vendor/llvm/llvm-release_900-r372316vendor/llvm-90

5 files changed, 52 insertions, 21 deletions

diff --git a/include/llvm/IR/Instructions.h b/include/llvm/IR/Instructions.h
index 215ce45c7b75..6773664104a3 100644
--- a/include/llvm/IR/Instructions.h
+++ b/include/llvm/IR/Instructions.h
@@ -3938,6 +3938,9 @@ class CallBrInst : public CallBase {
             ArrayRef<BasicBlock *> IndirectDests, ArrayRef<Value *> Args,
             ArrayRef<OperandBundleDef> Bundles, const Twine &NameStr);
 
+  /// Should the Indirect Destinations change, scan + update the Arg list.
+  void updateArgBlockAddresses(unsigned i, BasicBlock *B);
+
   /// Compute the number of operands to allocate.
   static int ComputeNumOperands(int NumArgs, int NumIndirectDests,
                                 int NumBundleInputs = 0) {
@@ -4075,7 +4078,7 @@ public:
     return cast<BasicBlock>(*(&Op<-1>() - getNumIndirectDests() - 1));
   }
   BasicBlock *getIndirectDest(unsigned i) const {
-    return cast<BasicBlock>(*(&Op<-1>() - getNumIndirectDests() + i));
+    return cast_or_null<BasicBlock>(*(&Op<-1>() - getNumIndirectDests() + i));
   }
   SmallVector<BasicBlock *, 16> getIndirectDests() const {
     SmallVector<BasicBlock *, 16> IndirectDests;
@@ -4087,6 +4090,7 @@ public:
     *(&Op<-1>() - getNumIndirectDests() - 1) = reinterpret_cast<Value *>(B);
   }
   void setIndirectDest(unsigned i, BasicBlock *B) {
+    updateArgBlockAddresses(i, B);
     *(&Op<-1>() - getNumIndirectDests() + i) = reinterpret_cast<Value *>(B);
   }
 
@@ -4096,11 +4100,10 @@ public:
     return i == 0 ? getDefaultDest() : getIndirectDest(i - 1);
   }
 
-  void setSuccessor(unsigned idx, BasicBlock *NewSucc) {
-    assert(idx < getNumIndirectDests() + 1 &&
+  void setSuccessor(unsigned i, BasicBlock *NewSucc) {
+    assert(i < getNumIndirectDests() + 1 &&
            "Successor # out of range for callbr!");
-    *(&Op<-1>() - getNumIndirectDests() -1 + idx) =
-        reinterpret_cast<Value *>(NewSucc);
+    return i == 0 ? setDefaultDest(NewSucc) : setIndirectDest(i - 1, NewSucc);
   }
 
   unsigned getNumSuccessors() const { return getNumIndirectDests() + 1; }
diff --git a/lib/CodeGen/IfConversion.cpp b/lib/CodeGen/IfConversion.cpp
index b17a253fe23f..af556e2c856e 100644
--- a/lib/CodeGen/IfConversion.cpp
+++ b/lib/CodeGen/IfConversion.cpp
@@ -912,6 +912,12 @@ void IfConverter::AnalyzeBranches(BBInfo &BBI) {
   BBI.BrCond.clear();
   BBI.IsBrAnalyzable =
       !TII->analyzeBranch(*BBI.BB, BBI.TrueBB, BBI.FalseBB, BBI.BrCond);
+  if (!BBI.IsBrAnalyzable) {
+    BBI.TrueBB = nullptr;
+    BBI.FalseBB = nullptr;
+    BBI.BrCond.clear();
+  }
+
   SmallVector<MachineOperand, 4> RevCond(BBI.BrCond.begin(), BBI.BrCond.end());
   BBI.IsBrReversible = (RevCond.size() == 0) ||
       !TII->reverseBranchCondition(RevCond);
@@ -1758,9 +1764,15 @@ bool IfConverter::IfConvertDiamondCommon(
   if (!BBI1->IsBrAnalyzable)
     verifySameBranchInstructions(&MBB1, &MBB2);
 #endif
-  BBI1->NonPredSize -= TII->removeBranch(*BBI1->BB);
-  // Remove duplicated instructions.
+  // Remove duplicated instructions from the tail of MBB1: any branch
+  // instructions, and the common instructions counted by NumDups2.
   DI1 = MBB1.end();
+  while (DI1 != MBB1.begin()) {
+    MachineBasicBlock::iterator Prev = std::prev(DI1);
+    if (!Prev->isBranch() && !Prev->isDebugInstr())
+      break;
+    DI1 = Prev;
+  }
   for (unsigned i = 0; i != NumDups2; ) {
     // NumDups2 only counted non-dbg_value instructions, so this won't
     // run off the head of the list.
diff --git a/lib/IR/Instructions.cpp b/lib/IR/Instructions.cpp
index 2e7cad103c12..90cb2c26e082 100644
--- a/lib/IR/Instructions.cpp
+++ b/lib/IR/Instructions.cpp
@@ -822,6 +822,17 @@ void CallBrInst::init(FunctionType *FTy, Value *Fn, BasicBlock *Fallthrough,
   setName(NameStr);
 }
 
+void CallBrInst::updateArgBlockAddresses(unsigned i, BasicBlock *B) {
+  assert(getNumIndirectDests() > i && "IndirectDest # out of range for callbr");
+  if (BasicBlock *OldBB = getIndirectDest(i)) {
+    BlockAddress *Old = BlockAddress::get(OldBB);
+    BlockAddress *New = BlockAddress::get(B);
+    for (unsigned ArgNo = 0, e = getNumArgOperands(); ArgNo != e; ++ArgNo)
+      if (dyn_cast<BlockAddress>(getArgOperand(ArgNo)) == Old)
+        setArgOperand(ArgNo, New);
+  }
+}
+
 CallBrInst::CallBrInst(const CallBrInst &CBI)
     : CallBase(CBI.Attrs, CBI.FTy, CBI.getType(), Instruction::CallBr,
                OperandTraits<CallBase>::op_end(this) - CBI.getNumOperands(),
diff --git a/lib/Target/X86/X86ISelLowering.cpp b/lib/Target/X86/X86ISelLowering.cpp
index 87ff5a719fca..0c5b8a79dd62 100644
--- a/lib/Target/X86/X86ISelLowering.cpp
+++ b/lib/Target/X86/X86ISelLowering.cpp
@@ -7650,17 +7650,22 @@ static SDValue EltsFromConsecutiveLoads(EVT VT, ArrayRef<SDValue> Elts,
     // IsConsecutiveLoadWithZeros - we need to create a shuffle of the loaded
     // vector and a zero vector to clear out the zero elements.
     if (!isAfterLegalize && VT.isVector()) {
-      SmallVector<int, 4> ClearMask(NumElems, -1);
-      for (unsigned i = 0; i < NumElems; ++i) {
-        if (ZeroMask[i])
-          ClearMask[i] = i + NumElems;
-        else if (LoadMask[i])
-          ClearMask[i] = i;
+      unsigned NumMaskElts = VT.getVectorNumElements();
+      if ((NumMaskElts % NumElems) == 0) {
+        unsigned Scale = NumMaskElts / NumElems;
+        SmallVector<int, 4> ClearMask(NumMaskElts, -1);
+        for (unsigned i = 0; i < NumElems; ++i) {
+          if (UndefMask[i])
+            continue;
+          int Offset = ZeroMask[i] ? NumMaskElts : 0;
+          for (unsigned j = 0; j != Scale; ++j)
+            ClearMask[(i * Scale) + j] = (i * Scale) + j + Offset;
+        }
+        SDValue V = CreateLoad(VT, LDBase);
+        SDValue Z = VT.isInteger() ? DAG.getConstant(0, DL, VT)
+                                   : DAG.getConstantFP(0.0, DL, VT);
+        return DAG.getVectorShuffle(VT, DL, V, Z, ClearMask);
       }
-      SDValue V = CreateLoad(VT, LDBase);
-      SDValue Z = VT.isInteger() ? DAG.getConstant(0, DL, VT)
-                                 : DAG.getConstantFP(0.0, DL, VT);
-      return DAG.getVectorShuffle(VT, DL, V, Z, ClearMask);
     }
   }
 
@@ -31664,8 +31669,8 @@ static bool matchUnaryPermuteShuffle(MVT MaskVT, ArrayRef<int> Mask,
   if (!ContainsZeros && AllowIntDomain && MaskScalarSizeInBits == 16) {
     SmallVector<int, 4> RepeatedMask;
     if (is128BitLaneRepeatedShuffleMask(MaskEltVT, Mask, RepeatedMask)) {
-      ArrayRef<int> LoMask(Mask.data() + 0, 4);
-      ArrayRef<int> HiMask(Mask.data() + 4, 4);
+      ArrayRef<int> LoMask(RepeatedMask.data() + 0, 4);
+      ArrayRef<int> HiMask(RepeatedMask.data() + 4, 4);
 
       // PSHUFLW: permute lower 4 elements only.
       if (isUndefOrInRange(LoMask, 0, 4) &&
diff --git a/lib/Transforms/Utils/SimplifyLibCalls.cpp b/lib/Transforms/Utils/SimplifyLibCalls.cpp
index e0def81d5eee..e938ae6cb42f 100644
--- a/lib/Transforms/Utils/SimplifyLibCalls.cpp
+++ b/lib/Transforms/Utils/SimplifyLibCalls.cpp
@@ -1480,8 +1480,8 @@ Value *LibCallSimplifier::optimizePow(CallInst *Pow, IRBuilder<> &B) {
   if (match(Expo, m_SpecificFP(-1.0)))
     return B.CreateFDiv(ConstantFP::get(Ty, 1.0), Base, "reciprocal");
 
-  // pow(x, 0.0) -> 1.0
-  if (match(Expo, m_SpecificFP(0.0)))
+  // pow(x, +/-0.0) -> 1.0
+  if (match(Expo, m_AnyZeroFP()))
     return ConstantFP::get(Ty, 1.0);
 
   // pow(x, 1.0) -> x