17LinkerInitialized(%class.Foo* %this, i32)
   %3 = alloca i32, align 4
   store %class.Foo* %this, %class.Foo** %2, align 8
   store i32 %0, i32* %3, align 4
-  %4 = load %class.Foo** %2
+  %4 = load %class.Foo*, %class.Foo** %2
   ret void
 }
 
diff --git a/test/Transforms/GlobalDCE/indirectbr.ll b/test/Transforms/GlobalDCE/indirectbr.ll
index 90f1ae44b1ac..5671aea943f4 100644
--- a/test/Transforms/GlobalDCE/indirectbr.ll
+++ b/test/Transforms/GlobalDCE/indirectbr.ll
@@ -9,8 +9,8 @@ entry:
   br label %L1
 
 L1:
-  %arrayidx = getelementptr inbounds [3 x i8*]* @L, i32 0, i32 %idx
-  %l = load i8** %arrayidx
+  %arrayidx = getelementptr inbounds [3 x i8*], [3 x i8*]* @L, i32 0, i32 %idx
+  %l = load i8*, i8** %arrayidx
   indirectbr i8* %l, [label %L1, label %L2]
 
 L2:
diff --git a/test/Transforms/GlobalOpt/2004-10-10-CastStoreOnce.ll b/test/Transforms/GlobalOpt/2004-10-10-CastStoreOnce.ll
index bdcf1fa4778b..061b9b0670f0 100644
--- a/test/Transforms/GlobalOpt/2004-10-10-CastStoreOnce.ll
+++ b/test/Transforms/GlobalOpt/2004-10-10-CastStoreOnce.ll
@@ -4,8 +4,8 @@
 @G = internal global i32* null          ; <i32**> [#uses=2]
 
 define i32 @user() {
-        %P = load i32** @G              ; <i32*> [#uses=1]
-        %Q = load i32* %P               ; <i32> [#uses=1]
+        %P = load i32*, i32** @G              ; <i32*> [#uses=1]
+        %Q = load i32, i32* %P               ; <i32> [#uses=1]
         ret i32 %Q
 }
 
diff --git a/test/Transforms/GlobalOpt/2005-06-15-LocalizeConstExprCrash.ll b/test/Transforms/GlobalOpt/2005-06-15-LocalizeConstExprCrash.ll
index 7bcb1d430918..3efbde423416 100644
--- a/test/Transforms/GlobalOpt/2005-06-15-LocalizeConstExprCrash.ll
+++ b/test/Transforms/GlobalOpt/2005-06-15-LocalizeConstExprCrash.ll
@@ -4,7 +4,7 @@
 @g_40507551 = internal global i16 31038         ; <i16*> [#uses=1]
 
 define void @main() {
-        %tmp.4.i.1 = load i8* getelementptr (i8* bitcast (i16* @g_40507551 to i8*), i32 1)              ; <i8> [#uses=0]
+        %tmp.4.i.1 = load i8, i8* getelementptr (i8, i8* bitcast (i16* @g_40507551 to i8*), i32 1)              ; <i8> [#uses=0]
         ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/2005-09-27-Crash.ll b/test/Transforms/GlobalOpt/2005-09-27-Crash.ll
index 43597bfd3685..061c88159dd3 100644
--- a/test/Transforms/GlobalOpt/2005-09-27-Crash.ll
+++ b/test/Transforms/GlobalOpt/2005-09-27-Crash.ll
@@ -8,7 +8,7 @@
 
 define fastcc void @pypy_array_constant() {
 block0:
-        %tmp.9 = getelementptr %structtype.test* bitcast ({ i32, { i32, [2 x i32] } }* @structinstance.test to %structtype.test*), i32 0, i32 0          ; <i32*> [#uses=0]
+        %tmp.9 = getelementptr %structtype.test, %structtype.test* bitcast ({ i32, { i32, [2 x i32] } }* @structinstance.test to %structtype.test*), i32 0, i32 0          ; <i32*> [#uses=0]
         ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/2006-07-07-InlineAsmCrash.ll b/test/Transforms/GlobalOpt/2006-07-07-InlineAsmCrash.ll
index c9712198ce71..419ae101966a 100644
--- a/test/Transforms/GlobalOpt/2006-07-07-InlineAsmCrash.ll
+++ b/test/Transforms/GlobalOpt/2006-07-07-InlineAsmCrash.ll
@@ -106,12 +106,12 @@ entry:
 	%__priority_addr = alloca i32		; <i32*> [#uses=2]
 	store i32 %__initialize_p, i32* %__initialize_p_addr
 	store i32 %__priority, i32* %__priority_addr
-	%tmp = load i32* %__priority_addr		; <i32> [#uses=1]
+	%tmp = load i32, i32* %__priority_addr		; <i32> [#uses=1]
 	%tmp.upgrd.1 = icmp eq i32 %tmp, 65535		; <i1> [#uses=1]
 	br i1 %tmp.upgrd.1, label %cond_true, label %cond_next14
 
 cond_true:		; preds = %entry
-	%tmp8 = load i32* %__initialize_p_addr		; <i32> [#uses=1]
+	%tmp8 = load i32, i32* %__initialize_p_addr		; <i32> [#uses=1]
 	%tmp9 = icmp eq i32 %tmp8, 1		; <i1> [#uses=1]
 	br i1 %tmp9, label %cond_true10, label %cond_next14
 
diff --git a/test/Transforms/GlobalOpt/2006-11-01-ShrinkGlobalPhiCrash.ll b/test/Transforms/GlobalOpt/2006-11-01-ShrinkGlobalPhiCrash.ll
index 352639ac067c..7b62cf08a4b1 100644
--- a/test/Transforms/GlobalOpt/2006-11-01-ShrinkGlobalPhiCrash.ll
+++ b/test/Transforms/GlobalOpt/2006-11-01-ShrinkGlobalPhiCrash.ll
@@ -10,7 +10,7 @@ entry:
 
 cond_true16.i:          ; preds = %cond_true16.i, %entry
         %low.0.in.i.0 = phi i32* [ @nrow, %entry ], [ null, %cond_true16.i ]            ; <i32*> [#uses=1]
-        %low.0.i = load i32* %low.0.in.i.0              ; <i32> [#uses=0]
+        %low.0.i = load i32, i32* %low.0.in.i.0              ; <i32> [#uses=0]
         br label %cond_true16.i
 }
 
diff --git a/test/Transforms/GlobalOpt/2007-04-05-Crash.ll b/test/Transforms/GlobalOpt/2007-04-05-Crash.ll
index c7aca62895fa..f312fbbc9f93 100644
--- a/test/Transforms/GlobalOpt/2007-04-05-Crash.ll
+++ b/test/Transforms/GlobalOpt/2007-04-05-Crash.ll
@@ -9,7 +9,7 @@ target triple = "thumb-apple-darwin8"
 define zeroext i16 @__NSCharToUnicharCFWrapper(i8 zeroext  %ch)   {
 entry:
 	%iftmp.0.0.in.in = select i1 false, i16* @replacementUnichar, i16* null		; <i16*> [#uses=1]
-	%iftmp.0.0.in = load i16* %iftmp.0.0.in.in		; <i16> [#uses=1]
+	%iftmp.0.0.in = load i16, i16* %iftmp.0.0.in.in		; <i16> [#uses=1]
 	ret i16 %iftmp.0.0.in
 }
 
diff --git a/test/Transforms/GlobalOpt/2007-05-13-Crash.ll b/test/Transforms/GlobalOpt/2007-05-13-Crash.ll
index 57039093d1e4..90ba15c37074 100644
--- a/test/Transforms/GlobalOpt/2007-05-13-Crash.ll
+++ b/test/Transforms/GlobalOpt/2007-05-13-Crash.ll
@@ -9,9 +9,9 @@ target triple = "i686-apple-darwin8"
 @_ZZ19SFLGetVisibilityKeyvE19_kSFLLVisibilityKey = internal global %struct.__CFString* null             ; <%struct.__CFString**> [#uses=2]
 @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey = internal global %struct.__CFString* null               ; <%struct.__CFString**> [#uses=7]
 internal constant %struct.__builtin_CFString {
-    i32* getelementptr ([0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), 
+    i32* getelementptr ([0 x i32], [0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), 
     i32 1992, 
-    i8* getelementptr ([14 x i8]* @.str, i32 0, i32 0), 
+    i8* getelementptr ([14 x i8], [14 x i8]* @.str, i32 0, i32 0), 
     i32 13 }, section "__DATA,__cfstring"               ; <%struct.__builtin_CFString*>:0 [#uses=1]
 @__CFConstantStringClassReference = external global [0 x i32]           ; <[0 x i32]*> [#uses=1]
 @.str = internal constant [14 x i8] c"AlwaysVisible\00"         ; <[14 x i8]*> [#uses=1]
@@ -19,13 +19,13 @@ internal constant %struct.__builtin_CFString {
 
 define %struct.__CFString* @_Z19SFLGetVisibilityKeyv() {
 entry:
-        %tmp1 = load %struct.__CFString** @_ZZ19SFLGetVisibilityKeyvE19_kSFLLVisibilityKey              ; <%struct.__CFString*> [#uses=1]
+        %tmp1 = load %struct.__CFString*, %struct.__CFString** @_ZZ19SFLGetVisibilityKeyvE19_kSFLLVisibilityKey              ; <%struct.__CFString*> [#uses=1]
         ret %struct.__CFString* %tmp1
 }
 
 define %struct.__CFString* @_Z22SFLGetAlwaysVisibleKeyv() {
 entry:
-        %tmp1 = load %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
+        %tmp1 = load %struct.__CFString*, %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
         %tmp2 = icmp eq %struct.__CFString* %tmp1, null         ; <i1> [#uses=1]
         br i1 %tmp2, label %cond_true, label %cond_next
 
@@ -34,25 +34,25 @@ cond_true:              ; preds = %entry
         br label %cond_next
 
 cond_next:              ; preds = %entry, %cond_true
-        %tmp4 = load %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
+        %tmp4 = load %struct.__CFString*, %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
         ret %struct.__CFString* %tmp4
 }
 
 define %struct.__CFString* @_Z21SFLGetNeverVisibleKeyv() {
 entry:
-        %tmp1 = load %struct.__CFString** @_ZZ21SFLGetNeverVisibleKeyvE21_kSFLLNeverVisibleKey          ; <%struct.__CFString*> [#uses=1]
+        %tmp1 = load %struct.__CFString*, %struct.__CFString** @_ZZ21SFLGetNeverVisibleKeyvE21_kSFLLNeverVisibleKey          ; <%struct.__CFString*> [#uses=1]
         ret %struct.__CFString* %tmp1
 }
 
 define %struct.__CFDictionary* @_ZN18SFLMutableListItem18GetPrefsDictionaryEv(%struct.SFLMutableListItem* %this) {
 entry:
-        %tmp4 = getelementptr %struct.SFLMutableListItem* %this, i32 0, i32 0  ; <i16*> [#uses=1]
-        %tmp5 = load i16* %tmp4         ; <i16> [#uses=1]
+        %tmp4 = getelementptr %struct.SFLMutableListItem, %struct.SFLMutableListItem* %this, i32 0, i32 0  ; <i16*> [#uses=1]
+        %tmp5 = load i16, i16* %tmp4         ; <i16> [#uses=1]
         %tmp6 = icmp eq i16 %tmp5, 0            ; <i1> [#uses=1]
         br i1 %tmp6, label %cond_next22, label %cond_true
 
 cond_true:              ; preds = %entry
-        %tmp9 = load %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
+        %tmp9 = load %struct.__CFString*, %struct.__CFString** @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey                ; <%struct.__CFString*> [#uses=1]
         %tmp10 = icmp eq %struct.__CFString* %tmp9, null                ; <i1> [#uses=1]
         br i1 %tmp10, label %cond_true13, label %cond_next22
 
@@ -62,8 +62,8 @@ cond_true13:            ; preds = %cond_true
 
 cond_next22:            ; preds = %entry, %cond_true13, %cond_true
         %iftmp.1.0.in = phi %struct.__CFString** [ @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey, %cond_true ], [ @_ZZ22SFLGetAlwaysVisibleKeyvE22_kSFLLAlwaysVisibleKey, %cond_true13 ], [ @_ZZ21SFLGetNeverVisibleKeyvE21_kSFLLNeverVisibleKey, %entry ]             ; <%struct.__CFString**> [#uses=1]
-        %iftmp.1.0 = load %struct.__CFString** %iftmp.1.0.in            ; <%struct.__CFString*> [#uses=1]
-        %tmp24 = load %struct.__CFString** @_ZZ19SFLGetVisibilityKeyvE19_kSFLLVisibilityKey             ; <%struct.__CFString*> [#uses=1]
+        %iftmp.1.0 = load %struct.__CFString*, %struct.__CFString** %iftmp.1.0.in            ; <%struct.__CFString*> [#uses=1]
+        %tmp24 = load %struct.__CFString*, %struct.__CFString** @_ZZ19SFLGetVisibilityKeyvE19_kSFLLVisibilityKey             ; <%struct.__CFString*> [#uses=1]
         %tmp2728 = bitcast %struct.__CFString* %tmp24 to i8*            ; <i8*> [#uses=1]
         %tmp2930 = bitcast %struct.__CFString* %iftmp.1.0 to i8*               ; <i8*> [#uses=1]
         call void @_Z20CFDictionaryAddValuePKvS0_( i8* %tmp2728, i8* %tmp2930 )
diff --git a/test/Transforms/GlobalOpt/2007-06-04-PackedStruct.ll b/test/Transforms/GlobalOpt/2007-06-04-PackedStruct.ll
index 7036c158ba67..f6e0bb70d63f 100644
--- a/test/Transforms/GlobalOpt/2007-06-04-PackedStruct.ll
+++ b/test/Transforms/GlobalOpt/2007-06-04-PackedStruct.ll
@@ -28,7 +28,7 @@ entry:
 	br i1 %tmp7, label %cond_true, label %cond_next
 
 cond_true:		; preds = %entry
-	store i8 0, i8* getelementptr (%"struct.std::map<int,int,std::less<int>,std::allocator<std::pair<const int, int> > >"* @someMap, i32 0, i32 0, i32 0, i32 0, i32 0)
+	store i8 0, i8* getelementptr (%"struct.std::map<int,int,std::less<int>,std::allocator<std::pair<const int, int> > >", %"struct.std::map<int,int,std::less<int>,std::allocator<std::pair<const int, int> > >"* @someMap, i32 0, i32 0, i32 0, i32 0, i32 0)
 	ret void
 
 cond_next:		; preds = %entry
diff --git a/test/Transforms/GlobalOpt/2007-11-09-GEP-GEP-Crash.ll b/test/Transforms/GlobalOpt/2007-11-09-GEP-GEP-Crash.ll
index 442cb921d8a3..7a2de55771b2 100644
--- a/test/Transforms/GlobalOpt/2007-11-09-GEP-GEP-Crash.ll
+++ b/test/Transforms/GlobalOpt/2007-11-09-GEP-GEP-Crash.ll
@@ -9,9 +9,9 @@ target triple = "powerpc-apple-darwin8"
 
 define i8 @func() {
 entry:
-        %tmp10 = getelementptr [2 x i32]* getelementptr ([6 x [2 x i32]]* @aaui1, i32 0, i32 0), i32 5, i32 1           ; <i32*> [#uses=1]
-        %tmp11 = load i32* %tmp10, align 4              ; <i32> [#uses=1]
-        %tmp12 = call i32 (...)* @func3( i32* null, i32 0, i32 %tmp11 )         ; <i32> [#uses=0]
+        %tmp10 = getelementptr [2 x i32], [2 x i32]* getelementptr ([6 x [2 x i32]], [6 x [2 x i32]]* @aaui1, i32 0, i32 0), i32 5, i32 1           ; <i32*> [#uses=1]
+        %tmp11 = load i32, i32* %tmp10, align 4              ; <i32> [#uses=1]
+        %tmp12 = call i32 (...) @func3( i32* null, i32 0, i32 %tmp11 )         ; <i32> [#uses=0]
         ret i8 undef
 }
 
diff --git a/test/Transforms/GlobalOpt/2008-01-03-Crash.ll b/test/Transforms/GlobalOpt/2008-01-03-Crash.ll
index 4105ab1ed5b9..dc41fdb8de46 100644
--- a/test/Transforms/GlobalOpt/2008-01-03-Crash.ll
+++ b/test/Transforms/GlobalOpt/2008-01-03-Crash.ll
@@ -20,7 +20,7 @@ stack_restore:		; preds = %cond_next20.i
 	ret i32 0
 
 cond_next21.i.i23.i:		; preds = %cond_next20.i
-	%tmp6.i4.i = load i32* bitcast (void (i32)** @indirect1 to i32*), align 4		; <i32> [#uses=0]
+	%tmp6.i4.i = load i32, i32* bitcast (void (i32)** @indirect1 to i32*), align 4		; <i32> [#uses=0]
 	ret i32 0
 }
 
diff --git a/test/Transforms/GlobalOpt/2008-01-13-OutOfRangeSROA.ll b/test/Transforms/GlobalOpt/2008-01-13-OutOfRangeSROA.ll
index 7c07d5d9a23a..8df7050b38a7 100644
--- a/test/Transforms/GlobalOpt/2008-01-13-OutOfRangeSROA.ll
+++ b/test/Transforms/GlobalOpt/2008-01-13-OutOfRangeSROA.ll
@@ -4,13 +4,13 @@
 @mm = internal global [16 x [31 x double]] zeroinitializer, align 32
 
 define void @test(i32 %X) {
-	%P = getelementptr [16 x [31 x double]]* @mm, i32 0, i32 0, i32 %X
+	%P = getelementptr [16 x [31 x double]], [16 x [31 x double]]* @mm, i32 0, i32 0, i32 %X
 	store double 1.0, double* %P
 	ret void
 }
 
 define double @get(i32 %X) {
-	%P = getelementptr [16 x [31 x double]]* @mm, i32 0, i32 0, i32 %X
-	%V = load double* %P
+	%P = getelementptr [16 x [31 x double]], [16 x [31 x double]]* @mm, i32 0, i32 0, i32 %X
+	%V = load double, double* %P
 	ret double %V
 }
diff --git a/test/Transforms/GlobalOpt/2008-01-29-VolatileGlobal.ll b/test/Transforms/GlobalOpt/2008-01-29-VolatileGlobal.ll
index 08b2cb1eb6f2..ac663604854d 100644
--- a/test/Transforms/GlobalOpt/2008-01-29-VolatileGlobal.ll
+++ b/test/Transforms/GlobalOpt/2008-01-29-VolatileGlobal.ll
@@ -3,7 +3,7 @@
 
 define double @foo() nounwind  {
 entry:
-	%tmp1 = load volatile double* @t0.1441, align 8		; <double> [#uses=2]
+	%tmp1 = load volatile double, double* @t0.1441, align 8		; <double> [#uses=2]
 	%tmp4 = fmul double %tmp1, %tmp1		; <double> [#uses=1]
 	ret double %tmp4
 }
diff --git a/test/Transforms/GlobalOpt/2008-04-26-SROA-Global-Align.ll b/test/Transforms/GlobalOpt/2008-04-26-SROA-Global-Align.ll
index d58becd53078..934d928aa88e 100644
--- a/test/Transforms/GlobalOpt/2008-04-26-SROA-Global-Align.ll
+++ b/test/Transforms/GlobalOpt/2008-04-26-SROA-Global-Align.ll
@@ -16,16 +16,16 @@ target triple = "x86_64-apple-darwin8"
 
 
 define void @test() {
-  store double 1.0, double* getelementptr (%T* @G, i32 0, i32 0), align 16
-  store double 2.0, double* getelementptr (%T* @G, i32 0, i32 1), align 8
-  store double 3.0, double* getelementptr (%T* @G, i32 0, i32 2), align 16
+  store double 1.0, double* getelementptr (%T, %T* @G, i32 0, i32 0), align 16
+  store double 2.0, double* getelementptr (%T, %T* @G, i32 0, i32 1), align 8
+  store double 3.0, double* getelementptr (%T, %T* @G, i32 0, i32 2), align 16
   ret void
 }
 
 define double @test2() {
-  %V1 = load double* getelementptr (%T* @G, i32 0, i32 0), align 16
-  %V2 = load double* getelementptr (%T* @G, i32 0, i32 1), align 8
-  %V3 = load double* getelementptr (%T* @G, i32 0, i32 2), align 16
+  %V1 = load double, double* getelementptr (%T, %T* @G, i32 0, i32 0), align 16
+  %V2 = load double, double* getelementptr (%T, %T* @G, i32 0, i32 1), align 8
+  %V3 = load double, double* getelementptr (%T, %T* @G, i32 0, i32 2), align 16
   %R = fadd double %V1, %V2
   %R2 = fadd double %R, %V3
   ret double %R2
diff --git a/test/Transforms/GlobalOpt/2008-07-17-addrspace.ll b/test/Transforms/GlobalOpt/2008-07-17-addrspace.ll
index 0867ca9c5431..d602df1e701c 100644
--- a/test/Transforms/GlobalOpt/2008-07-17-addrspace.ll
+++ b/test/Transforms/GlobalOpt/2008-07-17-addrspace.ll
@@ -10,8 +10,8 @@
 @array = internal addrspace(1) global [ 2 x i32 ] zeroinitializer 
 
 define i32 @foo() {
-  %A = load i32 addrspace(1) * getelementptr ({ i32, i32 } addrspace(1) * @struct, i32 0, i32 0)
-  %B = load i32 addrspace(1) * getelementptr ([ 2 x i32 ] addrspace(1) * @array, i32 0, i32 0)
+  %A = load i32, i32 addrspace(1) * getelementptr ({ i32, i32 }, { i32, i32 } addrspace(1) * @struct, i32 0, i32 0)
+  %B = load i32, i32 addrspace(1) * getelementptr ([ 2 x i32 ], [ 2 x i32 ] addrspace(1) * @array, i32 0, i32 0)
   ; Use the loaded values, so they won't get removed completely
   %R = add i32 %A, %B
   ret i32 %R
@@ -20,8 +20,8 @@ define i32 @foo() {
 ; We put stores in a different function, so that the global variables won't get
 ; optimized away completely.
 define void @bar(i32 %R) {
-  store i32 %R, i32 addrspace(1) * getelementptr ([ 2 x i32 ] addrspace(1) * @array, i32 0, i32 0)
-  store i32 %R, i32 addrspace(1) * getelementptr ({ i32, i32 } addrspace(1) * @struct, i32 0, i32 0)
+  store i32 %R, i32 addrspace(1) * getelementptr ([ 2 x i32 ], [ 2 x i32 ] addrspace(1) * @array, i32 0, i32 0)
+  store i32 %R, i32 addrspace(1) * getelementptr ({ i32, i32 }, { i32, i32 } addrspace(1) * @struct, i32 0, i32 0)
   ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash-2.ll b/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash-2.ll
index b74e4fcd1ef0..bd32163e5f9f 100644
--- a/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash-2.ll
+++ b/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash-2.ll
@@ -6,9 +6,9 @@ target triple = "i386-apple-darwin7"
 
 define void @bar(i32 %Size) nounwind noinline {
 entry:
-        %malloccall = tail call i8* @malloc(i32 trunc (i64 mul (i64 ptrtoint (i32* getelementptr (i32* null, i32 1) to i64), i64 2000000) to i32))
+        %malloccall = tail call i8* @malloc(i32 trunc (i64 mul (i64 ptrtoint (i32* getelementptr (i32, i32* null, i32 1) to i64), i64 2000000) to i32))
         %tmp = bitcast i8* %malloccall to [1000000 x %struct.foo]*
-	%.sub = getelementptr [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
+	%.sub = getelementptr [1000000 x %struct.foo], [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
 	store %struct.foo* %.sub, %struct.foo** @X, align 4
 	ret void
 }
@@ -18,11 +18,11 @@ declare noalias i8* @malloc(i32)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%tmpLD1 = load %struct.foo** @X, align 4		; <%struct.foo*> [#uses=2]
+	%tmpLD1 = load %struct.foo*, %struct.foo** @X, align 4		; <%struct.foo*> [#uses=2]
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
 	%tmp = phi %struct.foo* [ %tmpLD1, %bb1.thread ], [ %tmpLD1, %bb1 ]		; <%struct.foo*> [#uses=1]
-	%0 = getelementptr %struct.foo* %tmp, i32 1		; <%struct.foo*> [#uses=0]
+	%0 = getelementptr %struct.foo, %struct.foo* %tmp, i32 1		; <%struct.foo*> [#uses=0]
 	br label %bb1
 }
diff --git a/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash.ll b/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash.ll
index 613cb7bcef02..e9c16785afae 100644
--- a/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash.ll
+++ b/test/Transforms/GlobalOpt/2008-12-16-HeapSRACrash.ll
@@ -6,9 +6,9 @@ target triple = "i386-apple-darwin7"
 
 define void @bar(i32 %Size) nounwind noinline {
 entry:
-        %malloccall = tail call i8* @malloc(i32 trunc (i64 mul (i64 ptrtoint (i32* getelementptr (i32* null, i32 1) to i64), i64 2000000) to i32))
+        %malloccall = tail call i8* @malloc(i32 trunc (i64 mul (i64 ptrtoint (i32* getelementptr (i32, i32* null, i32 1) to i64), i64 2000000) to i32))
         %tmp = bitcast i8* %malloccall to [1000000 x %struct.foo]*
-	%.sub = getelementptr [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
+	%.sub = getelementptr [1000000 x %struct.foo], [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
 	store %struct.foo* %.sub, %struct.foo** @X, align 4
 	ret void
 }
@@ -17,7 +17,7 @@ declare noalias i8* @malloc(i32)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%tmpLD1 = load %struct.foo** @X, align 4		; <%struct.foo*> [#uses=3]
+	%tmpLD1 = load %struct.foo*, %struct.foo** @X, align 4		; <%struct.foo*> [#uses=3]
 	store %struct.foo* %tmpLD1, %struct.foo** null
 	br label %bb1
 
diff --git a/test/Transforms/GlobalOpt/2009-01-13-phi-user.ll b/test/Transforms/GlobalOpt/2009-01-13-phi-user.ll
index e76c44dbd044..21ec5267b973 100644
--- a/test/Transforms/GlobalOpt/2009-01-13-phi-user.ll
+++ b/test/Transforms/GlobalOpt/2009-01-13-phi-user.ll
@@ -12,15 +12,15 @@ entry:
 	br label %bb1
 
 bb:		; preds = %bb1
-	%0 = getelementptr %struct.node* %t.0, i64 0, i32 1		; <i32*> [#uses=1]
-	%1 = load i32* %0, align 4		; <i32> [#uses=1]
-	%2 = getelementptr %struct.node* %t.0, i64 0, i32 0		; <%struct.node**> [#uses=1]
+	%0 = getelementptr %struct.node, %struct.node* %t.0, i64 0, i32 1		; <i32*> [#uses=1]
+	%1 = load i32, i32* %0, align 4		; <i32> [#uses=1]
+	%2 = getelementptr %struct.node, %struct.node* %t.0, i64 0, i32 0		; <%struct.node**> [#uses=1]
 	br label %bb1
 
 bb1:		; preds = %bb, %entry
 	%value.0 = phi i32 [ undef, %entry ], [ %1, %bb ]		; <i32> [#uses=1]
 	%t.0.in = phi %struct.node** [ @head, %entry ], [ %2, %bb ]		; <%struct.node**> [#uses=1]
-	%t.0 = load %struct.node** %t.0.in		; <%struct.node*> [#uses=3]
+	%t.0 = load %struct.node*, %struct.node** %t.0.in		; <%struct.node*> [#uses=3]
 	%3 = icmp eq %struct.node* %t.0, null		; <i1> [#uses=1]
 	br i1 %3, label %bb2, label %bb
 
diff --git a/test/Transforms/GlobalOpt/2009-02-15-BitcastAlias.ll b/test/Transforms/GlobalOpt/2009-02-15-BitcastAlias.ll
index d6a565ad10a8..a3e90045d64b 100644
--- a/test/Transforms/GlobalOpt/2009-02-15-BitcastAlias.ll
+++ b/test/Transforms/GlobalOpt/2009-02-15-BitcastAlias.ll
@@ -5,6 +5,6 @@
 @a = alias bitcast (i32* @g to i8*)
 
 define void @f() {
-	%tmp = load i8* @a
+	%tmp = load i8, i8* @a
 	ret void
 }
diff --git a/test/Transforms/GlobalOpt/2009-02-15-ResolveAlias.ll b/test/Transforms/GlobalOpt/2009-02-15-ResolveAlias.ll
index 8efd01807fdb..6933d4a8d96c 100644
--- a/test/Transforms/GlobalOpt/2009-02-15-ResolveAlias.ll
+++ b/test/Transforms/GlobalOpt/2009-02-15-ResolveAlias.ll
@@ -9,7 +9,7 @@ define internal void @f() {
 @a = alias void ()* @f
 
 define void @g() {
-	call void()* @a()
+	call void() @a()
 	ret void
 }
 
@@ -17,7 +17,7 @@ define void @g() {
 ; CHECK-NOT: @b
 
 define void @h() {
-	call void()* @b()
+	call void() @b()
 ; CHECK: call void @g
 	ret void
 }
diff --git a/test/Transforms/GlobalOpt/2009-03-05-dbg.ll b/test/Transforms/GlobalOpt/2009-03-05-dbg.ll
index 049eef10eb7a..a8d618ae522d 100644
--- a/test/Transforms/GlobalOpt/2009-03-05-dbg.ll
+++ b/test/Transforms/GlobalOpt/2009-03-05-dbg.ll
@@ -33,7 +33,7 @@ declare void @llvm.dbg.declare(metadata, metadata, metadata) nounwind readnone
 define i32 @bar() nounwind ssp {
 entry:
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
-  %0 = load i32* @Stop, align 4, !dbg !13         ; <i32> [#uses=1]
+  %0 = load i32, i32* @Stop, align 4, !dbg !13         ; <i32> [#uses=1]
   %1 = icmp eq i32 %0, 1, !dbg !13                ; <i1> [#uses=1]
   br i1 %1, label %bb, label %bb1, !dbg !13
 
@@ -55,25 +55,25 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 
 !llvm.dbg.gv = !{!0}
 
-!0 = !{!"0x34\00Stop\00Stop\00\002\001\001", !1, !1, !2, i32* @Stop} ; [ DW_TAG_variable ]
-!1 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build)\001\00\000\00\000", !20, !21, !21, null, null, null} ; [ DW_TAG_compile_unit ]
-!2 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !1} ; [ DW_TAG_base_type ]
-!3 = !{!"0x101\00i\004\000", !4, !1, !2} ; [ DW_TAG_arg_variable ]
-!4 = !{!"0x2e\00foo\00foo\00foo\004\000\001\000\006\000\000\000", i32 0, !1, !5, null, null, null, null, null} ; [ DW_TAG_subprogram ]
-!5 = !{!"0x15\00\000\000\000\000\000\000", !1, null, null, !6, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DIGlobalVariable(name: "Stop", line: 2, isLocal: true, isDefinition: true, scope: !1, file: !1, type: !2, variable: i32* @Stop)
+!1 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: 0, file: !20, enums: !21, retainedTypes: !21)
+!2 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!3 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i", line: 4, arg: 0, scope: !4, file: !1, type: !2)
+!4 = !DISubprogram(name: "foo", linkageName: "foo", line: 4, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, scope: !1, type: !5)
+!5 = !DISubroutineType(types: !6)
 !6 = !{!2, !2}
-!7 = !MDLocation(line: 5, scope: !8)
-!8 = !{!"0xb\000\000\000", !20, !4} ; [ DW_TAG_lexical_block ]
-!9 = !MDLocation(line: 6, scope: !8)
-!10 = !MDLocation(line: 7, scope: !8)
-!11 = !MDLocation(line: 9, scope: !8)
-!12 = !MDLocation(line: 11, scope: !8)
-!13 = !MDLocation(line: 14, scope: !14)
-!14 = !{!"0xb\000\000\000", !20, !15} ; [ DW_TAG_lexical_block ]
-!15 = !{!"0x2e\00bar\00bar\00bar\0013\000\001\000\006\000\000\000", i32 0, !1, !16, null, null, null, null, null} ; [ DW_TAG_subprogram ]
-!16 = !{!"0x15\00\000\000\000\000\000\000", !1, null, null, !17, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!7 = !DILocation(line: 5, scope: !8)
+!8 = distinct !DILexicalBlock(line: 0, column: 0, file: !20, scope: !4)
+!9 = !DILocation(line: 6, scope: !8)
+!10 = !DILocation(line: 7, scope: !8)
+!11 = !DILocation(line: 9, scope: !8)
+!12 = !DILocation(line: 11, scope: !8)
+!13 = !DILocation(line: 14, scope: !14)
+!14 = distinct !DILexicalBlock(line: 0, column: 0, file: !20, scope: !15)
+!15 = !DISubprogram(name: "bar", linkageName: "bar", line: 13, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, scope: !1, type: !16)
+!16 = !DISubroutineType(types: !17)
 !17 = !{!2}
-!18 = !MDLocation(line: 15, scope: !14)
-!19 = !MDLocation(line: 16, scope: !14)
-!20 = !{!"g.c", !"/tmp"}
+!18 = !DILocation(line: 15, scope: !14)
+!19 = !DILocation(line: 16, scope: !14)
+!20 = !DIFile(filename: "g.c", directory: "/tmp")
 !21 = !{i32 0}
diff --git a/test/Transforms/GlobalOpt/2009-03-07-PromotePtrToBool.ll b/test/Transforms/GlobalOpt/2009-03-07-PromotePtrToBool.ll
index 059af1cfeac8..469fa765c7af 100644
--- a/test/Transforms/GlobalOpt/2009-03-07-PromotePtrToBool.ll
+++ b/test/Transforms/GlobalOpt/2009-03-07-PromotePtrToBool.ll
@@ -12,6 +12,6 @@ entry:
 
 define i32* @get() nounwind {
 entry:
-	%0 = load i32** @X, align 4		; <i32*> [#uses=1]
+	%0 = load i32*, i32** @X, align 4		; <i32*> [#uses=1]
 	ret i32* %0
 }
diff --git a/test/Transforms/GlobalOpt/2009-06-01-RecursivePHI.ll b/test/Transforms/GlobalOpt/2009-06-01-RecursivePHI.ll
index d3c3ff59fea1..30e4d4226215 100644
--- a/test/Transforms/GlobalOpt/2009-06-01-RecursivePHI.ll
+++ b/test/Transforms/GlobalOpt/2009-06-01-RecursivePHI.ll
@@ -33,7 +33,7 @@ bb.i4.i:		; preds = %my_calloc.exit.i
 	unreachable
 
 my_calloc.exit5.i:		; preds = %my_calloc.exit.i
-	%.pre.i58 = load %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
+	%.pre.i58 = load %struct.s_net*, %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
 	br label %bb17.i78
 
 bb1.i61:		; preds = %bb4.preheader.i, %bb1.i61
@@ -58,11 +58,11 @@ bb.i1.i68:		; preds = %bb8.i67
 	unreachable
 
 my_malloc.exit.i70:		; preds = %bb8.i67
-	%0 = load %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
+	%0 = load %struct.s_net*, %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
 	br i1 undef, label %bb9.i71, label %bb16.i77
 
 bb9.i71:		; preds = %bb9.i71, %my_malloc.exit.i70
-	%1 = load %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
+	%1 = load %struct.s_net*, %struct.s_net** @net, align 4		; <%struct.s_net*> [#uses=1]
 	br i1 undef, label %bb9.i71, label %bb16.i77
 
 bb16.i77:		; preds = %bb9.i71, %my_malloc.exit.i70, %bb15.preheader.i
diff --git a/test/Transforms/GlobalOpt/2009-11-16-BrokenPerformHeapAllocSRoA.ll b/test/Transforms/GlobalOpt/2009-11-16-BrokenPerformHeapAllocSRoA.ll
index 40862bd038e1..461c25316e91 100644
--- a/test/Transforms/GlobalOpt/2009-11-16-BrokenPerformHeapAllocSRoA.ll
+++ b/test/Transforms/GlobalOpt/2009-11-16-BrokenPerformHeapAllocSRoA.ll
@@ -13,9 +13,9 @@ target triple = "x86_64-apple-darwin10.0"
 
 ; based on linit in office-ispell
 define void @test() nounwind ssp {
-  %1 = load i32* getelementptr inbounds (%struct.hashheader* @hashheader, i64 0, i32 13), align 8 ; <i32> [#uses=1]
+  %1 = load i32, i32* getelementptr inbounds (%struct.hashheader, %struct.hashheader* @hashheader, i64 0, i32 13), align 8 ; <i32> [#uses=1]
   %2 = sext i32 %1 to i64                         ; <i64> [#uses=1]
-  %3 = mul i64 %2, ptrtoint (%struct.strchartype* getelementptr (%struct.strchartype* null, i64 1) to i64) ; <i64> [#uses=1]
+  %3 = mul i64 %2, ptrtoint (%struct.strchartype* getelementptr (%struct.strchartype, %struct.strchartype* null, i64 1) to i64) ; <i64> [#uses=1]
   %4 = tail call i8* @malloc(i64 %3)              ; <i8*> [#uses=1]
 ; CHECK-NOT: call i8* @malloc(i64
   %5 = bitcast i8* %4 to %struct.strchartype*     ; <%struct.strchartype*> [#uses=1]
diff --git a/test/Transforms/GlobalOpt/2009-11-16-MallocSingleStoreToGlobalVar.ll b/test/Transforms/GlobalOpt/2009-11-16-MallocSingleStoreToGlobalVar.ll
index b73f62ba148b..25bb9769f49e 100644
--- a/test/Transforms/GlobalOpt/2009-11-16-MallocSingleStoreToGlobalVar.ll
+++ b/test/Transforms/GlobalOpt/2009-11-16-MallocSingleStoreToGlobalVar.ll
@@ -14,15 +14,15 @@ target triple = "x86_64-apple-darwin10.0"
 ; Derived from @DescribeChannel() in yacr2
 define void @test() nounwind ssp {
   store i64 2335, i64* @channelColumns, align 8
-  %1 = load i64* @channelColumns, align 8         ; <i64> [#uses=1]
+  %1 = load i64, i64* @channelColumns, align 8         ; <i64> [#uses=1]
   %2 = shl i64 %1, 3                              ; <i64> [#uses=1]
   %3 = add i64 %2, 8                              ; <i64> [#uses=1]
   %4 = call noalias i8* @malloc(i64 %3) nounwind  ; <i8*> [#uses=1]
 ; CHECK: call noalias i8* @malloc
   %5 = bitcast i8* %4 to i64*                     ; <i64*> [#uses=1]
   store i64* %5, i64** @TOP, align 8
-  %6 = load i64** @TOP, align 8                   ; <i64*> [#uses=1]
-  %7 = getelementptr inbounds i64* %6, i64 13     ; <i64*> [#uses=1]
+  %6 = load i64*, i64** @TOP, align 8                   ; <i64*> [#uses=1]
+  %7 = getelementptr inbounds i64, i64* %6, i64 13     ; <i64*> [#uses=1]
   store i64 0, i64* %7, align 8
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/2010-02-25-MallocPromote.ll b/test/Transforms/GlobalOpt/2010-02-25-MallocPromote.ll
index 629d57c88424..9f53ce402fd4 100644
--- a/test/Transforms/GlobalOpt/2010-02-25-MallocPromote.ll
+++ b/test/Transforms/GlobalOpt/2010-02-25-MallocPromote.ll
@@ -12,7 +12,7 @@ bb918:
   %malloccall.i10 = call i8* @malloc(i32 16) nounwind ; <i8*> [#uses=1]
   %0 = bitcast i8* %malloccall.i10 to i32*        ; <i32*> [#uses=1]
   store i32* %0, i32** @fixLRBT, align 8
-  %1 = load i32** @fixLRBT, align 8               ; <i32*> [#uses=0]
-  %A = load i32* %1
+  %1 = load i32*, i32** @fixLRBT, align 8               ; <i32*> [#uses=0]
+  %A = load i32, i32* %1
   ret i32 %A
 }
diff --git a/test/Transforms/GlobalOpt/2010-02-26-MallocSROA.ll b/test/Transforms/GlobalOpt/2010-02-26-MallocSROA.ll
index ab7721fd9720..12fa34191926 100644
--- a/test/Transforms/GlobalOpt/2010-02-26-MallocSROA.ll
+++ b/test/Transforms/GlobalOpt/2010-02-26-MallocSROA.ll
@@ -10,7 +10,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define fastcc void @init_net() nounwind {
 entry:
-  %0 = load i32* @numf2s, align 4                 ; <i32> [#uses=1]
+  %0 = load i32, i32* @numf2s, align 4                 ; <i32> [#uses=1]
   %mallocsize2 = shl i32 %0, 4                    ; <i32> [#uses=1]
   %malloccall3 = tail call i8* @malloc(i32 %mallocsize2) nounwind ; <i8*> [#uses=1]
   %1 = bitcast i8* %malloccall3 to %struct.xyz*   ; <%struct.xyz*> [#uses=1]
@@ -20,7 +20,7 @@ entry:
 
 define fastcc void @load_train(i8* %trainfile, i32 %mode, i32 %objects) nounwind {
 entry:
-  %0 = load %struct.xyz** @Y, align 8             ; <%struct.xyz*> [#uses=0]
+  %0 = load %struct.xyz*, %struct.xyz** @Y, align 8             ; <%struct.xyz*> [#uses=0]
   ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/alias-resolve.ll b/test/Transforms/GlobalOpt/alias-resolve.ll
index ebc20c66cf61..090d78455226 100644
--- a/test/Transforms/GlobalOpt/alias-resolve.ll
+++ b/test/Transforms/GlobalOpt/alias-resolve.ll
@@ -13,8 +13,8 @@
 ; CHECK: @weak1 = weak alias void ()* @bar2
 
 @bar4 = private unnamed_addr constant [2 x i8*] zeroinitializer
-@foo4 = linkonce_odr unnamed_addr alias getelementptr inbounds ([2 x i8*]* @bar4, i32 0, i32 1)
-; CHECK: @foo4 = linkonce_odr unnamed_addr alias getelementptr inbounds ([2 x i8*]* @bar4, i32 0, i32 1)
+@foo4 = linkonce_odr unnamed_addr alias getelementptr inbounds ([2 x i8*], [2 x i8*]* @bar4, i32 0, i32 1)
+; CHECK: @foo4 = linkonce_odr unnamed_addr alias getelementptr inbounds ([2 x i8*], [2 x i8*]* @bar4, i32 0, i32 1)
 
 define void @bar2() {
   ret void
diff --git a/test/Transforms/GlobalOpt/array-elem-refs.ll b/test/Transforms/GlobalOpt/array-elem-refs.ll
index ec472b0e99f2..c31965b656d3 100644
--- a/test/Transforms/GlobalOpt/array-elem-refs.ll
+++ b/test/Transforms/GlobalOpt/array-elem-refs.ll
@@ -4,16 +4,16 @@ target triple = "x86_64-unknown-linux-gnu"
 
 %struct.S = type { i8, i8 }
 
-@c = internal global i8** bitcast (i8* getelementptr (i8* bitcast ([8 x i8*]* @b to i8*), i64 48) to i8**), align 8
-@b = internal global [8 x i8*] [i8* null, i8* null, i8* null, i8* null, i8* null, i8* null, i8* getelementptr inbounds (%struct.S* @a, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S* @a, i32 0, i32 0), i64 1)], align 16
+@c = internal global i8** bitcast (i8* getelementptr (i8, i8* bitcast ([8 x i8*]* @b to i8*), i64 48) to i8**), align 8
+@b = internal global [8 x i8*] [i8* null, i8* null, i8* null, i8* null, i8* null, i8* null, i8* getelementptr inbounds (%struct.S, %struct.S* @a, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S, %struct.S* @a, i32 0, i32 0), i64 1)], align 16
 @a = internal global %struct.S zeroinitializer, align 1
 
 ; Function Attrs: nounwind uwtable
 define signext i8 @foo() #0 {
 entry:
-  %0 = load i8*** @c, align 8
-  %1 = load i8** %0, align 8
-  %2 = load i8* %1, align 1
+  %0 = load i8**, i8*** @c, align 8
+  %1 = load i8*, i8** %0, align 8
+  %2 = load i8, i8* %1, align 1
   ret i8 %2
 
 ; CHECK-LABEL: @foo
diff --git a/test/Transforms/GlobalOpt/atomic.ll b/test/Transforms/GlobalOpt/atomic.ll
index ac05bfd68d9d..563c1fec7d27 100644
--- a/test/Transforms/GlobalOpt/atomic.ll
+++ b/test/Transforms/GlobalOpt/atomic.ll
@@ -8,7 +8,7 @@
 
 define void @test1() {
 entry:
-  %0 = load atomic i8* bitcast (i64* @GV1 to i8*) acquire, align 8
+  %0 = load atomic i8, i8* bitcast (i64* @GV1 to i8*) acquire, align 8
   ret void
 }
 
@@ -20,6 +20,6 @@ entry:
 }
 define i32 @test2b() {
 entry:
-  %atomic-load = load atomic i32* @GV2 seq_cst, align 4
+  %atomic-load = load atomic i32, i32* @GV2 seq_cst, align 4
   ret i32 %atomic-load
 }
diff --git a/test/Transforms/GlobalOpt/basictest.ll b/test/Transforms/GlobalOpt/basictest.ll
index 4332d3dd38c2..4c25e4f9f407 100644
--- a/test/Transforms/GlobalOpt/basictest.ll
+++ b/test/Transforms/GlobalOpt/basictest.ll
@@ -3,7 +3,7 @@
 @X = internal global i32 4              ; <i32*> [#uses=1]
 
 define i32 @foo() {
-        %V = load i32* @X               ; <i32> [#uses=1]
+        %V = load i32, i32* @X               ; <i32> [#uses=1]
         ret i32 %V
 }
 
diff --git a/test/Transforms/GlobalOpt/blockaddress.ll b/test/Transforms/GlobalOpt/blockaddress.ll
index f7f830869b84..12e09fcd4815 100644
--- a/test/Transforms/GlobalOpt/blockaddress.ll
+++ b/test/Transforms/GlobalOpt/blockaddress.ll
@@ -13,6 +13,9 @@ define void @f() {
 }
 
 define void @g() {
+entry:
+  br label %here
+
 ; CHECK-LABEL: @g(
 
 here:
diff --git a/test/Transforms/GlobalOpt/constantexpr-dangle.ll b/test/Transforms/GlobalOpt/constantexpr-dangle.ll
index be13a9811898..fbbfe69c15ab 100644
--- a/test/Transforms/GlobalOpt/constantexpr-dangle.ll
+++ b/test/Transforms/GlobalOpt/constantexpr-dangle.ll
@@ -6,7 +6,7 @@ define internal float @foo() {
 }
 
 define float @bar() {
-        %tmp1 = call float (...)* bitcast (float ()* @foo to float (...)*)( )
+        %tmp1 = call float (...) bitcast (float ()* @foo to float (...)*)( )
         %tmp2 = fmul float %tmp1, 1.000000e+01           ; <float> [#uses=1]
         ret float %tmp2
 }
diff --git a/test/Transforms/GlobalOpt/constantfold-initializers.ll b/test/Transforms/GlobalOpt/constantfold-initializers.ll
index 36de19c1cd84..d0c18812cd94 100644
--- a/test/Transforms/GlobalOpt/constantfold-initializers.ll
+++ b/test/Transforms/GlobalOpt/constantfold-initializers.ll
@@ -5,7 +5,7 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 @.str91250 = global [3 x i8] zeroinitializer
 
 ; CHECK: @A = global i1 false
-@A = global i1 icmp ne (i64 sub nsw (i64 ptrtoint (i8* getelementptr inbounds ([3 x i8]* @.str91250, i64 0, i64 1) to i64), i64 ptrtoint ([3 x i8]* @.str91250 to i64)), i64 1)
+@A = global i1 icmp ne (i64 sub nsw (i64 ptrtoint (i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str91250, i64 0, i64 1) to i64), i64 ptrtoint ([3 x i8]* @.str91250 to i64)), i64 1)
 
 ; PR11352
 
@@ -15,13 +15,13 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; PR12642
 %PR12642.struct = type { i8 }
 @PR12642.s = global <{}> zeroinitializer, align 1
-@PR12642.p = constant %PR12642.struct* bitcast (i8* getelementptr (i8* bitcast (<{}>* @PR12642.s to i8*), i64 1) to %PR12642.struct*), align 8
+@PR12642.p = constant %PR12642.struct* bitcast (i8* getelementptr (i8, i8* bitcast (<{}>* @PR12642.s to i8*), i64 1) to %PR12642.struct*), align 8
 
 define internal void @test1() {
 entry:
-  store i32 1, i32* getelementptr inbounds ([2 x i32]* @xs, i64 0, i64 0)
-  %0 = load i32* getelementptr inbounds ([2 x i32]* @xs, i32 0, i64 0), align 4
-  store i32 %0, i32* getelementptr inbounds ([2 x i32]* @xs, i64 0, i64 1)
+  store i32 1, i32* getelementptr inbounds ([2 x i32], [2 x i32]* @xs, i64 0, i64 0)
+  %0 = load i32, i32* getelementptr inbounds ([2 x i32], [2 x i32]* @xs, i32 0, i64 0), align 4
+  store i32 %0, i32* getelementptr inbounds ([2 x i32], [2 x i32]* @xs, i64 0, i64 1)
   ret void
 }
 
@@ -36,15 +36,15 @@ entry:
 
 define internal i32 @test2_helper(%closure* %this, i32 %b) {
 entry:
-  %0 = getelementptr inbounds %closure* %this, i32 0, i32 0
-  %1 = load i32* %0, align 4
+  %0 = getelementptr inbounds %closure, %closure* %this, i32 0, i32 0
+  %1 = load i32, i32* %0, align 4
   %add = add nsw i32 %1, %b
   ret i32 %add
 }
 
 define internal void @test2() {
 entry:
-  store i32 4, i32* getelementptr inbounds (%closure* @f, i32 0, i32 0)
+  store i32 4, i32* getelementptr inbounds (%closure, %closure* @f, i32 0, i32 0)
   %call = call i32 @test2_helper(%closure* @f, i32 9)
   store i32 %call, i32* @m, align 4
   ret void
@@ -85,7 +85,7 @@ entry:
 @test6_v2 = global i32 0, align 4
 ; CHECK: @test6_v2 = global i32 42, align 4
 define internal void @test6() {
-  %load = load { i32, i32 }* @test6_v1, align 8
+  %load = load { i32, i32 }, { i32, i32 }* @test6_v1, align 8
   %xv0 = extractvalue { i32, i32 } %load, 0
   %iv = insertvalue { i32, i32 } %load, i32 %xv0, 1
   %xv1 = extractvalue { i32, i32 } %iv, 1
diff --git a/test/Transforms/GlobalOpt/crash-2.ll b/test/Transforms/GlobalOpt/crash-2.ll
index 684f6cee180b..748fb027c7d7 100644
--- a/test/Transforms/GlobalOpt/crash-2.ll
+++ b/test/Transforms/GlobalOpt/crash-2.ll
@@ -11,9 +11,9 @@
 @g_71 = internal global %struct.S0.1.7.13 { i8 1, i8 -93, i8 58, i8 -1, i16 -5, [2 x i8] undef }, align 4
 @g_114 = external global i8, align 1
 @g_30 = external global { i32, i8, i32, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8, i8 }, align 4
-@g_271 = internal global [7 x [6 x [5 x i8*]]] [[6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* null, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* @g_114, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25], [5 x i8*] [i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* @g_25, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* null, i8* @g_25, i8* @g_25, i8* @g_25, i8* null], [5 x i8*] [i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* null, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* null, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null], [5 x i8*] [i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* @g_25], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25], [5 x i8*] [i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_114, i8* @g_25, i8* @g_25, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_114], [5 x i8*] [i8* @g_25, i8* null, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* null], [5 x i8*] [i8* @g_114, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_25]], [6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_114, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_25, i8* @g_25, i8* @g_114], [5 x i8*] [i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25], [5 x i8*] [i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* @g_25, i8* getelementptr (i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25]]], align 4
+@g_271 = internal global [7 x [6 x [5 x i8*]]] [[6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* null, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* @g_114, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25], [5 x i8*] [i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* @g_25, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* null, i8* @g_25, i8* @g_25, i8* @g_25, i8* null], [5 x i8*] [i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* null, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* null, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null], [5 x i8*] [i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* @g_25], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25], [5 x i8*] [i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_114, i8* @g_25, i8* @g_25, i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_114], [5 x i8*] [i8* @g_25, i8* null, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* null], [5 x i8*] [i8* @g_114, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_25]], [6 x [5 x i8*]] [[5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_114, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0)], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25]], [6 x [5 x i8*]] [[5 x i8*] [i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* null], [5 x i8*] [i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* @g_25, i8* @g_25, i8* @g_114], [5 x i8*] [i8* null, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_25, i8* null, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_114, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* @g_114, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1)], [5 x i8*] [i8* @g_25, i8* @g_25, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25], [5 x i8*] [i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25, i8* @g_25, i8* getelementptr (i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), i64 1), i8* @g_25]]], align 4
 
 define i32 @func() {
-  %tmp = load i8* getelementptr inbounds (%struct.S0.1.7.13* @g_71, i32 0, i32 0), align 1
+  %tmp = load i8, i8* getelementptr inbounds (%struct.S0.1.7.13, %struct.S0.1.7.13* @g_71, i32 0, i32 0), align 1
   ret i32 0
 }
diff --git a/test/Transforms/GlobalOpt/crash.ll b/test/Transforms/GlobalOpt/crash.ll
index 80c777ccabc1..8e39931c8525 100644
--- a/test/Transforms/GlobalOpt/crash.ll
+++ b/test/Transforms/GlobalOpt/crash.ll
@@ -11,7 +11,7 @@ target triple = "i386-apple-darwin9.8"
 
 define internal void @_GLOBAL__I__ZN21btConeTwistConstraintC2Ev() nounwind section "__TEXT,__StaticInit,regular,pure_instructions" {
 entry:
-  store float 1.0, float* getelementptr inbounds (%struct.btSimdScalar* @_ZL6vTwist, i32 0, i32 0, i32 0, i32 3), align 4
+  store float 1.0, float* getelementptr inbounds (%struct.btSimdScalar, %struct.btSimdScalar* @_ZL6vTwist, i32 0, i32 0, i32 0, i32 3), align 4
   ret void
 }
 
@@ -31,11 +31,11 @@ entry:
   unreachable
 
 bb.nph.i: 
-  %scevgep.i539 = getelementptr i8* %C, i64 4
+  %scevgep.i539 = getelementptr i8, i8* %C, i64 4
   unreachable
 
 xx:
-  %E = load %T** @switch_inf, align 8 
+  %E = load %T*, %T** @switch_inf, align 8 
   unreachable
 }
 
@@ -46,7 +46,7 @@ declare noalias i8* @malloc(i64) nounwind
 @permute_bitrev.bitrev = internal global i32* null, align 8
 define void @permute_bitrev() nounwind {
 entry:
-  %tmp = load i32** @permute_bitrev.bitrev, align 8
+  %tmp = load i32*, i32** @permute_bitrev.bitrev, align 8
   %conv = sext i32 0 to i64
   %mul = mul i64 %conv, 4
   %call = call i8* @malloc(i64 %mul)
@@ -60,7 +60,7 @@ entry:
 
 @data8 = internal global [8000 x i8] zeroinitializer, align 16
 define void @memset_with_strange_user() ssp {
-  call void @llvm.memset.p0i8.i64(i8* getelementptr inbounds ([8000 x i8]* @data8, i64 0, i64 0), i8 undef, i64 ptrtoint (i8* getelementptr ([8000 x i8]* @data8, i64 1, i64 sub (i64 0, i64 ptrtoint ([8000 x i8]* @data8 to i64))) to i64), i32 16, i1 false)
+  call void @llvm.memset.p0i8.i64(i8* getelementptr inbounds ([8000 x i8], [8000 x i8]* @data8, i64 0, i64 0), i8 undef, i64 ptrtoint (i8* getelementptr ([8000 x i8], [8000 x i8]* @data8, i64 1, i64 sub (i64 0, i64 ptrtoint ([8000 x i8]* @data8 to i64))) to i64), i32 16, i1 false)
   ret void
 }
 declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
@@ -72,7 +72,7 @@ declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 
 define void @icmp_user_of_stored_once() nounwind ssp {
 entry:
-  %tmp4 = load i32*** @g_52, align 8
+  %tmp4 = load i32**, i32*** @g_52, align 8
   store i32** @g_90, i32*** @g_52
   %cmp17 = icmp ne i32*** undef, @g_52
   ret void
diff --git a/test/Transforms/GlobalOpt/ctor-list-opt-constexpr.ll b/test/Transforms/GlobalOpt/ctor-list-opt-constexpr.ll
index dd86f01924a5..0c3ff68a437e 100644
--- a/test/Transforms/GlobalOpt/ctor-list-opt-constexpr.ll
+++ b/test/Transforms/GlobalOpt/ctor-list-opt-constexpr.ll
@@ -16,7 +16,7 @@ target triple = "x86_64-apple-darwin10.0.0"
 ; arbitrary constant expression, the code generator can't handle it.
 define internal void @init1() {
 entry:
-  %tmp = getelementptr inbounds %struct.foo* @X, i32 0, i32 0
+  %tmp = getelementptr inbounds %struct.foo, %struct.foo* @X, i32 0, i32 0
   store i32* inttoptr (i64 sdiv (i64 ptrtoint (i32* @G to i64), i64 ptrtoint (i32* @H to i64)) to i32*), i32** %tmp, align 8
   ret void
 }
@@ -26,7 +26,7 @@ entry:
 ; PR11705 - ptrtoint isn't safe in general in global initializers.
 define internal void @init2() {
 entry:
-  %tmp = getelementptr inbounds %struct.bar* @X2, i32 0, i32 0
+  %tmp = getelementptr inbounds %struct.bar, %struct.bar* @X2, i32 0, i32 0
   store i128 ptrtoint (i32* @G to i128), i128* %tmp, align 16
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/ctor-list-opt-inbounds.ll b/test/Transforms/GlobalOpt/ctor-list-opt-inbounds.ll
index 9b11985693a1..63dc783ae196 100644
--- a/test/Transforms/GlobalOpt/ctor-list-opt-inbounds.ll
+++ b/test/Transforms/GlobalOpt/ctor-list-opt-inbounds.ll
@@ -13,11 +13,11 @@
 @I = global i32 90
 
 define internal void @CTOR() {
-  store i32 1, i32* getelementptr ([6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
-  store i32 2, i32* getelementptr inbounds ([6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
-  %t = load i32* getelementptr ([6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
+  store i32 1, i32* getelementptr ([6 x [5 x i32]], [6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
+  store i32 2, i32* getelementptr inbounds ([6 x [5 x i32]], [6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
+  %t = load i32, i32* getelementptr ([6 x [5 x i32]], [6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
   store i32 %t, i32* @H
-  %s = load i32* getelementptr inbounds ([6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
+  %s = load i32, i32* getelementptr inbounds ([6 x [5 x i32]], [6 x [5 x i32]]* @G, i64 0, i64 0, i64 0)
   store i32 %s, i32* @I
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/ctor-list-opt.ll b/test/Transforms/GlobalOpt/ctor-list-opt.ll
index 450bdb830284..95e7d4d1ab5b 100644
--- a/test/Transforms/GlobalOpt/ctor-list-opt.ll
+++ b/test/Transforms/GlobalOpt/ctor-list-opt.ll
@@ -51,16 +51,16 @@ T:		; preds = %Cont
 }
 
 define internal void @CTOR4() {
-	%X = load i32* @G3		; <i32> [#uses=1]
+	%X = load i32, i32* @G3		; <i32> [#uses=1]
 	%Y = add i32 %X, 123		; <i32> [#uses=1]
 	store i32 %Y, i32* @G3
 	ret void
 }
 
 define internal void @CTOR5() {
-	%X.2p = getelementptr inbounds { i32, [2 x i32] }* @X, i32 0, i32 1, i32 0		; <i32*> [#uses=2]
-	%X.2 = load i32* %X.2p		; <i32> [#uses=1]
-	%X.1p = getelementptr inbounds { i32, [2 x i32] }* @X, i32 0, i32 0		; <i32*> [#uses=1]
+	%X.2p = getelementptr inbounds { i32, [2 x i32] }, { i32, [2 x i32] }* @X, i32 0, i32 1, i32 0		; <i32*> [#uses=2]
+	%X.2 = load i32, i32* %X.2p		; <i32> [#uses=1]
+	%X.1p = getelementptr inbounds { i32, [2 x i32] }, { i32, [2 x i32] }* @X, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 %X.2, i32* %X.1p
 	store i32 42, i32* %X.2p
 	ret void
@@ -68,9 +68,9 @@ define internal void @CTOR5() {
 
 define internal void @CTOR6() {
 	%A = alloca i32		; <i32*> [#uses=2]
-	%y = load i32* @Y		; <i32> [#uses=1]
+	%y = load i32, i32* @Y		; <i32> [#uses=1]
 	store i32 %y, i32* %A
-	%Av = load i32* %A		; <i32> [#uses=1]
+	%Av = load i32, i32* %A		; <i32> [#uses=1]
 	%Av1 = add i32 %Av, 1		; <i32> [#uses=1]
 	store i32 %Av1, i32* @Y
 	ret void
@@ -95,7 +95,7 @@ define internal void @CTOR8() {
 }
 
 define i1 @accessor() {
-	%V = load i1* @CTORGV		; <i1> [#uses=1]
+	%V = load i1, i1* @CTORGV		; <i1> [#uses=1]
 	ret i1 %V
 }
 
@@ -107,9 +107,9 @@ define i1 @accessor() {
 define internal void @CTOR9() {
 entry:
   %0 = bitcast %struct.B* @GV1 to i8*
-  %1 = getelementptr inbounds i8* %0, i64 16
+  %1 = getelementptr inbounds i8, i8* %0, i64 16
   %2 = bitcast i8* %1 to %struct.A*
   %3 = bitcast %struct.B* @GV1 to i8***
-  store i8** getelementptr inbounds ([3 x i8*]* @GV2, i64 1, i64 0), i8*** %3
+  store i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @GV2, i64 1, i64 0), i8*** %3
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/cxx-dtor.ll b/test/Transforms/GlobalOpt/cxx-dtor.ll
index 7c6ae78d1bc0..c94c26b53485 100644
--- a/test/Transforms/GlobalOpt/cxx-dtor.ll
+++ b/test/Transforms/GlobalOpt/cxx-dtor.ll
@@ -11,7 +11,7 @@
 ; CHECK-NOT: call i32 @__cxa_atexit
 
 define internal void @__cxx_global_var_init() nounwind section "__TEXT,__StaticInit,regular,pure_instructions" {
-  %1 = call i32 @__cxa_atexit(void (i8*)* bitcast (void (%struct.A*)* @_ZN1AD1Ev to void (i8*)*), i8* getelementptr inbounds (%struct.A* @a, i32 0, i32 0), i8* bitcast (i8** @__dso_handle to i8*))
+  %1 = call i32 @__cxa_atexit(void (i8*)* bitcast (void (%struct.A*)* @_ZN1AD1Ev to void (i8*)*), i8* getelementptr inbounds (%struct.A, %struct.A* @a, i32 0, i32 0), i8* bitcast (i8** @__dso_handle to i8*))
   ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/deadfunction.ll b/test/Transforms/GlobalOpt/deadfunction.ll
index 5e003c63f77d..5771c4ccfb75 100644
--- a/test/Transforms/GlobalOpt/deadfunction.ll
+++ b/test/Transforms/GlobalOpt/deadfunction.ll
@@ -10,8 +10,8 @@ declare void @bb()
 define internal void @test(i32 %n) nounwind noinline {
 entry:
   %idxprom = sext i32 %n to i64
-  %arrayidx = getelementptr inbounds [3 x i8*]* @test.x, i64 0, i64 %idxprom
-  %0 = load i8** %arrayidx, align 8
+  %arrayidx = getelementptr inbounds [3 x i8*], [3 x i8*]* @test.x, i64 0, i64 %idxprom
+  %0 = load i8*, i8** %arrayidx, align 8
   indirectbr i8* %0, [label %a, label %b, label %c]
 
 a:
diff --git a/test/Transforms/GlobalOpt/deadglobal-2.ll b/test/Transforms/GlobalOpt/deadglobal-2.ll
index 4f8181983094..6b8717ee1831 100644
--- a/test/Transforms/GlobalOpt/deadglobal-2.ll
+++ b/test/Transforms/GlobalOpt/deadglobal-2.ll
@@ -5,7 +5,7 @@
 @G = internal global [4 x i32] zeroinitializer
 
 define void @foo(i32 %X) {
-	%Ptr = getelementptr [4 x i32]* @G, i32 0, i32 %X
+	%Ptr = getelementptr [4 x i32], [4 x i32]* @G, i32 0, i32 %X
 	store i32 1, i32* %Ptr
 	ret void
 }
diff --git a/test/Transforms/GlobalOpt/externally-initialized-global-ctr.ll b/test/Transforms/GlobalOpt/externally-initialized-global-ctr.ll
index 675211b64b46..2e22ff064444 100644
--- a/test/Transforms/GlobalOpt/externally-initialized-global-ctr.ll
+++ b/test/Transforms/GlobalOpt/externally-initialized-global-ctr.ll
@@ -7,14 +7,14 @@
 @_ZL14buttonInitData = internal global [1 x %struct.ButtonInitData] zeroinitializer, align 4
 
 @"\01L_OBJC_METH_VAR_NAME_40" = internal global [7 x i8] c"print:\00", section "__TEXT,__objc_methname,cstring_literals", align 1
-@"\01L_OBJC_SELECTOR_REFERENCES_41" = internal externally_initialized  global i8* getelementptr inbounds ([7 x i8]* @"\01L_OBJC_METH_VAR_NAME_40", i32 0, i32 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@"\01L_OBJC_SELECTOR_REFERENCES_41" = internal externally_initialized  global i8* getelementptr inbounds ([7 x i8], [7 x i8]* @"\01L_OBJC_METH_VAR_NAME_40", i32 0, i32 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
 
 @llvm.global_ctors = appending global [1 x { i32, void ()* }] [{ i32, void ()* } { i32 65535, void ()* @_GLOBAL__I_a }]
-@llvm.used = appending global [2 x i8*] [i8* getelementptr inbounds ([7 x i8]* @"\01L_OBJC_METH_VAR_NAME_40", i32 0, i32 0),  i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_41" to i8*)]
+@llvm.used = appending global [2 x i8*] [i8* getelementptr inbounds ([7 x i8], [7 x i8]* @"\01L_OBJC_METH_VAR_NAME_40", i32 0, i32 0),  i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_41" to i8*)]
 
 define internal void @__cxx_global_var_init() section "__TEXT,__StaticInit,regular,pure_instructions" {
-  %1 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_41", !invariant.load !2009
-  store i8* %1, i8** getelementptr inbounds ([1 x %struct.ButtonInitData]* @_ZL14buttonInitData, i32 0, i32 0, i32 0), align 4
+  %1 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_41", !invariant.load !2009
+  store i8* %1, i8** getelementptr inbounds ([1 x %struct.ButtonInitData], [1 x %struct.ButtonInitData]* @_ZL14buttonInitData, i32 0, i32 0, i32 0), align 4
   ret void
 }
 
@@ -26,8 +26,8 @@ define internal void @_GLOBAL__I_a() section "__TEXT,__StaticInit,regular,pure_i
 declare void @test(i8*)
 
 define void @print() {
-; CHECK: %1 = load i8** getelementptr inbounds ([1 x %struct.ButtonInitData]* @_ZL14buttonInitData, i32 0, i32 0, i32 0), align 4
-  %1 = load i8** getelementptr inbounds ([1 x %struct.ButtonInitData]* @_ZL14buttonInitData, i32 0, i32 0, i32 0), align 4
+; CHECK: %1 = load i8*, i8** @_ZL14buttonInitData.0.0, align 4
+  %1 = load i8*, i8** getelementptr inbounds ([1 x %struct.ButtonInitData], [1 x %struct.ButtonInitData]* @_ZL14buttonInitData, i32 0, i32 0, i32 0), align 4
   call void @test(i8* %1)
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/fastcc.ll b/test/Transforms/GlobalOpt/fastcc.ll
index 76122b203d79..26398d2bb900 100644
--- a/test/Transforms/GlobalOpt/fastcc.ll
+++ b/test/Transforms/GlobalOpt/fastcc.ll
@@ -2,13 +2,13 @@
 
 define internal i32 @f(i32* %m) {
 ; CHECK-LABEL: define internal fastcc i32 @f
-  %v = load i32* %m
+  %v = load i32, i32* %m
   ret i32 %v
 }
 
 define internal x86_thiscallcc i32 @g(i32* %m) {
 ; CHECK-LABEL: define internal fastcc i32 @g
-  %v = load i32* %m
+  %v = load i32, i32* %m
   ret i32 %v
 }
 
@@ -16,13 +16,13 @@ define internal x86_thiscallcc i32 @g(i32* %m) {
 ; convention.
 define internal coldcc i32 @h(i32* %m) {
 ; CHECK-LABEL: define internal coldcc i32 @h
-  %v = load i32* %m
+  %v = load i32, i32* %m
   ret i32 %v
 }
 
 define internal i32 @j(i32* %m) {
 ; CHECK-LABEL: define internal i32 @j
-  %v = load i32* %m
+  %v = load i32, i32* %m
   ret i32 %v
 }
 
diff --git a/test/Transforms/GlobalOpt/globalsra-partial.ll b/test/Transforms/GlobalOpt/globalsra-partial.ll
index 06485b53e0eb..9abcce6f4931 100644
--- a/test/Transforms/GlobalOpt/globalsra-partial.ll
+++ b/test/Transforms/GlobalOpt/globalsra-partial.ll
@@ -6,19 +6,19 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
 @G = internal global { i32, [4 x float] } zeroinitializer               ; <{ i32, [4 x float] }*> [#uses=3]
 
 define void @onlystore() {
-        store i32 12345, i32* getelementptr ({ i32, [4 x float] }* @G, i32 0, i32 0)
+        store i32 12345, i32* getelementptr ({ i32, [4 x float] }, { i32, [4 x float] }* @G, i32 0, i32 0)
         ret void
 }
 
 define void @storeinit(i32 %i) {
-        %Ptr = getelementptr { i32, [4 x float] }* @G, i32 0, i32 1, i32 %i             ; <float*> [#uses=1]
+        %Ptr = getelementptr { i32, [4 x float] }, { i32, [4 x float] }* @G, i32 0, i32 1, i32 %i             ; <float*> [#uses=1]
         store float 1.000000e+00, float* %Ptr
         ret void
 }
 
 define float @readval(i32 %i) {
-        %Ptr = getelementptr { i32, [4 x float] }* @G, i32 0, i32 1, i32 %i             ; <float*> [#uses=1]
-        %V = load float* %Ptr           ; <float> [#uses=1]
+        %Ptr = getelementptr { i32, [4 x float] }, { i32, [4 x float] }* @G, i32 0, i32 1, i32 %i             ; <float*> [#uses=1]
+        %V = load float, float* %Ptr           ; <float> [#uses=1]
         ret float %V
 }
 
diff --git a/test/Transforms/GlobalOpt/globalsra-unknown-index.ll b/test/Transforms/GlobalOpt/globalsra-unknown-index.ll
index cc655e9a2d1d..5194b2ca7413 100644
--- a/test/Transforms/GlobalOpt/globalsra-unknown-index.ll
+++ b/test/Transforms/GlobalOpt/globalsra-unknown-index.ll
@@ -12,30 +12,30 @@
 @addr = external global i8
 
 define void @frob() {
-  store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 ptrtoint (i8* @addr to i64)), align 4
+  store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 ptrtoint (i8* @addr to i64)), align 4
   ret void
 }
 define i32 @borf(i64 %i, i64 %j) {
-  %p = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0
-  %a = load i32* %p
-  %q = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 0
-  %b = load i32* %q
+  %p = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0
+  %a = load i32, i32* %p
+  %q = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 0
+  %b = load i32, i32* %q
   %c = add i32 %a, %b
   ret i32 %c
 }
 define i32 @borg(i64 %i, i64 %j) {
-  %p = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 1
-  %a = load i32* %p
-  %q = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 1
-  %b = load i32* %q
+  %p = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 1
+  %a = load i32, i32* %p
+  %q = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 1
+  %b = load i32, i32* %q
   %c = add i32 %a, %b
   ret i32 %c
 }
 define i32 @borh(i64 %i, i64 %j) {
-  %p = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 2
-  %a = load i32* %p
-  %q = getelementptr inbounds [3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 2
-  %b = load i32* %q
+  %p = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 2
+  %a = load i32, i32* %p
+  %q = getelementptr inbounds [3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 2
+  %b = load i32, i32* %q
   %c = add i32 %a, %b
   ret i32 %c
 }
diff --git a/test/Transforms/GlobalOpt/globalsra.ll b/test/Transforms/GlobalOpt/globalsra.ll
index 6d8f220d12b1..d593fcd9e15e 100644
--- a/test/Transforms/GlobalOpt/globalsra.ll
+++ b/test/Transforms/GlobalOpt/globalsra.ll
@@ -7,18 +7,18 @@ target datalayout = "E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:1
     { double } { double 1.727000e+01 } }                ; <{ i32, float, { double } }*> [#uses=3]
 
 define void @onlystore() {
-        store i32 123, i32* getelementptr ({ i32, float, { double } }* @G, i32 0, i32 0)
+        store i32 123, i32* getelementptr ({ i32, float, { double } }, { i32, float, { double } }* @G, i32 0, i32 0)
         ret void
 }
 
 define float @storeinit() {
-        store float 1.000000e+00, float* getelementptr ({ i32, float, { double } }* @G, i32 0, i32 1)
-        %X = load float* getelementptr ({ i32, float, { double } }* @G, i32 0, i32 1)           ; <float> [#uses=1]
+        store float 1.000000e+00, float* getelementptr ({ i32, float, { double } }, { i32, float, { double } }* @G, i32 0, i32 1)
+        %X = load float, float* getelementptr ({ i32, float, { double } }, { i32, float, { double } }* @G, i32 0, i32 1)           ; <float> [#uses=1]
         ret float %X
 }
 
 define double @constantize() {
-        %X = load double* getelementptr ({ i32, float, { double } }* @G, i32 0, i32 2, i32 0)           ; <double> [#uses=1]
+        %X = load double, double* getelementptr ({ i32, float, { double } }, { i32, float, { double } }* @G, i32 0, i32 2, i32 0)           ; <double> [#uses=1]
         ret double %X
 }
 
diff --git a/test/Transforms/GlobalOpt/heap-sra-1.ll b/test/Transforms/GlobalOpt/heap-sra-1.ll
index 9d5148f9be6f..6035eaaefccd 100644
--- a/test/Transforms/GlobalOpt/heap-sra-1.ll
+++ b/test/Transforms/GlobalOpt/heap-sra-1.ll
@@ -19,14 +19,14 @@ declare noalias i8* @malloc(i64)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%0 = load %struct.foo** @X, align 4		
+	%0 = load %struct.foo*, %struct.foo** @X, align 4		
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %indvar.next, %bb1 ]
 	%sum.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %3, %bb1 ]
-	%1 = getelementptr %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
-	%2 = load i32* %1, align 4
+	%1 = getelementptr %struct.foo, %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
+	%2 = load i32, i32* %1, align 4
 	%3 = add i32 %2, %sum.0.reg2mem.0	
 	%indvar.next = add i32 %i.0.reg2mem.0, 1	
 	%exitcond = icmp eq i32 %indvar.next, 1200		
diff --git a/test/Transforms/GlobalOpt/heap-sra-2.ll b/test/Transforms/GlobalOpt/heap-sra-2.ll
index fa8c36281ee9..d66c627b184c 100644
--- a/test/Transforms/GlobalOpt/heap-sra-2.ll
+++ b/test/Transforms/GlobalOpt/heap-sra-2.ll
@@ -10,7 +10,7 @@ define void @bar(i32 %Size) nounwind noinline {
 entry:
 	%malloccall = tail call i8* @malloc(i64 8000000) ; <i8*> [#uses=1]
 	%0 = bitcast i8* %malloccall to [1000000 x %struct.foo]* ; <[1000000 x %struct.foo]*> [#uses=1]
-	%.sub = getelementptr [1000000 x %struct.foo]* %0, i32 0, i32 0		; <%struct.foo*> [#uses=1]
+	%.sub = getelementptr [1000000 x %struct.foo], [1000000 x %struct.foo]* %0, i32 0, i32 0		; <%struct.foo*> [#uses=1]
 	store %struct.foo* %.sub, %struct.foo** @X, align 4
 	ret void
 }
@@ -19,14 +19,14 @@ declare noalias i8* @malloc(i64)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%0 = load %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
+	%0 = load %struct.foo*, %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %indvar.next, %bb1 ]		; <i32> [#uses=2]
 	%sum.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %3, %bb1 ]		; <i32> [#uses=1]
-	%1 = getelementptr %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0		; <i32*> [#uses=1]
-	%2 = load i32* %1, align 4		; <i32> [#uses=1]
+	%1 = getelementptr %struct.foo, %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0		; <i32*> [#uses=1]
+	%2 = load i32, i32* %1, align 4		; <i32> [#uses=1]
 	%3 = add i32 %2, %sum.0.reg2mem.0		; <i32> [#uses=2]
 	%indvar.next = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%exitcond = icmp eq i32 %indvar.next, 1200		; <i1> [#uses=1]
diff --git a/test/Transforms/GlobalOpt/heap-sra-3.ll b/test/Transforms/GlobalOpt/heap-sra-3.ll
index e7a877cda446..6a34364b49fe 100644
--- a/test/Transforms/GlobalOpt/heap-sra-3.ll
+++ b/test/Transforms/GlobalOpt/heap-sra-3.ll
@@ -20,14 +20,14 @@ declare noalias i8* @malloc(i64)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%0 = load %struct.foo** @X, align 4		
+	%0 = load %struct.foo*, %struct.foo** @X, align 4		
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %indvar.next, %bb1 ]
 	%sum.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %3, %bb1 ]
-	%1 = getelementptr %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
-	%2 = load i32* %1, align 4
+	%1 = getelementptr %struct.foo, %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
+	%2 = load i32, i32* %1, align 4
 	%3 = add i32 %2, %sum.0.reg2mem.0	
 	%indvar.next = add i32 %i.0.reg2mem.0, 1	
 	%exitcond = icmp eq i32 %indvar.next, 1200		
diff --git a/test/Transforms/GlobalOpt/heap-sra-4.ll b/test/Transforms/GlobalOpt/heap-sra-4.ll
index d5a58288e1a2..2176b9fcbee2 100644
--- a/test/Transforms/GlobalOpt/heap-sra-4.ll
+++ b/test/Transforms/GlobalOpt/heap-sra-4.ll
@@ -20,14 +20,14 @@ declare noalias i8* @malloc(i64)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%0 = load %struct.foo** @X, align 4		
+	%0 = load %struct.foo*, %struct.foo** @X, align 4		
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %indvar.next, %bb1 ]
 	%sum.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %3, %bb1 ]
-	%1 = getelementptr %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
-	%2 = load i32* %1, align 4
+	%1 = getelementptr %struct.foo, %struct.foo* %0, i32 %i.0.reg2mem.0, i32 0
+	%2 = load i32, i32* %1, align 4
 	%3 = add i32 %2, %sum.0.reg2mem.0	
 	%indvar.next = add i32 %i.0.reg2mem.0, 1	
 	%exitcond = icmp eq i32 %indvar.next, 1200		
diff --git a/test/Transforms/GlobalOpt/heap-sra-phi.ll b/test/Transforms/GlobalOpt/heap-sra-phi.ll
index 123ad851f717..bf9ca7b5311e 100644
--- a/test/Transforms/GlobalOpt/heap-sra-phi.ll
+++ b/test/Transforms/GlobalOpt/heap-sra-phi.ll
@@ -9,7 +9,7 @@ define void @bar(i32 %Size) nounwind noinline {
 entry:
 	%malloccall = tail call i8* @malloc(i64 8000000) ; <i8*> [#uses=1]
 	%tmp = bitcast i8* %malloccall to [1000000 x %struct.foo]* ; <[1000000 x %struct.foo]*> [#uses=1]
-	%.sub = getelementptr [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
+	%.sub = getelementptr [1000000 x %struct.foo], [1000000 x %struct.foo]* %tmp, i32 0, i32 0		; <%struct.foo*> [#uses=1]
 	store %struct.foo* %.sub, %struct.foo** @X, align 4
 	ret void
 }
@@ -18,22 +18,22 @@ declare noalias i8* @malloc(i64)
 
 define i32 @baz() nounwind readonly noinline {
 bb1.thread:
-	%tmpLD1 = load %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
+	%tmpLD1 = load %struct.foo*, %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
 	br label %bb1
 
 bb1:		; preds = %bb1, %bb1.thread
         %tmp = phi %struct.foo* [%tmpLD1, %bb1.thread ], [ %tmpLD2, %bb1 ]		; <i32> [#uses=2]
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %indvar.next, %bb1 ]		; <i32> [#uses=2]
 	%sum.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %tmp3, %bb1 ]		; <i32> [#uses=1]
-	%tmp1 = getelementptr %struct.foo* %tmp, i32 %i.0.reg2mem.0, i32 0		; <i32*> [#uses=1]
-	%tmp2 = load i32* %tmp1, align 4		; <i32> [#uses=1]
+	%tmp1 = getelementptr %struct.foo, %struct.foo* %tmp, i32 %i.0.reg2mem.0, i32 0		; <i32*> [#uses=1]
+	%tmp2 = load i32, i32* %tmp1, align 4		; <i32> [#uses=1]
 	%tmp6 = add i32 %tmp2, %sum.0.reg2mem.0		; <i32> [#uses=2]
-	%tmp4 = getelementptr %struct.foo* %tmp, i32 %i.0.reg2mem.0, i32 1		; <i32*> [#uses=1]
-        %tmp5 = load i32 * %tmp4
+	%tmp4 = getelementptr %struct.foo, %struct.foo* %tmp, i32 %i.0.reg2mem.0, i32 1		; <i32*> [#uses=1]
+        %tmp5 = load i32 , i32 * %tmp4
         %tmp3 = add i32 %tmp5, %tmp6
 	%indvar.next = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=2]
         
-      	%tmpLD2 = load %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
+      	%tmpLD2 = load %struct.foo*, %struct.foo** @X, align 4		; <%struct.foo*> [#uses=1]
 
 	%exitcond = icmp eq i32 %indvar.next, 1200		; <i1> [#uses=1]
 	br i1 %exitcond, label %bb2, label %bb1
diff --git a/test/Transforms/GlobalOpt/integer-bool.ll b/test/Transforms/GlobalOpt/integer-bool.ll
index abf5fdd2ef3c..617febdc0166 100644
--- a/test/Transforms/GlobalOpt/integer-bool.ll
+++ b/test/Transforms/GlobalOpt/integer-bool.ll
@@ -20,7 +20,7 @@ define void @set2() {
 
 define i1 @get() {
 ; CHECK-LABEL: @get(
-  %A = load i32 addrspace(1) * @G
+  %A = load i32, i32 addrspace(1) * @G
   %C = icmp slt i32 %A, 2
   ret i1 %C
 ; CHECK: ret i1 true
diff --git a/test/Transforms/GlobalOpt/iterate.ll b/test/Transforms/GlobalOpt/iterate.ll
index 746687427590..f7f34fc39d9a 100644
--- a/test/Transforms/GlobalOpt/iterate.ll
+++ b/test/Transforms/GlobalOpt/iterate.ll
@@ -4,8 +4,8 @@
 @H = internal global { i32* } { i32* @G }               ; <{ i32* }*> [#uses=1]
 
 define i32 @loadg() {
-        %G = load i32** getelementptr ({ i32* }* @H, i32 0, i32 0)              ; <i32*> [#uses=1]
-        %GV = load i32* %G              ; <i32> [#uses=1]
+        %G = load i32*, i32** getelementptr ({ i32* }, { i32* }* @H, i32 0, i32 0)              ; <i32*> [#uses=1]
+        %GV = load i32, i32* %G              ; <i32> [#uses=1]
         ret i32 %GV
 }
 
diff --git a/test/Transforms/GlobalOpt/load-store-global.ll b/test/Transforms/GlobalOpt/load-store-global.ll
index ad7326dc682c..e01358e8ed38 100644
--- a/test/Transforms/GlobalOpt/load-store-global.ll
+++ b/test/Transforms/GlobalOpt/load-store-global.ll
@@ -4,7 +4,7 @@
 ; CHECK-NOT: @G
 
 define void @foo() {
-        %V = load i32* @G               ; <i32> [#uses=1]
+        %V = load i32, i32* @G               ; <i32> [#uses=1]
         store i32 %V, i32* @G
         ret void
 ; CHECK-LABEL: @foo(
@@ -12,7 +12,7 @@ define void @foo() {
 }
 
 define i32 @bar() {
-        %X = load i32* @G               ; <i32> [#uses=1]
+        %X = load i32, i32* @G               ; <i32> [#uses=1]
         ret i32 %X
 ; CHECK-LABEL: @bar(
 ; CHECK-NEXT: ret i32 17
@@ -24,12 +24,12 @@ define i32 @bar() {
 ; PR13968
 define void @qux() nounwind {
   %b = bitcast i64** @a to i8*
-  %g = getelementptr i64** @a, i32 1
+  %g = getelementptr i64*, i64** @a, i32 1
   %cmp = icmp ne i8* null, %b
   %cmp2 = icmp eq i8* null, %b
   %cmp3 = icmp eq i64** null, %g
   store i64* inttoptr (i64 1 to i64*), i64** @a, align 8
-  %l = load i64** @a, align 8
+  %l = load i64*, i64** @a, align 8
   ret void
 ; CHECK-LABEL: @qux(
 ; CHECK-NOT: store
diff --git a/test/Transforms/GlobalOpt/malloc-promote-1.ll b/test/Transforms/GlobalOpt/malloc-promote-1.ll
index 51ccbbd43ca8..d5087dea3443 100644
--- a/test/Transforms/GlobalOpt/malloc-promote-1.ll
+++ b/test/Transforms/GlobalOpt/malloc-promote-1.ll
@@ -8,7 +8,7 @@ define void @init() {
         %malloccall = tail call i8* @malloc(i64 4)      ; <i8*> [#uses=1]
         %P = bitcast i8* %malloccall to i32*            ; <i32*> [#uses=1]
         store i32* %P, i32** @G
-        %GV = load i32** @G             ; <i32*> [#uses=1]
+        %GV = load i32*, i32** @G             ; <i32*> [#uses=1]
         store i32 0, i32* %GV
         ret void
 }
@@ -16,8 +16,8 @@ define void @init() {
 declare noalias i8* @malloc(i64)
 
 define i32 @get() {
-        %GV = load i32** @G             ; <i32*> [#uses=1]
-        %V = load i32* %GV              ; <i32> [#uses=1]
+        %GV = load i32*, i32** @G             ; <i32*> [#uses=1]
+        %V = load i32, i32* %GV              ; <i32> [#uses=1]
         ret i32 %V
 ; CHECK: ret i32 0
 }
diff --git a/test/Transforms/GlobalOpt/malloc-promote-2.ll b/test/Transforms/GlobalOpt/malloc-promote-2.ll
index 6cb44812d2b6..335ed82a8cf5 100644
--- a/test/Transforms/GlobalOpt/malloc-promote-2.ll
+++ b/test/Transforms/GlobalOpt/malloc-promote-2.ll
@@ -10,8 +10,8 @@ define void @t() {
   %malloccall = tail call i8* @malloc(i64 mul (i64 100, i64 4))
   %P = bitcast i8* %malloccall to i32*
   store i32* %P, i32** @G
-  %GV = load i32** @G
-  %GVe = getelementptr i32* %GV, i32 40
+  %GV = load i32*, i32** @G
+  %GVe = getelementptr i32, i32* %GV, i32 40
   store i32 20, i32* %GVe
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/malloc-promote-3.ll b/test/Transforms/GlobalOpt/malloc-promote-3.ll
index d44ee6460950..1e42c3b89a60 100644
--- a/test/Transforms/GlobalOpt/malloc-promote-3.ll
+++ b/test/Transforms/GlobalOpt/malloc-promote-3.ll
@@ -9,8 +9,8 @@ define void @t() {
   %malloccall = tail call i8* @malloc(i64 mul (i64 100, i64 4)) nobuiltin
   %P = bitcast i8* %malloccall to i32*
   store i32* %P, i32** @G
-  %GV = load i32** @G
-  %GVe = getelementptr i32* %GV, i32 40
+  %GV = load i32*, i32** @G
+  %GVe = getelementptr i32, i32* %GV, i32 40
   store i32 20, i32* %GVe
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/memcpy.ll b/test/Transforms/GlobalOpt/memcpy.ll
index dcfe009e330c..437142e8c6e1 100644
--- a/test/Transforms/GlobalOpt/memcpy.ll
+++ b/test/Transforms/GlobalOpt/memcpy.ll
@@ -5,8 +5,8 @@
 
 define void @foo() {
   %Blah = alloca [58 x i8]
-  %tmp.0 = getelementptr [58 x i8]* %Blah, i32 0, i32 0
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp.0, i8* getelementptr inbounds ([58 x i8]* @G1, i32 0, i32 0), i32 58, i32 1, i1 false)
+  %tmp.0 = getelementptr [58 x i8], [58 x i8]* %Blah, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp.0, i8* getelementptr inbounds ([58 x i8], [58 x i8]* @G1, i32 0, i32 0), i32 58, i32 1, i1 false)
   ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/memset-null.ll b/test/Transforms/GlobalOpt/memset-null.ll
index 53ec7551130e..838ac0979e30 100644
--- a/test/Transforms/GlobalOpt/memset-null.ll
+++ b/test/Transforms/GlobalOpt/memset-null.ll
@@ -23,7 +23,7 @@ entry:
 
 define internal void @_GLOBAL__I_b() nounwind {
 entry:
-  %tmp.i.i.i = load i8** @y, align 8
+  %tmp.i.i.i = load i8*, i8** @y, align 8
   tail call void @llvm.memset.p0i8.i64(i8* %tmp.i.i.i, i8 0, i64 10, i32 1, i1 false) nounwind
   ret void
 }
diff --git a/test/Transforms/GlobalOpt/memset.ll b/test/Transforms/GlobalOpt/memset.ll
index 85320b7f2461..1dfdd6422334 100644
--- a/test/Transforms/GlobalOpt/memset.ll
+++ b/test/Transforms/GlobalOpt/memset.ll
@@ -11,7 +11,7 @@ define void @foo() {
   %Blah = alloca [58 x i8]
   %tmp3 = bitcast [58 x i8]* %Blah to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* bitcast ([4 x i32]* @G1 to i8*), i8* %tmp3, i32 16, i32 1, i1 false)
-  call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([58 x i8]* @G0, i32 0, i32 0), i8 17, i32 58, i32 1, i1 false)
+  call void @llvm.memset.p0i8.i32(i8* getelementptr inbounds ([58 x i8], [58 x i8]* @G0, i32 0, i32 0), i8 17, i32 58, i32 1, i1 false)
   ret void
 }
 
@@ -22,7 +22,7 @@ define void @foo_as1() {
   %Blah = alloca [58 x i8]
   %tmp3 = bitcast [58 x i8]* %Blah to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* addrspacecast ([4 x i32] addrspace(1)* @G1_as1 to i8*), i8* %tmp3, i32 16, i32 1, i1 false)
-  call void @llvm.memset.p1i8.i32(i8 addrspace(1)* getelementptr inbounds ([58 x i8] addrspace(1)* @G0_as1, i32 0, i32 0), i8 17, i32 58, i32 1, i1 false)
+  call void @llvm.memset.p1i8.i32(i8 addrspace(1)* getelementptr inbounds ([58 x i8], [58 x i8] addrspace(1)* @G0_as1, i32 0, i32 0), i8 17, i32 58, i32 1, i1 false)
   ret void
 }
 
diff --git a/test/Transforms/GlobalOpt/phi-select.ll b/test/Transforms/GlobalOpt/phi-select.ll
index cd8a7dc990ff..e5cb9819ca0a 100644
--- a/test/Transforms/GlobalOpt/phi-select.ll
+++ b/test/Transforms/GlobalOpt/phi-select.ll
@@ -8,7 +8,7 @@
 
 define i32 @test1(i1 %C) {
         %P = select i1 %C, i32* @X, i32* @Y             ; <i32*> [#uses=1]
-        %V = load i32* %P               ; <i32> [#uses=1]
+        %V = load i32, i32* %P               ; <i32> [#uses=1]
         ret i32 %V
 }
 
@@ -21,7 +21,7 @@ T:              ; preds = %0
 
 Cont:           ; preds = %T, %0
         %P = phi i32* [ @X, %0 ], [ @Y, %T ]            ; <i32*> [#uses=1]
-        %V = load i32* %P               ; <i32> [#uses=1]
+        %V = load i32, i32* %P               ; <i32> [#uses=1]
         ret i32 %V
 }
 
diff --git a/test/Transforms/GlobalOpt/storepointer-compare.ll b/test/Transforms/GlobalOpt/storepointer-compare.ll
index 09e20a8adb26..bcf7cafd831c 100644
--- a/test/Transforms/GlobalOpt/storepointer-compare.ll
+++ b/test/Transforms/GlobalOpt/storepointer-compare.ll
@@ -16,7 +16,7 @@ define void @init() {
 }
 
 define void @doit() {
-        %FP = load void ()** @G         ; <void ()*> [#uses=2]
+        %FP = load void ()*, void ()** @G         ; <void ()*> [#uses=2]
         %CC = icmp eq void ()* %FP, null                ; <i1> [#uses=1]
         br i1 %CC, label %isNull, label %DoCall
 
diff --git a/test/Transforms/GlobalOpt/storepointer.ll b/test/Transforms/GlobalOpt/storepointer.ll
index 8019076f9463..9003004d431d 100644
--- a/test/Transforms/GlobalOpt/storepointer.ll
+++ b/test/Transforms/GlobalOpt/storepointer.ll
@@ -12,7 +12,7 @@ define void @init() {
 }
 
 define void @doit() {
-        %FP = load void ()** @G         ; <void ()*> [#uses=1]
+        %FP = load void ()*, void ()** @G         ; <void ()*> [#uses=1]
         call void %FP( )
         ret void
 }
diff --git a/test/Transforms/GlobalOpt/tls.ll b/test/Transforms/GlobalOpt/tls.ll
index 59ae23a4f6a8..f3cb4a65704d 100644
--- a/test/Transforms/GlobalOpt/tls.ll
+++ b/test/Transforms/GlobalOpt/tls.ll
@@ -14,34 +14,34 @@ declare void @start_thread(void ()*)
 define i32 @f() {
 entry:
   ; Set @ip to point to x[1] for thread 1.
-  store i32* getelementptr inbounds ([100 x i32]* @x, i64 0, i64 1), i32** @ip, align 8
+  store i32* getelementptr inbounds ([100 x i32], [100 x i32]* @x, i64 0, i64 1), i32** @ip, align 8
 
   ; Run g on a new thread.
   tail call void @start_thread(void ()* @g) nounwind
   tail call void @wait() nounwind
 
   ; Reset x[1] for thread 1.
-  store i32 0, i32* getelementptr inbounds ([100 x i32]* @x, i64 0, i64 1), align 4
+  store i32 0, i32* getelementptr inbounds ([100 x i32], [100 x i32]* @x, i64 0, i64 1), align 4
 
   ; Read the value of @ip, which now points at x[1] for thread 2.
-  %0 = load i32** @ip, align 8
+  %0 = load i32*, i32** @ip, align 8
 
-  %1 = load i32* %0, align 4
+  %1 = load i32, i32* %0, align 4
   ret i32 %1
 
 ; CHECK-LABEL: @f(
 ; Make sure that the load from @ip hasn't been removed.
-; CHECK: load i32** @ip
+; CHECK: load i32*, i32** @ip
 ; CHECK: ret
 }
 
 define internal void @g() nounwind uwtable {
 entry:
   ; Set @ip to point to x[1] for thread 2.
-  store i32* getelementptr inbounds ([100 x i32]* @x, i64 0, i64 1), i32** @ip, align 8
+  store i32* getelementptr inbounds ([100 x i32], [100 x i32]* @x, i64 0, i64 1), i32** @ip, align 8
 
   ; Store 50 in x[1] for thread 2.
-  store i32 50, i32* getelementptr inbounds ([100 x i32]* @x, i64 0, i64 1), align 4
+  store i32 50, i32* getelementptr inbounds ([100 x i32], [100 x i32]* @x, i64 0, i64 1), align 4
 
   tail call void @signal() nounwind
   ret void
diff --git a/test/Transforms/GlobalOpt/trivialstore.ll b/test/Transforms/GlobalOpt/trivialstore.ll
index 21437f33b26d..f907d8382d57 100644
--- a/test/Transforms/GlobalOpt/trivialstore.ll
+++ b/test/Transforms/GlobalOpt/trivialstore.ll
@@ -8,7 +8,7 @@ define void @foo() {
 }
 
 define i32 @bar() {
-        %X = load i32* @G               ; <i32> [#uses=1]
+        %X = load i32, i32* @G               ; <i32> [#uses=1]
         ret i32 %X
 }
 
diff --git a/test/Transforms/GlobalOpt/undef-init.ll b/test/Transforms/GlobalOpt/undef-init.ll
index c14949739508..d304821225ea 100644
--- a/test/Transforms/GlobalOpt/undef-init.ll
+++ b/test/Transforms/GlobalOpt/undef-init.ll
@@ -5,7 +5,7 @@
 
 define i32 @_Z3foov() {
 entry:
-        %tmp.1 = load i32* @X.0         ; <i32> [#uses=1]
+        %tmp.1 = load i32, i32* @X.0         ; <i32> [#uses=1]
         ret i32 %tmp.1
 }
 
diff --git a/test/Transforms/GlobalOpt/unnamed-addr.ll b/test/Transforms/GlobalOpt/unnamed-addr.ll
index 2ca91e50da2a..85ed829c7112 100644
--- a/test/Transforms/GlobalOpt/unnamed-addr.ll
+++ b/test/Transforms/GlobalOpt/unnamed-addr.ll
@@ -13,7 +13,7 @@
 ; CHECK: @e = linkonce_odr global i32 0
 
 define i32 @get_e() {
-       %t = load i32* @e
+       %t = load i32, i32* @e
        ret i32 %t
 }
 
@@ -24,10 +24,10 @@ define void @set_e(i32 %x) {
 
 define i1 @bah(i64 %i) nounwind readonly optsize ssp {
 entry:
-  %arrayidx4 = getelementptr inbounds [4 x i8]* @d, i64 0, i64 %i
-  %tmp5 = load i8* %arrayidx4, align 1
+  %arrayidx4 = getelementptr inbounds [4 x i8], [4 x i8]* @d, i64 0, i64 %i
+  %tmp5 = load i8, i8* %arrayidx4, align 1
   %array0 = bitcast [4 x i8]* @d to i8*
-  %tmp6 = load i8* %array0, align 1
+  %tmp6 = load i8, i8* %array0, align 1
   %cmp = icmp eq i8 %tmp5, %tmp6
   ret i1 %cmp
 }
@@ -63,6 +63,6 @@ return:
 
 define i32 @zed() {
 entry:
-  %tmp1 = load i32* @c, align 4
+  %tmp1 = load i32, i32* @c, align 4
   ret i32 %tmp1
 }
diff --git a/test/Transforms/GlobalOpt/zeroinitializer-gep-load.ll b/test/Transforms/GlobalOpt/zeroinitializer-gep-load.ll
index d9787232d971..51feb480e3bf 100644
--- a/test/Transforms/GlobalOpt/zeroinitializer-gep-load.ll
+++ b/test/Transforms/GlobalOpt/zeroinitializer-gep-load.ll
@@ -3,8 +3,8 @@
 @zero = internal global [10 x i32] zeroinitializer
 
 define i32 @test1(i64 %idx) nounwind {
-  %arrayidx = getelementptr inbounds [10 x i32]* @zero, i64 0, i64 %idx
-  %l = load i32* %arrayidx
+  %arrayidx = getelementptr inbounds [10 x i32], [10 x i32]* @zero, i64 0, i64 %idx
+  %l = load i32, i32* %arrayidx
   ret i32 %l
 ; CHECK-LABEL: @test1(
 ; CHECK: ret i32 0
diff --git a/test/Transforms/IPConstantProp/2009-09-24-byval-ptr.ll b/test/Transforms/IPConstantProp/2009-09-24-byval-ptr.ll
index 4ea0b88fd0de..61f4bf6d9f7f 100644
--- a/test/Transforms/IPConstantProp/2009-09-24-byval-ptr.ll
+++ b/test/Transforms/IPConstantProp/2009-09-24-byval-ptr.ll
@@ -5,10 +5,10 @@
 @mystr = internal global %struct.MYstr zeroinitializer ; <%struct.MYstr*> [#uses=3]
 define internal void @vfu1(%struct.MYstr* byval align 4 %u) nounwind {
 entry:
-  %0 = getelementptr %struct.MYstr* %u, i32 0, i32 1 ; <i32*> [#uses=1]
+  %0 = getelementptr %struct.MYstr, %struct.MYstr* %u, i32 0, i32 1 ; <i32*> [#uses=1]
   store i32 99, i32* %0, align 4
 ; CHECK: %struct.MYstr* %u
-  %1 = getelementptr %struct.MYstr* %u, i32 0, i32 0 ; <i8*> [#uses=1]
+  %1 = getelementptr %struct.MYstr, %struct.MYstr* %u, i32 0, i32 0 ; <i8*> [#uses=1]
   store i8 97, i8* %1, align 4
 ; CHECK: %struct.MYstr* %u
   br label %return
@@ -19,12 +19,12 @@ return:                                           ; preds = %entry
 
 define internal i32 @vfu2(%struct.MYstr* byval align 4 %u) nounwind readonly {
 entry:
-  %0 = getelementptr %struct.MYstr* %u, i32 0, i32 1 ; <i32*> [#uses=1]
-  %1 = load i32* %0
-; CHECK: load i32* getelementptr inbounds (%struct.MYstr* @mystr, i32 0, i32 1)
-  %2 = getelementptr %struct.MYstr* %u, i32 0, i32 0 ; <i8*> [#uses=1]
-  %3 = load i8* %2
-; CHECK: load i8* getelementptr inbounds (%struct.MYstr* @mystr, i32 0, i32 0)
+  %0 = getelementptr %struct.MYstr, %struct.MYstr* %u, i32 0, i32 1 ; <i32*> [#uses=1]
+  %1 = load i32, i32* %0
+; CHECK: load i32, i32* getelementptr inbounds (%struct.MYstr, %struct.MYstr* @mystr, i32 0, i32 1)
+  %2 = getelementptr %struct.MYstr, %struct.MYstr* %u, i32 0, i32 0 ; <i8*> [#uses=1]
+  %3 = load i8, i8* %2
+; CHECK: load i8, i8* getelementptr inbounds (%struct.MYstr, %struct.MYstr* @mystr, i32 0, i32 0)
   %4 = zext i8 %3 to i32
   %5 = add i32 %4, %1
   ret i32 %5
diff --git a/test/Transforms/IPConstantProp/dangling-block-address.ll b/test/Transforms/IPConstantProp/dangling-block-address.ll
index bb101333f801..abd0b0fc3136 100644
--- a/test/Transforms/IPConstantProp/dangling-block-address.ll
+++ b/test/Transforms/IPConstantProp/dangling-block-address.ll
@@ -29,10 +29,10 @@ end:                                              ; preds = %indirectgoto
 
 indirectgoto:                                     ; preds = %lab0, %entry
   %indvar = phi i32 [ %indvar.next, %lab0 ], [ 0, %entry ] ; <i32> [#uses=2]
-  %pc.addr.0 = getelementptr i32* %pc, i32 %indvar ; <i32*> [#uses=1]
-  %tmp1.pn = load i32* %pc.addr.0                 ; <i32> [#uses=1]
-  %indirect.goto.dest.in = getelementptr inbounds [2 x i8*]* @bar.l, i32 0, i32 %tmp1.pn ; <i8**> [#uses=1]
-  %indirect.goto.dest = load i8** %indirect.goto.dest.in ; <i8*> [#uses=1]
+  %pc.addr.0 = getelementptr i32, i32* %pc, i32 %indvar ; <i32*> [#uses=1]
+  %tmp1.pn = load i32, i32* %pc.addr.0                 ; <i32> [#uses=1]
+  %indirect.goto.dest.in = getelementptr inbounds [2 x i8*], [2 x i8*]* @bar.l, i32 0, i32 %tmp1.pn ; <i8**> [#uses=1]
+  %indirect.goto.dest = load i8*, i8** %indirect.goto.dest.in ; <i8*> [#uses=1]
   indirectbr i8* %indirect.goto.dest, [label %lab0, label %end]
 }
 
diff --git a/test/Transforms/IPConstantProp/global.ll b/test/Transforms/IPConstantProp/global.ll
index 67152937beec..d3ba14658f6e 100644
--- a/test/Transforms/IPConstantProp/global.ll
+++ b/test/Transforms/IPConstantProp/global.ll
@@ -4,7 +4,7 @@
 
 define void @_Z7test1f1v() nounwind {
 entry:
-  %tmp = load i32* @_ZL6test1g, align 4
+  %tmp = load i32, i32* @_ZL6test1g, align 4
   %cmp = icmp eq i32 %tmp, 0
   br i1 %cmp, label %if.then, label %if.end
 
@@ -21,6 +21,6 @@ if.end:                                           ; preds = %if.then, %entry
 ; CHECK-NEXT: ret i32 42
 define i32 @_Z7test1f2v() nounwind {
 entry:
-  %tmp = load i32* @_ZL6test1g, align 4
+  %tmp = load i32, i32* @_ZL6test1g, align 4
   ret i32 %tmp
 }
diff --git a/test/Transforms/IPConstantProp/return-argument.ll b/test/Transforms/IPConstantProp/return-argument.ll
index 2a14f05985d0..927531b996ea 100644
--- a/test/Transforms/IPConstantProp/return-argument.ll
+++ b/test/Transforms/IPConstantProp/return-argument.ll
@@ -4,7 +4,7 @@
 
 ;; This function returns its second argument on all return statements
 define internal i32* @incdec(i1 %C, i32* %V) {
-        %X = load i32* %V
+        %X = load i32, i32* %V
         br i1 %C, label %T, label %F
 
 T:              ; preds = %0
diff --git a/test/Transforms/IPConstantProp/return-constant.ll b/test/Transforms/IPConstantProp/return-constant.ll
index 499d38329522..d89579532e71 100644
--- a/test/Transforms/IPConstantProp/return-constant.ll
+++ b/test/Transforms/IPConstantProp/return-constant.ll
@@ -1,5 +1,4 @@
-; RUN: opt < %s -ipconstprop -instcombine | \
-; RUN:    llvm-dis | grep "ret i1 true" | count 2
+; RUN: opt < %s -ipconstprop -instcombine -S | grep "ret i1 true" | count 2
 define internal i32 @foo(i1 %C) {
         br i1 %C, label %T, label %F
 
diff --git a/test/Transforms/IRCE/bug-loop-varying-upper-limit.ll b/test/Transforms/IRCE/bug-loop-varying-upper-limit.ll
new file mode 100644
index 000000000000..4635bb4381ea
--- /dev/null
+++ b/test/Transforms/IRCE/bug-loop-varying-upper-limit.ll
@@ -0,0 +1,31 @@
+; RUN: opt -irce-print-changed-loops -S -irce -verify < %s 2>&1 | FileCheck %s
+
+; CHECK-NOT: constrained loop
+
+define void @single_access_no_preloop_no_offset(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %abc = icmp slt i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/bug-mismatched-types.ll b/test/Transforms/IRCE/bug-mismatched-types.ll
new file mode 100644
index 000000000000..9ff72493393e
--- /dev/null
+++ b/test/Transforms/IRCE/bug-mismatched-types.ll
@@ -0,0 +1,66 @@
+; RUN: opt -irce -S < %s
+
+; These test cases don't check the correctness of the transform, but
+; that the -irce does not crash in the presence of certain things in
+; the IR:
+
+define void @mismatched_types_1() {
+; In this test case, the safe range for the only range check in the
+; loop is of type [i32, i32) while the backedge taken count is of type
+; i64.
+
+; CHECK-LABEL: mismatched_types_1
+entry:
+  br label %for.body
+
+for.body:
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
+  %0 = trunc i64 %indvars.iv to i32
+  %1 = icmp ult i32 %0, 7
+  br i1 %1, label %switch.lookup, label %for.inc
+
+switch.lookup:
+  br label %for.inc
+
+for.inc:
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %cmp55 = icmp slt i64 %indvars.iv.next, 11
+  br i1 %cmp55, label %for.body, label %for.end
+
+for.end:
+  unreachable
+}
+
+define void @mismatched_types_2() {
+; In this test case, there are two range check in the loop, one with a
+; safe range of type [i32, i32) and one with a safe range of type
+; [i64, i64).
+
+; CHECK-LABEL: mismatched_types_2
+entry:
+  br label %for.body.a
+
+for.body.a:
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
+  %cond.a = icmp ult i64 %indvars.iv, 7
+  br i1 %cond.a, label %switch.lookup.a, label %for.body.b
+
+switch.lookup.a:
+  br label %for.body.b
+
+for.body.b:
+  %truncated = trunc i64 %indvars.iv to i32
+  %cond.b = icmp ult i32 %truncated, 7
+  br i1 %cond.b, label %switch.lookup.b, label %for.inc
+
+switch.lookup.b:
+  br label %for.inc
+
+for.inc:
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %cmp55 = icmp slt i64 %indvars.iv.next, 11
+  br i1 %cmp55, label %for.body.a, label %for.end
+
+for.end:
+  unreachable
+}
diff --git a/test/Transforms/IRCE/decrementing-loop.ll b/test/Transforms/IRCE/decrementing-loop.ll
new file mode 100644
index 000000000000..6ff3e76c7546
--- /dev/null
+++ b/test/Transforms/IRCE/decrementing-loop.ll
@@ -0,0 +1,43 @@
+; RUN: opt -irce -S < %s | FileCheck %s
+
+define void @decrementing_loop(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  %start = sub i32 %n, 1
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ %start, %entry ] , [ %idx.dec, %in.bounds ]
+  %idx.dec = sub i32 %idx, 1
+  %abc.high = icmp slt i32 %idx, %len
+  %abc.low = icmp sge i32 %idx, 0
+  %abc = and i1 %abc.low, %abc.high
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp sgt i32 %idx.dec, -1
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+
+; CHECK: loop.preheader:
+; CHECK:   [[indvar_start:[^ ]+]] = add i32 %n, -1
+; CHECK:   [[not_len:[^ ]+]] = sub i32 -1, %len
+; CHECK:   [[not_n:[^ ]+]] = sub i32 -1, %n
+; CHECK:   [[not_len_hiclamp_cmp:[^ ]+]] = icmp sgt i32 [[not_len]], [[not_n]]
+; CHECK:   [[not_len_hiclamp:[^ ]+]] = select i1 [[not_len_hiclamp_cmp]], i32 [[not_len]], i32 [[not_n]]
+; CHECK:   [[len_hiclamp:[^ ]+]] = sub i32 -1, [[not_len_hiclamp]]
+; CHECK:   [[not_exit_preloop_at_cmp:[^ ]+]] = icmp sgt i32 [[len_hiclamp]], 0
+; CHECK:   [[not_exit_preloop_at:[^ ]+]] = select i1 [[not_exit_preloop_at_cmp]], i32 [[len_hiclamp]], i32 0
+; CHECK:   %exit.preloop.at = add i32 [[not_exit_preloop_at]], -1
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/low-becount.ll b/test/Transforms/IRCE/low-becount.ll
new file mode 100644
index 000000000000..89b91d6496a8
--- /dev/null
+++ b/test/Transforms/IRCE/low-becount.ll
@@ -0,0 +1,32 @@
+; RUN: opt -irce-print-changed-loops -irce -S < %s 2>&1 | FileCheck %s
+
+; CHECK-NOT: constrained Loop
+
+define void @low_profiled_be_count(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %abc = icmp slt i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit, !prof !2
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
+!2 = !{!"branch_weights", i32 4, i32 64}
diff --git a/test/Transforms/IRCE/multiple-access-no-preloop.ll b/test/Transforms/IRCE/multiple-access-no-preloop.ll
new file mode 100644
index 000000000000..9963a57185bf
--- /dev/null
+++ b/test/Transforms/IRCE/multiple-access-no-preloop.ll
@@ -0,0 +1,66 @@
+; RUN: opt -irce -S < %s | FileCheck %s
+
+define void @multiple_access_no_preloop(
+    i32* %arr_a, i32* %a_len_ptr, i32* %arr_b, i32* %b_len_ptr, i32 %n) {
+
+ entry:
+  %len.a = load i32, i32* %a_len_ptr, !range !0
+  %len.b = load i32, i32* %b_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds.b ]
+  %idx.next = add i32 %idx, 1
+  %abc.a = icmp slt i32 %idx, %len.a
+  br i1 %abc.a, label %in.bounds.a, label %out.of.bounds, !prof !1
+
+ in.bounds.a:
+  %addr.a = getelementptr i32, i32* %arr_a, i32 %idx
+  store i32 0, i32* %addr.a
+  %abc.b = icmp slt i32 %idx, %len.b
+  br i1 %abc.b, label %in.bounds.b, label %out.of.bounds, !prof !1
+
+ in.bounds.b:
+  %addr.b = getelementptr i32, i32* %arr_b, i32 %idx
+  store i32 -1, i32* %addr.b
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+; CHECK-LABEL: multiple_access_no_preloop
+
+; CHECK-LABEL: loop.preheader:
+; CHECK: [[not_len_b:[^ ]+]] = sub i32 -1, %len.b
+; CHECK: [[not_len_a:[^ ]+]] = sub i32 -1, %len.a
+; CHECK: [[smax_not_len_cond:[^ ]+]] = icmp sgt i32 [[not_len_b]], [[not_len_a]]
+; CHECK: [[smax_not_len:[^ ]+]] = select i1 [[smax_not_len_cond]], i32 [[not_len_b]], i32 [[not_len_a]]
+; CHECK: [[not_n:[^ ]+]] = sub i32 -1, %n
+; CHECK: [[not_upper_limit_cond_loclamp:[^ ]+]] = icmp sgt i32 [[smax_not_len]], [[not_n]]
+; CHECK: [[not_upper_limit_loclamp:[^ ]+]] = select i1 [[not_upper_limit_cond_loclamp]], i32 [[smax_not_len]], i32 [[not_n]]
+; CHECK: [[upper_limit_loclamp:[^ ]+]] = sub i32 -1, [[not_upper_limit_loclamp]]
+; CHECK: [[upper_limit_cmp:[^ ]+]] = icmp sgt i32 [[upper_limit_loclamp]], 0
+; CHECK: [[upper_limit:[^ ]+]] = select i1 [[upper_limit_cmp]], i32 [[upper_limit_loclamp]], i32 0
+
+; CHECK-LABEL: loop:
+; CHECK: br i1 true, label %in.bounds.a, label %out.of.bounds
+
+; CHECK-LABEL: in.bounds.a:
+; CHECK: br i1 true, label %in.bounds.b, label %out.of.bounds
+
+; CHECK-LABEL: in.bounds.b:
+; CHECK: [[main_loop_cond:[^ ]+]] = icmp slt i32 %idx.next, [[upper_limit]]
+; CHECK: br i1 [[main_loop_cond]], label %loop, label %main.exit.selector
+
+; CHECK-LABEL: in.bounds.b.postloop:
+; CHECK: %next.postloop = icmp slt i32 %idx.next.postloop, %n
+; CHECK: br i1 %next.postloop, label %loop.postloop, label %exit.loopexit
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/not-likely-taken.ll b/test/Transforms/IRCE/not-likely-taken.ll
new file mode 100644
index 000000000000..50a64a286320
--- /dev/null
+++ b/test/Transforms/IRCE/not-likely-taken.ll
@@ -0,0 +1,40 @@
+; RUN: opt -verify-loop-info -irce-print-changed-loops -irce < %s 2>&1 | FileCheck %s
+
+; CHECK-NOT: constrained Loop
+
+define void @multiple_access_no_preloop(
+    i32* %arr_a, i32* %a_len_ptr, i32* %arr_b, i32* %b_len_ptr, i32 %n) {
+
+ entry:
+  %len.a = load i32, i32* %a_len_ptr, !range !0
+  %len.b = load i32, i32* %b_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds.b ]
+  %idx.next = add i32 %idx, 1
+  %abc.a = icmp slt i32 %idx, %len.a
+  br i1 %abc.a, label %in.bounds.a, label %out.of.bounds, !prof !1
+
+ in.bounds.a:
+  %addr.a = getelementptr i32, i32* %arr_a, i32 %idx
+  store i32 0, i32* %addr.a
+  %abc.b = icmp slt i32 %idx, %len.b
+  br i1 %abc.b, label %in.bounds.b, label %out.of.bounds, !prof !1
+
+ in.bounds.b:
+  %addr.b = getelementptr i32, i32* %arr_b, i32 %idx
+  store i32 -1, i32* %addr.b
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 1, i32 1}
\ No newline at end of file
diff --git a/test/Transforms/IRCE/only-lower-check.ll b/test/Transforms/IRCE/only-lower-check.ll
new file mode 100644
index 000000000000..69abc89c1a66
--- /dev/null
+++ b/test/Transforms/IRCE/only-lower-check.ll
@@ -0,0 +1,37 @@
+; RUN: opt -irce-print-range-checks -irce-print-changed-loops -irce < %s 2>&1 | FileCheck %s
+
+; CHECK: irce: loop has 1 inductive range checks:
+; CHECK-NEXT: InductiveRangeCheck:
+; CHECK-NEXT:   Kind: RANGE_CHECK_LOWER
+; CHECK-NEXT:   Offset: (-1 + %n)  Scale: -1  Length: (null)
+; CHECK-NEXT:   Branch:   br i1 %abc, label %in.bounds, label %out.of.bounds
+; CHECK-NEXT: irce: in function only_lower_check: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+
+define void @only_lower_check(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  %start = sub i32 %n, 1
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ %start, %entry ] , [ %idx.dec, %in.bounds ]
+  %idx.dec = sub i32 %idx, 1
+  %abc = icmp sge i32 %idx, 0
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp sgt i32 %idx.dec, -1
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/only-upper-check.ll b/test/Transforms/IRCE/only-upper-check.ll
new file mode 100644
index 000000000000..dda3f3f6dd87
--- /dev/null
+++ b/test/Transforms/IRCE/only-upper-check.ll
@@ -0,0 +1,37 @@
+; RUN: opt -irce -irce-print-range-checks -irce-print-changed-loops %s -S 2>&1 | FileCheck %s
+
+; CHECK: irce: loop has 1 inductive range checks:
+; CHECK-NEXT:InductiveRangeCheck:
+; CHECK-NEXT:  Kind: RANGE_CHECK_UPPER
+; CHECK-NEXT:  Offset: %offset  Scale: 1  Length:   %len = load i32, i32* %a_len_ptr, !range !0
+; CHECK-NEXT:  Branch:   br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+; CHECK-NEXT: irce: in function incrementing: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+
+define void @incrementing(i32 *%arr, i32 *%a_len_ptr, i32 %n, i32 %offset) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %array.idx = add i32 %idx, %offset
+  %abc = icmp slt i32 %array.idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %array.idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/single-access-no-preloop.ll b/test/Transforms/IRCE/single-access-no-preloop.ll
new file mode 100644
index 000000000000..4824d9560c69
--- /dev/null
+++ b/test/Transforms/IRCE/single-access-no-preloop.ll
@@ -0,0 +1,116 @@
+; RUN: opt -irce -S < %s | FileCheck %s
+
+define void @single_access_no_preloop_no_offset(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %abc = icmp slt i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+; CHECK-LABEL: single_access_no_preloop
+
+; CHECK-LABEL: loop:
+; CHECK: br i1 true, label %in.bounds, label %out.of.bounds
+
+; CHECK-LABEL: main.exit.selector:
+; CHECK-NEXT: [[continue:%[^ ]+]] = icmp slt i32 %idx.next, %n
+; CHECK-NEXT: br i1 [[continue]], label %main.pseudo.exit, label %exit.loopexit
+
+; CHECK-LABEL: main.pseudo.exit:
+; CHECK-NEXT: %idx.copy = phi i32 [ 0, %loop.preheader ], [ %idx.next, %main.exit.selector ]
+; CHECK-NEXT: %indvar.end = phi i32 [ 0, %loop.preheader ], [ %idx.next, %main.exit.selector ]
+; CHECK-NEXT: br label %postloop
+
+; CHECK-LABEL: postloop:
+; CHECK-NEXT: br label %loop.postloop
+
+; CHECK-LABEL: loop.postloop:
+; CHECK-NEXT: %idx.postloop = phi i32 [ %idx.next.postloop, %in.bounds.postloop ], [ %idx.copy, %postloop ]
+; CHECK-NEXT: %idx.next.postloop = add i32 %idx.postloop, 1
+; CHECK-NEXT: %abc.postloop = icmp slt i32 %idx.postloop, %len
+; CHECK-NEXT: br i1 %abc.postloop, label %in.bounds.postloop, label %out.of.bounds
+
+; CHECK-LABEL: in.bounds.postloop:
+; CHECK-NEXT: %addr.postloop = getelementptr i32, i32* %arr, i32 %idx.postloop
+; CHECK-NEXT: store i32 0, i32* %addr.postloop
+; CHECK-NEXT: %next.postloop = icmp slt i32 %idx.next.postloop, %n
+; CHECK-NEXT: br i1 %next.postloop, label %loop.postloop, label %exit.loopexit
+
+
+define void @single_access_no_preloop_with_offset(i32 *%arr, i32 *%a_len_ptr, i32 %n) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %idx.for.abc = add i32 %idx, 4
+  %abc = icmp slt i32 %idx.for.abc, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %idx.for.abc
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+; CHECK-LABEL: single_access_no_preloop_with_offset
+
+; CHECK-LABEL: loop.preheader:
+; CHECK: [[not_n:[^ ]+]] = sub i32 -1, %n
+; CHECK: [[not_safe_range_end:[^ ]+]] = sub i32 3, %len
+; CHECK: [[not_exit_main_loop_at_hiclamp_cmp:[^ ]+]] = icmp sgt i32 [[not_n]], [[not_safe_range_end]]
+; CHECK: [[not_exit_main_loop_at_hiclamp:[^ ]+]] = select i1 [[not_exit_main_loop_at_hiclamp_cmp]], i32 [[not_n]], i32 [[not_safe_range_end]]
+; CHECK: [[exit_main_loop_at_hiclamp:[^ ]+]] = sub i32 -1, [[not_exit_main_loop_at_hiclamp]]
+; CHECK: [[exit_main_loop_at_loclamp_cmp:[^ ]+]] = icmp sgt i32 [[exit_main_loop_at_hiclamp]], 0
+; CHECK: [[exit_main_loop_at_loclamp:[^ ]+]] = select i1 [[exit_main_loop_at_loclamp_cmp]], i32 [[exit_main_loop_at_hiclamp]], i32 0
+; CHECK: [[enter_main_loop:[^ ]+]] = icmp slt i32 0, [[exit_main_loop_at_loclamp]]
+; CHECK: br i1 [[enter_main_loop]], label %loop, label %main.pseudo.exit
+
+; CHECK-LABEL: loop:
+; CHECK: br i1 true, label %in.bounds, label %out.of.bounds
+
+; CHECK-LABEL: in.bounds:
+; CHECK: [[continue_main_loop:[^ ]+]] = icmp slt i32 %idx.next, [[exit_main_loop_at_loclamp]]
+; CHECK: br i1 [[continue_main_loop]], label %loop, label %main.exit.selector
+
+; CHECK-LABEL: main.pseudo.exit:
+; CHECK:  %idx.copy = phi i32 [ 0, %loop.preheader ], [ %idx.next, %main.exit.selector ]
+; CHECK:  br label %postloop
+
+; CHECK-LABEL: loop.postloop:
+; CHECK: %idx.postloop = phi i32 [ %idx.next.postloop, %in.bounds.postloop ], [ %idx.copy, %postloop ]
+
+; CHECK-LABEL: in.bounds.postloop:
+; CHECK: %next.postloop = icmp slt i32 %idx.next.postloop, %n
+; CHECK: br i1 %next.postloop, label %loop.postloop, label %exit.loopexit
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/single-access-with-preloop.ll b/test/Transforms/IRCE/single-access-with-preloop.ll
new file mode 100644
index 000000000000..e581014ad188
--- /dev/null
+++ b/test/Transforms/IRCE/single-access-with-preloop.ll
@@ -0,0 +1,71 @@
+; RUN: opt -irce -S < %s | FileCheck %s
+
+define void @single_access_with_preloop(i32 *%arr, i32 *%a_len_ptr, i32 %n, i32 %offset) {
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %array.idx = add i32 %idx, %offset
+  %abc.high = icmp slt i32 %array.idx, %len
+  %abc.low = icmp sge i32 %array.idx, 0
+  %abc = and i1 %abc.low, %abc.high
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %array.idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+; CHECK-LABEL: loop.preheader:
+; CHECK: [[not_safe_start:[^ ]+]] = add i32 %offset, -1
+; CHECK: [[not_n:[^ ]+]] = sub i32 -1, %n
+; CHECK: [[not_exit_preloop_at_cond_loclamp:[^ ]+]] = icmp sgt i32 [[not_safe_start]], [[not_n]]
+; CHECK: [[not_exit_preloop_at_loclamp:[^ ]+]] = select i1 [[not_exit_preloop_at_cond_loclamp]], i32 [[not_safe_start]], i32 [[not_n]]
+; CHECK: [[exit_preloop_at_loclamp:[^ ]+]] = sub i32 -1, [[not_exit_preloop_at_loclamp]]
+; CHECK: [[exit_preloop_at_cond:[^ ]+]] = icmp sgt i32 [[exit_preloop_at_loclamp]], 0
+; CHECK: [[exit_preloop_at:[^ ]+]] = select i1 [[exit_preloop_at_cond]], i32 [[exit_preloop_at_loclamp]], i32 0
+
+
+; CHECK: [[not_safe_start_2:[^ ]+]] = add i32 %offset, -1
+; CHECK: [[not_safe_end:[^ ]+]] = sub i32 [[not_safe_start_2]], %len
+; CHECK: [[not_exit_mainloop_at_cond_loclamp:[^ ]+]] = icmp sgt i32 [[not_safe_end]], [[not_n]]
+; CHECK: [[not_exit_mainloop_at_loclamp:[^ ]+]] = select i1 [[not_exit_mainloop_at_cond_loclamp]], i32 [[not_safe_end]], i32 [[not_n]]
+; CHECK: [[exit_mainloop_at_loclamp:[^ ]+]] = sub i32 -1, [[not_exit_mainloop_at_loclamp]]
+; CHECK: [[exit_mainloop_at_cmp:[^ ]+]] = icmp sgt i32 [[exit_mainloop_at_loclamp]], 0
+; CHECK: [[exit_mainloop_at:[^ ]+]] = select i1 [[exit_mainloop_at_cmp]], i32 [[exit_mainloop_at_loclamp]], i32 0
+
+
+; CHECK-LABEL: in.bounds:
+; CHECK: [[continue_mainloop_cond:[^ ]+]] = icmp slt i32 %idx.next, [[exit_mainloop_at]]
+; CHECK: br i1 [[continue_mainloop_cond]], label %loop, label %main.exit.selector
+
+; CHECK-LABEL: main.exit.selector:
+; CHECK: [[mainloop_its_left:[^ ]+]] = icmp slt i32 %idx.next, %n
+; CHECK: br i1 [[mainloop_its_left]], label %main.pseudo.exit, label %exit.loopexit
+
+; CHECK-LABEL: in.bounds.preloop:
+; CHECK: [[continue_preloop_cond:[^ ]+]] = icmp slt i32 %idx.next.preloop, [[exit_preloop_at]]
+; CHECK: br i1 [[continue_preloop_cond]], label %loop.preloop, label %preloop.exit.selector
+
+; CHECK-LABEL: preloop.exit.selector:
+; CHECK: [[preloop_its_left:[^ ]+]] = icmp slt i32 %idx.next.preloop, %n
+; CHECK: br i1 [[preloop_its_left]], label %preloop.pseudo.exit, label %exit.loopexit
+
+; CHECK-LABEL: in.bounds.postloop:
+; CHECK: %next.postloop = icmp slt i32 %idx.next.postloop, %n
+; CHECK: br i1 %next.postloop, label %loop.postloop, label %exit.loopexit
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IRCE/unhandled.ll b/test/Transforms/IRCE/unhandled.ll
new file mode 100644
index 000000000000..0d5ff953aa3b
--- /dev/null
+++ b/test/Transforms/IRCE/unhandled.ll
@@ -0,0 +1,37 @@
+; RUN: opt -irce-print-changed-loops -irce -S < %s 2>&1 | FileCheck %s
+
+; Demonstrates that we don't currently handle the general expression
+; `A * I + B'.
+
+define void @general_affine_expressions(i32 *%arr, i32 *%a_len_ptr, i32 %n,
+                                        i32 %scale, i32 %offset) {
+; CHECK-NOT: constrained Loop at depth
+ entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+ loop:
+  %idx = phi i32 [ 0, %entry ] , [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %idx.mul = mul i32 %idx, %scale
+  %array.idx = add i32 %idx.mul, %offset
+  %abc.high = icmp slt i32 %array.idx, %len
+  %abc.low = icmp sge i32 %array.idx, 0
+  %abc = and i1 %abc.low, %abc.high
+  br i1 %abc, label %in.bounds, label %out.of.bounds
+
+ in.bounds:
+  %addr = getelementptr i32, i32* %arr, i32 %array.idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+ out.of.bounds:
+  ret void
+
+ exit:
+  ret void
+}
+
+!0 = !{i32 0, i32 2147483647}
diff --git a/test/Transforms/IRCE/with-parent-loops.ll b/test/Transforms/IRCE/with-parent-loops.ll
new file mode 100644
index 000000000000..00af20b964f8
--- /dev/null
+++ b/test/Transforms/IRCE/with-parent-loops.ll
@@ -0,0 +1,345 @@
+; RUN: opt -verify-loop-info -irce-print-changed-loops -irce < %s 2>&1 | FileCheck %s
+
+; This test checks if we update the LoopInfo correctly in the presence
+; of parents, uncles and cousins.
+
+; Function Attrs: alwaysinline
+define void @inner_loop(i32* %arr, i32* %a_len_ptr, i32 %n) #0 {
+; CHECK: irce: in function inner_loop: constrained Loop at depth 1 containing: %loop<header><exiting>,%in.bounds<latch><exiting>
+
+entry:
+  %len = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check = icmp sgt i32 %n, 0
+  br i1 %first.itr.check, label %loop, label %exit
+
+loop:                                             ; preds = %in.bounds, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %in.bounds ]
+  %idx.next = add i32 %idx, 1
+  %abc = icmp slt i32 %idx, %len
+  br i1 %abc, label %in.bounds, label %out.of.bounds, !prof !1
+
+in.bounds:                                        ; preds = %loop
+  %addr = getelementptr i32, i32* %arr, i32 %idx
+  store i32 0, i32* %addr
+  %next = icmp slt i32 %idx.next, %n
+  br i1 %next, label %loop, label %exit
+
+out.of.bounds:                                    ; preds = %loop
+  ret void
+
+exit:                                             ; preds = %in.bounds, %entry
+  ret void
+}
+
+; Function Attrs: alwaysinline
+define void @with_parent(i32* %arr, i32* %a_len_ptr, i32 %n, i32 %parent.count) #0 {
+; CHECK: irce: in function with_parent: constrained Loop at depth 2 containing: %loop.i<header><exiting>,%in.bounds.i<latch><exiting>
+
+entry:
+  br label %loop
+
+loop:                                             ; preds = %inner_loop.exit, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %inner_loop.exit ]
+  %idx.next = add i32 %idx, 1
+  %next = icmp ult i32 %idx.next, %parent.count
+  %len.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i, label %loop.i, label %exit.i
+
+loop.i:                                           ; preds = %in.bounds.i, %loop
+  %idx.i = phi i32 [ 0, %loop ], [ %idx.next.i, %in.bounds.i ]
+  %idx.next.i = add i32 %idx.i, 1
+  %abc.i = icmp slt i32 %idx.i, %len.i
+  br i1 %abc.i, label %in.bounds.i, label %out.of.bounds.i, !prof !1
+
+in.bounds.i:                                      ; preds = %loop.i
+  %addr.i = getelementptr i32, i32* %arr, i32 %idx.i
+  store i32 0, i32* %addr.i
+  %next.i = icmp slt i32 %idx.next.i, %n
+  br i1 %next.i, label %loop.i, label %exit.i
+
+out.of.bounds.i:                                  ; preds = %loop.i
+  br label %inner_loop.exit
+
+exit.i:                                           ; preds = %in.bounds.i, %loop
+  br label %inner_loop.exit
+
+inner_loop.exit:                                  ; preds = %exit.i, %out.of.bounds.i
+  br i1 %next, label %loop, label %exit
+
+exit:                                             ; preds = %inner_loop.exit
+  ret void
+}
+
+; Function Attrs: alwaysinline
+define void @with_grandparent(i32* %arr, i32* %a_len_ptr, i32 %n, i32 %parent.count, i32 %grandparent.count) #0 {
+; CHECK: irce: in function with_grandparent: constrained Loop at depth 3 containing: %loop.i.i<header><exiting>,%in.bounds.i.i<latch><exiting>
+
+entry:
+  br label %loop
+
+loop:                                             ; preds = %with_parent.exit, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %with_parent.exit ]
+  %idx.next = add i32 %idx, 1
+  %next = icmp ult i32 %idx.next, %grandparent.count
+  br label %loop.i
+
+loop.i:                                           ; preds = %inner_loop.exit.i, %loop
+  %idx.i = phi i32 [ 0, %loop ], [ %idx.next.i, %inner_loop.exit.i ]
+  %idx.next.i = add i32 %idx.i, 1
+  %next.i = icmp ult i32 %idx.next.i, %parent.count
+  %len.i.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i.i, label %loop.i.i, label %exit.i.i
+
+loop.i.i:                                         ; preds = %in.bounds.i.i, %loop.i
+  %idx.i.i = phi i32 [ 0, %loop.i ], [ %idx.next.i.i, %in.bounds.i.i ]
+  %idx.next.i.i = add i32 %idx.i.i, 1
+  %abc.i.i = icmp slt i32 %idx.i.i, %len.i.i
+  br i1 %abc.i.i, label %in.bounds.i.i, label %out.of.bounds.i.i, !prof !1
+
+in.bounds.i.i:                                    ; preds = %loop.i.i
+  %addr.i.i = getelementptr i32, i32* %arr, i32 %idx.i.i
+  store i32 0, i32* %addr.i.i
+  %next.i.i = icmp slt i32 %idx.next.i.i, %n
+  br i1 %next.i.i, label %loop.i.i, label %exit.i.i
+
+out.of.bounds.i.i:                                ; preds = %loop.i.i
+  br label %inner_loop.exit.i
+
+exit.i.i:                                         ; preds = %in.bounds.i.i, %loop.i
+  br label %inner_loop.exit.i
+
+inner_loop.exit.i:                                ; preds = %exit.i.i, %out.of.bounds.i.i
+  br i1 %next.i, label %loop.i, label %with_parent.exit
+
+with_parent.exit:                                 ; preds = %inner_loop.exit.i
+  br i1 %next, label %loop, label %exit
+
+exit:                                             ; preds = %with_parent.exit
+  ret void
+}
+
+; Function Attrs: alwaysinline
+define void @with_sibling(i32* %arr, i32* %a_len_ptr, i32 %n, i32 %parent.count) #0 {
+; CHECK: irce: in function with_sibling: constrained Loop at depth 2 containing: %loop.i<header><exiting>,%in.bounds.i<latch><exiting>
+; CHECK: irce: in function with_sibling: constrained Loop at depth 2 containing: %loop.i6<header><exiting>,%in.bounds.i9<latch><exiting>
+
+entry:
+  br label %loop
+
+loop:                                             ; preds = %inner_loop.exit12, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %inner_loop.exit12 ]
+  %idx.next = add i32 %idx, 1
+  %next = icmp ult i32 %idx.next, %parent.count
+  %len.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i, label %loop.i, label %exit.i
+
+loop.i:                                           ; preds = %in.bounds.i, %loop
+  %idx.i = phi i32 [ 0, %loop ], [ %idx.next.i, %in.bounds.i ]
+  %idx.next.i = add i32 %idx.i, 1
+  %abc.i = icmp slt i32 %idx.i, %len.i
+  br i1 %abc.i, label %in.bounds.i, label %out.of.bounds.i, !prof !1
+
+in.bounds.i:                                      ; preds = %loop.i
+  %addr.i = getelementptr i32, i32* %arr, i32 %idx.i
+  store i32 0, i32* %addr.i
+  %next.i = icmp slt i32 %idx.next.i, %n
+  br i1 %next.i, label %loop.i, label %exit.i
+
+out.of.bounds.i:                                  ; preds = %loop.i
+  br label %inner_loop.exit
+
+exit.i:                                           ; preds = %in.bounds.i, %loop
+  br label %inner_loop.exit
+
+inner_loop.exit:                                  ; preds = %exit.i, %out.of.bounds.i
+  %len.i1 = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i2 = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i2, label %loop.i6, label %exit.i11
+
+loop.i6:                                          ; preds = %in.bounds.i9, %inner_loop.exit
+  %idx.i3 = phi i32 [ 0, %inner_loop.exit ], [ %idx.next.i4, %in.bounds.i9 ]
+  %idx.next.i4 = add i32 %idx.i3, 1
+  %abc.i5 = icmp slt i32 %idx.i3, %len.i1
+  br i1 %abc.i5, label %in.bounds.i9, label %out.of.bounds.i10, !prof !1
+
+in.bounds.i9:                                     ; preds = %loop.i6
+  %addr.i7 = getelementptr i32, i32* %arr, i32 %idx.i3
+  store i32 0, i32* %addr.i7
+  %next.i8 = icmp slt i32 %idx.next.i4, %n
+  br i1 %next.i8, label %loop.i6, label %exit.i11
+
+out.of.bounds.i10:                                ; preds = %loop.i6
+  br label %inner_loop.exit12
+
+exit.i11:                                         ; preds = %in.bounds.i9, %inner_loop.exit
+  br label %inner_loop.exit12
+
+inner_loop.exit12:                                ; preds = %exit.i11, %out.of.bounds.i10
+  br i1 %next, label %loop, label %exit
+
+exit:                                             ; preds = %inner_loop.exit12
+  ret void
+}
+
+; Function Attrs: alwaysinline
+define void @with_cousin(i32* %arr, i32* %a_len_ptr, i32 %n, i32 %parent.count, i32 %grandparent.count) #0 {
+; CHECK: irce: in function with_cousin: constrained Loop at depth 3 containing: %loop.i.i<header><exiting>,%in.bounds.i.i<latch><exiting>
+; CHECK: irce: in function with_cousin: constrained Loop at depth 3 containing: %loop.i.i10<header><exiting>,%in.bounds.i.i13<latch><exiting>
+
+entry:
+  br label %loop
+
+loop:                                             ; preds = %with_parent.exit17, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %with_parent.exit17 ]
+  %idx.next = add i32 %idx, 1
+  %next = icmp ult i32 %idx.next, %grandparent.count
+  br label %loop.i
+
+loop.i:                                           ; preds = %inner_loop.exit.i, %loop
+  %idx.i = phi i32 [ 0, %loop ], [ %idx.next.i, %inner_loop.exit.i ]
+  %idx.next.i = add i32 %idx.i, 1
+  %next.i = icmp ult i32 %idx.next.i, %parent.count
+  %len.i.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i.i, label %loop.i.i, label %exit.i.i
+
+loop.i.i:                                         ; preds = %in.bounds.i.i, %loop.i
+  %idx.i.i = phi i32 [ 0, %loop.i ], [ %idx.next.i.i, %in.bounds.i.i ]
+  %idx.next.i.i = add i32 %idx.i.i, 1
+  %abc.i.i = icmp slt i32 %idx.i.i, %len.i.i
+  br i1 %abc.i.i, label %in.bounds.i.i, label %out.of.bounds.i.i, !prof !1
+
+in.bounds.i.i:                                    ; preds = %loop.i.i
+  %addr.i.i = getelementptr i32, i32* %arr, i32 %idx.i.i
+  store i32 0, i32* %addr.i.i
+  %next.i.i = icmp slt i32 %idx.next.i.i, %n
+  br i1 %next.i.i, label %loop.i.i, label %exit.i.i
+
+out.of.bounds.i.i:                                ; preds = %loop.i.i
+  br label %inner_loop.exit.i
+
+exit.i.i:                                         ; preds = %in.bounds.i.i, %loop.i
+  br label %inner_loop.exit.i
+
+inner_loop.exit.i:                                ; preds = %exit.i.i, %out.of.bounds.i.i
+  br i1 %next.i, label %loop.i, label %with_parent.exit
+
+with_parent.exit:                                 ; preds = %inner_loop.exit.i
+  br label %loop.i6
+
+loop.i6:                                          ; preds = %inner_loop.exit.i16, %with_parent.exit
+  %idx.i1 = phi i32 [ 0, %with_parent.exit ], [ %idx.next.i2, %inner_loop.exit.i16 ]
+  %idx.next.i2 = add i32 %idx.i1, 1
+  %next.i3 = icmp ult i32 %idx.next.i2, %parent.count
+  %len.i.i4 = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i.i5 = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i.i5, label %loop.i.i10, label %exit.i.i15
+
+loop.i.i10:                                       ; preds = %in.bounds.i.i13, %loop.i6
+  %idx.i.i7 = phi i32 [ 0, %loop.i6 ], [ %idx.next.i.i8, %in.bounds.i.i13 ]
+  %idx.next.i.i8 = add i32 %idx.i.i7, 1
+  %abc.i.i9 = icmp slt i32 %idx.i.i7, %len.i.i4
+  br i1 %abc.i.i9, label %in.bounds.i.i13, label %out.of.bounds.i.i14, !prof !1
+
+in.bounds.i.i13:                                  ; preds = %loop.i.i10
+  %addr.i.i11 = getelementptr i32, i32* %arr, i32 %idx.i.i7
+  store i32 0, i32* %addr.i.i11
+  %next.i.i12 = icmp slt i32 %idx.next.i.i8, %n
+  br i1 %next.i.i12, label %loop.i.i10, label %exit.i.i15
+
+out.of.bounds.i.i14:                              ; preds = %loop.i.i10
+  br label %inner_loop.exit.i16
+
+exit.i.i15:                                       ; preds = %in.bounds.i.i13, %loop.i6
+  br label %inner_loop.exit.i16
+
+inner_loop.exit.i16:                              ; preds = %exit.i.i15, %out.of.bounds.i.i14
+  br i1 %next.i3, label %loop.i6, label %with_parent.exit17
+
+with_parent.exit17:                               ; preds = %inner_loop.exit.i16
+  br i1 %next, label %loop, label %exit
+
+exit:                                             ; preds = %with_parent.exit17
+  ret void
+}
+
+; Function Attrs: alwaysinline
+define void @with_uncle(i32* %arr, i32* %a_len_ptr, i32 %n, i32 %parent.count, i32 %grandparent.count) #0 {
+; CHECK: irce: in function with_uncle: constrained Loop at depth 2 containing: %loop.i<header><exiting>,%in.bounds.i<latch><exiting>
+; CHECK: irce: in function with_uncle: constrained Loop at depth 3 containing: %loop.i.i<header><exiting>,%in.bounds.i.i<latch><exiting>
+
+entry:
+  br label %loop
+
+loop:                                             ; preds = %with_parent.exit, %entry
+  %idx = phi i32 [ 0, %entry ], [ %idx.next, %with_parent.exit ]
+  %idx.next = add i32 %idx, 1
+  %next = icmp ult i32 %idx.next, %grandparent.count
+  %len.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i, label %loop.i, label %exit.i
+
+loop.i:                                           ; preds = %in.bounds.i, %loop
+  %idx.i = phi i32 [ 0, %loop ], [ %idx.next.i, %in.bounds.i ]
+  %idx.next.i = add i32 %idx.i, 1
+  %abc.i = icmp slt i32 %idx.i, %len.i
+  br i1 %abc.i, label %in.bounds.i, label %out.of.bounds.i, !prof !1
+
+in.bounds.i:                                      ; preds = %loop.i
+  %addr.i = getelementptr i32, i32* %arr, i32 %idx.i
+  store i32 0, i32* %addr.i
+  %next.i = icmp slt i32 %idx.next.i, %n
+  br i1 %next.i, label %loop.i, label %exit.i
+
+out.of.bounds.i:                                  ; preds = %loop.i
+  br label %inner_loop.exit
+
+exit.i:                                           ; preds = %in.bounds.i, %loop
+  br label %inner_loop.exit
+
+inner_loop.exit:                                  ; preds = %exit.i, %out.of.bounds.i
+  br label %loop.i4
+
+loop.i4:                                          ; preds = %inner_loop.exit.i, %inner_loop.exit
+  %idx.i1 = phi i32 [ 0, %inner_loop.exit ], [ %idx.next.i2, %inner_loop.exit.i ]
+  %idx.next.i2 = add i32 %idx.i1, 1
+  %next.i3 = icmp ult i32 %idx.next.i2, %parent.count
+  %len.i.i = load i32, i32* %a_len_ptr, !range !0
+  %first.itr.check.i.i = icmp sgt i32 %n, 0
+  br i1 %first.itr.check.i.i, label %loop.i.i, label %exit.i.i
+
+loop.i.i:                                         ; preds = %in.bounds.i.i, %loop.i4
+  %idx.i.i = phi i32 [ 0, %loop.i4 ], [ %idx.next.i.i, %in.bounds.i.i ]
+  %idx.next.i.i = add i32 %idx.i.i, 1
+  %abc.i.i = icmp slt i32 %idx.i.i, %len.i.i
+  br i1 %abc.i.i, label %in.bounds.i.i, label %out.of.bounds.i.i, !prof !1
+
+in.bounds.i.i:                                    ; preds = %loop.i.i
+  %addr.i.i = getelementptr i32, i32* %arr, i32 %idx.i.i
+  store i32 0, i32* %addr.i.i
+  %next.i.i = icmp slt i32 %idx.next.i.i, %n
+  br i1 %next.i.i, label %loop.i.i, label %exit.i.i
+
+out.of.bounds.i.i:                                ; preds = %loop.i.i
+  br label %inner_loop.exit.i
+
+exit.i.i:                                         ; preds = %in.bounds.i.i, %loop.i4
+  br label %inner_loop.exit.i
+
+inner_loop.exit.i:                                ; preds = %exit.i.i, %out.of.bounds.i.i
+  br i1 %next.i3, label %loop.i4, label %with_parent.exit
+
+with_parent.exit:                                 ; preds = %inner_loop.exit.i
+  br i1 %next, label %loop, label %exit
+
+exit:                                             ; preds = %with_parent.exit
+  ret void
+}
+
+attributes #0 = { alwaysinline }
+
+!0 = !{i32 0, i32 2147483647}
+!1 = !{!"branch_weights", i32 64, i32 4}
diff --git a/test/Transforms/IndVarSimplify/2005-02-17-TruncateExprCrash.ll b/test/Transforms/IndVarSimplify/2005-02-17-TruncateExprCrash.ll
index ce043beb7fd9..16ad635a9205 100644
--- a/test/Transforms/IndVarSimplify/2005-02-17-TruncateExprCrash.ll
+++ b/test/Transforms/IndVarSimplify/2005-02-17-TruncateExprCrash.ll
@@ -45,7 +45,7 @@ then.4:		; preds = %_ZNK7QString2atEi.exit
 	ret void
 
 endif.4:		; preds = %_ZNK7QString2atEi.exit
-	%tmp.115 = load i8* null		; <i8> [#uses=1]
+	%tmp.115 = load i8, i8* null		; <i8> [#uses=1]
 	br i1 false, label %loopexit.1, label %no_exit.0
 
 no_exit.0:		; preds = %no_exit.0, %endif.4
diff --git a/test/Transforms/IndVarSimplify/2005-11-18-Crash.ll b/test/Transforms/IndVarSimplify/2005-11-18-Crash.ll
index 7202c7ae9d8d..7ef351d4a456 100644
--- a/test/Transforms/IndVarSimplify/2005-11-18-Crash.ll
+++ b/test/Transforms/IndVarSimplify/2005-11-18-Crash.ll
@@ -7,8 +7,8 @@ entry:
         br label %no_exit.0
 
 no_exit.0:              ; preds = %no_exit.0, %entry
-        %p.0.0 = phi i32* [ getelementptr ([29 x [29 x [2 x i32]]]* @fixtab, i32 0, i32 0, i32 0, i32 0), %entry ], [ %inc.0, %no_exit.0 ]               ; <i32*> [#uses=1]
-        %inc.0 = getelementptr i32* %p.0.0, i32 1               ; <i32*> [#uses=1]
+        %p.0.0 = phi i32* [ getelementptr ([29 x [29 x [2 x i32]]], [29 x [29 x [2 x i32]]]* @fixtab, i32 0, i32 0, i32 0, i32 0), %entry ], [ %inc.0, %no_exit.0 ]               ; <i32*> [#uses=1]
+        %inc.0 = getelementptr i32, i32* %p.0.0, i32 1               ; <i32*> [#uses=1]
         br i1 undef, label %no_exit.0, label %no_exit.1
 
 no_exit.1:              ; preds = %no_exit.0
diff --git a/test/Transforms/IndVarSimplify/2006-06-16-Indvar-LCSSA-Crash.ll b/test/Transforms/IndVarSimplify/2006-06-16-Indvar-LCSSA-Crash.ll
index 36ec2b81ba73..2d40f88d0d0b 100644
--- a/test/Transforms/IndVarSimplify/2006-06-16-Indvar-LCSSA-Crash.ll
+++ b/test/Transforms/IndVarSimplify/2006-06-16-Indvar-LCSSA-Crash.ll
@@ -6,9 +6,9 @@ endif.0:
 
 no_exit.30:             ; preds = %no_exit.30, %endif.0
         %x.12.0 = phi i32 [ %inc.28, %no_exit.30 ], [ -2, %endif.0 ]            ; <i32> [#uses=1]
-        %tmp.583 = load i16* null               ; <i16> [#uses=1]
+        %tmp.583 = load i16, i16* null               ; <i16> [#uses=1]
         %tmp.584 = zext i16 %tmp.583 to i32             ; <i32> [#uses=1]
-        %tmp.588 = load i32* null               ; <i32> [#uses=1]
+        %tmp.588 = load i32, i32* null               ; <i32> [#uses=1]
         %tmp.589 = mul i32 %tmp.584, %tmp.588           ; <i32> [#uses=1]
         %tmp.591 = add i32 %tmp.589, 0          ; <i32> [#uses=1]
         %inc.28 = add i32 %x.12.0, 1            ; <i32> [#uses=2]
diff --git a/test/Transforms/IndVarSimplify/2007-01-06-TripCount.ll b/test/Transforms/IndVarSimplify/2007-01-06-TripCount.ll
index 6366c8c0516c..3dda657a6d70 100644
--- a/test/Transforms/IndVarSimplify/2007-01-06-TripCount.ll
+++ b/test/Transforms/IndVarSimplify/2007-01-06-TripCount.ll
@@ -24,8 +24,8 @@ cond_true:		; preds = %bb2
 	br label %bb
 
 cond_next:		; preds = %bb2
-	%tmp2 = getelementptr [5 x i8]* @foo, i32 0, i32 %i.0		; <i8*> [#uses=1]
-	%tmp3 = load i8* %tmp2		; <i8> [#uses=1]
+	%tmp2 = getelementptr [5 x i8], [5 x i8]* @foo, i32 0, i32 %i.0		; <i8*> [#uses=1]
+	%tmp3 = load i8, i8* %tmp2		; <i8> [#uses=1]
 	%tmp5 = icmp eq i8 %tmp3, 0		; <i1> [#uses=1]
 	br i1 %tmp5, label %bb6, label %bb
 
diff --git a/test/Transforms/IndVarSimplify/2008-09-02-IVType.ll b/test/Transforms/IndVarSimplify/2008-09-02-IVType.ll
index a00483164fb1..58934f5ef84b 100644
--- a/test/Transforms/IndVarSimplify/2008-09-02-IVType.ll
+++ b/test/Transforms/IndVarSimplify/2008-09-02-IVType.ll
@@ -1,6 +1,10 @@
 ; RUN: opt < %s -indvars -S | grep sext | count 1
 ; ModuleID = '<stdin>'
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 	%struct.App1Marker = type <{ i32, i32, i32, i32, i32, i32, i32, i32, i32, i32 }>
 	%struct.ComponentInstanceRecord = type <{ [1 x i32] }>
 	%struct.DCPredictors = type { [5 x i16] }
@@ -18,18 +22,18 @@
 
 define signext i16 @ExtractBufferedBlocksIgnored(%struct.JPEGGlobals* %globp)  nounwind {
 entry:
-	%tmp4311 = getelementptr %struct.JPEGGlobals* %globp, i32 0, i32 70		; <i32*> [#uses=1]
-	%tmp4412 = load i32* %tmp4311, align 16		; <i32> [#uses=2]
+	%tmp4311 = getelementptr %struct.JPEGGlobals, %struct.JPEGGlobals* %globp, i32 0, i32 70		; <i32*> [#uses=1]
+	%tmp4412 = load i32, i32* %tmp4311, align 16		; <i32> [#uses=2]
 	%tmp4613 = icmp sgt i32 %tmp4412, 0		; <i1> [#uses=1]
 	br i1 %tmp4613, label %bb, label %bb49
 
 bb:		; preds = %bb28, %entry
 	%component.09 = phi i16 [ 0, %entry ], [ %tmp37, %bb28 ]		; <i16> [#uses=2]
 	%tmp12 = sext i16 %component.09 to i32		; <i32> [#uses=2]
-	%tmp6 = getelementptr %struct.JPEGGlobals* %globp, i32 0, i32 77, i32 %tmp12		; <i16**> [#uses=2]
-	%tmp7 = load i16** %tmp6, align 4		; <i16*> [#uses=2]
-	%tmp235 = getelementptr %struct.JPEGGlobals* %globp, i32 0, i32 71, i32 %tmp12		; <i32*> [#uses=1]
-	%tmp246 = load i32* %tmp235, align 4		; <i32> [#uses=2]
+	%tmp6 = getelementptr %struct.JPEGGlobals, %struct.JPEGGlobals* %globp, i32 0, i32 77, i32 %tmp12		; <i16**> [#uses=2]
+	%tmp7 = load i16*, i16** %tmp6, align 4		; <i16*> [#uses=2]
+	%tmp235 = getelementptr %struct.JPEGGlobals, %struct.JPEGGlobals* %globp, i32 0, i32 71, i32 %tmp12		; <i32*> [#uses=1]
+	%tmp246 = load i32, i32* %tmp235, align 4		; <i32> [#uses=2]
 	%tmp267 = icmp sgt i32 %tmp246, 0		; <i1> [#uses=1]
 	br i1 %tmp267, label %bb8, label %bb28
 
@@ -37,7 +41,7 @@ bb8:		; preds = %bb8, %bb
 	%indvar = phi i32 [ 0, %bb ], [ %indvar.next2, %bb8 ]		; <i32> [#uses=3]
 	%theDCTBufferIter.01.rec = shl i32 %indvar, 6		; <i32> [#uses=1]
 	%tmp10.rec = add i32 %theDCTBufferIter.01.rec, 64		; <i32> [#uses=1]
-	%tmp10 = getelementptr i16* %tmp7, i32 %tmp10.rec		; <i16*> [#uses=1]
+	%tmp10 = getelementptr i16, i16* %tmp7, i32 %tmp10.rec		; <i16*> [#uses=1]
 	%i.02 = trunc i32 %indvar to i16		; <i16> [#uses=1]
 	%tmp13 = add i16 %i.02, 1		; <i16> [#uses=1]
 	%phitmp = sext i16 %tmp13 to i32		; <i32> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/2008-10-03-CouldNotCompute.ll b/test/Transforms/IndVarSimplify/2008-10-03-CouldNotCompute.ll
index 23e788496138..1248154c1551 100644
--- a/test/Transforms/IndVarSimplify/2008-10-03-CouldNotCompute.ll
+++ b/test/Transforms/IndVarSimplify/2008-10-03-CouldNotCompute.ll
@@ -8,7 +8,7 @@ entry:
 	br i1 false, label %bb.nph, label %return
 
 bb.nph:		; preds = %entry
-	%0 = load i32* @foo, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* @foo, align 4		; <i32> [#uses=1]
 	%1 = sext i32 %0 to i64		; <i64> [#uses=1]
 	br label %bb
 
diff --git a/test/Transforms/IndVarSimplify/2009-04-14-shorten_iv_vars.ll b/test/Transforms/IndVarSimplify/2009-04-14-shorten_iv_vars.ll
index b4615668136b..ab2af329f4d5 100644
--- a/test/Transforms/IndVarSimplify/2009-04-14-shorten_iv_vars.ll
+++ b/test/Transforms/IndVarSimplify/2009-04-14-shorten_iv_vars.ll
@@ -15,95 +15,95 @@ bb1.thread:
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %84, %bb1 ]		; <i32> [#uses=19]
-	%0 = load i32** @a, align 8		; <i32*> [#uses=1]
-	%1 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%0 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
+	%1 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%2 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%3 = getelementptr i32* %1, i64 %2		; <i32*> [#uses=1]
-	%4 = load i32* %3, align 1		; <i32> [#uses=1]
-	%5 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%3 = getelementptr i32, i32* %1, i64 %2		; <i32*> [#uses=1]
+	%4 = load i32, i32* %3, align 1		; <i32> [#uses=1]
+	%5 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%6 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%7 = getelementptr i32* %5, i64 %6		; <i32*> [#uses=1]
-	%8 = load i32* %7, align 1		; <i32> [#uses=1]
+	%7 = getelementptr i32, i32* %5, i64 %6		; <i32*> [#uses=1]
+	%8 = load i32, i32* %7, align 1		; <i32> [#uses=1]
 	%9 = add i32 %8, %4		; <i32> [#uses=1]
 	%10 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%11 = getelementptr i32* %0, i64 %10		; <i32*> [#uses=1]
+	%11 = getelementptr i32, i32* %0, i64 %10		; <i32*> [#uses=1]
 	store i32 %9, i32* %11, align 1
-	%12 = load i32** @a, align 8		; <i32*> [#uses=1]
+	%12 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
 	%13 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
-	%14 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%14 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%15 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%16 = sext i32 %15 to i64		; <i64> [#uses=1]
-	%17 = getelementptr i32* %14, i64 %16		; <i32*> [#uses=1]
-	%18 = load i32* %17, align 1		; <i32> [#uses=1]
-	%19 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%17 = getelementptr i32, i32* %14, i64 %16		; <i32*> [#uses=1]
+	%18 = load i32, i32* %17, align 1		; <i32> [#uses=1]
+	%19 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%20 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%21 = sext i32 %20 to i64		; <i64> [#uses=1]
-	%22 = getelementptr i32* %19, i64 %21		; <i32*> [#uses=1]
-	%23 = load i32* %22, align 1		; <i32> [#uses=1]
+	%22 = getelementptr i32, i32* %19, i64 %21		; <i32*> [#uses=1]
+	%23 = load i32, i32* %22, align 1		; <i32> [#uses=1]
 	%24 = add i32 %23, %18		; <i32> [#uses=1]
 	%25 = sext i32 %13 to i64		; <i64> [#uses=1]
-	%26 = getelementptr i32* %12, i64 %25		; <i32*> [#uses=1]
+	%26 = getelementptr i32, i32* %12, i64 %25		; <i32*> [#uses=1]
 	store i32 %24, i32* %26, align 1
-	%27 = load i32** @a, align 8		; <i32*> [#uses=1]
+	%27 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
 	%28 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
-	%29 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%29 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%30 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%31 = sext i32 %30 to i64		; <i64> [#uses=1]
-	%32 = getelementptr i32* %29, i64 %31		; <i32*> [#uses=1]
-	%33 = load i32* %32, align 1		; <i32> [#uses=1]
-	%34 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%32 = getelementptr i32, i32* %29, i64 %31		; <i32*> [#uses=1]
+	%33 = load i32, i32* %32, align 1		; <i32> [#uses=1]
+	%34 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%35 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%36 = sext i32 %35 to i64		; <i64> [#uses=1]
-	%37 = getelementptr i32* %34, i64 %36		; <i32*> [#uses=1]
-	%38 = load i32* %37, align 1		; <i32> [#uses=1]
+	%37 = getelementptr i32, i32* %34, i64 %36		; <i32*> [#uses=1]
+	%38 = load i32, i32* %37, align 1		; <i32> [#uses=1]
 	%39 = add i32 %38, %33		; <i32> [#uses=1]
 	%40 = sext i32 %28 to i64		; <i64> [#uses=1]
-	%41 = getelementptr i32* %27, i64 %40		; <i32*> [#uses=1]
+	%41 = getelementptr i32, i32* %27, i64 %40		; <i32*> [#uses=1]
 	store i32 %39, i32* %41, align 1
-	%42 = load i32** @d, align 8		; <i32*> [#uses=1]
-	%43 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%42 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
+	%43 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%44 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%45 = getelementptr i32* %43, i64 %44		; <i32*> [#uses=1]
-	%46 = load i32* %45, align 1		; <i32> [#uses=1]
-	%47 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%45 = getelementptr i32, i32* %43, i64 %44		; <i32*> [#uses=1]
+	%46 = load i32, i32* %45, align 1		; <i32> [#uses=1]
+	%47 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%48 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%49 = getelementptr i32* %47, i64 %48		; <i32*> [#uses=1]
-	%50 = load i32* %49, align 1		; <i32> [#uses=1]
+	%49 = getelementptr i32, i32* %47, i64 %48		; <i32*> [#uses=1]
+	%50 = load i32, i32* %49, align 1		; <i32> [#uses=1]
 	%51 = add i32 %50, %46		; <i32> [#uses=1]
 	%52 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%53 = getelementptr i32* %42, i64 %52		; <i32*> [#uses=1]
+	%53 = getelementptr i32, i32* %42, i64 %52		; <i32*> [#uses=1]
 	store i32 %51, i32* %53, align 1
-	%54 = load i32** @d, align 8		; <i32*> [#uses=1]
+	%54 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
 	%55 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
-	%56 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%56 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%57 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%58 = sext i32 %57 to i64		; <i64> [#uses=1]
-	%59 = getelementptr i32* %56, i64 %58		; <i32*> [#uses=1]
-	%60 = load i32* %59, align 1		; <i32> [#uses=1]
-	%61 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%59 = getelementptr i32, i32* %56, i64 %58		; <i32*> [#uses=1]
+	%60 = load i32, i32* %59, align 1		; <i32> [#uses=1]
+	%61 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%62 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%63 = sext i32 %62 to i64		; <i64> [#uses=1]
-	%64 = getelementptr i32* %61, i64 %63		; <i32*> [#uses=1]
-	%65 = load i32* %64, align 1		; <i32> [#uses=1]
+	%64 = getelementptr i32, i32* %61, i64 %63		; <i32*> [#uses=1]
+	%65 = load i32, i32* %64, align 1		; <i32> [#uses=1]
 	%66 = add i32 %65, %60		; <i32> [#uses=1]
 	%67 = sext i32 %55 to i64		; <i64> [#uses=1]
-	%68 = getelementptr i32* %54, i64 %67		; <i32*> [#uses=1]
+	%68 = getelementptr i32, i32* %54, i64 %67		; <i32*> [#uses=1]
 	store i32 %66, i32* %68, align 1
-	%69 = load i32** @d, align 8		; <i32*> [#uses=1]
+	%69 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
 	%70 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
-	%71 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%71 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%72 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%73 = sext i32 %72 to i64		; <i64> [#uses=1]
-	%74 = getelementptr i32* %71, i64 %73		; <i32*> [#uses=1]
-	%75 = load i32* %74, align 1		; <i32> [#uses=1]
-	%76 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%74 = getelementptr i32, i32* %71, i64 %73		; <i32*> [#uses=1]
+	%75 = load i32, i32* %74, align 1		; <i32> [#uses=1]
+	%76 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%77 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%78 = sext i32 %77 to i64		; <i64> [#uses=1]
-	%79 = getelementptr i32* %76, i64 %78		; <i32*> [#uses=1]
-	%80 = load i32* %79, align 1		; <i32> [#uses=1]
+	%79 = getelementptr i32, i32* %76, i64 %78		; <i32*> [#uses=1]
+	%80 = load i32, i32* %79, align 1		; <i32> [#uses=1]
 	%81 = add i32 %80, %75		; <i32> [#uses=1]
 	%82 = sext i32 %70 to i64		; <i64> [#uses=1]
-	%83 = getelementptr i32* %69, i64 %82		; <i32*> [#uses=1]
+	%83 = getelementptr i32, i32* %69, i64 %82		; <i32*> [#uses=1]
 	store i32 %81, i32* %83, align 1
 	%84 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%85 = icmp sgt i32 %84, 23646		; <i1> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/2009-04-15-shorten-iv-vars-2.ll b/test/Transforms/IndVarSimplify/2009-04-15-shorten-iv-vars-2.ll
index 0722d8958517..921a3937aa52 100644
--- a/test/Transforms/IndVarSimplify/2009-04-15-shorten-iv-vars-2.ll
+++ b/test/Transforms/IndVarSimplify/2009-04-15-shorten-iv-vars-2.ll
@@ -29,127 +29,127 @@ bb1.thread:
 
 bb1:		; preds = %bb1, %bb1.thread
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb1.thread ], [ %116, %bb1 ]		; <i32> [#uses=22]
-	%0 = load i32** @a, align 8		; <i32*> [#uses=1]
+	%0 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
 	%1 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
-	%2 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%2 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%3 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
 	%4 = zext i32 %3 to i64		; <i64> [#uses=1]
-	%5 = getelementptr i32* %2, i64 %4		; <i32*> [#uses=1]
-	%6 = load i32* %5, align 1		; <i32> [#uses=1]
-	%7 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%5 = getelementptr i32, i32* %2, i64 %4		; <i32*> [#uses=1]
+	%6 = load i32, i32* %5, align 1		; <i32> [#uses=1]
+	%7 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%8 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
 	%9 = zext i32 %8 to i64		; <i64> [#uses=1]
-	%10 = getelementptr i32* %7, i64 %9		; <i32*> [#uses=1]
-	%11 = load i32* %10, align 1		; <i32> [#uses=1]
+	%10 = getelementptr i32, i32* %7, i64 %9		; <i32*> [#uses=1]
+	%11 = load i32, i32* %10, align 1		; <i32> [#uses=1]
 	%12 = add i32 %11, %6		; <i32> [#uses=1]
 	%13 = zext i32 %1 to i64		; <i64> [#uses=1]
-	%14 = getelementptr i32* %0, i64 %13		; <i32*> [#uses=1]
+	%14 = getelementptr i32, i32* %0, i64 %13		; <i32*> [#uses=1]
 	store i32 %12, i32* %14, align 1
-	%15 = load i32** @a, align 8		; <i32*> [#uses=1]
+	%15 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
 	%16 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%17 = and i32 %16, 15		; <i32> [#uses=1]
-	%18 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%18 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%19 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%20 = and i32 %19, 15		; <i32> [#uses=1]
 	%21 = zext i32 %20 to i64		; <i64> [#uses=1]
-	%22 = getelementptr i32* %18, i64 %21		; <i32*> [#uses=1]
-	%23 = load i32* %22, align 1		; <i32> [#uses=1]
-	%24 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%22 = getelementptr i32, i32* %18, i64 %21		; <i32*> [#uses=1]
+	%23 = load i32, i32* %22, align 1		; <i32> [#uses=1]
+	%24 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%25 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%26 = and i32 %25, 15		; <i32> [#uses=1]
 	%27 = zext i32 %26 to i64		; <i64> [#uses=1]
-	%28 = getelementptr i32* %24, i64 %27		; <i32*> [#uses=1]
-	%29 = load i32* %28, align 1		; <i32> [#uses=1]
+	%28 = getelementptr i32, i32* %24, i64 %27		; <i32*> [#uses=1]
+	%29 = load i32, i32* %28, align 1		; <i32> [#uses=1]
 	%30 = add i32 %29, %23		; <i32> [#uses=1]
 	%31 = zext i32 %17 to i64		; <i64> [#uses=1]
-	%32 = getelementptr i32* %15, i64 %31		; <i32*> [#uses=1]
+	%32 = getelementptr i32, i32* %15, i64 %31		; <i32*> [#uses=1]
 	store i32 %30, i32* %32, align 1
-	%33 = load i32** @a, align 8		; <i32*> [#uses=1]
+	%33 = load i32*, i32** @a, align 8		; <i32*> [#uses=1]
 	%34 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%35 = and i32 %34, 15		; <i32> [#uses=1]
-	%36 = load i32** @b, align 8		; <i32*> [#uses=1]
+	%36 = load i32*, i32** @b, align 8		; <i32*> [#uses=1]
 	%37 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%38 = and i32 %37, 15		; <i32> [#uses=1]
 	%39 = zext i32 %38 to i64		; <i64> [#uses=1]
-	%40 = getelementptr i32* %36, i64 %39		; <i32*> [#uses=1]
-	%41 = load i32* %40, align 1		; <i32> [#uses=1]
-	%42 = load i32** @c, align 8		; <i32*> [#uses=1]
+	%40 = getelementptr i32, i32* %36, i64 %39		; <i32*> [#uses=1]
+	%41 = load i32, i32* %40, align 1		; <i32> [#uses=1]
+	%42 = load i32*, i32** @c, align 8		; <i32*> [#uses=1]
 	%43 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%44 = and i32 %43, 15		; <i32> [#uses=1]
 	%45 = zext i32 %44 to i64		; <i64> [#uses=1]
-	%46 = getelementptr i32* %42, i64 %45		; <i32*> [#uses=1]
-	%47 = load i32* %46, align 1		; <i32> [#uses=1]
+	%46 = getelementptr i32, i32* %42, i64 %45		; <i32*> [#uses=1]
+	%47 = load i32, i32* %46, align 1		; <i32> [#uses=1]
 	%48 = add i32 %47, %41		; <i32> [#uses=1]
 	%49 = zext i32 %35 to i64		; <i64> [#uses=1]
-	%50 = getelementptr i32* %33, i64 %49		; <i32*> [#uses=1]
+	%50 = getelementptr i32, i32* %33, i64 %49		; <i32*> [#uses=1]
 	store i32 %48, i32* %50, align 1
-	%51 = load i32** @d, align 8		; <i32*> [#uses=1]
+	%51 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
 	%52 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
-	%53 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%53 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%54 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
 	%55 = zext i32 %54 to i64		; <i64> [#uses=1]
-	%56 = getelementptr i32* %53, i64 %55		; <i32*> [#uses=1]
-	%57 = load i32* %56, align 1		; <i32> [#uses=1]
-	%58 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%56 = getelementptr i32, i32* %53, i64 %55		; <i32*> [#uses=1]
+	%57 = load i32, i32* %56, align 1		; <i32> [#uses=1]
+	%58 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%59 = and i32 %i.0.reg2mem.0, 15		; <i32> [#uses=1]
 	%60 = zext i32 %59 to i64		; <i64> [#uses=1]
-	%61 = getelementptr i32* %58, i64 %60		; <i32*> [#uses=1]
-	%62 = load i32* %61, align 1		; <i32> [#uses=1]
+	%61 = getelementptr i32, i32* %58, i64 %60		; <i32*> [#uses=1]
+	%62 = load i32, i32* %61, align 1		; <i32> [#uses=1]
 	%63 = sext i32 %i.0.reg2mem.0 to i64		; <i64> [#uses=1]
-	%64 = getelementptr [256 x i32]* @K, i64 0, i64 %63		; <i32*> [#uses=1]
-	%65 = load i32* %64, align 4		; <i32> [#uses=1]
+	%64 = getelementptr [256 x i32], [256 x i32]* @K, i64 0, i64 %63		; <i32*> [#uses=1]
+	%65 = load i32, i32* %64, align 4		; <i32> [#uses=1]
 	%66 = add i32 %62, %57		; <i32> [#uses=1]
 	%67 = add i32 %66, %65		; <i32> [#uses=1]
 	%68 = zext i32 %52 to i64		; <i64> [#uses=1]
-	%69 = getelementptr i32* %51, i64 %68		; <i32*> [#uses=1]
+	%69 = getelementptr i32, i32* %51, i64 %68		; <i32*> [#uses=1]
 	store i32 %67, i32* %69, align 1
-	%70 = load i32** @d, align 8		; <i32*> [#uses=1]
+	%70 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
 	%71 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%72 = and i32 %71, 15		; <i32> [#uses=1]
-	%73 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%73 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%74 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%75 = and i32 %74, 15		; <i32> [#uses=1]
 	%76 = zext i32 %75 to i64		; <i64> [#uses=1]
-	%77 = getelementptr i32* %73, i64 %76		; <i32*> [#uses=1]
-	%78 = load i32* %77, align 1		; <i32> [#uses=1]
-	%79 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%77 = getelementptr i32, i32* %73, i64 %76		; <i32*> [#uses=1]
+	%78 = load i32, i32* %77, align 1		; <i32> [#uses=1]
+	%79 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%80 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%81 = and i32 %80, 15		; <i32> [#uses=1]
 	%82 = zext i32 %81 to i64		; <i64> [#uses=1]
-	%83 = getelementptr i32* %79, i64 %82		; <i32*> [#uses=1]
-	%84 = load i32* %83, align 1		; <i32> [#uses=1]
+	%83 = getelementptr i32, i32* %79, i64 %82		; <i32*> [#uses=1]
+	%84 = load i32, i32* %83, align 1		; <i32> [#uses=1]
 	%85 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=1]
 	%86 = sext i32 %85 to i64		; <i64> [#uses=1]
-	%87 = getelementptr [256 x i32]* @K, i64 0, i64 %86		; <i32*> [#uses=1]
-	%88 = load i32* %87, align 4		; <i32> [#uses=1]
+	%87 = getelementptr [256 x i32], [256 x i32]* @K, i64 0, i64 %86		; <i32*> [#uses=1]
+	%88 = load i32, i32* %87, align 4		; <i32> [#uses=1]
 	%89 = add i32 %84, %78		; <i32> [#uses=1]
 	%90 = add i32 %89, %88		; <i32> [#uses=1]
 	%91 = zext i32 %72 to i64		; <i64> [#uses=1]
-	%92 = getelementptr i32* %70, i64 %91		; <i32*> [#uses=1]
+	%92 = getelementptr i32, i32* %70, i64 %91		; <i32*> [#uses=1]
 	store i32 %90, i32* %92, align 1
-	%93 = load i32** @d, align 8		; <i32*> [#uses=1]
+	%93 = load i32*, i32** @d, align 8		; <i32*> [#uses=1]
 	%94 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%95 = and i32 %94, 15		; <i32> [#uses=1]
-	%96 = load i32** @e, align 8		; <i32*> [#uses=1]
+	%96 = load i32*, i32** @e, align 8		; <i32*> [#uses=1]
 	%97 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%98 = and i32 %97, 15		; <i32> [#uses=1]
 	%99 = zext i32 %98 to i64		; <i64> [#uses=1]
-	%100 = getelementptr i32* %96, i64 %99		; <i32*> [#uses=1]
-	%101 = load i32* %100, align 1		; <i32> [#uses=1]
-	%102 = load i32** @f, align 8		; <i32*> [#uses=1]
+	%100 = getelementptr i32, i32* %96, i64 %99		; <i32*> [#uses=1]
+	%101 = load i32, i32* %100, align 1		; <i32> [#uses=1]
+	%102 = load i32*, i32** @f, align 8		; <i32*> [#uses=1]
 	%103 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%104 = and i32 %103, 15		; <i32> [#uses=1]
 	%105 = zext i32 %104 to i64		; <i64> [#uses=1]
-	%106 = getelementptr i32* %102, i64 %105		; <i32*> [#uses=1]
-	%107 = load i32* %106, align 1		; <i32> [#uses=1]
+	%106 = getelementptr i32, i32* %102, i64 %105		; <i32*> [#uses=1]
+	%107 = load i32, i32* %106, align 1		; <i32> [#uses=1]
 	%108 = add i32 %i.0.reg2mem.0, 2		; <i32> [#uses=1]
 	%109 = sext i32 %108 to i64		; <i64> [#uses=1]
-	%110 = getelementptr [256 x i32]* @K, i64 0, i64 %109		; <i32*> [#uses=1]
-	%111 = load i32* %110, align 4		; <i32> [#uses=1]
+	%110 = getelementptr [256 x i32], [256 x i32]* @K, i64 0, i64 %109		; <i32*> [#uses=1]
+	%111 = load i32, i32* %110, align 4		; <i32> [#uses=1]
 	%112 = add i32 %107, %101		; <i32> [#uses=1]
 	%113 = add i32 %112, %111		; <i32> [#uses=1]
 	%114 = zext i32 %95 to i64		; <i64> [#uses=1]
-	%115 = getelementptr i32* %93, i64 %114		; <i32*> [#uses=1]
+	%115 = getelementptr i32, i32* %93, i64 %114		; <i32*> [#uses=1]
 	store i32 %113, i32* %115, align 1
 	%116 = add i32 %i.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%117 = icmp sgt i32 %116, 23646		; <i1> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/2009-04-27-Floating.ll b/test/Transforms/IndVarSimplify/2009-04-27-Floating.ll
index 47164d86047e..44c43a3a0435 100644
--- a/test/Transforms/IndVarSimplify/2009-04-27-Floating.ll
+++ b/test/Transforms/IndVarSimplify/2009-04-27-Floating.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -indvars -S | FileCheck %s
 ; PR4086
+
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 declare void @foo()
 
 define void @test() {
diff --git a/test/Transforms/IndVarSimplify/2011-09-27-hoistsext.ll b/test/Transforms/IndVarSimplify/2011-09-27-hoistsext.ll
index 0737489aad5b..f0765e7d4c69 100644
--- a/test/Transforms/IndVarSimplify/2011-09-27-hoistsext.ll
+++ b/test/Transforms/IndVarSimplify/2011-09-27-hoistsext.ll
@@ -17,8 +17,8 @@ for.body:
   %i2.115 = phi i32 [ 0, %entry ], [ %add249, %for.body ]
   %add174 = add nsw i32 %i2.115, %x
   %idxprom177 = sext i32 %add174 to i64
-  %arrayidx179 = getelementptr inbounds double* %data, i64 %idxprom177
-  %tmp180 = load double* %arrayidx179, align 8
+  %arrayidx179 = getelementptr inbounds double, double* %data, i64 %idxprom177
+  %tmp180 = load double, double* %arrayidx179, align 8
   %add249 = add nsw i32 %i2.115, %y
   %cmp168 = icmp sgt i32 %add249, %n
   br i1 %cmp168, label %exit, label %for.body
diff --git a/test/Transforms/IndVarSimplify/2011-10-27-lftrnull.ll b/test/Transforms/IndVarSimplify/2011-10-27-lftrnull.ll
index e462712fe7a5..3d77a364f96f 100644
--- a/test/Transforms/IndVarSimplify/2011-10-27-lftrnull.ll
+++ b/test/Transforms/IndVarSimplify/2011-10-27-lftrnull.ll
@@ -6,7 +6,7 @@ target triple = "thumbv7-apple-darwin"
 
 ; CHECK-LABEL: @test(
 ; CHECK: if.end.i126:
-; CHECK: %exitcond = icmp ne i8* %destYPixelPtr.010.i, getelementptr (i8* null, i32 undef)
+; CHECK: %exitcond = icmp ne i8* %incdec.ptr.i, getelementptr (i8, i8* null, i32 undef)
 define void @test() nounwind {
 entry:
   br label %while.cond
@@ -40,7 +40,7 @@ if.else.i124:                                     ; preds = %for.body21.i
   br label %if.end.i126
 
 if.end.i126:                                      ; preds = %if.else.i124, %for.body21.i
-  %incdec.ptr.i = getelementptr inbounds i8* %destYPixelPtr.010.i, i32 1
+  %incdec.ptr.i = getelementptr inbounds i8, i8* %destYPixelPtr.010.i, i32 1
   %inc.i125 = add i32 %x.09.i, 1
   %cmp19.i = icmp ult i32 %inc.i125, undef
   br i1 %cmp19.i, label %for.body21.i, label %for.end.i129
diff --git a/test/Transforms/IndVarSimplify/2011-11-01-lftrptr.ll b/test/Transforms/IndVarSimplify/2011-11-01-lftrptr.ll
index 8247886e66d9..402ae8cc05d0 100644
--- a/test/Transforms/IndVarSimplify/2011-11-01-lftrptr.ll
+++ b/test/Transforms/IndVarSimplify/2011-11-01-lftrptr.ll
@@ -25,13 +25,14 @@ preheader:
 loop:
   %p.01.us.us = phi i8* [ null, %preheader ], [ %gep, %loop ]
   %s = phi i8 [0, %preheader], [%snext, %loop]
-  %gep = getelementptr inbounds i8* %p.01.us.us, i64 1
-  %snext = load i8* %gep
+  %gep = getelementptr inbounds i8, i8* %p.01.us.us, i64 1
+  %snext = load i8, i8* %gep
   %cmp = icmp ult i8* %gep, %end
   br i1 %cmp, label %loop, label %exit
 
 exit:
-  ret i8 %snext
+  %ret = phi i8 [0, %loopguard], [%snext, %loop]
+  ret i8 %ret
 }
 
 ; CHECK-LABEL: @testptrptr(
@@ -50,13 +51,14 @@ preheader:
 loop:
   %p.01.us.us = phi i8* [ %buf, %preheader ], [ %gep, %loop ]
   %s = phi i8 [0, %preheader], [%snext, %loop]
-  %gep = getelementptr inbounds i8* %p.01.us.us, i64 1
-  %snext = load i8* %gep
+  %gep = getelementptr inbounds i8, i8* %p.01.us.us, i64 1
+  %snext = load i8, i8* %gep
   %cmp = icmp ult i8* %gep, %end
   br i1 %cmp, label %loop, label %exit
 
 exit:
-  ret i8 %snext
+  %ret = phi i8 [0, %loopguard], [%snext, %loop]
+  ret i8 %ret
 }
 
 ; CHECK-LABEL: @testnullptrint(
@@ -79,14 +81,15 @@ loop:
   %p.01.us.us = phi i8* [ null, %preheader ], [ %gep, %loop ]
   %iv = phi i32 [ 0, %preheader ], [ %ivnext, %loop ]
   %s = phi i8 [0, %preheader], [%snext, %loop]
-  %gep = getelementptr inbounds i8* %p.01.us.us, i64 1
-  %snext = load i8* %gep
+  %gep = getelementptr inbounds i8, i8* %p.01.us.us, i64 1
+  %snext = load i8, i8* %gep
   %ivnext = add i32 %iv, 1
   %cmp = icmp ult i32 %ivnext, %cnt
   br i1 %cmp, label %loop, label %exit
 
 exit:
-  ret i8 %snext
+  %ret = phi i8 [0, %loopguard], [%snext, %loop]
+  ret i8 %ret
 }
 
 ; CHECK-LABEL: @testptrint(
@@ -109,20 +112,21 @@ loop:
   %p.01.us.us = phi i8* [ %buf, %preheader ], [ %gep, %loop ]
   %iv = phi i32 [ %bi, %preheader ], [ %ivnext, %loop ]
   %s = phi i8 [0, %preheader], [%snext, %loop]
-  %gep = getelementptr inbounds i8* %p.01.us.us, i64 1
-  %snext = load i8* %gep
+  %gep = getelementptr inbounds i8, i8* %p.01.us.us, i64 1
+  %snext = load i8, i8* %gep
   %ivnext = add i32 %iv, 1
   %cmp = icmp ult i32 %ivnext, %cnt
   br i1 %cmp, label %loop, label %exit
 
 exit:
-  ret i8 %snext
+  %ret = phi i8 [0, %loopguard], [%snext, %loop]
+  ret i8 %ret
 }
 
 ; IV and BECount have two different pointer types here.
 define void @testnullptr([512 x i8]* %base) nounwind {
 entry:
-  %add.ptr1603 = getelementptr [512 x i8]* %base, i64 0, i64 512
+  %add.ptr1603 = getelementptr [512 x i8], [512 x i8]* %base, i64 0, i64 512
   br label %preheader
 
 preheader:
@@ -131,7 +135,7 @@ preheader:
 
 for.body:
   %r.17193 = phi i8* [ %incdec.ptr1608, %for.body ], [ null, %preheader ]
-  %incdec.ptr1608 = getelementptr i8* %r.17193, i64 1
+  %incdec.ptr1608 = getelementptr i8, i8* %r.17193, i64 1
   %cmp1604 = icmp ult i8* %incdec.ptr1608, %add.ptr1603
   br i1 %cmp1604, label %for.body, label %for.end1609
 
diff --git a/test/Transforms/IndVarSimplify/2011-11-15-multiexit.ll b/test/Transforms/IndVarSimplify/2011-11-15-multiexit.ll
index c74d04e6a5d8..65b2cf684580 100644
--- a/test/Transforms/IndVarSimplify/2011-11-15-multiexit.ll
+++ b/test/Transforms/IndVarSimplify/2011-11-15-multiexit.ll
@@ -20,12 +20,12 @@ if.then:                                          ; preds = %entry
 
 do.body:                                          ; preds = %if.else, %if.then
   %firstIV = phi i32* [ %incdec.ptr2, %if.else ], [ %first, %if.then ]
-  %incdec.ptr1 = getelementptr inbounds i32* %firstIV, i64 1
+  %incdec.ptr1 = getelementptr inbounds i32, i32* %firstIV, i64 1
   %cmp1 = icmp eq i32* %incdec.ptr1, %last
   br i1 %cmp1, label %early.exit, label %if.else
 
 if.else:                                        ; preds = %do.body
-  %incdec.ptr2 = getelementptr inbounds i32* %firstIV, i64 2
+  %incdec.ptr2 = getelementptr inbounds i32, i32* %firstIV, i64 2
   %cmp2 = icmp eq i32* %incdec.ptr2, %last
   br i1 %cmp2, label %if.end, label %do.body
 
@@ -35,6 +35,6 @@ early.exit:
 
 if.end:
   %tmp = phi i32* [ %first.lcssa, %early.exit ], [ %first, %if.then ], [ %first, %entry ], [ undef, %if.else ]
-  %val = load i32* %tmp
+  %val = load i32, i32* %tmp
   ret i32 %val
 }
diff --git a/test/Transforms/IndVarSimplify/2012-07-17-lftr-undef.ll b/test/Transforms/IndVarSimplify/2012-07-17-lftr-undef.ll
index 643d3cb333ee..faecbfbe07ab 100644
--- a/test/Transforms/IndVarSimplify/2012-07-17-lftr-undef.ll
+++ b/test/Transforms/IndVarSimplify/2012-07-17-lftr-undef.ll
@@ -13,7 +13,7 @@ block9:                                           ; preds = %block9,%func_start
   %undef = phi i64 [ %next_undef, %block9 ], [ undef, %func_start ]
   %iter = phi i64 [ %next_iter, %block9 ], [ 1, %func_start ]
   %next_iter = add nsw i64 %iter, 1
-  %0 = tail call i32 (i8*, ...)* @printf(i8* noalias nocapture getelementptr inbounds ([6 x i8]* @.str3, i64 0, i64 0), i64 %next_iter, i64 %undef)
+  %0 = tail call i32 (i8*, ...) @printf(i8* noalias nocapture getelementptr inbounds ([6 x i8], [6 x i8]* @.str3, i64 0, i64 0), i64 %next_iter, i64 %undef)
   %next_undef = add nsw i64 %undef, 1
   %_tmp_3 = icmp slt i64 %next_iter, 100
   br i1 %_tmp_3, label %block9, label %exit
diff --git a/test/Transforms/IndVarSimplify/2014-06-21-congruent-constant.ll b/test/Transforms/IndVarSimplify/2014-06-21-congruent-constant.ll
index 2c738de32e0a..1d80e751dcab 100644
--- a/test/Transforms/IndVarSimplify/2014-06-21-congruent-constant.ll
+++ b/test/Transforms/IndVarSimplify/2014-06-21-congruent-constant.ll
@@ -13,9 +13,9 @@ entry:
   br i1 undef, label %for.end12, label %for.cond.preheader
 
 for.cond.preheader:                               ; preds = %entry
-  %0 = load i32*** @c, align 8
-  %1 = load i32** %0, align 8
-  %2 = load i32* %1, align 4
+  %0 = load i32**, i32*** @c, align 8
+  %1 = load i32*, i32** %0, align 8
+  %2 = load i32, i32* %1, align 4
   br label %for.body
 
 for.body:                                         ; preds = %for.cond.backedge, %for.body9.us, %for.cond.preheader
@@ -26,7 +26,7 @@ for.body:                                         ; preds = %for.cond.backedge,
   br i1 %tobool1, label %if.end, label %for.cond.backedge
 
 if.end:                                           ; preds = %for.body
-  %5 = load i32* %3, align 4
+  %5 = load i32, i32* %3, align 4
   %tobool4 = icmp eq i32 %5, 0
   br i1 %tobool4, label %for.cond3, label %for.body9.preheader
 
@@ -35,8 +35,8 @@ for.body9.preheader:                              ; preds = %if.end
   br i1 %tobool8, label %for.body9.us, label %for.body9
 
 for.body9.us:                                     ; preds = %for.body9.preheader
-  %6 = load i32** undef, align 8
-  %7 = load i32* %6, align 4
+  %6 = load i32*, i32** undef, align 8
+  %7 = load i32, i32* %6, align 4
   br label %for.body
 
 for.cond3:                                        ; preds = %for.cond3, %if.end
diff --git a/test/Transforms/IndVarSimplify/NVPTX/no-widen-expensive.ll b/test/Transforms/IndVarSimplify/NVPTX/no-widen-expensive.ll
index 8744b19d5e28..ae2cb7ffee8f 100644
--- a/test/Transforms/IndVarSimplify/NVPTX/no-widen-expensive.ll
+++ b/test/Transforms/IndVarSimplify/NVPTX/no-widen-expensive.ll
@@ -23,7 +23,7 @@ for.body:                                         ; preds = %for.body.preheader,
 ; CHECK: phi i32
   %mul = mul nsw i32 %i.06, %i.06
   %0 = sext i32 %i.06 to i64
-  %arrayidx = getelementptr inbounds i32* %output, i64 %0
+  %arrayidx = getelementptr inbounds i32, i32* %output, i64 %0
   store i32 %mul, i32* %arrayidx, align 4
   %add = add nsw i32 %i.06, 3
   %cmp = icmp slt i32 %add, %n
diff --git a/test/Transforms/IndVarSimplify/ada-loops.ll b/test/Transforms/IndVarSimplify/ada-loops.ll
index c0932982ee5e..e452f03bed92 100644
--- a/test/Transforms/IndVarSimplify/ada-loops.ll
+++ b/test/Transforms/IndVarSimplify/ada-loops.ll
@@ -18,7 +18,7 @@ target triple = "i686-pc-linux-gnu"
 
 define void @kinds__sbytezero([256 x i32]* nocapture %a) nounwind {
 bb.thread:
-	%tmp46 = getelementptr [256 x i32]* %a, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp46 = getelementptr [256 x i32], [256 x i32]* %a, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp46
 	br label %bb
 
@@ -27,7 +27,7 @@ bb:		; preds = %bb, %bb.thread
 	%tmp8 = add i8 %i.0.reg2mem.0, 1		; <i8> [#uses=3]
 	%tmp1 = sext i8 %tmp8 to i32		; <i32> [#uses=1]
 	%tmp3 = add i32 %tmp1, 128		; <i32> [#uses=1]
-	%tmp4 = getelementptr [256 x i32]* %a, i32 0, i32 %tmp3		; <i32*> [#uses=1]
+	%tmp4 = getelementptr [256 x i32], [256 x i32]* %a, i32 0, i32 %tmp3		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp4
 	%0 = icmp eq i8 %tmp8, 127		; <i1> [#uses=1]
 	br i1 %0, label %return, label %bb
@@ -38,7 +38,7 @@ return:		; preds = %bb
 
 define void @kinds__ubytezero([256 x i32]* nocapture %a) nounwind {
 bb.thread:
-	%tmp35 = getelementptr [256 x i32]* %a, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp35 = getelementptr [256 x i32], [256 x i32]* %a, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp35
 	br label %bb
 
@@ -46,7 +46,7 @@ bb:		; preds = %bb, %bb.thread
 	%i.0.reg2mem.0 = phi i8 [ 0, %bb.thread ], [ %tmp7, %bb ]		; <i8> [#uses=1]
 	%tmp7 = add i8 %i.0.reg2mem.0, 1		; <i8> [#uses=3]
 	%tmp1 = zext i8 %tmp7 to i32		; <i32> [#uses=1]
-	%tmp3 = getelementptr [256 x i32]* %a, i32 0, i32 %tmp1		; <i32*> [#uses=1]
+	%tmp3 = getelementptr [256 x i32], [256 x i32]* %a, i32 0, i32 %tmp1		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp3
 	%0 = icmp eq i8 %tmp7, -1		; <i1> [#uses=1]
 	br i1 %0, label %return, label %bb
@@ -63,7 +63,7 @@ bb:		; preds = %bb, %bb.thread
 	%i.0.reg2mem.0 = phi i8 [ -10, %bb.thread ], [ %tmp7, %bb ]		; <i8> [#uses=2]
 	%tmp12 = sext i8 %i.0.reg2mem.0 to i32		; <i32> [#uses=1]
 	%tmp4 = add i32 %tmp12, 10		; <i32> [#uses=1]
-	%tmp5 = getelementptr [21 x i32]* %a, i32 0, i32 %tmp4		; <i32*> [#uses=1]
+	%tmp5 = getelementptr [21 x i32], [21 x i32]* %a, i32 0, i32 %tmp4		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp5
 	%tmp7 = add i8 %i.0.reg2mem.0, 1		; <i8> [#uses=2]
 	%0 = icmp sgt i8 %tmp7, 10		; <i1> [#uses=1]
@@ -81,7 +81,7 @@ bb:		; preds = %bb, %bb.thread
 	%i.0.reg2mem.0 = phi i8 [ 10, %bb.thread ], [ %tmp7, %bb ]		; <i8> [#uses=2]
 	%tmp12 = sext i8 %i.0.reg2mem.0 to i32		; <i32> [#uses=1]
 	%tmp4 = add i32 %tmp12, -10		; <i32> [#uses=1]
-	%tmp5 = getelementptr [21 x i32]* %a, i32 0, i32 %tmp4		; <i32*> [#uses=1]
+	%tmp5 = getelementptr [21 x i32], [21 x i32]* %a, i32 0, i32 %tmp4		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp5
 	%tmp7 = add i8 %i.0.reg2mem.0, 1		; <i8> [#uses=2]
 	%0 = icmp sgt i8 %tmp7, 30		; <i1> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/ashr-tripcount.ll b/test/Transforms/IndVarSimplify/ashr-tripcount.ll
index b47c8ad47184..5f337301b404 100644
--- a/test/Transforms/IndVarSimplify/ashr-tripcount.ll
+++ b/test/Transforms/IndVarSimplify/ashr-tripcount.ll
@@ -29,10 +29,10 @@ bb2:		; preds = %bb1, %entry
 	%.pn2.in = and i32 %.pn2.in.in, 3		; <i32> [#uses=1]
 	%.pn3 = sext i32 %.pn3.in to i64		; <i64> [#uses=1]
 	%.pn2 = zext i32 %.pn2.in to i64		; <i64> [#uses=1]
-	%.pn.in = getelementptr [0 x float]* %pow_2_tab.pn, i64 0, i64 %.pn3		; <float*> [#uses=1]
-	%.pn1.in = getelementptr [0 x float]* %pow_2_025_tab.pn, i64 0, i64 %.pn2		; <float*> [#uses=1]
-	%.pn = load float* %.pn.in		; <float> [#uses=1]
-	%.pn1 = load float* %.pn1.in		; <float> [#uses=1]
+	%.pn.in = getelementptr [0 x float], [0 x float]* %pow_2_tab.pn, i64 0, i64 %.pn3		; <float*> [#uses=1]
+	%.pn1.in = getelementptr [0 x float], [0 x float]* %pow_2_025_tab.pn, i64 0, i64 %.pn2		; <float*> [#uses=1]
+	%.pn = load float, float* %.pn.in		; <float> [#uses=1]
+	%.pn1 = load float, float* %.pn1.in		; <float> [#uses=1]
 	%invQuantizer.0 = fmul float %.pn, %.pn1		; <float> [#uses=4]
 	%t3 = ashr i32 %noOfLines, 2		; <i32> [#uses=1]
 	%t4 = icmp sgt i32 %t3, 0		; <i1> [#uses=1]
@@ -46,50 +46,50 @@ bb3:		; preds = %bb4, %bb.nph
 	%i.05 = phi i32 [ %t49, %bb4 ], [ 0, %bb.nph ]		; <i32> [#uses=9]
 	%k.04 = phi i32 [ %t48, %bb4 ], [ 0, %bb.nph ]		; <i32> [#uses=1]
 	%t6 = sext i32 %i.05 to i64		; <i64> [#uses=1]
-	%t7 = getelementptr i32* %quaSpectrum, i64 %t6		; <i32*> [#uses=1]
-	%t8 = load i32* %t7, align 4		; <i32> [#uses=1]
+	%t7 = getelementptr i32, i32* %quaSpectrum, i64 %t6		; <i32*> [#uses=1]
+	%t8 = load i32, i32* %t7, align 4		; <i32> [#uses=1]
 	%t9 = zext i32 %t8 to i64		; <i64> [#uses=1]
-	%t10 = getelementptr float* %pow4_3_tab_ptr, i64 %t9		; <float*> [#uses=1]
-	%t11 = load float* %t10, align 4		; <float> [#uses=1]
+	%t10 = getelementptr float, float* %pow4_3_tab_ptr, i64 %t9		; <float*> [#uses=1]
+	%t11 = load float, float* %t10, align 4		; <float> [#uses=1]
 	%t12 = or i32 %i.05, 1		; <i32> [#uses=1]
 	%t13 = sext i32 %t12 to i64		; <i64> [#uses=1]
-	%t14 = getelementptr i32* %quaSpectrum, i64 %t13		; <i32*> [#uses=1]
-	%t15 = load i32* %t14, align 4		; <i32> [#uses=1]
+	%t14 = getelementptr i32, i32* %quaSpectrum, i64 %t13		; <i32*> [#uses=1]
+	%t15 = load i32, i32* %t14, align 4		; <i32> [#uses=1]
 	%t16 = zext i32 %t15 to i64		; <i64> [#uses=1]
-	%t17 = getelementptr float* %pow4_3_tab_ptr, i64 %t16		; <float*> [#uses=1]
-	%t18 = load float* %t17, align 4		; <float> [#uses=1]
+	%t17 = getelementptr float, float* %pow4_3_tab_ptr, i64 %t16		; <float*> [#uses=1]
+	%t18 = load float, float* %t17, align 4		; <float> [#uses=1]
 	%t19 = or i32 %i.05, 2		; <i32> [#uses=1]
 	%t20 = sext i32 %t19 to i64		; <i64> [#uses=1]
-	%t21 = getelementptr i32* %quaSpectrum, i64 %t20		; <i32*> [#uses=1]
-	%t22 = load i32* %t21, align 4		; <i32> [#uses=1]
+	%t21 = getelementptr i32, i32* %quaSpectrum, i64 %t20		; <i32*> [#uses=1]
+	%t22 = load i32, i32* %t21, align 4		; <i32> [#uses=1]
 	%t23 = zext i32 %t22 to i64		; <i64> [#uses=1]
-	%t24 = getelementptr float* %pow4_3_tab_ptr, i64 %t23		; <float*> [#uses=1]
-	%t25 = load float* %t24, align 4		; <float> [#uses=1]
+	%t24 = getelementptr float, float* %pow4_3_tab_ptr, i64 %t23		; <float*> [#uses=1]
+	%t25 = load float, float* %t24, align 4		; <float> [#uses=1]
 	%t26 = or i32 %i.05, 3		; <i32> [#uses=1]
 	%t27 = sext i32 %t26 to i64		; <i64> [#uses=1]
-	%t28 = getelementptr i32* %quaSpectrum, i64 %t27		; <i32*> [#uses=1]
-	%t29 = load i32* %t28, align 4		; <i32> [#uses=1]
+	%t28 = getelementptr i32, i32* %quaSpectrum, i64 %t27		; <i32*> [#uses=1]
+	%t29 = load i32, i32* %t28, align 4		; <i32> [#uses=1]
 	%t30 = zext i32 %t29 to i64		; <i64> [#uses=1]
-	%t31 = getelementptr float* %pow4_3_tab_ptr, i64 %t30		; <float*> [#uses=1]
-	%t32 = load float* %t31, align 4		; <float> [#uses=1]
+	%t31 = getelementptr float, float* %pow4_3_tab_ptr, i64 %t30		; <float*> [#uses=1]
+	%t32 = load float, float* %t31, align 4		; <float> [#uses=1]
 	%t33 = fmul float %t11, %invQuantizer.0		; <float> [#uses=1]
 	%t34 = sext i32 %i.05 to i64		; <i64> [#uses=1]
-	%t35 = getelementptr float* %iquaSpectrum, i64 %t34		; <float*> [#uses=1]
+	%t35 = getelementptr float, float* %iquaSpectrum, i64 %t34		; <float*> [#uses=1]
 	store float %t33, float* %t35, align 4
 	%t36 = or i32 %i.05, 1		; <i32> [#uses=1]
 	%t37 = fmul float %t18, %invQuantizer.0		; <float> [#uses=1]
 	%t38 = sext i32 %t36 to i64		; <i64> [#uses=1]
-	%t39 = getelementptr float* %iquaSpectrum, i64 %t38		; <float*> [#uses=1]
+	%t39 = getelementptr float, float* %iquaSpectrum, i64 %t38		; <float*> [#uses=1]
 	store float %t37, float* %t39, align 4
 	%t40 = or i32 %i.05, 2		; <i32> [#uses=1]
 	%t41 = fmul float %t25, %invQuantizer.0		; <float> [#uses=1]
 	%t42 = sext i32 %t40 to i64		; <i64> [#uses=1]
-	%t43 = getelementptr float* %iquaSpectrum, i64 %t42		; <float*> [#uses=1]
+	%t43 = getelementptr float, float* %iquaSpectrum, i64 %t42		; <float*> [#uses=1]
 	store float %t41, float* %t43, align 4
 	%t44 = or i32 %i.05, 3		; <i32> [#uses=1]
 	%t45 = fmul float %t32, %invQuantizer.0		; <float> [#uses=1]
 	%t46 = sext i32 %t44 to i64		; <i64> [#uses=1]
-	%t47 = getelementptr float* %iquaSpectrum, i64 %t46		; <float*> [#uses=1]
+	%t47 = getelementptr float, float* %iquaSpectrum, i64 %t46		; <float*> [#uses=1]
 	store float %t45, float* %t47, align 4
 	%t48 = add i32 %k.04, 1		; <i32> [#uses=2]
 	%t49 = add i32 %i.05, 4		; <i32> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/avoid-i0.ll b/test/Transforms/IndVarSimplify/avoid-i0.ll
index 22f2e4b718c2..cc38590c0998 100644
--- a/test/Transforms/IndVarSimplify/avoid-i0.ll
+++ b/test/Transforms/IndVarSimplify/avoid-i0.ll
@@ -34,25 +34,25 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store i32 %_si1, i32* %_si1_addr
 	store i8 %_si2, i8* %_si2_addr
-	%1 = load i8* %_si2_addr, align 1		; <i8> [#uses=1]
+	%1 = load i8, i8* %_si2_addr, align 1		; <i8> [#uses=1]
 	%2 = sext i8 %1 to i32		; <i32> [#uses=1]
-	%3 = load i32* %_si1_addr, align 4		; <i32> [#uses=1]
+	%3 = load i32, i32* %_si1_addr, align 4		; <i32> [#uses=1]
 	%4 = xor i32 %2, %3		; <i32> [#uses=1]
-	%5 = load i8* %_si2_addr, align 1		; <i8> [#uses=1]
+	%5 = load i8, i8* %_si2_addr, align 1		; <i8> [#uses=1]
 	%6 = sext i8 %5 to i32		; <i32> [#uses=1]
 	%7 = sub i32 7, %6		; <i32> [#uses=1]
-	%8 = load i32* %_si1_addr, align 4		; <i32> [#uses=1]
+	%8 = load i32, i32* %_si1_addr, align 4		; <i32> [#uses=1]
 	%9 = shl i32 %8, %7		; <i32> [#uses=1]
 	%10 = and i32 %4, %9		; <i32> [#uses=1]
 	%11 = icmp slt i32 %10, 0		; <i1> [#uses=1]
 	%12 = zext i1 %11 to i32		; <i32> [#uses=1]
 	store i32 %12, i32* %0, align 4
-	%13 = load i32* %0, align 4		; <i32> [#uses=1]
+	%13 = load i32, i32* %0, align 4		; <i32> [#uses=1]
 	store i32 %13, i32* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load i32* %retval		; <i32> [#uses=1]
+	%retval1 = load i32, i32* %retval		; <i32> [#uses=1]
 	%retval12 = trunc i32 %retval1 to i8		; <i8> [#uses=1]
 	ret i8 %retval12
 }
@@ -66,15 +66,15 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store i32 %_ui1, i32* %_ui1_addr
 	store i32 %_ui2, i32* %_ui2_addr
-	%1 = load i32* %_ui1_addr, align 4		; <i32> [#uses=1]
+	%1 = load i32, i32* %_ui1_addr, align 4		; <i32> [#uses=1]
 	%2 = sub i32 %1, 1		; <i32> [#uses=1]
 	store i32 %2, i32* %0, align 4
-	%3 = load i32* %0, align 4		; <i32> [#uses=1]
+	%3 = load i32, i32* %0, align 4		; <i32> [#uses=1]
 	store i32 %3, i32* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load i32* %retval		; <i32> [#uses=1]
+	%retval1 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval1
 }
 
@@ -90,31 +90,31 @@ entry:
 	br label %bb4
 
 bb:		; preds = %bb4
-	%0 = load volatile i32* @x, align 4		; <i32> [#uses=1]
+	%0 = load volatile i32, i32* @x, align 4		; <i32> [#uses=1]
 	store i32 %0, i32* %vol.0, align 4
 	store i32 0, i32* %l_52, align 4
 	br label %bb2
 
 bb1:		; preds = %bb2
-	%1 = load i32* %l_52, align 4		; <i32> [#uses=1]
+	%1 = load i32, i32* %l_52, align 4		; <i32> [#uses=1]
 	%2 = call i32 @safe_sub_func_uint64_t_u_u(i32 %1, i32 1) nounwind		; <i32> [#uses=1]
 	store i32 %2, i32* %l_52, align 4
 	br label %bb2
 
 bb2:		; preds = %bb1, %bb
-	%3 = load i32* %l_52, align 4		; <i32> [#uses=1]
+	%3 = load i32, i32* %l_52, align 4		; <i32> [#uses=1]
 	%4 = icmp eq i32 %3, 0		; <i1> [#uses=1]
 	br i1 %4, label %bb1, label %bb3
 
 bb3:		; preds = %bb2
-	%5 = load i32* %l_52, align 4		; <i32> [#uses=1]
+	%5 = load i32, i32* %l_52, align 4		; <i32> [#uses=1]
 	%6 = call signext i8 @safe_sub_func_int32_t_s_s(i32 %5, i8 signext 1) nounwind		; <i8> [#uses=1]
 	%7 = sext i8 %6 to i32		; <i32> [#uses=1]
 	store i32 %7, i32* %l_52, align 4
 	br label %bb4
 
 bb4:		; preds = %bb3, %entry
-	%8 = load i32* %l_52, align 4		; <i32> [#uses=1]
+	%8 = load i32, i32* %l_52, align 4		; <i32> [#uses=1]
 	%9 = icmp ne i32 %8, 0		; <i1> [#uses=1]
 	br i1 %9, label %bb, label %bb5
 
diff --git a/test/Transforms/IndVarSimplify/backedge-on-min-max.ll b/test/Transforms/IndVarSimplify/backedge-on-min-max.ll
index 250ff9a2eca6..bb26ca5bafb0 100644
--- a/test/Transforms/IndVarSimplify/backedge-on-min-max.ll
+++ b/test/Transforms/IndVarSimplify/backedge-on-min-max.ll
@@ -18,7 +18,7 @@ define void @min.signed.1(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -46,7 +46,7 @@ define void @min.signed.2(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -74,7 +74,7 @@ define void @min.signed.3(i32* %a, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -102,7 +102,7 @@ define void @min.signed.4(i32* %a, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -130,7 +130,7 @@ define void @max.signed.1(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -158,7 +158,7 @@ define void @max.signed.2(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -186,7 +186,7 @@ define void @max.signed.3(i32* %a, i32 %n, i32 %init) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -214,7 +214,7 @@ define void @max.signed.4(i32* %a, i32 %n, i32 %init) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -244,7 +244,7 @@ define void @min.unsigned.1(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -272,7 +272,7 @@ define void @min.unsigned.2(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -300,7 +300,7 @@ define void @min.unsigned.3(i32* %a, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -328,7 +328,7 @@ define void @min.unsigned.4(i32* %a, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -356,7 +356,7 @@ define void @max.unsigned.1(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -384,7 +384,7 @@ define void @max.unsigned.2(i32* %a, i32 %a_len, i32 %n) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -412,7 +412,7 @@ define void @max.unsigned.3(i32* %a, i32 %n, i32 %init) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
@@ -440,7 +440,7 @@ define void @max.unsigned.4(i32* %a, i32 %n, i32 %init) {
 ; CHECK: br i1 true, label %ok, label %latch
 
  ok:
-  %addr = getelementptr i32* %a, i32 %idx
+  %addr = getelementptr i32, i32* %a, i32 %idx
   store i32 %idx, i32* %addr
   br label %latch
 
diff --git a/test/Transforms/IndVarSimplify/casted-argument.ll b/test/Transforms/IndVarSimplify/casted-argument.ll
index a5e002b4578f..9d868e800e98 100644
--- a/test/Transforms/IndVarSimplify/casted-argument.ll
+++ b/test/Transforms/IndVarSimplify/casted-argument.ll
@@ -17,7 +17,7 @@ loop:		; preds = %loop, %if.else
 	%pn = phi i8* [ %ge, %loop ], [ null, %entry ]		; <i8*> [#uses=1]
 	%cp = ptrtoint i8* %to to i32		; <i32> [#uses=1]
 	%su = sub i32 0, %cp		; <i32> [#uses=1]
-	%ge = getelementptr i8* %pn, i32 %su		; <i8*> [#uses=2]
+	%ge = getelementptr i8, i8* %pn, i32 %su		; <i8*> [#uses=2]
 	tail call void @bcopy(i8* %ge) nounwind
 	br label %loop
 }
@@ -39,9 +39,9 @@ if.end54:		; preds = %if.end54, %if.else
 	%sub.ptr.rhs.cast46.pn = ptrtoint i8* %from to i32		; <i32> [#uses=1]
 	%sub.ptr.lhs.cast45.pn = ptrtoint i8* %to to i32		; <i32> [#uses=1]
 	%sub.ptr.sub47.pn = sub i32 %sub.ptr.rhs.cast46.pn, %sub.ptr.lhs.cast45.pn		; <i32> [#uses=1]
-	%sub.ptr4912 = getelementptr i8* %sub.ptr4912.pn, i32 %sub.ptr.sub47.pn		; <i8*> [#uses=2]
+	%sub.ptr4912 = getelementptr i8, i8* %sub.ptr4912.pn, i32 %sub.ptr.sub47.pn		; <i8*> [#uses=2]
 	tail call void @bcopy_4038(i8* %sub.ptr4912, i8* %sub.ptr7, i32 0) nounwind
-	%sub.ptr = getelementptr i8* %sub.ptr7, i32 %sub.ptr.rhs.cast40		; <i8*> [#uses=1]
+	%sub.ptr = getelementptr i8, i8* %sub.ptr7, i32 %sub.ptr.rhs.cast40		; <i8*> [#uses=1]
 	br label %if.end54
 }
 
diff --git a/test/Transforms/IndVarSimplify/dangling-use.ll b/test/Transforms/IndVarSimplify/dangling-use.ll
index 51c31204c6df..24a68a9e82c8 100644
--- a/test/Transforms/IndVarSimplify/dangling-use.ll
+++ b/test/Transforms/IndVarSimplify/dangling-use.ll
@@ -15,7 +15,7 @@ bb49:                                             ; preds = %bb48, %entry
 
 bb10:                                             ; preds = %bb49
   %tmp326 = mul nsw i32 %tmp1, %tmp2              ; <i32> [#uses=1]
-  %tmp351 = getelementptr inbounds i8* %x_addr.0, i32 %tmp326 ; <i8*> [#uses=1]
+  %tmp351 = getelementptr inbounds i8, i8* %x_addr.0, i32 %tmp326 ; <i8*> [#uses=1]
   br i1 false, label %bb.nph, label %bb48
 
 bb.nph:                                           ; preds = %bb10
@@ -25,7 +25,7 @@ bb23:                                             ; preds = %bb28, %bb.nph
   %pOriginHi.01 = phi i8* [ %tmp351, %bb.nph ], [ %pOriginHi.0, %bb28 ] ; <i8*> [#uses=2]
   %tmp378 = bitcast i8* %pOriginHi.01 to i8*      ; <i8*> [#uses=1]
   store i8* %tmp378, i8** null
-  %tmp385 = getelementptr inbounds i8* %pOriginHi.01, i32 %tmp2 ; <i8*> [#uses=1]
+  %tmp385 = getelementptr inbounds i8, i8* %pOriginHi.01, i32 %tmp2 ; <i8*> [#uses=1]
   br label %bb28
 
 bb28:                                             ; preds = %bb23
@@ -36,6 +36,6 @@ bb28.bb48_crit_edge:                              ; preds = %bb28
   br label %bb48
 
 bb48:                                             ; preds = %bb28.bb48_crit_edge, %bb10
-  %tmp481 = getelementptr inbounds i8* %x_addr.0, i32 1 ; <i8*> [#uses=1]
+  %tmp481 = getelementptr inbounds i8, i8* %x_addr.0, i32 1 ; <i8*> [#uses=1]
   br label %bb49
 }
diff --git a/test/Transforms/IndVarSimplify/elim-extend.ll b/test/Transforms/IndVarSimplify/elim-extend.ll
index ad5679f3e7a4..98701c338520 100644
--- a/test/Transforms/IndVarSimplify/elim-extend.ll
+++ b/test/Transforms/IndVarSimplify/elim-extend.ll
@@ -14,15 +14,15 @@ loop:
   %iv = phi i32 [ %postiv, %loop ], [ 0, %entry ]
   %ivnsw = phi i32 [ %postivnsw, %loop ], [ 0, %entry ]
   %preofs = sext i32 %iv to i64
-  %preadr = getelementptr i8* %base, i64 %preofs
+  %preadr = getelementptr i8, i8* %base, i64 %preofs
   store i8 0, i8* %preadr
   %postiv = add i32 %iv, 1
   %postofs = sext i32 %postiv to i64
-  %postadr = getelementptr i8* %base, i64 %postofs
+  %postadr = getelementptr i8, i8* %base, i64 %postofs
   store i8 0, i8* %postadr
   %postivnsw = add nsw i32 %ivnsw, 1
   %postofsnsw = sext i32 %postivnsw to i64
-  %postadrnsw = getelementptr i8* %base, i64 %postofsnsw
+  %postadrnsw = getelementptr i8, i8* %base, i64 %postofsnsw
   store i8 0, i8* %postadrnsw
   %cond = icmp sgt i32 %limit, %iv
   br i1 %cond, label %loop, label %exit
@@ -46,15 +46,15 @@ loop:
   %iv = phi i32 [ %postiv, %loop ], [ %init, %entry ]
   %ivnsw = phi i32 [ %postivnsw, %loop ], [ %init, %entry ]
   %preofs = sext i32 %iv to i64
-  %preadr = getelementptr i8* %base, i64 %preofs
+  %preadr = getelementptr i8, i8* %base, i64 %preofs
   store i8 0, i8* %preadr
   %postiv = add i32 %iv, 1
   %postofs = sext i32 %postiv to i64
-  %postadr = getelementptr i8* %base, i64 %postofs
+  %postadr = getelementptr i8, i8* %base, i64 %postofs
   store i8 0, i8* %postadr
   %postivnsw = add nsw i32 %ivnsw, 1
   %postofsnsw = sext i32 %postivnsw to i64
-  %postadrnsw = getelementptr i8* %base, i64 %postofsnsw
+  %postadrnsw = getelementptr i8, i8* %base, i64 %postofsnsw
   store i8 0, i8* %postadrnsw
   %cond = icmp sgt i32 %limit, %postiv
   br i1 %cond, label %loop, label %exit
@@ -88,7 +88,7 @@ outerloop:
 
   %outercountdec = add i32 %outercount, -1
   %ofs1 = sext i32 %outercountdec to i64
-  %adr1 = getelementptr i8* %address, i64 %ofs1
+  %adr1 = getelementptr i8, i8* %address, i64 %ofs1
   store i8 0, i8* %adr1
 
   br label %innerpreheader
@@ -112,11 +112,11 @@ innerloop:
   %innerpostiv = add i32 %inneriv, 1
 
   %ofs2 = sext i32 %inneriv to i64
-  %adr2 = getelementptr i8* %address, i64 %ofs2
+  %adr2 = getelementptr i8, i8* %address, i64 %ofs2
   store i8 0, i8* %adr2
 
   %ofs3 = sext i32 %innerpostiv to i64
-  %adr3 = getelementptr i8* %address, i64 %ofs3
+  %adr3 = getelementptr i8, i8* %address, i64 %ofs3
   store i8 0, i8* %adr3
 
   %innercmp = icmp sgt i32 %limitdec, %innerpostiv
@@ -137,11 +137,11 @@ outermerge:
   %innercount.merge = phi i32 [ %innercount.lcssa, %innerexit ], [ %innercount, %innerpreheader ]
 
   %ofs4 = sext i32 %outercount to i64
-  %adr4 = getelementptr i8* %address, i64 %ofs4
+  %adr4 = getelementptr i8, i8* %address, i64 %ofs4
   store i8 0, i8* %adr4
 
   %ofs5 = sext i32 %innercount.merge to i64
-  %adr5 = getelementptr i8* %address, i64 %ofs5
+  %adr5 = getelementptr i8, i8* %address, i64 %ofs5
   store i8 0, i8* %adr5
 
   %outerpostcount = add i32 %outercount, 1
diff --git a/test/Transforms/IndVarSimplify/eliminate-comparison.ll b/test/Transforms/IndVarSimplify/eliminate-comparison.ll
index b48403e9a23b..4d14b3681c5d 100644
--- a/test/Transforms/IndVarSimplify/eliminate-comparison.ll
+++ b/test/Transforms/IndVarSimplify/eliminate-comparison.ll
@@ -16,7 +16,7 @@ entry:
   br i1 %cmp9, label %pre, label %return
 
 pre:
-  %t3 = load i32* %p
+  %t3 = load i32, i32* %p
   %tobool.not = icmp ne i32 %t3, 0
   br label %loop
 
@@ -27,7 +27,7 @@ loop:
   br i1 %cond, label %if.then, label %for.inc
 
 if.then:
-  %arrayidx = getelementptr [0 x double]* @X, i64 0, i64 %i
+  %arrayidx = getelementptr [0 x double], [0 x double]* @X, i64 0, i64 %i
   store double 3.200000e+00, double* %arrayidx
   br label %for.inc
 
@@ -50,22 +50,22 @@ entry:
   br label %bb18
 
 bb13:
-  %tmp66 = load i64** %tmp65, align 4
-  %tmp68 = getelementptr inbounds i64* %tmp66, i32 %i
-  %tmp69 = load i64* %tmp68, align 4
-  %tmp74 = load i64** %tmp73, align 4
-  %tmp76 = getelementptr inbounds i64* %tmp74, i32 %i
-  %tmp77 = load i64* %tmp76, align 4
+  %tmp66 = load i64*, i64** %tmp65, align 4
+  %tmp68 = getelementptr inbounds i64, i64* %tmp66, i32 %i
+  %tmp69 = load i64, i64* %tmp68, align 4
+  %tmp74 = load i64*, i64** %tmp73, align 4
+  %tmp76 = getelementptr inbounds i64, i64* %tmp74, i32 %i
+  %tmp77 = load i64, i64* %tmp76, align 4
   %tmp78 = icmp ugt i64 %tmp69, %tmp77
   br i1 %tmp78, label %bb20.loopexit, label %bb15
 
 bb15:
-  %tmp83 = load i64** %tmp82, align 4
-  %tmp85 = getelementptr inbounds i64* %tmp83, i32 %i
-  %tmp86 = load i64* %tmp85, align 4
-  %tmp91 = load i64** %tmp90, align 4
-  %tmp93 = getelementptr inbounds i64* %tmp91, i32 %i
-  %tmp94 = load i64* %tmp93, align 4
+  %tmp83 = load i64*, i64** %tmp82, align 4
+  %tmp85 = getelementptr inbounds i64, i64* %tmp83, i32 %i
+  %tmp86 = load i64, i64* %tmp85, align 4
+  %tmp91 = load i64*, i64** %tmp90, align 4
+  %tmp93 = getelementptr inbounds i64, i64* %tmp91, i32 %i
+  %tmp94 = load i64, i64* %tmp93, align 4
   %tmp95 = icmp ult i64 %tmp86, %tmp94
   br i1 %tmp95, label %bb20.loopexit, label %bb17
 
diff --git a/test/Transforms/IndVarSimplify/eliminate-max.ll b/test/Transforms/IndVarSimplify/eliminate-max.ll
index 98510eaed3f4..5831b9378a9a 100644
--- a/test/Transforms/IndVarSimplify/eliminate-max.ll
+++ b/test/Transforms/IndVarSimplify/eliminate-max.ll
@@ -45,7 +45,7 @@ bb14:                                             ; preds = %bb11, %bb7
   br i1 %t20, label %bb1, label %bb21
 
 bb21:                                             ; preds = %bb14
-  %t22 = call i32 (i8*, ...)* @printf(i8* noalias getelementptr inbounds ([4 x i8]* @0, i32 0, i32 0), i32 %t18) nounwind
+  %t22 = call i32 (i8*, ...) @printf(i8* noalias getelementptr inbounds ([4 x i8], [4 x i8]* @0, i32 0, i32 0), i32 %t18) nounwind
   ret i32 0
 }
 
diff --git a/test/Transforms/IndVarSimplify/eliminate-rem.ll b/test/Transforms/IndVarSimplify/eliminate-rem.ll
index 64fe7107b633..6f8e6bbac775 100644
--- a/test/Transforms/IndVarSimplify/eliminate-rem.ll
+++ b/test/Transforms/IndVarSimplify/eliminate-rem.ll
@@ -18,7 +18,7 @@ bb4:                                              ; preds = %bb
 bb5:                                              ; preds = %bb4, %bb5
   %t6 = phi i64 [ %t9, %bb5 ], [ 0, %bb4 ]    ; <i64> [#uses=2]
   %t7 = srem i64 %t6, %arg                    ; <i64> [#uses=1]
-  %t8 = getelementptr inbounds double* %arg3, i64 %t7 ; <double*> [#uses=1]
+  %t8 = getelementptr inbounds double, double* %arg3, i64 %t7 ; <double*> [#uses=1]
   store double 0.000000e+00, double* %t8
   %t9 = add nsw i64 %t6, 1                    ; <i64> [#uses=2]
   %t10 = icmp slt i64 %t9, %arg               ; <i1> [#uses=1]
@@ -78,16 +78,16 @@ bb21:                                             ; preds = %bb21, %bb20
   %t26 = add nsw i64 %t24, %t22                   ; <i64> [#uses=1]
   %t27 = mul i64 %t11, %arg1                      ; <i64> [#uses=1]
   %t28 = add nsw i64 %t25, %t22                   ; <i64> [#uses=1]
-  %t29 = getelementptr inbounds i64* %arg, i64 %t26 ; <i64*> [#uses=1]
+  %t29 = getelementptr inbounds i64, i64* %arg, i64 %t26 ; <i64*> [#uses=1]
   %t30 = add nsw i64 %t27, %t22                   ; <i64> [#uses=1]
-  %t31 = getelementptr inbounds i64* %arg, i64 %t28 ; <i64*> [#uses=1]
+  %t31 = getelementptr inbounds i64, i64* %arg, i64 %t28 ; <i64*> [#uses=1]
   %t32 = zext i32 %t23 to i64                     ; <i64> [#uses=1]
-  %t33 = load i64* %t29                           ; <i64> [#uses=1]
-  %t34 = getelementptr inbounds i64* %arg, i64 %t30 ; <i64*> [#uses=1]
-  %t35 = load i64* %t31                           ; <i64> [#uses=1]
+  %t33 = load i64, i64* %t29                           ; <i64> [#uses=1]
+  %t34 = getelementptr inbounds i64, i64* %arg, i64 %t30 ; <i64*> [#uses=1]
+  %t35 = load i64, i64* %t31                           ; <i64> [#uses=1]
   %t36 = add nsw i64 %t32, %t33                   ; <i64> [#uses=1]
   %t37 = add nsw i64 %t36, %t35                   ; <i64> [#uses=1]
-  %t38 = load i64* %t34                           ; <i64> [#uses=1]
+  %t38 = load i64, i64* %t34                           ; <i64> [#uses=1]
   %t39 = add nsw i64 %t37, %t38                   ; <i64> [#uses=1]
   %t40 = trunc i64 %t39 to i32                    ; <i32> [#uses=2]
   %t41 = add nsw i64 %t22, 1                      ; <i64> [#uses=2]
diff --git a/test/Transforms/IndVarSimplify/indirectbr.ll b/test/Transforms/IndVarSimplify/indirectbr.ll
index a208ded22ed1..d5801690d5e0 100644
--- a/test/Transforms/IndVarSimplify/indirectbr.ll
+++ b/test/Transforms/IndVarSimplify/indirectbr.ll
@@ -32,8 +32,8 @@ bb14:                                             ; preds = %bb14, %bb7.preheade
 bb16:                                             ; preds = %bb16, %bb14, %bb7.preheader
   %S.31.0 = phi i64 [ %3, %bb16 ], [ 1, %bb7.preheader ], [ 1, %bb14 ] ; <i64> [#uses=2]
   %0 = add nsw i64 %S.31.0, -1                    ; <i64> [#uses=1]
-  %1 = getelementptr inbounds [3 x double]* undef, i64 0, i64 %0 ; <double*> [#uses=1]
-  %2 = load double* %1, align 8                   ; <double> [#uses=0]
+  %1 = getelementptr inbounds [3 x double], [3 x double]* undef, i64 0, i64 %0 ; <double*> [#uses=1]
+  %2 = load double, double* %1, align 8                   ; <double> [#uses=0]
   %3 = add nsw i64 %S.31.0, 1                     ; <i64> [#uses=1]
   br label %bb16
 }
diff --git a/test/Transforms/IndVarSimplify/iv-fold.ll b/test/Transforms/IndVarSimplify/iv-fold.ll
index 41a1f5ff91a2..af8a33b30ea9 100644
--- a/test/Transforms/IndVarSimplify/iv-fold.ll
+++ b/test/Transforms/IndVarSimplify/iv-fold.ll
@@ -13,12 +13,12 @@ entry:
 while.body:
   %0 = phi i32 [ 0, %entry ], [ %inc.2, %while.body ]
   %shr = lshr i32 %0, 5
-  %arrayidx = getelementptr inbounds i32* %bitmap, i32 %shr
-  %tmp6 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %bitmap, i32 %shr
+  %tmp6 = load i32, i32* %arrayidx, align 4
   %inc.1 = add i32 %0, 1
   %shr.1 = lshr i32 %inc.1, 5
-  %arrayidx.1 = getelementptr inbounds i32* %bitmap, i32 %shr.1
-  %tmp6.1 = load i32* %arrayidx.1, align 4
+  %arrayidx.1 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.1
+  %tmp6.1 = load i32, i32* %arrayidx.1, align 4
   %inc.2 = add i32 %inc.1, 1
   %exitcond.3 = icmp eq i32 %inc.2, 128
   br i1 %exitcond.3, label %while.end, label %while.body
@@ -40,12 +40,12 @@ entry:
 while.body:
   %0 = phi i32 [ 0, %entry ], [ %inc.3, %while.body ]
   %shr = lshr i32 %0, 5
-  %arrayidx = getelementptr inbounds i32* %bitmap, i32 %shr
-  %tmp6 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %bitmap, i32 %shr
+  %tmp6 = load i32, i32* %arrayidx, align 4
   %inc.1 = add i32 %0, 1
   %shr.1 = lshr i32 %inc.1, 5
-  %arrayidx.1 = getelementptr inbounds i32* %bitmap, i32 %shr.1
-  %tmp6.1 = load i32* %arrayidx.1, align 4
+  %arrayidx.1 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.1
+  %tmp6.1 = load i32, i32* %arrayidx.1, align 4
   %inc.3 = add i32 %inc.1, 2
   %exitcond.3 = icmp eq i32 %inc.3, 96
   br i1 %exitcond.3, label %while.end, label %while.body
diff --git a/test/Transforms/IndVarSimplify/iv-sext.ll b/test/Transforms/IndVarSimplify/iv-sext.ll
index 04df0f9aa5fa..89e21e1b2690 100644
--- a/test/Transforms/IndVarSimplify/iv-sext.ll
+++ b/test/Transforms/IndVarSimplify/iv-sext.ll
@@ -8,7 +8,7 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 define void @t(float* %pTmp1, float* %peakWeight, float* %nrgReducePeakrate, i32 %bandEdgeIndex, float %tmp1) nounwind {
 entry:
-	%tmp = load float* %peakWeight, align 4		; <float> [#uses=1]
+	%tmp = load float, float* %peakWeight, align 4		; <float> [#uses=1]
 	%tmp2 = icmp sgt i32 %bandEdgeIndex, 0		; <i1> [#uses=1]
 	br i1 %tmp2, label %bb.nph22, label %return
 
@@ -33,13 +33,13 @@ bb:		; preds = %bb8, %bb.nph22
 bb1:		; preds = %bb
 	%tmp5 = add i32 %part.016, -1		; <i32> [#uses=1]
 	%tmp6 = sext i32 %tmp5 to i64		; <i64> [#uses=1]
-	%tmp7 = getelementptr float* %pTmp1, i64 %tmp6		; <float*> [#uses=1]
-	%tmp8 = load float* %tmp7, align 4		; <float> [#uses=1]
+	%tmp7 = getelementptr float, float* %pTmp1, i64 %tmp6		; <float*> [#uses=1]
+	%tmp8 = load float, float* %tmp7, align 4		; <float> [#uses=1]
 	%tmp9 = fadd float %tmp8, %distERBlo.120		; <float> [#uses=1]
 	%tmp10 = add i32 %part.016, -1		; <i32> [#uses=1]
 	%tmp11 = sext i32 %tmp10 to i64		; <i64> [#uses=1]
-	%tmp12 = getelementptr float* %pTmp1, i64 %tmp11		; <float*> [#uses=1]
-	%tmp13 = load float* %tmp12, align 4		; <float> [#uses=1]
+	%tmp12 = getelementptr float, float* %pTmp1, i64 %tmp11		; <float*> [#uses=1]
+	%tmp13 = load float, float* %tmp12, align 4		; <float> [#uses=1]
 	%tmp14 = fsub float %distERBhi.121, %tmp13		; <float> [#uses=1]
 	br label %bb3.preheader
 
@@ -57,12 +57,12 @@ bb2:		; preds = %bb3, %bb.nph
 	%loPart.02 = phi i32 [ %tmp24, %bb3 ], [ %loPart.118, %bb.nph ]		; <i32> [#uses=3]
 	%peakCount.01 = phi float [ %tmp23, %bb3 ], [ %peakCount.117, %bb.nph ]		; <float> [#uses=1]
 	%tmp16 = sext i32 %loPart.02 to i64		; <i64> [#uses=1]
-	%tmp17 = getelementptr float* %pTmp1, i64 %tmp16		; <float*> [#uses=1]
-	%tmp18 = load float* %tmp17, align 4		; <float> [#uses=1]
+	%tmp17 = getelementptr float, float* %pTmp1, i64 %tmp16		; <float*> [#uses=1]
+	%tmp18 = load float, float* %tmp17, align 4		; <float> [#uses=1]
 	%tmp19 = fsub float %distERBlo.03, %tmp18		; <float> [#uses=3]
 	%tmp20 = sext i32 %loPart.02 to i64		; <i64> [#uses=1]
-	%tmp21 = getelementptr float* %peakWeight, i64 %tmp20		; <float*> [#uses=1]
-	%tmp22 = load float* %tmp21, align 4		; <float> [#uses=1]
+	%tmp21 = getelementptr float, float* %peakWeight, i64 %tmp20		; <float*> [#uses=1]
+	%tmp22 = load float, float* %tmp21, align 4		; <float> [#uses=1]
 	%tmp23 = fsub float %peakCount.01, %tmp22		; <float> [#uses=2]
 	%tmp24 = add i32 %loPart.02, 1		; <i32> [#uses=2]
 	br label %bb3
@@ -97,13 +97,13 @@ bb4:		; preds = %bb5, %bb.nph12
 	%hiPart.08 = phi i32 [ %tmp31, %bb5 ], [ %hiPart.119, %bb.nph12 ]		; <i32> [#uses=2]
 	%peakCount.27 = phi float [ %tmp35, %bb5 ], [ %peakCount.0.lcssa, %bb.nph12 ]		; <float> [#uses=1]
 	%tmp27 = sext i32 %hiPart.08 to i64		; <i64> [#uses=1]
-	%tmp28 = getelementptr float* %pTmp1, i64 %tmp27		; <float*> [#uses=1]
-	%tmp29 = load float* %tmp28, align 4		; <float> [#uses=1]
+	%tmp28 = getelementptr float, float* %pTmp1, i64 %tmp27		; <float*> [#uses=1]
+	%tmp29 = load float, float* %tmp28, align 4		; <float> [#uses=1]
 	%tmp30 = fadd float %tmp29, %distERBhi.29		; <float> [#uses=3]
 	%tmp31 = add i32 %hiPart.08, 1		; <i32> [#uses=4]
 	%tmp32 = sext i32 %tmp31 to i64		; <i64> [#uses=1]
-	%tmp33 = getelementptr float* %peakWeight, i64 %tmp32		; <float*> [#uses=1]
-	%tmp34 = load float* %tmp33, align 4		; <float> [#uses=1]
+	%tmp33 = getelementptr float, float* %peakWeight, i64 %tmp32		; <float*> [#uses=1]
+	%tmp34 = load float, float* %tmp33, align 4		; <float> [#uses=1]
 	%tmp35 = fadd float %tmp34, %peakCount.27		; <float> [#uses=2]
 	br label %bb5
 
@@ -132,7 +132,7 @@ bb7:		; preds = %bb5.bb7_crit_edge, %bb5.preheader
 	%tmp42 = fadd float %tmp41, 1.000000e+00		; <float> [#uses=1]
 	%tmp43 = fdiv float 1.000000e+00, %tmp42		; <float> [#uses=1]
 	%tmp44 = sext i32 %part.016 to i64		; <i64> [#uses=1]
-	%tmp45 = getelementptr float* %nrgReducePeakrate, i64 %tmp44		; <float*> [#uses=1]
+	%tmp45 = getelementptr float, float* %nrgReducePeakrate, i64 %tmp44		; <float*> [#uses=1]
 	store float %tmp43, float* %tmp45, align 4
 	%tmp46 = add i32 %part.016, 1		; <i32> [#uses=2]
 	br label %bb8
diff --git a/test/Transforms/IndVarSimplify/iv-widen.ll b/test/Transforms/IndVarSimplify/iv-widen.ll
index c899e2f0a580..464b03ce5595 100644
--- a/test/Transforms/IndVarSimplify/iv-widen.ll
+++ b/test/Transforms/IndVarSimplify/iv-widen.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -indvars -S | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 target triple = "x86_64-apple-darwin"
 
 ; CHECK-LABEL: @sloop
@@ -20,8 +24,8 @@ B18:                                        ; preds = %B24, %Prologue
   %.02 = phi i32 [ 0, %Prologue ], [ %tmp33, %B24 ]
   %tmp23 = zext i32 %.02 to i64
   %tmp33 = add i32 %.02, 1
-  %o = getelementptr i32* %a, i32 %.02
-  %v = load i32* %o
+  %o = getelementptr i32, i32* %a, i32 %.02
+  %v = load i32, i32* %o
   %t = icmp eq i32 %v, 0
   br i1 %t, label %exit24, label %B24
 
diff --git a/test/Transforms/IndVarSimplify/iv-zext.ll b/test/Transforms/IndVarSimplify/iv-zext.ll
index ed0514b08e33..629a85efe8e1 100644
--- a/test/Transforms/IndVarSimplify/iv-zext.ll
+++ b/test/Transforms/IndVarSimplify/iv-zext.ll
@@ -11,17 +11,17 @@ entry:
 loop:
 	%indvar = phi i64 [ 0, %entry ], [ %indvar.next, %loop ]
 	%indvar.i8 = and i64 %indvar, 255
-	%t0 = getelementptr double* %d, i64 %indvar.i8
-	%t1 = load double* %t0
+	%t0 = getelementptr double, double* %d, i64 %indvar.i8
+	%t1 = load double, double* %t0
 	%t2 = fmul double %t1, 0.1
 	store double %t2, double* %t0
 	%indvar.i24 = and i64 %indvar, 16777215
-	%t3 = getelementptr double* %d, i64 %indvar.i24
-	%t4 = load double* %t3
+	%t3 = getelementptr double, double* %d, i64 %indvar.i24
+	%t4 = load double, double* %t3
 	%t5 = fmul double %t4, 2.3
 	store double %t5, double* %t3
-	%t6 = getelementptr double* %d, i64 %indvar
-	%t7 = load double* %t6
+	%t6 = getelementptr double, double* %d, i64 %indvar
+	%t7 = load double, double* %t6
 	%t8 = fmul double %t7, 4.5
 	store double %t8, double* %t6
 	%indvar.next = add i64 %indvar, 1
diff --git a/test/Transforms/IndVarSimplify/lftr-address-space-pointers.ll b/test/Transforms/IndVarSimplify/lftr-address-space-pointers.ll
index 9e55a1791341..feb4b35a20a7 100644
--- a/test/Transforms/IndVarSimplify/lftr-address-space-pointers.ll
+++ b/test/Transforms/IndVarSimplify/lftr-address-space-pointers.ll
@@ -6,12 +6,12 @@ define void @ptriv_as2(i8 addrspace(2)* %base, i32 %n) nounwind {
 ; CHECK-LABEL: @ptriv_as2(
 entry:
   %idx.trunc = trunc i32 %n to i8
-  %add.ptr = getelementptr inbounds i8 addrspace(2)* %base, i8 %idx.trunc
+  %add.ptr = getelementptr inbounds i8, i8 addrspace(2)* %base, i8 %idx.trunc
   %cmp1 = icmp ult i8 addrspace(2)* %base, %add.ptr
   br i1 %cmp1, label %for.body, label %for.end
 
 ; Make sure the added GEP has the right index type
-; CHECK: %lftr.limit = getelementptr i8 addrspace(2)* %base, i8
+; CHECK: %lftr.limit = getelementptr i8, i8 addrspace(2)* %base, i8 %0
 
 ; CHECK: for.body:
 ; CHECK: phi i8 addrspace(2)*
@@ -26,7 +26,7 @@ for.body:
   %sub.ptr.rhs.cast = ptrtoint i8 addrspace(2)* %base to i8
   %sub.ptr.sub = sub i8 %sub.ptr.lhs.cast, %sub.ptr.rhs.cast
   store i8 %sub.ptr.sub, i8 addrspace(2)* %p.02
-  %incdec.ptr = getelementptr inbounds i8 addrspace(2)* %p.02, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8 addrspace(2)* %p.02, i32 1
   %cmp = icmp ult i8 addrspace(2)* %incdec.ptr, %add.ptr
   br i1 %cmp, label %for.body, label %for.end
 
@@ -38,12 +38,12 @@ define void @ptriv_as3(i8 addrspace(3)* %base, i32 %n) nounwind {
 ; CHECK-LABEL: @ptriv_as3(
 entry:
   %idx.trunc = trunc i32 %n to i16
-  %add.ptr = getelementptr inbounds i8 addrspace(3)* %base, i16 %idx.trunc
+  %add.ptr = getelementptr inbounds i8, i8 addrspace(3)* %base, i16 %idx.trunc
   %cmp1 = icmp ult i8 addrspace(3)* %base, %add.ptr
   br i1 %cmp1, label %for.body, label %for.end
 
 ; Make sure the added GEP has the right index type
-; CHECK: %lftr.limit = getelementptr i8 addrspace(3)* %base, i16
+; CHECK: %lftr.limit = getelementptr i8, i8 addrspace(3)* %base, i16 %0
 
 ; CHECK: for.body:
 ; CHECK: phi i8 addrspace(3)*
@@ -59,7 +59,7 @@ for.body:
   %sub.ptr.sub = sub i16 %sub.ptr.lhs.cast, %sub.ptr.rhs.cast
   %conv = trunc i16 %sub.ptr.sub to i8
   store i8 %conv, i8 addrspace(3)* %p.02
-  %incdec.ptr = getelementptr inbounds i8 addrspace(3)* %p.02, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8 addrspace(3)* %p.02, i32 1
   %cmp = icmp ult i8 addrspace(3)* %incdec.ptr, %add.ptr
   br i1 %cmp, label %for.body, label %for.end
 
diff --git a/test/Transforms/IndVarSimplify/lftr-extend-const.ll b/test/Transforms/IndVarSimplify/lftr-extend-const.ll
index f12c68cacb66..fa3166d4a3e2 100644
--- a/test/Transforms/IndVarSimplify/lftr-extend-const.ll
+++ b/test/Transforms/IndVarSimplify/lftr-extend-const.ll
@@ -1,5 +1,9 @@
 ;RUN: opt -S %s -indvars | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 ; CHECK-LABEL: @foo(
 ; CHECK-NOT: %lftr.wideiv = trunc i32 %indvars.iv.next to i16
 ; CHECK: %exitcond = icmp ne i32 %indvars.iv.next, 512
diff --git a/test/Transforms/IndVarSimplify/lftr-promote.ll b/test/Transforms/IndVarSimplify/lftr-promote.ll
index c4ecc845562b..4fe3191391b0 100644
--- a/test/Transforms/IndVarSimplify/lftr-promote.ll
+++ b/test/Transforms/IndVarSimplify/lftr-promote.ll
@@ -17,11 +17,11 @@ bb.nph:		; preds = %entry
 bb2:		; preds = %bb3, %bb.nph
 	%i.01 = phi i32 [ %7, %bb3 ], [ 0, %bb.nph ]		; <i32> [#uses=3]
 	%1 = sext i32 %i.01 to i64		; <i64> [#uses=1]
-	%2 = getelementptr double* %p, i64 %1		; <double*> [#uses=1]
-	%3 = load double* %2, align 8		; <double> [#uses=1]
+	%2 = getelementptr double, double* %p, i64 %1		; <double*> [#uses=1]
+	%3 = load double, double* %2, align 8		; <double> [#uses=1]
 	%4 = fmul double %3, 1.100000e+00		; <double> [#uses=1]
 	%5 = sext i32 %i.01 to i64		; <i64> [#uses=1]
-	%6 = getelementptr double* %p, i64 %5		; <double*> [#uses=1]
+	%6 = getelementptr double, double* %p, i64 %5		; <double*> [#uses=1]
 	store double %4, double* %6, align 8
 	%7 = add i32 %i.01, 1		; <i32> [#uses=2]
 	br label %bb3
diff --git a/test/Transforms/IndVarSimplify/lftr-reuse.ll b/test/Transforms/IndVarSimplify/lftr-reuse.ll
index efb96bdbe6b5..b4e1fdef4885 100644
--- a/test/Transforms/IndVarSimplify/lftr-reuse.ll
+++ b/test/Transforms/IndVarSimplify/lftr-reuse.ll
@@ -12,7 +12,7 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define void @ptriv(i8* %base, i32 %n) nounwind {
 entry:
   %idx.ext = sext i32 %n to i64
-  %add.ptr = getelementptr inbounds i8* %base, i64 %idx.ext
+  %add.ptr = getelementptr inbounds i8, i8* %base, i64 %idx.ext
   %cmp1 = icmp ult i8* %base, %add.ptr
   br i1 %cmp1, label %for.body, label %for.end
 
@@ -30,7 +30,7 @@ for.body:
   %sub.ptr.sub = sub i64 %sub.ptr.lhs.cast, %sub.ptr.rhs.cast
   %conv = trunc i64 %sub.ptr.sub to i8
   store i8 %conv, i8* %p.02
-  %incdec.ptr = getelementptr inbounds i8* %p.02, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %p.02, i32 1
   %cmp = icmp ult i8* %incdec.ptr, %add.ptr
   br i1 %cmp, label %for.body, label %for.end
 
@@ -82,23 +82,15 @@ exit:
 ; Perform LFTR without generating extra preheader code.
 define void @guardedloop([0 x double]* %matrix, [0 x double]* %vector,
                          i32 %irow, i32 %ilead) nounwind {
-; CHECK-LABEL: @guardedloop(
-; CHECK-LABEL: entry:
-; CHECK-NEXT: %[[cmp:.*]] = icmp slt i32 1, %irow
-; CHECK-NEXT: br i1 %[[cmp]], label %[[loop_preheader:.*]], label %[[return:.*]]
-
-; CHECK: [[loop_preheader]]:
-; CHECK-NEXT: %[[sext:.*]] = sext i32 %ilead to i64
-; CHECK-NEXT: %[[add:.*]] = add i32 %irow, -1
-; CHECK-NEXT: br label %[[loop:.*]]
-
-; CHECK: [[loop]]:
-; CHECK-NEXT: %[[indvars_iv2:.*]] = phi i64
-; CHECK-NEXT: phi i64
+; CHECK: entry:
+; CHECK-NOT: zext
+; CHECK-NOT: add
+; CHECK: loop:
+; CHECK: phi i64
+; CHECK: phi i64
 ; CHECK-NOT: phi
-; CHECK: %[[lftr_wideiv:.*]] = trunc i64 %[[indvars_iv2]] to i32
-; CHECK-NEXT: %[[exitcond:.*]] = icmp ne i32 %[[lftr_wideiv]], %[[add]]
-; CHECK-NEXT: br i1 %[[exitcond]], label %[[loop]], label
+; CHECK: icmp ne
+; CHECK: br i1
 entry:
   %cmp = icmp slt i32 1, %irow
   br i1 %cmp, label %loop, label %return
@@ -108,11 +100,11 @@ loop:
   %i = phi i32 [ 0, %entry ], [ %i.inc, %loop ]
   %diagidx = add nsw i32 %rowidx, %i
   %diagidxw = sext i32 %diagidx to i64
-  %matrixp = getelementptr inbounds [0 x double]* %matrix, i32 0, i64 %diagidxw
-  %v1 = load double* %matrixp
+  %matrixp = getelementptr inbounds [0 x double], [0 x double]* %matrix, i32 0, i64 %diagidxw
+  %v1 = load double, double* %matrixp
   %iw = sext i32 %i to i64
-  %vectorp = getelementptr inbounds [0 x double]* %vector, i32 0, i64 %iw
-  %v2 = load double* %vectorp
+  %vectorp = getelementptr inbounds [0 x double], [0 x double]* %vector, i32 0, i64 %iw
+  %v2 = load double, double* %vectorp
   %row.inc = add nsw i32 %rowidx, %ilead
   %i.inc = add nsw i32 %i, 1
   %cmp196 = icmp slt i32 %i.inc, %irow
@@ -142,11 +134,11 @@ loop:
   %i = phi i32 [ 0, %entry ], [ %i.inc, %loop ]
   %diagidx = add nsw i32 %rowidx, %i
   %diagidxw = sext i32 %diagidx to i64
-  %matrixp = getelementptr inbounds [0 x double]* %matrix, i32 0, i64 %diagidxw
-  %v1 = load double* %matrixp
+  %matrixp = getelementptr inbounds [0 x double], [0 x double]* %matrix, i32 0, i64 %diagidxw
+  %v1 = load double, double* %matrixp
   %iw = sext i32 %i to i64
-  %vectorp = getelementptr inbounds [0 x double]* %vector, i32 0, i64 %iw
-  %v2 = load double* %vectorp
+  %vectorp = getelementptr inbounds [0 x double], [0 x double]* %vector, i32 0, i64 %iw
+  %v2 = load double, double* %vectorp
   %row.inc = add nsw i32 %rowidx, %ilead
   %i.inc = add nsw i32 %i, 1
   %cmp196 = icmp slt i32 %i.inc, %irow
@@ -165,9 +157,9 @@ return:
 define void @geplftr(i8* %base, i32 %x, i32 %y, i32 %n) nounwind {
 entry:
   %x.ext = sext i32 %x to i64
-  %add.ptr = getelementptr inbounds i8* %base, i64 %x.ext
+  %add.ptr = getelementptr inbounds i8, i8* %base, i64 %x.ext
   %y.ext = sext i32 %y to i64
-  %add.ptr10 = getelementptr inbounds i8* %add.ptr, i64 %y.ext
+  %add.ptr10 = getelementptr inbounds i8, i8* %add.ptr, i64 %y.ext
   %lim = add i32 %x, %n
   %cmp.ph = icmp ult i32 %x, %lim
   br i1 %cmp.ph, label %loop, label %exit
@@ -182,7 +174,7 @@ entry:
 loop:
   %i = phi i32 [ %x, %entry ], [ %inc, %loop ]
   %aptr = phi i8* [ %add.ptr10, %entry ], [ %incdec.ptr, %loop ]
-  %incdec.ptr = getelementptr inbounds i8* %aptr, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %aptr, i32 1
   store i8 3, i8* %aptr
   %inc = add i32 %i, 1
   %cmp = icmp ult i32 %inc, %lim
@@ -215,8 +207,8 @@ exit:
 ; Test LFTR on an IV whose recurrence start is a non-unit pointer type.
 define void @aryptriv([256 x i8]* %base, i32 %n) nounwind {
 entry:
-  %ivstart = getelementptr inbounds [256 x i8]* %base, i32 0, i32 0
-  %ivend = getelementptr inbounds [256 x i8]* %base, i32 0, i32 %n
+  %ivstart = getelementptr inbounds [256 x i8], [256 x i8]* %base, i32 0, i32 0
+  %ivend = getelementptr inbounds [256 x i8], [256 x i8]* %base, i32 0, i32 %n
   %cmp.ph = icmp ult i8* %ivstart, %ivend
   br i1 %cmp.ph, label %loop, label %exit
 
@@ -229,7 +221,7 @@ entry:
 ; CHECK: br i1
 loop:
   %aptr = phi i8* [ %ivstart, %entry ], [ %incdec.ptr, %loop ]
-  %incdec.ptr = getelementptr inbounds i8* %aptr, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %aptr, i32 1
   store i8 3, i8* %aptr
   %cmp = icmp ult i8* %incdec.ptr, %ivend
   br i1 %cmp, label %loop, label %exit
diff --git a/test/Transforms/IndVarSimplify/lftr-udiv-tripcount.ll b/test/Transforms/IndVarSimplify/lftr-udiv-tripcount.ll
new file mode 100644
index 000000000000..8a1bb07bb0c2
--- /dev/null
+++ b/test/Transforms/IndVarSimplify/lftr-udiv-tripcount.ll
@@ -0,0 +1,29 @@
+; RUN: opt < %s -indvars -S | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
+; It is okay to do LFTR on this loop even though the trip count is a
+; division because in this case the division can be optimized to a
+; shift.
+
+define void @foo(i8* %a, i8 %n) nounwind uwtable ssp {
+; CHECK-LABEL: @foo(
+ entry:
+  %e = icmp sgt i8 %n, 3
+  br i1 %e, label %loop, label %exit
+
+ loop:
+; CHECK-LABEL: loop:
+  %i = phi i8 [ 0, %entry ], [ %i.inc, %loop ]
+  %i1 = phi i8 [ 0, %entry ], [ %i1.inc, %loop ]
+  %i.inc = add nsw i8 %i, 4
+  %i1.inc = add i8 %i1, 1
+  store volatile i8 0, i8* %a
+  %c = icmp slt i8 %i, %n
+; CHECK-LABEL:  %exitcond = icmp ne i8 %i1.inc
+  br i1 %c, label %loop, label %exit
+
+ exit:
+; CHECK-LABEL: exit:
+  ret void
+}
diff --git a/test/Transforms/IndVarSimplify/lftr-zext.ll b/test/Transforms/IndVarSimplify/lftr-zext.ll
index 32fa61af1801..e654e143cee3 100644
--- a/test/Transforms/IndVarSimplify/lftr-zext.ll
+++ b/test/Transforms/IndVarSimplify/lftr-zext.ll
@@ -11,11 +11,11 @@ define void @foo(i8* %a) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %0, %1
   %i.0 = phi i8 [ 0, %0 ], [ %5, %1 ]
-  %p.0 = phi i8* [ getelementptr inbounds ([240 x i8]* @data, i64 0, i64 0), %0 ], [ %4, %1 ]
+  %p.0 = phi i8* [ getelementptr inbounds ([240 x i8], [240 x i8]* @data, i64 0, i64 0), %0 ], [ %4, %1 ]
   %.0 = phi i8* [ %a, %0 ], [ %2, %1 ]
-  %2 = getelementptr inbounds i8* %.0, i64 1
-  %3 = load i8* %.0, align 1
-  %4 = getelementptr inbounds i8* %p.0, i64 1
+  %2 = getelementptr inbounds i8, i8* %.0, i64 1
+  %3 = load i8, i8* %.0, align 1
+  %4 = getelementptr inbounds i8, i8* %p.0, i64 1
   store i8 %3, i8* %p.0, align 1
   %5 = add i8 %i.0, 1
   %6 = icmp ult i8 %5, -16
diff --git a/test/Transforms/IndVarSimplify/lftr_simple.ll b/test/Transforms/IndVarSimplify/lftr_simple.ll
index e373013609be..6a8d93721835 100644
--- a/test/Transforms/IndVarSimplify/lftr_simple.ll
+++ b/test/Transforms/IndVarSimplify/lftr_simple.ll
@@ -1,7 +1,11 @@
-; LFTR should eliminate the need for the computation of i*i completely.  It 
+; LFTR should eliminate the need for the computation of i*i completely.  It
 ; is only used to compute the exit value.
 ; RUN: opt < %s -indvars -dce -S | not grep mul
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 @A = external global i32                ; <i32*> [#uses=1]
 
 define i32 @quadratic_setlt() {
diff --git a/test/Transforms/IndVarSimplify/loop_evaluate7.ll b/test/Transforms/IndVarSimplify/loop_evaluate7.ll
index b9c0b12f2d97..333ab7a4967b 100644
--- a/test/Transforms/IndVarSimplify/loop_evaluate7.ll
+++ b/test/Transforms/IndVarSimplify/loop_evaluate7.ll
@@ -21,7 +21,7 @@ bb19:		; preds = %bb30
 	br i1 undef, label %bb20, label %bb29
 
 bb20:		; preds = %bb19
-	%0 = load i32* undef, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* undef, align 4		; <i32> [#uses=1]
 	%1 = sub i32 %0, %n		; <i32> [#uses=1]
 	br label %bb23
 
@@ -30,7 +30,7 @@ bb21:		; preds = %bb23
 	br i1 %2, label %bb22, label %overflow2
 
 bb22:		; preds = %bb21
-	%3 = getelementptr i8* %q.0, i32 1		; <i8*> [#uses=1]
+	%3 = getelementptr i8, i8* %q.0, i32 1		; <i8*> [#uses=1]
 	br label %bb23
 
 bb23:		; preds = %bb22, %bb20
diff --git a/test/Transforms/IndVarSimplify/loop_evaluate8.ll b/test/Transforms/IndVarSimplify/loop_evaluate8.ll
index 2a9d20596233..28d05dff2ff8 100644
--- a/test/Transforms/IndVarSimplify/loop_evaluate8.ll
+++ b/test/Transforms/IndVarSimplify/loop_evaluate8.ll
@@ -23,7 +23,7 @@ bb19:		; preds = %bb30
 	br i1 undef, label %bb20, label %bb29
 
 bb20:		; preds = %bb19
-	%0 = load i32* undef, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* undef, align 4		; <i32> [#uses=1]
 	%1 = sub i32 %0, undef		; <i32> [#uses=1]
 	br label %bb23
 
@@ -32,7 +32,7 @@ bb21:		; preds = %bb23
 	br i1 %2, label %bb22, label %overflow2
 
 bb22:		; preds = %bb21
-	%3 = getelementptr i8* %q.0, i32 1		; <i8*> [#uses=1]
+	%3 = getelementptr i8, i8* %q.0, i32 1		; <i8*> [#uses=1]
 	br label %bb23
 
 bb23:		; preds = %bb22, %bb20
diff --git a/test/Transforms/IndVarSimplify/loop_evaluate9.ll b/test/Transforms/IndVarSimplify/loop_evaluate9.ll
index 21fb7ef93167..7a69bd2e131f 100644
--- a/test/Transforms/IndVarSimplify/loop_evaluate9.ll
+++ b/test/Transforms/IndVarSimplify/loop_evaluate9.ll
@@ -28,7 +28,7 @@ bb1.i:		; preds = %bb2.i, %entry
 	br i1 %tmp1, label %bb.i.i, label %bb1.i.i
 
 bb.i.i:		; preds = %bb1.i
-	tail call void @__gnat_rcheck_12(i8* getelementptr ([13 x i8]* @.str, i32 0, i32 0), i32 24) noreturn
+	tail call void @__gnat_rcheck_12(i8* getelementptr ([13 x i8], [13 x i8]* @.str, i32 0, i32 0), i32 24) noreturn
 	unreachable
 
 bb1.i.i:		; preds = %bb1.i
@@ -39,7 +39,7 @@ bb1.i.i:		; preds = %bb1.i
 	br i1 %tmp3, label %bb2.i.i, label %cc70a02__complex_integers__Oadd.153.exit.i
 
 bb2.i.i:		; preds = %bb1.i.i
-	tail call void @__gnat_rcheck_12(i8* getelementptr ([13 x i8]* @.str, i32 0, i32 0), i32 24) noreturn
+	tail call void @__gnat_rcheck_12(i8* getelementptr ([13 x i8], [13 x i8]* @.str, i32 0, i32 0), i32 24) noreturn
 	unreachable
 
 cc70a02__complex_integers__Oadd.153.exit.i:		; preds = %bb1.i.i
diff --git a/test/Transforms/IndVarSimplify/masked-iv.ll b/test/Transforms/IndVarSimplify/masked-iv.ll
index f1f5af96eb96..24fe33d4369f 100644
--- a/test/Transforms/IndVarSimplify/masked-iv.ll
+++ b/test/Transforms/IndVarSimplify/masked-iv.ll
@@ -16,7 +16,7 @@ bb.preheader:		; preds = %entry
 bb:		; preds = %bb, %bb.preheader
 	%i.01 = phi i64 [ %t6, %bb ], [ %a, %bb.preheader ]		; <i64> [#uses=3]
 	%t1 = and i64 %i.01, 255		; <i64> [#uses=1]
-	%t2 = getelementptr i64* %A, i64 %t1		; <i64*> [#uses=1]
+	%t2 = getelementptr i64, i64* %A, i64 %t1		; <i64*> [#uses=1]
 	store i64 %i.01, i64* %t2, align 8
 	%t6 = add i64 %i.01, %s		; <i64> [#uses=1]
 	br label %bb
diff --git a/test/Transforms/IndVarSimplify/no-iv-rewrite.ll b/test/Transforms/IndVarSimplify/no-iv-rewrite.ll
index a7023f24314a..ca7b89911662 100644
--- a/test/Transforms/IndVarSimplify/no-iv-rewrite.ll
+++ b/test/Transforms/IndVarSimplify/no-iv-rewrite.ll
@@ -28,8 +28,8 @@ loop:
   %i.02 = phi i32 [ 0, %ph ], [ %iinc, %loop ]
   %s.01 = phi i32 [ 0, %ph ], [ %sinc, %loop ]
   %ofs = sext i32 %i.02 to i64
-  %adr = getelementptr inbounds i32* %arr, i64 %ofs
-  %val = load i32* %adr
+  %adr = getelementptr inbounds i32, i32* %arr, i64 %ofs
+  %val = load i32, i32* %adr
   %sinc = add nsw i32 %s.01, %val
   %iinc = add nsw i32 %i.02, 1
   %cond = icmp slt i32 %iinc, %n
@@ -69,8 +69,8 @@ loop:
   %i.02 = phi i32 [ 0, %ph ], [ %iinc, %loop ]
   %s.01 = phi i64 [ 0, %ph ], [ %sinc, %loop ]
   %ofs = sext i32 %i.02 to i64
-  %adr = getelementptr inbounds i32* %arr, i64 %ofs
-  %val = load i32* %adr
+  %adr = getelementptr inbounds i32, i32* %arr, i64 %ofs
+  %val = load i32, i32* %adr
   %vall = sext i32 %val to i64
   %sinc = add nsw i64 %s.01, %vall
   %iinc = add nsw i32 %i.02, 1
@@ -106,15 +106,15 @@ ph:
 ; CHECK-NOT: add
 ;
 ; Preserve gep inboundsness, and don't factor it.
-; CHECK: getelementptr inbounds i32* %ptriv, i32 1
+; CHECK: getelementptr inbounds i32, i32* %ptriv, i32 1
 ; CHECK-NOT: add
 ; CHECK: exit:
 loop:
   %ptriv = phi i32* [ %first, %ph ], [ %ptrpost, %loop ]
   %ofs = sext i32 %idx to i64
-  %adr = getelementptr inbounds i32* %ptriv, i64 %ofs
+  %adr = getelementptr inbounds i32, i32* %ptriv, i64 %ofs
   store i32 3, i32* %adr
-  %ptrpost = getelementptr inbounds i32* %ptriv, i32 1
+  %ptrpost = getelementptr inbounds i32, i32* %ptriv, i32 1
   %cond = icmp ne i32* %ptrpost, %last
   br i1 %cond, label %loop, label %exit
 
@@ -143,11 +143,11 @@ entry:
 loop:
   %iv = phi i32 [%start, %entry], [%next, %loop]
   %p = phi %structI* [%base, %entry], [%pinc, %loop]
-  %adr = getelementptr %structI* %p, i32 0, i32 0
+  %adr = getelementptr %structI, %structI* %p, i32 0, i32 0
   store i32 3, i32* %adr
   %pp = bitcast %structI* %p to i32*
   store i32 4, i32* %pp
-  %pinc = getelementptr %structI* %p, i32 1
+  %pinc = getelementptr %structI, %structI* %p, i32 1
   %next = add i32 %iv, 1
   %cond = icmp ne i32 %next, %limit
   br i1 %cond, label %loop, label %exit
@@ -170,8 +170,8 @@ loop:
   %idx = phi i32 [ 0, %entry ], [ %idx.next, %loop.inc ]
   %max = phi i32 [ 0, %entry ], [ %max.next, %loop.inc ]
   %idxprom = sext i32 %idx to i64
-  %adr = getelementptr inbounds i32* %base, i64 %idxprom
-  %val = load i32* %adr
+  %adr = getelementptr inbounds i32, i32* %base, i64 %idxprom
+  %val = load i32, i32* %adr
   %cmp19 = icmp sgt i32 %val, %max
   br i1 %cmp19, label %if.then, label %if.else
 
@@ -239,8 +239,8 @@ entry:
 loop:
   %iv = phi i32 [ 0, %entry], [ %iv.next, %loop ]
   %t1 = sext i32 %iv to i64
-  %adr = getelementptr i64* %base, i64 %t1
-  %val = load i64* %adr
+  %adr = getelementptr i64, i64* %base, i64 %t1
+  %val = load i64, i64* %adr
   %t2 = or i32 %iv, 1
   %t3 = sext i32 %t2 to i64
   %iv.next = add i32 %iv, 2
@@ -333,7 +333,7 @@ return:
 
 define void @congruentgepiv(%structIF* %base) nounwind uwtable ssp {
 entry:
-  %first = getelementptr inbounds %structIF* %base, i64 0, i32 0
+  %first = getelementptr inbounds %structIF, %structIF* %base, i64 0, i32 0
   br label %loop
 
 ; CHECK: loop:
@@ -349,8 +349,8 @@ loop:
   br i1 undef, label %latch, label %exit
 
 latch:                         ; preds = %for.inc50.i
-  %ptr.inc = getelementptr inbounds %structIF* %ptr.iv, i64 1
-  %next.inc = getelementptr inbounds %structIF* %ptr.inc, i64 0, i32 0
+  %ptr.inc = getelementptr inbounds %structIF, %structIF* %ptr.iv, i64 1
+  %next.inc = getelementptr inbounds %structIF, %structIF* %ptr.inc, i64 0, i32 0
   br label %loop
 
 exit:
diff --git a/test/Transforms/IndVarSimplify/overflowcheck.ll b/test/Transforms/IndVarSimplify/overflowcheck.ll
index 2603f363ab60..c3c033dfaece 100644
--- a/test/Transforms/IndVarSimplify/overflowcheck.ll
+++ b/test/Transforms/IndVarSimplify/overflowcheck.ll
@@ -9,7 +9,7 @@ target triple = "x86_64-apple-macosx"
 ; CHECK: @llvm.sadd.with.overflow
 ; CHECK-LABEL: loop2:
 ; CHECK-NOT: extractvalue
-; CHECK: add nuw nsw
+; CHECK: add nuw
 ; CHECK: @llvm.sadd.with.overflow
 ; CHECK-LABEL: loop3:
 ; CHECK-NOT: extractvalue
@@ -27,8 +27,8 @@ loop0:
 loop1:
   %zxt = zext i32 %i to i64
   %ofs = shl nuw nsw i64 %zxt, 3
-  %gep = getelementptr i64* %a, i64 %zxt
-  %v = load i64* %gep, align 8
+  %gep = getelementptr i64, i64* %a, i64 %zxt
+  %v = load i64, i64* %gep, align 8
   %truncv = trunc i64 %v to i32
   %adds = call { i32, i1 } @llvm.sadd.with.overflow.i32(i32 %s, i32 %truncv)
   %ovflows = extractvalue { i32, i1 } %adds, 1
diff --git a/test/Transforms/IndVarSimplify/phi-uses-value-multiple-times.ll b/test/Transforms/IndVarSimplify/phi-uses-value-multiple-times.ll
index dc36b9948254..519d34d76cbd 100644
--- a/test/Transforms/IndVarSimplify/phi-uses-value-multiple-times.ll
+++ b/test/Transforms/IndVarSimplify/phi-uses-value-multiple-times.ll
@@ -13,7 +13,7 @@ entry:
 	br label %bb38.i
 
 bb14.i27:
-	%t0 = load i64* @ue, align 8
+	%t0 = load i64, i64* @ue, align 8
 	%t1 = sub i64 %t0, %i.0.i35
 	%t2 = add i64 %t1, 1
 	br i1 undef, label %bb15.i28, label %bb19.i31
diff --git a/test/Transforms/IndVarSimplify/polynomial-expand.ll b/test/Transforms/IndVarSimplify/polynomial-expand.ll
index 2087f6a6966b..5708c64d6c8e 100644
--- a/test/Transforms/IndVarSimplify/polynomial-expand.ll
+++ b/test/Transforms/IndVarSimplify/polynomial-expand.ll
@@ -20,8 +20,8 @@ bb30:                                             ; preds = %bb30, %bb24
   %tmp31 = phi i32 [ %tmp39, %bb30 ], [ %tmp28, %bb24 ] ; <i32> [#uses=2]
   %tmp32 = phi i32 [ %tmp37, %bb30 ], [ %tmp27, %bb24 ] ; <i32> [#uses=2]
   %tmp33 = sext i32 %tmp32 to i64                 ; <i64> [#uses=1]
-  %tmp35 = getelementptr float* %tmp4, i64 %tmp33 ; <%0*> [#uses=1]
-  %tmp36 = load float* %tmp35, align 4               ; <%0> [#uses=0]
+  %tmp35 = getelementptr float, float* %tmp4, i64 %tmp33 ; <%0*> [#uses=1]
+  %tmp36 = load float, float* %tmp35, align 4               ; <%0> [#uses=0]
   %tmp37 = add nsw i32 %tmp32, -1                 ; <i32> [#uses=1]
   %tmp39 = add nsw i32 %tmp31, -1                 ; <i32> [#uses=1]
   %tmp38 = icmp eq i32 %tmp31, 1                  ; <i1> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/pr18223.ll b/test/Transforms/IndVarSimplify/pr18223.ll
index 738f75c0fe03..f922aa424a17 100644
--- a/test/Transforms/IndVarSimplify/pr18223.ll
+++ b/test/Transforms/IndVarSimplify/pr18223.ll
@@ -8,7 +8,7 @@
 
 define i32 @main() #0 {
 entry:
-  %0 = load i32* @c, align 4
+  %0 = load i32, i32* @c, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.body, label %exit
 
diff --git a/test/Transforms/IndVarSimplify/pr20680.ll b/test/Transforms/IndVarSimplify/pr20680.ll
index 88a7fd765d0b..2c9eb54a789c 100644
--- a/test/Transforms/IndVarSimplify/pr20680.ll
+++ b/test/Transforms/IndVarSimplify/pr20680.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -indvars -S | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 @a = common global i32 0, align 4
 @c = common global i32 0, align 4
 @b = common global i32 0, align 4
@@ -13,9 +17,9 @@ define void @f() {
 ; CHECK-NEXT: %[[indvars_iv:.*]] = phi i32 [ %[[indvars_iv_next:.*]], %[[for_inc13:.*]] ], [ -14, %entry ]
 ; br i1 {{.*}}, label %[[for_inc13]], label %
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool2 = icmp eq i32 %0, 0
-  %1 = load i32* @a, align 4
+  %1 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %1, 0
   br label %for.cond2.preheader
 
@@ -51,7 +55,7 @@ cond.false.us.us:                                 ; preds = %for.body3.us.us
 
 cond.end.us.us:                                   ; preds = %cond.false.us.us, %for.body3.us.us
   %cond.us.us = phi i32 [ %div, %cond.false.us.us ], [ %conv7, %for.body3.us.us ]
-  %4 = load i32* @b, align 4
+  %4 = load i32, i32* @b, align 4
   %cmp91.us.us = icmp slt i32 %4, 1
   br i1 %cmp91.us.us, label %for.inc.lr.ph.us.us, label %for.cond2.loopexit.us.us
 
@@ -87,7 +91,7 @@ cond.false.us:                                    ; preds = %for.body3.us
 
 cond.end.us:                                      ; preds = %cond.false.us, %for.body3.us
   %cond.us = phi i32 [ %div, %cond.false.us ], [ %conv7, %for.body3.us ]
-  %6 = load i32* @b, align 4
+  %6 = load i32, i32* @b, align 4
   %cmp91.us = icmp slt i32 %6, 1
   br i1 %cmp91.us, label %for.inc.lr.ph.us, label %for.cond2.loopexit.us
 
@@ -133,7 +137,7 @@ cond.false.us4:                                   ; preds = %for.body3.us3
 
 cond.end.us5:                                     ; preds = %cond.false.us4, %for.body3.us3
   %cond.us6 = phi i32 [ %div, %cond.false.us4 ], [ %conv7, %for.body3.us3 ]
-  %8 = load i32* @b, align 4
+  %8 = load i32, i32* @b, align 4
   %cmp91.us7 = icmp slt i32 %8, 1
   br i1 %cmp91.us7, label %for.inc.lr.ph.us12, label %for.cond2.loopexit.us11
 
@@ -177,7 +181,7 @@ cond.false:                                       ; preds = %for.body3
 
 cond.end:                                         ; preds = %cond.false, %for.body3
   %cond = phi i32 [ %div, %cond.false ], [ %conv7, %for.body3 ]
-  %10 = load i32* @b, align 4
+  %10 = load i32, i32* @b, align 4
   %cmp91 = icmp slt i32 %10, 1
   br i1 %cmp91, label %for.inc.lr.ph, label %for.cond2.loopexit
 
@@ -204,8 +208,8 @@ for.cond2.for.inc13_crit_edge:                    ; preds = %for.cond2.for.inc13
   br label %for.inc13
 
 ; CHECK: [[for_inc13]]:
-; CHECK-NEXT: %[[indvars_iv_next]] = add nuw nsw i32 %[[indvars_iv]], 1
-; CHECK-NEXT: %[[exitcond4:.*]] = icmp ne i32 %[[indvars_iv]], -1
+; CHECK-NEXT: %[[indvars_iv_next]] = add nsw i32 %[[indvars_iv]], 1
+; CHECK-NEXT: %[[exitcond4:.*]] = icmp ne i32 %[[indvars_iv_next]], 0
 ; CHECK-NEXT: br i1 %[[exitcond4]], label %[[for_cond2_preheader]], label %[[for_end15:.*]]
 for.inc13:                                        ; preds = %for.cond2.for.inc13_crit_edge, %for.cond2.preheader
   %inc14 = add i8 %storemerge15, 1
diff --git a/test/Transforms/IndVarSimplify/pr22222.ll b/test/Transforms/IndVarSimplify/pr22222.ll
index ccdfe538dfa5..d1f04906ae4a 100644
--- a/test/Transforms/IndVarSimplify/pr22222.ll
+++ b/test/Transforms/IndVarSimplify/pr22222.ll
@@ -9,7 +9,7 @@ declare void @abort() #1
 ; Function Attrs: nounwind ssp uwtable
 define i32 @main() {
 entry:
-  %a.promoted13 = load i32* @a, align 4
+  %a.promoted13 = load i32, i32* @a, align 4
   br label %for.cond1.preheader
 
 for.cond1.preheader:                              ; preds = %entry, %for.end
diff --git a/test/Transforms/IndVarSimplify/preserve-signed-wrap.ll b/test/Transforms/IndVarSimplify/preserve-signed-wrap.ll
index f619e8d127ec..c8d34acb24ae 100644
--- a/test/Transforms/IndVarSimplify/preserve-signed-wrap.ll
+++ b/test/Transforms/IndVarSimplify/preserve-signed-wrap.ll
@@ -4,6 +4,10 @@
 ; sext for the addressing, however it shouldn't eliminate the sext
 ; on the other phi, since that value undergoes signed wrapping.
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 define void @foo(i32* nocapture %d, i32 %n) nounwind {
 entry:
 	%0 = icmp sgt i32 %n, 0		; <i1> [#uses=1]
@@ -21,7 +25,7 @@ bb:		; preds = %bb1, %bb.nph
 	%p.01 = phi i8 [ %4, %bb1 ], [ -1, %bb.nph ]		; <i8> [#uses=2]
 	%1 = sext i8 %p.01 to i32		; <i32> [#uses=1]
 	%2 = sext i32 %i.02 to i64		; <i64> [#uses=1]
-	%3 = getelementptr i32* %d, i64 %2		; <i32*> [#uses=1]
+	%3 = getelementptr i32, i32* %d, i64 %2		; <i32*> [#uses=1]
 	store i32 %1, i32* %3, align 4
 	%4 = add i8 %p.01, 1		; <i8> [#uses=1]
 	%5 = add i32 %i.02, 1		; <i32> [#uses=2]
diff --git a/test/Transforms/IndVarSimplify/promote-iv-to-eliminate-casts.ll b/test/Transforms/IndVarSimplify/promote-iv-to-eliminate-casts.ll
index a007ca60b08f..f2afaf42bc92 100644
--- a/test/Transforms/IndVarSimplify/promote-iv-to-eliminate-casts.ll
+++ b/test/Transforms/IndVarSimplify/promote-iv-to-eliminate-casts.ll
@@ -1,5 +1,8 @@
-; RUN: opt < %s -indvars -S > %t
-; RUN: not grep sext %t
+; RUN: opt < %s -indvars -S | not grep sext
+
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 
 define i64 @test(i64* nocapture %first, i32 %count) nounwind readonly {
 entry:
@@ -13,8 +16,8 @@ bb:		; preds = %bb1, %bb.nph
 	%result.02 = phi i64 [ %t5, %bb1 ], [ 0, %bb.nph ]		; <i64> [#uses=1]
 	%n.01 = phi i32 [ %t6, %bb1 ], [ 0, %bb.nph ]		; <i32> [#uses=2]
 	%t1 = sext i32 %n.01 to i64		; <i64> [#uses=1]
-	%t2 = getelementptr i64* %first, i64 %t1		; <i64*> [#uses=1]
-	%t3 = load i64* %t2, align 8		; <i64> [#uses=1]
+	%t2 = getelementptr i64, i64* %first, i64 %t1		; <i64*> [#uses=1]
+	%t3 = load i64, i64* %t2, align 8		; <i64> [#uses=1]
 	%t4 = lshr i64 %t3, 4		; <i64> [#uses=1]
 	%t5 = add i64 %t4, %result.02		; <i64> [#uses=2]
 	%t6 = add i32 %n.01, 1		; <i32> [#uses=2]
@@ -44,7 +47,7 @@ bb.nph:		; preds = %entry
 bb:		; preds = %bb1, %bb.nph
 	%i.01 = phi i16 [ %t3, %bb1 ], [ 0, %bb.nph ]		; <i16> [#uses=2]
 	%t1 = sext i16 %i.01 to i64		; <i64> [#uses=1]
-	%t2 = getelementptr i32* %P, i64 %t1		; <i32*> [#uses=1]
+	%t2 = getelementptr i32, i32* %P, i64 %t1		; <i32*> [#uses=1]
 	store i32 123, i32* %t2, align 4
 	%t3 = add i16 %i.01, 1		; <i16> [#uses=2]
 	br label %bb1
@@ -70,7 +73,7 @@ bb:             ; preds = %bb, %bb.thread
   %i.0.reg2mem.0 = phi i8 [ -10, %bb.thread ], [ %tmp7, %bb ]           ; <i8> [#uses=2]
   %tmp12 = sext i8 %i.0.reg2mem.0 to i32                ; <i32> [#uses=1]
   %tmp4 = add i32 %tmp12, 10            ; <i32> [#uses=1]
-  %tmp5 = getelementptr [21 x i32]* %a, i32 0, i32 %tmp4                ; <i32*> [#uses=1]
+  %tmp5 = getelementptr [21 x i32], [21 x i32]* %a, i32 0, i32 %tmp4                ; <i32*> [#uses=1]
   store i32 0, i32* %tmp5
   %tmp7 = add i8 %i.0.reg2mem.0, 1              ; <i8> [#uses=2]
   %0 = icmp sgt i8 %tmp7, 10            ; <i1> [#uses=1]
@@ -88,7 +91,7 @@ bb:             ; preds = %bb, %bb.thread
   %i.0.reg2mem.0 = phi i8 [ 10, %bb.thread ], [ %tmp7, %bb ]            ; <i8> [#uses=2]
   %tmp12 = sext i8 %i.0.reg2mem.0 to i32                ; <i32> [#uses=1]
   %tmp4 = add i32 %tmp12, -10           ; <i32> [#uses=1]
-  %tmp5 = getelementptr [21 x i32]* %a, i32 0, i32 %tmp4                ; <i32*> [#uses=1]
+  %tmp5 = getelementptr [21 x i32], [21 x i32]* %a, i32 0, i32 %tmp4                ; <i32*> [#uses=1]
   store i32 0, i32* %tmp5
   %tmp7 = add i8 %i.0.reg2mem.0, 1              ; <i8> [#uses=2]
   %0 = icmp sgt i8 %tmp7, 30            ; <i1> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/sharpen-range.ll b/test/Transforms/IndVarSimplify/sharpen-range.ll
index 5392dbc5fc04..c103da9cec70 100644
--- a/test/Transforms/IndVarSimplify/sharpen-range.ll
+++ b/test/Transforms/IndVarSimplify/sharpen-range.ll
@@ -8,7 +8,7 @@ declare void @abort()
 define i1 @bounded_below_slt(i32* nocapture readonly %buffer) {
 ; CHECK-LABEL: bounded_below_slt
 entry:
-  %length = load i32* %buffer, !range !0
+  %length = load i32, i32* %buffer, !range !0
   %entry.pred = icmp eq i32 %length, 0
   br i1 %entry.pred, label %abort, label %loop.preheader
 
@@ -42,7 +42,7 @@ oob:
 define i1 @bounded_below_sle(i32* nocapture readonly %buffer) {
 ; CHECK-LABEL: bounded_below_sle
 entry:
-  %length = load i32* %buffer, !range !0
+  %length = load i32, i32* %buffer, !range !0
   %entry.pred = icmp eq i32 %length, 0
   br i1 %entry.pred, label %abort, label %loop.preheader
 
@@ -92,7 +92,7 @@ loop.begin:
 loop:
 ; CHECK: loop
   %.sum = add i64 %i.01, -2
-  %v = getelementptr inbounds i8* null, i64 %.sum
+  %v = getelementptr inbounds i8, i8* null, i64 %.sum
   %r = tail call i32 @check(i8* %v)
   %c = icmp eq i32 %r, 0
   br i1 %c, label %loop.end, label %abort.now
diff --git a/test/Transforms/IndVarSimplify/signed-trip-count.ll b/test/Transforms/IndVarSimplify/signed-trip-count.ll
index 1a5e64ddc1b9..992e22e2a969 100644
--- a/test/Transforms/IndVarSimplify/signed-trip-count.ll
+++ b/test/Transforms/IndVarSimplify/signed-trip-count.ll
@@ -2,6 +2,10 @@
 ; RUN: not grep sext %t
 ; RUN: grep phi %t | count 1
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 define void @foo(i64* nocapture %x, i32 %n) nounwind {
 entry:
 	%tmp102 = icmp sgt i32 %n, 0		; <i1> [#uses=1]
@@ -13,7 +17,7 @@ bb.nph:		; preds = %entry
 bb:		; preds = %bb7, %bb.nph
 	%i.01 = phi i32 [ %tmp6, %bb7 ], [ 0, %bb.nph ]		; <i32> [#uses=3]
 	%tmp1 = sext i32 %i.01 to i64		; <i64> [#uses=1]
-	%tmp4 = getelementptr i64* %x, i32 %i.01		; <i64*> [#uses=1]
+	%tmp4 = getelementptr i64, i64* %x, i32 %i.01		; <i64*> [#uses=1]
 	store i64 %tmp1, i64* %tmp4, align 8
 	%tmp6 = add i32 %i.01, 1		; <i32> [#uses=2]
 	br label %bb7
diff --git a/test/Transforms/IndVarSimplify/single-element-range.ll b/test/Transforms/IndVarSimplify/single-element-range.ll
index 4b035eea1491..e047a0b254e1 100644
--- a/test/Transforms/IndVarSimplify/single-element-range.ll
+++ b/test/Transforms/IndVarSimplify/single-element-range.ll
@@ -8,7 +8,7 @@ entry:
 	br i1 undef, label %return, label %bb
 
 bb:		; preds = %entry
-	%0 = load i8** undef, align 4		; <i8*> [#uses=2]
+	%0 = load i8*, i8** undef, align 4		; <i8*> [#uses=2]
 	%1 = ptrtoint i8* %0 to i32		; <i32> [#uses=1]
 	%2 = icmp sgt i8* %0, inttoptr (i32 1 to i8*)		; <i1> [#uses=1]
 	br i1 %2, label %bb1, label %bb5
diff --git a/test/Transforms/IndVarSimplify/sink-alloca.ll b/test/Transforms/IndVarSimplify/sink-alloca.ll
index 64207d823d2c..38c2d3140de6 100644
--- a/test/Transforms/IndVarSimplify/sink-alloca.ll
+++ b/test/Transforms/IndVarSimplify/sink-alloca.ll
@@ -19,7 +19,7 @@ while.cond:                                       ; preds = %while.cond, %entry
 
 while.end:                                        ; preds = %while.cond
   store volatile i32 0, i32* %result.i
-  %tmp.i = load volatile i32* %result.i           ; <i32> [#uses=0]
+  %tmp.i = load volatile i32, i32* %result.i           ; <i32> [#uses=0]
   ret i32 0
 }
 declare i32 @bar()
@@ -42,8 +42,8 @@ entry:
 
 for.body.i:
   %indvars.iv37.i = phi i64 [ %indvars.iv.next38.i, %for.body.i ], [ 0, %entry ]
-  %call.i = call i8* (...)* @a() nounwind
-  %arrayidx.i = getelementptr inbounds i8** %vla.i, i64 %indvars.iv37.i
+  %call.i = call i8* (...) @a() nounwind
+  %arrayidx.i = getelementptr inbounds i8*, i8** %vla.i, i64 %indvars.iv37.i
   store i8* %call.i, i8** %arrayidx.i, align 8
   %indvars.iv.next38.i = add i64 %indvars.iv37.i, 1
   %exitcond5 = icmp eq i64 %indvars.iv.next38.i, %n
@@ -51,6 +51,6 @@ for.body.i:
 
 g.exit:
   call void @llvm.stackrestore(i8* %savedstack) nounwind
-  %call1 = call i8* (...)* @a(i8** %vla) nounwind
+  %call1 = call i8* (...) @a(i8** %vla) nounwind
   ret void
 }
diff --git a/test/Transforms/IndVarSimplify/udiv.ll b/test/Transforms/IndVarSimplify/udiv.ll
index aee455d08c80..45c703c99dde 100644
--- a/test/Transforms/IndVarSimplify/udiv.ll
+++ b/test/Transforms/IndVarSimplify/udiv.ll
@@ -17,8 +17,8 @@ entry:
   br i1 %cmp, label %cond.true, label %while.cond.preheader
 
 cond.true:                                        ; preds = %entry
-  %arrayidx = getelementptr inbounds i8** %argv, i64 1 ; <i8**> [#uses=1]
-  %tmp2 = load i8** %arrayidx                     ; <i8*> [#uses=1]
+  %arrayidx = getelementptr inbounds i8*, i8** %argv, i64 1 ; <i8**> [#uses=1]
+  %tmp2 = load i8*, i8** %arrayidx                     ; <i8*> [#uses=1]
   %call = tail call i32 @atoi(i8* %tmp2) nounwind readonly ; <i32> [#uses=1]
   br label %while.cond.preheader
 
@@ -59,7 +59,7 @@ bb.nph:                                           ; preds = %for.cond.preheader
 
 for.body:                                         ; preds = %bb.nph, %for.cond
   %i.02 = phi i64 [ 2, %bb.nph ], [ %inc, %for.cond ] ; <i64> [#uses=2]
-  %arrayidx10 = getelementptr inbounds [8193 x i8]* @main.flags, i64 0, i64 %i.02 ; <i8*> [#uses=1]
+  %arrayidx10 = getelementptr inbounds [8193 x i8], [8193 x i8]* @main.flags, i64 0, i64 %i.02 ; <i8*> [#uses=1]
   store i8 1, i8* %arrayidx10
   %inc = add nsw i64 %i.02, 1                     ; <i64> [#uses=2]
   br label %for.cond
@@ -81,8 +81,8 @@ bb.nph16:                                         ; preds = %for.cond12.loopexit
 for.body15:                                       ; preds = %bb.nph16, %for.cond12
   %count.212 = phi i32 [ 0, %bb.nph16 ], [ %count.1, %for.cond12 ] ; <i32> [#uses=2]
   %i.17 = phi i64 [ 2, %bb.nph16 ], [ %inc37, %for.cond12 ] ; <i64> [#uses=4]
-  %arrayidx17 = getelementptr inbounds [8193 x i8]* @main.flags, i64 0, i64 %i.17 ; <i8*> [#uses=1]
-  %tmp18 = load i8* %arrayidx17                   ; <i8> [#uses=1]
+  %arrayidx17 = getelementptr inbounds [8193 x i8], [8193 x i8]* @main.flags, i64 0, i64 %i.17 ; <i8*> [#uses=1]
+  %tmp18 = load i8, i8* %arrayidx17                   ; <i8> [#uses=1]
   %tobool19 = icmp eq i8 %tmp18, 0                ; <i1> [#uses=1]
   br i1 %tobool19, label %for.inc35, label %if.then
 
@@ -103,7 +103,7 @@ bb.nph5:                                          ; preds = %if.then
 
 for.body25:                                       ; preds = %bb.nph5, %for.cond22
   %k.04 = phi i64 [ %add, %bb.nph5 ], [ %add31, %for.cond22 ] ; <i64> [#uses=2]
-  %arrayidx27 = getelementptr inbounds [8193 x i8]* @main.flags, i64 0, i64 %k.04 ; <i8*> [#uses=1]
+  %arrayidx27 = getelementptr inbounds [8193 x i8], [8193 x i8]* @main.flags, i64 0, i64 %k.04 ; <i8*> [#uses=1]
   store i8 0, i8* %arrayidx27
   %add31 = add nsw i64 %k.04, %i.17               ; <i64> [#uses=2]
   br label %for.cond22
@@ -119,7 +119,7 @@ for.inc35:                                        ; preds = %for.body15, %for.en
 
 while.end:                                        ; preds = %while.cond.while.end_crit_edge, %while.cond.preheader
   %count.0.lcssa = phi i32 [ %count.2.lcssa.lcssa, %while.cond.while.end_crit_edge ], [ 0, %while.cond.preheader ] ; <i32> [#uses=1]
-  %call40 = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([11 x i8]* @.str, i64 0, i64 0), i32 %count.0.lcssa) nounwind ; <i32> [#uses=0]
+  %call40 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str, i64 0, i64 0), i32 %count.0.lcssa) nounwind ; <i32> [#uses=0]
   ret i32 0
 }
 
@@ -146,7 +146,7 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %i.03 = phi i64 [ %inc, %for.body ], [ 0, %for.body.preheader ] ; <i64> [#uses=2]
-  %arrayidx = getelementptr inbounds double* %p, i64 %i.03 ; <double*> [#uses=1]
+  %arrayidx = getelementptr inbounds double, double* %p, i64 %i.03 ; <double*> [#uses=1]
   store double 0.000000e+00, double* %arrayidx
   %inc = add i64 %i.03, 1                         ; <i64> [#uses=2]
   %divx = udiv i64 %n, 7                           ; <i64> [#uses=1]
diff --git a/test/Transforms/IndVarSimplify/uglygep.ll b/test/Transforms/IndVarSimplify/uglygep.ll
index 0014b683db4b..e4343891be65 100644
--- a/test/Transforms/IndVarSimplify/uglygep.ll
+++ b/test/Transforms/IndVarSimplify/uglygep.ll
@@ -27,10 +27,10 @@ bb2:                                              ; preds = %bb1, %bb
   br i1 %tmp8, label %bb1, label %bb3
 
 bb1:                                              ; preds = %bb2
-  %tmp = load double*** @tds, align 8             ; <double**> [#uses=1]
+  %tmp = load double**, double*** @tds, align 8             ; <double**> [#uses=1]
   %tmp1 = sext i32 %i.0 to i64                    ; <i64> [#uses=1]
-  %tmp2 = getelementptr inbounds double** %tmp, i64 %tmp1 ; <double**> [#uses=1]
-  %tmp3 = load double** %tmp2, align 1            ; <double*> [#uses=1]
+  %tmp2 = getelementptr inbounds double*, double** %tmp, i64 %tmp1 ; <double**> [#uses=1]
+  %tmp3 = load double*, double** %tmp2, align 1            ; <double*> [#uses=1]
   %tmp6 = add nsw i32 %j.0, 1                     ; <i32> [#uses=1]
   br label %bb2
 
diff --git a/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll b/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
index a421003aa30d..7272ef1ec135 100644
--- a/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
+++ b/test/Transforms/IndVarSimplify/ult-sub-to-eq.ll
@@ -1,5 +1,9 @@
 ; RUN: opt -S -indvars < %s | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 define void @test1(float* nocapture %autoc, float* nocapture %data, float %d, i32 %data_len, i32 %sample) nounwind {
 entry:
   %sub = sub i32 %data_len, %sample
@@ -11,11 +15,11 @@ for.body:                                         ; preds = %entry, %for.body
   %0 = trunc i64 %indvars.iv to i32
   %add = add i32 %0, %sample
   %idxprom = zext i32 %add to i64
-  %arrayidx = getelementptr inbounds float* %data, i64 %idxprom
-  %1 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %data, i64 %idxprom
+  %1 = load float, float* %arrayidx, align 4
   %mul = fmul float %1, %d
-  %arrayidx2 = getelementptr inbounds float* %autoc, i64 %indvars.iv
-  %2 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %autoc, i64 %indvars.iv
+  %2 = load float, float* %arrayidx2, align 4
   %add3 = fadd float %2, %mul
   store float %add3, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/IndVarSimplify/use-range-metadata.ll b/test/Transforms/IndVarSimplify/use-range-metadata.ll
index ea3b12dac391..1f0142608b5d 100644
--- a/test/Transforms/IndVarSimplify/use-range-metadata.ll
+++ b/test/Transforms/IndVarSimplify/use-range-metadata.ll
@@ -6,7 +6,7 @@ declare void @abort()
 
 define i1 @iterate(i32* nocapture readonly %buffer) {
 entry:
-  %length = load i32* %buffer, !range !0
+  %length = load i32, i32* %buffer, !range !0
   br label %loop.preheader
 
 loop.preheader:
diff --git a/test/Transforms/IndVarSimplify/variable-stride-ivs-0.ll b/test/Transforms/IndVarSimplify/variable-stride-ivs-0.ll
index fb9ef22549c0..5fa4a17b9159 100644
--- a/test/Transforms/IndVarSimplify/variable-stride-ivs-0.ll
+++ b/test/Transforms/IndVarSimplify/variable-stride-ivs-0.ll
@@ -6,17 +6,17 @@
 
 define void @vnum_test8(i32* %data) {
 entry:
-        %tmp.1 = getelementptr i32* %data, i32 3                ; <i32*> [#uses=1]
-        %tmp.2 = load i32* %tmp.1               ; <i32> [#uses=2]
-        %tmp.4 = getelementptr i32* %data, i32 4                ; <i32*> [#uses=1]
-        %tmp.5 = load i32* %tmp.4               ; <i32> [#uses=2]
-        %tmp.8 = getelementptr i32* %data, i32 2                ; <i32*> [#uses=1]
-        %tmp.9 = load i32* %tmp.8               ; <i32> [#uses=3]
+        %tmp.1 = getelementptr i32, i32* %data, i32 3                ; <i32*> [#uses=1]
+        %tmp.2 = load i32, i32* %tmp.1               ; <i32> [#uses=2]
+        %tmp.4 = getelementptr i32, i32* %data, i32 4                ; <i32*> [#uses=1]
+        %tmp.5 = load i32, i32* %tmp.4               ; <i32> [#uses=2]
+        %tmp.8 = getelementptr i32, i32* %data, i32 2                ; <i32*> [#uses=1]
+        %tmp.9 = load i32, i32* %tmp.8               ; <i32> [#uses=3]
         %tmp.125 = icmp sgt i32 %tmp.2, 0               ; <i1> [#uses=1]
         br i1 %tmp.125, label %no_exit.preheader, label %return
 
 no_exit.preheader:              ; preds = %entry
-        %tmp.16 = getelementptr i32* %data, i32 %tmp.9          ; <i32*> [#uses=1]
+        %tmp.16 = getelementptr i32, i32* %data, i32 %tmp.9          ; <i32*> [#uses=1]
         br label %no_exit
 
 ; CHECK: store i32 0
diff --git a/test/Transforms/IndVarSimplify/verify-scev.ll b/test/Transforms/IndVarSimplify/verify-scev.ll
index b9ce3d63ec99..ddf2e7f1e144 100644
--- a/test/Transforms/IndVarSimplify/verify-scev.ll
+++ b/test/Transforms/IndVarSimplify/verify-scev.ll
@@ -175,7 +175,7 @@ if.end250:                                        ; preds = %for.cond246
   br i1 undef, label %if.end256, label %for.end562
 
 if.end256:                                        ; preds = %if.end250
-  %0 = load i32* undef, align 4
+  %0 = load i32, i32* undef, align 4
   br i1 undef, label %if.then274, label %for.cond404.preheader
 
 for.cond404.preheader:                            ; preds = %if.end256
@@ -379,7 +379,7 @@ for.body48:                                       ; preds = %for.inc221, %for.bo
   br i1 undef, label %for.inc221, label %for.body65.lr.ph
 
 for.body65.lr.ph:                                 ; preds = %for.body48
-  %0 = load i32* undef, align 4
+  %0 = load i32, i32* undef, align 4
   %1 = sext i32 %0 to i64
   br label %for.body65.us
 
diff --git a/test/Transforms/IndVarSimplify/widen-loop-comp.ll b/test/Transforms/IndVarSimplify/widen-loop-comp.ll
index 0930a0c41390..eb81ceb700b0 100644
--- a/test/Transforms/IndVarSimplify/widen-loop-comp.ll
+++ b/test/Transforms/IndVarSimplify/widen-loop-comp.ll
@@ -1,6 +1,10 @@
 ; RUN: opt < %s -indvars -S | FileCheck %s
 target triple = "aarch64--linux-gnu"
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 ; Check the loop exit i32 compare instruction and operand are widened to i64
 ; instead of truncating IV before its use in the i32 compare instruction.
 
@@ -19,13 +23,13 @@ target triple = "aarch64--linux-gnu"
 define i32 @test1() {
 entry:
   store i32 -1, i32* @idx, align 4
-  %0 = load i32* @e, align 4
+  %0 = load i32, i32* @e, align 4
   %cmp4 = icmp slt i32 %0, 0
   br i1 %cmp4, label %for.end.loopexit, label %for.body.lr.ph
 
 for.body.lr.ph:
-  %1 = load i32** @ptr, align 8
-  %2 = load i32* @e, align 4
+  %1 = load i32*, i32** @ptr, align 8
+  %2 = load i32, i32* @e, align 4
   br label %for.body
 
 for.cond:
@@ -36,8 +40,8 @@ for.cond:
 for.body:
   %i.05 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.cond ]
   %idxprom = sext i32 %i.05 to i64
-  %arrayidx = getelementptr inbounds i32* %1, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %1, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %3, 0
   br i1 %tobool, label %if.then, label %for.cond
 
@@ -53,7 +57,7 @@ for.end.loopexit:
   br label %for.end
 
 for.end:
-  %4 = load i32* @idx, align 4
+  %4 = load i32, i32* @idx, align 4
   ret i32 %4
 }
 
@@ -81,11 +85,11 @@ for.inc13.us:
 for.body4.us:
   %storemerge14.us = phi i32 [ 0, %for.body4.lr.ph.us ], [ %inc.us, %for.body4.us ]
   %idxprom.us = sext i32 %storemerge14.us to i64
-  %arrayidx6.us = getelementptr inbounds [8 x i8]* %a, i64 %idxprom5.us, i64 %idxprom.us
-  %0 = load i8* %arrayidx6.us, align 1
+  %arrayidx6.us = getelementptr inbounds [8 x i8], [8 x i8]* %a, i64 %idxprom5.us, i64 %idxprom.us
+  %0 = load i8, i8* %arrayidx6.us, align 1
   %idxprom7.us = zext i8 %0 to i64
-  %arrayidx8.us = getelementptr inbounds i8* %b, i64 %idxprom7.us
-  %1 = load i8* %arrayidx8.us, align 1
+  %arrayidx8.us = getelementptr inbounds i8, i8* %b, i64 %idxprom7.us
+  %1 = load i8, i8* %arrayidx8.us, align 1
   store i8 %1, i8* %arrayidx6.us, align 1
   %inc.us = add nsw i32 %storemerge14.us, 1
   %cmp2.us = icmp slt i32 %inc.us, %conv
@@ -126,8 +130,8 @@ for.cond:
 
 for.body:
   %idxprom = sext i32 %i.0 to i64
-  %arrayidx = getelementptr inbounds i32* %a, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %sum.0, %0
   %inc = add nsw i32 %i.0, 1
   br label %for.cond
@@ -180,8 +184,8 @@ for.cond:
 
 for.body:
   %idxprom = zext i32 %i.0 to i64
-  %arrayidx = getelementptr inbounds i32* %a, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %sum.0, %0
   %inc = add nsw i32 %i.0, 1
   br label %for.cond
diff --git a/test/Transforms/IndVarSimplify/widen-nsw.ll b/test/Transforms/IndVarSimplify/widen-nsw.ll
index 56c3c0d9f128..8dbbb51ee51d 100644
--- a/test/Transforms/IndVarSimplify/widen-nsw.ll
+++ b/test/Transforms/IndVarSimplify/widen-nsw.ll
@@ -16,8 +16,8 @@ for.cond:                                         ; preds = %for.body, %entry
 
 for.body:                                         ; preds = %for.cond
   %idxprom = sext i32 %i.0 to i64
-  %arrayidx = getelementptr inbounds i32* %a, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %sum.0, %0
   %inc = add nsw i32 %i.0, 1
   br label %for.cond
diff --git a/test/Transforms/Inline/2003-09-22-PHINodeInlineFail.ll b/test/Transforms/Inline/2003-09-22-PHINodeInlineFail.ll
index b8ca56050dca..df0b472bb2a5 100644
--- a/test/Transforms/Inline/2003-09-22-PHINodeInlineFail.ll
+++ b/test/Transforms/Inline/2003-09-22-PHINodeInlineFail.ll
@@ -9,7 +9,7 @@ LongJmpBlkPost:
         ret i32 0
 
 LongJmpBlkPre:
-        %i.3 = phi i32 [ 0, %entry ], [ 0, %entry ]             ; <i32> [#uses=0]
+        %i.3 = phi i32 [ 0, %entry ]
         %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
                  cleanup
         ret i32 0
diff --git a/test/Transforms/Inline/2003-09-22-PHINodesInExceptionDest.ll b/test/Transforms/Inline/2003-09-22-PHINodesInExceptionDest.ll
index 9af93325f142..d5416a205979 100644
--- a/test/Transforms/Inline/2003-09-22-PHINodesInExceptionDest.ll
+++ b/test/Transforms/Inline/2003-09-22-PHINodesInExceptionDest.ll
@@ -9,7 +9,7 @@ Call2Invoke:            ; preds = %entry
         br label %exit
 
 LongJmpBlkPre:          ; preds = %Call2Invoke, %entry
-        %i.3 = phi i32 [ 0, %entry ], [ 0, %Call2Invoke ]               ; <i32> [#uses=0]
+        %i.3 = phi i32 [ 0, %entry ]
         %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
                  cleanup
         br label %exit
diff --git a/test/Transforms/Inline/2006-07-12-InlinePruneCGUpdate.ll b/test/Transforms/Inline/2006-07-12-InlinePruneCGUpdate.ll
index 69345627221b..a0ddacdbd5b1 100644
--- a/test/Transforms/Inline/2006-07-12-InlinePruneCGUpdate.ll
+++ b/test/Transforms/Inline/2006-07-12-InlinePruneCGUpdate.ll
@@ -214,7 +214,7 @@ bb:		; preds = %cond_true
 	br label %bb3
 
 bb3:		; preds = %bb, %entry
-	%tmp5 = load i8** null		; <i8*> [#uses=1]
+	%tmp5 = load i8*, i8** null		; <i8*> [#uses=1]
 	%tmp = icmp ne i8* null, %tmp5		; <i1> [#uses=1]
 	br i1 %tmp, label %cond_true, label %cond_false
 
diff --git a/test/Transforms/Inline/2009-01-08-NoInlineDynamicAlloca.ll b/test/Transforms/Inline/2009-01-08-NoInlineDynamicAlloca.ll
index db2a799225b9..1bfb55144f5f 100644
--- a/test/Transforms/Inline/2009-01-08-NoInlineDynamicAlloca.ll
+++ b/test/Transforms/Inline/2009-01-08-NoInlineDynamicAlloca.ll
@@ -10,15 +10,15 @@ entry:
 	%p = alloca i8*		; <i8**> [#uses=2]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store i32 %i, i32* %i_addr
-	%0 = load i32* %i_addr, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* %i_addr, align 4		; <i32> [#uses=1]
 	%1 = alloca i8, i32 %0		; <i8*> [#uses=1]
 	store i8* %1, i8** %p, align 4
-	%2 = load i8** %p, align 4		; <i8*> [#uses=1]
+	%2 = load i8*, i8** %p, align 4		; <i8*> [#uses=1]
 	store i8* %2, i8** @q, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load i8** %retval		; <i8*> [#uses=1]
+	%retval1 = load i8*, i8** %retval		; <i8*> [#uses=1]
 	ret i8* %retval1
 }
 
@@ -27,7 +27,7 @@ entry:
 	%i_addr = alloca i32		; <i32*> [#uses=2]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store i32 %i, i32* %i_addr
-	%0 = load i32* %i_addr, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* %i_addr, align 4		; <i32> [#uses=1]
 	%1 = call i8* @a(i32 %0) nounwind		; <i8*> [#uses=0]
 	br label %return
 
diff --git a/test/Transforms/Inline/2009-01-13-RecursiveInlineCrash.ll b/test/Transforms/Inline/2009-01-13-RecursiveInlineCrash.ll
index 7d8d16bacffe..8d8f20feb739 100644
--- a/test/Transforms/Inline/2009-01-13-RecursiveInlineCrash.ll
+++ b/test/Transforms/Inline/2009-01-13-RecursiveInlineCrash.ll
@@ -75,23 +75,23 @@ entry:
 	]
 
 bb:		; preds = %entry
-	%0 = getelementptr %struct.quad_struct* %tree, i32 0, i32 3		; <%struct.quad_struct**> [#uses=1]
-	%1 = load %struct.quad_struct** %0, align 4		; <%struct.quad_struct*> [#uses=1]
+	%0 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 3		; <%struct.quad_struct**> [#uses=1]
+	%1 = load %struct.quad_struct*, %struct.quad_struct** %0, align 4		; <%struct.quad_struct*> [#uses=1]
 	ret %struct.quad_struct* %1
 
 bb1:		; preds = %entry
-	%2 = getelementptr %struct.quad_struct* %tree, i32 0, i32 2		; <%struct.quad_struct**> [#uses=1]
-	%3 = load %struct.quad_struct** %2, align 4		; <%struct.quad_struct*> [#uses=1]
+	%2 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 2		; <%struct.quad_struct**> [#uses=1]
+	%3 = load %struct.quad_struct*, %struct.quad_struct** %2, align 4		; <%struct.quad_struct*> [#uses=1]
 	ret %struct.quad_struct* %3
 
 bb2:		; preds = %entry
-	%4 = getelementptr %struct.quad_struct* %tree, i32 0, i32 5		; <%struct.quad_struct**> [#uses=1]
-	%5 = load %struct.quad_struct** %4, align 4		; <%struct.quad_struct*> [#uses=1]
+	%4 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 5		; <%struct.quad_struct**> [#uses=1]
+	%5 = load %struct.quad_struct*, %struct.quad_struct** %4, align 4		; <%struct.quad_struct*> [#uses=1]
 	ret %struct.quad_struct* %5
 
 bb3:		; preds = %entry
-	%6 = getelementptr %struct.quad_struct* %tree, i32 0, i32 4		; <%struct.quad_struct**> [#uses=1]
-	%7 = load %struct.quad_struct** %6, align 4		; <%struct.quad_struct*> [#uses=1]
+	%6 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 4		; <%struct.quad_struct**> [#uses=1]
+	%7 = load %struct.quad_struct*, %struct.quad_struct** %6, align 4		; <%struct.quad_struct*> [#uses=1]
 	ret %struct.quad_struct* %7
 
 bb5:		; preds = %entry
@@ -100,10 +100,10 @@ bb5:		; preds = %entry
 
 define internal fastcc %struct.quad_struct* @gtequal_adj_neighbor(%struct.quad_struct* nocapture %tree, i32 %d) nounwind readonly {
 entry:
-	%0 = getelementptr %struct.quad_struct* %tree, i32 0, i32 6		; <%struct.quad_struct**> [#uses=1]
-	%1 = load %struct.quad_struct** %0, align 4		; <%struct.quad_struct*> [#uses=4]
-	%2 = getelementptr %struct.quad_struct* %tree, i32 0, i32 1		; <i32*> [#uses=1]
-	%3 = load i32* %2, align 4		; <i32> [#uses=2]
+	%0 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 6		; <%struct.quad_struct**> [#uses=1]
+	%1 = load %struct.quad_struct*, %struct.quad_struct** %0, align 4		; <%struct.quad_struct*> [#uses=4]
+	%2 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 1		; <i32*> [#uses=1]
+	%3 = load i32, i32* %2, align 4		; <i32> [#uses=2]
 	%4 = icmp eq %struct.quad_struct* %1, null		; <i1> [#uses=1]
 	br i1 %4, label %bb3, label %bb
 
@@ -122,8 +122,8 @@ bb3:		; preds = %bb1, %bb, %entry
 	br i1 %8, label %bb7, label %bb4
 
 bb4:		; preds = %bb3
-	%9 = getelementptr %struct.quad_struct* %q.0, i32 0, i32 0		; <i32*> [#uses=1]
-	%10 = load i32* %9, align 4		; <i32> [#uses=1]
+	%9 = getelementptr %struct.quad_struct, %struct.quad_struct* %q.0, i32 0, i32 0		; <i32*> [#uses=1]
+	%10 = load i32, i32* %9, align 4		; <i32> [#uses=1]
 	%11 = icmp eq i32 %10, 2		; <i1> [#uses=1]
 	br i1 %11, label %bb5, label %bb7
 
@@ -140,36 +140,36 @@ declare fastcc i32 @sum_adjacent(%struct.quad_struct* nocapture, i32, i32, i32)
 
 define i32 @perimeter(%struct.quad_struct* nocapture %tree, i32 %size) nounwind readonly {
 entry:
-	%0 = getelementptr %struct.quad_struct* %tree, i32 0, i32 0		; <i32*> [#uses=1]
-	%1 = load i32* %0, align 4		; <i32> [#uses=1]
+	%0 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 0		; <i32*> [#uses=1]
+	%1 = load i32, i32* %0, align 4		; <i32> [#uses=1]
 	%2 = icmp eq i32 %1, 2		; <i1> [#uses=1]
 	br i1 %2, label %bb, label %bb2
 
 bb:		; preds = %entry
-	%3 = getelementptr %struct.quad_struct* %tree, i32 0, i32 4		; <%struct.quad_struct**> [#uses=1]
-	%4 = load %struct.quad_struct** %3, align 4		; <%struct.quad_struct*> [#uses=1]
+	%3 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 4		; <%struct.quad_struct**> [#uses=1]
+	%4 = load %struct.quad_struct*, %struct.quad_struct** %3, align 4		; <%struct.quad_struct*> [#uses=1]
 	%5 = sdiv i32 %size, 2		; <i32> [#uses=1]
 	%6 = call i32 @perimeter(%struct.quad_struct* %4, i32 %5) nounwind		; <i32> [#uses=1]
-	%7 = getelementptr %struct.quad_struct* %tree, i32 0, i32 5		; <%struct.quad_struct**> [#uses=1]
-	%8 = load %struct.quad_struct** %7, align 4		; <%struct.quad_struct*> [#uses=1]
+	%7 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 5		; <%struct.quad_struct**> [#uses=1]
+	%8 = load %struct.quad_struct*, %struct.quad_struct** %7, align 4		; <%struct.quad_struct*> [#uses=1]
 	%9 = sdiv i32 %size, 2		; <i32> [#uses=1]
 	%10 = call i32 @perimeter(%struct.quad_struct* %8, i32 %9) nounwind		; <i32> [#uses=1]
 	%11 = add i32 %10, %6		; <i32> [#uses=1]
-	%12 = getelementptr %struct.quad_struct* %tree, i32 0, i32 3		; <%struct.quad_struct**> [#uses=1]
-	%13 = load %struct.quad_struct** %12, align 4		; <%struct.quad_struct*> [#uses=1]
+	%12 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 3		; <%struct.quad_struct**> [#uses=1]
+	%13 = load %struct.quad_struct*, %struct.quad_struct** %12, align 4		; <%struct.quad_struct*> [#uses=1]
 	%14 = sdiv i32 %size, 2		; <i32> [#uses=1]
 	%15 = call i32 @perimeter(%struct.quad_struct* %13, i32 %14) nounwind		; <i32> [#uses=1]
 	%16 = add i32 %15, %11		; <i32> [#uses=1]
-	%17 = getelementptr %struct.quad_struct* %tree, i32 0, i32 2		; <%struct.quad_struct**> [#uses=1]
-	%18 = load %struct.quad_struct** %17, align 4		; <%struct.quad_struct*> [#uses=1]
+	%17 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 2		; <%struct.quad_struct**> [#uses=1]
+	%18 = load %struct.quad_struct*, %struct.quad_struct** %17, align 4		; <%struct.quad_struct*> [#uses=1]
 	%19 = sdiv i32 %size, 2		; <i32> [#uses=1]
 	%20 = call i32 @perimeter(%struct.quad_struct* %18, i32 %19) nounwind		; <i32> [#uses=1]
 	%21 = add i32 %20, %16		; <i32> [#uses=1]
 	ret i32 %21
 
 bb2:		; preds = %entry
-	%22 = getelementptr %struct.quad_struct* %tree, i32 0, i32 0		; <i32*> [#uses=1]
-	%23 = load i32* %22, align 4		; <i32> [#uses=1]
+	%22 = getelementptr %struct.quad_struct, %struct.quad_struct* %tree, i32 0, i32 0		; <i32*> [#uses=1]
+	%23 = load i32, i32* %22, align 4		; <i32> [#uses=1]
 	%24 = icmp eq i32 %23, 0		; <i1> [#uses=1]
 	br i1 %24, label %bb3, label %bb23
 
@@ -179,14 +179,14 @@ bb3:		; preds = %bb2
 	br i1 %26, label %bb8, label %bb4
 
 bb4:		; preds = %bb3
-	%27 = getelementptr %struct.quad_struct* %25, i32 0, i32 0		; <i32*> [#uses=1]
-	%28 = load i32* %27, align 4		; <i32> [#uses=1]
+	%27 = getelementptr %struct.quad_struct, %struct.quad_struct* %25, i32 0, i32 0		; <i32*> [#uses=1]
+	%28 = load i32, i32* %27, align 4		; <i32> [#uses=1]
 	%29 = icmp eq i32 %28, 1		; <i1> [#uses=1]
 	br i1 %29, label %bb8, label %bb6
 
 bb6:		; preds = %bb4
-	%30 = getelementptr %struct.quad_struct* %25, i32 0, i32 0		; <i32*> [#uses=1]
-	%31 = load i32* %30, align 4		; <i32> [#uses=1]
+	%30 = getelementptr %struct.quad_struct, %struct.quad_struct* %25, i32 0, i32 0		; <i32*> [#uses=1]
+	%31 = load i32, i32* %30, align 4		; <i32> [#uses=1]
 	%32 = icmp eq i32 %31, 2		; <i1> [#uses=1]
 	br i1 %32, label %bb7, label %bb8
 
@@ -201,8 +201,8 @@ bb8:		; preds = %bb7, %bb6, %bb4, %bb3
 	br i1 %35, label %bb10, label %bb9
 
 bb9:		; preds = %bb8
-	%36 = getelementptr %struct.quad_struct* %34, i32 0, i32 0		; <i32*> [#uses=1]
-	%37 = load i32* %36, align 4		; <i32> [#uses=1]
+	%36 = getelementptr %struct.quad_struct, %struct.quad_struct* %34, i32 0, i32 0		; <i32*> [#uses=1]
+	%37 = load i32, i32* %36, align 4		; <i32> [#uses=1]
 	%38 = icmp eq i32 %37, 1		; <i1> [#uses=1]
 	br i1 %38, label %bb10, label %bb11
 
@@ -211,8 +211,8 @@ bb10:		; preds = %bb9, %bb8
 	br label %bb13
 
 bb11:		; preds = %bb9
-	%40 = getelementptr %struct.quad_struct* %34, i32 0, i32 0		; <i32*> [#uses=1]
-	%41 = load i32* %40, align 4		; <i32> [#uses=1]
+	%40 = getelementptr %struct.quad_struct, %struct.quad_struct* %34, i32 0, i32 0		; <i32*> [#uses=1]
+	%41 = load i32, i32* %40, align 4		; <i32> [#uses=1]
 	%42 = icmp eq i32 %41, 2		; <i1> [#uses=1]
 	br i1 %42, label %bb12, label %bb13
 
@@ -228,8 +228,8 @@ bb13:		; preds = %bb12, %bb11, %bb10
 	br i1 %46, label %bb15, label %bb14
 
 bb14:		; preds = %bb13
-	%47 = getelementptr %struct.quad_struct* %45, i32 0, i32 0		; <i32*> [#uses=1]
-	%48 = load i32* %47, align 4		; <i32> [#uses=1]
+	%47 = getelementptr %struct.quad_struct, %struct.quad_struct* %45, i32 0, i32 0		; <i32*> [#uses=1]
+	%48 = load i32, i32* %47, align 4		; <i32> [#uses=1]
 	%49 = icmp eq i32 %48, 1		; <i1> [#uses=1]
 	br i1 %49, label %bb15, label %bb16
 
@@ -238,8 +238,8 @@ bb15:		; preds = %bb14, %bb13
 	br label %bb18
 
 bb16:		; preds = %bb14
-	%51 = getelementptr %struct.quad_struct* %45, i32 0, i32 0		; <i32*> [#uses=1]
-	%52 = load i32* %51, align 4		; <i32> [#uses=1]
+	%51 = getelementptr %struct.quad_struct, %struct.quad_struct* %45, i32 0, i32 0		; <i32*> [#uses=1]
+	%52 = load i32, i32* %51, align 4		; <i32> [#uses=1]
 	%53 = icmp eq i32 %52, 2		; <i1> [#uses=1]
 	br i1 %53, label %bb17, label %bb18
 
@@ -255,8 +255,8 @@ bb18:		; preds = %bb17, %bb16, %bb15
 	br i1 %57, label %bb20, label %bb19
 
 bb19:		; preds = %bb18
-	%58 = getelementptr %struct.quad_struct* %56, i32 0, i32 0		; <i32*> [#uses=1]
-	%59 = load i32* %58, align 4		; <i32> [#uses=1]
+	%58 = getelementptr %struct.quad_struct, %struct.quad_struct* %56, i32 0, i32 0		; <i32*> [#uses=1]
+	%59 = load i32, i32* %58, align 4		; <i32> [#uses=1]
 	%60 = icmp eq i32 %59, 1		; <i1> [#uses=1]
 	br i1 %60, label %bb20, label %bb21
 
@@ -265,8 +265,8 @@ bb20:		; preds = %bb19, %bb18
 	ret i32 %61
 
 bb21:		; preds = %bb19
-	%62 = getelementptr %struct.quad_struct* %56, i32 0, i32 0		; <i32*> [#uses=1]
-	%63 = load i32* %62, align 4		; <i32> [#uses=1]
+	%62 = getelementptr %struct.quad_struct, %struct.quad_struct* %56, i32 0, i32 0		; <i32*> [#uses=1]
+	%63 = load i32, i32* %62, align 4		; <i32> [#uses=1]
 	%64 = icmp eq i32 %63, 2		; <i1> [#uses=1]
 	br i1 %64, label %bb22, label %bb23
 
diff --git a/test/Transforms/Inline/align.ll b/test/Transforms/Inline/align.ll
index 9ac6d54bfba9..c91fe8046174 100644
--- a/test/Transforms/Inline/align.ll
+++ b/test/Transforms/Inline/align.ll
@@ -4,8 +4,8 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @hello(float* align 128 nocapture %a, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx, align 4
   ret void
 }
@@ -13,8 +13,8 @@ entry:
 define void @foo(float* nocapture %a, float* nocapture readonly %c) #0 {
 entry:
   tail call void @hello(float* %a, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
@@ -25,11 +25,11 @@ entry:
 ; CHECK:   %maskedptr = and i64 %ptrint, 127
 ; CHECK:   %maskcond = icmp eq i64 %maskedptr, 0
 ; CHECK:   call void @llvm.assume(i1 %maskcond)
-; CHECK:   %0 = load float* %c, align 4
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
@@ -37,29 +37,29 @@ entry:
 define void @fooa(float* nocapture align 128 %a, float* nocapture readonly %c) #0 {
 entry:
   tail call void @hello(float* %a, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
 
 ; CHECK: define void @fooa(float* nocapture align 128 %a, float* nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
 
 define void @hello2(float* align 128 nocapture %a, float* align 128 nocapture %b, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds float* %b, i64 8
+  %arrayidx1 = getelementptr inbounds float, float* %b, i64 8
   store float %0, float* %arrayidx1, align 4
   ret void
 }
@@ -67,8 +67,8 @@ entry:
 define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 entry:
   tail call void @hello2(float* %a, float* %b, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
@@ -83,13 +83,13 @@ entry:
 ; CHECK:   %maskedptr2 = and i64 %ptrint1, 127
 ; CHECK:   %maskcond3 = icmp eq i64 %maskedptr2, 0
 ; CHECK:   call void @llvm.assume(i1 %maskcond3)
-; CHECK:   %0 = load float* %c, align 4
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4
-; CHECK:   %arrayidx1.i = getelementptr inbounds float* %b, i64 8
+; CHECK:   %arrayidx1.i = getelementptr inbounds float, float* %b, i64 8
 ; CHECK:   store float %0, float* %arrayidx1.i, align 4
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
diff --git a/test/Transforms/Inline/alloca-bonus.ll b/test/Transforms/Inline/alloca-bonus.ll
index 3c5052b883b2..649fac909fba 100644
--- a/test/Transforms/Inline/alloca-bonus.ll
+++ b/test/Transforms/Inline/alloca-bonus.ll
@@ -15,10 +15,10 @@ define void @outer1() {
 }
 
 define void @inner1(i32 *%ptr) {
-  %A = load i32* %ptr
+  %A = load i32, i32* %ptr
   store i32 0, i32* %ptr
-  %C = getelementptr inbounds i32* %ptr, i32 0
-  %D = getelementptr inbounds i32* %ptr, i32 1
+  %C = getelementptr inbounds i32, i32* %ptr, i32 0
+  %D = getelementptr inbounds i32, i32* %ptr, i32 1
   %E = bitcast i32* %ptr to i8*
   %F = select i1 false, i32* %ptr, i32* @glbl
   call void @llvm.lifetime.start(i64 0, i8* %E)
@@ -35,10 +35,10 @@ define void @outer2() {
 
 ; %D poisons this call, scalar-repl can't handle that instruction.
 define void @inner2(i32 *%ptr) {
-  %A = load i32* %ptr
+  %A = load i32, i32* %ptr
   store i32 0, i32* %ptr
-  %C = getelementptr inbounds i32* %ptr, i32 0
-  %D = getelementptr inbounds i32* %ptr, i32 %A
+  %C = getelementptr inbounds i32, i32* %ptr, i32 0
+  %D = getelementptr inbounds i32, i32* %ptr, i32 %A
   %E = bitcast i32* %ptr to i8*
   %F = select i1 false, i32* %ptr, i32* @glbl
   call void @llvm.lifetime.start(i64 0, i8* %E)
@@ -59,7 +59,7 @@ define void @inner3(i32 *%ptr, i1 %x) {
   br i1 %A, label %bb.true, label %bb.false
 bb.true:
   ; This block musn't be counted in the inline cost.
-  %t1 = load i32* %ptr
+  %t1 = load i32, i32* %ptr
   %t2 = add i32 %t1, 1
   %t3 = add i32 %t2, 1
   %t4 = add i32 %t3, 1
@@ -95,12 +95,12 @@ define void @outer4(i32 %A) {
 ; %B poisons this call, scalar-repl can't handle that instruction. However, we
 ; still want to detect that the icmp and branch *can* be handled.
 define void @inner4(i32 *%ptr, i32 %A) {
-  %B = getelementptr inbounds i32* %ptr, i32 %A
+  %B = getelementptr inbounds i32, i32* %ptr, i32 %A
   %C = icmp eq i32* %ptr, null
   br i1 %C, label %bb.true, label %bb.false
 bb.true:
   ; This block musn't be counted in the inline cost.
-  %t1 = load i32* %ptr
+  %t1 = load i32, i32* %ptr
   %t2 = add i32 %t1, 1
   %t3 = add i32 %t2, 1
   %t4 = add i32 %t3, 1
@@ -137,13 +137,13 @@ define void @outer5() {
 ; the flag is set appropriately, the poisoning instruction is inside of dead
 ; code, and so shouldn't be counted.
 define void @inner5(i1 %flag, i32 *%ptr) {
-  %A = load i32* %ptr
+  %A = load i32, i32* %ptr
   store i32 0, i32* %ptr
-  %C = getelementptr inbounds i32* %ptr, i32 0
+  %C = getelementptr inbounds i32, i32* %ptr, i32 0
   br i1 %flag, label %if.then, label %exit
 
 if.then:
-  %D = getelementptr inbounds i32* %ptr, i32 %A
+  %D = getelementptr inbounds i32, i32* %ptr, i32 %A
   %E = bitcast i32* %ptr to i8*
   %F = select i1 false, i32* %ptr, i32* @glbl
   call void @llvm.lifetime.start(i64 0, i8* %E)
diff --git a/test/Transforms/Inline/alloca-dbgdeclare.ll b/test/Transforms/Inline/alloca-dbgdeclare.ll
new file mode 100644
index 000000000000..286f2931ff22
--- /dev/null
+++ b/test/Transforms/Inline/alloca-dbgdeclare.ll
@@ -0,0 +1,141 @@
+; RUN: opt -inline -S < %s | FileCheck %s
+; struct A {
+;   int arg0;
+;   double arg1[2];
+; } a, b;
+;  
+; void fn3(A p1) {
+;   if (p1.arg0)
+;     a = p1;
+; }
+;  
+; void fn4() { fn3(b); }
+;  
+; void fn5() {
+;   while (1)
+;     fn4();
+; }
+; ModuleID = 'test.cpp'
+target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64-apple-darwin"
+
+%struct.A = type { i32, [2 x double] }
+
+@a = global %struct.A zeroinitializer, align 8
+@b = global %struct.A zeroinitializer, align 8
+
+; Function Attrs: nounwind
+declare void @_Z3fn31A(%struct.A* nocapture readonly %p1) #0
+
+; Function Attrs: nounwind readnone
+declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
+
+; Function Attrs: nounwind
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1) #2
+
+; Function Attrs: nounwind
+define void @_Z3fn4v() #0 {
+entry:
+; Test that the dbg.declare is moved together with the alloca.
+; CHECK: define void @_Z3fn5v()
+; CHECK-NEXT: entry:
+; CHECK-NEXT:   %agg.tmp.sroa.3.i = alloca [20 x i8], align 4
+; CHECK-NEXT:   call void @llvm.dbg.declare(metadata [20 x i8]* %agg.tmp.sroa.3.i,
+  %agg.tmp.sroa.3 = alloca [20 x i8], align 4
+  tail call void @llvm.dbg.declare(metadata [20 x i8]* %agg.tmp.sroa.3, metadata !46, metadata !48), !dbg !49
+  %agg.tmp.sroa.0.0.copyload = load i32, i32* getelementptr inbounds (%struct.A, %struct.A* @b, i64 0, i32 0), align 8, !dbg !50
+  tail call void @llvm.dbg.value(metadata i32 %agg.tmp.sroa.0.0.copyload, i64 0, metadata !46, metadata !51), !dbg !49
+  %agg.tmp.sroa.3.0..sroa_idx = getelementptr inbounds [20 x i8], [20 x i8]* %agg.tmp.sroa.3, i64 0, i64 0, !dbg !50
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %agg.tmp.sroa.3.0..sroa_idx, i8* getelementptr (i8, i8* bitcast (%struct.A* @b to i8*), i64 4), i64 20, i32 4, i1 false), !dbg !50
+  tail call void @llvm.dbg.declare(metadata %struct.A* undef, metadata !46, metadata !31) #2, !dbg !49
+  %tobool.i = icmp eq i32 %agg.tmp.sroa.0.0.copyload, 0, !dbg !52
+  br i1 %tobool.i, label %_Z3fn31A.exit, label %if.then.i, !dbg !53
+
+if.then.i:                                        ; preds = %entry
+  store i32 %agg.tmp.sroa.0.0.copyload, i32* getelementptr inbounds (%struct.A, %struct.A* @a, i64 0, i32 0), align 8, !dbg !54
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* getelementptr (i8, i8* bitcast (%struct.A* @a to i8*), i64 4), i8* %agg.tmp.sroa.3.0..sroa_idx, i64 20, i32 4, i1 false), !dbg !54
+  br label %_Z3fn31A.exit, !dbg !54
+
+_Z3fn31A.exit:                                    ; preds = %entry, %if.then.i
+  ret void, !dbg !50
+}
+
+; Function Attrs: noreturn nounwind
+define void @_Z3fn5v() #3 {
+entry:
+  br label %while.body, !dbg !55
+
+while.body:                                       ; preds = %entry, %while.body
+  call void @_Z3fn4v(), !dbg !56
+  br label %while.body, !dbg !55
+}
+
+; Function Attrs: nounwind readnone
+declare void @llvm.dbg.value(metadata, i64, metadata, metadata) #1
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readnone }
+attributes #2 = { nounwind }
+attributes #3 = { noreturn nounwind }
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!28, !29}
+!llvm.ident = !{!30}
+
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.7.0 (trunk 227480) (llvm/trunk 227517)", isOptimized: true, emissionKind: 1, file: !1, enums: !2, retainedTypes: !3, subprograms: !14, globals: !25, imports: !2)
+!1 = !DIFile(filename: "<stdin>", directory: "")
+!2 = !{}
+!3 = !{!4}
+!4 = !DICompositeType(tag: DW_TAG_structure_type, name: "A", line: 1, size: 192, align: 64, file: !5, elements: !6, identifier: "_ZTS1A")
+!5 = !DIFile(filename: "test.cpp", directory: "")
+!6 = !{!7, !9}
+!7 = !DIDerivedType(tag: DW_TAG_member, name: "arg0", line: 2, size: 32, align: 32, file: !5, scope: !"_ZTS1A", baseType: !8)
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!9 = !DIDerivedType(tag: DW_TAG_member, name: "arg1", line: 3, size: 128, align: 64, offset: 64, file: !5, scope: !"_ZTS1A", baseType: !10)
+!10 = !DICompositeType(tag: DW_TAG_array_type, size: 128, align: 64, baseType: !11, elements: !12)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
+!12 = !{!13}
+!13 = !DISubrange(count: 2)
+!14 = !{!15, !21, !24}
+!15 = !DISubprogram(name: "fn3", linkageName: "_Z3fn31A", line: 6, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 6, file: !5, scope: !16, type: !17, function: void (%struct.A*)* @_Z3fn31A, variables: !19)
+!16 = !DIFile(filename: "test.cpp", directory: "")
+!17 = !DISubroutineType(types: !18)
+!18 = !{null, !"_ZTS1A"}
+!19 = !{!20}
+!20 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "p1", line: 6, arg: 1, scope: !15, file: !16, type: !"_ZTS1A")
+!21 = !DISubprogram(name: "fn4", linkageName: "_Z3fn4v", line: 11, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 11, file: !5, scope: !16, type: !22, function: void ()* @_Z3fn4v, variables: !2)
+!22 = !DISubroutineType(types: !23)
+!23 = !{null}
+!24 = !DISubprogram(name: "fn5", linkageName: "_Z3fn5v", line: 13, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 13, file: !5, scope: !16, type: !22, function: void ()* @_Z3fn5v, variables: !2)
+!25 = !{!26, !27}
+!26 = !DIGlobalVariable(name: "a", line: 4, isLocal: false, isDefinition: true, scope: null, file: !16, type: !"_ZTS1A", variable: %struct.A* @a)
+!27 = !DIGlobalVariable(name: "b", line: 4, isLocal: false, isDefinition: true, scope: null, file: !16, type: !"_ZTS1A", variable: %struct.A* @b)
+!28 = !{i32 2, !"Dwarf Version", i32 4}
+!29 = !{i32 2, !"Debug Info Version", i32 3}
+!30 = !{!"clang version 3.7.0 (trunk 227480) (llvm/trunk 227517)"}
+!31 = !DIExpression(DW_OP_deref)
+!32 = !DILocation(line: 6, scope: !15)
+!33 = !DILocation(line: 7, scope: !34)
+!34 = distinct !DILexicalBlock(line: 7, column: 0, file: !5, scope: !15)
+!35 = !{!36, !37, i64 0}
+!36 = !{!"_ZTS1A", !37, i64 0, !38, i64 8}
+!37 = !{!"int", !38, i64 0}
+!38 = !{!"omnipotent char", !39, i64 0}
+!39 = !{!"Simple C/C++ TBAA"}
+!40 = !DILocation(line: 7, scope: !15)
+!41 = !DILocation(line: 8, scope: !34)
+!42 = !{i64 0, i64 4, !43, i64 8, i64 16, !44}
+!43 = !{!37, !37, i64 0}
+!44 = !{!38, !38, i64 0}
+!45 = !DILocation(line: 9, scope: !15)
+!46 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "p1", line: 6, arg: 1, scope: !15, file: !16, type: !"_ZTS1A")
+!47 = distinct !DILocation(line: 11, scope: !21)
+!48 = !DIExpression(DW_OP_bit_piece, 32, 160)
+!49 = !DILocation(line: 6, scope: !15, inlinedAt: !47)
+!50 = !DILocation(line: 11, scope: !21)
+!51 = !DIExpression(DW_OP_bit_piece, 0, 32)
+!52 = !DILocation(line: 7, scope: !34, inlinedAt: !47)
+!53 = !DILocation(line: 7, scope: !15, inlinedAt: !47)
+!54 = !DILocation(line: 8, scope: !34, inlinedAt: !47)
+!55 = !DILocation(line: 14, scope: !24)
+!56 = !DILocation(line: 15, scope: !24)
diff --git a/test/Transforms/Inline/alloca-in-scc.ll b/test/Transforms/Inline/alloca-in-scc.ll
index d539255e6e0a..92649a6e0734 100644
--- a/test/Transforms/Inline/alloca-in-scc.ll
+++ b/test/Transforms/Inline/alloca-in-scc.ll
@@ -12,7 +12,7 @@ entry:
 define internal fastcc void @a() nounwind ssp {
 entry:
   %al = alloca [3 x i32], align 4
-  %0 = getelementptr inbounds [3 x i32]* %al, i32 0, i32 2 
+  %0 = getelementptr inbounds [3 x i32], [3 x i32]* %al, i32 0, i32 2 
   
   call fastcc void @c() nounwind
   unreachable
diff --git a/test/Transforms/Inline/alloca-merge-align-nodl.ll b/test/Transforms/Inline/alloca-merge-align-nodl.ll
deleted file mode 100644
index 301505ff7910..000000000000
--- a/test/Transforms/Inline/alloca-merge-align-nodl.ll
+++ /dev/null
@@ -1,88 +0,0 @@
-; RUN: opt < %s -inline -S | FileCheck %s
-; This variant of the test has no data layout information.
-target triple = "powerpc64-unknown-linux-gnu"
-
-%struct.s = type { i32, i32 }
-
-define void @foo(%struct.s* byval nocapture readonly %a) {
-entry:
-  %x = alloca [2 x i32], align 4
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
-  store i32 %0, i32* %arrayidx, align 4
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
-  store i32 %1, i32* %arrayidx2, align 4
-  call void @bar(i32* %arrayidx) #2
-  ret void
-}
-
-define void @foo0(%struct.s* byval nocapture readonly %a) {
-entry:
-  %x = alloca [2 x i32]
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
-  store i32 %0, i32* %arrayidx, align 4
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
-  store i32 %1, i32* %arrayidx2, align 4
-  call void @bar(i32* %arrayidx) #2
-  ret void
-}
-
-declare void @bar(i32*) #1
-
-define void @goo(%struct.s* byval nocapture readonly %a) {
-entry:
-  %x = alloca [2 x i32], align 32
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
-  store i32 %0, i32* %arrayidx, align 32
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
-  store i32 %1, i32* %arrayidx2, align 4
-  call void @bar(i32* %arrayidx) #2
-  ret void
-}
-
-; CHECK-LABEL: @main
-; CHECK: alloca [2 x i32], align 32
-; CHECK-NOT: alloca [2 x i32]
-; CHECK: ret i32 0
-
-define signext i32 @main() {
-entry:
-  %a = alloca i64, align 8
-  %tmpcast = bitcast i64* %a to %struct.s*
-  store i64 0, i64* %a, align 8
-  %a1 = bitcast i64* %a to i32*
-  store i32 1, i32* %a1, align 8
-  call void @foo(%struct.s* byval %tmpcast)
-  store i32 2, i32* %a1, align 8
-  call void @goo(%struct.s* byval %tmpcast)
-  ret i32 0
-}
-
-; CHECK-LABEL: @test0
-; CHECK: alloca [2 x i32], align 32
-; CHECK: alloca [2 x i32]
-; CHECK: ret i32 0
-
-define signext i32 @test0() {
-entry:
-  %a = alloca i64, align 8
-  %tmpcast = bitcast i64* %a to %struct.s*
-  store i64 0, i64* %a, align 8
-  %a1 = bitcast i64* %a to i32*
-  store i32 1, i32* %a1, align 8
-  call void @foo0(%struct.s* byval %tmpcast)
-  store i32 2, i32* %a1, align 8
-  call void @goo(%struct.s* byval %tmpcast)
-  ret i32 0
-}
diff --git a/test/Transforms/Inline/alloca-merge-align.ll b/test/Transforms/Inline/alloca-merge-align.ll
index d357b3cd61a4..70b94f7b69a0 100644
--- a/test/Transforms/Inline/alloca-merge-align.ll
+++ b/test/Transforms/Inline/alloca-merge-align.ll
@@ -8,13 +8,13 @@ target triple = "powerpc64-unknown-linux-gnu"
 define void @foo(%struct.s* byval nocapture readonly %a) {
 entry:
   %x = alloca [2 x i32], align 4
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
+  %a1 = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 0
+  %0 = load i32, i32* %a1, align 4
+  %arrayidx = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 0
   store i32 %0, i32* %arrayidx, align 4
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
+  %b = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 1
+  %1 = load i32, i32* %b, align 4
+  %arrayidx2 = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 1
   store i32 %1, i32* %arrayidx2, align 4
   call void @bar(i32* %arrayidx) #2
   ret void
@@ -23,13 +23,13 @@ entry:
 define void @foo0(%struct.s* byval nocapture readonly %a) {
 entry:
   %x = alloca [2 x i32]
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
+  %a1 = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 0
+  %0 = load i32, i32* %a1, align 4
+  %arrayidx = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 0
   store i32 %0, i32* %arrayidx, align 4
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
+  %b = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 1
+  %1 = load i32, i32* %b, align 4
+  %arrayidx2 = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 1
   store i32 %1, i32* %arrayidx2, align 4
   call void @bar(i32* %arrayidx) #2
   ret void
@@ -38,13 +38,13 @@ entry:
 define void @foo1(%struct.s* byval nocapture readonly %a) {
 entry:
   %x = alloca [2 x i32], align 1
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
+  %a1 = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 0
+  %0 = load i32, i32* %a1, align 4
+  %arrayidx = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 0
   store i32 %0, i32* %arrayidx, align 4
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
+  %b = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 1
+  %1 = load i32, i32* %b, align 4
+  %arrayidx2 = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 1
   store i32 %1, i32* %arrayidx2, align 4
   call void @bar(i32* %arrayidx) #2
   ret void
@@ -55,13 +55,13 @@ declare void @bar(i32*) #1
 define void @goo(%struct.s* byval nocapture readonly %a) {
 entry:
   %x = alloca [2 x i32], align 32
-  %a1 = getelementptr inbounds %struct.s* %a, i64 0, i32 0
-  %0 = load i32* %a1, align 4
-  %arrayidx = getelementptr inbounds [2 x i32]* %x, i64 0, i64 0
+  %a1 = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 0
+  %0 = load i32, i32* %a1, align 4
+  %arrayidx = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 0
   store i32 %0, i32* %arrayidx, align 32
-  %b = getelementptr inbounds %struct.s* %a, i64 0, i32 1
-  %1 = load i32* %b, align 4
-  %arrayidx2 = getelementptr inbounds [2 x i32]* %x, i64 0, i64 1
+  %b = getelementptr inbounds %struct.s, %struct.s* %a, i64 0, i32 1
+  %1 = load i32, i32* %b, align 4
+  %arrayidx2 = getelementptr inbounds [2 x i32], [2 x i32]* %x, i64 0, i64 1
   store i32 %1, i32* %arrayidx2, align 4
   call void @bar(i32* %arrayidx) #2
   ret void
@@ -102,21 +102,3 @@ entry:
   call void @goo(%struct.s* byval %tmpcast)
   ret i32 0
 }
-
-; CHECK-LABEL: @test1
-; CHECK: {{alloca \[2 x i32\]$}}
-; CHECK-NOT: alloca [2 x i32]
-; CHECK: ret i32 0
-
-define signext i32 @test1() {
-entry:
-  %a = alloca i64, align 8
-  %tmpcast = bitcast i64* %a to %struct.s*
-  store i64 0, i64* %a, align 8
-  %a1 = bitcast i64* %a to i32*
-  store i32 1, i32* %a1, align 8
-  call void @foo0(%struct.s* byval %tmpcast)
-  store i32 2, i32* %a1, align 8
-  call void @foo1(%struct.s* byval %tmpcast)
-  ret i32 0
-}
diff --git a/test/Transforms/Inline/attributes.ll b/test/Transforms/Inline/attributes.ll
index 53fb13f2baf5..a97e6a60de75 100644
--- a/test/Transforms/Inline/attributes.ll
+++ b/test/Transforms/Inline/attributes.ll
@@ -110,3 +110,53 @@ define i32 @test_sanitize_thread(i32 %arg) sanitize_thread {
 ; CHECK-NEXT: @noattr_callee
 ; CHECK-NEXT: ret i32
 }
+
+; Check that a function doesn't get inlined if target-cpu strings don't match
+; exactly.
+define i32 @test_target_cpu_callee0(i32 %i) "target-cpu"="corei7" {
+  ret i32 %i
+}
+
+define i32 @test_target_cpu0(i32 %i) "target-cpu"="corei7" {
+  %1 = call i32 @test_target_cpu_callee0(i32 %i)
+  ret i32 %1
+; CHECK-LABEL: @test_target_cpu0(
+; CHECK-NOT: @test_target_cpu_callee0
+}
+
+define i32 @test_target_cpu_callee1(i32 %i) "target-cpu"="x86-64" {
+  ret i32 %i
+}
+
+define i32 @test_target_cpu1(i32 %i) "target-cpu"="corei7" {
+  %1 = call i32 @test_target_cpu_callee1(i32 %i)
+  ret i32 %1
+; CHECK-LABEL: @test_target_cpu1(
+; CHECK-NEXT: @test_target_cpu_callee1
+; CHECK-NEXT: ret i32
+}
+
+; Check that a function doesn't get inlined if target-features strings don't
+; match exactly.
+define i32 @test_target_features_callee0(i32 %i)  "target-features"="+sse4.2" {
+  ret i32 %i
+}
+
+define i32 @test_target_features0(i32 %i) "target-features"="+sse4.2" {
+  %1 = call i32 @test_target_features_callee0(i32 %i)
+  ret i32 %1
+; CHECK-LABEL: @test_target_features0(
+; CHECK-NOT: @test_target_features_callee0
+}
+
+define i32 @test_target_features_callee1(i32 %i) "target-features"="+avx2" {
+  ret i32 %i
+}
+
+define i32 @test_target_features1(i32 %i) "target-features"="+sse4.2" {
+  %1 = call i32 @test_target_features_callee1(i32 %i)
+  ret i32 %1
+; CHECK-LABEL: @test_target_features1(
+; CHECK-NEXT: @test_target_features_callee1
+; CHECK-NEXT: ret i32
+}
diff --git a/test/Transforms/Inline/basictest.ll b/test/Transforms/Inline/basictest.ll
index 085694febc6a..8f0b3eafaecb 100644
--- a/test/Transforms/Inline/basictest.ll
+++ b/test/Transforms/Inline/basictest.ll
@@ -25,7 +25,7 @@ define internal %T* @test2f(i1 %cond, %T* %P) {
   br i1 %cond, label %T, label %F
   
 T:
-  %A = getelementptr %T* %P, i32 0, i32 0
+  %A = getelementptr %T, %T* %P, i32 0, i32 0
   store i32 42, i32* %A
   ret %T* %P
   
@@ -37,8 +37,8 @@ define i32 @test2(i1 %cond) {
   %A = alloca %T
   
   %B = call %T* @test2f(i1 %cond, %T* %A)
-  %C = getelementptr %T* %B, i32 0, i32 0
-  %D = load i32* %C
+  %C = getelementptr %T, %T* %B, i32 0, i32 0
+  %D = load i32, i32* %C
   ret i32 %D
   
 ; CHECK-LABEL: @test2(
diff --git a/test/Transforms/Inline/byval-tail-call.ll b/test/Transforms/Inline/byval-tail-call.ll
index 154f3974b58d..1e50463ed80c 100644
--- a/test/Transforms/Inline/byval-tail-call.ll
+++ b/test/Transforms/Inline/byval-tail-call.ll
@@ -31,7 +31,7 @@ define internal void @qux(i32* byval %x) {
 define void @frob(i32* %x) {
 ; CHECK-LABEL: define void @frob(
 ; CHECK: %[[POS:.*]] = alloca i32
-; CHECK: %[[VAL:.*]] = load i32* %x
+; CHECK: %[[VAL:.*]] = load i32, i32* %x
 ; CHECK: store i32 %[[VAL]], i32* %[[POS]]
 ; CHECK: {{^ *}}call void @ext(i32* %[[POS]]
 ; CHECK: tail call void @ext(i32* null)
diff --git a/test/Transforms/Inline/byval.ll b/test/Transforms/Inline/byval.ll
index d7597adaf84f..d4edf2503cf6 100644
--- a/test/Transforms/Inline/byval.ll
+++ b/test/Transforms/Inline/byval.ll
@@ -7,8 +7,8 @@
 
 define internal void @f(%struct.ss* byval  %b) nounwind  {
 entry:
-	%tmp = getelementptr %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
-	%tmp1 = load i32* %tmp, align 4		; <i32> [#uses=1]
+	%tmp = getelementptr %struct.ss, %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
+	%tmp1 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
 	%tmp2 = add i32 %tmp1, 1		; <i32> [#uses=1]
 	store i32 %tmp2, i32* %tmp, align 4
 	ret void
@@ -19,9 +19,9 @@ declare i32 @printf(i8*, ...) nounwind
 define i32 @test1() nounwind  {
 entry:
 	%S = alloca %struct.ss		; <%struct.ss*> [#uses=4]
-	%tmp1 = getelementptr %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp1, align 8
-	%tmp4 = getelementptr %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
+	%tmp4 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
 	store i64 2, i64* %tmp4, align 4
 	call void @f( %struct.ss* byval  %S ) nounwind 
 	ret i32 0
@@ -37,8 +37,8 @@ entry:
 
 define internal i32 @f2(%struct.ss* byval  %b) nounwind readonly {
 entry:
-	%tmp = getelementptr %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
-	%tmp1 = load i32* %tmp, align 4		; <i32> [#uses=1]
+	%tmp = getelementptr %struct.ss, %struct.ss* %b, i32 0, i32 0		; <i32*> [#uses=2]
+	%tmp1 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
 	%tmp2 = add i32 %tmp1, 1		; <i32> [#uses=1]
 	ret i32 %tmp2
 }
@@ -46,9 +46,9 @@ entry:
 define i32 @test2() nounwind  {
 entry:
 	%S = alloca %struct.ss		; <%struct.ss*> [#uses=4]
-	%tmp1 = getelementptr %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp1 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp1, align 8
-	%tmp4 = getelementptr %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
+	%tmp4 = getelementptr %struct.ss, %struct.ss* %S, i32 0, i32 1		; <i64*> [#uses=1]
 	store i64 2, i64* %tmp4, align 4
 	%X = call i32 @f2( %struct.ss* byval  %S ) nounwind 
 	ret i32 %X
@@ -111,9 +111,9 @@ entry:
 
 define internal void @f5(%struct.S0* byval nocapture readonly align 4 %p) {
 entry:
-	store i32 0, i32* getelementptr inbounds (%struct.S0* @b, i64 0, i32 0), align 4
-	%f2 = getelementptr inbounds %struct.S0* %p, i64 0, i32 0
-	%0 = load i32* %f2, align 4
+	store i32 0, i32* getelementptr inbounds (%struct.S0, %struct.S0* @b, i64 0, i32 0), align 4
+	%f2 = getelementptr inbounds %struct.S0, %struct.S0* %p, i64 0, i32 0
+	%0 = load i32, i32* %f2, align 4
 	store i32 %0, i32* @a, align 4
 	ret void
 }
@@ -121,9 +121,9 @@ entry:
 define i32 @test5() {
 entry:
 	tail call void @f5(%struct.S0* byval align 4 @b)
-	%0 = load i32* @a, align 4
+	%0 = load i32, i32* @a, align 4
 	ret i32 %0
 ; CHECK: @test5()
-; CHECK: store i32 0, i32* getelementptr inbounds (%struct.S0* @b, i64 0, i32 0), align 4
-; CHECK-NOT: load i32* getelementptr inbounds (%struct.S0* @b, i64 0, i32 0), align 4
+; CHECK: store i32 0, i32* getelementptr inbounds (%struct.S0, %struct.S0* @b, i64 0, i32 0), align 4
+; CHECK-NOT: load i32, i32* getelementptr inbounds (%struct.S0, %struct.S0* @b, i64 0, i32 0), align 4
 }
diff --git a/test/Transforms/Inline/byval_lifetime.ll b/test/Transforms/Inline/byval_lifetime.ll
index e8dff2aa711d..b500eaea3317 100644
--- a/test/Transforms/Inline/byval_lifetime.ll
+++ b/test/Transforms/Inline/byval_lifetime.ll
@@ -10,9 +10,9 @@
 
 define i32 @foo(%struct.foo* byval align 8 %f, i32 %a) {
 entry:
-  %a1 = getelementptr inbounds %struct.foo* %f, i32 0, i32 1
-  %arrayidx = getelementptr inbounds [16 x i32]* %a1, i32 0, i32 %a
-  %tmp2 = load i32* %arrayidx, align 1
+  %a1 = getelementptr inbounds %struct.foo, %struct.foo* %f, i32 0, i32 1
+  %arrayidx = getelementptr inbounds [16 x i32], [16 x i32]* %a1, i32 0, i32 %a
+  %tmp2 = load i32, i32* %arrayidx, align 1
   ret i32 %tmp2
 }
 
diff --git a/test/Transforms/Inline/crash-lifetime-marker.ll b/test/Transforms/Inline/crash-lifetime-marker.ll
new file mode 100644
index 000000000000..42e926b1f63d
--- /dev/null
+++ b/test/Transforms/Inline/crash-lifetime-marker.ll
@@ -0,0 +1,24 @@
+; RUN: opt < %s -inline -S | FileCheck %s
+
+; InlineFunction would assert inside the loop that leaves lifetime markers if
+; there was an zero-sized AllocaInst. Check that it doesn't assert and doesn't
+; leave lifetime markers in that case.
+
+declare i32 @callee2(i8*)
+
+define i32 @callee1(i32 %count) {
+  %a0 = alloca i8, i32 %count, align 4
+  %call0 = call i32 @callee2(i8* %a0)
+  ret i32 %call0
+}
+
+; CHECK-LABEL: define i32 @caller1(
+; CHECK: [[ALLOCA:%[a-z0-9\.]+]] = alloca i8
+; CHECK-NOT: call void @llvm.lifetime.start(
+; CHECK: call i32 @callee2(i8* [[ALLOCA]])
+; CHECK-NOT: call void @llvm.lifetime.end(
+
+define i32 @caller1(i32 %count) {
+  %call0 = call i32 @callee1(i32 0)
+  ret i32 %call0
+}
diff --git a/test/Transforms/Inline/crash2.ll b/test/Transforms/Inline/crash2.ll
index be634f625633..4c0dfaea0364 100644
--- a/test/Transforms/Inline/crash2.ll
+++ b/test/Transforms/Inline/crash2.ll
@@ -14,7 +14,7 @@ entry:
   %__f_addr = alloca void (i8*)*, align 8
   store void (i8*)* %__f, void (i8*)** %__f_addr
 
-  %0 = load void (i8*)** %__f_addr, align 8
+  %0 = load void (i8*)*, void (i8*)** %__f_addr, align 8
   call void %0(i8* undef)
   call i8* @f1(i8* undef) ssp
   unreachable
diff --git a/test/Transforms/Inline/debug-info-duplicate-calls.ll b/test/Transforms/Inline/debug-info-duplicate-calls.ll
new file mode 100644
index 000000000000..4e3d9d92fdb3
--- /dev/null
+++ b/test/Transforms/Inline/debug-info-duplicate-calls.ll
@@ -0,0 +1,121 @@
+; RUN: opt < %s -always-inline -S | FileCheck %s
+
+; Original input generated from clang -emit-llvm -S -c -mllvm -disable-llvm-optzns
+;
+; #define CALLS1 f2(); f2();
+; #define CALLS2 f4(); f4();
+; void f1();
+; inline __attribute__((always_inline)) void f2() {
+;   f1();
+; }
+; inline __attribute__((always_inline)) void f3() {
+;   CALLS1
+; }
+; inline __attribute__((always_inline)) void f4() {
+;   f3();
+; }
+; void f() {
+;   CALLS2
+; }
+
+; There should be unique locations for all 4 of these instructions, correctly
+; describing the inlining that has occurred, even in the face of duplicate call
+; site locations.
+
+; The nomenclature used for the tags here is <function name>[cs<number>] where
+; 'cs' is an abbreviation for 'call site' and the number indicates which call
+; site from within the named function this is. (so, given the above inlining, we
+; should have 4 calls to 'f1', two from the first call to f4 and two from the
+; second call to f4)
+
+; CHECK: call void @_Z2f1v(), !dbg [[fcs1_f4_f3cs1_f2:![0-9]+]]
+; CHECK: call void @_Z2f1v(), !dbg [[fcs1_f4_f3cs2_f2:![0-9]+]]
+; CHECK: call void @_Z2f1v(), !dbg [[fcs2_f4_f3cs1_f2:![0-9]+]]
+; CHECK: call void @_Z2f1v(), !dbg [[fcs2_f4_f3cs2_f2:![0-9]+]]
+
+; CHECK-DAG: [[F:![0-9]+]]  = !DISubprogram(name: "f"
+; CHECK-DAG: [[F2:![0-9]+]] = !DISubprogram(name: "f2"
+; CHECK-DAG: [[F3:![0-9]+]] = !DISubprogram(name: "f3"
+; CHECK-DAG: [[F4:![0-9]+]] = !DISubprogram(name: "f4"
+
+; CHECK: [[fcs1_f4_f3cs1_f2]] = {{.*}}, scope: [[F2]], inlinedAt: [[fcs1_f4_f3cs1:![0-9]+]])
+; CHECK: [[fcs1_f4_f3cs1]] = {{.*}}, scope: [[F3]], inlinedAt: [[fcs1_f4:![0-9]+]])
+; CHECK: [[fcs1_f4]] = {{.*}}, scope: [[F4]], inlinedAt: [[fcs1:![0-9]+]])
+; CHECK: [[fcs1]] = {{.*}}, scope: [[F]])
+; CHECK: [[fcs1_f4_f3cs2_f2]] = {{.*}}, scope: [[F2]], inlinedAt: [[fcs1_f4_f3cs2:![0-9]+]])
+; CHECK: [[fcs1_f4_f3cs2]] = {{.*}}, scope: [[F3]], inlinedAt: [[fcs1_f4]])
+
+; CHECK: [[fcs2_f4_f3cs1_f2]] = {{.*}}, scope: [[F2]], inlinedAt: [[fcs2_f4_f3cs1:![0-9]+]])
+; CHECK: [[fcs2_f4_f3cs1]] = {{.*}}, scope: [[F3]], inlinedAt: [[fcs2_f4:![0-9]+]])
+; CHECK: [[fcs2_f4]] = {{.*}}, scope: [[F4]], inlinedAt: [[fcs2:![0-9]+]])
+; CHECK: [[fcs2]] = {{.*}}, scope: [[F]])
+; CHECK: [[fcs2_f4_f3cs2_f2]] = {{.*}}, scope: [[F2]], inlinedAt: [[fcs2_f4_f3cs2:![0-9]+]])
+; CHECK: [[fcs2_f4_f3cs2]] = {{.*}}, scope: [[F3]], inlinedAt: [[fcs2_f4]])
+
+$_Z2f4v = comdat any
+
+$_Z2f3v = comdat any
+
+$_Z2f2v = comdat any
+
+; Function Attrs: uwtable
+define void @_Z1fv() #0 {
+entry:
+  call void @_Z2f4v(), !dbg !13
+  call void @_Z2f4v(), !dbg !13
+  ret void, !dbg !14
+}
+
+; Function Attrs: alwaysinline inlinehint uwtable
+define linkonce_odr void @_Z2f4v() #1 comdat {
+entry:
+  call void @_Z2f3v(), !dbg !15
+  ret void, !dbg !16
+}
+
+; Function Attrs: alwaysinline inlinehint uwtable
+define linkonce_odr void @_Z2f3v() #1 comdat {
+entry:
+  call void @_Z2f2v(), !dbg !17
+  call void @_Z2f2v(), !dbg !17
+  ret void, !dbg !18
+}
+
+; Function Attrs: alwaysinline inlinehint uwtable
+define linkonce_odr void @_Z2f2v() #1 comdat {
+entry:
+  call void @_Z2f1v(), !dbg !19
+  ret void, !dbg !20
+}
+
+declare void @_Z2f1v() #2
+
+attributes #0 = { uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #1 = { alwaysinline inlinehint uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+attributes #2 = { "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!10, !11}
+!llvm.ident = !{!12}
+
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.7.0 (trunk 226474) (llvm/trunk 226478)", isOptimized: false, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "debug-info-duplicate-calls.cpp", directory: "/tmp/dbginfo")
+!2 = !{}
+!3 = !{!4, !7, !8, !9}
+!4 = !DISubprogram(name: "f", line: 13, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 13, file: !1, scope: !5, type: !6, function: void ()* @_Z1fv, variables: !2)
+!5 = !DIFile(filename: "debug-info-duplicate-calls.cpp", directory: "/tmp/dbginfo")
+!6 = !DISubroutineType(types: !2)
+!7 = !DISubprogram(name: "f4", line: 10, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 10, file: !1, scope: !5, type: !6, function: void ()* @_Z2f4v, variables: !2)
+!8 = !DISubprogram(name: "f3", line: 7, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 7, file: !1, scope: !5, type: !6, function: void ()* @_Z2f3v, variables: !2)
+!9 = !DISubprogram(name: "f2", line: 4, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 4, file: !1, scope: !5, type: !6, function: void ()* @_Z2f2v, variables: !2)
+!10 = !{i32 2, !"Dwarf Version", i32 4}
+!11 = !{i32 2, !"Debug Info Version", i32 3}
+!12 = !{!"clang version 3.7.0 (trunk 226474) (llvm/trunk 226478)"}
+!13 = !DILocation(line: 14, column: 3, scope: !4)
+!14 = !DILocation(line: 15, column: 1, scope: !4)
+!15 = !DILocation(line: 11, column: 3, scope: !7)
+!16 = !DILocation(line: 12, column: 1, scope: !7)
+!17 = !DILocation(line: 8, column: 3, scope: !8)
+!18 = !DILocation(line: 9, column: 1, scope: !8)
+!19 = !DILocation(line: 5, column: 3, scope: !9)
+!20 = !DILocation(line: 6, column: 1, scope: !9)
diff --git a/test/Transforms/Inline/debug-invoke.ll b/test/Transforms/Inline/debug-invoke.ll
index 4815e689b8a5..bb40091014cf 100644
--- a/test/Transforms/Inline/debug-invoke.ll
+++ b/test/Transforms/Inline/debug-invoke.ll
@@ -4,9 +4,9 @@
 
 ; CHECK: invoke void @test()
 ; CHECK-NEXT: to label {{.*}} unwind label {{.*}}, !dbg [[INL_LOC:!.*]]
-; CHECK: [[EMPTY:.*]] = !{}
-; CHECK: [[INL_LOC]] = !MDLocation(line: 1, scope: [[EMPTY]], inlinedAt: [[INL_AT:.*]])
-; CHECK: [[INL_AT]] = !MDLocation(line: 2, scope: [[EMPTY]])
+; CHECK: [[SP:.*]] = !DISubprogram(
+; CHECK: [[INL_LOC]] = !DILocation(line: 1, scope: [[SP]], inlinedAt: [[INL_AT:.*]])
+; CHECK: [[INL_AT]] = distinct !DILocation(line: 2, scope: [[SP]])
 
 declare void @test()
 declare i32 @__gxx_personality_v0(...)
@@ -31,7 +31,7 @@ lpad:
 }
 
 !llvm.module.flags = !{!1}
-!1 = !{i32 2, !"Debug Info Version", i32 2}
-!2 = !{}
-!3 = !MDLocation(line: 1, scope: !2)
-!4 = !MDLocation(line: 2, scope: !2)
+!1 = !{i32 2, !"Debug Info Version", i32 3}
+!2 = !DISubprogram()
+!3 = !DILocation(line: 1, scope: !2)
+!4 = !DILocation(line: 2, scope: !2)
diff --git a/test/Transforms/Inline/devirtualize-2.ll b/test/Transforms/Inline/devirtualize-2.ll
index b7eb1be19db2..cca4e7582b5f 100644
--- a/test/Transforms/Inline/devirtualize-2.ll
+++ b/test/Transforms/Inline/devirtualize-2.ll
@@ -2,7 +2,7 @@
 ; PR4834
 
 define i32 @test1() {
-  %funcall1_ = call fastcc i32 ()* ()* @f1()
+  %funcall1_ = call fastcc i32 ()* () @f1()
   %executecommandptr1_ = call i32 %funcall1_()
   ret i32 %executecommandptr1_
 }
diff --git a/test/Transforms/Inline/devirtualize-3.ll b/test/Transforms/Inline/devirtualize-3.ll
index 3f019676e4a4..76c8150de01f 100644
--- a/test/Transforms/Inline/devirtualize-3.ll
+++ b/test/Transforms/Inline/devirtualize-3.ll
@@ -14,12 +14,12 @@ target triple = "x86_64-apple-darwin10.0.0"
 define i32 @main() nounwind ssp {
 entry:
   %cont = alloca %struct.cont_t, align 8          ; <%struct.cont_t*> [#uses=4]
-  %tmp = getelementptr inbounds %struct.cont_t* %cont, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
-  %tmp1 = getelementptr inbounds %struct.cont_t* %cont, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=2]
+  %tmp = getelementptr inbounds %struct.cont_t, %struct.cont_t* %cont, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
+  %tmp1 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %cont, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=2]
   store void (i8*, i32)* bitcast (void (%struct.cont_t*, i32)* @quit to void (i8*, i32)*), void (i8*, i32)** %tmp1
-  %tmp2 = load void (i8*, i32)** %tmp1            ; <void (i8*, i32)*> [#uses=1]
+  %tmp2 = load void (i8*, i32)*, void (i8*, i32)** %tmp1            ; <void (i8*, i32)*> [#uses=1]
   store void (i8*, i32)* %tmp2, void (i8*, i32)** %tmp
-  %tmp3 = getelementptr inbounds %struct.cont_t* %cont, i32 0, i32 1 ; <i8**> [#uses=1]
+  %tmp3 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %cont, i32 0, i32 1 ; <i8**> [#uses=1]
   store i8* null, i8** %tmp3
   call void @foo(%struct.cont_t* %cont)
   ret i32 0
@@ -35,13 +35,13 @@ define internal void @foo(%struct.cont_t* %c) nounwind ssp {
 entry:
   %sf = alloca %struct.foo_sf_t, align 8          ; <%struct.foo_sf_t*> [#uses=3]
   %next = alloca %struct.cont_t, align 8          ; <%struct.cont_t*> [#uses=3]
-  %tmp = getelementptr inbounds %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
+  %tmp = getelementptr inbounds %struct.foo_sf_t, %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
   store %struct.cont_t* %c, %struct.cont_t** %tmp
-  %tmp2 = getelementptr inbounds %struct.foo_sf_t* %sf, i32 0, i32 1 ; <i32*> [#uses=1]
+  %tmp2 = getelementptr inbounds %struct.foo_sf_t, %struct.foo_sf_t* %sf, i32 0, i32 1 ; <i32*> [#uses=1]
   store i32 2, i32* %tmp2
-  %tmp4 = getelementptr inbounds %struct.cont_t* %next, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
+  %tmp4 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %next, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
   store void (i8*, i32)* bitcast (void (%struct.foo_sf_t*, i32)* @foo2 to void (i8*, i32)*), void (i8*, i32)** %tmp4
-  %tmp5 = getelementptr inbounds %struct.cont_t* %next, i32 0, i32 1 ; <i8**> [#uses=1]
+  %tmp5 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %next, i32 0, i32 1 ; <i8**> [#uses=1]
   %conv = bitcast %struct.foo_sf_t* %sf to i8*    ; <i8*> [#uses=1]
   store i8* %conv, i8** %tmp5
   call void @bar(%struct.cont_t* %next, i32 14)
@@ -50,15 +50,15 @@ entry:
 
 define internal void @foo2(%struct.foo_sf_t* %sf, i32 %y) nounwind ssp {
 entry:
-  %tmp1 = getelementptr inbounds %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
-  %tmp2 = load %struct.cont_t** %tmp1             ; <%struct.cont_t*> [#uses=1]
-  %tmp3 = getelementptr inbounds %struct.cont_t* %tmp2, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
-  %tmp4 = load void (i8*, i32)** %tmp3            ; <void (i8*, i32)*> [#uses=1]
-  %tmp6 = getelementptr inbounds %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
-  %tmp7 = load %struct.cont_t** %tmp6             ; <%struct.cont_t*> [#uses=1]
+  %tmp1 = getelementptr inbounds %struct.foo_sf_t, %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
+  %tmp2 = load %struct.cont_t*, %struct.cont_t** %tmp1             ; <%struct.cont_t*> [#uses=1]
+  %tmp3 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %tmp2, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
+  %tmp4 = load void (i8*, i32)*, void (i8*, i32)** %tmp3            ; <void (i8*, i32)*> [#uses=1]
+  %tmp6 = getelementptr inbounds %struct.foo_sf_t, %struct.foo_sf_t* %sf, i32 0, i32 0 ; <%struct.cont_t**> [#uses=1]
+  %tmp7 = load %struct.cont_t*, %struct.cont_t** %tmp6             ; <%struct.cont_t*> [#uses=1]
   %conv = bitcast %struct.cont_t* %tmp7 to i8*    ; <i8*> [#uses=1]
-  %tmp9 = getelementptr inbounds %struct.foo_sf_t* %sf, i32 0, i32 1 ; <i32*> [#uses=1]
-  %tmp10 = load i32* %tmp9                        ; <i32> [#uses=1]
+  %tmp9 = getelementptr inbounds %struct.foo_sf_t, %struct.foo_sf_t* %sf, i32 0, i32 1 ; <i32*> [#uses=1]
+  %tmp10 = load i32, i32* %tmp9                        ; <i32> [#uses=1]
   %mul = mul i32 %tmp10, %y                       ; <i32> [#uses=1]
   call void %tmp4(i8* %conv, i32 %mul)
   ret void
@@ -66,10 +66,10 @@ entry:
 
 define internal void @bar(%struct.cont_t* %c, i32 %y) nounwind ssp {
 entry:
-  %tmp1 = getelementptr inbounds %struct.cont_t* %c, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
-  %tmp2 = load void (i8*, i32)** %tmp1            ; <void (i8*, i32)*> [#uses=1]
-  %tmp4 = getelementptr inbounds %struct.cont_t* %c, i32 0, i32 1 ; <i8**> [#uses=1]
-  %tmp5 = load i8** %tmp4                         ; <i8*> [#uses=1]
+  %tmp1 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %c, i32 0, i32 0 ; <void (i8*, i32)**> [#uses=1]
+  %tmp2 = load void (i8*, i32)*, void (i8*, i32)** %tmp1            ; <void (i8*, i32)*> [#uses=1]
+  %tmp4 = getelementptr inbounds %struct.cont_t, %struct.cont_t* %c, i32 0, i32 1 ; <i8**> [#uses=1]
+  %tmp5 = load i8*, i8** %tmp4                         ; <i8*> [#uses=1]
   %add = add nsw i32 %y, 5                        ; <i32> [#uses=1]
   call void %tmp2(i8* %tmp5, i32 %add)
   ret void
diff --git a/test/Transforms/Inline/devirtualize.ll b/test/Transforms/Inline/devirtualize.ll
index 89482941267c..561bb62ae644 100644
--- a/test/Transforms/Inline/devirtualize.ll
+++ b/test/Transforms/Inline/devirtualize.ll
@@ -9,7 +9,7 @@ define i32 @foo(i32 ()** noalias %p, i64* noalias %q) nounwind ssp {
 entry:
   store i32 ()* @bar, i32 ()** %p
   store i64 0, i64* %q
-  %tmp3 = load i32 ()** %p                        ; <i32 ()*> [#uses=1]
+  %tmp3 = load i32 ()*, i32 ()** %p                        ; <i32 ()*> [#uses=1]
   %call = call i32 %tmp3()                        ; <i32> [#uses=1]
   %X = add i32 %call, 4
   ret i32 %X
@@ -47,11 +47,11 @@ entry:
 @_ZTS1C = linkonce_odr constant [3 x i8] c"1C\00"     ; <[3 x i8]*> [#uses=1]
 @_ZTVN10__cxxabiv117__class_type_infoE = external global i8* ; <i8**> [#uses=1]
 @_ZTS1A = linkonce_odr constant [3 x i8] c"1A\00"     ; <[3 x i8]*> [#uses=1]
-@_ZTI1A = linkonce_odr constant %0 { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8]* @_ZTS1A, i32 0, i32 0) } ; <%0*> [#uses=1]
+@_ZTI1A = linkonce_odr constant %0 { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @_ZTS1A, i32 0, i32 0) } ; <%0*> [#uses=1]
 @_ZTS1B = linkonce_odr constant [3 x i8] c"1B\00"     ; <[3 x i8]*> [#uses=1]
-@_ZTI1B = linkonce_odr constant %0 { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8]* @_ZTS1B, i32 0, i32 0) } ; <%0*> [#uses=1]
-@_ZTI1C = linkonce_odr constant %1 { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv121__vmi_class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8]* @_ZTS1C, i32 0, i32 0), i32 0, i32 2, i8* bitcast (%0* @_ZTI1A to i8*), i64 2, i8* bitcast (%0* @_ZTI1B to i8*), i64 2050 } ; <%1*> [#uses=1]
-@_ZTI1D = linkonce_odr constant %2 { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv120__si_class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8]* @_ZTS1D, i32 0, i32 0), i8* bitcast (%1* @_ZTI1C to i8*) } ; <%2*> [#uses=1]
+@_ZTI1B = linkonce_odr constant %0 { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @_ZTS1B, i32 0, i32 0) } ; <%0*> [#uses=1]
+@_ZTI1C = linkonce_odr constant %1 { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv121__vmi_class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @_ZTS1C, i32 0, i32 0), i32 0, i32 2, i8* bitcast (%0* @_ZTI1A to i8*), i64 2, i8* bitcast (%0* @_ZTI1B to i8*), i64 2050 } ; <%1*> [#uses=1]
+@_ZTI1D = linkonce_odr constant %2 { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv120__si_class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @_ZTS1D, i32 0, i32 0), i8* bitcast (%1* @_ZTI1C to i8*) } ; <%2*> [#uses=1]
 @_ZTV1C = linkonce_odr constant [6 x i8*] [i8* null, i8* bitcast (%1* @_ZTI1C to i8*), i8* bitcast (i32 (%struct.C*)* @_ZN1C1fEv to i8*), i8* inttoptr (i64 -8 to i8*), i8* bitcast (%1* @_ZTI1C to i8*), i8* bitcast (i32 (%struct.C*)* @_ZThn8_N1C1fEv to i8*)] ; <[6 x i8*]*> [#uses=2]
 @_ZTV1B = linkonce_odr constant [3 x i8*] [i8* null, i8* bitcast (%0* @_ZTI1B to i8*), i8* bitcast (i32 (%struct.A*)* @_ZN1B1fEv to i8*)] ; <[3 x i8*]*> [#uses=1]
 @_ZTV1A = linkonce_odr constant [3 x i8*] [i8* null, i8* bitcast (%0* @_ZTI1A to i8*), i8* bitcast (i32 (%struct.A*)* @_ZN1A1fEv to i8*)] ; <[3 x i8*]*> [#uses=1]
@@ -78,16 +78,16 @@ entry:
 
 cast.notnull:                                     ; preds = %entry
   %1 = bitcast %struct.C* %d to i8*               ; <i8*> [#uses=1]
-  %add.ptr = getelementptr i8* %1, i64 8          ; <i8*> [#uses=1]
+  %add.ptr = getelementptr i8, i8* %1, i64 8          ; <i8*> [#uses=1]
   %2 = bitcast i8* %add.ptr to %struct.A*         ; <%struct.A*> [#uses=1]
   br label %cast.end
 
 cast.end:                                         ; preds = %entry, %cast.notnull
   %3 = phi %struct.A* [ %2, %cast.notnull ], [ null, %entry ] ; <%struct.A*> [#uses=2]
   %4 = bitcast %struct.A* %3 to i32 (%struct.A*)*** ; <i32 (%struct.A*)***> [#uses=1]
-  %5 = load i32 (%struct.A*)*** %4                ; <i32 (%struct.A*)**> [#uses=1]
-  %vfn = getelementptr inbounds i32 (%struct.A*)** %5, i64 0 ; <i32 (%struct.A*)**> [#uses=1]
-  %6 = load i32 (%struct.A*)** %vfn               ; <i32 (%struct.A*)*> [#uses=1]
+  %5 = load i32 (%struct.A*)**, i32 (%struct.A*)*** %4                ; <i32 (%struct.A*)**> [#uses=1]
+  %vfn = getelementptr inbounds i32 (%struct.A*)*, i32 (%struct.A*)** %5, i64 0 ; <i32 (%struct.A*)**> [#uses=1]
+  %6 = load i32 (%struct.A*)*, i32 (%struct.A*)** %vfn               ; <i32 (%struct.A*)*> [#uses=1]
   %call = call i32 %6(%struct.A* %3)              ; <i32> [#uses=1]
   ret i32 %call
 }
@@ -100,7 +100,7 @@ entry:
 define linkonce_odr i32 @_ZThn8_N1D1fEv(%struct.C* %this) {
 entry:
   %0 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %1 = getelementptr inbounds i8* %0, i64 -8      ; <i8*> [#uses=1]
+  %1 = getelementptr inbounds i8, i8* %0, i64 -8      ; <i8*> [#uses=1]
   %2 = bitcast i8* %1 to %struct.C*               ; <%struct.C*> [#uses=1]
   %call = call i32 @_ZN1D1fEv(%struct.C* %2)      ; <i32> [#uses=1]
   ret i32 %call
@@ -110,13 +110,13 @@ define linkonce_odr void @_ZN1DC2Ev(%struct.C* %this) inlinehint ssp align 2 {
 entry:
   call void @_ZN1CC2Ev(%struct.C* %this)
   %0 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %1 = getelementptr inbounds i8* %0, i64 0       ; <i8*> [#uses=1]
+  %1 = getelementptr inbounds i8, i8* %0, i64 0       ; <i8*> [#uses=1]
   %2 = bitcast i8* %1 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([6 x i8*]* @_ZTV1D, i64 0, i64 2), i8*** %2
+  store i8** getelementptr inbounds ([6 x i8*], [6 x i8*]* @_ZTV1D, i64 0, i64 2), i8*** %2
   %3 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %4 = getelementptr inbounds i8* %3, i64 8       ; <i8*> [#uses=1]
+  %4 = getelementptr inbounds i8, i8* %3, i64 8       ; <i8*> [#uses=1]
   %5 = bitcast i8* %4 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([6 x i8*]* @_ZTV1D, i64 0, i64 5), i8*** %5
+  store i8** getelementptr inbounds ([6 x i8*], [6 x i8*]* @_ZTV1D, i64 0, i64 5), i8*** %5
   ret void
 }
 
@@ -125,17 +125,17 @@ entry:
   %0 = bitcast %struct.C* %this to %struct.A*     ; <%struct.A*> [#uses=1]
   call void @_ZN1AC2Ev(%struct.A* %0)
   %1 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %2 = getelementptr inbounds i8* %1, i64 8       ; <i8*> [#uses=1]
+  %2 = getelementptr inbounds i8, i8* %1, i64 8       ; <i8*> [#uses=1]
   %3 = bitcast i8* %2 to %struct.A*               ; <%struct.A*> [#uses=1]
   call void @_ZN1BC2Ev(%struct.A* %3)
   %4 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %5 = getelementptr inbounds i8* %4, i64 0       ; <i8*> [#uses=1]
+  %5 = getelementptr inbounds i8, i8* %4, i64 0       ; <i8*> [#uses=1]
   %6 = bitcast i8* %5 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([6 x i8*]* @_ZTV1C, i64 0, i64 2), i8*** %6
+  store i8** getelementptr inbounds ([6 x i8*], [6 x i8*]* @_ZTV1C, i64 0, i64 2), i8*** %6
   %7 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %8 = getelementptr inbounds i8* %7, i64 8       ; <i8*> [#uses=1]
+  %8 = getelementptr inbounds i8, i8* %7, i64 8       ; <i8*> [#uses=1]
   %9 = bitcast i8* %8 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([6 x i8*]* @_ZTV1C, i64 0, i64 5), i8*** %9
+  store i8** getelementptr inbounds ([6 x i8*], [6 x i8*]* @_ZTV1C, i64 0, i64 5), i8*** %9
   ret void
 }
 
@@ -147,7 +147,7 @@ entry:
 define linkonce_odr i32 @_ZThn8_N1C1fEv(%struct.C* %this) {
 entry:
   %0 = bitcast %struct.C* %this to i8*            ; <i8*> [#uses=1]
-  %1 = getelementptr inbounds i8* %0, i64 -8      ; <i8*> [#uses=1]
+  %1 = getelementptr inbounds i8, i8* %0, i64 -8      ; <i8*> [#uses=1]
   %2 = bitcast i8* %1 to %struct.C*               ; <%struct.C*> [#uses=1]
   %call = call i32 @_ZN1C1fEv(%struct.C* %2)      ; <i32> [#uses=1]
   ret i32 %call
@@ -156,18 +156,18 @@ entry:
 define linkonce_odr void @_ZN1AC2Ev(%struct.A* %this) inlinehint ssp align 2 {
 entry:
   %0 = bitcast %struct.A* %this to i8*            ; <i8*> [#uses=1]
-  %1 = getelementptr inbounds i8* %0, i64 0       ; <i8*> [#uses=1]
+  %1 = getelementptr inbounds i8, i8* %0, i64 0       ; <i8*> [#uses=1]
   %2 = bitcast i8* %1 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([3 x i8*]* @_ZTV1A, i64 0, i64 2), i8*** %2
+  store i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @_ZTV1A, i64 0, i64 2), i8*** %2
   ret void
 }
 
 define linkonce_odr void @_ZN1BC2Ev(%struct.A* %this) inlinehint ssp align 2 {
 entry:
   %0 = bitcast %struct.A* %this to i8*            ; <i8*> [#uses=1]
-  %1 = getelementptr inbounds i8* %0, i64 0       ; <i8*> [#uses=1]
+  %1 = getelementptr inbounds i8, i8* %0, i64 0       ; <i8*> [#uses=1]
   %2 = bitcast i8* %1 to i8***                    ; <i8***> [#uses=1]
-  store i8** getelementptr inbounds ([3 x i8*]* @_ZTV1B, i64 0, i64 2), i8*** %2
+  store i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @_ZTV1B, i64 0, i64 2), i8*** %2
   ret void
 }
 
diff --git a/test/Transforms/Inline/ephemeral.ll b/test/Transforms/Inline/ephemeral.ll
index d1135c6f0c3e..7e5ae034af35 100644
--- a/test/Transforms/Inline/ephemeral.ll
+++ b/test/Transforms/Inline/ephemeral.ll
@@ -3,7 +3,7 @@
 @a = global i32 4
 
 define i1 @inner() {
-  %a1 = load volatile i32* @a
+  %a1 = load volatile i32, i32* @a
   %x1 = add i32 %a1, %a1
   %c = icmp eq i32 %x1, 0
 
diff --git a/test/Transforms/Inline/frameescape.ll b/test/Transforms/Inline/frameescape.ll
new file mode 100644
index 000000000000..fb336024f937
--- /dev/null
+++ b/test/Transforms/Inline/frameescape.ll
@@ -0,0 +1,44 @@
+; RUN: opt -inline -S < %s | FileCheck %s
+
+; PR23216: We can't inline functions using llvm.frameescape.
+
+declare void @llvm.frameescape(...)
+declare i8* @llvm.frameaddress(i32)
+declare i8* @llvm.framerecover(i8*, i8*, i32)
+
+define internal void @foo(i8* %fp) {
+  %a.i8 = call i8* @llvm.framerecover(i8* bitcast (i32 ()* @bar to i8*), i8* %fp, i32 0)
+  %a = bitcast i8* %a.i8 to i32*
+  store i32 42, i32* %a
+  ret void
+}
+
+define internal i32 @bar() {
+entry:
+  %a = alloca i32
+  call void (...) @llvm.frameescape(i32* %a)
+  %fp = call i8* @llvm.frameaddress(i32 0)
+  tail call void @foo(i8* %fp)
+  %r = load i32, i32* %a
+  ret i32 %r
+}
+
+; We even bail when someone marks it alwaysinline.
+define internal i32 @bar_alwaysinline() alwaysinline {
+entry:
+  %a = alloca i32
+  call void (...) @llvm.frameescape(i32* %a)
+  tail call void @foo(i8* null)
+  ret i32 0
+}
+
+define i32 @bazz() {
+entry:
+  %r = tail call i32 @bar()
+  %r1 = tail call i32 @bar_alwaysinline()
+  ret i32 %r
+}
+
+; CHECK: define i32 @bazz()
+; CHECK: call i32 @bar()
+; CHECK: call i32 @bar_alwaysinline()
diff --git a/test/Transforms/Inline/gvn-inline-iteration.ll b/test/Transforms/Inline/gvn-inline-iteration.ll
index 7916a13df995..b87c0609ea2b 100644
--- a/test/Transforms/Inline/gvn-inline-iteration.ll
+++ b/test/Transforms/Inline/gvn-inline-iteration.ll
@@ -8,7 +8,7 @@ define i32 @foo(i32 ()** noalias nocapture %p, i64* noalias nocapture %q) nounwi
 entry:
   store i32 ()* @bar, i32 ()** %p
   store i64 0, i64* %q
-  %tmp3 = load i32 ()** %p                        ; <i32 ()*> [#uses=1]
+  %tmp3 = load i32 ()*, i32 ()** %p                        ; <i32 ()*> [#uses=1]
   %call = tail call i32 %tmp3() nounwind          ; <i32> [#uses=1]
   ret i32 %call
 }
diff --git a/test/Transforms/Inline/ignore-debug-info.ll b/test/Transforms/Inline/ignore-debug-info.ll
index 8bd6e7c4287b..eb92bc52cc94 100644
--- a/test/Transforms/Inline/ignore-debug-info.ll
+++ b/test/Transforms/Inline/ignore-debug-info.ll
@@ -12,11 +12,11 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) #1
 
 define <4 x float> @inner_vectors(<4 x float> %a, <4 x float> %b) {
 entry:
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
   %mul = fmul <4 x float> %a, <float 3.000000e+00, float 3.000000e+00, float 3.000000e+00, float 3.000000e+00>
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
   %mul1 = fmul <4 x float> %b, <float 5.000000e+00, float 5.000000e+00, float 5.000000e+00, float 5.000000e+00>
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
   %add = fadd <4 x float> %mul, %mul1
   ret <4 x float> %add
 }
@@ -27,10 +27,10 @@ define float @outer_vectors(<4 x float> %a, <4 x float> %b) {
 ; CHECK: ret float
 
 entry:
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
   %call = call <4 x float> @inner_vectors(<4 x float> %a, <4 x float> %b)
-  call void @llvm.dbg.value(metadata !{}, i64 0, metadata !{}, metadata !{})
+  call void @llvm.dbg.value(metadata i32 undef, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !6), metadata !DIExpression()), !dbg !DILocation(scope: !6)
   %vecext = extractelement <4 x float> %call, i32 0
   %vecext1 = extractelement <4 x float> %call, i32 1
   %add = fadd float %vecext, %vecext1
@@ -47,9 +47,10 @@ attributes #0 = { nounwind readnone }
 !llvm.module.flags = !{!3, !4}
 !llvm.ident = !{!5}
 
-!0 = !{!"0x11\004\00\000\00\000\00\000", !1, !2, !2, !{}, !2, !2} ; [ DW_TAG_compile_unit ]
-!1 = !{!"", !""}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !{!6}, globals: !2, imports: !2)
+!1 = !DIFile(filename: "test.c", directory: "")
+!2 = !{}
 !3 = !{i32 2, !"Dwarf Version", i32 4}
-!4 = !{i32 1, !"Debug Info Version", i32 2}
+!4 = !{i32 1, !"Debug Info Version", i32 3}
 !5 = !{!""}
+!6 = !DISubprogram()
diff --git a/test/Transforms/Inline/inline-byval-bonus.ll b/test/Transforms/Inline/inline-byval-bonus.ll
index 052a5f12a191..7f0c0e1fcbaa 100644
--- a/test/Transforms/Inline/inline-byval-bonus.ll
+++ b/test/Transforms/Inline/inline-byval-bonus.ll
@@ -29,37 +29,37 @@ define i32 @caller(%struct.sphere* %i) {
 declare void @fix(%struct.ray*)
 
 define i32 @ray_sphere(%struct.sphere* nocapture %sph, %struct.ray* nocapture byval align 8 %ray, %struct.spoint* %sp) nounwind uwtable ssp {
-  %1 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 1, i32 0
-  %2 = load double* %1, align 8
+  %1 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 1, i32 0
+  %2 = load double, double* %1, align 8
   %3 = fmul double %2, %2
-  %4 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 1, i32 1
-  %5 = load double* %4, align 8
+  %4 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 1, i32 1
+  %5 = load double, double* %4, align 8
   %6 = fmul double %5, %5
   %7 = fadd double %3, %6
-  %8 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 1, i32 2
-  %9 = load double* %8, align 8
+  %8 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 1, i32 2
+  %9 = load double, double* %8, align 8
   %10 = fmul double %9, %9
   %11 = fadd double %7, %10
   %12 = fmul double %2, 2.000000e+00
-  %13 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 0, i32 0
-  %14 = load double* %13, align 8
-  %15 = getelementptr inbounds %struct.sphere* %sph, i64 0, i32 0, i32 0
-  %16 = load double* %15, align 8
+  %13 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 0, i32 0
+  %14 = load double, double* %13, align 8
+  %15 = getelementptr inbounds %struct.sphere, %struct.sphere* %sph, i64 0, i32 0, i32 0
+  %16 = load double, double* %15, align 8
   %17 = fsub double %14, %16
   %18 = fmul double %12, %17
   %19 = fmul double %5, 2.000000e+00
-  %20 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 0, i32 1
-  %21 = load double* %20, align 8
-  %22 = getelementptr inbounds %struct.sphere* %sph, i64 0, i32 0, i32 1
-  %23 = load double* %22, align 8
+  %20 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 0, i32 1
+  %21 = load double, double* %20, align 8
+  %22 = getelementptr inbounds %struct.sphere, %struct.sphere* %sph, i64 0, i32 0, i32 1
+  %23 = load double, double* %22, align 8
   %24 = fsub double %21, %23
   %25 = fmul double %19, %24
   %26 = fadd double %18, %25
   %27 = fmul double %9, 2.000000e+00
-  %28 = getelementptr inbounds %struct.ray* %ray, i64 0, i32 0, i32 2
-  %29 = load double* %28, align 8
-  %30 = getelementptr inbounds %struct.sphere* %sph, i64 0, i32 0, i32 2
-  %31 = load double* %30, align 8
+  %28 = getelementptr inbounds %struct.ray, %struct.ray* %ray, i64 0, i32 0, i32 2
+  %29 = load double, double* %28, align 8
+  %30 = getelementptr inbounds %struct.sphere, %struct.sphere* %sph, i64 0, i32 0, i32 2
+  %31 = load double, double* %30, align 8
   %32 = fsub double %29, %31
   %33 = fmul double %27, %32
   %34 = fadd double %26, %33
@@ -82,8 +82,8 @@ define i32 @ray_sphere(%struct.sphere* nocapture %sph, %struct.ray* nocapture by
   %51 = fsub double %49, %50
   %52 = fmul double %51, 2.000000e+00
   %53 = fadd double %52, %45
-  %54 = getelementptr inbounds %struct.sphere* %sph, i64 0, i32 1
-  %55 = load double* %54, align 8
+  %54 = getelementptr inbounds %struct.sphere, %struct.sphere* %sph, i64 0, i32 1
+  %55 = load double, double* %54, align 8
   %56 = fmul double %55, %55
   %57 = fsub double %53, %56
   %58 = fmul double %34, %34
@@ -121,37 +121,37 @@ define i32 @ray_sphere(%struct.sphere* nocapture %sph, %struct.ray* nocapture by
   %t2.0 = select i1 %72, double %t1.0, double %70
   %79 = fcmp olt double %t1.0, %t2.0
   %80 = select i1 %79, double %t1.0, double %t2.0
-  %81 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 3
+  %81 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 3
   store double %80, double* %81, align 8
   %82 = fmul double %80, %2
   %83 = fadd double %14, %82
-  %84 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 0, i32 0
+  %84 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 0, i32 0
   store double %83, double* %84, align 8
   %85 = fmul double %5, %80
   %86 = fadd double %21, %85
-  %87 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 0, i32 1
+  %87 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 0, i32 1
   store double %86, double* %87, align 8
   %88 = fmul double %9, %80
   %89 = fadd double %29, %88
-  %90 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 0, i32 2
+  %90 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 0, i32 2
   store double %89, double* %90, align 8
-  %91 = load double* %15, align 8
+  %91 = load double, double* %15, align 8
   %92 = fsub double %83, %91
-  %93 = load double* %54, align 8
+  %93 = load double, double* %54, align 8
   %94 = fdiv double %92, %93
-  %95 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 1, i32 0
+  %95 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 1, i32 0
   store double %94, double* %95, align 8
-  %96 = load double* %22, align 8
+  %96 = load double, double* %22, align 8
   %97 = fsub double %86, %96
-  %98 = load double* %54, align 8
+  %98 = load double, double* %54, align 8
   %99 = fdiv double %97, %98
-  %100 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 1, i32 1
+  %100 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 1, i32 1
   store double %99, double* %100, align 8
-  %101 = load double* %30, align 8
+  %101 = load double, double* %30, align 8
   %102 = fsub double %89, %101
-  %103 = load double* %54, align 8
+  %103 = load double, double* %54, align 8
   %104 = fdiv double %102, %103
-  %105 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 1, i32 2
+  %105 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 1, i32 2
   store double %104, double* %105, align 8
   %106 = fmul double %2, %94
   %107 = fmul double %5, %99
@@ -168,9 +168,9 @@ define i32 @ray_sphere(%struct.sphere* nocapture %sph, %struct.ray* nocapture by
   %118 = fmul double %104, %111
   %119 = fsub double %118, %9
   %120 = fsub double -0.000000e+00, %119
-  %.06 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 2, i32 0
-  %.18 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 2, i32 1
-  %.210 = getelementptr inbounds %struct.spoint* %sp, i64 0, i32 2, i32 2
+  %.06 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 2, i32 0
+  %.18 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 2, i32 1
+  %.210 = getelementptr inbounds %struct.spoint, %struct.spoint* %sp, i64 0, i32 2, i32 2
   %121 = fmul double %113, %113
   %122 = fmul double %116, %116
   %123 = fadd double %121, %122
diff --git a/test/Transforms/Inline/inline-cold.ll b/test/Transforms/Inline/inline-cold.ll
index 57433771961e..b3c134538a45 100644
--- a/test/Transforms/Inline/inline-cold.ll
+++ b/test/Transforms/Inline/inline-cold.ll
@@ -17,29 +17,29 @@
 ; Function Attrs: nounwind readnone uwtable
 define i32 @simpleFunction(i32 %a) #0 {
 entry:
-  %a1 = load volatile i32* @a
+  %a1 = load volatile i32, i32* @a
   %x1 = add i32 %a1,  %a1
-  %a2 = load volatile i32* @a
+  %a2 = load volatile i32, i32* @a
   %x2 = add i32 %x1, %a2
-  %a3 = load volatile i32* @a
+  %a3 = load volatile i32, i32* @a
   %x3 = add i32 %x2, %a3
-  %a4 = load volatile i32* @a
+  %a4 = load volatile i32, i32* @a
   %x4 = add i32 %x3, %a4
-  %a5 = load volatile i32* @a
+  %a5 = load volatile i32, i32* @a
   %x5 = add i32 %x4, %a5
-  %a6 = load volatile i32* @a
+  %a6 = load volatile i32, i32* @a
   %x6 = add i32 %x5, %a6
-  %a7 = load volatile i32* @a
+  %a7 = load volatile i32, i32* @a
   %x7 = add i32 %x6, %a6
-  %a8 = load volatile i32* @a
+  %a8 = load volatile i32, i32* @a
   %x8 = add i32 %x7, %a8
-  %a9 = load volatile i32* @a
+  %a9 = load volatile i32, i32* @a
   %x9 = add i32 %x8, %a9
-  %a10 = load volatile i32* @a
+  %a10 = load volatile i32, i32* @a
   %x10 = add i32 %x9, %a10
-  %a11 = load volatile i32* @a
+  %a11 = load volatile i32, i32* @a
   %x11 = add i32 %x10, %a11
-  %a12 = load volatile i32* @a
+  %a12 = load volatile i32, i32* @a
   %x12 = add i32 %x11, %a12
   %add = add i32 %x12, %a
   ret i32 %add
@@ -54,29 +54,29 @@ define i32 @ColdFunction(i32 %a) #1 {
 ; DEFAULT-LABEL: @ColdFunction
 ; DEFAULT: ret
 entry:
-  %a1 = load volatile i32* @a
+  %a1 = load volatile i32, i32* @a
   %x1 = add i32 %a1,  %a1
-  %a2 = load volatile i32* @a
+  %a2 = load volatile i32, i32* @a
   %x2 = add i32 %x1, %a2
-  %a3 = load volatile i32* @a
+  %a3 = load volatile i32, i32* @a
   %x3 = add i32 %x2, %a3
-  %a4 = load volatile i32* @a
+  %a4 = load volatile i32, i32* @a
   %x4 = add i32 %x3, %a4
-  %a5 = load volatile i32* @a
+  %a5 = load volatile i32, i32* @a
   %x5 = add i32 %x4, %a5
-  %a6 = load volatile i32* @a
+  %a6 = load volatile i32, i32* @a
   %x6 = add i32 %x5, %a6
-  %a7 = load volatile i32* @a
+  %a7 = load volatile i32, i32* @a
   %x7 = add i32 %x6, %a6
-  %a8 = load volatile i32* @a
+  %a8 = load volatile i32, i32* @a
   %x8 = add i32 %x7, %a8
-  %a9 = load volatile i32* @a
+  %a9 = load volatile i32, i32* @a
   %x9 = add i32 %x8, %a9
-  %a10 = load volatile i32* @a
+  %a10 = load volatile i32, i32* @a
   %x10 = add i32 %x9, %a10
-  %a11 = load volatile i32* @a
+  %a11 = load volatile i32, i32* @a
   %x11 = add i32 %x10, %a11
-  %a12 = load volatile i32* @a
+  %a12 = load volatile i32, i32* @a
   %x12 = add i32 %x11, %a12
   %add = add i32 %x12, %a
   ret i32 %add
@@ -91,79 +91,79 @@ define i32 @ColdFunction2(i32 %a) #1 {
 ; DEFAULT-LABEL: @ColdFunction2
 ; DEFAULT: ret
 entry:
-  %a1 = load volatile i32* @a
+  %a1 = load volatile i32, i32* @a
   %x1 = add i32 %a1,  %a1
-  %a2 = load volatile i32* @a
+  %a2 = load volatile i32, i32* @a
   %x2 = add i32 %x1, %a2
-  %a3 = load volatile i32* @a
+  %a3 = load volatile i32, i32* @a
   %x3 = add i32 %x2, %a3
-  %a4 = load volatile i32* @a
+  %a4 = load volatile i32, i32* @a
   %x4 = add i32 %x3, %a4
-  %a5 = load volatile i32* @a
+  %a5 = load volatile i32, i32* @a
   %x5 = add i32 %x4, %a5
-  %a6 = load volatile i32* @a
+  %a6 = load volatile i32, i32* @a
   %x6 = add i32 %x5, %a6
-  %a7 = load volatile i32* @a
+  %a7 = load volatile i32, i32* @a
   %x7 = add i32 %x6, %a7
-  %a8 = load volatile i32* @a
+  %a8 = load volatile i32, i32* @a
   %x8 = add i32 %x7, %a8
-  %a9 = load volatile i32* @a
+  %a9 = load volatile i32, i32* @a
   %x9 = add i32 %x8, %a9
-  %a10 = load volatile i32* @a
+  %a10 = load volatile i32, i32* @a
   %x10 = add i32 %x9, %a10
-  %a11 = load volatile i32* @a
+  %a11 = load volatile i32, i32* @a
   %x11 = add i32 %x10, %a11
-  %a12 = load volatile i32* @a
+  %a12 = load volatile i32, i32* @a
   %x12 = add i32 %x11, %a12
 
-  %a21 = load volatile i32* @a
+  %a21 = load volatile i32, i32* @a
   %x21 = add i32 %x12, %a21
-  %a22 = load volatile i32* @a
+  %a22 = load volatile i32, i32* @a
   %x22 = add i32 %x21, %a22
-  %a23 = load volatile i32* @a
+  %a23 = load volatile i32, i32* @a
   %x23 = add i32 %x22, %a23
-  %a24 = load volatile i32* @a
+  %a24 = load volatile i32, i32* @a
   %x24 = add i32 %x23, %a24
-  %a25 = load volatile i32* @a
+  %a25 = load volatile i32, i32* @a
   %x25 = add i32 %x24, %a25
-  %a26 = load volatile i32* @a
+  %a26 = load volatile i32, i32* @a
   %x26 = add i32 %x25, %a26
-  %a27 = load volatile i32* @a
+  %a27 = load volatile i32, i32* @a
   %x27 = add i32 %x26, %a27
-  %a28 = load volatile i32* @a
+  %a28 = load volatile i32, i32* @a
   %x28 = add i32 %x27, %a28
-  %a29 = load volatile i32* @a
+  %a29 = load volatile i32, i32* @a
   %x29 = add i32 %x28, %a29
-  %a30 = load volatile i32* @a
+  %a30 = load volatile i32, i32* @a
   %x30 = add i32 %x29, %a30
-  %a31 = load volatile i32* @a
+  %a31 = load volatile i32, i32* @a
   %x31 = add i32 %x30, %a31
-  %a32 = load volatile i32* @a
+  %a32 = load volatile i32, i32* @a
   %x32 = add i32 %x31, %a32
 
-  %a41 = load volatile i32* @a
+  %a41 = load volatile i32, i32* @a
   %x41 = add i32 %x32, %a41
-  %a42 = load volatile i32* @a
+  %a42 = load volatile i32, i32* @a
   %x42 = add i32 %x41, %a42
-  %a43 = load volatile i32* @a
+  %a43 = load volatile i32, i32* @a
   %x43 = add i32 %x42, %a43
-  %a44 = load volatile i32* @a
+  %a44 = load volatile i32, i32* @a
   %x44 = add i32 %x43, %a44
-  %a45 = load volatile i32* @a
+  %a45 = load volatile i32, i32* @a
   %x45 = add i32 %x44, %a45
-  %a46 = load volatile i32* @a
+  %a46 = load volatile i32, i32* @a
   %x46 = add i32 %x45, %a46
-  %a47 = load volatile i32* @a
+  %a47 = load volatile i32, i32* @a
   %x47 = add i32 %x46, %a47
-  %a48 = load volatile i32* @a
+  %a48 = load volatile i32, i32* @a
   %x48 = add i32 %x47, %a48
-  %a49 = load volatile i32* @a
+  %a49 = load volatile i32, i32* @a
   %x49 = add i32 %x48, %a49
-  %a50 = load volatile i32* @a
+  %a50 = load volatile i32, i32* @a
   %x50 = add i32 %x49, %a50
-  %a51 = load volatile i32* @a
+  %a51 = load volatile i32, i32* @a
   %x51 = add i32 %x50, %a51
-  %a52 = load volatile i32* @a
+  %a52 = load volatile i32, i32* @a
   %x52 = add i32 %x51, %a52
 
   %add = add i32 %x52, %a
diff --git a/test/Transforms/Inline/inline-fast-math-flags.ll b/test/Transforms/Inline/inline-fast-math-flags.ll
new file mode 100644
index 000000000000..e5ff8f7bfe2b
--- /dev/null
+++ b/test/Transforms/Inline/inline-fast-math-flags.ll
@@ -0,0 +1,34 @@
+; RUN: opt < %s -S -inline -inline-threshold=20 | FileCheck %s
+; Check that we don't drop FastMathFlag when estimating inlining profitability.
+;
+; In this test we should inline 'foo'  to 'boo', because it'll fold to a
+; constant.
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+define float @foo(float* %a, float %b) {
+entry:
+  %a0 = load float, float* %a, align 4
+  %mul = fmul fast float %a0, %b
+  %tobool = fcmp une float %mul, 0.000000e+00
+  br i1 %tobool, label %if.then, label %if.end
+
+if.then:                                          ; preds = %entry
+  %a1 = load float, float* %a, align 8
+  %arrayidx1 = getelementptr inbounds float, float* %a, i64 1
+  %a2 = load float, float* %arrayidx1, align 4
+  %add = fadd fast float %a1, %a2
+  br label %if.end
+
+if.end:                                           ; preds = %if.then, %entry
+  %storemerge = phi float [ %add, %if.then ], [ 1.000000e+00, %entry ]
+  ret float %storemerge
+}
+
+; CHECK-LABEL: @boo
+; CHECK-NOT: call float @foo
+define float @boo(float* %a) {
+entry:
+  %call = call float @foo(float* %a, float 0.000000e+00)
+  ret float %call
+}
diff --git a/test/Transforms/Inline/inline-fp.ll b/test/Transforms/Inline/inline-fp.ll
new file mode 100644
index 000000000000..cf42d82379a2
--- /dev/null
+++ b/test/Transforms/Inline/inline-fp.ll
@@ -0,0 +1,136 @@
+; RUN: opt -S -inline < %s | FileCheck %s
+; Make sure that soft float implementations are calculated as being more expensive
+; to the inliner.
+
+define i32 @test_nofp() #0 {
+; f_nofp() has the "use-soft-float" attribute, so it should never get inlined.
+; CHECK-LABEL: test_nofp
+; CHECK: call float @f_nofp 
+entry:
+  %responseX = alloca i32, align 4
+  %responseY = alloca i32, align 4
+  %responseZ = alloca i32, align 4
+  %valueX = alloca i8, align 1
+  %valueY = alloca i8, align 1
+  %valueZ = alloca i8, align 1
+
+  call void @getX(i32* %responseX, i8* %valueX)
+  call void @getY(i32* %responseY, i8* %valueY)
+  call void @getZ(i32* %responseZ, i8* %valueZ)
+
+  %0 = load i32, i32* %responseX
+  %1 = load i8, i8* %valueX
+  %call = call float @f_nofp(i32 %0, i8 zeroext %1)
+  %2 = load i32, i32* %responseZ
+  %3 = load i8, i8* %valueZ
+  %call2 = call float @f_nofp(i32 %2, i8 zeroext %3)
+  %call3 = call float @fabsf(float %call)
+  %cmp = fcmp ogt float %call3, 0x3FC1EB8520000000
+  br i1 %cmp, label %if.end12, label %if.else
+
+if.else:                                          ; preds = %entry
+  %4 = load i32, i32* %responseY
+  %5 = load i8, i8* %valueY
+  %call1 = call float @f_nofp(i32 %4, i8 zeroext %5)
+  %call4 = call float @fabsf(float %call1)
+  %cmp5 = fcmp ogt float %call4, 0x3FC1EB8520000000
+  br i1 %cmp5, label %if.end12, label %if.else7
+
+if.else7:                                         ; preds = %if.else
+  %call8 = call float @fabsf(float %call2)
+  %cmp9 = fcmp ogt float %call8, 0x3FC1EB8520000000
+  br i1 %cmp9, label %if.then10, label %if.end12
+
+if.then10:                                        ; preds = %if.else7
+  br label %if.end12
+
+if.end12:                                         ; preds = %if.else, %entry, %if.then10, %if.else7
+  %success.0 = phi i32 [ 0, %if.then10 ], [ 1, %if.else7 ], [ 0, %entry ], [ 0, %if.else ]
+  ret i32 %success.0
+}
+
+define i32 @test_hasfp() #0 {
+; f_hasfp()  does not have the "use-soft-float" attribute, so it should get inlined.
+; CHECK-LABEL: test_hasfp
+; CHECK-NOT: call float @f_hasfp 
+entry:
+  %responseX = alloca i32, align 4
+  %responseY = alloca i32, align 4
+  %responseZ = alloca i32, align 4
+  %valueX = alloca i8, align 1
+  %valueY = alloca i8, align 1
+  %valueZ = alloca i8, align 1
+
+  call void @getX(i32* %responseX, i8* %valueX)
+  call void @getY(i32* %responseY, i8* %valueY)
+  call void @getZ(i32* %responseZ, i8* %valueZ)
+
+  %0 = load i32, i32* %responseX
+  %1 = load i8, i8* %valueX
+  %call = call float @f_hasfp(i32 %0, i8 zeroext %1)
+  %2 = load i32, i32* %responseZ
+  %3 = load i8, i8* %valueZ
+  %call2 = call float @f_hasfp(i32 %2, i8 zeroext %3)
+  %call3 = call float @fabsf(float %call)
+  %cmp = fcmp ogt float %call3, 0x3FC1EB8520000000
+  br i1 %cmp, label %if.end12, label %if.else
+
+if.else:                                          ; preds = %entry
+  %4 = load i32, i32* %responseY
+  %5 = load i8, i8* %valueY
+  %call1 = call float @f_hasfp(i32 %4, i8 zeroext %5)
+  %call4 = call float @fabsf(float %call1)
+  %cmp5 = fcmp ogt float %call4, 0x3FC1EB8520000000
+  br i1 %cmp5, label %if.end12, label %if.else7
+
+if.else7:                                         ; preds = %if.else
+  %call8 = call float @fabsf(float %call2)
+  %cmp9 = fcmp ogt float %call8, 0x3FC1EB8520000000
+  br i1 %cmp9, label %if.then10, label %if.end12
+
+if.then10:                                        ; preds = %if.else7
+  br label %if.end12
+
+if.end12:                                         ; preds = %if.else, %entry, %if.then10, %if.else7
+  %success.0 = phi i32 [ 0, %if.then10 ], [ 1, %if.else7 ], [ 0, %entry ], [ 0, %if.else ]
+  ret i32 %success.0
+}
+
+declare void @getX(i32*, i8*) #0
+
+declare void @getY(i32*, i8*) #0
+
+declare void @getZ(i32*, i8*) #0
+
+define internal float @f_hasfp(i32 %response, i8 zeroext %value1) #0 {
+entry:
+  %conv = zext i8 %value1 to i32
+  %sub = add nsw i32 %conv, -1
+  %conv1 = sitofp i32 %sub to float
+  %0 = tail call float @llvm.pow.f32(float 0x3FF028F5C0000000, float %conv1)
+  %mul = fmul float %0, 2.620000e+03
+  %conv2 = sitofp i32 %response to float
+  %sub3 = fsub float %conv2, %mul
+  %div = fdiv float %sub3, %mul
+  ret float %div
+}
+
+define internal float @f_nofp(i32 %response, i8 zeroext %value1) #1 {
+entry:
+  %conv = zext i8 %value1 to i32
+  %sub = add nsw i32 %conv, -1
+  %conv1 = sitofp i32 %sub to float
+  %0 = tail call float @llvm.pow.f32(float 0x3FF028F5C0000000, float %conv1)
+  %mul = fmul float %0, 2.620000e+03
+  %conv2 = sitofp i32 %response to float
+  %sub3 = fsub float %conv2, %mul
+  %div = fdiv float %sub3, %mul
+  ret float %div
+}
+
+declare float @fabsf(float) optsize minsize
+
+declare float @llvm.pow.f32(float, float) optsize minsize
+
+attributes #0 = { minsize optsize }
+attributes #1 = { minsize optsize "use-soft-float"="true" }
diff --git a/test/Transforms/Inline/inline-invoke-tail.ll b/test/Transforms/Inline/inline-invoke-tail.ll
index e07752396a30..5fef4baee9e6 100644
--- a/test/Transforms/Inline/inline-invoke-tail.ll
+++ b/test/Transforms/Inline/inline-invoke-tail.ll
@@ -19,7 +19,7 @@ define i32 @main() {
 			to label %invcont unwind label %lpad
 
 invcont:
-	%retval = load i32* %a, align 4
+	%retval = load i32, i32* %a, align 4
 	ret i32 %retval
 
 lpad:
diff --git a/test/Transforms/Inline/inline-musttail-varargs.ll b/test/Transforms/Inline/inline-musttail-varargs.ll
index 7a8957488e1c..e93ef763aeff 100644
--- a/test/Transforms/Inline/inline-musttail-varargs.ll
+++ b/test/Transforms/Inline/inline-musttail-varargs.ll
@@ -6,17 +6,17 @@
 declare void @ext_method(i8*, i32)
 
 define linkonce_odr void @thunk(i8* %this, ...) {
-  %this_adj = getelementptr i8* %this, i32 4
-  musttail call void (i8*, ...)* bitcast (void (i8*, i32)* @ext_method to void (i8*, ...)*)(i8* %this_adj, ...)
+  %this_adj = getelementptr i8, i8* %this, i32 4
+  musttail call void (i8*, ...) bitcast (void (i8*, i32)* @ext_method to void (i8*, ...)*)(i8* %this_adj, ...)
   ret void
 }
 
 define void @thunk_caller(i8* %p) {
-  call void (i8*, ...)* @thunk(i8* %p, i32 42)
+  call void (i8*, ...) @thunk(i8* %p, i32 42)
   ret void
 }
 ; CHECK-LABEL: define void @thunk_caller(i8* %p)
-; CHECK: call void (i8*, ...)* @thunk(i8* %p, i32 42)
+; CHECK: call void (i8*, ...) @thunk(i8* %p, i32 42)
 
 ; FIXME: Inline the thunk. This should be significantly easier than inlining
 ; general varargs functions.
diff --git a/test/Transforms/Inline/inline-optsize.ll b/test/Transforms/Inline/inline-optsize.ll
index 3ad573a04e42..820e56f7f8ef 100644
--- a/test/Transforms/Inline/inline-optsize.ll
+++ b/test/Transforms/Inline/inline-optsize.ll
@@ -11,15 +11,15 @@
 ; This function should be larger than the inline threshold for -Oz (25), but
 ; smaller than the inline threshold for optsize (75).
 define i32 @inner() {
-  %a1 = load volatile i32* @a
+  %a1 = load volatile i32, i32* @a
   %x1 = add i32 %a1,  %a1
-  %a2 = load volatile i32* @a
+  %a2 = load volatile i32, i32* @a
   %x2 = add i32 %x1, %a2
-  %a3 = load volatile i32* @a
+  %a3 = load volatile i32, i32* @a
   %x3 = add i32 %x2, %a3
-  %a4 = load volatile i32* @a
+  %a4 = load volatile i32, i32* @a
   %x4 = add i32 %x3, %a4
-  %a5 = load volatile i32* @a
+  %a5 = load volatile i32, i32* @a
   %x5 = add i32 %x3, %a5
   ret i32 %x5
 }
diff --git a/test/Transforms/Inline/inline-vla.ll b/test/Transforms/Inline/inline-vla.ll
index 7da448b83f3a..df21b3f30cff 100644
--- a/test/Transforms/Inline/inline-vla.ll
+++ b/test/Transforms/Inline/inline-vla.ll
@@ -10,9 +10,9 @@
 define i32 @main(i32 %argc, i8** nocapture readnone %argv) #0 {
 entry:
   %data = alloca [2 x i8], align 1
-  %arraydecay = getelementptr inbounds [2 x i8]* %data, i64 0, i64 0
-  call fastcc void @memcpy2(i8* %arraydecay, i8* getelementptr inbounds ([2 x i8]* @.str, i64 0, i64 0), i64 1)
-  call fastcc void @memcpy2(i8* %arraydecay, i8* getelementptr inbounds ([3 x i8]* @.str1, i64 0, i64 0), i64 2)
+  %arraydecay = getelementptr inbounds [2 x i8], [2 x i8]* %data, i64 0, i64 0
+  call fastcc void @memcpy2(i8* %arraydecay, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str, i64 0, i64 0), i64 1)
+  call fastcc void @memcpy2(i8* %arraydecay, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str1, i64 0, i64 0), i64 2)
   ret i32 0
 }
 
diff --git a/test/Transforms/Inline/inline_constprop.ll b/test/Transforms/Inline/inline_constprop.ll
index b59a270468e0..de23b6157a71 100644
--- a/test/Transforms/Inline/inline_constprop.ll
+++ b/test/Transforms/Inline/inline_constprop.ll
@@ -267,8 +267,8 @@ entry:
   br i1 %cmp, label %if.then, label %if.end3
 
 if.then:
-  %0 = load i32* %a
-  %1 = load i32* %b
+  %0 = load i32, i32* %a
+  %1 = load i32, i32* %b
   %cmp1 = icmp eq i32 %0, %1
   br i1 %cmp1, label %return, label %if.end3
 
diff --git a/test/Transforms/Inline/inline_dbg_declare.ll b/test/Transforms/Inline/inline_dbg_declare.ll
index 563cde314160..f3ad7ef8b8fc 100644
--- a/test/Transforms/Inline/inline_dbg_declare.ll
+++ b/test/Transforms/Inline/inline_dbg_declare.ll
@@ -1,96 +1,99 @@
-; RUN: opt < %s -S -inline | FileCheck %s

-;

-; The purpose of this test is to check that inline pass preserves debug info

-; for variable using the dbg.declare intrinsic.

-;

-;; This test was generated by running this command:

-;; clang.exe -S -O0 -emit-llvm -g foo.c

-;;

-;; foo.c

-;; ==========================

-;; float foo(float x)

-;; {

-;;    return x;

-;; }

-;;

-;; void bar(float *dst)

-;; {

-;;    dst[0] = foo(dst[0]);

-;; }

-;; ==========================

-

-target datalayout = "e-m:w-p:32:32-i64:64-f80:32-n8:16:32-S32"

-target triple = "i686-pc-windows-msvc"

-

-; Function Attrs: nounwind

-define float @foo(float %x) #0 {

-entry:

-  %x.addr = alloca float, align 4

-  store float %x, float* %x.addr, align 4

-  call void @llvm.dbg.declare(metadata float* %x.addr, metadata !16, metadata !17), !dbg !18

-  %0 = load float* %x.addr, align 4, !dbg !19

-  ret float %0, !dbg !19

-}

-

-; Function Attrs: nounwind readnone

-declare void @llvm.dbg.declare(metadata, metadata, metadata) #1

-

-; CHECK: define void @bar

-

-; Function Attrs: nounwind

-define void @bar(float* %dst) #0 {

-entry:

-

-; CHECK: [[x_addr_i:%[a-zA-Z0-9.]+]] = alloca float, align 4

-

-  %dst.addr = alloca float*, align 4

-  store float* %dst, float** %dst.addr, align 4

-  call void @llvm.dbg.declare(metadata float** %dst.addr, metadata !20, metadata !17), !dbg !21

-  %0 = load float** %dst.addr, align 4, !dbg !22

-  %arrayidx = getelementptr inbounds float* %0, i32 0, !dbg !22

-  %1 = load float* %arrayidx, align 4, !dbg !22

-  %call = call float @foo(float %1), !dbg !22

-

-; CHECK-NOT: call float @foo

-; CHECK: void @llvm.dbg.declare(metadata float* [[x_addr_i]], metadata [[m23:![0-9]+]], metadata !17), !dbg [[m24:![0-9]+]]

-

-  %2 = load float** %dst.addr, align 4, !dbg !22

-  %arrayidx1 = getelementptr inbounds float* %2, i32 0, !dbg !22

-  store float %call, float* %arrayidx1, align 4, !dbg !22

-  ret void, !dbg !23

-}

-

-attributes #0 = { nounwind }

-attributes #1 = { nounwind readnone }

-

-!llvm.dbg.cu = !{!0}

-!llvm.module.flags = !{!13, !14}

-!llvm.ident = !{!15}

-

-!0 = !{!"0x11\0012\00clang version 3.6.0 (trunk)\000\00\000\00\001", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [foo.c] [DW_LANG_C99]

-!1 = !{!"foo.c", !""}

-!2 = !{}

-!3 = !{!4, !9}

-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\000\00256\000\002", !1, !5, !6, null, float (float)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [scope 2] [foo]

-!5 = !{!"0x29", !1}    ; [ DW_TAG_file_type ] [foo.c]

-!6 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]

-!7 = !{!8, !8}

-!8 = !{!"0x24\00float\000\0032\0032\000\000\004", null, null} ; [ DW_TAG_base_type ] [float] [line 0, size 32, align 32, offset 0, enc DW_ATE_float]

-!9 = !{!"0x2e\00bar\00bar\00\006\000\001\000\000\00256\000\007", !1, !5, !10, null, void (float*)* @bar, null, null, !2} ; [ DW_TAG_subprogram ] [line 6] [def] [scope 7] [bar]

-!10 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !11, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]

-!11 = !{null, !12}

-!12 = !{!"0xf\00\000\0032\0032\000\000", null, null, !8} ; [ DW_TAG_pointer_type ] [line 0, size 32, align 32, offset 0] [from float]

-!13 = !{i32 2, !"Dwarf Version", i32 4}

-!14 = !{i32 2, !"Debug Info Version", i32 2}

-!15 = !{!"clang version 3.6.0 (trunk)"}

-!16 = !{!"0x101\00x\0016777217\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [x] [line 1]

-!17 = !{!"0x102"}               ; [ DW_TAG_expression ]

-!18 = !MDLocation(line: 1, column: 17, scope: !4)

-!19 = !MDLocation(line: 3, column: 5, scope: !4)

-!20 = !{!"0x101\00dst\0016777222\000", !9, !5, !12} ; [ DW_TAG_arg_variable ] [dst] [line 6]

-!21 = !MDLocation(line: 6, column: 17, scope: !9)

-!22 = !MDLocation(line: 8, column: 14, scope: !9)

-!23 = !MDLocation(line: 9, column: 1, scope: !9)

-

-; CHECK: [[m23]] = !{!"0x101\00x\0016777217\000", !4, !5, !8, !22} ; [ DW_TAG_arg_variable ] [x] [line 1]

-; CHECK: [[m24]] = !MDLocation(line: 1, column: 17, scope: !4, inlinedAt: !22)

+; RUN: opt < %s -S -inline | FileCheck %s
+;
+; The purpose of this test is to check that inline pass preserves debug info
+; for variable using the dbg.declare intrinsic.
+;
+;; This test was generated by running this command:
+;; clang.exe -S -O0 -emit-llvm -g foo.c
+;;
+;; foo.c
+;; ==========================
+;; float foo(float x)
+;; {
+;;    return x;
+;; }
+;;
+;; void bar(float *dst)
+;; {
+;;    dst[0] = foo(dst[0]);
+;; }
+;; ==========================
+
+target datalayout = "e-m:w-p:32:32-i64:64-f80:32-n8:16:32-S32"
+target triple = "i686-pc-windows-msvc"
+
+; Function Attrs: nounwind
+define float @foo(float %x) #0 {
+entry:
+  %x.addr = alloca float, align 4
+  store float %x, float* %x.addr, align 4
+  call void @llvm.dbg.declare(metadata float* %x.addr, metadata !16, metadata !17), !dbg !18
+  %0 = load float, float* %x.addr, align 4, !dbg !19
+  ret float %0, !dbg !19
+}
+
+; Function Attrs: nounwind readnone
+declare void @llvm.dbg.declare(metadata, metadata, metadata) #1
+
+; CHECK: define void @bar
+
+; Function Attrs: nounwind
+define void @bar(float* %dst) #0 {
+entry:
+
+; CHECK: [[x_addr_i:%[a-zA-Z0-9.]+]] = alloca float, align 4
+
+  %dst.addr = alloca float*, align 4
+  store float* %dst, float** %dst.addr, align 4
+  call void @llvm.dbg.declare(metadata float** %dst.addr, metadata !20, metadata !17), !dbg !21
+  %0 = load float*, float** %dst.addr, align 4, !dbg !22
+  %arrayidx = getelementptr inbounds float, float* %0, i32 0, !dbg !22
+  %1 = load float, float* %arrayidx, align 4, !dbg !22
+  %call = call float @foo(float %1), !dbg !22
+
+; CHECK-NOT: call float @foo
+; CHECK: void @llvm.dbg.declare(metadata float* [[x_addr_i]], metadata [[m23:![0-9]+]], metadata !{{[0-9]+}}), !dbg [[m24:![0-9]+]]
+
+  %2 = load float*, float** %dst.addr, align 4, !dbg !22
+  %arrayidx1 = getelementptr inbounds float, float* %2, i32 0, !dbg !22
+  store float %call, float* %arrayidx1, align 4, !dbg !22
+  ret void, !dbg !23
+}
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readnone }
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!13, !14}
+!llvm.ident = !{!15}
+
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.6.0 (trunk)", isOptimized: false, emissionKind: 1, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "foo.c", directory: "")
+!2 = !{}
+!3 = !{!4, !9}
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 2, file: !1, scope: !5, type: !6, function: float (float)* @foo, variables: !2)
+!5 = !DIFile(filename: "foo.c", directory: "")
+!6 = !DISubroutineType(types: !7)
+!7 = !{!8, !8}
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "float", size: 32, align: 32, encoding: DW_ATE_float)
+!9 = !DISubprogram(name: "bar", line: 6, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 7, file: !1, scope: !5, type: !10, function: void (float*)* @bar, variables: !2)
+!10 = !DISubroutineType(types: !11)
+!11 = !{null, !12}
+!12 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 32, align: 32, baseType: !8)
+!13 = !{i32 2, !"Dwarf Version", i32 4}
+!14 = !{i32 2, !"Debug Info Version", i32 3}
+!15 = !{!"clang version 3.6.0 (trunk)"}
+!16 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "x", line: 1, arg: 1, scope: !4, file: !5, type: !8)
+!17 = !DIExpression()
+!18 = !DILocation(line: 1, column: 17, scope: !4)
+!19 = !DILocation(line: 3, column: 5, scope: !4)
+!20 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "dst", line: 6, arg: 1, scope: !9, file: !5, type: !12)
+!21 = !DILocation(line: 6, column: 17, scope: !9)
+!22 = !DILocation(line: 8, column: 14, scope: !9)
+!23 = !DILocation(line: 9, column: 1, scope: !9)
+
+; CHECK: [[FOO:![0-9]+]] = !DISubprogram(name: "foo",
+; CHECK: [[BAR:![0-9]+]] = !DISubprogram(name: "bar",
+; CHECK: [[m23]] = !DILocalVariable(tag: DW_TAG_arg_variable, name: "x", arg: 1, scope: [[FOO]]
+; CHECK: [[CALL_SITE:![0-9]+]] = distinct !DILocation(line: 8, column: 14, scope: [[BAR]])
+; CHECK: [[m24]] = !DILocation(line: 1, column: 17, scope: [[FOO]], inlinedAt: [[CALL_SITE]])
diff --git a/test/Transforms/Inline/inline_minisize.ll b/test/Transforms/Inline/inline_minisize.ll
index b9aad6083399..0bf75d72bd13 100644
--- a/test/Transforms/Inline/inline_minisize.ll
+++ b/test/Transforms/Inline/inline_minisize.ll
@@ -8,17 +8,17 @@ entry:
   %res = alloca i32, align 4
   %i = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
-  %tmp = load i32* %a.addr, align 4
+  %tmp = load i32, i32* %a.addr, align 4
   %idxprom = sext i32 %tmp to i64
-  %tmp1 = load i32** @data, align 8
-  %arrayidx = getelementptr inbounds i32* %tmp1, i64 %idxprom
-  %tmp2 = load i32* %arrayidx, align 4
-  %tmp3 = load i32* %a.addr, align 4
+  %tmp1 = load i32*, i32** @data, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %tmp1, i64 %idxprom
+  %tmp2 = load i32, i32* %arrayidx, align 4
+  %tmp3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %tmp3, 1
   %idxprom1 = sext i32 %add to i64
-  %tmp4 = load i32** @data, align 8
-  %arrayidx2 = getelementptr inbounds i32* %tmp4, i64 %idxprom1
-  %tmp5 = load i32* %arrayidx2, align 4
+  %tmp4 = load i32*, i32** @data, align 8
+  %arrayidx2 = getelementptr inbounds i32, i32* %tmp4, i64 %idxprom1
+  %tmp5 = load i32, i32* %arrayidx2, align 4
   %mul = mul nsw i32 %tmp2, %tmp5
   store i32 %mul, i32* %res, align 4
   store i32 0, i32* %i, align 4
@@ -26,21 +26,21 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %tmp6 = load i32* %i, align 4
-  %tmp7 = load i32* %res, align 4
+  %tmp6 = load i32, i32* %i, align 4
+  %tmp7 = load i32, i32* %res, align 4
   %cmp = icmp slt i32 %tmp6, %tmp7
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %tmp8 = load i32* %i, align 4
+  %tmp8 = load i32, i32* %i, align 4
   %idxprom3 = sext i32 %tmp8 to i64
-  %tmp9 = load i32** @data, align 8
-  %arrayidx4 = getelementptr inbounds i32* %tmp9, i64 %idxprom3
+  %tmp9 = load i32*, i32** @data, align 8
+  %arrayidx4 = getelementptr inbounds i32, i32* %tmp9, i64 %idxprom3
   call void @fct0(i32* %arrayidx4)
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %tmp10 = load i32* %i, align 4
+  %tmp10 = load i32, i32* %i, align 4
   %inc = add nsw i32 %tmp10, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -50,21 +50,21 @@ for.end:                                          ; preds = %for.cond
   br label %for.cond5
 
 for.cond5:                                        ; preds = %for.inc10, %for.end
-  %tmp11 = load i32* %i, align 4
-  %tmp12 = load i32* %res, align 4
+  %tmp11 = load i32, i32* %i, align 4
+  %tmp12 = load i32, i32* %res, align 4
   %cmp6 = icmp slt i32 %tmp11, %tmp12
   br i1 %cmp6, label %for.body7, label %for.end12
 
 for.body7:                                        ; preds = %for.cond5
-  %tmp13 = load i32* %i, align 4
+  %tmp13 = load i32, i32* %i, align 4
   %idxprom8 = sext i32 %tmp13 to i64
-  %tmp14 = load i32** @data, align 8
-  %arrayidx9 = getelementptr inbounds i32* %tmp14, i64 %idxprom8
+  %tmp14 = load i32*, i32** @data, align 8
+  %arrayidx9 = getelementptr inbounds i32, i32* %tmp14, i64 %idxprom8
   call void @fct0(i32* %arrayidx9)
   br label %for.inc10
 
 for.inc10:                                        ; preds = %for.body7
-  %tmp15 = load i32* %i, align 4
+  %tmp15 = load i32, i32* %i, align 4
   %inc11 = add nsw i32 %tmp15, 1
   store i32 %inc11, i32* %i, align 4
   br label %for.cond5
@@ -74,27 +74,27 @@ for.end12:                                        ; preds = %for.cond5
   br label %for.cond13
 
 for.cond13:                                       ; preds = %for.inc18, %for.end12
-  %tmp16 = load i32* %i, align 4
-  %tmp17 = load i32* %res, align 4
+  %tmp16 = load i32, i32* %i, align 4
+  %tmp17 = load i32, i32* %res, align 4
   %cmp14 = icmp slt i32 %tmp16, %tmp17
   br i1 %cmp14, label %for.body15, label %for.end20
 
 for.body15:                                       ; preds = %for.cond13
-  %tmp18 = load i32* %i, align 4
+  %tmp18 = load i32, i32* %i, align 4
   %idxprom16 = sext i32 %tmp18 to i64
-  %tmp19 = load i32** @data, align 8
-  %arrayidx17 = getelementptr inbounds i32* %tmp19, i64 %idxprom16
+  %tmp19 = load i32*, i32** @data, align 8
+  %arrayidx17 = getelementptr inbounds i32, i32* %tmp19, i64 %idxprom16
   call void @fct0(i32* %arrayidx17)
   br label %for.inc18
 
 for.inc18:                                        ; preds = %for.body15
-  %tmp20 = load i32* %i, align 4
+  %tmp20 = load i32, i32* %i, align 4
   %inc19 = add nsw i32 %tmp20, 1
   store i32 %inc19, i32* %i, align 4
   br label %for.cond13
 
 for.end20:                                        ; preds = %for.cond13
-  %tmp21 = load i32* %res, align 4
+  %tmp21 = load i32, i32* %res, align 4
   ret i32 %tmp21
 }
 
@@ -106,19 +106,19 @@ entry:
   %res = alloca i32, align 4
   %i = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
-  %tmp = load i32* %a.addr, align 4
+  %tmp = load i32, i32* %a.addr, align 4
   %shl = shl i32 %tmp, 1
   %idxprom = sext i32 %shl to i64
-  %tmp1 = load i32** @data, align 8
-  %arrayidx = getelementptr inbounds i32* %tmp1, i64 %idxprom
-  %tmp2 = load i32* %arrayidx, align 4
-  %tmp3 = load i32* %a.addr, align 4
+  %tmp1 = load i32*, i32** @data, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %tmp1, i64 %idxprom
+  %tmp2 = load i32, i32* %arrayidx, align 4
+  %tmp3 = load i32, i32* %a.addr, align 4
   %shl1 = shl i32 %tmp3, 1
   %add = add nsw i32 %shl1, 13
   %idxprom2 = sext i32 %add to i64
-  %tmp4 = load i32** @data, align 8
-  %arrayidx3 = getelementptr inbounds i32* %tmp4, i64 %idxprom2
-  %tmp5 = load i32* %arrayidx3, align 4
+  %tmp4 = load i32*, i32** @data, align 8
+  %arrayidx3 = getelementptr inbounds i32, i32* %tmp4, i64 %idxprom2
+  %tmp5 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %tmp2, %tmp5
   store i32 %mul, i32* %res, align 4
   store i32 0, i32* %i, align 4
@@ -126,21 +126,21 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %tmp6 = load i32* %i, align 4
-  %tmp7 = load i32* %res, align 4
+  %tmp6 = load i32, i32* %i, align 4
+  %tmp7 = load i32, i32* %res, align 4
   %cmp = icmp slt i32 %tmp6, %tmp7
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %tmp8 = load i32* %i, align 4
+  %tmp8 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %tmp8 to i64
-  %tmp9 = load i32** @data, align 8
-  %arrayidx5 = getelementptr inbounds i32* %tmp9, i64 %idxprom4
+  %tmp9 = load i32*, i32** @data, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %tmp9, i64 %idxprom4
   call void @fct0(i32* %arrayidx5)
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %tmp10 = load i32* %i, align 4
+  %tmp10 = load i32, i32* %i, align 4
   %inc = add nsw i32 %tmp10, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -150,21 +150,21 @@ for.end:                                          ; preds = %for.cond
   br label %for.cond6
 
 for.cond6:                                        ; preds = %for.inc11, %for.end
-  %tmp11 = load i32* %i, align 4
-  %tmp12 = load i32* %res, align 4
+  %tmp11 = load i32, i32* %i, align 4
+  %tmp12 = load i32, i32* %res, align 4
   %cmp7 = icmp slt i32 %tmp11, %tmp12
   br i1 %cmp7, label %for.body8, label %for.end13
 
 for.body8:                                        ; preds = %for.cond6
-  %tmp13 = load i32* %i, align 4
+  %tmp13 = load i32, i32* %i, align 4
   %idxprom9 = sext i32 %tmp13 to i64
-  %tmp14 = load i32** @data, align 8
-  %arrayidx10 = getelementptr inbounds i32* %tmp14, i64 %idxprom9
+  %tmp14 = load i32*, i32** @data, align 8
+  %arrayidx10 = getelementptr inbounds i32, i32* %tmp14, i64 %idxprom9
   call void @fct0(i32* %arrayidx10)
   br label %for.inc11
 
 for.inc11:                                        ; preds = %for.body8
-  %tmp15 = load i32* %i, align 4
+  %tmp15 = load i32, i32* %i, align 4
   %inc12 = add nsw i32 %tmp15, 1
   store i32 %inc12, i32* %i, align 4
   br label %for.cond6
@@ -174,27 +174,27 @@ for.end13:                                        ; preds = %for.cond6
   br label %for.cond14
 
 for.cond14:                                       ; preds = %for.inc19, %for.end13
-  %tmp16 = load i32* %i, align 4
-  %tmp17 = load i32* %res, align 4
+  %tmp16 = load i32, i32* %i, align 4
+  %tmp17 = load i32, i32* %res, align 4
   %cmp15 = icmp slt i32 %tmp16, %tmp17
   br i1 %cmp15, label %for.body16, label %for.end21
 
 for.body16:                                       ; preds = %for.cond14
-  %tmp18 = load i32* %i, align 4
+  %tmp18 = load i32, i32* %i, align 4
   %idxprom17 = sext i32 %tmp18 to i64
-  %tmp19 = load i32** @data, align 8
-  %arrayidx18 = getelementptr inbounds i32* %tmp19, i64 %idxprom17
+  %tmp19 = load i32*, i32** @data, align 8
+  %arrayidx18 = getelementptr inbounds i32, i32* %tmp19, i64 %idxprom17
   call void @fct0(i32* %arrayidx18)
   br label %for.inc19
 
 for.inc19:                                        ; preds = %for.body16
-  %tmp20 = load i32* %i, align 4
+  %tmp20 = load i32, i32* %i, align 4
   %inc20 = add nsw i32 %tmp20, 1
   store i32 %inc20, i32* %i, align 4
   br label %for.cond14
 
 for.end21:                                        ; preds = %for.cond14
-  %tmp21 = load i32* %res, align 4
+  %tmp21 = load i32, i32* %res, align 4
   ret i32 %tmp21
 }
 
@@ -206,9 +206,9 @@ entry:
   ;CHECK-NOT: call i32 @fct2
   %c.addr = alloca i32, align 4
   store i32 %c, i32* %c.addr, align 4
-  %tmp = load i32* %c.addr, align 4
+  %tmp = load i32, i32* %c.addr, align 4
   %call = call i32 @fct1(i32 %tmp)
-  %tmp1 = load i32* %c.addr, align 4
+  %tmp1 = load i32, i32* %c.addr, align 4
   %call1 = call i32 @fct2(i32 %tmp1)
   %add = add nsw i32 %call, %call1
   ret i32 %add
@@ -223,9 +223,9 @@ entry:
   ;CHECK: call i32 @fct2
   %c.addr = alloca i32, align 4
   store i32 %c, i32* %c.addr, align 4
-  %tmp = load i32* %c.addr, align 4
+  %tmp = load i32, i32* %c.addr, align 4
   %call = call i32 @fct1(i32 %tmp)
-  %tmp1 = load i32* %c.addr, align 4
+  %tmp1 = load i32, i32* %c.addr, align 4
   %call1 = call i32 @fct2(i32 %tmp1)
   %add = add nsw i32 %call, %call1
   ret i32 %add
diff --git a/test/Transforms/Inline/inline_ssp.ll b/test/Transforms/Inline/inline_ssp.ll
index a4b43a77bad2..1a077230da09 100644
--- a/test/Transforms/Inline/inline_ssp.ll
+++ b/test/Transforms/Inline/inline_ssp.ll
@@ -13,25 +13,25 @@
 ;  sspreq > sspstrong > ssp > [no ssp]
 define internal void @fun_sspreq() nounwind sspreq uwtable {
 entry:
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([12 x i8]* @.str3, i32 0, i32 0))
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str3, i32 0, i32 0))
   ret void
 }
 
 define internal void @fun_sspstrong() nounwind sspstrong uwtable {
 entry:
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([15 x i8]* @.str2, i32 0, i32 0))
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([15 x i8], [15 x i8]* @.str2, i32 0, i32 0))
   ret void
 }
 
 define internal void @fun_ssp() nounwind ssp uwtable {
 entry:
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([9 x i8]* @.str1, i32 0, i32 0))
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str1, i32 0, i32 0))
   ret void
 }
 
 define internal void @fun_nossp() nounwind uwtable {
 entry:
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([11 x i8]* @.str, i32 0, i32 0))
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str, i32 0, i32 0))
   ret void
 }
 
diff --git a/test/Transforms/Inline/invoke-combine-clauses.ll b/test/Transforms/Inline/invoke-combine-clauses.ll
index 5f06039b9edc..89a4cc951c7a 100644
--- a/test/Transforms/Inline/invoke-combine-clauses.ll
+++ b/test/Transforms/Inline/invoke-combine-clauses.ll
@@ -20,7 +20,7 @@ cont:
 lpad:
   %lp = landingpad i32 personality i8* null
       catch i8* @exception_inner
-  %cond = load i1* @condition
+  %cond = load i1, i1* @condition
   br i1 %cond, label %resume1, label %resume2
 resume1:
   resume i32 1
diff --git a/test/Transforms/Inline/lifetime-no-datalayout.ll b/test/Transforms/Inline/lifetime-no-datalayout.ll
index 5abb77f2c3fc..4307064e1946 100644
--- a/test/Transforms/Inline/lifetime-no-datalayout.ll
+++ b/test/Transforms/Inline/lifetime-no-datalayout.ll
@@ -8,13 +8,13 @@ define void @helper() {
   ret void
 }
 
-; Size in llvm.lifetime.X should be -1 (unknown).
+; Size in llvm.lifetime.X should be 1 (default for i8).
 define void @test() {
 ; CHECK-LABEL: @test(
 ; CHECK-NOT: lifetime
-; CHECK: llvm.lifetime.start(i64 -1
+; CHECK: llvm.lifetime.start(i64 1
 ; CHECK-NOT: lifetime
-; CHECK: llvm.lifetime.end(i64 -1
+; CHECK: llvm.lifetime.end(i64 1
   call void @helper()
 ; CHECK-NOT: lifetime
 ; CHECK: ret void
diff --git a/test/Transforms/Inline/noalias-cs.ll b/test/Transforms/Inline/noalias-cs.ll
index da5ddd64ed06..0bff1882e832 100644
--- a/test/Transforms/Inline/noalias-cs.ll
+++ b/test/Transforms/Inline/noalias-cs.ll
@@ -5,13 +5,13 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4, !noalias !3
-  %arrayidx.i = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4, !noalias !3
+  %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx.i, align 4, !alias.scope !7, !noalias !8
-  %arrayidx1.i = getelementptr inbounds float* %b, i64 8
+  %arrayidx1.i = getelementptr inbounds float, float* %b, i64 8
   store float %0, float* %arrayidx1.i, align 4, !alias.scope !8, !noalias !7
-  %1 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %1 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %1, float* %arrayidx, align 4
   ret void
 }
@@ -25,22 +25,22 @@ entry:
 
 ; CHECK: define void @foo(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4, !noalias !6
-; CHECK:   %arrayidx.i.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4, !noalias !6
+; CHECK:   %arrayidx.i.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i.i, align 4, !alias.scope !12, !noalias !13
-; CHECK:   %arrayidx1.i.i = getelementptr inbounds float* %b, i64 8
+; CHECK:   %arrayidx1.i.i = getelementptr inbounds float, float* %b, i64 8
 ; CHECK:   store float %0, float* %arrayidx1.i.i, align 4, !alias.scope !14, !noalias !15
-; CHECK:   %1 = load float* %c, align 4, !noalias !16
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4, !noalias !16
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx.i, align 4, !noalias !16
-; CHECK:   %2 = load float* %a, align 4, !alias.scope !16, !noalias !17
-; CHECK:   %arrayidx.i.i1 = getelementptr inbounds float* %b, i64 5
-; CHECK:   store float %2, float* %arrayidx.i.i1, align 4, !alias.scope !21, !noalias !22
-; CHECK:   %arrayidx1.i.i2 = getelementptr inbounds float* %b, i64 8
-; CHECK:   store float %2, float* %arrayidx1.i.i2, align 4, !alias.scope !23, !noalias !24
-; CHECK:   %3 = load float* %a, align 4, !alias.scope !16
-; CHECK:   %arrayidx.i3 = getelementptr inbounds float* %b, i64 7
-; CHECK:   store float %3, float* %arrayidx.i3, align 4, !alias.scope !16
+; CHECK:   %2 = load float, float* %a, align 4, !alias.scope !16, !noalias !17
+; CHECK:   %arrayidx.i.i.1 = getelementptr inbounds float, float* %b, i64 5
+; CHECK:   store float %2, float* %arrayidx.i.i.1, align 4, !alias.scope !21, !noalias !22
+; CHECK:   %arrayidx1.i.i.2 = getelementptr inbounds float, float* %b, i64 8
+; CHECK:   store float %2, float* %arrayidx1.i.i.2, align 4, !alias.scope !23, !noalias !24
+; CHECK:   %3 = load float, float* %a, align 4, !alias.scope !16
+; CHECK:   %arrayidx.i.3 = getelementptr inbounds float, float* %b, i64 7
+; CHECK:   store float %3, float* %arrayidx.i.3, align 4, !alias.scope !16
 ; CHECK:   ret void
 ; CHECK: }
 
diff --git a/test/Transforms/Inline/noalias.ll b/test/Transforms/Inline/noalias.ll
index 674da1e2efbb..27e53afc2a75 100644
--- a/test/Transforms/Inline/noalias.ll
+++ b/test/Transforms/Inline/noalias.ll
@@ -4,8 +4,8 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @hello(float* noalias nocapture %a, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx, align 4
   ret void
 }
@@ -13,29 +13,29 @@ entry:
 define void @foo(float* nocapture %a, float* nocapture readonly %c) #0 {
 entry:
   tail call void @hello(float* %a, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
 
 ; CHECK: define void @foo(float* nocapture %a, float* nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4, !noalias !0
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4, !noalias !0
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4, !alias.scope !0
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
 
 define void @hello2(float* noalias nocapture %a, float* noalias nocapture %b, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds float* %b, i64 8
+  %arrayidx1 = getelementptr inbounds float, float* %b, i64 8
   store float %0, float* %arrayidx1, align 4
   ret void
 }
@@ -43,21 +43,21 @@ entry:
 define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 entry:
   tail call void @hello2(float* %a, float* %b, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
 
 ; CHECK: define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4, !noalias !3
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4, !noalias !3
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4, !alias.scope !7, !noalias !8
-; CHECK:   %arrayidx1.i = getelementptr inbounds float* %b, i64 8
+; CHECK:   %arrayidx1.i = getelementptr inbounds float, float* %b, i64 8
 ; CHECK:   store float %0, float* %arrayidx1.i, align 4, !alias.scope !8, !noalias !7
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
diff --git a/test/Transforms/Inline/noalias2.ll b/test/Transforms/Inline/noalias2.ll
index 9c8f8e22d388..df135b0a318a 100644
--- a/test/Transforms/Inline/noalias2.ll
+++ b/test/Transforms/Inline/noalias2.ll
@@ -4,8 +4,8 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @hello(float* noalias nocapture %a, float* noalias nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 5
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 5
   store float %0, float* %arrayidx, align 4
   ret void
 }
@@ -13,29 +13,29 @@ entry:
 define void @foo(float* noalias nocapture %a, float* noalias nocapture readonly %c) #0 {
 entry:
   tail call void @hello(float* %a, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
 
 ; CHECK: define void @foo(float* noalias nocapture %a, float* noalias nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4, !alias.scope !0, !noalias !3
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4, !alias.scope !0, !noalias !3
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i, align 4, !alias.scope !3, !noalias !0
-; CHECK:   %1 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
 
 define void @hello2(float* noalias nocapture %a, float* noalias nocapture %b, float* nocapture readonly %c) #0 {
 entry:
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 6
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 6
   store float %0, float* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds float* %b, i64 8
+  %arrayidx1 = getelementptr inbounds float, float* %b, i64 8
   store float %0, float* %arrayidx1, align 4
   ret void
 }
@@ -46,27 +46,27 @@ define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture rea
 entry:
   tail call void @foo(float* %a, float* %c)
   tail call void @hello2(float* %a, float* %b, float* %c)
-  %0 = load float* %c, align 4
-  %arrayidx = getelementptr inbounds float* %a, i64 7
+  %0 = load float, float* %c, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 7
   store float %0, float* %arrayidx, align 4
   ret void
 }
 
 ; CHECK: define void @foo2(float* nocapture %a, float* nocapture %b, float* nocapture readonly %c) #0 {
 ; CHECK: entry:
-; CHECK:   %0 = load float* %c, align 4, !alias.scope !5, !noalias !10
-; CHECK:   %arrayidx.i.i = getelementptr inbounds float* %a, i64 5
+; CHECK:   %0 = load float, float* %c, align 4, !alias.scope !5, !noalias !10
+; CHECK:   %arrayidx.i.i = getelementptr inbounds float, float* %a, i64 5
 ; CHECK:   store float %0, float* %arrayidx.i.i, align 4, !alias.scope !10, !noalias !5
-; CHECK:   %1 = load float* %c, align 4, !alias.scope !13, !noalias !14
-; CHECK:   %arrayidx.i = getelementptr inbounds float* %a, i64 7
+; CHECK:   %1 = load float, float* %c, align 4, !alias.scope !13, !noalias !14
+; CHECK:   %arrayidx.i = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %1, float* %arrayidx.i, align 4, !alias.scope !14, !noalias !13
-; CHECK:   %2 = load float* %c, align 4, !noalias !15
-; CHECK:   %arrayidx.i1 = getelementptr inbounds float* %a, i64 6
-; CHECK:   store float %2, float* %arrayidx.i1, align 4, !alias.scope !19, !noalias !20
-; CHECK:   %arrayidx1.i = getelementptr inbounds float* %b, i64 8
+; CHECK:   %2 = load float, float* %c, align 4, !noalias !15
+; CHECK:   %arrayidx.i.1 = getelementptr inbounds float, float* %a, i64 6
+; CHECK:   store float %2, float* %arrayidx.i.1, align 4, !alias.scope !19, !noalias !20
+; CHECK:   %arrayidx1.i = getelementptr inbounds float, float* %b, i64 8
 ; CHECK:   store float %2, float* %arrayidx1.i, align 4, !alias.scope !20, !noalias !19
-; CHECK:   %3 = load float* %c, align 4
-; CHECK:   %arrayidx = getelementptr inbounds float* %a, i64 7
+; CHECK:   %3 = load float, float* %c, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %a, i64 7
 ; CHECK:   store float %3, float* %arrayidx, align 4
 ; CHECK:   ret void
 ; CHECK: }
diff --git a/test/Transforms/Inline/optimization-remarks.ll b/test/Transforms/Inline/optimization-remarks.ll
index fb1b04788885..8a3e4d1b1fe4 100644
--- a/test/Transforms/Inline/optimization-remarks.ll
+++ b/test/Transforms/Inline/optimization-remarks.ll
@@ -12,8 +12,8 @@ entry:
   %y.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
   store i32 %y, i32* %y.addr, align 4
-  %0 = load i32* %x.addr, align 4
-  %1 = load i32* %y.addr, align 4
+  %0 = load i32, i32* %x.addr, align 4
+  %1 = load i32, i32* %y.addr, align 4
   %add = add nsw i32 %0, %1
   ret i32 %add
 }
@@ -25,8 +25,8 @@ entry:
   %y.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
   store i32 %y, i32* %y.addr, align 4
-  %0 = load i32* %x.addr, align 4
-  %1 = load i32* %y.addr, align 4
+  %0 = load i32, i32* %x.addr, align 4
+  %1 = load i32, i32* %y.addr, align 4
   %mul = mul nsw i32 %0, %1
   %conv = sitofp i32 %mul to float
   ret float %conv
@@ -37,14 +37,14 @@ define i32 @bar(i32 %j) #2 {
 entry:
   %j.addr = alloca i32, align 4
   store i32 %j, i32* %j.addr, align 4
-  %0 = load i32* %j.addr, align 4
-  %1 = load i32* %j.addr, align 4
+  %0 = load i32, i32* %j.addr, align 4
+  %1 = load i32, i32* %j.addr, align 4
   %sub = sub nsw i32 %1, 2
   %call = call i32 @foo(i32 %0, i32 %sub)
   %conv = sitofp i32 %call to float
-  %2 = load i32* %j.addr, align 4
+  %2 = load i32, i32* %j.addr, align 4
   %sub1 = sub nsw i32 %2, 2
-  %3 = load i32* %j.addr, align 4
+  %3 = load i32, i32* %j.addr, align 4
   %call2 = call float @foz(i32 %sub1, i32 %3)
   %mul = fmul float %conv, %call2
   %conv3 = fptosi float %mul to i32
diff --git a/test/Transforms/Inline/pr22285.ll b/test/Transforms/Inline/pr22285.ll
new file mode 100644
index 000000000000..1deae1860f08
--- /dev/null
+++ b/test/Transforms/Inline/pr22285.ll
@@ -0,0 +1,15 @@
+; RUN: opt < %s -inline -S | FileCheck %s
+
+$f1 = comdat any
+; CHECK-NOT: $f1 = comdat any
+
+define void @f2() {
+  call void @f1()
+  ret void
+}
+; CHECK-LABEL: define void @f2
+
+define linkonce_odr void @f1() comdat {
+  ret void
+}
+; CHECK-NOT: define linkonce_odr void @f1() comdat
diff --git a/test/Transforms/Inline/ptr-diff.ll b/test/Transforms/Inline/ptr-diff.ll
index 46c3bcd4dc53..728f3793ee2d 100644
--- a/test/Transforms/Inline/ptr-diff.ll
+++ b/test/Transforms/Inline/ptr-diff.ll
@@ -8,8 +8,8 @@ define i32 @outer1() {
 ; CHECK: ret i32
 
   %ptr = alloca i32
-  %ptr1 = getelementptr inbounds i32* %ptr, i32 0
-  %ptr2 = getelementptr inbounds i32* %ptr, i32 42
+  %ptr1 = getelementptr inbounds i32, i32* %ptr, i32 0
+  %ptr2 = getelementptr inbounds i32, i32* %ptr, i32 42
   %result = call i32 @inner1(i32* %ptr1, i32* %ptr2)
   ret i32 %result
 }
@@ -25,7 +25,7 @@ then:
   ret i32 3
 
 else:
-  %t = load i32* %begin
+  %t = load i32, i32* %begin
   ret i32 %t
 }
 
@@ -36,8 +36,8 @@ define i32 @outer2(i32* %ptr) {
 ; CHECK: call i32 @inner2
 ; CHECK: ret i32
 
-  %ptr1 = getelementptr i32* %ptr, i32 0
-  %ptr2 = getelementptr i32* %ptr, i32 42
+  %ptr1 = getelementptr i32, i32* %ptr, i32 0
+  %ptr2 = getelementptr i32, i32* %ptr, i32 42
   %result = call i32 @inner2(i32* %ptr1, i32* %ptr2)
   ret i32 %result
 }
@@ -53,7 +53,7 @@ then:
   ret i32 3
 
 else:
-  %t = load i32* %begin
+  %t = load i32, i32* %begin
   ret i32 %t
 }
 
@@ -63,9 +63,9 @@ define i32 @inttoptr_free_cost(i32 %a, i32 %b, i32 %c) {
   %p1 = inttoptr i32 %a to i32 addrspace(1)*
   %p2 = inttoptr i32 %b to i32 addrspace(1)*
   %p3 = inttoptr i32 %c to i32 addrspace(1)*
-  %t1 = load i32 addrspace(1)* %p1
-  %t2 = load i32 addrspace(1)* %p2
-  %t3 = load i32 addrspace(1)* %p3
+  %t1 = load i32, i32 addrspace(1)* %p1
+  %t2 = load i32, i32 addrspace(1)* %p2
+  %t3 = load i32, i32 addrspace(1)* %p3
   %s = add i32 %t1, %t2
   %s1 = add i32 %s, %t3
   ret i32 %s1
@@ -84,9 +84,9 @@ define i32 @inttoptr_cost_smaller_ptr(i32 %a, i32 %b, i32 %c) {
   %p1 = inttoptr i32 %a to i32 addrspace(2)*
   %p2 = inttoptr i32 %b to i32 addrspace(2)*
   %p3 = inttoptr i32 %c to i32 addrspace(2)*
-  %t1 = load i32 addrspace(2)* %p1
-  %t2 = load i32 addrspace(2)* %p2
-  %t3 = load i32 addrspace(2)* %p3
+  %t1 = load i32, i32 addrspace(2)* %p1
+  %t2 = load i32, i32 addrspace(2)* %p2
+  %t3 = load i32, i32 addrspace(2)* %p3
   %s = add i32 %t1, %t2
   %s1 = add i32 %s, %t3
   ret i32 %s1
diff --git a/test/Transforms/Inline/store-sroa.ll b/test/Transforms/Inline/store-sroa.ll
new file mode 100644
index 000000000000..6b1ca964d330
--- /dev/null
+++ b/test/Transforms/Inline/store-sroa.ll
@@ -0,0 +1,22 @@
+; RUN: opt -S -O2 -inline-threshold=1 < %s | FileCheck %s
+
+%class.A = type { i32 }
+
+define void @_Z3barP1A(%class.A* %a) #0 {
+entry:
+  %a1 = getelementptr inbounds %class.A, %class.A* %a, i64 0, i32 0
+  %0 = load i32, i32* %a1, align 4
+  %add = add nsw i32 %0, 10
+  store i32 %add, i32* %a1, align 4
+  ret void
+}
+
+define void @_Z3foov() #0 {
+; CHECK-LABEL: @_Z3foov(
+; CHECK-NOT: call void @_Z3barP1A
+; CHECK: ret
+entry:
+  %a = alloca %class.A, align 4
+  call void @_Z3barP1A(%class.A* %a)
+  ret void
+}
diff --git a/test/Transforms/Inline/vector-bonus.ll b/test/Transforms/Inline/vector-bonus.ll
new file mode 100644
index 000000000000..c6745d59e5e0
--- /dev/null
+++ b/test/Transforms/Inline/vector-bonus.ll
@@ -0,0 +1,37 @@
+; RUN: opt < %s -inline -inline-threshold=35  -S | FileCheck %s
+
+define i32 @bar(<4 x i32> %v, i32 %i) #0 {
+entry:
+  %cmp = icmp sgt i32 %i, 4
+  br i1 %cmp, label %if.then, label %if.else
+
+if.then:                                          ; preds = %entry
+  %mul1 = mul nsw i32 %i, %i
+  br label %return
+
+if.else:                                          ; preds = %entry
+  %add1 = add nsw i32 %i, %i
+  %add2 = add nsw i32 %i, %i
+  %add3 = add nsw i32 %i, %i
+  %add4 = add nsw i32 %i, %i
+  %add5 = add nsw i32 %i, %i
+  %add6 = add nsw i32 %i, %i
+  %vecext = extractelement <4 x i32> %v, i32 0
+  %vecext7 = extractelement <4 x i32> %v, i32 1
+  %add7 = add nsw i32 %vecext, %vecext7
+  br label %return
+
+return:                                           ; preds = %if.else, %if.then
+  %retval.0 = phi i32 [ %mul1, %if.then ], [ %add7, %if.else ]
+  ret i32 %retval.0
+}
+
+define i32 @foo(<4 x i32> %v, i32 %a) #1 {
+; CHECK-LABEL: @foo(
+; CHECK-NOT: call i32 @bar
+; CHECK: ret
+entry:
+  %call = call i32 @bar(<4 x i32> %v, i32 %a)
+  ret i32 %call
+}
+
diff --git a/test/Transforms/InstCombine/2003-07-21-ExternalConstant.ll b/test/Transforms/InstCombine/2003-07-21-ExternalConstant.ll
index f550c8349f84..b38154580893 100644
--- a/test/Transforms/InstCombine/2003-07-21-ExternalConstant.ll
+++ b/test/Transforms/InstCombine/2003-07-21-ExternalConstant.ll
@@ -31,14 +31,14 @@ entry:
         %a.0 = alloca i32               ; <i32*> [#uses=2]
         %result = alloca i32            ; <i32*> [#uses=2]
         store i32 %a.1, i32* %a.0
-        %tmp.0 = load i32* %a.0         ; <i32> [#uses=1]
-        %tmp.1 = load i32* @silly               ; <i32> [#uses=1]
+        %tmp.0 = load i32, i32* %a.0         ; <i32> [#uses=1]
+        %tmp.1 = load i32, i32* @silly               ; <i32> [#uses=1]
         %tmp.2 = add i32 %tmp.0, %tmp.1         ; <i32> [#uses=1]
         store i32 %tmp.2, i32* %result
         br label %return
 
 return:         ; preds = %entry
-        %tmp.3 = load i32* %result              ; <i32> [#uses=1]
+        %tmp.3 = load i32, i32* %result              ; <i32> [#uses=1]
         ret i32 %tmp.3
 }
 
diff --git a/test/Transforms/InstCombine/2003-08-12-AllocaNonNull.ll b/test/Transforms/InstCombine/2003-08-12-AllocaNonNull.ll
index bb9a8181ccd4..50b9fdb16458 100644
--- a/test/Transforms/InstCombine/2003-08-12-AllocaNonNull.ll
+++ b/test/Transforms/InstCombine/2003-08-12-AllocaNonNull.ll
@@ -12,7 +12,7 @@ entry:
         br i1 %tmp.1, label %then, label %UnifiedExitNode
 
 then:           ; preds = %entry
-        %tmp.4 = call i32 (...)* @bitmap_clear( i32* %live_head )               ; <i32> [#uses=0]
+        %tmp.4 = call i32 (...) @bitmap_clear( i32* %live_head )               ; <i32> [#uses=0]
         br label %UnifiedExitNode
 
 UnifiedExitNode:                ; preds = %then, %entry
diff --git a/test/Transforms/InstCombine/2003-09-09-VolatileLoadElim.ll b/test/Transforms/InstCombine/2003-09-09-VolatileLoadElim.ll
index 7f7390809c77..3e0cf1247d90 100644
--- a/test/Transforms/InstCombine/2003-09-09-VolatileLoadElim.ll
+++ b/test/Transforms/InstCombine/2003-09-09-VolatileLoadElim.ll
@@ -2,6 +2,6 @@
 
 define void @test(i32* %P) {
         ; Dead but not deletable!
-        %X = load volatile i32* %P              ; <i32> [#uses=0]
+        %X = load volatile i32, i32* %P              ; <i32> [#uses=0]
         ret void
 }
diff --git a/test/Transforms/InstCombine/2003-11-03-VarargsCallBug.ll b/test/Transforms/InstCombine/2003-11-03-VarargsCallBug.ll
index c1692f77abb2..aff39f8bf979 100644
--- a/test/Transforms/InstCombine/2003-11-03-VarargsCallBug.ll
+++ b/test/Transforms/InstCombine/2003-11-03-VarargsCallBug.ll
@@ -7,7 +7,7 @@ declare void @foo(...)
 
 define void @test(i64 %X) {
         %Y = inttoptr i64 %X to i32*            ; <i32*> [#uses=1]
-        call void (...)* @foo( i32* %Y )
+        call void (...) @foo( i32* %Y )
         ret void
 }
 
diff --git a/test/Transforms/InstCombine/2004-01-13-InstCombineInvokePHI.ll b/test/Transforms/InstCombine/2004-01-13-InstCombineInvokePHI.ll
index 6df30c7e7fd1..a086c0156504 100644
--- a/test/Transforms/InstCombine/2004-01-13-InstCombineInvokePHI.ll
+++ b/test/Transforms/InstCombine/2004-01-13-InstCombineInvokePHI.ll
@@ -19,7 +19,7 @@ call:           ; preds = %entry
 
 cont:           ; preds = %call, %entry
         %P2 = phi i32* [ %P, %call ], [ null, %entry ]          ; <i32*> [#uses=1]
-        %V = load i32* %P2              ; <i32> [#uses=1]
+        %V = load i32, i32* %P2              ; <i32> [#uses=1]
         ret i32 %V
 
 N:              ; preds = %call
diff --git a/test/Transforms/InstCombine/2004-05-07-UnsizedCastLoad.ll b/test/Transforms/InstCombine/2004-05-07-UnsizedCastLoad.ll
index 8b549374a70c..ff855dc13e5b 100644
--- a/test/Transforms/InstCombine/2004-05-07-UnsizedCastLoad.ll
+++ b/test/Transforms/InstCombine/2004-05-07-UnsizedCastLoad.ll
@@ -4,7 +4,7 @@
 
 define i32 @test(%Ty* %X) {
         %Y = bitcast %Ty* %X to i32*            ; <i32*> [#uses=1]
-        %Z = load i32* %Y               ; <i32> [#uses=1]
+        %Z = load i32, i32* %Y               ; <i32> [#uses=1]
         ret i32 %Z
 }
 
diff --git a/test/Transforms/InstCombine/2004-09-20-BadLoadCombine.ll b/test/Transforms/InstCombine/2004-09-20-BadLoadCombine.ll
index d17db8d7eac2..10122e48ab66 100644
--- a/test/Transforms/InstCombine/2004-09-20-BadLoadCombine.ll
+++ b/test/Transforms/InstCombine/2004-09-20-BadLoadCombine.ll
@@ -12,7 +12,7 @@ define i32 @test(i1 %C) {
         store i32 2, i32* %X2
         %Y = select i1 %C, i32* %X, i32* %X2            ; <i32*> [#uses=1]
         store i32 3, i32* %X
-        %Z = load i32* %Y               ; <i32> [#uses=1]
+        %Z = load i32, i32* %Y               ; <i32> [#uses=1]
         ret i32 %Z
 }
 
diff --git a/test/Transforms/InstCombine/2004-09-20-BadLoadCombine2.ll b/test/Transforms/InstCombine/2004-09-20-BadLoadCombine2.ll
index 0d5fc810a8ae..981a4f301a99 100644
--- a/test/Transforms/InstCombine/2004-09-20-BadLoadCombine2.ll
+++ b/test/Transforms/InstCombine/2004-09-20-BadLoadCombine2.ll
@@ -18,7 +18,7 @@ cond_true.i:            ; preds = %entry
 cond_continue.i:                ; preds = %cond_true.i, %entry
         %mem_tmp.i.0 = phi i32* [ %X, %cond_true.i ], [ %X2, %entry ]           ; <i32*> [#uses=1]
         store i32 3, i32* %X
-        %tmp.3 = load i32* %mem_tmp.i.0         ; <i32> [#uses=1]
+        %tmp.3 = load i32, i32* %mem_tmp.i.0         ; <i32> [#uses=1]
         ret i32 %tmp.3
 }
 
diff --git a/test/Transforms/InstCombine/2005-06-16-SetCCOrSetCCMiscompile.ll b/test/Transforms/InstCombine/2005-06-16-SetCCOrSetCCMiscompile.ll
index 0a513c647c12..9c989b9ecf8a 100644
--- a/test/Transforms/InstCombine/2005-06-16-SetCCOrSetCCMiscompile.ll
+++ b/test/Transforms/InstCombine/2005-06-16-SetCCOrSetCCMiscompile.ll
@@ -5,7 +5,7 @@
 @g_07918478 = external global i32               ; <i32*> [#uses=1]
 
 define i1 @test() {
-        %tmp.0 = load i32* @g_07918478          ; <i32> [#uses=2]
+        %tmp.0 = load i32, i32* @g_07918478          ; <i32> [#uses=2]
         %tmp.1 = icmp ne i32 %tmp.0, 0          ; <i1> [#uses=1]
         %tmp.4 = icmp ult i32 %tmp.0, 4111              ; <i1> [#uses=1]
         %bothcond = or i1 %tmp.1, %tmp.4                ; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2006-09-15-CastToBool.ll b/test/Transforms/InstCombine/2006-09-15-CastToBool.ll
index ee261ced5869..2eb28a3f6bb8 100644
--- a/test/Transforms/InstCombine/2006-09-15-CastToBool.ll
+++ b/test/Transforms/InstCombine/2006-09-15-CastToBool.ll
@@ -2,7 +2,7 @@
 ; PR913
 
 define i32 @test(i32* %tmp1) {
-        %tmp.i = load i32* %tmp1                ; <i32> [#uses=1]
+        %tmp.i = load i32, i32* %tmp1                ; <i32> [#uses=1]
         %tmp = bitcast i32 %tmp.i to i32                ; <i32> [#uses=1]
         %tmp2.ui = lshr i32 %tmp, 5             ; <i32> [#uses=1]
         %tmp2 = bitcast i32 %tmp2.ui to i32             ; <i32> [#uses=1]
diff --git a/test/Transforms/InstCombine/2006-12-08-Phi-ICmp-Op-Fold.ll b/test/Transforms/InstCombine/2006-12-08-Phi-ICmp-Op-Fold.ll
index 295006ca2318..60113fb56206 100644
--- a/test/Transforms/InstCombine/2006-12-08-Phi-ICmp-Op-Fold.ll
+++ b/test/Transforms/InstCombine/2006-12-08-Phi-ICmp-Op-Fold.ll
@@ -12,24 +12,24 @@ entry:
 	%p3_addr = alloca %struct.point		; <%struct.point*> [#uses=2]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	%tmp = bitcast %struct.point* %p1_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp.upgrd.1 = getelementptr { i64 }* %tmp, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp.upgrd.1 = getelementptr { i64 }, { i64 }* %tmp, i64 0, i32 0		; <i64*> [#uses=1]
 	store i64 %p1.0, i64* %tmp.upgrd.1
 	%tmp1 = bitcast %struct.point* %p2_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp2 = getelementptr { i64 }* %tmp1, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp2 = getelementptr { i64 }, { i64 }* %tmp1, i64 0, i32 0		; <i64*> [#uses=1]
 	store i64 %p2.0, i64* %tmp2
 	%tmp3 = bitcast %struct.point* %p3_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp4 = getelementptr { i64 }* %tmp3, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp4 = getelementptr { i64 }, { i64 }* %tmp3, i64 0, i32 0		; <i64*> [#uses=1]
 	store i64 %p3.0, i64* %tmp4
 	%tmp.upgrd.2 = icmp eq i32 %direction, 0		; <i1> [#uses=1]
 	%tmp5 = bitcast %struct.point* %p1_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp6 = getelementptr { i64 }* %tmp5, i64 0, i32 0		; <i64*> [#uses=1]
-	%tmp.upgrd.3 = load i64* %tmp6		; <i64> [#uses=1]
+	%tmp6 = getelementptr { i64 }, { i64 }* %tmp5, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp.upgrd.3 = load i64, i64* %tmp6		; <i64> [#uses=1]
 	%tmp7 = bitcast %struct.point* %p2_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp8 = getelementptr { i64 }* %tmp7, i64 0, i32 0		; <i64*> [#uses=1]
-	%tmp9 = load i64* %tmp8		; <i64> [#uses=1]
+	%tmp8 = getelementptr { i64 }, { i64 }* %tmp7, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp9 = load i64, i64* %tmp8		; <i64> [#uses=1]
 	%tmp10 = bitcast %struct.point* %p3_addr to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp11 = getelementptr { i64 }* %tmp10, i64 0, i32 0		; <i64*> [#uses=1]
-	%tmp12 = load i64* %tmp11		; <i64> [#uses=1]
+	%tmp11 = getelementptr { i64 }, { i64 }* %tmp10, i64 0, i32 0		; <i64*> [#uses=1]
+	%tmp12 = load i64, i64* %tmp11		; <i64> [#uses=1]
 	%tmp13 = call i32 @determinant( i64 %tmp.upgrd.3, i64 %tmp9, i64 %tmp12 )		; <i32> [#uses=2]
 	br i1 %tmp.upgrd.2, label %cond_true, label %cond_false
 
diff --git a/test/Transforms/InstCombine/2006-12-08-Select-ICmp.ll b/test/Transforms/InstCombine/2006-12-08-Select-ICmp.ll
index 272753cc98d6..f54416d14254 100644
--- a/test/Transforms/InstCombine/2006-12-08-Select-ICmp.ll
+++ b/test/Transforms/InstCombine/2006-12-08-Select-ICmp.ll
@@ -12,24 +12,24 @@ entry:
         %p2_addr = alloca %struct.point         ; <%struct.point*> [#uses=2]
         %p3_addr = alloca %struct.point         ; <%struct.point*> [#uses=2]
         %tmp = bitcast %struct.point* %p1_addr to { i64 }*              ; <{ i64 }*> [#uses=1]
-        %tmp.upgrd.1 = getelementptr { i64 }* %tmp, i32 0, i32 0                ; <i64*> [#uses=1]
+        %tmp.upgrd.1 = getelementptr { i64 }, { i64 }* %tmp, i32 0, i32 0                ; <i64*> [#uses=1]
         store i64 %p1.0, i64* %tmp.upgrd.1
         %tmp1 = bitcast %struct.point* %p2_addr to { i64 }*             ; <{ i64 }*> [#uses=1]
-        %tmp2 = getelementptr { i64 }* %tmp1, i32 0, i32 0              ; <i64*> [#uses=1]
+        %tmp2 = getelementptr { i64 }, { i64 }* %tmp1, i32 0, i32 0              ; <i64*> [#uses=1]
         store i64 %p2.0, i64* %tmp2
         %tmp3 = bitcast %struct.point* %p3_addr to { i64 }*             ; <{ i64 }*> [#uses=1]
-        %tmp4 = getelementptr { i64 }* %tmp3, i32 0, i32 0              ; <i64*> [#uses=1]
+        %tmp4 = getelementptr { i64 }, { i64 }* %tmp3, i32 0, i32 0              ; <i64*> [#uses=1]
         store i64 %p3.0, i64* %tmp4
         %tmp.upgrd.2 = icmp eq i32 %direction, 0                ; <i1> [#uses=1]
         %tmp5 = bitcast %struct.point* %p1_addr to { i64 }*             ; <{ i64 }*> [#uses=1]
-        %tmp6 = getelementptr { i64 }* %tmp5, i32 0, i32 0              ; <i64*> [#uses=1]
-        %tmp.upgrd.3 = load i64* %tmp6          ; <i64> [#uses=1]
+        %tmp6 = getelementptr { i64 }, { i64 }* %tmp5, i32 0, i32 0              ; <i64*> [#uses=1]
+        %tmp.upgrd.3 = load i64, i64* %tmp6          ; <i64> [#uses=1]
         %tmp7 = bitcast %struct.point* %p2_addr to { i64 }*             ; <{ i64 }*> [#uses=1]
-        %tmp8 = getelementptr { i64 }* %tmp7, i32 0, i32 0              ; <i64*> [#uses=1]
-        %tmp9 = load i64* %tmp8         ; <i64> [#uses=1]
+        %tmp8 = getelementptr { i64 }, { i64 }* %tmp7, i32 0, i32 0              ; <i64*> [#uses=1]
+        %tmp9 = load i64, i64* %tmp8         ; <i64> [#uses=1]
         %tmp10 = bitcast %struct.point* %p3_addr to { i64 }*            ; <{ i64 }*> [#uses=1]
-        %tmp11 = getelementptr { i64 }* %tmp10, i32 0, i32 0            ; <i64*> [#uses=1]
-        %tmp12 = load i64* %tmp11               ; <i64> [#uses=1]
+        %tmp11 = getelementptr { i64 }, { i64 }* %tmp10, i32 0, i32 0            ; <i64*> [#uses=1]
+        %tmp12 = load i64, i64* %tmp11               ; <i64> [#uses=1]
         %tmp13 = call i32 @determinant( i64 %tmp.upgrd.3, i64 %tmp9, i64 %tmp12 )         ; <i32> [#uses=2]
         %tmp14 = icmp slt i32 %tmp13, 0         ; <i1> [#uses=1]
         %tmp26 = icmp sgt i32 %tmp13, 0         ; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2006-12-15-Range-Test.ll b/test/Transforms/InstCombine/2006-12-15-Range-Test.ll
index 0c8eeceec81c..784b3e4fe687 100644
--- a/test/Transforms/InstCombine/2006-12-15-Range-Test.ll
+++ b/test/Transforms/InstCombine/2006-12-15-Range-Test.ll
@@ -21,8 +21,8 @@ cond_next23.exitStub:           ; preds = %cond_true
         ret i1 false
 
 cond_true:              ; preds = %newFuncRoot
-        %tmp15 = getelementptr [17 x i32]* @r, i32 0, i32 %tmp12.reload         ; <i32*> [#uses=1]
-        %tmp16 = load i32* %tmp15               ; <i32> [#uses=4]
+        %tmp15 = getelementptr [17 x i32], [17 x i32]* @r, i32 0, i32 %tmp12.reload         ; <i32*> [#uses=1]
+        %tmp16 = load i32, i32* %tmp15               ; <i32> [#uses=4]
         %tmp18 = icmp slt i32 %tmp16, -31               ; <i1> [#uses=1]
         %tmp21 = icmp sgt i32 %tmp16, 31                ; <i1> [#uses=1]
         %bothcond = or i1 %tmp18, %tmp21                ; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2006-12-23-Select-Cmp-Cmp.ll b/test/Transforms/InstCombine/2006-12-23-Select-Cmp-Cmp.ll
index eba1ac1298ce..9251e9b455d6 100644
--- a/test/Transforms/InstCombine/2006-12-23-Select-Cmp-Cmp.ll
+++ b/test/Transforms/InstCombine/2006-12-23-Select-Cmp-Cmp.ll
@@ -14,9 +14,9 @@ target triple = "i686-pc-linux-gnu"
 
 define void @mng_write_basi() {
 entry:
-	%tmp = load i8* null		; <i8> [#uses=1]
+	%tmp = load i8, i8* null		; <i8> [#uses=1]
 	%tmp.upgrd.1 = icmp ugt i8 %tmp, 8		; <i1> [#uses=1]
-	%tmp.upgrd.2 = load i16* null		; <i16> [#uses=2]
+	%tmp.upgrd.2 = load i16, i16* null		; <i16> [#uses=2]
 	%tmp3 = icmp eq i16 %tmp.upgrd.2, 255		; <i1> [#uses=1]
 	%tmp7 = icmp eq i16 %tmp.upgrd.2, -1		; <i1> [#uses=1]
 	%bOpaque.0.in = select i1 %tmp.upgrd.1, i1 %tmp7, i1 %tmp3		; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2007-02-01-LoadSinkAlloca.ll b/test/Transforms/InstCombine/2007-02-01-LoadSinkAlloca.ll
index e2bebecded52..113ada3df9d9 100644
--- a/test/Transforms/InstCombine/2007-02-01-LoadSinkAlloca.ll
+++ b/test/Transforms/InstCombine/2007-02-01-LoadSinkAlloca.ll
@@ -10,33 +10,33 @@ define i32 @test2(i32 %C) {
 entry:
 	%A = alloca i32
 	%B = alloca i32
-	%tmp = call i32 (...)* @bar( i32* %A )		; <i32> [#uses=0]
-	%T = load i32* %A		; <i32> [#uses=1]
+	%tmp = call i32 (...) @bar( i32* %A )		; <i32> [#uses=0]
+	%T = load i32, i32* %A		; <i32> [#uses=1]
 	%tmp2 = icmp eq i32 %C, 0		; <i1> [#uses=1]
 	br i1 %tmp2, label %cond_next, label %cond_true
 
 cond_true:		; preds = %entry
 	store i32 123, i32* %B
 	call i32 @test2( i32 123 )		; <i32>:0 [#uses=0]
-	%T1 = load i32* %B		; <i32> [#uses=1]
+	%T1 = load i32, i32* %B		; <i32> [#uses=1]
 	br label %cond_next
 
 cond_next:		; preds = %cond_true, %entry
 	%tmp1.0 = phi i32 [ %T1, %cond_true ], [ %T, %entry ]		; <i32> [#uses=1]
-	%tmp7 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp8 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp9 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp10 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp11 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp12 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp13 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp14 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp15 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp16 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp17 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp18 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp19 = call i32 (...)* @baq( )		; <i32> [#uses=0]
-	%tmp20 = call i32 (...)* @baq( )		; <i32> [#uses=0]
+	%tmp7 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp8 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp9 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp10 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp11 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp12 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp13 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp14 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp15 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp16 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp17 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp18 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp19 = call i32 (...) @baq( )		; <i32> [#uses=0]
+	%tmp20 = call i32 (...) @baq( )		; <i32> [#uses=0]
 	ret i32 %tmp1.0
 }
 
diff --git a/test/Transforms/InstCombine/2007-02-07-PointerCast.ll b/test/Transforms/InstCombine/2007-02-07-PointerCast.ll
index bf60991f7dd9..ddc1e03ddd20 100644
--- a/test/Transforms/InstCombine/2007-02-07-PointerCast.ll
+++ b/test/Transforms/InstCombine/2007-02-07-PointerCast.ll
@@ -12,11 +12,11 @@ declare i32 @printf(i8*, ...)
 
 define i32 @main(i32 %x, i8** %a) {
 entry:
-        %tmp = getelementptr [6 x i8]* @str, i32 0, i64 0               ; <i8*> [#uses=1]
-        %tmp1 = load i8** %a            ; <i8*> [#uses=1]
+        %tmp = getelementptr [6 x i8], [6 x i8]* @str, i32 0, i64 0               ; <i8*> [#uses=1]
+        %tmp1 = load i8*, i8** %a            ; <i8*> [#uses=1]
         %tmp2 = ptrtoint i8* %tmp1 to i32               ; <i32> [#uses=1]
         %tmp3 = zext i32 %tmp2 to i64           ; <i64> [#uses=1]
-        %tmp.upgrd.1 = call i32 (i8*, ...)* @printf( i8* %tmp, i64 %tmp3 )              ; <i32> [#uses=0]
+        %tmp.upgrd.1 = call i32 (i8*, ...) @printf( i8* %tmp, i64 %tmp3 )              ; <i32> [#uses=0]
         ret i32 0
 }
 
diff --git a/test/Transforms/InstCombine/2007-03-25-BadShiftMask.ll b/test/Transforms/InstCombine/2007-03-25-BadShiftMask.ll
index 7e9c9e28c94b..adaaf784ab6c 100644
--- a/test/Transforms/InstCombine/2007-03-25-BadShiftMask.ll
+++ b/test/Transforms/InstCombine/2007-03-25-BadShiftMask.ll
@@ -7,12 +7,12 @@
 define i32 @main() {
 entry:
 	%u = alloca %struct..1anon, align 8		; <%struct..1anon*> [#uses=4]
-	%tmp1 = getelementptr %struct..1anon* %u, i32 0, i32 0		; <double*> [#uses=1]
+	%tmp1 = getelementptr %struct..1anon, %struct..1anon* %u, i32 0, i32 0		; <double*> [#uses=1]
 	store double 0x7FF0000000000000, double* %tmp1
-	%tmp3 = getelementptr %struct..1anon* %u, i32 0, i32 0		; <double*> [#uses=1]
+	%tmp3 = getelementptr %struct..1anon, %struct..1anon* %u, i32 0, i32 0		; <double*> [#uses=1]
 	%tmp34 = bitcast double* %tmp3 to %struct..0anon*		; <%struct..0anon*> [#uses=1]
-	%tmp5 = getelementptr %struct..0anon* %tmp34, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp6 = load i32* %tmp5		; <i32> [#uses=1]
+	%tmp5 = getelementptr %struct..0anon, %struct..0anon* %tmp34, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp6 = load i32, i32* %tmp5		; <i32> [#uses=1]
 	%tmp7 = shl i32 %tmp6, 1		; <i32> [#uses=1]
 	%tmp8 = lshr i32 %tmp7, 21		; <i32> [#uses=1]
 	%tmp89 = trunc i32 %tmp8 to i16		; <i16> [#uses=1]
diff --git a/test/Transforms/InstCombine/2007-05-14-Crash.ll b/test/Transforms/InstCombine/2007-05-14-Crash.ll
index a3c010d2c427..a3d216591439 100644
--- a/test/Transforms/InstCombine/2007-05-14-Crash.ll
+++ b/test/Transforms/InstCombine/2007-05-14-Crash.ll
@@ -10,7 +10,7 @@ target triple = "powerpc-apple-darwin8.8.0"
 define i8* @foo(%struct.anon* %deviceRef, %struct.abc* %pCap) {
 entry:
         %tmp1 = bitcast %struct.anon* %deviceRef to %struct.def*            
-        %tmp3 = getelementptr %struct.def* %tmp1, i32 0, i32 1               
+        %tmp3 = getelementptr %struct.def, %struct.def* %tmp1, i32 0, i32 1               
         %tmp35 = bitcast %struct.abc* %tmp3 to i8*           
         ret i8* %tmp35
 }
diff --git a/test/Transforms/InstCombine/2007-06-06-AshrSignBit.ll b/test/Transforms/InstCombine/2007-06-06-AshrSignBit.ll
index 082b2155a010..2b89a9d721fb 100644
--- a/test/Transforms/InstCombine/2007-06-06-AshrSignBit.ll
+++ b/test/Transforms/InstCombine/2007-06-06-AshrSignBit.ll
@@ -9,13 +9,13 @@ return.exitStub:		; preds = %cond_true
 	ret void
 
 cond_true:		; preds = %newFuncRoot
-	%tmp30 = load i64* %tmp10		; <i64> [#uses=1]
+	%tmp30 = load i64, i64* %tmp10		; <i64> [#uses=1]
 	%.cast = zext i32 63 to i64		; <i64> [#uses=1]
 	%tmp31 = ashr i64 %tmp30, %.cast		; <i64> [#uses=1]
 	%tmp3132 = trunc i64 %tmp31 to i32		; <i32> [#uses=1]
 	%tmp33 = or i32 %tmp3132, 1		; <i32> [#uses=1]
 	store i32 %tmp33, i32* %tmp9
-	%tmp34 = load i32* %tmp9		; <i32> [#uses=1]
+	%tmp34 = load i32, i32* %tmp9		; <i32> [#uses=1]
 	store i32 %tmp34, i32* %retval
 	br label %return.exitStub
 }
diff --git a/test/Transforms/InstCombine/2007-09-10-AliasConstFold.ll b/test/Transforms/InstCombine/2007-09-10-AliasConstFold.ll
index 625989384d70..7f06f009515f 100644
--- a/test/Transforms/InstCombine/2007-09-10-AliasConstFold.ll
+++ b/test/Transforms/InstCombine/2007-09-10-AliasConstFold.ll
@@ -9,7 +9,7 @@ define weak i32 @pthread_cancel(i32) {
 
 define i1 @__gthread_active_p() {
 entry:
-	%tmp1 = load i8** @__gthread_active_ptr.5335, align 4		; <i8*> [#uses=1]
+	%tmp1 = load i8*, i8** @__gthread_active_ptr.5335, align 4		; <i8*> [#uses=1]
 	%tmp2 = icmp ne i8* %tmp1, null		; <i1> [#uses=1]
 	ret i1 %tmp2
 }
diff --git a/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll b/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
index 656fb34061d9..67a94e5c8ccd 100644
--- a/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
+++ b/test/Transforms/InstCombine/2007-10-10-EliminateMemCpy.ll
@@ -8,8 +8,8 @@ define void @foo(i8* %P) {
 entry:
   %P_addr = alloca i8*
   store i8* %P, i8** %P_addr
-  %tmp = load i8** %P_addr, align 4
-  %tmp1 = getelementptr [4 x i8]* @.str, i32 0, i32 0
+  %tmp = load i8*, i8** %P_addr, align 4
+  %tmp1 = getelementptr [4 x i8], [4 x i8]* @.str, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* %tmp1, i32 4, i32 1, i1 false)
   br label %return
 
diff --git a/test/Transforms/InstCombine/2007-10-12-Crash.ll b/test/Transforms/InstCombine/2007-10-12-Crash.ll
index b3d9f02b68aa..33d55e7e9e02 100644
--- a/test/Transforms/InstCombine/2007-10-12-Crash.ll
+++ b/test/Transforms/InstCombine/2007-10-12-Crash.ll
@@ -14,7 +14,7 @@ entry:
 
 bb29:		; preds = %bb62
 	%tmp322 = bitcast %struct.Ray* %tmp3 to %struct.Vec*		; <%struct.Vec*> [#uses=1]
-	%tmp322.0 = getelementptr %struct.Vec* %tmp322, i32 0, i32 0		; <double*> [#uses=1]
+	%tmp322.0 = getelementptr %struct.Vec, %struct.Vec* %tmp322, i32 0, i32 0		; <double*> [#uses=1]
 	store double 0.000000e+00, double* %tmp322.0
 	%tmp57 = call double @_Z9ray_traceRK3VecRK3RayRK5Scene( %struct.Vec* null, %struct.Ray* %tmp3, %struct.Scene* null )		; <double> [#uses=0]
 	br label %bb62
diff --git a/test/Transforms/InstCombine/2007-10-28-stacksave.ll b/test/Transforms/InstCombine/2007-10-28-stacksave.ll
index 95a445c54357..48e876510a9d 100644
--- a/test/Transforms/InstCombine/2007-10-28-stacksave.ll
+++ b/test/Transforms/InstCombine/2007-10-28-stacksave.ll
@@ -23,7 +23,7 @@ lab:		; preds = %cleanup31, %entry
 	%tmp17 = mul i32 %tmp5, 4		; <i32> [#uses=1]
 	%tmp18 = alloca i8, i32 %tmp17		; <i8*> [#uses=1]
 	%tmp1819 = bitcast i8* %tmp18 to i32*		; <i32*> [#uses=2]
-	%tmp21 = getelementptr i32* %tmp1819, i32 0		; <i32*> [#uses=1]
+	%tmp21 = getelementptr i32, i32* %tmp1819, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp21, align 4
 	%tmp2223 = bitcast i32* %tmp1819 to i8*		; <i8*> [#uses=1]
 	store volatile i8* %tmp2223, i8** @p, align 4
diff --git a/test/Transforms/InstCombine/2007-10-31-RangeCrash.ll b/test/Transforms/InstCombine/2007-10-31-RangeCrash.ll
index 8105b4ba5579..8c2cb07af224 100644
--- a/test/Transforms/InstCombine/2007-10-31-RangeCrash.ll
+++ b/test/Transforms/InstCombine/2007-10-31-RangeCrash.ll
@@ -11,7 +11,7 @@ bb.i:		; preds = %entry
 	br label %bb51.i.i
 
 bb27.i.i:		; preds = %bb51.i.i
-	%tmp31.i.i = load i16* null, align 2		; <i16> [#uses=2]
+	%tmp31.i.i = load i16, i16* null, align 2		; <i16> [#uses=2]
 	%tmp35.i.i = icmp ult i16 %tmp31.i.i, 1		; <i1> [#uses=1]
 	%tmp41.i.i = icmp ugt i16 %tmp31.i.i, -1		; <i1> [#uses=1]
 	%bothcond.i.i = or i1 %tmp35.i.i, %tmp41.i.i		; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2007-10-31-StringCrash.ll b/test/Transforms/InstCombine/2007-10-31-StringCrash.ll
index 220f3e22b991..876cdd53f524 100644
--- a/test/Transforms/InstCombine/2007-10-31-StringCrash.ll
+++ b/test/Transforms/InstCombine/2007-10-31-StringCrash.ll
@@ -7,7 +7,7 @@ declare void @__darwin_gcc3_preregister_frame_info()
 define void @_start(i32 %argc, i8** %argv, i8** %envp) {
 entry:
 	%tmp1 = bitcast void ()* @__darwin_gcc3_preregister_frame_info to i32*		; <i32*> [#uses=1]
-	%tmp2 = load i32* %tmp1, align 4		; <i32> [#uses=1]
+	%tmp2 = load i32, i32* %tmp1, align 4		; <i32> [#uses=1]
 	%tmp3 = icmp ne i32 %tmp2, 0		; <i1> [#uses=1]
 	%tmp34 = zext i1 %tmp3 to i8		; <i8> [#uses=1]
 	%toBool = icmp ne i8 %tmp34, 0		; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll b/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll
index e6c9bcd01e5c..ff3107297a12 100644
--- a/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll
+++ b/test/Transforms/InstCombine/2007-11-07-OpaqueAlignCrash.ll
@@ -13,8 +13,8 @@ target triple = "i686-pc-linux-gnu"
 
 define i32 @foo() {
 entry:
-        %x = load i8* bitcast (%opaque_t* @g to i8*)
-        %y = load i32* bitcast (%op_ts* @h to i32*)
+        %x = load i8, i8* bitcast (%opaque_t* @g to i8*)
+        %y = load i32, i32* bitcast (%op_ts* @h to i32*)
 	%z = zext i8 %x to i32
 	%r = add i32 %y, %z
         ret i32 %r
diff --git a/test/Transforms/InstCombine/2007-11-25-CompatibleAttributes.ll b/test/Transforms/InstCombine/2007-11-25-CompatibleAttributes.ll
index e3192a990ba0..1232005a5610 100644
--- a/test/Transforms/InstCombine/2007-11-25-CompatibleAttributes.ll
+++ b/test/Transforms/InstCombine/2007-11-25-CompatibleAttributes.ll
@@ -5,7 +5,7 @@
 
 define i32 @main(i32 %argc, i8** %argv) {
 entry:
-	%tmp32 = tail call i32 (i8*  , ...) * bitcast (i32 (i8*, ...)  * @printf to i32 (i8*  , ...)  *)( i8* getelementptr ([4 x i8]* @.str, i32 0, i32 0)  , i32 0 ) nounwind 		; <i32> [#uses=0]
+	%tmp32 = tail call i32 (i8*  , ...) bitcast (i32 (i8*, ...)  * @printf to i32 (i8*  , ...)  *)( i8* getelementptr ([4 x i8], [4 x i8]* @.str, i32 0, i32 0)  , i32 0 ) nounwind 		; <i32> [#uses=0]
 	ret i32 undef
 }
 
diff --git a/test/Transforms/InstCombine/2007-12-12-GEPScale.ll b/test/Transforms/InstCombine/2007-12-12-GEPScale.ll
index cea87f2a4666..60f715e2ef7d 100644
--- a/test/Transforms/InstCombine/2007-12-12-GEPScale.ll
+++ b/test/Transforms/InstCombine/2007-12-12-GEPScale.ll
@@ -5,6 +5,6 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define i8* @foo([100 x {i8,i8,i8}]* %x) {
 entry:
         %p = bitcast [100 x {i8,i8,i8}]* %x to i8*
-        %q = getelementptr i8* %p, i32 -4
+        %q = getelementptr i8, i8* %p, i32 -4
         ret i8* %q
 }
diff --git a/test/Transforms/InstCombine/2007-12-28-IcmpSub2.ll b/test/Transforms/InstCombine/2007-12-28-IcmpSub2.ll
index 8721c83521bf..7260c001b0db 100644
--- a/test/Transforms/InstCombine/2007-12-28-IcmpSub2.ll
+++ b/test/Transforms/InstCombine/2007-12-28-IcmpSub2.ll
@@ -4,7 +4,7 @@ define i32 @test1() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp ule i32 %sub, 0
 	%retval = select i1 %cmp, i32 0, i32 1
@@ -15,7 +15,7 @@ define i32 @test2() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp ugt i32 %sub, 0
 	%retval = select i1 %cmp, i32 1, i32 0
@@ -26,7 +26,7 @@ define i32 @test3() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp slt i32 %sub, 0
 	%retval = select i1 %cmp, i32 1, i32 0
@@ -37,7 +37,7 @@ define i32 @test4() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp sle i32 %sub, 0
 	%retval = select i1 %cmp, i32 1, i32 0
@@ -48,7 +48,7 @@ define i32 @test5() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp sge i32 %sub, 0
 	%retval = select i1 %cmp, i32 0, i32 1
@@ -59,7 +59,7 @@ define i32 @test6() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp sgt i32 %sub, 0
 	%retval = select i1 %cmp, i32 0, i32 1
@@ -70,7 +70,7 @@ define i32 @test7() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp eq i32 %sub, 0
 	%retval = select i1 %cmp, i32 0, i32 1
@@ -81,7 +81,7 @@ define i32 @test8() {
 entry:
 	%z = alloca i32
 	store i32 0, i32* %z
-	%tmp = load i32* %z
+	%tmp = load i32, i32* %z
 	%sub = sub i32 %tmp, 1
 	%cmp = icmp ne i32 %sub, 0
 	%retval = select i1 %cmp, i32 1, i32 0
diff --git a/test/Transforms/InstCombine/2008-01-14-VarArgTrampoline.ll b/test/Transforms/InstCombine/2008-01-14-VarArgTrampoline.ll
index aacea9df5b31..b111b85d59f0 100644
--- a/test/Transforms/InstCombine/2008-01-14-VarArgTrampoline.ll
+++ b/test/Transforms/InstCombine/2008-01-14-VarArgTrampoline.ll
@@ -12,15 +12,15 @@ define i32 @nest(i32 %n) {
 entry:
 	%FRAME.0 = alloca %struct.FRAME.nest, align 8		; <%struct.FRAME.nest*> [#uses=3]
 	%TRAMP.216 = alloca [10 x i8], align 16		; <[10 x i8]*> [#uses=1]
-	%TRAMP.216.sub = getelementptr [10 x i8]* %TRAMP.216, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp3 = getelementptr %struct.FRAME.nest* %FRAME.0, i32 0, i32 0		; <i32*> [#uses=1]
+	%TRAMP.216.sub = getelementptr [10 x i8], [10 x i8]* %TRAMP.216, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp3 = getelementptr %struct.FRAME.nest, %struct.FRAME.nest* %FRAME.0, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 %n, i32* %tmp3, align 8
 	%FRAME.06 = bitcast %struct.FRAME.nest* %FRAME.0 to i8*		; <i8*> [#uses=1]
 	call void @llvm.init.trampoline( i8* %TRAMP.216.sub, i8* bitcast (i32 (%struct.FRAME.nest*, ...)* @f to i8*), i8* %FRAME.06 )		; <i8*> [#uses=1]
         %tramp = call i8* @llvm.adjust.trampoline( i8* %TRAMP.216.sub)
-	%tmp7 = getelementptr %struct.FRAME.nest* %FRAME.0, i32 0, i32 1		; <i32 (...)**> [#uses=1]
+	%tmp7 = getelementptr %struct.FRAME.nest, %struct.FRAME.nest* %FRAME.0, i32 0, i32 1		; <i32 (...)**> [#uses=1]
 	%tmp89 = bitcast i8* %tramp to i32 (...)*		; <i32 (...)*> [#uses=2]
 	store i32 (...)* %tmp89, i32 (...)** %tmp7, align 8
-	%tmp2.i = call i32 (...)* %tmp89( i32 zeroext 0 )		; <i32> [#uses=1]
+	%tmp2.i = call i32 (...) %tmp89( i32 zeroext 0 )		; <i32> [#uses=1]
 	ret i32 %tmp2.i
 }
diff --git a/test/Transforms/InstCombine/2008-03-13-IntToPtr.ll b/test/Transforms/InstCombine/2008-03-13-IntToPtr.ll
index 0fa4d715f289..d086f4b63b71 100644
--- a/test/Transforms/InstCombine/2008-03-13-IntToPtr.ll
+++ b/test/Transforms/InstCombine/2008-03-13-IntToPtr.ll
@@ -1,7 +1,7 @@
 ; RUN: opt < %s -instcombine -S | grep "16" | count 1
 
 define i8* @bork(i8** %qux) {
-  %tmp275 = load i8** %qux, align 1
+  %tmp275 = load i8*, i8** %qux, align 1
   %tmp275276 = ptrtoint i8* %tmp275 to i32
   %tmp277 = add i32 %tmp275276, 16
   %tmp277278 = inttoptr i32 %tmp277 to i8*
diff --git a/test/Transforms/InstCombine/2008-04-22-ByValBitcast.ll b/test/Transforms/InstCombine/2008-04-22-ByValBitcast.ll
index aa38065a7149..1ea0998bf70e 100644
--- a/test/Transforms/InstCombine/2008-04-22-ByValBitcast.ll
+++ b/test/Transforms/InstCombine/2008-04-22-ByValBitcast.ll
@@ -8,7 +8,7 @@ target triple = "i386-apple-darwin9"
 define void @foo(i8* %context) nounwind  {
 entry:
 	%tmp1 = bitcast i8* %context to %struct.NSRect*		; <%struct.NSRect*> [#uses=1]
-	call void (i32, ...)* @bar( i32 3, %struct.NSRect* byval align 4  %tmp1 ) nounwind 
+	call void (i32, ...) @bar( i32 3, %struct.NSRect* byval align 4  %tmp1 ) nounwind 
 	ret void
 }
 
diff --git a/test/Transforms/InstCombine/2008-04-29-VolatileLoadDontMerge.ll b/test/Transforms/InstCombine/2008-04-29-VolatileLoadDontMerge.ll
index fd0217e9f4d7..af662bda1e6e 100644
--- a/test/Transforms/InstCombine/2008-04-29-VolatileLoadDontMerge.ll
+++ b/test/Transforms/InstCombine/2008-04-29-VolatileLoadDontMerge.ll
@@ -6,7 +6,7 @@ target triple = "i386-apple-darwin8"
 define i32 @main() nounwind  {
 entry:
 	%tmp93 = icmp slt i32 0, 10		; <i1> [#uses=0]
-	%tmp34 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp34 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br label %bb
 
 bb:		; preds = %bb, %entry
@@ -16,7 +16,7 @@ bb:		; preds = %bb, %entry
 	store volatile i32 %tmp4, i32* @g_1, align 4
 	%tmp6 = add i32 %b.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%tmp9 = icmp slt i32 %tmp6, 10		; <i1> [#uses=1]
-	%tmp3 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp3 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br i1 %tmp9, label %bb, label %bb11
 
 bb11:		; preds = %bb
diff --git a/test/Transforms/InstCombine/2008-04-29-VolatileLoadMerge.ll b/test/Transforms/InstCombine/2008-04-29-VolatileLoadMerge.ll
index 8022414d6fc9..3c67e513bcc1 100644
--- a/test/Transforms/InstCombine/2008-04-29-VolatileLoadMerge.ll
+++ b/test/Transforms/InstCombine/2008-04-29-VolatileLoadMerge.ll
@@ -7,11 +7,11 @@ target triple = "i386-apple-darwin8"
 define i32 @main(i32 %i) nounwind  {
 entry:
 	%tmp93 = icmp slt i32 %i, 10		; <i1> [#uses=0]
-	%tmp34 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp34 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br i1 %tmp93, label %bb11, label %bb
 
 bb:		; preds = %bb, %entry
-	%tmp3 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp3 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br label %bb11
 
 bb11:		; preds = %bb
diff --git a/test/Transforms/InstCombine/2008-05-08-LiveStoreDelete.ll b/test/Transforms/InstCombine/2008-05-08-LiveStoreDelete.ll
index 7a1c84474153..907382093d11 100644
--- a/test/Transforms/InstCombine/2008-05-08-LiveStoreDelete.ll
+++ b/test/Transforms/InstCombine/2008-05-08-LiveStoreDelete.ll
@@ -7,14 +7,14 @@ define i32 @a() nounwind  {
 entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	%tmp1 = call i8* @malloc( i32 10 ) nounwind 		; <i8*> [#uses=5]
-	%tmp3 = getelementptr i8* %tmp1, i32 1		; <i8*> [#uses=1]
+	%tmp3 = getelementptr i8, i8* %tmp1, i32 1		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp3, align 1
-	%tmp5 = getelementptr i8* %tmp1, i32 0		; <i8*> [#uses=1]
+	%tmp5 = getelementptr i8, i8* %tmp1, i32 0		; <i8*> [#uses=1]
 	store i8 1, i8* %tmp5, align 1
 	%tmp7 = call i32 @strlen( i8* %tmp1 ) nounwind readonly 		; <i32> [#uses=1]
-	%tmp9 = getelementptr i8* %tmp1, i32 0		; <i8*> [#uses=1]
+	%tmp9 = getelementptr i8, i8* %tmp1, i32 0		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp9, align 1
-	%tmp11 = call i32 (...)* @b( i8* %tmp1 ) nounwind 		; <i32> [#uses=0]
+	%tmp11 = call i32 (...) @b( i8* %tmp1 ) nounwind 		; <i32> [#uses=0]
 	ret i32 %tmp7
 }
 
diff --git a/test/Transforms/InstCombine/2008-05-08-StrLenSink.ll b/test/Transforms/InstCombine/2008-05-08-StrLenSink.ll
index d266164fd870..ce1923338c6b 100644
--- a/test/Transforms/InstCombine/2008-05-08-StrLenSink.ll
+++ b/test/Transforms/InstCombine/2008-05-08-StrLenSink.ll
@@ -7,18 +7,18 @@ define i32 @a() nounwind  {
 entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	%tmp1 = call i8* @malloc( i32 10 ) nounwind 		; <i8*> [#uses=5]
-	%tmp3 = getelementptr i8* %tmp1, i32 1		; <i8*> [#uses=1]
+	%tmp3 = getelementptr i8, i8* %tmp1, i32 1		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp3, align 1
-	%tmp5 = getelementptr i8* %tmp1, i32 0		; <i8*> [#uses=1]
+	%tmp5 = getelementptr i8, i8* %tmp1, i32 0		; <i8*> [#uses=1]
 	store i8 1, i8* %tmp5, align 1
 ; CHECK: store
 ; CHECK: store
 ; CHECK-NEXT: strlen
 ; CHECK-NEXT: store
 	%tmp7 = call i32 @strlen( i8* %tmp1 ) nounwind readonly 		; <i32> [#uses=1]
-	%tmp9 = getelementptr i8* %tmp1, i32 0		; <i8*> [#uses=1]
+	%tmp9 = getelementptr i8, i8* %tmp1, i32 0		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp9, align 1
-	%tmp11 = call i32 (...)* @b( i8* %tmp1 ) nounwind 		; <i32> [#uses=0]
+	%tmp11 = call i32 (...) @b( i8* %tmp1 ) nounwind 		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
diff --git a/test/Transforms/InstCombine/2008-05-09-SinkOfInvoke.ll b/test/Transforms/InstCombine/2008-05-09-SinkOfInvoke.ll
index f6eb24843fb8..722f8f0fabbd 100644
--- a/test/Transforms/InstCombine/2008-05-09-SinkOfInvoke.ll
+++ b/test/Transforms/InstCombine/2008-05-09-SinkOfInvoke.ll
@@ -23,8 +23,8 @@ invcont31:		; preds = %invcont
 			to label %invcont37 unwind label %lpad		; <i32**> [#uses=1]
 
 invcont37:		; preds = %invcont31
-	%tmp39 = load i32** %tmp38, align 8		; <i32*> [#uses=1]
-	%tmp41 = getelementptr %"struct.std::ctype<char>"* %this, i32 0, i32 4		; <i32**> [#uses=1]
+	%tmp39 = load i32*, i32** %tmp38, align 8		; <i32*> [#uses=1]
+	%tmp41 = getelementptr %"struct.std::ctype<char>", %"struct.std::ctype<char>"* %this, i32 0, i32 4		; <i32**> [#uses=1]
 	store i32* %tmp39, i32** %tmp41, align 8
 	ret void
 
diff --git a/test/Transforms/InstCombine/2008-05-17-InfLoop.ll b/test/Transforms/InstCombine/2008-05-17-InfLoop.ll
index 2939a4821654..af0f2a45e7ec 100644
--- a/test/Transforms/InstCombine/2008-05-17-InfLoop.ll
+++ b/test/Transforms/InstCombine/2008-05-17-InfLoop.ll
@@ -10,11 +10,11 @@ entry:
 	%blockSize100k = alloca i32		; <i32*> [#uses=2]
 	store i32 %0, i32* %blockSize100k
 	%n = alloca i32		; <i32*> [#uses=2]
-	load i32* %blockSize100k		; <i32>:1 [#uses=1]
+	load i32, i32* %blockSize100k		; <i32>:1 [#uses=1]
 	store i32 %1, i32* %n
-	load i32* %n		; <i32>:2 [#uses=1]
+	load i32, i32* %n		; <i32>:2 [#uses=1]
 	add i32 %2, 2		; <i32>:3 [#uses=1]
-	mul i32 %3, ptrtoint (i32* getelementptr (i32* null, i32 1) to i32)		; <i32>:4 [#uses=1]
+	mul i32 %3, ptrtoint (i32* getelementptr (i32, i32* null, i32 1) to i32)		; <i32>:4 [#uses=1]
 	call void @BZALLOC( i32 %4 )
 	br label %return
 
diff --git a/test/Transforms/InstCombine/2008-05-23-CompareFold.ll b/test/Transforms/InstCombine/2008-05-23-CompareFold.ll
index b72967767558..b10aac96599f 100644
--- a/test/Transforms/InstCombine/2008-05-23-CompareFold.ll
+++ b/test/Transforms/InstCombine/2008-05-23-CompareFold.ll
@@ -5,7 +5,7 @@
 ; CHECK: ret i1 false
 define i1 @f(i8* %x) {
 entry:
-       %tmp462 = load i8* %x, align 1          ; <i8> [#uses=1]
+       %tmp462 = load i8, i8* %x, align 1          ; <i8> [#uses=1]
        %tmp462463 = sitofp i8 %tmp462 to float         ; <float> [#uses=1]
        %tmp464 = fcmp ugt float %tmp462463, 0x47EFFFFFE0000000         ; <i1>
        ret i1 %tmp464
diff --git a/test/Transforms/InstCombine/2008-06-08-ICmpPHI.ll b/test/Transforms/InstCombine/2008-06-08-ICmpPHI.ll
index ec946238d897..7e8341b99f78 100644
--- a/test/Transforms/InstCombine/2008-06-08-ICmpPHI.ll
+++ b/test/Transforms/InstCombine/2008-06-08-ICmpPHI.ll
@@ -7,28 +7,28 @@ entry:
 bb:		; preds = %bb16, %entry
 	%i.0 = phi i32 [ 0, %entry ], [ %indvar.next, %somebb ]		; <i32> [#uses=1]
 	%x.0 = phi i32 [ 37, %entry ], [ %tmp17, %somebb ]		; <i32> [#uses=1]
-	%tmp = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=0]
-	%tmp1 = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=0]
-	%tmp2 = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=1]
+	%tmp = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=0]
+	%tmp1 = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=0]
+	%tmp2 = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=1]
 	%tmp3 = icmp eq i32 %tmp2, 0		; <i1> [#uses=1]
 	br i1 %tmp3, label %bb7, label %bb5
 
 bb5:		; preds = %bb
-	%tmp6 = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=0]
+	%tmp6 = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=0]
 	br label %bb7
 
 bb7:		; preds = %bb5, %bb
-	%tmp8 = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=0]
-	%tmp9 = tail call i32 (...)* @bork( ) nounwind 		; <i32> [#uses=0]
+	%tmp8 = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=0]
+	%tmp9 = tail call i32 (...) @bork( ) nounwind 		; <i32> [#uses=0]
 	%tmp11 = icmp eq i32 %x.0, 37		; <i1> [#uses=1]
 	br i1 %tmp11, label %bb14, label %bb16
 
 bb14:		; preds = %bb7
-	%tmp15 = tail call i32 (...)* @bar( ) nounwind 		; <i32> [#uses=0]
+	%tmp15 = tail call i32 (...) @bar( ) nounwind 		; <i32> [#uses=0]
 	br label %bb16
 
 bb16:		; preds = %bb14, %bb7
-	%tmp17 = tail call i32 (...)* @zap( ) nounwind 		; <i32> [#uses=1]
+	%tmp17 = tail call i32 (...) @zap( ) nounwind 		; <i32> [#uses=1]
 	%indvar.next = add i32 %i.0, 1		; <i32> [#uses=2]
 	%exitcond = icmp eq i32 %indvar.next, 42		; <i1> [#uses=1]
 	br i1 %exitcond, label %return, label %somebb
diff --git a/test/Transforms/InstCombine/2008-06-19-UncondLoad.ll b/test/Transforms/InstCombine/2008-06-19-UncondLoad.ll
index 05f1c5200081..c3aab464b87e 100644
--- a/test/Transforms/InstCombine/2008-06-19-UncondLoad.ll
+++ b/test/Transforms/InstCombine/2008-06-19-UncondLoad.ll
@@ -4,13 +4,13 @@
 declare i32 @x(i32*)
 define i32 @b(i32* %a, i32* %b) {
 entry:
-        %tmp1 = load i32* %a            
-        %tmp3 = load i32* %b           
+        %tmp1 = load i32, i32* %a            
+        %tmp3 = load i32, i32* %b           
         %add = add i32 %tmp1, %tmp3   
         %call = call i32 @x( i32* %a )
         %tobool = icmp ne i32 %add, 0
 	; not safe to turn into an uncond load
         %cond = select i1 %tobool, i32* %b, i32* %a             
-        %tmp8 = load i32* %cond       
+        %tmp8 = load i32, i32* %cond       
         ret i32 %tmp8
 }
diff --git a/test/Transforms/InstCombine/2008-06-24-StackRestore.ll b/test/Transforms/InstCombine/2008-06-24-StackRestore.ll
index 9c4c1b538c4b..f963b009671b 100644
--- a/test/Transforms/InstCombine/2008-06-24-StackRestore.ll
+++ b/test/Transforms/InstCombine/2008-06-24-StackRestore.ll
@@ -27,7 +27,7 @@ bb44:		; preds = %bb44, %entry
 	%tmp27 = alloca i32, i32 %tmp5		; <i32*> [#uses=3]
 	%tmpcast = bitcast i32* %tmp27 to i8*		; <i8*> [#uses=1]
 	store i32 1, i32* %tmp27, align 4
-	%tmp34 = getelementptr i32* %tmp27, i32 %tmp4		; <i32*> [#uses=1]
+	%tmp34 = getelementptr i32, i32* %tmp27, i32 %tmp4		; <i32*> [#uses=1]
 	store i32 2, i32* %tmp34, align 4
 	store volatile i8* %tmpcast, i8** @p, align 4
 	%exitcond = icmp eq i32 %tmp3857, 999999		; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2008-07-08-VolatileLoadMerge.ll b/test/Transforms/InstCombine/2008-07-08-VolatileLoadMerge.ll
index dcf4befa8605..17ec9cd1d82b 100644
--- a/test/Transforms/InstCombine/2008-07-08-VolatileLoadMerge.ll
+++ b/test/Transforms/InstCombine/2008-07-08-VolatileLoadMerge.ll
@@ -7,7 +7,7 @@ target triple = "i386-apple-darwin8"
 define i32 @main() nounwind  {
 entry:
 	%tmp93 = icmp slt i32 0, 10		; <i1> [#uses=0]
-	%tmp34 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp34 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br label %bb
 
 bb:		; preds = %bb, %entry
@@ -17,7 +17,7 @@ bb:		; preds = %bb, %entry
 	store volatile i32 %tmp4, i32* @g_1, align 4
 	%tmp6 = add i32 %b.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%tmp9 = icmp slt i32 %tmp6, 10		; <i1> [#uses=1]
-	%tmp3 = load volatile i32* @g_1, align 4		; <i32> [#uses=1]
+	%tmp3 = load volatile i32, i32* @g_1, align 4		; <i32> [#uses=1]
 	br i1 %tmp9, label %bb, label %bb11
 
 bb11:		; preds = %bb
diff --git a/test/Transforms/InstCombine/2008-08-05-And.ll b/test/Transforms/InstCombine/2008-08-05-And.ll
index 9773c2d76249..91f1c0b0a986 100644
--- a/test/Transforms/InstCombine/2008-08-05-And.ll
+++ b/test/Transforms/InstCombine/2008-08-05-And.ll
@@ -6,8 +6,8 @@ entry:
         br label %bb
 
 bb:
-	%g1 = getelementptr i8* %x, i32 0
-        %l1 = load i8* %g1, align 1
+	%g1 = getelementptr i8, i8* %x, i32 0
+        %l1 = load i8, i8* %g1, align 1
 	%s1 = sub i8 %l1, 6
 	%c1 = icmp ugt i8 %s1, 2
 	%s2 = sub i8 %l1, 10
diff --git a/test/Transforms/InstCombine/2009-01-08-AlignAlloca.ll b/test/Transforms/InstCombine/2009-01-08-AlignAlloca.ll
index 50ea2f42b0ca..9994b588b21c 100644
--- a/test/Transforms/InstCombine/2009-01-08-AlignAlloca.ll
+++ b/test/Transforms/InstCombine/2009-01-08-AlignAlloca.ll
@@ -12,16 +12,16 @@ entry:
 	%iospec = alloca %struct.Key		; <%struct.Key*> [#uses=3]
 	%ret = alloca i32		; <i32*> [#uses=2]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%0 = getelementptr %struct.Key* %iospec, i32 0, i32 0		; <{ i32, i32 }*> [#uses=2]
-	%1 = getelementptr { i32, i32 }* %0, i32 0, i32 0		; <i32*> [#uses=1]
+	%0 = getelementptr %struct.Key, %struct.Key* %iospec, i32 0, i32 0		; <{ i32, i32 }*> [#uses=2]
+	%1 = getelementptr { i32, i32 }, { i32, i32 }* %0, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %1, align 4
-	%2 = getelementptr { i32, i32 }* %0, i32 0, i32 1		; <i32*> [#uses=1]
+	%2 = getelementptr { i32, i32 }, { i32, i32 }* %0, i32 0, i32 1		; <i32*> [#uses=1]
 	store i32 0, i32* %2, align 4
-	%3 = getelementptr %struct.Key* %iospec, i32 0, i32 0		; <{ i32, i32 }*> [#uses=1]
+	%3 = getelementptr %struct.Key, %struct.Key* %iospec, i32 0, i32 0		; <{ i32, i32 }*> [#uses=1]
 	%4 = bitcast { i32, i32 }* %3 to i64*		; <i64*> [#uses=1]
 	store i64 %key_token2, i64* %4, align 4
-	%5 = call i32 (...)* @foo(%struct.Key* byval align 4 %iospec, i32* %ret) nounwind		; <i32> [#uses=0]
-	%6 = load i32* %ret, align 4		; <i32> [#uses=1]
+	%5 = call i32 (...) @foo(%struct.Key* byval align 4 %iospec, i32* %ret) nounwind		; <i32> [#uses=0]
+	%6 = load i32, i32* %ret, align 4		; <i32> [#uses=1]
 	ret i32 %6
 }
 
diff --git a/test/Transforms/InstCombine/2009-01-19-fmod-constant-float-specials.ll b/test/Transforms/InstCombine/2009-01-19-fmod-constant-float-specials.ll
index 68c51b43fc3a..b9aa0a254972 100644
--- a/test/Transforms/InstCombine/2009-01-19-fmod-constant-float-specials.ll
+++ b/test/Transforms/InstCombine/2009-01-19-fmod-constant-float-specials.ll
@@ -15,12 +15,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF0000000000000, float* %x, align 4
 	store float 0x7FF8000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -36,12 +36,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF0000000000000, float* %x, align 4
 	store float 0.000000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -55,12 +55,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF0000000000000, float* %x, align 4
 	store float 3.500000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -74,12 +74,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF0000000000000, float* %x, align 4
 	store float 0x7FF0000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -93,12 +93,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF8000000000000, float* %x, align 4
 	store float 0x7FF0000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -112,12 +112,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF8000000000000, float* %x, align 4
 	store float 0.000000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -131,12 +131,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF8000000000000, float* %x, align 4
 	store float 3.500000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -150,12 +150,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0x7FF8000000000000, float* %x, align 4
 	store float 0x7FF8000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -169,12 +169,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0.000000e+00, float* %x, align 4
 	store float 0x7FF8000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -188,12 +188,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0.000000e+00, float* %x, align 4
 	store float 0x7FF0000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -207,12 +207,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0.000000e+00, float* %x, align 4
 	store float 0.000000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -226,12 +226,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 0.000000e+00, float* %x, align 4
 	store float 3.500000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -245,12 +245,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 3.500000e+00, float* %x, align 4
 	store float 0x7FF8000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -264,12 +264,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 3.500000e+00, float* %x, align 4
 	store float 0x7FF0000000000000, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -283,12 +283,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 3.500000e+00, float* %x, align 4
 	store float 0.000000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
@@ -302,12 +302,12 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store float 3.500000e+00, float* %x, align 4
 	store float 3.500000e+00, float* %y, align 4
-	%0 = load float* %y, align 4		; <float> [#uses=1]
+	%0 = load float, float* %y, align 4		; <float> [#uses=1]
 	%1 = fpext float %0 to double		; <double> [#uses=1]
-	%2 = load float* %x, align 4		; <float> [#uses=1]
+	%2 = load float, float* %x, align 4		; <float> [#uses=1]
 	%3 = fpext float %2 to double		; <double> [#uses=1]
 	%4 = frem double %3, %1		; <double> [#uses=1]
-	%5 = call i32 (i8*, ...)* @printf(i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
+	%5 = call i32 (i8*, ...) @printf(i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), double %4) nounwind		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
diff --git a/test/Transforms/InstCombine/2009-01-19-fmod-constant-float.ll b/test/Transforms/InstCombine/2009-01-19-fmod-constant-float.ll
index 6bc7ce34d1c2..5adcb6bfa071 100644
--- a/test/Transforms/InstCombine/2009-01-19-fmod-constant-float.ll
+++ b/test/Transforms/InstCombine/2009-01-19-fmod-constant-float.ll
@@ -14,12 +14,12 @@ entry:
 	%1 = frem double 1.000000e-01, 1.000000e+00	; <double> [#uses=1]
 	%2 = fptrunc double %1 to float		; <float> [#uses=1]
 	store float %2, float* %0, align 4
-	%3 = load float* %0, align 4		; <float> [#uses=1]
+	%3 = load float, float* %0, align 4		; <float> [#uses=1]
 	store float %3, float* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load float* %retval		; <float> [#uses=1]
+	%retval1 = load float, float* %retval		; <float> [#uses=1]
 	ret float %retval1
 }
 
@@ -31,12 +31,12 @@ entry:
 	%1 = frem double -1.000000e-01, 1.000000e+00	; <double> [#uses=1]
 	%2 = fptrunc double %1 to float		; <float> [#uses=1]
 	store float %2, float* %0, align 4
-	%3 = load float* %0, align 4		; <float> [#uses=1]
+	%3 = load float, float* %0, align 4		; <float> [#uses=1]
 	store float %3, float* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load float* %retval		; <float> [#uses=1]
+	%retval1 = load float, float* %retval		; <float> [#uses=1]
 	ret float %retval1
 }
 
@@ -48,12 +48,12 @@ entry:
 	%1 = frem double 1.000000e-01, -1.000000e+00	; <double> [#uses=1]
 	%2 = fptrunc double %1 to float		; <float> [#uses=1]
 	store float %2, float* %0, align 4
-	%3 = load float* %0, align 4		; <float> [#uses=1]
+	%3 = load float, float* %0, align 4		; <float> [#uses=1]
 	store float %3, float* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load float* %retval		; <float> [#uses=1]
+	%retval1 = load float, float* %retval		; <float> [#uses=1]
 	ret float %retval1
 }
 
@@ -65,11 +65,11 @@ entry:
 	%1 = frem double -1.000000e-01, -1.000000e+00	; <double> [#uses=1]
 	%2 = fptrunc double %1 to float		; <float> [#uses=1]
 	store float %2, float* %0, align 4
-	%3 = load float* %0, align 4		; <float> [#uses=1]
+	%3 = load float, float* %0, align 4		; <float> [#uses=1]
 	store float %3, float* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load float* %retval		; <float> [#uses=1]
+	%retval1 = load float, float* %retval		; <float> [#uses=1]
 	ret float %retval1
 }
diff --git a/test/Transforms/InstCombine/2009-01-24-EmptyStruct.ll b/test/Transforms/InstCombine/2009-01-24-EmptyStruct.ll
index 4b64b4872918..3f3535b363d7 100644
--- a/test/Transforms/InstCombine/2009-01-24-EmptyStruct.ll
+++ b/test/Transforms/InstCombine/2009-01-24-EmptyStruct.ll
@@ -11,7 +11,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @handle_event(%struct.inode* %bar) nounwind {
 entry:
-	%0 = getelementptr %struct.inode* %bar, i64 -1, i32 1, i32 1		; <%struct.rwlock_t*> [#uses=1]
+	%0 = getelementptr %struct.inode, %struct.inode* %bar, i64 -1, i32 1, i32 1		; <%struct.rwlock_t*> [#uses=1]
 	%1 = bitcast %struct.rwlock_t* %0 to i32*		; <i32*> [#uses=1]
 	store i32 1, i32* %1, align 4
 	ret void
diff --git a/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll b/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
index 9146a8ee0117..0f8b38c8e9c1 100644
--- a/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
+++ b/test/Transforms/InstCombine/2009-02-20-InstCombine-SROA.ll
@@ -27,53 +27,53 @@ entry:
   %4 = alloca i32
   %"alloca point" = bitcast i32 0 to i32
   store i32 42, i32* %4, align 4
-  %5 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
-  %6 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %5, i32 0, i32 0
-  %7 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %6, i32 0, i32 1
-  %8 = load i32** %7, align 4
-  %9 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
+  %5 = getelementptr %"struct.std::vector<int,std::allocator<int> >", %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
+  %6 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >", %"struct.std::_Vector_base<int,std::allocator<int> >"* %5, i32 0, i32 0
+  %7 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl", %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %6, i32 0, i32 1
+  %8 = load i32*, i32** %7, align 4
+  %9 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
   store i32* %8, i32** %9, align 4
-  %10 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
-  %11 = load i32** %10, align 4
+  %10 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %3, i32 0, i32 0
+  %11 = load i32*, i32** %10, align 4
   %tmp2.i = ptrtoint i32* %11 to i32
   %tmp1.i = inttoptr i32 %tmp2.i to i32*
   %tmp3 = ptrtoint i32* %tmp1.i to i32
   %tmp2 = inttoptr i32 %tmp3 to i32*
-  %12 = getelementptr %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
-  %13 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >"* %12, i32 0, i32 0
-  %14 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %13, i32 0, i32 0
-  %15 = load i32** %14, align 4
-  %16 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
+  %12 = getelementptr %"struct.std::vector<int,std::allocator<int> >", %"struct.std::vector<int,std::allocator<int> >"* %X, i32 0, i32 0
+  %13 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >", %"struct.std::_Vector_base<int,std::allocator<int> >"* %12, i32 0, i32 0
+  %14 = getelementptr %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl", %"struct.std::_Vector_base<int,std::allocator<int> >::_Vector_impl"* %13, i32 0, i32 0
+  %15 = load i32*, i32** %14, align 4
+  %16 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
   store i32* %15, i32** %16, align 4
-  %17 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
-  %18 = load i32** %17, align 4
+  %17 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %0, i32 0, i32 0
+  %18 = load i32*, i32** %17, align 4
   %tmp2.i17 = ptrtoint i32* %18 to i32
   %tmp1.i18 = inttoptr i32 %tmp2.i17 to i32*
   %tmp8 = ptrtoint i32* %tmp1.i18 to i32
   %tmp6 = inttoptr i32 %tmp8 to i32*
-  %19 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
+  %19 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
   store i32* %tmp6, i32** %19
-  %20 = getelementptr %"struct.std::bidirectional_iterator_tag"* %1, i32 0, i32 0
-  %21 = load i8* %20, align 1
+  %20 = getelementptr %"struct.std::bidirectional_iterator_tag", %"struct.std::bidirectional_iterator_tag"* %1, i32 0, i32 0
+  %21 = load i8, i8* %20, align 1
   %22 = or i8 %21, 0
   %23 = or i8 %22, 0
   %24 = or i8 %23, 0
-  %25 = getelementptr %"struct.std::bidirectional_iterator_tag"* %2, i32 0, i32 0
+  %25 = getelementptr %"struct.std::bidirectional_iterator_tag", %"struct.std::bidirectional_iterator_tag"* %2, i32 0, i32 0
   store i8 0, i8* %25, align 1
-  %elt.i = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
-  %val.i = load i32** %elt.i
+  %elt.i = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i, i32 0, i32 0
+  %val.i = load i32*, i32** %elt.i
   %tmp.i = bitcast %"struct.std::bidirectional_iterator_tag"* %unnamed_arg.i to i8*
   %tmp9.i = bitcast %"struct.std::bidirectional_iterator_tag"* %2 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp.i, i8* %tmp9.i, i64 1, i32 1, i1 false)
-  %26 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %26 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %val.i, i32** %26
-  %27 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %27 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
   store i32* %tmp2, i32** %27
-  %28 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
-  %29 = load i32** %28, align 4
+  %28 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %29 = load i32*, i32** %28, align 4
   %30 = ptrtoint i32* %29 to i32
-  %31 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %32 = load i32** %31, align 4
+  %31 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %32 = load i32*, i32** %31, align 4
   %33 = ptrtoint i32* %32 to i32
   %34 = sub i32 %30, %33
   %35 = ashr i32 %34, 2
@@ -81,85 +81,85 @@ entry:
   br label %bb12.i.i
 
 bb.i.i:                                           ; preds = %bb12.i.i
-  %37 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %38 = load i32** %37, align 4
-  %39 = load i32* %38, align 4
-  %40 = load i32* %4, align 4
+  %37 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %38 = load i32*, i32** %37, align 4
+  %39 = load i32, i32* %38, align 4
+  %40 = load i32, i32* %4, align 4
   %41 = icmp eq i32 %39, %40
   %42 = zext i1 %41 to i8
   %toBool.i.i = icmp ne i8 %42, 0
   br i1 %toBool.i.i, label %bb1.i.i, label %bb2.i.i
 
 bb1.i.i:                                          ; preds = %bb.i.i
-  %43 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %44 = load i32** %43, align 4
+  %43 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %44 = load i32*, i32** %43, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb2.i.i:                                          ; preds = %bb.i.i
-  %45 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %46 = load i32** %45, align 4
-  %47 = getelementptr i32* %46, i64 1
-  %48 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %45 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %46 = load i32*, i32** %45, align 4
+  %47 = getelementptr i32, i32* %46, i64 1
+  %48 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %47, i32** %48, align 4
-  %49 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %50 = load i32** %49, align 4
-  %51 = load i32* %50, align 4
-  %52 = load i32* %4, align 4
+  %49 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %50 = load i32*, i32** %49, align 4
+  %51 = load i32, i32* %50, align 4
+  %52 = load i32, i32* %4, align 4
   %53 = icmp eq i32 %51, %52
   %54 = zext i1 %53 to i8
   %toBool3.i.i = icmp ne i8 %54, 0
   br i1 %toBool3.i.i, label %bb4.i.i, label %bb5.i.i
 
 bb4.i.i:                                          ; preds = %bb2.i.i
-  %55 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %56 = load i32** %55, align 4
+  %55 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %56 = load i32*, i32** %55, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb5.i.i:                                          ; preds = %bb2.i.i
-  %57 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %58 = load i32** %57, align 4
-  %59 = getelementptr i32* %58, i64 1
-  %60 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %57 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %58 = load i32*, i32** %57, align 4
+  %59 = getelementptr i32, i32* %58, i64 1
+  %60 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %59, i32** %60, align 4
-  %61 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %62 = load i32** %61, align 4
-  %63 = load i32* %62, align 4
-  %64 = load i32* %4, align 4
+  %61 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %62 = load i32*, i32** %61, align 4
+  %63 = load i32, i32* %62, align 4
+  %64 = load i32, i32* %4, align 4
   %65 = icmp eq i32 %63, %64
   %66 = zext i1 %65 to i8
   %toBool6.i.i = icmp ne i8 %66, 0
   br i1 %toBool6.i.i, label %bb7.i.i, label %bb8.i.i
 
 bb7.i.i:                                          ; preds = %bb5.i.i
-  %67 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %68 = load i32** %67, align 4
+  %67 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %68 = load i32*, i32** %67, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb8.i.i:                                          ; preds = %bb5.i.i
-  %69 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %70 = load i32** %69, align 4
-  %71 = getelementptr i32* %70, i64 1
-  %72 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %69 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %70 = load i32*, i32** %69, align 4
+  %71 = getelementptr i32, i32* %70, i64 1
+  %72 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %71, i32** %72, align 4
-  %73 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %74 = load i32** %73, align 4
-  %75 = load i32* %74, align 4
-  %76 = load i32* %4, align 4
+  %73 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %74 = load i32*, i32** %73, align 4
+  %75 = load i32, i32* %74, align 4
+  %76 = load i32, i32* %4, align 4
   %77 = icmp eq i32 %75, %76
   %78 = zext i1 %77 to i8
   %toBool9.i.i = icmp ne i8 %78, 0
   br i1 %toBool9.i.i, label %bb10.i.i, label %bb11.i.i
 
 bb10.i.i:                                         ; preds = %bb8.i.i
-  %79 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %80 = load i32** %79, align 4
+  %79 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %80 = load i32*, i32** %79, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb11.i.i:                                         ; preds = %bb8.i.i
-  %81 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %82 = load i32** %81, align 4
-  %83 = getelementptr i32* %82, i64 1
-  %84 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %81 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %82 = load i32*, i32** %81, align 4
+  %83 = getelementptr i32, i32* %82, i64 1
+  %84 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %83, i32** %84, align 4
   %85 = sub i32 %__trip_count.0.i.i, 1
   br label %bb12.i.i
@@ -170,11 +170,11 @@ bb12.i.i:                                         ; preds = %bb11.i.i, %entry
   br i1 %86, label %bb.i.i, label %bb13.i.i
 
 bb13.i.i:                                         ; preds = %bb12.i.i
-  %87 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
-  %88 = load i32** %87, align 4
+  %87 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %88 = load i32*, i32** %87, align 4
   %89 = ptrtoint i32* %88 to i32
-  %90 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %91 = load i32** %90, align 4
+  %90 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %91 = load i32*, i32** %90, align 4
   %92 = ptrtoint i32* %91 to i32
   %93 = sub i32 %89, %92
   %94 = ashr i32 %93, 2
@@ -185,77 +185,77 @@ bb13.i.i:                                         ; preds = %bb12.i.i
   ]
 
 bb14.i.i:                                         ; preds = %bb13.i.i
-  %95 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %96 = load i32** %95, align 4
-  %97 = load i32* %96, align 4
-  %98 = load i32* %4, align 4
+  %95 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %96 = load i32*, i32** %95, align 4
+  %97 = load i32, i32* %96, align 4
+  %98 = load i32, i32* %4, align 4
   %99 = icmp eq i32 %97, %98
   %100 = zext i1 %99 to i8
   %toBool15.i.i = icmp ne i8 %100, 0
   br i1 %toBool15.i.i, label %bb16.i.i, label %bb17.i.i
 
 bb16.i.i:                                         ; preds = %bb14.i.i
-  %101 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %102 = load i32** %101, align 4
+  %101 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %102 = load i32*, i32** %101, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb17.i.i:                                         ; preds = %bb14.i.i
-  %103 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %104 = load i32** %103, align 4
-  %105 = getelementptr i32* %104, i64 1
-  %106 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %103 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %104 = load i32*, i32** %103, align 4
+  %105 = getelementptr i32, i32* %104, i64 1
+  %106 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %105, i32** %106, align 4
   br label %bb18.i.i
 
 bb18.i.i:                                         ; preds = %bb17.i.i, %bb13.i.i
-  %107 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %108 = load i32** %107, align 4
-  %109 = load i32* %108, align 4
-  %110 = load i32* %4, align 4
+  %107 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %108 = load i32*, i32** %107, align 4
+  %109 = load i32, i32* %108, align 4
+  %110 = load i32, i32* %4, align 4
   %111 = icmp eq i32 %109, %110
   %112 = zext i1 %111 to i8
   %toBool19.i.i = icmp ne i8 %112, 0
   br i1 %toBool19.i.i, label %bb20.i.i, label %bb21.i.i
 
 bb20.i.i:                                         ; preds = %bb18.i.i
-  %113 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %114 = load i32** %113, align 4
+  %113 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %114 = load i32*, i32** %113, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb21.i.i:                                         ; preds = %bb18.i.i
-  %115 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %116 = load i32** %115, align 4
-  %117 = getelementptr i32* %116, i64 1
-  %118 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %115 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %116 = load i32*, i32** %115, align 4
+  %117 = getelementptr i32, i32* %116, i64 1
+  %118 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %117, i32** %118, align 4
   br label %bb22.i.i
 
 bb22.i.i:                                         ; preds = %bb21.i.i, %bb13.i.i
-  %119 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %120 = load i32** %119, align 4
-  %121 = load i32* %120, align 4
-  %122 = load i32* %4, align 4
+  %119 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %120 = load i32*, i32** %119, align 4
+  %121 = load i32, i32* %120, align 4
+  %122 = load i32, i32* %4, align 4
   %123 = icmp eq i32 %121, %122
   %124 = zext i1 %123 to i8
   %toBool23.i.i = icmp ne i8 %124, 0
   br i1 %toBool23.i.i, label %bb24.i.i, label %bb25.i.i
 
 bb24.i.i:                                         ; preds = %bb22.i.i
-  %125 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %126 = load i32** %125, align 4
+  %125 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %126 = load i32*, i32** %125, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 bb25.i.i:                                         ; preds = %bb22.i.i
-  %127 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
-  %128 = load i32** %127, align 4
-  %129 = getelementptr i32* %128, i64 1
-  %130 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %127 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
+  %128 = load i32*, i32** %127, align 4
+  %129 = getelementptr i32, i32* %128, i64 1
+  %130 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__first_addr.i.i, i32 0, i32 0
   store i32* %129, i32** %130, align 4
   br label %bb26.i.i
 
 bb26.i.i:                                         ; preds = %bb25.i.i, %bb13.i.i
-  %131 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
-  %132 = load i32** %131, align 4
+  %131 = getelementptr %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >", %"struct.__gnu_cxx::__normal_iterator<int*,std::vector<int, std::allocator<int> > >"* %__last_addr.i.i, i32 0, i32 0
+  %132 = load i32*, i32** %131, align 4
   br label %_ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit
 
 _ZSt4findIN9__gnu_cxx17__normal_iteratorIPiSt6vectorIiSaIiEEEEiET_S7_S7_RKT0_.exit: ; preds = %bb26.i.i, %bb24.i.i, %bb20.i.i, %bb16.i.i, %bb10.i.i, %bb7.i.i, %bb4.i.i, %bb1.i.i
diff --git a/test/Transforms/InstCombine/2009-02-21-LoadCST.ll b/test/Transforms/InstCombine/2009-02-21-LoadCST.ll
index cb8a77c23bac..90ec6d540e9c 100644
--- a/test/Transforms/InstCombine/2009-02-21-LoadCST.ll
+++ b/test/Transforms/InstCombine/2009-02-21-LoadCST.ll
@@ -7,6 +7,6 @@ target triple = "i386-pc-linux-gnu"
 @.str1 = internal constant [4 x i8] c"\B5%8\00"
 
 define i32 @test() {
-  %rhsv = load i32* bitcast ([4 x i8]* @.str1 to i32*), align 1
+  %rhsv = load i32, i32* bitcast ([4 x i8]* @.str1 to i32*), align 1
   ret i32 %rhsv
 }
diff --git a/test/Transforms/InstCombine/2009-02-25-CrashZeroSizeArray.ll b/test/Transforms/InstCombine/2009-02-25-CrashZeroSizeArray.ll
index a8349f042e42..ef1734ba7d17 100644
--- a/test/Transforms/InstCombine/2009-02-25-CrashZeroSizeArray.ll
+++ b/test/Transforms/InstCombine/2009-02-25-CrashZeroSizeArray.ll
@@ -9,18 +9,18 @@ entry:
 	%tmp6 = mul i32 %max289, 4		; <i32> [#uses=1]
 	%tmp7 = alloca i8, i32 0		; <i8*> [#uses=1]
 	%tmp8 = bitcast i8* %tmp7 to [0 x [0 x i32]]*		; <[0 x [0 x i32]]*> [#uses=1]
-	%tmp11 = load i32* null, align 1		; <i32> [#uses=1]
+	%tmp11 = load i32, i32* null, align 1		; <i32> [#uses=1]
 	%tmp12 = icmp eq i32 %tmp11, 3		; <i1> [#uses=1]
 	%tmp13 = zext i1 %tmp12 to i8		; <i8> [#uses=1]
 	%tmp14 = ashr i32 %tmp6, 2		; <i32> [#uses=1]
 	%tmp15 = bitcast [0 x [0 x i32]]* %tmp8 to i8*		; <i8*> [#uses=1]
 	%tmp16 = mul i32 %tmp14, 4		; <i32> [#uses=1]
 	%tmp17 = mul i32 1, %tmp16		; <i32> [#uses=1]
-	%tmp18 = getelementptr i8* %tmp15, i32 %tmp17		; <i8*> [#uses=1]
+	%tmp18 = getelementptr i8, i8* %tmp15, i32 %tmp17		; <i8*> [#uses=1]
 	%tmp19 = bitcast i8* %tmp18 to [0 x i32]*		; <[0 x i32]*> [#uses=1]
 	%tmp20 = bitcast [0 x i32]* %tmp19 to i32*		; <i32*> [#uses=1]
-	%tmp21 = getelementptr i32* %tmp20, i32 0		; <i32*> [#uses=1]
-	%tmp22 = load i32* %tmp21, align 1		; <i32> [#uses=1]
+	%tmp21 = getelementptr i32, i32* %tmp20, i32 0		; <i32*> [#uses=1]
+	%tmp22 = load i32, i32* %tmp21, align 1		; <i32> [#uses=1]
 	%tmp23 = icmp eq i32 %tmp22, 4		; <i1> [#uses=1]
 	%tmp24 = zext i1 %tmp23 to i8		; <i8> [#uses=1]
 	%toBool709 = icmp ne i8 %tmp13, 0		; <i1> [#uses=1]
diff --git a/test/Transforms/InstCombine/2009-03-18-vector-ashr-crash.ll b/test/Transforms/InstCombine/2009-03-18-vector-ashr-crash.ll
index c617ca4fcad3..3847abd30c80 100644
--- a/test/Transforms/InstCombine/2009-03-18-vector-ashr-crash.ll
+++ b/test/Transforms/InstCombine/2009-03-18-vector-ashr-crash.ll
@@ -3,9 +3,9 @@
 
 define void @0(<4 x i16>*, <4 x i16>*) {
 	%3 = alloca <4 x i16>*		; <<4 x i16>**> [#uses=1]
-	%4 = load <4 x i16>* null, align 1		; <<4 x i16>> [#uses=1]
+	%4 = load <4 x i16>, <4 x i16>* null, align 1		; <<4 x i16>> [#uses=1]
 	%5 = ashr <4 x i16> %4, <i16 5, i16 5, i16 5, i16 5>		; <<4 x i16>> [#uses=1]
-	%6 = load <4 x i16>** %3		; <<4 x i16>*> [#uses=1]
+	%6 = load <4 x i16>*, <4 x i16>** %3		; <<4 x i16>*> [#uses=1]
 	store <4 x i16> %5, <4 x i16>* %6, align 1
 	ret void
 }
diff --git a/test/Transforms/InstCombine/2009-05-23-FCmpToICmp.ll b/test/Transforms/InstCombine/2009-05-23-FCmpToICmp.ll
index dd14c6beec4c..ced317c4d43f 100644
--- a/test/Transforms/InstCombine/2009-05-23-FCmpToICmp.ll
+++ b/test/Transforms/InstCombine/2009-05-23-FCmpToICmp.ll
@@ -2,7 +2,7 @@
 ; rdar://6903175
 
 define i1 @f0(i32 *%a) nounwind {
-       %b = load i32* %a, align 4
+       %b = load i32, i32* %a, align 4
        %c = uitofp i32 %b to double
        %d = fcmp ogt double %c, 0x41EFFFFFFFE00000
        ret i1 %d
diff --git a/test/Transforms/InstCombine/2009-12-17-CmpSelectNull.ll b/test/Transforms/InstCombine/2009-12-17-CmpSelectNull.ll
index fb7497b395d4..c438ca5fd172 100644
--- a/test/Transforms/InstCombine/2009-12-17-CmpSelectNull.ll
+++ b/test/Transforms/InstCombine/2009-12-17-CmpSelectNull.ll
@@ -6,9 +6,9 @@
 define i8* @demangle_qualified(i32 %isfuncname) nounwind {
 entry:
   %tobool272 = icmp ne i32 %isfuncname, 0
-  %cond276 = select i1 %tobool272, i8* getelementptr inbounds ([2 x i8]* @.str254, i32 0, i32 0), i8* getelementptr inbounds ([3 x i8]* @.str557, i32 0, i32 0) ; <i8*> [#uses=4]
+  %cond276 = select i1 %tobool272, i8* getelementptr inbounds ([2 x i8], [2 x i8]* @.str254, i32 0, i32 0), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str557, i32 0, i32 0) ; <i8*> [#uses=4]
   %cmp.i504 = icmp eq i8* %cond276, null
-  %rval = getelementptr i8* %cond276, i1 %cmp.i504
+  %rval = getelementptr i8, i8* %cond276, i1 %cmp.i504
   ret i8* %rval
 }
 
diff --git a/test/Transforms/InstCombine/2010-03-03-ExtElim.ll b/test/Transforms/InstCombine/2010-03-03-ExtElim.ll
index e0def997d974..ad0fe5a21783 100644
--- a/test/Transforms/InstCombine/2010-03-03-ExtElim.ll
+++ b/test/Transforms/InstCombine/2010-03-03-ExtElim.ll
@@ -4,11 +4,11 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 target triple = "i386-unknown-linux-gnu"
 
 @g_92 = common global [2 x i32*] zeroinitializer, align 4 ; <[2 x i32*]*> [#uses=1]
-@g_177 = constant i32** bitcast (i8* getelementptr (i8* bitcast ([2 x i32*]* @g_92 to i8*), i64 4) to i32**), align 4 ; <i32***> [#uses=1]
+@g_177 = constant i32** bitcast (i8* getelementptr (i8, i8* bitcast ([2 x i32*]* @g_92 to i8*), i64 4) to i32**), align 4 ; <i32***> [#uses=1]
 
 define i1 @PR6486() nounwind {
 ; CHECK-LABEL: @PR6486(
-  %tmp = load i32*** @g_177                       ; <i32**> [#uses=1]
+  %tmp = load i32**, i32*** @g_177                       ; <i32**> [#uses=1]
   %cmp = icmp ne i32** null, %tmp                 ; <i1> [#uses=1]
   %conv = zext i1 %cmp to i32                     ; <i32> [#uses=1]
   %cmp1 = icmp sle i32 0, %conv                   ; <i1> [#uses=1]
@@ -21,12 +21,12 @@ define i1 @PR6486() nounwind {
 
 define i1 @PR16462_1() nounwind {
 ; CHECK-LABEL: @PR16462_1(
-  ret i1 icmp sgt (i32 sext (i16 trunc (i32 select (i1 icmp eq (i32* getelementptr inbounds ([1 x i32]* @a, i32 0, i32 0), i32* @d), i32 0, i32 1) to i16) to i32), i32 65535)
+  ret i1 icmp sgt (i32 sext (i16 trunc (i32 select (i1 icmp eq (i32* getelementptr inbounds ([1 x i32], [1 x i32]* @a, i32 0, i32 0), i32* @d), i32 0, i32 1) to i16) to i32), i32 65535)
 ; CHECK: ret i1 false
 }
 
 define i1 @PR16462_2() nounwind {
 ; CHECK-LABEL: @PR16462_2(
-  ret i1 icmp sgt (i32 sext (i16 trunc (i32 select (i1 icmp eq (i32* getelementptr inbounds ([1 x i32]* @a, i32 0, i32 0), i32* @d), i32 0, i32 1) to i16) to i32), i32 42)
+  ret i1 icmp sgt (i32 sext (i16 trunc (i32 select (i1 icmp eq (i32* getelementptr inbounds ([1 x i32], [1 x i32]* @a, i32 0, i32 0), i32* @d), i32 0, i32 1) to i16) to i32), i32 42)
 ; CHECK: ret i1 false
 }
diff --git a/test/Transforms/InstCombine/2010-05-30-memcpy-Struct.ll b/test/Transforms/InstCombine/2010-05-30-memcpy-Struct.ll
index 09a96749f260..51610698c244 100644
--- a/test/Transforms/InstCombine/2010-05-30-memcpy-Struct.ll
+++ b/test/Transforms/InstCombine/2010-05-30-memcpy-Struct.ll
@@ -10,7 +10,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @CopyEventArg(%union.anon* %ev) nounwind {
 entry:
-  %call = call i32 (i8*, i8*, ...)* @sprintf(i8* undef, i8* getelementptr inbounds ([3 x i8]* @.str, i64 0, i64 0), %union.anon* %ev) nounwind
+  %call = call i32 (i8*, i8*, ...) @sprintf(i8* undef, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i64 0, i64 0), %union.anon* %ev) nounwind
 ; CHECK: bitcast %union.anon* %ev to i8*
 ; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64
   ret void
diff --git a/test/Transforms/InstCombine/2010-11-21-SizeZeroTypeGEP.ll b/test/Transforms/InstCombine/2010-11-21-SizeZeroTypeGEP.ll
index 80983ef27098..798c726e566f 100644
--- a/test/Transforms/InstCombine/2010-11-21-SizeZeroTypeGEP.ll
+++ b/test/Transforms/InstCombine/2010-11-21-SizeZeroTypeGEP.ll
@@ -5,13 +5,13 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define {}* @foo({}* %x, i32 %n) {
 ; CHECK-LABEL: @foo(
 ; CHECK-NOT: getelementptr
-  %p = getelementptr {}* %x, i32 %n
+  %p = getelementptr {}, {}* %x, i32 %n
   ret {}* %p
 }
 
 define i8* @bar(i64 %n, {{}, [0 x {[0 x i8]}]}* %p) {
 ; CHECK-LABEL: @bar(
-  %g = getelementptr {{}, [0 x {[0 x i8]}]}* %p, i64 %n, i32 1, i64 %n, i32 0, i64 %n
+  %g = getelementptr {{}, [0 x {[0 x i8]}]}, {{}, [0 x {[0 x i8]}]}* %p, i64 %n, i32 1, i64 %n, i32 0, i64 %n
 ; CHECK: %p, i64 0, i32 1, i64 0, i32 0, i64 %n
   ret i8* %g
 }
diff --git a/test/Transforms/InstCombine/2011-05-02-VectorBoolean.ll b/test/Transforms/InstCombine/2011-05-02-VectorBoolean.ll
index 02b64e3bb8d3..116c9713d893 100644
--- a/test/Transforms/InstCombine/2011-05-02-VectorBoolean.ll
+++ b/test/Transforms/InstCombine/2011-05-02-VectorBoolean.ll
@@ -6,9 +6,9 @@ entry:
   %a.addr = alloca <2 x i16>, align 4
   %.compoundliteral = alloca <2 x i16>, align 4
   store <2 x i16> %a, <2 x i16>* %a.addr, align 4
-  %tmp = load <2 x i16>* %a.addr, align 4
+  %tmp = load <2 x i16>, <2 x i16>* %a.addr, align 4
   store <2 x i16> zeroinitializer, <2 x i16>* %.compoundliteral
-  %tmp1 = load <2 x i16>* %.compoundliteral
+  %tmp1 = load <2 x i16>, <2 x i16>* %.compoundliteral
   %cmp = icmp uge <2 x i16> %tmp, %tmp1
   %sext = sext <2 x i1> %cmp to <2 x i16>
   ret <2 x i16> %sext
diff --git a/test/Transforms/InstCombine/2011-05-13-InBoundsGEP.ll b/test/Transforms/InstCombine/2011-05-13-InBoundsGEP.ll
index fba7239e2d1e..69568705cfd5 100644
--- a/test/Transforms/InstCombine/2011-05-13-InBoundsGEP.ll
+++ b/test/Transforms/InstCombine/2011-05-13-InBoundsGEP.ll
@@ -9,9 +9,9 @@ entry:
     %tmp2 = add i32 %argc, 1879048192
     %p = alloca i8
 ; CHECK: getelementptr
-    %p1 = getelementptr i8* %p, i32 %tmp1
+    %p1 = getelementptr i8, i8* %p, i32 %tmp1
 ; CHECK: getelementptr
-    %p2 = getelementptr i8* %p, i32 %tmp2
+    %p2 = getelementptr i8, i8* %p, i32 %tmp2
     %cmp = icmp ult i8* %p1, %p2
     br i1 %cmp, label %bbtrue, label %bbfalse
 bbtrue:          ; preds = %entry
diff --git a/test/Transforms/InstCombine/2011-05-28-swapmulsub.ll b/test/Transforms/InstCombine/2011-05-28-swapmulsub.ll
index b096d1f952f9..639b64aacba0 100644
--- a/test/Transforms/InstCombine/2011-05-28-swapmulsub.ll
+++ b/test/Transforms/InstCombine/2011-05-28-swapmulsub.ll
@@ -7,14 +7,14 @@ entry:
   %on_off.addr = alloca i32, align 4
   %a = alloca i32, align 4
   store i32 %on_off, i32* %on_off.addr, align 4
-  %tmp = load i32* %on_off.addr, align 4
+  %tmp = load i32, i32* %on_off.addr, align 4
   %sub = sub i32 1, %tmp
 ; CHECK-NOT: mul i32
   %mul = mul i32 %sub, -2
 ; CHECK: shl
 ; CHECK-NEXT: add
   store i32 %mul, i32* %a, align 4
-  %tmp1 = load i32* %a, align 4
+  %tmp1 = load i32, i32* %a, align 4
   %conv = trunc i32 %tmp1 to i16
   ret i16 %conv
 }
@@ -26,15 +26,15 @@ entry:
   %a = alloca i32, align 4
   store i32 %on_off, i32* %on_off.addr, align 4
   store i32 %q, i32* %q.addr, align 4
-  %tmp = load i32* %q.addr, align 4
-  %tmp1 = load i32* %on_off.addr, align 4
+  %tmp = load i32, i32* %q.addr, align 4
+  %tmp1 = load i32, i32* %on_off.addr, align 4
   %sub = sub i32 %tmp, %tmp1
 ; CHECK-NOT: mul i32
   %mul = mul i32 %sub, -4
 ; CHECK: sub i32
 ; CHECK-NEXT: shl
   store i32 %mul, i32* %a, align 4
-  %tmp2 = load i32* %a, align 4
+  %tmp2 = load i32, i32* %a, align 4
   %conv = trunc i32 %tmp2 to i16
   ret i16 %conv
 }
@@ -44,14 +44,14 @@ entry:
   %on_off.addr = alloca i32, align 4
   %a = alloca i32, align 4
   store i32 %on_off, i32* %on_off.addr, align 4
-  %tmp = load i32* %on_off.addr, align 4
+  %tmp = load i32, i32* %on_off.addr, align 4
   %sub = sub i32 7, %tmp
 ; CHECK-NOT: mul i32
   %mul = mul i32 %sub, -4
 ; CHECK: shl
 ; CHECK-NEXT: add
   store i32 %mul, i32* %a, align 4
-  %tmp1 = load i32* %a, align 4
+  %tmp1 = load i32, i32* %a, align 4
   %conv = trunc i32 %tmp1 to i16
   ret i16 %conv
 }
diff --git a/test/Transforms/InstCombine/2011-06-13-nsw-alloca.ll b/test/Transforms/InstCombine/2011-06-13-nsw-alloca.ll
index c8f0351858ca..15c11db37f59 100644
--- a/test/Transforms/InstCombine/2011-06-13-nsw-alloca.ll
+++ b/test/Transforms/InstCombine/2011-06-13-nsw-alloca.ll
@@ -9,12 +9,12 @@ define void @fu1(i32 %parm) nounwind ssp {
   %ptr = alloca double*, align 4
   store i32 %parm, i32* %1, align 4
   store double* null, double** %ptr, align 4
-  %2 = load i32* %1, align 4
+  %2 = load i32, i32* %1, align 4
   %3 = icmp ne i32 %2, 0
   br i1 %3, label %4, label %10
 
 ; <label>:4                                       ; preds = %0
-  %5 = load i32* %1, align 4
+  %5 = load i32, i32* %1, align 4
   %6 = shl nsw i32 %5, 3
 ; With "nsw", the alloca and its bitcast can be fused:
   %7 = add nsw i32 %6, 2048
@@ -25,7 +25,7 @@ define void @fu1(i32 %parm) nounwind ssp {
   store double* %9, double** %ptr, align 4
   br label %10
 ; <label>:10                                      ; preds = %4, %0
-  %11 = load double** %ptr, align 4
+  %11 = load double*, double** %ptr, align 4
   call void @bar(double* %11)
 ; CHECK: ret
   ret void
@@ -39,12 +39,12 @@ define void @fu2(i32 %parm) nounwind ssp {
   %ptr = alloca double*, align 4
   store i32 %parm, i32* %1, align 4
   store double* null, double** %ptr, align 4
-  %2 = load i32* %1, align 4
+  %2 = load i32, i32* %1, align 4
   %3 = icmp ne i32 %2, 0
   br i1 %3, label %4, label %10
 
 ; <label>:4                                       ; preds = %0
-  %5 = load i32* %1, align 4
+  %5 = load i32, i32* %1, align 4
   %6 = mul nsw i32 %5, 8
 ; Without "nsw", the alloca and its bitcast cannot be fused:
   %7 = add  i32 %6, 2048
@@ -57,7 +57,7 @@ define void @fu2(i32 %parm) nounwind ssp {
   br label %10
 
 ; <label>:10                                      ; preds = %4, %0
-  %11 = load double** %ptr, align 4
+  %11 = load double*, double** %ptr, align 4
   call void @bar(double* %11)
   ret void
 }
diff --git a/test/Transforms/InstCombine/2011-09-03-Trampoline.ll b/test/Transforms/InstCombine/2011-09-03-Trampoline.ll
index 5456e0357393..1833558cbceb 100644
--- a/test/Transforms/InstCombine/2011-09-03-Trampoline.ll
+++ b/test/Transforms/InstCombine/2011-09-03-Trampoline.ll
@@ -7,7 +7,7 @@ declare i32 @f(i8 * nest, i32)
 ; Most common case
 define i32 @test0(i32 %n) {
   %alloca = alloca [10 x i8], align 16
-  %gep = getelementptr [10 x i8]* %alloca, i32 0, i32 0
+  %gep = getelementptr [10 x i8], [10 x i8]* %alloca, i32 0, i32 0
   call void @llvm.init.trampoline(i8* %gep, i8* bitcast (i32 (i8*, i32)* @f to i8*),
                                   i8* null)
   %tramp = call i8* @llvm.adjust.trampoline(i8* %gep)
@@ -62,7 +62,7 @@ define i32 @test3(i32 %n, i8* %trampmem) {
 
 define i32 @test4(i32 %n) {
   %alloca = alloca [10 x i8], align 16
-  %gep = getelementptr [10 x i8]* %alloca, i32 0, i32 0
+  %gep = getelementptr [10 x i8], [10 x i8]* %alloca, i32 0, i32 0
   call void @llvm.init.trampoline(i8* %gep, i8* bitcast (i32 (i8*, i32)* @f to i8*),
                                   i8* null)
 
diff --git a/test/Transforms/InstCombine/2011-10-07-AlignPromotion.ll b/test/Transforms/InstCombine/2011-10-07-AlignPromotion.ll
index 22061b272a85..122669ec04a7 100644
--- a/test/Transforms/InstCombine/2011-10-07-AlignPromotion.ll
+++ b/test/Transforms/InstCombine/2011-10-07-AlignPromotion.ll
@@ -10,7 +10,7 @@ define void @t(%struct.CGPoint* %a) nounwind {
   %Point = alloca %struct.CGPoint, align 4
   %1 = bitcast %struct.CGPoint* %a to i64*
   %2 = bitcast %struct.CGPoint* %Point to i64*
-  %3 = load i64* %1, align 4
+  %3 = load i64, i64* %1, align 4
   store i64 %3, i64* %2, align 4
   call void @foo(i64* %2) nounwind
   ret void
diff --git a/test/Transforms/InstCombine/2012-02-13-FCmp.ll b/test/Transforms/InstCombine/2012-02-13-FCmp.ll
index 39b05946515a..586f86de3c29 100644
--- a/test/Transforms/InstCombine/2012-02-13-FCmp.ll
+++ b/test/Transforms/InstCombine/2012-02-13-FCmp.ll
@@ -6,9 +6,9 @@
 declare i32 @printf(i8*, ...)
 define i64 @_Z8tempCastj(i32 %val) uwtable ssp {
 entry:
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([35 x i8]* @.str1, i64 0, i64 0), i32 %val)
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([35 x i8], [35 x i8]* @.str1, i64 0, i64 0), i32 %val)
   %conv = uitofp i32 %val to double
-  %call.i = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([35 x i8]* @.str, i64 0, i64 0), double %conv)
+  %call.i = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([35 x i8], [35 x i8]* @.str, i64 0, i64 0), double %conv)
   %cmp.i = fcmp oge double %conv, -1.000000e+00
   br i1 %cmp.i, label %land.rhs.i, label %if.end.critedge
 ; CHECK:  br i1 true, label %land.rhs.i, label %if.end.critedge
diff --git a/test/Transforms/InstCombine/2012-05-27-Negative-Shift-Crash.ll b/test/Transforms/InstCombine/2012-05-27-Negative-Shift-Crash.ll
index ba83fe9ec0ad..ba57baf23711 100644
--- a/test/Transforms/InstCombine/2012-05-27-Negative-Shift-Crash.ll
+++ b/test/Transforms/InstCombine/2012-05-27-Negative-Shift-Crash.ll
@@ -20,7 +20,7 @@ entry:
 
 define void @fn4() nounwind uwtable ssp {
 entry:
-  %0 = load i32* @d, align 4
+  %0 = load i32, i32* @d, align 4
   %cmp = icmp eq i32 %0, 0
   %conv = zext i1 %cmp to i32
   store i32 %conv, i32* @c, align 4
@@ -34,12 +34,12 @@ entry:
   store i32 %and, i32* @e, align 4
   %sub = add nsw i32 %and, -1
   store i32 %sub, i32* @f, align 4
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %if.else, label %if.then
 
 if.then:                                          ; preds = %entry
-  %1 = load i32* @b, align 4
+  %1 = load i32, i32* @b, align 4
   %.lobit = lshr i32 %1, 31
   %2 = trunc i32 %.lobit to i8
   %.not = xor i8 %2, 1
diff --git a/test/Transforms/InstCombine/2012-05-28-select-hang.ll b/test/Transforms/InstCombine/2012-05-28-select-hang.ll
index db1dbd5b6e2f..c514dd1f5ec7 100644
--- a/test/Transforms/InstCombine/2012-05-28-select-hang.ll
+++ b/test/Transforms/InstCombine/2012-05-28-select-hang.ll
@@ -6,7 +6,7 @@
 
 define void @func() nounwind uwtable ssp {
 entry:
-  %0 = load i8* @c, align 1
+  %0 = load i8, i8* @c, align 1
   %conv = zext i8 %0 to i32
   %or = or i32 %conv, 1
   %conv1 = trunc i32 %or to i8
@@ -16,7 +16,7 @@ entry:
   %and = and i32 1, %neg
   %conv3 = trunc i32 %and to i8
   store i8 %conv3, i8* @b, align 1
-  %1 = load i8* @a, align 1
+  %1 = load i8, i8* @a, align 1
   %conv4 = zext i8 %1 to i32
   %conv5 = zext i8 %conv3 to i32
   %tobool = icmp ne i32 %conv4, 0
diff --git a/test/Transforms/InstCombine/2012-06-06-LoadOfPHIs.ll b/test/Transforms/InstCombine/2012-06-06-LoadOfPHIs.ll
index 22466a96b667..4af1ca842c59 100644
--- a/test/Transforms/InstCombine/2012-06-06-LoadOfPHIs.ll
+++ b/test/Transforms/InstCombine/2012-06-06-LoadOfPHIs.ll
@@ -34,16 +34,16 @@ bb11:                                             ; preds = %bb8
 
 bb12:                                             ; preds = %bb11, %bb10, %bb7, %bb6
   %max.0 = phi double* [ %tmp, %bb6 ], [ %tmp2, %bb7 ], [ %tmp1, %bb10 ], [ %tmp2, %bb11 ]
-; CHECK: %tmp13 = load double* %tmp, align 8
-; CHECK: %tmp14 = load double* %tmp1, align 8
+; CHECK: %tmp13 = load double, double* %tmp, align 8
+; CHECK: %tmp14 = load double, double* %tmp1, align 8
 ; CHECK: %tmp15 = fcmp olt double %tmp13, %tmp14
-  %tmp13 = load double* %tmp, align 8
-  %tmp14 = load double* %tmp1, align 8
+  %tmp13 = load double, double* %tmp, align 8
+  %tmp14 = load double, double* %tmp1, align 8
   %tmp15 = fcmp olt double %tmp13, %tmp14
   br i1 %tmp15, label %bb16, label %bb21
 
 bb16:                                             ; preds = %bb12
-  %tmp17 = load double* %tmp2, align 8
+  %tmp17 = load double, double* %tmp2, align 8
   %tmp18 = fcmp olt double %tmp13, %tmp17
   br i1 %tmp18, label %bb19, label %bb20
 
@@ -54,7 +54,7 @@ bb20:                                             ; preds = %bb16
   br label %bb26
 
 bb21:                                             ; preds = %bb12
-  %tmp22 = load double* %tmp2, align 8
+  %tmp22 = load double, double* %tmp2, align 8
   %tmp23 = fcmp olt double %tmp14, %tmp22
   br i1 %tmp23, label %bb24, label %bb25
 
@@ -66,16 +66,16 @@ bb25:                                             ; preds = %bb21
 
 bb26:                                             ; preds = %bb25, %bb24, %bb20, %bb19
   %min.0 = phi double* [ %tmp, %bb19 ], [ %tmp2, %bb20 ], [ %tmp1, %bb24 ], [ %tmp2, %bb25 ]
-; CHECK: %tmp27 = load double* %min.0, align 8
-; CHECK: %tmp28 = load double* %max.0
+; CHECK: %tmp27 = load double, double* %min.0, align 8
+; CHECK: %tmp28 = load double, double* %max.0
 ; CHECK: %tmp29 = fadd double %tmp27, %tmp28
-  %tmp27 = load double* %min.0, align 8
-  %tmp28 = load double* %max.0
+  %tmp27 = load double, double* %min.0, align 8
+  %tmp28 = load double, double* %max.0
   %tmp29 = fadd double %tmp27, %tmp28
   %tmp30 = fdiv double %tmp29, 2.000000e+00
   store double %tmp30, double* %outL
-  %tmp31 = load double* %min.0
-  %tmp32 = load double* %max.0
+  %tmp31 = load double, double* %min.0
+  %tmp32 = load double, double* %max.0
   %tmp33 = fcmp oeq double %tmp31, %tmp32
   br i1 %tmp33, label %bb34, label %bb35
 
@@ -107,11 +107,11 @@ bb45:                                             ; preds = %bb41, %bb38
   br i1 %tmp46, label %bb47, label %bb55
 
 bb47:                                             ; preds = %bb45
-  %tmp48 = load double* %tmp1, align 8
-  %tmp49 = load double* %tmp2, align 8
+  %tmp48 = load double, double* %tmp1, align 8
+  %tmp49 = load double, double* %tmp2, align 8
   %tmp50 = fsub double %tmp48, %tmp49
-  %tmp51 = load double* %max.0
-  %tmp52 = load double* %min.0
+  %tmp51 = load double, double* %max.0
+  %tmp52 = load double, double* %min.0
   %tmp53 = fsub double %tmp51, %tmp52
   %tmp54 = fdiv double %tmp50, %tmp53
   store double %tmp54, double* %outH
@@ -122,11 +122,11 @@ bb55:                                             ; preds = %bb45
   br i1 %tmp56, label %bb57, label %bb66
 
 bb57:                                             ; preds = %bb55
-  %tmp58 = load double* %tmp2, align 8
-  %tmp59 = load double* %tmp, align 8
+  %tmp58 = load double, double* %tmp2, align 8
+  %tmp59 = load double, double* %tmp, align 8
   %tmp60 = fsub double %tmp58, %tmp59
-  %tmp61 = load double* %max.0
-  %tmp62 = load double* %min.0
+  %tmp61 = load double, double* %max.0
+  %tmp62 = load double, double* %min.0
   %tmp63 = fsub double %tmp61, %tmp62
   %tmp64 = fdiv double %tmp60, %tmp63
   %tmp65 = fadd double 2.000000e+00, %tmp64
@@ -134,11 +134,11 @@ bb57:                                             ; preds = %bb55
   br label %bb75
 
 bb66:                                             ; preds = %bb55
-  %tmp67 = load double* %tmp, align 8
-  %tmp68 = load double* %tmp1, align 8
+  %tmp67 = load double, double* %tmp, align 8
+  %tmp68 = load double, double* %tmp1, align 8
   %tmp69 = fsub double %tmp67, %tmp68
-  %tmp70 = load double* %max.0
-  %tmp71 = load double* %min.0
+  %tmp70 = load double, double* %max.0
+  %tmp71 = load double, double* %min.0
   %tmp72 = fsub double %tmp70, %tmp71
   %tmp73 = fdiv double %tmp69, %tmp72
   %tmp74 = fadd double 4.000000e+00, %tmp73
@@ -146,7 +146,7 @@ bb66:                                             ; preds = %bb55
   br label %bb75
 
 bb75:                                             ; preds = %bb66, %bb57, %bb47
-  %tmp76 = load double* %outH
+  %tmp76 = load double, double* %outH
   %tmp77 = fdiv double %tmp76, 6.000000e+00
   store double %tmp77, double* %outH
   %tmp78 = fcmp olt double %tmp77, 0.000000e+00
diff --git a/test/Transforms/InstCombine/2012-07-25-LoadPart.ll b/test/Transforms/InstCombine/2012-07-25-LoadPart.ll
index 18aab7f27efd..14fcf52fe9a7 100644
--- a/test/Transforms/InstCombine/2012-07-25-LoadPart.ll
+++ b/test/Transforms/InstCombine/2012-07-25-LoadPart.ll
@@ -5,7 +5,7 @@
 @test = constant [4 x i32] [i32 1, i32 2, i32 3, i32 4]
 
 define i64 @foo() {
-  %ret = load i64* bitcast (i8* getelementptr (i8* bitcast ([4 x i32]* @test to i8*), i64 2) to i64*), align 1
+  %ret = load i64, i64* bitcast (i8* getelementptr (i8, i8* bitcast ([4 x i32]* @test to i8*), i64 2) to i64*), align 1
   ret i64 %ret
   ; 0x00030000_00020000 in [01 00/00 00 02 00 00 00 03 00/00 00 04 00 00 00]
   ; LE: ret i64 844424930263040
diff --git a/test/Transforms/InstCombine/2012-07-30-addrsp-bitcast.ll b/test/Transforms/InstCombine/2012-07-30-addrsp-bitcast.ll
index ac9c555020d0..666538067884 100644
--- a/test/Transforms/InstCombine/2012-07-30-addrsp-bitcast.ll
+++ b/test/Transforms/InstCombine/2012-07-30-addrsp-bitcast.ll
@@ -5,6 +5,6 @@
 declare void @foo(i32*)
 
 define void @test() nounwind {
-  call void @foo(i32* getelementptr (i32* addrspacecast ([16 x i32] addrspace(3)* @base to i32*), i64 2147483647)) nounwind
+  call void @foo(i32* getelementptr (i32, i32* addrspacecast ([16 x i32] addrspace(3)* @base to i32*), i64 2147483647)) nounwind
   ret void
 }
diff --git a/test/Transforms/InstCombine/2012-09-17-ZeroSizedAlloca.ll b/test/Transforms/InstCombine/2012-09-17-ZeroSizedAlloca.ll
index 7015725feb16..1c5a9813e6b3 100644
--- a/test/Transforms/InstCombine/2012-09-17-ZeroSizedAlloca.ll
+++ b/test/Transforms/InstCombine/2012-09-17-ZeroSizedAlloca.ll
@@ -16,8 +16,8 @@ target triple = "x86_64-apple-macosx10.8.0"
 define void @f() {
   %1 = alloca [0 x i8], align 1
   %2 = alloca [0 x i8], align 1024
-  %3 = getelementptr inbounds [0 x i8]* %1, i64 0, i64 0
-  %4 = getelementptr inbounds [0 x i8]* %2, i64 0, i64 0
+  %3 = getelementptr inbounds [0 x i8], [0 x i8]* %1, i64 0, i64 0
+  %4 = getelementptr inbounds [0 x i8], [0 x i8]* %2, i64 0, i64 0
   store i8* %3, i8** @x, align 8
   store i8* %4, i8** @y, align 8
   ret void
diff --git a/test/Transforms/InstCombine/2012-09-24-MemcpyFromGlobalCrash.ll b/test/Transforms/InstCombine/2012-09-24-MemcpyFromGlobalCrash.ll
deleted file mode 100644
index 35b62850c626..000000000000
--- a/test/Transforms/InstCombine/2012-09-24-MemcpyFromGlobalCrash.ll
+++ /dev/null
@@ -1,19 +0,0 @@
-; RUN: opt < %s -instcombine -S | FileCheck %s
-
-; Check we don't crash due to lack of target data.
-
-@G = constant [100 x i8] zeroinitializer
-
-declare void @bar(i8*)
-declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32, i1) nounwind
-
-define void @test() {
-; CHECK-LABEL: @test(
-; CHECK: llvm.memcpy
-; CHECK: ret void
-  %A = alloca [100 x i8]
-  %a = getelementptr inbounds [100 x i8]* %A, i64 0, i64 0
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* getelementptr inbounds ([100 x i8]* @G, i64 0, i32 0), i64 100, i32 4, i1 false)
-  call void @bar(i8* %a) readonly
-  ret void
-}
diff --git a/test/Transforms/InstCombine/2012-10-25-vector-of-pointers.ll b/test/Transforms/InstCombine/2012-10-25-vector-of-pointers.ll
index 20ea28268742..23210650e2f3 100644
--- a/test/Transforms/InstCombine/2012-10-25-vector-of-pointers.ll
+++ b/test/Transforms/InstCombine/2012-10-25-vector-of-pointers.ll
@@ -9,7 +9,7 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define void @widget(%struct.hoge* nocapture %arg) nounwind uwtable ssp {
 bb:
-  %tmp = getelementptr inbounds %struct.hoge* %arg, i64 0, i32 0
+  %tmp = getelementptr inbounds %struct.hoge, %struct.hoge* %arg, i64 0, i32 0
   br i1 undef, label %bb1, label %bb17
 
 bb1:                                              ; preds = %bb
@@ -20,7 +20,7 @@ bb2:                                              ; preds = %bb1
 
 bb3:                                              ; preds = %bb1
   %tmp4 = bitcast double** %tmp to <2 x double*>*
-  %tmp5 = load <2 x double*>* %tmp4, align 8
+  %tmp5 = load <2 x double*>, <2 x double*>* %tmp4, align 8
   %tmp6 = ptrtoint <2 x double*> %tmp5 to <2 x i64>
   %tmp7 = sub <2 x i64> zeroinitializer, %tmp6
   %tmp8 = ashr exact <2 x i64> %tmp7, <i64 3, i64 3>
diff --git a/test/Transforms/InstCombine/2012-12-14-simp-vgep.ll b/test/Transforms/InstCombine/2012-12-14-simp-vgep.ll
index fc29b095e5ce..46702f80c0cc 100644
--- a/test/Transforms/InstCombine/2012-12-14-simp-vgep.ll
+++ b/test/Transforms/InstCombine/2012-12-14-simp-vgep.ll
@@ -3,7 +3,7 @@
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 
 define <4 x i32> @foo(<4 x i32*>* %in) {
-  %t17 = load <4 x i32*>* %in, align 8
+  %t17 = load <4 x i32*>, <4 x i32*>* %in, align 8
   %t18 = icmp eq <4 x i32*> %t17, zeroinitializer
   %t19 = zext <4 x i1> %t18 to <4 x i32>
   ret <4 x i32> %t19
diff --git a/test/Transforms/InstCombine/2013-03-05-Combine-BitcastTy-Into-Alloca.ll b/test/Transforms/InstCombine/2013-03-05-Combine-BitcastTy-Into-Alloca.ll
index 789e3172e7c4..9425c29a427d 100644
--- a/test/Transforms/InstCombine/2013-03-05-Combine-BitcastTy-Into-Alloca.ll
+++ b/test/Transforms/InstCombine/2013-03-05-Combine-BitcastTy-Into-Alloca.ll
@@ -16,8 +16,8 @@ define { i64, i64 } @function(i32 %x, i32 %y, i32 %z) nounwind {
 ; CHECK-NOT: bitcast i96* %retval to %struct._my_struct*
 entry:
   %retval = alloca %struct._my_struct, align 8
-  %k.sroa.0.0.copyload = load i96* bitcast (%struct._my_struct* @initval to i96*), align 1
-  %k.sroa.1.12.copyload = load i32* bitcast ([4 x i8]* getelementptr inbounds (%struct._my_struct* @initval, i64 0, i32 1) to i32*), align 1
+  %k.sroa.0.0.copyload = load i96, i96* bitcast (%struct._my_struct* @initval to i96*), align 1
+  %k.sroa.1.12.copyload = load i32, i32* bitcast ([4 x i8]* getelementptr inbounds (%struct._my_struct, %struct._my_struct* @initval, i64 0, i32 1) to i32*), align 1
   %0 = zext i32 %x to i96
   %bf.value = shl nuw nsw i96 %0, 6
   %bf.clear = and i96 %k.sroa.0.0.copyload, -288230376151711744
@@ -32,14 +32,14 @@ entry:
   %bf.set10 = or i96 %bf.set5, %bf.clear
   %retval.0.cast7 = bitcast %struct._my_struct* %retval to i96*
   store i96 %bf.set10, i96* %retval.0.cast7, align 8
-  %retval.12.idx8 = getelementptr inbounds %struct._my_struct* %retval, i64 0, i32 1
+  %retval.12.idx8 = getelementptr inbounds %struct._my_struct, %struct._my_struct* %retval, i64 0, i32 1
   %retval.12.cast9 = bitcast [4 x i8]* %retval.12.idx8 to i32*
   store i32 %k.sroa.1.12.copyload, i32* %retval.12.cast9, align 4
   %trunc = trunc i96 %bf.set10 to i64
   %.fca.0.insert = insertvalue { i64, i64 } undef, i64 %trunc, 0
-  %retval.8.idx12 = getelementptr inbounds %struct._my_struct* %retval, i64 0, i32 0, i64 8
+  %retval.8.idx12 = getelementptr inbounds %struct._my_struct, %struct._my_struct* %retval, i64 0, i32 0, i64 8
   %retval.8.cast13 = bitcast i8* %retval.8.idx12 to i64*
-  %retval.8.load14 = load i64* %retval.8.cast13, align 8
+  %retval.8.load14 = load i64, i64* %retval.8.cast13, align 8
   %.fca.1.insert = insertvalue { i64, i64 } %.fca.0.insert, i64 %retval.8.load14, 1
   ret { i64, i64 } %.fca.1.insert
 }
diff --git a/test/Transforms/InstCombine/CPP_min_max.ll b/test/Transforms/InstCombine/CPP_min_max.ll
index b3d081b613a6..04bf0ceefea9 100644
--- a/test/Transforms/InstCombine/CPP_min_max.ll
+++ b/test/Transforms/InstCombine/CPP_min_max.ll
@@ -10,11 +10,11 @@
 
 define void @_Z5test1RiS_(i32* %x, i32* %y) {
 entry:
-        %tmp.1.i = load i32* %y         ; <i32> [#uses=1]
-        %tmp.3.i = load i32* %x         ; <i32> [#uses=1]
+        %tmp.1.i = load i32, i32* %y         ; <i32> [#uses=1]
+        %tmp.3.i = load i32, i32* %x         ; <i32> [#uses=1]
         %tmp.4.i = icmp slt i32 %tmp.1.i, %tmp.3.i              ; <i1> [#uses=1]
         %retval.i = select i1 %tmp.4.i, i32* %y, i32* %x                ; <i32*> [#uses=1]
-        %tmp.4 = load i32* %retval.i            ; <i32> [#uses=1]
+        %tmp.4 = load i32, i32* %retval.i            ; <i32> [#uses=1]
         store i32 %tmp.4, i32* %x
         ret void
 }
@@ -22,12 +22,12 @@ entry:
 define void @_Z5test2RiS_(i32* %x, i32* %y) {
 entry:
         %tmp.0 = alloca i32             ; <i32*> [#uses=2]
-        %tmp.2 = load i32* %x           ; <i32> [#uses=2]
+        %tmp.2 = load i32, i32* %x           ; <i32> [#uses=2]
         store i32 %tmp.2, i32* %tmp.0
-        %tmp.3.i = load i32* %y         ; <i32> [#uses=1]
+        %tmp.3.i = load i32, i32* %y         ; <i32> [#uses=1]
         %tmp.4.i = icmp slt i32 %tmp.2, %tmp.3.i                ; <i1> [#uses=1]
         %retval.i = select i1 %tmp.4.i, i32* %y, i32* %tmp.0            ; <i32*> [#uses=1]
-        %tmp.6 = load i32* %retval.i            ; <i32> [#uses=1]
+        %tmp.6 = load i32, i32* %retval.i            ; <i32> [#uses=1]
         store i32 %tmp.6, i32* %y
         ret void
 }
diff --git a/test/Transforms/InstCombine/LandingPadClauses.ll b/test/Transforms/InstCombine/LandingPadClauses.ll
index 10af4bcadf68..0d42f7c737f3 100644
--- a/test/Transforms/InstCombine/LandingPadClauses.ll
+++ b/test/Transforms/InstCombine/LandingPadClauses.ll
@@ -7,6 +7,7 @@
 declare i32 @generic_personality(i32, i64, i8*, i8*)
 declare i32 @__gxx_personality_v0(i32, i64, i8*, i8*)
 declare i32 @__objc_personality_v0(i32, i64, i8*, i8*)
+declare i32 @__C_specific_handler(...)
 
 declare void @bar()
 
@@ -231,3 +232,54 @@ lpad.d:
 ; CHECK-NEXT: null
 ; CHECK-NEXT: unreachable
 }
+
+define void @foo_seh() {
+; CHECK-LABEL: @foo_seh(
+  invoke void @bar()
+    to label %cont.a unwind label %lpad.a
+cont.a:
+  invoke void @bar()
+    to label %cont.b unwind label %lpad.b
+cont.b:
+  invoke void @bar()
+    to label %cont.c unwind label %lpad.c
+cont.c:
+  invoke void @bar()
+    to label %cont.d unwind label %lpad.d
+cont.d:
+  ret void
+
+lpad.a:
+  %a = landingpad { i8*, i32 } personality i32 (...)* @__C_specific_handler
+          catch i32* null
+          catch i32* @T1
+  unreachable
+; CHECK: %a = landingpad
+; CHECK-NEXT: null
+; CHECK-NEXT: unreachable
+
+lpad.b:
+  %b = landingpad { i8*, i32 } personality i32 (...)* @__C_specific_handler
+          filter [1 x i32*] zeroinitializer
+  unreachable
+; CHECK: %b = landingpad
+; CHECK-NEXT: cleanup
+; CHECK-NEXT: unreachable
+
+lpad.c:
+  %c = landingpad { i8*, i32 } personality i32 (...)* @__C_specific_handler
+          filter [2 x i32*] [i32* @T1, i32* null]
+  unreachable
+; CHECK: %c = landingpad
+; CHECK-NEXT: cleanup
+; CHECK-NEXT: unreachable
+
+lpad.d:
+  %d = landingpad { i8*, i32 } personality i32 (...)* @__C_specific_handler
+          cleanup
+          catch i32* null
+  unreachable
+; CHECK: %d = landingpad
+; CHECK-NEXT: null
+; CHECK-NEXT: unreachable
+}
diff --git a/test/Transforms/InstCombine/PR7357.ll b/test/Transforms/InstCombine/PR7357.ll
index c397c5ee05e3..616c8225b2ac 100644
--- a/test/Transforms/InstCombine/PR7357.ll
+++ b/test/Transforms/InstCombine/PR7357.ll
@@ -8,7 +8,7 @@
 ; CHECK: call arm_aapcscc  i8* @strchr
 define arm_aapcscc i32 @foo(i32 %argc) nounwind {
 bb.nph:
-  %c = call arm_aapcscc  i8* @strchr(i8* getelementptr ([11 x i8]* @.str1, i32 0,
+  %c = call arm_aapcscc  i8* @strchr(i8* getelementptr ([11 x i8], [11 x i8]* @.str1, i32 0,
 i32 0), i32 %argc) nounwind readonly
   %p = ptrtoint i8* %c to i32
   ret i32 %p
diff --git a/test/Transforms/InstCombine/add2.ll b/test/Transforms/InstCombine/add2.ll
index fbbba5954b32..1af2b0ffbf0a 100644
--- a/test/Transforms/InstCombine/add2.ll
+++ b/test/Transforms/InstCombine/add2.ll
@@ -273,6 +273,35 @@ define i32 @mul_add_to_mul_6(i32 %x, i32 %y) {
 ; CHECK-NEXT: ret i32 %add
 }
 
+define i16 @mul_add_to_mul_7(i16 %x) {
+  %mul1 = mul nsw i16 %x, 32767
+  %add2 = add nsw i16 %x, %mul1
+  ret i16 %add2
+; CHECK-LABEL: @mul_add_to_mul_7(
+; CHECK-NEXT: %add2 = shl i16 %x, 15
+; CHECK-NEXT: ret i16 %add2
+}
+
+define i16 @mul_add_to_mul_8(i16 %a) {
+  %mul1 = mul nsw i16 %a, 16383
+  %mul2 = mul nsw i16 %a, 16384
+  %add = add nsw i16 %mul1, %mul2
+  ret i16 %add
+; CHECK-LABEL: @mul_add_to_mul_8(
+; CHECK-NEXT: %add = mul nsw i16 %a, 32767
+; CHECK-NEXT: ret i16 %add
+}
+
+define i16 @mul_add_to_mul_9(i16 %a) {
+  %mul1 = mul nsw i16 %a, 16384
+  %mul2 = mul nsw i16 %a, 16384
+  %add = add nsw i16 %mul1, %mul2
+  ret i16 %add
+; CHECK-LABEL: @mul_add_to_mul_9(
+; CHECK-NEXT: %add = shl i16 %a, 15
+; CHECK-NEXT: ret i16 %add
+}
+
 ; This test and the next test verify that when a range metadata is attached to
 ; llvm.cttz, ValueTracking correctly intersects the range specified by the
 ; metadata and the range implied by the intrinsic.
@@ -353,3 +382,16 @@ define i32 @add_nuw_nsw_or_and(i32 %x, i32 %y) {
 ; CHECK-NEXT: add nuw nsw i32 %x, %y
 ; CHECK-NEXT: ret i32
 }
+
+; A *nsw B + A *nsw C != A *nsw (B + C)
+; e.g. A = -1, B = 1, C = INT_SMAX
+
+define i8 @add_of_mul(i8 %x, i8 %y, i8 %z) {
+; CHECK-LABEL: @add_of_mul(
+ entry:
+  %mA = mul nsw i8 %x, %y
+  %mB = mul nsw i8 %x, %z
+; CHECK: %sum = mul i8
+  %sum = add nsw i8 %mA, %mB
+  ret i8 %sum
+}
diff --git a/test/Transforms/InstCombine/add3.ll b/test/Transforms/InstCombine/add3.ll
index cde3e24215f8..9d3842f1a31a 100644
--- a/test/Transforms/InstCombine/add3.ll
+++ b/test/Transforms/InstCombine/add3.ll
@@ -12,8 +12,8 @@ EntryBlock:
         store i32 1, i32* %1
         add i32 %.val24, -16
         inttoptr i32 %2 to i32*
-        getelementptr i32* %3, i32 1
-        load i32* %4
+        getelementptr i32, i32* %3, i32 1
+        load i32, i32* %4
         tail call i32 @callee( i32 %5 )
         ret void
 }
diff --git a/test/Transforms/InstCombine/addrspacecast.ll b/test/Transforms/InstCombine/addrspacecast.ll
index c1684361f99f..27f6b720d667 100644
--- a/test/Transforms/InstCombine/addrspacecast.ll
+++ b/test/Transforms/InstCombine/addrspacecast.ll
@@ -104,12 +104,12 @@ define <4 x float addrspace(2)*> @combine_addrspacecast_types_vector(<4 x i32 ad
 
 define i32 @canonicalize_addrspacecast([16 x i32] addrspace(1)* %arr) {
 ; CHECK-LABEL: @canonicalize_addrspacecast(
-; CHECK-NEXT: getelementptr inbounds [16 x i32] addrspace(1)* %arr, i32 0, i32 0
+; CHECK-NEXT: getelementptr inbounds [16 x i32], [16 x i32] addrspace(1)* %arr, i32 0, i32 0
 ; CHECK-NEXT: addrspacecast i32 addrspace(1)* %{{[a-zA-Z0-9]+}} to i32*
-; CHECK-NEXT: load i32*
+; CHECK-NEXT: load i32, i32*
 ; CHECK-NEXT: ret i32
   %p = addrspacecast [16 x i32] addrspace(1)* %arr to i32*
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -127,14 +127,14 @@ declare void @foo(i8*) nounwind
 define i32 @memcpy_addrspacecast() nounwind {
 entry:
   %alloca = alloca i8, i32 48
-  call void @llvm.memcpy.p0i8.p1i8.i32(i8* %alloca, i8 addrspace(1)* addrspacecast (i8 addrspace(2)* getelementptr inbounds ([60 x i8] addrspace(2)* @const_array, i16 0, i16 4) to i8 addrspace(1)*), i32 48, i32 4, i1 false) nounwind
+  call void @llvm.memcpy.p0i8.p1i8.i32(i8* %alloca, i8 addrspace(1)* addrspacecast (i8 addrspace(2)* getelementptr inbounds ([60 x i8], [60 x i8] addrspace(2)* @const_array, i16 0, i16 4) to i8 addrspace(1)*), i32 48, i32 4, i1 false) nounwind
   br label %loop.body
 
 loop.body:
   %i = phi i32 [ 0, %entry ], [ %i.inc, %loop.body ]
   %sum = phi i32 [ 0, %entry ], [ %sum.inc, %loop.body]
-  %ptr = getelementptr i8* %alloca, i32 %i
-  %load = load i8* %ptr
+  %ptr = getelementptr i8, i8* %alloca, i32 %i
+  %load = load i8, i8* %ptr
   %ext = zext i8 %load to i32
   %sum.inc = add i32 %sum, %ext
   %i.inc = add i32 %i, 1
diff --git a/test/Transforms/InstCombine/alias-recursion.ll b/test/Transforms/InstCombine/alias-recursion.ll
index fa63726d2d33..74254f3e8dff 100644
--- a/test/Transforms/InstCombine/alias-recursion.ll
+++ b/test/Transforms/InstCombine/alias-recursion.ll
@@ -7,7 +7,7 @@ target triple = "x86_64-pc-windows-msvc"
 
 @0 = constant [1 x i8*] zeroinitializer
 
-@vtbl = alias getelementptr inbounds ([1 x i8*]* @0, i32 0, i32 0)
+@vtbl = alias getelementptr inbounds ([1 x i8*], [1 x i8*]* @0, i32 0, i32 0)
 
 define i32 (%class.A*)* @test() {
 ; CHECK-LABEL: test
@@ -19,6 +19,6 @@ for.body:                                         ; preds = %for.body, %entry
 
 for.end:                                          ; preds = %for.body, %entry
   %A = phi i32 (%class.A*)** [ bitcast (i8** @vtbl to i32 (%class.A*)**), %for.body ], [ null, %entry ]
-  %B = load i32 (%class.A*)** %A
+  %B = load i32 (%class.A*)*, i32 (%class.A*)** %A
   ret i32 (%class.A*)* %B
 }
diff --git a/test/Transforms/InstCombine/align-2d-gep.ll b/test/Transforms/InstCombine/align-2d-gep.ll
index f6a877684ce4..bbdb3f94cb28 100644
--- a/test/Transforms/InstCombine/align-2d-gep.ll
+++ b/test/Transforms/InstCombine/align-2d-gep.ll
@@ -22,11 +22,11 @@ bb7.outer:
 bb1:
   %j = phi i64 [ 0, %bb7.outer ], [ %indvar.next, %bb1 ]
 
-  %t4 = getelementptr [1001 x [20000 x double]]* @Nice, i64 0, i64 %i, i64 %j
+  %t4 = getelementptr [1001 x [20000 x double]], [1001 x [20000 x double]]* @Nice, i64 0, i64 %i, i64 %j
   %q = bitcast double* %t4 to <2 x double>*
   store <2 x double><double 0.0, double 0.0>, <2 x double>* %q, align 8
 
-  %s4 = getelementptr [1001 x [20001 x double]]* @Awkward, i64 0, i64 %i, i64 %j
+  %s4 = getelementptr [1001 x [20001 x double]], [1001 x [20001 x double]]* @Awkward, i64 0, i64 %i, i64 %j
   %r = bitcast double* %s4 to <2 x double>*
   store <2 x double><double 0.0, double 0.0>, <2 x double>* %r, align 8
 
diff --git a/test/Transforms/InstCombine/align-addr.ll b/test/Transforms/InstCombine/align-addr.ll
index 4d22c2cd2ef2..ab0ae390b145 100644
--- a/test/Transforms/InstCombine/align-addr.ll
+++ b/test/Transforms/InstCombine/align-addr.ll
@@ -21,7 +21,7 @@ bb:
   %i = phi i64 [ %indvar.next, %bb ], [ 20, %entry ]
   %j = mul i64 %i, %v
   %h = add i64 %j, %z
-  %t8 = getelementptr double* %e, i64 %h
+  %t8 = getelementptr double, double* %e, i64 %h
   %p = bitcast double* %t8 to <2 x double>*
   store <2 x double><double 0.0, double 0.0>, <2 x double>* %p, align 8
   %indvar.next = add i64 %i, 1
@@ -43,7 +43,7 @@ return:
 
 define <16 x i8> @test1(<2 x i64> %x) {
 entry:
-	%tmp = load <16 x i8>* bitcast ([4 x i32]* @GLOBAL to <16 x i8>*), align 1
+	%tmp = load <16 x i8>, <16 x i8>* bitcast ([4 x i32]* @GLOBAL to <16 x i8>*), align 1
 	ret <16 x i8> %tmp
 }
 
@@ -53,7 +53,7 @@ define <16 x i8> @test1_as1(<2 x i64> %x) {
 ; CHECK-LABEL: @test1_as1(
 ; CHECK: tmp = load
 ; CHECK: GLOBAL_as1{{.*}}align 16
-  %tmp = load <16 x i8> addrspace(1)* bitcast ([4 x i32] addrspace(1)* @GLOBAL_as1 to <16 x i8> addrspace(1)*), align 1
+  %tmp = load <16 x i8>, <16 x i8> addrspace(1)* bitcast ([4 x i32] addrspace(1)* @GLOBAL_as1 to <16 x i8> addrspace(1)*), align 1
   ret <16 x i8> %tmp
 }
 
@@ -63,7 +63,7 @@ define <16 x i8> @test1_as1_gep(<2 x i64> %x) {
 ; CHECK-LABEL: @test1_as1_gep(
 ; CHECK: tmp = load
 ; CHECK: GLOBAL_as1_gep{{.*}}align 16
-  %tmp = load <16 x i8> addrspace(1)* bitcast (i32 addrspace(1)* getelementptr ([8 x i32] addrspace(1)* @GLOBAL_as1_gep, i16 0, i16 4) to <16 x i8> addrspace(1)*), align 1
+  %tmp = load <16 x i8>, <16 x i8> addrspace(1)* bitcast (i32 addrspace(1)* getelementptr ([8 x i32], [8 x i32] addrspace(1)* @GLOBAL_as1_gep, i16 0, i16 4) to <16 x i8> addrspace(1)*), align 1
   ret <16 x i8> %tmp
 }
 
@@ -71,11 +71,11 @@ define <16 x i8> @test1_as1_gep(<2 x i64> %x) {
 ; When a load or store lacks an explicit alignment, add one.
 
 ; CHECK-LABEL: @test2(
-; CHECK: load double* %p, align 8
+; CHECK: load double, double* %p, align 8
 ; CHECK: store double %n, double* %p, align 8
 
 define double @test2(double* %p, double %n) nounwind {
-  %t = load double* %p
+  %t = load double, double* %p
   store double %n, double* %p
   ret double %t
 }
diff --git a/test/Transforms/InstCombine/align-attr.ll b/test/Transforms/InstCombine/align-attr.ll
index 9f366bf8fab1..99a17db13c4c 100644
--- a/test/Transforms/InstCombine/align-attr.ll
+++ b/test/Transforms/InstCombine/align-attr.ll
@@ -5,11 +5,11 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define i32 @foo1(i32* align 32 %a) #0 {
 entry:
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 
 ; CHECK-LABEL: @foo1
-; CHECK-DAG: load i32* %a, align 32
+; CHECK-DAG: load i32, i32* %a, align 32
 ; CHECK: ret i32
 }
 
diff --git a/test/Transforms/InstCombine/align-external.ll b/test/Transforms/InstCombine/align-external.ll
index 66ff9c16e424..ee98a0120179 100644
--- a/test/Transforms/InstCombine/align-external.ll
+++ b/test/Transforms/InstCombine/align-external.ll
@@ -23,7 +23,7 @@ define i64 @foo(i64 %a) {
 
 define i32 @bar() {
 ; CHECK-LABEL: @bar(
-  %r = load i32* @B, align 1
+  %r = load i32, i32* @B, align 1
 ; CHECK: align 1
   ret i32 %r
 }
diff --git a/test/Transforms/InstCombine/aligned-altivec.ll b/test/Transforms/InstCombine/aligned-altivec.ll
new file mode 100644
index 000000000000..10b4e4d62631
--- /dev/null
+++ b/test/Transforms/InstCombine/aligned-altivec.ll
@@ -0,0 +1,131 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+declare <4 x i32> @llvm.ppc.altivec.lvx(i8*) #1
+
+define <4 x i32> @test1(<4 x i32>* %h) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  %vl = call <4 x i32> @llvm.ppc.altivec.lvx(i8* %hv)
+
+; CHECK-LABEL: @test1
+; CHECK: @llvm.ppc.altivec.lvx
+; CHECK: ret <4 x i32>
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  %a = add <4 x i32> %v0, %vl
+  ret <4 x i32> %a
+}
+
+define <4 x i32> @test1a(<4 x i32>* align 16 %h) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  %vl = call <4 x i32> @llvm.ppc.altivec.lvx(i8* %hv)
+
+; CHECK-LABEL: @test1a
+; CHECK-NOT: @llvm.ppc.altivec.lvx
+; CHECK: ret <4 x i32>
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  %a = add <4 x i32> %v0, %vl
+  ret <4 x i32> %a
+}
+
+declare void @llvm.ppc.altivec.stvx(<4 x i32>, i8*) #0
+
+define <4 x i32> @test2(<4 x i32>* %h, <4 x i32> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  call void @llvm.ppc.altivec.stvx(<4 x i32> %d, i8* %hv)
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  ret <4 x i32> %v0
+
+; CHECK-LABEL: @test2
+; CHECK: @llvm.ppc.altivec.stvx
+; CHECK: ret <4 x i32>
+}
+
+define <4 x i32> @test2a(<4 x i32>* align 16 %h, <4 x i32> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  call void @llvm.ppc.altivec.stvx(<4 x i32> %d, i8* %hv)
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  ret <4 x i32> %v0
+
+; CHECK-LABEL: @test2
+; CHECK-NOT: @llvm.ppc.altivec.stvx
+; CHECK: ret <4 x i32>
+}
+
+declare <4 x i32> @llvm.ppc.altivec.lvxl(i8*) #1
+
+define <4 x i32> @test1l(<4 x i32>* %h) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  %vl = call <4 x i32> @llvm.ppc.altivec.lvxl(i8* %hv)
+
+; CHECK-LABEL: @test1l
+; CHECK: @llvm.ppc.altivec.lvxl
+; CHECK: ret <4 x i32>
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  %a = add <4 x i32> %v0, %vl
+  ret <4 x i32> %a
+}
+
+define <4 x i32> @test1la(<4 x i32>* align 16 %h) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  %vl = call <4 x i32> @llvm.ppc.altivec.lvxl(i8* %hv)
+
+; CHECK-LABEL: @test1la
+; CHECK-NOT: @llvm.ppc.altivec.lvxl
+; CHECK: ret <4 x i32>
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  %a = add <4 x i32> %v0, %vl
+  ret <4 x i32> %a
+}
+
+declare void @llvm.ppc.altivec.stvxl(<4 x i32>, i8*) #0
+
+define <4 x i32> @test2l(<4 x i32>* %h, <4 x i32> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  call void @llvm.ppc.altivec.stvxl(<4 x i32> %d, i8* %hv)
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  ret <4 x i32> %v0
+
+; CHECK-LABEL: @test2l
+; CHECK: @llvm.ppc.altivec.stvxl
+; CHECK: ret <4 x i32>
+}
+
+define <4 x i32> @test2la(<4 x i32>* align 16 %h, <4 x i32> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x i32>, <4 x i32>* %h, i64 1
+  %hv = bitcast <4 x i32>* %h1 to i8*
+  call void @llvm.ppc.altivec.stvxl(<4 x i32> %d, i8* %hv)
+
+  %v0 = load <4 x i32>, <4 x i32>* %h, align 8
+  ret <4 x i32> %v0
+
+; CHECK-LABEL: @test2l
+; CHECK-NOT: @llvm.ppc.altivec.stvxl
+; CHECK: ret <4 x i32>
+}
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readonly }
+
diff --git a/test/Transforms/InstCombine/aligned-qpx.ll b/test/Transforms/InstCombine/aligned-qpx.ll
new file mode 100644
index 000000000000..e9710df5670c
--- /dev/null
+++ b/test/Transforms/InstCombine/aligned-qpx.ll
@@ -0,0 +1,165 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f128:128:128-v128:128:128-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+declare <4 x double> @llvm.ppc.qpx.qvlfs(i8*) #1
+
+define <4 x double> @test1(<4 x float>* %h) #0 {
+entry:
+  %h1 = getelementptr <4 x float>, <4 x float>* %h, i64 1
+  %hv = bitcast <4 x float>* %h1 to i8*
+  %vl = call <4 x double> @llvm.ppc.qpx.qvlfs(i8* %hv)
+
+; CHECK-LABEL: @test1
+; CHECK: @llvm.ppc.qpx.qvlfs
+; CHECK: ret <4 x double>
+
+  %v0 = load <4 x float>, <4 x float>* %h, align 8
+  %v0e = fpext <4 x float> %v0 to <4 x double>
+  %a = fadd <4 x double> %v0e, %vl
+  ret <4 x double> %a
+}
+
+define <4 x double> @test1a(<4 x float>* align 16 %h) #0 {
+entry:
+  %h1 = getelementptr <4 x float>, <4 x float>* %h, i64 1
+  %hv = bitcast <4 x float>* %h1 to i8*
+  %vl = call <4 x double> @llvm.ppc.qpx.qvlfs(i8* %hv)
+
+; CHECK-LABEL: @test1a
+; CHECK-NOT: @llvm.ppc.qpx.qvlfs
+; CHECK-NOT: load <4 x double>
+; CHECK: ret <4 x double>
+
+  %v0 = load <4 x float>, <4 x float>* %h, align 8
+  %v0e = fpext <4 x float> %v0 to <4 x double>
+  %a = fadd <4 x double> %v0e, %vl
+  ret <4 x double> %a
+}
+
+declare void @llvm.ppc.qpx.qvstfs(<4 x double>, i8*) #0
+
+define <4 x float> @test2(<4 x float>* %h, <4 x double> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x float>, <4 x float>* %h, i64 1
+  %hv = bitcast <4 x float>* %h1 to i8*
+  call void @llvm.ppc.qpx.qvstfs(<4 x double> %d, i8* %hv)
+
+  %v0 = load <4 x float>, <4 x float>* %h, align 8
+  ret <4 x float> %v0
+
+; CHECK-LABEL: @test2
+; CHECK: @llvm.ppc.qpx.qvstfs
+; CHECK: ret <4 x float>
+}
+
+define <4 x float> @test2a(<4 x float>* align 16 %h, <4 x double> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x float>, <4 x float>* %h, i64 1
+  %hv = bitcast <4 x float>* %h1 to i8*
+  call void @llvm.ppc.qpx.qvstfs(<4 x double> %d, i8* %hv)
+
+  %v0 = load <4 x float>, <4 x float>* %h, align 8
+  ret <4 x float> %v0
+
+; CHECK-LABEL: @test2
+; CHECK: fptrunc <4 x double> %d to <4 x float>
+; CHECK-NOT: @llvm.ppc.qpx.qvstfs
+; CHECK-NOT: store <4 x double>
+; CHECK: ret <4 x float>
+}
+
+declare <4 x double> @llvm.ppc.qpx.qvlfd(i8*) #1
+
+define <4 x double> @test1l(<4 x double>* %h) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  %vl = call <4 x double> @llvm.ppc.qpx.qvlfd(i8* %hv)
+
+; CHECK-LABEL: @test1l
+; CHECK: @llvm.ppc.qpx.qvlfd
+; CHECK: ret <4 x double>
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  %a = fadd <4 x double> %v0, %vl
+  ret <4 x double> %a
+}
+
+define <4 x double> @test1ln(<4 x double>* align 16 %h) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  %vl = call <4 x double> @llvm.ppc.qpx.qvlfd(i8* %hv)
+
+; CHECK-LABEL: @test1ln
+; CHECK: @llvm.ppc.qpx.qvlfd
+; CHECK: ret <4 x double>
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  %a = fadd <4 x double> %v0, %vl
+  ret <4 x double> %a
+}
+
+define <4 x double> @test1la(<4 x double>* align 32 %h) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  %vl = call <4 x double> @llvm.ppc.qpx.qvlfd(i8* %hv)
+
+; CHECK-LABEL: @test1la
+; CHECK-NOT: @llvm.ppc.qpx.qvlfd
+; CHECK: ret <4 x double>
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  %a = fadd <4 x double> %v0, %vl
+  ret <4 x double> %a
+}
+
+declare void @llvm.ppc.qpx.qvstfd(<4 x double>, i8*) #0
+
+define <4 x double> @test2l(<4 x double>* %h, <4 x double> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  call void @llvm.ppc.qpx.qvstfd(<4 x double> %d, i8* %hv)
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  ret <4 x double> %v0
+
+; CHECK-LABEL: @test2l
+; CHECK: @llvm.ppc.qpx.qvstfd
+; CHECK: ret <4 x double>
+}
+
+define <4 x double> @test2ln(<4 x double>* align 16 %h, <4 x double> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  call void @llvm.ppc.qpx.qvstfd(<4 x double> %d, i8* %hv)
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  ret <4 x double> %v0
+
+; CHECK-LABEL: @test2ln
+; CHECK: @llvm.ppc.qpx.qvstfd
+; CHECK: ret <4 x double>
+}
+
+define <4 x double> @test2la(<4 x double>* align 32 %h, <4 x double> %d) #0 {
+entry:
+  %h1 = getelementptr <4 x double>, <4 x double>* %h, i64 1
+  %hv = bitcast <4 x double>* %h1 to i8*
+  call void @llvm.ppc.qpx.qvstfd(<4 x double> %d, i8* %hv)
+
+  %v0 = load <4 x double>, <4 x double>* %h, align 8
+  ret <4 x double> %v0
+
+; CHECK-LABEL: @test2l
+; CHECK-NOT: @llvm.ppc.qpx.qvstfd
+; CHECK: ret <4 x double>
+}
+
+attributes #0 = { nounwind }
+attributes #1 = { nounwind readonly }
+
diff --git a/test/Transforms/InstCombine/alloca.ll b/test/Transforms/InstCombine/alloca.ll
index 6d0c131c51ed..b61b75e9f9f3 100644
--- a/test/Transforms/InstCombine/alloca.ll
+++ b/test/Transforms/InstCombine/alloca.ll
@@ -1,6 +1,6 @@
-; RUN: opt < %s -instcombine -S -default-data-layout="E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128" | FileCheck %s
-; RUN: opt < %s -instcombine -S -default-data-layout="E-p:32:32:32-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128" | FileCheck %s -check-prefix=P32
-; RUN: opt < %s -instcombine -S | FileCheck %s -check-prefix=NODL
+; RUN: opt < %s -instcombine -S -default-data-layout="E-p:64:64:64-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128" | FileCheck %s -check-prefix=CHECK -check-prefix=ALL
+; RUN: opt < %s -instcombine -S -default-data-layout="E-p:32:32:32-a0:0:8-f32:32:32-f64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-v64:64:64-v128:128:128" | FileCheck %s -check-prefix=P32 -check-prefix=ALL
+; RUN: opt < %s -instcombine -S | FileCheck %s -check-prefix=NODL -check-prefix=ALL
 
 
 declare void @use(...)
@@ -13,14 +13,14 @@ declare void @use(...)
 ; CHECK-NOT: alloca
 define void @test() {
         %X = alloca [0 x i32]           ; <[0 x i32]*> [#uses=1]
-        call void (...)* @use( [0 x i32]* %X )
+        call void (...) @use( [0 x i32]* %X )
         %Y = alloca i32, i32 0          ; <i32*> [#uses=1]
-        call void (...)* @use( i32* %Y )
+        call void (...) @use( i32* %Y )
         %Z = alloca {  }                ; <{  }*> [#uses=1]
-        call void (...)* @use( {  }* %Z )
-        %size = load i32* @int
+        call void (...) @use( {  }* %Z )
+        %size = load i32, i32* @int
         %A = alloca {{}}, i32 %size
-        call void (...)* @use( {{}}* %A )
+        call void (...) @use( {{}}* %A )
         ret void
 }
 
@@ -38,7 +38,7 @@ define void @test2() {
 ; CHECK-NOT: alloca
 define void @test3() {
         %A = alloca { i32 }             ; <{ i32 }*> [#uses=1]
-        %B = getelementptr { i32 }* %A, i32 0, i32 0            ; <i32*> [#uses=1]
+        %B = getelementptr { i32 }, { i32 }* %A, i32 0, i32 0            ; <i32*> [#uses=1]
         store i32 123, i32* %B
         ret void
 }
@@ -62,16 +62,16 @@ define void @test5() {
 entry:
   %a = alloca { i32 }
   %b = alloca i32*
-  %a.1 = getelementptr { i32 }* %a, i32 0, i32 0
+  %a.1 = getelementptr { i32 }, { i32 }* %a, i32 0, i32 0
   store i32 123, i32* %a.1
   store i32* %a.1, i32** %b
   %b.1 = bitcast i32** %b to i32*
   store i32 123, i32* %b.1
-  %a.2 = getelementptr { i32 }* %a, i32 0, i32 0
+  %a.2 = getelementptr { i32 }, { i32 }* %a, i32 0, i32 0
   store atomic i32 2, i32* %a.2 unordered, align 4
-  %a.3 = getelementptr { i32 }* %a, i32 0, i32 0
+  %a.3 = getelementptr { i32 }, { i32 }* %a, i32 0, i32 0
   store atomic i32 3, i32* %a.3 release, align 4
-  %a.4 = getelementptr { i32 }* %a, i32 0, i32 0
+  %a.4 = getelementptr { i32 }, { i32 }* %a, i32 0, i32 0
   store atomic i32 4, i32* %a.4 seq_cst, align 4
   ret void
 }
@@ -89,7 +89,7 @@ define void @test6() {
 entry:
   %a = alloca { i32 }
   %b = alloca i32
-  %a.1 = getelementptr { i32 }* %a, i32 0, i32 0
+  %a.1 = getelementptr { i32 }, { i32 }* %a, i32 0, i32 0
   store volatile i32 123, i32* %a.1
   tail call void @f(i32* %b)
   ret void
@@ -116,17 +116,17 @@ declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32,
 define void @test8() {
 ; CHECK-LABEL: @test8(
 ; CHECK: alloca [100 x i32]
-; CHECK: getelementptr inbounds [100 x i32]* %x1, i64 0, i64 0
+; CHECK: getelementptr inbounds [100 x i32], [100 x i32]* %x1, i64 0, i64 0
 
 ; P32-LABEL: @test8(
 ; P32: alloca [100 x i32]
-; P32: getelementptr inbounds [100 x i32]* %x1, i32 0, i32 0
+; P32: getelementptr inbounds [100 x i32], [100 x i32]* %x1, i32 0, i32 0
 
 ; NODL-LABEL: @test8(
 ; NODL: alloca [100 x i32]
-; NODL: getelementptr inbounds [100 x i32]* %x1, i64 0, i64 0
+; NODL: getelementptr inbounds [100 x i32], [100 x i32]* %x1, i64 0, i64 0
   %x = alloca i32, i32 100
-  call void (...)* @use(i32* %x)
+  call void (...) @use(i32* %x)
   ret void
 }
 
@@ -142,7 +142,7 @@ entry:
   %inalloca.save = call i8* @llvm.stacksave()
   %argmem = alloca inalloca <{ %struct_type }>
 ; CHECK: alloca inalloca i64, align 8
-  %0 = getelementptr inbounds <{ %struct_type }>* %argmem, i32 0, i32 0
+  %0 = getelementptr inbounds <{ %struct_type }>, <{ %struct_type }>* %argmem, i32 0, i32 0
   %1 = bitcast %struct_type* %0 to i8*
   %2 = bitcast %struct_type* %a to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %1, i8* %2, i32 8, i32 4, i1 false)
@@ -150,3 +150,16 @@ entry:
   call void @llvm.stackrestore(i8* %inalloca.save)
   ret void
 }
+
+define void @test10() {
+entry:
+; ALL-LABEL: @test10(
+; ALL: %v32 = alloca i1, align 8
+; ALL: %v64 = alloca i1, align 8
+; ALL: %v33 = alloca i1, align 8
+  %v32 = alloca i1, align 8
+  %v64 = alloca i1, i64 1, align 8
+  %v33 = alloca i1, i33 1, align 8
+  call void (...) @use(i1* %v32, i1* %v64, i1* %v33)
+  ret void
+}
diff --git a/test/Transforms/InstCombine/assume-loop-align.ll b/test/Transforms/InstCombine/assume-loop-align.ll
index 19190de2cddd..e803ba61774d 100644
--- a/test/Transforms/InstCombine/assume-loop-align.ll
+++ b/test/Transforms/InstCombine/assume-loop-align.ll
@@ -19,16 +19,16 @@ entry:
   br label %for.body
 
 ; CHECK-LABEL: @foo
-; CHECK: load i32* {{.*}} align 64
+; CHECK: load i32, i32* {{.*}} align 64
 ; CHECK: store i32 {{.*}}  align 64
 ; CHECK: ret
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, 1
-  %arrayidx5 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx5 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %add, i32* %arrayidx5, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 16
   %1 = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/InstCombine/assume-redundant.ll b/test/Transforms/InstCombine/assume-redundant.ll
index 81fe0945949b..4b869ef2c50e 100644
--- a/test/Transforms/InstCombine/assume-redundant.ll
+++ b/test/Transforms/InstCombine/assume-redundant.ll
@@ -15,8 +15,8 @@ define void @_Z3fooR1s(%struct.s* nocapture readonly dereferenceable(8) %x) #0 {
 ; CHECK-NOT: call void @llvm.assume
 
 entry:
-  %a = getelementptr inbounds %struct.s* %x, i64 0, i32 0
-  %0 = load double** %a, align 8
+  %a = getelementptr inbounds %struct.s, %struct.s* %x, i64 0, i32 0
+  %0 = load double*, double** %a, align 8
   %ptrint = ptrtoint double* %0 to i64
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
@@ -25,16 +25,16 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next.1, %for.body ]
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds double* %0, i64 %indvars.iv
-  %1 = load double* %arrayidx, align 16
+  %arrayidx = getelementptr inbounds double, double* %0, i64 %indvars.iv
+  %1 = load double, double* %arrayidx, align 16
   %add = fadd double %1, 1.000000e+00
   tail call void @llvm.assume(i1 %maskcond)
   %mul = fmul double %add, 2.000000e+00
   store double %mul, double* %arrayidx, align 16
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx.1 = getelementptr inbounds double* %0, i64 %indvars.iv.next
-  %2 = load double* %arrayidx.1, align 8
+  %arrayidx.1 = getelementptr inbounds double, double* %0, i64 %indvars.iv.next
+  %2 = load double, double* %arrayidx.1, align 8
   %add.1 = fadd double %2, 1.000000e+00
   tail call void @llvm.assume(i1 %maskcond)
   %mul.1 = fmul double %add.1, 2.000000e+00
diff --git a/test/Transforms/InstCombine/assume.ll b/test/Transforms/InstCombine/assume.ll
index 7e45c04622ac..2f9213820f2b 100644
--- a/test/Transforms/InstCombine/assume.ll
+++ b/test/Transforms/InstCombine/assume.ll
@@ -5,12 +5,12 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define i32 @foo1(i32* %a) #0 {
 entry:
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
 
 ; Check that the alignment has been upgraded and that the assume has not
 ; been removed:
 ; CHECK-LABEL: @foo1
-; CHECK-DAG: load i32* %a, align 32
+; CHECK-DAG: load i32, i32* %a, align 32
 ; CHECK-DAG: call void @llvm.assume
 ; CHECK: ret i32
 
@@ -27,7 +27,7 @@ define i32 @foo2(i32* %a) #0 {
 entry:
 ; Same check as in @foo1, but make sure it works if the assume is first too.
 ; CHECK-LABEL: @foo2
-; CHECK-DAG: load i32* %a, align 32
+; CHECK-DAG: load i32, i32* %a, align 32
 ; CHECK-DAG: call void @llvm.assume
 ; CHECK: ret i32
 
@@ -36,7 +36,7 @@ entry:
   %maskcond = icmp eq i64 %maskedptr, 0
   tail call void @llvm.assume(i1 %maskcond)
 
-  %0 = load i32* %a, align 4
+  %0 = load i32, i32* %a, align 4
   ret i32 %0
 }
 
@@ -192,7 +192,7 @@ declare void @escape(i32* %a)
 ; metadata form?
 define i1 @nonnull1(i32** %a) {
 entry:
-  %load = load i32** %a
+  %load = load i32*, i32** %a
   %cmp = icmp ne i32* %load, null
   tail call void @llvm.assume(i1 %cmp)
   tail call void @escape(i32* %load)
@@ -209,7 +209,7 @@ entry:
 ; to pointer types.  Doing otherwise would be illegal.
 define i1 @nonnull2(i32* %a) {
 entry:
-  %load = load i32* %a
+  %load = load i32, i32* %a
   %cmp = icmp ne i32 %load, 0
   tail call void @llvm.assume(i1 %cmp)
   %rval = icmp eq i32 %load, 0
@@ -224,7 +224,7 @@ entry:
 ; if the assume is control dependent on something else
 define i1 @nonnull3(i32** %a, i1 %control) {
 entry:
-  %load = load i32** %a
+  %load = load i32*, i32** %a
   %cmp = icmp ne i32* %load, null
   br i1 %control, label %taken, label %not_taken
 taken:
@@ -244,7 +244,7 @@ not_taken:
 ; interrupted by an exception being thrown
 define i1 @nonnull4(i32** %a) {
 entry:
-  %load = load i32** %a
+  %load = load i32*, i32** %a
   ;; This call may throw!
   tail call void @escape(i32* %load)
   %cmp = icmp ne i32* %load, null
diff --git a/test/Transforms/InstCombine/atomic.ll b/test/Transforms/InstCombine/atomic.ll
index 98cecefcc294..5754a5a4ba56 100644
--- a/test/Transforms/InstCombine/atomic.ll
+++ b/test/Transforms/InstCombine/atomic.ll
@@ -7,10 +7,10 @@ target triple = "x86_64-apple-macosx10.7.0"
 
 define i32 @test2(i32* %p) {
 ; CHECK-LABEL: define i32 @test2(
-; CHECK: %x = load atomic i32* %p seq_cst, align 4
+; CHECK: %x = load atomic i32, i32* %p seq_cst, align 4
 ; CHECK: shl i32 %x, 1
-  %x = load atomic i32* %p seq_cst, align 4
-  %y = load i32* %p, align 4
+  %x = load atomic i32, i32* %p seq_cst, align 4
+  %y = load i32, i32* %p, align 4
   %z = add i32 %x, %y
   ret i32 %z
 }
diff --git a/test/Transforms/InstCombine/bitcast-alias-function.ll b/test/Transforms/InstCombine/bitcast-alias-function.ll
index cfec09200dbc..1a598a5d4153 100644
--- a/test/Transforms/InstCombine/bitcast-alias-function.ll
+++ b/test/Transforms/InstCombine/bitcast-alias-function.ll
@@ -91,12 +91,12 @@ define void @bitcast_alias_scalar(float* noalias %source, float* noalias %dest)
 entry:
 ; CHECK-LABEL: @bitcast_alias_scalar
 ; CHECK: bitcast float* %source to i32*
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NOT: fptoui
 ; CHECK-NOT: uitofp
 ; CHECK: bitcast float* %dest to i32*
 ; CHECK: store i32
-  %tmp = load float* %source, align 8
+  %tmp = load float, float* %source, align 8
   %call = call float @alias_i32_to_f32(float %tmp) nounwind
   store float %call, float* %dest, align 8
   ret void
@@ -107,12 +107,12 @@ define void @bitcast_alias_vector(<2 x float>* noalias %source, <2 x float>* noa
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector
 ; CHECK: bitcast <2 x float>* %source to <2 x i32>*
-; CHECK: load <2 x i32>*
+; CHECK: load <2 x i32>, <2 x i32>*
 ; CHECK-NOT: fptoui
 ; CHECK-NOT: uitofp
 ; CHECK: bitcast <2 x float>* %dest to <2 x i32>*
 ; CHECK: store <2 x i32>
-  %tmp = load <2 x float>* %source, align 8
+  %tmp = load <2 x float>, <2 x float>* %source, align 8
   %call = call <2 x float> @alias_v2i32_to_v2f32(<2 x float> %tmp) nounwind
   store <2 x float> %call, <2 x float>* %dest, align 8
   ret void
@@ -123,11 +123,11 @@ define void @bitcast_alias_vector_scalar_same_size(<2 x float>* noalias %source,
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector_scalar_same_size
 ; CHECK: bitcast <2 x float>* %source to i64*
-; CHECK: load i64*
+; CHECK: load i64, i64*
 ; CHECK: %call = call i64 @func_i64
 ; CHECK: bitcast <2 x float>* %dest to i64*
 ; CHECK: store i64
-  %tmp = load <2 x float>* %source, align 8
+  %tmp = load <2 x float>, <2 x float>* %source, align 8
   %call = call <2 x float> @alias_v2f32_to_i64(<2 x float> %tmp) nounwind
   store <2 x float> %call, <2 x float>* %dest, align 8
   ret void
@@ -137,11 +137,11 @@ define void @bitcast_alias_scalar_vector_same_size(i64* noalias %source, i64* no
 entry:
 ; CHECK-LABEL: @bitcast_alias_scalar_vector_same_size
 ; CHECK: bitcast i64* %source to <2 x float>*
-; CHECK: load <2 x float>*
+; CHECK: load <2 x float>, <2 x float>*
 ; CHECK: call <2 x float> @func_v2f32
 ; CHECK: bitcast i64* %dest to <2 x float>*
 ; CHECK: store <2 x float>
-  %tmp = load i64* %source, align 8
+  %tmp = load i64, i64* %source, align 8
   %call = call i64 @alias_i64_to_v2f32(i64 %tmp) nounwind
   store i64 %call, i64* %dest, align 8
   ret void
@@ -151,11 +151,11 @@ define void @bitcast_alias_vector_ptrs_same_size(<2 x i64*>* noalias %source, <2
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector_ptrs_same_size
 ; CHECK: bitcast <2 x i64*>* %source to <2 x i32*>*
-; CHECK: load <2 x i32*>*
+; CHECK: load <2 x i32*>, <2 x i32*>*
 ; CHECK: call <2 x i32*> @func_v2i32p
 ; CHECK: bitcast <2 x i64*>* %dest to <2 x i32*>*
 ; CHECK: store <2 x i32*>
-  %tmp = load <2 x i64*>* %source, align 8
+  %tmp = load <2 x i64*>, <2 x i64*>* %source, align 8
   %call = call <2 x i64*> @alias_v2i32p_to_v2i64p(<2 x i64*> %tmp) nounwind
   store <2 x i64*> %call, <2 x i64*>* %dest, align 8
   ret void
@@ -169,7 +169,7 @@ entry:
 ; CHECK-NOT: fptoui
 ; CHECK: @alias_i64_to_f32
 ; CHECK-NOT: uitofp
-  %tmp = load float* %source, align 8
+  %tmp = load float, float* %source, align 8
   %call = call float @alias_i64_to_f32(float %tmp) nounwind
   store float %call, float* %dest, align 8
   ret void
@@ -181,7 +181,7 @@ entry:
 ; CHECK-NOT: fptoui <2 x float> %tmp to <2 x i64>
 ; CHECK: @alias_v2i64_to_v2f32
 ; CHECK-NOT: uitofp <2 x i64> %call to <2 x float>
-  %tmp = load <2 x float>* %source, align 8
+  %tmp = load <2 x float>, <2 x float>* %source, align 8
   %call = call <2 x float> @alias_v2i64_to_v2f32(<2 x float> %tmp) nounwind
   store <2 x float> %call, <2 x float>* %dest, align 8
   ret void
@@ -191,7 +191,7 @@ define void @bitcast_alias_vector_mismatched_number_elements(<4 x float>* noalia
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector_mismatched_number_elements
 ; CHECK:  %call = call <4 x float> @alias_v2i32_to_v4f32
-  %tmp = load <4 x float>* %source, align 8
+  %tmp = load <4 x float>, <4 x float>* %source, align 8
   %call = call <4 x float> @alias_v2i32_to_v4f32(<4 x float> %tmp) nounwind
   store <4 x float> %call, <4 x float>* %dest, align 8
   ret void
@@ -201,7 +201,7 @@ define void @bitcast_alias_vector_scalar_mismatched_bit_size(<4 x float>* noalia
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector_scalar_mismatched_bit_size
 ; CHECK:  %call = call <4 x float> @alias_v4f32_to_i64
-  %tmp = load <4 x float>* %source, align 8
+  %tmp = load <4 x float>, <4 x float>* %source, align 8
   %call = call <4 x float> @alias_v4f32_to_i64(<4 x float> %tmp) nounwind
   store <4 x float> %call, <4 x float>* %dest, align 8
   ret void
@@ -211,7 +211,7 @@ define void @bitcast_alias_vector_ptrs_scalar_mismatched_bit_size(<4 x i32*>* no
 entry:
 ; CHECK-LABEL: @bitcast_alias_vector_ptrs_scalar_mismatched_bit_size
 ; CHECK: @alias_v4i32p_to_i64
-  %tmp = load <4 x i32*>* %source, align 8
+  %tmp = load <4 x i32*>, <4 x i32*>* %source, align 8
   %call = call <4 x i32*> @alias_v4i32p_to_i64(<4 x i32*> %tmp) nounwind
   store <4 x i32*> %call, <4 x i32*>* %dest, align 8
   ret void
@@ -221,7 +221,7 @@ define void @bitcast_alias_scalar_vector_ptrs_same_size(i64* noalias %source, i6
 entry:
 ; CHECK-LABEL: @bitcast_alias_scalar_vector_ptrs_same_size
 ; CHECK: @alias_i64_to_v2i32p
-  %tmp = load i64* %source, align 8
+  %tmp = load i64, i64* %source, align 8
   %call = call i64 @alias_i64_to_v2i32p(i64 %tmp) nounwind
   store i64 %call, i64* %dest, align 8
   ret void
@@ -231,7 +231,7 @@ define void @bitcast_alias_scalar_vector_mismatched_bit_size(i64* noalias %sourc
 entry:
 ; CHECK-LABEL: @bitcast_alias_scalar_vector_mismatched_bit_size
 ; CHECK: call i64 @alias_i64_to_v4f32
-  %tmp = load i64* %source, align 8
+  %tmp = load i64, i64* %source, align 8
   %call = call i64 @alias_i64_to_v4f32(i64 %tmp) nounwind
   store i64 %call, i64* %dest, align 8
   ret void
diff --git a/test/Transforms/InstCombine/bitcast.ll b/test/Transforms/InstCombine/bitcast.ll
index c7a520bcf360..579839e4245b 100644
--- a/test/Transforms/InstCombine/bitcast.ll
+++ b/test/Transforms/InstCombine/bitcast.ll
@@ -148,7 +148,7 @@ define <2 x i16> @BitcastInsert(i32 %a) {
 ; PR17293
 define <2 x i64> @test7(<2 x i8*>* %arg) nounwind {
   %cast = bitcast <2 x i8*>* %arg to <2 x i64>*
-  %load = load <2 x i64>* %cast, align 16
+  %load = load <2 x i64>, <2 x i64>* %cast, align 16
   ret <2 x i64> %load
 ; CHECK: @test7
 ; CHECK: bitcast
diff --git a/test/Transforms/InstCombine/bittest.ll b/test/Transforms/InstCombine/bittest.ll
index 84ee7dd602d5..edf65d5a87ba 100644
--- a/test/Transforms/InstCombine/bittest.ll
+++ b/test/Transforms/InstCombine/bittest.ll
@@ -5,7 +5,7 @@
 
 define void @_Z12h000007_testv(i32* %P) {
 entry:
-        %tmp.2 = load i32* @b_rec.0             ; <i32> [#uses=1]
+        %tmp.2 = load i32, i32* @b_rec.0             ; <i32> [#uses=1]
         %tmp.9 = or i32 %tmp.2, -989855744              ; <i32> [#uses=2]
         %tmp.16 = and i32 %tmp.9, -805306369            ; <i32> [#uses=2]
         %tmp.17 = and i32 %tmp.9, -973078529            ; <i32> [#uses=1]
diff --git a/test/Transforms/InstCombine/branch.ll b/test/Transforms/InstCombine/branch.ll
new file mode 100644
index 000000000000..5be93cc20234
--- /dev/null
+++ b/test/Transforms/InstCombine/branch.ll
@@ -0,0 +1,16 @@
+; RUN: opt -instcombine -S < %s | FileCheck %s
+
+define i32 @test(i32 %x) {
+; CHECK-LABEL: @test
+entry:
+; CHECK-NOT: icmp
+; CHECK: br i1 undef, 
+  %cmp = icmp ult i32 %x, 7
+  br i1 %cmp, label %merge, label %merge
+merge:
+; CHECK-LABEL: merge:
+; CHECK: ret i32 %x
+  ret i32 %x
+}
+
+
diff --git a/test/Transforms/InstCombine/call-cast-target.ll b/test/Transforms/InstCombine/call-cast-target.ll
index 4a5c94961e29..881e80762ea6 100644
--- a/test/Transforms/InstCombine/call-cast-target.ll
+++ b/test/Transforms/InstCombine/call-cast-target.ll
@@ -72,3 +72,18 @@ entry:
   %call = tail call i32 bitcast (i32 (i32)* @fn4 to i32 (i32*)*)(i32* %a)
   ret i32 %call
 }
+
+declare i1 @fn5({ i32, i32 }* byval align 4 %r)
+
+define i1 @test5() {
+; CHECK-LABEL: @test5
+; CHECK:      %[[call:.*]] = call i1 bitcast (i1 ({ i32, i32 }*)* @fn5 to i1 (i32, i32)*)(i32 {{.*}}, i32 {{.*}})
+; CHECK-NEXT: ret i1 %[[call]]
+  %1 = alloca { i32, i32 }, align 4
+  %2 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %1, i32 0, i32 0
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %1, i32 0, i32 1
+  %5 = load i32, i32* %4, align 4
+  %6 = call i1 bitcast (i1 ({ i32, i32 }*)* @fn5 to i1 (i32, i32)*)(i32 %3, i32 %5)
+  ret i1 %6
+}
diff --git a/test/Transforms/InstCombine/call-intrinsics.ll b/test/Transforms/InstCombine/call-intrinsics.ll
index f9d108058063..3e37a71e1bf6 100644
--- a/test/Transforms/InstCombine/call-intrinsics.ll
+++ b/test/Transforms/InstCombine/call-intrinsics.ll
@@ -3,17 +3,17 @@
 @X = global i8 0                ; <i8*> [#uses=3]
 @Y = global i8 12               ; <i8*> [#uses=2]
 
-declare void @llvm.memmove.i32(i8*, i8*, i32, i32)
+declare void @llvm.memmove.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
 
-declare void @llvm.memcpy.i32(i8*, i8*, i32, i32)
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
 
-declare void @llvm.memset.i32(i8*, i8, i32, i32)
+declare void @llvm.memset.p0i8.i32(i8*, i8, i32, i32, i1)
 
 define void @zero_byte_test() {
         ; These process zero bytes, so they are a noop.
-        call void @llvm.memmove.i32( i8* @X, i8* @Y, i32 0, i32 100 )
-        call void @llvm.memcpy.i32( i8* @X, i8* @Y, i32 0, i32 100 )
-        call void @llvm.memset.i32( i8* @X, i8 123, i32 0, i32 100 )
+        call void @llvm.memmove.p0i8.p0i8.i32( i8* @X, i8* @Y, i32 0, i32 128, i1 false )
+        call void @llvm.memcpy.p0i8.p0i8.i32( i8* @X, i8* @Y, i32 0, i32 128, i1 false )
+        call void @llvm.memset.p0i8.i32( i8* @X, i8 123, i32 0, i32 128, i1 false )
         ret void
 }
 
diff --git a/test/Transforms/InstCombine/call.ll b/test/Transforms/InstCombine/call.ll
index e68c0ad9b208..47ae71f37fb2 100644
--- a/test/Transforms/InstCombine/call.ll
+++ b/test/Transforms/InstCombine/call.ll
@@ -61,7 +61,7 @@ define void @test3a(i8, ...) {unreachable }
 define void @test3(i8 %A, i8 %B) {
 ; CHECK-LABEL: @test3(
 ; CHECK: %1 = zext i8 %B to i32
-; CHECK: call void (i8, ...)* @test3a(i8 %A, i32 %1)
+; CHECK: call void (i8, ...) @test3a(i8 %A, i32 %1)
 ; CHECK: ret void
   call void bitcast (void (i8, ...)* @test3a to void (i8, i8)*)( i8 %A, i8 %B)
   ret void
diff --git a/test/Transforms/InstCombine/call2.ll b/test/Transforms/InstCombine/call2.ll
index 1f45c7ab75dd..70a5b3ce36e3 100644
--- a/test/Transforms/InstCombine/call2.ll
+++ b/test/Transforms/InstCombine/call2.ll
@@ -4,11 +4,11 @@
 define i32 @bar() {
 entry:
 	%retval = alloca i32, align 4		; <i32*> [#uses=1]
-	%tmp = call i32 (...)* bitcast (i32 (i8*)* @f to i32 (...)*)( double 3.000000e+00 )		; <i32> [#uses=0]
+	%tmp = call i32 (...) bitcast (i32 (i8*)* @f to i32 (...)*)( double 3.000000e+00 )		; <i32> [#uses=0]
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load i32* %retval		; <i32> [#uses=1]
+	%retval1 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval1
 }
 
@@ -20,6 +20,6 @@ entry:
 	br label %return
 
 return:		; preds = %entry
-	%retval1 = load i32* %retval		; <i32> [#uses=1]
+	%retval1 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval1
 }
diff --git a/test/Transforms/InstCombine/cast.ll b/test/Transforms/InstCombine/cast.ll
index aac7a531b356..68f86336580b 100644
--- a/test/Transforms/InstCombine/cast.ll
+++ b/test/Transforms/InstCombine/cast.ll
@@ -93,9 +93,9 @@ declare void @varargs(i32, ...)
 
 define void @test11(i32* %P) {
         %c = bitcast i32* %P to i16*            ; <i16*> [#uses=1]
-        call void (i32, ...)* @varargs( i32 5, i16* %c )
+        call void (i32, ...) @varargs( i32 5, i16* %c )
         ret void
-; CHECK: call void (i32, ...)* @varargs(i32 5, i32* %P)
+; CHECK: call void (i32, ...) @varargs(i32 5, i32* %P)
 ; CHECK: ret void
 }
 
@@ -104,7 +104,7 @@ define void @test_invoke_vararg_cast(i32* %a, i32* %b) {
 entry:
   %0 = bitcast i32* %b to i8*
   %1 = bitcast i32* %a to i64*
-  invoke void (i32, ...)* @varargs(i32 1, i8* %0, i64* %1)
+  invoke void (i32, ...) @varargs(i32 1, i8* %0, i64* %1)
           to label %invoke.cont unwind label %lpad
 
 invoke.cont:                                      ; preds = %entry
@@ -116,13 +116,13 @@ lpad:                                             ; preds = %entry
   ret void
 ; CHECK-LABEL: test_invoke_vararg_cast
 ; CHECK-LABEL: entry:
-; CHECK: invoke void (i32, ...)* @varargs(i32 1, i32* %b, i32* %a)
+; CHECK: invoke void (i32, ...) @varargs(i32 1, i32* %b, i32* %a)
 }
 
 define i8* @test13(i64 %A) {
-        %c = getelementptr [0 x i8]* bitcast ([32832 x i8]* @inbuf to [0 x i8]*), i64 0, i64 %A             ; <i8*> [#uses=1]
+        %c = getelementptr [0 x i8], [0 x i8]* bitcast ([32832 x i8]* @inbuf to [0 x i8]*), i64 0, i64 %A             ; <i8*> [#uses=1]
         ret i8* %c
-; CHECK: %c = getelementptr [32832 x i8]* @inbuf, i64 0, i64 %A
+; CHECK: %c = getelementptr [32832 x i8], [32832 x i8]* @inbuf, i64 0, i64 %A
 ; CHECK: ret i8* %c
 }
 
@@ -187,8 +187,8 @@ define i32 @test21(i32 %X) {
         %c2 = sext i8 %c1 to i32                ; <i32> [#uses=1]
         %RV = and i32 %c2, 255          ; <i32> [#uses=1]
         ret i32 %RV
-; CHECK: %c21 = and i32 %X, 255
-; CHECK: ret i32 %c21
+; CHECK: %c2.1 = and i32 %X, 255
+; CHECK: ret i32 %c2.1
 }
 
 define i32 @test22(i32 %X) {
@@ -231,14 +231,14 @@ define i32 @test26(float %F) {
 define [4 x float]* @test27([9 x [4 x float]]* %A) {
         %c = bitcast [9 x [4 x float]]* %A to [4 x float]*              ; <[4 x float]*> [#uses=1]
         ret [4 x float]* %c
-; CHECK: %c = getelementptr inbounds [9 x [4 x float]]* %A, i64 0, i64 0
+; CHECK: %c = getelementptr inbounds [9 x [4 x float]], [9 x [4 x float]]* %A, i64 0, i64 0
 ; CHECK: ret [4 x float]* %c
 }
 
 define float* @test28([4 x float]* %A) {
         %c = bitcast [4 x float]* %A to float*          ; <float*> [#uses=1]
         ret float* %c
-; CHECK: %c = getelementptr inbounds [4 x float]* %A, i64 0, i64 0
+; CHECK: %c = getelementptr inbounds [4 x float], [4 x float]* %A, i64 0, i64 0
 ; CHECK: ret float* %c
 }
 
@@ -359,7 +359,7 @@ define i16 @test40(i16 %a) {
 ; PR1263
 define i32* @test41(i32* %tmp1) {
         %tmp64 = bitcast i32* %tmp1 to { i32 }*
-        %tmp65 = getelementptr { i32 }* %tmp64, i32 0, i32 0
+        %tmp65 = getelementptr { i32 }, { i32 }* %tmp64, i32 0, i32 0
         ret i32* %tmp65
 ; CHECK-LABEL: @test41(
 ; CHECK: ret i32* %tmp1
@@ -367,7 +367,7 @@ define i32* @test41(i32* %tmp1) {
 
 define i32 addrspace(1)* @test41_addrspacecast_smaller(i32* %tmp1) {
   %tmp64 = addrspacecast i32* %tmp1 to { i32 } addrspace(1)*
-  %tmp65 = getelementptr { i32 } addrspace(1)* %tmp64, i32 0, i32 0
+  %tmp65 = getelementptr { i32 }, { i32 } addrspace(1)* %tmp64, i32 0, i32 0
   ret i32 addrspace(1)* %tmp65
 ; CHECK-LABEL: @test41_addrspacecast_smaller(
 ; CHECK: addrspacecast i32* %tmp1 to i32 addrspace(1)*
@@ -376,7 +376,7 @@ define i32 addrspace(1)* @test41_addrspacecast_smaller(i32* %tmp1) {
 
 define i32* @test41_addrspacecast_larger(i32 addrspace(1)* %tmp1) {
   %tmp64 = addrspacecast i32 addrspace(1)* %tmp1 to { i32 }*
-  %tmp65 = getelementptr { i32 }* %tmp64, i32 0, i32 0
+  %tmp65 = getelementptr { i32 }, { i32 }* %tmp64, i32 0, i32 0
   ret i32* %tmp65
 ; CHECK-LABEL: @test41_addrspacecast_larger(
 ; CHECK: addrspacecast i32 addrspace(1)* %tmp1 to i32*
@@ -728,11 +728,11 @@ define %s @test68(%s *%p, i64 %i) {
 ; CHECK-LABEL: @test68(
   %o = mul i64 %i, 12
   %q = bitcast %s* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o
-; CHECK-NEXT: getelementptr %s*
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr %s, %s*
   %r = bitcast i8* %pp to %s*
-  %l = load %s* %r
-; CHECK-NEXT: load %s*
+  %l = load %s, %s* %r
+; CHECK-NEXT: load %s, %s*
   ret %s %l
 ; CHECK-NEXT: ret %s
 }
@@ -740,28 +740,28 @@ define %s @test68(%s *%p, i64 %i) {
 ; addrspacecasts should be eliminated.
 define %s @test68_addrspacecast(%s* %p, i64 %i) {
 ; CHECK-LABEL: @test68_addrspacecast(
-; CHECK-NEXT: getelementptr %s*
-; CHECK-NEXT: load %s*
+; CHECK-NEXT: getelementptr %s, %s*
+; CHECK-NEXT: load %s, %s*
 ; CHECK-NEXT: ret %s
   %o = mul i64 %i, 12
   %q = addrspacecast %s* %p to i8 addrspace(2)*
-  %pp = getelementptr inbounds i8 addrspace(2)* %q, i64 %o
+  %pp = getelementptr inbounds i8, i8 addrspace(2)* %q, i64 %o
   %r = addrspacecast i8 addrspace(2)* %pp to %s*
-  %l = load %s* %r
+  %l = load %s, %s* %r
   ret %s %l
 }
 
 define %s @test68_addrspacecast_2(%s* %p, i64 %i) {
 ; CHECK-LABEL: @test68_addrspacecast_2(
-; CHECK-NEXT: getelementptr %s* %p
+; CHECK-NEXT: getelementptr %s, %s* %p
 ; CHECK-NEXT: addrspacecast
-; CHECK-NEXT: load %s addrspace(1)*
+; CHECK-NEXT: load %s, %s addrspace(1)*
 ; CHECK-NEXT: ret %s
   %o = mul i64 %i, 12
   %q = addrspacecast %s* %p to i8 addrspace(2)*
-  %pp = getelementptr inbounds i8 addrspace(2)* %q, i64 %o
+  %pp = getelementptr inbounds i8, i8 addrspace(2)* %q, i64 %o
   %r = addrspacecast i8 addrspace(2)* %pp to %s addrspace(1)*
-  %l = load %s addrspace(1)* %r
+  %l = load %s, %s addrspace(1)* %r
   ret %s %l
 }
 
@@ -769,11 +769,11 @@ define %s @test68_as1(%s addrspace(1)* %p, i32 %i) {
 ; CHECK-LABEL: @test68_as1(
   %o = mul i32 %i, 12
   %q = bitcast %s addrspace(1)* %p to i8 addrspace(1)*
-  %pp = getelementptr inbounds i8 addrspace(1)* %q, i32 %o
-; CHECK-NEXT: getelementptr %s addrspace(1)*
+  %pp = getelementptr inbounds i8, i8 addrspace(1)* %q, i32 %o
+; CHECK-NEXT: getelementptr %s, %s addrspace(1)*
   %r = bitcast i8 addrspace(1)* %pp to %s addrspace(1)*
-  %l = load %s addrspace(1)* %r
-; CHECK-NEXT: load %s addrspace(1)*
+  %l = load %s, %s addrspace(1)* %r
+; CHECK-NEXT: load %s, %s addrspace(1)*
   ret %s %l
 ; CHECK-NEXT: ret %s
 }
@@ -782,11 +782,11 @@ define double @test69(double *%p, i64 %i) {
 ; CHECK-LABEL: @test69(
   %o = shl nsw i64 %i, 3
   %q = bitcast double* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o
-; CHECK-NEXT: getelementptr inbounds double*
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr inbounds double, double*
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -796,11 +796,11 @@ define %s @test70(%s *%p, i64 %i) {
   %o = mul nsw i64 %i, 36
 ; CHECK-NEXT: mul nsw i64 %i, 3
   %q = bitcast %s* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o
-; CHECK-NEXT: getelementptr inbounds %s*
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr inbounds %s, %s*
   %r = bitcast i8* %pp to %s*
-  %l = load %s* %r
-; CHECK-NEXT: load %s*
+  %l = load %s, %s* %r
+; CHECK-NEXT: load %s, %s*
   ret %s %l
 ; CHECK-NEXT: ret %s
 }
@@ -810,11 +810,11 @@ define double @test71(double *%p, i64 %i) {
   %o = shl i64 %i, 5
 ; CHECK-NEXT: shl i64 %i, 2
   %q = bitcast double* %p to i8*
-  %pp = getelementptr i8* %q, i64 %o
-; CHECK-NEXT: getelementptr double*
+  %pp = getelementptr i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr double, double*
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -825,11 +825,11 @@ define double @test72(double *%p, i32 %i) {
   %o = sext i32 %so to i64
 ; CHECK-NEXT: sext i32 %i to i64
   %q = bitcast double* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o
-; CHECK-NEXT: getelementptr inbounds double*
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr inbounds double, double*
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -840,11 +840,11 @@ define double @test73(double *%p, i128 %i) {
   %o = trunc i128 %lo to i64
 ; CHECK-NEXT: trunc i128 %i to i64
   %q = bitcast double* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o
-; CHECK-NEXT: getelementptr double*
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o
+; CHECK-NEXT: getelementptr double, double*
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -852,11 +852,11 @@ define double @test73(double *%p, i128 %i) {
 define double @test74(double *%p, i64 %i) {
 ; CHECK-LABEL: @test74(
   %q = bitcast double* %p to i64*
-  %pp = getelementptr inbounds i64* %q, i64 %i
-; CHECK-NEXT: getelementptr inbounds double*
+  %pp = getelementptr inbounds i64, i64* %q, i64 %i
+; CHECK-NEXT: getelementptr inbounds double, double*
   %r = bitcast i64* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -868,7 +868,7 @@ define i32* @test75(i32* %p, i32 %x) {
   %z = sext i32 %y to i64
 ; CHECK-NEXT: sext i32 %y to i64
   %q = bitcast i32* %p to i8*
-  %r = getelementptr i8* %q, i64 %z
+  %r = getelementptr i8, i8* %q, i64 %z
   %s = bitcast i8* %r to i32*
   ret i32* %s
 }
@@ -879,11 +879,11 @@ define %s @test76(%s *%p, i64 %i, i64 %j) {
   %o2 = mul nsw i64 %o, %j
 ; CHECK-NEXT: %o2 = mul i64 %i, %j
   %q = bitcast %s* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o2
-; CHECK-NEXT: getelementptr %s* %p, i64 %o2
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o2
+; CHECK-NEXT: getelementptr %s, %s* %p, i64 %o2
   %r = bitcast i8* %pp to %s*
-  %l = load %s* %r
-; CHECK-NEXT: load %s*
+  %l = load %s, %s* %r
+; CHECK-NEXT: load %s, %s*
   ret %s %l
 ; CHECK-NEXT: ret %s
 }
@@ -895,11 +895,11 @@ define %s @test77(%s *%p, i64 %i, i64 %j) {
 ; CHECK-NEXT: %o = mul nsw i64 %i, 3
 ; CHECK-NEXT: %o2 = mul nsw i64 %o, %j
   %q = bitcast %s* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %o2
-; CHECK-NEXT: getelementptr inbounds %s* %p, i64 %o2
+  %pp = getelementptr inbounds i8, i8* %q, i64 %o2
+; CHECK-NEXT: getelementptr inbounds %s, %s* %p, i64 %o2
   %r = bitcast i8* %pp to %s*
-  %l = load %s* %r
-; CHECK-NEXT: load %s*
+  %l = load %s, %s* %r
+; CHECK-NEXT: load %s, %s*
   ret %s %l
 ; CHECK-NEXT: ret %s
 }
@@ -923,11 +923,11 @@ define %s @test78(%s *%p, i64 %i, i64 %j, i32 %k, i32 %l, i128 %m, i128 %n) {
   %h = mul nsw i64 %g, %j
 ; CHECK-NEXT: mul i64 %g, %j
   %q = bitcast %s* %p to i8*
-  %pp = getelementptr inbounds i8* %q, i64 %h
-; CHECK-NEXT: getelementptr %s* %p, i64 %h
+  %pp = getelementptr inbounds i8, i8* %q, i64 %h
+; CHECK-NEXT: getelementptr %s, %s* %p, i64 %h
   %r = bitcast i8* %pp to %s*
-  %load = load %s* %r
-; CHECK-NEXT: load %s*
+  %load = load %s, %s* %r
+; CHECK-NEXT: load %s, %s*
   ret %s %load
 ; CHECK-NEXT: ret %s
 }
@@ -940,9 +940,9 @@ define %s @test79(%s *%p, i64 %i, i32 %j) {
   %c = mul i32 %b, %j
   %q = bitcast %s* %p to i8*
 ; CHECK: bitcast
-  %pp = getelementptr inbounds i8* %q, i32 %c
+  %pp = getelementptr inbounds i8, i8* %q, i32 %c
   %r = bitcast i8* %pp to %s*
-  %l = load %s* %r
+  %l = load %s, %s* %r
   ret %s %l
 }
 
@@ -951,39 +951,39 @@ define double @test80([100 x double]* %p, i32 %i) {
   %tmp = shl nsw i32 %i, 3
 ; CHECK-NEXT: sext i32 %i to i64
   %q = bitcast [100 x double]* %p to i8*
-  %pp = getelementptr i8* %q, i32 %tmp
-; CHECK-NEXT: getelementptr [100 x double]*
+  %pp = getelementptr i8, i8* %q, i32 %tmp
+; CHECK-NEXT: getelementptr [100 x double], [100 x double]*
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
-; CHECK-NEXT: load double*
+  %l = load double, double* %r
+; CHECK-NEXT: load double, double*
   ret double %l
 ; CHECK-NEXT: ret double
 }
 
 define double @test80_addrspacecast([100 x double] addrspace(1)* %p, i32 %i) {
 ; CHECK-LABEL: @test80_addrspacecast(
-; CHECK-NEXT: getelementptr [100 x double] addrspace(1)* %p
-; CHECK-NEXT: load double addrspace(1)*
+; CHECK-NEXT: getelementptr [100 x double], [100 x double] addrspace(1)* %p
+; CHECK-NEXT: load double, double addrspace(1)*
 ; CHECK-NEXT: ret double
   %tmp = shl nsw i32 %i, 3
   %q = addrspacecast [100 x double] addrspace(1)* %p to i8 addrspace(2)*
-  %pp = getelementptr i8 addrspace(2)* %q, i32 %tmp
+  %pp = getelementptr i8, i8 addrspace(2)* %q, i32 %tmp
   %r = addrspacecast i8 addrspace(2)* %pp to double addrspace(1)*
-  %l = load double addrspace(1)* %r
+  %l = load double, double addrspace(1)* %r
   ret double %l
 }
 
 define double @test80_addrspacecast_2([100 x double] addrspace(1)* %p, i32 %i) {
 ; CHECK-LABEL: @test80_addrspacecast_2(
-; CHECK-NEXT: getelementptr [100 x double] addrspace(1)*
+; CHECK-NEXT: getelementptr [100 x double], [100 x double] addrspace(1)*
 ; CHECK-NEXT: addrspacecast double addrspace(1)*
-; CHECK-NEXT: load double addrspace(3)*
+; CHECK-NEXT: load double, double addrspace(3)*
 ; CHECK-NEXT: ret double
   %tmp = shl nsw i32 %i, 3
   %q = addrspacecast [100 x double] addrspace(1)* %p to i8 addrspace(2)*
-  %pp = getelementptr i8 addrspace(2)* %q, i32 %tmp
+  %pp = getelementptr i8, i8 addrspace(2)* %q, i32 %tmp
   %r = addrspacecast i8 addrspace(2)* %pp to double addrspace(3)*
-  %l = load double addrspace(3)* %r
+  %l = load double, double addrspace(3)* %r
   ret double %l
 }
 
@@ -992,11 +992,11 @@ define double @test80_as1([100 x double] addrspace(1)* %p, i16 %i) {
   %tmp = shl nsw i16 %i, 3
 ; CHECK-NEXT: sext i16 %i to i32
   %q = bitcast [100 x double] addrspace(1)* %p to i8 addrspace(1)*
-  %pp = getelementptr i8 addrspace(1)* %q, i16 %tmp
-; CHECK-NEXT: getelementptr [100 x double] addrspace(1)*
+  %pp = getelementptr i8, i8 addrspace(1)* %q, i16 %tmp
+; CHECK-NEXT: getelementptr [100 x double], [100 x double] addrspace(1)*
   %r = bitcast i8 addrspace(1)* %pp to double addrspace(1)*
-  %l = load double addrspace(1)* %r
-; CHECK-NEXT: load double addrspace(1)*
+  %l = load double, double addrspace(1)* %r
+; CHECK-NEXT: load double, double addrspace(1)*
   ret double %l
 ; CHECK-NEXT: ret double
 }
@@ -1004,9 +1004,9 @@ define double @test80_as1([100 x double] addrspace(1)* %p, i16 %i) {
 define double @test81(double *%p, float %f) {
   %i = fptosi float %f to i64
   %q = bitcast double* %p to i8*
-  %pp = getelementptr i8* %q, i64 %i
+  %pp = getelementptr i8, i8* %q, i64 %i
   %r = bitcast i8* %pp to double*
-  %l = load double* %r
+  %l = load double, double* %r
   ret double %l
 }
 
@@ -1104,3 +1104,36 @@ define i32 @PR21388(i32* %v) {
 ; CHECK-NEXT: %[[sext:.*]] = sext i1 %[[icmp]] to i32
 ; CHECK-NEXT: ret i32 %[[sext]]
 }
+
+define float @sitofp_zext(i16 %a) {
+; CHECK-LABEL: @sitofp_zext(
+; CHECK-NEXT: %[[sitofp:.*]] = uitofp i16 %a to float
+; CHECK-NEXT: ret float %[[sitofp]]
+  %zext = zext i16 %a to i32
+  %sitofp = sitofp i32 %zext to float
+  ret float %sitofp
+}
+
+define i1 @PR23309(i32 %A, i32 %B) {
+; CHECK-LABEL: @PR23309(
+; CHECK-NEXT: %[[sub:.*]] = sub i32 %A, %B
+; CHECK-NEXT: %[[and:.*]] = and i32 %[[sub]], 1
+; CHECK-NEXT: %[[cmp:.*]] = icmp ne i32 %[[and]], 0
+; CHECK-NEXT: ret i1 %[[cmp]]
+  %add = add i32 %A, -4
+  %sub = sub nsw i32 %add, %B
+  %trunc = trunc i32 %sub to i1
+  ret i1 %trunc
+}
+
+define i1 @PR23309v2(i32 %A, i32 %B) {
+; CHECK-LABEL: @PR23309v2(
+; CHECK-NEXT: %[[sub:.*]] = add i32 %A, %B
+; CHECK-NEXT: %[[and:.*]] = and i32 %[[sub]], 1
+; CHECK-NEXT: %[[cmp:.*]] = icmp ne i32 %[[and]], 0
+; CHECK-NEXT: ret i1 %[[cmp]]
+  %add = add i32 %A, -4
+  %sub = add nuw i32 %add, %B
+  %trunc = trunc i32 %sub to i1
+  ret i1 %trunc
+}
diff --git a/test/Transforms/InstCombine/cast_ptr.ll b/test/Transforms/InstCombine/cast_ptr.ll
index 23006a84604b..eaf946ef9259 100644
--- a/test/Transforms/InstCombine/cast_ptr.ll
+++ b/test/Transforms/InstCombine/cast_ptr.ll
@@ -3,6 +3,8 @@
 
 target datalayout = "p:32:32-p1:32:32-p2:16:16"
 
+@global = global i8 0
+
 ; This shouldn't convert to getelementptr because the relationship
 ; between the arithmetic and the layout of allocated memory is
 ; entirely unknown.
@@ -47,10 +49,29 @@ define i1 @test2_as2_larger(i8 addrspace(2)* %a, i8 addrspace(2)* %b) {
   ret i1 %r
 }
 
+; These casts should not be folded away.
+; CHECK-LABEL: @test2_diff_as
+; CHECK: icmp sge i32 %i0, %i1
+define i1 @test2_diff_as(i8* %p, i8 addrspace(1)* %q) {
+  %i0 = ptrtoint i8* %p to i32
+  %i1 = ptrtoint i8 addrspace(1)* %q to i32
+  %r0 = icmp sge i32 %i0, %i1
+  ret i1 %r0
+}
+
+; These casts should not be folded away.
+; CHECK-LABEL: @test2_diff_as_global
+; CHECK: icmp sge i32 %i1
+define i1 @test2_diff_as_global(i8 addrspace(1)* %q) {
+  %i0 = ptrtoint i8* @global to i32
+  %i1 = ptrtoint i8 addrspace(1)* %q to i32
+  %r0 = icmp sge i32 %i1, %i0
+  ret i1 %r0
+}
+
 ; These casts should also be folded away.
 ; CHECK-LABEL: @test3(
 ; CHECK: icmp eq i8* %a, @global
-@global = global i8 0
 define i1 @test3(i8* %a) {
         %tmpa = ptrtoint i8* %a to i32
         %r = icmp eq i32 %tmpa, ptrtoint (i8* @global to i32)
@@ -86,7 +107,7 @@ define i1 @test4_as2(i16 %A) {
 declare %op* @foo(%op* %X)
 
 define %unop* @test5(%op* %O) {
-        %tmp = load %unop* (%op*)** bitcast ([1 x %op* (%op*)*]* @Array to %unop* (%op*)**); <%unop* (%op*)*> [#uses=1]
+        %tmp = load %unop* (%op*)*, %unop* (%op*)** bitcast ([1 x %op* (%op*)*]* @Array to %unop* (%op*)**); <%unop* (%op*)*> [#uses=1]
         %tmp.2 = call %unop* %tmp( %op* %O )            ; <%unop*> [#uses=1]
         ret %unop* %tmp.2
 ; CHECK-LABEL: @test5(
@@ -101,8 +122,8 @@ define %unop* @test5(%op* %O) {
 define i8 @test6(i8 addrspace(1)* %source) {
 entry:
   %arrayidx223 = addrspacecast i8 addrspace(1)* %source to i8*
-  %tmp4 = load i8* %arrayidx223
+  %tmp4 = load i8, i8* %arrayidx223
   ret i8 %tmp4
 ; CHECK-LABEL: @test6(
-; CHECK: load i8* %arrayidx223
+; CHECK: load i8, i8* %arrayidx223
 }
diff --git a/test/Transforms/InstCombine/constant-expr-datalayout.ll b/test/Transforms/InstCombine/constant-expr-datalayout.ll
index 9a72c77afdb0..cdecfc91f592 100644
--- a/test/Transforms/InstCombine/constant-expr-datalayout.ll
+++ b/test/Transforms/InstCombine/constant-expr-datalayout.ll
@@ -6,7 +6,7 @@ target triple = "x86_64-unknown-linux-gnu"
 %test1.struct = type { i32, i32 }
 @test1.aligned_glbl = global %test1.struct zeroinitializer, align 4
 define void @test1(i64 *%ptr) {
-  store i64 and (i64 ptrtoint (i32* getelementptr (%test1.struct* @test1.aligned_glbl, i32 0, i32 1) to i64), i64 3), i64* %ptr
+  store i64 and (i64 ptrtoint (i32* getelementptr (%test1.struct, %test1.struct* @test1.aligned_glbl, i32 0, i32 1) to i64), i64 3), i64* %ptr
 ; CHECK: store i64 0, i64* %ptr
   ret void
 }
diff --git a/test/Transforms/InstCombine/constant-fold-address-space-pointer.ll b/test/Transforms/InstCombine/constant-fold-address-space-pointer.ll
index bb61f02c8e23..0ab76be17cb3 100644
--- a/test/Transforms/InstCombine/constant-fold-address-space-pointer.ll
+++ b/test/Transforms/InstCombine/constant-fold-address-space-pointer.ll
@@ -149,22 +149,22 @@ define i1 @constant_fold_inttoptr() {
 
 define float @constant_fold_bitcast_ftoi_load() {
 ; CHECK-LABEL: @constant_fold_bitcast_ftoi_load(
-; CHECK: load float addrspace(3)* bitcast (i32 addrspace(3)* @g to float addrspace(3)*), align 4
-  %a = load float addrspace(3)* bitcast (i32 addrspace(3)* @g to float addrspace(3)*), align 4
+; CHECK: load float, float addrspace(3)* bitcast (i32 addrspace(3)* @g to float addrspace(3)*), align 4
+  %a = load float, float addrspace(3)* bitcast (i32 addrspace(3)* @g to float addrspace(3)*), align 4
   ret float %a
 }
 
 define i32 @constant_fold_bitcast_itof_load() {
 ; CHECK-LABEL: @constant_fold_bitcast_itof_load(
-; CHECK: load i32 addrspace(3)* bitcast (float addrspace(3)* @g_float_as3 to i32 addrspace(3)*), align 4
-  %a = load i32 addrspace(3)* bitcast (float addrspace(3)* @g_float_as3 to i32 addrspace(3)*), align 4
+; CHECK: load i32, i32 addrspace(3)* bitcast (float addrspace(3)* @g_float_as3 to i32 addrspace(3)*), align 4
+  %a = load i32, i32 addrspace(3)* bitcast (float addrspace(3)* @g_float_as3 to i32 addrspace(3)*), align 4
   ret i32 %a
 }
 
 define <4 x float> @constant_fold_bitcast_vector_as() {
 ; CHECK-LABEL: @constant_fold_bitcast_vector_as(
-; CHECK: load <4 x float> addrspace(3)* @g_v4f_as3, align 16
-  %a = load <4 x float> addrspace(3)* bitcast (<4 x i32> addrspace(3)* bitcast (<4 x float> addrspace(3)* @g_v4f_as3 to <4 x i32> addrspace(3)*) to <4 x float> addrspace(3)*), align 4
+; CHECK: load <4 x float>, <4 x float> addrspace(3)* @g_v4f_as3, align 16
+  %a = load <4 x float>, <4 x float> addrspace(3)* bitcast (<4 x i32> addrspace(3)* bitcast (<4 x float> addrspace(3)* @g_v4f_as3 to <4 x i32> addrspace(3)*) to <4 x float> addrspace(3)*), align 4
   ret <4 x float> %a
 }
 
@@ -172,9 +172,9 @@ define <4 x float> @constant_fold_bitcast_vector_as() {
 
 define i32 @test_cast_gep_small_indices_as() {
 ; CHECK-LABEL: @test_cast_gep_small_indices_as(
-; CHECK: load i32 addrspace(3)* getelementptr inbounds ([10 x i32] addrspace(3)* @i32_array_as3, i16 0, i16 0), align 16
-   %p = getelementptr [10 x i32] addrspace(3)* @i32_array_as3, i7 0, i7 0
-   %x = load i32 addrspace(3)* %p, align 4
+; CHECK: load i32, i32 addrspace(3)* getelementptr inbounds ([10 x i32], [10 x i32] addrspace(3)* @i32_array_as3, i16 0, i16 0), align 16
+   %p = getelementptr [10 x i32], [10 x i32] addrspace(3)* @i32_array_as3, i7 0, i7 0
+   %x = load i32, i32 addrspace(3)* %p, align 4
    ret i32 %x
 }
 
@@ -184,22 +184,22 @@ define i32 @test_cast_gep_small_indices_as() {
   float 0.0,
   float 0.0,
   [4 x i32] zeroinitializer,
-  i32 addrspace(3)* getelementptr ([10 x i32] addrspace(3)* @i32_array_as3, i64 0, i64 0)
+  i32 addrspace(3)* getelementptr ([10 x i32], [10 x i32] addrspace(3)* @i32_array_as3, i64 0, i64 0)
 }
 
 define i32 @test_cast_gep_large_indices_as() {
 ; CHECK-LABEL: @test_cast_gep_large_indices_as(
-; CHECK: load i32 addrspace(3)* getelementptr inbounds ([10 x i32] addrspace(3)* @i32_array_as3, i16 0, i16 0), align 16
-   %p = getelementptr [10 x i32] addrspace(3)* @i32_array_as3, i64 0, i64 0
-   %x = load i32 addrspace(3)* %p, align 4
+; CHECK: load i32, i32 addrspace(3)* getelementptr inbounds ([10 x i32], [10 x i32] addrspace(3)* @i32_array_as3, i16 0, i16 0), align 16
+   %p = getelementptr [10 x i32], [10 x i32] addrspace(3)* @i32_array_as3, i64 0, i64 0
+   %x = load i32, i32 addrspace(3)* %p, align 4
    ret i32 %x
 }
 
 define i32 @test_constant_cast_gep_struct_indices_as() {
 ; CHECK-LABEL: @test_constant_cast_gep_struct_indices_as(
-; CHECK: load i32 addrspace(3)* getelementptr inbounds (%struct.foo addrspace(3)* @constant_fold_global_ptr, i16 0, i32 2, i16 2), align 8
-  %x = getelementptr %struct.foo addrspace(3)* @constant_fold_global_ptr, i18 0, i32 2, i12 2
-  %y = load i32 addrspace(3)* %x, align 4
+; CHECK: load i32, i32 addrspace(3)* getelementptr inbounds (%struct.foo, %struct.foo addrspace(3)* @constant_fold_global_ptr, i16 0, i32 2, i16 2), align 8
+  %x = getelementptr %struct.foo, %struct.foo addrspace(3)* @constant_fold_global_ptr, i18 0, i32 2, i12 2
+  %y = load i32, i32 addrspace(3)* %x, align 4
   ret i32 %y
 }
 
@@ -208,8 +208,8 @@ define i32 @test_constant_cast_gep_struct_indices_as() {
 define i32 @test_read_data_from_global_as3() {
 ; CHECK-LABEL: @test_read_data_from_global_as3(
 ; CHECK-NEXT: ret i32 2
-  %x = getelementptr [5 x i32] addrspace(3)* @constant_data_as3, i32 0, i32 1
-  %y = load i32 addrspace(3)* %x, align 4
+  %x = getelementptr [5 x i32], [5 x i32] addrspace(3)* @constant_data_as3, i32 0, i32 1
+  %y = load i32, i32 addrspace(3)* %x, align 4
   ret i32 %y
 }
 
@@ -219,14 +219,14 @@ define i32 @test_read_data_from_global_as3() {
 @d = addrspace(1) constant i32 99
 
 @ptr_array = addrspace(2) constant [4 x i32 addrspace(1)*] [ i32 addrspace(1)* @a, i32 addrspace(1)* @b, i32 addrspace(1)* @c, i32 addrspace(1)* @d]
-@indirect = addrspace(0) constant i32 addrspace(1)* addrspace(2)* getelementptr inbounds ([4 x i32 addrspace(1)*] addrspace(2)* @ptr_array, i1 0, i32 2)
+@indirect = addrspace(0) constant i32 addrspace(1)* addrspace(2)* getelementptr inbounds ([4 x i32 addrspace(1)*], [4 x i32 addrspace(1)*] addrspace(2)* @ptr_array, i1 0, i32 2)
 
 define i32 @constant_through_array_as_ptrs() {
 ; CHECK-LABEL: @constant_through_array_as_ptrs(
 ; CHECK-NEXT: ret i32 34
-  %p = load i32 addrspace(1)* addrspace(2)* addrspace(0)* @indirect, align 4
-  %a = load i32 addrspace(1)* addrspace(2)* %p, align 4
-  %b = load i32 addrspace(1)* %a, align 4
+  %p = load i32 addrspace(1)* addrspace(2)*, i32 addrspace(1)* addrspace(2)* addrspace(0)* @indirect, align 4
+  %a = load i32 addrspace(1)*, i32 addrspace(1)* addrspace(2)* %p, align 4
+  %b = load i32, i32 addrspace(1)* %a, align 4
   ret i32 %b
 }
 
@@ -234,8 +234,8 @@ define i32 @constant_through_array_as_ptrs() {
 
 define float @canonicalize_addrspacecast(i32 %i) {
 ; CHECK-LABEL: @canonicalize_addrspacecast
-; CHECK-NEXT: getelementptr inbounds float* addrspacecast (float addrspace(3)* bitcast ([0 x i8] addrspace(3)* @shared_mem to float addrspace(3)*) to float*), i32 %i
-  %p = getelementptr inbounds float* addrspacecast ([0 x i8] addrspace(3)* @shared_mem to float*), i32 %i
-  %v = load float* %p
+; CHECK-NEXT: getelementptr inbounds float, float* addrspacecast (float addrspace(3)* bitcast ([0 x i8] addrspace(3)* @shared_mem to float addrspace(3)*) to float*), i32 %i
+  %p = getelementptr inbounds float, float* addrspacecast ([0 x i8] addrspace(3)* @shared_mem to float*), i32 %i
+  %v = load float, float* %p
   ret float %v
 }
diff --git a/test/Transforms/InstCombine/constant-fold-alias.ll b/test/Transforms/InstCombine/constant-fold-alias.ll
index 13da0f445242..c872f57c37e1 100644
--- a/test/Transforms/InstCombine/constant-fold-alias.ll
+++ b/test/Transforms/InstCombine/constant-fold-alias.ll
@@ -6,8 +6,8 @@ target datalayout = "e-p1:16:16-p2:32:32-p3:64:64"
 @G2 = global i32 42
 @G3 = global [4 x i8] zeroinitializer, align 1
 
-@A1 = alias bitcast (i8* getelementptr inbounds ([4 x i8]* @G3, i32 0, i32 2) to i32*)
-@A2 = alias inttoptr (i64 and (i64 ptrtoint (i8* getelementptr inbounds ([4 x i8]* @G3, i32 0, i32 3) to i64), i64 -4) to i32*)
+@A1 = alias bitcast (i8* getelementptr inbounds ([4 x i8], [4 x i8]* @G3, i32 0, i32 2) to i32*)
+@A2 = alias inttoptr (i64 and (i64 ptrtoint (i8* getelementptr inbounds ([4 x i8], [4 x i8]* @G3, i32 0, i32 3) to i64), i64 -4) to i32*)
 
 define i64 @f1() {
 ; This cannot be constant folded because G1 is underaligned.
diff --git a/test/Transforms/InstCombine/constant-fold-gep.ll b/test/Transforms/InstCombine/constant-fold-gep.ll
index 5fb56023a4f6..77090529e130 100644
--- a/test/Transforms/InstCombine/constant-fold-gep.ll
+++ b/test/Transforms/InstCombine/constant-fold-gep.ll
@@ -9,48 +9,48 @@ target datalayout = "E-p:64:64:64-p1:16:16:16-i1:8:8-i8:8:8-i16:16:16-i32:32:32-
 @Y = internal global [3 x %struct.X] zeroinitializer
 
 define void @frob() {
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 16
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 2), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 2), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 0), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 3), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 1), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 4), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 2), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 5), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 0), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 6), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 1), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 7), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 2), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 8), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 0), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 9), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 1), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 10), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 2), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 11), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 0), align 16
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 12), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 1), align 4
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 13), align 4
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 2), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 14), align 8
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 0), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 15), align 8
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 1), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 16), align 8
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 2), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 17), align 8
-; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 18), align 8
-; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 2, i64 0, i32 0, i64 0), align 16
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 36), align 8
-; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 1), align 8
-  store i32 1, i32* getelementptr ([3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 19), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 16
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 0), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 1), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 2), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 2), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 0), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 3), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 1), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 4), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 1, i64 2), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 5), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 0), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 6), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 1), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 7), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 0, i64 2), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 8), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 0), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 9), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 1), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 10), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 1, i32 1, i64 2), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 11), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 0), align 16
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 12), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 1), align 4
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 13), align 4
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 0, i64 2), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 14), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 0), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 15), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 1), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 16), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 2, i32 1, i64 2), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 17), align 8
+; CHECK: store i32 1, i32* getelementptr inbounds ([3 x %struct.X], [3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 0), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 18), align 8
+; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 2, i64 0, i32 0, i64 0), align 16
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 36), align 8
+; CHECK: store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 1, i64 0, i32 0, i64 1), align 8
+  store i32 1, i32* getelementptr ([3 x %struct.X], [3 x %struct.X]* @Y, i64 0, i64 0, i32 0, i64 19), align 8
   ret void
 }
 
@@ -61,12 +61,12 @@ define void @frob() {
 
 define i64 @test2() {
 entry:
-  %A = bitcast i8* getelementptr inbounds ([1000 x i8]* @X, i64 1, i64 0) to i8*
-  %B = bitcast i8* getelementptr inbounds ([1000 x i8]* @X, i64 0, i64 0) to i8*
+  %A = bitcast i8* getelementptr inbounds ([1000 x i8], [1000 x i8]* @X, i64 1, i64 0) to i8*
+  %B = bitcast i8* getelementptr inbounds ([1000 x i8], [1000 x i8]* @X, i64 0, i64 0) to i8*
 
   %B2 = ptrtoint i8* %B to i64
   %C = sub i64 0, %B2
-  %D = getelementptr i8* %A, i64 %C
+  %D = getelementptr i8, i8* %A, i64 %C
   %E = ptrtoint i8* %D to i64
 
   ret i64 %E
@@ -80,12 +80,12 @@ define i16 @test2_as1() {
   ; CHECK: ret i16 1000
 
 entry:
-  %A = bitcast i8 addrspace(1)* getelementptr inbounds ([1000 x i8] addrspace(1)* @X_as1, i64 1, i64 0) to i8 addrspace(1)*
-  %B = bitcast i8 addrspace(1)* getelementptr inbounds ([1000 x i8] addrspace(1)* @X_as1, i64 0, i64 0) to i8 addrspace(1)*
+  %A = bitcast i8 addrspace(1)* getelementptr inbounds ([1000 x i8], [1000 x i8] addrspace(1)* @X_as1, i64 1, i64 0) to i8 addrspace(1)*
+  %B = bitcast i8 addrspace(1)* getelementptr inbounds ([1000 x i8], [1000 x i8] addrspace(1)* @X_as1, i64 0, i64 0) to i8 addrspace(1)*
 
   %B2 = ptrtoint i8 addrspace(1)* %B to i16
   %C = sub i16 0, %B2
-  %D = getelementptr i8 addrspace(1)* %A, i16 %C
+  %D = getelementptr i8, i8 addrspace(1)* %A, i16 %C
   %E = ptrtoint i8 addrspace(1)* %D to i16
 
   ret i16 %E
diff --git a/test/Transforms/InstCombine/constant-fold-hang.ll b/test/Transforms/InstCombine/constant-fold-hang.ll
new file mode 100644
index 000000000000..2ca6b86ccc2f
--- /dev/null
+++ b/test/Transforms/InstCombine/constant-fold-hang.ll
@@ -0,0 +1,14 @@
+; RUN: opt -instcombine < %s
+
+; Function Attrs: nounwind readnone ssp
+define void @mulByZero(<4 x i16> %x) #0 {
+entry:
+  %a = tail call <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16> %x, <4 x i16> zeroinitializer) #2
+  ret void
+}
+
+; Function Attrs: nounwind readnone
+declare <4 x i32> @llvm.arm.neon.vmulls.v4i32(<4 x i16>, <4 x i16>) #1
+
+attributes #0 = { nounwind readnone ssp }
+attributes #1 = { nounwind readnone }
diff --git a/test/Transforms/InstCombine/crash.ll b/test/Transforms/InstCombine/crash.ll
index 2ef6ac64c7f3..2d93ecd59bf5 100644
--- a/test/Transforms/InstCombine/crash.ll
+++ b/test/Transforms/InstCombine/crash.ll
@@ -35,12 +35,12 @@ entry:
 ; PR4908
 define void @test2(<1 x i16>* nocapture %b, i32* nocapture %c) nounwind ssp {
 entry:
-  %arrayidx = getelementptr inbounds <1 x i16>* %b, i64 undef ; <<1 x i16>*>
-  %tmp2 = load <1 x i16>* %arrayidx               ; <<1 x i16>> [#uses=1]
+  %arrayidx = getelementptr inbounds <1 x i16>, <1 x i16>* %b, i64 undef ; <<1 x i16>*>
+  %tmp2 = load <1 x i16>, <1 x i16>* %arrayidx               ; <<1 x i16>> [#uses=1]
   %tmp6 = bitcast <1 x i16> %tmp2 to i16          ; <i16> [#uses=1]
   %tmp7 = zext i16 %tmp6 to i32                   ; <i32> [#uses=1]
   %ins = or i32 0, %tmp7                          ; <i32> [#uses=1]
-  %arrayidx20 = getelementptr inbounds i32* %c, i64 undef ; <i32*> [#uses=1]
+  %arrayidx20 = getelementptr inbounds i32, i32* %c, i64 undef ; <i32*> [#uses=1]
   store i32 %ins, i32* %arrayidx20
   ret void
 }
@@ -60,7 +60,7 @@ define void @foo(i1) nounwind align 2 {
 ; <label>:3                                       ; preds = %2, %1
   %4 = phi i8 [ 1, %2 ], [ 0, %1 ]                ; <i8> [#uses=1]
   %5 = icmp eq i8 %4, 0                           ; <i1> [#uses=1]
-  %6 = load i64* @tmp2, align 8                   ; <i64> [#uses=1]
+  %6 = load i64, i64* @tmp2, align 8                   ; <i64> [#uses=1]
   %7 = select i1 %5, i64 0, i64 %6                ; <i64> [#uses=1]
   br label %8
 
@@ -79,9 +79,9 @@ define void @bar3(i1, i1) nounwind align 2 {
   br i1 %1, label %10, label %3
 
 ; <label>:3                                       ; preds = %2
-  %4 = getelementptr inbounds %t0* null, i64 0, i32 1 ; <i32*> [#uses=0]
-  %5 = getelementptr inbounds %t1* null, i64 0, i32 4 ; <i32**> [#uses=1]
-  %6 = load i32** %5, align 8                     ; <i32*> [#uses=1]
+  %4 = getelementptr inbounds %t0, %t0* null, i64 0, i32 1 ; <i32*> [#uses=0]
+  %5 = getelementptr inbounds %t1, %t1* null, i64 0, i32 4 ; <i32**> [#uses=1]
+  %6 = load i32*, i32** %5, align 8                     ; <i32*> [#uses=1]
   %7 = icmp ne i32* %6, null                      ; <i1> [#uses=1]
   %8 = zext i1 %7 to i32                          ; <i32> [#uses=1]
   %9 = add i32 %8, 0                              ; <i32> [#uses=1]
@@ -115,7 +115,7 @@ BB1:
 
 BB2:
   %v5_ = phi i1 [ true, %BB0], [false, %BB1]
-  %v6 = load i64* %P
+  %v6 = load i64, i64* %P
   br label %l8
 
 l8:
@@ -149,7 +149,7 @@ exit:
 
 define arm_aapcs_vfpcc i32 @test6(i32 %argc, i8** %argv) nounwind {
 entry:
-  store i32* getelementptr (i32* bitcast (i32 (i32, i8**)* @test6 to i32*), i32 -2048), i32** @test6g, align 4
+  store i32* getelementptr (i32, i32* bitcast (i32 (i32, i8**)* @test6 to i32*), i32 -2048), i32** @test6g, align 4
   unreachable
 }
 
@@ -182,8 +182,8 @@ cont:                                             ; preds = %ehcleanup
   resume { i8*, i32 } %exc1
 
 cond.false:                                       ; preds = %entry
-  %tmp4 = getelementptr inbounds %class.RuleBasedBreakIterator* %this, i32 0, i32 0 ; <i64 ()**> [#uses=1]
-  %tmp5 = load i64 ()** %tmp4                     ; <i64 ()*> [#uses=1]
+  %tmp4 = getelementptr inbounds %class.RuleBasedBreakIterator, %class.RuleBasedBreakIterator* %this, i32 0, i32 0 ; <i64 ()**> [#uses=1]
+  %tmp5 = load i64 ()*, i64 ()** %tmp4                     ; <i64 ()*> [#uses=1]
   %call = invoke i64 %tmp5()
           to label %cond.end unwind label %ehcleanup ; <i64> [#uses=1]
 
@@ -242,10 +242,10 @@ entry:
 ; PR6503
 define void @test12(i32* %A) nounwind {
 entry:
-  %tmp1 = load i32* %A
+  %tmp1 = load i32, i32* %A
   %cmp = icmp ugt i32 1, %tmp1                    ; <i1> [#uses=1]
   %conv = zext i1 %cmp to i32                     ; <i32> [#uses=1]
-  %tmp2 = load i32* %A
+  %tmp2 = load i32, i32* %A
   %cmp3 = icmp ne i32 %tmp2, 0                    ; <i1> [#uses=1]
   %conv4 = zext i1 %cmp3 to i32                   ; <i32> [#uses=1]
   %or = or i32 %conv, %conv4                      ; <i32> [#uses=1]
@@ -258,10 +258,10 @@ entry:
 %s2 = type { i64 }
 define void @test13() nounwind ssp {
 entry:
-  %0 = getelementptr inbounds %s1* null, i64 0, i32 2, i64 0, i32 0
+  %0 = getelementptr inbounds %s1, %s1* null, i64 0, i32 2, i64 0, i32 0
   %1 = bitcast i64* %0 to i32*
-  %2 = getelementptr inbounds %s1* null, i64 0, i32 2, i64 1, i32 0
-  %.pre = load i32* %1, align 8
+  %2 = getelementptr inbounds %s1, %s1* null, i64 0, i32 2, i64 1, i32 0
+  %.pre = load i32, i32* %1, align 8
   %3 = lshr i32 %.pre, 19
   %brmerge = or i1 undef, undef
   %4 = and i32 %3, 3
@@ -269,7 +269,7 @@ entry:
   %6 = shl i32 %5, 19
   %7 = add i32 %6, 1572864
   %8 = and i32 %7, 1572864
-  %9 = load i64* %2, align 8
+  %9 = load i64, i64* %2, align 8
   %trunc156 = trunc i64 %9 to i32
   %10 = and i32 %trunc156, -1537
   %11 = and i32 %10, -6145
@@ -304,7 +304,7 @@ entry:
 
 define void @test15(i32* %p_92) nounwind {
 entry:
-%0 = load i32* %p_92, align 4
+%0 = load i32, i32* %p_92, align 4
 %1 = icmp ne i32 %0, 0
 %2 = zext i1 %1 to i32
 %3 = call i32 @func_14() nounwind
@@ -349,7 +349,7 @@ define double @test16(i32 %a) nounwind {
 
 define %struct.basic_ios *@test17() ssp {
 entry:
-  %add.ptr.i = getelementptr i8* null, i64 undef
+  %add.ptr.i = getelementptr i8, i8* null, i64 undef
   %0 = bitcast i8* %add.ptr.i to %struct.basic_ios*
   ret %struct.basic_ios* %0
 }
diff --git a/test/Transforms/InstCombine/debug-line.ll b/test/Transforms/InstCombine/debug-line.ll
index 1946576183c4..823ec98ebe2b 100644
--- a/test/Transforms/InstCombine/debug-line.ll
+++ b/test/Transforms/InstCombine/debug-line.ll
@@ -5,7 +5,7 @@
 
 define void @foo() nounwind ssp {
 ;CHECK: call i32 @putchar{{.+}} !dbg
-  %1 = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([3 x i8]* @.str, i32 0, i32 0), i32 97), !dbg !5
+  %1 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str, i32 0, i32 0), i32 97), !dbg !5
   ret void, !dbg !7
 }
 
@@ -15,14 +15,14 @@ declare i32 @printf(i8*, ...)
 !llvm.module.flags = !{!10}
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00foo\00foo\00\004\000\001\000\006\000\000\000", !8, !1, !3, null, void ()* @foo, null, null, null} ; [ DW_TAG_subprogram ]
-!1 = !{!"0x29", !8} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang\001\00\000\00\000", !8, !4, !4, !9, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !8, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", line: 4, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, file: !8, scope: !1, type: !3, function: void ()* @foo)
+!1 = !DIFile(filename: "m.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang", isOptimized: true, emissionKind: 0, file: !8, enums: !{}, retainedTypes: !{}, subprograms: !9)
+!3 = !DISubroutineType(types: !4)
 !4 = !{null}
-!5 = !MDLocation(line: 5, column: 2, scope: !6)
-!6 = !{!"0xb\004\0012\000", !8, !0} ; [ DW_TAG_lexical_block ]
-!7 = !MDLocation(line: 6, column: 1, scope: !6)
-!8 = !{!"m.c", !"/private/tmp"}
+!5 = !DILocation(line: 5, column: 2, scope: !6)
+!6 = distinct !DILexicalBlock(line: 4, column: 12, file: !8, scope: !0)
+!7 = !DILocation(line: 6, column: 1, scope: !6)
+!8 = !DIFile(filename: "m.c", directory: "/private/tmp")
 !9 = !{!0}
-!10 = !{i32 1, !"Debug Info Version", i32 2}
+!10 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/InstCombine/debuginfo.ll b/test/Transforms/InstCombine/debuginfo.ll
index ae72f700039e..3875bcc9b8c6 100644
--- a/test/Transforms/InstCombine/debuginfo.ll
+++ b/test/Transforms/InstCombine/debuginfo.ll
@@ -14,15 +14,15 @@ entry:
   store i8* %__dest, i8** %__dest.addr, align 8
 ; CHECK-NOT: call void @llvm.dbg.declare
 ; CHECK: call void @llvm.dbg.value
-  call void @llvm.dbg.declare(metadata i8** %__dest.addr, metadata !0, metadata !{}), !dbg !16
+  call void @llvm.dbg.declare(metadata i8** %__dest.addr, metadata !0, metadata !DIExpression()), !dbg !16
   store i32 %__val, i32* %__val.addr, align 4
-  call void @llvm.dbg.declare(metadata i32* %__val.addr, metadata !7, metadata !{}), !dbg !18
+  call void @llvm.dbg.declare(metadata i32* %__val.addr, metadata !7, metadata !DIExpression()), !dbg !18
   store i64 %__len, i64* %__len.addr, align 8
-  call void @llvm.dbg.declare(metadata i64* %__len.addr, metadata !9, metadata !{}), !dbg !20
-  %tmp = load i8** %__dest.addr, align 8, !dbg !21
-  %tmp1 = load i32* %__val.addr, align 4, !dbg !21
-  %tmp2 = load i64* %__len.addr, align 8, !dbg !21
-  %tmp3 = load i8** %__dest.addr, align 8, !dbg !21
+  call void @llvm.dbg.declare(metadata i64* %__len.addr, metadata !9, metadata !DIExpression()), !dbg !20
+  %tmp = load i8*, i8** %__dest.addr, align 8, !dbg !21
+  %tmp1 = load i32, i32* %__val.addr, align 4, !dbg !21
+  %tmp2 = load i64, i64* %__len.addr, align 8, !dbg !21
+  %tmp3 = load i8*, i8** %__dest.addr, align 8, !dbg !21
   %0 = call i64 @llvm.objectsize.i64.p0i8(i8* %tmp3, i1 false), !dbg !21
   %call = call i8* @foo(i8* %tmp, i32 %tmp1, i64 %tmp2, i64 %0), !dbg !21
   ret i8* %call, !dbg !21
@@ -31,29 +31,29 @@ entry:
 !llvm.dbg.cu = !{!3}
 !llvm.module.flags = !{!30}
 
-!0 = !{!"0x101\00__dest\0016777294\000", !1, !2, !6} ; [ DW_TAG_arg_variable ]
-!1 = !{!"0x2e\00foobar\00foobar\00\0079\001\001\000\006\00256\001\0079", !27, !2, !4, null, i8* (i8*, i32, i64)* @foobar, null, null, !25} ; [ DW_TAG_subprogram ] [line 79] [local] [def] [foobar]
-!2 = !{!"0x29", !27} ; [ DW_TAG_file_type ]
-!3 = !{!"0x11\0012\00clang version 3.0 (trunk 127710)\001\00\000\00\000", !28, !29, !29, !24, null, null} ; [ DW_TAG_compile_unit ]
-!4 = !{!"0x15\00\000\000\000\000\000\000", !27, !2, null, !5, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "__dest", line: 78, arg: 1, scope: !1, file: !2, type: !6)
+!1 = !DISubprogram(name: "foobar", line: 79, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 79, file: !27, scope: !2, type: !4, function: i8* (i8*, i32, i64)* @foobar, variables: !25)
+!2 = !DIFile(filename: "string.h", directory: "Game")
+!3 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.0 (trunk 127710)", isOptimized: true, emissionKind: 0, file: !28, enums: !29, retainedTypes: !29, subprograms: !24)
+!4 = !DISubroutineType(types: !5)
 !5 = !{!6}
-!6 = !{!"0xf\00\000\0064\0064\000\000", null, !3, null} ; [ DW_TAG_pointer_type ]
-!7 = !{!"0x101\00__val\0033554510\000", !1, !2, !8} ; [ DW_TAG_arg_variable ]
-!8 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !3} ; [ DW_TAG_base_type ]
-!9 = !{!"0x101\00__len\0050331726\000", !1, !2, !10} ; [ DW_TAG_arg_variable ]
-!10 = !{!"0x16\00size_t\0080\000\000\000\000", !27, !3, !11} ; [ DW_TAG_typedef ]
-!11 = !{!"0x16\00__darwin_size_t\0090\000\000\000\000", !27, !3, !12} ; [ DW_TAG_typedef ]
-!12 = !{!"0x24\00long unsigned int\000\0064\0064\000\000\007", null, !3} ; [ DW_TAG_base_type ]
-!16 = !MDLocation(line: 78, column: 28, scope: !1)
-!18 = !MDLocation(line: 78, column: 40, scope: !1)
-!20 = !MDLocation(line: 78, column: 54, scope: !1)
-!21 = !MDLocation(line: 80, column: 3, scope: !22)
-!22 = !{!"0xb\0080\003\007", !27, !23} ; [ DW_TAG_lexical_block ]
-!23 = !{!"0xb\0079\001\006", !27, !1} ; [ DW_TAG_lexical_block ]
+!6 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, scope: !3, baseType: null)
+!7 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "__val", line: 78, arg: 2, scope: !1, file: !2, type: !8)
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!9 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "__len", line: 78, arg: 3, scope: !1, file: !2, type: !10)
+!10 = !DIDerivedType(tag: DW_TAG_typedef, name: "size_t", line: 80, file: !27, scope: !3, baseType: !11)
+!11 = !DIDerivedType(tag: DW_TAG_typedef, name: "__darwin_size_t", line: 90, file: !27, scope: !3, baseType: !12)
+!12 = !DIBasicType(tag: DW_TAG_base_type, name: "long unsigned int", size: 64, align: 64, encoding: DW_ATE_unsigned)
+!16 = !DILocation(line: 78, column: 28, scope: !1)
+!18 = !DILocation(line: 78, column: 40, scope: !1)
+!20 = !DILocation(line: 78, column: 54, scope: !1)
+!21 = !DILocation(line: 80, column: 3, scope: !22)
+!22 = distinct !DILexicalBlock(line: 80, column: 3, file: !27, scope: !23)
+!23 = distinct !DILexicalBlock(line: 79, column: 1, file: !27, scope: !1)
 !24 = !{!1}
 !25 = !{!0, !7, !9}
-!26 = !{!"0x29", !28} ; [ DW_TAG_file_type ]
-!27 = !{!"string.h", !"Game"}
-!28 = !{!"bits.c", !"Game"}
-!29 = !{i32 0}
-!30 = !{i32 1, !"Debug Info Version", i32 2}
+!26 = !DIFile(filename: "bits.c", directory: "Game")
+!27 = !DIFile(filename: "string.h", directory: "Game")
+!28 = !DIFile(filename: "bits.c", directory: "Game")
+!29 = !{}
+!30 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/InstCombine/default-alignment.ll b/test/Transforms/InstCombine/default-alignment.ll
new file mode 100644
index 000000000000..718da2136717
--- /dev/null
+++ b/test/Transforms/InstCombine/default-alignment.ll
@@ -0,0 +1,10 @@
+; RUN: opt -verify -instcombine < %s
+%Foo = type <{ i8, x86_fp80 }>
+
+define i8 @t(%Foo* %arg) {
+entry:
+  %0 = getelementptr %Foo, %Foo* %arg, i32 0, i32 0
+  %1 = load i8, i8* %0, align 1
+  ret i8 %1
+}
+
diff --git a/test/Transforms/InstCombine/demand_shrink_nsw.ll b/test/Transforms/InstCombine/demand_shrink_nsw.ll
new file mode 100644
index 000000000000..f49174295167
--- /dev/null
+++ b/test/Transforms/InstCombine/demand_shrink_nsw.ll
@@ -0,0 +1,26 @@
+; RUN: opt -instcombine -o - -S %s | FileCheck %s
+
+; The constant at %v35 should be shrunk, but this must lead to the nsw flag of
+; %v43 getting removed so that %v44 is not illegally optimized away.
+; CHECK-LABEL: @foo
+; CHECK: %v35 = add nuw i32 %v34, 1362915575
+; ...
+; CHECK: add nuw i32 %v42, 1533579450
+; CHECK-NEXT: %v44 = or i32 %v43, -2147483648
+; CHECK-NEXT: %v45 = xor i32 %v44, 749011377
+; CHECK-NEXT: ret i32 %v45
+define i32 @foo(i32 %arg) {
+  %v33 = and i32 %arg, 223
+  %v34 = xor i32 %v33, 29
+  %v35 = add nuw i32 %v34, 3510399223
+  %v37 = or i32 %v34, 1874836915
+  %v38 = and i32 %v34, 221
+  %v39 = xor i32 %v38, 1874836915
+  %v40 = xor i32 %v37, %v39
+  %v41 = shl nsw nuw i32 %v40, 1
+  %v42 = sub i32 %v35, %v41
+  %v43 = add nsw i32 %v42, 1533579450
+  %v44 = or i32 %v43, -2147483648
+  %v45 = xor i32 %v44, 749011377
+  ret i32 %v45
+}
diff --git a/test/Transforms/InstCombine/descale-zero.ll b/test/Transforms/InstCombine/descale-zero.ll
deleted file mode 100644
index 4656837cc05e..000000000000
--- a/test/Transforms/InstCombine/descale-zero.ll
+++ /dev/null
@@ -1,20 +0,0 @@
-; RUN: opt < %s -instcombine -S | FileCheck %s
-
-target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
-target triple = "x86_64-apple-macosx10.10.0"
-
-define internal i8* @descale_zero() {
-entry:
-; CHECK: load i8** inttoptr (i64 48 to i8**), align 16
-; CHECK-NEXT: ret i8*
-  %i16_ptr = load i16** inttoptr (i64 48 to i16**), align 16
-  %num = load i64* inttoptr (i64 64 to i64*), align 64
-  %num_times_2 = shl i64 %num, 1
-  %num_times_2_plus_4 = add i64 %num_times_2, 4
-  %i8_ptr = bitcast i16* %i16_ptr to i8*
-  %i8_ptr_num_times_2_plus_4 = getelementptr i8* %i8_ptr, i64 %num_times_2_plus_4
-  %num_times_neg2 = mul i64 %num, -2
-  %num_times_neg2_minus_4 = add i64 %num_times_neg2, -4
-  %addr = getelementptr i8* %i8_ptr_num_times_2_plus_4, i64 %num_times_neg2_minus_4
-  ret i8* %addr
-}
diff --git a/test/Transforms/InstCombine/disable-simplify-libcalls.ll b/test/Transforms/InstCombine/disable-simplify-libcalls.ll
index 665278861791..e25ce31cbfbf 100644
--- a/test/Transforms/InstCombine/disable-simplify-libcalls.ll
+++ b/test/Transforms/InstCombine/disable-simplify-libcalls.ll
@@ -87,7 +87,7 @@ define double @t5(double %x) {
 
 define i8* @t6(i8* %x) {
 ; CHECK-LABEL: @t6(
-  %empty = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %empty = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   %ret = call i8* @strcat(i8* %x, i8* %empty)
   ret i8* %ret
 ; CHECK: call i8* @strcat
@@ -95,7 +95,7 @@ define i8* @t6(i8* %x) {
 
 define i8* @t7(i8* %x) {
 ; CHECK-LABEL: @t7(
-  %empty = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %empty = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   %ret = call i8* @strncat(i8* %x, i8* %empty, i32 1)
   ret i8* %ret
 ; CHECK: call i8* @strncat
@@ -103,7 +103,7 @@ define i8* @t7(i8* %x) {
 
 define i8* @t8() {
 ; CHECK-LABEL: @t8(
-  %x = getelementptr inbounds [13 x i8]* @.str1, i32 0, i32 0
+  %x = getelementptr inbounds [13 x i8], [13 x i8]* @.str1, i32 0, i32 0
   %ret = call i8* @strchr(i8* %x, i32 119)
   ret i8* %ret
 ; CHECK: call i8* @strchr
@@ -111,7 +111,7 @@ define i8* @t8() {
 
 define i8* @t9() {
 ; CHECK-LABEL: @t9(
-  %x = getelementptr inbounds [13 x i8]* @.str1, i32 0, i32 0
+  %x = getelementptr inbounds [13 x i8], [13 x i8]* @.str1, i32 0, i32 0
   %ret = call i8* @strrchr(i8* %x, i32 119)
   ret i8* %ret
 ; CHECK: call i8* @strrchr
@@ -119,8 +119,8 @@ define i8* @t9() {
 
 define i32 @t10() {
 ; CHECK-LABEL: @t10(
-  %x = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
-  %y = getelementptr inbounds [4 x i8]* @.str3, i32 0, i32 0
+  %x = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
+  %y = getelementptr inbounds [4 x i8], [4 x i8]* @.str3, i32 0, i32 0
   %ret = call i32 @strcmp(i8* %x, i8* %y)
   ret i32 %ret
 ; CHECK: call i32 @strcmp
@@ -128,8 +128,8 @@ define i32 @t10() {
 
 define i32 @t11() {
 ; CHECK-LABEL: @t11(
-  %x = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
-  %y = getelementptr inbounds [4 x i8]* @.str3, i32 0, i32 0
+  %x = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
+  %y = getelementptr inbounds [4 x i8], [4 x i8]* @.str3, i32 0, i32 0
   %ret = call i32 @strncmp(i8* %x, i8* %y, i64 3)
   ret i32 %ret
 ; CHECK: call i32 @strncmp
@@ -137,7 +137,7 @@ define i32 @t11() {
 
 define i8* @t12(i8* %x) {
 ; CHECK-LABEL: @t12(
-  %y = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
+  %y = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
   %ret = call i8* @strcpy(i8* %x, i8* %y)
   ret i8* %ret
 ; CHECK: call i8* @strcpy
@@ -145,7 +145,7 @@ define i8* @t12(i8* %x) {
 
 define i8* @t13(i8* %x) {
 ; CHECK-LABEL: @t13(
-  %y = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
+  %y = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
   %ret = call i8* @stpcpy(i8* %x, i8* %y)
   ret i8* %ret
 ; CHECK: call i8* @stpcpy
@@ -153,7 +153,7 @@ define i8* @t13(i8* %x) {
 
 define i8* @t14(i8* %x) {
 ; CHECK-LABEL: @t14(
-  %y = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
+  %y = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
   %ret = call i8* @strncpy(i8* %x, i8* %y, i64 3)
   ret i8* %ret
 ; CHECK: call i8* @strncpy
@@ -161,7 +161,7 @@ define i8* @t14(i8* %x) {
 
 define i64 @t15() {
 ; CHECK-LABEL: @t15(
-  %x = getelementptr inbounds [4 x i8]* @.str2, i32 0, i32 0
+  %x = getelementptr inbounds [4 x i8], [4 x i8]* @.str2, i32 0, i32 0
   %ret = call i64 @strlen(i8* %x)
   ret i64 %ret
 ; CHECK: call i64 @strlen
@@ -169,7 +169,7 @@ define i64 @t15() {
 
 define i8* @t16(i8* %x) {
 ; CHECK-LABEL: @t16(
-  %y = getelementptr inbounds [1 x i8]* @.str, i32 0, i32 0
+  %y = getelementptr inbounds [1 x i8], [1 x i8]* @.str, i32 0, i32 0
   %ret = call i8* @strpbrk(i8* %x, i8* %y)
   ret i8* %ret
 ; CHECK: call i8* @strpbrk
@@ -177,7 +177,7 @@ define i8* @t16(i8* %x) {
 
 define i64 @t17(i8* %x) {
 ; CHECK-LABEL: @t17(
-  %y = getelementptr inbounds [1 x i8]* @.str, i32 0, i32 0
+  %y = getelementptr inbounds [1 x i8], [1 x i8]* @.str, i32 0, i32 0
   %ret = call i64 @strspn(i8* %x, i8* %y)
   ret i64 %ret
 ; CHECK: call i64 @strspn
@@ -185,7 +185,7 @@ define i64 @t17(i8* %x) {
 
 define double @t18(i8** %y) {
 ; CHECK-LABEL: @t18(
-  %x = getelementptr inbounds [6 x i8]* @.str4, i64 0, i64 0
+  %x = getelementptr inbounds [6 x i8], [6 x i8]* @.str4, i64 0, i64 0
   %ret = call double @strtod(i8* %x, i8** %y)
   ret double %ret
 ; CHECK: call double @strtod
@@ -193,7 +193,7 @@ define double @t18(i8** %y) {
 
 define float @t19(i8** %y) {
 ; CHECK-LABEL: @t19(
-  %x = getelementptr inbounds [6 x i8]* @.str4, i64 0, i64 0
+  %x = getelementptr inbounds [6 x i8], [6 x i8]* @.str4, i64 0, i64 0
   %ret = call float @strtof(i8* %x, i8** %y)
   ret float %ret
 ; CHECK: call float @strtof
@@ -201,7 +201,7 @@ define float @t19(i8** %y) {
 
 define x86_fp80 @t20(i8** %y) {
 ; CHECK-LABEL: @t20(
-  %x = getelementptr inbounds [6 x i8]* @.str4, i64 0, i64 0
+  %x = getelementptr inbounds [6 x i8], [6 x i8]* @.str4, i64 0, i64 0
   %ret = call x86_fp80 @strtold(i8* %x, i8** %y)
   ret x86_fp80 %ret
 ; CHECK: call x86_fp80 @strtold
@@ -209,7 +209,7 @@ define x86_fp80 @t20(i8** %y) {
 
 define i64 @t21(i8** %y) {
 ; CHECK-LABEL: @t21(
-  %x = getelementptr inbounds [5 x i8]* @.str5, i64 0, i64 0
+  %x = getelementptr inbounds [5 x i8], [5 x i8]* @.str5, i64 0, i64 0
   %ret = call i64 @strtol(i8* %x, i8** %y, i32 10)
   ret i64 %ret
 ; CHECK: call i64 @strtol
@@ -217,7 +217,7 @@ define i64 @t21(i8** %y) {
 
 define i64 @t22(i8** %y) {
 ; CHECK-LABEL: @t22(
-  %x = getelementptr inbounds [5 x i8]* @.str5, i64 0, i64 0
+  %x = getelementptr inbounds [5 x i8], [5 x i8]* @.str5, i64 0, i64 0
   %ret = call i64 @strtoll(i8* %x, i8** %y, i32 10)
   ret i64 %ret
 ; CHECK: call i64 @strtoll
@@ -225,7 +225,7 @@ define i64 @t22(i8** %y) {
 
 define i64 @t23(i8** %y) {
 ; CHECK-LABEL: @t23(
-  %x = getelementptr inbounds [5 x i8]* @.str5, i64 0, i64 0
+  %x = getelementptr inbounds [5 x i8], [5 x i8]* @.str5, i64 0, i64 0
   %ret = call i64 @strtoul(i8* %x, i8** %y, i32 10)
   ret i64 %ret
 ; CHECK: call i64 @strtoul
@@ -233,7 +233,7 @@ define i64 @t23(i8** %y) {
 
 define i64 @t24(i8** %y) {
 ; CHECK-LABEL: @t24(
-  %x = getelementptr inbounds [5 x i8]* @.str5, i64 0, i64 0
+  %x = getelementptr inbounds [5 x i8], [5 x i8]* @.str5, i64 0, i64 0
   %ret = call i64 @strtoull(i8* %x, i8** %y, i32 10)
   ret i64 %ret
 ; CHECK: call i64 @strtoull
@@ -241,7 +241,7 @@ define i64 @t24(i8** %y) {
 
 define i64 @t25(i8* %y) {
 ; CHECK-LABEL: @t25(
-  %x = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %x = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   %ret = call i64 @strcspn(i8* %x, i8* %y)
   ret i64 %ret
 ; CHECK: call i64 @strcspn
@@ -277,7 +277,7 @@ define i32 @t29(i64 %y) {
 
 define void @t30() {
 ; CHECK-LABEL: @t30(
-  %x = getelementptr inbounds [13 x i8]* @.str1, i32 0, i32 0
+  %x = getelementptr inbounds [13 x i8], [13 x i8]* @.str1, i32 0, i32 0
   call i32 @fprintf(i8* null, i8* %x)
   ret void
 ; CHECK: call i32 @fprintf
@@ -320,7 +320,7 @@ define i64 @t35(i64 %y) {
 
 define void @t36() {
 ; CHECK-LABEL: @t36(
-  %x = getelementptr inbounds [1 x i8]* @empty, i32 0, i32 0
+  %x = getelementptr inbounds [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   call i32 @printf(i8* %x)
   ret void
 ; CHECK: call i32 @printf
@@ -328,7 +328,7 @@ define void @t36() {
 
 define void @t37(i8* %x) {
 ; CHECK-LABEL: @t37(
-  %y = getelementptr inbounds [13 x i8]* @.str1, i32 0, i32 0
+  %y = getelementptr inbounds [13 x i8], [13 x i8]* @.str1, i32 0, i32 0
   call i32 @sprintf(i8* %x, i8* %y)
   ret void
 ; CHECK: call i32 @sprintf
diff --git a/test/Transforms/InstCombine/div-shift-crash.ll b/test/Transforms/InstCombine/div-shift-crash.ll
index a61972456403..936173cd6d48 100644
--- a/test/Transforms/InstCombine/div-shift-crash.ll
+++ b/test/Transforms/InstCombine/div-shift-crash.ll
@@ -14,7 +14,7 @@ land.lhs.true:                                    ; preds = %entry
   br label %for.cond.i
 
 for.cond.i:                                       ; preds = %land.lhs.true, %entry
-  %0 = getelementptr inbounds %struct.S0.0.1.2.3.4.13.22.31.44.48.53.54.55.56.58.59.60.66.68.70.74.77.106.107.108.109.110.113.117.118.128.129* %l_819.i.i, i64 0, i32 0
+  %0 = getelementptr inbounds %struct.S0.0.1.2.3.4.13.22.31.44.48.53.54.55.56.58.59.60.66.68.70.74.77.106.107.108.109.110.113.117.118.128.129, %struct.S0.0.1.2.3.4.13.22.31.44.48.53.54.55.56.58.59.60.66.68.70.74.77.106.107.108.109.110.113.117.118.128.129* %l_819.i.i, i64 0, i32 0
   br label %for.cond.i6.i.i
 
 for.cond.i6.i.i:                                  ; preds = %for.body.i8.i.i, %for.cond.i
@@ -39,7 +39,7 @@ for.body4.i.i.i.i:                                ; preds = %for.cond1.i.i.i.i
   br label %for.cond1.i.i.i.i
 
 func_39.exit.i.i:                                 ; preds = %for.cond1.i.i.i.i
-  %l_8191.sroa.0.0.copyload.i.i = load i64* %0, align 1
+  %l_8191.sroa.0.0.copyload.i.i = load i64, i64* %0, align 1
   br label %for.cond1.i.i.i
 
 for.cond1.i.i.i:                                  ; preds = %safe_div_func_uint32_t_u_u.exit.i.i.i, %func_39.exit.i.i
diff --git a/test/Transforms/InstCombine/dom-conditions.ll b/test/Transforms/InstCombine/dom-conditions.ll
new file mode 100644
index 000000000000..426404352681
--- /dev/null
+++ b/test/Transforms/InstCombine/dom-conditions.ll
@@ -0,0 +1,152 @@
+; RUN: opt -instcombine -value-tracking-dom-conditions=1 -S < %s | FileCheck %s
+
+target datalayout = "e-p:64:64:64-p1:16:16:16-p2:32:32:32-p3:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+
+define i1 @test_cmp_ult(i64 %A) {
+; CHECK-LABEL: @test_cmp_ult
+entry:
+  %cmp = icmp ult i64 %A, 64
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i1 false
+  %cmp2 = icmp ugt i64 %A, 64
+  ret i1 %cmp2
+untaken:
+  ret i1 true
+}
+
+define i1 @test_cmp_ule(i64 %A) {
+; CHECK-LABEL: @test_cmp_ule
+entry:
+  %cmp = icmp ule i64 %A, 64
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i1 false
+  %cmp2 = icmp ugt i64 %A, 128
+  ret i1 %cmp2
+untaken:
+  ret i1 true
+}
+
+define i1 @test_cmp_sgt(i32 %A) {
+; CHECK-LABEL: @test_cmp_sgt
+entry:
+  %cmp = icmp sgt i32 %A, 10
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i1 true
+  %cmp2 = icmp sgt i32 %A, -1
+  ret i1 %cmp2
+untaken:
+  ret i1 true
+}
+
+define i64 @test_add_zero_bits(i64 %A) {
+; CHECK-LABEL: @test_add_zero_bits
+entry:
+  %cmp = icmp eq i64 %A, 2
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: ret i64 3
+  %add = add i64 %A, 1
+  ret i64 %add
+untaken:
+  ret i64 %A
+}
+
+define i64 @test_add_nsw(i64 %A) {
+; CHECK-LABEL: @test_add_nsw
+entry:
+  %cmp = icmp ult i64 %A, 20
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: %add = add nuw nsw i64 %A, 1
+; CHECK-NEXT: ret i64 %add
+  %add = add i64 %A, 1
+  ret i64 %add
+untaken:
+  ret i64 %A
+}
+
+; After sinking the instructions into the if block, check that we 
+; can simplify some of them using dominating conditions.
+define i32 @test_add_zero_bits_sink(i32 %x) nounwind ssp {
+; CHECK-LABEL: @test_add_zero_bits_sink(
+; CHECK-NOT: sdiv i32
+entry:
+  %a = add nsw i32 %x, 16
+  %b = sdiv i32 %a, %x
+  %cmp = icmp ult i32 %x, 7
+  br i1 %cmp, label %bb1, label %bb2
+
+bb1:
+; CHECK-LABEL: bb1:
+; CHECK-NEXT: or i32 %x, 16
+; CHECK-NEXT: udiv i32
+  ret i32 %b
+
+bb2:
+  ret i32 %x
+}
+
+; A condition in the same block gives no information
+define i32 @test_neg1(i32 %x) nounwind ssp {
+; CHECK-LABEL: @test_neg1
+; CHECK: add
+; CHECK: sdiv
+; CHECK: icmp
+; CHECK: select
+entry:
+  %a = add nsw i32 %x, 16
+  %b = sdiv i32 %a, %x
+  %cmp = icmp ult i32 %x, 7
+  %ret = select i1 %cmp, i32 %a, i32 %b
+  ret i32 %ret
+}
+
+; A non-dominating edge gives no information 
+define i32 @test_neg2(i32 %x) {
+; CHECK-LABEL: @test_neg2
+entry:
+  %cmp = icmp ult i32 %x, 7
+  br i1 %cmp, label %bb1, label %merge
+
+bb1:
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK: icmp
+; CHECK: select
+  %cmp2 = icmp ult i32 %x, 7
+  %ret = select i1 %cmp2, i32 %x, i32 0
+  ret i32 %ret
+}
+
+; A unconditional branch expressed as a condition one gives no
+; information (and shouldn't trip any asserts.)
+define i32 @test_neg3(i32 %x) {
+; CHECK-LABEL: @test_neg3
+entry:
+  %cmp = icmp ult i32 %x, 7
+  br i1 %cmp, label %merge, label %merge
+merge:
+; CHECK-LABEL: merge:
+; CHECK: icmp
+; CHECK: select
+  %cmp2 = icmp ult i32 %x, 7
+  %ret = select i1 %cmp2, i32 %x, i32 0
+  ret i32 %ret
+}
+
+declare i32 @bar()
diff --git a/test/Transforms/InstCombine/enforce-known-alignment.ll b/test/Transforms/InstCombine/enforce-known-alignment.ll
index 46bb60569fe2..323a7ec2998c 100644
--- a/test/Transforms/InstCombine/enforce-known-alignment.ll
+++ b/test/Transforms/InstCombine/enforce-known-alignment.ll
@@ -8,12 +8,12 @@ define void @foo(i32) {
 ; CHECK: alloca
 ; CHECK: align 16
 	%2 = alloca [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>], align 16		; <[3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>]*> [#uses=1]
-	%3 = getelementptr [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>]* %2, i32 0, i32 0		; <<{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>*> [#uses=1]
-	%4 = getelementptr <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>* %3, i32 0, i32 0		; <{ { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }*> [#uses=1]
-	%5 = getelementptr { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }* %4, i32 0, i32 0		; <{ [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 }*> [#uses=1]
+	%3 = getelementptr [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>], [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>]* %2, i32 0, i32 0		; <<{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>*> [#uses=1]
+	%4 = getelementptr <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>, <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>* %3, i32 0, i32 0		; <{ { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }*> [#uses=1]
+	%5 = getelementptr { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }, { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }* %4, i32 0, i32 0		; <{ [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 }*> [#uses=1]
 	%6 = bitcast { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 }* %5 to { [8 x i16] }*		; <{ [8 x i16] }*> [#uses=1]
-	%7 = getelementptr { [8 x i16] }* %6, i32 0, i32 0		; <[8 x i16]*> [#uses=1]
-	%8 = getelementptr [8 x i16]* %7, i32 0, i32 0		; <i16*> [#uses=1]
+	%7 = getelementptr { [8 x i16] }, { [8 x i16] }* %6, i32 0, i32 0		; <[8 x i16]*> [#uses=1]
+	%8 = getelementptr [8 x i16], [8 x i16]* %7, i32 0, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %8, align 16
     call void @bar(i16* %8)
 	ret void
@@ -24,12 +24,12 @@ declare void @bar(i16*)
 define void @foo_as1(i32 %a, [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>] addrspace(1)* %b) {
 ; CHECK-LABEL: @foo_as1(
 ; CHECK: align 16
-  %1 = getelementptr [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>] addrspace(1)* %b, i32 0, i32 0        ; <<{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>*> [#uses=1]
-  %2 = getelementptr <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }> addrspace(1)* %1, i32 0, i32 0      ; <{ { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }*> [#uses=1]
-  %3 = getelementptr { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } addrspace(1)* %2, i32 0, i32 0        ; <{ [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 }*> [#uses=1]
+  %1 = getelementptr [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>], [3 x <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>] addrspace(1)* %b, i32 0, i32 0        ; <<{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>*> [#uses=1]
+  %2 = getelementptr <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }>, <{ { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } }> addrspace(1)* %1, i32 0, i32 0      ; <{ { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }*> [#uses=1]
+  %3 = getelementptr { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } }, { { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } } addrspace(1)* %2, i32 0, i32 0        ; <{ [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 }*> [#uses=1]
   %4 = bitcast { [2 x { { i32 } }], [2 x i8], { i16 }, [2 x i8], i8, i8 } addrspace(1)* %3 to { [8 x i16] } addrspace(1)*     ; <{ [8 x i16] }*> [#uses=1]
-  %5 = getelementptr { [8 x i16] } addrspace(1)* %4, i32 0, i32 0     ; <[8 x i16]*> [#uses=1]
-  %6 = getelementptr [8 x i16] addrspace(1)* %5, i32 0, i32 0     ; <i16*> [#uses=1]
+  %5 = getelementptr { [8 x i16] }, { [8 x i16] } addrspace(1)* %4, i32 0, i32 0     ; <[8 x i16]*> [#uses=1]
+  %6 = getelementptr [8 x i16], [8 x i16] addrspace(1)* %5, i32 0, i32 0     ; <i16*> [#uses=1]
   store i16 0, i16 addrspace(1)* %6, align 16
   call void @bar_as1(i16 addrspace(1)* %6)
   ret void
diff --git a/test/Transforms/InstCombine/err-rep-cold.ll b/test/Transforms/InstCombine/err-rep-cold.ll
index 0cbafc43e0f6..6c4e4e1ccb01 100644
--- a/test/Transforms/InstCombine/err-rep-cold.ll
+++ b/test/Transforms/InstCombine/err-rep-cold.ll
@@ -18,11 +18,11 @@ entry:
   br i1 %cmp, label %if.then, label %return
 
 if.then:                                          ; preds = %entry
-  %0 = load %struct._IO_FILE** @stderr, align 8
-  %call = tail call i32 (%struct._IO_FILE*, i8*, ...)* @fprintf(%struct._IO_FILE* %0, i8* getelementptr inbounds ([13 x i8]* @.str, i64 0, i64 0), i32 %a) #1
+  %0 = load %struct._IO_FILE*, %struct._IO_FILE** @stderr, align 8
+  %call = tail call i32 (%struct._IO_FILE*, i8*, ...) @fprintf(%struct._IO_FILE* %0, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @.str, i64 0, i64 0), i32 %a) #1
   br label %return
 
-; CHECK: %call = tail call i32 (%struct._IO_FILE*, i8*, ...)* @fprintf(%struct._IO_FILE* %0, i8* getelementptr inbounds ([13 x i8]* @.str, i64 0, i64 0), i32 %a) #[[AT1:[0-9]+]]
+; CHECK: %call = tail call i32 (%struct._IO_FILE*, i8*, ...) @fprintf(%struct._IO_FILE* %0, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @.str, i64 0, i64 0), i32 %a) #[[AT1:[0-9]+]]
 
 return:                                           ; preds = %entry, %if.then
   %retval.0 = phi i32 [ 1, %if.then ], [ 0, %entry ]
@@ -38,11 +38,11 @@ entry:
   br i1 %cmp, label %if.then, label %return
 
 if.then:                                          ; preds = %entry
-  %0 = load %struct._IO_FILE** @stderr, align 8
-  %1 = tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0)
+  %0 = load %struct._IO_FILE*, %struct._IO_FILE** @stderr, align 8
+  %1 = tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0)
   br label %return
 
-; CHECK: tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0) #[[AT2:[0-9]+]]
+; CHECK: tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0) #[[AT2:[0-9]+]]
 
 return:                                           ; preds = %entry, %if.then
   %retval.0 = phi i32 [ 1, %if.then ], [ 0, %entry ]
@@ -58,11 +58,11 @@ entry:
   br i1 %cmp, label %if.then, label %return
 
 if.then:                                          ; preds = %entry
-  %0 = load %struct._IO_FILE** @stdout, align 8
-  %1 = tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0)
+  %0 = load %struct._IO_FILE*, %struct._IO_FILE** @stdout, align 8
+  %1 = tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0)
   br label %return
 
-; CHECK-NOT: tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0) #[[AT2]]
+; CHECK-NOT: tail call i64 @fwrite(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str1, i64 0, i64 0), i64 8, i64 1, %struct._IO_FILE* %0) #[[AT2]]
 
 return:                                           ; preds = %entry, %if.then
   %retval.0 = phi i32 [ 1, %if.then ], [ 0, %entry ]
diff --git a/test/Transforms/InstCombine/extractvalue.ll b/test/Transforms/InstCombine/extractvalue.ll
index 04c7ffa219da..6319590873a2 100644
--- a/test/Transforms/InstCombine/extractvalue.ll
+++ b/test/Transforms/InstCombine/extractvalue.ll
@@ -13,7 +13,7 @@ define i32 @foo(i32 %a, i32 %b) {
         %s1 = insertvalue {i32, i32} %s1.1, i32 %b, 1
         %v1 = extractvalue {i32, i32} %s1, 0
         %v2 = extractvalue {i32, i32} %s1, 1
-        
+
         ; Build a nested struct and pull a sub struct out of it
         ; This requires instcombine to insert a few insertvalue instructions
         %ns1.1 = insertvalue {i32, {i32, i32}} undef, i32 %v1, 0
@@ -40,8 +40,8 @@ define i32 @foo(i32 %a, i32 %b) {
 }
 
 ; CHECK-LABEL: define i32 @extract2gep(
-; CHECK-NEXT: [[GEP:%[a-z0-9]+]] = getelementptr inbounds {{.*}}* %pair, i32 0, i32 1
-; CHECK-NEXT: [[LOAD:%[A-Za-z0-9]+]] = load i32* [[GEP]]
+; CHECK-NEXT: [[GEP:%[a-z0-9]+]] = getelementptr inbounds {{.*}}, {{.*}}* %pair, i64 0, i32 1
+; CHECK-NEXT: [[LOAD:%[A-Za-z0-9]+]] = load i32, i32* [[GEP]]
 ; CHECK-NEXT: store
 ; CHECK-NEXT: br label %loop
 ; CHECK-NOT: extractvalue
@@ -52,7 +52,7 @@ define i32 @extract2gep({i32, i32}* %pair, i32* %P) {
         ; The load + extractvalue should be converted
         ; to an inbounds gep + smaller load.
         ; The new load should be in the same spot as the old load.
-        %L = load {i32, i32}* %pair
+        %L = load {i32, i32}, {i32, i32}* %pair
         store i32 0, i32* %P
         br label %loop
 
@@ -68,13 +68,13 @@ end:
 }
 
 ; CHECK-LABEL: define i32 @doubleextract2gep(
-; CHECK-NEXT: [[GEP:%[a-z0-9]+]] = getelementptr inbounds {{.*}}* %arg, i32 0, i32 1, i32 1
-; CHECK-NEXT: [[LOAD:%[A-Za-z0-9]+]] = load i32* [[GEP]]
+; CHECK-NEXT: [[GEP:%[a-z0-9]+]] = getelementptr inbounds {{.*}}, {{.*}}* %arg, i64 0, i32 1, i32 1
+; CHECK-NEXT: [[LOAD:%[A-Za-z0-9]+]] = load i32, i32* [[GEP]]
 ; CHECK-NEXT: ret i32 [[LOAD]]
 define i32 @doubleextract2gep({i32, {i32, i32}}* %arg) {
         ; The load + extractvalues should be converted
         ; to a 3-index inbounds gep + smaller load.
-        %L = load {i32, {i32, i32}}* %arg
+        %L = load {i32, {i32, i32}}, {i32, {i32, i32}}* %arg
         %E1 = extractvalue {i32, {i32, i32}} %L, 1
         %E2 = extractvalue {i32, i32} %E1, 1
         ret i32 %E2
@@ -88,7 +88,7 @@ define i32 @doubleextract2gep({i32, {i32, i32}}* %arg) {
 ; CHECK-NEXT: ret
 define i32 @nogep-multiuse({i32, i32}* %pair) {
         ; The load should be left unchanged since both parts are needed.
-        %L = load volatile {i32, i32}* %pair
+        %L = load volatile {i32, i32}, {i32, i32}* %pair
         %LHS = extractvalue {i32, i32} %L, 0
         %RHS = extractvalue {i32, i32} %L, 1
         %R = add i32 %LHS, %RHS
@@ -101,7 +101,7 @@ define i32 @nogep-multiuse({i32, i32}* %pair) {
 ; CHECK-NEXT: ret
 define i32 @nogep-volatile({i32, i32}* %pair) {
         ; The load volatile should be left unchanged.
-        %L = load volatile {i32, i32}* %pair
+        %L = load volatile {i32, i32}, {i32, i32}* %pair
         %E = extractvalue {i32, i32} %L, 1
         ret i32 %E
 }
diff --git a/test/Transforms/InstCombine/fast-math.ll b/test/Transforms/InstCombine/fast-math.ll
index c6081c399253..4eebdbdfacf1 100644
--- a/test/Transforms/InstCombine/fast-math.ll
+++ b/test/Transforms/InstCombine/fast-math.ll
@@ -267,6 +267,14 @@ define <4 x float> @fmul3_vec(<4 x float> %f1, <4 x float> %f2) {
 ; CHECK: fmul fast <4 x float> %f1, <float 3.000000e+00, float 2.000000e+00, float 1.000000e+00, float 1.000000e+00>
 }
 
+; Make sure fmul with constant expression doesn't assert.
+define <4 x float> @fmul3_vec_constexpr(<4 x float> %f1, <4 x float> %f2) {
+  %constExprMul = bitcast i128 trunc (i160 bitcast (<5 x float> <float 6.0e+3, float 6.0e+3, float 2.0e+3, float 1.0e+3, float undef> to i160) to i128) to <4 x float>  
+  %t1 = fdiv <4 x float> %f1, <float 2.0e+3, float 3.0e+3, float 2.0e+3, float 1.0e+3>
+  %t3 = fmul fast <4 x float> %t1, %constExprMul
+  ret <4 x float> %t3
+}
+
 ; Rule "X/C1 * C2 => X * (C2/C1) is not applicable if C2/C1 is either a special
 ; value of a denormal. The 0x3810000000000000 here take value FLT_MIN
 ;
diff --git a/test/Transforms/InstCombine/fcmp.ll b/test/Transforms/InstCombine/fcmp.ll
index ee39d1084ebc..7fd46f228183 100644
--- a/test/Transforms/InstCombine/fcmp.ll
+++ b/test/Transforms/InstCombine/fcmp.ll
@@ -240,3 +240,38 @@ define i32 @test17(double %a, double (double)* %p) nounwind {
   %conv = zext i1 %cmp to i32
   ret i32 %conv
 }
+
+; Can fold fcmp with undef on one side by choosing NaN for the undef
+define i32 @test18_undef_unordered(float %a) nounwind {
+; CHECK-LABEL: @test18_undef_unordered
+; CHECK: ret i32 1
+  %cmp = fcmp ueq float %a, undef
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+; Can fold fcmp with undef on one side by choosing NaN for the undef
+define i32 @test18_undef_ordered(float %a) nounwind {
+; CHECK-LABEL: @test18_undef_ordered
+; CHECK: ret i32 0
+  %cmp = fcmp oeq float %a, undef
+  %conv = zext i1 %cmp to i32
+  ret i32 %conv
+}
+
+; Can fold fcmp with undef on both side
+;   fcmp u_pred undef, undef -> true
+;   fcmp o_pred undef, undef -> false
+; because whatever you choose for the first undef
+; you can choose NaN for the other undef
+define i1 @test19_undef_unordered() nounwind {
+; CHECK-LABEL: @test19_undef
+; CHECK: ret i1 true
+  %cmp = fcmp ueq float undef, undef
+  ret i1 %cmp
+}
+define i1 @test19_undef_ordered() nounwind {
+; CHECK-LABEL: @test19_undef
+; CHECK: ret i1 false
+  %cmp = fcmp oeq float undef, undef
+  ret i1 %cmp
+}
diff --git a/test/Transforms/InstCombine/fmul.ll b/test/Transforms/InstCombine/fmul.ll
index a776765ac2ca..ac3000fc0514 100644
--- a/test/Transforms/InstCombine/fmul.ll
+++ b/test/Transforms/InstCombine/fmul.ll
@@ -74,7 +74,7 @@ define float @test7(float %x, float %y) {
 ; Don't crash when attempting to cast a constant FMul to an instruction.
 define void @test8(i32* %inout) {
 entry:
-  %0 = load i32* %inout, align 4
+  %0 = load i32, i32* %inout, align 4
   %conv = uitofp i32 %0 to float
   %vecinit = insertelement <4 x float> <float 0.000000e+00, float 0.000000e+00, float 0.000000e+00, float undef>, float %conv, i32 3
   %sub = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %vecinit
diff --git a/test/Transforms/InstCombine/fold-vector-zero.ll b/test/Transforms/InstCombine/fold-vector-zero.ll
index e1d86b6cd07c..bf661df7795d 100644
--- a/test/Transforms/InstCombine/fold-vector-zero.ll
+++ b/test/Transforms/InstCombine/fold-vector-zero.ll
@@ -9,7 +9,7 @@ bb30:
 	%l0 = phi i64 [ -2222, %bb8 ], [ %r23, %bb30 ]
 	%r2 = add i64 %s0, %B
 	%r3 = inttoptr i64 %r2 to <2 x double>*
-	%r4 = load <2 x double>* %r3, align 8
+	%r4 = load <2 x double>, <2 x double>* %r3, align 8
 	%r6 = bitcast <2 x double> %r4 to <2 x i64>
 	%r7 = bitcast <2 x double> zeroinitializer to <2 x i64>
 	%r8 = insertelement <2 x i64> undef, i64 9223372036854775807, i32 0
diff --git a/test/Transforms/InstCombine/fp-ret-bitcast.ll b/test/Transforms/InstCombine/fp-ret-bitcast.ll
index b2fbc0b882db..7106933d2bde 100644
--- a/test/Transforms/InstCombine/fp-ret-bitcast.ll
+++ b/test/Transforms/InstCombine/fp-ret-bitcast.ll
@@ -13,11 +13,11 @@ define void @bork() nounwind  {
 entry:
 	%color = alloca %struct.NSArray*
 	%color.466 = alloca %struct.NSObject*
-	%tmp103 = load %struct.NSArray** %color, align 4
-	%tmp103104 = getelementptr %struct.NSArray* %tmp103, i32 0, i32 0
+	%tmp103 = load %struct.NSArray*, %struct.NSArray** %color, align 4
+	%tmp103104 = getelementptr %struct.NSArray, %struct.NSArray* %tmp103, i32 0, i32 0
 	store %struct.NSObject* %tmp103104, %struct.NSObject** %color.466, align 4
-	%tmp105 = load %struct.objc_selector** @"\01L_OBJC_SELECTOR_REFERENCES_81", align 4
-	%tmp106 = load %struct.NSObject** %color.466, align 4
+	%tmp105 = load %struct.objc_selector*, %struct.objc_selector** @"\01L_OBJC_SELECTOR_REFERENCES_81", align 4
+	%tmp106 = load %struct.NSObject*, %struct.NSObject** %color.466, align 4
 	%tmp107 = call float bitcast (void (%struct.NSObject*, ...)* @objc_msgSend_fpret to float (%struct.NSObject*, %struct.objc_selector*)*)( %struct.NSObject* %tmp106, %struct.objc_selector* %tmp105 ) nounwind
 	br label %exit
 
diff --git a/test/Transforms/InstCombine/fpextend.ll b/test/Transforms/InstCombine/fpextend.ll
index 8640cd2b6f5c..775caadb45a4 100644
--- a/test/Transforms/InstCombine/fpextend.ll
+++ b/test/Transforms/InstCombine/fpextend.ll
@@ -5,7 +5,7 @@
 
 define void @test() nounwind  {
 entry:
-	%tmp = load float* @X, align 4		; <float> [#uses=1]
+	%tmp = load float, float* @X, align 4		; <float> [#uses=1]
 	%tmp1 = fpext float %tmp to double		; <double> [#uses=1]
 	%tmp3 = fadd double %tmp1, 0.000000e+00		; <double> [#uses=1]
 	%tmp34 = fptrunc double %tmp3 to float		; <float> [#uses=1]
@@ -15,9 +15,9 @@ entry:
 
 define void @test2() nounwind  {
 entry:
-	%tmp = load float* @X, align 4		; <float> [#uses=1]
+	%tmp = load float, float* @X, align 4		; <float> [#uses=1]
 	%tmp1 = fpext float %tmp to double		; <double> [#uses=1]
-	%tmp2 = load float* @Y, align 4		; <float> [#uses=1]
+	%tmp2 = load float, float* @Y, align 4		; <float> [#uses=1]
 	%tmp23 = fpext float %tmp2 to double		; <double> [#uses=1]
 	%tmp5 = fmul double %tmp1, %tmp23		; <double> [#uses=1]
 	%tmp56 = fptrunc double %tmp5 to float		; <float> [#uses=1]
@@ -27,9 +27,9 @@ entry:
 
 define void @test3() nounwind  {
 entry:
-	%tmp = load float* @X, align 4		; <float> [#uses=1]
+	%tmp = load float, float* @X, align 4		; <float> [#uses=1]
 	%tmp1 = fpext float %tmp to double		; <double> [#uses=1]
-	%tmp2 = load float* @Y, align 4		; <float> [#uses=1]
+	%tmp2 = load float, float* @Y, align 4		; <float> [#uses=1]
 	%tmp23 = fpext float %tmp2 to double		; <double> [#uses=1]
 	%tmp5 = fdiv double %tmp1, %tmp23		; <double> [#uses=1]
 	%tmp56 = fptrunc double %tmp5 to float		; <float> [#uses=1]
@@ -39,7 +39,7 @@ entry:
 
 define void @test4() nounwind  {
 entry:
-	%tmp = load float* @X, align 4		; <float> [#uses=1]
+	%tmp = load float, float* @X, align 4		; <float> [#uses=1]
 	%tmp1 = fpext float %tmp to double		; <double> [#uses=1]
 	%tmp2 = fsub double -0.000000e+00, %tmp1		; <double> [#uses=1]
 	%tmp34 = fptrunc double %tmp2 to float		; <float> [#uses=1]
diff --git a/test/Transforms/InstCombine/fprintf-1.ll b/test/Transforms/InstCombine/fprintf-1.ll
index 6741345839a0..6d21f39d4fd6 100644
--- a/test/Transforms/InstCombine/fprintf-1.ll
+++ b/test/Transforms/InstCombine/fprintf-1.ll
@@ -19,9 +19,9 @@ declare i32 @fprintf(%FILE*, i8*, ...)
 
 define void @test_simplify1(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify1(
-  %fmt = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt)
-; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0), i32 12, i32 1, %FILE* %fp)
+  %fmt = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt)
+; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0), i32 12, i32 1, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -30,8 +30,8 @@ define void @test_simplify1(%FILE* %fp) {
 
 define void @test_simplify2(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify2(
-  %fmt = getelementptr [3 x i8]* @percent_c, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt, i8 104)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_c, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt, i8 104)
 ; CHECK-NEXT: call i32 @fputc(i32 104, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
@@ -42,10 +42,10 @@ define void @test_simplify2(%FILE* %fp) {
 
 define void @test_simplify3(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify3(
-  %fmt = getelementptr [3 x i8]* @percent_s, i32 0, i32 0
-  %str = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt, i8* %str)
-; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0), i32 12, i32 1, %FILE* %fp)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_s, i32 0, i32 0
+  %str = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt, i8* %str)
+; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0), i32 12, i32 1, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -54,36 +54,36 @@ define void @test_simplify3(%FILE* %fp) {
 
 define void @test_simplify4(%FILE* %fp) {
 ; CHECK-IPRINTF-LABEL: @test_simplify4(
-  %fmt = getelementptr [3 x i8]* @percent_d, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt, i32 187)
-; CHECK-IPRINTF-NEXT: call i32 (%FILE*, i8*, ...)* @fiprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8]* @percent_d, i32 0, i32 0), i32 187)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_d, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt, i32 187)
+; CHECK-IPRINTF-NEXT: call i32 (%FILE*, i8*, ...) @fiprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_d, i32 0, i32 0), i32 187)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify1(%FILE* %fp) {
 ; CHECK-IPRINTF-LABEL: @test_no_simplify1(
-  %fmt = getelementptr [3 x i8]* @percent_f, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt, double 1.87)
-; CHECK-IPRINTF-NEXT: call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_f, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt, double 1.87)
+; CHECK-IPRINTF-NEXT: call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify2(%FILE* %fp, double %d) {
 ; CHECK-LABEL: @test_no_simplify2(
-  %fmt = getelementptr [3 x i8]* @percent_f, i32 0, i32 0
-  call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt, double %d)
-; CHECK-NEXT: call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8]* @percent_f, i32 0, i32 0), double %d)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_f, i32 0, i32 0
+  call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt, double %d)
+; CHECK-NEXT: call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_f, i32 0, i32 0), double %d)
   ret void
 ; CHECK-NEXT: ret void
 }
 
 define i32 @test_no_simplify3(%FILE* %fp) {
 ; CHECK-LABEL: @test_no_simplify3(
-  %fmt = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  %1 = call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* %fmt)
-; CHECK-NEXT: call i32 (%FILE*, i8*, ...)* @fprintf(%FILE* %fp, i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0))
+  %fmt = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  %1 = call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* %fmt)
+; CHECK-NEXT: call i32 (%FILE*, i8*, ...) @fprintf(%FILE* %fp, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0))
   ret i32 %1
 ; CHECK-NEXT: ret i32 %1
 }
diff --git a/test/Transforms/InstCombine/fputs-1.ll b/test/Transforms/InstCombine/fputs-1.ll
index 473610e6120a..1c245753e61f 100644
--- a/test/Transforms/InstCombine/fputs-1.ll
+++ b/test/Transforms/InstCombine/fputs-1.ll
@@ -16,7 +16,7 @@ declare i32 @fputs(i8*, %FILE*)
 
 define void @test_simplify1(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify1(
-  %str = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   call i32 @fputs(i8* %str, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
@@ -26,7 +26,7 @@ define void @test_simplify1(%FILE* %fp) {
 
 define void @test_simplify2(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify2(
-  %str = getelementptr [2 x i8]* @A, i32 0, i32 0
+  %str = getelementptr [2 x i8], [2 x i8]* @A, i32 0, i32 0
   call i32 @fputs(i8* %str, %FILE* %fp)
 ; CHECK-NEXT: call i32 @fputc(i32 65, %FILE* %fp)
   ret void
@@ -35,9 +35,9 @@ define void @test_simplify2(%FILE* %fp) {
 
 define void @test_simplify3(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr [7 x i8]* @hello, i32 0, i32 0
+  %str = getelementptr [7 x i8], [7 x i8]* @hello, i32 0, i32 0
   call i32 @fputs(i8* %str, %FILE* %fp)
-; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([7 x i8]* @hello, i32 0, i32 0), i32 6, i32 1, %FILE* %fp)
+; CHECK-NEXT: call i32 @fwrite(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @hello, i32 0, i32 0), i32 6, i32 1, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
 }
diff --git a/test/Transforms/InstCombine/fwrite-1.ll b/test/Transforms/InstCombine/fwrite-1.ll
index 6f9a8e463a8c..10f0b23e1d8d 100644
--- a/test/Transforms/InstCombine/fwrite-1.ll
+++ b/test/Transforms/InstCombine/fwrite-1.ll
@@ -15,7 +15,7 @@ declare i64 @fwrite(i8*, i64, i64, %FILE *)
 
 define void @test_simplify1(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify1(
-  %str = getelementptr inbounds [1 x i8]* @str, i64 0, i64 0
+  %str = getelementptr inbounds [1 x i8], [1 x i8]* @str, i64 0, i64 0
   call i64 @fwrite(i8* %str, i64 1, i64 1, %FILE* %fp)
 ; CHECK-NEXT: call i32 @fputc(i32 0, %FILE* %fp)
   ret void
@@ -24,7 +24,7 @@ define void @test_simplify1(%FILE* %fp) {
 
 define void @test_simplify2(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify2(
-  %str = getelementptr inbounds [0 x i8]* @empty, i64 0, i64 0
+  %str = getelementptr inbounds [0 x i8], [0 x i8]* @empty, i64 0, i64 0
   call i64 @fwrite(i8* %str, i64 1, i64 0, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
@@ -32,7 +32,7 @@ define void @test_simplify2(%FILE* %fp) {
 
 define void @test_simplify3(%FILE* %fp) {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr inbounds [0 x i8]* @empty, i64 0, i64 0
+  %str = getelementptr inbounds [0 x i8], [0 x i8]* @empty, i64 0, i64 0
   call i64 @fwrite(i8* %str, i64 0, i64 1, %FILE* %fp)
   ret void
 ; CHECK-NEXT: ret void
@@ -40,7 +40,7 @@ define void @test_simplify3(%FILE* %fp) {
 
 define i64 @test_no_simplify1(%FILE* %fp) {
 ; CHECK-LABEL: @test_no_simplify1(
-  %str = getelementptr inbounds [1 x i8]* @str, i64 0, i64 0
+  %str = getelementptr inbounds [1 x i8], [1 x i8]* @str, i64 0, i64 0
   %ret = call i64 @fwrite(i8* %str, i64 1, i64 1, %FILE* %fp)
 ; CHECK-NEXT: call i64 @fwrite
   ret i64 %ret
@@ -49,7 +49,7 @@ define i64 @test_no_simplify1(%FILE* %fp) {
 
 define void @test_no_simplify2(%FILE* %fp, i64 %size) {
 ; CHECK-LABEL: @test_no_simplify2(
-  %str = getelementptr inbounds [1 x i8]* @str, i64 0, i64 0
+  %str = getelementptr inbounds [1 x i8], [1 x i8]* @str, i64 0, i64 0
   call i64 @fwrite(i8* %str, i64 %size, i64 1, %FILE* %fp)
 ; CHECK-NEXT: call i64 @fwrite
   ret void
diff --git a/test/Transforms/InstCombine/gc.relocate.ll b/test/Transforms/InstCombine/gc.relocate.ll
new file mode 100644
index 000000000000..a51aac10eb57
--- /dev/null
+++ b/test/Transforms/InstCombine/gc.relocate.ll
@@ -0,0 +1,21 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
+
+; Uses InstCombine with DataLayout to propagate dereferenceable
+; attribute via gc.relocate: if the derived ptr is dereferenceable(N),
+; then the return attribute of gc.relocate is dereferenceable(N).
+
+declare zeroext i1 @return_i1()
+declare i32 @llvm.experimental.gc.statepoint.p0f_i1f(i64, i32, i1 ()*, i32, i32, ...)
+declare i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32, i32, i32)
+
+define i32 addrspace(1)* @deref(i32 addrspace(1)* dereferenceable(8) %dparam) gc "statepoint-example" {
+; Checks that a dereferenceabler pointer
+; CHECK-LABEL: @deref
+; CHECK: call dereferenceable(8)
+entry:
+    %load = load i32, i32 addrspace(1)* %dparam
+    %tok = tail call i32 (i64, i32, i1 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i1f(i64 0, i32 0, i1 ()* @return_i1, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* %dparam)
+    %relocate = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %tok,  i32 7, i32 7)
+    ret i32 addrspace(1)* %relocate
+}
diff --git a/test/Transforms/InstCombine/gep-addrspace.ll b/test/Transforms/InstCombine/gep-addrspace.ll
index 29511a3c6677..aa46ea671302 100644
--- a/test/Transforms/InstCombine/gep-addrspace.ll
+++ b/test/Transforms/InstCombine/gep-addrspace.ll
@@ -8,11 +8,11 @@ target triple = "x86_64-pc-win32"
 ; make sure that we are not crashing when creating an illegal type
 define void @func(%myStruct addrspace(1)* nocapture %p) nounwind {
 ST:
-  %A = getelementptr inbounds %myStruct addrspace(1)* %p, i64 0
+  %A = getelementptr inbounds %myStruct, %myStruct addrspace(1)* %p, i64 0
   %B = addrspacecast %myStruct addrspace(1)* %A to %myStruct*
-  %C = getelementptr inbounds %myStruct* %B, i32 0, i32 1
-  %D = getelementptr inbounds [3 x float]* %C, i32 0, i32 2
-  %E = load float* %D, align 4
+  %C = getelementptr inbounds %myStruct, %myStruct* %B, i32 0, i32 1
+  %D = getelementptr inbounds [3 x float], [3 x float]* %C, i32 0, i32 2
+  %E = load float, float* %D, align 4
   %F = fsub float %E, undef
   ret void
 }
@@ -23,9 +23,9 @@ ST:
 define void @keep_necessary_addrspacecast(i64 %i, float** %out0, float** %out1) {
 entry:
 ; CHECK-LABEL: @keep_necessary_addrspacecast
-  %0 = getelementptr [256 x float]* addrspacecast ([256 x float] addrspace(3)* @array to [256 x float]*), i64 0, i64 %i
+  %0 = getelementptr [256 x float], [256 x float]* addrspacecast ([256 x float] addrspace(3)* @array to [256 x float]*), i64 0, i64 %i
 ; CHECK: addrspacecast float addrspace(3)* %{{[0-9]+}} to float*
-  %1 = getelementptr [0 x float]* addrspacecast (float addrspace(3)* @scalar to [0 x float]*), i64 0, i64 %i
+  %1 = getelementptr [0 x float], [0 x float]* addrspacecast (float addrspace(3)* @scalar to [0 x float]*), i64 0, i64 %i
 ; CHECK: addrspacecast float addrspace(3)* %{{[0-9]+}} to float*
   store float* %0, float** %out0, align 4
   store float* %1, float** %out1, align 4
diff --git a/test/Transforms/InstCombine/gep-sext.ll b/test/Transforms/InstCombine/gep-sext.ll
new file mode 100644
index 000000000000..36e2aeff02cf
--- /dev/null
+++ b/test/Transforms/InstCombine/gep-sext.ll
@@ -0,0 +1,61 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+target triple = "x86_64-pc-win32"
+
+declare void @use(i32) readonly
+
+; We prefer to canonicalize the machine width gep indices early
+define void @test(i32* %p, i32 %index) {
+; CHECK-LABEL: @test
+; CHECK-NEXT: %1 = sext i32 %index to i64
+; CHECK-NEXT: %addr = getelementptr i32, i32* %p, i64 %1
+  %addr = getelementptr i32, i32* %p, i32 %index
+  %val = load i32, i32* %addr
+  call void @use(i32 %val)
+  ret void
+}
+; If they've already been canonicalized via zext, that's fine
+define void @test2(i32* %p, i32 %index) {
+; CHECK-LABEL: @test2
+; CHECK-NEXT: %i = zext i32 %index to i64
+; CHECK-NEXT: %addr = getelementptr i32, i32* %p, i64 %i
+  %i = zext i32 %index to i64
+  %addr = getelementptr i32, i32* %p, i64 %i
+  %val = load i32, i32* %addr
+  call void @use(i32 %val)
+  ret void
+}
+; If we can use a zext, we prefer that.  This requires
+; knowing that the index is positive.
+define void @test3(i32* %p, i32 %index) {
+; CHECK-LABEL: @test3
+; CHECK:   zext
+; CHECK-NOT: sext
+  %addr_begin = getelementptr i32, i32* %p, i64 40
+  %addr_fixed = getelementptr i32, i32* %addr_begin, i64 48
+  %val_fixed = load i32, i32* %addr_fixed, !range !0
+  %addr = getelementptr i32, i32* %addr_begin, i32 %val_fixed
+  %val = load i32, i32* %addr
+  call void @use(i32 %val)
+  ret void
+}
+; Replace sext with zext where possible
+define void @test4(i32* %p, i32 %index) {
+; CHECK-LABEL: @test4
+; CHECK:   zext
+; CHECK-NOT: sext
+  %addr_begin = getelementptr i32, i32* %p, i64 40
+  %addr_fixed = getelementptr i32, i32* %addr_begin, i64 48
+  %val_fixed = load i32, i32* %addr_fixed, !range !0
+  %i = sext i32 %val_fixed to i64
+  %addr = getelementptr i32, i32* %addr_begin, i64 %i
+  %val = load i32, i32* %addr
+  call void @use(i32 %val)
+  ret void
+}
+
+;;  !range !0
+!0 = !{i32 0, i32 2147483647}
+
+
+
diff --git a/test/Transforms/InstCombine/gepgep.ll b/test/Transforms/InstCombine/gepgep.ll
index 9e681d28c424..24b81aaea3f1 100644
--- a/test/Transforms/InstCombine/gepgep.ll
+++ b/test/Transforms/InstCombine/gepgep.ll
@@ -8,6 +8,6 @@ target triple = "x86_64-unknown-linux-gnu"
 declare void @use(i8*)
 
 define void @f() {
-  call void @use(i8* getelementptr (i8* getelementptr (i8* bitcast ([64 x float]* @buffer to i8*), i64 and (i64 sub (i64 0, i64 ptrtoint ([64 x float]* @buffer to i64)), i64 63)), i64 64))
+  call void @use(i8* getelementptr (i8, i8* getelementptr (i8, i8* bitcast ([64 x float]* @buffer to i8*), i64 and (i64 sub (i64 0, i64 ptrtoint ([64 x float]* @buffer to i64)), i64 63)), i64 64))
   ret void
 }
diff --git a/test/Transforms/InstCombine/gepphigep.ll b/test/Transforms/InstCombine/gepphigep.ll
index 9aab609901e2..5ae3171b39f6 100644
--- a/test/Transforms/InstCombine/gepphigep.ll
+++ b/test/Transforms/InstCombine/gepphigep.ll
@@ -2,55 +2,135 @@
 
 %struct1 = type { %struct2*, i32, i32, i32 }
 %struct2 = type { i32, i32 }
+%struct3 = type { i32, %struct4, %struct4 }
+%struct4 = type { %struct2, %struct2 }
 
 define i32 @test1(%struct1* %dm, i1 %tmp4, i64 %tmp9, i64 %tmp19) {
 bb:
-  %tmp = getelementptr inbounds %struct1* %dm, i64 0, i32 0
-  %tmp1 = load %struct2** %tmp, align 8
+  %tmp = getelementptr inbounds %struct1, %struct1* %dm, i64 0, i32 0
+  %tmp1 = load %struct2*, %struct2** %tmp, align 8
   br i1 %tmp4, label %bb1, label %bb2
 
 bb1:
-  %tmp10 = getelementptr inbounds %struct2* %tmp1, i64 %tmp9
-  %tmp11 = getelementptr inbounds %struct2* %tmp10, i64 0, i32 0
+  %tmp10 = getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp9
+  %tmp11 = getelementptr inbounds %struct2, %struct2* %tmp10, i64 0, i32 0
   store i32 0, i32* %tmp11, align 4
   br label %bb3
 
 bb2:
-  %tmp20 = getelementptr inbounds %struct2* %tmp1, i64 %tmp19
-  %tmp21 = getelementptr inbounds %struct2* %tmp20, i64 0, i32 0
+  %tmp20 = getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp19
+  %tmp21 = getelementptr inbounds %struct2, %struct2* %tmp20, i64 0, i32 0
   store i32 0, i32* %tmp21, align 4
   br label %bb3
 
 bb3:
   %phi = phi %struct2* [ %tmp10, %bb1 ], [ %tmp20, %bb2 ]
-  %tmp24 = getelementptr inbounds %struct2* %phi, i64 0, i32 1
-  %tmp25 = load i32* %tmp24, align 4
+  %tmp24 = getelementptr inbounds %struct2, %struct2* %phi, i64 0, i32 1
+  %tmp25 = load i32, i32* %tmp24, align 4
   ret i32 %tmp25
 
 ; CHECK-LABEL: @test1(
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %tmp9, i32 0
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %tmp19, i32 0
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp9, i32 0
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp19, i32 0
 ; CHECK: %[[PHI:[0-9A-Za-z]+]] = phi i64 [ %tmp9, %bb1 ], [ %tmp19, %bb2 ]
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %[[PHI]], i32 1
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %[[PHI]], i32 1
 
 }
 
 define i32 @test2(%struct1* %dm, i1 %tmp4, i64 %tmp9, i64 %tmp19) {
 bb:
-  %tmp = getelementptr inbounds %struct1* %dm, i64 0, i32 0
-  %tmp1 = load %struct2** %tmp, align 8
-  %tmp10 = getelementptr inbounds %struct2* %tmp1, i64 %tmp9
-  %tmp11 = getelementptr inbounds %struct2* %tmp10, i64 0, i32 0
+  %tmp = getelementptr inbounds %struct1, %struct1* %dm, i64 0, i32 0
+  %tmp1 = load %struct2*, %struct2** %tmp, align 8
+  %tmp10 = getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp9
+  %tmp11 = getelementptr inbounds %struct2, %struct2* %tmp10, i64 0, i32 0
   store i32 0, i32* %tmp11, align 4
-  %tmp20 = getelementptr inbounds %struct2* %tmp1, i64 %tmp19
-  %tmp21 = getelementptr inbounds %struct2* %tmp20, i64 0, i32 0
+  %tmp20 = getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp19
+  %tmp21 = getelementptr inbounds %struct2, %struct2* %tmp20, i64 0, i32 0
   store i32 0, i32* %tmp21, align 4
-  %tmp24 = getelementptr inbounds %struct2* %tmp10, i64 0, i32 1
-  %tmp25 = load i32* %tmp24, align 4
+  %tmp24 = getelementptr inbounds %struct2, %struct2* %tmp10, i64 0, i32 1
+  %tmp25 = load i32, i32* %tmp24, align 4
   ret i32 %tmp25
 
 ; CHECK-LABEL: @test2(
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %tmp9, i32 0
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %tmp19, i32 0
-; CHECK: getelementptr inbounds %struct2* %tmp1, i64 %tmp9, i32 1
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp9, i32 0
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp19, i32 0
+; CHECK: getelementptr inbounds %struct2, %struct2* %tmp1, i64 %tmp9, i32 1
+}
+
+; Check that instcombine doesn't insert GEPs before landingpad.
+
+define i32 @test3(%struct3* %dm, i1 %tmp4, i64 %tmp9, i64 %tmp19, i64 %tmp20, i64 %tmp21) {
+bb:
+  %tmp = getelementptr inbounds %struct3, %struct3* %dm, i64 0
+  br i1 %tmp4, label %bb1, label %bb2
+
+bb1:
+  %tmp1 = getelementptr inbounds %struct3, %struct3* %tmp, i64 %tmp19, i32 1
+  %tmp11 = getelementptr inbounds %struct4, %struct4* %tmp1, i64 0, i32 0, i32 0
+  store i32 0, i32* %tmp11, align 4
+  br label %bb3
+
+bb2:
+  %tmp2 = getelementptr inbounds %struct3, %struct3* %tmp, i64 %tmp20, i32 1
+  %tmp12 = getelementptr inbounds %struct4, %struct4* %tmp2, i64 0, i32 0, i32 1
+  store i32 0, i32* %tmp12, align 4
+  br label %bb3
+
+bb3:
+  %phi = phi %struct4* [ %tmp1, %bb1 ], [ %tmp2, %bb2 ]
+  %tmp22 = invoke i32 @foo1(i32 11) to label %bb4 unwind label %bb5
+
+bb4:
+  ret i32 0
+
+bb5:
+  %tmp27 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*) catch i8* bitcast (i8** @_ZTIi to i8*)
+  %tmp34 = getelementptr inbounds %struct4, %struct4* %phi, i64 %tmp21, i32 1
+  %tmp35 = getelementptr inbounds %struct2, %struct2* %tmp34, i64 0, i32 1
+  %tmp25 = load i32, i32* %tmp35, align 4
+  ret i32 %tmp25
+
+; CHECK-LABEL: @test3(
+; CHECK: bb5:
+; CHECK-NEXT: {{.*}}landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
+}
+
+@_ZTIi = external constant i8*
+declare i32 @__gxx_personality_v0(...)
+declare i32 @foo1(i32)
+
+
+; Check that instcombine doesn't fold GEPs into themselves through a loop
+; back-edge.
+
+define i8* @test4(i32 %value, i8* %buffer) {
+entry:
+  %incptr = getelementptr inbounds i8, i8* %buffer, i64 1
+  %cmp = icmp ugt i32 %value, 127
+  br i1 %cmp, label %loop.header, label %exit
+
+loop.header:
+  br label %loop.body
+
+loop.body:
+  %loopptr = phi i8* [ %incptr, %loop.header ], [ %incptr2, %loop.body ]
+  %newval = phi i32 [ %value, %loop.header ], [ %shr, %loop.body ]
+  %shr = lshr i32 %newval, 7
+  %incptr2 = getelementptr inbounds i8, i8* %loopptr, i64 1
+  %cmp2 = icmp ugt i32 %shr, 127
+  br i1 %cmp2, label %loop.body, label %loop.exit
+
+loop.exit:
+  %exitptr = phi i8* [ %incptr2, %loop.body ]
+  br label %exit
+
+exit:
+  %ptr2 = phi i8* [ %exitptr, %loop.exit ], [ %incptr, %entry ]
+  %incptr3 = getelementptr inbounds i8, i8* %ptr2, i64 1
+  ret i8* %incptr3
+
+; CHECK-LABEL: @test4(
+; CHECK: loop.body:
+; CHECK: getelementptr{{.*}}i64 1
+; CHECK: exit:
 }
diff --git a/test/Transforms/InstCombine/getelementptr.ll b/test/Transforms/InstCombine/getelementptr.ll
index 94cc1803073a..276ada91f3c6 100644
--- a/test/Transforms/InstCombine/getelementptr.ll
+++ b/test/Transforms/InstCombine/getelementptr.ll
@@ -14,14 +14,14 @@ target datalayout = "e-p:64:64-p1:16:16-p2:32:32:32-p3:64:64:64"
 
 ; Test noop elimination
 define i32* @test1(i32* %I) {
-        %A = getelementptr i32* %I, i64 0
+        %A = getelementptr i32, i32* %I, i64 0
         ret i32* %A
 ; CHECK-LABEL: @test1(
 ; CHECK: ret i32* %I
 }
 
 define i32 addrspace(1)* @test1_as1(i32 addrspace(1)* %I) {
-  %A = getelementptr i32 addrspace(1)* %I, i64 0
+  %A = getelementptr i32, i32 addrspace(1)* %I, i64 0
   ret i32 addrspace(1)* %A
 ; CHECK-LABEL: @test1_as1(
 ; CHECK: ret i32 addrspace(1)* %I
@@ -29,7 +29,7 @@ define i32 addrspace(1)* @test1_as1(i32 addrspace(1)* %I) {
 
 ; Test noop elimination
 define i32* @test2(i32* %I) {
-        %A = getelementptr i32* %I
+        %A = getelementptr i32, i32* %I
         ret i32* %A
 ; CHECK-LABEL: @test2(
 ; CHECK: ret i32* %I
@@ -37,38 +37,38 @@ define i32* @test2(i32* %I) {
 
 ; Test that two array indexing geps fold
 define i32* @test3(i32* %I) {
-        %A = getelementptr i32* %I, i64 17
-        %B = getelementptr i32* %A, i64 4
+        %A = getelementptr i32, i32* %I, i64 17
+        %B = getelementptr i32, i32* %A, i64 4
         ret i32* %B
 ; CHECK-LABEL: @test3(
-; CHECK: getelementptr i32* %I, i64 21
+; CHECK: getelementptr i32, i32* %I, i64 21
 }
 
 ; Test that two getelementptr insts fold
 define i32* @test4({ i32 }* %I) {
-        %A = getelementptr { i32 }* %I, i64 1
-        %B = getelementptr { i32 }* %A, i64 0, i32 0
+        %A = getelementptr { i32 }, { i32 }* %I, i64 1
+        %B = getelementptr { i32 }, { i32 }* %A, i64 0, i32 0
         ret i32* %B
 ; CHECK-LABEL: @test4(
-; CHECK: getelementptr { i32 }* %I, i64 1, i32 0
+; CHECK: getelementptr { i32 }, { i32 }* %I, i64 1, i32 0
 }
 
 define void @test5(i8 %B) {
         ; This should be turned into a constexpr instead of being an instruction
-        %A = getelementptr [10 x i8]* @Global, i64 0, i64 4
+        %A = getelementptr [10 x i8], [10 x i8]* @Global, i64 0, i64 4
         store i8 %B, i8* %A
         ret void
 ; CHECK-LABEL: @test5(
-; CHECK: store i8 %B, i8* getelementptr inbounds ([10 x i8]* @Global, i64 0, i64 4)
+; CHECK: store i8 %B, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @Global, i64 0, i64 4)
 }
 
 define void @test5_as1(i8 %B) {
         ; This should be turned into a constexpr instead of being an instruction
-        %A = getelementptr [10 x i8] addrspace(1)* @Global_as1, i16 0, i16 4
+        %A = getelementptr [10 x i8], [10 x i8] addrspace(1)* @Global_as1, i16 0, i16 4
         store i8 %B, i8 addrspace(1)* %A
         ret void
 ; CHECK-LABEL: @test5_as1(
-; CHECK: store i8 %B, i8 addrspace(1)* getelementptr inbounds ([10 x i8] addrspace(1)* @Global_as1, i16 0, i16 4)
+; CHECK: store i8 %B, i8 addrspace(1)* getelementptr inbounds ([10 x i8], [10 x i8] addrspace(1)* @Global_as1, i16 0, i16 4)
 }
 
 %as1_ptr_struct = type { i32 addrspace(1)* }
@@ -80,9 +80,9 @@ define void @test5_as1(i8 %B) {
 ; This should be turned into a constexpr instead of being an instruction
 define void @test_evaluate_gep_nested_as_ptrs(i32 addrspace(2)* %B) {
 ; CHECK-LABEL: @test_evaluate_gep_nested_as_ptrs(
-; CHECK-NEXT: store i32 addrspace(2)* %B, i32 addrspace(2)* addrspace(1)* getelementptr inbounds (%as2_ptr_struct addrspace(1)* @global_as1_as2_ptr, i16 0, i32 0), align 8
+; CHECK-NEXT: store i32 addrspace(2)* %B, i32 addrspace(2)* addrspace(1)* getelementptr inbounds (%as2_ptr_struct, %as2_ptr_struct addrspace(1)* @global_as1_as2_ptr, i16 0, i32 0), align 8
 ; CHECK-NEXT: ret void
-  %A = getelementptr %as2_ptr_struct addrspace(1)* @global_as1_as2_ptr, i16 0, i32 0
+  %A = getelementptr %as2_ptr_struct, %as2_ptr_struct addrspace(1)* @global_as1_as2_ptr, i16 0, i32 0
   store i32 addrspace(2)* %B, i32 addrspace(2)* addrspace(1)* %A
   ret void
 }
@@ -91,26 +91,26 @@ define void @test_evaluate_gep_nested_as_ptrs(i32 addrspace(2)* %B) {
 
 define void @test_evaluate_gep_as_ptrs_array(i8 addrspace(2)* %B) {
 ; CHECK-LABEL: @test_evaluate_gep_as_ptrs_array(
-; CHECK-NEXT: store i8 addrspace(2)* %B, i8 addrspace(2)* addrspace(1)* getelementptr inbounds ([4 x i8 addrspace(2)*] addrspace(1)* @arst, i16 0, i16 2), align 4
+; CHECK-NEXT: store i8 addrspace(2)* %B, i8 addrspace(2)* addrspace(1)* getelementptr inbounds ([4 x i8 addrspace(2)*], [4 x i8 addrspace(2)*] addrspace(1)* @arst, i16 0, i16 2), align 4
 
 ; CHECK-NEXT: ret void
-  %A = getelementptr [4 x i8 addrspace(2)*] addrspace(1)* @arst, i16 0, i16 2
+  %A = getelementptr [4 x i8 addrspace(2)*], [4 x i8 addrspace(2)*] addrspace(1)* @arst, i16 0, i16 2
   store i8 addrspace(2)* %B, i8 addrspace(2)* addrspace(1)* %A
   ret void
 }
 
 define i32* @test7(i32* %I, i64 %C, i64 %D) {
-        %A = getelementptr i32* %I, i64 %C
-        %B = getelementptr i32* %A, i64 %D
+        %A = getelementptr i32, i32* %I, i64 %C
+        %B = getelementptr i32, i32* %A, i64 %D
         ret i32* %B
 ; CHECK-LABEL: @test7(
-; CHECK: %A.sum = add i64 %C, %D
-; CHECK: getelementptr i32* %I, i64 %A.sum
+; CHECK: %A = getelementptr i32, i32* %I, i64 %C
+; CHECK: %B = getelementptr i32, i32* %A, i64 %D
 }
 
 define i8* @test8([10 x i32]* %X) {
         ;; Fold into the cast.
-        %A = getelementptr [10 x i32]* %X, i64 0, i64 0
+        %A = getelementptr [10 x i32], [10 x i32]* %X, i64 0, i64 0
         %B = bitcast i32* %A to i8*
         ret i8* %B
 ; CHECK-LABEL: @test8(
@@ -118,7 +118,7 @@ define i8* @test8([10 x i32]* %X) {
 }
 
 define i32 @test9() {
-        %A = getelementptr { i32, double }* null, i32 0, i32 1
+        %A = getelementptr { i32, double }, { i32, double }* null, i32 0, i32 1
         %B = ptrtoint double* %A to i32
         ret i32 %B
 ; CHECK-LABEL: @test9(
@@ -126,8 +126,8 @@ define i32 @test9() {
 }
 
 define i1 @test10({ i32, i32 }* %x, { i32, i32 }* %y) {
-        %tmp.1 = getelementptr { i32, i32 }* %x, i32 0, i32 1
-        %tmp.3 = getelementptr { i32, i32 }* %y, i32 0, i32 1
+        %tmp.1 = getelementptr { i32, i32 }, { i32, i32 }* %x, i32 0, i32 1
+        %tmp.3 = getelementptr { i32, i32 }, { i32, i32 }* %y, i32 0, i32 1
         ;; seteq x, y
         %tmp.4 = icmp eq i32* %tmp.1, %tmp.3
         ret i1 %tmp.4
@@ -136,7 +136,7 @@ define i1 @test10({ i32, i32 }* %x, { i32, i32 }* %y) {
 }
 
 define i1 @test11({ i32, i32 }* %X) {
-        %P = getelementptr { i32, i32 }* %X, i32 0, i32 0
+        %P = getelementptr { i32, i32 }, { i32, i32 }* %X, i32 0, i32 0
         %Q = icmp eq i32* %P, null
         ret i1 %Q
 ; CHECK-LABEL: @test11(
@@ -147,18 +147,18 @@ define i1 @test11({ i32, i32 }* %X) {
 ; PR4748
 define i32 @test12(%struct.A* %a) {
 entry:
-  %g3 = getelementptr %struct.A* %a, i32 0, i32 1
+  %g3 = getelementptr %struct.A, %struct.A* %a, i32 0, i32 1
   store i32 10, i32* %g3, align 4
 
-  %g4 = getelementptr %struct.A* %a, i32 0, i32 0
+  %g4 = getelementptr %struct.A, %struct.A* %a, i32 0, i32 0
 
   %new_a = bitcast %struct.B* %g4 to %struct.A*
 
-  %g5 = getelementptr %struct.A* %new_a, i32 0, i32 1
-  %a_a = load i32* %g5, align 4
+  %g5 = getelementptr %struct.A, %struct.A* %new_a, i32 0, i32 1
+  %a_a = load i32, i32* %g5, align 4
   ret i32 %a_a
 ; CHECK-LABEL:      @test12(
-; CHECK:      getelementptr %struct.A* %a, i64 0, i32 1
+; CHECK:      getelementptr %struct.A, %struct.A* %a, i64 0, i32 1
 ; CHECK-NEXT: store i32 10, i32* %g3
 ; CHECK-NEXT: ret i32 10
 }
@@ -167,8 +167,8 @@ entry:
 ; PR2235
 %S = type { i32, [ 100 x i32] }
 define i1 @test13(i64 %X, %S* %P) {
-        %A = getelementptr inbounds %S* %P, i32 0, i32 1, i64 %X
-        %B = getelementptr inbounds %S* %P, i32 0, i32 0
+        %A = getelementptr inbounds %S, %S* %P, i32 0, i32 1, i64 %X
+        %B = getelementptr inbounds %S, %S* %P, i32 0, i32 0
 	%C = icmp eq i32* %A, %B
 	ret i1 %C
 ; CHECK-LABEL: @test13(
@@ -180,8 +180,8 @@ define <2 x i1> @test13_vector(<2 x i64> %X, <2 x %S*> %P) nounwind {
 ; CHECK-NEXT: shl nuw <2 x i64> %X, <i64 2, i64 2>
 ; CHECK-NEXT: add <2 x i64> %A.idx, <i64 4, i64 4>
 ; CHECK-NEXT: icmp eq <2 x i64> %A.offs, zeroinitializer
-  %A = getelementptr inbounds <2 x %S*> %P, <2 x i64> zeroinitializer, <2 x i32> <i32 1, i32 1>, <2 x i64> %X
-  %B = getelementptr inbounds <2 x %S*> %P, <2 x i64> <i64 0, i64 0>, <2 x i32> <i32 0, i32 0>
+  %A = getelementptr inbounds %S, <2 x %S*> %P, <2 x i64> zeroinitializer, <2 x i32> <i32 1, i32 1>, <2 x i64> %X
+  %B = getelementptr inbounds %S, <2 x %S*> %P, <2 x i64> <i64 0, i64 0>, <2 x i32> <i32 0, i32 0>
   %C = icmp eq <2 x i32*> %A, %B
   ret <2 x i1> %C
 }
@@ -190,8 +190,8 @@ define i1 @test13_as1(i16 %X, %S addrspace(1)* %P) {
 ; CHECK-LABEL: @test13_as1(
 ; CHECK-NEXT:  %C = icmp eq i16 %X, -1
 ; CHECK-NEXT: ret i1 %C
-  %A = getelementptr inbounds %S addrspace(1)* %P, i16 0, i32 1, i16 %X
-  %B = getelementptr inbounds %S addrspace(1)* %P, i16 0, i32 0
+  %A = getelementptr inbounds %S, %S addrspace(1)* %P, i16 0, i32 1, i16 %X
+  %B = getelementptr inbounds %S, %S addrspace(1)* %P, i16 0, i32 0
   %C = icmp eq i32 addrspace(1)* %A, %B
   ret i1 %C
 }
@@ -202,8 +202,8 @@ define <2 x i1> @test13_vector_as1(<2 x i16> %X, <2 x %S addrspace(1)*> %P) {
 ; CHECK-NEXT: add <2 x i16> %A.idx, <i16 4, i16 4>
 ; CHECK-NEXT: icmp eq <2 x i16> %A.offs, zeroinitializer
 ; CHECK-NEXT: ret <2 x i1>
-  %A = getelementptr inbounds <2 x %S addrspace(1)*> %P, <2 x i16> <i16 0, i16 0>, <2 x i32> <i32 1, i32 1>, <2 x i16> %X
-  %B = getelementptr inbounds <2 x %S addrspace(1)*> %P, <2 x i16> <i16 0, i16 0>, <2 x i32> <i32 0, i32 0>
+  %A = getelementptr inbounds %S, <2 x %S addrspace(1)*> %P, <2 x i16> <i16 0, i16 0>, <2 x i32> <i32 1, i32 1>, <2 x i16> %X
+  %B = getelementptr inbounds %S, <2 x %S addrspace(1)*> %P, <2 x i16> <i16 0, i16 0>, <2 x i32> <i32 0, i32 0>
   %C = icmp eq <2 x i32 addrspace(1)*> %A, %B
   ret <2 x i1> %C
 }
@@ -211,8 +211,8 @@ define <2 x i1> @test13_vector_as1(<2 x i16> %X, <2 x %S addrspace(1)*> %P) {
 define i1 @test13_i32(i32 %X, %S* %P) {
 ; CHECK-LABEL: @test13_i32(
 ; CHECK: %C = icmp eq i32 %X, -1
-  %A = getelementptr inbounds %S* %P, i32 0, i32 1, i32 %X
-  %B = getelementptr inbounds %S* %P, i32 0, i32 0
+  %A = getelementptr inbounds %S, %S* %P, i32 0, i32 1, i32 %X
+  %B = getelementptr inbounds %S, %S* %P, i32 0, i32 0
   %C = icmp eq i32* %A, %B
   ret i1 %C
 }
@@ -220,8 +220,8 @@ define i1 @test13_i32(i32 %X, %S* %P) {
 define i1 @test13_i16(i16 %X, %S* %P) {
 ; CHECK-LABEL: @test13_i16(
 ; CHECK: %C = icmp eq i16 %X, -1
-  %A = getelementptr inbounds %S* %P, i16 0, i32 1, i16 %X
-  %B = getelementptr inbounds %S* %P, i16 0, i32 0
+  %A = getelementptr inbounds %S, %S* %P, i16 0, i32 1, i16 %X
+  %B = getelementptr inbounds %S, %S* %P, i16 0, i32 0
   %C = icmp eq i32* %A, %B
   ret i1 %C
 }
@@ -229,8 +229,8 @@ define i1 @test13_i16(i16 %X, %S* %P) {
 define i1 @test13_i128(i128 %X, %S* %P) {
 ; CHECK-LABEL: @test13_i128(
 ; CHECK: %C = icmp eq i64 %1, -1
-  %A = getelementptr inbounds %S* %P, i128 0, i32 1, i128 %X
-  %B = getelementptr inbounds %S* %P, i128 0, i32 0
+  %A = getelementptr inbounds %S, %S* %P, i128 0, i32 1, i128 %X
+  %B = getelementptr inbounds %S, %S* %P, i128 0, i32 0
   %C = icmp eq i32* %A, %B
   ret i1 %C
 }
@@ -239,25 +239,25 @@ define i1 @test13_i128(i128 %X, %S* %P) {
 @G = external global [3 x i8]
 define i8* @test14(i32 %Idx) {
         %idx = zext i32 %Idx to i64
-        %tmp = getelementptr i8* getelementptr ([3 x i8]* @G, i32 0, i32 0), i64 %idx
+        %tmp = getelementptr i8, i8* getelementptr ([3 x i8], [3 x i8]* @G, i32 0, i32 0), i64 %idx
         ret i8* %tmp
 ; CHECK-LABEL: @test14(
-; CHECK: getelementptr [3 x i8]* @G, i64 0, i64 %idx
+; CHECK: getelementptr [3 x i8], [3 x i8]* @G, i64 0, i64 %idx
 }
 
 
 ; Test folding of constantexpr geps into normal geps.
 @Array = external global [40 x i32]
 define i32 *@test15(i64 %X) {
-        %A = getelementptr i32* getelementptr ([40 x i32]* @Array, i64 0, i64 0), i64 %X
+        %A = getelementptr i32, i32* getelementptr ([40 x i32], [40 x i32]* @Array, i64 0, i64 0), i64 %X
         ret i32* %A
 ; CHECK-LABEL: @test15(
-; CHECK: getelementptr [40 x i32]* @Array, i64 0, i64 %X
+; CHECK: getelementptr [40 x i32], [40 x i32]* @Array, i64 0, i64 %X
 }
 
 
 define i32* @test16(i32* %X, i32 %Idx) {
-        %R = getelementptr i32* %X, i32 %Idx
+        %R = getelementptr i32, i32* %X, i32 %Idx
         ret i32* %R
 ; CHECK-LABEL: @test16(
 ; CHECK: sext i32 %Idx to i64
@@ -265,8 +265,8 @@ define i32* @test16(i32* %X, i32 %Idx) {
 
 
 define i1 @test17(i16* %P, i32 %I, i32 %J) {
-        %X = getelementptr inbounds i16* %P, i32 %I
-        %Y = getelementptr inbounds i16* %P, i32 %J
+        %X = getelementptr inbounds i16, i16* %P, i32 %I
+        %Y = getelementptr inbounds i16, i16* %P, i32 %J
         %C = icmp ult i16* %X, %Y
         ret i1 %C
 ; CHECK-LABEL: @test17(
@@ -274,7 +274,7 @@ define i1 @test17(i16* %P, i32 %I, i32 %J) {
 }
 
 define i1 @test18(i16* %P, i32 %I) {
-        %X = getelementptr inbounds i16* %P, i32 %I
+        %X = getelementptr inbounds i16, i16* %P, i32 %I
         %C = icmp ult i16* %X, %P
         ret i1 %C
 ; CHECK-LABEL: @test18(
@@ -287,7 +287,7 @@ define i1 @test18_as1(i16 addrspace(1)* %P, i32 %I) {
 ; CHECK-NEXT: %1 = trunc i32 %I to i16
 ; CHECK-NEXT: %C = icmp slt i16 %1, 0
 ; CHECK-NEXT: ret i1 %C
-  %X = getelementptr inbounds i16 addrspace(1)* %P, i32 %I
+  %X = getelementptr inbounds i16, i16 addrspace(1)* %P, i32 %I
   %C = icmp ult i16 addrspace(1)* %X, %P
   ret i1 %C
 }
@@ -298,7 +298,7 @@ define i1 @test18_as1_i32(i16 addrspace(1)* %P, i32 %I) {
 ; CHECK-NEXT: %1 = trunc i32 %I to i16
 ; CHECK-NEXT: %C = icmp slt i16 %1, 0
 ; CHECK-NEXT: ret i1 %C
-  %X = getelementptr inbounds i16 addrspace(1)* %P, i32 %I
+  %X = getelementptr inbounds i16, i16 addrspace(1)* %P, i32 %I
   %C = icmp ult i16 addrspace(1)* %X, %P
   ret i1 %C
 }
@@ -307,7 +307,7 @@ define i1 @test18_as1_i32(i16 addrspace(1)* %P, i32 %I) {
 define i1 @test18_i16(i16* %P, i16 %I) {
 ; CHECK-LABEL: @test18_i16(
 ; CHECK: %C = icmp slt i16 %I, 0
-  %X = getelementptr inbounds i16* %P, i16 %I
+  %X = getelementptr inbounds i16, i16* %P, i16 %I
   %C = icmp ult i16* %X, %P
   ret i1 %C
 }
@@ -316,7 +316,7 @@ define i1 @test18_i16(i16* %P, i16 %I) {
 define i1 @test18_i64(i16* %P, i64 %I) {
 ; CHECK-LABEL: @test18_i64(
 ; CHECK: %C = icmp slt i64 %I, 0
-  %X = getelementptr inbounds i16* %P, i64 %I
+  %X = getelementptr inbounds i16, i16* %P, i64 %I
   %C = icmp ult i16* %X, %P
   ret i1 %C
 }
@@ -325,14 +325,14 @@ define i1 @test18_i64(i16* %P, i64 %I) {
 define i1 @test18_i128(i16* %P, i128 %I) {
 ; CHECK-LABEL: @test18_i128(
 ; CHECK: %C = icmp slt i64 %1, 0
-  %X = getelementptr inbounds i16* %P, i128 %I
+  %X = getelementptr inbounds i16, i16* %P, i128 %I
   %C = icmp ult i16* %X, %P
   ret i1 %C
 }
 
 define i32 @test19(i32* %P, i32 %A, i32 %B) {
-        %tmp.4 = getelementptr inbounds i32* %P, i32 %A
-        %tmp.9 = getelementptr inbounds i32* %P, i32 %B
+        %tmp.4 = getelementptr inbounds i32, i32* %P, i32 %A
+        %tmp.9 = getelementptr inbounds i32, i32* %P, i32 %B
         %tmp.10 = icmp eq i32* %tmp.4, %tmp.9
         %tmp.11 = zext i1 %tmp.10 to i32
         ret i32 %tmp.11
@@ -341,7 +341,7 @@ define i32 @test19(i32* %P, i32 %A, i32 %B) {
 }
 
 define i32 @test20(i32* %P, i32 %A, i32 %B) {
-        %tmp.4 = getelementptr inbounds i32* %P, i32 %A
+        %tmp.4 = getelementptr inbounds i32, i32* %P, i32 %A
         %tmp.6 = icmp eq i32* %tmp.4, %P
         %tmp.7 = zext i1 %tmp.6 to i32
         ret i32 %tmp.7
@@ -350,7 +350,7 @@ define i32 @test20(i32* %P, i32 %A, i32 %B) {
 }
 
 define i32 @test20_as1(i32 addrspace(1)* %P, i32 %A, i32 %B) {
-  %tmp.4 = getelementptr inbounds i32 addrspace(1)* %P, i32 %A
+  %tmp.4 = getelementptr inbounds i32, i32 addrspace(1)* %P, i32 %A
   %tmp.6 = icmp eq i32 addrspace(1)* %tmp.4, %P
   %tmp.7 = zext i1 %tmp.6 to i32
   ret i32 %tmp.7
@@ -361,12 +361,12 @@ define i32 @test20_as1(i32 addrspace(1)* %P, i32 %A, i32 %B) {
 
 define i32 @test21() {
         %pbob1 = alloca %intstruct
-        %pbob2 = getelementptr %intstruct* %pbob1
-        %pbobel = getelementptr %intstruct* %pbob2, i64 0, i32 0
-        %rval = load i32* %pbobel
+        %pbob2 = getelementptr %intstruct, %intstruct* %pbob1
+        %pbobel = getelementptr %intstruct, %intstruct* %pbob2, i64 0, i32 0
+        %rval = load i32, i32* %pbobel
         ret i32 %rval
 ; CHECK-LABEL: @test21(
-; CHECK: getelementptr %intstruct* %pbob1, i64 0, i32 0
+; CHECK: getelementptr %intstruct, %intstruct* %pbob1, i64 0, i32 0
 }
 
 
@@ -374,18 +374,18 @@ define i32 @test21() {
 @B = global i32 2               ; <i32*> [#uses=1]
 
 define i1 @test22() {
-        %C = icmp ult i32* getelementptr (i32* @A, i64 1),
-                           getelementptr (i32* @B, i64 2)
+        %C = icmp ult i32* getelementptr (i32, i32* @A, i64 1),
+                           getelementptr (i32, i32* @B, i64 2)
         ret i1 %C
 ; CHECK-LABEL: @test22(
-; CHECK: icmp ult (i32* getelementptr inbounds (i32* @A, i64 1), i32* getelementptr (i32* @B, i64 2))
+; CHECK: icmp ult (i32* getelementptr inbounds (i32, i32* @A, i64 1), i32* getelementptr (i32, i32* @B, i64 2))
 }
 
 
 %X = type { [10 x i32], float }
 
 define i1 @test23() {
-        %A = getelementptr %X* null, i64 0, i32 0, i64 0                ; <i32*> [#uses=1]
+        %A = getelementptr %X, %X* null, i64 0, i32 0, i64 0                ; <i32*> [#uses=1]
         %B = icmp ne i32* %A, null              ; <i1> [#uses=1]
         ret i1 %B
 ; CHECK-LABEL: @test23(
@@ -394,9 +394,9 @@ define i1 @test23() {
 
 define void @test25() {
 entry:
-        %tmp = getelementptr { i64, i64, i64, i64 }* null, i32 0, i32 3         ; <i64*> [#uses=1]
-        %tmp.upgrd.1 = load i64* %tmp           ; <i64> [#uses=1]
-        %tmp8.ui = load i64* null               ; <i64> [#uses=1]
+        %tmp = getelementptr { i64, i64, i64, i64 }, { i64, i64, i64, i64 }* null, i32 0, i32 3         ; <i64*> [#uses=1]
+        %tmp.upgrd.1 = load i64, i64* %tmp           ; <i64> [#uses=1]
+        %tmp8.ui = load i64, i64* null               ; <i64> [#uses=1]
         %tmp8 = bitcast i64 %tmp8.ui to i64             ; <i64> [#uses=1]
         %tmp9 = and i64 %tmp8, %tmp.upgrd.1             ; <i64> [#uses=1]
         %sext = trunc i64 %tmp9 to i32          ; <i32> [#uses=1]
@@ -411,8 +411,8 @@ declare void @foo25(i32, i64)
 
 ; PR1637
 define i1 @test26(i8* %arr) {
-        %X = getelementptr i8* %arr, i32 1
-        %Y = getelementptr i8* %arr, i32 1
+        %X = getelementptr i8, i8* %arr, i32 1
+        %Y = getelementptr i8, i8* %arr, i32 1
         %test = icmp uge i8* %X, %Y
         ret i1 %test
 ; CHECK-LABEL: @test26(
@@ -427,14 +427,14 @@ define i1 @test26(i8* %arr) {
 define i32 @test27(%struct.compat_siginfo* %to, %struct.siginfo_t* %from) {
 entry:
 	%from_addr = alloca %struct.siginfo_t*
-	%tmp344 = load %struct.siginfo_t** %from_addr, align 8
-	%tmp345 = getelementptr %struct.siginfo_t* %tmp344, i32 0, i32 3
-	%tmp346 = getelementptr { { i32, i32, [0 x i8], %struct.sigval_t, i32 }, [88 x i8] }* %tmp345, i32 0, i32 0
+	%tmp344 = load %struct.siginfo_t*, %struct.siginfo_t** %from_addr, align 8
+	%tmp345 = getelementptr %struct.siginfo_t, %struct.siginfo_t* %tmp344, i32 0, i32 3
+	%tmp346 = getelementptr { { i32, i32, [0 x i8], %struct.sigval_t, i32 }, [88 x i8] }, { { i32, i32, [0 x i8], %struct.sigval_t, i32 }, [88 x i8] }* %tmp345, i32 0, i32 0
 	%tmp346347 = bitcast { i32, i32, [0 x i8], %struct.sigval_t, i32 }* %tmp346 to { i32, i32, %struct.sigval_t }*
-	%tmp348 = getelementptr { i32, i32, %struct.sigval_t }* %tmp346347, i32 0, i32 2
-	%tmp349 = getelementptr %struct.sigval_t* %tmp348, i32 0, i32 0
+	%tmp348 = getelementptr { i32, i32, %struct.sigval_t }, { i32, i32, %struct.sigval_t }* %tmp346347, i32 0, i32 2
+	%tmp349 = getelementptr %struct.sigval_t, %struct.sigval_t* %tmp348, i32 0, i32 0
 	%tmp349350 = bitcast i8** %tmp349 to i32*
-	%tmp351 = load i32* %tmp349350, align 8
+	%tmp351 = load i32, i32* %tmp349350, align 8
 	%tmp360 = call i32 asm sideeffect "...",
         "=r,ir,*m,i,0,~{dirflag},~{fpsr},~{flags}"( i32 %tmp351,
          %struct.__large_struct* null, i32 -14, i32 0 )
@@ -450,17 +450,17 @@ entry:
 define i32 @test28() nounwind  {
 entry:
 	%orientations = alloca [1 x [1 x %struct.x]]
-	%tmp3 = call i32 @puts( i8* getelementptr ([6 x i8]* @.str, i32 0, i32 0) ) nounwind
-	%tmp45 = getelementptr inbounds [1 x [1 x %struct.x]]* %orientations, i32 1, i32 0, i32 0
-	%orientations62 = getelementptr [1 x [1 x %struct.x]]* %orientations, i32 0, i32 0, i32 0
+	%tmp3 = call i32 @puts( i8* getelementptr ([6 x i8], [6 x i8]* @.str, i32 0, i32 0) ) nounwind
+	%tmp45 = getelementptr inbounds [1 x [1 x %struct.x]], [1 x [1 x %struct.x]]* %orientations, i32 1, i32 0, i32 0
+	%orientations62 = getelementptr [1 x [1 x %struct.x]], [1 x [1 x %struct.x]]* %orientations, i32 0, i32 0, i32 0
 	br label %bb10
 
 bb10:
 	%indvar = phi i32 [ 0, %entry ], [ %indvar.next, %bb10 ]
 	%tmp.0.reg2mem.0.rec = mul i32 %indvar, -1
 	%tmp12.rec = add i32 %tmp.0.reg2mem.0.rec, -1
-	%tmp12 = getelementptr inbounds %struct.x* %tmp45, i32 %tmp12.rec
-	%tmp16 = call i32 (i8*, ...)* @printf( i8* getelementptr ([12 x i8]* @.str1, i32 0, i32 0), %struct.x* %tmp12 ) nounwind
+	%tmp12 = getelementptr inbounds %struct.x, %struct.x* %tmp45, i32 %tmp12.rec
+	%tmp16 = call i32 (i8*, ...) @printf( i8* getelementptr ([12 x i8], [12 x i8]* @.str1, i32 0, i32 0), %struct.x* %tmp12 ) nounwind
 	%tmp84 = icmp eq %struct.x* %tmp12, %orientations62
 	%indvar.next = add i32 %indvar, 1
 	br i1 %tmp84, label %bb17, label %bb10
@@ -482,12 +482,12 @@ declare i32 @printf(i8*, ...)
 	%T = type <{ i64, i64, i64 }>
 define i32 @test29(i8* %start, i32 %X) nounwind {
 entry:
-	%tmp3 = load i64* null
-	%add.ptr = getelementptr i8* %start, i64 %tmp3
-	%tmp158 = load i32* null
-	%add.ptr159 = getelementptr %T* null, i32 %tmp158
-	%add.ptr209 = getelementptr i8* %start, i64 0
-	%add.ptr212 = getelementptr i8* %add.ptr209, i32 %X
+	%tmp3 = load i64, i64* null
+	%add.ptr = getelementptr i8, i8* %start, i64 %tmp3
+	%tmp158 = load i32, i32* null
+	%add.ptr159 = getelementptr %T, %T* null, i32 %tmp158
+	%add.ptr209 = getelementptr i8, i8* %start, i64 0
+	%add.ptr212 = getelementptr i8, i8* %add.ptr209, i32 %X
 	%cmp214 = icmp ugt i8* %add.ptr212, %add.ptr
 	br i1 %cmp214, label %if.then216, label %if.end363
 
@@ -506,8 +506,8 @@ entry:
 	%0 = alloca i32, i32 %n, align 4
 	%1 = bitcast i32* %0 to [0 x i32]*
 	call void @test30f(i32* %0) nounwind
-	%2 = getelementptr [0 x i32]* %1, i32 0, i32 %m
-	%3 = load i32* %2, align 4
+	%2 = getelementptr [0 x i32], [0 x i32]* %1, i32 0, i32 %m
+	%3 = load i32, i32* %2, align 4
 	ret i32 %3
 ; CHECK-LABEL: @test30(
 ; CHECK: getelementptr i32
@@ -518,8 +518,8 @@ declare void @test30f(i32*)
 
 
 define i1 @test31(i32* %A) {
-        %B = getelementptr i32* %A, i32 1
-        %C = getelementptr i32* %A, i64 1
+        %B = getelementptr i32, i32* %A, i32 1
+        %C = getelementptr i32, i32* %A, i64 1
         %V = icmp eq i32* %B, %C
         ret i1 %V
 ; CHECK-LABEL: @test31(
@@ -530,18 +530,18 @@ define i1 @test31(i32* %A) {
 ; PR1345
 define i8* @test32(i8* %v) {
 	%A = alloca [4 x i8*], align 16
-	%B = getelementptr [4 x i8*]* %A, i32 0, i32 0
+	%B = getelementptr [4 x i8*], [4 x i8*]* %A, i32 0, i32 0
 	store i8* null, i8** %B
 	%C = bitcast [4 x i8*]* %A to { [16 x i8] }*
-	%D = getelementptr { [16 x i8] }* %C, i32 0, i32 0, i32 8
+	%D = getelementptr { [16 x i8] }, { [16 x i8] }* %C, i32 0, i32 0, i32 8
 	%E = bitcast i8* %D to i8**
 	store i8* %v, i8** %E
-	%F = getelementptr [4 x i8*]* %A, i32 0, i32 2
-	%G = load i8** %F
+	%F = getelementptr [4 x i8*], [4 x i8*]* %A, i32 0, i32 2
+	%G = load i8*, i8** %F
 	ret i8* %G
 ; CHECK-LABEL: @test32(
-; CHECK: %D = getelementptr [4 x i8*]* %A, i64 0, i64 1
-; CHECK: %F = getelementptr [4 x i8*]* %A, i64 0, i64 2
+; CHECK: %D = getelementptr [4 x i8*], [4 x i8*]* %A, i64 0, i64 1
+; CHECK: %F = getelementptr [4 x i8*], [4 x i8*]* %A, i64 0, i64 2
 }
 
 ; PR3290
@@ -550,44 +550,44 @@ define i8* @test32(i8* %v) {
 
 define i32* @test33(%struct.Key* %A) {
 ; CHECK-LABEL: @test33(
-; CHECK: getelementptr %struct.Key* %A, i64 0, i32 0, i32 1
+; CHECK: getelementptr %struct.Key, %struct.Key* %A, i64 0, i32 0, i32 1
   %B = bitcast %struct.Key* %A to %struct.anon*
-  %C = getelementptr %struct.anon* %B, i32 0, i32 2
+  %C = getelementptr %struct.anon, %struct.anon* %B, i32 0, i32 2
   ret i32* %C
 }
 
 define i32 addrspace(1)* @test33_as1(%struct.Key addrspace(1)* %A) {
 ; CHECK-LABEL: @test33_as1(
-; CHECK: getelementptr %struct.Key addrspace(1)* %A, i16 0, i32 0, i32 1
+; CHECK: getelementptr %struct.Key, %struct.Key addrspace(1)* %A, i16 0, i32 0, i32 1
   %B = bitcast %struct.Key addrspace(1)* %A to %struct.anon addrspace(1)*
-  %C = getelementptr %struct.anon addrspace(1)* %B, i32 0, i32 2
+  %C = getelementptr %struct.anon, %struct.anon addrspace(1)* %B, i32 0, i32 2
   ret i32 addrspace(1)* %C
 }
 
 define i32 addrspace(1)* @test33_array_as1([10 x i32] addrspace(1)* %A) {
 ; CHECK-LABEL: @test33_array_as1(
-; CHECK: getelementptr [10 x i32] addrspace(1)* %A, i16 0, i16 2
+; CHECK: getelementptr [10 x i32], [10 x i32] addrspace(1)* %A, i16 0, i16 2
   %B = bitcast [10 x i32] addrspace(1)* %A to [5 x i32] addrspace(1)*
-  %C = getelementptr [5 x i32] addrspace(1)* %B, i32 0, i32 2
+  %C = getelementptr [5 x i32], [5 x i32] addrspace(1)* %B, i32 0, i32 2
   ret i32 addrspace(1)* %C
 }
 
 ; Make sure the GEP indices use the right pointer sized integer
 define i32 addrspace(1)* @test33_array_struct_as1([10 x %struct.Key] addrspace(1)* %A) {
 ; CHECK-LABEL: @test33_array_struct_as1(
-; CHECK: getelementptr [10 x %struct.Key] addrspace(1)* %A, i16 0, i16 1, i32 0, i32 0
+; CHECK: getelementptr [10 x %struct.Key], [10 x %struct.Key] addrspace(1)* %A, i16 0, i16 1, i32 0, i32 0
   %B = bitcast [10 x %struct.Key] addrspace(1)* %A to [20 x i32] addrspace(1)*
-  %C = getelementptr [20 x i32] addrspace(1)* %B, i32 0, i32 2
+  %C = getelementptr [20 x i32], [20 x i32] addrspace(1)* %B, i32 0, i32 2
   ret i32 addrspace(1)* %C
 }
 
 define i32 addrspace(1)* @test33_addrspacecast(%struct.Key* %A) {
 ; CHECK-LABEL: @test33_addrspacecast(
-; CHECK: %C = getelementptr %struct.Key* %A, i64 0, i32 0, i32 1
+; CHECK: %C = getelementptr %struct.Key, %struct.Key* %A, i64 0, i32 0, i32 1
 ; CHECK-NEXT: addrspacecast i32* %C to i32 addrspace(1)*
 ; CHECK-NEXT: ret
   %B = addrspacecast %struct.Key* %A to %struct.anon addrspace(1)*
-  %C = getelementptr %struct.anon addrspace(1)* %B, i32 0, i32 2
+  %C = getelementptr %struct.anon, %struct.anon addrspace(1)* %B, i32 0, i32 2
   ret i32 addrspace(1)* %C
 }
 
@@ -596,10 +596,10 @@ define i8* @test34(i8* %Val, i64 %V) nounwind {
 entry:
 	%A = alloca %T2, align 8
 	%mrv_gep = bitcast %T2* %A to i64*
-	%B = getelementptr %T2* %A, i64 0, i32 0
+	%B = getelementptr %T2, %T2* %A, i64 0, i32 0
 
       	store i64 %V, i64* %mrv_gep
-	%C = load i8** %B, align 8
+	%C = load i8*, i8** %B, align 8
 	ret i8* %C
 ; CHECK-LABEL: @test34(
 ; CHECK: %[[C:.*]] = inttoptr i64 %V to i8*
@@ -617,11 +617,11 @@ entry:
 ; Instcombine should be able to fold this getelementptr.
 
 define i32 @test35() nounwind {
-  call i32 (i8*, ...)* @printf(i8* getelementptr ([17 x i8]* @"\01LC8", i32 0, i32 0),
-             i8* getelementptr (%t1* bitcast (%t0* @s to %t1*), i32 0, i32 1, i32 0)) nounwind
+  call i32 (i8*, ...) @printf(i8* getelementptr ([17 x i8], [17 x i8]* @"\01LC8", i32 0, i32 0),
+             i8* getelementptr (%t1, %t1* bitcast (%t0* @s to %t1*), i32 0, i32 1, i32 0)) nounwind
   ret i32 0
 ; CHECK-LABEL: @test35(
-; CHECK: call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([17 x i8]* @"\01LC8", i64 0, i64 0), i8* getelementptr inbounds (%t0* @s, i64 0, i32 1, i64 0)) [[NUW:#[0-9]+]]
+; CHECK: call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([17 x i8], [17 x i8]* @"\01LC8", i64 0, i64 0), i8* getelementptr inbounds (%t0, %t0* @s, i64 0, i32 1, i64 0)) [[NUW:#[0-9]+]]
 }
 
 ; Instcombine should constant-fold the GEP so that indices that have
@@ -630,9 +630,9 @@ define i32 @test35() nounwind {
 ; the same address is computed, but 3 is in the range of [0,11).
 
 define i8* @test36() nounwind {
-  ret i8* getelementptr ([11 x i8]* @array, i32 0, i64 -1)
+  ret i8* getelementptr ([11 x i8], [11 x i8]* @array, i32 0, i64 -1)
 ; CHECK-LABEL: @test36(
-; CHECK: ret i8* getelementptr ([11 x i8]* @array, i64 1676976733973595601, i64 4)
+; CHECK: ret i8* getelementptr ([11 x i8], [11 x i8]* @array, i64 1676976733973595601, i64 4)
 }
 
 ; Instcombine shouldn't assume that gep(A,0,1) != gep(A,1,0).
@@ -640,18 +640,18 @@ define i8* @test36() nounwind {
 define i1 @test37() nounwind {
 ; CHECK-LABEL: @test37(
 ; CHECK: ret i1 true
-  %t = icmp eq i8* getelementptr ([1 x i8]* @A37, i64 0, i64 1),
-                   getelementptr ([1 x i8]* @A37, i64 1, i64 0)
+  %t = icmp eq i8* getelementptr ([1 x i8], [1 x i8]* @A37, i64 0, i64 1),
+                   getelementptr ([1 x i8], [1 x i8]* @A37, i64 1, i64 0)
   ret i1 %t
 }
 
 ; Test index promotion
 define i32* @test38(i32* %I, i32 %n) {
-        %A = getelementptr i32* %I, i32 %n
+        %A = getelementptr i32, i32* %I, i32 %n
         ret i32* %A
 ; CHECK-LABEL: @test38(
 ; CHECK: = sext i32 %n to i64
-; CHECK: %A = getelementptr i32* %I, i64 %
+; CHECK: %A = getelementptr i32, i32* %I, i64 %
 }
 
 ; Test that we don't duplicate work when the second gep is a "bitcast".
@@ -660,14 +660,14 @@ declare void @pr10322_f2(%pr10322_t*)
 declare void @pr10322_f3(i8**)
 define void @pr10322_f1(%pr10322_t* %foo) {
 entry:
-  %arrayidx8 = getelementptr inbounds %pr10322_t* %foo, i64 2
+  %arrayidx8 = getelementptr inbounds %pr10322_t, %pr10322_t* %foo, i64 2
   call void @pr10322_f2(%pr10322_t* %arrayidx8) nounwind
-  %tmp2 = getelementptr inbounds %pr10322_t* %arrayidx8, i64 0, i32 0
+  %tmp2 = getelementptr inbounds %pr10322_t, %pr10322_t* %arrayidx8, i64 0, i32 0
   call void @pr10322_f3(i8** %tmp2) nounwind
   ret void
 
 ; CHECK-LABEL: @pr10322_f1(
-; CHECK: %tmp2 = getelementptr inbounds %pr10322_t* %arrayidx8, i64 0, i32 0
+; CHECK: %tmp2 = getelementptr inbounds %pr10322_t, %pr10322_t* %arrayidx8, i64 0, i32 0
 }
 
 ; Test that we combine the last two geps in this sequence, before we
@@ -676,14 +676,14 @@ entry:
 %three_gep_t2 = type {%three_gep_t}
 
 define void @three_gep_f(%three_gep_t2* %x) {
-  %gep1 = getelementptr %three_gep_t2* %x, i64 2
+  %gep1 = getelementptr %three_gep_t2, %three_gep_t2* %x, i64 2
   call void @three_gep_h(%three_gep_t2* %gep1)
-  %gep2 = getelementptr %three_gep_t2* %gep1, i64 0, i32 0
-  %gep3 = getelementptr %three_gep_t* %gep2, i64 0, i32 0
+  %gep2 = getelementptr %three_gep_t2, %three_gep_t2* %gep1, i64 0, i32 0
+  %gep3 = getelementptr %three_gep_t, %three_gep_t* %gep2, i64 0, i32 0
   call void @three_gep_g(i32* %gep3)
 
 ; CHECK-LABEL: @three_gep_f(
-; CHECK: %gep3 = getelementptr %three_gep_t2* %gep1, i64 0, i32 0, i32 0
+; CHECK: %gep3 = getelementptr %three_gep_t2, %three_gep_t2* %gep1, i64 0, i32 0, i32 0
   ret void
 }
 
@@ -694,21 +694,21 @@ declare void @three_gep_h(%three_gep_t2*)
 %struct.zot = type { i64, i8 }
 
 define void @test39(%struct.ham* %arg, i8 %arg1) nounwind {
-  %tmp = getelementptr inbounds %struct.ham* %arg, i64 0, i32 2
-  %tmp2 = load %struct.zot** %tmp, align 8
+  %tmp = getelementptr inbounds %struct.ham, %struct.ham* %arg, i64 0, i32 2
+  %tmp2 = load %struct.zot*, %struct.zot** %tmp, align 8
   %tmp3 = bitcast %struct.zot* %tmp2 to i8*
-  %tmp4 = getelementptr inbounds i8* %tmp3, i64 -8
+  %tmp4 = getelementptr inbounds i8, i8* %tmp3, i64 -8
   store i8 %arg1, i8* %tmp4, align 8
   ret void
 
 ; CHECK-LABEL: @test39(
-; CHECK: getelementptr inbounds %struct.ham* %arg, i64 0, i32 2
-; CHECK: getelementptr inbounds i8* %{{.+}}, i64 -8
+; CHECK: getelementptr inbounds %struct.ham, %struct.ham* %arg, i64 0, i32 2
+; CHECK: getelementptr inbounds i8, i8* %{{.+}}, i64 -8
 }
 
 define i1 @pr16483([1 x i8]* %a, [1 x i8]* %b) {
-  %c = getelementptr [1 x i8]* %a, i32 0, i32 0
-  %d = getelementptr [1 x i8]* %b, i32 0, i32 0
+  %c = getelementptr [1 x i8], [1 x i8]* %a, i32 0, i32 0
+  %d = getelementptr [1 x i8], [1 x i8]* %b, i32 0, i32 0
   %cmp = icmp ult i8* %c, %d
   ret i1 %cmp
 
@@ -718,82 +718,82 @@ define i1 @pr16483([1 x i8]* %a, [1 x i8]* %b) {
 
 define i8 @test_gep_bitcast_as1(i32 addrspace(1)* %arr, i16 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_as1(
-; CHECK: getelementptr i32 addrspace(1)* %arr, i16 %N
+; CHECK: getelementptr i32, i32 addrspace(1)* %arr, i16 %N
 ; CHECK: bitcast
   %cast = bitcast i32 addrspace(1)* %arr to i8 addrspace(1)*
   %V = mul i16 %N, 4
-  %t = getelementptr i8 addrspace(1)* %cast, i16 %V
-  %x = load i8 addrspace(1)* %t
+  %t = getelementptr i8, i8 addrspace(1)* %cast, i16 %V
+  %x = load i8, i8 addrspace(1)* %t
   ret i8 %x
 }
 
 ; The element size of the array matches the element size of the pointer
 define i64 @test_gep_bitcast_array_same_size_element([100 x double]* %arr, i64 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_array_same_size_element(
-; CHECK: getelementptr [100 x double]* %arr, i64 0, i64 %V
+; CHECK: getelementptr [100 x double], [100 x double]* %arr, i64 0, i64 %V
 ; CHECK: bitcast
   %cast = bitcast [100 x double]* %arr to i64*
   %V = mul i64 %N, 8
-  %t = getelementptr i64* %cast, i64 %V
-  %x = load i64* %t
+  %t = getelementptr i64, i64* %cast, i64 %V
+  %x = load i64, i64* %t
   ret i64 %x
 }
 
 ; gep should be done in the original address space.
 define i64 @test_gep_bitcast_array_same_size_element_addrspacecast([100 x double]* %arr, i64 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_array_same_size_element_addrspacecast(
-; CHECK: getelementptr [100 x double]* %arr, i64 0, i64 %V
+; CHECK: getelementptr [100 x double], [100 x double]* %arr, i64 0, i64 %V
 ; CHECK-NEXT: bitcast double*
 ; CHECK-NEXT: %t = addrspacecast i64*
-; CHECK: load i64 addrspace(3)* %t
+; CHECK: load i64, i64 addrspace(3)* %t
   %cast = addrspacecast [100 x double]* %arr to i64 addrspace(3)*
   %V = mul i64 %N, 8
-  %t = getelementptr i64 addrspace(3)* %cast, i64 %V
-  %x = load i64 addrspace(3)* %t
+  %t = getelementptr i64, i64 addrspace(3)* %cast, i64 %V
+  %x = load i64, i64 addrspace(3)* %t
   ret i64 %x
 }
 
 ; The element size of the array is different the element size of the pointer
 define i8 @test_gep_bitcast_array_different_size_element([100 x double]* %arr, i64 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_array_different_size_element(
-; CHECK: getelementptr [100 x double]* %arr, i64 0, i64 %N
+; CHECK: getelementptr [100 x double], [100 x double]* %arr, i64 0, i64 %N
 ; CHECK: bitcast
   %cast = bitcast [100 x double]* %arr to i8*
   %V = mul i64 %N, 8
-  %t = getelementptr i8* %cast, i64 %V
-  %x = load i8* %t
+  %t = getelementptr i8, i8* %cast, i64 %V
+  %x = load i8, i8* %t
   ret i8 %x
 }
 
 define i64 @test_gep_bitcast_array_same_size_element_as1([100 x double] addrspace(1)* %arr, i16 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_array_same_size_element_as1(
-; CHECK: getelementptr [100 x double] addrspace(1)* %arr, i16 0, i16 %V
+; CHECK: getelementptr [100 x double], [100 x double] addrspace(1)* %arr, i16 0, i16 %V
 ; CHECK: bitcast
   %cast = bitcast [100 x double] addrspace(1)* %arr to i64 addrspace(1)*
   %V = mul i16 %N, 8
-  %t = getelementptr i64 addrspace(1)* %cast, i16 %V
-  %x = load i64 addrspace(1)* %t
+  %t = getelementptr i64, i64 addrspace(1)* %cast, i16 %V
+  %x = load i64, i64 addrspace(1)* %t
   ret i64 %x
 }
 
 define i8 @test_gep_bitcast_array_different_size_element_as1([100 x double] addrspace(1)* %arr, i16 %N) {
 ; CHECK-LABEL: @test_gep_bitcast_array_different_size_element_as1(
-; CHECK: getelementptr [100 x double] addrspace(1)* %arr, i16 0, i16 %N
+; CHECK: getelementptr [100 x double], [100 x double] addrspace(1)* %arr, i16 0, i16 %N
 ; CHECK: bitcast
   %cast = bitcast [100 x double] addrspace(1)* %arr to i8 addrspace(1)*
   %V = mul i16 %N, 8
-  %t = getelementptr i8 addrspace(1)* %cast, i16 %V
-  %x = load i8 addrspace(1)* %t
+  %t = getelementptr i8, i8 addrspace(1)* %cast, i16 %V
+  %x = load i8, i8 addrspace(1)* %t
   ret i8 %x
 }
 
 define i64 @test40() {
   %array = alloca [3 x i32], align 4
-  %gep = getelementptr inbounds [3 x i32]* %array, i64 0, i64 2
+  %gep = getelementptr inbounds [3 x i32], [3 x i32]* %array, i64 0, i64 2
   %gepi8 = bitcast i32* %gep to i8*
   %p = ptrtoint [3 x i32]* %array to i64
   %np = sub i64 0, %p
-  %gep2 = getelementptr i8* %gepi8, i64 %np
+  %gep2 = getelementptr i8, i8* %gepi8, i64 %np
   %ret = ptrtoint i8* %gep2 to i64
   ret i64 %ret
 
@@ -802,11 +802,11 @@ define i64 @test40() {
 }
 
 define i16 @test41([3 x i32] addrspace(1)* %array) {
-  %gep = getelementptr inbounds [3 x i32] addrspace(1)* %array, i16 0, i16 2
+  %gep = getelementptr inbounds [3 x i32], [3 x i32] addrspace(1)* %array, i16 0, i16 2
   %gepi8 = bitcast i32 addrspace(1)* %gep to i8 addrspace(1)*
   %p = ptrtoint [3 x i32] addrspace(1)* %array to i16
   %np = sub i16 0, %p
-  %gep2 = getelementptr i8 addrspace(1)* %gepi8, i16 %np
+  %gep2 = getelementptr i8, i8 addrspace(1)* %gepi8, i16 %np
   %ret = ptrtoint i8 addrspace(1)* %gep2 to i16
   ret i16 %ret
 
@@ -817,7 +817,7 @@ define i16 @test41([3 x i32] addrspace(1)* %array) {
 define i8* @test42(i8* %c1, i8* %c2) {
   %ptrtoint = ptrtoint i8* %c1 to i64
   %sub = sub i64 0, %ptrtoint
-  %gep = getelementptr inbounds i8* %c2, i64 %sub
+  %gep = getelementptr inbounds i8, i8* %c2, i64 %sub
   ret i8* %gep
 
 ; CHECK-LABEL: @test42(
@@ -832,7 +832,7 @@ define i16* @test43(i16* %c1, i16* %c2) {
   %ptrtoint = ptrtoint i16* %c1 to i64
   %sub = sub i64 0, %ptrtoint
   %shr = ashr i64 %sub, 1
-  %gep = getelementptr inbounds i16* %c2, i64 %shr
+  %gep = getelementptr inbounds i16, i16* %c2, i64 %shr
   ret i16* %gep
 
 ; CHECK-LABEL: @test43(
@@ -847,7 +847,7 @@ define %struct.C* @test44(%struct.C* %c1, %struct.C* %c2) {
   %ptrtoint = ptrtoint %struct.C* %c1 to i64
   %sub = sub i64 0, %ptrtoint
   %shr = sdiv i64 %sub, 7
-  %gep = getelementptr inbounds %struct.C* %c2, i64 %shr
+  %gep = getelementptr inbounds %struct.C, %struct.C* %c2, i64 %shr
   ret %struct.C* %gep
 
 ; CHECK-LABEL: @test44(
@@ -863,7 +863,7 @@ define %struct.C* @test45(%struct.C* %c1, %struct.C** %c2) {
   %ptrtoint2 = ptrtoint %struct.C** %c2 to i64
   %sub = sub i64 %ptrtoint2, %ptrtoint1 ; C2 - C1
   %shr = sdiv i64 %sub, 7
-  %gep = getelementptr inbounds %struct.C* %c1, i64 %shr ; C1 + (C2 - C1)
+  %gep = getelementptr inbounds %struct.C, %struct.C* %c1, i64 %shr ; C1 + (C2 - C1)
   ret %struct.C* %gep
 
 ; CHECK-LABEL: @test45(
@@ -875,14 +875,14 @@ define %struct.C* @test46(%struct.C* %c1, %struct.C* %c2, i64 %N) {
   %ptrtoint = ptrtoint %struct.C* %c1 to i64
   %sub = sub i64 0, %ptrtoint
   %sdiv = sdiv i64 %sub, %N
-  %gep = getelementptr inbounds %struct.C* %c2, i64 %sdiv
+  %gep = getelementptr inbounds %struct.C, %struct.C* %c2, i64 %sdiv
   ret %struct.C* %gep
 
 ; CHECK-LABEL: @test46(
 ; CHECK-NEXT:  [[PTRTOINT:%.*]] = ptrtoint %struct.C* %c1 to i64
 ; CHECK-NEXT:  [[SUB:%.*]] = sub i64 0, [[PTRTOINT]]
 ; CHECK-NEXT:  [[SDIV:%.*]] = sdiv i64 [[SUB]], %N
-; CHECK-NEXT:  [[GEP:%.*]] = getelementptr inbounds %struct.C* %c2, i64 %sdiv
+; CHECK-NEXT:  [[GEP:%.*]] = getelementptr inbounds %struct.C, %struct.C* %c2, i64 %sdiv
 ; CHECK-NEXT:  ret %struct.C* [[GEP]]
 }
 
@@ -890,7 +890,7 @@ define i32 addrspace(1)* @ascast_0_gep(i32* %p) nounwind {
 ; CHECK-LABEL: @ascast_0_gep(
 ; CHECK-NOT: getelementptr
 ; CHECK: ret
-  %gep = getelementptr i32* %p, i32 0
+  %gep = getelementptr i32, i32* %p, i32 0
   %x = addrspacecast i32* %gep to i32 addrspace(1)*
   ret i32 addrspace(1)* %x
 }
@@ -902,7 +902,7 @@ define i32 addrspace(1)* @ascast_0_0_gep([128 x i32]* %p) nounwind {
 ; CHECK-NEXT: getelementptr [128 x i32]
 ; CHECK-NEXT: addrspacecast i32*
 ; CHECK-NEXT: ret i32 addrspace(1)*
-  %gep = getelementptr [128 x i32]* %p, i32 0, i32 0
+  %gep = getelementptr [128 x i32], [128 x i32]* %p, i32 0, i32 0
   %x = addrspacecast i32* %gep to i32 addrspace(1)*
   ret i32 addrspace(1)* %x
 }
diff --git a/test/Transforms/InstCombine/icmp-range.ll b/test/Transforms/InstCombine/icmp-range.ll
index 0911ab03c601..041adf76b5e1 100644
--- a/test/Transforms/InstCombine/icmp-range.ll
+++ b/test/Transforms/InstCombine/icmp-range.ll
@@ -6,14 +6,14 @@
 define i1 @test_nonzero(i32* nocapture readonly %arg) {
 ; CHECK-LABEL:test_nonzero
 ; CHECK: ret i1 true
-  %val = load i32* %arg, !range !0
+  %val = load i32, i32* %arg, !range !0
   %rval = icmp ne i32 %val, 0
   ret i1 %rval
 }
 define i1 @test_nonzero2(i32* nocapture readonly %arg) {
 ; CHECK-LABEL:test_nonzero2
 ; CHECK: ret i1 false
-  %val = load i32* %arg, !range !0
+  %val = load i32, i32* %arg, !range !0
   %rval = icmp eq i32 %val, 0
   ret i1 %rval
 }
@@ -23,7 +23,7 @@ define i1 @test_nonzero3(i32* nocapture readonly %arg) {
 ; CHECK-LABEL: test_nonzero3
 ; Check that this does not trigger - it wouldn't be legal
 ; CHECK: icmp
-  %val = load i32* %arg, !range !1
+  %val = load i32, i32* %arg, !range !1
   %rval = icmp ne i32 %val, 0
   ret i1 %rval
 }
@@ -32,7 +32,7 @@ define i1 @test_nonzero3(i32* nocapture readonly %arg) {
 define i1 @test_nonzero4(i8* nocapture readonly %arg) {
 ; CHECK-LABEL: test_nonzero4
 ; CHECK: ret i1 false
-  %val = load i8* %arg, !range !2
+  %val = load i8, i8* %arg, !range !2
   %rval = icmp ne i8 %val, 0
   ret i1 %rval
 }
@@ -40,7 +40,7 @@ define i1 @test_nonzero4(i8* nocapture readonly %arg) {
 define i1 @test_nonzero5(i8* nocapture readonly %arg) {
 ; CHECK-LABEL: test_nonzero5
 ; CHECK: ret i1 false
-  %val = load i8* %arg, !range !2
+  %val = load i8, i8* %arg, !range !2
   %rval = icmp ugt i8 %val, 0
   ret i1 %rval
 }
@@ -49,7 +49,7 @@ define i1 @test_nonzero5(i8* nocapture readonly %arg) {
 define i1 @test_nonzero6(i8* %argw) {
 ; CHECK-LABEL: test_nonzero6
 ; CHECK: icmp ne i8 %val, 0
-  %val = load i8* %argw, !range !3
+  %val = load i8, i8* %argw, !range !3
   %rval = icmp sgt i8 %val, 0
   ret i1 %rval
 }
diff --git a/test/Transforms/InstCombine/icmp.ll b/test/Transforms/InstCombine/icmp.ll
index 64741c5712eb..edcf76d5a7d2 100644
--- a/test/Transforms/InstCombine/icmp.ll
+++ b/test/Transforms/InstCombine/icmp.ll
@@ -244,8 +244,8 @@ define i1 @test23(i32 %x) nounwind {
 ; CHECK:    %cmp = icmp eq i64 %i, 1000
 ; CHECK:   ret i1 %cmp
 define i1 @test24(i64 %i) {
-  %p1 = getelementptr inbounds i32* getelementptr inbounds ([1000 x i32]* @X, i64 0, i64 0), i64 %i
-  %cmp = icmp eq i32* %p1, getelementptr inbounds ([1000 x i32]* @X, i64 1, i64 0)
+  %p1 = getelementptr inbounds i32, i32* getelementptr inbounds ([1000 x i32], [1000 x i32]* @X, i64 0, i64 0), i64 %i
+  %cmp = icmp eq i32* %p1, getelementptr inbounds ([1000 x i32], [1000 x i32]* @X, i64 1, i64 0)
   ret i1 %cmp
 }
 
@@ -256,8 +256,8 @@ define i1 @test24(i64 %i) {
 ; CHECK: %cmp = icmp eq i16 %1, 1000
 ; CHECK: ret i1 %cmp
 define i1 @test24_as1(i64 %i) {
-  %p1 = getelementptr inbounds i32 addrspace(1)* getelementptr inbounds ([1000 x i32] addrspace(1)* @X_as1, i64 0, i64 0), i64 %i
-  %cmp = icmp eq i32 addrspace(1)* %p1, getelementptr inbounds ([1000 x i32] addrspace(1)* @X_as1, i64 1, i64 0)
+  %p1 = getelementptr inbounds i32, i32 addrspace(1)* getelementptr inbounds ([1000 x i32], [1000 x i32] addrspace(1)* @X_as1, i64 0, i64 0), i64 %i
+  %cmp = icmp eq i32 addrspace(1)* %p1, getelementptr inbounds ([1000 x i32], [1000 x i32] addrspace(1)* @X_as1, i64 1, i64 0)
   ret i1 %cmp
 }
 
@@ -619,8 +619,8 @@ declare i32 @test58_d(i64)
 
 define i1 @test59(i8* %foo) {
   %bit = bitcast i8* %foo to i32*
-  %gep1 = getelementptr inbounds i32* %bit, i64 2
-  %gep2 = getelementptr inbounds i8* %foo, i64 10
+  %gep1 = getelementptr inbounds i32, i32* %bit, i64 2
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 10
   %cast1 = bitcast i32* %gep1 to i8*
   %cmp = icmp ult i8* %cast1, %gep2
   %use = ptrtoint i8* %cast1 to i64
@@ -632,23 +632,23 @@ define i1 @test59(i8* %foo) {
 
 define i1 @test59_as1(i8 addrspace(1)* %foo) {
   %bit = bitcast i8 addrspace(1)* %foo to i32 addrspace(1)*
-  %gep1 = getelementptr inbounds i32 addrspace(1)* %bit, i64 2
-  %gep2 = getelementptr inbounds i8 addrspace(1)* %foo, i64 10
+  %gep1 = getelementptr inbounds i32, i32 addrspace(1)* %bit, i64 2
+  %gep2 = getelementptr inbounds i8, i8 addrspace(1)* %foo, i64 10
   %cast1 = bitcast i32 addrspace(1)* %gep1 to i8 addrspace(1)*
   %cmp = icmp ult i8 addrspace(1)* %cast1, %gep2
   %use = ptrtoint i8 addrspace(1)* %cast1 to i64
   %call = call i32 @test58_d(i64 %use) nounwind
   ret i1 %cmp
 ; CHECK: @test59_as1
-; CHECK: %[[GEP:.+]] = getelementptr inbounds i8 addrspace(1)* %foo, i16 8
+; CHECK: %[[GEP:.+]] = getelementptr inbounds i8, i8 addrspace(1)* %foo, i16 8
 ; CHECK: ptrtoint i8 addrspace(1)* %[[GEP]] to i16
 ; CHECK: ret i1 true
 }
 
 define i1 @test60(i8* %foo, i64 %i, i64 %j) {
   %bit = bitcast i8* %foo to i32*
-  %gep1 = getelementptr inbounds i32* %bit, i64 %i
-  %gep2 = getelementptr inbounds i8* %foo, i64 %j
+  %gep1 = getelementptr inbounds i32, i32* %bit, i64 %i
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 %j
   %cast1 = bitcast i32* %gep1 to i8*
   %cmp = icmp ult i8* %cast1, %gep2
   ret i1 %cmp
@@ -660,8 +660,8 @@ define i1 @test60(i8* %foo, i64 %i, i64 %j) {
 
 define i1 @test60_as1(i8 addrspace(1)* %foo, i64 %i, i64 %j) {
   %bit = bitcast i8 addrspace(1)* %foo to i32 addrspace(1)*
-  %gep1 = getelementptr inbounds i32 addrspace(1)* %bit, i64 %i
-  %gep2 = getelementptr inbounds i8 addrspace(1)* %foo, i64 %j
+  %gep1 = getelementptr inbounds i32, i32 addrspace(1)* %bit, i64 %i
+  %gep2 = getelementptr inbounds i8, i8 addrspace(1)* %foo, i64 %j
   %cast1 = bitcast i32 addrspace(1)* %gep1 to i8 addrspace(1)*
   %cmp = icmp ult i8 addrspace(1)* %cast1, %gep2
   ret i1 %cmp
@@ -677,8 +677,8 @@ define i1 @test60_as1(i8 addrspace(1)* %foo, i64 %i, i64 %j) {
 ; bitcast. This uses the same sized addrspace.
 define i1 @test60_addrspacecast(i8* %foo, i64 %i, i64 %j) {
   %bit = addrspacecast i8* %foo to i32 addrspace(3)*
-  %gep1 = getelementptr inbounds i32 addrspace(3)* %bit, i64 %i
-  %gep2 = getelementptr inbounds i8* %foo, i64 %j
+  %gep1 = getelementptr inbounds i32, i32 addrspace(3)* %bit, i64 %i
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 %j
   %cast1 = addrspacecast i32 addrspace(3)* %gep1 to i8*
   %cmp = icmp ult i8* %cast1, %gep2
   ret i1 %cmp
@@ -690,8 +690,8 @@ define i1 @test60_addrspacecast(i8* %foo, i64 %i, i64 %j) {
 
 define i1 @test60_addrspacecast_smaller(i8* %foo, i16 %i, i64 %j) {
   %bit = addrspacecast i8* %foo to i32 addrspace(1)*
-  %gep1 = getelementptr inbounds i32 addrspace(1)* %bit, i16 %i
-  %gep2 = getelementptr inbounds i8* %foo, i64 %j
+  %gep1 = getelementptr inbounds i32, i32 addrspace(1)* %bit, i16 %i
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 %j
   %cast1 = addrspacecast i32 addrspace(1)* %gep1 to i8*
   %cmp = icmp ult i8* %cast1, %gep2
   ret i1 %cmp
@@ -704,8 +704,8 @@ define i1 @test60_addrspacecast_smaller(i8* %foo, i16 %i, i64 %j) {
 
 define i1 @test60_addrspacecast_larger(i8 addrspace(1)* %foo, i32 %i, i16 %j) {
   %bit = addrspacecast i8 addrspace(1)* %foo to i32 addrspace(2)*
-  %gep1 = getelementptr inbounds i32 addrspace(2)* %bit, i32 %i
-  %gep2 = getelementptr inbounds i8 addrspace(1)* %foo, i16 %j
+  %gep1 = getelementptr inbounds i32, i32 addrspace(2)* %bit, i32 %i
+  %gep2 = getelementptr inbounds i8, i8 addrspace(1)* %foo, i16 %j
   %cast1 = addrspacecast i32 addrspace(2)* %gep1 to i8 addrspace(1)*
   %cmp = icmp ult i8 addrspace(1)* %cast1, %gep2
   ret i1 %cmp
@@ -718,8 +718,8 @@ define i1 @test60_addrspacecast_larger(i8 addrspace(1)* %foo, i32 %i, i16 %j) {
 
 define i1 @test61(i8* %foo, i64 %i, i64 %j) {
   %bit = bitcast i8* %foo to i32*
-  %gep1 = getelementptr i32* %bit, i64 %i
-  %gep2 = getelementptr  i8* %foo, i64 %j
+  %gep1 = getelementptr i32, i32* %bit, i64 %i
+  %gep2 = getelementptr  i8,  i8* %foo, i64 %j
   %cast1 = bitcast i32* %gep1 to i8*
   %cmp = icmp ult i8* %cast1, %gep2
   ret i1 %cmp
@@ -731,8 +731,8 @@ define i1 @test61(i8* %foo, i64 %i, i64 %j) {
 
 define i1 @test61_as1(i8 addrspace(1)* %foo, i16 %i, i16 %j) {
   %bit = bitcast i8 addrspace(1)* %foo to i32 addrspace(1)*
-  %gep1 = getelementptr i32 addrspace(1)* %bit, i16 %i
-  %gep2 = getelementptr i8 addrspace(1)* %foo, i16 %j
+  %gep1 = getelementptr i32, i32 addrspace(1)* %bit, i16 %i
+  %gep2 = getelementptr i8, i8 addrspace(1)* %foo, i16 %j
   %cast1 = bitcast i32 addrspace(1)* %gep1 to i8 addrspace(1)*
   %cmp = icmp ult i8 addrspace(1)* %cast1, %gep2
   ret i1 %cmp
@@ -743,8 +743,8 @@ define i1 @test61_as1(i8 addrspace(1)* %foo, i16 %i, i16 %j) {
 }
 
 define i1 @test62(i8* %a) {
-  %arrayidx1 = getelementptr inbounds i8* %a, i64 1
-  %arrayidx2 = getelementptr inbounds i8* %a, i64 10
+  %arrayidx1 = getelementptr inbounds i8, i8* %a, i64 1
+  %arrayidx2 = getelementptr inbounds i8, i8* %a, i64 10
   %cmp = icmp slt i8* %arrayidx1, %arrayidx2
   ret i1 %cmp
 ; CHECK-LABEL: @test62(
@@ -754,8 +754,8 @@ define i1 @test62(i8* %a) {
 define i1 @test62_as1(i8 addrspace(1)* %a) {
 ; CHECK-LABEL: @test62_as1(
 ; CHECK-NEXT: ret i1 true
-  %arrayidx1 = getelementptr inbounds i8 addrspace(1)* %a, i64 1
-  %arrayidx2 = getelementptr inbounds i8 addrspace(1)* %a, i64 10
+  %arrayidx1 = getelementptr inbounds i8, i8 addrspace(1)* %a, i64 1
+  %arrayidx2 = getelementptr inbounds i8, i8 addrspace(1)* %a, i64 10
   %cmp = icmp slt i8 addrspace(1)* %arrayidx1, %arrayidx2
   ret i1 %cmp
 }
@@ -1115,8 +1115,8 @@ define i1 @icmp_add_and_shr_ne_0(i32 %X) {
 ; CHECK-LABEL: define i1 @test71(
 ; CHECK-NEXT: ret i1 false
 define i1 @test71(i8* %x) {
-  %a = getelementptr i8* %x, i64 8
-  %b = getelementptr inbounds i8* %x, i64 8
+  %a = getelementptr i8, i8* %x, i64 8
+  %b = getelementptr inbounds i8, i8* %x, i64 8
   %c = icmp ugt i8* %a, %b
   ret i1 %c
 }
@@ -1124,8 +1124,8 @@ define i1 @test71(i8* %x) {
 define i1 @test71_as1(i8 addrspace(1)* %x) {
 ; CHECK-LABEL: @test71_as1(
 ; CHECK-NEXT: ret i1 false
-  %a = getelementptr i8 addrspace(1)* %x, i64 8
-  %b = getelementptr inbounds i8 addrspace(1)* %x, i64 8
+  %a = getelementptr i8, i8 addrspace(1)* %x, i64 8
+  %b = getelementptr inbounds i8, i8 addrspace(1)* %x, i64 8
   %c = icmp ugt i8 addrspace(1)* %a, %b
   ret i1 %c
 }
@@ -1573,3 +1573,33 @@ define i32 @f5(i8 %a, i8 %b) {
   %sub7.sub = select i1 %cmp4, i32 %sub7, i32 %sub
   ret i32 %sub7.sub
 }
+
+; CHECK-LABEL: @f6
+; CHECK: %cmp.unshifted = xor i32 %a, %b
+; CHECK-NEXT: %cmp.mask = and i32 %cmp.unshifted, 255
+; CHECK-NEXT: %cmp = icmp eq i32 %cmp.mask, 0
+; CHECK-NEXT: %s = select i1 %cmp, i32 10000, i32 0
+; CHECK-NEXT: ret i32 %s
+define i32 @f6(i32 %a, i32 %b) {
+  %sext = shl i32 %a, 24
+  %conv = ashr i32 %sext, 24
+  %sext6 = shl i32 %b, 24
+  %conv4 = ashr i32 %sext6, 24
+  %cmp = icmp eq i32 %conv, %conv4
+  %s = select i1 %cmp, i32 10000, i32 0
+  ret i32 %s
+}
+
+; CHECK-LABEL: @f7
+; CHECK:   %cmp.unshifted = xor i32 %a, %b
+; CHECK-NEXT: %cmp.mask = and i32 %cmp.unshifted, 511
+; CHECK-NEXT: %cmp = icmp ne i32 %cmp.mask, 0
+; CHECK-NEXT: %s = select i1 %cmp, i32 10000, i32 0
+; CHECK-NEXT: ret i32 %s
+define i32 @f7(i32 %a, i32 %b) {
+  %sext = shl i32 %a, 23
+  %sext6 = shl i32 %b, 23
+  %cmp = icmp ne i32 %sext, %sext6
+  %s = select i1 %cmp, i32 10000, i32 0
+  ret i32 %s
+}
diff --git a/test/Transforms/InstCombine/inline-intrinsic-assert.ll b/test/Transforms/InstCombine/inline-intrinsic-assert.ll
new file mode 100644
index 000000000000..af34277563e0
--- /dev/null
+++ b/test/Transforms/InstCombine/inline-intrinsic-assert.ll
@@ -0,0 +1,30 @@
+; RUN: opt < %s -inline -instcombine -S | FileCheck %s
+
+; PR22857: http://llvm.org/bugs/show_bug.cgi?id=22857
+; The inliner should not add an edge to an intrinsic and
+; then assert that it did not add an edge to an intrinsic!
+
+define float @foo(float %f1) #0 {
+  %call = call float @bar(float %f1)
+  ret float %call
+
+; CHECK-LABEL: @foo(
+; CHECK-NEXT: call float @llvm.fabs.f32
+; CHECK-NEXT: ret float
+}
+
+define float @bar(float %f1) #0 {
+  %call = call float @sqr(float %f1)
+  %call1 = call float @sqrtf(float %call) #0
+  ret float %call1
+}
+
+define float @sqr(float %f) #0 {
+  %mul = fmul fast float %f, %f
+  ret float %mul
+}
+
+declare float @sqrtf(float) #0
+
+attributes #0 = { "unsafe-fp-math"="true" }
+
diff --git a/test/Transforms/InstCombine/intrinsics.ll b/test/Transforms/InstCombine/intrinsics.ll
index 8e7742f8c34a..9767704c85cf 100644
--- a/test/Transforms/InstCombine/intrinsics.ll
+++ b/test/Transforms/InstCombine/intrinsics.ll
@@ -349,7 +349,8 @@ define i32 @ctlz_select(i32 %Value) nounwind {
   ret i32 %s
 
 ; CHECK-LABEL: @ctlz_select(
-; CHECK: select i1 %tobool, i32 %ctlz, i32 32
+; CHECK-NEXT: call i32 @llvm.ctlz.i32(i32 %Value, i1 false)
+; CHECK-NEXT: ret i32
 }
 
 define i32 @cttz_select(i32 %Value) nounwind {
@@ -359,5 +360,68 @@ define i32 @cttz_select(i32 %Value) nounwind {
   ret i32 %s
 
 ; CHECK-LABEL: @cttz_select(
-; CHECK: select i1 %tobool, i32 %cttz, i32 32
+; CHECK-NEXT: call i32 @llvm.cttz.i32(i32 %Value, i1 false)
+; CHECK-NEXT: ret i32
+}
+
+; CHECK-LABEL: @overflow_div_add(
+; CHECK: ret i1 false
+define i1 @overflow_div_add(i32 %v1, i32 %v2) nounwind {
+entry:
+  %div = sdiv i32 %v1, 2
+  %t = call %ov.result.32 @llvm.sadd.with.overflow.i32(i32 %div, i32 1)
+  %obit = extractvalue %ov.result.32 %t, 1
+  ret i1 %obit
+}
+
+; CHECK-LABEL: @overflow_div_sub(
+; CHECK: ret i1 false
+define i1 @overflow_div_sub(i32 %v1, i32 %v2) nounwind {
+entry:
+  ; Check cases where the known sign bits are larger than the word size.
+  %a = ashr i32 %v1, 18
+  %div = sdiv i32 %a, 65536
+  %t = call %ov.result.32 @llvm.ssub.with.overflow.i32(i32 %div, i32 1)
+  %obit = extractvalue %ov.result.32 %t, 1
+  ret i1 %obit
+}
+
+; CHECK-LABEL: @overflow_mod_mul(
+; CHECK: ret i1 false
+define i1 @overflow_mod_mul(i32 %v1, i32 %v2) nounwind {
+entry:
+  %rem = srem i32 %v1, 1000
+  %t = call %ov.result.32 @llvm.smul.with.overflow.i32(i32 %rem, i32 %rem)
+  %obit = extractvalue %ov.result.32 %t, 1
+  ret i1 %obit
+}
+
+; CHECK-LABEL: @overflow_mod_overflow_mul(
+; CHECK-NOT: ret i1 false
+define i1 @overflow_mod_overflow_mul(i32 %v1, i32 %v2) nounwind {
+entry:
+  %rem = srem i32 %v1, 65537
+  ; This may overflow because the result of the mul operands may be greater than 16bits
+  ; and the result greater than 32.
+  %t = call %ov.result.32 @llvm.smul.with.overflow.i32(i32 %rem, i32 %rem)
+  %obit = extractvalue %ov.result.32 %t, 1
+  ret i1 %obit
+}
+
+define %ov.result.32 @ssubtest_reorder(i8 %a) {
+  %A = sext i8 %a to i32
+  %x = call %ov.result.32 @llvm.ssub.with.overflow.i32(i32 0, i32 %A)
+  ret %ov.result.32 %x
+; CHECK-LABEL: @ssubtest_reorder
+; CHECK: %x = sub nsw i32 0, %A
+; CHECK-NEXT: %1 = insertvalue %ov.result.32 { i32 undef, i1 false }, i32 %x, 0
+; CHECK-NEXT:  ret %ov.result.32 %1
+}
+
+define %ov.result.32 @never_overflows_ssub_test0(i32 %a) {
+  %x = call %ov.result.32 @llvm.ssub.with.overflow.i32(i32 %a, i32 0)
+  ret %ov.result.32 %x
+; CHECK-LABEL: @never_overflows_ssub_test0
+; CHECK-NEXT: %[[x:.*]] = insertvalue %ov.result.32 { i32 undef, i1 false }, i32 %a, 0
+; CHECK-NEXT:  ret %ov.result.32 %[[x]]
 }
diff --git a/test/Transforms/InstCombine/invariant.ll b/test/Transforms/InstCombine/invariant.ll
index 383238022692..246f9e640417 100644
--- a/test/Transforms/InstCombine/invariant.ll
+++ b/test/Transforms/InstCombine/invariant.ll
@@ -11,6 +11,6 @@ define i8 @f() {
   %i = call {}* @llvm.invariant.start(i64 1, i8* %a) ; <{}*> [#uses=0]
   ; CHECK: call {}* @llvm.invariant.start
   call void @g(i8* %a)
-  %r = load i8* %a                                ; <i8> [#uses=1]
+  %r = load i8, i8* %a                                ; <i8> [#uses=1]
   ret i8 %r
 }
diff --git a/test/Transforms/InstCombine/known_align.ll b/test/Transforms/InstCombine/known_align.ll
index 0249951966dc..653c4c51e2da 100644
--- a/test/Transforms/InstCombine/known_align.ll
+++ b/test/Transforms/InstCombine/known_align.ll
@@ -11,17 +11,17 @@ entry:
 	%tmp = alloca i32, align 4		; <i32*> [#uses=2]
 	%tmp1 = alloca i32, align 4		; <i32*> [#uses=3]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%tmp3 = load i32* getelementptr (%struct.p* @t, i32 0, i32 1), align 1		; <i32> [#uses=1]
+	%tmp3 = load i32, i32* getelementptr (%struct.p, %struct.p* @t, i32 0, i32 1), align 1		; <i32> [#uses=1]
 	store i32 %tmp3, i32* %tmp1, align 4
-	%tmp5 = load i32* %tmp1, align 4		; <i32> [#uses=1]
-	store i32 %tmp5, i32* getelementptr (%struct.p* @u, i32 0, i32 1), align 1
-	%tmp6 = load i32* %tmp1, align 4		; <i32> [#uses=1]
+	%tmp5 = load i32, i32* %tmp1, align 4		; <i32> [#uses=1]
+	store i32 %tmp5, i32* getelementptr (%struct.p, %struct.p* @u, i32 0, i32 1), align 1
+	%tmp6 = load i32, i32* %tmp1, align 4		; <i32> [#uses=1]
 	store i32 %tmp6, i32* %tmp, align 4
-	%tmp7 = load i32* %tmp, align 4		; <i32> [#uses=1]
+	%tmp7 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
 	store i32 %tmp7, i32* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval8 = load i32* %retval		; <i32> [#uses=1]
+	%retval8 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval8
 }
diff --git a/test/Transforms/InstCombine/load-cmp.ll b/test/Transforms/InstCombine/load-cmp.ll
index 98100263b48c..b0bfdc4c4c54 100644
--- a/test/Transforms/InstCombine/load-cmp.ll
+++ b/test/Transforms/InstCombine/load-cmp.ll
@@ -1,5 +1,4 @@
-; RUN: opt -instcombine -S < %s | FileCheck -check-prefix=NODL %s
-; RUN: opt -instcombine -S -default-data-layout="p:32:32:32-p1:16:16:16-n8:16:32:64" < %s | FileCheck -check-prefix=P32 %s
+; RUN: opt -instcombine -S -default-data-layout="p:32:32:32-p1:16:16:16-n8:16:32:64" < %s | FileCheck %s
 
 @G16 = internal constant [10 x i16] [i16 35, i16 82, i16 69, i16 81, i16 85,
                                      i16 73, i16 82, i16 69, i16 68, i16 0]
@@ -21,185 +20,137 @@
 
 
 define i1 @test1(i32 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = icmp eq i16 %Q, 0
   ret i1 %R
-; NODL-LABEL: @test1(
-; NODL-NEXT: %R = icmp eq i32 %X, 9
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test1(
-; P32-NEXT: %R = icmp eq i32 %X, 9
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test1(
+; CHECK-NEXT: %R = icmp eq i32 %X, 9
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test1_noinbounds(i32 %X) {
-  %P = getelementptr [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = icmp eq i16 %Q, 0
   ret i1 %R
-; NODL-LABEL: @test1_noinbounds(
-; NODL-NEXT: %P = getelementptr [10 x i16]* @G16, i32 0, i32 %X
-
-; P32-LABEL: @test1_noinbounds(
-; P32-NEXT: %R = icmp eq i32 %X, 9
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test1_noinbounds(
+; CHECK-NEXT: %R = icmp eq i32 %X, 9
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test1_noinbounds_i64(i64 %X) {
-  %P = getelementptr [10 x i16]* @G16, i64 0, i64 %X
-  %Q = load i16* %P
+  %P = getelementptr [10 x i16], [10 x i16]* @G16, i64 0, i64 %X
+  %Q = load i16, i16* %P
   %R = icmp eq i16 %Q, 0
   ret i1 %R
-; NODL-LABEL: @test1_noinbounds_i64(
-; NODL-NEXT: %P = getelementptr [10 x i16]* @G16, i64 0, i64 %X
-
-; P32-LABEL: @test1_noinbounds_i64(
-; P32: %R = icmp eq i32 %1, 9
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test1_noinbounds_i64(
+; CHECK: %R = icmp eq i32 %1, 9
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test1_noinbounds_as1(i32 %x) {
-  %p = getelementptr [10 x i16] addrspace(1)* @G16_as1, i16 0, i32 %x
-  %q = load i16 addrspace(1)* %p
+  %p = getelementptr [10 x i16], [10 x i16] addrspace(1)* @G16_as1, i16 0, i32 %x
+  %q = load i16, i16 addrspace(1)* %p
   %r = icmp eq i16 %q, 0
   ret i1 %r
 
-; P32-LABEL: @test1_noinbounds_as1(
-; P32-NEXT: trunc i32 %x to i16
-; P32-NEXT: %r = icmp eq i16 %1, 9
-; P32-NEXT: ret i1 %r
+; CHECK-LABEL: @test1_noinbounds_as1(
+; CHECK-NEXT: trunc i32 %x to i16
+; CHECK-NEXT: %r = icmp eq i16 %1, 9
+; CHECK-NEXT: ret i1 %r
 }
 
 define i1 @test2(i32 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = icmp slt i16 %Q, 85
   ret i1 %R
-; NODL-LABEL: @test2(
-; NODL-NEXT: %R = icmp ne i32 %X, 4
-; NODL-NEXT: ret i1 %R
+; CHECK-LABEL: @test2(
+; CHECK-NEXT:  %R = icmp ne i32 %X, 4
+; CHECK-NEXT:  ret i1 %R
 }
 
 define i1 @test3(i32 %X) {
-  %P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
-  %Q = load double* %P
+  %P = getelementptr inbounds [6 x double], [6 x double]* @GD, i32 0, i32 %X
+  %Q = load double, double* %P
   %R = fcmp oeq double %Q, 1.0
   ret i1 %R
-; NODL-LABEL: @test3(
-; NODL-NEXT: %R = icmp eq i32 %X, 1
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test3(
-; P32-NEXT: %R = icmp eq i32 %X, 1
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test3(
+; CHECK-NEXT: %R = icmp eq i32 %X, 1
+; CHECK-NEXT: ret i1 %R
 
 }
 
 define i1 @test4(i32 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = icmp sle i16 %Q, 73
   ret i1 %R
-; NODL-LABEL: @test4(
-; NODL-NEXT: lshr i32 933, %X
-; NODL-NEXT: and i32 {{.*}}, 1
-; NODL-NEXT: %R = icmp ne i32 {{.*}}, 0
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test4(
-; P32-NEXT: lshr i32 933, %X
-; P32-NEXT: and i32 {{.*}}, 1
-; P32-NEXT: %R = icmp ne i32 {{.*}}, 0
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test4(
+; CHECK-NEXT: lshr i32 933, %X
+; CHECK-NEXT: and i32 {{.*}}, 1
+; CHECK-NEXT: %R = icmp ne i32 {{.*}}, 0
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test4_i16(i16 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i16 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i16 %X
+  %Q = load i16, i16* %P
   %R = icmp sle i16 %Q, 73
   ret i1 %R
-
-; NODL-LABEL: @test4_i16(
-; NODL-NEXT: lshr i16 933, %X
-; NODL-NEXT: and i16 {{.*}}, 1
-; NODL-NEXT: %R = icmp ne i16 {{.*}}, 0
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test4_i16(
-; P32-NEXT: sext i16 %X to i32
-; P32-NEXT: lshr i32 933, %1
-; P32-NEXT: and i32 {{.*}}, 1
-; P32-NEXT: %R = icmp ne i32 {{.*}}, 0
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test4_i16(
+; CHECK-NEXT: sext i16 %X to i32
+; CHECK-NEXT: lshr i32 933, %1
+; CHECK-NEXT: and i32 {{.*}}, 1
+; CHECK-NEXT: %R = icmp ne i32 {{.*}}, 0
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test5(i32 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = icmp eq i16 %Q, 69
   ret i1 %R
-; NODL-LABEL: @test5(
-; NODL-NEXT: icmp eq i32 %X, 2
-; NODL-NEXT: icmp eq i32 %X, 7
-; NODL-NEXT: %R = or i1
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test5(
-; P32-NEXT: icmp eq i32 %X, 2
-; P32-NEXT: icmp eq i32 %X, 7
-; P32-NEXT: %R = or i1
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test5(
+; CHECK-NEXT: icmp eq i32 %X, 2
+; CHECK-NEXT: icmp eq i32 %X, 7
+; CHECK-NEXT: %R = or i1
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test6(i32 %X) {
-  %P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
-  %Q = load double* %P
+  %P = getelementptr inbounds [6 x double], [6 x double]* @GD, i32 0, i32 %X
+  %Q = load double, double* %P
   %R = fcmp ogt double %Q, 0.0
   ret i1 %R
-; NODL-LABEL: @test6(
-; NODL-NEXT: add i32 %X, -1
-; NODL-NEXT: %R = icmp ult i32 {{.*}}, 3
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test6(
-; P32-NEXT: add i32 %X, -1
-; P32-NEXT: %R = icmp ult i32 {{.*}}, 3
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test6(
+; CHECK-NEXT: add i32 %X, -1
+; CHECK-NEXT: %R = icmp ult i32 {{.*}}, 3
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test7(i32 %X) {
-  %P = getelementptr inbounds [6 x double]* @GD, i32 0, i32 %X
-  %Q = load double* %P
+  %P = getelementptr inbounds [6 x double], [6 x double]* @GD, i32 0, i32 %X
+  %Q = load double, double* %P
   %R = fcmp olt double %Q, 0.0
   ret i1 %R
-; NODL-LABEL: @test7(
-; NODL-NEXT: add i32 %X, -1
-; NODL-NEXT: %R = icmp ugt i32 {{.*}}, 2
-; NODL-NEXT: ret i1 %R
-
-; P32-LABEL: @test7(
-; P32-NEXT: add i32 %X, -1
-; P32-NEXT: %R = icmp ugt i32 {{.*}}, 2
-; P32-NEXT: ret i1 %R
+; CHECK-LABEL: @test7(
+; CHECK-NEXT: add i32 %X, -1
+; CHECK-NEXT: %R = icmp ugt i32 {{.*}}, 2
+; CHECK-NEXT: ret i1 %R
 }
 
 define i1 @test8(i32 %X) {
-  %P = getelementptr inbounds [10 x i16]* @G16, i32 0, i32 %X
-  %Q = load i16* %P
+  %P = getelementptr inbounds [10 x i16], [10 x i16]* @G16, i32 0, i32 %X
+  %Q = load i16, i16* %P
   %R = and i16 %Q, 3
   %S = icmp eq i16 %R, 0
   ret i1 %S
-; NODL-LABEL: @test8(
-; NODL-NEXT: and i32 %X, -2
-; NODL-NEXT: icmp eq i32 {{.*}}, 8
-; NODL-NEXT: ret i1
-
-; P32-LABEL: @test8(
-; P32-NEXT: and i32 %X, -2
-; P32-NEXT: icmp eq i32 {{.*}}, 8
-; P32-NEXT: ret i1
+; CHECK-LABEL: @test8(
+; CHECK-NEXT: and i32 %X, -2
+; CHECK-NEXT: icmp eq i32 {{.*}}, 8
+; CHECK-NEXT: ret i1
 }
 
 @GA = internal constant [4 x { i32, i32 } ] [
@@ -210,41 +161,30 @@ define i1 @test8(i32 %X) {
 ]
 
 define i1 @test9(i32 %X) {
-  %P = getelementptr inbounds [4 x { i32, i32 } ]* @GA, i32 0, i32 %X, i32 1
-  %Q = load i32* %P
+  %P = getelementptr inbounds [4 x { i32, i32 } ], [4 x { i32, i32 } ]* @GA, i32 0, i32 %X, i32 1
+  %Q = load i32, i32* %P
   %R = icmp eq i32 %Q, 1
   ret i1 %R
-; NODL-LABEL: @test9(
-; NODL-NEXT: add i32 %X, -1
-; NODL-NEXT: icmp ult i32 {{.*}}, 2
-; NODL-NEXT: ret i1
-
-; P32-LABEL: @test9(
-; P32-NEXT: add i32 %X, -1
-; P32-NEXT: icmp ult i32 {{.*}}, 2
-; P32-NEXT: ret i1
+; CHECK-LABEL: @test9(
+; CHECK-NEXT: add i32 %X, -1
+; CHECK-NEXT: icmp ult i32 {{.*}}, 2
+; CHECK-NEXT: ret i1
 }
 
 define i1 @test10_struct(i32 %x) {
-; NODL-LABEL: @test10_struct(
-; NODL: getelementptr inbounds %Foo* @GS, i32 %x, i32 0
-
-; P32-LABEL: @test10_struct(
-; P32: getelementptr inbounds %Foo* @GS, i32 %x, i32 0
-  %p = getelementptr inbounds %Foo* @GS, i32 %x, i32 0
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct(
+; CHECK: ret i1 false
+  %p = getelementptr inbounds %Foo, %Foo* @GS, i32 %x, i32 0
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_noinbounds(i32 %x) {
-; NODL-LABEL: @test10_struct_noinbounds(
-; NODL: getelementptr %Foo* @GS, i32 %x, i32 0
-
-; P32-LABEL: @test10_struct_noinbounds(
-; P32: getelementptr %Foo* @GS, i32 %x, i32 0
-  %p = getelementptr %Foo* @GS, i32 %x, i32 0
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_noinbounds(
+; CHECK: getelementptr %Foo, %Foo* @GS, i32 %x, i32 0
+  %p = getelementptr %Foo, %Foo* @GS, i32 %x, i32 0
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
@@ -252,14 +192,10 @@ define i1 @test10_struct_noinbounds(i32 %x) {
 ; Test that the GEP indices are converted before we ever get here
 ; Index < ptr size
 define i1 @test10_struct_i16(i16 %x){
-; NODL-LABEL: @test10_struct_i16(
-; NODL: getelementptr inbounds %Foo* @GS, i16 %x, i32 0
-
-; P32-LABEL: @test10_struct_i16(
-; P32: %1 = sext i16 %x to i32
-; P32: getelementptr inbounds %Foo* @GS, i32 %1, i32 0
-  %p = getelementptr inbounds %Foo* @GS, i16 %x, i32 0
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_i16(
+; CHECK: ret i1 false
+  %p = getelementptr inbounds %Foo, %Foo* @GS, i16 %x, i32 0
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 0
   ret i1 %r
 }
@@ -267,108 +203,79 @@ define i1 @test10_struct_i16(i16 %x){
 ; Test that the GEP indices are converted before we ever get here
 ; Index > ptr size
 define i1 @test10_struct_i64(i64 %x){
-; NODL-LABEL: @test10_struct_i64(
-; NODL: getelementptr inbounds %Foo* @GS, i64 %x, i32 0
-
-; P32-LABEL: @test10_struct_i64(
-; P32: %1 = trunc i64 %x to i32
-; P32: getelementptr inbounds %Foo* @GS, i32 %1, i32 0
-  %p = getelementptr inbounds %Foo* @GS, i64 %x, i32 0
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_i64(
+; CHECK: ret i1 false
+  %p = getelementptr inbounds %Foo, %Foo* @GS, i64 %x, i32 0
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 0
   ret i1 %r
 }
 
 define i1 @test10_struct_noinbounds_i16(i16 %x) {
-; NODL-LABEL: @test10_struct_noinbounds_i16(
-; NODL: getelementptr %Foo* @GS, i16 %x, i32 0
-
-; P32-LABEL: @test10_struct_noinbounds_i16(
-; P32: %1 = sext i16 %x to i32
-; P32: getelementptr %Foo* @GS, i32 %1, i32 0
-  %p = getelementptr %Foo* @GS, i16 %x, i32 0
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_noinbounds_i16(
+; CHECK: %1 = sext i16 %x to i32
+; CHECK: getelementptr %Foo, %Foo* @GS, i32 %1, i32 0
+  %p = getelementptr %Foo, %Foo* @GS, i16 %x, i32 0
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 0
   ret i1 %r
 }
 
 define i1 @test10_struct_arr(i32 %x) {
-; NODL-LABEL: @test10_struct_arr(
-; NODL-NEXT: %r = icmp ne i32 %x, 1
-; NODL-NEXT: ret i1 %r
-
-; P32-LABEL: @test10_struct_arr(
-; P32-NEXT: %r = icmp ne i32 %x, 1
-; P32-NEXT: ret i1 %r
-  %p = getelementptr inbounds [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr(
+; CHECK-NEXT: %r = icmp ne i32 %x, 1
+; CHECK-NEXT: ret i1 %r
+  %p = getelementptr inbounds [4 x %Foo], [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_arr_noinbounds(i32 %x) {
-; NODL-LABEL: @test10_struct_arr_noinbounds(
-; NODL-NEXT  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
-
-; P32-LABEL: @test10_struct_arr_noinbounds(
-; P32-NEXT  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
-  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr_noinbounds(
+; CHECK-NEXT  %p = getelementptr [4 x %Foo], [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
+  %p = getelementptr [4 x %Foo], [4 x %Foo]* @GStructArr, i32 0, i32 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_arr_i16(i16 %x) {
-; NODL-LABEL: @test10_struct_arr_i16(
-; NODL-NEXT: %r = icmp ne i16 %x, 1
-; NODL-NEXT: ret i1 %r
-
-; P32-LABEL: @test10_struct_arr_i16(
-; P32-NEXT: %r = icmp ne i16 %x, 1
-; P32-NEXT: ret i1 %r
-  %p = getelementptr inbounds [4 x %Foo]* @GStructArr, i16 0, i16 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr_i16(
+; CHECK-NEXT: %r = icmp ne i16 %x, 1
+; CHECK-NEXT: ret i1 %r
+  %p = getelementptr inbounds [4 x %Foo], [4 x %Foo]* @GStructArr, i16 0, i16 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_arr_i64(i64 %x) {
-; NODL-LABEL: @test10_struct_arr_i64(
-; NODL-NEXT: %r = icmp ne i64 %x, 1
-; NODL-NEXT: ret i1 %r
-
-; P32-LABEL: @test10_struct_arr_i64(
-; P32-NEXT: trunc i64 %x to i32
-; P32-NEXT: %r = icmp ne i32 %1, 1
-; P32-NEXT: ret i1 %r
-  %p = getelementptr inbounds [4 x %Foo]* @GStructArr, i64 0, i64 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr_i64(
+; CHECK-NEXT: trunc i64 %x to i32
+; CHECK-NEXT: %r = icmp ne i32 %1, 1
+; CHECK-NEXT: ret i1 %r
+  %p = getelementptr inbounds [4 x %Foo], [4 x %Foo]* @GStructArr, i64 0, i64 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_arr_noinbounds_i16(i16 %x) {
-; NODL-LABEL: @test10_struct_arr_noinbounds_i16(
-; NODL-NEXT:  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i16 %x, i32 2
-
-; P32-LABEL: @test10_struct_arr_noinbounds_i16(
-; P32-NEXT: %r = icmp ne i16 %x, 1
-  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i16 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr_noinbounds_i16(
+; CHECK-NEXT: %r = icmp ne i16 %x, 1
+  %p = getelementptr [4 x %Foo], [4 x %Foo]* @GStructArr, i32 0, i16 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
 
 define i1 @test10_struct_arr_noinbounds_i64(i64 %x) {
-; FIXME: Should be no trunc?
-; NODL-LABEL: @test10_struct_arr_noinbounds_i64(
-; NODL-NEXT:  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i64 %x, i32 2
-
-; P32-LABEL: @test10_struct_arr_noinbounds_i64(
-; P32: %r = icmp ne i32 %1, 1
-; P32-NEXT: ret i1 %r
-  %p = getelementptr [4 x %Foo]* @GStructArr, i32 0, i64 %x, i32 2
-  %q = load i32* %p
+; CHECK-LABEL: @test10_struct_arr_noinbounds_i64(
+; CHECK: %r = icmp ne i32 %1, 1
+; CHECK-NEXT: ret i1 %r
+  %p = getelementptr [4 x %Foo], [4 x %Foo]* @GStructArr, i32 0, i64 %x, i32 2
+  %q = load i32, i32* %p
   %r = icmp eq i32 %q, 9
   ret i1 %r
 }
diff --git a/test/Transforms/InstCombine/load-select.ll b/test/Transforms/InstCombine/load-select.ll
index e8cbad335dea..dfc07983eff4 100644
--- a/test/Transforms/InstCombine/load-select.ll
+++ b/test/Transforms/InstCombine/load-select.ll
@@ -10,7 +10,7 @@ define i32 @b(i32 %y) nounwind readonly {
 ; CHECK: ret i32
 entry:
   %0 = icmp eq i32 %y, 0                          ; <i1> [#uses=1]
-  %storemerge = select i1 %0, i32* getelementptr inbounds ([2 x i32]* @a, i32 0, i32 1), i32* getelementptr inbounds ([2 x i32]* @a, i32 0, i32 0) ; <i32*> [#uses=1]
-  %1 = load i32* %storemerge, align 4             ; <i32> [#uses=1]
+  %storemerge = select i1 %0, i32* getelementptr inbounds ([2 x i32], [2 x i32]* @a, i32 0, i32 1), i32* getelementptr inbounds ([2 x i32], [2 x i32]* @a, i32 0, i32 0) ; <i32*> [#uses=1]
+  %1 = load i32, i32* %storemerge, align 4             ; <i32> [#uses=1]
   ret i32 %1
 }
diff --git a/test/Transforms/InstCombine/load.ll b/test/Transforms/InstCombine/load.ll
index b4b7558a1a05..2cc55e99fd20 100644
--- a/test/Transforms/InstCombine/load.ll
+++ b/test/Transforms/InstCombine/load.ll
@@ -1,8 +1,9 @@
 ; RUN: opt -instcombine -S < %s | FileCheck %s
+; RUN: opt -passes=instcombine -S < %s | FileCheck %s
 
 ; This test makes sure that these instructions are properly eliminated.
 
-target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target datalayout = "e-m:e-p:64:64:64-i64:64-f80:128-n8:16:32:64-S128"
 
 @X = constant i32 42		; <i32*> [#uses=2]
 @X2 = constant i32 47		; <i32*> [#uses=1]
@@ -15,31 +16,31 @@ target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 ; CHECK-LABEL: @test1(
 ; CHECK-NOT: load
 define i32 @test1() {
-	%B = load i32* @X		; <i32> [#uses=1]
+	%B = load i32, i32* @X		; <i32> [#uses=1]
 	ret i32 %B
 }
 
 ; CHECK-LABEL: @test2(
 ; CHECK-NOT: load
 define float @test2() {
-	%A = getelementptr [2 x { i32, float }]* @Y, i64 0, i64 1, i32 1		; <float*> [#uses=1]
-	%B = load float* %A		; <float> [#uses=1]
+	%A = getelementptr [2 x { i32, float }], [2 x { i32, float }]* @Y, i64 0, i64 1, i32 1		; <float*> [#uses=1]
+	%B = load float, float* %A		; <float> [#uses=1]
 	ret float %B
 }
 
 ; CHECK-LABEL: @test3(
 ; CHECK-NOT: load
 define i32 @test3() {
-	%A = getelementptr [2 x { i32, float }]* @Y, i64 0, i64 0, i32 0		; <i32*> [#uses=1]
-	%B = load i32* %A		; <i32> [#uses=1]
+	%A = getelementptr [2 x { i32, float }], [2 x { i32, float }]* @Y, i64 0, i64 0, i32 0		; <i32*> [#uses=1]
+	%B = load i32, i32* %A		; <i32> [#uses=1]
 	ret i32 %B
 }
 
 ; CHECK-LABEL: @test4(
 ; CHECK-NOT: load
 define i32 @test4() {
-	%A = getelementptr [2 x { i32, float }]* @Z, i64 0, i64 1, i32 0		; <i32*> [#uses=1]
-	%B = load i32* %A		; <i32> [#uses=1]
+	%A = getelementptr [2 x { i32, float }], [2 x { i32, float }]* @Z, i64 0, i64 1, i32 0		; <i32*> [#uses=1]
+	%B = load i32, i32* %A		; <i32> [#uses=1]
 	ret i32 %B
 }
 
@@ -47,15 +48,15 @@ define i32 @test4() {
 ; CHECK-NOT: load
 define i32 @test5(i1 %C) {
 	%Y = select i1 %C, i32* @X, i32* @X2		; <i32*> [#uses=1]
-	%Z = load i32* %Y		; <i32> [#uses=1]
+	%Z = load i32, i32* %Y		; <i32> [#uses=1]
 	ret i32 %Z
 }
 
 ; CHECK-LABEL: @test7(
 ; CHECK-NOT: load
 define i32 @test7(i32 %X) {
-	%V = getelementptr i32* null, i32 %X		; <i32*> [#uses=1]
-	%R = load i32* %V		; <i32> [#uses=1]
+	%V = getelementptr i32, i32* null, i32 %X		; <i32*> [#uses=1]
+	%R = load i32, i32* %V		; <i32> [#uses=1]
 	ret i32 %R
 }
 
@@ -63,15 +64,15 @@ define i32 @test7(i32 %X) {
 ; CHECK-NOT: load
 define i32 @test8(i32* %P) {
 	store i32 1, i32* %P
-	%X = load i32* %P		; <i32> [#uses=1]
+	%X = load i32, i32* %P		; <i32> [#uses=1]
 	ret i32 %X
 }
 
 ; CHECK-LABEL: @test9(
 ; CHECK-NOT: load
 define i32 @test9(i32* %P) {
-	%X = load i32* %P		; <i32> [#uses=1]
-	%Y = load i32* %P		; <i32> [#uses=1]
+	%X = load i32, i32* %P		; <i32> [#uses=1]
+	%Y = load i32, i32* %P		; <i32> [#uses=1]
 	%Z = sub i32 %X, %Y		; <i32> [#uses=1]
 	ret i32 %Z
 }
@@ -88,17 +89,17 @@ F:		; preds = %0
 	store i32 0, i32* %P
 	br label %C
 C:		; preds = %F, %T
-	%V = load i32* %P		; <i32> [#uses=1]
+	%V = load i32, i32* %P		; <i32> [#uses=1]
 	ret i32 %V
 }
 
 ; CHECK-LABEL: @test11(
 ; CHECK-NOT: load
 define double @test11(double* %p) {
-  %t0 = getelementptr double* %p, i32 1
+  %t0 = getelementptr double, double* %p, i32 1
   store double 2.0, double* %t0
-  %t1 = getelementptr double* %p, i32 1
-  %x = load double* %t1
+  %t1 = getelementptr double, double* %p, i32 1
+  %x = load double, double* %t1
   ret double %x
 }
 
@@ -109,14 +110,14 @@ define i32 @test12(i32* %P) {
   store i32 123, i32* %A
   ; Cast the result of the load not the source
   %Q = bitcast i32* %A to i32*
-  %V = load i32* %Q
+  %V = load i32, i32* %Q
   ret i32 %V
 }
 
 ; CHECK-LABEL: @test13(
 ; CHECK-NOT: load
 define <16 x i8> @test13(<2 x i64> %x) {
-  %tmp = load <16 x i8>* bitcast ([4 x i32]* @GLOBAL to <16 x i8>*)
+  %tmp = load <16 x i8>, <16 x i8>* bitcast ([4 x i32]* @GLOBAL to <16 x i8>*)
   ret <16 x i8> %tmp
 }
 
@@ -127,13 +128,13 @@ define i8 @test14(i8 %x, i32 %y) {
 ; those confuse the analysis into thinking that the second store does not alias
 ; the first.
 ; CHECK-LABEL: @test14(
-; CHECK:         %[[R:.*]] = load i8*
+; CHECK:         %[[R:.*]] = load i8, i8*
 ; CHECK-NEXT:    ret i8 %[[R]]
   %a = alloca i32
   %a.i8 = bitcast i32* %a to i8*
   store i8 %x, i8* %a.i8
   store i32 %y, i32* %a
-  %r = load i8* %a.i8
+  %r = load i8, i8* %a.i8
   ret i8 %r
 }
 
@@ -142,11 +143,61 @@ define i8 @test14(i8 %x, i32 %y) {
 define i8 @test15(i8 %x, i32 %y) {
 ; Same test as @test14 essentially, but using a global instead of an alloca.
 ; CHECK-LABEL: @test15(
-; CHECK:         %[[R:.*]] = load i8*
+; CHECK:         %[[R:.*]] = load i8, i8*
 ; CHECK-NEXT:    ret i8 %[[R]]
   %g.i8 = bitcast i32* @test15_global to i8*
   store i8 %x, i8* %g.i8
   store i32 %y, i32* @test15_global
-  %r = load i8* %g.i8
+  %r = load i8, i8* %g.i8
   ret i8 %r
 }
+
+define void @test16(i8* %x, i8* %a, i8* %b, i8* %c) {
+; Check that we canonicalize loads which are only stored to use integer types
+; when there is a valid integer type.
+; CHECK-LABEL: @test16(
+; CHECK: %[[L1:.*]] = load i32, i32*
+; CHECK-NOT: load
+; CHECK: store i32 %[[L1]], i32*
+; CHECK: store i32 %[[L1]], i32*
+; CHECK-NOT: store
+; CHECK: %[[L1:.*]] = load i32, i32*
+; CHECK-NOT: load
+; CHECK: store i32 %[[L1]], i32*
+; CHECK: store i32 %[[L1]], i32*
+; CHECK-NOT: store
+; CHECK: ret
+
+entry:
+  %x.cast = bitcast i8* %x to float*
+  %a.cast = bitcast i8* %a to float*
+  %b.cast = bitcast i8* %b to float*
+  %c.cast = bitcast i8* %c to i32*
+
+  %x1 = load float, float* %x.cast
+  store float %x1, float* %a.cast
+  store float %x1, float* %b.cast
+
+  %x2 = load float, float* %x.cast
+  store float %x2, float* %b.cast
+  %x2.cast = bitcast float %x2 to i32
+  store i32 %x2.cast, i32* %c.cast
+
+  ret void
+}
+
+define void @test17(i8** %x, i8 %y) {
+; Check that in cases similar to @test16 we don't try to rewrite a load when
+; its only use is a store but it is used as the pointer to that store rather
+; than the value.
+;
+; CHECK-LABEL: @test17(
+; CHECK: %[[L:.*]] = load i8*, i8**
+; CHECK: store i8 %y, i8* %[[L]]
+
+entry:
+  %x.load = load i8*, i8** %x
+  store i8 %y, i8* %x.load
+
+  ret void
+}
diff --git a/test/Transforms/InstCombine/load3.ll b/test/Transforms/InstCombine/load3.ll
index f79ef9a695c2..6db8dd391fa1 100644
--- a/test/Transforms/InstCombine/load3.ll
+++ b/test/Transforms/InstCombine/load3.ll
@@ -5,10 +5,10 @@ target triple = "i386-apple-macosx10.0.0"
 ; Instcombine should be able to do trivial CSE of loads.
 
 define i32 @test1(i32* %p) {
-  %t0 = getelementptr i32* %p, i32 1
-  %y = load i32* %t0
-  %t1 = getelementptr i32* %p, i32 1
-  %x = load i32* %t1
+  %t0 = getelementptr i32, i32* %p, i32 1
+  %y = load i32, i32* %t0
+  %t1 = getelementptr i32, i32* %p, i32 1
+  %x = load i32, i32* %t1
   %a = sub i32 %y, %x
   ret i32 %a
 ; CHECK-LABEL: @test1(
@@ -19,7 +19,7 @@ define i32 @test1(i32* %p) {
 ; PR7429
 @.str = private constant [4 x i8] c"XYZ\00"
 define float @test2() {
-  %tmp = load float* bitcast ([4 x i8]* @.str to float*), align 1
+  %tmp = load float, float* bitcast ([4 x i8]* @.str to float*), align 1
   ret float %tmp
   
 ; CHECK-LABEL: @test2(
@@ -37,10 +37,10 @@ align 4
 ; PR14986
 define void @test3() nounwind {
 ; This is a weird way of computing zero.
-  %l = load i32* getelementptr ([36 x i32]* @expect32, i32 29826161, i32 28), align 4
-  store i32 %l, i32* getelementptr ([36 x i32]* @rslts32, i32 29826161, i32 28), align 4
+  %l = load i32, i32* getelementptr ([36 x i32], [36 x i32]* @expect32, i32 29826161, i32 28), align 4
+  store i32 %l, i32* getelementptr ([36 x i32], [36 x i32]* @rslts32, i32 29826161, i32 28), align 4
   ret void
 
 ; CHECK-LABEL: @test3(
-; CHECK: store i32 1, i32* getelementptr inbounds ([36 x i32]* @rslts32, i32 0, i32 0)
+; CHECK: store i32 1, i32* getelementptr inbounds ([36 x i32], [36 x i32]* @rslts32, i32 0, i32 0)
 }
diff --git a/test/Transforms/InstCombine/loadstore-alignment.ll b/test/Transforms/InstCombine/loadstore-alignment.ll
index e90bdb7cdd99..4afa82dcb4c5 100644
--- a/test/Transforms/InstCombine/loadstore-alignment.ll
+++ b/test/Transforms/InstCombine/loadstore-alignment.ll
@@ -9,58 +9,58 @@ target datalayout = "E-p:64:64:64-p1:64:64:64-p2:32:32:32-a0:0:8-f32:32:32-f64:6
 ; CHECK-LABEL: @static_hem(
 ; CHECK: , align 16
 define <2 x i64> @static_hem() {
-  %t = getelementptr <2 x i64>* @x, i32 7
-  %tmp1 = load <2 x i64>* %t, align 1
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 7
+  %tmp1 = load <2 x i64>, <2 x i64>* %t, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @static_hem_addrspacecast(
 ; CHECK: , align 16
 define <2 x i64> @static_hem_addrspacecast() {
-  %t = getelementptr <2 x i64>* @x, i32 7
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 7
   %t.asc = addrspacecast <2 x i64>* %t to <2 x i64> addrspace(1)*
-  %tmp1 = load <2 x i64> addrspace(1)* %t.asc, align 1
+  %tmp1 = load <2 x i64>, <2 x i64> addrspace(1)* %t.asc, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @static_hem_addrspacecast_smaller_ptr(
 ; CHECK: , align 16
 define <2 x i64> @static_hem_addrspacecast_smaller_ptr() {
-  %t = getelementptr <2 x i64>* @x, i32 7
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 7
   %t.asc = addrspacecast <2 x i64>* %t to <2 x i64> addrspace(2)*
-  %tmp1 = load <2 x i64> addrspace(2)* %t.asc, align 1
+  %tmp1 = load <2 x i64>, <2 x i64> addrspace(2)* %t.asc, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @static_hem_addrspacecast_larger_ptr(
 ; CHECK: , align 16
 define <2 x i64> @static_hem_addrspacecast_larger_ptr() {
-  %t = getelementptr <2 x i64> addrspace(2)* @x.as2, i32 7
+  %t = getelementptr <2 x i64>, <2 x i64> addrspace(2)* @x.as2, i32 7
   %t.asc = addrspacecast <2 x i64> addrspace(2)* %t to <2 x i64> addrspace(1)*
-  %tmp1 = load <2 x i64> addrspace(1)* %t.asc, align 1
+  %tmp1 = load <2 x i64>, <2 x i64> addrspace(1)* %t.asc, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @hem(
 ; CHECK: , align 16
 define <2 x i64> @hem(i32 %i) {
-  %t = getelementptr <2 x i64>* @x, i32 %i
-  %tmp1 = load <2 x i64>* %t, align 1
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 %i
+  %tmp1 = load <2 x i64>, <2 x i64>* %t, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @hem_2d(
 ; CHECK: , align 16
 define <2 x i64> @hem_2d(i32 %i, i32 %j) {
-  %t = getelementptr [13 x <2 x i64>]* @xx, i32 %i, i32 %j
-  %tmp1 = load <2 x i64>* %t, align 1
+  %t = getelementptr [13 x <2 x i64>], [13 x <2 x i64>]* @xx, i32 %i, i32 %j
+  %tmp1 = load <2 x i64>, <2 x i64>* %t, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @foo(
 ; CHECK: , align 16
 define <2 x i64> @foo() {
-  %tmp1 = load <2 x i64>* @x, align 1
+  %tmp1 = load <2 x i64>, <2 x i64>* @x, align 1
   ret <2 x i64> %tmp1
 }
 
@@ -70,14 +70,14 @@ define <2 x i64> @foo() {
 define <2 x i64> @bar() {
   %t = alloca <2 x i64>
   call void @kip(<2 x i64>* %t)
-  %tmp1 = load <2 x i64>* %t, align 1
+  %tmp1 = load <2 x i64>, <2 x i64>* %t, align 1
   ret <2 x i64> %tmp1
 }
 
 ; CHECK-LABEL: @static_hem_store(
 ; CHECK: , align 16
 define void @static_hem_store(<2 x i64> %y) {
-  %t = getelementptr <2 x i64>* @x, i32 7
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 7
   store <2 x i64> %y, <2 x i64>* %t, align 1
   ret void
 }
@@ -85,7 +85,7 @@ define void @static_hem_store(<2 x i64> %y) {
 ; CHECK-LABEL: @hem_store(
 ; CHECK: , align 16
 define void @hem_store(i32 %i, <2 x i64> %y) {
-  %t = getelementptr <2 x i64>* @x, i32 %i
+  %t = getelementptr <2 x i64>, <2 x i64>* @x, i32 %i
   store <2 x i64> %y, <2 x i64>* %t, align 1
   ret void
 }
@@ -93,7 +93,7 @@ define void @hem_store(i32 %i, <2 x i64> %y) {
 ; CHECK-LABEL: @hem_2d_store(
 ; CHECK: , align 16
 define void @hem_2d_store(i32 %i, i32 %j, <2 x i64> %y) {
-  %t = getelementptr [13 x <2 x i64>]* @xx, i32 %i, i32 %j
+  %t = getelementptr [13 x <2 x i64>], [13 x <2 x i64>]* @xx, i32 %i, i32 %j
   store <2 x i64> %y, <2 x i64>* %t, align 1
   ret void
 }
diff --git a/test/Transforms/InstCombine/loadstore-metadata.ll b/test/Transforms/InstCombine/loadstore-metadata.ll
index 3d18ac0e3344..a30c0bc852ea 100644
--- a/test/Transforms/InstCombine/loadstore-metadata.ll
+++ b/test/Transforms/InstCombine/loadstore-metadata.ll
@@ -5,9 +5,9 @@ target datalayout = "e-m:e-p:64:64:64-i64:64-f80:128-n8:16:32:64-S128"
 define i32 @test_load_cast_combine_tbaa(float* %ptr) {
 ; Ensure (cast (load (...))) -> (load (cast (...))) preserves TBAA.
 ; CHECK-LABEL: @test_load_cast_combine_tbaa(
-; CHECK: load i32* %{{.*}}, !tbaa !0
+; CHECK: load i32, i32* %{{.*}}, !tbaa !0
 entry:
-  %l = load float* %ptr, !tbaa !0
+  %l = load float, float* %ptr, !tbaa !0
   %c = bitcast float %l to i32
   ret i32 %c
 }
@@ -15,9 +15,9 @@ entry:
 define i32 @test_load_cast_combine_noalias(float* %ptr) {
 ; Ensure (cast (load (...))) -> (load (cast (...))) preserves no-alias metadata.
 ; CHECK-LABEL: @test_load_cast_combine_noalias(
-; CHECK: load i32* %{{.*}}, !alias.scope !2, !noalias !1
+; CHECK: load i32, i32* %{{.*}}, !alias.scope !2, !noalias !1
 entry:
-  %l = load float* %ptr, !alias.scope !2, !noalias !1
+  %l = load float, float* %ptr, !alias.scope !2, !noalias !1
   %c = bitcast float %l to i32
   ret i32 %c
 }
@@ -27,11 +27,11 @@ define float @test_load_cast_combine_range(i32* %ptr) {
 ; would be nice to preserve or update it somehow but this is hard when moving
 ; between types.
 ; CHECK-LABEL: @test_load_cast_combine_range(
-; CHECK: load float* %{{.*}}
+; CHECK: load float, float* %{{.*}}
 ; CHECK-NOT: !range
 ; CHECK: ret float
 entry:
-  %l = load i32* %ptr, !range !5
+  %l = load i32, i32* %ptr, !range !5
   %c = bitcast i32 %l to float
   ret float %c
 }
@@ -39,9 +39,9 @@ entry:
 define i32 @test_load_cast_combine_invariant(float* %ptr) {
 ; Ensure (cast (load (...))) -> (load (cast (...))) preserves invariant metadata.
 ; CHECK-LABEL: @test_load_cast_combine_invariant(
-; CHECK: load i32* %{{.*}}, !invariant.load !3
+; CHECK: load i32, i32* %{{.*}}, !invariant.load !3
 entry:
-  %l = load float* %ptr, !invariant.load !3
+  %l = load float, float* %ptr, !invariant.load !3
   %c = bitcast float %l to i32
   ret i32 %c
 }
@@ -50,9 +50,9 @@ define i32 @test_load_cast_combine_nontemporal(float* %ptr) {
 ; Ensure (cast (load (...))) -> (load (cast (...))) preserves nontemporal
 ; metadata.
 ; CHECK-LABEL: @test_load_cast_combine_nontemporal(
-; CHECK: load i32* %{{.*}}, !nontemporal !4
+; CHECK: load i32, i32* %{{.*}}, !nontemporal !4
 entry:
-  %l = load float* %ptr, !nontemporal !4
+  %l = load float, float* %ptr, !nontemporal !4
   %c = bitcast float %l to i32
   ret i32 %c
 }
@@ -61,15 +61,15 @@ define void @test_load_cast_combine_loop(float* %src, i32* %dst, i32 %n) {
 ; Ensure (cast (load (...))) -> (load (cast (...))) preserves loop access
 ; metadata.
 ; CHECK-LABEL: @test_load_cast_combine_loop(
-; CHECK: load i32* %{{.*}}, !llvm.mem.parallel_loop_access !1
+; CHECK: load i32, i32* %{{.*}}, !llvm.mem.parallel_loop_access !1
 entry:
   br label %loop
 
 loop:
   %i = phi i32 [ 0, %entry ], [ %i.next, %loop ]
-  %src.gep = getelementptr inbounds float* %src, i32 %i
-  %dst.gep = getelementptr inbounds i32* %dst, i32 %i
-  %l = load float* %src.gep, !llvm.mem.parallel_loop_access !1
+  %src.gep = getelementptr inbounds float, float* %src, i32 %i
+  %dst.gep = getelementptr inbounds i32, i32* %dst, i32 %i
+  %l = load float, float* %src.gep, !llvm.mem.parallel_loop_access !1
   %c = bitcast float %l to i32
   store i32 %c, i32* %dst.gep
   %i.next = add i32 %i, 1
@@ -80,6 +80,31 @@ exit:
   ret void
 }
 
+define void @test_load_cast_combine_nonnull(float** %ptr) {
+; We can't preserve nonnull metadata when converting a load of a pointer to
+; a load of an integer. Instead, we translate it to range metadata.
+; FIXME: We should also transform range metadata back into nonnull metadata.
+; FIXME: This test is very fragile. If any LABEL lines are added after
+; this point, the test will fail, because this test depends on a metadata tuple,
+; which is always emitted at the end of the file. At some point, we should
+; consider an option to the IR printer to emit MD tuples after the function
+; that first uses them--this will allow us to refer to them like this and not
+; have the tests break. For now, this function must always come last in this
+; file, and no LABEL lines are to be added after this point.
+;
+; CHECK-LABEL: @test_load_cast_combine_nonnull(
+; CHECK: %[[V:.*]] = load i64, i64* %{{.*}}, !range ![[MD:[0-9]+]]
+; CHECK-NOT: !nonnull
+; CHECK: store i64 %[[V]], i64*
+entry:
+  %p = load float*, float** %ptr, !nonnull !3
+  %gep = getelementptr float*, float** %ptr, i32 42
+  store float* %p, float** %gep
+  ret void
+}
+
+; This is the metadata tuple that we reference above:
+; CHECK: ![[MD]] = !{i64 1, i64 0}
 !0 = !{ !1, !1, i64 0 }
 !1 = !{ !1 }
 !2 = !{ !2, !1 }
diff --git a/test/Transforms/InstCombine/lshr-phi.ll b/test/Transforms/InstCombine/lshr-phi.ll
index 76a113face05..79dc8b33192d 100644
--- a/test/Transforms/InstCombine/lshr-phi.ll
+++ b/test/Transforms/InstCombine/lshr-phi.ll
@@ -8,24 +8,24 @@
 
 define i32 @hash_string(i8* nocapture %key) nounwind readonly {
 entry:
-	%t0 = load i8* %key, align 1		; <i8> [#uses=1]
+	%t0 = load i8, i8* %key, align 1		; <i8> [#uses=1]
 	%t1 = icmp eq i8 %t0, 0		; <i1> [#uses=1]
 	br i1 %t1, label %bb2, label %bb
 
 bb:		; preds = %bb, %entry
 	%indvar = phi i64 [ 0, %entry ], [ %tmp, %bb ]		; <i64> [#uses=2]
 	%k.04 = phi i32 [ 0, %entry ], [ %t8, %bb ]		; <i32> [#uses=2]
-	%cp.05 = getelementptr i8* %key, i64 %indvar		; <i8*> [#uses=1]
+	%cp.05 = getelementptr i8, i8* %key, i64 %indvar		; <i8*> [#uses=1]
 	%t2 = shl i32 %k.04, 1		; <i32> [#uses=1]
 	%t3 = lshr i32 %k.04, 14		; <i32> [#uses=1]
 	%t4 = add i32 %t2, %t3		; <i32> [#uses=1]
-	%t5 = load i8* %cp.05, align 1		; <i8> [#uses=1]
+	%t5 = load i8, i8* %cp.05, align 1		; <i8> [#uses=1]
 	%t6 = sext i8 %t5 to i32		; <i32> [#uses=1]
 	%t7 = xor i32 %t6, %t4		; <i32> [#uses=1]
 	%t8 = and i32 %t7, 16383		; <i32> [#uses=2]
 	%tmp = add i64 %indvar, 1		; <i64> [#uses=2]
-	%scevgep = getelementptr i8* %key, i64 %tmp		; <i8*> [#uses=1]
-	%t9 = load i8* %scevgep, align 1		; <i8> [#uses=1]
+	%scevgep = getelementptr i8, i8* %key, i64 %tmp		; <i8*> [#uses=1]
+	%t9 = load i8, i8* %scevgep, align 1		; <i8> [#uses=1]
 	%t10 = icmp eq i8 %t9, 0		; <i1> [#uses=1]
 	br i1 %t10, label %bb2, label %bb
 
diff --git a/test/Transforms/InstCombine/malloc-free-delete.ll b/test/Transforms/InstCombine/malloc-free-delete.ll
index ed25e4e49c84..dc04adb06966 100644
--- a/test/Transforms/InstCombine/malloc-free-delete.ll
+++ b/test/Transforms/InstCombine/malloc-free-delete.ll
@@ -3,9 +3,9 @@
 define i32 @main(i32 %argc, i8** %argv) {
 ; CHECK-LABEL: @main(
     %c_19 = alloca i8*
-    %malloc_206 = tail call i8* @malloc(i32 mul (i32 ptrtoint (i8* getelementptr (i8* null, i32 1) to i32), i32 10))
+    %malloc_206 = tail call i8* @malloc(i32 mul (i32 ptrtoint (i8* getelementptr (i8, i8* null, i32 1) to i32), i32 10))
     store i8* %malloc_206, i8** %c_19
-    %tmp_207 = load i8** %c_19
+    %tmp_207 = load i8*, i8** %c_19
     tail call void @free(i8* %tmp_207)
     ret i32 0
 ; CHECK-NEXT: ret i32 0
@@ -146,17 +146,36 @@ lpad.i:                                           ; preds = %entry
 }
 
 declare i8* @_Znwm(i64) nobuiltin
-declare void @_ZdlPvm(i8*, i64) nobuiltin
 declare i8* @_Znwj(i32) nobuiltin
-declare void @_ZdlPvj(i8*, i32) nobuiltin
 declare i8* @_Znam(i64) nobuiltin
-declare void @_ZdaPvm(i8*, i64) nobuiltin
 declare i8* @_Znaj(i32) nobuiltin
-declare void @_ZdaPvj(i8*, i32) nobuiltin
+declare void @_ZdlPv(i8*) nobuiltin
+declare void @_ZdaPv(i8*) nobuiltin
+
+define linkonce void @_ZdlPvm(i8* %p, i64) nobuiltin {
+  call void @_ZdlPv(i8* %p)
+  ret void
+}
+define linkonce void @_ZdlPvj(i8* %p, i32) nobuiltin {
+  call void @_ZdlPv(i8* %p)
+  ret void
+}
+define linkonce void @_ZdaPvm(i8* %p, i64) nobuiltin {
+  call void @_ZdaPv(i8* %p)
+  ret void
+}
+define linkonce void @_ZdaPvj(i8* %p, i32) nobuiltin {
+  call void @_ZdaPv(i8* %p)
+  ret void
+}
 
 ; CHECK-LABEL: @test8(
 define void @test8() {
   ; CHECK-NOT: call
+  %nw = call i8* @_Znwm(i64 32) builtin
+  call void @_ZdlPv(i8* %nw) builtin
+  %na = call i8* @_Znam(i64 32) builtin
+  call void @_ZdaPv(i8* %na) builtin
   %nwm = call i8* @_Znwm(i64 32) builtin
   call void @_ZdlPvm(i8* %nwm, i64 32) builtin
   %nwj = call i8* @_Znwj(i32 32) builtin
diff --git a/test/Transforms/InstCombine/max-of-nots.ll b/test/Transforms/InstCombine/max-of-nots.ll
new file mode 100644
index 000000000000..dd9c54b06739
--- /dev/null
+++ b/test/Transforms/InstCombine/max-of-nots.ll
@@ -0,0 +1,85 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+define i32 @compute_min_2(i32 %x, i32 %y) {
+; CHECK-LABEL: compute_min_2
+ entry:
+  %not_x = sub i32 -1, %x
+  %not_y = sub i32 -1, %y
+  %cmp = icmp sgt i32 %not_x, %not_y
+  %not_min = select i1 %cmp, i32 %not_x, i32 %not_y
+  %min = sub i32 -1, %not_min
+  ret i32 %min
+
+; CHECK: %0 = icmp slt i32 %x, %y
+; CHECK-NEXT: %1 = select i1 %0, i32 %x, i32 %y
+; CHECK-NEXT: ret i32 %1
+}
+
+define i32 @compute_min_3(i32 %x, i32 %y, i32 %z) {
+; CHECK-LABEL: compute_min_3
+ entry:
+  %not_x = sub i32 -1, %x
+  %not_y = sub i32 -1, %y
+  %not_z = sub i32 -1, %z
+  %cmp_1 = icmp sgt i32 %not_x, %not_y
+  %not_min_1 = select i1 %cmp_1, i32 %not_x, i32 %not_y
+  %cmp_2 = icmp sgt i32 %not_min_1, %not_z
+  %not_min_2 = select i1 %cmp_2, i32 %not_min_1, i32 %not_z
+  %min = sub i32 -1, %not_min_2
+  ret i32 %min
+
+; CHECK: %0 = icmp slt i32 %x, %y
+; CHECK-NEXT: %1 = select i1 %0, i32 %x, i32 %y
+; CHECK-NEXT: %2 = icmp slt i32 %1, %z
+; CHECK-NEXT: %3 = select i1 %2, i32 %1, i32 %z
+; CHECK-NEXT: ret i32 %3
+}
+
+define i32 @compute_min_arithmetic(i32 %x, i32 %y) {
+; CHECK-LABEL: compute_min_arithmetic
+ entry:
+  %not_value = sub i32 3, %x
+  %not_y = sub i32 -1, %y
+  %cmp = icmp sgt i32 %not_value, %not_y
+  %not_min = select i1 %cmp, i32 %not_value, i32 %not_y
+  ret i32 %not_min
+
+; CHECK: %0 = add i32 %x, -4
+; CHECK-NEXT: %1 = icmp slt i32 %0, %y
+; CHECK-NEXT: %2 = select i1 %1, i32 %0, i32 %y
+; CHECK-NEXT: %3 = xor i32 %2, -1
+; CHECK-NEXT: ret i32 %3
+}
+
+declare void @fake_use(i32)
+
+define i32 @compute_min_pessimization(i32 %x, i32 %y) {
+; CHECK-LABEL: compute_min_pessimization
+ entry:
+  %not_value = sub i32 3, %x
+  call void @fake_use(i32 %not_value)
+  %not_y = sub i32 -1, %y
+  %cmp = icmp sgt i32 %not_value, %not_y
+; CHECK: %not_value = sub i32 3, %x
+; CHECK: %cmp = icmp sgt i32 %not_value, %not_y
+  %not_min = select i1 %cmp, i32 %not_value, i32 %not_y
+  %min = sub i32 -1, %not_min
+  ret i32 %min
+}
+
+define i32 @max_of_nots(i32 %x, i32 %y) {
+; CHECK-LABEL: @max_of_nots(
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: xor
+; CHECK-NEXT: ret
+  %c0 = icmp sgt i32 %y, 0
+  %xor_y = xor i32 %y, -1
+  %s0 = select i1 %c0, i32 %xor_y, i32 -1
+  %xor_x = xor i32 %x, -1
+  %c1 = icmp slt i32 %s0, %xor_x
+  %smax96 = select i1 %c1, i32 %xor_x, i32 %s0
+  ret i32 %smax96
+}
diff --git a/test/Transforms/InstCombine/mem-gep-zidx.ll b/test/Transforms/InstCombine/mem-gep-zidx.ll
new file mode 100644
index 000000000000..cf021b133703
--- /dev/null
+++ b/test/Transforms/InstCombine/mem-gep-zidx.ll
@@ -0,0 +1,48 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+@f.a = private unnamed_addr constant [1 x i32] [i32 12], align 4
+@f.b = private unnamed_addr constant [1 x i32] [i32 55], align 4
+
+define signext i32 @test1(i32 signext %x) #0 {
+entry:
+  %idxprom = sext i32 %x to i64
+  %arrayidx = getelementptr inbounds [1 x i32], [1 x i32]* @f.a, i64 0, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
+  ret i32 %0
+
+; CHECK-LABEL: @test1
+; CHECK: ret i32 12
+}
+
+declare void @foo(i64* %p)
+define void @test2(i32 signext %x, i64 %v) #0 {
+entry:
+  %p = alloca i64
+  %idxprom = sext i32 %x to i64
+  %arrayidx = getelementptr inbounds i64, i64* %p, i64 %idxprom
+  store i64 %v, i64* %arrayidx
+  call void @foo(i64* %p)
+  ret void
+
+; CHECK-LABEL: @test2
+; CHECK: %p = alloca i64
+; CHECK: store i64 %v, i64* %p
+; CHECK: ret void
+}
+
+define signext i32 @test3(i32 signext %x, i1 %y) #0 {
+entry:
+  %idxprom = sext i32 %x to i64
+  %p = select i1 %y, [1 x i32]* @f.a, [1 x i32]* @f.b
+  %arrayidx = getelementptr inbounds [1 x i32], [1 x i32]* %p, i64 0, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
+  ret i32 %0
+
+; CHECK-LABEL: @test3
+; CHECK: getelementptr inbounds [1 x i32], [1 x i32]* %p, i64 0, i64 0
+}
+
+attributes #0 = { nounwind readnone }
+
diff --git a/test/Transforms/InstCombine/memchr.ll b/test/Transforms/InstCombine/memchr.ll
new file mode 100644
index 000000000000..216dba874ccb
--- /dev/null
+++ b/test/Transforms/InstCombine/memchr.ll
@@ -0,0 +1,200 @@
+; Test that the memchr library call simplifier works correctly.
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32"
+
+@hello = constant [14 x i8] c"hello world\5Cn\00"
+@hellonull = constant [14 x i8] c"hello\00world\5Cn\00"
+@null = constant [1 x i8] zeroinitializer
+@newlines = constant [3 x i8] c"\0D\0A\00"
+@single = constant [2 x i8] c"\1F\00"
+@spaces = constant [4 x i8] c" \0D\0A\00"
+@negative = constant [3 x i8] c"\FF\FE\00"
+@chp = global i8* zeroinitializer
+
+declare i8* @memchr(i8*, i32, i32)
+
+define void @test1() {
+; CHECK-LABEL: @test1
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 6)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %str, i32 119, i32 14)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test2() {
+; CHECK-LABEL: @test2
+; CHECK: store i8* null, i8** @chp, align 4
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %str, i32 119, i32 1)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test3() {
+; CHECK-LABEL: @test3
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %src, i32 0, i32 14)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test4(i32 %chr) {
+; CHECK-LABEL: @test4
+; CHECK: call i8* @memchr
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %src, i32 %chr, i32 14)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test5() {
+; CHECK-LABEL: @test5
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %src, i32 65280, i32 14)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test6() {
+; CHECK-LABEL: @test6
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 6)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+; Overflow, but we still find the right thing.
+  %dst = call i8* @memchr(i8* %src, i32 119, i32 100)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test7() {
+; CHECK-LABEL: @test7
+; CHECK: store i8* null, i8** @chp, align 4
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+; Overflow
+  %dst = call i8* @memchr(i8* %src, i32 120, i32 100)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test8() {
+; CHECK-LABEL: @test8
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hellonull, i32 0, i32 6)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %str = getelementptr [14 x i8], [14 x i8]* @hellonull, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %str, i32 119, i32 14)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test9() {
+; CHECK-LABEL: @test9
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hellonull, i32 0, i32 6)
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %str = getelementptr [14 x i8], [14 x i8]* @hellonull, i32 0, i32 2
+  %dst = call i8* @memchr(i8* %str, i32 119, i32 12)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+define void @test10() {
+; CHECK-LABEL: @test10
+; CHECK: store i8* null, i8** @chp, align 4
+; CHECK-NOT: call i8* @memchr
+; CHECK: ret void
+
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
+  %dst = call i8* @memchr(i8* %str, i32 119, i32 6)
+  store i8* %dst, i8** @chp
+  ret void
+}
+
+; Check transformation memchr("\r\n", C, 2) != nullptr -> (C & 9216) != 0
+define i1 @test11(i32 %C) {
+; CHECK-LABEL: @test11
+; CHECK-NEXT: [[TRUNC:%.*]] = trunc i32 %C to i16
+; CHECK-NEXT: %memchr.bounds = icmp ult i16 [[TRUNC]], 16
+; CHECK-NEXT: [[SHL:%.*]] = shl i16 1, [[TRUNC]]
+; CHECK-NEXT: [[AND:%.*]] = and i16 [[SHL]], 9216
+; CHECK-NEXT: %memchr.bits = icmp ne i16 [[AND]], 0
+; CHECK-NEXT: %memchr = and i1 %memchr.bounds, %memchr.bits
+; CHECK-NEXT: ret i1 %memchr
+
+  %dst = call i8* @memchr(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @newlines, i64 0, i64 0), i32 %C, i32 2)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
+
+; No 64 bits here
+define i1 @test12(i32 %C) {
+; CHECK-LABEL: @test12
+; CHECK-NEXT: %dst = call i8* @memchr(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @spaces, i32 0, i32 0), i32 %C, i32 3)
+; CHECK-NEXT: %cmp = icmp ne i8* %dst, null
+; CHECK-NEXT: ret i1 %cmp
+
+  %dst = call i8* @memchr(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @spaces, i64 0, i64 0), i32 %C, i32 3)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
+
+define i1 @test13(i32 %C) {
+; CHECK-LABEL: @test13
+; CHECK-NEXT: %memchr.bounds = icmp ult i32 %C, 32
+; CHECK-NEXT: [[SHL:%.*]] = shl i32 1, %C
+; CHECK-NEXT: [[AND:%.*]] = and i32 [[SHL]], -2147483647
+; CHECK-NEXT: %memchr.bits = icmp ne i32 [[AND]], 0
+; CHECK-NEXT: %memchr = and i1 %memchr.bounds, %memchr.bits
+; CHECK-NEXT: ret i1 %memchr
+
+  %dst = call i8* @memchr(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @single, i64 0, i64 0), i32 %C, i32 2)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
+
+define i1 @test14(i32 %C) {
+; CHECK-LABEL: @test14
+; CHECK-NEXT: icmp eq i32 %C, 31
+; CHECK-NEXT: ret
+
+  %dst = call i8* @memchr(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @single, i64 0, i64 0), i32 %C, i32 1)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
+
+define i1 @test15(i32 %C) {
+; CHECK-LABEL: @test15
+; CHECK-NEXT: %dst = call i8* @memchr(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @negative, i32 0, i32 0), i32 %C, i32 3)
+; CHECK-NEXT: %cmp = icmp ne i8* %dst, null
+; CHECK-NEXT: ret i1 %cmp
+
+  %dst = call i8* @memchr(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @negative, i64 0, i64 0), i32 %C, i32 3)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
diff --git a/test/Transforms/InstCombine/memcmp-1.ll b/test/Transforms/InstCombine/memcmp-1.ll
index d960693a05d4..db15bd66b715 100644
--- a/test/Transforms/InstCombine/memcmp-1.ll
+++ b/test/Transforms/InstCombine/memcmp-1.ll
@@ -33,9 +33,9 @@ define i32 @test_simplify2(i8* %mem1, i8* %mem2) {
 define i32 @test_simplify3(i8* %mem1, i8* %mem2) {
 ; CHECK-LABEL: @test_simplify3(
   %ret = call i32 @memcmp(i8* %mem1, i8* %mem2, i32 1)
-; CHECK: [[LOAD1:%[a-z]+]] = load i8* %mem1, align 1
+; CHECK: [[LOAD1:%[a-z]+]] = load i8, i8* %mem1, align 1
 ; CHECK: [[ZEXT1:%[a-z]+]] = zext i8 [[LOAD1]] to i32
-; CHECK: [[LOAD2:%[a-z]+]] = load i8* %mem2, align 1
+; CHECK: [[LOAD2:%[a-z]+]] = load i8, i8* %mem2, align 1
 ; CHECK: [[ZEXT2:%[a-z]+]] = zext i8 [[LOAD2]] to i32
 ; CHECK: [[RET:%[a-z]+]] = sub nsw i32 [[ZEXT1]], [[ZEXT2]]
   ret i32 %ret
@@ -46,8 +46,8 @@ define i32 @test_simplify3(i8* %mem1, i8* %mem2) {
 
 define i32 @test_simplify4() {
 ; CHECK-LABEL: @test_simplify4(
-  %mem1 = getelementptr [4 x i8]* @hel, i32 0, i32 0
-  %mem2 = getelementptr [8 x i8]* @hello_u, i32 0, i32 0
+  %mem1 = getelementptr [4 x i8], [4 x i8]* @hel, i32 0, i32 0
+  %mem2 = getelementptr [8 x i8], [8 x i8]* @hello_u, i32 0, i32 0
   %ret = call i32 @memcmp(i8* %mem1, i8* %mem2, i32 3)
   ret i32 %ret
 ; CHECK: ret i32 0
@@ -55,8 +55,8 @@ define i32 @test_simplify4() {
 
 define i32 @test_simplify5() {
 ; CHECK-LABEL: @test_simplify5(
-  %mem1 = getelementptr [4 x i8]* @hel, i32 0, i32 0
-  %mem2 = getelementptr [4 x i8]* @foo, i32 0, i32 0
+  %mem1 = getelementptr [4 x i8], [4 x i8]* @hel, i32 0, i32 0
+  %mem2 = getelementptr [4 x i8], [4 x i8]* @foo, i32 0, i32 0
   %ret = call i32 @memcmp(i8* %mem1, i8* %mem2, i32 3)
   ret i32 %ret
 ; CHECK: ret i32 1
@@ -64,8 +64,8 @@ define i32 @test_simplify5() {
 
 define i32 @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %mem1 = getelementptr [4 x i8]* @foo, i32 0, i32 0
-  %mem2 = getelementptr [4 x i8]* @hel, i32 0, i32 0
+  %mem1 = getelementptr [4 x i8], [4 x i8]* @foo, i32 0, i32 0
+  %mem2 = getelementptr [4 x i8], [4 x i8]* @hel, i32 0, i32 0
   %ret = call i32 @memcmp(i8* %mem1, i8* %mem2, i32 3)
   ret i32 %ret
 ; CHECK: ret i32 -1
diff --git a/test/Transforms/InstCombine/memcpy-from-global.ll b/test/Transforms/InstCombine/memcpy-from-global.ll
index 3bc1d36f486d..da38087d7397 100644
--- a/test/Transforms/InstCombine/memcpy-from-global.ll
+++ b/test/Transforms/InstCombine/memcpy-from-global.ll
@@ -14,23 +14,23 @@ entry:
 
 	%tmp3 = shl i32 %hash, 2		; <i32> [#uses=1]
 	%tmp5 = and i32 %tmp3, 124		; <i32> [#uses=4]
-	%tmp753 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp5		; <float*> [#uses=1]
-	%tmp9 = load float* %tmp753		; <float> [#uses=1]
+	%tmp753 = getelementptr [128 x float], [128 x float]* %lookupTable, i32 0, i32 %tmp5		; <float*> [#uses=1]
+	%tmp9 = load float, float* %tmp753		; <float> [#uses=1]
 	%tmp11 = fmul float %tmp9, %x		; <float> [#uses=1]
 	%tmp13 = fadd float %tmp11, 0.000000e+00		; <float> [#uses=1]
 	%tmp17.sum52 = or i32 %tmp5, 1		; <i32> [#uses=1]
-	%tmp1851 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp17.sum52		; <float*> [#uses=1]
-	%tmp19 = load float* %tmp1851		; <float> [#uses=1]
+	%tmp1851 = getelementptr [128 x float], [128 x float]* %lookupTable, i32 0, i32 %tmp17.sum52		; <float*> [#uses=1]
+	%tmp19 = load float, float* %tmp1851		; <float> [#uses=1]
 	%tmp21 = fmul float %tmp19, %y		; <float> [#uses=1]
 	%tmp23 = fadd float %tmp21, %tmp13		; <float> [#uses=1]
 	%tmp27.sum50 = or i32 %tmp5, 2		; <i32> [#uses=1]
-	%tmp2849 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp27.sum50		; <float*> [#uses=1]
-	%tmp29 = load float* %tmp2849		; <float> [#uses=1]
+	%tmp2849 = getelementptr [128 x float], [128 x float]* %lookupTable, i32 0, i32 %tmp27.sum50		; <float*> [#uses=1]
+	%tmp29 = load float, float* %tmp2849		; <float> [#uses=1]
 	%tmp31 = fmul float %tmp29, %z		; <float> [#uses=1]
 	%tmp33 = fadd float %tmp31, %tmp23		; <float> [#uses=1]
 	%tmp37.sum48 = or i32 %tmp5, 3		; <i32> [#uses=1]
-	%tmp3847 = getelementptr [128 x float]* %lookupTable, i32 0, i32 %tmp37.sum48		; <float*> [#uses=1]
-	%tmp39 = load float* %tmp3847		; <float> [#uses=1]
+	%tmp3847 = getelementptr [128 x float], [128 x float]* %lookupTable, i32 0, i32 %tmp37.sum48		; <float*> [#uses=1]
+	%tmp39 = load float, float* %tmp3847		; <float> [#uses=1]
 	%tmp41 = fmul float %tmp39, %w		; <float> [#uses=1]
 	%tmp43 = fadd float %tmp41, %tmp33		; <float> [#uses=1]
 	ret float %tmp43
@@ -57,10 +57,10 @@ define void @test2() {
 
 ; %A alloca is deleted
 ; CHECK-NEXT: alloca [124 x i8]
-; CHECK-NEXT: getelementptr inbounds [124 x i8]*
+; CHECK-NEXT: getelementptr inbounds [124 x i8], [124 x i8]*
 
 ; use @G instead of %A
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.*}}, i8* getelementptr inbounds (%T* @G, i64 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %{{.*}}, i8* getelementptr inbounds (%T, %T* @G, i64 0, i32 0)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %b, i8* %a, i64 124, i32 4, i1 false)
   call void @bar(i8* %b)
@@ -101,7 +101,7 @@ define void @test3() {
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test3(
-; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T* @G, i64 0, i32 0))
+; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T, %T* @G, i64 0, i32 0))
   ret void
 }
 
@@ -111,7 +111,7 @@ define void @test3_addrspacecast() {
   call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%T* @G to i8 addrspace(1)*), i64 124, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test3_addrspacecast(
-; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T* @G, i64 0, i32 0))
+; CHECK-NEXT: call void @bar(i8* getelementptr inbounds (%T, %T* @G, i64 0, i32 0))
   ret void
 }
 
@@ -122,7 +122,7 @@ define void @test4() {
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
   call void @baz(i8* byval %a)
 ; CHECK-LABEL: @test4(
-; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T* @G, i64 0, i32 0))
+; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T, %T* @G, i64 0, i32 0))
   ret void
 }
 
@@ -134,7 +134,7 @@ define void @test5() {
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%T* @G to i8*), i64 124, i32 4, i1 false)
   call void @baz(i8* byval %a)
 ; CHECK-LABEL: @test5(
-; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T* @G, i64 0, i32 0))
+; CHECK-NEXT: call void @baz(i8* byval getelementptr inbounds (%T, %T* @G, i64 0, i32 0))
   ret void
 }
 
@@ -155,7 +155,7 @@ define void @test6() {
 define void @test7() {
   %A = alloca %U, align 16
   %a = bitcast %U* %A to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 0) to i8*), i64 20, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U], [2 x %U]* @H, i64 0, i32 0) to i8*), i64 20, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test7(
 ; CHECK-NEXT: call void @bar(i8* bitcast ([2 x %U]* @H to i8*))
@@ -165,7 +165,7 @@ define void @test7() {
 define void @test8() {
   %A = alloca %U, align 16
   %a = bitcast %U* %A to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U], [2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test8(
 ; CHECK: llvm.memcpy
@@ -177,7 +177,7 @@ define void @test8() {
 define void @test8_addrspacecast() {
   %A = alloca %U, align 16
   %a = bitcast %U* %A to i8*
-  call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8 addrspace(1)*), i64 20, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%U* getelementptr ([2 x %U], [2 x %U]* @H, i64 0, i32 1) to i8 addrspace(1)*), i64 20, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test8_addrspacecast(
 ; CHECK: llvm.memcpy
@@ -188,19 +188,19 @@ define void @test8_addrspacecast() {
 define void @test9() {
   %A = alloca %U, align 4
   %a = bitcast %U* %A to i8*
-  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* bitcast (%U* getelementptr ([2 x %U], [2 x %U]* @H, i64 0, i32 1) to i8*), i64 20, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test9(
-; CHECK-NEXT: call void @bar(i8* bitcast (%U* getelementptr inbounds ([2 x %U]* @H, i64 0, i64 1) to i8*))
+; CHECK-NEXT: call void @bar(i8* bitcast (%U* getelementptr inbounds ([2 x %U], [2 x %U]* @H, i64 0, i64 1) to i8*))
   ret void
 }
 
 define void @test9_addrspacecast() {
   %A = alloca %U, align 4
   %a = bitcast %U* %A to i8*
-  call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%U* getelementptr ([2 x %U]* @H, i64 0, i32 1) to i8 addrspace(1)*), i64 20, i32 4, i1 false)
+  call void @llvm.memcpy.p0i8.p1i8.i64(i8* %a, i8 addrspace(1)* addrspacecast (%U* getelementptr ([2 x %U], [2 x %U]* @H, i64 0, i32 1) to i8 addrspace(1)*), i64 20, i32 4, i1 false)
   call void @bar(i8* %a) readonly
 ; CHECK-LABEL: @test9_addrspacecast(
-; CHECK-NEXT: call void @bar(i8* bitcast (%U* getelementptr inbounds ([2 x %U]* @H, i64 0, i64 1) to i8*))
+; CHECK-NEXT: call void @bar(i8* bitcast (%U* getelementptr inbounds ([2 x %U], [2 x %U]* @H, i64 0, i64 1) to i8*))
   ret void
 }
diff --git a/test/Transforms/InstCombine/memmove.ll b/test/Transforms/InstCombine/memmove.ll
index 9d51ea0e70bb..96f230eece06 100644
--- a/test/Transforms/InstCombine/memmove.ll
+++ b/test/Transforms/InstCombine/memmove.ll
@@ -14,16 +14,16 @@ define void @test1(i8* %A, i8* %B, i32 %N) {
 
 define void @test2(i8* %A, i32 %N) {
         ;; dest can't alias source since we can't write to source!
-	call void @llvm.memmove.p0i8.p0i8.i32(i8* %A, i8* getelementptr inbounds ([33 x i8]* @S, i32 0, i32 0), i32 %N, i32 1, i1 false)
+	call void @llvm.memmove.p0i8.p0i8.i32(i8* %A, i8* getelementptr inbounds ([33 x i8], [33 x i8]* @S, i32 0, i32 0), i32 %N, i32 1, i1 false)
 	ret void
 }
 
 define i32 @test3() {
-	%h_p = getelementptr [2 x i8]* @h, i32 0, i32 0		; <i8*> [#uses=1]
-	%hel_p = getelementptr [4 x i8]* @hel, i32 0, i32 0		; <i8*> [#uses=1]
-	%hello_u_p = getelementptr [8 x i8]* @hello_u, i32 0, i32 0		; <i8*> [#uses=1]
+	%h_p = getelementptr [2 x i8], [2 x i8]* @h, i32 0, i32 0		; <i8*> [#uses=1]
+	%hel_p = getelementptr [4 x i8], [4 x i8]* @hel, i32 0, i32 0		; <i8*> [#uses=1]
+	%hello_u_p = getelementptr [8 x i8], [8 x i8]* @hello_u, i32 0, i32 0		; <i8*> [#uses=1]
 	%target = alloca [1024 x i8]		; <[1024 x i8]*> [#uses=1]
-	%target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0		; <i8*> [#uses=3]
+	%target_p = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0		; <i8*> [#uses=3]
         call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %h_p, i32 2, i32 2, i1 false)
         call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %hel_p, i32 4, i32 4, i1 false)
         call void @llvm.memmove.p0i8.p0i8.i32(i8* %target_p, i8* %hello_u_p, i32 8, i32 8, i1 false)
diff --git a/test/Transforms/InstCombine/memset.ll b/test/Transforms/InstCombine/memset.ll
index 7f02dad58a7a..dfafcf948d1c 100644
--- a/test/Transforms/InstCombine/memset.ll
+++ b/test/Transforms/InstCombine/memset.ll
@@ -2,7 +2,7 @@
 
 define i32 @main() {
   %target = alloca [1024 x i8]
-  %target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  %target_p = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0
   call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 0, i32 1, i1 false)
   call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 1, i32 1, i1 false)
   call void @llvm.memset.p0i8.i32(i8* %target_p, i8 1, i32 2, i32 2, i1 false)
diff --git a/test/Transforms/InstCombine/memset2.ll b/test/Transforms/InstCombine/memset2.ll
index 87639f0d6a22..ad4c225846bf 100644
--- a/test/Transforms/InstCombine/memset2.ll
+++ b/test/Transforms/InstCombine/memset2.ll
@@ -7,7 +7,7 @@
 define i32 @test(%struct.Moves addrspace(1)* nocapture %moves) {
 entry:
 ; CHECK: bitcast i8 addrspace(1)* %gep to i64 addrspace(1)*
-	%gep = getelementptr inbounds %struct.Moves addrspace(1)* %moves, i32 1, i32 0, i32 9
+	%gep = getelementptr inbounds %struct.Moves, %struct.Moves addrspace(1)* %moves, i32 1, i32 0, i32 9
 	 call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %gep, i8 0, i64 8, i32 1, i1 false)                                                                     
 	ret i32 0
 }
diff --git a/test/Transforms/InstCombine/merge-icmp.ll b/test/Transforms/InstCombine/merge-icmp.ll
index b021fe0429a5..6a65b5befa38 100644
--- a/test/Transforms/InstCombine/merge-icmp.ll
+++ b/test/Transforms/InstCombine/merge-icmp.ll
@@ -1,7 +1,7 @@
 ; RUN: opt -S -instcombine < %s | FileCheck %s
 
 define i1 @test1(i16* %x) {
-  %load = load i16* %x, align 4
+  %load = load i16, i16* %x, align 4
   %trunc = trunc i16 %load to i8
   %cmp1 = icmp eq i8 %trunc, 127
   %and = and i16 %load, -256
@@ -15,7 +15,7 @@ define i1 @test1(i16* %x) {
 }
 
 define i1 @test2(i16* %x) {
-  %load = load i16* %x, align 4
+  %load = load i16, i16* %x, align 4
   %and = and i16 %load, -256
   %cmp1 = icmp eq i16 %and, 32512
   %trunc = trunc i16 %load to i8
diff --git a/test/Transforms/InstCombine/minmax-fold.ll b/test/Transforms/InstCombine/minmax-fold.ll
new file mode 100644
index 000000000000..4ed0c78216d6
--- /dev/null
+++ b/test/Transforms/InstCombine/minmax-fold.ll
@@ -0,0 +1,112 @@
+; RUN: opt -S -instcombine < %s | FileCheck %s
+
+; CHECK-LABEL: @t1
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: sext
+define i64 @t1(i32 %a) {
+  ; This is the canonical form for a type-changing min/max.
+  %1 = icmp slt i32 %a, 5
+  %2 = select i1 %1, i32 %a, i32 5
+  %3 = sext i32 %2 to i64
+  ret i64 %3
+}
+
+; CHECK-LABEL: @t2
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: sext
+define i64 @t2(i32 %a) {
+  ; Check this is converted into canonical form, as above.
+  %1 = icmp slt i32 %a, 5
+  %2 = sext i32 %a to i64
+  %3 = select i1 %1, i64 %2, i64 5
+  ret i64 %3
+}
+
+; CHECK-LABEL: @t3
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: zext
+define i64 @t3(i32 %a) {
+  ; Same as @t2, with flipped operands and zext instead of sext.
+  %1 = icmp ult i32 %a, 5
+  %2 = zext i32 %a to i64
+  %3 = select i1 %1, i64 5, i64 %2
+  ret i64 %3
+}
+
+; CHECK-LABEL: @t4
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: trunc
+define i32 @t4(i64 %a) {
+  ; Same again, with trunc.
+  %1 = icmp slt i64 %a, 5
+  %2 = trunc i64 %a to i32
+  %3 = select i1 %1, i32 %2, i32 5
+  ret i32 %3
+}
+
+; CHECK-LABEL: @t5
+; CHECK-NEXT: icmp
+; CHECK-NEXT: zext
+; CHECK-NEXT: select
+define i64 @t5(i32 %a) {
+  ; Same as @t3, but with mismatched signedness between icmp and zext.
+  ; InstCombine should leave this alone.
+  %1 = icmp slt i32 %a, 5
+  %2 = zext i32 %a to i64
+  %3 = select i1 %1, i64 5, i64 %2
+  ret i64 %3
+}
+
+; CHECK-LABEL: @t6
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: sitofp
+define float @t6(i32 %a) {
+  %1 = icmp slt i32 %a, 0
+  %2 = select i1 %1, i32 %a, i32 0
+  %3 = sitofp i32 %2 to float
+  ret float %3
+}
+
+; CHECK-LABEL: @t7
+; CHECK-NEXT: icmp
+; CHECK-NEXT: select
+; CHECK-NEXT: trunc
+define i16 @t7(i32 %a) {
+  %1 = icmp slt i32 %a, -32768
+  %2 = trunc i32 %a to i16
+  %3 = select i1 %1, i16 %2, i16 -32768
+  ret i16 %3
+}
+
+; Just check for no infinite loop. InstSimplify liked to
+; "simplify" -32767 by removing all the sign bits,
+; which led to a canonicalization fight between different
+; parts of instcombine.
+define i32 @t8(i64 %a, i32 %b) {
+  %1 = icmp slt i64 %a, -32767
+  %2 = select i1 %1, i64 %a, i64 -32767
+  %3 = trunc i64 %2 to i32
+  %4 = icmp slt i32 %b, 42
+  %5 = select i1 %4, i32 42, i32 %3
+  %6 = icmp ne i32 %5, %b
+  %7 = zext i1 %6 to i32
+  ret i32 %7
+}
+
+; Ensure this doesn't get converted to a min/max.
+; CHECK-LABEL: @t9
+; CHECK-NEXT: icmp
+; CHECK-NEXT: sext
+; CHECK-NEXT: 4294967295
+; CHECK-NEXT: ret
+define i64 @t9(i32 %a) {
+  %1 = icmp sgt i32 %a, -1
+  %2 = sext i32 %a to i64
+  %3 = select i1 %1, i64 %2, i64 4294967295
+  ret i64 %3
+}
diff --git a/test/Transforms/InstCombine/mul.ll b/test/Transforms/InstCombine/mul.ll
index 4d1e6c700bda..6e67c7fa08a0 100644
--- a/test/Transforms/InstCombine/mul.ll
+++ b/test/Transforms/InstCombine/mul.ll
@@ -107,7 +107,7 @@ define i32 @test12(i32 %a, i32 %b) {
 ; PR2642
 define internal void @test13(<4 x float>*) {
 ; CHECK-LABEL: @test13(
-	load <4 x float>* %0, align 1
+	load <4 x float>, <4 x float>* %0, align 1
 	fmul <4 x float> %2, < float 1.000000e+00, float 1.000000e+00, float 1.000000e+00, float 1.000000e+00 >
 	store <4 x float> %3, <4 x float>* %0, align 1
 	ret void
@@ -288,3 +288,19 @@ define i32 @test31(i32 %V) {
 ; CHECK:      %[[mul:.*]] = shl i32 %V, zext (i1 icmp ne (i32* inttoptr (i64 1 to i32*), i32* @PR22087) to i32)
 ; CHECK-NEXT: ret i32 %[[mul]]
 }
+
+define i32 @test32(i32 %X) {
+; CHECK-LABEL: @test32
+  %mul = mul nsw i32 %X, -2147483648
+  ret i32 %mul
+; CHECK:      %[[shl:.*]] = shl i32 %X, 31
+; CHECK-NEXT: ret i32 %[[shl]]
+}
+
+define i32 @test33(i32 %X) {
+; CHECK-LABEL: @test33
+  %mul = mul nsw i32 %X, 1073741824
+; CHECK:      %[[shl:.*]] = shl nsw i32 %X, 30
+; CHECK-NEXT: ret i32 %[[shl]]
+  ret i32 %mul
+}
diff --git a/test/Transforms/InstCombine/multi-size-address-space-pointer.ll b/test/Transforms/InstCombine/multi-size-address-space-pointer.ll
index 2d88bed4e7bf..4e5b2100f34a 100644
--- a/test/Transforms/InstCombine/multi-size-address-space-pointer.ll
+++ b/test/Transforms/InstCombine/multi-size-address-space-pointer.ll
@@ -4,43 +4,43 @@ target datalayout = "e-p:32:32:32-p1:64:64:64-p2:8:8:8-p3:16:16:16-p4:16:16:16-i
 
 define i32 @test_as0(i32 addrspace(0)* %a) {
 ; CHECK-LABEL: @test_as0(
-; CHECK: %arrayidx = getelementptr i32* %a, i32 1
-  %arrayidx = getelementptr i32 addrspace(0)* %a, i64 1
-  %y = load i32 addrspace(0)* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr i32, i32* %a, i32 1
+  %arrayidx = getelementptr i32, i32 addrspace(0)* %a, i64 1
+  %y = load i32, i32 addrspace(0)* %arrayidx, align 4
   ret i32 %y
 }
 
 define i32 @test_as1(i32 addrspace(1)* %a) {
 ; CHECK-LABEL: @test_as1(
-; CHECK: %arrayidx = getelementptr i32 addrspace(1)* %a, i64 1
-  %arrayidx = getelementptr i32 addrspace(1)* %a, i32 1
-  %y = load i32 addrspace(1)* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr i32, i32 addrspace(1)* %a, i64 1
+  %arrayidx = getelementptr i32, i32 addrspace(1)* %a, i32 1
+  %y = load i32, i32 addrspace(1)* %arrayidx, align 4
   ret i32 %y
 }
 
 define i32 @test_as2(i32 addrspace(2)* %a) {
 ; CHECK-LABEL: @test_as2(
-; CHECK: %arrayidx = getelementptr i32 addrspace(2)* %a, i8 1
-  %arrayidx = getelementptr i32 addrspace(2)* %a, i32 1
-  %y = load i32 addrspace(2)* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr i32, i32 addrspace(2)* %a, i8 1
+  %arrayidx = getelementptr i32, i32 addrspace(2)* %a, i32 1
+  %y = load i32, i32 addrspace(2)* %arrayidx, align 4
   ret i32 %y
 }
 
 define i32 @test_as3(i32 addrspace(3)* %a) {
 ; CHECK-LABEL: @test_as3(
-; CHECK: %arrayidx = getelementptr i32 addrspace(3)* %a, i16 1
-  %arrayidx = getelementptr i32 addrspace(3)* %a, i32 1
-  %y = load i32 addrspace(3)* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr i32, i32 addrspace(3)* %a, i16 1
+  %arrayidx = getelementptr i32, i32 addrspace(3)* %a, i32 1
+  %y = load i32, i32 addrspace(3)* %arrayidx, align 4
   ret i32 %y
 }
 
 define i32 @test_combine_ptrtoint(i32 addrspace(2)* %a) {
 ; CHECK-LABEL: @test_combine_ptrtoint(
-; CHECK-NEXT: %y = load i32 addrspace(2)* %a
+; CHECK-NEXT: %y = load i32, i32 addrspace(2)* %a
 ; CHECK-NEXT: ret i32 %y
   %cast = ptrtoint i32 addrspace(2)* %a to i8
   %castback = inttoptr i8 %cast to i32 addrspace(2)*
-  %y = load i32 addrspace(2)* %castback, align 4
+  %y = load i32, i32 addrspace(2)* %castback, align 4
   ret i32 %y
 }
 
@@ -55,12 +55,12 @@ define i8 @test_combine_inttoptr(i8 %a) {
 define i32 @test_combine_vector_ptrtoint(<2 x i32 addrspace(2)*> %a) {
 ; CHECK-LABEL: @test_combine_vector_ptrtoint(
 ; CHECK-NEXT: %p = extractelement <2 x i32 addrspace(2)*> %a, i32 0
-; CHECK-NEXT: %y = load i32 addrspace(2)* %p, align 4
+; CHECK-NEXT: %y = load i32, i32 addrspace(2)* %p, align 4
 ; CHECK-NEXT: ret i32 %y
   %cast = ptrtoint <2 x i32 addrspace(2)*> %a to <2 x i8>
   %castback = inttoptr <2 x i8> %cast to <2 x i32 addrspace(2)*>
   %p = extractelement <2 x i32 addrspace(2)*> %castback, i32 0
-  %y = load i32 addrspace(2)* %p, align 4
+  %y = load i32, i32 addrspace(2)* %p, align 4
   ret i32 %y
 }
 
@@ -75,38 +75,38 @@ define <2 x i8> @test_combine_vector_inttoptr(<2 x i8> %a) {
 ; Check that the GEP index is changed to the address space integer type (i64 -> i8)
 define i32 addrspace(2)* @shrink_gep_constant_index_64_as2(i32 addrspace(2)* %p) {
 ; CHECK-LABEL: @shrink_gep_constant_index_64_as2(
-; CHECK-NEXT: getelementptr i32 addrspace(2)* %p, i8 1
-  %ret = getelementptr i32 addrspace(2)* %p, i64 1
+; CHECK-NEXT: getelementptr i32, i32 addrspace(2)* %p, i8 1
+  %ret = getelementptr i32, i32 addrspace(2)* %p, i64 1
   ret i32 addrspace(2)* %ret
 }
 
 define i32 addrspace(2)* @shrink_gep_constant_index_32_as2(i32 addrspace(2)* %p) {
 ; CHECK-LABEL: @shrink_gep_constant_index_32_as2(
-; CHECK-NEXT: getelementptr i32 addrspace(2)* %p, i8 1
-  %ret = getelementptr i32 addrspace(2)* %p, i32 1
+; CHECK-NEXT: getelementptr i32, i32 addrspace(2)* %p, i8 1
+  %ret = getelementptr i32, i32 addrspace(2)* %p, i32 1
   ret i32 addrspace(2)* %ret
 }
 
 define i32 addrspace(3)* @shrink_gep_constant_index_64_as3(i32 addrspace(3)* %p) {
 ; CHECK-LABEL: @shrink_gep_constant_index_64_as3(
-; CHECK-NEXT: getelementptr i32 addrspace(3)* %p, i16 1
-  %ret = getelementptr i32 addrspace(3)* %p, i64 1
+; CHECK-NEXT: getelementptr i32, i32 addrspace(3)* %p, i16 1
+  %ret = getelementptr i32, i32 addrspace(3)* %p, i64 1
   ret i32 addrspace(3)* %ret
 }
 
 define i32 addrspace(2)* @shrink_gep_variable_index_64_as2(i32 addrspace(2)* %p, i64 %idx) {
 ; CHECK-LABEL: @shrink_gep_variable_index_64_as2(
 ; CHECK-NEXT: %1 = trunc i64 %idx to i8
-; CHECK-NEXT: getelementptr i32 addrspace(2)* %p, i8 %1
-  %ret = getelementptr i32 addrspace(2)* %p, i64 %idx
+; CHECK-NEXT: getelementptr i32, i32 addrspace(2)* %p, i8 %1
+  %ret = getelementptr i32, i32 addrspace(2)* %p, i64 %idx
   ret i32 addrspace(2)* %ret
 }
 
 define i32 addrspace(1)* @grow_gep_variable_index_8_as1(i32 addrspace(1)* %p, i8 %idx) {
 ; CHECK-LABEL: @grow_gep_variable_index_8_as1(
 ; CHECK-NEXT: %1 = sext i8 %idx to i64
-; CHECK-NEXT: getelementptr i32 addrspace(1)* %p, i64 %1
-  %ret = getelementptr i32 addrspace(1)* %p, i8 %idx
+; CHECK-NEXT: getelementptr i32, i32 addrspace(1)* %p, i64 %1
+  %ret = getelementptr i32, i32 addrspace(1)* %p, i8 %idx
   ret i32 addrspace(1)* %ret
 }
 
diff --git a/test/Transforms/InstCombine/no-negzero.ll b/test/Transforms/InstCombine/no-negzero.ll
index 4ed283603446..07e68251b5ae 100644
--- a/test/Transforms/InstCombine/no-negzero.ll
+++ b/test/Transforms/InstCombine/no-negzero.ll
@@ -14,17 +14,17 @@ entry:
   %0 = alloca double, align 8                     ; <double*> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store double %x, double* %x_addr
-  %1 = load double* %x_addr, align 8              ; <double> [#uses=1]
+  %1 = load double, double* %x_addr, align 8              ; <double> [#uses=1]
   %2 = call double @fabs(double %1) nounwind readnone ; <double> [#uses=1]
   %3 = call double @sqrt(double %2) nounwind readonly ; <double> [#uses=1]
   %4 = fadd double %3, 0.000000e+00               ; <double> [#uses=1]
   store double %4, double* %0, align 8
-  %5 = load double* %0, align 8                   ; <double> [#uses=1]
+  %5 = load double, double* %0, align 8                   ; <double> [#uses=1]
   store double %5, double* %retval, align 8
   br label %return
 
 return:                                           ; preds = %entry
-  %retval1 = load double* %retval                 ; <double> [#uses=1]
+  %retval1 = load double, double* %retval                 ; <double> [#uses=1]
   ret double %retval1
 }
 
diff --git a/test/Transforms/InstCombine/not.ll b/test/Transforms/InstCombine/not.ll
index 4012ce1ea4ba..9d59edd7934d 100644
--- a/test/Transforms/InstCombine/not.ll
+++ b/test/Transforms/InstCombine/not.ll
@@ -52,3 +52,9 @@ entry:
 	%retval67 = zext i1 %tmp3 to i8		; <i8> [#uses=1]
 	ret i8 %retval67
 }
+
+define <2 x i1> @test7(<2 x i32> %A, <2 x i32> %B) {
+        %cond = icmp sle <2 x i32> %A, %B
+        %Ret = xor <2 x i1> %cond, <i1 true, i1 true>
+        ret <2 x i1> %Ret
+}
diff --git a/test/Transforms/InstCombine/obfuscated_splat.ll b/test/Transforms/InstCombine/obfuscated_splat.ll
index fa9cb423d02c..c37456cc9cfa 100644
--- a/test/Transforms/InstCombine/obfuscated_splat.ll
+++ b/test/Transforms/InstCombine/obfuscated_splat.ll
@@ -1,7 +1,7 @@
 ; RUN: opt -instcombine -S < %s | FileCheck %s
 
 define void @test(<4 x float> *%in_ptr, <4 x float> *%out_ptr) {
-  %A = load <4 x float>* %in_ptr, align 16
+  %A = load <4 x float>, <4 x float>* %in_ptr, align 16
   %B = shufflevector <4 x float> %A, <4 x float> undef, <4 x i32> <i32 0, i32 0, i32 undef, i32 undef>
   %C = shufflevector <4 x float> %B, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 4, i32 undef>
   %D = shufflevector <4 x float> %C, <4 x float> %A, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
diff --git a/test/Transforms/InstCombine/objsize-address-space.ll b/test/Transforms/InstCombine/objsize-address-space.ll
index a971c913bd41..6046dad89790 100644
--- a/test/Transforms/InstCombine/objsize-address-space.ll
+++ b/test/Transforms/InstCombine/objsize-address-space.ll
@@ -21,20 +21,20 @@ declare i16 @llvm.objectsize.i16.p3i8(i8 addrspace(3)*, i1) nounwind readonly
 define i32 @foo_as3() nounwind {
 ; CHECK-LABEL: @foo_as3(
 ; CHECK-NEXT: ret i32 60
-  %1 = call i32 @llvm.objectsize.i32.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8] addrspace(3)* @a_as3, i32 0, i32 0), i1 false)
+  %1 = call i32 @llvm.objectsize.i32.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8], [60 x i8] addrspace(3)* @a_as3, i32 0, i32 0), i1 false)
   ret i32 %1
 }
 
 define i16 @foo_as3_i16() nounwind {
 ; CHECK-LABEL: @foo_as3_i16(
 ; CHECK-NEXT: ret i16 60
-  %1 = call i16 @llvm.objectsize.i16.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8] addrspace(3)* @a_as3, i32 0, i32 0), i1 false)
+  %1 = call i16 @llvm.objectsize.i16.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8], [60 x i8] addrspace(3)* @a_as3, i32 0, i32 0), i1 false)
   ret i16 %1
 }
 
 @a_alias = weak alias [60 x i8] addrspace(3)* @a_as3
 define i32 @foo_alias() nounwind {
-  %1 = call i32 @llvm.objectsize.i32.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8] addrspace(3)* @a_alias, i32 0, i32 0), i1 false)
+  %1 = call i32 @llvm.objectsize.i32.p3i8(i8 addrspace(3)* getelementptr inbounds ([60 x i8], [60 x i8] addrspace(3)* @a_alias, i32 0, i32 0), i1 false)
   ret i32 %1
 }
 
diff --git a/test/Transforms/InstCombine/objsize-noverify.ll b/test/Transforms/InstCombine/objsize-noverify.ll
new file mode 100644
index 000000000000..7e469bd25282
--- /dev/null
+++ b/test/Transforms/InstCombine/objsize-noverify.ll
@@ -0,0 +1,43 @@
+; Test objectsize bounds checking that won't verify until after -instcombine.
+; RUN: opt < %s -disable-verify -instcombine -S | opt -S | FileCheck %s
+; We need target data to get the sizes of the arrays and structures.
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
+
+declare i32 @llvm.objectsize.i32.p0i8(i8*, i1) nounwind readonly
+
+; CHECK-LABEL: @PR13390(
+define i32 @PR13390(i1 %bool, i8* %a) {
+entry:
+  %cond = or i1 %bool, true
+  br i1 %cond, label %return, label %xpto
+
+xpto:
+  %select = select i1 %bool, i8* %select, i8* %a
+  %select2 = select i1 %bool, i8* %a, i8* %select2
+  %0 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %select, i1 true)
+  %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %select2, i1 true)
+  %2 = add i32 %0, %1
+; CHECK: ret i32 undef
+  ret i32 %2
+
+return:
+  ret i32 42
+}
+
+; CHECK-LABEL: @PR13621(
+define i32 @PR13621(i1 %bool) nounwind {
+entry:
+  %cond = or i1 %bool, true
+  br i1 %cond, label %return, label %xpto
+
+; technically reachable, but this malformed IR may appear as a result of constant propagation
+xpto:
+  %gep2 = getelementptr i8, i8* %gep, i32 1
+  %gep = getelementptr i8, i8* %gep2, i32 1
+  %o = call i32 @llvm.objectsize.i32.p0i8(i8* %gep, i1 true)
+; CHECK: ret i32 undef
+  ret i32 %o
+
+return:
+  ret i32 7
+}
diff --git a/test/Transforms/InstCombine/objsize.ll b/test/Transforms/InstCombine/objsize.ll
index 1285b1c3aa6a..335a816e9ece 100644
--- a/test/Transforms/InstCombine/objsize.ll
+++ b/test/Transforms/InstCombine/objsize.ll
@@ -8,7 +8,7 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define i32 @foo() nounwind {
 ; CHECK-LABEL: @foo(
 ; CHECK-NEXT: ret i32 60
-  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i1 false)
+  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i1 false)
   ret i32 %1
 }
 
@@ -16,24 +16,24 @@ define i8* @bar() nounwind {
 ; CHECK-LABEL: @bar(
 entry:
   %retval = alloca i8*
-  %0 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i1 false)
+  %0 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i1 false)
   %cmp = icmp ne i32 %0, -1
 ; CHECK: br i1 true
   br i1 %cmp, label %cond.true, label %cond.false
 
 cond.true:
-  %1 = load i8** %retval
+  %1 = load i8*, i8** %retval
   ret i8* %1
 
 cond.false:
-  %2 = load i8** %retval
+  %2 = load i8*, i8** %retval
   ret i8* %2
 }
 
 define i32 @f() nounwind {
 ; CHECK-LABEL: @f(
 ; CHECK-NEXT: ret i32 0
-  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr ([60 x i8]* @a, i32 1, i32 0), i1 false)
+  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr ([60 x i8], [60 x i8]* @a, i32 1, i32 0), i1 false)
   ret i32 %1
 }
 
@@ -42,7 +42,7 @@ define i32 @f() nounwind {
 define i1 @baz() nounwind {
 ; CHECK-LABEL: @baz(
 ; CHECK-NEXT: objectsize
-  %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([0 x i8]* @window, i32 0, i32 0), i1 false)
+  %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([0 x i8], [0 x i8]* @window, i32 0, i32 0), i1 false)
   %2 = icmp eq i32 %1, -1
   ret i1 %2
 }
@@ -51,7 +51,7 @@ define void @test1(i8* %q, i32 %x) nounwind noinline {
 ; CHECK-LABEL: @test1(
 ; CHECK: objectsize.i32.p0i8
 entry:
-  %0 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([0 x i8]* @window, i32 0, i32 10), i1 false) ; <i64> [#uses=1]
+  %0 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([0 x i8], [0 x i8]* @window, i32 0, i32 10), i1 false) ; <i64> [#uses=1]
   %1 = icmp eq i32 %0, -1                         ; <i1> [#uses=1]
   br i1 %1, label %"47", label %"46"
 
@@ -67,7 +67,7 @@ entry:
 define i32 @test2() nounwind {
 ; CHECK-LABEL: @test2(
 ; CHECK-NEXT: ret i32 34
-  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr (i8* bitcast ([9 x i32]* @.str5 to i8*), i32 2), i1 false)
+  %1 = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr (i8, i8* bitcast ([9 x i32]* @.str5 to i8*), i32 2), i1 false)
   ret i32 %1
 }
 
@@ -86,7 +86,7 @@ entry:
   br i1 undef, label %bb11, label %bb12
 
 bb11:
-  %0 = getelementptr inbounds float* getelementptr inbounds ([480 x float]* @array, i32 0, i32 128), i32 -127 ; <float*> [#uses=1]
+  %0 = getelementptr inbounds float, float* getelementptr inbounds ([480 x float], [480 x float]* @array, i32 0, i32 128), i32 -127 ; <float*> [#uses=1]
   %1 = bitcast float* %0 to i8*                   ; <i8*> [#uses=1]
   %2 = call i32 @llvm.objectsize.i32.p0i8(i8* %1, i1 false) ; <i32> [#uses=1]
   %3 = call i8* @__memcpy_chk(i8* undef, i8* undef, i32 512, i32 %2) nounwind ; <i8*> [#uses=0]
@@ -94,7 +94,7 @@ bb11:
   unreachable
 
 bb12:
-  %4 = getelementptr inbounds float* getelementptr inbounds ([480 x float]* @array, i32 0, i32 128), i32 -127 ; <float*> [#uses=1]
+  %4 = getelementptr inbounds float, float* getelementptr inbounds ([480 x float], [480 x float]* @array, i32 0, i32 128), i32 -127 ; <float*> [#uses=1]
   %5 = bitcast float* %4 to i8*                   ; <i8*> [#uses=1]
   %6 = call i8* @__inline_memcpy_chk(i8* %5, i8* undef, i32 512) nounwind inlinehint ; <i8*> [#uses=0]
 ; CHECK: @__inline_memcpy_chk
@@ -126,7 +126,7 @@ define i8* @test5(i32 %n) nounwind ssp {
 entry:
   %0 = tail call noalias i8* @malloc(i32 20) nounwind
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %0, i1 false)
-  %2 = load i8** @s, align 8
+  %2 = load i8*, i8** @s, align 8
 ; CHECK-NOT: @llvm.objectsize
 ; CHECK: @llvm.memcpy.p0i8.p0i8.i32(i8* %0, i8* %1, i32 10, i32 1, i1 false)
   %3 = tail call i8* @__memcpy_chk(i8* %0, i8* %2, i32 10, i32 %1) nounwind
@@ -138,7 +138,7 @@ define void @test6(i32 %n) nounwind ssp {
 entry:
   %0 = tail call noalias i8* @malloc(i32 20) nounwind
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %0, i1 false)
-  %2 = load i8** @s, align 8
+  %2 = load i8*, i8** @s, align 8
 ; CHECK-NOT: @llvm.objectsize
 ; CHECK: @__memcpy_chk(i8* %0, i8* %1, i32 30, i32 20)
   %3 = tail call i8* @__memcpy_chk(i8* %0, i8* %2, i32 30, i32 %1) nounwind
@@ -153,7 +153,7 @@ define i32 @test7(i8** %esc) {
 ; CHECK-LABEL: @test7(
   %alloc = call noalias i8* @malloc(i32 48) nounwind
   store i8* %alloc, i8** %esc
-  %gep = getelementptr inbounds i8* %alloc, i32 16
+  %gep = getelementptr inbounds i8, i8* %alloc, i32 16
   %objsize = call i32 @llvm.objectsize.i32.p0i8(i8* %gep, i1 false) nounwind readonly
 ; CHECK: ret i32 32
   ret i32 %objsize
@@ -165,7 +165,7 @@ define i32 @test8(i8** %esc) {
 ; CHECK-LABEL: @test8(
   %alloc = call noalias i8* @calloc(i32 5, i32 7) nounwind
   store i8* %alloc, i8** %esc
-  %gep = getelementptr inbounds i8* %alloc, i32 5
+  %gep = getelementptr inbounds i8, i8* %alloc, i32 5
   %objsize = call i32 @llvm.objectsize.i32.p0i8(i8* %gep, i1 false) nounwind readonly
 ; CHECK: ret i32 30
   ret i32 %objsize
@@ -176,7 +176,7 @@ declare noalias i8* @strndup(i8* nocapture, i32) nounwind
 
 ; CHECK-LABEL: @test9(
 define i32 @test9(i8** %esc) {
-  %call = tail call i8* @strdup(i8* getelementptr inbounds ([8 x i8]* @.str, i64 0, i64 0)) nounwind
+  %call = tail call i8* @strdup(i8* getelementptr inbounds ([8 x i8], [8 x i8]* @.str, i64 0, i64 0)) nounwind
   store i8* %call, i8** %esc, align 8
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %call, i1 true)
 ; CHECK: ret i32 8
@@ -185,7 +185,7 @@ define i32 @test9(i8** %esc) {
 
 ; CHECK-LABEL: @test10(
 define i32 @test10(i8** %esc) {
-  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8]* @.str, i64 0, i64 0), i32 3) nounwind
+  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8], [8 x i8]* @.str, i64 0, i64 0), i32 3) nounwind
   store i8* %call, i8** %esc, align 8
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %call, i1 true)
 ; CHECK: ret i32 4
@@ -194,7 +194,7 @@ define i32 @test10(i8** %esc) {
 
 ; CHECK-LABEL: @test11(
 define i32 @test11(i8** %esc) {
-  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8]* @.str, i64 0, i64 0), i32 7) nounwind
+  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8], [8 x i8]* @.str, i64 0, i64 0), i32 7) nounwind
   store i8* %call, i8** %esc, align 8
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %call, i1 true)
 ; CHECK: ret i32 8
@@ -203,7 +203,7 @@ define i32 @test11(i8** %esc) {
 
 ; CHECK-LABEL: @test12(
 define i32 @test12(i8** %esc) {
-  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8]* @.str, i64 0, i64 0), i32 8) nounwind
+  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8], [8 x i8]* @.str, i64 0, i64 0), i32 8) nounwind
   store i8* %call, i8** %esc, align 8
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %call, i1 true)
 ; CHECK: ret i32 8
@@ -212,50 +212,13 @@ define i32 @test12(i8** %esc) {
 
 ; CHECK-LABEL: @test13(
 define i32 @test13(i8** %esc) {
-  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8]* @.str, i64 0, i64 0), i32 57) nounwind
+  %call = tail call i8* @strndup(i8* getelementptr inbounds ([8 x i8], [8 x i8]* @.str, i64 0, i64 0), i32 57) nounwind
   store i8* %call, i8** %esc, align 8
   %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %call, i1 true)
 ; CHECK: ret i32 8
   ret i32 %1
 }
 
-; CHECK-LABEL: @PR13390(
-define i32 @PR13390(i1 %bool, i8* %a) {
-entry:
-  %cond = or i1 %bool, true
-  br i1 %cond, label %return, label %xpto
-
-xpto:
-  %select = select i1 %bool, i8* %select, i8* %a
-  %select2 = select i1 %bool, i8* %a, i8* %select2
-  %0 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %select, i1 true)
-  %1 = tail call i32 @llvm.objectsize.i32.p0i8(i8* %select2, i1 true)
-  %2 = add i32 %0, %1
-; CHECK: ret i32 undef
-  ret i32 %2
-
-return:
-  ret i32 42
-}
-
-; CHECK-LABEL: @PR13621(
-define i32 @PR13621(i1 %bool) nounwind {
-entry:
-  %cond = or i1 %bool, true
-  br i1 %cond, label %return, label %xpto
-
-; technically reachable, but this malformed IR may appear as a result of constant propagation
-xpto:
-  %gep2 = getelementptr i8* %gep, i32 1
-  %gep = getelementptr i8* %gep2, i32 1
-  %o = call i32 @llvm.objectsize.i32.p0i8(i8* %gep, i1 true)
-; CHECK: ret i32 undef
-  ret i32 %o
-
-return:
-  ret i32 7
-}
-
 @globalalias = internal alias [60 x i8]* @a
 
 ; CHECK-LABEL: @test18(
diff --git a/test/Transforms/InstCombine/odr-linkage.ll b/test/Transforms/InstCombine/odr-linkage.ll
index 2ce62468dc77..73675efb08bb 100644
--- a/test/Transforms/InstCombine/odr-linkage.ll
+++ b/test/Transforms/InstCombine/odr-linkage.ll
@@ -6,10 +6,10 @@
 @g4 = internal constant i32 4
 
 define i32 @test() {
-  %A = load i32* @g1
-  %B = load i32* @g2
-  %C = load i32* @g3
-  %D = load i32* @g4
+  %A = load i32, i32* @g1
+  %B = load i32, i32* @g2
+  %C = load i32, i32* @g3
+  %D = load i32, i32* @g4
   
   %a = add i32 %A, %B
   %b = add i32 %a, %C
diff --git a/test/Transforms/InstCombine/or.ll b/test/Transforms/InstCombine/or.ll
index f604bafcc330..b91a5954d97e 100644
--- a/test/Transforms/InstCombine/or.ll
+++ b/test/Transforms/InstCombine/or.ll
@@ -394,7 +394,7 @@ define i32 @test37(i32* %xp, i32 %y) {
 ; CHECK: select i1 %tobool, i32 -1, i32 %x
   %tobool = icmp ne i32 %y, 0
   %sext = sext i1 %tobool to i32
-  %x = load i32* %xp
+  %x = load i32, i32* %xp
   %or = or i32 %sext, %x
   ret i32 %or
 }
@@ -404,7 +404,7 @@ define i32 @test38(i32* %xp, i32 %y) {
 ; CHECK: select i1 %tobool, i32 -1, i32 %x
   %tobool = icmp ne i32 %y, 0
   %sext = sext i1 %tobool to i32
-  %x = load i32* %xp
+  %x = load i32, i32* %xp
   %or = or i32 %x, %sext
   ret i32 %or
 }
diff --git a/test/Transforms/InstCombine/osx-names.ll b/test/Transforms/InstCombine/osx-names.ll
index 926caadc49d8..04d842d0ec28 100644
--- a/test/Transforms/InstCombine/osx-names.ll
+++ b/test/Transforms/InstCombine/osx-names.ll
@@ -16,14 +16,14 @@ target triple = "i386-apple-macosx10.7.2"
 define void @test1(%struct.__sFILE* %stream) nounwind {
 ; CHECK-LABEL: define void @test1(
 ; CHECK: call i32 @"fwrite$UNIX2003"
-  %call = tail call i32 (%struct.__sFILE*, i8*, ...)* @fprintf(%struct.__sFILE* %stream, i8* getelementptr inbounds ([13 x i8]* @.str, i32 0, i32 0)) nounwind
+  %call = tail call i32 (%struct.__sFILE*, i8*, ...) @fprintf(%struct.__sFILE* %stream, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @.str, i32 0, i32 0)) nounwind
   ret void
 }
 
 define void @test2(%struct.__sFILE* %stream, i8* %str) nounwind ssp {
 ; CHECK-LABEL: define void @test2(
 ; CHECK: call i32 @"fputs$UNIX2003"
-  %call = tail call i32 (%struct.__sFILE*, i8*, ...)* @fprintf(%struct.__sFILE* %stream, i8* getelementptr inbounds ([3 x i8]* @.str2, i32 0, i32 0), i8* %str) nounwind
+  %call = tail call i32 (%struct.__sFILE*, i8*, ...) @fprintf(%struct.__sFILE* %stream, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @.str2, i32 0, i32 0), i8* %str) nounwind
   ret void
 }
 
diff --git a/test/Transforms/InstCombine/overflow-mul.ll b/test/Transforms/InstCombine/overflow-mul.ll
index 6d8d40bcac3e..bc0504b0a96d 100644
--- a/test/Transforms/InstCombine/overflow-mul.ll
+++ b/test/Transforms/InstCombine/overflow-mul.ll
@@ -174,6 +174,17 @@ define <4 x i32> @pr20113(<4 x i16> %a, <4 x i16> %b) {
   ret <4 x i32> %vcgez.i
 }
 
+
+; The last test needs this weird datalayout.
+target datalayout = "i32:8:8"
+; Without it, InstCombine will align the pointed on 4 Bytes
+; The KnownBitsZero that result from the alignment allows to
+; turn:
+;    and i32 %mul, 255
+; to:
+;    and i32 %mul, 252
+; The mask is no longer in the form 2^n-1  and this prevents the transformation.
+
 @pr21445_data = external global i32
 define i1 @pr21445(i8 %a) {
 ; CHECK-LABEL: @pr21445(
diff --git a/test/Transforms/InstCombine/overflow.ll b/test/Transforms/InstCombine/overflow.ll
index 3eddc80a7048..4c0a8a2c0df7 100644
--- a/test/Transforms/InstCombine/overflow.ll
+++ b/test/Transforms/InstCombine/overflow.ll
@@ -97,39 +97,6 @@ if.end:                                           ; preds = %entry
 ; CHECK: ret i8
 }
 
-; CHECK-LABEL: @test5(
-; CHECK: llvm.uadd.with.overflow
-; CHECK: ret i64
-define i64 @test5(i64 %a, i64 %b) nounwind ssp {
-entry:
-  %add = add i64 %b, %a
-  %cmp = icmp ult i64 %add, %a
-  %Q = select i1 %cmp, i64 %b, i64 42
-  ret i64 %Q
-}
-
-; CHECK-LABEL: @test6(
-; CHECK: llvm.uadd.with.overflow
-; CHECK: ret i64
-define i64 @test6(i64 %a, i64 %b) nounwind ssp {
-entry:
-  %add = add i64 %b, %a
-  %cmp = icmp ult i64 %add, %b
-  %Q = select i1 %cmp, i64 %b, i64 42
-  ret i64 %Q
-}
-
-; CHECK-LABEL: @test7(
-; CHECK: llvm.uadd.with.overflow
-; CHECK: ret i64
-define i64 @test7(i64 %a, i64 %b) nounwind ssp {
-entry:
-  %add = add i64 %b, %a
-  %cmp = icmp ugt i64 %b, %add
-  %Q = select i1 %cmp, i64 %b, i64 42
-  ret i64 %Q
-}
-
 ; CHECK-LABEL: @test8(
 ; PR11438
 ; This is @test1, but the operands are not sign-extended.  Make sure
diff --git a/test/Transforms/InstCombine/phi-merge-gep.ll b/test/Transforms/InstCombine/phi-merge-gep.ll
index 25c9cea9b770..b548e51a56a0 100644
--- a/test/Transforms/InstCombine/phi-merge-gep.ll
+++ b/test/Transforms/InstCombine/phi-merge-gep.ll
@@ -7,39 +7,39 @@
 
 define void @foo(float* %Ar, float* %Ai, i64 %As, float* %Cr, float* %Ci, i64 %Cs, i64 %n) nounwind {
 entry:
-  %0 = getelementptr inbounds float* %Ar, i64 0   ; <float*> [#uses=1]
-  %1 = getelementptr inbounds float* %Ai, i64 0   ; <float*> [#uses=1]
+  %0 = getelementptr inbounds float, float* %Ar, i64 0   ; <float*> [#uses=1]
+  %1 = getelementptr inbounds float, float* %Ai, i64 0   ; <float*> [#uses=1]
   %2 = mul i64 %n, %As                            ; <i64> [#uses=1]
-  %3 = getelementptr inbounds float* %Ar, i64 %2  ; <float*> [#uses=1]
+  %3 = getelementptr inbounds float, float* %Ar, i64 %2  ; <float*> [#uses=1]
   %4 = mul i64 %n, %As                            ; <i64> [#uses=1]
-  %5 = getelementptr inbounds float* %Ai, i64 %4  ; <float*> [#uses=1]
+  %5 = getelementptr inbounds float, float* %Ai, i64 %4  ; <float*> [#uses=1]
   %6 = mul i64 %n, 2                              ; <i64> [#uses=1]
   %7 = mul i64 %6, %As                            ; <i64> [#uses=1]
-  %8 = getelementptr inbounds float* %Ar, i64 %7  ; <float*> [#uses=1]
+  %8 = getelementptr inbounds float, float* %Ar, i64 %7  ; <float*> [#uses=1]
   %9 = mul i64 %n, 2                              ; <i64> [#uses=1]
   %10 = mul i64 %9, %As                           ; <i64> [#uses=1]
-  %11 = getelementptr inbounds float* %Ai, i64 %10 ; <float*> [#uses=1]
-  %12 = getelementptr inbounds float* %Cr, i64 0  ; <float*> [#uses=1]
-  %13 = getelementptr inbounds float* %Ci, i64 0  ; <float*> [#uses=1]
+  %11 = getelementptr inbounds float, float* %Ai, i64 %10 ; <float*> [#uses=1]
+  %12 = getelementptr inbounds float, float* %Cr, i64 0  ; <float*> [#uses=1]
+  %13 = getelementptr inbounds float, float* %Ci, i64 0  ; <float*> [#uses=1]
   %14 = mul i64 %n, %Cs                           ; <i64> [#uses=1]
-  %15 = getelementptr inbounds float* %Cr, i64 %14 ; <float*> [#uses=1]
+  %15 = getelementptr inbounds float, float* %Cr, i64 %14 ; <float*> [#uses=1]
   %16 = mul i64 %n, %Cs                           ; <i64> [#uses=1]
-  %17 = getelementptr inbounds float* %Ci, i64 %16 ; <float*> [#uses=1]
+  %17 = getelementptr inbounds float, float* %Ci, i64 %16 ; <float*> [#uses=1]
   %18 = mul i64 %n, 2                             ; <i64> [#uses=1]
   %19 = mul i64 %18, %Cs                          ; <i64> [#uses=1]
-  %20 = getelementptr inbounds float* %Cr, i64 %19 ; <float*> [#uses=1]
+  %20 = getelementptr inbounds float, float* %Cr, i64 %19 ; <float*> [#uses=1]
   %21 = mul i64 %n, 2                             ; <i64> [#uses=1]
   %22 = mul i64 %21, %Cs                          ; <i64> [#uses=1]
-  %23 = getelementptr inbounds float* %Ci, i64 %22 ; <float*> [#uses=1]
+  %23 = getelementptr inbounds float, float* %Ci, i64 %22 ; <float*> [#uses=1]
   br label %bb13
 
 bb:                                               ; preds = %bb13
-  %24 = load float* %A0r.0, align 4               ; <float> [#uses=1]
-  %25 = load float* %A0i.0, align 4               ; <float> [#uses=1]
-  %26 = load float* %A1r.0, align 4               ; <float> [#uses=2]
-  %27 = load float* %A1i.0, align 4               ; <float> [#uses=2]
-  %28 = load float* %A2r.0, align 4               ; <float> [#uses=2]
-  %29 = load float* %A2i.0, align 4               ; <float> [#uses=2]
+  %24 = load float, float* %A0r.0, align 4               ; <float> [#uses=1]
+  %25 = load float, float* %A0i.0, align 4               ; <float> [#uses=1]
+  %26 = load float, float* %A1r.0, align 4               ; <float> [#uses=2]
+  %27 = load float, float* %A1i.0, align 4               ; <float> [#uses=2]
+  %28 = load float, float* %A2r.0, align 4               ; <float> [#uses=2]
+  %29 = load float, float* %A2i.0, align 4               ; <float> [#uses=2]
   %30 = fadd float %26, %28                       ; <float> [#uses=2]
   %31 = fadd float %27, %29                       ; <float> [#uses=2]
   %32 = fsub float %26, %28                       ; <float> [#uses=1]
@@ -62,18 +62,18 @@ bb:                                               ; preds = %bb13
   store float %43, float* %C1i.0, align 4
   store float %44, float* %C2r.0, align 4
   store float %45, float* %C2i.0, align 4
-  %46 = getelementptr inbounds float* %A0r.0, i64 %As ; <float*> [#uses=1]
-  %47 = getelementptr inbounds float* %A0i.0, i64 %As ; <float*> [#uses=1]
-  %48 = getelementptr inbounds float* %A1r.0, i64 %As ; <float*> [#uses=1]
-  %49 = getelementptr inbounds float* %A1i.0, i64 %As ; <float*> [#uses=1]
-  %50 = getelementptr inbounds float* %A2r.0, i64 %As ; <float*> [#uses=1]
-  %51 = getelementptr inbounds float* %A2i.0, i64 %As ; <float*> [#uses=1]
-  %52 = getelementptr inbounds float* %C0r.0, i64 %Cs ; <float*> [#uses=1]
-  %53 = getelementptr inbounds float* %C0i.0, i64 %Cs ; <float*> [#uses=1]
-  %54 = getelementptr inbounds float* %C1r.0, i64 %Cs ; <float*> [#uses=1]
-  %55 = getelementptr inbounds float* %C1i.0, i64 %Cs ; <float*> [#uses=1]
-  %56 = getelementptr inbounds float* %C2r.0, i64 %Cs ; <float*> [#uses=1]
-  %57 = getelementptr inbounds float* %C2i.0, i64 %Cs ; <float*> [#uses=1]
+  %46 = getelementptr inbounds float, float* %A0r.0, i64 %As ; <float*> [#uses=1]
+  %47 = getelementptr inbounds float, float* %A0i.0, i64 %As ; <float*> [#uses=1]
+  %48 = getelementptr inbounds float, float* %A1r.0, i64 %As ; <float*> [#uses=1]
+  %49 = getelementptr inbounds float, float* %A1i.0, i64 %As ; <float*> [#uses=1]
+  %50 = getelementptr inbounds float, float* %A2r.0, i64 %As ; <float*> [#uses=1]
+  %51 = getelementptr inbounds float, float* %A2i.0, i64 %As ; <float*> [#uses=1]
+  %52 = getelementptr inbounds float, float* %C0r.0, i64 %Cs ; <float*> [#uses=1]
+  %53 = getelementptr inbounds float, float* %C0i.0, i64 %Cs ; <float*> [#uses=1]
+  %54 = getelementptr inbounds float, float* %C1r.0, i64 %Cs ; <float*> [#uses=1]
+  %55 = getelementptr inbounds float, float* %C1i.0, i64 %Cs ; <float*> [#uses=1]
+  %56 = getelementptr inbounds float, float* %C2r.0, i64 %Cs ; <float*> [#uses=1]
+  %57 = getelementptr inbounds float, float* %C2i.0, i64 %Cs ; <float*> [#uses=1]
   %58 = add nsw i64 %i.0, 1                       ; <i64> [#uses=1]
   br label %bb13
 
diff --git a/test/Transforms/InstCombine/phi.ll b/test/Transforms/InstCombine/phi.ll
index 6e314651383c..54cc4cfe4594 100644
--- a/test/Transforms/InstCombine/phi.ll
+++ b/test/Transforms/InstCombine/phi.ll
@@ -125,11 +125,11 @@ Exit:           ; preds = %Loop
 
 define i32* @test8({ i32, i32 } *%A, i1 %b) {
 BB0:
-        %X = getelementptr inbounds { i32, i32 } *%A, i32 0, i32 1
+        %X = getelementptr inbounds { i32, i32 }, { i32, i32 } *%A, i32 0, i32 1
         br i1 %b, label %BB1, label %BB2
 
 BB1:
-        %Y = getelementptr { i32, i32 } *%A, i32 0, i32 1
+        %Y = getelementptr { i32, i32 }, { i32, i32 } *%A, i32 0, i32 1
         br label %BB2
 
 BB2:
@@ -139,7 +139,7 @@ BB2:
 ; CHECK-LABEL: @test8(
 ; CHECK-NOT: phi
 ; CHECK: BB2:
-; CHECK-NEXT: %B = getelementptr { i32, i32 }* %A 
+; CHECK-NEXT: %B = getelementptr { i32, i32 }, { i32, i32 }* %A 
 ; CHECK-NEXT: ret i32* %B
 }
 
@@ -149,11 +149,11 @@ entry:
   br i1 %c, label %bb1, label %bb
 
 bb:
-  %C = load i32* %B, align 1
+  %C = load i32, i32* %B, align 1
   br label %bb2
 
 bb1:
-  %D = load i32* %A, align 1
+  %D = load i32, i32* %A, align 1
   br label %bb2
 
 bb2:
@@ -162,7 +162,7 @@ bb2:
 ; CHECK-LABEL: @test9(
 ; CHECK:       bb2:
 ; CHECK-NEXT:        phi i32* [ %B, %bb ], [ %A, %bb1 ]
-; CHECK-NEXT:   %E = load i32* %{{[^,]*}}, align 1
+; CHECK-NEXT:   %E = load i32, i32* %{{[^,]*}}, align 1
 ; CHECK-NEXT:   ret i32 %E
 
 }
@@ -173,11 +173,11 @@ entry:
   br i1 %c, label %bb1, label %bb
 
 bb:
-  %C = load i32* %B, align 16
+  %C = load i32, i32* %B, align 16
   br label %bb2
 
 bb1:
-  %D = load i32* %A, align 32
+  %D = load i32, i32* %A, align 32
   br label %bb2
 
 bb2:
@@ -186,7 +186,7 @@ bb2:
 ; CHECK-LABEL: @test10(
 ; CHECK:       bb2:
 ; CHECK-NEXT:        phi i32* [ %B, %bb ], [ %A, %bb1 ]
-; CHECK-NEXT:   %E = load i32* %{{[^,]*}}, align 16
+; CHECK-NEXT:   %E = load i32, i32* %{{[^,]*}}, align 16
 ; CHECK-NEXT:   ret i32 %E
 }
 
@@ -375,30 +375,30 @@ entry:
   store i32 %flag, i32* %flag.addr
   store i32* %pointer2, i32** %pointer2.addr
   store i32 10, i32* %res
-  %tmp = load i32* %flag.addr                     ; <i32> [#uses=1]
+  %tmp = load i32, i32* %flag.addr                     ; <i32> [#uses=1]
   %tobool = icmp ne i32 %tmp, 0                   ; <i1> [#uses=1]
   br i1 %tobool, label %if.then, label %if.else
 
 return:                                           ; preds = %if.end
-  %tmp7 = load i32* %retval                       ; <i32> [#uses=1]
+  %tmp7 = load i32, i32* %retval                       ; <i32> [#uses=1]
   ret i32 %tmp7
 
 if.end:                                           ; preds = %if.else, %if.then
-  %tmp6 = load i32* %res                          ; <i32> [#uses=1]
+  %tmp6 = load i32, i32* %res                          ; <i32> [#uses=1]
   store i32 %tmp6, i32* %retval
   br label %return
 
 if.then:                                          ; preds = %entry
-  %tmp1 = load i32 addrspace(1)** %pointer1.addr  ; <i32 addrspace(1)*>
-  %arrayidx = getelementptr i32 addrspace(1)* %tmp1, i32 0 ; <i32 addrspace(1)*> [#uses=1]
-  %tmp2 = load i32 addrspace(1)* %arrayidx        ; <i32> [#uses=1]
+  %tmp1 = load i32 addrspace(1)*, i32 addrspace(1)** %pointer1.addr  ; <i32 addrspace(1)*>
+  %arrayidx = getelementptr i32, i32 addrspace(1)* %tmp1, i32 0 ; <i32 addrspace(1)*> [#uses=1]
+  %tmp2 = load i32, i32 addrspace(1)* %arrayidx        ; <i32> [#uses=1]
   store i32 %tmp2, i32* %res
   br label %if.end
 
 if.else:                                          ; preds = %entry
-  %tmp3 = load i32** %pointer2.addr               ; <i32*> [#uses=1]
-  %arrayidx4 = getelementptr i32* %tmp3, i32 0    ; <i32*> [#uses=1]
-  %tmp5 = load i32* %arrayidx4                    ; <i32> [#uses=1]
+  %tmp3 = load i32*, i32** %pointer2.addr               ; <i32*> [#uses=1]
+  %arrayidx4 = getelementptr i32, i32* %tmp3, i32 0    ; <i32*> [#uses=1]
+  %tmp5 = load i32, i32* %arrayidx4                    ; <i32> [#uses=1]
   store i32 %tmp5, i32* %res
   br label %if.end
 }
diff --git a/test/Transforms/InstCombine/pr12251.ll b/test/Transforms/InstCombine/pr12251.ll
index 8c382bb104ac..7197bda2e542 100644
--- a/test/Transforms/InstCombine/pr12251.ll
+++ b/test/Transforms/InstCombine/pr12251.ll
@@ -2,13 +2,13 @@
 
 define zeroext i1 @_Z3fooPb(i8* nocapture %x) {
 entry:
-  %a = load i8* %x, align 1, !range !0
+  %a = load i8, i8* %x, align 1, !range !0
   %b = and i8 %a, 1
   %tobool = icmp ne i8 %b, 0
   ret i1 %tobool
 }
 
-; CHECK: %a = load i8* %x, align 1, !range !0
+; CHECK: %a = load i8, i8* %x, align 1, !range !0
 ; CHECK-NEXT: %tobool = icmp ne i8 %a, 0
 ; CHECK-NEXT: ret i1 %tobool
 
diff --git a/test/Transforms/InstCombine/pr21651.ll b/test/Transforms/InstCombine/pr21651.ll
index 914785f329af..d2b8d312d8c6 100644
--- a/test/Transforms/InstCombine/pr21651.ll
+++ b/test/Transforms/InstCombine/pr21651.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -instcombine -S | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 define void @PR21651() {
   switch i2 0, label %out [
     i2 0, label %out
diff --git a/test/Transforms/InstCombine/pr2645-0.ll b/test/Transforms/InstCombine/pr2645-0.ll
index e8aeb2afb9cf..21bfa64a8604 100644
--- a/test/Transforms/InstCombine/pr2645-0.ll
+++ b/test/Transforms/InstCombine/pr2645-0.ll
@@ -6,10 +6,10 @@
 define internal void @""(i8*) {
 ; <label>:1
         bitcast i8* %0 to i32*          ; <i32*>:2 [#uses=1]
-        load i32* %2, align 1           ; <i32>:3 [#uses=1]
-        getelementptr i8* %0, i32 4             ; <i8*>:4 [#uses=1]
+        load i32, i32* %2, align 1           ; <i32>:3 [#uses=1]
+        getelementptr i8, i8* %0, i32 4             ; <i8*>:4 [#uses=1]
         bitcast i8* %4 to i32*          ; <i32*>:5 [#uses=1]
-        load i32* %5, align 1           ; <i32>:6 [#uses=1]
+        load i32, i32* %5, align 1           ; <i32>:6 [#uses=1]
         br label %7
 
 ; <label>:7             ; preds = %9, %1
@@ -22,7 +22,7 @@ define internal void @""(i8*) {
         sitofp i32 %.0 to float         ; <float>:10 [#uses=1]
         insertelement <4 x float> %.01, float %10, i32 0                ; <<4 x float>>:11 [#uses=1]
         shufflevector <4 x float> %11, <4 x float> undef, <4 x i32> zeroinitializer             ; <<4 x float>>:12 [#uses=2]
-        getelementptr i8* %0, i32 48            ; <i8*>:13 [#uses=1]
+        getelementptr i8, i8* %0, i32 48            ; <i8*>:13 [#uses=1]
         bitcast i8* %13 to <4 x float>*         ; <<4 x float>*>:14 [#uses=1]
         store <4 x float> %12, <4 x float>* %14, align 16
         add i32 %.0, 2          ; <i32>:15 [#uses=1]
diff --git a/test/Transforms/InstCombine/pr2645-1.ll b/test/Transforms/InstCombine/pr2645-1.ll
index d320dafcd252..2986d21866bf 100644
--- a/test/Transforms/InstCombine/pr2645-1.ll
+++ b/test/Transforms/InstCombine/pr2645-1.ll
@@ -13,9 +13,9 @@ define internal void @""(i8*, i32, i8*) {
         br i1 %5, label %6, label %20
 
 ; <label>:6             ; preds = %4
-        %7 = getelementptr i8* %2, i32 %.0              ; <i8*> [#uses=1]
+        %7 = getelementptr i8, i8* %2, i32 %.0              ; <i8*> [#uses=1]
         %8 = bitcast i8* %7 to <4 x i16>*               ; <<4 x i16>*> [#uses=1]
-        %9 = load <4 x i16>* %8, align 1                ; <<4 x i16>> [#uses=1]
+        %9 = load <4 x i16>, <4 x i16>* %8, align 1                ; <<4 x i16>> [#uses=1]
         %10 = bitcast <4 x i16> %9 to <1 x i64>         ; <<1 x i64>> [#uses=1]
         %11 = call <2 x i64> @foo(<1 x i64> %10)
 ; <<2 x i64>> [#uses=1]
@@ -24,7 +24,7 @@ define internal void @""(i8*, i32, i8*) {
         %14 = shufflevector <8 x i16> %13, <8 x i16> %13, <8 x i32> < i32 0, i32 0, i32 1, i32 1, i32 2, i32 2, i32 3, i32 3 >          ; <<8 x i16>> [#uses=1]
         %15 = bitcast <8 x i16> %14 to <4 x i32>                ; <<4 x i32>> [#uses=1]
         %16 = sitofp <4 x i32> %15 to <4 x float>               ; <<4 x float>> [#uses=1]
-        %17 = getelementptr i8* %0, i32 %.0             ; <i8*> [#uses=1]
+        %17 = getelementptr i8, i8* %0, i32 %.0             ; <i8*> [#uses=1]
         %18 = bitcast i8* %17 to <4 x float>*           ; <<4 x float>*> [#uses=1]
         store <4 x float> %16, <4 x float>* %18, align 1
         %19 = add i32 %.0, 1            ; <i32> [#uses=1]
diff --git a/test/Transforms/InstCombine/pr2996.ll b/test/Transforms/InstCombine/pr2996.ll
index ff3245d8668c..f5e1df4dd048 100644
--- a/test/Transforms/InstCombine/pr2996.ll
+++ b/test/Transforms/InstCombine/pr2996.ll
@@ -5,7 +5,7 @@ define void @func_53(i16 signext %p_56) nounwind {
 entry:
 	%0 = icmp sgt i16 %p_56, -1		; <i1> [#uses=1]
 	%iftmp.0.0 = select i1 %0, i32 -1, i32 0		; <i32> [#uses=1]
-	%1 = call i32 (...)* @func_4(i32 %iftmp.0.0) nounwind		; <i32> [#uses=0]
+	%1 = call i32 (...) @func_4(i32 %iftmp.0.0) nounwind		; <i32> [#uses=0]
 	ret void
 }
 
diff --git a/test/Transforms/InstCombine/pr8547.ll b/test/Transforms/InstCombine/pr8547.ll
index 7e9cbe17b68d..6d74b4002c11 100644
--- a/test/Transforms/InstCombine/pr8547.ll
+++ b/test/Transforms/InstCombine/pr8547.ll
@@ -21,6 +21,6 @@ for.cond:                                         ; preds = %for.cond, %codeRepl
   br i1 %tobool, label %for.cond, label %codeRepl2
 
 codeRepl2:                                        ; preds = %for.cond
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([10 x i8]* @.str, i64 0, i64 0), i32 %conv2) nounwind
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str, i64 0, i64 0), i32 %conv2) nounwind
   ret i32 0
 }
diff --git a/test/Transforms/InstCombine/printf-1.ll b/test/Transforms/InstCombine/printf-1.ll
index 483bc7a6b079..75e11ce7b7b4 100644
--- a/test/Transforms/InstCombine/printf-1.ll
+++ b/test/Transforms/InstCombine/printf-1.ll
@@ -21,8 +21,8 @@ declare i32 @printf(i8*, ...)
 
 define void @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %fmt = getelementptr [1 x i8]* @empty, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt)
+  %fmt = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt)
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -31,8 +31,8 @@ define void @test_simplify1() {
 
 define void @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %fmt = getelementptr [2 x i8]* @h, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt)
+  %fmt = getelementptr [2 x i8], [2 x i8]* @h, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt)
 ; CHECK-NEXT: call i32 @putchar(i32 104)
   ret void
 ; CHECK-NEXT: ret void
@@ -40,8 +40,8 @@ define void @test_simplify2() {
 
 define void @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %fmt = getelementptr [2 x i8]* @percent, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt)
+  %fmt = getelementptr [2 x i8], [2 x i8]* @percent, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt)
 ; CHECK-NEXT: call i32 @putchar(i32 37)
   ret void
 ; CHECK-NEXT: ret void
@@ -51,9 +51,9 @@ define void @test_simplify3() {
 
 define void @test_simplify4() {
 ; CHECK-LABEL: @test_simplify4(
-  %fmt = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt)
-; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([12 x i8]* [[STR]], i32 0, i32 0))
+  %fmt = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt)
+; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([12 x i8], [12 x i8]* [[STR]], i32 0, i32 0))
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -62,8 +62,8 @@ define void @test_simplify4() {
 
 define void @test_simplify5() {
 ; CHECK-LABEL: @test_simplify5(
-  %fmt = getelementptr [3 x i8]* @percent_c, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt, i8 104)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_c, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt, i8 104)
 ; CHECK-NEXT: call i32 @putchar(i32 104)
   ret void
 ; CHECK-NEXT: ret void
@@ -73,10 +73,10 @@ define void @test_simplify5() {
 
 define void @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %fmt = getelementptr [4 x i8]* @percent_s, i32 0, i32 0
-  %str = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt, i8* %str)
-; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0))
+  %fmt = getelementptr [4 x i8], [4 x i8]* @percent_s, i32 0, i32 0
+  %str = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt, i8* %str)
+; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0))
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -85,35 +85,35 @@ define void @test_simplify6() {
 
 define void @test_simplify7() {
 ; CHECK-IPRINTF-LABEL: @test_simplify7(
-  %fmt = getelementptr [3 x i8]* @percent_d, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt, i32 187)
-; CHECK-IPRINTF-NEXT: call i32 (i8*, ...)* @iprintf(i8* getelementptr inbounds ([3 x i8]* @percent_d, i32 0, i32 0), i32 187)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_d, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt, i32 187)
+; CHECK-IPRINTF-NEXT: call i32 (i8*, ...) @iprintf(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_d, i32 0, i32 0), i32 187)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify1() {
 ; CHECK-IPRINTF-LABEL: @test_no_simplify1(
-  %fmt = getelementptr [3 x i8]* @percent_f, i32 0, i32 0
-  call i32 (i8*, ...)* @printf(i8* %fmt, double 1.87)
-; CHECK-IPRINTF-NEXT: call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_f, i32 0, i32 0
+  call i32 (i8*, ...) @printf(i8* %fmt, double 1.87)
+; CHECK-IPRINTF-NEXT: call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify2(i8* %fmt, double %d) {
 ; CHECK-LABEL: @test_no_simplify2(
-  call i32 (i8*, ...)* @printf(i8* %fmt, double %d)
-; CHECK-NEXT: call i32 (i8*, ...)* @printf(i8* %fmt, double %d)
+  call i32 (i8*, ...) @printf(i8* %fmt, double %d)
+; CHECK-NEXT: call i32 (i8*, ...) @printf(i8* %fmt, double %d)
   ret void
 ; CHECK-NEXT: ret void
 }
 
 define i32 @test_no_simplify3() {
 ; CHECK-LABEL: @test_no_simplify3(
-  %fmt = getelementptr [2 x i8]* @h, i32 0, i32 0
-  %ret = call i32 (i8*, ...)* @printf(i8* %fmt)
-; CHECK-NEXT: call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([2 x i8]* @h, i32 0, i32 0))
+  %fmt = getelementptr [2 x i8], [2 x i8]* @h, i32 0, i32 0
+  %ret = call i32 (i8*, ...) @printf(i8* %fmt)
+; CHECK-NEXT: call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([2 x i8], [2 x i8]* @h, i32 0, i32 0))
   ret i32 %ret
 ; CHECK-NEXT: ret i32 %ret
 }
diff --git a/test/Transforms/InstCombine/printf-2.ll b/test/Transforms/InstCombine/printf-2.ll
index 7e018ebba04d..d6769856e3d8 100644
--- a/test/Transforms/InstCombine/printf-2.ll
+++ b/test/Transforms/InstCombine/printf-2.ll
@@ -14,8 +14,8 @@ declare void @printf(i8*, ...)
 
 define void @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %fmt = getelementptr [2 x i8]* @h, i32 0, i32 0
-  call void (i8*, ...)* @printf(i8* %fmt)
+  %fmt = getelementptr [2 x i8], [2 x i8]* @h, i32 0, i32 0
+  call void (i8*, ...) @printf(i8* %fmt)
 ; CHECK-NEXT: call i32 @putchar(i32 104)
   ret void
 ; CHECK-NEXT: ret void
@@ -23,19 +23,19 @@ define void @test_simplify1() {
 
 define void @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %fmt = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call void (i8*, ...)* @printf(i8* %fmt)
-; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([12 x i8]* @str, i32 0, i32 0))
+  %fmt = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call void (i8*, ...) @printf(i8* %fmt)
+; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([12 x i8], [12 x i8]* @str, i32 0, i32 0))
   ret void
 ; CHECK-NEXT: ret void
 }
 
 define void @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %fmt = getelementptr [4 x i8]* @percent_s, i32 0, i32 0
-  %str = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call void (i8*, ...)* @printf(i8* %fmt, i8* %str)
-; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0))
+  %fmt = getelementptr [4 x i8], [4 x i8]* @percent_s, i32 0, i32 0
+  %str = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call void (i8*, ...) @printf(i8* %fmt, i8* %str)
+; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0))
   ret void
 ; CHECK-NEXT: ret void
 }
diff --git a/test/Transforms/InstCombine/puts-1.ll b/test/Transforms/InstCombine/puts-1.ll
index bd7557ef650f..21028684851d 100644
--- a/test/Transforms/InstCombine/puts-1.ll
+++ b/test/Transforms/InstCombine/puts-1.ll
@@ -12,7 +12,7 @@ declare i32 @puts(i8*)
 
 define void @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %str = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   call i32 @puts(i8* %str)
 ; CHECK-NEXT: call i32 @putchar(i32 10)
   ret void
@@ -23,9 +23,9 @@ define void @test_simplify1() {
 
 define i32 @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %str = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   %ret = call i32 @puts(i8* %str)
-; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([1 x i8]* @empty, i32 0, i32 0))
+; CHECK-NEXT: call i32 @puts(i8* getelementptr inbounds ([1 x i8], [1 x i8]* @empty, i32 0, i32 0))
   ret i32 %ret
 ; CHECK-NEXT: ret i32 %ret
 }
diff --git a/test/Transforms/InstCombine/select-cmp-br.ll b/test/Transforms/InstCombine/select-cmp-br.ll
index f10d58783a77..1dc7e153f5fb 100644
--- a/test/Transforms/InstCombine/select-cmp-br.ll
+++ b/test/Transforms/InstCombine/select-cmp-br.ll
@@ -9,13 +9,13 @@ declare void @foobar()
 
 define void @test1(%C*) {
 entry:
-  %1 = getelementptr inbounds %C* %0, i64 0, i32 0, i32 0
-  %m = load i64** %1, align 8
-  %2 = getelementptr inbounds %C* %0, i64 1, i32 0, i32 0
-  %n = load i64** %2, align 8
-  %3 = getelementptr inbounds i64* %m, i64 9
+  %1 = getelementptr inbounds %C, %C* %0, i64 0, i32 0, i32 0
+  %m = load i64*, i64** %1, align 8
+  %2 = getelementptr inbounds %C, %C* %0, i64 1, i32 0, i32 0
+  %n = load i64*, i64** %2, align 8
+  %3 = getelementptr inbounds i64, i64* %m, i64 9
   %4 = bitcast i64* %3 to i64 (%C*)**
-  %5 = load i64 (%C*)** %4, align 8
+  %5 = load i64 (%C*)*, i64 (%C*)** %4, align 8
   %6 = icmp eq i64* %m, %n
   %7 = select i1 %6, %C* %0, %C* null
   %8 = icmp eq %C* %7, null
@@ -25,7 +25,7 @@ entry:
   ret void
 
 ; <label>:10                                      ; preds = %entry
-  %11 = getelementptr inbounds %C* %7, i64 0, i32 0
+  %11 = getelementptr inbounds %C, %C* %7, i64 0, i32 0
   tail call void @bar(%struct.S* %11)
   br label %9
 
@@ -40,13 +40,13 @@ entry:
 
 define void @test2(%C*) {
 entry:
-  %1 = getelementptr inbounds %C* %0, i64 0, i32 0, i32 0
-  %m = load i64** %1, align 8
-  %2 = getelementptr inbounds %C* %0, i64 1, i32 0, i32 0
-  %n = load i64** %2, align 8
-  %3 = getelementptr inbounds i64* %m, i64 9
+  %1 = getelementptr inbounds %C, %C* %0, i64 0, i32 0, i32 0
+  %m = load i64*, i64** %1, align 8
+  %2 = getelementptr inbounds %C, %C* %0, i64 1, i32 0, i32 0
+  %n = load i64*, i64** %2, align 8
+  %3 = getelementptr inbounds i64, i64* %m, i64 9
   %4 = bitcast i64* %3 to i64 (%C*)**
-  %5 = load i64 (%C*)** %4, align 8
+  %5 = load i64 (%C*)*, i64 (%C*)** %4, align 8
   %6 = icmp eq i64* %m, %n
   %7 = select i1 %6, %C* null, %C* %0
   %8 = icmp eq %C* %7, null
@@ -56,7 +56,7 @@ entry:
   ret void
 
 ; <label>:10                                      ; preds = %entry
-  %11 = getelementptr inbounds %C* %7, i64 0, i32 0
+  %11 = getelementptr inbounds %C, %C* %7, i64 0, i32 0
   tail call void @bar(%struct.S* %11)
   br label %9
 
@@ -71,13 +71,13 @@ entry:
 
 define void @test3(%C*) {
 entry:
-  %1 = getelementptr inbounds %C* %0, i64 0, i32 0, i32 0
-  %m = load i64** %1, align 8
-  %2 = getelementptr inbounds %C* %0, i64 1, i32 0, i32 0
-  %n = load i64** %2, align 8
-  %3 = getelementptr inbounds i64* %m, i64 9
+  %1 = getelementptr inbounds %C, %C* %0, i64 0, i32 0, i32 0
+  %m = load i64*, i64** %1, align 8
+  %2 = getelementptr inbounds %C, %C* %0, i64 1, i32 0, i32 0
+  %n = load i64*, i64** %2, align 8
+  %3 = getelementptr inbounds i64, i64* %m, i64 9
   %4 = bitcast i64* %3 to i64 (%C*)**
-  %5 = load i64 (%C*)** %4, align 8
+  %5 = load i64 (%C*)*, i64 (%C*)** %4, align 8
   %6 = icmp eq i64* %m, %n
   %7 = select i1 %6, %C* %0, %C* null
   %8 = icmp ne %C* %7, null
@@ -87,7 +87,7 @@ entry:
   ret void
 
 ; <label>:10                                      ; preds = %entry
-  %11 = getelementptr inbounds %C* %7, i64 0, i32 0
+  %11 = getelementptr inbounds %C, %C* %7, i64 0, i32 0
   tail call void @bar(%struct.S* %11)
   br label %9
 
@@ -102,13 +102,13 @@ entry:
 
 define void @test4(%C*) {
 entry:
-  %1 = getelementptr inbounds %C* %0, i64 0, i32 0, i32 0
-  %m = load i64** %1, align 8
-  %2 = getelementptr inbounds %C* %0, i64 1, i32 0, i32 0
-  %n = load i64** %2, align 8
-  %3 = getelementptr inbounds i64* %m, i64 9
+  %1 = getelementptr inbounds %C, %C* %0, i64 0, i32 0, i32 0
+  %m = load i64*, i64** %1, align 8
+  %2 = getelementptr inbounds %C, %C* %0, i64 1, i32 0, i32 0
+  %n = load i64*, i64** %2, align 8
+  %3 = getelementptr inbounds i64, i64* %m, i64 9
   %4 = bitcast i64* %3 to i64 (%C*)**
-  %5 = load i64 (%C*)** %4, align 8
+  %5 = load i64 (%C*)*, i64 (%C*)** %4, align 8
   %6 = icmp eq i64* %m, %n
   %7 = select i1 %6, %C* null, %C* %0
   %8 = icmp ne %C* %7, null
@@ -118,7 +118,7 @@ entry:
   ret void
 
 ; <label>:10                                      ; preds = %entry
-  %11 = getelementptr inbounds %C* %7, i64 0, i32 0
+  %11 = getelementptr inbounds %C, %C* %7, i64 0, i32 0
   tail call void @bar(%struct.S* %11)
   br label %9
 
@@ -141,7 +141,7 @@ entry:
   ret void
 
 ; <label>:5                                      ; preds = %entry
-  %6 = getelementptr inbounds %C* %2, i64 0, i32 0
+  %6 = getelementptr inbounds %C, %C* %2, i64 0, i32 0
   tail call void @bar(%struct.S* %6)
   br label %4
 
diff --git a/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll b/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll
new file mode 100644
index 000000000000..894bf6db0a42
--- /dev/null
+++ b/test/Transforms/InstCombine/select-cmp-cttz-ctlz.ll
@@ -0,0 +1,327 @@
+; RUN: opt -instcombine -S < %s | FileCheck %s
+
+; This test is to verify that the instruction combiner is able to fold
+; a cttz/ctlz followed by a icmp + select into a single cttz/ctlz with
+; the 'is_zero_undef' flag cleared.
+
+define i16 @test1(i16 %x) {
+; CHECK-LABEL: @test1(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %x, i1 false)
+; CHECK-NEXT: ret i16 [[VAR]]
+entry:
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i16 %0, i16 16
+  ret i16 %cond
+}
+
+define i32 @test2(i32 %x) {
+; CHECK-LABEL: @test2(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 false)
+; CHECK-NEXT: ret i32 [[VAR]]
+entry:
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i32 %0, i32 32
+  ret i32 %cond
+}
+
+define i64 @test3(i64 %x) {
+; CHECK-LABEL: @test3(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 false)
+; CHECK-NEXT: ret i64 [[VAR]]
+entry:
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i64 %0, i64 64
+  ret i64 %cond
+}
+
+define i16 @test4(i16 %x) {
+; CHECK-LABEL: @test4(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %x, i1 false)
+; CHECK-NEXT: ret i16 [[VAR]]
+entry:
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
+  %tobool = icmp eq i16 %x, 0
+  %cond = select i1 %tobool, i16 16, i16 %0
+  ret i16 %cond
+}
+
+define i32 @test5(i32 %x) {
+; CHECK-LABEL: @test5(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 false)
+; CHECK-NEXT: ret i32 [[VAR]]
+entry:
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %tobool = icmp eq i32 %x, 0
+  %cond = select i1 %tobool, i32 32, i32 %0
+  ret i32 %cond
+}
+
+define i64 @test6(i64 %x) {
+; CHECK-LABEL: @test6(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 false)
+; CHECK-NEXT: ret i64 [[VAR]]
+entry:
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %tobool = icmp eq i64 %x, 0
+  %cond = select i1 %tobool, i64 64, i64 %0
+  ret i64 %cond
+}
+
+define i16 @test1b(i16 %x) {
+; CHECK-LABEL: @test1b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %x, i1 false)
+; CHECK-NEXT: ret i16 [[VAR]]
+entry:
+  %0 = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i16 %0, i16 16
+  ret i16 %cond
+}
+
+define i32 @test2b(i32 %x) {
+; CHECK-LABEL: @test2b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: ret i32 [[VAR]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i32 %0, i32 32
+  ret i32 %cond
+}
+
+define i64 @test3b(i64 %x) {
+; CHECK-LABEL: @test3b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 false)
+; CHECK-NEXT: ret i64 [[VAR]]
+entry:
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i64 %0, i64 64
+  ret i64 %cond
+}
+
+define i16 @test4b(i16 %x) {
+; CHECK-LABEL: @test4b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %x, i1 false)
+; CHECK-NEXT: ret i16 [[VAR]]
+entry:
+  %0 = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
+  %tobool = icmp eq i16 %x, 0
+  %cond = select i1 %tobool, i16 16, i16 %0
+  ret i16 %cond
+}
+
+define i32 @test5b(i32 %x) {
+; CHECK-LABEL: @test5b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: ret i32 [[VAR]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %tobool = icmp eq i32 %x, 0
+  %cond = select i1 %tobool, i32 32, i32 %0
+  ret i32 %cond
+}
+
+define i64 @test6b(i64 %x) {
+; CHECK-LABEL: @test6b(
+; CHECK: [[VAR:%[a-zA-Z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 false)
+; CHECK-NEXT: ret i64 [[VAR]]
+entry:
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %tobool = icmp eq i64 %x, 0
+  %cond = select i1 %tobool, i64 64, i64 %0
+  ret i64 %cond
+}
+
+define i32 @test1c(i16 %x) {
+; CHECK-LABEL: @test1c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i16 [[VAR1]] to i32
+; CHECK-NEXT: ret i32 [[VAR2]]
+entry:
+  %0 = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
+  %cast2 = zext i16 %0 to i32
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i32 %cast2, i32 16
+  ret i32 %cond
+}
+
+define i64 @test2c(i16 %x) {
+; CHECK-LABEL: @test2c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i16 [[VAR1]] to i64
+; CHECK-NEXT: ret i64 [[VAR2]]
+entry:
+  %0 = tail call i16 @llvm.cttz.i16(i16 %x, i1 true)
+  %conv = zext i16 %0 to i64
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i64 %conv, i64 16
+  ret i64 %cond
+}
+
+define i64 @test3c(i32 %x) {
+; CHECK-LABEL: @test3c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i32 [[VAR1]] to i64
+; CHECK-NEXT: ret i64 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %conv = zext i32 %0 to i64
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i64 %conv, i64 32
+  ret i64 %cond
+}
+
+define i32 @test4c(i16 %x) {
+; CHECK-LABEL: @test4c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i16 [[VAR1]] to i32
+; CHECK-NEXT: ret i32 [[VAR2]]
+entry:
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
+  %cast = zext i16 %0 to i32
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i32 %cast, i32 16
+  ret i32 %cond
+}
+
+define i64 @test5c(i16 %x) {
+; CHECK-LABEL: @test5c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i16 [[VAR1]] to i64
+; CHECK-NEXT: ret i64 [[VAR2]]
+entry:
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %x, i1 true)
+  %cast = zext i16 %0 to i64
+  %tobool = icmp ne i16 %x, 0
+  %cond = select i1 %tobool, i64 %cast, i64 16
+  ret i64 %cond
+}
+
+define i64 @test6c(i32 %x) {
+; CHECK-LABEL: @test6c(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i32 [[VAR1]] to i64
+; CHECK-NEXT: ret i64 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %cast = zext i32 %0 to i64
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i64 %cast, i64 32
+  ret i64 %cond
+}
+
+define i16 @test1d(i64 %x) {
+; CHECK-LABEL: @test1d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i64 [[VAR1]] to i16
+; CHECK-NEXT: ret i16 [[VAR2]]
+entry:
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %conv = trunc i64 %0 to i16
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i16 %conv, i16 64
+  ret i16 %cond
+}
+
+define i32 @test2d(i64 %x) {
+; CHECK-LABEL: @test2d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i64 [[VAR1]] to i32
+; CHECK-NEXT: ret i32 [[VAR2]]
+entry:
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i32
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i32 %cast, i32 64
+  ret i32 %cond
+}
+
+define i16 @test3d(i32 %x) {
+; CHECK-LABEL: @test3d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i32 [[VAR1]] to i16
+; CHECK-NEXT: ret i16 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %cast = trunc i32 %0 to i16
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i16 %cast, i16 32
+  ret i16 %cond
+}
+
+define i16 @test4d(i64 %x) {
+; CHECK-LABEL: @test4d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i64 [[VAR1]] to i16
+; CHECK-NEXT: ret i16 [[VAR2]]
+entry:
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i16
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i16 %cast, i16 64
+  ret i16 %cond
+}
+
+define i32 @test5d(i64 %x) {
+; CHECK-LABEL: @test5d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i64 [[VAR1]] to i32
+; CHECK-NEXT: ret i32 [[VAR2]]
+entry:
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i32
+  %tobool = icmp ne i64 %x, 0
+  %cond = select i1 %tobool, i32 %cast, i32 64
+  ret i32 %cond
+}
+
+define i16 @test6d(i32 %x) {
+; CHECK-LABEL: @test6d(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i32 [[VAR1]] to i16
+; CHECK-NEXT: ret i16 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %cast = trunc i32 %0 to i16
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i16 %cast, i16 32
+  ret i16 %cond
+}
+
+define i64 @select_bug1(i32 %x) {
+; CHECK-LABEL: @select_bug1(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = zext i32 [[VAR1]] to i64
+; CHECK-NEXT: ret i64 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+  %conv = zext i32 %0 to i64
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i64 %conv, i64 32
+  ret i64 %cond
+}
+
+define i16 @select_bug2(i32 %x) {
+; CHECK-LABEL: @select_bug2(
+; CHECK: [[VAR1:%[a-zA-Z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+; CHECK-NEXT: [[VAR2:%[a-zA-Z0-9]+]] = trunc i32 [[VAR1]] to i16
+; CHECK-NEXT: ret i16 [[VAR2]]
+entry:
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 false)
+  %conv = trunc i32 %0 to i16
+  %tobool = icmp ne i32 %x, 0
+  %cond = select i1 %tobool, i16 %conv, i16 32
+  ret i16 %cond
+}
+
+
+declare i16 @llvm.ctlz.i16(i16, i1)
+declare i32 @llvm.ctlz.i32(i32, i1)
+declare i64 @llvm.ctlz.i64(i64, i1)
+declare i16 @llvm.cttz.i16(i16, i1)
+declare i32 @llvm.cttz.i32(i32, i1)
+declare i64 @llvm.cttz.i64(i64, i1)
diff --git a/test/Transforms/InstCombine/select-crash-noverify.ll b/test/Transforms/InstCombine/select-crash-noverify.ll
new file mode 100644
index 000000000000..4a366aa8fb83
--- /dev/null
+++ b/test/Transforms/InstCombine/select-crash-noverify.ll
@@ -0,0 +1,19 @@
+; RUN: opt < %s -disable-verify -instcombine -S | opt -S | FileCheck %s
+; Formerly crashed, PR8490.
+
+; CHECK-LABEL: @test3(
+define i32 @test3(i1 %bool, i32 %a) {
+entry:
+  %cond = or i1 %bool, true
+  br i1 %cond, label %return, label %xpto
+
+; technically reachable, but this malformed IR may appear as a result of constant propagation
+xpto:
+  %select = select i1 %bool, i32 %a, i32 %select
+  %select2 = select i1 %bool, i32 %select2, i32 %a
+  %sum = add i32 %select, %select2
+  ret i32 %sum
+
+return:
+  ret i32 7
+}
diff --git a/test/Transforms/InstCombine/select-crash.ll b/test/Transforms/InstCombine/select-crash.ll
index 77446cd8ba02..991635b63aa8 100644
--- a/test/Transforms/InstCombine/select-crash.ll
+++ b/test/Transforms/InstCombine/select-crash.ll
@@ -30,20 +30,3 @@ define <4 x float> @foo(i1 %b, <4 x float> %x, <4 x float> %y, <4 x float> %z) {
   %sel = select i1 %b, <4 x float> %a, <4 x float> %sub 
   ret <4 x float> %sel
 }
-
-; CHECK-LABEL: @test3(
-define i32 @test3(i1 %bool, i32 %a) {
-entry:
-  %cond = or i1 %bool, true
-  br i1 %cond, label %return, label %xpto
-
-; technically reachable, but this malformed IR may appear as a result of constant propagation
-xpto:
-  %select = select i1 %bool, i32 %a, i32 %select
-  %select2 = select i1 %bool, i32 %select2, i32 %a
-  %sum = add i32 %select, %select2
-  ret i32 %sum
-
-return:
-  ret i32 7
-}
diff --git a/test/Transforms/InstCombine/select-load-call.ll b/test/Transforms/InstCombine/select-load-call.ll
index b63468de53e1..ad0ef4f85eb2 100644
--- a/test/Transforms/InstCombine/select-load-call.ll
+++ b/test/Transforms/InstCombine/select-load-call.ll
@@ -10,6 +10,6 @@ define i32 @test(i1 %cond, i32 *%P) {
   call void @test2() readonly
 
   %P2 = select i1 %cond, i32 *%P, i32* %A
-  %V = load i32* %P2
+  %V = load i32, i32* %P2
   ret i32 %V
 }
diff --git a/test/Transforms/InstCombine/select-select.ll b/test/Transforms/InstCombine/select-select.ll
index 65820acf07b8..768d1c47c209 100644
--- a/test/Transforms/InstCombine/select-select.ll
+++ b/test/Transforms/InstCombine/select-select.ll
@@ -21,4 +21,14 @@ define float @foo2(float %a) #0 {
   ret float %f
 }
 
+; CHECK-LABEL: @foo3
+define <2 x i32> @foo3(<2 x i1> %vec_bool, i1 %bool, <2 x i32> %V) {
+; CHECK: %[[sel0:.*]] = select <2 x i1> %vec_bool, <2 x i32> zeroinitializer, <2 x i32> %V
+; CHECK: %[[sel1:.*]] = select i1 %bool, <2 x i32> %[[sel0]], <2 x i32> %V
+; CHECK: ret <2 x i32> %[[sel1]]
+  %sel0 = select <2 x i1> %vec_bool, <2 x i32> zeroinitializer, <2 x i32> %V
+  %sel1 = select i1 %bool, <2 x i32> %sel0, <2 x i32> %V
+  ret <2 x i32> %sel1
+}
+
 attributes #0 = { nounwind readnone ssp uwtable }
diff --git a/test/Transforms/InstCombine/select.ll b/test/Transforms/InstCombine/select.ll
index 054f00a0de49..e4bc96cff176 100644
--- a/test/Transforms/InstCombine/select.ll
+++ b/test/Transforms/InstCombine/select.ll
@@ -301,30 +301,30 @@ define i32 @test15j(i32 %X) {
 
 define i32 @test16(i1 %C, i32* %P) {
         %P2 = select i1 %C, i32* %P, i32* null          
-        %V = load i32* %P2              
+        %V = load i32, i32* %P2              
         ret i32 %V
 ; CHECK-LABEL: @test16(
-; CHECK-NEXT: %V = load i32* %P
+; CHECK-NEXT: %V = load i32, i32* %P
 ; CHECK: ret i32 %V
 }
 
 ;; It may be legal to load from a null address in a non-zero address space
 define i32 @test16_neg(i1 %C, i32 addrspace(1)* %P) {
         %P2 = select i1 %C, i32 addrspace(1)* %P, i32 addrspace(1)* null
-        %V = load i32 addrspace(1)* %P2
+        %V = load i32, i32 addrspace(1)* %P2
         ret i32 %V
 ; CHECK-LABEL: @test16_neg
 ; CHECK-NEXT: %P2 = select i1 %C, i32 addrspace(1)* %P, i32 addrspace(1)* null
-; CHECK-NEXT: %V = load i32 addrspace(1)* %P2
+; CHECK-NEXT: %V = load i32, i32 addrspace(1)* %P2
 ; CHECK: ret i32 %V
 }
 define i32 @test16_neg2(i1 %C, i32 addrspace(1)* %P) {
         %P2 = select i1 %C, i32 addrspace(1)* null, i32 addrspace(1)* %P
-        %V = load i32 addrspace(1)* %P2
+        %V = load i32, i32 addrspace(1)* %P2
         ret i32 %V
 ; CHECK-LABEL: @test16_neg2
 ; CHECK-NEXT: %P2 = select i1 %C, i32 addrspace(1)* null, i32 addrspace(1)* %P
-; CHECK-NEXT: %V = load i32 addrspace(1)* %P2
+; CHECK-NEXT: %V = load i32, i32 addrspace(1)* %P2
 ; CHECK: ret i32 %V
 }
 
@@ -423,11 +423,11 @@ jump:
   %c = or i1 false, false
   br label %ret 
 ret:
-  %a = phi i1 [true, %jump], [%c, %entry]
-  %b = select i1 %a, i32 10, i32 20
+  %a = phi i1 [true, %entry], [%c, %jump]
+  %b = select i1 %a, i32 20, i32 10
   ret i32 %b
 ; CHECK-LABEL: @test26(
-; CHECK: %a = phi i32 [ 10, %jump ], [ 20, %entry ]
+; CHECK: %a = phi i32 [ 20, %entry ], [ 10, %jump ]
 ; CHECK-NEXT: ret i32 %a
 }
 
@@ -813,7 +813,7 @@ define i32 @test59(i32 %x, i32 %y) nounwind {
 
 define i1 @test60(i32 %x, i1* %y) nounwind {
   %cmp = icmp eq i32 %x, 0
-  %load = load i1* %y, align 1
+  %load = load i1, i1* %y, align 1
   %cmp1 = icmp slt i32 %x, 1
   %sel = select i1 %cmp, i1 %load, i1 %cmp1
   ret i1 %sel
@@ -823,7 +823,7 @@ define i1 @test60(i32 %x, i1* %y) nounwind {
 
 @glbl = constant i32 10
 define i32 @test61(i32* %ptr) {
-  %A = load i32* %ptr
+  %A = load i32, i32* %ptr
   %B = icmp eq i32* %ptr, @glbl
   %C = select i1 %B, i32 %A, i32 10
   ret i32 %C
@@ -1249,11 +1249,11 @@ define i32 @test76(i1 %flag, i32* %x) {
 ; CHECK-LABEL: @test76(
 ; CHECK: store i32 0, i32* %x
 ; CHECK: %[[P:.*]] = select i1 %flag, i32* @under_aligned, i32* %x
-; CHECK: load i32* %[[P]]
+; CHECK: load i32, i32* %[[P]]
 
   store i32 0, i32* %x
   %p = select i1 %flag, i32* @under_aligned, i32* %x
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -1268,13 +1268,13 @@ define i32 @test77(i1 %flag, i32* %x) {
 ; CHECK: call void @scribble_on_i32(i32* %[[A]])
 ; CHECK: store i32 0, i32* %x
 ; CHECK: %[[P:.*]] = select i1 %flag, i32* %[[A]], i32* %x
-; CHECK: load i32* %[[P]]
+; CHECK: load i32, i32* %[[P]]
 
   %under_aligned = alloca i32, align 1
   call void @scribble_on_i32(i32* %under_aligned)
   store i32 0, i32* %x
   %p = select i1 %flag, i32* %under_aligned, i32* %x
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -1282,8 +1282,8 @@ define i32 @test78(i1 %flag, i32* %x, i32* %y, i32* %z) {
 ; Test that we can speculate the loads around the select even when we can't
 ; fold the load completely away.
 ; CHECK-LABEL: @test78(
-; CHECK:         %[[V1:.*]] = load i32* %x
-; CHECK-NEXT:    %[[V2:.*]] = load i32* %y
+; CHECK:         %[[V1:.*]] = load i32, i32* %x
+; CHECK-NEXT:    %[[V2:.*]] = load i32, i32* %y
 ; CHECK-NEXT:    %[[S:.*]] = select i1 %flag, i32 %[[V1]], i32 %[[V2]]
 ; CHECK-NEXT:    ret i32 %[[S]]
 entry:
@@ -1292,7 +1292,7 @@ entry:
   ; Block forwarding by storing to %z which could alias either %x or %y.
   store i32 42, i32* %z
   %p = select i1 %flag, i32* %x, i32* %y
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -1300,8 +1300,8 @@ define float @test79(i1 %flag, float* %x, i32* %y, i32* %z) {
 ; Test that we can speculate the loads around the select even when we can't
 ; fold the load completely away.
 ; CHECK-LABEL: @test79(
-; CHECK:         %[[V1:.*]] = load float* %x
-; CHECK-NEXT:    %[[V2:.*]] = load float* %y
+; CHECK:         %[[V1:.*]] = load float, float* %x
+; CHECK-NEXT:    %[[V2:.*]] = load float, float* %y
 ; CHECK-NEXT:    %[[S:.*]] = select i1 %flag, float %[[V1]], float %[[V2]]
 ; CHECK-NEXT:    ret float %[[S]]
 entry:
@@ -1312,7 +1312,7 @@ entry:
   ; Block forwarding by storing to %z which could alias either %x or %y.
   store i32 42, i32* %z
   %p = select i1 %flag, float* %x, float* %y1
-  %v = load float* %p
+  %v = load float, float* %p
   ret float %v
 }
 
@@ -1322,7 +1322,7 @@ define i32 @test80(i1 %flag) {
 ; CHECK-LABEL: @test80(
 ; CHECK:         %[[X:.*]] = alloca i32
 ; CHECK-NEXT:    %[[Y:.*]] = alloca i32
-; CHECK:         %[[V:.*]] = load i32* %[[X]]
+; CHECK:         %[[V:.*]] = load i32, i32* %[[X]]
 ; CHECK-NEXT:    store i32 %[[V]], i32* %[[Y]]
 ; CHECK-NEXT:    ret i32 %[[V]]
 entry:
@@ -1330,10 +1330,10 @@ entry:
   %y = alloca i32
   call void @scribble_on_i32(i32* %x)
   call void @scribble_on_i32(i32* %y)
-  %tmp = load i32* %x
+  %tmp = load i32, i32* %x
   store i32 %tmp, i32* %y
   %p = select i1 %flag, i32* %x, i32* %y
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -1343,7 +1343,7 @@ define float @test81(i1 %flag) {
 ; CHECK-LABEL: @test81(
 ; CHECK:         %[[X:.*]] = alloca i32
 ; CHECK-NEXT:    %[[Y:.*]] = alloca i32
-; CHECK:         %[[V:.*]] = load i32* %[[X]]
+; CHECK:         %[[V:.*]] = load i32, i32* %[[X]]
 ; CHECK-NEXT:    store i32 %[[V]], i32* %[[Y]]
 ; CHECK-NEXT:    %[[C:.*]] = bitcast i32 %[[V]] to float
 ; CHECK-NEXT:    ret float %[[C]]
@@ -1354,10 +1354,10 @@ entry:
   %y1 = bitcast i32* %y to float*
   call void @scribble_on_i32(i32* %x1)
   call void @scribble_on_i32(i32* %y)
-  %tmp = load i32* %x1
+  %tmp = load i32, i32* %x1
   store i32 %tmp, i32* %y
   %p = select i1 %flag, float* %x, float* %y1
-  %v = load float* %p
+  %v = load float, float* %p
   ret float %v
 }
 
@@ -1369,7 +1369,7 @@ define i32 @test82(i1 %flag) {
 ; CHECK-NEXT:    %[[Y:.*]] = alloca i32
 ; CHECK-NEXT:    %[[X1:.*]] = bitcast float* %[[X]] to i32*
 ; CHECK-NEXT:    %[[Y1:.*]] = bitcast i32* %[[Y]] to float*
-; CHECK:         %[[V:.*]] = load float* %[[X]]
+; CHECK:         %[[V:.*]] = load float, float* %[[X]]
 ; CHECK-NEXT:    store float %[[V]], float* %[[Y1]]
 ; CHECK-NEXT:    %[[C:.*]] = bitcast float %[[V]] to i32
 ; CHECK-NEXT:    ret i32 %[[C]]
@@ -1380,10 +1380,10 @@ entry:
   %y1 = bitcast i32* %y to float*
   call void @scribble_on_i32(i32* %x1)
   call void @scribble_on_i32(i32* %y)
-  %tmp = load float* %x
+  %tmp = load float, float* %x
   store float %tmp, float* %y1
   %p = select i1 %flag, i32* %x1, i32* %y
-  %v = load i32* %p
+  %v = load i32, i32* %p
   ret i32 %v
 }
 
@@ -1398,7 +1398,7 @@ define i8* @test83(i1 %flag) {
 ; CHECK-NEXT:    %[[Y:.*]] = alloca i8*
 ; CHECK-DAG:     %[[X2:.*]] = bitcast i8** %[[X]] to i64*
 ; CHECK-DAG:     %[[Y2:.*]] = bitcast i8** %[[Y]] to i64*
-; CHECK:         %[[V:.*]] = load i64* %[[X2]]
+; CHECK:         %[[V:.*]] = load i64, i64* %[[X2]]
 ; CHECK-NEXT:    store i64 %[[V]], i64* %[[Y2]]
 ; CHECK-NEXT:    %[[C:.*]] = inttoptr i64 %[[V]] to i8*
 ; CHECK-NEXT:    ret i8* %[[S]]
@@ -1409,10 +1409,10 @@ entry:
   %y1 = bitcast i64* %y to i8**
   call void @scribble_on_i64(i64* %x1)
   call void @scribble_on_i64(i64* %y)
-  %tmp = load i64* %x1
+  %tmp = load i64, i64* %x1
   store i64 %tmp, i64* %y
   %p = select i1 %flag, i8** %x, i8** %y1
-  %v = load i8** %p
+  %v = load i8*, i8** %p
   ret i8* %v
 }
 
@@ -1422,7 +1422,7 @@ define i64 @test84(i1 %flag) {
 ; CHECK-LABEL: @test84(
 ; CHECK:         %[[X:.*]] = alloca i8*
 ; CHECK-NEXT:    %[[Y:.*]] = alloca i8*
-; CHECK:         %[[V:.*]] = load i8** %[[X]]
+; CHECK:         %[[V:.*]] = load i8*, i8** %[[X]]
 ; CHECK-NEXT:    store i8* %[[V]], i8** %[[Y]]
 ; CHECK-NEXT:    %[[C:.*]] = ptrtoint i8* %[[V]] to i64
 ; CHECK-NEXT:    ret i64 %[[C]]
@@ -1433,10 +1433,10 @@ entry:
   %y1 = bitcast i64* %y to i8**
   call void @scribble_on_i64(i64* %x1)
   call void @scribble_on_i64(i64* %y)
-  %tmp = load i8** %x
+  %tmp = load i8*, i8** %x
   store i8* %tmp, i8** %y1
   %p = select i1 %flag, i64* %x1, i64* %y
-  %v = load i64* %p
+  %v = load i64, i64* %p
   ret i64 %v
 }
 
@@ -1445,10 +1445,10 @@ define i8* @test85(i1 %flag) {
 ; pointer doesn't load all of the stored integer bits. We could fix this, but it
 ; would require endianness checks and other nastiness.
 ; CHECK-LABEL: @test85(
-; CHECK:         %[[T:.*]] = load i128*
+; CHECK:         %[[T:.*]] = load i128, i128*
 ; CHECK-NEXT:    store i128 %[[T]], i128*
-; CHECK-NEXT:    %[[X:.*]] = load i8**
-; CHECK-NEXT:    %[[Y:.*]] = load i8**
+; CHECK-NEXT:    %[[X:.*]] = load i8*, i8**
+; CHECK-NEXT:    %[[Y:.*]] = load i8*, i8**
 ; CHECK-NEXT:    %[[V:.*]] = select i1 %flag, i8* %[[X]], i8* %[[Y]]
 ; CHECK-NEXT:    ret i8* %[[V]]
 entry:
@@ -1459,10 +1459,10 @@ entry:
   %y1 = bitcast i128* %y to i8**
   call void @scribble_on_i128(i128* %x2)
   call void @scribble_on_i128(i128* %y)
-  %tmp = load i128* %x2
+  %tmp = load i128, i128* %x2
   store i128 %tmp, i128* %y
   %p = select i1 %flag, i8** %x1, i8** %y1
-  %v = load i8** %p
+  %v = load i8*, i8** %p
   ret i8* %v
 }
 
@@ -1472,10 +1472,10 @@ define i128 @test86(i1 %flag) {
 ; the bits of the integer.
 ;
 ; CHECK-LABEL: @test86(
-; CHECK:         %[[T:.*]] = load i8**
+; CHECK:         %[[T:.*]] = load i8*, i8**
 ; CHECK-NEXT:    store i8* %[[T]], i8**
-; CHECK-NEXT:    %[[X:.*]] = load i128*
-; CHECK-NEXT:    %[[Y:.*]] = load i128*
+; CHECK-NEXT:    %[[X:.*]] = load i128, i128*
+; CHECK-NEXT:    %[[Y:.*]] = load i128, i128*
 ; CHECK-NEXT:    %[[V:.*]] = select i1 %flag, i128 %[[X]], i128 %[[Y]]
 ; CHECK-NEXT:    ret i128 %[[V]]
 entry:
@@ -1486,9 +1486,49 @@ entry:
   %y1 = bitcast i128* %y to i8**
   call void @scribble_on_i128(i128* %x2)
   call void @scribble_on_i128(i128* %y)
-  %tmp = load i8** %x1
+  %tmp = load i8*, i8** %x1
   store i8* %tmp, i8** %y1
   %p = select i1 %flag, i128* %x2, i128* %y
-  %v = load i128* %p
+  %v = load i128, i128* %p
   ret i128 %v
 }
+
+define i32 @test_select_select0(i32 %a, i32 %r0, i32 %r1, i32 %v1, i32 %v2) {
+  ; CHECK-LABEL: @test_select_select0(
+  ; CHECK: %[[C0:.*]] = icmp sge i32 %a, %v1
+  ; CHECK-NEXT: %[[C1:.*]] = icmp slt i32 %a, %v2
+  ; CHECK-NEXT: %[[C:.*]] = and i1 %[[C1]], %[[C0]]
+  ; CHECK-NEXT: %[[SEL:.*]] = select i1 %[[C]], i32 %r0, i32 %r1
+  ; CHECK-NEXT: ret i32 %[[SEL]]
+  %c0 = icmp sge i32 %a, %v1
+  %s0 = select i1 %c0, i32 %r0, i32 %r1
+  %c1 = icmp slt i32 %a, %v2
+  %s1 = select i1 %c1, i32 %s0, i32 %r1
+  ret i32 %s1
+}
+
+define i32 @test_select_select1(i32 %a, i32 %r0, i32 %r1, i32 %v1, i32 %v2) {
+  ; CHECK-LABEL: @test_select_select1(
+  ; CHECK: %[[C0:.*]] = icmp sge i32 %a, %v1
+  ; CHECK-NEXT: %[[C1:.*]] = icmp slt i32 %a, %v2
+  ; CHECK-NEXT: %[[C:.*]] = or i1 %[[C1]], %[[C0]]
+  ; CHECK-NEXT: %[[SEL:.*]] = select i1 %[[C]], i32 %r0, i32 %r1
+  ; CHECK-NEXT: ret i32 %[[SEL]]
+  %c0 = icmp sge i32 %a, %v1
+  %s0 = select i1 %c0, i32 %r0, i32 %r1
+  %c1 = icmp slt i32 %a, %v2
+  %s1 = select i1 %c1, i32 %r0, i32 %s0
+  ret i32 %s1
+}
+
+define i32 @test_max_of_min(i32 %a) {
+; MAX(MIN(%a, -1), -1) == -1
+; CHECK-LABEL: @test_max_of_min(
+; CHECK: ret i32 -1
+  %not_a = xor i32 %a, -1
+  %c0 = icmp sgt i32 %a, 0
+  %s0 = select i1 %c0, i32 %not_a, i32 -1
+  %c1 = icmp sgt i32 %s0, -1
+  %s1 = select i1 %c1, i32 %s0, i32 -1
+  ret i32 %s1
+}
diff --git a/test/Transforms/InstCombine/shift.ll b/test/Transforms/InstCombine/shift.ll
index 0b5b5deb68c5..79c2ae28105e 100644
--- a/test/Transforms/InstCombine/shift.ll
+++ b/test/Transforms/InstCombine/shift.ll
@@ -575,7 +575,7 @@ entry:
 ; CHECK: %0 = shl i8 %tmp4, 2
 ; CHECK: %tmp54 = and i8 %0, 16
   %tmp55 = xor i8 %tmp54, %tmp51
-; CHECK: ret i8 %tmp551
+; CHECK: ret i8 %tmp55.1
   ret i8 %tmp55
 }
 
@@ -743,7 +743,7 @@ define i32 @test57(i32 %x) {
   %or = or i32 %shl, 7
   ret i32 %or
 ; CHECK-LABEL: @test57(
-; CHECK: %shl = shl i32 %shr1, 4
+; CHECK: %shl = shl i32 %shr.1, 4
 }
 
 
diff --git a/test/Transforms/InstCombine/shufflemask-undef.ll b/test/Transforms/InstCombine/shufflemask-undef.ll
index aa6baa989bfc..10509a92941b 100644
--- a/test/Transforms/InstCombine/shufflemask-undef.ll
+++ b/test/Transforms/InstCombine/shufflemask-undef.ll
@@ -69,8 +69,8 @@ target triple = "i386-apple-darwin9"
 
 define i32 @foo(%struct.State* %dst, <4 x float>* %prgrm, <4 x float>** %buffs, %struct._VMConstants* %cnstn, %struct.PPStreamToken* %pstrm, %struct.PluginBufferData* %gpctx, %struct.VMTextures* %txtrs, %struct.VMGPStack* %gpstk) nounwind {
 bb266.i:
-	getelementptr <4 x float>* null, i32 11		; <<4 x float>*>:0 [#uses=1]
-	load <4 x float>* %0, align 16		; <<4 x float>>:1 [#uses=1]
+	getelementptr <4 x float>, <4 x float>* null, i32 11		; <<4 x float>*>:0 [#uses=1]
+	load <4 x float>, <4 x float>* %0, align 16		; <<4 x float>>:1 [#uses=1]
 	shufflevector <4 x float> %1, <4 x float> undef, <4 x i32> < i32 0, i32 1, i32 1, i32 1 >		; <<4 x float>>:2 [#uses=1]
 	shufflevector <4 x float> %2, <4 x float> undef, <4 x i32> < i32 0, i32 4, i32 1, i32 5 >		; <<4 x float>>:3 [#uses=1]
 	shufflevector <4 x float> undef, <4 x float> undef, <4 x i32> < i32 0, i32 4, i32 1, i32 5 >		; <<4 x float>>:4 [#uses=1]
diff --git a/test/Transforms/InstCombine/signed-comparison.ll b/test/Transforms/InstCombine/signed-comparison.ll
index ab0e7e7ff9fe..922f4dcb2d42 100644
--- a/test/Transforms/InstCombine/signed-comparison.ll
+++ b/test/Transforms/InstCombine/signed-comparison.ll
@@ -12,8 +12,8 @@ entry:
 bb:
 	%indvar = phi i64 [ 0, %entry ], [ %indvar.next, %bb ]
 	%t0 = and i64 %indvar, 65535
-	%t1 = getelementptr double* %p, i64 %t0
-	%t2 = load double* %t1, align 8
+	%t1 = getelementptr double, double* %p, i64 %t0
+	%t2 = load double, double* %t1, align 8
 	%t3 = fmul double %t2, 2.2
 	store double %t3, double* %t1, align 8
 	%i.04 = trunc i64 %indvar to i16
diff --git a/test/Transforms/InstCombine/simplify-demanded-bits-pointer.ll b/test/Transforms/InstCombine/simplify-demanded-bits-pointer.ll
index 6d2193fe448a..db8f17917d1e 100644
--- a/test/Transforms/InstCombine/simplify-demanded-bits-pointer.ll
+++ b/test/Transforms/InstCombine/simplify-demanded-bits-pointer.ll
@@ -29,7 +29,7 @@ bb95:		; preds = %bb88
 	unreachable
 
 bb107:		; preds = %bb88
-	%0 = load i16* undef, align 8		; <i16> [#uses=1]
+	%0 = load i16, i16* undef, align 8		; <i16> [#uses=1]
 	%1 = icmp eq i16 %0, 38		; <i1> [#uses=1]
 	%src_eqv_here.0 = select i1 %1, %struct.rtx_def* null, %struct.rtx_def* null		; <%struct.rtx_def*> [#uses=1]
 	br i1 undef, label %bb127, label %bb125
diff --git a/test/Transforms/InstCombine/simplify-libcalls.ll b/test/Transforms/InstCombine/simplify-libcalls.ll
index fae3e6e3d9be..7cc671061379 100644
--- a/test/Transforms/InstCombine/simplify-libcalls.ll
+++ b/test/Transforms/InstCombine/simplify-libcalls.ll
@@ -1,11 +1,12 @@
 ; RUN: opt -S < %s -instcombine | FileCheck %s
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-f80:128:128-v64:64:64-v128:128:128-a0:0:64-f80:32:32-n8:16:32-S32"
 
 @G = constant [3 x i8] c"%s\00"		; <[3 x i8]*> [#uses=1]
 
 declare i32 @sprintf(i8*, i8*, ...)
 
 define void @foo(i8* %P, i32* %X) {
-	call i32 (i8*, i8*, ...)* @sprintf( i8* %P, i8* getelementptr ([3 x i8]* @G, i32 0, i32 0), i32* %X )		; <i32>:1 [#uses=0]
+	call i32 (i8*, i8*, ...) @sprintf( i8* %P, i8* getelementptr ([3 x i8], [3 x i8]* @G, i32 0, i32 0), i32* %X )		; <i32>:1 [#uses=0]
 	ret void
 }
 
@@ -15,26 +16,26 @@ define void @foo(i8* %P, i32* %X) {
 @str2 = internal constant [5 x i8] c"Ponk\00"
 
 define i8* @test1() {
-        %tmp3 = tail call i8* @strchr( i8* getelementptr ([5 x i8]* @str, i32 0, i32 2), i32 103 )              ; <i8*> [#uses=1]
+        %tmp3 = tail call i8* @strchr( i8* getelementptr ([5 x i8], [5 x i8]* @str, i32 0, i32 2), i32 103 )              ; <i8*> [#uses=1]
         ret i8* %tmp3
 
 ; CHECK-LABEL: @test1(
-; CHECK: ret i8* getelementptr inbounds ([5 x i8]* @str, i32 0, i64 3)
+; CHECK: ret i8* getelementptr inbounds ([5 x i8], [5 x i8]* @str, i32 0, i32 3)
 }
 
 declare i8* @strchr(i8*, i32)
 
 define i8* @test2() {
-        %tmp3 = tail call i8* @strchr( i8* getelementptr ([8 x i8]* @str1, i32 0, i32 2), i32 0 )               ; <i8*> [#uses=1]
+        %tmp3 = tail call i8* @strchr( i8* getelementptr ([8 x i8], [8 x i8]* @str1, i32 0, i32 2), i32 0 )               ; <i8*> [#uses=1]
         ret i8* %tmp3
 
 ; CHECK-LABEL: @test2(
-; CHECK: ret i8* getelementptr inbounds ([8 x i8]* @str1, i32 0, i64 7)
+; CHECK: ret i8* getelementptr inbounds ([8 x i8], [8 x i8]* @str1, i32 0, i32 7)
 }
 
 define i8* @test3() {
 entry:
-        %tmp3 = tail call i8* @strchr( i8* getelementptr ([5 x i8]* @str2, i32 0, i32 1), i32 80 )              ; <i8*> [#uses=1]
+        %tmp3 = tail call i8* @strchr( i8* getelementptr ([5 x i8], [5 x i8]* @str2, i32 0, i32 1), i32 80 )              ; <i8*> [#uses=1]
         ret i8* %tmp3
 
 ; CHECK-LABEL: @test3(
@@ -47,8 +48,8 @@ declare i32 @memcmp(i8*, i8*, i32) nounwind readonly
 
 define i1 @PR2341(i8** %start_addr) {
 entry:
-	%tmp4 = load i8** %start_addr, align 4		; <i8*> [#uses=1]
-	%tmp5 = call i32 @memcmp( i8* %tmp4, i8* getelementptr ([5 x i8]* @_2E_str, i32 0, i32 0), i32 4 ) nounwind readonly 		; <i32> [#uses=1]
+	%tmp4 = load i8*, i8** %start_addr, align 4		; <i8*> [#uses=1]
+	%tmp5 = call i32 @memcmp( i8* %tmp4, i8* getelementptr ([5 x i8], [5 x i8]* @_2E_str, i32 0, i32 0), i32 4 ) nounwind readonly 		; <i32> [#uses=1]
 	%tmp6 = icmp eq i32 %tmp5, 0		; <i1> [#uses=1]
 	ret i1 %tmp6
 
@@ -80,7 +81,7 @@ entry:
 define i32 @PR4641(i32 %argc, i8** %argv) nounwind {
 entry:
 	call void @exit(i32 0) nounwind
-	%cond392 = select i1 undef, i8* getelementptr ([2 x i8]* @.str13, i32 0, i32 0), i8* getelementptr ([2 x i8]* @.str14, i32 0, i32 0)		; <i8*> [#uses=1]
+	%cond392 = select i1 undef, i8* getelementptr ([2 x i8], [2 x i8]* @.str13, i32 0, i32 0), i8* getelementptr ([2 x i8], [2 x i8]* @.str14, i32 0, i32 0)		; <i8*> [#uses=1]
 	%call393 = call %struct.__sFILE* @fopen(i8* undef, i8* %cond392) nounwind		; <%struct.__sFILE*> [#uses=0]
 	unreachable
 }
@@ -114,11 +115,11 @@ for.end:		; preds = %for.cond20
 @hello_u = constant [8 x i8] c"hello_u\00"		; <[8 x i8]*> [#uses=1]
 
 define i32 @MemCpy() {
-  %h_p = getelementptr [2 x i8]* @h, i32 0, i32 0
-  %hel_p = getelementptr [4 x i8]* @hel, i32 0, i32 0
-  %hello_u_p = getelementptr [8 x i8]* @hello_u, i32 0, i32 0
+  %h_p = getelementptr [2 x i8], [2 x i8]* @h, i32 0, i32 0
+  %hel_p = getelementptr [4 x i8], [4 x i8]* @hel, i32 0, i32 0
+  %hello_u_p = getelementptr [8 x i8], [8 x i8]* @hello_u, i32 0, i32 0
   %target = alloca [1024 x i8]
-  %target_p = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  %target_p = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %h_p, i32 2, i32 2, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %hel_p, i32 4, i32 4, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %target_p, i8* %hello_u_p, i32 8, i32 8, i1 false)
diff --git a/test/Transforms/InstCombine/sincospi.ll b/test/Transforms/InstCombine/sincospi.ll
index 739827f1962a..f49fb35cb76a 100644
--- a/test/Transforms/InstCombine/sincospi.ll
+++ b/test/Transforms/InstCombine/sincospi.ll
@@ -18,17 +18,17 @@ declare double @__cospi(double %x) #0
 @var64 = global double 0.0
 
 define float @test_instbased_f32() {
-       %val = load float* @var32
+       %val = load float, float* @var32
        %sin = call float @__sinpif(float %val) #0
        %cos = call float @__cospif(float %val) #0
        %res = fadd float %sin, %cos
        ret float %res
-; CHECK-FLOAT-IN-VEC: [[VAL:%[a-z0-9]+]] = load float* @var32
+; CHECK-FLOAT-IN-VEC: [[VAL:%[a-z0-9]+]] = load float, float* @var32
 ; CHECK-FLOAT-IN-VEC: [[SINCOS:%[a-z0-9]+]] = call <2 x float> @__sincospif_stret(float [[VAL]])
 ; CHECK-FLOAT-IN-VEC: extractelement <2 x float> [[SINCOS]], i32 0
 ; CHECK-FLOAT-IN-VEC: extractelement <2 x float> [[SINCOS]], i32 1
 
-; CHECK: [[VAL:%[a-z0-9]+]] = load float* @var32
+; CHECK: [[VAL:%[a-z0-9]+]] = load float, float* @var32
 ; CHECK: [[SINCOS:%[a-z0-9]+]] = call { float, float } @__sincospif_stret(float [[VAL]])
 ; CHECK: extractvalue { float, float } [[SINCOS]], 0
 ; CHECK: extractvalue { float, float } [[SINCOS]], 1
@@ -55,17 +55,17 @@ define float @test_constant_f32() {
 }
 
 define double @test_instbased_f64() {
-       %val = load double* @var64
+       %val = load double, double* @var64
        %sin = call double @__sinpi(double %val) #0
        %cos = call double @__cospi(double %val) #0
        %res = fadd double %sin, %cos
        ret double %res
-; CHECK-FLOAT-IN-VEC: [[VAL:%[a-z0-9]+]] = load double* @var64
+; CHECK-FLOAT-IN-VEC: [[VAL:%[a-z0-9]+]] = load double, double* @var64
 ; CHECK-FLOAT-IN-VEC: [[SINCOS:%[a-z0-9]+]] = call { double, double } @__sincospi_stret(double [[VAL]])
 ; CHECK-FLOAT-IN-VEC: extractvalue { double, double } [[SINCOS]], 0
 ; CHECK-FLOAT-IN-VEC: extractvalue { double, double } [[SINCOS]], 1
 
-; CHECK: [[VAL:%[a-z0-9]+]] = load double* @var64
+; CHECK: [[VAL:%[a-z0-9]+]] = load double, double* @var64
 ; CHECK: [[SINCOS:%[a-z0-9]+]] = call { double, double } @__sincospi_stret(double [[VAL]])
 ; CHECK: extractvalue { double, double } [[SINCOS]], 0
 ; CHECK: extractvalue { double, double } [[SINCOS]], 1
diff --git a/test/Transforms/InstCombine/sitofp.ll b/test/Transforms/InstCombine/sitofp.ll
index bd31b89f7879..820977838836 100644
--- a/test/Transforms/InstCombine/sitofp.ll
+++ b/test/Transforms/InstCombine/sitofp.ll
@@ -1,28 +1,40 @@
-; RUN: opt < %s -instcombine -S | not grep itofp
+; RUN: opt < %s -instcombine -S | FileCheck %s
 
+; CHECK-LABEL: test1
+; CHECK: ret i1 true
 define i1 @test1(i8 %A) {
   %B = sitofp i8 %A to double
   %C = fcmp ult double %B, 128.0
-  ret i1 %C  ;  True!
+  ret i1 %C
 }
+
+; CHECK-LABEL: test2
+; CHECK: ret i1 true
 define i1 @test2(i8 %A) {
   %B = sitofp i8 %A to double
   %C = fcmp ugt double %B, -128.1
-  ret i1 %C  ;  True!
+  ret i1 %C
 }
 
+; CHECK-LABEL: test3
+; CHECK: ret i1 true
 define i1 @test3(i8 %A) {
   %B = sitofp i8 %A to double
   %C = fcmp ule double %B, 127.0
-  ret i1 %C  ;  true!
+  ret i1 %C
 }
 
+; CHECK-LABEL: test4
+; CHECK: icmp ne i8 %A, 127
+; CHECK-NEXT: ret i1
 define i1 @test4(i8 %A) {
   %B = sitofp i8 %A to double
   %C = fcmp ult double %B, 127.0
-  ret i1 %C  ;  A != 127
+  ret i1 %C
 }
 
+; CHECK-LABEL: test5
+; CHECK: ret i32
 define i32 @test5(i32 %A) {
   %B = sitofp i32 %A to double
   %C = fptosi double %B to i32
@@ -31,25 +43,142 @@ define i32 @test5(i32 %A) {
   ret i32 %E
 }
 
+; CHECK-LABEL: test6
+; CHECK: and i32 %A, 39
+; CHECK-NEXT: ret i32
 define i32 @test6(i32 %A) {
-	%B = and i32 %A, 7		; <i32> [#uses=1]
-	%C = and i32 %A, 32		; <i32> [#uses=1]
-	%D = sitofp i32 %B to double		; <double> [#uses=1]
-	%E = sitofp i32 %C to double		; <double> [#uses=1]
-	%F = fadd double %D, %E		; <double> [#uses=1]
-	%G = fptosi double %F to i32		; <i32> [#uses=1]
-	ret i32 %G
+  %B = and i32 %A, 7
+  %C = and i32 %A, 32
+  %D = sitofp i32 %B to double
+  %E = sitofp i32 %C to double
+  %F = fadd double %D, %E
+  %G = fptosi double %F to i32
+  ret i32 %G
+}
+
+; CHECK-LABEL: test7
+; CHECK: ret i32
+define i32 @test7(i32 %A) nounwind {
+  %B = sitofp i32 %A to double
+  %C = fptoui double %B to i32
+  ret i32 %C
+}
+
+; CHECK-LABEL: test8
+; CHECK: ret i32
+define i32 @test8(i32 %A) nounwind {
+  %B = uitofp i32 %A to double
+  %C = fptosi double %B to i32
+  ret i32 %C
+}
+
+; CHECK-LABEL: test9
+; CHECK: zext i8
+; CHECK-NEXT: ret i32
+define i32 @test9(i8 %A) nounwind {
+  %B = sitofp i8 %A to float
+  %C = fptoui float %B to i32
+  ret i32 %C
+}
+
+; CHECK-LABEL: test10
+; CHECK: sext i8
+; CHECK-NEXT: ret i32
+define i32 @test10(i8 %A) nounwind {
+  %B = sitofp i8 %A to float
+  %C = fptosi float %B to i32
+  ret i32 %C
+}
+
+; If the input value is outside of the range of the output cast, it's
+; undefined behavior, so we can assume it fits.
+; CHECK-LABEL: test11
+; CHECK: trunc
+; CHECK-NEXT: ret i8
+define i8 @test11(i32 %A) nounwind {
+  %B = sitofp i32 %A to float
+  %C = fptosi float %B to i8
+  ret i8 %C
+}
+
+; If the input value is negative, it'll be outside the range of the
+; output cast, and thus undefined behavior.
+; CHECK-LABEL: test12
+; CHECK: zext i8
+; CHECK-NEXT: ret i32
+define i32 @test12(i8 %A) nounwind {
+  %B = sitofp i8 %A to float
+  %C = fptoui float %B to i32
+  ret i32 %C
+}
+
+; This can't fold because the 25-bit input doesn't fit in the mantissa.
+; CHECK-LABEL: test13
+; CHECK: uitofp
+; CHECK-NEXT: fptoui
+define i32 @test13(i25 %A) nounwind {
+  %B = uitofp i25 %A to float
+  %C = fptoui float %B to i32
+  ret i32 %C
+}
+
+; But this one can.
+; CHECK-LABEL: test14
+; CHECK: zext i24
+; CHECK-NEXT: ret i32
+define i32 @test14(i24 %A) nounwind {
+  %B = uitofp i24 %A to float
+  %C = fptoui float %B to i32
+  ret i32 %C
+}
+
+; And this one can too.
+; CHECK-LABEL: test15
+; CHECK: trunc i32
+; CHECK-NEXT: ret i24
+define i24 @test15(i32 %A) nounwind {
+  %B = uitofp i32 %A to float
+  %C = fptoui float %B to i24
+  ret i24 %C
+}
+
+; This can fold because the 25-bit input is signed and we disard the sign bit.
+; CHECK-LABEL: test16
+; CHECK: zext
+define i32 @test16(i25 %A) nounwind {
+ %B = sitofp i25 %A to float
+ %C = fptoui float %B to i32
+ ret i32 %C
+}
+
+; This can't fold because the 26-bit input won't fit the mantissa
+; even after disarding the signed bit.
+; CHECK-LABEL: test17
+; CHECK: sitofp
+; CHECK-NEXT: fptoui
+define i32 @test17(i26 %A) nounwind {
+ %B = sitofp i26 %A to float
+ %C = fptoui float %B to i32
+ ret i32 %C
 }
 
-define i32 @test7(i32 %a) nounwind {
-	%b = sitofp i32 %a to double		; <double> [#uses=1]
-	%c = fptoui double %b to i32		; <i32> [#uses=1]
-	ret i32 %c
+; This can fold because the 54-bit output is signed and we disard the sign bit.
+; CHECK-LABEL: test18
+; CHECK: trunc
+define i54 @test18(i64 %A) nounwind {
+ %B = sitofp i64 %A to double
+ %C = fptosi double %B to i54
+ ret i54 %C
 }
 
-define i32 @test8(i32 %a) nounwind {
-	%b = uitofp i32 %a to double		; <double> [#uses=1]
-	%c = fptosi double %b to i32		; <i32> [#uses=1]
-	ret i32 %c
+; This can't fold because the 55-bit output won't fit the mantissa
+; even after disarding the sign bit.
+; CHECK-LABEL: test19
+; CHECK: sitofp
+; CHECK-NEXT: fptosi
+define i55 @test19(i64 %A) nounwind {
+ %B = sitofp i64 %A to double
+ %C = fptosi double %B to i55
+ ret i55 %C
 }
 
diff --git a/test/Transforms/InstCombine/sprintf-1.ll b/test/Transforms/InstCombine/sprintf-1.ll
index afa38f35ae83..ddf2f2f88e73 100644
--- a/test/Transforms/InstCombine/sprintf-1.ll
+++ b/test/Transforms/InstCombine/sprintf-1.ll
@@ -20,17 +20,17 @@ declare i32 @sprintf(i8*, i8*, ...)
 
 define void @test_simplify1(i8* %dst) {
 ; CHECK-LABEL: @test_simplify1(
-  %fmt = getelementptr [13 x i8]* @hello_world, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt)
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* getelementptr inbounds ([13 x i8]* @hello_world, i32 0, i32 0), i32 13, i32 1, i1 false)
+  %fmt = getelementptr [13 x i8], [13 x i8]* @hello_world, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* getelementptr inbounds ([13 x i8], [13 x i8]* @hello_world, i32 0, i32 0), i32 13, i32 1, i1 false)
   ret void
 ; CHECK-NEXT: ret void
 }
 
 define void @test_simplify2(i8* %dst) {
 ; CHECK-LABEL: @test_simplify2(
-  %fmt = getelementptr [1 x i8]* @null, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt)
+  %fmt = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt)
 ; CHECK-NEXT: store i8 0, i8* %dst, align 1
   ret void
 ; CHECK-NEXT: ret void
@@ -38,8 +38,8 @@ define void @test_simplify2(i8* %dst) {
 
 define void @test_simplify3(i8* %dst) {
 ; CHECK-LABEL: @test_simplify3(
-  %fmt = getelementptr [7 x i8]* @null_hello, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt)
+  %fmt = getelementptr [7 x i8], [7 x i8]* @null_hello, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt)
 ; CHECK-NEXT: store i8 0, i8* %dst, align 1
   ret void
 ; CHECK-NEXT: ret void
@@ -49,10 +49,10 @@ define void @test_simplify3(i8* %dst) {
 
 define void @test_simplify4(i8* %dst) {
 ; CHECK-LABEL: @test_simplify4(
-  %fmt = getelementptr [3 x i8]* @percent_c, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, i8 104)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_c, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, i8 104)
 ; CHECK-NEXT: store i8 104, i8* %dst, align 1
-; CHECK-NEXT: [[NUL:%[a-z0-9]+]] = getelementptr i8* %dst, i32 1
+; CHECK-NEXT: [[NUL:%[a-z0-9]+]] = getelementptr i8, i8* %dst, i32 1
 ; CHECK-NEXT: store i8 0, i8* [[NUL]], align 1
   ret void
 ; CHECK-NEXT: ret void
@@ -62,8 +62,8 @@ define void @test_simplify4(i8* %dst) {
 
 define void @test_simplify5(i8* %dst, i8* %str) {
 ; CHECK-LABEL: @test_simplify5(
-  %fmt = getelementptr [3 x i8]* @percent_s, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, i8* %str)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_s, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, i8* %str)
 ; CHECK-NEXT: [[STRLEN:%[a-z0-9]+]] = call i32 @strlen(i8* %str)
 ; CHECK-NEXT: [[LENINC:%[a-z0-9]+]] = add i32 [[STRLEN]], 1
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %str, i32 [[LENINC]], i32 1, i1 false)
@@ -75,26 +75,26 @@ define void @test_simplify5(i8* %dst, i8* %str) {
 
 define void @test_simplify6(i8* %dst) {
 ; CHECK-IPRINTF-LABEL: @test_simplify6(
-  %fmt = getelementptr [3 x i8]* @percent_d, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, i32 187)
-; CHECK-IPRINTF-NEXT: call i32 (i8*, i8*, ...)* @siprintf(i8* %dst, i8* getelementptr inbounds ([3 x i8]* @percent_d, i32 0, i32 0), i32 187)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_d, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, i32 187)
+; CHECK-IPRINTF-NEXT: call i32 (i8*, i8*, ...) @siprintf(i8* %dst, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_d, i32 0, i32 0), i32 187)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify1(i8* %dst) {
 ; CHECK-IPRINTF-LABEL: @test_no_simplify1(
-  %fmt = getelementptr [3 x i8]* @percent_f, i32 0, i32 0
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, double 1.87)
-; CHECK-IPRINTF-NEXT: call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* getelementptr inbounds ([3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
+  %fmt = getelementptr [3 x i8], [3 x i8]* @percent_f, i32 0, i32 0
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, double 1.87)
+; CHECK-IPRINTF-NEXT: call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* getelementptr inbounds ([3 x i8], [3 x i8]* @percent_f, i32 0, i32 0), double 1.870000e+00)
   ret void
 ; CHECK-IPRINTF-NEXT: ret void
 }
 
 define void @test_no_simplify2(i8* %dst, i8* %fmt, double %d) {
 ; CHECK-LABEL: @test_no_simplify2(
-  call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, double %d)
-; CHECK-NEXT: call i32 (i8*, i8*, ...)* @sprintf(i8* %dst, i8* %fmt, double %d)
+  call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, double %d)
+; CHECK-NEXT: call i32 (i8*, i8*, ...) @sprintf(i8* %dst, i8* %fmt, double %d)
   ret void
 ; CHECK-NEXT: ret void
 }
diff --git a/test/Transforms/InstCombine/sqrt.ll b/test/Transforms/InstCombine/sqrt.ll
index 650b10c6e966..bf44e4f841ce 100644
--- a/test/Transforms/InstCombine/sqrt.ll
+++ b/test/Transforms/InstCombine/sqrt.ll
@@ -38,13 +38,13 @@ entry:
 ; CHECK: sqrt(
 ; CHECK-NOT: sqrtf(
 ; CHECK: fptrunc
-  %arrayidx13 = getelementptr inbounds float* %v, i64 2
-  %tmp14 = load float* %arrayidx13
+  %arrayidx13 = getelementptr inbounds float, float* %v, i64 2
+  %tmp14 = load float, float* %arrayidx13
   %mul18 = fmul float %tmp14, %tmp14
   %add19 = fadd float undef, %mul18
   %conv = fpext float %add19 to double
   %call34 = call double @sqrt(double %conv) readnone
-  %call36 = call i32 (double)* @foo(double %call34) nounwind
+  %call36 = call i32 (double) @foo(double %call34) nounwind
   %conv38 = fptrunc double %call34 to float
   ret float %conv38
 }
diff --git a/test/Transforms/InstCombine/srem1.ll b/test/Transforms/InstCombine/srem1.ll
index f18690c3fff2..31452d8f208a 100644
--- a/test/Transforms/InstCombine/srem1.ll
+++ b/test/Transforms/InstCombine/srem1.ll
@@ -5,7 +5,7 @@
 
 define i32 @func_56(i32 %p_58, i32 %p_59, i32 %p_61, i16 signext %p_62) nounwind {
 entry:
-	%call = call i32 (...)* @rshift_s_s( i32 %p_61, i32 1 )		; <i32> [#uses=1]
+	%call = call i32 (...) @rshift_s_s( i32 %p_61, i32 1 )		; <i32> [#uses=1]
 	%conv = sext i32 %call to i64		; <i64> [#uses=1]
 	%or = or i64 -1734012817166602727, %conv		; <i64> [#uses=1]
 	%rem = srem i64 %or, 1		; <i64> [#uses=1]
diff --git a/test/Transforms/InstCombine/stack-overalign.ll b/test/Transforms/InstCombine/stack-overalign.ll
index 80c2ee88f340..a8f086e6cb93 100644
--- a/test/Transforms/InstCombine/stack-overalign.ll
+++ b/test/Transforms/InstCombine/stack-overalign.ll
@@ -18,8 +18,8 @@
 define void @foo() nounwind {
 entry:
   %src = alloca [1024 x i8], align 1
-  %src1 = getelementptr [1024 x i8]* %src, i32 0, i32 0
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([1024 x i8]* @dst, i32 0, i32 0), i8* %src1, i32 1024, i32 1, i1 false)
+  %src1 = getelementptr [1024 x i8], [1024 x i8]* %src, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([1024 x i8], [1024 x i8]* @dst, i32 0, i32 0), i8* %src1, i32 1024, i32 1, i1 false)
   call void @frob(i8* %src1) nounwind
   ret void
 }
diff --git a/test/Transforms/InstCombine/stacksaverestore.ll b/test/Transforms/InstCombine/stacksaverestore.ll
index f5c7a6f9f3f9..de400e91a751 100644
--- a/test/Transforms/InstCombine/stacksaverestore.ll
+++ b/test/Transforms/InstCombine/stacksaverestore.ll
@@ -31,7 +31,7 @@ bb:		; preds = %bb, %bb.preheader
 	%i.0.reg2mem.0 = phi i32 [ 0, %bb.preheader ], [ %indvar.next, %bb ]		; <i32> [#uses=2]
 	%tmp = call i8* @llvm.stacksave( )		; <i8*> [#uses=1]
 	%tmp23 = alloca i8, i32 %size		; <i8*> [#uses=2]
-	%tmp27 = getelementptr i8* %tmp23, i32 %tmp25		; <i8*> [#uses=1]
+	%tmp27 = getelementptr i8, i8* %tmp23, i32 %tmp25		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp27, align 1
 	%tmp28 = call i8* @llvm.stacksave( )		; <i8*> [#uses=1]
 	%tmp52 = alloca i8, i32 %size		; <i8*> [#uses=1]
diff --git a/test/Transforms/InstCombine/statepoint.ll b/test/Transforms/InstCombine/statepoint.ll
index bee219db9c12..f904f207bfdc 100644
--- a/test/Transforms/InstCombine/statepoint.ll
+++ b/test/Transforms/InstCombine/statepoint.ll
@@ -5,10 +5,10 @@
 
 declare void @func()
 
-define i1 @test_negative(i32 addrspace(1)* %p) {
+define i1 @test_negative(i32 addrspace(1)* %p) gc "statepoint-example" {
 entry:
-  %safepoint_token = tail call i32 (void ()*, i32, i32, ...)* @llvm.experimental.gc.statepoint.p0f_isVoidf(void ()* @func, i32 0, i32 0, i32 0, i32 addrspace(1)* %p)
-  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token, i32 4, i32 4)
+  %safepoint_token = tail call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @func, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* %p)
+  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token,  i32 7, i32 7)
   %cmp = icmp eq i32 addrspace(1)* %pnew, null
   ret i1 %cmp
 ; CHECK-LABEL: test_negative
@@ -16,20 +16,20 @@ entry:
 ; CHECK: ret i1 %cmp
 }
 
-define i1 @test_nonnull(i32 addrspace(1)* nonnull %p) {
+define i1 @test_nonnull(i32 addrspace(1)* nonnull %p) gc "statepoint-example" {
 entry:
-  %safepoint_token = tail call i32 (void ()*, i32, i32, ...)* @llvm.experimental.gc.statepoint.p0f_isVoidf(void ()* @func, i32 0, i32 0, i32 0, i32 addrspace(1)* %p)
-  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token, i32 4, i32 4)
+  %safepoint_token = tail call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @func, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* %p)
+  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token,  i32 7, i32 7)
   %cmp = icmp eq i32 addrspace(1)* %pnew, null
   ret i1 %cmp
 ; CHECK-LABEL: test_nonnull
 ; CHECK: ret i1 false
 }
 
-define i1 @test_null() {
+define i1 @test_null() gc "statepoint-example" {
 entry:
-  %safepoint_token = tail call i32 (void ()*, i32, i32, ...)* @llvm.experimental.gc.statepoint.p0f_isVoidf(void ()* @func, i32 0, i32 0, i32 0, i32 addrspace(1)* null)
-  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token, i32 4, i32 4)
+  %safepoint_token = tail call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @func, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* null)
+  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token,  i32 7, i32 7)
   %cmp = icmp eq i32 addrspace(1)* %pnew, null
   ret i1 %cmp
 ; CHECK-LABEL: test_null
@@ -37,10 +37,10 @@ entry:
 ; CHECK: ret i1 true
 }
 
-define i1 @test_undef() {
+define i1 @test_undef() gc "statepoint-example" {
 entry:
-  %safepoint_token = tail call i32 (void ()*, i32, i32, ...)* @llvm.experimental.gc.statepoint.p0f_isVoidf(void ()* @func, i32 0, i32 0, i32 0, i32 addrspace(1)* undef)
-  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token, i32 4, i32 4)
+  %safepoint_token = tail call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @func, i32 0, i32 0, i32 0, i32 0, i32 addrspace(1)* undef)
+  %pnew = call i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32 %safepoint_token,  i32 7, i32 7)
   %cmp = icmp eq i32 addrspace(1)* %pnew, null
   ret i1 %cmp
 ; CHECK-LABEL: test_undef
@@ -48,5 +48,5 @@ entry:
 ; CHECK: ret i1 undef
 }
 
-declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(void ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
 declare i32 addrspace(1)* @llvm.experimental.gc.relocate.p1i32(i32, i32, i32) #3
diff --git a/test/Transforms/InstCombine/store.ll b/test/Transforms/InstCombine/store.ll
index 0bb175923e98..5dfbd7140901 100644
--- a/test/Transforms/InstCombine/store.ll
+++ b/test/Transforms/InstCombine/store.ll
@@ -12,7 +12,7 @@ define void @test1(i32* %P) {
 }
 
 define void @test2(i32* %P) {
-        %X = load i32* %P               ; <i32> [#uses=1]
+        %X = load i32, i32* %P               ; <i32> [#uses=1]
         %Y = add i32 %X, 0              ; <i32> [#uses=1]
         store i32 %Y, i32* %P
         ret void
@@ -36,12 +36,12 @@ Cond2:
 	br label %Cont
 
 Cont:
-	%V = load i32* %A
+	%V = load i32, i32* %A
 	ret i32 %V
 ; CHECK-LABEL: @test3(
 ; CHECK-NOT: alloca
 ; CHECK: Cont:
-; CHECK-NEXT:  %storemerge = phi i32 [ 47, %Cond2 ], [ -987654321, %Cond ]
+; CHECK-NEXT:  %storemerge = phi i32 [ -987654321, %Cond ], [ 47, %Cond2 ]
 ; CHECK-NEXT:  ret i32 %storemerge
 }
 
@@ -56,7 +56,7 @@ Cond:
         br label %Cont
 
 Cont:
-	%V = load i32* %A
+	%V = load i32, i32* %A
 	ret i32 %V
 ; CHECK-LABEL: @test4(
 ; CHECK-NOT: alloca
@@ -92,15 +92,15 @@ entry:
 
 for.cond:                                         ; preds = %for.body, %entry
   %storemerge = phi i32 [ 0, %entry ], [ %inc, %for.body ]
-  %0 = load i32* %gi, align 4, !tbaa !0
+  %0 = load i32, i32* %gi, align 4, !tbaa !0
   %cmp = icmp slt i32 %0, %n
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
   %idxprom = sext i32 %0 to i64
-  %arrayidx = getelementptr inbounds float* %a, i64 %idxprom
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %idxprom
   store float 0.000000e+00, float* %arrayidx, align 4, !tbaa !3
-  %1 = load i32* %gi, align 4, !tbaa !0
+  %1 = load i32, i32* %gi, align 4, !tbaa !0
   %inc = add nsw i32 %1, 1
   store i32 %inc, i32* %gi, align 4, !tbaa !0
   br label %for.cond
diff --git a/test/Transforms/InstCombine/stpcpy-1.ll b/test/Transforms/InstCombine/stpcpy-1.ll
index b918c9e9e890..cc8289937078 100644
--- a/test/Transforms/InstCombine/stpcpy-1.ll
+++ b/test/Transforms/InstCombine/stpcpy-1.ll
@@ -14,31 +14,31 @@ declare i8* @stpcpy(i8*, i8*)
 define i8* @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   %ret = call i8* @stpcpy(i8* %dst, i8* %src)
 ; CHECK: @llvm.memcpy.p0i8.p0i8.i32
-; CHECK-NEXT: getelementptr inbounds ([32 x i8]* @a, i32 0, i32 5)
+; CHECK-NEXT: getelementptr inbounds ([32 x i8], [32 x i8]* @a, i32 0, i32 5)
   ret i8* %ret
 }
 
 define i8* @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
 
   %ret = call i8* @stpcpy(i8* %dst, i8* %dst)
 ; CHECK: [[LEN:%[a-z]+]] = call i32 @strlen
-; CHECK-NEXT: getelementptr inbounds [32 x i8]* @a, i32 0, i32 [[LEN]]
+; CHECK-NEXT: getelementptr inbounds [32 x i8], [32 x i8]* @a, i32 0, i32 [[LEN]]
   ret i8* %ret
 }
 
 define i8* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [32 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [32 x i8], [32 x i8]* @b, i32 0, i32 0
 
   %ret = call i8* @stpcpy(i8* %dst, i8* %src)
 ; CHECK: call i8* @stpcpy
diff --git a/test/Transforms/InstCombine/stpcpy-2.ll b/test/Transforms/InstCombine/stpcpy-2.ll
index 6a0f7530d5a8..07e13a69e2d2 100644
--- a/test/Transforms/InstCombine/stpcpy-2.ll
+++ b/test/Transforms/InstCombine/stpcpy-2.ll
@@ -13,8 +13,8 @@ declare i16* @stpcpy(i8*, i8*)
 define void @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i16* @stpcpy(i8* %dst, i8* %src)
 ; CHECK: call i16* @stpcpy
diff --git a/test/Transforms/InstCombine/stpcpy_chk-1.ll b/test/Transforms/InstCombine/stpcpy_chk-1.ll
index 393c5d96d10b..2fcc34b05227 100644
--- a/test/Transforms/InstCombine/stpcpy_chk-1.ll
+++ b/test/Transforms/InstCombine/stpcpy_chk-1.ll
@@ -13,33 +13,33 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i8* @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 11)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 11)
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 60)
   ret i8* %ret
 }
 
 define i8* @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 11)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 11)
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 12)
   ret i8* %ret
 }
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 11)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 11)
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 -1)
   ret i8* %ret
 }
@@ -48,10 +48,10 @@ define i8* @test_simplify3() {
 
 define i8* @test_simplify4() {
 ; CHECK-LABEL: @test_simplify4(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %stpcpy = call i8* @stpcpy(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0))
+; CHECK-NEXT: %stpcpy = call i8* @stpcpy(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0))
 ; CHECK-NEXT: ret i8* %stpcpy
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 -1)
   ret i8* %ret
@@ -61,12 +61,12 @@ define i8* @test_simplify4() {
 
 define i8* @test_simplify5() {
 ; CHECK-LABEL: @test_simplify5(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i1 false)
-; CHECK-NEXT: %1 = call i8* @__memcpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 %len)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 11)
+; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i1 false)
+; CHECK-NEXT: %1 = call i8* @__memcpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 %len)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 11)
   %len = call i32 @llvm.objectsize.i32.p0i8(i8* %dst, i1 false)
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 %len)
   ret i8* %ret
@@ -76,10 +76,10 @@ define i8* @test_simplify5() {
 
 define i8* @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
 
-; CHECK-NEXT: %strlen = call i32 @strlen(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0))
-; CHECK-NEXT: %1 = getelementptr inbounds [60 x i8]* @a, i32 0, i32 %strlen
+; CHECK-NEXT: %strlen = call i32 @strlen(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0))
+; CHECK-NEXT: %1 = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 %strlen
 ; CHECK-NEXT: ret i8* %1
   %len = call i32 @llvm.objectsize.i32.p0i8(i8* %dst, i1 false)
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %dst, i32 %len)
@@ -90,10 +90,10 @@ define i8* @test_simplify6() {
 
 define i8* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %ret = call i8* @__stpcpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0), i32 8)
+; CHECK-NEXT: %ret = call i8* @__stpcpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0), i32 8)
 ; CHECK-NEXT: ret i8* %ret
   %ret = call i8* @__stpcpy_chk(i8* %dst, i8* %src, i32 8)
   ret i8* %ret
diff --git a/test/Transforms/InstCombine/stpcpy_chk-2.ll b/test/Transforms/InstCombine/stpcpy_chk-2.ll
index b503da9c191c..b4803f98b249 100644
--- a/test/Transforms/InstCombine/stpcpy_chk-2.ll
+++ b/test/Transforms/InstCombine/stpcpy_chk-2.ll
@@ -10,8 +10,8 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define void @test_no_simplify() {
 ; CHECK-LABEL: @test_no_simplify(
-  %dst = getelementptr inbounds [60 x i16]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [8 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i16], [60 x i16]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [8 x i8], [8 x i8]* @.str, i32 0, i32 0
 
 ; CHECK-NEXT: call i16* @__strcpy_chk
   call i16* @__strcpy_chk(i16* %dst, i8* %src, i32 8)
diff --git a/test/Transforms/InstCombine/strcat-1.ll b/test/Transforms/InstCombine/strcat-1.ll
index 131ad4834837..446a26e94ac4 100644
--- a/test/Transforms/InstCombine/strcat-1.ll
+++ b/test/Transforms/InstCombine/strcat-1.ll
@@ -18,19 +18,19 @@ define i32 @main() {
 ; CHECK: call i32 @puts
 
   %target = alloca [1024 x i8]
-  %arg1 = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  %arg1 = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0
   store i8 0, i8* %arg1
 
   ; rslt1 = strcat(target, "hello\00")
-  %arg2 = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %arg2 = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %rslt1 = call i8* @strcat(i8* %arg1, i8* %arg2)
 
   ; rslt2 = strcat(rslt1, "\00")
-  %arg3 = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %arg3 = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %rslt2 = call i8* @strcat(i8* %rslt1, i8* %arg3)
 
   ; rslt3 = strcat(rslt2, "\00hello\00")
-  %arg4 = getelementptr [7 x i8]* @null_hello, i32 0, i32 0
+  %arg4 = getelementptr [7 x i8], [7 x i8]* @null_hello, i32 0, i32 0
   %rslt3 = call i8* @strcat(i8* %rslt2, i8* %arg4)
 
   call i32 @puts( i8* %rslt3 )
diff --git a/test/Transforms/InstCombine/strcat-2.ll b/test/Transforms/InstCombine/strcat-2.ll
index 48f82670c325..287019770acf 100644
--- a/test/Transforms/InstCombine/strcat-2.ll
+++ b/test/Transforms/InstCombine/strcat-2.ll
@@ -15,8 +15,8 @@ define void @test_simplify1() {
 ; CHECK-NOT: call i8* @strcat
 ; CHECK: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i8* @strcat(i8* %dst, i8* %src)
   ret void
 }
@@ -25,8 +25,8 @@ define void @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
 ; CHECK-NEXT: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   call i8* @strcat(i8* %dst, i8* %src)
   ret void
 }
diff --git a/test/Transforms/InstCombine/strcat-3.ll b/test/Transforms/InstCombine/strcat-3.ll
index e3396df73d37..88cd162d0cd5 100644
--- a/test/Transforms/InstCombine/strcat-3.ll
+++ b/test/Transforms/InstCombine/strcat-3.ll
@@ -15,8 +15,8 @@ define void @test_nosimplify1() {
 ; CHECK: call i16* @strcat
 ; CHECK: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i16* @strcat(i8* %dst, i8* %src)
   ret void
 }
diff --git a/test/Transforms/InstCombine/strchr-1.ll b/test/Transforms/InstCombine/strchr-1.ll
index 66b3e2e51c7a..6c10ebdfcc43 100644
--- a/test/Transforms/InstCombine/strchr-1.ll
+++ b/test/Transforms/InstCombine/strchr-1.ll
@@ -1,20 +1,21 @@
 ; Test that the strchr library call simplifier works correctly.
 ; RUN: opt < %s -instcombine -S | FileCheck %s
 
-target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
+target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128-n8:16:32"
 
 @hello = constant [14 x i8] c"hello world\5Cn\00"
 @null = constant [1 x i8] zeroinitializer
+@newlines = constant [3 x i8] c"\0D\0A\00"
 @chp = global i8* zeroinitializer
 
 declare i8* @strchr(i8*, i32)
 
 define void @test_simplify1() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 6)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 6)
 ; CHECK-NOT: call i8* @strchr
 ; CHECK: ret void
 
-  %str = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strchr(i8* %str, i32 119)
   store i8* %dst, i8** @chp
   ret void
@@ -25,18 +26,18 @@ define void @test_simplify2() {
 ; CHECK-NOT: call i8* @strchr
 ; CHECK: ret void
 
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %dst = call i8* @strchr(i8* %str, i32 119)
   store i8* %dst, i8** @chp
   ret void
 }
 
 define void @test_simplify3() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 13)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
 ; CHECK-NOT: call i8* @strchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strchr(i8* %src, i32 0)
   store i8* %dst, i8** @chp
   ret void
@@ -47,18 +48,18 @@ define void @test_simplify4(i32 %chr) {
 ; CHECK-NOT: call i8* @strchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strchr(i8* %src, i32 %chr)
   store i8* %dst, i8** @chp
   ret void
 }
 
 define void @test_simplify5() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 13)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
 ; CHECK-NOT: call i8* @strchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strchr(i8* %src, i32 65280)
   store i8* %dst, i8** @chp
   ret void
@@ -68,7 +69,7 @@ define void @test_simplify5() {
 define void @test_simplify6(i8* %str) {
 ; CHECK: %strlen = call i32 @strlen(i8* %str)
 ; CHECK-NOT: call i8* @strchr
-; CHECK: %strchr = getelementptr i8* %str, i32 %strlen
+; CHECK: %strchr = getelementptr i8, i8* %str, i32 %strlen
 ; CHECK: store i8* %strchr, i8** @chp, align 4
 ; CHECK: ret void
 
@@ -76,3 +77,19 @@ define void @test_simplify6(i8* %str) {
   store i8* %dst, i8** @chp
   ret void
 }
+
+; Check transformation strchr("\r\n", C) != nullptr -> (C & 9217) != 0
+define i1 @test_simplify7(i32 %C) {
+; CHECK-LABEL: @test_simplify7
+; CHECK-NEXT: [[TRUNC:%.*]] = trunc i32 %C to i16
+; CHECK-NEXT: %memchr.bounds = icmp ult i16 [[TRUNC]], 16
+; CHECK-NEXT: [[SHL:%.*]] = shl i16 1, [[TRUNC]]
+; CHECK-NEXT: [[AND:%.*]] = and i16 [[SHL]], 9217
+; CHECK-NEXT: %memchr.bits = icmp ne i16 [[AND]], 0
+; CHECK-NEXT: %memchr1 = and i1 %memchr.bounds, %memchr.bits
+; CHECK-NEXT: ret i1 %memchr1
+
+  %dst = call i8* @strchr(i8* getelementptr inbounds ([3 x i8], [3 x i8]* @newlines, i64 0, i64 0), i32 %C)
+  %cmp = icmp ne i8* %dst, null
+  ret i1 %cmp
+}
diff --git a/test/Transforms/InstCombine/strchr-2.ll b/test/Transforms/InstCombine/strchr-2.ll
index 35bbd23e6d4d..dd86a16f77bc 100644
--- a/test/Transforms/InstCombine/strchr-2.ll
+++ b/test/Transforms/InstCombine/strchr-2.ll
@@ -14,7 +14,7 @@ define void @test_nosimplify1() {
 ; CHECK: call i8 @strchr
 ; CHECK: ret void
 
-  %str = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8 @strchr(i8* %str, i32 119)
   store i8 %dst, i8* @chr
   ret void
diff --git a/test/Transforms/InstCombine/strcmp-1.ll b/test/Transforms/InstCombine/strcmp-1.ll
index 9bbd7dbe7da8..9bc1fc693557 100644
--- a/test/Transforms/InstCombine/strcmp-1.ll
+++ b/test/Transforms/InstCombine/strcmp-1.ll
@@ -13,12 +13,12 @@ declare i32 @strcmp(i8*, i8*)
 ; strcmp("", x) -> -*x
 define i32 @test1(i8* %str2) {
 ; CHECK-LABEL: @test1(
-; CHECK: %strcmpload = load i8* %str
+; CHECK: %strcmpload = load i8, i8* %str
 ; CHECK: %1 = zext i8 %strcmpload to i32
 ; CHECK: %2 = sub nsw i32 0, %1
 ; CHECK: ret i32 %2
 
-  %str1 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str1 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2)
   ret i32 %temp1
 
@@ -27,11 +27,11 @@ define i32 @test1(i8* %str2) {
 ; strcmp(x, "") -> *x
 define i32 @test2(i8* %str1) {
 ; CHECK-LABEL: @test2(
-; CHECK: %strcmpload = load i8* %str
+; CHECK: %strcmpload = load i8, i8* %str
 ; CHECK: %1 = zext i8 %strcmpload to i32
 ; CHECK: ret i32 %1
 
-  %str2 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str2 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2)
   ret i32 %temp1
 }
@@ -41,8 +41,8 @@ define i32 @test3() {
 ; CHECK-LABEL: @test3(
 ; CHECK: ret i32 -1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2)
   ret i32 %temp1
 }
@@ -51,8 +51,8 @@ define i32 @test4() {
 ; CHECK-LABEL: @test4(
 ; CHECK: ret i32 1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2)
   ret i32 %temp1
 }
@@ -61,12 +61,12 @@ define i32 @test4() {
 ; (This transform is rather difficult to trigger in a useful manner)
 define i32 @test5(i1 %b) {
 ; CHECK-LABEL: @test5(
-; CHECK: %memcmp = call i32 @memcmp(i8* getelementptr inbounds ([6 x i8]* @hello, i32 0, i32 0), i8* %str2, i32 5)
+; CHECK: %memcmp = call i32 @memcmp(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @hello, i32 0, i32 0), i8* %str2, i32 5)
 ; CHECK: ret i32 %memcmp
 
-  %str1 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
-  %temp1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %temp2 = getelementptr inbounds [5 x i8]* @bell, i32 0, i32 0
+  %str1 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
+  %temp1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %temp2 = getelementptr inbounds [5 x i8], [5 x i8]* @bell, i32 0, i32 0
   %str2 = select i1 %b, i8* %temp1, i8* %temp2
   %temp3 = call i32 @strcmp(i8* %str1, i8* %str2)
   ret i32 %temp3
diff --git a/test/Transforms/InstCombine/strcmp-2.ll b/test/Transforms/InstCombine/strcmp-2.ll
index f0ef51652270..a537b10e0deb 100644
--- a/test/Transforms/InstCombine/strcmp-2.ll
+++ b/test/Transforms/InstCombine/strcmp-2.ll
@@ -13,8 +13,8 @@ define i16 @test_nosimplify() {
 ; CHECK: call i16 @strcmp
 ; CHECK: ret i16 %temp1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %temp1 = call i16 @strcmp(i8* %str1, i8* %str2)
   ret i16 %temp1
 }
diff --git a/test/Transforms/InstCombine/strcpy-1.ll b/test/Transforms/InstCombine/strcpy-1.ll
index 7c253f6f9fca..24c70c18fc05 100644
--- a/test/Transforms/InstCombine/strcpy-1.ll
+++ b/test/Transforms/InstCombine/strcpy-1.ll
@@ -15,8 +15,8 @@ declare i8* @strcpy(i8*, i8*)
 define void @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i8* @strcpy(i8* %dst, i8* %src)
 ; CHECK: @llvm.memcpy.p0i8.p0i8.i32
@@ -26,18 +26,18 @@ define void @test_simplify1() {
 define i8* @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
 
   %ret = call i8* @strcpy(i8* %dst, i8* %dst)
-; CHECK: ret i8* getelementptr inbounds ([32 x i8]* @a, i32 0, i32 0)
+; CHECK: ret i8* getelementptr inbounds ([32 x i8], [32 x i8]* @a, i32 0, i32 0)
   ret i8* %ret
 }
 
 define i8* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [32 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [32 x i8], [32 x i8]* @b, i32 0, i32 0
 
   %ret = call i8* @strcpy(i8* %dst, i8* %src)
 ; CHECK: call i8* @strcpy
diff --git a/test/Transforms/InstCombine/strcpy-2.ll b/test/Transforms/InstCombine/strcpy-2.ll
index bad392d7c648..cfc8a410b7ee 100644
--- a/test/Transforms/InstCombine/strcpy-2.ll
+++ b/test/Transforms/InstCombine/strcpy-2.ll
@@ -13,8 +13,8 @@ declare i16* @strcpy(i8*, i8*)
 define void @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i16* @strcpy(i8* %dst, i8* %src)
 ; CHECK: call i16* @strcpy
diff --git a/test/Transforms/InstCombine/strcpy_chk-1.ll b/test/Transforms/InstCombine/strcpy_chk-1.ll
index e3f163fd087d..7a21a49c993c 100644
--- a/test/Transforms/InstCombine/strcpy_chk-1.ll
+++ b/test/Transforms/InstCombine/strcpy_chk-1.ll
@@ -13,33 +13,33 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i8* @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0)
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 60)
   ret i8* %ret
 }
 
 define i8* @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0)
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 12)
   ret i8* %ret
 }
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0)
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 -1)
   ret i8* %ret
 }
@@ -48,10 +48,10 @@ define i8* @test_simplify3() {
 
 define i8* @test_simplify4() {
 ; CHECK-LABEL: @test_simplify4(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %strcpy = call i8* @strcpy(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0))
+; CHECK-NEXT: %strcpy = call i8* @strcpy(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0))
 ; CHECK-NEXT: ret i8* %strcpy
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 -1)
   ret i8* %ret
@@ -61,11 +61,11 @@ define i8* @test_simplify4() {
 
 define i8* @test_simplify5() {
 ; CHECK-LABEL: @test_simplify5(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i1 false)
-; CHECK-NEXT: %1 = call i8* @__memcpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 %len)
+; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i1 false)
+; CHECK-NEXT: %1 = call i8* @__memcpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 %len)
 ; CHECK-NEXT: ret i8* %1
   %len = call i32 @llvm.objectsize.i32.p0i8(i8* %dst, i1 false)
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 %len)
@@ -76,10 +76,10 @@ define i8* @test_simplify5() {
 
 define i8* @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
 
-; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i1 false)
-; CHECK-NEXT: %ret = call i8* @__strcpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i32 %len)
+; CHECK-NEXT: %len = call i32 @llvm.objectsize.i32.p0i8(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i1 false)
+; CHECK-NEXT: %ret = call i8* @__strcpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i32 %len)
 ; CHECK-NEXT: ret i8* %ret
   %len = call i32 @llvm.objectsize.i32.p0i8(i8* %dst, i1 false)
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %dst, i32 %len)
@@ -90,10 +90,10 @@ define i8* @test_simplify6() {
 
 define i8* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %ret = call i8* @__strcpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0), i32 8)
+; CHECK-NEXT: %ret = call i8* @__strcpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0), i32 8)
 ; CHECK-NEXT: ret i8* %ret
   %ret = call i8* @__strcpy_chk(i8* %dst, i8* %src, i32 8)
   ret i8* %ret
diff --git a/test/Transforms/InstCombine/strcpy_chk-2.ll b/test/Transforms/InstCombine/strcpy_chk-2.ll
index 1eff5a822e78..c2204a830bae 100644
--- a/test/Transforms/InstCombine/strcpy_chk-2.ll
+++ b/test/Transforms/InstCombine/strcpy_chk-2.ll
@@ -10,8 +10,8 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define void @test_no_simplify() {
 ; CHECK-LABEL: @test_no_simplify(
-  %dst = getelementptr inbounds [60 x i16]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [8 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i16], [60 x i16]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [8 x i8], [8 x i8]* @.str, i32 0, i32 0
 
 ; CHECK-NEXT: call i16* @__strcpy_chk
   call i16* @__strcpy_chk(i16* %dst, i8* %src, i32 8)
diff --git a/test/Transforms/InstCombine/strcpy_chk-64.ll b/test/Transforms/InstCombine/strcpy_chk-64.ll
index 31447d9569e1..8a2cdc2f2bb5 100644
--- a/test/Transforms/InstCombine/strcpy_chk-64.ll
+++ b/test/Transforms/InstCombine/strcpy_chk-64.ll
@@ -7,7 +7,7 @@ define void @func(i8* %i) nounwind ssp {
 ; CHECK: @__strcpy_chk(i8* %arraydecay, i8* %i, i64 32)
 entry:
   %s = alloca [32 x i8], align 16
-  %arraydecay = getelementptr inbounds [32 x i8]* %s, i32 0, i32 0
+  %arraydecay = getelementptr inbounds [32 x i8], [32 x i8]* %s, i32 0, i32 0
   %call = call i8* @__strcpy_chk(i8* %arraydecay, i8* %i, i64 32)
   call void @func2(i8* %arraydecay)
   ret void
diff --git a/test/Transforms/InstCombine/strcspn-1.ll b/test/Transforms/InstCombine/strcspn-1.ll
index b3b52b5025a5..8d441a9eecd0 100644
--- a/test/Transforms/InstCombine/strcspn-1.ll
+++ b/test/Transforms/InstCombine/strcspn-1.ll
@@ -14,7 +14,7 @@ declare i64 @strcspn(i8*, i8*)
 
 define i64 @test_simplify1(i8* %str) {
 ; CHECK-LABEL: @test_simplify1(
-  %pat = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %pat = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i64 @strcspn(i8* %str, i8* %pat)
 ; CHECK-NEXT: [[VAR:%[a-z]+]] = call i64 @strlen(i8* %str)
@@ -26,7 +26,7 @@ define i64 @test_simplify1(i8* %str) {
 
 define i64 @test_simplify2(i8* %pat) {
 ; CHECK-LABEL: @test_simplify2(
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i64 @strcspn(i8* %str, i8* %pat)
   ret i64 %ret
@@ -37,8 +37,8 @@ define i64 @test_simplify2(i8* %pat) {
 
 define i64 @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr [6 x i8]* @abcba, i32 0, i32 0
-  %pat = getelementptr [4 x i8]* @abc, i32 0, i32 0
+  %str = getelementptr [6 x i8], [6 x i8]* @abcba, i32 0, i32 0
+  %pat = getelementptr [4 x i8], [4 x i8]* @abc, i32 0, i32 0
 
   %ret = call i64 @strcspn(i8* %str, i8* %pat)
   ret i64 %ret
diff --git a/test/Transforms/InstCombine/strcspn-2.ll b/test/Transforms/InstCombine/strcspn-2.ll
index ecfa27d3b6ba..749860afcd4e 100644
--- a/test/Transforms/InstCombine/strcspn-2.ll
+++ b/test/Transforms/InstCombine/strcspn-2.ll
@@ -12,7 +12,7 @@ declare double @strcspn(i8*, i8*)
 
 define double @test_no_simplify1(i8* %pat) {
 ; CHECK-LABEL: @test_no_simplify1(
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call double @strcspn(i8* %str, i8* %pat)
 ; CHECK-NEXT: call double @strcspn
diff --git a/test/Transforms/InstCombine/strlen-1.ll b/test/Transforms/InstCombine/strlen-1.ll
index 4a3caf28a0fe..fe453699a60e 100644
--- a/test/Transforms/InstCombine/strlen-1.ll
+++ b/test/Transforms/InstCombine/strlen-1.ll
@@ -17,7 +17,7 @@ declare i32 @strlen(i8*)
 
 define i32 @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %hello_p = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %hello_p = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %hello_l = call i32 @strlen(i8* %hello_p)
   ret i32 %hello_l
 ; CHECK-NEXT: ret i32 5
@@ -25,7 +25,7 @@ define i32 @test_simplify1() {
 
 define i32 @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %null_p = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %null_p = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %null_l = call i32 @strlen(i8* %null_p)
   ret i32 %null_l
 ; CHECK-NEXT: ret i32 0
@@ -33,7 +33,7 @@ define i32 @test_simplify2() {
 
 define i32 @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %null_hello_p = getelementptr [7 x i8]* @null_hello, i32 0, i32 0
+  %null_hello_p = getelementptr [7 x i8], [7 x i8]* @null_hello, i32 0, i32 0
   %null_hello_l = call i32 @strlen(i8* %null_hello_p)
   ret i32 %null_hello_l
 ; CHECK-NEXT: ret i32 0
@@ -50,7 +50,7 @@ define i32 @test_simplify4() {
 
 define i1 @test_simplify5() {
 ; CHECK-LABEL: @test_simplify5(
-  %hello_p = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %hello_p = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %hello_l = call i32 @strlen(i8* %hello_p)
   %eq_hello = icmp eq i32 %hello_l, 0
   ret i1 %eq_hello
@@ -59,7 +59,7 @@ define i1 @test_simplify5() {
 
 define i1 @test_simplify6() {
 ; CHECK-LABEL: @test_simplify6(
-  %null_p = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %null_p = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %null_l = call i32 @strlen(i8* %null_p)
   %eq_null = icmp eq i32 %null_l, 0
   ret i1 %eq_null
@@ -70,7 +70,7 @@ define i1 @test_simplify6() {
 
 define i1 @test_simplify7() {
 ; CHECK-LABEL: @test_simplify7(
-  %hello_p = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %hello_p = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %hello_l = call i32 @strlen(i8* %hello_p)
   %ne_hello = icmp ne i32 %hello_l, 0
   ret i1 %ne_hello
@@ -79,7 +79,7 @@ define i1 @test_simplify7() {
 
 define i1 @test_simplify8() {
 ; CHECK-LABEL: @test_simplify8(
-  %null_p = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %null_p = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %null_l = call i32 @strlen(i8* %null_p)
   %ne_null = icmp ne i32 %null_l, 0
   ret i1 %ne_null
@@ -88,8 +88,8 @@ define i1 @test_simplify8() {
 
 define i32 @test_simplify9(i1 %x) {
 ; CHECK-LABEL: @test_simplify9
-  %hello = getelementptr [6 x i8]* @hello, i32 0, i32 0
-  %longer = getelementptr [7 x i8]* @longer, i32 0, i32 0
+  %hello = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
+  %longer = getelementptr [7 x i8], [7 x i8]* @longer, i32 0, i32 0
   %s = select i1 %x, i8* %hello, i8* %longer
   %l = call i32 @strlen(i8* %s)
 ; CHECK-NEXT: select i1 %x, i32 5, i32 6
@@ -101,7 +101,7 @@ define i32 @test_simplify9(i1 %x) {
 
 define i32 @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %a_p = getelementptr [32 x i8]* @a, i32 0, i32 0
+  %a_p = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
   %a_l = call i32 @strlen(i8* %a_p)
 ; CHECK-NEXT: %a_l = call i32 @strlen
   ret i32 %a_l
diff --git a/test/Transforms/InstCombine/strlen-2.ll b/test/Transforms/InstCombine/strlen-2.ll
index 6652a310ba6a..df5eee0caef7 100644
--- a/test/Transforms/InstCombine/strlen-2.ll
+++ b/test/Transforms/InstCombine/strlen-2.ll
@@ -10,7 +10,7 @@ declare i32 @strlen(i8*, i32)
 
 define i32 @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %hello_p = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %hello_p = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %hello_l = call i32 @strlen(i8* %hello_p, i32 187)
 ; CHECK-NEXT: %hello_l = call i32 @strlen
   ret i32 %hello_l
diff --git a/test/Transforms/InstCombine/strncat-1.ll b/test/Transforms/InstCombine/strncat-1.ll
index 8eae3dae2e23..4b8da81f6fb6 100644
--- a/test/Transforms/InstCombine/strncat-1.ll
+++ b/test/Transforms/InstCombine/strncat-1.ll
@@ -17,19 +17,19 @@ define i32 @main() {
 ; CHECK: call i32 @puts
 
   %target = alloca [1024 x i8]
-  %arg1 = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  %arg1 = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0
   store i8 0, i8* %arg1
 
   ; rslt1 = strncat(target, "hello\00")
-  %arg2 = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %arg2 = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %rslt1 = call i8* @strncat(i8* %arg1, i8* %arg2, i32 6)
 
   ; rslt2 = strncat(rslt1, "\00")
-  %arg3 = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %arg3 = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %rslt2 = call i8* @strncat(i8* %rslt1, i8* %arg3, i32 42)
 
   ; rslt3 = strncat(rslt2, "\00hello\00")
-  %arg4 = getelementptr [7 x i8]* @null_hello, i32 0, i32 0
+  %arg4 = getelementptr [7 x i8], [7 x i8]* @null_hello, i32 0, i32 0
   %rslt3 = call i8* @strncat(i8* %rslt2, i8* %arg4, i32 42)
 
   call i32 @puts(i8* %rslt3)
diff --git a/test/Transforms/InstCombine/strncat-2.ll b/test/Transforms/InstCombine/strncat-2.ll
index b09fa1260a03..3533e1eace70 100644
--- a/test/Transforms/InstCombine/strncat-2.ll
+++ b/test/Transforms/InstCombine/strncat-2.ll
@@ -15,8 +15,8 @@ define void @test_simplify1() {
 ; CHECK-NOT: call i8* @strncat
 ; CHECK: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i8* @strncat(i8* %dst, i8* %src, i32 13)
   ret void
 }
@@ -25,8 +25,8 @@ define void @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
 ; CHECK-NEXT: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [1 x i8]* @empty, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [1 x i8], [1 x i8]* @empty, i32 0, i32 0
   call i8* @strncat(i8* %dst, i8* %src, i32 13)
   ret void
 }
@@ -35,8 +35,8 @@ define void @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
 ; CHECK-NEXT: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i8* @strncat(i8* %dst, i8* %src, i32 0)
   ret void
 }
@@ -46,8 +46,8 @@ define void @test_nosimplify1() {
 ; CHECK: call i8* @strncat
 ; CHECK: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i8* @strncat(i8* %dst, i8* %src, i32 1)
   ret void
 }
diff --git a/test/Transforms/InstCombine/strncat-3.ll b/test/Transforms/InstCombine/strncat-3.ll
index 1b25b4aca1ae..0f6964df9154 100644
--- a/test/Transforms/InstCombine/strncat-3.ll
+++ b/test/Transforms/InstCombine/strncat-3.ll
@@ -15,8 +15,8 @@ define void @test_nosimplify1() {
 ; CHECK: call i16* @strncat
 ; CHECK: ret void
 
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   call i16* @strncat(i8* %dst, i8* %src, i32 13)
   ret void
 }
diff --git a/test/Transforms/InstCombine/strncmp-1.ll b/test/Transforms/InstCombine/strncmp-1.ll
index 49b095554c2c..a1121821fdf5 100644
--- a/test/Transforms/InstCombine/strncmp-1.ll
+++ b/test/Transforms/InstCombine/strncmp-1.ll
@@ -13,12 +13,12 @@ declare i32 @strncmp(i8*, i8*, i32)
 ; strncmp("", x, n) -> -*x
 define i32 @test1(i8* %str2) {
 ; CHECK-LABEL: @test1(
-; CHECK: %strcmpload = load i8* %str
+; CHECK: %strcmpload = load i8, i8* %str
 ; CHECK: %1 = zext i8 %strcmpload to i32
 ; CHECK: %2 = sub nsw i32 0, %1
 ; CHECK: ret i32 %2
 
-  %str1 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str1 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strncmp(i8* %str1, i8* %str2, i32 10)
   ret i32 %temp1
 }
@@ -26,11 +26,11 @@ define i32 @test1(i8* %str2) {
 ; strncmp(x, "", n) -> *x
 define i32 @test2(i8* %str1) {
 ; CHECK-LABEL: @test2(
-; CHECK: %strcmpload = load i8* %str1
+; CHECK: %strcmpload = load i8, i8* %str1
 ; CHECK: %1 = zext i8 %strcmpload to i32
 ; CHECK: ret i32 %1
 
-  %str2 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str2 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strncmp(i8* %str1, i8* %str2, i32 10)
   ret i32 %temp1
 }
@@ -40,8 +40,8 @@ define i32 @test3() {
 ; CHECK-LABEL: @test3(
 ; CHECK: ret i32 -1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %temp1 = call i32 @strncmp(i8* %str1, i8* %str2, i32 10)
   ret i32 %temp1
 }
@@ -50,8 +50,8 @@ define i32 @test4() {
 ; CHECK-LABEL: @test4(
 ; CHECK: ret i32 1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %temp1 = call i32 @strncmp(i8* %str1, i8* %str2, i32 10)
   ret i32 %temp1
 }
@@ -60,8 +60,8 @@ define i32 @test5() {
 ; CHECK-LABEL: @test5(
 ; CHECK: ret i32 0
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %temp1 = call i32 @strncmp(i8* %str1, i8* %str2, i32 4)
   ret i32 %temp1
 }
@@ -69,9 +69,9 @@ define i32 @test5() {
 ; strncmp(x,y,1) -> memcmp(x,y,1)
 define i32 @test6(i8* %str1, i8* %str2) {
 ; CHECK-LABEL: @test6(
-; CHECK: [[LOAD1:%[a-z]+]] = load i8* %str1, align 1
+; CHECK: [[LOAD1:%[a-z]+]] = load i8, i8* %str1, align 1
 ; CHECK: [[ZEXT1:%[a-z]+]] = zext i8 [[LOAD1]] to i32
-; CHECK: [[LOAD2:%[a-z]+]] = load i8* %str2, align 1
+; CHECK: [[LOAD2:%[a-z]+]] = load i8, i8* %str2, align 1
 ; CHECK: [[ZEXT2:%[a-z]+]] = zext i8 [[LOAD2]] to i32
 ; CHECK: [[RET:%[a-z]+]] = sub nsw i32 [[ZEXT1]], [[ZEXT2]]
 ; CHECK: ret i32 [[RET]]
diff --git a/test/Transforms/InstCombine/strncmp-2.ll b/test/Transforms/InstCombine/strncmp-2.ll
index 16ad8a4f62a7..9e19781e04c4 100644
--- a/test/Transforms/InstCombine/strncmp-2.ll
+++ b/test/Transforms/InstCombine/strncmp-2.ll
@@ -13,8 +13,8 @@ define i16 @test_nosimplify() {
 ; CHECK: call i16 @strncmp
 ; CHECK: ret i16 %temp1
 
-  %str1 = getelementptr inbounds [5 x i8]* @hell, i32 0, i32 0
-  %str2 = getelementptr inbounds [6 x i8]* @hello, i32 0, i32 0
+  %str1 = getelementptr inbounds [5 x i8], [5 x i8]* @hell, i32 0, i32 0
+  %str2 = getelementptr inbounds [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %temp1 = call i16 @strncmp(i8* %str1, i8* %str2, i32 10)
   ret i16 %temp1
 }
diff --git a/test/Transforms/InstCombine/strncpy-1.ll b/test/Transforms/InstCombine/strncpy-1.ll
index c70197f12e28..28cb26b17047 100644
--- a/test/Transforms/InstCombine/strncpy-1.ll
+++ b/test/Transforms/InstCombine/strncpy-1.ll
@@ -20,16 +20,16 @@ define i32 @test_simplify1() {
 ; CHECK-NOT: call i8* @strncpy
 ; CHECK: call i32 @puts
   %target = alloca [1024 x i8]
-  %arg1 = getelementptr [1024 x i8]* %target, i32 0, i32 0
+  %arg1 = getelementptr [1024 x i8], [1024 x i8]* %target, i32 0, i32 0
   store i8 0, i8* %arg1
 
-  %arg2 = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %arg2 = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
   %rslt1 = call i8* @strncpy(i8* %arg1, i8* %arg2, i32 6)
 
-  %arg3 = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %arg3 = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %rslt2 = call i8* @strncpy(i8* %rslt1, i8* %arg3, i32 42)
 
-  %arg4 = getelementptr [7 x i8]* @null_hello, i32 0, i32 0
+  %arg4 = getelementptr [7 x i8], [7 x i8]* @null_hello, i32 0, i32 0
   %rslt3 = call i8* @strncpy(i8* %rslt2, i8* %arg4, i32 42)
 
   call i32 @puts( i8* %rslt3 )
@@ -40,8 +40,8 @@ define i32 @test_simplify1() {
 
 define void @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   call i8* @strncpy(i8* %dst, i8* %src, i32 32)
 ; CHECK: call void @llvm.memset.p0i8.i32
@@ -52,20 +52,20 @@ define void @test_simplify2() {
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   %ret = call i8* @strncpy(i8* %dst, i8* %src, i32 0)
   ret i8* %ret
-; CHECK: ret i8* getelementptr inbounds ([32 x i8]* @a, i32 0, i32 0)
+; CHECK: ret i8* getelementptr inbounds ([32 x i8], [32 x i8]* @a, i32 0, i32 0)
 }
 
 ; Check  strncpy(x, s, c) -> memcpy(x, s, c, 1) [s and c are constant].
 
 define void @test_simplify4() {
 ; CHECK-LABEL: @test_simplify4(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i8* @strncpy(i8* %dst, i8* %src, i32 6)
 ; CHECK: call void @llvm.memcpy.p0i8.p0i8.i32
@@ -76,8 +76,8 @@ define void @test_simplify4() {
 
 define void @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [32 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [32 x i8], [32 x i8]* @b, i32 0, i32 0
 
   call i8* @strncpy(i8* %dst, i8* %src, i32 32)
 ; CHECK: call i8* @strncpy
@@ -86,8 +86,8 @@ define void @test_no_simplify1() {
 
 define void @test_no_simplify2() {
 ; CHECK-LABEL: @test_no_simplify2(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i8* @strncpy(i8* %dst, i8* %src, i32 8)
 ; CHECK: call i8* @strncpy
diff --git a/test/Transforms/InstCombine/strncpy-2.ll b/test/Transforms/InstCombine/strncpy-2.ll
index acc2878648fc..5c45f9f24aee 100644
--- a/test/Transforms/InstCombine/strncpy-2.ll
+++ b/test/Transforms/InstCombine/strncpy-2.ll
@@ -13,8 +13,8 @@ declare i16* @strncpy(i8*, i8*, i32)
 
 define void @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %dst = getelementptr [32 x i8]* @a, i32 0, i32 0
-  %src = getelementptr [6 x i8]* @hello, i32 0, i32 0
+  %dst = getelementptr [32 x i8], [32 x i8]* @a, i32 0, i32 0
+  %src = getelementptr [6 x i8], [6 x i8]* @hello, i32 0, i32 0
 
   call i16* @strncpy(i8* %dst, i8* %src, i32 6)
 ; CHECK: call i16* @strncpy
diff --git a/test/Transforms/InstCombine/strncpy_chk-1.ll b/test/Transforms/InstCombine/strncpy_chk-1.ll
index 9242a8acdbbe..03690b9b1e01 100644
--- a/test/Transforms/InstCombine/strncpy_chk-1.ll
+++ b/test/Transforms/InstCombine/strncpy_chk-1.ll
@@ -13,32 +13,32 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i8* @test_simplify1() {
 ; CHECK-LABEL: @test_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0)
   %ret = call i8* @__strncpy_chk(i8* %dst, i8* %src, i32 12, i32 60)
   ret i8* %ret
 }
 
 define i8* @test_simplify2() {
 ; CHECK-LABEL: @test_simplify2(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
-; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 12, i32 1, i1 false)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0)
   %ret = call i8* @__strncpy_chk(i8* %dst, i8* %src, i32 12, i32 12)
   ret i8* %ret
 }
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %strncpy = call i8* @strncpy(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0), i32 12)
+; CHECK-NEXT: %strncpy = call i8* @strncpy(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0), i32 12)
 ; CHECK-NEXT: ret i8* %strncpy
   %ret = call i8* @__strncpy_chk(i8* %dst, i8* %src, i32 12, i32 60)
   ret i8* %ret
@@ -48,10 +48,10 @@ define i8* @test_simplify3() {
 
 define i8* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [12 x i8]* @.str, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [12 x i8], [12 x i8]* @.str, i32 0, i32 0
 
-; CHECK-NEXT: %ret = call i8* @__strncpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8]* @.str, i32 0, i32 0), i32 8, i32 4)
+; CHECK-NEXT: %ret = call i8* @__strncpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str, i32 0, i32 0), i32 8, i32 4)
 ; CHECK-NEXT: ret i8* %ret
   %ret = call i8* @__strncpy_chk(i8* %dst, i8* %src, i32 8, i32 4)
   ret i8* %ret
@@ -59,10 +59,10 @@ define i8* @test_no_simplify1() {
 
 define i8* @test_no_simplify2() {
 ; CHECK-LABEL: @test_no_simplify2(
-  %dst = getelementptr inbounds [60 x i8]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i8]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i8], [60 x i8]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i8], [60 x i8]* @b, i32 0, i32 0
 
-; CHECK-NEXT: %ret = call i8* @__strncpy_chk(i8* getelementptr inbounds ([60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8]* @b, i32 0, i32 0), i32 8, i32 0)
+; CHECK-NEXT: %ret = call i8* @__strncpy_chk(i8* getelementptr inbounds ([60 x i8], [60 x i8]* @a, i32 0, i32 0), i8* getelementptr inbounds ([60 x i8], [60 x i8]* @b, i32 0, i32 0), i32 8, i32 0)
 ; CHECK-NEXT: ret i8* %ret
   %ret = call i8* @__strncpy_chk(i8* %dst, i8* %src, i32 8, i32 0)
   ret i8* %ret
diff --git a/test/Transforms/InstCombine/strncpy_chk-2.ll b/test/Transforms/InstCombine/strncpy_chk-2.ll
index 829a4798f09a..89ecd46c04a9 100644
--- a/test/Transforms/InstCombine/strncpy_chk-2.ll
+++ b/test/Transforms/InstCombine/strncpy_chk-2.ll
@@ -10,8 +10,8 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define void @test_no_simplify() {
 ; CHECK-LABEL: @test_no_simplify(
-  %dst = getelementptr inbounds [60 x i16]* @a, i32 0, i32 0
-  %src = getelementptr inbounds [60 x i16]* @b, i32 0, i32 0
+  %dst = getelementptr inbounds [60 x i16], [60 x i16]* @a, i32 0, i32 0
+  %src = getelementptr inbounds [60 x i16], [60 x i16]* @b, i32 0, i32 0
 
 ; CHECK-NEXT: call i16* @__strncpy_chk
   call i16* @__strncpy_chk(i16* %dst, i16* %src, i32 60, i32 60)
diff --git a/test/Transforms/InstCombine/strpbrk-1.ll b/test/Transforms/InstCombine/strpbrk-1.ll
index 58b2d9e88d2c..a61100deb875 100644
--- a/test/Transforms/InstCombine/strpbrk-1.ll
+++ b/test/Transforms/InstCombine/strpbrk-1.ll
@@ -14,7 +14,7 @@ declare i8* @strpbrk(i8*, i8*)
 
 define i8* @test_simplify1(i8* %str) {
 ; CHECK-LABEL: @test_simplify1(
-  %pat = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %pat = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i8* @strpbrk(i8* %str, i8* %pat)
   ret i8* %ret
@@ -25,7 +25,7 @@ define i8* @test_simplify1(i8* %str) {
 
 define i8* @test_simplify2(i8* %pat) {
 ; CHECK-LABEL: @test_simplify2(
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i8* @strpbrk(i8* %str, i8* %pat)
   ret i8* %ret
@@ -36,19 +36,19 @@ define i8* @test_simplify2(i8* %pat) {
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr [12 x i8]* @hello, i32 0, i32 0
-  %pat = getelementptr [2 x i8]* @w, i32 0, i32 0
+  %str = getelementptr [12 x i8], [12 x i8]* @hello, i32 0, i32 0
+  %pat = getelementptr [2 x i8], [2 x i8]* @w, i32 0, i32 0
 
   %ret = call i8* @strpbrk(i8* %str, i8* %pat)
   ret i8* %ret
-; CHECK-NEXT: ret i8* getelementptr inbounds ([12 x i8]* @hello, i32 0, i32 6)
+; CHECK-NEXT: ret i8* getelementptr inbounds ([12 x i8], [12 x i8]* @hello, i32 0, i32 6)
 }
 
 ; Check strpbrk(s, "a") -> strchr(s, 'a').
 
 define i8* @test_simplify4(i8* %str) {
 ; CHECK-LABEL: @test_simplify4(
-  %pat = getelementptr [2 x i8]* @w, i32 0, i32 0
+  %pat = getelementptr [2 x i8], [2 x i8]* @w, i32 0, i32 0
 
   %ret = call i8* @strpbrk(i8* %str, i8* %pat)
 ; CHECK-NEXT: [[VAR:%[a-z]+]] = call i8* @strchr(i8* %str, i32 119)
diff --git a/test/Transforms/InstCombine/strpbrk-2.ll b/test/Transforms/InstCombine/strpbrk-2.ll
index b797d7a59527..0af6faf40bc2 100644
--- a/test/Transforms/InstCombine/strpbrk-2.ll
+++ b/test/Transforms/InstCombine/strpbrk-2.ll
@@ -13,8 +13,8 @@ declare i16* @strpbrk(i8*, i8*)
 
 define i16* @test_no_simplify1() {
 ; CHECK-LABEL: @test_no_simplify1(
-  %str = getelementptr [12 x i8]* @hello, i32 0, i32 0
-  %pat = getelementptr [2 x i8]* @w, i32 0, i32 0
+  %str = getelementptr [12 x i8], [12 x i8]* @hello, i32 0, i32 0
+  %pat = getelementptr [2 x i8], [2 x i8]* @w, i32 0, i32 0
 
   %ret = call i16* @strpbrk(i8* %str, i8* %pat)
 ; CHECK-NEXT: %ret = call i16* @strpbrk
diff --git a/test/Transforms/InstCombine/strrchr-1.ll b/test/Transforms/InstCombine/strrchr-1.ll
index 4615f5f2bd25..3ae68fba96df 100644
--- a/test/Transforms/InstCombine/strrchr-1.ll
+++ b/test/Transforms/InstCombine/strrchr-1.ll
@@ -10,11 +10,11 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 declare i8* @strrchr(i8*, i32)
 
 define void @test_simplify1() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 6)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 6)
 ; CHECK-NOT: call i8* @strrchr
 ; CHECK: ret void
 
-  %str = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strrchr(i8* %str, i32 119)
   store i8* %dst, i8** @chp
   ret void
@@ -25,29 +25,29 @@ define void @test_simplify2() {
 ; CHECK-NOT: call i8* @strrchr
 ; CHECK: ret void
 
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %dst = call i8* @strrchr(i8* %str, i32 119)
   store i8* %dst, i8** @chp
   ret void
 }
 
 define void @test_simplify3() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 13)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
 ; CHECK-NOT: call i8* @strrchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strrchr(i8* %src, i32 0)
   store i8* %dst, i8** @chp
   ret void
 }
 
 define void @test_simplify4() {
-; CHECK: store i8* getelementptr inbounds ([14 x i8]* @hello, i32 0, i32 13)
+; CHECK: store i8* getelementptr inbounds ([14 x i8], [14 x i8]* @hello, i32 0, i32 13)
 ; CHECK-NOT: call i8* @strrchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strrchr(i8* %src, i32 65280)
   store i8* %dst, i8** @chp
   ret void
@@ -58,7 +58,7 @@ define void @test_nosimplify1(i32 %chr) {
 ; CHECK: call i8* @strrchr
 ; CHECK: ret void
 
-  %src = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %src = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8* @strrchr(i8* %src, i32 %chr)
   store i8* %dst, i8** @chp
   ret void
diff --git a/test/Transforms/InstCombine/strrchr-2.ll b/test/Transforms/InstCombine/strrchr-2.ll
index 1974f6ca6033..4c203d0e8e79 100644
--- a/test/Transforms/InstCombine/strrchr-2.ll
+++ b/test/Transforms/InstCombine/strrchr-2.ll
@@ -14,7 +14,7 @@ define void @test_nosimplify1() {
 ; CHECK: call i8 @strrchr
 ; CHECK: ret void
 
-  %str = getelementptr [14 x i8]* @hello, i32 0, i32 0
+  %str = getelementptr [14 x i8], [14 x i8]* @hello, i32 0, i32 0
   %dst = call i8 @strrchr(i8* %str, i32 119)
   store i8 %dst, i8* @chr
   ret void
diff --git a/test/Transforms/InstCombine/strspn-1.ll b/test/Transforms/InstCombine/strspn-1.ll
index ac940cce9e56..3d9c573ac617 100644
--- a/test/Transforms/InstCombine/strspn-1.ll
+++ b/test/Transforms/InstCombine/strspn-1.ll
@@ -14,7 +14,7 @@ declare i64 @strspn(i8*, i8*)
 
 define i64 @test_simplify1(i8* %str) {
 ; CHECK-LABEL: @test_simplify1(
-  %pat = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %pat = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i64 @strspn(i8* %str, i8* %pat)
   ret i64 %ret
@@ -25,7 +25,7 @@ define i64 @test_simplify1(i8* %str) {
 
 define i64 @test_simplify2(i8* %pat) {
 ; CHECK-LABEL: @test_simplify2(
-  %str = getelementptr [1 x i8]* @null, i32 0, i32 0
+  %str = getelementptr [1 x i8], [1 x i8]* @null, i32 0, i32 0
 
   %ret = call i64 @strspn(i8* %str, i8* %pat)
   ret i64 %ret
@@ -36,8 +36,8 @@ define i64 @test_simplify2(i8* %pat) {
 
 define i64 @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr [6 x i8]* @abcba, i32 0, i32 0
-  %pat = getelementptr [4 x i8]* @abc, i32 0, i32 0
+  %str = getelementptr [6 x i8], [6 x i8]* @abcba, i32 0, i32 0
+  %pat = getelementptr [4 x i8], [4 x i8]* @abc, i32 0, i32 0
 
   %ret = call i64 @strspn(i8* %str, i8* %pat)
   ret i64 %ret
diff --git a/test/Transforms/InstCombine/strstr-1.ll b/test/Transforms/InstCombine/strstr-1.ll
index a946dd32779a..d57e56ccdaf3 100644
--- a/test/Transforms/InstCombine/strstr-1.ll
+++ b/test/Transforms/InstCombine/strstr-1.ll
@@ -15,7 +15,7 @@ declare i8* @strstr(i8*, i8*)
 
 define i8* @test_simplify1(i8* %str) {
 ; CHECK-LABEL: @test_simplify1(
-  %pat = getelementptr inbounds [1 x i8]* @.str, i32 0, i32 0
+  %pat = getelementptr inbounds [1 x i8], [1 x i8]* @.str, i32 0, i32 0
   %ret = call i8* @strstr(i8* %str, i8* %pat)
   ret i8* %ret
 ; CHECK-NEXT: ret i8* %str
@@ -25,7 +25,7 @@ define i8* @test_simplify1(i8* %str) {
 
 define i8* @test_simplify2(i8* %str) {
 ; CHECK-LABEL: @test_simplify2(
-  %pat = getelementptr inbounds [2 x i8]* @.str1, i32 0, i32 0
+  %pat = getelementptr inbounds [2 x i8], [2 x i8]* @.str1, i32 0, i32 0
   %ret = call i8* @strstr(i8* %str, i8* %pat)
   ret i8* %ret
 ; CHECK-NEXT: @strchr(i8* %str, i32 97)
@@ -35,11 +35,11 @@ define i8* @test_simplify2(i8* %str) {
 
 define i8* @test_simplify3() {
 ; CHECK-LABEL: @test_simplify3(
-  %str = getelementptr inbounds [6 x i8]* @.str2, i32 0, i32 0
-  %pat = getelementptr inbounds [4 x i8]* @.str3, i32 0, i32 0
+  %str = getelementptr inbounds [6 x i8], [6 x i8]* @.str2, i32 0, i32 0
+  %pat = getelementptr inbounds [4 x i8], [4 x i8]* @.str3, i32 0, i32 0
   %ret = call i8* @strstr(i8* %str, i8* %pat)
   ret i8* %ret
-; CHECK-NEXT: getelementptr inbounds ([6 x i8]* @.str2, i64 0, i64 1)
+; CHECK-NEXT: getelementptr inbounds ([6 x i8], [6 x i8]* @.str2, i64 0, i64 1)
 }
 
 ; Check strstr(str, str) -> str.
diff --git a/test/Transforms/InstCombine/strstr-2.ll b/test/Transforms/InstCombine/strstr-2.ll
index 7b28ed0e691e..9cb33115c2ee 100644
--- a/test/Transforms/InstCombine/strstr-2.ll
+++ b/test/Transforms/InstCombine/strstr-2.ll
@@ -10,7 +10,7 @@ declare i8 @strstr(i8*, i8*)
 
 define i8 @test_no_simplify1(i8* %str) {
 ; CHECK-LABEL: @test_no_simplify1(
-  %pat = getelementptr inbounds [1 x i8]* @null, i32 0, i32 0
+  %pat = getelementptr inbounds [1 x i8], [1 x i8]* @null, i32 0, i32 0
   %ret = call i8 @strstr(i8* %str, i8* %pat)
 ; CHECK-NEXT: call i8 @strstr
   ret i8 %ret
diff --git a/test/Transforms/InstCombine/struct-assign-tbaa.ll b/test/Transforms/InstCombine/struct-assign-tbaa.ll
index 4e4b3f2a1a88..c75a839f3fb6 100644
--- a/test/Transforms/InstCombine/struct-assign-tbaa.ll
+++ b/test/Transforms/InstCombine/struct-assign-tbaa.ll
@@ -10,8 +10,8 @@ declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture, i64, i32,
 %struct.test1 = type { float }
 
 ; CHECK: @test
-; CHECK: %2 = load float* %0, align 4, !tbaa !0
-; CHECK: store float %2, float* %1, align 4, !tbaa !0
+; CHECK: %[[LOAD:.*]] = load i32, i32* %{{.*}}, align 4, !tbaa !0
+; CHECK: store i32 %[[LOAD:.*]], i32* %{{.*}}, align 4, !tbaa !0
 ; CHECK: ret
 define void @test1(%struct.test1* nocapture %a, %struct.test1* nocapture %b) {
 entry:
@@ -30,8 +30,8 @@ define i32 (i8*, i32*, double*)*** @test2() {
   %tmp = alloca %struct.test2, align 8
   %tmp1 = bitcast %struct.test2* %tmp to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp1, i8* undef, i64 8, i32 8, i1 false), !tbaa.struct !4
-  %tmp2 = getelementptr %struct.test2* %tmp, i32 0, i32 0
-  %tmp3 = load i32 (i8*, i32*, double*)*** %tmp2
+  %tmp2 = getelementptr %struct.test2, %struct.test2* %tmp, i32 0, i32 0
+  %tmp3 = load i32 (i8*, i32*, double*)**, i32 (i8*, i32*, double*)*** %tmp2
   ret i32 (i8*, i32*, double*)*** %tmp2
 }
 
diff --git a/test/Transforms/InstCombine/sub.ll b/test/Transforms/InstCombine/sub.ll
index 0e421f75f139..c76d8d0e2205 100644
--- a/test/Transforms/InstCombine/sub.ll
+++ b/test/Transforms/InstCombine/sub.ll
@@ -216,7 +216,7 @@ define zeroext i1 @test22(i32 %a, i32 %b)  nounwind  {
 
 ; rdar://7362831
 define i32 @test23(i8* %P, i64 %A){
-  %B = getelementptr inbounds i8* %P, i64 %A
+  %B = getelementptr inbounds i8, i8* %P, i64 %A
   %C = ptrtoint i8* %B to i64
   %D = trunc i64 %C to i32
   %E = ptrtoint i8* %P to i64
@@ -232,7 +232,7 @@ define i8 @test23_as1(i8 addrspace(1)* %P, i16 %A) {
 ; CHECK: @test23_as1
 ; CHECK-NEXT: = trunc i16 %A to i8
 ; CHECK-NEXT: ret i8
-  %B = getelementptr inbounds i8 addrspace(1)* %P, i16 %A
+  %B = getelementptr inbounds i8, i8 addrspace(1)* %P, i16 %A
   %C = ptrtoint i8 addrspace(1)* %B to i16
   %D = trunc i16 %C to i8
   %E = ptrtoint i8 addrspace(1)* %P to i16
@@ -242,7 +242,7 @@ define i8 @test23_as1(i8 addrspace(1)* %P, i16 %A) {
 }
 
 define i64 @test24(i8* %P, i64 %A){
-  %B = getelementptr inbounds i8* %P, i64 %A
+  %B = getelementptr inbounds i8, i8* %P, i64 %A
   %C = ptrtoint i8* %B to i64
   %E = ptrtoint i8* %P to i64
   %G = sub i64 %C, %E
@@ -254,7 +254,7 @@ define i64 @test24(i8* %P, i64 %A){
 define i16 @test24_as1(i8 addrspace(1)* %P, i16 %A) {
 ; CHECK: @test24_as1
 ; CHECK-NEXT: ret i16 %A
-  %B = getelementptr inbounds i8 addrspace(1)* %P, i16 %A
+  %B = getelementptr inbounds i8, i8 addrspace(1)* %P, i16 %A
   %C = ptrtoint i8 addrspace(1)* %B to i16
   %E = ptrtoint i8 addrspace(1)* %P to i16
   %G = sub i16 %C, %E
@@ -262,7 +262,7 @@ define i16 @test24_as1(i8 addrspace(1)* %P, i16 %A) {
 }
 
 define i64 @test24a(i8* %P, i64 %A){
-  %B = getelementptr inbounds i8* %P, i64 %A
+  %B = getelementptr inbounds i8, i8* %P, i64 %A
   %C = ptrtoint i8* %B to i64
   %E = ptrtoint i8* %P to i64
   %G = sub i64 %E, %C
@@ -276,7 +276,7 @@ define i16 @test24a_as1(i8 addrspace(1)* %P, i16 %A) {
 ; CHECK: @test24a_as1
 ; CHECK-NEXT: sub i16 0, %A
 ; CHECK-NEXT: ret i16
-  %B = getelementptr inbounds i8 addrspace(1)* %P, i16 %A
+  %B = getelementptr inbounds i8, i8 addrspace(1)* %P, i16 %A
   %C = ptrtoint i8 addrspace(1)* %B to i16
   %E = ptrtoint i8 addrspace(1)* %P to i16
   %G = sub i16 %E, %C
@@ -287,7 +287,7 @@ define i16 @test24a_as1(i8 addrspace(1)* %P, i16 %A) {
 @Arr = external global [42 x i16]
 
 define i64 @test24b(i8* %P, i64 %A){
-  %B = getelementptr inbounds [42 x i16]* @Arr, i64 0, i64 %A
+  %B = getelementptr inbounds [42 x i16], [42 x i16]* @Arr, i64 0, i64 %A
   %C = ptrtoint i16* %B to i64
   %G = sub i64 %C, ptrtoint ([42 x i16]* @Arr to i64)
   ret i64 %G
@@ -298,9 +298,9 @@ define i64 @test24b(i8* %P, i64 %A){
 
 
 define i64 @test25(i8* %P, i64 %A){
-  %B = getelementptr inbounds [42 x i16]* @Arr, i64 0, i64 %A
+  %B = getelementptr inbounds [42 x i16], [42 x i16]* @Arr, i64 0, i64 %A
   %C = ptrtoint i16* %B to i64
-  %G = sub i64 %C, ptrtoint (i16* getelementptr ([42 x i16]* @Arr, i64 1, i64 0) to i64)
+  %G = sub i64 %C, ptrtoint (i16* getelementptr ([42 x i16], [42 x i16]* @Arr, i64 1, i64 0) to i64)
   ret i64 %G
 ; CHECK-LABEL: @test25(
 ; CHECK-NEXT: shl nuw i64 %A, 1
@@ -316,9 +316,9 @@ define i16 @test25_as1(i8 addrspace(1)* %P, i64 %A) {
 ; CHECK-NEXT: shl nuw i16 %1, 1
 ; CHECK-NEXT: add i16 {{.*}}, -84
 ; CHECK-NEXT: ret i16
-  %B = getelementptr inbounds [42 x i16] addrspace(1)* @Arr_as1, i64 0, i64 %A
+  %B = getelementptr inbounds [42 x i16], [42 x i16] addrspace(1)* @Arr_as1, i64 0, i64 %A
   %C = ptrtoint i16 addrspace(1)* %B to i16
-  %G = sub i16 %C, ptrtoint (i16 addrspace(1)* getelementptr ([42 x i16] addrspace(1)* @Arr_as1, i64 1, i64 0) to i16)
+  %G = sub i16 %C, ptrtoint (i16 addrspace(1)* getelementptr ([42 x i16], [42 x i16] addrspace(1)* @Arr_as1, i64 1, i64 0) to i16)
   ret i16 %G
 }
 
@@ -353,8 +353,8 @@ define i32 @test28(i32 %x, i32 %y, i32 %z) {
 }
 
 define i64 @test29(i8* %foo, i64 %i, i64 %j) {
-  %gep1 = getelementptr inbounds i8* %foo, i64 %i
-  %gep2 = getelementptr inbounds i8* %foo, i64 %j
+  %gep1 = getelementptr inbounds i8, i8* %foo, i64 %i
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 %j
   %cast1 = ptrtoint i8* %gep1 to i64
   %cast2 = ptrtoint i8* %gep2 to i64
   %sub = sub i64 %cast1, %cast2
@@ -366,8 +366,8 @@ define i64 @test29(i8* %foo, i64 %i, i64 %j) {
 
 define i64 @test30(i8* %foo, i64 %i, i64 %j) {
   %bit = bitcast i8* %foo to i32*
-  %gep1 = getelementptr inbounds i32* %bit, i64 %i
-  %gep2 = getelementptr inbounds i8* %foo, i64 %j
+  %gep1 = getelementptr inbounds i32, i32* %bit, i64 %i
+  %gep2 = getelementptr inbounds i8, i8* %foo, i64 %j
   %cast1 = ptrtoint i32* %gep1 to i64
   %cast2 = ptrtoint i8* %gep2 to i64
   %sub = sub i64 %cast1, %cast2
@@ -384,8 +384,8 @@ define i16 @test30_as1(i8 addrspace(1)* %foo, i16 %i, i16 %j) {
 ; CHECK-NEXT: sub i16 %gep1.idx, %j
 ; CHECK-NEXT: ret i16
   %bit = bitcast i8 addrspace(1)* %foo to i32 addrspace(1)*
-  %gep1 = getelementptr inbounds i32 addrspace(1)* %bit, i16 %i
-  %gep2 = getelementptr inbounds i8 addrspace(1)* %foo, i16 %j
+  %gep1 = getelementptr inbounds i32, i32 addrspace(1)* %bit, i16 %i
+  %gep2 = getelementptr inbounds i8, i8 addrspace(1)* %foo, i16 %j
   %cast1 = ptrtoint i32 addrspace(1)* %gep1 to i16
   %cast2 = ptrtoint i8 addrspace(1)* %gep2 to i16
   %sub = sub i16 %cast1, %cast2
diff --git a/test/Transforms/InstCombine/switch-truncate-crash.ll b/test/Transforms/InstCombine/switch-truncate-crash.ll
new file mode 100644
index 000000000000..cc3c1ff28ed2
--- /dev/null
+++ b/test/Transforms/InstCombine/switch-truncate-crash.ll
@@ -0,0 +1,7 @@
+; RUN: opt -instcombine < %s
+
+define void @test() {
+  switch i32 0, label %out [i32 0, label %out]
+out:
+  ret void
+}
diff --git a/test/Transforms/InstCombine/type_pun.ll b/test/Transforms/InstCombine/type_pun.ll
new file mode 100644
index 000000000000..098164cd029f
--- /dev/null
+++ b/test/Transforms/InstCombine/type_pun.ll
@@ -0,0 +1,141 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+; Ensure that type punning using a union of vector and same-sized array
+; generates an extract instead of a shuffle with an uncommon vector size:
+;
+;   typedef uint32_t v4i32 __attribute__((vector_size(16)));
+;   union { v4i32 v; uint32_t a[4]; };
+;
+; This cleans up behind SROA, which inserts the uncommon vector size when
+; cleaning up the alloca/store/GEP/load.
+
+
+; Provide legal integer types.
+target datalayout = "p:32:32"
+
+
+; Extracting the zeroth element in an i32 array.
+define i32 @type_pun_zeroth(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_zeroth(
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <4 x i32>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <4 x i32> %[[BC]], i32 0
+; CHECK-NEXT: ret i32 %[[EXT]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %1 = bitcast <4 x i8> %sroa to i32
+  ret i32 %1
+}
+
+; Extracting the first element in an i32 array.
+define i32 @type_pun_first(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_first(
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <4 x i32>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <4 x i32> %[[BC]], i32 1
+; CHECK-NEXT: ret i32 %[[EXT]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 4, i32 5, i32 6, i32 7>
+  %1 = bitcast <4 x i8> %sroa to i32
+  ret i32 %1
+}
+
+; Extracting an i32 that isn't aligned to any natural boundary.
+define i32 @type_pun_misaligned(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_misaligned(
+; CHECK-NEXT: %[[SHUF:.*]] = shufflevector <16 x i8> %in, <16 x i8> undef, <16 x i32> <i32 6, i32 7, i32 8, i32 9, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef, i32 undef>
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %[[SHUF]] to <4 x i32>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <4 x i32> %[[BC]], i32 0
+; CHECK-NEXT: ret i32 %[[EXT]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 6, i32 7, i32 8, i32 9>
+  %1 = bitcast <4 x i8> %sroa to i32
+  ret i32 %1
+}
+
+; Type punning to an array of pointers.
+define i32* @type_pun_pointer(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_pointer(
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <4 x i32>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <4 x i32> %[[BC]], i32 0
+; CHECK-NEXT: %[[I2P:.*]] = inttoptr i32 %[[EXT]] to i32*
+; CHECK-NEXT: ret i32* %[[I2P]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %1 = bitcast <4 x i8> %sroa to i32
+  %2 = inttoptr i32 %1 to i32*
+  ret i32* %2
+}
+
+; Type punning to an array of 32-bit floating-point values.
+define float @type_pun_float(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_float(
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <4 x float>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <4 x float> %[[BC]], i32 0
+; CHECK-NEXT: ret float %[[EXT]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %1 = bitcast <4 x i8> %sroa to float
+  ret float %1
+}
+
+; Type punning to an array of 64-bit floating-point values.
+define double @type_pun_double(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_double(
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <2 x double>
+; CHECK-NEXT: %[[EXT:.*]] = extractelement <2 x double> %[[BC]], i32 0
+; CHECK-NEXT: ret double %[[EXT]]
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 4, i32 5, i32 6, i32 7>
+  %1 = bitcast <8 x i8> %sroa to double
+  ret double %1
+}
+
+; Type punning to same-size floating-point and integer values.
+; Verify that multiple uses with different bitcast types are properly handled.
+define { float, i32 } @type_pun_float_i32(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_float_i32(
+; CHECK-NEXT: %[[BCI:.*]] = bitcast <16 x i8> %in to <4 x i32>
+; CHECK-NEXT: %[[EXTI:.*]] = extractelement <4 x i32> %[[BCI]], i32 0
+; CHECK-NEXT: %[[BCF:.*]] = bitcast <16 x i8> %in to <4 x float>
+; CHECK-NEXT: %[[EXTF:.*]] = extractelement <4 x float> %[[BCF]], i32 0
+; CHECK-NEXT: %1 = insertvalue { float, i32 } undef, float %[[EXTF]], 0
+; CHECK-NEXT: %2 = insertvalue { float, i32 } %1, i32 %[[EXTI]], 1
+; CHECK-NEXT: ret { float, i32 } %2
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  %f = bitcast <4 x i8> %sroa to float
+  %i = bitcast <4 x i8> %sroa to i32
+  %1 = insertvalue { float, i32 } undef, float %f, 0
+  %2 = insertvalue { float, i32 } %1, i32 %i, 1
+  ret { float, i32 } %2
+}
+
+; Type punning two i32 values, with control flow.
+; Verify that the bitcast is shared and dominates usage.
+define i32 @type_pun_i32_ctrl(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_i32_ctrl(
+entry: ; CHECK-NEXT: entry:
+; CHECK-NEXT: %[[BC:.*]] = bitcast <16 x i8> %in to <4 x i32>
+; CHECK-NEXT: br
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <4 x i32> <i32 0, i32 1, i32 2, i32 3>
+  br i1 undef, label %left, label %right
+left: ; CHECK: left:
+; CHECK-NEXT: %[[EXTL:.*]] = extractelement <4 x i32> %[[BC]], i32 0
+; CHECK-NEXT: br
+  %lhs = bitcast <4 x i8> %sroa to i32
+  br label %tail
+right: ; CHECK: right:
+; CHECK-NEXT: %[[EXTR:.*]] = extractelement <4 x i32> %[[BC]], i32 0
+; CHECK-NEXT: br
+  %rhs = bitcast <4 x i8> %sroa to i32
+  br label %tail
+tail: ; CHECK: tail:
+; CHECK-NEXT: %i = phi i32 [ %[[EXTL]], %left ], [ %[[EXTR]], %right ]
+; CHECK-NEXT: ret i32 %i
+  %i = phi i32 [ %lhs, %left ], [ %rhs, %right ]
+  ret i32 %i
+}
+
+; Extracting a type that won't fit in a vector isn't handled. The function
+; should stay the same.
+define i40 @type_pun_unhandled(<16 x i8> %in) {
+; CHECK-LABEL: @type_pun_unhandled(
+; CHECK-NEXT: %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <5 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8>
+; CHECK-NEXT: %1 = bitcast <5 x i8> %sroa to i40
+; CHECK-NEXT: ret i40 %1
+  %sroa = shufflevector <16 x i8> %in, <16 x i8> undef, <5 x i32> <i32 4, i32 5, i32 6, i32 7, i32 8>
+  %1 = bitcast <5 x i8> %sroa to i40
+  ret i40 %1
+}
diff --git a/test/Transforms/InstCombine/unpack-fca.ll b/test/Transforms/InstCombine/unpack-fca.ll
new file mode 100644
index 000000000000..48bb157956aa
--- /dev/null
+++ b/test/Transforms/InstCombine/unpack-fca.ll
@@ -0,0 +1,114 @@
+; RUN: opt -instcombine -S < %s | FileCheck %s
+
+target datalayout = "e-i64:64-f80:128-n8:16:32:64"
+target triple = "x86_64-unknown-linux-gnu"
+
+%A__vtbl = type { i8*, i32 (%A*)* }
+%A = type { %A__vtbl* }
+
+@A__vtblZ = constant %A__vtbl { i8* null, i32 (%A*)* @A.foo }
+
+declare i32 @A.foo(%A* nocapture %this)
+
+declare i8* @allocmemory(i64)
+
+define void @storeA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to %A*
+; CHECK-LABEL: storeA
+; CHECK: store %A__vtbl* @A__vtblZ
+  store %A { %A__vtbl* @A__vtblZ }, %A* %1, align 8
+  ret void
+}
+
+define void @storeStructOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to { %A }*
+; CHECK-LABEL: storeStructOfA
+; CHECK: store %A__vtbl* @A__vtblZ
+  store { %A } { %A { %A__vtbl* @A__vtblZ } }, { %A }* %1, align 8
+  ret void
+}
+
+define void @storeArrayOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to [1 x %A]*
+; CHECK-LABEL: storeArrayOfA
+; CHECK: store %A__vtbl* @A__vtblZ
+  store [1 x %A] [%A { %A__vtbl* @A__vtblZ }], [1 x %A]* %1, align 8
+  ret void
+}
+
+define void @storeStructOfArrayOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to { [1 x %A] }*
+; CHECK-LABEL: storeStructOfArrayOfA
+; CHECK: store %A__vtbl* @A__vtblZ
+  store { [1 x %A] } { [1 x %A] [%A { %A__vtbl* @A__vtblZ }] }, { [1 x %A] }* %1, align 8
+  ret void
+}
+
+define %A @loadA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to %A*
+; CHECK-LABEL: loadA
+; CHECK: load %A__vtbl*,
+; CHECK: insertvalue %A undef, %A__vtbl* {{.*}}, 0
+  %2 = load %A, %A* %1, align 8
+  ret %A %2
+}
+
+define { %A } @loadStructOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to { %A }*
+; CHECK-LABEL: loadStructOfA
+; CHECK: load %A__vtbl*,
+; CHECK: insertvalue %A undef, %A__vtbl* {{.*}}, 0
+; CHECK: insertvalue { %A } undef, %A {{.*}}, 0
+  %2 = load { %A }, { %A }* %1, align 8
+  ret { %A } %2
+}
+
+define [1 x %A] @loadArrayOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to [1 x %A]*
+; CHECK-LABEL: loadArrayOfA
+; CHECK: load %A__vtbl*,
+; CHECK: insertvalue %A undef, %A__vtbl* {{.*}}, 0
+; CHECK: insertvalue [1 x %A] undef, %A {{.*}}, 0
+  %2 = load [1 x %A], [1 x %A]* %1, align 8
+  ret [1 x %A] %2
+}
+
+define { [1 x %A] } @loadStructOfArrayOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to { [1 x %A] }*
+; CHECK-LABEL: loadStructOfArrayOfA
+; CHECK: load %A__vtbl*,
+; CHECK: insertvalue %A undef, %A__vtbl* {{.*}}, 0
+; CHECK: insertvalue [1 x %A] undef, %A {{.*}}, 0
+; CHECK: insertvalue { [1 x %A] } undef, [1 x %A] {{.*}}, 0
+  %2 = load { [1 x %A] }, { [1 x %A] }* %1, align 8
+  ret { [1 x %A] } %2
+}
+
+define { %A } @structOfA() {
+body:
+  %0 = tail call i8* @allocmemory(i64 32)
+  %1 = bitcast i8* %0 to { %A }*
+; CHECK-LABEL: structOfA
+; CHECK: store %A__vtbl* @A__vtblZ
+  store { %A } { %A { %A__vtbl* @A__vtblZ } }, { %A }* %1, align 8
+  %2 = load { %A }, { %A }* %1, align 8
+; CHECK-NOT: load
+; CHECK: ret { %A } { %A { %A__vtbl* @A__vtblZ } }
+  ret { %A } %2
+}
diff --git a/test/Transforms/InstCombine/urem-simplify-bug.ll b/test/Transforms/InstCombine/urem-simplify-bug.ll
index 3e94ab5b2c33..1220dfdc77f0 100644
--- a/test/Transforms/InstCombine/urem-simplify-bug.ll
+++ b/test/Transforms/InstCombine/urem-simplify-bug.ll
@@ -22,8 +22,8 @@ bb14:		; preds = %entry
 	br label %bb15
 
 bb15:		; preds = %bb14, %bb
-	%iftmp.0.0 = phi i8* [ getelementptr ([5 x i8]* @.str1, i32 0, i32 0), %bb14 ], [ getelementptr ([5 x i8]* @.str, i32 0, i32 0), %bb ]		; <i8*> [#uses=1]
-	%tmp17 = call i32 (i8*, ...)* @printf( i8* %iftmp.0.0 ) nounwind 		; <i32> [#uses=0]
+	%iftmp.0.0 = phi i8* [ getelementptr ([5 x i8], [5 x i8]* @.str1, i32 0, i32 0), %bb14 ], [ getelementptr ([5 x i8], [5 x i8]* @.str, i32 0, i32 0), %bb ]		; <i8*> [#uses=1]
+	%tmp17 = call i32 (i8*, ...) @printf( i8* %iftmp.0.0 ) nounwind 		; <i32> [#uses=0]
 	ret i32 0
 }
 
diff --git a/test/Transforms/InstCombine/vec_demanded_elts.ll b/test/Transforms/InstCombine/vec_demanded_elts.ll
index 00a029aeab78..4245c7a3c134 100644
--- a/test/Transforms/InstCombine/vec_demanded_elts.ll
+++ b/test/Transforms/InstCombine/vec_demanded_elts.ll
@@ -110,7 +110,7 @@ define void @vac(<4 x float>* nocapture %a) nounwind {
 ; CHECK-NOT: load
 ; CHECK: ret
 entry:
-	%tmp1 = load <4 x float>* %a		; <<4 x float>> [#uses=1]
+	%tmp1 = load <4 x float>, <4 x float>* %a		; <<4 x float>> [#uses=1]
 	%vecins = insertelement <4 x float> %tmp1, float 0.000000e+00, i32 0	; <<4 x float>> [#uses=1]
 	%vecins4 = insertelement <4 x float> %vecins, float 0.000000e+00, i32 1; <<4 x float>> [#uses=1]
 	%vecins6 = insertelement <4 x float> %vecins4, float 0.000000e+00, i32 2; <<4 x float>> [#uses=1]
@@ -141,8 +141,8 @@ declare i64 @llvm.x86.sse2.cvttsd2si64(<2 x double>)
 ; <rdar://problem/6945110>
 define <4 x i32> @kernel3_vertical(<4 x i16> * %src, <8 x i16> * %foo) nounwind {
 entry:
-	%tmp = load <4 x i16>* %src
-	%tmp1 = load <8 x i16>* %foo
+	%tmp = load <4 x i16>, <4 x i16>* %src
+	%tmp1 = load <8 x i16>, <8 x i16>* %foo
 ; CHECK: %tmp2 = shufflevector
 	%tmp2 = shufflevector <4 x i16> %tmp, <4 x i16> undef, <8 x i32> <i32 0, i32 1, i32 2, i32 3, i32 undef, i32 undef, i32 undef, i32 undef>
 ; pmovzxwd ignores the upper 64-bits of its input; -instcombine should remove this shuffle:
@@ -350,19 +350,19 @@ define <8 x float> @test_vpermilvar_ps_256(<8 x float> %v) {
   ret <8 x float> %a
 }
 
-declare <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double>, <2 x i32>)
+declare <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double>, <2 x i64>)
 define <2 x double> @test_vpermilvar_pd(<2 x double> %v) {
 ; CHECK-LABEL: @test_vpermilvar_pd(
 ; CHECK: shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> <i32 1, i32 0>
-  %a = tail call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %v, <2 x i32> <i32 2, i32 0>)
+  %a = tail call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %v, <2 x i64> <i64 2, i64 0>)
   ret <2 x double> %a
 }
 
-declare <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double>, <4 x i32>)
+declare <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double>, <4 x i64>)
 define <4 x double> @test_vpermilvar_pd_256(<4 x double> %v) {
 ; CHECK-LABEL: @test_vpermilvar_pd_256(
 ; CHECK: shufflevector <4 x double> %v, <4 x double> undef, <4 x i32> <i32 1, i32 0, i32 3, i32 2>
-  %a = tail call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %v, <4 x i32> <i32 3, i32 1, i32 2, i32 0>)
+  %a = tail call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %v, <4 x i64> <i64 3, i64 1, i64 2, i64 0>)
   ret <4 x double> %a
 }
 
@@ -383,14 +383,14 @@ define <8 x float> @test_vpermilvar_ps_256_zero(<8 x float> %v) {
 define <2 x double> @test_vpermilvar_pd_zero(<2 x double> %v) {
 ; CHECK-LABEL: @test_vpermilvar_pd_zero(
 ; CHECK: shufflevector <2 x double> %v, <2 x double> undef, <2 x i32> zeroinitializer
-  %a = tail call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %v, <2 x i32> zeroinitializer)
+  %a = tail call <2 x double> @llvm.x86.avx.vpermilvar.pd(<2 x double> %v, <2 x i64> zeroinitializer)
   ret <2 x double> %a
 }
 
 define <4 x double> @test_vpermilvar_pd_256_zero(<4 x double> %v) {
 ; CHECK-LABEL: @test_vpermilvar_pd_256_zero(
 ; CHECK: shufflevector <4 x double> %v, <4 x double> undef, <4 x i32> <i32 0, i32 0, i32 2, i32 2>
-  %a = tail call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %v, <4 x i32> zeroinitializer)
+  %a = tail call <4 x double> @llvm.x86.avx.vpermilvar.pd.256(<4 x double> %v, <4 x i64> zeroinitializer)
   ret <4 x double> %a
 }
 
diff --git a/test/Transforms/InstCombine/vec_extract_var_elt.ll b/test/Transforms/InstCombine/vec_extract_var_elt.ll
index f6f9e0134a16..5c0610ff48bb 100644
--- a/test/Transforms/InstCombine/vec_extract_var_elt.ll
+++ b/test/Transforms/InstCombine/vec_extract_var_elt.ll
@@ -3,7 +3,7 @@
 define void @test (float %b, <8 x float> * %p)  {
 ; CHECK: extractelement
 ; CHECK: fptosi
-  %1 = load <8 x float> * %p
+  %1 = load <8 x float> , <8 x float> * %p
   %2 = bitcast <8 x float> %1 to <8 x i32>
   %3 = bitcast <8 x i32> %2 to <8 x float>
   %a = fptosi <8 x float> %3 to <8 x i32>
diff --git a/test/Transforms/InstCombine/vec_phi_extract.ll b/test/Transforms/InstCombine/vec_phi_extract.ll
index 1d778a000fb7..1079dc9e7198 100644
--- a/test/Transforms/InstCombine/vec_phi_extract.ll
+++ b/test/Transforms/InstCombine/vec_phi_extract.ll
@@ -16,7 +16,7 @@ loop:
   %end = icmp ult i32 %elt, %limit
   %3 = add i32 10, %elt
   %4 = sext i32 %elt to i64
-  %5 = getelementptr i32* %ptr, i64 %4
+  %5 = getelementptr i32, i32* %ptr, i64 %4
   store i32 %3, i32* %5
   %inc = add <16 x i32> %2, <i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16, i32 16>
   br i1 %end, label %loop, label %ret
diff --git a/test/Transforms/InstCombine/vec_shuffle.ll b/test/Transforms/InstCombine/vec_shuffle.ll
index eb4e9d6f8c30..d4d7f167ef07 100644
--- a/test/Transforms/InstCombine/vec_shuffle.ll
+++ b/test/Transforms/InstCombine/vec_shuffle.ll
@@ -190,11 +190,11 @@ define void @test14(i16 %conv10) {
   %tmp = alloca <4 x i16>, align 8
   %vecinit6 = insertelement <4 x i16> undef, i16 23, i32 3
   store <4 x i16> %vecinit6, <4 x i16>* undef
-  %tmp1 = load <4 x i16>* undef
+  %tmp1 = load <4 x i16>, <4 x i16>* undef
   %vecinit11 = insertelement <4 x i16> undef, i16 %conv10, i32 3
   %div = udiv <4 x i16> %tmp1, %vecinit11
   store <4 x i16> %div, <4 x i16>* %tmp
-  %tmp4 = load <4 x i16>* %tmp
+  %tmp4 = load <4 x i16>, <4 x i16>* %tmp
   %tmp5 = shufflevector <4 x i16> %tmp4, <4 x i16> undef, <2 x i32> <i32 2, i32 0>
   %cmp = icmp ule <2 x i16> %tmp5, undef
   %sext = sext <2 x i1> %cmp to <2 x i16>
@@ -414,3 +414,11 @@ define <4 x i32> @pr20114(<4 x i32> %__mask) {
   %masked_new.i.i.i = and <4 x i32> bitcast (<2 x i64> <i64 ptrtoint (<4 x i32> (<4 x i32>)* @pr20114 to i64), i64 ptrtoint (<4 x i32> (<4 x i32>)* @pr20114 to i64)> to <4 x i32>), %mask01.i
   ret <4 x i32> %masked_new.i.i.i
 }
+
+define <2 x i32*> @pr23113(<4 x i32*> %A) {
+; CHECK-LABEL: @pr23113
+; CHECK: %[[V:.*]] = shufflevector <4 x i32*> %A, <4 x i32*> undef, <2 x i32> <i32 0, i32 1>
+; CHECK-NEXT: ret <2 x i32*> %[[V]]
+  %1 = shufflevector <4 x i32*> %A, <4 x i32*> undef, <2 x i32> <i32 0, i32 1>
+  ret <2 x i32*> %1
+}
diff --git a/test/Transforms/InstCombine/vector-casts.ll b/test/Transforms/InstCombine/vector-casts.ll
index ca97b3407da6..727eb4ebb4c8 100644
--- a/test/Transforms/InstCombine/vector-casts.ll
+++ b/test/Transforms/InstCombine/vector-casts.ll
@@ -133,7 +133,7 @@ entry:
   %dim31 = insertelement <4 x i32> %dim30, i32 %a, i32 2
   %dim32 = insertelement <4 x i32> %dim31, i32 %a, i32 3
 
-  %offset_ptr = getelementptr <4 x float>* null, i32 1
+  %offset_ptr = getelementptr <4 x float>, <4 x float>* null, i32 1
   %offset_int = ptrtoint <4 x float>* %offset_ptr to i64
   %sizeof32 = trunc i64 %offset_int to i32
 
diff --git a/test/Transforms/InstCombine/vector_gep1.ll b/test/Transforms/InstCombine/vector_gep1.ll
index 90ca26212f2a..0f4e407dd7fe 100644
--- a/test/Transforms/InstCombine/vector_gep1.ll
+++ b/test/Transforms/InstCombine/vector_gep1.ll
@@ -17,26 +17,26 @@ define <2 x i1> @test2(<2 x i8*> %a) {
 }
 
 define <2 x i1> @test3(<2 x i8*> %a) {
-  %g = getelementptr <2 x i8*> %a, <2 x i32> <i32 1, i32 0>
+  %g = getelementptr i8, <2 x i8*> %a, <2 x i32> <i32 1, i32 0>
   %B = icmp ult <2 x i8*> %g, zeroinitializer
   ret <2 x i1> %B
 }
 
 define <1 x i1> @test4(<1 x i8*> %a) {
-  %g = getelementptr <1 x i8*> %a, <1 x i32> <i32 1>
+  %g = getelementptr i8, <1 x i8*> %a, <1 x i32> <i32 1>
   %B = icmp ult <1 x i8*> %g, zeroinitializer
   ret <1 x i1> %B
 }
 
 define <2 x i1> @test5(<2 x i8*> %a) {
-  %w = getelementptr <2 x i8*> %a, <2 x i32> zeroinitializer
-  %e = getelementptr <2 x i8*> %w, <2 x i32> <i32 5, i32 9>
-  %g = getelementptr <2 x i8*> %e, <2 x i32> <i32 1, i32 0>
+  %w = getelementptr i8, <2 x i8*> %a, <2 x i32> zeroinitializer
+  %e = getelementptr i8, <2 x i8*> %w, <2 x i32> <i32 5, i32 9>
+  %g = getelementptr i8, <2 x i8*> %e, <2 x i32> <i32 1, i32 0>
   %B = icmp ult <2 x i8*> %g, zeroinitializer
   ret <2 x i1> %B
 }
 
 define <2 x i32*> @test7(<2 x {i32, i32}*> %a) {
-  %w = getelementptr <2 x {i32, i32}*> %a, <2 x i32> <i32 5, i32 9>, <2 x i32> zeroinitializer
+  %w = getelementptr {i32, i32}, <2 x {i32, i32}*> %a, <2 x i32> <i32 5, i32 9>, <2 x i32> zeroinitializer
   ret <2 x i32*> %w
 }
diff --git a/test/Transforms/InstCombine/vector_gep2.ll b/test/Transforms/InstCombine/vector_gep2.ll
index 42057d60b8c8..d76a7d56cc7a 100644
--- a/test/Transforms/InstCombine/vector_gep2.ll
+++ b/test/Transforms/InstCombine/vector_gep2.ll
@@ -5,7 +5,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define <2 x i8*> @testa(<2 x i8*> %a) {
 ; CHECK-LABEL: @testa(
-  %g = getelementptr <2 x i8*> %a, <2 x i32> <i32 0, i32 1>
-; CHECK: getelementptr <2 x i8*> %a, <2 x i64> <i64 0, i64 1>
+  %g = getelementptr i8, <2 x i8*> %a, <2 x i32> <i32 0, i32 1>
+; CHECK: getelementptr i8, <2 x i8*> %a, <2 x i64> <i64 0, i64 1>
   ret <2 x i8*> %g
 }
diff --git a/test/Transforms/InstCombine/volatile_store.ll b/test/Transforms/InstCombine/volatile_store.ll
index 7cab199a58f4..7377b6815e29 100644
--- a/test/Transforms/InstCombine/volatile_store.ll
+++ b/test/Transforms/InstCombine/volatile_store.ll
@@ -5,7 +5,7 @@
 
 define void @self_assign_1() {
 entry:
-	%tmp = load volatile i32* @x		; <i32> [#uses=1]
+	%tmp = load volatile i32, i32* @x		; <i32> [#uses=1]
 	store volatile i32 %tmp, i32* @x
 	br label %return
 
diff --git a/test/Transforms/InstCombine/vsx-unaligned.ll b/test/Transforms/InstCombine/vsx-unaligned.ll
index 26e04268f446..ad264fb15b31 100644
--- a/test/Transforms/InstCombine/vsx-unaligned.ll
+++ b/test/Transforms/InstCombine/vsx-unaligned.ll
@@ -14,28 +14,28 @@ entry:
   %t1 = alloca <4 x float>*, align 8
   %t2 = alloca <2 x double>*, align 8
   store <4 x float>* @vf, <4 x float>** %t1, align 8
-  %0 = load <4 x float>** %t1, align 8
+  %0 = load <4 x float>*, <4 x float>** %t1, align 8
   %1 = bitcast <4 x float>* %0 to i8*
   %2 = call <4 x i32> @llvm.ppc.vsx.lxvw4x(i8* %1)
   store <4 x float>* @res_vf, <4 x float>** %t1, align 8
-  %3 = load <4 x float>** %t1, align 8
+  %3 = load <4 x float>*, <4 x float>** %t1, align 8
   %4 = bitcast <4 x float>* %3 to i8*
   call void @llvm.ppc.vsx.stxvw4x(<4 x i32> %2, i8* %4)
   store <2 x double>* @vd, <2 x double>** %t2, align 8
-  %5 = load <2 x double>** %t2, align 8
+  %5 = load <2 x double>*, <2 x double>** %t2, align 8
   %6 = bitcast <2 x double>* %5 to i8*
   %7 = call <2 x double> @llvm.ppc.vsx.lxvd2x(i8* %6)
   store <2 x double>* @res_vd, <2 x double>** %t2, align 8
-  %8 = load <2 x double>** %t2, align 8
+  %8 = load <2 x double>*, <2 x double>** %t2, align 8
   %9 = bitcast <2 x double>* %8 to i8*
   call void @llvm.ppc.vsx.stxvd2x(<2 x double> %7, i8* %9)
   ret void
 }
 
 ; CHECK-LABEL: @test1
-; CHECK: %0 = load <4 x i32>* bitcast (<4 x float>* @vf to <4 x i32>*), align 1
+; CHECK: %0 = load <4 x i32>, <4 x i32>* bitcast (<4 x float>* @vf to <4 x i32>*), align 1
 ; CHECK: store <4 x i32> %0, <4 x i32>* bitcast (<4 x float>* @res_vf to <4 x i32>*), align 1
-; CHECK: %1 = load <2 x double>* @vd, align 1
+; CHECK: %1 = load <2 x double>, <2 x double>* @vd, align 1
 ; CHECK: store <2 x double> %1, <2 x double>* @res_vd, align 1
 
 declare <4 x i32> @llvm.ppc.vsx.lxvw4x(i8*)
diff --git a/test/Transforms/InstCombine/weak-symbols.ll b/test/Transforms/InstCombine/weak-symbols.ll
index ec946ead8602..3f92e64d2ec0 100644
--- a/test/Transforms/InstCombine/weak-symbols.ll
+++ b/test/Transforms/InstCombine/weak-symbols.ll
@@ -13,8 +13,8 @@ define i32 @foo() nounwind {
 ; CHECK: ret i32 %temp1
 
 entry:
-  %str1 = getelementptr inbounds [2 x i8]* @fake_init, i64 0, i64 0
-  %str2 = getelementptr inbounds [2 x i8]* @.str, i64 0, i64 0
+  %str1 = getelementptr inbounds [2 x i8], [2 x i8]* @fake_init, i64 0, i64 0
+  %str2 = getelementptr inbounds [2 x i8], [2 x i8]* @.str, i64 0, i64 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2) nounwind readonly
   ret i32 %temp1
 }
@@ -24,8 +24,8 @@ define i32 @bar() nounwind {
 ; CHECK: ret i32 0
 
 entry:
-  %str1 = getelementptr inbounds [2 x i8]* @real_init, i64 0, i64 0
-  %str2 = getelementptr inbounds [2 x i8]* @.str, i64 0, i64 0
+  %str1 = getelementptr inbounds [2 x i8], [2 x i8]* @real_init, i64 0, i64 0
+  %str2 = getelementptr inbounds [2 x i8], [2 x i8]* @.str, i64 0, i64 0
   %temp1 = call i32 @strcmp(i8* %str1, i8* %str2) nounwind readonly
   ret i32 %temp1
 }
diff --git a/test/Transforms/InstCombine/x86-insertps.ll b/test/Transforms/InstCombine/x86-insertps.ll
new file mode 100644
index 000000000000..092bf7c6c624
--- /dev/null
+++ b/test/Transforms/InstCombine/x86-insertps.ll
@@ -0,0 +1,150 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+declare <4 x float> @llvm.x86.sse41.insertps(<4 x float>, <4 x float>, i8) nounwind readnone
+
+; This should never happen, but make sure we don't crash handling a non-constant immediate byte.
+
+define <4 x float> @insertps_non_const_imm(<4 x float> %v1, <4 x float> %v2, i8 %c) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 %c)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_non_const_imm
+; CHECK-NEXT:  call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 %c)
+; CHECK-NEXT:  ret <4 x float>
+}
+
+; If all zero mask bits are set, return a zero regardless of the other control bits.
+
+define <4 x float> @insertps_0x0f(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 15)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x0f
+; CHECK-NEXT:  ret <4 x float> zeroinitializer
+}
+define <4 x float> @insertps_0xff(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 255)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xff
+; CHECK-NEXT:  ret <4 x float> zeroinitializer
+}
+
+; If some zero mask bits are set that do not override the insertion, we do not change anything.
+
+define <4 x float> @insertps_0x0c(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 12)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x0c
+; CHECK-NEXT:  call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 12)
+; CHECK-NEXT:  ret <4 x float>
+}
+
+; ...unless both input vectors are the same operand.
+
+define <4 x float> @insertps_0x15_single_input(<4 x float> %v1) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v1, i8 21)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x15_single_input
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> <float 0.000000e+00, float undef, float 0.000000e+00, float undef>, <4 x i32> <i32 4, i32 0, i32 6, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+; The zero mask overrides the insertion lane.
+
+define <4 x float> @insertps_0x1a_single_input(<4 x float> %v1) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v1, i8 26)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x1a_single_input
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> <float undef, float 0.000000e+00, float undef, float 0.000000e+00>, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+; The zero mask overrides the insertion lane, so the second input vector is not used.
+
+define <4 x float> @insertps_0xc1(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 193)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xc1
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> <float 0.000000e+00, float undef, float undef, float undef>, <4 x i32> <i32 4, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+; If no zero mask bits are set, convert to a shuffle.
+
+define <4 x float> @insertps_0x00(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 0)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x00
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 4, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0x10(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 16)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x10
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 4, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0x20(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 32)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x20
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 1, i32 4, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0x30(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 48)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0x30
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 1, i32 2, i32 4>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0xc0(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 192)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xc0
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 7, i32 1, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0xd0(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 208)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xd0
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 7, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0xe0(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 224)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xe0
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 1, i32 7, i32 3>
+; CHECK-NEXT:  ret <4 x float>
+}
+
+define <4 x float> @insertps_0xf0(<4 x float> %v1, <4 x float> %v2) {
+  %res = call <4 x float> @llvm.x86.sse41.insertps(<4 x float> %v1, <4 x float> %v2, i8 240)
+  ret <4 x float> %res
+
+; CHECK-LABEL: @insertps_0xf0
+; CHECK-NEXT:  shufflevector <4 x float> %v1, <4 x float> %v2, <4 x i32> <i32 0, i32 1, i32 2, i32 7>
+; CHECK-NEXT:  ret <4 x float>
+}
+
diff --git a/test/Transforms/InstCombine/x86-vperm2.ll b/test/Transforms/InstCombine/x86-vperm2.ll
new file mode 100644
index 000000000000..5c60852dddb6
--- /dev/null
+++ b/test/Transforms/InstCombine/x86-vperm2.ll
@@ -0,0 +1,283 @@
+; RUN: opt < %s -instcombine -S | FileCheck %s
+
+; This should never happen, but make sure we don't crash handling a non-constant immediate byte.
+
+define <4 x double> @perm2pd_non_const_imm(<4 x double> %a0, <4 x double> %a1, i8 %b) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 %b) 
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_non_const_imm
+; CHECK-NEXT:  call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 %b)
+; CHECK-NEXT:  ret <4 x double>
+}
+
+
+; In the following 4 tests, both zero mask bits of the immediate are set.
+
+define <4 x double> @perm2pd_0x88(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 136) 
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x88
+; CHECK-NEXT:  ret <4 x double> zeroinitializer
+}
+
+define <8 x float> @perm2ps_0x88(<8 x float> %a0, <8 x float> %a1) {
+  %res = call <8 x float> @llvm.x86.avx.vperm2f128.ps.256(<8 x float> %a0, <8 x float> %a1, i8 136) 
+  ret <8 x float> %res
+
+; CHECK-LABEL: @perm2ps_0x88
+; CHECK-NEXT:  ret <8 x float> zeroinitializer
+}
+
+define <8 x i32> @perm2si_0x88(<8 x i32> %a0, <8 x i32> %a1) {
+  %res = call <8 x i32> @llvm.x86.avx.vperm2f128.si.256(<8 x i32> %a0, <8 x i32> %a1, i8 136) 
+  ret <8 x i32> %res
+
+; CHECK-LABEL: @perm2si_0x88
+; CHECK-NEXT:  ret <8 x i32> zeroinitializer
+}
+
+define <4 x i64> @perm2i_0x88(<4 x i64> %a0, <4 x i64> %a1) {
+  %res = call <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64> %a0, <4 x i64> %a1, i8 136) 
+  ret <4 x i64> %res
+
+; CHECK-LABEL: @perm2i_0x88
+; CHECK-NEXT:  ret <4 x i64> zeroinitializer
+}
+
+
+; The other control bits are ignored when zero mask bits of the immediate are set.
+
+define <4 x double> @perm2pd_0xff(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 255) 
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0xff
+; CHECK-NEXT:  ret <4 x double> zeroinitializer
+}
+
+
+; The following 16 tests are simple shuffles, except for 2 cases where we can just return one of the
+; source vectors. Verify that we generate the right shuffle masks and undef source operand where possible..
+
+define <4 x double> @perm2pd_0x00(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 0)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x00
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> undef, <4 x i32> <i32 0, i32 1, i32 0, i32 1>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x01(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 1)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x01
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> undef, <4 x i32> <i32 2, i32 3, i32 0, i32 1>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x02(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 2)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x02
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> %a0, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x03(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 3)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x03
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> %a0, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x10(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 16)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x10
+; CHECK-NEXT:  ret <4 x double> %a0
+}
+
+define <4 x double> @perm2pd_0x11(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 17)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x11
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> undef, <4 x i32> <i32 2, i32 3, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x12(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 18)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x12
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> %a0, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x13(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 19)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x13
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> %a0, <4 x i32> <i32 2, i32 3, i32 6, i32 7>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x20(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 32)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x20
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x21(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 33)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x21
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x22(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 34)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x22
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> undef, <4 x i32> <i32 0, i32 1, i32 0, i32 1>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x23(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 35)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x23
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> undef, <4 x i32> <i32 2, i32 3, i32 0, i32 1>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x30(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 48)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x30
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 0, i32 1, i32 6, i32 7>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x31(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 49)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x31
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a0, <4 x double> %a1, <4 x i32> <i32 2, i32 3, i32 6, i32 7>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+define <4 x double> @perm2pd_0x32(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 50)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x32
+; CHECK-NEXT:  ret <4 x double> %a1
+}
+
+define <4 x double> @perm2pd_0x33(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 51)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x33
+; CHECK-NEXT:  %1 = shufflevector <4 x double> %a1, <4 x double> undef, <4 x i32> <i32 2, i32 3, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x double> %1
+}
+
+; Confirm that a mask for 32-bit elements is also correct.
+
+define <8 x float> @perm2ps_0x31(<8 x float> %a0, <8 x float> %a1) {
+  %res = call <8 x float> @llvm.x86.avx.vperm2f128.ps.256(<8 x float> %a0, <8 x float> %a1, i8 49)
+  ret <8 x float> %res
+
+; CHECK-LABEL: @perm2ps_0x31
+; CHECK-NEXT:  %1 = shufflevector <8 x float> %a0, <8 x float> %a1, <8 x i32> <i32 4, i32 5, i32 6, i32 7, i32 12, i32 13, i32 14, i32 15>
+; CHECK-NEXT:  ret <8 x float> %1
+}
+
+
+; Confirm that the AVX2 version works the same.
+
+define <4 x i64> @perm2i_0x33(<4 x i64> %a0, <4 x i64> %a1) {
+  %res = call <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64> %a0, <4 x i64> %a1, i8 51)
+  ret <4 x i64> %res
+
+; CHECK-LABEL: @perm2i_0x33
+; CHECK-NEXT:  %1 = shufflevector <4 x i64> %a1, <4 x i64> undef, <4 x i32> <i32 2, i32 3, i32 2, i32 3>
+; CHECK-NEXT:  ret <4 x i64> %1
+}
+
+
+; Confirm that when a single zero mask bit is set, we replace a source vector with zeros.
+
+define <4 x double> @perm2pd_0x81(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 129)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x81
+; CHECK-NEXT:  shufflevector <4 x double> %a0, <4 x double> <double 0.0{{.*}}<4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double>
+}
+
+define <4 x double> @perm2pd_0x83(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 131)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x83
+; CHECK-NEXT:  shufflevector <4 x double> %a1, <4 x double> <double 0.0{{.*}}, <4 x i32> <i32 2, i32 3, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double>
+}
+
+define <4 x double> @perm2pd_0x28(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 40)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x28
+; CHECK-NEXT:  shufflevector <4 x double> <double 0.0{{.*}}, <4 x double> %a1, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double>
+}
+
+define <4 x double> @perm2pd_0x08(<4 x double> %a0, <4 x double> %a1) {
+  %res = call <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double> %a0, <4 x double> %a1, i8 8)
+  ret <4 x double> %res
+
+; CHECK-LABEL: @perm2pd_0x08
+; CHECK-NEXT:  shufflevector <4 x double> <double 0.0{{.*}}, <4 x double> %a0, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x double>
+}
+
+; Check one more with the AVX2 version.
+
+define <4 x i64> @perm2i_0x28(<4 x i64> %a0, <4 x i64> %a1) {
+  %res = call <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64> %a0, <4 x i64> %a1, i8 40)
+  ret <4 x i64> %res
+
+; CHECK-LABEL: @perm2i_0x28
+; CHECK-NEXT:  shufflevector <4 x i64> <i64 0{{.*}}, <4 x i64> %a1, <4 x i32> <i32 0, i32 1, i32 4, i32 5>
+; CHECK-NEXT:  ret <4 x i64>
+}
+
+declare <4 x double> @llvm.x86.avx.vperm2f128.pd.256(<4 x double>, <4 x double>, i8) nounwind readnone
+declare <8 x float> @llvm.x86.avx.vperm2f128.ps.256(<8 x float>, <8 x float>, i8) nounwind readnone
+declare <8 x i32> @llvm.x86.avx.vperm2f128.si.256(<8 x i32>, <8 x i32>, i8) nounwind readnone
+declare <4 x i64> @llvm.x86.avx2.vperm2i128(<4 x i64>, <4 x i64>, i8) nounwind readnone
+
diff --git a/test/Transforms/InstCombine/xor.ll b/test/Transforms/InstCombine/xor.ll
index c8debcbac226..33d5a2a9fda5 100644
--- a/test/Transforms/InstCombine/xor.ll
+++ b/test/Transforms/InstCombine/xor.ll
@@ -63,8 +63,8 @@ define i32 @test7(i32 %A, i32 %B) {
 ; CHECK-LABEL: @test7(
 ; CHECK-NEXT: %A1 = and i32 %A, 7
 ; CHECK-NEXT: %B1 = and i32 %B, 128
-; CHECK-NEXT: %C11 = or i32 %A1, %B1
-; CHECK-NEXT: ret i32 %C11
+; CHECK-NEXT: %C1.1 = or i32 %A1, %B1
+; CHECK-NEXT: ret i32 %C1.1
 	%A1 = and i32 %A, 7		; <i32> [#uses=1]
 	%B1 = and i32 %B, 128		; <i32> [#uses=1]
 	%C1 = xor i32 %A1, %B1		; <i32> [#uses=1]
@@ -96,8 +96,8 @@ define i1 @test9(i8 %A) {
 define i8 @test10(i8 %A) {
 ; CHECK-LABEL: @test10(
 ; CHECK-NEXT: %B = and i8 %A, 3
-; CHECK-NEXT: %C1 = or i8 %B, 4
-; CHECK-NEXT: ret i8 %C1
+; CHECK-NEXT: %C.1 = or i8 %B, 4
+; CHECK-NEXT: ret i8 %C.1
 	%B = and i8 %A, 3		; <i8> [#uses=1]
 	%C = xor i8 %B, 4		; <i8> [#uses=1]
 	ret i8 %C
diff --git a/test/Transforms/InstCombine/zext-or-icmp.ll b/test/Transforms/InstCombine/zext-or-icmp.ll
index ddc60834f586..3a27f9a124cb 100644
--- a/test/Transforms/InstCombine/zext-or-icmp.ll
+++ b/test/Transforms/InstCombine/zext-or-icmp.ll
@@ -10,19 +10,19 @@ entry:
 	%tmp3 = and i32 %tmp2, 2		; <i32> [#uses=1]
 	%tmp5 = and i32 %blk_i, 1		; <i32> [#uses=1]
 	%tmp6 = or i32 %tmp3, %tmp5		; <i32> [#uses=1]
-	%tmp8 = getelementptr %struct.FooBar* %up, i32 0, i32 7		; <i16*> [#uses=1]
-	%tmp9 = load i16* %tmp8, align 1		; <i16> [#uses=1]
+	%tmp8 = getelementptr %struct.FooBar, %struct.FooBar* %up, i32 0, i32 7		; <i16*> [#uses=1]
+	%tmp9 = load i16, i16* %tmp8, align 1		; <i16> [#uses=1]
 	%tmp910 = zext i16 %tmp9 to i32		; <i32> [#uses=1]
-	%tmp12 = getelementptr [4 x i8]* @some_idx, i32 0, i32 %tmp6		; <i8*> [#uses=1]
-	%tmp13 = load i8* %tmp12, align 1		; <i8> [#uses=1]
+	%tmp12 = getelementptr [4 x i8], [4 x i8]* @some_idx, i32 0, i32 %tmp6		; <i8*> [#uses=1]
+	%tmp13 = load i8, i8* %tmp12, align 1		; <i8> [#uses=1]
 	%tmp1314 = zext i8 %tmp13 to i32		; <i32> [#uses=1]
 	%tmp151 = lshr i32 %tmp910, %tmp1314		; <i32> [#uses=1]
 	%tmp1516 = trunc i32 %tmp151 to i8		; <i8> [#uses=1]
-	%tmp18 = getelementptr %struct.FooBar* %up, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp19 = load i8* %tmp18, align 1		; <i8> [#uses=1]
+	%tmp18 = getelementptr %struct.FooBar, %struct.FooBar* %up, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp19 = load i8, i8* %tmp18, align 1		; <i8> [#uses=1]
 	%tmp22 = and i8 %tmp1516, %tmp19		; <i8> [#uses=1]
-	%tmp24 = getelementptr %struct.FooBar* %up, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp25 = load i8* %tmp24, align 1		; <i8> [#uses=1]
+	%tmp24 = getelementptr %struct.FooBar, %struct.FooBar* %up, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp25 = load i8, i8* %tmp24, align 1		; <i8> [#uses=1]
 	%tmp26.mask = and i8 %tmp25, 1		; <i8> [#uses=1]
 	%toBool = icmp eq i8 %tmp26.mask, 0		; <i1> [#uses=1]
 	%toBool.not = xor i1 %toBool, true		; <i1> [#uses=1]
diff --git a/test/Transforms/InstMerge/ld_hoist1.ll b/test/Transforms/InstMerge/ld_hoist1.ll
index 715f1b824407..74c8900b8ab1 100644
--- a/test/Transforms/InstMerge/ld_hoist1.ll
+++ b/test/Transforms/InstMerge/ld_hoist1.ll
@@ -15,35 +15,35 @@ for.body.lr.ph:                                   ; preds = %entry
 
 ; CHECK-LABEL: for.body
 ; CHECK: load
-; CHECK:  %2 = getelementptr inbounds i32* %in, i64 %indvars.iv
-; CHECK:  %3 = load i32* %2, align 4
+; CHECK:  %2 = getelementptr inbounds i32, i32* %in, i64 %indvars.iv
+; CHECK:  %3 = load i32, i32* %2, align 4
 
 for.body:                                         ; preds = %for.body.lr.ph, %for.inc
   %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.inc ]
-  %arrayidx = getelementptr inbounds i32* %trigger, i64 %indvars.iv
-  %1 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %trigger, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %1, 0
   br i1 %cmp1, label %if.then, label %if.else
 
 ; CHECK-LABEL: if.then
 if.then:                                          ; preds = %for.body
 ; This load should be hoisted
-  %arrayidx3 = getelementptr inbounds i32* %in, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %in, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %conv = sitofp i32 %2 to float
   %add = fadd float %conv, 5.000000e-01
-  %arrayidx5 = getelementptr inbounds float* %out, i64 %indvars.iv
+  %arrayidx5 = getelementptr inbounds float, float* %out, i64 %indvars.iv
   store float %add, float* %arrayidx5, align 4
   br label %for.inc
 
 if.else:                                          ; preds = %for.body
-  %arrayidx7 = getelementptr inbounds float* %out, i64 %indvars.iv
-  %3 = load float* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %out, i64 %indvars.iv
+  %3 = load float, float* %arrayidx7, align 4
   %div = fdiv float %3, 3.000000e+00
   store float %div, float* %arrayidx7, align 4
 ; This load should be hoisted in spite of store 
-  %arrayidx9 = getelementptr inbounds i32* %in, i64 %indvars.iv
-  %4 = load i32* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds i32, i32* %in, i64 %indvars.iv
+  %4 = load i32, i32* %arrayidx9, align 4
   %conv10 = sitofp i32 %4 to float
   %add13 = fadd float %div, %conv10
   store float %add13, float* %arrayidx7, align 4
diff --git a/test/Transforms/InstMerge/ld_hoist_st_sink.ll b/test/Transforms/InstMerge/ld_hoist_st_sink.ll
index 978160aa002e..1d3f941882e5 100644
--- a/test/Transforms/InstMerge/ld_hoist_st_sink.ll
+++ b/test/Transforms/InstMerge/ld_hoist_st_sink.ll
@@ -8,8 +8,8 @@ target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 
 define i64 @foo(%struct.node* nocapture readonly %r) nounwind {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %tobool18 = icmp eq %struct.node* %node.017, null
   br i1 %tobool18, label %while.end, label %while.body.preheader
 
@@ -21,46 +21,46 @@ while.body.preheader:                             ; preds = %entry
 while.body:                                       ; preds = %while.body.preheader, %if.end
   %node.020 = phi %struct.node* [ %node.0, %if.end ], [ %node.017, %while.body.preheader ]
   %sum.019 = phi i64 [ %inc, %if.end ], [ 0, %while.body.preheader ]
-  %orientation = getelementptr inbounds %struct.node* %node.020, i64 0, i32 4
-  %0 = load i64* %orientation, align 8
+  %orientation = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 4
+  %0 = load i64, i64* %orientation, align 8
   %cmp = icmp eq i64 %0, 1
   br i1 %cmp, label %if.then, label %if.else
 ; CHECK: if.then
 if.then:                                          ; preds = %while.body
-  %a = getelementptr inbounds %struct.node* %node.020, i64 0, i32 5
+  %a = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 5
 ; CHECK-NOT: load %struct.arc
-  %1 = load %struct.arc** %a, align 8
-  %cost = getelementptr inbounds %struct.arc* %1, i64 0, i32 0
-; CHECK-NOT: load i64*
-  %2 = load i64* %cost, align 8
-  %pred = getelementptr inbounds %struct.node* %node.020, i64 0, i32 1
-; CHECK-NOT: load %struct.node**
-  %3 = load %struct.node** %pred, align 8
-  %p = getelementptr inbounds %struct.node* %3, i64 0, i32 6
-; CHECK-NOT: load i64*
-  %4 = load i64* %p, align 8
+  %1 = load %struct.arc*, %struct.arc** %a, align 8
+  %cost = getelementptr inbounds %struct.arc, %struct.arc* %1, i64 0, i32 0
+; CHECK-NOT: load i64, i64*
+  %2 = load i64, i64* %cost, align 8
+  %pred = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 1
+; CHECK-NOT: load %struct.node*, %struct.node**
+  %3 = load %struct.node*, %struct.node** %pred, align 8
+  %p = getelementptr inbounds %struct.node, %struct.node* %3, i64 0, i32 6
+; CHECK-NOT: load i64, i64*
+  %4 = load i64, i64* %p, align 8
   %add = add nsw i64 %4, %2
-  %p1 = getelementptr inbounds %struct.node* %node.020, i64 0, i32 6
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 6
 ; CHECK-NOT: store i64
   store i64 %add, i64* %p1, align 8
   br label %if.end
 
 ; CHECK: if.else
 if.else:                                          ; preds = %while.body
-  %pred2 = getelementptr inbounds %struct.node* %node.020, i64 0, i32 1
-; CHECK-NOT: load %struct.node**
-  %5 = load %struct.node** %pred2, align 8
-  %p3 = getelementptr inbounds %struct.node* %5, i64 0, i32 6
-; CHECK-NOT: load i64*
-  %6 = load i64* %p3, align 8
-  %a4 = getelementptr inbounds %struct.node* %node.020, i64 0, i32 5
-; CHECK-NOT: load %struct.arc**
-  %7 = load %struct.arc** %a4, align 8
-  %cost5 = getelementptr inbounds %struct.arc* %7, i64 0, i32 0
-; CHECK-NOT: load i64*
-  %8 = load i64* %cost5, align 8
+  %pred2 = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 1
+; CHECK-NOT: load %struct.node*, %struct.node**
+  %5 = load %struct.node*, %struct.node** %pred2, align 8
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %5, i64 0, i32 6
+; CHECK-NOT: load i64, i64*
+  %6 = load i64, i64* %p3, align 8
+  %a4 = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 5
+; CHECK-NOT: load %struct.arc*, %struct.arc**
+  %7 = load %struct.arc*, %struct.arc** %a4, align 8
+  %cost5 = getelementptr inbounds %struct.arc, %struct.arc* %7, i64 0, i32 0
+; CHECK-NOT: load i64, i64*
+  %8 = load i64, i64* %cost5, align 8
   %sub = sub nsw i64 %6, %8
-  %p6 = getelementptr inbounds %struct.node* %node.020, i64 0, i32 6
+  %p6 = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 6
 ; CHECK-NOT: store i64
   store i64 %sub, i64* %p6, align 8
   br label %if.end
@@ -69,8 +69,8 @@ if.else:                                          ; preds = %while.body
 if.end:                                           ; preds = %if.else, %if.then
 ; CHECK: store
   %inc = add nsw i64 %sum.019, 1
-  %node.0.in = getelementptr inbounds %struct.node* %node.020, i64 0, i32 2
-  %node.0 = load %struct.node** %node.0.in, align 8
+  %node.0.in = getelementptr inbounds %struct.node, %struct.node* %node.020, i64 0, i32 2
+  %node.0 = load %struct.node*, %struct.node** %node.0.in, align 8
   %tobool = icmp eq %struct.node* %node.0, null
   br i1 %tobool, label %while.end.loopexit, label %while.body
 
diff --git a/test/Transforms/InstMerge/st_sink_barrier_call.ll b/test/Transforms/InstMerge/st_sink_barrier_call.ll
index c158b006bee0..cdcc34624d5c 100644
--- a/test/Transforms/InstMerge/st_sink_barrier_call.ll
+++ b/test/Transforms/InstMerge/st_sink_barrier_call.ll
@@ -10,27 +10,27 @@ declare i32 @foo(i32 %x)
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK: store i32
   store i32 %1, i32* %p1, align 4
   br label %if.end
   
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK: store i32
   store i32 %add, i32* %p3, align 4
   call i32 @foo(i32 5)				  ;barrier
diff --git a/test/Transforms/InstMerge/st_sink_bugfix_22613.ll b/test/Transforms/InstMerge/st_sink_bugfix_22613.ll
index 6b817f20d16b..575f239efb23 100644
--- a/test/Transforms/InstMerge/st_sink_bugfix_22613.ll
+++ b/test/Transforms/InstMerge/st_sink_bugfix_22613.ll
@@ -20,52 +20,52 @@ target triple = "x86_64-unknown-linux-gnu"
 @a = common global i32 0, align 4
 
 ; Function Attrs: nounwind uwtable
-define void @fn1() #0 {
+define void @fn1() {
 entry:
   store i32 0, i32* @d, align 4
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc8, %entry
-  %0 = load i32* @d, align 4
+  %0 = load i32, i32* @d, align 4
   %cmp = icmp slt i32 %0, 2
   br i1 %cmp, label %for.body, label %for.end10
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* @d, align 4
+  %1 = load i32, i32* @d, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32* @b, align 4
+  %2 = load i32, i32* @b, align 4
   %idxprom1 = sext i32 %2 to i64
-  %arrayidx = getelementptr inbounds [1 x [3 x i8]]* @f, i32 0, i64 %idxprom1
-  %arrayidx2 = getelementptr inbounds [3 x i8]* %arrayidx, i32 0, i64 %idxprom
+  %arrayidx = getelementptr inbounds [1 x [3 x i8]], [1 x [3 x i8]]* @f, i32 0, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds [3 x i8], [3 x i8]* %arrayidx, i32 0, i64 %idxprom
   store i8 0, i8* %arrayidx2, align 1
   store i32 0, i32* @e, align 4
   br label %for.cond3
 
 for.cond3:                                        ; preds = %for.inc, %for.body
-  %3 = load i32* @e, align 4
+  %3 = load i32, i32* @e, align 4
   %cmp4 = icmp slt i32 %3, 3
   br i1 %cmp4, label %for.body5, label %for.end
 
 for.body5:                                        ; preds = %for.cond3
-  %4 = load i32* @c, align 4
+  %4 = load i32, i32* @c, align 4
   %tobool = icmp ne i32 %4, 0
   br i1 %tobool, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body5
-  %5 = load i32* @a, align 4
+  %5 = load i32, i32* @a, align 4
   %dec = add nsw i32 %5, -1
   store i32 %dec, i32* @a, align 4
   br label %if.end
 
 if.end:                                           ; preds = %if.then, %for.body5
-  %6 = load i32* @e, align 4
+  %6 = load i32, i32* @e, align 4
   %idxprom6 = sext i32 %6 to i64
-  %arrayidx7 = getelementptr inbounds [3 x i8]* getelementptr inbounds ([1 x [3 x i8]]* @f, i32 0, i64 0), i32 0, i64 %idxprom6
+  %arrayidx7 = getelementptr inbounds [3 x i8], [3 x i8]* getelementptr inbounds ([1 x [3 x i8]], [1 x [3 x i8]]* @f, i32 0, i64 0), i32 0, i64 %idxprom6
   store i8 1, i8* %arrayidx7, align 1
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %7 = load i32* @e, align 4
+  %7 = load i32, i32* @e, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* @e, align 4
   br label %for.cond3
@@ -74,7 +74,7 @@ for.end:                                          ; preds = %for.cond3
   br label %for.inc8
 
 for.inc8:                                         ; preds = %for.end
-  %8 = load i32* @d, align 4
+  %8 = load i32, i32* @d, align 4
   %inc9 = add nsw i32 %8, 1
   store i32 %inc9, i32* @d, align 4
   br label %for.cond
@@ -84,18 +84,18 @@ for.end10:                                        ; preds = %for.cond
 }
 
 ; Function Attrs: nounwind uwtable
-define i32 @main() #0 {
+define i32 @main() {
 entry:
   %retval = alloca i32, align 4
   store i32 0, i32* %retval
   call void @fn1()
-  %0 = load i8* getelementptr inbounds ([1 x [3 x i8]]* @f, i32 0, i64 0, i64 1), align 1
+  %0 = load i8, i8* getelementptr inbounds ([1 x [3 x i8]], [1 x [3 x i8]]* @f, i32 0, i64 0, i64 1), align 1
   %conv = sext i8 %0 to i32
   %cmp = icmp ne i32 %conv, 1
   br i1 %cmp, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  call void @abort() #2
+  call void @abort()
   unreachable
 
 if.end:                                           ; preds = %entry
@@ -103,12 +103,4 @@ if.end:                                           ; preds = %entry
 }
 
 ; Function Attrs: noreturn nounwind
-declare void @abort() #1
-
-attributes #0 = { nounwind uwtable "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
-attributes #1 = { noreturn nounwind "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "no-frame-pointer-elim-non-leaf" "no-infs-fp-math"="false" "no-nans-fp-math"="false" "stack-protector-buffer-size"="8" "unsafe-fp-math"="false" "use-soft-float"="false" }
-attributes #2 = { noreturn nounwind }
-
-!llvm.ident = !{!0}
-
-!0 = !{!"clang version 3.7.0 (trunk 229288) (llvm/trunk 229286:229290M)"}
+declare void @abort()
diff --git a/test/Transforms/InstMerge/st_sink_no_barrier_call.ll b/test/Transforms/InstMerge/st_sink_no_barrier_call.ll
index 72f1fdf77371..0ad90f8581eb 100644
--- a/test/Transforms/InstMerge/st_sink_no_barrier_call.ll
+++ b/test/Transforms/InstMerge/st_sink_no_barrier_call.ll
@@ -10,27 +10,27 @@ declare i32 @foo(i32 %x) #0
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %1, i32* %p1, align 4
   br label %if.end
   
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %add, i32* %p3, align 4
   call i32 @foo(i32 5)				  ;not a barrier
diff --git a/test/Transforms/InstMerge/st_sink_no_barrier_load.ll b/test/Transforms/InstMerge/st_sink_no_barrier_load.ll
index 5be0c254ba6f..b7236e4c6a06 100644
--- a/test/Transforms/InstMerge/st_sink_no_barrier_load.ll
+++ b/test/Transforms/InstMerge/st_sink_no_barrier_load.ll
@@ -8,30 +8,30 @@ target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %1, i32* %p1, align 4
-  %p2 = getelementptr inbounds %struct.node* %node.017, i32 5, i32 6
-  ; CHECK: load i32*
-  %not_barrier = load i32 * %p2, align 4
+  %p2 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 5, i32 6
+  ; CHECK: load i32, i32*
+  %not_barrier = load i32 , i32 * %p2, align 4
   br label %if.end
 
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %add, i32* %p3, align 4
   br label %if.end
diff --git a/test/Transforms/InstMerge/st_sink_no_barrier_store.ll b/test/Transforms/InstMerge/st_sink_no_barrier_store.ll
index 06e2b63ca3ed..e13f28aa5e17 100644
--- a/test/Transforms/InstMerge/st_sink_no_barrier_store.ll
+++ b/test/Transforms/InstMerge/st_sink_no_barrier_store.ll
@@ -8,29 +8,29 @@ target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %1, i32* %p1, align 4
   br label %if.end
 
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p2 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p2 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   store i32 %add, i32* %p2, align 4
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 5, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 5, i32 6
   ; CHECK: store i32
   store i32 %add, i32* %p3, align 4  			  ; This is not a barrier
   br label %if.end
diff --git a/test/Transforms/InstMerge/st_sink_two_stores.ll b/test/Transforms/InstMerge/st_sink_two_stores.ll
index 1f7c6aa478ba..5b5582f438c8 100644
--- a/test/Transforms/InstMerge/st_sink_two_stores.ll
+++ b/test/Transforms/InstMerge/st_sink_two_stores.ll
@@ -8,33 +8,33 @@ target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %1, i32* %p1, align 4
-  %p2 = getelementptr inbounds %struct.node* %node.017, i32 4, i32 6
+  %p2 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 4, i32 6
   ; CHECK-NOT: store i32
   store i32 %1, i32* %p2, align 4
   br label %if.end
 
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK-NOT: store i32
   store i32 %add, i32* %p3, align 4
-  %p4 = getelementptr inbounds %struct.node* %node.017, i32 4, i32 6
+  %p4 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 4, i32 6
   ; CHECK-NOT: store i32
   store i32 %2, i32* %p4, align 4  
   br label %if.end
diff --git a/test/Transforms/InstMerge/st_sink_with_barrier.ll b/test/Transforms/InstMerge/st_sink_with_barrier.ll
index d4efaa7efff7..a05ae88e3847 100644
--- a/test/Transforms/InstMerge/st_sink_with_barrier.ll
+++ b/test/Transforms/InstMerge/st_sink_with_barrier.ll
@@ -7,30 +7,30 @@ target datalayout = "e-m:o-i64:64-i128:128-n32:64-S128"
 ; Function Attrs: nounwind uwtable
 define void @sink_store(%struct.node* nocapture %r, i32 %index) {
 entry:
-  %node.0.in16 = getelementptr inbounds %struct.node* %r, i64 0, i32 2
-  %node.017 = load %struct.node** %node.0.in16, align 8
+  %node.0.in16 = getelementptr inbounds %struct.node, %struct.node* %r, i64 0, i32 2
+  %node.017 = load %struct.node*, %struct.node** %node.0.in16, align 8
   %index.addr = alloca i32, align 4
   store i32 %index, i32* %index.addr, align 4
-  %0 = load i32* %index.addr, align 4
+  %0 = load i32, i32* %index.addr, align 4
   %cmp = icmp slt i32 %0, 0
   br i1 %cmp, label %if.then, label %if.else
 
 ; CHECK: if.then
 if.then:                                          ; preds = %entry
-  %1 = load i32* %index.addr, align 4
-  %p1 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %1 = load i32, i32* %index.addr, align 4
+  %p1 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK: store i32
   store i32 %1, i32* %p1, align 4
-  %p2 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
-  ; CHECK: load i32*
-  %barrier = load i32 * %p2, align 4
+  %p2 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
+  ; CHECK: load i32, i32*
+  %barrier = load i32 , i32 * %p2, align 4
   br label %if.end
 
 ; CHECK: if.else
 if.else:                                          ; preds = %entry
-  %2 = load i32* %index.addr, align 4
+  %2 = load i32, i32* %index.addr, align 4
   %add = add nsw i32 %2, 1
-  %p3 = getelementptr inbounds %struct.node* %node.017, i32 0, i32 6
+  %p3 = getelementptr inbounds %struct.node, %struct.node* %node.017, i32 0, i32 6
   ; CHECK: store i32
   store i32 %add, i32* %p3, align 4
   br label %if.end
diff --git a/test/Transforms/InstSimplify/2011-10-27-BinOpCrash.ll b/test/Transforms/InstSimplify/2011-10-27-BinOpCrash.ll
index a10081a42dd5..5380a7bb6171 100644
--- a/test/Transforms/InstSimplify/2011-10-27-BinOpCrash.ll
+++ b/test/Transforms/InstSimplify/2011-10-27-BinOpCrash.ll
@@ -6,7 +6,7 @@
 @_ZN11xercesc_2_5L15gCombiningCharsE = external constant [163 x i16], align 2
 
 define i32 @_ZN11xercesc_2_515XMLRangeFactory11buildRangesEv(i32 %x) {
-  %a = add i32 %x, add (i32 add (i32 ashr (i32 add (i32 mul (i32 ptrtoint ([32 x i16]* @_ZN11xercesc_2_5L11gDigitCharsE to i32), i32 -1), i32 ptrtoint (i16* getelementptr inbounds ([32 x i16]* @_ZN11xercesc_2_5L11gDigitCharsE, i32 0, i32 30) to i32)), i32 1), i32 ashr (i32 add (i32 mul (i32 ptrtoint ([7 x i16]* @_ZN11xercesc_2_5L17gIdeographicCharsE to i32), i32 -1), i32 ptrtoint (i16* getelementptr inbounds ([7 x i16]* @_ZN11xercesc_2_5L17gIdeographicCharsE, i32 0, i32 4) to i32)), i32 1)), i32 8)
+  %a = add i32 %x, add (i32 add (i32 ashr (i32 add (i32 mul (i32 ptrtoint ([32 x i16]* @_ZN11xercesc_2_5L11gDigitCharsE to i32), i32 -1), i32 ptrtoint (i16* getelementptr inbounds ([32 x i16], [32 x i16]* @_ZN11xercesc_2_5L11gDigitCharsE, i32 0, i32 30) to i32)), i32 1), i32 ashr (i32 add (i32 mul (i32 ptrtoint ([7 x i16]* @_ZN11xercesc_2_5L17gIdeographicCharsE to i32), i32 -1), i32 ptrtoint (i16* getelementptr inbounds ([7 x i16], [7 x i16]* @_ZN11xercesc_2_5L17gIdeographicCharsE, i32 0, i32 4) to i32)), i32 1)), i32 8)
   %b = add i32 %a, %x
   ret i32 %b
 }
diff --git a/test/Transforms/InstSimplify/call-callconv.ll b/test/Transforms/InstSimplify/call-callconv.ll
index e475be781db9..56a8e44328fe 100644
--- a/test/Transforms/InstSimplify/call-callconv.ll
+++ b/test/Transforms/InstSimplify/call-callconv.ll
@@ -29,7 +29,7 @@ declare arm_aapcscc i32 @labs(i32) nounwind readnone
 
 define arm_aapcscc i32 @_strlen1() {
 ; CHECK: _strlen1
-  %call = tail call arm_aapcscc i32 @strlen(i8* getelementptr inbounds ([4 x i8]* @.str, i32 0, i32 0))
+  %call = tail call arm_aapcscc i32 @strlen(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i32 0, i32 0))
   ret i32 %call
 ; CHECK: ret i32 3
 }
@@ -42,7 +42,7 @@ define arm_aapcscc zeroext i1 @_strlen2(i8* %str) {
   %cmp = icmp ne i32 %call, 0
   ret i1 %cmp
 
-; CHECK: %[[STRLENFIRST:.*]] = load i8* %str
+; CHECK: %[[STRLENFIRST:.*]] = load i8, i8* %str
 ; CHECK: %[[CMP:.*]] = icmp ne i8 %[[STRLENFIRST]], 0
 ; CHECK: ret i1 %[[CMP]]
 }
diff --git a/test/Transforms/InstSimplify/call.ll b/test/Transforms/InstSimplify/call.ll
index fd854c5b4662..465529aef66a 100644
--- a/test/Transforms/InstSimplify/call.ll
+++ b/test/Transforms/InstSimplify/call.ll
@@ -1,6 +1,9 @@
 ; RUN: opt < %s -instsimplify -S | FileCheck %s
 
 declare {i8, i1} @llvm.uadd.with.overflow.i8(i8 %a, i8 %b)
+declare {i8, i1} @llvm.usub.with.overflow.i8(i8 %a, i8 %b)
+declare {i8, i1} @llvm.ssub.with.overflow.i8(i8 %a, i8 %b)
+declare {i8, i1} @llvm.umul.with.overflow.i8(i8 %a, i8 %b)
 
 define i1 @test_uadd1() {
 ; CHECK-LABEL: @test_uadd1(
@@ -18,6 +21,27 @@ define i8 @test_uadd2() {
 ; CHECK-NEXT: ret i8 42
 }
 
+define {i8, i1} @test_usub1(i8 %V) {
+; CHECK-LABEL: @test_usub1(
+  %x = call {i8, i1} @llvm.usub.with.overflow.i8(i8 %V, i8 %V)
+  ret {i8, i1} %x
+; CHECK-NEXT: ret { i8, i1 } zeroinitializer
+}
+
+define {i8, i1} @test_ssub1(i8 %V) {
+; CHECK-LABEL: @test_ssub1(
+  %x = call {i8, i1} @llvm.ssub.with.overflow.i8(i8 %V, i8 %V)
+  ret {i8, i1} %x
+; CHECK-NEXT: ret { i8, i1 } zeroinitializer
+}
+
+define {i8, i1} @test_umul1(i8 %V) {
+; CHECK-LABEL: @test_umul1(
+  %x = call {i8, i1} @llvm.umul.with.overflow.i8(i8 %V, i8 0)
+  ret {i8, i1} %x
+; CHECK-NEXT: ret { i8, i1 } zeroinitializer
+}
+
 declare i256 @llvm.cttz.i256(i256 %src, i1 %is_zero_undef)
 
 define i256 @test_cttz() {
@@ -109,7 +133,7 @@ entry:
   br i1 %cmp, label %cast.end, label %cast.notnull
 
 cast.notnull:                                     ; preds = %entry
-  %add.ptr = getelementptr inbounds i8* %call, i64 4
+  %add.ptr = getelementptr inbounds i8, i8* %call, i64 4
   br label %cast.end
 
 cast.end:                                         ; preds = %cast.notnull, %entry
@@ -132,7 +156,7 @@ entry:
   br i1 %cmp, label %cast.end, label %cast.notnull
 
 cast.notnull:                                     ; preds = %entry
-  %add.ptr = getelementptr inbounds i8* %call, i64 4
+  %add.ptr = getelementptr inbounds i8, i8* %call, i64 4
   br label %cast.end
 
 cast.end:                                         ; preds = %cast.notnull, %entry
@@ -152,7 +176,7 @@ entry:
   br i1 %cmp, label %cast.end, label %cast.notnull
 
 cast.notnull:                                     ; preds = %entry
-  %add.ptr = getelementptr inbounds i8* %call, i64 4
+  %add.ptr = getelementptr inbounds i8, i8* %call, i64 4
   br label %cast.end
 
 cast.end:                                         ; preds = %cast.notnull, %entry
diff --git a/test/Transforms/InstSimplify/compare.ll b/test/Transforms/InstSimplify/compare.ll
index 10c7ca6754e3..07c90d8f1eb8 100644
--- a/test/Transforms/InstSimplify/compare.ll
+++ b/test/Transforms/InstSimplify/compare.ll
@@ -24,7 +24,7 @@ define i1 @bitcast() {
 define i1 @gep() {
 ; CHECK-LABEL: @gep(
   %a = alloca [3 x i8], align 8
-  %x = getelementptr inbounds [3 x i8]* %a, i32 0, i32 0
+  %x = getelementptr inbounds [3 x i8], [3 x i8]* %a, i32 0, i32 0
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 false
@@ -33,8 +33,8 @@ define i1 @gep() {
 define i1 @gep2() {
 ; CHECK-LABEL: @gep2(
   %a = alloca [3 x i8], align 8
-  %x = getelementptr inbounds [3 x i8]* %a, i32 0, i32 0
-  %y = getelementptr inbounds [3 x i8]* %a, i32 0, i32 0
+  %x = getelementptr inbounds [3 x i8], [3 x i8]* %a, i32 0, i32 0
+  %y = getelementptr inbounds [3 x i8], [3 x i8]* %a, i32 0, i32 0
   %cmp = icmp eq i8* %x, %y
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 true
@@ -48,8 +48,8 @@ define i1 @gep2() {
 define i1 @gep3() {
 ; CHECK-LABEL: @gep3(
   %x = alloca %gept, align 8
-  %a = getelementptr %gept* %x, i64 0, i32 0
-  %b = getelementptr %gept* %x, i64 0, i32 1
+  %a = getelementptr %gept, %gept* %x, i64 0, i32 0
+  %b = getelementptr %gept, %gept* %x, i64 0, i32 1
   %equal = icmp eq i32* %a, %b
   ret i1 %equal
 ; CHECK-NEXT: ret i1 false
@@ -58,8 +58,8 @@ define i1 @gep3() {
 define i1 @gep4() {
 ; CHECK-LABEL: @gep4(
   %x = alloca %gept, align 8
-  %a = getelementptr %gept* @gepy, i64 0, i32 0
-  %b = getelementptr %gept* @gepy, i64 0, i32 1
+  %a = getelementptr %gept, %gept* @gepy, i64 0, i32 0
+  %b = getelementptr %gept, %gept* @gepy, i64 0, i32 1
   %equal = icmp eq i32* %a, %b
   ret i1 %equal
 ; CHECK-NEXT: ret i1 false
@@ -68,8 +68,8 @@ define i1 @gep4() {
 define i1 @gep5() {
 ; CHECK-LABEL: @gep5(
   %x = alloca %gept, align 8
-  %a = getelementptr inbounds %gept* %x, i64 0, i32 1
-  %b = getelementptr %gept* @gepy, i64 0, i32 0
+  %a = getelementptr inbounds %gept, %gept* %x, i64 0, i32 1
+  %b = getelementptr %gept, %gept* @gepy, i64 0, i32 0
   %equal = icmp eq i32* %a, %b
   ret i1 %equal
 ; CHECK-NEXT: ret i1 false
@@ -78,8 +78,8 @@ define i1 @gep5() {
 define i1 @gep6(%gept* %x) {
 ; Same as @gep3 but potentially null.
 ; CHECK-LABEL: @gep6(
-  %a = getelementptr %gept* %x, i64 0, i32 0
-  %b = getelementptr %gept* %x, i64 0, i32 1
+  %a = getelementptr %gept, %gept* %x, i64 0, i32 0
+  %b = getelementptr %gept, %gept* %x, i64 0, i32 1
   %equal = icmp eq i32* %a, %b
   ret i1 %equal
 ; CHECK-NEXT: ret i1 false
@@ -87,8 +87,8 @@ define i1 @gep6(%gept* %x) {
 
 define i1 @gep7(%gept* %x) {
 ; CHECK-LABEL: @gep7(
-  %a = getelementptr %gept* %x, i64 0, i32 0
-  %b = getelementptr %gept* @gepz, i64 0, i32 0
+  %a = getelementptr %gept, %gept* %x, i64 0, i32 0
+  %b = getelementptr %gept, %gept* @gepz, i64 0, i32 0
   %equal = icmp eq i32* %a, %b
   ret i1 %equal
 ; CHECK: ret i1 %equal
@@ -96,8 +96,8 @@ define i1 @gep7(%gept* %x) {
 
 define i1 @gep8(%gept* %x) {
 ; CHECK-LABEL: @gep8(
-  %a = getelementptr %gept* %x, i32 1
-  %b = getelementptr %gept* %x, i32 -1
+  %a = getelementptr %gept, %gept* %x, i32 1
+  %b = getelementptr %gept, %gept* %x, i32 -1
   %equal = icmp ugt %gept* %a, %b
   ret i1 %equal
 ; CHECK: ret i1 %equal
@@ -109,14 +109,14 @@ define i1 @gep9(i8* %ptr) {
 ; CHECK: ret i1 true
 
 entry:
-  %first1 = getelementptr inbounds i8* %ptr, i32 0
-  %first2 = getelementptr inbounds i8* %first1, i32 1
-  %first3 = getelementptr inbounds i8* %first2, i32 2
-  %first4 = getelementptr inbounds i8* %first3, i32 4
-  %last1 = getelementptr inbounds i8* %first2, i32 48
-  %last2 = getelementptr inbounds i8* %last1, i32 8
-  %last3 = getelementptr inbounds i8* %last2, i32 -4
-  %last4 = getelementptr inbounds i8* %last3, i32 -4
+  %first1 = getelementptr inbounds i8, i8* %ptr, i32 0
+  %first2 = getelementptr inbounds i8, i8* %first1, i32 1
+  %first3 = getelementptr inbounds i8, i8* %first2, i32 2
+  %first4 = getelementptr inbounds i8, i8* %first3, i32 4
+  %last1 = getelementptr inbounds i8, i8* %first2, i32 48
+  %last2 = getelementptr inbounds i8, i8* %last1, i32 8
+  %last3 = getelementptr inbounds i8, i8* %last2, i32 -4
+  %last4 = getelementptr inbounds i8, i8* %last3, i32 -4
   %first.int = ptrtoint i8* %first4 to i32
   %last.int = ptrtoint i8* %last4 to i32
   %cmp = icmp ne i32 %last.int, %first.int
@@ -129,10 +129,10 @@ define i1 @gep10(i8* %ptr) {
 ; CHECK: ret i1 true
 
 entry:
-  %first1 = getelementptr inbounds i8* %ptr, i32 -2
-  %first2 = getelementptr inbounds i8* %first1, i32 44
-  %last1 = getelementptr inbounds i8* %ptr, i32 48
-  %last2 = getelementptr inbounds i8* %last1, i32 -6
+  %first1 = getelementptr inbounds i8, i8* %ptr, i32 -2
+  %first2 = getelementptr inbounds i8, i8* %first1, i32 44
+  %last1 = getelementptr inbounds i8, i8* %ptr, i32 48
+  %last2 = getelementptr inbounds i8, i8* %last1, i32 -6
   %first.int = ptrtoint i8* %first2 to i32
   %last.int = ptrtoint i8* %last2 to i32
   %cmp = icmp eq i32 %last.int, %first.int
@@ -145,9 +145,9 @@ define i1 @gep11(i8* %ptr) {
 ; CHECK: ret i1 true
 
 entry:
-  %first1 = getelementptr inbounds i8* %ptr, i32 -2
-  %last1 = getelementptr inbounds i8* %ptr, i32 48
-  %last2 = getelementptr inbounds i8* %last1, i32 -6
+  %first1 = getelementptr inbounds i8, i8* %ptr, i32 -2
+  %last1 = getelementptr inbounds i8, i8* %ptr, i32 48
+  %last2 = getelementptr inbounds i8, i8* %last1, i32 -6
   %cmp = icmp ult i8* %first1, %last2
   ret i1 %cmp
 }
@@ -158,9 +158,9 @@ define i1 @gep12(i8* %ptr) {
 ; CHECK: ret i1 %cmp
 
 entry:
-  %first1 = getelementptr inbounds i8* %ptr, i32 -2
-  %last1 = getelementptr inbounds i8* %ptr, i32 48
-  %last2 = getelementptr inbounds i8* %last1, i32 -6
+  %first1 = getelementptr inbounds i8, i8* %ptr, i32 -2
+  %last1 = getelementptr inbounds i8, i8* %ptr, i32 48
+  %last2 = getelementptr inbounds i8, i8* %last1, i32 -6
   %cmp = icmp slt i8* %first1, %last2
   ret i1 %cmp
 }
@@ -168,7 +168,7 @@ entry:
 define i1 @gep13(i8* %ptr) {
 ; CHECK-LABEL: @gep13(
 ; We can prove this GEP is non-null because it is inbounds.
-  %x = getelementptr inbounds i8* %ptr, i32 1
+  %x = getelementptr inbounds i8, i8* %ptr, i32 1
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 false
@@ -178,7 +178,7 @@ define i1 @gep14({ {}, i8 }* %ptr) {
 ; CHECK-LABEL: @gep14(
 ; We can't simplify this because the offset of one in the GEP actually doesn't
 ; move the pointer.
-  %x = getelementptr inbounds { {}, i8 }* %ptr, i32 0, i32 1
+  %x = getelementptr inbounds { {}, i8 }, { {}, i8 }* %ptr, i32 0, i32 1
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NOT: ret i1 false
@@ -188,7 +188,7 @@ define i1 @gep15({ {}, [4 x {i8, i8}]}* %ptr, i32 %y) {
 ; CHECK-LABEL: @gep15(
 ; We can prove this GEP is non-null even though there is a user value, as we
 ; would necessarily violate inbounds on one side or the other.
-  %x = getelementptr inbounds { {}, [4 x {i8, i8}]}* %ptr, i32 0, i32 1, i32 %y, i32 1
+  %x = getelementptr inbounds { {}, [4 x {i8, i8}]}, { {}, [4 x {i8, i8}]}* %ptr, i32 0, i32 1, i32 %y, i32 1
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 false
@@ -199,7 +199,7 @@ define i1 @gep16(i8* %ptr, i32 %a) {
 ; We can prove this GEP is non-null because it is inbounds and because we know
 ; %b is non-zero even though we don't know its value.
   %b = or i32 %a, 1
-  %x = getelementptr inbounds i8* %ptr, i32 %b
+  %x = getelementptr inbounds i8, i8* %ptr, i32 %b
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 false
@@ -687,7 +687,7 @@ define <2 x i1> @vectorselectcrash(i32 %arg1) {
 ; PR12013
 define i1 @alloca_compare(i64 %idx) {
   %sv = alloca { i32, i32, [124 x i32] }
-  %1 = getelementptr inbounds { i32, i32, [124 x i32] }* %sv, i32 0, i32 2, i64 %idx
+  %1 = getelementptr inbounds { i32, i32, [124 x i32] }, { i32, i32, [124 x i32] }* %sv, i32 0, i32 2, i64 %idx
   %2 = icmp eq i32* %1, null
   ret i1 %2
   ; CHECK: alloca_compare
@@ -699,7 +699,7 @@ define i1 @infinite_gep() {
   ret i1 1
 
 unreachableblock:
-  %X = getelementptr i32 *%X, i32 1
+  %X = getelementptr i32, i32 *%X, i32 1
   %Y = icmp eq i32* %X, null
   ret i1 %Y
 }
@@ -744,7 +744,7 @@ define i1 @alloca_gep(i64 %a, i64 %b) {
 ; We can prove this GEP is non-null because it is inbounds and the pointer
 ; is non-null.
   %strs = alloca [1000 x [1001 x i8]], align 16
-  %x = getelementptr inbounds [1000 x [1001 x i8]]* %strs, i64 0, i64 %a, i64 %b
+  %x = getelementptr inbounds [1000 x [1001 x i8]], [1000 x [1001 x i8]]* %strs, i64 0, i64 %a, i64 %b
   %cmp = icmp eq i8* %x, null
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 false
@@ -753,10 +753,10 @@ define i1 @alloca_gep(i64 %a, i64 %b) {
 define i1 @non_inbounds_gep_compare(i64* %a) {
 ; CHECK-LABEL: @non_inbounds_gep_compare(
 ; Equality compares with non-inbounds GEPs can be folded.
-  %x = getelementptr i64* %a, i64 42
-  %y = getelementptr inbounds i64* %x, i64 -42
-  %z = getelementptr i64* %a, i64 -42
-  %w = getelementptr inbounds i64* %z, i64 42
+  %x = getelementptr i64, i64* %a, i64 42
+  %y = getelementptr inbounds i64, i64* %x, i64 -42
+  %z = getelementptr i64, i64* %a, i64 -42
+  %w = getelementptr inbounds i64, i64* %z, i64 42
   %cmp = icmp eq i64* %y, %w
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 true
@@ -765,8 +765,8 @@ define i1 @non_inbounds_gep_compare(i64* %a) {
 define i1 @non_inbounds_gep_compare2(i64* %a) {
 ; CHECK-LABEL: @non_inbounds_gep_compare2(
 ; Equality compares with non-inbounds GEPs can be folded.
-  %x = getelementptr i64* %a, i64 4294967297
-  %y = getelementptr i64* %a, i64 1
+  %x = getelementptr i64, i64* %a, i64 4294967297
+  %y = getelementptr i64, i64* %a, i64 1
   %cmp = icmp eq i64* %y, %y
   ret i1 %cmp
 ; CHECK-NEXT: ret i1 true
@@ -918,7 +918,7 @@ define i1 @nonnull_deref_as_arg(i32 addrspace(1)* dereferenceable(4) %i) {
   ret i1 %cmp
 ; CHECK-LABEL: @nonnull_deref_as_arg
 ; CHECK: icmp
-; CHECK ret
+; CHECK: ret
 }
 
 declare nonnull i32* @returns_nonnull_helper()
@@ -950,7 +950,7 @@ define i1 @returns_nonnull_as_deref() {
 }
 
 define i1 @nonnull_load(i32** %addr) {
-  %ptr = load i32** %addr, !nonnull !{}
+  %ptr = load i32*, i32** %addr, !nonnull !{}
   %cmp = icmp eq i32* %ptr, null
   ret i1 %cmp
 ; CHECK-LABEL: @nonnull_load
@@ -958,14 +958,14 @@ define i1 @nonnull_load(i32** %addr) {
 }
 
 define i1 @nonnull_load_as_outer(i32* addrspace(1)* %addr) {
-  %ptr = load i32* addrspace(1)* %addr, !nonnull !{}
+  %ptr = load i32*, i32* addrspace(1)* %addr, !nonnull !{}
   %cmp = icmp eq i32* %ptr, null
   ret i1 %cmp
 ; CHECK-LABEL: @nonnull_load_as_outer
 ; CHECK: ret i1 false
 }
 define i1 @nonnull_load_as_inner(i32 addrspace(1)** %addr) {
-  %ptr = load i32 addrspace(1)** %addr, !nonnull !{}
+  %ptr = load i32 addrspace(1)*, i32 addrspace(1)** %addr, !nonnull !{}
   %cmp = icmp eq i32 addrspace(1)* %ptr, null
   ret i1 %cmp
 ; CHECK-LABEL: @nonnull_load_as_inner
diff --git a/test/Transforms/InstSimplify/fast-math.ll b/test/Transforms/InstSimplify/fast-math.ll
index 71d1ed838491..e7fb14d7e786 100644
--- a/test/Transforms/InstSimplify/fast-math.ll
+++ b/test/Transforms/InstSimplify/fast-math.ll
@@ -105,3 +105,12 @@ define float @nofold_fadd_x_0(float %a) {
 ; CHECK: ret float %no_zero
   ret float %no_zero
 }
+
+; fdiv nsz nnan 0, X ==> 0
+define double @fdiv_zero_by_x(double %X) {
+; CHECK-LABEL: @fdiv_zero_by_x(
+; 0 / X -> 0
+  %r = fdiv nnan nsz double 0.0, %X
+  ret double %r
+; CHECK: ret double 0
+}
diff --git a/test/Transforms/InstSimplify/floating-point-arithmetic.ll b/test/Transforms/InstSimplify/floating-point-arithmetic.ll
index 8177440472cb..b0957a817739 100644
--- a/test/Transforms/InstSimplify/floating-point-arithmetic.ll
+++ b/test/Transforms/InstSimplify/floating-point-arithmetic.ll
@@ -33,3 +33,29 @@ define double @fmul_X_1(double %a) {
   ; CHECK: ret double %a
   ret double %b
 }
+
+; We can't optimize away the fadd in this test because the input
+; value to the function and subsequently to the fadd may be -0.0. 
+; In that one special case, the result of the fadd should be +0.0
+; rather than the first parameter of the fadd.
+
+; Fragile test warning: We need 6 sqrt calls to trigger the bug 
+; because the internal logic has a magic recursion limit of 6. 
+; This is presented without any explanation or ability to customize.
+
+declare float @sqrtf(float)
+
+define float @PR22688(float %x) {
+  %1 = call float @sqrtf(float %x)
+  %2 = call float @sqrtf(float %1)
+  %3 = call float @sqrtf(float %2)
+  %4 = call float @sqrtf(float %3)
+  %5 = call float @sqrtf(float %4)
+  %6 = call float @sqrtf(float %5)
+  %7 = fadd float %6, 0.0
+  ret float %7
+
+; CHECK-LABEL: @PR22688(
+; CHECK: fadd float %6, 0.0
+}
+
diff --git a/test/Transforms/InstSimplify/floating-point-compare.ll b/test/Transforms/InstSimplify/floating-point-compare.ll
new file mode 100644
index 000000000000..af48d062b4f6
--- /dev/null
+++ b/test/Transforms/InstSimplify/floating-point-compare.ll
@@ -0,0 +1,60 @@
+; RUN: opt < %s -instsimplify -S | FileCheck %s
+
+; These tests choose arbitrarily between float and double,
+; and between uge and olt, to give reasonble coverage 
+; without combinatorial explosion.
+
+declare float @llvm.fabs.f32(float)
+declare float @llvm.sqrt.f32(float)
+declare double @llvm.powi.f64(double,i32)
+declare float @llvm.exp.f32(float)
+declare double @llvm.exp2.f64(double)
+declare float @llvm.fma.f32(float,float,float)
+
+declare void @expect_equal(i1,i1)
+
+; CHECK-LABEL: @orderedLessZeroTree(
+define i1 @orderedLessZeroTree(float,float,float,float) {
+  %square = fmul float %0, %0
+  %abs = call float @llvm.fabs.f32(float %1)
+  %sqrt = call float @llvm.sqrt.f32(float %2)
+  %fma = call float @llvm.fma.f32(float %3, float %3, float %sqrt)
+  %div = fdiv float %square, %abs
+  %rem = frem float %sqrt, %fma
+  %add = fadd float %div, %rem
+  %uge = fcmp uge float %add, 0.000000e+00
+; CHECK: ret i1 true
+  ret i1 %uge
+}
+
+; CHECK-LABEL: @orderedLessZeroExpExt(
+define i1 @orderedLessZeroExpExt(float) {
+  %a = call float @llvm.exp.f32(float %0)
+  %b = fpext float %a to double
+  %uge = fcmp uge double %b, 0.000000e+00
+; CHECK: ret i1 true
+  ret i1 %uge
+}
+
+; CHECK-LABEL: @orderedLessZeroExp2Trunc(
+define i1 @orderedLessZeroExp2Trunc(double) {
+  %a = call double @llvm.exp2.f64(double %0)
+  %b = fptrunc double %a to float
+  %olt = fcmp olt float %b, 0.000000e+00
+; CHECK: ret i1 false
+  ret i1 %olt
+}
+
+; CHECK-LABEL: @orderedLessZeroPowi(
+define i1 @orderedLessZeroPowi(double,double) {
+  ; Even constant exponent
+  %a = call double @llvm.powi.f64(double %0, i32 2)
+  %square = fmul double %1, %1
+  ; Odd constant exponent with provably non-negative base
+  %b = call double @llvm.powi.f64(double %square, i32 3)
+  %c = fadd double %a, %b
+  %olt = fcmp olt double %b, 0.000000e+00
+; CHECK: ret i1 false
+  ret i1 %olt
+}
+
diff --git a/test/Transforms/InstSimplify/gep.ll b/test/Transforms/InstSimplify/gep.ll
index 49a97f133cef..13640e7631d5 100644
--- a/test/Transforms/InstSimplify/gep.ll
+++ b/test/Transforms/InstSimplify/gep.ll
@@ -9,7 +9,7 @@ define %struct.A* @test1(%struct.A* %b, %struct.A* %e) {
   %b_ptr = ptrtoint %struct.A* %b to i64
   %sub = sub i64 %e_ptr, %b_ptr
   %sdiv = sdiv exact i64 %sub, 7
-  %gep = getelementptr inbounds %struct.A* %b, i64 %sdiv
+  %gep = getelementptr inbounds %struct.A, %struct.A* %b, i64 %sdiv
   ret %struct.A* %gep
 ; CHECK-LABEL: @test1
 ; CHECK-NEXT: ret %struct.A* %e
@@ -19,7 +19,7 @@ define i8* @test2(i8* %b, i8* %e) {
   %e_ptr = ptrtoint i8* %e to i64
   %b_ptr = ptrtoint i8* %b to i64
   %sub = sub i64 %e_ptr, %b_ptr
-  %gep = getelementptr inbounds i8* %b, i64 %sub
+  %gep = getelementptr inbounds i8, i8* %b, i64 %sub
   ret i8* %gep
 ; CHECK-LABEL: @test2
 ; CHECK-NEXT: ret i8* %e
@@ -30,7 +30,7 @@ define i64* @test3(i64* %b, i64* %e) {
   %b_ptr = ptrtoint i64* %b to i64
   %sub = sub i64 %e_ptr, %b_ptr
   %ashr = ashr exact i64 %sub, 3
-  %gep = getelementptr inbounds i64* %b, i64 %ashr
+  %gep = getelementptr inbounds i64, i64* %b, i64 %ashr
   ret i64* %gep
 ; CHECK-LABEL: @test3
 ; CHECK-NEXT: ret i64* %e
@@ -40,7 +40,7 @@ define %struct.A* @test4(%struct.A* %b) {
   %b_ptr = ptrtoint %struct.A* %b to i64
   %sub = sub i64 0, %b_ptr
   %sdiv = sdiv exact i64 %sub, 7
-  %gep = getelementptr inbounds %struct.A* %b, i64 %sdiv
+  %gep = getelementptr inbounds %struct.A, %struct.A* %b, i64 %sdiv
   ret %struct.A* %gep
 ; CHECK-LABEL: @test4
 ; CHECK-NEXT: ret %struct.A* null
@@ -49,7 +49,7 @@ define %struct.A* @test4(%struct.A* %b) {
 define i8* @test5(i8* %b) {
   %b_ptr = ptrtoint i8* %b to i64
   %sub = sub i64 0, %b_ptr
-  %gep = getelementptr inbounds i8* %b, i64 %sub
+  %gep = getelementptr inbounds i8, i8* %b, i64 %sub
   ret i8* %gep
 ; CHECK-LABEL: @test5
 ; CHECK-NEXT: ret i8* null
@@ -59,7 +59,7 @@ define i64* @test6(i64* %b) {
   %b_ptr = ptrtoint i64* %b to i64
   %sub = sub i64 0, %b_ptr
   %ashr = ashr exact i64 %sub, 3
-  %gep = getelementptr inbounds i64* %b, i64 %ashr
+  %gep = getelementptr inbounds i64, i64* %b, i64 %ashr
   ret i64* %gep
 ; CHECK-LABEL: @test6
 ; CHECK-NEXT: ret i64* null
@@ -69,7 +69,7 @@ define i8* @test7(i8* %b, i8** %e) {
   %e_ptr = ptrtoint i8** %e to i64
   %b_ptr = ptrtoint i8* %b to i64
   %sub = sub i64 %e_ptr, %b_ptr
-  %gep = getelementptr inbounds i8* %b, i64 %sub
+  %gep = getelementptr inbounds i8, i8* %b, i64 %sub
   ret i8* %gep
 ; CHECK-LABEL: @test7
 ; CHECK-NEXT: ptrtoint
diff --git a/test/Transforms/InstSimplify/load.ll b/test/Transforms/InstSimplify/load.ll
index 92953cd0ebfb..ab87d4b9c535 100644
--- a/test/Transforms/InstSimplify/load.ll
+++ b/test/Transforms/InstSimplify/load.ll
@@ -6,14 +6,14 @@
 define i32 @crash_on_zeroinit() {
 ; CHECK-LABEL: @crash_on_zeroinit
 ; CHECK: ret i32 0
-  %load = load i32* bitcast ({}* @zeroinit to i32*)
+  %load = load i32, i32* bitcast ({}* @zeroinit to i32*)
   ret i32 %load
 }
 
 define i32 @crash_on_undef() {
 ; CHECK-LABEL: @crash_on_undef
 ; CHECK: ret i32 undef
-  %load = load i32* bitcast ({}* @undef to i32*)
+  %load = load i32, i32* bitcast ({}* @undef to i32*)
   ret i32 %load
 }
 
diff --git a/test/Transforms/InstSimplify/noalias-ptr.ll b/test/Transforms/InstSimplify/noalias-ptr.ll
index 7693e5542355..59b1dfa8de34 100644
--- a/test/Transforms/InstSimplify/noalias-ptr.ll
+++ b/test/Transforms/InstSimplify/noalias-ptr.ll
@@ -26,7 +26,7 @@ define void @_Z2p1v() #0 {
   %1 = bitcast [10 x i32]* %mStackData to i8*
   %2 = tail call noalias i8* @_Znam(i64 48) #4
   %3 = bitcast i8* %2 to i32*
-  %4 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %4 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %5 = icmp eq i32* %3, %4
   br i1 %5, label %7, label %6
 
@@ -47,7 +47,7 @@ define void @_Z2p1v() #0 {
 define void @_Z2p2bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g2
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -72,7 +72,7 @@ define void @_Z2p2bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
 define void @_Z2p4bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g3
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -97,7 +97,7 @@ define void @_Z2p4bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
 define void @_Z2p5bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g4
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -122,7 +122,7 @@ define void @_Z2p5bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
 define void @_Z2p6bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g5
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -151,7 +151,7 @@ define void @_Z2p6bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
 define void @_Z4nopebbPi(i1 zeroext %b1, i1 zeroext %b2, i32* readnone %q) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* %q
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -176,7 +176,7 @@ define void @_Z4nopebbPi(i1 zeroext %b1, i1 zeroext %b2, i32* readnone %q) #0 {
 define void @_Z2p3bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g1
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -201,7 +201,7 @@ define void @_Z2p3bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
 define void @_Z2p7bb(i1 zeroext %b1, i1 zeroext %b2) #0 {
   %mStackData = alloca [10 x i32], align 16
   %1 = bitcast [10 x i32]* %mStackData to i8*
-  %2 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %2 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %3 = select i1 %b1, i32* %2, i32* @g6
   %4 = tail call noalias i8* @_Znam(i64 48) #4
   %5 = tail call noalias i8* @_Znam(i64 48) #4
@@ -228,7 +228,7 @@ define void @_Z2p2v(i32 %c) #0 {
   %1 = bitcast [10 x i32]* %mStackData to i8*
   %2 = tail call noalias i8* @_Znam(i64 48) #4
   %3 = bitcast i8* %2 to i32*
-  %4 = getelementptr inbounds [10 x i32]* %mStackData, i64 0, i64 0
+  %4 = getelementptr inbounds [10 x i32], [10 x i32]* %mStackData, i64 0, i64 0
   %5 = icmp eq i32* %3, %4
   br i1 %5, label %7, label %6
 
diff --git a/test/Transforms/InstSimplify/past-the-end.ll b/test/Transforms/InstSimplify/past-the-end.ll
index 075da4a24be0..b676e9db2c77 100644
--- a/test/Transforms/InstSimplify/past-the-end.ll
+++ b/test/Transforms/InstSimplify/past-the-end.ll
@@ -18,8 +18,8 @@ define zeroext i1 @no_offsets() {
 ; Comparing past-the-end addresses of two distinct globals. Never equal.
 
 define zeroext i1 @both_past_the_end() {
-  %x = getelementptr i32* @opte_a, i32 1
-  %y = getelementptr i32* @opte_b, i32 1
+  %x = getelementptr i32, i32* @opte_a, i32 1
+  %y = getelementptr i32, i32* @opte_b, i32 1
   %t = icmp eq i32* %x, %y
   ret i1 %t
   ; CHECK: both_past_the_end(
@@ -31,11 +31,11 @@ define zeroext i1 @both_past_the_end() {
 ; of another. Can't fold this.
 
 define zeroext i1 @just_one_past_the_end() {
-  %x = getelementptr i32* @opte_a, i32 1
+  %x = getelementptr i32, i32* @opte_a, i32 1
   %t = icmp eq i32* %x, @opte_b
   ret i1 %t
   ; CHECK: just_one_past_the_end(
-  ; CHECK: ret i1 icmp eq (i32* getelementptr inbounds (i32* @opte_a, i32 1), i32* @opte_b)
+  ; CHECK: ret i1 icmp eq (i32* getelementptr inbounds (i32, i32* @opte_a, i32 1), i32* @opte_b)
 }
 
 ; Comparing base addresses of two distinct allocas. Never equal.
@@ -54,8 +54,8 @@ define zeroext i1 @no_alloca_offsets() {
 define zeroext i1 @both_past_the_end_alloca() {
   %m = alloca i32
   %n = alloca i32
-  %x = getelementptr i32* %m, i32 1
-  %y = getelementptr i32* %n, i32 1
+  %x = getelementptr i32, i32* %m, i32 1
+  %y = getelementptr i32, i32* %n, i32 1
   %t = icmp eq i32* %x, %y
   ret i1 %t
   ; CHECK: both_past_the_end_alloca(
@@ -69,7 +69,7 @@ define zeroext i1 @both_past_the_end_alloca() {
 define zeroext i1 @just_one_past_the_end_alloca() {
   %m = alloca i32
   %n = alloca i32
-  %x = getelementptr i32* %m, i32 1
+  %x = getelementptr i32, i32* %m, i32 1
   %t = icmp eq i32* %x, %n
   ret i1 %t
   ; CHECK: just_one_past_the_end_alloca(
diff --git a/test/Transforms/InstSimplify/ptr_diff.ll b/test/Transforms/InstSimplify/ptr_diff.ll
index 6a3f65237c5e..6c27e6f44678 100644
--- a/test/Transforms/InstSimplify/ptr_diff.ll
+++ b/test/Transforms/InstSimplify/ptr_diff.ll
@@ -6,8 +6,8 @@ define i64 @ptrdiff1(i8* %ptr) {
 ; CHECK-LABEL: @ptrdiff1(
 ; CHECK-NEXT: ret i64 42
 
-  %first = getelementptr inbounds i8* %ptr, i32 0
-  %last = getelementptr inbounds i8* %ptr, i32 42
+  %first = getelementptr inbounds i8, i8* %ptr, i32 0
+  %last = getelementptr inbounds i8, i8* %ptr, i32 42
   %first.int = ptrtoint i8* %first to i64
   %last.int = ptrtoint i8* %last to i64
   %diff = sub i64 %last.int, %first.int
@@ -18,14 +18,14 @@ define i64 @ptrdiff2(i8* %ptr) {
 ; CHECK-LABEL: @ptrdiff2(
 ; CHECK-NEXT: ret i64 42
 
-  %first1 = getelementptr inbounds i8* %ptr, i32 0
-  %first2 = getelementptr inbounds i8* %first1, i32 1
-  %first3 = getelementptr inbounds i8* %first2, i32 2
-  %first4 = getelementptr inbounds i8* %first3, i32 4
-  %last1 = getelementptr inbounds i8* %first2, i32 48
-  %last2 = getelementptr inbounds i8* %last1, i32 8
-  %last3 = getelementptr inbounds i8* %last2, i32 -4
-  %last4 = getelementptr inbounds i8* %last3, i32 -4
+  %first1 = getelementptr inbounds i8, i8* %ptr, i32 0
+  %first2 = getelementptr inbounds i8, i8* %first1, i32 1
+  %first3 = getelementptr inbounds i8, i8* %first2, i32 2
+  %first4 = getelementptr inbounds i8, i8* %first3, i32 4
+  %last1 = getelementptr inbounds i8, i8* %first2, i32 48
+  %last2 = getelementptr inbounds i8, i8* %last1, i32 8
+  %last3 = getelementptr inbounds i8, i8* %last2, i32 -4
+  %last4 = getelementptr inbounds i8, i8* %last3, i32 -4
   %first.int = ptrtoint i8* %first4 to i64
   %last.int = ptrtoint i8* %last4 to i64
   %diff = sub i64 %last.int, %first.int
@@ -39,8 +39,8 @@ define i64 @ptrdiff3(i8* %ptr) {
 ; CHECK: sub
 ; CHECK: ret
 
-  %first = getelementptr i8* %ptr, i32 0
-  %last = getelementptr i8* %ptr, i32 42
+  %first = getelementptr i8, i8* %ptr, i32 0
+  %last = getelementptr i8, i8* %ptr, i32 42
   %first.int = ptrtoint i8* %first to i64
   %last.int = ptrtoint i8* %last to i64
   %diff = sub i64 %last.int, %first.int
@@ -64,12 +64,12 @@ define <4 x i32> @ptrdiff4(<4 x i8*> %arg) nounwind {
 
 define i32 @ptrdiff5() nounwind {
 bb:
-  %tmp = getelementptr inbounds %struct.ham* @global, i32 0, i32 1
-  %tmp1 = getelementptr inbounds [2 x [2 x i32]]* %tmp, i32 0, i32 0
+  %tmp = getelementptr inbounds %struct.ham, %struct.ham* @global, i32 0, i32 1
+  %tmp1 = getelementptr inbounds [2 x [2 x i32]], [2 x [2 x i32]]* %tmp, i32 0, i32 0
   %tmp2 = bitcast [2 x i32]* %tmp1 to i32*
   %tmp3 = ptrtoint i32* %tmp2 to i32
-  %tmp4 = getelementptr inbounds %struct.ham* @global, i32 0, i32 1
-  %tmp5 = getelementptr inbounds [2 x [2 x i32]]* %tmp4, i32 0, i32 0
+  %tmp4 = getelementptr inbounds %struct.ham, %struct.ham* @global, i32 0, i32 1
+  %tmp5 = getelementptr inbounds [2 x [2 x i32]], [2 x [2 x i32]]* %tmp4, i32 0, i32 0
   %tmp6 = ptrtoint [2 x i32]* %tmp5 to i32
   %tmp7 = sub i32 %tmp3, %tmp6
   ret i32 %tmp7
diff --git a/test/Transforms/InstSimplify/undef.ll b/test/Transforms/InstSimplify/undef.ll
index e8b49b6397db..f1f0b037fdbd 100644
--- a/test/Transforms/InstSimplify/undef.ll
+++ b/test/Transforms/InstSimplify/undef.ll
@@ -150,7 +150,7 @@ define i64 @test17(i64 %a) {
 ; @test18
 ; CHECK: ret i64 undef
 define i64 @test18(i64 %a) {
-  %r = call i64 (i64)* undef(i64 %a)
+  %r = call i64 (i64) undef(i64 %a)
   ret i64 %r
 }
 
diff --git a/test/Transforms/InstSimplify/vector_gep.ll b/test/Transforms/InstSimplify/vector_gep.ll
index 17814637b3dd..5c3062047c30 100644
--- a/test/Transforms/InstSimplify/vector_gep.ll
+++ b/test/Transforms/InstSimplify/vector_gep.ll
@@ -4,13 +4,13 @@ target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 
 declare void @helper(<2 x i8*>)
 define void @test(<2 x i8*> %a) {
-  %A = getelementptr <2 x i8*> %a, <2 x i32> <i32 0, i32 0>
+  %A = getelementptr i8, <2 x i8*> %a, <2 x i32> <i32 0, i32 0>
   call void @helper(<2 x i8*> %A)
   ret void
 }
 
 define <4 x i8*> @test1(<4 x i8*> %a) {
-  %gep = getelementptr <4 x i8*> %a, <4 x i32> zeroinitializer
+  %gep = getelementptr i8, <4 x i8*> %a, <4 x i32> zeroinitializer
   ret <4 x i8*> %gep
 
 ; CHECK-LABEL: @test1
@@ -18,7 +18,7 @@ define <4 x i8*> @test1(<4 x i8*> %a) {
 }
 
 define <4 x i8*> @test2(<4 x i8*> %a) {
-  %gep = getelementptr <4 x i8*> %a
+  %gep = getelementptr i8, <4 x i8*> %a
   ret <4 x i8*> %gep
 
 ; CHECK-LABEL: @test2
@@ -28,7 +28,7 @@ define <4 x i8*> @test2(<4 x i8*> %a) {
 %struct = type { double, float }
 
 define <4 x float*> @test3() {
-  %gep = getelementptr <4 x %struct*> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
+  %gep = getelementptr %struct, <4 x %struct*> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
   ret <4 x float*> %gep
 
 ; CHECK-LABEL: @test3
@@ -38,7 +38,7 @@ define <4 x float*> @test3() {
 %struct.empty = type { }
 
 define <4 x %struct.empty*> @test4(<4 x %struct.empty*> %a) {
-  %gep = getelementptr <4 x %struct.empty*> %a, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
+  %gep = getelementptr %struct.empty, <4 x %struct.empty*> %a, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
   ret <4 x %struct.empty*> %gep
 
 ; CHECK-LABEL: @test4
@@ -47,9 +47,9 @@ define <4 x %struct.empty*> @test4(<4 x %struct.empty*> %a) {
 
 define <4 x i8*> @test5() {
   %c = inttoptr <4 x i64> <i64 1, i64 2, i64 3, i64 4> to <4 x i8*>
-  %gep = getelementptr <4 x i8*> %c, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
+  %gep = getelementptr i8, <4 x i8*> %c, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
   ret <4 x i8*> %gep
 
 ; CHECK-LABEL: @test5
-; CHECK-NEXT: ret <4 x i8*> getelementptr (<4 x i8*> <i8* inttoptr (i64 1 to i8*), i8* inttoptr (i64 2 to i8*), i8* inttoptr (i64 3 to i8*), i8* inttoptr (i64 4 to i8*)>, <4 x i32> <i32 1, i32 1, i32 1, i32 1>)
+; CHECK-NEXT: ret <4 x i8*> getelementptr (i8, <4 x i8*> <i8* inttoptr (i64 1 to i8*), i8* inttoptr (i64 2 to i8*), i8* inttoptr (i64 3 to i8*), i8* inttoptr (i64 4 to i8*)>, <4 x i32> <i32 1, i32 1, i32 1, i32 1>)
 }
diff --git a/test/Transforms/InstSimplify/vector_ptr_bitcast.ll b/test/Transforms/InstSimplify/vector_ptr_bitcast.ll
index 607892ae74d2..97c8343bfb9b 100644
--- a/test/Transforms/InstSimplify/vector_ptr_bitcast.ll
+++ b/test/Transforms/InstSimplify/vector_ptr_bitcast.ll
@@ -14,7 +14,7 @@ target datalayout = "e-i64:64-f80:128-n8:16:32:64-S128"
                                           align 8
 
 define i64 @fn() {
-  %x = load <2 x i8*>* bitcast (%mst* @a to <2 x i8*>*), align 8
+  %x = load <2 x i8*>, <2 x i8*>* bitcast (%mst* @a to <2 x i8*>*), align 8
   %b = extractelement <2 x i8*> %x, i32 0
   %c = ptrtoint i8* %b to i64
   ; CHECK-LABEL: @fn
@@ -23,7 +23,7 @@ define i64 @fn() {
 }
 
 define i64 @fn2() {
-  %x = load <4 x i32*>* bitcast (%mst2* @b to <4 x i32*>*), align 8
+  %x = load <4 x i32*>, <4 x i32*>* bitcast (%mst2* @b to <4 x i32*>*), align 8
   %b = extractelement <4 x i32*> %x, i32 0
   %c = extractelement <4 x i32*> %x, i32 3
   %d = ptrtoint i32* %b to i64
diff --git a/test/Transforms/Internalize/2009-01-05-InternalizeAliases.ll b/test/Transforms/Internalize/2009-01-05-InternalizeAliases.ll
index 16523886a068..c50b6fc61c8e 100644
--- a/test/Transforms/Internalize/2009-01-05-InternalizeAliases.ll
+++ b/test/Transforms/Internalize/2009-01-05-InternalizeAliases.ll
@@ -10,7 +10,7 @@
 ; CHECK: @C = internal alias i32* @A
 
 define i32 @main() {
-	%tmp = load i32* @C
+	%tmp = load i32, i32* @C
 	ret i32 %tmp
 }
 
diff --git a/test/Transforms/JumpThreading/2010-08-26-and.ll b/test/Transforms/JumpThreading/2010-08-26-and.ll
index 2d6caf752e09..cc56ac9cc84c 100644
--- a/test/Transforms/JumpThreading/2010-08-26-and.ll
+++ b/test/Transforms/JumpThreading/2010-08-26-and.ll
@@ -29,8 +29,8 @@ land.lhs.true.i:                                  ; preds = %_ZN12StringSwitchI5
   %retval.0.i.pre161 = phi i32 [ undef, %bb.nph ], [ %retval.0.i.pre, %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit134 ] ; <i32> [#uses=3]
   %indvar = phi i64 [ 0, %bb.nph ], [ %tmp146, %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit134 ] ; <i64> [#uses=1]
   %tmp146 = add i64 %indvar, 1                    ; <i64> [#uses=3]
-  %arrayidx = getelementptr i8** %argv, i64 %tmp146 ; <i8**> [#uses=1]
-  %tmp6 = load i8** %arrayidx, align 8            ; <i8*> [#uses=8]
+  %arrayidx = getelementptr i8*, i8** %argv, i64 %tmp146 ; <i8**> [#uses=1]
+  %tmp6 = load i8*, i8** %arrayidx, align 8            ; <i8*> [#uses=8]
   %call.i.i = call i64 @strlen(i8* %tmp6) nounwind ; <i64> [#uses=1]
   %conv.i.i = trunc i64 %call.i.i to i32          ; <i32> [#uses=6]\
 ; CHECK: switch i32 %conv.i.i
@@ -42,7 +42,7 @@ land.lhs.true.i:                                  ; preds = %_ZN12StringSwitchI5
   ]
 
 land.lhs.true5.i:                                 ; preds = %land.lhs.true.i
-  %call.i = call i32 @memcmp(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i8* %tmp6, i64 4) nounwind ; <i32> [#uses=1]
+  %call.i = call i32 @memcmp(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i8* %tmp6, i64 4) nounwind ; <i32> [#uses=1]
   %cmp9.i = icmp eq i32 %call.i, 0                ; <i1> [#uses=1]
   br i1 %cmp9.i, label %_ZN12StringSwitchI5ColorE4CaseILj4EEERS1_RAT__KcRKS0_.exit, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit
 
@@ -50,7 +50,7 @@ _ZN12StringSwitchI5ColorE4CaseILj4EEERS1_RAT__KcRKS0_.exit: ; preds = %land.lhs.
   br label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit
 
 land.lhs.true5.i37:                               ; preds = %land.lhs.true.i
-  %call.i35 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8]* @.str1, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
+  %call.i35 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str1, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
   %cmp9.i36 = icmp eq i32 %call.i35, 0            ; <i1> [#uses=1]
   br i1 %cmp9.i36, label %if.then.i40, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit
 
@@ -66,7 +66,7 @@ _ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit: ; preds = %if.then.i
   br i1 %or.cond, label %land.lhs.true5.i55, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit60
 
 land.lhs.true5.i55:                               ; preds = %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit
-  %call.i53 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8]* @.str2, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
+  %call.i53 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str2, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
   %cmp9.i54 = icmp eq i32 %call.i53, 0            ; <i1> [#uses=1]
   br i1 %cmp9.i54, label %if.then.i58, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit60
 
@@ -83,7 +83,7 @@ _ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit60: ; preds = %if.then
   br i1 %or.cond168, label %land.lhs.true5.i74, label %_ZN12StringSwitchI5ColorE4CaseILj6EEERS1_RAT__KcRKS0_.exit
 
 land.lhs.true5.i74:                               ; preds = %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit60
-  %call.i72 = call i32 @memcmp(i8* getelementptr inbounds ([6 x i8]* @.str3, i64 0, i64 0), i8* %tmp6, i64 6) nounwind ; <i32> [#uses=1]
+  %call.i72 = call i32 @memcmp(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str3, i64 0, i64 0), i8* %tmp6, i64 6) nounwind ; <i32> [#uses=1]
   %cmp9.i73 = icmp eq i32 %call.i72, 0            ; <i1> [#uses=1]
   br i1 %cmp9.i73, label %if.then.i77, label %_ZN12StringSwitchI5ColorE4CaseILj6EEERS1_RAT__KcRKS0_.exit
 
@@ -100,7 +100,7 @@ _ZN12StringSwitchI5ColorE4CaseILj6EEERS1_RAT__KcRKS0_.exit: ; preds = %if.then.i
   br i1 %or.cond169, label %land.lhs.true5.i92, label %_ZN12StringSwitchI5ColorE4CaseILj5EEERS1_RAT__KcRKS0_.exit
 
 land.lhs.true5.i92:                               ; preds = %_ZN12StringSwitchI5ColorE4CaseILj6EEERS1_RAT__KcRKS0_.exit
-  %call.i90 = call i32 @memcmp(i8* getelementptr inbounds ([5 x i8]* @.str4, i64 0, i64 0), i8* %tmp6, i64 5) nounwind ; <i32> [#uses=1]
+  %call.i90 = call i32 @memcmp(i8* getelementptr inbounds ([5 x i8], [5 x i8]* @.str4, i64 0, i64 0), i8* %tmp6, i64 5) nounwind ; <i32> [#uses=1]
   %cmp9.i91 = icmp eq i32 %call.i90, 0            ; <i1> [#uses=1]
   br i1 %cmp9.i91, label %if.then.i95, label %_ZN12StringSwitchI5ColorE4CaseILj5EEERS1_RAT__KcRKS0_.exit
 
@@ -117,7 +117,7 @@ _ZN12StringSwitchI5ColorE4CaseILj5EEERS1_RAT__KcRKS0_.exit: ; preds = %if.then.i
   br i1 %or.cond170, label %land.lhs.true5.i110, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit115
 
 land.lhs.true5.i110:                              ; preds = %_ZN12StringSwitchI5ColorE4CaseILj5EEERS1_RAT__KcRKS0_.exit
-  %call.i108 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8]* @.str5, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
+  %call.i108 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str5, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
   %cmp9.i109 = icmp eq i32 %call.i108, 0          ; <i1> [#uses=1]
   br i1 %cmp9.i109, label %if.then.i113, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit115
 
@@ -134,7 +134,7 @@ _ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit115: ; preds = %if.the
   br i1 %or.cond171, label %land.lhs.true5.i129, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit134
 
 land.lhs.true5.i129:                              ; preds = %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit115
-  %call.i127 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8]* @.str6, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
+  %call.i127 = call i32 @memcmp(i8* getelementptr inbounds ([7 x i8], [7 x i8]* @.str6, i64 0, i64 0), i8* %tmp6, i64 7) nounwind ; <i32> [#uses=1]
   %cmp9.i128 = icmp eq i32 %call.i127, 0          ; <i1> [#uses=1]
   br i1 %cmp9.i128, label %if.then.i132, label %_ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit134
 
@@ -147,7 +147,7 @@ _ZN12StringSwitchI5ColorE4CaseILj7EEERS1_RAT__KcRKS0_.exit134: ; preds = %if.the
   %tmp7.i138 = and i8 %tmp2.i137, 1               ; <i8> [#uses=1]
   %tobool.i139 = icmp eq i8 %tmp7.i138, 0         ; <i1> [#uses=1]
   %retval.0.i = select i1 %tobool.i139, i32 0, i32 %retval.0.i.pre ; <i32> [#uses=1]
-  %call22 = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([12 x i8]* @.str7, i64 0, i64 0), i32 %retval.0.i) ; <i32> [#uses=0]
+  %call22 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([12 x i8], [12 x i8]* @.str7, i64 0, i64 0), i32 %retval.0.i) ; <i32> [#uses=0]
   %exitcond = icmp eq i64 %tmp146, %tmp145        ; <i1> [#uses=1]
   br i1 %exitcond, label %for.end, label %land.lhs.true.i
 
diff --git a/test/Transforms/JumpThreading/2011-04-14-InfLoop.ll b/test/Transforms/JumpThreading/2011-04-14-InfLoop.ll
index 86a1321c3543..0323723a3a85 100644
--- a/test/Transforms/JumpThreading/2011-04-14-InfLoop.ll
+++ b/test/Transforms/JumpThreading/2011-04-14-InfLoop.ll
@@ -15,7 +15,7 @@ for.cond1177:
   br i1 %cmp1179, label %for.cond1177, label %land.rhs1320
 
 land.rhs1320:
-  %tmp1324 = load volatile i64* getelementptr inbounds (%0* @g_338, i64 0, i32 2), align 1
+  %tmp1324 = load volatile i64, i64* getelementptr inbounds (%0, %0* @g_338, i64 0, i32 2), align 1
   br label %if.end.i
 
 if.end.i:
diff --git a/test/Transforms/JumpThreading/assume.ll b/test/Transforms/JumpThreading/assume.ll
index 89dd0a954c01..53010b71c728 100644
--- a/test/Transforms/JumpThreading/assume.ll
+++ b/test/Transforms/JumpThreading/assume.ll
@@ -21,11 +21,11 @@ if.then:                                          ; preds = %entry
   br i1 %cmp2, label %if.then3, label %return
 
 if.then3:                                         ; preds = %if.then
-  tail call void (...)* @bar() #1
+  tail call void (...) @bar() #1
   br label %return
 
 if.else:                                          ; preds = %entry
-  tail call void (...)* @car() #1
+  tail call void (...) @car() #1
   br label %return
 
 return:                                           ; preds = %if.else, %if.then, %if.then3
@@ -43,12 +43,12 @@ entry:
 ; CHECK-LABEL: @test2
 ; CHECK: icmp sgt i32 %a, 5
 ; CHECK: tail call void @llvm.assume
-; CHECK: tail call void (...)* @bar()
+; CHECK: tail call void (...) @bar()
 ; CHECK: ret i32 1
 
 
 if.then:                                          ; preds = %entry
-  tail call void (...)* @bar() #1
+  tail call void (...) @bar() #1
   br label %return
 
 return:                                           ; preds = %entry, %if.then
diff --git a/test/Transforms/JumpThreading/crash.ll b/test/Transforms/JumpThreading/crash.ll
index 2fe87464c117..900a7738d3bf 100644
--- a/test/Transforms/JumpThreading/crash.ll
+++ b/test/Transforms/JumpThreading/crash.ll
@@ -356,7 +356,7 @@ B2:
   br label %BrBlock
 
 BrBlock:
-  %L = load i32* %P
+  %L = load i32, i32* %P
   %C = icmp eq i32 %L, 42
   br i1 %C, label %T, label %F
   
diff --git a/test/Transforms/JumpThreading/indirectbr.ll b/test/Transforms/JumpThreading/indirectbr.ll
index b87fb6c22287..197ca30f58ae 100644
--- a/test/Transforms/JumpThreading/indirectbr.ll
+++ b/test/Transforms/JumpThreading/indirectbr.ll
@@ -79,15 +79,15 @@ entry:
   br label %__here
 
 __here:                                           ; preds = %entry
-  %call = call i32 (...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here) to i64)) nounwind noredzone
+  %call = call i32 (...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here) to i64)) nounwind noredzone
   br label %__here1
 
 __here1:                                          ; preds = %__here
-  %call2 = call i32 (...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here1) to i64)) nounwind noredzone
+  %call2 = call i32 (...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here1) to i64)) nounwind noredzone
   br label %__here3
 
 __here3:                                          ; preds = %__here1
-  %call4 = call i32 (...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here3) to i64)) nounwind noredzone
+  %call4 = call i32 (...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i64 ptrtoint (i8* blockaddress(@test3, %__here3) to i64)) nounwind noredzone
   ret void
 }
 
diff --git a/test/Transforms/JumpThreading/landing-pad.ll b/test/Transforms/JumpThreading/landing-pad.ll
index 9ee0526d0446..4d49db04f6ab 100644
--- a/test/Transforms/JumpThreading/landing-pad.ll
+++ b/test/Transforms/JumpThreading/landing-pad.ll
@@ -13,8 +13,8 @@
 define void @_ZN15EditCommandImpl5applyEv(%class.E* %this) uwtable align 2 {
 entry:
   %0 = bitcast %class.E* %this to void (%class.E*)***
-  %vtable = load void (%class.E*)*** %0, align 8
-  %1 = load void (%class.E*)** %vtable, align 8
+  %vtable = load void (%class.E*)**, void (%class.E*)*** %0, align 8
+  %1 = load void (%class.E*)*, void (%class.E*)** %vtable, align 8
   call void %1(%class.E* %this)
   ret void
 }
@@ -22,16 +22,16 @@ entry:
 define void @_ZN1DC1Ev(%class.D* nocapture %this) unnamed_addr uwtable align 2 {
 entry:
   call void @_ZN24CompositeEditCommandImplC2Ev()
-  %0 = getelementptr inbounds %class.D* %this, i64 0, i32 0, i32 0, i32 0
-  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
+  %0 = getelementptr inbounds %class.D, %class.D* %this, i64 0, i32 0, i32 0, i32 0
+  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
   ret void
 }
 
 define void @_ZN1DC2Ev(%class.D* nocapture %this) unnamed_addr uwtable align 2 {
 entry:
   call void @_ZN24CompositeEditCommandImplC2Ev()
-  %0 = getelementptr inbounds %class.D* %this, i64 0, i32 0, i32 0, i32 0
-  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
+  %0 = getelementptr inbounds %class.D, %class.D* %this, i64 0, i32 0, i32 0, i32 0
+  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
   ret void
 }
 
@@ -50,10 +50,10 @@ entry:
 
 _ZN1DC1Ev.exit:                                   ; preds = %entry
   %0 = bitcast i8* %call to i32 (...)***
-  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
-  %_ref.i.i.i = getelementptr inbounds i8* %call, i64 8
+  store i32 (...)** bitcast (i8** getelementptr inbounds ([3 x i8*], [3 x i8*]* @_ZTV1D, i64 0, i64 2) to i32 (...)**), i32 (...)*** %0, align 8
+  %_ref.i.i.i = getelementptr inbounds i8, i8* %call, i64 8
   %1 = bitcast i8* %_ref.i.i.i to i32*
-  %2 = load i32* %1, align 4
+  %2 = load i32, i32* %1, align 4
   %inc.i.i.i = add nsw i32 %2, 1
   store i32 %inc.i.i.i, i32* %1, align 4
   %3 = bitcast i8* %call to %class.D*
@@ -76,7 +76,7 @@ lpad:                                             ; preds = %entry
 lpad1:                                            ; preds = %_ZN1DC1Ev.exit, %_ZN15EditCommandImpl5applyEv.exit
   %5 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
           cleanup
-  %6 = load i32* %1, align 4
+  %6 = load i32, i32* %1, align 4
   %tobool.i.i.i = icmp eq i32 %6, 0
   br i1 %tobool.i.i.i, label %_ZN1BI1DED1Ev.exit, label %if.then.i.i.i
 
@@ -98,10 +98,10 @@ terminate.lpad:                                   ; No predecessors!
 
 define void @_ZN1BI1DEC1EPS0_(%class.B* nocapture %this, %class.D* %p1) unnamed_addr uwtable align 2 {
 entry:
-  %m_ptr.i = getelementptr inbounds %class.B* %this, i64 0, i32 0
+  %m_ptr.i = getelementptr inbounds %class.B, %class.B* %this, i64 0, i32 0
   store %class.D* %p1, %class.D** %m_ptr.i, align 8
-  %_ref.i.i = getelementptr inbounds %class.D* %p1, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
-  %0 = load i32* %_ref.i.i, align 4
+  %_ref.i.i = getelementptr inbounds %class.D, %class.D* %p1, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
+  %0 = load i32, i32* %_ref.i.i, align 4
   %inc.i.i = add nsw i32 %0, 1
   store i32 %inc.i.i, i32* %_ref.i.i, align 4
   ret void
@@ -115,8 +115,8 @@ declare void @_ZdlPv()
 
 define %class.D* @_ZN1BI1DEptEv(%class.B* nocapture readonly %this) nounwind readonly uwtable align 2 {
 entry:
-  %m_ptr = getelementptr inbounds %class.B* %this, i64 0, i32 0
-  %0 = load %class.D** %m_ptr, align 8
+  %m_ptr = getelementptr inbounds %class.B, %class.B* %this, i64 0, i32 0
+  %0 = load %class.D*, %class.D** %m_ptr, align 8
   ret %class.D* %0
 }
 
@@ -124,10 +124,10 @@ declare void @_ZN1D16deleteKeyPressedEv()
 
 define void @_ZN1BI1DED1Ev(%class.B* nocapture readonly %this) unnamed_addr uwtable align 2 {
 entry:
-  %m_ptr.i = getelementptr inbounds %class.B* %this, i64 0, i32 0
-  %0 = load %class.D** %m_ptr.i, align 8
-  %_ref.i.i = getelementptr inbounds %class.D* %0, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
-  %1 = load i32* %_ref.i.i, align 4
+  %m_ptr.i = getelementptr inbounds %class.B, %class.B* %this, i64 0, i32 0
+  %0 = load %class.D*, %class.D** %m_ptr.i, align 8
+  %_ref.i.i = getelementptr inbounds %class.D, %class.D* %0, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
+  %1 = load i32, i32* %_ref.i.i, align 4
   %tobool.i.i = icmp eq i32 %1, 0
   br i1 %tobool.i.i, label %_ZN1BI1DED2Ev.exit, label %if.then.i.i
 
@@ -146,10 +146,10 @@ declare hidden void @__clang_call_terminate()
 
 define void @_ZN1BI1DED2Ev(%class.B* nocapture readonly %this) unnamed_addr uwtable align 2 {
 entry:
-  %m_ptr = getelementptr inbounds %class.B* %this, i64 0, i32 0
-  %0 = load %class.D** %m_ptr, align 8
-  %_ref.i = getelementptr inbounds %class.D* %0, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
-  %1 = load i32* %_ref.i, align 4
+  %m_ptr = getelementptr inbounds %class.B, %class.B* %this, i64 0, i32 0
+  %0 = load %class.D*, %class.D** %m_ptr, align 8
+  %_ref.i = getelementptr inbounds %class.D, %class.D* %0, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
+  %1 = load i32, i32* %_ref.i, align 4
   %tobool.i = icmp eq i32 %1, 0
   br i1 %tobool.i, label %_ZN1AI1CE5derefEv.exit, label %if.then.i
 
@@ -166,8 +166,8 @@ _ZN1AI1CE5derefEv.exit:                           ; preds = %entry, %if.then.i
 
 define void @_ZN1AI1CE5derefEv(%class.A* nocapture readonly %this) nounwind uwtable align 2 {
 entry:
-  %_ref = getelementptr inbounds %class.A* %this, i64 0, i32 0
-  %0 = load i32* %_ref, align 4
+  %_ref = getelementptr inbounds %class.A, %class.A* %this, i64 0, i32 0
+  %0 = load i32, i32* %_ref, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %if.end, label %if.then
 
@@ -184,10 +184,10 @@ if.end:                                           ; preds = %entry, %if.then
 
 define void @_ZN1BI1DEC2EPS0_(%class.B* nocapture %this, %class.D* %p1) unnamed_addr uwtable align 2 {
 entry:
-  %m_ptr = getelementptr inbounds %class.B* %this, i64 0, i32 0
+  %m_ptr = getelementptr inbounds %class.B, %class.B* %this, i64 0, i32 0
   store %class.D* %p1, %class.D** %m_ptr, align 8
-  %_ref.i = getelementptr inbounds %class.D* %p1, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
-  %0 = load i32* %_ref.i, align 4
+  %_ref.i = getelementptr inbounds %class.D, %class.D* %p1, i64 0, i32 0, i32 0, i32 1, i32 0, i32 0
+  %0 = load i32, i32* %_ref.i, align 4
   %inc.i = add nsw i32 %0, 1
   store i32 %inc.i, i32* %_ref.i, align 4
   ret void
@@ -195,8 +195,8 @@ entry:
 
 define void @_ZN1AI1CE3refEv(%class.A* nocapture %this) nounwind uwtable align 2 {
 entry:
-  %_ref = getelementptr inbounds %class.A* %this, i64 0, i32 0
-  %0 = load i32* %_ref, align 4
+  %_ref = getelementptr inbounds %class.A, %class.A* %this, i64 0, i32 0
+  %0 = load i32, i32* %_ref, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %_ref, align 4
   ret void
diff --git a/test/Transforms/JumpThreading/lvi-load.ll b/test/Transforms/JumpThreading/lvi-load.ll
index 2a4cf925102f..a36e263bf92c 100644
--- a/test/Transforms/JumpThreading/lvi-load.ll
+++ b/test/Transforms/JumpThreading/lvi-load.ll
@@ -16,8 +16,8 @@ target triple = "x86_64-apple-darwin10.4"
 ; CHECK: Z3fooPN4llvm5ValueE
 define zeroext i8 @_Z3fooPN4llvm5ValueE(%"struct.llvm::Value"* %V) ssp {
 entry:
-  %0 = getelementptr inbounds %"struct.llvm::Value"* %V, i64 0, i32 1 ; <i8*> [#uses=1]
-  %1 = load i8* %0, align 8                       ; <i8> [#uses=2]
+  %0 = getelementptr inbounds %"struct.llvm::Value", %"struct.llvm::Value"* %V, i64 0, i32 1 ; <i8*> [#uses=1]
+  %1 = load i8, i8* %0, align 8                       ; <i8> [#uses=2]
   %2 = icmp ugt i8 %1, 20                         ; <i1> [#uses=1]
   br i1 %2, label %bb.i, label %bb2
 
@@ -27,7 +27,7 @@ bb.i:                                             ; preds = %entry
 
 ; CHECK-NOT: assert
 bb6.i.i:                                          ; preds = %bb.i
-  tail call void @__assert_rtn(i8* getelementptr inbounds ([5 x i8]* @_ZZN4llvm4castINS_11InstructionEPNS_5ValueEEENS_10cast_rettyIT_T0_E8ret_typeERKS6_E8__func__, i64 0, i64 0), i8* getelementptr inbounds ([31 x i8]* @.str, i64 0, i64 0), i32 202, i8* getelementptr inbounds ([59 x i8]* @.str1, i64 0, i64 0)) noreturn
+  tail call void @__assert_rtn(i8* getelementptr inbounds ([5 x i8], [5 x i8]* @_ZZN4llvm4castINS_11InstructionEPNS_5ValueEEENS_10cast_rettyIT_T0_E8ret_typeERKS6_E8__func__, i64 0, i64 0), i8* getelementptr inbounds ([31 x i8], [31 x i8]* @.str, i64 0, i64 0), i32 202, i8* getelementptr inbounds ([59 x i8], [59 x i8]* @.str1, i64 0, i64 0)) noreturn
   unreachable
 
 _ZN4llvm8dyn_castINS_11InstructionEPNS_5ValueEEENS_10cast_rettyIT_T0_E8ret_typeERKS6_.exit: ; preds = %bb.i
diff --git a/test/Transforms/JumpThreading/or-undef.ll b/test/Transforms/JumpThreading/or-undef.ll
index 6311b6df4373..b55bdddc196d 100644
--- a/test/Transforms/JumpThreading/or-undef.ll
+++ b/test/Transforms/JumpThreading/or-undef.ll
@@ -32,13 +32,13 @@ bb2:                                              ; preds = %bb1, %bb, %entry
   br i1 %tmp7, label %bb7, label %bb5
 
 bb5:                                              ; preds = %bb2
-  %tmp8 = load i8** %argv.0, align 8              ; <i8*> [#uses=1]
+  %tmp8 = load i8*, i8** %argv.0, align 8              ; <i8*> [#uses=1]
   %tmp9 = icmp eq i8* %tmp8, null                 ; <i1> [#uses=1]
   br i1 %tmp9, label %bb7, label %bb6
 
 bb6:                                              ; preds = %bb5
-  %tmp10 = load i8** %argv.0, align 8             ; <i8*> [#uses=1]
-  %tmp11 = load i8* %tmp10, align 1               ; <i8> [#uses=1]
+  %tmp10 = load i8*, i8** %argv.0, align 8             ; <i8*> [#uses=1]
+  %tmp11 = load i8, i8* %tmp10, align 1               ; <i8> [#uses=1]
   %tmp12 = icmp eq i8 %tmp11, 0                   ; <i1> [#uses=1]
   br i1 %tmp12, label %bb7, label %bb8
 
@@ -47,7 +47,7 @@ bb7:                                              ; preds = %bb6, %bb5, %bb2
   br label %bb9
 
 bb8:                                              ; preds = %bb6
-  %tmp13 = load i8** %argv.0, align 8             ; <i8*> [#uses=1]
+  %tmp13 = load i8*, i8** %argv.0, align 8             ; <i8*> [#uses=1]
   %tmp14 = call i64 @f5(i8* %tmp13) nounwind      ; <i64> [#uses=0]
   br label %bb9
 
diff --git a/test/Transforms/JumpThreading/phi-eq.ll b/test/Transforms/JumpThreading/phi-eq.ll
index e05d5ee7c974..75e8cc81fed8 100644
--- a/test/Transforms/JumpThreading/phi-eq.ll
+++ b/test/Transforms/JumpThreading/phi-eq.ll
@@ -39,7 +39,7 @@ if.then:                                          ; preds = %do.body
   br label %if.end
 
 if.else:                                          ; preds = %do.body
-  call void @g_assertion_message_expr(i8* null, i8* getelementptr inbounds ([10 x i8]* @.str2, i32 0, i32 0), i32 581, i8* getelementptr inbounds ([62 x i8]* @__PRETTY_FUNCTION__.copy_filter_list, i32 0, i32 0), i8* getelementptr inbounds ([22 x i8]* @.str12, i32 0, i32 0)) noreturn
+  call void @g_assertion_message_expr(i8* null, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str2, i32 0, i32 0), i32 581, i8* getelementptr inbounds ([62 x i8], [62 x i8]* @__PRETTY_FUNCTION__.copy_filter_list, i32 0, i32 0), i8* getelementptr inbounds ([22 x i8], [22 x i8]* @.str12, i32 0, i32 0)) noreturn
   unreachable
 
 if.end:                                           ; preds = %if.then
@@ -66,7 +66,7 @@ sw.bb3.i:                                         ; preds = %do.end
   br label %get_filter_list.exit
 
 sw.default.i:                                     ; preds = %do.end
-  call void @g_assertion_message(i8* null, i8* getelementptr inbounds ([10 x i8]* @.str2, i32 0, i32 0), i32 408, i8* getelementptr inbounds ([44 x i8]* @__PRETTY_FUNCTION__.get_filter_list, i32 0, i32 0), i8* null) noreturn nounwind
+  call void @g_assertion_message(i8* null, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str2, i32 0, i32 0), i32 408, i8* getelementptr inbounds ([44 x i8], [44 x i8]* @__PRETTY_FUNCTION__.get_filter_list, i32 0, i32 0), i8* null) noreturn nounwind
   unreachable
 
 get_filter_list.exit:                             ; preds = %sw.bb3.i, %sw.bb2.i, %sw.bb1.i, %sw.bb.i
@@ -91,40 +91,40 @@ sw.bb3.i4:                                        ; preds = %get_filter_list.exi
   br label %get_filter_list.exit6
 
 sw.default.i5:                                    ; preds = %get_filter_list.exit
-  call void @g_assertion_message(i8* null, i8* getelementptr inbounds ([10 x i8]* @.str2, i32 0, i32 0), i32 408, i8* getelementptr inbounds ([44 x i8]* @__PRETTY_FUNCTION__.get_filter_list, i32 0, i32 0), i8* null) noreturn nounwind
+  call void @g_assertion_message(i8* null, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str2, i32 0, i32 0), i32 408, i8* getelementptr inbounds ([44 x i8], [44 x i8]* @__PRETTY_FUNCTION__.get_filter_list, i32 0, i32 0), i8* null) noreturn nounwind
   unreachable
 
 ; CHECK: get_filter_list.exit
 get_filter_list.exit6:                            ; preds = %sw.bb3.i4, %sw.bb2.i3, %sw.bb1.i2, %sw.bb.i1
   %1 = phi %struct._GList** [ @display_edited_filters, %sw.bb3.i4 ], [ @capture_edited_filters, %sw.bb2.i3 ], [ @display_filters, %sw.bb1.i2 ], [ @capture_filters, %sw.bb.i1 ]
 ; CHECK: %2 = load
-  %2 = load %struct._GList** %1, align 8
+  %2 = load %struct._GList*, %struct._GList** %1, align 8
 ; We should have jump-threading insert an additional load here for the value
 ; coming out of the first switch, which is picked up by a subsequent phi
-; CHECK: {{%\.pr = load %[^%]* %0}}
+; CHECK: %.pr = load %struct._GList*, %struct._GList** %0
 ; CHECK-NEXT:  br label %while.cond
   br label %while.cond
 
 ; CHECK: while.cond
 while.cond:                                       ; preds = %while.body, %get_filter_list.exit6
 ; CHECK: {{= phi .*%.pr}}
-  %3 = load %struct._GList** %0, align 8
+  %3 = load %struct._GList*, %struct._GList** %0, align 8
 ; CHECK: tobool
   %tobool = icmp ne %struct._GList* %3, null
   br i1 %tobool, label %while.body, label %while.end
 
 while.body:                                       ; preds = %while.cond
-  %4 = load %struct._GList** %0, align 8
-  %5 = load %struct._GList** %0, align 8
+  %4 = load %struct._GList*, %struct._GList** %0, align 8
+  %5 = load %struct._GList*, %struct._GList** %0, align 8
   %call2 = call %struct._GList* @g_list_first(%struct._GList* %5)
-  %data.i = getelementptr inbounds %struct._GList* %call2, i32 0, i32 0
-  %6 = load i8** %data.i, align 8
+  %data.i = getelementptr inbounds %struct._GList, %struct._GList* %call2, i32 0, i32 0
+  %6 = load i8*, i8** %data.i, align 8
   %7 = bitcast i8* %6 to %struct.filter_def*
-  %name.i = getelementptr inbounds %struct.filter_def* %7, i32 0, i32 0
-  %8 = load i8** %name.i, align 8
+  %name.i = getelementptr inbounds %struct.filter_def, %struct.filter_def* %7, i32 0, i32 0
+  %8 = load i8*, i8** %name.i, align 8
   call void @g_free(i8* %8) nounwind
-  %strval.i = getelementptr inbounds %struct.filter_def* %7, i32 0, i32 1
-  %9 = load i8** %strval.i, align 8
+  %strval.i = getelementptr inbounds %struct.filter_def, %struct.filter_def* %7, i32 0, i32 1
+  %9 = load i8*, i8** %strval.i, align 8
   call void @g_free(i8* %9) nounwind
   %10 = bitcast %struct.filter_def* %7 to i8*
   call void @g_free(i8* %10) nounwind
@@ -136,7 +136,7 @@ while.end:                                        ; preds = %while.cond
   br label %do.body4
 
 do.body4:                                         ; preds = %while.end
-  %11 = load %struct._GList** %0, align 8
+  %11 = load %struct._GList*, %struct._GList** %0, align 8
   %call5 = call i32 @g_list_length(%struct._GList* %11)
   %cmp6 = icmp eq i32 %call5, 0
   br i1 %cmp6, label %if.then7, label %if.else8
@@ -145,7 +145,7 @@ if.then7:                                         ; preds = %do.body4
   br label %if.end9
 
 if.else8:                                         ; preds = %do.body4
-  call void @g_assertion_message_expr(i8* null, i8* getelementptr inbounds ([10 x i8]* @.str2, i32 0, i32 0), i32 600, i8* getelementptr inbounds ([62 x i8]* @__PRETTY_FUNCTION__.copy_filter_list, i32 0, i32 0), i8* getelementptr inbounds ([31 x i8]* @.str13, i32 0, i32 0)) noreturn
+  call void @g_assertion_message_expr(i8* null, i8* getelementptr inbounds ([10 x i8], [10 x i8]* @.str2, i32 0, i32 0), i32 600, i8* getelementptr inbounds ([62 x i8], [62 x i8]* @__PRETTY_FUNCTION__.copy_filter_list, i32 0, i32 0), i8* getelementptr inbounds ([31 x i8], [31 x i8]* @.str13, i32 0, i32 0)) noreturn
   unreachable
 
 if.end9:                                          ; preds = %if.then7
@@ -160,21 +160,21 @@ while.cond11:                                     ; preds = %cond.end, %do.end10
   br i1 %tobool12, label %while.body13, label %while.end16
 
 while.body13:                                     ; preds = %while.cond11
-  %data = getelementptr inbounds %struct._GList* %cond10, i32 0, i32 0
-  %12 = load i8** %data, align 8
+  %data = getelementptr inbounds %struct._GList, %struct._GList* %cond10, i32 0, i32 0
+  %12 = load i8*, i8** %data, align 8
   %13 = bitcast i8* %12 to %struct.filter_def*
-  %14 = load %struct._GList** %0, align 8
-  %name = getelementptr inbounds %struct.filter_def* %13, i32 0, i32 0
-  %15 = load i8** %name, align 8
-  %strval = getelementptr inbounds %struct.filter_def* %13, i32 0, i32 1
-  %16 = load i8** %strval, align 8
+  %14 = load %struct._GList*, %struct._GList** %0, align 8
+  %name = getelementptr inbounds %struct.filter_def, %struct.filter_def* %13, i32 0, i32 0
+  %15 = load i8*, i8** %name, align 8
+  %strval = getelementptr inbounds %struct.filter_def, %struct.filter_def* %13, i32 0, i32 1
+  %16 = load i8*, i8** %strval, align 8
   %call.i7 = call noalias i8* @g_malloc(i64 16) nounwind
   %17 = bitcast i8* %call.i7 to %struct.filter_def*
   %call1.i = call noalias i8* @g_strdup(i8* %15) nounwind
-  %name.i8 = getelementptr inbounds %struct.filter_def* %17, i32 0, i32 0
+  %name.i8 = getelementptr inbounds %struct.filter_def, %struct.filter_def* %17, i32 0, i32 0
   store i8* %call1.i, i8** %name.i8, align 8
   %call2.i = call noalias i8* @g_strdup(i8* %16) nounwind
-  %strval.i9 = getelementptr inbounds %struct.filter_def* %17, i32 0, i32 1
+  %strval.i9 = getelementptr inbounds %struct.filter_def, %struct.filter_def* %17, i32 0, i32 1
   store i8* %call2.i, i8** %strval.i9, align 8
   %18 = bitcast %struct.filter_def* %17 to i8*
   %call3.i = call %struct._GList* @g_list_append(%struct._GList* %14, i8* %18) nounwind
@@ -183,8 +183,8 @@ while.body13:                                     ; preds = %while.cond11
   br i1 %tobool15, label %cond.true, label %cond.false
 
 cond.true:                                        ; preds = %while.body13
-  %next = getelementptr inbounds %struct._GList* %cond10, i32 0, i32 1
-  %19 = load %struct._GList** %next, align 8
+  %next = getelementptr inbounds %struct._GList, %struct._GList* %cond10, i32 0, i32 1
+  %19 = load %struct._GList*, %struct._GList** %next, align 8
   br label %cond.end
 
 cond.false:                                       ; preds = %while.body13
diff --git a/test/Transforms/JumpThreading/select.ll b/test/Transforms/JumpThreading/select.ll
index 545e86c082f4..d0df7725f722 100644
--- a/test/Transforms/JumpThreading/select.ll
+++ b/test/Transforms/JumpThreading/select.ll
@@ -135,7 +135,7 @@ L4:
 ; CHECK: icmp
 define void @test_switch_default(i32* nocapture %status) nounwind {
 entry:
-  %0 = load i32* %status, align 4
+  %0 = load i32, i32* %status, align 4
   switch i32 %0, label %L2 [
     i32 5061, label %L1
     i32 0, label %L2
@@ -146,7 +146,7 @@ L1:
   br label %L2
 
 L2:
-  %1 = load i32* %status, align 4
+  %1 = load i32, i32* %status, align 4
   %cmp57.i = icmp eq i32 %1, 0
   br i1 %cmp57.i, label %L3, label %L4
 
diff --git a/test/Transforms/JumpThreading/thread-cmp.ll b/test/Transforms/JumpThreading/thread-cmp.ll
new file mode 100644
index 000000000000..950896744762
--- /dev/null
+++ b/test/Transforms/JumpThreading/thread-cmp.ll
@@ -0,0 +1,69 @@
+; RUN: opt -S -jump-threading %s | FileCheck %s
+; When simplify a branch based on LVI predicates, we should replace the 
+; comparison itself with a constant (when possible) in case it's otherwise used.
+
+define i32 @test(i32* %p) {
+; CHECK-LABEL: @test
+; CHECK: icmp eq
+; CHECK-NEXT: br i1 %cmp, label %exit2, label %exit1
+; CHECK-NOT: icmp ne
+entry:
+  %cmp = icmp eq i32* %p, null
+  br i1 %cmp, label %is_null, label %not_null
+is_null:
+  %cmp2 = icmp ne i32* %p, null
+  br i1 %cmp2, label %exit1, label %exit2
+not_null:
+  %cmp3 = icmp ne i32* %p, null
+  br i1 %cmp3, label %exit1, label %exit2
+exit1:
+  ret i32 0
+exit2:
+  ret i32 1
+}
+
+declare void @use(i1)
+
+; It would not be legal to replace %cmp2 (well, in this case it actually is, 
+; but that's a CSE problem, not a LVI/jump threading problem)
+define i32 @test_negative(i32* %p) {
+; CHECK-LABEL: @test
+; CHECK: icmp ne
+; CHECK: icmp eq
+; CHECK-NEXT: br i1 %cmp, label %exit2, label %exit1
+; CHECK-NOT: icmp ne
+entry:
+  %cmp2 = icmp ne i32* %p, null
+  call void @use(i1 %cmp2)
+  %cmp = icmp eq i32* %p, null
+  br i1 %cmp, label %is_null, label %not_null
+is_null:
+  br i1 %cmp2, label %exit1, label %exit2
+not_null:
+  br i1 %cmp2, label %exit1, label %exit2
+exit1:
+  ret i32 0
+exit2:
+  ret i32 1
+}
+
+; In this case, we can remove cmp2 because it's otherwise unused
+define i32 @test2(i32* %p) {
+; CHECK-LABEL: @test
+; CHECK-LABEL: entry:
+; CHECK-NEXT: icmp eq
+; CHECK-NEXT: br i1 %cmp, label %exit2, label %exit1
+; CHECK-NOT: icmp ne
+entry:
+  %cmp2 = icmp ne i32* %p, null
+  %cmp = icmp eq i32* %p, null
+  br i1 %cmp, label %is_null, label %not_null
+is_null:
+  br i1 %cmp2, label %exit1, label %exit2
+not_null:
+  br i1 %cmp2, label %exit1, label %exit2
+exit1:
+  ret i32 0
+exit2:
+  ret i32 1
+}
diff --git a/test/Transforms/JumpThreading/thread-loads.ll b/test/Transforms/JumpThreading/thread-loads.ll
index 4351f9918109..f76c1ec51857 100644
--- a/test/Transforms/JumpThreading/thread-loads.ll
+++ b/test/Transforms/JumpThreading/thread-loads.ll
@@ -8,7 +8,7 @@ target triple = "i386-apple-darwin7"
 define i32 @test1(i32* %P) nounwind {
 ; CHECK-LABEL: @test1(
 entry:
-	%0 = tail call i32 (...)* @f1() nounwind		; <i32> [#uses=1]
+	%0 = tail call i32 (...) @f1() nounwind		; <i32> [#uses=1]
 	%1 = icmp eq i32 %0, 0		; <i1> [#uses=1]
 	br i1 %1, label %bb1, label %bb
 
@@ -21,12 +21,12 @@ bb:		; preds = %entry
 
 bb1:		; preds = %entry, %bb
 	%res.0 = phi i32 [ 1, %bb ], [ 0, %entry ]		; <i32> [#uses=2]
-	%2 = load i32* %P, align 4		; <i32> [#uses=1]
+	%2 = load i32, i32* %P, align 4		; <i32> [#uses=1]
 	%3 = icmp sgt i32 %2, 36		; <i1> [#uses=1]
 	br i1 %3, label %bb3, label %bb2
 
 bb2:		; preds = %bb1
-	%4 = tail call i32 (...)* @f2() nounwind		; <i32> [#uses=0]
+	%4 = tail call i32 (...) @f2() nounwind		; <i32> [#uses=0]
 	ret i32 %res.0
 
 bb3:		; preds = %bb1
@@ -47,7 +47,7 @@ declare i32 @f2(...)
 define i32 @test2(i32* %P) nounwind {
 ; CHECK-LABEL: @test2(
 entry:
-	%0 = tail call i32 (...)* @f1() nounwind		; <i32> [#uses=1]
+	%0 = tail call i32 (...) @f1() nounwind		; <i32> [#uses=1]
 	%1 = icmp eq i32 %0, 0		; <i1> [#uses=1]
 	br i1 %1, label %bb1, label %bb
 
@@ -60,12 +60,12 @@ bb:		; preds = %entry
 
 bb1:		; preds = %entry, %bb
 	%res.0 = phi i32 [ 1, %bb ], [ 0, %entry ]
-	%2 = load i32* %P, align 4, !tbaa !0
+	%2 = load i32, i32* %P, align 4, !tbaa !0
 	%3 = icmp sgt i32 %2, 36
 	br i1 %3, label %bb3, label %bb2
 
 bb2:		; preds = %bb1
-	%4 = tail call i32 (...)* @f2() nounwind
+	%4 = tail call i32 (...) @f2() nounwind
 	ret i32 %res.0
 
 bb3:		; preds = %bb1
@@ -83,16 +83,16 @@ define i32 @test3(i8** %x, i1 %f) {
 ; CHECK-LABEL: @test3(
 entry:
   %0 = bitcast i8** %x to i32**
-  %1 = load i32** %0, align 8
+  %1 = load i32*, i32** %0, align 8
   br i1 %f, label %if.end57, label %if.then56
-; CHECK: %[[LOAD:.*]] = load i32**
+; CHECK: %[[LOAD:.*]] = load i32*, i32**
 ; CHECK: %[[CAST:.*]] = bitcast i32* %[[LOAD]] to i8*
 
 if.then56:
   br label %if.end57
 
 if.end57:
-  %2 = load i8** %x, align 8
+  %2 = load i8*, i8** %x, align 8
   %tobool59 = icmp eq i8* %2, null
   br i1 %tobool59, label %return, label %if.then60
 ; CHECK: %[[PHI:.*]] = phi i8* [ %[[CAST]], %[[PRED:[^ ]+]] ], [ %[[CAST]], %[[PRED]] ]
diff --git a/test/Transforms/LCSSA/2006-06-03-IncorrectIDFPhis.ll b/test/Transforms/LCSSA/2006-06-03-IncorrectIDFPhis.ll
index dd43c8837901..a6abfa5f3c4d 100644
--- a/test/Transforms/LCSSA/2006-06-03-IncorrectIDFPhis.ll
+++ b/test/Transforms/LCSSA/2006-06-03-IncorrectIDFPhis.ll
@@ -10,10 +10,10 @@ no_exit:                ; preds = %endif, %entry
         %SJE.0.0 = phi %struct.SetJmpMapEntry* [ %tmp.24, %endif ], [ null, %entry ]            ; <%struct.SetJmpMapEntry*> [#uses=1]
         br i1 false, label %then, label %endif
 then:           ; preds = %no_exit
-        %tmp.20 = getelementptr %struct.SetJmpMapEntry* %SJE.0.0, i32 0, i32 1          ; <i32*> [#uses=0]
+        %tmp.20 = getelementptr %struct.SetJmpMapEntry, %struct.SetJmpMapEntry* %SJE.0.0, i32 0, i32 1          ; <i32*> [#uses=0]
         ret void
 endif:          ; preds = %no_exit
-        %tmp.24 = load %struct.SetJmpMapEntry** null            ; <%struct.SetJmpMapEntry*> [#uses=1]
+        %tmp.24 = load %struct.SetJmpMapEntry*, %struct.SetJmpMapEntry** null            ; <%struct.SetJmpMapEntry*> [#uses=1]
         br i1 false, label %UnifiedReturnBlock, label %no_exit
 UnifiedReturnBlock:             ; preds = %endif, %entry
         ret void
diff --git a/test/Transforms/LCSSA/2006-07-09-NoDominator.ll b/test/Transforms/LCSSA/2006-07-09-NoDominator.ll
index b03f09ac20e5..bc3d150fbdfe 100644
--- a/test/Transforms/LCSSA/2006-07-09-NoDominator.ll
+++ b/test/Transforms/LCSSA/2006-07-09-NoDominator.ll
@@ -11,10 +11,10 @@ loopentry:		; preds = %endif, %entry
 no_exit:		; preds = %loopentry
 	br i1 false, label %then, label %endif
 then:		; preds = %no_exit
-	%tmp.21 = getelementptr %struct.SetJmpMapEntry* %SJE.0, i32 0, i32 1		; <i32*> [#uses=0]
+	%tmp.21 = getelementptr %struct.SetJmpMapEntry, %struct.SetJmpMapEntry* %SJE.0, i32 0, i32 1		; <i32*> [#uses=0]
 	br label %return
 endif:		; preds = %no_exit
-	%tmp.25 = load %struct.SetJmpMapEntry** null		; <%struct.SetJmpMapEntry*> [#uses=1]
+	%tmp.25 = load %struct.SetJmpMapEntry*, %struct.SetJmpMapEntry** null		; <%struct.SetJmpMapEntry*> [#uses=1]
 	br label %loopentry
 loopexit:		; preds = %loopentry
 	br label %return
diff --git a/test/Transforms/LCSSA/2007-07-12-LICM-2.ll b/test/Transforms/LCSSA/2007-07-12-LICM-2.ll
index 2c5815cef33a..f5d3f7e2b1c8 100644
--- a/test/Transforms/LCSSA/2007-07-12-LICM-2.ll
+++ b/test/Transforms/LCSSA/2007-07-12-LICM-2.ll
@@ -4,7 +4,7 @@ entry:
 	br label %bb7
 
 bb7:		; preds = %bb7, %entry
-	%tmp39 = load <4 x float>* null		; <<4 x float>> [#uses=1]
+	%tmp39 = load <4 x float>, <4 x float>* null		; <<4 x float>> [#uses=1]
 	%tmp40 = fadd <4 x float> %tmp39, < float 2.000000e+00, float 3.000000e+00, float 1.000000e+00, float 0.000000e+00 >		; <<4 x float>> [#uses=1]
 	%tmp43 = fadd <4 x float> %tmp40, < float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 2.000000e+00 >		; <<4 x float>> [#uses=1]
 	%tmp46 = fadd <4 x float> %tmp43, < float 3.000000e+00, float 0.000000e+00, float 2.000000e+00, float 4.000000e+00 >		; <<4 x float>> [#uses=1]
diff --git a/test/Transforms/LCSSA/2007-07-12-LICM-3.ll b/test/Transforms/LCSSA/2007-07-12-LICM-3.ll
index 7e0d3c63b13c..689b6eaff8e0 100644
--- a/test/Transforms/LCSSA/2007-07-12-LICM-3.ll
+++ b/test/Transforms/LCSSA/2007-07-12-LICM-3.ll
@@ -8,7 +8,7 @@ bb:             ; preds = %bb56, %entry
         br label %bb7
 
 bb7:            ; preds = %bb7, %bb
-        %tmp39 = load <4 x float>* null         ; <<4 x float>> [#uses=1]
+        %tmp39 = load <4 x float>, <4 x float>* null         ; <<4 x float>> [#uses=1]
         %tmp40 = fadd <4 x float> %tmp39, < float 2.000000e+00, float 3.000000e+00, float 1.000000e+00, float 0.000000e+00 >             ; <<4 x float>> [#uses=1]
         %tmp43 = fadd <4 x float> %tmp40, < float 1.000000e+00, float 1.000000e+00, float 0.000000e+00, float 2.000000e+00 >             ; <<4 x float>> [#uses=1]
         %tmp46 = fadd <4 x float> %tmp43, < float 3.000000e+00, float 0.000000e+00, float 2.000000e+00, float 4.000000e+00 >             ; <<4 x float>> [#uses=1]
diff --git a/test/Transforms/LCSSA/2007-07-12-LICM.ll b/test/Transforms/LCSSA/2007-07-12-LICM.ll
index 8c07aa2ec352..f1785edc2af2 100644
--- a/test/Transforms/LCSSA/2007-07-12-LICM.ll
+++ b/test/Transforms/LCSSA/2007-07-12-LICM.ll
@@ -4,7 +4,7 @@ entry:
 	br label %bb7
 
 bb7:		; preds = %bb7, %entry
-	%tmp39 = load <4 x float>* null		; <<4 x float>> [#uses=1]
+	%tmp39 = load <4 x float>, <4 x float>* null		; <<4 x float>> [#uses=1]
 	%tmp40 = fadd <4 x float> %tmp39, < float 2.000000e+00, float 3.000000e+00, float 1.000000e+00, float 0.000000e+00 >		; <<4 x float>> [#uses=0]
 	store <4 x float> zeroinitializer, <4 x float>* null
 	br i1 false, label %bb7, label %bb56
diff --git a/test/Transforms/LCSSA/invoke-dest.ll b/test/Transforms/LCSSA/invoke-dest.ll
index 22b320296afd..1e3f178c7e6c 100644
--- a/test/Transforms/LCSSA/invoke-dest.ll
+++ b/test/Transforms/LCSSA/invoke-dest.ll
@@ -17,7 +17,7 @@ bb.i:		; preds = %entry
 	br label %_ZN7cObjectnwEj.exit
 
 _ZN7cObjectnwEj.exit:		; preds = %bb.i, %entry
-	invoke void @_ZN7cObjectC2EPKc(i8* undef, i8* getelementptr ([12 x i8]* @.str21179, i32 0, i32 0))
+	invoke void @_ZN7cObjectC2EPKc(i8* undef, i8* getelementptr ([12 x i8], [12 x i8]* @.str21179, i32 0, i32 0))
 			to label %bb1 unwind label %lpad
 
 bb1:		; preds = %_ZN7cObjectnwEj.exit
@@ -39,7 +39,7 @@ bb2:		; preds = %_ZNK5cGate4sizeEv.exit122
 	unreachable
 
 bb8:		; preds = %_ZNK5cGate4sizeEv.exit122
-	%tmp = invoke i8* @_ZN7cModule3parEPKc(i8* undef, i8* getelementptr ([10 x i8]* @.str25183, i32 0, i32 0))
+	%tmp = invoke i8* @_ZN7cModule3parEPKc(i8* undef, i8* getelementptr ([10 x i8], [10 x i8]* @.str25183, i32 0, i32 0))
 			to label %invcont9 unwind label %lpad119		; <i8*> [#uses=1]
 
 invcont9:		; preds = %bb8
@@ -54,7 +54,7 @@ invcont11:		; preds = %invcont10
 	br i1 undef, label %bb12, label %bb18
 
 bb12:		; preds = %invcont11
-	invoke void (i8*, i8*, ...)* @_ZN6cEnvir6printfEPKcz(i8* null, i8* getelementptr ([3 x i8]* @.str12, i32 0, i32 0), i32 undef)
+	invoke void (i8*, i8*, ...) @_ZN6cEnvir6printfEPKcz(i8* null, i8* getelementptr ([3 x i8], [3 x i8]* @.str12, i32 0, i32 0), i32 undef)
 			to label %bb.i.i159 unwind label %lpad119
 
 bb.i.i159:		; preds = %bb12
@@ -77,7 +77,7 @@ invcont35:		; preds = %bb34
 	br i1 undef, label %bb49, label %bb61
 
 bb49:		; preds = %invcont35
-	invoke void (i8*, i8*, ...)* @_ZNK13cSimpleModule5errorEPKcz(i8* undef, i8* getelementptr ([92 x i8]* @.str32190, i32 0, i32 0))
+	invoke void (i8*, i8*, ...) @_ZNK13cSimpleModule5errorEPKcz(i8* undef, i8* getelementptr ([92 x i8], [92 x i8]* @.str32190, i32 0, i32 0))
 			to label %bb51 unwind label %lpad119
 
 bb51:		; preds = %bb49
@@ -87,7 +87,7 @@ bb61:		; preds = %invcont35
 	br label %bb106
 
 .noexc:		; preds = %bb106
-	invoke void @_ZN7cObjectC2EPKc(i8* undef, i8* getelementptr ([25 x i8]* @.str41, i32 0, i32 0))
+	invoke void @_ZN7cObjectC2EPKc(i8* undef, i8* getelementptr ([25 x i8], [25 x i8]* @.str41, i32 0, i32 0))
 			to label %bb102 unwind label %lpad123
 
 bb102:		; preds = %.noexc
@@ -99,7 +99,7 @@ invcont103:		; preds = %bb102
 			to label %invcont104 unwind label %lpad119
 
 invcont104:		; preds = %invcont103
-	%tmp3 = invoke i32 @_ZN13cSimpleModule11sendDelayedEP8cMessagedPKci(i8* undef, i8* undef, double 0.000000e+00, i8* getelementptr ([4 x i8]* @.str17175, i32 0, i32 0), i32 undef)
+	%tmp3 = invoke i32 @_ZN13cSimpleModule11sendDelayedEP8cMessagedPKci(i8* undef, i8* undef, double 0.000000e+00, i8* getelementptr ([4 x i8], [4 x i8]* @.str17175, i32 0, i32 0), i32 undef)
 			to label %invcont105 unwind label %lpad119		; <i32> [#uses=0]
 
 invcont105:		; preds = %invcont104
diff --git a/test/Transforms/LCSSA/unreachable-use.ll b/test/Transforms/LCSSA/unreachable-use.ll
index 71ae134e314c..c9e456c17474 100644
--- a/test/Transforms/LCSSA/unreachable-use.ll
+++ b/test/Transforms/LCSSA/unreachable-use.ll
@@ -1,9 +1,11 @@
-; RUN: opt < %s -lcssa -S -verify-loop-info | grep "[%]tmp33 = load i1\*\* [%]tmp"
+; RUN: opt < %s -lcssa -S -verify-loop-info | FileCheck %s
 ; PR6546
 
 ; LCSSA doesn't need to transform uses in blocks not reachable
 ; from the entry block.
 
+; CHECK: %tmp33 = load i1*, i1** %tmp
+
 define fastcc void @dfs() nounwind {
 bb:
   br label %bb44
@@ -19,7 +21,7 @@ bb15:
   br label %bb44
 
 bb32:
-  %tmp33 = load i1** %tmp, align 8
+  %tmp33 = load i1*, i1** %tmp, align 8
   br label %bb45
 
 bb45:
diff --git a/test/Transforms/LICM/2003-02-26-LoopExitNotDominated.ll b/test/Transforms/LICM/2003-02-26-LoopExitNotDominated.ll
index ff20312ec372..1b50a6c87af7 100644
--- a/test/Transforms/LICM/2003-02-26-LoopExitNotDominated.ll
+++ b/test/Transforms/LICM/2003-02-26-LoopExitNotDominated.ll
@@ -7,7 +7,7 @@ bb3:
 	%X = alloca [2 x i64]		; <[2 x i64]*> [#uses=1]
 	br i1 false, label %bb13, label %bb4
 bb4:		; preds = %bb3
-	%reg3011 = getelementptr [2 x i64]* %X, i64 0, i64 0		; <i64*> [#uses=1]
+	%reg3011 = getelementptr [2 x i64], [2 x i64]* %X, i64 0, i64 0		; <i64*> [#uses=1]
 	br label %bb8
 bb8:		; preds = %bb8, %bb4
 	store i64 0, i64* %reg3011
diff --git a/test/Transforms/LICM/2003-05-02-LoadHoist.ll b/test/Transforms/LICM/2003-05-02-LoadHoist.ll
index 71d3e789aa08..2f94dff4efd7 100644
--- a/test/Transforms/LICM/2003-05-02-LoadHoist.ll
+++ b/test/Transforms/LICM/2003-05-02-LoadHoist.ll
@@ -10,12 +10,12 @@
 declare void @foo()
 
 define i32 @test(i1 %c) {
-	%A = load i32* @X		; <i32> [#uses=1]
+	%A = load i32, i32* @X		; <i32> [#uses=1]
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	call void @foo( )
         ;; Should not hoist this load!
-	%B = load i32* @X		; <i32> [#uses=1]
+	%B = load i32, i32* @X		; <i32> [#uses=1]
 	br i1 %c, label %Loop, label %Out
 Out:		; preds = %Loop
 	%C = sub i32 %A, %B		; <i32> [#uses=1]
diff --git a/test/Transforms/LICM/2004-09-14-AliasAnalysisInvalidate.ll b/test/Transforms/LICM/2004-09-14-AliasAnalysisInvalidate.ll
index 16f4fed34ec3..73862db69819 100644
--- a/test/Transforms/LICM/2004-09-14-AliasAnalysisInvalidate.ll
+++ b/test/Transforms/LICM/2004-09-14-AliasAnalysisInvalidate.ll
@@ -7,8 +7,8 @@ define void @test() {
 Outer:		; preds = %Next, %0
 	br label %Inner
 Inner:		; preds = %Inner, %Outer
-	%tmp.114.i.i.i = load i8** @PL_regcomp_parse		; <i8*> [#uses=1]
-	%tmp.115.i.i.i = load i8* %tmp.114.i.i.i		; <i8> [#uses=0]
+	%tmp.114.i.i.i = load i8*, i8** @PL_regcomp_parse		; <i8*> [#uses=1]
+	%tmp.115.i.i.i = load i8, i8* %tmp.114.i.i.i		; <i8> [#uses=0]
 	store i8* null, i8** @PL_regcomp_parse
 	br i1 false, label %Inner, label %Next
 Next:		; preds = %Inner
diff --git a/test/Transforms/LICM/2004-11-17-UndefIndexCrash.ll b/test/Transforms/LICM/2004-11-17-UndefIndexCrash.ll
index a119865f0519..9416028ef2e7 100644
--- a/test/Transforms/LICM/2004-11-17-UndefIndexCrash.ll
+++ b/test/Transforms/LICM/2004-11-17-UndefIndexCrash.ll
@@ -11,10 +11,10 @@ define void @main() {
 __main.entry:
 	br label %invoke_cont.3
 invoke_cont.3:		; preds = %invoke_cont.3, %__main.entry
-	%tmp.34.i.i502.7 = getelementptr %struct.roadlet* null, i32 0, i32 3, i32 7		; <%struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)**> [#uses=1]
+	%tmp.34.i.i502.7 = getelementptr %struct.roadlet, %struct.roadlet* null, i32 0, i32 3, i32 7		; <%struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)**> [#uses=1]
 	store %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)* @_Z11return_nullP7roadletP7vehicle9direction, %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)** %tmp.34.i.i502.7
 	store %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)* @_Z14lane_switch_okP7roadletP7vehicle9direction, %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)** null
-	%tmp.4.i.i339 = getelementptr %struct.roadlet* null, i32 0, i32 3, i32 undef		; <%struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)**> [#uses=1]
+	%tmp.4.i.i339 = getelementptr %struct.roadlet, %struct.roadlet* null, i32 0, i32 3, i32 undef		; <%struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)**> [#uses=1]
 	store %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)* @_Z11return_nullP7roadletP7vehicle9direction, %struct.roadlet* (%struct.roadlet*, %struct.vehicle*, i32)** %tmp.4.i.i339
 	br label %invoke_cont.3
 }
diff --git a/test/Transforms/LICM/2007-05-22-VolatileSink.ll b/test/Transforms/LICM/2007-05-22-VolatileSink.ll
index 94511cc9c156..f5ce86b7245a 100644
--- a/test/Transforms/LICM/2007-05-22-VolatileSink.ll
+++ b/test/Transforms/LICM/2007-05-22-VolatileSink.ll
@@ -10,8 +10,8 @@ entry:
 	br label %bb6
 
 bb:		; preds = %bb6
-	%tmp2 = load volatile i32* %DataIn		; <i32> [#uses=1]
-	%tmp3 = getelementptr [64 x i32]* %buffer, i32 0, i32 %i.0		; <i32*> [#uses=1]
+	%tmp2 = load volatile i32, i32* %DataIn		; <i32> [#uses=1]
+	%tmp3 = getelementptr [64 x i32], [64 x i32]* %buffer, i32 0, i32 %i.0		; <i32*> [#uses=1]
 	store i32 %tmp2, i32* %tmp3
 	%tmp5 = add i32 %i.0, 1		; <i32> [#uses=1]
 	br label %bb6
@@ -26,8 +26,8 @@ bb6:		; preds = %bb, %entry
 bb12:		; preds = %bb22
 	%tmp14 = mul i32 %j.1, 8		; <i32> [#uses=1]
 	%tmp16 = add i32 %tmp14, %i.1		; <i32> [#uses=1]
-	%tmp17 = getelementptr [64 x i32]* %buffer, i32 0, i32 %tmp16		; <i32*> [#uses=1]
-	%tmp18 = load i32* %tmp17		; <i32> [#uses=1]
+	%tmp17 = getelementptr [64 x i32], [64 x i32]* %buffer, i32 0, i32 %tmp16		; <i32*> [#uses=1]
+	%tmp18 = load i32, i32* %tmp17		; <i32> [#uses=1]
 	store volatile i32 %tmp18, i32* %DataOut
 	%tmp21 = add i32 %j.1, 1		; <i32> [#uses=1]
 	br label %bb22
diff --git a/test/Transforms/LICM/2007-07-30-AliasSet.ll b/test/Transforms/LICM/2007-07-30-AliasSet.ll
index 8ecd1bcee486..3e4fbb0a06e0 100644
--- a/test/Transforms/LICM/2007-07-30-AliasSet.ll
+++ b/test/Transforms/LICM/2007-07-30-AliasSet.ll
@@ -17,12 +17,12 @@ bb63.outer:		; preds = %bb73, %bb28
 
 bb35:		; preds = %cond_next60, %bb63.outer
 	%window.34 = phi i32 [ %tmp62, %cond_next60 ], [ 0, %bb63.outer ]		; <i32> [#uses=1]
-	%tmp44 = getelementptr [4 x i32]* null, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp46 = load i32* %tmp44, align 4		; <i32> [#uses=0]
+	%tmp44 = getelementptr [4 x i32], [4 x i32]* null, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp46 = load i32, i32* %tmp44, align 4		; <i32> [#uses=0]
 	br i1 false, label %cond_true50, label %cond_next60
 
 cond_true50:		; preds = %bb35
-	%tmp59 = getelementptr [4 x i32]* null, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp59 = getelementptr [4 x i32], [4 x i32]* null, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp59, align 4
 	br label %cond_next60
 
diff --git a/test/Transforms/LICM/2007-09-17-PromoteValue.ll b/test/Transforms/LICM/2007-09-17-PromoteValue.ll
index 31abd8c180a5..1d25d17c2e64 100644
--- a/test/Transforms/LICM/2007-09-17-PromoteValue.ll
+++ b/test/Transforms/LICM/2007-09-17-PromoteValue.ll
@@ -14,7 +14,7 @@ blah.i:		; preds = %cond_true.i, %entry
 	br i1 %tmp3.i, label %clear_modes.exit, label %cond_true.i
 
 cond_true.i:		; preds = %blah.i
-	%tmp1.i = getelementptr %struct.decision* null, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp1.i = getelementptr %struct.decision, %struct.decision* null, i32 0, i32 0		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp1.i
 	br label %blah.i
 
@@ -33,7 +33,7 @@ loop.head:              ; preds = %cond.true, %entry
         br i1 %tmp3.i, label %cond.true, label %exit
 
 cond.true:              ; preds = %loop.head
-        %ptr.i = getelementptr i8* %ptr, i32 0          ; <i8*> [#uses=2]
+        %ptr.i = getelementptr i8, i8* %ptr, i32 0          ; <i8*> [#uses=2]
         store i8 0, i8* %ptr.i
         br label %loop.head
 
@@ -50,7 +50,7 @@ loop.head:              ; preds = %cond.true, %entry
         br i1 %tmp3.i, label %exit, label %cond.true
 
 cond.true:              ; preds = %loop.head
-        %ptr.i = getelementptr i8* %p, i32 0          ; <i8*> [#uses=2]
+        %ptr.i = getelementptr i8, i8* %p, i32 0          ; <i8*> [#uses=2]
         store i8 0, i8* %ptr.i
         br label %loop.head
 
diff --git a/test/Transforms/LICM/2008-07-22-LoadGlobalConstant.ll b/test/Transforms/LICM/2008-07-22-LoadGlobalConstant.ll
index d4df26e67abc..a715af032a96 100644
--- a/test/Transforms/LICM/2008-07-22-LoadGlobalConstant.ll
+++ b/test/Transforms/LICM/2008-07-22-LoadGlobalConstant.ll
@@ -6,7 +6,7 @@ define void @test(i32 %count) {
 entry:
         br label %forcond
 
-; CHECK:  %tmp3 = load float** @a
+; CHECK:  %tmp3 = load float*, float** @a
 ; CHECK:  br label %forcond
 
 forcond:
@@ -19,14 +19,14 @@ forcond:
 ; CHECK:  br i1 %cmp, label %forbody, label %afterfor
 
 forbody:
-        %tmp3 = load float** @a
-        %arrayidx = getelementptr float* %tmp3, i32 %i.0
+        %tmp3 = load float*, float** @a
+        %arrayidx = getelementptr float, float* %tmp3, i32 %i.0
         %tmp7 = uitofp i32 %i.0 to float
         store float %tmp7, float* %arrayidx
         %inc = add i32 %i.0, 1
         br label %forcond
 
-; CHECK:  %arrayidx = getelementptr float* %tmp3, i32 %i.0
+; CHECK:  %arrayidx = getelementptr float, float* %tmp3, i32 %i.0
 ; CHECK:  %tmp7 = uitofp i32 %i.0 to float
 ; CHECK:  store float %tmp7, float* %arrayidx
 ; CHECK:  %inc = add i32 %i.0, 1
diff --git a/test/Transforms/LICM/2009-12-10-LICM-Indbr-Crash.ll b/test/Transforms/LICM/2009-12-10-LICM-Indbr-Crash.ll
index e3cdbb3d2a40..1b3ff5bbb318 100644
--- a/test/Transforms/LICM/2009-12-10-LICM-Indbr-Crash.ll
+++ b/test/Transforms/LICM/2009-12-10-LICM-Indbr-Crash.ll
@@ -12,7 +12,7 @@ define void @foo (i8* %v)
     indirectbr i8* undef, [label %preheader, label %stuff]
 
   stuff:
-    %0 = load i8* undef, align 1
+    %0 = load i8, i8* undef, align 1
     br label %loop
 
   return:
diff --git a/test/Transforms/LICM/2011-04-06-HoistMissedASTUpdate.ll b/test/Transforms/LICM/2011-04-06-HoistMissedASTUpdate.ll
index 2bbc6ab0414a..b462885e5fbc 100644
--- a/test/Transforms/LICM/2011-04-06-HoistMissedASTUpdate.ll
+++ b/test/Transforms/LICM/2011-04-06-HoistMissedASTUpdate.ll
@@ -15,11 +15,11 @@ for.body4.lr.ph:
   br label %for.body4
 
 ; CHECK: for.body4:
-; CHECK: load volatile i16* @g_39
+; CHECK: load volatile i16, i16* @g_39
 
 for.body4:
   %l_612.11 = phi i32* [ undef, %for.body4.lr.ph ], [ %call19, %for.body4 ]
-  %tmp7 = load volatile i16* @g_39, align 2
+  %tmp7 = load volatile i16, i16* @g_39, align 2
   %call = call i32** @func_108(i32*** undef)
   %call19 = call i32* @func_84(i32** %call)
   br i1 false, label %for.body4, label %for.cond.loopexit
diff --git a/test/Transforms/LICM/2011-04-06-PromoteResultOfPromotion.ll b/test/Transforms/LICM/2011-04-06-PromoteResultOfPromotion.ll
index 5587142dcf76..370491eab980 100644
--- a/test/Transforms/LICM/2011-04-06-PromoteResultOfPromotion.ll
+++ b/test/Transforms/LICM/2011-04-06-PromoteResultOfPromotion.ll
@@ -8,7 +8,7 @@ define void @f() nounwind {
 
 ; CHECK: entry:
 ; CHECK: alloca [9 x i16]
-; CHECK: load i32* @g_58
+; CHECK: load i32, i32* @g_58
 ; CHECK: br label %for.body
 
 entry:
@@ -18,8 +18,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.inc
   %inc12 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   store i32* @g_58, i32** @g_116, align 8, !tbaa !0
-  %tmp2 = load i32** @g_116, align 8, !tbaa !0
-  %tmp3 = load i32* %tmp2, !tbaa !4
+  %tmp2 = load i32*, i32** @g_116, align 8, !tbaa !0
+  %tmp3 = load i32, i32* %tmp2, !tbaa !4
   %or = or i32 %tmp3, 10
   store i32 %or, i32* %tmp2, !tbaa !4
   %inc = add nsw i32 %inc12, 1
diff --git a/test/Transforms/LICM/2011-04-09-RAUW-AST.ll b/test/Transforms/LICM/2011-04-09-RAUW-AST.ll
index 4285bd19e5f1..bf069c2a9917 100644
--- a/test/Transforms/LICM/2011-04-09-RAUW-AST.ll
+++ b/test/Transforms/LICM/2011-04-09-RAUW-AST.ll
@@ -6,7 +6,7 @@
 
 define i32 @main() nounwind {
 entry:
-  %tmp = load i32* @g_3, align 4
+  %tmp = load i32, i32* @g_3, align 4
   %tobool = icmp eq i32 %tmp, 0
   br i1 %tobool, label %for.cond, label %if.then
 
@@ -40,8 +40,8 @@ for.inc10:                                        ; preds = %for.cond4
   br label %for.cond
 
 for.end13:                                        ; preds = %for.cond
-  %tmp14 = load i32* @g_3, align 4
-  %call = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i32 %tmp14) nounwind
+  %tmp14 = load i32, i32* @g_3, align 4
+  %call = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i32 %tmp14) nounwind
   ret i32 0
 }
 
diff --git a/test/Transforms/LICM/2011-07-06-Alignment.ll b/test/Transforms/LICM/2011-07-06-Alignment.ll
index 569231489fec..c71dd802ea9b 100644
--- a/test/Transforms/LICM/2011-07-06-Alignment.ll
+++ b/test/Transforms/LICM/2011-07-06-Alignment.ll
@@ -8,7 +8,7 @@ entry:
 
 for.cond:
   %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr [1024 x float]* @A, i64 0, i64 3
+  %arrayidx = getelementptr [1024 x float], [1024 x float]* @A, i64 0, i64 3
   %vecidx = bitcast float* %arrayidx to <4 x float>*
   store <4 x float> zeroinitializer, <4 x float>* %vecidx, align 4
   %indvar.next = add i64 %indvar, 1
diff --git a/test/Transforms/LICM/PR21582.ll b/test/Transforms/LICM/PR21582.ll
index c068c2f8e32b..5664f2e2b6e0 100644
--- a/test/Transforms/LICM/PR21582.ll
+++ b/test/Transforms/LICM/PR21582.ll
@@ -19,8 +19,8 @@ for.body.preheader:                               ; preds = %for.cond
 
 for.body:                                         ; preds = %for.body, %for.body.preheader
   %g.15 = phi i32 [ undef, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx2 = getelementptr inbounds i32* @fn3.i, i64 0
-  %0 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* @fn3.i, i64 0
+  %0 = load i32, i32* %arrayidx2, align 4
   %call = call i32 @g()
   br i1 false, label %for.body, label %for.end.loopexit
 
diff --git a/test/Transforms/LICM/atomics.ll b/test/Transforms/LICM/atomics.ll
index acf605d2dad8..4fe197abf5d3 100644
--- a/test/Transforms/LICM/atomics.ll
+++ b/test/Transforms/LICM/atomics.ll
@@ -7,7 +7,7 @@ entry:
 
 loop:
   %i = phi i32 [ %inc, %loop ], [ 0, %entry ]
-  %val = load atomic i32* %y unordered, align 4
+  %val = load atomic i32, i32* %y unordered, align 4
   %inc = add nsw i32 %i, 1
   %exitcond = icmp eq i32 %inc, %val
   br i1 %exitcond, label %end, label %loop
@@ -27,7 +27,7 @@ entry:
   br label %loop
 
 loop:
-  %val = load atomic i32* %y monotonic, align 4
+  %val = load atomic i32, i32* %y monotonic, align 4
   %exitcond = icmp ne i32 %val, 0
   br i1 %exitcond, label %end, label %loop
 
@@ -47,15 +47,15 @@ entry:
   br label %loop
 
 loop:
-  %vala = load atomic i32* %y monotonic, align 4
-  %valb = load atomic i32* %x unordered, align 4
+  %vala = load atomic i32, i32* %y monotonic, align 4
+  %valb = load atomic i32, i32* %x unordered, align 4
   %exitcond = icmp ne i32 %vala, %valb
   br i1 %exitcond, label %end, label %loop
 
 end:
   ret i32 %vala
 ; CHECK-LABEL: define i32 @test3(
-; CHECK: load atomic i32* %x unordered
+; CHECK: load atomic i32, i32* %x unordered
 ; CHECK-NEXT: br label %loop
 }
 
@@ -66,7 +66,7 @@ entry:
   br label %loop
 
 loop:
-  %vala = load atomic i32* %y monotonic, align 4
+  %vala = load atomic i32, i32* %y monotonic, align 4
   store atomic i32 %vala, i32* %x unordered, align 4
   %exitcond = icmp ne i32 %vala, 0
   br i1 %exitcond, label %end, label %loop
@@ -74,6 +74,6 @@ loop:
 end:
   ret i32 %vala
 ; CHECK-LABEL: define i32 @test4(
-; CHECK: load atomic i32* %y monotonic
+; CHECK: load atomic i32, i32* %y monotonic
 ; CHECK-NEXT: store atomic
 }
diff --git a/test/Transforms/LICM/constexpr.ll b/test/Transforms/LICM/constexpr.ll
new file mode 100644
index 000000000000..506721f25f8f
--- /dev/null
+++ b/test/Transforms/LICM/constexpr.ll
@@ -0,0 +1,46 @@
+; RUN: opt < %s -S -basicaa -licm | FileCheck %s
+; This fixes PR22460
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-pc-windows-msvc"
+
+@in = internal unnamed_addr global i32* null, align 8
+@out = internal unnamed_addr global i32* null, align 8
+
+; CHECK-LABEL: @bar
+; CHECK: entry:
+; CHECK: load i64, i64* bitcast (i32** @in to i64*)
+; CHECK: do.body:
+; CHECK-NOT: load
+
+define i64 @bar(i32 %N) {
+entry:
+  br label %do.body
+
+do.body:                                          ; preds = %l2, %entry
+  %i.0 = phi i32 [ 0, %entry ], [ %inc, %l2 ]
+  %total = phi i64 [ 0, %entry ], [ %next, %l2 ]
+  %c = icmp eq i32 %N, 6
+  br i1 %c, label %l1, label %do.body.l2_crit_edge
+
+do.body.l2_crit_edge:                             ; preds = %do.body
+  %inval.pre = load i32*, i32** @in, align 8
+  br label %l2
+
+l1:                                               ; preds = %do.body
+  %v1 = load i64, i64* bitcast (i32** @in to i64*), align 8
+  store i64 %v1, i64* bitcast (i32** @out to i64*), align 8
+  %0 = inttoptr i64 %v1 to i32*
+  br label %l2
+
+l2:                                               ; preds = %do.body.l2_crit_edge, %l1
+  %inval = phi i32* [ %inval.pre, %do.body.l2_crit_edge ], [ %0, %l1 ]
+  %int = ptrtoint i32* %inval to i64
+  %next = add i64 %total, %int
+  %inc = add nsw i32 %i.0, 1
+  %cmp = icmp slt i32 %inc, %N
+  br i1 %cmp, label %do.body, label %do.end
+
+do.end:                                           ; preds = %l2
+  ret i64 %total
+}
diff --git a/test/Transforms/LICM/crash.ll b/test/Transforms/LICM/crash.ll
index b43477a56df5..7fa41157338d 100644
--- a/test/Transforms/LICM/crash.ll
+++ b/test/Transforms/LICM/crash.ll
@@ -12,7 +12,7 @@ entry:
 
 for.body:                                         ; preds = %for.cond, %bb.nph
   store i8 0, i8* @g_12, align 1
-  %tmp6 = load i8* @g_12, align 1
+  %tmp6 = load i8, i8* @g_12, align 1
   br label %for.cond
 
 for.cond:                                         ; preds = %for.body
@@ -34,7 +34,7 @@ entry:
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %entry
-  %tmp7 = load i32* @g_8, align 4
+  %tmp7 = load i32, i32* @g_8, align 4
   store i32* @g_8, i32** undef, align 16
   store i32 undef, i32* @g_8, align 4
   br label %for.body
@@ -47,10 +47,10 @@ entry:
   br i1 undef, label %for.cond, label %for.end
 
 for.cond:                                         ; preds = %for.cond, %entry
-  %tmp1 = getelementptr { i32*}* %__first, i32 0, i32 0
-  %tmp2 = load i32** %tmp1, align 4
+  %tmp1 = getelementptr { i32*}, { i32*}* %__first, i32 0, i32 0
+  %tmp2 = load i32*, i32** %tmp1, align 4
   %call = tail call i32* @test3helper(i32* %tmp2)
-  %tmp3 = getelementptr { i32*}* %__first, i32 0, i32 0
+  %tmp3 = getelementptr { i32*}, { i32*}* %__first, i32 0, i32 0
   store i32* %call, i32** %tmp3, align 4
   br i1 false, label %for.cond, label %for.end
 
diff --git a/test/Transforms/LICM/debug-value.ll b/test/Transforms/LICM/debug-value.ll
index b49c559672a6..62429fdc3216 100644
--- a/test/Transforms/LICM/debug-value.ll
+++ b/test/Transforms/LICM/debug-value.ll
@@ -15,7 +15,7 @@ if.then:                                          ; preds = %for.body
 
 if.then27:                                        ; preds = %if.then
 ; CHECK: tail call void @llvm.dbg.value
-  tail call void @llvm.dbg.value(metadata double undef, i64 0, metadata !19, metadata !{}), !dbg !21
+  tail call void @llvm.dbg.value(metadata double undef, i64 0, metadata !19, metadata !DIExpression()), !dbg !21
   br label %for.body61.us
 
 if.end.if.end.split_crit_edge.critedge:           ; preds = %if.then
@@ -36,30 +36,30 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 !llvm.module.flags = !{!26}
 !llvm.dbg.sp = !{!0, !6, !9, !10}
 
-!0 = !{!"0x2e\00idamax\00idamax\00\00112\000\001\000\006\00256\000\000", !25, !1, !3, i32 0, null, null, null, null} ; [ DW_TAG_subprogram ]
-!1 = !{!"0x29", !25} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang version 2.9 (trunk 127169)\001\00\000\00\000", !25, !8, !8, !8, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !25, !1, null, !4, i32 0} ; [ DW_TAG_subroutine_type ]
+!0 = !DISubprogram(name: "idamax", line: 112, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !25, scope: !1, type: !3)
+!1 = !DIFile(filename: "/Volumes/Lalgate/work/llvm/projects/llvm-test/SingleSource/Benchmarks/CoyoteBench/lpbench.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 2.9 (trunk 127169)", isOptimized: true, emissionKind: 0, file: !25, enums: !8, retainedTypes: !8, subprograms: !8)
+!3 = !DISubroutineType(types: !4)
 !4 = !{!5}
-!5 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !2} ; [ DW_TAG_base_type ]
-!6 = !{!"0x2e\00dscal\00dscal\00\00206\000\001\000\006\00256\000\000", !25, !1, !7, i32 0, null, null, null, null} ; [ DW_TAG_subprogram ]
-!7 = !{!"0x15\00\000\000\000\000\000\000", !25, !1, null, !8, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!5 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!6 = !DISubprogram(name: "dscal", line: 206, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !25, scope: !1, type: !7)
+!7 = !DISubroutineType(types: !8)
 !8 = !{null}
-!9 = !{!"0x2e\00daxpy\00daxpy\00\00230\000\001\000\006\00256\000\000", !25, !1, !7, i32 0, null, null, null, null} ; [ DW_TAG_subprogram ]
-!10 = !{!"0x2e\00dgefa\00dgefa\00\00267\000\001\000\006\00256\000\000", !25, !1, !7, null, null, null, null, null} ; [ DW_TAG_subprogram ] [line 267] [def] [scope 0] [dgefa]
-!11 = !MDLocation(line: 281, column: 9, scope: !12)
-!12 = !{!"0xb\00272\005\0032", !25, !13} ; [ DW_TAG_lexical_block ]
-!13 = !{!"0xb\00271\005\0031", !25, !14} ; [ DW_TAG_lexical_block ]
-!14 = !{!"0xb\00267\001\0030", !25, !10} ; [ DW_TAG_lexical_block ]
-!15 = !MDLocation(line: 271, column: 5, scope: !14)
-!16 = !MDLocation(line: 284, column: 10, scope: !17)
-!17 = !{!"0xb\00282\009\0033", !25, !12} ; [ DW_TAG_lexical_block ]
+!9 = !DISubprogram(name: "daxpy", line: 230, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !25, scope: !1, type: !7)
+!10 = !DISubprogram(name: "dgefa", line: 267, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !25, scope: !1, type: !7)
+!11 = !DILocation(line: 281, column: 9, scope: !12)
+!12 = distinct !DILexicalBlock(line: 272, column: 5, file: !25, scope: !13)
+!13 = distinct !DILexicalBlock(line: 271, column: 5, file: !25, scope: !14)
+!14 = distinct !DILexicalBlock(line: 267, column: 1, file: !25, scope: !10)
+!15 = !DILocation(line: 271, column: 5, scope: !14)
+!16 = !DILocation(line: 284, column: 10, scope: !17)
+!17 = distinct !DILexicalBlock(line: 282, column: 9, file: !25, scope: !12)
 !18 = !{double undef}
-!19 = !{!"0x100\00temp\00268\000", !14, !1, !20} ; [ DW_TAG_auto_variable ]
-!20 = !{!"0x24\00double\000\0064\0064\000\000\004", null, !2} ; [ DW_TAG_base_type ]
-!21 = !MDLocation(line: 286, column: 14, scope: !22)
-!22 = !{!"0xb\00285\0013\0034", !25, !17} ; [ DW_TAG_lexical_block ]
-!23 = !MDLocation(line: 296, column: 13, scope: !17)
-!24 = !MDLocation(line: 313, column: 1, scope: !14)
-!25 = !{!"/Volumes/Lalgate/work/llvm/projects/llvm-test/SingleSource/Benchmarks/CoyoteBench/lpbench.c", !"/private/tmp"}
-!26 = !{i32 1, !"Debug Info Version", i32 2}
+!19 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "temp", line: 268, scope: !14, file: !1, type: !20)
+!20 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
+!21 = !DILocation(line: 286, column: 14, scope: !22)
+!22 = distinct !DILexicalBlock(line: 285, column: 13, file: !25, scope: !17)
+!23 = !DILocation(line: 296, column: 13, scope: !17)
+!24 = !DILocation(line: 313, column: 1, scope: !14)
+!25 = !DIFile(filename: "/Volumes/Lalgate/work/llvm/projects/llvm-test/SingleSource/Benchmarks/CoyoteBench/lpbench.c", directory: "/private/tmp")
+!26 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/LICM/hoist-bitcast-load.ll b/test/Transforms/LICM/hoist-bitcast-load.ll
index fa61eaf5b46f..47c474c17dde 100644
--- a/test/Transforms/LICM/hoist-bitcast-load.ll
+++ b/test/Transforms/LICM/hoist-bitcast-load.ll
@@ -4,7 +4,7 @@ target triple = "x86_64-unknown-linux-gnu"
 
 ; Make sure the basic alloca pointer hoisting works:
 ; CHECK-LABEL: @test1
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 ; CHECK: for.body:
 
 ; Function Attrs: nounwind uwtable
@@ -16,15 +16,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -42,7 +42,7 @@ for.end:                                          ; preds = %for.inc, %entry
 ; Make sure the basic alloca pointer hoisting works through a bitcast to a
 ; pointer to a smaller type:
 ; CHECK-LABEL: @test2
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 ; CHECK: for.body:
 
 ; Function Attrs: nounwind uwtable
@@ -55,15 +55,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -80,7 +80,7 @@ for.end:                                          ; preds = %for.inc, %entry
 
 ; Make sure the basic alloca pointer hoisting works through an addrspacecast
 ; CHECK-LABEL: @test2_addrspacecast
-; CHECK: load i32 addrspace(1)* %c, align 4
+; CHECK: load i32, i32 addrspace(1)* %c, align 4
 ; CHECK: for.body:
 
 ; Function Attrs: nounwind uwtable
@@ -93,15 +93,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %a, i64 %indvars.iv
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %a, i64 %indvars.iv
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32 addrspace(1)* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32 addrspace(1)* %b, i64 %indvars.iv
-  %2 = load i32 addrspace(1)* %arrayidx3, align 4
+  %1 = load i32, i32 addrspace(1)* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %indvars.iv
+  %2 = load i32, i32 addrspace(1)* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32 addrspace(1)* %arrayidx, align 4
   br label %for.inc
@@ -119,7 +119,7 @@ for.end:                                          ; preds = %for.inc, %entry
 ; Make sure the basic alloca pointer hoisting works through a bitcast to a
 ; pointer to a smaller type (where the bitcast also needs to be hoisted):
 ; CHECK-LABEL: @test3
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 ; CHECK: for.body:
 
 ; Function Attrs: nounwind uwtable
@@ -131,16 +131,16 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
   %c = bitcast i64* %ca to i32*
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -159,7 +159,7 @@ for.end:                                          ; preds = %for.inc, %entry
 ; to a pointer to a larger type:
 ; CHECK-LABEL: @test4
 ; CHECK: for.body:
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 
 ; Function Attrs: nounwind uwtable
 define void @test4(i32* nocapture %a, i32* nocapture readonly %b, i32 %n) #0 {
@@ -171,15 +171,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -197,7 +197,7 @@ for.end:                                          ; preds = %for.inc, %entry
 ; Don't crash on bitcasts to unsized types.
 ; CHECK-LABEL: @test5
 ; CHECK: for.body:
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 
 %atype = type opaque
 
@@ -212,15 +212,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
diff --git a/test/Transforms/LICM/hoist-deref-load.ll b/test/Transforms/LICM/hoist-deref-load.ll
index c230d1dcd43b..c384a858d1e6 100644
--- a/test/Transforms/LICM/hoist-deref-load.ll
+++ b/test/Transforms/LICM/hoist-deref-load.ll
@@ -12,7 +12,7 @@ target triple = "x86_64-unknown-linux-gnu"
 ; because the dereferenceable attribute is on %c.
 
 ; CHECK-LABEL: @test1
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 ; CHECK: for.body:
 
 define void @test1(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* nocapture readonly nonnull dereferenceable(4) %c, i32 %n) #0 {
@@ -22,15 +22,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -50,7 +50,7 @@ for.end:                                          ; preds = %for.inc, %entry
 
 ; CHECK-LABEL: @test2
 ; CHECK: if.then:
-; CHECK: load i32* %c, align 4
+; CHECK: load i32, i32* %c, align 4
 
 define void @test2(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* nocapture readonly nonnull %c, i32 %n) #0 {
 entry:
@@ -59,15 +59,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %1 = load i32* %c, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -92,7 +92,7 @@ for.end:                                          ; preds = %for.inc, %entry
 ; because the dereferenceable attribute is on %c.
 
 ; CHECK-LABEL: @test3
-; CHECK: load i32* %c2, align 4
+; CHECK: load i32, i32* %c2, align 4
 ; CHECK: for.body:
 
 define void @test3(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* nocapture readonly dereferenceable(12) %c, i32 %n) #0 {
@@ -102,16 +102,16 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %c2 = getelementptr inbounds i32* %c, i64 2
-  %1 = load i32* %c2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %c2 = getelementptr inbounds i32, i32* %c, i64 2
+  %1 = load i32, i32* %c2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -131,7 +131,7 @@ for.end:                                          ; preds = %for.inc, %entry
 
 ; CHECK-LABEL: @test4
 ; CHECK: if.then:
-; CHECK: load i32* %c2, align 4
+; CHECK: load i32, i32* %c2, align 4
 
 define void @test4(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* nocapture readonly dereferenceable(11) %c, i32 %n) #0 {
 entry:
@@ -140,16 +140,16 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 0
   br i1 %cmp1, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %c2 = getelementptr inbounds i32* %c, i64 2
-  %1 = load i32* %c2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx3, align 4
+  %c2 = getelementptr inbounds i32, i32* %c, i64 2
+  %1 = load i32, i32* %c2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
   %mul = mul nsw i32 %2, %1
   store i32 %mul, i32* %arrayidx, align 4
   br label %for.inc
@@ -164,5 +164,229 @@ for.end:                                          ; preds = %for.inc, %entry
   ret void
 }
 
-attributes #0 = { nounwind uwtable }
+; This test represents the following function:
+; void test1(int * __restrict__ a, int *b, int &c, int n) {
+;   if (c != null)
+;     for (int i = 0; i < n; ++i)
+;       if (a[i] > 0)
+;         a[i] = c*b[i];
+; }
+; and we want to hoist the load of %c out of the loop. This can be done only
+; because the dereferenceable_or_null attribute is on %c and there is a null
+; check on %c.
+
+; CHECK-LABEL: @test5
+; CHECK: load i32, i32* %c, align 4
+; CHECK: for.body:
+
+define void @test5(i32* noalias %a, i32* %b, i32* dereferenceable_or_null(4) %c, i32 %n) #0 {
+entry:
+  %not_null = icmp ne i32* %c, null
+  br i1 %not_null, label %not.null, label %for.end
+
+not.null:
+  %cmp11 = icmp sgt i32 %n, 0
+  br i1 %cmp11, label %for.body, label %for.end
+
+for.body:                                         ; preds = %not.null, %for.inc
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %not.null ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %cmp1 = icmp sgt i32 %0, 0
+  br i1 %cmp1, label %if.then, label %for.inc
+
+if.then:                                          ; preds = %for.body
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
+  %mul = mul nsw i32 %2, %1
+  store i32 %mul, i32* %arrayidx, align 4
+  br label %for.inc
+
+for.inc:                                          ; preds = %for.body, %if.then
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.inc, %entry, %not.null
+  ret void
+}
+
+; This is the same as @test5, but without the null check on %c.
+; Without this check, we should not hoist the load of %c.
+
+; This test case has an icmp on c but the use of this comparison is
+; not a branch. 
+
+; CHECK-LABEL: @test6
+; CHECK: if.then:
+; CHECK: load i32, i32* %c, align 4
+
+define i1 @test6(i32* noalias %a, i32* %b, i32* dereferenceable_or_null(4) %c, i32 %n) #0 {
+entry:
+  %not_null = icmp ne i32* %c, null
+  %cmp11 = icmp sgt i32 %n, 0
+  br i1 %cmp11, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.inc
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %cmp1 = icmp sgt i32 %0, 0
+  br i1 %cmp1, label %if.then, label %for.inc
+
+if.then:                                          ; preds = %for.body
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
+  %mul = mul nsw i32 %2, %1
+  store i32 %mul, i32* %arrayidx, align 4
+  br label %for.inc
+
+for.inc:                                          ; preds = %for.body, %if.then
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.inc, %entry
+  ret i1 %not_null
+}
+
+; This test represents the following function:
+; void test1(int * __restrict__ a, int *b, int **cptr, int n) {
+;   c = *cptr;
+;   for (int i = 0; i < n; ++i)
+;     if (a[i] > 0)
+;       a[i] = (*c)*b[i];
+; }
+; and we want to hoist the load of %c out of the loop. This can be done only
+; because the dereferenceable meatdata on the c = *cptr load.
+
+; CHECK-LABEL: @test7
+; CHECK: load i32, i32* %c, align 4
+; CHECK: for.body:
+
+define void @test7(i32* noalias %a, i32* %b, i32** %cptr, i32 %n) #0 {
+entry:
+  %c = load i32*, i32** %cptr, !dereferenceable !0
+  %cmp11 = icmp sgt i32 %n, 0
+  br i1 %cmp11, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.inc
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %cmp1 = icmp sgt i32 %0, 0
+  br i1 %cmp1, label %if.then, label %for.inc
 
+if.then:                                          ; preds = %for.body
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
+  %mul = mul nsw i32 %2, %1
+  store i32 %mul, i32* %arrayidx, align 4
+  br label %for.inc
+
+for.inc:                                          ; preds = %for.body, %if.then
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.inc, %entry
+  ret void
+}
+
+; This test represents the following function:
+; void test1(int * __restrict__ a, int *b, int **cptr, int n) {
+;   c = *cptr;
+;   if (c != null)
+;     for (int i = 0; i < n; ++i)
+;       if (a[i] > 0)
+;         a[i] = (*c)*b[i];
+; }
+; and we want to hoist the load of %c out of the loop. This can be done only
+; because the dereferenceable_or_null meatdata on the c = *cptr load and there 
+; is a null check on %c.
+
+; CHECK-LABEL: @test8
+; CHECK: load i32, i32* %c, align 4
+; CHECK: for.body:
+
+define void @test8(i32* noalias %a, i32* %b, i32** %cptr, i32 %n) #0 {
+entry:
+  %c = load i32*, i32** %cptr, !dereferenceable_or_null !0
+  %not_null = icmp ne i32* %c, null
+  br i1 %not_null, label %not.null, label %for.end
+
+not.null:
+  %cmp11 = icmp sgt i32 %n, 0
+  br i1 %cmp11, label %for.body, label %for.end
+
+for.body:                                         ; preds = %not.null, %for.inc
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %not.null ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %cmp1 = icmp sgt i32 %0, 0
+  br i1 %cmp1, label %if.then, label %for.inc
+
+if.then:                                          ; preds = %for.body
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
+  %mul = mul nsw i32 %2, %1
+  store i32 %mul, i32* %arrayidx, align 4
+  br label %for.inc
+
+for.inc:                                          ; preds = %for.body, %if.then
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.inc, %entry, %not.null
+  ret void
+}
+
+; This is the same as @test8, but without the null check on %c.
+; Without this check, we should not hoist the load of %c.
+
+; CHECK-LABEL: @test9
+; CHECK: if.then:
+; CHECK: load i32, i32* %c, align 4
+
+define void @test9(i32* noalias %a, i32* %b, i32** %cptr, i32 %n) #0 {
+entry:
+  %c = load i32*, i32** %cptr, !dereferenceable_or_null !0
+  %cmp11 = icmp sgt i32 %n, 0
+  br i1 %cmp11, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.inc
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %cmp1 = icmp sgt i32 %0, 0
+  br i1 %cmp1, label %if.then, label %for.inc
+
+if.then:                                          ; preds = %for.body
+  %1 = load i32, i32* %c, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx3, align 4
+  %mul = mul nsw i32 %2, %1
+  store i32 %mul, i32* %arrayidx, align 4
+  br label %for.inc
+
+for.inc:                                          ; preds = %for.body, %if.then
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.inc, %entry
+  ret void
+}
+
+attributes #0 = { nounwind uwtable }
+!0 = !{i64 4}
diff --git a/test/Transforms/LICM/hoist-invariant-load.ll b/test/Transforms/LICM/hoist-invariant-load.ll
index 59904bad4ae6..aec155b5580a 100644
--- a/test/Transforms/LICM/hoist-invariant-load.ll
+++ b/test/Transforms/LICM/hoist-invariant-load.ll
@@ -2,9 +2,9 @@
 ; RUN: opt < %s -licm -stats -S 2>&1 | grep "1 licm"
 
 @"\01L_OBJC_METH_VAR_NAME_" = internal global [4 x i8] c"foo\00", section "__TEXT,__objc_methname,cstring_literals", align 1
-@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([4 x i8], [4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
 @"\01L_OBJC_IMAGE_INFO" = internal constant [2 x i32] [i32 0, i32 16], section "__DATA, __objc_imageinfo, regular, no_dead_strip"
-@llvm.used = appending global [3 x i8*] [i8* getelementptr inbounds ([4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
+@llvm.used = appending global [3 x i8*] [i8* getelementptr inbounds ([4 x i8], [4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
 
 define void @test(i8* %x) uwtable ssp {
 entry:
@@ -15,18 +15,18 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp ult i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i8** %x.addr, align 8
-  %2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", !invariant.load !0
+  %1 = load i8*, i8** %x.addr, align 8
+  %2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", !invariant.load !0
   %call = call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %1, i8* %2)
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %3 = load i32* %i, align 4
+  %3 = load i32, i32* %i, align 4
   %inc = add i32 %3, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
diff --git a/test/Transforms/LICM/hoisting.ll b/test/Transforms/LICM/hoisting.ll
index b4d297ac27bf..8609407cc599 100644
--- a/test/Transforms/LICM/hoisting.ll
+++ b/test/Transforms/LICM/hoisting.ll
@@ -8,7 +8,7 @@ declare void @foo()
 ; potentially trapping instructions when they are not guaranteed to execute.
 define i32 @test1(i1 %c) {
 ; CHECK-LABEL: @test1(
-	%A = load i32* @X		; <i32> [#uses=2]
+	%A = load i32, i32* @X		; <i32> [#uses=2]
 	br label %Loop
 Loop:		; preds = %LoopTail, %0
 	call void @foo( )
@@ -35,9 +35,9 @@ declare void @foo2(i32) nounwind
 ;; It is ok and desirable to hoist this potentially trapping instruction.
 define i32 @test2(i1 %c) {
 ; CHECK-LABEL: @test2(
-; CHECK-NEXT: load i32* @X
+; CHECK-NEXT: load i32, i32* @X
 ; CHECK-NEXT: %B = sdiv i32 4, %A
-	%A = load i32* @X		; <i32> [#uses=2]
+	%A = load i32, i32* @X		; <i32> [#uses=2]
 	br label %Loop
 Loop:
         ;; Should have hoisted this div!
@@ -54,7 +54,7 @@ Out:		; preds = %Loop
 define i32 @test3(i1 %c) {
 ; CHECK-LABEL: define i32 @test3(
 ; CHECK: call void @foo2(i32 6)
-	%A = load i32* @X		; <i32> [#uses=2]
+	%A = load i32, i32* @X		; <i32> [#uses=2]
 	br label %Loop
 Loop:
 	%B = add i32 4, 2		; <i32> [#uses=2]
diff --git a/test/Transforms/LICM/lcssa-ssa-promoter.ll b/test/Transforms/LICM/lcssa-ssa-promoter.ll
index 5df3ef12181b..b0cae8772f3e 100644
--- a/test/Transforms/LICM/lcssa-ssa-promoter.ll
+++ b/test/Transforms/LICM/lcssa-ssa-promoter.ll
@@ -44,7 +44,7 @@ inner.body.rhs:
 ; CHECK-NEXT: br label %inner.latch
 
 inner.latch:
-  %y_val = load i32* @y, align 4
+  %y_val = load i32, i32* @y, align 4
   %icmp = icmp eq i32 %y_val, 0
   br i1 %icmp, label %inner.exit, label %inner.header
 ; CHECK: inner.latch:
diff --git a/test/Transforms/LICM/pr23608.ll b/test/Transforms/LICM/pr23608.ll
new file mode 100644
index 000000000000..249bc6bf5f63
--- /dev/null
+++ b/test/Transforms/LICM/pr23608.ll
@@ -0,0 +1,50 @@
+; RUN: opt -S -licm %s | FileCheck %s
+; ModuleID = '../pr23608.ll'
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+%struct.PyFrameObject = type { i32 }
+
+@a = common global %struct.PyFrameObject* null, align 8
+@__msan_origin_tls = external thread_local(initialexec) global i32
+
+define void @fn1() {
+entry:
+  br label %indirectgoto
+
+while.cond:                                       ; preds = %indirectgoto, %bb15
+  %tmp = load %struct.PyFrameObject*, %struct.PyFrameObject** @a, align 8
+  %_msld = load i64, i64* inttoptr (i64 and (i64 ptrtoint (%struct.PyFrameObject** @a to i64), i64 -70368744177665) to i64*), align 8
+  %tmp1 = load i32, i32* inttoptr (i64 add (i64 and (i64 ptrtoint (%struct.PyFrameObject** @a to i64), i64 -70368744177665), i64 35184372088832) to i32*), align 8
+  %f_iblock = getelementptr inbounds %struct.PyFrameObject, %struct.PyFrameObject* %tmp, i64 0, i32 0
+  br label %bb2
+
+bb:                                               ; preds = %while.cond
+  call void @__msan_warning_noreturn()
+  unreachable
+
+bb2:                                              ; preds = %while.cond
+  %tmp3 = load i32, i32* %f_iblock, align 4
+  %tmp4 = ptrtoint i32* %f_iblock to i64
+  %tmp8 = inttoptr i64 %tmp4 to i32*
+  %tobool = icmp eq i64 %tmp4, 0
+  br i1 %tobool, label %bb13, label %bb15
+
+bb13:                                             ; preds = %bb2
+; CHECK-LABEL bb13:
+; CHECK: %tmp8.le = inttoptr
+  %.lcssa7 = phi i32* [ %tmp8, %bb2 ]
+  call void @__msan_warning_noreturn()
+  unreachable
+
+bb15:                                             ; preds = %bb2
+  br i1 %tobool, label %while.end, label %while.cond
+
+while.end:                                        ; preds = %bb15
+  ret void
+
+indirectgoto:                                     ; preds = %indirectgoto, %entry
+  indirectbr i8* null, [label %indirectgoto, label %while.cond]
+}
+
+declare void @__msan_warning_noreturn()
diff --git a/test/Transforms/LICM/scalar-promote-memmodel.ll b/test/Transforms/LICM/scalar-promote-memmodel.ll
index 23d70f51351a..3603c25ba23c 100644
--- a/test/Transforms/LICM/scalar-promote-memmodel.ll
+++ b/test/Transforms/LICM/scalar-promote-memmodel.ll
@@ -19,12 +19,12 @@ for.body:                                         ; preds = %for.cond
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
-  %tmp3 = load i32* @g, align 4
+  %tmp3 = load i32, i32* @g, align 4
   %inc = add nsw i32 %tmp3, 1
   store i32 %inc, i32* @g, align 4
   br label %for.inc
 
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NEXT: add
 ; CHECK-NEXT: store i32
 
diff --git a/test/Transforms/LICM/scalar_promote.ll b/test/Transforms/LICM/scalar_promote.ll
index 80afb3c01e9a..6ef4bac39bbc 100644
--- a/test/Transforms/LICM/scalar_promote.ll
+++ b/test/Transforms/LICM/scalar_promote.ll
@@ -8,13 +8,13 @@ Entry:
   br label %Loop
 ; CHECK-LABEL: @test1(
 ; CHECK: Entry:
-; CHECK-NEXT:   load i32* @X
+; CHECK-NEXT:   load i32, i32* @X
 ; CHECK-NEXT:   br label %Loop
 
 
 Loop:   ; preds = %Loop, %0
   %j = phi i32 [ 0, %Entry ], [ %Next, %Loop ]    ; <i32> [#uses=1]
-  %x = load i32* @X   ; <i32> [#uses=1]
+  %x = load i32, i32* @X   ; <i32> [#uses=1]
   %x2 = add i32 %x, 1   ; <i32> [#uses=1]
   store i32 %x2, i32* @X
   %Next = add i32 %j, 1   ; <i32> [#uses=2]
@@ -35,14 +35,14 @@ Entry:
   br label %Loop
 ; CHECK-LABEL: @test2(
 ; CHECK: Entry:
-; CHECK-NEXT:    %.promoted = load i32* getelementptr inbounds (i32* @X, i64 1)
+; CHECK-NEXT:    %.promoted = load i32, i32* getelementptr inbounds (i32, i32* @X, i64 1)
 ; CHECK-NEXT:    br label %Loop
 
 Loop:   ; preds = %Loop, %0
-  %X1 = getelementptr i32* @X, i64 1    ; <i32*> [#uses=1]
-  %A = load i32* %X1    ; <i32> [#uses=1]
+  %X1 = getelementptr i32, i32* @X, i64 1    ; <i32*> [#uses=1]
+  %A = load i32, i32* %X1    ; <i32> [#uses=1]
   %V = add i32 %A, 1    ; <i32> [#uses=1]
-  %X2 = getelementptr i32* @X, i64 1    ; <i32*> [#uses=1]
+  %X2 = getelementptr i32, i32* @X, i64 1    ; <i32*> [#uses=1]
   store i32 %V, i32* %X2
   br i1 false, label %Loop, label %Exit
 
@@ -50,7 +50,7 @@ Exit:   ; preds = %Loop
   ret void
 ; CHECK: Exit:
 ; CHECK-NEXT:   %[[LCSSAPHI:.*]] = phi i32 [ %V
-; CHECK-NEXT:   store i32 %[[LCSSAPHI]], i32* getelementptr inbounds (i32* @X, i64 1)
+; CHECK-NEXT:   store i32 %[[LCSSAPHI]], i32* getelementptr inbounds (i32, i32* @X, i64 1)
 ; CHECK-NEXT:   ret void
 }
 
@@ -61,7 +61,7 @@ define void @test3(i32 %i) {
   br label %Loop
 Loop:
         ; Should not promote this to a register
-  %x = load volatile i32* @X
+  %x = load volatile i32, i32* @X
   %x2 = add i32 %x, 1
   store i32 %x2, i32* @X
   br i1 true, label %Out, label %Loop
@@ -82,15 +82,15 @@ define void @test4(i8* %x, i8 %n) {
   br label %loop
 
 loop:
-  %tmp = getelementptr i8* %x, i64 8
+  %tmp = getelementptr i8, i8* %x, i64 8
   store i8* %tmp, i8** %handle2
   br label %subloop
 
 subloop:
   %count = phi i8 [ 0, %loop ], [ %nextcount, %subloop ]
-  %offsetx2 = load i8** %handle2
+  %offsetx2 = load i8*, i8** %handle2
   store i8 %n, i8* %offsetx2
-  %newoffsetx2 = getelementptr i8* %offsetx2, i64 -1
+  %newoffsetx2 = getelementptr i8, i8* %offsetx2, i64 -1
   store i8* %newoffsetx2, i8** %handle2
   %nextcount = add i8 %count, 1
   %innerexitcond = icmp sge i8 %nextcount, 8
@@ -105,14 +105,14 @@ subloop:
 ; CHECK: br i1
 
 innerexit:
-  %offsetx1 = load i8** %handle1
-  %val = load i8* %offsetx1
+  %offsetx1 = load i8*, i8** %handle1
+  %val = load i8, i8* %offsetx1
   %cond = icmp eq i8 %val, %n
   br i1 %cond, label %exit, label %loop
 
 ; Should not have promoted offsetx1 loads.
 ; CHECK: innerexit:
-; CHECK: %val = load i8* %offsetx1
+; CHECK: %val = load i8, i8* %offsetx1
 ; CHECK: %cond = icmp eq i8 %val, %n
 ; CHECK: br i1 %cond, label %exit, label %loop
 
@@ -125,13 +125,13 @@ Entry:
   br label %Loop
 ; CHECK-LABEL: @test5(
 ; CHECK: Entry:
-; CHECK-NEXT:   load i32* @X
+; CHECK-NEXT:   load i32, i32* @X
 ; CHECK-NEXT:   br label %Loop
 
 
 Loop:   ; preds = %Loop, %0
   %j = phi i32 [ 0, %Entry ], [ %Next, %Loop ]    ; <i32> [#uses=1]
-  %x = load i32* @X   ; <i32> [#uses=1]
+  %x = load i32, i32* @X   ; <i32> [#uses=1]
   %x2 = add i32 %x, 1   ; <i32> [#uses=1]
   store i32 %x2, i32* @X
 
@@ -164,9 +164,9 @@ for.body.lr.ph:                                   ; preds = %entry
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %storemerge2 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %idxprom = sext i32 %storemerge2 to i64
-  %arrayidx = getelementptr inbounds float* %a, i64 %idxprom
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %idxprom
   store float 0.000000e+00, float* %arrayidx, align 4, !tbaa !3
-  %0 = load i32* %gi, align 4, !tbaa !0
+  %0 = load i32, i32* %gi, align 4, !tbaa !0
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %gi, align 4, !tbaa !0
   %cmp = icmp slt i32 %inc, %n
@@ -179,7 +179,7 @@ for.end:                                          ; preds = %for.cond.for.end_cr
   ret void
 
 ; CHECK: for.body.lr.ph:
-; CHECK-NEXT:  %gi.promoted = load i32* %gi, align 4, !tbaa !0
+; CHECK-NEXT:  %gi.promoted = load i32, i32* %gi, align 4, !tbaa !0
 ; CHECK: for.cond.for.end_crit_edge:
 ; CHECK-NEXT:  %[[LCSSAPHI:.*]] = phi i32 [ %inc
 ; CHECK-NEXT:  store i32 %[[LCSSAPHI]], i32* %gi, align 4, !tbaa !0
diff --git a/test/Transforms/LICM/sinking.ll b/test/Transforms/LICM/sinking.ll
index d7a8fcdafcad..02bf5846a64c 100644
--- a/test/Transforms/LICM/sinking.ll
+++ b/test/Transforms/LICM/sinking.ll
@@ -93,7 +93,7 @@ Entry:
 	br label %Loop
 Loop:		; preds = %Loop, %Entry
 	%N_addr.0.pn = phi i32 [ %dec, %Loop ], [ %N, %Entry ]	
-	%tmp.6 = load i32* @X		; <i32> [#uses=1]
+	%tmp.6 = load i32, i32* @X		; <i32> [#uses=1]
 	%dec = add i32 %N_addr.0.pn, -1		; <i32> [#uses=1]
 	%tmp.1 = icmp ne i32 %N_addr.0.pn, 1		; <i1> [#uses=1]
 	br i1 %tmp.1, label %Loop, label %Out
@@ -101,7 +101,7 @@ Out:		; preds = %Loop
 	ret i32 %tmp.6
 ; CHECK-LABEL: @test5(
 ; CHECK:     Out:
-; CHECK-NEXT:  %tmp.6.le = load i32* @X
+; CHECK-NEXT:  %tmp.6.le = load i32, i32* @X
 ; CHECK-NEXT:  ret i32 %tmp.6.le
 }
 
@@ -118,15 +118,15 @@ Out:		; preds = %Loop
 define i32 @test6() {
 	br label %Loop
 Loop:
-	%dead = getelementptr %Ty* @X2, i64 0, i32 0
-	%sunk2 = load i32* %dead
+	%dead = getelementptr %Ty, %Ty* @X2, i64 0, i32 0
+	%sunk2 = load i32, i32* %dead
 	br i1 false, label %Loop, label %Out
 Out:		; preds = %Loop
 	ret i32 %sunk2
 ; CHECK-LABEL: @test6(
 ; CHECK:     Out:
-; CHECK-NEXT:  %dead.le = getelementptr %Ty* @X2, i64 0, i32 0
-; CHECK-NEXT:  %sunk2.le = load i32* %dead.le
+; CHECK-NEXT:  %dead.le = getelementptr %Ty, %Ty* @X2, i64 0, i32 0
+; CHECK-NEXT:  %sunk2.le = load i32, i32* %dead.le
 ; CHECK-NEXT:  ret i32 %sunk2.le
 }
 
@@ -174,7 +174,7 @@ Entry:
 Loop:		; preds = %Cont, %Entry
 	br i1 %C1, label %Cont, label %exit1
 Cont:		; preds = %Loop
-	%X = load i32* %P		; <i32> [#uses=2]
+	%X = load i32, i32* %P		; <i32> [#uses=2]
 	store i32 %X, i32* %Q
 	%V = add i32 %X, 1		; <i32> [#uses=1]
 	br i1 %C2, label %Loop, label %exit2
@@ -242,7 +242,7 @@ Out:		; preds = %Loop
 define void @test11() {
 	br label %Loop
 Loop:
-	%dead = getelementptr %Ty* @X2, i64 0, i32 0
+	%dead = getelementptr %Ty, %Ty* @X2, i64 0, i32 0
 	br i1 false, label %Loop, label %Out
 Out:
 	ret void
@@ -261,36 +261,36 @@ entry:
 
 l1.header:
   %iv = phi i64 [ %iv.next, %l1.latch ], [ 0, %entry ]
-  %arrayidx.i = getelementptr inbounds [1 x i32]* @c, i64 0, i64 %iv
+  %arrayidx.i = getelementptr inbounds [1 x i32], [1 x i32]* @c, i64 0, i64 %iv
   br label %l2.header
 
 l2.header:
-  %x0 = load i1* %c, align 4
+  %x0 = load i1, i1* %c, align 4
   br i1 %x0, label %l1.latch, label %l3.preheader
 
 l3.preheader:
   br label %l3.header
 
 l3.header:
-  %x1 = load i1* %d, align 4
+  %x1 = load i1, i1* %d, align 4
   br i1 %x1, label %l2.latch, label %l4.preheader
 
 l4.preheader:
   br label %l4.header
 
 l4.header:
-  %x2 = load i1* %a
+  %x2 = load i1, i1* %a
   br i1 %x2, label %l3.latch, label %l4.body
 
 l4.body:
   call void @f(i32* %arrayidx.i)
-  %x3 = load i1* %b
+  %x3 = load i1, i1* %b
   %l = trunc i64 %iv to i32
   br i1 %x3, label %l4.latch, label %exit
 
 l4.latch:
   call void @g()
-  %x4 = load i1* %b, align 4
+  %x4 = load i1, i1* %b, align 4
   br i1 %x4, label %l4.header, label %exit
 
 l3.latch:
diff --git a/test/Transforms/LICM/speculate.ll b/test/Transforms/LICM/speculate.ll
index 69266693c479..91b5a25ac0f5 100644
--- a/test/Transforms/LICM/speculate.ll
+++ b/test/Transforms/LICM/speculate.ll
@@ -12,14 +12,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %i.02 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %p, i64 %i.02
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 %i.02
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
   %div = udiv i64 %x, 2
-  %arrayidx1 = getelementptr inbounds i64* %q, i64 %i.02
+  %arrayidx1 = getelementptr inbounds i64, i64* %q, i64 %i.02
   store i64 %div, i64* %arrayidx1, align 8
   br label %for.inc
 
@@ -44,14 +44,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %i.02 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %p, i64 %i.02
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 %i.02
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
   %div = udiv i64 %x, %m
-  %arrayidx1 = getelementptr inbounds i64* %q, i64 %i.02
+  %arrayidx1 = getelementptr inbounds i64, i64* %q, i64 %i.02
   store i64 %div, i64* %arrayidx1, align 8
   br label %for.inc
 
@@ -78,14 +78,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %i.02 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %p, i64 %i.02
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 %i.02
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
   %div = sdiv i64 %x, 2
-  %arrayidx1 = getelementptr inbounds i64* %q, i64 %i.02
+  %arrayidx1 = getelementptr inbounds i64, i64* %q, i64 %i.02
   store i64 %div, i64* %arrayidx1, align 8
   br label %for.inc
 
@@ -111,14 +111,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %i.02 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %p, i64 %i.02
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 %i.02
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
   %div = sdiv i64 %x, %or
-  %arrayidx1 = getelementptr inbounds i64* %q, i64 %i.02
+  %arrayidx1 = getelementptr inbounds i64, i64* %q, i64 %i.02
   store i64 %div, i64* %arrayidx1, align 8
   br label %for.inc
 
@@ -144,14 +144,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.inc
   %i.02 = phi i64 [ %inc, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %p, i64 %i.02
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 %i.02
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %for.inc, label %if.then
 
 if.then:                                          ; preds = %for.body
   %div = sdiv i64 %x, %and
-  %arrayidx1 = getelementptr inbounds i64* %q, i64 %i.02
+  %arrayidx1 = getelementptr inbounds i64, i64* %q, i64 %i.02
   store i64 %div, i64* %arrayidx1, align 8
   br label %for.inc
 
diff --git a/test/Transforms/LICM/volatile-alias.ll b/test/Transforms/LICM/volatile-alias.ll
index df7f0a931eb8..fda930df933b 100644
--- a/test/Transforms/LICM/volatile-alias.ll
+++ b/test/Transforms/LICM/volatile-alias.ll
@@ -2,9 +2,9 @@
 ; The objects *p and *q are aliased to each other, but even though *q is
 ; volatile, *p can be considered invariant in the loop. Check if it is moved
 ; out of the loop.
-; CHECK: load i32* %p
+; CHECK: load i32, i32* %p
 ; CHECK: for.body:
-; CHECK: load volatile i32* %q
+; CHECK: load volatile i32, i32* %q
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
 
@@ -24,30 +24,30 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
-  %1 = load i32* %n.addr, align 4
+  %0 = load i32, i32* %i, align 4
+  %1 = load i32, i32* %n.addr, align 4
   %cmp = icmp slt i32 %0, %1
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %2 = load i32** %p.addr, align 8
-  %3 = load i32* %2, align 4
-  %4 = load i32** %q.addr, align 8
-  %5 = load volatile i32* %4, align 4
+  %2 = load i32*, i32** %p.addr, align 8
+  %3 = load i32, i32* %2, align 4
+  %4 = load i32*, i32** %q.addr, align 8
+  %5 = load volatile i32, i32* %4, align 4
   %add = add nsw i32 %3, %5
-  %6 = load i32* %s, align 4
+  %6 = load i32, i32* %s, align 4
   %add1 = add nsw i32 %6, %add
   store i32 %add1, i32* %s, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32* %s, align 4
+  %8 = load i32, i32* %s, align 4
   ret i32 %8
 }
 
diff --git a/test/Transforms/LoadCombine/load-combine-aa.ll b/test/Transforms/LoadCombine/load-combine-aa.ll
index 3542dcebf5e4..fc639c0bc05d 100644
--- a/test/Transforms/LoadCombine/load-combine-aa.ll
+++ b/test/Transforms/LoadCombine/load-combine-aa.ll
@@ -5,14 +5,14 @@ target triple = "x86_64-unknown-linux-gnu"
 define i64 @test1(i32* nocapture readonly noalias %a, i32* nocapture readonly noalias %b) {
 ; CHECK-LABEL: @test1
 
-; CHECK: load i64*
+; CHECK: load i64, i64*
 ; CHECK: ret i64
 
-  %load1 = load i32* %a, align 4
+  %load1 = load i32, i32* %a, align 4
   %conv = zext i32 %load1 to i64
-  %arrayidx1 = getelementptr inbounds i32* %a, i64 1
+  %arrayidx1 = getelementptr inbounds i32, i32* %a, i64 1
   store i32 %load1, i32* %b, align 4
-  %load2 = load i32* %arrayidx1, align 4
+  %load2 = load i32, i32* %arrayidx1, align 4
   %conv2 = zext i32 %load2 to i64
   %shl = shl nuw i64 %conv2, 32
   %add = or i64 %shl, %conv
@@ -22,15 +22,15 @@ define i64 @test1(i32* nocapture readonly noalias %a, i32* nocapture readonly no
 define i64 @test2(i32* nocapture readonly %a, i32* nocapture readonly %b) {
 ; CHECK-LABEL: @test2
 
-; CHECK: load i32*
-; CHECK: load i32*
+; CHECK: load i32, i32*
+; CHECK: load i32, i32*
 ; CHECK: ret i64
 
-  %load1 = load i32* %a, align 4
+  %load1 = load i32, i32* %a, align 4
   %conv = zext i32 %load1 to i64
-  %arrayidx1 = getelementptr inbounds i32* %a, i64 1
+  %arrayidx1 = getelementptr inbounds i32, i32* %a, i64 1
   store i32 %load1, i32* %b, align 4
-  %load2 = load i32* %arrayidx1, align 4
+  %load2 = load i32, i32* %arrayidx1, align 4
   %conv2 = zext i32 %load2 to i64
   %shl = shl nuw i64 %conv2, 32
   %add = or i64 %shl, %conv
diff --git a/test/Transforms/LoadCombine/load-combine-assume.ll b/test/Transforms/LoadCombine/load-combine-assume.ll
index 94f630072add..2d6d160f12fe 100644
--- a/test/Transforms/LoadCombine/load-combine-assume.ll
+++ b/test/Transforms/LoadCombine/load-combine-assume.ll
@@ -8,14 +8,14 @@ declare void @llvm.assume(i1) nounwind
 define i64 @test1(i32* nocapture readonly %a, i1 %b) {
 ; CHECK-LABEL: @test1
 
-; CHECK-DAG: load i64* %1, align 4
+; CHECK-DAG: load i64, i64* %1, align 4
 ; CHECK-DAG: tail call void @llvm.assume(i1 %b)
 ; CHECK: ret i64
 
-  %load1 = load i32* %a, align 4
+  %load1 = load i32, i32* %a, align 4
   %conv = zext i32 %load1 to i64
-  %arrayidx1 = getelementptr inbounds i32* %a, i64 1
-  %load2 = load i32* %arrayidx1, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %a, i64 1
+  %load2 = load i32, i32* %arrayidx1, align 4
   tail call void @llvm.assume(i1 %b)
   %conv2 = zext i32 %load2 to i64
   %shl = shl nuw i64 %conv2, 32
@@ -27,15 +27,15 @@ define i64 @test1(i32* nocapture readonly %a, i1 %b) {
 define i64 @test2(i32* nocapture readonly %a, i1 %b) {
 ; CHECK-LABEL: @test2
 
-; CHECK-DAG: load i64* %1, align 4
+; CHECK-DAG: load i64, i64* %1, align 4
 ; CHECK-DAG: tail call void @llvm.assume(i1 %b)
 ; CHECK: ret i64
 
-  %load1 = load i32* %a, align 4
+  %load1 = load i32, i32* %a, align 4
   %conv = zext i32 %load1 to i64
-  %arrayidx1 = getelementptr inbounds i32* %a, i64 1
+  %arrayidx1 = getelementptr inbounds i32, i32* %a, i64 1
   tail call void @llvm.assume(i1 %b)
-  %load2 = load i32* %arrayidx1, align 4
+  %load2 = load i32, i32* %arrayidx1, align 4
   %conv2 = zext i32 %load2 to i64
   %shl = shl nuw i64 %conv2, 32
   %add = or i64 %shl, %conv
diff --git a/test/Transforms/LoadCombine/load-combine.ll b/test/Transforms/LoadCombine/load-combine.ll
index c4d9241764d9..d5068787639f 100644
--- a/test/Transforms/LoadCombine/load-combine.ll
+++ b/test/Transforms/LoadCombine/load-combine.ll
@@ -6,138 +6,138 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Combine read from char* idiom.
 define i64 @LoadU64_x64_0(i64* %pData) {
   %1 = bitcast i64* %pData to i8*
-  %2 = load i8* %1, align 1
+  %2 = load i8, i8* %1, align 1
   %3 = zext i8 %2 to i64
   %4 = shl nuw i64 %3, 56
-  %5 = getelementptr inbounds i8* %1, i64 1
-  %6 = load i8* %5, align 1
+  %5 = getelementptr inbounds i8, i8* %1, i64 1
+  %6 = load i8, i8* %5, align 1
   %7 = zext i8 %6 to i64
   %8 = shl nuw nsw i64 %7, 48
   %9 = or i64 %8, %4
-  %10 = getelementptr inbounds i8* %1, i64 2
-  %11 = load i8* %10, align 1
+  %10 = getelementptr inbounds i8, i8* %1, i64 2
+  %11 = load i8, i8* %10, align 1
   %12 = zext i8 %11 to i64
   %13 = shl nuw nsw i64 %12, 40
   %14 = or i64 %9, %13
-  %15 = getelementptr inbounds i8* %1, i64 3
-  %16 = load i8* %15, align 1
+  %15 = getelementptr inbounds i8, i8* %1, i64 3
+  %16 = load i8, i8* %15, align 1
   %17 = zext i8 %16 to i64
   %18 = shl nuw nsw i64 %17, 32
   %19 = or i64 %14, %18
-  %20 = getelementptr inbounds i8* %1, i64 4
-  %21 = load i8* %20, align 1
+  %20 = getelementptr inbounds i8, i8* %1, i64 4
+  %21 = load i8, i8* %20, align 1
   %22 = zext i8 %21 to i64
   %23 = shl nuw nsw i64 %22, 24
   %24 = or i64 %19, %23
-  %25 = getelementptr inbounds i8* %1, i64 5
-  %26 = load i8* %25, align 1
+  %25 = getelementptr inbounds i8, i8* %1, i64 5
+  %26 = load i8, i8* %25, align 1
   %27 = zext i8 %26 to i64
   %28 = shl nuw nsw i64 %27, 16
   %29 = or i64 %24, %28
-  %30 = getelementptr inbounds i8* %1, i64 6
-  %31 = load i8* %30, align 1
+  %30 = getelementptr inbounds i8, i8* %1, i64 6
+  %31 = load i8, i8* %30, align 1
   %32 = zext i8 %31 to i64
   %33 = shl nuw nsw i64 %32, 8
   %34 = or i64 %29, %33
-  %35 = getelementptr inbounds i8* %1, i64 7
-  %36 = load i8* %35, align 1
+  %35 = getelementptr inbounds i8, i8* %1, i64 7
+  %36 = load i8, i8* %35, align 1
   %37 = zext i8 %36 to i64
   %38 = or i64 %34, %37
   ret i64 %38
 ; CHECK-LABEL: @LoadU64_x64_0(
-; CHECK: load i64* %{{.*}}, align 1
+; CHECK: load i64, i64* %{{.*}}, align 1
 ; CHECK-NOT: load
 }
 
 ; Combine simple adjacent loads.
 define i32 @"2xi16_i32"(i16* %x) {
-  %1 = load i16* %x, align 2
-  %2 = getelementptr inbounds i16* %x, i64 1
-  %3 = load i16* %2, align 2
+  %1 = load i16, i16* %x, align 2
+  %2 = getelementptr inbounds i16, i16* %x, i64 1
+  %3 = load i16, i16* %2, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw i32 %4, 16
   %6 = zext i16 %1 to i32
   %7 = or i32 %5, %6
   ret i32 %7
 ; CHECK-LABEL: @"2xi16_i32"(
-; CHECK: load i32* %{{.*}}, align 2
+; CHECK: load i32, i32* %{{.*}}, align 2
 ; CHECK-NOT: load
 }
 
 ; Don't combine loads across stores.
 define i32 @"2xi16_i32_store"(i16* %x, i16* %y) {
-  %1 = load i16* %x, align 2
+  %1 = load i16, i16* %x, align 2
   store i16 0, i16* %y, align 2
-  %2 = getelementptr inbounds i16* %x, i64 1
-  %3 = load i16* %2, align 2
+  %2 = getelementptr inbounds i16, i16* %x, i64 1
+  %3 = load i16, i16* %2, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw i32 %4, 16
   %6 = zext i16 %1 to i32
   %7 = or i32 %5, %6
   ret i32 %7
 ; CHECK-LABEL: @"2xi16_i32_store"(
-; CHECK: load i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
 ; CHECK: store
-; CHECK: load i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
 }
 
 ; Don't combine loads with a gap.
 define i32 @"2xi16_i32_gap"(i16* %x) {
-  %1 = load i16* %x, align 2
-  %2 = getelementptr inbounds i16* %x, i64 2
-  %3 = load i16* %2, align 2
+  %1 = load i16, i16* %x, align 2
+  %2 = getelementptr inbounds i16, i16* %x, i64 2
+  %3 = load i16, i16* %2, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw i32 %4, 16
   %6 = zext i16 %1 to i32
   %7 = or i32 %5, %6
   ret i32 %7
 ; CHECK-LABEL: @"2xi16_i32_gap"(
-; CHECK: load i16* %{{.*}}, align 2
-; CHECK: load i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
 }
 
 ; Combine out of order loads.
 define i32 @"2xi16_i32_order"(i16* %x) {
-  %1 = getelementptr inbounds i16* %x, i64 1
-  %2 = load i16* %1, align 2
+  %1 = getelementptr inbounds i16, i16* %x, i64 1
+  %2 = load i16, i16* %1, align 2
   %3 = zext i16 %2 to i32
-  %4 = load i16* %x, align 2
+  %4 = load i16, i16* %x, align 2
   %5 = shl nuw i32 %3, 16
   %6 = zext i16 %4 to i32
   %7 = or i32 %5, %6
   ret i32 %7
 ; CHECK-LABEL: @"2xi16_i32_order"(
-; CHECK: load i32* %{{.*}}, align 2
+; CHECK: load i32, i32* %{{.*}}, align 2
 ; CHECK-NOT: load
 }
 
 ; Overlapping loads.
 define i32 @"2xi16_i32_overlap"(i8* %x) {
   %1 = bitcast i8* %x to i16*
-  %2 = load i16* %1, align 2
-  %3 = getelementptr inbounds i8* %x, i64 1
+  %2 = load i16, i16* %1, align 2
+  %3 = getelementptr inbounds i8, i8* %x, i64 1
   %4 = bitcast i8* %3 to i16*
-  %5 = load i16* %4, align 2
+  %5 = load i16, i16* %4, align 2
   %6 = zext i16 %5 to i32
   %7 = shl nuw i32 %6, 16
   %8 = zext i16 %2 to i32
   %9 = or i32 %7, %8
   ret i32 %9
 ; CHECK-LABEL: @"2xi16_i32_overlap"(
-; CHECK: load i16* %{{.*}}, align 2
-; CHECK: load i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
+; CHECK: load i16, i16* %{{.*}}, align 2
 }
 
 ; Combine valid alignments.
 define i64 @"2xi16_i64_align"(i8* %x) {
   %1 = bitcast i8* %x to i32*
-  %2 = load i32* %1, align 4
-  %3 = getelementptr inbounds i8* %x, i64 4
+  %2 = load i32, i32* %1, align 4
+  %3 = getelementptr inbounds i8, i8* %x, i64 4
   %4 = bitcast i8* %3 to i16*
-  %5 = load i16* %4, align 2
-  %6 = getelementptr inbounds i8* %x, i64 6
+  %5 = load i16, i16* %4, align 2
+  %6 = getelementptr inbounds i8, i8* %x, i64 6
   %7 = bitcast i8* %6 to i16*
-  %8 = load i16* %7, align 2
+  %8 = load i16, i16* %7, align 2
   %9 = zext i16 %8 to i64
   %10 = shl nuw i64 %9, 48
   %11 = zext i16 %5 to i64
@@ -147,44 +147,44 @@ define i64 @"2xi16_i64_align"(i8* %x) {
   %15 = or i64 %14, %10
   ret i64 %15
 ; CHECK-LABEL: @"2xi16_i64_align"(
-; CHECK: load i64* %{{.*}}, align 4
+; CHECK: load i64, i64* %{{.*}}, align 4
 }
 
 ; Non power of two.
 define i64 @"2xi16_i64_npo2"(i8* %x) {
-  %1 = load i8* %x, align 1
+  %1 = load i8, i8* %x, align 1
   %2 = zext i8 %1 to i64
-  %3 = getelementptr inbounds i8* %x, i64 1
-  %4 = load i8* %3, align 1
+  %3 = getelementptr inbounds i8, i8* %x, i64 1
+  %4 = load i8, i8* %3, align 1
   %5 = zext i8 %4 to i64
   %6 = shl nuw nsw i64 %5, 8
   %7 = or i64 %6, %2
-  %8 = getelementptr inbounds i8* %x, i64 2
-  %9 = load i8* %8, align 1
+  %8 = getelementptr inbounds i8, i8* %x, i64 2
+  %9 = load i8, i8* %8, align 1
   %10 = zext i8 %9 to i64
   %11 = shl nuw nsw i64 %10, 16
   %12 = or i64 %11, %7
-  %13 = getelementptr inbounds i8* %x, i64 3
-  %14 = load i8* %13, align 1
+  %13 = getelementptr inbounds i8, i8* %x, i64 3
+  %14 = load i8, i8* %13, align 1
   %15 = zext i8 %14 to i64
   %16 = shl nuw nsw i64 %15, 24
   %17 = or i64 %16, %12
-  %18 = getelementptr inbounds i8* %x, i64 4
-  %19 = load i8* %18, align 1
+  %18 = getelementptr inbounds i8, i8* %x, i64 4
+  %19 = load i8, i8* %18, align 1
   %20 = zext i8 %19 to i64
   %21 = shl nuw nsw i64 %20, 32
   %22 = or i64 %21, %17
-  %23 = getelementptr inbounds i8* %x, i64 5
-  %24 = load i8* %23, align 1
+  %23 = getelementptr inbounds i8, i8* %x, i64 5
+  %24 = load i8, i8* %23, align 1
   %25 = zext i8 %24 to i64
   %26 = shl nuw nsw i64 %25, 40
   %27 = or i64 %26, %22
-  %28 = getelementptr inbounds i8* %x, i64 6
-  %29 = load i8* %28, align 1
+  %28 = getelementptr inbounds i8, i8* %x, i64 6
+  %29 = load i8, i8* %28, align 1
   %30 = zext i8 %29 to i64
   %31 = shl nuw nsw i64 %30, 48
   %32 = or i64 %31, %27
   ret i64 %32
 ; CHECK-LABEL: @"2xi16_i64_npo2"(
-; CHECK: load i32* %{{.*}}, align 1
+; CHECK: load i32, i32* %{{.*}}, align 1
 }
diff --git a/test/Transforms/LoopDeletion/2008-05-06-Phi.ll b/test/Transforms/LoopDeletion/2008-05-06-Phi.ll
index ff6a30d3a342..fcf5ede76b71 100644
--- a/test/Transforms/LoopDeletion/2008-05-06-Phi.ll
+++ b/test/Transforms/LoopDeletion/2008-05-06-Phi.ll
@@ -23,7 +23,7 @@ declare %struct.BF_PartHolder* @BF_addElement(%struct.BF_PartHolder*, %struct.BF
 define %struct.BF_PartHolder* @BF_addEntry(%struct.BF_PartHolder* %thePH, i32 %value, i32 %length) nounwind  {
 entry:
 	%myElement = alloca %struct.BF_BitstreamElement		; <%struct.BF_BitstreamElement*> [#uses=2]
-	%tmp1 = getelementptr %struct.BF_BitstreamElement* %myElement, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp1 = getelementptr %struct.BF_BitstreamElement, %struct.BF_BitstreamElement* %myElement, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 %value, i32* %tmp1, align 8
 	%tmp7 = icmp eq i32 %length, 0		; <i1> [#uses=1]
 	br i1 %tmp7, label %bb13, label %bb
@@ -44,15 +44,15 @@ entry:
 
 define internal fastcc void @encodeMainData(%struct.lame_global_flags* %gfp, [2 x [576 x i32]]* %l3_enc, %struct.III_side_info_t* %si, [2 x %struct.III_scalefac_t]* %scalefac) nounwind  {
 entry:
-	%tmp69 = getelementptr %struct.lame_global_flags* %gfp, i32 0, i32 43		; <i32*> [#uses=1]
-	%tmp70 = load i32* %tmp69, align 4		; <i32> [#uses=1]
+	%tmp69 = getelementptr %struct.lame_global_flags, %struct.lame_global_flags* %gfp, i32 0, i32 43		; <i32*> [#uses=1]
+	%tmp70 = load i32, i32* %tmp69, align 4		; <i32> [#uses=1]
 	%tmp71 = icmp eq i32 %tmp70, 1		; <i1> [#uses=1]
 	br i1 %tmp71, label %bb352, label %bb498
 
 bb113:		; preds = %bb132
-	%tmp123 = getelementptr [2 x %struct.III_scalefac_t]* %scalefac, i32 0, i32 0, i32 1, i32 %sfb.0, i32 %window.0		; <i32*> [#uses=1]
-	%tmp124 = load i32* %tmp123, align 4		; <i32> [#uses=1]
-	%tmp126 = load %struct.BF_PartHolder** %tmp80, align 4		; <%struct.BF_PartHolder*> [#uses=1]
+	%tmp123 = getelementptr [2 x %struct.III_scalefac_t], [2 x %struct.III_scalefac_t]* %scalefac, i32 0, i32 0, i32 1, i32 %sfb.0, i32 %window.0		; <i32*> [#uses=1]
+	%tmp124 = load i32, i32* %tmp123, align 4		; <i32> [#uses=1]
+	%tmp126 = load %struct.BF_PartHolder*, %struct.BF_PartHolder** %tmp80, align 4		; <%struct.BF_PartHolder*> [#uses=1]
 	%tmp128 = call %struct.BF_PartHolder* @BF_addEntry( %struct.BF_PartHolder* %tmp126, i32 %tmp124, i32 %tmp93 ) nounwind 		; <%struct.BF_PartHolder*> [#uses=1]
 	store %struct.BF_PartHolder* %tmp128, %struct.BF_PartHolder** %tmp80, align 4
 	%tmp131 = add i32 %window.0, 1		; <i32> [#uses=1]
@@ -73,7 +73,7 @@ bb140:		; preds = %bb341, %bb137
 	br i1 %tmp142, label %bb132, label %bb174
 
 bb166:		; preds = %bb174
-	%tmp160 = load %struct.BF_PartHolder** %tmp80, align 4		; <%struct.BF_PartHolder*> [#uses=1]
+	%tmp160 = load %struct.BF_PartHolder*, %struct.BF_PartHolder** %tmp80, align 4		; <%struct.BF_PartHolder*> [#uses=1]
 	%tmp162 = call %struct.BF_PartHolder* @BF_addEntry( %struct.BF_PartHolder* %tmp160, i32 0, i32 0 ) nounwind 		; <%struct.BF_PartHolder*> [#uses=0]
 	unreachable
 
@@ -82,9 +82,9 @@ bb174:		; preds = %bb140
 	br i1 %tmp176, label %bb166, label %bb341
 
 bb341:		; preds = %bb352, %bb174
-	%tmp80 = getelementptr [2 x [2 x %struct.BF_PartHolder*]]* @scaleFactorsPH, i32 0, i32 0, i32 0		; <%struct.BF_PartHolder**> [#uses=3]
-	%tmp92 = getelementptr [16 x i32]* @slen1_tab, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp93 = load i32* %tmp92, align 4		; <i32> [#uses=1]
+	%tmp80 = getelementptr [2 x [2 x %struct.BF_PartHolder*]], [2 x [2 x %struct.BF_PartHolder*]]* @scaleFactorsPH, i32 0, i32 0, i32 0		; <%struct.BF_PartHolder**> [#uses=3]
+	%tmp92 = getelementptr [16 x i32], [16 x i32]* @slen1_tab, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp93 = load i32, i32* %tmp92, align 4		; <i32> [#uses=1]
 	br label %bb140
 
 bb352:		; preds = %entry
diff --git a/test/Transforms/LoopDistribute/basic-with-memchecks.ll b/test/Transforms/LoopDistribute/basic-with-memchecks.ll
new file mode 100644
index 000000000000..4c1c1b822964
--- /dev/null
+++ b/test/Transforms/LoopDistribute/basic-with-memchecks.ll
@@ -0,0 +1,112 @@
+; RUN: opt -basicaa -loop-distribute -verify-loop-info -verify-dom-info -S \
+; RUN:   < %s | FileCheck %s
+
+; RUN: opt -basicaa -loop-distribute -loop-vectorize -force-vector-width=4 \
+; RUN:   -verify-loop-info -verify-dom-info -S < %s | \
+; RUN:   FileCheck --check-prefix=VECTORIZE %s
+
+; The memcheck version of basic.ll.  We should distribute and vectorize the
+; second part of this loop with 5 memchecks (A+1 x {C, D, E} + C x {A, B})
+;
+;   for (i = 0; i < n; i++) {
+;     A[i + 1] = A[i] * B[i];
+; -------------------------------
+;     C[i] = D[i] * E[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+@B = common global i32* null, align 8
+@A = common global i32* null, align 8
+@C = common global i32* null, align 8
+@D = common global i32* null, align 8
+@E = common global i32* null, align 8
+
+define void @f() {
+entry:
+  %a = load i32*, i32** @A, align 8
+  %b = load i32*, i32** @B, align 8
+  %c = load i32*, i32** @C, align 8
+  %d = load i32*, i32** @D, align 8
+  %e = load i32*, i32** @E, align 8
+  br label %for.body
+
+; We have two compares for each array overlap check which is a total of 10
+; compares.
+;
+; CHECK: for.body.ldist.memcheck:
+; CHECK:     = icmp
+; CHECK:     = icmp
+
+; CHECK:     = icmp
+; CHECK:     = icmp
+
+; CHECK:     = icmp
+; CHECK:     = icmp
+
+; CHECK:     = icmp
+; CHECK:     = icmp
+
+; CHECK:     = icmp
+; CHECK:     = icmp
+
+; CHECK-NOT: = icmp
+; CHECK:     br i1 %memcheck.conflict, label %for.body.ph.ldist.nondist, label %for.body.ph.ldist1
+
+; The non-distributed loop that the memchecks fall back on.
+
+; CHECK: for.body.ph.ldist.nondist:
+; CHECK:     br label %for.body.ldist.nondist
+; CHECK: for.body.ldist.nondist:
+; CHECK:    br i1 %exitcond.ldist.nondist, label %for.end, label %for.body.ldist.nondist
+
+; Verify the two distributed loops.
+
+; CHECK: for.body.ph.ldist1:
+; CHECK:     br label %for.body.ldist1
+; CHECK: for.body.ldist1:
+; CHECK:    %mulA.ldist1 = mul i32 %loadB.ldist1, %loadA.ldist1
+; CHECK:    br i1 %exitcond.ldist1, label %for.body.ph, label %for.body.ldist1
+
+; CHECK: for.body.ph:
+; CHECK:    br label %for.body
+; CHECK: for.body:
+; CHECK:    %mulC = mul i32 %loadD, %loadE
+; CHECK: for.end:
+
+
+; VECTORIZE: mul <4 x i32>
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind
+  %loadD = load i32, i32* %arrayidxD, align 4
+
+  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind
+  %loadE = load i32, i32* %arrayidxE, align 4
+
+  %mulC = mul i32 %loadD, %loadE
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+  store i32 %mulC, i32* %arrayidxC, align 4
+
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
diff --git a/test/Transforms/LoopDistribute/basic.ll b/test/Transforms/LoopDistribute/basic.ll
new file mode 100644
index 000000000000..f19fa8bb1b14
--- /dev/null
+++ b/test/Transforms/LoopDistribute/basic.ll
@@ -0,0 +1,83 @@
+; RUN: opt -basicaa -loop-distribute -verify-loop-info -verify-dom-info -S \
+; RUN:   < %s | FileCheck %s
+
+; RUN: opt -basicaa -loop-distribute -verify-loop-info -verify-dom-info \
+; RUN:   -loop-accesses -analyze < %s | FileCheck %s --check-prefix=ANALYSIS
+
+; RUN: opt -basicaa -loop-distribute -loop-vectorize -force-vector-width=4 -S \
+; RUN:   < %s | FileCheck %s --check-prefix=VECTORIZE
+
+; We should distribute this loop into a safe (2nd statement) and unsafe loop
+; (1st statement):
+;   for (i = 0; i < n; i++) {
+;     A[i + 1] = A[i] * B[i];
+;     =======================
+;     C[i] = D[i] * E[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+define void @f(i32* noalias %a,
+               i32* noalias %b,
+               i32* noalias %c,
+               i32* noalias %d,
+               i32* noalias %e) {
+entry:
+  br label %for.body
+
+; Verify the two distributed loops.
+
+; CHECK: entry.split.ldist1:
+; CHECK:    br label %for.body.ldist1
+; CHECK: for.body.ldist1:
+; CHECK:    %mulA.ldist1 = mul i32 %loadB.ldist1, %loadA.ldist1
+; CHECK:    br i1 %exitcond.ldist1, label %entry.split, label %for.body.ldist1
+
+; CHECK: entry.split:
+; CHECK:    br label %for.body
+; CHECK: for.body:
+; CHECK:    %mulC = mul i32 %loadD, %loadE
+; CHECK: for.end:
+
+
+; ANALYSIS: for.body:
+; ANALYSIS-NEXT: Memory dependences are safe{{$}}
+; ANALYSIS: for.body.ldist1:
+; ANALYSIS-NEXT: Report: unsafe dependent memory operations in loop
+
+
+; VECTORIZE: mul <4 x i32>
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind
+  %loadD = load i32, i32* %arrayidxD, align 4
+
+  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind
+  %loadE = load i32, i32* %arrayidxE, align 4
+
+  %mulC = mul i32 %loadD, %loadE
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+  store i32 %mulC, i32* %arrayidxC, align 4
+
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
diff --git a/test/Transforms/LoopDistribute/crash-in-memcheck-generation.ll b/test/Transforms/LoopDistribute/crash-in-memcheck-generation.ll
new file mode 100644
index 000000000000..c81ddf59fa7c
--- /dev/null
+++ b/test/Transforms/LoopDistribute/crash-in-memcheck-generation.ll
@@ -0,0 +1,59 @@
+; RUN: opt -basicaa -loop-distribute -loop-vectorize -force-vector-width=4 \
+; RUN:   -verify-loop-info -verify-dom-info -S < %s | FileCheck %s
+
+; If only A and B can alias here, we don't need memchecks to distribute since
+; A and B are in the same partition.  This used to cause a crash in the
+; memcheck generation.
+;
+;   for (i = 0; i < n; i++) {
+;     A[i + 1] = A[i] * B[i];
+; ------------------------------
+;     C[i] = D[i] * E[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+define void @f(i32*  %a,
+               i32*  %b,
+               i32* noalias %c,
+               i32* noalias %d,
+               i32* noalias %e) {
+entry:
+  br label %for.body
+
+; CHECK-NOT: memcheck:
+; CHECK: mul <4 x i32>
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind
+  %loadD = load i32, i32* %arrayidxD, align 4
+
+  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind
+  %loadE = load i32, i32* %arrayidxE, align 4
+
+  %mulC = mul i32 %loadD, %loadE
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+  store i32 %mulC, i32* %arrayidxC, align 4
+
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
diff --git a/test/Transforms/LoopDistribute/no-if-convert.ll b/test/Transforms/LoopDistribute/no-if-convert.ll
new file mode 100644
index 000000000000..fcd8b65731cc
--- /dev/null
+++ b/test/Transforms/LoopDistribute/no-if-convert.ll
@@ -0,0 +1,95 @@
+; RUN: opt -basicaa -loop-distribute -verify-loop-info -verify-dom-info -S < %s \
+; RUN:   | FileCheck %s
+
+; We should distribute this loop along === but not along ---.  The last
+; partition won't be vectorized due to conditional stores so it's better to
+; keep it with the second partition which has a dependence cycle.
+
+; (1st statement):
+;   for (i = 0; i < n; i++) {
+;     C[i] = D[i] * E[i];
+;=============================
+;     A[i + 1] = A[i] * B[i];
+;-----------------------------
+;     if (F[i])
+;        G[i] = H[i] * J[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+define void @f(i32* noalias %a,
+               i32* noalias %b,
+               i32* noalias %c,
+               i32* noalias %d,
+               i32* noalias %e,
+               i32* noalias %g,
+               i32* noalias %h,
+               i32* noalias %j,
+               i64 %x) {
+entry:
+  br label %for.body
+
+; Ensure that we have only two partitions, the first with one multiplication
+; and the second with two.
+
+; CHECK: for.body.ldist1:
+; CHECK:    %mulC.ldist1 = mul i32 %loadD.ldist1, %loadE.ldist1
+; CHECK:    br i1 %exitcond.ldist1, label %entry.split, label %for.body.ldist1
+; CHECK: entry.split:
+; CHECK:    br label %for.body
+; CHECK: for.body:
+; CHECK:    %mulA = mul i32 %loadB, %loadA
+; CHECK:    %mulG = mul i32 %loadH, %loadJ
+; CHECK: for.end:
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %if.end ]
+
+  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind
+  %loadD = load i32, i32* %arrayidxD, align 4
+
+  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind
+  %loadE = load i32, i32* %arrayidxE, align 4
+
+  %mulC = mul i32 %loadD, %loadE
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+  store i32 %mulC, i32* %arrayidxC, align 4
+
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %if.cond = icmp eq i64 %ind, %x
+  br i1 %if.cond, label %if.then, label %if.end
+
+if.then:
+  %arrayidxH = getelementptr inbounds i32, i32* %h, i64 %ind
+  %loadH = load i32, i32* %arrayidxH, align 4
+
+  %arrayidxJ = getelementptr inbounds i32, i32* %j, i64 %ind
+  %loadJ = load i32, i32* %arrayidxJ, align 4
+
+  %mulG = mul i32 %loadH, %loadJ
+
+  %arrayidxG = getelementptr inbounds i32, i32* %g, i64 %ind
+  store i32 %mulG, i32* %arrayidxG, align 4
+  br label %if.end
+
+if.end:
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
diff --git a/test/Transforms/LoopDistribute/outside-use.ll b/test/Transforms/LoopDistribute/outside-use.ll
new file mode 100644
index 000000000000..546050d5b046
--- /dev/null
+++ b/test/Transforms/LoopDistribute/outside-use.ll
@@ -0,0 +1,69 @@
+; RUN: opt -loop-distribute -verify-loop-info -verify-dom-info -S < %s \
+; RUN:   | FileCheck %s
+
+; Check that definitions used outside the loop are handled correctly: (1) they
+; are not dropped (2) when version the loop, a phi is added to merge the value
+; from the non-distributed loop and the distributed loop.
+;
+;   for (i = 0; i < n; i++) {
+;     A[i + 1] = A[i] * B[i];
+;   ==========================
+;     sum += C[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+@B = common global i32* null, align 8
+@A = common global i32* null, align 8
+@C = common global i32* null, align 8
+@D = common global i32* null, align 8
+@E = common global i32* null, align 8
+@SUM = common global i32 0, align 8
+
+define void @f() {
+entry:
+  %a = load i32*, i32** @A, align 8
+  %b = load i32*, i32** @B, align 8
+  %c = load i32*, i32** @C, align 8
+  %d = load i32*, i32** @D, align 8
+  %e = load i32*, i32** @E, align 8
+
+  br label %for.body
+
+; CHECK: for.body.ldist1:
+; CHECK:   %mulA.ldist1 = mul i32 %loadB.ldist1, %loadA.ldist1
+; CHECK: for.body.ph:
+; CHECK: for.body:
+; CHECK:   %sum_add = add nuw nsw i32 %sum, %loadC
+; CHECK: for.end:
+; CHECK:   %sum_add.ldist = phi i32 [ %sum_add, %for.body ], [ %sum_add.ldist.nondist, %for.body.ldist.nondist ]
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+  %sum = phi i32 [ 0, %entry ], [ %sum_add, %for.body ]
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+  %loadC = load i32, i32* %arrayidxC, align 4
+
+  %sum_add = add nuw nsw i32 %sum, %loadC
+
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  store i32 %sum_add, i32* @SUM, align 4
+  ret void
+}
diff --git a/test/Transforms/LoopDistribute/program-order.ll b/test/Transforms/LoopDistribute/program-order.ll
new file mode 100644
index 000000000000..b534b79af71a
--- /dev/null
+++ b/test/Transforms/LoopDistribute/program-order.ll
@@ -0,0 +1,65 @@
+; RUN: opt -loop-distribute -S -verify-loop-info -verify-dom-info < %s \
+; RUN:   | FileCheck %s
+
+; Distributing this loop to avoid the dependence cycle would require to
+; reorder S1 and S2 to form the two partitions: {S2} | {S1, S3}.  The analysis
+; provided by LoopAccessAnalysis does not allow us to reorder memory
+; operations so make sure we bail on this loop.
+;
+;   for (i = 0; i < n; i++) {
+;     S1: d = D[i];
+;     S2: A[i + 1] = A[i] * B[i];
+;     S3: C[i] = d * E[i];
+;   }
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+define void @f(i32* noalias %a,
+               i32* noalias %b,
+               i32* noalias %c,
+               i32* noalias %d,
+               i32* noalias %e) {
+entry:
+  br label %for.body
+
+; CHECK: entry:
+; CHECK:    br label %for.body
+; CHECK: for.body:
+; CHECK:    br i1 %exitcond, label %for.end, label %for.body
+; CHECK: for.end:
+; CHECK:    ret void
+
+for.body:                                         ; preds = %for.body, %entry
+  %ind = phi i64 [ 0, %entry ], [ %add, %for.body ]
+
+  %arrayidxA = getelementptr inbounds i32, i32* %a, i64 %ind
+  %loadA = load i32, i32* %arrayidxA, align 4
+
+  %arrayidxB = getelementptr inbounds i32, i32* %b, i64 %ind
+  %loadB = load i32, i32* %arrayidxB, align 4
+
+  %mulA = mul i32 %loadB, %loadA
+
+  %arrayidxD = getelementptr inbounds i32, i32* %d, i64 %ind
+  %loadD = load i32, i32* %arrayidxD, align 4
+
+  %add = add nuw nsw i64 %ind, 1
+  %arrayidxA_plus_4 = getelementptr inbounds i32, i32* %a, i64 %add
+  store i32 %mulA, i32* %arrayidxA_plus_4, align 4
+
+  %arrayidxC = getelementptr inbounds i32, i32* %c, i64 %ind
+
+  %arrayidxE = getelementptr inbounds i32, i32* %e, i64 %ind
+  %loadE = load i32, i32* %arrayidxE, align 4
+
+  %mulC = mul i32 %loadD, %loadE
+
+  store i32 %mulC, i32* %arrayidxC, align 4
+
+  %exitcond = icmp eq i64 %add, 20
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
diff --git a/test/Transforms/LoopIdiom/basic-address-space.ll b/test/Transforms/LoopIdiom/basic-address-space.ll
index 697ab3726807..67cabf36eec7 100644
--- a/test/Transforms/LoopIdiom/basic-address-space.ll
+++ b/test/Transforms/LoopIdiom/basic-address-space.ll
@@ -22,7 +22,7 @@ for.body5:                                        ; preds = %for.body5, %bb.nph
   %j.02 = phi i16 [ 0, %bb.nph ], [ %inc, %for.body5 ]
   %mul = mul nsw i16 %i.04, 100
   %add = add nsw i16 %j.02, %mul
-  %arrayidx = getelementptr inbounds i8 addrspace(2)* %X, i16 %add
+  %arrayidx = getelementptr inbounds i8, i8 addrspace(2)* %X, i16 %add
   store i8 0, i8 addrspace(2)* %arrayidx, align 1
   %inc = add nsw i16 %j.02, 1
   %cmp4 = icmp eq i16 %inc, 100
@@ -45,7 +45,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
-  %arrayidx = getelementptr i32 addrspace(2)* %P, i64 %indvar
+  %arrayidx = getelementptr i32, i32 addrspace(2)* %P, i64 %indvar
   store i32 1, i32 addrspace(2)* %arrayidx, align 4
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, 10000
@@ -74,18 +74,18 @@ for.body:                                         ; preds = %for.inc, %for.body.
   %tmp5 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %add = add nsw i32 %tmp5, 4
   %idxprom = sext i32 %add to i64
-  %arrayidx = getelementptr inbounds [7 x i32] addrspace(2)* @g_50, i32 0, i64 %idxprom
-  %tmp2 = load i32 addrspace(2)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [7 x i32], [7 x i32] addrspace(2)* @g_50, i32 0, i64 %idxprom
+  %tmp2 = load i32, i32 addrspace(2)* %arrayidx, align 4
   %add4 = add nsw i32 %tmp5, 5
   %idxprom5 = sext i32 %add4 to i64
-  %arrayidx6 = getelementptr inbounds [7 x i32] addrspace(2)* @g_50, i32 0, i64 %idxprom5
+  %arrayidx6 = getelementptr inbounds [7 x i32], [7 x i32] addrspace(2)* @g_50, i32 0, i64 %idxprom5
   store i32 %tmp2, i32 addrspace(2)* %arrayidx6, align 4
   %inc = add nsw i32 %tmp5, 1
   %cmp = icmp slt i32 %inc, 2
   br i1 %cmp, label %for.body, label %for.end
 
 for.end:                                          ; preds = %for.inc
-  %tmp8 = load i32 addrspace(2)* getelementptr inbounds ([7 x i32] addrspace(2)* @g_50, i32 0, i64 6), align 4
+  %tmp8 = load i32, i32 addrspace(2)* getelementptr inbounds ([7 x i32], [7 x i32] addrspace(2)* @g_50, i32 0, i64 6), align 4
   ret i32 %tmp8
 }
 
diff --git a/test/Transforms/LoopIdiom/basic.ll b/test/Transforms/LoopIdiom/basic.ll
index 835a9f695ca9..a8a2c8efa386 100644
--- a/test/Transforms/LoopIdiom/basic.ll
+++ b/test/Transforms/LoopIdiom/basic.ll
@@ -1,5 +1,12 @@
 ; RUN: opt -basicaa -loop-idiom < %s -S | FileCheck %s
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+
+; For @test11_pattern
+; CHECK: @.memset_pattern = private unnamed_addr constant [4 x i32] [i32 1, i32 1, i32 1, i32 1]
+
+; For @test13_pattern
+; CHECK: @.memset_pattern.1 = private unnamed_addr constant [2 x i32*] [i32* @G, i32* @G]
+
 target triple = "x86_64-apple-darwin10.0.0"
 
 define void @test1(i8* %Base, i64 %Size) nounwind ssp {
@@ -8,7 +15,7 @@ bb.nph:                                           ; preds = %entry
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
   store i8 0, i8* %I.0.014, align 1
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, %Size
@@ -29,7 +36,7 @@ bb.nph:                                           ; preds = %entry
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body.cont ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
   store i8 0, i8* %I.0.014, align 1
   %indvar.next = add i64 %indvar, 1
   br label %for.body.cont
@@ -52,7 +59,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.011 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
-  %add.ptr.i = getelementptr i32* %Base, i64 %i.011
+  %add.ptr.i = getelementptr i32, i32* %Base, i64 %i.011
   store i32 16843009, i32* %add.ptr.i, align 4
   %inc = add nsw i64 %i.011, 1
   %exitcond = icmp eq i64 %inc, %Size
@@ -75,7 +82,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.011 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
-  %add.ptr.i = getelementptr i32* %Base, i64 %i.011
+  %add.ptr.i = getelementptr i32, i32* %Base, i64 %i.011
   store i32 16843009, i32* %add.ptr.i, align 4
   
   store i8 42, i8* %MayAlias
@@ -94,12 +101,12 @@ for.end:                                          ; preds = %entry
 ;; TODO: We should be able to promote this memset.  Not yet though.
 define void @test4(i8* %Base) nounwind ssp {
 bb.nph:                                           ; preds = %entry
-  %Base100 = getelementptr i8* %Base, i64 1000
+  %Base100 = getelementptr i8, i8* %Base, i64 1000
   br label %for.body
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
   store i8 0, i8* %I.0.014, align 1
   
   ;; Store beyond the range memset, should be safe to promote.
@@ -123,7 +130,7 @@ bb.nph:                                           ; preds = %entry
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
   
   %V = trunc i64 %indvar to i8
   store i8 %V, i8* %I.0.014, align 1
@@ -148,9 +155,9 @@ bb.nph:
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
-  %DestI = getelementptr i8* %Dest, i64 %indvar
-  %V = load i8* %I.0.014, align 1
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
+  %DestI = getelementptr i8, i8* %Dest, i64 %indvar
+  %V = load i8, i8* %I.0.014, align 1
   store i8 %V, i8* %DestI, align 1
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, %Size
@@ -175,7 +182,7 @@ for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body.cont ]
   br label %for.body.cont
 for.body.cont:
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
   store i8 0, i8* %I.0.014, align 1
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, %Size
@@ -195,7 +202,7 @@ bb.nph:                                           ; preds = %entry
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %PI = getelementptr i64* %Ptr, i64 %indvar
+  %PI = getelementptr i64, i64* %Ptr, i64 %indvar
   store i64 0, i64 *%PI
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, 1
@@ -221,9 +228,9 @@ bb.nph:
 
 for.body:                                         ; preds = %bb.nph, %for.body
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %for.body ]
-  %I.0.014 = getelementptr i8* %Base, i64 %indvar
-  %DestI = getelementptr i8* %Dest, i64 %indvar
-  %V = load i8* %I.0.014, align 1
+  %I.0.014 = getelementptr i8, i8* %Base, i64 %indvar
+  %DestI = getelementptr i8, i8* %Dest, i64 %indvar
+  %V = load i8, i8* %I.0.014, align 1
   store i8 %V, i8* %DestI, align 1
 
   ;; This store can clobber the input.
@@ -254,7 +261,7 @@ for.body5:                                        ; preds = %for.body5, %bb.nph
   %mul = mul nsw i32 %i.04, 100
   %add = add nsw i32 %j.02, %mul
   %idxprom = sext i32 %add to i64
-  %arrayidx = getelementptr inbounds i8* %X, i64 %idxprom
+  %arrayidx = getelementptr inbounds i8, i8* %X, i64 %idxprom
   store i8 0, i8* %arrayidx, align 1
   %inc = add nsw i32 %j.02, 1
   %cmp4 = icmp eq i32 %inc, 100
@@ -283,7 +290,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
-  %arrayidx = getelementptr i32* %P, i64 %indvar
+  %arrayidx = getelementptr i32, i32* %P, i64 %indvar
   store i32 1, i32* %arrayidx, align 4
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, 10000
@@ -306,7 +313,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
-  %arrayidx = getelementptr i32** %P, i64 %indvar
+  %arrayidx = getelementptr i32*, i32** %P, i64 %indvar
   store i32* null, i32** %arrayidx, align 4
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, 10000
@@ -332,7 +339,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
-  %arrayidx = getelementptr i32** %P, i64 %indvar
+  %arrayidx = getelementptr i32*, i32** %P, i64 %indvar
   store i32* @G, i32** %arrayidx, align 4
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp eq i64 %indvar.next, 10000
@@ -362,18 +369,18 @@ for.body:                                         ; preds = %for.inc, %for.body.
   %tmp5 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %add = add nsw i32 %tmp5, 4
   %idxprom = sext i32 %add to i64
-  %arrayidx = getelementptr inbounds [7 x i32]* @g_50, i32 0, i64 %idxprom
-  %tmp2 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [7 x i32], [7 x i32]* @g_50, i32 0, i64 %idxprom
+  %tmp2 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %tmp5, 5
   %idxprom5 = sext i32 %add4 to i64
-  %arrayidx6 = getelementptr inbounds [7 x i32]* @g_50, i32 0, i64 %idxprom5
+  %arrayidx6 = getelementptr inbounds [7 x i32], [7 x i32]* @g_50, i32 0, i64 %idxprom5
   store i32 %tmp2, i32* %arrayidx6, align 4
   %inc = add nsw i32 %tmp5, 1
   %cmp = icmp slt i32 %inc, 2
   br i1 %cmp, label %for.body, label %for.end
 
 for.end:                                          ; preds = %for.inc
-  %tmp8 = load i32* getelementptr inbounds ([7 x i32]* @g_50, i32 0, i64 6), align 4
+  %tmp8 = load i32, i32* getelementptr inbounds ([7 x i32], [7 x i32]* @g_50, i32 0, i64 6), align 4
   ret i32 %tmp8
 ; CHECK-LABEL: @test14(
 ; CHECK: for.body:
@@ -393,7 +400,7 @@ define void @PR14241(i32* %s, i64 %size) {
 
 entry:
   %end.idx = add i64 %size, -1
-  %end.ptr = getelementptr inbounds i32* %s, i64 %end.idx
+  %end.ptr = getelementptr inbounds i32, i32* %s, i64 %end.idx
   br label %while.body
 ; CHECK-NOT: memcpy
 ;
@@ -403,13 +410,13 @@ entry:
 
 while.body:
   %phi.ptr = phi i32* [ %s, %entry ], [ %next.ptr, %while.body ]
-  %src.ptr = getelementptr inbounds i32* %phi.ptr, i64 1
-  %val = load i32* %src.ptr, align 4
+  %src.ptr = getelementptr inbounds i32, i32* %phi.ptr, i64 1
+  %val = load i32, i32* %src.ptr, align 4
 ; CHECK: load
-  %dst.ptr = getelementptr inbounds i32* %phi.ptr, i64 0
+  %dst.ptr = getelementptr inbounds i32, i32* %phi.ptr, i64 0
   store i32 %val, i32* %dst.ptr, align 4
 ; CHECK: store
-  %next.ptr = getelementptr inbounds i32* %phi.ptr, i64 1
+  %next.ptr = getelementptr inbounds i32, i32* %phi.ptr, i64 1
   %cmp = icmp eq i32* %next.ptr, %end.ptr
   br i1 %cmp, label %exit, label %while.body
 
diff --git a/test/Transforms/LoopIdiom/crash.ll b/test/Transforms/LoopIdiom/crash.ll
index 969adbcd7635..c964aacd00c2 100644
--- a/test/Transforms/LoopIdiom/crash.ll
+++ b/test/Transforms/LoopIdiom/crash.ll
@@ -8,13 +8,13 @@ define void @test1(i64* %iwork, i64 %x)  {
 bb0:
   %mul116 = mul nsw i64 %x, %x
   %incdec.ptr6.sum175 = add i64 42, %x
-  %arrayidx135 = getelementptr inbounds i64* %iwork, i64 %incdec.ptr6.sum175
+  %arrayidx135 = getelementptr inbounds i64, i64* %iwork, i64 %incdec.ptr6.sum175
   br label %bb1
 bb1:
   %storemerge4226 = phi i64 [ 0, %bb0 ], [ %inc139, %bb1 ]
   store i64 1, i64* %arrayidx135, align 8
   %incdec.ptr6.sum176 = add i64 %mul116, %storemerge4226
-  %arrayidx137 = getelementptr inbounds i64* %iwork, i64 %incdec.ptr6.sum176
+  %arrayidx137 = getelementptr inbounds i64, i64* %iwork, i64 %incdec.ptr6.sum176
   store i64 1, i64* %arrayidx137, align 8
   %inc139 = add nsw i64 %storemerge4226, 1
   %cmp131 = icmp sgt i64 %storemerge4226, 42
diff --git a/test/Transforms/LoopIdiom/debug-line.ll b/test/Transforms/LoopIdiom/debug-line.ll
index 863df3c72d50..bcd862d7a729 100644
--- a/test/Transforms/LoopIdiom/debug-line.ll
+++ b/test/Transforms/LoopIdiom/debug-line.ll
@@ -5,13 +5,13 @@ target triple = "x86_64-apple-darwin10.0.0"
 
 define void @foo(double* nocapture %a) nounwind ssp {
 entry:
-  tail call void @llvm.dbg.value(metadata double* %a, i64 0, metadata !5, metadata !{}), !dbg !8
-  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !10, metadata !{}), !dbg !14
+  tail call void @llvm.dbg.value(metadata double* %a, i64 0, metadata !5, metadata !DIExpression()), !dbg !8
+  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !10, metadata !DIExpression()), !dbg !14
   br label %for.body
 
 for.body:                                         ; preds = %entry, %for.body
   %indvar = phi i64 [ 0, %entry ], [ %indvar.next, %for.body ]
-  %arrayidx = getelementptr double* %a, i64 %indvar
+  %arrayidx = getelementptr double, double* %a, i64 %indvar
 ; CHECK: call void @llvm.memset{{.+}} !dbg 
   store double 0.000000e+00, double* %arrayidx, align 8, !dbg !15
   %indvar.next = add i64 %indvar, 1
@@ -19,7 +19,7 @@ for.body:                                         ; preds = %entry, %for.body
   br i1 %exitcond, label %for.body, label %for.end, !dbg !14
 
 for.end:                                          ; preds = %for.body
-  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !10, metadata !{}), !dbg !16
+  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !10, metadata !DIExpression()), !dbg !16
   ret void, !dbg !17
 }
 
@@ -30,23 +30,23 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 !llvm.module.flags = !{!19}
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00foo\00foo\00\002\000\001\000\006\00256\000\000", !18, !1, !3, null, void (double*)* @foo, null, null, null} ; [ DW_TAG_subprogram ] [line 2] [def] [scope 0] [foo]
-!1 = !{!"0x29", !18} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang version 2.9 (trunk 127165:127174)\001\00\000\00\000", !18, !9, !9, null, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !18, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", line: 2, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !18, scope: !1, type: !3, function: void (double*)* @foo)
+!1 = !DIFile(filename: "li.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 2.9 (trunk 127165:127174)", isOptimized: true, emissionKind: 0, file: !18, enums: !9, retainedTypes: !9)
+!3 = !DISubroutineType(types: !4)
 !4 = !{null}
-!5 = !{!"0x101\00a\0016777218\000", !0, !1, !6} ; [ DW_TAG_arg_variable ]
-!6 = !{!"0xf\00\000\0064\0064\000\000", null, !2, !7} ; [ DW_TAG_pointer_type ]
-!7 = !{!"0x24\00double\000\0064\0064\000\000\004", null, !2} ; [ DW_TAG_base_type ]
-!8 = !MDLocation(line: 2, column: 18, scope: !0)
-!9 = !{i32 0}
-!10 = !{!"0x100\00i\003\000", !11, !1, !13} ; [ DW_TAG_auto_variable ]
-!11 = !{!"0xb\003\003\001", !18, !12} ; [ DW_TAG_lexical_block ]
-!12 = !{!"0xb\002\0021\000", !18, !0} ; [ DW_TAG_lexical_block ]
-!13 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !2} ; [ DW_TAG_base_type ]
-!14 = !MDLocation(line: 3, column: 3, scope: !12)
-!15 = !MDLocation(line: 4, column: 5, scope: !11)
-!16 = !MDLocation(line: 3, column: 29, scope: !11)
-!17 = !MDLocation(line: 5, column: 1, scope: !12)
-!18 = !{!"li.c", !"/private/tmp"}
-!19 = !{i32 1, !"Debug Info Version", i32 2}
+!5 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "a", line: 2, arg: 1, scope: !0, file: !1, type: !6)
+!6 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, scope: !2, baseType: !7)
+!7 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
+!8 = !DILocation(line: 2, column: 18, scope: !0)
+!9 = !{}
+!10 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "i", line: 3, scope: !11, file: !1, type: !13)
+!11 = distinct !DILexicalBlock(line: 3, column: 3, file: !18, scope: !12)
+!12 = distinct !DILexicalBlock(line: 2, column: 21, file: !18, scope: !0)
+!13 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!14 = !DILocation(line: 3, column: 3, scope: !12)
+!15 = !DILocation(line: 4, column: 5, scope: !11)
+!16 = !DILocation(line: 3, column: 29, scope: !11)
+!17 = !DILocation(line: 5, column: 1, scope: !12)
+!18 = !DIFile(filename: "li.c", directory: "/private/tmp")
+!19 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/LoopIdiom/memset_noidiom.ll b/test/Transforms/LoopIdiom/memset_noidiom.ll
index f2b55aed5467..f039d99ca49a 100644
--- a/test/Transforms/LoopIdiom/memset_noidiom.ll
+++ b/test/Transforms/LoopIdiom/memset_noidiom.ll
@@ -15,7 +15,7 @@ for.body.lr.ph:                                   ; preds = %entry
 
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %indvar = phi i64 [ 0, %for.body.lr.ph ], [ %indvar.next, %for.body ]
-  %p.02 = getelementptr i8* %b, i64 %indvar
+  %p.02 = getelementptr i8, i8* %b, i64 %indvar
   store i8 %conv6, i8* %p.02, align 1
   %indvar.next = add i64 %indvar, 1
   %exitcond = icmp ne i64 %indvar.next, %len
diff --git a/test/Transforms/LoopIdiom/non-canonical-loop.ll b/test/Transforms/LoopIdiom/non-canonical-loop.ll
index a6a4f9227f9a..77c818f834ec 100644
--- a/test/Transforms/LoopIdiom/non-canonical-loop.ll
+++ b/test/Transforms/LoopIdiom/non-canonical-loop.ll
@@ -20,7 +20,7 @@ start.exit:                       ; preds = %entry
 
 for.bodyprime:                                    ; preds = %for.bodyprime, %start.exit
   %i.057375 = phi i32 [ 0, %start.exit ], [ %1, %for.bodyprime ]
-  %arrayidx8prime = getelementptr inbounds i32* %currMB, i32 %i.057375
+  %arrayidx8prime = getelementptr inbounds i32, i32* %currMB, i32 %i.057375
   store i32 0, i32* %arrayidx8prime, align 4
   %1 = add i32 %i.057375, 1
   %cmp5prime = icmp slt i32 %1, 4
diff --git a/test/Transforms/LoopIdiom/scev-invalidation.ll b/test/Transforms/LoopIdiom/scev-invalidation.ll
index a244d9a280b9..2fe8a3023c4c 100644
--- a/test/Transforms/LoopIdiom/scev-invalidation.ll
+++ b/test/Transforms/LoopIdiom/scev-invalidation.ll
@@ -11,7 +11,7 @@ for.cond:                                         ; preds = %for.inc, %entry
   %backslashes.0 = phi i32 [ undef, %entry ], [ %backslashes.2, %for.inc ]
   %p.0 = phi i8* [ undef, %entry ], [ %incdec.ptr3, %for.inc ]
   %q.0 = phi i8* [ undef, %entry ], [ %q.2, %for.inc ]
-  %0 = load i8* %p.0, align 1
+  %0 = load i8, i8* %p.0, align 1
   switch i8 %0, label %while.cond.preheader [
     i8 0, label %for.cond4.preheader
     i8 92, label %for.inc
@@ -36,7 +36,7 @@ for.body6.lr.ph:                                  ; preds = %for.cond4.preheader
 while.body:                                       ; preds = %while.body.lr.ph, %while.body
   %q.112 = phi i8* [ %q.0, %while.body.lr.ph ], [ %incdec.ptr, %while.body ]
   %backslashes.111 = phi i32 [ %backslashes.0, %while.body.lr.ph ], [ %dec, %while.body ]
-  %incdec.ptr = getelementptr inbounds i8* %q.112, i64 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %q.112, i64 1
   store i8 92, i8* %incdec.ptr, align 1
   %dec = add nsw i32 %backslashes.111, -1
   %tobool2 = icmp eq i32 %dec, 0
@@ -44,7 +44,7 @@ while.body:                                       ; preds = %while.body.lr.ph, %
 
 while.cond.for.inc.loopexit_crit_edge:            ; preds = %while.body
   %scevgep.sum = add i64 %2, 1
-  %scevgep13 = getelementptr i8* %q.0, i64 %scevgep.sum
+  %scevgep13 = getelementptr i8, i8* %q.0, i64 %scevgep.sum
   br label %for.inc.loopexit
 
 for.inc.loopexit:                                 ; preds = %while.cond.for.inc.loopexit_crit_edge, %while.cond.preheader
@@ -54,13 +54,13 @@ for.inc.loopexit:                                 ; preds = %while.cond.for.inc.
 for.inc:                                          ; preds = %for.inc.loopexit, %for.cond
   %backslashes.2 = phi i32 [ %backslashes.0, %for.cond ], [ 0, %for.inc.loopexit ]
   %q.2 = phi i8* [ %q.0, %for.cond ], [ %q.1.lcssa, %for.inc.loopexit ]
-  %incdec.ptr3 = getelementptr inbounds i8* %p.0, i64 1
+  %incdec.ptr3 = getelementptr inbounds i8, i8* %p.0, i64 1
   br label %for.cond
 
 for.body6:                                        ; preds = %for.body6.lr.ph, %for.body6
   %q.39 = phi i8* [ %q.0, %for.body6.lr.ph ], [ %incdec.ptr7, %for.body6 ]
   %backslashes.38 = phi i32 [ %backslashes.0, %for.body6.lr.ph ], [ %dec9, %for.body6 ]
-  %incdec.ptr7 = getelementptr inbounds i8* %q.39, i64 1
+  %incdec.ptr7 = getelementptr inbounds i8, i8* %q.39, i64 1
   store i8 92, i8* %incdec.ptr7, align 1
   %dec9 = add nsw i32 %backslashes.38, -1
   %tobool5 = icmp eq i32 %dec9, 0
diff --git a/test/Transforms/LoopInterchange/currentLimitation.ll b/test/Transforms/LoopInterchange/currentLimitation.ll
new file mode 100644
index 000000000000..8045dd87a75a
--- /dev/null
+++ b/test/Transforms/LoopInterchange/currentLimitation.ll
@@ -0,0 +1,58 @@
+; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
+;; These are test that fail to interchange due to current limitation. This will go off once we extend the loop interchange pass.
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+ 
+@A = common global [100 x [100 x i32]] zeroinitializer
+@B = common global [100 x [100 x [100 x i32]]] zeroinitializer
+ 
+;;--------------------------------------Test case 01------------------------------------
+;; [FIXME] This loop though valid is currently not interchanged due to the limitation that we cannot split the inner loop latch due to multiple use of inner induction
+;; variable.(used to increment the loop counter and to access A[j+1][i+1]
+;;  for(int i=0;i<N-1;i++)
+;;    for(int j=1;j<N-1;j++)
+;;      A[j+1][i+1] = A[j+1][i+1] + k;
+
+define void @interchange_01(i32 %k, i32 %N) {
+ entry:
+   %sub = add nsw i32 %N, -1
+   %cmp26 = icmp sgt i32 %N, 1
+   br i1 %cmp26, label %for.cond1.preheader.lr.ph, label %for.end17
+ 
+ for.cond1.preheader.lr.ph:
+   %cmp324 = icmp sgt i32 %sub, 1
+   %0 = add i32 %N, -2
+   %1 = sext i32 %sub to i64
+   br label %for.cond1.preheader
+ 
+ for.cond.loopexit:
+   %cmp = icmp slt i64 %indvars.iv.next29, %1
+   br i1 %cmp, label %for.cond1.preheader, label %for.end17
+ 
+ for.cond1.preheader:
+   %indvars.iv28 = phi i64 [ 0, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next29, %for.cond.loopexit ]
+   %indvars.iv.next29 = add nuw nsw i64 %indvars.iv28, 1
+   br i1 %cmp324, label %for.body4, label %for.cond.loopexit
+ 
+ for.body4:
+   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body4 ], [ 1, %for.cond1.preheader ]
+   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+   %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next, i64 %indvars.iv.next29
+   %2 = load i32, i32* %arrayidx7
+   %add8 = add nsw i32 %2, %k
+   store i32 %add8, i32* %arrayidx7
+   %lftr.wideiv = trunc i64 %indvars.iv to i32
+   %exitcond = icmp eq i32 %lftr.wideiv, %0
+   br i1 %exitcond, label %for.cond.loopexit, label %for.body4
+ 
+ for.end17: 
+   ret void
+}
+;; Inner loop not split so it is not interchanged.
+; CHECK-LABEL: @interchange_01
+; CHECK:      for.body4:
+; CHECK-NEXT:   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body4 ], [ 1, %for.body4.preheader ]
+; CHECK-NEXT:   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK-NEXT:   %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next, i64 %indvars.iv.next29
+ 
diff --git a/test/Transforms/LoopInterchange/interchange.ll b/test/Transforms/LoopInterchange/interchange.ll
new file mode 100644
index 000000000000..39941660e601
--- /dev/null
+++ b/test/Transforms/LoopInterchange/interchange.ll
@@ -0,0 +1,557 @@
+; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
+;; We test the complete .ll for adjustment in outer loop header/latch and inner loop header/latch.
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@A = common global [100 x [100 x i32]] zeroinitializer
+@B = common global [100 x i32] zeroinitializer
+@C = common global [100 x [100 x i32]] zeroinitializer
+@D = common global [100 x [100 x [100 x i32]]] zeroinitializer
+
+declare void @foo(...)
+
+;;--------------------------------------Test case 01------------------------------------
+;;  for(int i=0;i<N;i++)
+;;    for(int j=1;j<N;j++)
+;;      A[j][i] = A[j][i]+k;
+
+define void @interchange_01(i32 %k, i32 %N) {
+entry:
+  %cmp21 = icmp sgt i32 %N, 0
+  br i1 %cmp21, label %for.cond1.preheader.lr.ph, label %for.end12
+
+for.cond1.preheader.lr.ph:
+  %cmp219 = icmp sgt i32 %N, 1
+  %0 = add i32 %N, -1
+  br label %for.cond1.preheader
+
+for.cond1.preheader: 
+  %indvars.iv23 = phi i64 [ 0, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next24, %for.inc10 ]
+  br i1 %cmp219, label %for.body3, label %for.inc10
+
+for.body3:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.cond1.preheader ]
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv23
+  %1 = load i32, i32* %arrayidx5
+  %add = add nsw i32 %1, %k
+  store i32 %add, i32* %arrayidx5
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc10, label %for.body3
+
+for.inc10:
+  %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
+  %lftr.wideiv25 = trunc i64 %indvars.iv23 to i32
+  %exitcond26 = icmp eq i32 %lftr.wideiv25, %0
+  br i1 %exitcond26, label %for.end12, label %for.cond1.preheader
+
+for.end12:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_01
+; CHECK: entry:
+; CHECK:   %cmp21 = icmp sgt i32 %N, 0
+; CHECK:   br i1 %cmp21, label %for.body3.preheader, label %for.end12
+; CHECK: for.cond1.preheader.lr.ph:                        
+; CHECK:   br label %for.cond1.preheader
+; CHECK: for.cond1.preheader:                              
+; CHECK:   %indvars.iv23 = phi i64 [ 0, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next24, %for.inc10 ]
+; CHECK:   br i1 %cmp219, label %for.body3.split1, label %for.end12.loopexit
+; CHECK: for.body3.preheader:                              
+; CHECK:   %cmp219 = icmp sgt i32 %N, 1
+; CHECK:   %0 = add i32 %N, -1
+; CHECK:   br label %for.body3
+; CHECK: for.body3:                                        
+; CHECK:   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 1, %for.body3.preheader ]
+; CHECK:   br label %for.cond1.preheader.lr.ph
+; CHECK: for.body3.split1:                                 
+; CHECK:   %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv23
+; CHECK:   %1 = load i32, i32* %arrayidx5
+; CHECK:   %add = add nsw i32 %1, %k
+; CHECK:   store i32 %add, i32* %arrayidx5
+; CHECK:   br label %for.inc10.loopexit
+; CHECK: for.body3.split:                                  
+; CHECK:   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:   %lftr.wideiv = trunc i64 %indvars.iv to i32
+; CHECK:   %exitcond = icmp eq i32 %lftr.wideiv, %0
+; CHECK:   br i1 %exitcond, label %for.end12.loopexit, label %for.body3
+; CHECK: for.inc10.loopexit:                               
+; CHECK:   br label %for.inc10
+; CHECK: for.inc10:                                        
+; CHECK:   %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
+; CHECK:   %lftr.wideiv25 = trunc i64 %indvars.iv23 to i32
+; CHECK:   %exitcond26 = icmp eq i32 %lftr.wideiv25, %0
+; CHECK:   br i1 %exitcond26, label %for.body3.split, label %for.cond1.preheader
+; CHECK: for.end12.loopexit:                               
+; CHECK:   br label %for.end12
+; CHECK: for.end12:                                        
+; CHECK:   ret void
+
+;;--------------------------------------Test case 02-------------------------------------
+
+;; for(int i=0;i<100;i++)
+;;   for(int j=100;j>=0;j--)
+;;     A[j][i] = A[j][i]+k;
+
+define void @interchange_02(i32 %k) {
+entry:
+  br label %for.cond1.preheader
+
+for.cond1.preheader:
+  %indvars.iv19 = phi i64 [ 0, %entry ], [ %indvars.iv.next20, %for.inc10 ]
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ 100, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv19
+  %0 = load i32, i32* %arrayidx5
+  %add = add nsw i32 %0, %k
+  store i32 %add, i32* %arrayidx5
+  %indvars.iv.next = add nsw i64 %indvars.iv, -1
+  %cmp2 = icmp sgt i64 %indvars.iv, 0
+  br i1 %cmp2, label %for.body3, label %for.inc10
+
+for.inc10:
+  %indvars.iv.next20 = add nuw nsw i64 %indvars.iv19, 1
+  %exitcond = icmp eq i64 %indvars.iv.next20, 100
+  br i1 %exitcond, label %for.end11, label %for.cond1.preheader
+
+for.end11:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_02
+; CHECK: entry:
+; CHECK:   br label %for.body3.preheader
+; CHECK: for.cond1.preheader.preheader: 
+; CHECK:   br label %for.cond1.preheader
+; CHECK: for.cond1.preheader:  
+; CHECK:   %indvars.iv19 = phi i64 [ %indvars.iv.next20, %for.inc10 ], [ 0, %for.cond1.preheader.preheader ]
+; CHECK:   br label %for.body3.split1
+; CHECK: for.body3.preheader: 
+; CHECK:   br label %for.body3
+; CHECK: for.body3: 
+; CHECK:   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 100, %for.body3.preheader ]
+; CHECK:   br label %for.cond1.preheader.preheader
+; CHECK: for.body3.split1:                                 ; preds = %for.cond1.preheader
+; CHECK:   %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv19
+; CHECK:   %0 = load i32, i32* %arrayidx5
+; CHECK:   %add = add nsw i32 %0, %k
+; CHECK:   store i32 %add, i32* %arrayidx5
+; CHECK:   br label %for.inc10
+; CHECK: for.body3.split:
+; CHECK:   %indvars.iv.next = add nsw i64 %indvars.iv, -1
+; CHECK:   %cmp2 = icmp sgt i64 %indvars.iv, 0
+; CHECK:   br i1 %cmp2, label %for.body3, label %for.end11
+; CHECK: for.inc10:
+; CHECK:   %indvars.iv.next20 = add nuw nsw i64 %indvars.iv19, 1
+; CHECK:   %exitcond = icmp eq i64 %indvars.iv.next20, 100
+; CHECK:   br i1 %exitcond, label %for.body3.split, label %for.cond1.preheader
+; CHECK: for.end11: 
+; CHECK:   ret void
+
+;;--------------------------------------Test case 03-------------------------------------
+;; Loops should not be interchanged in this case as it is not profitable.
+;;  for(int i=0;i<100;i++)
+;;    for(int j=0;j<100;j++)
+;;      A[i][j] = A[i][j]+k;
+
+define void @interchange_03(i32 %k) {
+entry:
+  br label %for.cond1.preheader
+
+for.cond1.preheader:
+  %indvars.iv21 = phi i64 [ 0, %entry ], [ %indvars.iv.next22, %for.inc10 ]
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv21, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx5
+  %add = add nsw i32 %0, %k
+  store i32 %add, i32* %arrayidx5
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond = icmp eq i64 %indvars.iv.next, 100
+  br i1 %exitcond, label %for.inc10, label %for.body3
+
+for.inc10:
+  %indvars.iv.next22 = add nuw nsw i64 %indvars.iv21, 1
+  %exitcond23 = icmp eq i64 %indvars.iv.next22, 100
+  br i1 %exitcond23, label %for.end12, label %for.cond1.preheader
+
+for.end12:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_03
+; CHECK: entry:
+; CHECK:   br label %for.cond1.preheader.preheader
+; CHECK: for.cond1.preheader.preheader:                    ; preds = %entry
+; CHECK:   br label %for.cond1.preheader
+; CHECK: for.cond1.preheader:                              ; preds = %for.cond1.preheader.preheader, %for.inc10
+; CHECK:   %indvars.iv21 = phi i64 [ %indvars.iv.next22, %for.inc10 ], [ 0, %for.cond1.preheader.preheader ]
+; CHECK:  br label %for.body3.preheader
+; CHECK: for.body3.preheader:                              ; preds = %for.cond1.preheader
+; CHECK:   br label %for.body3
+; CHECK: for.body3:                                        ; preds = %for.body3.preheader, %for.body3
+; CHECK:   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 0, %for.body3.preheader ]
+; CHECK:   %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv21, i64 %indvars.iv
+; CHECK:   %0 = load i32, i32* %arrayidx5
+; CHECK:   %add = add nsw i32 %0, %k
+; CHECK:   store i32 %add, i32* %arrayidx5
+; CHECK:   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:   %exitcond = icmp eq i64 %indvars.iv.next, 100
+; CHECK:   br i1 %exitcond, label %for.inc10, label %for.body3
+; CHECK: for.inc10:                                        ; preds = %for.body3
+; CHECK:   %indvars.iv.next22 = add nuw nsw i64 %indvars.iv21, 1
+; CHECK:   %exitcond23 = icmp eq i64 %indvars.iv.next22, 100
+; CHECK:   br i1 %exitcond23, label %for.end12, label %for.cond1.preheader
+; CHECK: for.end12:                                        ; preds = %for.inc10
+; CHECK:   ret void
+
+
+;;--------------------------------------Test case 04-------------------------------------
+;; Loops should not be interchanged in this case as it is not legal due to dependency.
+;;  for(int j=0;j<99;j++)
+;;   for(int i=0;i<99;i++)
+;;       A[j][i+1] = A[j+1][i]+k;
+
+define void @interchange_04(i32 %k){
+entry:
+  br label %for.cond1.preheader
+
+for.cond1.preheader:
+  %indvars.iv23 = phi i64 [ 0, %entry ], [ %indvars.iv.next24, %for.inc12 ]
+  %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next24, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx5
+  %add6 = add nsw i32 %0, %k
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %arrayidx11 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv23, i64 %indvars.iv.next
+  store i32 %add6, i32* %arrayidx11
+  %exitcond = icmp eq i64 %indvars.iv.next, 99
+  br i1 %exitcond, label %for.inc12, label %for.body3
+
+for.inc12:
+  %exitcond25 = icmp eq i64 %indvars.iv.next24, 99
+  br i1 %exitcond25, label %for.end14, label %for.cond1.preheader
+
+for.end14:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_04
+; CHECK: entry:
+; CHECK:   br label %for.cond1.preheader
+; CHECK: for.cond1.preheader:                              ; preds = %for.inc12, %entry
+; CHECK:   %indvars.iv23 = phi i64 [ 0, %entry ], [ %indvars.iv.next24, %for.inc12 ]
+; CHECK:   %indvars.iv.next24 = add nuw nsw i64 %indvars.iv23, 1
+; CHECK:   br label %for.body3
+; CHECK: for.body3:                                        ; preds = %for.body3, %for.cond1.preheader
+; CHECK:   %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
+; CHECK:   %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv.next24, i64 %indvars.iv
+; CHECK:   %0 = load i32, i32* %arrayidx5
+; CHECK:   %add6 = add nsw i32 %0, %k
+; CHECK:   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:   %arrayidx11 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv23, i64 %indvars.iv.next
+; CHECK:   store i32 %add6, i32* %arrayidx11
+; CHECK:   %exitcond = icmp eq i64 %indvars.iv.next, 99
+; CHECK:   br i1 %exitcond, label %for.inc12, label %for.body3
+; CHECK: for.inc12:                                        ; preds = %for.body3
+; CHECK:   %exitcond25 = icmp eq i64 %indvars.iv.next24, 99
+; CHECK:   br i1 %exitcond25, label %for.end14, label %for.cond1.preheader
+; CHECK: for.end14:                                        ; preds = %for.inc12
+; CHECK:   ret void
+
+
+
+;;--------------------------------------Test case 05-------------------------------------
+;; Loops not tightly nested are not interchanged
+;;  for(int j=0;j<N;j++) {
+;;    B[j] = j+k;
+;;    for(int i=0;i<N;i++)
+;;      A[j][i] = A[j][i]+B[j];
+;;  }
+
+define void @interchange_05(i32 %k, i32 %N){
+entry:
+  %cmp30 = icmp sgt i32 %N, 0
+  br i1 %cmp30, label %for.body.lr.ph, label %for.end17
+
+for.body.lr.ph:
+  %0 = add i32 %N, -1
+  %1 = zext i32 %k to i64
+  br label %for.body
+
+for.body:
+  %indvars.iv32 = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next33, %for.inc15 ]
+  %2 = add nsw i64 %indvars.iv32, %1
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* @B, i64 0, i64 %indvars.iv32
+  %3 = trunc i64 %2 to i32
+  store i32 %3, i32* %arrayidx
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ 0, %for.body ], [ %indvars.iv.next, %for.body3 ]
+  %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv32, i64 %indvars.iv
+  %4 = load i32, i32* %arrayidx7
+  %add10 = add nsw i32 %3, %4
+  store i32 %add10, i32* %arrayidx7
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc15, label %for.body3
+
+for.inc15:
+  %indvars.iv.next33 = add nuw nsw i64 %indvars.iv32, 1
+  %lftr.wideiv35 = trunc i64 %indvars.iv32 to i32
+  %exitcond36 = icmp eq i32 %lftr.wideiv35, %0
+  br i1 %exitcond36, label %for.end17, label %for.body
+
+for.end17:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_05
+; CHECK: entry:
+; CHECK: %cmp30 = icmp sgt i32 %N, 0
+; CHECK: br i1 %cmp30, label %for.body.lr.ph, label %for.end17
+; CHECK: for.body.lr.ph:
+; CHECK: %0 = add i32 %N, -1
+; CHECK: %1 = zext i32 %k to i64
+; CHECK: br label %for.body
+; CHECK: for.body:
+; CHECK: %indvars.iv32 = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next33, %for.inc15 ]
+; CHECK: %2 = add nsw i64 %indvars.iv32, %1
+; CHECK: %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* @B, i64 0, i64 %indvars.iv32
+; CHECK: %3 = trunc i64 %2 to i32
+; CHECK: store i32 %3, i32* %arrayidx
+; CHECK: br label %for.body3.preheader
+; CHECK: for.body3.preheader:
+; CHECK: br label %for.body3
+; CHECK: for.body3:
+; CHECK: %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 0, %for.body3.preheader ]
+; CHECK: %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv32, i64 %indvars.iv
+; CHECK: %4 = load i32, i32* %arrayidx7
+; CHECK: %add10 = add nsw i32 %3, %4
+; CHECK: store i32 %add10, i32* %arrayidx7
+; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK: %lftr.wideiv = trunc i64 %indvars.iv to i32
+; CHECK: %exitcond = icmp eq i32 %lftr.wideiv, %0
+; CHECK: br i1 %exitcond, label %for.inc15, label %for.body3
+; CHECK: for.inc15:
+; CHECK: %indvars.iv.next33 = add nuw nsw i64 %indvars.iv32, 1
+; CHECK: %lftr.wideiv35 = trunc i64 %indvars.iv32 to i32
+; CHECK: %exitcond36 = icmp eq i32 %lftr.wideiv35, %0
+; CHECK: br i1 %exitcond36, label %for.end17.loopexit, label %for.body
+; CHECK: for.end17.loopexit:
+; CHECK: br label %for.end17
+; CHECK: for.end17:
+; CHECK: ret void
+
+
+;;--------------------------------------Test case 06-------------------------------------
+;; Loops not tightly nested are not interchanged
+;;  for(int j=0;j<N;j++) {
+;;    foo();
+;;    for(int i=2;i<N;i++)
+;;      A[j][i] = A[j][i]+k;
+;;  }
+
+define void @interchange_06(i32 %k, i32 %N) {
+entry:
+  %cmp22 = icmp sgt i32 %N, 0
+  br i1 %cmp22, label %for.body.lr.ph, label %for.end12
+
+for.body.lr.ph:
+  %0 = add i32 %N, -1
+  br label %for.body
+
+for.body:
+  %indvars.iv24 = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next25, %for.inc10 ]
+  tail call void (...) @foo()
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 2, %for.body ]
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv24, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx5
+  %add = add nsw i32 %1, %k
+  store i32 %add, i32* %arrayidx5
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc10, label %for.body3
+
+for.inc10:
+  %indvars.iv.next25 = add nuw nsw i64 %indvars.iv24, 1
+  %lftr.wideiv26 = trunc i64 %indvars.iv24 to i32
+  %exitcond27 = icmp eq i32 %lftr.wideiv26, %0
+  br i1 %exitcond27, label %for.end12, label %for.body
+
+for.end12:
+  ret void
+}
+;; Here we are checking if the inner phi is not split then we have not interchanged.
+; CHECK-LABEL: @interchange_06
+; CHECK:  phi i64 [ %indvars.iv.next, %for.body3 ], [ 2, %for.body3.preheader ]
+; CHECK-NEXT: getelementptr
+; CHECK-NEXT: %1 = load
+
+;;--------------------------------------Test case 07-------------------------------------
+;; FIXME:
+;; Test for interchange when we have an lcssa phi. This should ideally be interchanged but it is currently not supported.
+;;     for(gi=1;gi<N;gi++)
+;;       for(gj=1;gj<M;gj++)
+;;         A[gj][gi] = A[gj - 1][gi] + C[gj][gi];
+
+@gi = common global i32 0
+@gj = common global i32 0
+
+define void @interchange_07(i32 %N, i32 %M){
+entry:
+  store i32 1, i32* @gi
+  %cmp21 = icmp sgt i32 %N, 1
+  br i1 %cmp21, label %for.cond1.preheader.lr.ph, label %for.end16
+
+for.cond1.preheader.lr.ph: 
+  %cmp218 = icmp sgt i32 %M, 1
+  %gi.promoted = load i32, i32* @gi
+  %0 = add i32 %M, -1
+  %1 = sext i32 %gi.promoted to i64
+  %2 = sext i32 %N to i64
+  %3 = add i32 %gi.promoted, 1
+  %4 = icmp slt i32 %3, %N
+  %smax = select i1 %4, i32 %N, i32 %3
+  br label %for.cond1.preheader
+
+for.cond1.preheader: 
+  %indvars.iv25 = phi i64 [ %1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next26, %for.inc14 ]
+  br i1 %cmp218, label %for.body3, label %for.inc14
+
+for.body3: 
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.cond1.preheader ]
+  %5 = add nsw i64 %indvars.iv, -1
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %5, i64 %indvars.iv25
+  %6 = load i32, i32* %arrayidx5
+  %arrayidx9 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @C, i64 0, i64 %indvars.iv, i64 %indvars.iv25
+  %7 = load i32, i32* %arrayidx9
+  %add = add nsw i32 %7, %6
+  %arrayidx13 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv25
+  store i32 %add, i32* %arrayidx13
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc14, label %for.body3
+
+for.inc14: 
+  %inc.lcssa23 = phi i32 [ 1, %for.cond1.preheader ], [ %M, %for.body3 ]
+  %indvars.iv.next26 = add nsw i64 %indvars.iv25, 1
+  %cmp = icmp slt i64 %indvars.iv.next26, %2
+  br i1 %cmp, label %for.cond1.preheader, label %for.cond.for.end16_crit_edge
+
+for.cond.for.end16_crit_edge: 
+  store i32 %inc.lcssa23, i32* @gj
+  store i32 %smax, i32* @gi
+  br label %for.end16
+
+for.end16:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_07
+; CHECK: for.body3:                                        ; preds = %for.body3.preheader, %for.body3
+; CHECK:   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.body3.preheader ]
+; CHECK:   %5 = add nsw i64 %indvars.iv, -1
+; CHECK:   %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %5, i64 %indvars.iv25
+; CHECK:   %6 = load i32, i32* %arrayidx5
+; CHECK:   %arrayidx9 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @C, i64 0, i64 %indvars.iv, i64 %indvars.iv25
+
+;;------------------------------------------------Test case 08-------------------------------
+;; Test for interchange in loop nest greater than 2.
+;;  for(int i=0;i<100;i++)
+;;    for(int j=0;j<100;j++)
+;;      for(int k=0;k<100;k++)
+;;        D[i][k][j] = D[i][k][j]+t;
+
+define void @interchange_08(i32 %t){
+entry:
+  br label %for.cond1.preheader
+
+for.cond1.preheader:                              ; preds = %for.inc15, %entry
+  %i.028 = phi i32 [ 0, %entry ], [ %inc16, %for.inc15 ]
+  br label %for.cond4.preheader
+
+for.cond4.preheader:                              ; preds = %for.inc12, %for.cond1.preheader
+  %j.027 = phi i32 [ 0, %for.cond1.preheader ], [ %inc13, %for.inc12 ]
+  br label %for.body6
+
+for.body6:                                        ; preds = %for.body6, %for.cond4.preheader
+  %k.026 = phi i32 [ 0, %for.cond4.preheader ], [ %inc, %for.body6 ]
+  %arrayidx8 = getelementptr inbounds [100 x [100 x [100 x i32]]], [100 x [100 x [100 x i32]]]* @D, i32 0, i32 %i.028, i32 %k.026, i32 %j.027
+  %0 = load i32, i32* %arrayidx8
+  %add = add nsw i32 %0, %t
+  store i32 %add, i32* %arrayidx8
+  %inc = add nuw nsw i32 %k.026, 1
+  %exitcond = icmp eq i32 %inc, 100
+  br i1 %exitcond, label %for.inc12, label %for.body6
+
+for.inc12:                                        ; preds = %for.body6
+  %inc13 = add nuw nsw i32 %j.027, 1
+  %exitcond29 = icmp eq i32 %inc13, 100
+  br i1 %exitcond29, label %for.inc15, label %for.cond4.preheader
+
+for.inc15:                                        ; preds = %for.inc12
+  %inc16 = add nuw nsw i32 %i.028, 1
+  %exitcond30 = icmp eq i32 %inc16, 100
+  br i1 %exitcond30, label %for.end17, label %for.cond1.preheader
+
+for.end17:                                        ; preds = %for.inc15
+  ret void
+}
+; CHECK-LABEL: @interchange_08
+; CHECK:   entry:
+; CHECK:     br label %for.cond1.preheader.preheader
+; CHECK:   for.cond1.preheader.preheader:                    ; preds = %entry
+; CHECK:     br label %for.cond1.preheader
+; CHECK:   for.cond1.preheader:                              ; preds = %for.cond1.preheader.preheader, %for.inc15
+; CHECK:     %i.028 = phi i32 [ %inc16, %for.inc15 ], [ 0, %for.cond1.preheader.preheader ]
+; CHECK:     br label %for.body6.preheader
+; CHECK:   for.cond4.preheader.preheader:                    ; preds = %for.body6
+; CHECK:     br label %for.cond4.preheader
+; CHECK:   for.cond4.preheader:                              ; preds = %for.cond4.preheader.preheader, %for.inc12
+; CHECK:     %j.027 = phi i32 [ %inc13, %for.inc12 ], [ 0, %for.cond4.preheader.preheader ]
+; CHECK:     br label %for.body6.split1
+; CHECK:   for.body6.preheader:                              ; preds = %for.cond1.preheader
+; CHECK:     br label %for.body6
+; CHECK:   for.body6:                                        ; preds = %for.body6.preheader, %for.body6.split
+; CHECK:     %k.026 = phi i32 [ %inc, %for.body6.split ], [ 0, %for.body6.preheader ]
+; CHECK:     br label %for.cond4.preheader.preheader
+; CHECK:   for.body6.split1:                                 ; preds = %for.cond4.preheader
+; CHECK:     %arrayidx8 = getelementptr inbounds [100 x [100 x [100 x i32]]], [100 x [100 x [100 x i32]]]* @D, i32 0, i32 %i.028, i32 %k.026, i32 %j.027
+; CHECK:     %0 = load i32, i32* %arrayidx8
+; CHECK:     %add = add nsw i32 %0, %t
+; CHECK:     store i32 %add, i32* %arrayidx8
+; CHECK:     br label %for.inc12
+; CHECK:   for.body6.split:                                  ; preds = %for.inc12
+; CHECK:     %inc = add nuw nsw i32 %k.026, 1
+; CHECK:     %exitcond = icmp eq i32 %inc, 100
+; CHECK:     br i1 %exitcond, label %for.inc15, label %for.body6
+; CHECK:   for.inc12:                                        ; preds = %for.body6.split1
+; CHECK:     %inc13 = add nuw nsw i32 %j.027, 1
+; CHECK:     %exitcond29 = icmp eq i32 %inc13, 100
+; CHECK:     br i1 %exitcond29, label %for.body6.split, label %for.cond4.preheader
+; CHECK:   for.inc15:                                        ; preds = %for.body6.split
+; CHECK:     %inc16 = add nuw nsw i32 %i.028, 1
+; CHECK:     %exitcond30 = icmp eq i32 %inc16, 100
+; CHECK:     br i1 %exitcond30, label %for.end17, label %for.cond1.preheader
+; CHECK:   for.end17:                                        ; preds = %for.inc15
+; CHECK:     ret void
+
diff --git a/test/Transforms/LoopInterchange/profitability.ll b/test/Transforms/LoopInterchange/profitability.ll
new file mode 100644
index 000000000000..ae1f79314486
--- /dev/null
+++ b/test/Transforms/LoopInterchange/profitability.ll
@@ -0,0 +1,205 @@
+; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
+;; We test profitability model in these test cases.
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+@A = common global [100 x [100 x i32]] zeroinitializer
+@B = common global [100 x [100 x i32]] zeroinitializer
+
+;;---------------------------------------Test case 01---------------------------------
+;; Loops interchange will result in code vectorization and hence profitable. Check for interchange.
+;;   for(int i=1;i<N;i++)
+;;     for(int j=1;j<N;j++)
+;;       A[j][i] = A[j - 1][i] + B[j][i];
+
+define void @interchange_01(i32 %N) {
+entry:
+  %cmp27 = icmp sgt i32 %N, 1
+  br i1 %cmp27, label %for.cond1.preheader.lr.ph, label %for.end16
+
+for.cond1.preheader.lr.ph:
+  %0 = add i32 %N, -1
+  br label %for.body3.preheader
+
+for.body3.preheader:
+  %indvars.iv30 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next31, %for.inc14 ]
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.body3.preheader ]
+  %1 = add nsw i64 %indvars.iv, -1
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %1, i64 %indvars.iv30
+  %2 = load i32, i32* %arrayidx5
+  %arrayidx9 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %indvars.iv, i64 %indvars.iv30
+  %3 = load i32, i32* %arrayidx9
+  %add = add nsw i32 %3, %2
+  %arrayidx13 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv30
+  store i32 %add, i32* %arrayidx13
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc14, label %for.body3
+
+for.inc14:
+  %indvars.iv.next31 = add nuw nsw i64 %indvars.iv30, 1
+  %lftr.wideiv32 = trunc i64 %indvars.iv30 to i32
+  %exitcond33 = icmp eq i32 %lftr.wideiv32, %0
+  br i1 %exitcond33, label %for.end16, label %for.body3.preheader
+
+for.end16:
+  ret void
+}
+;; Here we are checking partial .ll to check if loop are interchanged.
+; CHECK-LABEL: @interchange_01
+; CHECK:  for.body3.preheader:                              ; preds = %for.inc14, %for.cond1.preheader.lr.ph
+; CHECK:    %indvars.iv30 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next31, %for.inc14 ]
+; CHECK:    br label %for.body3.split2
+
+; CHECK:  for.body3.preheader1:                             ; preds = %entry
+; CHECK:    br label %for.body3
+
+; CHECK:  for.body3:                                        ; preds = %for.body3.preheader1, %for.body3.split
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 1, %for.body3.preheader1 ]
+; CHECK:    br label %for.cond1.preheader.lr.ph
+
+; CHECK:  for.body3.split2:                                 ; preds = %for.body3.preheader
+; CHECK:    %1 = add nsw i64 %indvars.iv, -1
+; CHECK:    %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %1, i64 %indvars.iv30
+; CHECK:    %2 = load i32, i32* %arrayidx5
+; CHECK:    %arrayidx9 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %indvars.iv, i64 %indvars.iv30
+; CHECK:    %3 = load i32, i32* %arrayidx9
+; CHECK:    %add = add nsw i32 %3, %2
+; CHECK:    %arrayidx13 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv30
+; CHECK:    store i32 %add, i32* %arrayidx13
+; CHECK:    br label %for.inc14
+
+
+;; ---------------------------------------Test case 02---------------------------------
+;; Check loop interchange profitability model. 
+;; This tests profitability model when operands of getelementpointer and not exactly the induction variable but some 
+;; arithmetic operation on them.
+;;   for(int i=1;i<N;i++)
+;;    for(int j=1;j<N;j++)
+;;       A[j-1][i-1] = A[j - 1][i-1] + B[j-1][i-1];
+
+define void @interchange_02(i32 %N) {
+entry:
+  %cmp32 = icmp sgt i32 %N, 1
+  br i1 %cmp32, label %for.cond1.preheader.lr.ph, label %for.end21
+
+for.cond1.preheader.lr.ph:
+  %0 = add i32 %N, -1
+  br label %for.body3.lr.ph
+
+for.body3.lr.ph:
+  %indvars.iv35 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next36, %for.inc19 ]
+  %1 = add nsw i64 %indvars.iv35, -1
+  br label %for.body3
+
+for.body3: 
+  %indvars.iv = phi i64 [ 1, %for.body3.lr.ph ], [ %indvars.iv.next, %for.body3 ]
+  %2 = add nsw i64 %indvars.iv, -1
+  %arrayidx6 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %2, i64 %1
+  %3 = load i32, i32* %arrayidx6
+  %arrayidx12 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %2, i64 %1
+  %4 = load i32, i32* %arrayidx12
+  %add = add nsw i32 %4, %3
+  store i32 %add, i32* %arrayidx6
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc19, label %for.body3
+
+for.inc19:
+  %indvars.iv.next36 = add nuw nsw i64 %indvars.iv35, 1
+  %lftr.wideiv38 = trunc i64 %indvars.iv35 to i32
+  %exitcond39 = icmp eq i32 %lftr.wideiv38, %0
+  br i1 %exitcond39, label %for.end21, label %for.body3.lr.ph
+
+for.end21:
+  ret void
+}
+; CHECK-LABEL: @interchange_02
+; CHECK:  for.body3.lr.ph:                                  ; preds = %for.inc19, %for.cond1.preheader.lr.ph
+; CHECK:    %indvars.iv35 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next36, %for.inc19 ]
+; CHECK:    %0 = add nsw i64 %indvars.iv35, -1
+; CHECK:    br label %for.body3.split1
+
+; CHECK:  for.body3.preheader:                              ; preds = %entry
+; CHECK:    %1 = add i32 %N, -1
+; CHECK:    br label %for.body3
+
+; CHECK:  for.body3:                                        ; preds = %for.body3.preheader, %for.body3.split
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 1, %for.body3.preheader ]
+; CHECK:    br label %for.cond1.preheader.lr.ph
+
+; CHECK:  for.body3.split1:                                 ; preds = %for.body3.lr.ph
+; CHECK:    %2 = add nsw i64 %indvars.iv, -1
+; CHECK:    %arrayidx6 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %2, i64 %0
+; CHECK:    %3 = load i32, i32* %arrayidx6
+; CHECK:    %arrayidx12 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %2, i64 %0
+; CHECK:    %4 = load i32, i32* %arrayidx12
+; CHECK:    %add = add nsw i32 %4, %3
+; CHECK:    store i32 %add, i32* %arrayidx6
+; CHECK:    br label %for.inc19
+
+
+;;---------------------------------------Test case 03---------------------------------
+;; Loops interchange is not profitable.
+;;   for(int i=1;i<N;i++)
+;;     for(int j=1;j<N;j++)
+;;       A[i-1][j-1] = A[i - 1][j-1] + B[i][j];
+
+define void @interchange_03(i32 %N){
+entry:
+  %cmp31 = icmp sgt i32 %N, 1
+  br i1 %cmp31, label %for.cond1.preheader.lr.ph, label %for.end19
+
+for.cond1.preheader.lr.ph:
+  %0 = add i32 %N, -1
+  br label %for.body3.lr.ph
+
+for.body3.lr.ph:
+  %indvars.iv34 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next35, %for.inc17 ]
+  %1 = add nsw i64 %indvars.iv34, -1
+  br label %for.body3
+
+for.body3:
+  %indvars.iv = phi i64 [ 1, %for.body3.lr.ph ], [ %indvars.iv.next, %for.body3 ]
+  %2 = add nsw i64 %indvars.iv, -1
+  %arrayidx6 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %1, i64 %2
+  %3 = load i32, i32* %arrayidx6
+  %arrayidx10 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %indvars.iv34, i64 %indvars.iv
+  %4 = load i32, i32* %arrayidx10
+  %add = add nsw i32 %4, %3
+  store i32 %add, i32* %arrayidx6
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.inc17, label %for.body3
+
+for.inc17:
+  %indvars.iv.next35 = add nuw nsw i64 %indvars.iv34, 1
+  %lftr.wideiv37 = trunc i64 %indvars.iv34 to i32
+  %exitcond38 = icmp eq i32 %lftr.wideiv37, %0
+  br i1 %exitcond38, label %for.end19, label %for.body3.lr.ph
+
+for.end19:
+  ret void
+}
+
+; CHECK-LABEL: @interchange_03
+; CHECK:  for.body3.lr.ph:
+; CHECK:    %indvars.iv34 = phi i64 [ 1, %for.cond1.preheader.lr.ph ], [ %indvars.iv.next35, %for.inc17 ]
+; CHECK:    %1 = add nsw i64 %indvars.iv34, -1
+; CHECK:    br label %for.body3.preheader
+; CHECK:  for.body3.preheader:
+; CHECK:    br label %for.body3
+; CHECK:  for.body3:
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 1, %for.body3.preheader ]
+; CHECK:    %2 = add nsw i64 %indvars.iv, -1
+; CHECK:    %arrayidx6 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @A, i64 0, i64 %1, i64 %2
+; CHECK:    %3 = load i32, i32* %arrayidx6
+; CHECK:    %arrayidx10 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* @B, i64 0, i64 %indvars.iv34, i64 %indvars.iv
+; CHECK:    %4 = load i32, i32* %arrayidx10
diff --git a/test/Transforms/LoopInterchange/reductions.ll b/test/Transforms/LoopInterchange/reductions.ll
new file mode 100644
index 000000000000..86a44da7cf20
--- /dev/null
+++ b/test/Transforms/LoopInterchange/reductions.ll
@@ -0,0 +1,235 @@
+; RUN: opt < %s -basicaa -loop-interchange -S | FileCheck %s
+
+@A = common global [500 x [500 x i32]] zeroinitializer
+@X = common global i32 0
+@B = common global [500 x [500 x i32]] zeroinitializer
+@Y = common global i32 0
+
+;;  for( int i=1;i<N;i++)
+;;    for( int j=1;j<N;j++)
+;;      X+=A[j][i];
+
+define void @reduction_01(i32 %N) {
+entry:
+  %cmp16 = icmp sgt i32 %N, 1
+  br i1 %cmp16, label %for.body3.lr.ph, label %for.end8
+
+for.body3.lr.ph:                                  ; preds = %entry, %for.cond1.for.inc6_crit_edge
+  %indvars.iv18 = phi i64 [ %indvars.iv.next19, %for.cond1.for.inc6_crit_edge ], [ 1, %entry ]
+  %X.promoted = load i32, i32* @X
+  br label %for.body3
+
+for.body3:                                        ; preds = %for.body3, %for.body3.lr.ph
+  %indvars.iv = phi i64 [ 1, %for.body3.lr.ph ], [ %indvars.iv.next, %for.body3 ]
+  %add15 = phi i32 [ %X.promoted, %for.body3.lr.ph ], [ %add, %for.body3 ]
+  %arrayidx5 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv18
+  %0 = load i32, i32* %arrayidx5
+  %add = add nsw i32 %add15, %0
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %N
+  br i1 %exitcond, label %for.cond1.for.inc6_crit_edge, label %for.body3
+
+for.cond1.for.inc6_crit_edge:                     ; preds = %for.body3
+  store i32 %add, i32* @X
+  %indvars.iv.next19 = add nuw nsw i64 %indvars.iv18, 1
+  %lftr.wideiv20 = trunc i64 %indvars.iv.next19 to i32
+  %exitcond21 = icmp eq i32 %lftr.wideiv20, %N
+  br i1 %exitcond21, label %for.end8, label %for.body3.lr.ph
+
+for.end8:                                         ; preds = %for.cond1.for.inc6_crit_edge, %entry
+  ret void
+}
+
+;; Loop is interchanged check that the phi nodes are split and the promoted value is used instead of the reduction phi.
+; CHECK-LABEL: @reduction_01
+; CHECK:  for.body3:                                        ; preds = %for.body3.preheader, %for.body3.split
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3.split ], [ 1, %for.body3.preheader ]
+; CHECK:    br label %for.body3.lr.ph.preheader
+; CHECK:    %add = add nsw i32 %X.promoted
+
+
+;; Test for more than 1 reductions inside a loop.
+;;  for( int i=1;i<N;i++)
+;;    for( int j=1;j<N;j++)
+;;      for( int k=1;k<N;k++) {
+;;        X+=A[k][j];
+;;        Y+=B[k][i];
+;;      }
+
+define void @reduction_02(i32 %N)  {
+entry:
+  %cmp34 = icmp sgt i32 %N, 1
+  br i1 %cmp34, label %for.cond4.preheader.preheader, label %for.end19
+
+for.cond4.preheader.preheader:                    ; preds = %entry, %for.inc17
+  %indvars.iv40 = phi i64 [ %indvars.iv.next41, %for.inc17 ], [ 1, %entry ]
+  br label %for.body6.lr.ph
+
+for.body6.lr.ph:                                  ; preds = %for.cond4.for.inc14_crit_edge, %for.cond4.preheader.preheader
+  %indvars.iv36 = phi i64 [ %indvars.iv.next37, %for.cond4.for.inc14_crit_edge ], [ 1, %for.cond4.preheader.preheader ]
+  %X.promoted = load i32, i32* @X
+  %Y.promoted = load i32, i32* @Y
+  br label %for.body6
+
+for.body6:                                        ; preds = %for.body6, %for.body6.lr.ph
+  %indvars.iv = phi i64 [ 1, %for.body6.lr.ph ], [ %indvars.iv.next, %for.body6 ]
+  %add1331 = phi i32 [ %Y.promoted, %for.body6.lr.ph ], [ %add13, %for.body6 ]
+  %add30 = phi i32 [ %X.promoted, %for.body6.lr.ph ], [ %add, %for.body6 ]
+  %arrayidx8 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv36
+  %0 = load i32, i32* %arrayidx8
+  %add = add nsw i32 %add30, %0
+  %arrayidx12 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @B, i64 0, i64 %indvars.iv, i64 %indvars.iv40
+  %1 = load i32, i32* %arrayidx12
+  %add13 = add nsw i32 %add1331, %1
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %N
+  br i1 %exitcond, label %for.cond4.for.inc14_crit_edge, label %for.body6
+
+for.cond4.for.inc14_crit_edge:                    ; preds = %for.body6
+  store i32 %add, i32* @X
+  store i32 %add13, i32* @Y
+  %indvars.iv.next37 = add nuw nsw i64 %indvars.iv36, 1
+  %lftr.wideiv38 = trunc i64 %indvars.iv.next37 to i32
+  %exitcond39 = icmp eq i32 %lftr.wideiv38, %N
+  br i1 %exitcond39, label %for.inc17, label %for.body6.lr.ph
+
+for.inc17:                                        ; preds = %for.cond4.for.inc14_crit_edge
+  %indvars.iv.next41 = add nuw nsw i64 %indvars.iv40, 1
+  %lftr.wideiv42 = trunc i64 %indvars.iv.next41 to i32
+  %exitcond43 = icmp eq i32 %lftr.wideiv42, %N
+  br i1 %exitcond43, label %for.end19, label %for.cond4.preheader.preheader
+
+for.end19:                                        ; preds = %for.inc17, %entry
+  ret void
+}
+
+;; Loop is interchanged check that the phi nodes are split and the promoted value is used instead of the reduction phi.
+; CHECK-LABEL: @reduction_02
+; CHECK:  for.body6:                                        ; preds = %for.body6.preheader, %for.body6.split
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body6.split ], [ 1, %for.body6.preheader ]
+; CHECK:    br label %for.cond4.preheader.preheader.preheader
+; CHECK:    %add13 = add nsw i32 %Y.promoted
+
+
+;; Not tightly nested. Do not interchange.
+;;  for( int i=1;i<N;i++)
+;;    for( int j=1;j<N;j++) {
+;;      for( int k=1;k<N;k++) {
+;;        X+=A[k][j];
+;;      }
+;;      Y+=B[j][i];
+;;    }
+define void @reduction_03(i32 %N)  {
+entry:
+  %cmp35 = icmp sgt i32 %N, 1
+  br i1 %cmp35, label %for.cond4.preheader.lr.ph, label %for.end19
+
+for.cond4.preheader.lr.ph:                        ; preds = %entry, %for.cond1.for.inc17_crit_edge
+  %indvars.iv41 = phi i64 [ %indvars.iv.next42, %for.cond1.for.inc17_crit_edge ], [ 1, %entry ]
+  %Y.promoted = load i32, i32* @Y
+  br label %for.body6.lr.ph
+
+for.body6.lr.ph:                                  ; preds = %for.cond4.preheader.lr.ph, %for.cond4.for.end_crit_edge
+  %indvars.iv37 = phi i64 [ 1, %for.cond4.preheader.lr.ph ], [ %indvars.iv.next38, %for.cond4.for.end_crit_edge ]
+  %add1334 = phi i32 [ %Y.promoted, %for.cond4.preheader.lr.ph ], [ %add13, %for.cond4.for.end_crit_edge ]
+  %X.promoted = load i32, i32* @X
+  br label %for.body6
+
+for.body6:                                        ; preds = %for.body6, %for.body6.lr.ph
+  %indvars.iv = phi i64 [ 1, %for.body6.lr.ph ], [ %indvars.iv.next, %for.body6 ]
+  %add31 = phi i32 [ %X.promoted, %for.body6.lr.ph ], [ %add, %for.body6 ]
+  %arrayidx8 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv37
+  %0 = load i32, i32* %arrayidx8
+  %add = add nsw i32 %add31, %0
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %N
+  br i1 %exitcond, label %for.cond4.for.end_crit_edge, label %for.body6
+
+for.cond4.for.end_crit_edge:                      ; preds = %for.body6
+  store i32 %add, i32* @X
+  %arrayidx12 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @B, i64 0, i64 %indvars.iv37, i64 %indvars.iv41
+  %1 = load i32, i32* %arrayidx12
+  %add13 = add nsw i32 %add1334, %1
+  %indvars.iv.next38 = add nuw nsw i64 %indvars.iv37, 1
+  %lftr.wideiv39 = trunc i64 %indvars.iv.next38 to i32
+  %exitcond40 = icmp eq i32 %lftr.wideiv39, %N
+  br i1 %exitcond40, label %for.cond1.for.inc17_crit_edge, label %for.body6.lr.ph
+
+for.cond1.for.inc17_crit_edge:                    ; preds = %for.cond4.for.end_crit_edge
+  store i32 %add13, i32* @Y
+  %indvars.iv.next42 = add nuw nsw i64 %indvars.iv41, 1
+  %lftr.wideiv43 = trunc i64 %indvars.iv.next42 to i32
+  %exitcond44 = icmp eq i32 %lftr.wideiv43, %N
+  br i1 %exitcond44, label %for.end19, label %for.cond4.preheader.lr.ph
+
+for.end19:                                        ; preds = %for.cond1.for.inc17_crit_edge, %entry
+  ret void
+}
+;; Not tightly nested. Do not interchange.
+;; Not interchanged hence the phi's in the inner loop will not be split. Check for the same.
+; CHECK-LABEL: @reduction_03
+; CHECK:  for.body6:                                        ; preds = %for.body6.preheader, %for.body6
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body6 ], [ 1, %for.body6.preheader ]
+; CHECK:    %add31 = phi i32 [ %add, %for.body6 ], [ %X.promoted, %for.body6.preheader ]
+
+
+
+;; Multiple use of reduction not safe. Do not interchange.
+;;  for( int i=1;i<N;i++)
+;;    for( int j=1;j<N;j++)
+;;      for( int k=1;k<N;k++) {
+;;        X+=A[k][j];
+;;        Y+=X;
+;;      }
+define void @reduction_04(i32 %N) {
+entry:
+  %cmp28 = icmp sgt i32 %N, 1
+  br i1 %cmp28, label %for.cond4.preheader.preheader, label %for.end15
+
+for.cond4.preheader.preheader:                    ; preds = %entry, %for.inc13
+  %i.029 = phi i32 [ %inc14, %for.inc13 ], [ 1, %entry ]
+  br label %for.body6.lr.ph
+
+for.body6.lr.ph:                                  ; preds = %for.cond4.for.inc10_crit_edge, %for.cond4.preheader.preheader
+  %indvars.iv30 = phi i64 [ %indvars.iv.next31, %for.cond4.for.inc10_crit_edge ], [ 1, %for.cond4.preheader.preheader ]
+  %X.promoted = load i32, i32* @X
+  %Y.promoted = load i32, i32* @Y
+  br label %for.body6
+
+for.body6:                                        ; preds = %for.body6, %for.body6.lr.ph
+  %indvars.iv = phi i64 [ 1, %for.body6.lr.ph ], [ %indvars.iv.next, %for.body6 ]
+  %add925 = phi i32 [ %Y.promoted, %for.body6.lr.ph ], [ %add9, %for.body6 ]
+  %add24 = phi i32 [ %X.promoted, %for.body6.lr.ph ], [ %add, %for.body6 ]
+  %arrayidx8 = getelementptr inbounds [500 x [500 x i32]], [500 x [500 x i32]]* @A, i64 0, i64 %indvars.iv, i64 %indvars.iv30
+  %0 = load i32, i32* %arrayidx8
+  %add = add nsw i32 %add24, %0
+  %add9 = add nsw i32 %add925, %add
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %N
+  br i1 %exitcond, label %for.cond4.for.inc10_crit_edge, label %for.body6
+
+for.cond4.for.inc10_crit_edge:                    ; preds = %for.body6
+  store i32 %add, i32* @X
+  store i32 %add9, i32* @Y
+  %indvars.iv.next31 = add nuw nsw i64 %indvars.iv30, 1
+  %lftr.wideiv32 = trunc i64 %indvars.iv.next31 to i32
+  %exitcond33 = icmp eq i32 %lftr.wideiv32, %N
+  br i1 %exitcond33, label %for.inc13, label %for.body6.lr.ph
+
+for.inc13:                                        ; preds = %for.cond4.for.inc10_crit_edge
+  %inc14 = add nuw nsw i32 %i.029, 1
+  %exitcond34 = icmp eq i32 %inc14, %N
+  br i1 %exitcond34, label %for.end15, label %for.cond4.preheader.preheader
+
+for.end15:                                        ; preds = %for.inc13, %entry
+  ret void
+}
+;; Not interchanged hence the phi's in the inner loop will not be split. Check for the same.
+; CHECK-LABEL: @reduction_04
+; CHECK:  for.body6:                                        ; preds = %for.body6.preheader, %for.body6
+; CHECK:    %indvars.iv = phi i64 [ %indvars.iv.next, %for.body6 ], [ 1, %for.body6.preheader ]
+; CHECK:    %add925 = phi i32 [ %add9, %for.body6 ], [ %Y.promoted, %for.body6.preheader ]
diff --git a/test/Transforms/LoopReroll/basic.ll b/test/Transforms/LoopReroll/basic.ll
index 3bd6d7ae02e1..16a6dc81af79 100644
--- a/test/Transforms/LoopReroll/basic.ll
+++ b/test/Transforms/LoopReroll/basic.ll
@@ -60,15 +60,15 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %call = tail call i32 @foo(i32 0) #1
-  %arrayidx = getelementptr inbounds i32* %x, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
   store i32 %call, i32* %arrayidx, align 4
   %call1 = tail call i32 @foo(i32 0) #1
   %0 = add nsw i64 %indvars.iv, 1
-  %arrayidx3 = getelementptr inbounds i32* %x, i64 %0
+  %arrayidx3 = getelementptr inbounds i32, i32* %x, i64 %0
   store i32 %call1, i32* %arrayidx3, align 4
   %call4 = tail call i32 @foo(i32 0) #1
   %1 = add nsw i64 %indvars.iv, 2
-  %arrayidx7 = getelementptr inbounds i32* %x, i64 %1
+  %arrayidx7 = getelementptr inbounds i32, i32* %x, i64 %1
   store i32 %call4, i32* %arrayidx7, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 3
   %2 = trunc i64 %indvars.iv.next to i32
@@ -80,7 +80,7 @@ for.body:                                         ; preds = %entry, %for.body
 ; CHECK: for.body:
 ; CHECK: %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
 ; CHECK: %call = tail call i32 @foo(i32 0) #1
-; CHECK: %arrayidx = getelementptr inbounds i32* %x, i64 %indvar
+; CHECK: %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvar
 ; CHECK: store i32 %call, i32* %arrayidx, align 4
 ; CHECK: %indvar.next = add i64 %indvar, 1
 ; CHECK: %exitcond = icmp eq i64 %indvar, 1499
@@ -109,15 +109,15 @@ for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %call = tail call i32 @foo(i32 0) #1
   %0 = mul nsw i64 %indvars.iv, 3
-  %arrayidx = getelementptr inbounds i32* %x, i64 %0
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %0
   store i32 %call, i32* %arrayidx, align 4
   %call1 = tail call i32 @foo(i32 0) #1
   %1 = add nsw i64 %0, 1
-  %arrayidx4 = getelementptr inbounds i32* %x, i64 %1
+  %arrayidx4 = getelementptr inbounds i32, i32* %x, i64 %1
   store i32 %call1, i32* %arrayidx4, align 4
   %call5 = tail call i32 @foo(i32 0) #1
   %2 = add nsw i64 %0, 2
-  %arrayidx9 = getelementptr inbounds i32* %x, i64 %2
+  %arrayidx9 = getelementptr inbounds i32, i32* %x, i64 %2
   store i32 %call5, i32* %arrayidx9, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 500
@@ -128,7 +128,7 @@ for.body:                                         ; preds = %for.body, %entry
 ; CHECK: for.body:
 ; CHECK: %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
 ; CHECK: %call = tail call i32 @foo(i32 0) #1
-; CHECK: %arrayidx = getelementptr inbounds i32* %x, i64 %indvars.iv
+; CHECK: %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
 ; CHECK: store i32 %call, i32* %arrayidx, align 4
 ; CHECK: %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
 ; CHECK: %exitcond1 = icmp eq i64 %indvars.iv, 1499
@@ -157,43 +157,43 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %mul = fmul float %0, %alpha
-  %arrayidx2 = getelementptr inbounds float* %a, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %add = fadd float %1, %mul
   store float %add, float* %arrayidx2, align 4
   %2 = add nsw i64 %indvars.iv, 1
-  %arrayidx5 = getelementptr inbounds float* %b, i64 %2
-  %3 = load float* %arrayidx5, align 4
+  %arrayidx5 = getelementptr inbounds float, float* %b, i64 %2
+  %3 = load float, float* %arrayidx5, align 4
   %mul6 = fmul float %3, %alpha
-  %arrayidx9 = getelementptr inbounds float* %a, i64 %2
-  %4 = load float* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds float, float* %a, i64 %2
+  %4 = load float, float* %arrayidx9, align 4
   %add10 = fadd float %4, %mul6
   store float %add10, float* %arrayidx9, align 4
   %5 = add nsw i64 %indvars.iv, 2
-  %arrayidx13 = getelementptr inbounds float* %b, i64 %5
-  %6 = load float* %arrayidx13, align 4
+  %arrayidx13 = getelementptr inbounds float, float* %b, i64 %5
+  %6 = load float, float* %arrayidx13, align 4
   %mul14 = fmul float %6, %alpha
-  %arrayidx17 = getelementptr inbounds float* %a, i64 %5
-  %7 = load float* %arrayidx17, align 4
+  %arrayidx17 = getelementptr inbounds float, float* %a, i64 %5
+  %7 = load float, float* %arrayidx17, align 4
   %add18 = fadd float %7, %mul14
   store float %add18, float* %arrayidx17, align 4
   %8 = add nsw i64 %indvars.iv, 3
-  %arrayidx21 = getelementptr inbounds float* %b, i64 %8
-  %9 = load float* %arrayidx21, align 4
+  %arrayidx21 = getelementptr inbounds float, float* %b, i64 %8
+  %9 = load float, float* %arrayidx21, align 4
   %mul22 = fmul float %9, %alpha
-  %arrayidx25 = getelementptr inbounds float* %a, i64 %8
-  %10 = load float* %arrayidx25, align 4
+  %arrayidx25 = getelementptr inbounds float, float* %a, i64 %8
+  %10 = load float, float* %arrayidx25, align 4
   %add26 = fadd float %10, %mul22
   store float %add26, float* %arrayidx25, align 4
   %11 = add nsw i64 %indvars.iv, 4
-  %arrayidx29 = getelementptr inbounds float* %b, i64 %11
-  %12 = load float* %arrayidx29, align 4
+  %arrayidx29 = getelementptr inbounds float, float* %b, i64 %11
+  %12 = load float, float* %arrayidx29, align 4
   %mul30 = fmul float %12, %alpha
-  %arrayidx33 = getelementptr inbounds float* %a, i64 %11
-  %13 = load float* %arrayidx33, align 4
+  %arrayidx33 = getelementptr inbounds float, float* %a, i64 %11
+  %13 = load float, float* %arrayidx33, align 4
   %add34 = fadd float %13, %mul30
   store float %add34, float* %arrayidx33, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 5
@@ -205,11 +205,11 @@ for.body:                                         ; preds = %entry, %for.body
 
 ; CHECK: for.body:
 ; CHECK: %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
-; CHECK: %arrayidx = getelementptr inbounds float* %b, i64 %indvar
-; CHECK: %0 = load float* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr inbounds float, float* %b, i64 %indvar
+; CHECK: %0 = load float, float* %arrayidx, align 4
 ; CHECK: %mul = fmul float %0, %alpha
-; CHECK: %arrayidx2 = getelementptr inbounds float* %a, i64 %indvar
-; CHECK: %1 = load float* %arrayidx2, align 4
+; CHECK: %arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvar
+; CHECK: %1 = load float, float* %arrayidx2, align 4
 ; CHECK: %add = fadd float %1, %mul
 ; CHECK: store float %add, float* %arrayidx2, align 4
 ; CHECK: %indvar.next = add i64 %indvar, 1
@@ -239,58 +239,58 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %ip, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %ip, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %idxprom1 = sext i32 %0 to i64
-  %arrayidx2 = getelementptr inbounds float* %b, i64 %idxprom1
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %b, i64 %idxprom1
+  %1 = load float, float* %arrayidx2, align 4
   %mul = fmul float %1, %alpha
-  %arrayidx4 = getelementptr inbounds float* %a, i64 %indvars.iv
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %2 = load float, float* %arrayidx4, align 4
   %add = fadd float %2, %mul
   store float %add, float* %arrayidx4, align 4
   %3 = add nsw i64 %indvars.iv, 1
-  %arrayidx7 = getelementptr inbounds i32* %ip, i64 %3
-  %4 = load i32* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %ip, i64 %3
+  %4 = load i32, i32* %arrayidx7, align 4
   %idxprom8 = sext i32 %4 to i64
-  %arrayidx9 = getelementptr inbounds float* %b, i64 %idxprom8
-  %5 = load float* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds float, float* %b, i64 %idxprom8
+  %5 = load float, float* %arrayidx9, align 4
   %mul10 = fmul float %5, %alpha
-  %arrayidx13 = getelementptr inbounds float* %a, i64 %3
-  %6 = load float* %arrayidx13, align 4
+  %arrayidx13 = getelementptr inbounds float, float* %a, i64 %3
+  %6 = load float, float* %arrayidx13, align 4
   %add14 = fadd float %6, %mul10
   store float %add14, float* %arrayidx13, align 4
   %7 = add nsw i64 %indvars.iv, 2
-  %arrayidx17 = getelementptr inbounds i32* %ip, i64 %7
-  %8 = load i32* %arrayidx17, align 4
+  %arrayidx17 = getelementptr inbounds i32, i32* %ip, i64 %7
+  %8 = load i32, i32* %arrayidx17, align 4
   %idxprom18 = sext i32 %8 to i64
-  %arrayidx19 = getelementptr inbounds float* %b, i64 %idxprom18
-  %9 = load float* %arrayidx19, align 4
+  %arrayidx19 = getelementptr inbounds float, float* %b, i64 %idxprom18
+  %9 = load float, float* %arrayidx19, align 4
   %mul20 = fmul float %9, %alpha
-  %arrayidx23 = getelementptr inbounds float* %a, i64 %7
-  %10 = load float* %arrayidx23, align 4
+  %arrayidx23 = getelementptr inbounds float, float* %a, i64 %7
+  %10 = load float, float* %arrayidx23, align 4
   %add24 = fadd float %10, %mul20
   store float %add24, float* %arrayidx23, align 4
   %11 = add nsw i64 %indvars.iv, 3
-  %arrayidx27 = getelementptr inbounds i32* %ip, i64 %11
-  %12 = load i32* %arrayidx27, align 4
+  %arrayidx27 = getelementptr inbounds i32, i32* %ip, i64 %11
+  %12 = load i32, i32* %arrayidx27, align 4
   %idxprom28 = sext i32 %12 to i64
-  %arrayidx29 = getelementptr inbounds float* %b, i64 %idxprom28
-  %13 = load float* %arrayidx29, align 4
+  %arrayidx29 = getelementptr inbounds float, float* %b, i64 %idxprom28
+  %13 = load float, float* %arrayidx29, align 4
   %mul30 = fmul float %13, %alpha
-  %arrayidx33 = getelementptr inbounds float* %a, i64 %11
-  %14 = load float* %arrayidx33, align 4
+  %arrayidx33 = getelementptr inbounds float, float* %a, i64 %11
+  %14 = load float, float* %arrayidx33, align 4
   %add34 = fadd float %14, %mul30
   store float %add34, float* %arrayidx33, align 4
   %15 = add nsw i64 %indvars.iv, 4
-  %arrayidx37 = getelementptr inbounds i32* %ip, i64 %15
-  %16 = load i32* %arrayidx37, align 4
+  %arrayidx37 = getelementptr inbounds i32, i32* %ip, i64 %15
+  %16 = load i32, i32* %arrayidx37, align 4
   %idxprom38 = sext i32 %16 to i64
-  %arrayidx39 = getelementptr inbounds float* %b, i64 %idxprom38
-  %17 = load float* %arrayidx39, align 4
+  %arrayidx39 = getelementptr inbounds float, float* %b, i64 %idxprom38
+  %17 = load float, float* %arrayidx39, align 4
   %mul40 = fmul float %17, %alpha
-  %arrayidx43 = getelementptr inbounds float* %a, i64 %15
-  %18 = load float* %arrayidx43, align 4
+  %arrayidx43 = getelementptr inbounds float, float* %a, i64 %15
+  %18 = load float, float* %arrayidx43, align 4
   %add44 = fadd float %18, %mul40
   store float %add44, float* %arrayidx43, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 5
@@ -302,14 +302,14 @@ for.body:                                         ; preds = %entry, %for.body
 
 ; CHECK: for.body:
 ; CHECK: %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
-; CHECK: %arrayidx = getelementptr inbounds i32* %ip, i64 %indvar
-; CHECK: %0 = load i32* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr inbounds i32, i32* %ip, i64 %indvar
+; CHECK: %0 = load i32, i32* %arrayidx, align 4
 ; CHECK: %idxprom1 = sext i32 %0 to i64
-; CHECK: %arrayidx2 = getelementptr inbounds float* %b, i64 %idxprom1
-; CHECK: %1 = load float* %arrayidx2, align 4
+; CHECK: %arrayidx2 = getelementptr inbounds float, float* %b, i64 %idxprom1
+; CHECK: %1 = load float, float* %arrayidx2, align 4
 ; CHECK: %mul = fmul float %1, %alpha
-; CHECK: %arrayidx4 = getelementptr inbounds float* %a, i64 %indvar
-; CHECK: %2 = load float* %arrayidx4, align 4
+; CHECK: %arrayidx4 = getelementptr inbounds float, float* %a, i64 %indvar
+; CHECK: %2 = load float, float* %arrayidx4, align 4
 ; CHECK: %add = fadd float %2, %mul
 ; CHECK: store float %add, float* %arrayidx4, align 4
 ; CHECK: %indvar.next = add i64 %indvar, 1
@@ -322,6 +322,260 @@ for.end:                                          ; preds = %for.body
   ret void
 }
 
+; void multi1(int *x) {
+;   y = foo(0)
+;   for (int i = 0; i < 500; ++i) {
+;     x[3*i] = y;
+;     x[3*i+1] = y;
+;     x[3*i+2] = y;
+;     x[3*i+6] = y;
+;     x[3*i+7] = y;
+;     x[3*i+8] = y;
+;   }
+; }
+
+; Function Attrs: nounwind uwtable
+define void @multi1(i32* nocapture %x) #0 {
+entry:
+  %call = tail call i32 @foo(i32 0) #1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %0 = mul nsw i64 %indvars.iv, 3
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %0
+  store i32 %call, i32* %arrayidx, align 4
+  %1 = add nsw i64 %0, 1
+  %arrayidx4 = getelementptr inbounds i32, i32* %x, i64 %1
+  store i32 %call, i32* %arrayidx4, align 4
+  %2 = add nsw i64 %0, 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %x, i64 %2
+  store i32 %call, i32* %arrayidx9, align 4
+  %3 = add nsw i64 %0, 6
+  %arrayidx6 = getelementptr inbounds i32, i32* %x, i64 %3
+  store i32 %call, i32* %arrayidx6, align 4
+  %4 = add nsw i64 %0, 7
+  %arrayidx7 = getelementptr inbounds i32, i32* %x, i64 %4
+  store i32 %call, i32* %arrayidx7, align 4
+  %5 = add nsw i64 %0, 8
+  %arrayidx8 = getelementptr inbounds i32, i32* %x, i64 %5
+  store i32 %call, i32* %arrayidx8, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond = icmp eq i64 %indvars.iv.next, 500
+  br i1 %exitcond, label %for.end, label %for.body
+
+; CHECK-LABEL: @multi1
+
+; CHECK:for.body:
+; CHECK:  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+; CHECK:  %0 = add i64 %indvars.iv, 6
+; CHECK:  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
+; CHECK:  store i32 %call, i32* %arrayidx, align 4
+; CHECK:  %arrayidx6 = getelementptr inbounds i32, i32* %x, i64 %0
+; CHECK:  store i32 %call, i32* %arrayidx6, align 4
+; CHECK:  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:  %exitcond2 = icmp eq i64 %0, 1505
+; CHECK:  br i1 %exitcond2, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
+
+; void multi2(int *x) {
+;   y = foo(0)
+;   for (int i = 0; i < 500; ++i) {
+;     x[3*i] = y;
+;     x[3*i+1] = y;
+;     x[3*i+2] = y;
+;     x[3*(i+1)] = y;
+;     x[3*(i+1)+1] = y;
+;     x[3*(i+1)+2] = y;
+;   }
+; }
+
+; Function Attrs: nounwind uwtable
+define void @multi2(i32* nocapture %x) #0 {
+entry:
+  %call = tail call i32 @foo(i32 0) #1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %0 = mul nsw i64 %indvars.iv, 3
+  %add = add nsw i64 %indvars.iv, 1
+  %newmul = mul nsw i64 %add, 3
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %0
+  store i32 %call, i32* %arrayidx, align 4
+  %1 = add nsw i64 %0, 1
+  %arrayidx4 = getelementptr inbounds i32, i32* %x, i64 %1
+  store i32 %call, i32* %arrayidx4, align 4
+  %2 = add nsw i64 %0, 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %x, i64 %2
+  store i32 %call, i32* %arrayidx9, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %x, i64 %newmul
+  store i32 %call, i32* %arrayidx6, align 4
+  %3 = add nsw i64 %newmul, 1
+  %arrayidx7 = getelementptr inbounds i32, i32* %x, i64 %3
+  store i32 %call, i32* %arrayidx7, align 4
+  %4 = add nsw i64 %newmul, 2
+  %arrayidx8 = getelementptr inbounds i32, i32* %x, i64 %4
+  store i32 %call, i32* %arrayidx8, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond = icmp eq i64 %indvars.iv.next, 500
+  br i1 %exitcond, label %for.end, label %for.body
+
+; CHECK-LABEL: @multi2
+
+; CHECK:for.body:
+; CHECK:  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+; CHECK:  %0 = add i64 %indvars.iv, 3
+; CHECK:  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
+; CHECK:  store i32 %call, i32* %arrayidx, align 4
+; CHECK:  %arrayidx6 = getelementptr inbounds i32, i32* %x, i64 %0
+; CHECK:  store i32 %call, i32* %arrayidx6, align 4
+; CHECK:  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:  %exitcond2 = icmp eq i64 %indvars.iv, 1499
+; CHECK:  br i1 %exitcond2, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
+
+; void multi3(int *x) {
+;   y = foo(0)
+;   for (int i = 0; i < 500; ++i) {
+;     // Note: No zero index
+;     x[3*i+3] = y;
+;     x[3*i+4] = y;
+;     x[3*i+5] = y;
+;   }
+; }
+
+; Function Attrs: nounwind uwtable
+define void @multi3(i32* nocapture %x) #0 {
+entry:
+  %call = tail call i32 @foo(i32 0) #1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %0 = mul nsw i64 %indvars.iv, 3
+  %x0 = add nsw i64 %0, 3
+  %add = add nsw i64 %indvars.iv, 1
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %x0
+  store i32 %call, i32* %arrayidx, align 4
+  %1 = add nsw i64 %0, 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %x, i64 %1
+  store i32 %call, i32* %arrayidx4, align 4
+  %2 = add nsw i64 %0, 5
+  %arrayidx9 = getelementptr inbounds i32, i32* %x, i64 %2
+  store i32 %call, i32* %arrayidx9, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond = icmp eq i64 %indvars.iv.next, 500
+  br i1 %exitcond, label %for.end, label %for.body
+
+; CHECK-LABEL: @multi3
+; CHECK: for.body:
+; CHECK:   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+; CHECK:   %0 = add i64 %indvars.iv, 3
+; CHECK:   %arrayidx = getelementptr inbounds i32, i32* %x, i64 %0
+; CHECK:   store i32 %call, i32* %arrayidx, align 4
+; CHECK:   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+; CHECK:   %exitcond1 = icmp eq i64 %0, 1502
+; CHECK:   br i1 %exitcond1, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
+
+; int foo(int a);
+; void bar2(int *x, int y, int z) {
+;   for (int i = 0; i < 500; i += 3) {
+;     foo(i+y+i*z); // Slightly reordered instruction order
+;     foo(i+1+y+(i+1)*z);
+;     foo(i+2+y+(i+2)*z);
+;   }
+; }
+
+; Function Attrs: nounwind uwtable
+define void @bar2(i32* nocapture readnone %x, i32 %y, i32 %z) #0 {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %i.08 = phi i32 [ 0, %entry ], [ %add3, %for.body ]
+
+  %tmp1 = add i32 %i.08, %y
+  %tmp2 = mul i32 %i.08, %z
+  %tmp3 = add i32 %tmp2, %tmp1
+  %call = tail call i32 @foo(i32 %tmp3) #1
+
+  %add = add nsw i32 %i.08, 1
+  %tmp2a = mul i32 %add, %z
+  %tmp1a = add i32 %add, %y
+  %tmp3a = add i32 %tmp2a, %tmp1a
+  %calla = tail call i32 @foo(i32 %tmp3a) #1
+  
+  %add2 = add nsw i32 %i.08, 2
+  %tmp2b = mul i32 %add2, %z
+  %tmp1b = add i32 %add2, %y
+  %tmp3b = add i32 %tmp2b, %tmp1b
+  %callb = tail call i32 @foo(i32 %tmp3b) #1
+
+  %add3 = add nsw i32 %i.08, 3
+
+  %exitcond = icmp eq i32 %add3, 500
+  br i1 %exitcond, label %for.end, label %for.body
+
+; CHECK-LABEL: @bar2
+
+; CHECK: for.body:
+; CHECK: %indvar = phi i32 [ %indvar.next, %for.body ], [ 0, %entry ]
+; CHECK: %tmp1 = add i32 %indvar, %y
+; CHECK: %tmp2 = mul i32 %indvar, %z
+; CHECK: %tmp3 = add i32 %tmp2, %tmp1
+; CHECK: %call = tail call i32 @foo(i32 %tmp3) #1
+; CHECK: %indvar.next = add i32 %indvar, 1
+; CHECK: %exitcond1 = icmp eq i32 %indvar, 497
+; CHECK: br i1 %exitcond1, label %for.end, label %for.body
+
+; CHECK: ret
+
+for.end:                                          ; preds = %for.body
+  ret void
+}
+
+%struct.s = type { i32, i32 }
+
+; Function Attrs: nounwind uwtable
+define void @gep1(%struct.s* nocapture %x) #0 {
+entry:
+  %call = tail call i32 @foo(i32 0) #1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %0 = mul nsw i64 %indvars.iv, 3
+  %arrayidx = getelementptr inbounds %struct.s, %struct.s* %x, i64 %0, i32 0
+  store i32 %call, i32* %arrayidx, align 4
+  %1 = add nsw i64 %0, 1
+  %arrayidx4 = getelementptr inbounds %struct.s, %struct.s* %x, i64 %1, i32 0
+  store i32 %call, i32* %arrayidx4, align 4
+  %2 = add nsw i64 %0, 2
+  %arrayidx9 = getelementptr inbounds %struct.s, %struct.s* %x, i64 %2, i32 0
+  store i32 %call, i32* %arrayidx9, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %exitcond = icmp eq i64 %indvars.iv.next, 500
+  br i1 %exitcond, label %for.end, label %for.body
+
+; CHECK-LABEL: @gep1
+; This test is a crash test only.
+; CHECK: ret
+for.end:                                          ; preds = %for.body
+  ret void
+}
+
+
 attributes #0 = { nounwind uwtable }
 attributes #1 = { nounwind }
 
diff --git a/test/Transforms/LoopReroll/nonconst_lb.ll b/test/Transforms/LoopReroll/nonconst_lb.ll
index a45469bd2393..5effa42d07c4 100644
--- a/test/Transforms/LoopReroll/nonconst_lb.ll
+++ b/test/Transforms/LoopReroll/nonconst_lb.ll
@@ -17,28 +17,28 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.035 = phi i32 [ %add18, %for.body ], [ %m, %entry ]
-  %arrayidx = getelementptr inbounds i32* %B, i32 %i.035
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i32 %i.035
+  %0 = load i32, i32* %arrayidx, align 4
   %mul = shl nsw i32 %0, 2
-  %arrayidx2 = getelementptr inbounds i32* %A, i32 %i.035
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i32 %i.035
   store i32 %mul, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %i.035, 1
-  %arrayidx4 = getelementptr inbounds i32* %B, i32 %add3
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i32 %add3
+  %1 = load i32, i32* %arrayidx4, align 4
   %mul5 = shl nsw i32 %1, 2
-  %arrayidx7 = getelementptr inbounds i32* %A, i32 %add3
+  %arrayidx7 = getelementptr inbounds i32, i32* %A, i32 %add3
   store i32 %mul5, i32* %arrayidx7, align 4
   %add8 = add nsw i32 %i.035, 2
-  %arrayidx9 = getelementptr inbounds i32* %B, i32 %add8
-  %2 = load i32* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds i32, i32* %B, i32 %add8
+  %2 = load i32, i32* %arrayidx9, align 4
   %mul10 = shl nsw i32 %2, 2
-  %arrayidx12 = getelementptr inbounds i32* %A, i32 %add8
+  %arrayidx12 = getelementptr inbounds i32, i32* %A, i32 %add8
   store i32 %mul10, i32* %arrayidx12, align 4
   %add13 = add nsw i32 %i.035, 3
-  %arrayidx14 = getelementptr inbounds i32* %B, i32 %add13
-  %3 = load i32* %arrayidx14, align 4
+  %arrayidx14 = getelementptr inbounds i32, i32* %B, i32 %add13
+  %3 = load i32, i32* %arrayidx14, align 4
   %mul15 = shl nsw i32 %3, 2
-  %arrayidx17 = getelementptr inbounds i32* %A, i32 %add13
+  %arrayidx17 = getelementptr inbounds i32, i32* %A, i32 %add13
   store i32 %mul15, i32* %arrayidx17, align 4
   %add18 = add nsw i32 %i.035, 4
   %cmp = icmp slt i32 %add18, %n
@@ -60,10 +60,10 @@ for.end:                                          ; preds = %for.body, %entry
 ; CHECK: for.body:                                         ; preds = %for.body, %for.body.preheader
 ; CHECK:   %indvar = phi i32 [ %indvar.next, %for.body ], [ 0, %for.body.preheader ]
 ; CHECK:   %6 = add i32 %m, %indvar
-; CHECK:   %arrayidx = getelementptr inbounds i32* %B, i32 %6
-; CHECK:   %7 = load i32* %arrayidx, align 4
+; CHECK:   %arrayidx = getelementptr inbounds i32, i32* %B, i32 %6
+; CHECK:   %7 = load i32, i32* %arrayidx, align 4
 ; CHECK:   %mul = shl nsw i32 %7, 2
-; CHECK:   %arrayidx2 = getelementptr inbounds i32* %A, i32 %6
+; CHECK:   %arrayidx2 = getelementptr inbounds i32, i32* %A, i32 %6
 ; CHECK:   store i32 %mul, i32* %arrayidx2, align 4
 ; CHECK:   %indvar.next = add i32 %indvar, 1
 ; CHECK:   %exitcond = icmp eq i32 %6, %5
@@ -88,34 +88,34 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.056 = phi i32 [ %add27, %for.body ], [ %rem, %entry ]
-  %arrayidx = getelementptr inbounds float* %dy, i32 %i.056
-  %0 = load float* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds float* %dx, i32 %i.056
-  %1 = load float* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds float, float* %dy, i32 %i.056
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds float, float* %dx, i32 %i.056
+  %1 = load float, float* %arrayidx1, align 4
   %mul = fmul float %1, %da
   %add = fadd float %0, %mul
   store float %add, float* %arrayidx, align 4
   %add3 = add nsw i32 %i.056, 1
-  %arrayidx4 = getelementptr inbounds float* %dy, i32 %add3
-  %2 = load float* %arrayidx4, align 4
-  %arrayidx6 = getelementptr inbounds float* %dx, i32 %add3
-  %3 = load float* %arrayidx6, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %dy, i32 %add3
+  %2 = load float, float* %arrayidx4, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %dx, i32 %add3
+  %3 = load float, float* %arrayidx6, align 4
   %mul7 = fmul float %3, %da
   %add8 = fadd float %2, %mul7
   store float %add8, float* %arrayidx4, align 4
   %add11 = add nsw i32 %i.056, 2
-  %arrayidx12 = getelementptr inbounds float* %dy, i32 %add11
-  %4 = load float* %arrayidx12, align 4
-  %arrayidx14 = getelementptr inbounds float* %dx, i32 %add11
-  %5 = load float* %arrayidx14, align 4
+  %arrayidx12 = getelementptr inbounds float, float* %dy, i32 %add11
+  %4 = load float, float* %arrayidx12, align 4
+  %arrayidx14 = getelementptr inbounds float, float* %dx, i32 %add11
+  %5 = load float, float* %arrayidx14, align 4
   %mul15 = fmul float %5, %da
   %add16 = fadd float %4, %mul15
   store float %add16, float* %arrayidx12, align 4
   %add19 = add nsw i32 %i.056, 3
-  %arrayidx20 = getelementptr inbounds float* %dy, i32 %add19
-  %6 = load float* %arrayidx20, align 4
-  %arrayidx22 = getelementptr inbounds float* %dx, i32 %add19
-  %7 = load float* %arrayidx22, align 4
+  %arrayidx20 = getelementptr inbounds float, float* %dy, i32 %add19
+  %6 = load float, float* %arrayidx20, align 4
+  %arrayidx22 = getelementptr inbounds float, float* %dx, i32 %add19
+  %7 = load float, float* %arrayidx22, align 4
   %mul23 = fmul float %7, %da
   %add24 = fadd float %6, %mul23
   store float %add24, float* %arrayidx20, align 4
@@ -140,10 +140,10 @@ for.end:                                          ; preds = %for.body, %entry
 ; CHECK: for.body:
 ; CHECK:   %indvar = phi i32 [ %indvar.next, %for.body ], [ 0, %for.body.preheader ]
 ; CHECK:   %6 = add i32 %rem, %indvar
-; CHECK:   %arrayidx = getelementptr inbounds float* %dy, i32 %6
-; CHECK:   %7 = load float* %arrayidx, align 4
-; CHECK:   %arrayidx1 = getelementptr inbounds float* %dx, i32 %6
-; CHECK:   %8 = load float* %arrayidx1, align 4
+; CHECK:   %arrayidx = getelementptr inbounds float, float* %dy, i32 %6
+; CHECK:   %7 = load float, float* %arrayidx, align 4
+; CHECK:   %arrayidx1 = getelementptr inbounds float, float* %dx, i32 %6
+; CHECK:   %8 = load float, float* %arrayidx1, align 4
 ; CHECK:   %mul = fmul float %8, %da
 ; CHECK:   %add = fadd float %7, %mul
 ; CHECK:   store float %add, float* %arrayidx, align 4
diff --git a/test/Transforms/LoopReroll/reduction.ll b/test/Transforms/LoopReroll/reduction.ll
index c9991c723ec6..191518514b8d 100644
--- a/test/Transforms/LoopReroll/reduction.ll
+++ b/test/Transforms/LoopReroll/reduction.ll
@@ -9,20 +9,20 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.029 = phi i32 [ 0, %entry ], [ %add12, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %x, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %r.029
   %1 = or i64 %indvars.iv, 1
-  %arrayidx3 = getelementptr inbounds i32* %x, i64 %1
-  %2 = load i32* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %x, i64 %1
+  %2 = load i32, i32* %arrayidx3, align 4
   %add4 = add nsw i32 %add, %2
   %3 = or i64 %indvars.iv, 2
-  %arrayidx7 = getelementptr inbounds i32* %x, i64 %3
-  %4 = load i32* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %x, i64 %3
+  %4 = load i32, i32* %arrayidx7, align 4
   %add8 = add nsw i32 %add4, %4
   %5 = or i64 %indvars.iv, 3
-  %arrayidx11 = getelementptr inbounds i32* %x, i64 %5
-  %6 = load i32* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds i32, i32* %x, i64 %5
+  %6 = load i32, i32* %arrayidx11, align 4
   %add12 = add nsw i32 %add8, %6
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 4
   %7 = trunc i64 %indvars.iv.next to i32
@@ -34,8 +34,8 @@ for.body:                                         ; preds = %entry, %for.body
 ; CHECK: for.body:
 ; CHECK: %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
 ; CHECK: %r.029 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-; CHECK: %arrayidx = getelementptr inbounds i32* %x, i64 %indvar
-; CHECK: %0 = load i32* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvar
+; CHECK: %0 = load i32, i32* %arrayidx, align 4
 ; CHECK: %add = add nsw i32 %0, %r.029
 ; CHECK: %indvar.next = add i64 %indvar, 1
 ; CHECK: %exitcond = icmp eq i64 %indvar, 399
@@ -54,20 +54,20 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.029 = phi float [ 0.0, %entry ], [ %add12, %for.body ]
-  %arrayidx = getelementptr inbounds float* %x, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %add = fadd float %0, %r.029
   %1 = or i64 %indvars.iv, 1
-  %arrayidx3 = getelementptr inbounds float* %x, i64 %1
-  %2 = load float* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %x, i64 %1
+  %2 = load float, float* %arrayidx3, align 4
   %add4 = fadd float %add, %2
   %3 = or i64 %indvars.iv, 2
-  %arrayidx7 = getelementptr inbounds float* %x, i64 %3
-  %4 = load float* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %x, i64 %3
+  %4 = load float, float* %arrayidx7, align 4
   %add8 = fadd float %add4, %4
   %5 = or i64 %indvars.iv, 3
-  %arrayidx11 = getelementptr inbounds float* %x, i64 %5
-  %6 = load float* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds float, float* %x, i64 %5
+  %6 = load float, float* %arrayidx11, align 4
   %add12 = fadd float %add8, %6
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 4
   %7 = trunc i64 %indvars.iv.next to i32
@@ -79,8 +79,8 @@ for.body:                                         ; preds = %entry, %for.body
 ; CHECK: for.body:
 ; CHECK: %indvar = phi i64 [ %indvar.next, %for.body ], [ 0, %entry ]
 ; CHECK: %r.029 = phi float [ 0.000000e+00, %entry ], [ %add, %for.body ]
-; CHECK: %arrayidx = getelementptr inbounds float* %x, i64 %indvar
-; CHECK: %0 = load float* %arrayidx, align 4
+; CHECK: %arrayidx = getelementptr inbounds float, float* %x, i64 %indvar
+; CHECK: %0 = load float, float* %arrayidx, align 4
 ; CHECK: %add = fadd float %0, %r.029
 ; CHECK: %indvar.next = add i64 %indvar, 1
 ; CHECK: %exitcond = icmp eq i64 %indvar, 399
@@ -92,5 +92,41 @@ for.end:                                          ; preds = %for.body
   ret float %add12
 }
 
+define i32 @foo_unusedphi(i32* nocapture readonly %x) #0 {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %r.029 = phi i32 [ 0, %entry ], [ %add12, %for.body ]
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %0
+  %1 = or i64 %indvars.iv, 1
+  %arrayidx3 = getelementptr inbounds i32, i32* %x, i64 %1
+  %2 = load i32, i32* %arrayidx3, align 4
+  %add4 = add nsw i32 %add, %2
+  %3 = or i64 %indvars.iv, 2
+  %arrayidx7 = getelementptr inbounds i32, i32* %x, i64 %3
+  %4 = load i32, i32* %arrayidx7, align 4
+  %add8 = add nsw i32 %add4, %4
+  %5 = or i64 %indvars.iv, 3
+  %arrayidx11 = getelementptr inbounds i32, i32* %x, i64 %5
+  %6 = load i32, i32* %arrayidx11, align 4
+  %add12 = add nsw i32 %add8, %6
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 4
+  %7 = trunc i64 %indvars.iv.next to i32
+  %cmp = icmp slt i32 %7, 400
+  br i1 %cmp, label %for.body, label %for.end
+
+; CHECK-LABEL: @foo_unusedphi
+; The above is just testing for a crash - no specific output expected.
+
+; CHECK: ret
+
+for.end:                                          ; preds = %for.body
+  ret i32 %add12
+}
+
 attributes #0 = { nounwind readonly uwtable }
 
diff --git a/test/Transforms/LoopRotate/PhiRename-1.ll b/test/Transforms/LoopRotate/PhiRename-1.ll
index a2247775ea78..6d75888d70db 100644
--- a/test/Transforms/LoopRotate/PhiRename-1.ll
+++ b/test/Transforms/LoopRotate/PhiRename-1.ll
@@ -36,59 +36,59 @@ entry:
 	%op = alloca %struct.operator*, align 4		; <%struct.operator**> [#uses=3]
 	store i32 %arity, i32* %arity_addr
 	store i32 0, i32* %c
-	%tmp1 = load %struct.list** @operators		; <%struct.list*> [#uses=1]
+	%tmp1 = load %struct.list*, %struct.list** @operators		; <%struct.list*> [#uses=1]
 	store %struct.list* %tmp1, %struct.list** %l
 	br label %bb21
 
 bb:		; preds = %bb21
-	%tmp3 = getelementptr %struct.list* %tmp22, i32 0, i32 0		; <i8**> [#uses=1]
-	%tmp4 = load i8** %tmp3		; <i8*> [#uses=1]
+	%tmp3 = getelementptr %struct.list, %struct.list* %tmp22, i32 0, i32 0		; <i8**> [#uses=1]
+	%tmp4 = load i8*, i8** %tmp3		; <i8*> [#uses=1]
 	%tmp45 = bitcast i8* %tmp4 to %struct.operator*		; <%struct.operator*> [#uses=1]
 	store %struct.operator* %tmp45, %struct.operator** %op
-	%tmp6 = load %struct.operator** %op		; <%struct.operator*> [#uses=1]
-	%tmp7 = getelementptr %struct.operator* %tmp6, i32 0, i32 5		; <i32*> [#uses=1]
-	%tmp8 = load i32* %tmp7		; <i32> [#uses=1]
-	%tmp9 = load i32* %arity_addr		; <i32> [#uses=1]
+	%tmp6 = load %struct.operator*, %struct.operator** %op		; <%struct.operator*> [#uses=1]
+	%tmp7 = getelementptr %struct.operator, %struct.operator* %tmp6, i32 0, i32 5		; <i32*> [#uses=1]
+	%tmp8 = load i32, i32* %tmp7		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* %arity_addr		; <i32> [#uses=1]
 	icmp eq i32 %tmp8, %tmp9		; <i1>:0 [#uses=1]
 	zext i1 %0 to i8		; <i8>:1 [#uses=1]
 	icmp ne i8 %1, 0		; <i1>:2 [#uses=1]
 	br i1 %2, label %cond_true, label %cond_next
 
 cond_true:		; preds = %bb
-	%tmp10 = load %struct.operator** %op		; <%struct.operator*> [#uses=1]
-	%tmp11 = getelementptr %struct.operator* %tmp10, i32 0, i32 2		; <i32*> [#uses=1]
-	%tmp12 = load i32* %tmp11		; <i32> [#uses=1]
-	%tmp13 = load %struct.FILE** @outfile		; <%struct.FILE*> [#uses=1]
-	%tmp14 = getelementptr [11 x i8]* @str1, i32 0, i32 0		; <i8*> [#uses=1]
-	%tmp15 = call i32 (%struct.FILE*, i8*, ...)* @fprintf( %struct.FILE* %tmp13, i8* %tmp14, i32 %tmp12 )		; <i32> [#uses=0]
-	%tmp16 = load i32* %c		; <i32> [#uses=1]
+	%tmp10 = load %struct.operator*, %struct.operator** %op		; <%struct.operator*> [#uses=1]
+	%tmp11 = getelementptr %struct.operator, %struct.operator* %tmp10, i32 0, i32 2		; <i32*> [#uses=1]
+	%tmp12 = load i32, i32* %tmp11		; <i32> [#uses=1]
+	%tmp13 = load %struct.FILE*, %struct.FILE** @outfile		; <%struct.FILE*> [#uses=1]
+	%tmp14 = getelementptr [11 x i8], [11 x i8]* @str1, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp15 = call i32 (%struct.FILE*, i8*, ...) @fprintf( %struct.FILE* %tmp13, i8* %tmp14, i32 %tmp12 )		; <i32> [#uses=0]
+	%tmp16 = load i32, i32* %c		; <i32> [#uses=1]
 	%tmp17 = add i32 %tmp16, 1		; <i32> [#uses=1]
 	store i32 %tmp17, i32* %c
 	br label %cond_next
 
 cond_next:		; preds = %cond_true, %bb
-	%tmp19 = getelementptr %struct.list* %tmp22, i32 0, i32 1		; <%struct.list**> [#uses=1]
-	%tmp20 = load %struct.list** %tmp19		; <%struct.list*> [#uses=1]
+	%tmp19 = getelementptr %struct.list, %struct.list* %tmp22, i32 0, i32 1		; <%struct.list**> [#uses=1]
+	%tmp20 = load %struct.list*, %struct.list** %tmp19		; <%struct.list*> [#uses=1]
 	store %struct.list* %tmp20, %struct.list** %l
 	br label %bb21
 
 bb21:		; preds = %cond_next, %entry
         %l.in = phi %struct.list** [ @operators, %entry ], [ %tmp19, %cond_next ]
-	%tmp22 = load %struct.list** %l.in		; <%struct.list*> [#uses=1]
+	%tmp22 = load %struct.list*, %struct.list** %l.in		; <%struct.list*> [#uses=1]
 	icmp ne %struct.list* %tmp22, null		; <i1>:3 [#uses=1]
 	zext i1 %3 to i8		; <i8>:4 [#uses=1]
 	icmp ne i8 %4, 0		; <i1>:5 [#uses=1]
 	br i1 %5, label %bb, label %bb23
 
 bb23:		; preds = %bb21
-	%tmp24 = load i32* %c		; <i32> [#uses=1]
+	%tmp24 = load i32, i32* %c		; <i32> [#uses=1]
 	store i32 %tmp24, i32* %tmp
-	%tmp25 = load i32* %tmp		; <i32> [#uses=1]
+	%tmp25 = load i32, i32* %tmp		; <i32> [#uses=1]
 	store i32 %tmp25, i32* %retval
 	br label %return
 
 return:		; preds = %bb23
-	%retval26 = load i32* %retval		; <i32> [#uses=1]
+	%retval26 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval26
 }
 
diff --git a/test/Transforms/LoopRotate/PhiSelfReference-1.ll b/test/Transforms/LoopRotate/PhiSelfReference-1.ll
index aa1708e8475c..ed4944833913 100644
--- a/test/Transforms/LoopRotate/PhiSelfReference-1.ll
+++ b/test/Transforms/LoopRotate/PhiSelfReference-1.ll
@@ -28,7 +28,7 @@ bb243:		; preds = %bb307
 
 bb307:		; preds = %bb243, %bb52
 	%sx_addr.2.pn = phi float* [ %sx_addr.5, %bb243 ], [ null, %bb52 ]		; <float*> [#uses=1]
-	%sx_addr.5 = getelementptr float* %sx_addr.2.pn, i32 %incx		; <float*> [#uses=1]
+	%sx_addr.5 = getelementptr float, float* %sx_addr.2.pn, i32 %incx		; <float*> [#uses=1]
 	br i1 false, label %bb243, label %bb310
 
 bb310:		; preds = %bb307
diff --git a/test/Transforms/LoopRotate/alloca.ll b/test/Transforms/LoopRotate/alloca.ll
index fd217ea8dcf0..bbcfb39c3722 100644
--- a/test/Transforms/LoopRotate/alloca.ll
+++ b/test/Transforms/LoopRotate/alloca.ll
@@ -14,7 +14,7 @@ declare void @use(i8*)
 
 define void @test() {
 entry:
-  %end = load i16* @e
+  %end = load i16, i16* @e
   br label %loop
 
 loop:
diff --git a/test/Transforms/LoopRotate/basic.ll b/test/Transforms/LoopRotate/basic.ll
index 6b92a6ecac13..9c04fa28753e 100644
--- a/test/Transforms/LoopRotate/basic.ll
+++ b/test/Transforms/LoopRotate/basic.ll
@@ -13,7 +13,7 @@ entry:
 for.cond:                                         ; preds = %for.body, %entry
   %i.0 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %cmp = icmp slt i32 %i.0, 100
-  %arrayidx = getelementptr inbounds [20 x i32]* %array, i64 0, i64 0
+  %arrayidx = getelementptr inbounds [20 x i32], [20 x i32]* %array, i64 0, i64 0
   br i1 %cmp, label %for.body, label %for.end
 
 ; CHECK: for.body:
diff --git a/test/Transforms/LoopRotate/crash.ll b/test/Transforms/LoopRotate/crash.ll
index e95f9a1f350f..5e2b66d6803d 100644
--- a/test/Transforms/LoopRotate/crash.ll
+++ b/test/Transforms/LoopRotate/crash.ll
@@ -125,7 +125,7 @@ entry:
 	br label %bb15
 bb6:		; preds = %bb15
 	%gep.upgrd.1 = zext i32 %offset.1 to i64		; <i64> [#uses=1]
-	%tmp11 = getelementptr i8* %msg, i64 %gep.upgrd.1		; <i8*> [#uses=0]
+	%tmp11 = getelementptr i8, i8* %msg, i64 %gep.upgrd.1		; <i8*> [#uses=0]
 	br label %bb15
 bb15:		; preds = %bb6, %entry
 	%offset.1 = add i32 0, 1		; <i32> [#uses=2]
diff --git a/test/Transforms/LoopRotate/dbgvalue.ll b/test/Transforms/LoopRotate/dbgvalue.ll
index 846b366f2942..9bcca15ab551 100644
--- a/test/Transforms/LoopRotate/dbgvalue.ll
+++ b/test/Transforms/LoopRotate/dbgvalue.ll
@@ -15,9 +15,9 @@ tailrecurse:                                      ; preds = %if.then, %entry
   %x.tr = phi i32 [ %x, %entry ], [ %call, %if.then ]
   %y.tr = phi i32 [ %y, %entry ], [ %call9, %if.then ]
   %z.tr = phi i32 [ %z, %entry ], [ %call14, %if.then ]
-  tail call void @llvm.dbg.value(metadata i32 %x.tr, i64 0, metadata !6, metadata !{}), !dbg !7
-  tail call void @llvm.dbg.value(metadata i32 %y.tr, i64 0, metadata !8, metadata !{}), !dbg !9
-  tail call void @llvm.dbg.value(metadata i32 %z.tr, i64 0, metadata !10, metadata !{}), !dbg !11
+  tail call void @llvm.dbg.value(metadata i32 %x.tr, i64 0, metadata !6, metadata !DIExpression()), !dbg !7
+  tail call void @llvm.dbg.value(metadata i32 %y.tr, i64 0, metadata !8, metadata !DIExpression()), !dbg !9
+  tail call void @llvm.dbg.value(metadata i32 %z.tr, i64 0, metadata !10, metadata !DIExpression()), !dbg !11
   %cmp = icmp slt i32 %y.tr, %x.tr, !dbg !12
   br i1 %cmp, label %if.then, label %if.end, !dbg !12
 
@@ -61,18 +61,18 @@ for.cond:
   br i1 %cmp, label %for.end, label %for.body
 
 for.body:
-  %0 = load i64* @channelColumns, align 8
+  %0 = load i64, i64* @channelColumns, align 8
   %mul = mul i64 %0, %row
   %add = add i64 %mul, %i.0
-  %1 = load i8** @horzPlane, align 8
-  %arrayidx = getelementptr inbounds i8* %1, i64 %add
-  %2 = load i8* %arrayidx, align 1
+  %1 = load i8*, i8** @horzPlane, align 8
+  %arrayidx = getelementptr inbounds i8, i8* %1, i64 %add
+  %2 = load i8, i8* %arrayidx, align 1
   %tobool = icmp eq i8 %2, 0
   br i1 %tobool, label %for.inc, label %for.end
 
 for.inc:
   %dec = add i64 %i.0, -1
-  tail call void @llvm.dbg.value(metadata i64 %dec, i64 0, metadata !{!"undef"}, metadata !{})
+  tail call void @llvm.dbg.value(metadata i64 %dec, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !0), metadata !DIExpression()), !dbg !DILocation(scope: !0)
   br label %for.cond
 
 for.end:
@@ -84,24 +84,24 @@ for.end:
 !llvm.module.flags = !{!20}
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00tak\00tak\00\0032\000\001\000\006\00256\000\000", !18, !1, !3, null, i32 (i32, i32, i32)* @tak, null, null, null} ; [ DW_TAG_subprogram ] [line 32] [def] [scope 0] [tak]
-!1 = !{!"0x29", !18} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang version 2.9 (trunk 125492)\001\00\000\00\000", !18, !19, !19, null, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !18, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "tak", line: 32, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !18, scope: !1, type: !3, function: i32 (i32, i32, i32)* @tak)
+!1 = !DIFile(filename: "/Volumes/Lalgate/cj/llvm/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame/recursive.c", directory: "/Volumes/Lalgate/cj/D/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 2.9 (trunk 125492)", isOptimized: true, emissionKind: 0, file: !18, enums: !19, retainedTypes: !19)
+!3 = !DISubroutineType(types: !4)
 !4 = !{!5}
-!5 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !2} ; [ DW_TAG_base_type ]
-!6 = !{!"0x101\00x\0032\000", !0, !1, !5} ; [ DW_TAG_arg_variable ]
-!7 = !MDLocation(line: 32, column: 13, scope: !0)
-!8 = !{!"0x101\00y\0032\000", !0, !1, !5} ; [ DW_TAG_arg_variable ]
-!9 = !MDLocation(line: 32, column: 20, scope: !0)
-!10 = !{!"0x101\00z\0032\000", !0, !1, !5} ; [ DW_TAG_arg_variable ]
-!11 = !MDLocation(line: 32, column: 27, scope: !0)
-!12 = !MDLocation(line: 33, column: 3, scope: !13)
-!13 = !{!"0xb\0032\0030\006", !18, !0} ; [ DW_TAG_lexical_block ]
-!14 = !MDLocation(line: 34, column: 5, scope: !15)
-!15 = !{!"0xb\0033\0014\007", !18, !13} ; [ DW_TAG_lexical_block ]
-!16 = !MDLocation(line: 36, column: 3, scope: !13)
-!17 = !MDLocation(line: 37, column: 1, scope: !13)
-!18 = !{!"/Volumes/Lalgate/cj/llvm/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame/recursive.c", !"/Volumes/Lalgate/cj/D/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame"}
+!5 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!6 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "x", line: 32, arg: 0, scope: !0, file: !1, type: !5)
+!7 = !DILocation(line: 32, column: 13, scope: !0)
+!8 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "y", line: 32, arg: 0, scope: !0, file: !1, type: !5)
+!9 = !DILocation(line: 32, column: 20, scope: !0)
+!10 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "z", line: 32, arg: 0, scope: !0, file: !1, type: !5)
+!11 = !DILocation(line: 32, column: 27, scope: !0)
+!12 = !DILocation(line: 33, column: 3, scope: !13)
+!13 = distinct !DILexicalBlock(line: 32, column: 30, file: !18, scope: !0)
+!14 = !DILocation(line: 34, column: 5, scope: !15)
+!15 = distinct !DILexicalBlock(line: 33, column: 14, file: !18, scope: !13)
+!16 = !DILocation(line: 36, column: 3, scope: !13)
+!17 = !DILocation(line: 37, column: 1, scope: !13)
+!18 = !DIFile(filename: "/Volumes/Lalgate/cj/llvm/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame/recursive.c", directory: "/Volumes/Lalgate/cj/D/projects/llvm-test/SingleSource/Benchmarks/BenchmarkGame")
 !19 = !{i32 0}
-!20 = !{i32 1, !"Debug Info Version", i32 2}
+!20 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/LoopRotate/indirectbr.ll b/test/Transforms/LoopRotate/indirectbr.ll
index 9c82aa883463..2ccc54697aee 100644
--- a/test/Transforms/LoopRotate/indirectbr.ll
+++ b/test/Transforms/LoopRotate/indirectbr.ll
@@ -12,7 +12,7 @@ while.body:                                       ; preds = %while.cond
   br label %indirectgoto
 
 run_opcode:                                       ; preds = %indirectgoto
-  %tmp276 = load i8* undef                        ; <i8> [#uses=1]
+  %tmp276 = load i8, i8* undef                        ; <i8> [#uses=1]
   br label %indirectgoto
 
 if.else295:                                       ; preds = %divide_late
diff --git a/test/Transforms/LoopRotate/multiple-exits.ll b/test/Transforms/LoopRotate/multiple-exits.ll
index cc8738e479d4..f31ed7f1c5b5 100644
--- a/test/Transforms/LoopRotate/multiple-exits.ll
+++ b/test/Transforms/LoopRotate/multiple-exits.ll
@@ -21,8 +21,8 @@ for.cond1:                                        ; preds = %for.cond, %land.rhs
 
 land.rhs:                                         ; preds = %for.cond1
   %conv = zext i32 %i.1 to i64
-  %arrayidx = getelementptr inbounds [100 x i32]* %a, i64 0, i64 %conv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* %a, i64 0, i64 %conv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add i32 %0, %sum.1
   %cmp4 = icmp ugt i32 %add, 1000
   %inc = add i32 %i.1, 1
diff --git a/test/Transforms/LoopRotate/nosimplifylatch.ll b/test/Transforms/LoopRotate/nosimplifylatch.ll
index 8e858b4dfcc3..07ff6643c6d1 100644
--- a/test/Transforms/LoopRotate/nosimplifylatch.ll
+++ b/test/Transforms/LoopRotate/nosimplifylatch.ll
@@ -18,13 +18,13 @@ for.cond:                                         ; preds = %for.inc, %entry
   br i1 %lnot.i, label %for.end, label %for.body
 
 for.body:                                         ; preds = %for.cond
-  %0 = load i32* %coerce.val.ip9, align 4
-  %1 = load i32* %__value_, align 4
+  %0 = load i32, i32* %coerce.val.ip9, align 4
+  %1 = load i32, i32* %__value_, align 4
   %cmp = icmp eq i32 %0, %1
   br i1 %cmp, label %for.end, label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %incdec.ptr.i = getelementptr inbounds i32* %coerce.val.ip9, i64 1
+  %incdec.ptr.i = getelementptr inbounds i32, i32* %coerce.val.ip9, i64 1
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond, %for.body
diff --git a/test/Transforms/LoopRotate/phi-duplicate.ll b/test/Transforms/LoopRotate/phi-duplicate.ll
index 86a4f2a475fa..46ee5961ba54 100644
--- a/test/Transforms/LoopRotate/phi-duplicate.ll
+++ b/test/Transforms/LoopRotate/phi-duplicate.ll
@@ -13,13 +13,13 @@ for.cond:                                         ; preds = %for.body, %entry
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %arrayidx = getelementptr inbounds double* %G, i64 %j.0 ; <double*> [#uses=1]
-  %tmp3 = load double* %arrayidx                  ; <double> [#uses=1]
+  %arrayidx = getelementptr inbounds double, double* %G, i64 %j.0 ; <double*> [#uses=1]
+  %tmp3 = load double, double* %arrayidx                  ; <double> [#uses=1]
   %sub = sub i64 %j.0, 1                          ; <i64> [#uses=1]
-  %arrayidx6 = getelementptr inbounds double* %G, i64 %sub ; <double*> [#uses=1]
-  %tmp7 = load double* %arrayidx6                 ; <double> [#uses=1]
+  %arrayidx6 = getelementptr inbounds double, double* %G, i64 %sub ; <double*> [#uses=1]
+  %tmp7 = load double, double* %arrayidx6                 ; <double> [#uses=1]
   %add = fadd double %tmp3, %tmp7                 ; <double> [#uses=1]
-  %arrayidx10 = getelementptr inbounds double* %G, i64 %j.0 ; <double*> [#uses=1]
+  %arrayidx10 = getelementptr inbounds double, double* %G, i64 %j.0 ; <double*> [#uses=1]
   store double %add, double* %arrayidx10
   %inc = add nsw i64 %j.0, 1                      ; <i64> [#uses=1]
   br label %for.cond
diff --git a/test/Transforms/LoopRotate/pr22337.ll b/test/Transforms/LoopRotate/pr22337.ll
new file mode 100644
index 000000000000..03e804b775e4
--- /dev/null
+++ b/test/Transforms/LoopRotate/pr22337.ll
@@ -0,0 +1,24 @@
+; RUN: opt < %s -loop-rotate -S | FileCheck %s
+
+@a = external global i8, align 4
+@tmp = global i8* @a
+
+define void @f() {
+; CHECK-LABEL: define void @f(
+; CHECK: getelementptr i8, i8* @a, i32 0
+entry:
+  br label %for.preheader
+
+for.preheader:
+  br i1 undef, label %if.then8, label %for.body
+
+for.body:
+  br i1 undef, label %if.end, label %if.then8
+
+if.end:
+  %arrayidx = getelementptr i8, i8* @a, i32 0
+  br label %for.preheader
+
+if.then8:
+  unreachable
+}
diff --git a/test/Transforms/LoopRotate/simplifylatch.ll b/test/Transforms/LoopRotate/simplifylatch.ll
index 62e5b1ae11d5..215622fe3028 100644
--- a/test/Transforms/LoopRotate/simplifylatch.ll
+++ b/test/Transforms/LoopRotate/simplifylatch.ll
@@ -13,8 +13,8 @@ entry:
 bb:		; preds = %bb4, %entry
 	%mode.0 = phi i8 [ 0, %entry ], [ %indvar.next, %bb4 ]		; <i8> [#uses=4]
 	zext i8 %mode.0 to i32		; <i32>:1 [#uses=1]
-	getelementptr [4 x i32]* @mode_table, i32 0, i32 %1		; <i32*>:2 [#uses=1]
-	load i32* %2, align 4		; <i32>:3 [#uses=1]
+	getelementptr [4 x i32], [4 x i32]* @mode_table, i32 0, i32 %1		; <i32*>:2 [#uses=1]
+	load i32, i32* %2, align 4		; <i32>:3 [#uses=1]
 	icmp eq i32 %3, %0		; <i1>:4 [#uses=1]
 	br i1 %4, label %bb1, label %bb2
 
@@ -39,8 +39,8 @@ declare i32 @fegetround()
 declare void @raise_exception() noreturn
 
 ;CHECK: for.body.lr.ph:
-;CHECK-NEXT:  %arrayidx1 = getelementptr inbounds i8* %CurPtr, i64 0
-;CHECK-NEXT:  %0 = load i8* %arrayidx1, align 1
+;CHECK-NEXT:  %arrayidx1 = getelementptr inbounds i8, i8* %CurPtr, i64 0
+;CHECK-NEXT:  %0 = load i8, i8* %arrayidx1, align 1
 ;CHECK-NEXT:  %conv2 = sext i8 %0 to i32
 ;CHECK-NEXT:  br label %for.body
 
@@ -55,11 +55,11 @@ for.cond:					  ; preds = %for.inc, %entry
 
 for.body:					  ; preds = %for.cond
   %idxprom = zext i32 %i.0 to i64
-  %arrayidx = getelementptr inbounds i8* %CurPtr, i64 %idxprom
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %CurPtr, i64 %idxprom
+  %0 = load i8, i8* %arrayidx, align 1
   %conv = sext i8 %0 to i32
-  %arrayidx1 = getelementptr inbounds i8* %CurPtr, i64 0
-  %1 = load i8* %arrayidx1, align 1
+  %arrayidx1 = getelementptr inbounds i8, i8* %CurPtr, i64 0
+  %1 = load i8, i8* %arrayidx1, align 1
   %conv2 = sext i8 %1 to i32
   %cmp3 = icmp ne i32 %conv, %conv2
   br i1 %cmp3, label %return, label %for.inc
diff --git a/test/Transforms/LoopSimplify/2003-08-15-PreheadersFail.ll b/test/Transforms/LoopSimplify/2003-08-15-PreheadersFail.ll
index 73f0813285ba..4a69067f5f18 100644
--- a/test/Transforms/LoopSimplify/2003-08-15-PreheadersFail.ll
+++ b/test/Transforms/LoopSimplify/2003-08-15-PreheadersFail.ll
@@ -8,29 +8,29 @@ target datalayout = "e-p:32:32"
 define i32 @_yylex() {
 	br label %loopentry.0
 loopentry.0:		; preds = %else.26, %0
-	store i32* getelementptr ([16386 x i32]* @yy_state_buf, i64 0, i64 0), i32** @yy_state_ptr
-	%tmp.35 = load i32** @yy_state_ptr		; <i32*> [#uses=2]
-	%inc.0 = getelementptr i32* %tmp.35, i64 1		; <i32*> [#uses=1]
+	store i32* getelementptr ([16386 x i32], [16386 x i32]* @yy_state_buf, i64 0, i64 0), i32** @yy_state_ptr
+	%tmp.35 = load i32*, i32** @yy_state_ptr		; <i32*> [#uses=2]
+	%inc.0 = getelementptr i32, i32* %tmp.35, i64 1		; <i32*> [#uses=1]
 	store i32* %inc.0, i32** @yy_state_ptr
-	%tmp.36 = load i32* null		; <i32> [#uses=1]
+	%tmp.36 = load i32, i32* null		; <i32> [#uses=1]
 	store i32 %tmp.36, i32* %tmp.35
 	br label %loopexit.2
 loopexit.2:		; preds = %else.26, %loopexit.2, %loopentry.0
 	store i8* null, i8** null
-	%tmp.91 = load i32* null		; <i32> [#uses=1]
+	%tmp.91 = load i32, i32* null		; <i32> [#uses=1]
 	%tmp.92 = sext i32 %tmp.91 to i64		; <i64> [#uses=1]
-	%tmp.93 = getelementptr [787 x i16]* @yy_base, i64 0, i64 %tmp.92		; <i16*> [#uses=1]
-	%tmp.94 = load i16* %tmp.93		; <i16> [#uses=1]
+	%tmp.93 = getelementptr [787 x i16], [787 x i16]* @yy_base, i64 0, i64 %tmp.92		; <i16*> [#uses=1]
+	%tmp.94 = load i16, i16* %tmp.93		; <i16> [#uses=1]
 	%tmp.95 = icmp ne i16 %tmp.94, 4394		; <i1> [#uses=1]
 	br i1 %tmp.95, label %loopexit.2, label %yy_find_action
 yy_find_action:		; preds = %else.26, %loopexit.2
 	br label %loopentry.3
 loopentry.3:		; preds = %then.9, %shortcirc_done.0, %yy_find_action
-	%tmp.105 = load i32* @yy_lp		; <i32> [#uses=1]
+	%tmp.105 = load i32, i32* @yy_lp		; <i32> [#uses=1]
 	%tmp.106 = icmp ne i32 %tmp.105, 0		; <i1> [#uses=1]
 	br i1 %tmp.106, label %shortcirc_next.0, label %shortcirc_done.0
 shortcirc_next.0:		; preds = %loopentry.3
-	%tmp.114 = load i16* null		; <i16> [#uses=1]
+	%tmp.114 = load i16, i16* null		; <i16> [#uses=1]
 	%tmp.115 = sext i16 %tmp.114 to i32		; <i32> [#uses=1]
 	%tmp.116 = icmp slt i32 0, %tmp.115		; <i1> [#uses=1]
 	br label %shortcirc_done.0
@@ -38,7 +38,7 @@ shortcirc_done.0:		; preds = %shortcirc_next.0, %loopentry.3
 	%shortcirc_val.0 = phi i1 [ false, %loopentry.3 ], [ %tmp.116, %shortcirc_next.0 ]		; <i1> [#uses=1]
 	br i1 %shortcirc_val.0, label %else.0, label %loopentry.3
 else.0:		; preds = %shortcirc_done.0
-	%tmp.144 = load i32* null		; <i32> [#uses=1]
+	%tmp.144 = load i32, i32* null		; <i32> [#uses=1]
 	%tmp.145 = and i32 %tmp.144, 8192		; <i32> [#uses=1]
 	%tmp.146 = icmp ne i32 %tmp.145, 0		; <i1> [#uses=1]
 	br i1 %tmp.146, label %then.9, label %else.26
diff --git a/test/Transforms/LoopSimplify/2003-12-10-ExitBlocksProblem.ll b/test/Transforms/LoopSimplify/2003-12-10-ExitBlocksProblem.ll
index fb39f05c6dbb..32b632220d3c 100644
--- a/test/Transforms/LoopSimplify/2003-12-10-ExitBlocksProblem.ll
+++ b/test/Transforms/LoopSimplify/2003-12-10-ExitBlocksProblem.ll
@@ -14,15 +14,15 @@ entry:
 	store i32 123, i32* @G
 	br label %loopentry.i
 loopentry.i:		; preds = %endif.1.i, %entry
-	%tmp.0.i = load i32* @G		; <i32> [#uses=1]
+	%tmp.0.i = load i32, i32* @G		; <i32> [#uses=1]
 	%tmp.1.i = icmp eq i32 %tmp.0.i, 123		; <i1> [#uses=1]
 	br i1 %tmp.1.i, label %Out.i, label %endif.0.i
 endif.0.i:		; preds = %loopentry.i
-	%tmp.3.i = load i32* @G		; <i32> [#uses=1]
+	%tmp.3.i = load i32, i32* @G		; <i32> [#uses=1]
 	%tmp.4.i = icmp eq i32 %tmp.3.i, 126		; <i1> [#uses=1]
 	br i1 %tmp.4.i, label %ExitBlock.i, label %endif.1.i
 endif.1.i:		; preds = %endif.0.i
-	%tmp.6.i = load i32* @G		; <i32> [#uses=1]
+	%tmp.6.i = load i32, i32* @G		; <i32> [#uses=1]
 	%inc.i = add i32 %tmp.6.i, 1		; <i32> [#uses=1]
 	store i32 %inc.i, i32* @G
 	br label %loopentry.i
@@ -30,7 +30,7 @@ Out.i:		; preds = %loopentry.i
 	store i32 0, i32* @G
 	br label %ExitBlock.i
 ExitBlock.i:		; preds = %Out.i, %endif.0.i
-	%tmp.7.i = load i32* @G		; <i32> [#uses=1]
+	%tmp.7.i = load i32, i32* @G		; <i32> [#uses=1]
 	ret i32 %tmp.7.i
 }
 
diff --git a/test/Transforms/LoopSimplify/2011-12-14-LandingpadHeader.ll b/test/Transforms/LoopSimplify/2011-12-14-LandingpadHeader.ll
index 173a5825767b..39471eb0d11a 100644
--- a/test/Transforms/LoopSimplify/2011-12-14-LandingpadHeader.ll
+++ b/test/Transforms/LoopSimplify/2011-12-14-LandingpadHeader.ll
@@ -12,7 +12,7 @@ entry:
 ; CHECK-NEXT: landingpad
 ; CHECK: br label %catch
 
-; CHECK: catch.split-lp:
+; CHECK: catch.preheader.split-lp:
 ; CHECK-NEXT: landingpad
 ; CHECK: br label %catch
 
diff --git a/test/Transforms/LoopSimplify/ashr-crash.ll b/test/Transforms/LoopSimplify/ashr-crash.ll
index c58903d49d55..b5cc1449cec2 100644
--- a/test/Transforms/LoopSimplify/ashr-crash.ll
+++ b/test/Transforms/LoopSimplify/ashr-crash.ll
@@ -51,10 +51,10 @@ for.cond1:                                        ; preds = %for.cond, %for.body
   br i1 %cmp2, label %for.body3, label %for.inc7
 
 for.body3:                                        ; preds = %for.cond1
-  %0 = load i32* @c, align 4
+  %0 = load i32, i32* @c, align 4
   %cmp4 = icmp sge i32 %storemerge1, %0
   %conv = zext i1 %cmp4 to i32
-  %1 = load i32* @d, align 4
+  %1 = load i32, i32* @d, align 4
   %add = add nsw i32 %conv, %1
   %sext = shl i32 %add, 16
   %conv6 = ashr exact i32 %sext, 16
@@ -63,7 +63,7 @@ for.body3:                                        ; preds = %for.cond1
   br label %for.cond1
 
 for.inc7:                                         ; preds = %for.cond1
-  %2 = load i32* @d, align 4
+  %2 = load i32, i32* @d, align 4
   %inc8 = add nsw i32 %2, 1
   br label %for.cond
 
diff --git a/test/Transforms/LoopSimplify/dup-preds.ll b/test/Transforms/LoopSimplify/dup-preds.ll
index 3d1f1499b11c..c9253fa51a65 100644
--- a/test/Transforms/LoopSimplify/dup-preds.ll
+++ b/test/Transforms/LoopSimplify/dup-preds.ll
@@ -28,7 +28,7 @@ for.body305:                                      ; preds = %for.body305, %if.th
   br label %for.body305
 
 for.body344:                                      ; preds = %for.body344, %for.body276.lr.ph, %for.body276.lr.ph
-  %indvar = phi i64 [ %indvar.next, %for.body344 ], [ 0, %for.body276.lr.ph ]
+  %indvar = phi i64 [ %indvar.next, %for.body344 ], [ 0, %for.body276.lr.ph ], [ 0, %for.body276.lr.ph ]
   %indvars.iv552 = phi i64 [ %indvars.iv.next553, %for.body344 ], [ 0, %for.body276.lr.ph ], [ 0, %for.body276.lr.ph ]
   %indvars.iv.next553 = add nuw nsw i64 %indvars.iv552, 1
   %indvar.next = add i64 %indvar, 1
diff --git a/test/Transforms/LoopSimplify/merge-exits.ll b/test/Transforms/LoopSimplify/merge-exits.ll
index 9678148c3570..5cdf81487784 100644
--- a/test/Transforms/LoopSimplify/merge-exits.ll
+++ b/test/Transforms/LoopSimplify/merge-exits.ll
@@ -15,18 +15,18 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 define float @test1(float* %pTmp1, float* %peakWeight, i32 %bandEdgeIndex) nounwind {
 entry:
-	%t0 = load float* %peakWeight, align 4
+	%t0 = load float, float* %peakWeight, align 4
 	br label %bb1
 
 bb:		; preds = %bb2
 	%t1 = sext i32 %hiPart.0 to i64
-	%t2 = getelementptr float* %pTmp1, i64 %t1
-	%t3 = load float* %t2, align 4
+	%t2 = getelementptr float, float* %pTmp1, i64 %t1
+	%t3 = load float, float* %t2, align 4
 	%t4 = fadd float %t3, %distERBhi.0
 	%t5 = add i32 %hiPart.0, 1
 	%t6 = sext i32 %t5 to i64
-	%t7 = getelementptr float* %peakWeight, i64 %t6
-	%t8 = load float* %t7, align 4
+	%t7 = getelementptr float, float* %peakWeight, i64 %t6
+	%t8 = load float, float* %t7, align 4
 	%t9 = fadd float %t8, %peakCount.0
 	br label %bb1
 
diff --git a/test/Transforms/LoopSimplify/notify-scev.ll b/test/Transforms/LoopSimplify/notify-scev.ll
index ee8e2eec9b3f..059b589abefa 100644
--- a/test/Transforms/LoopSimplify/notify-scev.ll
+++ b/test/Transforms/LoopSimplify/notify-scev.ll
@@ -60,7 +60,7 @@ for.inc172.us:
   br i1 undef, label %for.cond142.preheader.us, label %for.end174.us
 
 for.body145.us:
-  %arrayidx163.us = getelementptr inbounds %struct.Params* undef, i64 0, i32 0, i64 %idxprom130, i64 %idxprom146.us
+  %arrayidx163.us = getelementptr inbounds %struct.Params, %struct.Params* undef, i64 0, i32 0, i64 %idxprom130, i64 %idxprom146.us
   br i1 undef, label %for.body145.us, label %for.inc172.us
 
 for.cond142.preheader.us:
diff --git a/test/Transforms/LoopSimplify/phi-node-simplify.ll b/test/Transforms/LoopSimplify/phi-node-simplify.ll
index 8eb63d9111a7..6536acb7b038 100644
--- a/test/Transforms/LoopSimplify/phi-node-simplify.ll
+++ b/test/Transforms/LoopSimplify/phi-node-simplify.ll
@@ -12,29 +12,29 @@ entry:
 	br label %loopentry.1
 loopentry.1:		; preds = %loopexit.1, %entry
 	%indvar20 = phi i32 [ 0, %entry ], [ %indvar.next21, %loopexit.1 ]		; <i32> [#uses=1]
-	%a.1 = phi i32* [ getelementptr ([3000000 x i32]* @A, i32 0, i32 0), %entry ], [ %inc.0, %loopexit.1 ]		; <i32*> [#uses=1]
+	%a.1 = phi i32* [ getelementptr ([3000000 x i32], [3000000 x i32]* @A, i32 0, i32 0), %entry ], [ %inc.0, %loopexit.1 ]		; <i32*> [#uses=1]
 	br label %no_exit.2
 no_exit.2:		; preds = %loopexit.2, %no_exit.2, %loopentry.1
 	%a.0.4.ph = phi i32* [ %a.1, %loopentry.1 ], [ %inc.0, %loopexit.2 ], [ %a.0.4.ph, %no_exit.2 ]		; <i32*> [#uses=3]
-	%b.1.4.ph = phi i32* [ getelementptr ([20000 x i32]* @B, i32 0, i32 0), %loopentry.1 ], [ %inc.1, %loopexit.2 ], [ %b.1.4.ph, %no_exit.2 ]		; <i32*> [#uses=3]
+	%b.1.4.ph = phi i32* [ getelementptr ([20000 x i32], [20000 x i32]* @B, i32 0, i32 0), %loopentry.1 ], [ %inc.1, %loopexit.2 ], [ %b.1.4.ph, %no_exit.2 ]		; <i32*> [#uses=3]
 	%indvar17 = phi i32 [ 0, %loopentry.1 ], [ %indvar.next18, %loopexit.2 ], [ %indvar17, %no_exit.2 ]		; <i32> [#uses=2]
 	%indvar = phi i32 [ %indvar.next, %no_exit.2 ], [ 0, %loopexit.2 ], [ 0, %loopentry.1 ]		; <i32> [#uses=5]
 	%b.1.4.rec = bitcast i32 %indvar to i32		; <i32> [#uses=1]
 	%gep.upgrd.1 = zext i32 %indvar to i64		; <i64> [#uses=1]
-	%c.2.4 = getelementptr [100 x i32]* @C, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
+	%c.2.4 = getelementptr [100 x i32], [100 x i32]* @C, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
 	%gep.upgrd.2 = zext i32 %indvar to i64		; <i64> [#uses=1]
-	%a.0.4 = getelementptr i32* %a.0.4.ph, i64 %gep.upgrd.2		; <i32*> [#uses=1]
+	%a.0.4 = getelementptr i32, i32* %a.0.4.ph, i64 %gep.upgrd.2		; <i32*> [#uses=1]
 	%gep.upgrd.3 = zext i32 %indvar to i64		; <i64> [#uses=1]
-	%b.1.4 = getelementptr i32* %b.1.4.ph, i64 %gep.upgrd.3		; <i32*> [#uses=1]
+	%b.1.4 = getelementptr i32, i32* %b.1.4.ph, i64 %gep.upgrd.3		; <i32*> [#uses=1]
 	%inc.0.rec = add i32 %b.1.4.rec, 1		; <i32> [#uses=2]
-	%inc.0 = getelementptr i32* %a.0.4.ph, i32 %inc.0.rec		; <i32*> [#uses=2]
-	%tmp.13 = load i32* %a.0.4		; <i32> [#uses=1]
-	%inc.1 = getelementptr i32* %b.1.4.ph, i32 %inc.0.rec		; <i32*> [#uses=1]
-	%tmp.15 = load i32* %b.1.4		; <i32> [#uses=1]
-	%tmp.18 = load i32* %c.2.4		; <i32> [#uses=1]
+	%inc.0 = getelementptr i32, i32* %a.0.4.ph, i32 %inc.0.rec		; <i32*> [#uses=2]
+	%tmp.13 = load i32, i32* %a.0.4		; <i32> [#uses=1]
+	%inc.1 = getelementptr i32, i32* %b.1.4.ph, i32 %inc.0.rec		; <i32*> [#uses=1]
+	%tmp.15 = load i32, i32* %b.1.4		; <i32> [#uses=1]
+	%tmp.18 = load i32, i32* %c.2.4		; <i32> [#uses=1]
 	%tmp.16 = mul i32 %tmp.15, %tmp.13		; <i32> [#uses=1]
 	%tmp.19 = mul i32 %tmp.16, %tmp.18		; <i32> [#uses=1]
-	%tmp.20 = load i32* @Z		; <i32> [#uses=1]
+	%tmp.20 = load i32, i32* @Z		; <i32> [#uses=1]
 	%tmp.21 = add i32 %tmp.19, %tmp.20		; <i32> [#uses=1]
 	store i32 %tmp.21, i32* @Z
 	%indvar.next = add i32 %indvar, 1		; <i32> [#uses=2]
diff --git a/test/Transforms/LoopSimplify/preserve-scev.ll b/test/Transforms/LoopSimplify/preserve-scev.ll
index 89626b2af518..f6fa8afc56b5 100644
--- a/test/Transforms/LoopSimplify/preserve-scev.ll
+++ b/test/Transforms/LoopSimplify/preserve-scev.ll
@@ -1,6 +1,9 @@
 ; RUN: opt -S < %s -indvars | opt -analyze -iv-users | grep "%cmp = icmp slt i32" | grep "= {%\.ph,+,1}<%for.cond>"
 ; PR8079
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 ; LoopSimplify should invalidate indvars when splitting out the
 ; inner loop.
 
@@ -15,7 +18,7 @@ for.cond:                                         ; preds = %if.then5, %if.end,
   %0 = phi i32 [ 0, %entry ], [ %add, %if.end ], [ %add, %if.then5 ]
   %add = add i32 %0, 1
   %cmp = icmp slt i32 %0, 1
-  %tmp1 = load i32* @maxStat, align 4
+  %tmp1 = load i32, i32* @maxStat, align 4
   br i1 %cmp, label %for.body, label %for.cond14.preheader
 
 for.cond14.preheader:                             ; preds = %for.cond
@@ -39,7 +42,7 @@ for.body18:                                       ; preds = %for.body18, %for.co
   %i13.027 = phi i32 [ %1, %for.body18 ], [ 0, %for.cond14.preheader ]
   call void @foo() nounwind
   %1 = add nsw i32 %i13.027, 1
-  %tmp16 = load i32* @maxStat, align 4
+  %tmp16 = load i32, i32* @maxStat, align 4
   %cmp17 = icmp slt i32 %1, %tmp16
   br i1 %cmp17, label %for.body18, label %return
 
diff --git a/test/Transforms/LoopStrengthReduce/2005-08-15-AddRecIV.ll b/test/Transforms/LoopStrengthReduce/2005-08-15-AddRecIV.ll
index 1f08a4367bde..7ee1e63ced0f 100644
--- a/test/Transforms/LoopStrengthReduce/2005-08-15-AddRecIV.ll
+++ b/test/Transforms/LoopStrengthReduce/2005-08-15-AddRecIV.ll
@@ -48,9 +48,9 @@ loopexit.3:		; preds = %loopentry.3
 loopentry.4:		; preds = %loopentry.4, %loopexit.3
 	%indvar340 = phi i32 [ 0, %loopexit.3 ], [ %indvar.next341, %loopentry.4 ]		; <i32> [#uses=2]
 	%tmp. = add i32 %indvar340, %indvar342		; <i32> [#uses=1]
-	%tmp.526 = load i32** null		; <i32*> [#uses=1]
+	%tmp.526 = load i32*, i32** null		; <i32*> [#uses=1]
 	%gep.upgrd.1 = zext i32 %tmp. to i64		; <i64> [#uses=1]
-	%tmp.528 = getelementptr i32* %tmp.526, i64 %gep.upgrd.1		; <i32*> [#uses=1]
+	%tmp.528 = getelementptr i32, i32* %tmp.526, i64 %gep.upgrd.1		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp.528
 	%indvar.next341 = add i32 %indvar340, 1		; <i32> [#uses=1]
 	br label %loopentry.4
diff --git a/test/Transforms/LoopStrengthReduce/2005-08-17-OutOfLoopVariant.ll b/test/Transforms/LoopStrengthReduce/2005-08-17-OutOfLoopVariant.ll
index f1c523ae6c60..3e52dff2a876 100644
--- a/test/Transforms/LoopStrengthReduce/2005-08-17-OutOfLoopVariant.ll
+++ b/test/Transforms/LoopStrengthReduce/2005-08-17-OutOfLoopVariant.ll
@@ -6,7 +6,7 @@ loopentry.1.outer:
 	br label %loopentry.1
 loopentry.1:		; preds = %loopentry.1, %loopentry.1.outer
 	%i.3 = phi i32 [ 0, %loopentry.1.outer ], [ %i.3.be, %loopentry.1 ]		; <i32> [#uses=2]
-	%tmp.390 = load i32* null		; <i32> [#uses=1]
+	%tmp.390 = load i32, i32* null		; <i32> [#uses=1]
 	%tmp.392 = mul i32 %tmp.390, %j.2.1.ph		; <i32> [#uses=1]
 	%tmp.394 = add i32 %tmp.392, %i.3		; <i32> [#uses=1]
 	%i.3.be = add i32 %i.3, 1		; <i32> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/2007-04-23-UseIterator.ll b/test/Transforms/LoopStrengthReduce/2007-04-23-UseIterator.ll
index 8c2cfaf32107..44f98011b258 100644
--- a/test/Transforms/LoopStrengthReduce/2007-04-23-UseIterator.ll
+++ b/test/Transforms/LoopStrengthReduce/2007-04-23-UseIterator.ll
@@ -24,7 +24,7 @@ bb16:		; preds = %cond_next102, %bb16.preheader
 
 cond_next102:		; preds = %bb16
 	%tmp138145.rec = add i32 %ABC.2146.0.rec, 3		; <i32> [#uses=1]
-	%tmp138145 = getelementptr i8* %ABC, i32 %tmp138145.rec		; <i8*> [#uses=4]
+	%tmp138145 = getelementptr i8, i8* %ABC, i32 %tmp138145.rec		; <i8*> [#uses=4]
 	%indvar.next = add i32 %indvar, 1		; <i32> [#uses=1]
 	switch i8 0, label %bb129.loopexit [
 		 i8 0, label %UnifiedReturnBlock.loopexit
diff --git a/test/Transforms/LoopStrengthReduce/2008-08-13-CmpStride.ll b/test/Transforms/LoopStrengthReduce/2008-08-13-CmpStride.ll
index ce56bd31018c..394969c9935a 100644
--- a/test/Transforms/LoopStrengthReduce/2008-08-13-CmpStride.ll
+++ b/test/Transforms/LoopStrengthReduce/2008-08-13-CmpStride.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -loop-reduce -S | grep add | count 2
 ; PR 2662
+
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 @g_3 = common global i16 0		; <i16*> [#uses=2]
 @"\01LC" = internal constant [4 x i8] c"%d\0A\00"		; <[4 x i8]*> [#uses=1]
 
@@ -22,9 +26,9 @@ return:		; preds = %bb
 define i32 @main() nounwind {
 entry:
 	tail call void @func_1( ) nounwind
-	load volatile i16* @g_3, align 2		; <i16>:0 [#uses=1]
+	load volatile i16, i16* @g_3, align 2		; <i16>:0 [#uses=1]
 	zext i16 %0 to i32		; <i32>:1 [#uses=1]
-	tail call i32 (i8*, ...)* @printf( i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), i32 %1 ) nounwind		; <i32>:2 [#uses=0]
+	tail call i32 (i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), i32 %1 ) nounwind		; <i32>:2 [#uses=0]
 	ret i32 0
 }
 
diff --git a/test/Transforms/LoopStrengthReduce/2008-09-09-Overflow.ll b/test/Transforms/LoopStrengthReduce/2008-09-09-Overflow.ll
index 1ee6b5cdf18e..5904434e14ff 100644
--- a/test/Transforms/LoopStrengthReduce/2008-09-09-Overflow.ll
+++ b/test/Transforms/LoopStrengthReduce/2008-09-09-Overflow.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -loop-reduce -S | grep phi | count 2
 ; PR 2779
+
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 @g_19 = common global i32 0		; <i32*> [#uses=3]
 @"\01LC" = internal constant [4 x i8] c"%d\0A\00"		; <[4 x i8]*> [#uses=1]
 
@@ -21,7 +25,7 @@ bb:		; preds = %bb, %entry
 	%indvar = phi i16 [ 0, %entry ], [ %indvar.next, %bb ]		; <i16> [#uses=2]
 	%tmp = sub i16 0, %indvar		; <i16> [#uses=1]
 	%tmp27 = trunc i16 %tmp to i8		; <i8> [#uses=1]
-	load i32* @g_19, align 4		; <i32>:0 [#uses=2]
+	load i32, i32* @g_19, align 4		; <i32>:0 [#uses=2]
 	add i32 %0, 1		; <i32>:1 [#uses=1]
 	store i32 %1, i32* @g_19, align 4
 	trunc i32 %0 to i8		; <i8>:2 [#uses=1]
@@ -40,8 +44,8 @@ return:		; preds = %bb
 define i32 @main() nounwind {
 entry:
 	tail call void @func_1( ) nounwind
-	load i32* @g_19, align 4		; <i32>:0 [#uses=1]
-	tail call i32 (i8*, ...)* @printf( i8* getelementptr ([4 x i8]* @"\01LC", i32 0, i32 0), i32 %0 ) nounwind		; <i32>:1 [#uses=0]
+	load i32, i32* @g_19, align 4		; <i32>:0 [#uses=1]
+	tail call i32 (i8*, ...) @printf( i8* getelementptr ([4 x i8], [4 x i8]* @"\01LC", i32 0, i32 0), i32 %0 ) nounwind		; <i32>:1 [#uses=0]
 	ret i32 0
 }
 
diff --git a/test/Transforms/LoopStrengthReduce/2009-01-13-nonconstant-stride-outside-loop.ll b/test/Transforms/LoopStrengthReduce/2009-01-13-nonconstant-stride-outside-loop.ll
index b2cf818dc45d..c287b10e6946 100644
--- a/test/Transforms/LoopStrengthReduce/2009-01-13-nonconstant-stride-outside-loop.ll
+++ b/test/Transforms/LoopStrengthReduce/2009-01-13-nonconstant-stride-outside-loop.ll
@@ -26,14 +26,14 @@ bb1.i:		; preds = %bb.i2, %entry
 	br i1 %0, label %bb2.i3, label %nactive_heaps.exit
 
 bb2.i3:		; preds = %bb1.i
-	%1 = load %struct.obj** null, align 4		; <%struct.obj*> [#uses=1]
+	%1 = load %struct.obj*, %struct.obj** null, align 4		; <%struct.obj*> [#uses=1]
 	%2 = icmp eq %struct.obj* %1, null		; <i1> [#uses=1]
 	br i1 %2, label %nactive_heaps.exit, label %bb.i2
 
 nactive_heaps.exit:		; preds = %bb2.i3, %bb1.i
-	%3 = load i32* @heap_size, align 4		; <i32> [#uses=1]
+	%3 = load i32, i32* @heap_size, align 4		; <i32> [#uses=1]
 	%4 = mul i32 %3, %m.0.i		; <i32> [#uses=1]
 	%5 = sub i32 %4, 0		; <i32> [#uses=1]
-	%6 = tail call i32 (i8*, i8*, ...)* @sprintf(i8* null, i8* getelementptr ([39 x i8]* @"\01LC85", i32 0, i32 0), i32 %m.0.i, i32 0, i32 %5, i32 0) nounwind		; <i32> [#uses=0]
+	%6 = tail call i32 (i8*, i8*, ...) @sprintf(i8* null, i8* getelementptr ([39 x i8], [39 x i8]* @"\01LC85", i32 0, i32 0), i32 %m.0.i, i32 0, i32 %5, i32 0) nounwind		; <i32> [#uses=0]
 	ret %struct.obj* null
 }
diff --git a/test/Transforms/LoopStrengthReduce/2009-04-28-no-reduce-mul.ll b/test/Transforms/LoopStrengthReduce/2009-04-28-no-reduce-mul.ll
index 002a878b7b24..8e890e84bccf 100644
--- a/test/Transforms/LoopStrengthReduce/2009-04-28-no-reduce-mul.ll
+++ b/test/Transforms/LoopStrengthReduce/2009-04-28-no-reduce-mul.ll
@@ -21,7 +21,7 @@ bb2:		; preds = %bb4, %bb2, %bb4.thread
 	%0 = trunc i32 %j.0.reg2mem.0 to i8		; <i8> [#uses=1]
 	%1 = sext i8 %0 to i32		; <i32> [#uses=1]
 	%2 = mul i32 %1, %i.0.reg2mem.0.ph		; <i32> [#uses=1]
-	%3 = getelementptr [32 x [256 x i32]]* @table, i32 0, i32 %i.0.reg2mem.0.ph, i32 %j.0.reg2mem.0		; <i32*> [#uses=1]
+	%3 = getelementptr [32 x [256 x i32]], [32 x [256 x i32]]* @table, i32 0, i32 %i.0.reg2mem.0.ph, i32 %j.0.reg2mem.0		; <i32*> [#uses=1]
 	store i32 %2, i32* %3, align 4
 	%indvar.next = add i32 %j.0.reg2mem.0, 1		; <i32> [#uses=2]
 	%exitcond = icmp eq i32 %indvar.next, 256		; <i1> [#uses=1]
@@ -33,7 +33,7 @@ bb4:		; preds = %bb2
 	br i1 %exitcond10, label %bb5, label %bb2
 
 bb5:		; preds = %bb4
-	%4 = load i32* getelementptr ([32 x [256 x i32]]* @table, i32 0, i32 9, i32 132), align 16		; <i32> [#uses=1]
+	%4 = load i32, i32* getelementptr ([32 x [256 x i32]], [32 x [256 x i32]]* @table, i32 0, i32 9, i32 132), align 16		; <i32> [#uses=1]
 	%5 = icmp eq i32 %4, -1116		; <i1> [#uses=1]
 	br i1 %5, label %bb7, label %bb6
 
diff --git a/test/Transforms/LoopStrengthReduce/2011-07-19-CritEdgeBreakCrash.ll b/test/Transforms/LoopStrengthReduce/2011-07-19-CritEdgeBreakCrash.ll
index b9bd7c9bf4a3..cf549fc38acd 100644
--- a/test/Transforms/LoopStrengthReduce/2011-07-19-CritEdgeBreakCrash.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-07-19-CritEdgeBreakCrash.ll
@@ -24,12 +24,12 @@ H_MPZ_LBL:                                        ; preds = %indirectgoto
 while.cond.i:                                     ; preds = %while.body.i15795, %if.then.i15791
   %phi = phi i64 [ %tmp20916, %while.body.i15795 ], [ 0, %H_MPZ_LBL ]
   %tmp20916 = add i64 %phi, 1
-  %incdec.ptr.i15793 = getelementptr i64* %pc.0.lcssa.i1610719352, i64 %tmp20916
+  %incdec.ptr.i15793 = getelementptr i64, i64* %pc.0.lcssa.i1610719352, i64 %tmp20916
   %boo2 = call i1 @foo()
   br i1 %boo2, label %indirectgoto, label %while.body.i15795
 
 while.body.i15795:                                ; preds = %while.cond.i
-  %tmp20.i = load i64* %incdec.ptr.i15793, align 8
+  %tmp20.i = load i64, i64* %incdec.ptr.i15793, align 8
   %boo1 = call i1 @foo()
   br i1 %boo1, label %while.cond.i, label %body_failed
 
diff --git a/test/Transforms/LoopStrengthReduce/2011-10-03-CritEdgeMerge.ll b/test/Transforms/LoopStrengthReduce/2011-10-03-CritEdgeMerge.ll
index ccf8ebdd5d13..5588995a8ce8 100644
--- a/test/Transforms/LoopStrengthReduce/2011-10-03-CritEdgeMerge.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-10-03-CritEdgeMerge.ll
@@ -4,6 +4,10 @@
 
 target triple = "x86-apple-darwin"
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 ; Verify that identical edges are merged. rdar://problem/6453893
 ; CHECK-LABEL: @test1(
 ; CHECK: bb89:
@@ -16,7 +20,7 @@ entry:
 loop:
   %rec = phi i32 [ %next, %loop ], [ 0, %entry ]
   %next = add i32 %rec, 1
-  %tmp75 = getelementptr i8* null, i32 %next
+  %tmp75 = getelementptr i8, i8* null, i32 %next
   br i1 false, label %loop, label %loopexit
 
 loopexit:
@@ -53,7 +57,7 @@ entry:
 loop:
   %rec = phi i32 [ %next, %loop ], [ 0, %entry ]
   %next = add i32 %rec, 1
-  %tmp75 = getelementptr i8* null, i32 %next
+  %tmp75 = getelementptr i8, i8* null, i32 %next
   br i1 false, label %loop, label %loopexit
 
 loopexit:
diff --git a/test/Transforms/LoopStrengthReduce/2011-10-06-ReusePhi.ll b/test/Transforms/LoopStrengthReduce/2011-10-06-ReusePhi.ll
index 83963e3126da..cc8226e12589 100644
--- a/test/Transforms/LoopStrengthReduce/2011-10-06-ReusePhi.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-10-06-ReusePhi.ll
@@ -5,6 +5,10 @@
 
 target triple = "x86-apple-darwin"
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 ; CHECK-LABEL: @test(
 ; multiplies are hoisted out of the loop
 ; CHECK: while.body.lr.ph:
@@ -17,10 +21,10 @@ target triple = "x86-apple-darwin"
 ; CHECK: phi
 ; CHECK: phi
 ; CHECK-NOT: phi
-; CHECK: bitcast float* {{.*}} to i8*
-; CHECK: bitcast float* {{.*}} to i8*
-; CHECK: getelementptr i8*
-; CHECK: getelementptr i8*
+; CHECK: bitcast float* {{.*}} to i1*
+; CHECK: bitcast float* {{.*}} to i1*
+; CHECK: getelementptr i1, i1*
+; CHECK: getelementptr i1, i1*
 
 define float @test(float* nocapture %A, float* nocapture %B, i32 %N, i32 %IA, i32 %IB) nounwind uwtable readonly ssp {
 entry:
@@ -37,12 +41,12 @@ while.body:                                       ; preds = %while.body.lr.ph, %
   %B.addr.04 = phi float* [ %B, %while.body.lr.ph ], [ %add.ptr3, %while.body ]
   %N.addr.03 = phi i32 [ %N, %while.body.lr.ph ], [ %sub, %while.body ]
   %Sum0.02 = phi float [ 0.000000e+00, %while.body.lr.ph ], [ %add, %while.body ]
-  %0 = load float* %A.addr.05, align 4
-  %1 = load float* %B.addr.04, align 4
+  %0 = load float, float* %A.addr.05, align 4
+  %1 = load float, float* %B.addr.04, align 4
   %mul = fmul float %0, %1
   %add = fadd float %Sum0.02, %mul
-  %add.ptr = getelementptr inbounds float* %A.addr.05, i64 %idx.ext
-  %add.ptr3 = getelementptr inbounds float* %B.addr.04, i64 %idx.ext2
+  %add.ptr = getelementptr inbounds float, float* %A.addr.05, i64 %idx.ext
+  %add.ptr3 = getelementptr inbounds float, float* %B.addr.04, i64 %idx.ext2
   %sub = add nsw i32 %N.addr.03, -1
   %cmp = icmp sgt i32 %sub, 0
   br i1 %cmp, label %while.body, label %while.end
diff --git a/test/Transforms/LoopStrengthReduce/2011-10-13-SCEVChain.ll b/test/Transforms/LoopStrengthReduce/2011-10-13-SCEVChain.ll
index 484fefaad413..a61de24370f5 100644
--- a/test/Transforms/LoopStrengthReduce/2011-10-13-SCEVChain.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-10-13-SCEVChain.ll
@@ -95,14 +95,14 @@ bb1:
   %t15 = icmp ugt i32 %n15, -4
   %m15 = select i1 %t15, i32 %n15, i32 -4
   %a16 = add i32 %m15, %a15
-  %gep = getelementptr i8* %base, i32 %a16
+  %gep = getelementptr i8, i8* %base, i32 %a16
   %ofs = add i32 %a16, 4
-  %limit = getelementptr i8* %base, i32 %ofs
+  %limit = getelementptr i8, i8* %base, i32 %ofs
   br label %loop
 
 loop:
   %iv = phi i8* [ %gep, %bb1 ], [ %inc, %loop ]
-  %inc = getelementptr inbounds i8* %iv, i64 1
+  %inc = getelementptr inbounds i8, i8* %iv, i64 1
   %exitcond = icmp eq i8* %inc, %limit
   br i1 %exitcond, label %loop, label %exit
 
diff --git a/test/Transforms/LoopStrengthReduce/2011-10-14-IntPtr.ll b/test/Transforms/LoopStrengthReduce/2011-10-14-IntPtr.ll
index 068b716651d8..fe9f8cb2de4c 100644
--- a/test/Transforms/LoopStrengthReduce/2011-10-14-IntPtr.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-10-14-IntPtr.ll
@@ -22,6 +22,6 @@ for.end:                                          ; preds = %for.end.critedge, %
 for.body83:                                       ; preds = %for.body83, %for.end
   %ptr.0157 = phi i8* [ %add.ptr96, %for.body83 ], [ null, %for.end ]
   store i8 undef, i8* %ptr.0157, align 1
-  %add.ptr96 = getelementptr inbounds i8* %ptr.0157, i32 %cond
+  %add.ptr96 = getelementptr inbounds i8, i8* %ptr.0157, i32 %cond
   br label %for.body83
 }
diff --git a/test/Transforms/LoopStrengthReduce/2011-12-19-PostincQuadratic.ll b/test/Transforms/LoopStrengthReduce/2011-12-19-PostincQuadratic.ll
index 6c128feb541c..317b0b0b2939 100644
--- a/test/Transforms/LoopStrengthReduce/2011-12-19-PostincQuadratic.ll
+++ b/test/Transforms/LoopStrengthReduce/2011-12-19-PostincQuadratic.ll
@@ -15,7 +15,7 @@ target triple = "i386-unknown-freebsd10.0"
 ;   Inner recurrence:
 ; CHECK: %lsr.iv = phi i32
 ;   Outer step (relative to inner recurrence):
-; CHECK: %scevgep = getelementptr i1* %{{.*}}, i32 %lsr.iv
+; CHECK: %scevgep = getelementptr i1, i1* %{{.*}}, i32 %lsr.iv
 ;   Outer use:
 ; CHECK: %lsr.iv3 = phi [121 x i32]* [ %lsr.iv1, %for.body43.preheader ]
 define void @vb() nounwind {
@@ -32,8 +32,8 @@ for.body7:
 for.body43:
   %bf.459 = phi i32 [ %inc44, %for.body43 ], [ %t1, %for.body7 ]
   %inc44 = add nsw i32 %bf.459, 1
-  %arrayidx45 = getelementptr inbounds [121 x i32]* @b, i32 0, i32 %bf.459
-  %t2 = load i32* %arrayidx45, align 4
+  %arrayidx45 = getelementptr inbounds [121 x i32], [121 x i32]* @b, i32 0, i32 %bf.459
+  %t2 = load i32, i32* %arrayidx45, align 4
   br label %for.body43
 }
 
diff --git a/test/Transforms/LoopStrengthReduce/2012-01-02-nopreheader.ll b/test/Transforms/LoopStrengthReduce/2012-01-02-nopreheader.ll
index 87dd39730ec1..d8aa26402f72 100644
--- a/test/Transforms/LoopStrengthReduce/2012-01-02-nopreheader.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-01-02-nopreheader.ll
@@ -35,7 +35,7 @@ bb7.lr.ph.us:                                     ; preds = %bb9.us, %bb8.prehea
 bb7.us:                                           ; preds = %bb7.lr.ph.us, %bb7.us
   %j.01.us = phi i32 [ 0, %bb7.lr.ph.us ], [ %1, %bb7.us ]
   %tmp31 = add i32 %tmp30, %j.01.us
-  %scevgep9 = getelementptr float* %a, i32 %tmp31
+  %scevgep9 = getelementptr float, float* %a, i32 %tmp31
   store float undef, float* %scevgep9, align 1
   %1 = add nsw i32 %j.01.us, 1
   indirectbr i8* undef, [label %bb9.us, label %bb7.us]
@@ -74,7 +74,7 @@ bb8.preheader:                                    ; preds = %bb9, %bb3
 
 bb7:                                              ; preds = %bb8.preheader, %bb7
   %indvar = phi i32 [ 0, %bb8.preheader ], [ %indvar.next, %bb7 ]
-  %scevgep = getelementptr [200 x i32]* %Array2, i32 %tmp26, i32 %indvar
+  %scevgep = getelementptr [200 x i32], [200 x i32]* %Array2, i32 %tmp26, i32 %indvar
   store i32 undef, i32* %scevgep, align 4
   %indvar.next = add i32 %indvar, 1
   indirectbr i8* undef, [label %bb9, label %bb7]
diff --git a/test/Transforms/LoopStrengthReduce/2012-01-16-nopreheader.ll b/test/Transforms/LoopStrengthReduce/2012-01-16-nopreheader.ll
index 94a037ec28eb..7f1eed03a713 100644
--- a/test/Transforms/LoopStrengthReduce/2012-01-16-nopreheader.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-01-16-nopreheader.ll
@@ -48,7 +48,7 @@ while.cond238.preheader:                          ; preds = %lor.lhs.false227, %
 while.cond238:                                    ; preds = %land.rhs243, %while.cond238.preheader
   %1 = phi i64 [ %indvar.next15, %land.rhs243 ], [ 0, %while.cond238.preheader ]
   %tmp36 = add i64 %tmp16, %1
-  %s.3 = getelementptr i8* %s.1, i64 %tmp36
+  %s.3 = getelementptr i8, i8* %s.1, i64 %tmp36
   %cmp241 = icmp ult i8* %s.3, %end
   indirectbr i8* undef, [label %land.rhs243, label %while.end256]
 
@@ -75,7 +75,7 @@ if.end340:                                        ; preds = %while.end334
   indirectbr i8* undef, [label %PREMATURE, label %if.end348]
 
 if.end348:                                        ; preds = %if.end340
-  %incdec.ptr356 = getelementptr inbounds i8* undef, i64 2
+  %incdec.ptr356 = getelementptr inbounds i8, i8* undef, i64 2
   indirectbr i8* undef, [label %while.cond179]
 
 if.else386:                                       ; preds = %if.else
diff --git a/test/Transforms/LoopStrengthReduce/2012-03-15-nopreheader.ll b/test/Transforms/LoopStrengthReduce/2012-03-15-nopreheader.ll
index 5fa3838c8297..62064cb78813 100644
--- a/test/Transforms/LoopStrengthReduce/2012-03-15-nopreheader.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-03-15-nopreheader.ll
@@ -34,8 +34,8 @@ lor.lhs.false184:                                 ; preds = %lor.lhs.false177
 for.cond:                                         ; preds = %for.inc, %lor.lhs.false184, %if.end152
   %indvar65 = phi i64 [ %indvar.next66, %for.inc ], [ 0, %lor.lhs.false184 ], [ 0, %if.end152 ]
   %tmp128 = add i64 %0, %indvar65
-  %s.4 = getelementptr i8* %cmd, i64 %tmp128
-  %tmp195 = load i8* %s.4, align 1
+  %s.4 = getelementptr i8, i8* %cmd, i64 %tmp128
+  %tmp195 = load i8, i8* %s.4, align 1
   indirectbr i8* undef, [label %return, label %land.rhs198]
 
 land.rhs198:                                      ; preds = %for.cond
diff --git a/test/Transforms/LoopStrengthReduce/2012-03-26-constexpr.ll b/test/Transforms/LoopStrengthReduce/2012-03-26-constexpr.ll
index c9b11a9feed0..ce6161cb1b45 100644
--- a/test/Transforms/LoopStrengthReduce/2012-03-26-constexpr.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-03-26-constexpr.ll
@@ -10,11 +10,11 @@ target datalayout = "e-p:64:64:64-n32:64"
 
 define internal fastcc void @someFunction(%struct.this_structure_s.0.5* nocapture %scratch, i32 %stage, i32 %cbSize) nounwind {
 entry:
-  %0 = getelementptr inbounds %struct.this_structure_s.0.5* %scratch, i32 0, i32 4, i32 %stage
-  %1 = load i8** %0, align 4
-  %2 = getelementptr inbounds %struct.this_structure_s.0.5* %scratch, i32 0, i32 5, i32 %stage
-  %3 = load i8** %2, align 4
-  %4 = getelementptr inbounds %struct.this_structure_s.0.5* %scratch, i32 0, i32 2, i32 0, i32 0
+  %0 = getelementptr inbounds %struct.this_structure_s.0.5, %struct.this_structure_s.0.5* %scratch, i32 0, i32 4, i32 %stage
+  %1 = load i8*, i8** %0, align 4
+  %2 = getelementptr inbounds %struct.this_structure_s.0.5, %struct.this_structure_s.0.5* %scratch, i32 0, i32 5, i32 %stage
+  %3 = load i8*, i8** %2, align 4
+  %4 = getelementptr inbounds %struct.this_structure_s.0.5, %struct.this_structure_s.0.5* %scratch, i32 0, i32 2, i32 0, i32 0
   %tmp11 = shl i32 %stage, 1
   %tmp1325 = or i32 %tmp11, 1
   br label %__label_D_1608
@@ -22,18 +22,18 @@ entry:
 __label_D_1608:                                   ; preds = %__label_D_1608, %entry
   %i.12 = phi i32 [ 0, %entry ], [ %10, %__label_D_1608 ]
   %tmp = shl i32 %i.12, 2
-  %lvar_g.13 = getelementptr i32* %4, i32 %tmp
+  %lvar_g.13 = getelementptr i32, i32* %4, i32 %tmp
   %tmp626 = or i32 %tmp, 1
-  %scevgep = getelementptr i32* %4, i32 %tmp626
+  %scevgep = getelementptr i32, i32* %4, i32 %tmp626
   %tmp727 = or i32 %tmp, 2
-  %scevgep8 = getelementptr i32* %4, i32 %tmp727
+  %scevgep8 = getelementptr i32, i32* %4, i32 %tmp727
   %tmp928 = or i32 %tmp, 3
-  %scevgep10 = getelementptr i32* %4, i32 %tmp928
-  %scevgep12 = getelementptr %struct.this_structure_s.0.5* %scratch, i32 0, i32 9, i32 %tmp11, i32 %i.12
-  %scevgep14 = getelementptr %struct.this_structure_s.0.5* %scratch, i32 0, i32 9, i32 %tmp1325, i32 %i.12
-  %5 = load i8* %scevgep12, align 1
+  %scevgep10 = getelementptr i32, i32* %4, i32 %tmp928
+  %scevgep12 = getelementptr %struct.this_structure_s.0.5, %struct.this_structure_s.0.5* %scratch, i32 0, i32 9, i32 %tmp11, i32 %i.12
+  %scevgep14 = getelementptr %struct.this_structure_s.0.5, %struct.this_structure_s.0.5* %scratch, i32 0, i32 9, i32 %tmp1325, i32 %i.12
+  %5 = load i8, i8* %scevgep12, align 1
   %6 = sext i8 %5 to i32
-  %7 = load i8* %scevgep14, align 1
+  %7 = load i8, i8* %scevgep14, align 1
   %8 = sext i8 %7 to i32
   store i32 0, i32* %lvar_g.13, align 4
   store i32 %8, i32* %scevgep, align 4
diff --git a/test/Transforms/LoopStrengthReduce/2012-07-13-ExpandUDiv.ll b/test/Transforms/LoopStrengthReduce/2012-07-13-ExpandUDiv.ll
index ea1d65b1652e..80095c3ac154 100644
--- a/test/Transforms/LoopStrengthReduce/2012-07-13-ExpandUDiv.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-07-13-ExpandUDiv.ll
@@ -17,12 +17,12 @@ target triple = "x86_64-apple-darwin"
 define i32 @main() nounwind uwtable ssp {
 entry:
   %l_2 = alloca [1 x i32], align 4
-  %arrayidx = getelementptr inbounds [1 x i32]* %l_2, i64 0, i64 0
+  %arrayidx = getelementptr inbounds [1 x i32], [1 x i32]* %l_2, i64 0, i64 0
   store i32 0, i32* %arrayidx, align 4
-  %tmp = load i32* @g_3, align 4
+  %tmp = load i32, i32* @g_3, align 4
   %idxprom = sext i32 %tmp to i64
-  %arrayidx1 = getelementptr inbounds [1 x i32]* %l_2, i64 0, i64 %idxprom
-  %tmp1 = load i32* %arrayidx1, align 4
+  %arrayidx1 = getelementptr inbounds [1 x i32], [1 x i32]* %l_2, i64 0, i64 %idxprom
+  %tmp1 = load i32, i32* %arrayidx1, align 4
   %conv.i.i = and i32 %tmp1, 65535
   %tobool.i.i.i = icmp ne i32 %tmp, 0
   br label %codeRepl
diff --git a/test/Transforms/LoopStrengthReduce/2012-07-18-LimitReassociate.ll b/test/Transforms/LoopStrengthReduce/2012-07-18-LimitReassociate.ll
index 9524be3ceee0..aaac868dfd28 100644
--- a/test/Transforms/LoopStrengthReduce/2012-07-18-LimitReassociate.ll
+++ b/test/Transforms/LoopStrengthReduce/2012-07-18-LimitReassociate.ll
@@ -1,5 +1,4 @@
-; RUN: opt -loop-reduce -disable-output -debug-only=loop-reduce < %s 2> %t
-; RUN: FileCheck %s < %t
+; RUN: opt -loop-reduce -disable-output -debug-only=loop-reduce < %s 2>&1 | FileCheck %s
 ; REQUIRES: asserts
 ;
 ; PR13361: LSR + SCEV "hangs" on reasonably sized test with sequence of loops
@@ -18,6 +17,10 @@
 ; CHECK-NOT:reg
 ; CHECK: Filtering for use
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 %struct.snork = type { %struct.fuga, i32, i32, i32, i32, i32, i32 }
 %struct.fuga = type { %struct.gork, i64 }
 %struct.gork = type { i8*, i32, i32, %struct.noot* }
@@ -44,18 +47,18 @@ bb:
   br i1 %tmp4, label %bb6, label %bb5
 
 bb5:                                              ; preds = %bb
-  tail call void (...)* @snork(i8* getelementptr inbounds ([52 x i8]* @global1, i64 0, i64 0), i32 2021) nounwind
-  tail call void (...)* @snork(i8* getelementptr inbounds (%struct.jim* @global3, i64 0, i32 3, i64 1), i32 -2146631418) nounwind
+  tail call void (...) @snork(i8* getelementptr inbounds ([52 x i8], [52 x i8]* @global1, i64 0, i64 0), i32 2021) nounwind
+  tail call void (...) @snork(i8* getelementptr inbounds (%struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 1), i32 -2146631418) nounwind
   unreachable
 
 bb6:                                              ; preds = %bb
-  tail call void @zot(i8* getelementptr inbounds (%struct.jim* @global3, i64 0, i32 5, i64 0), i8* getelementptr inbounds (%struct.jim* @global3, i64 0, i32 3, i64 1), i64 undef, i32 1, i1 false) nounwind
-  %tmp7 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 5, i64 undef
+  tail call void @zot(i8* getelementptr inbounds (%struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 0), i8* getelementptr inbounds (%struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 1), i64 undef, i32 1, i1 false) nounwind
+  %tmp7 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 undef
   store i8 0, i8* %tmp7, align 1
   %tmp8 = add nsw i32 0, 1
   %tmp9 = sext i32 %tmp8 to i64
   %tmp10 = add i64 %tmp9, 1
-  %tmp11 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 3, i64 %tmp10
+  %tmp11 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 %tmp10
   %tmp12 = sub i64 2047, %tmp9
   %tmp13 = icmp eq i32 undef, 1
   br i1 %tmp13, label %bb14, label %bb15
@@ -76,8 +79,8 @@ bb17:                                             ; preds = %bb26, %bb15
   br i1 %tmp21, label %bb22, label %bb32
 
 bb22:                                             ; preds = %bb17
-  %tmp23 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 3, i64 0
-  %tmp24 = load i8* %tmp23, align 1
+  %tmp23 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 0
+  %tmp24 = load i8, i8* %tmp23, align 1
   %tmp25 = icmp eq i8 %tmp24, 58
   br i1 %tmp25, label %bb30, label %bb26
 
@@ -92,18 +95,18 @@ bb30:                                             ; preds = %bb22
   br i1 %tmp31, label %bb33, label %bb32
 
 bb32:                                             ; preds = %bb30, %bb26, %bb17
-  tail call void (...)* @snork(i8* getelementptr inbounds ([52 x i8]* @global1, i64 0, i64 0), i32 2038) nounwind
-  tail call void (...)* @snork(i8* %tmp11, i32 -2146631418) nounwind
+  tail call void (...) @snork(i8* getelementptr inbounds ([52 x i8], [52 x i8]* @global1, i64 0, i64 0), i32 2038) nounwind
+  tail call void (...) @snork(i8* %tmp11, i32 -2146631418) nounwind
   unreachable
 
 bb33:                                             ; preds = %bb30
-  tail call void @zot(i8* getelementptr inbounds (%struct.jim* @global3, i64 0, i32 5, i64 0), i8* %tmp11, i64 undef, i32 1, i1 false) nounwind
-  %tmp34 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 5, i64 undef
+  tail call void @zot(i8* getelementptr inbounds (%struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 0), i8* %tmp11, i64 undef, i32 1, i1 false) nounwind
+  %tmp34 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 undef
   store i8 0, i8* %tmp34, align 1
   %tmp35 = add nsw i32 %tmp19, 1
   %tmp36 = sext i32 %tmp35 to i64
   %tmp37 = add i64 %tmp36, %tmp10
-  %tmp38 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 3, i64 %tmp37
+  %tmp38 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 %tmp37
   %tmp39 = sub i64 %tmp12, %tmp36
   br i1 false, label %bb40, label %bb41
 
@@ -123,7 +126,7 @@ bb43:                                             ; preds = %bb52, %bb41
 
 bb48:                                             ; preds = %bb43
   %tmp49 = add i64 %tmp44, %tmp37
-  %tmp50 = load i8* undef, align 1
+  %tmp50 = load i8, i8* undef, align 1
   %tmp51 = icmp eq i8 %tmp50, 58
   br i1 %tmp51, label %bb55, label %bb52
 
@@ -138,14 +141,14 @@ bb55:                                             ; preds = %bb48
   br i1 %tmp57, label %bb59, label %bb58
 
 bb58:                                             ; preds = %bb55, %bb52, %bb43
-  tail call void (...)* @snork(i8* getelementptr inbounds ([52 x i8]* @global1, i64 0, i64 0), i32 2055) nounwind
-  tail call void (...)* @snork(i8* %tmp38, i32 -2146631418) nounwind
+  tail call void (...) @snork(i8* getelementptr inbounds ([52 x i8], [52 x i8]* @global1, i64 0, i64 0), i32 2055) nounwind
+  tail call void (...) @snork(i8* %tmp38, i32 -2146631418) nounwind
   br label %bb247
 
 bb59:                                             ; preds = %bb55
   %tmp60 = sext i32 %tmp45 to i64
-  tail call void @zot(i8* getelementptr inbounds (%struct.jim* @global3, i64 0, i32 5, i64 0), i8* %tmp38, i64 %tmp60, i32 1, i1 false) nounwind
-  %tmp61 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 5, i64 %tmp60
+  tail call void @zot(i8* getelementptr inbounds (%struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 0), i8* %tmp38, i64 %tmp60, i32 1, i1 false) nounwind
+  %tmp61 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 5, i64 %tmp60
   store i8 0, i8* %tmp61, align 1
   %tmp62 = add nsw i32 %tmp45, 1
   %tmp63 = sext i32 %tmp62 to i64
@@ -165,12 +168,12 @@ bb68:                                             ; preds = %bb59
   ]
 
 bb69:                                             ; preds = %bb68
-  tail call void (...)* @snork(i8* getelementptr inbounds ([52 x i8]* @global1, i64 0, i64 0), i32 2071) nounwind
-  %tmp70 = load i32* getelementptr inbounds (%struct.snork* @global, i64 0, i32 2), align 4
+  tail call void (...) @snork(i8* getelementptr inbounds ([52 x i8], [52 x i8]* @global1, i64 0, i64 0), i32 2071) nounwind
+  %tmp70 = load i32, i32* getelementptr inbounds (%struct.snork, %struct.snork* @global, i64 0, i32 2), align 4
   unreachable
 
 bb71:                                             ; preds = %bb68
-  %tmp72 = load i32* getelementptr inbounds (%struct.snork* @global, i64 0, i32 4), align 4
+  %tmp72 = load i32, i32* getelementptr inbounds (%struct.snork, %struct.snork* @global, i64 0, i32 4), align 4
   %tmp73 = icmp eq i32 undef, 0
   br i1 %tmp73, label %bb247, label %bb74
 
@@ -461,8 +464,8 @@ bb221:                                            ; preds = %bb230, %bb219
 
 bb226:                                            ; preds = %bb221
   %tmp227 = add i64 %tmp222, %tmp216
-  %tmp228 = getelementptr inbounds %struct.jim* @global3, i64 0, i32 3, i64 %tmp227
-  %tmp229 = load i8* %tmp228, align 1
+  %tmp228 = getelementptr inbounds %struct.jim, %struct.jim* @global3, i64 0, i32 3, i64 %tmp227
+  %tmp229 = load i8, i8* %tmp228, align 1
   br i1 false, label %bb233, label %bb230
 
 bb230:                                            ; preds = %bb226
diff --git a/test/Transforms/LoopStrengthReduce/2013-01-14-ReuseCast.ll b/test/Transforms/LoopStrengthReduce/2013-01-14-ReuseCast.ll
index 79dbf0d53703..dcd068191e10 100644
--- a/test/Transforms/LoopStrengthReduce/2013-01-14-ReuseCast.ll
+++ b/test/Transforms/LoopStrengthReduce/2013-01-14-ReuseCast.ll
@@ -72,8 +72,8 @@ bb61:                                             ; preds = %bb63, %bb58
   br i1 %tmp62, label %_ZNK4llvm9StringRef4findEcm.exit._crit_edge, label %bb63
 
 bb63:                                             ; preds = %bb61
-  %tmp64 = getelementptr inbounds i8* %tmp3, i64 %i.0.i
-  %tmp65 = load i8* %tmp64, align 1
+  %tmp64 = getelementptr inbounds i8, i8* %tmp3, i64 %i.0.i
+  %tmp65 = load i8, i8* %tmp64, align 1
   %tmp67 = add i64 %i.0.i, 1
   br i1 undef, label %_ZNK4llvm9StringRef4findEcm.exit.loopexit, label %bb61
 
diff --git a/test/Transforms/LoopStrengthReduce/AArch64/lsr-memcpy.ll b/test/Transforms/LoopStrengthReduce/AArch64/lsr-memcpy.ll
index 9a175ad8d355..2120b2a3bfe7 100644
--- a/test/Transforms/LoopStrengthReduce/AArch64/lsr-memcpy.ll
+++ b/test/Transforms/LoopStrengthReduce/AArch64/lsr-memcpy.ll
@@ -19,9 +19,9 @@ while.body:                                       ; preds = %while.body, %entry
   %len.06 = phi i64 [ 1288, %entry ], [ %sub, %while.body ]
   %pDst.05 = phi i64* [ inttoptr (i64 6442450944 to i64*), %entry ], [ %incdec.ptr1, %while.body ]
   %pSrc.04 = phi i64* [ inttoptr (i64 4294967296 to i64*), %entry ], [ %incdec.ptr, %while.body ]
-  %incdec.ptr = getelementptr inbounds i64* %pSrc.04, i64 1
-  %tmp = load volatile i64* %pSrc.04, align 8
-  %incdec.ptr1 = getelementptr inbounds i64* %pDst.05, i64 1
+  %incdec.ptr = getelementptr inbounds i64, i64* %pSrc.04, i64 1
+  %tmp = load volatile i64, i64* %pSrc.04, align 8
+  %incdec.ptr1 = getelementptr inbounds i64, i64* %pDst.05, i64 1
   store volatile i64 %tmp, i64* %pDst.05, align 8
   %sub = add i64 %len.06, -8
   %cmp = icmp sgt i64 %sub, -1
diff --git a/test/Transforms/LoopStrengthReduce/AArch64/lsr-memset.ll b/test/Transforms/LoopStrengthReduce/AArch64/lsr-memset.ll
index 48b709429467..e84f675736d1 100644
--- a/test/Transforms/LoopStrengthReduce/AArch64/lsr-memset.ll
+++ b/test/Transforms/LoopStrengthReduce/AArch64/lsr-memset.ll
@@ -38,7 +38,7 @@ land.rhs:                                         ; preds = %while.cond
   br i1 %cmp5, label %if.end9, label %while.body
 
 while.body:                                       ; preds = %land.rhs
-  %incdec.ptr = getelementptr inbounds i8* %ptr.0, i64 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %ptr.0, i64 1
   store i8 %conv, i8* %ptr.0, align 1, !tbaa !0
   %dec = add i64 %len.addr.0, -1
   br label %while.cond
@@ -67,7 +67,7 @@ if.end9:                                          ; preds = %land.rhs
 while.body18:                                     ; preds = %if.end9, %while.body18
   %wideptr.038 = phi i64* [ %incdec.ptr19, %while.body18 ], [ %9, %if.end9 ]
   %len.addr.137 = phi i64 [ %sub, %while.body18 ], [ %len.addr.0, %if.end9 ]
-  %incdec.ptr19 = getelementptr inbounds i64* %wideptr.038, i64 1
+  %incdec.ptr19 = getelementptr inbounds i64, i64* %wideptr.038, i64 1
   store i64 %ins, i64* %wideptr.038, align 8, !tbaa !2
   %sub = add i64 %len.addr.137, -8
   %cmp16 = icmp ugt i64 %sub, 7
@@ -87,7 +87,7 @@ while.body29:                                     ; preds = %while.body29, %whil
   %len.addr.235 = phi i64 [ %len.addr.1.lcssa49, %while.body29.lr.ph ], [ %dec26, %while.body29 ]
   %ptr.134 = phi i8* [ %10, %while.body29.lr.ph ], [ %incdec.ptr31, %while.body29 ]
   %dec26 = add i64 %len.addr.235, -1
-  %incdec.ptr31 = getelementptr inbounds i8* %ptr.134, i64 1
+  %incdec.ptr31 = getelementptr inbounds i8, i8* %ptr.134, i64 1
   store i8 %conv, i8* %ptr.134, align 1, !tbaa !0
   %cmp27 = icmp eq i64 %dec26, 0
   br i1 %cmp27, label %done, label %while.body29
diff --git a/test/Transforms/LoopStrengthReduce/AArch64/req-regs.ll b/test/Transforms/LoopStrengthReduce/AArch64/req-regs.ll
index 217896e55c66..c877ace7abf3 100644
--- a/test/Transforms/LoopStrengthReduce/AArch64/req-regs.ll
+++ b/test/Transforms/LoopStrengthReduce/AArch64/req-regs.ll
@@ -11,7 +11,7 @@ target triple = "arm64-apple-ios"
 define void @do_integer_add(i64 %iterations, i8* nocapture readonly %cookie) {
 entry:
   %N = bitcast i8* %cookie to i32*
-  %0 = load i32* %N, align 4
+  %0 = load i32, i32* %N, align 4
   %add = add nsw i32 %0, 57
   %cmp56 = icmp eq i64 %iterations, 0
   br i1 %cmp56, label %while.end, label %for.cond.preheader.preheader
diff --git a/test/Transforms/LoopStrengthReduce/ARM/2012-06-15-lsr-noaddrmode.ll b/test/Transforms/LoopStrengthReduce/ARM/2012-06-15-lsr-noaddrmode.ll
index 1d56ddea2446..56ff69c977ee 100644
--- a/test/Transforms/LoopStrengthReduce/ARM/2012-06-15-lsr-noaddrmode.ll
+++ b/test/Transforms/LoopStrengthReduce/ARM/2012-06-15-lsr-noaddrmode.ll
@@ -50,7 +50,7 @@ declare %s* @getstruct() nounwind
 ; CHECK: ldr{{.*}}lsl #2
 define i32 @main() nounwind ssp {
 entry:
-  %v0 = load i32* @ncol, align 4
+  %v0 = load i32, i32* @ncol, align 4
   %v1 = tail call i32* @getptr() nounwind
   %cmp10.i = icmp eq i32 %v0, 0
   br label %while.cond.outer
@@ -64,12 +64,12 @@ while.cond:
   br label %while.body
 
 while.body:
-  %v3 = load i32* @ncol, align 4
+  %v3 = load i32, i32* @ncol, align 4
   br label %end_of_chain
 
 end_of_chain:
-  %state.i = getelementptr inbounds %s* %call18, i32 0, i32 0
-  %v4 = load i32** %state.i, align 4
+  %state.i = getelementptr inbounds %s, %s* %call18, i32 0, i32 0
+  %v4 = load i32*, i32** %state.i, align 4
   br label %while.cond.i.i
 
 while.cond.i.i:
@@ -79,10 +79,10 @@ while.cond.i.i:
   br i1 %tobool.i.i, label %where.exit, label %land.rhs.i.i
 
 land.rhs.i.i:
-  %arrayidx.i.i = getelementptr inbounds i32* %v4, i32 %dec.i.i
-  %v5 = load i32* %arrayidx.i.i, align 4
-  %arrayidx1.i.i = getelementptr inbounds i32* %v1, i32 %dec.i.i
-  %v6 = load i32* %arrayidx1.i.i, align 4
+  %arrayidx.i.i = getelementptr inbounds i32, i32* %v4, i32 %dec.i.i
+  %v5 = load i32, i32* %arrayidx.i.i, align 4
+  %arrayidx1.i.i = getelementptr inbounds i32, i32* %v1, i32 %dec.i.i
+  %v6 = load i32, i32* %arrayidx1.i.i, align 4
   %cmp.i.i = icmp eq i32 %v5, %v6
   br i1 %cmp.i.i, label %while.cond.i.i, label %equal_data.exit.i
 
diff --git a/test/Transforms/LoopStrengthReduce/ARM/ivchain-ARM.ll b/test/Transforms/LoopStrengthReduce/ARM/ivchain-ARM.ll
index f4edf092641f..2ad6c2ea52da 100644
--- a/test/Transforms/LoopStrengthReduce/ARM/ivchain-ARM.ll
+++ b/test/Transforms/LoopStrengthReduce/ARM/ivchain-ARM.ll
@@ -18,18 +18,18 @@ entry:
 loop:
   %iv = phi i32* [ %a, %entry ], [ %iv4, %loop ]
   %s = phi i32 [ 0, %entry ], [ %s4, %loop ]
-  %v = load i32* %iv
-  %iv1 = getelementptr inbounds i32* %iv, i32 %x
-  %v1 = load i32* %iv1
-  %iv2 = getelementptr inbounds i32* %iv1, i32 %x
-  %v2 = load i32* %iv2
-  %iv3 = getelementptr inbounds i32* %iv2, i32 %x
-  %v3 = load i32* %iv3
+  %v = load i32, i32* %iv
+  %iv1 = getelementptr inbounds i32, i32* %iv, i32 %x
+  %v1 = load i32, i32* %iv1
+  %iv2 = getelementptr inbounds i32, i32* %iv1, i32 %x
+  %v2 = load i32, i32* %iv2
+  %iv3 = getelementptr inbounds i32, i32* %iv2, i32 %x
+  %v3 = load i32, i32* %iv3
   %s1 = add i32 %s, %v
   %s2 = add i32 %s1, %v1
   %s3 = add i32 %s2, %v2
   %s4 = add i32 %s3, %v3
-  %iv4 = getelementptr inbounds i32* %iv3, i32 %x
+  %iv4 = getelementptr inbounds i32, i32* %iv3, i32 %x
   %cmp = icmp eq i32* %iv4, %b
   br i1 %cmp, label %exit, label %loop
 exit:
@@ -52,18 +52,18 @@ entry:
 loop:
   %iv = phi i32* [ %a, %entry ], [ %iv4, %loop ]
   %s = phi i32 [ 0, %entry ], [ %s4, %loop ]
-  %v = load i32* %iv
-  %iv1 = getelementptr inbounds i32* %iv, i32 %x
-  %v1 = load i32* %iv1
-  %iv2 = getelementptr inbounds i32* %iv1, i32 %x
-  %v2 = load i32* %iv2
-  %iv3 = getelementptr inbounds i32* %iv2, i32 %x
-  %v3 = load i32* %iv3
+  %v = load i32, i32* %iv
+  %iv1 = getelementptr inbounds i32, i32* %iv, i32 %x
+  %v1 = load i32, i32* %iv1
+  %iv2 = getelementptr inbounds i32, i32* %iv1, i32 %x
+  %v2 = load i32, i32* %iv2
+  %iv3 = getelementptr inbounds i32, i32* %iv2, i32 %x
+  %v3 = load i32, i32* %iv3
   %s1 = add i32 %s, %v
   %s2 = add i32 %s1, %v1
   %s3 = add i32 %s2, %v2
   %s4 = add i32 %s3, %v3
-  %iv4 = getelementptr inbounds i32* %iv3, i32 %x
+  %iv4 = getelementptr inbounds i32, i32* %iv3, i32 %x
   store i32 %s4, i32* %iv
   %cmp = icmp eq i32* %iv4, %b
   br i1 %cmp, label %exit, label %loop
@@ -103,26 +103,26 @@ for.body:                                         ; preds = %for.body.lr.ph, %fo
   %i.010 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %res.addr.09 = phi i32* [ %res, %for.body.lr.ph ], [ %add.ptr7, %for.body ]
   %0 = bitcast i8* %main.addr.011 to i32*
-  %1 = load i32* %0, align 4
-  %add.ptr = getelementptr inbounds i8* %main.addr.011, i32 %main_stride
+  %1 = load i32, i32* %0, align 4
+  %add.ptr = getelementptr inbounds i8, i8* %main.addr.011, i32 %main_stride
   %2 = bitcast i8* %add.ptr to i32*
-  %3 = load i32* %2, align 4
-  %add.ptr1 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr.sum
+  %3 = load i32, i32* %2, align 4
+  %add.ptr1 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr.sum
   %4 = bitcast i8* %add.ptr1 to i32*
-  %5 = load i32* %4, align 4
-  %add.ptr2 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr1.sum
+  %5 = load i32, i32* %4, align 4
+  %add.ptr2 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr1.sum
   %6 = bitcast i8* %add.ptr2 to i32*
-  %7 = load i32* %6, align 4
-  %add.ptr3 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr4.sum
+  %7 = load i32, i32* %6, align 4
+  %add.ptr3 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr4.sum
   %8 = bitcast i8* %add.ptr3 to i32*
-  %9 = load i32* %8, align 4
+  %9 = load i32, i32* %8, align 4
   %add = add i32 %3, %1
   %add4 = add i32 %add, %5
   %add5 = add i32 %add4, %7
   %add6 = add i32 %add5, %9
   store i32 %add6, i32* %res.addr.09, align 4
-  %add.ptr6 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr3.sum
-  %add.ptr7 = getelementptr inbounds i32* %res.addr.09, i32 %y
+  %add.ptr6 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr3.sum
+  %add.ptr7 = getelementptr inbounds i32, i32* %res.addr.09, i32 %y
   %inc = add i32 %i.010, 1
   %cmp = icmp eq i32 %inc, %z
   br i1 %cmp, label %for.end, label %for.body
@@ -146,48 +146,48 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %i.07 = phi i32 [ 0, %entry ], [ %inc.3, %for.body ]
-  %arrayidx = getelementptr inbounds i8* %a, i32 %i.07
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %a, i32 %i.07
+  %0 = load i8, i8* %arrayidx, align 1
   %conv5 = zext i8 %0 to i32
-  %arrayidx1 = getelementptr inbounds i8* %b, i32 %i.07
-  %1 = load i8* %arrayidx1, align 1
+  %arrayidx1 = getelementptr inbounds i8, i8* %b, i32 %i.07
+  %1 = load i8, i8* %arrayidx1, align 1
   %conv26 = zext i8 %1 to i32
   %add = add nsw i32 %conv26, %conv5
   %conv3 = trunc i32 %add to i8
-  %arrayidx4 = getelementptr inbounds i8* %c, i32 %i.07
+  %arrayidx4 = getelementptr inbounds i8, i8* %c, i32 %i.07
   store i8 %conv3, i8* %arrayidx4, align 1
   %inc1 = or i32 %i.07, 1
-  %arrayidx.1 = getelementptr inbounds i8* %a, i32 %inc1
-  %2 = load i8* %arrayidx.1, align 1
+  %arrayidx.1 = getelementptr inbounds i8, i8* %a, i32 %inc1
+  %2 = load i8, i8* %arrayidx.1, align 1
   %conv5.1 = zext i8 %2 to i32
-  %arrayidx1.1 = getelementptr inbounds i8* %b, i32 %inc1
-  %3 = load i8* %arrayidx1.1, align 1
+  %arrayidx1.1 = getelementptr inbounds i8, i8* %b, i32 %inc1
+  %3 = load i8, i8* %arrayidx1.1, align 1
   %conv26.1 = zext i8 %3 to i32
   %add.1 = add nsw i32 %conv26.1, %conv5.1
   %conv3.1 = trunc i32 %add.1 to i8
-  %arrayidx4.1 = getelementptr inbounds i8* %c, i32 %inc1
+  %arrayidx4.1 = getelementptr inbounds i8, i8* %c, i32 %inc1
   store i8 %conv3.1, i8* %arrayidx4.1, align 1
   %inc.12 = or i32 %i.07, 2
-  %arrayidx.2 = getelementptr inbounds i8* %a, i32 %inc.12
-  %4 = load i8* %arrayidx.2, align 1
+  %arrayidx.2 = getelementptr inbounds i8, i8* %a, i32 %inc.12
+  %4 = load i8, i8* %arrayidx.2, align 1
   %conv5.2 = zext i8 %4 to i32
-  %arrayidx1.2 = getelementptr inbounds i8* %b, i32 %inc.12
-  %5 = load i8* %arrayidx1.2, align 1
+  %arrayidx1.2 = getelementptr inbounds i8, i8* %b, i32 %inc.12
+  %5 = load i8, i8* %arrayidx1.2, align 1
   %conv26.2 = zext i8 %5 to i32
   %add.2 = add nsw i32 %conv26.2, %conv5.2
   %conv3.2 = trunc i32 %add.2 to i8
-  %arrayidx4.2 = getelementptr inbounds i8* %c, i32 %inc.12
+  %arrayidx4.2 = getelementptr inbounds i8, i8* %c, i32 %inc.12
   store i8 %conv3.2, i8* %arrayidx4.2, align 1
   %inc.23 = or i32 %i.07, 3
-  %arrayidx.3 = getelementptr inbounds i8* %a, i32 %inc.23
-  %6 = load i8* %arrayidx.3, align 1
+  %arrayidx.3 = getelementptr inbounds i8, i8* %a, i32 %inc.23
+  %6 = load i8, i8* %arrayidx.3, align 1
   %conv5.3 = zext i8 %6 to i32
-  %arrayidx1.3 = getelementptr inbounds i8* %b, i32 %inc.23
-  %7 = load i8* %arrayidx1.3, align 1
+  %arrayidx1.3 = getelementptr inbounds i8, i8* %b, i32 %inc.23
+  %7 = load i8, i8* %arrayidx1.3, align 1
   %conv26.3 = zext i8 %7 to i32
   %add.3 = add nsw i32 %conv26.3, %conv5.3
   %conv3.3 = trunc i32 %add.3 to i8
-  %arrayidx4.3 = getelementptr inbounds i8* %c, i32 %inc.23
+  %arrayidx4.3 = getelementptr inbounds i8, i8* %c, i32 %inc.23
   store i8 %conv3.3, i8* %arrayidx4.3, align 1
   %inc.3 = add nsw i32 %i.07, 4
   %exitcond.3 = icmp eq i32 %inc.3, 400
@@ -201,7 +201,7 @@ for.end:                                          ; preds = %for.body
 ;
 ; Currently we have three extra add.w's that keep the store address
 ; live past the next increment because ISEL is unfortunately undoing
-; the store chain. ISEL also fails to convert the stores to
+; the store chain. ISEL also fails to convert all but one of the stores to
 ; post-increment addressing. However, the loads should use
 ; post-increment addressing, no add's or add.w's beyond the three
 ; mentioned. Most importantly, there should be no spills or reloads!
@@ -210,7 +210,7 @@ for.end:                                          ; preds = %for.body
 ; A9: %.lr.ph
 ; A9-NOT: lsl.w
 ; A9-NOT: {{ldr|str|adds|add r}}
-; A9: add.w r
+; A9: vst1.8 {{.*}} [r{{[0-9]+}}]!
 ; A9-NOT: {{ldr|str|adds|add r}}
 ; A9: add.w r
 ; A9-NOT: {{ldr|str|adds|add r}}
@@ -240,31 +240,31 @@ define hidden void @testNeon(i8* %ref_data, i32 %ref_stride, i32 %limit, <16 x i
   %result.03 = phi <16 x i8> [ zeroinitializer, %.lr.ph ], [ %41, %11 ]
   %.012 = phi <16 x i8>* [ %data, %.lr.ph ], [ %43, %11 ]
   %12 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %.05, i32 1) nounwind
-  %13 = getelementptr inbounds i8* %.05, i32 %ref_stride
+  %13 = getelementptr inbounds i8, i8* %.05, i32 %ref_stride
   %14 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %13, i32 1) nounwind
   %15 = shufflevector <1 x i64> %12, <1 x i64> %14, <2 x i32> <i32 0, i32 1>
   %16 = bitcast <2 x i64> %15 to <16 x i8>
-  %17 = getelementptr inbounds <16 x i8>* %.012, i32 1
+  %17 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 1
   store <16 x i8> %16, <16 x i8>* %.012, align 4
-  %18 = getelementptr inbounds i8* %.05, i32 %2
+  %18 = getelementptr inbounds i8, i8* %.05, i32 %2
   %19 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %18, i32 1) nounwind
-  %20 = getelementptr inbounds i8* %.05, i32 %3
+  %20 = getelementptr inbounds i8, i8* %.05, i32 %3
   %21 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %20, i32 1) nounwind
   %22 = shufflevector <1 x i64> %19, <1 x i64> %21, <2 x i32> <i32 0, i32 1>
   %23 = bitcast <2 x i64> %22 to <16 x i8>
-  %24 = getelementptr inbounds <16 x i8>* %.012, i32 2
+  %24 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 2
   store <16 x i8> %23, <16 x i8>* %17, align 4
-  %25 = getelementptr inbounds i8* %.05, i32 %4
+  %25 = getelementptr inbounds i8, i8* %.05, i32 %4
   %26 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %25, i32 1) nounwind
-  %27 = getelementptr inbounds i8* %.05, i32 %5
+  %27 = getelementptr inbounds i8, i8* %.05, i32 %5
   %28 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %27, i32 1) nounwind
   %29 = shufflevector <1 x i64> %26, <1 x i64> %28, <2 x i32> <i32 0, i32 1>
   %30 = bitcast <2 x i64> %29 to <16 x i8>
-  %31 = getelementptr inbounds <16 x i8>* %.012, i32 3
+  %31 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 3
   store <16 x i8> %30, <16 x i8>* %24, align 4
-  %32 = getelementptr inbounds i8* %.05, i32 %6
+  %32 = getelementptr inbounds i8, i8* %.05, i32 %6
   %33 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %32, i32 1) nounwind
-  %34 = getelementptr inbounds i8* %.05, i32 %7
+  %34 = getelementptr inbounds i8, i8* %.05, i32 %7
   %35 = tail call <1 x i64> @llvm.arm.neon.vld1.v1i64(i8* %34, i32 1) nounwind
   %36 = shufflevector <1 x i64> %33, <1 x i64> %35, <2 x i32> <i32 0, i32 1>
   %37 = bitcast <2 x i64> %36 to <16 x i8>
@@ -273,14 +273,14 @@ define hidden void @testNeon(i8* %ref_data, i32 %ref_stride, i32 %limit, <16 x i
   %39 = add <16 x i8> %38, %30
   %40 = add <16 x i8> %39, %37
   %41 = add <16 x i8> %result.03, %40
-  %42 = getelementptr i8* %.05, i32 %9
-  %43 = getelementptr inbounds <16 x i8>* %.012, i32 -64
+  %42 = getelementptr i8, i8* %.05, i32 %9
+  %43 = getelementptr inbounds <16 x i8>, <16 x i8>* %.012, i32 -64
   %44 = add nsw i32 %counter.04, 1
   %exitcond = icmp eq i32 %44, %limit
   br i1 %exitcond, label %._crit_edge, label %11
 
 ._crit_edge:                                      ; preds = %11
-  %scevgep = getelementptr <16 x i8>* %data, i32 %10
+  %scevgep = getelementptr <16 x i8>, <16 x i8>* %data, i32 %10
   br label %45
 
 ; <label>:45                                      ; preds = %._crit_edge, %0
@@ -327,20 +327,20 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %i.0110 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %src.addr = phi i8* [ %src, %entry ], [ %add.ptr45, %for.body ]
-  %add.ptr = getelementptr inbounds i8* %src.addr, i32 %idx.neg
+  %add.ptr = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg
   %vld1 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr, i32 1)
-  %add.ptr3 = getelementptr inbounds i8* %src.addr, i32 %idx.neg2
+  %add.ptr3 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg2
   %vld2 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr3, i32 1)
-  %add.ptr7 = getelementptr inbounds i8* %src.addr, i32 %idx.neg6
+  %add.ptr7 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg6
   %vld3 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr7, i32 1)
-  %add.ptr11 = getelementptr inbounds i8* %src.addr, i32 %idx.neg10
+  %add.ptr11 = getelementptr inbounds i8, i8* %src.addr, i32 %idx.neg10
   %vld4 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr11, i32 1)
   %vld5 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %src.addr, i32 1)
-  %add.ptr17 = getelementptr inbounds i8* %src.addr, i32 %stride
+  %add.ptr17 = getelementptr inbounds i8, i8* %src.addr, i32 %stride
   %vld6 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr17, i32 1)
-  %add.ptr20 = getelementptr inbounds i8* %src.addr, i32 %mul5
+  %add.ptr20 = getelementptr inbounds i8, i8* %src.addr, i32 %mul5
   %vld7 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr20, i32 1)
-  %add.ptr23 = getelementptr inbounds i8* %src.addr, i32 %mul1
+  %add.ptr23 = getelementptr inbounds i8, i8* %src.addr, i32 %mul1
   %vld8 = tail call <8 x i8> @llvm.arm.neon.vld1.v8i8(i8* %add.ptr23, i32 1)
   %vadd1 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld1, <8 x i8> %vld2) nounwind
   %vadd2 = tail call <8 x i8> @llvm.arm.neon.vhaddu.v8i8(<8 x i8> %vld2, <8 x i8> %vld3) nounwind
@@ -355,7 +355,7 @@ for.body:                                         ; preds = %for.body, %entry
   tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr17, <8 x i8> %vadd5, i32 1)
   tail call void @llvm.arm.neon.vst1.v8i8(i8* %add.ptr20, <8 x i8> %vadd6, i32 1)
   %inc = add nsw i32 %i.0110, 1
-  %add.ptr45 = getelementptr inbounds i8* %src.addr, i32 8
+  %add.ptr45 = getelementptr inbounds i8, i8* %src.addr, i32 8
   %exitcond = icmp eq i32 %inc, 4
   br i1 %exitcond, label %for.end, label %for.body
 
diff --git a/test/Transforms/LoopStrengthReduce/X86/2008-08-14-ShadowIV.ll b/test/Transforms/LoopStrengthReduce/X86/2008-08-14-ShadowIV.ll
index 9a7f4865c591..f862c2635869 100644
--- a/test/Transforms/LoopStrengthReduce/X86/2008-08-14-ShadowIV.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/2008-08-14-ShadowIV.ll
@@ -1,5 +1,9 @@
 ; RUN: opt < %s -loop-reduce -S -mtriple=x86_64-unknown-unknown | grep "phi double" | count 1
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 define void @foobar(i32 %n) nounwind {
 entry:
 	icmp eq i32 %n, 0		; <i1>:0 [#uses=2]
@@ -50,7 +54,7 @@ return:		; preds = %bb, %entry
 ; Unable to eliminate cast due to potentional overflow.
 define void @foobar3() nounwind {
 entry:
-	tail call i32 (...)* @nn( ) nounwind		; <i32>:0 [#uses=1]
+	tail call i32 (...) @nn( ) nounwind		; <i32>:0 [#uses=1]
 	icmp eq i32 %0, 0		; <i1>:1 [#uses=1]
 	br i1 %1, label %return, label %bb
 
@@ -60,7 +64,7 @@ bb:		; preds = %bb, %entry
 	uitofp i32 %i.03 to double		; <double>:2 [#uses=1]
 	tail call void @foo( double %2 ) nounwind
 	add i32 %i.03, 1		; <i32>:3 [#uses=2]
-	tail call i32 (...)* @nn( ) nounwind		; <i32>:4 [#uses=1]
+	tail call i32 (...) @nn( ) nounwind		; <i32>:4 [#uses=1]
 	icmp ugt i32 %4, %3		; <i1>:5 [#uses=1]
 	br i1 %5, label %bb, label %return
 
diff --git a/test/Transforms/LoopStrengthReduce/X86/2011-07-20-DoubleIV.ll b/test/Transforms/LoopStrengthReduce/X86/2011-07-20-DoubleIV.ll
index 2fe62e39fc93..0fc928ca9b28 100644
--- a/test/Transforms/LoopStrengthReduce/X86/2011-07-20-DoubleIV.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/2011-07-20-DoubleIV.ll
@@ -4,6 +4,10 @@
 ; nonzero initial value.
 ; rdar://9786536
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
+
 ; First, make sure LSR doesn't crash on an empty IVUsers list.
 ; CHECK-LABEL: @dummyIV(
 ; CHECK-NOT: phi
diff --git a/test/Transforms/LoopStrengthReduce/X86/2011-12-04-loserreg.ll b/test/Transforms/LoopStrengthReduce/X86/2011-12-04-loserreg.ll
index 001a1d695c99..862fff29cefb 100644
--- a/test/Transforms/LoopStrengthReduce/X86/2011-12-04-loserreg.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/2011-12-04-loserreg.ll
@@ -29,34 +29,34 @@ for.body:                                         ; preds = %entry, %for.body
   %indvars.iv39 = phi i64 [ %indvars.iv.next40, %for.body ], [ 0, %entry ]
   %dp.036 = phi i32* [ %add.ptr, %for.body ], [ %destrow, %entry ]
   %p.035 = phi float* [ %incdec.ptr4, %for.body ], [ %srcrow, %entry ]
-  %incdec.ptr = getelementptr inbounds float* %p.035, i64 1
-  %0 = load float* %incdec.ptr, align 4
-  %incdec.ptr2 = getelementptr inbounds float* %p.035, i64 2
-  %1 = load float* %incdec.ptr2, align 4
-  %incdec.ptr3 = getelementptr inbounds float* %p.035, i64 3
-  %2 = load float* %incdec.ptr3, align 4
-  %incdec.ptr4 = getelementptr inbounds float* %p.035, i64 4
-  %3 = load float* %incdec.ptr4, align 4
-  %4 = load i32* %dp.036, align 4
+  %incdec.ptr = getelementptr inbounds float, float* %p.035, i64 1
+  %0 = load float, float* %incdec.ptr, align 4
+  %incdec.ptr2 = getelementptr inbounds float, float* %p.035, i64 2
+  %1 = load float, float* %incdec.ptr2, align 4
+  %incdec.ptr3 = getelementptr inbounds float, float* %p.035, i64 3
+  %2 = load float, float* %incdec.ptr3, align 4
+  %incdec.ptr4 = getelementptr inbounds float, float* %p.035, i64 4
+  %3 = load float, float* %incdec.ptr4, align 4
+  %4 = load i32, i32* %dp.036, align 4
   %conv5 = fptoui float %0 to i32
   %or = or i32 %4, %conv5
-  %arrayidx6 = getelementptr inbounds i32* %dp.036, i64 1
-  %5 = load i32* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %dp.036, i64 1
+  %5 = load i32, i32* %arrayidx6, align 4
   %conv7 = fptoui float %1 to i32
   %or8 = or i32 %5, %conv7
-  %arrayidx9 = getelementptr inbounds i32* %dp.036, i64 2
-  %6 = load i32* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds i32, i32* %dp.036, i64 2
+  %6 = load i32, i32* %arrayidx9, align 4
   %conv10 = fptoui float %2 to i32
   %or11 = or i32 %6, %conv10
-  %arrayidx12 = getelementptr inbounds i32* %dp.036, i64 3
-  %7 = load i32* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds i32, i32* %dp.036, i64 3
+  %7 = load i32, i32* %arrayidx12, align 4
   %conv13 = fptoui float %3 to i32
   %or14 = or i32 %7, %conv13
   store i32 %or, i32* %dp.036, align 4
   store i32 %or8, i32* %arrayidx6, align 4
   store i32 %or11, i32* %arrayidx9, align 4
   store i32 %or14, i32* %arrayidx12, align 4
-  %add.ptr = getelementptr inbounds i32* %dp.036, i64 4
+  %add.ptr = getelementptr inbounds i32, i32* %dp.036, i64 4
   %indvars.iv.next40 = add i64 %indvars.iv39, 4
   %dummycnt = add i64 %dummyiv, 1
   %cmp = icmp ult i64 %indvars.iv.next40, %count
@@ -76,14 +76,14 @@ for.body23:                                       ; preds = %for.body23, %for.bo
   %indvars.iv = phi i64 [ 0, %for.body23.lr.ph ], [ %indvars.iv.next, %for.body23 ]
   %dp.132 = phi i32* [ %add.ptr, %for.body23.lr.ph ], [ %incdec.ptr28, %for.body23 ]
   %p.131 = phi float* [ %incdec.ptr4, %for.body23.lr.ph ], [ %incdec.ptr24, %for.body23 ]
-  %incdec.ptr24 = getelementptr inbounds float* %p.131, i64 1
-  %9 = load float* %incdec.ptr24, align 4
-  %10 = load i32* %dp.132, align 4
+  %incdec.ptr24 = getelementptr inbounds float, float* %p.131, i64 1
+  %9 = load float, float* %incdec.ptr24, align 4
+  %10 = load i32, i32* %dp.132, align 4
   %conv25 = fptoui float %9 to i32
   %or26 = or i32 %10, %conv25
   store i32 %or26, i32* %dp.132, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
-  %incdec.ptr28 = getelementptr inbounds i32* %dp.132, i64 1
+  %incdec.ptr28 = getelementptr inbounds i32, i32* %dp.132, i64 1
   %exitcond = icmp eq i64 %indvars.iv.next, %8
   br i1 %exitcond, label %for.end29, label %for.body23
 
diff --git a/test/Transforms/LoopStrengthReduce/X86/2012-01-13-phielim.ll b/test/Transforms/LoopStrengthReduce/X86/2012-01-13-phielim.ll
index c3b8b8910d5c..184e300c7eb8 100644
--- a/test/Transforms/LoopStrengthReduce/X86/2012-01-13-phielim.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/2012-01-13-phielim.ll
@@ -27,14 +27,14 @@ while.body.i:                                     ; preds = %cond.true29.i, %whi
 for.body.i:                                       ; preds = %for.body.i, %while.body.i
   %indvars.iv.i = phi i64 [ 0, %while.body.i ], [ %indvars.iv.next.i, %for.body.i ]
   %add.ptr.sum = add i64 %add.ptr.sum.i, %indvars.iv.i
-  %arrayidx22.i = getelementptr inbounds i8* %base, i64 %add.ptr.sum
-  %0 = load i8* %arrayidx22.i, align 1
+  %arrayidx22.i = getelementptr inbounds i8, i8* %base, i64 %add.ptr.sum
+  %0 = load i8, i8* %arrayidx22.i, align 1
   %indvars.iv.next.i = add i64 %indvars.iv.i, 1
   %cmp = call i1 @check() nounwind
   br i1 %cmp, label %for.end.i, label %for.body.i
 
 for.end.i:                                        ; preds = %for.body.i
-  %add.ptr.i144 = getelementptr inbounds i8* %base, i64 %add.ptr.sum.i
+  %add.ptr.i144 = getelementptr inbounds i8, i8* %base, i64 %add.ptr.sum.i
   %cmp2 = tail call i1 @foo(i8* %add.ptr.i144, i8* %add.ptr.i144, i8* undef) nounwind
   br i1 %cmp2, label %cond.true29.i, label %cond.false35.i
 
@@ -68,19 +68,19 @@ entry:
 
 for.cond468:                                      ; preds = %if.then477, %entry
   %indvars.iv1163 = phi i64 [ %indvars.iv.next1164, %if.then477 ], [ 1, %entry ]
-  %k.0.in = phi i32* [ %last, %if.then477 ], [ getelementptr inbounds ([5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 0, i32 2), %entry ]
-  %k.0 = load i32* %k.0.in, align 4
+  %k.0.in = phi i32* [ %last, %if.then477 ], [ getelementptr inbounds ([5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771], [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 0, i32 2), %entry ]
+  %k.0 = load i32, i32* %k.0.in, align 4
   %0 = trunc i64 %indvars.iv1163 to i32
   %cmp469 = icmp slt i32 %0, %n
   br i1 %cmp469, label %for.body471, label %for.inc498
 
 for.body471:                                      ; preds = %for.cond468
-  %first = getelementptr inbounds [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 %indvars.iv1163, i32 1
-  %1 = load i32* %first, align 4
+  %first = getelementptr inbounds [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771], [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 %indvars.iv1163, i32 1
+  %1 = load i32, i32* %first, align 4
   br i1 undef, label %if.then477, label %for.inc498
 
 if.then477:                                       ; preds = %for.body471
-  %last = getelementptr inbounds [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 %indvars.iv1163, i32 2
+  %last = getelementptr inbounds [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771], [5000 x %struct.anon.7.91.199.307.415.475.559.643.751.835.943.1003.1111.1219.1351.1375.1399.1435.1471.1483.1519.1531.1651.1771]* @tags, i64 0, i64 %indvars.iv1163, i32 2
   %indvars.iv.next1164 = add i64 %indvars.iv1163, 1
   br label %for.cond468
 
@@ -118,14 +118,14 @@ for.body3.us.i:                                   ; preds = %meshBB, %for.body3.
   %0 = add nsw i64 %indvars.iv.i.SV.phi, %indvars.iv8.i.SV.phi26
   %1 = trunc i64 %0 to i32
   %mul.i.us.i = mul nsw i32 0, %1
-  %arrayidx5.us.i = getelementptr inbounds double* %u, i64 %indvars.iv.i.SV.phi
-  %2 = load double* %arrayidx5.us.i, align 8
+  %arrayidx5.us.i = getelementptr inbounds double, double* %u, i64 %indvars.iv.i.SV.phi
+  %2 = load double, double* %arrayidx5.us.i, align 8
   %indvars.iv.next.i = add i64 %indvars.iv.i.SV.phi, 1
   br i1 undef, label %for.inc8.us.i, label %meshBB
 
 for.body3.lr.ph.us.i:                             ; preds = %meshBB1, %meshBB
   %indvars.iv8.i.SV.phi26 = phi i64 [ undef, %meshBB1 ], [ %indvars.iv8.i.SV.phi24, %meshBB ]
-  %arrayidx.us.i = getelementptr inbounds double* undef, i64 %indvars.iv8.i.SV.phi26
+  %arrayidx.us.i = getelementptr inbounds double, double* undef, i64 %indvars.iv8.i.SV.phi26
   %3 = add i64 %indvars.iv8.i.SV.phi26, 1
   br label %for.body3.us.i
 
diff --git a/test/Transforms/LoopStrengthReduce/X86/ivchain-X86.ll b/test/Transforms/LoopStrengthReduce/X86/ivchain-X86.ll
index 937791dca413..c1099b23dcfc 100644
--- a/test/Transforms/LoopStrengthReduce/X86/ivchain-X86.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/ivchain-X86.ll
@@ -29,18 +29,18 @@ entry:
 loop:
   %iv = phi i32* [ %a, %entry ], [ %iv4, %loop ]
   %s = phi i32 [ 0, %entry ], [ %s4, %loop ]
-  %v = load i32* %iv
-  %iv1 = getelementptr inbounds i32* %iv, i32 %x
-  %v1 = load i32* %iv1
-  %iv2 = getelementptr inbounds i32* %iv1, i32 %x
-  %v2 = load i32* %iv2
-  %iv3 = getelementptr inbounds i32* %iv2, i32 %x
-  %v3 = load i32* %iv3
+  %v = load i32, i32* %iv
+  %iv1 = getelementptr inbounds i32, i32* %iv, i32 %x
+  %v1 = load i32, i32* %iv1
+  %iv2 = getelementptr inbounds i32, i32* %iv1, i32 %x
+  %v2 = load i32, i32* %iv2
+  %iv3 = getelementptr inbounds i32, i32* %iv2, i32 %x
+  %v3 = load i32, i32* %iv3
   %s1 = add i32 %s, %v
   %s2 = add i32 %s1, %v1
   %s3 = add i32 %s2, %v2
   %s4 = add i32 %s3, %v3
-  %iv4 = getelementptr inbounds i32* %iv3, i32 %x
+  %iv4 = getelementptr inbounds i32, i32* %iv3, i32 %x
   %cmp = icmp eq i32* %iv4, %b
   br i1 %cmp, label %exit, label %loop
 exit:
@@ -59,7 +59,9 @@ exit:
 ;
 ; X32: @user
 ; expensive address computation in the preheader
-; X32: imul
+; X32: shll $4
+; X32: lea
+; X32: lea
 ; X32: %loop
 ; complex address modes
 ; X32: (%{{[^)]+}},%{{[^)]+}},
@@ -69,18 +71,18 @@ entry:
 loop:
   %iv = phi i32* [ %a, %entry ], [ %iv4, %loop ]
   %s = phi i32 [ 0, %entry ], [ %s4, %loop ]
-  %v = load i32* %iv
-  %iv1 = getelementptr inbounds i32* %iv, i32 %x
-  %v1 = load i32* %iv1
-  %iv2 = getelementptr inbounds i32* %iv1, i32 %x
-  %v2 = load i32* %iv2
-  %iv3 = getelementptr inbounds i32* %iv2, i32 %x
-  %v3 = load i32* %iv3
+  %v = load i32, i32* %iv
+  %iv1 = getelementptr inbounds i32, i32* %iv, i32 %x
+  %v1 = load i32, i32* %iv1
+  %iv2 = getelementptr inbounds i32, i32* %iv1, i32 %x
+  %v2 = load i32, i32* %iv2
+  %iv3 = getelementptr inbounds i32, i32* %iv2, i32 %x
+  %v3 = load i32, i32* %iv3
   %s1 = add i32 %s, %v
   %s2 = add i32 %s1, %v1
   %s3 = add i32 %s2, %v2
   %s4 = add i32 %s3, %v3
-  %iv4 = getelementptr inbounds i32* %iv3, i32 %x
+  %iv4 = getelementptr inbounds i32, i32* %iv3, i32 %x
   store i32 %s4, i32* %iv
   %cmp = icmp eq i32* %iv4, %b
   br i1 %cmp, label %exit, label %loop
@@ -124,26 +126,26 @@ for.body:                                         ; preds = %for.body.lr.ph, %fo
   %i.010 = phi i32 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %res.addr.09 = phi i32* [ %res, %for.body.lr.ph ], [ %add.ptr7, %for.body ]
   %0 = bitcast i8* %main.addr.011 to i32*
-  %1 = load i32* %0, align 4
-  %add.ptr = getelementptr inbounds i8* %main.addr.011, i32 %main_stride
+  %1 = load i32, i32* %0, align 4
+  %add.ptr = getelementptr inbounds i8, i8* %main.addr.011, i32 %main_stride
   %2 = bitcast i8* %add.ptr to i32*
-  %3 = load i32* %2, align 4
-  %add.ptr1 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr.sum
+  %3 = load i32, i32* %2, align 4
+  %add.ptr1 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr.sum
   %4 = bitcast i8* %add.ptr1 to i32*
-  %5 = load i32* %4, align 4
-  %add.ptr2 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr1.sum
+  %5 = load i32, i32* %4, align 4
+  %add.ptr2 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr1.sum
   %6 = bitcast i8* %add.ptr2 to i32*
-  %7 = load i32* %6, align 4
-  %add.ptr3 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr4.sum
+  %7 = load i32, i32* %6, align 4
+  %add.ptr3 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr4.sum
   %8 = bitcast i8* %add.ptr3 to i32*
-  %9 = load i32* %8, align 4
+  %9 = load i32, i32* %8, align 4
   %add = add i32 %3, %1
   %add4 = add i32 %add, %5
   %add5 = add i32 %add4, %7
   %add6 = add i32 %add5, %9
   store i32 %add6, i32* %res.addr.09, align 4
-  %add.ptr6 = getelementptr inbounds i8* %main.addr.011, i32 %add.ptr3.sum
-  %add.ptr7 = getelementptr inbounds i32* %res.addr.09, i32 %y
+  %add.ptr6 = getelementptr inbounds i8, i8* %main.addr.011, i32 %add.ptr3.sum
+  %add.ptr7 = getelementptr inbounds i32, i32* %res.addr.09, i32 %y
   %inc = add i32 %i.010, 1
   %cmp = icmp eq i32 %inc, %z
   br i1 %cmp, label %for.end, label %for.body
@@ -170,48 +172,48 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %i.07 = phi i32 [ 0, %entry ], [ %inc.3, %for.body ]
-  %arrayidx = getelementptr inbounds i8* %a, i32 %i.07
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %a, i32 %i.07
+  %0 = load i8, i8* %arrayidx, align 1
   %conv5 = zext i8 %0 to i32
-  %arrayidx1 = getelementptr inbounds i8* %b, i32 %i.07
-  %1 = load i8* %arrayidx1, align 1
+  %arrayidx1 = getelementptr inbounds i8, i8* %b, i32 %i.07
+  %1 = load i8, i8* %arrayidx1, align 1
   %conv26 = zext i8 %1 to i32
   %add = add nsw i32 %conv26, %conv5
   %conv3 = trunc i32 %add to i8
-  %arrayidx4 = getelementptr inbounds i8* %c, i32 %i.07
+  %arrayidx4 = getelementptr inbounds i8, i8* %c, i32 %i.07
   store i8 %conv3, i8* %arrayidx4, align 1
   %inc1 = or i32 %i.07, 1
-  %arrayidx.1 = getelementptr inbounds i8* %a, i32 %inc1
-  %2 = load i8* %arrayidx.1, align 1
+  %arrayidx.1 = getelementptr inbounds i8, i8* %a, i32 %inc1
+  %2 = load i8, i8* %arrayidx.1, align 1
   %conv5.1 = zext i8 %2 to i32
-  %arrayidx1.1 = getelementptr inbounds i8* %b, i32 %inc1
-  %3 = load i8* %arrayidx1.1, align 1
+  %arrayidx1.1 = getelementptr inbounds i8, i8* %b, i32 %inc1
+  %3 = load i8, i8* %arrayidx1.1, align 1
   %conv26.1 = zext i8 %3 to i32
   %add.1 = add nsw i32 %conv26.1, %conv5.1
   %conv3.1 = trunc i32 %add.1 to i8
-  %arrayidx4.1 = getelementptr inbounds i8* %c, i32 %inc1
+  %arrayidx4.1 = getelementptr inbounds i8, i8* %c, i32 %inc1
   store i8 %conv3.1, i8* %arrayidx4.1, align 1
   %inc.12 = or i32 %i.07, 2
-  %arrayidx.2 = getelementptr inbounds i8* %a, i32 %inc.12
-  %4 = load i8* %arrayidx.2, align 1
+  %arrayidx.2 = getelementptr inbounds i8, i8* %a, i32 %inc.12
+  %4 = load i8, i8* %arrayidx.2, align 1
   %conv5.2 = zext i8 %4 to i32
-  %arrayidx1.2 = getelementptr inbounds i8* %b, i32 %inc.12
-  %5 = load i8* %arrayidx1.2, align 1
+  %arrayidx1.2 = getelementptr inbounds i8, i8* %b, i32 %inc.12
+  %5 = load i8, i8* %arrayidx1.2, align 1
   %conv26.2 = zext i8 %5 to i32
   %add.2 = add nsw i32 %conv26.2, %conv5.2
   %conv3.2 = trunc i32 %add.2 to i8
-  %arrayidx4.2 = getelementptr inbounds i8* %c, i32 %inc.12
+  %arrayidx4.2 = getelementptr inbounds i8, i8* %c, i32 %inc.12
   store i8 %conv3.2, i8* %arrayidx4.2, align 1
   %inc.23 = or i32 %i.07, 3
-  %arrayidx.3 = getelementptr inbounds i8* %a, i32 %inc.23
-  %6 = load i8* %arrayidx.3, align 1
+  %arrayidx.3 = getelementptr inbounds i8, i8* %a, i32 %inc.23
+  %6 = load i8, i8* %arrayidx.3, align 1
   %conv5.3 = zext i8 %6 to i32
-  %arrayidx1.3 = getelementptr inbounds i8* %b, i32 %inc.23
-  %7 = load i8* %arrayidx1.3, align 1
+  %arrayidx1.3 = getelementptr inbounds i8, i8* %b, i32 %inc.23
+  %7 = load i8, i8* %arrayidx1.3, align 1
   %conv26.3 = zext i8 %7 to i32
   %add.3 = add nsw i32 %conv26.3, %conv5.3
   %conv3.3 = trunc i32 %add.3 to i8
-  %arrayidx4.3 = getelementptr inbounds i8* %c, i32 %inc.23
+  %arrayidx4.3 = getelementptr inbounds i8, i8* %c, i32 %inc.23
   store i8 %conv3.3, i8* %arrayidx4.3, align 1
   %inc.3 = add nsw i32 %i.07, 4
   %exitcond.3 = icmp eq i32 %inc.3, 400
@@ -252,15 +254,15 @@ for.body:
   %i = phi i32 [ %inc4, %for.body ], [ 0, %entry ]
   store i32 %i, i32* %p, align 4
   %inc1 = or i32 %i, 1
-  %add.ptr.i1 = getelementptr inbounds i32* %p, i32 1
+  %add.ptr.i1 = getelementptr inbounds i32, i32* %p, i32 1
   store i32 %inc1, i32* %add.ptr.i1, align 4
   %inc2 = add nsw i32 %i, 2
-  %add.ptr.i2 = getelementptr inbounds i32* %p, i32 2
+  %add.ptr.i2 = getelementptr inbounds i32, i32* %p, i32 2
   store i32 %inc2, i32* %add.ptr.i2, align 4
   %inc3 = add nsw i32 %i, 3
-  %add.ptr.i3 = getelementptr inbounds i32* %p, i32 3
+  %add.ptr.i3 = getelementptr inbounds i32, i32* %p, i32 3
   store i32 %inc3, i32* %add.ptr.i3, align 4
-  %p.next = getelementptr inbounds i32* %p, i32 4
+  %p.next = getelementptr inbounds i32, i32* %p, i32 4
   %inc4 = add nsw i32 %i, 4
   %cmp = icmp slt i32 %inc4, %n
   br i1 %cmp, label %for.body, label %exit
@@ -279,20 +281,20 @@ exit:
 ; X32: jne
 define void @testCmpZero(i8* %src, i8* %dst, i32 %srcidx, i32 %dstidx, i32 %len) nounwind ssp {
 entry:
-  %dest0 = getelementptr inbounds i8* %src, i32 %srcidx
-  %source0 = getelementptr inbounds i8* %dst, i32 %dstidx
+  %dest0 = getelementptr inbounds i8, i8* %src, i32 %srcidx
+  %source0 = getelementptr inbounds i8, i8* %dst, i32 %dstidx
   %add.ptr79.us.sum = add i32 %srcidx, %len
-  %lftr.limit = getelementptr i8* %src, i32 %add.ptr79.us.sum
+  %lftr.limit = getelementptr i8, i8* %src, i32 %add.ptr79.us.sum
   br label %for.body82.us
 
 for.body82.us:
   %dest = phi i8* [ %dest0, %entry ], [ %incdec.ptr91.us, %for.body82.us ]
   %source = phi i8* [ %source0, %entry ], [ %add.ptr83.us, %for.body82.us ]
   %0 = bitcast i8* %source to i32*
-  %1 = load i32* %0, align 4
+  %1 = load i32, i32* %0, align 4
   %trunc = trunc i32 %1 to i8
-  %add.ptr83.us = getelementptr inbounds i8* %source, i32 4
-  %incdec.ptr91.us = getelementptr inbounds i8* %dest, i32 1
+  %add.ptr83.us = getelementptr inbounds i8, i8* %source, i32 4
+  %incdec.ptr91.us = getelementptr inbounds i8, i8* %dest, i32 1
   store i8 %trunc, i8* %dest, align 1
   %exitcond = icmp eq i8* %incdec.ptr91.us, %lftr.limit
   br i1 %exitcond, label %return, label %for.body82.us
diff --git a/test/Transforms/LoopStrengthReduce/X86/ivchain-stress-X86.ll b/test/Transforms/LoopStrengthReduce/X86/ivchain-stress-X86.ll
index d8e0aa9dc805..7925bf01020e 100644
--- a/test/Transforms/LoopStrengthReduce/X86/ivchain-stress-X86.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/ivchain-stress-X86.ll
@@ -32,15 +32,15 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body.3
   %i.09 = phi i32 [ %add5.3, %for.body.3 ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i8* %a, i32 %i.09
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %a, i32 %i.09
+  %0 = load i8, i8* %arrayidx, align 1
   %conv6 = zext i8 %0 to i32
-  %arrayidx1 = getelementptr inbounds i8* %b, i32 %i.09
-  %1 = load i8* %arrayidx1, align 1
+  %arrayidx1 = getelementptr inbounds i8, i8* %b, i32 %i.09
+  %1 = load i8, i8* %arrayidx1, align 1
   %conv27 = zext i8 %1 to i32
   %add = add nsw i32 %conv27, %conv6
   %conv3 = trunc i32 %add to i8
-  %arrayidx4 = getelementptr inbounds i8* %c, i32 %i.09
+  %arrayidx4 = getelementptr inbounds i8, i8* %c, i32 %i.09
   store i8 %conv3, i8* %arrayidx4, align 1
   %add5 = add i32 %i.09, %s
   %cmp = icmp ult i32 %add5, %len
@@ -50,45 +50,45 @@ for.end:                                          ; preds = %for.body, %for.body
   ret void
 
 for.body.1:                                       ; preds = %for.body
-  %arrayidx.1 = getelementptr inbounds i8* %a, i32 %add5
-  %2 = load i8* %arrayidx.1, align 1
+  %arrayidx.1 = getelementptr inbounds i8, i8* %a, i32 %add5
+  %2 = load i8, i8* %arrayidx.1, align 1
   %conv6.1 = zext i8 %2 to i32
-  %arrayidx1.1 = getelementptr inbounds i8* %b, i32 %add5
-  %3 = load i8* %arrayidx1.1, align 1
+  %arrayidx1.1 = getelementptr inbounds i8, i8* %b, i32 %add5
+  %3 = load i8, i8* %arrayidx1.1, align 1
   %conv27.1 = zext i8 %3 to i32
   %add.1 = add nsw i32 %conv27.1, %conv6.1
   %conv3.1 = trunc i32 %add.1 to i8
-  %arrayidx4.1 = getelementptr inbounds i8* %c, i32 %add5
+  %arrayidx4.1 = getelementptr inbounds i8, i8* %c, i32 %add5
   store i8 %conv3.1, i8* %arrayidx4.1, align 1
   %add5.1 = add i32 %add5, %s
   %cmp.1 = icmp ult i32 %add5.1, %len
   br i1 %cmp.1, label %for.body.2, label %for.end
 
 for.body.2:                                       ; preds = %for.body.1
-  %arrayidx.2 = getelementptr inbounds i8* %a, i32 %add5.1
-  %4 = load i8* %arrayidx.2, align 1
+  %arrayidx.2 = getelementptr inbounds i8, i8* %a, i32 %add5.1
+  %4 = load i8, i8* %arrayidx.2, align 1
   %conv6.2 = zext i8 %4 to i32
-  %arrayidx1.2 = getelementptr inbounds i8* %b, i32 %add5.1
-  %5 = load i8* %arrayidx1.2, align 1
+  %arrayidx1.2 = getelementptr inbounds i8, i8* %b, i32 %add5.1
+  %5 = load i8, i8* %arrayidx1.2, align 1
   %conv27.2 = zext i8 %5 to i32
   %add.2 = add nsw i32 %conv27.2, %conv6.2
   %conv3.2 = trunc i32 %add.2 to i8
-  %arrayidx4.2 = getelementptr inbounds i8* %c, i32 %add5.1
+  %arrayidx4.2 = getelementptr inbounds i8, i8* %c, i32 %add5.1
   store i8 %conv3.2, i8* %arrayidx4.2, align 1
   %add5.2 = add i32 %add5.1, %s
   %cmp.2 = icmp ult i32 %add5.2, %len
   br i1 %cmp.2, label %for.body.3, label %for.end
 
 for.body.3:                                       ; preds = %for.body.2
-  %arrayidx.3 = getelementptr inbounds i8* %a, i32 %add5.2
-  %6 = load i8* %arrayidx.3, align 1
+  %arrayidx.3 = getelementptr inbounds i8, i8* %a, i32 %add5.2
+  %6 = load i8, i8* %arrayidx.3, align 1
   %conv6.3 = zext i8 %6 to i32
-  %arrayidx1.3 = getelementptr inbounds i8* %b, i32 %add5.2
-  %7 = load i8* %arrayidx1.3, align 1
+  %arrayidx1.3 = getelementptr inbounds i8, i8* %b, i32 %add5.2
+  %7 = load i8, i8* %arrayidx1.3, align 1
   %conv27.3 = zext i8 %7 to i32
   %add.3 = add nsw i32 %conv27.3, %conv6.3
   %conv3.3 = trunc i32 %add.3 to i8
-  %arrayidx4.3 = getelementptr inbounds i8* %c, i32 %add5.2
+  %arrayidx4.3 = getelementptr inbounds i8, i8* %c, i32 %add5.2
   store i8 %conv3.3, i8* %arrayidx4.3, align 1
   %add5.3 = add i32 %add5.2, %s
   %cmp.3 = icmp ult i32 %add5.3, %len
diff --git a/test/Transforms/LoopStrengthReduce/X86/no_superflous_induction_vars.ll b/test/Transforms/LoopStrengthReduce/X86/no_superflous_induction_vars.ll
index 55069947240d..a6613c53d786 100644
--- a/test/Transforms/LoopStrengthReduce/X86/no_superflous_induction_vars.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/no_superflous_induction_vars.ll
@@ -38,9 +38,9 @@ for.body14.lr.ph:
 
 vector.body:
   %index = phi i64 [ %index.next, %vector.body ], [ %0, %for.body14.lr.ph ]
-  %4 = getelementptr inbounds i8* %rowsptr, i64 %index
+  %4 = getelementptr inbounds i8, i8* %rowsptr, i64 %index
   %5 = bitcast i8* %4 to <4 x i8>*
-  %wide.load = load <4 x i8>* %5, align 1
+  %wide.load = load <4 x i8>, <4 x i8>* %5, align 1
   %index.next = add i64 %index, 8
   %6 = icmp eq i64 %index.next, %end.idx.rnd.down
   br i1 %6, label %for.end24, label %vector.body
diff --git a/test/Transforms/LoopStrengthReduce/X86/pr17473.ll b/test/Transforms/LoopStrengthReduce/X86/pr17473.ll
index e7ebaa8ea61b..5b7bb8846044 100644
--- a/test/Transforms/LoopStrengthReduce/X86/pr17473.ll
+++ b/test/Transforms/LoopStrengthReduce/X86/pr17473.ll
@@ -25,9 +25,9 @@ target triple = "x86_64-apple-macosx10.9.0"
 define i32 @main() #0 {
 entry:
   store i8 0, i8* @h, align 1
-  %0 = load i32* @j, align 4
+  %0 = load i32, i32* @j, align 4
   %tobool.i = icmp eq i32 %0, 0
-  %1 = load i32* @d, align 4
+  %1 = load i32, i32* @d, align 4
   %cmp3 = icmp sgt i32 %1, -1
   %.lobit = lshr i32 %1, 31
   %.lobit.not = xor i32 %.lobit, 1
@@ -55,7 +55,7 @@ for.end:                                          ; preds = %fn3.exit
   %conv7 = sext i8 %inc to i32
   %add = add nsw i32 %conv7, %conv
   store i32 %add, i32* @e, align 4
-  %call = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i32 %add) #2
+  %call = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i32 %add) #2
   ret i32 0
 }
 
diff --git a/test/Transforms/LoopStrengthReduce/addrec-gep-address-space.ll b/test/Transforms/LoopStrengthReduce/addrec-gep-address-space.ll
index 6333291aa65d..5650f81a3087 100644
--- a/test/Transforms/LoopStrengthReduce/addrec-gep-address-space.ll
+++ b/test/Transforms/LoopStrengthReduce/addrec-gep-address-space.ll
@@ -1,15 +1,15 @@
 ; RUN: opt < %s -loop-reduce -S | FileCheck %s
 ; CHECK: bb1:
-; CHECK: load double addrspace(1)* [[IV:%[^,]+]]
+; CHECK: load double, double addrspace(1)* [[IV:%[^,]+]]
 ; CHECK: store double {{.*}}, double addrspace(1)* [[IV]]
 
 ; CHECK-NOT: cast
 ; Make sure the GEP has the right index type
-; CHECK: getelementptr double addrspace(1)* [[IV]], i16 1
+; CHECK: getelementptr double, double addrspace(1)* [[IV]], i16 1
 ; CHECK: br {{.*}} label %bb1
 
 ; Make sure the GEP has the right index type
-; CHECK: getelementptr double addrspace(1)* {{.*}}, i16
+; CHECK: getelementptr double, double addrspace(1)* {{.*}}, i16
 
 
 ; This test tests several things. The load and store should use the
@@ -36,11 +36,11 @@ bb1:		; preds = %bb2, %bb.nph
 	%tmp3 = add i64 %j.01, %tmp1		; <i64> [#uses=1]
 	%tmp4 = add i64 %j.01, %tmp2		; <i64> [#uses=1]
         %z0 = add i64 %tmp3, 5203
-	%tmp5 = getelementptr double addrspace(1)* %p, i64 %z0		; <double addrspace(1)*> [#uses=1]
-	%tmp6 = load double addrspace(1)* %tmp5, align 8		; <double> [#uses=1]
+	%tmp5 = getelementptr double, double addrspace(1)* %p, i64 %z0		; <double addrspace(1)*> [#uses=1]
+	%tmp6 = load double, double addrspace(1)* %tmp5, align 8		; <double> [#uses=1]
 	%tmp7 = fdiv double %tmp6, 2.100000e+00		; <double> [#uses=1]
         %z1 = add i64 %tmp4, 5203
-	%tmp8 = getelementptr double addrspace(1)* %p, i64 %z1		; <double addrspace(1)*> [#uses=1]
+	%tmp8 = getelementptr double, double addrspace(1)* %p, i64 %z1		; <double addrspace(1)*> [#uses=1]
 	store double %tmp7, double addrspace(1)* %tmp8, align 8
 	%tmp9 = add i64 %j.01, 1		; <i64> [#uses=2]
 	br label %bb2
diff --git a/test/Transforms/LoopStrengthReduce/addrec-gep.ll b/test/Transforms/LoopStrengthReduce/addrec-gep.ll
index 3e4e369657b0..6919a33d8a27 100644
--- a/test/Transforms/LoopStrengthReduce/addrec-gep.ll
+++ b/test/Transforms/LoopStrengthReduce/addrec-gep.ll
@@ -1,8 +1,8 @@
 ; RUN: opt < %s -loop-reduce -S | FileCheck %s
 ; CHECK: bb1:
-; CHECK: load double* [[IV:%[^,]+]]
+; CHECK: load double, double* [[IV:%[^,]+]]
 ; CHECK: store double {{.*}}, double* [[IV]]
-; CHECK: getelementptr double*
+; CHECK: getelementptr double, double*
 ; CHECK-NOT: cast
 ; CHECK: br {{.*}} label %bb1
 
@@ -30,11 +30,11 @@ bb1:		; preds = %bb2, %bb.nph
 	%tmp3 = add i64 %j.01, %tmp1		; <i64> [#uses=1]
 	%tmp4 = add i64 %j.01, %tmp2		; <i64> [#uses=1]
         %z0 = add i64 %tmp3, 5203
-	%tmp5 = getelementptr double* %p, i64 %z0		; <double*> [#uses=1]
-	%tmp6 = load double* %tmp5, align 8		; <double> [#uses=1]
+	%tmp5 = getelementptr double, double* %p, i64 %z0		; <double*> [#uses=1]
+	%tmp6 = load double, double* %tmp5, align 8		; <double> [#uses=1]
 	%tmp7 = fdiv double %tmp6, 2.100000e+00		; <double> [#uses=1]
         %z1 = add i64 %tmp4, 5203
-	%tmp8 = getelementptr double* %p, i64 %z1		; <double*> [#uses=1]
+	%tmp8 = getelementptr double, double* %p, i64 %z1		; <double*> [#uses=1]
 	store double %tmp7, double* %tmp8, align 8
 	%tmp9 = add i64 %j.01, 1		; <i64> [#uses=2]
 	br label %bb2
diff --git a/test/Transforms/LoopStrengthReduce/address-space-loop.ll b/test/Transforms/LoopStrengthReduce/address-space-loop.ll
index 9c1b213b5979..57ba665b5f45 100644
--- a/test/Transforms/LoopStrengthReduce/address-space-loop.ll
+++ b/test/Transforms/LoopStrengthReduce/address-space-loop.ll
@@ -27,7 +27,7 @@ bb3:                                              ; preds = %bb2, %bb
 ; CHECK: bb10:
 ; CHECK-NEXT: %t7 = icmp eq i16 %t4, 0
 ; Host %t2 computation outside the loop.
-; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8 addrspace(1)* undef, i16 %t4
+; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8, i8 addrspace(1)* undef, i16 %t4
 ; CHECK-NEXT: br label %bb14
 bb10:                                             ; preds = %bb9
   %t7 = icmp eq i16 %t4, 0                    ; <i1> [#uses=1]
@@ -36,20 +36,20 @@ bb10:                                             ; preds = %bb9
 
 ; CHECK: bb14:
 ; CHECK-NEXT: store i8 undef, i8 addrspace(1)* [[SCEVGEP]]
-; CHECK-NEXT: %t6 = load float addrspace(1)* addrspace(1)* undef
+; CHECK-NEXT: %t6 = load float addrspace(1)*, float addrspace(1)* addrspace(1)* undef
 ; Fold %t3's add within the address.
-; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float addrspace(1)* %t6, i16 4
+; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float, float addrspace(1)* %t6, i16 4
 ; CHECK-NEXT: [[SCEVGEP2:%[^ ]+]] = bitcast float addrspace(1)* [[SCEVGEP1]] to i8 addrspace(1)*
 ; Use the induction variable (%t4) to access the right element
-; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8 addrspace(1)* [[SCEVGEP2]], i16 %t4
+; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8, i8 addrspace(1)* [[SCEVGEP2]], i16 %t4
 ; CHECK-NEXT: store i8 undef, i8 addrspace(1)* [[ADDRESS]]
 ; CHECK-NEXT: br label %bb14
 bb14:                                             ; preds = %bb14, %bb10
-  %t2 = getelementptr inbounds i8 addrspace(1)* undef, i16 %t4 ; <i8*> [#uses=1]
+  %t2 = getelementptr inbounds i8, i8 addrspace(1)* undef, i16 %t4 ; <i8*> [#uses=1]
   store i8 undef, i8 addrspace(1)* %t2
-  %t6 = load float addrspace(1)* addrspace(1)* undef
+  %t6 = load float addrspace(1)*, float addrspace(1)* addrspace(1)* undef
   %t8 = bitcast float addrspace(1)* %t6 to i8 addrspace(1)*              ; <i8*> [#uses=1]
-  %t9 = getelementptr inbounds i8 addrspace(1)* %t8, i16 %t3 ; <i8*> [#uses=1]
+  %t9 = getelementptr inbounds i8, i8 addrspace(1)* %t8, i16 %t3 ; <i8*> [#uses=1]
   store i8 undef, i8 addrspace(1)* %t9
   br label %bb14
 }
diff --git a/test/Transforms/LoopStrengthReduce/count-to-zero.ll b/test/Transforms/LoopStrengthReduce/count-to-zero.ll
index feb79f8a0c71..ca93e38ab64f 100644
--- a/test/Transforms/LoopStrengthReduce/count-to-zero.ll
+++ b/test/Transforms/LoopStrengthReduce/count-to-zero.ll
@@ -1,6 +1,9 @@
 ; RUN: opt < %s -loop-reduce -S | FileCheck %s
 ; rdar://7382068
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 define void @t(i32 %c) nounwind optsize {
 entry:
   br label %bb6
@@ -19,7 +22,7 @@ bb3:                                              ; preds = %bb1
   %tmp4 = add i32 %c_addr.1, -1                   ; <i32> [#uses=1]
   %c_addr.1.be = select i1 %tmp2, i32 %tmp3, i32 %tmp4 ; <i32> [#uses=1]
   %indvar.next = add i32 %indvar, 1               ; <i32> [#uses=1]
-; CHECK: add i32 %lsr.iv, -1
+; CHECK: add nsw i32 %lsr.iv, -1
   br label %bb6
 
 bb6:                                              ; preds = %bb3, %entry
diff --git a/test/Transforms/LoopStrengthReduce/different-type-ivs.ll b/test/Transforms/LoopStrengthReduce/different-type-ivs.ll
index 8cdd264591c8..c24f877a5502 100644
--- a/test/Transforms/LoopStrengthReduce/different-type-ivs.ll
+++ b/test/Transforms/LoopStrengthReduce/different-type-ivs.ll
@@ -11,7 +11,7 @@ no_exit:		; preds = %no_exit, %entry
 	%X.0.0 = mul i16 %indvar.upgrd.1, 1234		; <i16> [#uses=1]
 	%tmp. = mul i32 %indvar, 1234		; <i32> [#uses=1]
 	%tmp.5 = sext i16 %X.0.0 to i32		; <i32> [#uses=1]
-	%tmp.3 = call i32 (...)* @bar( i32 %tmp.5, i32 %tmp. )		; <i32> [#uses=0]
+	%tmp.3 = call i32 (...) @bar( i32 %tmp.5, i32 %tmp. )		; <i32> [#uses=0]
 	%tmp.0 = call i1 @pred( )		; <i1> [#uses=1]
 	%indvar.next = add i32 %indvar, 1		; <i32> [#uses=1]
 	br i1 %tmp.0, label %return, label %no_exit
diff --git a/test/Transforms/LoopStrengthReduce/dominate-assert.ll b/test/Transforms/LoopStrengthReduce/dominate-assert.ll
index 3ba93ff74835..bca234c6f3c4 100644
--- a/test/Transforms/LoopStrengthReduce/dominate-assert.ll
+++ b/test/Transforms/LoopStrengthReduce/dominate-assert.ll
@@ -23,7 +23,7 @@ bb4:
   br label %bb9
 bb5:
   %v5 = bitcast i8* %v3 to i32**
-  %add.ptr.i = getelementptr inbounds i32** %v5, i64 %v0
+  %add.ptr.i = getelementptr inbounds i32*, i32** %v5, i64 %v0
   br label %bb6
 bb6:
   %v6 = phi i32** [ null, %bb2 ], [ %add.ptr.i, %bb5 ]
@@ -49,7 +49,7 @@ bb2:
   invoke void @g() optsize
           to label %bb3 unwind label %bb6
 bb3:
-  %arrayctor.next = getelementptr inbounds i8* %arrayctor.cur, i64 1
+  %arrayctor.next = getelementptr inbounds i8, i8* %arrayctor.cur, i64 1
   br label %bb2
 bb4:
   ret void
@@ -76,17 +76,17 @@ entry:
 
 while.cond:                                       ; preds = %while.cond, %entry
   %c.0 = phi i16* [ undef, %entry ], [ %incdec.ptr, %while.cond ]
-  %incdec.ptr = getelementptr inbounds i16* %c.0, i64 1
+  %incdec.ptr = getelementptr inbounds i16, i16* %c.0, i64 1
   br i1 undef, label %while.cond1, label %while.cond
 
 while.cond1:                                      ; preds = %while.cond1, %while.cond
   %c.1 = phi i16* [ %incdec.ptr5, %while.cond1 ], [ %c.0, %while.cond ]
-  %incdec.ptr5 = getelementptr inbounds i16* %c.1, i64 1
+  %incdec.ptr5 = getelementptr inbounds i16, i16* %c.1, i64 1
   br i1 undef, label %while.cond7, label %while.cond1
 
 while.cond7:                                      ; preds = %while.cond7, %while.cond1
   %0 = phi i16* [ %incdec.ptr10, %while.cond7 ], [ %c.1, %while.cond1 ]
-  %incdec.ptr10 = getelementptr inbounds i16* %0, i64 1
+  %incdec.ptr10 = getelementptr inbounds i16, i16* %0, i64 1
   br i1 undef, label %while.cond12.preheader, label %while.cond7
 
 while.cond12.preheader:                           ; preds = %while.cond7
@@ -104,7 +104,7 @@ while.body13.lr.ph:                               ; preds = %while.cond12.outer.
   br label %while.body13
 
 if.else:                                          ; preds = %while.body13
-  %incdec.ptr15 = getelementptr inbounds i16* %1, i64 1
+  %incdec.ptr15 = getelementptr inbounds i16, i16* %1, i64 1
   %cmp = icmp eq i16* %incdec.ptr15, %0
   br i1 %cmp, label %while.end16, label %while.body13
 
diff --git a/test/Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll b/test/Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll
index f7a82f6616d2..7b92acee5a91 100644
--- a/test/Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll
+++ b/test/Transforms/LoopStrengthReduce/dont-hoist-simple-loop-constants.ll
@@ -10,7 +10,7 @@ then.0:
 no_exit.2:		; preds = %no_exit.2, %then.0
 	%indvar630 = phi i32 [ 0, %then.0 ], [ %indvar.next631, %no_exit.2 ]		; <i32> [#uses=4]
 	%gep.upgrd.1 = zext i32 %indvar630 to i64		; <i64> [#uses=1]
-	%tmp.38 = getelementptr [700 x i32]* %nbeaux_.0__558, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
+	%tmp.38 = getelementptr [700 x i32], [700 x i32]* %nbeaux_.0__558, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp.38
 	%inc.2 = add i32 %indvar630, 2		; <i32> [#uses=2]
 	%tmp.34 = icmp slt i32 %inc.2, 701		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll b/test/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll
index 16bb508d2e27..4a6ec50ab97d 100644
--- a/test/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll
+++ b/test/Transforms/LoopStrengthReduce/dont_insert_redundant_ops.ll
@@ -12,9 +12,9 @@ define void @test1({ i32, i32 }* %P) {
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=3]
-	%gep1 = getelementptr { i32, i32 }* %P, i32 %INDVAR, i32 0		; <i32*> [#uses=1]
+	%gep1 = getelementptr { i32, i32 }, { i32, i32 }* %P, i32 %INDVAR, i32 0		; <i32*> [#uses=1]
 	store i32 0, i32* %gep1
-	%gep2 = getelementptr { i32, i32 }* %P, i32 %INDVAR, i32 1		; <i32*> [#uses=1]
+	%gep2 = getelementptr { i32, i32 }, { i32, i32 }* %P, i32 %INDVAR, i32 1		; <i32*> [#uses=1]
 	store i32 0, i32* %gep2
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
 	%cond = call i1 @pred( )		; <i1> [#uses=1]
@@ -28,9 +28,9 @@ define void @test2([2 x i32]* %P) {
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=3]
-	%gep1 = getelementptr [2 x i32]* %P, i32 %INDVAR, i64 0		; <i32*> [#uses=1]
+	%gep1 = getelementptr [2 x i32], [2 x i32]* %P, i32 %INDVAR, i64 0		; <i32*> [#uses=1]
 	store i32 0, i32* %gep1
-	%gep2 = getelementptr [2 x i32]* %P, i32 %INDVAR, i64 1		; <i32*> [#uses=1]
+	%gep2 = getelementptr [2 x i32], [2 x i32]* %P, i32 %INDVAR, i64 1		; <i32*> [#uses=1]
 	store i32 0, i32* %gep2
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
 	%cond = call i1 @pred( )		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/dont_reduce_bytes.ll b/test/Transforms/LoopStrengthReduce/dont_reduce_bytes.ll
index 20300002eb39..4974f485d1c1 100644
--- a/test/Transforms/LoopStrengthReduce/dont_reduce_bytes.ll
+++ b/test/Transforms/LoopStrengthReduce/dont_reduce_bytes.ll
@@ -11,7 +11,7 @@ define void @test(i8* %PTR) {
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=2]
-	%STRRED = getelementptr i8* %PTR, i32 %INDVAR		; <i8*> [#uses=1]
+	%STRRED = getelementptr i8, i8* %PTR, i32 %INDVAR		; <i8*> [#uses=1]
 	store i8 0, i8* %STRRED
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=2]
         ;; cannot eliminate indvar
diff --git a/test/Transforms/LoopStrengthReduce/dont_reverse.ll b/test/Transforms/LoopStrengthReduce/dont_reverse.ll
index d65213d06af3..551bd03f0ae3 100644
--- a/test/Transforms/LoopStrengthReduce/dont_reverse.ll
+++ b/test/Transforms/LoopStrengthReduce/dont_reverse.ll
@@ -4,6 +4,10 @@
 ; Don't reverse the iteration if the rhs of the compare is defined
 ; inside the loop.
 
+; Provide legal integer types.
+; Declare i2 as legal so that IVUsers accepts to consider %indvar3451
+target datalayout = "n2:8:16:32:64"
+
 define void @Fill_Buffer(i2* %p) nounwind {
 entry:
 	br label %bb8
@@ -11,7 +15,7 @@ entry:
 bb8:
 	%indvar34 = phi i32 [ 0, %entry ], [ %indvar.next35, %bb8 ]
 	%indvar3451 = trunc i32 %indvar34 to i2
-	%xmp4344 = load i2* %p
+	%xmp4344 = load i2, i2* %p
 	%xmp104 = icmp eq i2 %indvar3451, %xmp4344
 	%indvar.next35 = add i32 %indvar34, 1
 	br i1 %xmp104, label %bb10, label %bb8
diff --git a/test/Transforms/LoopStrengthReduce/hoist-parent-preheader.ll b/test/Transforms/LoopStrengthReduce/hoist-parent-preheader.ll
index 7982fbc8c4a5..27371b56e611 100644
--- a/test/Transforms/LoopStrengthReduce/hoist-parent-preheader.ll
+++ b/test/Transforms/LoopStrengthReduce/hoist-parent-preheader.ll
@@ -15,7 +15,7 @@ loop2:                                            ; preds = %loop1, %loop2.backe
   %indvar414 = phi i64 [ %indvar.next415, %loop2.backedge ], [ 0, %loop1 ]
   %tmp473 = mul i64 %indvar414, -4
   %tmp485 = add i64 %tmp484, %tmp473
-  %storemerge4 = getelementptr i8* %a, i64 %tmp485
+  %storemerge4 = getelementptr i8, i8* %a, i64 %tmp485
   %0 = icmp ugt i8* %storemerge4, %a
   br i1 false, label %loop2.exit, label %loop2.backedge
 
diff --git a/test/Transforms/LoopStrengthReduce/invariant_value_first.ll b/test/Transforms/LoopStrengthReduce/invariant_value_first.ll
index 5c18809e59c2..4d59ca9869c2 100644
--- a/test/Transforms/LoopStrengthReduce/invariant_value_first.ll
+++ b/test/Transforms/LoopStrengthReduce/invariant_value_first.ll
@@ -13,7 +13,7 @@ define void @test([10000 x i32]* %P) {
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=2]
-	%STRRED = getelementptr [10000 x i32]* %P, i32 %outer, i32 %INDVAR		; <i32*> [#uses=1]
+	%STRRED = getelementptr [10000 x i32], [10000 x i32]* %P, i32 %outer, i32 %INDVAR		; <i32*> [#uses=1]
 	store i32 0, i32* %STRRED
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
 	%cond = call i1 @pred( )		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/invariant_value_first_arg.ll b/test/Transforms/LoopStrengthReduce/invariant_value_first_arg.ll
index 8eb8f05a8a1f..57716408a298 100644
--- a/test/Transforms/LoopStrengthReduce/invariant_value_first_arg.ll
+++ b/test/Transforms/LoopStrengthReduce/invariant_value_first_arg.ll
@@ -10,7 +10,7 @@ define void @test([10000 x i32]* %P, i32 %outer) {
 	br label %Loop
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=2]
-	%STRRED = getelementptr [10000 x i32]* %P, i32 %outer, i32 %INDVAR		; <i32*> [#uses=1]
+	%STRRED = getelementptr [10000 x i32], [10000 x i32]* %P, i32 %outer, i32 %INDVAR		; <i32*> [#uses=1]
 	store i32 0, i32* %STRRED
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
 	%cond = call i1 @pred( )		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/ivchain.ll b/test/Transforms/LoopStrengthReduce/ivchain.ll
index 233800b71c64..d95220d238ac 100644
--- a/test/Transforms/LoopStrengthReduce/ivchain.ll
+++ b/test/Transforms/LoopStrengthReduce/ivchain.ll
@@ -4,6 +4,10 @@
 ; A sign extend feeds an IVUser and cannot be hoisted into the AddRec.
 ; CollectIVChains should bailout on this case.
 
+
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 %struct = type { i8*, i8*, i16, i64, i16, i16, i16, i64, i64, i16, i8*, i64, i64, i64 }
 
 ; CHECK-LABEL: @test(
@@ -27,13 +31,13 @@ for.body:                                         ; preds = %sw.epilog, %land.en
   br label %if.end388
 
 if.end388:                                        ; preds = %if.then380, %if.else356
-  %ColLength = getelementptr inbounds %struct* %fbh.0, i64 0, i32 7
+  %ColLength = getelementptr inbounds %struct, %struct* %fbh.0, i64 0, i32 7
   %call405 = call signext i16 @SQLColAttribute(i8* undef, i16 zeroext %conv258, i16 zeroext 1003, i8* null, i16 signext 0, i16* null, i64* %ColLength) nounwind
   br label %sw.epilog
 
 sw.epilog:                                        ; preds = %sw.bb542, %sw.bb523, %if.end475
   %inc601 = add i16 %column_n.0, 1
-  %incdec.ptr = getelementptr inbounds %struct* %fbh.0, i64 1
+  %incdec.ptr = getelementptr inbounds %struct, %struct* %fbh.0, i64 1
   br label %for.body
 
 return:                                           ; preds = %entry
diff --git a/test/Transforms/LoopStrengthReduce/negative-scale.ll b/test/Transforms/LoopStrengthReduce/negative-scale.ll
new file mode 100644
index 000000000000..fdde1b6bf9dc
--- /dev/null
+++ b/test/Transforms/LoopStrengthReduce/negative-scale.ll
@@ -0,0 +1,28 @@
+; RUN: opt < %s -loop-reduce -S | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
+
+declare void @foo(i8)
+
+define void @not_addressing_mode(i8* %input, i64 %n) {
+; CHECK-LABEL: @not_addressing_mode(
+entry:
+  br label %loop
+
+loop:
+; CHECK: loop:
+; CHECK: %lsr.iv = phi i8* [ {{%[^,]+}}, %loop ], [ %input, %entry ]
+  %i = phi i64 [ 0, %entry ], [ %i.next, %loop ]
+  %i.next = add i64 %i, 1
+  %j = mul i64 %i, -2
+  ; (%input - 2 * %j) is not foldable. Worth another indvar.
+  %p = getelementptr i8, i8* %input, i64 %j
+  %v = load i8, i8* %p
+; CHECK: %v = load i8, i8* %lsr.iv
+  call void @foo(i8 %v)
+  %exitcond = icmp slt i64 %i.next, %n
+  br i1 %exitcond, label %exit, label %loop
+
+exit:
+  ret void
+}
diff --git a/test/Transforms/LoopStrengthReduce/nested-reduce.ll b/test/Transforms/LoopStrengthReduce/nested-reduce.ll
index 58b8d3eecd03..c05b19d8fdd5 100644
--- a/test/Transforms/LoopStrengthReduce/nested-reduce.ll
+++ b/test/Transforms/LoopStrengthReduce/nested-reduce.ll
@@ -1,5 +1,8 @@
 ; RUN: opt < %s -loop-reduce -S | not grep mul
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 ; Make sure we don't get a multiply by 6 in this loop.
 
 define i32 @foo(i32 %A, i32 %B, i32 %C, i32 %D) {
diff --git a/test/Transforms/LoopStrengthReduce/ops_after_indvar.ll b/test/Transforms/LoopStrengthReduce/ops_after_indvar.ll
index ec55179d6495..ad2caeb274f4 100644
--- a/test/Transforms/LoopStrengthReduce/ops_after_indvar.ll
+++ b/test/Transforms/LoopStrengthReduce/ops_after_indvar.ll
@@ -15,7 +15,7 @@ define void @test([10000 x i32]* %P) {
 Loop:		; preds = %Loop, %0
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=2]
 	%idx = call i32 @getidx( )		; <i32> [#uses=1]
-	%STRRED = getelementptr [10000 x i32]* %P, i32 %INDVAR, i32 %idx		; <i32*> [#uses=1]
+	%STRRED = getelementptr [10000 x i32], [10000 x i32]* %P, i32 %INDVAR, i32 %idx		; <i32*> [#uses=1]
 	store i32 0, i32* %STRRED
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
 	%cond = call i1 @pred( )		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/phi_node_update_multiple_preds.ll b/test/Transforms/LoopStrengthReduce/phi_node_update_multiple_preds.ll
index 7ef494debdda..6943ab9f29a9 100644
--- a/test/Transforms/LoopStrengthReduce/phi_node_update_multiple_preds.ll
+++ b/test/Transforms/LoopStrengthReduce/phi_node_update_multiple_preds.ll
@@ -10,12 +10,12 @@ entry:
 shortcirc_next:		; preds = %no_exit.2, %entry
 	%indvar37 = phi i32 [ 0, %entry ], [ %indvar.next38, %no_exit.2 ]		; <i32> [#uses=3]
 	%gep.upgrd.1 = zext i32 %indvar37 to i64		; <i64> [#uses=1]
-	%wp.2.4 = getelementptr i8* null, i64 %gep.upgrd.1		; <i8*> [#uses=1]
+	%wp.2.4 = getelementptr i8, i8* null, i64 %gep.upgrd.1		; <i8*> [#uses=1]
 	br i1 false, label %loopexit.2, label %no_exit.2
 no_exit.2:		; preds = %shortcirc_next
 	%wp.2.4.rec = bitcast i32 %indvar37 to i32		; <i32> [#uses=1]
 	%inc.1.rec = add i32 %wp.2.4.rec, 1		; <i32> [#uses=1]
-	%inc.1 = getelementptr i8* null, i32 %inc.1.rec		; <i8*> [#uses=2]
+	%inc.1 = getelementptr i8, i8* null, i32 %inc.1.rec		; <i8*> [#uses=2]
 	%indvar.next38 = add i32 %indvar37, 1		; <i32> [#uses=1]
 	switch i8 0, label %shortcirc_next [
 		 i8 32, label %loopexit.2
diff --git a/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll b/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll
index 45aeb4e691a0..092b274bfc03 100644
--- a/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll
+++ b/test/Transforms/LoopStrengthReduce/post-inc-icmpzero.ll
@@ -23,19 +23,19 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define void @_Z15IntegerToStringjjR7Vector2(i32 %i, i32 %radix, %struct.Vector2* nocapture %result) nounwind noinline {
 entry:
   %buffer = alloca [33 x i16], align 16
-  %add.ptr = getelementptr inbounds [33 x i16]* %buffer, i64 0, i64 33
+  %add.ptr = getelementptr inbounds [33 x i16], [33 x i16]* %buffer, i64 0, i64 33
   br label %do.body
 
 do.body:                                          ; preds = %do.body, %entry
   %0 = phi i64 [ %indvar.next44, %do.body ], [ 0, %entry ]
   %i.addr.0 = phi i32 [ %div, %do.body ], [ %i, %entry ]
   %tmp51 = sub i64 32, %0
-  %incdec.ptr = getelementptr [33 x i16]* %buffer, i64 0, i64 %tmp51
+  %incdec.ptr = getelementptr [33 x i16], [33 x i16]* %buffer, i64 0, i64 %tmp51
   %rem = urem i32 %i.addr.0, 10
   %div = udiv i32 %i.addr.0, 10
   %idxprom = zext i32 %rem to i64
-  %arrayidx = getelementptr inbounds [37 x i8]* @.str, i64 0, i64 %idxprom
-  %tmp5 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds [37 x i8], [37 x i8]* @.str, i64 0, i64 %idxprom
+  %tmp5 = load i8, i8* %arrayidx, align 1
   %conv = sext i8 %tmp5 to i16
   store i16 %conv, i16* %incdec.ptr, align 2
   %1 = icmp ugt i32 %i.addr.0, 9
@@ -50,30 +50,30 @@ do.end:                                           ; preds = %do.body
   %sub.ptr.sub = sub i64 %sub.ptr.lhs.cast, %sub.ptr.rhs.cast
   %sub.ptr.div39 = lshr exact i64 %sub.ptr.sub, 1
   %conv11 = trunc i64 %sub.ptr.div39 to i32
-  %mLength = getelementptr inbounds %struct.Vector2* %result, i64 0, i32 2
+  %mLength = getelementptr inbounds %struct.Vector2, %struct.Vector2* %result, i64 0, i32 2
   %idx.ext21 = bitcast i64 %sub.ptr.div39 to i64
   %incdec.ptr.sum = add i64 %idx.ext21, -1
   %cp.0.sum = sub i64 %incdec.ptr.sum, %0
-  %add.ptr22 = getelementptr [33 x i16]* %buffer, i64 1, i64 %cp.0.sum
+  %add.ptr22 = getelementptr [33 x i16], [33 x i16]* %buffer, i64 1, i64 %cp.0.sum
   %cmp2740 = icmp eq i64 %idx.ext21, 0
   br i1 %cmp2740, label %for.end, label %for.body.lr.ph
 
 for.body.lr.ph:                                   ; preds = %do.end
-  %tmp16 = load i32* %mLength, align 4
-  %mBegin = getelementptr inbounds %struct.Vector2* %result, i64 0, i32 0
-  %tmp14 = load i16** %mBegin, align 8
+  %tmp16 = load i32, i32* %mLength, align 4
+  %mBegin = getelementptr inbounds %struct.Vector2, %struct.Vector2* %result, i64 0, i32 0
+  %tmp14 = load i16*, i16** %mBegin, align 8
   %tmp48 = zext i32 %tmp16 to i64
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %for.body.lr.ph
   %indvar = phi i64 [ 0, %for.body.lr.ph ], [ %indvar.next, %for.body ]
   %tmp46 = add i64 %tmp51, %indvar
-  %p.042 = getelementptr [33 x i16]* %buffer, i64 0, i64 %tmp46
+  %p.042 = getelementptr [33 x i16], [33 x i16]* %buffer, i64 0, i64 %tmp46
   %tmp47 = sub i64 %indvar, %0
-  %incdec.ptr32 = getelementptr [33 x i16]* %buffer, i64 1, i64 %tmp47
+  %incdec.ptr32 = getelementptr [33 x i16], [33 x i16]* %buffer, i64 1, i64 %tmp47
   %tmp49 = add i64 %tmp48, %indvar
-  %dst.041 = getelementptr i16* %tmp14, i64 %tmp49
-  %tmp29 = load i16* %p.042, align 2
+  %dst.041 = getelementptr i16, i16* %tmp14, i64 %tmp49
+  %tmp29 = load i16, i16* %p.042, align 2
   store i16 %tmp29, i16* %dst.041, align 2
   %cmp27 = icmp eq i16* %incdec.ptr32, %add.ptr22
   %indvar.next = add i64 %indvar, 1
@@ -83,7 +83,7 @@ for.end.loopexit:                                 ; preds = %for.body
   br label %for.end
 
 for.end:                                          ; preds = %for.end.loopexit, %do.end
-  %tmp38 = load i32* %mLength, align 4
+  %tmp38 = load i32, i32* %mLength, align 4
   %add = add i32 %tmp38, %conv11
   store i32 %add, i32* %mLength, align 4
   ret void
diff --git a/test/Transforms/LoopStrengthReduce/pr12018.ll b/test/Transforms/LoopStrengthReduce/pr12018.ll
index e493cf85fdef..b15961a77904 100644
--- a/test/Transforms/LoopStrengthReduce/pr12018.ll
+++ b/test/Transforms/LoopStrengthReduce/pr12018.ll
@@ -12,15 +12,15 @@ entry:
 for.body:                                         ; preds = %_ZN8nsTArray9ElementAtEi.exit, %entry
   %i.06 = phi i32 [ %add, %_ZN8nsTArray9ElementAtEi.exit ], [ 0, %entry ]
   %call.i = call %struct.nsTArrayHeader* @_ZN8nsTArray4Hdr2Ev() nounwind
-  %add.ptr.i = getelementptr inbounds %struct.nsTArrayHeader* %call.i, i32 1
+  %add.ptr.i = getelementptr inbounds %struct.nsTArrayHeader, %struct.nsTArrayHeader* %call.i, i32 1
   %tmp = bitcast %struct.nsTArrayHeader* %add.ptr.i to %struct.nsTArray*
-  %arrayidx = getelementptr inbounds %struct.nsTArray* %tmp, i32 %i.06
+  %arrayidx = getelementptr inbounds %struct.nsTArray, %struct.nsTArray* %tmp, i32 %i.06
   %add = add nsw i32 %i.06, 1
-  call void @llvm.dbg.value(metadata %struct.nsTArray* %aValues, i64 0, metadata !0, metadata !{}) nounwind
+  call void @llvm.dbg.value(metadata %struct.nsTArray* %aValues, i64 0, metadata !0, metadata !DIExpression()) nounwind, !dbg !DILocation(scope: !DISubprogram())
   br label %_ZN8nsTArray9ElementAtEi.exit
 
 _ZN8nsTArray9ElementAtEi.exit:                    ; preds = %for.body
-  %arrayidx.i = getelementptr inbounds %struct.nsTArray* %tmp, i32 %add
+  %arrayidx.i = getelementptr inbounds %struct.nsTArray, %struct.nsTArray* %tmp, i32 %add
   call void @_ZN11nsTArray15ComputeDistanceERKS_Rd(%struct.nsTArray* %arrayidx, %struct.nsTArray* %arrayidx.i) nounwind
   %cmp = icmp slt i32 %add, %foo
   br i1 %cmp, label %for.body, label %for.end
@@ -35,4 +35,4 @@ declare %struct.nsTArrayHeader* @_ZN8nsTArray4Hdr2Ev()
 
 declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnone
 
-!0 = !{!"0x101"}                       ; [ DW_TAG_arg_variable ]
+!0 = !DILocalVariable(tag: DW_TAG_arg_variable, scope: !DISubprogram())
diff --git a/test/Transforms/LoopStrengthReduce/pr12048.ll b/test/Transforms/LoopStrengthReduce/pr12048.ll
index 7e0f2ad7a3c7..212004b24f8d 100644
--- a/test/Transforms/LoopStrengthReduce/pr12048.ll
+++ b/test/Transforms/LoopStrengthReduce/pr12048.ll
@@ -13,8 +13,8 @@ while.cond40:
     i8 0, label %if.then59
   ]
 while.body51:                                     ; preds = %land.end50
-  %incdec.ptr = getelementptr inbounds i8* %tmp.1, i64 1
-  %scevgep = getelementptr i8* %indvars.iv194, i64 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %tmp.1, i64 1
+  %scevgep = getelementptr i8, i8* %indvars.iv194, i64 1
   br label %while.cond40
 if.then59:                                        ; preds = %while.end
   br i1 undef, label %if.then64, label %if.end113
@@ -29,7 +29,7 @@ while.cond132.preheader:                          ; preds = %if.end113
   %cmp133173 = icmp eq i8* %tmp.1, %tmp.4
   br i1 %cmp133173, label %while.cond40.preheader, label %while.body139.lr.ph
 while.body139.lr.ph:                              ; preds = %while.cond132.preheader
-  %scevgep198 = getelementptr i8* %indvars.iv194, i64 0
+  %scevgep198 = getelementptr i8, i8* %indvars.iv194, i64 0
   %scevgep198199 = ptrtoint i8* %scevgep198 to i64
   br label %while.body139
 while.body139:                                    ; preds = %while.body139, %while.body139.lr.ph
diff --git a/test/Transforms/LoopStrengthReduce/pr12691.ll b/test/Transforms/LoopStrengthReduce/pr12691.ll
index 8399434343eb..e33e40503a3e 100644
--- a/test/Transforms/LoopStrengthReduce/pr12691.ll
+++ b/test/Transforms/LoopStrengthReduce/pr12691.ll
@@ -1,5 +1,8 @@
 ; RUN: opt < %s -loop-reduce -S | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 @d = common global i32 0, align 4
 
 define void @fn2(i32 %x) nounwind uwtable {
@@ -13,13 +16,13 @@ for.cond:
   br i1 %tobool, label %for.cond, label %for.end
 
 for.end:
-; CHECK:  %tmp1 = load i32* @d, align 4
-; CHECK-NEXT:  %tmp2 = load i32* @d, align 4
+; CHECK:  %tmp1 = load i32, i32* @d, align 4
+; CHECK-NEXT:  %tmp2 = load i32, i32* @d, align 4
 ; CHECK-NEXT:  %0 = sub i32 %tmp1, %tmp2
 
-  %tmp1 = load i32* @d, align 4
+  %tmp1 = load i32, i32* @d, align 4
   %add = add nsw i32 %tmp1, %g.0
-  %tmp2 = load i32* @d, align 4
+  %tmp2 = load i32, i32* @d, align 4
   %tobool26 = icmp eq i32 %x, 0
   br i1 %tobool26, label %for.end5, label %for.body.lr.ph
 
diff --git a/test/Transforms/LoopStrengthReduce/pr18165.ll b/test/Transforms/LoopStrengthReduce/pr18165.ll
index cc878c48c4b9..11c9c4ec403f 100644
--- a/test/Transforms/LoopStrengthReduce/pr18165.ll
+++ b/test/Transforms/LoopStrengthReduce/pr18165.ll
@@ -20,10 +20,10 @@ target triple = "x86_64-apple-macosx10.9.0"
 ; Function Attrs: nounwind optsize ssp uwtable
 define i32 @main() #0 {
 entry:
-  %0 = load i32* getelementptr inbounds (%struct.anon* @a, i64 0, i32 0), align 4, !tbaa !1
+  %0 = load i32, i32* getelementptr inbounds (%struct.anon, %struct.anon* @a, i64 0, i32 0), align 4, !tbaa !1
   %tobool7.i = icmp eq i32 %0, 0
-  %.promoted.i = load i32* getelementptr inbounds (%struct.anon* @a, i64 0, i32 2), align 4, !tbaa !6
-  %f.promoted.i = load i32* @f, align 4, !tbaa !7
+  %.promoted.i = load i32, i32* getelementptr inbounds (%struct.anon, %struct.anon* @a, i64 0, i32 2), align 4, !tbaa !6
+  %f.promoted.i = load i32, i32* @f, align 4, !tbaa !7
   br label %for.body6.i.outer
 
 for.body6.i.outer:                                ; preds = %entry, %lor.end.i
@@ -42,7 +42,7 @@ if.end9.i:                                        ; preds = %for.body6.i.outer
   br i1 %tobool12.i, label %lor.rhs.i, label %lor.end.i
 
 lor.rhs.i:                                        ; preds = %if.end9.i
-  %1 = load i32* @b, align 4, !tbaa !7
+  %1 = load i32, i32* @b, align 4, !tbaa !7
   %dec.i = add nsw i32 %1, -1
   store i32 %dec.i, i32* @b, align 4, !tbaa !7
   %tobool13.i = icmp ne i32 %1, 0
@@ -59,12 +59,12 @@ lor.end.i:                                        ; preds = %lor.rhs.i, %if.end9
 
 fn1.exit:                                         ; preds = %lor.end.i
   store i32 0, i32* @g, align 4, !tbaa !7
-  store i32 %or14.i, i32* getelementptr inbounds (%struct.anon* @a, i64 0, i32 2), align 4, !tbaa !6
+  store i32 %or14.i, i32* getelementptr inbounds (%struct.anon, %struct.anon* @a, i64 0, i32 2), align 4, !tbaa !6
   store i32 %or15.i, i32* @f, align 4, !tbaa !7
-  store i32 %add.i, i32* getelementptr inbounds (%struct.anon* @e, i64 0, i32 1), align 4, !tbaa !8
+  store i32 %add.i, i32* getelementptr inbounds (%struct.anon, %struct.anon* @e, i64 0, i32 1), align 4, !tbaa !8
   store i32 0, i32* @h, align 4, !tbaa !7
-  %3 = load i32* @b, align 4, !tbaa !7
-  %call1 = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i32 %3) #2
+  %3 = load i32, i32* @b, align 4, !tbaa !7
+  %call1 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i32 %3) #2
   ret i32 0
 }
 
diff --git a/test/Transforms/LoopStrengthReduce/pr2570.ll b/test/Transforms/LoopStrengthReduce/pr2570.ll
index 7b569713a93d..671ffde671f7 100644
--- a/test/Transforms/LoopStrengthReduce/pr2570.ll
+++ b/test/Transforms/LoopStrengthReduce/pr2570.ll
@@ -23,7 +23,7 @@ define i32 @func_44(i32 %p_45, i32 %p_46, i16 zeroext  %p_48, i32 %p_49, i8 zero
 entry:
 	tail call i32 @func_116( i8 zeroext  2 ) nounwind 		; <i32>:0 [#uses=0]
 	tail call i32 @func_63( i16 signext  2 ) nounwind 		; <i32>:1 [#uses=1]
-	load i16* @g_39, align 2		; <i16>:2 [#uses=1]
+	load i16, i16* @g_39, align 2		; <i16>:2 [#uses=1]
 	tail call i32 @func_63( i16 signext  %2 ) nounwind 		; <i32>:3 [#uses=1]
 	trunc i32 %3 to i16		; <i16>:4 [#uses=1]
 	and i16 %4, 1		; <i16>:5 [#uses=1]
@@ -32,10 +32,10 @@ entry:
 	tail call i32 @func_74( i16 zeroext  %5, i8 zeroext  %6, i16 zeroext  %7, i16 zeroext  0 ) nounwind 		; <i32>:8 [#uses=0]
 	tail call i32 @func_124( i32 544824386 ) nounwind 		; <i32>:9 [#uses=0]
 	zext i8 %p_50 to i32		; <i32>:10 [#uses=1]
-	load i32* @g_43, align 4		; <i32>:11 [#uses=1]
+	load i32, i32* @g_43, align 4		; <i32>:11 [#uses=1]
 	icmp sle i32 %10, %11		; <i1>:12 [#uses=1]
 	zext i1 %12 to i32		; <i32>:13 [#uses=2]
-	load i8* @g_247, align 1		; <i8>:14 [#uses=1]
+	load i8, i8* @g_247, align 1		; <i8>:14 [#uses=1]
 	trunc i32 %p_45 to i16		; <i16>:15 [#uses=1]
 	zext i8 %14 to i16		; <i16>:16 [#uses=1]
 	tail call i32 @func_74( i16 zeroext  %15, i8 zeroext  0, i16 zeroext  %16, i16 zeroext  23618 ) nounwind 		; <i32>:17 [#uses=4]
@@ -72,7 +72,7 @@ bb172:		; preds = %bb168, %bb162
 	icmp eq i32 %.0343, 0		; <i1>:30 [#uses=1]
 	%.0341 = select i1 %30, i32 1, i32 %.0343		; <i32> [#uses=1]
 	urem i32 %23, %.0341		; <i32>:31 [#uses=1]
-	load i32* @g_137, align 4		; <i32>:32 [#uses=4]
+	load i32, i32* @g_137, align 4		; <i32>:32 [#uses=4]
 	icmp slt i32 %32, 0		; <i1>:33 [#uses=1]
 	br i1 %33, label %bb202, label %bb198
 
@@ -119,13 +119,13 @@ bb222:		; preds = %bb215, %bb205
 
 bb223:		; preds = %bb222, %bb215
 	%iftmp.437.0 = phi i32 [ 0, %bb222 ], [ 1, %bb215 ]		; <i32> [#uses=1]
-	load i32* @g_91, align 4		; <i32>:55 [#uses=3]
+	load i32, i32* @g_91, align 4		; <i32>:55 [#uses=3]
 	tail call i32 @func_103( i16 zeroext  4 ) nounwind 		; <i32>:56 [#uses=0]
 	tail call i32 @func_112( i32 0, i16 zeroext  -31374 ) nounwind 		; <i32>:57 [#uses=0]
-	load i32* @g_197, align 4		; <i32>:58 [#uses=1]
+	load i32, i32* @g_197, align 4		; <i32>:58 [#uses=1]
 	tail call i32 @func_124( i32 28156 ) nounwind 		; <i32>:59 [#uses=1]
-	load i32* @g_260, align 4		; <i32>:60 [#uses=1]
-	load i32* @g_43, align 4		; <i32>:61 [#uses=1]
+	load i32, i32* @g_260, align 4		; <i32>:60 [#uses=1]
+	load i32, i32* @g_43, align 4		; <i32>:61 [#uses=1]
 	xor i32 %61, %60		; <i32>:62 [#uses=1]
 	mul i32 %62, %59		; <i32>:63 [#uses=1]
 	trunc i32 %63 to i8		; <i8>:64 [#uses=1]
@@ -138,7 +138,7 @@ bb223:		; preds = %bb222, %bb215
 	%or.cond352 = or i1 %70, %67		; <i1> [#uses=1]
 	select i1 %or.cond352, i32 0, i32 %55		; <i32>:71 [#uses=1]
 	%.353 = ashr i32 %66, %71		; <i32> [#uses=2]
-	load i16* @g_221, align 2		; <i16>:72 [#uses=1]
+	load i16, i16* @g_221, align 2		; <i16>:72 [#uses=1]
 	zext i16 %72 to i32		; <i32>:73 [#uses=1]
 	icmp ugt i32 %.353, 31		; <i1>:74 [#uses=1]
 	select i1 %74, i32 0, i32 %.353		; <i32>:75 [#uses=1]
@@ -146,7 +146,7 @@ bb223:		; preds = %bb222, %bb215
 	add i32 %.0323, %iftmp.437.0		; <i32>:76 [#uses=1]
 	and i32 %48, 255		; <i32>:77 [#uses=2]
 	add i32 %77, 2042556439		; <i32>:78 [#uses=1]
-	load i32* @g_207, align 4		; <i32>:79 [#uses=2]
+	load i32, i32* @g_207, align 4		; <i32>:79 [#uses=2]
 	icmp ugt i32 %79, 31		; <i1>:80 [#uses=1]
 	select i1 %80, i32 0, i32 %79		; <i32>:81 [#uses=1]
 	%.0320 = lshr i32 %77, %81		; <i32> [#uses=1]
@@ -154,7 +154,7 @@ bb223:		; preds = %bb222, %bb215
 	zext i1 %82 to i8		; <i8>:83 [#uses=1]
 	tail call i32 @func_25( i8 zeroext  %83 ) nounwind 		; <i32>:84 [#uses=1]
 	xor i32 %84, 1		; <i32>:85 [#uses=1]
-	load i32* @g_197, align 4		; <i32>:86 [#uses=1]
+	load i32, i32* @g_197, align 4		; <i32>:86 [#uses=1]
 	add i32 %86, 1		; <i32>:87 [#uses=1]
 	add i32 %87, %85		; <i32>:88 [#uses=1]
 	icmp ugt i32 %76, %88		; <i1>:89 [#uses=1]
@@ -163,22 +163,22 @@ bb223:		; preds = %bb222, %bb215
 bb241:		; preds = %bb223
 	store i16 -9, i16* @g_221, align 2
 	udiv i32 %p_52, 1538244727		; <i32>:90 [#uses=1]
-	load i32* @g_207, align 4		; <i32>:91 [#uses=1]
+	load i32, i32* @g_207, align 4		; <i32>:91 [#uses=1]
 	sub i32 %91, %90		; <i32>:92 [#uses=1]
-	load i32* @g_14, align 4		; <i32>:93 [#uses=1]
+	load i32, i32* @g_14, align 4		; <i32>:93 [#uses=1]
 	trunc i32 %93 to i16		; <i16>:94 [#uses=1]
 	trunc i32 %p_46 to i16		; <i16>:95 [#uses=2]
 	sub i16 %94, %95		; <i16>:96 [#uses=1]
-	load i32* @g_197, align 4		; <i32>:97 [#uses=1]
+	load i32, i32* @g_197, align 4		; <i32>:97 [#uses=1]
 	trunc i32 %97 to i16		; <i16>:98 [#uses=1]
 	tail call i32 @func_55( i32 -346178830, i16 zeroext  %98, i16 zeroext  %95 ) nounwind 		; <i32>:99 [#uses=0]
 	zext i16 %p_48 to i32		; <i32>:100 [#uses=1]
-	load i8* @g_247, align 1		; <i8>:101 [#uses=1]
+	load i8, i8* @g_247, align 1		; <i8>:101 [#uses=1]
 	zext i8 %101 to i32		; <i32>:102 [#uses=1]
 	sub i32 %100, %102		; <i32>:103 [#uses=1]
 	tail call i32 @func_55( i32 %103, i16 zeroext  -2972, i16 zeroext  %96 ) nounwind 		; <i32>:104 [#uses=0]
 	xor i32 %92, 2968		; <i32>:105 [#uses=1]
-	load i32* @g_197, align 4		; <i32>:106 [#uses=1]
+	load i32, i32* @g_197, align 4		; <i32>:106 [#uses=1]
 	icmp ugt i32 %105, %106		; <i1>:107 [#uses=1]
 	zext i1 %107 to i32		; <i32>:108 [#uses=1]
 	store i32 %108, i32* @g_33, align 4
@@ -195,12 +195,12 @@ bb248:		; preds = %bb284, %bb241
 
 bb272.thread:		; preds = %bb248
 	store i32 1, i32* @g_82
-	load i16* @g_267, align 2		; <i16>:111 [#uses=1]
+	load i16, i16* @g_267, align 2		; <i16>:111 [#uses=1]
 	icmp eq i16 %111, 0		; <i1>:112 [#uses=1]
 	br i1 %112, label %bb311.loopexit.split, label %bb268
 
 bb255.thread:		; preds = %bb248
-	load i32* @g_260, align 4		; <i32>:113 [#uses=1]
+	load i32, i32* @g_260, align 4		; <i32>:113 [#uses=1]
 	sub i32 %113, %p_52		; <i32>:114 [#uses=1]
 	and i32 %114, -20753		; <i32>:115 [#uses=1]
 	icmp ne i32 %115, 0		; <i1>:116 [#uses=1]
@@ -237,7 +237,7 @@ bb284:		; preds = %bb279, %bb276, %bb255.thread
 	%p_49_addr.0 = phi i32 [ %p_49_addr.1.reg2mem.0, %bb279 ], [ %p_49_addr.1.reg2mem.0, %bb276 ], [ 0, %bb255.thread ]		; <i32> [#uses=1]
 	%p_48_addr.1 = phi i16 [ %124, %bb279 ], [ %118, %bb276 ], [ %p_48_addr.2.reg2mem.0, %bb255.thread ]		; <i16> [#uses=1]
 	%p_45_addr.0 = phi i32 [ %p_45_addr.1.reg2mem.0, %bb279 ], [ %p_45_addr.1.reg2mem.0, %bb276 ], [ 8, %bb255.thread ]		; <i32> [#uses=3]
-	load i32* @g_43, align 4		; <i32>:125 [#uses=1]
+	load i32, i32* @g_43, align 4		; <i32>:125 [#uses=1]
 	trunc i32 %125 to i8		; <i8>:126 [#uses=1]
 	tail call i32 @func_116( i8 zeroext  %126 ) nounwind 		; <i32>:127 [#uses=0]
 	lshr i32 65255, %p_45_addr.0		; <i32>:128 [#uses=1]
@@ -245,7 +245,7 @@ bb284:		; preds = %bb279, %bb276, %bb255.thread
 	%.op = lshr i32 %128, 31		; <i32> [#uses=1]
 	%.op.op = xor i32 %.op, 1		; <i32> [#uses=1]
 	%.354..lobit.not = select i1 %129, i32 1, i32 %.op.op		; <i32> [#uses=1]
-	load i16* @g_39, align 2		; <i16>:130 [#uses=1]
+	load i16, i16* @g_39, align 2		; <i16>:130 [#uses=1]
 	zext i16 %130 to i32		; <i32>:131 [#uses=1]
 	icmp slt i32 %.354..lobit.not, %131		; <i1>:132 [#uses=1]
 	zext i1 %132 to i32		; <i32>:133 [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/pr3086.ll b/test/Transforms/LoopStrengthReduce/pr3086.ll
index 599633a11fee..187c14f900a5 100644
--- a/test/Transforms/LoopStrengthReduce/pr3086.ll
+++ b/test/Transforms/LoopStrengthReduce/pr3086.ll
@@ -10,18 +10,18 @@ entry:
 	br label %bb11
 
 bb5:		; preds = %bb9
-	%0 = load %struct.Lit** %storemerge, align 8		; <%struct.Lit*> [#uses=0]
+	%0 = load %struct.Lit*, %struct.Lit** %storemerge, align 8		; <%struct.Lit*> [#uses=0]
 	%indvar.next8 = add i64 %storemerge.rec, 1		; <i64> [#uses=1]
 	br label %bb9
 
 bb9:		; preds = %bb22, %bb5
 	%storemerge.rec = phi i64 [ %indvar.next8, %bb5 ], [ 0, %bb22 ]		; <i64> [#uses=2]
-	%storemerge = getelementptr %struct.Lit** null, i64 %storemerge.rec		; <%struct.Lit**> [#uses=2]
+	%storemerge = getelementptr %struct.Lit*, %struct.Lit** null, i64 %storemerge.rec		; <%struct.Lit**> [#uses=2]
 	%1 = icmp ugt %struct.Lit** null, %storemerge		; <i1> [#uses=1]
 	br i1 %1, label %bb5, label %bb22
 
 bb11:		; preds = %bb22, %entry
-	%2 = load %struct.Cls** null, align 8		; <%struct.Cls*> [#uses=0]
+	%2 = load %struct.Cls*, %struct.Cls** null, align 8		; <%struct.Cls*> [#uses=0]
 	br label %bb22
 
 bb22:		; preds = %bb11, %bb9
diff --git a/test/Transforms/LoopStrengthReduce/pr3399.ll b/test/Transforms/LoopStrengthReduce/pr3399.ll
index 26c5002fdecf..1037768f95c3 100644
--- a/test/Transforms/LoopStrengthReduce/pr3399.ll
+++ b/test/Transforms/LoopStrengthReduce/pr3399.ll
@@ -13,7 +13,7 @@ bb:		; preds = %bb5, %bb5.thread
 
 bb1:		; preds = %bb
 	%l_2.0.reg2mem.0 = sub i32 0, %indvar		; <i32> [#uses=1]
-	%0 = load volatile i32* @g_53, align 4		; <i32> [#uses=1]
+	%0 = load volatile i32, i32* @g_53, align 4		; <i32> [#uses=1]
 	%1 = trunc i32 %l_2.0.reg2mem.0 to i16		; <i16> [#uses=1]
 	%2 = trunc i32 %0 to i16		; <i16> [#uses=1]
 	%3 = mul i16 %2, %1		; <i16> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/pr3571.ll b/test/Transforms/LoopStrengthReduce/pr3571.ll
index a23e4db49705..1615a818771d 100644
--- a/test/Transforms/LoopStrengthReduce/pr3571.ll
+++ b/test/Transforms/LoopStrengthReduce/pr3571.ll
@@ -12,7 +12,7 @@ bb.i:		; preds = %_ZNK11QModelIndex7isValidEv.exit.i
 
 _ZNK11QModelIndex7isValidEv.exit.i:		; preds = %bb.i, %entry
 	%result.0.i = phi i32 [ 0, %entry ], [ %indvar.next, %bb.i ]		; <i32> [#uses=2]
-	%0 = load i32** null, align 4		; <%struct.QAbstractItemDelegate*> [#uses=0]
+	%0 = load i32*, i32** null, align 4		; <%struct.QAbstractItemDelegate*> [#uses=0]
 	br i1 false, label %_ZN18qdesigner_internalL5levelEP18QAbstractItemModelRK11QModelIndex.exit, label %bb.i
 
 _ZN18qdesigner_internalL5levelEP18QAbstractItemModelRK11QModelIndex.exit:		; preds = %_ZNK11QModelIndex7isValidEv.exit.i
diff --git a/test/Transforms/LoopStrengthReduce/preserve-gep-loop-variant.ll b/test/Transforms/LoopStrengthReduce/preserve-gep-loop-variant.ll
index f90d03080056..2b9920c566b2 100644
--- a/test/Transforms/LoopStrengthReduce/preserve-gep-loop-variant.ll
+++ b/test/Transforms/LoopStrengthReduce/preserve-gep-loop-variant.ll
@@ -28,7 +28,7 @@ for.body:
   %conv3 = sext i8 %conv to i64
   %add = add nsw i64 %call, %storemerge1
   %add4 = add nsw i64 %add, %conv3
-  %arrayidx = getelementptr inbounds i8* %p, i64 %add4
+  %arrayidx = getelementptr inbounds i8, i8* %p, i64 %add4
   store i8 0, i8* %arrayidx
   %inc = add nsw i64 %storemerge1, 1
   br label %for.cond
diff --git a/test/Transforms/LoopStrengthReduce/quadradic-exit-value.ll b/test/Transforms/LoopStrengthReduce/quadradic-exit-value.ll
index 65aa61fb937e..483becc0e7b8 100644
--- a/test/Transforms/LoopStrengthReduce/quadradic-exit-value.ll
+++ b/test/Transforms/LoopStrengthReduce/quadradic-exit-value.ll
@@ -1,5 +1,8 @@
 ; RUN: opt < %s -analyze -iv-users | FileCheck %s
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 ; The value of %r is dependent on a polynomial iteration expression.
 ;
 ; CHECK-LABEL: IV Users for loop %foo.loop
diff --git a/test/Transforms/LoopStrengthReduce/related_indvars.ll b/test/Transforms/LoopStrengthReduce/related_indvars.ll
index 12942bf10a5a..fbe8ffde3547 100644
--- a/test/Transforms/LoopStrengthReduce/related_indvars.ll
+++ b/test/Transforms/LoopStrengthReduce/related_indvars.ll
@@ -14,9 +14,9 @@ entry:
 no_exit:		; preds = %no_exit, %entry
 	%indvar = phi i32 [ 0, %entry ], [ %indvar.next, %no_exit ]		; <i32> [#uses=2]
 	%D_addr.0.0.rec = bitcast i32 %indvar to i32		; <i32> [#uses=2]
-	%D_addr.0.0 = getelementptr double* %D, i32 %D_addr.0.0.rec		; <double*> [#uses=1]
+	%D_addr.0.0 = getelementptr double, double* %D, i32 %D_addr.0.0.rec		; <double*> [#uses=1]
 	%inc.rec = add i32 %D_addr.0.0.rec, 1		; <i32> [#uses=1]
-	%inc = getelementptr double* %D, i32 %inc.rec		; <double*> [#uses=1]
+	%inc = getelementptr double, double* %D, i32 %inc.rec		; <double*> [#uses=1]
 	store double %F, double* %D_addr.0.0
 	%tmp.2 = icmp eq double* %inc, %E		; <i1> [#uses=1]
 	%indvar.next = add i32 %indvar, 1		; <i32> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/remove_indvar.ll b/test/Transforms/LoopStrengthReduce/remove_indvar.ll
index bb395324d4e4..3b92c25389b0 100644
--- a/test/Transforms/LoopStrengthReduce/remove_indvar.ll
+++ b/test/Transforms/LoopStrengthReduce/remove_indvar.ll
@@ -1,6 +1,9 @@
 ; Check that this test makes INDVAR and related stuff dead.
 ; RUN: opt < %s -loop-reduce -S | not grep INDVAR
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 declare i1 @pred()
 
 define void @test(i32* %P) {
@@ -9,7 +12,7 @@ define void @test(i32* %P) {
 Loop:		; preds = %Loop, %0
         %i = phi i32 [ 0, %0 ], [ %i.next, %Loop ]
 	%INDVAR = phi i32 [ 0, %0 ], [ %INDVAR2, %Loop ]		; <i32> [#uses=2]
-	%STRRED = getelementptr i32* %P, i32 %INDVAR		; <i32*> [#uses=1]
+	%STRRED = getelementptr i32, i32* %P, i32 %INDVAR		; <i32*> [#uses=1]
 	store i32 0, i32* %STRRED
 	%INDVAR2 = add i32 %INDVAR, 1		; <i32> [#uses=1]
         %i.next = add i32 %i, 1
diff --git a/test/Transforms/LoopStrengthReduce/scaling_factor_cost_crash.ll b/test/Transforms/LoopStrengthReduce/scaling_factor_cost_crash.ll
index a652a7661e23..27212d7b97b0 100644
--- a/test/Transforms/LoopStrengthReduce/scaling_factor_cost_crash.ll
+++ b/test/Transforms/LoopStrengthReduce/scaling_factor_cost_crash.ll
@@ -55,7 +55,7 @@ define void @scalingFactorCrash() {
 ; <label>:16                                      ; preds = %16, %15
   %17 = phi i32 [ %21, %16 ], [ undef, %15 ]
   %18 = sub i32 %17, 1623127498
-  %19 = getelementptr inbounds i32* undef, i32 %18
+  %19 = getelementptr inbounds i32, i32* undef, i32 %18
   store i32 undef, i32* %19, align 4
   %20 = add i32 %17, 1623127499
   %21 = add i32 %20, -1623127498
diff --git a/test/Transforms/LoopStrengthReduce/share_code_in_preheader.ll b/test/Transforms/LoopStrengthReduce/share_code_in_preheader.ll
index 412a716bc431..1035ce19ff10 100644
--- a/test/Transforms/LoopStrengthReduce/share_code_in_preheader.ll
+++ b/test/Transforms/LoopStrengthReduce/share_code_in_preheader.ll
@@ -12,10 +12,10 @@ no_exit:		; preds = %no_exit, %entry
 	%indvar = bitcast i32 %indvar.ui to i32		; <i32> [#uses=1]
 	%N_addr.0.0 = sub i32 %N.s, %indvar		; <i32> [#uses=1]
 	%tmp.8 = add i32 %N_addr.0.0, %tmp.6		; <i32> [#uses=2]
-	%tmp.9 = getelementptr i8* %A, i32 %tmp.8		; <i8*> [#uses=1]
-	%tmp.10 = load i8* %tmp.9		; <i8> [#uses=1]
-	%tmp.17 = getelementptr i8* %B, i32 %tmp.8		; <i8*> [#uses=1]
-	%tmp.18 = load i8* %tmp.17		; <i8> [#uses=1]
+	%tmp.9 = getelementptr i8, i8* %A, i32 %tmp.8		; <i8*> [#uses=1]
+	%tmp.10 = load i8, i8* %tmp.9		; <i8> [#uses=1]
+	%tmp.17 = getelementptr i8, i8* %B, i32 %tmp.8		; <i8*> [#uses=1]
+	%tmp.18 = load i8, i8* %tmp.17		; <i8> [#uses=1]
 	%tmp.19 = sub i8 %tmp.10, %tmp.18		; <i8> [#uses=1]
 	%tmp.21 = add i8 %tmp.19, %Sum.0.0		; <i8> [#uses=2]
 	%indvar.next = add i32 %indvar.ui, 1		; <i32> [#uses=2]
diff --git a/test/Transforms/LoopStrengthReduce/uglygep-address-space.ll b/test/Transforms/LoopStrengthReduce/uglygep-address-space.ll
index 2c65261f57f5..a81e314bad86 100644
--- a/test/Transforms/LoopStrengthReduce/uglygep-address-space.ll
+++ b/test/Transforms/LoopStrengthReduce/uglygep-address-space.ll
@@ -27,7 +27,7 @@ bb3:                                              ; preds = %bb2, %bb
 ; CHECK: bb10:
 ; CHECK-NEXT: %t7 = icmp eq i16 %t4, 0
 ; Host %t2 computation outside the loop.
-; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8 addrspace(1)* undef, i16 %t4
+; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8, i8 addrspace(1)* undef, i16 %t4
 ; CHECK-NEXT: br label %bb14
 bb10:                                             ; preds = %bb9
   %t7 = icmp eq i16 %t4, 0                    ; <i1> [#uses=1]
@@ -36,20 +36,20 @@ bb10:                                             ; preds = %bb9
 
 ; CHECK: bb14:
 ; CHECK-NEXT: store i8 undef, i8 addrspace(1)* [[SCEVGEP]]
-; CHECK-NEXT: %t6 = load float addrspace(1)* addrspace(1)* undef
+; CHECK-NEXT: %t6 = load float addrspace(1)*, float addrspace(1)* addrspace(1)* undef
 ; Fold %t3's add within the address.
-; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float addrspace(1)* %t6, i16 4
+; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float, float addrspace(1)* %t6, i16 4
 ; CHECK-NEXT: [[SCEVGEP2:%[^ ]+]] = bitcast float addrspace(1)* [[SCEVGEP1]] to i8 addrspace(1)*
 ; Use the induction variable (%t4) to access the right element
-; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8 addrspace(1)* [[SCEVGEP2]], i16 %t4
+; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8, i8 addrspace(1)* [[SCEVGEP2]], i16 %t4
 ; CHECK-NEXT: store i8 undef, i8 addrspace(1)* [[ADDRESS]]
 ; CHECK-NEXT: br label %bb14
 bb14:                                             ; preds = %bb14, %bb10
-  %t2 = getelementptr inbounds i8 addrspace(1)* undef, i16 %t4 ; <i8*> [#uses=1]
+  %t2 = getelementptr inbounds i8, i8 addrspace(1)* undef, i16 %t4 ; <i8*> [#uses=1]
   store i8 undef, i8 addrspace(1)* %t2
-  %t6 = load float addrspace(1)* addrspace(1)* undef
+  %t6 = load float addrspace(1)*, float addrspace(1)* addrspace(1)* undef
   %t8 = bitcast float addrspace(1)* %t6 to i8 addrspace(1)*              ; <i8*> [#uses=1]
-  %t9 = getelementptr inbounds i8 addrspace(1)* %t8, i16 %t3 ; <i8*> [#uses=1]
+  %t9 = getelementptr inbounds i8, i8 addrspace(1)* %t8, i16 %t3 ; <i8*> [#uses=1]
   store i8 undef, i8 addrspace(1)* %t9
   br label %bb14
 }
diff --git a/test/Transforms/LoopStrengthReduce/uglygep.ll b/test/Transforms/LoopStrengthReduce/uglygep.ll
index 4562d29a0a20..430127b3cd5d 100644
--- a/test/Transforms/LoopStrengthReduce/uglygep.ll
+++ b/test/Transforms/LoopStrengthReduce/uglygep.ll
@@ -24,7 +24,7 @@ bb3:                                              ; preds = %bb2, %bb
 ; CHECK: bb10:
 ; CHECK-NEXT: %t7 = icmp eq i64 %t4, 0
 ; Host %t2 computation outside the loop.
-; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8* undef, i64 %t4
+; CHECK-NEXT: [[SCEVGEP:%[^ ]+]] = getelementptr i8, i8* undef, i64 %t4
 ; CHECK-NEXT: br label %bb14
 bb10:                                             ; preds = %bb9
   %t7 = icmp eq i64 %t4, 0                    ; <i1> [#uses=1]
@@ -33,20 +33,20 @@ bb10:                                             ; preds = %bb9
 
 ; CHECK: bb14:
 ; CHECK-NEXT: store i8 undef, i8* [[SCEVGEP]]
-; CHECK-NEXT: %t6 = load float** undef
+; CHECK-NEXT: %t6 = load float*, float** undef
 ; Fold %t3's add within the address.
-; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float* %t6, i64 4
+; CHECK-NEXT: [[SCEVGEP1:%[^ ]+]] = getelementptr float, float* %t6, i64 4
 ; CHECK-NEXT: [[SCEVGEP2:%[^ ]+]] = bitcast float* [[SCEVGEP1]] to i8*
 ; Use the induction variable (%t4) to access the right element
-; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8* [[SCEVGEP2]], i64 %t4
+; CHECK-NEXT: [[ADDRESS:%[^ ]+]] = getelementptr i8, i8* [[SCEVGEP2]], i64 %t4
 ; CHECK-NEXT: store i8 undef, i8* [[ADDRESS]]
 ; CHECK-NEXT: br label %bb14
 bb14:                                             ; preds = %bb14, %bb10
-  %t2 = getelementptr inbounds i8* undef, i64 %t4 ; <i8*> [#uses=1]
+  %t2 = getelementptr inbounds i8, i8* undef, i64 %t4 ; <i8*> [#uses=1]
   store i8 undef, i8* %t2
-  %t6 = load float** undef
+  %t6 = load float*, float** undef
   %t8 = bitcast float* %t6 to i8*              ; <i8*> [#uses=1]
-  %t9 = getelementptr inbounds i8* %t8, i64 %t3 ; <i8*> [#uses=1]
+  %t9 = getelementptr inbounds i8, i8* %t8, i64 %t3 ; <i8*> [#uses=1]
   store i8 undef, i8* %t9
   br label %bb14
 }
@@ -59,7 +59,7 @@ bb:
 ; CHECK: loop0:
 ; Induction variable is initialized to -2.
 ; CHECK-NEXT: [[PHIIV:%[^ ]+]] = phi i32 [ [[IVNEXT:%[^ ]+]], %loop0 ], [ -2, %bb ]
-; CHECK-NEXT: [[IVNEXT]] = add i32 [[PHIIV]], 1
+; CHECK-NEXT: [[IVNEXT]] = add nuw nsw i32 [[PHIIV]], 1
 ; CHECK-NEXT: br i1 false, label %loop0, label %bb0
 loop0:                                            ; preds = %loop0, %bb
   %i0 = phi i32 [ %i0.next, %loop0 ], [ 0, %bb ]  ; <i32> [#uses=2]
diff --git a/test/Transforms/LoopStrengthReduce/use_postinc_value_outside_loop.ll b/test/Transforms/LoopStrengthReduce/use_postinc_value_outside_loop.ll
index 005e4c6b22c1..a673768d770e 100644
--- a/test/Transforms/LoopStrengthReduce/use_postinc_value_outside_loop.ll
+++ b/test/Transforms/LoopStrengthReduce/use_postinc_value_outside_loop.ll
@@ -16,7 +16,7 @@ no_exit.2:		; preds = %no_exit.2, %then.0
 	%indvar630.ui = phi i32 [ 0, %then.0 ], [ %indvar.next631, %no_exit.2 ]		; <i32> [#uses=3]
 	%indvar630 = bitcast i32 %indvar630.ui to i32		; <i32> [#uses=2]
 	%gep.upgrd.1 = zext i32 %indvar630.ui to i64		; <i64> [#uses=1]
-	%tmp.38 = getelementptr [700 x i32]* %nbeaux_.0__558, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
+	%tmp.38 = getelementptr [700 x i32], [700 x i32]* %nbeaux_.0__558, i32 0, i64 %gep.upgrd.1		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp.38
 	%inc.2 = add i32 %indvar630, 2		; <i32> [#uses=1]
 	%tmp.34 = call i1 @pred( i32 %indvar630 )		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/var_stride_used_by_compare.ll b/test/Transforms/LoopStrengthReduce/var_stride_used_by_compare.ll
index 3405b26f1f38..2dd14a0e5986 100644
--- a/test/Transforms/LoopStrengthReduce/var_stride_used_by_compare.ll
+++ b/test/Transforms/LoopStrengthReduce/var_stride_used_by_compare.ll
@@ -28,7 +28,7 @@ bb:		; preds = %bb, %bb.preheader
 	%tmp.16 = add i32 %tmp.15, %tmp.		; <i32> [#uses=2]
 	%k_addr.0.0 = bitcast i32 %tmp.16 to i32		; <i32> [#uses=1]
 	%gep.upgrd.1 = zext i32 %tmp.16 to i64		; <i64> [#uses=1]
-	%tmp = getelementptr [8193 x i8]* @flags2, i32 0, i64 %gep.upgrd.1		; <i8*> [#uses=1]
+	%tmp = getelementptr [8193 x i8], [8193 x i8]* @flags2, i32 0, i64 %gep.upgrd.1		; <i8*> [#uses=1]
 	store i8 0, i8* %tmp
 	%k_addr.0 = add i32 %k_addr.0.0, %i.s		; <i32> [#uses=1]
 	%tmp.upgrd.2 = icmp sgt i32 %k_addr.0, 8192		; <i1> [#uses=1]
diff --git a/test/Transforms/LoopStrengthReduce/variable_stride.ll b/test/Transforms/LoopStrengthReduce/variable_stride.ll
index 7c0f053e4c34..f82b2fc29ac5 100644
--- a/test/Transforms/LoopStrengthReduce/variable_stride.ll
+++ b/test/Transforms/LoopStrengthReduce/variable_stride.ll
@@ -1,6 +1,9 @@
 ; Check that variable strides are reduced to adds instead of multiplies.
 ; RUN: opt < %s -loop-reduce -S | not grep mul
 
+; Provide legal integer types.
+target datalayout = "n8:16:32:64"
+
 declare i1 @pred(i32)
 
 define void @test([10000 x i32]* %P, i32 %STRIDE) {
diff --git a/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll b/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
index d4c8402bd2c6..95e9dde04e6f 100644
--- a/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
+++ b/test/Transforms/LoopUnroll/2007-05-05-UnrollMiscomp.ll
@@ -9,11 +9,11 @@ target triple = "i686-apple-darwin8"
 define void @Foo(%struct.__mpz_struct* %base) {
 entry:
         %want = alloca [1 x %struct.__mpz_struct], align 16             ; <[1 x %struct.__mpz_struct]*> [#uses=4]
-        %want1 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
+        %want1 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
         call void @__gmpz_init( %struct.__mpz_struct* %want1 )
-        %want27 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0         ; <%struct.__mpz_struct*> [#uses=1]
-        %want3 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
-        %want2 = getelementptr [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=2]
+        %want27 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0         ; <%struct.__mpz_struct*> [#uses=1]
+        %want3 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=1]
+        %want2 = getelementptr [1 x %struct.__mpz_struct], [1 x %struct.__mpz_struct]* %want, i32 0, i32 0          ; <%struct.__mpz_struct*> [#uses=2]
         br label %bb
 
 bb:             ; preds = %bb, %entry
diff --git a/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll b/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
index bf6d6d5989c9..a87b16a28b7c 100644
--- a/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
+++ b/test/Transforms/LoopUnroll/2011-08-08-PhiUpdate.ll
@@ -46,8 +46,8 @@ bb.nph:                                           ; preds = %entry
 bb:                                               ; preds = %bb.nph, %bb1
   %indvar = phi i64 [ 0, %bb.nph ], [ %indvar.next, %bb1 ] ; <i64> [#uses=2]
   %s.01 = phi i32 [ 0, %bb.nph ], [ %2, %bb1 ]    ; <i32> [#uses=1]
-  %scevgep = getelementptr i32* %p, i64 %indvar   ; <i32*> [#uses=1]
-  %1 = load i32* %scevgep, align 1                ; <i32> [#uses=1]
+  %scevgep = getelementptr i32, i32* %p, i64 %indvar   ; <i32*> [#uses=1]
+  %1 = load i32, i32* %scevgep, align 1                ; <i32> [#uses=1]
   %2 = add nsw i32 %1, %s.01                      ; <i32> [#uses=2]
   br label %bb1
 
@@ -84,7 +84,7 @@ do.body:                                          ; preds = %do.cond, %if.end
   br i1 %cond2, label %exit, label %do.cond
 
 exit:                  ; preds = %do.body
-  %tmp7.i = load i32* undef, align 8
+  %tmp7.i = load i32, i32* undef, align 8
   br i1 undef, label %do.cond, label %land.lhs.true
 
 land.lhs.true:                                    ; preds = %exit
diff --git a/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll b/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
index a43a4ffc557c..0b4840956501 100644
--- a/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
+++ b/test/Transforms/LoopUnroll/2011-08-09-IVSimplify.ll
@@ -12,11 +12,11 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; CHECK: while.body:
 ; CHECK-NOT: while.body.1:
 ; CHECK: %shr.1 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.1 = getelementptr inbounds i32* %bitmap, i32 %shr.1
+; CHECK: %arrayidx.1 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.1
 ; CHECK: %shr.2 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.2 = getelementptr inbounds i32* %bitmap, i32 %shr.2
+; CHECK: %arrayidx.2 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.2
 ; CHECK: %shr.3 = lshr i32 %bit_addr.addr.01, 5
-; CHECK: %arrayidx.3 = getelementptr inbounds i32* %bitmap, i32 %shr.3
+; CHECK: %arrayidx.3 = getelementptr inbounds i32, i32* %bitmap, i32 %shr.3
 define void @FlipBit(i32* nocapture %bitmap, i32 %bit_addr, i32 %nbits) nounwind {
 entry:
   br label %while.body
@@ -28,8 +28,8 @@ while.body:
   %shr = lshr i32 %bit_addr.addr.01, 5
   %rem = and i32 %bit_addr.addr.01, 31
   %shl = shl i32 1, %rem
-  %arrayidx = getelementptr inbounds i32* %bitmap, i32 %shr
-  %tmp6 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %bitmap, i32 %shr
+  %tmp6 = load i32, i32* %arrayidx, align 4
   %xor = xor i32 %tmp6, %shl
   store i32 %xor, i32* %arrayidx, align 4
   %inc = add i32 %bit_addr.addr.01, 1
diff --git a/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll b/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
index 617d4dbe8fdd..5f9eec72ba50 100644
--- a/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
+++ b/test/Transforms/LoopUnroll/2011-10-01-NoopTrunc.ll
@@ -21,8 +21,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.02 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i8* %arr, i64 %indvars.iv
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %arr, i64 %indvars.iv
+  %0 = load i8, i8* %arrayidx, align 1
   %conv = sext i8 %0 to i32
   %add = add nsw i32 %conv, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopUnroll/AArch64/lit.local.cfg b/test/Transforms/LoopUnroll/AArch64/lit.local.cfg
new file mode 100644
index 000000000000..cec29af5bbe4
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/lit.local.cfg
@@ -0,0 +1,3 @@
+if not 'AArch64' in config.root.targets:
+    config.unsupported = True
+
diff --git a/test/Transforms/LoopUnroll/AArch64/partial.ll b/test/Transforms/LoopUnroll/AArch64/partial.ll
new file mode 100644
index 000000000000..8a1ea80c9d5b
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/partial.ll
@@ -0,0 +1,76 @@
+; RUN: opt < %s -S -loop-unroll -mtriple aarch64 -mcpu=cortex-a57 | FileCheck %s
+
+; Partial unroll 8 times for this loop.
+define void @unroll1() nounwind {
+entry:
+  br label %loop
+
+loop:
+  %iv = phi i32 [ 0, %entry ], [ %inc, %loop ]
+  %inc = add i32 %iv, 1
+  %exitcnd = icmp uge i32 %inc, 1024
+  br i1 %exitcnd, label %exit, label %loop
+
+exit:
+  ret void
+}
+
+; CHECK:      add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: icmp
+
+; Partial unroll 16 times for this loop.
+define void @unroll2() nounwind {
+entry:
+  br label %loop1
+
+loop1:
+  %iv1 = phi i32 [ 0, %entry ], [ %inc1, %loop1.latch ]
+  br label %loop2.header
+
+loop2.header:
+  br label %loop2
+
+loop2:
+  %iv2 = phi i32 [ 0, %loop2.header ], [ %inc2, %loop2 ]
+  %inc2 = add i32 %iv2, 1
+  %exitcnd2 = icmp uge i32 %inc2, 1024
+  br i1 %exitcnd2, label %exit2, label %loop2
+
+exit2:
+  br label %loop1.latch
+
+loop1.latch:
+  %inc1 = add i32 %iv1, 1
+  %exitcnd1 = icmp uge i32 %inc1, 1024
+  br i1 %exitcnd2, label %exit, label %loop1
+
+exit:
+  ret void
+}
+
+
+
+; CHECK:      add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: add
+; CHECK-NEXT: icmp
diff --git a/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll b/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll
new file mode 100644
index 000000000000..d3dc081fa6f8
--- /dev/null
+++ b/test/Transforms/LoopUnroll/AArch64/runtime-loop.ll
@@ -0,0 +1,33 @@
+; RUN: opt < %s -S -loop-unroll -mtriple aarch64 -mcpu=cortex-a57 | FileCheck %s
+
+; Tests for unrolling loops with run-time trip counts
+
+; CHECK:  %xtraiter = and i32 %n
+; CHECK:  %lcmp.mod = icmp ne i32 %xtraiter, 0
+; CHECK:  br i1 %lcmp.mod, label %for.body.prol, label %for.body.preheader.split
+
+; CHECK:  for.body.prol:
+; CHECK:  for.body:
+
+define i32 @test(i32* nocapture %a, i32 %n) nounwind uwtable readonly {
+entry:
+  %cmp1 = icmp eq i32 %n, 0
+  br i1 %cmp1, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %add = add nsw i32 %0, %sum.02
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  %sum.0.lcssa = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  ret i32 %sum.0.lcssa
+}
+
+
diff --git a/test/Transforms/LoopUnroll/PowerPC/a2-high-cost-trip-count-computation.ll b/test/Transforms/LoopUnroll/PowerPC/a2-high-cost-trip-count-computation.ll
new file mode 100644
index 000000000000..bd5f68f9f200
--- /dev/null
+++ b/test/Transforms/LoopUnroll/PowerPC/a2-high-cost-trip-count-computation.ll
@@ -0,0 +1,27 @@
+; RUN: opt < %s -S -mtriple=powerpc64-unknown-linux-gnu -mcpu=a2 -loop-unroll | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
+;; Check that we do emit expensive instructions to compute trip
+;; counts when unrolling loops on the a2 (because we unroll a lot).
+
+define i32 @test(i64 %v12, i8* %array, i64* %loc) {
+; CHECK-LABEL: @test(
+; CHECK: udiv
+entry:
+  %step = load i64, i64* %loc, !range !0
+  br label %loop
+
+loop:                                           ; preds = %entry, %loop
+  %k.015 = phi i64 [ %v15, %loop ], [ %v12, %entry ]
+  %v14 = getelementptr inbounds i8, i8* %array, i64 %k.015
+  store i8 0, i8* %v14
+  %v15 = add nuw nsw i64 %k.015, %step
+  %v16 = icmp slt i64 %v15, 8193
+  br i1 %v16, label %loop, label %loopexit
+
+loopexit:                             ; preds = %loop
+  ret i32 0
+}
+
+!0 = !{i64 1, i64 100}
diff --git a/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll b/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
index aae79cbac789..e9aa1acd5fec 100644
--- a/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
+++ b/test/Transforms/LoopUnroll/PowerPC/a2-unrolling.ll
@@ -27,8 +27,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/X86/partial.ll b/test/Transforms/LoopUnroll/X86/partial.ll
index a2b04c7d85f8..4566f792deb4 100644
--- a/test/Transforms/LoopUnroll/X86/partial.ll
+++ b/test/Transforms/LoopUnroll/X86/partial.ll
@@ -9,20 +9,20 @@ entry:
 
 vector.body:                                      ; preds = %vector.body, %entry
   %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
-  %0 = getelementptr inbounds double* %b, i64 %index
+  %0 = getelementptr inbounds double, double* %b, i64 %index
   %1 = bitcast double* %0 to <2 x double>*
-  %wide.load = load <2 x double>* %1, align 8
+  %wide.load = load <2 x double>, <2 x double>* %1, align 8
   %.sum9 = or i64 %index, 2
-  %2 = getelementptr double* %b, i64 %.sum9
+  %2 = getelementptr double, double* %b, i64 %.sum9
   %3 = bitcast double* %2 to <2 x double>*
-  %wide.load8 = load <2 x double>* %3, align 8
+  %wide.load8 = load <2 x double>, <2 x double>* %3, align 8
   %4 = fadd <2 x double> %wide.load, <double 1.000000e+00, double 1.000000e+00>
   %5 = fadd <2 x double> %wide.load8, <double 1.000000e+00, double 1.000000e+00>
-  %6 = getelementptr inbounds double* %a, i64 %index
+  %6 = getelementptr inbounds double, double* %a, i64 %index
   %7 = bitcast double* %6 to <2 x double>*
   store <2 x double> %4, <2 x double>* %7, align 8
   %.sum10 = or i64 %index, 2
-  %8 = getelementptr double* %a, i64 %.sum10
+  %8 = getelementptr double, double* %a, i64 %.sum10
   %9 = bitcast double* %8 to <2 x double>*
   store <2 x double> %5, <2 x double>* %9, align 8
   %index.next = add i64 %index, 4
@@ -45,12 +45,12 @@ entry:
 
 vector.body:                                      ; preds = %vector.body, %entry
   %index = phi i64 [ 0, %entry ], [ %index.next, %vector.body ]
-  %v0 = getelementptr inbounds double* %b, i64 %index
+  %v0 = getelementptr inbounds double, double* %b, i64 %index
   %v1 = bitcast double* %v0 to <2 x double>*
-  %wide.load = load <2 x double>* %v1, align 8
+  %wide.load = load <2 x double>, <2 x double>* %v1, align 8
   %v4 = fadd <2 x double> %wide.load, <double 1.000000e+00, double 1.000000e+00>
   %v5 = fmul <2 x double> %v4, <double 8.000000e+00, double 8.000000e+00>
-  %v6 = getelementptr inbounds double* %a, i64 %index
+  %v6 = getelementptr inbounds double, double* %a, i64 %index
   %v7 = bitcast double* %v6 to <2 x double>*
   store <2 x double> %v5, <2 x double>* %v7, align 8
   %index.next = add i64 %index, 2
@@ -84,18 +84,18 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %reduction.026 = phi i16 [ %add14, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i16* %arr, i64 %indvars.iv
-  %0 = load i16* %arrayidx, align 2
+  %arrayidx = getelementptr inbounds i16, i16* %arr, i64 %indvars.iv
+  %0 = load i16, i16* %arrayidx, align 2
   %add = add i16 %0, %reduction.026
   %sext = mul i64 %indvars.iv, 12884901888
   %idxprom3 = ashr exact i64 %sext, 32
-  %arrayidx4 = getelementptr inbounds i16* %arr, i64 %idxprom3
-  %1 = load i16* %arrayidx4, align 2
+  %arrayidx4 = getelementptr inbounds i16, i16* %arr, i64 %idxprom3
+  %1 = load i16, i16* %arrayidx4, align 2
   %add7 = add i16 %add, %1
   %sext28 = mul i64 %indvars.iv, 21474836480
   %idxprom10 = ashr exact i64 %sext28, 32
-  %arrayidx11 = getelementptr inbounds i16* %arr, i64 %idxprom10
-  %2 = load i16* %arrayidx11, align 2
+  %arrayidx11 = getelementptr inbounds i16, i16* %arr, i64 %idxprom10
+  %2 = load i16, i16* %arrayidx11, align 2
   %add14 = add i16 %add7, %2
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/ephemeral.ll b/test/Transforms/LoopUnroll/ephemeral.ll
index 9d4061390537..d16eba7a9644 100644
--- a/test/Transforms/LoopUnroll/ephemeral.ll
+++ b/test/Transforms/LoopUnroll/ephemeral.ll
@@ -12,8 +12,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.01 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
 
   ; This loop will be completely unrolled, even with these extra instructions,
   ; but only because they're ephemeral (and, thus, free).
diff --git a/test/Transforms/LoopUnroll/full-unroll-bad-geps.ll b/test/Transforms/LoopUnroll/full-unroll-bad-geps.ll
new file mode 100644
index 000000000000..4c99bc73880b
--- /dev/null
+++ b/test/Transforms/LoopUnroll/full-unroll-bad-geps.ll
@@ -0,0 +1,34 @@
+; Check that we don't crash on corner cases.
+; RUN: opt < %s -S -loop-unroll -unroll-max-iteration-count-to-analyze=1000 -unroll-absolute-threshold=10  -unroll-threshold=10  -unroll-percent-of-optimized-for-complete-unroll=20 -o /dev/null
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+define void @foo1() {
+entry:
+  br label %for.body
+
+for.body:
+  %phi = phi i64 [ 0, %entry ], [ %inc, %for.body ]
+  %idx = zext i32 undef to i64
+  %add.ptr = getelementptr inbounds i64, i64* null, i64 %idx
+  %inc = add nuw nsw i64 %phi, 1
+  %cmp = icmp ult i64 %inc, 999
+  br i1 %cmp, label %for.body, label %for.exit
+
+for.exit:
+  ret void
+}
+
+define void @foo2() {
+entry:
+  br label %for.body
+
+for.body:
+  %phi = phi i64 [ 0, %entry ], [ %inc, %for.body ]
+  %x = getelementptr i32, <4 x i32*> undef, <4 x i32> <i32 1, i32 1, i32 1, i32 1>
+  %inc = add nuw nsw i64 %phi, 1
+  %cmp = icmp ult i64 %inc, 999
+  br i1 %cmp, label %for.body, label %for.exit
+
+for.exit:
+  ret void
+}
diff --git a/test/Transforms/LoopUnroll/full-unroll-heuristics.ll b/test/Transforms/LoopUnroll/full-unroll-heuristics.ll
new file mode 100644
index 000000000000..2dab2fbf2e49
--- /dev/null
+++ b/test/Transforms/LoopUnroll/full-unroll-heuristics.ll
@@ -0,0 +1,62 @@
+; In this test we check how heuristics for complete unrolling work. We have
+; three knobs:
+;  1) -unroll-threshold
+;  2) -unroll-absolute-threshold and
+;  3) -unroll-percent-of-optimized-for-complete-unroll
+;
+; They control loop-unrolling according to the following rules:
+;  * If size of unrolled loop exceeds the absoulte threshold, we don't unroll
+;    this loop under any circumstances.
+;  * If size of unrolled loop is below the '-unroll-threshold', then we'll
+;    consider this loop as a very small one, and completely unroll it.
+;  * If a loop size is between these two tresholds, we only do complete unroll
+;    it if estimated number of potentially optimized instructions is high (we
+;    specify the minimal percent of such instructions).
+
+; In this particular test-case, complete unrolling will allow later
+; optimizations to remove ~55% of the instructions, the loop body size is 9,
+; and unrolled size is 65.
+
+; RUN: opt < %s -S -loop-unroll -unroll-max-iteration-count-to-analyze=1000 -unroll-absolute-threshold=10  -unroll-threshold=10  -unroll-percent-of-optimized-for-complete-unroll=20 | FileCheck %s -check-prefix=TEST1
+; RUN: opt < %s -S -loop-unroll -unroll-max-iteration-count-to-analyze=1000 -unroll-absolute-threshold=100 -unroll-threshold=10  -unroll-percent-of-optimized-for-complete-unroll=20 | FileCheck %s -check-prefix=TEST2
+; RUN: opt < %s -S -loop-unroll -unroll-max-iteration-count-to-analyze=1000 -unroll-absolute-threshold=100 -unroll-threshold=10  -unroll-percent-of-optimized-for-complete-unroll=80 | FileCheck %s -check-prefix=TEST3
+; RUN: opt < %s -S -loop-unroll -unroll-max-iteration-count-to-analyze=1000 -unroll-absolute-threshold=100 -unroll-threshold=100 -unroll-percent-of-optimized-for-complete-unroll=80 | FileCheck %s -check-prefix=TEST4
+
+; If the absolute threshold is too low, or if we can't optimize away requested
+; percent of instructions, we shouldn't unroll:
+; TEST1: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+; TEST3: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+
+; Otherwise, we should:
+; TEST2-NOT: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+
+; Also, we should unroll if the 'unroll-threshold' is big enough:
+; TEST4-NOT: %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+
+; And check that we don't crash when we're not allowed to do any analysis.
+; RUN: opt < %s -loop-unroll -unroll-max-iteration-count-to-analyze=0 -disable-output
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+@known_constant = internal unnamed_addr constant [9 x i32] [i32 0, i32 -1, i32 0, i32 -1, i32 5, i32 -1, i32 0, i32 -1, i32 0], align 16
+
+define i32 @foo(i32* noalias nocapture readonly %src) {
+entry:
+  br label %loop
+
+loop:                                                ; preds = %loop, %entry
+  %iv = phi i64 [ 0, %entry ], [ %inc, %loop ]
+  %r  = phi i32 [ 0, %entry ], [ %add, %loop ]
+  %arrayidx = getelementptr inbounds i32, i32* %src, i64 %iv
+  %src_element = load i32, i32* %arrayidx, align 4
+  %array_const_idx = getelementptr inbounds [9 x i32], [9 x i32]* @known_constant, i64 0, i64 %iv
+  %const_array_element = load i32, i32* %array_const_idx, align 4
+  %mul = mul nsw i32 %src_element, %const_array_element
+  %add = add nsw i32 %mul, %r
+  %inc = add nuw nsw i64 %iv, 1
+  %exitcond86.i = icmp eq i64 %inc, 9
+  br i1 %exitcond86.i, label %loop.end, label %loop
+
+loop.end:                                            ; preds = %loop
+  %r.lcssa = phi i32 [ %r, %loop ]
+  ret i32 %r.lcssa
+}
diff --git a/test/Transforms/LoopUnroll/high-cost-trip-count-computation.ll b/test/Transforms/LoopUnroll/high-cost-trip-count-computation.ll
new file mode 100644
index 000000000000..6b1943f13bac
--- /dev/null
+++ b/test/Transforms/LoopUnroll/high-cost-trip-count-computation.ll
@@ -0,0 +1,27 @@
+; RUN: opt -S -unroll-runtime -loop-unroll < %s | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
+;; Check that we don't emit expensive instructions to compute trip
+;; counts when unrolling loops.
+
+define i32 @test(i64 %v12, i8* %array, i64* %loc) {
+; CHECK-LABEL: @test(
+; CHECK-NOT: udiv
+entry:
+  %step = load i64, i64* %loc, !range !0
+  br label %loop
+
+loop:                                           ; preds = %entry, %loop
+  %k.015 = phi i64 [ %v15, %loop ], [ %v12, %entry ]
+  %v14 = getelementptr inbounds i8, i8* %array, i64 %k.015
+  store i8 0, i8* %v14
+  %v15 = add nuw nsw i64 %k.015, %step
+  %v16 = icmp slt i64 %v15, 8193
+  br i1 %v16, label %loop, label %loopexit
+
+loopexit:                             ; preds = %loop
+  ret i32 0
+}
+
+!0 = !{i64 1, i64 100}
diff --git a/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll b/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
index dcb5d1c28fb2..2101b63e026c 100644
--- a/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
+++ b/test/Transforms/LoopUnroll/ignore-annotation-intrinsic-cost.ll
@@ -18,7 +18,7 @@ for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   ; The real loop.
   %mul = mul nsw i32 %B, %C
-  %arrayidx = getelementptr inbounds i32* %A, i32 %i.01
+  %arrayidx = getelementptr inbounds i32, i32* %A, i32 %i.01
   store i32 %mul, i32* %arrayidx, align 4
   %inc = add nsw i32 %i.01, 1
   %exitcond = icmp ne i32 %inc, 4
diff --git a/test/Transforms/LoopUnroll/runtime-loop.ll b/test/Transforms/LoopUnroll/runtime-loop.ll
index 80571ecbd882..fea15b65d5a5 100644
--- a/test/Transforms/LoopUnroll/runtime-loop.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop.ll
@@ -1,5 +1,7 @@
 ; RUN: opt < %s -S -loop-unroll -unroll-runtime=true | FileCheck %s
 
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
 ; Tests for unrolling loops with run-time trip counts
 
 ; CHECK: %xtraiter = and i32 %n
@@ -20,8 +22,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -47,8 +49,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.01 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.01
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -96,8 +98,8 @@ for.body:                                         ; preds = %for.body, %entry
   %p.addr.05 = phi i16* [ %incdec.ptr, %for.body ], [ %p, %entry ]
   %len.addr.04 = phi i32 [ %sub, %for.body ], [ %len, %entry ]
   %res.03 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %incdec.ptr = getelementptr inbounds i16* %p.addr.05, i64 1
-  %0 = load i16* %p.addr.05, align 2
+  %incdec.ptr = getelementptr inbounds i16, i16* %p.addr.05, i64 1
+  %0 = load i16, i16* %p.addr.05, align 2
   %conv = zext i16 %0 to i32
   %add = add i32 %conv, %res.03
   %sub = add nsw i32 %len.addr.04, -2
@@ -113,6 +115,39 @@ for.end:                                          ; preds = %for.cond.for.end_cr
   ret i16 %res.0.lcssa
 }
 
+; Test run-time unrolling disable metadata.
+; CHECK: for.body:
+; CHECK-NOT: for.body.prol:
+
+define zeroext i16 @test2(i16* nocapture %p, i32 %len) nounwind uwtable readonly {
+entry:
+  %cmp2 = icmp eq i32 %len, 0
+  br i1 %cmp2, label %for.end, label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %p.addr.05 = phi i16* [ %incdec.ptr, %for.body ], [ %p, %entry ]
+  %len.addr.04 = phi i32 [ %sub, %for.body ], [ %len, %entry ]
+  %res.03 = phi i32 [ %add, %for.body ], [ 0, %entry ]
+  %incdec.ptr = getelementptr inbounds i16, i16* %p.addr.05, i64 1
+  %0 = load i16, i16* %p.addr.05, align 2
+  %conv = zext i16 %0 to i32
+  %add = add i32 %conv, %res.03
+  %sub = add nsw i32 %len.addr.04, -2
+  %cmp = icmp eq i32 %sub, 0
+  br i1 %cmp, label %for.cond.for.end_crit_edge, label %for.body, !llvm.loop !0
+
+for.cond.for.end_crit_edge:                       ; preds = %for.body
+  %phitmp = trunc i32 %add to i16
+  br label %for.end
+
+for.end:                                          ; preds = %for.cond.for.end_crit_edge, %entry
+  %res.0.lcssa = phi i16 [ %phitmp, %for.cond.for.end_crit_edge ], [ 0, %entry ]
+  ret i16 %res.0.lcssa
+}
+
+!0 = distinct !{!0, !1}
+!1 = !{!"llvm.loop.unroll.runtime.disable"}
+
 ; CHECK: !0 = distinct !{!0, !1}
 ; CHECK: !1 = !{!"llvm.loop.unroll.disable"}
 
diff --git a/test/Transforms/LoopUnroll/runtime-loop1.ll b/test/Transforms/LoopUnroll/runtime-loop1.ll
index 5ff75e33f7f8..7684e394290b 100644
--- a/test/Transforms/LoopUnroll/runtime-loop1.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop1.ll
@@ -16,8 +16,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop2.ll b/test/Transforms/LoopUnroll/runtime-loop2.ll
index 176362a34456..7c6bb9690551 100644
--- a/test/Transforms/LoopUnroll/runtime-loop2.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop2.ll
@@ -16,8 +16,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.02 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %sum.02
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop3.ll b/test/Transforms/LoopUnroll/runtime-loop3.ll
index aa928ccc60c1..fd13ebfa0b84 100644
--- a/test/Transforms/LoopUnroll/runtime-loop3.ll
+++ b/test/Transforms/LoopUnroll/runtime-loop3.ll
@@ -23,8 +23,8 @@ for.body3:                                        ; preds = %for.cond1.preheader
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body3 ], [ 0, %for.cond1.preheader ]
   %sum.19 = phi i32 [ %add4, %for.body3 ], [ %sum.012, %for.cond1.preheader ]
   %0 = add nsw i64 %indvars.iv, %indvars.iv16
-  %arrayidx = getelementptr inbounds i32* %a, i64 %0
-  %1 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %0
+  %1 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %1, %sum.19
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopUnroll/runtime-loop4.ll b/test/Transforms/LoopUnroll/runtime-loop4.ll
new file mode 100644
index 000000000000..5014c739e4c9
--- /dev/null
+++ b/test/Transforms/LoopUnroll/runtime-loop4.ll
@@ -0,0 +1,44 @@
+; RUN: opt < %s -S -O2 -unroll-runtime=true | FileCheck %s
+
+; Check runtime unrolling prologue can be promoted by LICM pass.
+
+; CHECK: entry:
+; CHECK: %xtraiter
+; CHECK: %lcmp.mod
+; CHECK: loop1:
+; CHECK: br i1 %lcmp.mod
+; CHECK: loop2.prol:
+
+define void @unroll(i32 %iter, i32* %addr1, i32* %addr2) nounwind {
+entry:
+  br label %loop1
+
+loop1:
+  %iv1 = phi i32 [ 0, %entry ], [ %inc1, %loop1.latch ]
+  %offset1 = getelementptr i32, i32* %addr1, i32 %iv1
+  store i32 %iv1, i32* %offset1, align 4
+  br label %loop2.header
+
+loop2.header:
+  %e = icmp uge i32 %iter, 1
+  br i1 %e, label %loop2, label %exit2
+
+loop2:
+  %iv2 = phi i32 [ 0, %loop2.header ], [ %inc2, %loop2 ]
+  %offset2 = getelementptr i32, i32* %addr2, i32 %iv2
+  store i32 %iv2, i32* %offset2, align 4
+  %inc2 = add i32 %iv2, 1
+  %exitcnd2 = icmp uge i32 %inc2, %iter
+  br i1 %exitcnd2, label %exit2, label %loop2
+
+exit2:
+  br label %loop1.latch
+
+loop1.latch:
+  %inc1 = add i32 %iv1, 1
+  %exitcnd1 = icmp uge i32 %inc1, 1024
+  br i1 %exitcnd1, label %exit, label %loop1
+
+exit:
+  ret void
+}
diff --git a/test/Transforms/LoopUnroll/runtime-loop5.ll b/test/Transforms/LoopUnroll/runtime-loop5.ll
new file mode 100644
index 000000000000..e8d51775ce18
--- /dev/null
+++ b/test/Transforms/LoopUnroll/runtime-loop5.ll
@@ -0,0 +1,45 @@
+; RUN: opt < %s -S -loop-unroll -unroll-runtime=true -unroll-count=16 | FileCheck --check-prefix=UNROLL-16 %s
+; RUN: opt < %s -S -loop-unroll -unroll-runtime=true -unroll-count=4 | FileCheck --check-prefix=UNROLL-4 %s
+
+; Given that the trip-count of this loop is a 3-bit value, we cannot
+; safely unroll it with a count of anything more than 8.
+
+define i3 @test(i3* %a, i3 %n) {
+; UNROLL-16-LABEL: @test(
+; UNROLL-4-LABEL: @test(
+entry:
+  %cmp1 = icmp eq i3 %n, 0
+  br i1 %cmp1, label %for.end, label %for.body
+
+; UNROLL-16-NOT: for.body.prol:
+; UNROLL-4: for.body.prol:
+
+for.body:                                         ; preds = %for.body, %entry
+; UNROLL-16-LABEL: for.body:
+; UNROLL-4-LABEL: for.body:
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %sum.02 = phi i3 [ %add, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds i3, i3* %a, i64 %indvars.iv
+
+; UNROLL-16-LABEL: for.body
+; UNROLL-16-LABEL: getelementptr
+; UNROLL-16-LABEL-NOT: getelementptr
+
+; UNROLL-4-LABEL: getelementptr
+; UNROLL-4-LABEL: getelementptr
+; UNROLL-4-LABEL: getelementptr
+; UNROLL-4-LABEL: getelementptr
+
+  %0 = load i3, i3* %arrayidx
+  %add = add nsw i3 %0, %sum.02
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i3
+  %exitcond = icmp eq i3 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+; UNROLL-16-LABEL: for.end
+; UNROLL-4-LABEL: for.end
+for.end:                                          ; preds = %for.body, %entry
+  %sum.0.lcssa = phi i3 [ 0, %entry ], [ %add, %for.body ]
+  ret i3 %sum.0.lcssa
+}
diff --git a/test/Transforms/LoopUnroll/scevunroll.ll b/test/Transforms/LoopUnroll/scevunroll.ll
index 20161d798e9e..a5c9a6efacf3 100644
--- a/test/Transforms/LoopUnroll/scevunroll.ll
+++ b/test/Transforms/LoopUnroll/scevunroll.ll
@@ -19,8 +19,8 @@ while.body:
   %iv = phi i64 [ 10, %entry ], [ %iv.next, %while.body ]
   %sum = phi i32 [ 0, %entry ], [ %sum.next, %while.body ]
   %iv.next = add i64 %iv, -1
-  %adr = getelementptr inbounds i32* %base, i64 %iv.next
-  %tmp = load i32* %adr, align 8
+  %adr = getelementptr inbounds i32, i32* %base, i64 %iv.next
+  %tmp = load i32, i32* %adr, align 8
   %sum.next = add i32 %sum, %tmp
   %iv.narrow = trunc i64 %iv.next to i32
   %cmp.i65 = icmp sgt i32 %iv.narrow, 0
@@ -46,8 +46,8 @@ entry:
 loop:
   %iv = phi i64 [ 0, %entry ], [ %inc, %tail ]
   %s = phi i64 [ 0, %entry ], [ %s.next, %tail ]
-  %adr = getelementptr i64* %base, i64 %iv
-  %val = load i64* %adr
+  %adr = getelementptr i64, i64* %base, i64 %iv
+  %val = load i64, i64* %adr
   %s.next = add i64 %s, %val
   %inc = add i64 %iv, 1
   %cmp = icmp ne i64 %inc, 4
@@ -67,8 +67,8 @@ exit2:
 ; SCEV properly unrolls multi-exit loops.
 ;
 ; CHECK-LABEL: @multiExit(
-; CHECK: getelementptr i32* %base, i32 10
-; CHECK-NEXT: load i32*
+; CHECK: getelementptr i32, i32* %base, i32 10
+; CHECK-NEXT: load i32, i32*
 ; CHECK: br i1 false, label %l2.10, label %exit1
 ; CHECK: l2.10:
 ; CHECK-NOT: br
@@ -81,8 +81,8 @@ l1:
   %iv2 = phi i32 [ 0, %entry ], [ %inc2, %l2 ]
   %inc1 = add i32 %iv1, 1
   %inc2 = add i32 %iv2, 1
-  %adr = getelementptr i32* %base, i32 %iv1
-  %val = load i32* %adr
+  %adr = getelementptr i32, i32* %base, i32 %iv1
+  %val = load i32, i32* %adr
   %cmp1 = icmp slt i32 %iv1, 5
   br i1 %cmp1, label %l2, label %exit1
 l2:
@@ -112,8 +112,8 @@ l1:
   %iv2 = phi i32 [ 0, %entry ], [ %inc2, %l3 ]
   %inc1 = add i32 %iv1, 1
   %inc2 = add i32 %iv2, 1
-  %adr = getelementptr i32* %base, i32 %iv1
-  %val = load i32* %adr
+  %adr = getelementptr i32, i32* %base, i32 %iv1
+  %val = load i32, i32* %adr
   %cmp1 = icmp slt i32 %iv1, 5
   br i1 %cmp1, label %l2, label %exit1
 l2:
diff --git a/test/Transforms/LoopUnroll/shifted-tripcount.ll b/test/Transforms/LoopUnroll/shifted-tripcount.ll
index a118a463172e..4c216983af51 100644
--- a/test/Transforms/LoopUnroll/shifted-tripcount.ll
+++ b/test/Transforms/LoopUnroll/shifted-tripcount.ll
@@ -13,11 +13,11 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.013 = phi i64 [ %tmp16, %for.body ], [ 0, %entry ] ; <i64> [#uses=2]
-  %arrayidx7 = getelementptr double* %p, i64 %i.013 ; <double*> [#uses=2]
+  %arrayidx7 = getelementptr double, double* %p, i64 %i.013 ; <double*> [#uses=2]
   %tmp16 = add i64 %i.013, 1                      ; <i64> [#uses=3]
-  %arrayidx = getelementptr double* %p, i64 %tmp16 ; <double*> [#uses=1]
-  %tmp4 = load double* %arrayidx                  ; <double> [#uses=1]
-  %tmp8 = load double* %arrayidx7                 ; <double> [#uses=1]
+  %arrayidx = getelementptr double, double* %p, i64 %tmp16 ; <double*> [#uses=1]
+  %tmp4 = load double, double* %arrayidx                  ; <double> [#uses=1]
+  %tmp8 = load double, double* %arrayidx7                 ; <double> [#uses=1]
   %mul9 = fmul double %tmp8, %tmp4                ; <double> [#uses=1]
   store double %mul9, double* %arrayidx7
   %exitcond = icmp eq i64 %tmp16, %mul10          ; <i1> [#uses=1]
diff --git a/test/Transforms/LoopUnroll/unroll-cleanup.ll b/test/Transforms/LoopUnroll/unroll-cleanup.ll
new file mode 100644
index 000000000000..1e42203876ec
--- /dev/null
+++ b/test/Transforms/LoopUnroll/unroll-cleanup.ll
@@ -0,0 +1,85 @@
+; PR23524
+; The test is to check redundency produced by loop unroll pass
+; should be cleaned up by later pass.
+; RUN: opt < %s -O2 -S | FileCheck %s
+
+; After loop unroll:
+;       %dec18 = add nsw i32 %dec18.in, -1
+;       ...
+;       %dec18.1 = add nsw i32 %dec18, -1
+; should be merged to:
+;       %dec18.1 = add nsw i32 %dec18.in, -2
+;
+; CHECK-LABEL: @_Z3fn1v(
+; CHECK: %dec18.1 = add nsw i32 %dec18.in, -2
+
+; ModuleID = '<stdin>'
+target triple = "x86_64-unknown-linux-gnu"
+
+@b = global i32 0, align 4
+@c = global i32 0, align 4
+
+; Function Attrs: nounwind uwtable
+define void @_Z3fn1v() #0 {
+entry:
+  %tmp = load i32, i32* @b, align 4
+  %tobool20 = icmp eq i32 %tmp, 0
+  br i1 %tobool20, label %for.end6, label %for.body.lr.ph
+
+for.body.lr.ph:                                   ; preds = %entry
+  br label %for.body
+
+for.cond1.for.cond.loopexit_crit_edge:            ; preds = %for.inc
+  %add.ptr.lcssa = phi i16* [ %add.ptr, %for.inc ]
+  %incdec.ptr.lcssa = phi i8* [ %incdec.ptr, %for.inc ]
+  br label %for.cond.loopexit
+
+for.cond.loopexit:                                ; preds = %for.body, %for.cond1.for.cond.loopexit_crit_edge
+  %r.1.lcssa = phi i16* [ %add.ptr.lcssa, %for.cond1.for.cond.loopexit_crit_edge ], [ %r.022, %for.body ]
+  %a.1.lcssa = phi i8* [ %incdec.ptr.lcssa, %for.cond1.for.cond.loopexit_crit_edge ], [ %a.021, %for.body ]
+  %tmp1 = load i32, i32* @b, align 4
+  %tobool = icmp eq i32 %tmp1, 0
+  br i1 %tobool, label %for.cond.for.end6_crit_edge, label %for.body
+
+for.body:                                         ; preds = %for.cond.loopexit, %for.body.lr.ph
+  %r.022 = phi i16* [ undef, %for.body.lr.ph ], [ %r.1.lcssa, %for.cond.loopexit ]
+  %a.021 = phi i8* [ undef, %for.body.lr.ph ], [ %a.1.lcssa, %for.cond.loopexit ]
+  %tmp2 = load i32, i32* @c, align 4
+  %tobool215 = icmp eq i32 %tmp2, 0
+  br i1 %tobool215, label %for.cond.loopexit, label %for.body3.lr.ph
+
+for.body3.lr.ph:                                  ; preds = %for.body
+  br label %for.body3
+
+for.body3:                                        ; preds = %for.inc, %for.body3.lr.ph
+  %dec18.in = phi i32 [ %tmp2, %for.body3.lr.ph ], [ %dec18, %for.inc ]
+  %r.117 = phi i16* [ %r.022, %for.body3.lr.ph ], [ %add.ptr, %for.inc ]
+  %a.116 = phi i8* [ %a.021, %for.body3.lr.ph ], [ %incdec.ptr, %for.inc ]
+  %dec18 = add nsw i32 %dec18.in, -1
+  %tmp3 = load i8, i8* %a.116, align 1
+  %cmp = icmp eq i8 %tmp3, 0
+  br i1 %cmp, label %if.then, label %for.inc
+
+if.then:                                          ; preds = %for.body3
+  %arrayidx = getelementptr inbounds i16, i16* %r.117, i64 1
+  store i16 0, i16* %arrayidx, align 2
+  store i16 0, i16* %r.117, align 2
+  %arrayidx5 = getelementptr inbounds i16, i16* %r.117, i64 2
+  store i16 0, i16* %arrayidx5, align 2
+  br label %for.inc
+
+for.inc:                                          ; preds = %if.then, %for.body3
+  %incdec.ptr = getelementptr inbounds i8, i8* %a.116, i64 1
+  %add.ptr = getelementptr inbounds i16, i16* %r.117, i64 3
+  %tobool2 = icmp eq i32 %dec18, 0
+  br i1 %tobool2, label %for.cond1.for.cond.loopexit_crit_edge, label %for.body3, !llvm.loop !0
+
+for.cond.for.end6_crit_edge:                      ; preds = %for.cond.loopexit
+  br label %for.end6
+
+for.end6:                                         ; preds = %for.cond.for.end6_crit_edge, %entry
+  ret void
+}
+
+!0 = !{!0, !1}
+!1 = !{!"llvm.loop.unroll.count", i32 2}
diff --git a/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll b/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
index 4f934a688be8..dc812fb4065f 100644
--- a/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
+++ b/test/Transforms/LoopUnroll/unroll-pragmas-disabled.ll
@@ -19,8 +19,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -51,8 +51,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -78,8 +78,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -110,8 +110,8 @@ entry:
 
 for.body3:                                        ; preds = %for.body3, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body3 ]
-  %arrayidx = getelementptr inbounds i32* %List, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %List, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %0, 10
   store i32 %add4, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -124,8 +124,8 @@ for.body3.1.preheader:                            ; preds = %for.body3
 for.body3.1:                                      ; preds = %for.body3.1.preheader, %for.body3.1
   %indvars.iv.1 = phi i64 [ %1, %for.body3.1 ], [ 0, %for.body3.1.preheader ]
   %1 = add nsw i64 %indvars.iv.1, 1
-  %arrayidx.1 = getelementptr inbounds i32* %List, i64 %1
-  %2 = load i32* %arrayidx.1, align 4
+  %arrayidx.1 = getelementptr inbounds i32, i32* %List, i64 %1
+  %2 = load i32, i32* %arrayidx.1, align 4
   %add4.1 = add nsw i32 %2, 10
   store i32 %add4.1, i32* %arrayidx.1, align 4
   %exitcond.1 = icmp eq i64 %1, 4
diff --git a/test/Transforms/LoopUnroll/unroll-pragmas.ll b/test/Transforms/LoopUnroll/unroll-pragmas.ll
index d53fa49cc9f5..1354181becd0 100644
--- a/test/Transforms/LoopUnroll/unroll-pragmas.ll
+++ b/test/Transforms/LoopUnroll/unroll-pragmas.ll
@@ -1,5 +1,5 @@
-; RUN: opt < %s -loop-unroll -S | FileCheck %s
-; RUN: opt < %s -loop-unroll -loop-unroll -S | FileCheck %s
+; RUN: opt < %s -loop-unroll -pragma-unroll-threshold=1024 -S | FileCheck %s
+; RUN: opt < %s -loop-unroll -loop-unroll -pragma-unroll-threshold=1024 -S | FileCheck %s
 ;
 ; Run loop unrolling twice to verify that loop unrolling metadata is properly
 ; removed and further unrolling is disabled after the pass is run once.
@@ -19,8 +19,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -43,8 +43,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -71,8 +71,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -94,8 +94,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -124,8 +124,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -153,8 +153,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -190,8 +190,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -217,8 +217,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -245,8 +245,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %inc = add nsw i32 %0, 1
   store i32 %inc, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopUnswitch/2007-07-12-ExitDomInfo.ll b/test/Transforms/LoopUnswitch/2007-07-12-ExitDomInfo.ll
index bf5a61b36300..9da8dc2f3e85 100644
--- a/test/Transforms/LoopUnswitch/2007-07-12-ExitDomInfo.ll
+++ b/test/Transforms/LoopUnswitch/2007-07-12-ExitDomInfo.ll
@@ -17,7 +17,7 @@ bb36.outer:		; preds = %bb41, %bb36.preheader
 bb.nph:		; preds = %bb36.outer
 	%ttmp8 = icmp eq i8* null, null		; <i1> [#uses=1]
 	%ttmp6 = icmp eq i8* null, null		; <i1> [#uses=1]
-	%tmp31 = call i32 @strcspn( i8* null, i8* getelementptr ([3 x i8]* @str3, i64 0, i64 0) )		; <i32> [#uses=1]
+	%tmp31 = call i32 @strcspn( i8* null, i8* getelementptr ([3 x i8], [3 x i8]* @str3, i64 0, i64 0) )		; <i32> [#uses=1]
 	br i1 %ttmp8, label %cond_next, label %cond_true
 
 cond_true:		; preds = %bb.nph
diff --git a/test/Transforms/LoopUnswitch/2007-07-18-DomInfo.ll b/test/Transforms/LoopUnswitch/2007-07-18-DomInfo.ll
index dfca15403bfa..c01bed142fc7 100644
--- a/test/Transforms/LoopUnswitch/2007-07-18-DomInfo.ll
+++ b/test/Transforms/LoopUnswitch/2007-07-18-DomInfo.ll
@@ -52,7 +52,7 @@ cond_next1929:          ; preds = %cond_next1915
         br i1 false, label %cond_next1961, label %cond_next2009
 
 cond_next1961:          ; preds = %cond_next1929
-        %tmp1992 = getelementptr i8* %b.10.ph, i32 0            ; <i8*> [#uses=0]
+        %tmp1992 = getelementptr i8, i8* %b.10.ph, i32 0            ; <i8*> [#uses=0]
         br label %cond_next1915
 
 cond_next2009:          ; preds = %cond_next1929
diff --git a/test/Transforms/LoopUnswitch/2007-08-01-LCSSA.ll b/test/Transforms/LoopUnswitch/2007-08-01-LCSSA.ll
index f83acaa6b0b5..49b30bda50fa 100644
--- a/test/Transforms/LoopUnswitch/2007-08-01-LCSSA.ll
+++ b/test/Transforms/LoopUnswitch/2007-08-01-LCSSA.ll
@@ -19,7 +19,7 @@ declare i32 @strcmp(i8*, i8*)
 
 define i32 @_ZN9Generator6strregEPKc(%struct.Generator* %this, i8* %s) {
 entry:
-	%s_addr.0 = select i1 false, i8* getelementptr ([1 x i8]* @.str9, i32 0, i32 0), i8* %s		; <i8*> [#uses=2]
+	%s_addr.0 = select i1 false, i8* getelementptr ([1 x i8], [1 x i8]* @.str9, i32 0, i32 0), i8* %s		; <i8*> [#uses=2]
 	%tmp122 = icmp eq i8* %s_addr.0, null		; <i1> [#uses=1]
 	br label %bb184
 
diff --git a/test/Transforms/LoopUnswitch/2008-06-02-DomInfo.ll b/test/Transforms/LoopUnswitch/2008-06-02-DomInfo.ll
index 906c2c581e1f..8ea4ed33fc5c 100644
--- a/test/Transforms/LoopUnswitch/2008-06-02-DomInfo.ll
+++ b/test/Transforms/LoopUnswitch/2008-06-02-DomInfo.ll
@@ -19,7 +19,7 @@ bb54:		; preds = %bb31, %bb5, %bb
 	br i1 false, label %bb64, label %bb
 bb64:		; preds = %bb54
 	%tmp6566 = sext i16 %p_6 to i32		; <i32> [#uses=1]
-	%tmp68 = tail call i32 (...)* @func_18( i32 1, i32 %tmp6566, i32 1 ) nounwind 		; <i32> [#uses=0]
+	%tmp68 = tail call i32 (...) @func_18( i32 1, i32 %tmp6566, i32 1 ) nounwind 		; <i32> [#uses=0]
 	ret i32 undef
 }
 
diff --git a/test/Transforms/LoopUnswitch/2008-06-17-DomFrontier.ll b/test/Transforms/LoopUnswitch/2008-06-17-DomFrontier.ll
index f74054a0589c..d606ea9c6d20 100644
--- a/test/Transforms/LoopUnswitch/2008-06-17-DomFrontier.ll
+++ b/test/Transforms/LoopUnswitch/2008-06-17-DomFrontier.ll
@@ -17,6 +17,6 @@ bb36:		; preds = %bb19
 	store i16 0, i16* @g_56, align 2
 	br i1 false, label %bb44, label %bb3
 bb44:		; preds = %bb44, %bb36
-	%tmp46 = load i16* @g_56, align 2		; <i16> [#uses=0]
+	%tmp46 = load i16, i16* @g_56, align 2		; <i16> [#uses=0]
 	br i1 false, label %bb, label %bb44
 }
diff --git a/test/Transforms/LoopUnswitch/2008-11-03-Invariant.ll b/test/Transforms/LoopUnswitch/2008-11-03-Invariant.ll
index 31dba79be1f8..90c0944e38d8 100644
--- a/test/Transforms/LoopUnswitch/2008-11-03-Invariant.ll
+++ b/test/Transforms/LoopUnswitch/2008-11-03-Invariant.ll
@@ -15,7 +15,7 @@ bb:		; preds = %bb.nph, %bb3
 	br i1 %1, label %bb2, label %bb1
 
 bb1:		; preds = %bb
-	%2 = tail call i32 (...)* @b() nounwind		; <i32> [#uses=0]
+	%2 = tail call i32 (...) @b() nounwind		; <i32> [#uses=0]
 	br label %bb2
 
 bb2:		; preds = %bb, %bb1
diff --git a/test/Transforms/LoopUnswitch/2010-11-18-LCSSA.ll b/test/Transforms/LoopUnswitch/2010-11-18-LCSSA.ll
index a976d18d4445..3b89fa915833 100644
--- a/test/Transforms/LoopUnswitch/2010-11-18-LCSSA.ll
+++ b/test/Transforms/LoopUnswitch/2010-11-18-LCSSA.ll
@@ -7,7 +7,7 @@ entry:
   br i1 true, label %for.end12, label %bb.nph
 
 bb.nph:                                           ; preds = %entry
-  %g_38.promoted = load i32* @g_38
+  %g_38.promoted = load i32, i32* @g_38
   br label %for.body
 
 for.body:                                         ; preds = %for.cond, %bb.nph
diff --git a/test/Transforms/LoopUnswitch/2011-09-26-EHCrash.ll b/test/Transforms/LoopUnswitch/2011-09-26-EHCrash.ll
index e8feef383b05..0b7f91fee150 100644
--- a/test/Transforms/LoopUnswitch/2011-09-26-EHCrash.ll
+++ b/test/Transforms/LoopUnswitch/2011-09-26-EHCrash.ll
@@ -22,9 +22,9 @@ for.body:                                         ; preds = %for.cond
 
 if.then:                                          ; preds = %for.body
   %idxprom = sext i32 %inc1 to i64
-  %array_ = getelementptr inbounds %class.MyContainer.1.3.19.29* %this, i32 0, i32 0
-  %arrayidx = getelementptr inbounds [6 x %class.MyMemVarClass.0.2.18.28*]* %array_, i32 0, i64 %idxprom
-  %tmp4 = load %class.MyMemVarClass.0.2.18.28** %arrayidx, align 8
+  %array_ = getelementptr inbounds %class.MyContainer.1.3.19.29, %class.MyContainer.1.3.19.29* %this, i32 0, i32 0
+  %arrayidx = getelementptr inbounds [6 x %class.MyMemVarClass.0.2.18.28*], [6 x %class.MyMemVarClass.0.2.18.28*]* %array_, i32 0, i64 %idxprom
+  %tmp4 = load %class.MyMemVarClass.0.2.18.28*, %class.MyMemVarClass.0.2.18.28** %arrayidx, align 8
   %isnull = icmp eq %class.MyMemVarClass.0.2.18.28* %tmp4, null
   br i1 %isnull, label %for.inc, label %delete.notnull
 
diff --git a/test/Transforms/LoopUnswitch/2011-11-18-SimpleSwitch.ll b/test/Transforms/LoopUnswitch/2011-11-18-SimpleSwitch.ll
index a8608b877205..abeea209f639 100644
--- a/test/Transforms/LoopUnswitch/2011-11-18-SimpleSwitch.ll
+++ b/test/Transforms/LoopUnswitch/2011-11-18-SimpleSwitch.ll
@@ -15,7 +15,7 @@
 ; CHECK-NEXT:   br label %loop_begin.us
 
 ; CHECK:      loop_begin.us:                                    ; preds = %loop_begin.backedge.us, %.split.us
-; CHECK-NEXT:   %var_val.us = load i32* %var
+; CHECK-NEXT:   %var_val.us = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 1, label %default.us-lcssa.us [
 ; CHECK-NEXT:     i32 1, label %inc.us
 
@@ -34,7 +34,7 @@
 ; CHECK-NEXT:   br label %loop_begin.us1
 
 ; CHECK:      loop_begin.us1:                                   ; preds = %loop_begin.backedge.us5, %.split.split.us
-; CHECK-NEXT:   %var_val.us2 = load i32* %var
+; CHECK-NEXT:   %var_val.us.2 = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 2, label %default.us-lcssa.us-lcssa.us [
 ; CHECK-NEXT:     i32 1, label %inc.us4
 ; CHECK-NEXT:     i32 2, label %dec.us3
@@ -48,7 +48,7 @@
 ; CHECK-NEXT:   br label %loop_begin
 
 ; CHECK:      loop_begin:                                       ; preds = %loop_begin.backedge, %.split.split
-; CHECK-NEXT:   %var_val = load i32* %var
+; CHECK-NEXT:   %var_val = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 %c, label %default.us-lcssa.us-lcssa [
 ; CHECK-NEXT:     i32 1, label %inc
 ; CHECK-NEXT:     i32 2, label %dec
@@ -63,13 +63,13 @@
 define i32 @test(i32* %var) {
   %mem = alloca i32
   store i32 2, i32* %mem
-  %c = load i32* %mem
+  %c = load i32, i32* %mem
 
   br label %loop_begin
 
 loop_begin:
 
-  %var_val = load i32* %var
+  %var_val = load i32, i32* %var
 
   switch i32 %c, label %default [
       i32 1, label %inc
diff --git a/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches-Threshold.ll b/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches-Threshold.ll
index 686cedbbc51a..393dd5c313a4 100644
--- a/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches-Threshold.ll
+++ b/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches-Threshold.ll
@@ -52,14 +52,14 @@
 define i32 @test(i32* %var) {
   %mem = alloca i32
   store i32 2, i32* %mem
-  %c = load i32* %mem
-  %d = load i32* %mem
+  %c = load i32, i32* %mem
+  %d = load i32, i32* %mem
 
   br label %loop_begin
 
 loop_begin:
 
-  %var_val = load i32* %var
+  %var_val = load i32, i32* %var
 
   switch i32 %c, label %second_switch [
       i32 1, label %inc
diff --git a/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches.ll b/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches.ll
index 3ba9fc2f5cf1..02552ea5cc40 100644
--- a/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches.ll
+++ b/test/Transforms/LoopUnswitch/2011-11-18-TwoSwitches.ll
@@ -22,7 +22,7 @@
 ; CHECK-NEXT:   br label %loop_begin.us.us
 
 ; CHECK:      loop_begin.us.us:                                 ; preds = %loop_begin.backedge.us.us, %.split.us.split.us
-; CHECK-NEXT:   %var_val.us.us = load i32* %var
+; CHECK-NEXT:   %var_val.us.us = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 1, label %second_switch.us.us [
 ; CHECK-NEXT:     i32 1, label %inc.us.us
 
@@ -38,7 +38,7 @@
 ; CHECK-NEXT:   br label %loop_begin.us
 
 ; CHECK:      loop_begin.us:                                    ; preds = %loop_begin.backedge.us, %.split.us.split
-; CHECK-NEXT:   %var_val.us = load i32* %var
+; CHECK-NEXT:   %var_val.us = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 1, label %second_switch.us [
 ; CHECK-NEXT:     i32 1, label %inc.us
 
@@ -65,7 +65,7 @@
 ; CHECK-NEXT:   br label %loop_begin.us1
 
 ; CHECK:      loop_begin.us1:                                   ; preds = %loop_begin.backedge.us6, %.split.split.us
-; CHECK-NEXT:   %var_val.us2 = load i32* %var
+; CHECK-NEXT:   %var_val.us.2 = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 %c, label %second_switch.us3 [
 ; CHECK-NEXT:     i32 1, label %loop_begin.inc_crit_edge.us
 ; CHECK-NEXT:   ]
@@ -86,7 +86,7 @@
 ; CHECK-NEXT:   br label %loop_begin
 
 ; CHECK:      loop_begin:                                       ; preds = %loop_begin.backedge, %.split.split
-; CHECK-NEXT:   %var_val = load i32* %var
+; CHECK-NEXT:   %var_val = load i32, i32* %var
 ; CHECK-NEXT:   switch i32 %c, label %second_switch [
 ; CHECK-NEXT:     i32 1, label %loop_begin.inc_crit_edge
 ; CHECK-NEXT:   ]
@@ -106,14 +106,14 @@
 define i32 @test(i32* %var) {
   %mem = alloca i32
   store i32 2, i32* %mem
-  %c = load i32* %mem
-  %d = load i32* %mem
+  %c = load i32, i32* %mem
+  %d = load i32, i32* %mem
 
   br label %loop_begin
 
 loop_begin:
 
-  %var_val = load i32* %var
+  %var_val = load i32, i32* %var
 
   switch i32 %c, label %second_switch [
       i32 1, label %inc
diff --git a/test/Transforms/LoopUnswitch/2012-04-30-LoopUnswitch-LPad-Crash.ll b/test/Transforms/LoopUnswitch/2012-04-30-LoopUnswitch-LPad-Crash.ll
index a6c0d83484b3..223fbf18bf53 100644
--- a/test/Transforms/LoopUnswitch/2012-04-30-LoopUnswitch-LPad-Crash.ll
+++ b/test/Transforms/LoopUnswitch/2012-04-30-LoopUnswitch-LPad-Crash.ll
@@ -46,9 +46,9 @@ define void @_ZN1DptEv(%class.D.22.42.66.102.138.158.178.198.238.242.246.250.262
 entry:
   %this.addr = alloca %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379*, align 8
   store %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379* %this, %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379** %this.addr, align 8
-  %this1 = load %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379** %this.addr
-  %px = getelementptr inbounds %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379* %this1, i32 0, i32 0
-  %0 = load %class.C.23.43.67.103.139.159.179.199.239.243.247.251.263.295.303.339.347.376** %px, align 8
+  %this1 = load %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379*, %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379** %this.addr
+  %px = getelementptr inbounds %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379, %class.D.22.42.66.102.138.158.178.198.238.242.246.250.262.294.302.338.346.379* %this1, i32 0, i32 0
+  %0 = load %class.C.23.43.67.103.139.159.179.199.239.243.247.251.263.295.303.339.347.376*, %class.C.23.43.67.103.139.159.179.199.239.243.247.251.263.295.303.339.347.376** %px, align 8
   %tobool = icmp ne %class.C.23.43.67.103.139.159.179.199.239.243.247.251.263.295.303.339.347.376* %0, null
   br i1 %tobool, label %cond.end, label %cond.false
 
diff --git a/test/Transforms/LoopUnswitch/2012-05-20-Phi.ll b/test/Transforms/LoopUnswitch/2012-05-20-Phi.ll
index 4c63a56bfdc5..96bc28c9a358 100644
--- a/test/Transforms/LoopUnswitch/2012-05-20-Phi.ll
+++ b/test/Transforms/LoopUnswitch/2012-05-20-Phi.ll
@@ -9,9 +9,9 @@ target triple = "x86_64-unknown-linux-gnu"
 
 define void @func() noreturn nounwind uwtable {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %0, 0
-  %1 = load i32* @b, align 4
+  %1 = load i32, i32* @b, align 4
   br label %while.body
 
 while.body:                                       ; preds = %while.body, %entry
diff --git a/test/Transforms/LoopUnswitch/basictest.ll b/test/Transforms/LoopUnswitch/basictest.ll
index 85e44eca9b63..e990144d5ccc 100644
--- a/test/Transforms/LoopUnswitch/basictest.ll
+++ b/test/Transforms/LoopUnswitch/basictest.ll
@@ -6,13 +6,13 @@ entry:
 no_exit:		; preds = %no_exit.backedge, %entry
 	%i.0.0 = phi i32 [ 0, %entry ], [ %i.0.0.be, %no_exit.backedge ]		; <i32> [#uses=3]
 	%gep.upgrd.1 = zext i32 %i.0.0 to i64		; <i64> [#uses=1]
-	%tmp.7 = getelementptr i32* %A, i64 %gep.upgrd.1		; <i32*> [#uses=4]
-	%tmp.13 = load i32* %tmp.7		; <i32> [#uses=2]
+	%tmp.7 = getelementptr i32, i32* %A, i64 %gep.upgrd.1		; <i32*> [#uses=4]
+	%tmp.13 = load i32, i32* %tmp.7		; <i32> [#uses=2]
 	%tmp.14 = add i32 %tmp.13, 1		; <i32> [#uses=1]
 	store i32 %tmp.14, i32* %tmp.7
 	br i1 %C, label %then, label %endif
 then:		; preds = %no_exit
-	%tmp.29 = load i32* %tmp.7		; <i32> [#uses=1]
+	%tmp.29 = load i32, i32* %tmp.7		; <i32> [#uses=1]
 	%tmp.30 = add i32 %tmp.29, 2		; <i32> [#uses=1]
 	store i32 %tmp.30, i32* %tmp.7
 	%inc9 = add i32 %i.0.0, 1		; <i32> [#uses=2]
@@ -36,13 +36,13 @@ return:		; preds = %endif, %then
 define i32 @test2(i32* %var) {
   %mem = alloca i32
   store i32 2, i32* %mem
-  %c = load i32* %mem
+  %c = load i32, i32* %mem
 
   br label %loop_begin
 
 loop_begin:
 
-  %var_val = load i32* %var
+  %var_val = load i32, i32* %var
 
   switch i32 %c, label %default [
       i32 1, label %inc
diff --git a/test/Transforms/LoopUnswitch/preserve-analyses.ll b/test/Transforms/LoopUnswitch/preserve-analyses.ll
index f79612bef51e..e3774a1e907a 100644
--- a/test/Transforms/LoopUnswitch/preserve-analyses.ll
+++ b/test/Transforms/LoopUnswitch/preserve-analyses.ll
@@ -11,8 +11,8 @@ target triple = "armv6-apple-darwin9"
 
 define i32 @ineqn(i8* %s, i8* %p) nounwind readonly {
 entry:
-  %0 = load i32* @delim1, align 4                 ; <i32> [#uses=1]
-  %1 = load i32* @delim2, align 4                 ; <i32> [#uses=1]
+  %0 = load i32, i32* @delim1, align 4                 ; <i32> [#uses=1]
+  %1 = load i32, i32* @delim2, align 4                 ; <i32> [#uses=1]
   br label %bb8.outer
 
 bb:                                               ; preds = %bb8
@@ -20,7 +20,7 @@ bb:                                               ; preds = %bb8
   br i1 %2, label %bb10, label %bb2
 
 bb2:                                              ; preds = %bb
-  %3 = getelementptr inbounds i8* %p_addr.0, i32 1 ; <i8*> [#uses=3]
+  %3 = getelementptr inbounds i8, i8* %p_addr.0, i32 1 ; <i8*> [#uses=3]
   switch i32 %ineq.0.ph, label %bb8.backedge [
     i32 0, label %bb3
     i32 1, label %bb6
@@ -61,7 +61,7 @@ bb8.outer:                                        ; preds = %bb8.outer.backedge,
 
 bb8:                                              ; preds = %bb8.outer, %bb8.backedge
   %p_addr.0 = phi i8* [ %p_addr.0.ph, %bb8.outer ], [ %3, %bb8.backedge ] ; <i8*> [#uses=3]
-  %7 = load i8* %p_addr.0, align 1                ; <i8> [#uses=2]
+  %7 = load i8, i8* %p_addr.0, align 1                ; <i8> [#uses=2]
   %8 = sext i8 %7 to i32                          ; <i32> [#uses=2]
   %9 = icmp eq i8 %7, 0                           ; <i1> [#uses=1]
   br i1 %9, label %bb10, label %bb
diff --git a/test/Transforms/LoopVectorize/12-12-11-if-conv.ll b/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
index a292afbd5882..d5e020c36e98 100644
--- a/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
+++ b/test/Transforms/LoopVectorize/12-12-11-if-conv.ll
@@ -14,8 +14,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %if.end
   %indvars.iv = phi i64 [ %indvars.iv.next, %if.end ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %if.end, label %if.then
 
diff --git a/test/Transforms/LoopVectorize/2012-10-22-isconsec.ll b/test/Transforms/LoopVectorize/2012-10-22-isconsec.ll
index 16d64eab25c3..9d848bdede2c 100644
--- a/test/Transforms/LoopVectorize/2012-10-22-isconsec.ll
+++ b/test/Transforms/LoopVectorize/2012-10-22-isconsec.ll
@@ -22,7 +22,7 @@ entry:
 "3":                                              ; preds = %"3.lr.ph", %"3"
   %indvars.iv = phi i64 [ 0, %"3.lr.ph" ], [ %indvars.iv.next, %"3" ]
   %3 = shl nsw i64 %indvars.iv, 2
-  %4 = getelementptr inbounds i8* %1, i64 %3
+  %4 = getelementptr inbounds i8, i8* %1, i64 %3
   %5 = bitcast i8* %4 to float*
   store float %value, float* %5, align 4
   %indvars.iv.next = add i64 %indvars.iv, %2
@@ -41,7 +41,7 @@ entry:
 "3":                                              ; preds = %"3", %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %"3" ]
   %0 = shl nsw i64 %indvars.iv, 2
-  %1 = getelementptr inbounds i8* bitcast (float* getelementptr inbounds ([32000 x float]* @b, i64 0, i64 16000) to i8*), i64 %0
+  %1 = getelementptr inbounds i8, i8* bitcast (float* getelementptr inbounds ([32000 x float], [32000 x float]* @b, i64 0, i64 16000) to i8*), i64 %0
   %2 = bitcast i8* %1 to float*
   store float -1.000000e+00, float* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/AArch64/aarch64-unroll.ll b/test/Transforms/LoopVectorize/AArch64/aarch64-unroll.ll
index 9962c3d76a66..a689f44e912c 100644
--- a/test/Transforms/LoopVectorize/AArch64/aarch64-unroll.ll
+++ b/test/Transforms/LoopVectorize/AArch64/aarch64-unroll.ll
@@ -22,12 +22,12 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
-  %arrayidx4 = getelementptr inbounds i32* %c, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i32, i32* %c, i64 %indvars.iv
   store i32 %add, i32* %arrayidx4, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/AArch64/arbitrary-induction-step.ll b/test/Transforms/LoopVectorize/AArch64/arbitrary-induction-step.ll
new file mode 100644
index 000000000000..4cd703f64582
--- /dev/null
+++ b/test/Transforms/LoopVectorize/AArch64/arbitrary-induction-step.ll
@@ -0,0 +1,150 @@
+; RUN: opt -S < %s -loop-vectorize 2>&1 | FileCheck %s
+; RUN: opt -S < %s -loop-vectorize -force-vector-interleave=1 -force-vector-width=2 | FileCheck %s --check-prefix=FORCE-VEC
+
+target datalayout = "e-m:e-i64:64-i128:128-n32:64-S128"
+target triple = "aarch64--linux-gnueabi"
+
+; Test integer induction variable of step 2:
+;   for (int i = 0; i < 1024; i+=2) {
+;     int tmp = *A++;
+;     sum += i * tmp;
+;   }
+
+; CHECK-LABEL: @ind_plus2(
+; CHECK: load <4 x i32>, <4 x i32>*
+; CHECK: load <4 x i32>, <4 x i32>*
+; CHECK: mul nsw <4 x i32>
+; CHECK: mul nsw <4 x i32>
+; CHECK: add nsw <4 x i32>
+; CHECK: add nsw <4 x i32>
+; CHECK: %index.next = add i64 %index, 8
+; CHECK: icmp eq i64 %index.next, 512
+
+; FORCE-VEC-LABEL: @ind_plus2(
+; FORCE-VEC: %wide.load = load <2 x i32>, <2 x i32>*
+; FORCE-VEC: mul nsw <2 x i32>
+; FORCE-VEC: add nsw <2 x i32>
+; FORCE-VEC: %index.next = add i64 %index, 2
+; FORCE-VEC: icmp eq i64 %index.next, 512
+define i32 @ind_plus2(i32* %A) {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %for.body
+  %A.addr = phi i32* [ %A, %entry ], [ %inc.ptr, %for.body ]
+  %i = phi i32 [ 0, %entry ], [ %add1, %for.body ]
+  %sum = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %inc.ptr = getelementptr inbounds i32, i32* %A.addr, i64 1
+  %0 = load i32, i32* %A.addr, align 4
+  %mul = mul nsw i32 %0, %i
+  %add = add nsw i32 %mul, %sum
+  %add1 = add nsw i32 %i, 2
+  %cmp = icmp slt i32 %add1, 1024
+  br i1 %cmp, label %for.body, label %for.end
+
+for.end:                                          ; preds = %for.body
+  %add.lcssa = phi i32 [ %add, %for.body ]
+  ret i32 %add.lcssa
+}
+
+
+; Test integer induction variable of step -2:
+;   for (int i = 1024; i > 0; i-=2) {
+;     int tmp = *A++;
+;     sum += i * tmp;
+;   }
+
+; CHECK-LABEL: @ind_minus2(
+; CHECK: load <4 x i32>, <4 x i32>*
+; CHECK: load <4 x i32>, <4 x i32>*
+; CHECK: mul nsw <4 x i32>
+; CHECK: mul nsw <4 x i32>
+; CHECK: add nsw <4 x i32>
+; CHECK: add nsw <4 x i32>
+; CHECK: %index.next = add i64 %index, 8
+; CHECK: icmp eq i64 %index.next, 512
+
+; FORCE-VEC-LABEL: @ind_minus2(
+; FORCE-VEC: %wide.load = load <2 x i32>, <2 x i32>*
+; FORCE-VEC: mul nsw <2 x i32>
+; FORCE-VEC: add nsw <2 x i32>
+; FORCE-VEC: %index.next = add i64 %index, 2
+; FORCE-VEC: icmp eq i64 %index.next, 512
+define i32 @ind_minus2(i32* %A) {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %entry, %for.body
+  %A.addr = phi i32* [ %A, %entry ], [ %inc.ptr, %for.body ]
+  %i = phi i32 [ 1024, %entry ], [ %sub, %for.body ]
+  %sum = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %inc.ptr = getelementptr inbounds i32, i32* %A.addr, i64 1
+  %0 = load i32, i32* %A.addr, align 4
+  %mul = mul nsw i32 %0, %i
+  %add = add nsw i32 %mul, %sum
+  %sub = add nsw i32 %i, -2
+  %cmp = icmp sgt i32 %i, 2
+  br i1 %cmp, label %for.body, label %for.end
+
+for.end:                                          ; preds = %for.body
+  %add.lcssa = phi i32 [ %add, %for.body ]
+  ret i32 %add.lcssa
+}
+
+
+; Test pointer induction variable of step 2. As currently we don't support
+; masked load/store, vectorization is possible but not beneficial. If loop
+; vectorization is not enforced, LV will only do interleave.
+;   for (int i = 0; i < 1024; i++) {
+;     int tmp0 = *A++;
+;     int tmp1 = *A++;
+;     sum += tmp0 * tmp1;
+;   }
+
+; CHECK-LABEL: @ptr_ind_plus2(
+; CHECK: load i32, i32*
+; CHECK: load i32, i32*
+; CHECK: load i32, i32*
+; CHECK: load i32, i32*
+; CHECK: mul nsw i32
+; CHECK: mul nsw i32
+; CHECK: add nsw i32
+; CHECK: add nsw i32
+; CHECK: %index.next = add i64 %index, 2
+; CHECK: %21 = icmp eq i64 %index.next, 1024
+
+; FORCE-VEC-LABEL: @ptr_ind_plus2(
+; FORCE-VEC: load i32, i32*
+; FORCE-VEC: insertelement <2 x i32>
+; FORCE-VEC: load i32, i32*
+; FORCE-VEC: insertelement <2 x i32>
+; FORCE-VEC: load i32, i32*
+; FORCE-VEC: insertelement <2 x i32>
+; FORCE-VEC: load i32, i32*
+; FORCE-VEC: insertelement <2 x i32>
+; FORCE-VEC: mul nsw <2 x i32>
+; FORCE-VEC: add nsw <2 x i32>
+; FORCE-VEC: %index.next = add i64 %index, 2
+; FORCE-VEC: icmp eq i64 %index.next, 1024
+define i32 @ptr_ind_plus2(i32* %A) {
+entry:
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %entry
+  %A.addr = phi i32* [ %A, %entry ], [ %inc.ptr1, %for.body ]
+  %sum = phi i32 [ 0, %entry ], [ %add, %for.body ]
+  %i = phi i32 [ 0, %entry ], [ %inc, %for.body ]
+  %inc.ptr = getelementptr inbounds i32, i32* %A.addr, i64 1
+  %0 = load i32, i32* %A.addr, align 4
+  %inc.ptr1 = getelementptr inbounds i32, i32* %A.addr, i64 2
+  %1 = load i32, i32* %inc.ptr, align 4
+  %mul = mul nsw i32 %1, %0
+  %add = add nsw i32 %mul, %sum
+  %inc = add nsw i32 %i, 1
+  %exitcond = icmp eq i32 %inc, 1024
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body
+  %add.lcssa = phi i32 [ %add, %for.body ]
+  ret i32 %add.lcssa
+}
diff --git a/test/Transforms/LoopVectorize/AArch64/arm64-unroll.ll b/test/Transforms/LoopVectorize/AArch64/arm64-unroll.ll
index f8eb3ed1f35f..395b468c509c 100644
--- a/test/Transforms/LoopVectorize/AArch64/arm64-unroll.ll
+++ b/test/Transforms/LoopVectorize/AArch64/arm64-unroll.ll
@@ -22,12 +22,12 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
-  %arrayidx4 = getelementptr inbounds i32* %c, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i32, i32* %c, i64 %indvars.iv
   store i32 %add, i32* %arrayidx4, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/AArch64/gather-cost.ll b/test/Transforms/LoopVectorize/AArch64/gather-cost.ll
index bb285382e53c..46b8ef1a09bf 100644
--- a/test/Transforms/LoopVectorize/AArch64/gather-cost.ll
+++ b/test/Transforms/LoopVectorize/AArch64/gather-cost.ll
@@ -30,32 +30,32 @@ for.body:
   %b.054 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add30, %for.body ]
   %add = add i64 %v.055, %offset
   %mul = mul i64 %add, 3
-  %arrayidx = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %mul
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds [512 x float]* @kernel, i64 0, i64 %v.055
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %mul
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds [512 x float], [512 x float]* @kernel, i64 0, i64 %v.055
+  %1 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %0, %1
-  %arrayidx4 = getelementptr inbounds [512 x float]* @kernel2, i64 0, i64 %v.055
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds [512 x float], [512 x float]* @kernel2, i64 0, i64 %v.055
+  %2 = load float, float* %arrayidx4, align 4
   %mul5 = fmul fast float %mul3, %2
-  %arrayidx6 = getelementptr inbounds [512 x float]* @kernel3, i64 0, i64 %v.055
-  %3 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds [512 x float], [512 x float]* @kernel3, i64 0, i64 %v.055
+  %3 = load float, float* %arrayidx6, align 4
   %mul7 = fmul fast float %mul5, %3
-  %arrayidx8 = getelementptr inbounds [512 x float]* @kernel4, i64 0, i64 %v.055
-  %4 = load float* %arrayidx8, align 4
+  %arrayidx8 = getelementptr inbounds [512 x float], [512 x float]* @kernel4, i64 0, i64 %v.055
+  %4 = load float, float* %arrayidx8, align 4
   %mul9 = fmul fast float %mul7, %4
   %add10 = fadd fast float %r.057, %mul9
   %arrayidx.sum = add i64 %mul, 1
-  %arrayidx11 = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum
-  %5 = load float* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum
+  %5 = load float, float* %arrayidx11, align 4
   %mul13 = fmul fast float %1, %5
   %mul15 = fmul fast float %2, %mul13
   %mul17 = fmul fast float %3, %mul15
   %mul19 = fmul fast float %4, %mul17
   %add20 = fadd fast float %g.056, %mul19
   %arrayidx.sum52 = add i64 %mul, 2
-  %arrayidx21 = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum52
-  %6 = load float* %arrayidx21, align 4
+  %arrayidx21 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum52
+  %6 = load float, float* %arrayidx21, align 4
   %mul23 = fmul fast float %1, %6
   %mul25 = fmul fast float %2, %mul23
   %mul27 = fmul fast float %3, %mul25
diff --git a/test/Transforms/LoopVectorize/AArch64/sdiv-pow2.ll b/test/Transforms/LoopVectorize/AArch64/sdiv-pow2.ll
index a01d54312a60..f3c6548c68ed 100644
--- a/test/Transforms/LoopVectorize/AArch64/sdiv-pow2.ll
+++ b/test/Transforms/LoopVectorize/AArch64/sdiv-pow2.ll
@@ -7,7 +7,7 @@ target triple = "aarch64--linux-gnu"
 @Foo = common global %struct.anon zeroinitializer, align 4
 
 ; CHECK-LABEL: @foo(
-; CHECK: load <4 x i32>*
+; CHECK: load <4 x i32>, <4 x i32>*
 ; CHECK: sdiv <4 x i32>
 ; CHECK: store <4 x i32>
 
@@ -17,10 +17,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds %struct.anon* @Foo, i64 0, i32 2, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds %struct.anon, %struct.anon* @Foo, i64 0, i32 2, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %div = sdiv i32 %0, 2
-  %arrayidx2 = getelementptr inbounds %struct.anon* @Foo, i64 0, i32 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds %struct.anon, %struct.anon* @Foo, i64 0, i32 0, i64 %indvars.iv
   store i32 %div, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 100
diff --git a/test/Transforms/LoopVectorize/ARM/arm-unroll.ll b/test/Transforms/LoopVectorize/ARM/arm-unroll.ll
index 8843fc2d2b1a..7b09913636f7 100644
--- a/test/Transforms/LoopVectorize/ARM/arm-unroll.ll
+++ b/test/Transforms/LoopVectorize/ARM/arm-unroll.ll
@@ -20,8 +20,8 @@ define i32 @foo(i32* nocapture %A, i32 %n) nounwind readonly ssp {
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %i.02 = phi i32 [ %5, %.lr.ph ], [ 0, %0 ]
   %sum.01 = phi i32 [ %4, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i32 %i.02
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i32 %i.02
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i32 %3, %sum.01
   %5 = add nsw i32 %i.02, 1
   %exitcond = icmp eq i32 %5, %n
@@ -48,8 +48,8 @@ define i32 @register_limit(i32* nocapture %A, i32 %n) {
   %sum.04 = phi i32 [ %8, %.lr.ph ], [ 0, %0 ]
   %sum.05 = phi i32 [ %9, %.lr.ph ], [ 0, %0 ]
   %sum.06 = phi i32 [ %10, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i32 %i.02
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i32 %i.02
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i32 %3, %sum.01
   %5 = add nsw i32 %i.02, 1
   %6 = add nsw i32 %3, %sum.02
diff --git a/test/Transforms/LoopVectorize/ARM/gather-cost.ll b/test/Transforms/LoopVectorize/ARM/gather-cost.ll
index 239a28fa6c7c..f14a8cc6ca2d 100644
--- a/test/Transforms/LoopVectorize/ARM/gather-cost.ll
+++ b/test/Transforms/LoopVectorize/ARM/gather-cost.ll
@@ -33,32 +33,32 @@ for.body:
   %b.054 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add30, %for.body ]
   %add = add i32 %v.055, %offset
   %mul = mul i32 %add, 3
-  %arrayidx = getelementptr inbounds [1536 x float]* @src_data, i32 0, i32 %mul
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds [512 x float]* @kernel, i32 0, i32 %v.055
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i32 0, i32 %mul
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds [512 x float], [512 x float]* @kernel, i32 0, i32 %v.055
+  %1 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %0, %1
-  %arrayidx4 = getelementptr inbounds [512 x float]* @kernel2, i32 0, i32 %v.055
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds [512 x float], [512 x float]* @kernel2, i32 0, i32 %v.055
+  %2 = load float, float* %arrayidx4, align 4
   %mul5 = fmul fast float %mul3, %2
-  %arrayidx6 = getelementptr inbounds [512 x float]* @kernel3, i32 0, i32 %v.055
-  %3 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds [512 x float], [512 x float]* @kernel3, i32 0, i32 %v.055
+  %3 = load float, float* %arrayidx6, align 4
   %mul7 = fmul fast float %mul5, %3
-  %arrayidx8 = getelementptr inbounds [512 x float]* @kernel4, i32 0, i32 %v.055
-  %4 = load float* %arrayidx8, align 4
+  %arrayidx8 = getelementptr inbounds [512 x float], [512 x float]* @kernel4, i32 0, i32 %v.055
+  %4 = load float, float* %arrayidx8, align 4
   %mul9 = fmul fast float %mul7, %4
   %add10 = fadd fast float %r.057, %mul9
   %arrayidx.sum = add i32 %mul, 1
-  %arrayidx11 = getelementptr inbounds [1536 x float]* @src_data, i32 0, i32 %arrayidx.sum
-  %5 = load float* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i32 0, i32 %arrayidx.sum
+  %5 = load float, float* %arrayidx11, align 4
   %mul13 = fmul fast float %1, %5
   %mul15 = fmul fast float %2, %mul13
   %mul17 = fmul fast float %3, %mul15
   %mul19 = fmul fast float %4, %mul17
   %add20 = fadd fast float %g.056, %mul19
   %arrayidx.sum52 = add i32 %mul, 2
-  %arrayidx21 = getelementptr inbounds [1536 x float]* @src_data, i32 0, i32 %arrayidx.sum52
-  %6 = load float* %arrayidx21, align 4
+  %arrayidx21 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i32 0, i32 %arrayidx.sum52
+  %6 = load float, float* %arrayidx21, align 4
   %mul23 = fmul fast float %1, %6
   %mul25 = fmul fast float %2, %mul23
   %mul27 = fmul fast float %3, %mul25
diff --git a/test/Transforms/LoopVectorize/ARM/gcc-examples.ll b/test/Transforms/LoopVectorize/ARM/gcc-examples.ll
index f2bd0ac200d4..783156d7399d 100644
--- a/test/Transforms/LoopVectorize/ARM/gcc-examples.ll
+++ b/test/Transforms/LoopVectorize/ARM/gcc-examples.ll
@@ -18,12 +18,12 @@ define void @example1() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -44,10 +44,10 @@ define void @example10b(i16* noalias nocapture %sa, i16* noalias nocapture %sb,
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i16* %sb, i64 %indvars.iv
-  %3 = load i16* %2, align 2
+  %2 = getelementptr inbounds i16, i16* %sb, i64 %indvars.iv
+  %3 = load i16, i16* %2, align 2
   %4 = sext i16 %3 to i32
-  %5 = getelementptr inbounds i32* %ia, i64 %indvars.iv
+  %5 = getelementptr inbounds i32, i32* %ia, i64 %indvars.iv
   store i32 %4, i32* %5, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/ARM/mul-cast-vect.ll b/test/Transforms/LoopVectorize/ARM/mul-cast-vect.ll
index d2e3de279f7c..e88fcca1225c 100644
--- a/test/Transforms/LoopVectorize/ARM/mul-cast-vect.ll
+++ b/test/Transforms/LoopVectorize/ARM/mul-cast-vect.ll
@@ -16,9 +16,9 @@ target triple = "armv7--linux-gnueabihf"
 
 define void @direct(%T432* %loadaddr, %T432* %loadaddr2, %T432* %storeaddr) {
 ; COST: function 'direct':
-  %v0 = load %T432* %loadaddr
+  %v0 = load %T432, %T432* %loadaddr
 ; ASM: vld1.64
-  %v1 = load %T432* %loadaddr2
+  %v1 = load %T432, %T432* %loadaddr2
 ; ASM: vld1.64
   %r3 = mul %T432 %v0, %v1 
 ; COST: cost of 2 for instruction: {{.*}} mul <4 x i32>
@@ -30,9 +30,9 @@ define void @direct(%T432* %loadaddr, %T432* %loadaddr2, %T432* %storeaddr) {
 
 define void @ups1632(%T416* %loadaddr, %T416* %loadaddr2, %T432* %storeaddr) {
 ; COST: function 'ups1632':
-  %v0 = load %T416* %loadaddr
+  %v0 = load %T416, %T416* %loadaddr
 ; ASM: vldr
-  %v1 = load %T416* %loadaddr2
+  %v1 = load %T416, %T416* %loadaddr2
 ; ASM: vldr
   %r1 = sext %T416 %v0 to %T432
   %r2 = sext %T416 %v1 to %T432
@@ -47,9 +47,9 @@ define void @ups1632(%T416* %loadaddr, %T416* %loadaddr2, %T432* %storeaddr) {
 
 define void @upu1632(%T416* %loadaddr, %T416* %loadaddr2, %T432* %storeaddr) {
 ; COST: function 'upu1632':
-  %v0 = load %T416* %loadaddr
+  %v0 = load %T416, %T416* %loadaddr
 ; ASM: vldr
-  %v1 = load %T416* %loadaddr2
+  %v1 = load %T416, %T416* %loadaddr2
 ; ASM: vldr
   %r1 = zext %T416 %v0 to %T432
   %r2 = zext %T416 %v1 to %T432
@@ -64,9 +64,9 @@ define void @upu1632(%T416* %loadaddr, %T416* %loadaddr2, %T432* %storeaddr) {
 
 define void @ups3264(%T232* %loadaddr, %T232* %loadaddr2, %T264* %storeaddr) {
 ; COST: function 'ups3264':
-  %v0 = load %T232* %loadaddr
+  %v0 = load %T232, %T232* %loadaddr
 ; ASM: vldr
-  %v1 = load %T232* %loadaddr2
+  %v1 = load %T232, %T232* %loadaddr2
 ; ASM: vldr
   %r3 = mul %T232 %v0, %v1 
 ; ASM: vmul.i32
@@ -81,9 +81,9 @@ define void @ups3264(%T232* %loadaddr, %T232* %loadaddr2, %T264* %storeaddr) {
 
 define void @upu3264(%T232* %loadaddr, %T232* %loadaddr2, %T264* %storeaddr) {
 ; COST: function 'upu3264':
-  %v0 = load %T232* %loadaddr
+  %v0 = load %T232, %T232* %loadaddr
 ; ASM: vldr
-  %v1 = load %T232* %loadaddr2
+  %v1 = load %T232, %T232* %loadaddr2
 ; ASM: vldr
   %r3 = mul %T232 %v0, %v1 
 ; ASM: vmul.i32
@@ -98,9 +98,9 @@ define void @upu3264(%T232* %loadaddr, %T232* %loadaddr2, %T264* %storeaddr) {
 
 define void @dn3216(%T432* %loadaddr, %T432* %loadaddr2, %T416* %storeaddr) {
 ; COST: function 'dn3216':
-  %v0 = load %T432* %loadaddr
+  %v0 = load %T432, %T432* %loadaddr
 ; ASM: vld1.64
-  %v1 = load %T432* %loadaddr2
+  %v1 = load %T432, %T432* %loadaddr2
 ; ASM: vld1.64
   %r3 = mul %T432 %v0, %v1 
 ; ASM: vmul.i32
diff --git a/test/Transforms/LoopVectorize/ARM/width-detect.ll b/test/Transforms/LoopVectorize/ARM/width-detect.ll
index 99d7fa75ee33..66d2556dfb83 100644
--- a/test/Transforms/LoopVectorize/ARM/width-detect.ll
+++ b/test/Transforms/LoopVectorize/ARM/width-detect.ll
@@ -13,8 +13,8 @@ define float @foo_F32(float* nocapture %A, i32 %n) nounwind uwtable readonly ssp
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %prod.01 = phi float [ %4, %.lr.ph ], [ 0.000000e+00, %0 ]
-  %2 = getelementptr inbounds float* %A, i64 %indvars.iv
-  %3 = load float* %2, align 8
+  %2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %3 = load float, float* %2, align 8
   %4 = fmul fast float %prod.01, %3
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -36,8 +36,8 @@ define signext i8 @foo_I8(i8* nocapture %A, i32 %n) nounwind uwtable readonly ss
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %red.01 = phi i8 [ %4, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i8* %A, i64 %indvars.iv
-  %3 = load i8* %2, align 1
+  %2 = getelementptr inbounds i8, i8* %A, i64 %indvars.iv
+  %3 = load i8, i8* %2, align 1
   %4 = xor i8 %3, %red.01
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/PowerPC/large-loop-rdx.ll b/test/Transforms/LoopVectorize/PowerPC/large-loop-rdx.ll
new file mode 100644
index 000000000000..de6595f64ed3
--- /dev/null
+++ b/test/Transforms/LoopVectorize/PowerPC/large-loop-rdx.ll
@@ -0,0 +1,73 @@
+; RUN: opt < %s -loop-vectorize -S | FileCheck %s
+
+; CHECK: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT-NOT: fadd
+
+target datalayout = "e-m:e-i64:64-n32:64"
+target triple = "powerpc64le-ibm-linux-gnu"
+
+define void @QLA_F3_r_veq_norm2_V(float* noalias nocapture %r, [3 x { float, float }]* noalias nocapture readonly %a, i32 signext %n) #0 {
+entry:
+  %cmp24 = icmp sgt i32 %n, 0
+  br i1 %cmp24, label %for.cond1.preheader.preheader, label %for.end13
+
+for.cond1.preheader.preheader:                    ; preds = %entry
+  br label %for.cond1.preheader
+
+for.cond1.preheader:                              ; preds = %for.cond1.preheader.preheader, %for.cond1.preheader
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.cond1.preheader ], [ 0, %for.cond1.preheader.preheader ]
+  %sum.026 = phi double [ %add10.2, %for.cond1.preheader ], [ 0.000000e+00, %for.cond1.preheader.preheader ]
+  %arrayidx5.realp = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 0, i32 0
+  %arrayidx5.real = load float, float* %arrayidx5.realp, align 8
+  %arrayidx5.imagp = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 0, i32 1
+  %arrayidx5.imag = load float, float* %arrayidx5.imagp, align 8
+  %mul = fmul fast float %arrayidx5.real, %arrayidx5.real
+  %mul9 = fmul fast float %arrayidx5.imag, %arrayidx5.imag
+  %add = fadd fast float %mul9, %mul
+  %conv = fpext float %add to double
+  %add10 = fadd fast double %conv, %sum.026
+  %arrayidx5.realp.1 = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 1, i32 0
+  %arrayidx5.real.1 = load float, float* %arrayidx5.realp.1, align 8
+  %arrayidx5.imagp.1 = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 1, i32 1
+  %arrayidx5.imag.1 = load float, float* %arrayidx5.imagp.1, align 8
+  %mul.1 = fmul fast float %arrayidx5.real.1, %arrayidx5.real.1
+  %mul9.1 = fmul fast float %arrayidx5.imag.1, %arrayidx5.imag.1
+  %add.1 = fadd fast float %mul9.1, %mul.1
+  %conv.1 = fpext float %add.1 to double
+  %add10.1 = fadd fast double %conv.1, %add10
+  %arrayidx5.realp.2 = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 2, i32 0
+  %arrayidx5.real.2 = load float, float* %arrayidx5.realp.2, align 8
+  %arrayidx5.imagp.2 = getelementptr inbounds [3 x { float, float }], [3 x { float, float }]* %a, i64 %indvars.iv, i64 2, i32 1
+  %arrayidx5.imag.2 = load float, float* %arrayidx5.imagp.2, align 8
+  %mul.2 = fmul fast float %arrayidx5.real.2, %arrayidx5.real.2
+  %mul9.2 = fmul fast float %arrayidx5.imag.2, %arrayidx5.imag.2
+  %add.2 = fadd fast float %mul9.2, %mul.2
+  %conv.2 = fpext float %add.2 to double
+  %add10.2 = fadd fast double %conv.2, %add10.1
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.cond.for.end13_crit_edge, label %for.cond1.preheader
+
+for.cond.for.end13_crit_edge:                     ; preds = %for.cond1.preheader
+  %add10.2.lcssa = phi double [ %add10.2, %for.cond1.preheader ]
+  %phitmp = fptrunc double %add10.2.lcssa to float
+  br label %for.end13
+
+for.end13:                                        ; preds = %for.cond.for.end13_crit_edge, %entry
+  %sum.0.lcssa = phi float [ %phitmp, %for.cond.for.end13_crit_edge ], [ 0.000000e+00, %entry ]
+  store float %sum.0.lcssa, float* %r, align 4
+  ret void
+}
+
diff --git a/test/Transforms/LoopVectorize/PowerPC/small-loop-rdx.ll b/test/Transforms/LoopVectorize/PowerPC/small-loop-rdx.ll
new file mode 100644
index 000000000000..2898af2986d3
--- /dev/null
+++ b/test/Transforms/LoopVectorize/PowerPC/small-loop-rdx.ll
@@ -0,0 +1,47 @@
+; RUN: opt < %s -loop-vectorize -S | FileCheck %s
+
+; CHECK: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT: fadd
+; CHECK-NEXT-NOT: fadd
+
+target datalayout = "e-m:e-i64:64-n32:64"
+target triple = "powerpc64le-ibm-linux-gnu"
+
+define void @test(double* nocapture readonly %arr, i32 signext %len) #0 {
+entry:
+  %cmp4 = icmp sgt i32 %len, 0
+  br i1 %cmp4, label %for.body.lr.ph, label %for.end
+
+for.body.lr.ph:                                   ; preds = %entry
+  %0 = add i32 %len, -1
+  br label %for.body
+
+for.body:                                         ; preds = %for.body, %for.body.lr.ph
+  %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
+  %redx.05 = phi double [ 0.000000e+00, %for.body.lr.ph ], [ %add, %for.body ]
+  %arrayidx = getelementptr inbounds double, double* %arr, i64 %indvars.iv
+  %1 = load double, double* %arrayidx, align 8
+  %add = fadd fast double %1, %redx.05
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %0
+  br i1 %exitcond, label %for.end.loopexit, label %for.body
+
+for.end.loopexit:                                 ; preds = %for.body
+  %add.lcssa = phi double [ %add, %for.body ]
+  br label %for.end
+
+for.end:                                          ; preds = %for.end.loopexit, %entry
+  %redx.0.lcssa = phi double [ 0.000000e+00, %entry ], [ %add.lcssa, %for.end.loopexit ]
+  ret void
+}
diff --git a/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll b/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
index 6cd9c4d610b5..65b3919585e3 100644
--- a/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
+++ b/test/Transforms/LoopVectorize/PowerPC/vsx-tsvc-s173.ll
@@ -9,7 +9,7 @@ target triple = "powerpc64-unknown-linux-gnu"
 
 define signext i32 @s173() #0 {
 entry:
-  %0 = load i32* @ntimes, align 4
+  %0 = load i32, i32* @ntimes, align 4
   %cmp21 = icmp sgt i32 %0, 0
   br i1 %cmp21, label %for.cond1.preheader, label %for.end12
 
@@ -19,13 +19,13 @@ for.cond1.preheader:                              ; preds = %for.end, %entry
 
 for.body3:                                        ; preds = %for.body3, %for.cond1.preheader
   %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
-  %arrayidx = getelementptr inbounds %struct.GlobalData* @global_data, i64 0, i32 0, i64 %indvars.iv
-  %1 = load float* %arrayidx, align 4
-  %arrayidx5 = getelementptr inbounds %struct.GlobalData* @global_data, i64 0, i32 3, i64 %indvars.iv
-  %2 = load float* %arrayidx5, align 4
+  %arrayidx = getelementptr inbounds %struct.GlobalData, %struct.GlobalData* @global_data, i64 0, i32 0, i64 %indvars.iv
+  %1 = load float, float* %arrayidx, align 4
+  %arrayidx5 = getelementptr inbounds %struct.GlobalData, %struct.GlobalData* @global_data, i64 0, i32 3, i64 %indvars.iv
+  %2 = load float, float* %arrayidx5, align 4
   %add = fadd float %1, %2
   %3 = add nsw i64 %indvars.iv, 16000
-  %arrayidx8 = getelementptr inbounds %struct.GlobalData* @global_data, i64 0, i32 0, i64 %3
+  %arrayidx8 = getelementptr inbounds %struct.GlobalData, %struct.GlobalData* @global_data, i64 0, i32 0, i64 %3
   store float %add, float* %arrayidx8, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 16000
@@ -33,7 +33,7 @@ for.body3:                                        ; preds = %for.body3, %for.con
 
 for.end:                                          ; preds = %for.body3
   %inc11 = add nsw i32 %nl.022, 1
-  %4 = load i32* @ntimes, align 4
+  %4 = load i32, i32* @ntimes, align 4
   %mul = mul nsw i32 %4, 10
   %cmp = icmp slt i32 %inc11, %mul
   br i1 %cmp, label %for.cond1.preheader, label %for.end12
@@ -42,7 +42,7 @@ for.end12:                                        ; preds = %for.end, %entry
   ret i32 0
 
 ; CHECK-LABEL: @s173
-; CHECK: load <4 x float>*
+; CHECK: load <4 x float>, <4 x float>*
 ; CHECK: add i64 %index, 16000
 ; CHECK: ret i32 0
 }
diff --git a/test/Transforms/LoopVectorize/X86/already-vectorized.ll b/test/Transforms/LoopVectorize/X86/already-vectorized.ll
index 29d74a0645e8..c400c76461bb 100644
--- a/test/Transforms/LoopVectorize/X86/already-vectorized.ll
+++ b/test/Transforms/LoopVectorize/X86/already-vectorized.ll
@@ -21,8 +21,8 @@ for.body:                                         ; preds = %for.body, %entry
 ; CHECK: LV: We can vectorize this loop!
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %red.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds [255 x i32]* @a, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [255 x i32], [255 x i32]* @a, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %red.05
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 255
@@ -42,5 +42,6 @@ for.end:                                          ; preds = %for.body
 ; CHECK: [[vect]] = distinct !{[[vect]], [[width:![0-9]+]], [[unroll:![0-9]+]]}
 ; CHECK: [[width]] = !{!"llvm.loop.vectorize.width", i32 1}
 ; CHECK: [[unroll]] = !{!"llvm.loop.interleave.count", i32 1}
-; CHECK: [[scalar]] = distinct !{[[scalar]], [[width]], [[unroll]]}
+; CHECK: [[scalar]] = distinct !{[[scalar]], [[runtime_unroll:![0-9]+]], [[width]], [[unroll]]}
+; CHECK: [[runtime_unroll]] = !{!"llvm.loop.unroll.runtime.disable"}
 
diff --git a/test/Transforms/LoopVectorize/X86/assume.ll b/test/Transforms/LoopVectorize/X86/assume.ll
index a94e24dd7e9c..4fd378d1a0a9 100644
--- a/test/Transforms/LoopVectorize/X86/assume.ll
+++ b/test/Transforms/LoopVectorize/X86/assume.ll
@@ -22,12 +22,12 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp1 = fcmp ogt float %0, 1.000000e+02
   tail call void @llvm.assume(i1 %cmp1)
   %add = fadd float %0, 1.000000e+00
-  %arrayidx5 = getelementptr inbounds float* %a, i64 %indvars.iv
+  %arrayidx5 = getelementptr inbounds float, float* %a, i64 %indvars.iv
   store float %add, float* %arrayidx5, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv, 1599
@@ -48,13 +48,13 @@ attributes #1 = { nounwind }
 ; Function Attrs: nounwind uwtable
 define void @test2(%struct.data* nocapture readonly %d) #0 {
 entry:
-  %b = getelementptr inbounds %struct.data* %d, i64 0, i32 1
-  %0 = load float** %b, align 8
+  %b = getelementptr inbounds %struct.data, %struct.data* %d, i64 0, i32 1
+  %0 = load float*, float** %b, align 8
   %ptrint = ptrtoint float* %0 to i64
   %maskedptr = and i64 %ptrint, 31
   %maskcond = icmp eq i64 %maskedptr, 0
-  %a = getelementptr inbounds %struct.data* %d, i64 0, i32 0
-  %1 = load float** %a, align 8
+  %a = getelementptr inbounds %struct.data, %struct.data* %d, i64 0, i32 0
+  %1 = load float*, float** %a, align 8
   %ptrint2 = ptrtoint float* %1 to i64
   %maskedptr3 = and i64 %ptrint2, 31
   %maskcond4 = icmp eq i64 %maskedptr3, 0
@@ -84,11 +84,11 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   tail call void @llvm.assume(i1 %maskcond)
-  %arrayidx = getelementptr inbounds float* %0, i64 %indvars.iv
-  %2 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %0, i64 %indvars.iv
+  %2 = load float, float* %arrayidx, align 4
   %add = fadd float %2, 1.000000e+00
   tail call void @llvm.assume(i1 %maskcond4)
-  %arrayidx5 = getelementptr inbounds float* %1, i64 %indvars.iv
+  %arrayidx5 = getelementptr inbounds float, float* %1, i64 %indvars.iv
   store float %add, float* %arrayidx5, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv, 1599
diff --git a/test/Transforms/LoopVectorize/X86/avx1.ll b/test/Transforms/LoopVectorize/X86/avx1.ll
index 01c912567b61..37977c43ac30 100644
--- a/test/Transforms/LoopVectorize/X86/avx1.ll
+++ b/test/Transforms/LoopVectorize/X86/avx1.ll
@@ -12,8 +12,8 @@ define i32 @read_mod_write_single_ptr(float* nocapture %a, i32 %n) nounwind uwta
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds float* %a, i64 %indvars.iv
-  %3 = load float* %2, align 4
+  %2 = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %3 = load float, float* %2, align 4
   %4 = fmul float %3, 3.000000e+00
   store float %4, float* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -35,8 +35,8 @@ define i32 @read_mod_i64(i64* nocapture %a, i32 %n) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i64* %a, i64 %indvars.iv
-  %3 = load i64* %2, align 4
+  %2 = getelementptr inbounds i64, i64* %a, i64 %indvars.iv
+  %3 = load i64, i64* %2, align 4
   %4 = add i64 %3, 3
   store i64 %4, i64* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/X86/avx512.ll b/test/Transforms/LoopVectorize/X86/avx512.ll
index a2208668177d..754e859cd8fe 100644
--- a/test/Transforms/LoopVectorize/X86/avx512.ll
+++ b/test/Transforms/LoopVectorize/X86/avx512.ll
@@ -20,7 +20,7 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %n, i32* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/constant-vector-operand.ll b/test/Transforms/LoopVectorize/X86/constant-vector-operand.ll
index f4c07b4b24d9..d75b1d940ae2 100644
--- a/test/Transforms/LoopVectorize/X86/constant-vector-operand.ll
+++ b/test/Transforms/LoopVectorize/X86/constant-vector-operand.ll
@@ -15,10 +15,10 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @B, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @B, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %shl = ashr i32 %0, 3
-  %arrayidx2 = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
   store i32 %shl, i32* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/conversion-cost.ll b/test/Transforms/LoopVectorize/X86/conversion-cost.ll
index 0af562db8479..eb2a2a56fae1 100644
--- a/test/Transforms/LoopVectorize/X86/conversion-cost.ll
+++ b/test/Transforms/LoopVectorize/X86/conversion-cost.ll
@@ -13,7 +13,7 @@ define i32 @conversion_cost1(i32 %n, i8* nocapture %A, float* nocapture %B) noun
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 3, %0 ]
   %2 = trunc i64 %indvars.iv to i8
-  %3 = getelementptr inbounds i8* %A, i64 %indvars.iv
+  %3 = getelementptr inbounds i8, i8* %A, i64 %indvars.iv
   store i8 %2, i8* %3, align 1
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -35,7 +35,7 @@ define i32 @conversion_cost2(i32 %n, i8* nocapture %A, float* nocapture %B) noun
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 9, %0 ]
   %add = add nsw i64 %indvars.iv, 3
   %tofp = sitofp i64 %add to float
-  %gep = getelementptr inbounds float* %B, i64 %indvars.iv
+  %gep = getelementptr inbounds float, float* %B, i64 %indvars.iv
   store float %tofp, float* %gep, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/cost-model.ll b/test/Transforms/LoopVectorize/X86/cost-model.ll
index 98718e1e9708..013657102e6b 100644
--- a/test/Transforms/LoopVectorize/X86/cost-model.ll
+++ b/test/Transforms/LoopVectorize/X86/cost-model.ll
@@ -21,15 +21,15 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %0 = shl nsw i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %0
-  %1 = load i32* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %0
+  %1 = load i32, i32* %arrayidx, align 8
   %idxprom1 = sext i32 %1 to i64
-  %arrayidx2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %idxprom1
-  %2 = load i32* %arrayidx2, align 4
-  %arrayidx4 = getelementptr inbounds [2048 x i32]* @d, i64 0, i64 %indvars.iv
-  %3 = load i32* %arrayidx4, align 4
+  %arrayidx2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %idxprom1
+  %2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @d, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %arrayidx4, align 4
   %idxprom5 = sext i32 %3 to i64
-  %arrayidx6 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %idxprom5
+  %arrayidx6 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %idxprom5
   store i32 %2, i32* %arrayidx6, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/fp32_to_uint32-cost-model.ll b/test/Transforms/LoopVectorize/X86/fp32_to_uint32-cost-model.ll
index 529ed883c3b4..4a56d6b5ebf6 100644
--- a/test/Transforms/LoopVectorize/X86/fp32_to_uint32-cost-model.ll
+++ b/test/Transforms/LoopVectorize/X86/fp32_to_uint32-cost-model.ll
@@ -20,10 +20,10 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds [10000 x float]* @float_array, i64 0, i64 %indvars.iv
-  %1 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [10000 x float], [10000 x float]* @float_array, i64 0, i64 %indvars.iv
+  %1 = load float, float* %arrayidx, align 4
   %conv = fptoui float %1 to i32
-  %arrayidx2 = getelementptr inbounds [10000 x i32]* @unsigned_array, i64 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds [10000 x i32], [10000 x i32]* @unsigned_array, i64 0, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/fp64_to_uint32-cost-model.ll b/test/Transforms/LoopVectorize/X86/fp64_to_uint32-cost-model.ll
index ef3e3bec793a..c066afcfa63e 100644
--- a/test/Transforms/LoopVectorize/X86/fp64_to_uint32-cost-model.ll
+++ b/test/Transforms/LoopVectorize/X86/fp64_to_uint32-cost-model.ll
@@ -13,7 +13,7 @@ target triple = "x86_64-apple-macosx"
 
 define void @convert() {
 entry:
-  %0 = load i32* @n, align 4
+  %0 = load i32, i32* @n, align 4
   %cmp4 = icmp eq i32 %0, 0
   br i1 %cmp4, label %for.end, label %for.body.preheader
 
@@ -22,10 +22,10 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds [10000 x double]* @double_array, i64 0, i64 %indvars.iv
-  %1 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds [10000 x double], [10000 x double]* @double_array, i64 0, i64 %indvars.iv
+  %1 = load double, double* %arrayidx, align 8
   %conv = fptoui double %1 to i32
-  %arrayidx2 = getelementptr inbounds [10000 x i32]* @unsigned_array, i64 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds [10000 x i32], [10000 x i32]* @unsigned_array, i64 0, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %2 = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/fp_to_sint8-cost-model.ll b/test/Transforms/LoopVectorize/X86/fp_to_sint8-cost-model.ll
index 23e62275ce04..b3a0710545d7 100644
--- a/test/Transforms/LoopVectorize/X86/fp_to_sint8-cost-model.ll
+++ b/test/Transforms/LoopVectorize/X86/fp_to_sint8-cost-model.ll
@@ -11,10 +11,10 @@ entry:
   br label %for.body
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %tmp = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %tmp = load float, float* %arrayidx, align 4
   %conv = fptosi float %tmp to i8
-  %arrayidx2 = getelementptr inbounds i8* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i8, i8* %a, i64 %indvars.iv
   store i8 %conv, i8* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 256
diff --git a/test/Transforms/LoopVectorize/X86/gather-cost.ll b/test/Transforms/LoopVectorize/X86/gather-cost.ll
index 09363d65eefc..f0e6c8f205cf 100644
--- a/test/Transforms/LoopVectorize/X86/gather-cost.ll
+++ b/test/Transforms/LoopVectorize/X86/gather-cost.ll
@@ -31,32 +31,32 @@ for.body:
   %b.054 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add30, %for.body ]
   %add = add i64 %v.055, %offset
   %mul = mul i64 %add, 3
-  %arrayidx = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %mul
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds [512 x float]* @kernel, i64 0, i64 %v.055
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %mul
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds [512 x float], [512 x float]* @kernel, i64 0, i64 %v.055
+  %1 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %0, %1
-  %arrayidx4 = getelementptr inbounds [512 x float]* @kernel2, i64 0, i64 %v.055
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds [512 x float], [512 x float]* @kernel2, i64 0, i64 %v.055
+  %2 = load float, float* %arrayidx4, align 4
   %mul5 = fmul fast float %mul3, %2
-  %arrayidx6 = getelementptr inbounds [512 x float]* @kernel3, i64 0, i64 %v.055
-  %3 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds [512 x float], [512 x float]* @kernel3, i64 0, i64 %v.055
+  %3 = load float, float* %arrayidx6, align 4
   %mul7 = fmul fast float %mul5, %3
-  %arrayidx8 = getelementptr inbounds [512 x float]* @kernel4, i64 0, i64 %v.055
-  %4 = load float* %arrayidx8, align 4
+  %arrayidx8 = getelementptr inbounds [512 x float], [512 x float]* @kernel4, i64 0, i64 %v.055
+  %4 = load float, float* %arrayidx8, align 4
   %mul9 = fmul fast float %mul7, %4
   %add10 = fadd fast float %r.057, %mul9
   %arrayidx.sum = add i64 %mul, 1
-  %arrayidx11 = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum
-  %5 = load float* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum
+  %5 = load float, float* %arrayidx11, align 4
   %mul13 = fmul fast float %1, %5
   %mul15 = fmul fast float %2, %mul13
   %mul17 = fmul fast float %3, %mul15
   %mul19 = fmul fast float %4, %mul17
   %add20 = fadd fast float %g.056, %mul19
   %arrayidx.sum52 = add i64 %mul, 2
-  %arrayidx21 = getelementptr inbounds [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum52
-  %6 = load float* %arrayidx21, align 4
+  %arrayidx21 = getelementptr inbounds [1536 x float], [1536 x float]* @src_data, i64 0, i64 %arrayidx.sum52
+  %6 = load float, float* %arrayidx21, align 4
   %mul23 = fmul fast float %1, %6
   %mul25 = fmul fast float %2, %mul23
   %mul27 = fmul fast float %3, %mul25
diff --git a/test/Transforms/LoopVectorize/X86/gcc-examples.ll b/test/Transforms/LoopVectorize/X86/gcc-examples.ll
index 05403cd5b61b..c581f4bf2a63 100644
--- a/test/Transforms/LoopVectorize/X86/gcc-examples.ll
+++ b/test/Transforms/LoopVectorize/X86/gcc-examples.ll
@@ -28,12 +28,12 @@ define void @example1() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -61,10 +61,10 @@ define void @example10b(i16* noalias nocapture %sa, i16* noalias nocapture %sb,
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i16* %sb, i64 %indvars.iv
-  %3 = load i16* %2, align 2
+  %2 = getelementptr inbounds i16, i16* %sb, i64 %indvars.iv
+  %3 = load i16, i16* %2, align 2
   %4 = sext i16 %3 to i32
-  %5 = getelementptr inbounds i32* %ia, i64 %indvars.iv
+  %5 = getelementptr inbounds i32, i32* %ia, i64 %indvars.iv
   store i32 %4, i32* %5, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll b/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll
index 46efaf0e8531..cbba5300b9cb 100644
--- a/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll
+++ b/test/Transforms/LoopVectorize/X86/illegal-parallel-loop-uniform-write.ll
@@ -14,8 +14,8 @@ entry:
   br i1 %cmp27, label %for.body3.lr.ph.us, label %for.end15
 
 for.end.us:                                       ; preds = %for.body3.us
-  %arrayidx9.us = getelementptr inbounds i32* %b, i64 %indvars.iv33
-  %0 = load i32* %arrayidx9.us, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx9.us = getelementptr inbounds i32, i32* %b, i64 %indvars.iv33
+  %0 = load i32, i32* %arrayidx9.us, align 4, !llvm.mem.parallel_loop_access !3
   %add10.us = add nsw i32 %0, 3
   store i32 %add10.us, i32* %arrayidx9.us, align 4, !llvm.mem.parallel_loop_access !3
   %indvars.iv.next34 = add i64 %indvars.iv33, 1
@@ -28,8 +28,8 @@ for.body3.us:                                     ; preds = %for.body3.us, %for.
   %1 = trunc i64 %indvars.iv29 to i32
   %add4.us = add i32 %add.us, %1
   %idxprom.us = sext i32 %add4.us to i64
-  %arrayidx.us = getelementptr inbounds i32* %a, i64 %idxprom.us
-  %2 = load i32* %arrayidx.us, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx.us = getelementptr inbounds i32, i32* %a, i64 %idxprom.us
+  %2 = load i32, i32* %arrayidx.us, align 4, !llvm.mem.parallel_loop_access !3
   %add5.us = add nsw i32 %2, 1
   store i32 %add5.us, i32* %arrayidx7.us, align 4, !llvm.mem.parallel_loop_access !3
   %indvars.iv.next30 = add i64 %indvars.iv29, 1
@@ -41,7 +41,7 @@ for.body3.lr.ph.us:                               ; preds = %for.end.us, %entry
   %indvars.iv33 = phi i64 [ %indvars.iv.next34, %for.end.us ], [ 0, %entry ]
   %3 = trunc i64 %indvars.iv33 to i32
   %add.us = add i32 %3, %k
-  %arrayidx7.us = getelementptr inbounds i32* %a, i64 %indvars.iv33
+  %arrayidx7.us = getelementptr inbounds i32, i32* %a, i64 %indvars.iv33
   br label %for.body3.us
 
 for.end15:                                        ; preds = %for.end.us, %entry
diff --git a/test/Transforms/LoopVectorize/X86/masked_load_store.ll b/test/Transforms/LoopVectorize/X86/masked_load_store.ll
index 9e2de8000587..8c375ccfd315 100644
--- a/test/Transforms/LoopVectorize/X86/masked_load_store.ll
+++ b/test/Transforms/LoopVectorize/X86/masked_load_store.ll
@@ -46,35 +46,35 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp slt i32 %3, 100
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %idxprom2 = sext i32 %4 to i64
-  %5 = load i32** %B.addr, align 8
-  %arrayidx3 = getelementptr inbounds i32* %5, i64 %idxprom2
-  %6 = load i32* %arrayidx3, align 4
-  %7 = load i32* %i, align 4
+  %5 = load i32*, i32** %B.addr, align 8
+  %arrayidx3 = getelementptr inbounds i32, i32* %5, i64 %idxprom2
+  %6 = load i32, i32* %arrayidx3, align 4
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load i32** %trigger.addr, align 8
-  %arrayidx5 = getelementptr inbounds i32* %8, i64 %idxprom4
-  %9 = load i32* %arrayidx5, align 4
+  %8 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %8, i64 %idxprom4
+  %9 = load i32, i32* %arrayidx5, align 4
   %add = add nsw i32 %6, %9
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %idxprom6 = sext i32 %10 to i64
-  %11 = load i32** %A.addr, align 8
-  %arrayidx7 = getelementptr inbounds i32* %11, i64 %idxprom6
+  %11 = load i32*, i32** %A.addr, align 8
+  %arrayidx7 = getelementptr inbounds i32, i32* %11, i64 %idxprom6
   store i32 %add, i32* %arrayidx7, align 4
   br label %if.end
 
@@ -82,7 +82,7 @@ if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %12 = load i32* %i, align 4
+  %12 = load i32, i32* %i, align 4
   %inc = add nsw i32 %12, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -130,36 +130,36 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp slt i32 %3, 100
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %idxprom2 = sext i32 %4 to i64
-  %5 = load float** %B.addr, align 8
-  %arrayidx3 = getelementptr inbounds float* %5, i64 %idxprom2
-  %6 = load float* %arrayidx3, align 4
-  %7 = load i32* %i, align 4
+  %5 = load float*, float** %B.addr, align 8
+  %arrayidx3 = getelementptr inbounds float, float* %5, i64 %idxprom2
+  %6 = load float, float* %arrayidx3, align 4
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load i32** %trigger.addr, align 8
-  %arrayidx5 = getelementptr inbounds i32* %8, i64 %idxprom4
-  %9 = load i32* %arrayidx5, align 4
+  %8 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %8, i64 %idxprom4
+  %9 = load i32, i32* %arrayidx5, align 4
   %conv = sitofp i32 %9 to float
   %add = fadd float %6, %conv
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %idxprom6 = sext i32 %10 to i64
-  %11 = load float** %A.addr, align 8
-  %arrayidx7 = getelementptr inbounds float* %11, i64 %idxprom6
+  %11 = load float*, float** %A.addr, align 8
+  %arrayidx7 = getelementptr inbounds float, float* %11, i64 %idxprom6
   store float %add, float* %arrayidx7, align 4
   br label %if.end
 
@@ -167,7 +167,7 @@ if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %12 = load i32* %i, align 4
+  %12 = load i32, i32* %i, align 4
   %inc = add nsw i32 %12, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -218,36 +218,36 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp slt i32 %3, 100
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %idxprom2 = sext i32 %4 to i64
-  %5 = load double** %B.addr, align 8
-  %arrayidx3 = getelementptr inbounds double* %5, i64 %idxprom2
-  %6 = load double* %arrayidx3, align 8
-  %7 = load i32* %i, align 4
+  %5 = load double*, double** %B.addr, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %5, i64 %idxprom2
+  %6 = load double, double* %arrayidx3, align 8
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load i32** %trigger.addr, align 8
-  %arrayidx5 = getelementptr inbounds i32* %8, i64 %idxprom4
-  %9 = load i32* %arrayidx5, align 4
+  %8 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %8, i64 %idxprom4
+  %9 = load i32, i32* %arrayidx5, align 4
   %conv = sitofp i32 %9 to double
   %add = fadd double %6, %conv
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %idxprom6 = sext i32 %10 to i64
-  %11 = load double** %A.addr, align 8
-  %arrayidx7 = getelementptr inbounds double* %11, i64 %idxprom6
+  %11 = load double*, double** %A.addr, align 8
+  %arrayidx7 = getelementptr inbounds double, double* %11, i64 %idxprom6
   store double %add, double* %arrayidx7, align 8
   br label %if.end
 
@@ -255,7 +255,7 @@ if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %12 = load i32* %i, align 4
+  %12 = load i32, i32* %i, align 4
   %inc = add nsw i32 %12, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -297,37 +297,37 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp slt i32 %3, 100
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %mul = mul nsw i32 %4, 2
   %idxprom2 = sext i32 %mul to i64
-  %5 = load double** %B.addr, align 8
-  %arrayidx3 = getelementptr inbounds double* %5, i64 %idxprom2
-  %6 = load double* %arrayidx3, align 8
-  %7 = load i32* %i, align 4
+  %5 = load double*, double** %B.addr, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %5, i64 %idxprom2
+  %6 = load double, double* %arrayidx3, align 8
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load i32** %trigger.addr, align 8
-  %arrayidx5 = getelementptr inbounds i32* %8, i64 %idxprom4
-  %9 = load i32* %arrayidx5, align 4
+  %8 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %8, i64 %idxprom4
+  %9 = load i32, i32* %arrayidx5, align 4
   %conv = sitofp i32 %9 to double
   %add = fadd double %6, %conv
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %idxprom6 = sext i32 %10 to i64
-  %11 = load double** %A.addr, align 8
-  %arrayidx7 = getelementptr inbounds double* %11, i64 %idxprom6
+  %11 = load double*, double** %A.addr, align 8
+  %arrayidx7 = getelementptr inbounds double, double* %11, i64 %idxprom6
   store double %add, double* %arrayidx7, align 8
   br label %if.end
 
@@ -335,7 +335,7 @@ if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %12 = load i32* %i, align 4
+  %12 = load i32, i32* %i, align 4
   %inc = add nsw i32 %12, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -373,43 +373,43 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 10000
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp slt i32 %3, 100
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %idxprom2 = sext i32 %4 to i64
-  %5 = load i32** %B.addr, align 8
-  %arrayidx3 = getelementptr inbounds i32* %5, i64 %idxprom2
-  %6 = load i32* %arrayidx3, align 4
-  %7 = load i32* %i, align 4
+  %5 = load i32*, i32** %B.addr, align 8
+  %arrayidx3 = getelementptr inbounds i32, i32* %5, i64 %idxprom2
+  %6 = load i32, i32* %arrayidx3, align 4
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load i32** %trigger.addr, align 8
-  %arrayidx5 = getelementptr inbounds i32* %8, i64 %idxprom4
-  %9 = load i32* %arrayidx5, align 4
+  %8 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx5 = getelementptr inbounds i32, i32* %8, i64 %idxprom4
+  %9 = load i32, i32* %arrayidx5, align 4
   %add = add nsw i32 %6, %9
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %idxprom6 = sext i32 %10 to i64
-  %11 = load i32** %A.addr, align 8
-  %arrayidx7 = getelementptr inbounds i32* %11, i64 %idxprom6
-  store i32 sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*]* @a, i64 0, i64 1), i32** @c) to i32)), i32* %arrayidx7, align 4
+  %11 = load i32*, i32** %A.addr, align 8
+  %arrayidx7 = getelementptr inbounds i32, i32* %11, i64 %idxprom6
+  store i32 sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*], [1 x i32*]* @a, i64 0, i64 1), i32** @c) to i32)), i32* %arrayidx7, align 4
   br label %if.end
 
 if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %12 = load i32* %i, align 4
+  %12 = load i32, i32* %i, align 4
   %inc = add nsw i32 %12, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
@@ -459,30 +459,30 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp sge i32 %0, 0
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %idxprom = sext i32 %1 to i64
-  %2 = load i32** %trigger.addr, align 8
-  %arrayidx = getelementptr inbounds i32* %2, i64 %idxprom
-  %3 = load i32* %arrayidx, align 4
+  %2 = load i32*, i32** %trigger.addr, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %2, i64 %idxprom
+  %3 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %3, 0
   br i1 %cmp1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %for.body
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %idxprom2 = sext i32 %4 to i64
-  %5 = load double** %in.addr, align 8
-  %arrayidx3 = getelementptr inbounds double* %5, i64 %idxprom2
-  %6 = load double* %arrayidx3, align 8
+  %5 = load double*, double** %in.addr, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %5, i64 %idxprom2
+  %6 = load double, double* %arrayidx3, align 8
   %add = fadd double %6, 5.000000e-01
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %idxprom4 = sext i32 %7 to i64
-  %8 = load double** %out.addr, align 8
-  %arrayidx5 = getelementptr inbounds double* %8, i64 %idxprom4
+  %8 = load double*, double** %out.addr, align 8
+  %arrayidx5 = getelementptr inbounds double, double* %8, i64 %idxprom4
   store double %add, double* %arrayidx5, align 8
   br label %if.end
 
@@ -490,7 +490,7 @@ if.end:                                           ; preds = %if.then, %for.body
   br label %for.inc
 
 for.inc:                                          ; preds = %if.end
-  %9 = load i32* %i, align 4
+  %9 = load i32, i32* %i, align 4
   %dec = add nsw i32 %9, -1
   store i32 %dec, i32* %i, align 4
   br label %for.cond
diff --git a/test/Transforms/LoopVectorize/X86/metadata-enable.ll b/test/Transforms/LoopVectorize/X86/metadata-enable.ll
index 7feb66ce6a98..ba8e11e58749 100644
--- a/test/Transforms/LoopVectorize/X86/metadata-enable.ll
+++ b/test/Transforms/LoopVectorize/X86/metadata-enable.ll
@@ -54,17 +54,17 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %N
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %add, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 32
   br i1 %exitcond, label %for.end, label %for.body, !llvm.loop !0
 
 for.end:                                          ; preds = %for.body
-  %1 = load i32* %a, align 4
+  %1 = load i32, i32* %a, align 4
   ret i32 %1
 }
 
@@ -105,17 +105,17 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %N
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %add, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 32
   br i1 %exitcond, label %for.end, label %for.body
 
 for.end:                                          ; preds = %for.body
-  %1 = load i32* %a, align 4
+  %1 = load i32, i32* %a, align 4
   ret i32 %1
 }
 
@@ -156,17 +156,17 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %N
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %add, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 32
   br i1 %exitcond, label %for.end, label %for.body, !llvm.loop !2
 
 for.end:                                          ; preds = %for.body
-  %1 = load i32* %a, align 4
+  %1 = load i32, i32* %a, align 4
   ret i32 %1
 }
 
diff --git a/test/Transforms/LoopVectorize/X86/min-trip-count-switch.ll b/test/Transforms/LoopVectorize/X86/min-trip-count-switch.ll
index fd69dc46070e..bb972c4488a4 100644
--- a/test/Transforms/LoopVectorize/X86/min-trip-count-switch.ll
+++ b/test/Transforms/LoopVectorize/X86/min-trip-count-switch.ll
@@ -10,8 +10,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %a, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %add = fadd float %0, 1.000000e+00
   store float %add, float* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/X86/no-vector.ll b/test/Transforms/LoopVectorize/X86/no-vector.ll
index 692eec989591..4b464b012675 100644
--- a/test/Transforms/LoopVectorize/X86/no-vector.ll
+++ b/test/Transforms/LoopVectorize/X86/no-vector.ll
@@ -8,8 +8,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.06 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %r.05 = phi i32 [ %xor, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i8* %s, i32 %i.06
-  %0 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %s, i32 %i.06
+  %0 = load i8, i8* %arrayidx, align 1
   %conv = sext i8 %0 to i32
   %xor = xor i32 %conv, %r.05
   %inc = add nsw i32 %i.06, 1
diff --git a/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll b/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll
index ad010443a5e0..631361c9d47d 100644
--- a/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll
+++ b/test/Transforms/LoopVectorize/X86/parallel-loops-after-reg2mem.ll
@@ -17,28 +17,28 @@ entry:
   br label %for.body
 
 for.body:                                         ; preds = %for.body.for.body_crit_edge, %entry
-  %indvars.iv.reload = load i64* %indvars.iv.reg2mem
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv.reload
-  %0 = load i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv.reload
-  %1 = load i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
+  %indvars.iv.reload = load i64, i64* %indvars.iv.reg2mem
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.reload
+  %0 = load i32, i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv.reload
+  %1 = load i32, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
   %idxprom3 = sext i32 %1 to i64
-  %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i64 %idxprom3
   store i32 %0, i32* %arrayidx4, align 4, !llvm.mem.parallel_loop_access !3
   %indvars.iv.next = add i64 %indvars.iv.reload, 1
   ; A new store without the parallel metadata here:
   store i64 %indvars.iv.next, i64* %indvars.iv.next.reg2mem
-  %indvars.iv.next.reload1 = load i64* %indvars.iv.next.reg2mem
-  %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next.reload1
-  %2 = load i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !3
+  %indvars.iv.next.reload1 = load i64, i64* %indvars.iv.next.reg2mem
+  %arrayidx6 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.next.reload1
+  %2 = load i32, i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !3
   store i32 %2, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
-  %indvars.iv.next.reload = load i64* %indvars.iv.next.reg2mem
+  %indvars.iv.next.reload = load i64, i64* %indvars.iv.next.reg2mem
   %lftr.wideiv = trunc i64 %indvars.iv.next.reload to i32
   %exitcond = icmp eq i32 %lftr.wideiv, 512
   br i1 %exitcond, label %for.end, label %for.body.for.body_crit_edge, !llvm.loop !3
 
 for.body.for.body_crit_edge:                      ; preds = %for.body
-  %indvars.iv.next.reload2 = load i64* %indvars.iv.next.reg2mem
+  %indvars.iv.next.reload2 = load i64, i64* %indvars.iv.next.reg2mem
   store i64 %indvars.iv.next.reload2, i64* %indvars.iv.reg2mem
   br label %for.body
 
diff --git a/test/Transforms/LoopVectorize/X86/parallel-loops.ll b/test/Transforms/LoopVectorize/X86/parallel-loops.ll
index 22ab52157498..53061edc35c5 100644
--- a/test/Transforms/LoopVectorize/X86/parallel-loops.ll
+++ b/test/Transforms/LoopVectorize/X86/parallel-loops.ll
@@ -20,16 +20,16 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %idxprom3 = sext i32 %1 to i64
-  %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i64 %idxprom3
   store i32 %0, i32* %arrayidx4, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
-  %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next
-  %2 = load i32* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.next
+  %2 = load i32, i32* %arrayidx6, align 4
   store i32 %2, i32* %arrayidx2, align 4
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, 512
@@ -50,18 +50,18 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
   %idxprom3 = sext i32 %1 to i64
-  %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i64 %idxprom3
   ; This store might have originated from inlining a function with a parallel
   ; loop. Refers to a list with the "original loop reference" (!4) also included.
   store i32 %0, i32* %arrayidx4, align 4, !llvm.mem.parallel_loop_access !5
   %indvars.iv.next = add i64 %indvars.iv, 1
-  %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next
-  %2 = load i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx6 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.next
+  %2 = load i32, i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !3
   store i32 %2, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, 512
@@ -83,18 +83,18 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !6
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !6
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4, !llvm.mem.parallel_loop_access !6
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !6
   %idxprom3 = sext i32 %1 to i64
-  %arrayidx4 = getelementptr inbounds i32* %a, i64 %idxprom3
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i64 %idxprom3
   ; This refers to the loop marked with !7 which we are not in at the moment.
   ; It should prevent detecting as a parallel loop.
   store i32 %0, i32* %arrayidx4, align 4, !llvm.mem.parallel_loop_access !7
   %indvars.iv.next = add i64 %indvars.iv, 1
-  %arrayidx6 = getelementptr inbounds i32* %b, i64 %indvars.iv.next
-  %2 = load i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !6
+  %arrayidx6 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.next
+  %2 = load i32, i32* %arrayidx6, align 4, !llvm.mem.parallel_loop_access !6
   store i32 %2, i32* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !6
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp eq i32 %lftr.wideiv, 512
diff --git a/test/Transforms/LoopVectorize/X86/powof2div.ll b/test/Transforms/LoopVectorize/X86/powof2div.ll
index 054da8ed20ae..6bc738a7d143 100644
--- a/test/Transforms/LoopVectorize/X86/powof2div.ll
+++ b/test/Transforms/LoopVectorize/X86/powof2div.ll
@@ -7,7 +7,7 @@ target triple = "x86_64-unknown-linux-gnu"
 @Foo = common global %struct.anon zeroinitializer, align 4
 
 ;CHECK-LABEL: @foo(
-;CHECK: load <4 x i32>*
+;CHECK: load <4 x i32>, <4 x i32>*
 ;CHECK: sdiv <4 x i32>
 ;CHECK: store <4 x i32>
 
@@ -17,10 +17,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds %struct.anon* @Foo, i64 0, i32 2, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds %struct.anon, %struct.anon* @Foo, i64 0, i32 2, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %div = sdiv i32 %0, 2
-  %arrayidx2 = getelementptr inbounds %struct.anon* @Foo, i64 0, i32 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds %struct.anon, %struct.anon* @Foo, i64 0, i32 0, i64 %indvars.iv
   store i32 %div, i32* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 100
diff --git a/test/Transforms/LoopVectorize/X86/reduction-crash.ll b/test/Transforms/LoopVectorize/X86/reduction-crash.ll
index 3957a5541422..3741b95d9859 100644
--- a/test/Transforms/LoopVectorize/X86/reduction-crash.ll
+++ b/test/Transforms/LoopVectorize/X86/reduction-crash.ll
@@ -14,17 +14,17 @@ bb:
   br label %bb2
 
 bb2:                                              ; preds = %bb
-  %tmp = load double* null, align 8
+  %tmp = load double, double* null, align 8
   br i1 undef, label %bb3, label %bb12
 
 bb3:                                              ; preds = %bb3, %bb2
   %tmp4 = phi double [ %tmp9, %bb3 ], [ %tmp, %bb2 ]
   %tmp5 = phi i32 [ %tmp8, %bb3 ], [ 0, %bb2 ]
-  %tmp6 = getelementptr inbounds [16 x double]* undef, i32 0, i32 %tmp5
-  %tmp7 = load double* %tmp6, align 4
+  %tmp6 = getelementptr inbounds [16 x double], [16 x double]* undef, i32 0, i32 %tmp5
+  %tmp7 = load double, double* %tmp6, align 4
   %tmp8 = add nsw i32 %tmp5, 1
   %tmp9 = fadd fast double %tmp4, undef
-  %tmp10 = getelementptr inbounds float* %arg, i32 %tmp5
+  %tmp10 = getelementptr inbounds float, float* %arg, i32 %tmp5
   store float undef, float* %tmp10, align 4
   %tmp11 = icmp eq i32 %tmp8, %arg1
   br i1 %tmp11, label %bb12, label %bb3
diff --git a/test/Transforms/LoopVectorize/X86/small-size.ll b/test/Transforms/LoopVectorize/X86/small-size.ll
index 8c7a881be46f..47c262b11b46 100644
--- a/test/Transforms/LoopVectorize/X86/small-size.ll
+++ b/test/Transforms/LoopVectorize/X86/small-size.ll
@@ -30,12 +30,12 @@ define void @example1() optsize {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -65,7 +65,7 @@ define void @example2(i32 %n, i32 %x) optsize {
 
 .lr.ph5:                                          ; preds = %0, %.lr.ph5
   %indvars.iv6 = phi i64 [ %indvars.iv.next7, %.lr.ph5 ], [ 0, %0 ]
-  %3 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv6
+  %3 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv6
   store i32 %x, i32* %3, align 4
   %indvars.iv.next7 = add i64 %indvars.iv6, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next7 to i32
@@ -76,12 +76,12 @@ define void @example2(i32 %n, i32 %x) optsize {
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ %i.0.lcssa, %.preheader ]
   %.02 = phi i32 [ %4, %.lr.ph ], [ %n, %.preheader ]
   %4 = add nsw i32 %.02, -1
-  %5 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %6 = load i32* %5, align 4
-  %7 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %8 = load i32* %7, align 4
+  %5 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %6 = load i32, i32* %5, align 4
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %8 = load i32, i32* %7, align 4
   %9 = and i32 %8, %6
-  %10 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %10 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %9, i32* %10, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %11 = icmp eq i32 %4, 0
@@ -104,9 +104,9 @@ define void @example3(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
   %.014 = phi i32* [ %5, %.lr.ph ], [ %p, %0 ]
   %.023 = phi i32* [ %3, %.lr.ph ], [ %q, %0 ]
   %2 = add nsw i32 %.05, -1
-  %3 = getelementptr inbounds i32* %.023, i64 1
-  %4 = load i32* %.023, align 16
-  %5 = getelementptr inbounds i32* %.014, i64 1
+  %3 = getelementptr inbounds i32, i32* %.023, i64 1
+  %4 = load i32, i32* %.023, align 16
+  %5 = getelementptr inbounds i32, i32* %.014, i64 1
   store i32 %4, i32* %.014, align 16
   %6 = icmp eq i32 %2, 0
   br i1 %6, label %._crit_edge, label %.lr.ph
@@ -128,9 +128,9 @@ define void @example4(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
   %.014 = phi i32* [ %5, %.lr.ph ], [ %p, %0 ]
   %.023 = phi i32* [ %3, %.lr.ph ], [ %q, %0 ]
   %2 = add nsw i32 %.05, -1
-  %3 = getelementptr inbounds i32* %.023, i64 1
-  %4 = load i32* %.023, align 16
-  %5 = getelementptr inbounds i32* %.014, i64 1
+  %3 = getelementptr inbounds i32, i32* %.023, i64 1
+  %4 = load i32, i32* %.023, align 16
+  %5 = getelementptr inbounds i32, i32* %.014, i64 1
   store i32 %4, i32* %.014, align 16
   %6 = icmp eq i32 %2, 0
   br i1 %6, label %._crit_edge, label %.lr.ph
@@ -152,11 +152,11 @@ define void @example23(i16* nocapture %src, i32* nocapture %dst) optsize {
   %.04 = phi i16* [ %src, %0 ], [ %2, %1 ]
   %.013 = phi i32* [ %dst, %0 ], [ %6, %1 ]
   %i.02 = phi i32 [ 0, %0 ], [ %7, %1 ]
-  %2 = getelementptr inbounds i16* %.04, i64 1
-  %3 = load i16* %.04, align 2
+  %2 = getelementptr inbounds i16, i16* %.04, i64 1
+  %3 = load i16, i16* %.04, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw nsw i32 %4, 7
-  %6 = getelementptr inbounds i32* %.013, i64 1
+  %6 = getelementptr inbounds i32, i32* %.013, i64 1
   store i32 %5, i32* %.013, align 4
   %7 = add nsw i32 %i.02, 1
   %exitcond = icmp eq i32 %7, 256
@@ -178,11 +178,11 @@ define void @example23b(i16* noalias nocapture %src, i32* noalias nocapture %dst
   %.04 = phi i16* [ %src, %0 ], [ %2, %1 ]
   %.013 = phi i32* [ %dst, %0 ], [ %6, %1 ]
   %i.02 = phi i32 [ 0, %0 ], [ %7, %1 ]
-  %2 = getelementptr inbounds i16* %.04, i64 1
-  %3 = load i16* %.04, align 2
+  %2 = getelementptr inbounds i16, i16* %.04, i64 1
+  %3 = load i16, i16* %.04, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw nsw i32 %4, 7
-  %6 = getelementptr inbounds i32* %.013, i64 1
+  %6 = getelementptr inbounds i32, i32* %.013, i64 1
   store i32 %5, i32* %.013, align 4
   %7 = add nsw i32 %i.02, 1
   %exitcond = icmp eq i32 %7, 256
diff --git a/test/Transforms/LoopVectorize/X86/struct-store.ll b/test/Transforms/LoopVectorize/X86/struct-store.ll
index a995e43a5ab1..4ff3b0e4c05d 100644
--- a/test/Transforms/LoopVectorize/X86/struct-store.ll
+++ b/test/Transforms/LoopVectorize/X86/struct-store.ll
@@ -15,7 +15,7 @@ entry:
 
 loop:
   %indvars.iv = phi i64 [ %indvars.iv.next, %loop ], [ 0, %entry ]
-  %tmp = getelementptr inbounds [16 x { i64, i64 }]* @glbl, i64 0, i64 %indvars.iv
+  %tmp = getelementptr inbounds [16 x { i64, i64 }], [16 x { i64, i64 }]* @glbl, i64 0, i64 %indvars.iv
   store { i64, i64 } { i64 ptrtoint (void ()* @fn to i64), i64 0 }, { i64, i64 }* %tmp, align 16
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/X86/tripcount.ll b/test/Transforms/LoopVectorize/X86/tripcount.ll
index a4ec6948a2f0..c0bbb92c2c53 100644
--- a/test/Transforms/LoopVectorize/X86/tripcount.ll
+++ b/test/Transforms/LoopVectorize/X86/tripcount.ll
@@ -22,8 +22,8 @@ for.body.preheader:
 
 for.body:
   %i.07 = phi i32 [ %inc, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds [0 x i32]* @big, i32 0, i32 %i.07
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [0 x i32], [0 x i32]* @big, i32 0, i32 %i.07
+  %0 = load i32, i32* %arrayidx, align 4
   %neg = xor i32 %0, -1
   store i32 %neg, i32* %arrayidx, align 4
   %inc = add nsw i32 %i.07, 1
diff --git a/test/Transforms/LoopVectorize/X86/uint64_to_fp64-cost-model.ll b/test/Transforms/LoopVectorize/X86/uint64_to_fp64-cost-model.ll
index 86c32b2d2ee4..38af11c443d0 100644
--- a/test/Transforms/LoopVectorize/X86/uint64_to_fp64-cost-model.ll
+++ b/test/Transforms/LoopVectorize/X86/uint64_to_fp64-cost-model.ll
@@ -12,10 +12,10 @@ entry:
   br label %for.body
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i64* %a, i64 %indvars.iv
-  %tmp = load i64* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i64, i64* %a, i64 %indvars.iv
+  %tmp = load i64, i64* %arrayidx, align 4
   %conv = uitofp i64 %tmp to double
-  %arrayidx2 = getelementptr inbounds double* %b, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %b, i64 %indvars.iv
   store double %conv, double* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 256
diff --git a/test/Transforms/LoopVectorize/X86/unroll-pm.ll b/test/Transforms/LoopVectorize/X86/unroll-pm.ll
index 5064fec286ce..52914b6a7c67 100644
--- a/test/Transforms/LoopVectorize/X86/unroll-pm.ll
+++ b/test/Transforms/LoopVectorize/X86/unroll-pm.ll
@@ -17,8 +17,8 @@ define i32 @bar(i32* nocapture %A, i32 %n) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i32 %3, 6
   store i32 %4, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/X86/unroll-small-loops.ll b/test/Transforms/LoopVectorize/X86/unroll-small-loops.ll
index 716dc08a55ac..69d2a319a8ca 100644
--- a/test/Transforms/LoopVectorize/X86/unroll-small-loops.ll
+++ b/test/Transforms/LoopVectorize/X86/unroll-small-loops.ll
@@ -16,8 +16,8 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK-VECTOR: ret
 ;
 ; CHECK-SCALAR-LABEL: @foo(
-; CHECK-SCALAR: load i32*
-; CHECK-SCALAR-NOT: load i32*
+; CHECK-SCALAR: load i32, i32*
+; CHECK-SCALAR-NOT: load i32, i32*
 ; CHECK-SCALAR: store i32
 ; CHECK-SCALAR-NOT: store i32
 ; CHECK-SCALAR: ret
@@ -26,8 +26,8 @@ define i32 @foo(i32* nocapture %A) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i32 %3, 6
   store i32 %4, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -47,9 +47,11 @@ define i32 @foo(i32* nocapture %A) nounwind uwtable ssp {
 ; CHECK-VECTOR: store <4 x i32>
 ; CHECK-VECTOR: ret
 ;
+; For x86, loop unroll in loop vectorizer is disabled when VF==1.
+;
 ; CHECK-SCALAR-LABEL: @bar(
 ; CHECK-SCALAR: store i32
-; CHECK-SCALAR: store i32
+; CHECK-SCALAR-NOT: store i32
 ; CHECK-SCALAR: ret
 define i32 @bar(i32* nocapture %A, i32 %n) nounwind uwtable ssp {
   %1 = icmp sgt i32 %n, 0
@@ -57,8 +59,8 @@ define i32 @bar(i32* nocapture %A, i32 %n) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i32 %3, 6
   store i32 %4, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -86,10 +88,10 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %mul = fmul float %0, %N
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
   store float %mul, float* %arrayidx2, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 256
diff --git a/test/Transforms/LoopVectorize/X86/unroll_selection.ll b/test/Transforms/LoopVectorize/X86/unroll_selection.ll
index c684b4e79b14..71b829071e22 100644
--- a/test/Transforms/LoopVectorize/X86/unroll_selection.ll
+++ b/test/Transforms/LoopVectorize/X86/unroll_selection.ll
@@ -16,8 +16,8 @@ define void @reg_pressure(double* nocapture %A, i32 %n) nounwind uwtable ssp {
 
 ; <label>:2                                       ; preds = %2, %0
   %indvars.iv = phi i64 [ %indvars.iv.next, %2 ], [ %1, %0 ]
-  %3 = getelementptr inbounds double* %A, i64 %indvars.iv
-  %4 = load double* %3, align 8
+  %3 = getelementptr inbounds double, double* %A, i64 %indvars.iv
+  %4 = load double, double* %3, align 8
   %5 = fadd double %4, 3.000000e+00
   %6 = fmul double %4, 2.000000e+00
   %7 = fadd double %5, %6
@@ -58,8 +58,8 @@ define void @small_loop(i16* nocapture %A, i64 %n) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %i.01 = phi i64 [ %5, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i16* %A, i64 %i.01
-  %3 = load i16* %2, align 2
+  %2 = getelementptr inbounds i16, i16* %A, i64 %i.01
+  %3 = load i16, i16* %2, align 2
   %4 = xor i16 %3, 3
   store i16 %4, i16* %2, align 2
   %5 = add i64 %i.01, 1
diff --git a/test/Transforms/LoopVectorize/X86/veclib-calls.ll b/test/Transforms/LoopVectorize/X86/veclib-calls.ll
new file mode 100644
index 000000000000..6f8f5223ce45
--- /dev/null
+++ b/test/Transforms/LoopVectorize/X86/veclib-calls.ll
@@ -0,0 +1,632 @@
+; RUN: opt < %s -vector-library=Accelerate -loop-vectorize -S | FileCheck %s
+
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+;CHECK-LABEL: @sqrt_f32(
+;CHECK: vsqrtf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @sqrtf(float) nounwind readnone
+define void @sqrt_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @sqrtf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @exp_f32(
+;CHECK: vexpf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @expf(float) nounwind readnone
+define void @exp_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @expf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @log_f32(
+;CHECK: vlogf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @logf(float) nounwind readnone
+define void @log_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @logf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+; For abs instruction we'll generate vector intrinsic, as it's cheaper than a lib call.
+;CHECK-LABEL: @fabs_f32(
+;CHECK: fabs{{.*}}<4 x float>
+;CHECK: ret void
+declare float @fabsf(float) nounwind readnone
+define void @fabs_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @fabsf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+; Test that we can vectorize an intrinsic into a vector call.
+;CHECK-LABEL: @exp_f32_intrin(
+;CHECK: vexpf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @llvm.exp.f32(float) nounwind readnone
+define void @exp_f32_intrin(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @llvm.exp.f32(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+; Test that we don't vectorize arbitrary functions.
+;CHECK-LABEL: @foo_f32(
+;CHECK-NOT: foo{{.*}}<4 x float>
+;CHECK: ret void
+declare float @foo(float) nounwind readnone
+define void @foo_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @foo(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+; Test that we don't vectorize calls with nobuiltin attribute.
+;CHECK-LABEL: @sqrt_f32_nobuiltin(
+;CHECK-NOT: vsqrtf{{.*}}<4 x float>
+;CHECK: ret void
+define void @sqrt_f32_nobuiltin(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @sqrtf(float %0) nounwind readnone nobuiltin
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @ceil_f32(
+;CHECK: vceilf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @ceilf(float) nounwind readnone
+define void @ceil_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @ceilf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @floor_f32(
+;CHECK: vfloorf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @floorf(float) nounwind readnone
+define void @floor_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @floorf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @expm1_f32(
+;CHECK: vexpm1f{{.*}}<4 x float>
+;CHECK: ret void
+declare float @expm1f(float) nounwind readnone
+define void @expm1_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @expm1f(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @log1p_f32(
+;CHECK: vlog1pf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @log1pf(float) nounwind readnone
+define void @log1p_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @log1pf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @log10_f32(
+;CHECK: vlog10f{{.*}}<4 x float>
+;CHECK: ret void
+declare float @log10f(float) nounwind readnone
+define void @log10_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @log10f(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @logb_f32(
+;CHECK: vlogbf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @logbf(float) nounwind readnone
+define void @logb_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @logbf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @sin_f32(
+;CHECK: vsinf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @sinf(float) nounwind readnone
+define void @sin_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @sinf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @cos_f32(
+;CHECK: vcosf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @cosf(float) nounwind readnone
+define void @cos_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @cosf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @tan_f32(
+;CHECK: vtanf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @tanf(float) nounwind readnone
+define void @tan_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @tanf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @asin_f32(
+;CHECK: vasinf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @asinf(float) nounwind readnone
+define void @asin_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @asinf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @acos_f32(
+;CHECK: vacosf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @acosf(float) nounwind readnone
+define void @acos_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @acosf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @atan_f32(
+;CHECK: vatanf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @atanf(float) nounwind readnone
+define void @atan_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @atanf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @sinh_f32(
+;CHECK: vsinhf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @sinhf(float) nounwind readnone
+define void @sinh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @sinhf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @cosh_f32(
+;CHECK: vcoshf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @coshf(float) nounwind readnone
+define void @cosh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @coshf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @tanh_f32(
+;CHECK: vtanhf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @tanhf(float) nounwind readnone
+define void @tanh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @tanhf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @asinh_f32(
+;CHECK: vasinhf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @asinhf(float) nounwind readnone
+define void @asinh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @asinhf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @acosh_f32(
+;CHECK: vacoshf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @acoshf(float) nounwind readnone
+define void @acosh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @acoshf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
+
+;CHECK-LABEL: @atanh_f32(
+;CHECK: vatanhf{{.*}}<4 x float>
+;CHECK: ret void
+declare float @atanhf(float) nounwind readnone
+define void @atanh_f32(i32 %n, float* noalias %y, float* noalias %x) nounwind uwtable {
+entry:
+  %cmp6 = icmp sgt i32 %n, 0
+  br i1 %cmp6, label %for.body, label %for.end
+
+for.body:                                         ; preds = %entry, %for.body
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %call = tail call float @atanhf(float %0) nounwind readnone
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  store float %call, float* %arrayidx2, align 4
+  %indvars.iv.next = add i64 %indvars.iv, 1
+  %lftr.wideiv = trunc i64 %indvars.iv.next to i32
+  %exitcond = icmp eq i32 %lftr.wideiv, %n
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:                                          ; preds = %for.body, %entry
+  ret void
+}
diff --git a/test/Transforms/LoopVectorize/X86/vect.omp.force.ll b/test/Transforms/LoopVectorize/X86/vect.omp.force.ll
index a781fbedb404..c2a0fed1fe6f 100644
--- a/test/Transforms/LoopVectorize/X86/vect.omp.force.ll
+++ b/test/Transforms/LoopVectorize/X86/vect.omp.force.ll
@@ -15,9 +15,9 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; The source code for the test:
 ;
 ; #include <math.h>
-; void foo(float* restrict A, float * restrict B, int size)
+; void foo(float* restrict A, float * restrict B)
 ; {
-;   for (int i = 0; i < size; ++i) A[i] = sinf(B[i]);
+;   for (int i = 0; i < 1000; i+=2) A[i] = sinf(B[i]);
 ; }
 ;
 
@@ -25,24 +25,20 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; This loop will be vectorized, although the scalar cost is lower than any of vector costs, but vectorization is explicitly forced in metadata.
 ;
 
-define void @vectorized(float* noalias nocapture %A, float* noalias nocapture %B, i32 %size) {
+define void @vectorized(float* noalias nocapture %A, float* noalias nocapture %B) {
 entry:
-  %cmp6 = icmp sgt i32 %size, 0
-  br i1 %cmp6, label %for.body.preheader, label %for.end
-
-for.body.preheader:
   br label %for.body
 
 for.body:
-  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
   %call = tail call float @llvm.sin.f32(float %0)
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
-  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 2
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
-  %exitcond = icmp eq i32 %lftr.wideiv, %size
+  %exitcond = icmp eq i32 %lftr.wideiv, 1000
   br i1 %exitcond, label %for.end.loopexit, label %for.body, !llvm.loop !1
 
 for.end.loopexit:
@@ -59,24 +55,20 @@ for.end:
 ; This method will not be vectorized, as scalar cost is lower than any of vector costs.
 ;
 
-define void @not_vectorized(float* noalias nocapture %A, float* noalias nocapture %B, i32 %size) {
+define void @not_vectorized(float* noalias nocapture %A, float* noalias nocapture %B) {
 entry:
-  %cmp6 = icmp sgt i32 %size, 0
-  br i1 %cmp6, label %for.body.preheader, label %for.end
-
-for.body.preheader:
   br label %for.body
 
 for.body:
-  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
+  %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
   %call = tail call float @llvm.sin.f32(float %0)
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
-  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 2
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
-  %exitcond = icmp eq i32 %lftr.wideiv, %size
+  %exitcond = icmp eq i32 %lftr.wideiv, 1000
   br i1 %exitcond, label %for.end.loopexit, label %for.body, !llvm.loop !3
 
 for.end.loopexit:
diff --git a/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll b/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll
index e39e6b57dd96..8d139ac7e5af 100644
--- a/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll
+++ b/test/Transforms/LoopVectorize/X86/vect.omp.force.small-tc.ll
@@ -29,10 +29,10 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
   %add = fadd fast float %0, %1
   store float %add, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -55,10 +55,10 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !3
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
   %add = fadd fast float %0, %1
   store float %add, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !3
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/X86/vector-scalar-select-cost.ll b/test/Transforms/LoopVectorize/X86/vector-scalar-select-cost.ll
index 3b3a7875ab36..5efabe16d32b 100644
--- a/test/Transforms/LoopVectorize/X86/vector-scalar-select-cost.ll
+++ b/test/Transforms/LoopVectorize/X86/vector-scalar-select-cost.ll
@@ -15,12 +15,12 @@ define void @scalarselect(i1 %cond) {
 
 ; <label>:1
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
 
 ; A scalar select has a cost of 1 on core2
 ; CHECK: cost of 1 for VF 2 {{.*}}  select i1 %cond, i32 %6, i32 0
@@ -42,16 +42,16 @@ define void @vectorselect(i1 %cond) {
 
 ; <label>:1
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   %8 = icmp ult i64 %indvars.iv, 8
 
-; A vector select has a cost of 4 on core2
-; CHECK: cost of 4 for VF 2 {{.*}}  select i1 %8, i32 %6, i32 0
+; A vector select has a cost of 1 on core2
+; CHECK: cost of 1 for VF 2 {{.*}}  select i1 %8, i32 %6, i32 0
 
   %sel = select i1 %8, i32 %6, i32 zeroinitializer
   store i32 %sel, i32* %7, align 4
diff --git a/test/Transforms/LoopVectorize/X86/vector_ptr_load_store.ll b/test/Transforms/LoopVectorize/X86/vector_ptr_load_store.ll
index e57cfefec07c..6cd3c9c3bc01 100644
--- a/test/Transforms/LoopVectorize/X86/vector_ptr_load_store.ll
+++ b/test/Transforms/LoopVectorize/X86/vector_ptr_load_store.ll
@@ -19,7 +19,7 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: test_consecutive_store
 ; CHECK: The Widest type: 64 bits
 define void @test_consecutive_store(%0**, %0**, %0** nocapture) nounwind ssp uwtable align 2 {
-  %4 = load %0** %2, align 8
+  %4 = load %0*, %0** %2, align 8
   %5 = icmp eq %0** %0, %1
   br i1 %5, label %12, label %6
 
@@ -29,7 +29,7 @@ define void @test_consecutive_store(%0**, %0**, %0** nocapture) nounwind ssp uwt
 ; <label>:7                                       ; preds = %7, %6
   %8 = phi %0** [ %0, %6 ], [ %9, %7 ]
   store %0* %4, %0** %8, align 8
-  %9 = getelementptr inbounds %0** %8, i64 1
+  %9 = getelementptr inbounds %0*, %0** %8, i64 1
   %10 = icmp eq %0** %9, %1
   br i1 %10, label %11, label %7
 
@@ -61,12 +61,12 @@ define void @test_nonconsecutive_store() nounwind ssp uwtable {
 
 ; <label>:3                                       ; preds = %3, %1
   %4 = phi i64 [ 0, %1 ], [ %11, %3 ]
-  %5 = getelementptr inbounds [2048 x i16]* @q, i64 0, i64 %4
-  %6 = load i16* %5, align 2
+  %5 = getelementptr inbounds [2048 x i16], [2048 x i16]* @q, i64 0, i64 %4
+  %6 = load i16, i16* %5, align 2
   %7 = sext i16 %6 to i64
   %8 = add i64 %7, 1
   %9 = inttoptr i64 %8 to i32*
-  %10 = getelementptr inbounds [2048 x [8 x i32*]]* @p, i64 0, i64 %4, i64 %2
+  %10 = getelementptr inbounds [2048 x [8 x i32*]], [2048 x [8 x i32*]]* @p, i64 0, i64 %4, i64 %2
   store i32* %9, i32** %10, align 8
   %11 = add i64 %4, 1
   %12 = trunc i64 %11 to i32
@@ -100,8 +100,8 @@ define i8 @test_consecutive_ptr_load() nounwind readonly ssp uwtable {
 ; <label>:1                                       ; preds = %1, %0
   %2 = phi i64 [ 0, %0 ], [ %10, %1 ]
   %3 = phi i8 [ 0, %0 ], [ %9, %1 ]
-  %4 = getelementptr inbounds [1024 x i32*]* @ia, i32 0, i64 %2
-  %5 = load i32** %4, align 4
+  %4 = getelementptr inbounds [1024 x i32*], [1024 x i32*]* @ia, i32 0, i64 %2
+  %5 = load i32*, i32** %4, align 4
   %6 = ptrtoint i32* %5 to i64
   %7 = trunc i64 %6 to i8
   %8 = add i8 %3, 1
@@ -127,9 +127,9 @@ define void @test_nonconsecutive_ptr_load() nounwind ssp uwtable {
 
 ; <label>:3                                       ; preds = %3, %1
   %4 = phi i64 [ 0, %1 ], [ %10, %3 ]
-  %5 = getelementptr inbounds [2048 x [8 x i32*]]* @p2, i64 0, i64 %4, i64 %2
-  %6 = getelementptr inbounds [2048 x i16]* @q2, i64 0, i64 %4
-  %7 = load i32** %5, align 2
+  %5 = getelementptr inbounds [2048 x [8 x i32*]], [2048 x [8 x i32*]]* @p2, i64 0, i64 %4, i64 %2
+  %6 = getelementptr inbounds [2048 x i16], [2048 x i16]* @q2, i64 0, i64 %4
+  %7 = load i32*, i32** %5, align 2
   %8 = ptrtoint i32* %7 to i64
   %9 = trunc i64 %8 to i16
   store i16 %9, i16* %6, align 8
diff --git a/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll b/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll
index 011ce8eac556..65cabb05f2fb 100644
--- a/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll
+++ b/test/Transforms/LoopVectorize/X86/vectorization-remarks-missed.ll
@@ -52,7 +52,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !16
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !16
   %0 = trunc i64 %indvars.iv to i32, !dbg !16
   store i32 %0, i32* %arrayidx, align 4, !dbg !16, !tbaa !18
   %cmp3 = icmp sle i32 %0, %Length, !dbg !22
@@ -74,7 +74,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !30
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !30
   %0 = trunc i64 %indvars.iv to i32, !dbg !30
   store i32 %0, i32* %arrayidx, align 4, !dbg !30, !tbaa !18
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !25
@@ -97,12 +97,12 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv, !dbg !35
-  %0 = load i32* %arrayidx, align 4, !dbg !35, !tbaa !18
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %indvars.iv, !dbg !35
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !35, !tbaa !18
   %idxprom1 = sext i32 %0 to i64, !dbg !35
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %idxprom1, !dbg !35
-  %1 = load i32* %arrayidx2, align 4, !dbg !35, !tbaa !18
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !35
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %idxprom1, !dbg !35
+  %1 = load i32, i32* %arrayidx2, align 4, !dbg !35, !tbaa !18
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !35
   store i32 %1, i32* %arrayidx4, align 4, !dbg !35, !tbaa !18
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !32
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !32
@@ -122,40 +122,40 @@ attributes #0 = { nounwind }
 !llvm.module.flags = !{!9, !10}
 !llvm.ident = !{!11}
 
-!0 = !{!"0x11\004\00clang version 3.5.0\001\00\006\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./source.cpp] [DW_LANG_C_plus_plus]
-!1 = !{!"source.cpp", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0", isOptimized: true, runtimeVersion: 6, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "source.cpp", directory: ".")
 !2 = !{}
 !3 = !{!4, !7, !8}
-!4 = !{!"0x2e\00test\00test\00\001\000\001\000\006\00256\001\001", !1, !5, !6, null, void (i32*, i32)* @_Z4testPii, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [test]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./source.cpp]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!7 = !{!"0x2e\00test_disabled\00test_disabled\00\0010\000\001\000\006\00256\001\0010", !1, !5, !6, null, void (i32*, i32)* @_Z13test_disabledPii, null, null, !2} ; [ DW_TAG_subprogram ] [line 10] [def] [test_disabled]
-!8 = !{!"0x2e\00test_array_bounds\00test_array_bounds\00\0016\000\001\000\006\00256\001\0016", !1, !5, !6, null, void (i32*, i32*, i32)* @_Z17test_array_boundsPiS_i, null, null, !2} ; [ DW_TAG_subprogram ] [line 16] [def] [test_array_bounds]
+!4 = !DISubprogram(name: "test", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32*, i32)* @_Z4testPii, variables: !2)
+!5 = !DIFile(filename: "source.cpp", directory: ".")
+!6 = !DISubroutineType(types: !2)
+!7 = !DISubprogram(name: "test_disabled", line: 10, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 10, file: !1, scope: !5, type: !6, function: void (i32*, i32)* @_Z13test_disabledPii, variables: !2)
+!8 = !DISubprogram(name: "test_array_bounds", line: 16, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 16, file: !1, scope: !5, type: !6, function: void (i32*, i32*, i32)* @_Z17test_array_boundsPiS_i, variables: !2)
 !9 = !{i32 2, !"Dwarf Version", i32 2}
-!10 = !{i32 2, !"Debug Info Version", i32 2}
+!10 = !{i32 2, !"Debug Info Version", i32 3}
 !11 = !{!"clang version 3.5.0"}
-!12 = !MDLocation(line: 3, column: 8, scope: !13)
-!13 = !{!"0xb\003\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
+!12 = !DILocation(line: 3, column: 8, scope: !13)
+!13 = distinct !DILexicalBlock(line: 3, column: 3, file: !1, scope: !4)
 !14 = !{!14, !15, !15}
 !15 = !{!"llvm.loop.vectorize.enable", i1 true}
-!16 = !MDLocation(line: 4, column: 5, scope: !17)
-!17 = !{!"0xb\003\0036\000", !1, !13} ; [ DW_TAG_lexical_block ]
+!16 = !DILocation(line: 4, column: 5, scope: !17)
+!17 = distinct !DILexicalBlock(line: 3, column: 36, file: !1, scope: !13)
 !18 = !{!19, !19, i64 0}
 !19 = !{!"int", !20, i64 0}
 !20 = !{!"omnipotent char", !21, i64 0}
 !21 = !{!"Simple C/C++ TBAA"}
-!22 = !MDLocation(line: 5, column: 9, scope: !23)
-!23 = !{!"0xb\005\009\000", !1, !17} ; [ DW_TAG_lexical_block ]
-!24 = !MDLocation(line: 8, column: 1, scope: !4)
-!25 = !MDLocation(line: 12, column: 8, scope: !26)
-!26 = !{!"0xb\0012\003\000", !1, !7} ; [ DW_TAG_lexical_block ]
+!22 = !DILocation(line: 5, column: 9, scope: !23)
+!23 = distinct !DILexicalBlock(line: 5, column: 9, file: !1, scope: !17)
+!24 = !DILocation(line: 8, column: 1, scope: !4)
+!25 = !DILocation(line: 12, column: 8, scope: !26)
+!26 = distinct !DILexicalBlock(line: 12, column: 3, file: !1, scope: !7)
 !27 = !{!27, !28, !29}
 !28 = !{!"llvm.loop.interleave.count", i32 1}
 !29 = !{!"llvm.loop.vectorize.width", i32 1}
-!30 = !MDLocation(line: 13, column: 5, scope: !26)
-!31 = !MDLocation(line: 14, column: 1, scope: !7)
-!32 = !MDLocation(line: 18, column: 8, scope: !33)
-!33 = !{!"0xb\0018\003\000", !1, !8} ; [ DW_TAG_lexical_block ]
+!30 = !DILocation(line: 13, column: 5, scope: !26)
+!31 = !DILocation(line: 14, column: 1, scope: !7)
+!32 = !DILocation(line: 18, column: 8, scope: !33)
+!33 = distinct !DILexicalBlock(line: 18, column: 3, file: !1, scope: !8)
 !34 = !{!34, !15}
-!35 = !MDLocation(line: 19, column: 5, scope: !33)
-!36 = !MDLocation(line: 20, column: 1, scope: !8)
+!35 = !DILocation(line: 19, column: 5, scope: !33)
+!36 = !DILocation(line: 20, column: 1, scope: !8)
diff --git a/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll b/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll
index 16fe370464b0..1c21748d8bdd 100644
--- a/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll
+++ b/test/Transforms/LoopVectorize/X86/vectorization-remarks.ll
@@ -5,7 +5,7 @@
 ; This code has all the !dbg annotations needed to track source line information,
 ; but is missing the llvm.dbg.cu annotation. This prevents code generation from
 ; emitting debug info in the final output.
-; RUN: llc -mtriple x86_64-pc-linux-gnu %s -o - | FileCheck -check-prefix=DEBUG-OUTPUT %s
+; RUN: llc < %s -mtriple x86_64-pc-linux-gnu -o - | FileCheck -check-prefix=DEBUG-OUTPUT %s
 ; DEBUG-OUTPUT-NOT: .loc
 ; DEBUG-OUTPUT-NOT: {{.*}}.debug_info
 
@@ -26,11 +26,11 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %add8 = phi i32 [ 0, %entry ], [ %add, %for.body ], !dbg !19
-  %arrayidx = getelementptr inbounds [16 x i8]* %cb, i64 0, i64 %indvars.iv, !dbg !19
-  %0 = load i8* %arrayidx, align 1, !dbg !19, !tbaa !21
+  %arrayidx = getelementptr inbounds [16 x i8], [16 x i8]* %cb, i64 0, i64 %indvars.iv, !dbg !19
+  %0 = load i8, i8* %arrayidx, align 1, !dbg !19, !tbaa !21
   %conv = sext i8 %0 to i32, !dbg !19
-  %arrayidx2 = getelementptr inbounds [16 x i8]* %cc, i64 0, i64 %indvars.iv, !dbg !19
-  %1 = load i8* %arrayidx2, align 1, !dbg !19, !tbaa !21
+  %arrayidx2 = getelementptr inbounds [16 x i8], [16 x i8]* %cc, i64 0, i64 %indvars.iv, !dbg !19
+  %1 = load i8, i8* %arrayidx2, align 1, !dbg !19, !tbaa !21
   %conv3 = sext i8 %1 to i32, !dbg !19
   %sub = sub i32 %conv, %conv3, !dbg !19
   %add = add nsw i32 %sub, %add8, !dbg !19
@@ -49,26 +49,26 @@ declare void @ibar(i32*) #1
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!1 = !{!"vectorization-remarks.c", !"."}
+!1 = !DIFile(filename: "vectorization-remarks.c", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\005\000\001\000\006\00256\001\006", !1, !5, !6, null, i32 (i32)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 5] [def] [scope 6] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./vectorization-remarks.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 5, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 6, file: !1, scope: !5, type: !6, function: i32 (i32)* @foo, variables: !2)
+!5 = !DIFile(filename: "vectorization-remarks.c", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 4}
-!8 = !{i32 1, !"Debug Info Version", i32 2}
+!8 = !{i32 1, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5.0 "}
-!10 = !MDLocation(line: 8, column: 3, scope: !4)
+!10 = !DILocation(line: 8, column: 3, scope: !4)
 !11 = !{!12, !12, i64 0}
 !12 = !{!"int", !13, i64 0}
 !13 = !{!"omnipotent char", !14, i64 0}
 !14 = !{!"Simple C/C++ TBAA"}
-!15 = !MDLocation(line: 17, column: 8, scope: !16)
-!16 = !{!"0xb\0017\008\002", !1, !17} ; [ DW_TAG_lexical_block ] [./vectorization-remarks.c]
-!17 = !{!"0xb\0017\008\001", !1, !18} ; [ DW_TAG_lexical_block ] [./vectorization-remarks.c]
-!18 = !{!"0xb\0017\003\000", !1, !4} ; [ DW_TAG_lexical_block ] [./vectorization-remarks.c]
-!19 = !MDLocation(line: 18, column: 5, scope: !20)
-!20 = !{!"0xb\0017\0027\000", !1, !18} ; [ DW_TAG_lexical_block ] [./vectorization-remarks.c]
+!15 = !DILocation(line: 17, column: 8, scope: !16)
+!16 = distinct !DILexicalBlock(line: 17, column: 8, file: !1, scope: !17)
+!17 = distinct !DILexicalBlock(line: 17, column: 8, file: !1, scope: !18)
+!18 = distinct !DILexicalBlock(line: 17, column: 3, file: !1, scope: !4)
+!19 = !DILocation(line: 18, column: 5, scope: !20)
+!20 = distinct !DILexicalBlock(line: 17, column: 27, file: !1, scope: !18)
 !21 = !{!13, !13, i64 0}
-!22 = !MDLocation(line: 20, column: 3, scope: !4)
-!23 = !MDLocation(line: 21, column: 3, scope: !4)
+!22 = !DILocation(line: 20, column: 3, scope: !4)
+!23 = !DILocation(line: 21, column: 3, scope: !4)
diff --git a/test/Transforms/LoopVectorize/X86/x86_fp80-vector-store.ll b/test/Transforms/LoopVectorize/X86/x86_fp80-vector-store.ll
index d8e5403d2d4e..0debb338486c 100644
--- a/test/Transforms/LoopVectorize/X86/x86_fp80-vector-store.ll
+++ b/test/Transforms/LoopVectorize/X86/x86_fp80-vector-store.ll
@@ -17,7 +17,7 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %conv = sitofp i32 1 to x86_fp80
-  %arrayidx = getelementptr inbounds [1024 x x86_fp80]* @x, i64 0, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds [1024 x x86_fp80], [1024 x x86_fp80]* @x, i64 0, i64 %indvars.iv
   store x86_fp80 %conv, x86_fp80* %arrayidx, align 16
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/XCore/no-vector-registers.ll b/test/Transforms/LoopVectorize/XCore/no-vector-registers.ll
index cab333d06406..afb3322dd67e 100644
--- a/test/Transforms/LoopVectorize/XCore/no-vector-registers.ll
+++ b/test/Transforms/LoopVectorize/XCore/no-vector-registers.ll
@@ -13,7 +13,7 @@ entry:
 do.body:
   %ptr.addr.0 = phi i8* [ %ptr, %entry ], [ %incdec.ptr, %do.body ]
   %len.addr.0 = phi i32 [ %len, %entry ], [ %dec, %do.body ]
-  %incdec.ptr = getelementptr inbounds i8* %ptr.addr.0, i32 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %ptr.addr.0, i32 1
   store i8 0, i8* %ptr.addr.0, align 1
   %dec = add nsw i32 %len.addr.0, -1
   %tobool = icmp eq i32 %len.addr.0, 0
diff --git a/test/Transforms/LoopVectorize/align.ll b/test/Transforms/LoopVectorize/align.ll
index f2fb8b91b46e..7ee401deef4e 100644
--- a/test/Transforms/LoopVectorize/align.ll
+++ b/test/Transforms/LoopVectorize/align.ll
@@ -6,8 +6,8 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; Make sure we output the abi alignment if no alignment is specified.
 
 ;CHECK-LABEL: @align
-;CHECK: load <4 x i32>* {{.*}} align  4
-;CHECK: load <4 x i32>* {{.*}} align  4
+;CHECK: load <4 x i32>, <4 x i32>* {{.*}} align  4
+;CHECK: load <4 x i32>, <4 x i32>* {{.*}} align  4
 ;CHECK: store <4 x i32> {{.*}} align  4
 
 define void @align(i32* %a, i32* %b, i32* %c) nounwind uwtable ssp {
@@ -15,12 +15,12 @@ define void @align(i32* %a, i32* %b, i32* %c) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %3 = load i32* %2
-  %4 = getelementptr inbounds i32* %c, i64 %indvars.iv
-  %5 = load i32* %4
+  %2 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %3 = load i32, i32* %2
+  %4 = getelementptr inbounds i32, i32* %c, i64 %indvars.iv
+  %5 = load i32, i32* %4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %7 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %6, i32* %7
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/bsd_regex.ll b/test/Transforms/LoopVectorize/bsd_regex.ll
index 7a3e79893da3..ddb00ba31e14 100644
--- a/test/Transforms/LoopVectorize/bsd_regex.ll
+++ b/test/Transforms/LoopVectorize/bsd_regex.ll
@@ -24,7 +24,7 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %0 = shl nsw i64 %indvars.iv, 2
-  %arrayidx = getelementptr inbounds i32* %A, i64 %0
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %0
   store i32 4, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/bzip_reverse_loops.ll b/test/Transforms/LoopVectorize/bzip_reverse_loops.ll
index d7cbad06ddf0..f1efb256391b 100644
--- a/test/Transforms/LoopVectorize/bzip_reverse_loops.ll
+++ b/test/Transforms/LoopVectorize/bzip_reverse_loops.ll
@@ -16,8 +16,8 @@ entry:
 do.body:                                          ; preds = %cond.end, %entry
   %n.addr.0 = phi i32 [ %n, %entry ], [ %dec, %cond.end ]
   %p.addr.0 = phi i16* [ %p, %entry ], [ %incdec.ptr, %cond.end ]
-  %incdec.ptr = getelementptr inbounds i16* %p.addr.0, i64 -1
-  %0 = load i16* %incdec.ptr, align 2
+  %incdec.ptr = getelementptr inbounds i16, i16* %p.addr.0, i64 -1
+  %0 = load i16, i16* %incdec.ptr, align 2
   %conv = zext i16 %0 to i32
   %cmp = icmp ult i32 %conv, %size
   br i1 %cmp, label %cond.end, label %cond.true
@@ -51,8 +51,8 @@ entry:
 do.body:                                          ; preds = %do.body, %entry
   %n.addr.0 = phi i32 [ %n, %entry ], [ %dec, %do.body ]
   %p.0 = phi i32* [ %a, %entry ], [ %incdec.ptr, %do.body ]
-  %incdec.ptr = getelementptr inbounds i32* %p.0, i64 -1
-  %0 = load i32* %incdec.ptr, align 4
+  %incdec.ptr = getelementptr inbounds i32, i32* %p.0, i64 -1
+  %0 = load i32, i32* %incdec.ptr, align 4
   %cmp = icmp slt i32 %0, %wsize
   %sub = sub nsw i32 %0, %wsize
   %cond = select i1 %cmp, i32 0, i32 %sub
diff --git a/test/Transforms/LoopVectorize/calloc.ll b/test/Transforms/LoopVectorize/calloc.ll
index 5f441f3e238f..a41e51787f32 100644
--- a/test/Transforms/LoopVectorize/calloc.ll
+++ b/test/Transforms/LoopVectorize/calloc.ll
@@ -22,8 +22,8 @@ for.body.lr.ph:                                   ; preds = %entry
 for.body:                                         ; preds = %for.body, %for.body.lr.ph
   %i.030 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %shr = lshr i64 %i.030, 1
-  %arrayidx = getelementptr inbounds i8* %bytes, i64 %shr
-  %1 = load i8* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds i8, i8* %bytes, i64 %shr
+  %1 = load i8, i8* %arrayidx, align 1
   %conv = zext i8 %1 to i32
   %and = shl i64 %i.030, 2
   %neg = and i64 %and, 4
@@ -37,7 +37,7 @@ for.body:                                         ; preds = %for.body, %for.body
   %cond = select i1 %cmp15, i32 87, i32 48
   %add17 = add nsw i32 %cond, %shr11
   %conv18 = trunc i32 %add17 to i8
-  %arrayidx19 = getelementptr inbounds i8* %call, i64 %i.030
+  %arrayidx19 = getelementptr inbounds i8, i8* %call, i64 %i.030
   store i8 %conv18, i8* %arrayidx19, align 1
   %inc = add i64 %i.030, 1
   %exitcond = icmp eq i64 %inc, %0
diff --git a/test/Transforms/LoopVectorize/cast-induction.ll b/test/Transforms/LoopVectorize/cast-induction.ll
index 4f92d33ff528..fae89976a7bc 100644
--- a/test/Transforms/LoopVectorize/cast-induction.ll
+++ b/test/Transforms/LoopVectorize/cast-induction.ll
@@ -16,7 +16,7 @@ define void @example12() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   %3 = trunc i64 %indvars.iv to i32
   store i32 %3, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/conditional-assignment.ll b/test/Transforms/LoopVectorize/conditional-assignment.ll
index 38e9c4f9a61c..f41f08df07a6 100644
--- a/test/Transforms/LoopVectorize/conditional-assignment.ll
+++ b/test/Transforms/LoopVectorize/conditional-assignment.ll
@@ -12,8 +12,8 @@ entry:
 
 for.body:                                         ; preds = %for.inc, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
-  %arrayidx = getelementptr inbounds i32* %indices, i64 %indvars.iv, !dbg !12
-  %0 = load i32* %arrayidx, align 4, !dbg !12, !tbaa !14
+  %arrayidx = getelementptr inbounds i32, i32* %indices, i64 %indvars.iv, !dbg !12
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !12, !tbaa !14
   %cmp1 = icmp eq i32 %0, 1024, !dbg !12
   br i1 %cmp1, label %if.then, label %for.inc, !dbg !12
 
@@ -36,23 +36,23 @@ attributes #0 = { nounwind }
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\0012\00clang version 3.6.0\001\00\000\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ]
-!1 = !{!"source.c", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.6.0", isOptimized: true, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "source.c", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00conditional_store\00conditional_store\00\001\000\001\000\006\00256\001\001", !1, !5, !6, null, void (i32*)* @conditional_store, null, null, !2} ; [ DW_TAG_subprogram ]
-!5 = !{!"0x29", !1} ; [ DW_TAG_file_type ]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ]
+!4 = !DISubprogram(name: "conditional_store", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32*)* @conditional_store, variables: !2)
+!5 = !DIFile(filename: "source.c", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 2}
-!8 = !{i32 2, !"Debug Info Version", i32 2}
+!8 = !{i32 2, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.6.0"}
-!10 = !MDLocation(line: 2, column: 8, scope: !11)
-!11 = !{!"0xb\002\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
-!12 = !MDLocation(line: 3, column: 9, scope: !13)
-!13 = !{!"0xb\003\009\000", !1, !11} ; [ DW_TAG_lexical_block ]
+!10 = !DILocation(line: 2, column: 8, scope: !11)
+!11 = distinct !DILexicalBlock(line: 2, column: 3, file: !1, scope: !4)
+!12 = !DILocation(line: 3, column: 9, scope: !13)
+!13 = distinct !DILexicalBlock(line: 3, column: 9, file: !1, scope: !11)
 !14 = !{!15, !15, i64 0}
 !15 = !{!"int", !16, i64 0}
 !16 = !{!"omnipotent char", !17, i64 0}
 !17 = !{!"Simple C/C++ TBAA"}
-!18 = !MDLocation(line: 3, column: 29, scope: !13)
-!19 = !MDLocation(line: 4, column: 1, scope: !4)
+!18 = !DILocation(line: 3, column: 29, scope: !13)
+!19 = !DILocation(line: 4, column: 1, scope: !4)
diff --git a/test/Transforms/LoopVectorize/control-flow.ll b/test/Transforms/LoopVectorize/control-flow.ll
index 1882c3ff9a83..c56f9122e462 100644
--- a/test/Transforms/LoopVectorize/control-flow.ll
+++ b/test/Transforms/LoopVectorize/control-flow.ll
@@ -30,8 +30,8 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %if.else
   %indvars.iv = phi i64 [ %indvars.iv.next, %if.else ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !12
-  %0 = load i32* %arrayidx, align 4, !dbg !12, !tbaa !15
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !12
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !12, !tbaa !15
   %cmp1 = icmp sgt i32 %0, 10, !dbg !12
   br i1 %cmp1, label %end.loopexit, label %if.else, !dbg !12
 
@@ -55,24 +55,24 @@ attributes #0 = { nounwind }
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\004\00clang version 3.5.0\001\00\006\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./source.cpp] [DW_LANG_C_plus_plus]
-!1 = !{!"source.cpp", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0", isOptimized: true, runtimeVersion: 6, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "source.cpp", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00test\00test\00\001\000\001\000\006\00256\001\002", !1, !5, !6, null, i32 (i32*, i32)* @_Z4testPii, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [scope 2] [test]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./source.cpp]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "test", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 2, file: !1, scope: !5, type: !6, function: i32 (i32*, i32)* @_Z4testPii, variables: !2)
+!5 = !DIFile(filename: "source.cpp", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 2}
-!8 = !{i32 2, !"Debug Info Version", i32 2}
+!8 = !{i32 2, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5.0"}
-!10 = !MDLocation(line: 3, column: 8, scope: !11)
-!11 = !{!"0xb\003\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
-!12 = !MDLocation(line: 5, column: 9, scope: !13)
-!13 = !{!"0xb\005\009\000", !1, !14} ; [ DW_TAG_lexical_block ]
-!14 = !{!"0xb\004\003\000", !1, !11} ; [ DW_TAG_lexical_block ]
+!10 = !DILocation(line: 3, column: 8, scope: !11)
+!11 = distinct !DILexicalBlock(line: 3, column: 3, file: !1, scope: !4)
+!12 = !DILocation(line: 5, column: 9, scope: !13)
+!13 = distinct !DILexicalBlock(line: 5, column: 9, file: !1, scope: !14)
+!14 = distinct !DILexicalBlock(line: 4, column: 3, file: !1, scope: !11)
 !15 = !{!16, !16, i64 0}
 !16 = !{!"int", !17, i64 0}
 !17 = !{!"omnipotent char", !18, i64 0}
 !18 = !{!"Simple C/C++ TBAA"}
-!19 = !MDLocation(line: 8, column: 7, scope: !13)
-!20 = !MDLocation(line: 12, column: 3, scope: !4)
+!19 = !DILocation(line: 8, column: 7, scope: !13)
+!20 = !DILocation(line: 12, column: 3, scope: !4)
diff --git a/test/Transforms/LoopVectorize/cpp-new-array.ll b/test/Transforms/LoopVectorize/cpp-new-array.ll
index f32f610b1cd6..22896d33428a 100644
--- a/test/Transforms/LoopVectorize/cpp-new-array.ll
+++ b/test/Transforms/LoopVectorize/cpp-new-array.ll
@@ -24,21 +24,21 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %idxprom = sext i32 %i.01 to i64
-  %arrayidx = getelementptr inbounds float* %0, i64 %idxprom
-  %3 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %0, i64 %idxprom
+  %3 = load float, float* %arrayidx, align 4
   %idxprom5 = sext i32 %i.01 to i64
-  %arrayidx6 = getelementptr inbounds float* %1, i64 %idxprom5
-  %4 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %1, i64 %idxprom5
+  %4 = load float, float* %arrayidx6, align 4
   %add = fadd float %3, %4
   %idxprom7 = sext i32 %i.01 to i64
-  %arrayidx8 = getelementptr inbounds float* %2, i64 %idxprom7
+  %arrayidx8 = getelementptr inbounds float, float* %2, i64 %idxprom7
   store float %add, float* %arrayidx8, align 4
   %inc = add nsw i32 %i.01, 1
   %cmp = icmp slt i32 %inc, 1000
   br i1 %cmp, label %for.body, label %for.end
 
 for.end:                                          ; preds = %for.body
-  %5 = load float* %2, align 4
+  %5 = load float, float* %2, align 4
   %conv10 = fptosi float %5 to i32
   ret i32 %conv10
 }
diff --git a/test/Transforms/LoopVectorize/dbg.value.ll b/test/Transforms/LoopVectorize/dbg.value.ll
index 92d315481266..c7440f84b2c9 100644
--- a/test/Transforms/LoopVectorize/dbg.value.ll
+++ b/test/Transforms/LoopVectorize/dbg.value.ll
@@ -11,21 +11,21 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK-LABEL: @test(
 define i32 @test() #0 {
 entry:
-  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !9, metadata !{}), !dbg !18
+  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !9, metadata !DIExpression()), !dbg !18
   br label %for.body, !dbg !18
 
 for.body:
   ;CHECK: load <4 x i32>
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @B, i64 0, i64 %indvars.iv, !dbg !19
-  %0 = load i32* %arrayidx, align 4, !dbg !19
-  %arrayidx2 = getelementptr inbounds [1024 x i32]* @C, i64 0, i64 %indvars.iv, !dbg !19
-  %1 = load i32* %arrayidx2, align 4, !dbg !19
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @B, i64 0, i64 %indvars.iv, !dbg !19
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !19
+  %arrayidx2 = getelementptr inbounds [1024 x i32], [1024 x i32]* @C, i64 0, i64 %indvars.iv, !dbg !19
+  %1 = load i32, i32* %arrayidx2, align 4, !dbg !19
   %add = add nsw i32 %1, %0, !dbg !19
-  %arrayidx4 = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv, !dbg !19
+  %arrayidx4 = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv, !dbg !19
   store i32 %add, i32* %arrayidx4, align 4, !dbg !19
   %indvars.iv.next = add i64 %indvars.iv, 1, !dbg !18
-  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !9, metadata !{}), !dbg !18
+  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !9, metadata !DIExpression()), !dbg !18
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !18
   %exitcond = icmp ne i32 %lftr.wideiv, 1024, !dbg !18
   br i1 %exitcond, label %for.body, label %for.end, !dbg !18
@@ -44,27 +44,27 @@ attributes #1 = { nounwind readnone }
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!26}
 
-!0 = !{!"0x11\004\00clang\001\00\000\00\000", !25, !1, !1, !2, !11, null} ; [ DW_TAG_compile_unit ]
-!1 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang", isOptimized: true, emissionKind: 0, file: !25, enums: !1, retainedTypes: !1, subprograms: !2, globals: !11)
+!1 = !{}
 !2 = !{!3}
-!3 = !{!"0x2e\00test\00test\00test\005\000\001\000\006\00256\001\005", !25, !4, !5, null, i32 ()* @test, null, null, !8} ; [ DW_TAG_subprogram ]
-!4 = !{!"0x29", !25} ; [ DW_TAG_file_type ]
-!5 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !6, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!3 = !DISubprogram(name: "test", linkageName: "test", line: 5, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 5, file: !25, scope: !4, type: !5, function: i32 ()* @test, variables: !8)
+!4 = !DIFile(filename: "test", directory: "/path/to/somewhere")
+!5 = !DISubroutineType(types: !6)
 !6 = !{!7}
-!7 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ]
+!7 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !8 = !{!9}
-!9 = !{!"0x100\00i\006\000", !10, !4, !7} ; [ DW_TAG_auto_variable ]
-!10 = !{!"0xb\006\000\000", !25, !3} ; [ DW_TAG_lexical_block ]
+!9 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "i", line: 6, scope: !10, file: !4, type: !7)
+!10 = distinct !DILexicalBlock(line: 6, column: 0, file: !25, scope: !3)
 !11 = !{!12, !16, !17}
-!12 = !{!"0x34\00A\00A\00\001\000\001", null, !4, !13, [1024 x i32]* @A, null} ; [ DW_TAG_variable ]
-!13 = !{!"0x1\00\000\0032768\0032\000\000", null, null, !7, !14, i32 0, null, null, null} ; [ DW_TAG_array_type ] [line 0, size 32768, align 32, offset 0] [from int]
+!12 = !DIGlobalVariable(name: "A", line: 1, isLocal: false, isDefinition: true, scope: null, file: !4, type: !13, variable: [1024 x i32]* @A)
+!13 = !DICompositeType(tag: DW_TAG_array_type, size: 32768, align: 32, baseType: !7, elements: !14)
 !14 = !{!15}
 !15 = !{i32 786465, i64 0, i64 1024}
-!16 = !{!"0x34\00B\00B\00\002\000\001", null, !4, !13, [1024 x i32]* @B, null} ; [ DW_TAG_variable ]
-!17 = !{!"0x34\00C\00C\00\003\000\001", null, !4, !13, [1024 x i32]* @C, null} ; [ DW_TAG_variable ]
-!18 = !MDLocation(line: 6, scope: !10)
-!19 = !MDLocation(line: 7, scope: !20)
-!20 = !{!"0xb\006\000\001", !25, !10} ; [ DW_TAG_lexical_block ]
-!24 = !MDLocation(line: 9, scope: !3)
-!25 = !{!"test", !"/path/to/somewhere"}
-!26 = !{i32 1, !"Debug Info Version", i32 2}
+!16 = !DIGlobalVariable(name: "B", line: 2, isLocal: false, isDefinition: true, scope: null, file: !4, type: !13, variable: [1024 x i32]* @B)
+!17 = !DIGlobalVariable(name: "C", line: 3, isLocal: false, isDefinition: true, scope: null, file: !4, type: !13, variable: [1024 x i32]* @C)
+!18 = !DILocation(line: 6, scope: !10)
+!19 = !DILocation(line: 7, scope: !20)
+!20 = distinct !DILexicalBlock(line: 6, column: 0, file: !25, scope: !10)
+!24 = !DILocation(line: 9, scope: !3)
+!25 = !DIFile(filename: "test", directory: "/path/to/somewhere")
+!26 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/LoopVectorize/debugloc.ll b/test/Transforms/LoopVectorize/debugloc.ll
index 634bf79ddf0e..e691afdd6933 100644
--- a/test/Transforms/LoopVectorize/debugloc.ll
+++ b/test/Transforms/LoopVectorize/debugloc.ll
@@ -8,8 +8,8 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 ; CHECK:   cmp.zero = icmp eq i64 {{.*}}, 0, !dbg ![[LOC:[0-9]+]]
 ; CHECK: vector.body
 ; CHECK:   index {{.*}}, !dbg ![[LOC]]
-; CHECK:   getelementptr inbounds i32* %a, {{.*}}, !dbg ![[LOC2:[0-9]+]]
-; CHECK:   load <2 x i32>* {{.*}}, !dbg ![[LOC2]]
+; CHECK:   getelementptr inbounds i32, i32* %a, {{.*}}, !dbg ![[LOC2:[0-9]+]]
+; CHECK:   load <2 x i32>, <2 x i32>* {{.*}}, !dbg ![[LOC2]]
 ; CHECK:   add <2 x i32> {{.*}}, !dbg ![[LOC2]]
 ; CHECK:   add i64 %index, 2, !dbg ![[LOC]]
 ; CHECK:   icmp eq i64 %index.next, %end.idx.rnd.down, !dbg ![[LOC]]
@@ -19,10 +19,10 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 define i32 @f(i32* nocapture %a, i32 %size) #0 {
 entry:
-  tail call void @llvm.dbg.value(metadata i32* %a, i64 0, metadata !13, metadata !{}), !dbg !19
-  tail call void @llvm.dbg.value(metadata i32 %size, i64 0, metadata !14, metadata !{}), !dbg !19
-  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !15, metadata !{}), !dbg !20
-  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !16, metadata !{}), !dbg !21
+  tail call void @llvm.dbg.value(metadata i32* %a, i64 0, metadata !13, metadata !DIExpression()), !dbg !19
+  tail call void @llvm.dbg.value(metadata i32 %size, i64 0, metadata !14, metadata !DIExpression()), !dbg !19
+  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !15, metadata !DIExpression()), !dbg !20
+  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !16, metadata !DIExpression()), !dbg !21
   %cmp4 = icmp eq i32 %size, 0, !dbg !21
   br i1 %cmp4, label %for.end, label %for.body.lr.ph, !dbg !21
 
@@ -32,12 +32,12 @@ for.body.lr.ph:                                   ; preds = %entry
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
   %sum.05 = phi i32 [ 0, %for.body.lr.ph ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv, !dbg !22
-  %0 = load i32* %arrayidx, align 4, !dbg !22
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv, !dbg !22
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !22
   %add = add i32 %0, %sum.05, !dbg !22
-  tail call void @llvm.dbg.value(metadata i32 %add.lcssa, i64 0, metadata !15, metadata !{}), !dbg !22
+  tail call void @llvm.dbg.value(metadata i32 %add.lcssa, i64 0, metadata !15, metadata !DIExpression()), !dbg !22
   %indvars.iv.next = add i64 %indvars.iv, 1, !dbg !21
-  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !16, metadata !{}), !dbg !21
+  tail call void @llvm.dbg.value(metadata !{null}, i64 0, metadata !16, metadata !DIExpression()), !dbg !21
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32, !dbg !21
   %exitcond = icmp ne i32 %lftr.wideiv, %size, !dbg !21
   br i1 %exitcond, label %for.body, label %for.cond.for.end_crit_edge, !dbg !21
@@ -63,28 +63,28 @@ attributes #1 = { nounwind readnone }
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!18, !27}
 
-!0 = !{!"0x11\0012\00clang version 3.4 (trunk 185038) (llvm/trunk 185097)\001\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [/Volumes/Data/backedup/dev/os/llvm/debug/-] [DW_LANG_C99]
-!1 = !{!"-", !"/Volumes/Data/backedup/dev/os/llvm/debug"}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.4 (trunk 185038) (llvm/trunk 185097)", isOptimized: true, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "-", directory: "/Volumes/Data/backedup/dev/os/llvm/debug")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00f\00f\00\003\000\001\000\006\00256\001\003", !5, !6, !7, null, i32 (i32*, i32)* @f, null, null, !12} ; [ DW_TAG_subprogram ] [line 3] [def] [f]
-!5 = !{!"<stdin>", !"/Volumes/Data/backedup/dev/os/llvm/debug"}
-!6 = !{!"0x29", !5}          ; [ DW_TAG_file_type ] [/Volumes/Data/backedup/dev/os/llvm/debug/<stdin>]
-!7 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !8, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "f", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 3, file: !5, scope: !6, type: !7, function: i32 (i32*, i32)* @f, variables: !12)
+!5 = !DIFile(filename: "<stdin>", directory: "/Volumes/Data/backedup/dev/os/llvm/debug")
+!6 = !DIFile(filename: "<stdin>", directory: "/Volumes/Data/backedup/dev/os/llvm/debug")
+!7 = !DISubroutineType(types: !8)
 !8 = !{!9, !10, !11}
-!9 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!10 = !{!"0xf\00\000\0064\0064\000\000", null, null, !9} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from int]
-!11 = !{!"0x24\00unsigned int\000\0032\0032\000\000\007", null, null} ; [ DW_TAG_base_type ] [unsigned int] [line 0, size 32, align 32, offset 0, enc DW_ATE_unsigned]
+!9 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!10 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !9)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "unsigned int", size: 32, align: 32, encoding: DW_ATE_unsigned)
 !12 = !{!13, !14, !15, !16}
-!13 = !{!"0x101\00a\0016777219\000", !4, !6, !10} ; [ DW_TAG_arg_variable ] [a] [line 3]
-!14 = !{!"0x101\00size\0033554435\000", !4, !6, !11} ; [ DW_TAG_arg_variable ] [size] [line 3]
-!15 = !{!"0x100\00sum\004\000", !4, !6, !11} ; [ DW_TAG_auto_variable ] [sum] [line 4]
-!16 = !{!"0x100\00i\005\000", !17, !6, !11} ; [ DW_TAG_auto_variable ] [i] [line 5]
-!17 = !{!"0xb\005\000\000", !5, !4} ; [ DW_TAG_lexical_block ] [/Volumes/Data/backedup/dev/os/llvm/debug/<stdin>]
+!13 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "a", line: 3, arg: 1, scope: !4, file: !6, type: !10)
+!14 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "size", line: 3, arg: 2, scope: !4, file: !6, type: !11)
+!15 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "sum", line: 4, scope: !4, file: !6, type: !11)
+!16 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "i", line: 5, scope: !17, file: !6, type: !11)
+!17 = distinct !DILexicalBlock(line: 5, column: 0, file: !5, scope: !4)
 !18 = !{i32 2, !"Dwarf Version", i32 3}
-!19 = !MDLocation(line: 3, scope: !4)
-!20 = !MDLocation(line: 4, scope: !4)
-!21 = !MDLocation(line: 5, scope: !17)
-!22 = !MDLocation(line: 6, scope: !17)
-!26 = !MDLocation(line: 7, scope: !4)
-!27 = !{i32 1, !"Debug Info Version", i32 2}
+!19 = !DILocation(line: 3, scope: !4)
+!20 = !DILocation(line: 4, scope: !4)
+!21 = !DILocation(line: 5, scope: !17)
+!22 = !DILocation(line: 6, scope: !17)
+!26 = !DILocation(line: 7, scope: !4)
+!27 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/LoopVectorize/duplicated-metadata.ll b/test/Transforms/LoopVectorize/duplicated-metadata.ll
index bf2f899dff04..9f7cdef51fe3 100644
--- a/test/Transforms/LoopVectorize/duplicated-metadata.ll
+++ b/test/Transforms/LoopVectorize/duplicated-metadata.ll
@@ -12,8 +12,8 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %a, i64 %indvars.iv
-  %p = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %p = load float, float* %arrayidx, align 4
   %mul = fmul float %p, 2.000000e+00
   store float %mul, float* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/ee-crash.ll b/test/Transforms/LoopVectorize/ee-crash.ll
index a3c0bb89b128..0c387346f867 100644
--- a/test/Transforms/LoopVectorize/ee-crash.ll
+++ b/test/Transforms/LoopVectorize/ee-crash.ll
@@ -11,18 +11,18 @@ target triple = "x86_64-apple-macosx10.8.0"
 define i32 @_Z4foo1Pii(i32* %A, i32 %n, <2 x i32> %q) #0 {
 entry:
   %idx.ext = sext i32 %n to i64
-  %add.ptr = getelementptr inbounds i32* %A, i64 %idx.ext
+  %add.ptr = getelementptr inbounds i32, i32* %A, i64 %idx.ext
   %cmp3.i = icmp eq i32 %n, 0
   br i1 %cmp3.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
 
 for.body.i:                                       ; preds = %entry, %for.body.i
   %__init.addr.05.i = phi i32 [ %add.i, %for.body.i ], [ 0, %entry ]
   %__first.addr.04.i = phi i32* [ %incdec.ptr.i, %for.body.i ], [ %A, %entry ]
-  %0 = load i32* %__first.addr.04.i, align 4
+  %0 = load i32, i32* %__first.addr.04.i, align 4
   %q1 = extractelement <2 x i32> %q, i32 %n
   %q2 = add nsw i32 %0, %q1
   %add.i = add nsw i32 %q2, %__init.addr.05.i
-  %incdec.ptr.i = getelementptr inbounds i32* %__first.addr.04.i, i64 1
+  %incdec.ptr.i = getelementptr inbounds i32, i32* %__first.addr.04.i, i64 1
   %cmp.i = icmp eq i32* %incdec.ptr.i, %add.ptr
   br i1 %cmp.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
 
diff --git a/test/Transforms/LoopVectorize/exact.ll b/test/Transforms/LoopVectorize/exact.ll
index 0a8fbf337347..0a4e0dc17dfa 100644
--- a/test/Transforms/LoopVectorize/exact.ll
+++ b/test/Transforms/LoopVectorize/exact.ll
@@ -11,8 +11,8 @@ entry:
 
 for.body:
   %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %x, i64 %iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %x, i64 %iv
+  %0 = load i32, i32* %arrayidx, align 4
   %conv1 = lshr exact i32 %0, 1
   store i32 %conv1, i32* %arrayidx, align 4
   %iv.next = add nuw nsw i64 %iv, 1
diff --git a/test/Transforms/LoopVectorize/flags.ll b/test/Transforms/LoopVectorize/flags.ll
index 0fc55c897310..fcbc8741741e 100644
--- a/test/Transforms/LoopVectorize/flags.ll
+++ b/test/Transforms/LoopVectorize/flags.ll
@@ -14,8 +14,8 @@ define i32 @flags1(i32 %n, i32* nocapture %A) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 9, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = mul nsw i32 %3, 3
   store i32 %4, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -39,8 +39,8 @@ define i32 @flags2(i32 %n, i32* nocapture %A) nounwind uwtable ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 9, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = mul i32 %3, 3
   store i32 %4, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -66,8 +66,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %q.04 = phi float [ 0.000000e+00, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds float* %s, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %s, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %add = fadd fast float %q.04, %0
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 256
diff --git a/test/Transforms/LoopVectorize/float-reduction.ll b/test/Transforms/LoopVectorize/float-reduction.ll
index 0f064ee2dc85..1310b279f61e 100644
--- a/test/Transforms/LoopVectorize/float-reduction.ll
+++ b/test/Transforms/LoopVectorize/float-reduction.ll
@@ -12,8 +12,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.04 = phi float [ 0.000000e+00, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds float* %A, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %add = fadd fast float %sum.04, %0
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -34,8 +34,8 @@ entry:
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum.04 = phi float [ 0.000000e+00, %entry ], [ %sub, %for.body ]
-  %arrayidx = getelementptr inbounds float* %A, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %sub = fsub fast float %sum.04, %0
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/funcall.ll b/test/Transforms/LoopVectorize/funcall.ll
index e03534fbac37..35c2dfca4b2f 100644
--- a/test/Transforms/LoopVectorize/funcall.ll
+++ b/test/Transforms/LoopVectorize/funcall.ll
@@ -16,8 +16,8 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double* %d, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %d, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %1 = tail call double @llvm.pow.f64(double %0, double %t)
   store double %1, double* %arrayidx, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/gcc-examples.ll b/test/Transforms/LoopVectorize/gcc-examples.ll
index 6c8af0bcd7c9..188090186158 100644
--- a/test/Transforms/LoopVectorize/gcc-examples.ll
+++ b/test/Transforms/LoopVectorize/gcc-examples.ll
@@ -44,12 +44,12 @@ define void @example1() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -84,7 +84,7 @@ define void @example2(i32 %n, i32 %x) nounwind uwtable ssp {
 
 .lr.ph5:                                          ; preds = %0, %.lr.ph5
   %indvars.iv6 = phi i64 [ %indvars.iv.next7, %.lr.ph5 ], [ 0, %0 ]
-  %3 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv6
+  %3 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv6
   store i32 %x, i32* %3, align 4
   %indvars.iv.next7 = add i64 %indvars.iv6, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next7 to i32
@@ -95,12 +95,12 @@ define void @example2(i32 %n, i32 %x) nounwind uwtable ssp {
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ %i.0.lcssa, %.preheader ]
   %.02 = phi i32 [ %4, %.lr.ph ], [ %n, %.preheader ]
   %4 = add nsw i32 %.02, -1
-  %5 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %6 = load i32* %5, align 4
-  %7 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %8 = load i32* %7, align 4
+  %5 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %6 = load i32, i32* %5, align 4
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %8 = load i32, i32* %7, align 4
   %9 = and i32 %8, %6
-  %10 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %10 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %9, i32* %10, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %11 = icmp eq i32 %4, 0
@@ -128,9 +128,9 @@ define void @example3(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
   %.014 = phi i32* [ %5, %.lr.ph ], [ %p, %0 ]
   %.023 = phi i32* [ %3, %.lr.ph ], [ %q, %0 ]
   %2 = add nsw i32 %.05, -1
-  %3 = getelementptr inbounds i32* %.023, i64 1
-  %4 = load i32* %.023, align 16
-  %5 = getelementptr inbounds i32* %.014, i64 1
+  %3 = getelementptr inbounds i32, i32* %.023, i64 1
+  %4 = load i32, i32* %.023, align 16
+  %5 = getelementptr inbounds i32, i32* %.014, i64 1
   store i32 %4, i32* %.014, align 16
   %6 = icmp eq i32 %2, 0
   br i1 %6, label %._crit_edge, label %.lr.ph
@@ -161,10 +161,10 @@ define void @example4(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
   %4 = phi i32 [ %9, %.lr.ph10 ], [ %1, %0 ]
   %.018 = phi i32* [ %8, %.lr.ph10 ], [ %p, %0 ]
   %.027 = phi i32* [ %5, %.lr.ph10 ], [ %q, %0 ]
-  %5 = getelementptr inbounds i32* %.027, i64 1
-  %6 = load i32* %.027, align 16
+  %5 = getelementptr inbounds i32, i32* %.027, i64 1
+  %6 = load i32, i32* %.027, align 16
   %7 = add nsw i32 %6, 5
-  %8 = getelementptr inbounds i32* %.018, i64 1
+  %8 = getelementptr inbounds i32, i32* %.018, i64 1
   store i32 %7, i32* %.018, align 16
   %9 = add nsw i32 %4, -1
   %10 = icmp eq i32 %4, 0
@@ -176,13 +176,13 @@ define void @example4(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
 .lr.ph6:                                          ; preds = %.preheader4, %.lr.ph6
   %indvars.iv11 = phi i64 [ %indvars.iv.next12, %.lr.ph6 ], [ 0, %.preheader4 ]
   %indvars.iv.next12 = add i64 %indvars.iv11, 1
-  %11 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv.next12
-  %12 = load i32* %11, align 4
+  %11 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv.next12
+  %12 = load i32, i32* %11, align 4
   %13 = add nsw i64 %indvars.iv11, 3
-  %14 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %13
-  %15 = load i32* %14, align 4
+  %14 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %13
+  %15 = load i32, i32* %14, align 4
   %16 = add nsw i32 %15, %12
-  %17 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv11
+  %17 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv11
   store i32 %16, i32* %17, align 4
   %lftr.wideiv13 = trunc i64 %indvars.iv.next12 to i32
   %exitcond14 = icmp eq i32 %lftr.wideiv13, %1
@@ -190,11 +190,11 @@ define void @example4(i32 %n, i32* noalias nocapture %p, i32* noalias nocapture
 
 .lr.ph:                                           ; preds = %.preheader, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %.preheader ]
-  %18 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
-  %19 = load i32* %18, align 4
+  %18 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %19 = load i32, i32* %18, align 4
   %20 = icmp sgt i32 %19, 4
   %21 = select i1 %20, i32 4, i32 0
-  %22 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %22 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
   store i32 %21, i32* %22, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -223,7 +223,7 @@ define void @example8(i32 %x) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %.preheader
   %indvars.iv = phi i64 [ 0, %.preheader ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [32 x [1024 x i32]]* @G, i64 0, i64 %indvars.iv3, i64 %indvars.iv
+  %2 = getelementptr inbounds [32 x [1024 x i32]], [32 x [1024 x i32]]* @G, i64 0, i64 %indvars.iv3, i64 %indvars.iv
   store i32 %x, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -249,10 +249,10 @@ define i32 @example9() nounwind uwtable readonly ssp {
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
   %diff.01 = phi i32 [ 0, %0 ], [ %7, %1 ]
-  %2 = getelementptr inbounds [1024 x i32]* @ub, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [1024 x i32]* @uc, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [1024 x i32], [1024 x i32]* @ub, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [1024 x i32], [1024 x i32]* @uc, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add i32 %3, %diff.01
   %7 = sub i32 %6, %5
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -276,19 +276,19 @@ define void @example10a(i16* noalias nocapture %sa, i16* noalias nocapture %sb,
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i32* %ib, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds i32* %ic, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds i32, i32* %ib, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds i32, i32* %ic, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds i32* %ia, i64 %indvars.iv
+  %7 = getelementptr inbounds i32, i32* %ia, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
-  %8 = getelementptr inbounds i16* %sb, i64 %indvars.iv
-  %9 = load i16* %8, align 2
-  %10 = getelementptr inbounds i16* %sc, i64 %indvars.iv
-  %11 = load i16* %10, align 2
+  %8 = getelementptr inbounds i16, i16* %sb, i64 %indvars.iv
+  %9 = load i16, i16* %8, align 2
+  %10 = getelementptr inbounds i16, i16* %sc, i64 %indvars.iv
+  %11 = load i16, i16* %10, align 2
   %12 = add i16 %11, %9
-  %13 = getelementptr inbounds i16* %sa, i64 %indvars.iv
+  %13 = getelementptr inbounds i16, i16* %sa, i64 %indvars.iv
   store i16 %12, i16* %13, align 2
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -309,10 +309,10 @@ define void @example10b(i16* noalias nocapture %sa, i16* noalias nocapture %sb,
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds i16* %sb, i64 %indvars.iv
-  %3 = load i16* %2, align 2
+  %2 = getelementptr inbounds i16, i16* %sb, i64 %indvars.iv
+  %3 = load i16, i16* %2, align 2
   %4 = sext i16 %3 to i32
-  %5 = getelementptr inbounds i32* %ia, i64 %indvars.iv
+  %5 = getelementptr inbounds i32, i32* %ia, i64 %indvars.iv
   store i32 %4, i32* %5, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -340,23 +340,23 @@ define void @example11() nounwind uwtable ssp {
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
   %2 = shl nsw i64 %indvars.iv, 1
   %3 = or i64 %2, 1
-  %4 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %3
-  %5 = load i32* %4, align 4
-  %6 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %3
-  %7 = load i32* %6, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %3
+  %5 = load i32, i32* %4, align 4
+  %6 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %3
+  %7 = load i32, i32* %6, align 4
   %8 = mul nsw i32 %7, %5
-  %9 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %2
-  %10 = load i32* %9, align 8
-  %11 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %2
-  %12 = load i32* %11, align 8
+  %9 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %2
+  %10 = load i32, i32* %9, align 8
+  %11 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %2
+  %12 = load i32, i32* %11, align 8
   %13 = mul nsw i32 %12, %10
   %14 = sub nsw i32 %8, %13
-  %15 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %15 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %14, i32* %15, align 4
   %16 = mul nsw i32 %7, %10
   %17 = mul nsw i32 %12, %5
   %18 = add nsw i32 %17, %16
-  %19 = getelementptr inbounds [2048 x i32]* @d, i64 0, i64 %indvars.iv
+  %19 = getelementptr inbounds [2048 x i32], [2048 x i32]* @d, i64 0, i64 %indvars.iv
   store i32 %18, i32* %19, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -376,7 +376,7 @@ define void @example12() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   %3 = trunc i64 %indvars.iv to i32
   store i32 %3, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -388,28 +388,27 @@ define void @example12() nounwind uwtable ssp {
   ret void
 }
 
-; Can't vectorize because of reductions.
 ;CHECK-LABEL: @example13(
-;CHECK-NOT: <4 x i32>
+;CHECK: <4 x i32>
 ;CHECK: ret void
 define void @example13(i32** nocapture %A, i32** nocapture %B, i32* nocapture %out) nounwind uwtable ssp {
   br label %.preheader
 
 .preheader:                                       ; preds = %14, %0
   %indvars.iv4 = phi i64 [ 0, %0 ], [ %indvars.iv.next5, %14 ]
-  %1 = getelementptr inbounds i32** %A, i64 %indvars.iv4
-  %2 = load i32** %1, align 8
-  %3 = getelementptr inbounds i32** %B, i64 %indvars.iv4
-  %4 = load i32** %3, align 8
+  %1 = getelementptr inbounds i32*, i32** %A, i64 %indvars.iv4
+  %2 = load i32*, i32** %1, align 8
+  %3 = getelementptr inbounds i32*, i32** %B, i64 %indvars.iv4
+  %4 = load i32*, i32** %3, align 8
   br label %5
 
 ; <label>:5                                       ; preds = %.preheader, %5
   %indvars.iv = phi i64 [ 0, %.preheader ], [ %indvars.iv.next, %5 ]
   %diff.02 = phi i32 [ 0, %.preheader ], [ %11, %5 ]
-  %6 = getelementptr inbounds i32* %2, i64 %indvars.iv
-  %7 = load i32* %6, align 4
-  %8 = getelementptr inbounds i32* %4, i64 %indvars.iv
-  %9 = load i32* %8, align 4
+  %6 = getelementptr inbounds i32, i32* %2, i64 %indvars.iv
+  %7 = load i32, i32* %6, align 4
+  %8 = getelementptr inbounds i32, i32* %4, i64 %indvars.iv
+  %9 = load i32, i32* %8, align 4
   %10 = add i32 %7, %diff.02
   %11 = sub i32 %10, %9
   %indvars.iv.next = add i64 %indvars.iv, 8
@@ -418,7 +417,7 @@ define void @example13(i32** nocapture %A, i32** nocapture %B, i32* nocapture %o
   br i1 %13, label %5, label %14
 
 ; <label>:14                                      ; preds = %5
-  %15 = getelementptr inbounds i32* %out, i64 %indvars.iv4
+  %15 = getelementptr inbounds i32, i32* %out, i64 %indvars.iv4
   store i32 %11, i32* %15, align 4
   %indvars.iv.next5 = add i64 %indvars.iv4, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next5 to i32
@@ -445,14 +444,14 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
 ; <label>:0                                       ; preds = %0, %.preheader
   %indvars.iv = phi i64 [ 0, %.preheader ], [ %indvars.iv.next, %0 ]
   %sum.12 = phi i32 [ %sum.05, %.preheader ], [ %10, %0 ]
-  %1 = getelementptr inbounds i32** %in, i64 %indvars.iv
-  %2 = load i32** %1, align 8
-  %3 = getelementptr inbounds i32* %2, i64 %indvars.iv7
-  %4 = load i32* %3, align 4
-  %5 = getelementptr inbounds i32** %coeff, i64 %indvars.iv
-  %6 = load i32** %5, align 8
-  %7 = getelementptr inbounds i32* %6, i64 %indvars.iv7
-  %8 = load i32* %7, align 4
+  %1 = getelementptr inbounds i32*, i32** %in, i64 %indvars.iv
+  %2 = load i32*, i32** %1, align 8
+  %3 = getelementptr inbounds i32, i32* %2, i64 %indvars.iv7
+  %4 = load i32, i32* %3, align 4
+  %5 = getelementptr inbounds i32*, i32** %coeff, i64 %indvars.iv
+  %6 = load i32*, i32** %5, align 8
+  %7 = getelementptr inbounds i32, i32* %6, i64 %indvars.iv7
+  %8 = load i32, i32* %7, align 4
   %9 = mul nsw i32 %8, %4
   %10 = add nsw i32 %9, %sum.12
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -479,14 +478,14 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   %indvars.iv.1 = phi i64 [ 0, %.preheader.1 ], [ %13, %12 ]
   %sum.12.1 = phi i32 [ %sum.05.1, %.preheader.1 ], [ %23, %12 ]
   %13 = add nsw i64 %indvars.iv.1, 1
-  %14 = getelementptr inbounds i32** %in, i64 %13
-  %15 = load i32** %14, align 8
-  %16 = getelementptr inbounds i32* %15, i64 %indvars.iv7.1
-  %17 = load i32* %16, align 4
-  %18 = getelementptr inbounds i32** %coeff, i64 %indvars.iv.1
-  %19 = load i32** %18, align 8
-  %20 = getelementptr inbounds i32* %19, i64 %indvars.iv7.1
-  %21 = load i32* %20, align 4
+  %14 = getelementptr inbounds i32*, i32** %in, i64 %13
+  %15 = load i32*, i32** %14, align 8
+  %16 = getelementptr inbounds i32, i32* %15, i64 %indvars.iv7.1
+  %17 = load i32, i32* %16, align 4
+  %18 = getelementptr inbounds i32*, i32** %coeff, i64 %indvars.iv.1
+  %19 = load i32*, i32** %18, align 8
+  %20 = getelementptr inbounds i32, i32* %19, i64 %indvars.iv7.1
+  %21 = load i32, i32* %20, align 4
   %22 = mul nsw i32 %21, %17
   %23 = add nsw i32 %22, %sum.12.1
   %lftr.wideiv.1 = trunc i64 %13 to i32
@@ -500,7 +499,7 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   br i1 %exitcond10.1, label %.preheader3.2, label %.preheader.1
 
 .preheader3.2:                                    ; preds = %24
-  %25 = getelementptr inbounds i32* %out, i64 1
+  %25 = getelementptr inbounds i32, i32* %out, i64 1
   store i32 %23, i32* %25, align 4
   br label %.preheader.2
 
@@ -513,14 +512,14 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   %indvars.iv.2 = phi i64 [ 0, %.preheader.2 ], [ %indvars.iv.next.2, %26 ]
   %sum.12.2 = phi i32 [ %sum.05.2, %.preheader.2 ], [ %37, %26 ]
   %27 = add nsw i64 %indvars.iv.2, 2
-  %28 = getelementptr inbounds i32** %in, i64 %27
-  %29 = load i32** %28, align 8
-  %30 = getelementptr inbounds i32* %29, i64 %indvars.iv7.2
-  %31 = load i32* %30, align 4
-  %32 = getelementptr inbounds i32** %coeff, i64 %indvars.iv.2
-  %33 = load i32** %32, align 8
-  %34 = getelementptr inbounds i32* %33, i64 %indvars.iv7.2
-  %35 = load i32* %34, align 4
+  %28 = getelementptr inbounds i32*, i32** %in, i64 %27
+  %29 = load i32*, i32** %28, align 8
+  %30 = getelementptr inbounds i32, i32* %29, i64 %indvars.iv7.2
+  %31 = load i32, i32* %30, align 4
+  %32 = getelementptr inbounds i32*, i32** %coeff, i64 %indvars.iv.2
+  %33 = load i32*, i32** %32, align 8
+  %34 = getelementptr inbounds i32, i32* %33, i64 %indvars.iv7.2
+  %35 = load i32, i32* %34, align 4
   %36 = mul nsw i32 %35, %31
   %37 = add nsw i32 %36, %sum.12.2
   %indvars.iv.next.2 = add i64 %indvars.iv.2, 1
@@ -535,7 +534,7 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   br i1 %exitcond10.2, label %.preheader3.3, label %.preheader.2
 
 .preheader3.3:                                    ; preds = %38
-  %39 = getelementptr inbounds i32* %out, i64 2
+  %39 = getelementptr inbounds i32, i32* %out, i64 2
   store i32 %37, i32* %39, align 4
   br label %.preheader.3
 
@@ -548,14 +547,14 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   %indvars.iv.3 = phi i64 [ 0, %.preheader.3 ], [ %indvars.iv.next.3, %40 ]
   %sum.12.3 = phi i32 [ %sum.05.3, %.preheader.3 ], [ %51, %40 ]
   %41 = add nsw i64 %indvars.iv.3, 3
-  %42 = getelementptr inbounds i32** %in, i64 %41
-  %43 = load i32** %42, align 8
-  %44 = getelementptr inbounds i32* %43, i64 %indvars.iv7.3
-  %45 = load i32* %44, align 4
-  %46 = getelementptr inbounds i32** %coeff, i64 %indvars.iv.3
-  %47 = load i32** %46, align 8
-  %48 = getelementptr inbounds i32* %47, i64 %indvars.iv7.3
-  %49 = load i32* %48, align 4
+  %42 = getelementptr inbounds i32*, i32** %in, i64 %41
+  %43 = load i32*, i32** %42, align 8
+  %44 = getelementptr inbounds i32, i32* %43, i64 %indvars.iv7.3
+  %45 = load i32, i32* %44, align 4
+  %46 = getelementptr inbounds i32*, i32** %coeff, i64 %indvars.iv.3
+  %47 = load i32*, i32** %46, align 8
+  %48 = getelementptr inbounds i32, i32* %47, i64 %indvars.iv7.3
+  %49 = load i32, i32* %48, align 4
   %50 = mul nsw i32 %49, %45
   %51 = add nsw i32 %50, %sum.12.3
   %indvars.iv.next.3 = add i64 %indvars.iv.3, 1
@@ -570,7 +569,7 @@ define void @example14(i32** nocapture %in, i32** nocapture %coeff, i32* nocaptu
   br i1 %exitcond10.3, label %53, label %.preheader.3
 
 ; <label>:53                                      ; preds = %52
-  %54 = getelementptr inbounds i32* %out, i64 3
+  %54 = getelementptr inbounds i32, i32* %out, i64 3
   store i32 %51, i32* %54, align 4
   ret void
 }
@@ -591,8 +590,8 @@ define i32 @example21(i32* nocapture %b, i32 %n) nounwind uwtable readonly ssp {
   %indvars.iv = phi i64 [ %2, %.lr.ph ], [ %indvars.iv.next, %3 ]
   %a.02 = phi i32 [ 0, %.lr.ph ], [ %6, %3 ]
   %indvars.iv.next = add i64 %indvars.iv, -1
-  %4 = getelementptr inbounds i32* %b, i64 %indvars.iv.next
-  %5 = load i32* %4, align 4
+  %4 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv.next
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %a.02
   %7 = trunc i64 %indvars.iv.next to i32
   %8 = icmp sgt i32 %7, 0
@@ -613,11 +612,11 @@ define void @example23(i16* nocapture %src, i32* nocapture %dst) nounwind uwtabl
   %.04 = phi i16* [ %src, %0 ], [ %2, %1 ]
   %.013 = phi i32* [ %dst, %0 ], [ %6, %1 ]
   %i.02 = phi i32 [ 0, %0 ], [ %7, %1 ]
-  %2 = getelementptr inbounds i16* %.04, i64 1
-  %3 = load i16* %.04, align 2
+  %2 = getelementptr inbounds i16, i16* %.04, i64 1
+  %3 = load i16, i16* %.04, align 2
   %4 = zext i16 %3 to i32
   %5 = shl nuw nsw i32 %4, 7
-  %6 = getelementptr inbounds i32* %.013, i64 1
+  %6 = getelementptr inbounds i32, i32* %.013, i64 1
   store i32 %5, i32* %.013, align 4
   %7 = add nsw i32 %i.02, 1
   %exitcond = icmp eq i32 %7, 256
@@ -635,14 +634,14 @@ define void @example24(i16 signext %x, i16 signext %y) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [1024 x float]* @fa, i64 0, i64 %indvars.iv
-  %3 = load float* %2, align 4
-  %4 = getelementptr inbounds [1024 x float]* @fb, i64 0, i64 %indvars.iv
-  %5 = load float* %4, align 4
+  %2 = getelementptr inbounds [1024 x float], [1024 x float]* @fa, i64 0, i64 %indvars.iv
+  %3 = load float, float* %2, align 4
+  %4 = getelementptr inbounds [1024 x float], [1024 x float]* @fb, i64 0, i64 %indvars.iv
+  %5 = load float, float* %4, align 4
   %6 = fcmp olt float %3, %5
   %x.y = select i1 %6, i16 %x, i16 %y
   %7 = sext i16 %x.y to i32
-  %8 = getelementptr inbounds [1024 x i32]* @ic, i64 0, i64 %indvars.iv
+  %8 = getelementptr inbounds [1024 x i32], [1024 x i32]* @ic, i64 0, i64 %indvars.iv
   store i32 %7, i32* %8, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -662,19 +661,19 @@ define void @example25() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [1024 x float]* @da, i64 0, i64 %indvars.iv
-  %3 = load float* %2, align 4
-  %4 = getelementptr inbounds [1024 x float]* @db, i64 0, i64 %indvars.iv
-  %5 = load float* %4, align 4
+  %2 = getelementptr inbounds [1024 x float], [1024 x float]* @da, i64 0, i64 %indvars.iv
+  %3 = load float, float* %2, align 4
+  %4 = getelementptr inbounds [1024 x float], [1024 x float]* @db, i64 0, i64 %indvars.iv
+  %5 = load float, float* %4, align 4
   %6 = fcmp olt float %3, %5
-  %7 = getelementptr inbounds [1024 x float]* @dc, i64 0, i64 %indvars.iv
-  %8 = load float* %7, align 4
-  %9 = getelementptr inbounds [1024 x float]* @dd, i64 0, i64 %indvars.iv
-  %10 = load float* %9, align 4
+  %7 = getelementptr inbounds [1024 x float], [1024 x float]* @dc, i64 0, i64 %indvars.iv
+  %8 = load float, float* %7, align 4
+  %9 = getelementptr inbounds [1024 x float], [1024 x float]* @dd, i64 0, i64 %indvars.iv
+  %10 = load float, float* %9, align 4
   %11 = fcmp olt float %8, %10
   %12 = and i1 %6, %11
   %13 = zext i1 %12 to i32
-  %14 = getelementptr inbounds [1024 x i32]* @dj, i64 0, i64 %indvars.iv
+  %14 = getelementptr inbounds [1024 x i32], [1024 x i32]* @dj, i64 0, i64 %indvars.iv
   store i32 %13, i32* %14, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/global_alias.ll b/test/Transforms/LoopVectorize/global_alias.ll
index 3f11ce8bf213..84fa48cd5148 100644
--- a/test/Transforms/LoopVectorize/global_alias.ll
+++ b/test/Transforms/LoopVectorize/global_alias.ll
@@ -35,31 +35,31 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %1 = load i32, i32* %i, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
-  %arrayidx1 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
+  %4 = load i32, i32* %i, align 4
+  %arrayidx1 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
   store i32 %add, i32* %arrayidx1, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx2, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx2, align 4
   ret i32 %7
 }
 
@@ -83,32 +83,32 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 90
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %add = add nsw i32 %1, 10
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %add
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %add
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add1 = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
+  %4 = load i32, i32* %i, align 4
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
   store i32 %add1, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -132,32 +132,32 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %1 = load i32, i32* %i, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %add1 = add nsw i32 %4, 10
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add1
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add1
   store i32 %add, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -184,34 +184,34 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32** @PB, align 4
-  %2 = load i32* %i, align 4
-  %add.ptr = getelementptr inbounds i32* %1, i32 %2
-  %3 = load i32* %add.ptr, align 4
-  %4 = load i32* %a.addr, align 4
+  %1 = load i32*, i32** @PB, align 4
+  %2 = load i32, i32* %i, align 4
+  %add.ptr = getelementptr inbounds i32, i32* %1, i32 %2
+  %3 = load i32, i32* %add.ptr, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %3, %4
-  %5 = load i32** @PA, align 4
-  %6 = load i32* %i, align 4
-  %add.ptr1 = getelementptr inbounds i32* %5, i32 %6
+  %5 = load i32*, i32** @PA, align 4
+  %6 = load i32, i32* %i, align 4
+  %add.ptr1 = getelementptr inbounds i32, i32* %5, i32 %6
   store i32 %add, i32* %add.ptr1, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32** @PA, align 4
-  %9 = load i32* %a.addr, align 4
-  %add.ptr2 = getelementptr inbounds i32* %8, i32 %9
-  %10 = load i32* %add.ptr2, align 4
+  %8 = load i32*, i32** @PA, align 4
+  %9 = load i32, i32* %a.addr, align 4
+  %add.ptr2 = getelementptr inbounds i32, i32* %8, i32 %9
+  %10 = load i32, i32* %add.ptr2, align 4
   ret i32 %10
 }
 
@@ -237,37 +237,37 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
-  %2 = load i32* %N, align 4
-  %arrayidx = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 2), i32 0, i32 %2
-  %arrayidx1 = getelementptr inbounds [100 x i32]* %arrayidx, i32 0, i32 %1
-  %3 = load i32* %arrayidx1, align 4
-  %4 = load i32* %a.addr, align 4
+  %1 = load i32, i32* %i, align 4
+  %2 = load i32, i32* %N, align 4
+  %arrayidx = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 2), i32 0, i32 %2
+  %arrayidx1 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx, i32 0, i32 %1
+  %3 = load i32, i32* %arrayidx1, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %3, %4
-  %5 = load i32* %i, align 4
-  %6 = load i32* %N, align 4
-  %arrayidx2 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
-  %arrayidx3 = getelementptr inbounds [100 x i32]* %arrayidx2, i32 0, i32 %5
+  %5 = load i32, i32* %i, align 4
+  %6 = load i32, i32* %N, align 4
+  %arrayidx2 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx2, i32 0, i32 %5
   store i32 %add, i32* %arrayidx3, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32* %a.addr, align 4
-  %9 = load i32* %N, align 4
-  %arrayidx4 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
-  %arrayidx5 = getelementptr inbounds [100 x i32]* %arrayidx4, i32 0, i32 %8
-  %10 = load i32* %arrayidx5, align 4
+  %8 = load i32, i32* %a.addr, align 4
+  %9 = load i32, i32* %N, align 4
+  %arrayidx4 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
+  %arrayidx5 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx4, i32 0, i32 %8
+  %10 = load i32, i32* %arrayidx5, align 4
   ret i32 %10
 }
 
@@ -293,38 +293,38 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
-  %2 = load i32* %N, align 4
+  %1 = load i32, i32* %i, align 4
+  %2 = load i32, i32* %N, align 4
   %add = add nsw i32 %2, 1
-  %arrayidx = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %add
-  %arrayidx1 = getelementptr inbounds [100 x i32]* %arrayidx, i32 0, i32 %1
-  %3 = load i32* %arrayidx1, align 4
-  %4 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %add
+  %arrayidx1 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx, i32 0, i32 %1
+  %3 = load i32, i32* %arrayidx1, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add2 = add nsw i32 %3, %4
-  %5 = load i32* %i, align 4
-  %6 = load i32* %N, align 4
-  %arrayidx3 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
-  %arrayidx4 = getelementptr inbounds [100 x i32]* %arrayidx3, i32 0, i32 %5
+  %5 = load i32, i32* %i, align 4
+  %6 = load i32, i32* %N, align 4
+  %arrayidx3 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx3, i32 0, i32 %5
   store i32 %add2, i32* %arrayidx4, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32* %a.addr, align 4
-  %9 = load i32* %N, align 4
-  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
-  %arrayidx6 = getelementptr inbounds [100 x i32]* %arrayidx5, i32 0, i32 %8
-  %10 = load i32* %arrayidx6, align 4
+  %8 = load i32, i32* %a.addr, align 4
+  %9 = load i32, i32* %N, align 4
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
+  %arrayidx6 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx5, i32 0, i32 %8
+  %10 = load i32, i32* %arrayidx6, align 4
   ret i32 %10
 }
 
@@ -347,35 +347,35 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %sub2 = sub nsw i32 100, %4
   %sub3 = sub nsw i32 %sub2, 1
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
   store i32 %add, i32* %arrayidx4, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx5 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx5, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx5 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx5, align 4
   ret i32 %7
 }
 
@@ -399,35 +399,35 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 90
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 10
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %sub2 = sub nsw i32 100, %4
   %sub3 = sub nsw i32 %sub2, 1
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
   store i32 %add, i32* %arrayidx4, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx5 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx5, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx5 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx5, align 4
   ret i32 %7
 }
 
@@ -451,35 +451,35 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %sub2 = sub nsw i32 100, %4
   %sub3 = sub nsw i32 %sub2, 10
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
   store i32 %add, i32* %arrayidx4, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx5 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx5, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx5 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx5, align 4
   ret i32 %7
 }
 
@@ -506,40 +506,40 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32** @PB, align 4
-  %add.ptr = getelementptr inbounds i32* %1, i32 100
-  %2 = load i32* %i, align 4
+  %1 = load i32*, i32** @PB, align 4
+  %add.ptr = getelementptr inbounds i32, i32* %1, i32 100
+  %2 = load i32, i32* %i, align 4
   %idx.neg = sub i32 0, %2
-  %add.ptr1 = getelementptr inbounds i32* %add.ptr, i32 %idx.neg
-  %add.ptr2 = getelementptr inbounds i32* %add.ptr1, i32 -1
-  %3 = load i32* %add.ptr2, align 4
-  %4 = load i32* %a.addr, align 4
+  %add.ptr1 = getelementptr inbounds i32, i32* %add.ptr, i32 %idx.neg
+  %add.ptr2 = getelementptr inbounds i32, i32* %add.ptr1, i32 -1
+  %3 = load i32, i32* %add.ptr2, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %3, %4
-  %5 = load i32** @PA, align 4
-  %add.ptr3 = getelementptr inbounds i32* %5, i32 100
-  %6 = load i32* %i, align 4
+  %5 = load i32*, i32** @PA, align 4
+  %add.ptr3 = getelementptr inbounds i32, i32* %5, i32 100
+  %6 = load i32, i32* %i, align 4
   %idx.neg4 = sub i32 0, %6
-  %add.ptr5 = getelementptr inbounds i32* %add.ptr3, i32 %idx.neg4
-  %add.ptr6 = getelementptr inbounds i32* %add.ptr5, i32 -1
+  %add.ptr5 = getelementptr inbounds i32, i32* %add.ptr3, i32 %idx.neg4
+  %add.ptr6 = getelementptr inbounds i32, i32* %add.ptr5, i32 -1
   store i32 %add, i32* %add.ptr6, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32** @PA, align 4
-  %9 = load i32* %a.addr, align 4
-  %add.ptr7 = getelementptr inbounds i32* %8, i32 %9
-  %10 = load i32* %add.ptr7, align 4
+  %8 = load i32*, i32** @PA, align 4
+  %9 = load i32, i32* %a.addr, align 4
+  %add.ptr7 = getelementptr inbounds i32, i32* %8, i32 %9
+  %10 = load i32, i32* %add.ptr7, align 4
   ret i32 %10
 }
 
@@ -565,41 +565,41 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %2 = load i32* %N, align 4
-  %arrayidx = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 2), i32 0, i32 %2
-  %arrayidx2 = getelementptr inbounds [100 x i32]* %arrayidx, i32 0, i32 %sub1
-  %3 = load i32* %arrayidx2, align 4
-  %4 = load i32* %a.addr, align 4
+  %2 = load i32, i32* %N, align 4
+  %arrayidx = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 2), i32 0, i32 %2
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx, i32 0, i32 %sub1
+  %3 = load i32, i32* %arrayidx2, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %3, %4
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %sub3 = sub nsw i32 100, %5
   %sub4 = sub nsw i32 %sub3, 1
-  %6 = load i32* %N, align 4
-  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
-  %arrayidx6 = getelementptr inbounds [100 x i32]* %arrayidx5, i32 0, i32 %sub4
+  %6 = load i32, i32* %N, align 4
+  %arrayidx5 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
+  %arrayidx6 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx5, i32 0, i32 %sub4
   store i32 %add, i32* %arrayidx6, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32* %a.addr, align 4
-  %9 = load i32* %N, align 4
-  %arrayidx7 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
-  %arrayidx8 = getelementptr inbounds [100 x i32]* %arrayidx7, i32 0, i32 %8
-  %10 = load i32* %arrayidx8, align 4
+  %8 = load i32, i32* %a.addr, align 4
+  %9 = load i32, i32* %N, align 4
+  %arrayidx7 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
+  %arrayidx8 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx7, i32 0, i32 %8
+  %10 = load i32, i32* %arrayidx8, align 4
   ret i32 %10
 }
 
@@ -625,42 +625,42 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %2 = load i32* %N, align 4
+  %2 = load i32, i32* %N, align 4
   %add = add nsw i32 %2, 1
-  %arrayidx = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %add
-  %arrayidx2 = getelementptr inbounds [100 x i32]* %arrayidx, i32 0, i32 %sub1
-  %3 = load i32* %arrayidx2, align 4
-  %4 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %add
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx, i32 0, i32 %sub1
+  %3 = load i32, i32* %arrayidx2, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add3 = add nsw i32 %3, %4
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %sub4 = sub nsw i32 100, %5
   %sub5 = sub nsw i32 %sub4, 1
-  %6 = load i32* %N, align 4
-  %arrayidx6 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
-  %arrayidx7 = getelementptr inbounds [100 x i32]* %arrayidx6, i32 0, i32 %sub5
+  %6 = load i32, i32* %N, align 4
+  %arrayidx6 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %6
+  %arrayidx7 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx6, i32 0, i32 %sub5
   store i32 %add3, i32* %arrayidx7, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32* %a.addr, align 4
-  %9 = load i32* %N, align 4
-  %arrayidx8 = getelementptr inbounds [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
-  %arrayidx9 = getelementptr inbounds [100 x i32]* %arrayidx8, i32 0, i32 %8
-  %10 = load i32* %arrayidx9, align 4
+  %8 = load i32, i32* %a.addr, align 4
+  %9 = load i32, i32* %N, align 4
+  %arrayidx8 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* getelementptr inbounds (%struct.anon.0, %struct.anon.0* @Bar, i32 0, i32 0), i32 0, i32 %9
+  %arrayidx9 = getelementptr inbounds [100 x i32], [100 x i32]* %arrayidx8, i32 0, i32 %8
+  %10 = load i32, i32* %arrayidx9, align 4
   ret i32 %10
 }
 
@@ -684,32 +684,32 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %add = add nsw i32 %1, 4
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add1 = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
+  %4 = load i32, i32* %i, align 4
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
   store i32 %add1, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -733,35 +733,35 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 5
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %sub2 = sub nsw i32 100, %4
   %sub3 = sub nsw i32 %sub2, 1
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub3
   store i32 %add, i32* %arrayidx4, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx5 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx5, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx5 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx5, align 4
   ret i32 %7
 }
 
@@ -789,33 +789,33 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
+  %4 = load i32, i32* %i, align 4
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
   store i32 %add, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -839,33 +839,33 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %1 = load i32, i32* %i, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %4
   %sub1 = sub nsw i32 %sub, 1
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub1
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %sub1
   store i32 %add, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -889,37 +889,37 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32** @PB, align 4
-  %add.ptr = getelementptr inbounds i32* %1, i32 100
-  %2 = load i32* %i, align 4
+  %1 = load i32*, i32** @PB, align 4
+  %add.ptr = getelementptr inbounds i32, i32* %1, i32 100
+  %2 = load i32, i32* %i, align 4
   %idx.neg = sub i32 0, %2
-  %add.ptr1 = getelementptr inbounds i32* %add.ptr, i32 %idx.neg
-  %add.ptr2 = getelementptr inbounds i32* %add.ptr1, i32 -1
-  %3 = load i32* %add.ptr2, align 4
-  %4 = load i32* %a.addr, align 4
+  %add.ptr1 = getelementptr inbounds i32, i32* %add.ptr, i32 %idx.neg
+  %add.ptr2 = getelementptr inbounds i32, i32* %add.ptr1, i32 -1
+  %3 = load i32, i32* %add.ptr2, align 4
+  %4 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %3, %4
-  %5 = load i32** @PA, align 4
-  %6 = load i32* %i, align 4
-  %add.ptr3 = getelementptr inbounds i32* %5, i32 %6
+  %5 = load i32*, i32** @PA, align 4
+  %6 = load i32, i32* %i, align 4
+  %add.ptr3 = getelementptr inbounds i32, i32* %5, i32 %6
   store i32 %add, i32* %add.ptr3, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %7 = load i32* %i, align 4
+  %7 = load i32, i32* %i, align 4
   %inc = add nsw i32 %7, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %8 = load i32** @PA, align 4
-  %9 = load i32* %a.addr, align 4
-  %add.ptr4 = getelementptr inbounds i32* %8, i32 %9
-  %10 = load i32* %add.ptr4, align 4
+  %8 = load i32*, i32** @PA, align 4
+  %9 = load i32, i32* %a.addr, align 4
+  %add.ptr4 = getelementptr inbounds i32, i32* %8, i32 %9
+  %10 = load i32, i32* %add.ptr4, align 4
   ret i32 %10
 }
 
@@ -946,34 +946,34 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 1
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %add2 = add nsw i32 %4, 10
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add2
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add2
   store i32 %add, i32* %arrayidx3, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx4, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx4, align 4
   ret i32 %7
 }
 
@@ -996,33 +996,33 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 10
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
-  %arrayidx2 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
+  %4 = load i32, i32* %i, align 4
+  %arrayidx2 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %4
   store i32 %add, i32* %arrayidx2, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx3, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx3, align 4
   ret i32 %7
 }
 
@@ -1045,33 +1045,33 @@ entry:
   br label %for.cond
 
 for.cond:                                         ; preds = %for.inc, %entry
-  %0 = load i32* %i, align 4
+  %0 = load i32, i32* %i, align 4
   %cmp = icmp slt i32 %0, 100
   br i1 %cmp, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.cond
-  %1 = load i32* %i, align 4
+  %1 = load i32, i32* %i, align 4
   %sub = sub nsw i32 100, %1
   %sub1 = sub nsw i32 %sub, 10
-  %arrayidx = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
-  %2 = load i32* %arrayidx, align 4
-  %3 = load i32* %a.addr, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 2), i32 0, i32 %sub1
+  %2 = load i32, i32* %arrayidx, align 4
+  %3 = load i32, i32* %a.addr, align 4
   %add = add nsw i32 %2, %3
-  %4 = load i32* %i, align 4
+  %4 = load i32, i32* %i, align 4
   %add2 = add nsw i32 %4, 10
-  %arrayidx3 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add2
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %add2
   store i32 %add, i32* %arrayidx3, align 4
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
-  %5 = load i32* %i, align 4
+  %5 = load i32, i32* %i, align 4
   %inc = add nsw i32 %5, 1
   store i32 %inc, i32* %i, align 4
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %6 = load i32* %a.addr, align 4
-  %arrayidx4 = getelementptr inbounds [100 x i32]* getelementptr inbounds (%struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
-  %7 = load i32* %arrayidx4, align 4
+  %6 = load i32, i32* %a.addr, align 4
+  %arrayidx4 = getelementptr inbounds [100 x i32], [100 x i32]* getelementptr inbounds (%struct.anon, %struct.anon* @Foo, i32 0, i32 0), i32 0, i32 %6
+  %7 = load i32, i32* %arrayidx4, align 4
   ret i32 %7
 }
diff --git a/test/Transforms/LoopVectorize/hoist-loads.ll b/test/Transforms/LoopVectorize/hoist-loads.ll
index d0b27f1eae7c..a20b0f6cdf37 100644
--- a/test/Transforms/LoopVectorize/hoist-loads.ll
+++ b/test/Transforms/LoopVectorize/hoist-loads.ll
@@ -14,14 +14,14 @@ entry:
   br label %for.body
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %if.end9 ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @A, i64 0, i64 %indvars.iv
-  %arrayidx2 = getelementptr inbounds [1024 x float]* @B, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @A, i64 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds [1024 x float], [1024 x float]* @B, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx2, align 4
   %cmp3 = fcmp oeq float %0, 0.000000e+00
   br i1 %cmp3, label %if.end9, label %if.else
 
 if.else:
-  %1 = load float* %arrayidx, align 4
+  %1 = load float, float* %arrayidx, align 4
   br label %if.end9
 
 if.end9:
@@ -46,14 +46,14 @@ entry:
   br label %for.body
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %if.end9 ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @A, i64 0, i64 %indvars.iv
-  %arrayidx2 = getelementptr inbounds [1024 x float]* @B, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @A, i64 0, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds [1024 x float], [1024 x float]* @B, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx2, align 4
   %cmp3 = fcmp oeq float %0, 0.000000e+00
   br i1 %cmp3, label %if.end9, label %if.else
 
 if.else:
-  %1 = load float* %arrayidx, align 4
+  %1 = load float, float* %arrayidx, align 4
   br label %if.end9
 
 if.end9:
diff --git a/test/Transforms/LoopVectorize/i8-induction.ll b/test/Transforms/LoopVectorize/i8-induction.ll
index 90e3ec00cde2..d9e8a430be18 100644
--- a/test/Transforms/LoopVectorize/i8-induction.ll
+++ b/test/Transforms/LoopVectorize/i8-induction.ll
@@ -9,7 +9,7 @@ target triple = "x86_64-apple-macosx10.8.0"
 define void @f() nounwind uwtable ssp {
 scalar.ph:
   store i8 0, i8* inttoptr (i64 1 to i8*), align 1
-  %0 = load i8* @a, align 1
+  %0 = load i8, i8* @a, align 1
   br label %for.body
 
 for.body:
diff --git a/test/Transforms/LoopVectorize/if-conversion-edgemasks.ll b/test/Transforms/LoopVectorize/if-conversion-edgemasks.ll
index 27c274d557eb..8d435f513537 100644
--- a/test/Transforms/LoopVectorize/if-conversion-edgemasks.ll
+++ b/test/Transforms/LoopVectorize/if-conversion-edgemasks.ll
@@ -20,12 +20,15 @@ entry:
   br i1 %cmp88, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load i32** @b, align 8  %1 = load i32** @a, align 8  %2 = load i32** @c, align 8  br label %for.body
+  %0 = load i32*, i32** @b, align 8
+  %1 = load i32*, i32** @a, align 8
+  %2 = load i32*, i32** @c, align 8
+  br label %for.body
 
 for.body:
   %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %_ZL3fn3ii.exit58 ]
-  %arrayidx = getelementptr inbounds i32* %0, i64 %indvars.iv
-  %3 = load i32* %arrayidx, align 4  %4 = trunc i64 %indvars.iv to i32
+  %arrayidx = getelementptr inbounds i32, i32* %0, i64 %indvars.iv
+  %3 = load i32, i32* %arrayidx, align 4  %4 = trunc i64 %indvars.iv to i32
   %and.i = and i32 %4, 1
   %tobool.i.i = icmp eq i32 %and.i, 0
   br i1 %tobool.i.i, label %if.end.i, label %if.then.i
@@ -131,9 +134,9 @@ if.then.i15.i:
 
 _ZL3fn3ii.exit:
   %p1.addr.0.i16.i = phi i32 [ %or.i14.i, %if.then.i15.i ], [ %p1.addr.3.i.i, %_Z3fn2iii.exit.i ]
-  %arrayidx2 = getelementptr inbounds i32* %1, i64 %indvars.iv
-  store i32 %p1.addr.0.i16.i, i32* %arrayidx2, align 4  %arrayidx4 = getelementptr inbounds i32* %0, i64 %indvars.iv
-  %10 = load i32* %arrayidx4, align 4  br i1 %tobool.i.i, label %_Z3fn1ii.exit.i26, label %if.then.i.i21
+  %arrayidx2 = getelementptr inbounds i32, i32* %1, i64 %indvars.iv
+  store i32 %p1.addr.0.i16.i, i32* %arrayidx2, align 4  %arrayidx4 = getelementptr inbounds i32, i32* %0, i64 %indvars.iv
+  %10 = load i32, i32* %arrayidx4, align 4  br i1 %tobool.i.i, label %_Z3fn1ii.exit.i26, label %if.then.i.i21
 
 if.then.i.i21:
   %and.i.i18 = lshr i32 %10, 2
@@ -229,7 +232,7 @@ if.then.i15.i56:
 
 _ZL3fn3ii.exit58:
   %p1.addr.0.i16.i57 = phi i32 [ %or.i14.i55, %if.then.i15.i56 ], [ %p1.addr.3.i.i50, %_Z3fn2iii.exit.i52 ]
-  %arrayidx7 = getelementptr inbounds i32* %2, i64 %indvars.iv
+  %arrayidx7 = getelementptr inbounds i32, i32* %2, i64 %indvars.iv
   store i32 %p1.addr.0.i16.i57, i32* %arrayidx7, align 4  %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp ne i32 %lftr.wideiv, %p1
diff --git a/test/Transforms/LoopVectorize/if-conversion-nest.ll b/test/Transforms/LoopVectorize/if-conversion-nest.ll
index b5ac8fca23b3..3a581ebf847e 100644
--- a/test/Transforms/LoopVectorize/if-conversion-nest.ll
+++ b/test/Transforms/LoopVectorize/if-conversion-nest.ll
@@ -19,10 +19,10 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ %indvars.iv.next, %if.end14 ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %cmp3 = icmp sgt i32 %0, %1
   br i1 %cmp3, label %if.then, label %if.end14
 
diff --git a/test/Transforms/LoopVectorize/if-conversion-reduction.ll b/test/Transforms/LoopVectorize/if-conversion-reduction.ll
index 455699c78030..20333b94345f 100644
--- a/test/Transforms/LoopVectorize/if-conversion-reduction.ll
+++ b/test/Transforms/LoopVectorize/if-conversion-reduction.ll
@@ -14,8 +14,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
   %sum.011 = phi i32 [ %sum.1, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 30
   br i1 %cmp1, label %if.then, label %for.inc
 
diff --git a/test/Transforms/LoopVectorize/if-conversion.ll b/test/Transforms/LoopVectorize/if-conversion.ll
index 9e185284919a..fb5416976525 100644
--- a/test/Transforms/LoopVectorize/if-conversion.ll
+++ b/test/Transforms/LoopVectorize/if-conversion.ll
@@ -19,9 +19,9 @@ target triple = "x86_64-apple-macosx10.9.0"
 
 ;CHECK-LABEL: @function0(
 ;CHECK: load <4 x i32>
-;CHECK: icmp sgt <4 x i32>
 ;CHECK: mul <4 x i32>
 ;CHECK: add <4 x i32>
+;CHECK: icmp sle <4 x i32>
 ;CHECK: select <4 x i1>
 ;CHECK: ret i32
 define i32 @function0(i32* nocapture %a, i32* nocapture %b, i32 %start, i32 %end) nounwind uwtable ssp {
@@ -35,10 +35,10 @@ for.body.lr.ph:
 
 for.body:
   %indvars.iv = phi i64 [ %0, %for.body.lr.ph ], [ %indvars.iv.next, %if.end ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx, align 4
-  %arrayidx4 = getelementptr inbounds i32* %b, i64 %indvars.iv
-  %2 = load i32* %arrayidx4, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %b, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx4, align 4
   %cmp5 = icmp sgt i32 %1, %2
   br i1 %cmp5, label %if.then, label %if.end
 
@@ -72,8 +72,8 @@ for.end:
 
 ;CHECK-LABEL: @reduction_func(
 ;CHECK: load <4 x i32>
-;CHECK: icmp sgt <4 x i32>
 ;CHECK: add <4 x i32>
+;CHECK: icmp sle <4 x i32>
 ;CHECK: select <4 x i1>
 ;CHECK: ret i32
 define i32 @reduction_func(i32* nocapture %A, i32 %n) nounwind uwtable readonly ssp {
@@ -84,8 +84,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %entry ]
   %sum.011 = phi i32 [ %sum.1, %for.inc ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 30
   br i1 %cmp1, label %if.then, label %for.inc
 
@@ -123,13 +123,13 @@ entry:
 for.body:
   %inc3 = phi i32 [ 0, %entry ], [ %inc, %cond.end ]
   %or2 = phi i32 [ 0, %entry ], [ %or, %cond.end ]
-  br i1 icmp eq (i32** getelementptr inbounds ([1 x i32*]* @a, i64 0, i64 0), i32** @c), label %cond.false, label %cond.end
+  br i1 icmp eq (i32** getelementptr inbounds ([1 x i32*], [1 x i32*]* @a, i64 0, i64 0), i32** @c), label %cond.false, label %cond.end
 
 cond.false:
   br label %cond.end
 
 cond.end:
-  %cond = phi i32 [ sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*]* @a, i64 0, i64 0), i32** @c) to i32)), %cond.false ], [ 0, %for.body ]
+  %cond = phi i32 [ sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*], [1 x i32*]* @a, i64 0, i64 0), i32** @c) to i32)), %cond.false ], [ 0, %for.body ]
   %or = or i32 %or2, %cond
   %inc = add nsw i32 %inc3, 1
   %cmp = icmp slt i32 %inc, 128
@@ -153,10 +153,10 @@ entry:
 for.body:
   %inc3 = phi i32 [ 0, %entry ], [ %inc, %cond.end ]
   %or2 = phi i32 [ 0, %entry ], [ %or, %cond.end ]
-  br i1 icmp eq (i32** getelementptr inbounds ([1 x i32*]* @a, i64 0, i64 0), i32** @c), label %cond.false, label %cond.end
+  br i1 icmp eq (i32** getelementptr inbounds ([1 x i32*], [1 x i32*]* @a, i64 0, i64 0), i32** @c), label %cond.false, label %cond.end
 
 cond.false:
-  %cond.1 = or i32 %inc3, sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*]* @a, i64 0, i64 1), i32** @c) to i32))
+  %cond.1 = or i32 %inc3, sdiv (i32 1, i32 zext (i1 icmp eq (i32** getelementptr inbounds ([1 x i32*], [1 x i32*]* @a, i64 0, i64 1), i32** @c) to i32))
   br label %cond.end
 
 cond.end:
diff --git a/test/Transforms/LoopVectorize/if-pred-stores.ll b/test/Transforms/LoopVectorize/if-pred-stores.ll
index c6067e01ee21..991d027ada5c 100644
--- a/test/Transforms/LoopVectorize/if-pred-stores.ll
+++ b/test/Transforms/LoopVectorize/if-pred-stores.ll
@@ -39,10 +39,10 @@ entry:
 ; UNROLL: vector.body:
 ; UNROLL:   %[[IND:[a-zA-Z0-9]+]] = add i64 %{{.*}}, 0
 ; UNROLL:   %[[IND1:[a-zA-Z0-9]+]] = add i64 %{{.*}}, 1
-; UNROLL:   %[[v0:[a-zA-Z0-9]+]] = getelementptr inbounds i32* %f, i64 %[[IND]]
-; UNROLL:   %[[v1:[a-zA-Z0-9]+]] = getelementptr inbounds i32* %f, i64 %[[IND1]]
-; UNROLL:   %[[v2:[a-zA-Z0-9]+]] = load i32* %[[v0]], align 4
-; UNROLL:   %[[v3:[a-zA-Z0-9]+]] = load i32* %[[v1]], align 4
+; UNROLL:   %[[v0:[a-zA-Z0-9]+]] = getelementptr inbounds i32, i32* %f, i64 %[[IND]]
+; UNROLL:   %[[v1:[a-zA-Z0-9]+]] = getelementptr inbounds i32, i32* %f, i64 %[[IND1]]
+; UNROLL:   %[[v2:[a-zA-Z0-9]+]] = load i32, i32* %[[v0]], align 4
+; UNROLL:   %[[v3:[a-zA-Z0-9]+]] = load i32, i32* %[[v1]], align 4
 ; UNROLL:   %[[v4:[a-zA-Z0-9]+]] = icmp sgt i32 %[[v2]], 100
 ; UNROLL:   %[[v5:[a-zA-Z0-9]+]] = icmp sgt i32 %[[v3]], 100
 ; UNROLL:   %[[v6:[a-zA-Z0-9]+]] = add nsw i32 %[[v2]], 20
@@ -66,8 +66,8 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
-  %arrayidx = getelementptr inbounds i32* %f, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %f, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp1 = icmp sgt i32 %0, 100
   br i1 %cmp1, label %if.then, label %for.inc
 
@@ -104,8 +104,8 @@ for.body9:
 for.body14:
   %indvars.iv3 = phi i64 [ %indvars.iv.next4, %for.inc23 ], [ undef, %for.body9 ]
   %iNewChunks.120 = phi i32 [ %iNewChunks.2, %for.inc23 ], [ undef, %for.body9 ]
-  %arrayidx16 = getelementptr inbounds [768 x i32]* undef, i64 0, i64 %indvars.iv3
-  %tmp = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds [768 x i32], [768 x i32]* undef, i64 0, i64 %indvars.iv3
+  %tmp = load i32, i32* %arrayidx16, align 4
   br i1 undef, label %if.then18, label %for.inc23
 
 if.then18:
diff --git a/test/Transforms/LoopVectorize/incorrect-dom-info.ll b/test/Transforms/LoopVectorize/incorrect-dom-info.ll
index b8624fd00774..798793a641bc 100644
--- a/test/Transforms/LoopVectorize/incorrect-dom-info.ll
+++ b/test/Transforms/LoopVectorize/incorrect-dom-info.ll
@@ -45,7 +45,7 @@ thread-pre-split.preheader:                       ; preds = %9
 
 .thread-pre-split.loopexit_crit_edge:             ; preds = %19
   %scevgep.sum = xor i64 %umax, -1
-  %scevgep45 = getelementptr i8* %d.020, i64 %scevgep.sum
+  %scevgep45 = getelementptr i8, i8* %d.020, i64 %scevgep.sum
   br label %thread-pre-split.loopexit
 
 thread-pre-split.loopexit:                        ; preds = %11, %.thread-pre-split.loopexit_crit_edge
@@ -58,8 +58,8 @@ thread-pre-split.loopexit:                        ; preds = %11, %.thread-pre-sp
   br i1 undef, label %11, label %22
 
 ; <label>:11                                      ; preds = %.lr.ph21
-  %12 = getelementptr inbounds [0 x i8]* @PL_utf8skip, i64 0, i64 undef
-  %13 = load i8* %12, align 1
+  %12 = getelementptr inbounds [0 x i8], [0 x i8]* @PL_utf8skip, i64 0, i64 undef
+  %13 = load i8, i8* %12, align 1
   %14 = zext i8 %13 to i64
   %15 = icmp ugt i64 %14, %10
   %. = select i1 %15, i64 %10, i64 %14
@@ -91,7 +91,7 @@ thread-pre-split.loopexit:                        ; preds = %11, %.thread-pre-sp
   br label %26
 
 ; <label>:26                                      ; preds = %25, %24, %23, %22
-  %27 = load i64* %len, align 8
+  %27 = load i64, i64* %len, align 8
   %28 = add i64 %27, -1
   br i1 undef, label %thread-pre-split._crit_edge, label %.lr.ph21
 
diff --git a/test/Transforms/LoopVectorize/increment.ll b/test/Transforms/LoopVectorize/increment.ll
index 067a76b51ad0..d0b25096e37a 100644
--- a/test/Transforms/LoopVectorize/increment.ll
+++ b/test/Transforms/LoopVectorize/increment.ll
@@ -20,8 +20,8 @@ define void @inc(i32 %n) nounwind uwtable noinline ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = trunc i64 %indvars.iv to i32
   %5 = add nsw i32 %3, %4
   store i32 %5, i32* %2, align 4
@@ -49,11 +49,11 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %idxprom1 = sext i32 %0 to i64
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %idxprom1
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %idxprom1
+  %1 = load i32, i32* %arrayidx2, align 4
   %inc = add nsw i32 %1, 1
   store i32 %inc, i32* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/induction.ll b/test/Transforms/LoopVectorize/induction.ll
index 3f3491835b54..2fbb2de797ae 100644
--- a/test/Transforms/LoopVectorize/induction.ll
+++ b/test/Transforms/LoopVectorize/induction.ll
@@ -16,7 +16,7 @@ for.body.lr.ph:
 for.body:
   %indvars.iv = phi i64 [ 0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
   %count.09 = phi i32 [ 190, %for.body.lr.ph ], [ %inc, %for.body ]
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
   store i32 %count.09, i32* %arrayidx2, align 4
   %inc = add nsw i32 %count.09, 1
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -51,11 +51,11 @@ entry:
 for.body:
   %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
   %ind.sum = add i64 %iv, %offset
-  %arr.idx = getelementptr inbounds float* %a, i64 %ind.sum
-  %l1 = load float* %arr.idx, align 4
+  %arr.idx = getelementptr inbounds float, float* %a, i64 %ind.sum
+  %l1 = load float, float* %arr.idx, align 4
   %ind.sum2 = add i64 %iv, %offset2
-  %arr.idx2 = getelementptr inbounds float* %a, i64 %ind.sum2
-  %l2 = load float* %arr.idx2, align 4
+  %arr.idx2 = getelementptr inbounds float, float* %a, i64 %ind.sum2
+  %l2 = load float, float* %arr.idx2, align 4
   %m = fmul fast float %b, %l2
   %ad = fadd fast float %l1, %m
   store float %ad, float* %arr.idx, align 4
@@ -153,9 +153,9 @@ define i32 @max_i32_backedgetaken() nounwind readnone ssp uwtable {
 @c = common global i32 0, align 4
 define i32 @testoverflowcheck() {
 entry:
-  %.pr.i = load i8* @e, align 1
-  %0 = load i32* @d, align 4
-  %c.promoted.i = load i32* @c, align 4
+  %.pr.i = load i8, i8* @e, align 1
+  %0 = load i32, i32* @d, align 4
+  %c.promoted.i = load i32, i32* @c, align 4
   br label %cond.end.i
 
 cond.end.i:
diff --git a/test/Transforms/LoopVectorize/induction_plus.ll b/test/Transforms/LoopVectorize/induction_plus.ll
index ce64c5b620d5..7c4c8f2edcbf 100644
--- a/test/Transforms/LoopVectorize/induction_plus.ll
+++ b/test/Transforms/LoopVectorize/induction_plus.ll
@@ -16,7 +16,7 @@ define i32 @array_at_plus_one(i32 %n) nounwind uwtable ssp {
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %2 = add nsw i64 %indvars.iv, 12
-  %3 = getelementptr inbounds [1024 x i32]* @array, i64 0, i64 %2
+  %3 = getelementptr inbounds [1024 x i32], [1024 x i32]* @array, i64 0, i64 %2
   %4 = trunc i64 %indvars.iv to i32
   store i32 %4, i32* %3, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/intrinsic.ll b/test/Transforms/LoopVectorize/intrinsic.ll
index d48731a07436..fae6a8cb714b 100644
--- a/test/Transforms/LoopVectorize/intrinsic.ll
+++ b/test/Transforms/LoopVectorize/intrinsic.ll
@@ -13,10 +13,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.sqrt.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -39,10 +39,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.sqrt.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -65,10 +65,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.sin.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -91,10 +91,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.sin.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -117,10 +117,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.cos.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -143,10 +143,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.cos.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -169,10 +169,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.exp.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -195,10 +195,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.exp.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -221,10 +221,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.exp2.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -247,10 +247,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.exp2.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -273,10 +273,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.log.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -299,10 +299,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.log.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -325,10 +325,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.log10.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -351,10 +351,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.log10.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -377,10 +377,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.log2.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -403,10 +403,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.log2.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -429,10 +429,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.fabs.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -452,10 +452,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.fabs(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -478,12 +478,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %1 = load float* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %1 = load float, float* %arrayidx1, align 4
   %call = tail call float @llvm.copysign.f32(float %0, float %1) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -503,12 +503,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx1 = getelementptr inbounds double* %z, i64 %indvars.iv
-  %1 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx1 = getelementptr inbounds double, double* %z, i64 %indvars.iv
+  %1 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.copysign(double %0, double %1) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -531,10 +531,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.floor.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -557,10 +557,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.floor.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -583,10 +583,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.ceil.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -609,10 +609,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.ceil.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -635,10 +635,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.trunc.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -661,10 +661,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.trunc.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -687,10 +687,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.rint.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -713,10 +713,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.rint.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -739,10 +739,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.nearbyint.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -765,10 +765,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.nearbyint.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -791,10 +791,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @llvm.round.f32(float %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -817,10 +817,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.round.f64(double %0) nounwind readnone
-  %arrayidx2 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx2, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -843,14 +843,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %w, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
-  %arrayidx4 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %w, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %2 = load float, float* %arrayidx4, align 4
   %3 = tail call float @llvm.fma.f32(float %0, float %2, float %1)
-  %arrayidx6 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx6 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %3, float* %arrayidx6, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -873,14 +873,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx2 = getelementptr inbounds double* %w, i64 %indvars.iv
-  %1 = load double* %arrayidx2, align 8
-  %arrayidx4 = getelementptr inbounds double* %z, i64 %indvars.iv
-  %2 = load double* %arrayidx4, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %w, i64 %indvars.iv
+  %1 = load double, double* %arrayidx2, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %z, i64 %indvars.iv
+  %2 = load double, double* %arrayidx4, align 8
   %3 = tail call double @llvm.fma.f64(double %0, double %2, double %1)
-  %arrayidx6 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx6 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %3, double* %arrayidx6, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -903,14 +903,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %w, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
-  %arrayidx4 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %w, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %2 = load float, float* %arrayidx4, align 4
   %3 = tail call float @llvm.fmuladd.f32(float %0, float %2, float %1)
-  %arrayidx6 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx6 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %3, float* %arrayidx6, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -933,14 +933,14 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx2 = getelementptr inbounds double* %w, i64 %indvars.iv
-  %1 = load double* %arrayidx2, align 8
-  %arrayidx4 = getelementptr inbounds double* %z, i64 %indvars.iv
-  %2 = load double* %arrayidx4, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %w, i64 %indvars.iv
+  %1 = load double, double* %arrayidx2, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %z, i64 %indvars.iv
+  %2 = load double, double* %arrayidx4, align 8
   %3 = tail call double @llvm.fmuladd.f64(double %0, double %2, double %1)
-  %arrayidx6 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx6 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %3, double* %arrayidx6, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -963,12 +963,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %call = tail call float @llvm.pow.f32(float %0, float %1) nounwind readnone
-  %arrayidx4 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx4, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -991,12 +991,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
-  %arrayidx2 = getelementptr inbounds double* %z, i64 %indvars.iv
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %z, i64 %indvars.iv
+  %1 = load double, double* %arrayidx2, align 8
   %call = tail call double @llvm.pow.f64(double %0, double %1) nounwind readnone
-  %arrayidx4 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx4, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1016,8 +1016,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %x, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @fabsf(float %0) nounwind readnone
   store float %call, float* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -1050,8 +1050,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %x, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %x, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %call = tail call float @roundf(float %0) nounwind readnone
   store float %call, float* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -1077,8 +1077,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds double* %x, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds double, double* %x, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 4
   store double %0, double* %arrayidx, align 4
   tail call void @round(double %0) nounwind readnone
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -1102,10 +1102,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %call = tail call double @llvm.powi.f64(double %0, i32  %P) nounwind readnone
-  %arrayidx4 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx4, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1126,11 +1126,11 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds double* %y, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %y, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 8
   %1 = trunc i64 %indvars.iv to i32
   %call = tail call double @llvm.powi.f64(double %0, i32  %1) nounwind readnone
-  %arrayidx4 = getelementptr inbounds double* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds double, double* %x, i64 %indvars.iv
   store double %call, double* %arrayidx4, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1153,10 +1153,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i64* %y, i64 %indvars.iv
-  %0 = load i64* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds i64, i64* %y, i64 %indvars.iv
+  %0 = load i64, i64* %arrayidx, align 8
   %call = tail call i64 @llvm.cttz.i64(i64 %0, i1 true) nounwind readnone
-  %arrayidx4 = getelementptr inbounds i64* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i64, i64* %x, i64 %indvars.iv
   store i64 %call, i64* %arrayidx4, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1179,10 +1179,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i64* %y, i64 %indvars.iv
-  %0 = load i64* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds i64, i64* %y, i64 %indvars.iv
+  %0 = load i64, i64* %arrayidx, align 8
   %call = tail call i64 @llvm.ctlz.i64(i64 %0, i1 true) nounwind readnone
-  %arrayidx4 = getelementptr inbounds i64* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i64, i64* %x, i64 %indvars.iv
   store i64 %call, i64* %arrayidx4, align 8
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1205,12 +1205,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %call = tail call float @llvm.minnum.f32(float %0, float %1) nounwind readnone
-  %arrayidx4 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx4, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -1233,12 +1233,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %y, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %z, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds float, float* %y, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %z, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %call = tail call float @llvm.maxnum.f32(float %0, float %1) nounwind readnone
-  %arrayidx4 = getelementptr inbounds float* %x, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds float, float* %x, i64 %indvars.iv
   store float %call, float* %arrayidx4, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/lifetime.ll b/test/Transforms/LoopVectorize/lifetime.ll
index ba36cc4f7aa4..6e525ca1d822 100644
--- a/test/Transforms/LoopVectorize/lifetime.ll
+++ b/test/Transforms/LoopVectorize/lifetime.ll
@@ -19,8 +19,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   call void @llvm.lifetime.end(i64 4096, i8* %0) #1
-  %arrayidx = getelementptr inbounds i32* %d, i64 %indvars.iv
-  %1 = load i32* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %d, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx, align 8
   store i32 100, i32* %arrayidx, align 8
   call void @llvm.lifetime.start(i64 4096, i8* %0) #1
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -49,8 +49,8 @@ for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %1 = bitcast [1024 x i32]* %arr to i8*
   call void @llvm.lifetime.end(i64 4096, i8* %1) #1
-  %arrayidx = getelementptr inbounds i32* %d, i64 %indvars.iv
-  %2 = load i32* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %d, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx, align 8
   store i32 100, i32* %arrayidx, align 8
   call void @llvm.lifetime.start(i64 4096, i8* %1) #1
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -75,11 +75,11 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %0 = getelementptr [1024 x i32]* %arr, i32 0, i64 %indvars.iv
+  %0 = getelementptr [1024 x i32], [1024 x i32]* %arr, i32 0, i64 %indvars.iv
   %1 = bitcast [1024 x i32]* %arr to i8*
   call void @llvm.lifetime.end(i64 4096, i8* %1) #1
-  %arrayidx = getelementptr inbounds i32* %d, i64 %indvars.iv
-  %2 = load i32* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds i32, i32* %d, i64 %indvars.iv
+  %2 = load i32, i32* %arrayidx, align 8
   store i32 100, i32* %arrayidx, align 8
   call void @llvm.lifetime.start(i64 4096, i8* %1) #1
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/loop-form.ll b/test/Transforms/LoopVectorize/loop-form.ll
index 138df1d96138..3bbe8100e34e 100644
--- a/test/Transforms/LoopVectorize/loop-form.ll
+++ b/test/Transforms/LoopVectorize/loop-form.ll
@@ -21,7 +21,7 @@ for.cond:
 
 for.body:
   %iprom = sext i32 %i to i64
-  %b = getelementptr inbounds %struct.X* undef, i64 %iprom, i32 1
+  %b = getelementptr inbounds %struct.X, %struct.X* undef, i64 %iprom, i32 1
   store i16 0, i16* %b, align 4
   %inc = add nsw i32 %i, 1
   br label %for.cond
diff --git a/test/Transforms/LoopVectorize/loop-vect-memdep.ll b/test/Transforms/LoopVectorize/loop-vect-memdep.ll
index e2c7524238d1..d9efaa5c085f 100644
--- a/test/Transforms/LoopVectorize/loop-vect-memdep.ll
+++ b/test/Transforms/LoopVectorize/loop-vect-memdep.ll
@@ -6,16 +6,16 @@ target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 
 define void @test_loop_novect(double** %arr, i64 %n) {
 for.body.lr.ph:
-  %t = load double** %arr, align 8
+  %t = load double*, double** %arr, align 8
   br label %for.body
 
 for.body:                                      ; preds = %for.body, %for.body.lr.ph
   %i = phi i64 [ 0, %for.body.lr.ph ], [ %i.next, %for.body ]
-  %a = getelementptr inbounds double* %t, i64 %i
+  %a = getelementptr inbounds double, double* %t, i64 %i
   %i.next = add nuw nsw i64 %i, 1
-  %a.next = getelementptr inbounds double* %t, i64 %i.next
-  %t1 = load double* %a, align 8
-  %t2 = load double* %a.next, align 8
+  %a.next = getelementptr inbounds double, double* %t, i64 %i.next
+  %t1 = load double, double* %a, align 8
+  %t2 = load double, double* %a.next, align 8
   store double %t1, double* %a.next, align 8
   store double %t2, double* %a, align 8
   %c = icmp eq i64 %i, %n
diff --git a/test/Transforms/LoopVectorize/memdep.ll b/test/Transforms/LoopVectorize/memdep.ll
index f857e800bea7..fb608836dfef 100644
--- a/test/Transforms/LoopVectorize/memdep.ll
+++ b/test/Transforms/LoopVectorize/memdep.ll
@@ -19,10 +19,10 @@ entry:
 for.body:
   %indvars.iv = phi i32 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %indvars.iv.next = add i32 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds i32* %A, i32 %indvars.iv.next
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i32 %indvars.iv.next
+  %0 = load i32, i32* %arrayidx, align 4
   %add1 = add nsw i32 %0, 1
-  %arrayidx3 = getelementptr inbounds i32* %A, i32 %indvars.iv
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i32 %indvars.iv
   store i32 %add1, i32* %arrayidx3, align 4
   %exitcond = icmp ne i32 %indvars.iv.next, 1024
   br i1 %exitcond, label %for.body, label %for.end
@@ -44,11 +44,11 @@ entry:
 
 for.body:
   %indvars.iv = phi i32 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %A, i32 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i32 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, 1
   %indvars.iv.next = add i32 %indvars.iv, 1
-  %arrayidx3 = getelementptr inbounds i32* %A, i32 %indvars.iv.next
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i32 %indvars.iv.next
   store i32 %add, i32* %arrayidx3, align 4
   %exitcond = icmp ne i32 %indvars.iv.next, 1024
   br i1 %exitcond, label %for.body, label %for.end
@@ -74,12 +74,12 @@ entry:
 for.body:
   %i.01 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %idxprom = sext i32 %i.01 to i64
-  %arrayidx = getelementptr inbounds i32* %A, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, 1
   %add1 = add nsw i32 %i.01, 2
   %idxprom2 = sext i32 %add1 to i64
-  %arrayidx3 = getelementptr inbounds i32* %A, i64 %idxprom2
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i64 %idxprom2
   store i32 %add, i32* %arrayidx3, align 4
   %inc = add nsw i32 %i.01, 1
   %cmp = icmp slt i32 %inc, 1024
@@ -105,13 +105,13 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %B, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
   store i32 %0, i32* %arrayidx2, align 4
   %indvars.iv.next = add nsw i64 %indvars.iv, 1
-  %arrayidx4 = getelementptr inbounds i32* %B, i64 %indvars.iv.next
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv.next
+  %1 = load i32, i32* %arrayidx4, align 4
   store i32 %1, i32* %arrayidx, align 4
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp ne i32 %lftr.wideiv, 1024
@@ -138,10 +138,10 @@ for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %tmp.addr.08 = phi i32 [ %tmp, %entry ], [ %0, %for.body ]
   %indvars.iv.next = add nsw i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv.next
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv.next
   store i32 %tmp.addr.08, i32* %arrayidx, align 4
-  %arrayidx3 = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx3, align 4
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
   %exitcond = icmp ne i32 %lftr.wideiv, 1024
   br i1 %exitcond, label %for.body, label %for.end
@@ -169,13 +169,13 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 16, %entry ], [ %indvars.iv.next, %for.body ]
   %0 = add nsw i64 %indvars.iv, -3
-  %arrayidx = getelementptr inbounds i32* %A, i64 %0
-  %1 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %0
+  %1 = load i32, i32* %arrayidx, align 4
   %2 = add nsw i64 %indvars.iv, 4
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %2
-  %3 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %2
+  %3 = load i32, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %3, %1
-  %arrayidx5 = getelementptr inbounds i32* %A, i64 %indvars.iv
+  %arrayidx5 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
   store i32 %add3, i32* %arrayidx5, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -203,14 +203,14 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 16, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
   store i32 %0, i32* %arrayidx2, align 4
   %1 = add nsw i64 %indvars.iv, -3
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 %1
-  %2 = load i32* %arrayidx4, align 4
-  %arrayidx6 = getelementptr inbounds i32* %C, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 %1
+  %2 = load i32, i32* %arrayidx4, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %C, i64 %indvars.iv
   store i32 %2, i32* %arrayidx6, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/metadata-unroll.ll b/test/Transforms/LoopVectorize/metadata-unroll.ll
index 36a231454f08..3c80ae0cb54f 100644
--- a/test/Transforms/LoopVectorize/metadata-unroll.ll
+++ b/test/Transforms/LoopVectorize/metadata-unroll.ll
@@ -23,8 +23,8 @@ define void @inc(i32 %n) nounwind uwtable noinline ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = trunc i64 %indvars.iv to i32
   %5 = add nsw i32 %3, %4
   store i32 %5, i32* %2, align 4
diff --git a/test/Transforms/LoopVectorize/metadata-width.ll b/test/Transforms/LoopVectorize/metadata-width.ll
index dee4feedd6b8..69d902fad640 100644
--- a/test/Transforms/LoopVectorize/metadata-width.ll
+++ b/test/Transforms/LoopVectorize/metadata-width.ll
@@ -13,7 +13,7 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   %0 = trunc i64 %indvars.iv to i32
   store i32 %0, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/metadata.ll b/test/Transforms/LoopVectorize/metadata.ll
index a258f7c06c07..9a791aedbbe4 100644
--- a/test/Transforms/LoopVectorize/metadata.ll
+++ b/test/Transforms/LoopVectorize/metadata.ll
@@ -9,10 +9,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !tbaa !0
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !tbaa !0
   %conv = fptosi float %0 to i32
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx2, align 4, !tbaa !4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 1600
@@ -23,7 +23,7 @@ for.end:                                          ; preds = %for.body
 }
 
 ; CHECK-LABEL: @test1
-; CHECK: load <4 x float>* %{{.*}}, align 4, !tbaa ![[TFLT:[0-9]+]]
+; CHECK: load <4 x float>, <4 x float>* %{{.*}}, align 4, !tbaa ![[TFLT:[0-9]+]]
 ; CHECK: store <4 x i32> %{{.*}}, <4 x i32>* %{{.*}}, align 4, !tbaa ![[TINT:[0-9]+]]
 ; CHECK: ret i32 0
 
diff --git a/test/Transforms/LoopVectorize/minmax_reduction.ll b/test/Transforms/LoopVectorize/minmax_reduction.ll
index 1984cdd892d1..5a0356fe11a2 100644
--- a/test/Transforms/LoopVectorize/minmax_reduction.ll
+++ b/test/Transforms/LoopVectorize/minmax_reduction.ll
@@ -26,8 +26,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp sgt i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -55,8 +55,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp slt i32 %max.red.08, %0
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -83,8 +83,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp slt i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -112,8 +112,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp sgt i32 %max.red.08, %0
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -142,8 +142,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp ugt i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -171,8 +171,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp ult i32 %max.red.08, %0
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -199,8 +199,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp ult i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -228,8 +228,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp ugt i32 %max.red.08, %0
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -257,8 +257,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp sge i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %max.red.08, i32 %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -286,8 +286,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp sle i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %max.red.08, i32 %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -315,8 +315,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp uge i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %max.red.08, i32 %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -344,8 +344,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %cmp3 = icmp ule i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %max.red.08, i32 %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -367,10 +367,10 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %arrayidx1 = getelementptr inbounds [1024 x i32]* @A, i64 1, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %arrayidx1 = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 1, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %1 = load i32, i32* %arrayidx1, align 4
   %cmp3 = icmp sgt i32 %0, %1
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -391,10 +391,10 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi i32 [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x i32]* @A, i64 0, i64 %indvars.iv
-  %arrayidx1 = getelementptr inbounds [1024 x i32]* @A, i64 1, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 0, i64 %indvars.iv
+  %arrayidx1 = getelementptr inbounds [1024 x i32], [1024 x i32]* @A, i64 1, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %1 = load i32, i32* %arrayidx1, align 4
   %cmp3 = icmp sgt i32 %0, %max.red.08
   %max.red.0 = select i1 %cmp3, i32 %0, i32 %1
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -425,8 +425,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ogt float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %0, float %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -451,8 +451,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp oge float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %0, float %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -477,8 +477,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp olt float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %max.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -503,8 +503,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ole float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %max.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -529,8 +529,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ugt float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %0, float %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -555,8 +555,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp uge float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %0, float %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -581,8 +581,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ult float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %max.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -607,8 +607,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ule float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %max.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -636,8 +636,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp olt float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %0, float %min.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -662,8 +662,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ole float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %0, float %min.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -688,8 +688,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ogt float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %min.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -714,8 +714,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp oge float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %min.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -740,8 +740,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ult float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %0, float %min.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -766,8 +766,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ule float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %0, float %min.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -792,8 +792,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ugt float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %min.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -818,8 +818,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi float [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp uge float %0, %min.red.08
   %min.red.0 = select i1 %cmp3, float %min.red.08, float %0
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -845,8 +845,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %min.red.08 = phi double [ %min, %entry ], [ %min.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x double]* @dA, i64 0, i64 %indvars.iv
-  %0 = load double* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x double], [1024 x double]* @dA, i64 0, i64 %indvars.iv
+  %0 = load double, double* %arrayidx, align 4
   %cmp3 = fcmp olt double %0, %min.red.08
   %min.red.0 = select i1 %cmp3, double %0, double %min.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -869,8 +869,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %max.red.08 = phi float [ %max, %entry ], [ %max.red.0, %for.body ]
-  %arrayidx = getelementptr inbounds [1024 x float]* @fA, i64 0, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x float], [1024 x float]* @fA, i64 0, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %cmp3 = fcmp ogt float %0, %max.red.08
   %max.red.0 = select i1 %cmp3, float %0, float %max.red.08
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/multi-use-reduction-bug.ll b/test/Transforms/LoopVectorize/multi-use-reduction-bug.ll
index cd022adc9399..9f7fb39dc7ca 100644
--- a/test/Transforms/LoopVectorize/multi-use-reduction-bug.ll
+++ b/test/Transforms/LoopVectorize/multi-use-reduction-bug.ll
@@ -16,10 +16,10 @@ entry:
   %n = alloca i32, align 4
   %k7 = alloca i32, align 4
   %nf = alloca i32, align 4
-  %0 = load i32* %k7, align 4
+  %0 = load i32, i32* %k7, align 4
   %.neg1 = sub i32 0, %0
-  %n.promoted = load i32* %n, align 4
-  %nf.promoted = load i32* %nf, align 4
+  %n.promoted = load i32, i32* %n, align 4
+  %nf.promoted = load i32, i32* %nf, align 4
   br label %for.body
 
 for.body:
diff --git a/test/Transforms/LoopVectorize/multiple-address-spaces.ll b/test/Transforms/LoopVectorize/multiple-address-spaces.ll
index bb2af1e57ccc..e79c93146b19 100644
--- a/test/Transforms/LoopVectorize/multiple-address-spaces.ll
+++ b/test/Transforms/LoopVectorize/multiple-address-spaces.ll
@@ -27,10 +27,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds [40000 x i8] addrspace(1)* @Y, i64 0, i64 %indvars.iv
-  %0 = load i8 addrspace(1)* %arrayidx, align 1
+  %arrayidx = getelementptr inbounds [40000 x i8], [40000 x i8] addrspace(1)* @Y, i64 0, i64 %indvars.iv
+  %0 = load i8, i8 addrspace(1)* %arrayidx, align 1
   %add = add i8 %0, 1
-  %arrayidx3 = getelementptr inbounds [40000 x i8]* @X, i64 0, i64 %indvars.iv
+  %arrayidx3 = getelementptr inbounds [40000 x i8], [40000 x i8]* @X, i64 0, i64 %indvars.iv
   store i8 %add, i8* %arrayidx3, align 1
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/no_array_bounds.ll b/test/Transforms/LoopVectorize/no_array_bounds.ll
index d3bd755e6cf1..f7c7ff7732b9 100644
--- a/test/Transforms/LoopVectorize/no_array_bounds.ll
+++ b/test/Transforms/LoopVectorize/no_array_bounds.ll
@@ -33,11 +33,11 @@ for.body7.preheader:                              ; preds = %for.cond5.preheader
 
 for.body:                                         ; preds = %for.body.preheader, %for.body
   %indvars.iv27 = phi i64 [ %indvars.iv.next28, %for.body ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv27, !dbg !14
-  %0 = load i32* %arrayidx, align 4, !dbg !14, !tbaa !22
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %indvars.iv27, !dbg !14
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !14, !tbaa !22
   %idxprom1 = sext i32 %0 to i64, !dbg !14
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 %idxprom1, !dbg !14
-  %1 = load i32* %arrayidx2, align 4, !dbg !14, !tbaa !22
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 %idxprom1, !dbg !14
+  %1 = load i32, i32* %arrayidx2, align 4, !dbg !14, !tbaa !22
   %inc = add nsw i32 %1, 1, !dbg !14
   store i32 %inc, i32* %arrayidx2, align 4, !dbg !14, !tbaa !22
   %indvars.iv.next28 = add nuw nsw i64 %indvars.iv27, 1, !dbg !10
@@ -47,11 +47,11 @@ for.body:                                         ; preds = %for.body.preheader,
 
 for.body7:                                        ; preds = %for.body7.preheader, %for.body7
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body7 ], [ 0, %for.body7.preheader ]
-  %arrayidx9 = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !20
-  %2 = load i32* %arrayidx9, align 4, !dbg !20, !tbaa !22
+  %arrayidx9 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !20
+  %2 = load i32, i32* %arrayidx9, align 4, !dbg !20, !tbaa !22
   %idxprom10 = sext i32 %2 to i64, !dbg !20
-  %arrayidx11 = getelementptr inbounds i32* %B, i64 %idxprom10, !dbg !20
-  %3 = load i32* %arrayidx11, align 4, !dbg !20, !tbaa !22
+  %arrayidx11 = getelementptr inbounds i32, i32* %B, i64 %idxprom10, !dbg !20
+  %3 = load i32, i32* %arrayidx11, align 4, !dbg !20, !tbaa !22
   %inc12 = add nsw i32 %3, 1, !dbg !20
   store i32 %inc12, i32* %arrayidx11, align 4, !dbg !20, !tbaa !22
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1, !dbg !16
@@ -72,30 +72,30 @@ attributes #0 = { nounwind }
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\004\00clang version 3.5.0\001\00\000\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ]
-!1 = !{!"no_array_bounds.cpp", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0", isOptimized: true, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "no_array_bounds.cpp", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00test\00test\00\001\000\001\000\006\00256\001\002", !1, !5, !6, null, void (i32*, i32*, i32)* @_Z4testPiS_i, null, null, !2} ; [ DW_TAG_subprogram ]
-!5 = !{!"0x29", !1} ; [ DW_TAG_file_type ]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ]
+!4 = !DISubprogram(name: "test", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 2, file: !1, scope: !5, type: !6, function: void (i32*, i32*, i32)* @_Z4testPiS_i, variables: !2)
+!5 = !DIFile(filename: "no_array_bounds.cpp", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 2}
-!8 = !{i32 2, !"Debug Info Version", i32 2}
+!8 = !{i32 2, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5.0"}
-!10 = !MDLocation(line: 4, column: 8, scope: !11)
-!11 = !{!"0xb\004\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
+!10 = !DILocation(line: 4, column: 8, scope: !11)
+!11 = distinct !DILexicalBlock(line: 4, column: 3, file: !1, scope: !4)
 !12 = !{!12, !13}
 !13 = !{!"llvm.loop.vectorize.enable", i1 true}
-!14 = !MDLocation(line: 5, column: 5, scope: !15)
-!15 = !{!"0xb\004\0036\000", !1, !11} ; [ DW_TAG_lexical_block ]
-!16 = !MDLocation(line: 9, column: 8, scope: !17)
-!17 = !{!"0xb\009\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
+!14 = !DILocation(line: 5, column: 5, scope: !15)
+!15 = distinct !DILexicalBlock(line: 4, column: 36, file: !1, scope: !11)
+!16 = !DILocation(line: 9, column: 8, scope: !17)
+!17 = distinct !DILexicalBlock(line: 9, column: 3, file: !1, scope: !4)
 !18 = !{!18, !13, !19}
 !19 = !{!"llvm.loop.vectorize.width", i32 1}
-!20 = !MDLocation(line: 10, column: 5, scope: !21)
-!21 = !{!"0xb\009\0036\000", !1, !17} ; [ DW_TAG_lexical_block ]
+!20 = !DILocation(line: 10, column: 5, scope: !21)
+!21 = distinct !DILexicalBlock(line: 9, column: 36, file: !1, scope: !17)
 !22 = !{!23, !23, i64 0}
 !23 = !{!"int", !24, i64 0}
 !24 = !{!"omnipotent char", !25, i64 0}
 !25 = !{!"Simple C/C++ TBAA"}
-!26 = !MDLocation(line: 12, column: 1, scope: !4)
+!26 = !DILocation(line: 12, column: 1, scope: !4)
diff --git a/test/Transforms/LoopVectorize/no_idiv_reduction.ll b/test/Transforms/LoopVectorize/no_idiv_reduction.ll
index 5c721a6806d9..bfa48a2529b9 100644
--- a/test/Transforms/LoopVectorize/no_idiv_reduction.ll
+++ b/test/Transforms/LoopVectorize/no_idiv_reduction.ll
@@ -11,8 +11,8 @@ for.body:
   ; CHECK-NOT: sdiv <2 x i32>
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %r.05 = phi i32 [ 80, %entry ], [ %div, %for.body ]
-  %arrayidx = getelementptr inbounds [128 x i32]* @a, i64 0, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [128 x i32], [128 x i32]* @a, i64 0, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %div = sdiv i32 %r.05, %0
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/no_int_induction.ll b/test/Transforms/LoopVectorize/no_int_induction.ll
index 1275915f5e36..7e6b26cd9006 100644
--- a/test/Transforms/LoopVectorize/no_int_induction.ll
+++ b/test/Transforms/LoopVectorize/no_int_induction.ll
@@ -14,16 +14,16 @@ target datalayout = "e-p:64:64:64-p1:16:16:16-i1:8:8-i8:8:8-i16:16:16-i32:32:32-
 ;CHECK: ret i32
 define i32 @sum_array(i32* %A, i32 %n) nounwind uwtable readonly noinline ssp {
   %1 = sext i32 %n to i64
-  %2 = getelementptr inbounds i32* %A, i64 %1
+  %2 = getelementptr inbounds i32, i32* %A, i64 %1
   %3 = icmp eq i32 %n, 0
   br i1 %3, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %.lr.ph.i
 
 .lr.ph.i:                                         ; preds = %0, %.lr.ph.i
   %.03.i = phi i32* [ %6, %.lr.ph.i ], [ %A, %0 ]
   %.012.i = phi i32 [ %5, %.lr.ph.i ], [ 0, %0 ]
-  %4 = load i32* %.03.i, align 4
+  %4 = load i32, i32* %.03.i, align 4
   %5 = add nsw i32 %4, %.012.i
-  %6 = getelementptr inbounds i32* %.03.i, i64 1
+  %6 = getelementptr inbounds i32, i32* %.03.i, i64 1
   %7 = icmp eq i32* %6, %2
   br i1 %7, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %.lr.ph.i
 
@@ -41,16 +41,16 @@ _ZSt10accumulateIPiiET0_T_S2_S1_.exit:            ; preds = %.lr.ph.i, %0
 ;CHECK: ret i32
 define i32 @sum_array_as1(i32 addrspace(1)* %A, i32 %n) nounwind uwtable readonly noinline ssp {
   %1 = sext i32 %n to i64
-  %2 = getelementptr inbounds i32 addrspace(1)* %A, i64 %1
+  %2 = getelementptr inbounds i32, i32 addrspace(1)* %A, i64 %1
   %3 = icmp eq i32 %n, 0
   br i1 %3, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %.lr.ph.i
 
 .lr.ph.i:                                         ; preds = %0, %.lr.ph.i
   %.03.i = phi i32 addrspace(1)* [ %6, %.lr.ph.i ], [ %A, %0 ]
   %.012.i = phi i32 [ %5, %.lr.ph.i ], [ 0, %0 ]
-  %4 = load i32 addrspace(1)* %.03.i, align 4
+  %4 = load i32, i32 addrspace(1)* %.03.i, align 4
   %5 = add nsw i32 %4, %.012.i
-  %6 = getelementptr inbounds i32 addrspace(1)* %.03.i, i64 1
+  %6 = getelementptr inbounds i32, i32 addrspace(1)* %.03.i, i64 1
   %7 = icmp eq i32 addrspace(1)* %6, %2
   br i1 %7, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %.lr.ph.i
 
diff --git a/test/Transforms/LoopVectorize/no_outside_user.ll b/test/Transforms/LoopVectorize/no_outside_user.ll
index bcd29c1a439a..7030b6b4df2d 100644
--- a/test/Transforms/LoopVectorize/no_outside_user.ll
+++ b/test/Transforms/LoopVectorize/no_outside_user.ll
@@ -20,7 +20,7 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i32 @main()  {
 bb:
-  %b.promoted = load i32* @b, align 4
+  %b.promoted = load i32, i32* @b, align 4
   br label %.lr.ph.i
 
 .lr.ph.i:
@@ -56,7 +56,7 @@ f1.exit.loopexit:
 define i32 @test2()  {
 entry:
   store i32 0, i32* @x1, align 4
-  %0 = load i32* @x0, align 4
+  %0 = load i32, i32* @x0, align 4
   br label %for.cond1.preheader
 
 for.cond1.preheader:
diff --git a/test/Transforms/LoopVectorize/no_switch.ll b/test/Transforms/LoopVectorize/no_switch.ll
index 64aab37592f6..1f139c26d790 100644
--- a/test/Transforms/LoopVectorize/no_switch.ll
+++ b/test/Transforms/LoopVectorize/no_switch.ll
@@ -21,8 +21,8 @@ for.body.preheader:                               ; preds = %entry
 
 for.body:                                         ; preds = %for.body.preheader, %for.inc
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.inc ], [ 0, %for.body.preheader ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv, !dbg !14
-  %0 = load i32* %arrayidx, align 4, !dbg !14, !tbaa !16
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv, !dbg !14
+  %0 = load i32, i32* %arrayidx, align 4, !dbg !14, !tbaa !16
   switch i32 %0, label %for.inc [
     i32 0, label %sw.bb
     i32 1, label %sw.bb3
@@ -59,28 +59,28 @@ attributes #0 = { nounwind }
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\004\00clang version 3.5.0\001\00\006\00\002", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./source.cpp] [DW_LANG_C_plus_plus]
-!1 = !{!"source.cpp", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5.0", isOptimized: true, runtimeVersion: 6, emissionKind: 2, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "source.cpp", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00test_switch\00test_switch\00\001\000\001\000\006\00256\001\001", !1, !5, !6, null, void (i32*, i32)* @_Z11test_switchPii, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [test_switch]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./source.cpp]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "test_switch", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 1, file: !1, scope: !5, type: !6, function: void (i32*, i32)* @_Z11test_switchPii, variables: !2)
+!5 = !DIFile(filename: "source.cpp", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 2}
-!8 = !{i32 2, !"Debug Info Version", i32 2}
+!8 = !{i32 2, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5.0"}
-!10 = !MDLocation(line: 3, column: 8, scope: !11)
-!11 = !{!"0xb\003\003\000", !1, !4} ; [ DW_TAG_lexical_block ]
+!10 = !DILocation(line: 3, column: 8, scope: !11)
+!11 = distinct !DILexicalBlock(line: 3, column: 3, file: !1, scope: !4)
 !12 = !{!12, !13, !13}
 !13 = !{!"llvm.loop.vectorize.enable", i1 true}
-!14 = !MDLocation(line: 4, column: 5, scope: !15)
-!15 = !{!"0xb\003\0036\000", !1, !11} ; [ DW_TAG_lexical_block ]
+!14 = !DILocation(line: 4, column: 5, scope: !15)
+!15 = distinct !DILexicalBlock(line: 3, column: 36, file: !1, scope: !11)
 !16 = !{!17, !17, i64 0}
 !17 = !{!"int", !18, i64 0}
 !18 = !{!"omnipotent char", !19, i64 0}
 !19 = !{!"Simple C/C++ TBAA"}
-!20 = !MDLocation(line: 6, column: 7, scope: !21)
-!21 = !{!"0xb\004\0018\000", !1, !15} ; [ DW_TAG_lexical_block ]
-!22 = !MDLocation(line: 7, column: 5, scope: !21)
-!23 = !MDLocation(line: 9, column: 7, scope: !21)
-!24 = !MDLocation(line: 14, column: 1, scope: !4)
+!20 = !DILocation(line: 6, column: 7, scope: !21)
+!21 = distinct !DILexicalBlock(line: 4, column: 18, file: !1, scope: !15)
+!22 = !DILocation(line: 7, column: 5, scope: !21)
+!23 = !DILocation(line: 9, column: 7, scope: !21)
+!24 = !DILocation(line: 14, column: 1, scope: !4)
diff --git a/test/Transforms/LoopVectorize/nofloat.ll b/test/Transforms/LoopVectorize/nofloat.ll
index e9f4c5f23e08..6956bd9c0b45 100644
--- a/test/Transforms/LoopVectorize/nofloat.ll
+++ b/test/Transforms/LoopVectorize/nofloat.ll
@@ -15,7 +15,7 @@ define void @example12() noimplicitfloat { ;           <--------- "noimplicitflo
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   %3 = trunc i64 %indvars.iv to i32
   store i32 %3, i32* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/non-const-n.ll b/test/Transforms/LoopVectorize/non-const-n.ll
index b03d4f05e16f..9007b1b00fd0 100644
--- a/test/Transforms/LoopVectorize/non-const-n.ll
+++ b/test/Transforms/LoopVectorize/non-const-n.ll
@@ -20,12 +20,12 @@ define void @example1(i32 %n) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/nsw-crash.ll b/test/Transforms/LoopVectorize/nsw-crash.ll
index 68d993340bdb..3631f0be6904 100644
--- a/test/Transforms/LoopVectorize/nsw-crash.ll
+++ b/test/Transforms/LoopVectorize/nsw-crash.ll
@@ -13,7 +13,7 @@ while.body.lr.ph:
 
 while.body:
   %it.sroa.0.091 = phi i32* [ undef, %while.body.lr.ph ], [ %incdec.ptr.i, %while.body ]
-  %incdec.ptr.i = getelementptr inbounds i32* %it.sroa.0.091, i64 1
+  %incdec.ptr.i = getelementptr inbounds i32, i32* %it.sroa.0.091, i64 1
   %inc32 = add i32 undef, 1                                        ; <------------- Make sure we don't set NSW flags to the undef.
   %cmp.i11 = icmp eq i32* %incdec.ptr.i, undef
   br i1 %cmp.i11, label %while.end, label %while.body
diff --git a/test/Transforms/LoopVectorize/opt.ll b/test/Transforms/LoopVectorize/opt.ll
index a9be80fbfee6..71b7e1f906b7 100644
--- a/test/Transforms/LoopVectorize/opt.ll
+++ b/test/Transforms/LoopVectorize/opt.ll
@@ -16,8 +16,8 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %red.05 = phi i32 [ 0, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %red.05
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 255
diff --git a/test/Transforms/LoopVectorize/ptr_loops.ll b/test/Transforms/LoopVectorize/ptr_loops.ll
index 3fb38fe1ef13..bc804133e552 100644
--- a/test/Transforms/LoopVectorize/ptr_loops.ll
+++ b/test/Transforms/LoopVectorize/ptr_loops.ll
@@ -15,14 +15,14 @@ define i32 @_Z5test1v() nounwind uwtable ssp {
   br label %1
 
 ; <label>:1                                       ; preds = %0, %1
-  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32]* @A, i64 0, i64 18), %0 ], [ %4, %1 ]
-  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32]* @B, i64 0, i64 0), %0 ], [ %5, %1 ]
-  %2 = load i32* %b.01, align 4
+  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @A, i64 0, i64 18), %0 ], [ %4, %1 ]
+  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @B, i64 0, i64 0), %0 ], [ %5, %1 ]
+  %2 = load i32, i32* %b.01, align 4
   %3 = shl nsw i32 %2, 1
   store i32 %3, i32* %p.02, align 4
-  %4 = getelementptr inbounds i32* %p.02, i64 -1
-  %5 = getelementptr inbounds i32* %b.01, i64 1
-  %6 = icmp eq i32* %4, getelementptr ([36 x i32]* @A, i64 128102389400760775, i64 3)
+  %4 = getelementptr inbounds i32, i32* %p.02, i64 -1
+  %5 = getelementptr inbounds i32, i32* %b.01, i64 1
+  %6 = icmp eq i32* %4, getelementptr ([36 x i32], [36 x i32]* @A, i64 128102389400760775, i64 3)
   br i1 %6, label %7, label %1
 
 ; <label>:7                                       ; preds = %1
@@ -38,13 +38,13 @@ define i32 @_Z5test2v() nounwind uwtable ssp {
   br label %1
 
 ; <label>:1                                       ; preds = %0, %1
-  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32]* @A, i64 0, i64 25), %0 ], [ %3, %1 ]
-  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32]* @B, i64 0, i64 2), %0 ], [ %4, %1 ]
-  %2 = load i32* %b.01, align 4
+  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @A, i64 0, i64 25), %0 ], [ %3, %1 ]
+  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @B, i64 0, i64 2), %0 ], [ %4, %1 ]
+  %2 = load i32, i32* %b.01, align 4
   store i32 %2, i32* %p.02, align 4
-  %3 = getelementptr inbounds i32* %p.02, i64 -1
-  %4 = getelementptr inbounds i32* %b.01, i64 1
-  %5 = icmp eq i32* %4, getelementptr inbounds ([36 x i32]* @A, i64 0, i64 18)
+  %3 = getelementptr inbounds i32, i32* %p.02, i64 -1
+  %4 = getelementptr inbounds i32, i32* %b.01, i64 1
+  %5 = icmp eq i32* %4, getelementptr inbounds ([36 x i32], [36 x i32]* @A, i64 0, i64 18)
   br i1 %5, label %6, label %1
 
 ; <label>:6                                       ; preds = %1
@@ -60,13 +60,13 @@ define i32 @_Z5test3v() nounwind uwtable ssp {
   br label %1
 
 ; <label>:1                                       ; preds = %0, %1
-  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32]* @A, i64 0, i64 29), %0 ], [ %3, %1 ]
-  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32]* @B, i64 0, i64 5), %0 ], [ %4, %1 ]
-  %2 = load i32* %b.01, align 4
+  %p.02 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @A, i64 0, i64 29), %0 ], [ %3, %1 ]
+  %b.01 = phi i32* [ getelementptr inbounds ([36 x i32], [36 x i32]* @B, i64 0, i64 5), %0 ], [ %4, %1 ]
+  %2 = load i32, i32* %b.01, align 4
   store i32 %2, i32* %p.02, align 4
-  %3 = getelementptr inbounds i32* %p.02, i64 -1
-  %4 = getelementptr inbounds i32* %b.01, i64 1
-  %5 = icmp eq i32* %3, getelementptr ([36 x i32]* @A, i64 128102389400760775, i64 3)
+  %3 = getelementptr inbounds i32, i32* %p.02, i64 -1
+  %4 = getelementptr inbounds i32, i32* %b.01, i64 1
+  %5 = icmp eq i32* %3, getelementptr ([36 x i32], [36 x i32]* @A, i64 128102389400760775, i64 3)
   br i1 %5, label %6, label %1
 
 ; <label>:6                                       ; preds = %1
diff --git a/test/Transforms/LoopVectorize/read-only.ll b/test/Transforms/LoopVectorize/read-only.ll
index 2f7a96a09d22..f81afd6ad348 100644
--- a/test/Transforms/LoopVectorize/read-only.ll
+++ b/test/Transforms/LoopVectorize/read-only.ll
@@ -13,11 +13,11 @@ define i32 @read_only_func(i32* nocapture %A, i32* nocapture %B, i32 %n) nounwin
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = add nsw i64 %indvars.iv, 13
-  %5 = getelementptr inbounds i32* %B, i64 %4
-  %6 = load i32* %5, align 4
+  %5 = getelementptr inbounds i32, i32* %B, i64 %4
+  %6 = load i32, i32* %5, align 4
   %7 = shl i32 %6, 1
   %8 = add i32 %3, %sum.02
   %9 = add i32 %8, %7
diff --git a/test/Transforms/LoopVectorize/reduction.ll b/test/Transforms/LoopVectorize/reduction.ll
index 5e6b7fa332b2..647e58a7e41f 100644
--- a/test/Transforms/LoopVectorize/reduction.ll
+++ b/test/Transforms/LoopVectorize/reduction.ll
@@ -20,10 +20,10 @@ define i32 @reduction_sum(i32 %n, i32* noalias nocapture %A, i32* noalias nocapt
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = trunc i64 %indvars.iv to i32
   %7 = add i32 %sum.02, %6
   %8 = add i32 %7, %3
@@ -55,10 +55,10 @@ define i32 @reduction_prod(i32 %n, i32* noalias nocapture %A, i32* noalias nocap
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %prod.02 = phi i32 [ %9, %.lr.ph ], [ 1, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = trunc i64 %indvars.iv to i32
   %7 = mul i32 %prod.02, %6
   %8 = mul i32 %7, %3
@@ -90,10 +90,10 @@ define i32 @reduction_mix(i32 %n, i32* noalias nocapture %A, i32* noalias nocapt
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %sum.02 = phi i32 [ %9, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = mul nsw i32 %5, %3
   %7 = trunc i64 %indvars.iv to i32
   %8 = add i32 %sum.02, %7
@@ -123,10 +123,10 @@ define i32 @reduction_mul(i32 %n, i32* noalias nocapture %A, i32* noalias nocapt
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
   %sum.02 = phi i32 [ %9, %.lr.ph ], [ 19, %0 ]
-  %2 = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = trunc i64 %indvars.iv to i32
   %7 = add i32 %3, %6
   %8 = add i32 %7, %5
@@ -158,10 +158,10 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.09 = phi i32 [ %add, %for.body ], [ 120, %entry ]
-  %arrayidx = getelementptr inbounds i32* %in, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %coeff, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %in, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %coeff, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %mul = mul nsw i32 %1, %0
   %add = add nsw i32 %mul, %sum.09
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -191,10 +191,10 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %result.08 = phi i32 [ %and, %for.body ], [ -1, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
   %and = and i32 %add, %result.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -223,10 +223,10 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %result.08 = phi i32 [ %or, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
   %or = or i32 %add, %result.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -255,10 +255,10 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %result.08 = phi i32 [ %xor, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
   %xor = xor i32 %add, %result.08
   %indvars.iv.next = add i64 %indvars.iv, 1
@@ -284,8 +284,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %x.05 = phi i32 [ %sub, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %sub = sub nsw i32 %0, %x.05
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -311,8 +311,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %x.05 = phi i32 [ %sub, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %sub = sub nsw i32 %x.05, %0
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -335,10 +335,10 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
   %sum.033 = phi float [ %S, %entry ], [ %sum.1, %for.inc ]
-  %arrayidx = getelementptr inbounds float* %A, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %B, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %cmp3 = fcmp ogt float %0, %1
   br i1 %cmp3, label %if.then, label %for.inc
 
@@ -380,10 +380,10 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
   %sum.033 = phi float [ %S, %entry ], [ %sum.1, %for.inc ]
-  %arrayidx = getelementptr inbounds float* %A, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds float* %B, i64 %indvars.iv
-  %1 = load float* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds float, float* %A, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %1 = load float, float* %arrayidx2, align 4
   %cmp3 = fcmp ogt float %0, %1
   br i1 %cmp3, label %if.then, label %for.inc
 
@@ -427,8 +427,8 @@ for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %sum2.09 = phi float [ 0.000000e+00, %entry ], [ %add1, %for.body ]
   %sum.08 = phi float [ %S, %entry ], [ %add, %for.body ]
-  %arrayidx = getelementptr inbounds float* %B, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %B, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %add = fadd fast float %sum.08, %0
   %add1 = fadd fast float %sum2.09, %add
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/reverse_induction.ll b/test/Transforms/LoopVectorize/reverse_induction.ll
index da02d01706ca..6b63a0d8db6c 100644
--- a/test/Transforms/LoopVectorize/reverse_induction.ll
+++ b/test/Transforms/LoopVectorize/reverse_induction.ll
@@ -18,8 +18,8 @@ for.body:
   %i.06 = phi i32 [ 0, %entry ], [ %inc4, %for.body ]
   %redux5 = phi i32 [ 0, %entry ], [ %inc.redux, %for.body ]
   %add.i = add i64 %add.i7, -1
-  %kind_.i = getelementptr inbounds i32* %ptr, i64 %add.i
-  %tmp.i1 = load i32* %kind_.i, align 4
+  %kind_.i = getelementptr inbounds i32, i32* %ptr, i64 %add.i
+  %tmp.i1 = load i32, i32* %kind_.i, align 4
   %inc.redux = add i32 %tmp.i1, %redux5
   %inc4 = add i32 %i.06, 1
   %exitcond = icmp ne i32 %inc4, 1024
@@ -41,8 +41,8 @@ for.body:
   %i.06 = phi i32 [ 0, %entry ], [ %inc4, %for.body ]
   %redux5 = phi i32 [ 0, %entry ], [ %inc.redux, %for.body ]
   %add.i = add i128 %add.i7, -1
-  %kind_.i = getelementptr inbounds i32* %ptr, i128 %add.i
-  %tmp.i1 = load i32* %kind_.i, align 4
+  %kind_.i = getelementptr inbounds i32, i32* %ptr, i128 %add.i
+  %tmp.i1 = load i32, i32* %kind_.i, align 4
   %inc.redux = add i32 %tmp.i1, %redux5
   %inc4 = add i32 %i.06, 1
   %exitcond = icmp ne i32 %inc4, 1024
@@ -65,8 +65,8 @@ for.body:
   %i.06 = phi i32 [ 0, %entry ], [ %inc4, %for.body ]
   %redux5 = phi i32 [ 0, %entry ], [ %inc.redux, %for.body ]
   %add.i = add i16 %add.i7, -1
-  %kind_.i = getelementptr inbounds i32* %ptr, i16 %add.i
-  %tmp.i1 = load i32* %kind_.i, align 4
+  %kind_.i = getelementptr inbounds i32, i32* %ptr, i16 %add.i
+  %tmp.i1 = load i32, i32* %kind_.i, align 4
   %inc.redux = add i32 %tmp.i1, %redux5
   %inc4 = add i32 %i.06, 1
   %exitcond = icmp ne i32 %inc4, 1024
@@ -97,7 +97,7 @@ loopend:
 ; CHECK: vector.body
 ; CHECK: %index = phi i64 [ 0, %vector.ph ], [ %index.next, %vector.body ]
 ; CHECK: %normalized.idx = sub i64 %index, 0
-; CHECK: %reverse.idx = sub i64 1023, %normalized.idx
+; CHECK: %offset.idx = sub i64 1023, %normalized.idx
 ; CHECK: trunc i64 %index to i8
 
 define void @reverse_forward_induction_i64_i8() {
@@ -109,7 +109,7 @@ while.body:
   %forward_induction.05 = phi i8 [ 0, %entry ], [ %inc, %while.body ]
   %inc = add i8 %forward_induction.05, 1
   %conv = zext i8 %inc to i32
-  %arrayidx = getelementptr inbounds [1024 x i32]* @a, i64 0, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, -1
   %0 = trunc i64 %indvars.iv to i32
@@ -124,7 +124,7 @@ while.end:
 ; CHECK: vector.body:
 ; CHECK:  %index = phi i64 [ 129, %vector.ph ], [ %index.next, %vector.body ]
 ; CHECK:  %normalized.idx = sub i64 %index, 129
-; CHECK:  %reverse.idx = sub i64 1023, %normalized.idx
+; CHECK:  %offset.idx = sub i64 1023, %normalized.idx
 ; CHECK:  trunc i64 %index to i8
 
 define void @reverse_forward_induction_i64_i8_signed() {
@@ -136,7 +136,7 @@ while.body:
   %forward_induction.05 = phi i8 [ -127, %entry ], [ %inc, %while.body ]
   %inc = add i8 %forward_induction.05, 1
   %conv = sext i8 %inc to i32
-  %arrayidx = getelementptr inbounds [1024 x i32]* @a, i64 0, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, -1
   %0 = trunc i64 %indvars.iv to i32
diff --git a/test/Transforms/LoopVectorize/reverse_iter.ll b/test/Transforms/LoopVectorize/reverse_iter.ll
index 13172bb17aeb..5bbc769a20aa 100644
--- a/test/Transforms/LoopVectorize/reverse_iter.ll
+++ b/test/Transforms/LoopVectorize/reverse_iter.ll
@@ -30,8 +30,8 @@ define i32 @foo(i32 %n, i32* nocapture %A) {
   %4 = trunc i64 %indvars.iv to i32
   %5 = shl nsw i32 %4, 1
   %6 = sext i32 %5 to i64
-  %7 = getelementptr inbounds i32* %A, i64 %6
-  %8 = load i32* %7, align 4
+  %7 = getelementptr inbounds i32, i32* %A, i64 %6
+  %8 = load i32, i32* %7, align 4
   %9 = add nsw i32 %8, %sum.01
   %indvars.iv.next = add i64 %indvars.iv, -1
   %10 = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/runtime-check-address-space.ll b/test/Transforms/LoopVectorize/runtime-check-address-space.ll
index ec56f80144c4..8e7ac1f865a8 100644
--- a/test/Transforms/LoopVectorize/runtime-check-address-space.ll
+++ b/test/Transforms/LoopVectorize/runtime-check-address-space.ll
@@ -37,11 +37,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %idxprom
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %idxprom
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds i32 addrspace(1)* %a, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %a, i64 %idxprom1
   store i32 %mul, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -64,11 +64,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %idxprom
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %idxprom
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %idxprom1
   store i32 %mul, i32* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -91,11 +91,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds i32* %b, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds i32 addrspace(1)* %a, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %a, i64 %idxprom1
   store i32 %mul, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -119,11 +119,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %idxprom
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %idxprom
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %idxprom1
   store i32 %mul, i32* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -147,11 +147,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds i32* %b, i64 %idxprom
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %idxprom
+  %0 = load i32, i32* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds [1024 x i32], [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom1
   store i32 %mul, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -176,11 +176,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds i32* %b, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds i32, i32* %b, i64 %idxprom1
   store i32 %mul, i32* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
@@ -204,11 +204,11 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.02 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %idxprom = sext i32 %i.02 to i64
-  %arrayidx = getelementptr inbounds [1024 x i32] addrspace(2)* @q_as2, i64 0, i64 %idxprom
-  %0 = load i32 addrspace(2)* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [1024 x i32], [1024 x i32] addrspace(2)* @q_as2, i64 0, i64 %idxprom
+  %0 = load i32, i32 addrspace(2)* %arrayidx, align 4
   %mul = mul nsw i32 %0, 3
   %idxprom1 = sext i32 %i.02 to i64
-  %arrayidx2 = getelementptr inbounds [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom1
+  %arrayidx2 = getelementptr inbounds [1024 x i32], [1024 x i32] addrspace(1)* @g_as1, i64 0, i64 %idxprom1
   store i32 %mul, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add nsw i32 %i.02, 1
   %cmp = icmp slt i32 %inc, %n
diff --git a/test/Transforms/LoopVectorize/runtime-check-readonly-address-space.ll b/test/Transforms/LoopVectorize/runtime-check-readonly-address-space.ll
index 12ba3ce6c667..6ee983d22c6d 100644
--- a/test/Transforms/LoopVectorize/runtime-check-readonly-address-space.ll
+++ b/test/Transforms/LoopVectorize/runtime-check-readonly-address-space.ll
@@ -14,12 +14,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %i.01
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32 addrspace(1)* %c, i64 %i.01
-  %1 = load i32 addrspace(1)* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %i.01
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %c, i64 %i.01
+  %1 = load i32, i32 addrspace(1)* %arrayidx1, align 4
   %add = add nsw i32 %0, %1
-  %arrayidx2 = getelementptr inbounds i32 addrspace(1)* %a, i64 %i.01
+  %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %a, i64 %i.01
   store i32 %add, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add i64 %i.01, 1
   %cmp = icmp ult i64 %inc, 200
@@ -39,12 +39,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %b, i64 %i.01
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32* %c, i64 %i.01
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %b, i64 %i.01
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %c, i64 %i.01
+  %1 = load i32, i32* %arrayidx1, align 4
   %add = add nsw i32 %0, %1
-  %arrayidx2 = getelementptr inbounds i32 addrspace(1)* %a, i64 %i.01
+  %arrayidx2 = getelementptr inbounds i32, i32 addrspace(1)* %a, i64 %i.01
   store i32 %add, i32 addrspace(1)* %arrayidx2, align 4
   %inc = add i64 %i.01, 1
   %cmp = icmp ult i64 %inc, 200
@@ -64,12 +64,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %i.01
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32* %c, i64 %i.01
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %i.01
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %c, i64 %i.01
+  %1 = load i32, i32* %arrayidx1, align 4
   %add = add nsw i32 %0, %1
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %i.01
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %i.01
   store i32 %add, i32* %arrayidx2, align 4
   %inc = add i64 %i.01, 1
   %cmp = icmp ult i64 %inc, 200
@@ -89,12 +89,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %i.01
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32 addrspace(1)* %c, i64 %i.01
-  %1 = load i32 addrspace(1)* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %i.01
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32 addrspace(1)* %c, i64 %i.01
+  %1 = load i32, i32 addrspace(1)* %arrayidx1, align 4
   %add = add nsw i32 %0, %1
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %i.01
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %i.01
   store i32 %add, i32* %arrayidx2, align 4
   %inc = add i64 %i.01, 1
   %cmp = icmp ult i64 %inc, 200
@@ -114,12 +114,12 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.01 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %b, i64 %i.01
-  %0 = load i32 addrspace(1)* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32 addrspace(2)* %c, i64 %i.01
-  %1 = load i32 addrspace(2)* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %b, i64 %i.01
+  %0 = load i32, i32 addrspace(1)* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32 addrspace(2)* %c, i64 %i.01
+  %1 = load i32, i32 addrspace(2)* %arrayidx1, align 4
   %add = add nsw i32 %0, %1
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %i.01
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %i.01
   store i32 %add, i32* %arrayidx2, align 4
   %inc = add i64 %i.01, 1
   %cmp = icmp ult i64 %inc, 200
diff --git a/test/Transforms/LoopVectorize/runtime-check-readonly.ll b/test/Transforms/LoopVectorize/runtime-check-readonly.ll
index 9d02a6accd67..a3b5a598d220 100644
--- a/test/Transforms/LoopVectorize/runtime-check-readonly.ll
+++ b/test/Transforms/LoopVectorize/runtime-check-readonly.ll
@@ -22,12 +22,12 @@ entry:
 
 for.body:
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %B, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx2 = getelementptr inbounds i32* %C, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %C, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4
   %add = add nsw i32 %1, %0
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
   store i32 %add, i32* %arrayidx4, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/runtime-check.ll b/test/Transforms/LoopVectorize/runtime-check.ll
index 1edafb4ed7a9..b1c5d402c458 100644
--- a/test/Transforms/LoopVectorize/runtime-check.ll
+++ b/test/Transforms/LoopVectorize/runtime-check.ll
@@ -21,10 +21,10 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
   %mul = fmul float %0, 3.000000e+00
-  %arrayidx2 = getelementptr inbounds float* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds float, float* %a, i64 %indvars.iv
   store float %mul, float* %arrayidx2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -47,11 +47,11 @@ entry:
 for.body:
   %iv = phi i64 [ 0, %entry ], [ %iv.next, %for.body ]
   %ind.sum = add i64 %iv, %offset
-  %arr.idx = getelementptr inbounds float* %a, i64 %ind.sum
-  %l1 = load float* %arr.idx, align 4
+  %arr.idx = getelementptr inbounds float, float* %a, i64 %ind.sum
+  %l1 = load float, float* %arr.idx, align 4
   %ind.sum2 = add i64 %iv, %offset2
-  %arr.idx2 = getelementptr inbounds float* %a, i64 %ind.sum2
-  %l2 = load float* %arr.idx2, align 4
+  %arr.idx2 = getelementptr inbounds float, float* %a, i64 %ind.sum2
+  %l2 = load float, float* %arr.idx2, align 4
   %m = fmul fast float %b, %l2
   %ad = fadd fast float %l1, %m
   store float %ad, float* %arr.idx, align 4
diff --git a/test/Transforms/LoopVectorize/runtime-limit.ll b/test/Transforms/LoopVectorize/runtime-limit.ll
index 324949dafd0a..6bc71e160ccd 100644
--- a/test/Transforms/LoopVectorize/runtime-limit.ll
+++ b/test/Transforms/LoopVectorize/runtime-limit.ll
@@ -13,21 +13,21 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %i.016 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %i.016
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32* %B, i64 %i.016
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %i.016
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %B, i64 %i.016
+  %1 = load i32, i32* %arrayidx1, align 4
   %add = add nsw i32 %1, %0
-  %arrayidx2 = getelementptr inbounds i32* %C, i64 %i.016
-  %2 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %C, i64 %i.016
+  %2 = load i32, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %add, %2
-  %arrayidx4 = getelementptr inbounds i32* %E, i64 %i.016
-  %3 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %E, i64 %i.016
+  %3 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %add3, %3
-  %arrayidx6 = getelementptr inbounds i32* %F, i64 %i.016
-  %4 = load i32* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %F, i64 %i.016
+  %4 = load i32, i32* %arrayidx6, align 4
   %add7 = add nsw i32 %add5, %4
-  %arrayidx8 = getelementptr inbounds i32* %out, i64 %i.016
+  %arrayidx8 = getelementptr inbounds i32, i32* %out, i64 %i.016
   store i32 %add7, i32* %arrayidx8, align 4
   %inc = add i64 %i.016, 1
   %exitcond = icmp eq i64 %inc, 256
@@ -47,32 +47,32 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %i.037 = phi i64 [ 0, %entry ], [ %inc, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %i.037
-  %0 = load i32* %arrayidx, align 4
-  %arrayidx1 = getelementptr inbounds i32* %B, i64 %i.037
-  %1 = load i32* %arrayidx1, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %i.037
+  %0 = load i32, i32* %arrayidx, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %B, i64 %i.037
+  %1 = load i32, i32* %arrayidx1, align 4
   %add = add nsw i32 %1, %0
-  %arrayidx2 = getelementptr inbounds i32* %C, i64 %i.037
-  %2 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %C, i64 %i.037
+  %2 = load i32, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %add, %2
-  %arrayidx4 = getelementptr inbounds i32* %E, i64 %i.037
-  %3 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %E, i64 %i.037
+  %3 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %add3, %3
-  %arrayidx6 = getelementptr inbounds i32* %F, i64 %i.037
-  %4 = load i32* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %F, i64 %i.037
+  %4 = load i32, i32* %arrayidx6, align 4
   %add7 = add nsw i32 %add5, %4
-  %arrayidx8 = getelementptr inbounds i32* %out, i64 %i.037
+  %arrayidx8 = getelementptr inbounds i32, i32* %out, i64 %i.037
   store i32 %add7, i32* %arrayidx8, align 4
-  %5 = load i32* %arrayidx, align 4
-  %6 = load i32* %arrayidx1, align 4
+  %5 = load i32, i32* %arrayidx, align 4
+  %6 = load i32, i32* %arrayidx1, align 4
   %add11 = add nsw i32 %6, %5
-  %7 = load i32* %arrayidx2, align 4
+  %7 = load i32, i32* %arrayidx2, align 4
   %add13 = add nsw i32 %add11, %7
-  %8 = load i32* %arrayidx4, align 4
+  %8 = load i32, i32* %arrayidx4, align 4
   %add15 = add nsw i32 %add13, %8
-  %9 = load i32* %arrayidx6, align 4
+  %9 = load i32, i32* %arrayidx6, align 4
   %add17 = add nsw i32 %add15, %9
-  %arrayidx18 = getelementptr inbounds i32* %out2, i64 %i.037
+  %arrayidx18 = getelementptr inbounds i32, i32* %out2, i64 %i.037
   store i32 %add17, i32* %arrayidx18, align 4
   %inc = add i64 %i.037, 1
   %exitcond = icmp eq i64 %inc, 256
diff --git a/test/Transforms/LoopVectorize/safegep.ll b/test/Transforms/LoopVectorize/safegep.ll
index f853afd2bba6..ecef8138f0b2 100644
--- a/test/Transforms/LoopVectorize/safegep.ll
+++ b/test/Transforms/LoopVectorize/safegep.ll
@@ -17,10 +17,10 @@ entry:
 
 "<bb 3>":
   %i_15 = phi i32 [ 0, %entry ], [ %i_19, %"<bb 3>" ]
-  %pp3 = getelementptr float* %A, i32 %i_15
-  %D.1396_10 = load float* %pp3, align 4
-  %pp24 = getelementptr float* %B, i32 %i_15
-  %D.1398_15 = load float* %pp24, align 4
+  %pp3 = getelementptr float, float* %A, i32 %i_15
+  %D.1396_10 = load float, float* %pp3, align 4
+  %pp24 = getelementptr float, float* %B, i32 %i_15
+  %D.1398_15 = load float, float* %pp24, align 4
   %D.1399_17 = fadd float %D.1398_15, %K
   %D.1400_18 = fmul float %D.1396_10, %D.1399_17
   store float %D.1400_18, float* %pp3, align 4
@@ -43,10 +43,10 @@ entry:
 
 "<bb 3>":
   %i_15 = phi i32 [ 0, %entry ], [ %i_19, %"<bb 3>" ]
-  %pp3 = getelementptr float addrspace(5) * %A, i32 %i_15
-  %D.1396_10 = load float addrspace(5) * %pp3, align 4
-  %pp24 = getelementptr float* %B, i32 %i_15
-  %D.1398_15 = load float* %pp24, align 4
+  %pp3 = getelementptr float, float addrspace(5) * %A, i32 %i_15
+  %D.1396_10 = load float, float addrspace(5) * %pp3, align 4
+  %pp24 = getelementptr float, float* %B, i32 %i_15
+  %D.1398_15 = load float, float* %pp24, align 4
   %D.1399_17 = fadd float %D.1398_15, %K
   %D.1400_18 = fmul float %D.1396_10, %D.1399_17
   store float %D.1400_18, float addrspace(5) * %pp3, align 4
diff --git a/test/Transforms/LoopVectorize/same-base-access.ll b/test/Transforms/LoopVectorize/same-base-access.ll
index d19458f4aa92..31cff0ee653f 100644
--- a/test/Transforms/LoopVectorize/same-base-access.ll
+++ b/test/Transforms/LoopVectorize/same-base-access.ll
@@ -26,33 +26,33 @@ define i32 @kernel11(double* %x, double* %y, i32 %n) nounwind uwtable ssp {
   br label %4
 
 ; <label>:4                                       ; preds = %25, %0
-  %5 = load i32* %k, align 4
-  %6 = load i32* %3, align 4
+  %5 = load i32, i32* %k, align 4
+  %6 = load i32, i32* %3, align 4
   %7 = icmp slt i32 %5, %6
   br i1 %7, label %8, label %28
 
 ; <label>:8                                       ; preds = %4
-  %9 = load i32* %k, align 4
+  %9 = load i32, i32* %k, align 4
   %10 = sub nsw i32 %9, 1
   %11 = sext i32 %10 to i64
-  %12 = load double** %1, align 8
-  %13 = getelementptr inbounds double* %12, i64 %11
-  %14 = load double* %13, align 8
-  %15 = load i32* %k, align 4
+  %12 = load double*, double** %1, align 8
+  %13 = getelementptr inbounds double, double* %12, i64 %11
+  %14 = load double, double* %13, align 8
+  %15 = load i32, i32* %k, align 4
   %16 = sext i32 %15 to i64
-  %17 = load double** %2, align 8
-  %18 = getelementptr inbounds double* %17, i64 %16
-  %19 = load double* %18, align 8
+  %17 = load double*, double** %2, align 8
+  %18 = getelementptr inbounds double, double* %17, i64 %16
+  %19 = load double, double* %18, align 8
   %20 = fadd double %14, %19
-  %21 = load i32* %k, align 4
+  %21 = load i32, i32* %k, align 4
   %22 = sext i32 %21 to i64
-  %23 = load double** %1, align 8
-  %24 = getelementptr inbounds double* %23, i64 %22
+  %23 = load double*, double** %1, align 8
+  %24 = getelementptr inbounds double, double* %23, i64 %22
   store double %20, double* %24, align 8
   br label %25
 
 ; <label>:25                                      ; preds = %8
-  %26 = load i32* %k, align 4
+  %26 = load i32, i32* %k, align 4
   %27 = add nsw i32 %26, 1
   store i32 %27, i32* %k, align 4
   br label %4
@@ -86,8 +86,8 @@ define i32 @func2(i32* nocapture %a) nounwind uwtable ssp {
 ; <label>:1                                       ; preds = %7, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %7 ]
   %2 = mul nsw i64 %indvars.iv, 7
-  %3 = getelementptr inbounds i32* %a, i64 %2
-  %4 = load i32* %3, align 4
+  %3 = getelementptr inbounds i32, i32* %a, i64 %2
+  %4 = load i32, i32* %3, align 4
   %5 = icmp sgt i32 %4, 3
   br i1 %5, label %6, label %7
 
diff --git a/test/Transforms/LoopVectorize/scalar-select.ll b/test/Transforms/LoopVectorize/scalar-select.ll
index 6b37cc23ae31..b17b20288217 100644
--- a/test/Transforms/LoopVectorize/scalar-select.ll
+++ b/test/Transforms/LoopVectorize/scalar-select.ll
@@ -18,12 +18,12 @@ define void @example1(i1 %cond) nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   %sel = select i1 %cond, i32 %6, i32 zeroinitializer
   store i32 %sel, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/scev-exitlim-crash.ll b/test/Transforms/LoopVectorize/scev-exitlim-crash.ll
index 515477180772..cfc1d6d40a7e 100644
--- a/test/Transforms/LoopVectorize/scev-exitlim-crash.ll
+++ b/test/Transforms/LoopVectorize/scev-exitlim-crash.ll
@@ -29,14 +29,14 @@ for.cond4.preheader:                              ; preds = %for.cond
   br i1 %cmp514, label %for.cond7.preheader.lr.ph, label %for.end26
 
 for.cond7.preheader.lr.ph:                        ; preds = %for.cond4.preheader
-  %0 = load i32** @e, align 8, !tbaa !4
+  %0 = load i32*, i32** @e, align 8, !tbaa !4
   br label %for.cond7.preheader
 
 for.cond7.preheader:                              ; preds = %for.cond7.preheader.lr.ph, %for.inc23
   %y.017 = phi i32 [ 0, %for.cond7.preheader.lr.ph ], [ %inc24, %for.inc23 ]
   %i.116 = phi i32 [ 0, %for.cond7.preheader.lr.ph ], [ %i.2.lcssa, %for.inc23 ]
   %n.015 = phi i32 [ undef, %for.cond7.preheader.lr.ph ], [ %inc25, %for.inc23 ]
-  %1 = load i32* @b, align 4, !tbaa !5
+  %1 = load i32, i32* @b, align 4, !tbaa !5
   %tobool11 = icmp eq i32 %1, 0
   br i1 %tobool11, label %for.inc23, label %for.body8.lr.ph
 
@@ -51,7 +51,7 @@ for.body8:                                        ; preds = %for.body8.lr.ph, %f
   %add10 = add i32 %add9, %2
   store i32 %add10, i32* @f, align 4, !tbaa !5
   %idx.ext = sext i32 %add10 to i64
-  %add.ptr = getelementptr inbounds i32* @a, i64 %idx.ext
+  %add.ptr = getelementptr inbounds i32, i32* @a, i64 %idx.ext
   %tobool129 = icmp eq i32 %i.213, 0
   br i1 %tobool129, label %for.inc19, label %for.body13.lr.ph
 
@@ -62,9 +62,9 @@ for.body13.lr.ph:                                 ; preds = %for.body8
 for.body13:                                       ; preds = %for.body13.lr.ph, %for.body13
   %indvars.iv = phi i64 [ %3, %for.body13.lr.ph ], [ %indvars.iv.next, %for.body13 ]
   %add.ptr.sum = add i64 %idx.ext, %indvars.iv
-  %arrayidx = getelementptr inbounds i32* @a, i64 %add.ptr.sum
-  %4 = load i32* %arrayidx, align 4, !tbaa !5
-  %arrayidx15 = getelementptr inbounds i32* %0, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds i32, i32* @a, i64 %add.ptr.sum
+  %4 = load i32, i32* %arrayidx, align 4, !tbaa !5
+  %arrayidx15 = getelementptr inbounds i32, i32* %0, i64 %indvars.iv
   store i32 %4, i32* %arrayidx15, align 4, !tbaa !5
   %indvars.iv.next = add i64 %indvars.iv, 1
   %5 = trunc i64 %indvars.iv.next to i32
@@ -75,11 +75,11 @@ for.cond11.for.inc19_crit_edge:                   ; preds = %for.body13
   br label %for.inc19
 
 for.inc19:                                        ; preds = %for.cond11.for.inc19_crit_edge, %for.body8
-  %6 = load i32* @c, align 4, !tbaa !5
+  %6 = load i32, i32* @c, align 4, !tbaa !5
   %inc20 = add nsw i32 %6, 1
   store i32 %inc20, i32* @c, align 4, !tbaa !5
   %indvars.iv.next20 = add i64 %indvars.iv19, 1
-  %7 = load i32* @b, align 4, !tbaa !5
+  %7 = load i32, i32* @b, align 4, !tbaa !5
   %tobool = icmp eq i32 %7, 0
   br i1 %tobool, label %for.cond7.for.inc23_crit_edge, label %for.body8
 
diff --git a/test/Transforms/LoopVectorize/simple-unroll.ll b/test/Transforms/LoopVectorize/simple-unroll.ll
index 8bf680ac7c5f..fc996ed3818d 100644
--- a/test/Transforms/LoopVectorize/simple-unroll.ll
+++ b/test/Transforms/LoopVectorize/simple-unroll.ll
@@ -23,8 +23,8 @@ define void @inc(i32 %n) nounwind uwtable noinline ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = trunc i64 %indvars.iv to i32
   %5 = add nsw i32 %3, %4
   store i32 %5, i32* %2, align 4
diff --git a/test/Transforms/LoopVectorize/small-loop.ll b/test/Transforms/LoopVectorize/small-loop.ll
index 1d30102c3f97..ce606d1b3606 100644
--- a/test/Transforms/LoopVectorize/small-loop.ll
+++ b/test/Transforms/LoopVectorize/small-loop.ll
@@ -15,12 +15,12 @@ define void @example1() nounwind uwtable ssp {
 
 ; <label>:1                                       ; preds = %1, %0
   %indvars.iv = phi i64 [ 0, %0 ], [ %indvars.iv.next, %1 ]
-  %2 = getelementptr inbounds [2048 x i32]* @b, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
-  %4 = getelementptr inbounds [2048 x i32]* @c, i64 0, i64 %indvars.iv
-  %5 = load i32* %4, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @b, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
+  %4 = getelementptr inbounds [2048 x i32], [2048 x i32]* @c, i64 0, i64 %indvars.iv
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %5, %3
-  %7 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %7 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
   store i32 %6, i32* %7, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/start-non-zero.ll b/test/Transforms/LoopVectorize/start-non-zero.ll
index cc474945a9d0..b444a692319a 100644
--- a/test/Transforms/LoopVectorize/start-non-zero.ll
+++ b/test/Transforms/LoopVectorize/start-non-zero.ll
@@ -17,8 +17,8 @@ for.body.lr.ph:                                   ; preds = %entry
 
 for.body:                                         ; preds = %for.body.lr.ph, %for.body
   %indvars.iv = phi i64 [ %0, %for.body.lr.ph ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx, align 4
   %mul = mul nuw i32 %1, 333
   store i32 %mul, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/store-shuffle-bug.ll b/test/Transforms/LoopVectorize/store-shuffle-bug.ll
index 6d3d11349b14..4d62df157175 100644
--- a/test/Transforms/LoopVectorize/store-shuffle-bug.ll
+++ b/test/Transforms/LoopVectorize/store-shuffle-bug.ll
@@ -30,14 +30,14 @@ entry:
 for.body:
   %indvars.iv = phi i64 [ 93, %entry ], [ %indvars.iv.next, %for.body ]
   %0 = add i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds [100 x i32]* @uf, i64 0, i64 %0
-  %arrayidx3 = getelementptr inbounds [100 x i32]* @xi, i64 0, i64 %0
-  %1 = load i32* %arrayidx3, align 4
-  %2 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [100 x i32], [100 x i32]* @uf, i64 0, i64 %0
+  %arrayidx3 = getelementptr inbounds [100 x i32], [100 x i32]* @xi, i64 0, i64 %0
+  %1 = load i32, i32* %arrayidx3, align 4
+  %2 = load i32, i32* %arrayidx, align 4
   %add4 = add nsw i32 %2, %1
   store i32 %add4, i32* %arrayidx, align 4
-  %arrayidx7 = getelementptr inbounds [100 x i32]* @q, i64 0, i64 %0
-  %3 = load i32* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds [100 x i32], [100 x i32]* @q, i64 0, i64 %0
+  %3 = load i32, i32* %arrayidx7, align 4
   %add8 = add nsw i32 %add4, %3
   store i32 %add8, i32* %arrayidx, align 4
   %indvars.iv.next = add i64 %indvars.iv, -1
diff --git a/test/Transforms/LoopVectorize/struct_access.ll b/test/Transforms/LoopVectorize/struct_access.ll
index cf6f325ae306..1e4019a7810a 100644
--- a/test/Transforms/LoopVectorize/struct_access.ll
+++ b/test/Transforms/LoopVectorize/struct_access.ll
@@ -32,8 +32,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.05 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %x = getelementptr inbounds %struct.coordinate* %A, i64 %indvars.iv, i32 0
-  %0 = load i32* %x, align 4
+  %x = getelementptr inbounds %struct.coordinate, %struct.coordinate* %A, i64 %indvars.iv, i32 0
+  %0 = load i32, i32* %x, align 4
   %add = add nsw i32 %0, %sum.05
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -74,8 +74,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %sum.05 = phi i32 [ %add, %for.body ], [ 0, %entry ]
-  %x = getelementptr inbounds %struct.lit* %A, i64 %indvars.iv, i32 0
-  %0 = load i32* %x, align 4
+  %x = getelementptr inbounds %struct.lit, %struct.lit* %A, i64 %indvars.iv, i32 0
+  %0 = load i32, i32* %x, align 4
   %add = add nsw i32 %0, %sum.05
   %indvars.iv.next = add i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/tbaa-nodep.ll b/test/Transforms/LoopVectorize/tbaa-nodep.ll
index be3e74f5d68a..06d000230027 100644
--- a/test/Transforms/LoopVectorize/tbaa-nodep.ll
+++ b/test/Transforms/LoopVectorize/tbaa-nodep.ll
@@ -10,10 +10,10 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !tbaa !0
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !tbaa !0
   %conv = fptosi float %0 to i32
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
   store i32 %conv, i32* %arrayidx2, align 4, !tbaa !4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 1600
@@ -30,7 +30,7 @@ for.end:                                          ; preds = %for.body
 ; CHECK-NEXT: br label %vector.body
 ; CHECK: vector.body:
 
-; CHECK: load <4 x float>* %{{.*}}, align 4, !tbaa
+; CHECK: load <4 x float>, <4 x float>* %{{.*}}, align 4, !tbaa
 ; CHECK: store <4 x i32> %{{.*}}, <4 x i32>* %{{.*}}, align 4, !tbaa
 
 ; CHECK: ret i32 0
@@ -38,7 +38,7 @@ for.end:                                          ; preds = %for.body
 ; CHECK-NOTBAA-LABEL: @test1
 ; CHECK-NOTBAA: icmp uge i32*
 
-; CHECK-NOTBAA: load <4 x float>* %{{.*}}, align 4, !tbaa
+; CHECK-NOTBAA: load <4 x float>, <4 x float>* %{{.*}}, align 4, !tbaa
 ; CHECK-NOTBAA: store <4 x i32> %{{.*}}, <4 x i32>* %{{.*}}, align 4, !tbaa
 
 ; CHECK-NOTBAA: ret i32 0
@@ -51,13 +51,13 @@ entry:
 
 for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
-  %arrayidx = getelementptr inbounds float* %b, i64 %indvars.iv
-  %0 = load float* %arrayidx, align 4, !tbaa !0
-  %arrayidx2 = getelementptr inbounds i32* %a, i64 %indvars.iv
-  %1 = load i32* %arrayidx2, align 4, !tbaa !4
+  %arrayidx = getelementptr inbounds float, float* %b, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4, !tbaa !0
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i64 %indvars.iv
+  %1 = load i32, i32* %arrayidx2, align 4, !tbaa !4
   %conv = sitofp i32 %1 to float
   %mul = fmul float %0, %conv
-  %arrayidx4 = getelementptr inbounds float* %c, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds float, float* %c, i64 %indvars.iv
   store float %mul, float* %arrayidx4, align 4, !tbaa !0
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %exitcond = icmp eq i64 %indvars.iv.next, 1600
@@ -74,7 +74,7 @@ for.end:                                          ; preds = %for.body
 ; CHECK: icmp uge float*
 ; CHECK-NOT: icmp uge i32*
 
-; CHECK: load <4 x float>* %{{.*}}, align 4, !tbaa
+; CHECK: load <4 x float>, <4 x float>* %{{.*}}, align 4, !tbaa
 ; CHECK: store <4 x float> %{{.*}}, <4 x float>* %{{.*}}, align 4, !tbaa
 
 ; CHECK: ret i32 0
@@ -85,7 +85,7 @@ for.end:                                          ; preds = %for.body
 ; CHECK-NOTBAA-DAG: icmp uge float*
 ; CHECK-NOTBAA-DAG: icmp uge i32*
 
-; CHECK-NOTBAA: load <4 x float>* %{{.*}}, align 4, !tbaa
+; CHECK-NOTBAA: load <4 x float>, <4 x float>* %{{.*}}, align 4, !tbaa
 ; CHECK-NOTBAA: store <4 x float> %{{.*}}, <4 x float>* %{{.*}}, align 4, !tbaa
 
 ; CHECK-NOTBAA: ret i32 0
diff --git a/test/Transforms/LoopVectorize/undef-inst-bug.ll b/test/Transforms/LoopVectorize/undef-inst-bug.ll
index 0444fe894ed4..e9d053cf0bef 100644
--- a/test/Transforms/LoopVectorize/undef-inst-bug.ll
+++ b/test/Transforms/LoopVectorize/undef-inst-bug.ll
@@ -24,7 +24,7 @@ for.body:
   ; Loop invariant anchored in loop.
   %idxprom21 = zext i32 undef to i64
 
-  %arrayidx23 = getelementptr inbounds [100 x [100 x i32]]* undef, i64 0, i64 %idxprom21, i64 %indvars.iv17
+  %arrayidx23 = getelementptr inbounds [100 x [100 x i32]], [100 x [100 x i32]]* undef, i64 0, i64 %idxprom21, i64 %indvars.iv17
   store i32 undef, i32* %arrayidx23, align 4
   %indvars.next= add i64 %indvars.iv17, -1
   %0 = trunc i64 %indvars.next to i32
diff --git a/test/Transforms/LoopVectorize/unroll.ll b/test/Transforms/LoopVectorize/unroll.ll
new file mode 100644
index 000000000000..74076f695104
--- /dev/null
+++ b/test/Transforms/LoopVectorize/unroll.ll
@@ -0,0 +1,37 @@
+; This test makes sure that loop will not be unrolled in vectorization if VF computed
+; equals to 1.
+; RUN: opt < %s -loop-vectorize -S | FileCheck %s
+
+; Make sure there are no geps being merged.
+; CHECK-LABEL: @foo(
+; CHECK: getelementptr
+; CHECK-NOT: getelementptr
+
+@N = common global i32 0, align 4
+@a = common global [1000 x i32] zeroinitializer, align 16
+
+define void @foo() #0 {
+entry:
+  %0 = load i32, i32* @N, align 4
+  %cmp5 = icmp sgt i32 %0, 0
+  br i1 %cmp5, label %for.body.lr.ph, label %for.end
+
+for.body.lr.ph:                                   ; preds = %entry
+  %conv = sext i32 %0 to i64
+  br label %for.body
+
+for.body:                                         ; preds = %for.body.lr.ph, %for.body
+  %i.06 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
+  %mul = mul nuw nsw i64 %i.06, 7
+  %arrayidx = getelementptr inbounds [1000 x i32], [1000 x i32]* @a, i64 0, i64 %mul
+  store i32 3, i32* %arrayidx, align 4
+  %inc = add nuw nsw i64 %i.06, 1
+  %cmp = icmp slt i64 %inc, %conv
+  br i1 %cmp, label %for.body, label %for.end.loopexit
+
+for.end.loopexit:                                 ; preds = %for.body
+  br label %for.end
+
+for.end:                                          ; preds = %for.end.loopexit, %entry
+  ret void
+}
diff --git a/test/Transforms/LoopVectorize/unroll_novec.ll b/test/Transforms/LoopVectorize/unroll_novec.ll
index 257b4e63a915..c23ad77a31f8 100644
--- a/test/Transforms/LoopVectorize/unroll_novec.ll
+++ b/test/Transforms/LoopVectorize/unroll_novec.ll
@@ -10,11 +10,11 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;    a[i] += i;
 ;  }
 ;CHECK-LABEL: @inc(
-;CHECK: load i32*
-;CHECK: load i32*
-;CHECK: load i32*
-;CHECK: load i32*
-;CHECK-NOT: load i32*
+;CHECK: load i32, i32*
+;CHECK: load i32, i32*
+;CHECK: load i32, i32*
+;CHECK: load i32, i32*
+;CHECK-NOT: load i32, i32*
 ;CHECK: add nsw i32
 ;CHECK: add nsw i32
 ;CHECK: add nsw i32
@@ -33,8 +33,8 @@ define void @inc(i32 %n) nounwind uwtable noinline ssp {
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds [2048 x i32]* @a, i64 0, i64 %indvars.iv
-  %3 = load i32* %2, align 4
+  %2 = getelementptr inbounds [2048 x i32], [2048 x i32]* @a, i64 0, i64 %indvars.iv
+  %3 = load i32, i32* %2, align 4
   %4 = trunc i64 %indvars.iv to i32
   %5 = add nsw i32 %3, %4
   store i32 %5, i32* %2, align 4
diff --git a/test/Transforms/LoopVectorize/unsized-pointee-crash.ll b/test/Transforms/LoopVectorize/unsized-pointee-crash.ll
index 5cc98378b952..7763c9c68420 100644
--- a/test/Transforms/LoopVectorize/unsized-pointee-crash.ll
+++ b/test/Transforms/LoopVectorize/unsized-pointee-crash.ll
@@ -11,10 +11,10 @@ for.body:
   %b.05 = phi i32 (...)* [ undef, %entry ], [ %1, %for.body ]
   %a.04 = phi i32 [ 0, %entry ], [ %inc, %for.body ]
   %0 = bitcast i32 (...)* %b.05 to i8*
-  %add.ptr = getelementptr i8* %0, i64 1
+  %add.ptr = getelementptr i8, i8* %0, i64 1
   %1 = bitcast i8* %add.ptr to i32 (...)*
 ; CHECK:      %[[cst:.*]] = bitcast i32 (...)* {{.*}} to i8*
-; CHECK-NEXT: %[[gep:.*]] = getelementptr i8* %[[cst]], i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr i8, i8* %[[cst]], i64 1
   %inc = add nsw i32 %a.04, 1
   %exitcond = icmp eq i32 %a.04, 63
   br i1 %exitcond, label %for.end, label %for.body
diff --git a/test/Transforms/LoopVectorize/value-ptr-bug.ll b/test/Transforms/LoopVectorize/value-ptr-bug.ll
index 7fb9095b7c10..ce4601f7b92d 100644
--- a/test/Transforms/LoopVectorize/value-ptr-bug.ll
+++ b/test/Transforms/LoopVectorize/value-ptr-bug.ll
@@ -19,8 +19,8 @@ entry:
 loop:
   %inc = phi i32 [ %sub267, %entry ], [ %add, %loop]
   %ext.inc = sext i32 %inc to i64
-  %add.ptr265 = getelementptr inbounds i32* %ptr265, i64 %ext.inc
-  %add.ptr266 = getelementptr inbounds i32* %ptr266, i64 %ext.inc
+  %add.ptr265 = getelementptr inbounds i32, i32* %ptr265, i64 %ext.inc
+  %add.ptr266 = getelementptr inbounds i32, i32* %ptr266, i64 %ext.inc
   %add = add i32 %inc, 9
   %cmp = icmp slt i32 %add, 140
   br i1 %cmp, label %block1, label %loop
@@ -37,9 +37,9 @@ do.body272:
   %row_width.5 = phi i32 [ %sub267.lcssa, %block1 ], [ %dec, %do.body272 ]
   %sp.4 = phi i8* [ %tmp30, %block1 ], [ %incdec.ptr273, %do.body272 ]
   %dp.addr.4 = phi i8* [ %tmp29, %block1 ], [ %incdec.ptr274, %do.body272 ]
-  %incdec.ptr273 = getelementptr inbounds i8* %sp.4, i64 1
-  %tmp31 = load i8* %sp.4, align 1
-  %incdec.ptr274 = getelementptr inbounds i8* %dp.addr.4, i64 1
+  %incdec.ptr273 = getelementptr inbounds i8, i8* %sp.4, i64 1
+  %tmp31 = load i8, i8* %sp.4, align 1
+  %incdec.ptr274 = getelementptr inbounds i8, i8* %dp.addr.4, i64 1
   store i8 %tmp31, i8* %dp.addr.4, align 1
   %dec = add i32 %row_width.5, -1
   %cmp276 = icmp eq i32 %dec, 0
diff --git a/test/Transforms/LoopVectorize/vect.omp.persistence.ll b/test/Transforms/LoopVectorize/vect.omp.persistence.ll
index ea7be9c106b3..0d7f8b9398f6 100644
--- a/test/Transforms/LoopVectorize/vect.omp.persistence.ll
+++ b/test/Transforms/LoopVectorize/vect.omp.persistence.ll
@@ -45,13 +45,13 @@ for.header:
 for.body:
 
   %0 = add nsw i64 %indvars.iv, -5
-  %arrayidx = getelementptr inbounds float* %a, i64 %0
-  %1 = load float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %0
+  %1 = load float, float* %arrayidx, align 4, !llvm.mem.parallel_loop_access !1
   %2 = add nsw i64 %indvars.iv, 2
-  %arrayidx2 = getelementptr inbounds float* %a, i64 %2
-  %3 = load float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
+  %arrayidx2 = getelementptr inbounds float, float* %a, i64 %2
+  %3 = load float, float* %arrayidx2, align 4, !llvm.mem.parallel_loop_access !1
   %mul = fmul float %1, %3
-  %arrayidx4 = getelementptr inbounds float* %a, i64 %indvars.iv
+  %arrayidx4 = getelementptr inbounds float, float* %a, i64 %indvars.iv
   store float %mul, float* %arrayidx4, align 4, !llvm.mem.parallel_loop_access !1
 
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
diff --git a/test/Transforms/LoopVectorize/vect.stats.ll b/test/Transforms/LoopVectorize/vect.stats.ll
index c5b6e64bc9fb..7bf5a6084847 100644
--- a/test/Transforms/LoopVectorize/vect.stats.ll
+++ b/test/Transforms/LoopVectorize/vect.stats.ll
@@ -20,8 +20,8 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv2 = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds float* %a, i64 %indvars.iv2
-  %0 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %indvars.iv2
+  %0 = load float, float* %arrayidx, align 4
   %mul = fmul float %0, %0
   store float %mul, float* %arrayidx, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv2, 1
@@ -42,13 +42,13 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv2 = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
   %0 = add nsw i64 %indvars.iv2, -5
-  %arrayidx = getelementptr inbounds float* %a, i64 %0
-  %1 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %a, i64 %0
+  %1 = load float, float* %arrayidx, align 4
   %2 = add nsw i64 %indvars.iv2, 2
-  %arrayidx2 = getelementptr inbounds float* %a, i64 %2
-  %3 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %a, i64 %2
+  %3 = load float, float* %arrayidx2, align 4
   %mul = fmul float %1, %3
-  %arrayidx4 = getelementptr inbounds float* %a, i64 %indvars.iv2
+  %arrayidx4 = getelementptr inbounds float, float* %a, i64 %indvars.iv2
   store float %mul, float* %arrayidx4, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv2, 1
   %cmp2 = icmp sgt i64 %indvars.iv.next, %size
diff --git a/test/Transforms/LoopVectorize/vectorize-once.ll b/test/Transforms/LoopVectorize/vectorize-once.ll
index a9b2a53f74b3..9a9125039966 100644
--- a/test/Transforms/LoopVectorize/vectorize-once.ll
+++ b/test/Transforms/LoopVectorize/vectorize-once.ll
@@ -22,16 +22,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 define i32 @_Z4foo1Pii(i32* %A, i32 %n) #0 {
 entry:
   %idx.ext = sext i32 %n to i64
-  %add.ptr = getelementptr inbounds i32* %A, i64 %idx.ext
+  %add.ptr = getelementptr inbounds i32, i32* %A, i64 %idx.ext
   %cmp3.i = icmp eq i32 %n, 0
   br i1 %cmp3.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
 
 for.body.i:                                       ; preds = %entry, %for.body.i
   %__init.addr.05.i = phi i32 [ %add.i, %for.body.i ], [ 0, %entry ]
   %__first.addr.04.i = phi i32* [ %incdec.ptr.i, %for.body.i ], [ %A, %entry ]
-  %0 = load i32* %__first.addr.04.i, align 4
+  %0 = load i32, i32* %__first.addr.04.i, align 4
   %add.i = add nsw i32 %0, %__init.addr.05.i
-  %incdec.ptr.i = getelementptr inbounds i32* %__first.addr.04.i, i64 1
+  %incdec.ptr.i = getelementptr inbounds i32, i32* %__first.addr.04.i, i64 1
   %cmp.i = icmp eq i32* %incdec.ptr.i, %add.ptr
   br i1 %cmp.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
 
@@ -48,16 +48,16 @@ _ZSt10accumulateIPiiET0_T_S2_S1_.exit:            ; preds = %for.body.i, %entry
 define i32 @_Z4foo2Pii(i32* %A, i32 %n) #0 {
 entry:
   %idx.ext = sext i32 %n to i64
-  %add.ptr = getelementptr inbounds i32* %A, i64 %idx.ext
+  %add.ptr = getelementptr inbounds i32, i32* %A, i64 %idx.ext
   %cmp3.i = icmp eq i32 %n, 0
   br i1 %cmp3.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i
 
 for.body.i:                                       ; preds = %entry, %for.body.i
   %__init.addr.05.i = phi i32 [ %add.i, %for.body.i ], [ 0, %entry ]
   %__first.addr.04.i = phi i32* [ %incdec.ptr.i, %for.body.i ], [ %A, %entry ]
-  %0 = load i32* %__first.addr.04.i, align 4
+  %0 = load i32, i32* %__first.addr.04.i, align 4
   %add.i = add nsw i32 %0, %__init.addr.05.i
-  %incdec.ptr.i = getelementptr inbounds i32* %__first.addr.04.i, i64 1
+  %incdec.ptr.i = getelementptr inbounds i32, i32* %__first.addr.04.i, i64 1
   %cmp.i = icmp eq i32* %incdec.ptr.i, %add.ptr
   br i1 %cmp.i, label %_ZSt10accumulateIPiiET0_T_S2_S1_.exit, label %for.body.i, !llvm.loop !0
 
@@ -71,7 +71,8 @@ attributes #0 = { nounwind readonly ssp uwtable "fp-contract-model"="standard" "
 ; CHECK: !0 = distinct !{!0, !1, !2}
 ; CHECK: !1 = !{!"llvm.loop.vectorize.width", i32 1}
 ; CHECK: !2 = !{!"llvm.loop.interleave.count", i32 1}
-; CHECK: !3 = distinct !{!3, !1, !2}
+; CHECK: !3 = distinct !{!3, !4, !1, !2}
+; CHECK: !4 = !{!"llvm.loop.unroll.runtime.disable"}
 
 !0 = !{!0, !1}
 !1 = !{!"llvm.loop.vectorize.width", i32 1}
diff --git a/test/Transforms/LoopVectorize/version-mem-access.ll b/test/Transforms/LoopVectorize/version-mem-access.ll
index 7ac2fca2533f..a9d319e5a2dd 100644
--- a/test/Transforms/LoopVectorize/version-mem-access.ll
+++ b/test/Transforms/LoopVectorize/version-mem-access.ll
@@ -2,10 +2,16 @@
 
 target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 
+; Check that we version this loop with speculating the value 1 for symbolic
+; strides.  This also checks that the symbolic stride information is correctly
+; propagated to the memcheck generation.  Without this the loop wouldn't
+; vectorize because we couldn't determine the array bounds for the required
+; memchecks.
+
 ; CHECK-LABEL: test
-define void @test(i32* noalias %A, i64 %AStride,
-                  i32* noalias %B, i32 %BStride,
-                  i32* noalias %C, i64 %CStride, i32 %N) {
+define void @test(i32*  %A, i64 %AStride,
+                  i32*  %B, i32 %BStride,
+                  i32*  %C, i64 %CStride, i32 %N) {
 entry:
   %cmp13 = icmp eq i32 %N, 0
   br i1 %cmp13, label %for.end, label %for.body.preheader
@@ -28,14 +34,14 @@ for.body:
   %iv.trunc = trunc i64 %indvars.iv to i32
   %mul = mul i32 %iv.trunc, %BStride
   %mul64 = zext i32 %mul to i64
-  %arrayidx = getelementptr inbounds i32* %B, i64 %mul64
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 %mul64
+  %0 = load i32, i32* %arrayidx, align 4
   %mul2 = mul nsw i64 %indvars.iv, %CStride
-  %arrayidx3 = getelementptr inbounds i32* %C, i64 %mul2
-  %1 = load i32* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %C, i64 %mul2
+  %1 = load i32, i32* %arrayidx3, align 4
   %mul4 = mul nsw i32 %1, %0
   %mul3 = mul nsw i64 %indvars.iv, %AStride
-  %arrayidx7 = getelementptr inbounds i32* %A, i64 %mul3
+  %arrayidx7 = getelementptr inbounds i32, i32* %A, i64 %mul3
   store i32 %mul4, i32* %arrayidx7, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
@@ -70,9 +76,9 @@ for.body:
   %0 = trunc i64 %indvars.iv to i32
   %mul = mul nsw i32 %0, %conv
   %idxprom = sext i32 %mul to i64
-  %arrayidx = getelementptr inbounds double* %x, i64 %idxprom
-  %1 = load double* %arrayidx, align 8
-  %arrayidx3 = getelementptr inbounds double* %c, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds double, double* %x, i64 %idxprom
+  %1 = load double, double* %arrayidx, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %c, i64 %indvars.iv
   store double %1, double* %arrayidx3, align 8
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
   %lftr.wideiv = trunc i64 %indvars.iv.next to i32
diff --git a/test/Transforms/LoopVectorize/write-only.ll b/test/Transforms/LoopVectorize/write-only.ll
index 2f100de82370..b2bc0458a23a 100644
--- a/test/Transforms/LoopVectorize/write-only.ll
+++ b/test/Transforms/LoopVectorize/write-only.ll
@@ -12,8 +12,8 @@ define i32 @read_mod_write_single_ptr(float* nocapture %a, i32 %n) nounwind uwta
 
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %indvars.iv = phi i64 [ %indvars.iv.next, %.lr.ph ], [ 0, %0 ]
-  %2 = getelementptr inbounds float* %a, i64 %indvars.iv
-  %3 = load float* %2, align 4
+  %2 = getelementptr inbounds float, float* %a, i64 %indvars.iv
+  %3 = load float, float* %2, align 4
   %4 = fmul float %3, 3.000000e+00
   store float %4, float* %2, align 4
   %indvars.iv.next = add i64 %indvars.iv, 1
diff --git a/test/Transforms/LowerAtomic/atomic-swap.ll b/test/Transforms/LowerAtomic/atomic-swap.ll
index cb1124136f86..77000527a11f 100644
--- a/test/Transforms/LowerAtomic/atomic-swap.ll
+++ b/test/Transforms/LowerAtomic/atomic-swap.ll
@@ -5,7 +5,7 @@ define i8 @cmpswap() {
   %i = alloca i8
   %pair = cmpxchg i8* %i, i8 0, i8 42 monotonic monotonic
   %j = extractvalue { i8, i1 } %pair, 0
-; CHECK: [[OLDVAL:%[a-z0-9]+]] = load i8* [[ADDR:%[a-z0-9]+]]
+; CHECK: [[OLDVAL:%[a-z0-9]+]] = load i8, i8* [[ADDR:%[a-z0-9]+]]
 ; CHECK-NEXT: [[SAME:%[a-z0-9]+]] = icmp eq i8 [[OLDVAL]], 0
 ; CHECK-NEXT: [[TO_STORE:%[a-z0-9]+]] = select i1 [[SAME]], i8 42, i8 [[OLDVAL]]
 ; CHECK-NEXT: store i8 [[TO_STORE]], i8* [[ADDR]]
diff --git a/test/Transforms/LowerBitSets/constant.ll b/test/Transforms/LowerBitSets/constant.ll
new file mode 100644
index 000000000000..99c925914f72
--- /dev/null
+++ b/test/Transforms/LowerBitSets/constant.ll
@@ -0,0 +1,34 @@
+; RUN: opt -S -lowerbitsets < %s | FileCheck %s
+
+target datalayout = "e-p:32:32"
+
+@a = constant i32 1
+@b = constant [2 x i32] [i32 2, i32 3]
+
+!0 = !{!"bitset1", i32* @a, i32 0}
+!1 = !{!"bitset1", [2 x i32]* @b, i32 4}
+
+!llvm.bitsets = !{ !0, !1 }
+
+declare i1 @llvm.bitset.test(i8* %ptr, metadata %bitset) nounwind readnone
+
+; CHECK: @foo(
+define i1 @foo() {
+  ; CHECK: ret i1 true
+  %x = call i1 @llvm.bitset.test(i8* bitcast (i32* @a to i8*), metadata !"bitset1")
+  ret i1 %x
+}
+
+; CHECK: @bar(
+define i1 @bar() {
+  ; CHECK: ret i1 true
+  %x = call i1 @llvm.bitset.test(i8* bitcast (i32* getelementptr ([2 x i32], [2 x i32]* @b, i32 0, i32 1) to i8*), metadata !"bitset1")
+  ret i1 %x
+}
+
+; CHECK: @baz(
+define i1 @baz() {
+  ; CHECK-NOT: ret i1 true
+  %x = call i1 @llvm.bitset.test(i8* bitcast (i32* getelementptr ([2 x i32], [2 x i32]* @b, i32 0, i32 0) to i8*), metadata !"bitset1")
+  ret i1 %x
+}
diff --git a/test/Transforms/LowerBitSets/layout.ll b/test/Transforms/LowerBitSets/layout.ll
new file mode 100644
index 000000000000..a0c6e77a57fe
--- /dev/null
+++ b/test/Transforms/LowerBitSets/layout.ll
@@ -0,0 +1,35 @@
+; RUN: opt -S -lowerbitsets < %s | FileCheck %s
+
+target datalayout = "e-p:32:32"
+
+; Tests that this set of globals is laid out according to our layout algorithm
+; (see GlobalLayoutBuilder in include/llvm/Transforms/IPO/LowerBitSets.h).
+; The chosen layout in this case is a, e, b, d, c.
+
+; CHECK: private constant { i32, [0 x i8], i32, [0 x i8], i32, [0 x i8], i32, [0 x i8], i32 } { i32 1, [0 x i8] zeroinitializer, i32 5, [0 x i8] zeroinitializer, i32 2, [0 x i8] zeroinitializer, i32 4, [0 x i8] zeroinitializer, i32 3 }
+@a = constant i32 1
+@b = constant i32 2
+@c = constant i32 3
+@d = constant i32 4
+@e = constant i32 5
+
+!0 = !{!"bitset1", i32* @a, i32 0}
+!1 = !{!"bitset1", i32* @b, i32 0}
+!2 = !{!"bitset1", i32* @c, i32 0}
+
+!3 = !{!"bitset2", i32* @b, i32 0}
+!4 = !{!"bitset2", i32* @d, i32 0}
+
+!5 = !{!"bitset3", i32* @a, i32 0}
+!6 = !{!"bitset3", i32* @e, i32 0}
+
+!llvm.bitsets = !{ !0, !1, !2, !3, !4, !5, !6 }
+
+declare i1 @llvm.bitset.test(i8* %ptr, metadata %bitset) nounwind readnone
+
+define void @foo() {
+  %x = call i1 @llvm.bitset.test(i8* undef, metadata !"bitset1")
+  %y = call i1 @llvm.bitset.test(i8* undef, metadata !"bitset2")
+  %z = call i1 @llvm.bitset.test(i8* undef, metadata !"bitset3")
+  ret void
+}
diff --git a/test/Transforms/LowerBitSets/simple.ll b/test/Transforms/LowerBitSets/simple.ll
new file mode 100644
index 000000000000..0fcdf0b36d63
--- /dev/null
+++ b/test/Transforms/LowerBitSets/simple.ll
@@ -0,0 +1,137 @@
+; RUN: opt -S -lowerbitsets < %s | FileCheck %s
+; RUN: opt -S -lowerbitsets -mtriple=x86_64-apple-macosx10.8.0 < %s | FileCheck -check-prefix=CHECK-DARWIN %s
+; RUN: opt -S -O3 < %s | FileCheck -check-prefix=CHECK-NODISCARD %s
+
+target datalayout = "e-p:32:32"
+
+; CHECK: [[G:@[^ ]*]] = private constant { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] } { i32 1, [0 x i8] zeroinitializer, [63 x i32] zeroinitializer, [4 x i8] zeroinitializer, i32 3, [0 x i8] zeroinitializer, [2 x i32] [i32 4, i32 5] }
+@a = constant i32 1
+@b = constant [63 x i32] zeroinitializer
+@c = constant i32 3
+@d = constant [2 x i32] [i32 4, i32 5]
+
+; CHECK: [[BA:@[^ ]*]] = private constant [68 x i8] c"\03\01\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\00\02\00\01"
+
+; Offset 0, 4 byte alignment
+!0 = !{!"bitset1", i32* @a, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset1", i32* @a, i32 0}
+!1 = !{!"bitset1", [63 x i32]* @b, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset1", [63 x i32]* @b, i32 0}
+!2 = !{!"bitset1", [2 x i32]* @d, i32 4}
+; CHECK-NODISCARD-DAG: !{!"bitset1", [2 x i32]* @d, i32 4}
+
+; Offset 4, 256 byte alignment
+!3 = !{!"bitset2", [63 x i32]* @b, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset2", [63 x i32]* @b, i32 0}
+!4 = !{!"bitset2", i32* @c, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset2", i32* @c, i32 0}
+
+; Offset 0, 4 byte alignment
+!5 = !{!"bitset3", i32* @a, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset3", i32* @a, i32 0}
+!6 = !{!"bitset3", i32* @c, i32 0}
+; CHECK-NODISCARD-DAG: !{!"bitset3", i32* @c, i32 0}
+
+; Entries whose second operand is null (the result of a global being DCE'd)
+; should be ignored.
+!7 = !{!"bitset2", null, i32 0}
+
+!llvm.bitsets = !{ !0, !1, !2, !3, !4, !5, !6, !7 }
+
+; CHECK: @bits_use{{[0-9]*}} = private alias i8* @bits{{[0-9]*}}
+; CHECK: @bits_use.{{[0-9]*}} = private alias i8* @bits{{[0-9]*}}
+; CHECK: @bits_use.{{[0-9]*}} = private alias i8* @bits{{[0-9]*}}
+
+; CHECK: @a = alias getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 0)
+; CHECK: @b = alias getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 2)
+; CHECK: @c = alias getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 4)
+; CHECK: @d = alias getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 6)
+
+; CHECK-DARWIN: @aptr = constant i32* getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G:@[^ ]*]], i32 0, i32 0)
+@aptr = constant i32* @a
+
+; CHECK-DARWIN: @bptr = constant [63 x i32]* getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 2)
+@bptr = constant [63 x i32]* @b
+
+; CHECK-DARWIN: @cptr = constant i32* getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 4)
+@cptr = constant i32* @c
+
+; CHECK-DARWIN: @dptr = constant [2 x i32]* getelementptr inbounds ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }, { i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]], i32 0, i32 6)
+@dptr = constant [2 x i32]* @d
+
+; CHECK-DARWIN: [[G]] = private constant
+
+; CHECK: @bits{{[0-9]*}} = private alias getelementptr inbounds ([68 x i8], [68 x i8]* [[BA]], i32 0, i32 0)
+; CHECK: @bits.{{[0-9]*}} = private alias getelementptr inbounds ([68 x i8], [68 x i8]* [[BA]], i32 0, i32 0)
+
+declare i1 @llvm.bitset.test(i8* %ptr, metadata %bitset) nounwind readnone
+
+; CHECK: @foo(i32* [[A0:%[^ ]*]])
+define i1 @foo(i32* %p) {
+  ; CHECK-NOT: llvm.bitset.test
+
+  ; CHECK: [[R0:%[^ ]*]] = bitcast i32* [[A0]] to i8*
+  %pi8 = bitcast i32* %p to i8*
+  ; CHECK: [[R1:%[^ ]*]] = ptrtoint i8* [[R0]] to i32
+  ; CHECK: [[R2:%[^ ]*]] = sub i32 [[R1]], ptrtoint ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]] to i32)
+  ; CHECK: [[R3:%[^ ]*]] = lshr i32 [[R2]], 2
+  ; CHECK: [[R4:%[^ ]*]] = shl i32 [[R2]], 30
+  ; CHECK: [[R5:%[^ ]*]] = or i32 [[R3]], [[R4]]
+  ; CHECK: [[R6:%[^ ]*]] = icmp ult i32 [[R5]], 68
+  ; CHECK: br i1 [[R6]]
+
+  ; CHECK: [[R8:%[^ ]*]] = getelementptr i8, i8* @bits_use.{{[0-9]*}}, i32 [[R5]]
+  ; CHECK: [[R9:%[^ ]*]] = load i8, i8* [[R8]]
+  ; CHECK: [[R10:%[^ ]*]] = and i8 [[R9]], 1
+  ; CHECK: [[R11:%[^ ]*]] = icmp ne i8 [[R10]], 0
+
+  ; CHECK: [[R16:%[^ ]*]] = phi i1 [ false, {{%[^ ]*}} ], [ [[R11]], {{%[^ ]*}} ]
+  %x = call i1 @llvm.bitset.test(i8* %pi8, metadata !"bitset1")
+
+  ; CHECK-NOT: llvm.bitset.test
+  %y = call i1 @llvm.bitset.test(i8* %pi8, metadata !"bitset1")
+
+  ; CHECK: ret i1 [[R16]]
+  ret i1 %x
+}
+
+; CHECK: @bar(i32* [[B0:%[^ ]*]])
+define i1 @bar(i32* %p) {
+  ; CHECK: [[S0:%[^ ]*]] = bitcast i32* [[B0]] to i8*
+  %pi8 = bitcast i32* %p to i8*
+  ; CHECK: [[S1:%[^ ]*]] = ptrtoint i8* [[S0]] to i32
+  ; CHECK: [[S2:%[^ ]*]] = sub i32 [[S1]], add (i32 ptrtoint ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]] to i32), i32 4)
+  ; CHECK: [[S3:%[^ ]*]] = lshr i32 [[S2]], 8
+  ; CHECK: [[S4:%[^ ]*]] = shl i32 [[S2]], 24
+  ; CHECK: [[S5:%[^ ]*]] = or i32 [[S3]], [[S4]]
+  ; CHECK: [[S6:%[^ ]*]] = icmp ult i32 [[S5]], 2
+  %x = call i1 @llvm.bitset.test(i8* %pi8, metadata !"bitset2")
+
+  ; CHECK: ret i1 [[S6]]
+  ret i1 %x
+}
+
+; CHECK: @baz(i32* [[C0:%[^ ]*]])
+define i1 @baz(i32* %p) {
+  ; CHECK: [[T0:%[^ ]*]] = bitcast i32* [[C0]] to i8*
+  %pi8 = bitcast i32* %p to i8*
+  ; CHECK: [[T1:%[^ ]*]] = ptrtoint i8* [[T0]] to i32
+  ; CHECK: [[T2:%[^ ]*]] = sub i32 [[T1]], ptrtoint ({ i32, [0 x i8], [63 x i32], [4 x i8], i32, [0 x i8], [2 x i32] }* [[G]] to i32)
+  ; CHECK: [[T3:%[^ ]*]] = lshr i32 [[T2]], 2
+  ; CHECK: [[T4:%[^ ]*]] = shl i32 [[T2]], 30
+  ; CHECK: [[T5:%[^ ]*]] = or i32 [[T3]], [[T4]]
+  ; CHECK: [[T6:%[^ ]*]] = icmp ult i32 [[T5]], 66
+  ; CHECK: br i1 [[T6]]
+
+  ; CHECK: [[T8:%[^ ]*]] = getelementptr i8, i8* @bits_use.{{[0-9]*}}, i32 [[T5]]
+  ; CHECK: [[T9:%[^ ]*]] = load i8, i8* [[T8]]
+  ; CHECK: [[T10:%[^ ]*]] = and i8 [[T9]], 2
+  ; CHECK: [[T11:%[^ ]*]] = icmp ne i8 [[T10]], 0
+
+  ; CHECK: [[T16:%[^ ]*]] = phi i1 [ false, {{%[^ ]*}} ], [ [[T11]], {{%[^ ]*}} ]
+  %x = call i1 @llvm.bitset.test(i8* %pi8, metadata !"bitset3")
+  ; CHECK: ret i1 [[T16]]
+  ret i1 %x
+}
+
+; CHECK-NOT: !llvm.bitsets
diff --git a/test/Transforms/LowerBitSets/single-offset.ll b/test/Transforms/LowerBitSets/single-offset.ll
new file mode 100644
index 000000000000..57194f42e096
--- /dev/null
+++ b/test/Transforms/LowerBitSets/single-offset.ll
@@ -0,0 +1,40 @@
+; RUN: opt -S -lowerbitsets < %s | FileCheck %s
+
+target datalayout = "e-p:32:32"
+
+; CHECK: [[G:@[^ ]*]] = private constant { i32, [0 x i8], i32 }
+@a = constant i32 1
+@b = constant i32 2
+
+!0 = !{!"bitset1", i32* @a, i32 0}
+!1 = !{!"bitset1", i32* @b, i32 0}
+!2 = !{!"bitset2", i32* @a, i32 0}
+!3 = !{!"bitset3", i32* @b, i32 0}
+
+!llvm.bitsets = !{ !0, !1, !2, !3 }
+
+declare i1 @llvm.bitset.test(i8* %ptr, metadata %bitset) nounwind readnone
+
+; CHECK: @foo(i8* [[A0:%[^ ]*]])
+define i1 @foo(i8* %p) {
+  ; CHECK: [[R0:%[^ ]*]] = ptrtoint i8* [[A0]] to i32
+  ; CHECK: [[R1:%[^ ]*]] = icmp eq i32 [[R0]], ptrtoint ({ i32, [0 x i8], i32 }* [[G]] to i32)
+  %x = call i1 @llvm.bitset.test(i8* %p, metadata !"bitset2")
+  ; CHECK: ret i1 [[R1]]
+  ret i1 %x
+}
+
+; CHECK: @bar(i8* [[B0:%[^ ]*]])
+define i1 @bar(i8* %p) {
+  ; CHECK: [[S0:%[^ ]*]] = ptrtoint i8* [[B0]] to i32
+  ; CHECK: [[S1:%[^ ]*]] = icmp eq i32 [[S0]], add (i32 ptrtoint ({ i32, [0 x i8], i32 }* [[G]] to i32), i32 4)
+  %x = call i1 @llvm.bitset.test(i8* %p, metadata !"bitset3")
+  ; CHECK: ret i1 [[S1]]
+  ret i1 %x
+}
+
+; CHECK: @x(
+define i1 @x(i8* %p) {
+  %x = call i1 @llvm.bitset.test(i8* %p, metadata !"bitset1")
+  ret i1 %x
+}
diff --git a/test/Transforms/LowerExpectIntrinsic/basic.ll b/test/Transforms/LowerExpectIntrinsic/basic.ll
index b15edc715141..73d9f44ee7e5 100644
--- a/test/Transforms/LowerExpectIntrinsic/basic.ll
+++ b/test/Transforms/LowerExpectIntrinsic/basic.ll
@@ -1,4 +1,5 @@
 ; RUN: opt -lower-expect -strip-dead-prototypes -S -o - < %s | FileCheck %s
+; RUN: opt -S -passes=lower-expect < %s | opt -strip-dead-prototypes -S | FileCheck %s
 
 ; CHECK-LABEL: @test1(
 define i32 @test1(i32 %x) nounwind uwtable ssp {
@@ -6,7 +7,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %cmp = icmp sgt i32 %tmp, 1
   %conv = zext i1 %cmp to i32
   %conv1 = sext i32 %conv to i64
@@ -17,7 +18,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -26,7 +27,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -40,7 +41,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %conv = sext i32 %tmp to i64
   %expval = call i64 @llvm.expect.i64(i64 %conv, i64 1)
   %tobool = icmp ne i64 %expval, 0
@@ -49,7 +50,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -58,7 +59,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -68,7 +69,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %tobool = icmp ne i32 %tmp, 0
   %lnot = xor i1 %tobool, true
   %lnot.ext = zext i1 %lnot to i32
@@ -80,7 +81,7 @@ entry:
   br i1 %tobool1, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -89,7 +90,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -99,7 +100,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %tobool = icmp ne i32 %tmp, 0
   %lnot = xor i1 %tobool, true
   %lnot1 = xor i1 %lnot, true
@@ -112,7 +113,7 @@ entry:
   br i1 %tobool2, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -121,7 +122,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -131,7 +132,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %cmp = icmp slt i32 %tmp, 0
   %conv = zext i1 %cmp to i32
   %conv1 = sext i32 %conv to i64
@@ -142,7 +143,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -151,7 +152,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -161,7 +162,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %conv = sext i32 %tmp to i64
   %expval = call i64 @llvm.expect.i64(i64 %conv, i64 1)
 ; CHECK: !prof !2
@@ -180,7 +181,7 @@ sw.epilog:                                        ; preds = %entry
   br label %return
 
 return:                                           ; preds = %sw.epilog, %sw.bb
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -190,7 +191,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %conv = sext i32 %tmp to i64
   %expval = call i64 @llvm.expect.i64(i64 %conv, i64 1)
 ; CHECK: !prof !3
@@ -201,7 +202,7 @@ entry:
   ]
 
 sw.bb:                                            ; preds = %entry, %entry
-  %tmp1 = load i32* %x.addr, align 4
+  %tmp1 = load i32, i32* %x.addr, align 4
   store i32 %tmp1, i32* %retval
   br label %return
 
@@ -210,7 +211,7 @@ sw.epilog:                                        ; preds = %entry
   br label %return
 
 return:                                           ; preds = %sw.epilog, %sw.bb
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -220,7 +221,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %cmp = icmp sgt i32 %tmp, 1
   %conv = zext i1 %cmp to i32
   %expval = call i32 @llvm.expect.i32(i32 %conv, i32 1)
@@ -230,7 +231,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -239,7 +240,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -251,7 +252,7 @@ entry:
   %retval = alloca i32, align 4
   %x.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
-  %tmp = load i32* %x.addr, align 4
+  %tmp = load i32, i32* %x.addr, align 4
   %cmp = icmp sgt i32 %tmp, 1
   %expval = call i1 @llvm.expect.i1(i1 %cmp, i1 1)
 ; CHECK: !prof !0
@@ -259,7 +260,7 @@ entry:
   br i1 %expval, label %if.then, label %if.end
 
 if.then:                                          ; preds = %entry
-  %call = call i32 (...)* @f()
+  %call = call i32 (...) @f()
   store i32 %call, i32* %retval
   br label %return
 
@@ -268,7 +269,7 @@ if.end:                                           ; preds = %entry
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
diff --git a/test/Transforms/LowerInvoke/2003-12-10-Crash.ll b/test/Transforms/LowerInvoke/2003-12-10-Crash.ll
index 31f3d42225ab..fca8e868018a 100644
--- a/test/Transforms/LowerInvoke/2003-12-10-Crash.ll
+++ b/test/Transforms/LowerInvoke/2003-12-10-Crash.ll
@@ -15,8 +15,11 @@ invoke_cont.0:		; preds = %then
 			to label %try_exit unwind label %try_catch
 try_catch:		; preds = %invoke_cont.0, %then
 	%__tmp.0 = phi i32* [ null, %invoke_cont.0 ], [ null, %then ]		; <i32*> [#uses=0]
+  %res = landingpad { i8* } personality i32 (...)* @__gxx_personality_v0
+          cleanup
 	ret void
 try_exit:		; preds = %invoke_cont.0
 	ret void
 }
 
+declare i32 @__gxx_personality_v0(...)
diff --git a/test/Transforms/LowerSwitch/2014-06-10-SwitchContiguousOpt.ll b/test/Transforms/LowerSwitch/2014-06-10-SwitchContiguousOpt.ll
index 3673c04a8b81..22173b492c4c 100644
--- a/test/Transforms/LowerSwitch/2014-06-10-SwitchContiguousOpt.ll
+++ b/test/Transforms/LowerSwitch/2014-06-10-SwitchContiguousOpt.ll
@@ -6,7 +6,7 @@ entry:
   %retval = alloca i32, align 4
   %a.addr = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
-  %0 = load i32* %a.addr, align 4
+  %0 = load i32, i32* %a.addr, align 4
   switch i32 %0, label %sw.default [
     i32 0, label %sw.bb
     i32 1, label %sw.bb1
diff --git a/test/Transforms/LowerSwitch/2014-06-11-SwitchDefaultUnreachableOpt.ll b/test/Transforms/LowerSwitch/2014-06-11-SwitchDefaultUnreachableOpt.ll
index 0f737211f598..2652a6cc5e09 100644
--- a/test/Transforms/LowerSwitch/2014-06-11-SwitchDefaultUnreachableOpt.ll
+++ b/test/Transforms/LowerSwitch/2014-06-11-SwitchDefaultUnreachableOpt.ll
@@ -1,5 +1,8 @@
 ; RUN: opt < %s -lowerswitch -S | FileCheck %s
-; CHECK-NOT: {{.*}}icmp eq{{.*}}
+;
+; The switch is lowered with a single icmp.
+; CHECK: icmp
+; CHECK-NOT: icmp
 ;
 ;int foo(int a) {
 ;
@@ -14,11 +17,11 @@
 ;
 ;}
 
-define i32 @foo(i32 %a) nounwind ssp uwtable {
+define i32 @foo(i32 %a) {
   %1 = alloca i32, align 4
   %2 = alloca i32, align 4
   store i32 %a, i32* %2, align 4
-  %3 = load i32* %2, align 4
+  %3 = load i32, i32* %2, align 4
   switch i32 %3, label %6 [
     i32 0, label %4
     i32 1, label %5
@@ -36,6 +39,6 @@ define i32 @foo(i32 %a) nounwind ssp uwtable {
   unreachable
 
 ; <label>:7
-  %8 = load i32* %1
+  %8 = load i32, i32* %1
   ret i32 %8
 }
diff --git a/test/Transforms/LowerSwitch/feature.ll b/test/Transforms/LowerSwitch/feature.ll
index 09d25f0b06d4..b82d93455436 100644
--- a/test/Transforms/LowerSwitch/feature.ll
+++ b/test/Transforms/LowerSwitch/feature.ll
@@ -4,49 +4,49 @@
 ; On output we should got binary comparison tree. Check that all is fine.
 
 ;CHECK:     entry:
-;CHECK-NEXT:  br label %NodeBlock19
+;CHECK-NEXT:  br label %NodeBlock.19
 
-;CHECK:     NodeBlock19:                                      ; preds = %entry
-;CHECK-NEXT:  %Pivot20 = icmp slt i32 %tmp158, 10
-;CHECK-NEXT:  br i1 %Pivot20, label %NodeBlock5, label %NodeBlock17
+;CHECK:     NodeBlock.19:                                      ; preds = %entry
+;CHECK-NEXT:  %Pivot.20 = icmp slt i32 %tmp158, 10
+;CHECK-NEXT:  br i1 %Pivot.20, label %NodeBlock.5, label %NodeBlock.17
 
-;CHECK:     NodeBlock17:                                      ; preds = %NodeBlock19
-;CHECK-NEXT:  %Pivot18 = icmp slt i32 %tmp158, 13
-;CHECK-NEXT:  br i1 %Pivot18, label %NodeBlock9, label %NodeBlock15
+;CHECK:     NodeBlock.17:                                      ; preds = %NodeBlock.19
+;CHECK-NEXT:  %Pivot.18 = icmp slt i32 %tmp158, 13
+;CHECK-NEXT:  br i1 %Pivot.18, label %NodeBlock.9, label %NodeBlock.15
 
-;CHECK:     NodeBlock15:                                      ; preds = %NodeBlock17
-;CHECK-NEXT:  %Pivot16 = icmp slt i32 %tmp158, 14
-;CHECK-NEXT:  br i1 %Pivot16, label %bb330, label %NodeBlock13
+;CHECK:     NodeBlock.15:                                      ; preds = %NodeBlock.17
+;CHECK-NEXT:  %Pivot.16 = icmp slt i32 %tmp158, 14
+;CHECK-NEXT:  br i1 %Pivot.16, label %bb330, label %NodeBlock.13
 
-;CHECK:     NodeBlock13:                                      ; preds = %NodeBlock15
-;CHECK-NEXT:  %Pivot14 = icmp slt i32 %tmp158, 15
-;CHECK-NEXT:  br i1 %Pivot14, label %bb332, label %LeafBlock11
+;CHECK:     NodeBlock.13:                                      ; preds = %NodeBlock.15
+;CHECK-NEXT:  %Pivot.14 = icmp slt i32 %tmp158, 15
+;CHECK-NEXT:  br i1 %Pivot.14, label %bb332, label %LeafBlock.11
 
-;CHECK:     LeafBlock11:                                      ; preds = %NodeBlock13
+;CHECK:     LeafBlock.11:                                      ; preds = %NodeBlock.13
 ;CHECK-NEXT:  %SwitchLeaf12 = icmp eq i32 %tmp158, 15
 ;CHECK-NEXT:  br i1 %SwitchLeaf12, label %bb334, label %NewDefault
 
-;CHECK:     NodeBlock9:                                       ; preds = %NodeBlock17
-;CHECK-NEXT:  %Pivot10 = icmp slt i32 %tmp158, 11
-;CHECK-NEXT:  br i1 %Pivot10, label %bb324, label %NodeBlock7
+;CHECK:     NodeBlock.9:                                       ; preds = %NodeBlock.17
+;CHECK-NEXT:  %Pivot.10 = icmp slt i32 %tmp158, 11
+;CHECK-NEXT:  br i1 %Pivot.10, label %bb324, label %NodeBlock.7
 
-;CHECK:     NodeBlock7:                                       ; preds = %NodeBlock9
-;CHECK-NEXT:  %Pivot8 = icmp slt i32 %tmp158, 12
-;CHECK-NEXT:  br i1 %Pivot8, label %bb326, label %bb328
+;CHECK:     NodeBlock.7:                                       ; preds = %NodeBlock.9
+;CHECK-NEXT:  %Pivot.8 = icmp slt i32 %tmp158, 12
+;CHECK-NEXT:  br i1 %Pivot.8, label %bb326, label %bb328
 
-;CHECK:     NodeBlock5:                                       ; preds = %NodeBlock19
-;CHECK-NEXT:  %Pivot6 = icmp slt i32 %tmp158, 7
-;CHECK-NEXT:  br i1 %Pivot6, label %NodeBlock, label %NodeBlock3
+;CHECK:     NodeBlock.5:                                       ; preds = %NodeBlock.19
+;CHECK-NEXT:  %Pivot.6 = icmp slt i32 %tmp158, 7
+;CHECK-NEXT:  br i1 %Pivot.6, label %NodeBlock, label %NodeBlock.3
 
-;CHECK:     NodeBlock3:                                       ; preds = %NodeBlock5
-;CHECK-NEXT:  %Pivot4 = icmp slt i32 %tmp158, 8
-;CHECK-NEXT:  br i1 %Pivot4, label %bb, label %NodeBlock1
+;CHECK:     NodeBlock.3:                                       ; preds = %NodeBlock.5
+;CHECK-NEXT:  %Pivot.4 = icmp slt i32 %tmp158, 8
+;CHECK-NEXT:  br i1 %Pivot.4, label %bb, label %NodeBlock.1
 
-;CHECK:     NodeBlock1:                                       ; preds = %NodeBlock3
-;CHECK-NEXT:  %Pivot2 = icmp slt i32 %tmp158, 9
-;CHECK-NEXT:  br i1 %Pivot2, label %bb338, label %bb322
+;CHECK:     NodeBlock.1:                                       ; preds = %NodeBlock.3
+;CHECK-NEXT:  %Pivot.2 = icmp slt i32 %tmp158, 9
+;CHECK-NEXT:  br i1 %Pivot.2, label %bb338, label %bb322
 
-;CHECK:     NodeBlock:                                        ; preds = %NodeBlock5
+;CHECK:     NodeBlock:                                        ; preds = %NodeBlock.5
 ;CHECK-NEXT:  %Pivot = icmp slt i32 %tmp158, 0
 ;CHECK-NEXT:  br i1 %Pivot, label %LeafBlock, label %bb338
 
diff --git a/test/Transforms/LowerSwitch/fold-popular-case-to-unreachable-default.ll b/test/Transforms/LowerSwitch/fold-popular-case-to-unreachable-default.ll
new file mode 100644
index 000000000000..54929c5bfca3
--- /dev/null
+++ b/test/Transforms/LowerSwitch/fold-popular-case-to-unreachable-default.ll
@@ -0,0 +1,110 @@
+; RUN: opt %s -lowerswitch -S | FileCheck %s
+
+define void @foo(i32 %x, i32* %p) {
+; Cases 2 and 4 are removed and become the new default case.
+; It is now enough to use two icmps to lower the switch.
+;
+; CHECK-LABEL: @foo
+; CHECK:       icmp slt i32 %x, 5
+; CHECK:       icmp eq i32 %x, 1
+; CHECK-NOT:   icmp
+;
+entry:
+  switch i32 %x, label %default [
+    i32 1, label %bb0
+    i32 2, label %popular
+    i32 4, label %popular
+    i32 5, label %bb1
+  ]
+bb0:
+  store i32 0, i32* %p
+  br label %exit
+bb1:
+  store i32 1, i32* %p
+  br label %exit
+popular:
+  store i32 2, i32* %p
+  br label %exit
+exit:
+  ret void
+default:
+  unreachable
+}
+
+define void @unreachable_gap(i64 %x, i32* %p) {
+; Cases 6 and INT64_MAX become the new default, but we still exploit the fact
+; that 3-4 is unreachable, so four icmps is enough.
+
+; CHECK-LABEL: @unreachable_gap
+; CHECK:       icmp slt i64 %x, 2
+; CHECK:       icmp slt i64 %x, 5
+; CHECK:       icmp eq  i64 %x, 5
+; CHECK:       icmp slt i64 %x, 1
+; CHECK-NOT:   icmp
+
+entry:
+  switch i64 %x, label %default [
+    i64 -9223372036854775808, label %bb0
+    i64 1, label %bb1
+    i64 2, label %bb2
+    i64 5, label %bb3
+    i64 6, label %bb4
+    i64 9223372036854775807, label %bb4
+  ]
+bb0:
+  store i32 0, i32* %p
+  br label %exit
+bb1:
+  store i32 1, i32* %p
+  br label %exit
+bb2:
+  store i32 2, i32* %p
+  br label %exit
+bb3:
+  store i32 3, i32* %p
+  br label %exit
+bb4:
+  store i32 4, i32* %p
+  br label %exit
+exit:
+  ret void
+default:
+  unreachable
+}
+
+
+
+define void @nocases(i32 %x, i32* %p) {
+; Don't fall over when there are no cases.
+;
+; CHECK-LABEL: @nocases
+; CHECK-LABEL: entry
+; CHECK-NEXT:  br label %default
+;
+entry:
+  switch i32 %x, label %default [
+  ]
+default:
+  unreachable
+}
+
+define void @nocasesleft(i32 %x, i32* %p) {
+; Cases 2 and 4 are removed and we are left with no cases.
+;
+; CHECK-LABEL: @nocasesleft
+; CHECK-LABEL: entry
+; CHECK-NEXT:  br label %popular
+;
+entry:
+  switch i32 %x, label %default [
+    i32 2, label %popular
+    i32 4, label %popular
+  ]
+popular:
+  store i32 2, i32* %p
+  br label %exit
+exit:
+  ret void
+default:
+  unreachable
+}
diff --git a/test/Transforms/Mem2Reg/2002-03-28-UninitializedVal.ll b/test/Transforms/Mem2Reg/2002-03-28-UninitializedVal.ll
index 777f3757bb89..49b560551d4e 100644
--- a/test/Transforms/Mem2Reg/2002-03-28-UninitializedVal.ll
+++ b/test/Transforms/Mem2Reg/2002-03-28-UninitializedVal.ll
@@ -6,6 +6,6 @@
 define i32 @test() {
         ; To be promoted
 	%X = alloca i32		; <i32*> [#uses=1]
-	%Y = load i32* %X		; <i32> [#uses=1]
+	%Y = load i32, i32* %X		; <i32> [#uses=1]
 	ret i32 %Y
 }
diff --git a/test/Transforms/Mem2Reg/2003-04-24-MultipleIdenticalSuccessors.ll b/test/Transforms/Mem2Reg/2003-04-24-MultipleIdenticalSuccessors.ll
index f5f1ee34365d..a013ff401f6c 100644
--- a/test/Transforms/Mem2Reg/2003-04-24-MultipleIdenticalSuccessors.ll
+++ b/test/Transforms/Mem2Reg/2003-04-24-MultipleIdenticalSuccessors.ll
@@ -10,7 +10,7 @@ B2:		; preds = %0
 	store i32 2, i32* %X
 	br i1 %c2, label %Exit, label %Exit
 Exit:		; preds = %B2, %B2, %0
-	%Y = load i32* %X		; <i32> [#uses=1]
+	%Y = load i32, i32* %X		; <i32> [#uses=1]
 	ret i32 %Y
 }
 
diff --git a/test/Transforms/Mem2Reg/2003-06-26-IterativePromote.ll b/test/Transforms/Mem2Reg/2003-06-26-IterativePromote.ll
index e82caa9fe0f6..de7280e4d8e3 100644
--- a/test/Transforms/Mem2Reg/2003-06-26-IterativePromote.ll
+++ b/test/Transforms/Mem2Reg/2003-06-26-IterativePromote.ll
@@ -7,10 +7,10 @@ define i32 @test2() {
 	%p = alloca i32*		; <i32**> [#uses=2]
 	store i32 0, i32* %a
 	store i32* %a, i32** %p
-	%tmp.0 = load i32** %p		; <i32*> [#uses=1]
-	%tmp.1 = load i32* %tmp.0		; <i32> [#uses=1]
+	%tmp.0 = load i32*, i32** %p		; <i32*> [#uses=1]
+	%tmp.1 = load i32, i32* %tmp.0		; <i32> [#uses=1]
 	store i32 %tmp.1, i32* %result
-	%tmp.2 = load i32* %result		; <i32> [#uses=1]
+	%tmp.2 = load i32, i32* %result		; <i32> [#uses=1]
 	ret i32 %tmp.2
 }
 
diff --git a/test/Transforms/Mem2Reg/2003-10-05-DeadPHIInsertion.ll b/test/Transforms/Mem2Reg/2003-10-05-DeadPHIInsertion.ll
index 1d38efc74576..8d55a1d0fa10 100644
--- a/test/Transforms/Mem2Reg/2003-10-05-DeadPHIInsertion.ll
+++ b/test/Transforms/Mem2Reg/2003-10-05-DeadPHIInsertion.ll
@@ -9,11 +9,11 @@ define void @test(i32 %B, i1 %C) {
 	br i1 %C, label %L1, label %L2
 L1:		; preds = %0
 	store i32 %B, i32* %A
-	%D = load i32* %A		; <i32> [#uses=1]
+	%D = load i32, i32* %A		; <i32> [#uses=1]
 	call void @test( i32 %D, i1 false )
 	br label %L3
 L2:		; preds = %0
-	%E = load i32* %A		; <i32> [#uses=1]
+	%E = load i32, i32* %A		; <i32> [#uses=1]
 	call void @test( i32 %E, i1 true )
 	br label %L3
 L3:		; preds = %L2, %L1
diff --git a/test/Transforms/Mem2Reg/2005-06-30-ReadBeforeWrite.ll b/test/Transforms/Mem2Reg/2005-06-30-ReadBeforeWrite.ll
index 74355961fbf9..f0f1fdc7ee2b 100644
--- a/test/Transforms/Mem2Reg/2005-06-30-ReadBeforeWrite.ll
+++ b/test/Transforms/Mem2Reg/2005-06-30-ReadBeforeWrite.ll
@@ -14,31 +14,31 @@ entry:
 	store i32 0, i32* %i
 	br label %loopentry
 loopentry:		; preds = %endif, %entry
-	%tmp.0 = load i32* %n_addr		; <i32> [#uses=1]
+	%tmp.0 = load i32, i32* %n_addr		; <i32> [#uses=1]
 	%tmp.1 = add i32 %tmp.0, 1		; <i32> [#uses=1]
-	%tmp.2 = load i32* %i		; <i32> [#uses=1]
+	%tmp.2 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp.3 = icmp sgt i32 %tmp.1, %tmp.2		; <i1> [#uses=2]
 	%tmp.4 = zext i1 %tmp.3 to i32		; <i32> [#uses=0]
 	br i1 %tmp.3, label %no_exit, label %return
 no_exit:		; preds = %loopentry
-	%tmp.5 = load i32* %undef		; <i32> [#uses=1]
+	%tmp.5 = load i32, i32* %undef		; <i32> [#uses=1]
 	store i32 %tmp.5, i32* %out
 	store i32 0, i32* %undef
-	%tmp.6 = load i32* %i		; <i32> [#uses=1]
+	%tmp.6 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp.7 = icmp sgt i32 %tmp.6, 0		; <i1> [#uses=2]
 	%tmp.8 = zext i1 %tmp.7 to i32		; <i32> [#uses=0]
 	br i1 %tmp.7, label %then, label %endif
 then:		; preds = %no_exit
-	%tmp.9 = load i8** %p_addr		; <i8*> [#uses=1]
-	%tmp.10 = load i32* %i		; <i32> [#uses=1]
+	%tmp.9 = load i8*, i8** %p_addr		; <i8*> [#uses=1]
+	%tmp.10 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp.11 = sub i32 %tmp.10, 1		; <i32> [#uses=1]
-	%tmp.12 = getelementptr i8* %tmp.9, i32 %tmp.11		; <i8*> [#uses=1]
-	%tmp.13 = load i32* %out		; <i32> [#uses=1]
+	%tmp.12 = getelementptr i8, i8* %tmp.9, i32 %tmp.11		; <i8*> [#uses=1]
+	%tmp.13 = load i32, i32* %out		; <i32> [#uses=1]
 	%tmp.14 = trunc i32 %tmp.13 to i8		; <i8> [#uses=1]
 	store i8 %tmp.14, i8* %tmp.12
 	br label %endif
 endif:		; preds = %then, %no_exit
-	%tmp.15 = load i32* %i		; <i32> [#uses=1]
+	%tmp.15 = load i32, i32* %i		; <i32> [#uses=1]
 	%inc = add i32 %tmp.15, 1		; <i32> [#uses=1]
 	store i32 %inc, i32* %i
 	br label %loopentry
diff --git a/test/Transforms/Mem2Reg/2005-11-28-Crash.ll b/test/Transforms/Mem2Reg/2005-11-28-Crash.ll
index 8fd3351ba42d..4b1d7f6651f3 100644
--- a/test/Transforms/Mem2Reg/2005-11-28-Crash.ll
+++ b/test/Transforms/Mem2Reg/2005-11-28-Crash.ll
@@ -41,7 +41,7 @@ endif.3:		; preds = %then.3, %endif.1
 loopexit:		; preds = %loopentry
 	br label %endif.4
 then.4:		; No predecessors!
-	%tmp.61 = load i32* %flags		; <i32> [#uses=0]
+	%tmp.61 = load i32, i32* %flags		; <i32> [#uses=0]
 	br label %out
 dead_block_after_goto:		; No predecessors!
 	br label %endif.4
diff --git a/test/Transforms/Mem2Reg/2007-08-27-VolatileLoadsStores.ll b/test/Transforms/Mem2Reg/2007-08-27-VolatileLoadsStores.ll
index ea581d197e58..891af9843ed7 100644
--- a/test/Transforms/Mem2Reg/2007-08-27-VolatileLoadsStores.ll
+++ b/test/Transforms/Mem2Reg/2007-08-27-VolatileLoadsStores.ll
@@ -15,14 +15,14 @@ entry:
 	%tmp = alloca i32, align 4		; <i32*> [#uses=3]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store volatile i32 0, i32* %v, align 4
-	%tmp1 = call i32 @_setjmp( %struct.__jmp_buf_tag* getelementptr ([1 x %struct.__jmp_buf_tag]* @j, i32 0, i32 0) )		; <i32> [#uses=1]
+	%tmp1 = call i32 @_setjmp( %struct.__jmp_buf_tag* getelementptr ([1 x %struct.__jmp_buf_tag], [1 x %struct.__jmp_buf_tag]* @j, i32 0, i32 0) )		; <i32> [#uses=1]
 	%tmp2 = icmp ne i32 %tmp1, 0		; <i1> [#uses=1]
 	%tmp23 = zext i1 %tmp2 to i8		; <i8> [#uses=1]
 	%toBool = icmp ne i8 %tmp23, 0		; <i1> [#uses=1]
 	br i1 %toBool, label %bb, label %bb5
 
 bb:		; preds = %entry
-	%tmp4 = load volatile i32* %v, align 4		; <i32> [#uses=1]
+	%tmp4 = load volatile i32, i32* %v, align 4		; <i32> [#uses=1]
 	store i32 %tmp4, i32* %tmp, align 4
 	br label %bb6
 
@@ -33,12 +33,12 @@ bb5:		; preds = %entry
 	br label %bb6
 
 bb6:		; preds = %bb5, %bb
-	%tmp7 = load i32* %tmp, align 4		; <i32> [#uses=1]
+	%tmp7 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
 	store i32 %tmp7, i32* %retval, align 4
 	br label %return
 
 return:		; preds = %bb6
-	%retval8 = load i32* %retval		; <i32> [#uses=1]
+	%retval8 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval8
 }
 
diff --git a/test/Transforms/Mem2Reg/ConvertDebugInfo.ll b/test/Transforms/Mem2Reg/ConvertDebugInfo.ll
index a7369c0902b9..8a2eedd96baf 100644
--- a/test/Transforms/Mem2Reg/ConvertDebugInfo.ll
+++ b/test/Transforms/Mem2Reg/ConvertDebugInfo.ll
@@ -7,26 +7,26 @@ entry:
   %retval = alloca double                         ; <double*> [#uses=2]
   %0 = alloca double                              ; <double*> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
-  call void @llvm.dbg.declare(metadata i32* %i_addr, metadata !0, metadata !{}), !dbg !8
+  call void @llvm.dbg.declare(metadata i32* %i_addr, metadata !0, metadata !DIExpression()), !dbg !8
 ; CHECK: call void @llvm.dbg.value(metadata i32 %i, i64 0, metadata ![[IVAR:[0-9]*]], metadata {{.*}})
 ; CHECK: call void @llvm.dbg.value(metadata double %j, i64 0, metadata ![[JVAR:[0-9]*]], metadata {{.*}})
-; CHECK: ![[IVAR]] = {{.*}} ; [ DW_TAG_arg_variable ] [i]
-; CHECK: ![[JVAR]] = {{.*}} ; [ DW_TAG_arg_variable ] [j]
+; CHECK: ![[IVAR]] = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i"
+; CHECK: ![[JVAR]] = !DILocalVariable(tag: DW_TAG_arg_variable, name: "j"
   store i32 %i, i32* %i_addr
-  call void @llvm.dbg.declare(metadata double* %j_addr, metadata !9, metadata !{}), !dbg !8
+  call void @llvm.dbg.declare(metadata double* %j_addr, metadata !9, metadata !DIExpression()), !dbg !8
   store double %j, double* %j_addr
-  %1 = load i32* %i_addr, align 4, !dbg !10       ; <i32> [#uses=1]
+  %1 = load i32, i32* %i_addr, align 4, !dbg !10       ; <i32> [#uses=1]
   %2 = add nsw i32 %1, 1, !dbg !10                ; <i32> [#uses=1]
   %3 = sitofp i32 %2 to double, !dbg !10          ; <double> [#uses=1]
-  %4 = load double* %j_addr, align 8, !dbg !10    ; <double> [#uses=1]
+  %4 = load double, double* %j_addr, align 8, !dbg !10    ; <double> [#uses=1]
   %5 = fadd double %3, %4, !dbg !10               ; <double> [#uses=1]
   store double %5, double* %0, align 8, !dbg !10
-  %6 = load double* %0, align 8, !dbg !10         ; <double> [#uses=1]
+  %6 = load double, double* %0, align 8, !dbg !10         ; <double> [#uses=1]
   store double %6, double* %retval, align 8, !dbg !10
   br label %return, !dbg !10
 
 return:                                           ; preds = %entry
-  %retval1 = load double* %retval, !dbg !10       ; <double> [#uses=1]
+  %retval1 = load double, double* %retval, !dbg !10       ; <double> [#uses=1]
   ret double %retval1, !dbg !10
 }
 
@@ -35,18 +35,18 @@ declare void @llvm.dbg.declare(metadata, metadata, metadata) nounwind readnone
 !llvm.dbg.cu = !{!3}
 !llvm.module.flags = !{!14}
 
-!0 = !{!"0x101\00i\002\000", !1, !2, !7} ; [ DW_TAG_arg_variable ]
-!1 = !{!"0x2e\00testfunc\00testfunc\00testfunc\002\000\001\000\006\000\000\002", !12, !2, !4, null, double (i32, double)* @testfunc, null, null, null} ; [ DW_TAG_subprogram ]
-!2 = !{!"0x29", !12} ; [ DW_TAG_file_type ]
-!3 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build)\001\00\000\00\000", !12, !13, !13, null, null, null} ; [ DW_TAG_compile_unit ]
-!4 = !{!"0x15\00\000\000\000\000\000\000", !12, !2, null, !5, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i", line: 2, arg: 0, scope: !1, file: !2, type: !7)
+!1 = !DISubprogram(name: "testfunc", linkageName: "testfunc", line: 2, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, scopeLine: 2, file: !12, scope: !2, type: !4, function: double (i32, double)* @testfunc)
+!2 = !DIFile(filename: "testfunc.c", directory: "/tmp")
+!3 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: 0, file: !12, enums: !13, retainedTypes: !13)
+!4 = !DISubroutineType(types: !5)
 !5 = !{!6, !7, !6}
-!6 = !{!"0x24\00double\000\0064\0064\000\000\004", !12, !2} ; [ DW_TAG_base_type ]
-!7 = !{!"0x24\00int\000\0032\0032\000\000\005", !12, !2} ; [ DW_TAG_base_type ]
-!8 = !MDLocation(line: 2, scope: !1)
-!9 = !{!"0x101\00j\002\000", !1, !2, !6} ; [ DW_TAG_arg_variable ]
-!10 = !MDLocation(line: 3, scope: !11)
-!11 = !{!"0xb\002\000\000", !12, !1} ; [ DW_TAG_lexical_block ]
-!12 = !{!"testfunc.c", !"/tmp"}
-!13 = !{i32 0}
-!14 = !{i32 1, !"Debug Info Version", i32 2}
+!6 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
+!7 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!8 = !DILocation(line: 2, scope: !1)
+!9 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "j", line: 2, arg: 0, scope: !1, file: !2, type: !6)
+!10 = !DILocation(line: 3, scope: !11)
+!11 = distinct !DILexicalBlock(line: 2, column: 0, file: !12, scope: !1)
+!12 = !DIFile(filename: "testfunc.c", directory: "/tmp")
+!13 = !{}
+!14 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/Mem2Reg/ConvertDebugInfo2.ll b/test/Transforms/Mem2Reg/ConvertDebugInfo2.ll
index 76d2a1aa7d2b..831221b7f97f 100644
--- a/test/Transforms/Mem2Reg/ConvertDebugInfo2.ll
+++ b/test/Transforms/Mem2Reg/ConvertDebugInfo2.ll
@@ -11,18 +11,18 @@ entry:
   %z_addr.i = alloca i8*                          ; <i8**> [#uses=2]
   %a_addr = alloca i32                            ; <i32*> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
-  call void @llvm.dbg.declare(metadata i32* %a_addr, metadata !0, metadata !{}), !dbg !7
+  call void @llvm.dbg.declare(metadata i32* %a_addr, metadata !0, metadata !DIExpression()), !dbg !7
   store i32 %a, i32* %a_addr
-  %0 = load i32* %a_addr, align 4, !dbg !8        ; <i32> [#uses=1]
-  call void @llvm.dbg.declare(metadata i32* %x_addr.i, metadata !9, metadata !{}) nounwind, !dbg !15
+  %0 = load i32, i32* %a_addr, align 4, !dbg !8        ; <i32> [#uses=1]
+  call void @llvm.dbg.declare(metadata i32* %x_addr.i, metadata !9, metadata !DIExpression()) nounwind, !dbg !15
   store i32 %0, i32* %x_addr.i
-  call void @llvm.dbg.declare(metadata i64* %y_addr.i, metadata !16, metadata !{}) nounwind, !dbg !15
+  call void @llvm.dbg.declare(metadata i64* %y_addr.i, metadata !16, metadata !DIExpression()) nounwind, !dbg !15
   store i64 55, i64* %y_addr.i
-  call void @llvm.dbg.declare(metadata i8** %z_addr.i, metadata !17, metadata !{}) nounwind, !dbg !15
+  call void @llvm.dbg.declare(metadata i8** %z_addr.i, metadata !17, metadata !DIExpression()) nounwind, !dbg !15
   store i8* bitcast (void (i32)* @baz to i8*), i8** %z_addr.i
-  %1 = load i32* %x_addr.i, align 4, !dbg !18     ; <i32> [#uses=1]
-  %2 = load i64* %y_addr.i, align 8, !dbg !18     ; <i64> [#uses=1]
-  %3 = load i8** %z_addr.i, align 8, !dbg !18     ; <i8*> [#uses=1]
+  %1 = load i32, i32* %x_addr.i, align 4, !dbg !18     ; <i32> [#uses=1]
+  %2 = load i64, i64* %y_addr.i, align 8, !dbg !18     ; <i64> [#uses=1]
+  %3 = load i8*, i8** %z_addr.i, align 8, !dbg !18     ; <i8*> [#uses=1]
   call void @foo(i32 %1, i64 %2, i8* %3) nounwind, !dbg !18
   br label %return, !dbg !19
 
@@ -32,26 +32,26 @@ return:                                           ; preds = %entry
 
 !llvm.dbg.cu = !{!3}
 !llvm.module.flags = !{!22}
-!0 = !{!"0x101\00a\008\000", !1, !2, !6} ; [ DW_TAG_arg_variable ]
-!1 = !{!"0x2e\00baz\00baz\00baz\008\000\001\000\006\000\000\008", !20, !2, !4, null, void (i32)* @baz, null, null, null} ; [ DW_TAG_subprogram ]
-!2 = !{!"0x29", !20} ; [ DW_TAG_file_type ]
-!3 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build)\001\00\000\00\000", !20, !21, !21, null, null, null} ; [ DW_TAG_compile_unit ]
-!4 = !{!"0x15\00\000\000\000\000\000\000", !20, !2, null, !5, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "a", line: 8, arg: 0, scope: !1, file: !2, type: !6)
+!1 = !DISubprogram(name: "baz", linkageName: "baz", line: 8, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, scopeLine: 8, file: !20, scope: !2, type: !4, function: void (i32)* @baz)
+!2 = !DIFile(filename: "bar.c", directory: "/tmp/")
+!3 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: 0, file: !20, enums: !21, retainedTypes: !21)
+!4 = !DISubroutineType(types: !5)
 !5 = !{null, !6}
-!6 = !{!"0x24\00int\000\0032\0032\000\000\005", !20, !2} ; [ DW_TAG_base_type ]
-!7 = !MDLocation(line: 8, scope: !1)
-!8 = !MDLocation(line: 9, scope: !1)
-!9 = !{!"0x101\00x\004\000", !10, !2, !6} ; [ DW_TAG_arg_variable ]
-!10 = !{!"0x2e\00bar\00bar\00bar\004\001\001\000\006\000\000\004", !20, !2, !11, null, null, null, null, null} ; [ DW_TAG_subprogram ]
-!11 = !{!"0x15\00\000\000\000\000\000\000", !20, !2, null, !12, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!6 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!7 = !DILocation(line: 8, scope: !1)
+!8 = !DILocation(line: 9, scope: !1)
+!9 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "x", line: 4, arg: 0, scope: !10, file: !2, type: !6)
+!10 = !DISubprogram(name: "bar", linkageName: "bar", line: 4, isLocal: true, isDefinition: true, virtualIndex: 6, isOptimized: false, scopeLine: 4, file: !20, scope: !2, type: !11)
+!11 = !DISubroutineType(types: !12)
 !12 = !{null, !6, !13, !14}
-!13 = !{!"0x24\00long int\000\0064\0064\000\000\005", !20, !2} ; [ DW_TAG_base_type ]
-!14 = !{!"0xf\00\000\0064\0064\000\000", !20, !2, null} ; [ DW_TAG_pointer_type ]
-!15 = !MDLocation(line: 4, scope: !10, inlinedAt: !8)
-!16 = !{!"0x101\00y\004\000", !10, !2, !13} ; [ DW_TAG_arg_variable ]
-!17 = !{!"0x101\00z\004\000", !10, !2, !14} ; [ DW_TAG_arg_variable ]
-!18 = !MDLocation(line: 5, scope: !10, inlinedAt: !8)
-!19 = !MDLocation(line: 10, scope: !1)
-!20 = !{!"bar.c", !"/tmp/"}
-!21 = !{i32 0}
-!22 = !{i32 1, !"Debug Info Version", i32 2}
+!13 = !DIBasicType(tag: DW_TAG_base_type, name: "long int", size: 64, align: 64, encoding: DW_ATE_signed)
+!14 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, file: !20, scope: !2, baseType: null)
+!15 = !DILocation(line: 4, scope: !10, inlinedAt: !8)
+!16 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "y", line: 4, arg: 0, scope: !10, file: !2, type: !13)
+!17 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "z", line: 4, arg: 0, scope: !10, file: !2, type: !14)
+!18 = !DILocation(line: 5, scope: !10, inlinedAt: !8)
+!19 = !DILocation(line: 10, scope: !1)
+!20 = !DIFile(filename: "bar.c", directory: "/tmp/")
+!21 = !{}
+!22 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/Mem2Reg/PromoteMemToRegister.ll b/test/Transforms/Mem2Reg/PromoteMemToRegister.ll
index 1be6b03beec6..b7f39947afb4 100644
--- a/test/Transforms/Mem2Reg/PromoteMemToRegister.ll
+++ b/test/Transforms/Mem2Reg/PromoteMemToRegister.ll
@@ -6,12 +6,12 @@ define double @testfunc(i32 %i, double %j) {
 	%J = alloca double		; <double*> [#uses=2]
 	store i32 %i, i32* %I
 	store double %j, double* %J
-	%t1 = load i32* %I		; <i32> [#uses=1]
+	%t1 = load i32, i32* %I		; <i32> [#uses=1]
 	%t2 = add i32 %t1, 1		; <i32> [#uses=1]
 	store i32 %t2, i32* %I
-	%t3 = load i32* %I		; <i32> [#uses=1]
+	%t3 = load i32, i32* %I		; <i32> [#uses=1]
 	%t4 = sitofp i32 %t3 to double		; <double> [#uses=1]
-	%t5 = load double* %J		; <double> [#uses=1]
+	%t5 = load double, double* %J		; <double> [#uses=1]
 	%t6 = fmul double %t4, %t5		; <double> [#uses=1]
 	ret double %t6
 }
diff --git a/test/Transforms/Mem2Reg/UndefValuesMerge.ll b/test/Transforms/Mem2Reg/UndefValuesMerge.ll
index 5013229b77f9..eeeb72f8e5a4 100644
--- a/test/Transforms/Mem2Reg/UndefValuesMerge.ll
+++ b/test/Transforms/Mem2Reg/UndefValuesMerge.ll
@@ -7,7 +7,7 @@ T:		; preds = %0
 	store i32 %i, i32* %I
 	br label %Cont
 Cont:		; preds = %T, %0
-	%Y = load i32* %I		; <i32> [#uses=1]
+	%Y = load i32, i32* %I		; <i32> [#uses=1]
 	ret i32 %Y
 }
 
diff --git a/test/Transforms/Mem2Reg/atomic.ll b/test/Transforms/Mem2Reg/atomic.ll
index 5bc9e9281b27..f20043d9716a 100644
--- a/test/Transforms/Mem2Reg/atomic.ll
+++ b/test/Transforms/Mem2Reg/atomic.ll
@@ -7,6 +7,6 @@ define i32 @test1(i32 %x) {
 ; CHECK: ret i32 %x
   %a = alloca i32
   store atomic i32 %x, i32* %a seq_cst, align 4
-  %r = load atomic i32* %a seq_cst, align 4
+  %r = load atomic i32, i32* %a seq_cst, align 4
   ret i32 %r
 }
diff --git a/test/Transforms/Mem2Reg/crash.ll b/test/Transforms/Mem2Reg/crash.ll
index 59e2c0b3108d..a4a31b112ee8 100644
--- a/test/Transforms/Mem2Reg/crash.ll
+++ b/test/Transforms/Mem2Reg/crash.ll
@@ -14,7 +14,7 @@ invcont2:
   br label %bb15
 
 bb15:
-  %B = load i32* %whichFlag
+  %B = load i32, i32* %whichFlag
   ret i32 %B
 
 lpad86:
@@ -33,11 +33,11 @@ entry:
   br label %bb15
 
 bb15:
-  %B = load i32* %whichFlag
+  %B = load i32, i32* %whichFlag
   ret i32 %B
 
 invcont2:
-  %C = load i32* %whichFlag
+  %C = load i32, i32* %whichFlag
   store i32 %C, i32* %whichFlag
   br label %bb15
 }
diff --git a/test/Transforms/Mem2Reg/ignore-lifetime.ll b/test/Transforms/Mem2Reg/ignore-lifetime.ll
index 5e4f9bfd8c5b..12adaffc7714 100644
--- a/test/Transforms/Mem2Reg/ignore-lifetime.ll
+++ b/test/Transforms/Mem2Reg/ignore-lifetime.ll
@@ -18,7 +18,7 @@ define void @test2() {
 ; CHECK: test2
 ; CHECK-NOT: alloca
   %A = alloca {i8, i16}
-  %B = getelementptr {i8, i16}* %A, i32 0, i32 0
+  %B = getelementptr {i8, i16}, {i8, i16}* %A, i32 0, i32 0
   call void @llvm.lifetime.start(i64 2, i8* %B)
   store {i8, i16} zeroinitializer, {i8, i16}* %A
   call void @llvm.lifetime.end(i64 2, i8* %B)
diff --git a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
index 00ac34d93e79..f73addd62687 100644
--- a/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
+++ b/test/Transforms/MemCpyOpt/2008-02-24-MultipleUseofSRet.ll
@@ -8,9 +8,9 @@ target triple = "i386-pc-linux-gnu"
 
 define internal fastcc void @initialize(%0* noalias nocapture sret %agg.result) nounwind {
 entry:
-  %agg.result.03 = getelementptr %0* %agg.result, i32 0, i32 0
+  %agg.result.03 = getelementptr %0, %0* %agg.result, i32 0, i32 0
   store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.03
-  %agg.result.15 = getelementptr %0* %agg.result, i32 0, i32 1
+  %agg.result.15 = getelementptr %0, %0* %agg.result, i32 0, i32 1
   store x86_fp80 0xK00000000000000000000, x86_fp80* %agg.result.15
   ret void
 }
diff --git a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
index 6982c8bf2f85..ffbb2992557a 100644
--- a/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
+++ b/test/Transforms/MemCpyOpt/2008-03-13-ReturnSlotBitcast.ll
@@ -14,8 +14,8 @@ entry:
   %a_i8 = bitcast %a* %a_var to i8*
   %b_i8 = bitcast %b* %b_var to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b_i8, i8* %a_i8, i32 4, i32 1, i1 false)
-  %tmp1 = getelementptr %b* %b_var, i32 0, i32 0
-  %tmp2 = load float* %tmp1
+  %tmp1 = getelementptr %b, %b* %b_var, i32 0, i32 0
+  %tmp2 = load float, float* %tmp1
   ret float %tmp2
 }
 
diff --git a/test/Transforms/MemCpyOpt/2011-06-02-CallSlotOverwritten.ll b/test/Transforms/MemCpyOpt/2011-06-02-CallSlotOverwritten.ll
index 132966ef6b98..8e4a0230d7fd 100644
--- a/test/Transforms/MemCpyOpt/2011-06-02-CallSlotOverwritten.ll
+++ b/test/Transforms/MemCpyOpt/2011-06-02-CallSlotOverwritten.ll
@@ -17,20 +17,20 @@ define i32 @foo() nounwind {
   call void @bar(%struct1* sret %x) nounwind
 ; CHECK: call void @bar(%struct1* sret %x)
 
-  %gepn1 = getelementptr inbounds %struct2* %y, i32 0, i32 0, i32 0
+  %gepn1 = getelementptr inbounds %struct2, %struct2* %y, i32 0, i32 0, i32 0
   store i32 0, i32* %gepn1, align 8
-  %gepn2 = getelementptr inbounds %struct2* %y, i32 0, i32 0, i32 1
+  %gepn2 = getelementptr inbounds %struct2, %struct2* %y, i32 0, i32 0, i32 1
   store i32 0, i32* %gepn2, align 4
 
   %bit1 = bitcast %struct1* %x to i64*
   %bit2 = bitcast %struct2* %y to i64*
-  %load = load i64* %bit1, align 8
+  %load = load i64, i64* %bit1, align 8
   store i64 %load, i64* %bit2, align 8
 
-; CHECK: %load = load i64* %bit1, align 8
+; CHECK: %load = load i64, i64* %bit1, align 8
 ; CHECK: store i64 %load, i64* %bit2, align 8
 
-  %gep1 = getelementptr %struct2* %y, i32 0, i32 0, i32 0
-  %ret = load i32* %gep1
+  %gep1 = getelementptr %struct2, %struct2* %y, i32 0, i32 0, i32 0
+  %ret = load i32, i32* %gep1
   ret i32 %ret
 }
diff --git a/test/Transforms/MemCpyOpt/align.ll b/test/Transforms/MemCpyOpt/align.ll
index 5d5bfbdafc5b..9074684f6c76 100644
--- a/test/Transforms/MemCpyOpt/align.ll
+++ b/test/Transforms/MemCpyOpt/align.ll
@@ -10,13 +10,13 @@ declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 define void @foo(i32* %p) {
 ; CHECK-LABEL: @foo(
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* {{.*}}, i8 0, i64 16, i32 4, i1 false)
-  %a0 = getelementptr i32* %p, i64 0
+  %a0 = getelementptr i32, i32* %p, i64 0
   store i32 0, i32* %a0, align 4
-  %a1 = getelementptr i32* %p, i64 1
+  %a1 = getelementptr i32, i32* %p, i64 1
   store i32 0, i32* %a1, align 16
-  %a2 = getelementptr i32* %p, i64 2
+  %a2 = getelementptr i32, i32* %p, i64 2
   store i32 0, i32* %a2, align 4
-  %a3 = getelementptr i32* %p, i64 3
+  %a3 = getelementptr i32, i32* %p, i64 3
   store i32 0, i32* %a3, align 4
   ret void
 }
diff --git a/test/Transforms/MemCpyOpt/atomic.ll b/test/Transforms/MemCpyOpt/atomic.ll
index f351e6792e22..5be6b1555b56 100644
--- a/test/Transforms/MemCpyOpt/atomic.ll
+++ b/test/Transforms/MemCpyOpt/atomic.ll
@@ -16,9 +16,9 @@ define void @test1() nounwind uwtable ssp {
   %x = alloca [101 x i32], align 16
   %bc = bitcast [101 x i32]* %x to i8*
   call void @llvm.memset.p0i8.i64(i8* %bc, i8 0, i64 400, i32 16, i1 false)
-  %gep1 = getelementptr inbounds [101 x i32]* %x, i32 0, i32 100
+  %gep1 = getelementptr inbounds [101 x i32], [101 x i32]* %x, i32 0, i32 100
   store atomic i32 0, i32* %gep1 unordered, align 4
-  %gep2 = getelementptr inbounds [101 x i32]* %x, i32 0, i32 0
+  %gep2 = getelementptr inbounds [101 x i32], [101 x i32]* %x, i32 0, i32 0
   call void @otherf(i32* %gep2)
   ret void
 }
@@ -33,7 +33,7 @@ define void @test2() nounwind uwtable ssp {
   %new = alloca i32
   call void @otherf(i32* nocapture %old)
   store atomic i32 0, i32* @x unordered, align 4
-  %v = load i32* %old
+  %v = load i32, i32* %old
   store i32 %v, i32* %new
   call void @otherf(i32* nocapture %new)  
   ret void
diff --git a/test/Transforms/MemCpyOpt/callslot_deref.ll b/test/Transforms/MemCpyOpt/callslot_deref.ll
index 4d51552d0151..4c3e3e86bf38 100644
--- a/test/Transforms/MemCpyOpt/callslot_deref.ll
+++ b/test/Transforms/MemCpyOpt/callslot_deref.ll
@@ -10,19 +10,20 @@ define void @must_remove_memcpy(i8* noalias nocapture dereferenceable(4096) %dst
 ; CHECK: call void @llvm.memset.p0i8.i64
 ; CHECK-NOT: call void @llvm.memcpy.p0i8.p0i8.i64
   %src = alloca [4096 x i8], align 1
-  %p = getelementptr inbounds [4096 x i8]* %src, i64 0, i64 0
+  %p = getelementptr inbounds [4096 x i8], [4096 x i8]* %src, i64 0, i64 0
   call void @llvm.memset.p0i8.i64(i8* %p, i8 0, i64 4096, i32 1, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %p, i64 4096, i32 1, i1 false) #2
   ret void
 }
 
 ; memset touch more bytes than those guaranteed to be dereferenceable
+; We can't remove the memcpy, but we can turn it into an independent memset.
 define void @must_not_remove_memcpy(i8* noalias nocapture dereferenceable(1024) %dst) {
 ; CHECK-LABEL: @must_not_remove_memcpy(
 ; CHECK: call void @llvm.memset.p0i8.i64
-; CHECK: call void @llvm.memcpy.p0i8.p0i8.i64
+; CHECK: call void @llvm.memset.p0i8.i64
   %src = alloca [4096 x i8], align 1
-  %p = getelementptr inbounds [4096 x i8]* %src, i64 0, i64 0
+  %p = getelementptr inbounds [4096 x i8], [4096 x i8]* %src, i64 0, i64 0
   call void @llvm.memset.p0i8.i64(i8* %p, i8 0, i64 4096, i32 1, i1 false)
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %p, i64 4096, i32 1, i1 false) #2
   ret void
diff --git a/test/Transforms/MemCpyOpt/crash.ll b/test/Transforms/MemCpyOpt/crash.ll
index cc3a6b05278a..56757a6d1f8b 100644
--- a/test/Transforms/MemCpyOpt/crash.ll
+++ b/test/Transforms/MemCpyOpt/crash.ll
@@ -9,37 +9,37 @@ target triple = "armv7-eabi"
 ; PR4882
 define void @test1(%struct.bar* %this) {
 entry:
-  %0 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 0
+  %0 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 0, i32 0, i32 0
   store float 0.000000e+00, float* %0, align 4
-  %1 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 1
+  %1 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 0, i32 0, i32 1
   store float 0.000000e+00, float* %1, align 4
-  %2 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 2
+  %2 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 0, i32 0, i32 2
   store float 0.000000e+00, float* %2, align 4
-  %3 = getelementptr inbounds %struct.bar* %this, i32 0, i32 0, i32 0, i32 3
+  %3 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 0, i32 0, i32 3
   store float 0.000000e+00, float* %3, align 4
-  %4 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 0
+  %4 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 1, i32 0, i32 0
   store float 0.000000e+00, float* %4, align 4
-  %5 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 1
+  %5 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 1, i32 0, i32 1
   store float 0.000000e+00, float* %5, align 4
-  %6 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 2
+  %6 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 1, i32 0, i32 2
   store float 0.000000e+00, float* %6, align 4
-  %7 = getelementptr inbounds %struct.bar* %this, i32 0, i32 1, i32 0, i32 3
+  %7 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 1, i32 0, i32 3
   store float 0.000000e+00, float* %7, align 4
-  %8 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 1
+  %8 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 3, i32 0, i32 1
   store float 0.000000e+00, float* %8, align 4
-  %9 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 2
+  %9 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 3, i32 0, i32 2
   store float 0.000000e+00, float* %9, align 4
-  %10 = getelementptr inbounds %struct.bar* %this, i32 0, i32 3, i32 0, i32 3
+  %10 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 3, i32 0, i32 3
   store float 0.000000e+00, float* %10, align 4
-  %11 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 0
+  %11 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 4, i32 0, i32 0
   store float 0.000000e+00, float* %11, align 4
-  %12 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 1
+  %12 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 4, i32 0, i32 1
   store float 0.000000e+00, float* %12, align 4
-  %13 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 2
+  %13 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 4, i32 0, i32 2
   store float 0.000000e+00, float* %13, align 4
-  %14 = getelementptr inbounds %struct.bar* %this, i32 0, i32 4, i32 0, i32 3
+  %14 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 4, i32 0, i32 3
   store float 0.000000e+00, float* %14, align 4
-  %15 = getelementptr inbounds %struct.bar* %this, i32 0, i32 5
+  %15 = getelementptr inbounds %struct.bar, %struct.bar* %this, i32 0, i32 5
   store float 0.000000e+00, float* %15, align 4
   unreachable
 }
diff --git a/test/Transforms/MemCpyOpt/form-memset.ll b/test/Transforms/MemCpyOpt/form-memset.ll
index d980b7fdbd69..7d7f3a6dac3c 100644
--- a/test/Transforms/MemCpyOpt/form-memset.ll
+++ b/test/Transforms/MemCpyOpt/form-memset.ll
@@ -8,45 +8,45 @@ target triple = "i386-apple-darwin8"
 define void @test1(i8 signext  %c) nounwind  {
 entry:
 	%x = alloca [19 x i8]		; <[19 x i8]*> [#uses=20]
-	%tmp = getelementptr [19 x i8]* %x, i32 0, i32 0		; <i8*> [#uses=1]
+	%tmp = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 0		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp, align 1
-	%tmp5 = getelementptr [19 x i8]* %x, i32 0, i32 1		; <i8*> [#uses=1]
+	%tmp5 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 1		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp5, align 1
-	%tmp9 = getelementptr [19 x i8]* %x, i32 0, i32 2		; <i8*> [#uses=1]
+	%tmp9 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 2		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp9, align 1
-	%tmp13 = getelementptr [19 x i8]* %x, i32 0, i32 3		; <i8*> [#uses=1]
+	%tmp13 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 3		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp13, align 1
-	%tmp17 = getelementptr [19 x i8]* %x, i32 0, i32 4		; <i8*> [#uses=1]
+	%tmp17 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 4		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp17, align 1
-	%tmp21 = getelementptr [19 x i8]* %x, i32 0, i32 5		; <i8*> [#uses=1]
+	%tmp21 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 5		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp21, align 1
-	%tmp25 = getelementptr [19 x i8]* %x, i32 0, i32 6		; <i8*> [#uses=1]
+	%tmp25 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 6		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp25, align 1
-	%tmp29 = getelementptr [19 x i8]* %x, i32 0, i32 7		; <i8*> [#uses=1]
+	%tmp29 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 7		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp29, align 1
-	%tmp33 = getelementptr [19 x i8]* %x, i32 0, i32 8		; <i8*> [#uses=1]
+	%tmp33 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 8		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp33, align 1
-	%tmp37 = getelementptr [19 x i8]* %x, i32 0, i32 9		; <i8*> [#uses=1]
+	%tmp37 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 9		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp37, align 1
-	%tmp41 = getelementptr [19 x i8]* %x, i32 0, i32 10		; <i8*> [#uses=1]
+	%tmp41 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 10		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp41, align 1
-	%tmp45 = getelementptr [19 x i8]* %x, i32 0, i32 11		; <i8*> [#uses=1]
+	%tmp45 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 11		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp45, align 1
-	%tmp49 = getelementptr [19 x i8]* %x, i32 0, i32 12		; <i8*> [#uses=1]
+	%tmp49 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 12		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp49, align 1
-	%tmp53 = getelementptr [19 x i8]* %x, i32 0, i32 13		; <i8*> [#uses=1]
+	%tmp53 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 13		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp53, align 1
-	%tmp57 = getelementptr [19 x i8]* %x, i32 0, i32 14		; <i8*> [#uses=1]
+	%tmp57 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 14		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp57, align 1
-	%tmp61 = getelementptr [19 x i8]* %x, i32 0, i32 15		; <i8*> [#uses=1]
+	%tmp61 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 15		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp61, align 1
-	%tmp65 = getelementptr [19 x i8]* %x, i32 0, i32 16		; <i8*> [#uses=1]
+	%tmp65 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 16		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp65, align 1
-	%tmp69 = getelementptr [19 x i8]* %x, i32 0, i32 17		; <i8*> [#uses=1]
+	%tmp69 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 17		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp69, align 1
-	%tmp73 = getelementptr [19 x i8]* %x, i32 0, i32 18		; <i8*> [#uses=1]
+	%tmp73 = getelementptr [19 x i8], [19 x i8]* %x, i32 0, i32 18		; <i8*> [#uses=1]
 	store i8 %c, i8* %tmp73, align 1
-	%tmp76 = call i32 (...)* @bar( [19 x i8]* %x ) nounwind
+	%tmp76 = call i32 (...) @bar( [19 x i8]* %x ) nounwind
 	ret void
 ; CHECK-LABEL: @test1(
 ; CHECK-NOT: store
@@ -65,88 +65,88 @@ entry:
 	%ref_idx = alloca [8 x i8]		; <[8 x i8]*> [#uses=8]
 	%left_mvd = alloca [8 x %struct.MV]		; <[8 x %struct.MV]*> [#uses=17]
 	%up_mvd = alloca [8 x %struct.MV]		; <[8 x %struct.MV]*> [#uses=17]
-	%tmp20 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 7		; <i8*> [#uses=1]
+	%tmp20 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 7		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp20, align 1
-	%tmp23 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 6		; <i8*> [#uses=1]
+	%tmp23 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 6		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp23, align 1
-	%tmp26 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 5		; <i8*> [#uses=1]
+	%tmp26 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 5		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp26, align 1
-	%tmp29 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 4		; <i8*> [#uses=1]
+	%tmp29 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 4		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp29, align 1
-	%tmp32 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 3		; <i8*> [#uses=1]
+	%tmp32 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 3		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp32, align 1
-	%tmp35 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 2		; <i8*> [#uses=1]
+	%tmp35 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 2		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp35, align 1
-	%tmp38 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 1		; <i8*> [#uses=1]
+	%tmp38 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 1		; <i8*> [#uses=1]
 	store i8 -1, i8* %tmp38, align 1
-	%tmp41 = getelementptr [8 x i8]* %ref_idx, i32 0, i32 0		; <i8*> [#uses=2]
+	%tmp41 = getelementptr [8 x i8], [8 x i8]* %ref_idx, i32 0, i32 0		; <i8*> [#uses=2]
 	store i8 -1, i8* %tmp41, align 1
-	%tmp43 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 7, i32 0		; <i16*> [#uses=1]
+	%tmp43 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 7, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp43, align 2
-	%tmp46 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 7, i32 1		; <i16*> [#uses=1]
+	%tmp46 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 7, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp46, align 2
-	%tmp57 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 6, i32 0		; <i16*> [#uses=1]
+	%tmp57 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 6, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp57, align 2
-	%tmp60 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 6, i32 1		; <i16*> [#uses=1]
+	%tmp60 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 6, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp60, align 2
-	%tmp71 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 5, i32 0		; <i16*> [#uses=1]
+	%tmp71 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 5, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp71, align 2
-	%tmp74 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 5, i32 1		; <i16*> [#uses=1]
+	%tmp74 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 5, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp74, align 2
-	%tmp85 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 4, i32 0		; <i16*> [#uses=1]
+	%tmp85 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 4, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp85, align 2
-	%tmp88 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 4, i32 1		; <i16*> [#uses=1]
+	%tmp88 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 4, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp88, align 2
-	%tmp99 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 3, i32 0		; <i16*> [#uses=1]
+	%tmp99 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 3, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp99, align 2
-	%tmp102 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 3, i32 1		; <i16*> [#uses=1]
+	%tmp102 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 3, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp102, align 2
-	%tmp113 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 2, i32 0		; <i16*> [#uses=1]
+	%tmp113 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 2, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp113, align 2
-	%tmp116 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 2, i32 1		; <i16*> [#uses=1]
+	%tmp116 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 2, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp116, align 2
-	%tmp127 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 1, i32 0		; <i16*> [#uses=1]
+	%tmp127 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 1, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp127, align 2
-	%tmp130 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 1, i32 1		; <i16*> [#uses=1]
+	%tmp130 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 1, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp130, align 2
-	%tmp141 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 0, i32 0		; <i16*> [#uses=1]
+	%tmp141 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 0, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp141, align 8
-	%tmp144 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 0, i32 1		; <i16*> [#uses=1]
+	%tmp144 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 0, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp144, align 2
-	%tmp148 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 7, i32 0		; <i16*> [#uses=1]
+	%tmp148 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 7, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp148, align 2
-	%tmp151 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 7, i32 1		; <i16*> [#uses=1]
+	%tmp151 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 7, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp151, align 2
-	%tmp162 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 6, i32 0		; <i16*> [#uses=1]
+	%tmp162 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 6, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp162, align 2
-	%tmp165 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 6, i32 1		; <i16*> [#uses=1]
+	%tmp165 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 6, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp165, align 2
-	%tmp176 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 5, i32 0		; <i16*> [#uses=1]
+	%tmp176 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 5, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp176, align 2
-	%tmp179 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 5, i32 1		; <i16*> [#uses=1]
+	%tmp179 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 5, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp179, align 2
-	%tmp190 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 4, i32 0		; <i16*> [#uses=1]
+	%tmp190 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 4, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp190, align 2
-	%tmp193 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 4, i32 1		; <i16*> [#uses=1]
+	%tmp193 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 4, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp193, align 2
-	%tmp204 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 3, i32 0		; <i16*> [#uses=1]
+	%tmp204 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 3, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp204, align 2
-	%tmp207 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 3, i32 1		; <i16*> [#uses=1]
+	%tmp207 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 3, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp207, align 2
-	%tmp218 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 2, i32 0		; <i16*> [#uses=1]
+	%tmp218 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 2, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp218, align 2
-	%tmp221 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 2, i32 1		; <i16*> [#uses=1]
+	%tmp221 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 2, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp221, align 2
-	%tmp232 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 1, i32 0		; <i16*> [#uses=1]
+	%tmp232 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 1, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp232, align 2
-	%tmp235 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 1, i32 1		; <i16*> [#uses=1]
+	%tmp235 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 1, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp235, align 2
-	%tmp246 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 0, i32 0		; <i16*> [#uses=1]
+	%tmp246 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 0, i32 0		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp246, align 8
-	%tmp249 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 0, i32 1		; <i16*> [#uses=1]
+	%tmp249 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 0, i32 1		; <i16*> [#uses=1]
 	store i16 0, i16* %tmp249, align 2
-	%up_mvd252 = getelementptr [8 x %struct.MV]* %up_mvd, i32 0, i32 0		; <%struct.MV*> [#uses=1]
-	%left_mvd253 = getelementptr [8 x %struct.MV]* %left_mvd, i32 0, i32 0		; <%struct.MV*> [#uses=1]
+	%up_mvd252 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %up_mvd, i32 0, i32 0		; <%struct.MV*> [#uses=1]
+	%left_mvd253 = getelementptr [8 x %struct.MV], [8 x %struct.MV]* %left_mvd, i32 0, i32 0		; <%struct.MV*> [#uses=1]
 	call void @foo( %struct.MV* %up_mvd252, %struct.MV* %left_mvd253, i8* %tmp41 ) nounwind 
 	ret void
         
@@ -167,9 +167,9 @@ declare void @foo(%struct.MV*, %struct.MV*, i8*)
 ; Store followed by memset.
 define void @test3(i32* nocapture %P) nounwind ssp {
 entry:
-  %arrayidx = getelementptr inbounds i32* %P, i64 1
+  %arrayidx = getelementptr inbounds i32, i32* %P, i64 1
   store i32 0, i32* %arrayidx, align 4
-  %add.ptr = getelementptr inbounds i32* %P, i64 2
+  %add.ptr = getelementptr inbounds i32, i32* %P, i64 2
   %0 = bitcast i32* %add.ptr to i8*
   tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 11, i32 1, i1 false)
   ret void
@@ -182,7 +182,7 @@ entry:
 define void @test4(i32* nocapture %P) nounwind ssp {
 entry:
   store i32 0, i32* %P, align 4
-  %add.ptr = getelementptr inbounds i32* %P, i64 1
+  %add.ptr = getelementptr inbounds i32, i32* %P, i64 1
   %0 = bitcast i32* %add.ptr to i8*
   tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 11, i32 1, i1 false)
   ret void
@@ -196,10 +196,10 @@ declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1) nounwind
 ; Memset followed by store.
 define void @test5(i32* nocapture %P) nounwind ssp {
 entry:
-  %add.ptr = getelementptr inbounds i32* %P, i64 2
+  %add.ptr = getelementptr inbounds i32, i32* %P, i64 2
   %0 = bitcast i32* %add.ptr to i8*
   tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 11, i32 1, i1 false)
-  %arrayidx = getelementptr inbounds i32* %P, i64 1
+  %arrayidx = getelementptr inbounds i32, i32* %P, i64 1
   store i32 0, i32* %arrayidx, align 4
   ret void
 ; CHECK-LABEL: @test5(
@@ -212,7 +212,7 @@ define void @test6(i32* nocapture %P) nounwind ssp {
 entry:
   %0 = bitcast i32* %P to i8*
   tail call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 12, i32 1, i1 false)
-  %add.ptr = getelementptr inbounds i32* %P, i64 3
+  %add.ptr = getelementptr inbounds i32, i32* %P, i64 3
   %1 = bitcast i32* %add.ptr to i8*
   tail call void @llvm.memset.p0i8.i64(i8* %1, i8 0, i64 12, i32 1, i1 false)
   ret void
@@ -224,13 +224,13 @@ entry:
 ; rdar://9892684
 define void @test7(i32* nocapture %c) nounwind optsize {
   store i32 -1, i32* %c, align 4
-  %1 = getelementptr inbounds i32* %c, i32 1
+  %1 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 -1, i32* %1, align 4
-  %2 = getelementptr inbounds i32* %c, i32 2
+  %2 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 -1, i32* %2, align 4
-  %3 = getelementptr inbounds i32* %c, i32 3
+  %3 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 -1, i32* %3, align 4
-  %4 = getelementptr inbounds i32* %c, i32 4
+  %4 = getelementptr inbounds i32, i32* %c, i32 4
   store i32 -1, i32* %4, align 4
 ; CHECK-LABEL: @test7(
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* %5, i8 -1, i64 20, i32 4, i1 false)
@@ -253,21 +253,21 @@ entry:
 
 define void @test9() nounwind {
   store i8 -1, i8* bitcast ([16 x i64]* @test9buf to i8*), align 16
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 1), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 2), align 2
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 3), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 4), align 4
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 5), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 6), align 2
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 7), align 1
-  store i8 -1, i8* bitcast (i64* getelementptr inbounds ([16 x i64]* @test9buf, i64 0, i64 1) to i8*), align 8
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 9), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 10), align 2
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 11), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 12), align 4
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 13), align 1
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 14), align 2
-  store i8 -1, i8* getelementptr (i8* bitcast ([16 x i64]* @test9buf to i8*), i64 15), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 1), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 2), align 2
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 3), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 4), align 4
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 5), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 6), align 2
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 7), align 1
+  store i8 -1, i8* bitcast (i64* getelementptr inbounds ([16 x i64], [16 x i64]* @test9buf, i64 0, i64 1) to i8*), align 8
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 9), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 10), align 2
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 11), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 12), align 4
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 13), align 1
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 14), align 2
+  store i8 -1, i8* getelementptr (i8, i8* bitcast ([16 x i64]* @test9buf to i8*), i64 15), align 1
   ret void
 ; CHECK-LABEL: @test9(
 ; CHECK: call void @llvm.memset.p0i8.i64(i8* bitcast ([16 x i64]* @test9buf to i8*), i8 -1, i64 16, i32 16, i1 false)
@@ -284,3 +284,18 @@ define void @test10(i8* nocapture %P) nounwind {
 ; CHECK-NOT: memset
 ; CHECK: ret void
 }
+
+; Memset followed by odd store.
+define void @test11(i32* nocapture %P) nounwind ssp {
+entry:
+  %add.ptr = getelementptr inbounds i32, i32* %P, i64 3
+  %0 = bitcast i32* %add.ptr to i8*
+  tail call void @llvm.memset.p0i8.i64(i8* %0, i8 1, i64 11, i32 1, i1 false)
+  %arrayidx = getelementptr inbounds i32, i32* %P, i64 0
+  %arrayidx.cast = bitcast i32* %arrayidx to i96*
+  store i96 310698676526526814092329217, i96* %arrayidx.cast, align 4
+  ret void
+; CHECK-LABEL: @test11(
+; CHECK-NOT: store
+; CHECK: call void @llvm.memset.p0i8.i64(i8* %1, i8 1, i64 23, i32 4, i1 false)
+}
diff --git a/test/Transforms/MemCpyOpt/loadstore-sret.ll b/test/Transforms/MemCpyOpt/loadstore-sret.ll
index d4a700d03115..55cbe59651f4 100644
--- a/test/Transforms/MemCpyOpt/loadstore-sret.ll
+++ b/test/Transforms/MemCpyOpt/loadstore-sret.ll
@@ -12,10 +12,10 @@ _ZNSt8auto_ptrIiED1Ev.exit:
   %temp.lvalue = alloca %"class.std::auto_ptr", align 8
 ; CHECK: call void @_Z3barv(%"class.std::auto_ptr"* sret %agg.result)
   call void @_Z3barv(%"class.std::auto_ptr"* sret %temp.lvalue)
-  %tmp.i.i = getelementptr inbounds %"class.std::auto_ptr"* %temp.lvalue, i64 0, i32 0
+  %tmp.i.i = getelementptr inbounds %"class.std::auto_ptr", %"class.std::auto_ptr"* %temp.lvalue, i64 0, i32 0
 ; CHECK-NOT: load
-  %tmp2.i.i = load i32** %tmp.i.i, align 8
-  %tmp.i.i4 = getelementptr inbounds %"class.std::auto_ptr"* %agg.result, i64 0, i32 0
+  %tmp2.i.i = load i32*, i32** %tmp.i.i, align 8
+  %tmp.i.i4 = getelementptr inbounds %"class.std::auto_ptr", %"class.std::auto_ptr"* %agg.result, i64 0, i32 0
 ; CHECK-NOT: store
   store i32* %tmp2.i.i, i32** %tmp.i.i4, align 8
 ; CHECK: ret void
diff --git a/test/Transforms/MemCpyOpt/memcpy-to-memset-with-lifetimes.ll b/test/Transforms/MemCpyOpt/memcpy-to-memset-with-lifetimes.ll
index 626317614c94..e3e57f09d88f 100644
--- a/test/Transforms/MemCpyOpt/memcpy-to-memset-with-lifetimes.ll
+++ b/test/Transforms/MemCpyOpt/memcpy-to-memset-with-lifetimes.ll
@@ -40,7 +40,7 @@ entry-block:
 ; CHECK:         %[[a_cast:[^=]+]] = bitcast [8 x i64]* %[[a]] to i8*
 ; CHECK:         call void @llvm.memset.p0i8.i64(i8* %[[a_cast]], i8 0, i64 64
 ; CHECK:         %[[sret_cast:[^=]+]] = bitcast [8 x i64]* %sret to i8*
-; CHECK:         call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[sret_cast]], i8* %[[a_cast]], i64 64
+; CHECK:         call void @llvm.memset.p0i8.i64(i8* %[[sret_cast]], i8 0, i64 64
 ; CHECK:         call void @llvm.memset.p0i8.i64(i8* %[[a_cast]], i8 42, i64 32
 ; CHECK:         %[[out_cast:[^=]+]] = bitcast [8 x i64]* %out to i8*
 ; CHECK:         call void @llvm.memcpy.p0i8.p0i8.i64(i8* %[[out_cast]], i8* %[[a_cast]], i64 64
diff --git a/test/Transforms/MemCpyOpt/memcpy-to-memset.ll b/test/Transforms/MemCpyOpt/memcpy-to-memset.ll
index 8409de7ad206..fd8b93c19b1d 100644
--- a/test/Transforms/MemCpyOpt/memcpy-to-memset.ll
+++ b/test/Transforms/MemCpyOpt/memcpy-to-memset.ll
@@ -9,7 +9,7 @@ define void @test1() nounwind {
   %arr = alloca [3 x i32], align 4
   %arr_i8 = bitcast [3 x i32]* %arr to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %arr_i8, i8* bitcast ([3 x i32]* @cst to i8*), i64 12, i32 4, i1 false)
-  %arraydecay = getelementptr inbounds [3 x i32]* %arr, i64 0, i64 0
+  %arraydecay = getelementptr inbounds [3 x i32], [3 x i32]* %arr, i64 0, i64 0
   call void @foo(i32* %arraydecay) nounwind
   ret void
 ; CHECK-LABEL: @test1(
diff --git a/test/Transforms/MemCpyOpt/memcpy-undef.ll b/test/Transforms/MemCpyOpt/memcpy-undef.ll
index 663b8dcfd3df..c75d020c0786 100644
--- a/test/Transforms/MemCpyOpt/memcpy-undef.ll
+++ b/test/Transforms/MemCpyOpt/memcpy-undef.ll
@@ -7,12 +7,12 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define i32 @test1(%struct.foo* nocapture %foobie) nounwind noinline ssp uwtable {
   %bletch.sroa.1 = alloca [7 x i8], align 1
-  %1 = getelementptr inbounds %struct.foo* %foobie, i64 0, i32 0
+  %1 = getelementptr inbounds %struct.foo, %struct.foo* %foobie, i64 0, i32 0
   store i8 98, i8* %1, align 4
-  %2 = getelementptr inbounds %struct.foo* %foobie, i64 0, i32 1, i64 0
-  %3 = getelementptr inbounds [7 x i8]* %bletch.sroa.1, i64 0, i64 0
+  %2 = getelementptr inbounds %struct.foo, %struct.foo* %foobie, i64 0, i32 1, i64 0
+  %3 = getelementptr inbounds [7 x i8], [7 x i8]* %bletch.sroa.1, i64 0, i64 0
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %2, i8* %3, i64 7, i32 1, i1 false)
-  %4 = getelementptr inbounds %struct.foo* %foobie, i64 0, i32 2
+  %4 = getelementptr inbounds %struct.foo, %struct.foo* %foobie, i64 0, i32 2
   store i32 20, i32* %4, align 4
   ret i32 undef
 
diff --git a/test/Transforms/MemCpyOpt/memcpy.ll b/test/Transforms/MemCpyOpt/memcpy.ll
index ee04f1951164..731847440d76 100644
--- a/test/Transforms/MemCpyOpt/memcpy.ll
+++ b/test/Transforms/MemCpyOpt/memcpy.ll
@@ -93,7 +93,7 @@ entry:
   %y = alloca %struct.S, align 16
   %tmp = bitcast %struct.S* %y to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp, i8* bitcast (%struct.S* @sS to i8*), i64 32, i32 16, i1 false)
-  %a = getelementptr %struct.S* %y, i64 0, i32 1, i64 0
+  %a = getelementptr %struct.S, %struct.S* %y, i64 0, i32 1, i64 0
   store i8 4, i8* %a
   call void @test5a(%struct.S* align 16 byval %y)
   ret i32 0
@@ -139,10 +139,10 @@ define void @test8() {
 ; CHECK: test8
 ; CHECK-NOT: memcpy
   %A = tail call i8* @malloc(i32 10)
-  %B = getelementptr inbounds i8* %A, i64 2
-  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %B, i8* getelementptr inbounds ([7 x i8]* @test8.str, i64 0, i64 0), i32 7, i32 1, i1 false)
+  %B = getelementptr inbounds i8, i8* %A, i64 2
+  tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %B, i8* getelementptr inbounds ([7 x i8], [7 x i8]* @test8.str, i64 0, i64 0), i32 7, i32 1, i1 false)
   %C = tail call i8* @malloc(i32 10)
-  %D = getelementptr inbounds i8* %C, i64 2
+  %D = getelementptr inbounds i8, i8* %C, i64 2
   tail call void @llvm.memcpy.p0i8.p0i8.i32(i8* %D, i8* %B, i32 7, i32 1, i1 false)
   ret void
 ; CHECK: ret void
@@ -196,7 +196,7 @@ define void @test10(%opaque* noalias nocapture sret %x, i32 %y) {
   %a = alloca i32, align 4
   store i32 %y, i32* %a
   call void @foo(i32* noalias nocapture %a)
-  %c = load i32* %a
+  %c = load i32, i32* %a
   %d = bitcast %opaque* %x to i32*
   store i32 %c, i32* %d
   ret void
diff --git a/test/Transforms/MemCpyOpt/memmove.ll b/test/Transforms/MemCpyOpt/memmove.ll
index 2057760efa01..1af85a1edc56 100644
--- a/test/Transforms/MemCpyOpt/memmove.ll
+++ b/test/Transforms/MemCpyOpt/memmove.ll
@@ -11,9 +11,9 @@ entry:
 ; CHECK-LABEL: @test1(
 ; CHECK: call void @llvm.memcpy
 
-  %malloccall = tail call i8* @malloc(i32 trunc (i64 mul nuw (i64 ptrtoint (i8* getelementptr (i8* null, i32 1) to i64), i64 13) to i32))
+  %malloccall = tail call i8* @malloc(i32 trunc (i64 mul nuw (i64 ptrtoint (i8* getelementptr (i8, i8* null, i32 1) to i64), i64 13) to i32))
   %call3 = bitcast i8* %malloccall to [13 x i8]*
-  %call3.sub = getelementptr inbounds [13 x i8]* %call3, i64 0, i64 0
+  %call3.sub = getelementptr inbounds [13 x i8], [13 x i8]* %call3, i64 0, i64 0
   tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %call3.sub, i8* %src, i64 13, i32 1, i1 false)
   ret i8* %call3.sub
 }
@@ -24,7 +24,7 @@ define void @test2(i8* %P) nounwind {
 entry:
 ; CHECK-LABEL: @test2(
 ; CHECK: call void @llvm.memcpy
-  %add.ptr = getelementptr i8* %P, i64 16
+  %add.ptr = getelementptr i8, i8* %P, i64 16
   tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %P, i8* %add.ptr, i64 16, i32 1, i1 false)
   ret void
 }
@@ -34,7 +34,7 @@ define void @test3(i8* %P) nounwind {
 entry:
 ; CHECK-LABEL: @test3(
 ; CHECK: call void @llvm.memmove
-  %add.ptr = getelementptr i8* %P, i64 16
+  %add.ptr = getelementptr i8, i8* %P, i64 16
   tail call void @llvm.memmove.p0i8.p0i8.i64(i8* %P, i8* %add.ptr, i64 17, i32 1, i1 false)
   ret void
 }
diff --git a/test/Transforms/MemCpyOpt/memset-memcpy-redundant-memset.ll b/test/Transforms/MemCpyOpt/memset-memcpy-redundant-memset.ll
new file mode 100644
index 000000000000..993c8a1bb2d0
--- /dev/null
+++ b/test/Transforms/MemCpyOpt/memset-memcpy-redundant-memset.ll
@@ -0,0 +1,168 @@
+; RUN: opt -basicaa -memcpyopt -S %s | FileCheck %s
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+; CHECK-LABEL: define void @test
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i64 %src_size
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i64 %dst_size, %src_size
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i64 %dst_size, %src_size
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i64 0, i64 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* [[DST]], i8 %c, i64 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test(i8* %src, i64 %src_size, i8* %dst, i64 %dst_size, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 %c, i64 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types_i32_i64
+; CHECK-DAG: [[DSTSIZE:%[0-9]+]] = zext i32 %dst_size to i64
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i64 %src_size
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i64 [[DSTSIZE]], %src_size
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i64 [[DSTSIZE]], %src_size
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i64 0, i64 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* [[DST]], i8 %c, i64 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types_i32_i64(i8* %dst, i8* %src, i32 %dst_size, i64 %src_size, i8 %c) {
+  call void @llvm.memset.p0i8.i32(i8* %dst, i8 %c, i32 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types_i128_i32
+; CHECK-DAG: [[SRCSIZE:%[0-9]+]] = zext i32 %src_size to i128
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i128 [[SRCSIZE]]
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i128 %dst_size, [[SRCSIZE]]
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i128 %dst_size, [[SRCSIZE]]
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i128 0, i128 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i128(i8* [[DST]], i8 %c, i128 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %src, i32 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types_i128_i32(i8* %dst, i8* %src, i128 %dst_size, i32 %src_size, i8 %c) {
+  call void @llvm.memset.p0i8.i128(i8* %dst, i8 %c, i128 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %src, i32 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types_i32_i128
+; CHECK-DAG: [[DSTSIZE:%[0-9]+]] = zext i32 %dst_size to i128
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i128 %src_size
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i128 [[DSTSIZE]], %src_size
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i128 [[DSTSIZE]], %src_size
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i128 0, i128 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i128(i8* [[DST]], i8 %c, i128 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i128(i8* %dst, i8* %src, i128 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types_i32_i128(i8* %dst, i8* %src, i32 %dst_size, i128 %src_size, i8 %c) {
+  call void @llvm.memset.p0i8.i32(i8* %dst, i8 %c, i32 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i128(i8* %dst, i8* %src, i128 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types_i64_i32
+; CHECK-DAG: [[SRCSIZE:%[0-9]+]] = zext i32 %src_size to i64
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i64 [[SRCSIZE]]
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i64 %dst_size, [[SRCSIZE]]
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i64 %dst_size, [[SRCSIZE]]
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i64 0, i64 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* [[DST]], i8 %c, i64 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %src, i32 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types_i64_i32(i8* %dst, i8* %src, i64 %dst_size, i32 %src_size, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 %c, i64 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %src, i32 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_align_same
+; CHECK: call void @llvm.memset.p0i8.i64(i8* {{.*}}, i8 0, i64 {{.*}}, i32 8, i1 false)
+define void @test_align_same(i8* %src, i8* %dst, i64 %dst_size) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %dst_size, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 80, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_align_min
+; CHECK: call void @llvm.memset.p0i8.i64(i8* {{.*}}, i8 0, i64 {{.*}}, i32 4, i1 false)
+define void @test_align_min(i8* %src, i8* %dst, i64 %dst_size) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %dst_size, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 36, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_align_memcpy
+; CHECK: call void @llvm.memset.p0i8.i64(i8* {{.*}}, i8 0, i64 {{.*}}, i32 8, i1 false)
+define void @test_align_memcpy(i8* %src, i8* %dst, i64 %dst_size) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 80, i32 8, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_non_i8_dst_type
+; CHECK-NEXT: %dst = bitcast i64* %dst_pi64 to i8*
+; CHECK-DAG: [[DST:%[0-9]+]] = getelementptr i8, i8* %dst, i64 %src_size
+; CHECK-DAG: [[ULE:%[0-9]+]] = icmp ule i64 %dst_size, %src_size
+; CHECK-DAG: [[SIZEDIFF:%[0-9]+]] = sub i64 %dst_size, %src_size
+; CHECK-DAG: [[SIZE:%[0-9]+]] = select i1 [[ULE]], i64 0, i64 [[SIZEDIFF]]
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* [[DST]], i8 %c, i64 [[SIZE]], i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_non_i8_dst_type(i8* %src, i64 %src_size, i64* %dst_pi64, i64 %dst_size, i8 %c) {
+  %dst = bitcast i64* %dst_pi64 to i8*
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 %c, i64 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst, i8* %src, i64 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_dst
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %dst_size, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %src, i64 %src_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_dst(i8* %dst2, i8* %src, i64 %src_size, i8* %dst, i64 %dst_size) {
+  call void @llvm.memset.p0i8.i64(i8* %dst, i8 0, i64 %dst_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %src, i64 %src_size, i32 1, i1 false)
+  ret void
+}
+
+; Make sure we also take into account dependencies on the destination.
+
+; CHECK-LABEL: define i8 @test_intermediate_read
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 64, i32 1, i1 false)
+; CHECK-NEXT: %r = load i8, i8* %a
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 24, i32 1, i1 false)
+; CHECK-NEXT: ret i8 %r
+define i8 @test_intermediate_read(i8* %a, i8* %b) #0 {
+  call void @llvm.memset.p0i8.i64(i8* %a, i8 0, i64 64, i32 1, i1 false)
+  %r = load i8, i8* %a
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a, i8* %b, i64 24, i32 1, i1 false)
+  ret i8 %r
+}
+
+%struct = type { [8 x i8], [8 x i8] }
+
+; CHECK-LABEL: define void @test_intermediate_write
+; CHECK-NEXT: %a = alloca %struct
+; CHECK-NEXT: %a0 = getelementptr %struct, %struct* %a, i32 0, i32 0, i32 0
+; CHECK-NEXT: %a1 = getelementptr %struct, %struct* %a, i32 0, i32 1, i32 0
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %a0, i8 0, i64 16, i32 1, i1 false)
+; CHECK-NEXT: store i8 1, i8* %a1
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a0, i8* %b, i64 8, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_intermediate_write(i8* %b) #0 {
+  %a = alloca %struct
+  %a0 = getelementptr %struct, %struct* %a, i32 0, i32 0, i32 0
+  %a1 = getelementptr %struct, %struct* %a, i32 0, i32 1, i32 0
+  call void @llvm.memset.p0i8.i64(i8* %a0, i8 0, i64 16, i32 1, i1 false)
+  store i8 1, i8* %a1
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %a0, i8* %b, i64 8, i32 1, i1 false)
+  ret void
+}
+
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1)
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1)
+declare void @llvm.memset.p0i8.i128(i8* nocapture, i8, i128, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i128(i8* nocapture, i8* nocapture readonly, i128, i32, i1)
diff --git a/test/Transforms/MemCpyOpt/memset-memcpy-to-2x-memset.ll b/test/Transforms/MemCpyOpt/memset-memcpy-to-2x-memset.ll
new file mode 100644
index 000000000000..1c567042ce79
--- /dev/null
+++ b/test/Transforms/MemCpyOpt/memset-memcpy-to-2x-memset.ll
@@ -0,0 +1,101 @@
+; RUN: opt -memcpyopt -S %s | FileCheck %s
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+; CHECK-LABEL: define void @test(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst2, i8 %c, i64 128, i32 8, i1 false)
+; CHECK-NEXT: ret void
+define void @test(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 8, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_smaller_memcpy(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst2, i8 %c, i64 100, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_smaller_memcpy(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 100, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_smaller_memset(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 100, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_smaller_memset(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 100, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_align_memset(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 8, i1 false)
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst2, i8 %c, i64 128, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_align_memset(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 8, i1 false)
+; CHECK-NEXT: call void @llvm.memset.p0i8.i32(i8* %dst2, i8 %c, i32 100, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst2, i8* %dst1, i32 100, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_types_2(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i32(i8* %dst1, i8 %c, i32 128, i32 8, i1 false)
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst2, i8 %c, i64 100, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_types_2(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i32(i8* %dst1, i8 %c, i32 128, i32 8, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 100, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_different_source_gep(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+; CHECK-NEXT: %p = getelementptr i8, i8* %dst1, i64 64
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %p, i64 64, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_different_source_gep(i8* %dst1, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+  ; FIXME: We could optimize this as well.
+  %p = getelementptr i8, i8* %dst1, i64 64
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %p, i64 64, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_variable_size_1(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 %dst1_size, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_variable_size_1(i8* %dst1, i64 %dst1_size, i8* %dst2, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 %dst1_size, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 128, i32 1, i1 false)
+  ret void
+}
+
+; CHECK-LABEL: define void @test_variable_size_2(
+; CHECK-NEXT: call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 %dst2_size, i32 1, i1 false)
+; CHECK-NEXT: ret void
+define void @test_variable_size_2(i8* %dst1, i8* %dst2, i64 %dst2_size, i8 %c) {
+  call void @llvm.memset.p0i8.i64(i8* %dst1, i8 %c, i64 128, i32 1, i1 false)
+  call void @llvm.memcpy.p0i8.p0i8.i64(i8* %dst2, i8* %dst1, i64 %dst2_size, i32 1, i1 false)
+  ret void
+}
+
+declare void @llvm.memset.p0i8.i64(i8* nocapture, i8, i64, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i64(i8* nocapture, i8* nocapture readonly, i64, i32, i1)
+declare void @llvm.memset.p0i8.i32(i8* nocapture, i8, i32, i32, i1)
+declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture readonly, i32, i32, i1)
diff --git a/test/Transforms/MemCpyOpt/smaller.ll b/test/Transforms/MemCpyOpt/smaller.ll
index 1d35582d993e..8f6cafa6c012 100644
--- a/test/Transforms/MemCpyOpt/smaller.ll
+++ b/test/Transforms/MemCpyOpt/smaller.ll
@@ -4,7 +4,7 @@
 ; Memcpyopt shouldn't optimize the second memcpy using the first
 ; because the first has a smaller size.
 
-; CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* getelementptr inbounds (%struct.s* @cell, i32 0, i32 0, i32 0), i32 16, i32 4, i1 false)
+; CHECK: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* getelementptr inbounds (%struct.s, %struct.s* @cell, i32 0, i32 0, i32 0), i32 16, i32 4, i1 false)
 
 target datalayout = "e-p:32:32:32"
 
@@ -19,10 +19,10 @@ declare void @llvm.memcpy.p0i8.p0i8.i32(i8* nocapture, i8* nocapture, i32, i32,
 define void @foo() nounwind {
 entry:
   %agg.tmp = alloca %struct.s, align 4
-  store i32 99, i32* getelementptr inbounds (%struct.s* @cell, i32 0, i32 1), align 4
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds (%struct.s* @cell, i32 0, i32 0, i32 0), i8* getelementptr inbounds ([11 x i8]* @.str, i32 0, i32 0), i32 11, i32 1, i1 false)
-  %tmp = getelementptr inbounds %struct.s* %agg.tmp, i32 0, i32 0, i32 0
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* getelementptr inbounds (%struct.s* @cell, i32 0, i32 0, i32 0), i32 16, i32 4, i1 false)
+  store i32 99, i32* getelementptr inbounds (%struct.s, %struct.s* @cell, i32 0, i32 1), align 4
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds (%struct.s, %struct.s* @cell, i32 0, i32 0, i32 0), i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str, i32 0, i32 0), i32 11, i32 1, i1 false)
+  %tmp = getelementptr inbounds %struct.s, %struct.s* %agg.tmp, i32 0, i32 0, i32 0
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp, i8* getelementptr inbounds (%struct.s, %struct.s* @cell, i32 0, i32 0, i32 0), i32 16, i32 4, i1 false)
   call void @check(%struct.s* byval %agg.tmp)
   ret void
 }
diff --git a/test/Transforms/MemCpyOpt/sret.ll b/test/Transforms/MemCpyOpt/sret.ll
index bfe5e0fbb993..34ba4c4d8bb7 100644
--- a/test/Transforms/MemCpyOpt/sret.ll
+++ b/test/Transforms/MemCpyOpt/sret.ll
@@ -9,13 +9,13 @@ define void @ccosl(%0* noalias sret %agg.result, %0* byval align 8 %z) nounwind
 entry:
   %iz = alloca %0
   %memtmp = alloca %0, align 16
-  %tmp1 = getelementptr %0* %z, i32 0, i32 1
-  %tmp2 = load x86_fp80* %tmp1, align 16
+  %tmp1 = getelementptr %0, %0* %z, i32 0, i32 1
+  %tmp2 = load x86_fp80, x86_fp80* %tmp1, align 16
   %tmp3 = fsub x86_fp80 0xK80000000000000000000, %tmp2
-  %tmp4 = getelementptr %0* %iz, i32 0, i32 1
-  %real = getelementptr %0* %iz, i32 0, i32 0
-  %tmp7 = getelementptr %0* %z, i32 0, i32 0
-  %tmp8 = load x86_fp80* %tmp7, align 16
+  %tmp4 = getelementptr %0, %0* %iz, i32 0, i32 1
+  %real = getelementptr %0, %0* %iz, i32 0, i32 0
+  %tmp7 = getelementptr %0, %0* %z, i32 0, i32 0
+  %tmp8 = load x86_fp80, x86_fp80* %tmp7, align 16
   store x86_fp80 %tmp3, x86_fp80* %real, align 16
   store x86_fp80 %tmp8, x86_fp80* %tmp4, align 16
   call void @ccoshl(%0* noalias sret %memtmp, %0* byval align 8 %iz) nounwind
diff --git a/test/Transforms/MergeFunc/2011-02-08-RemoveEqual.ll b/test/Transforms/MergeFunc/2011-02-08-RemoveEqual.ll
index 19cd6a5171da..4735ea5e61d2 100644
--- a/test/Transforms/MergeFunc/2011-02-08-RemoveEqual.ll
+++ b/test/Transforms/MergeFunc/2011-02-08-RemoveEqual.ll
@@ -36,15 +36,15 @@ entry:
   store %"struct.kc::impl_fileline_FileLine"* %this, %"struct.kc::impl_fileline_FileLine"** %this_addr
   store %"struct.kc::impl_casestring__Str"* %_file, %"struct.kc::impl_casestring__Str"** %_file_addr
   store i32 %_line, i32* %_line_addr
-  %0 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %1 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %0, i32 0, i32 0
+  %0 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %1 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %0, i32 0, i32 0
   call void @_ZN2kc13impl_filelineC2Ev() nounwind
-  %2 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %3 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %2, i32 0, i32 0
-  %4 = getelementptr inbounds %"struct.kc::impl_fileline"* %3, i32 0, i32 0
-  %5 = getelementptr inbounds %"struct.kc::impl_abstract_phylum"* %4, i32 0, i32 0
-  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*]* @_ZTVN2kc22impl_fileline_FileLineE, i32 0, i32 2), i32 (...)*** %5, align 4
-  %6 = load %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
+  %2 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %3 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %2, i32 0, i32 0
+  %4 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %3, i32 0, i32 0
+  %5 = getelementptr inbounds %"struct.kc::impl_abstract_phylum", %"struct.kc::impl_abstract_phylum"* %4, i32 0, i32 0
+  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*], [13 x i32 (...)*]* @_ZTVN2kc22impl_fileline_FileLineE, i32 0, i32 2), i32 (...)*** %5, align 4
+  %6 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
   %7 = icmp eq %"struct.kc::impl_casestring__Str"* %6, null
   br i1 %7, label %bb, label %bb1
 
@@ -57,20 +57,20 @@ invcont:                                          ; preds = %bb
   br label %bb2
 
 bb1:                                              ; preds = %entry
-  %9 = load %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
+  %9 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
   store %"struct.kc::impl_casestring__Str"* %9, %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
   br label %bb2
 
 bb2:                                              ; preds = %bb1, %invcont
-  %10 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %11 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %10, i32 0, i32 0
-  %12 = getelementptr inbounds %"struct.kc::impl_fileline"* %11, i32 0, i32 1
-  %13 = load %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
+  %10 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %11 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %10, i32 0, i32 0
+  %12 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %11, i32 0, i32 1
+  %13 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
   store %"struct.kc::impl_casestring__Str"* %13, %"struct.kc::impl_casestring__Str"** %12, align 4
-  %14 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %15 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %14, i32 0, i32 0
-  %16 = getelementptr inbounds %"struct.kc::impl_fileline"* %15, i32 0, i32 2
-  %17 = load i32* %_line_addr, align 4
+  %14 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %15 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %14, i32 0, i32 0
+  %16 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %15, i32 0, i32 2
+  %17 = load i32, i32* %_line_addr, align 4
   store i32 %17, i32* %16, align 4
   ret void
 
@@ -79,21 +79,21 @@ lpad:                                             ; preds = %bb
               cleanup
   %exn = extractvalue { i8*, i32 } %eh_ptr, 0
   store i8* %exn, i8** %eh_exception
-  %eh_ptr4 = load i8** %eh_exception
+  %eh_ptr4 = load i8*, i8** %eh_exception
   %eh_select5 = extractvalue { i8*, i32 } %eh_ptr, 1
   store i32 %eh_select5, i32* %eh_selector
-  %eh_select = load i32* %eh_selector
+  %eh_select = load i32, i32* %eh_selector
   store i32 %eh_select, i32* %save_filt.150, align 4
-  %eh_value = load i8** %eh_exception
+  %eh_value = load i8*, i8** %eh_exception
   store i8* %eh_value, i8** %save_eptr.149, align 4
-  %18 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %18 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
   %19 = bitcast %"struct.kc::impl_fileline_FileLine"* %18 to %"struct.kc::impl_fileline"*
   call void @_ZN2kc13impl_filelineD2Ev(%"struct.kc::impl_fileline"* %19) nounwind
-  %20 = load i8** %save_eptr.149, align 4
+  %20 = load i8*, i8** %save_eptr.149, align 4
   store i8* %20, i8** %eh_exception, align 4
-  %21 = load i32* %save_filt.150, align 4
+  %21 = load i32, i32* %save_filt.150, align 4
   store i32 %21, i32* %eh_selector, align 4
-  %eh_ptr6 = load i8** %eh_exception
+  %eh_ptr6 = load i8*, i8** %eh_exception
   call void @_Unwind_Resume_or_Rethrow()
   unreachable
 }
@@ -105,16 +105,16 @@ entry:
   %this_addr = alloca %"struct.kc::impl_fileline"*, align 4
   %"alloca point" = bitcast i32 0 to i32
   store %"struct.kc::impl_fileline"* %this, %"struct.kc::impl_fileline"** %this_addr
-  %0 = load %"struct.kc::impl_fileline"** %this_addr, align 4
-  %1 = getelementptr inbounds %"struct.kc::impl_fileline"* %0, i32 0, i32 0
-  %2 = getelementptr inbounds %"struct.kc::impl_abstract_phylum"* %1, i32 0, i32 0
-  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*]* @_ZTVN2kc13impl_filelineE, i32 0, i32 2), i32 (...)*** %2, align 4
+  %0 = load %"struct.kc::impl_fileline"*, %"struct.kc::impl_fileline"** %this_addr, align 4
+  %1 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %0, i32 0, i32 0
+  %2 = getelementptr inbounds %"struct.kc::impl_abstract_phylum", %"struct.kc::impl_abstract_phylum"* %1, i32 0, i32 0
+  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*], [13 x i32 (...)*]* @_ZTVN2kc13impl_filelineE, i32 0, i32 2), i32 (...)*** %2, align 4
   %3 = trunc i32 0 to i8
   %toBool = icmp ne i8 %3, 0
   br i1 %toBool, label %bb1, label %return
 
 bb1:                                              ; preds = %entry
-  %4 = load %"struct.kc::impl_fileline"** %this_addr, align 4
+  %4 = load %"struct.kc::impl_fileline"*, %"struct.kc::impl_fileline"** %this_addr, align 4
   %5 = bitcast %"struct.kc::impl_fileline"* %4 to i8*
   call void @_ZdlPv() nounwind
   br label %return
@@ -130,16 +130,16 @@ entry:
   %this_addr = alloca %"struct.kc::impl_fileline"*, align 4
   %"alloca point" = bitcast i32 0 to i32
   store %"struct.kc::impl_fileline"* %this, %"struct.kc::impl_fileline"** %this_addr
-  %0 = load %"struct.kc::impl_fileline"** %this_addr, align 4
-  %1 = getelementptr inbounds %"struct.kc::impl_fileline"* %0, i32 0, i32 0
-  %2 = getelementptr inbounds %"struct.kc::impl_abstract_phylum"* %1, i32 0, i32 0
-  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*]* @_ZTVN2kc13impl_filelineE, i32 0, i32 2), i32 (...)*** %2, align 4
+  %0 = load %"struct.kc::impl_fileline"*, %"struct.kc::impl_fileline"** %this_addr, align 4
+  %1 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %0, i32 0, i32 0
+  %2 = getelementptr inbounds %"struct.kc::impl_abstract_phylum", %"struct.kc::impl_abstract_phylum"* %1, i32 0, i32 0
+  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*], [13 x i32 (...)*]* @_ZTVN2kc13impl_filelineE, i32 0, i32 2), i32 (...)*** %2, align 4
   %3 = trunc i32 0 to i8
   %toBool = icmp ne i8 %3, 0
   br i1 %toBool, label %bb1, label %return
 
 bb1:                                              ; preds = %entry
-  %4 = load %"struct.kc::impl_fileline"** %this_addr, align 4
+  %4 = load %"struct.kc::impl_fileline"*, %"struct.kc::impl_fileline"** %this_addr, align 4
   %5 = bitcast %"struct.kc::impl_fileline"* %4 to i8*
   call void @_ZdlPv() nounwind
   br label %return
@@ -162,15 +162,15 @@ entry:
   store %"struct.kc::impl_fileline_FileLine"* %this, %"struct.kc::impl_fileline_FileLine"** %this_addr
   store %"struct.kc::impl_casestring__Str"* %_file, %"struct.kc::impl_casestring__Str"** %_file_addr
   store i32 %_line, i32* %_line_addr
-  %0 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %1 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %0, i32 0, i32 0
+  %0 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %1 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %0, i32 0, i32 0
   call void @_ZN2kc13impl_filelineC2Ev() nounwind
-  %2 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %3 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %2, i32 0, i32 0
-  %4 = getelementptr inbounds %"struct.kc::impl_fileline"* %3, i32 0, i32 0
-  %5 = getelementptr inbounds %"struct.kc::impl_abstract_phylum"* %4, i32 0, i32 0
-  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*]* @_ZTVN2kc22impl_fileline_FileLineE, i32 0, i32 2), i32 (...)*** %5, align 4
-  %6 = load %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
+  %2 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %3 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %2, i32 0, i32 0
+  %4 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %3, i32 0, i32 0
+  %5 = getelementptr inbounds %"struct.kc::impl_abstract_phylum", %"struct.kc::impl_abstract_phylum"* %4, i32 0, i32 0
+  store i32 (...)** getelementptr inbounds ([13 x i32 (...)*], [13 x i32 (...)*]* @_ZTVN2kc22impl_fileline_FileLineE, i32 0, i32 2), i32 (...)*** %5, align 4
+  %6 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
   %7 = icmp eq %"struct.kc::impl_casestring__Str"* %6, null
   br i1 %7, label %bb, label %bb1
 
@@ -183,20 +183,20 @@ invcont:                                          ; preds = %bb
   br label %bb2
 
 bb1:                                              ; preds = %entry
-  %9 = load %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
+  %9 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %_file_addr, align 4
   store %"struct.kc::impl_casestring__Str"* %9, %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
   br label %bb2
 
 bb2:                                              ; preds = %bb1, %invcont
-  %10 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %11 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %10, i32 0, i32 0
-  %12 = getelementptr inbounds %"struct.kc::impl_fileline"* %11, i32 0, i32 1
-  %13 = load %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
+  %10 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %11 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %10, i32 0, i32 0
+  %12 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %11, i32 0, i32 1
+  %13 = load %"struct.kc::impl_casestring__Str"*, %"struct.kc::impl_casestring__Str"** %iftmp.99, align 4
   store %"struct.kc::impl_casestring__Str"* %13, %"struct.kc::impl_casestring__Str"** %12, align 4
-  %14 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
-  %15 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine"* %14, i32 0, i32 0
-  %16 = getelementptr inbounds %"struct.kc::impl_fileline"* %15, i32 0, i32 2
-  %17 = load i32* %_line_addr, align 4
+  %14 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %15 = getelementptr inbounds %"struct.kc::impl_fileline_FileLine", %"struct.kc::impl_fileline_FileLine"* %14, i32 0, i32 0
+  %16 = getelementptr inbounds %"struct.kc::impl_fileline", %"struct.kc::impl_fileline"* %15, i32 0, i32 2
+  %17 = load i32, i32* %_line_addr, align 4
   store i32 %17, i32* %16, align 4
   ret void
 
@@ -205,21 +205,21 @@ lpad:                                             ; preds = %bb
               cleanup
   %exn = extractvalue { i8*, i32 } %eh_ptr, 0
   store i8* %exn, i8** %eh_exception
-  %eh_ptr4 = load i8** %eh_exception
+  %eh_ptr4 = load i8*, i8** %eh_exception
   %eh_select5 = extractvalue { i8*, i32 } %eh_ptr, 1
   store i32 %eh_select5, i32* %eh_selector
-  %eh_select = load i32* %eh_selector
+  %eh_select = load i32, i32* %eh_selector
   store i32 %eh_select, i32* %save_filt.148, align 4
-  %eh_value = load i8** %eh_exception
+  %eh_value = load i8*, i8** %eh_exception
   store i8* %eh_value, i8** %save_eptr.147, align 4
-  %18 = load %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
+  %18 = load %"struct.kc::impl_fileline_FileLine"*, %"struct.kc::impl_fileline_FileLine"** %this_addr, align 4
   %19 = bitcast %"struct.kc::impl_fileline_FileLine"* %18 to %"struct.kc::impl_fileline"*
   call void @_ZN2kc13impl_filelineD2Ev(%"struct.kc::impl_fileline"* %19) nounwind
-  %20 = load i8** %save_eptr.147, align 4
+  %20 = load i8*, i8** %save_eptr.147, align 4
   store i8* %20, i8** %eh_exception, align 4
-  %21 = load i32* %save_filt.148, align 4
+  %21 = load i32, i32* %save_filt.148, align 4
   store i32 %21, i32* %eh_selector, align 4
-  %eh_ptr6 = load i8** %eh_exception
+  %eh_ptr6 = load i8*, i8** %eh_exception
   call void @_Unwind_Resume_or_Rethrow()
   unreachable
 }
diff --git a/test/Transforms/MergeFunc/address-spaces.ll b/test/Transforms/MergeFunc/address-spaces.ll
index 0d66b8281fb2..1cfecae979d2 100644
--- a/test/Transforms/MergeFunc/address-spaces.ll
+++ b/test/Transforms/MergeFunc/address-spaces.ll
@@ -9,8 +9,8 @@ declare void @foo(i32) nounwind
 define i32 @store_as0(i32* %x) {
 ; CHECK-LABEL: @store_as0(
 ; CHECK: call void @foo(
-  %gep = getelementptr i32* %x, i32 4
-  %y = load i32* %gep
+  %gep = getelementptr i32, i32* %x, i32 4
+  %y = load i32, i32* %gep
   call void @foo(i32 %y) nounwind
   ret i32 %y
 }
@@ -18,8 +18,8 @@ define i32 @store_as0(i32* %x) {
 define i32 @store_as1(i32 addrspace(1)* %x) {
 ; CHECK-LABEL: @store_as1(
 ; CHECK: call void @foo(
-  %gep = getelementptr i32 addrspace(1)* %x, i32 4
-  %y = load i32 addrspace(1)* %gep
+  %gep = getelementptr i32, i32 addrspace(1)* %x, i32 4
+  %y = load i32, i32 addrspace(1)* %gep
   call void @foo(i32 %y) nounwind
   ret i32 %y
 }
@@ -27,8 +27,8 @@ define i32 @store_as1(i32 addrspace(1)* %x) {
 define i32 @store_as2(i32 addrspace(2)* %x) {
 ; CHECK-LABEL: @store_as2(
 ; CHECK: call void @foo(
-  %gep = getelementptr i32 addrspace(2)* %x, i32 4
-  %y = load i32 addrspace(2)* %gep
+  %gep = getelementptr i32, i32 addrspace(2)* %x, i32 4
+  %y = load i32, i32 addrspace(2)* %gep
   call void @foo(i32 %y) nounwind
   ret i32 %y
 }
diff --git a/test/Transforms/MergeFunc/alloca.ll b/test/Transforms/MergeFunc/alloca.ll
new file mode 100644
index 000000000000..d9f66d0911da
--- /dev/null
+++ b/test/Transforms/MergeFunc/alloca.ll
@@ -0,0 +1,33 @@
+; RUN: opt -mergefunc -S < %s | FileCheck %s
+
+;; Make sure that two different sized allocas are not treated as equal.
+
+target datalayout = "e-m:w-p:32:32-i64:64-f80:32-n8:16:32-S32"
+
+%kv1 = type { i32, i32 }
+%kv2 = type { i8 }
+
+
+define void @a(i8 *%f) {
+  %v = alloca %kv1, align 8
+  %f_2 = bitcast i8* %f to void (%kv1 *)*
+  call void %f_2(%kv1 * %v)
+  call void %f_2(%kv1 * %v)
+  call void %f_2(%kv1 * %v)
+  call void %f_2(%kv1 * %v)
+  ret void
+}
+
+; CHECK-LABEL: define void @b
+; CHECK-NOT: call @a
+; CHECK: ret
+
+define void @b(i8 *%f) {
+  %v = alloca %kv2, align 8
+  %f_2 = bitcast i8* %f to void (%kv2 *)*
+  call void %f_2(%kv2 * %v)
+  call void %f_2(%kv2 * %v)
+  call void %f_2(%kv2 * %v)
+  call void %f_2(%kv2 * %v)
+  ret void
+}
diff --git a/test/Transforms/MergeFunc/crash.ll b/test/Transforms/MergeFunc/crash.ll
index 3475e28a6565..73802203dc51 100644
--- a/test/Transforms/MergeFunc/crash.ll
+++ b/test/Transforms/MergeFunc/crash.ll
@@ -21,13 +21,13 @@ define internal i32 @func1(i32* %ptr, { i32, i32 }* nocapture %method) align 2 {
 }
 
 define internal i32 @func10(%.qux.2496* nocapture %this) align 2 {
-  %1 = getelementptr inbounds %.qux.2496* %this, i32 0, i32 1, i32 1
-  %2 = load i32* %1, align 4
+  %1 = getelementptr inbounds %.qux.2496, %.qux.2496* %this, i32 0, i32 1, i32 1
+  %2 = load i32, i32* %1, align 4
   ret i32 %2
 }
 
 define internal i8* @func29(i32* nocapture %this) align 2 {
-  ret i8* getelementptr inbounds ([9 x i8]* @g2, i32 0, i32 0)
+  ret i8* getelementptr inbounds ([9 x i8], [9 x i8]* @g2, i32 0, i32 0)
 }
 
 define internal i32* @func33(%.qux.2585* nocapture %this) align 2 {
@@ -35,12 +35,12 @@ define internal i32* @func33(%.qux.2585* nocapture %this) align 2 {
 }
 
 define internal i32* @func34(%.qux.2585* nocapture %this) align 2 {
-  %1 = getelementptr inbounds %.qux.2585* %this, i32 0
+  %1 = getelementptr inbounds %.qux.2585, %.qux.2585* %this, i32 0
   ret i32* undef
 }
 
 define internal i8* @func35(%.qux.2585* nocapture %this) align 2 {
-  %1 = getelementptr inbounds %.qux.2585* %this, i32 0, i32 2
-  %2 = load i8** %1, align 4
+  %1 = getelementptr inbounds %.qux.2585, %.qux.2585* %this, i32 0, i32 2
+  %2 = load i8*, i8** %1, align 4
   ret i8* %2
 }
diff --git a/test/Transforms/MergeFunc/inttoptr-address-space.ll b/test/Transforms/MergeFunc/inttoptr-address-space.ll
index 2e5e2fcc4865..5f672debd919 100644
--- a/test/Transforms/MergeFunc/inttoptr-address-space.ll
+++ b/test/Transforms/MergeFunc/inttoptr-address-space.ll
@@ -11,8 +11,8 @@ target datalayout = "e-p:32:32:32-p1:16:16:16-i1:8:8-i8:8:8-i16:16:16-i32:32:32-
 
 define internal i32 @func10(%.qux.2496 addrspace(1)* nocapture %this) align 2 {
 bb:
-  %tmp = getelementptr inbounds %.qux.2496 addrspace(1)* %this, i32 0, i32 1, i32 1
-  %tmp1 = load i32 addrspace(1)* %tmp, align 4
+  %tmp = getelementptr inbounds %.qux.2496, %.qux.2496 addrspace(1)* %this, i32 0, i32 1, i32 1
+  %tmp1 = load i32, i32 addrspace(1)* %tmp, align 4
   ret i32 %tmp1
 }
 
@@ -23,7 +23,7 @@ bb:
 ; CHECK: %[[V2:.+]] = bitcast %.qux.2585 addrspace(1)* %{{.*}} to %.qux.2496 addrspace(1)*
 ; CHECK: %[[V3:.+]] = tail call i32 @func10(%.qux.2496 addrspace(1)* %[[V2]])
 ; CHECK: %{{.*}} = inttoptr i32 %[[V3]] to i8*
-  %tmp = getelementptr inbounds %.qux.2585 addrspace(1)* %this, i32 0, i32 2
-  %tmp1 = load i8* addrspace(1)* %tmp, align 4
+  %tmp = getelementptr inbounds %.qux.2585, %.qux.2585 addrspace(1)* %this, i32 0, i32 2
+  %tmp1 = load i8*, i8* addrspace(1)* %tmp, align 4
   ret i8* %tmp1
 }
diff --git a/test/Transforms/MergeFunc/inttoptr.ll b/test/Transforms/MergeFunc/inttoptr.ll
index 86c18a0c122b..0abbf6239a34 100644
--- a/test/Transforms/MergeFunc/inttoptr.ll
+++ b/test/Transforms/MergeFunc/inttoptr.ll
@@ -23,14 +23,14 @@ bb2:                                              ; preds = %bb1
 
 define internal i32 @func10(%.qux.2496* nocapture %this) align 2 {
 bb:
-  %tmp = getelementptr inbounds %.qux.2496* %this, i32 0, i32 1, i32 1
-  %tmp1 = load i32* %tmp, align 4
+  %tmp = getelementptr inbounds %.qux.2496, %.qux.2496* %this, i32 0, i32 1, i32 1
+  %tmp1 = load i32, i32* %tmp, align 4
   ret i32 %tmp1
 }
 
 define internal i8* @func29(i32* nocapture %this) align 2 {
 bb:
-  ret i8* getelementptr inbounds ([9 x i8]* @g2, i32 0, i32 0)
+  ret i8* getelementptr inbounds ([9 x i8], [9 x i8]* @g2, i32 0, i32 0)
 }
 
 define internal i32* @func33(%.qux.2585* nocapture %this) align 2 {
@@ -40,7 +40,7 @@ bb:
 
 define internal i32* @func34(%.qux.2585* nocapture %this) align 2 {
 bb:
-  %tmp = getelementptr inbounds %.qux.2585* %this, i32 0
+  %tmp = getelementptr inbounds %.qux.2585, %.qux.2585* %this, i32 0
   ret i32* undef
 }
 
@@ -50,7 +50,7 @@ bb:
 ; CHECK: %[[V2:.+]] = bitcast %.qux.2585* %{{.*}} to %.qux.2496*
 ; CHECK: %[[V3:.+]] = tail call i32 @func10(%.qux.2496* %[[V2]])
 ; CHECK: %{{.*}} = inttoptr i32 %[[V3]] to i8*
-  %tmp = getelementptr inbounds %.qux.2585* %this, i32 0, i32 2
-  %tmp1 = load i8** %tmp, align 4
+  %tmp = getelementptr inbounds %.qux.2585, %.qux.2585* %this, i32 0, i32 2
+  %tmp1 = load i8*, i8** %tmp, align 4
   ret i8* %tmp1
 }
diff --git a/test/Transforms/MergeFunc/mergefunc-struct-return.ll b/test/Transforms/MergeFunc/mergefunc-struct-return.ll
index d2cbe43da0c4..14db39995e72 100644
--- a/test/Transforms/MergeFunc/mergefunc-struct-return.ll
+++ b/test/Transforms/MergeFunc/mergefunc-struct-return.ll
@@ -13,12 +13,12 @@ declare void @noop()
 define %kv1 @fn1() {
 ; CHECK-LABEL: @fn1(
   %tmp = alloca %kv1
-  %v1 = getelementptr %kv1* %tmp, i32 0, i32 0
+  %v1 = getelementptr %kv1, %kv1* %tmp, i32 0, i32 0
   store i32* null, i32** %v1
-  %v2 = getelementptr %kv1* %tmp, i32 0, i32 0
+  %v2 = getelementptr %kv1, %kv1* %tmp, i32 0, i32 0
   store i32* null, i32** %v2
   call void @noop()
-  %v3 = load %kv1* %tmp
+  %v3 = load %kv1, %kv1* %tmp
   ret %kv1 %v3
 }
 
@@ -29,12 +29,12 @@ define %kv2 @fn2() {
 ; CHECK: %3 = bitcast i32* %2 to i8*
 ; CHECK: %4 = insertvalue %kv2 undef, i8* %3, 0
   %tmp = alloca %kv2
-  %v1 = getelementptr %kv2* %tmp, i32 0, i32 0
+  %v1 = getelementptr %kv2, %kv2* %tmp, i32 0, i32 0
   store i8* null, i8** %v1
-  %v2 = getelementptr %kv2* %tmp, i32 0, i32 0
+  %v2 = getelementptr %kv2, %kv2* %tmp, i32 0, i32 0
   store i8* null, i8** %v2
   call void @noop()
 
-  %v3 = load %kv2* %tmp
+  %v3 = load %kv2, %kv2* %tmp
   ret %kv2 %v3
 }
diff --git a/test/Transforms/MergeFunc/ranges.ll b/test/Transforms/MergeFunc/ranges.ll
index d3e4d942a9d5..46a0c76cc7d1 100644
--- a/test/Transforms/MergeFunc/ranges.ll
+++ b/test/Transforms/MergeFunc/ranges.ll
@@ -1,31 +1,31 @@
 ; RUN: opt -mergefunc -S < %s | FileCheck %s
 define i1 @cmp_with_range(i8*, i8*) {
-  %v1 = load i8* %0, !range !0
-  %v2 = load i8* %1, !range !0
+  %v1 = load i8, i8* %0, !range !0
+  %v2 = load i8, i8* %1, !range !0
   %out = icmp eq i8 %v1, %v2
   ret i1 %out
 }
 
 define i1 @cmp_no_range(i8*, i8*) {
 ; CHECK-LABEL: @cmp_no_range
-; CHECK-NEXT  %v1 = load i8* %0
-; CHECK-NEXT  %v2 = load i8* %1
+; CHECK-NEXT  %v1 = load i8, i8* %0
+; CHECK-NEXT  %v2 = load i8, i8* %1
 ; CHECK-NEXT  %out = icmp eq i8 %v1, %v2
 ; CHECK-NEXT  ret i1 %out
-  %v1 = load i8* %0
-  %v2 = load i8* %1
+  %v1 = load i8, i8* %0
+  %v2 = load i8, i8* %1
   %out = icmp eq i8 %v1, %v2
   ret i1 %out
 }
 
 define i1 @cmp_different_range(i8*, i8*) {
 ; CHECK-LABEL: @cmp_different_range
-; CHECK-NEXT:  %v1 = load i8* %0, !range !1
-; CHECK-NEXT:  %v2 = load i8* %1, !range !1
+; CHECK-NEXT:  %v1 = load i8, i8* %0, !range !1
+; CHECK-NEXT:  %v2 = load i8, i8* %1, !range !1
 ; CHECK-NEXT:  %out = icmp eq i8 %v1, %v2
 ; CHECK-NEXT:  ret i1 %out
-  %v1 = load i8* %0, !range !1
-  %v2 = load i8* %1, !range !1
+  %v1 = load i8, i8* %0, !range !1
+  %v2 = load i8, i8* %1, !range !1
   %out = icmp eq i8 %v1, %v2
   ret i1 %out
 }
@@ -33,8 +33,8 @@ define i1 @cmp_different_range(i8*, i8*) {
 define i1 @cmp_with_same_range(i8*, i8*) {
 ; CHECK-LABEL: @cmp_with_same_range
 ; CHECK: tail call i1 @cmp_with_range
-  %v1 = load i8* %0, !range !0
-  %v2 = load i8* %1, !range !0
+  %v1 = load i8, i8* %0, !range !0
+  %v2 = load i8, i8* %1, !range !0
   %out = icmp eq i8 %v1, %v2
   ret i1 %out
 }
diff --git a/test/Transforms/MergeFunc/vector-GEP-crash.ll b/test/Transforms/MergeFunc/vector-GEP-crash.ll
index a1eefa0bc983..d0e86ebf54b9 100644
--- a/test/Transforms/MergeFunc/vector-GEP-crash.ll
+++ b/test/Transforms/MergeFunc/vector-GEP-crash.ll
@@ -2,11 +2,11 @@
 ; This used to cause a crash when compairing the GEPs
 
 define void @foo(<2 x i64*>) {
-  %tmp = getelementptr <2 x i64*> %0, <2 x i64> <i64 0, i64 0>
+  %tmp = getelementptr i64, <2 x i64*> %0, <2 x i64> <i64 0, i64 0>
   ret void
 }
 
 define void @bar(<2 x i64*>) {
-  %tmp = getelementptr <2 x i64*> %0, <2 x i64> <i64 0, i64 0>
+  %tmp = getelementptr i64, <2 x i64*> %0, <2 x i64> <i64 0, i64 0>
   ret void
 }
diff --git a/test/Transforms/MergeFunc/vector.ll b/test/Transforms/MergeFunc/vector.ll
index 56f74e65c60d..ef1375311e60 100644
--- a/test/Transforms/MergeFunc/vector.ll
+++ b/test/Transforms/MergeFunc/vector.ll
@@ -22,7 +22,7 @@ target triple = "x86_64-unknown-linux-gnu"
 define linkonce_odr void @_ZNSt6vectorIlSaIlEED1Ev(%"class.std::vector"* nocapture %this) unnamed_addr align 2 {
 entry:
   %tmp2.i.i = bitcast %"class.std::vector"* %this to i64**
-  %tmp3.i.i = load i64** %tmp2.i.i, align 8
+  %tmp3.i.i = load i64*, i64** %tmp2.i.i, align 8
   %tobool.i.i.i = icmp eq i64* %tmp3.i.i, null
   br i1 %tobool.i.i.i, label %_ZNSt6vectorIlSaIlEED2Ev.exit, label %if.then.i.i.i
 
@@ -40,7 +40,7 @@ declare i32 @__cxa_atexit(void (i8*)*, i8*, i8*)
 define linkonce_odr void @_ZNSt6vectorIPvSaIS0_EED1Ev(%"class.std::vector"* nocapture %this) unnamed_addr align 2 {
 entry:
   %tmp2.i.i = bitcast %"class.std::vector"* %this to i8***
-  %tmp3.i.i = load i8*** %tmp2.i.i, align 8
+  %tmp3.i.i = load i8**, i8*** %tmp2.i.i, align 8
   %tobool.i.i.i = icmp eq i8** %tmp3.i.i, null
   br i1 %tobool.i.i.i, label %_ZNSt6vectorIPvSaIS0_EED2Ev.exit, label %if.then.i.i.i
 
diff --git a/test/Transforms/MetaRenamer/metarenamer.ll b/test/Transforms/MetaRenamer/metarenamer.ll
index 4010f31ff772..e126bed9b232 100644
--- a/test/Transforms/MetaRenamer/metarenamer.ll
+++ b/test/Transforms/MetaRenamer/metarenamer.ll
@@ -24,14 +24,14 @@ define i32 @func_3_xxx() nounwind uwtable ssp {
 
 define void @func_4_xxx(%struct.foo_xxx* sret %agg.result) nounwind uwtable ssp {
   %1 = alloca %struct.foo_xxx, align 8
-  %2 = getelementptr inbounds %struct.foo_xxx* %1, i32 0, i32 0
+  %2 = getelementptr inbounds %struct.foo_xxx, %struct.foo_xxx* %1, i32 0, i32 0
   store i32 1, i32* %2, align 4
-  %3 = getelementptr inbounds %struct.foo_xxx* %1, i32 0, i32 1
+  %3 = getelementptr inbounds %struct.foo_xxx, %struct.foo_xxx* %1, i32 0, i32 1
   store float 2.000000e+00, float* %3, align 4
-  %4 = getelementptr inbounds %struct.foo_xxx* %1, i32 0, i32 2
-  %5 = getelementptr inbounds %struct.bar_xxx* %4, i32 0, i32 0
+  %4 = getelementptr inbounds %struct.foo_xxx, %struct.foo_xxx* %1, i32 0, i32 2
+  %5 = getelementptr inbounds %struct.bar_xxx, %struct.bar_xxx* %4, i32 0, i32 0
   store i32 3, i32* %5, align 4
-  %6 = getelementptr inbounds %struct.bar_xxx* %4, i32 0, i32 1
+  %6 = getelementptr inbounds %struct.bar_xxx, %struct.bar_xxx* %4, i32 0, i32 1
   store double 4.000000e+00, double* %6, align 8
   %7 = bitcast %struct.foo_xxx* %agg.result to i8*
   %8 = bitcast %struct.foo_xxx* %1 to i8*
@@ -59,7 +59,7 @@ define i32 @func_5_xxx(i32 %arg_1_xxx, i32 %arg_2_xxx, i32 %arg_3_xxx, i32 %arg_
   br label %5
 
 ; <label>:5                                       ; preds = %9, %0
-  %6 = load i32* %i, align 4
+  %6 = load i32, i32* %i, align 4
   %7 = icmp slt i32 %6, 10
   br i1 %7, label %8, label %12
 
@@ -67,24 +67,24 @@ define i32 @func_5_xxx(i32 %arg_1_xxx, i32 %arg_2_xxx, i32 %arg_3_xxx, i32 %arg_
   br label %9
 
 ; <label>:9                                       ; preds = %8
-  %10 = load i32* %i, align 4
+  %10 = load i32, i32* %i, align 4
   %11 = add nsw i32 %10, 1
   store i32 %11, i32* %i, align 4
   br label %5
 
 ; <label>:12                                      ; preds = %5
-  %13 = load i32* %local_1_xxx, align 4
-  %14 = load i32* %1, align 4
+  %13 = load i32, i32* %local_1_xxx, align 4
+  %14 = load i32, i32* %1, align 4
   %15 = add nsw i32 %13, %14
-  %16 = load i32* %local_2_xxx, align 4
+  %16 = load i32, i32* %local_2_xxx, align 4
   %17 = add nsw i32 %15, %16
-  %18 = load i32* %2, align 4
+  %18 = load i32, i32* %2, align 4
   %19 = add nsw i32 %17, %18
-  %20 = load i32* @func_5_xxx.static_local_3_xxx, align 4
+  %20 = load i32, i32* @func_5_xxx.static_local_3_xxx, align 4
   %21 = add nsw i32 %19, %20
-  %22 = load i32* %3, align 4
+  %22 = load i32, i32* %3, align 4
   %23 = add nsw i32 %21, %22
-  %24 = load i32* %4, align 4
+  %24 = load i32, i32* %4, align 4
   %25 = add nsw i32 %23, %24
   ret i32 %25
 }
diff --git a/test/Transforms/NaryReassociate/NVPTX/lit.local.cfg b/test/Transforms/NaryReassociate/NVPTX/lit.local.cfg
new file mode 100644
index 000000000000..2cb98eb371b2
--- /dev/null
+++ b/test/Transforms/NaryReassociate/NVPTX/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'NVPTX' in config.root.targets:
+    config.unsupported = True
diff --git a/test/Transforms/NaryReassociate/NVPTX/nary-gep.ll b/test/Transforms/NaryReassociate/NVPTX/nary-gep.ll
new file mode 100644
index 000000000000..a620c98ec18f
--- /dev/null
+++ b/test/Transforms/NaryReassociate/NVPTX/nary-gep.ll
@@ -0,0 +1,77 @@
+; RUN: opt < %s -nary-reassociate -S | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+target triple = "nvptx64-unknown-unknown"
+
+declare void @foo(float*)
+
+; foo(&a[i]);
+; foo(&a[i + j]);
+;   =>
+; t = &a[i];
+; foo(t);
+; foo(t + j);
+define void @reassociate_gep(float* %a, i64 %i, i64 %j) {
+; CHECK-LABEL: @reassociate_gep(
+  %1 = add i64 %i, %j
+  %2 = getelementptr float, float* %a, i64 %i
+; CHECK: [[t1:[^ ]+]] = getelementptr float, float* %a, i64 %i
+  call void @foo(float* %2)
+; CHECK: call void @foo(float* [[t1]])
+  %3 = getelementptr float, float* %a, i64 %1
+; CHECK: [[t2:[^ ]+]] = getelementptr float, float* [[t1]], i64 %j
+  call void @foo(float* %3)
+; CHECK: call void @foo(float* [[t2]])
+  ret void
+}
+
+; foo(&a[sext(j)]);
+; foo(&a[sext(i +nsw j)]);
+;   =>
+; t = &a[sext(j)];
+; foo(t);
+; foo(t + sext(i));
+define void @reassociate_gep_nsw(float* %a, i32 %i, i32 %j) {
+; CHECK-LABEL: @reassociate_gep_nsw(
+  %1 = add nsw i32 %i, %j
+  %idxprom.1 = sext i32 %1 to i64
+  %idxprom.j = sext i32 %j to i64
+  %2 = getelementptr float, float* %a, i64 %idxprom.j
+; CHECK: [[t1:[^ ]+]] = getelementptr float, float* %a, i64 %idxprom.j
+  call void @foo(float* %2)
+; CHECK: call void @foo(float* [[t1]])
+  %3 = getelementptr float, float* %a, i64 %idxprom.1
+; CHECK: [[sexti:[^ ]+]] = sext i32 %i to i64
+; CHECK: [[t2:[^ ]+]] = getelementptr float, float* [[t1]], i64 [[sexti]]
+  call void @foo(float* %3)
+; CHECK: call void @foo(float* [[t2]])
+  ret void
+}
+
+; Do not split the second GEP because sext(i + j) != sext(i) + sext(j).
+define void @reassociate_gep_no_nsw(float* %a, i32 %i, i32 %j) {
+; CHECK-LABEL: @reassociate_gep_no_nsw(
+  %1 = add i32 %i, %j
+  %2 = getelementptr float, float* %a, i32 %j
+; CHECK: getelementptr float, float* %a, i32 %j
+  call void @foo(float* %2)
+  %3 = getelementptr float, float* %a, i32 %1
+; CHECK: getelementptr float, float* %a, i32 %1
+  call void @foo(float* %3)
+  ret void
+}
+
+define void @reassociate_gep_128(float* %a, i128 %i, i128 %j) {
+; CHECK-LABEL: @reassociate_gep_128(
+  %1 = add i128 %i, %j
+  %2 = getelementptr float, float* %a, i128 %i
+; CHECK: [[t1:[^ ]+]] = getelementptr float, float* %a, i128 %i
+  call void @foo(float* %2)
+; CHECK: call void @foo(float* [[t1]])
+  %3 = getelementptr float, float* %a, i128 %1
+; CHECK: [[truncj:[^ ]+]] = trunc i128 %j to i64
+; CHECK: [[t2:[^ ]+]] = getelementptr float, float* [[t1]], i64 [[truncj]]
+  call void @foo(float* %3)
+; CHECK: call void @foo(float* [[t2]])
+  ret void
+}
diff --git a/test/Transforms/NaryReassociate/NVPTX/nary-slsr.ll b/test/Transforms/NaryReassociate/NVPTX/nary-slsr.ll
new file mode 100644
index 000000000000..2360fe7f1b78
--- /dev/null
+++ b/test/Transforms/NaryReassociate/NVPTX/nary-slsr.ll
@@ -0,0 +1,47 @@
+; RUN: opt < %s -slsr -nary-reassociate -S | FileCheck %s
+; RUN: llc < %s -march=nvptx64 -mcpu=sm_20 | FileCheck %s --check-prefix=PTX
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+; foo((a + b) + c);
+; foo((a + b * 2) + c);
+; foo((a + b * 3) + c);
+;   =>
+; abc = (a + b) + c;
+; foo(abc);
+; ab2c = abc + b;
+; foo(ab2c);
+; ab3c = ab2c + b;
+; foo(ab3c);
+define void @nary_reassociate_after_slsr(i32 %a, i32 %b, i32 %c) {
+; CHECK-LABEL: @nary_reassociate_after_slsr(
+; PTX-LABEL: .visible .func nary_reassociate_after_slsr(
+; PTX: ld.param.u32 [[b:%r[0-9]+]], [nary_reassociate_after_slsr_param_1];
+  %ab = add i32 %a, %b
+  %abc = add i32 %ab, %c
+  call void @foo(i32 %abc)
+; CHECK: call void @foo(i32 %abc)
+; PTX: st.param.b32 [param0+0], [[abc:%r[0-9]+]];
+
+  %b2 = shl i32 %b, 1
+  %ab2 = add i32 %a, %b2
+  %ab2c = add i32 %ab2, %c
+; CHECK-NEXT: %ab2c = add i32 %abc, %b
+; PTX: add.s32 [[ab2c:%r[0-9]+]], [[abc]], [[b]]
+  call void @foo(i32 %ab2c)
+; CHECK-NEXT: call void @foo(i32 %ab2c)
+; PTX: st.param.b32 [param0+0], [[ab2c]];
+
+  %b3 = mul i32 %b, 3
+  %ab3 = add i32 %a, %b3
+  %ab3c = add i32 %ab3, %c
+; CHECK-NEXT: %ab3c = add i32 %ab2c, %b
+; PTX: add.s32 [[ab3c:%r[0-9]+]], [[ab2c]], [[b]]
+  call void @foo(i32 %ab3c)
+; CHECK-NEXT: call void @foo(i32 %ab3c)
+; PTX: st.param.b32 [param0+0], [[ab3c]];
+
+  ret void
+}
+
+declare void @foo(i32)
diff --git a/test/Transforms/NaryReassociate/nary-add.ll b/test/Transforms/NaryReassociate/nary-add.ll
new file mode 100644
index 000000000000..b3093ff6ecd6
--- /dev/null
+++ b/test/Transforms/NaryReassociate/nary-add.ll
@@ -0,0 +1,209 @@
+; RUN: opt < %s -nary-reassociate -S | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+declare void @foo(i32)
+
+; foo(a + c);
+; foo((a + (b + c));
+;   =>
+; t = a + c;
+; foo(t);
+; foo(t + b);
+define void @left_reassociate(i32 %a, i32 %b, i32 %c) {
+; CHECK-LABEL: @left_reassociate(
+  %1 = add i32 %a, %c
+; CHECK: [[BASE:%[a-zA-Z0-9]+]] = add i32 %a, %c
+  call void @foo(i32 %1)
+  %2 = add i32 %b, %c
+  %3 = add i32 %a, %2
+; CHECK: add i32 [[BASE]], %b
+  call void @foo(i32 %3)
+  ret void
+}
+
+; foo(a + c);
+; foo((a + b) + c);
+;   =>
+; t = a + c;
+; foo(t);
+; foo(t + b);
+define void @right_reassociate(i32 %a, i32 %b, i32 %c) {
+; CHECK-LABEL: @right_reassociate(
+  %1 = add i32 %a, %c
+; CHECK: [[BASE:%[a-zA-Z0-9]+]] = add i32 %a, %c
+  call void @foo(i32 %1)
+  %2 = add i32 %a, %b
+  %3 = add i32 %2, %c
+; CHECK: add i32 [[BASE]], %b
+  call void @foo(i32 %3)
+  ret void
+}
+
+; t1 = a + c;
+; foo(t1);
+; t2 = a + b;
+; foo(t2);
+; t3 = t2 + c;
+; foo(t3);
+;
+; Do not rewrite t3 into t1 + b because t2 is used elsewhere and is likely free.
+define void @no_reassociate(i32 %a, i32 %b, i32 %c) {
+; CHECK-LABEL: @no_reassociate(
+  %1 = add i32 %a, %c
+; CHECK: add i32 %a, %c
+  call void @foo(i32 %1)
+  %2 = add i32 %a, %b
+; CHECK: add i32 %a, %b
+  call void @foo(i32 %2)
+  %3 = add i32 %2, %c
+; CHECK: add i32 %2, %c
+  call void @foo(i32 %3)
+  ret void
+}
+
+; if (p1)
+;   foo(a + c);
+; if (p2)
+;   foo(a + c);
+; if (p3)
+;   foo((a + b) + c);
+;
+; No action because (a + c) does not dominate ((a + b) + c).
+define void @conditional(i1 %p1, i1 %p2, i1 %p3, i32 %a, i32 %b, i32 %c) {
+; CHECK-LABEL: @conditional(
+entry:
+  br i1 %p1, label %then1, label %branch1
+
+then1:
+  %0 = add i32 %a, %c
+; CHECK: add i32 %a, %c
+  call void @foo(i32 %0)
+  br label %branch1
+
+branch1:
+  br i1 %p2, label %then2, label %branch2
+
+then2:
+  %1 = add i32 %a, %c
+; CHECK: add i32 %a, %c
+  call void @foo(i32 %1)
+  br label %branch2
+
+branch2:
+  br i1 %p3, label %then3, label %return
+
+then3:
+  %2 = add i32 %a, %b
+; CHECK: %2 = add i32 %a, %b
+  %3 = add i32 %2, %c
+; CHECK: add i32 %2, %c
+  call void @foo(i32 %3)
+  br label %return
+
+return:
+  ret void
+}
+
+; This test involves more conditional reassociation candidates. It exercises
+; the stack optimization in tryReassociatedAdd that pops the candidates that
+; do not dominate the current instruction.
+;
+;       def1
+;      cond1
+;      /  \
+;     /    \
+;   cond2  use2
+;   /  \
+;  /    \
+; def2  def3
+;      cond3
+;       /  \
+;      /    \
+;    def4   use1
+;
+; NaryReassociate should match use1 with def3, and use2 with def1.
+define void @conditional2(i32 %a, i32 %b, i32 %c, i1 %cond1, i1 %cond2, i1 %cond3) {
+entry:
+  %def1 = add i32 %a, %b
+  br i1 %cond1, label %bb1, label %bb6
+bb1:
+  br i1 %cond2, label %bb2, label %bb3
+bb2:
+  %def2 = add i32 %a, %b
+  call void @foo(i32 %def2)
+  ret void
+bb3:
+  %def3 = add i32 %a, %b
+  br i1 %cond3, label %bb4, label %bb5
+bb4:
+  %def4 = add i32 %a, %b
+  call void @foo(i32 %def4)
+  ret void
+bb5:
+  %0 = add i32 %a, %c
+  %1 = add i32 %0, %b
+; CHECK: [[t1:%[0-9]+]] = add i32 %def3, %c
+  call void @foo(i32 %1) ; foo((a + c) + b);
+; CHECK-NEXT: call void @foo(i32 [[t1]])
+  ret void
+bb6:
+  %2 = add i32 %a, %c
+  %3 = add i32 %2, %b
+; CHECK: [[t2:%[0-9]+]] = add i32 %def1, %c
+  call void @foo(i32 %3) ; foo((a + c) + b);
+; CHECK-NEXT: call void @foo(i32 [[t2]])
+  ret void
+}
+
+; foo((a + b) + c)
+; foo(((a + d) + b) + c)
+;   =>
+; t = (a + b) + c;
+; foo(t);
+; foo(t + d);
+define void @quaternary(i32 %a, i32 %b, i32 %c, i32 %d) {
+; CHECK-LABEL: @quaternary(
+  %1 = add i32 %a, %b
+  %2 = add i32 %1, %c
+  call void @foo(i32 %2)
+; CHECK: call void @foo(i32 [[TMP1:%[a-zA-Z0-9]]])
+  %3 = add i32 %a, %d
+  %4 = add i32 %3, %b
+  %5 = add i32 %4, %c
+; CHECK: [[TMP2:%[a-zA-Z0-9]]] = add i32 [[TMP1]], %d
+  call void @foo(i32 %5)
+; CHECK: call void @foo(i32 [[TMP2]]
+  ret void
+}
+
+define void @iterative(i32 %a, i32 %b, i32 %c) {
+  %ab = add i32 %a, %b
+  %abc = add i32 %ab, %c
+  call void @foo(i32 %abc)
+
+  %ab2 = add i32 %ab, %b
+  %ab2c = add i32 %ab2, %c
+; CHECK: %ab2c = add i32 %abc, %b
+  call void @foo(i32 %ab2c)
+; CHECK-NEXT: call void @foo(i32 %ab2c)
+
+  %ab3 = add i32 %ab2, %b
+  %ab3c = add i32 %ab3, %c
+; CHECK-NEXT: %ab3c = add i32 %ab2c, %b
+  call void @foo(i32 %ab3c)
+; CHECK-NEXT: call void @foo(i32 %ab3c)
+
+  ret void
+}
+
+define void @avoid_infinite_loop(i32 %a, i32 %b) {
+; CHECK-LABEL: @avoid_infinite_loop
+  %ab = add i32 %a, %b
+; CHECK-NEXT: %ab
+  %ab2 = add i32 %ab, %b
+; CHECK-NEXT: %ab2
+  call void @foo(i32 %ab2)
+; CHECK-NEXT: @foo(i32 %ab2)
+  ret void
+}
diff --git a/test/Transforms/ObjCARC/allocas.ll b/test/Transforms/ObjCARC/allocas.ll
index 7b671df70649..ac26a097d49d 100644
--- a/test/Transforms/ObjCARC/allocas.ll
+++ b/test/Transforms/ObjCARC/allocas.ll
@@ -56,7 +56,7 @@ entry:
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %A, align 8
-  %y = load i8** %A
+  %y = load i8*, i8** %A
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -74,11 +74,11 @@ entry:
 define void @test1b(i8* %x) {
 entry:
   %A = alloca i8*
-  %gep = getelementptr i8** %A, i32 0
+  %gep = getelementptr i8*, i8** %A, i32 0
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %gep, align 8
-  %y = load i8** %A
+  %y = load i8*, i8** %A
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -97,11 +97,11 @@ entry:
 define void @test1c(i8* %x) {
 entry:
   %A = alloca i8*, i32 3
-  %gep = getelementptr i8** %A, i32 2
+  %gep = getelementptr i8*, i8** %A, i32 2
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %gep, align 8
-  %y = load i8** %gep
+  %y = load i8*, i8** %gep
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -131,11 +131,11 @@ use_allocaB:
 
 exit:
   %A = phi i8** [ %allocaA, %use_allocaA ], [ %allocaB, %use_allocaB ]
-  %gep = getelementptr i8** %A, i32 0
+  %gep = getelementptr i8*, i8** %A, i32 0
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %gep, align 8
-  %y = load i8** %gep
+  %y = load i8*, i8** %gep
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -164,11 +164,11 @@ use_allocaB:
 
 exit:
   %A = phi i8** [ %allocaA, %use_allocaA ], [ %allocaB, %use_allocaB ]
-  %gep = getelementptr i8** %A, i32 2
+  %gep = getelementptr i8*, i8** %A, i32 2
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %gep, align 8
-  %y = load i8** %gep
+  %y = load i8*, i8** %gep
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -191,7 +191,7 @@ entry:
   tail call i8* @objc_retain(i8* %x)
   tail call i8* @objc_retain(i8* %x)
   store i8* %x, i8** %A, align 8
-  %y = load i8** %A
+  %y = load i8*, i8** %A
   call void @use_alloca(i8** %A)
   call void @objc_release(i8* %y), !clang.imprecise_release !0
   call void @use_pointer(i8* %x)
@@ -214,7 +214,7 @@ define void @test2a(i8* %x) {
 entry:
   %A = alloca i8*
   store i8* %x, i8** %A, align 8
-  %y = load i8** %A
+  %y = load i8*, i8** %A
   br label %bb1
 
 bb1:
@@ -243,10 +243,10 @@ bb3:
 define void @test2b(i8* %x) {
 entry:
   %A = alloca i8*
-  %gep1 = getelementptr i8** %A, i32 0
+  %gep1 = getelementptr i8*, i8** %A, i32 0
   store i8* %x, i8** %gep1, align 8
-  %gep2 = getelementptr i8** %A, i32 0
-  %y = load i8** %gep2
+  %gep2 = getelementptr i8*, i8** %A, i32 0
+  %y = load i8*, i8** %gep2
   br label %bb1
 
 bb1:
@@ -275,10 +275,10 @@ bb3:
 define void @test2c(i8* %x) {
 entry:
   %A = alloca i8*, i32 3
-  %gep1 = getelementptr i8** %A, i32 2
+  %gep1 = getelementptr i8*, i8** %A, i32 2
   store i8* %x, i8** %gep1, align 8
-  %gep2 = getelementptr i8** %A, i32 2
-  %y = load i8** %gep2
+  %gep2 = getelementptr i8*, i8** %A, i32 2
+  %y = load i8*, i8** %gep2
   tail call i8* @objc_retain(i8* %x)
   br label %bb1
 
@@ -311,18 +311,18 @@ entry:
 
 bb1:
   %Abb1 = alloca i8*, i32 3
-  %gepbb11 = getelementptr i8** %Abb1, i32 2
+  %gepbb11 = getelementptr i8*, i8** %Abb1, i32 2
   store i8* %x, i8** %gepbb11, align 8
-  %gepbb12 = getelementptr i8** %Abb1, i32 2
-  %ybb1 = load i8** %gepbb12
+  %gepbb12 = getelementptr i8*, i8** %Abb1, i32 2
+  %ybb1 = load i8*, i8** %gepbb12
   br label %bb3
 
 bb2:
   %Abb2 = alloca i8*, i32 4
-  %gepbb21 = getelementptr i8** %Abb2, i32 2
+  %gepbb21 = getelementptr i8*, i8** %Abb2, i32 2
   store i8* %x, i8** %gepbb21, align 8
-  %gepbb22 = getelementptr i8** %Abb2, i32 2
-  %ybb2 = load i8** %gepbb22
+  %gepbb22 = getelementptr i8*, i8** %Abb2, i32 2
+  %ybb2 = load i8*, i8** %gepbb22
   br label %bb3
 
 bb3:
@@ -369,44 +369,44 @@ entry:
   %call1 = call i8* @returner()
   %tmp0 = tail call i8* @objc_retainAutoreleasedReturnValue(i8* %call1)
 
-  %objs.begin = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 0
+  %objs.begin = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 0
   tail call i8* @objc_retain(i8* %call1)
   store i8* %call1, i8** %objs.begin, align 8
-  %objs.elt = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 1
+  %objs.elt = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 1
   tail call i8* @objc_retain(i8* %call1)
   store i8* %call1, i8** %objs.elt
 
   %call2 = call i8* @returner1()
   %call3 = call i8* @returner2()
-  %keys.begin = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 0
+  %keys.begin = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 0
   tail call i8* @objc_retain(i8* %call2)
   store i8* %call2, i8** %keys.begin, align 8
-  %keys.elt = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 1
+  %keys.elt = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 1
   tail call i8* @objc_retain(i8* %call3)
   store i8* %call3, i8** %keys.elt  
   
-  %gep = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 2
+  %gep = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 2
   br label %arraydestroy.body
 
 arraydestroy.body:
   %arraydestroy.elementPast = phi i8** [ %gep, %entry ], [ %arraydestroy.element, %arraydestroy.body ]
-  %arraydestroy.element = getelementptr inbounds i8** %arraydestroy.elementPast, i64 -1
-  %destroy_tmp = load i8** %arraydestroy.element, align 8
+  %arraydestroy.element = getelementptr inbounds i8*, i8** %arraydestroy.elementPast, i64 -1
+  %destroy_tmp = load i8*, i8** %arraydestroy.element, align 8
   call void @objc_release(i8* %destroy_tmp), !clang.imprecise_release !0
-  %objs_ptr = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 0
+  %objs_ptr = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 0
   %arraydestroy.cmp = icmp eq i8** %arraydestroy.element, %objs_ptr
   br i1 %arraydestroy.cmp, label %arraydestroy.done, label %arraydestroy.body
 
 arraydestroy.done:
-  %gep1 = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 2
+  %gep1 = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 2
   br label %arraydestroy.body1
 
 arraydestroy.body1:
   %arraydestroy.elementPast1 = phi i8** [ %gep1, %arraydestroy.done ], [ %arraydestroy.element1, %arraydestroy.body1 ]
-  %arraydestroy.element1 = getelementptr inbounds i8** %arraydestroy.elementPast1, i64 -1
-  %destroy_tmp1 = load i8** %arraydestroy.element1, align 8
+  %arraydestroy.element1 = getelementptr inbounds i8*, i8** %arraydestroy.elementPast1, i64 -1
+  %destroy_tmp1 = load i8*, i8** %arraydestroy.element1, align 8
   call void @objc_release(i8* %destroy_tmp1), !clang.imprecise_release !0
-  %keys_ptr = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 0
+  %keys_ptr = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 0
   %arraydestroy.cmp1 = icmp eq i8** %arraydestroy.element1, %keys_ptr
   br i1 %arraydestroy.cmp1, label %arraydestroy.done1, label %arraydestroy.body1
 
@@ -448,44 +448,44 @@ entry:
   %tmp0 = tail call i8* @objc_retainAutoreleasedReturnValue(i8* %call1)
   %tmp1 = tail call i8* @objc_retain(i8* %call1)
 
-  %objs.begin = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 0
+  %objs.begin = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 0
   tail call i8* @objc_retain(i8* %call1)
   store i8* %call1, i8** %objs.begin, align 8
-  %objs.elt = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 1
+  %objs.elt = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 1
   tail call i8* @objc_retain(i8* %call1)
   store i8* %call1, i8** %objs.elt
 
   %call2 = call i8* @returner1()
   %call3 = call i8* @returner2()
-  %keys.begin = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 0
+  %keys.begin = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 0
   tail call i8* @objc_retain(i8* %call2)
   store i8* %call2, i8** %keys.begin, align 8
-  %keys.elt = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 1
+  %keys.elt = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 1
   tail call i8* @objc_retain(i8* %call3)
   store i8* %call3, i8** %keys.elt  
   
-  %gep = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 2
+  %gep = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 2
   br label %arraydestroy.body
 
 arraydestroy.body:
   %arraydestroy.elementPast = phi i8** [ %gep, %entry ], [ %arraydestroy.element, %arraydestroy.body ]
-  %arraydestroy.element = getelementptr inbounds i8** %arraydestroy.elementPast, i64 -1
-  %destroy_tmp = load i8** %arraydestroy.element, align 8
+  %arraydestroy.element = getelementptr inbounds i8*, i8** %arraydestroy.elementPast, i64 -1
+  %destroy_tmp = load i8*, i8** %arraydestroy.element, align 8
   call void @objc_release(i8* %destroy_tmp), !clang.imprecise_release !0
-  %objs_ptr = getelementptr inbounds [2 x i8*]* %objs, i64 0, i64 0
+  %objs_ptr = getelementptr inbounds [2 x i8*], [2 x i8*]* %objs, i64 0, i64 0
   %arraydestroy.cmp = icmp eq i8** %arraydestroy.element, %objs_ptr
   br i1 %arraydestroy.cmp, label %arraydestroy.done, label %arraydestroy.body
 
 arraydestroy.done:
-  %gep1 = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 2
+  %gep1 = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 2
   br label %arraydestroy.body1
 
 arraydestroy.body1:
   %arraydestroy.elementPast1 = phi i8** [ %gep1, %arraydestroy.done ], [ %arraydestroy.element1, %arraydestroy.body1 ]
-  %arraydestroy.element1 = getelementptr inbounds i8** %arraydestroy.elementPast1, i64 -1
-  %destroy_tmp1 = load i8** %arraydestroy.element1, align 8
+  %arraydestroy.element1 = getelementptr inbounds i8*, i8** %arraydestroy.elementPast1, i64 -1
+  %destroy_tmp1 = load i8*, i8** %arraydestroy.element1, align 8
   call void @objc_release(i8* %destroy_tmp1), !clang.imprecise_release !0
-  %keys_ptr = getelementptr inbounds [2 x i8*]* %keys, i64 0, i64 0
+  %keys_ptr = getelementptr inbounds [2 x i8*], [2 x i8*]* %keys, i64 0, i64 0
   %arraydestroy.cmp1 = icmp eq i8** %arraydestroy.element1, %keys_ptr
   br i1 %arraydestroy.cmp1, label %arraydestroy.done1, label %arraydestroy.body1
 
diff --git a/test/Transforms/ObjCARC/arc-annotations.ll b/test/Transforms/ObjCARC/arc-annotations.ll
deleted file mode 100644
index c0ce44f284ca..000000000000
--- a/test/Transforms/ObjCARC/arc-annotations.ll
+++ /dev/null
@@ -1,83 +0,0 @@
-; This file consists of various tests which ensure that the objc-arc-annotations
-; are working correctly. In the future, I will use this in other lit tests to
-; check the data flow analysis of ARC.
-
-; REQUIRES: asserts
-; RUN: opt -S -objc-arc -enable-objc-arc-annotations < %s | FileCheck %s
-
-declare i8* @objc_retain(i8*)
-declare i8* @objc_retainAutoreleasedReturnValue(i8*)
-declare void @objc_release(i8*)
-declare i8* @objc_autorelease(i8*)
-declare i8* @objc_autoreleaseReturnValue(i8*)
-declare void @objc_autoreleasePoolPop(i8*)
-declare i8* @objc_autoreleasePoolPush()
-declare i8* @objc_retainBlock(i8*)
-
-declare i8* @objc_retainedObject(i8*)
-declare i8* @objc_unretainedObject(i8*)
-declare i8* @objc_unretainedPointer(i8*)
-
-declare void @use_pointer(i8*)
-declare void @callee()
-declare void @callee_fnptr(void ()*)
-declare void @invokee()
-declare i8* @returner()
-
-; Simple retain+release pair deletion, with some intervening control
-; flow and harmless instructions.
-
-; CHECK-LABEL: define void @test0(
-; CHECK: entry:
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbstart(i8** @x, i8** @S_None)
-; CHECK:   %0 = tail call i8* @objc_retain(i8* %a) #0, !llvm.arc.annotation.bottomup ![[ANN0:[0-9]+]], !llvm.arc.annotation.topdown ![[ANN1:[0-9]+]]
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbend(i8** @x, i8** @S_Use)
-; CHECK:   call void @llvm.arc.annotation.topdown.bbend(i8** @x, i8** @S_Retain)
-; CHECK: t:
-; CHECK:   call void @llvm.arc.annotation.topdown.bbstart(i8** @x, i8** @S_Retain)
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbstart(i8** @x, i8** @S_Use)
-; CHECK:   store float 2.000000e+00, float* %b, !llvm.arc.annotation.bottomup ![[ANN2:[0-9]+]]
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbend(i8** @x, i8** @S_Release)
-; CHECK:   call void @llvm.arc.annotation.topdown.bbend(i8** @x, i8** @S_Retain)
-; CHECK: f:
-; CHECK:   call void @llvm.arc.annotation.topdown.bbstart(i8** @x, i8** @S_Retain)
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbstart(i8** @x, i8** @S_Use)
-; CHECK:   store i32 7, i32* %x, !llvm.arc.annotation.bottomup ![[ANN2]]
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbend(i8** @x, i8** @S_Release)
-; CHECK:   call void @llvm.arc.annotation.topdown.bbend(i8** @x, i8** @S_Retain)
-; CHECK: return:
-; CHECK:   call void @llvm.arc.annotation.topdown.bbstart(i8** @x, i8** @S_Retain)
-; CHECK:   call void @llvm.arc.annotation.bottomup.bbstart(i8** @x, i8** @S_Release)
-; CHECK:   call void @objc_release(i8* %c) #0, !llvm.arc.annotation.bottomup ![[ANN3:[0-9]+]], !llvm.arc.annotation.topdown ![[ANN4:[0-9]+]]
-; CHECK:   call void @llvm.arc.annotation.topdown.bbend(i8** @x, i8** @S_None)
-; CHECK: }
-define void @test0(i32* %x, i1 %p) nounwind {
-entry:
-  %a = bitcast i32* %x to i8*
-  %0 = call i8* @objc_retain(i8* %a) nounwind
-  br i1 %p, label %t, label %f
-
-t:
-  store i8 3, i8* %a
-  %b = bitcast i32* %x to float*
-  store float 2.0, float* %b
-  br label %return
-
-f:
-  store i32 7, i32* %x
-  br label %return
-
-return:
-  %c = bitcast i32* %x to i8*
-  call void @objc_release(i8* %c) nounwind
-  ret void
-}
-
-!0 = !{}
-
-; CHECK: ![[ANN0]] = !{!"(test0,%x)", !"S_Use", !"S_None"}
-; CHECK: ![[ANN1]] = !{!"(test0,%x)", !"S_None", !"S_Retain"}
-; CHECK: ![[ANN2]] = !{!"(test0,%x)", !"S_Release", !"S_Use"}
-; CHECK: ![[ANN3]] = !{!"(test0,%x)", !"S_None", !"S_Release"}
-; CHECK: ![[ANN4]] = !{!"(test0,%x)", !"S_Retain", !"S_None"}
-
diff --git a/test/Transforms/ObjCARC/basic.ll b/test/Transforms/ObjCARC/basic.ll
index 7bc58c42d532..ff6c69cd9ed0 100644
--- a/test/Transforms/ObjCARC/basic.ll
+++ b/test/Transforms/ObjCARC/basic.ll
@@ -286,7 +286,7 @@ entry:
 loop:
   %c = bitcast i32* %x to i8*
   call void @objc_release(i8* %c) nounwind
-  %j = load volatile i1* %q
+  %j = load volatile i1, i1* %q
   br i1 %j, label %loop, label %return
 
 return:
@@ -306,7 +306,7 @@ entry:
 loop:
   %c = bitcast i32* %x to i8*
   call void @objc_release(i8* %c) nounwind, !clang.imprecise_release !0
-  %j = load volatile i1* %q
+  %j = load volatile i1, i1* %q
   br i1 %j, label %loop, label %return
 
 return:
@@ -330,7 +330,7 @@ entry:
 loop:
   %a = bitcast i32* %x to i8*
   %0 = call i8* @objc_retain(i8* %a) nounwind
-  %j = load volatile i1* %q
+  %j = load volatile i1, i1* %q
   br i1 %j, label %loop, label %return
 
 return:
@@ -350,7 +350,7 @@ entry:
 loop:
   %a = bitcast i32* %x to i8*
   %0 = call i8* @objc_retain(i8* %a) nounwind
-  %j = load volatile i1* %q
+  %j = load volatile i1, i1* %q
   br i1 %j, label %loop, label %return
 
 return:
@@ -900,13 +900,14 @@ entry:
   ret i8* %x
 }
 
-; Trivial retain,release pair with intervening call, but it's dominated
-; by another retain - delete!
+; We can not delete this retain, release since we do not have a post-dominating
+; use of the release.
 
 ; CHECK-LABEL: define void @test12(
 ; CHECK-NEXT: entry:
 ; CHECK-NEXT: @objc_retain(i8* %x)
-; CHECK-NOT: @objc_
+; CHECK-NEXT: @objc_retain
+; CHECK: @objc_release
 ; CHECK: }
 define void @test12(i8* %x, i64 %n) {
 entry:
@@ -942,6 +943,8 @@ entry:
 ; CHECK-NEXT: @objc_retain(i8* %x)
 ; CHECK-NEXT: @use_pointer
 ; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @objc_release
 ; CHECK-NEXT: ret void
 ; CHECK-NEXT: }
 define void @test13b(i8* %x, i64 %n) {
@@ -951,6 +954,8 @@ entry:
   call void @use_pointer(i8* %x)
   call void @use_pointer(i8* %x)
   call void @objc_release(i8* %x) nounwind
+  call void @use_pointer(i8* %x)
+  call void @objc_release(i8* %x) nounwind
   ret void
 }
 
@@ -984,6 +989,8 @@ entry:
 ; CHECK-NEXT: @objc_autoreleasePoolPush
 ; CHECK-NEXT: @use_pointer
 ; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @objc_release
 ; CHECK-NEXT: ret void
 ; CHECK-NEXT: }
 define void @test13d(i8* %x, i64 %n) {
@@ -994,17 +1001,22 @@ entry:
   call void @use_pointer(i8* %x)
   call void @use_pointer(i8* %x)
   call void @objc_release(i8* %x) nounwind
+  call void @use_pointer(i8* %x)
+  call void @objc_release(i8* %x) nounwind
   ret void
 }
 
-; Trivial retain,release pair with intervening call, but it's post-dominated
-; by another release - delete!
+; Trivial retain,release pair with intervening call, and it's post-dominated by
+; another release. But it is not known safe in the top down direction. We can
+; not eliminate it.
 
 ; CHECK-LABEL: define void @test14(
 ; CHECK-NEXT: entry:
+; CHECK-NEXT: @objc_retain
 ; CHECK-NEXT: @use_pointer
 ; CHECK-NEXT: @use_pointer
 ; CHECK-NEXT: @objc_release
+; CHECK-NEXT: @objc_release
 ; CHECK-NEXT: ret void
 ; CHECK-NEXT: }
 define void @test14(i8* %x, i64 %n) {
@@ -1073,6 +1085,9 @@ entry:
 ; CHECK-LABEL: define void @test16a(
 ; CHECK: @objc_retain(i8* %x)
 ; CHECK-NOT: @objc
+; CHECK: purple:
+; CHECK: @use_pointer
+; CHECK: @objc_release
 ; CHECK: }
 define void @test16a(i1 %a, i1 %b, i8* %x) {
 entry:
@@ -1101,12 +1116,18 @@ blue:
   br label %purple
 
 purple:
+  call void @use_pointer(i8* %x)
+  call void @objc_release(i8* %x) nounwind
   ret void
 }
 
 ; CHECK-LABEL: define void @test16b(
 ; CHECK: @objc_retain(i8* %x)
 ; CHECK-NOT: @objc
+; CHECK: purple:
+; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @use_pointer
+; CHECK-NEXT: @objc_release
 ; CHECK: }
 define void @test16b(i1 %a, i1 %b, i8* %x) {
 entry:
@@ -1135,12 +1156,18 @@ blue:
   br label %purple
 
 purple:
+  call void @use_pointer(i8* %x)
+  call void @use_pointer(i8* %x)
+  call void @objc_release(i8* %x) nounwind
   ret void
 }
 
 ; CHECK-LABEL: define void @test16c(
 ; CHECK: @objc_retain(i8* %x)
 ; CHECK-NOT: @objc
+; CHECK: purple:
+; CHECK: @use_pointer
+; CHECK: @objc_release
 ; CHECK: }
 define void @test16c(i1 %a, i1 %b, i8* %x) {
 entry:
@@ -1169,12 +1196,14 @@ blue:
   br label %purple
 
 purple:
+  call void @use_pointer(i8* %x)
+  call void @objc_release(i8* %x) nounwind, !clang.imprecise_release !0
   ret void
 }
 
 ; CHECK-LABEL: define void @test16d(
 ; CHECK: @objc_retain(i8* %x)
-; CHECK-NOT: @objc
+; CHECK: @objc
 ; CHECK: }
 define void @test16d(i1 %a, i1 %b, i8* %x) {
 entry:
@@ -1206,44 +1235,6 @@ purple:
   ret void
 }
 
-
-; Retain+release pairs in diamonds, all post-dominated by a release.
-
-; CHECK-LABEL: define void @test17(
-; CHECK-NOT: @objc_
-; CHECK: purple:
-; CHECK: @objc_release
-; CHECK: }
-define void @test17(i1 %a, i1 %b, i8* %x) {
-entry:
-  br i1 %a, label %red, label %orange
-
-red:
-  call i8* @objc_retain(i8* %x) nounwind
-  br label %yellow
-
-orange:
-  call i8* @objc_retain(i8* %x) nounwind
-  br label %yellow
-
-yellow:
-  call void @use_pointer(i8* %x)
-  call void @use_pointer(i8* %x)
-  br i1 %b, label %green, label %blue
-
-green:
-  call void @objc_release(i8* %x) nounwind
-  br label %purple
-
-blue:
-  call void @objc_release(i8* %x) nounwind
-  br label %purple
-
-purple:
-  call void @objc_release(i8* %x) nounwind
-  ret void
-}
-
 ; Delete no-ops.
 
 ; CHECK-LABEL: define void @test18(
@@ -1366,7 +1357,7 @@ C:
 define void @test24(i8* %r, i8* %a) {
   call i8* @objc_retain(i8* %a)
   call void @use_pointer(i8* %r)
-  %q = load i8* %a
+  %q = load i8, i8* %a
   call void @objc_release(i8* %a)
   ret void
 }
@@ -1659,7 +1650,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g)
   ret void
 }
@@ -1688,7 +1679,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g)
   ret void
 }
@@ -1713,7 +1704,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g)
   ret void
 }
@@ -1732,7 +1723,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g), !clang.imprecise_release !0
   ret void
 }
@@ -1760,7 +1751,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g)
   ret void
 }
@@ -1780,7 +1771,7 @@ true:
 
 done:
   %g = bitcast i8* %p to i8*
-  %h = getelementptr i8* %g, i64 0
+  %h = getelementptr i8, i8* %g, i64 0
   call void @objc_release(i8* %g), !clang.imprecise_release !0
   ret void
 }
@@ -1936,6 +1927,9 @@ exit:                                             ; preds = %loop
 ; CHECK-NEXT: call i8* @objc_autorelease(i8* %p)
 ; CHECK-NEXT: call void @use_pointer(i8* %p)
 ; CHECK-NEXT: call void @use_pointer(i8* %p)
+; CHECK-NEXT: call void @use_pointer(i8* %p)
+; CHECK-NEXT: call void @use_pointer(i8* %p)
+; CHECK-NEXT: call void @objc_release(i8* %p)
 ; CHECK-NEXT: ret void
 ; CHECK-NEXT: }
 define void @test42(i8* %p) {
@@ -1946,6 +1940,9 @@ entry:
   call void @use_pointer(i8* %p)
   call void @use_pointer(i8* %p)
   call void @objc_release(i8* %p)
+  call void @use_pointer(i8* %p)
+  call void @use_pointer(i8* %p)
+  call void @objc_release(i8* %p)
   ret void
 }
 
@@ -1985,6 +1982,8 @@ entry:
 ; CHECK-NEXT: call void @use_pointer(i8* %p)
 ; CHECK-NEXT: call void @use_pointer(i8* %p)
 ; CHECK-NEXT: call i8* @objc_autoreleasePoolPush()
+; CHECK-NEXT: call void @use_pointer(i8* %p)
+; CHECK-NEXT: call void @objc_release
 ; CHECK-NEXT: ret void
 ; CHECK-NEXT: }
 define void @test43b(i8* %p) {
@@ -1996,6 +1995,8 @@ entry:
   call void @use_pointer(i8* %p)
   call i8* @objc_autoreleasePoolPush()
   call void @objc_release(i8* %p)
+  call void @use_pointer(i8* %p)
+  call void @objc_release(i8* %p)
   ret void
 }
 
@@ -2005,7 +2006,7 @@ entry:
 ; CHECK-NOT: objc_
 ; CHECK: }
 define void @test44(i8** %pp) {
-  %p = load i8** %pp
+  %p = load i8*, i8** %pp
   %q = call i8* @objc_retain(i8* %p)
   call void @objc_release(i8* %q)
   ret void
@@ -2021,8 +2022,8 @@ define void @test44(i8** %pp) {
 ; CHECK: call void @objc_release(i8* %p)
 ; CHECK: }
 define void @test45(i8** %pp, i8** %qq) {
-  %p = load i8** %pp
-  %q = load i8** %qq
+  %p = load i8*, i8** %pp
+  %q = load i8*, i8** %qq
   call i8* @objc_retain(i8* %p)
   call void @objc_release(i8* %q)
   call void @use_pointer(i8* %p)
@@ -2154,10 +2155,10 @@ define void @test51b(i8* %p) {
 ; CHECK: ret void
 ; CHECK: }
 define void @test52a(i8** %zz, i8** %pp) {
-  %p = load i8** %pp
+  %p = load i8*, i8** %pp
   %1 = call i8* @objc_retain(i8* %p)
   call void @callee()
-  %z = load i8** %zz
+  %z = load i8*, i8** %zz
   call void @use_pointer(i8* %z)
   call void @objc_release(i8* %p)
   ret void
@@ -2171,10 +2172,10 @@ define void @test52a(i8** %zz, i8** %pp) {
 ; CHECK: ret void
 ; CHECK: }
 define void @test52b(i8** %zz, i8** %pp) {
-  %p = load i8** %pp
+  %p = load i8*, i8** %pp
   %1 = call i8* @objc_retain(i8* %p)
   call void @callee()
-  %z = load i8** %zz
+  %z = load i8*, i8** %zz
   call void @use_pointer(i8* %z)
   call void @objc_release(i8* %p), !clang.imprecise_release !0
   ret void
@@ -2189,10 +2190,10 @@ define void @test52b(i8** %zz, i8** %pp) {
 ; CHECK: @objc_
 ; CHECK: }
 define void @test53(void ()** %zz, i8** %pp) {
-  %p = load i8** %pp
+  %p = load i8*, i8** %pp
   %1 = call i8* @objc_retain(i8* %p)
   call void @callee()
-  %z = load void ()** %zz
+  %z = load void ()*, void ()** %zz
   call void @callee_fnptr(void ()* %z)
   call void @objc_release(i8* %p)
   ret void
@@ -2260,15 +2261,16 @@ if.end:                                           ; preds = %entry, %if.then
   ret void
 }
 
-; When there are adjacent retain+release pairs, the first one is
-; known unnecessary because the presence of the second one means that
-; the first one won't be deleting the object.
+; When there are adjacent retain+release pairs, the first one is known
+; unnecessary because the presence of the second one means that the first one
+; won't be deleting the object.
 
 ; CHECK-LABEL:      define void @test57(
 ; CHECK-NEXT: entry:
+; CHECK-NEXT:   tail call i8* @objc_retain(i8* %x) [[NUW]]
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
-; CHECK-NEXT:   %0 = tail call i8* @objc_retain(i8* %x) [[NUW]]
+; CHECK-NEXT:   tail call i8* @objc_retain(i8* %x) [[NUW]]
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
 ; CHECK-NEXT:   call void @objc_release(i8* %x) [[NUW]]
@@ -2277,6 +2279,7 @@ if.end:                                           ; preds = %entry, %if.then
 define void @test57(i8* %x) nounwind {
 entry:
   call i8* @objc_retain(i8* %x) nounwind
+  call i8* @objc_retain(i8* %x) nounwind
   call void @use_pointer(i8* %x)
   call void @use_pointer(i8* %x)
   call void @objc_release(i8* %x) nounwind
@@ -2292,6 +2295,7 @@ entry:
 
 ; CHECK-LABEL:      define void @test58(
 ; CHECK-NEXT: entry:
+; CHECK-NEXT:   @objc_retain
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
 ; CHECK-NEXT:   call void @use_pointer(i8* %x)
 ; CHECK-NEXT:   ret void
@@ -2299,6 +2303,7 @@ entry:
 define void @test58(i8* %x) nounwind {
 entry:
   call i8* @objc_retain(i8* %x) nounwind
+  call i8* @objc_retain(i8* %x) nounwind
   call void @use_pointer(i8* %x)
   call void @use_pointer(i8* %x)
   call void @objc_release(i8* %x) nounwind
@@ -2341,8 +2346,8 @@ entry:
 ; CHECK: call void @objc_release
 ; CHECK: }
 define void @test60a() {
-  %t = load i8** @constptr
-  %s = load i8** @something
+  %t = load i8*, i8** @constptr
+  %s = load i8*, i8** @something
   call i8* @objc_retain(i8* %s)
   call void @callee()
   call void @use_pointer(i8* %t)
@@ -2353,16 +2358,16 @@ define void @test60a() {
 ; CHECK-LABEL: define void @test60b(
 ; CHECK: call i8* @objc_retain
 ; CHECK-NOT: call i8* @objc_retain
-; CHECK-NOT: call i8* @objc_rrelease
+; CHECK-NOT: call i8* @objc_release
 ; CHECK: }
 define void @test60b() {
-  %t = load i8** @constptr
-  %s = load i8** @something
-  call i8* @objc_retain(i8* %s)
-  call i8* @objc_retain(i8* %s)
+  %t = load i8*, i8** @constptr
+  %s = load i8*, i8** @something
+  call i8* @objc_retain(i8* %t)
+  call i8* @objc_retain(i8* %t)
   call void @callee()
-  call void @use_pointer(i8* %t)
-  call void @objc_release(i8* %s)
+  call void @use_pointer(i8* %s)
+  call void @objc_release(i8* %t)
   ret void
 }
 
@@ -2370,12 +2375,12 @@ define void @test60b() {
 ; CHECK-NOT: @objc_
 ; CHECK: }
 define void @test60c() {
-  %t = load i8** @constptr
-  %s = load i8** @something
-  call i8* @objc_retain(i8* %s)
+  %t = load i8*, i8** @constptr
+  %s = load i8*, i8** @something
+  call i8* @objc_retain(i8* %t)
   call void @callee()
-  call void @use_pointer(i8* %t)
-  call void @objc_release(i8* %s), !clang.imprecise_release !0
+  call void @use_pointer(i8* %s)
+  call void @objc_release(i8* %t), !clang.imprecise_release !0
   ret void
 }
 
@@ -2383,8 +2388,8 @@ define void @test60c() {
 ; CHECK-NOT: @objc_
 ; CHECK: }
 define void @test60d() {
-  %t = load i8** @constptr
-  %s = load i8** @something
+  %t = load i8*, i8** @constptr
+  %s = load i8*, i8** @something
   call i8* @objc_retain(i8* %t)
   call void @callee()
   call void @use_pointer(i8* %s)
@@ -2396,8 +2401,8 @@ define void @test60d() {
 ; CHECK-NOT: @objc_
 ; CHECK: }
 define void @test60e() {
-  %t = load i8** @constptr
-  %s = load i8** @something
+  %t = load i8*, i8** @constptr
+  %s = load i8*, i8** @something
   call i8* @objc_retain(i8* %t)
   call void @callee()
   call void @use_pointer(i8* %s)
@@ -2412,7 +2417,7 @@ define void @test60e() {
 ; CHECK-NOT: @objc_
 ; CHECK: }
 define void @test61() {
-  %t = load i8** @constptr
+  %t = load i8*, i8** @constptr
   call i8* @objc_retain(i8* %t)
   call void @callee()
   call void @use_pointer(i8* %t)
@@ -2432,7 +2437,7 @@ entry:
 
 loop:
   call i8* @objc_retain(i8* %x)
-  %q = load i1* %p
+  %q = load i1, i1* %p
   br i1 %q, label %loop.more, label %exit
 
 loop.more:
@@ -2459,7 +2464,7 @@ entry:
 
 loop:
   call i8* @objc_retain(i8* %x)
-  %q = load i1* %p
+  %q = load i1, i1* %p
   br i1 %q, label %loop.more, label %exit
 
 loop.more:
@@ -2485,7 +2490,7 @@ entry:
 
 loop:
   call i8* @objc_retain(i8* %x)
-  %q = load i1* %p
+  %q = load i1, i1* %p
   br i1 %q, label %loop.more, label %exit
 
 loop.more:
@@ -2679,33 +2684,33 @@ define {<2 x float>, <2 x float>} @"\01-[A z]"({}* %self, i8* nocapture %_cmd) n
 invoke.cont:
   %0 = bitcast {}* %self to i8*
   %1 = tail call i8* @objc_retain(i8* %0) nounwind
-  tail call void @llvm.dbg.value(metadata {}* %self, i64 0, metadata !0, metadata !{})
-  tail call void @llvm.dbg.value(metadata {}* %self, i64 0, metadata !0, metadata !{})
-  %ivar = load i64* @"OBJC_IVAR_$_A.myZ", align 8
-  %add.ptr = getelementptr i8* %0, i64 %ivar
+  tail call void @llvm.dbg.value(metadata {}* %self, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !2), metadata !DIExpression()), !dbg !DILocation(scope: !2)
+  tail call void @llvm.dbg.value(metadata {}* %self, i64 0, metadata !DILocalVariable(tag: DW_TAG_auto_variable, scope: !2), metadata !DIExpression()), !dbg !DILocation(scope: !2)
+  %ivar = load i64, i64* @"OBJC_IVAR_$_A.myZ", align 8
+  %add.ptr = getelementptr i8, i8* %0, i64 %ivar
   %tmp1 = bitcast i8* %add.ptr to float*
-  %tmp2 = load float* %tmp1, align 4
+  %tmp2 = load float, float* %tmp1, align 4
   %conv = fpext float %tmp2 to double
   %add.ptr.sum = add i64 %ivar, 4
-  %tmp6 = getelementptr inbounds i8* %0, i64 %add.ptr.sum
+  %tmp6 = getelementptr inbounds i8, i8* %0, i64 %add.ptr.sum
   %2 = bitcast i8* %tmp6 to float*
-  %tmp7 = load float* %2, align 4
+  %tmp7 = load float, float* %2, align 4
   %conv8 = fpext float %tmp7 to double
   %add.ptr.sum36 = add i64 %ivar, 8
-  %tmp12 = getelementptr inbounds i8* %0, i64 %add.ptr.sum36
+  %tmp12 = getelementptr inbounds i8, i8* %0, i64 %add.ptr.sum36
   %arrayidx = bitcast i8* %tmp12 to float*
-  %tmp13 = load float* %arrayidx, align 4
+  %tmp13 = load float, float* %arrayidx, align 4
   %conv14 = fpext float %tmp13 to double
   %tmp12.sum = add i64 %ivar, 12
-  %arrayidx19 = getelementptr inbounds i8* %0, i64 %tmp12.sum
+  %arrayidx19 = getelementptr inbounds i8, i8* %0, i64 %tmp12.sum
   %3 = bitcast i8* %arrayidx19 to float*
-  %tmp20 = load float* %3, align 4
+  %tmp20 = load float, float* %3, align 4
   %conv21 = fpext float %tmp20 to double
-  %call = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([33 x i8]* @.str4, i64 0, i64 0), double %conv, double %conv8, double %conv14, double %conv21)
-  %ivar23 = load i64* @"OBJC_IVAR_$_A.myZ", align 8
-  %add.ptr24 = getelementptr i8* %0, i64 %ivar23
+  %call = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([33 x i8], [33 x i8]* @.str4, i64 0, i64 0), double %conv, double %conv8, double %conv14, double %conv21)
+  %ivar23 = load i64, i64* @"OBJC_IVAR_$_A.myZ", align 8
+  %add.ptr24 = getelementptr i8, i8* %0, i64 %ivar23
   %4 = bitcast i8* %add.ptr24 to i128*
-  %srcval = load i128* %4, align 4
+  %srcval = load i128, i128* %4, align 4
   tail call void @objc_release(i8* %0) nounwind
   %tmp29 = trunc i128 %srcval to i64
   %tmp30 = bitcast i64 %tmp29 to <2 x float>
@@ -2725,14 +2730,14 @@ define i32 @"\01-[Top0 _getX]"({}* %self, i8* nocapture %_cmd) nounwind {
 invoke.cont:
   %0 = bitcast {}* %self to i8*
   %1 = tail call i8* @objc_retain(i8* %0) nounwind
-  %puts = tail call i32 @puts(i8* getelementptr inbounds ([16 x i8]* @str, i64 0, i64 0))
+  %puts = tail call i32 @puts(i8* getelementptr inbounds ([16 x i8], [16 x i8]* @str, i64 0, i64 0))
   tail call void @objc_release(i8* %0) nounwind
   ret i32 0
 }
 
-@"\01L_OBJC_METH_VAR_NAME_" = internal global [5 x i8] c"frob\00", section "__TEXT,__cstring,cstring_literals", align 1@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([5 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@"\01L_OBJC_METH_VAR_NAME_" = internal global [5 x i8] c"frob\00", section "__TEXT,__cstring,cstring_literals", align 1@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([5 x i8], [5 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
 @"\01L_OBJC_IMAGE_INFO" = internal constant [2 x i32] [i32 0, i32 16], section "__DATA, __objc_imageinfo, regular, no_dead_strip"
-@llvm.used = appending global [3 x i8*] [i8* getelementptr inbounds ([5 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
+@llvm.used = appending global [3 x i8*] [i8* getelementptr inbounds ([5 x i8], [5 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
 
 ; A simple loop. Eliminate the retain and release inside of it!
 
@@ -2752,8 +2757,8 @@ entry:
 for.body:                                         ; preds = %entry, %for.body
   %i.010 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
   %1 = tail call i8* @objc_retain(i8* %x) nounwind
-  %tmp5 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
-  %call = tail call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %1, i8* %tmp5)
+  %tmp5 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %call = tail call i8* (i8*, i8*, ...) @objc_msgSend(i8* %1, i8* %tmp5)
   tail call void @objc_release(i8* %1) nounwind, !clang.imprecise_release !0
   %inc = add nsw i64 %i.010, 1
   %exitcond = icmp eq i64 %inc, %n
@@ -2828,21 +2833,21 @@ entry:
   %tmp7 = bitcast %2* %self to i8*
   %tmp8 = call i8* @objc_retain(i8* %tmp7) nounwind
   store %4* null, %4** %err, align 8
-  %tmp1 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_17", align 8
-  %tmp2 = load %struct.__CFString** @kUTTypePlainText, align 8
-  %tmp3 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_19", align 8
+  %tmp1 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_17", align 8
+  %tmp2 = load %struct.__CFString*, %struct.__CFString** @kUTTypePlainText, align 8
+  %tmp3 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_19", align 8
   %tmp4 = bitcast %struct._class_t* %tmp1 to i8*
-  %call5 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp4, i8* %tmp3, %struct.__CFString* %tmp2)
-  %tmp5 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_21", align 8
+  %call5 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp4, i8* %tmp3, %struct.__CFString* %tmp2)
+  %tmp5 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_21", align 8
   %tmp6 = bitcast %3* %pboard to i8*
-  %call76 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp6, i8* %tmp5, i8* %call5)
+  %call76 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp6, i8* %tmp5, i8* %call5)
   %tmp9 = call i8* @objc_retain(i8* %call76) nounwind
   %tobool = icmp eq i8* %tmp9, null
   br i1 %tobool, label %end, label %land.lhs.true
 
 land.lhs.true:                                    ; preds = %entry
-  %tmp11 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_23", align 8
-  %call137 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp6, i8* %tmp11, i8* %tmp9)
+  %tmp11 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_23", align 8
+  %call137 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp6, i8* %tmp11, i8* %tmp9)
   %tmp = bitcast i8* %call137 to %1*
   %tmp10 = call i8* @objc_retain(i8* %call137) nounwind
   call void @objc_release(i8* null) nounwind
@@ -2852,16 +2857,16 @@ land.lhs.true:                                    ; preds = %entry
   br i1 %tobool16, label %end, label %if.then
 
 if.then:                                          ; preds = %land.lhs.true
-  %tmp19 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_25", align 8
+  %tmp19 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_25", align 8
   %call21 = call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*)*)(i8* %call137, i8* %tmp19)
   %tobool22 = icmp eq i8 %call21, 0
   br i1 %tobool22, label %if.then44, label %land.lhs.true23
 
 land.lhs.true23:                                  ; preds = %if.then
-  %tmp24 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_26", align 8
-  %tmp26 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_28", align 8
+  %tmp24 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_26", align 8
+  %tmp26 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_28", align 8
   %tmp27 = bitcast %struct._class_t* %tmp24 to i8*
-  %call2822 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp27, i8* %tmp26, i8* %call137)
+  %call2822 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp27, i8* %tmp26, i8* %call137)
   %tmp13 = bitcast i8* %call2822 to %5*
   %tmp14 = call i8* @objc_retain(i8* %call2822) nounwind
   call void @objc_release(i8* null) nounwind
@@ -2869,39 +2874,39 @@ land.lhs.true23:                                  ; preds = %if.then
   br i1 %tobool30, label %if.then44, label %if.end
 
 if.end:                                           ; preds = %land.lhs.true23
-  %tmp32 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_29", align 8
-  %tmp33 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_31", align 8
+  %tmp32 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_29", align 8
+  %tmp33 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_31", align 8
   %tmp34 = bitcast %struct._class_t* %tmp32 to i8*
-  %call35 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp34, i8* %tmp33)
-  %tmp37 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_33", align 8
-  %call3923 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call35, i8* %tmp37, i8* %call2822, i32 signext 1, %4** %err)
+  %call35 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp34, i8* %tmp33)
+  %tmp37 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_33", align 8
+  %call3923 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call35, i8* %tmp37, i8* %call2822, i32 signext 1, %4** %err)
   %cmp = icmp eq i8* %call3923, null
   br i1 %cmp, label %if.then44, label %end
 
 if.then44:                                        ; preds = %if.end, %land.lhs.true23, %if.then
   %url.025 = phi %5* [ %tmp13, %if.end ], [ %tmp13, %land.lhs.true23 ], [ null, %if.then ]
-  %tmp49 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_35", align 8
+  %tmp49 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_35", align 8
   %call51 = call %struct._NSRange bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %struct._NSRange (i8*, i8*, i64, i64)*)(i8* %call137, i8* %tmp49, i64 0, i64 0)
   %call513 = extractvalue %struct._NSRange %call51, 0
   %call514 = extractvalue %struct._NSRange %call51, 1
-  %tmp52 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_37", align 8
-  %call548 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call137, i8* %tmp52, i64 %call513, i64 %call514)
-  %tmp55 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_38", align 8
-  %tmp56 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_40", align 8
+  %tmp52 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_37", align 8
+  %call548 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call137, i8* %tmp52, i64 %call513, i64 %call514)
+  %tmp55 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_38", align 8
+  %tmp56 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_40", align 8
   %tmp57 = bitcast %struct._class_t* %tmp55 to i8*
-  %call58 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp57, i8* %tmp56)
-  %tmp59 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_42", align 8
-  %call6110 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call548, i8* %tmp59, i8* %call58)
+  %call58 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp57, i8* %tmp56)
+  %tmp59 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_42", align 8
+  %call6110 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call548, i8* %tmp59, i8* %call58)
   %tmp15 = call i8* @objc_retain(i8* %call6110) nounwind
   call void @objc_release(i8* %call137) nounwind
-  %tmp64 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_46", align 8
+  %tmp64 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_46", align 8
   %call66 = call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*, %1*)*)(i8* %call6110, i8* %tmp64, %1* bitcast (%struct.NSConstantString* @_unnamed_cfstring_44 to %1*))
   %tobool67 = icmp eq i8 %call66, 0
   br i1 %tobool67, label %if.end74, label %if.then68
 
 if.then68:                                        ; preds = %if.then44
-  %tmp70 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_48", align 8
-  %call7220 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call6110, i8* %tmp70)
+  %tmp70 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_48", align 8
+  %call7220 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call6110, i8* %tmp70)
   %tmp16 = call i8* @objc_retain(i8* %call7220) nounwind
   call void @objc_release(i8* %call6110) nounwind
   br label %if.end74
@@ -2909,54 +2914,54 @@ if.then68:                                        ; preds = %if.then44
 if.end74:                                         ; preds = %if.then68, %if.then44
   %filename.0.in = phi i8* [ %call7220, %if.then68 ], [ %call6110, %if.then44 ]
   %filename.0 = bitcast i8* %filename.0.in to %1*
-  %tmp17 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_isEqual_" to i8**), align 16
+  %tmp17 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_isEqual_" to i8**), align 16
   %tmp18 = bitcast i8* %tmp17 to i8 (i8*, %struct._message_ref_t*, i8*, ...)*
-  %call78 = call signext i8 (i8*, %struct._message_ref_t*, i8*, ...)* %tmp18(i8* %call137, %struct._message_ref_t* bitcast (%0* @"\01l_objc_msgSend_fixup_isEqual_" to %struct._message_ref_t*), i8* %filename.0.in)
+  %call78 = call signext i8 (i8*, %struct._message_ref_t*, i8*, ...) %tmp18(i8* %call137, %struct._message_ref_t* bitcast (%0* @"\01l_objc_msgSend_fixup_isEqual_" to %struct._message_ref_t*), i8* %filename.0.in)
   %tobool79 = icmp eq i8 %call78, 0
   br i1 %tobool79, label %land.lhs.true80, label %if.then109
 
 land.lhs.true80:                                  ; preds = %if.end74
-  %tmp82 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_25", align 8
+  %tmp82 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_25", align 8
   %call84 = call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*)*)(i8* %filename.0.in, i8* %tmp82)
   %tobool86 = icmp eq i8 %call84, 0
   br i1 %tobool86, label %if.then109, label %if.end106
 
 if.end106:                                        ; preds = %land.lhs.true80
-  %tmp88 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_26", align 8
-  %tmp90 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_28", align 8
+  %tmp88 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_26", align 8
+  %tmp90 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_28", align 8
   %tmp91 = bitcast %struct._class_t* %tmp88 to i8*
-  %call9218 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp91, i8* %tmp90, i8* %filename.0.in)
+  %call9218 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp91, i8* %tmp90, i8* %filename.0.in)
   %tmp20 = bitcast i8* %call9218 to %5*
   %tmp21 = call i8* @objc_retain(i8* %call9218) nounwind
   %tmp22 = bitcast %5* %url.025 to i8*
   call void @objc_release(i8* %tmp22) nounwind
-  %tmp94 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_29", align 8
-  %tmp95 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_31", align 8
+  %tmp94 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_29", align 8
+  %tmp95 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_31", align 8
   %tmp96 = bitcast %struct._class_t* %tmp94 to i8*
-  %call97 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp96, i8* %tmp95)
-  %tmp99 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_33", align 8
-  %call10119 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call97, i8* %tmp99, i8* %call9218, i32 signext 1, %4** %err)
+  %call97 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp96, i8* %tmp95)
+  %tmp99 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_33", align 8
+  %call10119 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call97, i8* %tmp99, i8* %call9218, i32 signext 1, %4** %err)
   %phitmp = icmp eq i8* %call10119, null
   br i1 %phitmp, label %if.then109, label %end
 
 if.then109:                                       ; preds = %if.end106, %land.lhs.true80, %if.end74
   %url.129 = phi %5* [ %tmp20, %if.end106 ], [ %url.025, %if.end74 ], [ %url.025, %land.lhs.true80 ]
-  %tmp110 = load %4** %err, align 8
+  %tmp110 = load %4*, %4** %err, align 8
   %tobool111 = icmp eq %4* %tmp110, null
   br i1 %tobool111, label %if.then112, label %if.end125
 
 if.then112:                                       ; preds = %if.then109
-  %tmp113 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_50", align 8
-  %tmp114 = load %1** @NSCocoaErrorDomain, align 8
-  %tmp115 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_51", align 8
+  %tmp113 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_50", align 8
+  %tmp114 = load %1*, %1** @NSCocoaErrorDomain, align 8
+  %tmp115 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_51", align 8
   %call117 = call %1* @truncatedString(%1* %filename.0, i64 1034)
-  %tmp118 = load %1** @NSFilePathErrorKey, align 8
-  %tmp119 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_53", align 8
+  %tmp118 = load %1*, %1** @NSFilePathErrorKey, align 8
+  %tmp119 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_53", align 8
   %tmp120 = bitcast %struct._class_t* %tmp115 to i8*
-  %call12113 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp120, i8* %tmp119, %1* %call117, %1* %tmp118, i8* null)
-  %tmp122 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_55", align 8
+  %call12113 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp120, i8* %tmp119, %1* %call117, %1* %tmp118, i8* null)
+  %tmp122 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_55", align 8
   %tmp123 = bitcast %struct._class_t* %tmp113 to i8*
-  %call12414 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp123, i8* %tmp122, %1* %tmp114, i64 258, i8* %call12113)
+  %call12414 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp123, i8* %tmp122, %1* %tmp114, i64 258, i8* %call12113)
   %tmp23 = call i8* @objc_retain(i8* %call12414) nounwind
   %tmp25 = call i8* @objc_autorelease(i8* %tmp23) nounwind
   %tmp28 = bitcast i8* %tmp25 to %4*
@@ -2965,12 +2970,12 @@ if.then112:                                       ; preds = %if.then109
 
 if.end125:                                        ; preds = %if.then112, %if.then109
   %tmp127 = phi %4* [ %tmp110, %if.then109 ], [ %tmp28, %if.then112 ]
-  %tmp126 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_56", align 8
-  %tmp128 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_58", align 8
+  %tmp126 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_56", align 8
+  %tmp128 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_58", align 8
   %tmp129 = bitcast %struct._class_t* %tmp126 to i8*
-  %call13015 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %tmp129, i8* %tmp128, %4* %tmp127)
-  %tmp131 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_60", align 8
-  %call13317 = call i8* (i8*, i8*, ...)* @objc_msgSend(i8* %call13015, i8* %tmp131)
+  %call13015 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %tmp129, i8* %tmp128, %4* %tmp127)
+  %tmp131 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_60", align 8
+  %call13317 = call i8* (i8*, i8*, ...) @objc_msgSend(i8* %call13015, i8* %tmp131)
   br label %end
 
 end:                                              ; preds = %if.end125, %if.end106, %if.end, %land.lhs.true, %entry
@@ -3012,7 +3017,8 @@ define void @test67(i8* %x) {
 !llvm.module.flags = !{!1}
 
 !0 = !{}
-!1 = !{i32 1, !"Debug Info Version", i32 2}
+!1 = !{i32 1, !"Debug Info Version", i32 3}
+!2 = !DISubprogram()
 
 ; CHECK: attributes #0 = { nounwind readnone }
 ; CHECK: attributes [[NUW]] = { nounwind }
diff --git a/test/Transforms/ObjCARC/cfg-hazards.ll b/test/Transforms/ObjCARC/cfg-hazards.ll
index 746d56df8fad..8407e446b4f3 100644
--- a/test/Transforms/ObjCARC/cfg-hazards.ll
+++ b/test/Transforms/ObjCARC/cfg-hazards.ll
@@ -421,7 +421,7 @@ loop:
   store i8* %a, i8** %block, align 8
   %casted_block = bitcast i8** %block to void ()*
   call void @block_callee(void ()* %casted_block)
-  %reloaded_a = load i8** %block, align 8
+  %reloaded_a = load i8*, i8** %block, align 8
   call void @objc_release(i8* %reloaded_a) nounwind, !clang.imprecise_release !0
   br i1 undef, label %loop, label %exit
   
diff --git a/test/Transforms/ObjCARC/contract-storestrong-ivar.ll b/test/Transforms/ObjCARC/contract-storestrong-ivar.ll
index 4a9b3140f101..8b1a02f3feb9 100644
--- a/test/Transforms/ObjCARC/contract-storestrong-ivar.ll
+++ b/test/Transforms/ObjCARC/contract-storestrong-ivar.ll
@@ -16,11 +16,11 @@ declare void @objc_release(i8*)
 
 define hidden void @y(%0* nocapture %self, %1* %preferencesController) nounwind {
 entry:
-  %ivar = load i64* @"OBJC_IVAR_$_Controller.preferencesController", align 8
+  %ivar = load i64, i64* @"OBJC_IVAR_$_Controller.preferencesController", align 8
   %tmp = bitcast %0* %self to i8*
-  %add.ptr = getelementptr inbounds i8* %tmp, i64 %ivar
+  %add.ptr = getelementptr inbounds i8, i8* %tmp, i64 %ivar
   %tmp1 = bitcast i8* %add.ptr to %1**
-  %tmp2 = load %1** %tmp1, align 8
+  %tmp2 = load %1*, %1** %tmp1, align 8
   %tmp3 = bitcast %1* %preferencesController to i8*
   %tmp4 = tail call i8* @objc_retain(i8* %tmp3) nounwind
   %tmp5 = bitcast %1* %tmp2 to i8*
diff --git a/test/Transforms/ObjCARC/contract-storestrong.ll b/test/Transforms/ObjCARC/contract-storestrong.ll
index f9308c8fdee2..aadc3a26539a 100644
--- a/test/Transforms/ObjCARC/contract-storestrong.ll
+++ b/test/Transforms/ObjCARC/contract-storestrong.ll
@@ -16,7 +16,7 @@ declare void @use_pointer(i8*)
 define void @test0(i8* %p) {
 entry:
   %0 = tail call i8* @objc_retain(i8* %p) nounwind
-  %tmp = load i8** @x, align 8
+  %tmp = load i8*, i8** @x, align 8
   store i8* %0, i8** @x, align 8
   tail call void @objc_release(i8* %tmp) nounwind
   ret void
@@ -24,10 +24,10 @@ entry:
 
 ; Don't do this if the load is volatile.
 
-;      CHECK: define void @test1(i8* %p) {
+; CHECK-LABEL: define void @test1(i8* %p) {
 ; CHECK-NEXT: entry:
 ; CHECK-NEXT:   %0 = tail call i8* @objc_retain(i8* %p) [[NUW]]
-; CHECK-NEXT:   %tmp = load volatile i8** @x, align 8
+; CHECK-NEXT:   %tmp = load volatile i8*, i8** @x, align 8
 ; CHECK-NEXT:   store i8* %0, i8** @x, align 8
 ; CHECK-NEXT:   tail call void @objc_release(i8* %tmp) [[NUW]]
 ; CHECK-NEXT:   ret void
@@ -35,7 +35,7 @@ entry:
 define void @test1(i8* %p) {
 entry:
   %0 = tail call i8* @objc_retain(i8* %p) nounwind
-  %tmp = load volatile i8** @x, align 8
+  %tmp = load volatile i8*, i8** @x, align 8
   store i8* %0, i8** @x, align 8
   tail call void @objc_release(i8* %tmp) nounwind
   ret void
@@ -43,10 +43,10 @@ entry:
 
 ; Don't do this if the store is volatile.
 
-;      CHECK: define void @test2(i8* %p) {
+; CHECK-LABEL: define void @test2(i8* %p) {
 ; CHECK-NEXT: entry:
 ; CHECK-NEXT:   %0 = tail call i8* @objc_retain(i8* %p) [[NUW]]
-; CHECK-NEXT:   %tmp = load i8** @x, align 8
+; CHECK-NEXT:   %tmp = load i8*, i8** @x, align 8
 ; CHECK-NEXT:   store volatile i8* %0, i8** @x, align 8
 ; CHECK-NEXT:   tail call void @objc_release(i8* %tmp) [[NUW]]
 ; CHECK-NEXT:   ret void
@@ -54,7 +54,7 @@ entry:
 define void @test2(i8* %p) {
 entry:
   %0 = tail call i8* @objc_retain(i8* %p) nounwind
-  %tmp = load i8** @x, align 8
+  %tmp = load i8*, i8** @x, align 8
   store volatile i8* %0, i8** @x, align 8
   tail call void @objc_release(i8* %tmp) nounwind
   ret void
@@ -63,19 +63,19 @@ entry:
 ; Don't do this if there's a use of the old pointer value between the store
 ; and the release.
 
-; CHECK:      define void @test3(i8* %newValue) {
-; CHECK-NEXT: entry:
-; CHECK-NEXT:   %x0 = tail call i8* @objc_retain(i8* %newValue) [[NUW]]
-; CHECK-NEXT:   %x1 = load i8** @x, align 8
-; CHECK-NEXT:   store i8* %x0, i8** @x, align 8
-; CHECK-NEXT:   tail call void @use_pointer(i8* %x1), !clang.arc.no_objc_arc_exceptions !0
-; CHECK-NEXT:   tail call void @objc_release(i8* %x1) [[NUW]], !clang.imprecise_release !0
-; CHECK-NEXT:   ret void
-; CHECK-NEXT: }
+; CHECK-LABEL: define void @test3(i8* %newValue) {
+; CHECK-NEXT:  entry:
+; CHECK-NEXT:    %x0 = tail call i8* @objc_retain(i8* %newValue) [[NUW]]
+; CHECK-NEXT:    %x1 = load i8*, i8** @x, align 8
+; CHECK-NEXT:    store i8* %x0, i8** @x, align 8
+; CHECK-NEXT:    tail call void @use_pointer(i8* %x1), !clang.arc.no_objc_arc_exceptions !0
+; CHECK-NEXT:    tail call void @objc_release(i8* %x1) [[NUW]], !clang.imprecise_release !0
+; CHECK-NEXT:    ret void
+; CHECK-NEXT:  }
 define void @test3(i8* %newValue) {
 entry:
   %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
-  %x1 = load i8** @x, align 8
+  %x1 = load i8*, i8** @x, align 8
   store i8* %newValue, i8** @x, align 8
   tail call void @use_pointer(i8* %x1), !clang.arc.no_objc_arc_exceptions !0
   tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
@@ -84,19 +84,19 @@ entry:
 
 ; Like test3, but with an icmp use instead of a call, for good measure.
 
-; CHECK:      define i1 @test4(i8* %newValue, i8* %foo) {
-; CHECK-NEXT: entry:
-; CHECK-NEXT:   %x0 = tail call i8* @objc_retain(i8* %newValue) [[NUW]]
-; CHECK-NEXT:   %x1 = load i8** @x, align 8
-; CHECK-NEXT:   store i8* %x0, i8** @x, align 8
-; CHECK-NEXT:   %t = icmp eq i8* %x1, %foo
-; CHECK-NEXT:   tail call void @objc_release(i8* %x1) [[NUW]], !clang.imprecise_release !0
-; CHECK-NEXT:   ret i1 %t
-; CHECK-NEXT: }
+; CHECK-LABEL:  define i1 @test4(i8* %newValue, i8* %foo) {
+; CHECK-NEXT:   entry:
+; CHECK-NEXT:     %x0 = tail call i8* @objc_retain(i8* %newValue) [[NUW]]
+; CHECK-NEXT:     %x1 = load i8*, i8** @x, align 8
+; CHECK-NEXT:     store i8* %x0, i8** @x, align 8
+; CHECK-NEXT:     %t = icmp eq i8* %x1, %foo
+; CHECK-NEXT:     tail call void @objc_release(i8* %x1) [[NUW]], !clang.imprecise_release !0
+; CHECK-NEXT:     ret i1 %t
+; CHECK-NEXT:   }
 define i1 @test4(i8* %newValue, i8* %foo) {
 entry:
   %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
-  %x1 = load i8** @x, align 8
+  %x1 = load i8*, i8** @x, align 8
   store i8* %newValue, i8** @x, align 8
   %t = icmp eq i8* %x1, %foo
   tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
@@ -105,14 +105,14 @@ entry:
 
 ; Do form an objc_storeStrong here, because the use is before the store.
 
-; CHECK: define i1 @test5(i8* %newValue, i8* %foo) {
+; CHECK-LABEL: define i1 @test5(i8* %newValue, i8* %foo) {
 ; CHECK: %t = icmp eq i8* %x1, %foo
 ; CHECK: tail call void @objc_storeStrong(i8** @x, i8* %newValue) [[NUW]]
 ; CHECK: }
 define i1 @test5(i8* %newValue, i8* %foo) {
 entry:
   %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
-  %x1 = load i8** @x, align 8
+  %x1 = load i8*, i8** @x, align 8
   %t = icmp eq i8* %x1, %foo
   store i8* %newValue, i8** @x, align 8
   tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
@@ -121,14 +121,14 @@ entry:
 
 ; Like test5, but the release is before the store.
 
-; CHECK: define i1 @test6(i8* %newValue, i8* %foo) {
+; CHECK-LABEL: define i1 @test6(i8* %newValue, i8* %foo) {
 ; CHECK: %t = icmp eq i8* %x1, %foo
 ; CHECK: tail call void @objc_storeStrong(i8** @x, i8* %newValue) [[NUW]]
 ; CHECK: }
 define i1 @test6(i8* %newValue, i8* %foo) {
 entry:
   %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
-  %x1 = load i8** @x, align 8
+  %x1 = load i8*, i8** @x, align 8
   tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
   %t = icmp eq i8* %x1, %foo
   store i8* %newValue, i8** @x, align 8
@@ -137,38 +137,86 @@ entry:
 
 ; Like test0, but there's no store, so don't form an objc_storeStrong.
 
-;      CHECK-LABEL: define void @test7(
+; CHECK-LABEL: define void @test7(
 ; CHECK-NEXT: entry:
 ; CHECK-NEXT:   %0 = tail call i8* @objc_retain(i8* %p) [[NUW]]
-; CHECK-NEXT:   %tmp = load i8** @x, align 8
+; CHECK-NEXT:   %tmp = load i8*, i8** @x, align 8
 ; CHECK-NEXT:   tail call void @objc_release(i8* %tmp) [[NUW]]
 ; CHECK-NEXT:   ret void
 ; CHECK-NEXT: }
 define void @test7(i8* %p) {
 entry:
   %0 = tail call i8* @objc_retain(i8* %p) nounwind
-  %tmp = load i8** @x, align 8
+  %tmp = load i8*, i8** @x, align 8
   tail call void @objc_release(i8* %tmp) nounwind
   ret void
 }
 
 ; Like test0, but there's no retain, so don't form an objc_storeStrong.
 
-;      CHECK-LABEL: define void @test8(
+; CHECK-LABEL: define void @test8(
 ; CHECK-NEXT: entry:
-; CHECK-NEXT:   %tmp = load i8** @x, align 8
+; CHECK-NEXT:   %tmp = load i8*, i8** @x, align 8
 ; CHECK-NEXT:   store i8* %p, i8** @x, align 8
 ; CHECK-NEXT:   tail call void @objc_release(i8* %tmp) [[NUW]]
 ; CHECK-NEXT:   ret void
 ; CHECK-NEXT: }
 define void @test8(i8* %p) {
 entry:
-  %tmp = load i8** @x, align 8
+  %tmp = load i8*, i8** @x, align 8
   store i8* %p, i8** @x, align 8
   tail call void @objc_release(i8* %tmp) nounwind
   ret void
 }
 
+; Make sure that we properly handle release that *may* release our new
+; value in between the retain and the store. We need to be sure that
+; this we can safely move the retain to the store. This specific test
+; makes sure that we properly handled a release of an unrelated
+; pointer.
+;
+; CHECK-LABEL: define i1 @test9(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+; CHECK-NOT: objc_storeStrong
+define i1 @test9(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+entry:
+  %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
+  tail call void @objc_release(i8* %unrelated_ptr) nounwind, !clang.imprecise_release !0
+  %x1 = load i8*, i8** @x, align 8
+  tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
+  %t = icmp eq i8* %x1, %foo
+  store i8* %newValue, i8** @x, align 8
+  ret i1 %t  
+}
+
+; Make sure that we don't perform the optimization when we just have a call.
+;
+; CHECK-LABEL: define i1 @test10(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+; CHECK-NOT: objc_storeStrong
+define i1 @test10(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+entry:
+  %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
+  call void @use_pointer(i8* %unrelated_ptr)
+  %x1 = load i8*, i8** @x, align 8
+  tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
+  %t = icmp eq i8* %x1, %foo
+  store i8* %newValue, i8** @x, align 8
+  ret i1 %t
+}
+
+; Make sure we form the store strong if the use in between the retain
+; and the store does not touch reference counts.
+; CHECK-LABEL: define i1 @test11(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+; CHECK: objc_storeStrong
+define i1 @test11(i8* %newValue, i8* %foo, i8* %unrelated_ptr) {
+entry:
+  %x0 = tail call i8* @objc_retain(i8* %newValue) nounwind
+  %t = icmp eq i8* %newValue, %foo
+  %x1 = load i8*, i8** @x, align 8
+  tail call void @objc_release(i8* %x1) nounwind, !clang.imprecise_release !0
+  store i8* %newValue, i8** @x, align 8
+  ret i1 %t
+}
+
 !0 = !{}
 
 ; CHECK: attributes [[NUW]] = { nounwind }
diff --git a/test/Transforms/ObjCARC/contract.ll b/test/Transforms/ObjCARC/contract.ll
index 2259e17ec58c..6ad46f2c85c5 100644
--- a/test/Transforms/ObjCARC/contract.ll
+++ b/test/Transforms/ObjCARC/contract.ll
@@ -169,7 +169,7 @@ return:                                           ; preds = %if.then, %entry
 ; CHECK-NOT: clang.arc.use
 ; CHECK: }
 define void @test9(i8* %a, i8* %b) {
-  call void (...)* @clang.arc.use(i8* %a, i8* %b) nounwind
+  call void (...) @clang.arc.use(i8* %a, i8* %b) nounwind
   ret void
 }
 
diff --git a/test/Transforms/ObjCARC/ensure-that-exception-unwind-path-is-visited.ll b/test/Transforms/ObjCARC/ensure-that-exception-unwind-path-is-visited.ll
index c72566cd2065..b875c3f039e6 100644
--- a/test/Transforms/ObjCARC/ensure-that-exception-unwind-path-is-visited.ll
+++ b/test/Transforms/ObjCARC/ensure-that-exception-unwind-path-is-visited.ll
@@ -22,29 +22,29 @@ target triple = "x86_64-apple-macosx10.9.0"
 @"OBJC_CLASS_$_NSObject" = external global %struct._class_t
 @"\01L_OBJC_CLASSLIST_REFERENCES_$_" = internal global %struct._class_t* @"OBJC_CLASS_$_NSObject", section "__DATA, __objc_classrefs, regular, no_dead_strip", align 8
 @"\01L_OBJC_METH_VAR_NAME_" = internal global [4 x i8] c"new\00", section "__TEXT,__objc_methname,cstring_literals", align 1
-@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([4 x i8], [4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
 @__CFConstantStringClassReference = external global [0 x i32]
 @.str = private unnamed_addr constant [11 x i8] c"Failed: %@\00", align 1
-@_unnamed_cfstring_ = private constant %struct.NSConstantString { i32* getelementptr inbounds ([0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 1992, i8* getelementptr inbounds ([11 x i8]* @.str, i32 0, i32 0), i64 10 }, section "__DATA,__cfstring"
+@_unnamed_cfstring_ = private constant %struct.NSConstantString { i32* getelementptr inbounds ([0 x i32], [0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 1992, i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str, i32 0, i32 0), i64 10 }, section "__DATA,__cfstring"
 @"OBJC_CLASS_$_NSException" = external global %struct._class_t
 @"\01L_OBJC_CLASSLIST_REFERENCES_$_1" = internal global %struct._class_t* @"OBJC_CLASS_$_NSException", section "__DATA, __objc_classrefs, regular, no_dead_strip", align 8
 @.str2 = private unnamed_addr constant [4 x i8] c"Foo\00", align 1
-@_unnamed_cfstring_3 = private constant %struct.NSConstantString { i32* getelementptr inbounds ([0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 1992, i8* getelementptr inbounds ([4 x i8]* @.str2, i32 0, i32 0), i64 3 }, section "__DATA,__cfstring"
+@_unnamed_cfstring_3 = private constant %struct.NSConstantString { i32* getelementptr inbounds ([0 x i32], [0 x i32]* @__CFConstantStringClassReference, i32 0, i32 0), i32 1992, i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str2, i32 0, i32 0), i64 3 }, section "__DATA,__cfstring"
 @"\01L_OBJC_METH_VAR_NAME_4" = internal global [14 x i8] c"raise:format:\00", section "__TEXT,__objc_methname,cstring_literals", align 1
-@"\01L_OBJC_SELECTOR_REFERENCES_5" = internal global i8* getelementptr inbounds ([14 x i8]* @"\01L_OBJC_METH_VAR_NAME_4", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
-@llvm.used = appending global [6 x i8*] [i8* bitcast (%struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_" to i8*), i8* getelementptr inbounds ([4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast (%struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_1" to i8*), i8* getelementptr inbounds ([14 x i8]* @"\01L_OBJC_METH_VAR_NAME_4", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_5" to i8*)], section "llvm.metadata"
+@"\01L_OBJC_SELECTOR_REFERENCES_5" = internal global i8* getelementptr inbounds ([14 x i8], [14 x i8]* @"\01L_OBJC_METH_VAR_NAME_4", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@llvm.used = appending global [6 x i8*] [i8* bitcast (%struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_" to i8*), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_" to i8*), i8* bitcast (%struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_1" to i8*), i8* getelementptr inbounds ([14 x i8], [14 x i8]* @"\01L_OBJC_METH_VAR_NAME_4", i32 0, i32 0), i8* bitcast (i8** @"\01L_OBJC_SELECTOR_REFERENCES_5" to i8*)], section "llvm.metadata"
 
 define i32 @main() uwtable ssp {
 entry:
-  %tmp = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_", align 8, !dbg !37
-  %tmp1 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8, !dbg !37, !invariant.load !38
+  %tmp = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_", align 8, !dbg !37
+  %tmp1 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8, !dbg !37, !invariant.load !38
   %tmp2 = bitcast %struct._class_t* %tmp to i8*, !dbg !37
 ; CHECK: call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp2, i8* %tmp1)
   %call = call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp2, i8* %tmp1), !dbg !37, !clang.arc.no_objc_arc_exceptions !38
-  call void @llvm.dbg.value(metadata i8* %call, i64 0, metadata i32 02, metadata !{}), !dbg !37
+  call void @llvm.dbg.value(metadata i8* %call, i64 0, metadata !25, metadata !DIExpression()), !dbg !37
 ; CHECK: call i8* @objc_retain(i8* %call) [[NUW:#[0-9]+]]
   %tmp3 = call i8* @objc_retain(i8* %call) nounwind, !dbg !39
-  call void @llvm.dbg.value(metadata i8* %call, i64 0, metadata !25, metadata !{}), !dbg !39
+  call void @llvm.dbg.value(metadata i8* %call, i64 0, metadata !25, metadata !DIExpression()), !dbg !39
   invoke fastcc void @ThrowFunc(i8* %call)
           to label %eh.cont unwind label %lpad, !dbg !40, !clang.arc.no_objc_arc_exceptions !38
 
@@ -58,15 +58,15 @@ lpad:                                             ; preds = %entry
           catch i8* null, !dbg !40
   %tmp5 = extractvalue { i8*, i32 } %tmp4, 0, !dbg !40
   %exn.adjusted = call i8* @objc_begin_catch(i8* %tmp5) nounwind, !dbg !44
-  call void @llvm.dbg.value(metadata i8 0, i64 0, metadata !21, metadata !{}), !dbg !46
+  call void @llvm.dbg.value(metadata i8 0, i64 0, metadata !21, metadata !DIExpression()), !dbg !46
   call void @objc_end_catch(), !dbg !49, !clang.arc.no_objc_arc_exceptions !38
 ; CHECK: call void @objc_release(i8* %call)
   call void @objc_release(i8* %call) nounwind, !dbg !42, !clang.imprecise_release !38
-  call void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring_ to i8*), i8* %call), !dbg !50, !clang.arc.no_objc_arc_exceptions !38
+  call void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring_ to i8*), i8* %call), !dbg !50, !clang.arc.no_objc_arc_exceptions !38
   br label %if.end, !dbg !52
 
 if.end:                                           ; preds = %lpad, %eh.cont
-  call void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring_ to i8*), i8* %call), !dbg !53, !clang.arc.no_objc_arc_exceptions !38
+  call void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring_ to i8*), i8* %call), !dbg !53, !clang.arc.no_objc_arc_exceptions !38
 ; CHECK: call void @objc_release(i8* %call)
   call void @objc_release(i8* %call) nounwind, !dbg !54, !clang.imprecise_release !38
   ret i32 0, !dbg !54
@@ -87,11 +87,11 @@ declare void @objc_exception_rethrow()
 define internal fastcc void @ThrowFunc(i8* %obj) uwtable noinline ssp {
 entry:
   %tmp = call i8* @objc_retain(i8* %obj) nounwind
-  call void @llvm.dbg.value(metadata i8* %obj, i64 0, metadata !32, metadata !{}), !dbg !55
-  %tmp1 = load %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_1", align 8, !dbg !56
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_5", align 8, !dbg !56, !invariant.load !38
+  call void @llvm.dbg.value(metadata i8* %obj, i64 0, metadata !32, metadata !DIExpression()), !dbg !55
+  %tmp1 = load %struct._class_t*, %struct._class_t** @"\01L_OBJC_CLASSLIST_REFERENCES_$_1", align 8, !dbg !56
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_5", align 8, !dbg !56, !invariant.load !38
   %tmp3 = bitcast %struct._class_t* %tmp1 to i8*, !dbg !56
-  call void (i8*, i8*, %0*, %0*, ...)* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %0*, %0*, ...)*)(i8* %tmp3, i8* %tmp2, %0* bitcast (%struct.NSConstantString* @_unnamed_cfstring_3 to %0*), %0* bitcast (%struct.NSConstantString* @_unnamed_cfstring_3 to %0*)), !dbg !56, !clang.arc.no_objc_arc_exceptions !38
+  call void (i8*, i8*, %0*, %0*, ...) bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %0*, %0*, ...)*)(i8* %tmp3, i8* %tmp2, %0* bitcast (%struct.NSConstantString* @_unnamed_cfstring_3 to %0*), %0* bitcast (%struct.NSConstantString* @_unnamed_cfstring_3 to %0*)), !dbg !56, !clang.arc.no_objc_arc_exceptions !38
   call void @objc_release(i8* %obj) nounwind, !dbg !58, !clang.imprecise_release !38
   ret void, !dbg !58
 }
@@ -113,62 +113,60 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!33, !34, !35, !36, !61}
 
-!0 = !{!"0x11\0016\00clang version 3.3 \001\00\002\00\000", !60, !1, !1, !3, !1, null} ; [ DW_TAG_compile_unit ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m] [DW_LANG_ObjC]
-!1 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_ObjC, producer: "clang version 3.3 ", isOptimized: true, runtimeVersion: 2, emissionKind: 0, file: !60, enums: !1, retainedTypes: !1, subprograms: !3, globals: !1)
+!1 = !{}
 !3 = !{!5, !27}
-!5 = !{!"0x2e\00main\00main\00\009\000\001\000\006\000\001\0010", !60, !6, !7, null, i32 ()* @main, null, null, !10} ; [ DW_TAG_subprogram ] [line 9] [def] [scope 10] [main]
-!6 = !{!"0x29", !60} ; [ DW_TAG_file_type ]
-!7 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !8, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!5 = !DISubprogram(name: "main", line: 9, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: true, scopeLine: 10, file: !60, scope: !6, type: !7, function: i32 ()* @main, variables: !11)
+!6 = !DIFile(filename: "test.m", directory: "/Volumes/Files/gottesmmcab/Radar/12906997")
+!7 = !DISubroutineType(types: !8)
 !8 = !{!9}
-!9 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!10 = !{!11}
+!9 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !11 = !{!12, !21, !25}
-!12 = !{!"0x100\00obj\0011\000", !13, !6, !14} ; [ DW_TAG_auto_variable ] [obj] [line 11]
-!13 = !{!"0xb\0010\000\000", !60, !5} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!14 = !{!"0x16\00id\0011\000\000\000\000", !60, null, !15} ; [ DW_TAG_typedef ] [id] [line 11, size 0, align 0, offset 0] [from ]
-!15 = !{!"0xf\00\000\0064\0064\000\000", !60, null, !16} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from objc_object]
-!16 = !{!"0x13\00objc_object\000\000\000\000\000\000", !60, null, null, !17, null, i32 0, null} ; [ DW_TAG_structure_type ] [objc_object] [line 0, size 0, align 0, offset 0] [def] [from ]
+!12 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "obj", line: 11, scope: !13, file: !6, type: !14)
+!13 = distinct !DILexicalBlock(line: 10, column: 0, file: !60, scope: !5)
+!14 = !DIDerivedType(tag: DW_TAG_typedef, name: "id", line: 11, file: !60, baseType: !15)
+!15 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, file: !60, baseType: !16)
+!16 = !DICompositeType(tag: DW_TAG_structure_type, name: "objc_object", file: !60, elements: !17)
 !17 = !{!18}
-!18 = !{!"0xd\00isa\000\0064\000\000\000", !60, !16, !19} ; [ DW_TAG_member ] [isa] [line 0, size 64, align 0, offset 0] [from ]
-!19 = !{!"0xf\00\000\0064\000\000\000", null, null, !20} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 0, offset 0] [from objc_class]
-!20 = !{!"0x13\00objc_class\000\000\000\000\004\000", !60, null, null, null, null, null, null} ; [ DW_TAG_structure_type ] [objc_class] [line 0, size 0, align 0, offset 0] [decl] [from ]
-!21 = !{!"0x100\00ok\0013\000", !22, !6, !23} ; [ DW_TAG_auto_variable ] [ok] [line 13]
-!22 = !{!"0xb\0012\000\001", !60, !13} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!23 = !{!"0x16\00BOOL\0062\000\000\000\000", !60, null, !24} ; [ DW_TAG_typedef ] [BOOL] [line 62, size 0, align 0, offset 0] [from signed char]
-!24 = !{!"0x24\00signed char\000\008\008\000\000\006", null, null} ; [ DW_TAG_base_type ] [signed char] [line 0, size 8, align 8, offset 0, enc DW_ATE_signed_char]
-!25 = !{!"0x100\00obj2\0015\000", !26, !6, !14} ; [ DW_TAG_auto_variable ] [obj2] [line 15]
-!26 = !{!"0xb\0014\000\002", !60, !22} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!27 = !{!"0x2e\00ThrowFunc\00ThrowFunc\00\004\001\001\000\006\00256\001\005", !60, !6, !28, null, void (i8*)* @ThrowFunc, null, null, !30} ; [ DW_TAG_subprogram ] [line 4] [local] [def] [scope 5] [ThrowFunc]
-!28 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !29, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!18 = !DIDerivedType(tag: DW_TAG_member, name: "isa", size: 64, file: !60, scope: !16, baseType: !19)
+!19 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, baseType: !20)
+!20 = !DICompositeType(tag: DW_TAG_structure_type, name: "objc_class", flags: DIFlagFwdDecl, file: !60)
+!21 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "ok", line: 13, scope: !22, file: !6, type: !23)
+!22 = distinct !DILexicalBlock(line: 12, column: 0, file: !60, scope: !13)
+!23 = !DIDerivedType(tag: DW_TAG_typedef, name: "BOOL", line: 62, file: !60, baseType: !24)
+!24 = !DIBasicType(tag: DW_TAG_base_type, name: "signed char", size: 8, align: 8, encoding: DW_ATE_signed_char)
+!25 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "obj2", line: 15, scope: !26, file: !6, type: !14)
+!26 = distinct !DILexicalBlock(line: 14, column: 0, file: !60, scope: !22)
+!27 = !DISubprogram(name: "ThrowFunc", line: 4, isLocal: true, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 5, file: !60, scope: !6, type: !28, function: void (i8*)* @ThrowFunc, variables: !31)
+!28 = !DISubroutineType(types: !29)
 !29 = !{null, !14}
-!30 = !{!31}
 !31 = !{!32}
-!32 = !{!"0x101\00obj\0016777220\000", !27, !6, !14} ; [ DW_TAG_arg_variable ] [obj] [line 4]
+!32 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "obj", line: 4, arg: 1, scope: !27, file: !6, type: !14)
 !33 = !{i32 1, !"Objective-C Version", i32 2}
 !34 = !{i32 1, !"Objective-C Image Info Version", i32 0}
 !35 = !{i32 1, !"Objective-C Image Info Section", !"__DATA, __objc_imageinfo, regular, no_dead_strip"}
 !36 = !{i32 4, !"Objective-C Garbage Collection", i32 0}
-!37 = !MDLocation(line: 11, scope: !13)
+!37 = !DILocation(line: 11, scope: !13)
 !38 = !{}
-!39 = !MDLocation(line: 15, scope: !26)
-!40 = !MDLocation(line: 17, scope: !41)
-!41 = !{!"0xb\0016\000\003", !60, !26} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!42 = !MDLocation(line: 22, scope: !26)
-!43 = !MDLocation(line: 23, scope: !22)
-!44 = !MDLocation(line: 19, scope: !41)
+!39 = !DILocation(line: 15, scope: !26)
+!40 = !DILocation(line: 17, scope: !41)
+!41 = distinct !DILexicalBlock(line: 16, column: 0, file: !60, scope: !26)
+!42 = !DILocation(line: 22, scope: !26)
+!43 = !DILocation(line: 23, scope: !22)
+!44 = !DILocation(line: 19, scope: !41)
 !45 = !{i8 0}
-!46 = !MDLocation(line: 20, scope: !47)
-!47 = !{!"0xb\0019\000\005", !60, !48} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!48 = !{!"0xb\0019\000\004", !60, !26} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!49 = !MDLocation(line: 21, scope: !47)
-!50 = !MDLocation(line: 24, scope: !51)
-!51 = !{!"0xb\0023\000\006", !60, !22} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!52 = !MDLocation(line: 25, scope: !51)
-!53 = !MDLocation(line: 27, scope: !13)
-!54 = !MDLocation(line: 28, scope: !13)
-!55 = !MDLocation(line: 4, scope: !27)
-!56 = !MDLocation(line: 6, scope: !57)
-!57 = !{!"0xb\005\000\007", !60, !27} ; [ DW_TAG_lexical_block ] [/Volumes/Files/gottesmmcab/Radar/12906997/test.m]
-!58 = !MDLocation(line: 7, scope: !57)
-!60 = !{!"test.m", !"/Volumes/Files/gottesmmcab/Radar/12906997"}
-!61 = !{i32 1, !"Debug Info Version", i32 2}
+!46 = !DILocation(line: 20, scope: !47)
+!47 = distinct !DILexicalBlock(line: 19, column: 0, file: !60, scope: !48)
+!48 = distinct !DILexicalBlock(line: 19, column: 0, file: !60, scope: !26)
+!49 = !DILocation(line: 21, scope: !47)
+!50 = !DILocation(line: 24, scope: !51)
+!51 = distinct !DILexicalBlock(line: 23, column: 0, file: !60, scope: !22)
+!52 = !DILocation(line: 25, scope: !51)
+!53 = !DILocation(line: 27, scope: !13)
+!54 = !DILocation(line: 28, scope: !13)
+!55 = !DILocation(line: 4, scope: !27)
+!56 = !DILocation(line: 6, scope: !57)
+!57 = distinct !DILexicalBlock(line: 5, column: 0, file: !60, scope: !27)
+!58 = !DILocation(line: 7, scope: !57)
+!60 = !DIFile(filename: "test.m", directory: "/Volumes/Files/gottesmmcab/Radar/12906997")
+!61 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/ObjCARC/escape.ll b/test/Transforms/ObjCARC/escape.ll
index 357f7592d83c..c7a1b03c16fc 100644
--- a/test/Transforms/ObjCARC/escape.ll
+++ b/test/Transforms/ObjCARC/escape.ll
@@ -17,43 +17,43 @@ define void @test0() nounwind {
 entry:
   %weakLogNTimes = alloca %struct.__block_byref_weakLogNTimes, align 8
   %block = alloca <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, align 8
-  %byref.isa = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 0
+  %byref.isa = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 0
   store i8* null, i8** %byref.isa, align 8
-  %byref.forwarding = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 1
+  %byref.forwarding = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 1
   store %struct.__block_byref_weakLogNTimes* %weakLogNTimes, %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
-  %byref.flags = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 2
+  %byref.flags = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 2
   store i32 33554432, i32* %byref.flags, align 8
-  %byref.size = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 3
+  %byref.size = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 3
   store i32 48, i32* %byref.size, align 4
-  %tmp1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 4
+  %tmp1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 4
   store i8* bitcast (void (i8*, i8*)* @__Block_byref_object_copy_ to i8*), i8** %tmp1, align 8
-  %tmp2 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 5
+  %tmp2 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 5
   store i8* bitcast (void (i8*)* @__Block_byref_object_dispose_ to i8*), i8** %tmp2, align 8
-  %weakLogNTimes1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 6
+  %weakLogNTimes1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 6
   %tmp3 = bitcast void (...)** %weakLogNTimes1 to i8**
   %tmp4 = call i8* @objc_initWeak(i8** %tmp3, i8* null) nounwind
-  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 0
+  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 0
   store i8* null, i8** %block.isa, align 8
-  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 1
+  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 1
   store i32 1107296256, i32* %block.flags, align 8
-  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 2
+  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 2
   store i32 0, i32* %block.reserved, align 4
-  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 3
+  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 3
   store i8* bitcast (void (i8*, i32)* @__main_block_invoke_0 to i8*), i8** %block.invoke, align 8
-  %block.descriptor = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 4
+  %block.descriptor = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 4
   store %struct.__block_descriptor* null, %struct.__block_descriptor** %block.descriptor, align 8
-  %block.captured = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 5
+  %block.captured = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 5
   %tmp5 = bitcast %struct.__block_byref_weakLogNTimes* %weakLogNTimes to i8*
   store i8* %tmp5, i8** %block.captured, align 8
   %tmp6 = bitcast <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block to i8*
   %tmp7 = call i8* @objc_retainBlock(i8* %tmp6) nounwind, !clang.arc.copy_on_escape !0
-  %tmp8 = load %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
-  %weakLogNTimes3 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %tmp8, i64 0, i32 6
+  %tmp8 = load %struct.__block_byref_weakLogNTimes*, %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
+  %weakLogNTimes3 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %tmp8, i64 0, i32 6
   %tmp9 = bitcast void (...)** %weakLogNTimes3 to i8**
   %tmp10 = call i8* @objc_storeWeak(i8** %tmp9, i8* %tmp7) nounwind
-  %tmp11 = getelementptr inbounds i8* %tmp7, i64 16
+  %tmp11 = getelementptr inbounds i8, i8* %tmp7, i64 16
   %tmp12 = bitcast i8* %tmp11 to i8**
-  %tmp13 = load i8** %tmp12, align 8
+  %tmp13 = load i8*, i8** %tmp12, align 8
   %tmp14 = bitcast i8* %tmp13 to void (i8*, i32)*
   call void %tmp14(i8* %tmp7, i32 10) nounwind, !clang.arc.no_objc_arc_exceptions !0
   call void @objc_release(i8* %tmp7) nounwind, !clang.imprecise_release !0
@@ -72,43 +72,43 @@ define void @test1() nounwind {
 entry:
   %weakLogNTimes = alloca %struct.__block_byref_weakLogNTimes, align 8
   %block = alloca <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, align 8
-  %byref.isa = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 0
+  %byref.isa = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 0
   store i8* null, i8** %byref.isa, align 8
-  %byref.forwarding = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 1
+  %byref.forwarding = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 1
   store %struct.__block_byref_weakLogNTimes* %weakLogNTimes, %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
-  %byref.flags = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 2
+  %byref.flags = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 2
   store i32 33554432, i32* %byref.flags, align 8
-  %byref.size = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 3
+  %byref.size = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 3
   store i32 48, i32* %byref.size, align 4
-  %tmp1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 4
+  %tmp1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 4
   store i8* bitcast (void (i8*, i8*)* @__Block_byref_object_copy_ to i8*), i8** %tmp1, align 8
-  %tmp2 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 5
+  %tmp2 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 5
   store i8* bitcast (void (i8*)* @__Block_byref_object_dispose_ to i8*), i8** %tmp2, align 8
-  %weakLogNTimes1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 6
+  %weakLogNTimes1 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %weakLogNTimes, i64 0, i32 6
   %tmp3 = bitcast void (...)** %weakLogNTimes1 to i8**
   %tmp4 = call i8* @objc_initWeak(i8** %tmp3, i8* null) nounwind
-  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 0
+  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 0
   store i8* null, i8** %block.isa, align 8
-  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 1
+  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 1
   store i32 1107296256, i32* %block.flags, align 8
-  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 2
+  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 2
   store i32 0, i32* %block.reserved, align 4
-  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 3
+  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 3
   store i8* bitcast (void (i8*, i32)* @__main_block_invoke_0 to i8*), i8** %block.invoke, align 8
-  %block.descriptor = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 4
+  %block.descriptor = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 4
   store %struct.__block_descriptor* null, %struct.__block_descriptor** %block.descriptor, align 8
-  %block.captured = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 5
+  %block.captured = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block, i64 0, i32 5
   %tmp5 = bitcast %struct.__block_byref_weakLogNTimes* %weakLogNTimes to i8*
   store i8* %tmp5, i8** %block.captured, align 8
   %tmp6 = bitcast <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, i8* }>* %block to i8*
   %tmp7 = call i8* @objc_retainBlock(i8* %tmp6) nounwind, !clang.arc.copy_on_escape !0
-  %tmp8 = load %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
-  %weakLogNTimes3 = getelementptr inbounds %struct.__block_byref_weakLogNTimes* %tmp8, i64 0, i32 6
+  %tmp8 = load %struct.__block_byref_weakLogNTimes*, %struct.__block_byref_weakLogNTimes** %byref.forwarding, align 8
+  %weakLogNTimes3 = getelementptr inbounds %struct.__block_byref_weakLogNTimes, %struct.__block_byref_weakLogNTimes* %tmp8, i64 0, i32 6
   %tmp9 = bitcast void (...)** %weakLogNTimes3 to i8**
   %tmp10 = call i8* @not_really_objc_storeWeak(i8** %tmp9, i8* %tmp7) nounwind
-  %tmp11 = getelementptr inbounds i8* %tmp7, i64 16
+  %tmp11 = getelementptr inbounds i8, i8* %tmp7, i64 16
   %tmp12 = bitcast i8* %tmp11 to i8**
-  %tmp13 = load i8** %tmp12, align 8
+  %tmp13 = load i8*, i8** %tmp12, align 8
   %tmp14 = bitcast i8* %tmp13 to void (i8*, i32)*
   call void %tmp14(i8* %tmp7, i32 10) nounwind, !clang.arc.no_objc_arc_exceptions !0
   call void @objc_release(i8* %tmp7) nounwind, !clang.imprecise_release !0
diff --git a/test/Transforms/ObjCARC/gvn.ll b/test/Transforms/ObjCARC/gvn.ll
index 2d120e7345a5..6f828545bc6f 100644
--- a/test/Transforms/ObjCARC/gvn.ll
+++ b/test/Transforms/ObjCARC/gvn.ll
@@ -11,15 +11,15 @@ declare i32 @objc_sync_exit(i8*)
 
 ; CHECK: define i8* @test0(i32 %n)
 ; CHECK-NEXT: entry:
-; CHECK-NEXT: %s = load i8** @x
+; CHECK-NEXT: %s = load i8*, i8** @x
 ; CHECK-NOT: load
 ; CHECK: ret i8* %s
 ; CHECK-NEXT: }
 define i8* @test0(i32 %n) nounwind {
 entry:
-  %s = load i8** @x
+  %s = load i8*, i8** @x
   %0 = tail call i8* @objc_retain(i8* %s) nounwind
-  %t = load i8** @x
+  %t = load i8*, i8** @x
   ret i8* %t
 }
 
@@ -33,9 +33,9 @@ entry:
 ; CHECK: }
 define i8* @test1(i32 %n) nounwind {
 entry:
-  %s = load i8** @x
+  %s = load i8*, i8** @x
   %0 = call i32 @objc_sync_enter(i8* %s)
-  %t = load i8** @x
+  %t = load i8*, i8** @x
   %1 = call i32 @objc_sync_exit(i8* %s)
   ret i8* %t
 }
diff --git a/test/Transforms/ObjCARC/intrinsic-use-isolated.ll b/test/Transforms/ObjCARC/intrinsic-use-isolated.ll
index f5c31fd82ecb..03d7520dde91 100644
--- a/test/Transforms/ObjCARC/intrinsic-use-isolated.ll
+++ b/test/Transforms/ObjCARC/intrinsic-use-isolated.ll
@@ -10,7 +10,7 @@ declare void @clang.arc.use(...) nounwind
 ; CHECK-NOT: clang.arc.use
 ; CHECK: }
 define void @test0(i8* %a, i8* %b) {
-  call void (...)* @clang.arc.use(i8* %a, i8* %b) nounwind
+  call void (...) @clang.arc.use(i8* %a, i8* %b) nounwind
   ret void
 }
 
diff --git a/test/Transforms/ObjCARC/intrinsic-use.ll b/test/Transforms/ObjCARC/intrinsic-use.ll
index b1e56c8e9de1..f75b1872b172 100644
--- a/test/Transforms/ObjCARC/intrinsic-use.ll
+++ b/test/Transforms/ObjCARC/intrinsic-use.ll
@@ -22,14 +22,14 @@ declare void @test0_helper(i8*, i8**)
 ; CHECK-NEXT:   store i8* %y, i8** %temp0
 ; CHECK-NEXT:   @objc_retain(i8* %y)
 ; CHECK-NEXT:   call void @test0_helper
-; CHECK-NEXT:   [[VAL1:%.*]] = load i8** %temp0
-; CHECK-NEXT:   call void (...)* @clang.arc.use(i8* %y)
+; CHECK-NEXT:   [[VAL1:%.*]] = load i8*, i8** %temp0
+; CHECK-NEXT:   call void (...) @clang.arc.use(i8* %y)
 ; CHECK-NEXT:   @objc_retain(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_release(i8* %y)
 ; CHECK-NEXT:   store i8* [[VAL1]], i8** %temp1
 ; CHECK-NEXT:   call void @test0_helper
-; CHECK-NEXT:   [[VAL2:%.*]] = load i8** %temp1
-; CHECK-NEXT:   call void (...)* @clang.arc.use(i8* [[VAL1]])
+; CHECK-NEXT:   [[VAL2:%.*]] = load i8*, i8** %temp1
+; CHECK-NEXT:   call void (...) @clang.arc.use(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_retain(i8* [[VAL2]])
 ; CHECK-NEXT:   @objc_release(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_autorelease(i8* %x)
@@ -47,15 +47,15 @@ entry:
   %1 = call i8* @objc_retain(i8* %y) nounwind
   store i8* %y, i8** %temp0
   call void @test0_helper(i8* %x, i8** %temp0)
-  %val1 = load i8** %temp0
+  %val1 = load i8*, i8** %temp0
   %2 = call i8* @objc_retain(i8* %val1) nounwind
-  call void (...)* @clang.arc.use(i8* %y) nounwind
+  call void (...) @clang.arc.use(i8* %y) nounwind
   call void @objc_release(i8* %y) nounwind
   store i8* %val1, i8** %temp1
   call void @test0_helper(i8* %x, i8** %temp1)
-  %val2 = load i8** %temp1
+  %val2 = load i8*, i8** %temp1
   %3 = call i8* @objc_retain(i8* %val2) nounwind
-  call void (...)* @clang.arc.use(i8* %val1) nounwind
+  call void (...) @clang.arc.use(i8* %val1) nounwind
   call void @objc_release(i8* %val1) nounwind
   %4 = call i8* @objc_retain(i8* %x) nounwind
   %5 = call i8* @objc_autorelease(i8* %x) nounwind
@@ -70,14 +70,14 @@ entry:
 ; CHECK-NEXT:   store i8* %y, i8** %temp0
 ; CHECK-NEXT:   @objc_retain(i8* %y)
 ; CHECK-NEXT:   call void @test0_helper
-; CHECK-NEXT:   [[VAL1:%.*]] = load i8** %temp0
-; CHECK-NEXT:   call void (...)* @clang.arc.use(i8* %y)
+; CHECK-NEXT:   [[VAL1:%.*]] = load i8*, i8** %temp0
+; CHECK-NEXT:   call void (...) @clang.arc.use(i8* %y)
 ; CHECK-NEXT:   @objc_retain(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_release(i8* %y)
 ; CHECK-NEXT:   store i8* [[VAL1]], i8** %temp1
 ; CHECK-NEXT:   call void @test0_helper
-; CHECK-NEXT:   [[VAL2:%.*]] = load i8** %temp1
-; CHECK-NEXT:   call void (...)* @clang.arc.use(i8* [[VAL1]])
+; CHECK-NEXT:   [[VAL2:%.*]] = load i8*, i8** %temp1
+; CHECK-NEXT:   call void (...) @clang.arc.use(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_retain(i8* [[VAL2]])
 ; CHECK-NEXT:   @objc_release(i8* [[VAL1]])
 ; CHECK-NEXT:   @objc_autorelease(i8* %x)
@@ -93,15 +93,15 @@ entry:
   %1 = call i8* @objc_retain(i8* %y) nounwind
   store i8* %y, i8** %temp0
   call void @test0_helper(i8* %x, i8** %temp0)
-  %val1 = load i8** %temp0
+  %val1 = load i8*, i8** %temp0
   %2 = call i8* @objc_retain(i8* %val1) nounwind
-  call void (...)* @clang.arc.use(i8* %y) nounwind
+  call void (...) @clang.arc.use(i8* %y) nounwind
   call void @objc_release(i8* %y) nounwind, !clang.imprecise_release !0
   store i8* %val1, i8** %temp1
   call void @test0_helper(i8* %x, i8** %temp1)
-  %val2 = load i8** %temp1
+  %val2 = load i8*, i8** %temp1
   %3 = call i8* @objc_retain(i8* %val2) nounwind
-  call void (...)* @clang.arc.use(i8* %val1) nounwind
+  call void (...) @clang.arc.use(i8* %val1) nounwind
   call void @objc_release(i8* %val1) nounwind, !clang.imprecise_release !0
   %4 = call i8* @objc_retain(i8* %x) nounwind
   %5 = call i8* @objc_autorelease(i8* %x) nounwind
diff --git a/test/Transforms/ObjCARC/move-and-form-retain-autorelease.ll b/test/Transforms/ObjCARC/move-and-form-retain-autorelease.ll
index 5d058257c6ed..9894eb4f5345 100644
--- a/test/Transforms/ObjCARC/move-and-form-retain-autorelease.ll
+++ b/test/Transforms/ObjCARC/move-and-form-retain-autorelease.ll
@@ -80,20 +80,20 @@ declare i8* @objc_explicit_autorelease(i8*)
 
 define hidden %14* @foo(%15* %arg, %16* %arg2) {
 bb:
-  %tmp = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_3725", align 8
+  %tmp = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_3725", align 8
   %tmp4 = bitcast %15* %arg to i8*
   %tmp5 = tail call %18* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %18* (i8*, i8*)*)(i8* %tmp4, i8* %tmp)
   %tmp6 = bitcast %18* %tmp5 to i8*
   %tmp7 = tail call i8* @objc_retain(i8* %tmp6) nounwind
-  %tmp8 = load %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_40", align 8
-  %tmp9 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_4227", align 8
+  %tmp8 = load %2*, %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_40", align 8
+  %tmp9 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_4227", align 8
   %tmp10 = bitcast %2* %tmp8 to i8*
   %tmp11 = tail call %19* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %19* (i8*, i8*)*)(i8* %tmp10, i8* %tmp9)
-  %tmp12 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_4631", align 8
+  %tmp12 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_4631", align 8
   %tmp13 = bitcast %19* %tmp11 to i8*
   %tmp14 = tail call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*, %13*)*)(i8* %tmp13, i8* %tmp12, %13* bitcast (%12* @_unnamed_cfstring_386 to %13*))
   %tmp15 = bitcast %16* %arg2 to i8*
-  %tmp16 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_count" to i8**), align 16
+  %tmp16 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_count" to i8**), align 16
   %tmp17 = bitcast i8* %tmp16 to i64 (i8*, %1*)*
   %tmp18 = tail call i64 %tmp17(i8* %tmp15, %1* bitcast (%0* @"\01l_objc_msgSend_fixup_count" to %1*))
   %tmp19 = icmp eq i64 %tmp18, 0
@@ -110,104 +110,104 @@ bb22:                                             ; preds = %bb
 
 bb25:                                             ; preds = %bb22, %bb20
   %tmp26 = phi i1 [ %tmp21, %bb20 ], [ false, %bb22 ]
-  %tmp27 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_188", align 8
+  %tmp27 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_188", align 8
   %tmp28 = tail call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp7, i8* %tmp27)
   %tmp29 = tail call i8* @objc_explicit_autorelease(i8* %tmp28) nounwind
   %tmp30 = bitcast i8* %tmp29 to %18*
   tail call void @objc_release(i8* %tmp7) nounwind
-  %tmp31 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_389", align 8
+  %tmp31 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_389", align 8
   %tmp32 = tail call %20* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %20* (i8*, i8*)*)(i8* %tmp29, i8* %tmp31)
-  %tmp33 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_391", align 8
+  %tmp33 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_391", align 8
   %tmp34 = bitcast %20* %tmp32 to i8*
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %16*)*)(i8* %tmp34, i8* %tmp33, %16* %arg2)
   br i1 %tmp26, label %bb46, label %bb35
 
 bb35:                                             ; preds = %bb25
-  %tmp36 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_389", align 8
+  %tmp36 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_389", align 8
   %tmp37 = tail call %20* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %20* (i8*, i8*)*)(i8* %tmp29, i8* %tmp36)
-  %tmp38 = load %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_70", align 8
-  %tmp39 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_393", align 8
+  %tmp38 = load %2*, %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_70", align 8
+  %tmp39 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_393", align 8
   %tmp40 = bitcast %2* %tmp38 to i8*
   %tmp41 = tail call %21* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %21* (i8*, i8*, i8)*)(i8* %tmp40, i8* %tmp39, i8 signext 1)
   %tmp42 = bitcast %21* %tmp41 to i8*
-  %tmp43 = load %13** @NSPrintHeaderAndFooter, align 8
-  %tmp44 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_159", align 8
+  %tmp43 = load %13*, %13** @NSPrintHeaderAndFooter, align 8
+  %tmp44 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_159", align 8
   %tmp45 = bitcast %20* %tmp37 to i8*
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i8*, %13*)*)(i8* %tmp45, i8* %tmp44, i8* %tmp42, %13* %tmp43)
   br label %bb46
 
 bb46:                                             ; preds = %bb35, %bb25, %bb22
   %tmp47 = phi %18* [ %tmp30, %bb35 ], [ %tmp30, %bb25 ], [ %tmp23, %bb22 ]
-  %tmp48 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
+  %tmp48 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
   %tmp49 = tail call %22* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %22* (i8*, i8*)*)(i8* %tmp4, i8* %tmp48)
   %tmp50 = bitcast %22* %tmp49 to i8*
-  %tmp51 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_count" to i8**), align 16
+  %tmp51 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_count" to i8**), align 16
   %tmp52 = bitcast i8* %tmp51 to i64 (i8*, %1*)*
   %tmp53 = tail call i64 %tmp52(i8* %tmp50, %1* bitcast (%0* @"\01l_objc_msgSend_fixup_count" to %1*))
   %tmp54 = icmp eq i64 %tmp53, 0
   br i1 %tmp54, label %bb55, label %bb57
 
 bb55:                                             ; preds = %bb46
-  %tmp56 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_395", align 8
+  %tmp56 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_395", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*)*)(i8* %tmp4, i8* %tmp56)
   br label %bb57
 
 bb57:                                             ; preds = %bb55, %bb46
-  %tmp58 = load %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_396", align 8
-  %tmp59 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
+  %tmp58 = load %2*, %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_396", align 8
+  %tmp59 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
   %tmp60 = tail call %22* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %22* (i8*, i8*)*)(i8* %tmp4, i8* %tmp59)
   %tmp61 = bitcast %22* %tmp60 to i8*
-  %tmp62 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to i8**), align 16
+  %tmp62 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to i8**), align 16
   %tmp63 = bitcast i8* %tmp62 to i8* (i8*, %1*, i64)*
   %tmp64 = tail call i8* %tmp63(i8* %tmp61, %1* bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to %1*), i64 0)
-  %tmp65 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_398", align 8
+  %tmp65 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_398", align 8
   %tmp66 = tail call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp64, i8* %tmp65)
   %tmp67 = bitcast i8* %tmp66 to %23*
-  %tmp68 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_400", align 8
+  %tmp68 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_400", align 8
   %tmp69 = bitcast %2* %tmp58 to i8*
   %tmp70 = tail call %14* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %14* (i8*, i8*, %23*, %18*)*)(i8* %tmp69, i8* %tmp68, %23* %tmp67, %18* %tmp47)
   %tmp71 = bitcast %14* %tmp70 to i8*
   ; hack to prevent the optimize from using objc_retainAutoreleasedReturnValue.
-  %tmp71x = getelementptr i8* %tmp71, i64 1
+  %tmp71x = getelementptr i8, i8* %tmp71, i64 1
   %tmp72 = tail call i8* @objc_retain(i8* %tmp71x) nounwind
-  %tmp73 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_402", align 8
+  %tmp73 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_402", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i8)*)(i8* %tmp72, i8* %tmp73, i8 signext 1)
-  %tmp74 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_404", align 8
+  %tmp74 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_404", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i8)*)(i8* %tmp72, i8* %tmp74, i8 signext 1)
-  %tmp75 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
+  %tmp75 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_328", align 8
   %tmp76 = tail call %22* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %22* (i8*, i8*)*)(i8* %tmp4, i8* %tmp75)
   %tmp77 = bitcast %22* %tmp76 to i8*
-  %tmp78 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to i8**), align 16
+  %tmp78 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to i8**), align 16
   %tmp79 = bitcast i8* %tmp78 to i8* (i8*, %1*, i64)*
   %tmp80 = tail call i8* %tmp79(i8* %tmp77, %1* bitcast (%0* @"\01l_objc_msgSend_fixup_objectAtIndex_" to %1*), i64 0)
-  %tmp81 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_406", align 8
+  %tmp81 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_406", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i64)*)(i8* %tmp80, i8* %tmp81, i64 9223372036854775807)
-  %tmp82 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_408", align 8
+  %tmp82 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_408", align 8
   %tmp83 = tail call %24* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %24* (i8*, i8*)*)(i8* %tmp72, i8* %tmp82)
   %tmp84 = bitcast %24* %tmp83 to i8*
   %tmp85 = tail call i8* @objc_retain(i8* %tmp84) nounwind
-  %tmp86 = load %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_409", align 8
+  %tmp86 = load %2*, %2** @"\01L_OBJC_CLASSLIST_REFERENCES_$_409", align 8
   %tmp87 = bitcast %2* %tmp86 to i8*
-  %tmp88 = load i8** bitcast (%0* @"\01l_objc_msgSend_fixup_alloc" to i8**), align 16
+  %tmp88 = load i8*, i8** bitcast (%0* @"\01l_objc_msgSend_fixup_alloc" to i8**), align 16
   %tmp89 = bitcast i8* %tmp88 to i8* (i8*, %1*)*
   %tmp90 = tail call i8* %tmp89(i8* %tmp87, %1* bitcast (%0* @"\01l_objc_msgSend_fixup_alloc" to %1*))
-  %tmp91 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_8", align 8
+  %tmp91 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_8", align 8
   %tmp92 = tail call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp90, i8* %tmp91)
   %tmp93 = tail call i8* @objc_explicit_autorelease(i8* %tmp92) nounwind
   %tmp94 = bitcast i8* %tmp93 to %25*
-  %tmp95 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_411", align 8
+  %tmp95 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_411", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, %25*)*)(i8* %tmp85, i8* %tmp95, %25* %tmp94)
   tail call void @objc_release(i8* %tmp93) nounwind
-  %tmp96 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_148", align 8
+  %tmp96 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_148", align 8
   %tmp97 = tail call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*)*)(i8* %tmp4, i8* %tmp96)
   %tmp98 = icmp eq i8 %tmp97, 0
   br i1 %tmp98, label %bb99, label %bb104
 
 bb99:                                             ; preds = %bb57
-  %tmp100 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_413", align 8
+  %tmp100 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_413", align 8
   %tmp101 = tail call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*)*)(i8* %tmp85, i8* %tmp100)
   %tmp102 = or i64 %tmp101, 12
-  %tmp103 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_415", align 8
+  %tmp103 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_415", align 8
   tail call void bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to void (i8*, i8*, i64)*)(i8* %tmp85, i8* %tmp103, i64 %tmp102)
   br label %bb104
 
diff --git a/test/Transforms/ObjCARC/move-and-merge-autorelease.ll b/test/Transforms/ObjCARC/move-and-merge-autorelease.ll
index e5d2f07e45a2..0a68541d9350 100644
--- a/test/Transforms/ObjCARC/move-and-merge-autorelease.ll
+++ b/test/Transforms/ObjCARC/move-and-merge-autorelease.ll
@@ -35,10 +35,10 @@ declare i8* @objc_autorelease(i8*)
 
 define hidden %0* @foo(%1* %arg, %3* %arg3) {
 bb:
-  %tmp16 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_620", align 8
+  %tmp16 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_620", align 8
   %tmp17 = bitcast %3* %arg3 to i8*
   %tmp18 = call %4* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %4* (i8*, i8*)*)(i8* %tmp17, i8* %tmp16)
-  %tmp19 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_622", align 8
+  %tmp19 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_622", align 8
   %tmp20 = bitcast %4* %tmp18 to i8*
   %tmp21 = call %5* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %5* (i8*, i8*)*)(i8* %tmp20, i8* %tmp19)
   %tmp22 = bitcast %5* %tmp21 to i8*
@@ -48,11 +48,11 @@ bb:
   br i1 %tmp26, label %bb81, label %bb27
 
 bb27:                                             ; preds = %bb
-  %tmp29 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_11", align 8
+  %tmp29 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_11", align 8
   %tmp30 = bitcast %1* %arg to i8*
   %tmp31 = call i8* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*)*)(i8* %tmp30, i8* %tmp29)
   %tmp34 = call i8* @objc_retain(i8* %tmp31) nounwind
-  %tmp37 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_421455", align 8
+  %tmp37 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_421455", align 8
   %tmp39 = call %0* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %0* (i8*, i8*)*)(i8* %tmp34, i8* %tmp37)
   %tmp40 = bitcast %0* %tmp39 to i8*
   %tmp41 = call i8* @objc_retain(i8* %tmp40) nounwind
@@ -61,7 +61,7 @@ bb27:                                             ; preds = %bb
   br i1 %tmp44, label %bb45, label %bb55
 
 bb45:                                             ; preds = %bb27
-  %tmp47 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_624", align 8
+  %tmp47 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_624", align 8
   %tmp49 = call %0* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %0* (i8*, i8*)*)(i8* %tmp34, i8* %tmp47)
   %tmp51 = bitcast %0* %tmp49 to i8*
   %tmp52 = call i8* @objc_retain(i8* %tmp51) nounwind
@@ -74,14 +74,14 @@ bb55:                                             ; preds = %bb27, %bb45
   br i1 %tmp57, label %bb76, label %bb58
 
 bb58:                                             ; preds = %bb55
-  %tmp60 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_598", align 8
+  %tmp60 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_598", align 8
   %tmp61 = bitcast %0* %tmp13.0 to i8*
   %tmp62 = call signext i8 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8 (i8*, i8*)*)(i8* %tmp61, i8* %tmp60)
   %tmp64 = icmp eq i8 %tmp62, 0
   br i1 %tmp64, label %bb76, label %bb65
 
 bb65:                                             ; preds = %bb58
-  %tmp68 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_626", align 8
+  %tmp68 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_626", align 8
   %tmp69 = bitcast %0* %tmp13.0 to i8*
   %tmp70 = call %0* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to %0* (i8*, i8*, %5*)*)(i8* %tmp69, i8* %tmp68, %5* %tmp24)
   %tmp72 = bitcast %0* %tmp70 to i8*
@@ -99,9 +99,9 @@ bb81:                                             ; preds = %bb, %bb76
   %tmp10.1 = phi %0* [ %tmp10.0, %bb76 ], [ null, %bb ]
   %tmp83 = bitcast %0* %tmp10.1 to i8*
   %tmp84 = call i8* @objc_retain(i8* %tmp83) nounwind
-  %tmp88 = bitcast i8* %tmp87 to %0*
   call void @objc_release(i8* %tmp23) nounwind
   %tmp87 = call i8* @objc_autorelease(i8* %tmp84) nounwind
+  %tmp88 = bitcast i8* %tmp87 to %0*
   %tmp92 = bitcast %0* %tmp10.1 to i8*
   call void @objc_release(i8* %tmp92) nounwind
   ret %0* %tmp88
diff --git a/test/Transforms/ObjCARC/nested.ll b/test/Transforms/ObjCARC/nested.ll
index 7d72e37f8034..464426abfb06 100644
--- a/test/Transforms/ObjCARC/nested.ll
+++ b/test/Transforms/ObjCARC/nested.ll
@@ -3,7 +3,7 @@
 %struct.__objcFastEnumerationState = type { i64, i8**, i64*, [5 x i64] }
 
 @"\01L_OBJC_METH_VAR_NAME_" = internal global [43 x i8] c"countByEnumeratingWithState:objects:count:\00", section "__TEXT,__objc_methname,cstring_literals", align 1
-@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([43 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
+@"\01L_OBJC_SELECTOR_REFERENCES_" = internal global i8* getelementptr inbounds ([43 x i8], [43 x i8]* @"\01L_OBJC_METH_VAR_NAME_", i64 0, i64 0), section "__DATA, __objc_selrefs, literal_pointers, no_dead_strip"
 @g = common global i8* null, align 8
 @"\01L_OBJC_IMAGE_INFO" = internal constant [2 x i32] [i32 0, i32 16], section "__DATA, __objc_imageinfo, regular, no_dead_strip"
 
@@ -37,16 +37,16 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -57,8 +57,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr3 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr3, align 8
+  %mutationsptr3 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr3, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -67,16 +67,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp5 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp5 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call6 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp5, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call6, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -102,16 +102,16 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call3 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call3, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -122,8 +122,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr4 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr4, align 8
+  %mutationsptr4 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr4, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -132,16 +132,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call7, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -162,21 +162,21 @@ define void @test4() nounwind {
 entry:
   %state.ptr = alloca %struct.__objcFastEnumerationState, align 8
   %items.ptr = alloca [16 x i8*], align 8
-  %tmp = load i8** @g, align 8
+  %tmp = load i8*, i8** @g, align 8
   %0 = call i8* @objc_retain(i8* %tmp) nounwind
   %tmp2 = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp2, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp4 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp4 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp4, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -187,8 +187,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr5 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr5, align 8
+  %mutationsptr5 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr5, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -197,16 +197,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp7 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp7 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call8 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp7, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call8, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -229,20 +229,19 @@ entry:
   %items.ptr = alloca [16 x i8*], align 8
   %call = call i8* @returner()
   %0 = call i8* @objc_retainAutoreleasedReturnValue(i8* %call) nounwind
-  call void @callee()
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call3 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call3, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -253,8 +252,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr4 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr4, align 8
+  %mutationsptr4 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr4, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -263,16 +262,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call7, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -283,13 +282,12 @@ forcoll.empty:
   ret void
 }
 
-; TODO: Delete a nested retain+release pair.
-; The optimizer currently can't do this, because isn't isn't sophisticated enough in
-; reasnoning about nesting.
-
+; We handle this now due to the fact that a release just needs a post dominating
+; use.
+;
 ; CHECK-LABEL: define void @test6(
 ; CHECK: call i8* @objc_retain
-; CHECK: @objc_retain
+; CHECK-NOT: @objc_retain
 ; CHECK: }
 define void @test6() nounwind {
 entry:
@@ -300,16 +298,16 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call3 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call3, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -320,8 +318,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr4 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr4, align 8
+  %mutationsptr4 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr4, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -330,16 +328,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call7, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -369,16 +367,16 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call3 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call3, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -389,8 +387,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.notmutated ]
-  %mutationsptr4 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr4, align 8
+  %mutationsptr4 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr4, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -399,16 +397,16 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   call void @use(i8* %3)
   %4 = add i64 %forcoll.index, 1
   %exitcond = icmp eq i64 %4, %umax
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call7, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -435,16 +433,16 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %1 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp2 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp2 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call3 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp2, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call3, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
-  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
+  %stateitems.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 1
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -455,8 +453,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ 0, %forcoll.loopbody.outer ], [ %4, %forcoll.next ]
-  %mutationsptr4 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr4, align 8
+  %mutationsptr4 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr4, align 8
   %2 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %2, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -465,9 +463,9 @@ forcoll.mutated:
   br label %forcoll.notmutated
 
 forcoll.notmutated:
-  %stateitems = load i8*** %stateitems.ptr, align 8
-  %currentitem.ptr = getelementptr i8** %stateitems, i64 %forcoll.index
-  %3 = load i8** %currentitem.ptr, align 8
+  %stateitems = load i8**, i8*** %stateitems.ptr, align 8
+  %currentitem.ptr = getelementptr i8*, i8** %stateitems, i64 %forcoll.index
+  %3 = load i8*, i8** %currentitem.ptr, align 8
   %tobool = icmp eq i8* %3, null
   br i1 %tobool, label %forcoll.next, label %if.then
 
@@ -481,7 +479,7 @@ forcoll.next:
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %1, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %5 = icmp eq i64 %call7, 0
   br i1 %5, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -512,15 +510,15 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %2 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp3 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp3 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call4 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp3, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call4, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -531,8 +529,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ %phitmp, %forcoll.notmutated.forcoll.loopbody_crit_edge ], [ 1, %forcoll.loopbody.outer ]
-  %mutationsptr5 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr5, align 8
+  %mutationsptr5 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr5, align 8
   %3 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %3, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -549,7 +547,7 @@ forcoll.notmutated.forcoll.loopbody_crit_edge:
   br label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %4 = icmp eq i64 %call7, 0
   br i1 %4, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -579,15 +577,15 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %2 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp3 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp3 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call4 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp3, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call4, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -598,8 +596,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ %phitmp, %forcoll.notmutated ], [ 0, %forcoll.loopbody.outer ]
-  %mutationsptr5 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr5, align 8
+  %mutationsptr5 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr5, align 8
   %3 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %3, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -613,7 +611,7 @@ forcoll.notmutated:
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %4 = icmp eq i64 %call7, 0
   br i1 %4, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -646,15 +644,15 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %2 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp3 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp3 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call4 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp3, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call4, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -665,8 +663,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ %phitmp, %forcoll.notmutated.forcoll.loopbody_crit_edge ], [ 1, %forcoll.loopbody.outer ]
-  %mutationsptr5 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr5, align 8
+  %mutationsptr5 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr5, align 8
   %3 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %3, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -683,7 +681,7 @@ forcoll.notmutated.forcoll.loopbody_crit_edge:
   br label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %4 = icmp eq i64 %call7, 0
   br i1 %4, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -714,15 +712,15 @@ entry:
   %tmp = bitcast %struct.__objcFastEnumerationState* %state.ptr to i8*
   call void @llvm.memset.p0i8.i64(i8* %tmp, i8 0, i64 64, i32 8, i1 false)
   %2 = call i8* @objc_retain(i8* %0) nounwind
-  %tmp3 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp3 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call4 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp3, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %iszero = icmp eq i64 %call4, 0
   br i1 %iszero, label %forcoll.empty, label %forcoll.loopinit
 
 forcoll.loopinit:
-  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
-  %mutationsptr = load i64** %mutationsptr.ptr, align 8
-  %forcoll.initial-mutations = load i64* %mutationsptr, align 8
+  %mutationsptr.ptr = getelementptr inbounds %struct.__objcFastEnumerationState, %struct.__objcFastEnumerationState* %state.ptr, i64 0, i32 2
+  %mutationsptr = load i64*, i64** %mutationsptr.ptr, align 8
+  %forcoll.initial-mutations = load i64, i64* %mutationsptr, align 8
   br label %forcoll.loopbody.outer
 
 forcoll.loopbody.outer:
@@ -733,8 +731,8 @@ forcoll.loopbody.outer:
 
 forcoll.loopbody:
   %forcoll.index = phi i64 [ %phitmp, %forcoll.notmutated ], [ 0, %forcoll.loopbody.outer ]
-  %mutationsptr5 = load i64** %mutationsptr.ptr, align 8
-  %statemutations = load i64* %mutationsptr5, align 8
+  %mutationsptr5 = load i64*, i64** %mutationsptr.ptr, align 8
+  %statemutations = load i64, i64* %mutationsptr5, align 8
   %3 = icmp eq i64 %statemutations, %forcoll.initial-mutations
   br i1 %3, label %forcoll.notmutated, label %forcoll.mutated
 
@@ -748,7 +746,7 @@ forcoll.notmutated:
   br i1 %exitcond, label %forcoll.refetch, label %forcoll.loopbody
 
 forcoll.refetch:
-  %tmp6 = load i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
+  %tmp6 = load i8*, i8** @"\01L_OBJC_SELECTOR_REFERENCES_", align 8
   %call7 = call i64 bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i64 (i8*, i8*, %struct.__objcFastEnumerationState*, [16 x i8*]*, i64)*)(i8* %2, i8* %tmp6, %struct.__objcFastEnumerationState* %state.ptr, [16 x i8*]* %items.ptr, i64 16)
   %4 = icmp eq i64 %call7, 0
   br i1 %4, label %forcoll.empty, label %forcoll.loopbody.outer
@@ -779,16 +777,16 @@ entry:
   %block = alloca <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, align 8
   %block9 = alloca <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, align 8
   %call = call i8* @def(), !clang.arc.no_objc_arc_exceptions !0
-  %foo = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 5
-  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 0
+  %foo = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 5
+  %block.isa = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 0
   store i8* bitcast (i8** @_NSConcreteStackBlock to i8*), i8** %block.isa, align 8
-  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 1
+  %block.flags = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 1
   store i32 1107296256, i32* %block.flags, align 8
-  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 2
+  %block.reserved = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 2
   store i32 0, i32* %block.reserved, align 4
-  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 3
+  %block.invoke = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 3
   store i8* bitcast (void (i8*)* @__crasher_block_invoke to i8*), i8** %block.invoke, align 8
-  %block.d = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 4
+  %block.d = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block, i64 0, i32 4
   store %struct.__block_d* bitcast ({ i64, i64, i8*, i8*, i8*, i8* }* @__block_d_tmp to %struct.__block_d*), %struct.__block_d** %block.d, align 8
   %foo2 = tail call i8* @objc_retain(i8* %call) nounwind
   store i8* %foo2, i8** %foo, align 8
@@ -796,18 +794,18 @@ entry:
   %foo5 = call i8* @objc_retainBlock(i8* %foo4) nounwind
   call void @use(i8* %foo5), !clang.arc.no_objc_arc_exceptions !0
   call void @objc_release(i8* %foo5) nounwind
-  %strongdestroy = load i8** %foo, align 8
+  %strongdestroy = load i8*, i8** %foo, align 8
   call void @objc_release(i8* %strongdestroy) nounwind, !clang.imprecise_release !0
-  %foo10 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 5
-  %block.isa11 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 0
+  %foo10 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 5
+  %block.isa11 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 0
   store i8* bitcast (i8** @_NSConcreteStackBlock to i8*), i8** %block.isa11, align 8
-  %block.flags12 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 1
+  %block.flags12 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 1
   store i32 1107296256, i32* %block.flags12, align 8
-  %block.reserved13 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 2
+  %block.reserved13 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 2
   store i32 0, i32* %block.reserved13, align 4
-  %block.invoke14 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 3
+  %block.invoke14 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 3
   store i8* bitcast (void (i8*)* @__crasher_block_invoke1 to i8*), i8** %block.invoke14, align 8
-  %block.d15 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 4
+  %block.d15 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_d*, i8* }>* %block9, i64 0, i32 4
   store %struct.__block_d* bitcast ({ i64, i64, i8*, i8*, i8*, i8* }* @__block_d_tmp5 to %struct.__block_d*), %struct.__block_d** %block.d15, align 8
   %foo18 = call i8* @objc_retain(i8* %call) nounwind
   store i8* %call, i8** %foo10, align 8
@@ -815,7 +813,7 @@ entry:
   %foo21 = call i8* @objc_retainBlock(i8* %foo20) nounwind
   call void @use(i8* %foo21), !clang.arc.no_objc_arc_exceptions !0
   call void @objc_release(i8* %foo21) nounwind
-  %strongdestroy25 = load i8** %foo10, align 8
+  %strongdestroy25 = load i8*, i8** %foo10, align 8
   call void @objc_release(i8* %strongdestroy25) nounwind, !clang.imprecise_release !0
   call void @objc_release(i8* %call) nounwind, !clang.imprecise_release !0
   ret void
diff --git a/test/Transforms/ObjCARC/path-overflow.ll b/test/Transforms/ObjCARC/path-overflow.ll
index d23965310127..31f0e3b034b5 100644
--- a/test/Transforms/ObjCARC/path-overflow.ll
+++ b/test/Transforms/ObjCARC/path-overflow.ll
@@ -881,7 +881,7 @@ land.rhs:                                         ; preds = %invoke.cont4
           to label %land.end unwind label %lpad3
 
 land.end:                                         ; preds = %land.rhs, %invoke.cont4
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i unwind label %lpad.i
 
 invoke.cont.i:                                    ; preds = %land.end
@@ -896,7 +896,7 @@ lpad.i:                                           ; preds = %land.end
   unreachable
 
 invoke.cont8:                                     ; preds = %if.then.i, %invoke.cont.i
-  %call18 = invoke i8* (i8*, i8*, i8*, ...)* bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*, i8*, ...)*)(i8* undef, i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef)
+  %call18 = invoke i8* (i8*, i8*, i8*, ...) bitcast (i8* (i8*, i8*, ...)* @objc_msgSend to i8* (i8*, i8*, i8*, ...)*)(i8* undef, i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef)
           to label %invoke.cont17 unwind label %lpad16
 
 invoke.cont17:                                    ; preds = %invoke.cont8
@@ -904,7 +904,7 @@ invoke.cont17:                                    ; preds = %invoke.cont8
           to label %invoke.cont21 unwind label %lpad20
 
 invoke.cont21:                                    ; preds = %invoke.cont17
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i1980 unwind label %lpad.i1982
 
 invoke.cont.i1980:                                ; preds = %invoke.cont21
@@ -930,7 +930,7 @@ land.rhs39:                                       ; preds = %invoke.cont36
           to label %land.end43 unwind label %lpad35
 
 land.end43:                                       ; preds = %land.rhs39, %invoke.cont36
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i1986 unwind label %lpad.i1988
 
 invoke.cont.i1986:                                ; preds = %land.end43
@@ -960,7 +960,7 @@ invoke.cont62:                                    ; preds = %land.rhs58
           to label %land.end70 unwind label %lpad66.body.thread
 
 land.end70:                                       ; preds = %invoke.cont62, %invoke.cont52
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i1992 unwind label %lpad66.body
 
 invoke.cont.i1992:                                ; preds = %land.end70
@@ -970,7 +970,7 @@ if.then.i1993:                                    ; preds = %invoke.cont.i1992
   br label %invoke.cont71
 
 invoke.cont71:                                    ; preds = %if.then.i1993, %invoke.cont.i1992
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i1998 unwind label %lpad.i2000
 
 invoke.cont.i1998:                                ; preds = %invoke.cont71
@@ -993,7 +993,7 @@ invoke.cont95:                                    ; preds = %invoke.cont91
           to label %invoke.cont97 unwind label %lpad94
 
 invoke.cont97:                                    ; preds = %invoke.cont95
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2004 unwind label %lpad.i2006
 
 invoke.cont.i2004:                                ; preds = %invoke.cont97
@@ -1012,7 +1012,7 @@ invoke.cont100:                                   ; preds = %if.then.i2005, %inv
           to label %invoke.cont110 unwind label %lpad109
 
 invoke.cont110:                                   ; preds = %invoke.cont100
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2010 unwind label %lpad.i2012
 
 invoke.cont.i2010:                                ; preds = %invoke.cont110
@@ -1027,7 +1027,7 @@ lpad.i2012:                                       ; preds = %invoke.cont110
   unreachable
 
 invoke.cont117:                                   ; preds = %if.then.i2011, %invoke.cont.i2010
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2022 unwind label %lpad156.body
 
 lpad:                                             ; preds = %entry
@@ -1101,7 +1101,7 @@ if.then.i2023:                                    ; preds = %invoke.cont.i2022
   br label %invoke.cont157
 
 invoke.cont157:                                   ; preds = %if.then.i2023, %invoke.cont.i2022
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2028 unwind label %lpad164.body
 
 invoke.cont.i2028:                                ; preds = %invoke.cont157
@@ -1119,7 +1119,7 @@ invoke.cont184:                                   ; preds = %invoke.cont165
           to label %invoke.cont185 unwind label %lpad183
 
 invoke.cont185:                                   ; preds = %invoke.cont184
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2034 unwind label %lpad.i2036
 
 invoke.cont.i2034:                                ; preds = %invoke.cont185
@@ -1146,7 +1146,7 @@ invoke.cont201:                                   ; preds = %invoke.cont197
           to label %invoke.cont204 unwind label %lpad203
 
 invoke.cont204:                                   ; preds = %invoke.cont201
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2040 unwind label %lpad.i2042
 
 invoke.cont.i2040:                                ; preds = %invoke.cont204
@@ -1165,7 +1165,7 @@ invoke.cont207:                                   ; preds = %if.then.i2041, %inv
           to label %invoke.cont208 unwind label %lpad203
 
 invoke.cont208:                                   ; preds = %invoke.cont207
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2046 unwind label %lpad212.body
 
 invoke.cont.i2046:                                ; preds = %invoke.cont208
@@ -1183,7 +1183,7 @@ invoke.cont221:                                   ; preds = %invoke.cont213
           to label %invoke.cont228 unwind label %lpad227
 
 invoke.cont228:                                   ; preds = %invoke.cont221
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2052 unwind label %lpad.i2054
 
 invoke.cont.i2052:                                ; preds = %invoke.cont228
@@ -1202,7 +1202,7 @@ invoke.cont231:                                   ; preds = %if.then.i2053, %inv
           to label %invoke.cont232 unwind label %lpad227
 
 invoke.cont232:                                   ; preds = %invoke.cont231
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2058 unwind label %lpad236.body
 
 invoke.cont.i2058:                                ; preds = %invoke.cont232
@@ -1248,7 +1248,7 @@ invoke.cont274:                                   ; preds = %invoke.cont266
           to label %invoke.cont278 unwind label %lpad277
 
 invoke.cont278:                                   ; preds = %invoke.cont274
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2064 unwind label %lpad.i2066
 
 invoke.cont.i2064:                                ; preds = %invoke.cont278
@@ -1294,7 +1294,7 @@ land.rhs335:                                      ; preds = %invoke.cont321
           to label %land.end344 unwind label %lpad340.body.thread
 
 land.end344:                                      ; preds = %land.rhs335, %invoke.cont321
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2070 unwind label %lpad340.body
 
 invoke.cont.i2070:                                ; preds = %land.end344
@@ -1316,7 +1316,7 @@ invoke.cont364:                                   ; preds = %invoke.cont361
           to label %invoke.cont370 unwind label %lpad369
 
 invoke.cont370:                                   ; preds = %invoke.cont364
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2076 unwind label %lpad.i2078
 
 invoke.cont.i2076:                                ; preds = %invoke.cont370
@@ -1343,7 +1343,7 @@ invoke.cont382:                                   ; preds = %invoke.cont376
           to label %invoke.cont383 unwind label %lpad381
 
 invoke.cont383:                                   ; preds = %invoke.cont382
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2082 unwind label %lpad.i2084
 
 invoke.cont.i2082:                                ; preds = %invoke.cont383
@@ -1374,7 +1374,7 @@ invoke.cont399:                                   ; preds = %invoke.cont396
           to label %invoke.cont402 unwind label %lpad401
 
 invoke.cont402:                                   ; preds = %invoke.cont399
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2088 unwind label %lpad.i2090
 
 invoke.cont.i2088:                                ; preds = %invoke.cont402
@@ -1401,7 +1401,7 @@ invoke.cont409:                                   ; preds = %invoke.cont408
           to label %invoke.cont412 unwind label %lpad411
 
 invoke.cont412:                                   ; preds = %invoke.cont409
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2094 unwind label %lpad.i2096
 
 invoke.cont.i2094:                                ; preds = %invoke.cont412
@@ -1432,7 +1432,7 @@ invoke.cont426:                                   ; preds = %invoke.cont423
           to label %invoke.cont429 unwind label %lpad428
 
 invoke.cont429:                                   ; preds = %invoke.cont426
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2100 unwind label %lpad.i2102
 
 invoke.cont.i2100:                                ; preds = %invoke.cont429
@@ -1451,7 +1451,7 @@ invoke.cont432:                                   ; preds = %if.then.i2101, %inv
           to label %invoke.cont435 unwind label %lpad381
 
 invoke.cont435:                                   ; preds = %invoke.cont432
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2106 unwind label %lpad.i2108
 
 invoke.cont.i2106:                                ; preds = %invoke.cont435
@@ -1464,7 +1464,7 @@ lpad.i2108:                                       ; preds = %invoke.cont435
   unreachable
 
 invoke.cont443:                                   ; preds = %invoke.cont.i2106
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2112 unwind label %lpad.i2114
 
 invoke.cont.i2112:                                ; preds = %invoke.cont443
@@ -1487,7 +1487,7 @@ invoke.cont452:                                   ; preds = %invoke.cont449
           to label %invoke.cont455 unwind label %lpad454
 
 invoke.cont455:                                   ; preds = %invoke.cont452
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2118 unwind label %lpad.i2120
 
 invoke.cont.i2118:                                ; preds = %invoke.cont455
@@ -1506,7 +1506,7 @@ invoke.cont458:                                   ; preds = %if.then.i2119, %inv
           to label %invoke.cont460 unwind label %lpad381
 
 invoke.cont460:                                   ; preds = %invoke.cont458
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2124 unwind label %lpad.i2126
 
 invoke.cont.i2124:                                ; preds = %invoke.cont460
@@ -1525,7 +1525,7 @@ invoke.cont466:                                   ; preds = %if.then.i2125, %inv
           to label %invoke.cont469 unwind label %lpad381
 
 invoke.cont469:                                   ; preds = %invoke.cont466
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2130 unwind label %lpad.i2132
 
 invoke.cont.i2130:                                ; preds = %invoke.cont469
@@ -1574,7 +1574,7 @@ msgSend.null-receiver:                            ; preds = %invoke.cont512
   br label %msgSend.cont
 
 msgSend.cont:                                     ; preds = %msgSend.null-receiver, %msgSend.call
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2136 unwind label %lpad.i2138
 
 invoke.cont.i2136:                                ; preds = %msgSend.cont
@@ -1601,7 +1601,7 @@ invoke.cont531:                                   ; preds = %invoke.cont528
           to label %invoke.cont534 unwind label %lpad533
 
 invoke.cont534:                                   ; preds = %invoke.cont531
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* undef, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2142 unwind label %lpad.i2144
 
 invoke.cont.i2142:                                ; preds = %invoke.cont534
@@ -1633,7 +1633,7 @@ invoke.cont548:                                   ; preds = %invoke.cont546
 
 invoke.cont554:                                   ; preds = %invoke.cont548
   %tmp499 = call i8* @objc_retain(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*)) #3
-  invoke void (i8*, ...)* @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* %tmp499, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
+  invoke void (i8*, ...) @NSLog(i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*), i8* %tmp499, i8* bitcast (%struct.NSConstantString* @_unnamed_cfstring to i8*))
           to label %invoke.cont.i2148 unwind label %lpad.i2150
 
 invoke.cont.i2148:                                ; preds = %invoke.cont554
diff --git a/test/Transforms/ObjCARC/provenance.ll b/test/Transforms/ObjCARC/provenance.ll
index 937c68924357..aa5a932a86e9 100644
--- a/test/Transforms/ObjCARC/provenance.ll
+++ b/test/Transforms/ObjCARC/provenance.ll
@@ -10,28 +10,28 @@
 declare void @g(i8)
 
 define void @f(i8* %a, i8** %b, i8** %c) {
-  %y1 = load i8* %a
+  %y1 = load i8, i8* %a
   call void @g(i8 %y1)
 
-  %y2 = load i8** %b
-  %y3 = load i8** %c
+  %y2 = load i8*, i8** %b
+  %y3 = load i8*, i8** %c
 
-  %x0 = load i8* @"\01l_objc_msgSend_fixup_"
+  %x0 = load i8, i8* @"\01l_objc_msgSend_fixup_"
   call void @g(i8 %x0)
 
-  %x1 = load i8* @g1
+  %x1 = load i8, i8* @g1
   call void @g(i8 %x1)
 
-  %x2 = load i8* @g2
+  %x2 = load i8, i8* @g2
   call void @g(i8 %x2)
 
-  %x3 = load i8* @g3
+  %x3 = load i8, i8* @g3
   call void @g(i8 %x3)
 
-  %x4 = load i8* @g4
+  %x4 = load i8, i8* @g4
   call void @g(i8 %x4)
 
-  %x5 = load i8* @g5
+  %x5 = load i8, i8* @g5
   call void @g(i8 %x5)
   ret void
 }
diff --git a/test/Transforms/ObjCARC/retain-block-side-effects.ll b/test/Transforms/ObjCARC/retain-block-side-effects.ll
index 7fa73cbfef15..5f5def9ff036 100644
--- a/test/Transforms/ObjCARC/retain-block-side-effects.ll
+++ b/test/Transforms/ObjCARC/retain-block-side-effects.ll
@@ -6,8 +6,8 @@
 
 ; CHECK: %tmp16 = call i8* @objc_retainBlock(i8* %tmp15) [[NUW:#[0-9]+]]
 ; CHECK: %tmp17 = bitcast i8* %tmp16 to void ()*
-; CHECK: %tmp18 = load %struct.__block_byref_repeater** %byref.forwarding, align 8
-; CHECK: %repeater12 = getelementptr inbounds %struct.__block_byref_repeater* %tmp18, i64 0, i32 6
+; CHECK: %tmp18 = load %struct.__block_byref_repeater*, %struct.__block_byref_repeater** %byref.forwarding, align 8
+; CHECK: %repeater12 = getelementptr inbounds %struct.__block_byref_repeater, %struct.__block_byref_repeater* %tmp18, i64 0, i32 6
 ; CHECK: store void ()* %tmp17, void ()** %repeater12, align 8
 
 target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64"
@@ -20,18 +20,18 @@ define void @foo() noreturn {
 entry:
   %repeater = alloca %struct.__block_byref_repeater, align 8
   %block = alloca <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0*, i8* }>, align 8
-  %byref.forwarding = getelementptr inbounds %struct.__block_byref_repeater* %repeater, i64 0, i32 1
-  %tmp10 = getelementptr inbounds %struct.__block_byref_repeater* %repeater, i64 0, i32 6
+  %byref.forwarding = getelementptr inbounds %struct.__block_byref_repeater, %struct.__block_byref_repeater* %repeater, i64 0, i32 1
+  %tmp10 = getelementptr inbounds %struct.__block_byref_repeater, %struct.__block_byref_repeater* %repeater, i64 0, i32 6
   store void ()* null, void ()** %tmp10, align 8
-  %block.captured11 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0*, i8* }>* %block, i64 0, i32 6
+  %block.captured11 = getelementptr inbounds <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0*, i8* }>, <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0*, i8* }>* %block, i64 0, i32 6
   %tmp14 = bitcast %struct.__block_byref_repeater* %repeater to i8*
   store i8* %tmp14, i8** %block.captured11, align 8
   %tmp15 = bitcast <{ i8*, i32, i32, i8*, %struct.__block_descriptor*, %0*, i8* }>* %block to i8*
   %tmp16 = call i8* @objc_retainBlock(i8* %tmp15) nounwind
   %tmp17 = bitcast i8* %tmp16 to void ()*
-  %tmp18 = load %struct.__block_byref_repeater** %byref.forwarding, align 8
-  %repeater12 = getelementptr inbounds %struct.__block_byref_repeater* %tmp18, i64 0, i32 6
-  %tmp13 = load void ()** %repeater12, align 8
+  %tmp18 = load %struct.__block_byref_repeater*, %struct.__block_byref_repeater** %byref.forwarding, align 8
+  %repeater12 = getelementptr inbounds %struct.__block_byref_repeater, %struct.__block_byref_repeater* %tmp18, i64 0, i32 6
+  %tmp13 = load void ()*, void ()** %repeater12, align 8
   store void ()* %tmp17, void ()** %repeater12, align 8
   ret void
 }
diff --git a/test/Transforms/ObjCARC/weak-copies.ll b/test/Transforms/ObjCARC/weak-copies.ll
index 13d0b0a0c6b7..d3177bb727fc 100644
--- a/test/Transforms/ObjCARC/weak-copies.ll
+++ b/test/Transforms/ObjCARC/weak-copies.ll
@@ -10,9 +10,9 @@ target triple = "x86_64-apple-darwin11.0.0"
 @_NSConcreteStackBlock = external global i8*
 @.str = private unnamed_addr constant [6 x i8] c"v8@?0\00"
 @"\01L_OBJC_CLASS_NAME_" = internal global [3 x i8] c"\01@\00", section "__TEXT,__objc_classname,cstring_literals", align 1
-@__block_descriptor_tmp = internal constant %0 { i64 0, i64 40, i8* bitcast (void (i8*, i8*)* @__copy_helper_block_ to i8*), i8* bitcast (void (i8*)* @__destroy_helper_block_ to i8*), i8* getelementptr inbounds ([6 x i8]* @.str, i32 0, i32 0), i8* getelementptr inbounds ([3 x i8]* @"\01L_OBJC_CLASS_NAME_", i32 0, i32 0) }
+@__block_descriptor_tmp = internal constant %0 { i64 0, i64 40, i8* bitcast (void (i8*, i8*)* @__copy_helper_block_ to i8*), i8* bitcast (void (i8*)* @__destroy_helper_block_ to i8*), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str, i32 0, i32 0), i8* getelementptr inbounds ([3 x i8], [3 x i8]* @"\01L_OBJC_CLASS_NAME_", i32 0, i32 0) }
 @"\01L_OBJC_IMAGE_INFO" = internal constant [2 x i32] [i32 0, i32 16], section "__DATA, __objc_imageinfo, regular, no_dead_strip"
-@llvm.used = appending global [2 x i8*] [i8* getelementptr inbounds ([3 x i8]* @"\01L_OBJC_CLASS_NAME_", i32 0, i32 0), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
+@llvm.used = appending global [2 x i8*] [i8* getelementptr inbounds ([3 x i8], [3 x i8]* @"\01L_OBJC_CLASS_NAME_", i32 0, i32 0), i8* bitcast ([2 x i32]* @"\01L_OBJC_IMAGE_INFO" to i8*)], section "llvm.metadata"
 
 ; Eliminate unnecessary weak pointer copies.
 
@@ -50,17 +50,17 @@ entry:
   %block = alloca %1, align 8
   %0 = call i8* @objc_retain(i8* %me) nounwind
   %1 = call i8* @objc_initWeak(i8** %w, i8* %0) nounwind
-  %block.isa = getelementptr inbounds %1* %block, i64 0, i32 0
+  %block.isa = getelementptr inbounds %1, %1* %block, i64 0, i32 0
   store i8* bitcast (i8** @_NSConcreteStackBlock to i8*), i8** %block.isa, align 8
-  %block.flags = getelementptr inbounds %1* %block, i64 0, i32 1
+  %block.flags = getelementptr inbounds %1, %1* %block, i64 0, i32 1
   store i32 1107296256, i32* %block.flags, align 8
-  %block.reserved = getelementptr inbounds %1* %block, i64 0, i32 2
+  %block.reserved = getelementptr inbounds %1, %1* %block, i64 0, i32 2
   store i32 0, i32* %block.reserved, align 4
-  %block.invoke = getelementptr inbounds %1* %block, i64 0, i32 3
+  %block.invoke = getelementptr inbounds %1, %1* %block, i64 0, i32 3
   store i8* bitcast (void (i8*)* @__qux_block_invoke_0 to i8*), i8** %block.invoke, align 8
-  %block.descriptor = getelementptr inbounds %1* %block, i64 0, i32 4
+  %block.descriptor = getelementptr inbounds %1, %1* %block, i64 0, i32 4
   store %struct.__block_descriptor* bitcast (%0* @__block_descriptor_tmp to %struct.__block_descriptor*), %struct.__block_descriptor** %block.descriptor, align 8
-  %block.captured = getelementptr inbounds %1* %block, i64 0, i32 5
+  %block.captured = getelementptr inbounds %1, %1* %block, i64 0, i32 5
   %2 = call i8* @objc_loadWeak(i8** %w) nounwind
   %3 = call i8* @objc_initWeak(i8** %block.captured, i8* %2) nounwind
   %4 = bitcast %1* %block to void ()*
diff --git a/test/Transforms/PhaseOrdering/2010-03-22-empty-baseclass.ll b/test/Transforms/PhaseOrdering/2010-03-22-empty-baseclass.ll
index 53d98e02ec88..13404a8b6a71 100644
--- a/test/Transforms/PhaseOrdering/2010-03-22-empty-baseclass.ll
+++ b/test/Transforms/PhaseOrdering/2010-03-22-empty-baseclass.ll
@@ -29,29 +29,29 @@ entry:
   store %struct.empty_base_t* %4, %struct.empty_base_t** %2, align 8
   call void @_ZN7empty_tC1Ev(%struct.empty_base_t* %1) nounwind
   %5 = call i32* @_ZN5boost15compressed_pairI7empty_tiE6secondEv(%"struct.boost::compressed_pair<empty_t,int>"* %x) ssp ; <i32*> [#uses=1]
-  %6 = load i32* %5, align 4                      ; <i32> [#uses=1]
+  %6 = load i32, i32* %5, align 4                      ; <i32> [#uses=1]
   %7 = icmp ne i32 %6, -3                         ; <i1> [#uses=1]
   %8 = zext i1 %7 to i8                           ; <i8> [#uses=1]
   store i8 %8, i8* %retval.1, align 1
-  %9 = load i8* %retval.1, align 1                ; <i8> [#uses=1]
+  %9 = load i8, i8* %retval.1, align 1                ; <i8> [#uses=1]
   %toBool = icmp ne i8 %9, 0                      ; <i1> [#uses=1]
   br i1 %toBool, label %bb, label %bb1
 
 bb:                                               ; preds = %entry
-  %10 = call i32 @puts(i8* getelementptr inbounds ([25 x i8]* @.str, i64 0, i64 0)) ; <i32> [#uses=0]
+  %10 = call i32 @puts(i8* getelementptr inbounds ([25 x i8], [25 x i8]* @.str, i64 0, i64 0)) ; <i32> [#uses=0]
   call void @abort() noreturn
   unreachable
 
 bb1:                                              ; preds = %entry
   store i32 0, i32* %0, align 4
-  %11 = load i32* %0, align 4                     ; <i32> [#uses=1]
+  %11 = load i32, i32* %0, align 4                     ; <i32> [#uses=1]
   store i32 %11, i32* %retval, align 4
   br label %return
 
 ; CHECK-NOT: x.second() was clobbered
 ; CHECK: ret i32
 return:                                           ; preds = %bb1
-  %retval2 = load i32* %retval                    ; <i32> [#uses=1]
+  %retval2 = load i32, i32* %retval                    ; <i32> [#uses=1]
   ret i32 %retval2
 }
 
@@ -71,7 +71,7 @@ entry:
   %this_addr = alloca %struct.empty_base_t*, align 8 ; <%struct.empty_base_t**> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store %struct.empty_base_t* %this, %struct.empty_base_t** %this_addr
-  %0 = load %struct.empty_base_t** %this_addr, align 8 ; <%struct.empty_base_t*> [#uses=1]
+  %0 = load %struct.empty_base_t*, %struct.empty_base_t** %this_addr, align 8 ; <%struct.empty_base_t*> [#uses=1]
   call void @_ZN12empty_base_tC2Ev(%struct.empty_base_t* %0) nounwind
   br label %return
 
@@ -86,15 +86,15 @@ entry:
   %0 = alloca i32*                                ; <i32**> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %this, %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr
-  %1 = load %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr, align 8 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
-  %2 = getelementptr inbounds %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %1, i32 0, i32 0 ; <i32*> [#uses=1]
+  %1 = load %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*, %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr, align 8 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
+  %2 = getelementptr inbounds %"struct.boost::details::compressed_pair_imp<empty_t,int,1>", %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %1, i32 0, i32 0 ; <i32*> [#uses=1]
   store i32* %2, i32** %0, align 8
-  %3 = load i32** %0, align 8                     ; <i32*> [#uses=1]
+  %3 = load i32*, i32** %0, align 8                     ; <i32*> [#uses=1]
   store i32* %3, i32** %retval, align 8
   br label %return
 
 return:                                           ; preds = %entry
-  %retval1 = load i32** %retval                   ; <i32*> [#uses=1]
+  %retval1 = load i32*, i32** %retval                   ; <i32*> [#uses=1]
   ret i32* %retval1
 }
 
@@ -105,16 +105,16 @@ entry:
   %0 = alloca i32*                                ; <i32**> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store %"struct.boost::compressed_pair<empty_t,int>"* %this, %"struct.boost::compressed_pair<empty_t,int>"** %this_addr
-  %1 = load %"struct.boost::compressed_pair<empty_t,int>"** %this_addr, align 8 ; <%"struct.boost::compressed_pair<empty_t,int>"*> [#uses=1]
-  %2 = getelementptr inbounds %"struct.boost::compressed_pair<empty_t,int>"* %1, i32 0, i32 0 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
+  %1 = load %"struct.boost::compressed_pair<empty_t,int>"*, %"struct.boost::compressed_pair<empty_t,int>"** %this_addr, align 8 ; <%"struct.boost::compressed_pair<empty_t,int>"*> [#uses=1]
+  %2 = getelementptr inbounds %"struct.boost::compressed_pair<empty_t,int>", %"struct.boost::compressed_pair<empty_t,int>"* %1, i32 0, i32 0 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
   %3 = call i32* @_ZN5boost7details19compressed_pair_impI7empty_tiLi1EE6secondEv(%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %2) nounwind ; <i32*> [#uses=1]
   store i32* %3, i32** %0, align 8
-  %4 = load i32** %0, align 8                     ; <i32*> [#uses=1]
+  %4 = load i32*, i32** %0, align 8                     ; <i32*> [#uses=1]
   store i32* %4, i32** %retval, align 8
   br label %return
 
 return:                                           ; preds = %entry
-  %retval1 = load i32** %retval                   ; <i32*> [#uses=1]
+  %retval1 = load i32*, i32** %retval                   ; <i32*> [#uses=1]
   ret i32* %retval1
 }
 
@@ -125,15 +125,15 @@ entry:
   %0 = alloca %struct.empty_base_t*               ; <%struct.empty_base_t**> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %this, %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr
-  %1 = load %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr, align 8 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
+  %1 = load %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*, %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"** %this_addr, align 8 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
   %2 = bitcast %"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %1 to %struct.empty_base_t* ; <%struct.empty_base_t*> [#uses=1]
   store %struct.empty_base_t* %2, %struct.empty_base_t** %0, align 8
-  %3 = load %struct.empty_base_t** %0, align 8    ; <%struct.empty_base_t*> [#uses=1]
+  %3 = load %struct.empty_base_t*, %struct.empty_base_t** %0, align 8    ; <%struct.empty_base_t*> [#uses=1]
   store %struct.empty_base_t* %3, %struct.empty_base_t** %retval, align 8
   br label %return
 
 return:                                           ; preds = %entry
-  %retval1 = load %struct.empty_base_t** %retval  ; <%struct.empty_base_t*> [#uses=1]
+  %retval1 = load %struct.empty_base_t*, %struct.empty_base_t** %retval  ; <%struct.empty_base_t*> [#uses=1]
   ret %struct.empty_base_t* %retval1
 }
 
@@ -144,16 +144,16 @@ entry:
   %0 = alloca %struct.empty_base_t*               ; <%struct.empty_base_t**> [#uses=2]
   %"alloca point" = bitcast i32 0 to i32          ; <i32> [#uses=0]
   store %"struct.boost::compressed_pair<empty_t,int>"* %this, %"struct.boost::compressed_pair<empty_t,int>"** %this_addr
-  %1 = load %"struct.boost::compressed_pair<empty_t,int>"** %this_addr, align 8 ; <%"struct.boost::compressed_pair<empty_t,int>"*> [#uses=1]
-  %2 = getelementptr inbounds %"struct.boost::compressed_pair<empty_t,int>"* %1, i32 0, i32 0 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
+  %1 = load %"struct.boost::compressed_pair<empty_t,int>"*, %"struct.boost::compressed_pair<empty_t,int>"** %this_addr, align 8 ; <%"struct.boost::compressed_pair<empty_t,int>"*> [#uses=1]
+  %2 = getelementptr inbounds %"struct.boost::compressed_pair<empty_t,int>", %"struct.boost::compressed_pair<empty_t,int>"* %1, i32 0, i32 0 ; <%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"*> [#uses=1]
   %3 = call %struct.empty_base_t* @_ZN5boost7details19compressed_pair_impI7empty_tiLi1EE5firstEv(%"struct.boost::details::compressed_pair_imp<empty_t,int,1>"* %2) nounwind ; <%struct.empty_base_t*> [#uses=1]
   store %struct.empty_base_t* %3, %struct.empty_base_t** %0, align 8
-  %4 = load %struct.empty_base_t** %0, align 8    ; <%struct.empty_base_t*> [#uses=1]
+  %4 = load %struct.empty_base_t*, %struct.empty_base_t** %0, align 8    ; <%struct.empty_base_t*> [#uses=1]
   store %struct.empty_base_t* %4, %struct.empty_base_t** %retval, align 8
   br label %return
 
 return:                                           ; preds = %entry
-  %retval1 = load %struct.empty_base_t** %retval  ; <%struct.empty_base_t*> [#uses=1]
+  %retval1 = load %struct.empty_base_t*, %struct.empty_base_t** %retval  ; <%struct.empty_base_t*> [#uses=1]
   ret %struct.empty_base_t* %retval1
 }
 
diff --git a/test/Transforms/PhaseOrdering/PR6627.ll b/test/Transforms/PhaseOrdering/PR6627.ll
index cf95363e655a..2774d208352c 100644
--- a/test/Transforms/PhaseOrdering/PR6627.ll
+++ b/test/Transforms/PhaseOrdering/PR6627.ll
@@ -8,42 +8,42 @@ declare i32 @doo(...)
 define void @test2(i8* %arrayidx) nounwind ssp {
 entry:
   %xx = bitcast i8* %arrayidx to i32*
-  %x1 = load i32* %xx, align 4
+  %x1 = load i32, i32* %xx, align 4
   %tmp = trunc i32 %x1 to i8
   %conv = zext i8 %tmp to i32
   %cmp = icmp eq i32 %conv, 127
   br i1 %cmp, label %land.lhs.true, label %if.end
 
 land.lhs.true:                                    ; preds = %entry
-  %arrayidx4 = getelementptr inbounds i8* %arrayidx, i64 1
-  %tmp5 = load i8* %arrayidx4, align 1
+  %arrayidx4 = getelementptr inbounds i8, i8* %arrayidx, i64 1
+  %tmp5 = load i8, i8* %arrayidx4, align 1
   %conv6 = zext i8 %tmp5 to i32
   %cmp7 = icmp eq i32 %conv6, 69
   br i1 %cmp7, label %land.lhs.true9, label %if.end
 
 land.lhs.true9:                                   ; preds = %land.lhs.true
-  %arrayidx12 = getelementptr inbounds i8* %arrayidx, i64 2
-  %tmp13 = load i8* %arrayidx12, align 1
+  %arrayidx12 = getelementptr inbounds i8, i8* %arrayidx, i64 2
+  %tmp13 = load i8, i8* %arrayidx12, align 1
   %conv14 = zext i8 %tmp13 to i32
   %cmp15 = icmp eq i32 %conv14, 76
   br i1 %cmp15, label %land.lhs.true17, label %if.end
 
 land.lhs.true17:                                  ; preds = %land.lhs.true9
-  %arrayidx20 = getelementptr inbounds i8* %arrayidx, i64 3
-  %tmp21 = load i8* %arrayidx20, align 1
+  %arrayidx20 = getelementptr inbounds i8, i8* %arrayidx, i64 3
+  %tmp21 = load i8, i8* %arrayidx20, align 1
   %conv22 = zext i8 %tmp21 to i32
   %cmp23 = icmp eq i32 %conv22, 70
   br i1 %cmp23, label %if.then, label %if.end
 
 if.then:                                          ; preds = %land.lhs.true17
-  %call25 = call i32 (...)* @doo()
+  %call25 = call i32 (...) @doo()
   br label %if.end
 
 if.end:
   ret void
 
 ; CHECK-LABEL: @test2(
-; CHECK: %x1 = load i32* %xx, align 4
+; CHECK: %x1 = load i32, i32* %xx, align 4
 ; CHECK-NEXT: icmp eq i32 %x1, 1179403647
 ; CHECK-NEXT: br i1 {{.*}}, label %if.then, label %if.end 
 }
@@ -53,41 +53,41 @@ if.end:
 ; requiring widening.
 define void @test2a(i8* %arrayidx) nounwind ssp {
 entry:
-  %x1 = load i8* %arrayidx, align 4
+  %x1 = load i8, i8* %arrayidx, align 4
   %conv = zext i8 %x1 to i32
   %cmp = icmp eq i32 %conv, 127
   br i1 %cmp, label %land.lhs.true, label %if.end
 
 land.lhs.true:                                    ; preds = %entry
-  %arrayidx4 = getelementptr inbounds i8* %arrayidx, i64 1
-  %tmp5 = load i8* %arrayidx4, align 1
+  %arrayidx4 = getelementptr inbounds i8, i8* %arrayidx, i64 1
+  %tmp5 = load i8, i8* %arrayidx4, align 1
   %conv6 = zext i8 %tmp5 to i32
   %cmp7 = icmp eq i32 %conv6, 69
   br i1 %cmp7, label %land.lhs.true9, label %if.end
 
 land.lhs.true9:                                   ; preds = %land.lhs.true
-  %arrayidx12 = getelementptr inbounds i8* %arrayidx, i64 2
-  %tmp13 = load i8* %arrayidx12, align 1
+  %arrayidx12 = getelementptr inbounds i8, i8* %arrayidx, i64 2
+  %tmp13 = load i8, i8* %arrayidx12, align 1
   %conv14 = zext i8 %tmp13 to i32
   %cmp15 = icmp eq i32 %conv14, 76
   br i1 %cmp15, label %land.lhs.true17, label %if.end
 
 land.lhs.true17:                                  ; preds = %land.lhs.true9
-  %arrayidx20 = getelementptr inbounds i8* %arrayidx, i64 3
-  %tmp21 = load i8* %arrayidx20, align 1
+  %arrayidx20 = getelementptr inbounds i8, i8* %arrayidx, i64 3
+  %tmp21 = load i8, i8* %arrayidx20, align 1
   %conv22 = zext i8 %tmp21 to i32
   %cmp23 = icmp eq i32 %conv22, 70
   br i1 %cmp23, label %if.then, label %if.end
 
 if.then:                                          ; preds = %land.lhs.true17
-  %call25 = call i32 (...)* @doo()
+  %call25 = call i32 (...) @doo()
   br label %if.end
 
 if.end:
   ret void
 
 ; CHECK-LABEL: @test2a(
-; CHECK: %x1 = load i32* {{.*}}, align 4
+; CHECK: %x1 = load i32, i32* {{.*}}, align 4
 ; CHECK-NEXT: icmp eq i32 %x1, 1179403647
 ; CHECK-NEXT: br i1 {{.*}}, label %if.then, label %if.end 
 }
diff --git a/test/Transforms/PhaseOrdering/basic.ll b/test/Transforms/PhaseOrdering/basic.ll
index 2deefa618109..ef57e55e15e3 100644
--- a/test/Transforms/PhaseOrdering/basic.ll
+++ b/test/Transforms/PhaseOrdering/basic.ll
@@ -13,9 +13,9 @@ define void @test1() nounwind ssp {
   %i = alloca i8*, align 8
   %call = call i8* @malloc(i64 1)
   store i8* %call, i8** %i, align 8
-  %tmp = load i8** %i, align 8
+  %tmp = load i8*, i8** %i, align 8
   store i8 1, i8* %tmp
-  %tmp1 = load i8** %i, align 8
+  %tmp1 = load i8*, i8** %i, align 8
   call void @free(i8* %tmp1)
   ret void
 
@@ -31,15 +31,15 @@ define void @test1() nounwind ssp {
 define i32 @test2(i32 %a, i32* %p) nounwind uwtable ssp {
 entry:
   %div = udiv i32 %a, 4
-  %arrayidx = getelementptr inbounds i32* %p, i64 0
+  %arrayidx = getelementptr inbounds i32, i32* %p, i64 0
   store i32 %div, i32* %arrayidx, align 4
   %add = add i32 %div, %div
-  %arrayidx1 = getelementptr inbounds i32* %p, i64 1
+  %arrayidx1 = getelementptr inbounds i32, i32* %p, i64 1
   store i32 %add, i32* %arrayidx1, align 4
-  %arrayidx2 = getelementptr inbounds i32* %p, i64 1
-  %0 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %p, i64 0
-  %1 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %p, i64 1
+  %0 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %p, i64 0
+  %1 = load i32, i32* %arrayidx3, align 4
   %mul = mul i32 2, %1
   %sub = sub i32 %0, %mul
   ret i32 %sub
diff --git a/test/Transforms/PhaseOrdering/gdce.ll b/test/Transforms/PhaseOrdering/gdce.ll
index 95f06757a788..6f79eb815832 100644
--- a/test/Transforms/PhaseOrdering/gdce.ll
+++ b/test/Transforms/PhaseOrdering/gdce.ll
@@ -15,7 +15,7 @@
 @_ZTV4Base = linkonce_odr unnamed_addr constant [4 x i8*] [i8* null, i8* bitcast ({ i8*, i8* }* @_ZTI4Base to i8*), i8* bitcast (void (%class.Base*)* @_ZN4BaseD1Ev to i8*), i8* bitcast (void (%class.Base*)* @_ZN4BaseD0Ev to i8*)]
 @_ZTVN10__cxxabiv117__class_type_infoE = external global i8*
 @_ZTS4Base = linkonce_odr constant [6 x i8] c"4Base\00"
-@_ZTI4Base = linkonce_odr unnamed_addr constant { i8*, i8* } { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([6 x i8]* @_ZTS4Base, i32 0, i32 0) }
+@_ZTI4Base = linkonce_odr unnamed_addr constant { i8*, i8* } { i8* bitcast (i8** getelementptr inbounds (i8*, i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([6 x i8], [6 x i8]* @_ZTS4Base, i32 0, i32 0) }
 
 define i32 @main() uwtable ssp {
 entry:
@@ -27,7 +27,7 @@ entry:
   store i32 0, i32* %retval
   store i32 1, i32* %cleanup.dest.slot
   call void @_ZN4BaseD1Ev(%class.Base* %b)
-  %0 = load i32* %retval
+  %0 = load i32, i32* %retval
   ret i32 %0
 }
 
@@ -35,7 +35,7 @@ define linkonce_odr void @_ZN4BaseC1Ev(%class.Base* %this) unnamed_addr uwtable
 entry:
   %this.addr = alloca %class.Base*, align 8
   store %class.Base* %this, %class.Base** %this.addr, align 8
-  %this1 = load %class.Base** %this.addr
+  %this1 = load %class.Base*, %class.Base** %this.addr
   call void @_ZN4BaseC2Ev(%class.Base* %this1)
   ret void
 }
@@ -44,7 +44,7 @@ define linkonce_odr void @_ZN4BaseD1Ev(%class.Base* %this) unnamed_addr uwtable
 entry:
   %this.addr = alloca %class.Base*, align 8
   store %class.Base* %this, %class.Base** %this.addr, align 8
-  %this1 = load %class.Base** %this.addr
+  %this1 = load %class.Base*, %class.Base** %this.addr
   call void @_ZN4BaseD2Ev(%class.Base* %this1)
   ret void
 }
@@ -53,7 +53,7 @@ define linkonce_odr void @_ZN4BaseD2Ev(%class.Base* %this) unnamed_addr nounwind
 entry:
   %this.addr = alloca %class.Base*, align 8
   store %class.Base* %this, %class.Base** %this.addr, align 8
-  %this1 = load %class.Base** %this.addr
+  %this1 = load %class.Base*, %class.Base** %this.addr
   ret void
 }
 
@@ -61,9 +61,9 @@ define linkonce_odr void @_ZN4BaseC2Ev(%class.Base* %this) unnamed_addr nounwind
 entry:
   %this.addr = alloca %class.Base*, align 8
   store %class.Base* %this, %class.Base** %this.addr, align 8
-  %this1 = load %class.Base** %this.addr
+  %this1 = load %class.Base*, %class.Base** %this.addr
   %0 = bitcast %class.Base* %this1 to i8***
-  store i8** getelementptr inbounds ([4 x i8*]* @_ZTV4Base, i64 0, i64 2), i8*** %0
+  store i8** getelementptr inbounds ([4 x i8*], [4 x i8*]* @_ZTV4Base, i64 0, i64 2), i8*** %0
   ret void
 }
 
@@ -73,7 +73,7 @@ entry:
   %exn.slot = alloca i8*
   %ehselector.slot = alloca i32
   store %class.Base* %this, %class.Base** %this.addr, align 8
-  %this1 = load %class.Base** %this.addr
+  %this1 = load %class.Base*, %class.Base** %this.addr
   invoke void @_ZN4BaseD1Ev(%class.Base* %this1)
           to label %invoke.cont unwind label %lpad
 
@@ -94,8 +94,8 @@ lpad:                                             ; preds = %entry
   br label %eh.resume
 
 eh.resume:                                        ; preds = %lpad
-  %exn = load i8** %exn.slot
-  %sel = load i32* %ehselector.slot
+  %exn = load i8*, i8** %exn.slot
+  %sel = load i32, i32* %ehselector.slot
   %lpad.val = insertvalue { i8*, i32 } undef, i8* %exn, 0
   %lpad.val2 = insertvalue { i8*, i32 } %lpad.val, i32 %sel, 1
   resume { i8*, i32 } %lpad.val2
diff --git a/test/Transforms/PhaseOrdering/scev.ll b/test/Transforms/PhaseOrdering/scev.ll
index 39adb6b73d3a..c616ca2d7684 100644
--- a/test/Transforms/PhaseOrdering/scev.ll
+++ b/test/Transforms/PhaseOrdering/scev.ll
@@ -7,7 +7,7 @@
 ; CHECK: test1
 ; The loop body contains two increments by %div.
 ; Make sure that 2*%div is recognizable, and not expressed as a bit mask of %d.
-; CHECK: -->  {%p,+,(2 * (%d /u 4) * sizeof(i32))}
+; CHECK: -->  {%p,+,(8 * (%d /u 4))}
 define void @test1(i64 %d, i32* %p) nounwind uwtable ssp {
 entry:
   %div = udiv i64 %d, 4
@@ -21,9 +21,9 @@ for.cond:                                         ; preds = %for.inc, %entry
 
 for.body:                                         ; preds = %for.cond
   store i32 0, i32* %p.addr.0, align 4
-  %add.ptr = getelementptr inbounds i32* %p.addr.0, i64 %div
+  %add.ptr = getelementptr inbounds i32, i32* %p.addr.0, i64 %div
   store i32 1, i32* %add.ptr, align 4
-  %add.ptr1 = getelementptr inbounds i32* %add.ptr, i64 %div
+  %add.ptr1 = getelementptr inbounds i32, i32* %add.ptr, i64 %div
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
@@ -36,7 +36,7 @@ for.end:                                          ; preds = %for.cond
 
 ; CHECK: test1a
 ; Same thing as test1, but it is even more tempting to fold 2 * (%d /u 2)
-; CHECK: -->  {%p,+,(2 * (%d /u 2) * sizeof(i32))}
+; CHECK: -->  {%p,+,(8 * (%d /u 2))}
 define void @test1a(i64 %d, i32* %p) nounwind uwtable ssp {
 entry:
   %div = udiv i64 %d, 2
@@ -50,9 +50,9 @@ for.cond:                                         ; preds = %for.inc, %entry
 
 for.body:                                         ; preds = %for.cond
   store i32 0, i32* %p.addr.0, align 4
-  %add.ptr = getelementptr inbounds i32* %p.addr.0, i64 %div
+  %add.ptr = getelementptr inbounds i32, i32* %p.addr.0, i64 %div
   store i32 1, i32* %add.ptr, align 4
-  %add.ptr1 = getelementptr inbounds i32* %add.ptr, i64 %div
+  %add.ptr1 = getelementptr inbounds i32, i32* %add.ptr, i64 %div
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body
diff --git a/test/Transforms/PlaceSafepoints/basic.ll b/test/Transforms/PlaceSafepoints/basic.ll
new file mode 100644
index 000000000000..32aa4da68f21
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/basic.ll
@@ -0,0 +1,94 @@
+; RUN: opt %s -S -place-safepoints | FileCheck %s
+
+
+; Do we insert a simple entry safepoint?
+define void @test_entry() gc "statepoint-example" {
+; CHECK-LABEL: @test_entry
+entry:
+; CHECK-LABEL: entry
+; CHECK: statepoint
+  ret void
+}
+
+; On a non-gc function, we should NOT get an entry safepoint
+define void @test_negative() {
+; CHECK-LABEL: @test_negative
+entry:
+; CHECK-NOT: statepoint
+  ret void
+}
+
+; Do we insert a backedge safepoint in a statically
+; infinite loop?
+define void @test_backedge() gc "statepoint-example" {
+; CHECK-LABEL: test_backedge
+entry:
+; CHECK-LABEL: entry
+; This statepoint is technically not required, but we don't exploit that yet.
+; CHECK: statepoint
+  br label %other
+
+; CHECK-LABEL: other
+; CHECK: statepoint
+other:
+  call void undef()
+  br label %other
+}
+
+; Check that we remove an unreachable block rather than trying
+; to insert a backedge safepoint
+define void @test_unreachable() gc "statepoint-example" {
+; CHECK-LABEL: test_unreachable
+entry:
+; CHECK-LABEL: entry
+; CHECK: statepoint
+  ret void
+
+; CHECK-NOT: other
+; CHECK-NOT: statepoint
+other:
+  br label %other
+}
+
+declare void @foo()
+
+; Do we turn a call into it's own statepoint
+define void @test_simple_call() gc "statepoint-example" {
+; CHECK-LABEL: test_simple_call
+entry:
+  br label %other
+other:
+; CHECK-LABEL: other
+; CHECK: statepoint
+; CHECK-NOT: gc.result
+  call void @foo()
+  ret void
+}
+
+declare zeroext i1 @i1_return_i1(i1)
+
+define i1 @test_call_with_result() gc "statepoint-example" {
+; CHECK-LABEL: test_call_with_result
+; This is checking that a statepoint_poll + statepoint + result is
+; inserted for a function that takes 1 argument.
+; CHECK: gc.statepoint.p0f_isVoidf
+; CHECK: gc.statepoint.p0f_i1i1f
+; CHECK: (i64 2882400000, i32 0, i1 (i1)* @i1_return_i1, i32 1, i32 0, i1 false, i32 0, i32 0)
+; CHECK: %call1.2 = call i1 @llvm.experimental.gc.result.i1
+entry:
+  %call1 = tail call i1 (i1) @i1_return_i1(i1 false)
+  ret i1 %call1
+}
+
+; This function is inlined when inserting a poll.  To avoid recursive 
+; issues, make sure we don't place safepoints in it.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+; CHECK-LABEL: entry
+; CHECK-NEXT: do_safepoint
+; CHECK-NEXT: ret void 
+entry:
+  call void @do_safepoint()
+  ret void
+}
diff --git a/test/Transforms/PlaceSafepoints/call-in-loop.ll b/test/Transforms/PlaceSafepoints/call-in-loop.ll
new file mode 100644
index 000000000000..9edfeb75e0c4
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/call-in-loop.ll
@@ -0,0 +1,31 @@
+; If there's a call in the loop which dominates the backedge, we 
+; don't need a safepoint poll (since the callee must contain a 
+; poll test).
+;; RUN: opt %s -place-safepoints -S | FileCheck %s
+
+declare void @foo()
+
+define void @test1() gc "statepoint-example" {
+; CHECK-LABEL: test1
+
+entry:
+; CHECK-LABEL: entry
+; CHECK: statepoint
+  br label %loop
+
+loop:
+; CHECK-LABEL: loop
+; CHECK: @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 2882400000, i32 0, void ()* @foo
+; CHECK-NOT: statepoint
+  call void @foo()
+  br label %loop
+}
+
+; This function is inlined when inserting a poll.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+entry:
+  call void @do_safepoint()
+  ret void
+}
diff --git a/test/Transforms/PlaceSafepoints/call_gc_result.ll b/test/Transforms/PlaceSafepoints/call_gc_result.ll
new file mode 100644
index 000000000000..d78a0989c3b1
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/call_gc_result.ll
@@ -0,0 +1,38 @@
+;; RUN: opt %s -place-safepoints -S | FileCheck %s
+
+;; This test is to verify that gc_result from a call statepoint
+;; can have preceding phis in its parent basic block. Unlike
+;; invoke statepoint, call statepoint does not terminate the
+;; block, and thus its gc_result is in the same block with the
+;; call statepoint.
+
+declare i32 @foo()
+
+define i32 @test1(i1 %cond, i32 %a) gc "statepoint-example" {
+entry:
+  br i1 %cond, label %branch1, label %branch2
+  
+branch1:
+  %b = add i32 %a, 1
+  br label %merge
+ 
+branch2:
+  br label %merge
+
+merge:
+;; CHECK: 		%phi = phi i32 [ %a, %branch2 ], [ %b, %branch1 ]
+;; CHECK-NEXT:  %safepoint_token.1 = call i32 (i64, i32, i32 ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i32f(i64 2882400000, i32 0, i32 ()* @foo, i32 0, i32 0, i32 0, i32 0)
+;; CHECK-NEXT:  %ret.2 = call i32 @llvm.experimental.gc.result.i32(i32 %safepoint_token.1)
+  %phi = phi i32 [ %a, %branch2 ], [ %b, %branch1 ]
+  %ret = call i32 @foo()
+  ret i32 %ret
+}
+
+; This function is inlined when inserting a poll.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+entry:
+  call void @do_safepoint()
+  ret void
+}
diff --git a/test/Transforms/PlaceSafepoints/finite-loops.ll b/test/Transforms/PlaceSafepoints/finite-loops.ll
new file mode 100644
index 000000000000..3cc7158afcfe
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/finite-loops.ll
@@ -0,0 +1,80 @@
+; Tests to ensure that we are not placing backedge safepoints in
+; loops which are clearly finite.
+;; RUN: opt %s -place-safepoints -S | FileCheck %s
+
+
+; A simple counted loop with trivially known range
+define void @test1(i32) gc "statepoint-example" {
+; CHECK-LABEL: test1
+; CHECK-LABEL: entry
+; CHECK: statepoint
+; CHECK-LABEL: loop
+; CHECK-NOT: statepoint
+
+entry:
+  br label %loop
+
+loop:
+  %counter = phi i32 [ 0 , %entry ], [ %counter.inc , %loop ]
+  %counter.inc = add i32 %counter, 1
+  %counter.cmp = icmp slt i32 %counter.inc, 16
+  br i1 %counter.cmp, label %loop, label %exit
+
+exit:
+  ret void
+}
+
+; The same counted loop, but with an unknown early exit
+define void @test2(i32) gc "statepoint-example" {
+; CHECK-LABEL: test2
+; CHECK-LABEL: entry
+; CHECK: statepoint
+; CHECK-LABEL: loop
+; CHECK-NOT: statepoint
+
+entry:
+  br label %loop
+
+loop:
+  %counter = phi i32 [ 0 , %entry ], [ %counter.inc , %continue ]
+  %counter.inc = add i32 %counter, 1
+  %counter.cmp = icmp slt i32 %counter.inc, 16
+  br i1 undef, label %continue, label %exit
+
+continue:
+  br i1 %counter.cmp, label %loop, label %exit
+
+exit:
+  ret void
+}
+
+; The range is a 8 bit value and we can't overflow
+define void @test3(i8 %upper) gc "statepoint-example" {
+; CHECK-LABEL: test3
+; CHECK-LABEL: entry
+; CHECK: statepoint
+; CHECK-LABEL: loop
+; CHECK-NOT: statepoint
+
+entry:
+  br label %loop
+
+loop:
+  %counter = phi i8 [ 0 , %entry ], [ %counter.inc , %loop ]
+  %counter.inc = add nsw i8 %counter, 1
+  %counter.cmp = icmp slt i8 %counter.inc, %upper
+  br i1 %counter.cmp, label %loop, label %exit
+
+exit:
+  ret void
+}
+
+
+; This function is inlined when inserting a poll.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+entry:
+  call void @do_safepoint()
+  ret void
+}
\ No newline at end of file
diff --git a/test/Transforms/PlaceSafepoints/invokes.ll b/test/Transforms/PlaceSafepoints/invokes.ll
new file mode 100644
index 000000000000..913e148d9bcc
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/invokes.ll
@@ -0,0 +1,111 @@
+; RUN: opt %s -S -place-safepoints | FileCheck %s
+
+declare i64 addrspace(1)* @"some_call"(i64 addrspace(1)*)
+declare i32 @"personality_function"()
+
+define i64 addrspace(1)* @test_basic(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: entry:
+entry:
+  ; CHECK: invoke
+  ; CHECK: statepoint
+  ; CHECK: some_call
+  %ret_val = invoke i64 addrspace(1)* @"some_call"(i64 addrspace(1)* %obj)
+               to label %normal_return unwind label %exceptional_return
+
+; CHECK-LABEL: normal_return:
+; CHECK: gc.result
+; CHECK: ret i64
+
+normal_return:
+  ret i64 addrspace(1)* %ret_val
+
+; CHECK-LABEL: exceptional_return:
+; CHECK: landingpad
+; CHECK: ret i64
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @"personality_function"
+          cleanup
+  ret i64 addrspace(1)* %obj1
+}
+
+define i64 addrspace(1)* @test_two_invokes(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: entry:
+entry:
+  ; CHECK: invoke 
+  ; CHECK: statepoint
+  ; CHECK: some_call
+  %ret_val1 = invoke i64 addrspace(1)* @"some_call"(i64 addrspace(1)* %obj)
+               to label %second_invoke unwind label %exceptional_return
+
+; CHECK-LABEL: second_invoke:
+second_invoke:
+  ; CHECK: invoke
+  ; CHECK: statepoint
+  ; CHECK: some_call
+  %ret_val2 = invoke i64 addrspace(1)* @"some_call"(i64 addrspace(1)* %ret_val1)
+                to label %normal_return unwind label %exceptional_return
+
+; CHECK-LABEL: normal_return:
+normal_return:
+  ; CHECK: gc.result
+  ; CHECK: ret i64
+  ret i64 addrspace(1)* %ret_val2
+
+; CHECK: exceptional_return:
+; CHECK: ret i64
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @"personality_function"
+          cleanup
+  ret i64 addrspace(1)* %obj1
+}
+
+define i64 addrspace(1)* @test_phi_node(i1 %cond, i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test_phi_node
+; CHECK-LABEL: entry:
+entry:
+  br i1 %cond, label %left, label %right
+
+left:
+  %ret_val_left = invoke i64 addrspace(1)* @"some_call"(i64 addrspace(1)* %obj)
+                    to label %merge unwind label %exceptional_return
+
+right:
+  %ret_val_right = invoke i64 addrspace(1)* @"some_call"(i64 addrspace(1)* %obj)
+                     to label %merge unwind label %exceptional_return
+
+; CHECK: merge[[A:[0-9]]]:
+; CHECK: gc.result
+; CHECK: br label %[[with_phi:merge[0-9]*]]
+
+; CHECK: merge[[B:[0-9]]]:
+; CHECK: gc.result
+; CHECK: br label %[[with_phi]]
+
+; CHECK: [[with_phi]]:
+; CHECK: phi
+; CHECK: ret i64 addrspace(1)* %ret_val
+merge:
+  %ret_val = phi i64 addrspace(1)* [%ret_val_left, %left], [%ret_val_right, %right]
+  ret i64 addrspace(1)* %ret_val
+
+; CHECK-LABEL: exceptional_return:
+; CHECK: ret i64 addrspace(1)*
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @"personality_function"
+          cleanup
+  ret i64 addrspace(1)* %obj
+}
+
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+; CHECK-LABEL: entry
+; CHECK-NEXT: do_safepoint
+; CHECK-NEXT: ret void 
+entry:
+  call void @do_safepoint()
+  ret void
+}
\ No newline at end of file
diff --git a/test/Transforms/PlaceSafepoints/memset.ll b/test/Transforms/PlaceSafepoints/memset.ll
new file mode 100644
index 000000000000..534b2f120581
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/memset.ll
@@ -0,0 +1,20 @@
+; RUN: opt -S -place-safepoints %s | FileCheck %s
+
+define void @test(i32, i8 addrspace(1)* %ptr) gc "statepoint-example" {
+; CHECK-LABEL: @test
+; CHECK-NEXT: llvm.memset
+; CHECK: do_safepoint
+; CHECK: @foo
+  call void @llvm.memset.p1i8.i64(i8 addrspace(1)* %ptr, i8 0, i64 24, i32 8, i1 false)
+  call void @foo()
+  ret void
+}
+
+declare void @foo()
+declare void @llvm.memset.p1i8.i64(i8 addrspace(1)*, i8, i64, i32, i1)
+
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+  call void @do_safepoint()
+  ret void
+}
diff --git a/test/Transforms/PlaceSafepoints/patchable-statepoints.ll b/test/Transforms/PlaceSafepoints/patchable-statepoints.ll
new file mode 100644
index 000000000000..ac0aa29e5049
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/patchable-statepoints.ll
@@ -0,0 +1,44 @@
+; RUN: opt -place-safepoints -S < %s | FileCheck %s
+
+declare void @f()
+declare i32 @personality_function()
+
+define void @test_id() gc "statepoint-example" {
+; CHECK-LABEL: @test_id(
+entry:
+; CHECK-LABEL: entry:
+; CHECK: invoke i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 100, i32 0, void ()* @f
+  invoke void @f()  "statepoint-id"="100" to label %normal_return unwind label %exceptional_return
+
+normal_return:
+  ret void
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @personality_function cleanup
+  ret void
+}
+
+define void @test_num_patch_bytes() gc "statepoint-example" {
+; CHECK-LABEL: @test_num_patch_bytes(
+entry:
+; CHECK-LABEL: entry:
+; CHECK: invoke i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 2882400000, i32 99, void ()* null,
+  invoke void @f()  "statepoint-num-patch-bytes"="99" to label %normal_return unwind label %exceptional_return
+
+normal_return:
+  ret void
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @personality_function cleanup
+  ret void
+}
+
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+entry:
+  call void @do_safepoint()
+  ret void
+}
+
+; CHECK-NOT: statepoint-id
+; CHECK-NOT: statepoint-num-patch_bytes
diff --git a/test/Transforms/PlaceSafepoints/split-backedge.ll b/test/Transforms/PlaceSafepoints/split-backedge.ll
new file mode 100644
index 000000000000..b9fad45709d0
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/split-backedge.ll
@@ -0,0 +1,46 @@
+;; A very basic test to make sure that splitting the backedge keeps working
+;; RUN: opt -place-safepoints -spp-split-backedge=1 -S %s | FileCheck %s
+
+define void @test(i32, i1 %cond) gc "statepoint-example" {
+; CHECK-LABEL: @test
+; CHECK-LABEL: loop.loop_crit_edge
+; CHECK: gc.statepoint
+; CHECK-NEXT: br label %loop
+entry:
+  br label %loop
+
+loop:
+  br i1 %cond, label %loop, label %exit
+
+exit:
+  ret void
+}
+
+; Test for the case where a single conditional branch jumps to two
+; different loop header blocks.  Since we're currently using LoopSimplfy
+; this doesn't hit the interesting case, but once we remove that, we need
+; to be sure this keeps working.
+define void @test2(i32, i1 %cond) gc "statepoint-example" {
+; CHECK-LABEL: @test2
+; CHECK-LABEL: loop2.loop2_crit_edge:
+; CHECK: gc.statepoint
+; CHECK-NEXT: br label %loop2
+; CHECK-LABEL: loop2.loop_crit_edge:
+; CHECK: gc.statepoint
+; CHECK-NEXT: br label %loop
+entry:
+  br label %loop
+
+loop:
+  br label %loop2
+
+loop2:
+  br i1 %cond, label %loop, label %loop2
+}
+
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+entry:
+  call void @do_safepoint()
+  ret void
+}
\ No newline at end of file
diff --git a/test/Transforms/PlaceSafepoints/statepoint-calling-conventions.ll b/test/Transforms/PlaceSafepoints/statepoint-calling-conventions.ll
new file mode 100644
index 000000000000..e31c2aa5166a
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/statepoint-calling-conventions.ll
@@ -0,0 +1,42 @@
+; RUN: opt -place-safepoints -S < %s | FileCheck %s
+
+; Ensure that the gc.statepoint calls / invokes we generate carry over
+; the right calling conventions.
+
+define i64 addrspace(1)* @test_invoke_format(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: @test_invoke_format(
+; CHECK-LABEL: entry:
+; CHECK: invoke coldcc i32 (i64, i32, i64 addrspace(1)* (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64p1i64f(i64 2882400000, i32 0, i64 addrspace(1)* (i64 addrspace(1)*)* @callee, i32 1, i32 0, i64 addrspace(1)* %obj, i32 0, i32 0)
+entry:
+  %ret_val = invoke coldcc i64 addrspace(1)* @callee(i64 addrspace(1)* %obj)
+               to label %normal_return unwind label %exceptional_return
+
+normal_return:
+  ret i64 addrspace(1)* %ret_val
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @personality
+          cleanup
+  ret i64 addrspace(1)* %obj1
+}
+
+define i64 addrspace(1)* @test_call_format(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: @test_call_format(
+; CHECK-LABEL: entry:
+; CHECK: call coldcc i32 (i64, i32, i64 addrspace(1)* (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64p1i64f(i64 2882400000, i32 0, i64 addrspace(1)* (i64 addrspace(1)*)* @callee, i32 1, i32 0, i64 addrspace(1)* %obj, i32 0, i32 0)
+entry:
+  %ret_val = call coldcc i64 addrspace(1)* @callee(i64 addrspace(1)* %obj)
+  ret i64 addrspace(1)* %ret_val
+}
+
+; This function is inlined when inserting a poll.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+entry:
+  call void @do_safepoint()
+  ret void
+}
+
+declare coldcc i64 addrspace(1)* @callee(i64 addrspace(1)*)
+declare i32 @personality()
diff --git a/test/Transforms/PlaceSafepoints/statepoint-coreclr.ll b/test/Transforms/PlaceSafepoints/statepoint-coreclr.ll
new file mode 100644
index 000000000000..dc468966877f
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/statepoint-coreclr.ll
@@ -0,0 +1,31 @@
+; RUN: opt %s -S -place-safepoints | FileCheck %s

+

+; Basic test to make sure that safepoints are placed

+; for CoreCLR GC

+

+declare void @foo()

+

+define void @test_simple_call() gc "coreclr" {

+; CHECK-LABEL: test_simple_call

+entry:

+  br label %other

+other:

+; CHECK-LABEL: other

+; CHECK: statepoint

+; CHECK-NOT: gc.result

+  call void @foo()

+  ret void

+}

+

+; This function is inlined when inserting a poll.  To avoid recursive

+; issues, make sure we don't place safepoints in it.

+declare void @do_safepoint()

+define void @gc.safepoint_poll() {

+; CHECK-LABEL: gc.safepoint_poll

+; CHECK-LABEL: entry

+; CHECK-NEXT: do_safepoint

+; CHECK-NEXT: ret void

+entry:

+  call void @do_safepoint()

+  ret void

+}

diff --git a/test/Transforms/PlaceSafepoints/statepoint-format.ll b/test/Transforms/PlaceSafepoints/statepoint-format.ll
new file mode 100644
index 000000000000..328b670873eb
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/statepoint-format.ll
@@ -0,0 +1,42 @@
+; RUN: opt -place-safepoints -S < %s | FileCheck %s
+
+; Ensure that the gc.statepoint calls / invokes we generate have the
+; set of arguments we expect it to have.
+
+define i64 addrspace(1)* @test_invoke_format(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: @test_invoke_format(
+; CHECK-LABEL: entry:
+; CHECK: invoke i32 (i64, i32, i64 addrspace(1)* (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64p1i64f(i64 2882400000, i32 0, i64 addrspace(1)* (i64 addrspace(1)*)* @callee, i32 1, i32 0, i64 addrspace(1)* %obj, i32 0, i32 0)
+entry:
+  %ret_val = invoke i64 addrspace(1)* @callee(i64 addrspace(1)* %obj)
+               to label %normal_return unwind label %exceptional_return
+
+normal_return:
+  ret i64 addrspace(1)* %ret_val
+
+exceptional_return:
+  %landing_pad4 = landingpad {i8*, i32} personality i32 ()* @personality
+          cleanup
+  ret i64 addrspace(1)* %obj1
+}
+
+define i64 addrspace(1)* @test_call_format(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj1) gc "statepoint-example" {
+; CHECK-LABEL: @test_call_format(
+; CHECK-LABEL: entry:
+; CHECK: call i32 (i64, i32, i64 addrspace(1)* (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64p1i64f(i64 2882400000, i32 0, i64 addrspace(1)* (i64 addrspace(1)*)* @callee, i32 1, i32 0, i64 addrspace(1)* %obj, i32 0, i32 0)
+entry:
+  %ret_val = call i64 addrspace(1)* @callee(i64 addrspace(1)* %obj)
+  ret i64 addrspace(1)* %ret_val
+}
+
+; This function is inlined when inserting a poll.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+entry:
+  call void @do_safepoint()
+  ret void
+}
+
+declare i64 addrspace(1)* @callee(i64 addrspace(1)*)
+declare i32 @personality()
diff --git a/test/Transforms/PlaceSafepoints/statepoint-frameescape.ll b/test/Transforms/PlaceSafepoints/statepoint-frameescape.ll
new file mode 100644
index 000000000000..a6ee5ee078fb
--- /dev/null
+++ b/test/Transforms/PlaceSafepoints/statepoint-frameescape.ll
@@ -0,0 +1,29 @@
+; RUN: opt %s -S -place-safepoints | FileCheck %s
+
+declare void @llvm.frameescape(...)
+
+; Do we insert the entry safepoint after the frameescape intrinsic?
+define void @parent() gc "statepoint-example" {
+; CHECK-LABEL: @parent
+entry:
+; CHECK-LABEL: entry
+; CHECK-NEXT: alloca
+; CHECK-NEXT: frameescape
+; CHECK-NEXT: statepoint
+  %ptr = alloca i32
+  call void (...) @llvm.frameescape(i32* %ptr)
+  ret void
+}
+
+; This function is inlined when inserting a poll.  To avoid recursive 
+; issues, make sure we don't place safepoints in it.
+declare void @do_safepoint()
+define void @gc.safepoint_poll() {
+; CHECK-LABEL: gc.safepoint_poll
+; CHECK-LABEL: entry
+; CHECK-NEXT: do_safepoint
+; CHECK-NEXT: ret void 
+entry:
+  call void @do_safepoint()
+  ret void
+}
\ No newline at end of file
diff --git a/test/Transforms/PruneEH/2003-11-21-PHIUpdate.ll b/test/Transforms/PruneEH/2003-11-21-PHIUpdate.ll
deleted file mode 100644
index a01070308bed..000000000000
--- a/test/Transforms/PruneEH/2003-11-21-PHIUpdate.ll
+++ /dev/null
@@ -1,15 +0,0 @@
-; RUN: opt < %s -prune-eh -disable-output
-
-define internal void @callee() {
-	ret void
-}
-
-define i32 @caller() {
-; <label>:0
-	invoke void @callee( )
-			to label %E unwind label %E
-E:		; preds = %0, %0
-	%X = phi i32 [ 0, %0 ], [ 0, %0 ]		; <i32> [#uses=1]
-	ret i32 %X
-}
-
diff --git a/test/Transforms/PruneEH/recursivetest.ll b/test/Transforms/PruneEH/recursivetest.ll
index 724c7cf224c1..bc002ae70d3b 100644
--- a/test/Transforms/PruneEH/recursivetest.ll
+++ b/test/Transforms/PruneEH/recursivetest.ll
@@ -6,6 +6,8 @@ define internal i32 @foo() {
 Normal:		; preds = %0
 	ret i32 12
 Except:		; preds = %0
+        landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
+                catch i8* null
 	ret i32 123
 }
 
@@ -15,6 +17,9 @@ define i32 @caller() {
 Normal:		; preds = %0
 	ret i32 0
 Except:		; preds = %0
+        landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
+                catch i8* null
 	ret i32 1
 }
 
+declare i32 @__gxx_personality_v0(...)
diff --git a/test/Transforms/PruneEH/seh-nounwind.ll b/test/Transforms/PruneEH/seh-nounwind.ll
new file mode 100644
index 000000000000..4b69ae4b28dd
--- /dev/null
+++ b/test/Transforms/PruneEH/seh-nounwind.ll
@@ -0,0 +1,31 @@
+; RUN: opt -S -prune-eh < %s | FileCheck %s
+
+; Don't remove invokes of nounwind functions if the personality handles async
+; exceptions. The @div function in this test can fault, even though it can't
+; throw a synchronous exception.
+
+define i32 @div(i32 %n, i32 %d) nounwind {
+entry:
+  %div = sdiv i32 %n, %d
+  ret i32 %div
+}
+
+define i32 @main() nounwind {
+entry:
+  %call = invoke i32 @div(i32 10, i32 0)
+          to label %__try.cont unwind label %lpad
+
+lpad:
+  %0 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
+          catch i8* null
+  br label %__try.cont
+
+__try.cont:
+  %retval.0 = phi i32 [ %call, %entry ], [ 0, %lpad ]
+  ret i32 %retval.0
+}
+
+; CHECK-LABEL: define i32 @main()
+; CHECK: invoke i32 @div(i32 10, i32 0)
+
+declare i32 @__C_specific_handler(...)
diff --git a/test/Transforms/PruneEH/simpletest.ll b/test/Transforms/PruneEH/simpletest.ll
index 77c429dae6d6..6154a80b68a3 100644
--- a/test/Transforms/PruneEH/simpletest.ll
+++ b/test/Transforms/PruneEH/simpletest.ll
@@ -15,5 +15,9 @@ Normal:		; preds = %0
 	ret i32 0
 
 Except:		; preds = %0
+        landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
+                catch i8* null
 	ret i32 1
 }
+
+declare i32 @__gxx_personality_v0(...)
diff --git a/test/Transforms/Reassociate/2011-01-26-UseAfterFree.ll b/test/Transforms/Reassociate/2011-01-26-UseAfterFree.ll
index 003fbb1b3bb3..1c8f0d2ea219 100644
--- a/test/Transforms/Reassociate/2011-01-26-UseAfterFree.ll
+++ b/test/Transforms/Reassociate/2011-01-26-UseAfterFree.ll
@@ -5,7 +5,7 @@ target triple = "i386-gnu-linux"
 
 define void @exp_averages_intraday__deviation() {
 entry:
-  %0 = load i32* undef, align 4
+  %0 = load i32, i32* undef, align 4
   %1 = shl i32 %0, 2
   %2 = add nsw i32 undef, %1
   %3 = add nsw i32 %2, undef
@@ -22,7 +22,7 @@ entry:
   unreachable
 
 "8":                                              ; preds = %"4"
-  %8 = getelementptr inbounds i8* undef, i32 %6
+  %8 = getelementptr inbounds i8, i8* undef, i32 %6
   br i1 undef, label %"13", label %"12"
 
 "12":                                             ; preds = %"8", %entry
diff --git a/test/Transforms/Reassociate/basictest.ll b/test/Transforms/Reassociate/basictest.ll
index 0194ce2e302c..015d3b0bee9f 100644
--- a/test/Transforms/Reassociate/basictest.ll
+++ b/test/Transforms/Reassociate/basictest.ll
@@ -28,9 +28,9 @@ define i32 @test2(i32 %reg109, i32 %reg1111) {
 @f = external global i32
 
 define void @test3() {
-  %A = load i32* @a
-  %B = load i32* @b
-  %C = load i32* @c
+  %A = load i32, i32* @a
+  %B = load i32, i32* @b
+  %C = load i32, i32* @c
   %t1 = add i32 %A, %B
   %t2 = add i32 %t1, %C
   %t3 = add i32 %C, %A
@@ -49,9 +49,9 @@ define void @test3() {
 }
 
 define void @test4() {
-  %A = load i32* @a
-  %B = load i32* @b
-  %C = load i32* @c
+  %A = load i32, i32* @a
+  %B = load i32, i32* @b
+  %C = load i32, i32* @c
   %t1 = add i32 %A, %B
   %t2 = add i32 %t1, %C
   %t3 = add i32 %C, %A
@@ -70,9 +70,9 @@ define void @test4() {
 }
 
 define void @test5() {
-  %A = load i32* @a
-  %B = load i32* @b
-  %C = load i32* @c
+  %A = load i32, i32* @a
+  %B = load i32, i32* @b
+  %C = load i32, i32* @c
   %t1 = add i32 %B, %A
   %t2 = add i32 %t1, %C
   %t3 = add i32 %C, %A
@@ -91,11 +91,11 @@ define void @test5() {
 }
 
 define i32 @test6() {
-  %tmp.0 = load i32* @a
-  %tmp.1 = load i32* @b
+  %tmp.0 = load i32, i32* @a
+  %tmp.1 = load i32, i32* @b
   ; (a+b)
   %tmp.2 = add i32 %tmp.0, %tmp.1
-  %tmp.4 = load i32* @c
+  %tmp.4 = load i32, i32* @c
   ; (a+b)+c
   %tmp.5 = add i32 %tmp.2, %tmp.4
   ; (a+c)
diff --git a/test/Transforms/Reassociate/crash.ll b/test/Transforms/Reassociate/crash.ll
index 770f97371d7e..f8774ea509a4 100644
--- a/test/Transforms/Reassociate/crash.ll
+++ b/test/Transforms/Reassociate/crash.ll
@@ -17,7 +17,7 @@ for.cond:                                         ; preds = %for.body, %entry
 ; PR5981
 define i32 @test2() nounwind ssp {
 entry:
-  %0 = load i32* undef, align 4
+  %0 = load i32, i32* undef, align 4
   %1 = mul nsw i32 undef, %0
   %2 = mul nsw i32 undef, %0
   %3 = add nsw i32 undef, %1
@@ -50,7 +50,7 @@ _:
   br label %_33
 
 _33:                                              ; preds = %_33, %_
-  %tmp348 = load i8* %arg, align 1
+  %tmp348 = load i8, i8* %arg, align 1
   %tmp349 = lshr i8 %tmp348, 7
   %tmp350 = or i8 %tmp349, 42
   %tmp351 = add i8 %tmp350, -42
@@ -123,7 +123,7 @@ for.cond:                                         ; preds = %for.cond, %entry
 ; PR12963
 @a = external global i8
 define i8 @f0(i8 %x) {
-  %t0 = load i8* @a
+  %t0 = load i8, i8* @a
   %t1 = mul i8 %x, %x
   %t2 = mul i8 %t1, %t1
   %t3 = mul i8 %t2, %t2
diff --git a/test/Transforms/Reassociate/crash2.ll b/test/Transforms/Reassociate/crash2.ll
new file mode 100644
index 000000000000..b51a88ca674d
--- /dev/null
+++ b/test/Transforms/Reassociate/crash2.ll
@@ -0,0 +1,25 @@
+; RUN: opt -reassociate %s -S -o - | FileCheck %s
+
+; Reassociate pass used to crash on these example
+
+
+define float @undef1() {
+wrapper_entry:
+; CHECK-LABEL: @undef1
+; CHECK: ret float fadd (float undef, float fadd (float undef, float fadd (float fsub (float -0.000000e+00, float undef), float fsub (float -0.000000e+00, float undef))))
+  %0 = fadd fast float undef, undef
+  %1 = fsub fast float undef, %0
+  %2 = fadd fast float undef, %1
+  ret float %2
+}
+
+define void @undef2() {
+wrapper_entry:
+; CHECK-LABEL: @undef2
+; CHECK: unreachable
+  %0 = fadd fast float undef, undef
+  %1 = fadd fast float %0, 1.000000e+00
+  %2 = fsub fast float %0, %1
+  %3 = fmul fast float %2, 2.000000e+00
+  unreachable
+}
diff --git a/test/Transforms/Reassociate/fast-ReassociateVector.ll b/test/Transforms/Reassociate/fast-ReassociateVector.ll
index eeae096bf944..9fbb5ccfe9a2 100644
--- a/test/Transforms/Reassociate/fast-ReassociateVector.ll
+++ b/test/Transforms/Reassociate/fast-ReassociateVector.ll
@@ -1,46 +1,192 @@
 ; RUN: opt < %s -reassociate -S | FileCheck %s
 
-; Canonicalize operands, but don't optimize floating point vector operations.
-define <4 x float> @test1() {
-; CHECK-LABEL: test1
-; CHECK-NEXT: %tmp1 = fsub fast <4 x float> zeroinitializer, zeroinitializer
-; CHECK-NEXT: %tmp2 = fmul fast <4 x float> %tmp1, zeroinitializer
+; Check that a*c+b*c is turned into (a+b)*c
+define <4 x float> @test1(<4 x float> %a, <4 x float> %b, <4 x float> %c) {
+; CHECK-LABEL: @test1
+; CHECK-NEXT: %tmp = fadd fast <4 x float> %b, %a
+; CHECK-NEXT: %tmp1 = fmul fast <4 x float> %tmp, %c
+; CHECK-NEXT: ret <4 x float> %tmp1
 
-  %tmp1 = fsub fast <4 x float> zeroinitializer, zeroinitializer
-  %tmp2 = fmul fast <4 x float> zeroinitializer, %tmp1
-  ret <4 x float> %tmp2
+  %mul = fmul fast <4 x float> %a, %c
+  %mul1 = fmul fast <4 x float> %b, %c
+  %add = fadd fast <4 x float> %mul, %mul1
+  ret <4 x float> %add
 }
 
-; Commute integer vector operations.
-define <2 x i32> @test2(<2 x i32> %x, <2 x i32> %y) {
-; CHECK-LABEL: test2
-; CHECK-NEXT: %tmp1 = add <2 x i32> %x, %y
-; CHECK-NEXT: %tmp2 = add <2 x i32> %x, %y
-; CHECK-NEXT: %tmp3 = add <2 x i32> %tmp1, %tmp2
+; Check that a*a*b+a*a*c is turned into a*(a*(b+c)).
+define <2 x float> @test2(<2 x float> %a, <2 x float> %b, <2 x float> %c) {
+; CHECK-LABEL: @test2
+; CHECK-NEXT: fadd fast <2 x float> %c, %b
+; CHECK-NEXT: fmul fast <2 x float> %a, %tmp2
+; CHECK-NEXT: fmul fast <2 x float> %tmp3, %a
+; CHECK-NEXT: ret <2 x float>
 
-  %tmp1 = add <2 x i32> %x, %y
-  %tmp2 = add <2 x i32> %y, %x
-  %tmp3 = add <2 x i32> %tmp1, %tmp2
-  ret <2 x i32> %tmp3
+  %t0 = fmul fast <2 x float> %a, %b
+  %t1 = fmul fast <2 x float> %a, %t0
+  %t2 = fmul fast <2 x float> %a, %c
+  %t3 = fmul fast <2 x float> %a, %t2
+  %t4 = fadd fast <2 x float> %t1, %t3
+  ret <2 x float> %t4
 }
 
-define <2 x i32> @test3(<2 x i32> %x, <2 x i32> %y) {
-; CHECK-LABEL: test3
-; CHECK-NEXT: %tmp1 = mul <2 x i32> %x, %y
-; CHECK-NEXT: %tmp2 = mul <2 x i32> %x, %y
-; CHECK-NEXT: %tmp3 = mul <2 x i32> %tmp1, %tmp2
+; Check that a*b+a*c+d is turned into a*(b+c)+d.
+define <2 x double> @test3(<2 x double> %a, <2 x double> %b, <2 x double> %c, <2 x double> %d) {
+; CHECK-LABEL: @test3
+; CHECK-NEXT: fadd fast <2 x double> %c, %b
+; CHECK-NEXT: fmul fast <2 x double> %tmp, %a
+; CHECK-NEXT: fadd fast <2 x double> %tmp1, %d
+; CHECK-NEXT: ret <2 x double>
 
-  %tmp1 = mul <2 x i32> %x, %y
-  %tmp2 = mul <2 x i32> %y, %x
-  %tmp3 = mul <2 x i32> %tmp1, %tmp2
-  ret <2 x i32> %tmp3
+  %t0 = fmul fast <2 x double> %a, %b
+  %t1 = fmul fast <2 x double> %a, %c
+  %t2 = fadd fast <2 x double> %t1, %d
+  %t3 = fadd fast <2 x double> %t0, %t2
+  ret <2 x double> %t3
+}
+
+; No fast-math.
+define <2 x float> @test4(<2 x float> %A) {
+; CHECK-LABEL: @test4
+; CHECK-NEXT: %X = fadd <2 x float> %A, <float 1.000000e+00, float 1.000000e+00>
+; CHECK-NEXT: %Y = fadd <2 x float> %A, <float 1.000000e+00, float 1.000000e+00>
+; CHECK-NEXT: %R = fsub <2 x float> %X, %Y
+; CHECK-NEXT: ret <2 x float> %R
+
+  %X = fadd <2 x float> %A, < float 1.000000e+00, float 1.000000e+00 >
+  %Y = fadd <2 x float> %A, < float 1.000000e+00, float 1.000000e+00 >
+  %R = fsub <2 x float> %X, %Y
+  ret <2 x float> %R
+}
+
+; Check 47*X + 47*X -> 94*X.
+define <2 x float> @test5(<2 x float> %X) {
+; CHECK-LABEL: @test5
+; CHECK-NEXT: fmul fast <2 x float> %X, <float 9.400000e+01, float 9.400000e+01>
+; CHECK-NEXT: ret <2 x float>
+
+  %Y = fmul fast <2 x float> %X, <float 4.700000e+01, float 4.700000e+01>
+  %Z = fadd fast <2 x float> %Y, %Y
+  ret <2 x float> %Z
+}
+
+; Check X+X+X -> 3*X.
+define <2 x float> @test6(<2 x float> %X) {
+; CHECK-LABEL: @test6
+; CHECK-NEXT: fmul fast <2 x float> %X, <float 3.000000e+00, float 3.000000e+00>
+; CHECK-NEXT: ret <2 x float>
+
+  %Y = fadd fast <2 x float> %X ,%X
+  %Z = fadd fast <2 x float> %Y, %X
+  ret <2 x float> %Z
+}
+
+; Check 127*W+50*W -> 177*W.
+define <2 x double> @test7(<2 x double> %W) {
+; CHECK-LABEL: @test7
+; CHECK-NEXT: fmul fast <2 x double> %W, <double 1.770000e+02, double 1.770000e+02>
+; CHECK-NEXT: ret <2 x double>
+
+  %X = fmul fast <2 x double> %W, <double 127.0, double 127.0>
+  %Y = fmul fast <2 x double> %W, <double 50.0, double 50.0>
+  %Z = fadd fast <2 x double> %Y, %X
+  ret <2 x double> %Z
+}
+
+; Check X*12*12 -> X*144.
+define <2 x float> @test8(<2 x float> %arg) {
+; CHECK-LABEL: @test8
+; CHECK: fmul fast <2 x float> %arg, <float 1.440000e+02, float 1.440000e+02>
+; CHECK-NEXT: ret <2 x float> %tmp2
+
+  %tmp1 = fmul fast <2 x float> <float 1.200000e+01, float 1.200000e+01>, %arg
+  %tmp2 = fmul fast <2 x float> %tmp1, <float 1.200000e+01, float 1.200000e+01>
+  ret <2 x float> %tmp2
+}
+
+; Check (b+(a+1234))+-a -> b+1234.
+define <2 x double> @test9(<2 x double> %b, <2 x double> %a) {
+; CHECK-LABEL: @test9
+; CHECK: fadd fast <2 x double> %b, <double 1.234000e+03, double 1.234000e+03>
+; CHECK-NEXT: ret <2 x double>
+
+  %1 = fadd fast <2 x double> %a, <double 1.234000e+03, double 1.234000e+03>
+  %2 = fadd fast <2 x double> %b, %1
+  %3 = fsub fast <2 x double> <double 0.000000e+00, double 0.000000e+00>, %a
+  %4 = fadd fast <2 x double> %2, %3
+  ret <2 x double> %4
+}
+
+; Check -(-(z*40)*a) -> a*40*z.
+define <2 x float> @test10(<2 x float> %a, <2 x float> %b, <2 x float> %z) {
+; CHECK-LABEL: @test10
+; CHECK: fmul fast <2 x float> %a, <float 4.000000e+01, float 4.000000e+01>
+; CHECK-NEXT: fmul fast <2 x float> %e, %z
+; CHECK-NEXT: ret <2 x float>
+
+  %d = fmul fast <2 x float> %z, <float 4.000000e+01, float 4.000000e+01>
+  %c = fsub fast <2 x float> <float 0.000000e+00, float 0.000000e+00>, %d
+  %e = fmul fast <2 x float> %a, %c
+  %f = fsub fast <2 x float> <float 0.000000e+00, float 0.000000e+00>, %e
+  ret <2 x float> %f
+}
+
+; Check x*y+y*x -> x*y*2.
+define <2 x double> @test11(<2 x double> %x, <2 x double> %y) {
+; CHECK-LABEL: @test11
+; CHECK-NEXT: %factor = fmul fast <2 x double> %y, <double 2.000000e+00, double 2.000000e+00>
+; CHECK-NEXT: %tmp1 = fmul fast <2 x double> %factor, %x
+; CHECK-NEXT: ret <2 x double> %tmp1
+
+  %1 = fmul fast <2 x double> %x, %y
+  %2 = fmul fast <2 x double> %y, %x
+  %3 = fadd fast <2 x double> %1, %2
+  ret <2 x double> %3
+}
+
+; FIXME: shifts should be converted to mul to assist further reassociation.
+define <2 x i64> @test12(<2 x i64> %b, <2 x i64> %c) {
+; CHECK-LABEL: @test12
+; CHECK-NEXT:  %mul = mul <2 x i64> %c, %b
+; CHECK-NEXT:  %shl = shl <2 x i64> %mul, <i64 5, i64 5>
+; CHECK-NEXT:  ret <2 x i64> %shl
+
+  %mul = mul <2 x i64> %c, %b
+  %shl = shl <2 x i64> %mul, <i64 5, i64 5>
+  ret <2 x i64> %shl
+}
+
+; FIXME: expressions with a negative const should be canonicalized to assist
+; further reassociation.
+; We would expect (-5*b)+a -> a-(5*b) but only the constant operand is commuted.
+define <4 x float> @test13(<4 x float> %a, <4 x float> %b) {
+; CHECK-LABEL: @test13
+; CHECK-NEXT:  %mul = fmul fast <4 x float> %b, <float -5.000000e+00, float -5.000000e+00, float -5.000000e+00, float -5.000000e+00>
+; CHECK-NEXT:  %add = fadd fast <4 x float> %mul, %a
+; CHECK-NEXT:  ret <4 x float> %add
+
+  %mul = fmul fast <4 x float> <float -5.000000e+00, float -5.000000e+00, float -5.000000e+00, float -5.000000e+00>, %b
+  %add = fadd fast <4 x float> %mul, %a
+  ret <4 x float> %add
+}
+
+; Break up subtract to assist further reassociation.
+; Check a+b-c -> a+b+-c.
+define <2 x i64> @test14(<2 x i64> %a, <2 x i64> %b, <2 x i64> %c) {
+; CHECK-LABEL: @test14
+; CHECK-NEXT: %add = add <2 x i64> %b, %a
+; CHECK-NEXT: %c.neg = sub <2 x i64> zeroinitializer, %c
+; CHECK-NEXT: %sub = add <2 x i64> %add, %c.neg
+; CHECK-NEXT: ret <2 x i64> %sub
+
+  %add = add <2 x i64> %b, %a
+  %sub = sub <2 x i64> %add, %c
+  ret <2 x i64> %sub
 }
 
-define <2 x i32> @test4(<2 x i32> %x, <2 x i32> %y) {
-; CHECK-LABEL: test4
-; CHECK-NEXT: %tmp1 = and <2 x i32> %x, %y
-; CHECK-NEXT: %tmp2 = and <2 x i32> %x, %y
-; CHECK-NEXT: %tmp3 = and <2 x i32> %tmp1, %tmp2
+define <2 x i32> @test15(<2 x i32> %x, <2 x i32> %y) {
+; CHECK-LABEL: test15
+; CHECK-NEXT: %tmp3 = and <2 x i32> %y, %x
+; CHECK-NEXT: ret <2 x i32> %tmp3
 
   %tmp1 = and <2 x i32> %x, %y
   %tmp2 = and <2 x i32> %y, %x
@@ -48,11 +194,10 @@ define <2 x i32> @test4(<2 x i32> %x, <2 x i32> %y) {
   ret <2 x i32> %tmp3
 }
 
-define <2 x i32> @test5(<2 x i32> %x, <2 x i32> %y) {
-; CHECK-LABEL: test5
-; CHECK-NEXT: %tmp1 = or <2 x i32> %x, %y
-; CHECK-NEXT: %tmp2 = or <2 x i32> %x, %y
-; CHECK-NEXT: %tmp3 = or <2 x i32> %tmp1, %tmp2
+define <2 x i32> @test16(<2 x i32> %x, <2 x i32> %y) {
+; CHECK-LABEL: test16
+; CHECK-NEXT: %tmp3 = or <2 x i32> %y, %x
+; CHECK-NEXT: ret <2 x i32> %tmp3
 
   %tmp1 = or <2 x i32> %x, %y
   %tmp2 = or <2 x i32> %y, %x
@@ -60,8 +205,9 @@ define <2 x i32> @test5(<2 x i32> %x, <2 x i32> %y) {
   ret <2 x i32> %tmp3
 }
 
-define <2 x i32> @test6(<2 x i32> %x, <2 x i32> %y) {
-; CHECK-LABEL: test6
+; FIXME: Optimize vector xor.  Currently only commute operands.
+define <2 x i32> @test17(<2 x i32> %x, <2 x i32> %y) {
+; CHECK-LABEL: test17
 ; CHECK-NEXT: %tmp1 = xor <2 x i32> %x, %y
 ; CHECK-NEXT: %tmp2 = xor <2 x i32> %x, %y
 ; CHECK-NEXT: %tmp3 = xor <2 x i32> %tmp1, %tmp2
diff --git a/test/Transforms/Reassociate/fast-basictest.ll b/test/Transforms/Reassociate/fast-basictest.ll
index 67b07f4d2699..64b74e3e8c16 100644
--- a/test/Transforms/Reassociate/fast-basictest.ll
+++ b/test/Transforms/Reassociate/fast-basictest.ll
@@ -48,9 +48,9 @@ define void @test4() {
 ; CHECK-NOT: fadd fast float
 ; CHECK: ret void
 
-  %A = load float* @fa
-  %B = load float* @fb
-  %C = load float* @fc
+  %A = load float, float* @fa
+  %B = load float, float* @fb
+  %C = load float, float* @fc
   %t1 = fadd fast float %A, %B
   %t2 = fadd fast float %t1, %C
   %t3 = fadd fast float %C, %A
@@ -69,9 +69,9 @@ define void @test5() {
 ; CHECK-NOT: fadd
 ; CHECK: ret void
 
-  %A = load float* @fa
-  %B = load float* @fb
-  %C = load float* @fc
+  %A = load float, float* @fa
+  %B = load float, float* @fb
+  %C = load float, float* @fc
   %t1 = fadd fast float %A, %B
   %t2 = fadd fast float %t1, %C
   %t3 = fadd fast float %C, %A
@@ -90,9 +90,9 @@ define void @test6() {
 ; CHECK-NOT: fadd
 ; CHECK: ret void
 
-  %A = load float* @fa
-  %B = load float* @fb
-  %C = load float* @fc
+  %A = load float, float* @fa
+  %B = load float, float* @fb
+  %C = load float, float* @fc
   %t1 = fadd fast float %B, %A
   %t2 = fadd fast float %t1, %C
   %t3 = fadd fast float %C, %A
diff --git a/test/Transforms/Reassociate/looptest.ll b/test/Transforms/Reassociate/looptest.ll
index aad3b206f69f..5563070d5245 100644
--- a/test/Transforms/Reassociate/looptest.ll
+++ b/test/Transforms/Reassociate/looptest.ll
@@ -33,8 +33,8 @@ bb4:		; preds = %bb4, %bb3
 	%reg117 = phi i32 [ %reg118, %bb4 ], [ 0, %bb3 ]		; <i32> [#uses=2]
 	%reg113 = add i32 %reg115, %reg117		; <i32> [#uses=1]
 	%reg114 = add i32 %reg113, %reg116		; <i32> [#uses=1]
-	%cast227 = getelementptr [4 x i8]* @.LC0, i64 0, i64 0		; <i8*> [#uses=1]
-	call i32 (i8*, ...)* @printf( i8* %cast227, i32 %reg114 )		; <i32>:0 [#uses=0]
+	%cast227 = getelementptr [4 x i8], [4 x i8]* @.LC0, i64 0, i64 0		; <i8*> [#uses=1]
+	call i32 (i8*, ...) @printf( i8* %cast227, i32 %reg114 )		; <i32>:0 [#uses=0]
 	%reg118 = add i32 %reg117, 1		; <i32> [#uses=2]
 	%cond224 = icmp ne i32 %reg118, %Num		; <i1> [#uses=1]
 	br i1 %cond224, label %bb4, label %bb5
diff --git a/test/Transforms/Reassociate/min_int.ll b/test/Transforms/Reassociate/min_int.ll
new file mode 100644
index 000000000000..52dab3a531f7
--- /dev/null
+++ b/test/Transforms/Reassociate/min_int.ll
@@ -0,0 +1,13 @@
+; RUN: opt < %s -reassociate -dce -S | FileCheck %s
+
+; MIN_INT cannot be negated during reassociation
+
+define i32 @minint(i32 %i) {
+; CHECK:  %mul = mul i32 %i, -2147483648
+; CHECK-NEXT:  %add = add i32 %mul, 1
+; CHECK-NEXT:  ret i32 %add
+  %mul = mul i32 %i, -2147483648
+  %add = add i32 %mul, 1
+  ret i32 %add
+}
+
diff --git a/test/Transforms/Reassociate/pr12245.ll b/test/Transforms/Reassociate/pr12245.ll
index e9b5355cceb1..0e7152e2a177 100644
--- a/test/Transforms/Reassociate/pr12245.ll
+++ b/test/Transforms/Reassociate/pr12245.ll
@@ -6,34 +6,34 @@
 
 define i32 @fn2() nounwind uwtable ssp {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %dec = add nsw i32 %0, -1
   store i32 %dec, i32* @a, align 4
-  %1 = load i32* @d, align 4
+  %1 = load i32, i32* @d, align 4
   %sub = sub nsw i32 %dec, %1
   store i32 %sub, i32* @d, align 4
-  %2 = load i32* @a, align 4
+  %2 = load i32, i32* @a, align 4
   %dec1 = add nsw i32 %2, -1
   store i32 %dec1, i32* @a, align 4
-  %3 = load i32* @d, align 4
+  %3 = load i32, i32* @d, align 4
   %sub2 = sub nsw i32 %dec1, %3
   store i32 %sub2, i32* @d, align 4
-  %4 = load i32* @a, align 4
+  %4 = load i32, i32* @a, align 4
   %dec3 = add nsw i32 %4, -1
   store i32 %dec3, i32* @a, align 4
-  %5 = load i32* @d, align 4
+  %5 = load i32, i32* @d, align 4
   %sub4 = sub nsw i32 %dec3, %5
   store i32 %sub4, i32* @d, align 4
-  %6 = load i32* @a, align 4
+  %6 = load i32, i32* @a, align 4
   %dec5 = add nsw i32 %6, -1
   store i32 %dec5, i32* @a, align 4
-  %7 = load i32* @d, align 4
+  %7 = load i32, i32* @d, align 4
   %sub6 = sub nsw i32 %dec5, %7
   store i32 %sub6, i32* @d, align 4
-  %8 = load i32* @a, align 4
+  %8 = load i32, i32* @a, align 4
   %dec7 = add nsw i32 %8, -1
   store i32 %dec7, i32* @a, align 4
-  %9 = load i32* @d, align 4
+  %9 = load i32, i32* @d, align 4
   %sub8 = sub nsw i32 %dec7, %9
   store i32 %sub8, i32* @d, align 4
   ret i32 0
diff --git a/test/Transforms/Reassociate/pr21205.ll b/test/Transforms/Reassociate/pr21205.ll
index fcc7150478eb..0c6fd3ab5693 100644
--- a/test/Transforms/Reassociate/pr21205.ll
+++ b/test/Transforms/Reassociate/pr21205.ll
@@ -11,7 +11,7 @@
 
 define i32 @test1() {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %conv = sitofp i32 %0 to float
   %sub = fsub fast float %conv, undef
   %sub1 = fadd fast float %sub, -1.000000e+00
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-1.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-1.ll
new file mode 100644
index 000000000000..1ff1abedd31c
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-1.ll
@@ -0,0 +1,28 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %merged_value base %base_phi
+
+declare void @site_for_call_safpeoint()
+
+define i64 addrspace(1)* @test(i64 addrspace(1)* %base_obj_x, i64 addrspace(1)* %base_obj_y, i1 %runtime_condition) gc "statepoint-example" {
+entry:
+  br i1 %runtime_condition, label %here, label %there
+
+here:
+  %x = getelementptr i64, i64 addrspace(1)* %base_obj_x, i32 1
+  br label %merge
+
+there:
+  %y = getelementptr i64, i64 addrspace(1)* %base_obj_y, i32 1
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK:   %base_phi = phi i64 addrspace(1)* [ %base_obj_x, %here ], [ %base_obj_y, %there ]
+  %merged_value = phi i64 addrspace(1)* [ %x, %here ], [ %y, %there ]
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @site_for_call_safpeoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %merged_value
+}
+
+declare void @foo()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-10.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-10.ll
new file mode 100644
index 000000000000..f4292a998485
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-10.ll
@@ -0,0 +1,37 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %next_x base %base_obj_x
+; CHECK: derived %next_y base %base_obj_y
+; CHECK: derived %next base %base_phi
+
+declare i1 @runtime_value()
+declare void @do_safepoint()
+
+define void @select_of_phi(i64 addrspace(1)* %base_obj_x, i64 addrspace(1)* %base_obj_y) gc "statepoint-example" {
+entry:
+  br label %loop
+
+loop:
+  %current_x = phi i64 addrspace(1)* [ %base_obj_x , %entry ], [ %next_x, %merge ]
+  %current_y = phi i64 addrspace(1)* [ %base_obj_y , %entry ], [ %next_y, %merge ]
+  %current = phi i64 addrspace(1)* [ null , %entry ], [ %next , %merge ]
+
+  %condition = call i1 @runtime_value()
+  %next_x = getelementptr i64, i64 addrspace(1)* %current_x, i32 1
+  %next_y = getelementptr i64, i64 addrspace(1)* %current_y, i32 1
+
+  br i1 %condition, label %true, label %false
+
+true:
+  br label %merge
+
+false:
+  br label %merge
+
+merge:
+  %next = phi i64 addrspace(1)* [ %next_x, %true ], [ %next_y, %false ]
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-11.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-11.ll
new file mode 100644
index 000000000000..2a7a7444adec
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-11.ll
@@ -0,0 +1,26 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers  -S 2>&1 | FileCheck %s
+
+; CHECK: derived %next base %base_obj
+
+declare void @do_safepoint()
+
+define void @test(i64 addrspace(1)* %base_obj) gc "statepoint-example" {
+entry:
+  %obj = getelementptr i64, i64 addrspace(1)* %base_obj, i32 1
+  br label %loop
+
+loop:                                             ; preds = %loop, %entry
+; CHECK-LABEL: loop:
+; CHECK: phi i64 addrspace(1)*
+; CHECK-DAG:  [ %base_obj.relocated.casted, %loop ] 
+; CHECK-DAG:  [ %base_obj, %entry ]
+; CHECK:  %current = phi i64 addrspace(1)* 
+; CHECK-DAG:  [ %obj, %entry ]
+; CHECK-DAG:  [ %next.relocated.casted, %loop ]
+  %current = phi i64 addrspace(1)* [ %obj, %entry ], [ %next, %loop ]
+  %next = getelementptr i64, i64 addrspace(1)* %current, i32 1
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-2.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-2.ll
new file mode 100644
index 000000000000..c2877d8b5393
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-2.ll
@@ -0,0 +1,21 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %merged_value base %base_obj
+
+
+define i64 addrspace(1)* @test(i64 addrspace(1)* %base_obj, i1 %runtime_condition) gc "statepoint-example" {
+entry:
+  br i1 %runtime_condition, label %merge, label %there
+
+there:
+  %derived_obj = getelementptr i64, i64 addrspace(1)* %base_obj, i32 1
+  br label %merge
+
+merge:
+  %merged_value = phi i64 addrspace(1)* [ %base_obj, %entry ], [ %derived_obj, %there ]
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %merged_value
+}
+
+declare void @foo()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-3.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-3.ll
new file mode 100644
index 000000000000..f72201b5138e
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-3.ll
@@ -0,0 +1,20 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %next.i64 base %base_obj
+
+define void @test(i64 addrspace(1)* %base_obj) gc "statepoint-example" {
+entry:
+  %obj = getelementptr i64, i64 addrspace(1)* %base_obj, i32 1
+  br label %loop
+
+loop:
+  %current = phi i64 addrspace(1)* [ %obj, %entry ], [ %next.i64, %loop ]
+  %current.i32 = bitcast i64 addrspace(1)* %current to i32 addrspace(1)*
+  %next.i32 = getelementptr i32, i32 addrspace(1)* %current.i32, i32 1
+  %next.i64 = bitcast i32 addrspace(1)* %next.i32 to i64 addrspace(1)*
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare void @do_safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-4.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-4.ll
new file mode 100644
index 000000000000..90d91d21d7b4
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-4.ll
@@ -0,0 +1,53 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+
+; CHECK: derived %obj_to_consume base %base_phi
+
+declare void @foo()
+declare i64 addrspace(1)* @generate_obj()
+declare void @consume_obj(i64 addrspace(1)*)
+
+define void @test(i32 %condition) gc "statepoint-example" {
+entry:
+  br label %loop
+
+loop:
+; CHECK: loop:
+; CHECK:  %safepoint_token1 = call i32 (i64, i32, i64 addrspace(1)* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64f(i64 0, i32 0, i64 addrspace(1)* ()* @generate_obj, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i3
+; CHECK-NEXT:  %obj2 = call i64 addrspace(1)* @llvm.experimental.gc.result
+  %safepoint_token1 = call i32 (i64, i32, i64 addrspace(1)* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i64f(i64 0, i32 0, i64 addrspace(1)* ()* @generate_obj, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  %obj2 = call i64 addrspace(1)* @llvm.experimental.gc.result.p1i64(i32 %safepoint_token1)
+  switch i32 %condition, label %dest_a [
+    i32 0, label %dest_b
+    i32 1, label %dest_c
+  ]
+
+dest_a:
+  br label %merge
+
+dest_b:
+  br label %merge
+
+dest_c:
+  br label %merge
+
+merge:
+; CHECK: merge:
+; CHECK:  %base_phi = phi i64 addrspace(1)* [ %obj2, %dest_a ], [ null, %dest_b ], [ null, %dest_c ]
+; CHECK:  %obj_to_consume = phi i64 addrspace(1)* [ %obj2, %dest_a ], [ null, %dest_b ], [ null, %dest_c ]
+
+  %obj_to_consume = phi i64 addrspace(1)* [ %obj2, %dest_a ], [ null, %dest_b ], [ null, %dest_c ]
+  %safepoint_token3 = call i32 (i64, i32, void (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidp1i64f(i64 0, i32 0, void (i64 addrspace(1)*)* @consume_obj, i32 1, i32 0, i64 addrspace(1)* %obj_to_consume, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %merge.split
+
+merge.split:                                      ; preds = %merge
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+
+; Function Attrs: nounwind
+declare i64 addrspace(1)* @llvm.experimental.gc.result.p1i64(i32) #0
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_p1i64f(i64, i32, i64 addrspace(1)* ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidp1i64f(i64, i32, void (i64 addrspace(1)*)*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-5.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-5.ll
new file mode 100644
index 000000000000..9b73377e31cb
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-5.ll
@@ -0,0 +1,30 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %merged_value base %base_phi
+
+declare void @foo()
+
+define i64 addrspace(1)* @test(i64 addrspace(1)* %base_obj_x, i64 addrspace(1)* %base_obj_y, i1 %runtime_condition) gc "statepoint-example" {
+entry:
+  br i1 %runtime_condition, label %here, label %there
+
+here:
+  br label %bump
+
+bump:
+  br label %merge
+
+there:
+  %y = getelementptr i64, i64 addrspace(1)* %base_obj_y, i32 1
+  br label %merge
+
+merge:
+; CHECK: merge:
+; CHECK:  %base_phi = phi i64 addrspace(1)* [ %base_obj_x, %bump ], [ %base_obj_y, %there ]
+; CHECK-NEXT:  %merged_value = phi i64 addrspace(1)* [ %base_obj_x, %bump ], [ %y, %there ]  
+  %merged_value = phi i64 addrspace(1)* [ %base_obj_x, %bump ], [ %y, %there ]
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %merged_value
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-6.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-6.ll
new file mode 100644
index 000000000000..0b785d72d63c
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-6.ll
@@ -0,0 +1,41 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %merged_value base %base_phi
+
+declare void @site_for_call_safpeoint()
+
+define i64 addrspace(1)* @test(i64 addrspace(1)* %base_obj_x, i64 addrspace(1)* %base_obj_y, i1 %runtime_condition_x, i1 %runtime_condition_y) gc "statepoint-example" {
+entry:
+  br i1 %runtime_condition_x, label %here, label %there
+
+here:
+ br i1 %runtime_condition_y, label %bump_here_a, label %bump_here_b
+
+bump_here_a:
+  %x_a = getelementptr i64, i64 addrspace(1)* %base_obj_x, i32 1
+  br label %merge_here
+
+bump_here_b:
+  %x_b = getelementptr i64, i64 addrspace(1)* %base_obj_x, i32 2
+  br label %merge_here
+  
+
+merge_here:
+  %x = phi i64 addrspace(1)* [ %x_a , %bump_here_a ], [ %x_b , %bump_here_b ]
+  br label %merge
+
+there:
+  %y = getelementptr i64, i64 addrspace(1)* %base_obj_y, i32 1
+  br label %merge
+
+merge:
+; CHECK: merge:
+; CHECK:  %base_phi = phi i64 addrspace(1)* [ %base_obj_x, %merge_here ], [ %base_obj_y, %there ]
+; CHECK-NEXT:  %merged_value = phi i64 addrspace(1)* [ %x, %merge_here ], [ %y, %there ]  
+  %merged_value = phi i64 addrspace(1)* [ %x, %merge_here ], [ %y, %there ]
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @site_for_call_safpeoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %merged_value
+}
+
+declare void @do_safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-7.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-7.ll
new file mode 100644
index 000000000000..467429b4d27d
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-7.ll
@@ -0,0 +1,52 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %merged_value base %base_phi
+
+declare void @site_for_call_safpeoint()
+
+define i64 addrspace(1)* @test(i64 addrspace(1)* %base_obj_x, 
+   i64 addrspace(1)* %base_obj_y, i1 %runtime_condition_x, 
+   i1 %runtime_condition_y) gc "statepoint-example" {
+entry:
+  br i1 %runtime_condition_x, label %here, label %there
+
+here:
+ br i1 %runtime_condition_y, label %bump_here_a, label %bump_here_b
+
+bump_here_a:
+  %x_a = getelementptr i64, i64 addrspace(1)* %base_obj_x, i32 1
+  br label %merge_here
+
+bump_here_b:
+  %x_b = getelementptr i64, i64 addrspace(1)* %base_obj_y, i32 2
+  br label %merge_here
+  
+
+merge_here:
+; CHECK: merge_here:
+; CHECK-DAG: %base_phi
+; CHECK-DAG: phi i64 addrspace(1)*
+; CHECK-DAG: [ %base_obj_x, %bump_here_a ]
+; CHECK-DAG: [ %base_obj_y, %bump_here_b ]
+  %x = phi i64 addrspace(1)* [ %x_a , %bump_here_a ], [ %x_b , %bump_here_b ]
+  br label %merge
+
+there:
+  %y = getelementptr i64, i64 addrspace(1)* %base_obj_y, i32 1
+  br label %merge
+
+merge:
+; CHECK: merge:
+; CHECK-DAG:  %base_phi1
+; CHECK-DAG: phi i64 addrspace(1)*
+; CHECK-DAG: %merge_here
+; CHECK-DAG: [ %base_obj_y, %there ]
+; CHECK:  %merged_value = phi i64 addrspace(1)* [ %x, %merge_here ], [ %y, %there ]  
+  %merged_value = phi i64 addrspace(1)* [ %x, %merge_here ], [ %y, %there ]
+
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @site_for_call_safpeoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %merged_value
+}
+
+declare void @do_safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-8.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-8.ll
new file mode 100644
index 000000000000..b89be7daa67b
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-8.ll
@@ -0,0 +1,38 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S 2>&1 | FileCheck %s
+
+; CHECK: derived %next_element_ptr base %array_obj
+
+define i32 @null_in_array(i64 addrspace(1)* %array_obj) gc "statepoint-example" {
+entry:
+  %array_len_pointer.i64 = getelementptr i64, i64 addrspace(1)* %array_obj, i32 1
+  %array_len_pointer.i32 = bitcast i64 addrspace(1)* %array_len_pointer.i64 to i32 addrspace(1)*
+  %array_len = load i32, i32 addrspace(1)* %array_len_pointer.i32
+  %array_elems = bitcast i32 addrspace(1)* %array_len_pointer.i32 to i64 addrspace(1)* addrspace(1)*
+  br label %loop_check
+
+loop_check:
+  %index = phi i32 [ 0, %entry ], [ %next_index, %loop_back ]
+  %current_element_ptr = phi i64 addrspace(1)* addrspace(1)* [ %array_elems, %entry ], [ %next_element_ptr, %loop_back ]
+  %index_lt = icmp ult i32 %index, %array_len
+  br i1 %index_lt, label %check_for_null, label %not_found
+
+check_for_null:
+  %current_element = load i64 addrspace(1)*, i64 addrspace(1)* addrspace(1)* %current_element_ptr
+  %is_null = icmp eq i64 addrspace(1)* %current_element, null
+  br i1 %is_null, label %found, label %loop_back
+
+loop_back:
+  %next_element_ptr = getelementptr i64 addrspace(1)*, i64 addrspace(1)* addrspace(1)* %current_element_ptr, i32 1
+  %next_index = add i32 %index, 1
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop_check
+
+not_found:
+  ret i32 -1
+
+found:
+  ret i32 %index
+}
+
+declare void @do_safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers-9.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers-9.ll
new file mode 100644
index 000000000000..848633b2a275
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers-9.ll
@@ -0,0 +1,21 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-print-base-pointers -S  2>&1 | FileCheck %s
+
+; CHECK: derived %next base %base_obj
+
+declare i1 @runtime_value()
+
+define void @maybe_GEP(i64 addrspace(1)* %base_obj) gc "statepoint-example" {
+entry:
+  br label %loop
+
+loop:
+  %current = phi i64 addrspace(1)* [ %base_obj, %entry ], [ %next, %loop ]
+  %condition = call i1 @runtime_value()
+  %maybe_next = getelementptr i64, i64 addrspace(1)* %current, i32 1
+  %next = select i1 %condition, i64 addrspace(1)* %maybe_next, i64 addrspace(1)* %current
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare void @do_safepoint()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/base-pointers.ll b/test/Transforms/RewriteStatepointsForGC/base-pointers.ll
new file mode 100644
index 000000000000..879f3f9e24da
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/base-pointers.ll
@@ -0,0 +1,100 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -S 2>&1 | FileCheck %s
+
+declare i64 addrspace(1)* @generate_obj()
+declare void @use_obj(i64 addrspace(1)*)
+
+; The rewriting needs to make %obj loop variant by inserting a phi 
+; of the original value and it's relocation.
+define void @def_use_safepoint() gc "statepoint-example" {
+; CHECK-LABEL: def_use_safepoint
+entry:
+  %obj = call i64 addrspace(1)* @generate_obj()
+  br label %loop
+
+loop:
+; CHECK: phi i64 addrspace(1)* 
+; CHECK-DAG: [ %obj.relocated.casted, %loop ]
+; CHECK-DAG: [ %obj, %entry ]
+  call void @use_obj(i64 addrspace(1)* %obj)
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare void @do_safepoint()
+
+declare void @parse_point(i64 addrspace(1)*)
+
+define i64 addrspace(1)* @test1(i32 %caller, i8 addrspace(1)* %a, i8 addrspace(1)* %b, i32 %unknown) gc "statepoint-example" {
+; CHECK-LABEL: test1
+ entry:
+  br i1 undef, label %left, label %right
+
+ left:
+  %a.cast = bitcast i8 addrspace(1)* %a to i64 addrspace(1)*
+; CHECK: left:
+; CHECK-NEXT: %a.cast = bitcast i8 addrspace(1)* %a to i64 addrspace(1)*
+; CHECK-NEXT: [[CAST_L:%.*]] = bitcast i8 addrspace(1)* %a to i64 addrspace(1)*
+
+; Our safepoint placement pass calls removeUnreachableBlocks, which does a bunch
+; of simplifications to branch instructions.  This bug is visible only when
+; there are multiple branches into the same block from the same predecessor, and
+; the following ceremony is to make that artefact survive a call to 
+; removeUnreachableBlocks.  As an example, "br i1 undef, label %merge, label %merge"
+; will get simplified to "br label %merge" by removeUnreachableBlocks.
+  switch i32 %unknown, label %right [ i32 0, label %merge
+                                      i32 1, label %merge
+                                      i32 5, label %merge
+                                      i32 3, label %right ]
+
+ right:
+  %b.cast = bitcast i8 addrspace(1)* %b to i64 addrspace(1)*
+  br label %merge
+; CHECK: right:
+; CHECK-NEXT: %b.cast = bitcast i8 addrspace(1)* %b to i64 addrspace(1)*
+; CHECK-NEXT: [[CAST_R:%.*]] = bitcast i8 addrspace(1)* %b to i64 addrspace(1)*
+
+ merge:
+; CHECK: merge:
+; CHECK-NEXT: %base_phi = phi i64 addrspace(1)* [ [[CAST_L]], %left ], [ [[CAST_L]], %left ], [ [[CAST_L]], %left ], [ [[CAST_R]], %right ], !is_base_value !0
+  %value = phi i64 addrspace(1)* [ %a.cast, %left], [ %a.cast, %left], [ %a.cast, %left], [ %b.cast, %right]
+  %safepoint_token = call i32 (i64, i32, void (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidp1i64f(i64 0, i32 0, void (i64 addrspace(1)*)* @parse_point, i32 1, i32 0, i64 addrspace(1)* %value, i32 0, i32 5, i32 0, i32 0, i32 0, i32 0, i32 0)
+
+  ret i64 addrspace(1)* %value
+}
+
+;; The purpose of this test is to ensure that when two live values share a
+;;  base defining value with inherent conflicts, we end up with a *single*
+;;  base phi/select per such node.  This is testing an optimization, not a
+;;  fundemental correctness criteria
+define void @test2(i1 %cnd, i64 addrspace(1)* %base_obj, i64 addrspace(1)* %base_arg2)  gc "statepoint-example" {
+; CHECK-LABEL: @test2
+entry:
+  %obj = getelementptr i64, i64 addrspace(1)* %base_obj, i32 1
+  br label %loop
+
+loop:                                             ; preds = %loop, %entry
+; CHECK-LABEL: loop
+; CHECK:   %base_phi = phi i64 addrspace(1)*
+; CHECK-DAG: [ %base_obj, %entry ]
+; Given the two selects are equivelent, so are their base phis - ideally,
+; we'd have commoned these, but that's a missed optimization, not correctness.
+; CHECK-DAG: [ [[DISCARD:%base_select.*.relocated.casted]], %loop ]
+; CHECK-NOT: base_phi2
+; CHECK: next = select
+; CHECK: base_select
+; CHECK: extra2 = select
+; CHECK: base_select
+; CHECK: statepoint
+;; Both 'next' and 'extra2' are live across the backedge safepoint...
+  %current = phi i64 addrspace(1)* [ %obj, %entry ], [ %next, %loop ]
+  %extra = phi i64 addrspace(1)* [ %obj, %entry ], [ %extra2, %loop ]
+  %nexta = getelementptr i64, i64 addrspace(1)* %current, i32 1
+  %next = select i1 %cnd, i64 addrspace(1)* %nexta, i64 addrspace(1)* %base_arg2
+  %extra2 = select i1 %cnd, i64 addrspace(1)* %nexta, i64 addrspace(1)* %base_arg2
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %loop
+}
+
+declare void @foo()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidp1i64f(i64, i32, void (i64 addrspace(1)*)*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/basics.ll b/test/Transforms/RewriteStatepointsForGC/basics.ll
new file mode 100644
index 000000000000..2a61924a5927
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/basics.ll
@@ -0,0 +1,88 @@
+; This is a collection of really basic tests for gc.statepoint rewriting.
+; RUN:  opt %s -rewrite-statepoints-for-gc -spp-rematerialization-threshold=0 -S | FileCheck %s
+
+declare void @foo()
+
+; Trivial relocation over a single call
+define i8 addrspace(1)* @test1(i8 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test1
+; CHECK-LABEL: entry:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i8 addrspace(1)* %obj
+}
+
+; Two safepoints in a row (i.e. consistent liveness)
+define i8 addrspace(1)* @test2(i8 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test2
+; CHECK-LABEL: entry:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated1 = call coldcc i8 addrspace(1)*
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i8 addrspace(1)* %obj
+}
+
+; A simple derived pointer
+define i8 @test3(i8 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: entry:
+; CHECK-NEXT: getelementptr
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %derived.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: load i8, i8 addrspace(1)* %derived.relocated
+; CHECK-NEXT: load i8, i8 addrspace(1)* %obj.relocated
+entry:
+  %derived = getelementptr i8, i8 addrspace(1)* %obj, i64 10
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+
+  %a = load i8, i8 addrspace(1)* %derived
+  %b = load i8, i8 addrspace(1)* %obj
+  %c = sub i8 %a, %b
+  ret i8 %c
+}
+
+; Tests to make sure we visit both the taken and untaken predeccessor 
+; of merge.  This was a bug in the dataflow liveness at one point.
+define i8 addrspace(1)* @test4(i1 %cmp, i8 addrspace(1)* %obj) gc "statepoint-example" {
+entry:
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %merge
+
+untaken:
+; CHECK-LABEL: untaken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated1 = call coldcc i8 addrspace(1)*
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: %.0 = phi i8 addrspace(1)* [ %obj.relocated, %taken ], [ %obj.relocated1, %untaken ]
+; CHECK-NEXT: ret i8 addrspace(1)* %.0
+  ret i8 addrspace(1)* %obj
+}
+
+; When run over a function which doesn't opt in, should do nothing!
+define i8 addrspace(1)* @test5(i8 addrspace(1)* %obj) gc "ocaml" {
+; CHECK-LABEL: @test5
+; CHECK-LABEL: entry:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NOT: %obj.relocated = call coldcc i8 addrspace(1)*
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i8 addrspace(1)* %obj
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
\ No newline at end of file
diff --git a/test/Transforms/RewriteStatepointsForGC/constants.ll b/test/Transforms/RewriteStatepointsForGC/constants.ll
new file mode 100644
index 000000000000..a30fdd7034a4
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/constants.ll
@@ -0,0 +1,60 @@
+; RUN: opt -S -rewrite-statepoints-for-gc %s | FileCheck %s
+
+declare void @foo()
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+
+; constants don't get relocated.
+define i8 @test() gc "statepoint-example" {
+; CHECK-LABEL: @test
+; CHECK: gc.statepoint
+; CHECK-NEXT: load i8, i8 addrspace(1)* inttoptr (i64 15 to i8 addrspace(1)*)
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  %res = load i8, i8 addrspace(1)* inttoptr (i64 15 to i8 addrspace(1)*)
+  ret i8 %res
+}
+
+
+; Mostly just here to show reasonable code test can come from.  
+define i8 @test2(i8 addrspace(1)* %p) gc "statepoint-example" {
+; CHECK-LABEL: @test2
+; CHECK: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: icmp
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  %cmp = icmp eq i8 addrspace(1)* %p, null
+  br i1 %cmp, label %taken, label %not_taken
+
+taken:
+  ret i8 0
+
+not_taken:
+  %cmp2 = icmp ne i8 addrspace(1)* %p, null
+  br i1 %cmp2, label %taken, label %dead
+
+dead:
+  ; We see that dead can't be reached, but the optimizer might not.  It's 
+  ; completely legal for it to exploit the fact that if dead executed, %p 
+  ; would have to equal null.  This can produce intermediate states which 
+  ; look like that of test above, even if arbitrary constant addresses aren't
+  ; legal in the source language
+  %addr = getelementptr i8, i8 addrspace(1)* %p, i32 15
+  %res = load i8, i8addrspace(1)* %addr
+  ret i8 %res
+}
+
+@G = addrspace(1) global i8 5
+
+; Globals don't move and thus don't get relocated
+define i8 @test3(i1 %always_true) gc "statepoint-example" {
+; CHECK-LABEL: @test3
+; CHECK: gc.statepoint
+; CHECK-NEXT: load i8, i8 addrspace(1)* @G
+entry:
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  %res = load i8, i8 addrspace(1)* @G, align 1
+  ret i8 %res
+}
+
+
diff --git a/test/Transforms/RewriteStatepointsForGC/gc_relocate_creation.ll b/test/Transforms/RewriteStatepointsForGC/gc_relocate_creation.ll
new file mode 100644
index 000000000000..355ffa634f3c
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/gc_relocate_creation.ll
@@ -0,0 +1,20 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -S 2>&1 | FileCheck %s
+; This test is to verify gc.relocate can handle pointer to vector of
+; pointers (<2 x i32 addrspace(1)*> addrspace(1)* in this case).
+; The old scheme to create a gc.relocate of <2 x i32 addrspace(1)*> addrspace(1)*
+; type will fail because llvm does not support mangling vector of pointers.
+; The new scheme will create all gc.relocate to i8 addrspace(1)* type and
+; then bitcast to the correct type.
+
+declare void @foo()
+declare void @use(...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+
+define void @test1(<2 x i32 addrspace(1)*> addrspace(1)* %obj) gc "statepoint-example" {
+entry:
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+; CHECK: %obj.relocated = call coldcc i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(i32 %safepoint_token, i32 7, i32 7)
+; CHECK-NEXT:  %obj.relocated.casted = bitcast i8 addrspace(1)* %obj.relocated to <2 x i32 addrspace(1)*> addrspace(1)*
+  call void (...) @use(<2 x i32 addrspace(1)*> addrspace(1)* %obj)
+  ret void
+}
diff --git a/test/Transforms/RewriteStatepointsForGC/live-vector.ll b/test/Transforms/RewriteStatepointsForGC/live-vector.ll
new file mode 100644
index 000000000000..c49e101abe14
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/live-vector.ll
@@ -0,0 +1,121 @@
+; Test that we can correctly handle vectors of pointers in statepoint 
+; rewriting.  Currently, we scalarize, but that's an implementation detail.
+; RUN: opt %s -rewrite-statepoints-for-gc -S | FileCheck  %s
+
+; A non-vector relocation for comparison
+define i64 addrspace(1)* @test(i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: test
+; CHECK: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: ret i64 addrspace(1)* %obj.relocated.casted
+entry:
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %obj
+}
+
+; A base vector from a argument
+define <2 x i64 addrspace(1)*> @test2(<2 x i64 addrspace(1)*> %obj) gc "statepoint-example" {
+; CHECK-LABEL: test2
+; CHECK: extractelement
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: ret <2 x i64 addrspace(1)*> %7
+entry:
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ret <2 x i64 addrspace(1)*> %obj
+}
+
+; A base vector from a load
+define <2 x i64 addrspace(1)*> @test3(<2 x i64 addrspace(1)*>* %ptr) gc "statepoint-example" {
+; CHECK-LABEL: test3
+; CHECK: load
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: ret <2 x i64 addrspace(1)*> %7
+entry:
+  %obj = load <2 x i64 addrspace(1)*>, <2 x i64 addrspace(1)*>* %ptr
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ret <2 x i64 addrspace(1)*> %obj
+}
+
+declare i32 @fake_personality_function()
+
+; When a statepoint is an invoke rather than a call
+define <2 x i64 addrspace(1)*> @test4(<2 x i64 addrspace(1)*>* %ptr) gc "statepoint-example" {
+; CHECK-LABEL: test4
+; CHECK: load
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: gc.statepoint
+entry:
+  %obj = load <2 x i64 addrspace(1)*>, <2 x i64 addrspace(1)*>* %ptr
+  invoke i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+          to label %normal_return unwind label %exceptional_return
+
+; CHECK-LABEL: normal_return:
+; CHECK: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: ret <2 x i64 addrspace(1)*> %8
+normal_return:                                    ; preds = %entry
+  ret <2 x i64 addrspace(1)*> %obj
+
+; CHECK-LABEL: exceptional_return:
+; CHECK: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: ret <2 x i64 addrspace(1)*> %14
+exceptional_return:                               ; preds = %entry
+  %landing_pad4 = landingpad { i8*, i32 } personality i32 ()* @fake_personality_function
+          cleanup
+  ret <2 x i64 addrspace(1)*> %obj
+}
+
+; Can we handle an insert element with a constant offset?  This effectively
+; tests both the equal and inequal case since we have to relocate both indices
+; in the vector.
+define <2 x i64 addrspace(1)*> @test5(i64 addrspace(1)* %p) 
+     gc "statepoint-example" {
+; CHECK-LABEL: test5
+; CHECK: insertelement
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: extractelement
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: insertelement
+; CHECK-NEXT: ret <2 x i64 addrspace(1)*> %7
+entry:
+  %vec = insertelement <2 x i64 addrspace(1)*> undef, i64 addrspace(1)* %p, i32 0
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ret <2 x i64 addrspace(1)*> %vec
+}
+
+declare void @do_safepoint()
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/RewriteStatepointsForGC/liveness-basics.ll b/test/Transforms/RewriteStatepointsForGC/liveness-basics.ll
new file mode 100644
index 000000000000..6bc4d5324494
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/liveness-basics.ll
@@ -0,0 +1,166 @@
+; A collection of liveness test cases to ensure we're reporting the
+; correct live values at statepoints
+; RUN: opt -rewrite-statepoints-for-gc -spp-rematerialization-threshold=0 -S < %s | FileCheck %s
+
+
+; Tests to make sure we consider %obj live in both the taken and untaken 
+; predeccessor of merge.
+define i64 addrspace(1)* @test1(i1 %cmp, i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test1
+entry:
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: br label %merge
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %merge
+
+untaken:
+; CHECK-LABEL: untaken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated1 = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: br label %merge
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: %.0 = phi i64 addrspace(1)* [ %obj.relocated.casted, %taken ], [ %obj.relocated1.casted, %untaken ]
+; CHECK-NEXT: ret i64 addrspace(1)* %.0
+  ret i64 addrspace(1)* %obj
+}
+
+; A local kill should not effect liveness in predecessor block
+define i64 addrspace(1)* @test2(i1 %cmp, i64 addrspace(1)** %loc) gc "statepoint-example" {
+; CHECK-LABEL: @test2
+entry:
+; CHECK-LABEL: entry:
+; CHECK-NEXT:  gc.statepoint
+; CHECK-NEXT:  br
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT:  %obj = load
+; CHECK-NEXT:  gc.statepoint
+; CHECK-NEXT:  gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT:  ret i64 addrspace(1)* %obj.relocated.casted
+
+  %obj = load i64 addrspace(1)*, i64 addrspace(1)** %loc
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %obj
+
+untaken:
+  ret i64 addrspace(1)* null
+}
+
+; A local kill should effect values live from a successor phi.  Also, we
+; should only propagate liveness from a phi to the appropriate predecessors.
+define i64 addrspace(1)* @test3(i1 %cmp, i64 addrspace(1)** %loc) gc "statepoint-example" {
+; CHECK-LABEL: @test3
+entry:
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj = load
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: br label %merge
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  %obj = load i64 addrspace(1)*, i64 addrspace(1)** %loc
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %merge
+
+untaken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: br label %merge
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %merge
+
+merge:
+  %phi = phi i64 addrspace(1)* [ %obj, %taken ], [ null, %untaken ]
+  ret i64 addrspace(1)* %phi
+}
+
+; A base pointer must be live if it is needed at a later statepoint,
+; even if the base pointer is otherwise unused.
+define i64 addrspace(1)* @test4(i1 %cmp, i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test4
+entry:
+; CHECK-LABEL: entry:
+; CHECK-NEXT:  %derived = getelementptr
+; CHECK-NEXT:  gc.statepoint
+; CHECK-NEXT:  %derived.relocated =
+; CHECK-NEXT:  bitcast 
+; CHECK-NEXT:  %obj.relocated =
+; CHECK-NEXT:  bitcast
+; CHECK-NEXT:  gc.statepoint
+; CHECK-NEXT:  %derived.relocated1 =
+; CHECK-NEXT:  bitcast 
+; Note: It's legal to relocate obj again, but not strictly needed
+; CHECK-NEXT:  %obj.relocated2 =
+; CHECK-NEXT:  bitcast
+; CHECK-NEXT:  ret i64 addrspace(1)* %derived.relocated1.casted
+; 
+  %derived = getelementptr i64, i64 addrspace(1)* %obj, i64 8
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  ret i64 addrspace(1)* %derived
+}
+
+declare void @consume(...) readonly
+
+; Make sure that a phi def visited during iteration is considered a kill.
+; Also, liveness after base pointer analysis can change based on new uses,
+; not just new defs.
+define i64 addrspace(1)* @test5(i1 %cmp, i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test5
+entry:
+  br i1 %cmp, label %taken, label %untaken
+
+taken:
+; CHECK-LABEL: taken:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: br label %merge
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %merge
+
+untaken:
+; CHECK-LABEL: untaken:
+; CHECK-NEXT: br label %merge
+  br label %merge
+
+merge:
+; CHECK-LABEL: merge:
+; CHECK-NEXT: %.0 = phi i64 addrspace(1)*
+; CHECK-NEXT: %obj2a = phi
+; CHECK-NEXT: @consume
+; CHECK-NEXT: br label %final
+  %obj2a = phi i64 addrspace(1)* [ %obj, %taken ], [null, %untaken]
+  call void (...) @consume(i64 addrspace(1)* %obj2a)
+  br label %final
+final:
+; CHECK-LABEL: final:
+; CHECK-NEXT: @consume
+; CHECK-NEXT: ret i64 addrspace(1)* %.0
+  call void (...) @consume(i64 addrspace(1)* %obj2a)
+  ret i64 addrspace(1)* %obj
+}
+
+declare void @foo()
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/RewriteStatepointsForGC/preprocess.ll b/test/Transforms/RewriteStatepointsForGC/preprocess.ll
new file mode 100644
index 000000000000..60215310af02
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/preprocess.ll
@@ -0,0 +1,65 @@
+; RUN: opt -rewrite-statepoints-for-gc -S < %s | FileCheck %s
+
+declare void @consume(...)
+
+; Test to make sure we destroy LCSSA's single entry phi nodes before
+; running liveness
+define void @test6(i64 addrspace(1)* %obj) gc "statepoint-example" {
+; CHECK-LABEL: @test6
+entry:
+  br label %next
+
+next:
+; CHECK-LABEL: next:
+; CHECK-NEXT: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: @consume(i64 addrspace(1)* %obj.relocated.casted)
+; CHECK-NEXT: @consume(i64 addrspace(1)* %obj.relocated.casted)
+  %obj2 = phi i64 addrspace(1)* [ %obj, %entry ]
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  call void (...) @consume(i64 addrspace(1)* %obj2)
+  call void (...) @consume(i64 addrspace(1)* %obj)
+  ret void
+}
+
+declare void @some_call(i64 addrspace(1)*)
+
+; Need to delete unreachable gc.statepoint call
+define void @test7() gc "statepoint-example" {
+; CHECK-LABEL: test7
+; CHECK-NOT: gc.statepoint
+  ret void
+
+unreached:
+  %obj = phi i64 addrspace(1)* [null, %unreached]
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  call void (...) @consume(i64 addrspace(1)* %obj)
+  br label %unreached
+}
+
+; Need to delete unreachable gc.statepoint invoke - tested seperately given
+; a correct implementation could only remove the instructions, not the block
+define void @test8() gc "statepoint-example" {
+; CHECK-LABEL: test8
+; CHECK-NOT: gc.statepoint
+  ret void
+
+unreached:
+  invoke i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+          to label %normal_return unwind label %exceptional_return
+
+normal_return:                                    ; preds = %entry
+  ret void
+
+exceptional_return:                               ; preds = %entry
+  %landing_pad4 = landingpad { i8*, i32 } personality i32 ()* undef
+          cleanup
+  ret void
+}
+
+declare void @foo()
+; Bound the last check-not
+; CHECK-LABEL: @foo
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/RewriteStatepointsForGC/relocate_invoke_result.ll b/test/Transforms/RewriteStatepointsForGC/relocate_invoke_result.ll
new file mode 100644
index 000000000000..9c00b53dfbb5
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/relocate_invoke_result.ll
@@ -0,0 +1,33 @@
+;; RUN: opt -rewrite-statepoints-for-gc -verify -S < %s | FileCheck %s
+
+;; This test is to verify that RewriteStatepointsForGC correctly relocates values
+;; defined by invoke instruction results. 
+
+declare i64* addrspace(1)* @non_gc_call()
+
+declare void @gc_call()
+
+declare i32* @fake_personality_function()
+
+; Function Attrs: nounwind
+define i64* addrspace(1)* @test() gc "statepoint-example" {
+entry:
+  %obj = invoke i64* addrspace(1)* @non_gc_call()
+          to label %normal_dest unwind label %unwind_dest
+
+unwind_dest: 
+  %lpad = landingpad { i8*, i32 } personality i32* ()* @fake_personality_function
+          cleanup
+  resume { i8*, i32 } undef
+
+normal_dest:
+;; CHECK-LABEL: normal_dest:
+;; CHECK-NEXT: gc.statepoint
+;; CHECK-NEXT: %obj.relocated = call coldcc i8 addrspace(1)*
+;; CHECK-NEXT: bitcast
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @gc_call, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret i64* addrspace(1)* %obj
+}
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+
diff --git a/test/Transforms/RewriteStatepointsForGC/relocation.ll b/test/Transforms/RewriteStatepointsForGC/relocation.ll
new file mode 100644
index 000000000000..d7a84e5820c8
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/relocation.ll
@@ -0,0 +1,295 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -spp-rematerialization-threshold=0 -S 2>&1 | FileCheck %s
+
+
+declare void @foo()
+declare void @use(...)
+
+define i64 addrspace(1)* @test1(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj2, i1 %condition) gc "statepoint-example" {
+entry:
+; CHECK-LABEL: @test1
+; CHECK-DAG: %obj.relocated
+; CHECK-DAG: %obj2.relocated
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %joint
+
+joint:
+; CHECK-LABEL: joint:
+; CHECK: %phi1 = phi i64 addrspace(1)* [ %obj.relocated.casted, %entry ], [ %obj3, %joint2 ]
+  %phi1 = phi i64 addrspace(1)* [ %obj, %entry ], [ %obj3, %joint2 ]
+  br i1 %condition, label %use, label %joint2
+
+use:
+  br label %joint2
+
+joint2:
+; CHECK-LABEL: joint2:
+; CHECK: %phi2 = phi i64 addrspace(1)* [ %obj.relocated.casted, %use ], [ %obj2.relocated.casted, %joint ]
+; CHECK: %obj3 = getelementptr i64, i64 addrspace(1)* %obj2.relocated.casted, i32 1
+  %phi2 = phi i64 addrspace(1)* [ %obj, %use ], [ %obj2, %joint ]
+  %obj3 = getelementptr i64, i64 addrspace(1)* %obj2, i32 1
+  br label %joint
+}
+
+declare i64 addrspace(1)* @generate_obj()
+
+declare void @consume_obj(i64 addrspace(1)*)
+
+declare i1 @rt()
+
+define void @test2() gc "statepoint-example" {
+; CHECK-LABEL: @test2
+entry:
+  %obj_init = call i64 addrspace(1)* @generate_obj()
+  %obj = getelementptr i64, i64 addrspace(1)* %obj_init, i32 42
+  br label %loop
+
+loop:
+; CHECK: loop:
+; CHECK-DAG: [ %obj_init.relocated.casted, %loop.backedge ]
+; CHECK-DAG: [ %obj_init, %entry ]
+; CHECK-DAG: [ %obj.relocated.casted, %loop.backedge ]
+; CHECK-DAG: [ %obj, %entry ]
+  %index = phi i32 [ 0, %entry ], [ %index.inc, %loop.backedge ]
+; CHECK-NOT: %location = getelementptr i64, i64 addrspace(1)* %obj, i32 %index
+  %location = getelementptr i64, i64 addrspace(1)* %obj, i32 %index
+  call void @consume_obj(i64 addrspace(1)* %location)
+  %index.inc = add i32 %index, 1
+  %condition = call i1 @rt()
+  br i1 %condition, label %loop_x, label %loop_y
+
+loop_x:
+  br label %loop.backedge
+
+loop.backedge:
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  br label %loop
+
+loop_y:
+  br label %loop.backedge
+}
+
+declare void @some_call(i8 addrspace(1)*)
+
+define void @relocate_merge(i1 %cnd, i8 addrspace(1)* %arg) gc "statepoint-example" {
+; CHECK-LABEL: @relocate_merge
+bci_0:
+  br i1 %cnd, label %if_branch, label %else_branch
+
+if_branch:
+; CHECK-LABEL: if_branch:
+; CHECK: gc.statepoint
+; CHECK: gc.relocate
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %join
+
+else_branch:
+; CHECK-LABEL: else_branch:
+; CHECK: gc.statepoint
+; CHECK: gc.relocate
+  %safepoint_token1 = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 0)
+  br label %join
+
+join:
+; We need to end up with a single relocation phi updated from both paths 
+; CHECK-LABEL: join:
+; CHECK: phi i8 addrspace(1)*
+; CHECK-DAG: [ %arg.relocated, %if_branch ]
+; CHECK-DAG: [ %arg.relocated4, %else_branch ]
+; CHECK-NOT: phi
+  call void (i8 addrspace(1)*) @some_call(i8 addrspace(1)* %arg)
+  ret void
+}
+
+; Make sure a use in a statepoint gets properly relocated at a previous one.  
+; This is basically just making sure that statepoints aren't accidentally 
+; treated specially.
+define void @test3(i64 addrspace(1)* %obj) gc "statepoint-example" {
+entry:
+; CHECK-LABEL: @test3
+; CHECK: gc.statepoint
+; CHECK-NEXT: gc.relocate
+; CHECK-NEXT: bitcast
+; CHECK-NEXT: gc.statepoint
+  %safepoint_token = call i32 (i64, i32, void (i64)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidi64f(i64 0, i32 0, void (i64)* undef, i32 1, i32 0, i64 undef, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  %safepoint_token1 = call i32 (i64, i32, i32 (i64 addrspace(1)*)*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_i32p1i64f(i64 0, i32 0, i32 (i64 addrspace(1)*)* undef, i32 1, i32 0, i64 addrspace(1)* %obj, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  ret void
+}
+
+; Check specifically for the case where the result of a statepoint needs to 
+; be relocated itself
+define void @test4() gc "statepoint-example" {
+; CHECK-LABEL: @test4
+; CHECK: gc.statepoint
+; CHECK: gc.result
+; CHECK: gc.statepoint
+; CHECK: gc.relocate
+; CHECK: @use(i8 addrspace(1)* %res.relocated)
+  %safepoint_token2 = tail call i32 (i64, i32, i8 addrspace(1)* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i8f(i64 0, i32 0, i8 addrspace(1)* ()* undef, i32 0, i32 0, i32 0, i32 0)
+  %res = call i8 addrspace(1)* @llvm.experimental.gc.result.ptr.p1i8(i32 %safepoint_token2)
+  call i32 (i64, i32, i8 addrspace(1)* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i8f(i64 0, i32 0, i8 addrspace(1)* ()* undef, i32 0, i32 0, i32 0, i32 0)
+  call void (...) @use(i8 addrspace(1)* %res)
+  unreachable
+}
+
+
+; Test updating a phi where not all inputs are live to begin with
+define void @test5(i8 addrspace(1)* %arg) gc "statepoint-example" {
+; CHECK-LABEL: test5
+entry:
+  call i32 (i64, i32, i8 addrspace(1)* ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_p1i8f(i64 0, i32 0, i8 addrspace(1)* ()* undef, i32 0, i32 0, i32 0, i32 0)
+  switch i32 undef, label %kill [
+    i32 10, label %merge
+    i32 13, label %merge
+  ]
+
+kill:
+  br label %merge
+
+merge:
+; CHECK: merge:
+; CHECK: %test = phi i8 addrspace(1)
+; CHECK-DAG: [ null, %kill ]
+; CHECK-DAG: [ %arg.relocated, %entry ]
+; CHECK-DAG: [ %arg.relocated, %entry ]
+  %test = phi i8 addrspace(1)* [ null, %kill ], [ %arg, %entry ], [ %arg, %entry ]
+  call void (...) @use(i8 addrspace(1)* %test)
+  unreachable
+}
+
+
+; Check to make sure we handle values live over an entry statepoint
+define void @test6(i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2, 
+                  i8 addrspace(1)* %arg3) gc "statepoint-example" {
+; CHECK-LABEL: @test6
+entry:
+  br i1 undef, label %gc.safepoint_poll.exit2, label %do_safepoint
+
+do_safepoint:
+; CHECK-LABEL: do_safepoint:
+; CHECK: gc.statepoint
+; CHECK: arg1.relocated = 
+; CHECK: arg2.relocated = 
+; CHECK: arg3.relocated = 
+  call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 3, i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2, i8 addrspace(1)* %arg3)
+  br label %gc.safepoint_poll.exit2
+
+gc.safepoint_poll.exit2:
+; CHECK-LABEL: gc.safepoint_poll.exit2:
+; CHECK: phi i8 addrspace(1)*
+; CHECK-DAG: [ %arg3, %entry ]
+; CHECK-DAG: [ %arg3.relocated, %do_safepoint ]
+; CHECK: phi i8 addrspace(1)*
+; CHECK-DAG: [ %arg2, %entry ]
+; CHECK-DAG: [ %arg2.relocated, %do_safepoint ]
+; CHECK: phi i8 addrspace(1)*
+; CHECK-DAG: [ %arg1, %entry ]
+; CHECK-DAG:  [ %arg1.relocated, %do_safepoint ]
+  call void (...) @use(i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2, i8 addrspace(1)* %arg3)
+  ret void
+}
+
+; Check relocation in a loop nest where a relocation happens in the outer
+; but not the inner loop
+define void @test_outer_loop(i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2, 
+                  i1 %cmp) gc "statepoint-example" {
+; CHECK-LABEL: @test_outer_loop
+bci_0:
+  br label %outer-loop
+
+outer-loop:
+; CHECK-LABEL: outer-loop:
+; CHECK: phi i8 addrspace(1)* [ %arg2, %bci_0 ], [ %arg2.relocated, %outer-inc ]
+; CHECK: phi i8 addrspace(1)* [ %arg1, %bci_0 ], [ %arg1.relocated, %outer-inc ]
+  br label %inner-loop
+
+inner-loop:
+  br i1 %cmp, label %inner-loop, label %outer-inc
+
+outer-inc:
+; CHECK-LABEL: outer-inc:
+; CHECK: %arg1.relocated
+; CHECK: %arg2.relocated
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 2, i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2)
+  br label %outer-loop
+}
+
+; Check that both inner and outer loops get phis when relocation is in
+;  inner loop
+define void @test_inner_loop(i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2, 
+                  i1 %cmp) gc "statepoint-example" {
+; CHECK-LABEL: @test_inner_loop
+bci_0:
+  br label %outer-loop
+
+outer-loop:
+; CHECK-LABEL: outer-loop:
+; CHECK: phi i8 addrspace(1)* [ %arg2, %bci_0 ], [ %arg2.relocated, %outer-inc ]
+; CHECK: phi i8 addrspace(1)* [ %arg1, %bci_0 ], [ %arg1.relocated, %outer-inc ]
+  br label %inner-loop
+
+inner-loop:
+; CHECK-LABEL: inner-loop
+; CHECK: phi i8 addrspace(1)* 
+; CHECK-DAG: %outer-loop ]
+; CHECK-DAG: [ %arg2.relocated, %inner-loop ]
+; CHECKL phi i8 addrspace(1)* 
+; CHECK-DAG: %outer-loop ]
+; CHECK-DAG: [ %arg1.relocated, %inner-loop ]
+; CHECK: gc.statepoint
+; CHECK: %arg1.relocated
+; CHECK: %arg2.relocated
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 2, i8 addrspace(1)* %arg1, i8 addrspace(1)* %arg2)
+  br i1 %cmp, label %inner-loop, label %outer-inc
+
+outer-inc:
+; CHECK-LABEL: outer-inc:
+  br label %outer-loop
+}
+
+
+; This test shows why updating just those uses of the original value being
+; relocated dominated by the inserted relocation is not always sufficient.
+define i64 addrspace(1)* @test7(i64 addrspace(1)* %obj, i64 addrspace(1)* %obj2, i1 %condition) gc "statepoint-example" {
+; CHECK-LABEL: @test7
+entry:
+  br i1 %condition, label %branch2, label %join
+
+branch2:
+  br i1 %condition, label %callbb, label %join2
+
+callbb:
+  %safepoint_token = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @foo, i32 0, i32 0, i32 0, i32 5, i32 0, i32 -1, i32 0, i32 0, i32 0)
+  br label %join
+
+join:
+; CHECK-LABEL: join:
+; CHECK: phi i64 addrspace(1)* [ %obj.relocated.casted, %callbb ], [ %obj, %entry ]
+; CHECK: phi i64 addrspace(1)* 
+; CHECK-DAG: [ %obj, %entry ]
+; CHECK-DAG: [ %obj2.relocated.casted, %callbb ]
+  ; This is a phi outside the dominator region of the new defs inserted by
+  ; the safepoint, BUT we can't stop the search here or we miss the second
+  ; phi below.
+  %phi1 = phi i64 addrspace(1)* [ %obj, %entry ], [ %obj2, %callbb ]
+  br label %join2
+
+join2:
+; CHECK-LABEL: join2:
+; CHECK: phi2 = phi i64 addrspace(1)* 
+; CHECK-DAG: %join ] 
+; CHECK-DAG:  [ %obj2, %branch2 ]
+  %phi2 = phi i64 addrspace(1)* [ %obj, %join ], [ %obj2, %branch2 ]
+  ret i64 addrspace(1)* %phi2
+}
+
+
+declare void @do_safepoint()
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_p1i8f(i64, i32, i8 addrspace(1)* ()*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidi64f(i64, i32, void (i64)*, i32, i32, ...)
+declare i32 @llvm.experimental.gc.statepoint.p0f_i32p1i64f(i64, i32, i32 (i64 addrspace(1)*)*, i32, i32, ...)
+declare i8 addrspace(1)* @llvm.experimental.gc.result.ptr.p1i8(i32) #3
+
+
+
diff --git a/test/Transforms/RewriteStatepointsForGC/rematerialize-derived-pointers.ll b/test/Transforms/RewriteStatepointsForGC/rematerialize-derived-pointers.ll
new file mode 100644
index 000000000000..dc52bd88776b
--- /dev/null
+++ b/test/Transforms/RewriteStatepointsForGC/rematerialize-derived-pointers.ll
@@ -0,0 +1,222 @@
+; RUN: opt %s -rewrite-statepoints-for-gc -S 2>&1 | FileCheck %s
+
+declare void @use_obj16(i16 addrspace(1)*)
+declare void @use_obj32(i32 addrspace(1)*)
+declare void @use_obj64(i64 addrspace(1)*)
+declare void @do_safepoint()
+
+define void @"test_gep_const"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_gep_const
+entry:
+  %ptr = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr i32, i32 addrspace(1)* %base, i32 15
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: %base.relocated = call coldcc i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(i32 %sp, i32 7, i32 7)
+  ; CHECK: bitcast i8 addrspace(1)* %base.relocated to i32 addrspace(1)*
+  ; CHECK: getelementptr i32, i32 addrspace(1)* %base.relocated.casted, i32 15
+  call void @use_obj32(i32 addrspace(1)* %base)
+  call void @use_obj32(i32 addrspace(1)* %ptr)
+  ret void
+}
+
+define void @"test_gep_idx"(i32 addrspace(1)* %base, i32 %idx) gc "statepoint-example" {
+; CHECK-LABEL: test_gep_idx
+entry:
+  %ptr = getelementptr i32, i32 addrspace(1)* %base, i32 %idx
+  ; CHECK: getelementptr
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: %base.relocated = call coldcc i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(i32 %sp, i32 7, i32 7)
+  ; CHECK: %base.relocated.casted = bitcast i8 addrspace(1)* %base.relocated to i32 addrspace(1)*
+  ; CHECK: getelementptr i32, i32 addrspace(1)* %base.relocated.casted, i32 %idx
+  call void @use_obj32(i32 addrspace(1)* %base)
+  call void @use_obj32(i32 addrspace(1)* %ptr)
+  ret void
+}
+
+define void @"test_bitcast"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_bitcast
+entry:
+  %ptr = bitcast i32 addrspace(1)* %base to i64 addrspace(1)*
+  ; CHECK: bitcast i32 addrspace(1)* %base to i64 addrspace(1)*
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: %base.relocated = call coldcc i8 addrspace(1)* @llvm.experimental.gc.relocate.p1i8(i32 %sp, i32 7, i32 7)
+  ; CHECK: %base.relocated.casted = bitcast i8 addrspace(1)* %base.relocated to i32 addrspace(1)*
+  ; CHECK: bitcast i32 addrspace(1)* %base.relocated.casted to i64 addrspace(1)*
+  call void @use_obj32(i32 addrspace(1)* %base)
+  call void @use_obj64(i64 addrspace(1)* %ptr)
+  ret void
+}
+
+define void @"test_bitcast_gep"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_bitcast_gep
+entry:
+  %ptr.gep = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr
+  %ptr.cast = bitcast i32 addrspace(1)* %ptr.gep to i64 addrspace(1)*
+  ; CHECK: bitcast
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  call void @use_obj32(i32 addrspace(1)* %base)
+  call void @use_obj64(i64 addrspace(1)* %ptr.cast)
+  ret void
+}
+
+define void @"test_intersecting_chains"(i32 addrspace(1)* %base, i32 %idx) gc "statepoint-example" {
+; CHECK-LABEL: test_intersecting_chains
+entry:
+  %ptr.gep = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr
+  %ptr.cast = bitcast i32 addrspace(1)* %ptr.gep to i64 addrspace(1)*
+  ; CHECK: bitcast
+  %ptr.cast2 = bitcast i32 addrspace(1)* %ptr.gep to i16 addrspace(1)*
+  ; CHECK: bitcast
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  call void @use_obj64(i64 addrspace(1)* %ptr.cast)
+  call void @use_obj16(i16 addrspace(1)* %ptr.cast2)
+  ret void
+}
+
+define void @"test_cost_threshold"(i32 addrspace(1)* %base, i32 %idx1, i32 %idx2, i32 %idx3) gc "statepoint-example" {
+; CHECK-LABEL: test_cost_threshold
+entry:
+  %ptr.gep = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr
+  %ptr.gep2 = getelementptr i32, i32 addrspace(1)* %ptr.gep, i32 %idx1
+  ; CHECK: getelementptr
+  %ptr.gep3 = getelementptr i32, i32 addrspace(1)* %ptr.gep2, i32 %idx2
+  ; CHECK: getelementptr
+  %ptr.gep4 = getelementptr i32, i32 addrspace(1)* %ptr.gep3, i32 %idx3
+  ; CHECK: getelementptr
+  %ptr.cast = bitcast i32 addrspace(1)* %ptr.gep4 to i64 addrspace(1)*
+  ; CHECK: bitcast
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  call void @use_obj64(i64 addrspace(1)* %ptr.cast)
+  ret void
+}
+
+define void @"test_two_derived"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_two_derived
+entry:
+  %ptr = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  %ptr2 = getelementptr i32, i32 addrspace(1)* %base, i32 12
+  ; CHECK: getelementptr
+  ; CHECK: getelementptr
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: getelementptr
+  call void @use_obj32(i32 addrspace(1)* %ptr)
+  call void @use_obj32(i32 addrspace(1)* %ptr2)
+  ret void
+}
+
+define void @"test_gep_smallint_array"([3 x i32] addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_gep_smallint_array
+entry:
+  %ptr = getelementptr [3 x i32], [3 x i32] addrspace(1)* %base, i32 0, i32 2
+  ; CHECK: getelementptr
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  call void @use_obj32(i32 addrspace(1)* %ptr)
+  ret void
+}
+
+declare i32 @fake_personality_function()
+
+define void @"test_invoke"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_invoke
+entry:
+  %ptr.gep = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr
+  %ptr.cast = bitcast i32 addrspace(1)* %ptr.gep to i64 addrspace(1)*
+  ; CHECK: bitcast
+  %ptr.cast2 = bitcast i32 addrspace(1)* %ptr.gep to i16 addrspace(1)*
+  ; CHECK: bitcast
+  %sp = invoke i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+                to label %normal unwind label %exception
+
+normal:
+  ; CHECK-LABEL: normal:
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  call void @use_obj64(i64 addrspace(1)* %ptr.cast)
+  call void @use_obj16(i16 addrspace(1)* %ptr.cast2)
+  ret void
+
+exception:
+  ; CHECK-LABEL: exception:
+  %landing_pad4 = landingpad { i8*, i32 } personality i32 ()* @fake_personality_function
+          cleanup
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  ; CHECK: bitcast
+  call void @use_obj64(i64 addrspace(1)* %ptr.cast)
+  call void @use_obj16(i16 addrspace(1)* %ptr.cast2)
+  ret void
+}
+
+define void @"test_loop"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_loop
+entry:
+  %ptr.gep = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  ; CHECK: getelementptr
+  br label %loop
+
+loop:
+  ; CHECK: phi i32 addrspace(1)* [ %ptr.gep, %entry ], [ %ptr.gep.remat, %loop ]
+  ; CHECK: phi i32 addrspace(1)* [ %base, %entry ], [ %base.relocated.casted, %loop ]
+  call void @use_obj32(i32 addrspace(1)* %ptr.gep)
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: getelementptr
+  br label %loop
+}
+
+define void @"test_too_long"(i32 addrspace(1)* %base) gc "statepoint-example" {
+; CHECK-LABEL: test_too_long
+entry:
+  %ptr.gep   = getelementptr i32, i32 addrspace(1)* %base, i32 15
+  %ptr.gep1  = getelementptr i32, i32 addrspace(1)* %ptr.gep, i32 15
+  %ptr.gep2  = getelementptr i32, i32 addrspace(1)* %ptr.gep1, i32 15
+  %ptr.gep3  = getelementptr i32, i32 addrspace(1)* %ptr.gep2, i32 15
+  %ptr.gep4  = getelementptr i32, i32 addrspace(1)* %ptr.gep3, i32 15
+  %ptr.gep5  = getelementptr i32, i32 addrspace(1)* %ptr.gep4, i32 15
+  %ptr.gep6  = getelementptr i32, i32 addrspace(1)* %ptr.gep5, i32 15
+  %ptr.gep7  = getelementptr i32, i32 addrspace(1)* %ptr.gep6, i32 15
+  %ptr.gep8  = getelementptr i32, i32 addrspace(1)* %ptr.gep7, i32 15
+  %ptr.gep9  = getelementptr i32, i32 addrspace(1)* %ptr.gep8, i32 15
+  %ptr.gep10 = getelementptr i32, i32 addrspace(1)* %ptr.gep9, i32 15
+  %ptr.gep11 = getelementptr i32, i32 addrspace(1)* %ptr.gep10, i32 15
+  %sp = call i32 (i64, i32, void ()*, i32, i32, ...) @llvm.experimental.gc.statepoint.p0f_isVoidf(i64 0, i32 0, void ()* @do_safepoint, i32 0, i32 0, i32 0, i32 0)
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  ; CHECK: gc.relocate
+  ; CHECK: bitcast
+  call void @use_obj32(i32 addrspace(1)* %ptr.gep11)
+  ret void
+}
+
+
+declare i32 @llvm.experimental.gc.statepoint.p0f_isVoidf(i64, i32, void ()*, i32, i32, ...)
diff --git a/test/Transforms/SCCP/2002-08-30-GetElementPtrTest.ll b/test/Transforms/SCCP/2002-08-30-GetElementPtrTest.ll
index 6aaf33ec206c..7b88a669ffba 100644
--- a/test/Transforms/SCCP/2002-08-30-GetElementPtrTest.ll
+++ b/test/Transforms/SCCP/2002-08-30-GetElementPtrTest.ll
@@ -3,7 +3,7 @@
 @G = external global [40 x i32]		; <[40 x i32]*> [#uses=1]
 
 define i32* @test() {
-	%X = getelementptr [40 x i32]* @G, i64 0, i64 0		; <i32*> [#uses=1]
+	%X = getelementptr [40 x i32], [40 x i32]* @G, i64 0, i64 0		; <i32*> [#uses=1]
 	ret i32* %X
 }
 
diff --git a/test/Transforms/SCCP/2003-06-24-OverdefinedPHIValue.ll b/test/Transforms/SCCP/2003-06-24-OverdefinedPHIValue.ll
index 576f5d6504d1..e5a1d679fc29 100644
--- a/test/Transforms/SCCP/2003-06-24-OverdefinedPHIValue.ll
+++ b/test/Transforms/SCCP/2003-06-24-OverdefinedPHIValue.ll
@@ -3,7 +3,7 @@
 
 define void @cprop_test11(i32* %data.1) {
 entry:
-	%tmp.1 = load i32* %data.1		; <i32> [#uses=3]
+	%tmp.1 = load i32, i32* %data.1		; <i32> [#uses=3]
 	%tmp.41 = icmp sgt i32 %tmp.1, 1		; <i1> [#uses=1]
 	br i1 %tmp.41, label %no_exit, label %loopexit
 no_exit:		; preds = %endif, %then, %entry
@@ -22,9 +22,9 @@ endif:		; preds = %no_exit
 loopexit:		; preds = %endif, %then, %entry
 	%j.1 = phi i32 [ 1, %entry ], [ %j.0, %endif ], [ %i.0, %then ]		; <i32> [#uses=1]
 	%i.1 = phi i32 [ 1, %entry ], [ %inc, %endif ], [ %inc1, %then ]		; <i32> [#uses=1]
-	%tmp.17 = getelementptr i32* %data.1, i64 1		; <i32*> [#uses=1]
+	%tmp.17 = getelementptr i32, i32* %data.1, i64 1		; <i32*> [#uses=1]
 	store i32 %j.1, i32* %tmp.17
-	%tmp.23 = getelementptr i32* %data.1, i64 2		; <i32*> [#uses=1]
+	%tmp.23 = getelementptr i32, i32* %data.1, i64 2		; <i32*> [#uses=1]
 	store i32 %i.1, i32* %tmp.23
 	ret void
 }
diff --git a/test/Transforms/SCCP/2006-10-23-IPSCCP-Crash.ll b/test/Transforms/SCCP/2006-10-23-IPSCCP-Crash.ll
index 47f9cb45cc67..9724e6ebc24e 100644
--- a/test/Transforms/SCCP/2006-10-23-IPSCCP-Crash.ll
+++ b/test/Transforms/SCCP/2006-10-23-IPSCCP-Crash.ll
@@ -37,7 +37,7 @@ cond_next50:		; preds = %entry
 	%tmp52 = icmp sgt i32 %D, 0		; <i1> [#uses=1]
 	br i1 %tmp52, label %cond_true53, label %cond_next71
 cond_true53:		; preds = %cond_next50
-	%tmp54 = load i32* @JUMP		; <i32> [#uses=1]
+	%tmp54 = load i32, i32* @JUMP		; <i32> [#uses=1]
 	%tmp55 = icmp eq i32 %tmp54, 1		; <i1> [#uses=1]
 	br i1 %tmp55, label %cond_true56, label %cond_next63
 cond_true56:		; preds = %cond_true53
@@ -67,9 +67,9 @@ bb217:		; preds = %cond_true260
 	ret void
 cond_next252:		; preds = %cond_next208, %entry
 	%D.0.0 = phi i32 [ 0, %entry ], [ %tmp229, %cond_next208 ]		; <i32> [#uses=1]
-	%tmp254 = getelementptr i8** null, i32 1		; <i8**> [#uses=1]
-	%tmp256 = load i8** %tmp254		; <i8*> [#uses=1]
-	%tmp258 = load i8* %tmp256		; <i8> [#uses=1]
+	%tmp254 = getelementptr i8*, i8** null, i32 1		; <i8**> [#uses=1]
+	%tmp256 = load i8*, i8** %tmp254		; <i8*> [#uses=1]
+	%tmp258 = load i8, i8* %tmp256		; <i8> [#uses=1]
 	%tmp259 = icmp eq i8 %tmp258, 45		; <i1> [#uses=1]
 	br i1 %tmp259, label %cond_true260, label %bb263
 cond_true260:		; preds = %cond_next252
diff --git a/test/Transforms/SCCP/2006-12-04-PackedType.ll b/test/Transforms/SCCP/2006-12-04-PackedType.ll
index cee334987218..05cf5fffd2b6 100644
--- a/test/Transforms/SCCP/2006-12-04-PackedType.ll
+++ b/test/Transforms/SCCP/2006-12-04-PackedType.ll
@@ -103,14 +103,14 @@ target triple = "powerpc-apple-darwin8"
 
 define void @gldLLVMVecPointRender(%struct.GLDContextRec* %ctx) {
 entry:
-	%tmp.uip = getelementptr %struct.GLDContextRec* %ctx, i32 0, i32 22		; <i32*> [#uses=1]
-	%tmp = load i32* %tmp.uip		; <i32> [#uses=3]
+	%tmp.uip = getelementptr %struct.GLDContextRec, %struct.GLDContextRec* %ctx, i32 0, i32 22		; <i32*> [#uses=1]
+	%tmp = load i32, i32* %tmp.uip		; <i32> [#uses=3]
 	%tmp91 = lshr i32 %tmp, 5		; <i32> [#uses=1]
 	%tmp92 = trunc i32 %tmp91 to i1		; <i1> [#uses=1]
 	br i1 %tmp92, label %cond_true93, label %cond_next116
 cond_true93:		; preds = %entry
-	%tmp.upgrd.1 = getelementptr %struct.GLDContextRec* %ctx, i32 0, i32 31, i32 14		; <i32*> [#uses=1]
-	%tmp95 = load i32* %tmp.upgrd.1		; <i32> [#uses=1]
+	%tmp.upgrd.1 = getelementptr %struct.GLDContextRec, %struct.GLDContextRec* %ctx, i32 0, i32 31, i32 14		; <i32*> [#uses=1]
+	%tmp95 = load i32, i32* %tmp.upgrd.1		; <i32> [#uses=1]
 	%tmp95.upgrd.2 = sitofp i32 %tmp95 to float		; <float> [#uses=1]
 	%tmp108 = fmul float undef, %tmp95.upgrd.2		; <float> [#uses=1]
 	br label %cond_next116
diff --git a/test/Transforms/SCCP/2009-05-27-VectorOperandZero.ll b/test/Transforms/SCCP/2009-05-27-VectorOperandZero.ll
index 7aced6635272..1ac6bcd8b4ec 100644
--- a/test/Transforms/SCCP/2009-05-27-VectorOperandZero.ll
+++ b/test/Transforms/SCCP/2009-05-27-VectorOperandZero.ll
@@ -3,7 +3,7 @@
 
 define i32 @main() nounwind {
 entry:
-	%0 = tail call signext i8 (...)* @sin() nounwind
+	%0 = tail call signext i8 (...) @sin() nounwind
 	ret i32 0
 }
 
diff --git a/test/Transforms/SCCP/apint-array.ll b/test/Transforms/SCCP/apint-array.ll
index 888b9e1d22d5..eff6cc997e35 100644
--- a/test/Transforms/SCCP/apint-array.ll
+++ b/test/Transforms/SCCP/apint-array.ll
@@ -6,16 +6,16 @@
 define i101 @array()
 {
 Head:
-   %A = getelementptr [6 x i101]* @Y, i32 0, i32 1
+   %A = getelementptr [6 x i101], [6 x i101]* @Y, i32 0, i32 1
 
-   %B = load i101* %A
+   %B = load i101, i101* %A
    %C = icmp sge i101 %B, 1
    br i1 %C, label %True, label %False
 True:
    %D = and i101 %B, 1
    %E = trunc i101 %D to i32
-   %F = getelementptr [6 x i101]* @Y, i32 0, i32 %E
-   %G = load i101* %F
+   %F = getelementptr [6 x i101], [6 x i101]* @Y, i32 0, i32 %E
+   %G = load i101, i101* %F
    br label %False
 False:
    %H = phi i101 [%G, %True], [-1, %Head]
diff --git a/test/Transforms/SCCP/apint-bigarray.ll b/test/Transforms/SCCP/apint-bigarray.ll
index 0dd9ad331a07..e0231997f1bf 100644
--- a/test/Transforms/SCCP/apint-bigarray.ll
+++ b/test/Transforms/SCCP/apint-bigarray.ll
@@ -3,7 +3,7 @@
 @G =  global [1000000 x i10000] zeroinitializer
 
 define internal i10000* @test(i10000 %Arg) {
-	%X = getelementptr [1000000 x i10000]* @G, i32 0, i32 999
+	%X = getelementptr [1000000 x i10000], [1000000 x i10000]* @G, i32 0, i32 999
         store i10000 %Arg, i10000* %X
 	ret i10000* %X
 }
@@ -11,13 +11,13 @@ define internal i10000* @test(i10000 %Arg) {
 define i10000 @caller()
 {
         %Y = call i10000* @test(i10000 -1)
-        %Z = load i10000* %Y
+        %Z = load i10000, i10000* %Y
         ret i10000 %Z 
 }
 
 define i10000 @caller2()
 {
         %Y = call i10000* @test(i10000 1)
-        %Z = load i10000* %Y
+        %Z = load i10000, i10000* %Y
         ret i10000 %Z 
 }
diff --git a/test/Transforms/SCCP/apint-bigint2.ll b/test/Transforms/SCCP/apint-bigint2.ll
index 660eaad7bc12..f28b966aceba 100644
--- a/test/Transforms/SCCP/apint-bigint2.ll
+++ b/test/Transforms/SCCP/apint-bigint2.ll
@@ -6,13 +6,13 @@
 define i101 @array()
 {
 Head:
-   %A = getelementptr [6 x i101]* @Y, i32 0, i32 1
-   %B = load i101* %A
+   %A = getelementptr [6 x i101], [6 x i101]* @Y, i32 0, i32 1
+   %B = load i101, i101* %A
    %D = and i101 %B, 1
    %DD = or i101 %D, 1
    %E = trunc i101 %DD to i32
-   %F = getelementptr [6 x i101]* @Y, i32 0, i32 %E
-   %G = load i101* %F
+   %F = getelementptr [6 x i101], [6 x i101]* @Y, i32 0, i32 %E
+   %G = load i101, i101* %F
  
    ret i101 %G
 }
diff --git a/test/Transforms/SCCP/apint-ipsccp3.ll b/test/Transforms/SCCP/apint-ipsccp3.ll
index 68987aee249a..c99ae5820b2d 100644
--- a/test/Transforms/SCCP/apint-ipsccp3.ll
+++ b/test/Transforms/SCCP/apint-ipsccp3.ll
@@ -5,13 +5,13 @@
 
 
 define void @foo() {
-	%X = load i66* @G
+	%X = load i66, i66* @G
 	store i66 %X, i66* @G
 	ret void
 }
 
 define i66 @bar() {
-	%V = load i66* @G
+	%V = load i66, i66* @G
 	%C = icmp eq i66 %V, 17
 	br i1 %C, label %T, label %F
 T:
diff --git a/test/Transforms/SCCP/apint-ipsccp4.ll b/test/Transforms/SCCP/apint-ipsccp4.ll
index 75875ff642d2..be06d03f3914 100644
--- a/test/Transforms/SCCP/apint-ipsccp4.ll
+++ b/test/Transforms/SCCP/apint-ipsccp4.ll
@@ -9,14 +9,14 @@
                                      { i212, float } { i212 37, float 2.0 } ]
 
 define internal float @test2() {
-	%A = getelementptr [2 x { i212, float}]* @Y, i32 0, i32 1, i32 1
-	%B = load float* %A
+	%A = getelementptr [2 x { i212, float}], [2 x { i212, float}]* @Y, i32 0, i32 1, i32 1
+	%B = load float, float* %A
 	ret float %B
 }
 
 define internal float  @test3() {
-	%A = getelementptr [2 x { i212, float}]* @Y, i32 0, i32 0, i32 1
-	%B = load float* %A
+	%A = getelementptr [2 x { i212, float}], [2 x { i212, float}]* @Y, i32 0, i32 0, i32 1
+	%B = load float, float* %A
 	ret float %B
 }
 
diff --git a/test/Transforms/SCCP/apint-load.ll b/test/Transforms/SCCP/apint-load.ll
index 56fdb3513f15..17506fc043cc 100644
--- a/test/Transforms/SCCP/apint-load.ll
+++ b/test/Transforms/SCCP/apint-load.ll
@@ -7,19 +7,19 @@
 @Y = constant [2 x { i212, float }] [ { i212, float } { i212 12, float 1.0 }, 
                                      { i212, float } { i212 37, float 0x3FF3B2FEC0000000 } ]
 define i212 @test1() {
-	%B = load i212* @X
+	%B = load i212, i212* @X
 	ret i212 %B
 }
 
 define internal float @test2() {
-	%A = getelementptr [2 x { i212, float}]* @Y, i32 0, i32 1, i32 1
-	%B = load float* %A
+	%A = getelementptr [2 x { i212, float}], [2 x { i212, float}]* @Y, i32 0, i32 1, i32 1
+	%B = load float, float* %A
 	ret float %B
 }
 
 define internal i212 @test3() {
-	%A = getelementptr [2 x { i212, float}]* @Y, i32 0, i32 0, i32 0
-	%B = load i212* %A
+	%A = getelementptr [2 x { i212, float}], [2 x { i212, float}]* @Y, i32 0, i32 0, i32 0
+	%B = load i212, i212* %A
 	ret i212 %B
 }
 
diff --git a/test/Transforms/SCCP/apint-select.ll b/test/Transforms/SCCP/apint-select.ll
index c79751910fcb..893331ea9867 100644
--- a/test/Transforms/SCCP/apint-select.ll
+++ b/test/Transforms/SCCP/apint-select.ll
@@ -3,7 +3,7 @@
 @A = constant i32 10
 
 define i712 @test1() {
-        %P = getelementptr i32* @A, i32 0
+        %P = getelementptr i32, i32* @A, i32 0
         %B = ptrtoint i32* %P to i64
         %BB = and i64 %B, undef
         %C = icmp sge i64 %BB, 0
diff --git a/test/Transforms/SCCP/atomic-load-store.ll b/test/Transforms/SCCP/atomic-load-store.ll
index 53e4c10a720a..45b5d7c80feb 100644
--- a/test/Transforms/SCCP/atomic-load-store.ll
+++ b/test/Transforms/SCCP/atomic-load-store.ll
@@ -6,7 +6,7 @@
 @C = internal constant i32 222
 
 define i32 @test1() {
-	%V = load atomic i32* @G seq_cst, align 4
+	%V = load atomic i32, i32* @G seq_cst, align 4
 	%C = icmp eq i32 %V, 17
 	br i1 %C, label %T, label %F
 T:
@@ -21,7 +21,7 @@ F:
 ; CHECK: ret i32 17
 
 define i32 @test2() {
-	%V = load atomic i32* @C seq_cst, align 4
+	%V = load atomic i32, i32* @C seq_cst, align 4
 	ret i32 %V
 }
 
diff --git a/test/Transforms/SCCP/ipsccp-basic.ll b/test/Transforms/SCCP/ipsccp-basic.ll
index 107b7af2c1dc..c74063f33009 100644
--- a/test/Transforms/SCCP/ipsccp-basic.ll
+++ b/test/Transforms/SCCP/ipsccp-basic.ll
@@ -50,7 +50,7 @@ define i32 @test2b() {
 @G = internal global i32 undef
 
 define void @test3a() {
-	%X = load i32* @G
+	%X = load i32, i32* @G
 	store i32 %X, i32* @G
 	ret void
 }
@@ -59,7 +59,7 @@ define void @test3a() {
 
 
 define i32 @test3b() {
-	%V = load i32* @G
+	%V = load i32, i32* @G
 	%C = icmp eq i32 %V, 17
 	br i1 %C, label %T, label %F
 T:
@@ -203,7 +203,7 @@ define void @test8b(i32* %P) {
 define void @test9() {
 entry:
         %local_foo = alloca {  }
-        load {  }* @test9g
+        load {  }, {  }* @test9g
         store {  } %0, {  }* %local_foo
         ret void
 }
diff --git a/test/Transforms/SCCP/loadtest.ll b/test/Transforms/SCCP/loadtest.ll
index dd1dba69143c..b88b44b76040 100644
--- a/test/Transforms/SCCP/loadtest.ll
+++ b/test/Transforms/SCCP/loadtest.ll
@@ -10,25 +10,25 @@
 @Y = constant [2 x { i32, float }] [ { i32, float } { i32 12, float 1.000000e+00 }, { i32, float } { i32 37, float 0x3FF3B2FEC0000000 } ]		; <[2 x { i32, float }]*> [#uses=2]
 
 define i32 @test1() {
-	%B = load i32* @X		; <i32> [#uses=1]
+	%B = load i32, i32* @X		; <i32> [#uses=1]
 	ret i32 %B
 }
 
 define float @test2() {
-	%A = getelementptr [2 x { i32, float }]* @Y, i64 0, i64 1, i32 1		; <float*> [#uses=1]
-	%B = load float* %A		; <float> [#uses=1]
+	%A = getelementptr [2 x { i32, float }], [2 x { i32, float }]* @Y, i64 0, i64 1, i32 1		; <float*> [#uses=1]
+	%B = load float, float* %A		; <float> [#uses=1]
 	ret float %B
 }
 
 define i32 @test3() {
-	%A = getelementptr [2 x { i32, float }]* @Y, i64 0, i64 0, i32 0		; <i32*> [#uses=1]
-	%B = load i32* %A
+	%A = getelementptr [2 x { i32, float }], [2 x { i32, float }]* @Y, i64 0, i64 0, i32 0		; <i32*> [#uses=1]
+	%B = load i32, i32* %A
 	ret i32 %B
 }
 
 define i8 @test4() {
 	%A = bitcast i32* @X to i8*
-	%B = load i8* %A
+	%B = load i8, i8* %A
 	ret i8 %B
 }
 
diff --git a/test/Transforms/SCCP/retvalue-undef.ll b/test/Transforms/SCCP/retvalue-undef.ll
index 5a4ba113b7c0..f0e9e676cc04 100644
--- a/test/Transforms/SCCP/retvalue-undef.ll
+++ b/test/Transforms/SCCP/retvalue-undef.ll
@@ -15,7 +15,7 @@ define internal i32 @g() {
 ; CHECK-NEXT: ret i32 8
 
 define internal void @outer_mod() {
-  %1 = call i32 ()* ()* @f()                      ; <i32 ()*> [#uses=1]
+  %1 = call i32 ()* () @f()                      ; <i32 ()*> [#uses=1]
   %2 = call i32 %1()                              ; <i32> [#uses=0]
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/AArch64/commute.ll b/test/Transforms/SLPVectorizer/AArch64/commute.ll
index 4ee91a5ed4c0..1cff73d9f695 100644
--- a/test/Transforms/SLPVectorizer/AArch64/commute.ll
+++ b/test/Transforms/SLPVectorizer/AArch64/commute.ll
@@ -6,10 +6,10 @@ target triple = "aarch64--linux-gnu"
 
 define void @test1(%structA* nocapture readonly %J, i32 %xmin, i32 %ymin) {
 ; CHECK-LABEL: test1
-; CHECK: %arrayidx4 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 0
-; CHECK: %arrayidx9 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 1
+; CHECK: %arrayidx4 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 0
+; CHECK: %arrayidx9 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 1
 ; CHECK: %3 = bitcast float* %arrayidx4 to <2 x float>*
-; CHECK: %4 = load <2 x float>* %3, align 4
+; CHECK: %4 = load <2 x float>, <2 x float>* %3, align 4
 ; CHECK: %5 = fsub fast <2 x float> %2, %4
 ; CHECK: %6 = fmul fast <2 x float> %5, %5
 ; CHECK: %7 = extractelement <2 x float> %6, i32 0
@@ -23,11 +23,11 @@ entry:
 for.body3.lr.ph:
   %conv5 = sitofp i32 %ymin to float
   %conv = sitofp i32 %xmin to float
-  %arrayidx4 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 0
-  %0 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 0
+  %0 = load float, float* %arrayidx4, align 4
   %sub = fsub fast float %conv, %0
-  %arrayidx9 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 1
-  %1 = load float* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 1
+  %1 = load float, float* %arrayidx9, align 4
   %sub10 = fsub fast float %conv5, %1
   %mul11 = fmul fast float %sub, %sub
   %mul12 = fmul fast float %sub10, %sub10
@@ -41,10 +41,10 @@ for.end27:
 
 define void @test2(%structA* nocapture readonly %J, i32 %xmin, i32 %ymin) {
 ; CHECK-LABEL: test2
-; CHECK: %arrayidx4 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 0
-; CHECK: %arrayidx9 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 1
+; CHECK: %arrayidx4 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 0
+; CHECK: %arrayidx9 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 1
 ; CHECK: %3 = bitcast float* %arrayidx4 to <2 x float>*
-; CHECK: %4 = load <2 x float>* %3, align 4
+; CHECK: %4 = load <2 x float>, <2 x float>* %3, align 4
 ; CHECK: %5 = fsub fast <2 x float> %2, %4
 ; CHECK: %6 = fmul fast <2 x float> %5, %5
 ; CHECK: %7 = extractelement <2 x float> %6, i32 0
@@ -58,11 +58,11 @@ entry:
 for.body3.lr.ph:
   %conv5 = sitofp i32 %ymin to float
   %conv = sitofp i32 %xmin to float
-  %arrayidx4 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 0
-  %0 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 0
+  %0 = load float, float* %arrayidx4, align 4
   %sub = fsub fast float %conv, %0
-  %arrayidx9 = getelementptr inbounds %structA* %J, i64 0, i32 0, i64 1
-  %1 = load float* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds %structA, %structA* %J, i64 0, i32 0, i64 1
+  %1 = load float, float* %arrayidx9, align 4
   %sub10 = fsub fast float %conv5, %1
   %mul11 = fmul fast float %sub, %sub
   %mul12 = fmul fast float %sub10, %sub10
diff --git a/test/Transforms/SLPVectorizer/AArch64/load-store-q.ll b/test/Transforms/SLPVectorizer/AArch64/load-store-q.ll
index 45fa2f917f3a..6ff1118d3cc6 100644
--- a/test/Transforms/SLPVectorizer/AArch64/load-store-q.ll
+++ b/test/Transforms/SLPVectorizer/AArch64/load-store-q.ll
@@ -13,10 +13,10 @@ target triple = "arm64-apple-ios5.0.0"
 ; CHECK: store double
 ; CHECK: store double
 define void @f(double* %p, double* %q) {
-  %addr2 = getelementptr double* %q, i32 1
-  %addr = getelementptr double* %p, i32 1
-  %x = load double* %p
-  %y = load double* %addr
+  %addr2 = getelementptr double, double* %q, i32 1
+  %addr = getelementptr double, double* %p, i32 1
+  %x = load double, double* %p
+  %y = load double, double* %addr
   call void @g()
   store double %x, double* %q
   store double %y, double* %addr2
@@ -35,12 +35,12 @@ entry:
 loop:
   %p1 = phi double [0.0, %entry], [%x, %loop]
   %p2 = phi double [0.0, %entry], [%y, %loop]
-  %addr2 = getelementptr double* %q, i32 1
-  %addr = getelementptr double* %p, i32 1
+  %addr2 = getelementptr double, double* %q, i32 1
+  %addr = getelementptr double, double* %p, i32 1
   store double %p1, double* %q
   store double %p2, double* %addr2
 
-  %x = load double* %p
-  %y = load double* %addr
+  %x = load double, double* %p
+  %y = load double, double* %addr
   br label %loop
 }
diff --git a/test/Transforms/SLPVectorizer/AArch64/sdiv-pow2.ll b/test/Transforms/SLPVectorizer/AArch64/sdiv-pow2.ll
index e49c7adda400..72c70823e698 100644
--- a/test/Transforms/SLPVectorizer/AArch64/sdiv-pow2.ll
+++ b/test/Transforms/SLPVectorizer/AArch64/sdiv-pow2.ll
@@ -9,34 +9,34 @@ target triple = "aarch64--linux-gnu"
 
 define void @test1(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c) {
 entry:
-  %0 = load i32* %b, align 4
-  %1 = load i32* %c, align 4
+  %0 = load i32, i32* %b, align 4
+  %1 = load i32, i32* %c, align 4
   %add = add nsw i32 %1, %0
   %div = sdiv i32 %add, 2
   store i32 %div, i32* %a, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 1
-  %2 = load i32* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds i32* %c, i64 1
-  %3 = load i32* %arrayidx4, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 1
+  %2 = load i32, i32* %arrayidx3, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %c, i64 1
+  %3 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %3, %2
   %div6 = sdiv i32 %add5, 2
-  %arrayidx7 = getelementptr inbounds i32* %a, i64 1
+  %arrayidx7 = getelementptr inbounds i32, i32* %a, i64 1
   store i32 %div6, i32* %arrayidx7, align 4
-  %arrayidx8 = getelementptr inbounds i32* %b, i64 2
-  %4 = load i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i64 2
-  %5 = load i32* %arrayidx9, align 4
+  %arrayidx8 = getelementptr inbounds i32, i32* %b, i64 2
+  %4 = load i32, i32* %arrayidx8, align 4
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i64 2
+  %5 = load i32, i32* %arrayidx9, align 4
   %add10 = add nsw i32 %5, %4
   %div11 = sdiv i32 %add10, 2
-  %arrayidx12 = getelementptr inbounds i32* %a, i64 2
+  %arrayidx12 = getelementptr inbounds i32, i32* %a, i64 2
   store i32 %div11, i32* %arrayidx12, align 4
-  %arrayidx13 = getelementptr inbounds i32* %b, i64 3
-  %6 = load i32* %arrayidx13, align 4
-  %arrayidx14 = getelementptr inbounds i32* %c, i64 3
-  %7 = load i32* %arrayidx14, align 4
+  %arrayidx13 = getelementptr inbounds i32, i32* %b, i64 3
+  %6 = load i32, i32* %arrayidx13, align 4
+  %arrayidx14 = getelementptr inbounds i32, i32* %c, i64 3
+  %7 = load i32, i32* %arrayidx14, align 4
   %add15 = add nsw i32 %7, %6
   %div16 = sdiv i32 %add15, 2
-  %arrayidx17 = getelementptr inbounds i32* %a, i64 3
+  %arrayidx17 = getelementptr inbounds i32, i32* %a, i64 3
   store i32 %div16, i32* %arrayidx17, align 4
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/ARM/memory.ll b/test/Transforms/SLPVectorizer/ARM/memory.ll
index 383c808d21cf..57d7cceac6b9 100644
--- a/test/Transforms/SLPVectorizer/ARM/memory.ll
+++ b/test/Transforms/SLPVectorizer/ARM/memory.ll
@@ -10,11 +10,11 @@ target datalayout = "e-p:32:32:32-i1:8:32-i8:8:32-i16:16:32-i32:32:32-i64:32:64-
 ; CHECK-NOT: store <2 x double>
 define void @expensive_double_store(double* noalias %dst, double* noalias %src, i64 %count) {
 entry:
-  %0 = load double* %src, align 8
+  %0 = load double, double* %src, align 8
   store double %0, double* %dst, align 8
-  %arrayidx2 = getelementptr inbounds double* %src, i64 1
-  %1 = load double* %arrayidx2, align 8
-  %arrayidx3 = getelementptr inbounds double* %dst, i64 1
+  %arrayidx2 = getelementptr inbounds double, double* %src, i64 1
+  %1 = load double, double* %arrayidx2, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %dst, i64 1
   store double %1, double* %arrayidx3, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/ARM/sroa.ll b/test/Transforms/SLPVectorizer/ARM/sroa.ll
index 899cfb1f82cf..65e02608b290 100644
--- a/test/Transforms/SLPVectorizer/ARM/sroa.ll
+++ b/test/Transforms/SLPVectorizer/ARM/sroa.ll
@@ -44,9 +44,9 @@ entry:
   %3 = bitcast i64 %b.sroa.3.12.insert.insert to double
   %add = fadd double %0, %2
   %add3 = fadd double %1, %3
-  %re.i.i = getelementptr inbounds %class.Complex* %agg.result, i32 0, i32 0
+  %re.i.i = getelementptr inbounds %class.Complex, %class.Complex* %agg.result, i32 0, i32 0
   store double %add, double* %re.i.i, align 4
-  %im.i.i = getelementptr inbounds %class.Complex* %agg.result, i32 0, i32 1
+  %im.i.i = getelementptr inbounds %class.Complex, %class.Complex* %agg.result, i32 0, i32 1
   store double %add3, double* %im.i.i, align 4
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/R600/simplebb.ll b/test/Transforms/SLPVectorizer/R600/simplebb.ll
index b6d794b994d4..9ed86f881473 100644
--- a/test/Transforms/SLPVectorizer/R600/simplebb.ll
+++ b/test/Transforms/SLPVectorizer/R600/simplebb.ll
@@ -6,60 +6,60 @@ target datalayout = "e-p:32:32:32-p3:16:16:16-i1:8:8-i8:8:8-i16:16:16-i32:32:32-
 ; Simple 3-pair chain with loads and stores
 define void @test1_as_3_3_3(double addrspace(3)* %a, double addrspace(3)* %b, double addrspace(3)* %c) {
 ; CHECK-LABEL: @test1_as_3_3_3(
-; CHECK: load <2 x double> addrspace(3)*
-; CHECK: load <2 x double> addrspace(3)*
+; CHECK: load <2 x double>, <2 x double> addrspace(3)*
+; CHECK: load <2 x double>, <2 x double> addrspace(3)*
 ; CHECK: store <2 x double> %{{.*}}, <2 x double> addrspace(3)* %
 ; CHECK: ret
-  %i0 = load double addrspace(3)* %a, align 8
-  %i1 = load double addrspace(3)* %b, align 8
+  %i0 = load double, double addrspace(3)* %a, align 8
+  %i1 = load double, double addrspace(3)* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double addrspace(3)* %a, i64 1
-  %i3 = load double addrspace(3)* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double addrspace(3)* %b, i64 1
-  %i4 = load double addrspace(3)* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double addrspace(3)* %a, i64 1
+  %i3 = load double, double addrspace(3)* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double addrspace(3)* %b, i64 1
+  %i4 = load double, double addrspace(3)* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double addrspace(3)* %c, align 8
-  %arrayidx5 = getelementptr inbounds double addrspace(3)* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double addrspace(3)* %c, i64 1
   store double %mul5, double addrspace(3)* %arrayidx5, align 8
   ret void
 }
 
 define void @test1_as_3_0_0(double addrspace(3)* %a, double* %b, double* %c) {
 ; CHECK-LABEL: @test1_as_3_0_0(
-; CHECK: load <2 x double> addrspace(3)*
-; CHECK: load <2 x double>*
+; CHECK: load <2 x double>, <2 x double> addrspace(3)*
+; CHECK: load <2 x double>, <2 x double>*
 ; CHECK: store <2 x double> %{{.*}}, <2 x double>* %
 ; CHECK: ret
-  %i0 = load double addrspace(3)* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double addrspace(3)* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double addrspace(3)* %a, i64 1
-  %i3 = load double addrspace(3)* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double addrspace(3)* %a, i64 1
+  %i3 = load double, double addrspace(3)* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
 
 define void @test1_as_0_0_3(double* %a, double* %b, double addrspace(3)* %c) {
 ; CHECK-LABEL: @test1_as_0_0_3(
-; CHECK: load <2 x double>*
-; CHECK: load <2 x double>*
+; CHECK: load <2 x double>, <2 x double>*
+; CHECK: load <2 x double>, <2 x double>*
 ; CHECK: store <2 x double> %{{.*}}, <2 x double> addrspace(3)* %
 ; CHECK: ret
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double addrspace(3)* %c, align 8
-  %arrayidx5 = getelementptr inbounds double addrspace(3)* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double addrspace(3)* %c, i64 1
   store double %mul5, double addrspace(3)* %arrayidx5, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/addsub.ll b/test/Transforms/SLPVectorizer/X86/addsub.ll
index 174d40046849..7056a14c1c4f 100644
--- a/test/Transforms/SLPVectorizer/X86/addsub.ll
+++ b/test/Transforms/SLPVectorizer/X86/addsub.ll
@@ -10,6 +10,7 @@ target triple = "x86_64-unknown-linux-gnu"
 @fb = common global [4 x float] zeroinitializer, align 16
 @fc = common global [4 x float] zeroinitializer, align 16
 @fa = common global [4 x float] zeroinitializer, align 16
+@fd = common global [4 x float] zeroinitializer, align 16
 
 ; CHECK-LABEL: @addsub
 ; CHECK: %5 = add nsw <4 x i32> %3, %4
@@ -20,38 +21,38 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define void @addsub() #0 {
 entry:
-  %0 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 0), align 4
-  %1 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 0), align 4
+  %0 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 0), align 4
+  %1 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 0), align 4
   %add = add nsw i32 %0, %1
-  %2 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 0), align 4
-  %3 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 0), align 4
+  %2 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 0), align 4
+  %3 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 0), align 4
   %add1 = add nsw i32 %2, %3
   %add2 = add nsw i32 %add, %add1
-  store i32 %add2, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 0), align 4
-  %4 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 1), align 4
-  %5 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 1), align 4
+  store i32 %add2, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 0), align 4
+  %4 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 1), align 4
+  %5 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 1), align 4
   %add3 = add nsw i32 %4, %5
-  %6 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 1), align 4
-  %7 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 1), align 4
+  %6 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 1), align 4
+  %7 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 1), align 4
   %add4 = add nsw i32 %6, %7
   %sub = sub nsw i32 %add3, %add4
-  store i32 %sub, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 1), align 4
-  %8 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 2), align 4
-  %9 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 2), align 4
+  store i32 %sub, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 1), align 4
+  %8 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 2), align 4
+  %9 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 2), align 4
   %add5 = add nsw i32 %8, %9
-  %10 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 2), align 4
-  %11 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 2), align 4
+  %10 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 2), align 4
+  %11 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 2), align 4
   %add6 = add nsw i32 %10, %11
   %add7 = add nsw i32 %add5, %add6
-  store i32 %add7, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 2), align 4
-  %12 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 3), align 4
-  %13 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 3), align 4
+  store i32 %add7, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 2), align 4
+  %12 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 3), align 4
+  %13 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 3), align 4
   %add8 = add nsw i32 %12, %13
-  %14 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 3), align 4
-  %15 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 3), align 4
+  %14 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 3), align 4
+  %15 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 3), align 4
   %add9 = add nsw i32 %14, %15
   %sub10 = sub nsw i32 %add8, %add9
-  store i32 %sub10, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 3), align 4
+  store i32 %sub10, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 3), align 4
   ret void
 }
 
@@ -64,38 +65,38 @@ entry:
 ; Function Attrs: nounwind uwtable
 define void @subadd() #0 {
 entry:
-  %0 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 0), align 4
-  %1 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 0), align 4
+  %0 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 0), align 4
+  %1 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 0), align 4
   %add = add nsw i32 %0, %1
-  %2 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 0), align 4
-  %3 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 0), align 4
+  %2 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 0), align 4
+  %3 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 0), align 4
   %add1 = add nsw i32 %2, %3
   %sub = sub nsw i32 %add, %add1
-  store i32 %sub, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 0), align 4
-  %4 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 1), align 4
-  %5 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 1), align 4
+  store i32 %sub, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 0), align 4
+  %4 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 1), align 4
+  %5 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 1), align 4
   %add2 = add nsw i32 %4, %5
-  %6 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 1), align 4
-  %7 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 1), align 4
+  %6 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 1), align 4
+  %7 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 1), align 4
   %add3 = add nsw i32 %6, %7
   %add4 = add nsw i32 %add2, %add3
-  store i32 %add4, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 1), align 4
-  %8 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 2), align 4
-  %9 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 2), align 4
+  store i32 %add4, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 1), align 4
+  %8 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 2), align 4
+  %9 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 2), align 4
   %add5 = add nsw i32 %8, %9
-  %10 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 2), align 4
-  %11 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 2), align 4
+  %10 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 2), align 4
+  %11 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 2), align 4
   %add6 = add nsw i32 %10, %11
   %sub7 = sub nsw i32 %add5, %add6
-  store i32 %sub7, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 2), align 4
-  %12 = load i32* getelementptr inbounds ([4 x i32]* @b, i32 0, i64 3), align 4
-  %13 = load i32* getelementptr inbounds ([4 x i32]* @c, i32 0, i64 3), align 4
+  store i32 %sub7, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 2), align 4
+  %12 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @b, i32 0, i64 3), align 4
+  %13 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @c, i32 0, i64 3), align 4
   %add8 = add nsw i32 %12, %13
-  %14 = load i32* getelementptr inbounds ([4 x i32]* @d, i32 0, i64 3), align 4
-  %15 = load i32* getelementptr inbounds ([4 x i32]* @e, i32 0, i64 3), align 4
+  %14 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @d, i32 0, i64 3), align 4
+  %15 = load i32, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @e, i32 0, i64 3), align 4
   %add9 = add nsw i32 %14, %15
   %add10 = add nsw i32 %add8, %add9
-  store i32 %add10, i32* getelementptr inbounds ([4 x i32]* @a, i32 0, i64 3), align 4
+  store i32 %add10, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i32 0, i64 3), align 4
   ret void
 }
 
@@ -106,22 +107,22 @@ entry:
 ; Function Attrs: nounwind uwtable
 define void @faddfsub() #0 {
 entry:
-  %0 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 0), align 4
-  %1 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 0), align 4
+  %0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
   %add = fadd float %0, %1
-  store float %add, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 0), align 4
-  %2 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 1), align 4
-  %3 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 1), align 4
+  store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
   %sub = fsub float %2, %3
-  store float %sub, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 1), align 4
-  %4 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 2), align 4
-  %5 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 2), align 4
+  store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
   %add1 = fadd float %4, %5
-  store float %add1, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 2), align 4
-  %6 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 3), align 4
-  %7 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 3), align 4
+  store float %add1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
   %sub2 = fsub float %6, %7
-  store float %sub2, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 3), align 4
+  store float %sub2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
   ret void
 }
 
@@ -132,22 +133,22 @@ entry:
 ; Function Attrs: nounwind uwtable
 define void @fsubfadd() #0 {
 entry:
-  %0 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 0), align 4
-  %1 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 0), align 4
+  %0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
   %sub = fsub float %0, %1
-  store float %sub, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 0), align 4
-  %2 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 1), align 4
-  %3 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 1), align 4
+  store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
   %add = fadd float %2, %3
-  store float %add, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 1), align 4
-  %4 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 2), align 4
-  %5 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 2), align 4
+  store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
   %sub1 = fsub float %4, %5
-  store float %sub1, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 2), align 4
-  %6 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 3), align 4
-  %7 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 3), align 4
+  store float %sub1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
   %add2 = fadd float %6, %7
-  store float %add2, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 3), align 4
+  store float %add2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
   ret void
 }
 
@@ -158,24 +159,156 @@ entry:
 ; Function Attrs: nounwind uwtable
 define void @No_faddfsub() #0 {
 entry:
-  %0 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 0), align 4
-  %1 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 0), align 4
+  %0 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
   %add = fadd float %0, %1
-  store float %add, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 0), align 4
-  %2 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 1), align 4
-  %3 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 1), align 4
+  store float %add, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
   %add1 = fadd float %2, %3
-  store float %add1, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 1), align 4
-  %4 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 2), align 4
-  %5 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 2), align 4
+  store float %add1, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
   %add2 = fadd float %4, %5
-  store float %add2, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 2), align 4
-  %6 = load float* getelementptr inbounds ([4 x float]* @fb, i32 0, i64 3), align 4
-  %7 = load float* getelementptr inbounds ([4 x float]* @fc, i32 0, i64 3), align 4
+  store float %add2, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
   %sub = fsub float %6, %7
-  store float %sub, float* getelementptr inbounds ([4 x float]* @fa, i32 0, i64 3), align 4
+  store float %sub, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
   ret void
 }
 
+; Check vectorization of following code for float data type-
+;  fc[0] = fb[0]+fa[0]; //swapped fb and fa
+;  fc[1] = fa[1]-fb[1];
+;  fc[2] = fa[2]+fb[2];
+;  fc[3] = fa[3]-fb[3];
+
+; CHECK-LABEL: @reorder_alt
+; CHECK: %3 = fadd <4 x float> %1, %2
+; CHECK: %4 = fsub <4 x float> %1, %2
+; CHECK: %5 = shufflevector <4 x float> %3, <4 x float> %4, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+define void @reorder_alt() #0 {
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %3 = fadd float %1, %2
+  store float %3, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
+  %4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %6 = fsub float %4, %5
+  store float %6, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %9 = fadd float %7, %8
+  store float %9, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
+  %10 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
+  %11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %12 = fsub float %10, %11
+  store float %12, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
+  ret void
+}
+
+; Check vectorization of following code for float data type-
+;  fc[0] = fa[0]+(fb[0]-fd[0]);
+;  fc[1] = fa[1]-(fb[1]+fd[1]);
+;  fc[2] = fa[2]+(fb[2]-fd[2]);
+;  fc[3] = fa[3]-(fd[3]+fb[3]); //swapped fd and fb 
+
+; CHECK-LABEL: @reorder_alt_subTree
+; CHECK: %4 = fsub <4 x float> %3, %2
+; CHECK: %5 = fadd <4 x float> %3, %2
+; CHECK: %6 = shufflevector <4 x float> %4, <4 x float> %5, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+; CHECK: %7 = fadd <4 x float> %1, %6
+; CHECK: %8 = fsub <4 x float> %1, %6
+; CHECK: %9 = shufflevector <4 x float> %7, <4 x float> %8, <4 x i32> <i32 0, i32 5, i32 2, i32 7>
+define void @reorder_alt_subTree() #0 {
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %3 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 0), align 4
+  %4 = fsub float %2, %3
+  %5 = fadd float %1, %4
+  store float %5, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
+  %6 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 1), align 4
+  %9 = fadd float %7, %8
+  %10 = fsub float %6, %9
+  store float %10, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
+  %11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %12 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %13 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 2), align 4
+  %14 = fsub float %12, %13
+  %15 = fadd float %11, %14
+  store float %15, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
+  %16 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
+  %17 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fd, i32 0, i64 3), align 4
+  %18 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %19 = fadd float %17, %18
+  %20 = fsub float %16, %19
+  store float %20, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
+  ret void
+}
+
+; Check vectorization of following code for double data type-
+;  c[0] = (a[0]+b[0])-d[0];
+;  c[1] = d[1]+(a[1]+b[1]); //swapped d[1] and (a[1]+b[1]) 
+
+; CHECK-LABEL: @reorder_alt_rightsubTree
+; CHECK: fadd <2 x double>
+; CHECK: fsub <2 x double>
+; CHECK: shufflevector <2 x double> 
+define void @reorder_alt_rightsubTree(double* nocapture %c, double* noalias nocapture readonly %a, double* noalias nocapture readonly %b, double* noalias nocapture readonly %d) {
+  %1 = load double, double* %a
+  %2 = load double, double* %b
+  %3 = fadd double %1, %2
+  %4 = load double, double* %d
+  %5 = fsub double %3, %4
+  store double %5, double* %c
+  %6 = getelementptr inbounds double, double* %d, i64 1
+  %7 = load double, double* %6
+  %8 = getelementptr inbounds double, double* %a, i64 1
+  %9 = load double, double* %8
+  %10 = getelementptr inbounds double, double* %b, i64 1
+  %11 = load double, double* %10
+  %12 = fadd double %9, %11
+  %13 = fadd double %7, %12
+  %14 = getelementptr inbounds double, double* %c, i64 1
+  store double %13, double* %14
+  ret void
+}
+
+; Dont vectorization of following code for float data type as sub is not commutative-
+;  fc[0] = fb[0]+fa[0];
+;  fc[1] = fa[1]-fb[1];
+;  fc[2] = fa[2]+fb[2];
+;  fc[3] = fb[3]-fa[3];
+;  In the above code we can swap the 1st and 2nd operation as fadd is commutative
+;  but not 2nd or 4th as fsub is not commutative. 
+
+; CHECK-LABEL: @no_vec_shuff_reorder
+; CHECK-NOT: fadd <4 x float>
+; CHECK-NOT: fsub <4 x float>
+; CHECK-NOT: shufflevector
+define void @no_vec_shuff_reorder() #0 {
+  %1 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 0), align 4
+  %2 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 0), align 4
+  %3 = fadd float %1, %2
+  store float %3, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 0), align 4
+  %4 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 1), align 4
+  %5 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 1), align 4
+  %6 = fsub float %4, %5
+  store float %6, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 1), align 4
+  %7 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 2), align 4
+  %8 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 2), align 4
+  %9 = fadd float %7, %8
+  store float %9, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 2), align 4
+  %10 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fb, i32 0, i64 3), align 4
+  %11 = load float, float* getelementptr inbounds ([4 x float], [4 x float]* @fa, i32 0, i64 3), align 4
+  %12 = fsub float %10, %11
+  store float %12, float* getelementptr inbounds ([4 x float], [4 x float]* @fc, i32 0, i64 3), align 4
+  ret void
+}
+
+
 attributes #0 = { nounwind }
 
diff --git a/test/Transforms/SLPVectorizer/X86/align.ll b/test/Transforms/SLPVectorizer/X86/align.ll
index ce8062025064..b74b70900ee9 100644
--- a/test/Transforms/SLPVectorizer/X86/align.ll
+++ b/test/Transforms/SLPVectorizer/X86/align.ll
@@ -8,16 +8,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 define void @test1(double* %a, double* %b, double* %c) {
 entry:
   %agg.tmp.i.i.sroa.0 = alloca [3 x double], align 16
-; CHECK: %[[V0:[0-9]+]] = load <2 x double>* %[[V2:[0-9]+]], align 8
-  %i0 = load double* %a 
-  %i1 = load double* %b 
+; CHECK: %[[V0:[0-9]+]] = load <2 x double>, <2 x double>* %[[V2:[0-9]+]], align 8
+  %i0 = load double, double* %a 
+  %i1 = load double, double* %b 
   %mul = fmul double %i0, %i1
-  %store1 = getelementptr inbounds [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 1
-  %store2 = getelementptr inbounds [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 2
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %store1 = getelementptr inbounds [3 x double], [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 1
+  %store2 = getelementptr inbounds [3 x double], [3 x double]* %agg.tmp.i.i.sroa.0, i64 0, i64 2
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
 ; CHECK: store <2 x double> %[[V1:[0-9]+]], <2 x double>* %[[V2:[0-9]+]], align 8
   store double %mul, double* %store1
@@ -37,19 +37,19 @@ entry:
 
 define void @test2(float * %a, float * %b) {
 entry:
-  %l0 = load float* %a
-  %a1 = getelementptr inbounds float* %a, i64 1
-  %l1 = load float* %a1
-  %a2 = getelementptr inbounds float* %a, i64 2
-  %l2 = load float* %a2
-  %a3 = getelementptr inbounds float* %a, i64 3
-  %l3 = load float* %a3
+  %l0 = load float, float* %a
+  %a1 = getelementptr inbounds float, float* %a, i64 1
+  %l1 = load float, float* %a1
+  %a2 = getelementptr inbounds float, float* %a, i64 2
+  %l2 = load float, float* %a2
+  %a3 = getelementptr inbounds float, float* %a, i64 3
+  %l3 = load float, float* %a3
   store float %l0, float* %b
-  %b1 = getelementptr inbounds float* %b, i64 1
+  %b1 = getelementptr inbounds float, float* %b, i64 1
   store float %l1, float* %b1
-  %b2 = getelementptr inbounds float* %b, i64 2
+  %b2 = getelementptr inbounds float, float* %b, i64 2
   store float %l2, float* %b2
-  %b3 = getelementptr inbounds float* %b, i64 3
+  %b3 = getelementptr inbounds float, float* %b, i64 3
   store float %l3, float* %b3
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/atomics.ll b/test/Transforms/SLPVectorizer/X86/atomics.ll
new file mode 100644
index 000000000000..a48b793922f4
--- /dev/null
+++ b/test/Transforms/SLPVectorizer/X86/atomics.ll
@@ -0,0 +1,31 @@
+; RUN: opt < %s -basicaa -slp-vectorizer -S |FileCheck %s
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+
+@x = global [4 x i32] zeroinitializer, align 16
+@a = global [4 x i32] zeroinitializer, align 16
+
+; The SLPVectorizer should not vectorize atomic stores and it should not
+; schedule regular stores around atomic stores.
+
+; CHECK-LABEL: test
+; CHECK: store i32
+; CHECK: store atomic i32
+; CHECK: store i32
+; CHECK: store atomic i32
+; CHECK: store i32
+; CHECK: store atomic i32
+; CHECK: store i32
+; CHECK: store atomic i32
+define void @test() {
+entry:
+  store i32 0, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i64 0, i64 0), align 16
+  store atomic i32 0, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @x, i64 0, i64 0) release, align 16
+  store i32 0, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i64 0, i64 1), align 4
+  store atomic i32 1, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @x, i64 0, i64 1) release, align 4
+  store i32 0, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i64 0, i64 2), align 8
+  store atomic i32 2, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @x, i64 0, i64 2) release, align 8
+  store i32 0, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @a, i64 0, i64 3), align 4
+  store atomic i32 3, i32* getelementptr inbounds ([4 x i32], [4 x i32]* @x, i64 0, i64 3) release, align 4
+  ret void
+}
+
diff --git a/test/Transforms/SLPVectorizer/X86/bad_types.ll b/test/Transforms/SLPVectorizer/X86/bad_types.ll
index 38ed18dad2ac..2d8f3832ee29 100644
--- a/test/Transforms/SLPVectorizer/X86/bad_types.ll
+++ b/test/Transforms/SLPVectorizer/X86/bad_types.ll
@@ -16,7 +16,7 @@ entry:
   %b.cast = bitcast x86_mmx %b to i64
   %a.and = and i64 %a.cast, 42
   %b.and = and i64 %b.cast, 42
-  %gep = getelementptr i64* %ptr, i32 1
+  %gep = getelementptr i64, i64* %ptr, i32 1
   store i64 %a.and, i64* %ptr
   store i64 %b.and, i64* %gep
   ret void
diff --git a/test/Transforms/SLPVectorizer/X86/barriercall.ll b/test/Transforms/SLPVectorizer/X86/barriercall.ll
index bba285526a4b..382a43fddf58 100644
--- a/test/Transforms/SLPVectorizer/X86/barriercall.ll
+++ b/test/Transforms/SLPVectorizer/X86/barriercall.ll
@@ -8,21 +8,21 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK: ret
 define i32 @foo(i32* nocapture %A, i32 %n) {
 entry:
-  %call = tail call i32 (...)* @bar() #2
+  %call = tail call i32 (...) @bar() #2
   %mul = mul nsw i32 %n, 5
   %add = add nsw i32 %mul, 9
   store i32 %add, i32* %A, align 4
   %mul1 = mul nsw i32 %n, 9
   %add2 = add nsw i32 %mul1, 9
-  %arrayidx3 = getelementptr inbounds i32* %A, i64 1
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i64 1
   store i32 %add2, i32* %arrayidx3, align 4
   %mul4 = shl i32 %n, 3
   %add5 = add nsw i32 %mul4, 9
-  %arrayidx6 = getelementptr inbounds i32* %A, i64 2
+  %arrayidx6 = getelementptr inbounds i32, i32* %A, i64 2
   store i32 %add5, i32* %arrayidx6, align 4
   %mul7 = mul nsw i32 %n, 10
   %add8 = add nsw i32 %mul7, 9
-  %arrayidx9 = getelementptr inbounds i32* %A, i64 3
+  %arrayidx9 = getelementptr inbounds i32, i32* %A, i64 3
   store i32 %add8, i32* %arrayidx9, align 4
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/call.ll b/test/Transforms/SLPVectorizer/X86/call.ll
index 83d45c0a9d72..a55cc1bf793f 100644
--- a/test/Transforms/SLPVectorizer/X86/call.ll
+++ b/test/Transforms/SLPVectorizer/X86/call.ll
@@ -15,18 +15,18 @@ declare i64 @round(i64)
 ; CHECK: ret void
 define void @sin_libm(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %call = tail call double @sin(double %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %call5 = tail call double @sin(double %mul5) nounwind readnone
   store double %call, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
   ret void
 }
@@ -36,18 +36,18 @@ entry:
 ; CHECK: ret void
 define void @cos_libm(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %call = tail call double @cos(double %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %call5 = tail call double @cos(double %mul5) nounwind readnone
   store double %call, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
   ret void
 }
@@ -57,18 +57,18 @@ entry:
 ; CHECK: ret void
 define void @pow_libm(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %call = tail call double @pow(double %mul,double %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %call5 = tail call double @pow(double %mul5,double %mul5) nounwind readnone
   store double %call, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
   ret void
 }
@@ -79,18 +79,18 @@ entry:
 ; CHECK: ret void
 define void @exp2_libm(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %call = tail call double @exp2(double %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %call5 = tail call double @exp2(double %mul5) nounwind readnone
   store double %call, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
   ret void
 }
@@ -102,27 +102,27 @@ entry:
 ; CHECK: ret void
 define void @round_custom(i64* %a, i64* %b, i64* %c) {
 entry:
-  %i0 = load i64* %a, align 8
-  %i1 = load i64* %b, align 8
+  %i0 = load i64, i64* %a, align 8
+  %i1 = load i64, i64* %b, align 8
   %mul = mul i64 %i0, %i1
   %call = tail call i64 @round(i64 %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds i64* %a, i64 1
-  %i3 = load i64* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds i64* %b, i64 1
-  %i4 = load i64* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds i64, i64* %a, i64 1
+  %i3 = load i64, i64* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds i64, i64* %b, i64 1
+  %i4 = load i64, i64* %arrayidx4, align 8
   %mul5 = mul i64 %i3, %i4
   %call5 = tail call i64 @round(i64 %mul5) nounwind readnone
   store i64 %call, i64* %c, align 8
-  %arrayidx5 = getelementptr inbounds i64* %c, i64 1
+  %arrayidx5 = getelementptr inbounds i64, i64* %c, i64 1
   store i64 %call5, i64* %arrayidx5, align 8
   ret void
 }
 
 
-; CHECK: declare <2 x double> @llvm.sin.v2f64(<2 x double>) #0
-; CHECK: declare <2 x double> @llvm.cos.v2f64(<2 x double>) #0
-; CHECK: declare <2 x double> @llvm.pow.v2f64(<2 x double>, <2 x double>) #0
-; CHECK: declare <2 x double> @llvm.exp2.v2f64(<2 x double>) #0
+; CHECK: declare <2 x double> @llvm.sin.v2f64(<2 x double>) [[ATTR0:#[0-9]+]]
+; CHECK: declare <2 x double> @llvm.cos.v2f64(<2 x double>) [[ATTR0]]
+; CHECK: declare <2 x double> @llvm.pow.v2f64(<2 x double>, <2 x double>) [[ATTR0]]
+; CHECK: declare <2 x double> @llvm.exp2.v2f64(<2 x double>) [[ATTR0]]
 
-; CHECK: attributes #0 = { nounwind readnone }
+; CHECK: attributes [[ATTR0]] = { nounwind readnone }
 
diff --git a/test/Transforms/SLPVectorizer/X86/cast.ll b/test/Transforms/SLPVectorizer/X86/cast.ll
index e340fba351a3..044db5d694b6 100644
--- a/test/Transforms/SLPVectorizer/X86/cast.ll
+++ b/test/Transforms/SLPVectorizer/X86/cast.ll
@@ -15,23 +15,23 @@ target triple = "x86_64-apple-macosx10.9.0"
 ;CHECK: store <4 x i32>
 define i32 @foo(i32* noalias nocapture %A, i8* noalias nocapture %B) {
 entry:
-  %0 = load i8* %B, align 1
+  %0 = load i8, i8* %B, align 1
   %conv = sext i8 %0 to i32
   store i32 %conv, i32* %A, align 4
-  %arrayidx2 = getelementptr inbounds i8* %B, i64 1
-  %1 = load i8* %arrayidx2, align 1
+  %arrayidx2 = getelementptr inbounds i8, i8* %B, i64 1
+  %1 = load i8, i8* %arrayidx2, align 1
   %conv3 = sext i8 %1 to i32
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 1
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
   store i32 %conv3, i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i8* %B, i64 2
-  %2 = load i8* %arrayidx5, align 1
+  %arrayidx5 = getelementptr inbounds i8, i8* %B, i64 2
+  %2 = load i8, i8* %arrayidx5, align 1
   %conv6 = sext i8 %2 to i32
-  %arrayidx7 = getelementptr inbounds i32* %A, i64 2
+  %arrayidx7 = getelementptr inbounds i32, i32* %A, i64 2
   store i32 %conv6, i32* %arrayidx7, align 4
-  %arrayidx8 = getelementptr inbounds i8* %B, i64 3
-  %3 = load i8* %arrayidx8, align 1
+  %arrayidx8 = getelementptr inbounds i8, i8* %B, i64 3
+  %3 = load i8, i8* %arrayidx8, align 1
   %conv9 = sext i8 %3 to i32
-  %arrayidx10 = getelementptr inbounds i32* %A, i64 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %A, i64 3
   store i32 %conv9, i32* %arrayidx10, align 4
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/cmp_sel.ll b/test/Transforms/SLPVectorizer/X86/cmp_sel.ll
index 0c124a75d417..a3e2b21ea6e0 100644
--- a/test/Transforms/SLPVectorizer/X86/cmp_sel.ll
+++ b/test/Transforms/SLPVectorizer/X86/cmp_sel.ll
@@ -16,16 +16,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK: ret i32 undef
 define i32 @foo(double* noalias nocapture %A, double* noalias nocapture %B, double %G) {
 entry:
-  %arrayidx = getelementptr inbounds double* %B, i64 10
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %B, i64 10
+  %0 = load double, double* %arrayidx, align 8
   %tobool = fcmp une double %0, 0.000000e+00
   %cond = select i1 %tobool, double %G, double 1.000000e+00
   store double %cond, double* %A, align 8
-  %arrayidx2 = getelementptr inbounds double* %B, i64 11
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %B, i64 11
+  %1 = load double, double* %arrayidx2, align 8
   %tobool3 = fcmp une double %1, 0.000000e+00
   %cond7 = select i1 %tobool3, double %G, double 1.000000e+00
-  %arrayidx8 = getelementptr inbounds double* %A, i64 1
+  %arrayidx8 = getelementptr inbounds double, double* %A, i64 1
   store double %cond7, double* %arrayidx8, align 8
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/compare-reduce.ll b/test/Transforms/SLPVectorizer/X86/compare-reduce.ll
index 9653d18db566..be17c5db2e86 100644
--- a/test/Transforms/SLPVectorizer/X86/compare-reduce.ll
+++ b/test/Transforms/SLPVectorizer/X86/compare-reduce.ll
@@ -21,14 +21,14 @@ entry:
 for.body:                                         ; preds = %for.inc, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
   %0 = shl nsw i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds double* %A, i64 %0
-  %1 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 %0
+  %1 = load double, double* %arrayidx, align 8
   %mul1 = fmul double %conv, %1
   %mul2 = fmul double %mul1, 7.000000e+00
   %add = fadd double %mul2, 5.000000e+00
   %2 = or i64 %0, 1
-  %arrayidx6 = getelementptr inbounds double* %A, i64 %2
-  %3 = load double* %arrayidx6, align 8
+  %arrayidx6 = getelementptr inbounds double, double* %A, i64 %2
+  %3 = load double, double* %arrayidx6, align 8
   %mul8 = fmul double %conv, %3
   %mul9 = fmul double %mul8, 4.000000e+00
   %add10 = fadd double %mul9, 9.000000e+00
@@ -36,7 +36,7 @@ for.body:                                         ; preds = %for.inc, %entry
   br i1 %cmp11, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %call = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([6 x i8]* @.str, i64 0, i64 0))
+  %call = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str, i64 0, i64 0))
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body, %if.then
diff --git a/test/Transforms/SLPVectorizer/X86/consecutive-access.ll b/test/Transforms/SLPVectorizer/X86/consecutive-access.ll
index aa59429a77d1..1ad4d694d34a 100644
--- a/test/Transforms/SLPVectorizer/X86/consecutive-access.ll
+++ b/test/Transforms/SLPVectorizer/X86/consecutive-access.ll
@@ -20,26 +20,26 @@ entry:
   store i32 %u, i32* %u.addr, align 4
   %mul = mul nsw i32 %u, 3
   %idxprom = sext i32 %mul to i64
-  %arrayidx = getelementptr inbounds [2000 x double]* @A, i32 0, i64 %idxprom
-  %0 = load double* %arrayidx, align 8
-  %arrayidx4 = getelementptr inbounds [2000 x double]* @B, i32 0, i64 %idxprom
-  %1 = load double* %arrayidx4, align 8
+  %arrayidx = getelementptr inbounds [2000 x double], [2000 x double]* @A, i32 0, i64 %idxprom
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx4 = getelementptr inbounds [2000 x double], [2000 x double]* @B, i32 0, i64 %idxprom
+  %1 = load double, double* %arrayidx4, align 8
   %add5 = fadd double %0, %1
   store double %add5, double* %arrayidx, align 8
   %add11 = add nsw i32 %mul, 1
   %idxprom12 = sext i32 %add11 to i64
-  %arrayidx13 = getelementptr inbounds [2000 x double]* @A, i32 0, i64 %idxprom12
-  %2 = load double* %arrayidx13, align 8
-  %arrayidx17 = getelementptr inbounds [2000 x double]* @B, i32 0, i64 %idxprom12
-  %3 = load double* %arrayidx17, align 8
+  %arrayidx13 = getelementptr inbounds [2000 x double], [2000 x double]* @A, i32 0, i64 %idxprom12
+  %2 = load double, double* %arrayidx13, align 8
+  %arrayidx17 = getelementptr inbounds [2000 x double], [2000 x double]* @B, i32 0, i64 %idxprom12
+  %3 = load double, double* %arrayidx17, align 8
   %add18 = fadd double %2, %3
   store double %add18, double* %arrayidx13, align 8
   %add24 = add nsw i32 %mul, 2
   %idxprom25 = sext i32 %add24 to i64
-  %arrayidx26 = getelementptr inbounds [2000 x double]* @A, i32 0, i64 %idxprom25
-  %4 = load double* %arrayidx26, align 8
-  %arrayidx30 = getelementptr inbounds [2000 x double]* @B, i32 0, i64 %idxprom25
-  %5 = load double* %arrayidx30, align 8
+  %arrayidx26 = getelementptr inbounds [2000 x double], [2000 x double]* @A, i32 0, i64 %idxprom25
+  %4 = load double, double* %arrayidx26, align 8
+  %arrayidx30 = getelementptr inbounds [2000 x double], [2000 x double]* @B, i32 0, i64 %idxprom25
+  %5 = load double, double* %arrayidx30, align 8
   %add31 = fadd double %4, %5
   store double %add31, double* %arrayidx26, align 8
   ret void
@@ -57,18 +57,18 @@ entry:
   store i32 %u, i32* %u.addr, align 4
   %mul = mul nsw i32 %u, 2
   %idxprom = sext i32 %mul to i64
-  %arrayidx = getelementptr inbounds [2000 x double]* @A, i32 0, i64 %idxprom
-  %0 = load double* %arrayidx, align 8
-  %arrayidx4 = getelementptr inbounds [2000 x double]* @B, i32 0, i64 %idxprom
-  %1 = load double* %arrayidx4, align 8
+  %arrayidx = getelementptr inbounds [2000 x double], [2000 x double]* @A, i32 0, i64 %idxprom
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx4 = getelementptr inbounds [2000 x double], [2000 x double]* @B, i32 0, i64 %idxprom
+  %1 = load double, double* %arrayidx4, align 8
   %add5 = fadd double %0, %1
   store double %add5, double* %arrayidx, align 8
   %add11 = add nsw i32 %mul, 1
   %idxprom12 = sext i32 %add11 to i64
-  %arrayidx13 = getelementptr inbounds [2000 x double]* @A, i32 0, i64 %idxprom12
-  %2 = load double* %arrayidx13, align 8
-  %arrayidx17 = getelementptr inbounds [2000 x double]* @B, i32 0, i64 %idxprom12
-  %3 = load double* %arrayidx17, align 8
+  %arrayidx13 = getelementptr inbounds [2000 x double], [2000 x double]* @A, i32 0, i64 %idxprom12
+  %2 = load double, double* %arrayidx13, align 8
+  %arrayidx17 = getelementptr inbounds [2000 x double], [2000 x double]* @B, i32 0, i64 %idxprom12
+  %3 = load double, double* %arrayidx17, align 8
   %add18 = fadd double %2, %3
   store double %add18, double* %arrayidx13, align 8
   ret void
@@ -84,34 +84,34 @@ entry:
   store i32 %u, i32* %u.addr, align 4
   %mul = mul nsw i32 %u, 4
   %idxprom = sext i32 %mul to i64
-  %arrayidx = getelementptr inbounds [2000 x float]* @C, i32 0, i64 %idxprom
-  %0 = load float* %arrayidx, align 4
-  %arrayidx4 = getelementptr inbounds [2000 x float]* @D, i32 0, i64 %idxprom
-  %1 = load float* %arrayidx4, align 4
+  %arrayidx = getelementptr inbounds [2000 x float], [2000 x float]* @C, i32 0, i64 %idxprom
+  %0 = load float, float* %arrayidx, align 4
+  %arrayidx4 = getelementptr inbounds [2000 x float], [2000 x float]* @D, i32 0, i64 %idxprom
+  %1 = load float, float* %arrayidx4, align 4
   %add5 = fadd float %0, %1
   store float %add5, float* %arrayidx, align 4
   %add11 = add nsw i32 %mul, 1
   %idxprom12 = sext i32 %add11 to i64
-  %arrayidx13 = getelementptr inbounds [2000 x float]* @C, i32 0, i64 %idxprom12
-  %2 = load float* %arrayidx13, align 4
-  %arrayidx17 = getelementptr inbounds [2000 x float]* @D, i32 0, i64 %idxprom12
-  %3 = load float* %arrayidx17, align 4
+  %arrayidx13 = getelementptr inbounds [2000 x float], [2000 x float]* @C, i32 0, i64 %idxprom12
+  %2 = load float, float* %arrayidx13, align 4
+  %arrayidx17 = getelementptr inbounds [2000 x float], [2000 x float]* @D, i32 0, i64 %idxprom12
+  %3 = load float, float* %arrayidx17, align 4
   %add18 = fadd float %2, %3
   store float %add18, float* %arrayidx13, align 4
   %add24 = add nsw i32 %mul, 2
   %idxprom25 = sext i32 %add24 to i64
-  %arrayidx26 = getelementptr inbounds [2000 x float]* @C, i32 0, i64 %idxprom25
-  %4 = load float* %arrayidx26, align 4
-  %arrayidx30 = getelementptr inbounds [2000 x float]* @D, i32 0, i64 %idxprom25
-  %5 = load float* %arrayidx30, align 4
+  %arrayidx26 = getelementptr inbounds [2000 x float], [2000 x float]* @C, i32 0, i64 %idxprom25
+  %4 = load float, float* %arrayidx26, align 4
+  %arrayidx30 = getelementptr inbounds [2000 x float], [2000 x float]* @D, i32 0, i64 %idxprom25
+  %5 = load float, float* %arrayidx30, align 4
   %add31 = fadd float %4, %5
   store float %add31, float* %arrayidx26, align 4
   %add37 = add nsw i32 %mul, 3
   %idxprom38 = sext i32 %add37 to i64
-  %arrayidx39 = getelementptr inbounds [2000 x float]* @C, i32 0, i64 %idxprom38
-  %6 = load float* %arrayidx39, align 4
-  %arrayidx43 = getelementptr inbounds [2000 x float]* @D, i32 0, i64 %idxprom38
-  %7 = load float* %arrayidx43, align 4
+  %arrayidx39 = getelementptr inbounds [2000 x float], [2000 x float]* @C, i32 0, i64 %idxprom38
+  %6 = load float, float* %arrayidx39, align 4
+  %arrayidx43 = getelementptr inbounds [2000 x float], [2000 x float]* @D, i32 0, i64 %idxprom38
+  %7 = load float, float* %arrayidx43, align 4
   %add44 = fadd float %6, %7
   store float %add44, float* %arrayidx39, align 4
   ret void
@@ -142,13 +142,13 @@ for.body:                                         ; preds = %for.body.lr.ph, %fo
   %1 = phi double [ 0.000000e+00, %for.body.lr.ph ], [ %add7, %for.body ]
   %mul = mul nsw i32 %0, 2
   %idxprom = sext i32 %mul to i64
-  %arrayidx = getelementptr inbounds double* %A, i64 %idxprom
-  %2 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 %idxprom
+  %2 = load double, double* %arrayidx, align 8
   %mul1 = fmul double 7.000000e+00, %2
   %add = add nsw i32 %mul, 1
   %idxprom3 = sext i32 %add to i64
-  %arrayidx4 = getelementptr inbounds double* %A, i64 %idxprom3
-  %3 = load double* %arrayidx4, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %A, i64 %idxprom3
+  %3 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double 7.000000e+00, %3
   %add6 = fadd double %mul1, %mul5
   %add7 = fadd double %1, %add6
diff --git a/test/Transforms/SLPVectorizer/X86/continue_vectorizing.ll b/test/Transforms/SLPVectorizer/X86/continue_vectorizing.ll
index ed225743a6e2..ecae70ecc918 100644
--- a/test/Transforms/SLPVectorizer/X86/continue_vectorizing.ll
+++ b/test/Transforms/SLPVectorizer/X86/continue_vectorizing.ll
@@ -9,21 +9,21 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: ret
 define void @test1(double* %a, double* %b, double* %c, double* %d) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   %0 = bitcast double* %a to <4 x i32>*
-  %1 = load <4 x i32>* %0, align 8
+  %1 = load <4 x i32>, <4 x i32>* %0, align 8
   %2 = bitcast double* %b to <4 x i32>*
-  %3 = load <4 x i32>* %2, align 8
+  %3 = load <4 x i32>, <4 x i32>* %2, align 8
   %4 = mul <4 x i32> %1, %3
   %5 = bitcast double* %d to <4 x i32>*
   store <4 x i32> %4, <4 x i32>* %5, align 8
diff --git a/test/Transforms/SLPVectorizer/X86/crash_7zip.ll b/test/Transforms/SLPVectorizer/X86/crash_7zip.ll
index 51b1c08fb36f..54e5d5a53080 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_7zip.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_7zip.ll
@@ -8,8 +8,8 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define fastcc void @LzmaDec_DecodeReal2(%struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334* %p) {
 entry:
-  %range20.i = getelementptr inbounds %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334* %p, i64 0, i32 4
-  %code21.i = getelementptr inbounds %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334* %p, i64 0, i32 5
+  %range20.i = getelementptr inbounds %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334, %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334* %p, i64 0, i32 4
+  %code21.i = getelementptr inbounds %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334, %struct.CLzmaDec.1.28.55.82.103.124.145.166.181.196.229.259.334* %p, i64 0, i32 5
   br label %do.body66.i
 
 do.body66.i:                                      ; preds = %do.cond.i, %entry
diff --git a/test/Transforms/SLPVectorizer/X86/crash_binaryop.ll b/test/Transforms/SLPVectorizer/X86/crash_binaryop.ll
index dc99366e1da7..9046c3562821 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_binaryop.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_binaryop.ll
@@ -7,7 +7,7 @@ target triple = "x86_64-apple-darwin13.3.0"
 
 define i32 @fn1() {
 entry:
-  %init = load double* @a, align 8
+  %init = load double, double* @a, align 8
   br label %loop
 
 loop:
diff --git a/test/Transforms/SLPVectorizer/X86/crash_bullet.ll b/test/Transforms/SLPVectorizer/X86/crash_bullet.ll
index 389892115ced..1bad671fd824 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_bullet.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_bullet.ll
@@ -13,8 +13,8 @@ if.then:                                          ; preds = %entry
   ret void
 
 if.else:                                          ; preds = %entry
-  %m_numConstraintRows4 = getelementptr inbounds %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960"* %info, i64 0, i32 0
-  %nub5 = getelementptr inbounds %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960"* %info, i64 0, i32 1
+  %m_numConstraintRows4 = getelementptr inbounds %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960", %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960"* %info, i64 0, i32 0
+  %nub5 = getelementptr inbounds %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960", %"struct.btTypedConstraint::btConstraintInfo1.17.157.357.417.477.960"* %info, i64 0, i32 1
   br i1 undef, label %land.lhs.true.i.1, label %if.then7.1
 
 land.lhs.true.i.1:                                ; preds = %if.else
@@ -43,9 +43,9 @@ for.inc.1:                                        ; preds = %if.then7.1, %land.l
 
 define void @_ZN30GIM_TRIANGLE_CALCULATION_CACHE18triangle_collisionERK9btVector3S2_S2_fS2_S2_S2_fR25GIM_TRIANGLE_CONTACT_DATA(%class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332* %this) {
 entry:
-  %arrayidx26 = getelementptr inbounds %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332* %this, i64 0, i32 2, i64 0, i32 0, i64 1
-  %arrayidx36 = getelementptr inbounds %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332* %this, i64 0, i32 2, i64 0, i32 0, i64 2
-  %0 = load float* %arrayidx36, align 4
+  %arrayidx26 = getelementptr inbounds %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332, %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332* %this, i64 0, i32 2, i64 0, i32 0, i64 1
+  %arrayidx36 = getelementptr inbounds %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332, %class.GIM_TRIANGLE_CALCULATION_CACHE.9.34.69.94.119.144.179.189.264.284.332* %this, i64 0, i32 2, i64 0, i32 0, i64 2
+  %0 = load float, float* %arrayidx36, align 4
   %add587 = fadd float undef, undef
   %sub600 = fsub float %add587, undef
   store float %sub600, float* undef, align 4
diff --git a/test/Transforms/SLPVectorizer/X86/crash_bullet3.ll b/test/Transforms/SLPVectorizer/X86/crash_bullet3.ll
index 25c65457946b..810276929530 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_bullet3.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_bullet3.ll
@@ -59,9 +59,9 @@ if.end332:                                        ; preds = %if.then329, %if.end
   %dy276.1 = phi float [ undef, %if.then329 ], [ undef, %if.end327 ], [ 0x3F847AE140000000, %if.then291 ]
   %sub334 = fsub float %add294, %dx272.1
   %sub338 = fsub float %add297, %dy276.1
-  %arrayidx.i.i606 = getelementptr inbounds %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113* %vertices, i64 0, i32 0, i64 0
+  %arrayidx.i.i606 = getelementptr inbounds %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113, %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113* %vertices, i64 0, i32 0, i64 0
   store float %sub334, float* %arrayidx.i.i606, align 4
-  %arrayidx3.i607 = getelementptr inbounds %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113* %vertices, i64 0, i32 0, i64 1
+  %arrayidx3.i607 = getelementptr inbounds %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113, %class.btVector3.23.221.463.485.507.573.595.683.727.749.815.837.991.1585.1607.1629.1651.1849.2047.2069.2091.2113* %vertices, i64 0, i32 0, i64 1
   store float %sub338, float* %arrayidx3.i607, align 4
   br label %return
 
diff --git a/test/Transforms/SLPVectorizer/X86/crash_cmpop.ll b/test/Transforms/SLPVectorizer/X86/crash_cmpop.ll
new file mode 100644
index 000000000000..f10c8626d417
--- /dev/null
+++ b/test/Transforms/SLPVectorizer/X86/crash_cmpop.ll
@@ -0,0 +1,56 @@
+; RUN: opt < %s -basicaa -slp-vectorizer -S
+
+target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-apple-macosx10.10.0"
+
+define void @testfunc(float* nocapture %dest, float* nocapture readonly %src) {
+entry:
+  br label %for.body
+
+for.body:
+  %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
+  %acc1.056 = phi float [ 0.000000e+00, %entry ], [ %add13, %for.body ]
+  %s1.055 = phi float [ 0.000000e+00, %entry ], [ %cond.i40, %for.body ]
+  %s0.054 = phi float [ 0.000000e+00, %entry ], [ %cond.i44, %for.body ]
+  %arrayidx = getelementptr inbounds float, float* %src, i64 %indvars.iv
+  %0 = load float, float* %arrayidx, align 4
+  %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
+  %arrayidx2 = getelementptr inbounds float, float* %dest, i64 %indvars.iv
+  store float %acc1.056, float* %arrayidx2, align 4
+  %add = fadd float %s0.054, %0
+  %add3 = fadd float %s1.055, %0
+  %mul = fmul float %s0.054, 0.000000e+00
+  %add4 = fadd float %mul, %add3
+  %mul5 = fmul float %s1.055, 0.000000e+00
+  %add6 = fadd float %mul5, %add
+  %cmp.i = fcmp olt float %add6, 1.000000e+00
+  %cond.i = select i1 %cmp.i, float %add6, float 1.000000e+00
+  %cmp.i51 = fcmp olt float %cond.i, -1.000000e+00
+  %cmp.i49 = fcmp olt float %add4, 1.000000e+00
+  %cond.i50 = select i1 %cmp.i49, float %add4, float 1.000000e+00
+  %cmp.i47 = fcmp olt float %cond.i50, -1.000000e+00
+  %cond.i.op = fmul float %cond.i, 0.000000e+00
+  %mul10 = select i1 %cmp.i51, float -0.000000e+00, float %cond.i.op
+  %cond.i50.op = fmul float %cond.i50, 0.000000e+00
+  %mul11 = select i1 %cmp.i47, float -0.000000e+00, float %cond.i50.op
+  %add13 = fadd float %mul10, %mul11
+
+  ; The SLPVectorizer crashed in vectorizeChainsInBlock() because it tried
+  ; to access the second operand of the following cmp after the cmp itself
+  ; was already vectorized and deleted.
+  %cmp.i45 = fcmp olt float %add13, 1.000000e+00
+
+  %cond.i46 = select i1 %cmp.i45, float %add13, float 1.000000e+00
+  %cmp.i43 = fcmp olt float %cond.i46, -1.000000e+00
+  %cond.i44 = select i1 %cmp.i43, float -1.000000e+00, float %cond.i46
+  %cmp.i41 = fcmp olt float %mul11, 1.000000e+00
+  %cond.i42 = select i1 %cmp.i41, float %mul11, float 1.000000e+00
+  %cmp.i39 = fcmp olt float %cond.i42, -1.000000e+00
+  %cond.i40 = select i1 %cmp.i39, float -1.000000e+00, float %cond.i42
+  %exitcond = icmp eq i64 %indvars.iv.next, 32
+  br i1 %exitcond, label %for.end, label %for.body
+
+for.end:
+  ret void
+}
+
diff --git a/test/Transforms/SLPVectorizer/X86/crash_dequeue.ll b/test/Transforms/SLPVectorizer/X86/crash_dequeue.ll
index ce0159071c60..28b7aa3c4de3 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_dequeue.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_dequeue.ll
@@ -7,13 +7,13 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; Function Attrs: nounwind ssp uwtable
 define void @_ZSt6uniqueISt15_Deque_iteratorIdRdPdEET_S4_S4_(%"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__first, %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* nocapture %__last) {
 entry:
-  %_M_cur2.i.i = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__first, i64 0, i32 0
-  %0 = load double** %_M_cur2.i.i, align 8
-  %_M_first3.i.i = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__first, i64 0, i32 1
-  %_M_cur2.i.i81 = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__last, i64 0, i32 0
-  %1 = load double** %_M_cur2.i.i81, align 8
-  %_M_first3.i.i83 = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__last, i64 0, i32 1
-  %2 = load double** %_M_first3.i.i83, align 8
+  %_M_cur2.i.i = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731", %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__first, i64 0, i32 0
+  %0 = load double*, double** %_M_cur2.i.i, align 8
+  %_M_first3.i.i = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731", %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__first, i64 0, i32 1
+  %_M_cur2.i.i81 = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731", %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__last, i64 0, i32 0
+  %1 = load double*, double** %_M_cur2.i.i81, align 8
+  %_M_first3.i.i83 = getelementptr inbounds %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731", %"struct.std::_Deque_iterator.4.157.174.208.259.276.344.731"* %__last, i64 0, i32 1
+  %2 = load double*, double** %_M_first3.i.i83, align 8
   br i1 undef, label %_ZSt13adjacent_findISt15_Deque_iteratorIdRdPdEET_S4_S4_.exit, label %while.cond.i.preheader
 
 while.cond.i.preheader:                           ; preds = %entry
diff --git a/test/Transforms/SLPVectorizer/X86/crash_gep.ll b/test/Transforms/SLPVectorizer/X86/crash_gep.ll
index dd4034c116e4..bd1e8f7cc19d 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_gep.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_gep.ll
@@ -8,10 +8,10 @@ target triple = "x86_64-unknown-linux-gnu"
 ; Function Attrs: nounwind uwtable
 define i32 @fn1() {
 entry:
-  %0 = load i64** @a, align 8
-  %add.ptr = getelementptr inbounds i64* %0, i64 1
+  %0 = load i64*, i64** @a, align 8
+  %add.ptr = getelementptr inbounds i64, i64* %0, i64 1
   %1 = ptrtoint i64* %add.ptr to i64
-  %arrayidx = getelementptr inbounds i64* %0, i64 2
+  %arrayidx = getelementptr inbounds i64, i64* %0, i64 2
   store i64 %1, i64* %arrayidx, align 8
   %2 = ptrtoint i64* %arrayidx to i64
   store i64 %2, i64* %add.ptr, align 8
diff --git a/test/Transforms/SLPVectorizer/X86/crash_lencod.ll b/test/Transforms/SLPVectorizer/X86/crash_lencod.ll
index c02e1fa607d5..70b13fd75f14 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_lencod.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_lencod.ll
@@ -68,23 +68,23 @@ if.end103:                                        ; preds = %if.else, %if.then88
 define void @intrapred_luma() {
 entry:
   %conv153 = trunc i32 undef to i16
-  %arrayidx154 = getelementptr inbounds [13 x i16]* undef, i64 0, i64 12
+  %arrayidx154 = getelementptr inbounds [13 x i16], [13 x i16]* undef, i64 0, i64 12
   store i16 %conv153, i16* %arrayidx154, align 8
-  %arrayidx155 = getelementptr inbounds [13 x i16]* undef, i64 0, i64 11
+  %arrayidx155 = getelementptr inbounds [13 x i16], [13 x i16]* undef, i64 0, i64 11
   store i16 %conv153, i16* %arrayidx155, align 2
-  %arrayidx156 = getelementptr inbounds [13 x i16]* undef, i64 0, i64 10
+  %arrayidx156 = getelementptr inbounds [13 x i16], [13 x i16]* undef, i64 0, i64 10
   store i16 %conv153, i16* %arrayidx156, align 4
   ret void
 }
 
 define fastcc void @dct36(double* %inbuf) {
 entry:
-  %arrayidx41 = getelementptr inbounds double* %inbuf, i64 2
-  %arrayidx44 = getelementptr inbounds double* %inbuf, i64 1
-  %0 = load double* %arrayidx44, align 8
+  %arrayidx41 = getelementptr inbounds double, double* %inbuf, i64 2
+  %arrayidx44 = getelementptr inbounds double, double* %inbuf, i64 1
+  %0 = load double, double* %arrayidx44, align 8
   %add46 = fadd double %0, undef
   store double %add46, double* %arrayidx41, align 8
-  %1 = load double* %inbuf, align 8
+  %1 = load double, double* %inbuf, align 8
   %add49 = fadd double %1, %0
   store double %add49, double* %arrayidx44, align 8
   ret void
diff --git a/test/Transforms/SLPVectorizer/X86/crash_mandeltext.ll b/test/Transforms/SLPVectorizer/X86/crash_mandeltext.ll
index d6915e2dc5d6..f82343fb433f 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_mandeltext.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_mandeltext.ll
@@ -55,18 +55,18 @@ for.end48:                                        ; preds = %for.end44
 
 define void @zot(%struct.hoge* %arg) {
 bb:
-  %tmp = load double* undef, align 8
+  %tmp = load double, double* undef, align 8
   %tmp1 = fsub double %tmp, undef
-  %tmp2 = load double* undef, align 8
+  %tmp2 = load double, double* undef, align 8
   %tmp3 = fsub double %tmp2, undef
   %tmp4 = fmul double %tmp3, undef
   %tmp5 = fmul double %tmp3, undef
   %tmp6 = fsub double %tmp5, undef
-  %tmp7 = getelementptr inbounds %struct.hoge* %arg, i64 0, i32 1
+  %tmp7 = getelementptr inbounds %struct.hoge, %struct.hoge* %arg, i64 0, i32 1
   store double %tmp6, double* %tmp7, align 8
   %tmp8 = fmul double %tmp1, undef
   %tmp9 = fsub double %tmp8, undef
-  %tmp10 = getelementptr inbounds %struct.hoge* %arg, i64 0, i32 2
+  %tmp10 = getelementptr inbounds %struct.hoge, %struct.hoge* %arg, i64 0, i32 2
   store double %tmp9, double* %tmp10, align 8
   br i1 undef, label %bb11, label %bb12
 
@@ -86,8 +86,8 @@ bb14:                                             ; preds = %bb12, %bb11
 
 define void @rc4_crypt(%struct.rc4_state.0.24* nocapture %s) {
 entry:
-  %x1 = getelementptr inbounds %struct.rc4_state.0.24* %s, i64 0, i32 0
-  %y2 = getelementptr inbounds %struct.rc4_state.0.24* %s, i64 0, i32 1
+  %x1 = getelementptr inbounds %struct.rc4_state.0.24, %struct.rc4_state.0.24* %s, i64 0, i32 0
+  %y2 = getelementptr inbounds %struct.rc4_state.0.24, %struct.rc4_state.0.24* %s, i64 0, i32 1
   br i1 undef, label %for.body, label %for.end
 
 for.body:                                         ; preds = %for.body, %entry
diff --git a/test/Transforms/SLPVectorizer/X86/crash_netbsd_decompress.ll b/test/Transforms/SLPVectorizer/X86/crash_netbsd_decompress.ll
index 8da3c34a0279..e1df98d83bdb 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_netbsd_decompress.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_netbsd_decompress.ll
@@ -13,14 +13,14 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define i32 @fn1() {
 entry:
-  %0 = load i32* getelementptr inbounds (%struct.DState* @b, i32 0, i32 0), align 4
-  %1 = load i32* getelementptr inbounds (%struct.DState* @b, i32 0, i32 1), align 4
-  %2 = load i32* @d, align 4
+  %0 = load i32, i32* getelementptr inbounds (%struct.DState, %struct.DState* @b, i32 0, i32 0), align 4
+  %1 = load i32, i32* getelementptr inbounds (%struct.DState, %struct.DState* @b, i32 0, i32 1), align 4
+  %2 = load i32, i32* @d, align 4
   %cond = icmp eq i32 %2, 0
   br i1 %cond, label %sw.bb, label %save_state_and_return
 
 sw.bb:                                            ; preds = %entry
-  %3 = load i32* @c, align 4
+  %3 = load i32, i32* @c, align 4
   %and = and i32 %3, 7
   store i32 %and, i32* @a, align 4
   switch i32 %and, label %if.end [
@@ -34,8 +34,8 @@ if.end:                                           ; preds = %sw.bb
 save_state_and_return:                            ; preds = %sw.bb, %sw.bb, %if.end, %entry
   %t.0 = phi i32 [ 0, %if.end ], [ %0, %entry ], [ %0, %sw.bb ], [ %0, %sw.bb ]
   %f.0 = phi i32 [ 0, %if.end ], [ %1, %entry ], [ 0, %sw.bb ], [ 0, %sw.bb ]
-  store i32 %t.0, i32* getelementptr inbounds (%struct.DState* @b, i32 0, i32 0), align 4
-  store i32 %f.0, i32* getelementptr inbounds (%struct.DState* @b, i32 0, i32 1), align 4
+  store i32 %t.0, i32* getelementptr inbounds (%struct.DState, %struct.DState* @b, i32 0, i32 0), align 4
+  store i32 %f.0, i32* getelementptr inbounds (%struct.DState, %struct.DState* @b, i32 0, i32 1), align 4
   ret i32 undef
 }
 
diff --git a/test/Transforms/SLPVectorizer/X86/crash_scheduling.ll b/test/Transforms/SLPVectorizer/X86/crash_scheduling.ll
index e6cc0f783d79..14a8b35f550a 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_scheduling.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_scheduling.ll
@@ -25,11 +25,11 @@ for.body:
   %p3.addr.0258 = phi double [ %add, %bb1 ], [ %add28, %for.body ]
   %vecinit.i.i237 = insertelement <2 x double> undef, double %t.0259, i32 0
   %x13 = tail call i32 @_xfn(<2 x double> %vecinit.i.i237) #2
-  %arrayidx = getelementptr inbounds [256 x i32]* %tab1, i64 0, i64 %indvars.iv266
+  %arrayidx = getelementptr inbounds [256 x i32], [256 x i32]* %tab1, i64 0, i64 %indvars.iv266
   store i32 %x13, i32* %arrayidx, align 4, !tbaa !4
   %vecinit.i.i = insertelement <2 x double> undef, double %p3.addr.0258, i32 0
   %x14 = tail call i32 @_xfn(<2 x double> %vecinit.i.i) #2
-  %arrayidx26 = getelementptr inbounds [256 x i32]* %tab2, i64 0, i64 %indvars.iv266
+  %arrayidx26 = getelementptr inbounds [256 x i32], [256 x i32]* %tab2, i64 0, i64 %indvars.iv266
   store i32 %x14, i32* %arrayidx26, align 4, !tbaa !4
   %add27 = fadd double %mul19, %t.0259
   %add28 = fadd double %mul21, %p3.addr.0258
diff --git a/test/Transforms/SLPVectorizer/X86/crash_sim4b1.ll b/test/Transforms/SLPVectorizer/X86/crash_sim4b1.ll
index 05415456cf0f..5a576c2a7222 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_sim4b1.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_sim4b1.ll
@@ -34,13 +34,13 @@ land.rhs.lr.ph:                                   ; preds = %if.then17
   unreachable
 
 if.end98:                                         ; preds = %if.then17
-  %from299 = getelementptr inbounds %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171* undef, i64 0, i32 1
+  %from299 = getelementptr inbounds %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171, %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171* undef, i64 0, i32 1
   br i1 undef, label %land.lhs.true167, label %if.then103
 
 if.then103:                                       ; preds = %if.end98
   %.sub100 = select i1 undef, i32 250, i32 undef
   %mul114 = shl nsw i32 %.sub100, 2
-  %from1115 = getelementptr inbounds %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171* undef, i64 0, i32 0
+  %from1115 = getelementptr inbounds %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171, %struct._exon_t.12.103.220.363.480.649.740.857.1039.1065.1078.1091.1117.1130.1156.1169.1195.1221.1234.1286.1299.1312.1338.1429.1455.1468.1494.1520.1884.1897.1975.2066.2105.2170.2171* undef, i64 0, i32 0
   %cond125 = select i1 undef, i32 undef, i32 %mul114
   br label %for.cond.i
 
diff --git a/test/Transforms/SLPVectorizer/X86/crash_smallpt.ll b/test/Transforms/SLPVectorizer/X86/crash_smallpt.ll
index 915c41bb9c59..273584c11a9c 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_smallpt.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_smallpt.ll
@@ -21,10 +21,10 @@ invoke.cont:                                      ; preds = %invoke.cont, %cond.
   br i1 undef, label %arrayctor.cont, label %invoke.cont
 
 arrayctor.cont:                                   ; preds = %invoke.cont
-  %agg.tmp99208.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 0, i32 0
-  %agg.tmp99208.sroa.1.8.idx388 = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 0, i32 1
-  %agg.tmp101211.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 1, i32 0
-  %agg.tmp101211.sroa.1.8.idx390 = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 1, i32 1
+  %agg.tmp99208.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416, %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 0, i32 0
+  %agg.tmp99208.sroa.1.8.idx388 = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416, %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 0, i32 1
+  %agg.tmp101211.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416, %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 1, i32 0
+  %agg.tmp101211.sroa.1.8.idx390 = getelementptr inbounds %struct.Ray.5.11.53.113.119.137.149.185.329.389.416, %struct.Ray.5.11.53.113.119.137.149.185.329.389.416* undef, i64 0, i32 1, i32 1
   br label %for.cond36.preheader
 
 for.cond36.preheader:                             ; preds = %_Z5clampd.exit.1, %arrayctor.cont
@@ -89,9 +89,9 @@ if.then38:                                        ; preds = %entry
   %add4.i698 = fadd double undef, %add4.i719
   %mul.i.i679 = fmul double undef, %add.i695
   %mul2.i.i680 = fmul double undef, %add4.i698
-  %agg.tmp74663.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601* undef, i64 0, i32 1, i32 0
+  %agg.tmp74663.sroa.0.0.idx = getelementptr inbounds %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601, %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601* undef, i64 0, i32 1, i32 0
   store double %mul.i.i679, double* %agg.tmp74663.sroa.0.0.idx, align 8
-  %agg.tmp74663.sroa.1.8.idx943 = getelementptr inbounds %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601* undef, i64 0, i32 1, i32 1
+  %agg.tmp74663.sroa.1.8.idx943 = getelementptr inbounds %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601, %struct.Ray.5.11.53.95.137.191.197.203.239.257.263.269.275.281.287.293.383.437.443.455.461.599.601* undef, i64 0, i32 1, i32 1
   store double %mul2.i.i680, double* %agg.tmp74663.sroa.1.8.idx943, align 8
   br label %return
 
diff --git a/test/Transforms/SLPVectorizer/X86/crash_vectorizeTree.ll b/test/Transforms/SLPVectorizer/X86/crash_vectorizeTree.ll
index 9f1fb71e096f..45ca99a3ea16 100644
--- a/test/Transforms/SLPVectorizer/X86/crash_vectorizeTree.ll
+++ b/test/Transforms/SLPVectorizer/X86/crash_vectorizeTree.ll
@@ -17,12 +17,12 @@ target triple = "x86_64-apple-macosx10.9.0"
 
 ;define fastcc void @bar() {
 define void @bar() {
-  %1 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 0
-  %2 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 1
-  %3 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 0
-  %4 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 1
-  %5 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 0
-  %6 = getelementptr inbounds %0* undef, i64 0, i32 1, i32 1
+  %1 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 0
+  %2 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 1
+  %3 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 0
+  %4 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 1
+  %5 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 0
+  %6 = getelementptr inbounds %0, %0* undef, i64 0, i32 1, i32 1
   br label %7
 
 ; <label>:7                                       ; preds = %18, %17, %17, %0
@@ -30,8 +30,8 @@ define void @bar() {
   %9 = phi double [ 1.800000e+01, %0 ], [ %10, %18 ], [ %10, %17 ], [ %10, %17 ]
   store double %9, double* %1, align 8
   store double %8, double* %2, align 8
-  %10 = load double* %3, align 8
-  %11 = load double* %4, align 8
+  %10 = load double, double* %3, align 8
+  %11 = load double, double* %4, align 8
   br i1 undef, label %12, label %13
 
 ; <label>:12                                      ; preds = %7
diff --git a/test/Transforms/SLPVectorizer/X86/cross_block_slp.ll b/test/Transforms/SLPVectorizer/X86/cross_block_slp.ll
index 06c4b524ee95..eac20a006450 100644
--- a/test/Transforms/SLPVectorizer/X86/cross_block_slp.ll
+++ b/test/Transforms/SLPVectorizer/X86/cross_block_slp.ll
@@ -26,26 +26,26 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK: ret
 define i32 @foo(double* nocapture %A, float* nocapture %B, i32 %g) {
 entry:
-  %0 = load float* %B, align 4
-  %arrayidx1 = getelementptr inbounds float* %B, i64 1
-  %1 = load float* %arrayidx1, align 4
+  %0 = load float, float* %B, align 4
+  %arrayidx1 = getelementptr inbounds float, float* %B, i64 1
+  %1 = load float, float* %arrayidx1, align 4
   %add = fadd float %0, 5.000000e+00
   %add2 = fadd float %1, 8.000000e+00
   %tobool = icmp eq i32 %g, 0
   br i1 %tobool, label %if.end, label %if.then
 
 if.then:
-  %call = tail call i32 (...)* @bar()
+  %call = tail call i32 (...) @bar()
   br label %if.end
 
 if.end:
   %conv = fpext float %add to double
-  %2 = load double* %A, align 8
+  %2 = load double, double* %A, align 8
   %add4 = fadd double %conv, %2
   store double %add4, double* %A, align 8
   %conv5 = fpext float %add2 to double
-  %arrayidx6 = getelementptr inbounds double* %A, i64 1
-  %3 = load double* %arrayidx6, align 8
+  %arrayidx6 = getelementptr inbounds double, double* %A, i64 1
+  %3 = load double, double* %arrayidx6, align 8
   %add7 = fadd double %conv5, %3
   store double %add7, double* %arrayidx6, align 8
   ret i32 undef
diff --git a/test/Transforms/SLPVectorizer/X86/cse.ll b/test/Transforms/SLPVectorizer/X86/cse.ll
index d2ad7eb1a82a..9f56e2195991 100644
--- a/test/Transforms/SLPVectorizer/X86/cse.ll
+++ b/test/Transforms/SLPVectorizer/X86/cse.ll
@@ -21,23 +21,23 @@ target triple = "i386-apple-macosx10.8.0"
 
 define i32 @test(double* nocapture %G) {
 entry:
-  %arrayidx = getelementptr inbounds double* %G, i64 5
-  %0 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %G, i64 5
+  %0 = load double, double* %arrayidx, align 8
   %mul = fmul double %0, 4.000000e+00
   %add = fadd double %mul, 1.000000e+00
   store double %add, double* %G, align 8
-  %arrayidx2 = getelementptr inbounds double* %G, i64 6
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %G, i64 6
+  %1 = load double, double* %arrayidx2, align 8
   %mul3 = fmul double %1, 3.000000e+00
   %add4 = fadd double %mul3, 6.000000e+00
-  %arrayidx5 = getelementptr inbounds double* %G, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %G, i64 1
   store double %add4, double* %arrayidx5, align 8
   %add8 = fadd double %mul, 7.000000e+00
-  %arrayidx9 = getelementptr inbounds double* %G, i64 2
+  %arrayidx9 = getelementptr inbounds double, double* %G, i64 2
   store double %add8, double* %arrayidx9, align 8
   %mul11 = fmul double %1, 4.000000e+00
   %add12 = fadd double %mul11, 8.000000e+00
-  %arrayidx13 = getelementptr inbounds double* %G, i64 3
+  %arrayidx13 = getelementptr inbounds double, double* %G, i64 3
   store double %add12, double* %arrayidx13, align 8
   ret i32 undef
 }
@@ -55,26 +55,26 @@ entry:
 ;CHECK: ret
 define i32 @foo(double* nocapture %A, i32 %n) {
 entry:
-  %0 = load double* %A, align 8
+  %0 = load double, double* %A, align 8
   %mul = fmul double %0, 7.900000e+00
   %conv = sitofp i32 %n to double
   %mul1 = fmul double %conv, %mul
   %add = fadd double %mul1, 6.000000e+00
   store double %add, double* %A, align 8
-  %arrayidx3 = getelementptr inbounds double* %A, i64 1
-  %1 = load double* %arrayidx3, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %A, i64 1
+  %1 = load double, double* %arrayidx3, align 8
   %mul4 = fmul double %1, 7.700000e+00
   %mul6 = fmul double %conv, %mul4
   %add7 = fadd double %mul6, 2.000000e+00
   store double %add7, double* %arrayidx3, align 8
-  %arrayidx9 = getelementptr inbounds double* %A, i64 2
-  %2 = load double* %arrayidx9, align 8
+  %arrayidx9 = getelementptr inbounds double, double* %A, i64 2
+  %2 = load double, double* %arrayidx9, align 8
   %mul10 = fmul double %2, 7.600000e+00
   %mul12 = fmul double %conv, %mul10
   %add13 = fadd double %mul12, 3.000000e+00
   store double %add13, double* %arrayidx9, align 8
-  %arrayidx15 = getelementptr inbounds double* %A, i64 3
-  %3 = load double* %arrayidx15, align 8
+  %arrayidx15 = getelementptr inbounds double, double* %A, i64 3
+  %3 = load double, double* %arrayidx15, align 8
   %mul16 = fmul double %3, 7.400000e+00
   %mul18 = fmul double %conv, %mul16
   %add19 = fadd double %mul18, 4.000000e+00
@@ -101,31 +101,31 @@ entry:
 ; CHECK: ret
 define i32 @test2(double* nocapture %G, i32 %k) {
   %1 = icmp eq i32 %k, 0
-  %2 = getelementptr inbounds double* %G, i64 5
-  %3 = load double* %2, align 8
+  %2 = getelementptr inbounds double, double* %G, i64 5
+  %3 = load double, double* %2, align 8
   %4 = fmul double %3, 4.000000e+00
   br i1 %1, label %12, label %5
 
 ; <label>:5                                       ; preds = %0
   %6 = fadd double %4, 1.000000e+00
   store double %6, double* %G, align 8
-  %7 = getelementptr inbounds double* %G, i64 6
-  %8 = load double* %7, align 8
+  %7 = getelementptr inbounds double, double* %G, i64 6
+  %8 = load double, double* %7, align 8
   %9 = fmul double %8, 3.000000e+00
   %10 = fadd double %9, 6.000000e+00
-  %11 = getelementptr inbounds double* %G, i64 1
+  %11 = getelementptr inbounds double, double* %G, i64 1
   store double %10, double* %11, align 8
   br label %20
 
 ; <label>:12                                      ; preds = %0
   %13 = fadd double %4, 7.000000e+00
-  %14 = getelementptr inbounds double* %G, i64 2
+  %14 = getelementptr inbounds double, double* %G, i64 2
   store double %13, double* %14, align 8
-  %15 = getelementptr inbounds double* %G, i64 6
-  %16 = load double* %15, align 8
+  %15 = getelementptr inbounds double, double* %G, i64 6
+  %16 = load double, double* %15, align 8
   %17 = fmul double %16, 3.000000e+00
   %18 = fadd double %17, 8.000000e+00
-  %19 = getelementptr inbounds double* %G, i64 3
+  %19 = getelementptr inbounds double, double* %G, i64 3
   store double %18, double* %19, align 8
   br label %20
 
@@ -147,26 +147,26 @@ define i32 @test2(double* nocapture %G, i32 %k) {
 ;CHECK: ret
 define i32 @foo4(double* nocapture %A, i32 %n) {
 entry:
-  %0 = load double* %A, align 8
+  %0 = load double, double* %A, align 8
   %mul = fmul double %0, 7.900000e+00
   %conv = sitofp i32 %n to double
   %mul1 = fmul double %conv, %mul
   %add = fadd double %mul1, 6.000000e+00
   store double %add, double* %A, align 8
-  %arrayidx3 = getelementptr inbounds double* %A, i64 1
-  %1 = load double* %arrayidx3, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %A, i64 1
+  %1 = load double, double* %arrayidx3, align 8
   %mul4 = fmul double %1, 7.900000e+00
   %mul6 = fmul double %conv, %mul4
   %add7 = fadd double %mul6, 6.000000e+00
   store double %add7, double* %arrayidx3, align 8
-  %arrayidx9 = getelementptr inbounds double* %A, i64 2
-  %2 = load double* %arrayidx9, align 8
+  %arrayidx9 = getelementptr inbounds double, double* %A, i64 2
+  %2 = load double, double* %arrayidx9, align 8
   %mul10 = fmul double %2, 7.900000e+00
   %mul12 = fmul double %conv, %mul10
   %add13 = fadd double %mul12, 6.000000e+00
   store double %add13, double* %arrayidx9, align 8
-  %arrayidx15 = getelementptr inbounds double* %A, i64 3
-  %3 = load double* %arrayidx15, align 8
+  %arrayidx15 = getelementptr inbounds double, double* %A, i64 3
+  %3 = load double, double* %arrayidx15, align 8
   %mul16 = fmul double %3, 7.900000e+00
   %mul18 = fmul double %conv, %mul16
   %add19 = fadd double %mul18, 6.000000e+00
@@ -189,24 +189,24 @@ entry:
 ;CHECK: ret
 define i32 @partial_mrg(double* nocapture %A, i32 %n) {
 entry:
-  %0 = load double* %A, align 8
+  %0 = load double, double* %A, align 8
   %conv = sitofp i32 %n to double
   %mul = fmul double %conv, %0
   store double %mul, double* %A, align 8
-  %arrayidx2 = getelementptr inbounds double* %A, i64 1
-  %1 = load double* %arrayidx2, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %A, i64 1
+  %1 = load double, double* %arrayidx2, align 8
   %mul4 = fmul double %conv, %1
   store double %mul4, double* %arrayidx2, align 8
   %cmp = icmp slt i32 %n, 4
   br i1 %cmp, label %return, label %if.end
 
 if.end:                                           ; preds = %entry
-  %arrayidx7 = getelementptr inbounds double* %A, i64 2
-  %2 = load double* %arrayidx7, align 8
+  %arrayidx7 = getelementptr inbounds double, double* %A, i64 2
+  %2 = load double, double* %arrayidx7, align 8
   %mul9 = fmul double %conv, %2
   store double %mul9, double* %arrayidx7, align 8
-  %arrayidx11 = getelementptr inbounds double* %A, i64 3
-  %3 = load double* %arrayidx11, align 8
+  %arrayidx11 = getelementptr inbounds double, double* %A, i64 3
+  %3 = load double, double* %arrayidx11, align 8
   %add = add nsw i32 %n, 4
   %conv12 = sitofp i32 %add to double
   %mul13 = fmul double %conv12, %3
@@ -227,19 +227,19 @@ entry:
   br i1 undef, label %if.end13, label %if.end13
 
 sw.epilog7:                                       ; No predecessors!
-  %.in = getelementptr inbounds %class.B.53.55* %this, i64 0, i32 0, i32 1
-  %0 = load double* %.in, align 8
+  %.in = getelementptr inbounds %class.B.53.55, %class.B.53.55* %this, i64 0, i32 0, i32 1
+  %0 = load double, double* %.in, align 8
   %add = fadd double undef, 0.000000e+00
   %add6 = fadd double %add, %0
-  %1 = load double* @a, align 8
+  %1 = load double, double* @a, align 8
   %add8 = fadd double %1, 0.000000e+00
-  %_dy = getelementptr inbounds %class.B.53.55* %this, i64 0, i32 0, i32 2
-  %2 = load double* %_dy, align 8
+  %_dy = getelementptr inbounds %class.B.53.55, %class.B.53.55* %this, i64 0, i32 0, i32 2
+  %2 = load double, double* %_dy, align 8
   %add10 = fadd double %add8, %2
   br i1 undef, label %if.then12, label %if.end13
 
 if.then12:                                        ; preds = %sw.epilog7
-  %3 = load double* undef, align 8
+  %3 = load double, double* undef, align 8
   br label %if.end13
 
 if.end13:                                         ; preds = %if.then12, %sw.epilog7, %entry
diff --git a/test/Transforms/SLPVectorizer/X86/cycle_dup.ll b/test/Transforms/SLPVectorizer/X86/cycle_dup.ll
index bac2c3c0df34..0a4e961c2e88 100644
--- a/test/Transforms/SLPVectorizer/X86/cycle_dup.ll
+++ b/test/Transforms/SLPVectorizer/X86/cycle_dup.ll
@@ -23,15 +23,15 @@ target triple = "x86_64-apple-macosx10.9.0"
 ;CHECK-NEXT:ret i32 undef
 define i32 @foo(i32* nocapture %A) #0 {
 entry:
-  %0 = load i32* %A, align 4
-  %arrayidx1 = getelementptr inbounds i32* %A, i64 1
-  %1 = load i32* %arrayidx1, align 4
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 2
-  %2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %A, i64 3
-  %3 = load i32* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 13
-  %4 = load i32* %arrayidx4, align 4
+  %0 = load i32, i32* %A, align 4
+  %arrayidx1 = getelementptr inbounds i32, i32* %A, i64 1
+  %1 = load i32, i32* %arrayidx1, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 2
+  %2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i64 3
+  %3 = load i32, i32* %arrayidx3, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 13
+  %4 = load i32, i32* %arrayidx4, align 4
   %cmp24 = icmp sgt i32 %4, 0
   br i1 %cmp24, label %for.body, label %for.end
 
diff --git a/test/Transforms/SLPVectorizer/X86/debug_info.ll b/test/Transforms/SLPVectorizer/X86/debug_info.ll
index 21f51d7c6125..1d349fbc98b5 100644
--- a/test/Transforms/SLPVectorizer/X86/debug_info.ll
+++ b/test/Transforms/SLPVectorizer/X86/debug_info.ll
@@ -15,35 +15,35 @@ target triple = "x86_64-apple-macosx10.7.0"
 ;CHECK: @depth
 ;CHECK: getelementptr inbounds {{.*}}, !dbg ![[LOC:[0-9]+]]
 ;CHECK: bitcast double* {{.*}}, !dbg ![[LOC]]
-;CHECK: load <2 x double>* {{.*}}, !dbg ![[LOC]]
+;CHECK: load <2 x double>, <2 x double>* {{.*}}, !dbg ![[LOC]]
 ;CHECK: store <2 x double> {{.*}}, !dbg ![[LOC2:[0-9]+]]
 ;CHECK: ret
-;CHECK: ![[LOC]] = !MDLocation(line: 4, scope:
-;CHECK: ![[LOC2]] = !MDLocation(line: 7, scope:
+;CHECK: ![[LOC]] = !DILocation(line: 4, scope:
+;CHECK: ![[LOC2]] = !DILocation(line: 7, scope:
 
 define i32 @depth(double* nocapture %A, i32 %m) #0 {
 entry:
-  tail call void @llvm.dbg.value(metadata double* %A, i64 0, metadata !12, metadata !{}), !dbg !19
-  tail call void @llvm.dbg.value(metadata i32 %m, i64 0, metadata !13, metadata !{}), !dbg !19
-  tail call void @llvm.dbg.value(metadata i32 00, i64 0, metadata !14, metadata !{}), !dbg !21
-  tail call void @llvm.dbg.value(metadata i32 02, i64 0, metadata !15, metadata !{}), !dbg !21
-  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !16, metadata !{}), !dbg !23
+  tail call void @llvm.dbg.value(metadata double* %A, i64 0, metadata !12, metadata !DIExpression()), !dbg !19
+  tail call void @llvm.dbg.value(metadata i32 %m, i64 0, metadata !13, metadata !DIExpression()), !dbg !19
+  tail call void @llvm.dbg.value(metadata i32 00, i64 0, metadata !14, metadata !DIExpression()), !dbg !21
+  tail call void @llvm.dbg.value(metadata i32 02, i64 0, metadata !15, metadata !DIExpression()), !dbg !21
+  tail call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !16, metadata !DIExpression()), !dbg !23
   %cmp8 = icmp sgt i32 %m, 0, !dbg !23
   br i1 %cmp8, label %for.body.lr.ph, label %for.end, !dbg !23
 
 for.body.lr.ph:                                   ; preds = %entry
-  %arrayidx = getelementptr inbounds double* %A, i64 4, !dbg !24
-  %0 = load double* %arrayidx, align 8, !dbg !24
-  %arrayidx1 = getelementptr inbounds double* %A, i64 5, !dbg !29
-  %1 = load double* %arrayidx1, align 8, !dbg !29
+  %arrayidx = getelementptr inbounds double, double* %A, i64 4, !dbg !24
+  %0 = load double, double* %arrayidx, align 8, !dbg !24
+  %arrayidx1 = getelementptr inbounds double, double* %A, i64 5, !dbg !29
+  %1 = load double, double* %arrayidx1, align 8, !dbg !29
   br label %for.end, !dbg !23
 
 for.end:                                          ; preds = %for.body.lr.ph, %entry
   %y1.0.lcssa = phi double [ %1, %for.body.lr.ph ], [ 1.000000e+00, %entry ]
   %y0.0.lcssa = phi double [ %0, %for.body.lr.ph ], [ 0.000000e+00, %entry ]
-  %arrayidx2 = getelementptr inbounds double* %A, i64 8, !dbg !30
+  %arrayidx2 = getelementptr inbounds double, double* %A, i64 8, !dbg !30
   store double %y0.0.lcssa, double* %arrayidx2, align 8, !dbg !30
-  %arrayidx3 = getelementptr inbounds double* %A, i64 9, !dbg !30
+  %arrayidx3 = getelementptr inbounds double, double* %A, i64 9, !dbg !30
   store double %y1.0.lcssa, double* %arrayidx3, align 8, !dbg !30
   ret i32 undef, !dbg !31
 }
@@ -57,33 +57,33 @@ attributes #1 = { nounwind readnone }
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!18, !32}
 
-!0 = !{!"0x11\0012\00clang version 3.4 (trunk 187335) (llvm/trunk 187335:187340M)\001\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [/Users/nadav/file.c] [DW_LANG_C99]
-!1 = !{!"file.c", !"/Users/nadav"}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.4 (trunk 187335) (llvm/trunk 187335:187340M)", isOptimized: true, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "file.c", directory: "/Users/nadav")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00depth\00depth\00\001\000\001\000\006\00256\001\001", !1, !5, !6, null, i32 (double*, i32)* @depth, null, null, !11} ; [ DW_TAG_subprogram ] [line 1] [def] [depth]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [/Users/nadav/file.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "depth", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 1, file: !1, scope: !5, type: !6, function: i32 (double*, i32)* @depth, variables: !11)
+!5 = !DIFile(filename: "file.c", directory: "/Users/nadav")
+!6 = !DISubroutineType(types: !7)
 !7 = !{!8, !9, !8}
-!8 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!9 = !{!"0xf\00\000\0064\0064\000\000", null, null, !10} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from double]
-!10 = !{!"0x24\00double\000\0064\0064\000\000\004", null, null} ; [ DW_TAG_base_type ] [double] [line 0, size 64, align 64, offset 0, enc DW_ATE_float]
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!9 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !10)
+!10 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
 !11 = !{!12, !13, !14, !15, !16}
-!12 = !{!"0x101\00A\0016777217\000", !4, !5, !9} ; [ DW_TAG_arg_variable ] [A] [line 1]
-!13 = !{!"0x101\00m\0033554433\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [m] [line 1]
-!14 = !{!"0x100\00y0\002\000", !4, !5, !10} ; [ DW_TAG_auto_variable ] [y0] [line 2]
-!15 = !{!"0x100\00y1\002\000", !4, !5, !10} ; [ DW_TAG_auto_variable ] [y1] [line 2]
-!16 = !{!"0x100\00i\003\000", !17, !5, !8} ; [ DW_TAG_auto_variable ] [i] [line 3]
-!17 = !{!"0xb\003\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [/Users/nadav/file.c]
+!12 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "A", line: 1, arg: 1, scope: !4, file: !5, type: !9)
+!13 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "m", line: 1, arg: 2, scope: !4, file: !5, type: !8)
+!14 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "y0", line: 2, scope: !4, file: !5, type: !10)
+!15 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "y1", line: 2, scope: !4, file: !5, type: !10)
+!16 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "i", line: 3, scope: !17, file: !5, type: !8)
+!17 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !4)
 !18 = !{i32 2, !"Dwarf Version", i32 2}
-!19 = !MDLocation(line: 1, scope: !4)
+!19 = !DILocation(line: 1, scope: !4)
 !20 = !{double 0.000000e+00}
-!21 = !MDLocation(line: 2, scope: !4)
+!21 = !DILocation(line: 2, scope: !4)
 !22 = !{double 1.000000e+00}
-!23 = !MDLocation(line: 3, scope: !17)
-!24 = !MDLocation(line: 4, scope: !25)
-!25 = !{!"0xb\003\000\001", !1, !17} ; [ DW_TAG_lexical_block ] [/Users/nadav/file.c]
-!29 = !MDLocation(line: 5, scope: !25)
-!30 = !MDLocation(line: 7, scope: !4)
-!31 = !MDLocation(line: 8, scope: !4)
-!32 = !{i32 1, !"Debug Info Version", i32 2}
+!23 = !DILocation(line: 3, scope: !17)
+!24 = !DILocation(line: 4, scope: !25)
+!25 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !17)
+!29 = !DILocation(line: 5, scope: !25)
+!30 = !DILocation(line: 7, scope: !4)
+!31 = !DILocation(line: 8, scope: !4)
+!32 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/SLPVectorizer/X86/diamond.ll b/test/Transforms/SLPVectorizer/X86/diamond.ll
index 5135a92a7bdb..4e2c02f6965a 100644
--- a/test/Transforms/SLPVectorizer/X86/diamond.ll
+++ b/test/Transforms/SLPVectorizer/X86/diamond.ll
@@ -18,24 +18,24 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: ret
 define i32 @foo(i32* noalias nocapture %B, i32* noalias nocapture %A, i32 %n, i32 %m) #0 {
 entry:
-  %0 = load i32* %A, align 4
+  %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add8 = mul i32 %1, %mul238
-  %arrayidx9 = getelementptr inbounds i32* %B, i64 1
+  %arrayidx9 = getelementptr inbounds i32, i32* %B, i64 1
   store i32 %add8, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %A, i64 2
-  %2 = load i32* %arrayidx10, align 4
+  %arrayidx10 = getelementptr inbounds i32, i32* %A, i64 2
+  %2 = load i32, i32* %arrayidx10, align 4
   %add14 = mul i32 %2, %mul238
-  %arrayidx15 = getelementptr inbounds i32* %B, i64 2
+  %arrayidx15 = getelementptr inbounds i32, i32* %B, i64 2
   store i32 %add14, i32* %arrayidx15, align 4
-  %arrayidx16 = getelementptr inbounds i32* %A, i64 3
-  %3 = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds i32, i32* %A, i64 3
+  %3 = load i32, i32* %arrayidx16, align 4
   %add20 = mul i32 %3, %mul238
-  %arrayidx21 = getelementptr inbounds i32* %B, i64 3
+  %arrayidx21 = getelementptr inbounds i32, i32* %B, i64 3
   store i32 %add20, i32* %arrayidx21, align 4
   ret i32 0
 }
@@ -56,24 +56,24 @@ entry:
 ; CHECK-NEXT: ret
 define i32 @extr_user(i32* noalias nocapture %B, i32* noalias nocapture %A, i32 %n, i32 %m) {
 entry:
-  %0 = load i32* %A, align 4
+  %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add8 = mul i32 %1, %mul238
-  %arrayidx9 = getelementptr inbounds i32* %B, i64 1
+  %arrayidx9 = getelementptr inbounds i32, i32* %B, i64 1
   store i32 %add8, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %A, i64 2
-  %2 = load i32* %arrayidx10, align 4
+  %arrayidx10 = getelementptr inbounds i32, i32* %A, i64 2
+  %2 = load i32, i32* %arrayidx10, align 4
   %add14 = mul i32 %2, %mul238
-  %arrayidx15 = getelementptr inbounds i32* %B, i64 2
+  %arrayidx15 = getelementptr inbounds i32, i32* %B, i64 2
   store i32 %add14, i32* %arrayidx15, align 4
-  %arrayidx16 = getelementptr inbounds i32* %A, i64 3
-  %3 = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds i32, i32* %A, i64 3
+  %3 = load i32, i32* %arrayidx16, align 4
   %add20 = mul i32 %3, %mul238
-  %arrayidx21 = getelementptr inbounds i32* %B, i64 3
+  %arrayidx21 = getelementptr inbounds i32, i32* %B, i64 3
   store i32 %add20, i32* %arrayidx21, align 4
   ret i32 %0  ;<--------- This value has multiple users
 }
@@ -86,24 +86,24 @@ entry:
 ; CHECK-NEXT: ret
 define i32 @extr_user1(i32* noalias nocapture %B, i32* noalias nocapture %A, i32 %n, i32 %m) {
 entry:
-  %0 = load i32* %A, align 4
+  %0 = load i32, i32* %A, align 4
   %mul238 = add i32 %m, %n
   %add = mul i32 %0, %mul238
   store i32 %add, i32* %B, align 4
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add8 = mul i32 %1, %mul238
-  %arrayidx9 = getelementptr inbounds i32* %B, i64 1
+  %arrayidx9 = getelementptr inbounds i32, i32* %B, i64 1
   store i32 %add8, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %A, i64 2
-  %2 = load i32* %arrayidx10, align 4
+  %arrayidx10 = getelementptr inbounds i32, i32* %A, i64 2
+  %2 = load i32, i32* %arrayidx10, align 4
   %add14 = mul i32 %2, %mul238
-  %arrayidx15 = getelementptr inbounds i32* %B, i64 2
+  %arrayidx15 = getelementptr inbounds i32, i32* %B, i64 2
   store i32 %add14, i32* %arrayidx15, align 4
-  %arrayidx16 = getelementptr inbounds i32* %A, i64 3
-  %3 = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds i32, i32* %A, i64 3
+  %3 = load i32, i32* %arrayidx16, align 4
   %add20 = mul i32 %3, %mul238
-  %arrayidx21 = getelementptr inbounds i32* %B, i64 3
+  %arrayidx21 = getelementptr inbounds i32, i32* %B, i64 3
   store i32 %add20, i32* %arrayidx21, align 4
   ret i32 %1  ;<--------- This value has multiple users
 }
diff --git a/test/Transforms/SLPVectorizer/X86/external_user.ll b/test/Transforms/SLPVectorizer/X86/external_user.ll
index 6d09aa61bf35..bf2febda86b4 100644
--- a/test/Transforms/SLPVectorizer/X86/external_user.ll
+++ b/test/Transforms/SLPVectorizer/X86/external_user.ll
@@ -33,9 +33,9 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define double @ext_user(double* noalias nocapture %B, double* noalias nocapture %A, i32 %n, i32 %m) {
 entry:
-  %arrayidx = getelementptr inbounds double* %A, i64 1
-  %0 = load double* %arrayidx, align 8
-  %1 = load double* %A, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 1
+  %0 = load double, double* %arrayidx, align 8
+  %1 = load double, double* %A, align 8
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %entry
@@ -54,7 +54,7 @@ for.body:                                         ; preds = %for.body, %entry
 
 for.end:                                          ; preds = %for.body
   store double %add5, double* %B, align 8
-  %arrayidx7 = getelementptr inbounds double* %B, i64 1
+  %arrayidx7 = getelementptr inbounds double, double* %B, i64 1
   store double %add4, double* %arrayidx7, align 8
   ret double %mul3
 }
@@ -69,9 +69,9 @@ for.end:                                          ; preds = %for.body
 define i32 @needtogather(double *noalias %a, i32 *noalias %b,  float * noalias %c,
                 i32 * noalias %d) {
 entry:
-  %0 = load i32* %d, align 4
+  %0 = load i32, i32* %d, align 4
   %conv = sitofp i32 %0 to float
-  %1 = load float* %c
+  %1 = load float, float* %c
   %sub = fsub float 0.000000e+00, %1
   %mul = fmul float %sub, 0.000000e+00
   %add = fadd float %conv, %mul
diff --git a/test/Transforms/SLPVectorizer/X86/extract.ll b/test/Transforms/SLPVectorizer/X86/extract.ll
index f611fd4ec24f..9a6ee2afc8eb 100644
--- a/test/Transforms/SLPVectorizer/X86/extract.ll
+++ b/test/Transforms/SLPVectorizer/X86/extract.ll
@@ -9,11 +9,11 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK: ret void
 define void @fextr(double* %ptr) {
 entry:
-  %LD = load <2 x double>* undef
+  %LD = load <2 x double>, <2 x double>* undef
   %V0 = extractelement <2 x double> %LD, i32 0
   %V1 = extractelement <2 x double> %LD, i32 1
-  %P0 = getelementptr inbounds double* %ptr, i64 0
-  %P1 = getelementptr inbounds double* %ptr, i64 1
+  %P0 = getelementptr inbounds double, double* %ptr, i64 0
+  %P1 = getelementptr inbounds double, double* %ptr, i64 1
   %A0 = fadd double %V0, 0.0
   %A1 = fadd double %V1, 1.1
   store double %A0, double* %P0, align 4
@@ -27,11 +27,11 @@ entry:
 ;CHECK: ret void
 define void @fextr1(double* %ptr) {
 entry:
-  %LD = load <2 x double>* undef
+  %LD = load <2 x double>, <2 x double>* undef
   %V0 = extractelement <2 x double> %LD, i32 0
   %V1 = extractelement <2 x double> %LD, i32 1
-  %P0 = getelementptr inbounds double* %ptr, i64 1  ; <--- incorrect order
-  %P1 = getelementptr inbounds double* %ptr, i64 0
+  %P0 = getelementptr inbounds double, double* %ptr, i64 1  ; <--- incorrect order
+  %P1 = getelementptr inbounds double, double* %ptr, i64 0
   %A0 = fadd double %V0, 1.2
   %A1 = fadd double %V1, 3.4
   store double %A0, double* %P0, align 4
@@ -45,11 +45,11 @@ entry:
 ;CHECK: ret void
 define void @fextr2(double* %ptr) {
 entry:
-  %LD = load <4 x double>* undef
+  %LD = load <4 x double>, <4 x double>* undef
   %V0 = extractelement <4 x double> %LD, i32 0  ; <--- invalid size.
   %V1 = extractelement <4 x double> %LD, i32 1
-  %P0 = getelementptr inbounds double* %ptr, i64 0
-  %P1 = getelementptr inbounds double* %ptr, i64 1
+  %P0 = getelementptr inbounds double, double* %ptr, i64 0
+  %P1 = getelementptr inbounds double, double* %ptr, i64 1
   %A0 = fadd double %V0, 5.5
   %A1 = fadd double %V1, 6.6
   store double %A0, double* %P0, align 4
diff --git a/test/Transforms/SLPVectorizer/X86/extract_in_tree_user.ll b/test/Transforms/SLPVectorizer/X86/extract_in_tree_user.ll
index 3628042cdb4d..6e5415b3c701 100644
--- a/test/Transforms/SLPVectorizer/X86/extract_in_tree_user.ll
+++ b/test/Transforms/SLPVectorizer/X86/extract_in_tree_user.ll
@@ -7,13 +7,13 @@ target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 ; Function Attrs: nounwind ssp uwtable
 define i32 @fn1() {
 entry:
-  %0 = load i64** @a, align 8
-  %add.ptr = getelementptr inbounds i64* %0, i64 11
+  %0 = load i64*, i64** @a, align 8
+  %add.ptr = getelementptr inbounds i64, i64* %0, i64 11
   %1 = ptrtoint i64* %add.ptr to i64
   store i64 %1, i64* %add.ptr, align 8
-  %add.ptr1 = getelementptr inbounds i64* %0, i64 56
+  %add.ptr1 = getelementptr inbounds i64, i64* %0, i64 56
   %2 = ptrtoint i64* %add.ptr1 to i64
-  %arrayidx2 = getelementptr inbounds i64* %0, i64 12
+  %arrayidx2 = getelementptr inbounds i64, i64* %0, i64 12
   store i64 %2, i64* %arrayidx2, align 8
   ret i32 undef
 ; CHECK-LABEL: @fn1(
@@ -25,42 +25,42 @@ entry:
 declare float @llvm.powi.f32(float, i32)
 define void @fn2(i32* %a, i32* %b, float* %c) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %fp1 = sitofp i32 %add1 to float
   %call1 = tail call float @llvm.powi.f32(float %fp1,i32 %add1) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %fp2 = sitofp i32 %add2 to float
   %call2 = tail call float @llvm.powi.f32(float %fp2,i32 %add1) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %fp3 = sitofp i32 %add3 to float
   %call3 = tail call float @llvm.powi.f32(float %fp3,i32 %add1) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %fp4 = sitofp i32 %add4 to float
   %call4 = tail call float @llvm.powi.f32(float %fp4,i32 %add1) nounwind readnone
 
   store float %call1, float* %c, align 4
-  %arrayidx8 = getelementptr inbounds float* %c, i32 1
+  %arrayidx8 = getelementptr inbounds float, float* %c, i32 1
   store float %call2, float* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds float* %c, i32 2
+  %arrayidx9 = getelementptr inbounds float, float* %c, i32 2
   store float %call3, float* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds float* %c, i32 3
+  %arrayidx10 = getelementptr inbounds float, float* %c, i32 3
   store float %call4, float* %arrayidx10, align 4
   ret void
 
diff --git a/test/Transforms/SLPVectorizer/X86/extractcost.ll b/test/Transforms/SLPVectorizer/X86/extractcost.ll
index 01baf66e2f95..164ddf38c192 100644
--- a/test/Transforms/SLPVectorizer/X86/extractcost.ll
+++ b/test/Transforms/SLPVectorizer/X86/extractcost.ll
@@ -13,15 +13,15 @@ entry:
   store i32 %add, i32* %A, align 4
   %mul1 = mul nsw i32 %n, 9
   %add2 = add nsw i32 %mul1, 9
-  %arrayidx3 = getelementptr inbounds i32* %A, i64 1
+  %arrayidx3 = getelementptr inbounds i32, i32* %A, i64 1
   store i32 %add2, i32* %arrayidx3, align 4
   %mul4 = shl i32 %n, 3
   %add5 = add nsw i32 %mul4, 9
-  %arrayidx6 = getelementptr inbounds i32* %A, i64 2
+  %arrayidx6 = getelementptr inbounds i32, i32* %A, i64 2
   store i32 %add5, i32* %arrayidx6, align 4
   %mul7 = mul nsw i32 %n, 10
   %add8 = add nsw i32 %mul7, 9
-  %arrayidx9 = getelementptr inbounds i32* %A, i64 3
+  %arrayidx9 = getelementptr inbounds i32, i32* %A, i64 3
   store i32 %add8, i32* %arrayidx9, align 4
   %externaluse1 = add nsw i32 %add, %m  
   %externaluse2 = mul nsw i32 %add, %m  ; we should add the extract cost only once and the store will be vectorized
diff --git a/test/Transforms/SLPVectorizer/X86/flag.ll b/test/Transforms/SLPVectorizer/X86/flag.ll
index 3ca540706744..7db8d75c20a7 100644
--- a/test/Transforms/SLPVectorizer/X86/flag.ll
+++ b/test/Transforms/SLPVectorizer/X86/flag.ll
@@ -15,17 +15,17 @@ define i32 @rollable(i32* noalias nocapture %in, i32* noalias nocapture %out, i6
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %i.019 = phi i64 [ %26, %.lr.ph ], [ 0, %0 ]
   %2 = shl i64 %i.019, 2
-  %3 = getelementptr inbounds i32* %in, i64 %2
-  %4 = load i32* %3, align 4
+  %3 = getelementptr inbounds i32, i32* %in, i64 %2
+  %4 = load i32, i32* %3, align 4
   %5 = or i64 %2, 1
-  %6 = getelementptr inbounds i32* %in, i64 %5
-  %7 = load i32* %6, align 4
+  %6 = getelementptr inbounds i32, i32* %in, i64 %5
+  %7 = load i32, i32* %6, align 4
   %8 = or i64 %2, 2
-  %9 = getelementptr inbounds i32* %in, i64 %8
-  %10 = load i32* %9, align 4
+  %9 = getelementptr inbounds i32, i32* %in, i64 %8
+  %10 = load i32, i32* %9, align 4
   %11 = or i64 %2, 3
-  %12 = getelementptr inbounds i32* %in, i64 %11
-  %13 = load i32* %12, align 4
+  %12 = getelementptr inbounds i32, i32* %in, i64 %11
+  %13 = load i32, i32* %12, align 4
   %14 = mul i32 %4, 7
   %15 = add i32 %14, 7
   %16 = mul i32 %7, 7
@@ -34,13 +34,13 @@ define i32 @rollable(i32* noalias nocapture %in, i32* noalias nocapture %out, i6
   %19 = add i32 %18, 21
   %20 = mul i32 %13, 7
   %21 = add i32 %20, 28
-  %22 = getelementptr inbounds i32* %out, i64 %2
+  %22 = getelementptr inbounds i32, i32* %out, i64 %2
   store i32 %15, i32* %22, align 4
-  %23 = getelementptr inbounds i32* %out, i64 %5
+  %23 = getelementptr inbounds i32, i32* %out, i64 %5
   store i32 %17, i32* %23, align 4
-  %24 = getelementptr inbounds i32* %out, i64 %8
+  %24 = getelementptr inbounds i32, i32* %out, i64 %8
   store i32 %19, i32* %24, align 4
-  %25 = getelementptr inbounds i32* %out, i64 %11
+  %25 = getelementptr inbounds i32, i32* %out, i64 %11
   store i32 %21, i32* %25, align 4
   %26 = add i64 %i.019, 1
   %exitcond = icmp eq i64 %26, %n
diff --git a/test/Transforms/SLPVectorizer/X86/gep.ll b/test/Transforms/SLPVectorizer/X86/gep.ll
index 9e105ec98489..3f952d7b242b 100644
--- a/test/Transforms/SLPVectorizer/X86/gep.ll
+++ b/test/Transforms/SLPVectorizer/X86/gep.ll
@@ -9,15 +9,15 @@ target datalayout = "e-m:o-i64:64-f80:128-n8:16:32:64-S128"
 ; CHECK-LABEL: foo1
 ; CHECK: <2 x i32*>
 define void @foo1 ({ i32*, i32* }* noalias %x, { i32*, i32* }* noalias %y) {
-  %1 = getelementptr inbounds { i32*, i32* }* %y, i64 0, i32 0
-  %2 = load i32** %1, align 8
-  %3 = getelementptr inbounds i32* %2, i64 16
-  %4 = getelementptr inbounds { i32*, i32* }* %x, i64 0, i32 0
+  %1 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %y, i64 0, i32 0
+  %2 = load i32*, i32** %1, align 8
+  %3 = getelementptr inbounds i32, i32* %2, i64 16
+  %4 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %x, i64 0, i32 0
   store i32* %3, i32** %4, align 8
-  %5 = getelementptr inbounds { i32*, i32* }* %y, i64 0, i32 1
-  %6 = load i32** %5, align 8
-  %7 = getelementptr inbounds i32* %6, i64 16
-  %8 = getelementptr inbounds { i32*, i32* }* %x, i64 0, i32 1
+  %5 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %y, i64 0, i32 1
+  %6 = load i32*, i32** %5, align 8
+  %7 = getelementptr inbounds i32, i32* %6, i64 16
+  %8 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %x, i64 0, i32 1
   store i32* %7, i32** %8, align 8
   ret void
 }
@@ -27,15 +27,15 @@ define void @foo1 ({ i32*, i32* }* noalias %x, { i32*, i32* }* noalias %y) {
 ; CHECK-LABEL: foo2
 ; CHECK-NOT: <2 x i32*>
 define void @foo2 ({ i32*, i32* }* noalias %x, { i32*, i32* }* noalias %y, i32 %i) {
-  %1 = getelementptr inbounds { i32*, i32* }* %y, i64 0, i32 0
-  %2 = load i32** %1, align 8
-  %3 = getelementptr inbounds i32* %2, i32 %i
-  %4 = getelementptr inbounds { i32*, i32* }* %x, i64 0, i32 0
+  %1 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %y, i64 0, i32 0
+  %2 = load i32*, i32** %1, align 8
+  %3 = getelementptr inbounds i32, i32* %2, i32 %i
+  %4 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %x, i64 0, i32 0
   store i32* %3, i32** %4, align 8
-  %5 = getelementptr inbounds { i32*, i32* }* %y, i64 0, i32 1
-  %6 = load i32** %5, align 8
-  %7 = getelementptr inbounds i32* %6, i32 %i
-  %8 = getelementptr inbounds { i32*, i32* }* %x, i64 0, i32 1
+  %5 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %y, i64 0, i32 1
+  %6 = load i32*, i32** %5, align 8
+  %7 = getelementptr inbounds i32, i32* %6, i32 %i
+  %8 = getelementptr inbounds { i32*, i32* }, { i32*, i32* }* %x, i64 0, i32 1
   store i32* %7, i32** %8, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/hoist.ll b/test/Transforms/SLPVectorizer/X86/hoist.ll
index 78c58f1d84e9..36c939b597e0 100644
--- a/test/Transforms/SLPVectorizer/X86/hoist.ll
+++ b/test/Transforms/SLPVectorizer/X86/hoist.ll
@@ -30,23 +30,23 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %i.024 = phi i32 [ 0, %entry ], [ %add10, %for.body ]
-  %arrayidx = getelementptr inbounds i32* %A, i32 %i.024
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i32 %i.024
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %n
   store i32 %add, i32* %arrayidx, align 4
   %add121 = or i32 %i.024, 1
-  %arrayidx2 = getelementptr inbounds i32* %A, i32 %add121
-  %1 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i32 %add121
+  %1 = load i32, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %1, %k
   store i32 %add3, i32* %arrayidx2, align 4
   %add422 = or i32 %i.024, 2
-  %arrayidx5 = getelementptr inbounds i32* %A, i32 %add422
-  %2 = load i32* %arrayidx5, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %A, i32 %add422
+  %2 = load i32, i32* %arrayidx5, align 4
   %add6 = add nsw i32 %2, %n
   store i32 %add6, i32* %arrayidx5, align 4
   %add723 = or i32 %i.024, 3
-  %arrayidx8 = getelementptr inbounds i32* %A, i32 %add723
-  %3 = load i32* %arrayidx8, align 4
+  %arrayidx8 = getelementptr inbounds i32, i32* %A, i32 %add723
+  %3 = load i32, i32* %arrayidx8, align 4
   %add9 = add nsw i32 %3, %k
   store i32 %add9, i32* %arrayidx8, align 4
   %add10 = add nsw i32 %i.024, 4
diff --git a/test/Transforms/SLPVectorizer/X86/horizontal.ll b/test/Transforms/SLPVectorizer/X86/horizontal.ll
index 18360471c190..83b2e01f04e0 100644
--- a/test/Transforms/SLPVectorizer/X86/horizontal.ll
+++ b/test/Transforms/SLPVectorizer/X86/horizontal.ll
@@ -32,22 +32,22 @@ for.body:
   %i.033 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %sum.032 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add17, %for.body ]
   %mul = shl nsw i64 %i.033, 2
-  %arrayidx = getelementptr inbounds float* %A, i64 %mul
-  %1 = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds float, float* %A, i64 %mul
+  %1 = load float, float* %arrayidx, align 4
   %mul2 = fmul float %1, 7.000000e+00
   %add28 = or i64 %mul, 1
-  %arrayidx4 = getelementptr inbounds float* %A, i64 %add28
-  %2 = load float* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %A, i64 %add28
+  %2 = load float, float* %arrayidx4, align 4
   %mul5 = fmul float %2, 7.000000e+00
   %add6 = fadd fast float %mul2, %mul5
   %add829 = or i64 %mul, 2
-  %arrayidx9 = getelementptr inbounds float* %A, i64 %add829
-  %3 = load float* %arrayidx9, align 4
+  %arrayidx9 = getelementptr inbounds float, float* %A, i64 %add829
+  %3 = load float, float* %arrayidx9, align 4
   %mul10 = fmul float %3, 7.000000e+00
   %add11 = fadd fast float %add6, %mul10
   %add1330 = or i64 %mul, 3
-  %arrayidx14 = getelementptr inbounds float* %A, i64 %add1330
-  %4 = load float* %arrayidx14, align 4
+  %arrayidx14 = getelementptr inbounds float, float* %A, i64 %add1330
+  %4 = load float, float* %arrayidx14, align 4
   %mul15 = fmul float %4, 7.000000e+00
   %add16 = fadd fast float %add11, %mul15
   %add17 = fadd fast float %sum.032, %add16
@@ -85,13 +85,13 @@ entry:
   br i1 %cmp38, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load float* %B, align 4
-  %arrayidx4 = getelementptr inbounds float* %B, i64 1
-  %1 = load float* %arrayidx4, align 4
-  %arrayidx9 = getelementptr inbounds float* %B, i64 2
-  %2 = load float* %arrayidx9, align 4
-  %arrayidx15 = getelementptr inbounds float* %B, i64 3
-  %3 = load float* %arrayidx15, align 4
+  %0 = load float, float* %B, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %B, i64 1
+  %1 = load float, float* %arrayidx4, align 4
+  %arrayidx9 = getelementptr inbounds float, float* %B, i64 2
+  %2 = load float, float* %arrayidx9, align 4
+  %arrayidx15 = getelementptr inbounds float, float* %B, i64 3
+  %3 = load float, float* %arrayidx15, align 4
   %4 = sext i32 %n to i64
   br label %for.body
 
@@ -99,22 +99,22 @@ for.body:
   %i.040 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %sum.039 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %mul21, %for.body ]
   %mul = shl nsw i64 %i.040, 2
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %mul
-  %5 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %mul
+  %5 = load float, float* %arrayidx2, align 4
   %mul3 = fmul float %0, %5
   %add35 = or i64 %mul, 1
-  %arrayidx6 = getelementptr inbounds float* %A, i64 %add35
-  %6 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %A, i64 %add35
+  %6 = load float, float* %arrayidx6, align 4
   %mul7 = fmul float %1, %6
   %add8 = fadd fast float %mul3, %mul7
   %add1136 = or i64 %mul, 2
-  %arrayidx12 = getelementptr inbounds float* %A, i64 %add1136
-  %7 = load float* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds float, float* %A, i64 %add1136
+  %7 = load float, float* %arrayidx12, align 4
   %mul13 = fmul float %2, %7
   %add14 = fadd fast float %add8, %mul13
   %add1737 = or i64 %mul, 3
-  %arrayidx18 = getelementptr inbounds float* %A, i64 %add1737
-  %8 = load float* %arrayidx18, align 4
+  %arrayidx18 = getelementptr inbounds float, float* %A, i64 %add1737
+  %8 = load float, float* %arrayidx18, align 4
   %mul19 = fmul float %3, %8
   %add20 = fadd fast float %add14, %mul19
   %mul21 = fmul float %sum.039, %add20
@@ -157,23 +157,23 @@ entry:
   br i1 %cmp81, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load float* %B, align 4
-  %arrayidx4 = getelementptr inbounds float* %B, i64 1
-  %1 = load float* %arrayidx4, align 4
-  %arrayidx9 = getelementptr inbounds float* %B, i64 2
-  %2 = load float* %arrayidx9, align 4
-  %arrayidx15 = getelementptr inbounds float* %B, i64 3
-  %3 = load float* %arrayidx15, align 4
-  %arrayidx21 = getelementptr inbounds float* %B, i64 4
-  %4 = load float* %arrayidx21, align 4
-  %arrayidx27 = getelementptr inbounds float* %B, i64 5
-  %5 = load float* %arrayidx27, align 4
-  %arrayidx33 = getelementptr inbounds float* %B, i64 6
-  %6 = load float* %arrayidx33, align 4
-  %arrayidx39 = getelementptr inbounds float* %B, i64 7
-  %7 = load float* %arrayidx39, align 4
-  %arrayidx45 = getelementptr inbounds float* %B, i64 8
-  %8 = load float* %arrayidx45, align 4
+  %0 = load float, float* %B, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %B, i64 1
+  %1 = load float, float* %arrayidx4, align 4
+  %arrayidx9 = getelementptr inbounds float, float* %B, i64 2
+  %2 = load float, float* %arrayidx9, align 4
+  %arrayidx15 = getelementptr inbounds float, float* %B, i64 3
+  %3 = load float, float* %arrayidx15, align 4
+  %arrayidx21 = getelementptr inbounds float, float* %B, i64 4
+  %4 = load float, float* %arrayidx21, align 4
+  %arrayidx27 = getelementptr inbounds float, float* %B, i64 5
+  %5 = load float, float* %arrayidx27, align 4
+  %arrayidx33 = getelementptr inbounds float, float* %B, i64 6
+  %6 = load float, float* %arrayidx33, align 4
+  %arrayidx39 = getelementptr inbounds float, float* %B, i64 7
+  %7 = load float, float* %arrayidx39, align 4
+  %arrayidx45 = getelementptr inbounds float, float* %B, i64 8
+  %8 = load float, float* %arrayidx45, align 4
   %9 = sext i32 %n to i64
   br label %for.body
 
@@ -181,47 +181,47 @@ for.body:
   %i.083 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %sum.082 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add51, %for.body ]
   %mul = mul nsw i64 %i.083, 6
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %mul
-  %10 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %mul
+  %10 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %0, %10
   %add80 = or i64 %mul, 1
-  %arrayidx6 = getelementptr inbounds float* %A, i64 %add80
-  %11 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %A, i64 %add80
+  %11 = load float, float* %arrayidx6, align 4
   %mul7 = fmul fast float %1, %11
   %add8 = fadd fast float %mul3, %mul7
   %add11 = add nsw i64 %mul, 2
-  %arrayidx12 = getelementptr inbounds float* %A, i64 %add11
-  %12 = load float* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds float, float* %A, i64 %add11
+  %12 = load float, float* %arrayidx12, align 4
   %mul13 = fmul fast float %2, %12
   %add14 = fadd fast float %add8, %mul13
   %add17 = add nsw i64 %mul, 3
-  %arrayidx18 = getelementptr inbounds float* %A, i64 %add17
-  %13 = load float* %arrayidx18, align 4
+  %arrayidx18 = getelementptr inbounds float, float* %A, i64 %add17
+  %13 = load float, float* %arrayidx18, align 4
   %mul19 = fmul fast float %3, %13
   %add20 = fadd fast float %add14, %mul19
   %add23 = add nsw i64 %mul, 4
-  %arrayidx24 = getelementptr inbounds float* %A, i64 %add23
-  %14 = load float* %arrayidx24, align 4
+  %arrayidx24 = getelementptr inbounds float, float* %A, i64 %add23
+  %14 = load float, float* %arrayidx24, align 4
   %mul25 = fmul fast float %4, %14
   %add26 = fadd fast float %add20, %mul25
   %add29 = add nsw i64 %mul, 5
-  %arrayidx30 = getelementptr inbounds float* %A, i64 %add29
-  %15 = load float* %arrayidx30, align 4
+  %arrayidx30 = getelementptr inbounds float, float* %A, i64 %add29
+  %15 = load float, float* %arrayidx30, align 4
   %mul31 = fmul fast float %5, %15
   %add32 = fadd fast float %add26, %mul31
   %add35 = add nsw i64 %mul, 6
-  %arrayidx36 = getelementptr inbounds float* %A, i64 %add35
-  %16 = load float* %arrayidx36, align 4
+  %arrayidx36 = getelementptr inbounds float, float* %A, i64 %add35
+  %16 = load float, float* %arrayidx36, align 4
   %mul37 = fmul fast float %6, %16
   %add38 = fadd fast float %add32, %mul37
   %add41 = add nsw i64 %mul, 7
-  %arrayidx42 = getelementptr inbounds float* %A, i64 %add41
-  %17 = load float* %arrayidx42, align 4
+  %arrayidx42 = getelementptr inbounds float, float* %A, i64 %add41
+  %17 = load float, float* %arrayidx42, align 4
   %mul43 = fmul fast float %7, %17
   %add44 = fadd fast float %add38, %mul43
   %add47 = add nsw i64 %mul, 8
-  %arrayidx48 = getelementptr inbounds float* %A, i64 %add47
-  %18 = load float* %arrayidx48, align 4
+  %arrayidx48 = getelementptr inbounds float, float* %A, i64 %add47
+  %18 = load float, float* %arrayidx48, align 4
   %mul49 = fmul fast float %8, %18
   %add50 = fadd fast float %add44, %mul49
   %add51 = fadd fast float %sum.082, %add50
@@ -259,13 +259,13 @@ entry:
   br i1 %cmp41, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load float* %B, align 4
-  %arrayidx4 = getelementptr inbounds float* %B, i64 1
-  %1 = load float* %arrayidx4, align 4
-  %arrayidx10 = getelementptr inbounds float* %B, i64 2
-  %2 = load float* %arrayidx10, align 4
-  %arrayidx16 = getelementptr inbounds float* %B, i64 3
-  %3 = load float* %arrayidx16, align 4
+  %0 = load float, float* %B, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %B, i64 1
+  %1 = load float, float* %arrayidx4, align 4
+  %arrayidx10 = getelementptr inbounds float, float* %B, i64 2
+  %2 = load float, float* %arrayidx10, align 4
+  %arrayidx16 = getelementptr inbounds float, float* %B, i64 3
+  %3 = load float, float* %arrayidx16, align 4
   %4 = sext i32 %n to i64
   br label %for.body
 
@@ -273,23 +273,23 @@ for.body:
   %i.043 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %sum.042 = phi float [ 0.000000e+00, %for.body.lr.ph ], [ %add21, %for.body ]
   %mul = shl nsw i64 %i.043, 2
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %mul
-  %5 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %mul
+  %5 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %0, %5
   %add = fadd fast float %sum.042, %mul3
   %add638 = or i64 %mul, 1
-  %arrayidx7 = getelementptr inbounds float* %A, i64 %add638
-  %6 = load float* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %A, i64 %add638
+  %6 = load float, float* %arrayidx7, align 4
   %mul8 = fmul fast float %1, %6
   %add9 = fadd fast float %add, %mul8
   %add1239 = or i64 %mul, 2
-  %arrayidx13 = getelementptr inbounds float* %A, i64 %add1239
-  %7 = load float* %arrayidx13, align 4
+  %arrayidx13 = getelementptr inbounds float, float* %A, i64 %add1239
+  %7 = load float, float* %arrayidx13, align 4
   %mul14 = fmul fast float %2, %7
   %add15 = fadd fast float %add9, %mul14
   %add1840 = or i64 %mul, 3
-  %arrayidx19 = getelementptr inbounds float* %A, i64 %add1840
-  %8 = load float* %arrayidx19, align 4
+  %arrayidx19 = getelementptr inbounds float, float* %A, i64 %add1840
+  %8 = load float, float* %arrayidx19, align 4
   %mul20 = fmul fast float %3, %8
   %add21 = fadd fast float %add15, %mul20
   %inc = add nsw i64 %i.043, 1
@@ -326,40 +326,40 @@ entry:
   br i1 %cmp37, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %arrayidx4 = getelementptr inbounds float* %B, i64 1
-  %arrayidx9 = getelementptr inbounds float* %B, i64 2
-  %arrayidx15 = getelementptr inbounds float* %B, i64 3
+  %arrayidx4 = getelementptr inbounds float, float* %B, i64 1
+  %arrayidx9 = getelementptr inbounds float, float* %B, i64 2
+  %arrayidx15 = getelementptr inbounds float, float* %B, i64 3
   %0 = sext i32 %n to i64
   br label %for.body
 
 for.body:
   %i.039 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %C.addr.038 = phi float* [ %C, %for.body.lr.ph ], [ %incdec.ptr, %for.body ]
-  %1 = load float* %B, align 4
+  %1 = load float, float* %B, align 4
   %mul = shl nsw i64 %i.039, 2
-  %arrayidx2 = getelementptr inbounds float* %A, i64 %mul
-  %2 = load float* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 %mul
+  %2 = load float, float* %arrayidx2, align 4
   %mul3 = fmul fast float %1, %2
-  %3 = load float* %arrayidx4, align 4
+  %3 = load float, float* %arrayidx4, align 4
   %add34 = or i64 %mul, 1
-  %arrayidx6 = getelementptr inbounds float* %A, i64 %add34
-  %4 = load float* %arrayidx6, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %A, i64 %add34
+  %4 = load float, float* %arrayidx6, align 4
   %mul7 = fmul fast float %3, %4
   %add8 = fadd fast float %mul3, %mul7
-  %5 = load float* %arrayidx9, align 4
+  %5 = load float, float* %arrayidx9, align 4
   %add1135 = or i64 %mul, 2
-  %arrayidx12 = getelementptr inbounds float* %A, i64 %add1135
-  %6 = load float* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds float, float* %A, i64 %add1135
+  %6 = load float, float* %arrayidx12, align 4
   %mul13 = fmul fast float %5, %6
   %add14 = fadd fast float %add8, %mul13
-  %7 = load float* %arrayidx15, align 4
+  %7 = load float, float* %arrayidx15, align 4
   %add1736 = or i64 %mul, 3
-  %arrayidx18 = getelementptr inbounds float* %A, i64 %add1736
-  %8 = load float* %arrayidx18, align 4
+  %arrayidx18 = getelementptr inbounds float, float* %A, i64 %add1736
+  %8 = load float, float* %arrayidx18, align 4
   %mul19 = fmul fast float %7, %8
   %add20 = fadd fast float %add14, %mul19
   store float %add20, float* %C.addr.038, align 4
-  %incdec.ptr = getelementptr inbounds float* %C.addr.038, i64 1
+  %incdec.ptr = getelementptr inbounds float, float* %C.addr.038, i64 1
   %inc = add nsw i64 %i.039, 1
   %exitcond = icmp eq i64 %inc, %0
   br i1 %exitcond, label %for.end, label %for.body
@@ -389,24 +389,24 @@ entry:
   br i1 %cmp17, label %for.body.lr.ph, label %for.end
 
 for.body.lr.ph:
-  %0 = load double* %B, align 8
-  %arrayidx4 = getelementptr inbounds double* %B, i64 1
-  %1 = load double* %arrayidx4, align 8
+  %0 = load double, double* %B, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %B, i64 1
+  %1 = load double, double* %arrayidx4, align 8
   %2 = sext i32 %n to i64
   br label %for.body
 
 for.body:
   %i.018 = phi i64 [ 0, %for.body.lr.ph ], [ %inc, %for.body ]
   %mul = shl nsw i64 %i.018, 2
-  %arrayidx2 = getelementptr inbounds double* %A, i64 %mul
-  %3 = load double* %arrayidx2, align 8
+  %arrayidx2 = getelementptr inbounds double, double* %A, i64 %mul
+  %3 = load double, double* %arrayidx2, align 8
   %mul3 = fmul fast double %0, %3
   %add16 = or i64 %mul, 1
-  %arrayidx6 = getelementptr inbounds double* %A, i64 %add16
-  %4 = load double* %arrayidx6, align 8
+  %arrayidx6 = getelementptr inbounds double, double* %A, i64 %add16
+  %4 = load double, double* %arrayidx6, align 8
   %mul7 = fmul fast double %1, %4
   %add8 = fadd fast double %mul3, %mul7
-  %arrayidx9 = getelementptr inbounds double* %C, i64 %i.018
+  %arrayidx9 = getelementptr inbounds double, double* %C, i64 %i.018
   store double %add8, double* %arrayidx9, align 8
   %inc = add nsw i64 %i.018, 1
   %exitcond = icmp eq i64 %inc, %2
diff --git a/test/Transforms/SLPVectorizer/X86/implicitfloat.ll b/test/Transforms/SLPVectorizer/X86/implicitfloat.ll
index f63f2683b10e..f7283f0d0271 100644
--- a/test/Transforms/SLPVectorizer/X86/implicitfloat.ll
+++ b/test/Transforms/SLPVectorizer/X86/implicitfloat.ll
@@ -9,16 +9,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: ret
 define void @test1(double* %a, double* %b, double* %c) noimplicitfloat { ; <------ noimplicitfloat attribute here!
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/in-tree-user.ll b/test/Transforms/SLPVectorizer/X86/in-tree-user.ll
index 194a0fbd5ed0..b0ce07421b57 100644
--- a/test/Transforms/SLPVectorizer/X86/in-tree-user.ll
+++ b/test/Transforms/SLPVectorizer/X86/in-tree-user.ll
@@ -19,15 +19,15 @@ entry:
 for.body:                                         ; preds = %for.inc, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.inc ]
   %0 = shl nsw i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds double* %A, i64 %0
-  %1 = load double* %arrayidx, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 %0
+  %1 = load double, double* %arrayidx, align 8
   %mul1 = fmul double %conv, %1
   %mul2 = fmul double %mul1, 7.000000e+00
   %add = fadd double %mul2, 5.000000e+00
   %InTreeUser = fadd double %add, %add    ; <------------------ In tree user.
   %2 = or i64 %0, 1
-  %arrayidx6 = getelementptr inbounds double* %A, i64 %2
-  %3 = load double* %arrayidx6, align 8
+  %arrayidx6 = getelementptr inbounds double, double* %A, i64 %2
+  %3 = load double, double* %arrayidx6, align 8
   %mul8 = fmul double %conv, %3
   %mul9 = fmul double %mul8, 4.000000e+00
   %add10 = fadd double %mul9, 9.000000e+00
@@ -35,7 +35,7 @@ for.body:                                         ; preds = %for.inc, %entry
   br i1 %cmp11, label %if.then, label %for.inc
 
 if.then:                                          ; preds = %for.body
-  %call = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([6 x i8]* @.str, i64 0, i64 0))
+  %call = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str, i64 0, i64 0))
   br label %for.inc
 
 for.inc:                                          ; preds = %for.body, %if.then
diff --git a/test/Transforms/SLPVectorizer/X86/intrinsic.ll b/test/Transforms/SLPVectorizer/X86/intrinsic.ll
index 937252f4146b..cc5a4afe43d9 100644
--- a/test/Transforms/SLPVectorizer/X86/intrinsic.ll
+++ b/test/Transforms/SLPVectorizer/X86/intrinsic.ll
@@ -13,18 +13,18 @@ declare double @llvm.fabs.f64(double) nounwind readnone
 ;CHECK: ret
 define void @vec_fabs_f64(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
   %call = tail call double @llvm.fabs.f64(double %mul) nounwind readnone
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %call5 = tail call double @llvm.fabs.f64(double %mul5) nounwind readnone
   store double %call, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %call5, double* %arrayidx5, align 8
   ret void
 }
@@ -39,33 +39,33 @@ declare float @llvm.copysign.f32(float, float) nounwind readnone
 ;CHECK: ret
 define void @vec_copysign_f32(float* %a, float* %b, float* noalias %c) {
 entry:
-  %0 = load float* %a, align 4
-  %1 = load float* %b, align 4
+  %0 = load float, float* %a, align 4
+  %1 = load float, float* %b, align 4
   %call0 = tail call float @llvm.copysign.f32(float %0, float %1) nounwind readnone
   store float %call0, float* %c, align 4
 
-  %ix2 = getelementptr inbounds float* %a, i64 1
-  %2 = load float* %ix2, align 4
-  %ix3 = getelementptr inbounds float* %b, i64 1
-  %3 = load float* %ix3, align 4
+  %ix2 = getelementptr inbounds float, float* %a, i64 1
+  %2 = load float, float* %ix2, align 4
+  %ix3 = getelementptr inbounds float, float* %b, i64 1
+  %3 = load float, float* %ix3, align 4
   %call1 = tail call float @llvm.copysign.f32(float %2, float %3) nounwind readnone
-  %c1 = getelementptr inbounds float* %c, i64 1
+  %c1 = getelementptr inbounds float, float* %c, i64 1
   store float %call1, float* %c1, align 4
 
-  %ix4 = getelementptr inbounds float* %a, i64 2
-  %4 = load float* %ix4, align 4
-  %ix5 = getelementptr inbounds float* %b, i64 2
-  %5 = load float* %ix5, align 4
+  %ix4 = getelementptr inbounds float, float* %a, i64 2
+  %4 = load float, float* %ix4, align 4
+  %ix5 = getelementptr inbounds float, float* %b, i64 2
+  %5 = load float, float* %ix5, align 4
   %call2 = tail call float @llvm.copysign.f32(float %4, float %5) nounwind readnone
-  %c2 = getelementptr inbounds float* %c, i64 2
+  %c2 = getelementptr inbounds float, float* %c, i64 2
   store float %call2, float* %c2, align 4
 
-  %ix6 = getelementptr inbounds float* %a, i64 3
-  %6 = load float* %ix6, align 4
-  %ix7 = getelementptr inbounds float* %b, i64 3
-  %7 = load float* %ix7, align 4
+  %ix6 = getelementptr inbounds float, float* %a, i64 3
+  %6 = load float, float* %ix6, align 4
+  %ix7 = getelementptr inbounds float, float* %b, i64 3
+  %7 = load float, float* %ix7, align 4
   %call3 = tail call float @llvm.copysign.f32(float %6, float %7) nounwind readnone
-  %c3 = getelementptr inbounds float* %c, i64 3
+  %c3 = getelementptr inbounds float, float* %c, i64 3
   store float %call3, float* %c3, align 4
 
   ret void
@@ -75,38 +75,38 @@ declare i32 @llvm.bswap.i32(i32) nounwind readnone
 
 define void @vec_bswap_i32(i32* %a, i32* %b, i32* %c) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %call1 = tail call i32 @llvm.bswap.i32(i32 %add1) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %call2 = tail call i32 @llvm.bswap.i32(i32 %add2) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %call3 = tail call i32 @llvm.bswap.i32(i32 %add3) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %call4 = tail call i32 @llvm.bswap.i32(i32 %add4) nounwind readnone
 
   store i32 %call1, i32* %c, align 4
-  %arrayidx8 = getelementptr inbounds i32* %c, i32 1
+  %arrayidx8 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 %call2, i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i32 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 %call3, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %c, i32 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 %call4, i32* %arrayidx10, align 4
   ret void
 
@@ -122,38 +122,38 @@ declare i32 @llvm.ctlz.i32(i32,i1) nounwind readnone
 
 define void @vec_ctlz_i32(i32* %a, i32* %b, i32* %c, i1) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %call1 = tail call i32 @llvm.ctlz.i32(i32 %add1,i1 true) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %call2 = tail call i32 @llvm.ctlz.i32(i32 %add2,i1 true) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %call3 = tail call i32 @llvm.ctlz.i32(i32 %add3,i1 true) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %call4 = tail call i32 @llvm.ctlz.i32(i32 %add4,i1 true) nounwind readnone
 
   store i32 %call1, i32* %c, align 4
-  %arrayidx8 = getelementptr inbounds i32* %c, i32 1
+  %arrayidx8 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 %call2, i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i32 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 %call3, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %c, i32 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 %call4, i32* %arrayidx10, align 4
   ret void
 
@@ -167,38 +167,38 @@ entry:
 
 define void @vec_ctlz_i32_neg(i32* %a, i32* %b, i32* %c, i1) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %call1 = tail call i32 @llvm.ctlz.i32(i32 %add1,i1 true) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %call2 = tail call i32 @llvm.ctlz.i32(i32 %add2,i1 false) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %call3 = tail call i32 @llvm.ctlz.i32(i32 %add3,i1 true) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %call4 = tail call i32 @llvm.ctlz.i32(i32 %add4,i1 false) nounwind readnone
 
   store i32 %call1, i32* %c, align 4
-  %arrayidx8 = getelementptr inbounds i32* %c, i32 1
+  %arrayidx8 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 %call2, i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i32 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 %call3, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %c, i32 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 %call4, i32* %arrayidx10, align 4
   ret void
 
@@ -212,38 +212,38 @@ declare i32 @llvm.cttz.i32(i32,i1) nounwind readnone
 
 define void @vec_cttz_i32(i32* %a, i32* %b, i32* %c, i1) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %call1 = tail call i32 @llvm.cttz.i32(i32 %add1,i1 true) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %call2 = tail call i32 @llvm.cttz.i32(i32 %add2,i1 true) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %call3 = tail call i32 @llvm.cttz.i32(i32 %add3,i1 true) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %call4 = tail call i32 @llvm.cttz.i32(i32 %add4,i1 true) nounwind readnone
 
   store i32 %call1, i32* %c, align 4
-  %arrayidx8 = getelementptr inbounds i32* %c, i32 1
+  %arrayidx8 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 %call2, i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i32 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 %call3, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %c, i32 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 %call4, i32* %arrayidx10, align 4
   ret void
 
@@ -257,38 +257,38 @@ entry:
 
 define void @vec_cttz_i32_neg(i32* %a, i32* %b, i32* %c, i1) {
 entry:
-  %i0 = load i32* %a, align 4
-  %i1 = load i32* %b, align 4
+  %i0 = load i32, i32* %a, align 4
+  %i1 = load i32, i32* %b, align 4
   %add1 = add i32 %i0, %i1
   %call1 = tail call i32 @llvm.cttz.i32(i32 %add1,i1 true) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds i32* %a, i32 1
-  %i2 = load i32* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i32 1
-  %i3 = load i32* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %a, i32 1
+  %i2 = load i32, i32* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i32 1
+  %i3 = load i32, i32* %arrayidx3, align 4
   %add2 = add i32 %i2, %i3
   %call2 = tail call i32 @llvm.cttz.i32(i32 %add2,i1 false) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds i32* %a, i32 2
-  %i4 = load i32* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds i32* %b, i32 2
-  %i5 = load i32* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %a, i32 2
+  %i4 = load i32, i32* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds i32, i32* %b, i32 2
+  %i5 = load i32, i32* %arrayidx5, align 4
   %add3 = add i32 %i4, %i5
   %call3 = tail call i32 @llvm.cttz.i32(i32 %add3,i1 true) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds i32* %a, i32 3
-  %i6 = load i32* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds i32* %b, i32 3
-  %i7 = load i32* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds i32, i32* %a, i32 3
+  %i6 = load i32, i32* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds i32, i32* %b, i32 3
+  %i7 = load i32, i32* %arrayidx7, align 4
   %add4 = add i32 %i6, %i7
   %call4 = tail call i32 @llvm.cttz.i32(i32 %add4,i1 false) nounwind readnone
 
   store i32 %call1, i32* %c, align 4
-  %arrayidx8 = getelementptr inbounds i32* %c, i32 1
+  %arrayidx8 = getelementptr inbounds i32, i32* %c, i32 1
   store i32 %call2, i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i32 2
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i32 2
   store i32 %call3, i32* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds i32* %c, i32 3
+  %arrayidx10 = getelementptr inbounds i32, i32* %c, i32 3
   store i32 %call4, i32* %arrayidx10, align 4
   ret void
 
@@ -300,38 +300,38 @@ entry:
 declare float @llvm.powi.f32(float, i32)
 define void @vec_powi_f32(float* %a, float* %b, float* %c, i32 %P) {
 entry:
-  %i0 = load float* %a, align 4
-  %i1 = load float* %b, align 4
+  %i0 = load float, float* %a, align 4
+  %i1 = load float, float* %b, align 4
   %add1 = fadd float %i0, %i1
   %call1 = tail call float @llvm.powi.f32(float %add1,i32 %P) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds float* %a, i32 1
-  %i2 = load float* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds float* %b, i32 1
-  %i3 = load float* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %a, i32 1
+  %i2 = load float, float* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %b, i32 1
+  %i3 = load float, float* %arrayidx3, align 4
   %add2 = fadd float %i2, %i3
   %call2 = tail call float @llvm.powi.f32(float %add2,i32 %P) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds float* %a, i32 2
-  %i4 = load float* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds float* %b, i32 2
-  %i5 = load float* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %a, i32 2
+  %i4 = load float, float* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds float, float* %b, i32 2
+  %i5 = load float, float* %arrayidx5, align 4
   %add3 = fadd float %i4, %i5
   %call3 = tail call float @llvm.powi.f32(float %add3,i32 %P) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds float* %a, i32 3
-  %i6 = load float* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds float* %b, i32 3
-  %i7 = load float* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %a, i32 3
+  %i6 = load float, float* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %b, i32 3
+  %i7 = load float, float* %arrayidx7, align 4
   %add4 = fadd float %i6, %i7
   %call4 = tail call float @llvm.powi.f32(float %add4,i32 %P) nounwind readnone
 
   store float %call1, float* %c, align 4
-  %arrayidx8 = getelementptr inbounds float* %c, i32 1
+  %arrayidx8 = getelementptr inbounds float, float* %c, i32 1
   store float %call2, float* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds float* %c, i32 2
+  %arrayidx9 = getelementptr inbounds float, float* %c, i32 2
   store float %call3, float* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds float* %c, i32 3
+  %arrayidx10 = getelementptr inbounds float, float* %c, i32 3
   store float %call4, float* %arrayidx10, align 4
   ret void
 
@@ -346,38 +346,38 @@ entry:
 
 define void @vec_powi_f32_neg(float* %a, float* %b, float* %c, i32 %P, i32 %Q) {
 entry:
-  %i0 = load float* %a, align 4
-  %i1 = load float* %b, align 4
+  %i0 = load float, float* %a, align 4
+  %i1 = load float, float* %b, align 4
   %add1 = fadd float %i0, %i1
   %call1 = tail call float @llvm.powi.f32(float %add1,i32 %P) nounwind readnone
 
-  %arrayidx2 = getelementptr inbounds float* %a, i32 1
-  %i2 = load float* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds float* %b, i32 1
-  %i3 = load float* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %a, i32 1
+  %i2 = load float, float* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %b, i32 1
+  %i3 = load float, float* %arrayidx3, align 4
   %add2 = fadd float %i2, %i3
   %call2 = tail call float @llvm.powi.f32(float %add2,i32 %Q) nounwind readnone
 
-  %arrayidx4 = getelementptr inbounds float* %a, i32 2
-  %i4 = load float* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds float* %b, i32 2
-  %i5 = load float* %arrayidx5, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %a, i32 2
+  %i4 = load float, float* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds float, float* %b, i32 2
+  %i5 = load float, float* %arrayidx5, align 4
   %add3 = fadd float %i4, %i5
   %call3 = tail call float @llvm.powi.f32(float %add3,i32 %P) nounwind readnone
 
-  %arrayidx6 = getelementptr inbounds float* %a, i32 3
-  %i6 = load float* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds float* %b, i32 3
-  %i7 = load float* %arrayidx7, align 4
+  %arrayidx6 = getelementptr inbounds float, float* %a, i32 3
+  %i6 = load float, float* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %b, i32 3
+  %i7 = load float, float* %arrayidx7, align 4
   %add4 = fadd float %i6, %i7
   %call4 = tail call float @llvm.powi.f32(float %add4,i32 %Q) nounwind readnone
 
   store float %call1, float* %c, align 4
-  %arrayidx8 = getelementptr inbounds float* %c, i32 1
+  %arrayidx8 = getelementptr inbounds float, float* %c, i32 1
   store float %call2, float* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds float* %c, i32 2
+  %arrayidx9 = getelementptr inbounds float, float* %c, i32 2
   store float %call3, float* %arrayidx9, align 4
-  %arrayidx10 = getelementptr inbounds float* %c, i32 3
+  %arrayidx10 = getelementptr inbounds float, float* %c, i32 3
   store float %call4, float* %arrayidx10, align 4
   ret void
 
diff --git a/test/Transforms/SLPVectorizer/X86/long_chains.ll b/test/Transforms/SLPVectorizer/X86/long_chains.ll
index 5af3e6d6e903..f87dabf4c9fe 100644
--- a/test/Transforms/SLPVectorizer/X86/long_chains.ll
+++ b/test/Transforms/SLPVectorizer/X86/long_chains.ll
@@ -13,9 +13,9 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: ret
 define i32 @test(double* nocapture %A, i8* nocapture %B) {
 entry:
-  %0 = load i8* %B, align 1
-  %arrayidx1 = getelementptr inbounds i8* %B, i64 1
-  %1 = load i8* %arrayidx1, align 1
+  %0 = load i8, i8* %B, align 1
+  %arrayidx1 = getelementptr inbounds i8, i8* %B, i64 1
+  %1 = load i8, i8* %arrayidx1, align 1
   %add = add i8 %0, 3
   %add4 = add i8 %1, 3
   %conv6 = sitofp i8 %add to double
@@ -41,7 +41,7 @@ entry:
   %mul25 = fmul double %add22, %add22
   %add26 = fadd double %mul25, 1.000000e+00
   store double %add24, double* %A, align 8
-  %arrayidx28 = getelementptr inbounds double* %A, i64 1
+  %arrayidx28 = getelementptr inbounds double, double* %A, i64 1
   store double %add26, double* %arrayidx28, align 8
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/loopinvariant.ll b/test/Transforms/SLPVectorizer/X86/loopinvariant.ll
index bc12926e3fe6..0c16c34a1888 100644
--- a/test/Transforms/SLPVectorizer/X86/loopinvariant.ll
+++ b/test/Transforms/SLPVectorizer/X86/loopinvariant.ll
@@ -18,43 +18,43 @@ entry:
 
 for.body:                                         ; preds = %entry, %for.body
   %indvars.iv = phi i64 [ %indvars.iv.next, %for.body ], [ 0, %entry ]
-  %arrayidx = getelementptr inbounds i32* %A, i64 %indvars.iv
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %A, i64 %indvars.iv
+  %0 = load i32, i32* %arrayidx, align 4
   %add1 = add nsw i32 %0, %n
   store i32 %add1, i32* %arrayidx, align 4
   %1 = or i64 %indvars.iv, 1
-  %arrayidx4 = getelementptr inbounds i32* %A, i64 %1
-  %2 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %A, i64 %1
+  %2 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %2, %n
   store i32 %add5, i32* %arrayidx4, align 4
   %3 = or i64 %indvars.iv, 2
-  %arrayidx8 = getelementptr inbounds i32* %A, i64 %3
-  %4 = load i32* %arrayidx8, align 4
+  %arrayidx8 = getelementptr inbounds i32, i32* %A, i64 %3
+  %4 = load i32, i32* %arrayidx8, align 4
   %add9 = add nsw i32 %4, %n
   store i32 %add9, i32* %arrayidx8, align 4
   %5 = or i64 %indvars.iv, 3
-  %arrayidx12 = getelementptr inbounds i32* %A, i64 %5
-  %6 = load i32* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds i32, i32* %A, i64 %5
+  %6 = load i32, i32* %arrayidx12, align 4
   %add13 = add nsw i32 %6, %n
   store i32 %add13, i32* %arrayidx12, align 4
   %7 = or i64 %indvars.iv, 4
-  %arrayidx16 = getelementptr inbounds i32* %A, i64 %7
-  %8 = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds i32, i32* %A, i64 %7
+  %8 = load i32, i32* %arrayidx16, align 4
   %add17 = add nsw i32 %8, %n
   store i32 %add17, i32* %arrayidx16, align 4
   %9 = or i64 %indvars.iv, 5
-  %arrayidx20 = getelementptr inbounds i32* %A, i64 %9
-  %10 = load i32* %arrayidx20, align 4
+  %arrayidx20 = getelementptr inbounds i32, i32* %A, i64 %9
+  %10 = load i32, i32* %arrayidx20, align 4
   %add21 = add nsw i32 %10, %n
   store i32 %add21, i32* %arrayidx20, align 4
   %11 = or i64 %indvars.iv, 6
-  %arrayidx24 = getelementptr inbounds i32* %A, i64 %11
-  %12 = load i32* %arrayidx24, align 4
+  %arrayidx24 = getelementptr inbounds i32, i32* %A, i64 %11
+  %12 = load i32, i32* %arrayidx24, align 4
   %add25 = add nsw i32 %12, %n
   store i32 %add25, i32* %arrayidx24, align 4
   %13 = or i64 %indvars.iv, 7
-  %arrayidx28 = getelementptr inbounds i32* %A, i64 %13
-  %14 = load i32* %arrayidx28, align 4
+  %arrayidx28 = getelementptr inbounds i32, i32* %A, i64 %13
+  %14 = load i32, i32* %arrayidx28, align 4
   %add29 = add nsw i32 %14, %n
   store i32 %add29, i32* %arrayidx28, align 4
   %indvars.iv.next = add i64 %indvars.iv, 8
diff --git a/test/Transforms/SLPVectorizer/X86/metadata.ll b/test/Transforms/SLPVectorizer/X86/metadata.ll
index e021cca23e34..ebef6b53c6a4 100644
--- a/test/Transforms/SLPVectorizer/X86/metadata.ll
+++ b/test/Transforms/SLPVectorizer/X86/metadata.ll
@@ -12,16 +12,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define void @test1(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8, !tbaa !4
-  %i1 = load double* %b, align 8, !tbaa !4
+  %i0 = load double, double* %a, align 8, !tbaa !4
+  %i1 = load double, double* %b, align 8, !tbaa !4
   %mul = fmul double %i0, %i1, !fpmath !0
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8, !tbaa !4
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8, !tbaa !4
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8, !tbaa !4
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8, !tbaa !4
   %mul5 = fmul double %i3, %i4, !fpmath !0
   store double %mul, double* %c, align 8, !tbaa !4
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8, !tbaa !4
   ret void
 }
@@ -35,17 +35,17 @@ entry:
 
 define void @test2(double* %a, double* %b, i8* %e) {
 entry:
-  %i0 = load double* %a, align 8, !tbaa !4
-  %i1 = load double* %b, align 8, !tbaa !4
+  %i0 = load double, double* %a, align 8, !tbaa !4
+  %i1 = load double, double* %b, align 8, !tbaa !4
   %mul = fmul double %i0, %i1, !fpmath !1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8, !tbaa !4
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8, !tbaa !4
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8, !tbaa !4
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8, !tbaa !4
   %mul5 = fmul double %i3, %i4, !fpmath !1
   %c = bitcast i8* %e to double*
   store double %mul, double* %c, align 8, !tbaa !4
-  %carrayidx5 = getelementptr inbounds i8* %e, i64 8
+  %carrayidx5 = getelementptr inbounds i8, i8* %e, i64 8
   %arrayidx5 = bitcast i8* %carrayidx5 to double*
   store double %mul5, double* %arrayidx5, align 8, !tbaa !4
   ret void
diff --git a/test/Transforms/SLPVectorizer/X86/multi_block.ll b/test/Transforms/SLPVectorizer/X86/multi_block.ll
index 2f1cc74d05ff..b381d06688bd 100644
--- a/test/Transforms/SLPVectorizer/X86/multi_block.ll
+++ b/test/Transforms/SLPVectorizer/X86/multi_block.ll
@@ -25,16 +25,16 @@ target triple = "x86_64-apple-macosx10.7.0"
 ;CHECK: store <2 x double>
 ;CHECK: ret
 define i32 @bar(double* nocapture %A, i32 %d) {
-  %1 = load double* %A, align 8
-  %2 = getelementptr inbounds double* %A, i64 1
-  %3 = load double* %2, align 8
+  %1 = load double, double* %A, align 8
+  %2 = getelementptr inbounds double, double* %A, i64 1
+  %3 = load double, double* %2, align 8
   %4 = fptrunc double %1 to float
   %5 = fptrunc double %3 to float
   %6 = icmp eq i32 %d, 0
   br i1 %6, label %9, label %7
 
 ; <label>:7                                       ; preds = %0
-  %8 = tail call i32 (...)* @foo()
+  %8 = tail call i32 (...) @foo()
   br label %9
 
 ; <label>:9                                       ; preds = %0, %7
@@ -42,11 +42,11 @@ define i32 @bar(double* nocapture %A, i32 %d) {
   %11 = fadd float %5, 5.000000e+00
   %12 = fpext float %10 to double
   %13 = fadd double %12, 9.000000e+00
-  %14 = getelementptr inbounds double* %A, i64 8
+  %14 = getelementptr inbounds double, double* %A, i64 8
   store double %13, double* %14, align 8
   %15 = fpext float %11 to double
   %16 = fadd double %15, 5.000000e+00
-  %17 = getelementptr inbounds double* %A, i64 9
+  %17 = getelementptr inbounds double, double* %A, i64 9
   store double %16, double* %17, align 8
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/multi_user.ll b/test/Transforms/SLPVectorizer/X86/multi_user.ll
index 63a77e4b6733..3197f6db266f 100644
--- a/test/Transforms/SLPVectorizer/X86/multi_user.ll
+++ b/test/Transforms/SLPVectorizer/X86/multi_user.ll
@@ -20,27 +20,27 @@ target triple = "x86_64-apple-macosx10.7.0"
 define i32 @foo(i32* nocapture %A, i32 %n) {
   %1 = mul nsw i32 %n, 5
   %2 = add nsw i32 %1, 7
-  %3 = load i32* %A, align 4
+  %3 = load i32, i32* %A, align 4
   %4 = add nsw i32 %2, %3
   store i32 %4, i32* %A, align 4
   %5 = add nsw i32 %1, 8
-  %6 = getelementptr inbounds i32* %A, i64 1
-  %7 = load i32* %6, align 4
+  %6 = getelementptr inbounds i32, i32* %A, i64 1
+  %7 = load i32, i32* %6, align 4
   %8 = add nsw i32 %5, %7
   store i32 %8, i32* %6, align 4
   %9 = add nsw i32 %1, 9
-  %10 = getelementptr inbounds i32* %A, i64 2
-  %11 = load i32* %10, align 4
+  %10 = getelementptr inbounds i32, i32* %A, i64 2
+  %11 = load i32, i32* %10, align 4
   %12 = add nsw i32 %9, %11
   store i32 %12, i32* %10, align 4
   %13 = add nsw i32 %1, 10
-  %14 = getelementptr inbounds i32* %A, i64 3
-  %15 = load i32* %14, align 4
+  %14 = getelementptr inbounds i32, i32* %A, i64 3
+  %15 = load i32, i32* %14, align 4
   %16 = add nsw i32 %13, %15
   store i32 %16, i32* %14, align 4
   %17 = add nsw i32 %1, 11
-  %18 = getelementptr inbounds i32* %A, i64 4
-  %19 = load i32* %18, align 4
+  %18 = getelementptr inbounds i32, i32* %A, i64 4
+  %19 = load i32, i32* %18, align 4
   %20 = add nsw i32 %17, %19
   store i32 %20, i32* %18, align 4
   ret i32 undef
diff --git a/test/Transforms/SLPVectorizer/X86/odd_store.ll b/test/Transforms/SLPVectorizer/X86/odd_store.ll
index 027f6016e2b9..25f049adf479 100644
--- a/test/Transforms/SLPVectorizer/X86/odd_store.ll
+++ b/test/Transforms/SLPVectorizer/X86/odd_store.ll
@@ -18,28 +18,28 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK-NOT: store <3 x i8>
 ;CHECK: ret
 define i32 @foo(i8* noalias nocapture %A, float* noalias nocapture %B, float %T) {
-  %1 = getelementptr inbounds float* %B, i64 10
-  %2 = load float* %1, align 4
+  %1 = getelementptr inbounds float, float* %B, i64 10
+  %2 = load float, float* %1, align 4
   %3 = fmul float %2, %T
   %4 = fpext float %3 to double
   %5 = fadd double %4, 4.000000e+00
   %6 = fptosi double %5 to i8
   store i8 %6, i8* %A, align 1
-  %7 = getelementptr inbounds float* %B, i64 11
-  %8 = load float* %7, align 4
+  %7 = getelementptr inbounds float, float* %B, i64 11
+  %8 = load float, float* %7, align 4
   %9 = fmul float %8, %T
   %10 = fpext float %9 to double
   %11 = fadd double %10, 5.000000e+00
   %12 = fptosi double %11 to i8
-  %13 = getelementptr inbounds i8* %A, i64 1
+  %13 = getelementptr inbounds i8, i8* %A, i64 1
   store i8 %12, i8* %13, align 1
-  %14 = getelementptr inbounds float* %B, i64 12
-  %15 = load float* %14, align 4
+  %14 = getelementptr inbounds float, float* %B, i64 12
+  %15 = load float, float* %14, align 4
   %16 = fmul float %15, %T
   %17 = fpext float %16 to double
   %18 = fadd double %17, 6.000000e+00
   %19 = fptosi double %18 to i8
-  %20 = getelementptr inbounds i8* %A, i64 2
+  %20 = getelementptr inbounds i8, i8* %A, i64 2
   store i8 %19, i8* %20, align 1
   ret i32 undef
 }
diff --git a/test/Transforms/SLPVectorizer/X86/operandorder.ll b/test/Transforms/SLPVectorizer/X86/operandorder.ll
index c5322a839ed1..109540bc769d 100644
--- a/test/Transforms/SLPVectorizer/X86/operandorder.ll
+++ b/test/Transforms/SLPVectorizer/X86/operandorder.ll
@@ -13,12 +13,12 @@ target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define void @shuffle_operands1(double * noalias %from, double * noalias %to,
                                double %v1, double %v2) {
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %v0_1, %v1
   %v1_2 = fadd double %v2, %v0_2
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
   ret void
@@ -35,12 +35,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %v0_1, %p
   %v1_2 = fadd double %v0_1, %v0_2
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -60,12 +60,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %p, %v0_1
   %v1_2 = fadd double %v0_2, %v0_1
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -85,12 +85,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %p, %v0_1
   %v1_2 = fadd double %v0_1, %v0_2
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -111,12 +111,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %v0_2, %v0_1
   %v1_2 = fadd double %p, %v0_1
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -136,12 +136,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %v0_1, %v0_2
   %v1_2 = fadd double %p, %v0_1
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -162,12 +162,12 @@ br label %lp
 
 lp:
   %p = phi double [ 1.000000e+00, %lp ], [ 0.000000e+00, %entry ]
-  %from_1 = getelementptr double *%from, i64 1
-  %v0_1 = load double * %from
-  %v0_2 = load double * %from_1
+  %from_1 = getelementptr double, double *%from, i64 1
+  %v0_1 = load double , double * %from
+  %v0_2 = load double , double * %from_1
   %v1_1 = fadd double %v0_1, %v0_2
   %v1_2 = fadd double %v0_1, %p
-  %to_2 = getelementptr double * %to, i64 1
+  %to_2 = getelementptr double, double * %to, i64 1
   store double %v1_1, double *%to
   store double %v1_2, double *%to_2
 br i1 undef, label %lp, label %ext
@@ -181,7 +181,7 @@ ext:
 
 ; CHECK-LABEL: good_load_order
 
-; CHECK: %[[V1:[0-9]+]] = load <4 x float>*
+; CHECK: %[[V1:[0-9]+]] = load <4 x float>, <4 x float>*
 ; CHECK: %[[V2:[0-9]+]] = insertelement <4 x float> undef, float %1, i32 0
 ; CHECK: %[[V3:[0-9]+]] = shufflevector <4 x float> %[[V2]], <4 x float> %[[V1]], <4 x i32> <i32 0, i32 4, i32 5, i32 6>
 ; CHECK:                = fmul <4 x float> %[[V1]], %[[V3]]
@@ -193,36 +193,36 @@ entry:
   br label %for.cond1.preheader
 
 for.cond1.preheader:
-  %0 = load float* getelementptr inbounds ([32000 x float]* @a, i64 0, i64 0), align 16
+  %0 = load float, float* getelementptr inbounds ([32000 x float], [32000 x float]* @a, i64 0, i64 0), align 16
   br label %for.body3
 
 for.body3:
   %1 = phi float [ %0, %for.cond1.preheader ], [ %10, %for.body3 ]
   %indvars.iv = phi i64 [ 0, %for.cond1.preheader ], [ %indvars.iv.next, %for.body3 ]
   %2 = add nsw i64 %indvars.iv, 1
-  %arrayidx = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %2
-  %3 = load float* %arrayidx, align 4
-  %arrayidx5 = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %indvars.iv
+  %arrayidx = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %2
+  %3 = load float, float* %arrayidx, align 4
+  %arrayidx5 = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %indvars.iv
   %mul6 = fmul float %3, %1
   store float %mul6, float* %arrayidx5, align 4
   %4 = add nsw i64 %indvars.iv, 2
-  %arrayidx11 = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %4
-  %5 = load float* %arrayidx11, align 4
+  %arrayidx11 = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %4
+  %5 = load float, float* %arrayidx11, align 4
   %mul15 = fmul float %5, %3
   store float %mul15, float* %arrayidx, align 4
   %6 = add nsw i64 %indvars.iv, 3
-  %arrayidx21 = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %6
-  %7 = load float* %arrayidx21, align 4
+  %arrayidx21 = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %6
+  %7 = load float, float* %arrayidx21, align 4
   %mul25 = fmul float %7, %5
   store float %mul25, float* %arrayidx11, align 4
   %8 = add nsw i64 %indvars.iv, 4
-  %arrayidx31 = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %8
-  %9 = load float* %arrayidx31, align 4
+  %arrayidx31 = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %8
+  %9 = load float, float* %arrayidx31, align 4
   %mul35 = fmul float %9, %7
   store float %mul35, float* %arrayidx21, align 4
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 5
-  %arrayidx41 = getelementptr inbounds [32000 x float]* @a, i64 0, i64 %indvars.iv.next
-  %10 = load float* %arrayidx41, align 4
+  %arrayidx41 = getelementptr inbounds [32000 x float], [32000 x float]* @a, i64 0, i64 %indvars.iv.next
+  %10 = load float, float* %arrayidx41, align 4
   %mul45 = fmul float %10, %9
   store float %mul45, float* %arrayidx31, align 4
   %11 = trunc i64 %indvars.iv.next to i32
@@ -232,3 +232,113 @@ for.body3:
 for.end:
   ret void
 }
+
+; Check vectorization of following code for double data type-
+;  c[0] = a[0]+b[0];
+;  c[1] = b[1]+a[1]; // swapped b[1] and a[1]
+
+; CHECK-LABEL: load_reorder_double
+; CHECK: load <2 x double>, <2 x double>*
+; CHECK: fadd <2 x double>
+define void @load_reorder_double(double* nocapture %c, double* noalias nocapture readonly %a, double* noalias nocapture readonly %b){
+  %1 = load double, double* %a
+  %2 = load double, double* %b
+  %3 = fadd double %1, %2
+  store double %3, double* %c
+  %4 = getelementptr inbounds double, double* %b, i64 1
+  %5 = load double, double* %4
+  %6 = getelementptr inbounds double, double* %a, i64 1
+  %7 = load double, double* %6
+  %8 = fadd double %5, %7
+  %9 = getelementptr inbounds double, double* %c, i64 1
+  store double %8, double* %9
+  ret void
+}
+
+; Check vectorization of following code for float data type-
+;  c[0] = a[0]+b[0];
+;  c[1] = b[1]+a[1]; // swapped b[1] and a[1]
+;  c[2] = a[2]+b[2];
+;  c[3] = a[3]+b[3];
+
+; CHECK-LABEL: load_reorder_float
+; CHECK: load <4 x float>, <4 x float>*
+; CHECK: fadd <4 x float>
+define void @load_reorder_float(float* nocapture %c, float* noalias nocapture readonly %a, float* noalias nocapture readonly %b){
+  %1 = load float, float* %a
+  %2 = load float, float* %b
+  %3 = fadd float %1, %2
+  store float %3, float* %c
+  %4 = getelementptr inbounds float, float* %b, i64 1
+  %5 = load float, float* %4
+  %6 = getelementptr inbounds float, float* %a, i64 1
+  %7 = load float, float* %6
+  %8 = fadd float %5, %7
+  %9 = getelementptr inbounds float, float* %c, i64 1
+  store float %8, float* %9
+  %10 = getelementptr inbounds float, float* %a, i64 2
+  %11 = load float, float* %10
+  %12 = getelementptr inbounds float, float* %b, i64 2
+  %13 = load float, float* %12
+  %14 = fadd float %11, %13
+  %15 = getelementptr inbounds float, float* %c, i64 2
+  store float %14, float* %15
+  %16 = getelementptr inbounds float, float* %a, i64 3
+  %17 = load float, float* %16
+  %18 = getelementptr inbounds float, float* %b, i64 3
+  %19 = load float, float* %18
+  %20 = fadd float %17, %19
+  %21 = getelementptr inbounds float, float* %c, i64 3
+  store float %20, float* %21
+  ret void
+}
+
+; Check we properly reorder the below code so that it gets vectorized optimally-
+; a[0] = (b[0]+c[0])+d[0];
+; a[1] = d[1]+(b[1]+c[1]);
+; a[2] = (b[2]+c[2])+d[2];
+; a[3] = (b[3]+c[3])+d[3];
+
+; CHECK-LABEL: opcode_reorder
+; CHECK: load <4 x float>, <4 x float>*
+; CHECK: fadd <4 x float>
+define void @opcode_reorder(float* noalias nocapture %a, float* noalias nocapture readonly %b, 
+                            float* noalias nocapture readonly %c,float* noalias nocapture readonly %d){
+  %1 = load float, float* %b
+  %2 = load float, float* %c
+  %3 = fadd float %1, %2
+  %4 = load float, float* %d
+  %5 = fadd float %3, %4
+  store float %5, float* %a
+  %6 = getelementptr inbounds float, float* %d, i64 1
+  %7 = load float, float* %6
+  %8 = getelementptr inbounds float, float* %b, i64 1
+  %9 = load float, float* %8
+  %10 = getelementptr inbounds float, float* %c, i64 1
+  %11 = load float, float* %10
+  %12 = fadd float %9, %11
+  %13 = fadd float %7, %12
+  %14 = getelementptr inbounds float, float* %a, i64 1
+  store float %13, float* %14
+  %15 = getelementptr inbounds float, float* %b, i64 2
+  %16 = load float, float* %15
+  %17 = getelementptr inbounds float, float* %c, i64 2
+  %18 = load float, float* %17
+  %19 = fadd float %16, %18
+  %20 = getelementptr inbounds float, float* %d, i64 2
+  %21 = load float, float* %20
+  %22 = fadd float %19, %21
+  %23 = getelementptr inbounds float, float* %a, i64 2
+  store float %22, float* %23
+  %24 = getelementptr inbounds float, float* %b, i64 3
+  %25 = load float, float* %24
+  %26 = getelementptr inbounds float, float* %c, i64 3
+  %27 = load float, float* %26
+  %28 = fadd float %25, %27
+  %29 = getelementptr inbounds float, float* %d, i64 3
+  %30 = load float, float* %29
+  %31 = fadd float %28, %30
+  %32 = getelementptr inbounds float, float* %a, i64 3
+  store float %31, float* %32
+  ret void
+}
diff --git a/test/Transforms/SLPVectorizer/X86/opt.ll b/test/Transforms/SLPVectorizer/X86/opt.ll
index 14137c11ee41..824e9992af03 100644
--- a/test/Transforms/SLPVectorizer/X86/opt.ll
+++ b/test/Transforms/SLPVectorizer/X86/opt.ll
@@ -15,16 +15,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define void @test1(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/ordering.ll b/test/Transforms/SLPVectorizer/X86/ordering.ll
index d2ecd4546ddb..0fa72c94c272 100644
--- a/test/Transforms/SLPVectorizer/X86/ordering.ll
+++ b/test/Transforms/SLPVectorizer/X86/ordering.ll
@@ -5,7 +5,7 @@ target triple = "x86_64-apple-macosx10.8.0"
 
 define void @updateModelQPFrame(i32 %m_Bits) {
 entry:
-  %0 = load double* undef, align 8
+  %0 = load double, double* undef, align 8
   %mul = fmul double undef, %0
   %mul2 = fmul double undef, %mul
   %mul4 = fmul double %0, %mul2
diff --git a/test/Transforms/SLPVectorizer/X86/phi.ll b/test/Transforms/SLPVectorizer/X86/phi.ll
index 0c53b60f3df1..ef94467f509a 100644
--- a/test/Transforms/SLPVectorizer/X86/phi.ll
+++ b/test/Transforms/SLPVectorizer/X86/phi.ll
@@ -29,17 +29,17 @@ entry:
   br i1 %tobool, label %if.else, label %if.end
 
 if.else:                                          ; preds = %entry
-  %arrayidx = getelementptr inbounds double* %A, i64 10
-  %0 = load double* %arrayidx, align 8
-  %arrayidx1 = getelementptr inbounds double* %A, i64 11
-  %1 = load double* %arrayidx1, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 10
+  %0 = load double, double* %arrayidx, align 8
+  %arrayidx1 = getelementptr inbounds double, double* %A, i64 11
+  %1 = load double, double* %arrayidx1, align 8
   br label %if.end
 
 if.end:                                           ; preds = %entry, %if.else
   %A0.0 = phi double [ %0, %if.else ], [ 3.000000e+00, %entry ]
   %A1.0 = phi double [ %1, %if.else ], [ 5.000000e+00, %entry ]
   store double %A0.0, double* %A, align 8
-  %arrayidx3 = getelementptr inbounds double* %A, i64 1
+  %arrayidx3 = getelementptr inbounds double, double* %A, i64 1
   store double %A1.0, double* %arrayidx3, align 8
   ret i32 undef
 }
@@ -69,9 +69,9 @@ if.end:                                           ; preds = %entry, %if.else
 ;CHECK: ret
 define i32 @foo2(double* noalias nocapture %B, double* noalias nocapture %A, i32 %n, i32 %m) #0 {
 entry:
-  %arrayidx = getelementptr inbounds double* %A, i64 1
-  %0 = load double* %arrayidx, align 8
-  %1 = load double* %A, align 8
+  %arrayidx = getelementptr inbounds double, double* %A, i64 1
+  %0 = load double, double* %arrayidx, align 8
+  %1 = load double, double* %A, align 8
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %entry
@@ -90,7 +90,7 @@ for.body:                                         ; preds = %for.body, %entry
 
 for.end:                                          ; preds = %for.body
   store double %add5, double* %B, align 8
-  %arrayidx7 = getelementptr inbounds double* %B, i64 1
+  %arrayidx7 = getelementptr inbounds double, double* %B, i64 1
   store double %add4, double* %arrayidx7, align 8
   ret i32 0
 }
@@ -123,15 +123,15 @@ for.end:                                          ; preds = %for.body
 
 define float @foo3(float* nocapture readonly %A) #0 {
 entry:
-  %0 = load float* %A, align 4
-  %arrayidx1 = getelementptr inbounds float* %A, i64 1
-  %1 = load float* %arrayidx1, align 4
-  %arrayidx2 = getelementptr inbounds float* %A, i64 2
-  %2 = load float* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds float* %A, i64 3
-  %3 = load float* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds float* %A, i64 4
-  %4 = load float* %arrayidx4, align 4
+  %0 = load float, float* %A, align 4
+  %arrayidx1 = getelementptr inbounds float, float* %A, i64 1
+  %1 = load float, float* %arrayidx1, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 2
+  %2 = load float, float* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %A, i64 3
+  %3 = load float, float* %arrayidx3, align 4
+  %arrayidx4 = getelementptr inbounds float, float* %A, i64 4
+  %4 = load float, float* %arrayidx4, align 4
   br label %for.body
 
 for.body:                                         ; preds = %for.body, %entry
@@ -148,18 +148,18 @@ for.body:                                         ; preds = %for.body, %entry
   %mul10 = fmul float %5, 8.000000e+00
   %add11 = fadd float %G.053, %mul10
   %7 = add nsw i64 %indvars.iv, 2
-  %arrayidx14 = getelementptr inbounds float* %A, i64 %7
-  %8 = load float* %arrayidx14, align 4
+  %arrayidx14 = getelementptr inbounds float, float* %A, i64 %7
+  %8 = load float, float* %arrayidx14, align 4
   %mul15 = fmul float %8, 9.000000e+00
   %add16 = fadd float %B.054, %mul15
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 3
-  %arrayidx19 = getelementptr inbounds float* %A, i64 %indvars.iv.next
-  %9 = load float* %arrayidx19, align 4
+  %arrayidx19 = getelementptr inbounds float, float* %A, i64 %indvars.iv.next
+  %9 = load float, float* %arrayidx19, align 4
   %mul20 = fmul float %9, 1.000000e+01
   %add21 = fadd float %Y.055, %mul20
   %10 = add nsw i64 %indvars.iv, 4
-  %arrayidx24 = getelementptr inbounds float* %A, i64 %10
-  %11 = load float* %arrayidx24, align 4
+  %arrayidx24 = getelementptr inbounds float, float* %A, i64 %10
+  %11 = load float, float* %arrayidx24, align 4
   %mul25 = fmul float %11, 1.100000e+01
   %add26 = fadd float %P.056, %mul25
   %12 = trunc i64 %indvars.iv.next to i32
@@ -215,22 +215,22 @@ define void @test(x86_fp80* %i1, x86_fp80* %i2, x86_fp80* %o) {
 ; We disable the vectorization of x86_fp80 for now. 
 
 entry:
-  %i1.0 = load x86_fp80* %i1, align 16
-  %i1.gep1 = getelementptr x86_fp80* %i1, i64 1
-  %i1.1 = load x86_fp80* %i1.gep1, align 16
-; CHECK: load x86_fp80*
-; CHECK: load x86_fp80*
+  %i1.0 = load x86_fp80, x86_fp80* %i1, align 16
+  %i1.gep1 = getelementptr x86_fp80, x86_fp80* %i1, i64 1
+  %i1.1 = load x86_fp80, x86_fp80* %i1.gep1, align 16
+; CHECK: load x86_fp80, x86_fp80*
+; CHECK: load x86_fp80, x86_fp80*
 ; CHECK-NOT: insertelement <2 x x86_fp80>
 ; CHECK-NOT: insertelement <2 x x86_fp80>
   br i1 undef, label %then, label %end
 
 then:
-  %i2.gep0 = getelementptr inbounds x86_fp80* %i2, i64 0
-  %i2.0 = load x86_fp80* %i2.gep0, align 16
-  %i2.gep1 = getelementptr inbounds x86_fp80* %i2, i64 1
-  %i2.1 = load x86_fp80* %i2.gep1, align 16
-; CHECK: load x86_fp80*
-; CHECK: load x86_fp80*
+  %i2.gep0 = getelementptr inbounds x86_fp80, x86_fp80* %i2, i64 0
+  %i2.0 = load x86_fp80, x86_fp80* %i2.gep0, align 16
+  %i2.gep1 = getelementptr inbounds x86_fp80, x86_fp80* %i2, i64 1
+  %i2.1 = load x86_fp80, x86_fp80* %i2.gep1, align 16
+; CHECK: load x86_fp80, x86_fp80*
+; CHECK: load x86_fp80, x86_fp80*
 ; CHECK-NOT: insertelement <2 x x86_fp80>
 ; CHECK-NOT: insertelement <2 x x86_fp80>
   br label %end
@@ -242,7 +242,7 @@ end:
 ; CHECK-NOT: extractelement <2 x x86_fp80>
 ; CHECK-NOT: extractelement <2 x x86_fp80>
   store x86_fp80 %phi0, x86_fp80* %o, align 16
-  %o.gep1 = getelementptr inbounds x86_fp80* %o, i64 1
+  %o.gep1 = getelementptr inbounds x86_fp80, x86_fp80* %o, i64 1
   store x86_fp80 %phi1, x86_fp80* %o.gep1, align 16
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/phi3.ll b/test/Transforms/SLPVectorizer/X86/phi3.ll
index fd8d36137201..61628301aec7 100644
--- a/test/Transforms/SLPVectorizer/X86/phi3.ll
+++ b/test/Transforms/SLPVectorizer/X86/phi3.ll
@@ -12,7 +12,7 @@ declare %struct.GPar.0.16.26* @Rf_gpptr(...)
 define void @Rf_GReset() {
 entry:
   %sub = fsub double -0.000000e+00, undef
-  %0 = load double* @d, align 8
+  %0 = load double, double* @d, align 8
   %sub1 = fsub double -0.000000e+00, %0
   br i1 icmp eq (%struct.GPar.0.16.26* (...)* inttoptr (i64 115 to %struct.GPar.0.16.26* (...)*), %struct.GPar.0.16.26* (...)* @Rf_gpptr), label %if.then, label %if.end7
 
diff --git a/test/Transforms/SLPVectorizer/X86/phi_overalignedtype.ll b/test/Transforms/SLPVectorizer/X86/phi_overalignedtype.ll
index 520e6729de0c..fa08effcd648 100644
--- a/test/Transforms/SLPVectorizer/X86/phi_overalignedtype.ll
+++ b/test/Transforms/SLPVectorizer/X86/phi_overalignedtype.ll
@@ -12,22 +12,22 @@ define void @test(double* %i1, double* %i2, double* %o) {
 ; size is less than the alignment, and through various different GEP formations.
 
 entry:
-  %i1.0 = load double* %i1, align 16
-  %i1.gep1 = getelementptr double* %i1, i64 1
-  %i1.1 = load double* %i1.gep1, align 16
-; CHECK: load double*
-; CHECK: load double*
+  %i1.0 = load double, double* %i1, align 16
+  %i1.gep1 = getelementptr double, double* %i1, i64 1
+  %i1.1 = load double, double* %i1.gep1, align 16
+; CHECK: load double, double*
+; CHECK: load double, double*
 ; CHECK: insertelement <2 x double>
 ; CHECK: insertelement <2 x double>
   br i1 undef, label %then, label %end
 
 then:
-  %i2.gep0 = getelementptr inbounds double* %i2, i64 0
-  %i2.0 = load double* %i2.gep0, align 16
-  %i2.gep1 = getelementptr inbounds double* %i2, i64 1
-  %i2.1 = load double* %i2.gep1, align 16
-; CHECK: load double*
-; CHECK: load double*
+  %i2.gep0 = getelementptr inbounds double, double* %i2, i64 0
+  %i2.0 = load double, double* %i2.gep0, align 16
+  %i2.gep1 = getelementptr inbounds double, double* %i2, i64 1
+  %i2.1 = load double, double* %i2.gep1, align 16
+; CHECK: load double, double*
+; CHECK: load double, double*
 ; CHECK: insertelement <2 x double>
 ; CHECK: insertelement <2 x double>
   br label %end
@@ -39,7 +39,7 @@ end:
 ; CHECK: extractelement <2 x double>
 ; CHECK: extractelement <2 x double>
   store double %phi0, double* %o, align 16
-  %o.gep1 = getelementptr inbounds double* %o, i64 1
+  %o.gep1 = getelementptr inbounds double, double* %o, i64 1
   store double %phi1, double* %o.gep1, align 16
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/powof2div.ll b/test/Transforms/SLPVectorizer/X86/powof2div.ll
index 7aa1efde6f06..a97b870f3f44 100644
--- a/test/Transforms/SLPVectorizer/X86/powof2div.ll
+++ b/test/Transforms/SLPVectorizer/X86/powof2div.ll
@@ -4,39 +4,39 @@ target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
 
 ;CHECK-LABEL: @powof2div(
-;CHECK: load <4 x i32>*
+;CHECK: load <4 x i32>, <4 x i32>*
 ;CHECK: add nsw <4 x i32>
 ;CHECK: sdiv <4 x i32>
 define void @powof2div(i32* noalias nocapture %a, i32* noalias nocapture readonly %b, i32* noalias nocapture readonly %c){
 entry:
-  %0 = load i32* %b, align 4
-  %1 = load i32* %c, align 4
+  %0 = load i32, i32* %b, align 4
+  %1 = load i32, i32* %c, align 4
   %add = add nsw i32 %1, %0
   %div = sdiv i32 %add, 2
   store i32 %div, i32* %a, align 4
-  %arrayidx3 = getelementptr inbounds i32* %b, i64 1
-  %2 = load i32* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds i32* %c, i64 1
-  %3 = load i32* %arrayidx4, align 4
+  %arrayidx3 = getelementptr inbounds i32, i32* %b, i64 1
+  %2 = load i32, i32* %arrayidx3, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %c, i64 1
+  %3 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %3, %2
   %div6 = sdiv i32 %add5, 2
-  %arrayidx7 = getelementptr inbounds i32* %a, i64 1
+  %arrayidx7 = getelementptr inbounds i32, i32* %a, i64 1
   store i32 %div6, i32* %arrayidx7, align 4
-  %arrayidx8 = getelementptr inbounds i32* %b, i64 2
-  %4 = load i32* %arrayidx8, align 4
-  %arrayidx9 = getelementptr inbounds i32* %c, i64 2
-  %5 = load i32* %arrayidx9, align 4
+  %arrayidx8 = getelementptr inbounds i32, i32* %b, i64 2
+  %4 = load i32, i32* %arrayidx8, align 4
+  %arrayidx9 = getelementptr inbounds i32, i32* %c, i64 2
+  %5 = load i32, i32* %arrayidx9, align 4
   %add10 = add nsw i32 %5, %4
   %div11 = sdiv i32 %add10, 2
-  %arrayidx12 = getelementptr inbounds i32* %a, i64 2
+  %arrayidx12 = getelementptr inbounds i32, i32* %a, i64 2
   store i32 %div11, i32* %arrayidx12, align 4
-  %arrayidx13 = getelementptr inbounds i32* %b, i64 3
-  %6 = load i32* %arrayidx13, align 4
-  %arrayidx14 = getelementptr inbounds i32* %c, i64 3
-  %7 = load i32* %arrayidx14, align 4
+  %arrayidx13 = getelementptr inbounds i32, i32* %b, i64 3
+  %6 = load i32, i32* %arrayidx13, align 4
+  %arrayidx14 = getelementptr inbounds i32, i32* %c, i64 3
+  %7 = load i32, i32* %arrayidx14, align 4
   %add15 = add nsw i32 %7, %6
   %div16 = sdiv i32 %add15, 2
-  %arrayidx17 = getelementptr inbounds i32* %a, i64 3
+  %arrayidx17 = getelementptr inbounds i32, i32* %a, i64 3
   store i32 %div16, i32* %arrayidx17, align 4
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/pr16628.ll b/test/Transforms/SLPVectorizer/X86/pr16628.ll
index 3f9d775eeeb6..06abe9127e6f 100644
--- a/test/Transforms/SLPVectorizer/X86/pr16628.ll
+++ b/test/Transforms/SLPVectorizer/X86/pr16628.ll
@@ -10,11 +10,11 @@ target triple = "x86_64-apple-macosx10.9.0"
 ; Function Attrs: nounwind ssp uwtable
 define void @f() {
 entry:
-  %call = tail call i32 (...)* @g()
-  %0 = load i32* @c, align 4
+  %call = tail call i32 (...) @g()
+  %0 = load i32, i32* @c, align 4
   %lnot = icmp eq i32 %0, 0
   %lnot.ext = zext i1 %lnot to i32
-  %1 = load i16* @a, align 2
+  %1 = load i16, i16* @a, align 2
   %lnot2 = icmp eq i16 %1, 0
   %lnot.ext3 = zext i1 %lnot2 to i32
   %or = or i32 %lnot.ext3, %lnot.ext
diff --git a/test/Transforms/SLPVectorizer/X86/pr16899.ll b/test/Transforms/SLPVectorizer/X86/pr16899.ll
index c642f3cfe1d4..0de14ec35856 100644
--- a/test/Transforms/SLPVectorizer/X86/pr16899.ll
+++ b/test/Transforms/SLPVectorizer/X86/pr16899.ll
@@ -7,10 +7,10 @@ target triple = "i386--netbsd"
 ; Function Attrs: noreturn nounwind readonly
 define i32 @fn1() #0 {
 entry:
-  %0 = load i32** @a, align 4, !tbaa !4
-  %1 = load i32* %0, align 4, !tbaa !5
-  %arrayidx1 = getelementptr inbounds i32* %0, i32 1
-  %2 = load i32* %arrayidx1, align 4, !tbaa !5
+  %0 = load i32*, i32** @a, align 4, !tbaa !4
+  %1 = load i32, i32* %0, align 4, !tbaa !5
+  %arrayidx1 = getelementptr inbounds i32, i32* %0, i32 1
+  %2 = load i32, i32* %arrayidx1, align 4, !tbaa !5
   br label %do.body
 
 do.body:                                          ; preds = %do.body, %entry
diff --git a/test/Transforms/SLPVectorizer/X86/pr19657.ll b/test/Transforms/SLPVectorizer/X86/pr19657.ll
index 93523089c28b..a687aec76103 100644
--- a/test/Transforms/SLPVectorizer/X86/pr19657.ll
+++ b/test/Transforms/SLPVectorizer/X86/pr19657.ll
@@ -3,7 +3,7 @@
 target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 target triple = "x86_64-unknown-linux-gnu"
 
-;CHECK: load <2 x double>*
+;CHECK: load <2 x double>, <2 x double>*
 ;CHECK: fadd <2 x double>
 ;CHECK: store <2 x double>
 
@@ -11,61 +11,61 @@ target triple = "x86_64-unknown-linux-gnu"
 define void @foo(double* %x) #0 {
   %1 = alloca double*, align 8
   store double* %x, double** %1, align 8
-  %2 = load double** %1, align 8
-  %3 = getelementptr inbounds double* %2, i64 0
-  %4 = load double* %3, align 8
-  %5 = load double** %1, align 8
-  %6 = getelementptr inbounds double* %5, i64 0
-  %7 = load double* %6, align 8
+  %2 = load double*, double** %1, align 8
+  %3 = getelementptr inbounds double, double* %2, i64 0
+  %4 = load double, double* %3, align 8
+  %5 = load double*, double** %1, align 8
+  %6 = getelementptr inbounds double, double* %5, i64 0
+  %7 = load double, double* %6, align 8
   %8 = fadd double %4, %7
-  %9 = load double** %1, align 8
-  %10 = getelementptr inbounds double* %9, i64 0
-  %11 = load double* %10, align 8
+  %9 = load double*, double** %1, align 8
+  %10 = getelementptr inbounds double, double* %9, i64 0
+  %11 = load double, double* %10, align 8
   %12 = fadd double %8, %11
-  %13 = load double** %1, align 8
-  %14 = getelementptr inbounds double* %13, i64 0
+  %13 = load double*, double** %1, align 8
+  %14 = getelementptr inbounds double, double* %13, i64 0
   store double %12, double* %14, align 8
-  %15 = load double** %1, align 8
-  %16 = getelementptr inbounds double* %15, i64 1
-  %17 = load double* %16, align 8
-  %18 = load double** %1, align 8
-  %19 = getelementptr inbounds double* %18, i64 1
-  %20 = load double* %19, align 8
+  %15 = load double*, double** %1, align 8
+  %16 = getelementptr inbounds double, double* %15, i64 1
+  %17 = load double, double* %16, align 8
+  %18 = load double*, double** %1, align 8
+  %19 = getelementptr inbounds double, double* %18, i64 1
+  %20 = load double, double* %19, align 8
   %21 = fadd double %17, %20
-  %22 = load double** %1, align 8
-  %23 = getelementptr inbounds double* %22, i64 1
-  %24 = load double* %23, align 8
+  %22 = load double*, double** %1, align 8
+  %23 = getelementptr inbounds double, double* %22, i64 1
+  %24 = load double, double* %23, align 8
   %25 = fadd double %21, %24
-  %26 = load double** %1, align 8
-  %27 = getelementptr inbounds double* %26, i64 1
+  %26 = load double*, double** %1, align 8
+  %27 = getelementptr inbounds double, double* %26, i64 1
   store double %25, double* %27, align 8
-  %28 = load double** %1, align 8
-  %29 = getelementptr inbounds double* %28, i64 2
-  %30 = load double* %29, align 8
-  %31 = load double** %1, align 8
-  %32 = getelementptr inbounds double* %31, i64 2
-  %33 = load double* %32, align 8
+  %28 = load double*, double** %1, align 8
+  %29 = getelementptr inbounds double, double* %28, i64 2
+  %30 = load double, double* %29, align 8
+  %31 = load double*, double** %1, align 8
+  %32 = getelementptr inbounds double, double* %31, i64 2
+  %33 = load double, double* %32, align 8
   %34 = fadd double %30, %33
-  %35 = load double** %1, align 8
-  %36 = getelementptr inbounds double* %35, i64 2
-  %37 = load double* %36, align 8
+  %35 = load double*, double** %1, align 8
+  %36 = getelementptr inbounds double, double* %35, i64 2
+  %37 = load double, double* %36, align 8
   %38 = fadd double %34, %37
-  %39 = load double** %1, align 8
-  %40 = getelementptr inbounds double* %39, i64 2
+  %39 = load double*, double** %1, align 8
+  %40 = getelementptr inbounds double, double* %39, i64 2
   store double %38, double* %40, align 8
-  %41 = load double** %1, align 8
-  %42 = getelementptr inbounds double* %41, i64 3
-  %43 = load double* %42, align 8
-  %44 = load double** %1, align 8
-  %45 = getelementptr inbounds double* %44, i64 3
-  %46 = load double* %45, align 8
+  %41 = load double*, double** %1, align 8
+  %42 = getelementptr inbounds double, double* %41, i64 3
+  %43 = load double, double* %42, align 8
+  %44 = load double*, double** %1, align 8
+  %45 = getelementptr inbounds double, double* %44, i64 3
+  %46 = load double, double* %45, align 8
   %47 = fadd double %43, %46
-  %48 = load double** %1, align 8
-  %49 = getelementptr inbounds double* %48, i64 3
-  %50 = load double* %49, align 8
+  %48 = load double*, double** %1, align 8
+  %49 = getelementptr inbounds double, double* %48, i64 3
+  %50 = load double, double* %49, align 8
   %51 = fadd double %47, %50
-  %52 = load double** %1, align 8
-  %53 = getelementptr inbounds double* %52, i64 3
+  %52 = load double*, double** %1, align 8
+  %53 = getelementptr inbounds double, double* %52, i64 3
   store double %51, double* %53, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/propagate_ir_flags.ll b/test/Transforms/SLPVectorizer/X86/propagate_ir_flags.ll
index 3843ef7f62c7..a3b0c8ff0279 100644
--- a/test/Transforms/SLPVectorizer/X86/propagate_ir_flags.ll
+++ b/test/Transforms/SLPVectorizer/X86/propagate_ir_flags.ll
@@ -10,15 +10,15 @@ target triple = "x86_64-unknown-unknown"
 ; CHECK-LABEL: @exact(
 ; CHECK: lshr exact <4 x i32>
 define void @exact(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = lshr exact i32 %load1, 1
   %op2 = lshr exact i32 %load2, 1
@@ -36,15 +36,15 @@ define void @exact(i32* %x) {
 ; CHECK-LABEL: @not_exact(
 ; CHECK: lshr <4 x i32>
 define void @not_exact(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = lshr exact i32 %load1, 1
   %op2 = lshr i32 %load2, 1
@@ -62,15 +62,15 @@ define void @not_exact(i32* %x) {
 ; CHECK-LABEL: @nsw(
 ; CHECK: add nsw <4 x i32>
 define void @nsw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nsw i32 %load1, 1
   %op2 = add nsw i32 %load2, 1
@@ -88,15 +88,15 @@ define void @nsw(i32* %x) {
 ; CHECK-LABEL: @not_nsw(
 ; CHECK: add <4 x i32>
 define void @not_nsw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nsw i32 %load1, 1
   %op2 = add nsw i32 %load2, 1
@@ -114,15 +114,15 @@ define void @not_nsw(i32* %x) {
 ; CHECK-LABEL: @nuw(
 ; CHECK: add nuw <4 x i32>
 define void @nuw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nuw i32 %load1, 1
   %op2 = add nuw i32 %load2, 1
@@ -140,15 +140,15 @@ define void @nuw(i32* %x) {
 ; CHECK-LABEL: @not_nuw(
 ; CHECK: add <4 x i32>
 define void @not_nuw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nuw i32 %load1, 1
   %op2 = add i32 %load2, 1
@@ -166,15 +166,15 @@ define void @not_nuw(i32* %x) {
 ; CHECK-LABEL: @nnan(
 ; CHECK: fadd nnan <4 x float>
 define void @nnan(float* %x) {
-  %idx1 = getelementptr inbounds float* %x, i64 0
-  %idx2 = getelementptr inbounds float* %x, i64 1
-  %idx3 = getelementptr inbounds float* %x, i64 2
-  %idx4 = getelementptr inbounds float* %x, i64 3
+  %idx1 = getelementptr inbounds float, float* %x, i64 0
+  %idx2 = getelementptr inbounds float, float* %x, i64 1
+  %idx3 = getelementptr inbounds float, float* %x, i64 2
+  %idx4 = getelementptr inbounds float, float* %x, i64 3
 
-  %load1 = load float* %idx1, align 4
-  %load2 = load float* %idx2, align 4
-  %load3 = load float* %idx3, align 4
-  %load4 = load float* %idx4, align 4
+  %load1 = load float, float* %idx1, align 4
+  %load2 = load float, float* %idx2, align 4
+  %load3 = load float, float* %idx3, align 4
+  %load4 = load float, float* %idx4, align 4
 
   %op1 = fadd fast nnan float %load1, 1.0
   %op2 = fadd nnan ninf float %load2, 1.0
@@ -192,15 +192,15 @@ define void @nnan(float* %x) {
 ; CHECK-LABEL: @not_nnan(
 ; CHECK: fadd <4 x float>
 define void @not_nnan(float* %x) {
-  %idx1 = getelementptr inbounds float* %x, i64 0
-  %idx2 = getelementptr inbounds float* %x, i64 1
-  %idx3 = getelementptr inbounds float* %x, i64 2
-  %idx4 = getelementptr inbounds float* %x, i64 3
+  %idx1 = getelementptr inbounds float, float* %x, i64 0
+  %idx2 = getelementptr inbounds float, float* %x, i64 1
+  %idx3 = getelementptr inbounds float, float* %x, i64 2
+  %idx4 = getelementptr inbounds float, float* %x, i64 3
 
-  %load1 = load float* %idx1, align 4
-  %load2 = load float* %idx2, align 4
-  %load3 = load float* %idx3, align 4
-  %load4 = load float* %idx4, align 4
+  %load1 = load float, float* %idx1, align 4
+  %load2 = load float, float* %idx2, align 4
+  %load3 = load float, float* %idx3, align 4
+  %load4 = load float, float* %idx4, align 4
 
   %op1 = fadd nnan float %load1, 1.0
   %op2 = fadd ninf float %load2, 1.0
@@ -218,15 +218,15 @@ define void @not_nnan(float* %x) {
 ; CHECK-LABEL: @only_fast(
 ; CHECK: fadd fast <4 x float>
 define void @only_fast(float* %x) {
-  %idx1 = getelementptr inbounds float* %x, i64 0
-  %idx2 = getelementptr inbounds float* %x, i64 1
-  %idx3 = getelementptr inbounds float* %x, i64 2
-  %idx4 = getelementptr inbounds float* %x, i64 3
+  %idx1 = getelementptr inbounds float, float* %x, i64 0
+  %idx2 = getelementptr inbounds float, float* %x, i64 1
+  %idx3 = getelementptr inbounds float, float* %x, i64 2
+  %idx4 = getelementptr inbounds float, float* %x, i64 3
 
-  %load1 = load float* %idx1, align 4
-  %load2 = load float* %idx2, align 4
-  %load3 = load float* %idx3, align 4
-  %load4 = load float* %idx4, align 4
+  %load1 = load float, float* %idx1, align 4
+  %load2 = load float, float* %idx2, align 4
+  %load3 = load float, float* %idx3, align 4
+  %load4 = load float, float* %idx4, align 4
 
   %op1 = fadd fast nnan float %load1, 1.0
   %op2 = fadd fast nnan ninf float %load2, 1.0
@@ -244,15 +244,15 @@ define void @only_fast(float* %x) {
 ; CHECK-LABEL: @only_arcp(
 ; CHECK: fadd arcp <4 x float>
 define void @only_arcp(float* %x) {
-  %idx1 = getelementptr inbounds float* %x, i64 0
-  %idx2 = getelementptr inbounds float* %x, i64 1
-  %idx3 = getelementptr inbounds float* %x, i64 2
-  %idx4 = getelementptr inbounds float* %x, i64 3
+  %idx1 = getelementptr inbounds float, float* %x, i64 0
+  %idx2 = getelementptr inbounds float, float* %x, i64 1
+  %idx3 = getelementptr inbounds float, float* %x, i64 2
+  %idx4 = getelementptr inbounds float, float* %x, i64 3
 
-  %load1 = load float* %idx1, align 4
-  %load2 = load float* %idx2, align 4
-  %load3 = load float* %idx3, align 4
-  %load4 = load float* %idx4, align 4
+  %load1 = load float, float* %idx1, align 4
+  %load2 = load float, float* %idx2, align 4
+  %load3 = load float, float* %idx3, align 4
+  %load4 = load float, float* %idx4, align 4
 
   %op1 = fadd fast float %load1, 1.0
   %op2 = fadd fast float %load2, 1.0
@@ -271,15 +271,15 @@ define void @only_arcp(float* %x) {
 ; CHECK: add nsw <4 x i32>
 ; CHECK: sub nsw <4 x i32>
 define void @addsub_all_nsw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nsw i32 %load1, 1
   %op2 = sub nsw i32 %load2, 1
@@ -298,15 +298,15 @@ define void @addsub_all_nsw(i32* %x) {
 ; CHECK: add nsw <4 x i32>
 ; CHECK: sub <4 x i32>
 define void @addsub_some_nsw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add nsw i32 %load1, 1
   %op2 = sub nsw i32 %load2, 1
@@ -325,15 +325,15 @@ define void @addsub_some_nsw(i32* %x) {
 ; CHECK: add <4 x i32>
 ; CHECK: sub <4 x i32>
 define void @addsub_no_nsw(i32* %x) {
-  %idx1 = getelementptr inbounds i32* %x, i64 0
-  %idx2 = getelementptr inbounds i32* %x, i64 1
-  %idx3 = getelementptr inbounds i32* %x, i64 2
-  %idx4 = getelementptr inbounds i32* %x, i64 3
+  %idx1 = getelementptr inbounds i32, i32* %x, i64 0
+  %idx2 = getelementptr inbounds i32, i32* %x, i64 1
+  %idx3 = getelementptr inbounds i32, i32* %x, i64 2
+  %idx4 = getelementptr inbounds i32, i32* %x, i64 3
 
-  %load1 = load i32* %idx1, align 4
-  %load2 = load i32* %idx2, align 4
-  %load3 = load i32* %idx3, align 4
-  %load4 = load i32* %idx4, align 4
+  %load1 = load i32, i32* %idx1, align 4
+  %load2 = load i32, i32* %idx2, align 4
+  %load3 = load i32, i32* %idx3, align 4
+  %load4 = load i32, i32* %idx4, align 4
 
   %op1 = add i32 %load1, 1
   %op2 = sub nsw i32 %load2, 1
diff --git a/test/Transforms/SLPVectorizer/X86/reduction.ll b/test/Transforms/SLPVectorizer/X86/reduction.ll
index 70b7c3a0b998..4c5f1266336b 100644
--- a/test/Transforms/SLPVectorizer/X86/reduction.ll
+++ b/test/Transforms/SLPVectorizer/X86/reduction.ll
@@ -23,12 +23,12 @@ for.body:                                         ; preds = %entry, %for.body
   %i.015 = phi i32 [ %inc, %for.body ], [ 0, %entry ]
   %sum.014 = phi double [ %add6, %for.body ], [ 0.000000e+00, %entry ]
   %mul = shl nsw i32 %i.015, 1
-  %arrayidx = getelementptr inbounds double* %A, i32 %mul
-  %0 = load double* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds double, double* %A, i32 %mul
+  %0 = load double, double* %arrayidx, align 4
   %mul1 = fmul double %0, 7.000000e+00
   %add12 = or i32 %mul, 1
-  %arrayidx3 = getelementptr inbounds double* %A, i32 %add12
-  %1 = load double* %arrayidx3, align 4
+  %arrayidx3 = getelementptr inbounds double, double* %A, i32 %add12
+  %1 = load double, double* %arrayidx3, align 4
   %mul4 = fmul double %1, 7.000000e+00
   %add5 = fadd double %mul1, %mul4
   %add6 = fadd double %sum.014, %add5
diff --git a/test/Transforms/SLPVectorizer/X86/reduction2.ll b/test/Transforms/SLPVectorizer/X86/reduction2.ll
index f21e86c5646c..507a61aa16f4 100644
--- a/test/Transforms/SLPVectorizer/X86/reduction2.ll
+++ b/test/Transforms/SLPVectorizer/X86/reduction2.ll
@@ -13,13 +13,13 @@ define double @foo(double* nocapture %D) {
   %i.02 = phi i32 [ 0, %0 ], [ %10, %1 ]
   %sum.01 = phi double [ 0.000000e+00, %0 ], [ %9, %1 ]
   %2 = shl nsw i32 %i.02, 1
-  %3 = getelementptr inbounds double* %D, i32 %2
-  %4 = load double* %3, align 4
+  %3 = getelementptr inbounds double, double* %D, i32 %2
+  %4 = load double, double* %3, align 4
   %A4 = fmul double %4, %4
   %A42 = fmul double %A4, %A4
   %5 = or i32 %2, 1
-  %6 = getelementptr inbounds double* %D, i32 %5
-  %7 = load double* %6, align 4
+  %6 = getelementptr inbounds double, double* %D, i32 %5
+  %7 = load double, double* %6, align 4
   %A7 = fmul double %7, %7
   %A72 = fmul double %A7, %A7
   %8 = fadd double %A42, %A72
diff --git a/test/Transforms/SLPVectorizer/X86/return.ll b/test/Transforms/SLPVectorizer/X86/return.ll
index 1a81c235bdfb..e2b263718e02 100644
--- a/test/Transforms/SLPVectorizer/X86/return.ll
+++ b/test/Transforms/SLPVectorizer/X86/return.ll
@@ -13,17 +13,17 @@ target triple = "x86_64--linux-gnu"
 ; }
 
 ; CHECK-LABEL: @return1
-; CHECK: %0 = load <2 x double>*
-; CHECK: %1 = load <2 x double>*
+; CHECK: %0 = load <2 x double>, <2 x double>*
+; CHECK: %1 = load <2 x double>, <2 x double>*
 ; CHECK: %2 = fadd <2 x double>
 
 define double @return1() {
 entry:
-  %a0 = load double* getelementptr inbounds ([4 x double]* @a, i32 0, i32 0), align 8
-  %b0 = load double* getelementptr inbounds ([4 x double]* @b, i32 0, i32 0), align 8
+  %a0 = load double, double* getelementptr inbounds ([4 x double], [4 x double]* @a, i32 0, i32 0), align 8
+  %b0 = load double, double* getelementptr inbounds ([4 x double], [4 x double]* @b, i32 0, i32 0), align 8
   %add0 = fadd double %a0, %b0
-  %a1 = load double* getelementptr inbounds ([4 x double]* @a, i32 0, i32 1), align 8
-  %b1 = load double* getelementptr inbounds ([4 x double]* @b, i32 0, i32 1), align 8
+  %a1 = load double, double* getelementptr inbounds ([4 x double], [4 x double]* @a, i32 0, i32 1), align 8
+  %b1 = load double, double* getelementptr inbounds ([4 x double], [4 x double]* @b, i32 0, i32 1), align 8
   %add1 = fadd double %a1, %b1
   %add2 = fadd double %add0, %add1
   ret double %add2
@@ -34,20 +34,20 @@ entry:
 ; }
 
 ; CHECK-LABEL: @return2
-; CHECK: %1 = load <2 x double>*
-; CHECK: %3 = load <2 x double>* %2
+; CHECK: %1 = load <2 x double>, <2 x double>*
+; CHECK: %3 = load <2 x double>, <2 x double>* %2
 ; CHECK: %4 = fadd <2 x double> %1, %3
 
 define double @return2(double* nocapture readonly %x) {
 entry:
-  %x0 = load double* %x, align 4
-  %arrayidx1 = getelementptr inbounds double* %x, i32 2
-  %x2 = load double* %arrayidx1, align 4
+  %x0 = load double, double* %x, align 4
+  %arrayidx1 = getelementptr inbounds double, double* %x, i32 2
+  %x2 = load double, double* %arrayidx1, align 4
   %add3 = fadd double %x0, %x2
-  %arrayidx2 = getelementptr inbounds double* %x, i32 1
-  %x1 = load double* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds double* %x, i32 3
-  %x3 = load double* %arrayidx3, align 4
+  %arrayidx2 = getelementptr inbounds double, double* %x, i32 1
+  %x1 = load double, double* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds double, double* %x, i32 3
+  %x3 = load double, double* %arrayidx3, align 4
   %add4 = fadd double %x1, %x3
   %add5 = fadd double %add3, %add4
   ret double %add5
diff --git a/test/Transforms/SLPVectorizer/X86/rgb_phi.ll b/test/Transforms/SLPVectorizer/X86/rgb_phi.ll
index 6aea5d3c6f6b..0bdb7dab1726 100644
--- a/test/Transforms/SLPVectorizer/X86/rgb_phi.ll
+++ b/test/Transforms/SLPVectorizer/X86/rgb_phi.ll
@@ -33,11 +33,11 @@ target triple = "i386-apple-macosx10.9.0"
 
 define float @foo(float* nocapture readonly %A) {
 entry:
-  %0 = load float* %A, align 4
-  %arrayidx1 = getelementptr inbounds float* %A, i64 1
-  %1 = load float* %arrayidx1, align 4
-  %arrayidx2 = getelementptr inbounds float* %A, i64 2
-  %2 = load float* %arrayidx2, align 4
+  %0 = load float, float* %A, align 4
+  %arrayidx1 = getelementptr inbounds float, float* %A, i64 1
+  %1 = load float, float* %arrayidx1, align 4
+  %arrayidx2 = getelementptr inbounds float, float* %A, i64 2
+  %2 = load float, float* %arrayidx2, align 4
   br label %for.body
 
 for.body:                                         ; preds = %for.body.for.body_crit_edge, %entry
@@ -49,13 +49,13 @@ for.body:                                         ; preds = %for.body.for.body_c
   %mul = fmul float %3, 7.000000e+00
   %add4 = fadd float %R.030, %mul
   %4 = add nsw i64 %indvars.iv, 1
-  %arrayidx7 = getelementptr inbounds float* %A, i64 %4
-  %5 = load float* %arrayidx7, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %A, i64 %4
+  %5 = load float, float* %arrayidx7, align 4
   %mul8 = fmul float %5, 8.000000e+00
   %add9 = fadd float %G.031, %mul8
   %6 = add nsw i64 %indvars.iv, 2
-  %arrayidx12 = getelementptr inbounds float* %A, i64 %6
-  %7 = load float* %arrayidx12, align 4
+  %arrayidx12 = getelementptr inbounds float, float* %A, i64 %6
+  %7 = load float, float* %arrayidx12, align 4
   %mul13 = fmul float %7, 9.000000e+00
   %add14 = fadd float %B.032, %mul13
   %indvars.iv.next = add i64 %indvars.iv, 3
@@ -64,8 +64,8 @@ for.body:                                         ; preds = %for.body.for.body_c
   br i1 %cmp, label %for.body.for.body_crit_edge, label %for.end
 
 for.body.for.body_crit_edge:                      ; preds = %for.body
-  %arrayidx3.phi.trans.insert = getelementptr inbounds float* %A, i64 %indvars.iv.next
-  %.pre = load float* %arrayidx3.phi.trans.insert, align 4
+  %arrayidx3.phi.trans.insert = getelementptr inbounds float, float* %A, i64 %indvars.iv.next
+  %.pre = load float, float* %arrayidx3.phi.trans.insert, align 4
   br label %for.body
 
 for.end:                                          ; preds = %for.body
diff --git a/test/Transforms/SLPVectorizer/X86/saxpy.ll b/test/Transforms/SLPVectorizer/X86/saxpy.ll
index 4b39d46f89db..a9ca093c0cdf 100644
--- a/test/Transforms/SLPVectorizer/X86/saxpy.ll
+++ b/test/Transforms/SLPVectorizer/X86/saxpy.ll
@@ -9,35 +9,35 @@ target triple = "x86_64-apple-macosx10.8.0"
 ;CHECK: ret
 
 define void @SAXPY(i32* noalias nocapture %x, i32* noalias nocapture %y, i32 %a, i64 %i) {
-  %1 = getelementptr inbounds i32* %x, i64 %i
-  %2 = load i32* %1, align 4
+  %1 = getelementptr inbounds i32, i32* %x, i64 %i
+  %2 = load i32, i32* %1, align 4
   %3 = mul nsw i32 %2, %a
-  %4 = getelementptr inbounds i32* %y, i64 %i
-  %5 = load i32* %4, align 4
+  %4 = getelementptr inbounds i32, i32* %y, i64 %i
+  %5 = load i32, i32* %4, align 4
   %6 = add nsw i32 %3, %5
   store i32 %6, i32* %1, align 4
   %7 = add i64 %i, 1
-  %8 = getelementptr inbounds i32* %x, i64 %7
-  %9 = load i32* %8, align 4
+  %8 = getelementptr inbounds i32, i32* %x, i64 %7
+  %9 = load i32, i32* %8, align 4
   %10 = mul nsw i32 %9, %a
-  %11 = getelementptr inbounds i32* %y, i64 %7
-  %12 = load i32* %11, align 4
+  %11 = getelementptr inbounds i32, i32* %y, i64 %7
+  %12 = load i32, i32* %11, align 4
   %13 = add nsw i32 %10, %12
   store i32 %13, i32* %8, align 4
   %14 = add i64 %i, 2
-  %15 = getelementptr inbounds i32* %x, i64 %14
-  %16 = load i32* %15, align 4
+  %15 = getelementptr inbounds i32, i32* %x, i64 %14
+  %16 = load i32, i32* %15, align 4
   %17 = mul nsw i32 %16, %a
-  %18 = getelementptr inbounds i32* %y, i64 %14
-  %19 = load i32* %18, align 4
+  %18 = getelementptr inbounds i32, i32* %y, i64 %14
+  %19 = load i32, i32* %18, align 4
   %20 = add nsw i32 %17, %19
   store i32 %20, i32* %15, align 4
   %21 = add i64 %i, 3
-  %22 = getelementptr inbounds i32* %x, i64 %21
-  %23 = load i32* %22, align 4
+  %22 = getelementptr inbounds i32, i32* %x, i64 %21
+  %23 = load i32, i32* %22, align 4
   %24 = mul nsw i32 %23, %a
-  %25 = getelementptr inbounds i32* %y, i64 %21
-  %26 = load i32* %25, align 4
+  %25 = getelementptr inbounds i32, i32* %y, i64 %21
+  %26 = load i32, i32* %25, align 4
   %27 = add nsw i32 %24, %26
   store i32 %27, i32* %22, align 4
   ret void
@@ -46,15 +46,15 @@ define void @SAXPY(i32* noalias nocapture %x, i32* noalias nocapture %y, i32 %a,
 ; Make sure we don't crash on this one.
 define void @SAXPY_crash(i32* noalias nocapture %x, i32* noalias nocapture %y, i64 %i) {
   %1 = add i64 %i, 1
-  %2 = getelementptr inbounds i32* %x, i64 %1
-  %3 = getelementptr inbounds i32* %y, i64 %1
-  %4 = load i32* %3, align 4
+  %2 = getelementptr inbounds i32, i32* %x, i64 %1
+  %3 = getelementptr inbounds i32, i32* %y, i64 %1
+  %4 = load i32, i32* %3, align 4
   %5 = add nsw i32 undef, %4
   store i32 %5, i32* %2, align 4
   %6 = add i64 %i, 2
-  %7 = getelementptr inbounds i32* %x, i64 %6
-  %8 = getelementptr inbounds i32* %y, i64 %6
-  %9 = load i32* %8, align 4
+  %7 = getelementptr inbounds i32, i32* %x, i64 %6
+  %8 = getelementptr inbounds i32, i32* %y, i64 %6
+  %9 = load i32, i32* %8, align 4
   %10 = add nsw i32 undef, %9
   store i32 %10, i32* %7, align 4
   ret void
diff --git a/test/Transforms/SLPVectorizer/X86/scheduling.ll b/test/Transforms/SLPVectorizer/X86/scheduling.ll
index 3b3bd804e057..33bdc6a6b04a 100644
--- a/test/Transforms/SLPVectorizer/X86/scheduling.ll
+++ b/test/Transforms/SLPVectorizer/X86/scheduling.ll
@@ -24,43 +24,43 @@ for.body:                                         ; preds = %for.body, %entry
   %indvars.iv = phi i64 [ 0, %entry ], [ %indvars.iv.next, %for.body ]
   %a.088 = phi i32 [ 0, %entry ], [ %add52, %for.body ]
   %1 = shl i64 %indvars.iv, 3
-  %arrayidx = getelementptr inbounds i32* %diff, i64 %1
-  %2 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %diff, i64 %1
+  %2 = load i32, i32* %arrayidx, align 4
   %3 = or i64 %1, 4
-  %arrayidx2 = getelementptr inbounds i32* %diff, i64 %3
-  %4 = load i32* %arrayidx2, align 4
+  %arrayidx2 = getelementptr inbounds i32, i32* %diff, i64 %3
+  %4 = load i32, i32* %arrayidx2, align 4
   %add3 = add nsw i32 %4, %2
-  %arrayidx6 = getelementptr inbounds [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 0
+  %arrayidx6 = getelementptr inbounds [8 x [8 x i32]], [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 0
   store i32 %add3, i32* %arrayidx6, align 16
   %add10 = add nsw i32 %add3, %a.088
   %5 = or i64 %1, 1
-  %arrayidx13 = getelementptr inbounds i32* %diff, i64 %5
-  %6 = load i32* %arrayidx13, align 4
+  %arrayidx13 = getelementptr inbounds i32, i32* %diff, i64 %5
+  %6 = load i32, i32* %arrayidx13, align 4
   %7 = or i64 %1, 5
-  %arrayidx16 = getelementptr inbounds i32* %diff, i64 %7
-  %8 = load i32* %arrayidx16, align 4
+  %arrayidx16 = getelementptr inbounds i32, i32* %diff, i64 %7
+  %8 = load i32, i32* %arrayidx16, align 4
   %add17 = add nsw i32 %8, %6
-  %arrayidx20 = getelementptr inbounds [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 1
+  %arrayidx20 = getelementptr inbounds [8 x [8 x i32]], [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 1
   store i32 %add17, i32* %arrayidx20, align 4
   %add24 = add nsw i32 %add10, %add17
   %9 = or i64 %1, 2
-  %arrayidx27 = getelementptr inbounds i32* %diff, i64 %9
-  %10 = load i32* %arrayidx27, align 4
+  %arrayidx27 = getelementptr inbounds i32, i32* %diff, i64 %9
+  %10 = load i32, i32* %arrayidx27, align 4
   %11 = or i64 %1, 6
-  %arrayidx30 = getelementptr inbounds i32* %diff, i64 %11
-  %12 = load i32* %arrayidx30, align 4
+  %arrayidx30 = getelementptr inbounds i32, i32* %diff, i64 %11
+  %12 = load i32, i32* %arrayidx30, align 4
   %add31 = add nsw i32 %12, %10
-  %arrayidx34 = getelementptr inbounds [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 2
+  %arrayidx34 = getelementptr inbounds [8 x [8 x i32]], [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 2
   store i32 %add31, i32* %arrayidx34, align 8
   %add38 = add nsw i32 %add24, %add31
   %13 = or i64 %1, 3
-  %arrayidx41 = getelementptr inbounds i32* %diff, i64 %13
-  %14 = load i32* %arrayidx41, align 4
+  %arrayidx41 = getelementptr inbounds i32, i32* %diff, i64 %13
+  %14 = load i32, i32* %arrayidx41, align 4
   %15 = or i64 %1, 7
-  %arrayidx44 = getelementptr inbounds i32* %diff, i64 %15
-  %16 = load i32* %arrayidx44, align 4
+  %arrayidx44 = getelementptr inbounds i32, i32* %diff, i64 %15
+  %16 = load i32, i32* %arrayidx44, align 4
   %add45 = add nsw i32 %16, %14
-  %arrayidx48 = getelementptr inbounds [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 3
+  %arrayidx48 = getelementptr inbounds [8 x [8 x i32]], [8 x [8 x i32]]* %m2, i64 0, i64 %indvars.iv, i64 3
   store i32 %add45, i32* %arrayidx48, align 4
   %add52 = add nsw i32 %add38, %add45
   %indvars.iv.next = add nuw nsw i64 %indvars.iv, 1
@@ -68,7 +68,7 @@ for.body:                                         ; preds = %for.body, %entry
   br i1 %exitcond, label %for.end, label %for.body
 
 for.end:                                          ; preds = %for.body
-  %arraydecay = getelementptr inbounds [8 x [8 x i32]]* %m2, i64 0, i64 0
+  %arraydecay = getelementptr inbounds [8 x [8 x i32]], [8 x [8 x i32]]* %m2, i64 0, i64 0
   call void @ff([8 x i32]* %arraydecay) #1
   ret i32 %add52
 }
diff --git a/test/Transforms/SLPVectorizer/X86/simple-loop.ll b/test/Transforms/SLPVectorizer/X86/simple-loop.ll
index 0111b946d496..c9bb88497ac1 100644
--- a/test/Transforms/SLPVectorizer/X86/simple-loop.ll
+++ b/test/Transforms/SLPVectorizer/X86/simple-loop.ll
@@ -11,18 +11,18 @@ define i32 @rollable(i32* noalias nocapture %in, i32* noalias nocapture %out, i6
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %i.019 = phi i64 [ %26, %.lr.ph ], [ 0, %0 ]
   %2 = shl i64 %i.019, 2
-  %3 = getelementptr inbounds i32* %in, i64 %2
+  %3 = getelementptr inbounds i32, i32* %in, i64 %2
 ;CHECK:load <4 x i32>
-  %4 = load i32* %3, align 4
+  %4 = load i32, i32* %3, align 4
   %5 = or i64 %2, 1
-  %6 = getelementptr inbounds i32* %in, i64 %5
-  %7 = load i32* %6, align 4
+  %6 = getelementptr inbounds i32, i32* %in, i64 %5
+  %7 = load i32, i32* %6, align 4
   %8 = or i64 %2, 2
-  %9 = getelementptr inbounds i32* %in, i64 %8
-  %10 = load i32* %9, align 4
+  %9 = getelementptr inbounds i32, i32* %in, i64 %8
+  %10 = load i32, i32* %9, align 4
   %11 = or i64 %2, 3
-  %12 = getelementptr inbounds i32* %in, i64 %11
-  %13 = load i32* %12, align 4
+  %12 = getelementptr inbounds i32, i32* %in, i64 %11
+  %13 = load i32, i32* %12, align 4
 ;CHECK:mul <4 x i32>
   %14 = mul i32 %4, 7
 ;CHECK:add <4 x i32>
@@ -33,14 +33,14 @@ define i32 @rollable(i32* noalias nocapture %in, i32* noalias nocapture %out, i6
   %19 = add i32 %18, 21
   %20 = mul i32 %13, 7
   %21 = add i32 %20, 28
-  %22 = getelementptr inbounds i32* %out, i64 %2
+  %22 = getelementptr inbounds i32, i32* %out, i64 %2
 ;CHECK:store <4 x i32>
   store i32 %15, i32* %22, align 4
-  %23 = getelementptr inbounds i32* %out, i64 %5
+  %23 = getelementptr inbounds i32, i32* %out, i64 %5
   store i32 %17, i32* %23, align 4
-  %24 = getelementptr inbounds i32* %out, i64 %8
+  %24 = getelementptr inbounds i32, i32* %out, i64 %8
   store i32 %19, i32* %24, align 4
-  %25 = getelementptr inbounds i32* %out, i64 %11
+  %25 = getelementptr inbounds i32, i32* %out, i64 %11
   store i32 %21, i32* %25, align 4
   %26 = add i64 %i.019, 1
   %exitcond = icmp eq i64 %26, %n
@@ -61,17 +61,17 @@ define i32 @unrollable(i32* %in, i32* %out, i64 %n) nounwind ssp uwtable {
 .lr.ph:                                           ; preds = %0, %.lr.ph
   %i.019 = phi i64 [ %26, %.lr.ph ], [ 0, %0 ]
   %2 = shl i64 %i.019, 2
-  %3 = getelementptr inbounds i32* %in, i64 %2
-  %4 = load i32* %3, align 4
+  %3 = getelementptr inbounds i32, i32* %in, i64 %2
+  %4 = load i32, i32* %3, align 4
   %5 = or i64 %2, 1
-  %6 = getelementptr inbounds i32* %in, i64 %5
-  %7 = load i32* %6, align 4
+  %6 = getelementptr inbounds i32, i32* %in, i64 %5
+  %7 = load i32, i32* %6, align 4
   %8 = or i64 %2, 2
-  %9 = getelementptr inbounds i32* %in, i64 %8
-  %10 = load i32* %9, align 4
+  %9 = getelementptr inbounds i32, i32* %in, i64 %8
+  %10 = load i32, i32* %9, align 4
   %11 = or i64 %2, 3
-  %12 = getelementptr inbounds i32* %in, i64 %11
-  %13 = load i32* %12, align 4
+  %12 = getelementptr inbounds i32, i32* %in, i64 %11
+  %13 = load i32, i32* %12, align 4
   %14 = mul i32 %4, 7
   %15 = add i32 %14, 7
   %16 = mul i32 %7, 7
@@ -80,14 +80,14 @@ define i32 @unrollable(i32* %in, i32* %out, i64 %n) nounwind ssp uwtable {
   %19 = add i32 %18, 21
   %20 = mul i32 %13, 7
   %21 = add i32 %20, 28
-  %22 = getelementptr inbounds i32* %out, i64 %2
+  %22 = getelementptr inbounds i32, i32* %out, i64 %2
   store i32 %15, i32* %22, align 4
-  %23 = getelementptr inbounds i32* %out, i64 %5
+  %23 = getelementptr inbounds i32, i32* %out, i64 %5
   store i32 %17, i32* %23, align 4
   %barrier = call i32 @goo(i32 0)                      ; <---------------- memory barrier.
-  %24 = getelementptr inbounds i32* %out, i64 %8
+  %24 = getelementptr inbounds i32, i32* %out, i64 %8
   store i32 %19, i32* %24, align 4
-  %25 = getelementptr inbounds i32* %out, i64 %11
+  %25 = getelementptr inbounds i32, i32* %out, i64 %11
   store i32 %21, i32* %25, align 4
   %26 = add i64 %i.019, 1
   %exitcond = icmp eq i64 %26, %n
diff --git a/test/Transforms/SLPVectorizer/X86/simplebb.ll b/test/Transforms/SLPVectorizer/X86/simplebb.ll
index 7d682e5e4676..a5d9ad9685c0 100644
--- a/test/Transforms/SLPVectorizer/X86/simplebb.ll
+++ b/test/Transforms/SLPVectorizer/X86/simplebb.ll
@@ -9,16 +9,16 @@ target triple = "x86_64-apple-macosx10.8.0"
 ; CHECK: ret
 define void @test1(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
@@ -29,17 +29,17 @@ entry:
 ; CHECK: ret
 define void @test2(double* %a, double* %b, i8* %e) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   %c = bitcast i8* %e to double*
   store double %mul, double* %c, align 8
-  %carrayidx5 = getelementptr inbounds i8* %e, i64 8
+  %carrayidx5 = getelementptr inbounds i8, i8* %e, i64 8
   %arrayidx5 = bitcast i8* %carrayidx5 to double*
   store double %mul5, double* %arrayidx5, align 8
   ret void
@@ -52,16 +52,16 @@ entry:
 ; CHECK: ret
 define void @test_volatile_load(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load volatile double* %a, align 8
-  %i1 = load volatile double* %b, align 8
+  %i0 = load volatile double, double* %a, align 8
+  %i1 = load volatile double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
@@ -72,16 +72,16 @@ entry:
 ; CHECK: ret
 define void @test_volatile_store(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store volatile double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store volatile double %mul5, double* %arrayidx5, align 8
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/tiny-tree.ll b/test/Transforms/SLPVectorizer/X86/tiny-tree.ll
index 10c3130b424c..6c93222ef93a 100644
--- a/test/Transforms/SLPVectorizer/X86/tiny-tree.ll
+++ b/test/Transforms/SLPVectorizer/X86/tiny-tree.ll
@@ -17,14 +17,14 @@ for.body:                                         ; preds = %entry, %for.body
   %i.015 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
   %dst.addr.014 = phi double* [ %add.ptr4, %for.body ], [ %dst, %entry ]
   %src.addr.013 = phi double* [ %add.ptr, %for.body ], [ %src, %entry ]
-  %0 = load double* %src.addr.013, align 8
+  %0 = load double, double* %src.addr.013, align 8
   store double %0, double* %dst.addr.014, align 8
-  %arrayidx2 = getelementptr inbounds double* %src.addr.013, i64 1
-  %1 = load double* %arrayidx2, align 8
-  %arrayidx3 = getelementptr inbounds double* %dst.addr.014, i64 1
+  %arrayidx2 = getelementptr inbounds double, double* %src.addr.013, i64 1
+  %1 = load double, double* %arrayidx2, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %dst.addr.014, i64 1
   store double %1, double* %arrayidx3, align 8
-  %add.ptr = getelementptr inbounds double* %src.addr.013, i64 %i.015
-  %add.ptr4 = getelementptr inbounds double* %dst.addr.014, i64 %i.015
+  %add.ptr = getelementptr inbounds double, double* %src.addr.013, i64 %i.015
+  %add.ptr4 = getelementptr inbounds double, double* %dst.addr.014, i64 %i.015
   %inc = add i64 %i.015, 1
   %exitcond = icmp eq i64 %inc, %count
   br i1 %exitcond, label %for.end, label %for.body
@@ -47,22 +47,22 @@ for.body:                                         ; preds = %entry, %for.body
   %i.023 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
   %dst.addr.022 = phi float* [ %add.ptr8, %for.body ], [ %dst, %entry ]
   %src.addr.021 = phi float* [ %add.ptr, %for.body ], [ %src, %entry ]
-  %0 = load float* %src.addr.021, align 4
+  %0 = load float, float* %src.addr.021, align 4
   store float %0, float* %dst.addr.022, align 4
-  %arrayidx2 = getelementptr inbounds float* %src.addr.021, i64 1
-  %1 = load float* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds float* %dst.addr.022, i64 1
+  %arrayidx2 = getelementptr inbounds float, float* %src.addr.021, i64 1
+  %1 = load float, float* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %dst.addr.022, i64 1
   store float %1, float* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds float* %src.addr.021, i64 2
-  %2 = load float* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds float* %dst.addr.022, i64 2
+  %arrayidx4 = getelementptr inbounds float, float* %src.addr.021, i64 2
+  %2 = load float, float* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds float, float* %dst.addr.022, i64 2
   store float %2, float* %arrayidx5, align 4
-  %arrayidx6 = getelementptr inbounds float* %src.addr.021, i64 3
-  %3 = load float* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds float* %dst.addr.022, i64 3
+  %arrayidx6 = getelementptr inbounds float, float* %src.addr.021, i64 3
+  %3 = load float, float* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %dst.addr.022, i64 3
   store float %3, float* %arrayidx7, align 4
-  %add.ptr = getelementptr inbounds float* %src.addr.021, i64 %i.023
-  %add.ptr8 = getelementptr inbounds float* %dst.addr.022, i64 %i.023
+  %add.ptr = getelementptr inbounds float, float* %src.addr.021, i64 %i.023
+  %add.ptr8 = getelementptr inbounds float, float* %dst.addr.022, i64 %i.023
   %inc = add i64 %i.023, 1
   %exitcond = icmp eq i64 %inc, %count
   br i1 %exitcond, label %for.end, label %for.body
@@ -85,14 +85,14 @@ for.body:                                         ; preds = %entry, %for.body
   %i.015 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
   %dst.addr.014 = phi double* [ %add.ptr4, %for.body ], [ %dst, %entry ]
   %src.addr.013 = phi double* [ %add.ptr, %for.body ], [ %src, %entry ]
-  %0 = load double* %src.addr.013, align 8
+  %0 = load double, double* %src.addr.013, align 8
   store double %0, double* %dst.addr.014, align 8
-  %arrayidx2 = getelementptr inbounds double* %src.addr.013, i64 2
-  %1 = load double* %arrayidx2, align 8
-  %arrayidx3 = getelementptr inbounds double* %dst.addr.014, i64 1 
+  %arrayidx2 = getelementptr inbounds double, double* %src.addr.013, i64 2
+  %1 = load double, double* %arrayidx2, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %dst.addr.014, i64 1 
   store double %1, double* %arrayidx3, align 8
-  %add.ptr = getelementptr inbounds double* %src.addr.013, i64 %i.015
-  %add.ptr4 = getelementptr inbounds double* %dst.addr.014, i64 %i.015
+  %add.ptr = getelementptr inbounds double, double* %src.addr.013, i64 %i.015
+  %add.ptr4 = getelementptr inbounds double, double* %dst.addr.014, i64 %i.015
   %inc = add i64 %i.015, 1
   %exitcond = icmp eq i64 %inc, %count
   br i1 %exitcond, label %for.end, label %for.body
@@ -115,22 +115,22 @@ for.body:                                         ; preds = %entry, %for.body
   %i.023 = phi i64 [ %inc, %for.body ], [ 0, %entry ]
   %dst.addr.022 = phi float* [ %add.ptr8, %for.body ], [ %dst, %entry ]
   %src.addr.021 = phi float* [ %add.ptr, %for.body ], [ %src, %entry ]
-  %0 = load float* %src.addr.021, align 4
+  %0 = load float, float* %src.addr.021, align 4
   store float %0, float* %dst.addr.022, align 4
-  %arrayidx2 = getelementptr inbounds float* %src.addr.021, i64 4 
-  %1 = load float* %arrayidx2, align 4
-  %arrayidx3 = getelementptr inbounds float* %dst.addr.022, i64 1
+  %arrayidx2 = getelementptr inbounds float, float* %src.addr.021, i64 4 
+  %1 = load float, float* %arrayidx2, align 4
+  %arrayidx3 = getelementptr inbounds float, float* %dst.addr.022, i64 1
   store float %1, float* %arrayidx3, align 4
-  %arrayidx4 = getelementptr inbounds float* %src.addr.021, i64 2
-  %2 = load float* %arrayidx4, align 4
-  %arrayidx5 = getelementptr inbounds float* %dst.addr.022, i64 2
+  %arrayidx4 = getelementptr inbounds float, float* %src.addr.021, i64 2
+  %2 = load float, float* %arrayidx4, align 4
+  %arrayidx5 = getelementptr inbounds float, float* %dst.addr.022, i64 2
   store float %2, float* %arrayidx5, align 4
-  %arrayidx6 = getelementptr inbounds float* %src.addr.021, i64 3
-  %3 = load float* %arrayidx6, align 4
-  %arrayidx7 = getelementptr inbounds float* %dst.addr.022, i64 3
+  %arrayidx6 = getelementptr inbounds float, float* %src.addr.021, i64 3
+  %3 = load float, float* %arrayidx6, align 4
+  %arrayidx7 = getelementptr inbounds float, float* %dst.addr.022, i64 3
   store float %3, float* %arrayidx7, align 4
-  %add.ptr = getelementptr inbounds float* %src.addr.021, i64 %i.023
-  %add.ptr8 = getelementptr inbounds float* %dst.addr.022, i64 %i.023
+  %add.ptr = getelementptr inbounds float, float* %src.addr.021, i64 %i.023
+  %add.ptr8 = getelementptr inbounds float, float* %dst.addr.022, i64 %i.023
   %inc = add i64 %i.023, 1
   %exitcond = icmp eq i64 %inc, %count
   br i1 %exitcond, label %for.end, label %for.body
@@ -143,13 +143,13 @@ for.end:                                          ; preds = %for.body, %entry
 ; CHECK-LABEL: store_splat
 ; CHECK: store <4 x float>
 define void @store_splat(float*, float) {
-  %3 = getelementptr inbounds float* %0, i64 0
+  %3 = getelementptr inbounds float, float* %0, i64 0
   store float %1, float* %3, align 4
-  %4 = getelementptr inbounds float* %0, i64 1
+  %4 = getelementptr inbounds float, float* %0, i64 1
   store float %1, float* %4, align 4
-  %5 = getelementptr inbounds float* %0, i64 2
+  %5 = getelementptr inbounds float, float* %0, i64 2
   store float %1, float* %5, align 4
-  %6 = getelementptr inbounds float* %0, i64 3
+  %6 = getelementptr inbounds float, float* %0, i64 3
   store float %1, float* %6, align 4
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/X86/unreachable.ll b/test/Transforms/SLPVectorizer/X86/unreachable.ll
index 8d609571be17..f29f69d7e82f 100644
--- a/test/Transforms/SLPVectorizer/X86/unreachable.ll
+++ b/test/Transforms/SLPVectorizer/X86/unreachable.ll
@@ -11,16 +11,16 @@ entry:
   br label %bb2
 
 bb1:                                    ; an unreachable block
-  %t3 = getelementptr inbounds i32* %x, i64 4
-  %t4 = load i32* %t3, align 4
-  %t5 = getelementptr inbounds i32* %x, i64 5
-  %t6 = load i32* %t5, align 4
+  %t3 = getelementptr inbounds i32, i32* %x, i64 4
+  %t4 = load i32, i32* %t3, align 4
+  %t5 = getelementptr inbounds i32, i32* %x, i64 5
+  %t6 = load i32, i32* %t5, align 4
   %bad = fadd float %bad, 0.000000e+00  ; <- an instruction with self dependency,
                                         ;    but legal in unreachable code
-  %t7 = getelementptr inbounds i32* %x, i64 6
-  %t8 = load i32* %t7, align 4
-  %t9 = getelementptr inbounds i32* %x, i64 7
-  %t10 = load i32* %t9, align 4
+  %t7 = getelementptr inbounds i32, i32* %x, i64 6
+  %t8 = load i32, i32* %t7, align 4
+  %t9 = getelementptr inbounds i32, i32* %x, i64 7
+  %t10 = load i32, i32* %t9, align 4
   br label %bb2
 
 bb2:
@@ -29,11 +29,11 @@ bb2:
   %t3.0 = phi i32 [ %t8, %bb1 ], [ 2, %entry ]
   %t4.0 = phi i32 [ %t10, %bb1 ], [ 2, %entry ]
   store i32 %t1.0, i32* %x, align 4
-  %t12 = getelementptr inbounds i32* %x, i64 1
+  %t12 = getelementptr inbounds i32, i32* %x, i64 1
   store i32 %t2.0, i32* %t12, align 4
-  %t13 = getelementptr inbounds i32* %x, i64 2
+  %t13 = getelementptr inbounds i32, i32* %x, i64 2
   store i32 %t3.0, i32* %t13, align 4
-  %t14 = getelementptr inbounds i32* %x, i64 3
+  %t14 = getelementptr inbounds i32, i32* %x, i64 3
   store i32 %t4.0, i32* %t14, align 4
   ret void
 }
diff --git a/test/Transforms/SLPVectorizer/XCore/no-vector-registers.ll b/test/Transforms/SLPVectorizer/XCore/no-vector-registers.ll
index 66392e74cb31..efd5386f520e 100644
--- a/test/Transforms/SLPVectorizer/XCore/no-vector-registers.ll
+++ b/test/Transforms/SLPVectorizer/XCore/no-vector-registers.ll
@@ -8,16 +8,16 @@ target triple = "xcore"
 ; CHECK-NOT: <2 x double>
 define void @test1(double* %a, double* %b, double* %c) {
 entry:
-  %i0 = load double* %a, align 8
-  %i1 = load double* %b, align 8
+  %i0 = load double, double* %a, align 8
+  %i1 = load double, double* %b, align 8
   %mul = fmul double %i0, %i1
-  %arrayidx3 = getelementptr inbounds double* %a, i64 1
-  %i3 = load double* %arrayidx3, align 8
-  %arrayidx4 = getelementptr inbounds double* %b, i64 1
-  %i4 = load double* %arrayidx4, align 8
+  %arrayidx3 = getelementptr inbounds double, double* %a, i64 1
+  %i3 = load double, double* %arrayidx3, align 8
+  %arrayidx4 = getelementptr inbounds double, double* %b, i64 1
+  %i4 = load double, double* %arrayidx4, align 8
   %mul5 = fmul double %i3, %i4
   store double %mul, double* %c, align 8
-  %arrayidx5 = getelementptr inbounds double* %c, i64 1
+  %arrayidx5 = getelementptr inbounds double, double* %c, i64 1
   store double %mul5, double* %arrayidx5, align 8
   ret void
 }
diff --git a/test/Transforms/SROA/address-spaces.ll b/test/Transforms/SROA/address-spaces.ll
index 847f2851bbbf..004695dceddc 100644
--- a/test/Transforms/SROA/address-spaces.ll
+++ b/test/Transforms/SROA/address-spaces.ll
@@ -10,7 +10,7 @@ declare void @llvm.memcpy.p1i8.p1i8.i32(i8 addrspace(1)* nocapture, i8 addrspace
 ; Make sure an illegal bitcast isn't introduced
 define void @test_address_space_1_1(<2 x i64> addrspace(1)* %a, i16 addrspace(1)* %b) {
 ; CHECK-LABEL: @test_address_space_1_1(
-; CHECK: load <2 x i64> addrspace(1)* %a, align 2
+; CHECK: load <2 x i64>, <2 x i64> addrspace(1)* %a, align 2
 ; CHECK: store <2 x i64> {{.*}}, <2 x i64> addrspace(1)* {{.*}}, align 2
 ; CHECK: ret void
   %aa = alloca <2 x i64>, align 16
@@ -24,7 +24,7 @@ define void @test_address_space_1_1(<2 x i64> addrspace(1)* %a, i16 addrspace(1)
 
 define void @test_address_space_1_0(<2 x i64> addrspace(1)* %a, i16* %b) {
 ; CHECK-LABEL: @test_address_space_1_0(
-; CHECK: load <2 x i64> addrspace(1)* %a, align 2
+; CHECK: load <2 x i64>, <2 x i64> addrspace(1)* %a, align 2
 ; CHECK: store <2 x i64> {{.*}}, <2 x i64>* {{.*}}, align 2
 ; CHECK: ret void
   %aa = alloca <2 x i64>, align 16
@@ -38,7 +38,7 @@ define void @test_address_space_1_0(<2 x i64> addrspace(1)* %a, i16* %b) {
 
 define void @test_address_space_0_1(<2 x i64>* %a, i16 addrspace(1)* %b) {
 ; CHECK-LABEL: @test_address_space_0_1(
-; CHECK: load <2 x i64>* %a, align 2
+; CHECK: load <2 x i64>, <2 x i64>* %a, align 2
 ; CHECK: store <2 x i64> {{.*}}, <2 x i64> addrspace(1)* {{.*}}, align 2
 ; CHECK: ret void
   %aa = alloca <2 x i64>, align 16
@@ -60,7 +60,7 @@ for.end:
   %in = alloca %struct.struct_test_27.0.13, align 8
   %0 = bitcast %struct.struct_test_27.0.13* %in to [5 x i64]*
   store [5 x i64] %in.coerce, [5 x i64]* %0, align 8
-  %scevgep9 = getelementptr %struct.struct_test_27.0.13* %in, i32 0, i32 4, i32 0
+  %scevgep9 = getelementptr %struct.struct_test_27.0.13, %struct.struct_test_27.0.13* %in, i32 0, i32 4, i32 0
   %scevgep910 = bitcast i32* %scevgep9 to i8*
   call void @llvm.memcpy.p1i8.p0i8.i32(i8 addrspace(1)* undef, i8* %scevgep910, i32 16, i32 4, i1 false)
   ret void
diff --git a/test/Transforms/SROA/alignment.ll b/test/Transforms/SROA/alignment.ll
index 4f4e40cf76a7..455d14246e57 100644
--- a/test/Transforms/SROA/alignment.ll
+++ b/test/Transforms/SROA/alignment.ll
@@ -5,21 +5,21 @@ declare void @llvm.memcpy.p0i8.p0i8.i32(i8*, i8*, i32, i32, i1)
 
 define void @test1({ i8, i8 }* %a, { i8, i8 }* %b) {
 ; CHECK-LABEL: @test1(
-; CHECK: %[[gep_a0:.*]] = getelementptr inbounds { i8, i8 }* %a, i64 0, i32 0
-; CHECK: %[[a0:.*]] = load i8* %[[gep_a0]], align 16
-; CHECK: %[[gep_a1:.*]] = getelementptr inbounds { i8, i8 }* %a, i64 0, i32 1
-; CHECK: %[[a1:.*]] = load i8* %[[gep_a1]], align 1
-; CHECK: %[[gep_b0:.*]] = getelementptr inbounds { i8, i8 }* %b, i64 0, i32 0
+; CHECK: %[[gep_a0:.*]] = getelementptr inbounds { i8, i8 }, { i8, i8 }* %a, i64 0, i32 0
+; CHECK: %[[a0:.*]] = load i8, i8* %[[gep_a0]], align 16
+; CHECK: %[[gep_a1:.*]] = getelementptr inbounds { i8, i8 }, { i8, i8 }* %a, i64 0, i32 1
+; CHECK: %[[a1:.*]] = load i8, i8* %[[gep_a1]], align 1
+; CHECK: %[[gep_b0:.*]] = getelementptr inbounds { i8, i8 }, { i8, i8 }* %b, i64 0, i32 0
 ; CHECK: store i8 %[[a0]], i8* %[[gep_b0]], align 16
-; CHECK: %[[gep_b1:.*]] = getelementptr inbounds { i8, i8 }* %b, i64 0, i32 1
+; CHECK: %[[gep_b1:.*]] = getelementptr inbounds { i8, i8 }, { i8, i8 }* %b, i64 0, i32 1
 ; CHECK: store i8 %[[a1]], i8* %[[gep_b1]], align 1
 ; CHECK: ret void
 
 entry:
   %alloca = alloca { i8, i8 }, align 16
-  %gep_a = getelementptr { i8, i8 }* %a, i32 0, i32 0
-  %gep_alloca = getelementptr { i8, i8 }* %alloca, i32 0, i32 0
-  %gep_b = getelementptr { i8, i8 }* %b, i32 0, i32 0
+  %gep_a = getelementptr { i8, i8 }, { i8, i8 }* %a, i32 0, i32 0
+  %gep_alloca = getelementptr { i8, i8 }, { i8, i8 }* %alloca, i32 0, i32 0
+  %gep_b = getelementptr { i8, i8 }, { i8, i8 }* %b, i32 0, i32 0
 
   store i8 420, i8* %gep_alloca, align 16
 
@@ -31,17 +31,17 @@ entry:
 define void @test2() {
 ; CHECK-LABEL: @test2(
 ; CHECK: alloca i16
-; CHECK: load i8* %{{.*}}
+; CHECK: load i8, i8* %{{.*}}
 ; CHECK: store i8 42, i8* %{{.*}}
 ; CHECK: ret void
 
 entry:
   %a = alloca { i8, i8, i8, i8 }, align 2
-  %gep1 = getelementptr { i8, i8, i8, i8 }* %a, i32 0, i32 1
+  %gep1 = getelementptr { i8, i8, i8, i8 }, { i8, i8, i8, i8 }* %a, i32 0, i32 1
   %cast1 = bitcast i8* %gep1 to i16*
   store volatile i16 0, i16* %cast1
-  %gep2 = getelementptr { i8, i8, i8, i8 }* %a, i32 0, i32 2
-  %result = load i8* %gep2
+  %gep2 = getelementptr { i8, i8, i8, i8 }, { i8, i8, i8, i8 }* %a, i32 0, i32 2
+  %result = load i8, i8* %gep2
   store i8 42, i8* %gep2
   ret void
 }
@@ -49,7 +49,7 @@ entry:
 define void @PR13920(<2 x i64>* %a, i16* %b) {
 ; Test that alignments on memcpy intrinsics get propagated to loads and stores.
 ; CHECK-LABEL: @PR13920(
-; CHECK: load <2 x i64>* %a, align 2
+; CHECK: load <2 x i64>, <2 x i64>* %a, align 2
 ; CHECK: store <2 x i64> {{.*}}, <2 x i64>* {{.*}}, align 2
 ; CHECK: ret void
 
@@ -79,7 +79,7 @@ entry:
   %a_raw = bitcast { i8*, i8*, i8* }* %a to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a_raw, i8* %x, i32 22, i32 8, i1 false)
   %b_raw = bitcast { i8*, i8*, i8* }* %b to i8*
-  %b_gep = getelementptr i8* %b_raw, i32 6
+  %b_gep = getelementptr i8, i8* %b_raw, i32 6
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b_gep, i8* %x, i32 18, i32 2, i1 false)
   ret void
 }
@@ -93,28 +93,28 @@ define void @test5() {
 ; CHECK: alloca [9 x i8]
 ; CHECK: alloca [9 x i8]
 ; CHECK: store volatile double 0.0{{.*}}, double* %{{.*}}, align 1
-; CHECK: load volatile i16* %{{.*}}, align 1
-; CHECK: load double* %{{.*}}, align 1
+; CHECK: load volatile i16, i16* %{{.*}}, align 1
+; CHECK: load double, double* %{{.*}}, align 1
 ; CHECK: store volatile double %{{.*}}, double* %{{.*}}, align 1
-; CHECK: load volatile i16* %{{.*}}, align 1
+; CHECK: load volatile i16, i16* %{{.*}}, align 1
 ; CHECK: ret void
 
 entry:
   %a = alloca [18 x i8]
-  %raw1 = getelementptr inbounds [18 x i8]* %a, i32 0, i32 0
+  %raw1 = getelementptr inbounds [18 x i8], [18 x i8]* %a, i32 0, i32 0
   %ptr1 = bitcast i8* %raw1 to double*
   store volatile double 0.0, double* %ptr1, align 1
-  %weird_gep1 = getelementptr inbounds [18 x i8]* %a, i32 0, i32 7
+  %weird_gep1 = getelementptr inbounds [18 x i8], [18 x i8]* %a, i32 0, i32 7
   %weird_cast1 = bitcast i8* %weird_gep1 to i16*
-  %weird_load1 = load volatile i16* %weird_cast1, align 1
+  %weird_load1 = load volatile i16, i16* %weird_cast1, align 1
 
-  %raw2 = getelementptr inbounds [18 x i8]* %a, i32 0, i32 9
+  %raw2 = getelementptr inbounds [18 x i8], [18 x i8]* %a, i32 0, i32 9
   %ptr2 = bitcast i8* %raw2 to double*
-  %d1 = load double* %ptr1, align 1
+  %d1 = load double, double* %ptr1, align 1
   store volatile double %d1, double* %ptr2, align 1
-  %weird_gep2 = getelementptr inbounds [18 x i8]* %a, i32 0, i32 16
+  %weird_gep2 = getelementptr inbounds [18 x i8], [18 x i8]* %a, i32 0, i32 16
   %weird_cast2 = bitcast i8* %weird_gep2 to i16*
-  %weird_load2 = load volatile i16* %weird_cast2, align 1
+  %weird_load2 = load volatile i16, i16* %weird_cast2, align 1
 
   ret void
 }
@@ -130,13 +130,13 @@ define void @test6() {
 
 entry:
   %a = alloca [16 x i8]
-  %raw1 = getelementptr inbounds [16 x i8]* %a, i32 0, i32 0
+  %raw1 = getelementptr inbounds [16 x i8], [16 x i8]* %a, i32 0, i32 0
   %ptr1 = bitcast i8* %raw1 to double*
   store volatile double 0.0, double* %ptr1, align 1
 
-  %raw2 = getelementptr inbounds [16 x i8]* %a, i32 0, i32 8
+  %raw2 = getelementptr inbounds [16 x i8], [16 x i8]* %a, i32 0, i32 8
   %ptr2 = bitcast i8* %raw2 to double*
-  %val = load double* %ptr1, align 1
+  %val = load double, double* %ptr1, align 1
   store volatile double %val, double* %ptr2, align 1
 
   ret void
@@ -150,17 +150,17 @@ define void @test7(i8* %out) {
 
 entry:
   %a = alloca [16 x i8]
-  %raw1 = getelementptr inbounds [16 x i8]* %a, i32 0, i32 0
+  %raw1 = getelementptr inbounds [16 x i8], [16 x i8]* %a, i32 0, i32 0
   %ptr1 = bitcast i8* %raw1 to double*
-  %raw2 = getelementptr inbounds [16 x i8]* %a, i32 0, i32 8
+  %raw2 = getelementptr inbounds [16 x i8], [16 x i8]* %a, i32 0, i32 8
   %ptr2 = bitcast i8* %raw2 to double*
 
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %raw1, i8* %out, i32 16, i32 0, i1 false)
-; CHECK: %[[val2:.*]] = load double* %{{.*}}, align 1
-; CHECK: %[[val1:.*]] = load double* %{{.*}}, align 1
+; CHECK: %[[val2:.*]] = load double, double* %{{.*}}, align 1
+; CHECK: %[[val1:.*]] = load double, double* %{{.*}}, align 1
 
-  %val1 = load double* %ptr2, align 1
-  %val2 = load double* %ptr1, align 1
+  %val1 = load double, double* %ptr2, align 1
+  %val2 = load double, double* %ptr1, align 1
 
   store double %val1, double* %ptr1, align 1
   store double %val2, double* %ptr2, align 1
diff --git a/test/Transforms/SROA/basictest.ll b/test/Transforms/SROA/basictest.ll
index e3f762abfa39..a59192d718cf 100644
--- a/test/Transforms/SROA/basictest.ll
+++ b/test/Transforms/SROA/basictest.ll
@@ -19,7 +19,7 @@ entry:
   call void @llvm.lifetime.start(i64 4, i8* %a1.i8)
 
   store i32 0, i32* %a1
-  %v1 = load i32* %a1
+  %v1 = load i32, i32* %a1
 
   call void @llvm.lifetime.end(i64 4, i8* %a1.i8)
 
@@ -27,7 +27,7 @@ entry:
   call void @llvm.lifetime.start(i64 4, i8* %a2.i8)
 
   store float 0.0, float* %a2
-  %v2 = load float * %a2
+  %v2 = load float , float * %a2
   %v2.int = bitcast float %v2 to i32
   %sum1 = add i32 %v1, %v2.int
 
@@ -43,9 +43,9 @@ define i32 @test1() {
 
 entry:
   %X = alloca { i32, float }
-  %Y = getelementptr { i32, float }* %X, i64 0, i32 0
+  %Y = getelementptr { i32, float }, { i32, float }* %X, i64 0, i32 0
   store i32 0, i32* %Y
-  %Z = load i32* %Y
+  %Z = load i32, i32* %Y
   ret i32 %Z
 }
 
@@ -61,7 +61,7 @@ entry:
   br label %L2
 
 L2:
-  %Z = load i64* %B
+  %Z = load i64, i64* %B
   ret i64 %Z
 }
 
@@ -79,48 +79,48 @@ entry:
 ; CHECK-NEXT: %[[test3_a6:.*]] = alloca [7 x i8]
 ; CHECK-NEXT: %[[test3_a7:.*]] = alloca [85 x i8]
 
-  %b = getelementptr [300 x i8]* %a, i64 0, i64 0
+  %b = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b, i8* %src, i32 300, i32 1, i1 false)
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8]* %[[test3_a1]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8], [42 x i8]* %[[test3_a1]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %src, i32 42
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 42
-; CHECK-NEXT: %[[test3_r1:.*]] = load i8* %[[gep]]
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 43
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [99 x i8]* %[[test3_a2]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 42
+; CHECK-NEXT: %[[test3_r1:.*]] = load i8, i8* %[[gep]]
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 43
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [99 x i8], [99 x i8]* %[[test3_a2]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 99
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 142
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 142
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 16
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 158
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8]* %[[test3_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 158
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8], [42 x i8]* %[[test3_a4]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 42
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 200
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 200
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 207
-; CHECK-NEXT: %[[test3_r2:.*]] = load i8* %[[gep]]
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 208
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 207
+; CHECK-NEXT: %[[test3_r2:.*]] = load i8, i8* %[[gep]]
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 208
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 215
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [85 x i8]* %[[test3_a7]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 215
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [85 x i8], [85 x i8]* %[[test3_a7]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 85
 
   ; Clobber a single element of the array, this should be promotable.
-  %c = getelementptr [300 x i8]* %a, i64 0, i64 42
+  %c = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 42
   store i8 0, i8* %c
 
   ; Make a sequence of overlapping stores to the array. These overlap both in
   ; forward strides and in shrinking accesses.
-  %overlap.1.i8 = getelementptr [300 x i8]* %a, i64 0, i64 142
-  %overlap.2.i8 = getelementptr [300 x i8]* %a, i64 0, i64 143
-  %overlap.3.i8 = getelementptr [300 x i8]* %a, i64 0, i64 144
-  %overlap.4.i8 = getelementptr [300 x i8]* %a, i64 0, i64 145
-  %overlap.5.i8 = getelementptr [300 x i8]* %a, i64 0, i64 146
-  %overlap.6.i8 = getelementptr [300 x i8]* %a, i64 0, i64 147
-  %overlap.7.i8 = getelementptr [300 x i8]* %a, i64 0, i64 148
-  %overlap.8.i8 = getelementptr [300 x i8]* %a, i64 0, i64 149
-  %overlap.9.i8 = getelementptr [300 x i8]* %a, i64 0, i64 150
+  %overlap.1.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 142
+  %overlap.2.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 143
+  %overlap.3.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 144
+  %overlap.4.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 145
+  %overlap.5.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 146
+  %overlap.6.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 147
+  %overlap.7.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 148
+  %overlap.8.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 149
+  %overlap.9.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 150
   %overlap.1.i16 = bitcast i8* %overlap.1.i8 to i16*
   %overlap.1.i32 = bitcast i8* %overlap.1.i8 to i32*
   %overlap.1.i64 = bitcast i8* %overlap.1.i8 to i64*
@@ -133,7 +133,7 @@ entry:
   %overlap.8.i64 = bitcast i8* %overlap.8.i8 to i64*
   %overlap.9.i64 = bitcast i8* %overlap.9.i8 to i64*
   store i8 1, i8* %overlap.1.i8
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 0
 ; CHECK-NEXT: store i8 1, i8* %[[gep]]
   store i16 1, i16* %overlap.1.i16
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast [16 x i8]* %[[test3_a3]] to i16*
@@ -145,48 +145,48 @@ entry:
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast [16 x i8]* %[[test3_a3]] to i64*
 ; CHECK-NEXT: store i64 1, i64* %[[bitcast]]
   store i64 2, i64* %overlap.2.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 1
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 2, i64* %[[bitcast]]
   store i64 3, i64* %overlap.3.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 2
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 2
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 3, i64* %[[bitcast]]
   store i64 4, i64* %overlap.4.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 3
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 3
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 4, i64* %[[bitcast]]
   store i64 5, i64* %overlap.5.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 4
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 4
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 5, i64* %[[bitcast]]
   store i64 6, i64* %overlap.6.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 5
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 5
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 6, i64* %[[bitcast]]
   store i64 7, i64* %overlap.7.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 6
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 6
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 7, i64* %[[bitcast]]
   store i64 8, i64* %overlap.8.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 7
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 7
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 8, i64* %[[bitcast]]
   store i64 9, i64* %overlap.9.i64
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 8
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 8
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i64*
 ; CHECK-NEXT: store i64 9, i64* %[[bitcast]]
 
   ; Make two sequences of overlapping stores with more gaps and irregularities.
-  %overlap2.1.0.i8 = getelementptr [300 x i8]* %a, i64 0, i64 200
-  %overlap2.1.1.i8 = getelementptr [300 x i8]* %a, i64 0, i64 201
-  %overlap2.1.2.i8 = getelementptr [300 x i8]* %a, i64 0, i64 202
-  %overlap2.1.3.i8 = getelementptr [300 x i8]* %a, i64 0, i64 203
+  %overlap2.1.0.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 200
+  %overlap2.1.1.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 201
+  %overlap2.1.2.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 202
+  %overlap2.1.3.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 203
 
-  %overlap2.2.0.i8 = getelementptr [300 x i8]* %a, i64 0, i64 208
-  %overlap2.2.1.i8 = getelementptr [300 x i8]* %a, i64 0, i64 209
-  %overlap2.2.2.i8 = getelementptr [300 x i8]* %a, i64 0, i64 210
-  %overlap2.2.3.i8 = getelementptr [300 x i8]* %a, i64 0, i64 211
+  %overlap2.2.0.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 208
+  %overlap2.2.1.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 209
+  %overlap2.2.2.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 210
+  %overlap2.2.3.i8 = getelementptr [300 x i8], [300 x i8]* %a, i64 0, i64 211
 
   %overlap2.1.0.i16 = bitcast i8* %overlap2.1.0.i8 to i16*
   %overlap2.1.0.i32 = bitcast i8* %overlap2.1.0.i8 to i32*
@@ -194,7 +194,7 @@ entry:
   %overlap2.1.2.i32 = bitcast i8* %overlap2.1.2.i8 to i32*
   %overlap2.1.3.i32 = bitcast i8* %overlap2.1.3.i8 to i32*
   store i8 1,  i8*  %overlap2.1.0.i8
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 0
 ; CHECK-NEXT: store i8 1, i8* %[[gep]]
   store i16 1, i16* %overlap2.1.0.i16
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast [7 x i8]* %[[test3_a5]] to i16*
@@ -203,15 +203,15 @@ entry:
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast [7 x i8]* %[[test3_a5]] to i32*
 ; CHECK-NEXT: store i32 1, i32* %[[bitcast]]
   store i32 2, i32* %overlap2.1.1.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 1
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 2, i32* %[[bitcast]]
   store i32 3, i32* %overlap2.1.2.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 2
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 2
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 3, i32* %[[bitcast]]
   store i32 4, i32* %overlap2.1.3.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 3
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 3
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 4, i32* %[[bitcast]]
 
@@ -224,78 +224,78 @@ entry:
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast [7 x i8]* %[[test3_a6]] to i32*
 ; CHECK-NEXT: store i32 1, i32* %[[bitcast]]
   store i8 1,  i8*  %overlap2.2.1.i8
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 1
 ; CHECK-NEXT: store i8 1, i8* %[[gep]]
   store i16 1, i16* %overlap2.2.1.i16
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 1
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
 ; CHECK-NEXT: store i16 1, i16* %[[bitcast]]
   store i32 1, i32* %overlap2.2.1.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 1
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 1, i32* %[[bitcast]]
   store i32 3, i32* %overlap2.2.2.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 2
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 2
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 3, i32* %[[bitcast]]
   store i32 4, i32* %overlap2.2.3.i32
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 3
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 3
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i32*
 ; CHECK-NEXT: store i32 4, i32* %[[bitcast]]
 
-  %overlap2.prefix = getelementptr i8* %overlap2.1.1.i8, i64 -4
+  %overlap2.prefix = getelementptr i8, i8* %overlap2.1.1.i8, i64 -4
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %overlap2.prefix, i8* %src, i32 8, i32 1, i1 false)
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8]* %[[test3_a4]], i64 0, i64 39
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [42 x i8], [42 x i8]* %[[test3_a4]], i64 0, i64 39
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %src, i32 3
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 3
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 3
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 5
 
   ; Bridge between the overlapping areas
   call void @llvm.memset.p0i8.i32(i8* %overlap2.1.2.i8, i8 42, i32 8, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 2
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 2
 ; CHECK-NEXT: call void @llvm.memset.p0i8.i32(i8* %[[gep]], i8 42, i32 5
 ; ...promoted i8 store...
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memset.p0i8.i32(i8* %[[gep]], i8 42, i32 2
 
   ; Entirely within the second overlap.
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %overlap2.2.1.i8, i8* %src, i32 5, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 1
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 1
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep]], i8* %src, i32 5
 
   ; Trailing past the second overlap.
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %overlap2.2.2.i8, i8* %src, i32 8, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 2
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 2
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep]], i8* %src, i32 5
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 5
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [85 x i8]* %[[test3_a7]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 5
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [85 x i8], [85 x i8]* %[[test3_a7]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 3
 
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %b, i32 300, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [42 x i8]* %[[test3_a1]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [42 x i8], [42 x i8]* %[[test3_a1]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %[[gep]], i32 42
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 42
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 42
 ; CHECK-NEXT: store i8 0, i8* %[[gep]]
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 43
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [99 x i8]* %[[test3_a2]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 43
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [99 x i8], [99 x i8]* %[[test3_a2]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 99
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 142
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [16 x i8]* %[[test3_a3]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 142
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [16 x i8], [16 x i8]* %[[test3_a3]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 16
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 158
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [42 x i8]* %[[test3_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 158
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [42 x i8], [42 x i8]* %[[test3_a4]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 42
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 200
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 200
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 207
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 207
 ; CHECK-NEXT: store i8 42, i8* %[[gep]]
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 208
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test3_a6]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 208
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test3_a6]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 215
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [85 x i8]* %[[test3_a7]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 215
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [85 x i8], [85 x i8]* %[[test3_a7]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 85
 
   ret void
@@ -314,90 +314,90 @@ entry:
 ; CHECK-NEXT: %[[test4_a5:.*]] = alloca [7 x i8]
 ; CHECK-NEXT: %[[test4_a6:.*]] = alloca [40 x i8]
 
-  %b = getelementptr [100 x i8]* %a, i64 0, i64 0
+  %b = getelementptr [100 x i8], [100 x i8]* %a, i64 0, i64 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b, i8* %src, i32 100, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [20 x i8]* %[[test4_a1]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [20 x i8], [20 x i8]* %[[test4_a1]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep]], i8* %src, i32 20
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 20
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 20
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
-; CHECK-NEXT: %[[test4_r1:.*]] = load i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 22
-; CHECK-NEXT: %[[test4_r2:.*]] = load i8* %[[gep]]
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 23
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a2]], i64 0, i64 0
+; CHECK-NEXT: %[[test4_r1:.*]] = load i16, i16* %[[bitcast]]
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 22
+; CHECK-NEXT: %[[test4_r2:.*]] = load i8, i8* %[[gep]]
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 23
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a2]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 30
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [10 x i8]* %[[test4_a3]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 30
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [10 x i8], [10 x i8]* %[[test4_a3]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 10
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 40
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 40
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
-; CHECK-NEXT: %[[test4_r3:.*]] = load i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 42
-; CHECK-NEXT: %[[test4_r4:.*]] = load i8* %[[gep]]
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 43
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[test4_r3:.*]] = load i16, i16* %[[bitcast]]
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 42
+; CHECK-NEXT: %[[test4_r4:.*]] = load i8, i8* %[[gep]]
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 43
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a4]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 50
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 50
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
-; CHECK-NEXT: %[[test4_r5:.*]] = load i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %src, i64 52
-; CHECK-NEXT: %[[test4_r6:.*]] = load i8* %[[gep]]
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 53
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[test4_r5:.*]] = load i16, i16* %[[bitcast]]
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %src, i64 52
+; CHECK-NEXT: %[[test4_r6:.*]] = load i8, i8* %[[gep]]
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 53
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8* %src, i64 60
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [40 x i8]* %[[test4_a6]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds i8, i8* %src, i64 60
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [40 x i8], [40 x i8]* %[[test4_a6]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 40
 
-  %a.src.1 = getelementptr [100 x i8]* %a, i64 0, i64 20
-  %a.dst.1 = getelementptr [100 x i8]* %a, i64 0, i64 40
+  %a.src.1 = getelementptr [100 x i8], [100 x i8]* %a, i64 0, i64 20
+  %a.dst.1 = getelementptr [100 x i8], [100 x i8]* %a, i64 0, i64 40
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.dst.1, i8* %a.src.1, i32 10, i32 1, i1 false)
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a4]], i64 0, i64 0
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a2]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a2]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
 
   ; Clobber a single element of the array, this should be promotable, and be deleted.
-  %c = getelementptr [100 x i8]* %a, i64 0, i64 42
+  %c = getelementptr [100 x i8], [100 x i8]* %a, i64 0, i64 42
   store i8 0, i8* %c
 
-  %a.src.2 = getelementptr [100 x i8]* %a, i64 0, i64 50
+  %a.src.2 = getelementptr [100 x i8], [100 x i8]* %a, i64 0, i64 50
   call void @llvm.memmove.p0i8.p0i8.i32(i8* %a.dst.1, i8* %a.src.2, i32 10, i32 1, i1 false)
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a4]], i64 0, i64 0
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
 
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %b, i32 100, i32 1, i1 false)
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [20 x i8]* %[[test4_a1]], i64 0, i64 0
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds [20 x i8], [20 x i8]* %[[test4_a1]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %[[gep]], i32 20
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 20
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 20
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
 ; CHECK-NEXT: store i16 %[[test4_r1]], i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 22
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 22
 ; CHECK-NEXT: store i8 %[[test4_r2]], i8* %[[gep]]
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 23
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a2]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 23
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a2]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 30
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [10 x i8]* %[[test4_a3]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 30
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [10 x i8], [10 x i8]* %[[test4_a3]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 10
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 40
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 40
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
 ; CHECK-NEXT: store i16 %[[test4_r5]], i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 42
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 42
 ; CHECK-NEXT: store i8 %[[test4_r6]], i8* %[[gep]]
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 43
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a4]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 43
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a4]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 50
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 50
 ; CHECK-NEXT: %[[bitcast:.*]] = bitcast i8* %[[gep]] to i16*
 ; CHECK-NEXT: store i16 %[[test4_r5]], i16* %[[bitcast]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8* %dst, i64 52
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds i8, i8* %dst, i64 52
 ; CHECK-NEXT: store i8 %[[test4_r6]], i8* %[[gep]]
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 53
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8]* %[[test4_a5]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 53
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [7 x i8], [7 x i8]* %[[test4_a5]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 7
-; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8* %dst, i64 60
-; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [40 x i8]* %[[test4_a6]], i64 0, i64 0
+; CHECK-NEXT: %[[gep_dst:.*]] = getelementptr inbounds i8, i8* %dst, i64 60
+; CHECK-NEXT: %[[gep_src:.*]] = getelementptr inbounds [40 x i8], [40 x i8]* %[[test4_a6]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[gep_dst]], i8* %[[gep_src]], i32 40
 
   ret void
@@ -420,9 +420,9 @@ entry:
   %a = alloca [4 x i8]
   %fptr = bitcast [4 x i8]* %a to float*
   store float 0.0, float* %fptr
-  %ptr = getelementptr [4 x i8]* %a, i32 0, i32 2
+  %ptr = getelementptr [4 x i8], [4 x i8]* %a, i32 0, i32 2
   %iptr = bitcast i8* %ptr to i16*
-  %val = load i16* %iptr
+  %val = load i16, i16* %iptr
   ret i16 %val
 }
 
@@ -430,15 +430,15 @@ define i32 @test6() {
 ; CHECK-LABEL: @test6(
 ; CHECK: alloca i32
 ; CHECK-NEXT: store volatile i32
-; CHECK-NEXT: load i32*
+; CHECK-NEXT: load i32, i32*
 ; CHECK-NEXT: ret i32
 
 entry:
   %a = alloca [4 x i8]
-  %ptr = getelementptr [4 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [4 x i8], [4 x i8]* %a, i32 0, i32 0
   call void @llvm.memset.p0i8.i32(i8* %ptr, i8 42, i32 4, i32 1, i1 true)
   %iptr = bitcast i8* %ptr to i32*
-  %val = load i32* %iptr
+  %val = load i32, i32* %iptr
   ret i32 %val
 }
 
@@ -446,16 +446,16 @@ define void @test7(i8* %src, i8* %dst) {
 ; CHECK-LABEL: @test7(
 ; CHECK: alloca i32
 ; CHECK-NEXT: bitcast i8* %src to i32*
-; CHECK-NEXT: load volatile i32*
+; CHECK-NEXT: load volatile i32, i32*
 ; CHECK-NEXT: store volatile i32
 ; CHECK-NEXT: bitcast i8* %dst to i32*
-; CHECK-NEXT: load volatile i32*
+; CHECK-NEXT: load volatile i32, i32*
 ; CHECK-NEXT: store volatile i32
 ; CHECK-NEXT: ret
 
 entry:
   %a = alloca [4 x i8]
-  %ptr = getelementptr [4 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [4 x i8], [4 x i8]* %a, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %ptr, i8* %src, i32 4, i32 1, i1 true)
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %ptr, i32 4, i32 1, i1 true)
   ret void
@@ -471,28 +471,28 @@ entry:
   %new = alloca %S2
 ; CHECK-NOT: alloca
 
-  %s2.next.ptr = getelementptr %S2* %s2, i64 0, i32 1
-  %s2.next = load %S2** %s2.next.ptr
-; CHECK:      %[[gep:.*]] = getelementptr %S2* %s2, i64 0, i32 1
-; CHECK-NEXT: %[[next:.*]] = load %S2** %[[gep]]
+  %s2.next.ptr = getelementptr %S2, %S2* %s2, i64 0, i32 1
+  %s2.next = load %S2*, %S2** %s2.next.ptr
+; CHECK:      %[[gep:.*]] = getelementptr %S2, %S2* %s2, i64 0, i32 1
+; CHECK-NEXT: %[[next:.*]] = load %S2*, %S2** %[[gep]]
 
-  %s2.next.s1.ptr = getelementptr %S2* %s2.next, i64 0, i32 0
-  %s2.next.s1 = load %S1** %s2.next.s1.ptr
-  %new.s1.ptr = getelementptr %S2* %new, i64 0, i32 0
+  %s2.next.s1.ptr = getelementptr %S2, %S2* %s2.next, i64 0, i32 0
+  %s2.next.s1 = load %S1*, %S1** %s2.next.s1.ptr
+  %new.s1.ptr = getelementptr %S2, %S2* %new, i64 0, i32 0
   store %S1* %s2.next.s1, %S1** %new.s1.ptr
-  %s2.next.next.ptr = getelementptr %S2* %s2.next, i64 0, i32 1
-  %s2.next.next = load %S2** %s2.next.next.ptr
-  %new.next.ptr = getelementptr %S2* %new, i64 0, i32 1
+  %s2.next.next.ptr = getelementptr %S2, %S2* %s2.next, i64 0, i32 1
+  %s2.next.next = load %S2*, %S2** %s2.next.next.ptr
+  %new.next.ptr = getelementptr %S2, %S2* %new, i64 0, i32 1
   store %S2* %s2.next.next, %S2** %new.next.ptr
-; CHECK-NEXT: %[[gep:.*]] = getelementptr %S2* %[[next]], i64 0, i32 0
-; CHECK-NEXT: %[[next_s1:.*]] = load %S1** %[[gep]]
-; CHECK-NEXT: %[[gep:.*]] = getelementptr %S2* %[[next]], i64 0, i32 1
-; CHECK-NEXT: %[[next_next:.*]] = load %S2** %[[gep]]
+; CHECK-NEXT: %[[gep:.*]] = getelementptr %S2, %S2* %[[next]], i64 0, i32 0
+; CHECK-NEXT: %[[next_s1:.*]] = load %S1*, %S1** %[[gep]]
+; CHECK-NEXT: %[[gep:.*]] = getelementptr %S2, %S2* %[[next]], i64 0, i32 1
+; CHECK-NEXT: %[[next_next:.*]] = load %S2*, %S2** %[[gep]]
 
-  %new.s1 = load %S1** %new.s1.ptr
+  %new.s1 = load %S1*, %S1** %new.s1.ptr
   %result1 = insertvalue %S2 undef, %S1* %new.s1, 0
 ; CHECK-NEXT: %[[result1:.*]] = insertvalue %S2 undef, %S1* %[[next_s1]], 0
-  %new.next = load %S2** %new.next.ptr
+  %new.next = load %S2*, %S2** %new.next.ptr
   %result2 = insertvalue %S2 %result1, %S2* %new.next, 1
 ; CHECK-NEXT: %[[result2:.*]] = insertvalue %S2 %[[result1]], %S2* %[[next_next]], 1
   ret %S2 %result2
@@ -522,15 +522,15 @@ define i64 @test9() {
 
 entry:
   %a = alloca { [3 x i8] }, align 8
-  %gep1 = getelementptr inbounds { [3 x i8] }* %a, i32 0, i32 0, i32 0
+  %gep1 = getelementptr inbounds { [3 x i8] }, { [3 x i8] }* %a, i32 0, i32 0, i32 0
   store i8 0, i8* %gep1, align 1
-  %gep2 = getelementptr inbounds { [3 x i8] }* %a, i32 0, i32 0, i32 1
+  %gep2 = getelementptr inbounds { [3 x i8] }, { [3 x i8] }* %a, i32 0, i32 0, i32 1
   store i8 0, i8* %gep2, align 1
-  %gep3 = getelementptr inbounds { [3 x i8] }* %a, i32 0, i32 0, i32 2
+  %gep3 = getelementptr inbounds { [3 x i8] }, { [3 x i8] }* %a, i32 0, i32 0, i32 2
   store i8 26, i8* %gep3, align 1
   %cast = bitcast { [3 x i8] }* %a to { i64 }*
-  %elt = getelementptr inbounds { i64 }* %cast, i32 0, i32 0
-  %load = load i64* %elt
+  %elt = getelementptr inbounds { i64 }, { i64 }* %cast, i32 0, i32 0
+  %load = load i64, i64* %elt
   %result = and i64 %load, 16777215
   ret i64 %result
 }
@@ -542,10 +542,10 @@ define %S2* @test10() {
 
 entry:
   %a = alloca [8 x i8]
-  %ptr = getelementptr [8 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [8 x i8], [8 x i8]* %a, i32 0, i32 0
   call void @llvm.memset.p0i8.i32(i8* %ptr, i8 0, i32 8, i32 1, i1 false)
   %s2ptrptr = bitcast i8* %ptr to %S2**
-  %s2ptr = load %S2** %s2ptrptr
+  %s2ptr = load %S2*, %S2** %s2ptrptr
   ret %S2* %s2ptr
 }
 
@@ -559,15 +559,15 @@ entry:
   br i1 undef, label %good, label %bad
 
 good:
-  %Y = getelementptr i32* %X, i64 0
+  %Y = getelementptr i32, i32* %X, i64 0
   store i32 0, i32* %Y
-  %Z = load i32* %Y
+  %Z = load i32, i32* %Y
   ret i32 %Z
 
 bad:
-  %Y2 = getelementptr i32* %X, i64 1
+  %Y2 = getelementptr i32, i32* %X, i64 1
   store i32 0, i32* %Y2
-  %Z2 = load i32* %Y2
+  %Z2 = load i32, i32* %Y2
   ret i32 %Z2
 }
 
@@ -582,14 +582,14 @@ entry:
   %b = alloca [3 x i8]
 ; CHECK-NOT: alloca
 
-  %a0ptr = getelementptr [3 x i8]* %a, i64 0, i32 0
+  %a0ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 0
   store i8 0, i8* %a0ptr
-  %a1ptr = getelementptr [3 x i8]* %a, i64 0, i32 1
+  %a1ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 1
   store i8 0, i8* %a1ptr
-  %a2ptr = getelementptr [3 x i8]* %a, i64 0, i32 2
+  %a2ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 2
   store i8 0, i8* %a2ptr
   %aiptr = bitcast [3 x i8]* %a to i24*
-  %ai = load i24* %aiptr
+  %ai = load i24, i24* %aiptr
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 ; CHECK:      %[[ext2:.*]] = zext i8 0 to i24
@@ -606,12 +606,12 @@ entry:
 
   %biptr = bitcast [3 x i8]* %b to i24*
   store i24 %ai, i24* %biptr
-  %b0ptr = getelementptr [3 x i8]* %b, i64 0, i32 0
-  %b0 = load i8* %b0ptr
-  %b1ptr = getelementptr [3 x i8]* %b, i64 0, i32 1
-  %b1 = load i8* %b1ptr
-  %b2ptr = getelementptr [3 x i8]* %b, i64 0, i32 2
-  %b2 = load i8* %b2ptr
+  %b0ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 0
+  %b0 = load i8, i8* %b0ptr
+  %b1ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 1
+  %b1 = load i8, i8* %b1ptr
+  %b2ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 2
+  %b2 = load i8, i8* %b2ptr
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 ; CHECK:      %[[trunc0:.*]] = trunc i24 %[[insert0]] to i8
@@ -638,15 +638,15 @@ define i32 @test13() {
 
 entry:
   %a = alloca [3 x i8], align 2
-  %b0ptr = getelementptr [3 x i8]* %a, i64 0, i32 0
+  %b0ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 0
   store i8 0, i8* %b0ptr
-  %b1ptr = getelementptr [3 x i8]* %a, i64 0, i32 1
+  %b1ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 1
   store i8 0, i8* %b1ptr
-  %b2ptr = getelementptr [3 x i8]* %a, i64 0, i32 2
+  %b2ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 2
   store i8 0, i8* %b2ptr
   %iptrcast = bitcast [3 x i8]* %a to i16*
-  %iptrgep = getelementptr i16* %iptrcast, i64 1
-  %i = load i16* %iptrgep
+  %iptrgep = getelementptr i16, i16* %iptrcast, i64 1
+  %i = load i16, i16* %iptrgep
   %ret = zext i16 %i to i32
   ret i32 %ret
 }
@@ -666,21 +666,21 @@ entry:
   %a = alloca %test14.struct
   %p = alloca %test14.struct*
   %0 = bitcast %test14.struct* %a to i8*
-  %1 = getelementptr i8* %0, i64 12
+  %1 = getelementptr i8, i8* %0, i64 12
   %2 = bitcast i8* %1 to %test14.struct*
-  %3 = getelementptr inbounds %test14.struct* %2, i32 0, i32 0
-  %4 = getelementptr inbounds %test14.struct* %a, i32 0, i32 0
+  %3 = getelementptr inbounds %test14.struct, %test14.struct* %2, i32 0, i32 0
+  %4 = getelementptr inbounds %test14.struct, %test14.struct* %a, i32 0, i32 0
   %5 = bitcast [3 x i32]* %3 to i32*
   %6 = bitcast [3 x i32]* %4 to i32*
-  %7 = load i32* %6, align 4
+  %7 = load i32, i32* %6, align 4
   store i32 %7, i32* %5, align 4
-  %8 = getelementptr inbounds i32* %5, i32 1
-  %9 = getelementptr inbounds i32* %6, i32 1
-  %10 = load i32* %9, align 4
+  %8 = getelementptr inbounds i32, i32* %5, i32 1
+  %9 = getelementptr inbounds i32, i32* %6, i32 1
+  %10 = load i32, i32* %9, align 4
   store i32 %10, i32* %8, align 4
-  %11 = getelementptr inbounds i32* %5, i32 2
-  %12 = getelementptr inbounds i32* %6, i32 2
-  %13 = load i32* %12, align 4
+  %11 = getelementptr inbounds i32, i32* %5, i32 2
+  %12 = getelementptr inbounds i32, i32* %6, i32 2
+  %13 = load i32, i32* %12, align 4
   store i32 %13, i32* %11, align 4
   ret void
 }
@@ -707,25 +707,25 @@ loop:
 
   store i64 1879048192, i64* %l0, align 8
   %bc0 = bitcast i64* %l0 to i8*
-  %gep0 = getelementptr i8* %bc0, i64 3
+  %gep0 = getelementptr i8, i8* %bc0, i64 3
   %dead0 = bitcast i8* %gep0 to i64*
 
   store i64 1879048192, i64* %l1, align 8
   %bc1 = bitcast i64* %l1 to i8*
-  %gep1 = getelementptr i8* %bc1, i64 3
-  %dead1 = getelementptr i8* %gep1, i64 1
+  %gep1 = getelementptr i8, i8* %bc1, i64 3
+  %dead1 = getelementptr i8, i8* %gep1, i64 1
 
   store i64 1879048192, i64* %l2, align 8
   %bc2 = bitcast i64* %l2 to i8*
-  %gep2.1 = getelementptr i8* %bc2, i64 1
-  %gep2.2 = getelementptr i8* %bc2, i64 3
+  %gep2.1 = getelementptr i8, i8* %bc2, i64 1
+  %gep2.2 = getelementptr i8, i8* %bc2, i64 3
   ; Note that this select should get visited multiple times due to using two
   ; different GEPs off the same alloca. We should only delete it once.
   %dead2 = select i1 %flag, i8* %gep2.1, i8* %gep2.2
 
   store i64 1879048192, i64* %l3, align 8
   %bc3 = bitcast i64* %l3 to i8*
-  %gep3 = getelementptr i8* %bc3, i64 3
+  %gep3 = getelementptr i8, i8* %bc3, i64 3
 
   br label %loop
 }
@@ -735,14 +735,14 @@ define void @test16(i8* %src, i8* %dst) {
 ; CHECK-LABEL: @test16(
 ; CHECK-NOT: alloca
 ; CHECK:      %[[srccast:.*]] = bitcast i8* %src to i24*
-; CHECK-NEXT: load i24* %[[srccast]]
+; CHECK-NEXT: load i24, i24* %[[srccast]]
 ; CHECK-NEXT: %[[dstcast:.*]] = bitcast i8* %dst to i24*
 ; CHECK-NEXT: store i24 0, i24* %[[dstcast]]
 ; CHECK-NEXT: ret void
 
 entry:
   %a = alloca [3 x i8]
-  %ptr = getelementptr [3 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [3 x i8], [3 x i8]* %a, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %ptr, i8* %src, i32 4, i32 1, i1 false)
   %cast = bitcast i8* %ptr to i24*
   store i24 0, i24* %cast
@@ -755,14 +755,14 @@ define void @test17(i8* %src, i8* %dst) {
 ; the alloca.
 ; CHECK-LABEL: @test17(
 ; CHECK:      %[[a:.*]] = alloca [3 x i8]
-; CHECK-NEXT: %[[ptr:.*]] = getelementptr [3 x i8]* %[[a]], i32 0, i32 0
+; CHECK-NEXT: %[[ptr:.*]] = getelementptr [3 x i8], [3 x i8]* %[[a]], i32 0, i32 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[ptr]], i8* %src,
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %[[ptr]],
 ; CHECK-NEXT: ret void
 
 entry:
   %a = alloca [3 x i8]
-  %ptr = getelementptr [3 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [3 x i8], [3 x i8]* %a, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %ptr, i8* %src, i32 4, i32 1, i1 true)
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %ptr, i32 4, i32 1, i1 true)
   ret void
@@ -774,27 +774,27 @@ define void @test18(i8* %src, i8* %dst, i32 %size) {
 ; the variable sized intrinsic.
 ; CHECK-LABEL: @test18(
 ; CHECK:      %[[a:.*]] = alloca [34 x i8]
-; CHECK:      %[[srcgep1:.*]] = getelementptr inbounds i8* %src, i64 4
+; CHECK:      %[[srcgep1:.*]] = getelementptr inbounds i8, i8* %src, i64 4
 ; CHECK-NEXT: %[[srccast1:.*]] = bitcast i8* %[[srcgep1]] to i32*
-; CHECK-NEXT: %[[srcload:.*]] = load i32* %[[srccast1]]
-; CHECK-NEXT: %[[agep1:.*]] = getelementptr inbounds [34 x i8]* %[[a]], i64 0, i64 0
+; CHECK-NEXT: %[[srcload:.*]] = load i32, i32* %[[srccast1]]
+; CHECK-NEXT: %[[agep1:.*]] = getelementptr inbounds [34 x i8], [34 x i8]* %[[a]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %[[agep1]], i8* %src, i32 %size,
-; CHECK-NEXT: %[[agep2:.*]] = getelementptr inbounds [34 x i8]* %[[a]], i64 0, i64 0
+; CHECK-NEXT: %[[agep2:.*]] = getelementptr inbounds [34 x i8], [34 x i8]* %[[a]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memset.p0i8.i32(i8* %[[agep2]], i8 42, i32 %size,
 ; CHECK-NEXT: %[[dstcast1:.*]] = bitcast i8* %dst to i32*
 ; CHECK-NEXT: store i32 42, i32* %[[dstcast1]]
-; CHECK-NEXT: %[[dstgep1:.*]] = getelementptr inbounds i8* %dst, i64 4
+; CHECK-NEXT: %[[dstgep1:.*]] = getelementptr inbounds i8, i8* %dst, i64 4
 ; CHECK-NEXT: %[[dstcast2:.*]] = bitcast i8* %[[dstgep1]] to i32*
 ; CHECK-NEXT: store i32 %[[srcload]], i32* %[[dstcast2]]
-; CHECK-NEXT: %[[agep3:.*]] = getelementptr inbounds [34 x i8]* %[[a]], i64 0, i64 0
+; CHECK-NEXT: %[[agep3:.*]] = getelementptr inbounds [34 x i8], [34 x i8]* %[[a]], i64 0, i64 0
 ; CHECK-NEXT: call void @llvm.memcpy.p0i8.p0i8.i32(i8* %dst, i8* %[[agep3]], i32 %size,
 ; CHECK-NEXT: ret void
 
 entry:
   %a = alloca [42 x i8]
-  %ptr = getelementptr [42 x i8]* %a, i32 0, i32 0
+  %ptr = getelementptr [42 x i8], [42 x i8]* %a, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %ptr, i8* %src, i32 8, i32 1, i1 false)
-  %ptr2 = getelementptr [42 x i8]* %a, i32 0, i32 8
+  %ptr2 = getelementptr [42 x i8], [42 x i8]* %a, i32 0, i32 8
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %ptr2, i8* %src, i32 %size, i32 1, i1 false)
   call void @llvm.memset.p0i8.i32(i8* %ptr2, i8 42, i32 %size, i32 1, i1 false)
   %cast = bitcast i8* %ptr to i32*
@@ -820,8 +820,8 @@ entry:
   %cast1 = bitcast %opaque* %x to i8*
   %cast2 = bitcast { i64, i8* }* %a to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %cast2, i8* %cast1, i32 16, i32 1, i1 false)
-  %gep = getelementptr inbounds { i64, i8* }* %a, i32 0, i32 0
-  %val = load i64* %gep
+  %gep = getelementptr inbounds { i64, i8* }, { i64, i8* }* %a, i32 0, i32 0
+  %val = load i64, i64* %gep
   ret i32 undef
 }
 
@@ -836,18 +836,18 @@ define i32 @test20() {
 
 entry:
   %a = alloca [3 x i32]
-  %gep1 = getelementptr [3 x i32]* %a, i32 0, i32 0
+  %gep1 = getelementptr [3 x i32], [3 x i32]* %a, i32 0, i32 0
   store i32 1, i32* %gep1
-  %gep2.1 = getelementptr [3 x i32]* %a, i32 0, i32 -2
-  %gep2.2 = getelementptr i32* %gep2.1, i32 3
+  %gep2.1 = getelementptr [3 x i32], [3 x i32]* %a, i32 0, i32 -2
+  %gep2.2 = getelementptr i32, i32* %gep2.1, i32 3
   store i32 2, i32* %gep2.2
-  %gep3.1 = getelementptr [3 x i32]* %a, i32 0, i32 14
-  %gep3.2 = getelementptr i32* %gep3.1, i32 -12
+  %gep3.1 = getelementptr [3 x i32], [3 x i32]* %a, i32 0, i32 14
+  %gep3.2 = getelementptr i32, i32* %gep3.1, i32 -12
   store i32 3, i32* %gep3.2
 
-  %load1 = load i32* %gep1
-  %load2 = load i32* %gep2.2
-  %load3 = load i32* %gep3.2
+  %load1 = load i32, i32* %gep1
+  %load2 = load i32, i32* %gep2.2
+  %load3 = load i32, i32* %gep3.2
   %sum1 = add i32 %load1, %load2
   %sum2 = add i32 %sum1, %load3
   ret i32 %sum2
@@ -865,20 +865,20 @@ define i8 @test21() {
 
 entry:
   %a = alloca [2305843009213693951 x i8]
-  %gep0 = getelementptr [2305843009213693951 x i8]* %a, i64 0, i64 2305843009213693949
+  %gep0 = getelementptr [2305843009213693951 x i8], [2305843009213693951 x i8]* %a, i64 0, i64 2305843009213693949
   store i8 255, i8* %gep0
-  %gep1 = getelementptr [2305843009213693951 x i8]* %a, i64 0, i64 -9223372036854775807
-  %gep2 = getelementptr i8* %gep1, i64 -1
+  %gep1 = getelementptr [2305843009213693951 x i8], [2305843009213693951 x i8]* %a, i64 0, i64 -9223372036854775807
+  %gep2 = getelementptr i8, i8* %gep1, i64 -1
   call void @llvm.memset.p0i8.i64(i8* %gep2, i8 0, i64 18446744073709551615, i32 1, i1 false)
-  %gep3 = getelementptr i8* %gep1, i64 9223372036854775807
-  %gep4 = getelementptr i8* %gep3, i64 9223372036854775807
-  %gep5 = getelementptr i8* %gep4, i64 -6917529027641081857
+  %gep3 = getelementptr i8, i8* %gep1, i64 9223372036854775807
+  %gep4 = getelementptr i8, i8* %gep3, i64 9223372036854775807
+  %gep5 = getelementptr i8, i8* %gep4, i64 -6917529027641081857
   store i8 255, i8* %gep5
   %cast1 = bitcast i8* %gep4 to i32*
   store i32 0, i32* %cast1
-  %load = load i8* %gep0
-  %gep6 = getelementptr i8* %gep0, i32 1
-  %load2 = load i8* %gep6
+  %load = load i8, i8* %gep0
+  %gep6 = getelementptr i8, i8* %gep0, i32 1
+  %load2 = load i8, i8* %gep6
   %result = or i8 %load, %load2
   ret i8 %result
 }
@@ -895,7 +895,7 @@ define void @PR13916.1() {
 entry:
   %a = alloca i8
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a, i8* %a, i32 1, i32 1, i1 false)
-  %tmp2 = load i8* %a
+  %tmp2 = load i8, i8* %a
   ret void
 }
 
@@ -918,8 +918,8 @@ if.then:
   br label %if.end
 
 if.end:
-  %gep = getelementptr %PR13916.struct* %a, i32 0, i32 0
-  %tmp2 = load i8* %gep
+  %gep = getelementptr %PR13916.struct, %PR13916.struct* %a, i32 0, i32 0
+  %tmp2 = load i8, i8* %gep
   ret void
 }
 
@@ -968,7 +968,7 @@ entry:
   store double* %c, double** %b
   store double* %a, double** %b
   store double %x, double* %c
-  %ret = load double* %a
+  %ret = load double, double* %a
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 
@@ -987,8 +987,8 @@ define void @PR14034() {
 
 entry:
   %a = alloca %PR14034.struct
-  %list = getelementptr %PR14034.struct* %a, i32 0, i32 2
-  %prev = getelementptr %PR14034.list* %list, i32 0, i32 1
+  %list = getelementptr %PR14034.struct, %PR14034.struct* %a, i32 0, i32 2
+  %prev = getelementptr %PR14034.list, %PR14034.list* %list, i32 0, i32 1
   store %PR14034.list* undef, %PR14034.list** %prev
   %cast0 = bitcast %PR14034.struct* undef to i8*
   %cast1 = bitcast %PR14034.struct* %a to i8*
@@ -1008,31 +1008,31 @@ entry:
 ; CHECK-NOT: alloca
 
   %wrap1 = insertvalue [1 x { i32 }] undef, i32 %x, 0, 0
-  %gep1 = getelementptr { { [1 x { i32 }] } }* %a1, i32 0, i32 0, i32 0
+  %gep1 = getelementptr { { [1 x { i32 }] } }, { { [1 x { i32 }] } }* %a1, i32 0, i32 0, i32 0
   store [1 x { i32 }] %wrap1, [1 x { i32 }]* %gep1
 
-  %gep2 = getelementptr { { [1 x { i32 }] } }* %a1, i32 0, i32 0
+  %gep2 = getelementptr { { [1 x { i32 }] } }, { { [1 x { i32 }] } }* %a1, i32 0, i32 0
   %ptrcast1 = bitcast { [1 x { i32 }] }* %gep2 to { [1 x { float }] }*
-  %load1 = load { [1 x { float }] }* %ptrcast1
+  %load1 = load { [1 x { float }] }, { [1 x { float }] }* %ptrcast1
   %unwrap1 = extractvalue { [1 x { float }] } %load1, 0, 0
 
   %wrap2 = insertvalue { {}, { float }, [0 x i8] } undef, { float } %unwrap1, 1
   store { {}, { float }, [0 x i8] } %wrap2, { {}, { float }, [0 x i8] }* %a2
 
-  %gep3 = getelementptr { {}, { float }, [0 x i8] }* %a2, i32 0, i32 1, i32 0
+  %gep3 = getelementptr { {}, { float }, [0 x i8] }, { {}, { float }, [0 x i8] }* %a2, i32 0, i32 1, i32 0
   %ptrcast2 = bitcast float* %gep3 to <4 x i8>*
-  %load3 = load <4 x i8>* %ptrcast2
+  %load3 = load <4 x i8>, <4 x i8>* %ptrcast2
   %valcast1 = bitcast <4 x i8> %load3 to i32
 
   %wrap3 = insertvalue [1 x [1 x i32]] undef, i32 %valcast1, 0, 0
   %wrap4 = insertvalue { [1 x [1 x i32]], {} } undef, [1 x [1 x i32]] %wrap3, 0
-  %gep4 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1
+  %gep4 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }, { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1
   %ptrcast3 = bitcast { [0 x double], [1 x [1 x <4 x i8>]], {} }* %gep4 to { [1 x [1 x i32]], {} }*
   store { [1 x [1 x i32]], {} } %wrap4, { [1 x [1 x i32]], {} }* %ptrcast3
 
-  %gep5 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1, i32 1, i32 0
+  %gep5 = getelementptr { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }, { [0 x i8], { [0 x double], [1 x [1 x <4 x i8>]], {} }, { { {} } } }* %a3, i32 0, i32 1, i32 1, i32 0
   %ptrcast4 = bitcast [1 x <4 x i8>]* %gep5 to { {}, float, {} }*
-  %load4 = load { {}, float, {} }* %ptrcast4
+  %load4 = load { {}, float, {} }, { {}, float, {} }* %ptrcast4
   %unwrap2 = extractvalue { {}, float, {} } %load4, 1
   %valcast2 = bitcast float %unwrap2 to i32
 
@@ -1064,14 +1064,14 @@ entry:
   store i32 0, i32* %X.sroa.0.0.cast2.i, align 8
 
   ; Also use a memset to the middle 32-bits for fun.
-  %X.sroa.0.2.raw_idx2.i = getelementptr inbounds i8* %0, i32 2
+  %X.sroa.0.2.raw_idx2.i = getelementptr inbounds i8, i8* %0, i32 2
   call void @llvm.memset.p0i8.i64(i8* %X.sroa.0.2.raw_idx2.i, i8 0, i64 4, i32 1, i1 false)
 
   ; Or a memset of the whole thing.
   call void @llvm.memset.p0i8.i64(i8* %0, i8 0, i64 8, i32 1, i1 false)
 
   ; Write to the high 32-bits with a memcpy.
-  %X.sroa.0.4.raw_idx4.i = getelementptr inbounds i8* %0, i32 4
+  %X.sroa.0.4.raw_idx4.i = getelementptr inbounds i8, i8* %0, i32 4
   %d.raw = bitcast double* %d to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %X.sroa.0.4.raw_idx4.i, i8* %d.raw, i32 4, i32 1, i1 false)
 
@@ -1080,8 +1080,8 @@ entry:
   store i32 1072693248, i32* %X.sroa.0.4.cast5.i, align 4
 
   ; Do the actual math...
-  %X.sroa.0.0.load1.i = load double* %X.sroa.0.i, align 8
-  %accum.real.i = load double* %d, align 8
+  %X.sroa.0.0.load1.i = load double, double* %X.sroa.0.i, align 8
+  %accum.real.i = load double, double* %d, align 8
   %add.r.i = fadd double %accum.real.i, %X.sroa.0.0.load1.i
   store double %add.r.i, double* %d, align 8
   call void @llvm.lifetime.end(i64 -1, i8* %0)
@@ -1103,17 +1103,17 @@ entry:
   store i64 0, i64* %0
   ; CHECK-NOT: store
 
-  %phi.realp = getelementptr inbounds { float, float }* %phi, i32 0, i32 0
-  %phi.real = load float* %phi.realp
-  %phi.imagp = getelementptr inbounds { float, float }* %phi, i32 0, i32 1
-  %phi.imag = load float* %phi.imagp
-  ; CHECK:      %[[realp:.*]] = getelementptr inbounds { float, float }* %phi, i32 0, i32 0
-  ; CHECK-NEXT: %[[real:.*]] = load float* %[[realp]]
-  ; CHECK-NEXT: %[[imagp:.*]] = getelementptr inbounds { float, float }* %phi, i32 0, i32 1
-  ; CHECK-NEXT: %[[imag:.*]] = load float* %[[imagp]]
+  %phi.realp = getelementptr inbounds { float, float }, { float, float }* %phi, i32 0, i32 0
+  %phi.real = load float, float* %phi.realp
+  %phi.imagp = getelementptr inbounds { float, float }, { float, float }* %phi, i32 0, i32 1
+  %phi.imag = load float, float* %phi.imagp
+  ; CHECK:      %[[realp:.*]] = getelementptr inbounds { float, float }, { float, float }* %phi, i32 0, i32 0
+  ; CHECK-NEXT: %[[real:.*]] = load float, float* %[[realp]]
+  ; CHECK-NEXT: %[[imagp:.*]] = getelementptr inbounds { float, float }, { float, float }* %phi, i32 0, i32 1
+  ; CHECK-NEXT: %[[imag:.*]] = load float, float* %[[imagp]]
 
-  %real = getelementptr inbounds { float, float }* %retval, i32 0, i32 0
-  %imag = getelementptr inbounds { float, float }* %retval, i32 0, i32 1
+  %real = getelementptr inbounds { float, float }, { float, float }* %retval, i32 0, i32 0
+  %imag = getelementptr inbounds { float, float }, { float, float }* %retval, i32 0, i32 1
   store float %phi.real, float* %real
   store float %phi.imag, float* %imag
   ; CHECK-NEXT: %[[real_convert:.*]] = bitcast float %[[real]] to i32
@@ -1126,7 +1126,7 @@ entry:
   ; CHECK-NEXT: %[[real_mask:.*]] = and i64 %[[imag_insert]], -4294967296
   ; CHECK-NEXT: %[[real_insert:.*]] = or i64 %[[real_mask]], %[[real_ext]]
 
-  %1 = load i64* %0, align 1
+  %1 = load i64, i64* %0, align 1
   ret i64 %1
   ; CHECK-NEXT: ret i64 %[[real_insert]]
 }
@@ -1141,8 +1141,8 @@ entry:
   %a = alloca { [16 x i8] }, align 8
 ; CHECK: alloca [16 x i8], align 8
 
-  %gep = getelementptr inbounds { [16 x i8] }* %ptr, i64 -1
-; CHECK-NEXT: getelementptr inbounds { [16 x i8] }* %ptr, i64 -1, i32 0, i64 0
+  %gep = getelementptr inbounds { [16 x i8] }, { [16 x i8] }* %ptr, i64 -1
+; CHECK-NEXT: getelementptr inbounds { [16 x i8] }, { [16 x i8] }* %ptr, i64 -1, i32 0, i64 0
 
   %cast1 = bitcast { [16 x i8 ] }* %gep to i8*
   %cast2 = bitcast { [16 x i8 ] }* %a to i8*
@@ -1159,8 +1159,8 @@ entry:
   %a = alloca { [16 x i8] }, align 8
 ; CHECK: alloca [16 x i8], align 8
 
-  %gep = getelementptr inbounds { [16 x i8] } addrspace(1)* %ptr, i64 -1
-; CHECK-NEXT: getelementptr inbounds { [16 x i8] } addrspace(1)* %ptr, i16 -1, i32 0, i16 0
+  %gep = getelementptr inbounds { [16 x i8] }, { [16 x i8] } addrspace(1)* %ptr, i64 -1
+; CHECK-NEXT: getelementptr inbounds { [16 x i8] }, { [16 x i8] } addrspace(1)* %ptr, i16 -1, i32 0, i16 0
 
   %cast1 = bitcast { [16 x i8 ] } addrspace(1)* %gep to i8 addrspace(1)*
   %cast2 = bitcast { [16 x i8 ] }* %a to i8*
@@ -1199,18 +1199,18 @@ entry:
   %b.i1 = bitcast <{ i1 }>* %b to i1*
   store i1 %x, i1* %b.i1, align 8
   %b.i8 = bitcast <{ i1 }>* %b to i8*
-  %foo = load i8* %b.i8, align 1
+  %foo = load i8, i8* %b.i8, align 1
 ; CHECK-NEXT: %[[ext:.*]] = zext i1 %x to i8
 ; CHECK-NEXT: store i8 %[[ext]], i8* %[[a]], align 8
-; CHECK-NEXT: {{.*}} = load i8* %[[a]], align 8
+; CHECK-NEXT: {{.*}} = load i8, i8* %[[a]], align 8
 
   %a.i8 = bitcast <{ i1 }>* %a to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.i8, i8* %b.i8, i32 1, i32 1, i1 false) nounwind
-  %bar = load i8* %a.i8, align 1
-  %a.i1 = getelementptr inbounds <{ i1 }>* %a, i32 0, i32 0
-  %baz = load i1* %a.i1, align 1
+  %bar = load i8, i8* %a.i8, align 1
+  %a.i1 = getelementptr inbounds <{ i1 }>, <{ i1 }>* %a, i32 0, i32 0
+  %baz = load i1, i1* %a.i1, align 1
 ; CHECK-NEXT: %[[a_cast:.*]] = bitcast i8* %[[a]] to i1*
-; CHECK-NEXT: {{.*}} = load i1* %[[a_cast]], align 8
+; CHECK-NEXT: {{.*}} = load i1, i1* %[[a_cast]], align 8
 
   ret void
 }
@@ -1226,7 +1226,7 @@ entry:
 
   %cast = bitcast <3 x i8>* %a to i32*
   store i32 %x, i32* %cast, align 1
-  %y = load <3 x i8>* %a, align 4
+  %y = load <3 x i8>, <3 x i8>* %a, align 4
   ret <3 x i8> %y
 ; CHECK: ret <3 x i8>
 }
@@ -1242,7 +1242,7 @@ entry:
 
   store <3 x i8> %x, <3 x i8>* %a, align 1
   %cast = bitcast <3 x i8>* %a to i32*
-  %y = load i32* %cast, align 4
+  %y = load i32, i32* %cast, align 4
   ret i32 %y
 ; CHECK: ret i32
 }
@@ -1258,7 +1258,7 @@ entry:
 
   %a.i8 = bitcast i32* %a to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.i8, i8 0, i32 %x, i32 1, i1 false)
-  %v = load i32* %a
+  %v = load i32, i32* %a
   ret i32 %v
 }
 
@@ -1282,36 +1282,36 @@ entry:
   ]
 
 bb4:
-  %src.gep3 = getelementptr inbounds i8* %src, i32 3
-  %src.3 = load i8* %src.gep3
-  %tmp.gep3 = getelementptr inbounds [4 x i8]* %tmp, i32 0, i32 3
+  %src.gep3 = getelementptr inbounds i8, i8* %src, i32 3
+  %src.3 = load i8, i8* %src.gep3
+  %tmp.gep3 = getelementptr inbounds [4 x i8], [4 x i8]* %tmp, i32 0, i32 3
   store i8 %src.3, i8* %tmp.gep3
 ; CHECK: store i8
 
   br label %bb3
 
 bb3:
-  %src.gep2 = getelementptr inbounds i8* %src, i32 2
-  %src.2 = load i8* %src.gep2
-  %tmp.gep2 = getelementptr inbounds [4 x i8]* %tmp, i32 0, i32 2
+  %src.gep2 = getelementptr inbounds i8, i8* %src, i32 2
+  %src.2 = load i8, i8* %src.gep2
+  %tmp.gep2 = getelementptr inbounds [4 x i8], [4 x i8]* %tmp, i32 0, i32 2
   store i8 %src.2, i8* %tmp.gep2
 ; CHECK: store i8
 
   br label %bb2
 
 bb2:
-  %src.gep1 = getelementptr inbounds i8* %src, i32 1
-  %src.1 = load i8* %src.gep1
-  %tmp.gep1 = getelementptr inbounds [4 x i8]* %tmp, i32 0, i32 1
+  %src.gep1 = getelementptr inbounds i8, i8* %src, i32 1
+  %src.1 = load i8, i8* %src.gep1
+  %tmp.gep1 = getelementptr inbounds [4 x i8], [4 x i8]* %tmp, i32 0, i32 1
   store i8 %src.1, i8* %tmp.gep1
 ; CHECK: store i8
 
   br label %bb1
 
 bb1:
-  %src.gep0 = getelementptr inbounds i8* %src, i32 0
-  %src.0 = load i8* %src.gep0
-  %tmp.gep0 = getelementptr inbounds [4 x i8]* %tmp, i32 0, i32 0
+  %src.gep0 = getelementptr inbounds i8, i8* %src, i32 0
+  %src.0 = load i8, i8* %src.gep0
+  %tmp.gep0 = getelementptr inbounds [4 x i8], [4 x i8]* %tmp, i32 0, i32 0
   store i8 %src.0, i8* %tmp.gep0
 ; CHECK: store i8
 
@@ -1332,7 +1332,7 @@ define void @PR15805(i1 %a, i1 %b) {
   %c = alloca i64, align 8
   %p.0.c = select i1 undef, i64* %c, i64* %c
   %cond.in = select i1 undef, i64* %p.0.c, i64* %c
-  %cond = load i64* %cond.in, align 8
+  %cond = load i64, i64* %cond.in, align 8
   ret void
 }
 
@@ -1351,7 +1351,7 @@ define void @PR15805.1(i1 %a, i1 %b) {
 loop:
   %cond.in = select i1 undef, i64* %c, i64* %p.0.c
   %p.0.c = select i1 undef, i64* %c, i64* %c
-  %cond = load i64* %cond.in, align 8
+  %cond = load i64, i64* %cond.in, align 8
   br i1 undef, label %loop, label %exit
 
 exit:
@@ -1373,8 +1373,8 @@ entry:
   %b = alloca i32, align 4
   %b.cast = bitcast i32* %b to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %b.cast, i8* %a, i32 4, i32 4, i1 true)
-  %b.gep = getelementptr inbounds i8* %b.cast, i32 2
-  load i8* %b.gep, align 2
+  %b.gep = getelementptr inbounds i8, i8* %b.cast, i32 2
+  load i8, i8* %b.gep, align 2
   unreachable
 }
 
@@ -1390,11 +1390,11 @@ define void @PR16651.2() {
 
 entry:
   %tv1 = alloca { <2 x float>, <2 x float> }, align 8
-  %0 = getelementptr { <2 x float>, <2 x float> }* %tv1, i64 0, i32 1
+  %0 = getelementptr { <2 x float>, <2 x float> }, { <2 x float>, <2 x float> }* %tv1, i64 0, i32 1
   store <2 x float> undef, <2 x float>* %0, align 8
-  %1 = getelementptr inbounds { <2 x float>, <2 x float> }* %tv1, i64 0, i32 1, i64 0
+  %1 = getelementptr inbounds { <2 x float>, <2 x float> }, { <2 x float>, <2 x float> }* %tv1, i64 0, i32 1, i64 0
   %cond105.in.i.i = select i1 undef, float* null, float* %1
-  %cond105.i.i = load float* %cond105.in.i.i, align 8
+  %cond105.i.i = load float, float* %cond105.in.i.i, align 8
   ret void
 }
 
@@ -1405,8 +1405,8 @@ define void @test23(i32 %x) {
 entry:
   %a = alloca i32, align 4
   store i32 %x, i32* %a, align 4
-  %gep1 = getelementptr inbounds i32* %a, i32 1
-  %gep0 = getelementptr inbounds i32* %a, i32 0
+  %gep1 = getelementptr inbounds i32, i32* %a, i32 1
+  %gep0 = getelementptr inbounds i32, i32* %a, i32 0
   %cast1 = bitcast i32* %gep1 to i8*
   %cast0 = bitcast i32* %gep0 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %cast1, i8* %cast0, i32 4, i32 1, i1 false)
@@ -1419,7 +1419,7 @@ define void @PR18615() {
 ; CHECK: ret void
 entry:
   %f = alloca i8
-  %gep = getelementptr i8* %f, i64 -1
+  %gep = getelementptr i8, i8* %f, i64 -1
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* undef, i8* %gep, i32 1, i32 1, i1 false)
   ret void
 }
@@ -1427,9 +1427,9 @@ entry:
 define void @test24(i8* %src, i8* %dst) {
 ; CHECK-LABEL: @test24(
 ; CHECK: alloca i64, align 16
-; CHECK: load volatile i64* %{{[^,]*}}, align 1
+; CHECK: load volatile i64, i64* %{{[^,]*}}, align 1
 ; CHECK: store volatile i64 %{{[^,]*}}, i64* %{{[^,]*}}, align 16
-; CHECK: load volatile i64* %{{[^,]*}}, align 16
+; CHECK: load volatile i64, i64* %{{[^,]*}}, align 16
 ; CHECK: store volatile i64 %{{[^,]*}}, i64* %{{[^,]*}}, align 1
 
 entry:
@@ -1459,17 +1459,17 @@ entry:
   %a = alloca i64
   %b = alloca i64
   %a.cast = bitcast i64* %a to [2 x float]*
-  %a.gep1 = getelementptr [2 x float]* %a.cast, i32 0, i32 0
-  %a.gep2 = getelementptr [2 x float]* %a.cast, i32 0, i32 1
+  %a.gep1 = getelementptr [2 x float], [2 x float]* %a.cast, i32 0, i32 0
+  %a.gep2 = getelementptr [2 x float], [2 x float]* %a.cast, i32 0, i32 1
   %b.cast = bitcast i64* %b to [2 x float]*
-  %b.gep1 = getelementptr [2 x float]* %b.cast, i32 0, i32 0
-  %b.gep2 = getelementptr [2 x float]* %b.cast, i32 0, i32 1
+  %b.gep1 = getelementptr [2 x float], [2 x float]* %b.cast, i32 0, i32 0
+  %b.gep2 = getelementptr [2 x float], [2 x float]* %b.cast, i32 0, i32 1
   store float 0.0, float* %a.gep1
   store float 1.0, float* %a.gep2
-  %v = load i64* %a
+  %v = load i64, i64* %a
   store i64 %v, i64* %b
-  %f1 = load float* %b.gep1
-  %f2 = load float* %b.gep2
+  %f1 = load float, float* %b.gep1
+  %f2 = load float, float* %b.gep2
   %ret = fadd float %f1, %f2
   ret float %ret
 }
@@ -1482,8 +1482,8 @@ define void @test26() {
 ;
 ; CHECK-LABEL: @test26(
 ; CHECK-NOT: alloca
-; CHECK: %[[L1:.*]] = load i32* bitcast
-; CHECK: %[[L2:.*]] = load i32* bitcast
+; CHECK: %[[L1:.*]] = load i32, i32* bitcast
+; CHECK: %[[L2:.*]] = load i32, i32* bitcast
 ; CHECK: %[[F1:.*]] = bitcast i32 %[[L1]] to float
 ; CHECK: %[[F2:.*]] = bitcast i32 %[[L2]] to float
 ; CHECK: %[[SUM:.*]] = fadd float %[[F1]], %[[F2]]
@@ -1496,16 +1496,16 @@ define void @test26() {
 entry:
   %a = alloca i64
   %a.cast = bitcast i64* %a to [2 x float]*
-  %a.gep1 = getelementptr [2 x float]* %a.cast, i32 0, i32 0
-  %a.gep2 = getelementptr [2 x float]* %a.cast, i32 0, i32 1
-  %v1 = load i64* bitcast ([2 x float]* @complex1 to i64*)
+  %a.gep1 = getelementptr [2 x float], [2 x float]* %a.cast, i32 0, i32 0
+  %a.gep2 = getelementptr [2 x float], [2 x float]* %a.cast, i32 0, i32 1
+  %v1 = load i64, i64* bitcast ([2 x float]* @complex1 to i64*)
   store i64 %v1, i64* %a
-  %f1 = load float* %a.gep1
-  %f2 = load float* %a.gep2
+  %f1 = load float, float* %a.gep1
+  %f2 = load float, float* %a.gep2
   %sum = fadd float %f1, %f2
   store float %sum, float* %a.gep1
   store float %sum, float* %a.gep2
-  %v2 = load i64* %a
+  %v2 = load i64, i64* %a
   store i64 %v2, i64* bitcast ([2 x float]* @complex2 to i64*)
   ret void
 }
@@ -1524,9 +1524,9 @@ define float @test27() {
 
 entry:
   %a = alloca [12 x i8]
-  %gep1 = getelementptr [12 x i8]* %a, i32 0, i32 0
-  %gep2 = getelementptr [12 x i8]* %a, i32 0, i32 4
-  %gep3 = getelementptr [12 x i8]* %a, i32 0, i32 8
+  %gep1 = getelementptr [12 x i8], [12 x i8]* %a, i32 0, i32 0
+  %gep2 = getelementptr [12 x i8], [12 x i8]* %a, i32 0, i32 4
+  %gep3 = getelementptr [12 x i8], [12 x i8]* %a, i32 0, i32 8
   %iptr1 = bitcast i8* %gep1 to i64*
   %iptr2 = bitcast i8* %gep2 to i64*
   %fptr1 = bitcast i8* %gep1 to float*
@@ -1534,10 +1534,10 @@ entry:
   %fptr3 = bitcast i8* %gep3 to float*
   store float 0.0, float* %fptr1
   store float 1.0, float* %fptr2
-  %v = load i64* %iptr1
+  %v = load i64, i64* %iptr1
   store i64 %v, i64* %iptr2
-  %f1 = load float* %fptr2
-  %f2 = load float* %fptr3
+  %f1 = load float, float* %fptr2
+  %f2 = load float, float* %fptr3
   %ret = fadd float %f1, %f2
   ret float %ret
 }
@@ -1560,7 +1560,7 @@ entry:
   %a = alloca i32
   %a.cast = bitcast i32* %a to i16*
   store volatile i16 42, i16* %a.cast
-  %load = load i32* %a
+  %load = load i32, i32* %a
   store i32 %load, i32* %a
   ret i32 %load
 }
@@ -1587,9 +1587,9 @@ entry:
   %a.cast1 = bitcast i64* %a to i32*
   %a.cast2 = bitcast i64* %a to i16*
   store volatile i16 42, i16* %a.cast2
-  %load = load i32* %a.cast1
+  %load = load i32, i32* %a.cast1
   store i32 %load, i32* %a.cast1
-  %a.gep1 = getelementptr i32* %a.cast1, i32 1
+  %a.gep1 = getelementptr i32, i32* %a.cast1, i32 1
   %a.cast3 = bitcast i32* %a.gep1 to i8*
   store volatile i8 13, i8* %a.cast3
   store i32 %load, i32* %a.gep1
diff --git a/test/Transforms/SROA/big-endian.ll b/test/Transforms/SROA/big-endian.ll
index 9e87a9f073c5..b5a04ca8e64a 100644
--- a/test/Transforms/SROA/big-endian.ll
+++ b/test/Transforms/SROA/big-endian.ll
@@ -16,14 +16,14 @@ entry:
   %b = alloca [3 x i8]
 ; CHECK-NOT: alloca
 
-  %a0ptr = getelementptr [3 x i8]* %a, i64 0, i32 0
+  %a0ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 0
   store i8 0, i8* %a0ptr
-  %a1ptr = getelementptr [3 x i8]* %a, i64 0, i32 1
+  %a1ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 1
   store i8 0, i8* %a1ptr
-  %a2ptr = getelementptr [3 x i8]* %a, i64 0, i32 2
+  %a2ptr = getelementptr [3 x i8], [3 x i8]* %a, i64 0, i32 2
   store i8 0, i8* %a2ptr
   %aiptr = bitcast [3 x i8]* %a to i24*
-  %ai = load i24* %aiptr
+  %ai = load i24, i24* %aiptr
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 ; CHECK:      %[[ext2:.*]] = zext i8 0 to i24
@@ -40,12 +40,12 @@ entry:
 
   %biptr = bitcast [3 x i8]* %b to i24*
   store i24 %ai, i24* %biptr
-  %b0ptr = getelementptr [3 x i8]* %b, i64 0, i32 0
-  %b0 = load i8* %b0ptr
-  %b1ptr = getelementptr [3 x i8]* %b, i64 0, i32 1
-  %b1 = load i8* %b1ptr
-  %b2ptr = getelementptr [3 x i8]* %b, i64 0, i32 2
-  %b2 = load i8* %b2ptr
+  %b0ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 0
+  %b0 = load i8, i8* %b0ptr
+  %b1ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 1
+  %b1 = load i8, i8* %b1ptr
+  %b2ptr = getelementptr [3 x i8], [3 x i8]* %b, i64 0, i32 2
+  %b2 = load i8, i8* %b2ptr
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 ; CHECK:      %[[shift0:.*]] = lshr i24 %[[insert0]], 16
@@ -72,10 +72,10 @@ entry:
   %a = alloca [7 x i8]
 ; CHECK-NOT: alloca
 
-  %a0ptr = getelementptr [7 x i8]* %a, i64 0, i32 0
-  %a1ptr = getelementptr [7 x i8]* %a, i64 0, i32 1
-  %a2ptr = getelementptr [7 x i8]* %a, i64 0, i32 2
-  %a3ptr = getelementptr [7 x i8]* %a, i64 0, i32 3
+  %a0ptr = getelementptr [7 x i8], [7 x i8]* %a, i64 0, i32 0
+  %a1ptr = getelementptr [7 x i8], [7 x i8]* %a, i64 0, i32 1
+  %a2ptr = getelementptr [7 x i8], [7 x i8]* %a, i64 0, i32 2
+  %a3ptr = getelementptr [7 x i8], [7 x i8]* %a, i64 0, i32 3
 
 ; CHECK-NOT: store
 ; CHECK-NOT: load
@@ -102,7 +102,7 @@ entry:
 ; CHECK-NOT: load
 
   %aiptr = bitcast [7 x i8]* %a to i56*
-  %ai = load i56* %aiptr
+  %ai = load i56, i56* %aiptr
   %ret = zext i56 %ai to i64
   ret i64 %ret
 ; CHECK-NEXT: %[[ext4:.*]] = zext i16 1 to i56
diff --git a/test/Transforms/SROA/fca.ll b/test/Transforms/SROA/fca.ll
index e8b4c6c13c88..6eaa73f53189 100644
--- a/test/Transforms/SROA/fca.ll
+++ b/test/Transforms/SROA/fca.ll
@@ -14,12 +14,12 @@ entry:
 
   store { i32, i32 } undef, { i32, i32 }* %a
 
-  %gep1 = getelementptr inbounds { i32, i32 }* %a, i32 0, i32 0
+  %gep1 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %a, i32 0, i32 0
   store i32 %x, i32* %gep1
-  %gep2 = getelementptr inbounds { i32, i32 }* %a, i32 0, i32 1
+  %gep2 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %a, i32 0, i32 1
   store i32 %y, i32* %gep2
 
-  %result = load { i32, i32 }* %a
+  %result = load { i32, i32 }, { i32, i32 }* %a
   ret { i32, i32 } %result
 }
 
@@ -30,7 +30,7 @@ define { i32, i32 } @test1(i32 %x, i32 %y) {
 ; CHECK-LABEL: @test1(
 ; CHECK: alloca
 ; CHECK: alloca
-; CHECK: load volatile { i32, i32 }*
+; CHECK: load volatile { i32, i32 }, { i32, i32 }*
 ; CHECK: store volatile { i32, i32 }
 ; CHECK: ret { i32, i32 }
 
@@ -38,12 +38,12 @@ entry:
   %a = alloca { i32, i32 }
   %b = alloca { i32, i32 }
 
-  %gep1 = getelementptr inbounds { i32, i32 }* %a, i32 0, i32 0
+  %gep1 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %a, i32 0, i32 0
   store i32 %x, i32* %gep1
-  %gep2 = getelementptr inbounds { i32, i32 }* %a, i32 0, i32 1
+  %gep2 = getelementptr inbounds { i32, i32 }, { i32, i32 }* %a, i32 0, i32 1
   store i32 %y, i32* %gep2
 
-  %result = load volatile { i32, i32 }* %a
+  %result = load volatile { i32, i32 }, { i32, i32 }* %a
   store volatile { i32, i32 } %result, { i32, i32 }* %b
   ret { i32, i32 } %result
 }
diff --git a/test/Transforms/SROA/phi-and-select.ll b/test/Transforms/SROA/phi-and-select.ll
index f2870127352c..e97bd66d052a 100644
--- a/test/Transforms/SROA/phi-and-select.ll
+++ b/test/Transforms/SROA/phi-and-select.ll
@@ -7,12 +7,12 @@ entry:
 	%a = alloca [2 x i32]
 ; CHECK-NOT: alloca
 
-  %a0 = getelementptr [2 x i32]* %a, i64 0, i32 0
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a0 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 0
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 0, i32* %a0
 	store i32 1, i32* %a1
-	%v0 = load i32* %a0
-	%v1 = load i32* %a1
+	%v0 = load i32, i32* %a0
+	%v1 = load i32, i32* %a1
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 
@@ -26,7 +26,7 @@ exit:
 	%phi = phi i32* [ %a1, %then ], [ %a0, %entry ]
 ; CHECK: phi i32 [ 1, %{{.*}} ], [ 0, %{{.*}} ]
 
-	%result = load i32* %phi
+	%result = load i32, i32* %phi
 	ret i32 %result
 }
 
@@ -36,12 +36,12 @@ entry:
 	%a = alloca [2 x i32]
 ; CHECK-NOT: alloca
 
-  %a0 = getelementptr [2 x i32]* %a, i64 0, i32 0
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a0 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 0
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 0, i32* %a0
 	store i32 1, i32* %a1
-	%v0 = load i32* %a0
-	%v1 = load i32* %a1
+	%v0 = load i32, i32* %a0
+	%v1 = load i32, i32* %a1
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 
@@ -49,7 +49,7 @@ entry:
 	%select = select i1 %cond, i32* %a1, i32* %a0
 ; CHECK: select i1 %{{.*}}, i32 1, i32 0
 
-	%result = load i32* %select
+	%result = load i32, i32* %select
 	ret i32 %result
 }
 
@@ -62,10 +62,10 @@ entry:
   ; Note that we build redundant GEPs here to ensure that having different GEPs
   ; into the same alloca partation continues to work with PHI speculation. This
   ; was the underlying cause of PR13926.
-  %a0 = getelementptr [2 x i32]* %a, i64 0, i32 0
-  %a0b = getelementptr [2 x i32]* %a, i64 0, i32 0
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
-  %a1b = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a0 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 0
+  %a0b = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 0
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
+  %a1b = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 0, i32* %a0
 	store i32 1, i32* %a1
 ; CHECK-NOT: store
@@ -100,7 +100,7 @@ exit:
                   [ %a1b, %bb4 ], [ %a0b, %bb5 ], [ %a0b, %bb6 ], [ %a1b, %bb7 ]
 ; CHECK: phi i32 [ 1, %{{.*}} ], [ 0, %{{.*}} ], [ 0, %{{.*}} ], [ 1, %{{.*}} ], [ 1, %{{.*}} ], [ 0, %{{.*}} ], [ 0, %{{.*}} ], [ 1, %{{.*}} ]
 
-	%result = load i32* %phi
+	%result = load i32, i32* %phi
 	ret i32 %result
 }
 
@@ -110,12 +110,12 @@ entry:
 	%a = alloca [2 x i32]
 ; CHECK-NOT: alloca
 
-  %a0 = getelementptr [2 x i32]* %a, i64 0, i32 0
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a0 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 0
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 0, i32* %a0
 	store i32 1, i32* %a1
-	%v0 = load i32* %a0
-	%v1 = load i32* %a1
+	%v0 = load i32, i32* %a0
+	%v1 = load i32, i32* %a1
 ; CHECK-NOT: store
 ; CHECK-NOT: load
 
@@ -123,7 +123,7 @@ entry:
 	%select = select i1 %cond, i32* %a0, i32* %a0
 ; CHECK-NOT: select
 
-	%result = load i32* %select
+	%result = load i32, i32* %select
 	ret i32 %result
 ; CHECK: ret i32 0
 }
@@ -134,14 +134,14 @@ entry:
 	%a = alloca [2 x i32]
 ; CHECK-NOT: alloca
 
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 1, i32* %a1
 ; CHECK-NOT: store
 
 	%select = select i1 true, i32* %a1, i32* %b
 ; CHECK-NOT: select
 
-	%result = load i32* %select
+	%result = load i32, i32* %select
 ; CHECK-NOT: load
 
 	ret i32 %result
@@ -157,7 +157,7 @@ entry:
   %c = alloca i32
 ; CHECK-NOT: alloca
 
-  %a1 = getelementptr [2 x i32]* %a, i64 0, i32 1
+  %a1 = getelementptr [2 x i32], [2 x i32]* %a, i64 0, i32 1
 	store i32 1, i32* %a1
 
 	%select = select i1 true, i32* %a1, i32* %b
@@ -172,10 +172,10 @@ entry:
 ; CHECK: call void @f(i32* %[[select2]], i32* %[[select3]])
 
 
-	%result = load i32* %select
+	%result = load i32, i32* %select
 ; CHECK-NOT: load
 
-  %dead = load i32* %c
+  %dead = load i32, i32* %c
 
 	ret i32 %result
 ; CHECK: ret i32 1
@@ -190,19 +190,19 @@ entry:
   br i1 undef, label %good, label %bad
 
 good:
-  %Y1 = getelementptr i32* %X, i64 0
+  %Y1 = getelementptr i32, i32* %X, i64 0
   store i32 0, i32* %Y1
   br label %exit
 
 bad:
-  %Y2 = getelementptr i32* %X, i64 1
+  %Y2 = getelementptr i32, i32* %X, i64 1
   store i32 0, i32* %Y2
   br label %exit
 
 exit:
 	%P = phi i32* [ %Y1, %good ], [ %Y2, %bad ]
 ; CHECK: %[[phi:.*]] = phi i32 [ 0, %good ],
-  %Z2 = load i32* %P
+  %Z2 = load i32, i32* %P
   ret i32 %Z2
 ; CHECK: ret i32 %[[phi]]
 }
@@ -213,7 +213,7 @@ define i32 @test8(i32 %b, i32* %ptr) {
 ; CHECK-LABEL: @test8(
 ; CHECK-NOT: alloca
 ; CHECK-NOT: load
-; CHECK: %[[value:.*]] = load i32* %ptr
+; CHECK: %[[value:.*]] = load i32, i32* %ptr
 ; CHECK-NOT: load
 ; CHECK: %[[result:.*]] = phi i32 [ undef, %else ], [ %[[value]], %then ]
 ; CHECK-NEXT: ret i32 %[[result]]
@@ -232,7 +232,7 @@ else:
 
 exit:
   %phi = phi i32* [ %bitcast, %else ], [ %ptr, %then ]
-  %loaded = load i32* %phi, align 4
+  %loaded = load i32, i32* %phi, align 4
   ret i32 %loaded
 }
 
@@ -241,7 +241,7 @@ define i32 @test9(i32 %b, i32* %ptr) {
 ; CHECK-LABEL: @test9(
 ; CHECK-NOT: alloca
 ; CHECK-NOT: load
-; CHECK: %[[value:.*]] = load i32* %ptr
+; CHECK: %[[value:.*]] = load i32, i32* %ptr
 ; CHECK-NOT: load
 ; CHECK: %[[result:.*]] = select i1 %{{.*}}, i32 undef, i32 %[[value]]
 ; CHECK-NEXT: ret i32 %[[result]]
@@ -252,7 +252,7 @@ entry:
   %test = icmp ne i32 %b, 0
   %bitcast = bitcast float* %f to i32*
   %select = select i1 %test, i32* %bitcast, i32* %ptr
-  %loaded = load i32* %select, align 4
+  %loaded = load i32, i32* %select, align 4
   ret i32 %loaded
 }
 
@@ -262,9 +262,9 @@ define float @test10(i32 %b, float* %ptr) {
 ; node.
 ; CHECK-LABEL: @test10(
 ; CHECK: %[[alloca:.*]] = alloca
-; CHECK: %[[argvalue:.*]] = load float* %ptr
+; CHECK: %[[argvalue:.*]] = load float, float* %ptr
 ; CHECK: %[[cast:.*]] = bitcast double* %[[alloca]] to float*
-; CHECK: %[[allocavalue:.*]] = load float* %[[cast]]
+; CHECK: %[[allocavalue:.*]] = load float, float* %[[cast]]
 ; CHECK: %[[result:.*]] = phi float [ %[[allocavalue]], %else ], [ %[[argvalue]], %then ]
 ; CHECK-NEXT: ret float %[[result]]
 
@@ -283,7 +283,7 @@ else:
 
 exit:
   %phi = phi float* [ %bitcast, %else ], [ %ptr, %then ]
-  %loaded = load float* %phi, align 4
+  %loaded = load float, float* %phi, align 4
   ret float %loaded
 }
 
@@ -292,8 +292,8 @@ define float @test11(i32 %b, float* %ptr) {
 ; CHECK-LABEL: @test11(
 ; CHECK: %[[alloca:.*]] = alloca
 ; CHECK: %[[cast:.*]] = bitcast double* %[[alloca]] to float*
-; CHECK: %[[allocavalue:.*]] = load float* %[[cast]]
-; CHECK: %[[argvalue:.*]] = load float* %ptr
+; CHECK: %[[allocavalue:.*]] = load float, float* %[[cast]]
+; CHECK: %[[argvalue:.*]] = load float, float* %ptr
 ; CHECK: %[[result:.*]] = select i1 %{{.*}}, float %[[allocavalue]], float %[[argvalue]]
 ; CHECK-NEXT: ret float %[[result]]
 
@@ -304,7 +304,7 @@ entry:
   %test = icmp ne i32 %b, 0
   %bitcast = bitcast double* %f to float*
   %select = select i1 %test, float* %bitcast, float* %ptr
-  %loaded = load float* %select, align 4
+  %loaded = load float, float* %select, align 4
   ret float %loaded
 }
 
@@ -320,7 +320,7 @@ entry:
   %a = alloca i32
   store i32 %x, i32* %a
   %dead = select i1 undef, i32* %a, i32* %p
-  %load = load i32* %a
+  %load = load i32, i32* %a
   ret i32 %load
 }
 
@@ -342,7 +342,7 @@ loop:
   br i1 undef, label %loop, label %exit
 
 exit:
-  %load = load i32* %a
+  %load = load i32, i32* %a
   ret i32 %load
 }
 
@@ -376,9 +376,9 @@ else:
 exit:
   %f.phi = phi i32* [ %f, %then ], [ %f.select, %else ]
   %g.phi = phi i32* [ %g, %then ], [ %ptr, %else ]
-  %f.loaded = load i32* %f.phi
+  %f.loaded = load i32, i32* %f.phi
   %g.select = select i1 %b1, i32* %g, i32* %g.phi
-  %g.loaded = load i32* %g.select
+  %g.loaded = load i32, i32* %g.select
   %result = add i32 %f.loaded, %g.loaded
   ret i32 %result
 }
@@ -456,8 +456,8 @@ if.then:
 ; CHECK: %[[ext:.*]] = zext i8 1 to i64
 
 if.end:
-  %tmp = load i64** %ptr
-  %result = load i64* %tmp
+  %tmp = load i64*, i64** %ptr
+  %result = load i64, i64* %tmp
 ; CHECK-NOT: load
 ; CHECK: %[[result:.*]] = phi i64 [ %[[ext]], %if.then ], [ 0, %entry ]
 
@@ -488,14 +488,14 @@ then:
 
 else:
   %a.raw = bitcast i64* %a to i8*
-  %a.raw.4 = getelementptr i8* %a.raw, i64 4
+  %a.raw.4 = getelementptr i8, i8* %a.raw, i64 4
   %a.raw.4.f = bitcast i8* %a.raw.4 to float*
   br label %end
 ; CHECK: %[[hi_cast:.*]] = bitcast i32 %[[hi]] to float
 
 end:
   %a.phi.f = phi float* [ %a.f, %then ], [ %a.raw.4.f, %else ]
-  %f = load float* %a.phi.f
+  %f = load float, float* %a.phi.f
   ret float %f
 ; CHECK: %[[phi:.*]] = phi float [ %[[lo_cast]], %then ], [ %[[hi_cast]], %else ]
 ; CHECK-NOT: load
@@ -516,19 +516,19 @@ entry:
   br i1 %cond, label %then, label %else
 
 then:
-  %0 = getelementptr inbounds [4 x float]* %arr, i64 0, i64 3
+  %0 = getelementptr inbounds [4 x float], [4 x float]* %arr, i64 0, i64 3
   store float 1.000000e+00, float* %0, align 4
   br label %merge
 
 else:
-  %1 = getelementptr inbounds [4 x float]* %arr, i64 0, i64 3
+  %1 = getelementptr inbounds [4 x float], [4 x float]* %arr, i64 0, i64 3
   store float 2.000000e+00, float* %1, align 4
   br label %merge
 
 merge:
   %2 = phi float* [ %0, %then ], [ %1, %else ]
   store float 0.000000e+00, float* %temp, align 4
-  %3 = load float* %2, align 4
+  %3 = load float, float* %2, align 4
   ret float %3
 }
 
@@ -546,7 +546,7 @@ entry:
   br i1 %cond, label %then, label %else
 
 then:
-  %0 = getelementptr inbounds [4 x float]* %arr, i64 0, i64 3
+  %0 = getelementptr inbounds [4 x float], [4 x float]* %arr, i64 0, i64 3
   store float 1.000000e+00, float* %0, align 4
   br label %then2
 
@@ -556,14 +556,14 @@ then2:
   br label %merge
 
 else:
-  %2 = getelementptr inbounds [4 x float]* %arr, i64 0, i64 3
+  %2 = getelementptr inbounds [4 x float], [4 x float]* %arr, i64 0, i64 3
   store float 3.000000e+00, float* %2, align 4
   br label %merge
 
 merge:
   %3 = phi float* [ %1, %then2 ], [ %2, %else ]
   store float 0.000000e+00, float* %temp, align 4
-  %4 = load float* %3, align 4
+  %4 = load float, float* %3, align 4
   ret float %4
 }
 
diff --git a/test/Transforms/SROA/ppcf128-no-fold.ll b/test/Transforms/SROA/ppcf128-no-fold.ll
new file mode 100644
index 000000000000..3f2934cbe166
--- /dev/null
+++ b/test/Transforms/SROA/ppcf128-no-fold.ll
@@ -0,0 +1,36 @@
+; RUN: opt < %s -sroa -S | FileCheck %s 
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+%struct.ld2 = type { [2 x ppc_fp128] }
+declare void @bar(i8*, [2 x i128])
+
+define void @foo(i8* %v) #0 {
+entry:
+  %v.addr = alloca i8*, align 8
+  %z = alloca %struct.ld2, align 16
+  store i8* %v, i8** %v.addr, align 8
+  %dat = getelementptr inbounds %struct.ld2, %struct.ld2* %z, i32 0, i32 0
+  %arrayidx = getelementptr inbounds [2 x ppc_fp128], [2 x ppc_fp128]* %dat, i32 0, i64 0
+  store ppc_fp128 0xM403B0000000000000000000000000000, ppc_fp128* %arrayidx, align 16
+  %dat1 = getelementptr inbounds %struct.ld2, %struct.ld2* %z, i32 0, i32 0
+  %arrayidx2 = getelementptr inbounds [2 x ppc_fp128], [2 x ppc_fp128]* %dat1, i32 0, i64 1
+  store ppc_fp128 0xM4093B400000000000000000000000000, ppc_fp128* %arrayidx2, align 16
+  %0 = load i8*, i8** %v.addr, align 8
+  %coerce.dive = getelementptr %struct.ld2, %struct.ld2* %z, i32 0, i32 0
+  %1 = bitcast [2 x ppc_fp128]* %coerce.dive to [2 x i128]*
+  %2 = load [2 x i128], [2 x i128]* %1, align 1
+  call void @bar(i8* %0, [2 x i128] %2)
+  ret void
+}
+
+; CHECK-LABEL: @foo
+; CHECK-NOT: i128 4628293042053316608
+; CHECK-NOT: i128 4653260752096854016
+; CHECK-DAG: i128 bitcast (ppc_fp128 0xM403B0000000000000000000000000000 to i128)
+; CHECK-DAG: i128 bitcast (ppc_fp128 0xM4093B400000000000000000000000000 to i128)
+; CHECK: call void @bar(i8* %v, [2 x i128]
+; CHECK: ret void
+
+attributes #0 = { nounwind }
+
diff --git a/test/Transforms/SROA/slice-order-independence.ll b/test/Transforms/SROA/slice-order-independence.ll
index 364ef85f1d14..7d57be6d81cc 100644
--- a/test/Transforms/SROA/slice-order-independence.ll
+++ b/test/Transforms/SROA/slice-order-independence.ll
@@ -13,11 +13,11 @@ define void @skipped_inttype_first({ i16*, i32 }*) {
   %2 = bitcast { i16*, i32 }* %0 to i8*
   %3 = bitcast { i16*, i32 }* %arg to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %3, i8* %2, i32 16, i32 8, i1 false)
-  %b = getelementptr inbounds { i16*, i32 }* %arg, i64 0, i32 0
+  %b = getelementptr inbounds { i16*, i32 }, { i16*, i32 }* %arg, i64 0, i32 0
   %pb0 = bitcast i16** %b to i63*
-  %b0 = load i63* %pb0
+  %b0 = load i63, i63* %pb0
   %pb1 = bitcast i16** %b to i8**
-  %b1 = load i8** %pb1
+  %b1 = load i8*, i8** %pb1
   ret void
 }
 
@@ -28,10 +28,10 @@ define void @skipped_inttype_last({ i16*, i32 }*) {
   %2 = bitcast { i16*, i32 }* %0 to i8*
   %3 = bitcast { i16*, i32 }* %arg to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %3, i8* %2, i32 16, i32 8, i1 false)
-  %b = getelementptr inbounds { i16*, i32 }* %arg, i64 0, i32 0
+  %b = getelementptr inbounds { i16*, i32 }, { i16*, i32 }* %arg, i64 0, i32 0
   %pb1 = bitcast i16** %b to i8**
-  %b1 = load i8** %pb1
+  %b1 = load i8*, i8** %pb1
   %pb0 = bitcast i16** %b to i63*
-  %b0 = load i63* %pb0
+  %b0 = load i63, i63* %pb0
   ret void
 }
diff --git a/test/Transforms/SROA/slice-width.ll b/test/Transforms/SROA/slice-width.ll
index ff66dcccee4b..6b6ab93dd0d2 100644
--- a/test/Transforms/SROA/slice-width.ll
+++ b/test/Transforms/SROA/slice-width.ll
@@ -14,14 +14,14 @@ load_i32:
 ; CHECK-LABEL: load_i32:
 ; CHECK-NOT: bitcast {{.*}} to i1
 ; CHECK-NOT: zext i1
-  %r0 = load i32* %arg
+  %r0 = load i32, i32* %arg
   br label %load_i1
 
 load_i1:
 ; CHECK-LABEL: load_i1:
 ; CHECK: bitcast {{.*}} to i1
   %p1 = bitcast i32* %arg to i1*
-  %t1 = load i1* %p1
+  %t1 = load i1, i1* %p1
   ret void
 }
 
@@ -42,16 +42,16 @@ define void @memcpy_fp80_padding() {
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %x_i8, i8* bitcast (%union.Foo* @foo_copy_source to i8*), i32 32, i32 16, i1 false)
 
   ; Access a slice of the alloca to trigger SROA.
-  %mid_p = getelementptr %union.Foo* %x, i32 0, i32 1
-  %elt = load i64* %mid_p
+  %mid_p = getelementptr %union.Foo, %union.Foo* %x, i32 0, i32 1
+  %elt = load i64, i64* %mid_p
   store i64 %elt, i64* @i64_sink
   ret void
 }
 ; CHECK-LABEL: define void @memcpy_fp80_padding
 ; CHECK: alloca x86_fp80
 ; CHECK: call void @llvm.memcpy.p0i8.p0i8.i32
-; CHECK: load i64* getelementptr inbounds (%union.Foo* @foo_copy_source, i64 0, i32 1)
-; CHECK: load i64* getelementptr inbounds (%union.Foo* @foo_copy_source, i64 0, i32 2)
+; CHECK: load i64, i64* getelementptr inbounds (%union.Foo, %union.Foo* @foo_copy_source, i64 0, i32 1)
+; CHECK: load i64, i64* getelementptr inbounds (%union.Foo, %union.Foo* @foo_copy_source, i64 0, i32 2)
 
 define void @memset_fp80_padding() {
   %x = alloca %union.Foo
@@ -61,8 +61,8 @@ define void @memset_fp80_padding() {
   call void @llvm.memset.p0i8.i32(i8* %x_i8, i8 -1, i32 32, i32 16, i1 false)
 
   ; Access a slice of the alloca to trigger SROA.
-  %mid_p = getelementptr %union.Foo* %x, i32 0, i32 1
-  %elt = load i64* %mid_p
+  %mid_p = getelementptr %union.Foo, %union.Foo* %x, i32 0, i32 1
+  %elt = load i64, i64* %mid_p
   store i64 %elt, i64* @i64_sink
   ret void
 }
@@ -89,8 +89,8 @@ entry:
 
   ; The following block does nothing; but appears to confuse SROA
   %unused1 = bitcast %S.vec3float* %tmp1 to %U.vec3float*
-  %unused2 = getelementptr inbounds %U.vec3float* %unused1, i32 0, i32 0
-  %unused3 = load <4 x float>* %unused2, align 1
+  %unused2 = getelementptr inbounds %U.vec3float, %U.vec3float* %unused1, i32 0, i32 0
+  %unused3 = load <4 x float>, <4 x float>* %unused2, align 1
 
   ; Create a second temporary and copy %tmp1 into it
   %tmp2 = alloca %S.vec3float, align 4
diff --git a/test/Transforms/SROA/vector-conversion.ll b/test/Transforms/SROA/vector-conversion.ll
index 08d796087317..91ae5be6c3d2 100644
--- a/test/Transforms/SROA/vector-conversion.ll
+++ b/test/Transforms/SROA/vector-conversion.ll
@@ -10,7 +10,7 @@ define <4 x i64> @vector_ptrtoint({<2 x i32*>, <2 x i32*>} %x) {
 ; CHECK-NOT: store
 
   %cast = bitcast {<2 x i32*>, <2 x i32*>}* %a to <4 x i64>*
-  %vec = load <4 x i64>* %cast
+  %vec = load <4 x i64>, <4 x i64>* %cast
 ; CHECK-NOT: load
 ; CHECK: ptrtoint
 
@@ -26,7 +26,7 @@ define <4 x i32*> @vector_inttoptr({<2 x i64>, <2 x i64>} %x) {
 ; CHECK-NOT: store
 
   %cast = bitcast {<2 x i64>, <2 x i64>}* %a to <4 x i32*>*
-  %vec = load <4 x i32*>* %cast
+  %vec = load <4 x i32*>, <4 x i32*>* %cast
 ; CHECK-NOT: load
 ; CHECK: inttoptr
 
@@ -42,7 +42,7 @@ define <2 x i64> @vector_ptrtointbitcast({<1 x i32*>, <1 x i32*>} %x) {
 ; CHECK-NOT: store
 
   %cast = bitcast {<1 x i32*>, <1 x i32*>}* %a to <2 x i64>*
-  %vec = load <2 x i64>* %cast
+  %vec = load <2 x i64>, <2 x i64>* %cast
 ; CHECK-NOT: load
 ; CHECK: ptrtoint
 ; CHECK: bitcast
diff --git a/test/Transforms/SROA/vector-lifetime-intrinsic.ll b/test/Transforms/SROA/vector-lifetime-intrinsic.ll
index 30c93b054ecc..37cf394382ac 100644
--- a/test/Transforms/SROA/vector-lifetime-intrinsic.ll
+++ b/test/Transforms/SROA/vector-lifetime-intrinsic.ll
@@ -18,7 +18,7 @@ bb:
   call void @llvm.lifetime.start(i64 16, i8* %tmp8)
   store <4 x float> %arg1, <4 x float>* %tmp, align 16
   %tmp17 = bitcast <4 x float>* %tmp to <3 x float>*
-  %tmp18 = load <3 x float>* %tmp17
+  %tmp18 = load <3 x float>, <3 x float>* %tmp17
   %tmp20 = bitcast <4 x float>* %tmp to i8*
   call void @llvm.lifetime.end(i64 16, i8* %tmp20)
   call void @wombat3(<3 x float> %tmp18)
diff --git a/test/Transforms/SROA/vector-promotion.ll b/test/Transforms/SROA/vector-promotion.ll
index c20c6352e099..2d9b26b5309c 100644
--- a/test/Transforms/SROA/vector-promotion.ll
+++ b/test/Transforms/SROA/vector-promotion.ll
@@ -9,18 +9,18 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: load
 ; CHECK:      extractelement <4 x i32> %x, i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %y, i32 3
@@ -40,19 +40,19 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
   %a.tmp3.cast = bitcast i32* %a.tmp3 to <2 x i32>*
-  %tmp3.vec = load <2 x i32>* %a.tmp3.cast
+  %tmp3.vec = load <2 x i32>, <2 x i32>* %a.tmp3.cast
   %tmp3 = extractelement <2 x i32> %tmp3.vec, i32 0
 ; CHECK-NOT: load
 ; CHECK:      %[[extract1:.*]] = extractelement <4 x i32> %x, i32 2
@@ -74,9 +74,9 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
@@ -84,14 +84,14 @@ entry:
   call void @llvm.memset.p0i8.i32(i8* %a.y.cast, i8 0, i32 16, i32 1, i1 false)
 ; CHECK-NOT: memset
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
   %a.tmp1.cast = bitcast i32* %a.tmp1 to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.tmp1.cast, i8 -1, i32 4, i32 1, i1 false)
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: load
 ; CHECK:      %[[insert:.*]] = insertelement <4 x i32> %x, i32 -1, i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %[[insert]], i32 2
@@ -112,9 +112,9 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
@@ -123,20 +123,20 @@ entry:
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.y.cast, i8* %z.cast, i32 16, i32 1, i1 false)
 ; CHECK-NOT: memcpy
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
   %a.tmp1.cast = bitcast i32* %a.tmp1 to i8*
-  %z.tmp1 = getelementptr inbounds <4 x i32>* %z, i64 0, i64 2
+  %z.tmp1 = getelementptr inbounds <4 x i32>, <4 x i32>* %z, i64 0, i64 2
   %z.tmp1.cast = bitcast i32* %z.tmp1 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.tmp1.cast, i8* %z.tmp1.cast, i32 4, i32 1, i1 false)
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: memcpy
-; CHECK:      %[[load:.*]] = load <4 x i32>* %z
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds <4 x i32>* %z, i64 0, i64 2
-; CHECK-NEXT: %[[element_load:.*]] = load i32* %[[gep]]
+; CHECK:      %[[load:.*]] = load <4 x i32>, <4 x i32>* %z
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds <4 x i32>, <4 x i32>* %z, i64 0, i64 2
+; CHECK-NEXT: %[[element_load:.*]] = load i32, i32* %[[gep]]
 ; CHECK-NEXT: %[[insert:.*]] = insertelement <4 x i32> %x, i32 %[[element_load]], i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %[[insert]], i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %[[load]], i32 3
@@ -159,9 +159,9 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
@@ -170,20 +170,20 @@ entry:
   call void @llvm.memcpy.p0i8.p1i8.i32(i8* %a.y.cast, i8 addrspace(1)* %z.cast, i32 16, i32 1, i1 false)
 ; CHECK-NOT: memcpy
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
   %a.tmp1.cast = bitcast i32* %a.tmp1 to i8*
-  %z.tmp1 = getelementptr inbounds <4 x i32> addrspace(1)* %z, i16 0, i16 2
+  %z.tmp1 = getelementptr inbounds <4 x i32>, <4 x i32> addrspace(1)* %z, i16 0, i16 2
   %z.tmp1.cast = bitcast i32 addrspace(1)* %z.tmp1 to i8 addrspace(1)*
   call void @llvm.memcpy.p0i8.p1i8.i32(i8* %a.tmp1.cast, i8 addrspace(1)* %z.tmp1.cast, i32 4, i32 1, i1 false)
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: memcpy
-; CHECK:      %[[load:.*]] = load <4 x i32> addrspace(1)* %z
-; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds <4 x i32> addrspace(1)* %z, i64 0, i64 2
-; CHECK-NEXT: %[[element_load:.*]] = load i32 addrspace(1)* %[[gep]]
+; CHECK:      %[[load:.*]] = load <4 x i32>, <4 x i32> addrspace(1)* %z
+; CHECK-NEXT: %[[gep:.*]] = getelementptr inbounds <4 x i32>, <4 x i32> addrspace(1)* %z, i64 0, i64 2
+; CHECK-NEXT: %[[element_load:.*]] = load i32, i32 addrspace(1)* %[[gep]]
 ; CHECK-NEXT: %[[insert:.*]] = insertelement <4 x i32> %x, i32 %[[element_load]], i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %[[insert]], i32 2
 ; CHECK-NEXT: extractelement <4 x i32> %[[load]], i32 3
@@ -205,9 +205,9 @@ entry:
 	%a = alloca [2 x <4 x i32>]
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0
   store <4 x i32> %x, <4 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1
   store <4 x i32> %y, <4 x i32>* %a.y
 ; CHECK-NOT: store
 
@@ -216,18 +216,18 @@ entry:
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.x.cast, i8* %a.y.cast, i32 16, i32 1, i1 false)
 ; CHECK-NOT: memcpy
 
-  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
+  %a.tmp1 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 0, i64 2
   %a.tmp1.cast = bitcast i32* %a.tmp1 to i8*
-  %z.tmp1 = getelementptr inbounds <4 x i32>* %z, i64 0, i64 2
+  %z.tmp1 = getelementptr inbounds <4 x i32>, <4 x i32>* %z, i64 0, i64 2
   %z.tmp1.cast = bitcast i32* %z.tmp1 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %z.tmp1.cast, i8* %a.tmp1.cast, i32 4, i32 1, i1 false)
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 3
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <4 x i32>], [2 x <4 x i32>]* %a, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: memcpy
-; CHECK:      %[[gep:.*]] = getelementptr inbounds <4 x i32>* %z, i64 0, i64 2
+; CHECK:      %[[gep:.*]] = getelementptr inbounds <4 x i32>, <4 x i32>* %z, i64 0, i64 2
 ; CHECK-NEXT: %[[extract:.*]] = extractelement <4 x i32> %y, i32 2
 ; CHECK-NEXT: store i32 %[[extract]], i32* %[[gep]]
 ; CHECK-NEXT: extractelement <4 x i32> %y, i32 2
@@ -250,14 +250,14 @@ define i64 @test6(<4 x i64> %x, <4 x i64> %y, i64 %n) {
 ; The old scalarrepl pass would wrongly drop the store to the second alloca.
 ; PR13254
   %tmp = alloca { <4 x i64>, <4 x i64> }
-  %p0 = getelementptr inbounds { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 0
+  %p0 = getelementptr inbounds { <4 x i64>, <4 x i64> }, { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 0
   store <4 x i64> %x, <4 x i64>* %p0
 ; CHECK: store <4 x i64> %x,
-  %p1 = getelementptr inbounds { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 1
+  %p1 = getelementptr inbounds { <4 x i64>, <4 x i64> }, { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 1
   store <4 x i64> %y, <4 x i64>* %p1
 ; CHECK: store <4 x i64> %y,
-  %addr = getelementptr inbounds { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 0, i64 %n
-  %res = load i64* %addr, align 4
+  %addr = getelementptr inbounds { <4 x i64>, <4 x i64> }, { <4 x i64>, <4 x i64> }* %tmp, i32 0, i32 0, i64 %n
+  %res = load i64, i64* %addr, align 4
   ret i64 %res
 }
 
@@ -267,27 +267,27 @@ entry:
   %a = alloca <4 x i32>
 ; CHECK-NOT: alloca
 
-  %a.gep0 = getelementptr <4 x i32>* %a, i32 0, i32 0
+  %a.gep0 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 0
   %a.cast0 = bitcast i32* %a.gep0 to <2 x i32>*
   store <2 x i32> <i32 0, i32 0>, <2 x i32>* %a.cast0
 ; CHECK-NOT: store
 ; CHECK:     select <4 x i1> <i1 true, i1 true, i1 false, i1 false> 
 
-  %a.gep1 = getelementptr <4 x i32>* %a, i32 0, i32 1
+  %a.gep1 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 1
   %a.cast1 = bitcast i32* %a.gep1 to <2 x i32>*
   store <2 x i32> <i32 1, i32 1>, <2 x i32>* %a.cast1
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 true, i1 true, i1 false>
 
-  %a.gep2 = getelementptr <4 x i32>* %a, i32 0, i32 2
+  %a.gep2 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 2
   %a.cast2 = bitcast i32* %a.gep2 to <2 x i32>*
   store <2 x i32> <i32 2, i32 2>, <2 x i32>* %a.cast2
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 false, i1 true, i1 true>
 
-  %a.gep3 = getelementptr <4 x i32>* %a, i32 0, i32 3
+  %a.gep3 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 3
   store i32 3, i32* %a.gep3
 ; CHECK-NEXT: insertelement <4 x i32>
 
-  %ret = load <4 x i32>* %a
+  %ret = load <4 x i32>, <4 x i32>* %a
 
   ret <4 x i32> %ret
 ; CHECK-NEXT: ret <4 x i32> 
@@ -301,20 +301,20 @@ entry:
   store <4 x i32> <i32 0, i32 1, i32 2, i32 3>, <4 x i32>* %a
 ; CHECK-NOT: store
 
-  %a.gep0 = getelementptr <4 x i32>* %a, i32 0, i32 0
+  %a.gep0 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 0
   %a.cast0 = bitcast i32* %a.gep0 to <2 x i32>*
-  %first = load <2 x i32>* %a.cast0
+  %first = load <2 x i32>, <2 x i32>* %a.cast0
 ; CHECK-NOT: load
 ; CHECK:      %[[extract1:.*]] = shufflevector <4 x i32> <i32 0, i32 1, i32 2, i32 3>, <4 x i32> undef, <2 x i32> <i32 0, i32 1>
 
-  %a.gep1 = getelementptr <4 x i32>* %a, i32 0, i32 1
+  %a.gep1 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 1
   %a.cast1 = bitcast i32* %a.gep1 to <2 x i32>*
-  %second = load <2 x i32>* %a.cast1
+  %second = load <2 x i32>, <2 x i32>* %a.cast1
 ; CHECK-NEXT: %[[extract2:.*]] = shufflevector <4 x i32> <i32 0, i32 1, i32 2, i32 3>, <4 x i32> undef, <2 x i32> <i32 1, i32 2>
 
-  %a.gep2 = getelementptr <4 x i32>* %a, i32 0, i32 2
+  %a.gep2 = getelementptr <4 x i32>, <4 x i32>* %a, i32 0, i32 2
   %a.cast2 = bitcast i32* %a.gep2 to <2 x i32>*
-  %third = load <2 x i32>* %a.cast2
+  %third = load <2 x i32>, <2 x i32>* %a.cast2
 ; CHECK-NEXT: %[[extract3:.*]] = shufflevector <4 x i32> <i32 0, i32 1, i32 2, i32 3>, <4 x i32> undef, <2 x i32> <i32 2, i32 3>
 
   %tmp = shufflevector <2 x i32> %first, <2 x i32> %second, <2 x i32> <i32 0, i32 2>
@@ -334,28 +334,28 @@ entry:
   %a = alloca <4 x float>
 ; CHECK-NOT: alloca
 
-  %a.gep0 = getelementptr <4 x float>* %a, i32 0, i32 0
+  %a.gep0 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 0
   %a.cast0 = bitcast float* %a.gep0 to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.cast0, i8 0, i32 8, i32 0, i1 false)
 ; CHECK-NOT: store
 ; CHECK: select <4 x i1> <i1 true, i1 true, i1 false, i1 false>
 
-  %a.gep1 = getelementptr <4 x float>* %a, i32 0, i32 1
+  %a.gep1 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 1
   %a.cast1 = bitcast float* %a.gep1 to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.cast1, i8 1, i32 8, i32 0, i1 false)
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 true, i1 true, i1 false>
 
-  %a.gep2 = getelementptr <4 x float>* %a, i32 0, i32 2
+  %a.gep2 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 2
   %a.cast2 = bitcast float* %a.gep2 to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.cast2, i8 3, i32 8, i32 0, i1 false)
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 false, i1 true, i1 true>
 
-  %a.gep3 = getelementptr <4 x float>* %a, i32 0, i32 3
+  %a.gep3 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 3
   %a.cast3 = bitcast float* %a.gep3 to i8*
   call void @llvm.memset.p0i8.i32(i8* %a.cast3, i8 7, i32 4, i32 0, i1 false)
 ; CHECK-NEXT: insertelement <4 x float> 
 
-  %ret = load <4 x float>* %a
+  %ret = load <4 x float>, <4 x float>* %a
 
   ret <4 x float> %ret
 ; CHECK-NEXT: ret <4 x float> 
@@ -367,35 +367,35 @@ entry:
   %a = alloca <4 x float>
 ; CHECK-NOT: alloca
 
-  %a.gep0 = getelementptr <4 x float>* %a, i32 0, i32 0
+  %a.gep0 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 0
   %a.cast0 = bitcast float* %a.gep0 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.cast0, i8* %x, i32 8, i32 0, i1 false)
 ; CHECK:      %[[xptr:.*]] = bitcast i8* %x to <2 x float>*
-; CHECK-NEXT: %[[x:.*]] = load <2 x float>* %[[xptr]]
+; CHECK-NEXT: %[[x:.*]] = load <2 x float>, <2 x float>* %[[xptr]]
 ; CHECK-NEXT: %[[expand_x:.*]] = shufflevector <2 x float> %[[x]], <2 x float> undef, <4 x i32> <i32 0, i32 1, i32 undef, i32 undef>
 ; CHECK-NEXT: select <4 x i1> <i1 true, i1 true, i1 false, i1 false>  
 
-  %a.gep1 = getelementptr <4 x float>* %a, i32 0, i32 1
+  %a.gep1 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 1
   %a.cast1 = bitcast float* %a.gep1 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.cast1, i8* %y, i32 8, i32 0, i1 false)
 ; CHECK-NEXT: %[[yptr:.*]] = bitcast i8* %y to <2 x float>*
-; CHECK-NEXT: %[[y:.*]] = load <2 x float>* %[[yptr]]
+; CHECK-NEXT: %[[y:.*]] = load <2 x float>, <2 x float>* %[[yptr]]
 ; CHECK-NEXT: %[[expand_y:.*]] = shufflevector <2 x float> %[[y]], <2 x float> undef, <4 x i32> <i32 undef, i32 0, i32 1, i32 undef>
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 true, i1 true, i1 false>
 
-  %a.gep2 = getelementptr <4 x float>* %a, i32 0, i32 2
+  %a.gep2 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 2
   %a.cast2 = bitcast float* %a.gep2 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.cast2, i8* %z, i32 8, i32 0, i1 false)
 ; CHECK-NEXT: %[[zptr:.*]] = bitcast i8* %z to <2 x float>*
-; CHECK-NEXT: %[[z:.*]] = load <2 x float>* %[[zptr]]
+; CHECK-NEXT: %[[z:.*]] = load <2 x float>, <2 x float>* %[[zptr]]
 ; CHECK-NEXT: %[[expand_z:.*]] = shufflevector <2 x float> %[[z]], <2 x float> undef, <4 x i32> <i32 undef, i32 undef, i32 0, i32 1>
 ; CHECK-NEXT: select <4 x i1> <i1 false, i1 false, i1 true, i1 true>
 
-  %a.gep3 = getelementptr <4 x float>* %a, i32 0, i32 3
+  %a.gep3 = getelementptr <4 x float>, <4 x float>* %a, i32 0, i32 3
   %a.cast3 = bitcast float* %a.gep3 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %a.cast3, i8* %f, i32 4, i32 0, i1 false)
 ; CHECK-NEXT: %[[fptr:.*]] = bitcast i8* %f to float*
-; CHECK-NEXT: %[[f:.*]] = load float* %[[fptr]]
+; CHECK-NEXT: %[[f:.*]] = load float, float* %[[fptr]]
 ; CHECK-NEXT: %[[insert_f:.*]] = insertelement <4 x float> 
 
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %out, i8* %a.cast2, i32 8, i32 0, i1 false)
@@ -403,7 +403,7 @@ entry:
 ; CHECK-NEXT: %[[extract_out:.*]] = shufflevector <4 x float> %[[insert_f]], <4 x float> undef, <2 x i32> <i32 2, i32 3>
 ; CHECK-NEXT: store <2 x float> %[[extract_out]], <2 x float>* %[[outptr]]
 
-  %ret = load <4 x float>* %a
+  %ret = load <4 x float>, <4 x float>* %a
 
   ret <4 x float> %ret
 ; CHECK-NEXT: ret <4 x float> %[[insert_f]]
@@ -419,7 +419,7 @@ entry:
 
   store <3 x i8> undef, <3 x i8>* %retval, align 4
   %cast = bitcast <3 x i8>* %retval to i32*
-  %load = load i32* %cast, align 4
+  %load = load i32, i32* %cast, align 4
   ret i32 %load
 ; CHECK: ret i32
 }
@@ -437,7 +437,7 @@ entry:
 ; CHECK-NOT: store
 
   %cast = bitcast i32* %a to <2 x i8>*
-  %vec = load <2 x i8>* %cast
+  %vec = load <2 x i8>, <2 x i8>* %cast
 ; CHECK-NOT: load
 
   ret <2 x i8> %vec
@@ -459,7 +459,7 @@ entry:
   store <2 x i8> %x, <2 x i8>* %cast
 ; CHECK-NOT: store
 
-  %int = load i32* %a
+  %int = load i32, i32* %a
 ; CHECK-NOT: load
 
   ret i32 %int
@@ -477,18 +477,18 @@ entry:
   %a.cast = bitcast [2 x i64]* %a to [2 x <2 x i32>]*
 ; CHECK-NOT: alloca
 
-  %a.x = getelementptr inbounds [2 x <2 x i32>]* %a.cast, i64 0, i64 0
+  %a.x = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* %a.cast, i64 0, i64 0
   store <2 x i32> %x, <2 x i32>* %a.x
-  %a.y = getelementptr inbounds [2 x <2 x i32>]* %a.cast, i64 0, i64 1
+  %a.y = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* %a.cast, i64 0, i64 1
   store <2 x i32> %y, <2 x i32>* %a.y
 ; CHECK-NOT: store
 
-  %a.tmp1 = getelementptr inbounds [2 x <2 x i32>]* %a.cast, i64 0, i64 0, i64 1
-  %tmp1 = load i32* %a.tmp1
-  %a.tmp2 = getelementptr inbounds [2 x <2 x i32>]* %a.cast, i64 0, i64 1, i64 1
-  %tmp2 = load i32* %a.tmp2
-  %a.tmp3 = getelementptr inbounds [2 x <2 x i32>]* %a.cast, i64 0, i64 1, i64 0
-  %tmp3 = load i32* %a.tmp3
+  %a.tmp1 = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* %a.cast, i64 0, i64 0, i64 1
+  %tmp1 = load i32, i32* %a.tmp1
+  %a.tmp2 = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* %a.cast, i64 0, i64 1, i64 1
+  %tmp2 = load i32, i32* %a.tmp2
+  %a.tmp3 = getelementptr inbounds [2 x <2 x i32>], [2 x <2 x i32>]* %a.cast, i64 0, i64 1, i64 0
+  %tmp3 = load i32, i32* %a.tmp3
 ; CHECK-NOT: load
 ; CHECK:      extractelement <2 x i32> %x, i32 1
 ; CHECK-NEXT: extractelement <2 x i32> %y, i32 1
@@ -515,9 +515,9 @@ entry:
   store <2 x i32> %x, <2 x i32>* %a.vec
 ; CHECK-NOT: store
 
-  %tmp1 = load i32* %a.i32
-  %a.tmp2 = getelementptr inbounds i32* %a.i32, i64 1
-  %tmp2 = load i32* %a.tmp2
+  %tmp1 = load i32, i32* %a.i32
+  %a.tmp2 = getelementptr inbounds i32, i32* %a.i32, i64 1
+  %tmp2 = load i32, i32* %a.tmp2
 ; CHECK-NOT: load
 ; CHECK:      extractelement <2 x i32> %x, i32 0
 ; CHECK-NEXT: extractelement <2 x i32> %x, i32 1
@@ -539,13 +539,13 @@ entry:
 ; CHECK-NOT: alloca
 
   store i32 %x, i32* %a.i32
-  %a.tmp2 = getelementptr inbounds i32* %a.i32, i64 1
+  %a.tmp2 = getelementptr inbounds i32, i32* %a.i32, i64 1
   store i32 %y, i32* %a.tmp2
 ; CHECK-NOT: store
 ; CHECK:      %[[V1:.*]] = insertelement <2 x i32> undef, i32 %x, i32 0
 ; CHECK-NEXT: %[[V2:.*]] = insertelement <2 x i32> %[[V1]], i32 %y, i32 1
 
-  %result = load <2 x i32>* %a.vec
+  %result = load <2 x i32>, <2 x i32>* %a.vec
 ; CHECK-NOT:  load
 
   ret <2 x i32> %result
@@ -564,13 +564,13 @@ entry:
 ; CHECK-NOT: alloca
 
   store <4 x i16> %x, <4 x i16>* %a.vec2
-  %a.tmp2 = getelementptr inbounds i32* %a.i32, i64 1
+  %a.tmp2 = getelementptr inbounds i32, i32* %a.i32, i64 1
   store i32 %y, i32* %a.tmp2
 ; CHECK-NOT: store
 ; CHECK:      %[[V1:.*]] = bitcast <4 x i16> %x to <2 x i32>
 ; CHECK-NEXT: %[[V2:.*]] = insertelement <2 x i32> %[[V1]], i32 %y, i32 1
 
-  %result = load <2 x i32>* %a.vec1
+  %result = load <2 x i32>, <2 x i32>* %a.vec1
 ; CHECK-NOT:  load
 
   ret <2 x i32> %result
@@ -590,7 +590,7 @@ entry:
 ; CHECK-NOT: alloca
 
   store <4 x i16> %x, <4 x i16>* %a.vec2
-  %a.tmp2 = getelementptr inbounds i32* %a.i32, i64 1
+  %a.tmp2 = getelementptr inbounds i32, i32* %a.i32, i64 1
   store i32 %y, i32* %a.tmp2
 ; CHECK-NOT: store
 ; CHECK:      %[[V1:.*]] = bitcast i32 %y to <2 x i16>
@@ -598,7 +598,7 @@ entry:
 ; CHECK-NEXT: %[[V3:.*]] = select <4 x i1> <i1 false, i1 false, i1 true, i1 true>, <4 x i16> %[[V2]], <4 x i16> %x
 ; CHECK-NEXT: %[[V4:.*]] = bitcast <4 x i16> %[[V3]] to <2 x float>
 
-  %result = load <2 x float>* %a.vec1
+  %result = load <2 x float>, <2 x float>* %a.vec1
 ; CHECK-NOT:  load
 
   ret <2 x float> %result
@@ -616,7 +616,7 @@ define <4 x float> @test12() {
 
   %cast2 = bitcast <3 x i32>* %a to <3 x float>*
   %cast3 = bitcast <3 x float>* %cast2 to <4 x float>*
-  %vec = load <4 x float>* %cast3
+  %vec = load <4 x float>, <4 x float>* %cast3
 ; CHECK-NOT: load
 
 ; CHECK:      %[[ret:.*]] = bitcast <4 x i32> undef to <4 x float>
diff --git a/test/Transforms/SROA/vectors-of-pointers.ll b/test/Transforms/SROA/vectors-of-pointers.ll
index 7e995b9e4476..ff09e9598965 100644
--- a/test/Transforms/SROA/vectors-of-pointers.ll
+++ b/test/Transforms/SROA/vectors-of-pointers.ll
@@ -20,6 +20,6 @@ if.then.i.i.i.i.i237:
   unreachable
 
 bb0.exit257:
-  %0 = load <2 x i32*>* %Args.i, align 16
+  %0 = load <2 x i32*>, <2 x i32*>* %Args.i, align 16
   unreachable
 }
diff --git a/test/Transforms/SampleProfile/Inputs/entry_counts.prof b/test/Transforms/SampleProfile/Inputs/entry_counts.prof
new file mode 100644
index 000000000000..5c2172b5a4d3
--- /dev/null
+++ b/test/Transforms/SampleProfile/Inputs/entry_counts.prof
@@ -0,0 +1,3 @@
+empty:100:13293
+0: 0
+1: 100
diff --git a/test/Transforms/SampleProfile/branch.ll b/test/Transforms/SampleProfile/branch.ll
index 6391fc5158cf..25bd455a044c 100644
--- a/test/Transforms/SampleProfile/branch.ll
+++ b/test/Transforms/SampleProfile/branch.ll
@@ -32,22 +32,22 @@ define i32 @main(i32 %argc, i8** nocapture readonly %argv) #0 {
 ; CHECK: Printing analysis 'Branch Probability Analysis' for function 'main':
 
 entry:
-  tail call void @llvm.dbg.value(metadata i32 %argc, i64 0, metadata !13, metadata !{}), !dbg !27
-  tail call void @llvm.dbg.value(metadata i8** %argv, i64 0, metadata !14, metadata !{}), !dbg !27
+  tail call void @llvm.dbg.value(metadata i32 %argc, i64 0, metadata !13, metadata !DIExpression()), !dbg !27
+  tail call void @llvm.dbg.value(metadata i8** %argv, i64 0, metadata !14, metadata !DIExpression()), !dbg !27
   %cmp = icmp slt i32 %argc, 2, !dbg !28
   br i1 %cmp, label %return, label %if.end, !dbg !28
-; CHECK: edge entry -> return probability is 1 / 2 = 50%
-; CHECK: edge entry -> if.end probability is 1 / 2 = 50%
+; CHECK: edge entry -> return probability is 0 / 1 = 0%
+; CHECK: edge entry -> if.end probability is 1 / 1 = 100%
 
 if.end:                                           ; preds = %entry
-  %arrayidx = getelementptr inbounds i8** %argv, i64 1, !dbg !30
-  %0 = load i8** %arrayidx, align 8, !dbg !30, !tbaa !31
+  %arrayidx = getelementptr inbounds i8*, i8** %argv, i64 1, !dbg !30
+  %0 = load i8*, i8** %arrayidx, align 8, !dbg !30, !tbaa !31
   %call = tail call i32 @atoi(i8* %0) #4, !dbg !30
-  tail call void @llvm.dbg.value(metadata i32 %call, i64 0, metadata !17, metadata !{}), !dbg !30
+  tail call void @llvm.dbg.value(metadata i32 %call, i64 0, metadata !17, metadata !DIExpression()), !dbg !30
   %cmp1 = icmp sgt i32 %call, 100, !dbg !35
   br i1 %cmp1, label %for.body, label %if.end6, !dbg !35
-; CHECK: edge if.end -> for.body probability is 1 / 2 = 50%
-; CHECK: edge if.end -> if.end6 probability is 1 / 2 = 50%
+; CHECK: edge if.end -> for.body probability is 0 / 1 = 0%
+; CHECK: edge if.end -> if.end6 probability is 1 / 1 = 100%
 
 for.body:                                         ; preds = %if.end, %for.body
   %u.016 = phi i32 [ %inc, %for.body ], [ 0, %if.end ]
@@ -55,22 +55,22 @@ for.body:                                         ; preds = %if.end, %for.body
   %add = fadd double %s.015, 3.049000e+00, !dbg !36
   %conv = sitofp i32 %u.016 to double, !dbg !36
   %add4 = fadd double %add, %conv, !dbg !36
-  tail call void @llvm.dbg.value(metadata double %add4, i64 0, metadata !18, metadata !{}), !dbg !36
+  tail call void @llvm.dbg.value(metadata double %add4, i64 0, metadata !18, metadata !DIExpression()), !dbg !36
   %div = fdiv double 3.940000e+00, %s.015, !dbg !37
   %mul = fmul double %div, 3.200000e-01, !dbg !37
   %add5 = fadd double %add4, %mul, !dbg !37
   %sub = fsub double %add4, %add5, !dbg !37
-  tail call void @llvm.dbg.value(metadata double %sub, i64 0, metadata !18, metadata !{}), !dbg !37
+  tail call void @llvm.dbg.value(metadata double %sub, i64 0, metadata !18, metadata !DIExpression()), !dbg !37
   %inc = add nsw i32 %u.016, 1, !dbg !38
-  tail call void @llvm.dbg.value(metadata i32 %inc, i64 0, metadata !21, metadata !{}), !dbg !38
+  tail call void @llvm.dbg.value(metadata i32 %inc, i64 0, metadata !21, metadata !DIExpression()), !dbg !38
   %exitcond = icmp eq i32 %inc, %call, !dbg !38
   br i1 %exitcond, label %if.end6, label %for.body, !dbg !38
-; CHECK: edge for.body -> if.end6 probability is 1 / 10227 = 0.00977804
-; CHECK: edge for.body -> for.body probability is 10226 / 10227 = 99.9902% [HOT edge]
+; CHECK: edge for.body -> if.end6 probability is 0 / 10226 = 0%
+; CHECK: edge for.body -> for.body probability is 10226 / 10226 = 100% [HOT edge]
 
 if.end6:                                          ; preds = %for.body, %if.end
   %result.0 = phi double [ 0.000000e+00, %if.end ], [ %sub, %for.body ]
-  %call7 = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([15 x i8]* @.str, i64 0, i64 0), double %result.0), !dbg !39
+  %call7 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([15 x i8], [15 x i8]* @.str, i64 0, i64 0), double %result.0), !dbg !39
   br label %return, !dbg !40
 ; CHECK: edge if.end6 -> return probability is 16 / 16 = 100% [HOT edge]
 
@@ -98,46 +98,46 @@ attributes #4 = { nounwind readonly }
 !llvm.module.flags = !{!25, !42}
 !llvm.ident = !{!26}
 
-!0 = !{!"0x11\004\00clang version 3.4 (trunk 192896) (llvm/trunk 192895)\001\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./branch.cc] [DW_LANG_C_plus_plus]
-!1 = !{!"branch.cc", !"."}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.4 (trunk 192896) (llvm/trunk 192895)", isOptimized: true, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "branch.cc", directory: ".")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00main\00main\00\004\000\001\000\006\00256\001\004", !1, !5, !6, null, i32 (i32, i8**)* @main, null, null, !12} ; [ DW_TAG_subprogram ] [line 4] [def] [main]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./branch.cc]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "main", line: 4, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 4, file: !1, scope: !5, type: !6, function: i32 (i32, i8**)* @main, variables: !12)
+!5 = !DIFile(filename: "branch.cc", directory: ".")
+!6 = !DISubroutineType(types: !7)
 !7 = !{!8, !8, !9}
-!8 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!9 = !{!"0xf\00\000\0064\0064\000\000", null, null, !10} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from ]
-!10 = !{!"0xf\00\000\0064\0064\000\000", null, null, !11} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from char]
-!11 = !{!"0x24\00char\000\008\008\000\000\006", null, null} ; [ DW_TAG_base_type ] [char] [line 0, size 8, align 8, offset 0, enc DW_ATE_signed_char]
+!8 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!9 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !10)
+!10 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !11)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "char", size: 8, align: 8, encoding: DW_ATE_signed_char)
 !12 = !{!13, !14, !15, !17, !18, !21, !23}
-!13 = !{!"0x101\00argc\0016777220\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [argc] [line 4]
-!14 = !{!"0x101\00argv\0033554436\000", !4, !5, !9} ; [ DW_TAG_arg_variable ] [argv] [line 4]
-!15 = !{!"0x100\00result\007\000", !4, !5, !16} ; [ DW_TAG_auto_variable ] [result] [line 7]
-!16 = !{!"0x24\00double\000\0064\0064\000\000\004", null, null} ; [ DW_TAG_base_type ] [double] [line 0, size 64, align 64, offset 0, enc DW_ATE_float]
-!17 = !{!"0x100\00limit\008\000", !4, !5, !8} ; [ DW_TAG_auto_variable ] [limit] [line 8]
-!18 = !{!"0x100\00s\0010\000", !19, !5, !16} ; [ DW_TAG_auto_variable ] [s] [line 10]
-!19 = !{!"0xb\009\000\000", !1, !20} ; [ DW_TAG_lexical_block ] [./branch.cc]
-!20 = !{!"0xb\009\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [./branch.cc]
-!21 = !{!"0x100\00u\0011\000", !22, !5, !8} ; [ DW_TAG_auto_variable ] [u] [line 11]
-!22 = !{!"0xb\0011\000\000", !1, !19} ; [ DW_TAG_lexical_block ] [./branch.cc]
-!23 = !{!"0x100\00x\0012\000", !24, !5, !16} ; [ DW_TAG_auto_variable ] [x] [line 12]
-!24 = !{!"0xb\0011\000\000", !1, !22} ; [ DW_TAG_lexical_block ] [./branch.cc]
+!13 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "argc", line: 4, arg: 1, scope: !4, file: !5, type: !8)
+!14 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "argv", line: 4, arg: 2, scope: !4, file: !5, type: !9)
+!15 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "result", line: 7, scope: !4, file: !5, type: !16)
+!16 = !DIBasicType(tag: DW_TAG_base_type, name: "double", size: 64, align: 64, encoding: DW_ATE_float)
+!17 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "limit", line: 8, scope: !4, file: !5, type: !8)
+!18 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "s", line: 10, scope: !19, file: !5, type: !16)
+!19 = distinct !DILexicalBlock(line: 9, column: 0, file: !1, scope: !20)
+!20 = distinct !DILexicalBlock(line: 9, column: 0, file: !1, scope: !4)
+!21 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "u", line: 11, scope: !22, file: !5, type: !8)
+!22 = distinct !DILexicalBlock(line: 11, column: 0, file: !1, scope: !19)
+!23 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "x", line: 12, scope: !24, file: !5, type: !16)
+!24 = distinct !DILexicalBlock(line: 11, column: 0, file: !1, scope: !22)
 !25 = !{i32 2, !"Dwarf Version", i32 4}
 !26 = !{!"clang version 3.4 (trunk 192896) (llvm/trunk 192895)"}
-!27 = !MDLocation(line: 4, scope: !4)
-!28 = !MDLocation(line: 5, scope: !29)
-!29 = !{!"0xb\005\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [./branch.cc]
-!30 = !MDLocation(line: 8, scope: !4)
+!27 = !DILocation(line: 4, scope: !4)
+!28 = !DILocation(line: 5, scope: !29)
+!29 = distinct !DILexicalBlock(line: 5, column: 0, file: !1, scope: !4)
+!30 = !DILocation(line: 8, scope: !4)
 !31 = !{!32, !32, i64 0}
 !32 = !{!"any pointer", !33, i64 0}
 !33 = !{!"omnipotent char", !34, i64 0}
 !34 = !{!"Simple C/C++ TBAA"}
-!35 = !MDLocation(line: 9, scope: !20)
-!36 = !MDLocation(line: 13, scope: !24)
-!37 = !MDLocation(line: 14, scope: !24)
-!38 = !MDLocation(line: 11, scope: !22)
-!39 = !MDLocation(line: 20, scope: !4)
-!40 = !MDLocation(line: 21, scope: !4)
-!41 = !MDLocation(line: 22, scope: !4)
-!42 = !{i32 1, !"Debug Info Version", i32 2}
+!35 = !DILocation(line: 9, scope: !20)
+!36 = !DILocation(line: 13, scope: !24)
+!37 = !DILocation(line: 14, scope: !24)
+!38 = !DILocation(line: 11, scope: !22)
+!39 = !DILocation(line: 20, scope: !4)
+!40 = !DILocation(line: 21, scope: !4)
+!41 = !DILocation(line: 22, scope: !4)
+!42 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/SampleProfile/calls.ll b/test/Transforms/SampleProfile/calls.ll
index d56660937bc0..10f43a1a6126 100644
--- a/test/Transforms/SampleProfile/calls.ll
+++ b/test/Transforms/SampleProfile/calls.ll
@@ -30,8 +30,8 @@ entry:
   %y.addr = alloca i32, align 4
   store i32 %x, i32* %x.addr, align 4
   store i32 %y, i32* %y.addr, align 4
-  %0 = load i32* %x.addr, align 4, !dbg !11
-  %1 = load i32* %y.addr, align 4, !dbg !11
+  %0 = load i32, i32* %x.addr, align 4, !dbg !11
+  %1 = load i32, i32* %y.addr, align 4, !dbg !11
   %add = add nsw i32 %0, %1, !dbg !11
   ret i32 %add, !dbg !11
 }
@@ -47,29 +47,29 @@ entry:
   br label %while.cond, !dbg !13
 
 while.cond:                                       ; preds = %if.end, %entry
-  %0 = load i32* %i, align 4, !dbg !14
+  %0 = load i32, i32* %i, align 4, !dbg !14
   %inc = add nsw i32 %0, 1, !dbg !14
   store i32 %inc, i32* %i, align 4, !dbg !14
   %cmp = icmp slt i32 %0, 400000000, !dbg !14
   br i1 %cmp, label %while.body, label %while.end, !dbg !14
-; CHECK: edge while.cond -> while.body probability is 5391 / 5392 = 99.9815% [HOT edge]
-; CHECK: edge while.cond -> while.end probability is 1 / 5392 = 0.018546%
+; CHECK: edge while.cond -> while.body probability is 5391 / 5391 = 100% [HOT edge]
+; CHECK: edge while.cond -> while.end probability is 0 / 5391 = 0%
 
 while.body:                                       ; preds = %while.cond
-  %1 = load i32* %i, align 4, !dbg !16
+  %1 = load i32, i32* %i, align 4, !dbg !16
   %cmp1 = icmp ne i32 %1, 100, !dbg !16
   br i1 %cmp1, label %if.then, label %if.else, !dbg !16
 ; Without discriminator information, the profiler used to think that
 ; both branches out of while.body had the same weight. In reality,
 ; the edge while.body->if.then is taken most of the time.
 ;
-; CHECK: edge while.body -> if.then probability is 5752 / 5753 = 99.9826% [HOT edge]
-; CHECK: edge while.body -> if.else probability is 1 / 5753 = 0.0173822%
+; CHECK: edge while.body -> if.then probability is 5752 / 5752 = 100% [HOT edge]
+; CHECK: edge while.body -> if.else probability is 0 / 5752 = 0%
 
 
 if.then:                                          ; preds = %while.body
-  %2 = load i32* %i, align 4, !dbg !18
-  %3 = load i32* %s, align 4, !dbg !18
+  %2 = load i32, i32* %i, align 4, !dbg !18
+  %3 = load i32, i32* %s, align 4, !dbg !18
   %call = call i32 @_Z3sumii(i32 %2, i32 %3), !dbg !18
   store i32 %call, i32* %s, align 4, !dbg !18
   br label %if.end, !dbg !18
@@ -82,8 +82,8 @@ if.end:                                           ; preds = %if.else, %if.then
   br label %while.cond, !dbg !22
 
 while.end:                                        ; preds = %while.cond
-  %4 = load i32* %s, align 4, !dbg !24
-  %call2 = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([11 x i8]* @.str, i32 0, i32 0), i32 %4), !dbg !24
+  %4 = load i32, i32* %s, align 4, !dbg !24
+  %call2 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([11 x i8], [11 x i8]* @.str, i32 0, i32 0), i32 %4), !dbg !24
   ret i32 0, !dbg !25
 }
 
@@ -92,29 +92,29 @@ declare i32 @printf(i8*, ...) #2
 !llvm.module.flags = !{!8, !9}
 !llvm.ident = !{!10}
 
-!0 = !{!"0x11\004\00clang version 3.5 \000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [./calls.cc] [DW_LANG_C_plus_plus]
-!1 = !{!"calls.cc", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5 ", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "calls.cc", directory: ".")
 !2 = !{}
 !3 = !{!4, !7}
-!4 = !{!"0x2e\00sum\00sum\00\003\000\001\000\006\00256\000\003", !1, !5, !6, null, i32 (i32, i32)* @_Z3sumii, null, null, !2} ; [ DW_TAG_subprogram ] [line 3] [def] [sum]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [./calls.cc]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!7 = !{!"0x2e\00main\00main\00\007\000\001\000\006\00256\000\007", !1, !5, !6, null, i32 ()* @main, null, null, !2} ; [ DW_TAG_subprogram ] [line 7] [def] [main]
+!4 = !DISubprogram(name: "sum", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 3, file: !1, scope: !5, type: !6, function: i32 (i32, i32)* @_Z3sumii, variables: !2)
+!5 = !DIFile(filename: "calls.cc", directory: ".")
+!6 = !DISubroutineType(types: !2)
+!7 = !DISubprogram(name: "main", line: 7, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 7, file: !1, scope: !5, type: !6, function: i32 ()* @main, variables: !2)
 !8 = !{i32 2, !"Dwarf Version", i32 4}
-!9 = !{i32 1, !"Debug Info Version", i32 2}
+!9 = !{i32 1, !"Debug Info Version", i32 3}
 !10 = !{!"clang version 3.5 "}
-!11 = !MDLocation(line: 4, scope: !4)
-!12 = !MDLocation(line: 8, scope: !7)
-!13 = !MDLocation(line: 9, scope: !7)
-!14 = !MDLocation(line: 9, scope: !15)
-!15 = !{!"0xb\001", !1, !7} ; [ DW_TAG_lexical_block ] [./calls.cc]
-!16 = !MDLocation(line: 10, scope: !17)
-!17 = !{!"0xb\0010\000\000", !1, !7} ; [ DW_TAG_lexical_block ] [./calls.cc]
-!18 = !MDLocation(line: 10, scope: !19)
-!19 = !{!"0xb\001", !1, !17} ; [ DW_TAG_lexical_block ] [./calls.cc]
-!20 = !MDLocation(line: 10, scope: !21)
-!21 = !{!"0xb\002", !1, !17} ; [ DW_TAG_lexical_block ] [./calls.cc]
-!22 = !MDLocation(line: 10, scope: !23)
-!23 = !{!"0xb\003", !1, !17} ; [ DW_TAG_lexical_block ] [./calls.cc]
-!24 = !MDLocation(line: 11, scope: !7)
-!25 = !MDLocation(line: 12, scope: !7)
+!11 = !DILocation(line: 4, scope: !4)
+!12 = !DILocation(line: 8, scope: !7)
+!13 = !DILocation(line: 9, scope: !7)
+!14 = !DILocation(line: 9, scope: !15)
+!15 = !DILexicalBlockFile(discriminator: 1, file: !1, scope: !7)
+!16 = !DILocation(line: 10, scope: !17)
+!17 = distinct !DILexicalBlock(line: 10, column: 0, file: !1, scope: !7)
+!18 = !DILocation(line: 10, scope: !19)
+!19 = !DILexicalBlockFile(discriminator: 1, file: !1, scope: !17)
+!20 = !DILocation(line: 10, scope: !21)
+!21 = !DILexicalBlockFile(discriminator: 2, file: !1, scope: !17)
+!22 = !DILocation(line: 10, scope: !23)
+!23 = !DILexicalBlockFile(discriminator: 3, file: !1, scope: !17)
+!24 = !DILocation(line: 11, scope: !7)
+!25 = !DILocation(line: 12, scope: !7)
diff --git a/test/Transforms/SampleProfile/discriminator.ll b/test/Transforms/SampleProfile/discriminator.ll
index cafc69d14923..2445c5c5d609 100644
--- a/test/Transforms/SampleProfile/discriminator.ll
+++ b/test/Transforms/SampleProfile/discriminator.ll
@@ -31,33 +31,33 @@ entry:
   br label %while.cond, !dbg !11
 
 while.cond:                                       ; preds = %if.end, %entry
-  %0 = load i32* %i.addr, align 4, !dbg !12
+  %0 = load i32, i32* %i.addr, align 4, !dbg !12
   %cmp = icmp slt i32 %0, 100, !dbg !12
   br i1 %cmp, label %while.body, label %while.end, !dbg !12
 ; CHECK: edge while.cond -> while.body probability is 100 / 101 = 99.0099% [HOT edge]
 ; CHECK: edge while.cond -> while.end probability is 1 / 101 = 0.990099%
 
 while.body:                                       ; preds = %while.cond
-  %1 = load i32* %i.addr, align 4, !dbg !14
+  %1 = load i32, i32* %i.addr, align 4, !dbg !14
   %cmp1 = icmp slt i32 %1, 50, !dbg !14
   br i1 %cmp1, label %if.then, label %if.end, !dbg !14
 ; CHECK: edge while.body -> if.then probability is 5 / 100 = 5%
 ; CHECK: edge while.body -> if.end probability is 95 / 100 = 95% [HOT edge]
 
 if.then:                                          ; preds = %while.body
-  %2 = load i32* %x, align 4, !dbg !17
+  %2 = load i32, i32* %x, align 4, !dbg !17
   %dec = add nsw i32 %2, -1, !dbg !17
   store i32 %dec, i32* %x, align 4, !dbg !17
   br label %if.end, !dbg !17
 
 if.end:                                           ; preds = %if.then, %while.body
-  %3 = load i32* %i.addr, align 4, !dbg !19
+  %3 = load i32, i32* %i.addr, align 4, !dbg !19
   %inc = add nsw i32 %3, 1, !dbg !19
   store i32 %inc, i32* %i.addr, align 4, !dbg !19
   br label %while.cond, !dbg !20
 
 while.end:                                        ; preds = %while.cond
-  %4 = load i32* %x, align 4, !dbg !21
+  %4 = load i32, i32* %x, align 4, !dbg !21
   ret i32 %4, !dbg !21
 }
 
@@ -66,25 +66,25 @@ while.end:                                        ; preds = %while.cond
 !llvm.module.flags = !{!7, !8}
 !llvm.ident = !{!9}
 
-!0 = !{!"0x11\0012\00clang version 3.5 \000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [discriminator.c] [DW_LANG_C99]
-!1 = !{!"discriminator.c", !"."}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.5 ", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "discriminator.c", directory: ".")
 !2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00foo\00foo\00\001\000\001\000\006\00256\000\001", !1, !5, !6, null, i32 (i32)* @foo, null, null, !2} ; [ DW_TAG_subprogram ] [line 1] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [discriminator.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "foo", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !1, scope: !5, type: !6, function: i32 (i32)* @foo, variables: !2)
+!5 = !DIFile(filename: "discriminator.c", directory: ".")
+!6 = !DISubroutineType(types: !2)
 !7 = !{i32 2, !"Dwarf Version", i32 4}
-!8 = !{i32 1, !"Debug Info Version", i32 2}
+!8 = !{i32 1, !"Debug Info Version", i32 3}
 !9 = !{!"clang version 3.5 "}
-!10 = !MDLocation(line: 2, scope: !4)
-!11 = !MDLocation(line: 3, scope: !4)
-!12 = !MDLocation(line: 3, scope: !13)
-!13 = !{!"0xb\001", !1, !4} ; [ DW_TAG_lexical_block ] [discriminator.c]
-!14 = !MDLocation(line: 4, scope: !15)
-!15 = !{!"0xb\004\000\001", !1, !16} ; [ DW_TAG_lexical_block ] [discriminator.c]
-!16 = !{!"0xb\003\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [discriminator.c]
-!17 = !MDLocation(line: 4, scope: !18)
-!18 = !{!"0xb\001", !1, !15} ; [ DW_TAG_lexical_block ] [discriminator.c]
-!19 = !MDLocation(line: 5, scope: !16)
-!20 = !MDLocation(line: 6, scope: !16)
-!21 = !MDLocation(line: 7, scope: !4)
+!10 = !DILocation(line: 2, scope: !4)
+!11 = !DILocation(line: 3, scope: !4)
+!12 = !DILocation(line: 3, scope: !13)
+!13 = !DILexicalBlockFile(discriminator: 1, file: !1, scope: !4)
+!14 = !DILocation(line: 4, scope: !15)
+!15 = distinct !DILexicalBlock(line: 4, column: 0, file: !1, scope: !16)
+!16 = distinct !DILexicalBlock(line: 3, column: 0, file: !1, scope: !4)
+!17 = !DILocation(line: 4, scope: !18)
+!18 = !DILexicalBlockFile(discriminator: 1, file: !1, scope: !15)
+!19 = !DILocation(line: 5, scope: !16)
+!20 = !DILocation(line: 6, scope: !16)
+!21 = !DILocation(line: 7, scope: !4)
diff --git a/test/Transforms/SampleProfile/entry_counts.ll b/test/Transforms/SampleProfile/entry_counts.ll
new file mode 100644
index 000000000000..bf66e693a9d1
--- /dev/null
+++ b/test/Transforms/SampleProfile/entry_counts.ll
@@ -0,0 +1,24 @@
+; RUN: opt < %s -sample-profile -sample-profile-file=%S/Inputs/entry_counts.prof -S | FileCheck %s
+
+; According to the profile, function empty() was called 13,293 times.
+; CHECK: {{.*}} = !{!"function_entry_count", i64 13293}
+
+define void @empty() {
+entry:
+  ret void, !dbg !9
+}
+
+!llvm.dbg.cu = !{!0}
+!llvm.module.flags = !{!6, !7}
+!llvm.ident = !{!8}
+
+!0 = !DICompileUnit(language: DW_LANG_C99, file: !1, producer: "clang version 3.7.0 (trunk 237249) (llvm/trunk 237261)", isOptimized: false, runtimeVersion: 0, emissionKind: 2, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "entry_counts.c", directory: "/usr/local/google/home/dnovillo/llvm/test/pgo")
+!2 = !{}
+!3 = !{!4}
+!4 = !DISubprogram(name: "empty", scope: !1, file: !1, line: 1, type: !5, isLocal: false, isDefinition: true, scopeLine: 1, isOptimized: false, function: void ()* @empty, variables: !2)
+!5 = !DISubroutineType(types: !2)
+!6 = !{i32 2, !"Dwarf Version", i32 4}
+!7 = !{i32 2, !"Debug Info Version", i32 3}
+!8 = !{!"clang version 3.7.0 (trunk 237249) (llvm/trunk 237261)"}
+!9 = !DILocation(line: 1, column: 15, scope: !4)
diff --git a/test/Transforms/SampleProfile/fnptr.ll b/test/Transforms/SampleProfile/fnptr.ll
index 096033bd35fc..368da42fc8a1 100644
--- a/test/Transforms/SampleProfile/fnptr.ll
+++ b/test/Transforms/SampleProfile/fnptr.ll
@@ -114,7 +114,7 @@ for.inc12:                                        ; preds = %for.inc
 
 for.end14:                                        ; preds = %for.inc12
   %S.2.lcssa.lcssa = phi double [ %S.2.lcssa, %for.inc12 ]
-  %call15 = tail call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([9 x i8]* @.str, i64 0, i64 0), double %S.2.lcssa.lcssa), !dbg !24
+  %call15 = tail call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([9 x i8], [9 x i8]* @.str, i64 0, i64 0), double %S.2.lcssa.lcssa), !dbg !24
   ret i32 0, !dbg !25
 }
 
@@ -127,29 +127,29 @@ declare i32 @printf(i8* nocapture readonly, ...) #1
 !llvm.module.flags = !{!0}
 !llvm.ident = !{!1}
 
-!0 = !{i32 2, !"Debug Info Version", i32 2}
+!0 = !{i32 2, !"Debug Info Version", i32 3}
 !1 = !{!"clang version 3.6.0 "}
-!2 = !MDLocation(line: 9, column: 3, scope: !3)
-!3 = !{!"0x2e\00foo\00foo\00\008\000\001\000\000\00256\001\008", !4, !5, !6, null, double (i32)* @_Z3fooi, null, null, !7} ; [ DW_TAG_subprogram ] [line 8] [def] [foo]
-!4 = !{!"fnptr.cc", !"."}
-!5 = !{!"0x29", !4}    ; [ DW_TAG_file_type ] [./fnptr.cc]
-!6 = !{!"0x15\00\000\000\000\000\000\000", null, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!2 = !DILocation(line: 9, column: 3, scope: !3)
+!3 = !DISubprogram(name: "foo", line: 8, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 8, file: !4, scope: !5, type: !6, function: double (i32)* @_Z3fooi, variables: !7)
+!4 = !DIFile(filename: "fnptr.cc", directory: ".")
+!5 = !DIFile(filename: "fnptr.cc", directory: ".")
+!6 = !DISubroutineType(types: !7)
 !7 = !{}
-!8 = !MDLocation(line: 9, column: 14, scope: !3)
-!9 = !MDLocation(line: 13, column: 3, scope: !10)
-!10 = !{!"0x2e\00bar\00bar\00\0012\000\001\000\000\00256\001\0012", !4, !5, !6, null, double (i32)* @_Z3bari, null, null, !7} ; [ DW_TAG_subprogram ] [line 12] [def] [bar]
-!11 = !MDLocation(line: 13, column: 14, scope: !10)
-!12 = !MDLocation(line: 19, column: 3, scope: !13)
-!13 = !{!"0x2e\00main\00main\00\0016\000\001\000\000\00256\001\0016", !4, !5, !6, null, i32 ()* @main, null, null, !7} ; [ DW_TAG_subprogram ] [line 16] [def] [main]
-!14 = !MDLocation(line: 20, column: 5, scope: !13)
-!15 = !MDLocation(line: 21, column: 15, scope: !13)
-!16 = !MDLocation(line: 22, column: 11, scope: !13)
+!8 = !DILocation(line: 9, column: 14, scope: !3)
+!9 = !DILocation(line: 13, column: 3, scope: !10)
+!10 = !DISubprogram(name: "bar", line: 12, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 12, file: !4, scope: !5, type: !6, function: double (i32)* @_Z3bari, variables: !7)
+!11 = !DILocation(line: 13, column: 14, scope: !10)
+!12 = !DILocation(line: 19, column: 3, scope: !13)
+!13 = !DISubprogram(name: "main", line: 16, isLocal: false, isDefinition: true, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 16, file: !4, scope: !5, type: !6, function: i32 ()* @main, variables: !7)
+!14 = !DILocation(line: 20, column: 5, scope: !13)
+!15 = !DILocation(line: 21, column: 15, scope: !13)
+!16 = !DILocation(line: 22, column: 11, scope: !13)
 !17 = !{!"branch_weights", i32 534, i32 2064}
-!18 = !MDLocation(line: 23, column: 14, scope: !13)
-!19 = !MDLocation(line: 25, column: 14, scope: !13)
-!20 = !MDLocation(line: 20, column: 28, scope: !13)
+!18 = !DILocation(line: 23, column: 14, scope: !13)
+!19 = !DILocation(line: 25, column: 14, scope: !13)
+!20 = !DILocation(line: 20, column: 28, scope: !13)
 !21 = !{!"branch_weights", i32 0, i32 1075}
-!22 = !MDLocation(line: 19, column: 26, scope: !13)
+!22 = !DILocation(line: 19, column: 26, scope: !13)
 !23 = !{!"branch_weights", i32 0, i32 534}
-!24 = !MDLocation(line: 27, column: 3, scope: !13)
-!25 = !MDLocation(line: 28, column: 3, scope: !13)
+!24 = !DILocation(line: 27, column: 3, scope: !13)
+!25 = !DILocation(line: 28, column: 3, scope: !13)
diff --git a/test/Transforms/SampleProfile/propagate.ll b/test/Transforms/SampleProfile/propagate.ll
index 594645f7f2be..620d125875f7 100644
--- a/test/Transforms/SampleProfile/propagate.ll
+++ b/test/Transforms/SampleProfile/propagate.ll
@@ -51,14 +51,14 @@ entry:
   store i32 %x, i32* %x.addr, align 4
   store i32 %y, i32* %y.addr, align 4
   store i64 %N, i64* %N.addr, align 8
-  %0 = load i32* %x.addr, align 4, !dbg !11
-  %1 = load i32* %y.addr, align 4, !dbg !11
+  %0 = load i32, i32* %x.addr, align 4, !dbg !11
+  %1 = load i32, i32* %y.addr, align 4, !dbg !11
   %cmp = icmp slt i32 %0, %1, !dbg !11
   br i1 %cmp, label %if.then, label %if.else, !dbg !11
 
 if.then:                                          ; preds = %entry
-  %2 = load i32* %y.addr, align 4, !dbg !13
-  %3 = load i32* %x.addr, align 4, !dbg !13
+  %2 = load i32, i32* %y.addr, align 4, !dbg !13
+  %3 = load i32, i32* %x.addr, align 4, !dbg !13
   %sub = sub nsw i32 %2, %3, !dbg !13
   %conv = sext i32 %sub to i64, !dbg !13
   store i64 %conv, i64* %retval, !dbg !13
@@ -69,16 +69,16 @@ if.else:                                          ; preds = %entry
   br label %for.cond, !dbg !15
 
 for.cond:                                         ; preds = %for.inc16, %if.else
-  %4 = load i64* %i, align 8, !dbg !15
-  %5 = load i64* %N.addr, align 8, !dbg !15
+  %4 = load i64, i64* %i, align 8, !dbg !15
+  %5 = load i64, i64* %N.addr, align 8, !dbg !15
   %cmp1 = icmp slt i64 %4, %5, !dbg !15
   br i1 %cmp1, label %for.body, label %for.end18, !dbg !15
-; CHECK: edge for.cond -> for.body probability is 10 / 11 = 90.9091% [HOT edge]
-; CHECK: edge for.cond -> for.end18 probability is 1 / 11 = 9.09091%
+; CHECK: edge for.cond -> for.body probability is 10 / 10 = 100% [HOT edge]
+; CHECK: edge for.cond -> for.end18 probability is 0 / 10 = 0%
 
 for.body:                                         ; preds = %for.cond
-  %6 = load i64* %i, align 8, !dbg !18
-  %7 = load i64* %N.addr, align 8, !dbg !18
+  %6 = load i64, i64* %i, align 8, !dbg !18
+  %7 = load i64, i64* %N.addr, align 8, !dbg !18
   %div = sdiv i64 %7, 3, !dbg !18
   %cmp2 = icmp sgt i64 %6, %div, !dbg !18
   br i1 %cmp2, label %if.then3, label %if.end, !dbg !18
@@ -86,14 +86,14 @@ for.body:                                         ; preds = %for.cond
 ; CHECK: edge for.body -> if.end probability is 4 / 5 = 80%
 
 if.then3:                                         ; preds = %for.body
-  %8 = load i32* %x.addr, align 4, !dbg !21
+  %8 = load i32, i32* %x.addr, align 4, !dbg !21
   %dec = add nsw i32 %8, -1, !dbg !21
   store i32 %dec, i32* %x.addr, align 4, !dbg !21
   br label %if.end, !dbg !21
 
 if.end:                                           ; preds = %if.then3, %for.body
-  %9 = load i64* %i, align 8, !dbg !22
-  %10 = load i64* %N.addr, align 8, !dbg !22
+  %9 = load i64, i64* %i, align 8, !dbg !22
+  %10 = load i64, i64* %N.addr, align 8, !dbg !22
   %div4 = sdiv i64 %10, 4, !dbg !22
   %cmp5 = icmp sgt i64 %9, %div4, !dbg !22
   br i1 %cmp5, label %if.then6, label %if.else7, !dbg !22
@@ -101,10 +101,10 @@ if.end:                                           ; preds = %if.then3, %for.body
 ; CHECK: edge if.end -> if.else7 probability is 6339 / 6342 = 99.9527% [HOT edge]
 
 if.then6:                                         ; preds = %if.end
-  %11 = load i32* %y.addr, align 4, !dbg !24
+  %11 = load i32, i32* %y.addr, align 4, !dbg !24
   %inc = add nsw i32 %11, 1, !dbg !24
   store i32 %inc, i32* %y.addr, align 4, !dbg !24
-  %12 = load i32* %x.addr, align 4, !dbg !26
+  %12 = load i32, i32* %x.addr, align 4, !dbg !26
   %add = add nsw i32 %12, 3, !dbg !26
   store i32 %add, i32* %x.addr, align 4, !dbg !26
   br label %if.end15, !dbg !27
@@ -114,26 +114,26 @@ if.else7:                                         ; preds = %if.end
   br label %for.cond8, !dbg !28
 
 for.cond8:                                        ; preds = %for.inc, %if.else7
-  %13 = load i32* %j, align 4, !dbg !28
+  %13 = load i32, i32* %j, align 4, !dbg !28
   %conv9 = zext i32 %13 to i64, !dbg !28
-  %14 = load i64* %i, align 8, !dbg !28
+  %14 = load i64, i64* %i, align 8, !dbg !28
   %cmp10 = icmp slt i64 %conv9, %14, !dbg !28
   br i1 %cmp10, label %for.body11, label %for.end, !dbg !28
-; CHECK: edge for.cond8 -> for.body11 probability is 16191 / 16192 = 99.9938% [HOT edge]
-; CHECK: edge for.cond8 -> for.end probability is 1 / 16192 = 0.00617589%
+; CHECK: edge for.cond8 -> for.body11 probability is 16191 / 16191 = 100% [HOT edge]
+; CHECK: edge for.cond8 -> for.end probability is 0 / 16191 = 0%
 
 for.body11:                                       ; preds = %for.cond8
-  %15 = load i32* %j, align 4, !dbg !31
-  %16 = load i32* %x.addr, align 4, !dbg !31
+  %15 = load i32, i32* %j, align 4, !dbg !31
+  %16 = load i32, i32* %x.addr, align 4, !dbg !31
   %add12 = add i32 %16, %15, !dbg !31
   store i32 %add12, i32* %x.addr, align 4, !dbg !31
-  %17 = load i32* %y.addr, align 4, !dbg !33
+  %17 = load i32, i32* %y.addr, align 4, !dbg !33
   %sub13 = sub nsw i32 %17, 3, !dbg !33
   store i32 %sub13, i32* %y.addr, align 4, !dbg !33
   br label %for.inc, !dbg !34
 
 for.inc:                                          ; preds = %for.body11
-  %18 = load i32* %j, align 4, !dbg !28
+  %18 = load i32, i32* %j, align 4, !dbg !28
   %inc14 = add i32 %18, 1, !dbg !28
   store i32 %inc14, i32* %j, align 4, !dbg !28
   br label %for.cond8, !dbg !28
@@ -145,7 +145,7 @@ if.end15:                                         ; preds = %for.end, %if.then6
   br label %for.inc16, !dbg !35
 
 for.inc16:                                        ; preds = %if.end15
-  %19 = load i64* %i, align 8, !dbg !15
+  %19 = load i64, i64* %i, align 8, !dbg !15
   %inc17 = add nsw i64 %19, 1, !dbg !15
   store i64 %inc17, i64* %i, align 8, !dbg !15
   br label %for.cond, !dbg !15
@@ -154,15 +154,15 @@ for.end18:                                        ; preds = %for.cond
   br label %if.end19
 
 if.end19:                                         ; preds = %for.end18
-  %20 = load i32* %y.addr, align 4, !dbg !36
-  %21 = load i32* %x.addr, align 4, !dbg !36
+  %20 = load i32, i32* %y.addr, align 4, !dbg !36
+  %21 = load i32, i32* %x.addr, align 4, !dbg !36
   %mul = mul nsw i32 %20, %21, !dbg !36
   %conv20 = sext i32 %mul to i64, !dbg !36
   store i64 %conv20, i64* %retval, !dbg !36
   br label %return, !dbg !36
 
 return:                                           ; preds = %if.end19, %if.then
-  %22 = load i64* %retval, !dbg !37
+  %22 = load i64, i64* %retval, !dbg !37
   ret i64 %22, !dbg !37
 }
 
@@ -177,14 +177,14 @@ entry:
   store i32 5678, i32* %x, align 4, !dbg !38
   store i32 1234, i32* %y, align 4, !dbg !39
   store i64 999999, i64* %N, align 8, !dbg !40
-  %0 = load i32* %x, align 4, !dbg !41
-  %1 = load i32* %y, align 4, !dbg !41
-  %2 = load i64* %N, align 8, !dbg !41
-  %3 = load i32* %x, align 4, !dbg !41
-  %4 = load i32* %y, align 4, !dbg !41
-  %5 = load i64* %N, align 8, !dbg !41
+  %0 = load i32, i32* %x, align 4, !dbg !41
+  %1 = load i32, i32* %y, align 4, !dbg !41
+  %2 = load i64, i64* %N, align 8, !dbg !41
+  %3 = load i32, i32* %x, align 4, !dbg !41
+  %4 = load i32, i32* %y, align 4, !dbg !41
+  %5 = load i64, i64* %N, align 8, !dbg !41
   %call = call i64 @_Z3fooiil(i32 %3, i32 %4, i64 %5), !dbg !41
-  %call1 = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([24 x i8]* @.str, i32 0, i32 0), i32 %0, i32 %1, i64 %2, i64 %call), !dbg !41
+  %call1 = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([24 x i8], [24 x i8]* @.str, i32 0, i32 0), i32 %0, i32 %1, i64 %2, i64 %call), !dbg !41
   ret i32 0, !dbg !42
 }
 
@@ -198,46 +198,46 @@ attributes #2 = { "less-precise-fpmad"="false" "no-frame-pointer-elim"="true" "n
 !llvm.module.flags = !{!8, !9}
 !llvm.ident = !{!10}
 
-!0 = !{!"0x11\004\00clang version 3.5 \000\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [propagate.cc] [DW_LANG_C_plus_plus]
-!1 = !{!"propagate.cc", !"."}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C_plus_plus, producer: "clang version 3.5 ", isOptimized: false, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "propagate.cc", directory: ".")
+!2 = !{}
 !3 = !{!4, !7}
-!4 = !{!"0x2e\00foo\00foo\00\003\000\001\000\006\00256\000\003", !1, !5, !6, null, i64 (i32, i32, i64)* @_Z3fooiil, null, null, !2} ; [ DW_TAG_subprogram ] [line 3] [def] [foo]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [propagate.cc]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!7 = !{!"0x2e\00main\00main\00\0024\000\001\000\006\00256\000\0024", !1, !5, !6, null, i32 ()* @main, null, null, !2} ; [ DW_TAG_subprogram ] [line 24] [def] [main]
+!4 = !DISubprogram(name: "foo", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 3, file: !1, scope: !5, type: !6, function: i64 (i32, i32, i64)* @_Z3fooiil, variables: !2)
+!5 = !DIFile(filename: "propagate.cc", directory: ".")
+!6 = !DISubroutineType(types: !{null})
+!7 = !DISubprogram(name: "main", line: 24, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 24, file: !1, scope: !5, type: !6, function: i32 ()* @main, variables: !2)
 !8 = !{i32 2, !"Dwarf Version", i32 4}
-!9 = !{i32 1, !"Debug Info Version", i32 2}
+!9 = !{i32 1, !"Debug Info Version", i32 3}
 !10 = !{!"clang version 3.5 "}
-!11 = !MDLocation(line: 4, scope: !12)
-!12 = !{!"0xb\004\000\000", !1, !4} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!13 = !MDLocation(line: 5, scope: !14)
-!14 = !{!"0xb\004\000\000", !1, !12} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!15 = !MDLocation(line: 7, scope: !16)
-!16 = !{!"0xb\007\000\000", !1, !17} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!17 = !{!"0xb\006\000\000", !1, !12} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!18 = !MDLocation(line: 8, scope: !19)
-!19 = !{!"0xb\008\000\000", !1, !20} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!20 = !{!"0xb\007\000\000", !1, !16} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!21 = !MDLocation(line: 9, scope: !19)
-!22 = !MDLocation(line: 10, scope: !23)
-!23 = !{!"0xb\0010\000\000", !1, !20} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!24 = !MDLocation(line: 11, scope: !25)
-!25 = !{!"0xb\0010\000\000", !1, !23} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!26 = !MDLocation(line: 12, scope: !25)
-!27 = !MDLocation(line: 13, scope: !25)
-!28 = !MDLocation(line: 14, scope: !29)
-!29 = !{!"0xb\0014\000\000", !1, !30} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!30 = !{!"0xb\0013\000\000", !1, !23} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!31 = !MDLocation(line: 15, scope: !32)
-!32 = !{!"0xb\0014\000\000", !1, !29} ; [ DW_TAG_lexical_block ] [propagate.cc]
-!33 = !MDLocation(line: 16, scope: !32)
-!34 = !MDLocation(line: 17, scope: !32)
-!35 = !MDLocation(line: 19, scope: !20)
-!36 = !MDLocation(line: 21, scope: !4)
-!37 = !MDLocation(line: 22, scope: !4)
-!38 = !MDLocation(line: 25, scope: !7)
-!39 = !MDLocation(line: 26, scope: !7)
-!40 = !MDLocation(line: 27, scope: !7)
-!41 = !MDLocation(line: 28, scope: !7)
-!42 = !MDLocation(line: 29, scope: !7)
+!11 = !DILocation(line: 4, scope: !12)
+!12 = distinct !DILexicalBlock(line: 4, column: 0, file: !1, scope: !4)
+!13 = !DILocation(line: 5, scope: !14)
+!14 = distinct !DILexicalBlock(line: 4, column: 0, file: !1, scope: !12)
+!15 = !DILocation(line: 7, scope: !16)
+!16 = distinct !DILexicalBlock(line: 7, column: 0, file: !1, scope: !17)
+!17 = distinct !DILexicalBlock(line: 6, column: 0, file: !1, scope: !12)
+!18 = !DILocation(line: 8, scope: !19)
+!19 = distinct !DILexicalBlock(line: 8, column: 0, file: !1, scope: !20)
+!20 = distinct !DILexicalBlock(line: 7, column: 0, file: !1, scope: !16)
+!21 = !DILocation(line: 9, scope: !19)
+!22 = !DILocation(line: 10, scope: !23)
+!23 = distinct !DILexicalBlock(line: 10, column: 0, file: !1, scope: !20)
+!24 = !DILocation(line: 11, scope: !25)
+!25 = distinct !DILexicalBlock(line: 10, column: 0, file: !1, scope: !23)
+!26 = !DILocation(line: 12, scope: !25)
+!27 = !DILocation(line: 13, scope: !25)
+!28 = !DILocation(line: 14, scope: !29)
+!29 = distinct !DILexicalBlock(line: 14, column: 0, file: !1, scope: !30)
+!30 = distinct !DILexicalBlock(line: 13, column: 0, file: !1, scope: !23)
+!31 = !DILocation(line: 15, scope: !32)
+!32 = distinct !DILexicalBlock(line: 14, column: 0, file: !1, scope: !29)
+!33 = !DILocation(line: 16, scope: !32)
+!34 = !DILocation(line: 17, scope: !32)
+!35 = !DILocation(line: 19, scope: !20)
+!36 = !DILocation(line: 21, scope: !4)
+!37 = !DILocation(line: 22, scope: !4)
+!38 = !DILocation(line: 25, scope: !7)
+!39 = !DILocation(line: 26, scope: !7)
+!40 = !DILocation(line: 27, scope: !7)
+!41 = !DILocation(line: 28, scope: !7)
+!42 = !DILocation(line: 29, scope: !7)
diff --git a/test/Transforms/ScalarRepl/2003-05-29-ArrayFail.ll b/test/Transforms/ScalarRepl/2003-05-29-ArrayFail.ll
index 7116199d021f..336c0a9dfa6b 100644
--- a/test/Transforms/ScalarRepl/2003-05-29-ArrayFail.ll
+++ b/test/Transforms/ScalarRepl/2003-05-29-ArrayFail.ll
@@ -5,9 +5,9 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i32 @test() nounwind {
 	%X = alloca [4 x i32]		; <[4 x i32]*> [#uses=1]
-	%Y = getelementptr [4 x i32]* %X, i64 0, i64 0		; <i32*> [#uses=1]
+	%Y = getelementptr [4 x i32], [4 x i32]* %X, i64 0, i64 0		; <i32*> [#uses=1]
         ; Must preserve arrayness!
-	%Z = getelementptr i32* %Y, i64 1		; <i32*> [#uses=1]
-	%A = load i32* %Z		; <i32> [#uses=1]
+	%Z = getelementptr i32, i32* %Y, i64 1		; <i32*> [#uses=1]
+	%A = load i32, i32* %Z		; <i32> [#uses=1]
 	ret i32 %A
 }
diff --git a/test/Transforms/ScalarRepl/2003-09-12-IncorrectPromote.ll b/test/Transforms/ScalarRepl/2003-09-12-IncorrectPromote.ll
index 3f28cb187f86..c5ca428be4bc 100644
--- a/test/Transforms/ScalarRepl/2003-09-12-IncorrectPromote.ll
+++ b/test/Transforms/ScalarRepl/2003-09-12-IncorrectPromote.ll
@@ -4,8 +4,8 @@
 
 define i8* @test() {
 	%A = alloca [30 x i8]		; <[30 x i8]*> [#uses=1]
-	%B = getelementptr [30 x i8]* %A, i64 0, i64 0		; <i8*> [#uses=2]
-	%C = getelementptr i8* %B, i64 1		; <i8*> [#uses=1]
+	%B = getelementptr [30 x i8], [30 x i8]* %A, i64 0, i64 0		; <i8*> [#uses=2]
+	%C = getelementptr i8, i8* %B, i64 1		; <i8*> [#uses=1]
 	store i8 0, i8* %B
 	ret i8* %C
 }
diff --git a/test/Transforms/ScalarRepl/2003-10-29-ArrayProblem.ll b/test/Transforms/ScalarRepl/2003-10-29-ArrayProblem.ll
index 77c7b54b63ff..ce652737bc1e 100644
--- a/test/Transforms/ScalarRepl/2003-10-29-ArrayProblem.ll
+++ b/test/Transforms/ScalarRepl/2003-10-29-ArrayProblem.ll
@@ -7,9 +7,9 @@ declare void @.iter_2(i32 (i8*)*, i8*)
 
 define i32 @main() {
 	%d = alloca %T		; <{ [80 x i8], i32, i32 }*> [#uses=2]
-	%tmp.0 = getelementptr %T* %d, i64 0, i32 2		; <i32*> [#uses=1]
+	%tmp.0 = getelementptr %T, %T* %d, i64 0, i32 2		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp.0
-	%tmp.1 = getelementptr %T* %d, i64 0, i32 0, i64 0		; <i8*> [#uses=1]
+	%tmp.1 = getelementptr %T, %T* %d, i64 0, i32 0, i64 0		; <i8*> [#uses=1]
 	call void @.iter_2( i32 (i8*)* @.callback_1, i8* %tmp.1 )
 	ret i32 0
 }
diff --git a/test/Transforms/ScalarRepl/2006-11-07-InvalidArrayPromote.ll b/test/Transforms/ScalarRepl/2006-11-07-InvalidArrayPromote.ll
index 99c9fb9ef326..2701fdaea516 100644
--- a/test/Transforms/ScalarRepl/2006-11-07-InvalidArrayPromote.ll
+++ b/test/Transforms/ScalarRepl/2006-11-07-InvalidArrayPromote.ll
@@ -5,14 +5,14 @@ define i32 @func(<4 x float> %v0, <4 x float> %v1) nounwind {
 	%vsiidx = alloca [2 x <4 x i32>], align 16		; <[2 x <4 x i32>]*> [#uses=3]
 	%tmp = call <4 x i32> @llvm.x86.sse2.cvttps2dq( <4 x float> %v0 )		; <<4 x i32>> [#uses=2]
 	%tmp.upgrd.1 = bitcast <4 x i32> %tmp to <2 x i64>		; <<2 x i64>> [#uses=0]
-	%tmp.upgrd.2 = getelementptr [2 x <4 x i32>]* %vsiidx, i32 0, i32 0		; <<4 x i32>*> [#uses=1]
+	%tmp.upgrd.2 = getelementptr [2 x <4 x i32>], [2 x <4 x i32>]* %vsiidx, i32 0, i32 0		; <<4 x i32>*> [#uses=1]
 	store <4 x i32> %tmp, <4 x i32>* %tmp.upgrd.2
 	%tmp10 = call <4 x i32> @llvm.x86.sse2.cvttps2dq( <4 x float> %v1 )		; <<4 x i32>> [#uses=2]
 	%tmp10.upgrd.3 = bitcast <4 x i32> %tmp10 to <2 x i64>		; <<2 x i64>> [#uses=0]
-	%tmp14 = getelementptr [2 x <4 x i32>]* %vsiidx, i32 0, i32 1		; <<4 x i32>*> [#uses=1]
+	%tmp14 = getelementptr [2 x <4 x i32>], [2 x <4 x i32>]* %vsiidx, i32 0, i32 1		; <<4 x i32>*> [#uses=1]
 	store <4 x i32> %tmp10, <4 x i32>* %tmp14
-	%tmp15 = getelementptr [2 x <4 x i32>]* %vsiidx, i32 0, i32 0, i32 4		; <i32*> [#uses=1]
-	%tmp.upgrd.4 = load i32* %tmp15		; <i32> [#uses=1]
+	%tmp15 = getelementptr [2 x <4 x i32>], [2 x <4 x i32>]* %vsiidx, i32 0, i32 0, i32 4		; <i32*> [#uses=1]
+	%tmp.upgrd.4 = load i32, i32* %tmp15		; <i32> [#uses=1]
 	ret i32 %tmp.upgrd.4
 }
 
diff --git a/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll b/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
index cf96c4cc205c..966b17939fec 100644
--- a/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
+++ b/test/Transforms/ScalarRepl/2007-05-29-MemcpyPreserve.ll
@@ -11,12 +11,12 @@ target triple = "i686-apple-darwin8"
 define void @_Z4testP9UnionTypePS0_(%struct.UnionType* %p, %struct.UnionType** %pointerToUnion) {
 entry:
   %tmp = alloca %struct.UnionType, align 8
-  %tmp2 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0
-  %tmp13 = getelementptr %struct.UnionType* %p, i32 0, i32 0, i32 0
+  %tmp2 = getelementptr %struct.UnionType, %struct.UnionType* %tmp, i32 0, i32 0, i32 0
+  %tmp13 = getelementptr %struct.UnionType, %struct.UnionType* %p, i32 0, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp13, i32 8, i32 0, i1 false)
-  %tmp5 = load %struct.UnionType** %pointerToUnion
-  %tmp56 = getelementptr %struct.UnionType* %tmp5, i32 0, i32 0, i32 0
-  %tmp7 = getelementptr %struct.UnionType* %tmp, i32 0, i32 0, i32 0
+  %tmp5 = load %struct.UnionType*, %struct.UnionType** %pointerToUnion
+  %tmp56 = getelementptr %struct.UnionType, %struct.UnionType* %tmp5, i32 0, i32 0, i32 0
+  %tmp7 = getelementptr %struct.UnionType, %struct.UnionType* %tmp, i32 0, i32 0, i32 0
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp56, i8* %tmp7, i32 8, i32 0, i1 false)
   ret void
 }
diff --git a/test/Transforms/ScalarRepl/2007-11-03-bigendian_apint.ll b/test/Transforms/ScalarRepl/2007-11-03-bigendian_apint.ll
index 48abffe5f571..28f503a210ca 100644
--- a/test/Transforms/ScalarRepl/2007-11-03-bigendian_apint.ll
+++ b/test/Transforms/ScalarRepl/2007-11-03-bigendian_apint.ll
@@ -1,5 +1,11 @@
 ; RUN: opt < %s -scalarrepl -S | not grep shr
 
+; FIXME: I think this test is no longer valid.
+; It was working because SROA was aborting when
+; no datalayout was supplied
+; XFAIL: *
+
+
 %struct.S = type { i16 }
 
 define zeroext i1 @f(i16 signext  %b)   {
@@ -10,21 +16,21 @@ entry:
 	%tmp = alloca i32		; <i32*> [#uses=2]
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	store i16 %b, i16* %b_addr
-	%tmp1 = getelementptr %struct.S* %s, i32 0, i32 0		; <i16*> [#uses=1]
-	%tmp2 = load i16* %b_addr, align 2		; <i16> [#uses=1]
+	%tmp1 = getelementptr %struct.S, %struct.S* %s, i32 0, i32 0		; <i16*> [#uses=1]
+	%tmp2 = load i16, i16* %b_addr, align 2		; <i16> [#uses=1]
 	store i16 %tmp2, i16* %tmp1, align 2
-	%tmp3 = getelementptr %struct.S* %s, i32 0, i32 0		; <i16*> [#uses=1]
+	%tmp3 = getelementptr %struct.S, %struct.S* %s, i32 0, i32 0		; <i16*> [#uses=1]
 	%tmp34 = bitcast i16* %tmp3 to [2 x i1]*		; <[2 x i1]*> [#uses=1]
-	%tmp5 = getelementptr [2 x i1]* %tmp34, i32 0, i32 1		; <i1*> [#uses=1]
-	%tmp6 = load i1* %tmp5, align 1		; <i1> [#uses=1]
+	%tmp5 = getelementptr [2 x i1], [2 x i1]* %tmp34, i32 0, i32 1		; <i1*> [#uses=1]
+	%tmp6 = load i1, i1* %tmp5, align 1		; <i1> [#uses=1]
 	%tmp67 = zext i1 %tmp6 to i32		; <i32> [#uses=1]
 	store i32 %tmp67, i32* %tmp, align 4
-	%tmp8 = load i32* %tmp, align 4		; <i32> [#uses=1]
+	%tmp8 = load i32, i32* %tmp, align 4		; <i32> [#uses=1]
 	store i32 %tmp8, i32* %retval, align 4
 	br label %return
 
 return:		; preds = %entry
-	%retval9 = load i32* %retval		; <i32> [#uses=1]
+	%retval9 = load i32, i32* %retval		; <i32> [#uses=1]
 	%retval910 = trunc i32 %retval9 to i1		; <i1> [#uses=1]
 	ret i1 %retval910
 }
diff --git a/test/Transforms/ScalarRepl/2008-01-29-PromoteBug.ll b/test/Transforms/ScalarRepl/2008-01-29-PromoteBug.ll
index a53f3deadc72..99366b36442b 100644
--- a/test/Transforms/ScalarRepl/2008-01-29-PromoteBug.ll
+++ b/test/Transforms/ScalarRepl/2008-01-29-PromoteBug.ll
@@ -10,10 +10,10 @@ entry:
 	%T3 = bitcast [1 x %struct.T]* %s to i32*
 	store i32 -61184, i32* %T3
 
-	%tmp16 = getelementptr [1 x %struct.T]* %s, i32 0, i32 0		; <%struct.T*> [#uses=1]
-	%tmp17 = getelementptr %struct.T* %tmp16, i32 0, i32 1		; <[3 x i8]*> [#uses=1]
+	%tmp16 = getelementptr [1 x %struct.T], [1 x %struct.T]* %s, i32 0, i32 0		; <%struct.T*> [#uses=1]
+	%tmp17 = getelementptr %struct.T, %struct.T* %tmp16, i32 0, i32 1		; <[3 x i8]*> [#uses=1]
 	%tmp1718 = bitcast [3 x i8]* %tmp17 to i32*		; <i32*> [#uses=1]
-	%tmp19 = load i32* %tmp1718, align 4		; <i32> [#uses=1]
+	%tmp19 = load i32, i32* %tmp1718, align 4		; <i32> [#uses=1]
 	%mask = and i32 %tmp19, 16777215		; <i32> [#uses=2]
 	%mask2324 = trunc i32 %mask to i8		; <i8> [#uses=1]
 	ret i8 %mask2324
diff --git a/test/Transforms/ScalarRepl/2008-02-28-SubElementExtractCrash.ll b/test/Transforms/ScalarRepl/2008-02-28-SubElementExtractCrash.ll
index 7f8ef8310522..f37b6529a546 100644
--- a/test/Transforms/ScalarRepl/2008-02-28-SubElementExtractCrash.ll
+++ b/test/Transforms/ScalarRepl/2008-02-28-SubElementExtractCrash.ll
@@ -6,11 +6,11 @@ target triple = "i686-apple-darwin8"
 define i32 @main(i32 %argc, i8** %argv) {
 entry:
 	%c = alloca %struct..0anon		; <%struct..0anon*> [#uses=2]
-	%tmp2 = getelementptr %struct..0anon* %c, i32 0, i32 0		; <<1 x i64>*> [#uses=1]
+	%tmp2 = getelementptr %struct..0anon, %struct..0anon* %c, i32 0, i32 0		; <<1 x i64>*> [#uses=1]
 	store <1 x i64> zeroinitializer, <1 x i64>* %tmp2, align 8
-	%tmp7 = getelementptr %struct..0anon* %c, i32 0, i32 0		; <<1 x i64>*> [#uses=1]
+	%tmp7 = getelementptr %struct..0anon, %struct..0anon* %c, i32 0, i32 0		; <<1 x i64>*> [#uses=1]
 	%tmp78 = bitcast <1 x i64>* %tmp7 to [2 x i32]*		; <[2 x i32]*> [#uses=1]
-	%tmp9 = getelementptr [2 x i32]* %tmp78, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp10 = load i32* %tmp9, align 4		; <i32> [#uses=0]
+	%tmp9 = getelementptr [2 x i32], [2 x i32]* %tmp78, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp10 = load i32, i32* %tmp9, align 4		; <i32> [#uses=0]
 	unreachable
 }
diff --git a/test/Transforms/ScalarRepl/2008-06-05-loadstore-agg.ll b/test/Transforms/ScalarRepl/2008-06-05-loadstore-agg.ll
index ce70a1b13b81..d1f331211745 100644
--- a/test/Transforms/ScalarRepl/2008-06-05-loadstore-agg.ll
+++ b/test/Transforms/ScalarRepl/2008-06-05-loadstore-agg.ll
@@ -14,8 +14,8 @@ define i32 @foo() {
         ; And store it
 	store { i32, i32 } %res2, { i32, i32 }* %target
         ; Actually use %target, so it doesn't get removed altogether
-        %ptr = getelementptr { i32, i32 }* %target, i32 0, i32 0
-        %val = load i32* %ptr
+        %ptr = getelementptr { i32, i32 }, { i32, i32 }* %target, i32 0, i32 0
+        %val = load i32, i32* %ptr
 	ret i32 %val
 }
 
@@ -27,7 +27,7 @@ define i32 @bar() {
         ; And store it
 	store [ 2 x i32 ] %res2, [ 2 x i32 ]* %target
         ; Actually use %target, so it doesn't get removed altogether
-        %ptr = getelementptr [ 2 x i32 ]* %target, i32 0, i32 0
-        %val = load i32* %ptr
+        %ptr = getelementptr [ 2 x i32 ], [ 2 x i32 ]* %target, i32 0, i32 0
+        %val = load i32, i32* %ptr
 	ret i32 %val
 }
diff --git a/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll b/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
index b2a9d43b2277..c0ff25f3541c 100644
--- a/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
+++ b/test/Transforms/ScalarRepl/2008-08-22-out-of-range-array-promote.ll
@@ -13,8 +13,8 @@ entry:
   %r1 = bitcast %struct.x* %r to i8*
   %s2 = bitcast %struct.x* %s to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %r1, i8* %s2, i32 12, i32 8, i1 false)
-  %1 = getelementptr %struct.x* %r, i32 0, i32 0, i32 1
-  %2 = load i32* %1, align 4
+  %1 = getelementptr %struct.x, %struct.x* %r, i32 0, i32 0, i32 1
+  %2 = load i32, i32* %1, align 4
   ret i32 %2
 }
 
diff --git a/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll b/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
index 7554b7f2b0b8..16d9108bb24c 100644
--- a/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
+++ b/test/Transforms/ScalarRepl/2008-09-22-vector-gep.ll
@@ -12,11 +12,11 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define void @main(%struct.two* %D, i16 %V) {
 entry:
 	%S = alloca %struct.two
-        %S.2 = getelementptr %struct.two* %S, i32 0, i32 1
+        %S.2 = getelementptr %struct.two, %struct.two* %S, i32 0, i32 1
         store i16 %V, i16* %S.2
         ; This gep is effectively a bitcast to i8*, but is sometimes generated
         ; because the type of the first element in %struct.two is i8.
-	%tmpS = getelementptr %struct.two* %S, i32 0, i32 0, i32 0 
+	%tmpS = getelementptr %struct.two, %struct.two* %S, i32 0, i32 0, i32 0 
 	%tmpD = bitcast %struct.two* %D to i8*
         call void @llvm.memmove.p0i8.p0i8.i32(i8* %tmpD, i8* %tmpS, i32 4, i32 1, i1 false)
         ret void
diff --git a/test/Transforms/ScalarRepl/2009-02-02-ScalarPromoteOutOfRange.ll b/test/Transforms/ScalarRepl/2009-02-02-ScalarPromoteOutOfRange.ll
index 3c8a364edcdc..f0af1caa4610 100644
--- a/test/Transforms/ScalarRepl/2009-02-02-ScalarPromoteOutOfRange.ll
+++ b/test/Transforms/ScalarRepl/2009-02-02-ScalarPromoteOutOfRange.ll
@@ -6,11 +6,11 @@ target triple = "i386-pc-linux-gnu"
 
 define i32 @f(i32 %x, i32 %y) {
        %instance = alloca %pair
-       %first = getelementptr %pair* %instance, i32 0, i32 0
+       %first = getelementptr %pair, %pair* %instance, i32 0, i32 0
        %cast = bitcast [1 x i32]* %first to i32*
        store i32 %x, i32* %cast
-       %second = getelementptr %pair* %instance, i32 0, i32 1
+       %second = getelementptr %pair, %pair* %instance, i32 0, i32 1
        store i32 %y, i32* %second
-       %v = load i32* %cast
+       %v = load i32, i32* %cast
        ret i32 %v
 }
diff --git a/test/Transforms/ScalarRepl/2009-02-05-LoadFCA.ll b/test/Transforms/ScalarRepl/2009-02-05-LoadFCA.ll
index 67228a7a3cec..56375ffe7933 100644
--- a/test/Transforms/ScalarRepl/2009-02-05-LoadFCA.ll
+++ b/test/Transforms/ScalarRepl/2009-02-05-LoadFCA.ll
@@ -9,7 +9,7 @@ entry:
 	%tmp = alloca { i64, i64 }, align 8		; <{ i64, i64 }*> [#uses=2]
 	store { i64, i64 } %0, { i64, i64 }* %tmp
 	%1 = bitcast { i64, i64 }* %tmp to %struct.anon*		; <%struct.anon*> [#uses=1]
-	%2 = load %struct.anon* %1, align 8		; <%struct.anon> [#uses=1]
+	%2 = load %struct.anon, %struct.anon* %1, align 8		; <%struct.anon> [#uses=1]
         %tmp3 = extractvalue %struct.anon %2, 0
 	ret i32 %tmp3
 }
diff --git a/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll b/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
index a4182d4c1d7c..025578c7f44d 100644
--- a/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
+++ b/test/Transforms/ScalarRepl/2009-03-04-MemCpyAlign.ll
@@ -9,7 +9,7 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define void @f(i8* %p) nounwind {
 entry:
         %s = alloca %struct.st, align 4  ; <%struct.st*> [#uses=2]
-        %0 = getelementptr %struct.st* %s, i32 0, i32 0  ; <i16*> [#uses=1]
+        %0 = getelementptr %struct.st, %struct.st* %s, i32 0, i32 0  ; <i16*> [#uses=1]
         store i16 1, i16* %0, align 4
         %s1 = bitcast %struct.st* %s to i8*  ; <i8*> [#uses=1]
         call void @llvm.memcpy.p0i8.p0i8.i32(i8* %p, i8* %s1, i32 2, i32 1, i1 false)
diff --git a/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll b/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
index 1f921914fcb0..d1cc4244ccfc 100644
--- a/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
+++ b/test/Transforms/ScalarRepl/2009-12-11-NeonTypes.ll
@@ -25,37 +25,37 @@ entry:
   %tmp2 = alloca %struct.int16x8x2_t
   %0 = alloca %struct.int16x8x2_t
   %"alloca point" = bitcast i32 0 to i32
-  %1 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
+  %1 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %tmp_addr, i32 0, i32 0
   store <8 x i16> %tmp.0, <8 x i16>* %1
   store %struct.int16x8x2_t* %dst, %struct.int16x8x2_t** %dst_addr
-  %2 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0
-  %3 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
-  %4 = load <8 x i16>* %3, align 16
+  %2 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %__ax, i32 0, i32 0
+  %3 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %tmp_addr, i32 0, i32 0
+  %4 = load <8 x i16>, <8 x i16>* %3, align 16
   store <8 x i16> %4, <8 x i16>* %2, align 16
-  %5 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0
-  %6 = getelementptr inbounds %struct.int16x8_t* %tmp_addr, i32 0, i32 0
-  %7 = load <8 x i16>* %6, align 16
+  %5 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %__bx, i32 0, i32 0
+  %6 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %tmp_addr, i32 0, i32 0
+  %7 = load <8 x i16>, <8 x i16>* %6, align 16
   store <8 x i16> %7, <8 x i16>* %5, align 16
-  %8 = getelementptr inbounds %struct.int16x8_t* %__ax, i32 0, i32 0
-  %9 = load <8 x i16>* %8, align 16
-  %10 = getelementptr inbounds %struct.int16x8_t* %__bx, i32 0, i32 0
-  %11 = load <8 x i16>* %10, align 16
-  %12 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0
+  %8 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %__ax, i32 0, i32 0
+  %9 = load <8 x i16>, <8 x i16>* %8, align 16
+  %10 = getelementptr inbounds %struct.int16x8_t, %struct.int16x8_t* %__bx, i32 0, i32 0
+  %11 = load <8 x i16>, <8 x i16>* %10, align 16
+  %12 = getelementptr inbounds %union..0anon, %union..0anon* %__rv, i32 0, i32 0
   %13 = bitcast %struct.int16x8x2_t* %12 to %struct.__neon_int16x8x2_t*
   %14 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 0, i32 8, i32 2, i32 10, i32 4, i32 12, i32 6, i32 14>
-  %15 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 0
+  %15 = getelementptr inbounds %struct.__neon_int16x8x2_t, %struct.__neon_int16x8x2_t* %13, i32 0, i32 0
   store <8 x i16> %14, <8 x i16>* %15
   %16 = shufflevector <8 x i16> %9, <8 x i16> %11, <8 x i32> <i32 1, i32 9, i32 3, i32 11, i32 5, i32 13, i32 7, i32 15>
-  %17 = getelementptr inbounds %struct.__neon_int16x8x2_t* %13, i32 0, i32 1
+  %17 = getelementptr inbounds %struct.__neon_int16x8x2_t, %struct.__neon_int16x8x2_t* %13, i32 0, i32 1
   store <8 x i16> %16, <8 x i16>* %17
-  %18 = getelementptr inbounds %union..0anon* %__rv, i32 0, i32 0
+  %18 = getelementptr inbounds %union..0anon, %union..0anon* %__rv, i32 0, i32 0
   %19 = bitcast %struct.int16x8x2_t* %0 to i8*
   %20 = bitcast %struct.int16x8x2_t* %18 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %19, i8* %20, i32 32, i32 16, i1 false)
   %tmp21 = bitcast %struct.int16x8x2_t* %tmp2 to i8*
   %21 = bitcast %struct.int16x8x2_t* %0 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp21, i8* %21, i32 32, i32 16, i1 false)
-  %22 = load %struct.int16x8x2_t** %dst_addr, align 4
+  %22 = load %struct.int16x8x2_t*, %struct.int16x8x2_t** %dst_addr, align 4
   %23 = bitcast %struct.int16x8x2_t* %22 to i8*
   %tmp22 = bitcast %struct.int16x8x2_t* %tmp2 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %23, i8* %tmp22, i32 32, i32 16, i1 false)
diff --git a/test/Transforms/ScalarRepl/2011-05-06-CapturedAlloca.ll b/test/Transforms/ScalarRepl/2011-05-06-CapturedAlloca.ll
index 816cb60e82a4..997d03b059e9 100644
--- a/test/Transforms/ScalarRepl/2011-05-06-CapturedAlloca.ll
+++ b/test/Transforms/ScalarRepl/2011-05-06-CapturedAlloca.ll
@@ -17,7 +17,7 @@ entry:
   %tmp = bitcast [4 x i32]* %l_10 to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp, i8* bitcast ([4 x i32]* @func_1.l_10 to i8*), i64 16, i32 16, i1 false)
 ; CHECK: call void @llvm.memcpy
-  %arrayidx = getelementptr inbounds [4 x i32]* %l_10, i64 0, i64 0
+  %arrayidx = getelementptr inbounds [4 x i32], [4 x i32]* %l_10, i64 0, i64 0
   %call = call i32* @noop(i32* %arrayidx)
   store i32 0, i32* %call
   ret i32 0
diff --git a/test/Transforms/ScalarRepl/2011-06-08-VectorExtractValue.ll b/test/Transforms/ScalarRepl/2011-06-08-VectorExtractValue.ll
index 459688547fb4..dee27f8e3068 100644
--- a/test/Transforms/ScalarRepl/2011-06-08-VectorExtractValue.ll
+++ b/test/Transforms/ScalarRepl/2011-06-08-VectorExtractValue.ll
@@ -16,12 +16,12 @@ define void @main() uwtable ssp {
 entry:
   %ref.tmp2 = alloca %0, align 16
   %tmpcast = bitcast %0* %ref.tmp2 to %struct.Point_3*
-  %0 = getelementptr %0* %ref.tmp2, i64 0, i32 0
+  %0 = getelementptr %0, %0* %ref.tmp2, i64 0, i32 0
   store <2 x float> zeroinitializer, <2 x float>* %0, align 16
-  %1 = getelementptr inbounds %struct.Point_3* %tmpcast, i64 0, i32 0
-  %base.i.i.i = getelementptr inbounds %struct.PointC3* %1, i64 0, i32 0
-  %arrayidx.i.i.i.i = getelementptr inbounds %struct.array* %base.i.i.i, i64 0, i32 0, i64 0
-  %tmp5.i.i = load float* %arrayidx.i.i.i.i, align 4
+  %1 = getelementptr inbounds %struct.Point_3, %struct.Point_3* %tmpcast, i64 0, i32 0
+  %base.i.i.i = getelementptr inbounds %struct.PointC3, %struct.PointC3* %1, i64 0, i32 0
+  %arrayidx.i.i.i.i = getelementptr inbounds %struct.array, %struct.array* %base.i.i.i, i64 0, i32 0, i64 0
+  %tmp5.i.i = load float, float* %arrayidx.i.i.i.i, align 4
   ret void
 }
 
@@ -33,9 +33,9 @@ define void @test1() uwtable ssp {
 entry:
   %ref.tmp2 = alloca {<2 x float>, float}, align 16
   %tmpcast = bitcast {<2 x float>, float}* %ref.tmp2 to float*
-  %0 = getelementptr {<2 x float>, float}* %ref.tmp2, i64 0, i32 0
+  %0 = getelementptr {<2 x float>, float}, {<2 x float>, float}* %ref.tmp2, i64 0, i32 0
   store <2 x float> zeroinitializer, <2 x float>* %0, align 16
-  %tmp5.i.i = load float* %tmpcast, align 4
+  %tmp5.i.i = load float, float* %tmpcast, align 4
   ret void
 }
 
@@ -50,12 +50,12 @@ define float @test2() uwtable ssp {
 entry:
   %ref.tmp2 = alloca {<2 x float>, float}, align 16
   %tmpcast = bitcast {<2 x float>, float}* %ref.tmp2 to float*
-  %tmpcast2 = getelementptr {<2 x float>, float}* %ref.tmp2, i64 0, i32 1
-  %0 = getelementptr {<2 x float>, float}* %ref.tmp2, i64 0, i32 0
+  %tmpcast2 = getelementptr {<2 x float>, float}, {<2 x float>, float}* %ref.tmp2, i64 0, i32 1
+  %0 = getelementptr {<2 x float>, float}, {<2 x float>, float}* %ref.tmp2, i64 0, i32 0
   store <2 x float> zeroinitializer, <2 x float>* %0, align 16
   store float 1.0, float* %tmpcast2, align 4
-  %r1 = load float* %tmpcast, align 4
-  %r2 = load float* %tmpcast2, align 4
+  %r1 = load float, float* %tmpcast, align 4
+  %r2 = load float, float* %tmpcast2, align 4
   %r = fadd float %r1, %r2
   ret float %r
 }
@@ -69,7 +69,7 @@ entry:
   %ai = alloca { <2 x float>, <2 x float> }, align 8
   store { <2 x float>, <2 x float> } {<2 x float> <float 0.0, float 1.0>, <2 x float> <float 2.0, float 3.0>}, { <2 x float>, <2 x float> }* %ai, align 8
   %tmpcast = bitcast { <2 x float>, <2 x float> }* %ai to [4 x float]*
-  %arrayidx = getelementptr inbounds [4 x float]* %tmpcast, i64 0, i64 3
-  %f = load float* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds [4 x float], [4 x float]* %tmpcast, i64 0, i64 3
+  %f = load float, float* %arrayidx, align 4
   ret float %f
 }
diff --git a/test/Transforms/ScalarRepl/2011-06-17-VectorPartialMemset.ll b/test/Transforms/ScalarRepl/2011-06-17-VectorPartialMemset.ll
index f8530d68a8fb..af6d1f36faea 100644
--- a/test/Transforms/ScalarRepl/2011-06-17-VectorPartialMemset.ll
+++ b/test/Transforms/ScalarRepl/2011-06-17-VectorPartialMemset.ll
@@ -14,7 +14,7 @@ entry:
   %a = alloca <4 x float>, align 16
   %p = bitcast <4 x float>* %a to i8*
   call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 3, i32 16, i1 false)
-  %vec = load <4 x float>* %a, align 8
+  %vec = load <4 x float>, <4 x float>* %a, align 8
   %val = extractelement <4 x float> %vec, i32 0
   ret float %val
 }
@@ -29,7 +29,7 @@ entry:
   %p = bitcast { <4 x float> }* %a to i8*
   call void @llvm.memset.p0i8.i32(i8* %p, i8 0, i32 16, i32 16, i1 false)
   %q = bitcast { <4 x float> }* %a to [2 x <2 x float>]*
-  %arrayidx = getelementptr inbounds [2 x <2 x float>]* %q, i32 0, i32 0
+  %arrayidx = getelementptr inbounds [2 x <2 x float>], [2 x <2 x float>]* %q, i32 0, i32 0
   store <2 x float> undef, <2 x float>* %arrayidx, align 8
   ret void
 }
diff --git a/test/Transforms/ScalarRepl/2011-09-22-PHISpeculateInvoke.ll b/test/Transforms/ScalarRepl/2011-09-22-PHISpeculateInvoke.ll
index 5f4d0fc7e1c8..9a24662f4114 100644
--- a/test/Transforms/ScalarRepl/2011-09-22-PHISpeculateInvoke.ll
+++ b/test/Transforms/ScalarRepl/2011-09-22-PHISpeculateInvoke.ll
@@ -25,7 +25,7 @@ else:                                             ; preds = %2
 
 join:                                             ; preds = %then, %else
   %storemerge.in = phi i32* [ %retptr2, %else ], [ %retptr1, %then ]
-  %storemerge = load i32* %storemerge.in
+  %storemerge = load i32, i32* %storemerge.in
   %x3 = call i32 @extern_fn2(i32 %storemerge)
   ret void
 
diff --git a/test/Transforms/ScalarRepl/2011-10-22-VectorCrash.ll b/test/Transforms/ScalarRepl/2011-10-22-VectorCrash.ll
index e4456361a66d..c9c1a148a484 100644
--- a/test/Transforms/ScalarRepl/2011-10-22-VectorCrash.ll
+++ b/test/Transforms/ScalarRepl/2011-10-22-VectorCrash.ll
@@ -11,7 +11,7 @@ define void @test() nounwind {
 entry:
   %u = alloca %union.anon, align 16
   %u164 = bitcast %union.anon* %u to [4 x i32]*
-  %arrayidx165 = getelementptr inbounds [4 x i32]* %u164, i32 0, i32 0
+  %arrayidx165 = getelementptr inbounds [4 x i32], [4 x i32]* %u164, i32 0, i32 0
   store i32 undef, i32* %arrayidx165, align 4
   %v186 = bitcast %union.anon* %u to <4 x float>*
   store <4 x float> undef, <4 x float>* %v186, align 16
diff --git a/test/Transforms/ScalarRepl/2011-11-11-EmptyStruct.ll b/test/Transforms/ScalarRepl/2011-11-11-EmptyStruct.ll
index da707b706d6d..51d1d146a90b 100644
--- a/test/Transforms/ScalarRepl/2011-11-11-EmptyStruct.ll
+++ b/test/Transforms/ScalarRepl/2011-11-11-EmptyStruct.ll
@@ -13,13 +13,13 @@ define double @test() nounwind uwtable ssp {
 entry:
   %retval = alloca %struct.S, align 8
   %ret = alloca %struct.S, align 8
-  %b = getelementptr inbounds %struct.S* %ret, i32 0, i32 1
+  %b = getelementptr inbounds %struct.S, %struct.S* %ret, i32 0, i32 1
   store double 1.000000e+00, double* %b, align 8
   %0 = bitcast %struct.S* %retval to i8*
   %1 = bitcast %struct.S* %ret to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %0, i8* %1, i64 8, i32 8, i1 false)
   %2 = bitcast %struct.S* %retval to double*
-  %3 = load double* %2, align 1
+  %3 = load double, double* %2, align 1
   ret double %3
 }
 
diff --git a/test/Transforms/ScalarRepl/AggregatePromote.ll b/test/Transforms/ScalarRepl/AggregatePromote.ll
index 16b327356bbd..f6dfdf55346a 100644
--- a/test/Transforms/ScalarRepl/AggregatePromote.ll
+++ b/test/Transforms/ScalarRepl/AggregatePromote.ll
@@ -10,7 +10,7 @@ define i64 @test1(i64 %X) {
 	%B = bitcast i64* %A to i32*		; <i32*> [#uses=1]
 	%C = bitcast i32* %B to i8*		; <i8*> [#uses=1]
 	store i8 0, i8* %C
-	%Y = load i64* %A		; <i64> [#uses=1]
+	%Y = load i64, i64* %A		; <i64> [#uses=1]
 	ret i64 %Y
 }
 
@@ -18,22 +18,22 @@ define i8 @test2(i64 %X) {
 	%X_addr = alloca i64		; <i64*> [#uses=2]
 	store i64 %X, i64* %X_addr
 	%tmp.0 = bitcast i64* %X_addr to i32*		; <i32*> [#uses=1]
-	%tmp.1 = getelementptr i32* %tmp.0, i32 1		; <i32*> [#uses=1]
+	%tmp.1 = getelementptr i32, i32* %tmp.0, i32 1		; <i32*> [#uses=1]
 	%tmp.2 = bitcast i32* %tmp.1 to i8*		; <i8*> [#uses=1]
-	%tmp.3 = getelementptr i8* %tmp.2, i32 3		; <i8*> [#uses=1]
-	%tmp.2.upgrd.1 = load i8* %tmp.3		; <i8> [#uses=1]
+	%tmp.3 = getelementptr i8, i8* %tmp.2, i32 3		; <i8*> [#uses=1]
+	%tmp.2.upgrd.1 = load i8, i8* %tmp.3		; <i8> [#uses=1]
 	ret i8 %tmp.2.upgrd.1
 }
 
 define i16 @crafty(i64 %X) {
 	%a = alloca { i64 }		; <{ i64 }*> [#uses=2]
-	%tmp.0 = getelementptr { i64 }* %a, i32 0, i32 0		; <i64*> [#uses=1]
+	%tmp.0 = getelementptr { i64 }, { i64 }* %a, i32 0, i32 0		; <i64*> [#uses=1]
 	store i64 %X, i64* %tmp.0
 	%tmp.3 = bitcast { i64 }* %a to [4 x i16]*		; <[4 x i16]*> [#uses=2]
-	%tmp.4 = getelementptr [4 x i16]* %tmp.3, i32 0, i32 3		; <i16*> [#uses=1]
-	%tmp.5 = load i16* %tmp.4		; <i16> [#uses=1]
-	%tmp.8 = getelementptr [4 x i16]* %tmp.3, i32 0, i32 2		; <i16*> [#uses=1]
-	%tmp.9 = load i16* %tmp.8		; <i16> [#uses=1]
+	%tmp.4 = getelementptr [4 x i16], [4 x i16]* %tmp.3, i32 0, i32 3		; <i16*> [#uses=1]
+	%tmp.5 = load i16, i16* %tmp.4		; <i16> [#uses=1]
+	%tmp.8 = getelementptr [4 x i16], [4 x i16]* %tmp.3, i32 0, i32 2		; <i16*> [#uses=1]
+	%tmp.9 = load i16, i16* %tmp.8		; <i16> [#uses=1]
 	%tmp.10 = or i16 %tmp.9, %tmp.5		; <i16> [#uses=1]
 	ret i16 %tmp.10
 }
@@ -42,10 +42,10 @@ define i16 @crafty2(i64 %X) {
 	%a = alloca i64		; <i64*> [#uses=2]
 	store i64 %X, i64* %a
 	%tmp.3 = bitcast i64* %a to [4 x i16]*		; <[4 x i16]*> [#uses=2]
-	%tmp.4 = getelementptr [4 x i16]* %tmp.3, i32 0, i32 3		; <i16*> [#uses=1]
-	%tmp.5 = load i16* %tmp.4		; <i16> [#uses=1]
-	%tmp.8 = getelementptr [4 x i16]* %tmp.3, i32 0, i32 2		; <i16*> [#uses=1]
-	%tmp.9 = load i16* %tmp.8		; <i16> [#uses=1]
+	%tmp.4 = getelementptr [4 x i16], [4 x i16]* %tmp.3, i32 0, i32 3		; <i16*> [#uses=1]
+	%tmp.5 = load i16, i16* %tmp.4		; <i16> [#uses=1]
+	%tmp.8 = getelementptr [4 x i16], [4 x i16]* %tmp.3, i32 0, i32 2		; <i16*> [#uses=1]
+	%tmp.9 = load i16, i16* %tmp.8		; <i16> [#uses=1]
 	%tmp.10 = or i16 %tmp.9, %tmp.5		; <i16> [#uses=1]
 	ret i16 %tmp.10
 }
diff --git a/test/Transforms/ScalarRepl/DifferingTypes.ll b/test/Transforms/ScalarRepl/DifferingTypes.ll
index 933c47f79874..3860f6cd757f 100644
--- a/test/Transforms/ScalarRepl/DifferingTypes.ll
+++ b/test/Transforms/ScalarRepl/DifferingTypes.ll
@@ -10,7 +10,7 @@ define i32 @testfunc(i32 %i, i8 %j) {
 	store i32 %i, i32* %I
 	%P = bitcast i32* %I to i8*		; <i8*> [#uses=1]
 	store i8 %j, i8* %P
-	%t = load i32* %I		; <i32> [#uses=1]
+	%t = load i32, i32* %I		; <i32> [#uses=1]
 	ret i32 %t
 }
 
diff --git a/test/Transforms/ScalarRepl/address-space.ll b/test/Transforms/ScalarRepl/address-space.ll
index d8efc1774b7e..b8b90efefc3b 100644
--- a/test/Transforms/ScalarRepl/address-space.ll
+++ b/test/Transforms/ScalarRepl/address-space.ll
@@ -7,22 +7,22 @@ target triple = "x86_64-apple-darwin10"
 %struct.anon = type { [1 x float] }
 
 ; CHECK-LABEL: define void @Test(
-; CHECK: load float addrspace(2)*
+; CHECK: load float, float addrspace(2)*
 ; CHECK-NEXT: fsub float
 ; CHECK: store float {{.*}}, float addrspace(2)* 
 define void @Test(%struct.anon addrspace(2)* %pPtr) nounwind {
 entry:
   %s = alloca %struct.anon, align 4               ; <%struct.anon*> [#uses=3]
-  %arrayidx = getelementptr inbounds %struct.anon addrspace(2)* %pPtr, i64 0 ; <%struct.anon addrspace(2)*> [#uses=1]
+  %arrayidx = getelementptr inbounds %struct.anon, %struct.anon addrspace(2)* %pPtr, i64 0 ; <%struct.anon addrspace(2)*> [#uses=1]
   %tmp1 = bitcast %struct.anon* %s to i8*         ; <i8*> [#uses=1]
   %tmp2 = bitcast %struct.anon addrspace(2)* %arrayidx to i8 addrspace(2)* ; <i8 addrspace(2)*> [#uses=1]
   call void @llvm.memcpy.p0i8.p2i8.i64(i8* %tmp1, i8 addrspace(2)* %tmp2, i64 4, i32 4, i1 false)
-  %tmp3 = getelementptr inbounds %struct.anon* %s, i32 0, i32 0 ; <[1 x float]*> [#uses=1]
-  %arrayidx4 = getelementptr inbounds [1 x float]* %tmp3, i32 0, i64 0 ; <float*> [#uses=2]
-  %tmp5 = load float* %arrayidx4                  ; <float> [#uses=1]
+  %tmp3 = getelementptr inbounds %struct.anon, %struct.anon* %s, i32 0, i32 0 ; <[1 x float]*> [#uses=1]
+  %arrayidx4 = getelementptr inbounds [1 x float], [1 x float]* %tmp3, i32 0, i64 0 ; <float*> [#uses=2]
+  %tmp5 = load float, float* %arrayidx4                  ; <float> [#uses=1]
   %sub = fsub float %tmp5, 5.000000e+00           ; <float> [#uses=1]
   store float %sub, float* %arrayidx4
-  %arrayidx7 = getelementptr inbounds %struct.anon addrspace(2)* %pPtr, i64 0 ; <%struct.anon addrspace(2)*> [#uses=1]
+  %arrayidx7 = getelementptr inbounds %struct.anon, %struct.anon addrspace(2)* %pPtr, i64 0 ; <%struct.anon addrspace(2)*> [#uses=1]
   %tmp8 = bitcast %struct.anon addrspace(2)* %arrayidx7 to i8 addrspace(2)* ; <i8 addrspace(2)*> [#uses=1]
   %tmp9 = bitcast %struct.anon* %s to i8*         ; <i8*> [#uses=1]
   call void @llvm.memcpy.p2i8.p0i8.i64(i8 addrspace(2)* %tmp8, i8* %tmp9, i64 4, i32 4, i1 false)
diff --git a/test/Transforms/ScalarRepl/arraytest.ll b/test/Transforms/ScalarRepl/arraytest.ll
index 06a928c6d821..486e725fa6a5 100644
--- a/test/Transforms/ScalarRepl/arraytest.ll
+++ b/test/Transforms/ScalarRepl/arraytest.ll
@@ -3,9 +3,9 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i32 @test() {
 	%X = alloca [4 x i32]		; <[4 x i32]*> [#uses=1]
-	%Y = getelementptr [4 x i32]* %X, i64 0, i64 0		; <i32*> [#uses=2]
+	%Y = getelementptr [4 x i32], [4 x i32]* %X, i64 0, i64 0		; <i32*> [#uses=2]
 	store i32 0, i32* %Y
-	%Z = load i32* %Y		; <i32> [#uses=1]
+	%Z = load i32, i32* %Y		; <i32> [#uses=1]
 	ret i32 %Z
 }
 
diff --git a/test/Transforms/ScalarRepl/badarray.ll b/test/Transforms/ScalarRepl/badarray.ll
index 480e12b8d445..6f5bc95b43ba 100644
--- a/test/Transforms/ScalarRepl/badarray.ll
+++ b/test/Transforms/ScalarRepl/badarray.ll
@@ -10,9 +10,9 @@ define i32 @test1() {
 ; CHECK-LABEL: @test1(
 ; CHECK-NOT: = alloca
 	%X = alloca [4 x i32]
-	%Y = getelementptr [4 x i32]* %X, i64 0, i64 6		; <i32*> [#uses=2]
+	%Y = getelementptr [4 x i32], [4 x i32]* %X, i64 0, i64 6		; <i32*> [#uses=2]
 	store i32 0, i32* %Y
-	%Z = load i32* %Y		; <i32> [#uses=1]
+	%Z = load i32, i32* %Y		; <i32> [#uses=1]
 	ret i32 %Z
 }
 
@@ -24,7 +24,7 @@ entry:
 ; CHECK-NOT: = alloca
         %yx2.i = alloca float, align 4          ; <float*> [#uses=1]            
         %yx26.i = bitcast float* %yx2.i to i64*         ; <i64*> [#uses=1]      
-        %0 = load i64* %yx26.i, align 8         ; <i64> [#uses=0]               
+        %0 = load i64, i64* %yx26.i, align 8         ; <i64> [#uses=0]               
         unreachable
 }
 
@@ -38,9 +38,9 @@ entry:
 ; CHECK-NOT: = alloca
 ; CHECK: store i64
   %var_1 = alloca %padded, align 8                ; <%padded*> [#uses=3]
-  %0 = getelementptr inbounds %padded* %var_1, i32 0, i32 0 ; <%base*> [#uses=2]
+  %0 = getelementptr inbounds %padded, %padded* %var_1, i32 0, i32 0 ; <%base*> [#uses=2]
   
-  %p2 = getelementptr inbounds %base* %0, i32 0, i32 1, i32 0 ; <i8*> [#uses=1]
+  %p2 = getelementptr inbounds %base, %base* %0, i32 0, i32 1, i32 0 ; <i8*> [#uses=1]
   store i8 72, i8* %p2, align 1
   
   ; 72 -> a[0].
diff --git a/test/Transforms/ScalarRepl/basictest.ll b/test/Transforms/ScalarRepl/basictest.ll
index af3c237d4914..35d4d3ba86fe 100644
--- a/test/Transforms/ScalarRepl/basictest.ll
+++ b/test/Transforms/ScalarRepl/basictest.ll
@@ -3,9 +3,9 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 
 define i32 @test1() {
 	%X = alloca { i32, float }		; <{ i32, float }*> [#uses=1]
-	%Y = getelementptr { i32, float }* %X, i64 0, i32 0		; <i32*> [#uses=2]
+	%Y = getelementptr { i32, float }, { i32, float }* %X, i64 0, i32 0		; <i32*> [#uses=2]
 	store i32 0, i32* %Y
-	%Z = load i32* %Y		; <i32> [#uses=1]
+	%Z = load i32, i32* %Y		; <i32> [#uses=1]
 	ret i32 %Z
 ; CHECK-LABEL: @test1(
 ; CHECK-NOT: alloca
@@ -21,7 +21,7 @@ define i64 @test2(i64 %X) {
         br label %L2
         
 L2:
-	%Z = load i64* %B		; <i32> [#uses=1]
+	%Z = load i64, i64* %B		; <i32> [#uses=1]
 	ret i64 %Z
 ; CHECK-LABEL: @test2(
 ; CHECK-NOT: alloca
diff --git a/test/Transforms/ScalarRepl/bitfield-sroa.ll b/test/Transforms/ScalarRepl/bitfield-sroa.ll
index 3728658caaee..52986b0a49d7 100644
--- a/test/Transforms/ScalarRepl/bitfield-sroa.ll
+++ b/test/Transforms/ScalarRepl/bitfield-sroa.ll
@@ -7,11 +7,11 @@ define i8 @foo(i64 %A) {
         %ALL = alloca %t, align 8 
         %tmp59172 = bitcast %t* %ALL to i64*
         store i64 %A, i64* %tmp59172, align 8
-        %C = getelementptr %t* %ALL, i32 0, i32 0, i32 1             
+        %C = getelementptr %t, %t* %ALL, i32 0, i32 0, i32 1             
         %D = bitcast i16* %C to i32*    
-        %E = load i32* %D, align 4     
+        %E = load i32, i32* %D, align 4     
         %F = bitcast %t* %ALL to i8* 
-        %G = load i8* %F, align 8 
+        %G = load i8, i8* %F, align 8 
 	ret i8 %G
 }
 
diff --git a/test/Transforms/ScalarRepl/copy-aggregate.ll b/test/Transforms/ScalarRepl/copy-aggregate.ll
index 51ba8109b39b..97977dbf11f1 100644
--- a/test/Transforms/ScalarRepl/copy-aggregate.ll
+++ b/test/Transforms/ScalarRepl/copy-aggregate.ll
@@ -10,10 +10,10 @@ define i32 @test1(i64 %V) nounwind {
 	%Y = bitcast {{i32,i32}}* %X to i64*
 	store i64 %V, i64* %Y
 
-	%A = getelementptr {{i32,i32}}* %X, i32 0, i32 0, i32 0
-	%B = getelementptr {{i32,i32}}* %X, i32 0, i32 0, i32 1
-	%a = load i32* %A
-	%b = load i32* %B
+	%A = getelementptr {{i32,i32}}, {{i32,i32}}* %X, i32 0, i32 0, i32 0
+	%B = getelementptr {{i32,i32}}, {{i32,i32}}* %X, i32 0, i32 0, i32 1
+	%a = load i32, i32* %A
+	%b = load i32, i32* %B
 	%c = add i32 %a, %b
 	ret i32 %c
 }
@@ -26,10 +26,10 @@ define float @test2(i128 %V) nounwind {
 	%Y = bitcast {[4 x float]}* %X to i128*
 	store i128 %V, i128* %Y
 
-	%A = getelementptr {[4 x float]}* %X, i32 0, i32 0, i32 0
-	%B = getelementptr {[4 x float]}* %X, i32 0, i32 0, i32 3
-	%a = load float* %A
-	%b = load float* %B
+	%A = getelementptr {[4 x float]}, {[4 x float]}* %X, i32 0, i32 0, i32 0
+	%B = getelementptr {[4 x float]}, {[4 x float]}* %X, i32 0, i32 0, i32 3
+	%a = load float, float* %A
+	%b = load float, float* %B
 	%c = fadd float %a, %b
 	ret float %c
 }
@@ -40,13 +40,13 @@ define i64 @test3(i32 %a, i32 %b) nounwind {
 ; CHECK-NOT: alloca
 	%X = alloca {{i32, i32}}
 
-	%A = getelementptr {{i32,i32}}* %X, i32 0, i32 0, i32 0
-	%B = getelementptr {{i32,i32}}* %X, i32 0, i32 0, i32 1
+	%A = getelementptr {{i32,i32}}, {{i32,i32}}* %X, i32 0, i32 0, i32 0
+	%B = getelementptr {{i32,i32}}, {{i32,i32}}* %X, i32 0, i32 0, i32 1
         store i32 %a, i32* %A
         store i32 %b, i32* %B
 
 	%Y = bitcast {{i32,i32}}* %X to i64*
-        %Z = load i64* %Y
+        %Z = load i64, i64* %Y
 	ret i64 %Z
 }
 
@@ -55,13 +55,13 @@ define i128 @test4(float %a, float %b) nounwind {
 ; CHECK: test4
 ; CHECK-NOT: alloca
 	%X = alloca {[4 x float]}
-	%A = getelementptr {[4 x float]}* %X, i32 0, i32 0, i32 0
-	%B = getelementptr {[4 x float]}* %X, i32 0, i32 0, i32 3
+	%A = getelementptr {[4 x float]}, {[4 x float]}* %X, i32 0, i32 0, i32 0
+	%B = getelementptr {[4 x float]}, {[4 x float]}* %X, i32 0, i32 0, i32 3
 	store float %a, float* %A
 	store float %b, float* %B
         
       	%Y = bitcast {[4 x float]}* %X to i128*
-	%V = load i128* %Y
+	%V = load i128, i128* %Y
 	ret i128 %V
 }
 
diff --git a/test/Transforms/ScalarRepl/crash.ll b/test/Transforms/ScalarRepl/crash.ll
index 8c60dceb8b07..72e9f090fb68 100644
--- a/test/Transforms/ScalarRepl/crash.ll
+++ b/test/Transforms/ScalarRepl/crash.ll
@@ -11,7 +11,7 @@ entry:
   unreachable
 
 for.cond:                                         ; preds = %for.cond
-  %tmp1.i = load i32** %l_72, align 8
+  %tmp1.i = load i32*, i32** %l_72, align 8
   store i32* %tmp1.i, i32** %l_72, align 8
   br label %for.cond
 
@@ -22,15 +22,15 @@ if.end:                                           ; No predecessors!
 
 define void @test2() {
   %E = alloca { { i32, float, double, i64 }, { i32, float, double, i64 } }        ; <{ { i32, float, double, i64 }, { i32, float, double, i64 } }*> [#uses=1]
-  %tmp.151 = getelementptr { { i32, float, double, i64 }, { i32, float, double, i64 } }* %E, i64 0, i32 1, i32 3          ; <i64*> [#uses=0]
+  %tmp.151 = getelementptr { { i32, float, double, i64 }, { i32, float, double, i64 } }, { { i32, float, double, i64 }, { i32, float, double, i64 } }* %E, i64 0, i32 1, i32 3          ; <i64*> [#uses=0]
   ret void
 }
 
 define i32 @test3() {
         %X = alloca { [4 x i32] }               ; <{ [4 x i32] }*> [#uses=1]
-        %Y = getelementptr { [4 x i32] }* %X, i64 0, i32 0, i64 2               ; <i32*> [#uses=2]
+        %Y = getelementptr { [4 x i32] }, { [4 x i32] }* %X, i64 0, i32 0, i64 2               ; <i32*> [#uses=2]
         store i32 4, i32* %Y
-        %Z = load i32* %Y               ; <i32> [#uses=1]
+        %Z = load i32, i32* %Y               ; <i32> [#uses=1]
         ret i32 %Z
 }
 
@@ -102,11 +102,11 @@ bb9131:         ; preds = %bb1365
         ret void
 bb9875:         ; preds = %bb1365
         %source_ptr9884 = bitcast i8** %source_ptr to i8**              ; <i8**> [#uses=1]
-        %tmp9885 = load i8** %source_ptr9884            ; <i8*> [#uses=0]
+        %tmp9885 = load i8*, i8** %source_ptr9884            ; <i8*> [#uses=0]
         ret void
 bb10249:                ; preds = %bb1365
         %source_ptr10257 = bitcast i8** %source_ptr to i16**            ; <i16**> [#uses=1]
-        %tmp10258 = load i16** %source_ptr10257         ; <i16*> [#uses=0]
+        %tmp10258 = load i16*, i16** %source_ptr10257         ; <i16*> [#uses=0]
         ret void
 cond_next10377:         ; preds = %bb1365
         ret void
@@ -125,10 +125,10 @@ entry:
         %this_addr.i = alloca %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*                ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"**> [#uses=3]
         %tmp = alloca %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>", align 4                ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*> [#uses=1]
         store %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"* %tmp, %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"** %this_addr.i
-        %tmp.i = load %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"** %this_addr.i          ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*> [#uses=1]
+        %tmp.i = load %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*, %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"** %this_addr.i          ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*> [#uses=1]
         %tmp.i.upgrd.1 = bitcast %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"* %tmp.i to %"struct.__gnu_cxx::bitmap_allocator<char>"*              ; <%"struct.__gnu_cxx::bitmap_allocator<char>"*> [#uses=0]
-        %tmp1.i = load %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"** %this_addr.i         ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*> [#uses=1]
-        %tmp.i.upgrd.2 = getelementptr %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"* %tmp1.i, i32 0, i32 0         ; <%"struct.__gnu_cxx::bitmap_allocator<char>::_Alloc_block"**> [#uses=0]
+        %tmp1.i = load %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*, %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"** %this_addr.i         ; <%"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"*> [#uses=1]
+        %tmp.i.upgrd.2 = getelementptr %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>", %"struct.__gnu_cxx::balloc::_Inclusive_between<__gnu_cxx::bitmap_allocator<char>::_Alloc_block*>"* %tmp1.i, i32 0, i32 0         ; <%"struct.__gnu_cxx::bitmap_allocator<char>::_Alloc_block"**> [#uses=0]
         unreachable
 }
 
@@ -161,8 +161,8 @@ cond_next:              ; preds = %cond_true
         br i1 false, label %cond_next34, label %cond_next79
 
 cond_next34:            ; preds = %cond_next
-        %i.2.reload22 = load i32* null          ; <i32> [#uses=1]
-        %tmp51 = getelementptr %struct.aal_spanbucket_t* %SB, i32 0, i32 2, i32 0, i32 0, i32 %i.2.reload22, i32 1      
+        %i.2.reload22 = load i32, i32* null          ; <i32> [#uses=1]
+        %tmp51 = getelementptr %struct.aal_spanbucket_t, %struct.aal_spanbucket_t* %SB, i32 0, i32 2, i32 0, i32 0, i32 %i.2.reload22, i32 1      
         ; <i16*> [#uses=0]
         ret void
 
@@ -180,7 +180,7 @@ cond_next79:            ; preds = %cond_next, %entry
 define void @test8() {
 entry:
         %v = alloca %struct.c37304a__vrec
-        %0 = getelementptr %struct.c37304a__vrec* %v, i32 0, i32 0             
+        %0 = getelementptr %struct.c37304a__vrec, %struct.c37304a__vrec* %v, i32 0, i32 0             
         store i8 8, i8* %0, align 1
         unreachable
 }
@@ -193,7 +193,7 @@ entry:
 define i32 @test9() {
 entry:
         %.compoundliteral = alloca %0           
-        %tmp228 = getelementptr %0* %.compoundliteral, i32 0, i32 7
+        %tmp228 = getelementptr %0, %0* %.compoundliteral, i32 0, i32 7
         %tmp229 = bitcast [0 x i16]* %tmp228 to i8*             
         call void @llvm.memset.p0i8.i64(i8* %tmp229, i8 0, i64 0, i32 2, i1 false)
         unreachable
@@ -207,7 +207,7 @@ declare void @llvm.memset.i64(i8* nocapture, i8, i64, i32) nounwind
 define void @test10() {
 entry:
         %w = alloca %wrapper, align 8           ; <%wrapper*> [#uses=1]
-        %0 = getelementptr %wrapper* %w, i64 0, i32 0           ; <i1*>
+        %0 = getelementptr %wrapper, %wrapper* %w, i64 0, i32 0           ; <i1*>
         store i1 true, i1* %0
         ret void
 }
@@ -220,8 +220,8 @@ entry:
         %a = alloca %struct.singlebool, align 1         ; <%struct.singlebool*> [#uses=2]
         %storetmp.i = bitcast %struct.singlebool* %a to i1*             ; <i1*> [#uses=1]
         store i1 true, i1* %storetmp.i
-        %tmp = getelementptr %struct.singlebool* %a, i64 0, i32 0               ; <i8*> [#uses=1]
-        %tmp1 = load i8* %tmp           ; <i8> [#uses=1]
+        %tmp = getelementptr %struct.singlebool, %struct.singlebool* %a, i64 0, i32 0               ; <i8*> [#uses=1]
+        %tmp1 = load i8, i8* %tmp           ; <i8> [#uses=1]
         ret i8 %tmp1
 }
 
@@ -246,7 +246,7 @@ define void @test12() {
 bb4.i:
         %malloccall = tail call i8* @malloc(i32 0)
         %0 = bitcast i8* %malloccall to [0 x %struct.Item]*
-        %.sub.i.c.i = getelementptr [0 x %struct.Item]* %0, i32 0, i32 0                ; <%struct.Item*> [#uses=0]
+        %.sub.i.c.i = getelementptr [0 x %struct.Item], [0 x %struct.Item]* %0, i32 0, i32 0                ; <%struct.Item*> [#uses=0]
         unreachable
 }
 declare noalias i8* @malloc(i32)
@@ -268,13 +268,13 @@ define fastcc void @test() optsize inlinehint ssp align 2 {
 entry:
   %alloc.0.0 = alloca <4 x float>, align 16
   %bitcast = bitcast <4 x float>* %alloc.0.0 to [4 x float]*
-  %idx3 = getelementptr inbounds [4 x float]* %bitcast, i32 0, i32 3
+  %idx3 = getelementptr inbounds [4 x float], [4 x float]* %bitcast, i32 0, i32 3
   store float 0.000000e+00, float* %idx3, align 4
   br label %for.body10
 
 for.body10:                                       ; preds = %for.body10, %entry
   %loopidx = phi i32 [ 0, %entry ], [ undef, %for.body10 ]
-  %unusedidx = getelementptr inbounds <4 x float>* %alloc.0.0, i32 0, i32 %loopidx
+  %unusedidx = getelementptr inbounds <4 x float>, <4 x float>* %alloc.0.0, i32 0, i32 %loopidx
   br i1 undef, label %for.end, label %for.body10
 
 for.end:                                          ; preds = %for.body10
diff --git a/test/Transforms/ScalarRepl/debuginfo-preserved.ll b/test/Transforms/ScalarRepl/debuginfo-preserved.ll
index b0c459e21b10..8e5d4ff773dd 100644
--- a/test/Transforms/ScalarRepl/debuginfo-preserved.ll
+++ b/test/Transforms/ScalarRepl/debuginfo-preserved.ll
@@ -17,22 +17,22 @@ entry:
   %b.addr = alloca i32, align 4
   %c = alloca i32, align 4
   store i32 %a, i32* %a.addr, align 4
-  call void @llvm.dbg.declare(metadata i32* %a.addr, metadata !6, metadata !{}), !dbg !7
+  call void @llvm.dbg.declare(metadata i32* %a.addr, metadata !6, metadata !DIExpression()), !dbg !7
   store i32 %b, i32* %b.addr, align 4
-  call void @llvm.dbg.declare(metadata i32* %b.addr, metadata !8, metadata !{}), !dbg !9
-  call void @llvm.dbg.declare(metadata i32* %c, metadata !10, metadata !{}), !dbg !12
-  %tmp = load i32* %a.addr, align 4, !dbg !13
+  call void @llvm.dbg.declare(metadata i32* %b.addr, metadata !8, metadata !DIExpression()), !dbg !9
+  call void @llvm.dbg.declare(metadata i32* %c, metadata !10, metadata !DIExpression()), !dbg !12
+  %tmp = load i32, i32* %a.addr, align 4, !dbg !13
   store i32 %tmp, i32* %c, align 4, !dbg !13
-  %tmp1 = load i32* %a.addr, align 4, !dbg !14
-  %tmp2 = load i32* %b.addr, align 4, !dbg !14
+  %tmp1 = load i32, i32* %a.addr, align 4, !dbg !14
+  %tmp2 = load i32, i32* %b.addr, align 4, !dbg !14
   %add = add nsw i32 %tmp1, %tmp2, !dbg !14
   store i32 %add, i32* %a.addr, align 4, !dbg !14
-  %tmp3 = load i32* %c, align 4, !dbg !15
-  %tmp4 = load i32* %b.addr, align 4, !dbg !15
+  %tmp3 = load i32, i32* %c, align 4, !dbg !15
+  %tmp4 = load i32, i32* %b.addr, align 4, !dbg !15
   %sub = sub nsw i32 %tmp3, %tmp4, !dbg !15
   store i32 %sub, i32* %b.addr, align 4, !dbg !15
-  %tmp5 = load i32* %a.addr, align 4, !dbg !16
-  %tmp6 = load i32* %b.addr, align 4, !dbg !16
+  %tmp5 = load i32, i32* %a.addr, align 4, !dbg !16
+  %tmp6 = load i32, i32* %b.addr, align 4, !dbg !16
   %add7 = add nsw i32 %tmp5, %tmp6, !dbg !16
   ret i32 %add7, !dbg !16
 }
@@ -42,24 +42,24 @@ declare void @llvm.dbg.declare(metadata, metadata, metadata) nounwind readnone
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!20}
 
-!0 = !{!"0x11\0012\00clang version 3.0 (trunk 131941)\000\00\000\00\000", !18, !19, !19, !17, null, null} ; [ DW_TAG_compile_unit ]
-!1 = !{!"0x2e\00f\00f\00\001\000\001\000\006\00256\000\001", !18, !2, !3, null, i32 (i32, i32)* @f, null, null, null} ; [ DW_TAG_subprogram ] [line 1] [def] [f]
-!2 = !{!"0x29", !18} ; [ DW_TAG_file_type ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !18, !2, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.0 (trunk 131941)", isOptimized: false, emissionKind: 0, file: !18, enums: !19, retainedTypes: !19, subprograms: !17)
+!1 = !DISubprogram(name: "f", line: 1, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, scopeLine: 1, file: !18, scope: !2, type: !3, function: i32 (i32, i32)* @f)
+!2 = !DIFile(filename: "/d/j/debug-test.c", directory: "/Volumes/Data/b")
+!3 = !DISubroutineType(types: !4)
 !4 = !{!5}
-!5 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !0} ; [ DW_TAG_base_type ]
-!6 = !{!"0x101\00a\0016777217\000", !1, !2, !5} ; [ DW_TAG_arg_variable ]
-!7 = !MDLocation(line: 1, column: 11, scope: !1)
-!8 = !{!"0x101\00b\0033554433\000", !1, !2, !5} ; [ DW_TAG_arg_variable ]
-!9 = !MDLocation(line: 1, column: 18, scope: !1)
-!10 = !{!"0x100\00c\002\000", !11, !2, !5} ; [ DW_TAG_auto_variable ]
-!11 = !{!"0xb\001\0021\000", !18, !1} ; [ DW_TAG_lexical_block ]
-!12 = !MDLocation(line: 2, column: 9, scope: !11)
-!13 = !MDLocation(line: 2, column: 14, scope: !11)
-!14 = !MDLocation(line: 3, column: 5, scope: !11)
-!15 = !MDLocation(line: 4, column: 5, scope: !11)
-!16 = !MDLocation(line: 5, column: 5, scope: !11)
+!5 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!6 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "a", line: 1, arg: 1, scope: !1, file: !2, type: !5)
+!7 = !DILocation(line: 1, column: 11, scope: !1)
+!8 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "b", line: 1, arg: 2, scope: !1, file: !2, type: !5)
+!9 = !DILocation(line: 1, column: 18, scope: !1)
+!10 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "c", line: 2, scope: !11, file: !2, type: !5)
+!11 = distinct !DILexicalBlock(line: 1, column: 21, file: !18, scope: !1)
+!12 = !DILocation(line: 2, column: 9, scope: !11)
+!13 = !DILocation(line: 2, column: 14, scope: !11)
+!14 = !DILocation(line: 3, column: 5, scope: !11)
+!15 = !DILocation(line: 4, column: 5, scope: !11)
+!16 = !DILocation(line: 5, column: 5, scope: !11)
 !17 = !{!1}
-!18 = !{!"/d/j/debug-test.c", !"/Volumes/Data/b"}
-!19 = !{i32 0}
-!20 = !{i32 1, !"Debug Info Version", i32 2}
+!18 = !DIFile(filename: "/d/j/debug-test.c", directory: "/Volumes/Data/b")
+!19 = !{}
+!20 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/ScalarRepl/inline-vector.ll b/test/Transforms/ScalarRepl/inline-vector.ll
index f7c70dcd5575..85f37414e656 100644
--- a/test/Transforms/ScalarRepl/inline-vector.ll
+++ b/test/Transforms/ScalarRepl/inline-vector.ll
@@ -30,22 +30,22 @@ for.body:                                         ; preds = %for.cond
   %tmp3 = bitcast %struct.Vector4* %vector to i8*
   call void @llvm.memcpy.p0i8.p0i8.i32(i8* %tmp2, i8* %tmp3, i32 16, i32 16, i1 false)
   %0 = bitcast %struct.Vector4* %agg.tmp to [2 x i64]*
-  %1 = load [2 x i64]* %0, align 16
+  %1 = load [2 x i64], [2 x i64]* %0, align 16
   %tmp2.i = extractvalue [2 x i64] %1, 0
   %tmp3.i = zext i64 %tmp2.i to i128
   %tmp10.i = bitcast i128 %tmp3.i to <4 x float>
   %sub.i.i = fsub <4 x float> <float -0.000000e+00, float -0.000000e+00, float -0.000000e+00, float -0.000000e+00>, %tmp10.i
   %2 = bitcast %struct.Vector4* %vector to <4 x float>*
   store <4 x float> %sub.i.i, <4 x float>* %2, align 16
-  %tmp4 = load i32* %i, align 4
+  %tmp4 = load i32, i32* %i, align 4
   %inc = add nsw i32 %tmp4, 1
   br label %for.cond
 
 for.end:                                          ; preds = %for.cond
-  %x = getelementptr inbounds %struct.Vector4* %vector, i32 0, i32 0
-  %tmp5 = load float* %x, align 16
+  %x = getelementptr inbounds %struct.Vector4, %struct.Vector4* %vector, i32 0, i32 0
+  %tmp5 = load float, float* %x, align 16
   %conv = fpext float %tmp5 to double
-  %call = call i32 (...)* @printf(double %conv) nounwind
+  %call = call i32 (...) @printf(double %conv) nounwind
   ret void
 }
 
diff --git a/test/Transforms/ScalarRepl/lifetime.ll b/test/Transforms/ScalarRepl/lifetime.ll
index 47cb8549822b..c0ddfb58bbdf 100644
--- a/test/Transforms/ScalarRepl/lifetime.ll
+++ b/test/Transforms/ScalarRepl/lifetime.ll
@@ -11,9 +11,9 @@ declare void @llvm.lifetime.end(i64, i8*)
 define void @test1() {
 ; CHECK-LABEL: @test1(
   %A = alloca %t1
-  %A1 = getelementptr %t1* %A, i32 0, i32 0
-  %A2 = getelementptr %t1* %A, i32 0, i32 1
-  %A3 = getelementptr %t1* %A, i32 0, i32 2
+  %A1 = getelementptr %t1, %t1* %A, i32 0, i32 0
+  %A2 = getelementptr %t1, %t1* %A, i32 0, i32 1
+  %A3 = getelementptr %t1, %t1* %A, i32 0, i32 2
   %B = bitcast i32* %A1 to i8*
   store i32 0, i32* %A1
   call void @llvm.lifetime.start(i64 -1, i8* %B)
@@ -24,13 +24,13 @@ define void @test1() {
 define void @test2() {
 ; CHECK-LABEL: @test2(
   %A = alloca %t1
-  %A1 = getelementptr %t1* %A, i32 0, i32 0
-  %A2 = getelementptr %t1* %A, i32 0, i32 1
-  %A3 = getelementptr %t1* %A, i32 0, i32 2
+  %A1 = getelementptr %t1, %t1* %A, i32 0, i32 0
+  %A2 = getelementptr %t1, %t1* %A, i32 0, i32 1
+  %A3 = getelementptr %t1, %t1* %A, i32 0, i32 2
   %B = bitcast i32* %A2 to i8*
   store i32 0, i32* %A2
   call void @llvm.lifetime.start(i64 -1, i8* %B)
-  %C = load i32* %A2
+  %C = load i32, i32* %A2
   ret void
 ; CHECK: ret void
 }
@@ -38,13 +38,13 @@ define void @test2() {
 define void @test3() {
 ; CHECK-LABEL: @test3(
   %A = alloca %t1
-  %A1 = getelementptr %t1* %A, i32 0, i32 0
-  %A2 = getelementptr %t1* %A, i32 0, i32 1
-  %A3 = getelementptr %t1* %A, i32 0, i32 2
+  %A1 = getelementptr %t1, %t1* %A, i32 0, i32 0
+  %A2 = getelementptr %t1, %t1* %A, i32 0, i32 1
+  %A3 = getelementptr %t1, %t1* %A, i32 0, i32 2
   %B = bitcast i32* %A2 to i8*
   store i32 0, i32* %A2
   call void @llvm.lifetime.start(i64 6, i8* %B)
-  %C = load i32* %A2
+  %C = load i32, i32* %A2
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -52,13 +52,13 @@ define void @test3() {
 define void @test4() {
 ; CHECK-LABEL: @test4(
   %A = alloca %t1
-  %A1 = getelementptr %t1* %A, i32 0, i32 0
-  %A2 = getelementptr %t1* %A, i32 0, i32 1
-  %A3 = getelementptr %t1* %A, i32 0, i32 2
+  %A1 = getelementptr %t1, %t1* %A, i32 0, i32 0
+  %A2 = getelementptr %t1, %t1* %A, i32 0, i32 1
+  %A3 = getelementptr %t1, %t1* %A, i32 0, i32 2
   %B = bitcast i32* %A2 to i8*
   store i32 0, i32* %A2
   call void @llvm.lifetime.start(i64 1, i8* %B)
-  %C = load i32* %A2
+  %C = load i32, i32* %A2
   ret void
 ; CHECK-NEXT: ret void
 }
@@ -72,25 +72,25 @@ define void @test5() {
 ; CHECK: alloca{{.*}}i8
 ; CHECK: alloca{{.*}}i8
 
-  %A21 = getelementptr %t2* %A, i32 0, i32 1, i32 0
-  %A22 = getelementptr %t2* %A, i32 0, i32 1, i32 1
-  %A23 = getelementptr %t2* %A, i32 0, i32 1, i32 2
-  %A24 = getelementptr %t2* %A, i32 0, i32 1, i32 3
+  %A21 = getelementptr %t2, %t2* %A, i32 0, i32 1, i32 0
+  %A22 = getelementptr %t2, %t2* %A, i32 0, i32 1, i32 1
+  %A23 = getelementptr %t2, %t2* %A, i32 0, i32 1, i32 2
+  %A24 = getelementptr %t2, %t2* %A, i32 0, i32 1, i32 3
 ; CHECK-NOT: store i8 1
   store i8 1, i8* %A21
   store i8 2, i8* %A22
   store i8 3, i8* %A23
   store i8 4, i8* %A24
 
-  %A1 = getelementptr %t2* %A, i32 0, i32 0
-  %A2 = getelementptr %t2* %A, i32 0, i32 1, i32 1
-  %A3 = getelementptr %t2* %A, i32 0, i32 2
+  %A1 = getelementptr %t2, %t2* %A, i32 0, i32 0
+  %A2 = getelementptr %t2, %t2* %A, i32 0, i32 1, i32 1
+  %A3 = getelementptr %t2, %t2* %A, i32 0, i32 2
   store i8 0, i8* %A2
   call void @llvm.lifetime.start(i64 5, i8* %A2)
 ; CHECK: llvm.lifetime{{.*}}i64 1
 ; CHECK: llvm.lifetime{{.*}}i64 1
 ; CHECK: llvm.lifetime{{.*}}i64 1
-  %C = load i8* %A2
+  %C = load i8, i8* %A2
   ret void
 }
 
@@ -103,10 +103,10 @@ define void @test6() {
 ; CHECK: alloca i8
 ; CHECK: alloca i8
 
-  %A11 = getelementptr %t3* %A, i32 0, i32 0, i32 0
-  %A12 = getelementptr %t3* %A, i32 0, i32 0, i32 1
-  %A13 = getelementptr %t3* %A, i32 0, i32 0, i32 2
-  %A14 = getelementptr %t3* %A, i32 0, i32 0, i32 3
+  %A11 = getelementptr %t3, %t3* %A, i32 0, i32 0, i32 0
+  %A12 = getelementptr %t3, %t3* %A, i32 0, i32 0, i32 1
+  %A13 = getelementptr %t3, %t3* %A, i32 0, i32 0, i32 2
+  %A14 = getelementptr %t3, %t3* %A, i32 0, i32 0, i32 3
   store i16 11, i16* %A11
   store i16 12, i16* %A12
   store i16 13, i16* %A13
@@ -116,10 +116,10 @@ define void @test6() {
 ; CHECK-NOT: store i16 13
 ; CHECK-NOT: store i16 14
 
-  %A21 = getelementptr %t3* %A, i32 0, i32 1, i32 0
-  %A22 = getelementptr %t3* %A, i32 0, i32 1, i32 1
-  %A23 = getelementptr %t3* %A, i32 0, i32 1, i32 2
-  %A24 = getelementptr %t3* %A, i32 0, i32 1, i32 3
+  %A21 = getelementptr %t3, %t3* %A, i32 0, i32 1, i32 0
+  %A22 = getelementptr %t3, %t3* %A, i32 0, i32 1, i32 1
+  %A23 = getelementptr %t3, %t3* %A, i32 0, i32 1, i32 2
+  %A24 = getelementptr %t3, %t3* %A, i32 0, i32 1, i32 3
   store i8 21, i8* %A21
   store i8 22, i8* %A22
   store i8 23, i8* %A23
diff --git a/test/Transforms/ScalarRepl/load-store-aggregate.ll b/test/Transforms/ScalarRepl/load-store-aggregate.ll
index c5008ac1312f..88299f3679c2 100644
--- a/test/Transforms/ScalarRepl/load-store-aggregate.ll
+++ b/test/Transforms/ScalarRepl/load-store-aggregate.ll
@@ -11,21 +11,21 @@ target datalayout = "E-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f3
 define i32 @test(%struct.foo* %P) {
 entry:
 	%L = alloca %struct.foo, align 8		; <%struct.foo*> [#uses=2]
-        %V = load %struct.foo* %P
+        %V = load %struct.foo, %struct.foo* %P
         store %struct.foo %V, %struct.foo* %L
 
-	%tmp4 = getelementptr %struct.foo* %L, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
+	%tmp4 = getelementptr %struct.foo, %struct.foo* %L, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp5 = load i32, i32* %tmp4		; <i32> [#uses=1]
 	ret i32 %tmp5
 }
 
 define %struct.foo @test2(i32 %A, i32 %B) {
 entry:
 	%L = alloca %struct.foo, align 8		; <%struct.foo*> [#uses=2]
-        %L.0 = getelementptr %struct.foo* %L, i32 0, i32 0
+        %L.0 = getelementptr %struct.foo, %struct.foo* %L, i32 0, i32 0
         store i32 %A, i32* %L.0
-        %L.1 = getelementptr %struct.foo* %L, i32 0, i32 1
+        %L.1 = getelementptr %struct.foo, %struct.foo* %L, i32 0, i32 1
         store i32 %B, i32* %L.1
-        %V = load %struct.foo* %L
+        %V = load %struct.foo, %struct.foo* %L
         ret %struct.foo %V
 }
diff --git a/test/Transforms/ScalarRepl/memcpy-align.ll b/test/Transforms/ScalarRepl/memcpy-align.ll
index 6046e1295d9e..29a1bb8a4b5a 100644
--- a/test/Transforms/ScalarRepl/memcpy-align.ll
+++ b/test/Transforms/ScalarRepl/memcpy-align.ll
@@ -15,7 +15,7 @@ entry:
   %tmp = bitcast %struct.anon* %x0 to i8*         ; <i8*> [#uses=1]
   call void @llvm.memset.p0i8.i32(i8* %tmp, i8 0, i32 4, i32 4, i1 false)
   %tmp1 = bitcast %struct.anon* %x0 to i8*        ; <i8*> [#uses=1]
-  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds (%0* @c, i32
+  call void @llvm.memcpy.p0i8.p0i8.i32(i8* getelementptr inbounds (%0, %0* @c, i32
 0, i32 0, i32 0, i32 0), i8* %tmp1, i32 4, i32 4, i1 false)
   ret void
   
diff --git a/test/Transforms/ScalarRepl/memset-aggregate-byte-leader.ll b/test/Transforms/ScalarRepl/memset-aggregate-byte-leader.ll
index 8ac1d2570a1c..e8088c121fbc 100644
--- a/test/Transforms/ScalarRepl/memset-aggregate-byte-leader.ll
+++ b/test/Transforms/ScalarRepl/memset-aggregate-byte-leader.ll
@@ -12,10 +12,10 @@ target triple = "powerpc-apple-darwin8.8.0"
 define i32 @test1(%struct.foo* %P) {
 entry:
 	%L = alloca %struct.foo, align 2		; <%struct.foo*> [#uses=1]
-	%L2 = getelementptr %struct.foo* %L, i32 0, i32 0		; <i8*> [#uses=2]
-	%tmp13 = getelementptr %struct.foo* %P, i32 0, i32 0		; <i8*> [#uses=1]
+	%L2 = getelementptr %struct.foo, %struct.foo* %L, i32 0, i32 0		; <i8*> [#uses=2]
+	%tmp13 = getelementptr %struct.foo, %struct.foo* %P, i32 0, i32 0		; <i8*> [#uses=1]
 	call void @llvm.memcpy.p0i8.p0i8.i32( i8* %L2, i8* %tmp13, i32 2, i32 1, i1 false)
-	%tmp5 = load i8* %L2		; <i8> [#uses=1]
+	%tmp5 = load i8, i8* %L2		; <i8> [#uses=1]
 	%tmp56 = sext i8 %tmp5 to i32		; <i32> [#uses=1]
 	ret i32 %tmp56
 }
diff --git a/test/Transforms/ScalarRepl/memset-aggregate.ll b/test/Transforms/ScalarRepl/memset-aggregate.ll
index 3a5c37c04673..98e2dddefe7f 100644
--- a/test/Transforms/ScalarRepl/memset-aggregate.ll
+++ b/test/Transforms/ScalarRepl/memset-aggregate.ll
@@ -15,8 +15,8 @@ entry:
 	%L2 = bitcast %struct.foo* %L to i8*		; <i8*> [#uses=1]
 	%tmp13 = bitcast %struct.foo* %P to i8*		; <i8*> [#uses=1]
         call void @llvm.memcpy.p0i8.p0i8.i32(i8* %L2, i8* %tmp13, i32 8, i32 4, i1 false)
-	%tmp4 = getelementptr %struct.foo* %L, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
+	%tmp4 = getelementptr %struct.foo, %struct.foo* %L, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp5 = load i32, i32* %tmp4		; <i32> [#uses=1]
 	ret i32 %tmp5
 }
 
@@ -26,8 +26,8 @@ entry:
 	%L = alloca [4 x %struct.foo], align 16		; <[4 x %struct.foo]*> [#uses=2]
 	%L12 = bitcast [4 x %struct.foo]* %L to i8*		; <i8*> [#uses=1]
         call void @llvm.memset.p0i8.i32(i8* %L12, i8 0, i32 32, i32 16, i1 false)
-	%tmp4 = getelementptr [4 x %struct.foo]* %L, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
-	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
+	%tmp4 = getelementptr [4 x %struct.foo], [4 x %struct.foo]* %L, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp5 = load i32, i32* %tmp4		; <i32> [#uses=1]
 	ret i32 %tmp5
 }
 
@@ -37,12 +37,12 @@ entry:
 	%B = alloca %struct.bar, align 16		; <%struct.bar*> [#uses=4]
 	%B1 = bitcast %struct.bar* %B to i8*		; <i8*> [#uses=1]
 	call void @llvm.memset.p0i8.i32(i8* %B1, i8 1, i32 24, i32 16, i1 false)
-	%tmp3 = getelementptr %struct.bar* %B, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp3 = getelementptr %struct.bar, %struct.bar* %B, i32 0, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp3
-	%tmp4 = getelementptr %struct.bar* %B, i32 0, i32 2		; <double*> [#uses=1]
+	%tmp4 = getelementptr %struct.bar, %struct.bar* %B, i32 0, i32 2		; <double*> [#uses=1]
 	store double 1.000000e+01, double* %tmp4
-	%tmp6 = getelementptr %struct.bar* %B, i32 0, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp7 = load i32* %tmp6		; <i32> [#uses=1]
+	%tmp6 = getelementptr %struct.bar, %struct.bar* %B, i32 0, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp7 = load i32, i32* %tmp6		; <i32> [#uses=1]
 	ret i32 %tmp7
 }
 
@@ -52,13 +52,13 @@ entry:
 define i16 @test4() nounwind {
 entry:
 	%A = alloca %struct.f, align 8		; <%struct.f*> [#uses=3]
-	%0 = getelementptr %struct.f* %A, i32 0, i32 0		; <i32*> [#uses=1]
+	%0 = getelementptr %struct.f, %struct.f* %A, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %0, align 8
-	%1 = getelementptr %struct.f* %A, i32 0, i32 1		; <i32*> [#uses=1]
+	%1 = getelementptr %struct.f, %struct.f* %A, i32 0, i32 1		; <i32*> [#uses=1]
 	%2 = bitcast i32* %1 to i8*		; <i8*> [#uses=1]
 	call void @llvm.memset.p0i8.i32(i8* %2, i8 2, i32 12, i32 4, i1 false)
-	%3 = getelementptr %struct.f* %A, i32 0, i32 2		; <i32*> [#uses=1]
-	%4 = load i32* %3, align 8		; <i32> [#uses=1]
+	%3 = getelementptr %struct.f, %struct.f* %A, i32 0, i32 2		; <i32*> [#uses=1]
+	%4 = load i32, i32* %3, align 8		; <i32> [#uses=1]
 	%retval12 = trunc i32 %4 to i16		; <i16> [#uses=1]
 	ret i16 %retval12
 }
diff --git a/test/Transforms/ScalarRepl/negative-memset.ll b/test/Transforms/ScalarRepl/negative-memset.ll
index e52ab468fd2c..458d9610cd73 100644
--- a/test/Transforms/ScalarRepl/negative-memset.ll
+++ b/test/Transforms/ScalarRepl/negative-memset.ll
@@ -12,7 +12,7 @@ entry:
   store i32 0, i32* %retval
   %0 = bitcast [1 x i8]* %buff to i8*
   call void @llvm.memset.p0i8.i32(i8* %0, i8 0, i32 1, i32 1, i1 false)
-  %arraydecay = getelementptr inbounds [1 x i8]* %buff, i32 0, i32 0
+  %arraydecay = getelementptr inbounds [1 x i8], [1 x i8]* %buff, i32 0, i32 0
   call void @llvm.memset.p0i8.i32(i8* %arraydecay, i8 -1, i32 -8, i32 1, i1 false)	; Negative 8!
   ret i32 0
 }
diff --git a/test/Transforms/ScalarRepl/nonzero-first-index.ll b/test/Transforms/ScalarRepl/nonzero-first-index.ll
index b2e93feb6cc2..da757b08d454 100644
--- a/test/Transforms/ScalarRepl/nonzero-first-index.ll
+++ b/test/Transforms/ScalarRepl/nonzero-first-index.ll
@@ -12,9 +12,9 @@ define i32 @test1() {
 ; CHECK-NOT: = i160
 ; CHECK: ret i32 undef
 	%A = alloca %nested
-	%B = getelementptr %nested* %A, i32 0, i32 1, i32 0
-	%C = getelementptr i32* %B, i32 2
-	%D = load i32* %C
+	%B = getelementptr %nested, %nested* %A, i32 0, i32 1, i32 0
+	%C = getelementptr i32, i32* %B, i32 2
+	%D = load i32, i32* %C
 	ret i32 %D
 }
 
@@ -23,9 +23,9 @@ define i32 @test2() {
 ; CHECK-LABEL: @test2(
 ; CHECK: i160
 	%A = alloca %nested
-	%B = getelementptr %nested* %A, i32 0, i32 1, i32 0
-	%C = getelementptr i32* %B, i32 4
-	%D = load i32* %C
+	%B = getelementptr %nested, %nested* %A, i32 0, i32 1, i32 0
+	%C = getelementptr i32, i32* %B, i32 4
+	%D = load i32, i32* %C
 	ret i32 %D
 }
 
@@ -36,8 +36,8 @@ define i32 @test3() {
 ; CHECK: ret i32 undef
 	%A = alloca %nested
 	%B = bitcast %nested* %A to i32*
-	%C = getelementptr i32* %B, i32 2
-	%D = load i32* %C
+	%C = getelementptr i32, i32* %B, i32 2
+	%D = load i32, i32* %C
 	ret i32 %D
 }
 
@@ -47,7 +47,7 @@ define i32 @test4() {
 ; CHECK: i160
 	%A = alloca %nested
 	%B = bitcast %nested* %A to i32*
-	%C = getelementptr i32* %B, i32 -1
-	%D = load i32* %C
+	%C = getelementptr i32, i32* %B, i32 -1
+	%D = load i32, i32* %C
 	ret i32 %D
 }
diff --git a/test/Transforms/ScalarRepl/not-a-vector.ll b/test/Transforms/ScalarRepl/not-a-vector.ll
index 67fefb4841b3..04c1f93617b9 100644
--- a/test/Transforms/ScalarRepl/not-a-vector.ll
+++ b/test/Transforms/ScalarRepl/not-a-vector.ll
@@ -8,12 +8,12 @@ define double @test(double %A, double %B) {
 	%C = bitcast [7 x i64]* %ARR to double*
 	store double %A, double* %C
 
-	%D = getelementptr [7 x i64]* %ARR, i32 0, i32 4
+	%D = getelementptr [7 x i64], [7 x i64]* %ARR, i32 0, i32 4
 	%E = bitcast i64* %D to double*
 	store double %B, double* %E
 
-	%F = getelementptr double* %C, i32 4
-	%G = load double* %F
+	%F = getelementptr double, double* %C, i32 4
+	%G = load double, double* %F
 	ret double %G
 }
 
diff --git a/test/Transforms/ScalarRepl/only-memcpy-uses.ll b/test/Transforms/ScalarRepl/only-memcpy-uses.ll
index 935c2895520c..d0ed20b26bca 100644
--- a/test/Transforms/ScalarRepl/only-memcpy-uses.ll
+++ b/test/Transforms/ScalarRepl/only-memcpy-uses.ll
@@ -18,7 +18,7 @@ entry:
   %tmp2 = bitcast %struct.S* %agg.tmp to i8*
   %tmp3 = bitcast %struct.S* %t to i8*
   call void @llvm.memcpy.p0i8.p0i8.i64(i8* %tmp2, i8* %tmp3, i64 48, i32 4, i1 false)
-  %call = call i32 (...)* @bazz(%struct.S* byval %agg.tmp)
+  %call = call i32 (...) @bazz(%struct.S* byval %agg.tmp)
   ret void
 }
 
diff --git a/test/Transforms/ScalarRepl/phi-cycle.ll b/test/Transforms/ScalarRepl/phi-cycle.ll
index 05d9382cec40..a44f08180643 100644
--- a/test/Transforms/ScalarRepl/phi-cycle.ll
+++ b/test/Transforms/ScalarRepl/phi-cycle.ll
@@ -11,9 +11,9 @@ target triple = "x86_64-unknown-linux-gnu"
 define i32 @main(i32 %argc, i8** nocapture %argv) nounwind uwtable {
 entry:
   %f = alloca %struct.foo, align 4
-  %x.i = getelementptr inbounds %struct.foo* %f, i64 0, i32 0
+  %x.i = getelementptr inbounds %struct.foo, %struct.foo* %f, i64 0, i32 0
   store i32 1, i32* %x.i, align 4
-  %y.i = getelementptr inbounds %struct.foo* %f, i64 0, i32 1
+  %y.i = getelementptr inbounds %struct.foo, %struct.foo* %f, i64 0, i32 1
   br label %while.cond.i
 
 ; CHECK: while.cond.i:
@@ -67,10 +67,10 @@ while.cond.backedge.i:                            ; preds = %if.end.i, %while.bo
 
 ; CHECK: func.exit:
 ; CHECK-NOT: load
-; CHECK: %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([6 x i8]* @.str, i64 0, i64 0), i32 %tmp) [[NUW:#[0-9]+]]
+; CHECK: %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str, i64 0, i64 0), i32 %tmp) [[NUW:#[0-9]+]]
 func.exit:                                        ; preds = %while.body.i.func.exit_crit_edge, %while.cond.i.func.exit_crit_edge
-  %tmp3 = load i32* %x.i, align 4
-  %call = call i32 (i8*, ...)* @printf(i8* getelementptr inbounds ([6 x i8]* @.str, i64 0, i64 0), i32 %tmp3) nounwind
+  %tmp3 = load i32, i32* %x.i, align 4
+  %call = call i32 (i8*, ...) @printf(i8* getelementptr inbounds ([6 x i8], [6 x i8]* @.str, i64 0, i64 0), i32 %tmp3) nounwind
   ret i32 0
 }
 
diff --git a/test/Transforms/ScalarRepl/phi-select.ll b/test/Transforms/ScalarRepl/phi-select.ll
index a5da2dcf49d1..a6c7135a4925 100644
--- a/test/Transforms/ScalarRepl/phi-select.ll
+++ b/test/Transforms/ScalarRepl/phi-select.ll
@@ -13,14 +13,14 @@ define i32 @test1(i32 %x) nounwind readnone ssp {
 entry:
   %a = alloca %struct.X, align 8                  ; <%struct.X*> [#uses=2]
   %b = alloca %struct.X, align 8                  ; <%struct.X*> [#uses=2]
-  %0 = getelementptr inbounds %struct.X* %a, i64 0, i32 0 ; <i32*> [#uses=1]
+  %0 = getelementptr inbounds %struct.X, %struct.X* %a, i64 0, i32 0 ; <i32*> [#uses=1]
   store i32 1, i32* %0, align 8
-  %1 = getelementptr inbounds %struct.X* %b, i64 0, i32 0 ; <i32*> [#uses=1]
+  %1 = getelementptr inbounds %struct.X, %struct.X* %b, i64 0, i32 0 ; <i32*> [#uses=1]
   store i32 2, i32* %1, align 8
   %2 = icmp eq i32 %x, 0                          ; <i1> [#uses=1]
   %p.0 = select i1 %2, %struct.X* %b, %struct.X* %a ; <%struct.X*> [#uses=1]
-  %3 = getelementptr inbounds %struct.X* %p.0, i64 0, i32 0 ; <i32*> [#uses=1]
-  %4 = load i32* %3, align 8                      ; <i32> [#uses=1]
+  %3 = getelementptr inbounds %struct.X, %struct.X* %p.0, i64 0, i32 0 ; <i32*> [#uses=1]
+  %4 = load i32, i32* %3, align 8                      ; <i32> [#uses=1]
   ret i32 %4
 }
 
@@ -30,16 +30,16 @@ entry:
 define i32 @test2(i1 %c) {
 entry:
   %A = alloca {i32, i32}
-  %B = getelementptr {i32, i32}* %A, i32 0, i32 0
+  %B = getelementptr {i32, i32}, {i32, i32}* %A, i32 0, i32 0
   store i32 1, i32* %B
   br i1 %c, label %T, label %F
 T:
-  %C = getelementptr {i32, i32}* %A, i32 0, i32 1
+  %C = getelementptr {i32, i32}, {i32, i32}* %A, i32 0, i32 1
   store i32 2, i32* %C
   br label %F
 F:
   %X = phi i32* [%B, %entry], [%C, %T]
-  %Q = load i32* %X
+  %Q = load i32, i32* %X
   ret i32 %Q
 }
 
@@ -49,13 +49,13 @@ F:
 ; rdar://8904039
 define i32 @test3(i1 %c) {
   %A = alloca {i32, i32}
-  %B = getelementptr {i32, i32}* %A, i32 0, i32 0
+  %B = getelementptr {i32, i32}, {i32, i32}* %A, i32 0, i32 0
   store i32 1, i32* %B
-  %C = getelementptr {i32, i32}* %A, i32 0, i32 1
+  %C = getelementptr {i32, i32}, {i32, i32}* %A, i32 0, i32 1
   store i32 2, i32* %C
   
   %X = select i1 %c, i32* %B, i32* %C
-  %Q = load i32* %X
+  %Q = load i32, i32* %X
   ret i32 %Q
 }
 
@@ -65,14 +65,14 @@ entry:
   %A = alloca %PairTy
   ; CHECK-LABEL: @test4(
   ; CHECK: %A = alloca %PairTy
-  %B = getelementptr %PairTy* %A, i32 0, i32 0
+  %B = getelementptr %PairTy, %PairTy* %A, i32 0, i32 0
   store i32 1, i32* %B
-  %C = getelementptr %PairTy* %A, i32 0, i32 1
+  %C = getelementptr %PairTy, %PairTy* %A, i32 0, i32 1
   store i32 2, i32* %B
   
   %X = select i1 %c, i32* %B, i32* %C
   %Y = bitcast i32* %X to i64*
-  %Q = load i64* %Y
+  %Q = load i64, i64* %Y
   ret i64 %Q
 }
 
@@ -91,7 +91,7 @@ entry:
   %p.0 = select i1 false, i32* %b, i32* %P
   store i32 123, i32* %p.0
   
-  %r = load i32* %b, align 8
+  %r = load i32, i32* %b, align 8
   ret i32 %r
   
 ; CHECK-LABEL: @test5(
@@ -105,7 +105,7 @@ define i32 @test6(i32 %x, i1 %c) nounwind readnone ssp {
   store i32 1, i32* %a, align 8
   store i32 2, i32* %b, align 8
   %p.0 = select i1 %c, i32* %b, i32* %a
-  %r = load i32* %p.0, align 8
+  %r = load i32, i32* %p.0, align 8
   ret i32 %r
 ; CHECK-LABEL: @test6(
 ; CHECK-NEXT: %r = select i1 %c, i32 2, i32 1
@@ -122,7 +122,7 @@ define i32 @test7(i32 %x, i1 %c) nounwind readnone ssp {
   
   store i32 0, i32* %a
   
-  %r = load i32* %p.0, align 8
+  %r = load i32, i32* %p.0, align 8
   ret i32 %r
 ; CHECK-LABEL: @test7(
 ; CHECK-NOT: alloca i32
@@ -148,6 +148,6 @@ T:
   br label %Cont
 Cont:
   %p.0 = phi i32* [%b, %entry],[%a, %T]
-  %r = load i32* %p.0, align 8
+  %r = load i32, i32* %p.0, align 8
   ret i32 %r
 }
diff --git a/test/Transforms/ScalarRepl/phinodepromote.ll b/test/Transforms/ScalarRepl/phinodepromote.ll
index 9c6e8b92f3e5..c3af62485db1 100644
--- a/test/Transforms/ScalarRepl/phinodepromote.ll
+++ b/test/Transforms/ScalarRepl/phinodepromote.ll
@@ -21,14 +21,14 @@ entry:
 	%mem_tmp.1 = alloca i32		; <i32*> [#uses=3]
 	store i32 0, i32* %mem_tmp.0
 	store i32 1, i32* %mem_tmp.1
-	%tmp.1.i = load i32* %mem_tmp.1		; <i32> [#uses=1]
-	%tmp.3.i = load i32* %mem_tmp.0		; <i32> [#uses=1]
+	%tmp.1.i = load i32, i32* %mem_tmp.1		; <i32> [#uses=1]
+	%tmp.3.i = load i32, i32* %mem_tmp.0		; <i32> [#uses=1]
 	%tmp.4.i = icmp sle i32 %tmp.1.i, %tmp.3.i		; <i1> [#uses=1]
 	br i1 %tmp.4.i, label %cond_true.i, label %cond_continue.i
 cond_true.i:		; preds = %entry
 	br label %cond_continue.i
 cond_continue.i:		; preds = %cond_true.i, %entry
 	%mem_tmp.i.0 = phi i32* [ %mem_tmp.1, %cond_true.i ], [ %mem_tmp.0, %entry ]		; <i32*> [#uses=1]
-	%tmp.3 = load i32* %mem_tmp.i.0		; <i32> [#uses=1]
+	%tmp.3 = load i32, i32* %mem_tmp.i.0		; <i32> [#uses=1]
 	ret i32 %tmp.3
 }
diff --git a/test/Transforms/ScalarRepl/select_promote.ll b/test/Transforms/ScalarRepl/select_promote.ll
index d6b2b75327c5..b4ef8c46fa09 100644
--- a/test/Transforms/ScalarRepl/select_promote.ll
+++ b/test/Transforms/ScalarRepl/select_promote.ll
@@ -8,11 +8,11 @@ define i32 @main() {
 	%mem_tmp.1 = alloca i32		; <i32*> [#uses=3]
 	store i32 0, i32* %mem_tmp.0
 	store i32 1, i32* %mem_tmp.1
-	%tmp.1.i = load i32* %mem_tmp.1		; <i32> [#uses=1]
-	%tmp.3.i = load i32* %mem_tmp.0		; <i32> [#uses=1]
+	%tmp.1.i = load i32, i32* %mem_tmp.1		; <i32> [#uses=1]
+	%tmp.3.i = load i32, i32* %mem_tmp.0		; <i32> [#uses=1]
 	%tmp.4.i = icmp sle i32 %tmp.1.i, %tmp.3.i		; <i1> [#uses=1]
 	%mem_tmp.i.0 = select i1 %tmp.4.i, i32* %mem_tmp.1, i32* %mem_tmp.0		; <i32*> [#uses=1]
-	%tmp.3 = load i32* %mem_tmp.i.0		; <i32> [#uses=1]
+	%tmp.3 = load i32, i32* %mem_tmp.i.0		; <i32> [#uses=1]
 	ret i32 %tmp.3
 }
 
diff --git a/test/Transforms/ScalarRepl/sroa-fca.ll b/test/Transforms/ScalarRepl/sroa-fca.ll
index 2df3b9be1e48..c6e7c23ab066 100644
--- a/test/Transforms/ScalarRepl/sroa-fca.ll
+++ b/test/Transforms/ScalarRepl/sroa-fca.ll
@@ -6,7 +6,7 @@ define i64 @test({i32, i32} %A) {
 	%Y = bitcast i64* %X to {i32,i32}*
 	store {i32,i32} %A, {i32,i32}* %Y
 	
-	%Q = load i64* %X
+	%Q = load i64, i64* %X
 	ret i64 %Q
 }
 
@@ -15,7 +15,7 @@ define {i32,i32} @test2(i64 %A) {
 	%Y = bitcast i64* %X to {i32,i32}*
 	store i64 %A, i64* %X
 	
-	%Q = load {i32,i32}* %Y
+	%Q = load {i32,i32}, {i32,i32}* %Y
 	ret {i32,i32} %Q
 }
 
diff --git a/test/Transforms/ScalarRepl/sroa_two.ll b/test/Transforms/ScalarRepl/sroa_two.ll
index d8aa26dce637..f2285ef21a49 100644
--- a/test/Transforms/ScalarRepl/sroa_two.ll
+++ b/test/Transforms/ScalarRepl/sroa_two.ll
@@ -2,12 +2,12 @@
 
 define i32 @test(i32 %X) {
 	%Arr = alloca [2 x i32]		; <[2 x i32]*> [#uses=3]
-	%tmp.0 = getelementptr [2 x i32]* %Arr, i32 0, i32 0		; <i32*> [#uses=1]
+	%tmp.0 = getelementptr [2 x i32], [2 x i32]* %Arr, i32 0, i32 0		; <i32*> [#uses=1]
 	store i32 1, i32* %tmp.0
-	%tmp.1 = getelementptr [2 x i32]* %Arr, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp.1 = getelementptr [2 x i32], [2 x i32]* %Arr, i32 0, i32 1		; <i32*> [#uses=1]
 	store i32 2, i32* %tmp.1
-	%tmp.3 = getelementptr [2 x i32]* %Arr, i32 0, i32 %X		; <i32*> [#uses=1]
-	%tmp.4 = load i32* %tmp.3		; <i32> [#uses=1]
+	%tmp.3 = getelementptr [2 x i32], [2 x i32]* %Arr, i32 0, i32 %X		; <i32*> [#uses=1]
+	%tmp.4 = load i32, i32* %tmp.3		; <i32> [#uses=1]
 	ret i32 %tmp.4
 }
 
diff --git a/test/Transforms/ScalarRepl/union-fp-int.ll b/test/Transforms/ScalarRepl/union-fp-int.ll
index 6a4991887186..fa64b60685f4 100644
--- a/test/Transforms/ScalarRepl/union-fp-int.ll
+++ b/test/Transforms/ScalarRepl/union-fp-int.ll
@@ -8,7 +8,7 @@ define i32 @test(float %X) {
 	%X_addr = alloca float		; <float*> [#uses=2]
 	store float %X, float* %X_addr
 	%X_addr.upgrd.1 = bitcast float* %X_addr to i32*		; <i32*> [#uses=1]
-	%tmp = load i32* %X_addr.upgrd.1		; <i32> [#uses=1]
+	%tmp = load i32, i32* %X_addr.upgrd.1		; <i32> [#uses=1]
 	ret i32 %tmp
 }
 
diff --git a/test/Transforms/ScalarRepl/union-packed.ll b/test/Transforms/ScalarRepl/union-packed.ll
index b272abfc3d5b..741de76d581b 100644
--- a/test/Transforms/ScalarRepl/union-packed.ll
+++ b/test/Transforms/ScalarRepl/union-packed.ll
@@ -8,7 +8,7 @@ define <4 x i32> @test(<4 x float> %X) {
 	%X_addr = alloca <4 x float>		; <<4 x float>*> [#uses=2]
 	store <4 x float> %X, <4 x float>* %X_addr
 	%X_addr.upgrd.1 = bitcast <4 x float>* %X_addr to <4 x i32>*		; <<4 x i32>*> [#uses=1]
-	%tmp = load <4 x i32>* %X_addr.upgrd.1		; <<4 x i32>> [#uses=1]
+	%tmp = load <4 x i32>, <4 x i32>* %X_addr.upgrd.1		; <<4 x i32>> [#uses=1]
 	ret <4 x i32> %tmp
 }
 
diff --git a/test/Transforms/ScalarRepl/union-pointer.ll b/test/Transforms/ScalarRepl/union-pointer.ll
index f0dc141aca8d..6a5db1c4d6bd 100644
--- a/test/Transforms/ScalarRepl/union-pointer.ll
+++ b/test/Transforms/ScalarRepl/union-pointer.ll
@@ -14,7 +14,7 @@ define i8* @test(i16* %X) {
 	%X_addr = alloca i16*		; <i16**> [#uses=2]
 	store i16* %X, i16** %X_addr
 	%X_addr.upgrd.1 = bitcast i16** %X_addr to i8**		; <i8**> [#uses=1]
-	%tmp = load i8** %X_addr.upgrd.1		; <i8*> [#uses=1]
+	%tmp = load i8*, i8** %X_addr.upgrd.1		; <i8*> [#uses=1]
 	ret i8* %tmp
 }
 
@@ -26,7 +26,7 @@ define i8 addrspace(1)* @test_as1(i16 addrspace(1)* %x) {
     %x_addr = alloca i16 addrspace(1)*
 	store i16 addrspace(1)* %x, i16 addrspace(1)** %x_addr
 	%x_addr.upgrd.1 = bitcast i16 addrspace(1)** %x_addr to i8 addrspace(1)**
-	%tmp = load i8 addrspace(1)** %x_addr.upgrd.1
+	%tmp = load i8 addrspace(1)*, i8 addrspace(1)** %x_addr.upgrd.1
 	ret i8 addrspace(1)* %tmp
 }
 
@@ -36,10 +36,10 @@ define i8 addrspace(1)* @test_as1_array(i16 addrspace(1)* %x) {
 ; CHECK-NEXT: %2 = inttoptr i16 %1 to i8 addrspace(1)*
 ; CHECK-NEXT: ret i8 addrspace(1)* %2
   %as_ptr_array = alloca [4 x i16 addrspace(1)*]
-  %elem1 = getelementptr [4 x i16 addrspace(1)*]* %as_ptr_array, i32 0, i32 1
+  %elem1 = getelementptr [4 x i16 addrspace(1)*], [4 x i16 addrspace(1)*]* %as_ptr_array, i32 0, i32 1
   store i16 addrspace(1)* %x, i16 addrspace(1)** %elem1
   %elem1.cast = bitcast i16 addrspace(1)** %elem1 to i8 addrspace(1)**
-  %tmp = load i8 addrspace(1)** %elem1.cast
+  %tmp = load i8 addrspace(1)*, i8 addrspace(1)** %elem1.cast
   ret i8 addrspace(1)* %tmp
 }
 
@@ -54,17 +54,17 @@ define void @test2(i64 %Op.0) {
 	%tmp.upgrd.2 = call i64 @_Z3foov( )		; <i64> [#uses=1]
 	%tmp1.upgrd.3 = bitcast %struct.Val* %tmp1 to i64*		; <i64*> [#uses=1]
 	store i64 %tmp.upgrd.2, i64* %tmp1.upgrd.3
-	%tmp.upgrd.4 = getelementptr %struct.Val* %tmp, i32 0, i32 0		; <i32**> [#uses=1]
-	%tmp2 = getelementptr %struct.Val* %tmp1, i32 0, i32 0		; <i32**> [#uses=1]
-	%tmp.upgrd.5 = load i32** %tmp2		; <i32*> [#uses=1]
+	%tmp.upgrd.4 = getelementptr %struct.Val, %struct.Val* %tmp, i32 0, i32 0		; <i32**> [#uses=1]
+	%tmp2 = getelementptr %struct.Val, %struct.Val* %tmp1, i32 0, i32 0		; <i32**> [#uses=1]
+	%tmp.upgrd.5 = load i32*, i32** %tmp2		; <i32*> [#uses=1]
 	store i32* %tmp.upgrd.5, i32** %tmp.upgrd.4
-	%tmp3 = getelementptr %struct.Val* %tmp, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp4 = getelementptr %struct.Val* %tmp1, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp.upgrd.6 = load i32* %tmp4		; <i32> [#uses=1]
+	%tmp3 = getelementptr %struct.Val, %struct.Val* %tmp, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp4 = getelementptr %struct.Val, %struct.Val* %tmp1, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp.upgrd.6 = load i32, i32* %tmp4		; <i32> [#uses=1]
 	store i32 %tmp.upgrd.6, i32* %tmp3
 	%tmp7 = bitcast %struct.Val* %tmp to { i64 }*		; <{ i64 }*> [#uses=1]
-	%tmp8 = getelementptr { i64 }* %tmp7, i32 0, i32 0		; <i64*> [#uses=1]
-	%tmp9 = load i64* %tmp8		; <i64> [#uses=1]
+	%tmp8 = getelementptr { i64 }, { i64 }* %tmp7, i32 0, i32 0		; <i64*> [#uses=1]
+	%tmp9 = load i64, i64* %tmp8		; <i64> [#uses=1]
 	call void @_Z3bar3ValS_( i64 %Op.0, i64 %tmp9 )
 	ret void
 }
diff --git a/test/Transforms/ScalarRepl/vector_memcpy.ll b/test/Transforms/ScalarRepl/vector_memcpy.ll
index dfba9e2c8074..031ad5e38ca7 100644
--- a/test/Transforms/ScalarRepl/vector_memcpy.ll
+++ b/test/Transforms/ScalarRepl/vector_memcpy.ll
@@ -10,7 +10,7 @@ define <16 x float> @foo(<16 x float> %A) nounwind {
 	%s = bitcast <16 x float>* %tmp to i8*
 	%s2 = bitcast <16 x float>* %tmp2 to i8*
 	call void @llvm.memcpy.p0i8.p0i8.i64(i8* %s2, i8* %s, i64 64, i32 16, i1 false)
-	%R = load <16 x float>* %tmp2
+	%R = load <16 x float>, <16 x float>* %tmp2
 	ret <16 x float> %R
 }
 
@@ -20,7 +20,7 @@ define <16 x float> @foo2(<16 x float> %A) nounwind {
 	%s2 = bitcast <16 x float>* %tmp2 to i8*
 	call void @llvm.memset.p0i8.i64(i8* %s2, i8 0, i64 64, i32 16, i1 false)
 	
-	%R = load <16 x float>* %tmp2
+	%R = load <16 x float>, <16 x float>* %tmp2
 	ret <16 x float> %R
 }
 
diff --git a/test/Transforms/ScalarRepl/vector_promote.ll b/test/Transforms/ScalarRepl/vector_promote.ll
index 8ca1ed5005b3..3c2377fc0f28 100644
--- a/test/Transforms/ScalarRepl/vector_promote.ll
+++ b/test/Transforms/ScalarRepl/vector_promote.ll
@@ -5,18 +5,18 @@ target triple = "x86_64-apple-darwin10.0.0"
 define void @test1(<4 x float>* %F, float %f) {
 entry:
 	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=3]
-	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp = load <4 x float>, <4 x float>* %F		; <<4 x float>> [#uses=2]
 	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp3, <4 x float>* %G
-	%G.upgrd.1 = getelementptr <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
+	%G.upgrd.1 = getelementptr <4 x float>, <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
 	store float %f, float* %G.upgrd.1
-	%tmp4 = load <4 x float>* %G		; <<4 x float>> [#uses=2]
+	%tmp4 = load <4 x float>, <4 x float>* %G		; <<4 x float>> [#uses=2]
 	%tmp6 = fadd <4 x float> %tmp4, %tmp4		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp6, <4 x float>* %F
 	ret void
 ; CHECK-LABEL: @test1(
 ; CHECK-NOT: alloca
-; CHECK: %tmp = load <4 x float>* %F
+; CHECK: %tmp = load <4 x float>, <4 x float>* %F
 ; CHECK: fadd <4 x float> %tmp, %tmp
 ; CHECK-NEXT: insertelement <4 x float> %tmp3, float %f, i32 0
 }
@@ -24,18 +24,18 @@ entry:
 define void @test2(<4 x float>* %F, float %f) {
 entry:
 	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=3]
-	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp = load <4 x float>, <4 x float>* %F		; <<4 x float>> [#uses=2]
 	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp3, <4 x float>* %G
-	%tmp.upgrd.2 = getelementptr <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
+	%tmp.upgrd.2 = getelementptr <4 x float>, <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
 	store float %f, float* %tmp.upgrd.2
-	%tmp4 = load <4 x float>* %G		; <<4 x float>> [#uses=2]
+	%tmp4 = load <4 x float>, <4 x float>* %G		; <<4 x float>> [#uses=2]
 	%tmp6 = fadd <4 x float> %tmp4, %tmp4		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp6, <4 x float>* %F
 	ret void
 ; CHECK-LABEL: @test2(
 ; CHECK-NOT: alloca
-; CHECK: %tmp = load <4 x float>* %F
+; CHECK: %tmp = load <4 x float>, <4 x float>* %F
 ; CHECK: fadd <4 x float> %tmp, %tmp
 ; CHECK-NEXT: insertelement <4 x float> %tmp3, float %f, i32 2
 }
@@ -43,16 +43,16 @@ entry:
 define void @test3(<4 x float>* %F, float* %f) {
 entry:
 	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=2]
-	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp = load <4 x float>, <4 x float>* %F		; <<4 x float>> [#uses=2]
 	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp3, <4 x float>* %G
-	%tmp.upgrd.3 = getelementptr <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
-	%tmp.upgrd.4 = load float* %tmp.upgrd.3		; <float> [#uses=1]
+	%tmp.upgrd.3 = getelementptr <4 x float>, <4 x float>* %G, i32 0, i32 2		; <float*> [#uses=1]
+	%tmp.upgrd.4 = load float, float* %tmp.upgrd.3		; <float> [#uses=1]
 	store float %tmp.upgrd.4, float* %f
 	ret void
 ; CHECK-LABEL: @test3(
 ; CHECK-NOT: alloca
-; CHECK: %tmp = load <4 x float>* %F
+; CHECK: %tmp = load <4 x float>, <4 x float>* %F
 ; CHECK: fadd <4 x float> %tmp, %tmp
 ; CHECK-NEXT: extractelement <4 x float> %tmp3, i32 2
 }
@@ -60,26 +60,26 @@ entry:
 define void @test4(<4 x float>* %F, float* %f) {
 entry:
 	%G = alloca <4 x float>, align 16		; <<4 x float>*> [#uses=2]
-	%tmp = load <4 x float>* %F		; <<4 x float>> [#uses=2]
+	%tmp = load <4 x float>, <4 x float>* %F		; <<4 x float>> [#uses=2]
 	%tmp3 = fadd <4 x float> %tmp, %tmp		; <<4 x float>> [#uses=1]
 	store <4 x float> %tmp3, <4 x float>* %G
-	%G.upgrd.5 = getelementptr <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
-	%tmp.upgrd.6 = load float* %G.upgrd.5		; <float> [#uses=1]
+	%G.upgrd.5 = getelementptr <4 x float>, <4 x float>* %G, i32 0, i32 0		; <float*> [#uses=1]
+	%tmp.upgrd.6 = load float, float* %G.upgrd.5		; <float> [#uses=1]
 	store float %tmp.upgrd.6, float* %f
 	ret void
 ; CHECK-LABEL: @test4(
 ; CHECK-NOT: alloca
-; CHECK: %tmp = load <4 x float>* %F
+; CHECK: %tmp = load <4 x float>, <4 x float>* %F
 ; CHECK: fadd <4 x float> %tmp, %tmp
 ; CHECK-NEXT: extractelement <4 x float> %tmp3, i32 0
 }
 
 define i32 @test5(float %X) {  ;; should turn into bitcast.
 	%X_addr = alloca [4 x float]
-        %X1 = getelementptr [4 x float]* %X_addr, i32 0, i32 2
+        %X1 = getelementptr [4 x float], [4 x float]* %X_addr, i32 0, i32 2
 	store float %X, float* %X1
 	%a = bitcast float* %X1 to i32*
-	%tmp = load i32* %a
+	%tmp = load i32, i32* %a
 	ret i32 %tmp
 ; CHECK-LABEL: @test5(
 ; CHECK-NEXT: bitcast float %X to i32
@@ -90,7 +90,7 @@ define i64 @test6(<2 x float> %X) {
 	%X_addr = alloca <2 x float>
         store <2 x float> %X, <2 x float>* %X_addr
 	%P = bitcast <2 x float>* %X_addr to i64*
-	%tmp = load i64* %P
+	%tmp = load i64, i64* %P
 	ret i64 %tmp
 ; CHECK-LABEL: @test6(
 ; CHECK: bitcast <2 x float> %X to i64
@@ -104,7 +104,7 @@ entry:
   %memtmp = alloca %struct.test7, align 16
   %0 = bitcast %struct.test7* %memtmp to <4 x i32>*
   store <4 x i32> zeroinitializer, <4 x i32>* %0, align 16
-  %1 = getelementptr inbounds %struct.test7* %memtmp, i64 0, i32 0, i64 5
+  %1 = getelementptr inbounds %struct.test7, %struct.test7* %memtmp, i64 0, i32 0, i64 5
   store i32 0, i32* %1, align 4
   ret void
 ; CHECK-LABEL: @test7(
@@ -121,14 +121,14 @@ entry:
   %__a = alloca <1 x i64>, align 8
   %tmp = alloca <1 x i64>, align 8
   store <1 x i64> %a, <1 x i64>* %a.addr, align 8
-  %0 = load <1 x i64>* %a.addr, align 8
+  %0 = load <1 x i64>, <1 x i64>* %a.addr, align 8
   store <1 x i64> %0, <1 x i64>* %__a, align 8
-  %1 = load <1 x i64>* %__a, align 8
+  %1 = load <1 x i64>, <1 x i64>* %__a, align 8
   %2 = bitcast <1 x i64> %1 to <8 x i8>
   %3 = bitcast <8 x i8> %2 to <1 x i64>
   %vshl_n = shl <1 x i64> %3, <i64 4>
   store <1 x i64> %vshl_n, <1 x i64>* %tmp
-  %4 = load <1 x i64>* %tmp
+  %4 = load <1 x i64>, <1 x i64>* %tmp
   ret <1 x i64> %4
 ; CHECK-LABEL: @test8(
 ; CHECK-NOT: alloca
diff --git a/test/Transforms/ScalarRepl/vectors-with-mismatched-elements.ll b/test/Transforms/ScalarRepl/vectors-with-mismatched-elements.ll
index c3fbdf5f8632..154883122df6 100644
--- a/test/Transforms/ScalarRepl/vectors-with-mismatched-elements.ll
+++ b/test/Transforms/ScalarRepl/vectors-with-mismatched-elements.ll
@@ -18,10 +18,10 @@ define <2 x i64> @foo() nounwind {
 entry:
   %retval = alloca <3 x i32>, align 16
   %z = alloca <4 x i32>, align 16
-  %tmp = load <4 x i32>* %z
+  %tmp = load <4 x i32>, <4 x i32>* %z
   %tmp1 = shufflevector <4 x i32> %tmp, <4 x i32> undef, <3 x i32> <i32 0, i32 1, i32 2>
   store <3 x i32> %tmp1, <3 x i32>* %retval
   %0 = bitcast <3 x i32>* %retval to <2 x i64>*
-  %1 = load <2 x i64>* %0, align 1
+  %1 = load <2 x i64>, <2 x i64>* %0, align 1
   ret <2 x i64> %1
 }
diff --git a/test/Transforms/ScalarRepl/volatile.ll b/test/Transforms/ScalarRepl/volatile.ll
index d506cdfbd87a..2a600b3b2797 100644
--- a/test/Transforms/ScalarRepl/volatile.ll
+++ b/test/Transforms/ScalarRepl/volatile.ll
@@ -2,12 +2,12 @@
 
 define i32 @voltest(i32 %T) {
 	%A = alloca {i32, i32}
-	%B = getelementptr {i32,i32}* %A, i32 0, i32 0
+	%B = getelementptr {i32,i32}, {i32,i32}* %A, i32 0, i32 0
 	store volatile i32 %T, i32* %B
 ; CHECK: store volatile
 
-	%C = getelementptr {i32,i32}* %A, i32 0, i32 1
-	%X = load volatile i32* %C
+	%C = getelementptr {i32,i32}, {i32,i32}* %A, i32 0, i32 1
+	%X = load volatile i32, i32* %C
 ; CHECK: load volatile
 	ret i32 %X
 }
diff --git a/test/Transforms/Scalarizer/basic.ll b/test/Transforms/Scalarizer/basic.ll
index a94cf9f61142..150eb7d43c46 100644
--- a/test/Transforms/Scalarizer/basic.ll
+++ b/test/Transforms/Scalarizer/basic.ll
@@ -19,15 +19,15 @@ define void @f1(<4 x float> %init, <4 x float> *%base, i32 %count) {
 ; CHECK:   %acc.i2 = phi float [ %init.i2, %entry ], [ %sel.i2, %loop ]
 ; CHECK:   %acc.i3 = phi float [ %init.i3, %entry ], [ %sel.i3, %loop ]
 ; CHECK:   %nexti = sub i32 %i, 1
-; CHECK:   %ptr = getelementptr <4 x float>* %base, i32 %i
+; CHECK:   %ptr = getelementptr <4 x float>, <4 x float>* %base, i32 %i
 ; CHECK:   %ptr.i0 = bitcast <4 x float>* %ptr to float*
-; CHECK:   %val.i0 = load float* %ptr.i0, align 16
-; CHECK:   %ptr.i1 = getelementptr float* %ptr.i0, i32 1
-; CHECK:   %val.i1 = load float* %ptr.i1, align 4
-; CHECK:   %ptr.i2 = getelementptr float* %ptr.i0, i32 2
-; CHECK:   %val.i2 = load float* %ptr.i2, align 8
-; CHECK:   %ptr.i3 = getelementptr float* %ptr.i0, i32 3
-; CHECK:   %val.i3 = load float* %ptr.i3, align 4
+; CHECK:   %val.i0 = load float, float* %ptr.i0, align 16
+; CHECK:   %ptr.i1 = getelementptr float, float* %ptr.i0, i32 1
+; CHECK:   %val.i1 = load float, float* %ptr.i1, align 4
+; CHECK:   %ptr.i2 = getelementptr float, float* %ptr.i0, i32 2
+; CHECK:   %val.i2 = load float, float* %ptr.i2, align 8
+; CHECK:   %ptr.i3 = getelementptr float, float* %ptr.i0, i32 3
+; CHECK:   %val.i3 = load float, float* %ptr.i3, align 4
 ; CHECK:   %add.i0 = fadd float %val.i0, %val.i2
 ; CHECK:   %add.i1 = fadd float %val.i1, %val.i3
 ; CHECK:   %add.i2 = fadd float %acc.i0, %acc.i2
@@ -65,8 +65,8 @@ loop:
   %acc = phi <4 x float> [ %init, %entry ], [ %sel, %loop ]
   %nexti = sub i32 %i, 1
 
-  %ptr = getelementptr <4 x float> *%base, i32 %i
-  %val = load <4 x float> *%ptr
+  %ptr = getelementptr <4 x float>, <4 x float> *%base, i32 %i
+  %val = load <4 x float> , <4 x float> *%ptr
   %dval = bitcast <4 x float> %val to <2 x double>
   %dacc = bitcast <4 x float> %acc to <2 x double>
   %shuffle1 = shufflevector <2 x double> %dval, <2 x double> %dacc,
@@ -105,15 +105,15 @@ define void @f2(<4 x i32> %init, <4 x i8> *%base, i32 %count) {
 ; CHECK:   %acc.i2 = phi i32 [ %init.i2, %entry ], [ %sel.i2, %loop ]
 ; CHECK:   %acc.i3 = phi i32 [ %init.i3, %entry ], [ %sel.i3, %loop ]
 ; CHECK:   %nexti = sub i32 %i, 1
-; CHECK:   %ptr = getelementptr <4 x i8>* %base, i32 %i
+; CHECK:   %ptr = getelementptr <4 x i8>, <4 x i8>* %base, i32 %i
 ; CHECK:   %ptr.i0 = bitcast <4 x i8>* %ptr to i8*
-; CHECK:   %val.i0 = load i8* %ptr.i0, align 4
-; CHECK:   %ptr.i1 = getelementptr i8* %ptr.i0, i32 1
-; CHECK:   %val.i1 = load i8* %ptr.i1, align 1
-; CHECK:   %ptr.i2 = getelementptr i8* %ptr.i0, i32 2
-; CHECK:   %val.i2 = load i8* %ptr.i2, align 2
-; CHECK:   %ptr.i3 = getelementptr i8* %ptr.i0, i32 3
-; CHECK:   %val.i3 = load i8* %ptr.i3, align 1
+; CHECK:   %val.i0 = load i8, i8* %ptr.i0, align 4
+; CHECK:   %ptr.i1 = getelementptr i8, i8* %ptr.i0, i32 1
+; CHECK:   %val.i1 = load i8, i8* %ptr.i1, align 1
+; CHECK:   %ptr.i2 = getelementptr i8, i8* %ptr.i0, i32 2
+; CHECK:   %val.i2 = load i8, i8* %ptr.i2, align 2
+; CHECK:   %ptr.i3 = getelementptr i8, i8* %ptr.i0, i32 3
+; CHECK:   %val.i3 = load i8, i8* %ptr.i3, align 1
 ; CHECK:   %ext.i0 = sext i8 %val.i0 to i32
 ; CHECK:   %ext.i1 = sext i8 %val.i1 to i32
 ; CHECK:   %ext.i2 = sext i8 %val.i2 to i32
@@ -150,8 +150,8 @@ loop:
   %acc = phi <4 x i32> [ %init, %entry ], [ %sel, %loop ]
   %nexti = sub i32 %i, 1
 
-  %ptr = getelementptr <4 x i8> *%base, i32 %i
-  %val = load <4 x i8> *%ptr
+  %ptr = getelementptr <4 x i8>, <4 x i8> *%base, i32 %i
+  %val = load <4 x i8> , <4 x i8> *%ptr
   %ext = sext <4 x i8> %val to <4 x i32>
   %add = add <4 x i32> %ext, %acc
   %cmp = icmp slt <4 x i32> %add, <i32 -10, i32 -11, i32 -12, i32 -13>
@@ -172,16 +172,16 @@ exit:
 ; Check that !tbaa information is preserved.
 define void @f3(<4 x i32> *%src, <4 x i32> *%dst) {
 ; CHECK-LABEL: @f3(
-; CHECK: %val.i0 = load i32* %src.i0, align 16, !tbaa ![[TAG:[0-9]*]]
-; CHECK: %val.i1 = load i32* %src.i1, align 4, !tbaa ![[TAG]]
-; CHECK: %val.i2 = load i32* %src.i2, align 8, !tbaa ![[TAG]]
-; CHECK: %val.i3 = load i32* %src.i3, align 4, !tbaa ![[TAG]]
+; CHECK: %val.i0 = load i32, i32* %src.i0, align 16, !tbaa ![[TAG:[0-9]*]]
+; CHECK: %val.i1 = load i32, i32* %src.i1, align 4, !tbaa ![[TAG]]
+; CHECK: %val.i2 = load i32, i32* %src.i2, align 8, !tbaa ![[TAG]]
+; CHECK: %val.i3 = load i32, i32* %src.i3, align 4, !tbaa ![[TAG]]
 ; CHECK: store i32 %add.i0, i32* %dst.i0, align 16, !tbaa ![[TAG:[0-9]*]]
 ; CHECK: store i32 %add.i1, i32* %dst.i1, align 4, !tbaa ![[TAG]]
 ; CHECK: store i32 %add.i2, i32* %dst.i2, align 8, !tbaa ![[TAG]]
 ; CHECK: store i32 %add.i3, i32* %dst.i3, align 4, !tbaa ![[TAG]]
 ; CHECK: ret void
-  %val = load <4 x i32> *%src, !tbaa !1
+  %val = load <4 x i32> , <4 x i32> *%src, !tbaa !1
   %add = add <4 x i32> %val, %val
   store <4 x i32> %add, <4 x i32> *%dst, !tbaa !2
   ret void
@@ -190,16 +190,16 @@ define void @f3(<4 x i32> *%src, <4 x i32> *%dst) {
 ; Check that !tbaa.struct information is preserved.
 define void @f4(<4 x i32> *%src, <4 x i32> *%dst) {
 ; CHECK-LABEL: @f4(
-; CHECK: %val.i0 = load i32* %src.i0, align 16, !tbaa.struct ![[TAG:[0-9]*]]
-; CHECK: %val.i1 = load i32* %src.i1, align 4, !tbaa.struct ![[TAG]]
-; CHECK: %val.i2 = load i32* %src.i2, align 8, !tbaa.struct ![[TAG]]
-; CHECK: %val.i3 = load i32* %src.i3, align 4, !tbaa.struct ![[TAG]]
+; CHECK: %val.i0 = load i32, i32* %src.i0, align 16, !tbaa.struct ![[TAG:[0-9]*]]
+; CHECK: %val.i1 = load i32, i32* %src.i1, align 4, !tbaa.struct ![[TAG]]
+; CHECK: %val.i2 = load i32, i32* %src.i2, align 8, !tbaa.struct ![[TAG]]
+; CHECK: %val.i3 = load i32, i32* %src.i3, align 4, !tbaa.struct ![[TAG]]
 ; CHECK: store i32 %add.i0, i32* %dst.i0, align 16, !tbaa.struct ![[TAG]]
 ; CHECK: store i32 %add.i1, i32* %dst.i1, align 4, !tbaa.struct ![[TAG]]
 ; CHECK: store i32 %add.i2, i32* %dst.i2, align 8, !tbaa.struct ![[TAG]]
 ; CHECK: store i32 %add.i3, i32* %dst.i3, align 4, !tbaa.struct ![[TAG]]
 ; CHECK: ret void
-  %val = load <4 x i32> *%src, !tbaa.struct !5
+  %val = load <4 x i32> , <4 x i32> *%src, !tbaa.struct !5
   %add = add <4 x i32> %val, %val
   store <4 x i32> %add, <4 x i32> *%dst, !tbaa.struct !5
   ret void
@@ -208,10 +208,10 @@ define void @f4(<4 x i32> *%src, <4 x i32> *%dst) {
 ; Check that llvm.mem.parallel_loop_access information is preserved.
 define void @f5(i32 %count, <4 x i32> *%src, <4 x i32> *%dst) {
 ; CHECK-LABEL: @f5(
-; CHECK: %val.i0 = load i32* %this_src.i0, align 16, !llvm.mem.parallel_loop_access ![[TAG:[0-9]*]]
-; CHECK: %val.i1 = load i32* %this_src.i1, align 4, !llvm.mem.parallel_loop_access ![[TAG]]
-; CHECK: %val.i2 = load i32* %this_src.i2, align 8, !llvm.mem.parallel_loop_access ![[TAG]]
-; CHECK: %val.i3 = load i32* %this_src.i3, align 4, !llvm.mem.parallel_loop_access ![[TAG]]
+; CHECK: %val.i0 = load i32, i32* %this_src.i0, align 16, !llvm.mem.parallel_loop_access ![[TAG:[0-9]*]]
+; CHECK: %val.i1 = load i32, i32* %this_src.i1, align 4, !llvm.mem.parallel_loop_access ![[TAG]]
+; CHECK: %val.i2 = load i32, i32* %this_src.i2, align 8, !llvm.mem.parallel_loop_access ![[TAG]]
+; CHECK: %val.i3 = load i32, i32* %this_src.i3, align 4, !llvm.mem.parallel_loop_access ![[TAG]]
 ; CHECK: store i32 %add.i0, i32* %this_dst.i0, align 16, !llvm.mem.parallel_loop_access ![[TAG]]
 ; CHECK: store i32 %add.i1, i32* %this_dst.i1, align 4, !llvm.mem.parallel_loop_access ![[TAG]]
 ; CHECK: store i32 %add.i2, i32* %this_dst.i2, align 8, !llvm.mem.parallel_loop_access ![[TAG]]
@@ -222,9 +222,9 @@ entry:
 
 loop:
   %index = phi i32 [ 0, %entry ], [ %next_index, %loop ]
-  %this_src = getelementptr <4 x i32> *%src, i32 %index
-  %this_dst = getelementptr <4 x i32> *%dst, i32 %index
-  %val = load <4 x i32> *%this_src, !llvm.mem.parallel_loop_access !3
+  %this_src = getelementptr <4 x i32>, <4 x i32> *%src, i32 %index
+  %this_dst = getelementptr <4 x i32>, <4 x i32> *%dst, i32 %index
+  %val = load <4 x i32> , <4 x i32> *%this_src, !llvm.mem.parallel_loop_access !3
   %add = add <4 x i32> %val, %val
   store <4 x i32> %add, <4 x i32> *%this_dst, !llvm.mem.parallel_loop_access !3
   %next_index = add i32 %index, -1
@@ -261,7 +261,7 @@ define void @f7(<4 x i32> *%src, <4 x i32> *%dst) {
 ; CHECK-LABEL: @f7(
 ; CHECK-NOT: !foo
 ; CHECK: ret void
-  %val = load <4 x i32> *%src, !foo !5
+  %val = load <4 x i32> , <4 x i32> *%src, !foo !5
   %add = add <4 x i32> %val, %val
   store <4 x i32> %add, <4 x i32> *%dst, !foo !5
   ret void
@@ -272,18 +272,18 @@ define void @f8(<4 x float *> *%dest, <4 x float *> %ptr0, <4 x i32> %i0,
                 float *%other) {
 ; CHECK-LABEL: @f8(
 ; CHECK: %dest.i0 = bitcast <4 x float*>* %dest to float**
-; CHECK: %dest.i1 = getelementptr float** %dest.i0, i32 1
-; CHECK: %dest.i2 = getelementptr float** %dest.i0, i32 2
-; CHECK: %dest.i3 = getelementptr float** %dest.i0, i32 3
+; CHECK: %dest.i1 = getelementptr float*, float** %dest.i0, i32 1
+; CHECK: %dest.i2 = getelementptr float*, float** %dest.i0, i32 2
+; CHECK: %dest.i3 = getelementptr float*, float** %dest.i0, i32 3
 ; CHECK: %i0.i1 = extractelement <4 x i32> %i0, i32 1
 ; CHECK: %i0.i3 = extractelement <4 x i32> %i0, i32 3
 ; CHECK: %ptr0.i0 = extractelement <4 x float*> %ptr0, i32 0
-; CHECK: %val.i0 = getelementptr float* %ptr0.i0, i32 100
-; CHECK: %val.i1 = getelementptr float* %other, i32 %i0.i1
+; CHECK: %val.i0 = getelementptr float, float* %ptr0.i0, i32 100
+; CHECK: %val.i1 = getelementptr float, float* %other, i32 %i0.i1
 ; CHECK: %ptr0.i2 = extractelement <4 x float*> %ptr0, i32 2
-; CHECK: %val.i2 = getelementptr float* %ptr0.i2, i32 100
+; CHECK: %val.i2 = getelementptr float, float* %ptr0.i2, i32 100
 ; CHECK: %ptr0.i3 = extractelement <4 x float*> %ptr0, i32 3
-; CHECK: %val.i3 = getelementptr float* %ptr0.i3, i32 %i0.i3
+; CHECK: %val.i3 = getelementptr float, float* %ptr0.i3, i32 %i0.i3
 ; CHECK: store float* %val.i0, float** %dest.i0, align 32
 ; CHECK: store float* %val.i1, float** %dest.i1, align 8
 ; CHECK: store float* %val.i2, float** %dest.i2, align 16
@@ -292,7 +292,7 @@ define void @f8(<4 x float *> *%dest, <4 x float *> %ptr0, <4 x i32> %i0,
   %i1 = insertelement <4 x i32> %i0, i32 100, i32 0
   %i2 = insertelement <4 x i32> %i1, i32 100, i32 2
   %ptr1 = insertelement <4 x float *> %ptr0, float *%other, i32 1
-  %val = getelementptr <4 x float *> %ptr1, <4 x i32> %i2
+  %val = getelementptr float, <4 x float *> %ptr1, <4 x i32> %i2
   store <4 x float *> %val, <4 x float *> *%dest
   ret void
 }
@@ -301,23 +301,23 @@ define void @f8(<4 x float *> *%dest, <4 x float *> %ptr0, <4 x i32> %i0,
 define void @f9(<4 x float> *%dest, <4 x float> *%src) {
 ; CHECK: @f9(
 ; CHECK: %dest.i0 = bitcast <4 x float>* %dest to float*
-; CHECK: %dest.i1 = getelementptr float* %dest.i0, i32 1
-; CHECK: %dest.i2 = getelementptr float* %dest.i0, i32 2
-; CHECK: %dest.i3 = getelementptr float* %dest.i0, i32 3
+; CHECK: %dest.i1 = getelementptr float, float* %dest.i0, i32 1
+; CHECK: %dest.i2 = getelementptr float, float* %dest.i0, i32 2
+; CHECK: %dest.i3 = getelementptr float, float* %dest.i0, i32 3
 ; CHECK: %src.i0 = bitcast <4 x float>* %src to float*
-; CHECK: %val.i0 = load float* %src.i0, align 4
-; CHECK: %src.i1 = getelementptr float* %src.i0, i32 1
-; CHECK: %val.i1 = load float* %src.i1, align 4
-; CHECK: %src.i2 = getelementptr float* %src.i0, i32 2
-; CHECK: %val.i2 = load float* %src.i2, align 4
-; CHECK: %src.i3 = getelementptr float* %src.i0, i32 3
-; CHECK: %val.i3 = load float* %src.i3, align 4
+; CHECK: %val.i0 = load float, float* %src.i0, align 4
+; CHECK: %src.i1 = getelementptr float, float* %src.i0, i32 1
+; CHECK: %val.i1 = load float, float* %src.i1, align 4
+; CHECK: %src.i2 = getelementptr float, float* %src.i0, i32 2
+; CHECK: %val.i2 = load float, float* %src.i2, align 4
+; CHECK: %src.i3 = getelementptr float, float* %src.i0, i32 3
+; CHECK: %val.i3 = load float, float* %src.i3, align 4
 ; CHECK: store float %val.i0, float* %dest.i0, align 8
 ; CHECK: store float %val.i1, float* %dest.i1, align 4
 ; CHECK: store float %val.i2, float* %dest.i2, align 8
 ; CHECK: store float %val.i3, float* %dest.i3, align 4
 ; CHECK: ret void
-  %val = load <4 x float> *%src, align 4
+  %val = load <4 x float> , <4 x float> *%src, align 4
   store <4 x float> %val, <4 x float> *%dest, align 8
   ret void
 }
@@ -326,23 +326,23 @@ define void @f9(<4 x float> *%dest, <4 x float> *%src) {
 define void @f10(<4 x float> *%dest, <4 x float> *%src) {
 ; CHECK: @f10(
 ; CHECK: %dest.i0 = bitcast <4 x float>* %dest to float*
-; CHECK: %dest.i1 = getelementptr float* %dest.i0, i32 1
-; CHECK: %dest.i2 = getelementptr float* %dest.i0, i32 2
-; CHECK: %dest.i3 = getelementptr float* %dest.i0, i32 3
+; CHECK: %dest.i1 = getelementptr float, float* %dest.i0, i32 1
+; CHECK: %dest.i2 = getelementptr float, float* %dest.i0, i32 2
+; CHECK: %dest.i3 = getelementptr float, float* %dest.i0, i32 3
 ; CHECK: %src.i0 = bitcast <4 x float>* %src to float*
-; CHECK: %val.i0 = load float* %src.i0, align 1
-; CHECK: %src.i1 = getelementptr float* %src.i0, i32 1
-; CHECK: %val.i1 = load float* %src.i1, align 1
-; CHECK: %src.i2 = getelementptr float* %src.i0, i32 2
-; CHECK: %val.i2 = load float* %src.i2, align 1
-; CHECK: %src.i3 = getelementptr float* %src.i0, i32 3
-; CHECK: %val.i3 = load float* %src.i3, align 1
+; CHECK: %val.i0 = load float, float* %src.i0, align 1
+; CHECK: %src.i1 = getelementptr float, float* %src.i0, i32 1
+; CHECK: %val.i1 = load float, float* %src.i1, align 1
+; CHECK: %src.i2 = getelementptr float, float* %src.i0, i32 2
+; CHECK: %val.i2 = load float, float* %src.i2, align 1
+; CHECK: %src.i3 = getelementptr float, float* %src.i0, i32 3
+; CHECK: %val.i3 = load float, float* %src.i3, align 1
 ; CHECK: store float %val.i0, float* %dest.i0, align 2
 ; CHECK: store float %val.i1, float* %dest.i1, align 2
 ; CHECK: store float %val.i2, float* %dest.i2, align 2
 ; CHECK: store float %val.i3, float* %dest.i3, align 2
 ; CHECK: ret void
-  %val = load <4 x float> *%src, align 1
+  %val = load <4 x float> , <4 x float> *%src, align 1
   store <4 x float> %val, <4 x float> *%dest, align 2
   ret void
 }
@@ -350,13 +350,13 @@ define void @f10(<4 x float> *%dest, <4 x float> *%src) {
 ; Test that sub-byte loads aren't scalarized.
 define void @f11(<32 x i1> *%dest, <32 x i1> *%src0) {
 ; CHECK: @f11(
-; CHECK: %val0 = load <32 x i1>* %src0
-; CHECK: %val1 = load <32 x i1>* %src1
+; CHECK: %val0 = load <32 x i1>, <32 x i1>* %src0
+; CHECK: %val1 = load <32 x i1>, <32 x i1>* %src1
 ; CHECK: store <32 x i1> %and, <32 x i1>* %dest
 ; CHECK: ret void
-  %src1 = getelementptr <32 x i1> *%src0, i32 1
-  %val0 = load <32 x i1> *%src0
-  %val1 = load <32 x i1> *%src1
+  %src1 = getelementptr <32 x i1>, <32 x i1> *%src0, i32 1
+  %val0 = load <32 x i1> , <32 x i1> *%src0
+  %val1 = load <32 x i1> , <32 x i1> *%src1
   %and = and <32 x i1> %val0, %val1
   store <32 x i1> %and, <32 x i1> *%dest
   ret void
@@ -375,7 +375,7 @@ define void @f12(<4 x i32> *%dest, <4 x i32> *%src, i32 %index) {
 ; CHECK-DAG: %val2.i2 = shl i32 3, %val1.i2
 ; CHECK-DAG: %val2.i3 = shl i32 4, %val1.i3
 ; CHECK: ret void
-  %val0 = load <4 x i32> *%src
+  %val0 = load <4 x i32> , <4 x i32> *%src
   %val1 = insertelement <4 x i32> %val0, i32 1, i32 %index
   %val2 = shl <4 x i32> <i32 1, i32 2, i32 3, i32 4>, %val1
   store <4 x i32> %val2, <4 x i32> *%dest
@@ -387,27 +387,27 @@ define void @f13(<4 x float *> *%dest, <4 x [4 x float] *> %ptr, <4 x i32> %i,
                  float *%other) {
 ; CHECK-LABEL: @f13(
 ; CHECK: %dest.i0 = bitcast <4 x float*>* %dest to float**
-; CHECK: %dest.i1 = getelementptr float** %dest.i0, i32 1
-; CHECK: %dest.i2 = getelementptr float** %dest.i0, i32 2
-; CHECK: %dest.i3 = getelementptr float** %dest.i0, i32 3
+; CHECK: %dest.i1 = getelementptr float*, float** %dest.i0, i32 1
+; CHECK: %dest.i2 = getelementptr float*, float** %dest.i0, i32 2
+; CHECK: %dest.i3 = getelementptr float*, float** %dest.i0, i32 3
 ; CHECK: %i.i0 = extractelement <4 x i32> %i, i32 0
 ; CHECK: %ptr.i0 = extractelement <4 x [4 x float]*> %ptr, i32 0
-; CHECK: %val.i0 = getelementptr inbounds [4 x float]* %ptr.i0, i32 0, i32 %i.i0
+; CHECK: %val.i0 = getelementptr inbounds [4 x float], [4 x float]* %ptr.i0, i32 0, i32 %i.i0
 ; CHECK: %i.i1 = extractelement <4 x i32> %i, i32 1
 ; CHECK: %ptr.i1 = extractelement <4 x [4 x float]*> %ptr, i32 1
-; CHECK: %val.i1 = getelementptr inbounds [4 x float]* %ptr.i1, i32 1, i32 %i.i1
+; CHECK: %val.i1 = getelementptr inbounds [4 x float], [4 x float]* %ptr.i1, i32 1, i32 %i.i1
 ; CHECK: %i.i2 = extractelement <4 x i32> %i, i32 2
 ; CHECK: %ptr.i2 = extractelement <4 x [4 x float]*> %ptr, i32 2
-; CHECK: %val.i2 = getelementptr inbounds [4 x float]* %ptr.i2, i32 2, i32 %i.i2
+; CHECK: %val.i2 = getelementptr inbounds [4 x float], [4 x float]* %ptr.i2, i32 2, i32 %i.i2
 ; CHECK: %i.i3 = extractelement <4 x i32> %i, i32 3
 ; CHECK: %ptr.i3 = extractelement <4 x [4 x float]*> %ptr, i32 3
-; CHECK: %val.i3 = getelementptr inbounds [4 x float]* %ptr.i3, i32 3, i32 %i.i3
+; CHECK: %val.i3 = getelementptr inbounds [4 x float], [4 x float]* %ptr.i3, i32 3, i32 %i.i3
 ; CHECK: store float* %val.i0, float** %dest.i0, align 32
 ; CHECK: store float* %val.i1, float** %dest.i1, align 8
 ; CHECK: store float* %val.i2, float** %dest.i2, align 16
 ; CHECK: store float* %val.i3, float** %dest.i3, align 8
 ; CHECK: ret void
-  %val = getelementptr inbounds <4 x [4 x float] *> %ptr,
+  %val = getelementptr inbounds [4 x float], <4 x [4 x float] *> %ptr,
                                 <4 x i32> <i32 0, i32 1, i32 2, i32 3>,
                                 <4 x i32> %i
   store <4 x float *> %val, <4 x float *> *%dest
diff --git a/test/Transforms/Scalarizer/dbginfo.ll b/test/Transforms/Scalarizer/dbginfo.ll
index ed65aaace2c4..3770a3e8c642 100644
--- a/test/Transforms/Scalarizer/dbginfo.ll
+++ b/test/Transforms/Scalarizer/dbginfo.ll
@@ -5,28 +5,28 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 define void @f1(<4 x i32>* nocapture %a, <4 x i32>* nocapture readonly %b, <4 x i32>* nocapture readonly %c) #0 {
 ; CHECK: @f1(
 ; CHECK: %a.i0 = bitcast <4 x i32>* %a to i32*
-; CHECK: %a.i1 = getelementptr i32* %a.i0, i32 1
-; CHECK: %a.i2 = getelementptr i32* %a.i0, i32 2
-; CHECK: %a.i3 = getelementptr i32* %a.i0, i32 3
+; CHECK: %a.i1 = getelementptr i32, i32* %a.i0, i32 1
+; CHECK: %a.i2 = getelementptr i32, i32* %a.i0, i32 2
+; CHECK: %a.i3 = getelementptr i32, i32* %a.i0, i32 3
 ; CHECK: %c.i0 = bitcast <4 x i32>* %c to i32*
-; CHECK: %c.i1 = getelementptr i32* %c.i0, i32 1
-; CHECK: %c.i2 = getelementptr i32* %c.i0, i32 2
-; CHECK: %c.i3 = getelementptr i32* %c.i0, i32 3
+; CHECK: %c.i1 = getelementptr i32, i32* %c.i0, i32 1
+; CHECK: %c.i2 = getelementptr i32, i32* %c.i0, i32 2
+; CHECK: %c.i3 = getelementptr i32, i32* %c.i0, i32 3
 ; CHECK: %b.i0 = bitcast <4 x i32>* %b to i32*
-; CHECK: %b.i1 = getelementptr i32* %b.i0, i32 1
-; CHECK: %b.i2 = getelementptr i32* %b.i0, i32 2
-; CHECK: %b.i3 = getelementptr i32* %b.i0, i32 3
+; CHECK: %b.i1 = getelementptr i32, i32* %b.i0, i32 1
+; CHECK: %b.i2 = getelementptr i32, i32* %b.i0, i32 2
+; CHECK: %b.i3 = getelementptr i32, i32* %b.i0, i32 3
 ; CHECK: tail call void @llvm.dbg.value(metadata <4 x i32>* %a, i64 0, metadata !{{[0-9]+}}, metadata {{.*}}), !dbg !{{[0-9]+}}
 ; CHECK: tail call void @llvm.dbg.value(metadata <4 x i32>* %b, i64 0, metadata !{{[0-9]+}}, metadata {{.*}}), !dbg !{{[0-9]+}}
 ; CHECK: tail call void @llvm.dbg.value(metadata <4 x i32>* %c, i64 0, metadata !{{[0-9]+}}, metadata {{.*}}), !dbg !{{[0-9]+}}
-; CHECK: %bval.i0 = load i32* %b.i0, align 16, !dbg ![[TAG1:[0-9]+]], !tbaa ![[TAG2:[0-9]+]]
-; CHECK: %bval.i1 = load i32* %b.i1, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %bval.i2 = load i32* %b.i2, align 8, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %bval.i3 = load i32* %b.i3, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %cval.i0 = load i32* %c.i0, align 16, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %cval.i1 = load i32* %c.i1, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %cval.i2 = load i32* %c.i2, align 8, !dbg ![[TAG1]], !tbaa ![[TAG2]]
-; CHECK: %cval.i3 = load i32* %c.i3, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %bval.i0 = load i32, i32* %b.i0, align 16, !dbg ![[TAG1:[0-9]+]], !tbaa ![[TAG2:[0-9]+]]
+; CHECK: %bval.i1 = load i32, i32* %b.i1, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %bval.i2 = load i32, i32* %b.i2, align 8, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %bval.i3 = load i32, i32* %b.i3, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %cval.i0 = load i32, i32* %c.i0, align 16, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %cval.i1 = load i32, i32* %c.i1, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %cval.i2 = load i32, i32* %c.i2, align 8, !dbg ![[TAG1]], !tbaa ![[TAG2]]
+; CHECK: %cval.i3 = load i32, i32* %c.i3, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
 ; CHECK: %add.i0 = add i32 %bval.i0, %cval.i0, !dbg ![[TAG1]]
 ; CHECK: %add.i1 = add i32 %bval.i1, %cval.i1, !dbg ![[TAG1]]
 ; CHECK: %add.i2 = add i32 %bval.i2, %cval.i2, !dbg ![[TAG1]]
@@ -37,11 +37,11 @@ define void @f1(<4 x i32>* nocapture %a, <4 x i32>* nocapture readonly %b, <4 x
 ; CHECK: store i32 %add.i3, i32* %a.i3, align 4, !dbg ![[TAG1]], !tbaa ![[TAG2]]
 ; CHECK: ret void
 entry:
-  tail call void @llvm.dbg.value(metadata <4 x i32>* %a, i64 0, metadata !15, metadata !{}), !dbg !20
-  tail call void @llvm.dbg.value(metadata <4 x i32>* %b, i64 0, metadata !16, metadata !{}), !dbg !20
-  tail call void @llvm.dbg.value(metadata <4 x i32>* %c, i64 0, metadata !17, metadata !{}), !dbg !20
-  %bval = load <4 x i32>* %b, align 16, !dbg !21, !tbaa !22
-  %cval = load <4 x i32>* %c, align 16, !dbg !21, !tbaa !22
+  tail call void @llvm.dbg.value(metadata <4 x i32>* %a, i64 0, metadata !15, metadata !DIExpression()), !dbg !20
+  tail call void @llvm.dbg.value(metadata <4 x i32>* %b, i64 0, metadata !16, metadata !DIExpression()), !dbg !20
+  tail call void @llvm.dbg.value(metadata <4 x i32>* %c, i64 0, metadata !17, metadata !DIExpression()), !dbg !20
+  %bval = load <4 x i32>, <4 x i32>* %b, align 16, !dbg !21, !tbaa !22
+  %cval = load <4 x i32>, <4 x i32>* %c, align 16, !dbg !21, !tbaa !22
   %add = add <4 x i32> %bval, %cval, !dbg !21
   store <4 x i32> %add, <4 x i32>* %a, align 16, !dbg !21, !tbaa !22
   ret void, !dbg !25
@@ -57,30 +57,30 @@ attributes #1 = { nounwind readnone }
 !llvm.module.flags = !{!18, !26}
 !llvm.ident = !{!19}
 
-!0 = !{!"0x11\0012\00clang version 3.4 (trunk 194134) (llvm/trunk 194126)\001\00\000\00\000", !1, !2, !2, !3, !2, !2} ; [ DW_TAG_compile_unit ] [/home/richards/llvm/build//tmp/add.c] [DW_LANG_C99]
-!1 = !{!"/tmp/add.c", !"/home/richards/llvm/build"}
-!2 = !{i32 0}
+!0 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 3.4 (trunk 194134) (llvm/trunk 194126)", isOptimized: true, emissionKind: 0, file: !1, enums: !2, retainedTypes: !2, subprograms: !3, globals: !2, imports: !2)
+!1 = !DIFile(filename: "/tmp/add.c", directory: "/home/richards/llvm/build")
+!2 = !{}
 !3 = !{!4}
-!4 = !{!"0x2e\00f1\00f1\00\003\000\001\000\006\00256\001\004", !1, !5, !6, null, void (<4 x i32>*, <4 x i32>*, <4 x i32>*)* @f1, null, null, !14} ; [ DW_TAG_subprogram ] [line 3] [def] [scope 4] [f]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [/home/richards/llvm/build//tmp/add.c]
-!6 = !{!"0x15\00\000\000\000\000\000\000", i32 0, null, null, !7, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!4 = !DISubprogram(name: "f1", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: true, scopeLine: 4, file: !1, scope: !5, type: !6, function: void (<4 x i32>*, <4 x i32>*, <4 x i32>*)* @f1, variables: !14)
+!5 = !DIFile(filename: "/tmp/add.c", directory: "/home/richards/llvm/build")
+!6 = !DISubroutineType(types: !7)
 !7 = !{null, !8, !8, !8}
-!8 = !{!"0xf\00\000\0064\0064\000\000", null, null, !9} ; [ DW_TAG_pointer_type ] [line 0, size 64, align 64, offset 0] [from V4SI]
-!9 = !{!"0x16\00V4SI\001\000\000\000\000", !1, null, !10} ; [ DW_TAG_typedef ] [V4SI] [line 1, size 0, align 0, offset 0] [from ]
-!10 = !{!"0x1\00\000\00128\00128\000\002048", null, null, !11, !12, i32 0, null, null, null} ; [ DW_TAG_array_type ] [line 0, size 128, align 128, offset 0] [vector] [from int]
-!11 = !{!"0x24\00int\000\0032\0032\000\000\005", null, null} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
+!8 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 64, align: 64, baseType: !9)
+!9 = !DIDerivedType(tag: DW_TAG_typedef, name: "V4SI", line: 1, file: !1, baseType: !10)
+!10 = !DICompositeType(tag: DW_TAG_array_type, size: 128, align: 128, flags: DIFlagVector, baseType: !11, elements: !12)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
 !12 = !{!13}
-!13 = !{!"0x21\000\004"}        ; [ DW_TAG_subrange_type ] [0, 3]
+!13 = !DISubrange(count: 4)
 !14 = !{!15, !16, !17}
-!15 = !{!"0x101\00a\0016777219\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [a] [line 3]
-!16 = !{!"0x101\00b\0033554435\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [b] [line 3]
-!17 = !{!"0x101\00c\0050331651\000", !4, !5, !8} ; [ DW_TAG_arg_variable ] [c] [line 3]
+!15 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "a", line: 3, arg: 1, scope: !4, file: !5, type: !8)
+!16 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "b", line: 3, arg: 2, scope: !4, file: !5, type: !8)
+!17 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "c", line: 3, arg: 3, scope: !4, file: !5, type: !8)
 !18 = !{i32 2, !"Dwarf Version", i32 4}
 !19 = !{!"clang version 3.4 (trunk 194134) (llvm/trunk 194126)"}
-!20 = !MDLocation(line: 3, scope: !4)
-!21 = !MDLocation(line: 5, scope: !4)
+!20 = !DILocation(line: 3, scope: !4)
+!21 = !DILocation(line: 5, scope: !4)
 !22 = !{!23, !23, i64 0}
 !23 = !{!"omnipotent char", !24, i64 0}
 !24 = !{!"Simple C/C++ TBAA"}
-!25 = !MDLocation(line: 6, scope: !4)
-!26 = !{i32 1, !"Debug Info Version", i32 2}
+!25 = !DILocation(line: 6, scope: !4)
+!26 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/Scalarizer/no-data-layout.ll b/test/Transforms/Scalarizer/no-data-layout.ll
deleted file mode 100644
index 3eaf669ef9dd..000000000000
--- a/test/Transforms/Scalarizer/no-data-layout.ll
+++ /dev/null
@@ -1,25 +0,0 @@
-; RUN: opt %s -scalarizer -scalarize-load-store -S | FileCheck %s
-
-; Test the handling of loads and stores when no data layout is available.
-define void @f1(<4 x float> *%dest, <4 x float> *%src) {
-; CHECK: @f1(
-; CHECK: %val = load <4 x float>* %src, align 4
-; CHECK: %val.i0 = extractelement <4 x float> %val, i32 0
-; CHECK: %add.i0 = fadd float %val.i0, %val.i0
-; CHECK: %val.i1 = extractelement <4 x float> %val, i32 1
-; CHECK: %add.i1 = fadd float %val.i1, %val.i1
-; CHECK: %val.i2 = extractelement <4 x float> %val, i32 2
-; CHECK: %add.i2 = fadd float %val.i2, %val.i2
-; CHECK: %val.i3 = extractelement <4 x float> %val, i32 3
-; CHECK: %add.i3 = fadd float %val.i3, %val.i3
-; CHECK: %add.upto0 = insertelement <4 x float> undef, float %add.i0, i32 0
-; CHECK: %add.upto1 = insertelement <4 x float> %add.upto0, float %add.i1, i32 1
-; CHECK: %add.upto2 = insertelement <4 x float> %add.upto1, float %add.i2, i32 2
-; CHECK: %add = insertelement <4 x float> %add.upto2, float %add.i3, i32 3
-; CHECK: store <4 x float> %add, <4 x float>* %dest, align 8
-; CHECK: ret void
-  %val = load <4 x float> *%src, align 4
-  %add = fadd <4 x float> %val, %val
-  store <4 x float> %add, <4 x float> *%dest, align 8
-  ret void
-}
diff --git a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
index d054a3b1b9f2..073313d40e77 100644
--- a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
+++ b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep-and-gvn.ll
@@ -1,5 +1,5 @@
 ; RUN: llc < %s -march=nvptx64 -mcpu=sm_20 | FileCheck %s --check-prefix=PTX
-; RUN: opt < %s -S -separate-const-offset-from-gep -gvn -dce | FileCheck %s --check-prefix=IR
+; RUN: opt < %s -S -separate-const-offset-from-gep -reassociate-geps-verify-no-dead-code -gvn | FileCheck %s --check-prefix=IR
 
 ; Verifies the SeparateConstOffsetFromGEP pass.
 ; The following code computes
@@ -21,25 +21,25 @@ define void @sum_of_array(i32 %x, i32 %y, float* nocapture %output) {
 .preheader:
   %0 = sext i32 %y to i64
   %1 = sext i32 %x to i64
-  %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+  %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
   %3 = addrspacecast float addrspace(3)* %2 to float*
-  %4 = load float* %3, align 4
+  %4 = load float, float* %3, align 4
   %5 = fadd float %4, 0.000000e+00
   %6 = add i32 %y, 1
   %7 = sext i32 %6 to i64
-  %8 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
+  %8 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
   %9 = addrspacecast float addrspace(3)* %8 to float*
-  %10 = load float* %9, align 4
+  %10 = load float, float* %9, align 4
   %11 = fadd float %5, %10
   %12 = add i32 %x, 1
   %13 = sext i32 %12 to i64
-  %14 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
+  %14 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
   %15 = addrspacecast float addrspace(3)* %14 to float*
-  %16 = load float* %15, align 4
+  %16 = load float, float* %15, align 4
   %17 = fadd float %11, %16
-  %18 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
+  %18 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
   %19 = addrspacecast float addrspace(3)* %18 to float*
-  %20 = load float* %19, align 4
+  %20 = load float, float* %19, align 4
   %21 = fadd float %17, %20
   store float %21, float* %output, align 4
   ret void
@@ -51,10 +51,10 @@ define void @sum_of_array(i32 %x, i32 %y, float* nocapture %output) {
 ; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
 
 ; IR-LABEL: @sum_of_array(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
 
 ; @sum_of_array2 is very similar to @sum_of_array. The only difference is in
 ; the order of "sext" and "add" when computing the array indices. @sum_of_array
@@ -66,23 +66,23 @@ define void @sum_of_array2(i32 %x, i32 %y, float* nocapture %output) {
 .preheader:
   %0 = sext i32 %y to i64
   %1 = sext i32 %x to i64
-  %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+  %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
   %3 = addrspacecast float addrspace(3)* %2 to float*
-  %4 = load float* %3, align 4
+  %4 = load float, float* %3, align 4
   %5 = fadd float %4, 0.000000e+00
   %6 = add i64 %0, 1
-  %7 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
+  %7 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
   %8 = addrspacecast float addrspace(3)* %7 to float*
-  %9 = load float* %8, align 4
+  %9 = load float, float* %8, align 4
   %10 = fadd float %5, %9
   %11 = add i64 %1, 1
-  %12 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
+  %12 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
   %13 = addrspacecast float addrspace(3)* %12 to float*
-  %14 = load float* %13, align 4
+  %14 = load float, float* %13, align 4
   %15 = fadd float %10, %14
-  %16 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
+  %16 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
   %17 = addrspacecast float addrspace(3)* %16 to float*
-  %18 = load float* %17, align 4
+  %18 = load float, float* %17, align 4
   %19 = fadd float %15, %18
   store float %19, float* %output, align 4
   ret void
@@ -94,10 +94,10 @@ define void @sum_of_array2(i32 %x, i32 %y, float* nocapture %output) {
 ; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
 
 ; IR-LABEL: @sum_of_array2(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
 
 
 ; This function loads
@@ -114,25 +114,25 @@ define void @sum_of_array3(i32 %x, i32 %y, float* nocapture %output) {
 .preheader:
   %0 = zext i32 %y to i64
   %1 = zext i32 %x to i64
-  %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+  %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
   %3 = addrspacecast float addrspace(3)* %2 to float*
-  %4 = load float* %3, align 4
+  %4 = load float, float* %3, align 4
   %5 = fadd float %4, 0.000000e+00
   %6 = add nuw i32 %y, 1
   %7 = zext i32 %6 to i64
-  %8 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
+  %8 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %7
   %9 = addrspacecast float addrspace(3)* %8 to float*
-  %10 = load float* %9, align 4
+  %10 = load float, float* %9, align 4
   %11 = fadd float %5, %10
   %12 = add nuw i32 %x, 1
   %13 = zext i32 %12 to i64
-  %14 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
+  %14 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %0
   %15 = addrspacecast float addrspace(3)* %14 to float*
-  %16 = load float* %15, align 4
+  %16 = load float, float* %15, align 4
   %17 = fadd float %11, %16
-  %18 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
+  %18 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %13, i64 %7
   %19 = addrspacecast float addrspace(3)* %18 to float*
-  %20 = load float* %19, align 4
+  %20 = load float, float* %19, align 4
   %21 = fadd float %17, %20
   store float %21, float* %output, align 4
   ret void
@@ -144,10 +144,10 @@ define void @sum_of_array3(i32 %x, i32 %y, float* nocapture %output) {
 ; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
 
 ; IR-LABEL: @sum_of_array3(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
 
 
 ; This function loads
@@ -162,23 +162,23 @@ define void @sum_of_array4(i32 %x, i32 %y, float* nocapture %output) {
 .preheader:
   %0 = zext i32 %y to i64
   %1 = zext i32 %x to i64
-  %2 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
+  %2 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %0
   %3 = addrspacecast float addrspace(3)* %2 to float*
-  %4 = load float* %3, align 4
+  %4 = load float, float* %3, align 4
   %5 = fadd float %4, 0.000000e+00
   %6 = add i64 %0, 1
-  %7 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
+  %7 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %1, i64 %6
   %8 = addrspacecast float addrspace(3)* %7 to float*
-  %9 = load float* %8, align 4
+  %9 = load float, float* %8, align 4
   %10 = fadd float %5, %9
   %11 = add i64 %1, 1
-  %12 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
+  %12 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %0
   %13 = addrspacecast float addrspace(3)* %12 to float*
-  %14 = load float* %13, align 4
+  %14 = load float, float* %13, align 4
   %15 = fadd float %10, %14
-  %16 = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
+  %16 = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %11, i64 %6
   %17 = addrspacecast float addrspace(3)* %16 to float*
-  %18 = load float* %17, align 4
+  %18 = load float, float* %17, align 4
   %19 = fadd float %15, %18
   store float %19, float* %output, align 4
   ret void
@@ -190,7 +190,7 @@ define void @sum_of_array4(i32 %x, i32 %y, float* nocapture %output) {
 ; PTX: ld.shared.f32 {{%f[0-9]+}}, {{\[}}[[BASE_REG]]+132{{\]}}
 
 ; IR-LABEL: @sum_of_array4(
-; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 1
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 32
-; IR: getelementptr float addrspace(3)* [[BASE_PTR]], i64 33
+; IR: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]] addrspace(3)* @array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 1
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 32
+; IR: getelementptr float, float addrspace(3)* [[BASE_PTR]], i64 33
diff --git a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep.ll b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep.ll
index ea0d3f5673a5..2fdd158a35ed 100644
--- a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep.ll
+++ b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/split-gep.ll
@@ -1,4 +1,4 @@
-; RUN: opt < %s -separate-const-offset-from-gep -dce -S | FileCheck %s
+; RUN: opt < %s -separate-const-offset-from-gep -reassociate-geps-verify-no-dead-code -S | FileCheck %s
 
 ; Several unit tests for -separate-const-offset-from-gep. The transformation
 ; heavily relies on TargetTransformInfo, so we put these tests under
@@ -20,11 +20,11 @@ define double* @struct(i32 %i) {
 entry:
   %add = add nsw i32 %i, 5
   %idxprom = sext i32 %add to i64
-  %p = getelementptr inbounds [1024 x %struct.S]* @struct_array, i64 0, i64 %idxprom, i32 1
+  %p = getelementptr inbounds [1024 x %struct.S], [1024 x %struct.S]* @struct_array, i64 0, i64 %idxprom, i32 1
   ret double* %p
 }
 ; CHECK-LABEL: @struct(
-; CHECK: getelementptr [1024 x %struct.S]* @struct_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i32 1
+; CHECK: getelementptr [1024 x %struct.S], [1024 x %struct.S]* @struct_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i32 1
 
 ; We should be able to trace into sext(a + b) if a + b is non-negative
 ; (e.g., used as an index of an inbounds GEP) and one of a and b is
@@ -36,15 +36,15 @@ entry:
   %2 = add i32 %j, -2
   ; However, inbound sext(j + -2) != sext(j) + -2, e.g., j = INT_MIN
   %3 = sext i32 %2 to i64
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 %1, i64 %3
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %1, i64 %3
   ret float* %p
 }
 ; CHECK-LABEL: @sext_add(
 ; CHECK-NOT: = add
 ; CHECK: add i32 %j, -2
 ; CHECK: sext
-; CHECK: getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* %{{[a-zA-Z0-9]+}}, i64 32
+; CHECK: getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* %{{[a-zA-Z0-9]+}}, i64 32
 
 ; We should be able to trace into sext/zext if it can be distributed to both
 ; operands, e.g., sext (add nsw a, b) == add nsw (sext a), (sext b)
@@ -60,12 +60,12 @@ define float* @ext_add_no_overflow(i64 %a, i32 %b, i64 %c, i32 %d) {
   %d1 = add nuw i32 %d, 1
   %d2 = zext i32 %d1 to i64
   %j = add i64 %c, %d2       ; j = c + zext(d +nuw 1)
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %j
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %j
   ret float* %p
 }
 ; CHECK-LABEL: @ext_add_no_overflow(
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* [[BASE_PTR]], i64 33
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* [[BASE_PTR]], i64 33
 
 ; Verifies we handle nested sext/zext correctly.
 define void @sext_zext(i32 %a, i32 %b, float** %out1, float** %out2) {
@@ -76,7 +76,7 @@ entry:
   %3 = add nsw i32 %b, 2
   %4 = sext i32 %3 to i48
   %5 = zext i48 %4 to i64    ; zext(sext(b +nsw 2)) != zext(sext(b)) + 2
-  %p1 = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %2, i64 %5
+  %p1 = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %2, i64 %5
   store float* %p1, float** %out1
   %6 = add nuw i32 %a, 3
   %7 = zext i32 %6 to i48
@@ -84,15 +84,15 @@ entry:
   %9 = add nsw i32 %b, 4
   %10 = zext i32 %9 to i48
   %11 = sext i48 %10 to i64  ; sext(zext(b +nsw 4)) != zext(b) + 4
-  %p2 = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %8, i64 %11
+  %p2 = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %8, i64 %11
   store float* %p2, float** %out2
   ret void
 }
 ; CHECK-LABEL: @sext_zext(
-; CHECK: [[BASE_PTR_1:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* [[BASE_PTR_1]], i64 32
-; CHECK: [[BASE_PTR_2:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* [[BASE_PTR_2]], i64 96
+; CHECK: [[BASE_PTR_1:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* [[BASE_PTR_1]], i64 32
+; CHECK: [[BASE_PTR_2:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* [[BASE_PTR_2]], i64 96
 
 ; Similar to @ext_add_no_overflow, we should be able to trace into s/zext if
 ; its operand is an OR and the two operands of the OR have no common bits.
@@ -105,12 +105,12 @@ entry:
   %b3.ext = sext i32 %b3 to i64
   %i = add i64 %a, %b2.ext
   %j = add i64 %a, %b3.ext
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %j
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %j
   ret float* %p
 }
 ; CHECK-LABEL: @sext_or(
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* [[BASE_PTR]], i64 32
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* [[BASE_PTR]], i64 32
 
 ; The subexpression (b + 5) is used in both "i = a + (b + 5)" and "*out = b +
 ; 5". When extracting the constant offset 5, make sure "*out = b + 5" isn't
@@ -119,13 +119,13 @@ define float* @expr(i64 %a, i64 %b, i64* %out) {
 entry:
   %b5 = add i64 %b, 5
   %i = add i64 %b5, %a
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 0
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 0
   store i64 %b5, i64* %out
   ret float* %p
 }
 ; CHECK-LABEL: @expr(
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 0
-; CHECK: getelementptr float* [[BASE_PTR]], i64 160
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %{{[a-zA-Z0-9]+}}, i64 0
+; CHECK: getelementptr float, float* [[BASE_PTR]], i64 160
 ; CHECK: store i64 %b5, i64* %out
 
 ; d + sext(a +nsw (b +nsw (c +nsw 8))) => (d + sext(a) + sext(b) + sext(c)) + 8
@@ -136,26 +136,26 @@ entry:
   %2 = add nsw i32 %a, %1
   %3 = sext i32 %2 to i64
   %i = add i64 %d, %3
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
   ret float* %p
 }
 ; CHECK-LABEL: @sext_expr(
 ; CHECK: sext i32
 ; CHECK: sext i32
 ; CHECK: sext i32
-; CHECK: getelementptr float* %{{[a-zA-Z0-9]+}}, i64 8
+; CHECK: getelementptr float, float* %{{[a-zA-Z0-9]+}}, i64 8
 
 ; Verifies we handle "sub" correctly.
 define float* @sub(i64 %i, i64 %j) {
   %i2 = sub i64 %i, 5 ; i - 5
   %j2 = sub i64 5, %j ; 5 - i
-  %p = getelementptr inbounds [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i2, i64 %j2
+  %p = getelementptr inbounds [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i2, i64 %j2
   ret float* %p
 }
 ; CHECK-LABEL: @sub(
 ; CHECK: %[[j2:[a-zA-Z0-9]+]] = sub i64 0, %j
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %[[j2]]
-; CHECK: getelementptr float* [[BASE_PTR]], i64 -155
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 %i, i64 %[[j2]]
+; CHECK: getelementptr float, float* [[BASE_PTR]], i64 -155
 
 %struct.Packed = type <{ [3 x i32], [8 x i64] }> ; <> means packed
 
@@ -167,13 +167,13 @@ entry:
   %idxprom = sext i32 %add to i64
   %add1 = add nsw i32 %i, 1
   %idxprom2 = sext i32 %add1 to i64
-  %arrayidx3 = getelementptr inbounds [1024 x %struct.Packed]* %s, i64 0, i64 %idxprom2, i32 1, i64 %idxprom
+  %arrayidx3 = getelementptr inbounds [1024 x %struct.Packed], [1024 x %struct.Packed]* %s, i64 0, i64 %idxprom2, i32 1, i64 %idxprom
   ret i64* %arrayidx3
 }
 ; CHECK-LABEL: @packed_struct(
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [1024 x %struct.Packed]* %s, i64 0, i64 %{{[a-zA-Z0-9]+}}, i32 1, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [1024 x %struct.Packed], [1024 x %struct.Packed]* %s, i64 0, i64 %{{[a-zA-Z0-9]+}}, i32 1, i64 %{{[a-zA-Z0-9]+}}
 ; CHECK: [[CASTED_PTR:%[a-zA-Z0-9]+]] = bitcast i64* [[BASE_PTR]] to i8*
-; CHECK: %uglygep = getelementptr i8* [[CASTED_PTR]], i64 100
+; CHECK: %uglygep = getelementptr i8, i8* [[CASTED_PTR]], i64 100
 ; CHECK: bitcast i8* %uglygep to i64*
 
 ; We shouldn't be able to extract the 8 from "zext(a +nuw (b + 8))",
@@ -183,11 +183,11 @@ entry:
   %0 = add i32 %b, 8
   %1 = add nuw i32 %a, %0
   %i = zext i32 %1 to i64
-  %p = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
+  %p = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
   ret float* %p
 }
 ; CHECK-LABEL: zext_expr(
-; CHECK: getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
+; CHECK: getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %i
 
 ; Per http://llvm.org/docs/LangRef.html#id181, the indices of a off-bound gep
 ; should be considered sign-extended to the pointer size. Therefore,
@@ -200,11 +200,11 @@ entry:
 define float* @i32_add(i32 %a) {
 entry:
   %i = add i32 %a, 8
-  %p = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i32 %i
+  %p = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i32 %i
   ret float* %p
 }
 ; CHECK-LABEL: @i32_add(
-; CHECK: getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %{{[a-zA-Z0-9]+}}
 ; CHECK-NOT: getelementptr
 
 ; Verifies that we compute the correct constant offset when the index is
@@ -216,23 +216,23 @@ entry:
   %0 = add nsw nuw i1 %a, 1
   %1 = sext i1 %0 to i4
   %2 = zext i4 %1 to i64         ; zext (sext i1 1 to i4) to i64 = 15
-  %p = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %2
+  %p = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %2
   ret float* %p
 }
 ; CHECK-LABEL: @apint(
-; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %{{[a-zA-Z0-9]+}}
-; CHECK: getelementptr float* [[BASE_PTR]], i64 15
+; CHECK: [[BASE_PTR:%[a-zA-Z0-9]+]] = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %{{[a-zA-Z0-9]+}}
+; CHECK: getelementptr float, float* [[BASE_PTR]], i64 15
 
 ; Do not trace into binary operators other than ADD, SUB, and OR.
 define float* @and(i64 %a) {
 entry:
   %0 = shl i64 %a, 2
   %1 = and i64 %0, 1
-  %p = getelementptr [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %1
+  %p = getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array, i64 0, i64 0, i64 %1
   ret float* %p
 }
 ; CHECK-LABEL: @and(
-; CHECK: getelementptr [32 x [32 x float]]* @float_2d_array
+; CHECK: getelementptr [32 x [32 x float]], [32 x [32 x float]]* @float_2d_array
 ; CHECK-NOT: getelementptr
 
 ; The code that rebuilds an OR expression used to be buggy, and failed on this
@@ -247,9 +247,9 @@ entry:
   ; ((a << 2) + 12) and 1 have no common bits. Therefore,
   ; SeparateConstOffsetFromGEP is able to extract the 12.
   ; TODO(jingyue): We could reassociate the expression to combine 12 and 1.
-  %p = getelementptr float* %ptr, i64 %or
-; CHECK: [[PTR:%[a-zA-Z0-9]+]] = getelementptr float* %ptr, i64 [[OR]]
-; CHECK: getelementptr float* [[PTR]], i64 12
+  %p = getelementptr float, float* %ptr, i64 %or
+; CHECK: [[PTR:%[a-zA-Z0-9]+]] = getelementptr float, float* %ptr, i64 [[OR]]
+; CHECK: getelementptr float, float* [[PTR]], i64 12
   ret float* %p
 ; CHECK-NEXT: ret
 }
@@ -269,10 +269,10 @@ define %struct2* @sign_mod_unsign(%struct0* %ptr, i64 %idx) {
 entry:
   %arrayidx = add nsw i64 %idx, -2
 ; CHECK-NOT: add
-  %ptr2 = getelementptr inbounds %struct0* %ptr, i64 0, i32 3, i64 %arrayidx, i32 1
-; CHECK: [[PTR:%[a-zA-Z0-9]+]] = getelementptr %struct0* %ptr, i64 0, i32 3, i64 %idx, i32 1
+  %ptr2 = getelementptr inbounds %struct0, %struct0* %ptr, i64 0, i32 3, i64 %arrayidx, i32 1
+; CHECK: [[PTR:%[a-zA-Z0-9]+]] = getelementptr %struct0, %struct0* %ptr, i64 0, i32 3, i64 %idx, i32 1
 ; CHECK: [[PTR1:%[a-zA-Z0-9]+]] = bitcast %struct2* [[PTR]] to i8*
-; CHECK: getelementptr i8* [[PTR1]], i64 -64
+; CHECK: getelementptr i8, i8* [[PTR1]], i64 -64
 ; CHECK: bitcast
   ret %struct2* %ptr2
 ; CHECK-NEXT: ret
diff --git a/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/value-tracking-domtree.ll b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/value-tracking-domtree.ll
new file mode 100644
index 000000000000..d1a0f33d5a21
--- /dev/null
+++ b/test/Transforms/SeparateConstOffsetFromGEP/NVPTX/value-tracking-domtree.ll
@@ -0,0 +1,33 @@
+; RUN: opt < %s -separate-const-offset-from-gep -value-tracking-dom-conditions -reassociate-geps-verify-no-dead-code -S | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "nvptx64-unknown-unknown"
+
+; if (i == 4)
+;   p = &input[i | 3];
+;
+; =>
+;
+; if (i == 4) {
+;   base = &input[i];
+;   p = &base[3];
+; }
+;
+; We should treat (i | 3) as (i + 3) because i is guaranteed to be 4, which
+; does not share any set bits with 3.
+define float* @guarded_or(float* %input, i64 %i) {
+; CHECK-LABEL: @guarded_or(
+entry:
+  %is4 = icmp eq i64 %i, 4
+  br i1 %is4, label %then, label %exit
+
+then:
+  %or = or i64 %i, 3
+  %p = getelementptr inbounds float, float* %input, i64 %or
+; CHECK: [[base:[^ ]+]] = getelementptr float, float* %input, i64 %i
+; CHECK: getelementptr float, float* [[base]], i64 3
+  ret float* %p
+
+exit:
+  ret float* null
+}
diff --git a/test/Transforms/SimplifyCFG/2003-08-17-FoldSwitch-dbg.ll b/test/Transforms/SimplifyCFG/2003-08-17-FoldSwitch-dbg.ll
index fbfb100274f6..0626ea178e5c 100644
--- a/test/Transforms/SimplifyCFG/2003-08-17-FoldSwitch-dbg.ll
+++ b/test/Transforms/SimplifyCFG/2003-08-17-FoldSwitch-dbg.ll
@@ -9,7 +9,7 @@
 @.str = internal constant [4 x i8] c"a.c\00", section "llvm.metadata"		; <[4 x i8]*> [#uses=1]
 @.str1 = internal constant [6 x i8] c"/tmp/\00", section "llvm.metadata"	; <[6 x i8]*> [#uses=1]
 @.str2 = internal constant [55 x i8] c"4.2.1 (Based on Apple Inc. build 5636) (LLVM build 00)\00", section "llvm.metadata"		; <[55 x i8]*> [#uses=1]
-@llvm.dbg.compile_unit = internal constant %llvm.dbg.compile_unit.type { i32 458769, { }* bitcast (%llvm.dbg.anchor.type* @llvm.dbg.compile_units to { }*), i32 1, i8* getelementptr ([4 x i8]* @.str, i32 0, i32 0), i8* getelementptr ([6 x i8]* @.str1, i32 0, i32 0), i8* getelementptr ([55 x i8]* @.str2, i32 0, i32 0), i1 true, i1 false, i8* null }, section "llvm.metadata"		; <%llvm.dbg.compile_unit.type*> [#uses=1]
+@llvm.dbg.compile_unit = internal constant %llvm.dbg.compile_unit.type { i32 458769, { }* bitcast (%llvm.dbg.anchor.type* @llvm.dbg.compile_units to { }*), i32 1, i8* getelementptr ([4 x i8], [4 x i8]* @.str, i32 0, i32 0), i8* getelementptr ([6 x i8], [6 x i8]* @.str1, i32 0, i32 0), i8* getelementptr ([55 x i8], [55 x i8]* @.str2, i32 0, i32 0), i1 true, i1 false, i8* null }, section "llvm.metadata"		; <%llvm.dbg.compile_unit.type*> [#uses=1]
 
 declare void @llvm.dbg.stoppoint(i32, i32, { }*) nounwind
 
diff --git a/test/Transforms/SimplifyCFG/2005-06-16-PHICrash.ll b/test/Transforms/SimplifyCFG/2005-06-16-PHICrash.ll
index 90be6803a5c3..8fd1fae34503 100644
--- a/test/Transforms/SimplifyCFG/2005-06-16-PHICrash.ll
+++ b/test/Transforms/SimplifyCFG/2005-06-16-PHICrash.ll
@@ -30,7 +30,7 @@ endif.0:		; preds = %loopentry.0
 	%tmp.14 = sext i8 %tmp.1 to i32		; <i32> [#uses=1]
 	%tmp.16 = zext i8 %l_88173906 to i32		; <i32> [#uses=1]
 	%tmp.17 = icmp sgt i32 %tmp.14, %tmp.16		; <i1> [#uses=1]
-	%tmp.19 = load i32* @g_59182229		; <i32> [#uses=2]
+	%tmp.19 = load i32, i32* @g_59182229		; <i32> [#uses=2]
 	br i1 %tmp.17, label %cond_true, label %cond_false
 cond_true:		; preds = %endif.0
 	%tmp.20 = icmp ne i32 %tmp.19, 1		; <i1> [#uses=1]
@@ -53,7 +53,7 @@ loopentry.1:		; preds = %endif.3, %else.2
 	%tmp.29 = icmp sgt i32 %i.1.1, 99		; <i1> [#uses=1]
 	br i1 %tmp.29, label %endif.2, label %no_exit.1
 no_exit.1:		; preds = %loopentry.1
-	%tmp.30 = load i32* @g_38098584		; <i32> [#uses=1]
+	%tmp.30 = load i32, i32* @g_38098584		; <i32> [#uses=1]
 	%tmp.31 = icmp eq i32 %tmp.30, 0		; <i1> [#uses=1]
 	br i1 %tmp.31, label %else.3, label %then.3
 then.3:		; preds = %no_exit.1
diff --git a/test/Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll b/test/Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll
index c30bfa1379c7..c45986749ab0 100644
--- a/test/Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll
+++ b/test/Transforms/SimplifyCFG/2005-08-01-PHIUpdateFail.ll
@@ -58,7 +58,7 @@ then.2.i:		; preds = %endif.1.i, %then.1.i
 getfree.exit:		; preds = %endif.1.i, %then.1.i
 	ret void
 endif.1:		; preds = %read_min.exit
-	%tmp.27.i = getelementptr i32* null, i32 0		; <i32*> [#uses=0]
+	%tmp.27.i = getelementptr i32, i32* null, i32 0		; <i32*> [#uses=0]
 	br i1 false, label %loopexit.0.i15, label %no_exit.0.i14
 no_exit.0.i14:		; preds = %endif.1
 	ret void
diff --git a/test/Transforms/SimplifyCFG/2005-12-03-IncorrectPHIFold.ll b/test/Transforms/SimplifyCFG/2005-12-03-IncorrectPHIFold.ll
index 907261bd438a..2606e08da9a2 100644
--- a/test/Transforms/SimplifyCFG/2005-12-03-IncorrectPHIFold.ll
+++ b/test/Transforms/SimplifyCFG/2005-12-03-IncorrectPHIFold.ll
@@ -36,8 +36,8 @@ entry:
 	%guess = alloca %struct.anon*		; <%struct.anon**> [#uses=7]
 	%guess1 = alloca %struct.anon*		; <%struct.anon**> [#uses=7]
 	%point5 = alloca %struct.anon*		; <%struct.anon**> [#uses=3]
-	%tmp = load %struct.anon** %num		; <%struct.anon*> [#uses=1]
-	%tmp1 = load %struct.anon** @_zero_		; <%struct.anon*> [#uses=1]
+	%tmp = load %struct.anon*, %struct.anon** %num		; <%struct.anon*> [#uses=1]
+	%tmp1 = load %struct.anon*, %struct.anon** @_zero_		; <%struct.anon*> [#uses=1]
 	%tmp.upgrd.1 = call i32 @bc_compare( %struct.anon* %tmp, %struct.anon* %tmp1 )		; <i32> [#uses=2]
 	%tmp.upgrd.2 = icmp slt i32 %tmp.upgrd.1, 0		; <i1> [#uses=1]
 	br i1 %tmp.upgrd.2, label %cond_true, label %cond_false
@@ -48,26 +48,26 @@ cond_false:		; preds = %entry
 	br i1 %tmp5, label %cond_true6, label %cond_next13
 cond_true6:		; preds = %cond_false
 	call void @free_num( %struct.anon** %num )
-	%tmp8 = load %struct.anon** @_zero_		; <%struct.anon*> [#uses=1]
+	%tmp8 = load %struct.anon*, %struct.anon** @_zero_		; <%struct.anon*> [#uses=1]
 	%tmp9 = call %struct.anon* @copy_num( %struct.anon* %tmp8 )		; <%struct.anon*> [#uses=1]
 	store %struct.anon* %tmp9, %struct.anon** %num
 	ret i32 1
 cond_next13:		; preds = %cond_false
-	%tmp15 = load %struct.anon** %num		; <%struct.anon*> [#uses=1]
-	%tmp16 = load %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
+	%tmp15 = load %struct.anon*, %struct.anon** %num		; <%struct.anon*> [#uses=1]
+	%tmp16 = load %struct.anon*, %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
 	%tmp17 = call i32 @bc_compare( %struct.anon* %tmp15, %struct.anon* %tmp16 )		; <i32> [#uses=2]
 	%tmp19 = icmp eq i32 %tmp17, 0		; <i1> [#uses=1]
 	br i1 %tmp19, label %cond_true20, label %cond_next27
 cond_true20:		; preds = %cond_next13
 	call void @free_num( %struct.anon** %num )
-	%tmp22 = load %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
+	%tmp22 = load %struct.anon*, %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
 	%tmp23 = call %struct.anon* @copy_num( %struct.anon* %tmp22 )		; <%struct.anon*> [#uses=1]
 	store %struct.anon* %tmp23, %struct.anon** %num
 	ret i32 1
 cond_next27:		; preds = %cond_next13
-	%tmp29 = load %struct.anon** %num		; <%struct.anon*> [#uses=1]
-	%tmp30 = getelementptr %struct.anon* %tmp29, i32 0, i32 2		; <i32*> [#uses=1]
-	%tmp31 = load i32* %tmp30		; <i32> [#uses=2]
+	%tmp29 = load %struct.anon*, %struct.anon** %num		; <%struct.anon*> [#uses=1]
+	%tmp30 = getelementptr %struct.anon, %struct.anon* %tmp29, i32 0, i32 2		; <i32*> [#uses=1]
+	%tmp31 = load i32, i32* %tmp30		; <i32> [#uses=2]
 	%tmp33 = icmp sge i32 %tmp31, %scale		; <i1> [#uses=1]
 	%max = select i1 %tmp33, i32 %tmp31, i32 %scale		; <i32> [#uses=4]
 	%tmp35 = add i32 %max, 2		; <i32> [#uses=0]
@@ -75,29 +75,29 @@ cond_next27:		; preds = %cond_next13
 	call void @init_num( %struct.anon** %guess1 )
 	%tmp36 = call %struct.anon* @new_num( i32 1, i32 1 )		; <%struct.anon*> [#uses=2]
 	store %struct.anon* %tmp36, %struct.anon** %point5
-	%tmp.upgrd.3 = getelementptr %struct.anon* %tmp36, i32 0, i32 4, i32 1		; <i8*> [#uses=1]
+	%tmp.upgrd.3 = getelementptr %struct.anon, %struct.anon* %tmp36, i32 0, i32 4, i32 1		; <i8*> [#uses=1]
 	store i8 5, i8* %tmp.upgrd.3
 	%tmp39 = icmp slt i32 %tmp17, 0		; <i1> [#uses=1]
 	br i1 %tmp39, label %cond_true40, label %cond_false43
 cond_true40:		; preds = %cond_next27
-	%tmp41 = load %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
+	%tmp41 = load %struct.anon*, %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
 	%tmp42 = call %struct.anon* @copy_num( %struct.anon* %tmp41 )		; <%struct.anon*> [#uses=1]
 	store %struct.anon* %tmp42, %struct.anon** %guess
 	br label %bb80.outer
 cond_false43:		; preds = %cond_next27
 	call void @int2num( %struct.anon** %guess, i32 10 )
-	%tmp45 = load %struct.anon** %num		; <%struct.anon*> [#uses=1]
-	%tmp46 = getelementptr %struct.anon* %tmp45, i32 0, i32 1		; <i32*> [#uses=1]
-	%tmp47 = load i32* %tmp46		; <i32> [#uses=1]
+	%tmp45 = load %struct.anon*, %struct.anon** %num		; <%struct.anon*> [#uses=1]
+	%tmp46 = getelementptr %struct.anon, %struct.anon* %tmp45, i32 0, i32 1		; <i32*> [#uses=1]
+	%tmp47 = load i32, i32* %tmp46		; <i32> [#uses=1]
 	call void @int2num( %struct.anon** %guess1, i32 %tmp47 )
-	%tmp48 = load %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
-	%tmp49 = load %struct.anon** %point5		; <%struct.anon*> [#uses=1]
+	%tmp48 = load %struct.anon*, %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
+	%tmp49 = load %struct.anon*, %struct.anon** %point5		; <%struct.anon*> [#uses=1]
 	call void @bc_multiply( %struct.anon* %tmp48, %struct.anon* %tmp49, %struct.anon** %guess1, i32 %max )
-	%tmp51 = load %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
-	%tmp52 = getelementptr %struct.anon* %tmp51, i32 0, i32 2		; <i32*> [#uses=1]
+	%tmp51 = load %struct.anon*, %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
+	%tmp52 = getelementptr %struct.anon, %struct.anon* %tmp51, i32 0, i32 2		; <i32*> [#uses=1]
 	store i32 0, i32* %tmp52
-	%tmp53 = load %struct.anon** %guess		; <%struct.anon*> [#uses=1]
-	%tmp54 = load %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
+	%tmp53 = load %struct.anon*, %struct.anon** %guess		; <%struct.anon*> [#uses=1]
+	%tmp54 = load %struct.anon*, %struct.anon** %guess1		; <%struct.anon*> [#uses=1]
 	call void @bc_raise( %struct.anon* %tmp53, %struct.anon* %tmp54, %struct.anon** %guess, i32 %max )
 	br label %bb80.outer
 bb80.outer:		; preds = %cond_true83, %cond_false43, %cond_true40
@@ -113,8 +113,8 @@ cond_true83:		; preds = %bb80
 ; CHECK: bb86
 bb86:		; preds = %bb80
 	call void @free_num( %struct.anon** %num )
-	%tmp88 = load %struct.anon** %guess		; <%struct.anon*> [#uses=1]
-	%tmp89 = load %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
+	%tmp88 = load %struct.anon*, %struct.anon** %guess		; <%struct.anon*> [#uses=1]
+	%tmp89 = load %struct.anon*, %struct.anon** @_one_		; <%struct.anon*> [#uses=1]
 	%tmp92 = call i32 @bc_divide( %struct.anon* %tmp88, %struct.anon* %tmp89, %struct.anon** %num, i32 %max )		; <i32> [#uses=0]
 	call void @free_num( %struct.anon** %guess )
 	call void @free_num( %struct.anon** %guess1 )
diff --git a/test/Transforms/SimplifyCFG/2006-08-03-Crash.ll b/test/Transforms/SimplifyCFG/2006-08-03-Crash.ll
index 70fbddfe33b4..0820e9c9825f 100644
--- a/test/Transforms/SimplifyCFG/2006-08-03-Crash.ll
+++ b/test/Transforms/SimplifyCFG/2006-08-03-Crash.ll
@@ -34,7 +34,7 @@ target triple = "powerpc-apple-darwin8"
 
 define void @fold_builtin_classify() {
 entry:
-	%tmp63 = load i32* null		; <i32> [#uses=1]
+	%tmp63 = load i32, i32* null		; <i32> [#uses=1]
 	switch i32 %tmp63, label %bb276 [
 		 i32 414, label %bb145
 		 i32 417, label %bb
@@ -42,54 +42,54 @@ entry:
 bb:		; preds = %entry
 	ret void
 bb145:		; preds = %entry
-	%tmp146 = load %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
-	%tmp148 = getelementptr %struct.tree_node* %tmp146, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
-	%tmp149 = load %struct.tree_node** %tmp148		; <%struct.tree_node*> [#uses=1]
+	%tmp146 = load %struct.tree_node*, %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
+	%tmp148 = getelementptr %struct.tree_node, %struct.tree_node* %tmp146, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
+	%tmp149 = load %struct.tree_node*, %struct.tree_node** %tmp148		; <%struct.tree_node*> [#uses=1]
 	%tmp150 = bitcast %struct.tree_node* %tmp149 to %struct.tree_type*		; <%struct.tree_type*> [#uses=1]
-	%tmp151 = getelementptr %struct.tree_type* %tmp150, i32 0, i32 6		; <i16*> [#uses=1]
+	%tmp151 = getelementptr %struct.tree_type, %struct.tree_type* %tmp150, i32 0, i32 6		; <i16*> [#uses=1]
 	%tmp151.upgrd.1 = bitcast i16* %tmp151 to i32*		; <i32*> [#uses=1]
-	%tmp152 = load i32* %tmp151.upgrd.1		; <i32> [#uses=1]
+	%tmp152 = load i32, i32* %tmp151.upgrd.1		; <i32> [#uses=1]
 	%tmp154 = lshr i32 %tmp152, 16		; <i32> [#uses=1]
 	%tmp154.mask = and i32 %tmp154, 127		; <i32> [#uses=1]
 	%gep.upgrd.2 = zext i32 %tmp154.mask to i64		; <i64> [#uses=1]
-	%tmp155 = getelementptr [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.2		; <i8*> [#uses=1]
-	%tmp156 = load i8* %tmp155		; <i8> [#uses=1]
+	%tmp155 = getelementptr [35 x i8], [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.2		; <i8*> [#uses=1]
+	%tmp156 = load i8, i8* %tmp155		; <i8> [#uses=1]
 	%tmp157 = icmp eq i8 %tmp156, 4		; <i1> [#uses=1]
 	br i1 %tmp157, label %cond_next241, label %cond_true158
 cond_true158:		; preds = %bb145
-	%tmp172 = load %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
-	%tmp174 = getelementptr %struct.tree_node* %tmp172, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
-	%tmp175 = load %struct.tree_node** %tmp174		; <%struct.tree_node*> [#uses=1]
+	%tmp172 = load %struct.tree_node*, %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
+	%tmp174 = getelementptr %struct.tree_node, %struct.tree_node* %tmp172, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
+	%tmp175 = load %struct.tree_node*, %struct.tree_node** %tmp174		; <%struct.tree_node*> [#uses=1]
 	%tmp176 = bitcast %struct.tree_node* %tmp175 to %struct.tree_type*		; <%struct.tree_type*> [#uses=1]
-	%tmp177 = getelementptr %struct.tree_type* %tmp176, i32 0, i32 6		; <i16*> [#uses=1]
+	%tmp177 = getelementptr %struct.tree_type, %struct.tree_type* %tmp176, i32 0, i32 6		; <i16*> [#uses=1]
 	%tmp177.upgrd.3 = bitcast i16* %tmp177 to i32*		; <i32*> [#uses=1]
-	%tmp178 = load i32* %tmp177.upgrd.3		; <i32> [#uses=1]
+	%tmp178 = load i32, i32* %tmp177.upgrd.3		; <i32> [#uses=1]
 	%tmp180 = lshr i32 %tmp178, 16		; <i32> [#uses=1]
 	%tmp180.mask = and i32 %tmp180, 127		; <i32> [#uses=1]
 	%gep.upgrd.4 = zext i32 %tmp180.mask to i64		; <i64> [#uses=1]
-	%tmp181 = getelementptr [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.4		; <i8*> [#uses=1]
-	%tmp182 = load i8* %tmp181		; <i8> [#uses=1]
+	%tmp181 = getelementptr [35 x i8], [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.4		; <i8*> [#uses=1]
+	%tmp182 = load i8, i8* %tmp181		; <i8> [#uses=1]
 	%tmp183 = icmp eq i8 %tmp182, 8		; <i1> [#uses=1]
 	br i1 %tmp183, label %cond_next241, label %cond_true184
 cond_true184:		; preds = %cond_true158
-	%tmp185 = load %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
-	%tmp187 = getelementptr %struct.tree_node* %tmp185, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
-	%tmp188 = load %struct.tree_node** %tmp187		; <%struct.tree_node*> [#uses=1]
+	%tmp185 = load %struct.tree_node*, %struct.tree_node** null		; <%struct.tree_node*> [#uses=1]
+	%tmp187 = getelementptr %struct.tree_node, %struct.tree_node* %tmp185, i32 0, i32 0, i32 0, i32 1		; <%struct.tree_node**> [#uses=1]
+	%tmp188 = load %struct.tree_node*, %struct.tree_node** %tmp187		; <%struct.tree_node*> [#uses=1]
 	%tmp189 = bitcast %struct.tree_node* %tmp188 to %struct.tree_type*		; <%struct.tree_type*> [#uses=1]
-	%tmp190 = getelementptr %struct.tree_type* %tmp189, i32 0, i32 6		; <i16*> [#uses=1]
+	%tmp190 = getelementptr %struct.tree_type, %struct.tree_type* %tmp189, i32 0, i32 6		; <i16*> [#uses=1]
 	%tmp190.upgrd.5 = bitcast i16* %tmp190 to i32*		; <i32*> [#uses=1]
-	%tmp191 = load i32* %tmp190.upgrd.5		; <i32> [#uses=1]
+	%tmp191 = load i32, i32* %tmp190.upgrd.5		; <i32> [#uses=1]
 	%tmp193 = lshr i32 %tmp191, 16		; <i32> [#uses=1]
 	%tmp193.mask = and i32 %tmp193, 127		; <i32> [#uses=1]
 	%gep.upgrd.6 = zext i32 %tmp193.mask to i64		; <i64> [#uses=1]
-	%tmp194 = getelementptr [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.6		; <i8*> [#uses=1]
-	%tmp195 = load i8* %tmp194		; <i8> [#uses=1]
+	%tmp194 = getelementptr [35 x i8], [35 x i8]* @mode_class, i32 0, i64 %gep.upgrd.6		; <i8*> [#uses=1]
+	%tmp195 = load i8, i8* %tmp194		; <i8> [#uses=1]
 	%tmp196 = icmp eq i8 %tmp195, 4		; <i1> [#uses=1]
 	br i1 %tmp196, label %cond_next241, label %cond_true197
 cond_true197:		; preds = %cond_true184
 	ret void
 cond_next241:		; preds = %cond_true184, %cond_true158, %bb145
-	%tmp245 = load i32* null		; <i32> [#uses=0]
+	%tmp245 = load i32, i32* null		; <i32> [#uses=0]
 	ret void
 bb276:		; preds = %entry
 	ret void
diff --git a/test/Transforms/SimplifyCFG/2006-12-08-Ptr-ICmp-Branch.ll b/test/Transforms/SimplifyCFG/2006-12-08-Ptr-ICmp-Branch.ll
index af865ce7fac0..dcf241255d85 100644
--- a/test/Transforms/SimplifyCFG/2006-12-08-Ptr-ICmp-Branch.ll
+++ b/test/Transforms/SimplifyCFG/2006-12-08-Ptr-ICmp-Branch.ll
@@ -45,42 +45,42 @@ entry:
 	store %struct.trie_s* %t, %struct.trie_s** %t_addr
 	store %struct.FILE* %f, %struct.FILE** %f_addr
 	store i32 0, i32* %wstate
-	%tmp = getelementptr %struct.charsequence* %cs, i64 0, i32 0		; <i8**> [#uses=1]
-	%tmp1 = getelementptr %struct.charsequence* @C.0.2294, i64 0, i32 0		; <i8**> [#uses=1]
-	%tmp.upgrd.5 = load i8** %tmp1		; <i8*> [#uses=1]
+	%tmp = getelementptr %struct.charsequence, %struct.charsequence* %cs, i64 0, i32 0		; <i8**> [#uses=1]
+	%tmp1 = getelementptr %struct.charsequence, %struct.charsequence* @C.0.2294, i64 0, i32 0		; <i8**> [#uses=1]
+	%tmp.upgrd.5 = load i8*, i8** %tmp1		; <i8*> [#uses=1]
 	store i8* %tmp.upgrd.5, i8** %tmp
-	%tmp.upgrd.6 = getelementptr %struct.charsequence* %cs, i64 0, i32 1		; <i32*> [#uses=1]
-	%tmp2 = getelementptr %struct.charsequence* @C.0.2294, i64 0, i32 1		; <i32*> [#uses=1]
-	%tmp.upgrd.7 = load i32* %tmp2		; <i32> [#uses=1]
+	%tmp.upgrd.6 = getelementptr %struct.charsequence, %struct.charsequence* %cs, i64 0, i32 1		; <i32*> [#uses=1]
+	%tmp2 = getelementptr %struct.charsequence, %struct.charsequence* @C.0.2294, i64 0, i32 1		; <i32*> [#uses=1]
+	%tmp.upgrd.7 = load i32, i32* %tmp2		; <i32> [#uses=1]
 	store i32 %tmp.upgrd.7, i32* %tmp.upgrd.6
-	%tmp3 = getelementptr %struct.charsequence* %cs, i64 0, i32 2		; <i32*> [#uses=1]
-	%tmp4 = getelementptr %struct.charsequence* @C.0.2294, i64 0, i32 2		; <i32*> [#uses=1]
-	%tmp5 = load i32* %tmp4		; <i32> [#uses=1]
+	%tmp3 = getelementptr %struct.charsequence, %struct.charsequence* %cs, i64 0, i32 2		; <i32*> [#uses=1]
+	%tmp4 = getelementptr %struct.charsequence, %struct.charsequence* @C.0.2294, i64 0, i32 2		; <i32*> [#uses=1]
+	%tmp5 = load i32, i32* %tmp4		; <i32> [#uses=1]
 	store i32 %tmp5, i32* %tmp3
 	br label %bb33
 bb:		; preds = %bb33
-	%tmp.upgrd.8 = load %struct.FILE** %f_addr		; <%struct.FILE*> [#uses=1]
+	%tmp.upgrd.8 = load %struct.FILE*, %struct.FILE** %f_addr		; <%struct.FILE*> [#uses=1]
 	%tmp.upgrd.9 = call i32 @_IO_getc( %struct.FILE* %tmp.upgrd.8 )		; <i32> [#uses=1]
 	%tmp6 = call i32 @tolower( i32 %tmp.upgrd.9 )		; <i32> [#uses=1]
 	%tmp6.upgrd.10 = trunc i32 %tmp6 to i8		; <i8> [#uses=1]
 	store i8 %tmp6.upgrd.10, i8* %c
-	%tmp7 = load i32* %wstate		; <i32> [#uses=1]
+	%tmp7 = load i32, i32* %wstate		; <i32> [#uses=1]
 	%tmp.upgrd.11 = icmp ne i32 %tmp7, 0		; <i1> [#uses=1]
 	br i1 %tmp.upgrd.11, label %cond_true, label %cond_false
 cond_true:		; preds = %bb
-	%tmp.upgrd.12 = load i8* %c		; <i8> [#uses=1]
+	%tmp.upgrd.12 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp8 = icmp sle i8 %tmp.upgrd.12, 96		; <i1> [#uses=1]
 	br i1 %tmp8, label %cond_true9, label %cond_next
 cond_true9:		; preds = %cond_true
 	br label %bb16
 cond_next:		; preds = %cond_true
-	%tmp10 = load i8* %c		; <i8> [#uses=1]
+	%tmp10 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp11 = icmp sgt i8 %tmp10, 122		; <i1> [#uses=1]
 	br i1 %tmp11, label %cond_true12, label %cond_next13
 cond_true12:		; preds = %cond_next
 	br label %bb16
 cond_next13:		; preds = %cond_next
-	%tmp14 = load i8* %c		; <i8> [#uses=1]
+	%tmp14 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp14.upgrd.13 = sext i8 %tmp14 to i32		; <i32> [#uses=1]
 	%tmp1415 = trunc i32 %tmp14.upgrd.13 to i8		; <i8> [#uses=1]
 	call void @charsequence_push( %struct.charsequence* %cs, i8 %tmp1415 )
@@ -88,26 +88,26 @@ cond_next13:		; preds = %cond_next
 bb16:		; preds = %cond_true12, %cond_true9
 	%tmp17 = call i8* @charsequence_val( %struct.charsequence* %cs )		; <i8*> [#uses=1]
 	store i8* %tmp17, i8** %str
-	%tmp.upgrd.14 = load %struct.trie_s** %t_addr		; <%struct.trie_s*> [#uses=1]
-	%tmp18 = load i8** %str		; <i8*> [#uses=1]
+	%tmp.upgrd.14 = load %struct.trie_s*, %struct.trie_s** %t_addr		; <%struct.trie_s*> [#uses=1]
+	%tmp18 = load i8*, i8** %str		; <i8*> [#uses=1]
 	%tmp19 = call %struct.trie_s* @trie_insert( %struct.trie_s* %tmp.upgrd.14, i8* %tmp18 )		; <%struct.trie_s*> [#uses=0]
-	%tmp20 = load i8** %str		; <i8*> [#uses=1]
+	%tmp20 = load i8*, i8** %str		; <i8*> [#uses=1]
 	call void @free( i8* %tmp20 )
 	store i32 0, i32* %wstate
 	br label %bb21
 bb21:		; preds = %bb16, %cond_next13
 	br label %cond_next32
 cond_false:		; preds = %bb
-	%tmp22 = load i8* %c		; <i8> [#uses=1]
+	%tmp22 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp23 = icmp sgt i8 %tmp22, 96		; <i1> [#uses=1]
 	br i1 %tmp23, label %cond_true24, label %cond_next31
 cond_true24:		; preds = %cond_false
-	%tmp25 = load i8* %c		; <i8> [#uses=1]
+	%tmp25 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp26 = icmp sle i8 %tmp25, 122		; <i1> [#uses=1]
 	br i1 %tmp26, label %cond_true27, label %cond_next30
 cond_true27:		; preds = %cond_true24
 	call void @charsequence_reset( %struct.charsequence* %cs )
-	%tmp28 = load i8* %c		; <i8> [#uses=1]
+	%tmp28 = load i8, i8* %c		; <i8> [#uses=1]
 	%tmp28.upgrd.15 = sext i8 %tmp28 to i32		; <i32> [#uses=1]
 	%tmp2829 = trunc i32 %tmp28.upgrd.15 to i8		; <i8> [#uses=1]
 	call void @charsequence_push( %struct.charsequence* %cs, i8 %tmp2829 )
@@ -120,7 +120,7 @@ cond_next31:		; preds = %cond_next30, %cond_false
 cond_next32:		; preds = %cond_next31, %bb21
 	br label %bb33
 bb33:		; preds = %cond_next32, %entry
-	%tmp34 = load %struct.FILE** %f_addr		; <%struct.FILE*> [#uses=1]
+	%tmp34 = load %struct.FILE*, %struct.FILE** %f_addr		; <%struct.FILE*> [#uses=1]
 	%tmp35 = call i32 @feof( %struct.FILE* %tmp34 )		; <i32> [#uses=1]
 	%tmp36 = icmp eq i32 %tmp35, 0		; <i1> [#uses=1]
 	br i1 %tmp36, label %bb, label %bb37
diff --git a/test/Transforms/SimplifyCFG/2007-11-22-InvokeNoUnwind.ll b/test/Transforms/SimplifyCFG/2007-11-22-InvokeNoUnwind.ll
index a90e07276253..76f41e8fc218 100644
--- a/test/Transforms/SimplifyCFG/2007-11-22-InvokeNoUnwind.ll
+++ b/test/Transforms/SimplifyCFG/2007-11-22-InvokeNoUnwind.ll
@@ -12,5 +12,9 @@ Cont:		; preds = %0
 	ret i32 0
 
 Other:		; preds = %0
+	landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
+		catch i8* null
 	ret i32 1
 }
+
+declare i32 @__gxx_personality_v0(...)
diff --git a/test/Transforms/SimplifyCFG/2007-12-21-Crash.ll b/test/Transforms/SimplifyCFG/2007-12-21-Crash.ll
index 46df0f0ed071..99041ede5d98 100644
--- a/test/Transforms/SimplifyCFG/2007-12-21-Crash.ll
+++ b/test/Transforms/SimplifyCFG/2007-12-21-Crash.ll
@@ -13,7 +13,7 @@ bb5.outer:		; preds = %bb5.outer.loopexit, %entry
 	br label %bb5
 
 bb5:		; preds = %bb5, %bb5.outer
-	%tmp6 = tail call i32 (...)* @foo( ) nounwind 		; <i32> [#uses=1]
+	%tmp6 = tail call i32 (...) @foo( ) nounwind 		; <i32> [#uses=1]
 	switch i32 %tmp6, label %bb13 [
 		 i32 -1, label %bb10
 		 i32 102, label %bb5
@@ -21,7 +21,7 @@ bb5:		; preds = %bb5, %bb5.outer
 	]
 
 bb10:		; preds = %bb5
-	%tmp12 = tail call i32 (...)* @bar( i32 %undo.0.ph ) nounwind 		; <i32> [#uses=0]
+	%tmp12 = tail call i32 (...) @bar( i32 %undo.0.ph ) nounwind 		; <i32> [#uses=0]
 	br label %UnifiedReturnBlock
 
 bb13:		; preds = %bb5
diff --git a/test/Transforms/SimplifyCFG/2008-01-02-hoist-fp-add.ll b/test/Transforms/SimplifyCFG/2008-01-02-hoist-fp-add.ll
index cf29b7159793..7625d935ab3d 100644
--- a/test/Transforms/SimplifyCFG/2008-01-02-hoist-fp-add.ll
+++ b/test/Transforms/SimplifyCFG/2008-01-02-hoist-fp-add.ll
@@ -1,27 +1,27 @@
-; The phi should not be eliminated in this case, because the fp op could trap.
+; The phi should not be eliminated in this case, because the divide op could trap.
 ; RUN: opt < %s -simplifycfg -S | FileCheck %s
 
 target datalayout = "e-p:32:32:32-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:32:64-f32:32:32-f64:32:64-v64:64:64-v128:128:128-a0:0:64-f80:128:128"
 target triple = "i686-apple-darwin8"
-@G = weak global double 0.000000e+00, align 8		; <double*> [#uses=2]
+@G = weak global i32 0, align 8		; <i32*> [#uses=2]
 
-define void @test(i32 %X, i32 %Y, double %Z) {
+define void @test(i32 %X, i32 %Y, i32 %Z) {
 entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
-	%tmp = load double* @G, align 8		; <double> [#uses=2]
+	%tmp = load i32, i32* @G, align 8		; <i32> [#uses=2]
 	%tmp3 = icmp eq i32 %X, %Y		; <i1> [#uses=1]
 	%tmp34 = zext i1 %tmp3 to i8		; <i8> [#uses=1]
 	%toBool = icmp ne i8 %tmp34, 0		; <i1> [#uses=1]
 	br i1 %toBool, label %cond_true, label %cond_next
 
 cond_true:		; preds = %entry
-	%tmp7 = fadd double %tmp, %Z		; <double> [#uses=1]
+	%tmp7 = udiv i32 %tmp, %Z		; <i32> [#uses=1]
 	br label %cond_next
 
 cond_next:		; preds = %cond_true, %entry
-; CHECK: = phi double
-	%F.0 = phi double [ %tmp, %entry ], [ %tmp7, %cond_true ]		; <double> [#uses=1]
-	store double %F.0, double* @G, align 8
+; CHECK: = phi i32
+	%F.0 = phi i32 [ %tmp, %entry ], [ %tmp7, %cond_true ]		; <i32> [#uses=1]
+	store i32 %F.0, i32* @G, align 8
 	ret void
 }
 
diff --git a/test/Transforms/SimplifyCFG/2008-07-13-InfLoopMiscompile.ll b/test/Transforms/SimplifyCFG/2008-07-13-InfLoopMiscompile.ll
index 9b6084f0e1af..154677b0747f 100644
--- a/test/Transforms/SimplifyCFG/2008-07-13-InfLoopMiscompile.ll
+++ b/test/Transforms/SimplifyCFG/2008-07-13-InfLoopMiscompile.ll
@@ -9,7 +9,7 @@ target triple = "i386-pc-linux-gnu"
 
 define i32 @main() nounwind  {
 entry:
-	%l = load i32* @g_37, align 4		; <i32> [#uses=1]
+	%l = load i32, i32* @g_37, align 4		; <i32> [#uses=1]
 	%cmpa = icmp ne i32 %l, 0		; <i1> [#uses=3]
 	br i1 %cmpa, label %func_1.exit, label %mooseblock
 
@@ -29,7 +29,7 @@ cowblock:		; preds = %beeblock, %monkeyblock
 
 func_1.exit:		; preds = %cowblock, %entry
 	%outval = phi i32 [ %cowval, %cowblock ], [ 1, %entry ]		; <i32> [#uses=1]
-	%pout = tail call i32 (i8*, ...)* @printf( i8* noalias  getelementptr ([4 x i8]* @.str, i32 0, i32 0), i32 %outval ) nounwind 		; <i32> [#uses=0]
+	%pout = tail call i32 (i8*, ...) @printf( i8* noalias  getelementptr ([4 x i8], [4 x i8]* @.str, i32 0, i32 0), i32 %outval ) nounwind 		; <i32> [#uses=0]
 	ret i32 0
 }
 
diff --git a/test/Transforms/SimplifyCFG/2008-09-08-MultiplePred.ll b/test/Transforms/SimplifyCFG/2008-09-08-MultiplePred.ll
index ac9622d43c31..6b216f598efb 100644
--- a/test/Transforms/SimplifyCFG/2008-09-08-MultiplePred.ll
+++ b/test/Transforms/SimplifyCFG/2008-09-08-MultiplePred.ll
@@ -4,7 +4,7 @@
 
 define i32 @func_127(i32 %p_129) nounwind {
 entry:
-	load i32* @g_103, align 4		; <i32>:0 [#uses=1]
+	load i32, i32* @g_103, align 4		; <i32>:0 [#uses=1]
 	icmp eq i32 %0, 0		; <i1>:1 [#uses=2]
 	br i1 %1, label %bb6.preheader, label %entry.return_crit_edge
 
diff --git a/test/Transforms/SimplifyCFG/2008-12-16-DCECond.ll b/test/Transforms/SimplifyCFG/2008-12-16-DCECond.ll
index 727102435fc7..4fc21d942237 100644
--- a/test/Transforms/SimplifyCFG/2008-12-16-DCECond.ll
+++ b/test/Transforms/SimplifyCFG/2008-12-16-DCECond.ll
@@ -9,7 +9,7 @@ entry:
 	br i1 %cmp, label %ifthen, label %ifend
 
 ifthen:		; preds = %entry
-	%call = call i32 (...)* @foo()		; <i32> [#uses=0]
+	%call = call i32 (...) @foo()		; <i32> [#uses=0]
 	br label %ifend
 
 ifend:		; preds = %ifthen, %entry
@@ -17,7 +17,7 @@ ifend:		; preds = %ifthen, %entry
 	br i1 %cmp2, label %ifthen3, label %ifend5
 
 ifthen3:		; preds = %ifend
-	%call4 = call i32 (...)* @foo()		; <i32> [#uses=0]
+	%call4 = call i32 (...) @foo()		; <i32> [#uses=0]
 	br label %ifend5
 
 ifend5:		; preds = %ifthen3, %ifend
@@ -25,7 +25,7 @@ ifend5:		; preds = %ifthen3, %ifend
 	br i1 %cmp7, label %ifthen8, label %ifend10
 
 ifthen8:		; preds = %ifend5
-	%call9 = call i32 (...)* @bar()		; <i32> [#uses=0]
+	%call9 = call i32 (...) @bar()		; <i32> [#uses=0]
 	br label %ifend10
 
 ifend10:		; preds = %ifthen8, %ifend5
@@ -33,7 +33,7 @@ ifend10:		; preds = %ifthen8, %ifend5
 	br i1 %cmp12, label %ifthen13, label %ifend15
 
 ifthen13:		; preds = %ifend10
-	%call14 = call i32 (...)* @bar()		; <i32> [#uses=0]
+	%call14 = call i32 (...) @bar()		; <i32> [#uses=0]
 	br label %ifend15
 
 ifend15:		; preds = %ifthen13, %ifend10
diff --git a/test/Transforms/SimplifyCFG/2009-05-12-externweak.ll b/test/Transforms/SimplifyCFG/2009-05-12-externweak.ll
index 419feb6b56a1..faf3f5f8b2a7 100644
--- a/test/Transforms/SimplifyCFG/2009-05-12-externweak.ll
+++ b/test/Transforms/SimplifyCFG/2009-05-12-externweak.ll
@@ -18,7 +18,7 @@ bb:		; preds = %bb4
 	br i1 icmp ne (i32* @i, i32* null), label %bb1, label %bb2
 
 bb1:		; preds = %bb
-	%0 = load i32* @i, align 4		; <i32> [#uses=1]
+	%0 = load i32, i32* @i, align 4		; <i32> [#uses=1]
 	br label %bb3
 
 bb2:		; preds = %bb
diff --git a/test/Transforms/SimplifyCFG/AArch64/lit.local.cfg b/test/Transforms/SimplifyCFG/AArch64/lit.local.cfg
new file mode 100644
index 000000000000..6642d2870680
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/AArch64/lit.local.cfg
@@ -0,0 +1,5 @@
+config.suffixes = ['.ll']
+
+targets = set(config.root.targets_to_build.split())
+if not 'AArch64' in targets:
+    config.unsupported = True
diff --git a/test/Transforms/SimplifyCFG/AArch64/prefer-fma.ll b/test/Transforms/SimplifyCFG/AArch64/prefer-fma.ll
new file mode 100644
index 000000000000..cfbe219c4c2d
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/AArch64/prefer-fma.ll
@@ -0,0 +1,72 @@
+; RUN: opt < %s -mtriple=aarch64-linux-gnu -simplifycfg -enable-unsafe-fp-math -S >%t
+; RUN: FileCheck %s < %t
+; ModuleID = 't.cc'
+
+; Function Attrs: nounwind
+define double @_Z3fooRdS_S_S_(double* dereferenceable(8) %x, double* dereferenceable(8) %y, double* dereferenceable(8) %a) #0 {
+entry:
+  %0 = load double, double* %y, align 8
+  %cmp = fcmp oeq double %0, 0.000000e+00
+  %1 = load double, double* %x, align 8
+  br i1 %cmp, label %if.then, label %if.else
+
+; fadd (const, (fmul x, y))
+if.then:                                          ; preds = %entry
+; CHECK-LABEL: if.then:
+; CHECK:   %3 = fmul fast double %1, %2
+; CHECK-NEXT:   %mul = fadd fast double 1.000000e+00, %3
+  %2 = load double, double* %a, align 8
+  %3 = fmul fast double %1, %2
+  %mul = fadd fast double 1.000000e+00, %3
+  store double %mul, double* %y, align 8
+  br label %if.end
+
+; fsub ((fmul x, y), z)
+if.else:                                          ; preds = %entry
+; CHECK-LABEL: if.else:
+; CHECK:   %mul1 = fmul fast double %1, %2
+; CHECK-NEXT:   %sub1 = fsub fast double %mul1, %0
+  %4 = load double, double* %a, align 8
+  %mul1 = fmul fast double %1, %4
+  %sub1 = fsub fast double %mul1, %0
+  store double %sub1, double* %y, align 8
+  br label %if.end
+
+if.end:                                           ; preds = %if.else, %if.then
+  %5 = load double, double* %y, align 8
+  %cmp2 = fcmp oeq double %5, 2.000000e+00
+  %6 = load double, double* %x, align 8
+  br i1 %cmp2, label %if.then2, label %if.else2
+
+; fsub (x, (fmul y, z))
+if.then2:                                         ; preds = %entry
+; CHECK-LABEL: if.then2:
+; CHECK:   %7 = fmul fast double %5, 3.000000e+00
+; CHECK-NEXT:   %mul2 = fsub fast double %6, %7
+  %7 = load double, double* %a, align 8
+  %8 = fmul fast double %6, 3.0000000e+00
+  %mul2 = fsub fast double %7, %8
+  store double %mul2, double* %y, align 8
+  br label %if.end2
+
+; fsub (fneg((fmul x, y)), const)
+if.else2:                                         ; preds = %entry
+; CHECK-LABEL: if.else2:
+; CHECK:   %mul3 = fmul fast double %5, 3.000000e+00
+; CHECK-NEXT:   %neg = fsub fast double 0.000000e+00, %mul3
+; CHECK-NEXT:   %sub2 = fsub fast double %neg, 3.000000e+00
+  %mul3 = fmul fast double %6, 3.0000000e+00
+  %neg = fsub fast double 0.0000000e+00, %mul3
+  %sub2 = fsub fast double %neg, 3.0000000e+00
+  store double %sub2, double* %y, align 8
+  br label %if.end2
+
+if.end2:                                           ; preds = %if.else, %if.then
+  %9 = load double, double* %x, align 8
+  %10 = load double, double* %y, align 8
+  %add = fadd fast double %9, %10
+  %11 = load double, double* %a, align 8
+  %add2 = fadd fast double %add, %11
+  ret double %add2
+}
+
diff --git a/test/Transforms/SimplifyCFG/EmptyBlockMerge.ll b/test/Transforms/SimplifyCFG/EmptyBlockMerge.ll
index aba08dc073a8..32a0202364d9 100644
--- a/test/Transforms/SimplifyCFG/EmptyBlockMerge.ll
+++ b/test/Transforms/SimplifyCFG/EmptyBlockMerge.ll
@@ -6,7 +6,7 @@ declare void @foo()
 
 define void @cprop_test12(i32* %data) {
 bb0:
-	%reg108 = load i32* %data		; <i32> [#uses=2]
+	%reg108 = load i32, i32* %data		; <i32> [#uses=2]
 	%cond218 = icmp ne i32 %reg108, 5		; <i1> [#uses=1]
 	br i1 %cond218, label %bb3, label %bb2
 bb2:		; preds = %bb0
diff --git a/test/Transforms/SimplifyCFG/MagicPointer.ll b/test/Transforms/SimplifyCFG/MagicPointer.ll
index b8b8cbd9dc63..7789600f3f0c 100644
--- a/test/Transforms/SimplifyCFG/MagicPointer.ll
+++ b/test/Transforms/SimplifyCFG/MagicPointer.ll
@@ -33,7 +33,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.else
 
 if.then:                                          ; preds = %entry
-  %call = call i32 @puts(i8* getelementptr inbounds ([5 x i8]* @.str, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call = call i32 @puts(i8* getelementptr inbounds ([5 x i8], [5 x i8]* @.str, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end21
 
 if.else:                                          ; preds = %entry
@@ -41,7 +41,7 @@ if.else:                                          ; preds = %entry
   br i1 %cmp, label %if.then2, label %if.else4
 
 if.then2:                                         ; preds = %if.else
-  %call3 = call i32 @puts(i8* getelementptr inbounds ([4 x i8]* @.str1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call3 = call i32 @puts(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end20
 
 if.else4:                                         ; preds = %if.else
@@ -53,7 +53,7 @@ lor.lhs.false:                                    ; preds = %if.else4
   br i1 %cmp8, label %if.then9, label %if.else11
 
 if.then9:                                         ; preds = %lor.lhs.false, %if.else4
-  %call10 = call i32 @puts(i8* getelementptr inbounds ([4 x i8]* @.str2, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call10 = call i32 @puts(i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str2, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end19
 
 if.else11:                                        ; preds = %lor.lhs.false
@@ -61,7 +61,7 @@ if.else11:                                        ; preds = %lor.lhs.false
   br i1 %cmp13, label %if.then14, label %if.else16
 
 if.then14:                                        ; preds = %if.else11
-  %call15 = call i32 @puts(i8* getelementptr inbounds ([5 x i8]* @.str3, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call15 = call i32 @puts(i8* getelementptr inbounds ([5 x i8], [5 x i8]* @.str3, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end
 
 if.else16:                                        ; preds = %if.else11
@@ -99,7 +99,7 @@ entry:
   br i1 %tobool, label %if.then, label %if.else
 
 if.then:                                          ; preds = %entry
-  %call = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([5 x i8] addrspace(1)* @.str_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([5 x i8], [5 x i8] addrspace(1)* @.str_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end21
 
 if.else:                                          ; preds = %entry
@@ -107,7 +107,7 @@ if.else:                                          ; preds = %entry
   br i1 %cmp, label %if.then2, label %if.else4
 
 if.then2:                                         ; preds = %if.else
-  %call3 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([4 x i8] addrspace(1)* @.str1_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call3 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([4 x i8], [4 x i8] addrspace(1)* @.str1_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end20
 
 if.else4:                                         ; preds = %if.else
@@ -119,7 +119,7 @@ lor.lhs.false:                                    ; preds = %if.else4
   br i1 %cmp8, label %if.then9, label %if.else11
 
 if.then9:                                         ; preds = %lor.lhs.false, %if.else4
-  %call10 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([4 x i8] addrspace(1)* @.str2_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call10 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([4 x i8], [4 x i8] addrspace(1)* @.str2_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end19
 
 if.else11:                                        ; preds = %lor.lhs.false
@@ -127,7 +127,7 @@ if.else11:                                        ; preds = %lor.lhs.false
   br i1 %cmp13, label %if.then14, label %if.else16
 
 if.then14:                                        ; preds = %if.else11
-  %call15 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([5 x i8] addrspace(1)* @.str3_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
+  %call15 = call i32 @puts_as1(i8 addrspace(1)* getelementptr inbounds ([5 x i8], [5 x i8] addrspace(1)* @.str3_as1, i64 0, i64 0)) nounwind ; <i32> [#uses=0]
   br label %if.end
 
 if.else16:                                        ; preds = %if.else11
diff --git a/test/Transforms/SimplifyCFG/PR17073.ll b/test/Transforms/SimplifyCFG/PR17073.ll
index 8dc9fb28d617..e6e98b266879 100644
--- a/test/Transforms/SimplifyCFG/PR17073.ll
+++ b/test/Transforms/SimplifyCFG/PR17073.ll
@@ -18,7 +18,7 @@ target triple = "i386-apple-macosx10.9.0"
 ; CHECK-NOT: select i1 %tobool, i32* null, i32* select (i1 icmp eq (i64 urem (i64 2, i64 zext (i1 icmp eq (i32* bitcast (i8* @b to i32*), i32* @a) to i64)), i64 0), i32* null, i32* @a) 
 define i32* @can_trap1() {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %exit, label %block1
 
@@ -38,7 +38,7 @@ exit:
 ; CHECK-NOT: select i1 %tobool, i32* select (i1 icmp eq (i64 urem (i64 2, i64 zext (i1 icmp eq (i32* bitcast (i8* @b to i32*), i32* @a) to i64)), i64 0), i32* null, i32* @a), i32* null
 define i32* @can_trap2() {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %exit, label %block1
 
@@ -57,7 +57,7 @@ exit:
 ; CHECK: select i1 icmp eq (i32* bitcast (i8* @b to i32*), i32* @a), i32* select (i1 icmp eq (i64 add (i64 zext (i1 icmp eq (i32* bitcast (i8* @b to i32*), i32* @a) to i64), i64 2), i64 0), i32* null, i32* @a), i32* null
 define i32* @cannot_trap() {
 entry:
-  %0 = load i32* @a, align 4
+  %0 = load i32, i32* @a, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %exit, label %block1
 
diff --git a/test/Transforms/SimplifyCFG/PhiBlockMerge.ll b/test/Transforms/SimplifyCFG/PhiBlockMerge.ll
index 36b52f52d490..555082921b96 100644
--- a/test/Transforms/SimplifyCFG/PhiBlockMerge.ll
+++ b/test/Transforms/SimplifyCFG/PhiBlockMerge.ll
@@ -4,9 +4,7 @@
 ;
 
 define i32 @test(i1 %a, i1 %b) {
-; CHECK: br i1 %a
         br i1 %a, label %M, label %O
-; CHECK: O:
 O:              ; preds = %0
 ; CHECK: select i1 %b, i32 0, i32 1
 ; CHECK-NOT: phi
@@ -18,9 +16,9 @@ N:              ; preds = %Q, %O
         %Wp = phi i32 [ 0, %O ], [ 1, %Q ]              ; <i32> [#uses=1]
         br label %M
 M:              ; preds = %N, %0
-; CHECK: %W = phi i32
         %W = phi i32 [ %Wp, %N ], [ 2, %0 ]             ; <i32> [#uses=1]
         %R = add i32 %W, 1              ; <i32> [#uses=1]
         ret i32 %R
+; CHECK: ret
 }
 
diff --git a/test/Transforms/SimplifyCFG/PowerPC/cttz-ctlz-spec.ll b/test/Transforms/SimplifyCFG/PowerPC/cttz-ctlz-spec.ll
new file mode 100644
index 000000000000..fa74549710f4
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/PowerPC/cttz-ctlz-spec.ll
@@ -0,0 +1,45 @@
+; RUN: opt -S -simplifycfg < %s | FileCheck %s
+target datalayout = "E-m:e-i64:64-n32:64"
+target triple = "powerpc64-unknown-linux-gnu"
+
+define i64 @test1(i64 %A) {
+; CHECK-LABEL: @test1(
+; CHECK: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; CHECK-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+; CHECK-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 64, i64 [[CTLZ]]
+; CHECK-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+define i64 @test1b(i64 %A) {
+; CHECK-LABEL: @test1b(
+; CHECK: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; CHECK-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+; CHECK-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 64, i64 [[CTLZ]]
+; CHECK-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+declare i64 @llvm.ctlz.i64(i64, i1)
+declare i64 @llvm.cttz.i64(i64, i1)
+
diff --git a/test/Transforms/SimplifyCFG/PowerPC/lit.local.cfg b/test/Transforms/SimplifyCFG/PowerPC/lit.local.cfg
new file mode 100644
index 000000000000..091332439b18
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/PowerPC/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'PowerPC' in config.root.targets:
+    config.unsupported = True
diff --git a/test/Transforms/SimplifyCFG/R600/cttz-ctlz.ll b/test/Transforms/SimplifyCFG/R600/cttz-ctlz.ll
new file mode 100644
index 000000000000..5b2799494647
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/R600/cttz-ctlz.ll
@@ -0,0 +1,249 @@
+; RUN: opt -S -simplifycfg -mtriple=r600-unknown-unknown -mcpu=SI < %s | FileCheck -check-prefix=SI -check-prefix=ALL %s
+; RUN: opt -S -simplifycfg -mtriple=r600-unknown-unknown -mcpu=tonga < %s | FileCheck -check-prefix=SI -check-prefix=ALL %s
+
+
+define i64 @test1(i64 %A) {
+; ALL-LABEL: @test1(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; SI-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 64, i64 [[CTLZ]]
+; SI-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+
+define i32 @test2(i32 %A) {
+; ALL-LABEL: @test2(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; SI-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i32 32, i32 [[CTLZ]]
+; SI-NEXT: ret i32 [[SEL]]
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 32, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3(i16 signext %A) {
+; ALL-LABEL: @test3(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; SI-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i16 16, i16 [[CTLZ]]
+; SI-NEXT: ret i16 [[SEL]]
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 16, %entry ]
+  ret i16 %cond
+}
+
+
+define i64 @test1b(i64 %A) {
+; ALL-LABEL: @test1b(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; SI-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 64, i64 [[CTTZ]]
+; SI-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+
+define i32 @test2b(i32 %A) {
+; ALL-LABEL: @test2b(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; SI-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i32 32, i32 [[CTTZ]]
+; SI-NEXT: ret i32 [[SEL]]
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 32, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3b(i16 signext %A) {
+; ALL-LABEL: @test3b(
+; SI: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; SI-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+; SI-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i16 16, i16 [[CTTZ]]
+; SI-NEXT: ret i16 [[SEL]]
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 16, %entry ]
+  ret i16 %cond
+}
+
+
+define i64 @test1c(i64 %A) {
+; ALL-LABEL: @test1c(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; ALL-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 63, i64 [[CTLZ]]
+; ALL-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 63, %entry ]
+  ret i64 %cond
+}
+
+define i32 @test2c(i32 %A) {
+; ALL-LABEL: @test2c(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; ALL-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i32 31, i32 [[CTLZ]]
+; ALL-NEXT: ret i32 [[SEL]]
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 31, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3c(i16 signext %A) {
+; ALL-LABEL: @test3c(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; ALL-NEXT: [[CTLZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i16 15, i16 [[CTLZ]]
+; ALL-NEXT: ret i16 [[SEL]]
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 15, %entry ]
+  ret i16 %cond
+}
+
+
+define i64 @test1d(i64 %A) {
+; ALL-LABEL: @test1d(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; ALL-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i64 63, i64 [[CTTZ]]
+; ALL-NEXT: ret i64 [[SEL]]
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 63, %entry ]
+  ret i64 %cond
+}
+
+
+define i32 @test2d(i32 %A) {
+; ALL-LABEL: @test2d(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; ALL-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i32 31, i32 [[CTTZ]]
+; ALL-NEXT: ret i32 [[SEL]]
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 31, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3d(i16 signext %A) {
+; ALL-LABEL: @test3d(
+; ALL: [[ICMP:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; ALL-NEXT: [[CTTZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+; ALL-NEXT: [[SEL:%[A-Za-z0-9.]+]] = select i1 [[ICMP]], i16 15, i16 [[CTTZ]]
+; ALL-NEXT: ret i16 [[SEL]]
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 15, %entry ]
+  ret i16 %cond
+}
+
+
+declare i64 @llvm.ctlz.i64(i64, i1)
+declare i32 @llvm.ctlz.i32(i32, i1)
+declare i16 @llvm.ctlz.i16(i16, i1)
+declare i64 @llvm.cttz.i64(i64, i1)
+declare i32 @llvm.cttz.i32(i32, i1)
+declare i16 @llvm.cttz.i16(i16, i1)
diff --git a/test/Transforms/SimplifyCFG/R600/lit.local.cfg b/test/Transforms/SimplifyCFG/R600/lit.local.cfg
new file mode 100644
index 000000000000..ad9ce2541ef7
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/R600/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'R600' in config.root.targets:
+    config.unsupported = True
diff --git a/test/Transforms/SimplifyCFG/SpeculativeExec.ll b/test/Transforms/SimplifyCFG/SpeculativeExec.ll
index 83fa41900035..c23a96df52ee 100644
--- a/test/Transforms/SimplifyCFG/SpeculativeExec.ll
+++ b/test/Transforms/SimplifyCFG/SpeculativeExec.ll
@@ -28,33 +28,17 @@ bb3:		; preds = %bb2, %entry
 	ret i32 %tmp5
 }
 
-declare i8 @llvm.cttz.i8(i8, i1)
-
-define i8 @test2(i8 %a) {
-; CHECK-LABEL: @test2(
-  br i1 undef, label %bb_true, label %bb_false
-bb_true:
-  %b = tail call i8 @llvm.cttz.i8(i8 %a, i1 false)
-  br label %join
-bb_false:
-  br label %join
-join:
-  %c = phi i8 [%b, %bb_true], [%a, %bb_false]
-; CHECK: select
-  ret i8 %c
-}
-
 define i8* @test4(i1* %dummy, i8* %a, i8* %b) {
 ; Test that we don't speculate an arbitrarily large number of unfolded constant
 ; expressions.
 ; CHECK-LABEL: @test4(
 
 entry:
-  %cond1 = load volatile i1* %dummy
+  %cond1 = load volatile i1, i1* %dummy
   br i1 %cond1, label %if, label %end
 
 if:
-  %cond2 = load volatile i1* %dummy
+  %cond2 = load volatile i1, i1* %dummy
   br i1 %cond2, label %then, label %end
 
 then:
diff --git a/test/Transforms/SimplifyCFG/UnreachableEliminate.ll b/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
index 21428c62f531..8718c552c693 100644
--- a/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
+++ b/test/Transforms/SimplifyCFG/UnreachableEliminate.ll
@@ -22,11 +22,15 @@ entry:
         invoke void @test2( )
                         to label %N unwind label %U
 U:
+  %res = landingpad { i8* } personality i32 (...)* @__gxx_personality_v0
+          cleanup
         unreachable
 N:
         ret void
 }
 
+declare i32 @__gxx_personality_v0(...)
+
 define i32 @test3(i32 %v) {
 ; CHECK-LABEL: @test3(
 ; CHECK: entry:
@@ -46,32 +50,6 @@ T:
         ret i32 2
 }
 
-; PR9450
-define i32 @test4(i32 %v, i32 %w) {
-; CHECK: entry:
-; CHECK-NEXT:  switch i32 %v, label %T [
-; CHECK-NEXT:    i32 3, label %V
-; CHECK-NEXT:    i32 2, label %U
-; CHECK-NEXT:  ]
-
-entry:
-        br label %SWITCH
-V:
-        ret i32 7
-SWITCH:
-        switch i32 %v, label %default [
-                 i32 1, label %T
-                 i32 2, label %U
-                 i32 3, label %V
-        ]
-default:
-        unreachable
-U:
-        ret i32 %w
-T:
-        ret i32 2
-}
-
 
 ;; We can either convert the following control-flow to a select or remove the
 ;; unreachable control flow because of the undef store of null. Make sure we do
diff --git a/test/Transforms/SimplifyCFG/X86/speculate-cttz-ctlz.ll b/test/Transforms/SimplifyCFG/X86/speculate-cttz-ctlz.ll
new file mode 100644
index 000000000000..69f6c69059d4
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/X86/speculate-cttz-ctlz.ll
@@ -0,0 +1,330 @@
+; RUN: opt -S -simplifycfg -mtriple=x86_64-unknown-unknown -mattr=+bmi < %s | FileCheck %s --check-prefix=ALL --check-prefix=BMI
+; RUN: opt -S -simplifycfg -mtriple=x86_64-unknown-unknown -mattr=+lzcnt < %s | FileCheck %s --check-prefix=ALL --check-prefix=LZCNT
+; RUN: opt -S -simplifycfg -mtriple=x86_64-unknown-unknown < %s | FileCheck %s --check-prefix=ALL --check-prefix=GENERIC
+
+
+define i64 @test1(i64 %A) {
+; ALL-LABEL: @test1(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+; LZCNT-NEXT: select i1 [[COND]], i64 64, i64 [[CTLZ]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+define i32 @test2(i32 %A) {
+; ALL-LABEL: @test2(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+; LZCNT-NEXT: select i1 [[COND]], i32 32, i32 [[CTLZ]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 32, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3(i16 signext %A) {
+; ALL-LABEL: @test3(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+; LZCNT-NEXT: select i1 [[COND]], i16 16, i16 [[CTLZ]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.ctlz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 16, %entry ]
+  ret i16 %cond
+}
+
+
+define i64 @test1b(i64 %A) {
+; ALL-LABEL: @test1b(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %A, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+; BMI-NEXT: select i1 [[COND]], i64 64, i64 [[CTTZ]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %0, %cond.true ], [ 64, %entry ]
+  ret i64 %cond
+}
+
+
+define i32 @test2b(i32 %A) {
+; ALL-LABEL: @test2b(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %A, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+; BMI-NEXT: select i1 [[COND]], i32 32, i32 [[CTTZ]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.cttz.i32(i32 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %0, %cond.true ], [ 32, %entry ]
+  ret i32 %cond
+}
+
+
+define signext i16 @test3b(i16 signext %A) {
+; ALL-LABEL: @test3b(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i16 %A, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+; BMI-NEXT: select i1 [[COND]], i16 16, i16 [[CTTZ]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i16 %A, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i16 @llvm.cttz.i16(i16 %A, i1 true)
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %0, %cond.true ], [ 16, %entry ]
+  ret i16 %cond
+}
+
+; The following tests verify that calls to cttz/ctlz are speculated even if
+; basic block %cond.true has an extra zero extend/truncate which is "free"
+; for the target.
+
+define i64 @test1e(i32 %x) {
+; ALL-LABEL: @test1e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %x, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+; ALL: [[ZEXT:%[A-Za-z0-9]+]] = zext i32 [[CTTZ]] to i64
+; BMI-NEXT: select i1 [[COND]], i64 32, i64 [[ZEXT]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %phitmp2 = zext i32 %0 to i64
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %phitmp2, %cond.true ], [ 32, %entry ]
+  ret i64 %cond
+}
+
+define i32 @test2e(i64 %x) {
+; ALL-LABEL: @test2e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %x, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i64 [[CTTZ]] to i32
+; BMI-NEXT: select i1 [[COND]], i32 64, i32 [[TRUNC]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i32
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %cast, %cond.true ], [ 64, %entry ]
+  ret i32 %cond
+}
+
+define i64 @test3e(i32 %x) {
+; ALL-LABEL: @test3e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %x, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+; ALL: [[ZEXT:%[A-Za-z0-9]+]] = zext i32 [[CTLZ]] to i64
+; LZCNT-NEXT: select i1 [[COND]], i64 32, i64 [[ZEXT]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %phitmp2 = zext i32 %0 to i64
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i64 [ %phitmp2, %cond.true ], [ 32, %entry ]
+  ret i64 %cond
+}
+
+define i32 @test4e(i64 %x) {
+; ALL-LABEL: @test4e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %x, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i64 [[CTLZ]] to i32
+; LZCNT-NEXT: select i1 [[COND]], i32 64, i32 [[TRUNC]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i32
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i32 [ %cast, %cond.true ], [ 64, %entry ]
+  ret i32 %cond
+}
+
+define i16 @test5e(i64 %x) {
+; ALL-LABEL: @test5e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %x, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i64 [[CTLZ]] to i16
+; LZCNT-NEXT: select i1 [[COND]], i16 64, i16 [[TRUNC]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.ctlz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i16
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %cast, %cond.true ], [ 64, %entry ]
+  ret i16 %cond
+}
+
+define i16 @test6e(i32 %x) {
+; ALL-LABEL: @test6e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %x, 0
+; ALL: [[CTLZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i32 [[CTLZ]] to i16
+; LZCNT-NEXT: select i1 [[COND]], i16 32, i16 [[TRUNC]]
+; BMI-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.ctlz.i32(i32 %x, i1 true)
+  %cast = trunc i32 %0 to i16
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %cast, %cond.true ], [ 32, %entry ]
+  ret i16 %cond
+}
+
+define i16 @test7e(i64 %x) {
+; ALL-LABEL: @test7e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i64 %x, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i64 [[CTTZ]] to i16
+; BMI-NEXT: select i1 [[COND]], i16 64, i16 [[TRUNC]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i64 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i64 @llvm.cttz.i64(i64 %x, i1 true)
+  %cast = trunc i64 %0 to i16
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %cast, %cond.true ], [ 64, %entry ]
+  ret i16 %cond
+}
+
+define i16 @test8e(i32 %x) {
+; ALL-LABEL: @test8e(
+; ALL: [[COND:%[A-Za-z0-9]+]] = icmp eq i32 %x, 0
+; ALL: [[CTTZ:%[A-Za-z0-9]+]] = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+; ALL: [[TRUNC:%[A-Za-z0-9]+]] = trunc i32 [[CTTZ]] to i16
+; BMI-NEXT: select i1 [[COND]], i16 32, i16 [[TRUNC]]
+; LZCNT-NOT: select
+; GENERIC-NOT: select
+; ALL: ret
+entry:
+  %tobool = icmp eq i32 %x, 0
+  br i1 %tobool, label %cond.end, label %cond.true
+
+cond.true:                                        ; preds = %entry
+  %0 = tail call i32 @llvm.cttz.i32(i32 %x, i1 true)
+  %cast = trunc i32 %0 to i16
+  br label %cond.end
+
+cond.end:                                         ; preds = %entry, %cond.true
+  %cond = phi i16 [ %cast, %cond.true ], [ 32, %entry ]
+  ret i16 %cond
+}
+
+
+declare i64 @llvm.ctlz.i64(i64, i1)
+declare i32 @llvm.ctlz.i32(i32, i1)
+declare i16 @llvm.ctlz.i16(i16, i1)
+declare i64 @llvm.cttz.i64(i64, i1)
+declare i32 @llvm.cttz.i32(i32, i1)
+declare i16 @llvm.cttz.i16(i16, i1)
diff --git a/test/Transforms/SimplifyCFG/X86/switch-covered-bug.ll b/test/Transforms/SimplifyCFG/X86/switch-covered-bug.ll
index 22599b397ed4..f3e5506ad933 100644
--- a/test/Transforms/SimplifyCFG/X86/switch-covered-bug.ll
+++ b/test/Transforms/SimplifyCFG/X86/switch-covered-bug.ll
@@ -9,8 +9,8 @@ target triple = "x86_64-apple-darwin12.0.0"
 ; CHECK: entry:
 ; CHECK-NEXT: sub i3 %arg, -4
 ; CHECK-NEXT: zext i3 %switch.tableidx to i4
-; CHECK-NEXT: getelementptr inbounds [8 x i64]* @switch.table, i32 0, i4 %switch.tableidx.zext
-; CHECK-NEXT: load i64* %switch.gep
+; CHECK-NEXT: getelementptr inbounds [8 x i64], [8 x i64]* @switch.table, i32 0, i4 %switch.tableidx.zext
+; CHECK-NEXT: load i64, i64* %switch.gep
 ; CHECK-NEXT: add i64
 ; CHECK-NEXT: ret i64
 define i64 @test(i3 %arg) {
diff --git a/test/Transforms/SimplifyCFG/X86/switch-table-bug.ll b/test/Transforms/SimplifyCFG/X86/switch-table-bug.ll
index d0b8ab29d26b..26008700f5be 100644
--- a/test/Transforms/SimplifyCFG/X86/switch-table-bug.ll
+++ b/test/Transforms/SimplifyCFG/X86/switch-table-bug.ll
@@ -8,8 +8,8 @@ target triple = "x86_64-apple-darwin12.0.0"
 ; CHECK: entry:
 ; CHECK-NEXT: sub i2 %0, -2
 ; CHECK-NEXT: zext i2 %switch.tableidx to i3
-; CHECK-NEXT: getelementptr inbounds [4 x i64]* @switch.table, i32 0, i3 %switch.tableidx.zext
-; CHECK-NEXT: load i64* %switch.gep
+; CHECK-NEXT: getelementptr inbounds [4 x i64], [4 x i64]* @switch.table, i32 0, i3 %switch.tableidx.zext
+; CHECK-NEXT: load i64, i64* %switch.gep
 ; CHECK-NEXT: ret i64 %switch.load
 define i64 @_TFO6reduce1E5toRawfS0_FT_Si(i2) {
 entry:
diff --git a/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll b/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll
index 22f18cd4d432..cae1a91bd43d 100644
--- a/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll
+++ b/test/Transforms/SimplifyCFG/X86/switch_to_lookup_table.ll
@@ -7,22 +7,22 @@ target triple = "x86_64-unknown-linux-gnu"
 ; CHECK: @switch.table = private unnamed_addr constant [7 x i32] [i32 55, i32 123, i32 0, i32 -1, i32 27, i32 62, i32 1]
 
 ; The float table for @h
-; CHECK: @switch.table1 = private unnamed_addr constant [4 x float] [float 0x40091EB860000000, float 0x3FF3BE76C0000000, float 0x4012449BA0000000, float 0x4001AE1480000000]
+; CHECK: @switch.table.1 = private unnamed_addr constant [4 x float] [float 0x40091EB860000000, float 0x3FF3BE76C0000000, float 0x4012449BA0000000, float 0x4001AE1480000000]
 
 ; The table for @foostring
-; CHECK: @switch.table2 = private unnamed_addr constant [4 x i8*] [i8* getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str1, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str2, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8]* @.str3, i64 0, i64 0)]
+; CHECK: @switch.table.2 = private unnamed_addr constant [4 x i8*] [i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str1, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str2, i64 0, i64 0), i8* getelementptr inbounds ([4 x i8], [4 x i8]* @.str3, i64 0, i64 0)]
 
 ; The table for @earlyreturncrash
-; CHECK: @switch.table3 = private unnamed_addr constant [4 x i32] [i32 42, i32 9, i32 88, i32 5]
+; CHECK: @switch.table.3 = private unnamed_addr constant [4 x i32] [i32 42, i32 9, i32 88, i32 5]
 
 ; The table for @large.
-; CHECK: @switch.table4 = private unnamed_addr constant [199 x i32] [i32 1, i32 4, i32 9,
+; CHECK: @switch.table.4 = private unnamed_addr constant [199 x i32] [i32 1, i32 4, i32 9,
 
 ; The table for @cprop
-; CHECK: @switch.table5 = private unnamed_addr constant [7 x i32] [i32 5, i32 42, i32 126, i32 -452, i32 128, i32 6, i32 7]
+; CHECK: @switch.table.5 = private unnamed_addr constant [7 x i32] [i32 5, i32 42, i32 126, i32 -452, i32 128, i32 6, i32 7]
 
-; The table for @unreachable
-; CHECK: @switch.table6 = private unnamed_addr constant [5 x i32] [i32 0, i32 0, i32 0, i32 1, i32 -1]
+; The table for @unreachable_case
+; CHECK: @switch.table.6 = private unnamed_addr constant [9 x i32] [i32 0, i32 0, i32 0, i32 2, i32 -1, i32 1, i32 1, i32 1, i32 1]
 
 ; A simple int-to-int selection switch.
 ; It is dense enough to be replaced by table lookup.
@@ -58,8 +58,8 @@ return:
 ; CHECK-NEXT: %0 = icmp ult i32 %switch.tableidx, 7
 ; CHECK-NEXT: br i1 %0, label %switch.lookup, label %return
 ; CHECK: switch.lookup:
-; CHECK-NEXT: %switch.gep = getelementptr inbounds [7 x i32]* @switch.table, i32 0, i32 %switch.tableidx
-; CHECK-NEXT: %switch.load = load i32* %switch.gep
+; CHECK-NEXT: %switch.gep = getelementptr inbounds [7 x i32], [7 x i32]* @switch.table, i32 0, i32 %switch.tableidx
+; CHECK-NEXT: %switch.load = load i32, i32* %switch.gep
 ; CHECK-NEXT: ret i32 %switch.load
 ; CHECK: return:
 ; CHECK-NEXT: ret i32 15
@@ -97,8 +97,8 @@ sw.epilog:
 ; CHECK-NEXT: %switch.shiftamt = mul i32 %switch.tableidx, 8
 ; CHECK-NEXT: %switch.downshift = lshr i32 89655594, %switch.shiftamt
 ; CHECK-NEXT: %switch.masked = trunc i32 %switch.downshift to i8
-; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x float]* @switch.table1, i32 0, i32 %switch.tableidx
-; CHECK-NEXT: %switch.load = load float* %switch.gep
+; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x float], [4 x float]* @switch.table.1, i32 0, i32 %switch.tableidx
+; CHECK-NEXT: %switch.load = load float, float* %switch.gep
 ; CHECK-NEXT: br label %sw.epilog
 ; CHECK: sw.epilog:
 ; CHECK-NEXT: %a.0 = phi i8 [ %switch.masked, %switch.lookup ], [ 7, %entry ]
@@ -131,11 +131,11 @@ sw.bb3: br label %return
 sw.default: br label %return
 
 return:
-  %retval.0 = phi i8* [ getelementptr inbounds ([6 x i8]* @.str4, i64 0, i64 0), %sw.default ],
-                      [ getelementptr inbounds ([4 x i8]* @.str3, i64 0, i64 0), %sw.bb3 ],
-                      [ getelementptr inbounds ([4 x i8]* @.str2, i64 0, i64 0), %sw.bb2 ],
-                      [ getelementptr inbounds ([4 x i8]* @.str1, i64 0, i64 0), %sw.bb1 ],
-                      [ getelementptr inbounds ([4 x i8]* @.str, i64 0, i64 0), %entry ]
+  %retval.0 = phi i8* [ getelementptr inbounds ([6 x i8], [6 x i8]* @.str4, i64 0, i64 0), %sw.default ],
+                      [ getelementptr inbounds ([4 x i8], [4 x i8]* @.str3, i64 0, i64 0), %sw.bb3 ],
+                      [ getelementptr inbounds ([4 x i8], [4 x i8]* @.str2, i64 0, i64 0), %sw.bb2 ],
+                      [ getelementptr inbounds ([4 x i8], [4 x i8]* @.str1, i64 0, i64 0), %sw.bb1 ],
+                      [ getelementptr inbounds ([4 x i8], [4 x i8]* @.str, i64 0, i64 0), %entry ]
   ret i8* %retval.0
 
 ; CHECK-LABEL: @foostring(
@@ -144,8 +144,8 @@ return:
 ; CHECK-NEXT: %0 = icmp ult i32 %switch.tableidx, 4
 ; CHECK-NEXT: br i1 %0, label %switch.lookup, label %return
 ; CHECK: switch.lookup:
-; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x i8*]* @switch.table2, i32 0, i32 %switch.tableidx
-; CHECK-NEXT: %switch.load = load i8** %switch.gep
+; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x i8*], [4 x i8*]* @switch.table.2, i32 0, i32 %switch.tableidx
+; CHECK-NEXT: %switch.load = load i8*, i8** %switch.gep
 ; CHECK-NEXT: ret i8* %switch.load
 }
 
@@ -173,8 +173,8 @@ sw.epilog:
 
 ; CHECK-LABEL: @earlyreturncrash(
 ; CHECK: switch.lookup:
-; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x i32]* @switch.table3, i32 0, i32 %switch.tableidx
-; CHECK-NEXT: %switch.load = load i32* %switch.gep
+; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x i32], [4 x i32]* @switch.table.3, i32 0, i32 %switch.tableidx
+; CHECK-NEXT: %switch.load = load i32, i32* %switch.gep
 ; CHECK-NEXT: ret i32 %switch.load
 ; CHECK: sw.epilog:
 ; CHECK-NEXT: ret i32 7
@@ -749,10 +749,10 @@ return:
 
 ; CHECK-LABEL: @cprop(
 ; CHECK: switch.lookup:
-; CHECK: %switch.gep = getelementptr inbounds [7 x i32]* @switch.table5, i32 0, i32 %switch.tableidx
+; CHECK: %switch.gep = getelementptr inbounds [7 x i32], [7 x i32]* @switch.table.5, i32 0, i32 %switch.tableidx
 }
 
-define i32 @unreachable(i32 %x)  {
+define i32 @unreachable_case(i32 %x)  {
 entry:
   switch i32 %x, label %sw.default [
     i32 0, label %sw.bb
@@ -770,15 +770,44 @@ sw.bb: br label %return
 sw.bb1: unreachable
 sw.bb2: br label %return
 sw.bb3: br label %return
-sw.default: unreachable
+sw.default: br label %return
 
 return:
-  %retval.0 = phi i32 [ 1, %sw.bb3 ], [ -1, %sw.bb2 ], [ 0, %sw.bb ]
+  %retval.0 = phi i32 [ 1, %sw.bb3 ], [ -1, %sw.bb2 ], [ 0, %sw.bb ], [ 2, %sw.default ]
   ret i32 %retval.0
 
-; CHECK-LABEL: @unreachable(
+; CHECK-LABEL: @unreachable_case(
 ; CHECK: switch.lookup:
-; CHECK: getelementptr inbounds [5 x i32]* @switch.table6, i32 0, i32 %switch.tableidx
+; CHECK: getelementptr inbounds [9 x i32], [9 x i32]* @switch.table.6, i32 0, i32 %switch.tableidx
+}
+
+define i32 @unreachable_default(i32 %x)  {
+entry:
+  switch i32 %x, label %default [
+    i32 0, label %bb0
+    i32 1, label %bb1
+    i32 2, label %bb2
+    i32 3, label %bb3
+  ]
+
+bb0: br label %return
+bb1: br label %return
+bb2: br label %return
+bb3: br label %return
+default: unreachable
+
+return:
+  %retval = phi i32 [ 42, %bb0 ], [ 52, %bb1 ], [ 1, %bb2 ], [ 2, %bb3 ]
+  ret i32 %retval
+
+; CHECK-LABEL: @unreachable_default(
+; CHECK: entry:
+; CHECK-NEXT: %switch.tableidx = sub i32 %x, 0
+; CHECK-NOT: icmp
+; CHECK-NOT: br 1i
+; CHECK-NEXT: %switch.gep = getelementptr inbounds [4 x i32], [4 x i32]* @switch.table.7, i32 0, i32 %switch.tableidx
+; CHECK-NEXT: %switch.load = load i32, i32* %switch.gep
+; CHECK-NEXT: ret i32 %switch.load
 }
 
 ; Don't create a table with illegal type
@@ -967,10 +996,10 @@ sw.bb2:
 sw.default:
   br label %return
 return:
-  %retval.0 = phi i32* [ getelementptr inbounds ([3 x i32]* @dllimport_d, i32 0, i32 0), %sw.default ],
-                       [ getelementptr inbounds ([3 x i32]* @dllimport_c, i32 0, i32 0), %sw.bb2 ],
-                       [ getelementptr inbounds ([3 x i32]* @dllimport_b, i32 0, i32 0), %sw.bb1 ],
-                       [ getelementptr inbounds ([3 x i32]* @dllimport_a, i32 0, i32 0), %entry ]
+  %retval.0 = phi i32* [ getelementptr inbounds ([3 x i32], [3 x i32]* @dllimport_d, i32 0, i32 0), %sw.default ],
+                       [ getelementptr inbounds ([3 x i32], [3 x i32]* @dllimport_c, i32 0, i32 0), %sw.bb2 ],
+                       [ getelementptr inbounds ([3 x i32], [3 x i32]* @dllimport_b, i32 0, i32 0), %sw.bb1 ],
+                       [ getelementptr inbounds ([3 x i32], [3 x i32]* @dllimport_a, i32 0, i32 0), %entry ]
   ret i32* %retval.0
 ; CHECK-LABEL: @dllimport(
 ; CHECK: switch i32
@@ -1244,3 +1273,32 @@ end:
 ; CHECK-LABEL: @pr20210
 ; CHECK: switch i8 %x
 }
+
+; Make sure we do not crash due to trying to generate an unguarded
+; lookup (since i3 can only hold values in the range of explicit
+; values) and simultaneously trying to generate a branch to deal with
+; the fact that we have holes in the range.
+define i32 @covered_switch_with_bit_tests(i3) {
+entry:
+  switch i3 %0, label %l6 [
+    i3 -3, label %l5
+    i3 -4, label %l5
+    i3 3, label %l1
+    i3 2, label %l1
+  ]
+
+l1: br label %l2
+
+l2:
+  %x = phi i32 [ 1, %l1 ], [ 2, %l5 ]
+  br label %l6
+
+l5: br label %l2
+
+l6:
+  %r = phi i32 [ %x, %l2 ], [ 0, %entry ]
+  ret i32 %r
+; CHECK-LABEL: @covered_switch_with_bit_tests
+; CHECK: entry
+; CHECK-NEXT: switch
+}
diff --git a/test/Transforms/SimplifyCFG/attr-noduplicate.ll b/test/Transforms/SimplifyCFG/attr-noduplicate.ll
index 523aa51bb84e..5f8456ba4054 100644
--- a/test/Transforms/SimplifyCFG/attr-noduplicate.ll
+++ b/test/Transforms/SimplifyCFG/attr-noduplicate.ll
@@ -8,8 +8,8 @@
 ; CHECK-NOT: call void @barrier
 define void @noduplicate(i32 %cond, i32* %out) {
 entry:
-  %out1 = getelementptr i32* %out, i32 1
-  %out2 = getelementptr i32* %out, i32 2
+  %out1 = getelementptr i32, i32* %out, i32 1
+  %out2 = getelementptr i32, i32* %out, i32 2
   %cmp = icmp eq i32 %cond, 0
   br i1 %cmp, label %if.then, label %if.end
 
diff --git a/test/Transforms/SimplifyCFG/basictest.ll b/test/Transforms/SimplifyCFG/basictest.ll
index 5de7cc5e31d5..d228499b2ec5 100644
--- a/test/Transforms/SimplifyCFG/basictest.ll
+++ b/test/Transforms/SimplifyCFG/basictest.ll
@@ -1,6 +1,7 @@
 ; Test CFG simplify removal of branch instructions.
 ;
 ; RUN: opt < %s -simplifycfg -S | FileCheck %s
+; RUN: opt < %s -passes=simplify-cfg -S | FileCheck %s
 
 define void @test1() {
         br label %1
@@ -49,7 +50,7 @@ define i8 @test6f() {
 ; CHECK: alloca i8, align 1
 ; CHECK-NEXT: call i8 @test6g
 ; CHECK-NEXT: icmp eq i8 %tmp, 0
-; CHECK-NEXT: load i8* %r, align 1{{$}}
+; CHECK-NEXT: load i8, i8* %r, align 1{{$}}
 
 bb0:
   %r = alloca i8, align 1
@@ -57,7 +58,7 @@ bb0:
   %tmp1 = icmp eq i8 %tmp, 0
   br i1 %tmp1, label %bb2, label %bb1
 bb1:
-  %tmp3 = load i8* %r, align 1, !range !2, !tbaa !1
+  %tmp3 = load i8, i8* %r, align 1, !range !2, !tbaa !1
   %tmp4 = icmp eq i8 %tmp3, 1
   br i1 %tmp4, label %bb2, label %bb3
 bb2:
diff --git a/test/Transforms/SimplifyCFG/branch-fold-dbg.ll b/test/Transforms/SimplifyCFG/branch-fold-dbg.ll
index f715a0cda93c..fac5b186e89d 100644
--- a/test/Transforms/SimplifyCFG/branch-fold-dbg.ll
+++ b/test/Transforms/SimplifyCFG/branch-fold-dbg.ll
@@ -24,7 +24,7 @@ BB2:                                              ; preds = %BB1
 ;CHECK-NEXT: icmp eq
 
 BB3:                                              ; preds = %BB2
-  %6 = getelementptr inbounds [5 x %0]* @0, i32 0, i32 %0, !dbg !6
+  %6 = getelementptr inbounds [5 x %0], [5 x %0]* @0, i32 0, i32 %0, !dbg !6
   call void @llvm.dbg.value(metadata %0* %6, i64 0, metadata !7, metadata !{}), !dbg !12
   %7 = icmp eq %0* %6, null, !dbg !13
   br i1 %7, label %BB5, label %BB4, !dbg !13
@@ -41,19 +41,19 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00foo\00foo\00\00231\000\001\000\006\00256\000\000", !15, !1, !3, null, void (i32)* @foo, null, null, null} ; [ DW_TAG_subprogram ] [line 231] [def] [scope 0] [foo]
-!1 = !{!"0x29", !15} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang (trunk 129006)\001\00\000\00\000", !15, !4, !4, null, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !15, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", line: 231, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !15, scope: !1, type: !3, function: void (i32)* @foo)
+!1 = !DIFile(filename: "a.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang (trunk 129006)", isOptimized: true, emissionKind: 0, file: !15, enums: !4, retainedTypes: !4)
+!3 = !DISubroutineType(types: !4)
 !4 = !{null}
-!5 = !MDLocation(line: 131, column: 2, scope: !0)
-!6 = !MDLocation(line: 134, column: 2, scope: !0)
-!7 = !{!"0x100\00bar\00232\000", !8, !1, !9} ; [ DW_TAG_auto_variable ]
-!8 = !{!"0xb\00231\001\003", !15, !0} ; [ DW_TAG_lexical_block ]
-!9 = !{!"0xf\00\000\0032\0032\000\000", null, !2, !10} ; [ DW_TAG_pointer_type ]
-!10 = !{!"0x26\00\000\000\000\000\000", null, !2, !11} ; [ DW_TAG_const_type ]
-!11 = !{!"0x24\00unsigned int\000\0032\0032\000\000\007", null, !2} ; [ DW_TAG_base_type ]
-!12 = !MDLocation(line: 232, column: 40, scope: !8)
-!13 = !MDLocation(line: 234, column: 2, scope: !8)
-!14 = !MDLocation(line: 274, column: 1, scope: !8)
-!15 = !{!"a.c", !"/private/tmp"}
+!5 = !DILocation(line: 131, column: 2, scope: !0)
+!6 = !DILocation(line: 134, column: 2, scope: !0)
+!7 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "bar", line: 232, scope: !8, file: !1, type: !9)
+!8 = distinct !DILexicalBlock(line: 231, column: 1, file: !15, scope: !0)
+!9 = !DIDerivedType(tag: DW_TAG_pointer_type, size: 32, align: 32, scope: !2, baseType: !10)
+!10 = !DIDerivedType(tag: DW_TAG_const_type, scope: !2, baseType: !11)
+!11 = !DIBasicType(tag: DW_TAG_base_type, name: "unsigned int", size: 32, align: 32, encoding: DW_ATE_unsigned)
+!12 = !DILocation(line: 232, column: 40, scope: !8)
+!13 = !DILocation(line: 234, column: 2, scope: !8)
+!14 = !DILocation(line: 274, column: 1, scope: !8)
+!15 = !DIFile(filename: "a.c", directory: "/private/tmp")
diff --git a/test/Transforms/SimplifyCFG/branch-fold-threshold.ll b/test/Transforms/SimplifyCFG/branch-fold-threshold.ll
index 878c0a4837ae..fad5fce1d4b0 100644
--- a/test/Transforms/SimplifyCFG/branch-fold-threshold.ll
+++ b/test/Transforms/SimplifyCFG/branch-fold-threshold.ll
@@ -19,7 +19,7 @@ lor.lhs.false:
 ; AGGRESSIVE-NOT: br i1
 
 cond.false:
-  %0 = load i32* %input, align 4
+  %0 = load i32, i32* %input, align 4
   br label %cond.end
 
 cond.end:
diff --git a/test/Transforms/SimplifyCFG/branch-phi-thread.ll b/test/Transforms/SimplifyCFG/branch-phi-thread.ll
index c19ba69866b4..4c1b7e68e25f 100644
--- a/test/Transforms/SimplifyCFG/branch-phi-thread.ll
+++ b/test/Transforms/SimplifyCFG/branch-phi-thread.ll
@@ -51,7 +51,7 @@ E:
 	br i1 %C, label %T, label %F
 T:		; preds = %A, %E
 	call void @f3( )
-	%XX = load i32* %AP		; <i32> [#uses=1]
+	%XX = load i32, i32* %AP		; <i32> [#uses=1]
 	store i32 %XX, i32* %BP
 	br i1 %C, label %B, label %A
 A:		; preds = %T
diff --git a/test/Transforms/SimplifyCFG/clamp.ll b/test/Transforms/SimplifyCFG/clamp.ll
new file mode 100644
index 000000000000..d21894a12699
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/clamp.ll
@@ -0,0 +1,22 @@
+; RUN: opt < %s -simplifycfg -S | FileCheck %s
+
+define float @clamp(float %a, float %b, float %c) {
+; CHECK-LABEL: @clamp
+; CHECK:  %cmp = fcmp ogt float %a, %c
+; CHECK:  %cmp1 = fcmp olt float %a, %b
+; CHECK:  %cond = select i1 %cmp1, float %b, float %a
+; CHECK:  %cond5 = select i1 %cmp, float %c, float %cond
+; CHECK:  ret float %cond5
+entry:
+  %cmp = fcmp ogt float %a, %c
+  br i1 %cmp, label %cond.end4, label %cond.false
+
+cond.false:                                       ; preds = %entry
+  %cmp1 = fcmp olt float %a, %b
+  %cond = select i1 %cmp1, float %b, float %a
+  br label %cond.end4
+
+cond.end4:                                        ; preds = %entry, %cond.false
+  %cond5 = phi float [ %cond, %cond.false ], [ %c, %entry ]
+  ret float %cond5
+}
diff --git a/test/Transforms/SimplifyCFG/common-dest-folding.ll b/test/Transforms/SimplifyCFG/common-dest-folding.ll
index 0aa3b2c560a5..e3e27c706d1d 100644
--- a/test/Transforms/SimplifyCFG/common-dest-folding.ll
+++ b/test/Transforms/SimplifyCFG/common-dest-folding.ll
@@ -18,7 +18,7 @@ define i32 @foo(i32 %k, i32 %c1, i32 %c2) {
   br i1 %5, label %8, label %6
 
 ; <label>:6                                       ; preds = %3
-  %7 = tail call i32 (...)* @bar() nounwind
+  %7 = tail call i32 (...) @bar() nounwind
   br label %8
 
 ; <label>:8                                       ; preds = %3, %0, %6
@@ -47,7 +47,7 @@ bb3:                                              ; preds = %bb
   br i1 %tmp6, label %bb9, label %bb7
 
 bb7:                                              ; preds = %bb3
-  %tmp8 = tail call i32 (...)* @bar() #1
+  %tmp8 = tail call i32 (...) @bar() #1
   br label %bb9
 
 bb9:                                              ; preds = %bb7, %bb3, %bb
diff --git a/test/Transforms/SimplifyCFG/dbginfo.ll b/test/Transforms/SimplifyCFG/dbginfo.ll
index 1a9f20ac871e..12aec91c6c55 100644
--- a/test/Transforms/SimplifyCFG/dbginfo.ll
+++ b/test/Transforms/SimplifyCFG/dbginfo.ll
@@ -58,7 +58,7 @@ entry:
 	%"alloca point" = bitcast i32 0 to i32		; <i32> [#uses=0]
 	call void @llvm.dbg.func.start({ }* bitcast (%llvm.dbg.subprogram.type* @llvm.dbg.subprogram947 to { }*))
 	store %struct.__false_type* %this, %struct.__false_type** %this_addr
-	%0 = load %struct.__false_type** %this_addr, align 4		; <%struct.__false_type*> [#uses=1]
+	%0 = load %struct.__false_type*, %struct.__false_type** %this_addr, align 4		; <%struct.__false_type*> [#uses=1]
 	call void @_ZN9__gnu_cxx13new_allocatorIP5SceneED2Ev(%struct.__false_type* %0) nounwind
 	br label %bb
 
diff --git a/test/Transforms/SimplifyCFG/duplicate-landingpad.ll b/test/Transforms/SimplifyCFG/duplicate-landingpad.ll
new file mode 100644
index 000000000000..54028774d20e
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/duplicate-landingpad.ll
@@ -0,0 +1,110 @@
+; RUN: opt < %s -simplifycfg -S | FileCheck %s
+target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f32:32:32-f64:64:64-v64:64:64-v128:128:128-a0:0:64-s0:64:64-f80:128:128-n8:16:32:64-S128"
+
+declare i32 @__gxx_personality_v0(...)
+declare void @fn()
+
+
+; CHECK-LABEL: @test1
+define void @test1() {
+entry:
+; CHECK-LABEL: entry:
+; CHECK: to label %invoke2 unwind label %lpad2
+  invoke void @fn()
+    to label %invoke2 unwind label %lpad1
+
+invoke2:
+; CHECK-LABEL: invoke2:
+; CHECK: to label %invoke.cont unwind label %lpad2
+  invoke void @fn()
+    to label %invoke.cont unwind label %lpad2
+
+invoke.cont:
+  ret void
+
+lpad1:
+  %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+         cleanup
+  br label %shared_resume
+
+lpad2:
+; CHECK-LABEL: lpad2:
+; CHECK: landingpad { i8*, i32 } personality i32 (...)* @__gxx_personality_v0
+; CHECK-NEXT: cleanup
+; CHECK-NEXT: call void @fn()
+; CHECK-NEXT: ret void
+  %exn2 = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+          cleanup
+  br label %shared_resume
+
+shared_resume:
+  call void @fn()
+  ret void
+}
+
+; Don't trigger if blocks aren't the same/empty
+define void @neg1() {
+; CHECK-LABEL: @neg1
+entry:
+; CHECK-LABEL: entry:
+; CHECK: to label %invoke2 unwind label %lpad1
+  invoke void @fn()
+    to label %invoke2 unwind label %lpad1
+
+invoke2:
+; CHECK-LABEL: invoke2:
+; CHECK: to label %invoke.cont unwind label %lpad2
+  invoke void @fn()
+    to label %invoke.cont unwind label %lpad2
+
+invoke.cont:
+  ret void
+
+lpad1:
+  %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+         filter [0 x i8*] zeroinitializer
+  call void @fn()
+  br label %shared_resume
+
+lpad2:
+  %exn2 = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+          cleanup
+  br label %shared_resume
+
+shared_resume:
+  call void @fn()
+  ret void
+}
+
+; Should not trigger when the landing pads are not the exact same
+define void @neg2() {
+; CHECK-LABEL: @neg2
+entry:
+; CHECK-LABEL: entry:
+; CHECK: to label %invoke2 unwind label %lpad1
+  invoke void @fn()
+    to label %invoke2 unwind label %lpad1
+
+invoke2:
+; CHECK-LABEL: invoke2:
+; CHECK: to label %invoke.cont unwind label %lpad2
+  invoke void @fn()
+    to label %invoke.cont unwind label %lpad2
+
+invoke.cont:
+  ret void
+
+lpad1:
+  %exn = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+         filter [0 x i8*] zeroinitializer
+  br label %shared_resume
+
+lpad2:
+  %exn2 = landingpad {i8*, i32} personality i32 (...)* @__gxx_personality_v0
+          cleanup
+  br label %shared_resume
+
+shared_resume:
+  call void @fn()
+  ret void
+}
diff --git a/test/Transforms/SimplifyCFG/hoist-common-code.ll b/test/Transforms/SimplifyCFG/hoist-common-code.ll
index 5c83e2a3aa49..c1ca605cc206 100644
--- a/test/Transforms/SimplifyCFG/hoist-common-code.ll
+++ b/test/Transforms/SimplifyCFG/hoist-common-code.ll
@@ -6,12 +6,12 @@ define void @test(i1 %P, i32* %Q) {
         br i1 %P, label %T, label %F
 T:              ; preds = %0
         store i32 1, i32* %Q
-        %A = load i32* %Q               ; <i32> [#uses=1]
+        %A = load i32, i32* %Q               ; <i32> [#uses=1]
         call void @bar( i32 %A )
         ret void
 F:              ; preds = %0
         store i32 1, i32* %Q
-        %B = load i32* %Q               ; <i32> [#uses=1]
+        %B = load i32, i32* %Q               ; <i32> [#uses=1]
         call void @bar( i32 %B )
         ret void
 }
diff --git a/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll b/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
index 869ce099d075..345cf6282e3c 100644
--- a/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
+++ b/test/Transforms/SimplifyCFG/hoist-dbgvalue.ll
@@ -1,21 +1,21 @@
 ; RUN: opt -simplifycfg -S < %s | FileCheck %s
 
 define i32 @foo(i32 %i) nounwind ssp {
-  call void @llvm.dbg.value(metadata i32 %i, i64 0, metadata !6, metadata !{}), !dbg !7
-  call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !9, metadata !{}), !dbg !11
+  call void @llvm.dbg.value(metadata i32 %i, i64 0, metadata !6, metadata !DIExpression()), !dbg !7
+  call void @llvm.dbg.value(metadata i32 0, i64 0, metadata !9, metadata !DIExpression()), !dbg !11
   %1 = icmp ne i32 %i, 0, !dbg !12
-;CHECK: call i32 (...)* @bar()
+;CHECK: call i32 (...) @bar()
 ;CHECK-NEXT: llvm.dbg.value
   br i1 %1, label %2, label %4, !dbg !12
 
 ; <label>:2                                       ; preds = %0
-  %3 = call i32 (...)* @bar(), !dbg !13
-  call void @llvm.dbg.value(metadata i32 %3, i64 0, metadata !9, metadata !{}), !dbg !13
+  %3 = call i32 (...) @bar(), !dbg !13
+  call void @llvm.dbg.value(metadata i32 %3, i64 0, metadata !9, metadata !DIExpression()), !dbg !13
   br label %6, !dbg !15
 
 ; <label>:4                                       ; preds = %0
-  %5 = call i32 (...)* @bar(), !dbg !16
-  call void @llvm.dbg.value(metadata i32 %5, i64 0, metadata !9, metadata !{}), !dbg !16
+  %5 = call i32 (...) @bar(), !dbg !16
+  call void @llvm.dbg.value(metadata i32 %5, i64 0, metadata !9, metadata !DIExpression()), !dbg !16
   br label %6, !dbg !18
 
 ; <label>:6                                       ; preds = %4, %2
@@ -32,25 +32,25 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 !llvm.module.flags = !{!21}
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00foo\00foo\00\002\000\001\000\006\00256\000\000", !20, !1, !3, null, i32 (i32)* @foo, null, null, null} ; [ DW_TAG_subprogram ] [line 2] [def] [scope 0] [foo]
-!1 = !{!"0x29", !20} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang\001\00\000\00\000", !20, !8, !8, null, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !20, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", line: 2, isLocal: false, isDefinition: true, virtualIndex: 6, flags: DIFlagPrototyped, isOptimized: false, file: !20, scope: !1, type: !3, function: i32 (i32)* @foo)
+!1 = !DIFile(filename: "b.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang", isOptimized: true, emissionKind: 0, file: !20, enums: !8, retainedTypes: !8)
+!3 = !DISubroutineType(types: !4)
 !4 = !{!5}
-!5 = !{!"0x24\00int\000\0032\0032\000\000\005", null, !2} ; [ DW_TAG_base_type ]
-!6 = !{!"0x101\00i\0016777218\000", !0, !1, !5} ; [ DW_TAG_arg_variable ]
-!7 = !MDLocation(line: 2, column: 13, scope: !0)
+!5 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!6 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i", line: 2, arg: 1, scope: !0, file: !1, type: !5)
+!7 = !DILocation(line: 2, column: 13, scope: !0)
 !8 = !{i32 0}
-!9 = !{!"0x100\00k\003\000", !10, !1, !5} ; [ DW_TAG_auto_variable ]
-!10 = !{!"0xb\002\0016\000", !20, !0} ; [ DW_TAG_lexical_block ]
-!11 = !MDLocation(line: 3, column: 12, scope: !10)
-!12 = !MDLocation(line: 4, column: 3, scope: !10)
-!13 = !MDLocation(line: 5, column: 5, scope: !14)
-!14 = !{!"0xb\004\0010\001", !20, !10} ; [ DW_TAG_lexical_block ]
-!15 = !MDLocation(line: 6, column: 3, scope: !14)
-!16 = !MDLocation(line: 7, column: 5, scope: !17)
-!17 = !{!"0xb\006\0010\002", !20, !10} ; [ DW_TAG_lexical_block ]
-!18 = !MDLocation(line: 8, column: 3, scope: !17)
-!19 = !MDLocation(line: 9, column: 3, scope: !10)
-!20 = !{!"b.c", !"/private/tmp"}
-!21 = !{i32 1, !"Debug Info Version", i32 2}
+!9 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "k", line: 3, scope: !10, file: !1, type: !5)
+!10 = distinct !DILexicalBlock(line: 2, column: 16, file: !20, scope: !0)
+!11 = !DILocation(line: 3, column: 12, scope: !10)
+!12 = !DILocation(line: 4, column: 3, scope: !10)
+!13 = !DILocation(line: 5, column: 5, scope: !14)
+!14 = distinct !DILexicalBlock(line: 4, column: 10, file: !20, scope: !10)
+!15 = !DILocation(line: 6, column: 3, scope: !14)
+!16 = !DILocation(line: 7, column: 5, scope: !17)
+!17 = distinct !DILexicalBlock(line: 6, column: 10, file: !20, scope: !10)
+!18 = !DILocation(line: 8, column: 3, scope: !17)
+!19 = !DILocation(line: 9, column: 3, scope: !10)
+!20 = !DIFile(filename: "b.c", directory: "/private/tmp")
+!21 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/SimplifyCFG/hoist-with-range.ll b/test/Transforms/SimplifyCFG/hoist-with-range.ll
index 7ca3ff247dc8..0a2b28267ef1 100644
--- a/test/Transforms/SimplifyCFG/hoist-with-range.ll
+++ b/test/Transforms/SimplifyCFG/hoist-with-range.ll
@@ -2,15 +2,15 @@
 
 define void @foo(i1 %c, i8* %p) {
 ; CHECK: if:
-; CHECK-NEXT: load i8* %p, !range !0
+; CHECK-NEXT: load i8, i8* %p, !range !0
 ; CHECK: !0 = !{i8 0, i8 1, i8 3, i8 5}
 if:
   br i1 %c, label %then, label %else
 then:
-  %t = load i8* %p, !range !0
+  %t = load i8, i8* %p, !range !0
   br label %out
 else:
-  %e = load i8* %p, !range !1
+  %e = load i8, i8* %p, !range !1
   br label %out
 out:
   ret void
diff --git a/test/Transforms/SimplifyCFG/indirectbr.ll b/test/Transforms/SimplifyCFG/indirectbr.ll
index d0020d001b03..67e23d2fe935 100644
--- a/test/Transforms/SimplifyCFG/indirectbr.ll
+++ b/test/Transforms/SimplifyCFG/indirectbr.ll
@@ -17,7 +17,7 @@ entry:
   store i8* blockaddress(@indbrtest0, %BB1), i8** %P
   store i8* blockaddress(@indbrtest0, %BB2), i8** %P
   call void @foo()
-  %t = load i8** %Q
+  %t = load i8*, i8** %Q
   indirectbr i8* %t, [label %BB0, label %BB1, label %BB2, label %BB0, label %BB1, label %BB2]
 BB0:
   call void @A()
@@ -42,7 +42,7 @@ define void @indbrtest1(i8** %P, i8** %Q) {
 entry:
   store i8* blockaddress(@indbrtest1, %BB0), i8** %P
   call void @foo()
-  %t = load i8** %Q
+  %t = load i8*, i8** %Q
   indirectbr i8* %t, [label %BB0, label %BB0]
 BB0:
   call void @A()
@@ -192,8 +192,8 @@ escape-string.top:
 
 xlab8x:                                           ; preds = %xlab5x
   %xvaluex = call i32 @xselectorx()
-  %xblkx.x = getelementptr [9 x i8*]* @xblkx.bbs, i32 0, i32 %xvaluex
-  %xblkx.load = load i8** %xblkx.x
+  %xblkx.x = getelementptr [9 x i8*], [9 x i8*]* @xblkx.bbs, i32 0, i32 %xvaluex
+  %xblkx.load = load i8*, i8** %xblkx.x
   indirectbr i8* %xblkx.load, [label %xblkx.begin, label %xblkx.begin3, label %xblkx.begin4, label %xblkx.begin5, label %xblkx.begin6, label %xblkx.begin7, label %xblkx.begin8, label %xblkx.begin9, label %xblkx.end]
 
 xblkx.begin:
diff --git a/test/Transforms/SimplifyCFG/iterative-simplify.ll b/test/Transforms/SimplifyCFG/iterative-simplify.ll
index a39741105764..60728b9a9578 100644
--- a/test/Transforms/SimplifyCFG/iterative-simplify.ll
+++ b/test/Transforms/SimplifyCFG/iterative-simplify.ll
@@ -17,13 +17,13 @@ cond_true:		; preds = %entry
 	br label %bb
 
 bb:		; preds = %cond_next, %cond_true
-	%tmp = load i32* %z		; <i32> [#uses=1]
+	%tmp = load i32, i32* %z		; <i32> [#uses=1]
 	%tmp1 = sub i32 %tmp, 16384		; <i32> [#uses=1]
 	store i32 %tmp1, i32* %z
-	%tmp2 = load i32* %i		; <i32> [#uses=1]
+	%tmp2 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp3 = add i32 %tmp2, 1		; <i32> [#uses=1]
 	store i32 %tmp3, i32* %i
-	%tmp4 = load i32* %i		; <i32> [#uses=1]
+	%tmp4 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp5 = icmp sgt i32 %tmp4, 262144		; <i1> [#uses=1]
 	%tmp56 = zext i1 %tmp5 to i8		; <i8> [#uses=1]
 	%toBool7 = icmp ne i8 %tmp56, 0		; <i1> [#uses=1]
@@ -34,7 +34,7 @@ cond_true8:		; preds = %bb
 	unreachable
 
 cond_next:		; preds = %bb
-	%tmp9 = load i32* %z		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* %z		; <i32> [#uses=1]
 	%tmp10 = icmp ne i32 %tmp9, 0		; <i1> [#uses=1]
 	%tmp1011 = zext i1 %tmp10 to i8		; <i8> [#uses=1]
 	%toBool12 = icmp ne i8 %tmp1011, 0		; <i1> [#uses=1]
@@ -53,13 +53,13 @@ cond_true15:		; preds = %cond_false
 	br label %bb17
 
 bb17:		; preds = %cond_next27, %cond_true15
-	%tmp18 = load i32* %z16		; <i32> [#uses=1]
+	%tmp18 = load i32, i32* %z16		; <i32> [#uses=1]
 	%tmp19 = sub i32 %tmp18, 16384		; <i32> [#uses=1]
 	store i32 %tmp19, i32* %z16
-	%tmp20 = load i32* %i		; <i32> [#uses=1]
+	%tmp20 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp21 = add i32 %tmp20, 1		; <i32> [#uses=1]
 	store i32 %tmp21, i32* %i
-	%tmp22 = load i32* %i		; <i32> [#uses=1]
+	%tmp22 = load i32, i32* %i		; <i32> [#uses=1]
 	%tmp23 = icmp sgt i32 %tmp22, 262144		; <i1> [#uses=1]
 	%tmp2324 = zext i1 %tmp23 to i8		; <i8> [#uses=1]
 	%toBool25 = icmp ne i8 %tmp2324, 0		; <i1> [#uses=1]
@@ -70,7 +70,7 @@ cond_true26:		; preds = %bb17
 	unreachable
 
 cond_next27:		; preds = %bb17
-	%tmp28 = load i32* %z16		; <i32> [#uses=1]
+	%tmp28 = load i32, i32* %z16		; <i32> [#uses=1]
 	%tmp29 = icmp ne i32 %tmp28, 0		; <i1> [#uses=1]
 	%tmp2930 = zext i1 %tmp29 to i8		; <i8> [#uses=1]
 	%toBool31 = icmp ne i8 %tmp2930, 0		; <i1> [#uses=1]
@@ -91,7 +91,7 @@ cond_next35:		; preds = %cond_next34
 	br label %return
 
 return:		; preds = %cond_next35
-	%retval36 = load i32* %retval		; <i32> [#uses=1]
+	%retval36 = load i32, i32* %retval		; <i32> [#uses=1]
 	ret i32 %retval36
 }
 
diff --git a/test/Transforms/SimplifyCFG/multiple-phis.ll b/test/Transforms/SimplifyCFG/multiple-phis.ll
index 78454232461a..a6eef09ae646 100644
--- a/test/Transforms/SimplifyCFG/multiple-phis.ll
+++ b/test/Transforms/SimplifyCFG/multiple-phis.ll
@@ -22,8 +22,8 @@ while.body:                                       ; preds = %while.cond
   %add = add i32 %low.0, %high.addr.0
   %div = udiv i32 %add, 2
   %idxprom = zext i32 %div to i64
-  %arrayidx = getelementptr inbounds i32* %r, i64 %idxprom
-  %0 = load i32* %arrayidx
+  %arrayidx = getelementptr inbounds i32, i32* %r, i64 %idxprom
+  %0 = load i32, i32* %arrayidx
   %cmp1 = icmp ult i32 %k, %0
   br i1 %cmp1, label %if.then, label %if.else
 
diff --git a/test/Transforms/SimplifyCFG/no_speculative_loads_with_tsan.ll b/test/Transforms/SimplifyCFG/no_speculative_loads_with_tsan.ll
index b388cc5830f4..4792e9565b2d 100644
--- a/test/Transforms/SimplifyCFG/no_speculative_loads_with_tsan.ll
+++ b/test/Transforms/SimplifyCFG/no_speculative_loads_with_tsan.ll
@@ -8,7 +8,7 @@ entry:
   br i1 %tobool, label %return, label %if.then
 
 if.then:                                          ; preds = %entry
-  %0 = load i32* @g, align 4
+  %0 = load i32, i32* @g, align 4
   br label %return
 
 return:                                           ; preds = %entry, %if.then
@@ -26,7 +26,7 @@ entry:
   br i1 %tobool, label %return, label %if.then
 
 if.then:                                          ; preds = %entry
-  %0 = load i32* @g, align 4
+  %0 = load i32, i32* @g, align 4
   br label %return
 
 return:                                           ; preds = %entry, %if.then
@@ -34,7 +34,7 @@ return:                                           ; preds = %entry, %if.then
   ret i32 %retval
 ; CHECK-LABEL: @TestTsan
 ; CHECK: br i1
-; CHECK: load i32* @g
+; CHECK: load i32, i32* @g
 ; CHECK: br label
 ; CHECK: ret i32
 }
diff --git a/test/Transforms/SimplifyCFG/phi-undef-loadstore.ll b/test/Transforms/SimplifyCFG/phi-undef-loadstore.ll
index f34aec525df8..c0f0046a1d2c 100644
--- a/test/Transforms/SimplifyCFG/phi-undef-loadstore.ll
+++ b/test/Transforms/SimplifyCFG/phi-undef-loadstore.ll
@@ -21,7 +21,7 @@ if.then4:                                         ; preds = %if.else
 
 if.end7:                                          ; preds = %if.else, %if.then4, %if.then
   %x.0 = phi i32* [ %a, %if.then ], [ %c, %if.then4 ], [ null, %if.else ]
-  %tmp9 = load i32* %x.0
+  %tmp9 = load i32, i32* %x.0
   ret i32 %tmp9
 
 ; CHECK-LABEL: @test1(
@@ -50,7 +50,7 @@ if.then4:                                         ; preds = %if.else
 
 if.end7:                                          ; preds = %if.else, %if.then4, %if.then
   %x.0 = phi i32* [ %a, %if.then ], [ null, %if.then4 ], [ null, %if.else ]
-  %tmp9 = load i32* %x.0
+  %tmp9 = load i32, i32* %x.0
   ret i32 %tmp9
 ; CHECK-LABEL: @test2(
 ; CHECK: if.else:
@@ -79,7 +79,7 @@ if.then4:                                         ; preds = %if.else
 if.end7:                                          ; preds = %if.else, %if.then4, %if.then
   %x.0 = phi i32* [ %a, %if.then ], [ null, %if.then4 ], [ null, %if.else ]
   tail call void @bar() nounwind
-  %tmp9 = load i32* %x.0
+  %tmp9 = load i32, i32* %x.0
   ret i32 %tmp9
 ; CHECK-LABEL: @test3(
 ; CHECK: if.end7:
@@ -105,8 +105,8 @@ if.then4:                                         ; preds = %if.else
 
 if.end7:                                          ; preds = %if.else, %if.then4, %if.then
   %x.0 = phi i32* [ %a, %if.then ], [ null, %if.then4 ], [ null, %if.else ]
-  %gep = getelementptr i32* %x.0, i32 10
-  %tmp9 = load i32* %gep
+  %gep = getelementptr i32, i32* %x.0, i32 10
+  %tmp9 = load i32, i32* %gep
   %tmp10 = or i32 %tmp9, 1
   store i32 %tmp10, i32* %gep
   ret i32 %tmp9
diff --git a/test/Transforms/SimplifyCFG/preserve-branchweights.ll b/test/Transforms/SimplifyCFG/preserve-branchweights.ll
index 7802a0593574..ae1794b1c61a 100644
--- a/test/Transforms/SimplifyCFG/preserve-branchweights.ll
+++ b/test/Transforms/SimplifyCFG/preserve-branchweights.ll
@@ -353,7 +353,7 @@ for.cond2:
   %tobool = icmp eq i32 %bit.0, 0
   br i1 %tobool, label %for.exit, label %for.body3, !prof !10
 for.body3:
-  %v3 = load i32* @max_regno, align 4
+  %v3 = load i32, i32* @max_regno, align 4
   %cmp4 = icmp eq i32 %i.1, %v3
   br i1 %cmp4, label %for.exit, label %for.inc, !prof !11
 for.inc:
diff --git a/test/Transforms/SimplifyCFG/seh-nounwind.ll b/test/Transforms/SimplifyCFG/seh-nounwind.ll
new file mode 100644
index 000000000000..3845e3198909
--- /dev/null
+++ b/test/Transforms/SimplifyCFG/seh-nounwind.ll
@@ -0,0 +1,31 @@
+; RUN: opt -S -simplifycfg < %s | FileCheck %s
+
+; Don't remove invokes of nounwind functions if the personality handles async
+; exceptions. The @div function in this test can fault, even though it can't
+; throw a synchronous exception.
+
+define i32 @div(i32 %n, i32 %d) nounwind {
+entry:
+  %div = sdiv i32 %n, %d
+  ret i32 %div
+}
+
+define i32 @main() nounwind {
+entry:
+  %call = invoke i32 @div(i32 10, i32 0)
+          to label %__try.cont unwind label %lpad
+
+lpad:
+  %0 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__C_specific_handler to i8*)
+          catch i8* null
+  br label %__try.cont
+
+__try.cont:
+  %retval.0 = phi i32 [ %call, %entry ], [ 0, %lpad ]
+  ret i32 %retval.0
+}
+
+; CHECK-LABEL: define i32 @main()
+; CHECK: invoke i32 @div(i32 10, i32 0)
+
+declare i32 @__C_specific_handler(...)
diff --git a/test/Transforms/SimplifyCFG/select-gep.ll b/test/Transforms/SimplifyCFG/select-gep.ll
index 96c214cbc81e..f2bc5abbf738 100644
--- a/test/Transforms/SimplifyCFG/select-gep.ll
+++ b/test/Transforms/SimplifyCFG/select-gep.ll
@@ -1,40 +1,21 @@
 ; RUN: opt -S -simplifycfg < %s | FileCheck %s
 
-define i8* @test1(i8* %x, i64 %y) nounwind {
-entry:
-  %tmp1 = load i8* %x, align 1
-  %cmp = icmp eq i8 %tmp1, 47
-  br i1 %cmp, label %if.then, label %if.end
-
-if.then:
-  %incdec.ptr = getelementptr inbounds i8* %x, i64 %y
-  br label %if.end
-
-if.end:
-  %x.addr = phi i8* [ %incdec.ptr, %if.then ], [ %x, %entry ]
-  ret i8* %x.addr
-
-; CHECK-LABEL: @test1(
-; CHECK-NOT: select
-; CHECK: ret i8* %x.addr
-}
-
 %ST = type { i8, i8 }
 
-define i8* @test2(%ST* %x, i8* %y) nounwind {
+define i8* @test1(%ST* %x, i8* %y) nounwind {
 entry:
   %cmp = icmp eq %ST* %x, null
   br i1 %cmp, label %if.then, label %if.end
 
 if.then:
-  %incdec.ptr = getelementptr %ST* %x, i32 0, i32 1
+  %incdec.ptr = getelementptr %ST, %ST* %x, i32 0, i32 1
   br label %if.end
 
 if.end:
   %x.addr = phi i8* [ %incdec.ptr, %if.then ], [ %y, %entry ]
   ret i8* %x.addr
 
-; CHECK-LABEL: @test2(
+; CHECK-LABEL: @test1(
 ; CHECK: %incdec.ptr.y = select i1 %cmp, i8* %incdec.ptr, i8* %y
 ; CHECK: ret i8* %incdec.ptr.y
 }
diff --git a/test/Transforms/SimplifyCFG/speculate-store.ll b/test/Transforms/SimplifyCFG/speculate-store.ll
index e241901a8789..c1ac7bcea249 100644
--- a/test/Transforms/SimplifyCFG/speculate-store.ll
+++ b/test/Transforms/SimplifyCFG/speculate-store.ll
@@ -2,15 +2,15 @@
 
 define void @ifconvertstore(i32 %m, i32* %A, i32* %B, i32 %C, i32 %D) {
 entry:
-  %arrayidx = getelementptr inbounds i32* %B, i64 0
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 0
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %C
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 0
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 0
 
 ; First store to the location.
   store i32 %add, i32* %arrayidx2, align 4
-  %arrayidx4 = getelementptr inbounds i32* %B, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %1, %D
   %cmp6 = icmp sgt i32 %add5, %C
   br i1 %cmp6, label %if.then, label %ret.end
@@ -30,15 +30,15 @@ ret.end:
 
 define void @noifconvertstore1(i32 %m, i32* %A, i32* %B, i32 %C, i32 %D) {
 entry:
-  %arrayidx = getelementptr inbounds i32* %B, i64 0
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 0
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %C
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 0
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 0
 
 ; Store to a different location.
   store i32 %add, i32* %arrayidx, align 4
-  %arrayidx4 = getelementptr inbounds i32* %B, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %1, %D
   %cmp6 = icmp sgt i32 %add5, %C
   br i1 %cmp6, label %if.then, label %ret.end
@@ -57,16 +57,16 @@ declare void @unknown_fun()
 
 define void @noifconvertstore2(i32 %m, i32* %A, i32* %B, i32 %C, i32 %D) {
 entry:
-  %arrayidx = getelementptr inbounds i32* %B, i64 0
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 0
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %C
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 0
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 0
 
 ; First store to the location.
   store i32 %add, i32* %arrayidx2, align 4
   call void @unknown_fun()
-  %arrayidx4 = getelementptr inbounds i32* %B, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %1, %D
   %cmp6 = icmp sgt i32 %add5, %C
   br i1 %cmp6, label %if.then, label %ret.end
@@ -83,15 +83,15 @@ ret.end:
 
 define void @noifconvertstore_volatile(i32 %m, i32* %A, i32* %B, i32 %C, i32 %D) {
 entry:
-  %arrayidx = getelementptr inbounds i32* %B, i64 0
-  %0 = load i32* %arrayidx, align 4
+  %arrayidx = getelementptr inbounds i32, i32* %B, i64 0
+  %0 = load i32, i32* %arrayidx, align 4
   %add = add nsw i32 %0, %C
-  %arrayidx2 = getelementptr inbounds i32* %A, i64 0
+  %arrayidx2 = getelementptr inbounds i32, i32* %A, i64 0
 
 ; First store to the location.
   store i32 %add, i32* %arrayidx2, align 4
-  %arrayidx4 = getelementptr inbounds i32* %B, i64 1
-  %1 = load i32* %arrayidx4, align 4
+  %arrayidx4 = getelementptr inbounds i32, i32* %B, i64 1
+  %1 = load i32, i32* %arrayidx4, align 4
   %add5 = add nsw i32 %1, %D
   %cmp6 = icmp sgt i32 %add5, %C
   br i1 %cmp6, label %if.then, label %ret.end
diff --git a/test/Transforms/SimplifyCFG/speculate-with-offset.ll b/test/Transforms/SimplifyCFG/speculate-with-offset.ll
index 64fed85c795c..65ebb5c2c3f6 100644
--- a/test/Transforms/SimplifyCFG/speculate-with-offset.ll
+++ b/test/Transforms/SimplifyCFG/speculate-with-offset.ll
@@ -9,7 +9,7 @@
 define void @yes(i1 %c) nounwind {
 entry:
   %a = alloca [4 x i64*], align 8
-  %__a.addr = getelementptr [4 x i64*]* %a, i64 0, i64 3
+  %__a.addr = getelementptr [4 x i64*], [4 x i64*]* %a, i64 0, i64 3
   call void @frob(i64** %__a.addr)
   br i1 %c, label %if.then, label %if.end
 
@@ -17,7 +17,7 @@ if.then:                                          ; preds = %entry
   br label %return
 
 if.end:                                           ; preds = %entry
-  %tmp5 = load i64** %__a.addr, align 8
+  %tmp5 = load i64*, i64** %__a.addr, align 8
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
@@ -31,7 +31,7 @@ return:                                           ; preds = %if.end, %if.then
 define void @no0(i1 %c) nounwind {
 entry:
   %a = alloca [4 x i64*], align 8
-  %__a.addr = getelementptr [4 x i64*]* %a, i64 0, i64 4
+  %__a.addr = getelementptr [4 x i64*], [4 x i64*]* %a, i64 0, i64 4
   call void @frob(i64** %__a.addr)
   br i1 %c, label %if.then, label %if.end
 
@@ -39,7 +39,7 @@ if.then:                                          ; preds = %entry
   br label %return
 
 if.end:                                           ; preds = %entry
-  %tmp5 = load i64** %__a.addr, align 8
+  %tmp5 = load i64*, i64** %__a.addr, align 8
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
@@ -53,7 +53,7 @@ return:                                           ; preds = %if.end, %if.then
 define void @no1(i1 %c, i64 %n) nounwind {
 entry:
   %a = alloca [4 x i64*], align 8
-  %__a.addr = getelementptr [4 x i64*]* %a, i64 0, i64 %n
+  %__a.addr = getelementptr [4 x i64*], [4 x i64*]* %a, i64 0, i64 %n
   call void @frob(i64** %__a.addr)
   br i1 %c, label %if.then, label %if.end
 
@@ -61,7 +61,7 @@ if.then:                                          ; preds = %entry
   br label %return
 
 if.end:                                           ; preds = %entry
-  %tmp5 = load i64** %__a.addr, align 8
+  %tmp5 = load i64*, i64** %__a.addr, align 8
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
@@ -75,7 +75,7 @@ return:                                           ; preds = %if.end, %if.then
 define void @no2(i1 %c, i64 %n) nounwind {
 entry:
   %a = alloca [4 x i64*], align 8
-  %__a.addr = getelementptr [4 x i64*]* %a, i64 1, i64 0
+  %__a.addr = getelementptr [4 x i64*], [4 x i64*]* %a, i64 1, i64 0
   call void @frob(i64** %__a.addr)
   br i1 %c, label %if.then, label %if.end
 
@@ -83,7 +83,7 @@ if.then:                                          ; preds = %entry
   br label %return
 
 if.end:                                           ; preds = %entry
-  %tmp5 = load i64** %__a.addr, align 8
+  %tmp5 = load i64*, i64** %__a.addr, align 8
   br label %return
 
 return:                                           ; preds = %if.end, %if.then
diff --git a/test/Transforms/SimplifyCFG/switch-range-to-icmp.ll b/test/Transforms/SimplifyCFG/switch-range-to-icmp.ll
index 8bf2b04520b9..a109b317c732 100644
--- a/test/Transforms/SimplifyCFG/switch-range-to-icmp.ll
+++ b/test/Transforms/SimplifyCFG/switch-range-to-icmp.ll
@@ -48,3 +48,30 @@ b:
   %1 = call i32 @f(i32 1)
   ret i32 %1
 }
+
+
+define i32 @unreachable2(i32 %x) {
+; CHECK-LABEL: @unreachable2
+; CHECK: x.off = add i32 %x, -5
+; CHECK: %switch = icmp ult i32 %x.off, 3
+; CHECK: br i1 %switch, label %a, label %b
+
+entry:
+  ; Note: folding the most popular case destination into the default
+  ; would prevent switch-to-icmp here.
+  switch i32 %x, label %unreachable [
+    i32 5, label %a
+    i32 6, label %a
+    i32 7, label %a
+    i32 10, label %b
+    i32 20, label %b
+  ]
+unreachable:
+  unreachable
+a:
+  %0 = call i32 @f(i32 0)
+  ret i32 %0
+b:
+  %1 = call i32 @f(i32 1)
+  ret i32 %1
+}
diff --git a/test/Transforms/SimplifyCFG/switch-to-select-multiple-edge-per-block-phi.ll b/test/Transforms/SimplifyCFG/switch-to-select-multiple-edge-per-block-phi.ll
index ddf5d1f06ce7..f5359b5b2b32 100644
--- a/test/Transforms/SimplifyCFG/switch-to-select-multiple-edge-per-block-phi.ll
+++ b/test/Transforms/SimplifyCFG/switch-to-select-multiple-edge-per-block-phi.ll
@@ -20,12 +20,12 @@ define i32 @fn1() {
 ; CHECK: %switch.selectcmp1 = icmp eq i32 %1, 5
 ; CHECK: %switch.select2 = select i1 %switch.selectcmp1, i32 5, i32 %switch.select
 entry:
-  %0 = load i32* @b, align 4
+  %0 = load i32, i32* @b, align 4
   %tobool = icmp eq i32 %0, 0
   br i1 %tobool, label %if.end3, label %if.then
 
 if.then:
-  %1 = load i32* @a, align 4
+  %1 = load i32, i32* @a, align 4
   switch i32 %1, label %if.end3 [
     i32 5, label %return
     i32 0, label %return
diff --git a/test/Transforms/SimplifyCFG/switch-to-select-two-case.ll b/test/Transforms/SimplifyCFG/switch-to-select-two-case.ll
index 69f97e5f9f9f..f4d171ad2b18 100644
--- a/test/Transforms/SimplifyCFG/switch-to-select-two-case.ll
+++ b/test/Transforms/SimplifyCFG/switch-to-select-two-case.ll
@@ -35,38 +35,3 @@ return:
   %retval.0 = phi i32 [ 4, %sw.epilog ], [ 2, %sw.bb1 ], [ 10, %sw.bb ]
   ret i32 %retval.0
 }
-
-; int foo1_without_default(int a) {
-;   switch(a) {
-;     case 10:
-;       return 10;
-;     case 20:
-;       return 2;
-;   }
-;   __builtin_unreachable();
-; }
-
-define i32 @foo1_without_default(i32 %a) {
-; CHECK-LABEL: @foo1_without_default
-; CHECK: %switch.selectcmp = icmp eq i32 %a, 10
-; CHECK-NEXT: %switch.select = select i1 %switch.selectcmp, i32 10, i32 2
-; CHECK-NOT: %switch.selectcmp1
-entry:
-  switch i32 %a, label %sw.epilog [
-    i32 10, label %sw.bb
-    i32 20, label %sw.bb1
-  ]
-
-sw.bb:
-  br label %return
-
-sw.bb1:
-  br label %return
-
-sw.epilog:
-  unreachable
-
-return:
-  %retval.0 = phi i32 [ 2, %sw.bb1 ], [ 10, %sw.bb ]
-  ret i32 %retval.0
-}
diff --git a/test/Transforms/SimplifyCFG/switch_create.ll b/test/Transforms/SimplifyCFG/switch_create.ll
index e1e91570300c..490b7513a944 100644
--- a/test/Transforms/SimplifyCFG/switch_create.ll
+++ b/test/Transforms/SimplifyCFG/switch_create.ll
@@ -154,8 +154,8 @@ lor.end:                                          ; preds = %entry, %entry, %ent
 
 define i1 @test6({ i32, i32 }* %I) {
 entry:
-        %tmp.1.i = getelementptr { i32, i32 }* %I, i64 0, i32 1         ; <i32*> [#uses=1]
-        %tmp.2.i = load i32* %tmp.1.i           ; <i32> [#uses=6]
+        %tmp.1.i = getelementptr { i32, i32 }, { i32, i32 }* %I, i64 0, i32 1         ; <i32*> [#uses=1]
+        %tmp.2.i = load i32, i32* %tmp.1.i           ; <i32> [#uses=6]
         %tmp.2 = icmp eq i32 %tmp.2.i, 14               ; <i1> [#uses=1]
         br i1 %tmp.2, label %shortcirc_done.4, label %shortcirc_next.0
 shortcirc_next.0:               ; preds = %entry
diff --git a/test/Transforms/SimplifyCFG/trap-debugloc.ll b/test/Transforms/SimplifyCFG/trap-debugloc.ll
index 24a286fd9735..fedf6b172d75 100644
--- a/test/Transforms/SimplifyCFG/trap-debugloc.ll
+++ b/test/Transforms/SimplifyCFG/trap-debugloc.ll
@@ -11,14 +11,14 @@ define void @foo() nounwind ssp {
 !llvm.module.flags = !{!10}
 !llvm.dbg.sp = !{!0}
 
-!0 = !{!"0x2e\00foo\00foo\00\003\000\001\000\006\000\000\000", !8, !1, !3, null, void ()* @foo, null, null, null} ; [ DW_TAG_subprogram ] [line 3] [def] [scope 0] [foo]
-!1 = !{!"0x29", !8} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00Apple clang version 3.0 (tags/Apple/clang-206.1) (based on LLVM 3.0svn)\001\00\000\00\000", !8, !4, !4, !9, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !8, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, file: !8, scope: !1, type: !3, function: void ()* @foo)
+!1 = !DIFile(filename: "foo.c", directory: "/private/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "Apple clang version 3.0 (tags/Apple/clang-206.1) (based on LLVM 3.0svn)", isOptimized: true, emissionKind: 0, file: !8, enums: !{}, retainedTypes: !{}, subprograms: !9)
+!3 = !DISubroutineType(types: !4)
 !4 = !{null}
-!5 = !MDLocation(line: 4, column: 2, scope: !6)
-!6 = !{!"0xb\003\0012\000", !8, !0} ; [ DW_TAG_lexical_block ]
-!7 = !MDLocation(line: 5, column: 1, scope: !6)
-!8 = !{!"foo.c", !"/private/tmp"}
+!5 = !DILocation(line: 4, column: 2, scope: !6)
+!6 = distinct !DILexicalBlock(line: 3, column: 12, file: !8, scope: !0)
+!7 = !DILocation(line: 5, column: 1, scope: !6)
+!8 = !DIFile(filename: "foo.c", directory: "/private/tmp")
 !9 = !{!0}
-!10 = !{i32 1, !"Debug Info Version", i32 2}
+!10 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/SimplifyCFG/trapping-load-unreachable.ll b/test/Transforms/SimplifyCFG/trapping-load-unreachable.ll
index 5ae62af54581..5881367d9610 100644
--- a/test/Transforms/SimplifyCFG/trapping-load-unreachable.ll
+++ b/test/Transforms/SimplifyCFG/trapping-load-unreachable.ll
@@ -11,7 +11,7 @@ entry:
         br i1 %0, label %bb, label %return
 
 bb:             ; preds = %entry
-        %1 = load volatile i32* null
+        %1 = load volatile i32, i32* null
         unreachable
         
         br label %return
diff --git a/test/Transforms/SimplifyCFG/trivial-throw.ll b/test/Transforms/SimplifyCFG/trivial-throw.ll
deleted file mode 100644
index ca2b5693e600..000000000000
--- a/test/Transforms/SimplifyCFG/trivial-throw.ll
+++ /dev/null
@@ -1,77 +0,0 @@
-; RUN: opt -simplifycfg -S < %s | FileCheck %s
-; <rdar://problem/13360379>
-
-@_ZTVN10__cxxabiv117__class_type_infoE = external global i8*
-@_ZTS13TestException = linkonce_odr constant [16 x i8] c"13TestException\00"
-@_ZTI13TestException = linkonce_odr unnamed_addr constant { i8*, i8* } { i8* bitcast (i8** getelementptr inbounds (i8** @_ZTVN10__cxxabiv117__class_type_infoE, i64 2) to i8*), i8* getelementptr inbounds ([16 x i8]* @_ZTS13TestException, i32 0, i32 0) }
-
-define void @throw(i32 %n) #0 {
-entry:
-  %exception = call i8* @__cxa_allocate_exception(i64 1) #4
-  call void @__cxa_throw(i8* %exception, i8* bitcast ({ i8*, i8* }* @_ZTI13TestException to i8*), i8* null) #2
-  unreachable
-}
-
-define void @func() #0 {
-entry:
-; CHECK: func()
-; CHECK: invoke void @throw
-; CHECK-NOT: call void @throw
-  invoke void @throw(i32 42) #0
-          to label %exit unwind label %lpad
-
-lpad:
-  %tmp0 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
-          cleanup
-  resume { i8*, i32 } %tmp0
-
-exit:
-  invoke void @abort() #2
-          to label %invoke.cont unwind label %lpad1
-
-invoke.cont:
-  unreachable
-
-lpad1:
-  %tmp1 = landingpad { i8*, i32 } personality i8* bitcast (i32 (...)* @__gxx_personality_v0 to i8*)
-          catch i8* bitcast ({ i8*, i8* }* @_ZTI13TestException to i8*)
-  %tmp2 = extractvalue { i8*, i32 } %tmp1, 1
-  %tmp3 = call i32 @llvm.eh.typeid.for(i8* bitcast ({ i8*, i8* }* @_ZTI13TestException to i8*)) #4
-  %matches = icmp eq i32 %tmp2, %tmp3
-  br i1 %matches, label %catch, label %eh.resume
-
-catch:
-  ret void
-
-eh.resume:
-  resume { i8*, i32 } %tmp1
-}
-
-define linkonce_odr hidden void @__clang_call_terminate(i8*) #1 {
-  %2 = call i8* @__cxa_begin_catch(i8* %0) #4
-  call void @_ZSt9terminatev() #5
-  unreachable
-}
-
-declare void @abort() #2
-
-declare i32 @llvm.eh.typeid.for(i8*) #3
-
-declare void @__cxa_end_catch()
-
-declare i8* @__cxa_allocate_exception(i64)
-
-declare i32 @__gxx_personality_v0(...)
-
-declare void @__cxa_throw(i8*, i8*, i8*)
-
-declare i8* @__cxa_begin_catch(i8*)
-
-declare void @_ZSt9terminatev()
-
-attributes #0 = { ssp uwtable }
-attributes #1 = { noinline noreturn nounwind }
-attributes #2 = { noreturn }
-attributes #3 = { nounwind readnone }
-attributes #4 = { nounwind }
-attributes #5 = { noreturn nounwind }
diff --git a/test/Transforms/SimplifyCFG/unreachable-blocks.ll b/test/Transforms/SimplifyCFG/unreachable-blocks.ll
index 1df0eab6259f..87a64ad47c62 100644
--- a/test/Transforms/SimplifyCFG/unreachable-blocks.ll
+++ b/test/Transforms/SimplifyCFG/unreachable-blocks.ll
@@ -10,12 +10,12 @@ entry:
 
 while_block:                                      ; preds = %and_if_cont2, %and_if_cont
   %newlen = sub i32 %newlen, 1
-  %newptr = getelementptr i8* %newptr, i64 1
+  %newptr = getelementptr i8, i8* %newptr, i64 1
   %test = icmp sgt i32 %newlen, 0
   br i1 %test, label %and_if1, label %and_if_cont2
 
 and_if1:                                          ; preds = %while_block
-  %char = load i8* %newptr
+  %char = load i8, i8* %newptr
   %test2 = icmp ule i8 %char, 32
   br label %and_if_cont2
 
diff --git a/test/Transforms/SimplifyCFG/volatile-phioper.ll b/test/Transforms/SimplifyCFG/volatile-phioper.ll
index 6367451433c1..c366d059d334 100644
--- a/test/Transforms/SimplifyCFG/volatile-phioper.ll
+++ b/test/Transforms/SimplifyCFG/volatile-phioper.ll
@@ -17,19 +17,19 @@ target datalayout = "e-p:64:64:64-i1:8:8-i8:8:8-i16:16:16-i32:32:32-i64:64:64-f3
 
 define void @test(i8** nocapture %PeiServices) #0 {
 entry:
-  %call = tail call i32 (...)* @Trace() #2
+  %call = tail call i32 (...) @Trace() #2
   %tobool = icmp eq i32 %call, 0
   br i1 %tobool, label %while.body, label %if.then
 
 if.then:                                          ; preds = %entry
-  %call1 = tail call i32 (...)* @Trace() #2
+  %call1 = tail call i32 (...) @Trace() #2
   br label %while.body
 
 while.body:                                       ; preds = %entry, %if.then, %while.body
   %Addr.017 = phi i8* [ %incdec.ptr, %while.body ], [ null, %if.then ], [ null, %entry ]
   %x.016 = phi i8 [ %inc, %while.body ], [ 0, %if.then ], [ 0, %entry ]
   %inc = add i8 %x.016, 1
-  %incdec.ptr = getelementptr inbounds i8* %Addr.017, i64 1
+  %incdec.ptr = getelementptr inbounds i8, i8* %Addr.017, i64 1
   store volatile i8 %x.016, i8* %Addr.017, align 1
   %0 = ptrtoint i8* %incdec.ptr to i64
   %1 = trunc i64 %0 to i32
diff --git a/test/Transforms/Sink/basic.ll b/test/Transforms/Sink/basic.ll
index 4aac6d613a8c..1bbf161921d1 100644
--- a/test/Transforms/Sink/basic.ll
+++ b/test/Transforms/Sink/basic.ll
@@ -8,11 +8,11 @@
 
 ;      CHECK-LABEL: @foo(
 ;      CHECK: true:
-; CHECK-NEXT: %l = load i32* @A
+; CHECK-NEXT: %l = load i32, i32* @A
 ; CHECK-NEXT: ret i32 %l
 
 define i32 @foo(i1 %z) {
-  %l = load i32* @A
+  %l = load i32, i32* @A
   store i32 0, i32* @B
   br i1 %z, label %true, label %false
 true:
@@ -28,7 +28,7 @@ false:
 ; CHECK-NEXT: store i32
 
 define i32 @foo2(i1 %z) {
-  %l = load volatile i32* @A
+  %l = load volatile i32, i32* @A
   store i32 0, i32* @B
   br i1 %z, label %true, label %false
 true:
@@ -75,11 +75,11 @@ entry:
   br i1 %1, label %if, label %endif
 
 if:
-  %2 = getelementptr i32* %0, i32 1
+  %2 = getelementptr i32, i32* %0, i32 1
   store i32 0, i32* %0
   store i32 1, i32* %2
-  %3 = getelementptr i32* %0, i32 %b
-  %4 = load i32* %3
+  %3 = getelementptr i32, i32* %0, i32 %b
+  %4 = load i32, i32* %3
   ret i32 %4
 
 endif:
@@ -100,11 +100,11 @@ entry:
   br i1 %1, label %if, label %endif
 
 if:
-  %2 = getelementptr i32* %0, i32 1
+  %2 = getelementptr i32, i32* %0, i32 1
   store i32 0, i32* %0
   store i32 1, i32* %2
-  %3 = getelementptr i32* %0, i32 %b
-  %4 = load i32* %3
+  %3 = getelementptr i32, i32* %0, i32 %b
+  %4 = load i32, i32* %3
   ret i32 %4
 
 endif:
@@ -131,11 +131,11 @@ if0:
   br i1 %1, label %if, label %endif
 
 if:
-  %2 = getelementptr i32* %0, i32 1
+  %2 = getelementptr i32, i32* %0, i32 1
   store i32 0, i32* %0
   store i32 1, i32* %2
-  %3 = getelementptr i32* %0, i32 %b
-  %4 = load i32* %3
+  %3 = getelementptr i32, i32* %0, i32 %b
+  %4 = load i32, i32* %3
   ret i32 %4
 
 endif:
diff --git a/test/Transforms/SpeculativeExecution/spec.ll b/test/Transforms/SpeculativeExecution/spec.ll
new file mode 100644
index 000000000000..b0953c85dd40
--- /dev/null
+++ b/test/Transforms/SpeculativeExecution/spec.ll
@@ -0,0 +1,195 @@
+; RUN: opt < %s -S -speculative-execution \
+; RUN:   -spec-exec-max-speculation-cost 4 -spec-exec-max-not-hoisted 3 \
+; RUN:   | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+; Hoist in if-then pattern.
+define void @ifThen() {
+; CHECK-LABEL: @ifThen(
+; CHECK: %x = add i32 2, 3
+; CHECK: br i1 true
+  br i1 true, label %a, label %b
+; CHECK: a:
+a:
+  %x = add i32 2, 3
+; CHECK: br label
+  br label %b
+; CHECK: b:
+b:
+; CHECK: ret void
+  ret void
+}
+
+; Hoist in if-else pattern.
+define void @ifElse() {
+; CHECK-LABEL: @ifElse(
+; CHECK: %x = add i32 2, 3
+; CHECK: br i1 true
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+  %x = add i32 2, 3
+; CHECK: br label
+  br label %b
+; CHECK: b:
+b:
+; CHECK: ret void
+  ret void
+}
+
+; Hoist in if-then-else pattern if it is equivalent to if-then.
+define void @ifElseThenAsIfThen() {
+; CHECK-LABEL: @ifElseThenAsIfThen(
+; CHECK: %x = add i32 2, 3
+; CHECK: br
+  br i1 true, label %a, label %b
+; CHECK: a:
+a:
+  %x = add i32 2, 3
+; CHECK: br label
+  br label %c
+; CHECK: b:
+b:
+  br label %c
+; CHECK: c
+c:
+  ret void
+}
+
+; Hoist in if-then-else pattern if it is equivalent to if-else.
+define void @ifElseThenAsIfElse() {
+; CHECK-LABEL: @ifElseThenAsIfElse(
+; CHECK: %x = add i32 2, 3
+; CHECK: br
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+  %x = add i32 2, 3
+; CHECK: br label
+  br label %c
+; CHECK: b:
+b:
+  br label %c
+; CHECK: c
+c:
+  ret void
+}
+
+; Do not hoist if-then-else pattern if it is not equivalent to if-then
+; or if-else.
+define void @ifElseThen() {
+; CHECK-LABEL: @ifElseThen(
+; CHECK: br
+  br i1 true, label %a, label %b
+; CHECK: a:
+a:
+; CHECK: %x = add
+  %x = add i32 2, 3
+; CHECK: br label
+  br label %c
+; CHECK: b:
+b:
+; CHECK: %y = add
+  %y = add i32 2, 3
+  br label %c
+; CHECK: c
+c:
+  ret void
+}
+
+; Do not hoist loads and do not hoist an instruction past a definition of
+; an operand.
+define void @doNotHoistPastDef() {
+; CHECK-LABEL: @doNotHoistPastDef(
+  br i1 true, label %b, label %a
+; CHECK-NOT: load
+; CHECK-NOT: add
+; CHECK: a:
+a:
+; CHECK: %def = load
+  %def = load i32, i32* null
+; CHECK: %use = add
+  %use = add i32 %def, 0
+  br label %b
+; CHECK: b:
+b:
+  ret void
+}
+
+; Case with nothing to speculate.
+define void @nothingToSpeculate() {
+; CHECK-LABEL: @nothingToSpeculate(
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+; CHECK: %def = load
+  %def = load i32, i32* null
+  br label %b
+; CHECK: b:
+b:
+  ret void
+}
+
+; Still hoist if an operand is defined before the block or is itself hoisted.
+define void @hoistIfNotPastDef() {
+; CHECK-LABEL: @hoistIfNotPastDef(
+; CHECK: %x = load
+  %x = load i32, i32* null
+; CHECK: %y = add i32 %x, 1
+; CHECK: %z = add i32 %y, 1
+; CHECK: br
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+  %y = add i32 %x, 1
+  %z = add i32 %y, 1
+  br label %b
+; CHECK: b:
+b:
+  ret void
+}
+
+; Do not hoist if the speculation cost is too high.
+define void @costTooHigh() {
+; CHECK-LABEL: @costTooHigh(
+; CHECK: br
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+; CHECK: %r1 = add
+  %r1 = add i32 1, 1
+; CHECK: %r2 = add
+  %r2 = add i32 1, 1
+; CHECK: %r3 = add
+  %r3 = add i32 1, 1
+; CHECK: %r4 = add
+  %r4 = add i32 1, 1
+; CHECK: %r5 = add
+  %r5 = add i32 1, 1
+  br label %b
+; CHECK: b:
+b:
+  ret void
+}
+
+; Do not hoist if too many instructions are left behind.
+define void @tooMuchLeftBehind() {
+; CHECK-LABEL: @tooMuchLeftBehind(
+; CHECK: br
+  br i1 true, label %b, label %a
+; CHECK: a:
+a:
+; CHECK: %x = load
+  %x = load i32, i32* null
+; CHECK: %r1 = add
+  %r1 = add i32 %x, 1
+; CHECK: %r2 = add
+  %r2 = add i32 %x, 1
+; CHECK: %r3 = add
+  %r3 = add i32 %x, 1
+  br label %b
+; CHECK: b:
+b:
+  ret void
+}
diff --git a/test/Transforms/StraightLineStrengthReduce/NVPTX/lit.local.cfg b/test/Transforms/StraightLineStrengthReduce/NVPTX/lit.local.cfg
new file mode 100644
index 000000000000..2cb98eb371b2
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/NVPTX/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'NVPTX' in config.root.targets:
+    config.unsupported = True
diff --git a/test/Transforms/StraightLineStrengthReduce/NVPTX/reassociate-geps-and-slsr.ll b/test/Transforms/StraightLineStrengthReduce/NVPTX/reassociate-geps-and-slsr.ll
new file mode 100644
index 000000000000..03c0356eb5b4
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/NVPTX/reassociate-geps-and-slsr.ll
@@ -0,0 +1,74 @@
+; RUN: opt < %s -separate-const-offset-from-gep -slsr -gvn -S | FileCheck %s
+; RUN: llc < %s -march=nvptx64 -mcpu=sm_35 | FileCheck %s --check-prefix=PTX
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+target triple = "nvptx64-unknown-unknown"
+
+; arr[i + 5]
+; arr[i * 2 + 5]
+; arr[i * 3 + 5]
+; arr[i * 4 + 5]
+;
+;   => reassociate-geps
+;
+; *(&arr[i] + 5)
+; *(&arr[i * 2] + 5)
+; *(&arr[i * 3] + 5)
+; *(&arr[i * 4] + 5)
+;
+;   => slsr
+;
+; p1 = &arr[i]
+; *(p1 + 5)
+; p2 = p1 + i
+; *(p2 + 5)
+; p3 = p2 + i
+; *(p3 + 5)
+; p4 = p3 + i
+; *(p4 + 5)
+define void @slsr_after_reassociate_geps(float* %arr, i32 %i) {
+; CHECK-LABEL: @slsr_after_reassociate_geps(
+; PTX-LABEL: .visible .func slsr_after_reassociate_geps(
+; PTX: ld.param.u64 [[arr:%rd[0-9]+]], [slsr_after_reassociate_geps_param_0];
+; PTX: ld.param.u32 [[i:%r[0-9]+]], [slsr_after_reassociate_geps_param_1];
+  %i2 = shl nsw i32 %i, 1
+  %i3 = mul nsw i32 %i, 3
+  %i4 = shl nsw i32 %i, 2
+
+  %j1 = add nsw i32 %i, 5
+  %p1 = getelementptr inbounds float, float* %arr, i32 %j1
+; CHECK: [[b1:%[0-9]+]] = getelementptr float, float* %arr, i64 [[bump:%[0-9]+]]
+; PTX: mul.wide.s32 [[i4:%rd[0-9]+]], [[i]], 4;
+; PTX: add.s64 [[base1:%rd[0-9]+]], [[arr]], [[i4]];
+  %v1 = load float, float* %p1, align 4
+; PTX: ld.f32 {{%f[0-9]+}}, {{\[}}[[base1]]+20];
+  call void @foo(float %v1)
+
+  %j2 = add nsw i32 %i2, 5
+  %p2 = getelementptr inbounds float, float* %arr, i32 %j2
+; CHECK: [[b2:%[0-9]+]] = getelementptr float, float* [[b1]], i64 [[bump]]
+; PTX: add.s64 [[base2:%rd[0-9]+]], [[base1]], [[i4]];
+  %v2 = load float, float* %p2, align 4
+; PTX: ld.f32 {{%f[0-9]+}}, {{\[}}[[base2]]+20];
+  call void @foo(float %v2)
+
+  %j3 = add nsw i32 %i3, 5
+  %p3 = getelementptr inbounds float, float* %arr, i32 %j3
+; CHECK: [[b3:%[0-9]+]] = getelementptr float, float* [[b2]], i64 [[bump]]
+; PTX: add.s64 [[base3:%rd[0-9]+]], [[base2]], [[i4]];
+  %v3 = load float, float* %p3, align 4
+; PTX: ld.f32 {{%f[0-9]+}}, {{\[}}[[base3]]+20];
+  call void @foo(float %v3)
+
+  %j4 = add nsw i32 %i4, 5
+  %p4 = getelementptr inbounds float, float* %arr, i32 %j4
+; CHECK: [[b4:%[0-9]+]] = getelementptr float, float* [[b3]], i64 [[bump]]
+; PTX: add.s64 [[base4:%rd[0-9]+]], [[base3]], [[i4]];
+  %v4 = load float, float* %p4, align 4
+; PTX: ld.f32 {{%f[0-9]+}}, {{\[}}[[base4]]+20];
+  call void @foo(float %v4)
+
+  ret void
+}
+
+declare void @foo(float)
diff --git a/test/Transforms/StraightLineStrengthReduce/X86/lit.local.cfg b/test/Transforms/StraightLineStrengthReduce/X86/lit.local.cfg
new file mode 100644
index 000000000000..c8625f4d9d24
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/X86/lit.local.cfg
@@ -0,0 +1,2 @@
+if not 'X86' in config.root.targets:
+    config.unsupported = True
diff --git a/test/Transforms/StraightLineStrengthReduce/X86/no-slsr.ll b/test/Transforms/StraightLineStrengthReduce/X86/no-slsr.ll
new file mode 100644
index 000000000000..f11cbc5897a6
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/X86/no-slsr.ll
@@ -0,0 +1,44 @@
+; RUN: opt < %s -slsr -gvn -S | FileCheck %s
+
+target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
+target triple = "x86_64-unknown-linux-gnu"
+
+; Do not perform SLSR on &input[s] and &input[s * 2] which fit into addressing
+; modes of X86.
+define i32 @no_slsr_gep(i32* %input, i64 %s) {
+; CHECK-LABEL: @no_slsr_gep(
+  ; v0 = input[0];
+  %p0 = getelementptr inbounds i32, i32* %input, i64 0
+  %v0 = load i32, i32* %p0
+
+  ; v1 = input[s];
+  %p1 = getelementptr inbounds i32, i32* %input, i64 %s
+; CHECK: %p1 = getelementptr inbounds i32, i32* %input, i64 %s
+  %v1 = load i32, i32* %p1
+
+  ; v2 = input[s * 2];
+  %s2 = mul nsw i64 %s, 2
+  %p2 = getelementptr inbounds i32, i32* %input, i64 %s2
+; CHECK: %p2 = getelementptr inbounds i32, i32* %input, i64 %s2
+  %v2 = load i32, i32* %p2
+
+  ; return v0 + v1 + v2;
+  %1 = add i32 %v0, %v1
+  %2 = add i32 %1, %v2
+  ret i32 %2
+}
+
+define void @no_slsr_add(i32 %b, i32 %s) {
+; CHECK-LABEL: @no_slsr_add(
+  %1 = add i32 %b, %s
+; CHECK: add i32 %b, %s
+  call void @foo(i32 %1)
+  %s2 = mul i32 %s, 2
+; CHECK: %s2 = mul i32 %s, 2
+  %2 = add i32 %b, %s2
+; CHECK: add i32 %b, %s2
+  call void @foo(i32 %2)
+  ret void
+}
+
+declare void @foo(i32 %a)
diff --git a/test/Transforms/StraightLineStrengthReduce/slsr-add.ll b/test/Transforms/StraightLineStrengthReduce/slsr-add.ll
new file mode 100644
index 000000000000..e25ddc2888a3
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/slsr-add.ll
@@ -0,0 +1,101 @@
+; RUN: opt < %s -slsr -gvn -S | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+define void @shl(i32 %b, i32 %s) {
+; CHECK-LABEL: @shl(
+  %1 = add i32 %b, %s
+; [[BASIS:%[a-zA-Z0-9]+]] = add i32 %b, %s
+  call void @foo(i32 %1)
+  %s2 = shl i32 %s, 1
+  %2 = add i32 %b, %s2
+; add i32 [[BASIS]], %s
+  call void @foo(i32 %2)
+  ret void
+}
+
+define void @stride_is_2s(i32 %b, i32 %s) {
+; CHECK-LABEL: @stride_is_2s(
+  %s2 = shl i32 %s, 1
+; CHECK: %s2 = shl i32 %s, 1
+  %1 = add i32 %b, %s2
+; CHECK: [[t1:%[a-zA-Z0-9]+]] = add i32 %b, %s2
+  call void @foo(i32 %1)
+  %s4 = shl i32 %s, 2
+  %2 = add i32 %b, %s4
+; CHECK: [[t2:%[a-zA-Z0-9]+]] = add i32 [[t1]], %s2
+  call void @foo(i32 %2)
+  %s6 = mul i32 %s, 6
+  %3 = add i32 %b, %s6
+; CHECK: add i32 [[t2]], %s2
+  call void @foo(i32 %3)
+  ret void
+}
+
+define void @stride_is_3s(i32 %b, i32 %s) {
+; CHECK-LABEL: @stride_is_3s(
+  %1 = add i32 %s, %b
+; CHECK: [[t1:%[a-zA-Z0-9]+]] = add i32 %s, %b
+  call void @foo(i32 %1)
+  %s4 = shl i32 %s, 2
+  %2 = add i32 %s4, %b
+; CHECK: [[bump:%[a-zA-Z0-9]+]] = mul i32 %s, 3
+; CHECK: [[t2:%[a-zA-Z0-9]+]] = add i32 [[t1]], [[bump]]
+  call void @foo(i32 %2)
+  %s7 = mul i32 %s, 7
+  %3 = add i32 %s7, %b
+; CHECK: add i32 [[t2]], [[bump]]
+  call void @foo(i32 %3)
+  ret void
+}
+
+; foo(b + 6 * s);
+; foo(b + 4 * s);
+; foo(b + 2 * s);
+;   =>
+; t1 = b + 6 * s;
+; foo(t1);
+; s2 = 2 * s;
+; t2 = t1 - s2;
+; foo(t2);
+; t3 = t2 - s2;
+; foo(t3);
+define void @stride_is_minus_2s(i32 %b, i32 %s) {
+; CHECK-LABEL: @stride_is_minus_2s(
+  %s6 = mul i32 %s, 6
+  %1 = add i32 %b, %s6
+; CHECK: [[t1:%[a-zA-Z0-9]+]] = add i32 %b, %s6
+; CHECK: call void @foo(i32 [[t1]])
+  call void @foo(i32 %1)
+  %s4 = shl i32 %s, 2
+  %2 = add i32 %b, %s4
+; CHECK: [[bump:%[a-zA-Z0-9]+]] = shl i32 %s, 1
+; CHECK: [[t2:%[a-zA-Z0-9]+]] = sub i32 [[t1]], [[bump]]
+  call void @foo(i32 %2)
+; CHECK: call void @foo(i32 [[t2]])
+  %s2 = shl i32 %s, 1
+  %3 = add i32 %b, %s2
+; CHECK: [[t3:%[a-zA-Z0-9]+]] = sub i32 [[t2]], [[bump]]
+  call void @foo(i32 %3)
+; CHECK: call void @foo(i32 [[t3]])
+  ret void
+}
+
+; t = b + (s << 3);
+; foo(t);
+; foo(b + s);
+;
+; do not rewrite b + s to t - 7 * s because the latter is more complicated.
+define void @simple_enough(i32 %b, i32 %s) {
+; CHECK-LABEL: @simple_enough(
+  %s8 = shl i32 %s, 3
+  %1 = add i32 %b, %s8
+  call void @foo(i32 %1)
+  %2 = add i32 %b, %s
+; CHECK: [[t:%[a-zA-Z0-9]+]] = add i32 %b, %s{{$}}
+  call void @foo(i32 %2)
+; CHECK: call void @foo(i32 [[t]])
+  ret void
+}
+
+declare void @foo(i32)
diff --git a/test/Transforms/StraightLineStrengthReduce/slsr-gep.ll b/test/Transforms/StraightLineStrengthReduce/slsr-gep.ll
new file mode 100644
index 000000000000..bd92780a036c
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/slsr-gep.ll
@@ -0,0 +1,168 @@
+; RUN: opt < %s -slsr -gvn -S | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+; foo(input[0]);
+; foo(input[s]);
+; foo(input[s * 2]);
+;   =>
+; p0 = &input[0];
+; foo(*p);
+; p1 = p0 + s;
+; foo(*p1);
+; p2 = p1 + s;
+; foo(*p2);
+define void @slsr_gep(i32* %input, i64 %s) {
+; CHECK-LABEL: @slsr_gep(
+  ; v0 = input[0];
+  %p0 = getelementptr inbounds i32, i32* %input, i64 0
+  %v0 = load i32, i32* %p0
+  call void @foo(i32 %v0)
+
+  ; v1 = input[s];
+  %p1 = getelementptr inbounds i32, i32* %input, i64 %s
+; CHECK: %p1 = getelementptr inbounds i32, i32* %input, i64 %s
+  %v1 = load i32, i32* %p1
+  call void @foo(i32 %v1)
+
+  ; v2 = input[s * 2];
+  %s2 = shl nsw i64 %s, 1
+  %p2 = getelementptr inbounds i32, i32* %input, i64 %s2
+; CHECK: %p2 = getelementptr inbounds i32, i32* %p1, i64 %s
+  %v2 = load i32, i32* %p2
+  call void @foo(i32 %v2)
+
+  ret void
+}
+
+; foo(input[0]);
+; foo(input[(long)s]);
+; foo(input[(long)(s * 2)]);
+;   =>
+; p0 = &input[0];
+; foo(*p);
+; p1 = p0 + (long)s;
+; foo(*p1);
+; p2 = p1 + (long)s;
+; foo(*p2);
+define void @slsr_gep_sext(i32* %input, i32 %s) {
+; CHECK-LABEL: @slsr_gep_sext(
+  ; v0 = input[0];
+  %p0 = getelementptr inbounds i32, i32* %input, i64 0
+  %v0 = load i32, i32* %p0
+  call void @foo(i32 %v0)
+
+  ; v1 = input[s];
+  %t = sext i32 %s to i64
+  %p1 = getelementptr inbounds i32, i32* %input, i64 %t
+; CHECK: %p1 = getelementptr inbounds i32, i32* %input, i64 %t
+  %v1 = load i32, i32* %p1
+  call void @foo(i32 %v1)
+
+  ; v2 = input[s * 2];
+  %s2 = shl nsw i32 %s, 1
+  %t2 = sext i32 %s2 to i64
+  %p2 = getelementptr inbounds i32, i32* %input, i64 %t2
+; CHECK: %p2 = getelementptr inbounds i32, i32* %p1, i64 %t
+  %v2 = load i32, i32* %p2
+  call void @foo(i32 %v2)
+
+  ret void
+}
+
+; int input[10][5];
+; foo(input[s][t]);
+; foo(input[s * 2][t]);
+; foo(input[s * 3][t]);
+;   =>
+; p0 = &input[s][t];
+; foo(*p0);
+; p1 = p0 + 5s;
+; foo(*p1);
+; p2 = p1 + 5s;
+; foo(*p2);
+define void @slsr_gep_2d([10 x [5 x i32]]* %input, i64 %s, i64 %t) {
+; CHECK-LABEL: @slsr_gep_2d(
+  ; v0 = input[s][t];
+  %p0 = getelementptr inbounds [10 x [5 x i32]], [10 x [5 x i32]]* %input, i64 0, i64 %s, i64 %t
+  %v0 = load i32, i32* %p0
+  call void @foo(i32 %v0)
+
+  ; v1 = input[s * 2][t];
+  %s2 = shl nsw i64 %s, 1
+; CHECK: [[BUMP:%[a-zA-Z0-9]+]] = mul i64 %s, 5
+  %p1 = getelementptr inbounds [10 x [5 x i32]], [10 x [5 x i32]]* %input, i64 0, i64 %s2, i64 %t
+; CHECK: %p1 = getelementptr inbounds i32, i32* %p0, i64 [[BUMP]]
+  %v1 = load i32, i32* %p1
+  call void @foo(i32 %v1)
+
+  ; v3 = input[s * 3][t];
+  %s3 = mul nsw i64 %s, 3
+  %p2 = getelementptr inbounds [10 x [5 x i32]], [10 x [5 x i32]]* %input, i64 0, i64 %s3, i64 %t
+; CHECK: %p2 = getelementptr inbounds i32, i32* %p1, i64 [[BUMP]]
+  %v2 = load i32, i32* %p2
+  call void @foo(i32 %v2)
+
+  ret void
+}
+
+%struct.S = type <{ i64, i32 }>
+
+; In this case, the bump
+;     = (char *)&input[s * 2][t].f1 - (char *)&input[s][t].f1
+;     = 60 * s
+; which may not be divisible by typeof(input[s][t].f1) = 8. Therefore, we
+; rewrite the candidates using byte offset instead of index offset as in
+; @slsr_gep_2d.
+define void @slsr_gep_uglygep([10 x [5 x %struct.S]]* %input, i64 %s, i64 %t) {
+; CHECK-LABEL: @slsr_gep_uglygep(
+  ; v0 = input[s][t].f1;
+  %p0 = getelementptr inbounds [10 x [5 x %struct.S]], [10 x [5 x %struct.S]]* %input, i64 0, i64 %s, i64 %t, i32 0
+  %v0 = load i64, i64* %p0
+  call void @bar(i64 %v0)
+
+  ; v1 = input[s * 2][t].f1;
+  %s2 = shl nsw i64 %s, 1
+; CHECK: [[BUMP:%[a-zA-Z0-9]+]] = mul i64 %s, 60
+  %p1 = getelementptr inbounds [10 x [5 x %struct.S]], [10 x [5 x %struct.S]]* %input, i64 0, i64 %s2, i64 %t, i32 0
+; CHECK: getelementptr inbounds i8, i8* %{{[0-9]+}}, i64 [[BUMP]]
+  %v1 = load i64, i64* %p1
+  call void @bar(i64 %v1)
+
+  ; v2 = input[s * 3][t].f1;
+  %s3 = mul nsw i64 %s, 3
+  %p2 = getelementptr inbounds [10 x [5 x %struct.S]], [10 x [5 x %struct.S]]* %input, i64 0, i64 %s3, i64 %t, i32 0
+; CHECK: getelementptr inbounds i8, i8* %{{[0-9]+}}, i64 [[BUMP]]
+  %v2 = load i64, i64* %p2
+  call void @bar(i64 %v2)
+
+  ret void
+}
+
+define void @slsr_out_of_bounds_gep(i32* %input, i32 %s) {
+; CHECK-LABEL: @slsr_out_of_bounds_gep(
+  ; v0 = input[0];
+  %p0 = getelementptr i32, i32* %input, i64 0
+  %v0 = load i32, i32* %p0
+  call void @foo(i32 %v0)
+
+  ; v1 = input[(long)s];
+  %t = sext i32 %s to i64
+  %p1 = getelementptr i32, i32* %input, i64 %t
+; CHECK: %p1 = getelementptr i32, i32* %input, i64 %t
+  %v1 = load i32, i32* %p1
+  call void @foo(i32 %v1)
+
+  ; v2 = input[(long)(s * 2)];
+  %s2 = shl nsw i32 %s, 1
+  %t2 = sext i32 %s2 to i64
+  %p2 = getelementptr i32, i32* %input, i64 %t2
+; CHECK: %p2 = getelementptr i32, i32* %p1, i64 %t
+  %v2 = load i32, i32* %p2
+  call void @foo(i32 %v2)
+
+  ret void
+}
+
+declare void @foo(i32)
+declare void @bar(i64)
diff --git a/test/Transforms/StraightLineStrengthReduce/slsr-mul.ll b/test/Transforms/StraightLineStrengthReduce/slsr-mul.ll
new file mode 100644
index 000000000000..56b1d1fc5cdd
--- /dev/null
+++ b/test/Transforms/StraightLineStrengthReduce/slsr-mul.ll
@@ -0,0 +1,147 @@
+; RUN: opt < %s -slsr -gvn -S | FileCheck %s
+
+target datalayout = "e-i64:64-v16:16-v32:32-n16:32:64"
+
+define void @slsr1(i32 %b, i32 %s) {
+; CHECK-LABEL: @slsr1(
+  ; foo(b * s);
+  %mul0 = mul i32 %b, %s
+; CHECK: mul i32
+; CHECK-NOT: mul i32
+  call void @foo(i32 %mul0)
+
+  ; foo((b + 1) * s);
+  %b1 = add i32 %b, 1
+  %mul1 = mul i32 %b1, %s
+  call void @foo(i32 %mul1)
+
+  ; foo((b + 2) * s);
+  %b2 = add i32 %b, 2
+  %mul2 = mul i32 %b2, %s
+  call void @foo(i32 %mul2)
+
+  ret void
+}
+
+define void @non_canonicalized(i32 %b, i32 %s) {
+; CHECK-LABEL: @non_canonicalized(
+  ; foo(b * s);
+  %mul0 = mul i32 %b, %s
+; CHECK: mul i32
+; CHECK-NOT: mul i32
+  call void @foo(i32 %mul0)
+
+  ; foo((1 + b) * s);
+  %b1 = add i32 1, %b
+  %mul1 = mul i32 %b1, %s
+  call void @foo(i32 %mul1)
+
+  ; foo((2 + b) * s);
+  %b2 = add i32 2, %b
+  %mul2 = mul i32 %b2, %s
+  call void @foo(i32 %mul2)
+
+  ret void
+}
+
+define void @or(i32 %a, i32 %s) {
+  %b = shl i32 %a, 1
+; CHECK-LABEL: @or(
+  ; foo(b * s);
+  %mul0 = mul i32 %b, %s
+; CHECK: [[base:[^ ]+]] = mul i32
+  call void @foo(i32 %mul0)
+
+  ; foo((b | 1) * s);
+  %b1 = or i32 %b, 1
+  %mul1 = mul i32 %b1, %s
+; CHECK: add i32 [[base]], %s
+  call void @foo(i32 %mul1)
+
+  ; foo((b | 2) * s);
+  %b2 = or i32 %b, 2
+  %mul2 = mul i32 %b2, %s
+; CHECK: mul i32 %b2, %s
+  call void @foo(i32 %mul2)
+
+  ret void
+}
+
+; foo(a * b)
+; foo((a + 1) * b)
+; foo(a * (b + 1))
+; foo((a + 1) * (b + 1))
+define void @slsr2(i32 %a, i32 %b) {
+; CHECK-LABEL: @slsr2(
+  %a1 = add i32 %a, 1
+  %b1 = add i32 %b, 1
+  %mul0 = mul i32 %a, %b
+; CHECK: mul i32
+; CHECK-NOT: mul i32
+  %mul1 = mul i32 %a1, %b
+  %mul2 = mul i32 %a, %b1
+  %mul3 = mul i32 %a1, %b1
+
+  call void @foo(i32 %mul0)
+  call void @foo(i32 %mul1)
+  call void @foo(i32 %mul2)
+  call void @foo(i32 %mul3)
+
+  ret void
+}
+
+; The bump is a multiple of the stride.
+;
+; foo(b * s);
+; foo((b + 2) * s);
+; foo((b + 4) * s);
+;   =>
+; mul0 = b * s;
+; bump = s * 2;
+; mul1 = mul0 + bump; // GVN ensures mul1 and mul2 use the same bump.
+; mul2 = mul1 + bump;
+define void @slsr3(i32 %b, i32 %s) {
+; CHECK-LABEL: @slsr3(
+  %mul0 = mul i32 %b, %s
+; CHECK: mul i32
+  call void @foo(i32 %mul0)
+
+  %b1 = add i32 %b, 2
+  %mul1 = mul i32 %b1, %s
+; CHECK: [[BUMP:%[a-zA-Z0-9]+]] = shl i32 %s, 1
+; CHECK: %mul1 = add i32 %mul0, [[BUMP]]
+  call void @foo(i32 %mul1)
+
+  %b2 = add i32 %b, 4
+  %mul2 = mul i32 %b2, %s
+; CHECK: %mul2 = add i32 %mul1, [[BUMP]]
+  call void @foo(i32 %mul2)
+
+  ret void
+}
+
+; Do not rewrite a candidate if its potential basis does not dominate it.
+;
+; if (cond)
+;   foo(a * b);
+; foo((a + 1) * b);
+define void @not_dominate(i1 %cond, i32 %a, i32 %b) {
+; CHECK-LABEL: @not_dominate(
+entry:
+  %a1 = add i32 %a, 1
+  br i1 %cond, label %then, label %merge
+
+then:
+  %mul0 = mul i32 %a, %b
+; CHECK: %mul0 = mul i32 %a, %b
+  call void @foo(i32 %mul0)
+  br label %merge
+
+merge:
+  %mul1 = mul i32 %a1, %b
+; CHECK: %mul1 = mul i32 %a1, %b
+  call void @foo(i32 %mul1)
+  ret void
+}
+
+declare void @foo(i32)
diff --git a/test/Transforms/StripSymbols/2010-06-30-StripDebug.ll b/test/Transforms/StripSymbols/2010-06-30-StripDebug.ll
index f2c705a317af..3b1fd74b9813 100644
--- a/test/Transforms/StripSymbols/2010-06-30-StripDebug.ll
+++ b/test/Transforms/StripSymbols/2010-06-30-StripDebug.ll
@@ -18,17 +18,17 @@ declare void @llvm.dbg.value(metadata, i64, metadata, metadata) nounwind readnon
 !llvm.dbg.lv.foo = !{!5}
 !llvm.dbg.gv = !{!8}
 
-!0 = !{!"0x2e\00foo\00foo\00foo\002\000\001\000\006\000\001\000", !12, !1, !3, null, void ()* @foo, null, null, null} ; [ DW_TAG_subprogram ]
-!1 = !{!"0x29", !12} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build)\001\00\000\00\000", !12, !4, !4, null, null, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !12, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", linkageName: "foo", line: 2, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: true, file: !12, scope: !1, type: !3, function: void ()* @foo)
+!1 = !DIFile(filename: "b.c", directory: "/tmp")
+!2 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: 0, file: !12, enums: !4, retainedTypes: !4)
+!3 = !DISubroutineType(types: !4)
 !4 = !{null}
-!5 = !{!"0x100\00y\003\000", !6, !1, !7} ; [ DW_TAG_auto_variable ]
-!6 = !{!"0xb\002\000\000", !12, !0} ; [ DW_TAG_lexical_block ]
-!7 = !{!"0x24\00int\000\0032\0032\000\000\005", !12, !1} ; [ DW_TAG_base_type ]
-!8 = !{!"0x34\00x\00x\00\001\000\001", !1, !1, !7, i32* @x} ; [ DW_TAG_variable ]
+!5 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "y", line: 3, scope: !6, file: !1, type: !7)
+!6 = distinct !DILexicalBlock(line: 2, column: 0, file: !12, scope: !0)
+!7 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!8 = !DIGlobalVariable(name: "x", line: 1, isLocal: false, isDefinition: true, scope: !1, file: !1, type: !7, variable: i32* @x)
 !9 = !{i32 0}
-!10 = !MDLocation(line: 3, scope: !6)
-!11 = !MDLocation(line: 4, scope: !6)
-!12 = !{!"b.c", !"/tmp"}
-!13 = !{i32 1, !"Debug Info Version", i32 2}
+!10 = !DILocation(line: 3, scope: !6)
+!11 = !DILocation(line: 4, scope: !6)
+!12 = !DIFile(filename: "b.c", directory: "/tmp")
+!13 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/StripSymbols/2010-08-25-crash.ll b/test/Transforms/StripSymbols/2010-08-25-crash.ll
index 1534647c2a71..f77ed11d912c 100644
--- a/test/Transforms/StripSymbols/2010-08-25-crash.ll
+++ b/test/Transforms/StripSymbols/2010-08-25-crash.ll
@@ -7,18 +7,18 @@ entry:
 !llvm.dbg.cu = !{!2}
 !llvm.module.flags = !{!14}
 
-!0 = !{!"0x2e\00foo\00foo\00foo\003\000\001\000\006\000\000\000", !10, !1, !3, null, i32 ()* @foo, null, null, null} ; [ DW_TAG_subprogram ]
-!1 = !{!"0x29", !10} ; [ DW_TAG_file_type ]
-!2 = !{!"0x11\0012\00clang version 2.8 (trunk 112062)\001\00\000\00\001", !10, !11, !11, !12, !13, null} ; [ DW_TAG_compile_unit ]
-!3 = !{!"0x15\00\000\000\000\000\000\000", !10, !1, null, !4, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!0 = !DISubprogram(name: "foo", linkageName: "foo", line: 3, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: false, file: !10, scope: !1, type: !3, function: i32 ()* @foo)
+!1 = !DIFile(filename: "/tmp/a.c", directory: "/Volumes/Lalgate/clean/D.CW")
+!2 = !DICompileUnit(language: DW_LANG_C99, producer: "clang version 2.8 (trunk 112062)", isOptimized: true, emissionKind: 1, file: !10, enums: !11, retainedTypes: !11, subprograms: !12, globals: !13)
+!3 = !DISubroutineType(types: !4)
 !4 = !{!5}
-!5 = !{!"0x24\00int\000\0032\0032\000\000\005", !10, !1} ; [ DW_TAG_base_type ]
-!6 = !{!"0x34\00i\00i\00i\002\001\001", !1, !1, !7, i32 0, null} ; [ DW_TAG_variable ]
-!7 = !{!"0x26\00\000\000\000\000\000", !10, !1, !5} ; [ DW_TAG_const_type ]
-!8 = !MDLocation(line: 3, column: 13, scope: !9)
-!9 = !{!"0xb\003\0011\000", !10, !0} ; [ DW_TAG_lexical_block ]
-!10 = !{!"/tmp/a.c", !"/Volumes/Lalgate/clean/D.CW"}
-!11 = !{i32 0}
+!5 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!6 = !DIGlobalVariable(name: "i", linkageName: "i", line: 2, isLocal: true, isDefinition: true, scope: !1, file: !1, type: !7)
+!7 = !DIDerivedType(tag: DW_TAG_const_type, file: !10, scope: !1, baseType: !5)
+!8 = !DILocation(line: 3, column: 13, scope: !9)
+!9 = distinct !DILexicalBlock(line: 3, column: 11, file: !10, scope: !0)
+!10 = !DIFile(filename: "/tmp/a.c", directory: "/Volumes/Lalgate/clean/D.CW")
+!11 = !{}
 !12 = !{!0}
 !13 = !{!6}
-!14 = !{i32 1, !"Debug Info Version", i32 2}
+!14 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/StripSymbols/strip-dead-debug-info.ll b/test/Transforms/StripSymbols/strip-dead-debug-info.ll
index aca7cd608470..08eff003dfca 100644
--- a/test/Transforms/StripSymbols/strip-dead-debug-info.ll
+++ b/test/Transforms/StripSymbols/strip-dead-debug-info.ll
@@ -18,8 +18,8 @@ entry:
 ; Function Attrs: nounwind readonly ssp
 define i32 @foo(i32 %i) #2 {
 entry:
-  tail call void @llvm.dbg.value(metadata i32 %i, i64 0, metadata !15, metadata !{}), !dbg !20
-  %.0 = load i32* @xyz, align 4
+  tail call void @llvm.dbg.value(metadata i32 %i, i64 0, metadata !15, metadata !DIExpression()), !dbg !20
+  %.0 = load i32, i32* @xyz, align 4
   ret i32 %.0, !dbg !21
 }
 
@@ -30,29 +30,29 @@ attributes #2 = { nounwind readonly ssp }
 !llvm.dbg.cu = !{!0}
 !llvm.module.flags = !{!25}
 
-!0 = !{!"0x11\001\004.2.1 (Based on Apple Inc. build 5658) (LLVM build)\001\00\000\00\001", !1, !2, !2, !23, !24, null} ; [ DW_TAG_compile_unit ] [/tmp//g.c] [DW_LANG_C89]
-!1 = !{!"g.c", !"/tmp/"}
+!0 = !DICompileUnit(language: DW_LANG_C89, producer: "4.2.1 (Based on Apple Inc. build 5658) (LLVM build)", isOptimized: true, emissionKind: 1, file: !1, enums: !{}, retainedTypes: !{}, subprograms: !23, globals: !24)
+!1 = !DIFile(filename: "g.c", directory: "/tmp/")
 !2 = !{null}
-!3 = !{!"0x2e\00bar\00bar\00\005\001\001\000\006\000\001\000", !1, null, !4, null, null, null, null, null} ; [ DW_TAG_subprogram ] [line 5] [local] [def] [scope 0] [bar]
-!4 = !{!"0x15\00\000\000\000\000\000\000", !1, !5, null, !2, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
-!5 = !{!"0x29", !1}          ; [ DW_TAG_file_type ] [/tmp//g.c]
-!6 = !{!"0x2e\00fn\00fn\00fn\006\000\001\000\006\000\001\000", !1, null, !7, null, i32 ()* @fn, null, null, null} ; [ DW_TAG_subprogram ] [line 6] [def] [scope 0] [fn]
-!7 = !{!"0x15\00\000\000\000\000\000\000", !1, !5, null, !8, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!3 = !DISubprogram(name: "bar", line: 5, isLocal: true, isDefinition: true, virtualIndex: 6, isOptimized: true, file: !1, scope: null, type: !4)
+!4 = !DISubroutineType(types: !2)
+!5 = !DIFile(filename: "g.c", directory: "/tmp/")
+!6 = !DISubprogram(name: "fn", linkageName: "fn", line: 6, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: true, file: !1, scope: null, type: !7, function: i32 ()* @fn)
+!7 = !DISubroutineType(types: !8)
 !8 = !{!9}
-!9 = !{!"0x24\00int\000\0032\0032\000\000\005", !1, !5} ; [ DW_TAG_base_type ] [int] [line 0, size 32, align 32, offset 0, enc DW_ATE_signed]
-!10 = !{!"0x2e\00foo\00foo\00foo\007\000\001\000\006\000\001\000", !1, null, !11, null, i32 (i32)* @foo, null, null, null} ; [ DW_TAG_subprogram ] [line 7] [def] [scope 0] [foo]
-!11 = !{!"0x15\00\000\000\000\000\000\000", !1, !5, null, !12, null, null, null} ; [ DW_TAG_subroutine_type ] [line 0, size 0, align 0, offset 0] [from ]
+!9 = !DIBasicType(tag: DW_TAG_base_type, name: "int", size: 32, align: 32, encoding: DW_ATE_signed)
+!10 = !DISubprogram(name: "foo", linkageName: "foo", line: 7, isLocal: false, isDefinition: true, virtualIndex: 6, isOptimized: true, file: !1, scope: null, type: !11, function: i32 (i32)* @foo)
+!11 = !DISubroutineType(types: !12)
 !12 = !{!9, !9}
-!13 = !{!"0x100\00bb\005\000", !14, !5, !9} ; [ DW_TAG_auto_variable ]
-!14 = !{!"0xb\005\000\000", !1, !3} ; [ DW_TAG_lexical_block ] [/tmp//g.c]
-!15 = !{!"0x101\00i\007\000", !10, !5, !9} ; [ DW_TAG_arg_variable ]
-!16 = !{!"0x34\00abcd\00abcd\00\002\001\001", !5, !5, !9, null, null} ; [ DW_TAG_variable ]
-!17 = !{!"0x34\00xyz\00xyz\00\003\000\001", !5, !5, !9, i32* @xyz, null} ; [ DW_TAG_variable ]
-!18 = !MDLocation(line: 6, scope: !19)
-!19 = !{!"0xb\006\000\000", !1, !6} ; [ DW_TAG_lexical_block ] [/tmp//g.c]
-!20 = !MDLocation(line: 7, scope: !10)
-!21 = !MDLocation(line: 10, scope: !22)
-!22 = !{!"0xb\007\000\000", !1, !10} ; [ DW_TAG_lexical_block ] [/tmp//g.c]
+!13 = !DILocalVariable(tag: DW_TAG_auto_variable, name: "bb", line: 5, scope: !14, file: !5, type: !9)
+!14 = distinct !DILexicalBlock(line: 5, column: 0, file: !1, scope: !3)
+!15 = !DILocalVariable(tag: DW_TAG_arg_variable, name: "i", line: 7, arg: 0, scope: !10, file: !5, type: !9)
+!16 = !DIGlobalVariable(name: "abcd", line: 2, isLocal: true, isDefinition: true, scope: !5, file: !5, type: !9)
+!17 = !DIGlobalVariable(name: "xyz", line: 3, isLocal: false, isDefinition: true, scope: !5, file: !5, type: !9, variable: i32* @xyz)
+!18 = !DILocation(line: 6, scope: !19)
+!19 = distinct !DILexicalBlock(line: 6, column: 0, file: !1, scope: !6)
+!20 = !DILocation(line: 7, scope: !10)
+!21 = !DILocation(line: 10, scope: !22)
+!22 = distinct !DILexicalBlock(line: 7, column: 0, file: !1, scope: !10)
 !23 = !{!3, !6, !10}
 !24 = !{!16, !17}
-!25 = !{i32 1, !"Debug Info Version", i32 2}
+!25 = !{i32 1, !"Debug Info Version", i32 3}
diff --git a/test/Transforms/StructurizeCFG/branch-on-argument.ll b/test/Transforms/StructurizeCFG/branch-on-argument.ll
index 4eba0cd8fe15..386994f1fd90 100644
--- a/test/Transforms/StructurizeCFG/branch-on-argument.ll
+++ b/test/Transforms/StructurizeCFG/branch-on-argument.ll
@@ -28,7 +28,7 @@ entry:
 
 for.body:
   %i = phi i32 [0, %entry], [%i.inc, %end.loop]
-  %ptr = getelementptr i32 addrspace(1)* %out, i32 %i
+  %ptr = getelementptr i32, i32 addrspace(1)* %out, i32 %i
   store i32 %i, i32 addrspace(1)* %ptr, align 4
   br i1 %arg, label %mid.loop, label %end.loop
 
diff --git a/test/Transforms/StructurizeCFG/loop-multiple-exits.ll b/test/Transforms/StructurizeCFG/loop-multiple-exits.ll
index 45f3165671e1..40f6be9670a3 100644
--- a/test/Transforms/StructurizeCFG/loop-multiple-exits.ll
+++ b/test/Transforms/StructurizeCFG/loop-multiple-exits.ll
@@ -23,7 +23,7 @@ for.cond:                                         ; preds = %for.inc, %entry
 
 ; CHECK: for.body:
 for.body:                                         ; preds = %for.cond
-  %arrayidx = getelementptr inbounds i32 addrspace(1)* %out, i32 %i.0
+  %arrayidx = getelementptr inbounds i32, i32 addrspace(1)* %out, i32 %i.0
   store i32 %i.0, i32 addrspace(1)* %arrayidx, align 4
   %cmp1 = icmp ugt i32 %i.0, %cond_b
 ; CHECK: br i1 %{{[0-9a-zA-Z_]+}}, label %for.inc, label %[[FLOW1:[0-9a-zA-Z_]+]]
@@ -37,7 +37,7 @@ for.body:                                         ; preds = %for.cond
 
 for.inc:                                          ; preds = %for.body
   %0 = add i32 %cond_a, %i.0
-  %arrayidx3 = getelementptr inbounds i32 addrspace(1)* %out, i32 %0
+  %arrayidx3 = getelementptr inbounds i32, i32 addrspace(1)* %out, i32 %0
   store i32 %i.0, i32 addrspace(1)* %arrayidx3, align 4
   %inc = add i32 %i.0, 1
   br label %for.cond
diff --git a/test/Transforms/StructurizeCFG/nested-loop-order.ll b/test/Transforms/StructurizeCFG/nested-loop-order.ll
new file mode 100644
index 000000000000..fee1ff0433b5
--- /dev/null
+++ b/test/Transforms/StructurizeCFG/nested-loop-order.ll
@@ -0,0 +1,79 @@
+; RUN: opt -S -structurizecfg %s -o - | FileCheck %s
+
+define void @main(float addrspace(1)* %out) {
+
+; CHECK: main_body:
+; CHECK: br label %LOOP.outer
+main_body:
+  br label %LOOP.outer
+
+; CHECK: LOOP.outer:
+; CHECK: br label %LOOP
+LOOP.outer:                                       ; preds = %ENDIF28, %main_body
+  %temp8.0.ph = phi float [ 0.000000e+00, %main_body ], [ %tmp35, %ENDIF28 ]
+  %temp4.0.ph = phi i32 [ 0, %main_body ], [ %tmp20, %ENDIF28 ]
+  br label %LOOP
+
+; CHECK: LOOP:
+; br i1 %{{[0-9]+}}, label %ENDIF, label %Flow
+LOOP:                                             ; preds = %IF29, %LOOP.outer
+  %temp4.0 = phi i32 [ %temp4.0.ph, %LOOP.outer ], [ %tmp20, %IF29 ]
+  %tmp20 = add i32 %temp4.0, 1
+  %tmp22 = icmp sgt i32 %tmp20, 3
+  br i1 %tmp22, label %ENDLOOP, label %ENDIF
+
+; CHECK: Flow3
+; CHECK: br i1 %{{[0-9]+}}, label %ENDLOOP, label %LOOP.outer
+
+; CHECK: ENDLOOP:
+; CHECK: ret void
+ENDLOOP:                                          ; preds = %ENDIF28, %IF29, %LOOP
+  %temp8.1 = phi float [ %temp8.0.ph, %LOOP ], [ %temp8.0.ph, %IF29 ], [ %tmp35, %ENDIF28 ]
+  %tmp23 = icmp eq i32 %tmp20, 3
+  %.45 = select i1 %tmp23, float 0.000000e+00, float 1.000000e+00
+  store float %.45, float addrspace(1)* %out
+  ret void
+
+; CHECK: ENDIF:
+; CHECK: br i1 %tmp31, label %IF29, label %Flow1
+ENDIF:                                            ; preds = %LOOP
+  %tmp31 = icmp sgt i32 %tmp20, 1
+  br i1 %tmp31, label %IF29, label %ENDIF28
+
+; CHECK: Flow:
+; CHECK br i1 %{{[0-9]+}}, label %Flow, label %LOOP
+
+; CHECK: IF29:
+; CHECK: br label %Flow1
+IF29:                                             ; preds = %ENDIF
+  %tmp32 = icmp sgt i32 %tmp20, 2
+  br i1 %tmp32, label %ENDLOOP, label %LOOP
+
+; CHECK: Flow1:
+; CHECK: br label %Flow
+
+; CHECK: Flow2:
+; CHECK: br i1 %{{[0-9]+}}, label %ENDIF28, label %Flow3
+
+; CHECK: ENDIF28:
+; CHECK: br label %Flow3
+ENDIF28:                                          ; preds = %ENDIF
+  %tmp35 = fadd float %temp8.0.ph, 1.0
+  %tmp36 = icmp sgt i32 %tmp20, 2
+  br i1 %tmp36, label %ENDLOOP, label %LOOP.outer
+}
+
+; Function Attrs: nounwind readnone
+declare <4 x float> @llvm.SI.vs.load.input(<16 x i8>, i32, i32) #1
+
+; Function Attrs: readnone
+declare float @llvm.AMDIL.clamp.(float, float, float) #2
+
+declare void @llvm.SI.export(i32, i32, i32, i32, i32, float, float, float, float)
+
+attributes #0 = { "ShaderType"="1" "enable-no-nans-fp-math"="true" "unsafe-fp-math"="true" }
+attributes #1 = { nounwind readnone }
+attributes #2 = { readnone }
+
+!0 = !{!1, !1, i64 0, i32 1}
+!1 = !{!"const", null}
diff --git a/test/Transforms/StructurizeCFG/one-loop-multiple-backedges.ll b/test/Transforms/StructurizeCFG/one-loop-multiple-backedges.ll
index 8ebd47aefd15..668a1e99d814 100644
--- a/test/Transforms/StructurizeCFG/one-loop-multiple-backedges.ll
+++ b/test/Transforms/StructurizeCFG/one-loop-multiple-backedges.ll
@@ -11,28 +11,29 @@ bb:
 bb3:                                              ; preds = %bb7, %bb
   %tmp = phi i64 [ 0, %bb ], [ %tmp8, %bb7 ]
   %tmp4 = fcmp ult float %arg1, 3.500000e+00
-; CHECK: br i1 %tmp4, label %bb7, label %Flow
+; CHECK: %0 = xor i1 %tmp4, true
+; CHECK: br i1 %0, label %bb5, label %Flow
   br i1 %tmp4, label %bb7, label %bb5
 
-; CHECK: Flow:
-; CHECK: br i1 %2, label %Flow1, label %bb3
-
-; CHECK: Flow1:
-; CHECK: br i1 %3, label %bb5, label %bb10
-
 ; CHECK: bb5:
 bb5:                                              ; preds = %bb3
   %tmp6 = fcmp olt float 0.000000e+00, %arg2
-; CHECK: br label %bb10
+; CHECK: br label %Flow
   br i1 %tmp6, label %bb10, label %bb7
 
+; CHECK: Flow:
+; CHECK: br i1 %3, label %bb7, label %Flow1
+
 ; CHECK: bb7
 bb7:                                              ; preds = %bb5, %bb3
   %tmp8 = add nuw nsw i64 %tmp, 1
   %tmp9 = icmp slt i64 %tmp8, 5
-; CHECK: br label %Flow
+; CHECK: br label %Flow1
   br i1 %tmp9, label %bb3, label %bb10
 
+; CHECK: Flow1:
+; CHECK: br i1 %7, label %bb10, label %bb3
+
 ; CHECK: bb10
 bb10:                                             ; preds = %bb7, %bb5
   %tmp11 = phi i32 [ 15, %bb5 ], [ 255, %bb7 ]
diff --git a/test/Transforms/StructurizeCFG/post-order-traversal-bug.ll b/test/Transforms/StructurizeCFG/post-order-traversal-bug.ll
new file mode 100644
index 000000000000..ba9aa2913061
--- /dev/null
+++ b/test/Transforms/StructurizeCFG/post-order-traversal-bug.ll
@@ -0,0 +1,100 @@
+; RUN: opt -S -structurizecfg %s -o - | FileCheck %s
+
+; The structurize cfg pass used to do a post-order traversal to generate a list
+; of ; basic blocks and then operate on the list in reverse.  This led to bugs,
+; because sometimes successors would be visited before their predecessors.
+; The fix for this was to do a reverse post-order traversal which is what the
+; algorithm requires.
+
+; Function Attrs: nounwind
+define void @test(float* nocapture %out, i32 %K1, float* nocapture readonly %nr) {
+
+; CHECK: entry:
+; CHECK: br label %for.body
+entry:
+  br label %for.body
+
+; CHECK: for.body:
+; CHECK: br i1 %{{[0-9]+}}, label %lor.lhs.false, label %Flow
+for.body:                                         ; preds = %for.body.backedge, %entry
+  %indvars.iv = phi i64 [ %indvars.iv.be, %for.body.backedge ], [ 1, %entry ]
+  %best_val.027 = phi float [ %best_val.027.be, %for.body.backedge ], [ 5.000000e+01, %entry ]
+  %prev_start.026 = phi i32 [ %tmp26, %for.body.backedge ], [ 0, %entry ]
+  %best_count.025 = phi i32 [ %best_count.025.be, %for.body.backedge ], [ 0, %entry ]
+  %tmp0 = trunc i64 %indvars.iv to i32
+  %cmp1 = icmp eq i32 %tmp0, %K1
+  br i1 %cmp1, label %if.then, label %lor.lhs.false
+
+; CHECK: lor.lhs.false:
+; CHECK: br label %Flow
+lor.lhs.false:                                    ; preds = %for.body
+  %arrayidx = getelementptr inbounds float, float* %nr, i64 %indvars.iv
+  %tmp1 = load float, float* %arrayidx, align 4
+  %tmp2 = add nsw i64 %indvars.iv, -1
+  %arrayidx2 = getelementptr inbounds float, float* %nr, i64 %tmp2
+  %tmp3 = load float, float* %arrayidx2, align 4
+  %cmp3 = fcmp une float %tmp1, %tmp3
+  br i1 %cmp3, label %if.then, label %for.body.1
+
+; CHECK: Flow:
+; CHECK: br i1 %{{[0-9]+}}, label %if.then, label %Flow1
+
+; CHECK: if.then:
+; CHECK: br label %Flow1
+if.then:                                          ; preds = %lor.lhs.false, %for.body
+  %sub4 = sub nsw i32 %tmp0, %prev_start.026
+  %tmp4 = add nsw i64 %indvars.iv, -1
+  %arrayidx8 = getelementptr inbounds float, float* %nr, i64 %tmp4
+  %tmp5 = load float, float* %arrayidx8, align 4
+  br i1 %cmp1, label %for.end, label %for.body.1
+
+; CHECK: for.end:
+; CHECK: ret void
+for.end:                                          ; preds = %for.body.1, %if.then
+  %best_val.0.lcssa = phi float [ %best_val.233, %for.body.1 ], [ %tmp5, %if.then ]
+  store float %best_val.0.lcssa, float* %out, align 4
+  ret void
+
+; CHECK: Flow1
+; CHECK: br i1 %{{[0-9]}}, label %for.body.1, label %Flow2
+
+; CHECK: for.body.1:
+; CHECK: br i1 %{{[0-9]+}}, label %for.body.6, label %Flow3
+for.body.1:                                       ; preds = %if.then, %lor.lhs.false
+  %best_val.233 = phi float [ %tmp5, %if.then ], [ %best_val.027, %lor.lhs.false ]
+  %best_count.231 = phi i32 [ %sub4, %if.then ], [ %best_count.025, %lor.lhs.false ]
+  %indvars.iv.next.454 = add nsw i64 %indvars.iv, 5
+  %tmp22 = trunc i64 %indvars.iv.next.454 to i32
+  %cmp1.5 = icmp eq i32 %tmp22, %K1
+  br i1 %cmp1.5, label %for.end, label %for.body.6
+
+; CHECK: Flow2:
+; CHECK: br i1 %{{[0-9]+}}, label %for.end, label %for.body
+
+; CHECK: for.body.6:
+; CHECK: br i1 %cmp5.6, label %if.then6.6, label %for.body.backedge
+for.body.6:                                       ; preds = %for.body.1
+  %indvars.iv.next.559 = add nsw i64 %indvars.iv, 6
+  %tmp26 = trunc i64 %indvars.iv.next.559 to i32
+  %sub4.6 = sub nsw i32 %tmp26, %tmp22
+  %cmp5.6 = icmp slt i32 %best_count.231, %sub4.6
+  br i1 %cmp5.6, label %if.then6.6, label %for.body.backedge
+
+; CHECK: if.then6.6
+; CHECK: br label %for.body.backedge
+if.then6.6:                                       ; preds = %for.body.6
+  %arrayidx8.6 = getelementptr inbounds float, float* %nr, i64 %indvars.iv.next.454
+  %tmp29 = load float, float* %arrayidx8.6, align 4
+  br label %for.body.backedge
+
+; CHECK: Flow3:
+; CHECK: br label %Flow2
+
+; CHECK: for.body.backedge:
+; CHECK: br label %Flow3
+for.body.backedge:                                ; preds = %if.then6.6, %for.body.6
+  %best_val.027.be = phi float [ %tmp29, %if.then6.6 ], [ %best_val.233, %for.body.6 ]
+  %best_count.025.be = phi i32 [ %sub4.6, %if.then6.6 ], [ %best_count.231, %for.body.6 ]
+  %indvars.iv.be = add nsw i64 %indvars.iv, 7
+  br label %for.body
+}
diff --git a/test/Transforms/TailCallElim/basic.ll b/test/Transforms/TailCallElim/basic.ll
index 8e9814b52bbc..2488b552d8f3 100644
--- a/test/Transforms/TailCallElim/basic.ll
+++ b/test/Transforms/TailCallElim/basic.ll
@@ -183,7 +183,7 @@ define void @test11() {
   %a = alloca i8*
   %b = alloca i8
   call void @test11_helper1(i8** %a, i8* %b)  ; a = &b
-  %c = load i8** %a
+  %c = load i8*, i8** %a
   call void @test11_helper2(i8* %c)
 ; CHECK: call void @test11_helper2
   ret void
diff --git a/test/Transforms/TailCallElim/dont_reorder_load.ll b/test/Transforms/TailCallElim/dont_reorder_load.ll
index a29b72e94239..ac399a1bf5a9 100644
--- a/test/Transforms/TailCallElim/dont_reorder_load.ll
+++ b/test/Transforms/TailCallElim/dont_reorder_load.ll
@@ -21,7 +21,7 @@ if:		; preds = %entry
 else:		; preds = %entry
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @no_tailrecelim_1(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load i32* @extern_weak_global		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* @extern_weak_global		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
@@ -40,7 +40,7 @@ if:		; preds = %entry
 else:		; preds = %entry
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @no_tailrecelim_2(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load i32* %a_arg		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* %a_arg		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
@@ -58,7 +58,7 @@ if:		; preds = %entry
 else:		; preds = %entry
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @no_tailrecelim_3(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load volatile i32* %a_arg		; <i32> [#uses=1]
+	%tmp9 = load volatile i32, i32* %a_arg		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
diff --git a/test/Transforms/TailCallElim/inf-recursion.ll b/test/Transforms/TailCallElim/inf-recursion.ll
index 157226f93d3f..c121c25aee98 100644
--- a/test/Transforms/TailCallElim/inf-recursion.ll
+++ b/test/Transforms/TailCallElim/inf-recursion.ll
@@ -31,3 +31,24 @@ define float @fabsf(float %f) {
 }
 
 declare x86_fp80 @fabsl(x86_fp80 %f)
+
+; Don't crash while transforming a function with infinite recursion.
+define i32 @PR22704(i1 %bool) {
+entry:
+  br i1 %bool, label %t, label %f
+
+t:
+  %call1 = call i32 @PR22704(i1 1)
+  br label %return
+
+f:
+  %call = call i32 @PR22704(i1 1)
+  br label %return
+
+return:
+  ret i32 0
+
+; CHECK-LABEL: @PR22704(
+; CHECK:       %bool.tr = phi i1 [ %bool, %entry ], [ true, %t ], [ true, %f ]
+; CHECK:       br i1 %bool.tr, label %t, label %f
+}
diff --git a/test/Transforms/TailCallElim/reorder_load.ll b/test/Transforms/TailCallElim/reorder_load.ll
index 2e350d662a39..b989bbf9547a 100644
--- a/test/Transforms/TailCallElim/reorder_load.ll
+++ b/test/Transforms/TailCallElim/reorder_load.ll
@@ -16,7 +16,7 @@ target datalayout = "e-m:e-i64:64-f80:128-n8:16:32:64-S128"
 define fastcc i32 @raise_load_1(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) nounwind readonly {
 ; CHECK-LABEL: @raise_load_1(
 ; CHECK-NOT: call
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NOT: call
 ; CHECK: }
 entry:
@@ -29,7 +29,7 @@ if:		; preds = %entry
 else:		; preds = %entry
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @raise_load_1(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load i32* %a_arg		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* %a_arg		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
@@ -40,7 +40,7 @@ else:		; preds = %entry
 define fastcc i32 @raise_load_2(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) readonly {
 ; CHECK-LABEL: @raise_load_2(
 ; CHECK-NOT: call
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NOT: call
 ; CHECK: }
 entry:
@@ -60,7 +60,7 @@ unwind:		; preds = %else
 recurse:		; preds = %else
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @raise_load_2(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load i32* @global		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* @global		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
@@ -71,7 +71,7 @@ recurse:		; preds = %else
 define fastcc i32 @raise_load_3(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) nounwind readonly {
 ; CHECK-LABEL: @raise_load_3(
 ; CHECK-NOT: call
-; CHECK: load i32*
+; CHECK: load i32, i32*
 ; CHECK-NOT: call
 ; CHECK: }
 entry:
@@ -84,7 +84,7 @@ if:		; preds = %entry
 else:		; preds = %entry
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @raise_load_3(i32* %a_arg, i32 %a_len_arg, i32 %tmp7)		; <i32> [#uses=1]
-	%tmp9 = load i32* @extern_weak_global		; <i32> [#uses=1]
+	%tmp9 = load i32, i32* @extern_weak_global		; <i32> [#uses=1]
 	%tmp10 = add i32 %tmp9, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
@@ -96,8 +96,8 @@ else:		; preds = %entry
 define fastcc i32 @raise_load_4(i32* %a_arg, i32 %a_len_arg, i32 %start_arg) readonly {
 ; CHECK-LABEL: @raise_load_4(
 ; CHECK-NOT: call
-; CHECK: load i32*
-; CHECK-NEXT: load i32*
+; CHECK: load i32, i32*
+; CHECK-NEXT: load i32, i32*
 ; CHECK-NOT: call
 ; CHECK: }
 entry:
@@ -116,9 +116,9 @@ unwind:		; preds = %else
 
 recurse:		; preds = %else
 	%tmp7 = add i32 %start_arg, 1		; <i32> [#uses=1]
-	%first = load i32* %a_arg		; <i32> [#uses=1]
+	%first = load i32, i32* %a_arg		; <i32> [#uses=1]
 	%tmp8 = call fastcc i32 @raise_load_4(i32* %a_arg, i32 %first, i32 %tmp7)		; <i32> [#uses=1]
-	%second = load i32* %a_arg		; <i32> [#uses=1]
+	%second = load i32, i32* %a_arg		; <i32> [#uses=1]
 	%tmp10 = add i32 %second, %tmp8		; <i32> [#uses=1]
 	ret i32 %tmp10
 }
diff --git a/test/Transforms/TailDup/2008-06-11-AvoidDupLoopHeader.ll b/test/Transforms/TailDup/2008-06-11-AvoidDupLoopHeader.ll
index 292186020f4f..4f7a3ca8ce6a 100644
--- a/test/Transforms/TailDup/2008-06-11-AvoidDupLoopHeader.ll
+++ b/test/Transforms/TailDup/2008-06-11-AvoidDupLoopHeader.ll
@@ -11,7 +11,7 @@ bb:		; preds = %entry
 	br label %bb9
 
 bb5:		; preds = %bb9
-	%tmp7 = call i32 (...)* @bar( i32 %x.0 ) nounwind 		; <i32> [#uses=1]
+	%tmp7 = call i32 (...) @bar( i32 %x.0 ) nounwind 		; <i32> [#uses=1]
 	br label %bb9
 
 bb9:		; preds = %bb5, %bb, %entry
diff --git a/test/Transforms/Util/lowerswitch.ll b/test/Transforms/Util/lowerswitch.ll
index 17c12028705f..6e21f916c038 100644
--- a/test/Transforms/Util/lowerswitch.ll
+++ b/test/Transforms/Util/lowerswitch.ll
@@ -3,7 +3,7 @@
 ; Test that we don't crash and have a different basic block for each incoming edge.
 define void @test0() {
 ; CHECK-LABEL: @test0
-; CHECK: %merge = phi i64 [ 1, %BB3 ], [ 0, %NewDefault ], [ 0, %NodeBlock5 ], [ 0, %LeafBlock1 ]
+; CHECK: %merge = phi i64 [ 1, %BB3 ], [ 0, %NewDefault ], [ 0, %NodeBlock.5 ], [ 0, %LeafBlock.1 ]
 BB1:
   switch i32 undef, label %BB2 [
     i32 3, label %BB2
@@ -43,12 +43,146 @@ bb2:
 
 bb3:
 ; CHECK-LABEL: bb3
-; CHECK: %tmp = phi i32 [ 1, %NodeBlock ], [ 0, %bb2 ], [ 1, %LeafBlock3 ]
+; CHECK: %tmp = phi i32 [ 1, %NodeBlock ], [ 0, %bb2 ], [ 1, %LeafBlock.3 ]
   %tmp = phi i32 [ 1, %bb1 ], [ 0, %bb2 ], [ 1, %bb1 ], [ 1, %bb1 ]
-; CHECK-NEXT: %tmp2 = phi i32 [ 2, %NodeBlock ], [ 5, %bb2 ], [ 2, %LeafBlock3 ]
+; CHECK-NEXT: %tmp2 = phi i32 [ 2, %NodeBlock ], [ 5, %bb2 ], [ 2, %LeafBlock.3 ]
   %tmp2 = phi i32 [ 2, %bb1 ], [ 2, %bb1 ], [ 5, %bb2 ], [ 2, %bb1 ]
   br label %exit
 
 exit:
   ret void
 }
+
+; Test that we don't crash.
+define void @test2(i32 %mode) {
+; CHECK-LABEL: @test2
+  br i1 undef, label %1, label %._crit_edge
+
+; <label>:1                                       ; preds = %0
+  switch i32 %mode, label %33 [
+    i32 2, label %2
+    i32 3, label %3
+    i32 4, label %4
+    i32 5, label %5
+    i32 6, label %6
+    i32 7, label %7
+    i32 8, label %8
+    i32 9, label %9
+    i32 10, label %10
+    i32 11, label %14
+    i32 12, label %18
+    i32 13, label %22
+    i32 14, label %26
+    i32 15, label %27
+    i32 16, label %34
+    i32 17, label %34
+    i32 18, label %34
+    i32 19, label %34
+    i32 22, label %34
+    i32 20, label %31
+    i32 21, label %32
+  ]
+
+; <label>:2                                       ; preds = %1
+  br label %34
+
+; <label>:3                                       ; preds = %1
+  br label %34
+
+; <label>:4                                       ; preds = %1
+  br label %34
+
+; <label>:5                                       ; preds = %1
+  br label %34
+
+; <label>:6                                       ; preds = %1
+  br label %34
+
+; <label>:7                                       ; preds = %1
+  br label %34
+
+; <label>:8                                       ; preds = %1
+  br label %34
+
+; <label>:9                                       ; preds = %1
+  br label %34
+
+; <label>:10                                      ; preds = %1
+  br i1 undef, label %11, label %12
+
+; <label>:11                                      ; preds = %10
+  br label %13
+
+; <label>:12                                      ; preds = %10
+  br label %13
+
+; <label>:13                                      ; preds = %12, %11
+  br label %34
+
+; <label>:14                                      ; preds = %1
+  br i1 undef, label %15, label %16
+
+; <label>:15                                      ; preds = %14
+  br label %17
+
+; <label>:16                                      ; preds = %14
+  br label %17
+
+; <label>:17                                      ; preds = %16, %15
+  br label %34
+
+; <label>:18                                      ; preds = %1
+  br i1 undef, label %19, label %20
+
+; <label>:19                                      ; preds = %18
+  br label %21
+
+; <label>:20                                      ; preds = %18
+  br label %21
+
+; <label>:21                                      ; preds = %20, %19
+  br label %34
+
+; <label>:22                                      ; preds = %1
+  br i1 undef, label %23, label %24
+
+; <label>:23                                      ; preds = %22
+  br label %25
+
+; <label>:24                                      ; preds = %22
+  br label %25
+
+; <label>:25                                      ; preds = %24, %23
+  br label %34
+
+; <label>:26                                      ; preds = %1
+  br label %34
+
+; <label>:27                                      ; preds = %1
+  br i1 undef, label %28, label %29
+
+; <label>:28                                      ; preds = %27
+  br label %30
+
+; <label>:29                                      ; preds = %27
+  br label %30
+
+; <label>:30                                      ; preds = %29, %28
+  br label %34
+
+; <label>:31                                      ; preds = %1
+  br label %34
+
+; <label>:32                                      ; preds = %1
+  br label %34
+
+; <label>:33                                      ; preds = %1
+  br label %34
+
+; <label>:34                                      ; preds = %33, %32, %31, %30, %26, %25, %21, %17, %13, %9, %8, %7, %6, %5, %4, %3, %2, %1, %1, %1, %1, %1
+  %o.0 = phi float [ undef, %33 ], [ undef, %32 ], [ undef, %31 ], [ undef, %30 ], [ undef, %26 ], [ undef, %25 ], [ undef, %21 ], [ undef, %17 ], [ undef, %13 ], [ undef, %9 ], [ undef, %8 ], [ undef, %7 ], [ undef, %6 ], [ undef, %5 ], [ undef, %4 ], [ undef, %3 ], [ undef, %2 ], [ undef, %1 ], [ undef, %1 ], [ undef, %1 ], [ undef, %1 ], [ undef, %1 ]
+  br label %._crit_edge
+
+._crit_edge:                                      ; preds = %34, %0
+  ret void
+}