summaryrefslogtreecommitdiff
path: root/test/OpenMP/target_teams_distribute_codegen.cpp
diff options
context:
space:
mode:
Diffstat (limited to 'test/OpenMP/target_teams_distribute_codegen.cpp')
-rw-r--r--test/OpenMP/target_teams_distribute_codegen.cpp253
1 files changed, 148 insertions, 105 deletions
diff --git a/test/OpenMP/target_teams_distribute_codegen.cpp b/test/OpenMP/target_teams_distribute_codegen.cpp
index 17baa603d12be..f9500a73ef795 100644
--- a/test/OpenMP/target_teams_distribute_codegen.cpp
+++ b/test/OpenMP/target_teams_distribute_codegen.cpp
@@ -1,28 +1,46 @@
// Test host codegen.
-// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix CHECK --check-prefix CHECK-64
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
-// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
-// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix CHECK --check-prefix CHECK-64
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix CHECK --check-prefix CHECK-32
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
-// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix CHECK --check-prefix CHECK-32
+
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY0 %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY0 %s
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY0 %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY0 %s
+// SIMD-ONLY0-NOT: {{__kmpc|__tgt}}
// Test target codegen - host bc file has to be created first.
// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
-// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix TCHECK --check-prefix TCHECK-64
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix TCHECK --check-prefix TCHECK-64
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
-// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix TCHECK --check-prefix TCHECK-64
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix TCHECK --check-prefix TCHECK-64
// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
-// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --check-prefix TCHECK --check-prefix TCHECK-32
+// RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix TCHECK --check-prefix TCHECK-32
// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
-// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck %s --check-prefix TCHECK --check-prefix TCHECK-32
+// RUN: %clang_cc1 -fopenmp -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap %s --check-prefix TCHECK --check-prefix TCHECK-32
+
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY1 %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o %t %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY1 %s
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc
+// RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY1 %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o %t %s
+// RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=45 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -include-pch %t -verify %s -emit-llvm -o - | FileCheck -allow-deprecated-dag-overlap --check-prefix SIMD-ONLY1 %s
+// SIMD-ONLY1-NOT: {{__kmpc|__tgt}}
// expected-no-diagnostics
#ifndef HEADER
#define HEADER
-// CHECK-DAG: %ident_t = type { i32, i32, i32, i32, i8* }
+// CHECK-DAG: %struct.ident_t = type { i32, i32, i32, i32, i8* }
// CHECK-DAG: [[STR:@.+]] = private unnamed_addr constant [23 x i8] c";unknown;unknown;0;0;;\00"
-// CHECK-DAG: [[DEF_LOC:@.+]] = private unnamed_addr constant %ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
+// CHECK-DAG: [[DEF_LOC:@.+]] = private unnamed_addr constant %struct.ident_t { i32 0, i32 2, i32 0, i32 0, i8* getelementptr inbounds ([23 x i8], [23 x i8]* [[STR]], i32 0, i32 0) }
// CHECK-DAG: [[TT:%.+]] = type { i64, i8 }
// CHECK-DAG: [[S1:%.+]] = type { double }
@@ -39,33 +57,33 @@
// sizes.
// CHECK-DAG: [[SIZET:@.+]] = private unnamed_addr constant [3 x i[[SZ]]] [i[[SZ]] 2, i[[SZ]] 4, i[[SZ]] 4]
-// CHECK-DAG: [[MAPT:@.+]] = private unnamed_addr constant [3 x i64] [i64 288, i64 288, i64 288]
+// CHECK-DAG: [[MAPT:@.+]] = private unnamed_addr constant [3 x i64] [i64 800, i64 800, i64 800]
// CHECK-DAG: [[SIZET2:@.+]] = private unnamed_addr constant [1 x i[[SZ]]] [i[[SZ]] 2]
-// CHECK-DAG: [[MAPT2:@.+]] = private unnamed_addr constant [1 x i64] [i64 288]
+// CHECK-DAG: [[MAPT2:@.+]] = private unnamed_addr constant [1 x i64] [i64 800]
// CHECK-DAG: [[SIZET3:@.+]] = private unnamed_addr constant [2 x i[[SZ]]] [i[[SZ]] 4, i[[SZ]] 2]
-// CHECK-DAG: [[MAPT3:@.+]] = private unnamed_addr constant [2 x i64] [i64 288, i64 288]
-// CHECK-DAG: [[MAPT4:@.+]] = private unnamed_addr constant [9 x i64] [i64 288, i64 547, i64 288, i64 547, i64 547, i64 288, i64 288, i64 547, i64 547]
-// CHECK-DAG: [[MAPT5:@.+]] = private unnamed_addr constant [5 x i64] [i64 547, i64 288, i64 288, i64 288, i64 547]
+// CHECK-DAG: [[MAPT3:@.+]] = private unnamed_addr constant [2 x i64] [i64 800, i64 800]
+// CHECK-DAG: [[MAPT4:@.+]] = private unnamed_addr constant [10 x i64] [i64 800, i64 547, i64 288, i64 547, i64 547, i64 288, i64 288, i64 547, i64 547, i64 800]
+// CHECK-DAG: [[MAPT5:@.+]] = private unnamed_addr constant [6 x i64] [i64 32, i64 281474976711171, i64 800, i64 288, i64 288, i64 547]
// CHECK-DAG: [[SIZET6:@.+]] = private unnamed_addr constant [5 x i[[SZ]]] [i[[SZ]] 4, i[[SZ]] 4, i[[SZ]] 2, i[[SZ]] 1, i[[SZ]] 40]
-// CHECK-DAG: [[MAPT6:@.+]] = private unnamed_addr constant [5 x i64] [i64 288, i64 288, i64 288, i64 288, i64 547]
+// CHECK-DAG: [[MAPT6:@.+]] = private unnamed_addr constant [5 x i64] [i64 800, i64 800, i64 800, i64 800, i64 547]
// CHECK-DAG: [[SIZET7:@.+]] = private unnamed_addr constant [3 x i[[SZ]]] [i[[SZ]] 4, i[[SZ]] 2, i[[SZ]] 40]
-// CHECK-DAG: [[MAPT7:@.+]] = private unnamed_addr constant [3 x i64] [i64 288, i64 288, i64 547]
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-// CHECK-DAG: @{{.*}} = private constant i8 0
-
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK: @{{.+}} = constant [[ENTTY]]
-// TCHECK-NOT: @{{.+}} = constant [[ENTTY]]
+// CHECK-DAG: [[MAPT7:@.+]] = private unnamed_addr constant [3 x i64] [i64 800, i64 800, i64 547]
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+// CHECK-DAG: @{{.*}} = weak constant i8 0
+
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK: @{{.+}} = weak constant [[ENTTY]]
+// TCHECK-NOT: @{{.+}} = weak constant [[ENTTY]]
// Check if offloading descriptor is created.
// CHECK: [[ENTBEGIN:@.+]] = external constant [[ENTTY]]
@@ -76,7 +94,7 @@
// CHECK: [[DESC:@.+]] = internal constant [[DSCTY]] { i32 1, [[DEVTY]]* getelementptr inbounds ([1 x [[DEVTY]]], [1 x [[DEVTY]]]* [[IMAGES]], i32 0, i32 0), [[ENTTY]]* [[ENTBEGIN]], [[ENTTY]]* [[ENTEND]] }, comdat($[[REGFN]])
// Check target registration is registered as a Ctor.
-// CHECK: appending global [1 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 0, void ()* bitcast (void (i8*)* @[[REGFN]] to void ()*), i8* bitcast (void (i8*)* @[[REGFN]] to i8*) }]
+// CHECK: appending global [1 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 0, void ()* @[[REGFN]], i8* bitcast (void ()* @[[REGFN]] to i8*) }]
template<typename tx, typename ty>
@@ -193,6 +211,7 @@ int foo(int n) {
}
// We capture 3 VLA sizes in this target region
+ // CHECK: load i32, i32* %
// CHECK-64: [[A_VAL:%.+]] = load i32, i32* %{{.+}},
// CHECK-64: [[A_ADDR:%.+]] = bitcast i[[SZ]]* [[A_CADDR:%.+]] to i32*
// CHECK-64: store i32 [[A_VAL]], i32* [[A_ADDR]],
@@ -202,45 +221,48 @@ int foo(int n) {
// CHECK-32: store i32 [[A_VAL]], i32* [[A_CADDR:%.+]],
// CHECK-32: [[A_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[A_CADDR]],
+ // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 20
+ // CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]]
+ // CHECK: [[TRY]]
// CHECK: [[BNSIZE:%.+]] = mul nuw i[[SZ]] [[VLA0:%.+]], 4
// CHECK: [[CNELEMSIZE2:%.+]] = mul nuw i[[SZ]] 5, [[VLA1:%.+]]
// CHECK: [[CNSIZE:%.+]] = mul nuw i[[SZ]] [[CNELEMSIZE2]], 8
- // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 20
- // CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]]
- // CHECK: [[TRY]]
- // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams(i64 -1, i8* @{{[^,]+}}, i32 9, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i[[SZ]]* [[SR:%[^,]+]], i64* getelementptr inbounds ([9 x i64], [9 x i64]* [[MAPT4]], i32 0, i32 0), i32 0, i32 0)
- // CHECK-DAG: [[BPR]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP:%[^,]+]], i32 0, i32 0
- // CHECK-DAG: [[PR]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P:%[^,]+]], i32 0, i32 0
- // CHECK-DAG: [[SR]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S:%[^,]+]], i32 0, i32 0
-
- // CHECK-DAG: [[SADDR0:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX0:[0-9]+]]
- // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX0]]
- // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX0]]
- // CHECK-DAG: [[SADDR1:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX1:[0-9]+]]
- // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX1]]
- // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX1]]
- // CHECK-DAG: [[SADDR2:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX2:[0-9]+]]
- // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX2]]
- // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX2]]
- // CHECK-DAG: [[SADDR3:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX3:[0-9]+]]
- // CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX3]]
- // CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX3]]
- // CHECK-DAG: [[SADDR4:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX4:[0-9]+]]
- // CHECK-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX4]]
- // CHECK-DAG: [[PADDR4:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX4]]
- // CHECK-DAG: [[SADDR5:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX5:[0-9]+]]
- // CHECK-DAG: [[BPADDR5:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX5]]
- // CHECK-DAG: [[PADDR5:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX5]]
- // CHECK-DAG: [[SADDR6:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX6:[0-9]+]]
- // CHECK-DAG: [[BPADDR6:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX6]]
- // CHECK-DAG: [[PADDR6:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX6]]
- // CHECK-DAG: [[SADDR7:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX7:[0-9]+]]
- // CHECK-DAG: [[BPADDR7:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX7]]
- // CHECK-DAG: [[PADDR7:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX7]]
- // CHECK-DAG: [[SADDR8:%.+]] = getelementptr inbounds [9 x i[[SZ]]], [9 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX8:[0-9]+]]
- // CHECK-DAG: [[BPADDR8:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX8]]
- // CHECK-DAG: [[PADDR8:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX8]]
+ // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams(i64 -1, i8* @{{[^,]+}}, i32 10, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i[[SZ]]* [[SR:%[^,]+]], i64* getelementptr inbounds ([10 x i64], [10 x i64]* [[MAPT4]], i32 0, i32 0), i32 0, i32 0)
+ // CHECK-DAG: [[BPR]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP:%[^,]+]], i32 0, i32 0
+ // CHECK-DAG: [[PR]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P:%[^,]+]], i32 0, i32 0
+ // CHECK-DAG: [[SR]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S:%[^,]+]], i32 0, i32 0
+
+ // CHECK-DAG: [[SADDR0:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX0:[0-9]+]]
+ // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX0]]
+ // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX0]]
+ // CHECK-DAG: [[SADDR1:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX1:[0-9]+]]
+ // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX1]]
+ // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX1]]
+ // CHECK-DAG: [[SADDR2:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX2:[0-9]+]]
+ // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX2]]
+ // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX2]]
+ // CHECK-DAG: [[SADDR3:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX3:[0-9]+]]
+ // CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX3]]
+ // CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX3]]
+ // CHECK-DAG: [[SADDR4:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX4:[0-9]+]]
+ // CHECK-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX4]]
+ // CHECK-DAG: [[PADDR4:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX4]]
+ // CHECK-DAG: [[SADDR5:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX5:[0-9]+]]
+ // CHECK-DAG: [[BPADDR5:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX5]]
+ // CHECK-DAG: [[PADDR5:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX5]]
+ // CHECK-DAG: [[SADDR6:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX6:[0-9]+]]
+ // CHECK-DAG: [[BPADDR6:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX6]]
+ // CHECK-DAG: [[PADDR6:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX6]]
+ // CHECK-DAG: [[SADDR7:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX7:[0-9]+]]
+ // CHECK-DAG: [[BPADDR7:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX7]]
+ // CHECK-DAG: [[PADDR7:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX7]]
+ // CHECK-DAG: [[SADDR8:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX8:[0-9]+]]
+ // CHECK-DAG: [[BPADDR8:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX8]]
+ // CHECK-DAG: [[PADDR8:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX8]]
+ // CHECK-DAG: [[SADDR9:%.+]] = getelementptr inbounds [10 x i[[SZ]]], [10 x i[[SZ]]]* [[S]], i32 0, i32 [[IDX9:[0-9]+]]
+ // CHECK-DAG: [[BPADDR9:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[BP]], i32 0, i32 [[IDX9]]
+ // CHECK-DAG: [[PADDR9:%.+]] = getelementptr inbounds [10 x i8*], [10 x i8*]* [[P]], i32 0, i32 [[IDX9]]
// The names below are not necessarily consistent with the names used for the
// addresses above as some are repeated.
@@ -298,14 +320,20 @@ int foo(int n) {
// CHECK-DAG: [[CPADDR8]] = bitcast i8** {{%[^,]+}} to [[TT]]**
// CHECK-DAG: store i[[SZ]] {{12|16}}, i[[SZ]]* {{%[^,]+}}
+ // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CBPADDR9:%.+]],
+ // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CPADDR9:%.+]],
+ // CHECK-DAG: [[CBPADDR9]] = bitcast i8** {{%[^,]+}} to i[[SZ]]*
+ // CHECK-DAG: [[CPADDR9]] = bitcast i8** {{%[^,]+}} to i[[SZ]]*
+ // CHECK-DAG: store i[[SZ]] {{4|8}}, i[[SZ]]* {{%[^,]+}}
+
// CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0
// CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:[^,]+]], label %[[END:[^,]+]]
// CHECK: [[FAIL]]
- // CHECK: call void [[HVT4:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}})
+ // CHECK: call void [[HVT4:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}})
// CHECK-NEXT: br label %[[END]]
// CHECK: [[END]]
- #pragma omp target teams distribute if(target: n>20)
+ #pragma omp target teams distribute if(target: n>20) dist_schedule(static, n)
for (int i = 0; i < 10; ++i) {
a += 1;
b[2] += 1.0;
@@ -323,11 +351,12 @@ int foo(int n) {
// correct and loaded correctly for the target regions in foo().
// CHECK: define internal void [[HVT0]](i[[SZ]] {{[^,]+}}, i[[SZ]] {{[^,]+}}, i[[SZ]] {{[^)]+}})
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] {{[^)]+}})
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] {{[^)]+}})
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] {{[^)]+}})
-// CHECK: alloca i16,
+// CHECK: alloca i[[SZ]],
+// CHECK: bitcast i[[SZ]]* {{.+}} to i16*
// CHECK: ret void
// CHECK-NEXT: }
@@ -344,7 +373,7 @@ int foo(int n) {
// CHECK-64: store i32 [[AA]], i32* [[AA_C]], align
// CHECK-32: store i32 [[AA]], i32* [[AA_CASTED]], align
// CHECK: [[PARAM:%.+]] = load i[[SZ]], i[[SZ]]* [[AA_CASTED]], align
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED1:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED1:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED1]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}})
@@ -366,7 +395,7 @@ int foo(int n) {
// CHECK: [[AA_C:%.+]] = bitcast i[[SZ]]* [[AA_CASTED]] to i16*
// CHECK: store i16 [[AA]], i16* [[AA_C]], align
// CHECK: [[PARAM:%.+]] = load i[[SZ]], i[[SZ]]* [[AA_CASTED]], align
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED2:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 1, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]])* [[OMP_OUTLINED2:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED2]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}})
@@ -397,7 +426,7 @@ int foo(int n) {
// CHECK-DAG: store i16 [[AA]], i16* [[AA_C]], align
// CHECK-DAG: [[PARAM1:%.+]] = load i[[SZ]], i[[SZ]]* [[A_CASTED]], align
// CHECK-DAG: [[PARAM2:%.+]] = load i[[SZ]], i[[SZ]]* [[AA_CASTED]], align
-// CHECK-DAG: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]])* [[OMP_OUTLINED3:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM1]], i[[SZ]] [[PARAM2]])
+// CHECK-DAG: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 2, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]])* [[OMP_OUTLINED3:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[PARAM1]], i[[SZ]] [[PARAM2]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED3]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}})
@@ -421,6 +450,7 @@ int foo(int n) {
// CHECK: [[LOCAL_VLA3:%.+]] = alloca i[[SZ]]
// CHECK: [[LOCAL_CN:%.+]] = alloca double*
// CHECK: [[LOCAL_D:%.+]] = alloca [[TT]]*
+// CHECK: alloca i[[SZ]],
// CHECK: [[LOCAL_A_CASTED:%.+]] = alloca i[[SZ]]
// CHECK-DAG: store i[[SZ]] [[ARG_A:%.+]], i[[SZ]]* [[LOCAL_A]]
// CHECK-DAG: store [10 x float]* [[ARG_B:%.+]], [10 x float]** [[LOCAL_B]]
@@ -449,7 +479,7 @@ int foo(int n) {
// CHECK-32-DAG:store i32 [[LOCAL_AV]], i32* [[LOCAL_A_CASTED]], align
// CHECK-DAG: [[REF_A:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_A_CASTED]],
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 9, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], [10 x float]*, i[[SZ]], float*, [5 x [10 x double]]*, i[[SZ]], i[[SZ]], double*, [[TT]]*)* [[OMP_OUTLINED4:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], [10 x float]* [[REF_B]], i[[SZ]] [[VAL_VLA1]], float* [[REF_BN]], [5 x [10 x double]]* [[REF_C]], i[[SZ]] [[VAL_VLA2]], i[[SZ]] [[VAL_VLA3]], double* [[REF_CN]], [[TT]]* [[REF_D]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 10, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], [10 x float]*, i[[SZ]], float*, [5 x [10 x double]]*, i[[SZ]], i[[SZ]], double*, [[TT]]*, i[[SZ]])* [[OMP_OUTLINED4:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], [10 x float]* [[REF_B]], i[[SZ]] [[VAL_VLA1]], float* [[REF_BN]], [5 x [10 x double]]* [[REF_C]], i[[SZ]] [[VAL_VLA2]], i[[SZ]] [[VAL_VLA3]], double* [[REF_CN]], [[TT]]* [[REF_D]], i[[SZ]] %{{.+}})
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED4]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, [10 x float]* {{.+}}, i[[SZ]] %{{.+}}, float* {{.+}}, [5 x [10 x double]]* {{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, double* {{.+}}, [[TT]]* {{.+}})
@@ -534,32 +564,39 @@ int bar(int n){
// CHECK-64: store i32 %{{.+}}, i32* [[B_ADDR]],
// CHECK-64: [[B_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[B_CADDR]],
+// CHECK-32: store i32 %{{.+}}, i32* %__vla_expr
// CHECK-32: store i32 %{{.+}}, i32* [[B_ADDR:%.+]],
// CHECK-32: [[B_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[B_ADDR]],
+// CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 60
+// CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]]
+// CHECK: [[TRY]]
// We capture 2 VLA sizes in this target region
// CHECK: [[CELEMSIZE2:%.+]] = mul nuw i[[SZ]] 2, [[VLA0:%.+]]
// CHECK: [[CSIZE:%.+]] = mul nuw i[[SZ]] [[CELEMSIZE2]], 2
-// CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 60
-// CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]]
-// CHECK: [[TRY]]
-// CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams(i64 -1, i8* @{{[^,]+}}, i32 5, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i[[SZ]]* [[SR:%[^,]+]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* [[MAPT5]], i32 0, i32 0), i32 0, i32 0)
-// CHECK-DAG: [[BPR]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP:%.+]], i32 0, i32 0
-// CHECK-DAG: [[PR]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P:%.+]], i32 0, i32 0
-// CHECK-DAG: [[SR]] = getelementptr inbounds [5 x i[[SZ]]], [5 x i[[SZ]]]* [[S:%.+]], i32 0, i32 0
-// CHECK-DAG: [[SADDR0:%.+]] = getelementptr inbounds [5 x i[[SZ]]], [5 x i[[SZ]]]* [[S]], i32 [[IDX0:[0-9]+]]
-// CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX0]]
-// CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX0]]
-// CHECK-DAG: [[SADDR1:%.+]] = getelementptr inbounds [5 x i[[SZ]]], [5 x i[[SZ]]]* [[S]], i32 [[IDX1:[0-9]+]]
-// CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX1]]
-// CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX1]]
-// CHECK-DAG: [[SADDR2:%.+]] = getelementptr inbounds [5 x i[[SZ]]], [5 x i[[SZ]]]* [[S]], i32 [[IDX2:[0-9]+]]
-// CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX2]]
-// CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX2]]
-// CHECK-DAG: [[SADDR3:%.+]] = getelementptr inbounds [5 x i[[SZ]]], [5 x i[[SZ]]]* [[S]], i32 [[IDX3:[0-9]+]]
-// CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX3]]
-// CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX3]]
+// CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams(i64 -1, i8* @{{[^,]+}}, i32 6, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i[[SZ]]* [[SR:%[^,]+]], i64* getelementptr inbounds ([6 x i64], [6 x i64]* [[MAPT5]], i32 0, i32 0), i32 0, i32 0)
+// CHECK-DAG: [[BPR]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP:%.+]], i32 0, i32 0
+// CHECK-DAG: [[PR]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P:%.+]], i32 0, i32 0
+// CHECK-DAG: [[SR]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S:%.+]], i32 0, i32 0
+// CHECK-DAG: [[SADDR0:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX0:[0-9]+]]
+// CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX0]]
+// CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX0]]
+// CHECK-DAG: [[SADDR1:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX1:[0-9]+]]
+// CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX1]]
+// CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX1]]
+// CHECK-DAG: [[SADDR2:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX2:[0-9]+]]
+// CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX2]]
+// CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX2]]
+// CHECK-DAG: [[SADDR3:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX3:[0-9]+]]
+// CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX3]]
+// CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX3]]
+// CHECK-DAG: [[SADDR4:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX3:[0-9]+]]
+// CHECK-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX3]]
+// CHECK-DAG: [[PADDR4:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX3]]
+// CHECK-DAG: [[SADDR5:%.+]] = getelementptr inbounds [6 x i[[SZ]]], [6 x i[[SZ]]]* [[S]], i32 [[IDX3:[0-9]+]]
+// CHECK-DAG: [[BPADDR5:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX3]]
+// CHECK-DAG: [[PADDR5:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX3]]
// The names below are not necessarily consistent with the names used for the
// addresses above as some are repeated.
@@ -582,15 +619,21 @@ int bar(int n){
// CHECK-DAG: store i[[SZ]] 4, i[[SZ]]* {{%[^,]+}}
// CHECK-DAG: store [[S1]]* %{{.+}}, [[S1]]** [[CBPADDR3:%.+]],
-// CHECK-DAG: store [[S1]]* %{{.+}}, [[S1]]** [[CPADDR3:%.+]],
+// CHECK-DAG: store double* %{{.+}}, double** [[CPADDR3:%.+]],
// CHECK-DAG: [[CBPADDR3]] = bitcast i8** {{%[^,]+}} to [[S1]]**
-// CHECK-DAG: [[CPADDR3]] = bitcast i8** {{%[^,]+}} to [[S1]]**
+// CHECK-DAG: [[CPADDR3]] = bitcast i8** {{%[^,]+}} to double**
+// CHECK-DAG: store i[[SZ]] 8, i[[SZ]]* {{%[^,]+}}
+
+// CHECK-DAG: store [[S1]]* %{{.+}}, [[S1]]** [[CBPADDR4:%.+]],
+// CHECK-DAG: store double* %{{.+}}, double** [[CPADDR4:%.+]],
+// CHECK-DAG: [[CBPADDR4]] = bitcast i8** {{%[^,]+}} to [[S1]]**
+// CHECK-DAG: [[CPADDR4]] = bitcast i8** {{%[^,]+}} to double**
// CHECK-DAG: store i[[SZ]] 8, i[[SZ]]* {{%[^,]+}}
-// CHECK-DAG: store i16* %{{.+}}, i16** [[CBPADDR4:%.+]],
-// CHECK-DAG: store i16* %{{.+}}, i16** [[CPADDR4:%.+]],
-// CHECK-DAG: [[CBPADDR4]] = bitcast i8** {{%[^,]+}} to i16**
-// CHECK-DAG: [[CPADDR4]] = bitcast i8** {{%[^,]+}} to i16**
+// CHECK-DAG: store i16* %{{.+}}, i16** [[CBPADDR5:%.+]],
+// CHECK-DAG: store i16* %{{.+}}, i16** [[CPADDR5:%.+]],
+// CHECK-DAG: [[CBPADDR5]] = bitcast i8** {{%[^,]+}} to i16**
+// CHECK-DAG: [[CPADDR5]] = bitcast i8** {{%[^,]+}} to i16**
// CHECK-DAG: store i[[SZ]] [[CSIZE]], i[[SZ]]* {{%[^,]+}}
// CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0
@@ -733,7 +776,7 @@ int bar(int n){
// CHECK-32-DAG:store i32 [[LOCAL_BV]], i32* [[LOCAL_B_CASTED]], align
// CHECK-DAG: [[REF_B:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_B_CASTED]],
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, [[S1]]*, i[[SZ]], i[[SZ]], i[[SZ]], i16*)* [[OMP_OUTLINED5:@.+]] to void (i32*, i32*, ...)*), [[S1]]* [[REF_THIS]], i[[SZ]] [[REF_B]], i[[SZ]] [[VAL_VLA1]], i[[SZ]] [[VAL_VLA2]], i16* [[REF_C]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, [[S1]]*, i[[SZ]], i[[SZ]], i[[SZ]], i16*)* [[OMP_OUTLINED5:@.+]] to void (i32*, i32*, ...)*), [[S1]]* [[REF_THIS]], i[[SZ]] [[REF_B]], i[[SZ]] [[VAL_VLA1]], i[[SZ]] [[VAL_VLA2]], i16* [[REF_C]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED5]](i32* noalias %.global_tid., i32* noalias %.bound_tid., [[S1]]* %{{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, i16* {{.+}})
@@ -778,7 +821,7 @@ int bar(int n){
// CHECK-DAG: store i8 [[CONV_AAA]], i8* [[CONV]], align
// CHECK-DAG: [[REF_AAA:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_AAA_CASTED]],
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]], i[[SZ]], i[[SZ]], [10 x i32]*)* [[OMP_OUTLINED6:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], i[[SZ]] {{.+}}, i[[SZ]] [[REF_AA]], i[[SZ]] [[REF_AAA]], [10 x i32]* [[REF_B]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 5, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]], i[[SZ]], i[[SZ]], [10 x i32]*)* [[OMP_OUTLINED6:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], i[[SZ]] {{.+}}, i[[SZ]] [[REF_AA]], i[[SZ]] [[REF_AAA]], [10 x i32]* [[REF_B]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED6]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, [10 x i32]* {{.+}})
@@ -811,7 +854,7 @@ int bar(int n){
// CHECK-DAG: store i16 [[CONV_AA]], i16* [[CONV]], align
// CHECK-DAG: [[REF_AA:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_AA_CASTED]],
-// CHECK: call {{.*}}void (%ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%ident_t* [[DEF_LOC]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]], [10 x i32]*)* [[OMP_OUTLINED7:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], i[[SZ]] [[REF_AA]], [10 x i32]* [[REF_B]])
+// CHECK: call {{.*}}void (%struct.ident_t*, i32, void (i32*, i32*, ...)*, ...) @__kmpc_fork_teams(%struct.ident_t* [[DEF_LOC]], i32 3, void (i32*, i32*, ...)* bitcast (void (i32*, i32*, i[[SZ]], i[[SZ]], [10 x i32]*)* [[OMP_OUTLINED7:@.+]] to void (i32*, i32*, ...)*), i[[SZ]] [[REF_A]], i[[SZ]] [[REF_AA]], [10 x i32]* [[REF_B]])
//
//
// CHECK: define internal {{.*}}void [[OMP_OUTLINED7]](i32* noalias %.global_tid., i32* noalias %.bound_tid., i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, [10 x i32]* {{.+}})