1 // Test host codegen. 2 // 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 --check-prefix OMP45 3 // 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 4 // 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 --check-prefix OMP45 5 // 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 --check-prefix OMP45 6 // 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 7 // 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 --check-prefix OMP45 8 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix OMP50 9 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s 10 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix OMP50 11 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix OMP50 12 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s 13 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix OMP50 14 15 // 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 --check-prefix SIMD-ONLY0 %s 16 // 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 17 // 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 --check-prefix SIMD-ONLY0 %s 18 // 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 --check-prefix SIMD-ONLY0 %s 19 // 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 20 // 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 --check-prefix SIMD-ONLY0 %s 21 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm %s -o - | FileCheck --check-prefix SIMD-ONLY0 %s 22 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -std=c++11 -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-pch -o %t %s 23 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix SIMD-ONLY0 %s 24 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm %s -o - | FileCheck --check-prefix SIMD-ONLY0 %s 25 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -std=c++11 -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-pch -o %t %s 26 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -std=c++11 -include-pch %t -verify %s -emit-llvm -o - | FileCheck --check-prefix SIMD-ONLY0 %s 27 // SIMD-ONLY0-NOT: {{__kmpc|__tgt}} 28 29 // Test target codegen - host bc file has to be created first. 30 // 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 31 // 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 --check-prefix TOMP45 32 // 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 33 // 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 --check-prefix TOMP45 34 // 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 35 // 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 --check-prefix TOMP45 36 // 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 37 // 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 --check-prefix TOMP45 38 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc 39 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix TOMP50 40 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 41 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix TOMP50 42 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc 43 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix TOMP50 44 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 45 // RUN: %clang_cc1 -fopenmp -fopenmp-version=50 -DOMP5 -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 --check-prefix TOMP50 46 47 // 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 48 // 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 --check-prefix SIMD-ONLY1 %s 49 // 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 50 // 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 --check-prefix SIMD-ONLY1 %s 51 // 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 52 // 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 --check-prefix SIMD-ONLY1 %s 53 // 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 54 // 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 --check-prefix SIMD-ONLY1 %s 55 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=powerpc64le-ibm-linux-gnu -emit-llvm-bc %s -o %t-ppc-host.bc 56 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 --check-prefix SIMD-ONLY1 %s 57 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 58 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 --check-prefix SIMD-ONLY1 %s 59 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -x c++ -triple i386-unknown-unknown -fopenmp-targets=i386-pc-linux-gnu -emit-llvm-bc %s -o %t-x86-host.bc 60 // RUN: %clang_cc1 -verify -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 --check-prefix SIMD-ONLY1 %s 61 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 62 // RUN: %clang_cc1 -fopenmp-simd -fopenmp-version=50 -DOMP5 -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 --check-prefix SIMD-ONLY1 %s 63 // SIMD-ONLY1-NOT: {{__kmpc|__tgt}} 64 65 // expected-no-diagnostics 66 #ifndef HEADER 67 #define HEADER 68 69 // CHECK-DAG: [[TT:%.+]] = type { i64, i8 } 70 // CHECK-DAG: [[S1:%.+]] = type { double } 71 // CHECK-DAG: [[ENTTY:%.+]] = type { i8*, i8*, i[[SZ:32|64]], i32, i32 } 72 73 // TCHECK: [[ENTTY:%.+]] = type { i8*, i8*, i{{32|64}}, i32, i32 } 74 75 // We have 8 target regions, but only 7 that actually will generate offloading 76 // code, only 6 will have mapped arguments, and only 4 have all-constant map 77 // sizes. 78 79 // CHECK-DAG: [[SIZET2:@.+]] = private unnamed_addr constant [3 x i64] [i64 2, i64 4, i64 4] 80 // CHECK-DAG: [[MAPT2:@.+]] = private unnamed_addr constant [3 x i64] [i64 800, i64 800, i64 800] 81 // CHECK-DAG: [[SIZET3:@.+]] = private unnamed_addr constant [2 x i64] [i64 4, i64 2] 82 // CHECK-DAG: [[MAPT3:@.+]] = private unnamed_addr constant [2 x i64] [i64 800, i64 800] 83 // CHECK-DAG: [[MAPT4:@.+]] = private unnamed_addr constant [9 x i64] [i64 800, i64 547, i64 800, i64 547, i64 547, i64 800, i64 800, i64 547, i64 547] 84 // CHECK-DAG: [[SIZET5:@.+]] = private unnamed_addr constant [3 x i64] [i64 4, i64 2, i64 40] 85 // CHECK-DAG: [[MAPT5:@.+]] = private unnamed_addr constant [3 x i64] [i64 800, i64 800, i64 547] 86 // CHECK-DAG: [[SIZET6:@.+]] = private unnamed_addr constant [4 x i64] [i64 4, i64 2, i64 1, i64 40] 87 // CHECK-DAG: [[MAPT6:@.+]] = private unnamed_addr constant [4 x i64] [i64 800, i64 800, i64 800, i64 547] 88 // OMP45-DAG: [[MAPT7:@.+]] = private unnamed_addr constant [5 x i64] [i64 547, i64 800, i64 800, i64 800, i64 547] 89 // OMP50-DAG: [[MAPT7:@.+]] = private unnamed_addr constant [6 x i64] [i64 547, i64 800, i64 800, i64 800, i64 547, i64 800] 90 // CHECK-DAG: @{{.*}} = weak constant i8 0 91 // CHECK-DAG: @{{.*}} = weak constant i8 0 92 // CHECK-DAG: @{{.*}} = weak constant i8 0 93 // CHECK-DAG: @{{.*}} = weak constant i8 0 94 // CHECK-DAG: @{{.*}} = weak constant i8 0 95 // CHECK-DAG: @{{.*}} = weak constant i8 0 96 // CHECK-DAG: @{{.*}} = weak constant i8 0 97 98 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 99 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 100 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 101 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 102 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 103 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 104 // TCHECK: @{{.+}} = weak constant [[ENTTY]] 105 // TCHECK-NOT: @{{.+}} = weak constant [[ENTTY]] 106 107 // Check target registration is registered as a Ctor. 108 // CHECK: appending global [1 x { i32, void ()*, i8* }] [{ i32, void ()*, i8* } { i32 0, void ()* @.omp_offloading.requires_reg, i8* null }] 109 110 111 template<typename tx, typename ty> 112 struct TT{ 113 tx X; 114 ty Y; 115 }; 116 117 // CHECK-LABEL: get_val 118 long long get_val() { return 0; } 119 120 // CHECK: define {{.*}}[[FOO:@.+]]( 121 int foo(int n) { 122 int a = 0; 123 short aa = 0; 124 float b[10]; 125 float bn[n]; 126 double c[5][10]; 127 double cn[5][n]; 128 TT<long long, char> d; 129 130 // CHECK: [[RET:%.+]] = call i32 @__tgt_target_teams_nowait_mapper(i64 -1, i8* @{{[^,]+}}, i32 0, i8** null, i8** null, i64* null, i64* null, i8** null, i32 1, i32 1) 131 // CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 132 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:[^,]+]], label %[[END:[^,]+]] 133 // CHECK: [[FAIL]] 134 // CHECK: call void [[HVT0:@.+]]() 135 // CHECK-NEXT: br label %[[END]] 136 // CHECK: [[END]] 137 #pragma omp target simd nowait 138 for (int i = 3; i < 32; i += 5) { 139 } 140 141 // CHECK: call void [[HVT1:@.+]](i[[SZ]] {{[^,]+}}, i{{32|64}}{{[*]*}} {{[^)]+}}) 142 long long k = get_val(); 143 #pragma omp target simd if(target: 0) linear(k : 3) 144 for (int i = 10; i > 1; i--) { 145 a += 1; 146 } 147 148 // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 3, i8** [[BP:%[^,]+]], i8** [[P:%[^,]+]], i64* getelementptr inbounds ([3 x i64], [3 x i64]* [[SIZET2]], i32 0, i32 0), i64* getelementptr inbounds ([3 x i64], [3 x i64]* [[MAPT2]], i32 0, i32 0), i8** null, i32 1, i32 1) 149 // CHECK-DAG: [[BP]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BPR:%[^,]+]], i32 0, i32 0 150 // CHECK-DAG: [[P]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[PR:%[^,]+]], i32 0, i32 0 151 // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BPR]], i32 0, i32 0 152 // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[PR]], i32 0, i32 0 153 // CHECK-DAG: [[CBPADDR0:%.+]] = bitcast i8** [[BPADDR0]] to i[[SZ]]* 154 // CHECK-DAG: [[CPADDR0:%.+]] = bitcast i8** [[PADDR0]] to i[[SZ]]* 155 // CHECK-DAG: store i[[SZ]] [[VAL0:%.+]], i[[SZ]]* [[CBPADDR0]], 156 // CHECK-DAG: store i[[SZ]] [[VAL0]], i[[SZ]]* [[CPADDR0]], 157 // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BPR]], i32 0, i32 1 158 // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[PR]], i32 0, i32 1 159 // CHECK-DAG: [[CBPADDR1:%.+]] = bitcast i8** [[BPADDR1]] to i[[SZ]]* 160 // CHECK-DAG: [[CPADDR1:%.+]] = bitcast i8** [[PADDR1]] to i[[SZ]]* 161 // CHECK-DAG: store i[[SZ]] [[VAL1:%.+]], i[[SZ]]* [[CBPADDR1]], 162 // CHECK-DAG: store i[[SZ]] [[VAL1]], i[[SZ]]* [[CPADDR1]], 163 // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BPR]], i32 0, i32 2 164 // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[PR]], i32 0, i32 2 165 // CHECK-DAG: [[CBPADDR2:%.+]] = bitcast i8** [[BPADDR2]] to i[[SZ]]* 166 // CHECK-DAG: [[CPADDR2:%.+]] = bitcast i8** [[PADDR2]] to i[[SZ]]* 167 // CHECK-DAG: store i[[SZ]] [[VAL2:%.+]], i[[SZ]]* [[CBPADDR2]], 168 // CHECK-DAG: store i[[SZ]] [[VAL2]], i[[SZ]]* [[CPADDR2]], 169 170 // CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 171 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:[^,]+]], label %[[END:[^,]+]] 172 // CHECK: [[FAIL]] 173 // CHECK: call void [[HVT2:@.+]](i[[SZ]] {{[^,]+}}, i[[SZ]] {{[^)]+}}) 174 // CHECK-NEXT: br label %[[END]] 175 // CHECK: [[END]] 176 int lin = 12; 177 #pragma omp target simd if(target: 1) linear(lin, a : get_val()) 178 for (unsigned long long it = 2000; it >= 600; it-=400) { 179 aa += 1; 180 } 181 182 // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 10 183 // CHECK: br i1 [[IF]], label %[[IFTHEN:[^,]+]], label %[[IFELSE:[^,]+]] 184 // CHECK: [[IFTHEN]] 185 // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 2, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* getelementptr inbounds ([2 x i64], [2 x i64]* [[SIZET3]], i32 0, i32 0), i64* getelementptr inbounds ([2 x i64], [2 x i64]* [[MAPT3]], i32 0, i32 0), i8** null, i32 1, i32 1) 186 // CHECK-DAG: [[BPR]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[BP:%[^,]+]], i32 0, i32 0 187 // CHECK-DAG: [[PR]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[P:%[^,]+]], i32 0, i32 0 188 189 // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[BP]], i32 0, i32 0 190 // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[P]], i32 0, i32 0 191 // CHECK-DAG: [[CBPADDR0:%.+]] = bitcast i8** [[BPADDR0]] to i[[SZ]]* 192 // CHECK-DAG: [[CPADDR0:%.+]] = bitcast i8** [[PADDR0]] to i[[SZ]]* 193 // CHECK-DAG: store i[[SZ]] [[VAL0:%.+]], i[[SZ]]* [[CBPADDR0]], 194 // CHECK-DAG: store i[[SZ]] [[VAL0]], i[[SZ]]* [[CPADDR0]], 195 196 // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[BP]], i32 0, i32 1 197 // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [2 x i8*], [2 x i8*]* [[P]], i32 0, i32 1 198 // CHECK-DAG: [[CBPADDR1:%.+]] = bitcast i8** [[BPADDR1]] to i[[SZ]]* 199 // CHECK-DAG: [[CPADDR1:%.+]] = bitcast i8** [[PADDR1]] to i[[SZ]]* 200 // CHECK-DAG: store i[[SZ]] [[VAL1:%.+]], i[[SZ]]* [[CBPADDR1]], 201 // CHECK-DAG: store i[[SZ]] [[VAL1]], i[[SZ]]* [[CPADDR1]], 202 // CHECK: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 203 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:.+]], label %[[END:[^,]+]] 204 // CHECK: [[FAIL]] 205 // CHECK: call void [[HVT3:@.+]]({{[^,]+}}, {{[^,]+}}) 206 // CHECK-NEXT: br label %[[END]] 207 // CHECK: [[END]] 208 // CHECK-NEXT: br label %[[IFEND:.+]] 209 // CHECK: [[IFELSE]] 210 // CHECK: call void [[HVT3]]({{[^,]+}}, {{[^,]+}}) 211 // CHECK-NEXT: br label %[[IFEND]] 212 // CHECK: [[IFEND]] 213 214 #pragma omp target simd if(target: n>10) 215 for (short it = 6; it <= 20; it-=-4) { 216 a += 1; 217 aa += 1; 218 } 219 220 // We capture 3 VLA sizes in this target region 221 // CHECK-64: [[A_VAL:%.+]] = load i32, i32* %{{.+}}, 222 // CHECK-64: [[A_ADDR:%.+]] = bitcast i[[SZ]]* [[A_CADDR:%.+]] to i32* 223 // CHECK-64: store i32 [[A_VAL]], i32* [[A_ADDR]], 224 // CHECK-64: [[A_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[A_CADDR]], 225 226 // CHECK-32: [[A_VAL:%.+]] = load i32, i32* %{{.+}}, 227 // CHECK-32: store i32 [[A_VAL]], i32* [[A_CADDR:%.+]], 228 // CHECK-32: [[A_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[A_CADDR]], 229 230 // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 20 231 // CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]] 232 // CHECK: [[TRY]] 233 // CHECK-64: [[BNSIZE:%.+]] = mul nuw i64 [[VLA0:%.+]], 4 234 // CHECK-32: [[BNSZSIZE:%.+]] = mul nuw i32 [[VLA0:%.+]], 4 235 // CHECK-32: [[BNSIZE:%.+]] = sext i32 [[BNSZSIZE]] to i64 236 // CHECK: [[CNELEMSIZE2:%.+]] = mul nuw i[[SZ]] 5, [[VLA1:%.+]] 237 // CHECK-64: [[CNSIZE:%.+]] = mul nuw i64 [[CNELEMSIZE2]], 8 238 // CHECK-32: [[CNSZSIZE:%.+]] = mul nuw i32 [[CNELEMSIZE2]], 8 239 // CHECK-32: [[CNSIZE:%.+]] = sext i32 [[CNSZSIZE]] to i64 240 241 // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 9, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* [[SR:%[^,]+]], i64* getelementptr inbounds ([9 x i64], [9 x i64]* [[MAPT4]], i32 0, i32 0), i8** null, i32 1, i32 1) 242 // CHECK-DAG: [[BPR]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP:%[^,]+]], i32 0, i32 0 243 // CHECK-DAG: [[PR]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P:%[^,]+]], i32 0, i32 0 244 // CHECK-DAG: [[SR]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S:%[^,]+]], i32 0, i32 0 245 246 // CHECK-DAG: [[SADDR0:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX0:[0-9]+]] 247 // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX0]] 248 // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX0]] 249 // CHECK-DAG: [[SADDR1:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX1:[0-9]+]] 250 // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX1]] 251 // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX1]] 252 // CHECK-DAG: [[SADDR2:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX2:[0-9]+]] 253 // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX2]] 254 // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX2]] 255 // CHECK-DAG: [[SADDR3:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX3:[0-9]+]] 256 // CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX3]] 257 // CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX3]] 258 // CHECK-DAG: [[SADDR4:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX4:[0-9]+]] 259 // CHECK-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX4]] 260 // CHECK-DAG: [[PADDR4:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX4]] 261 // CHECK-DAG: [[SADDR5:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX5:[0-9]+]] 262 // CHECK-DAG: [[BPADDR5:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX5]] 263 // CHECK-DAG: [[PADDR5:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX5]] 264 // CHECK-DAG: [[SADDR6:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX6:[0-9]+]] 265 // CHECK-DAG: [[BPADDR6:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX6]] 266 // CHECK-DAG: [[PADDR6:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX6]] 267 // CHECK-DAG: [[SADDR7:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX7:[0-9]+]] 268 // CHECK-DAG: [[BPADDR7:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX7]] 269 // CHECK-DAG: [[PADDR7:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX7]] 270 // CHECK-DAG: [[SADDR8:%.+]] = getelementptr inbounds [9 x i64], [9 x i64]* [[S]], i32 0, i32 [[IDX8:[0-9]+]] 271 // CHECK-DAG: [[BPADDR8:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[BP]], i32 0, i32 [[IDX8]] 272 // CHECK-DAG: [[PADDR8:%.+]] = getelementptr inbounds [9 x i8*], [9 x i8*]* [[P]], i32 0, i32 [[IDX8]] 273 274 // The names below are not necessarily consistent with the names used for the 275 // addresses above as some are repeated. 276 // CHECK-DAG: store i[[SZ]] [[A_CVAL]], i[[SZ]]* [[CBPADDR0:%.+]], 277 // CHECK-DAG: store i[[SZ]] [[A_CVAL]], i[[SZ]]* [[CPADDR0:%.+]], 278 // CHECK-DAG: [[CBPADDR0]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 279 // CHECK-DAG: [[CPADDR0]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 280 // CHECK-DAG: store i64 4, i64* {{%[^,]+}} 281 282 // CHECK-DAG: store [10 x float]* %{{.+}}, [10 x float]** [[CBPADDR1:%.+]], 283 // CHECK-DAG: store [10 x float]* %{{.+}}, [10 x float]** [[CPADDR1:%.+]], 284 // CHECK-DAG: [[CBPADDR1]] = bitcast i8** {{%[^,]+}} to [10 x float]** 285 // CHECK-DAG: [[CPADDR1]] = bitcast i8** {{%[^,]+}} to [10 x float]** 286 // CHECK-DAG: store i64 40, i64* {{%[^,]+}} 287 288 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CBPADDR2:%.+]], 289 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CPADDR2:%.+]], 290 // CHECK-DAG: [[CBPADDR2]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 291 // CHECK-DAG: [[CPADDR2]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 292 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 293 294 // CHECK-DAG: store float* %{{.+}}, float** [[CBPADDR3:%.+]], 295 // CHECK-DAG: store float* %{{.+}}, float** [[CPADDR3:%.+]], 296 // CHECK-DAG: [[CBPADDR3]] = bitcast i8** {{%[^,]+}} to float** 297 // CHECK-DAG: [[CPADDR3]] = bitcast i8** {{%[^,]+}} to float** 298 // CHECK-DAG: store i64 [[BNSIZE]], i64* {{%[^,]+}} 299 300 // CHECK-DAG: store [5 x [10 x double]]* %{{.+}}, [5 x [10 x double]]** [[CBPADDR4:%.+]], 301 // CHECK-DAG: store [5 x [10 x double]]* %{{.+}}, [5 x [10 x double]]** [[CPADDR4:%.+]], 302 // CHECK-DAG: [[CBPADDR4]] = bitcast i8** {{%[^,]+}} to [5 x [10 x double]]** 303 // CHECK-DAG: [[CPADDR4]] = bitcast i8** {{%[^,]+}} to [5 x [10 x double]]** 304 // CHECK-DAG: store i64 400, i64* {{%[^,]+}} 305 306 // CHECK-DAG: store i[[SZ]] 5, i[[SZ]]* [[CBPADDR5:%.+]], 307 // CHECK-DAG: store i[[SZ]] 5, i[[SZ]]* [[CPADDR5:%.+]], 308 // CHECK-DAG: [[CBPADDR5]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 309 // CHECK-DAG: [[CPADDR5]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 310 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 311 312 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CBPADDR6:%.+]], 313 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CPADDR6:%.+]], 314 // CHECK-DAG: [[CBPADDR6]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 315 // CHECK-DAG: [[CPADDR6]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 316 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 317 318 // CHECK-DAG: store double* %{{.+}}, double** [[CBPADDR7:%.+]], 319 // CHECK-DAG: store double* %{{.+}}, double** [[CPADDR7:%.+]], 320 // CHECK-DAG: [[CBPADDR7]] = bitcast i8** {{%[^,]+}} to double** 321 // CHECK-DAG: [[CPADDR7]] = bitcast i8** {{%[^,]+}} to double** 322 // CHECK-DAG: store i64 [[CNSIZE]], i64* {{%[^,]+}} 323 324 // CHECK-DAG: store [[TT]]* %{{.+}}, [[TT]]** [[CBPADDR8:%.+]], 325 // CHECK-DAG: store [[TT]]* %{{.+}}, [[TT]]** [[CPADDR8:%.+]], 326 // CHECK-DAG: [[CBPADDR8]] = bitcast i8** {{%[^,]+}} to [[TT]]** 327 // CHECK-DAG: [[CPADDR8]] = bitcast i8** {{%[^,]+}} to [[TT]]** 328 // CHECK-DAG: store i64 {{12|16}}, i64* {{%[^,]+}} 329 330 // CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 331 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:[^,]+]], label %[[END:[^,]+]] 332 333 // CHECK: [[FAIL]] 334 // CHECK: call void [[HVT4:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 335 // CHECK-NEXT: br label %[[END]] 336 // CHECK: [[END]] 337 #pragma omp target simd if(target: n>20) 338 for (unsigned char it = 'z'; it >= 'a'; it+=-1) { 339 a += 1; 340 b[2] += 1.0; 341 bn[3] += 1.0; 342 c[1][2] += 1.0; 343 cn[1][3] += 1.0; 344 d.X += 1; 345 d.Y += 1; 346 } 347 348 return a; 349 } 350 351 // Check that the offloading functions are emitted and that the arguments are 352 // correct and loaded correctly for the target regions in foo(). 353 354 // CHECK: define internal void [[HVT0]]() 355 // CHECK: !llvm.loop 356 // CHECK: ret void 357 // CHECK-NEXT: } 358 359 360 // CHECK: define internal void [[HVT1]](i[[SZ]] %{{.+}}, i{{32|64}}{{[*]*.*}} %{{.+}}) 361 // CHECK: [[AA_ADDR:%.+]] = alloca i[[SZ]], align 362 // CHECK: store i[[SZ]] %{{.+}}, i[[SZ]]* [[AA_ADDR]], align 363 // CHECK-64: [[AA_CADDR:%.+]] = bitcast i[[SZ]]* [[AA_ADDR]] to i32* 364 // CHECK-64: [[AA:%.+]] = load i32, i32* [[AA_CADDR]], align 365 // CHECK-32: [[AA:%.+]] = load i32, i32* [[AA_ADDR]], align 366 // CHECK: !llvm.access.group 367 // CHECK: !llvm.loop 368 // CHECK: ret void 369 // CHECK-NEXT: } 370 371 // CHECK: define internal void [[HVT2]](i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}, i[[SZ]] %{{.+}}) 372 // CHECK: [[AA_ADDR:%.+]] = alloca i[[SZ]], align 373 // CHECK: store i[[SZ]] %{{.+}}, i[[SZ]]* [[AA_ADDR]], align 374 // CHECK: [[AA_CADDR:%.+]] = bitcast i[[SZ]]* [[AA_ADDR]] to i16* 375 // CHECK: [[AA:%.+]] = load i16, i16* [[AA_CADDR]], align 376 // CHECK: !llvm.loop 377 // CHECK: ret void 378 // CHECK-NEXT: } 379 380 // CHECK: define internal void [[HVT3]] 381 // CHECK: [[A_ADDR:%.+]] = alloca i[[SZ]], align 382 // CHECK: [[AA_ADDR:%.+]] = alloca i[[SZ]], align 383 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[A_ADDR]], align 384 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[AA_ADDR]], align 385 // CHECK-64-DAG:[[A_CADDR:%.+]] = bitcast i[[SZ]]* [[A_ADDR]] to i32* 386 // CHECK-DAG: [[AA_CADDR:%.+]] = bitcast i[[SZ]]* [[AA_ADDR]] to i16* 387 // CHECK: !llvm.loop 388 // CHECK: ret void 389 // CHECK-NEXT: } 390 391 // CHECK: define internal void [[HVT4]] 392 // Create local storage for each capture. 393 // CHECK: [[LOCAL_A:%.+]] = alloca i[[SZ]] 394 // CHECK: [[LOCAL_B:%.+]] = alloca [10 x float]* 395 // CHECK: [[LOCAL_VLA1:%.+]] = alloca i[[SZ]] 396 // CHECK: [[LOCAL_BN:%.+]] = alloca float* 397 // CHECK: [[LOCAL_C:%.+]] = alloca [5 x [10 x double]]* 398 // CHECK: [[LOCAL_VLA2:%.+]] = alloca i[[SZ]] 399 // CHECK: [[LOCAL_VLA3:%.+]] = alloca i[[SZ]] 400 // CHECK: [[LOCAL_CN:%.+]] = alloca double* 401 // CHECK: [[LOCAL_D:%.+]] = alloca [[TT]]* 402 // CHECK-DAG: store i[[SZ]] [[ARG_A:%.+]], i[[SZ]]* [[LOCAL_A]] 403 // CHECK-DAG: store [10 x float]* [[ARG_B:%.+]], [10 x float]** [[LOCAL_B]] 404 // CHECK-DAG: store i[[SZ]] [[ARG_VLA1:%.+]], i[[SZ]]* [[LOCAL_VLA1]] 405 // CHECK-DAG: store float* [[ARG_BN:%.+]], float** [[LOCAL_BN]] 406 // CHECK-DAG: store [5 x [10 x double]]* [[ARG_C:%.+]], [5 x [10 x double]]** [[LOCAL_C]] 407 // CHECK-DAG: store i[[SZ]] [[ARG_VLA2:%.+]], i[[SZ]]* [[LOCAL_VLA2]] 408 // CHECK-DAG: store i[[SZ]] [[ARG_VLA3:%.+]], i[[SZ]]* [[LOCAL_VLA3]] 409 // CHECK-DAG: store double* [[ARG_CN:%.+]], double** [[LOCAL_CN]] 410 // CHECK-DAG: store [[TT]]* [[ARG_D:%.+]], [[TT]]** [[LOCAL_D]] 411 412 // CHECK-64-DAG:[[CONV_AP:%.+]] = bitcast i[[SZ]]* [[LOCAL_A]] to i32* 413 // CHECK-DAG: [[REF_B:%.+]] = load [10 x float]*, [10 x float]** [[LOCAL_B]], 414 // CHECK-DAG: [[VAL_VLA1:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_VLA1]], 415 // CHECK-DAG: [[REF_BN:%.+]] = load float*, float** [[LOCAL_BN]], 416 // CHECK-DAG: [[REF_C:%.+]] = load [5 x [10 x double]]*, [5 x [10 x double]]** [[LOCAL_C]], 417 // CHECK-DAG: [[VAL_VLA2:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_VLA2]], 418 // CHECK-DAG: [[VAL_VLA3:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_VLA3]], 419 // CHECK-DAG: [[REF_CN:%.+]] = load double*, double** [[LOCAL_CN]], 420 // CHECK-DAG: [[REF_D:%.+]] = load [[TT]]*, [[TT]]** [[LOCAL_D]], 421 422 423 template<typename tx> 424 tx ftemplate(int n) { 425 tx a = 0; 426 short aa = 0; 427 tx b[10]; 428 429 #pragma omp target simd if(target: n>40) 430 for (long long i = -10; i < 10; i += 3) { 431 a += 1; 432 aa += 1; 433 b[2] += 1; 434 } 435 436 return a; 437 } 438 439 static 440 int fstatic(int n) { 441 int a = 0; 442 short aa = 0; 443 char aaa = 0; 444 int b[10]; 445 446 #pragma omp target simd if(target: n>50) 447 for (unsigned i=100; i<10; i+=10) { 448 a += 1; 449 aa += 1; 450 aaa += 1; 451 b[2] += 1; 452 } 453 454 return a; 455 } 456 457 struct S1 { 458 double a; 459 460 int r1(int n){ 461 int b = n+1; 462 short int c[2][n]; 463 464 #ifdef OMP5 465 #pragma omp target simd if(n>60) nontemporal(a) private(a) 466 #else 467 #pragma omp target simd if(n>60) private(a) 468 #endif // OMP5 469 for (unsigned long long it = 2000; it >= 600; it -= 400) { 470 this->a = (double)b + 1.5; 471 c[1][1] = ++a; 472 } 473 474 return c[1][1] + (int)b; 475 } 476 }; 477 478 // CHECK: define {{.*}}@{{.*}}bar{{.*}} 479 int bar(int n){ 480 int a = 0; 481 482 // CHECK: call {{.*}}i32 [[FOO]](i32 {{.*}}) 483 a += foo(n); 484 485 S1 S; 486 // CHECK: call {{.*}}i32 [[FS1:@.+]]([[S1]]* {{.*}}, i32 {{.*}}) 487 a += S.r1(n); 488 489 // CHECK: call {{.*}}i32 [[FSTATIC:@.+]](i32 {{.*}}) 490 a += fstatic(n); 491 492 // CHECK: call {{.*}}i32 [[FTEMPLATE:@.+]](i32 {{.*}}) 493 a += ftemplate<int>(n); 494 495 return a; 496 } 497 498 // 499 // CHECK: define {{.*}}[[FS1]] 500 // 501 // CHECK: i8* @llvm.stacksave() 502 // CHECK-32: store i32 %{{.+}}, i32* %__vla_expr 503 // OMP50: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 60 504 // CHECK-64: [[B_ADDR:%.+]] = bitcast i[[SZ]]* [[B_CADDR:%.+]] to i32* 505 // CHECK-64: store i32 %{{.+}}, i32* [[B_ADDR]], 506 // CHECK-64: [[B_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[B_CADDR]], 507 508 // CHECK-32: store i32 %{{.+}}, i32* [[B_ADDR:%.+]], 509 // CHECK-32: [[B_CVAL:%.+]] = load i[[SZ]], i[[SZ]]* [[B_ADDR]], 510 511 // OMP45: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 60 512 // OMP50: [[TOBOOL:%.+]] = trunc i8 %{{.+}} to i1 513 // OMP50: [[CONV:%.+]] = bitcast i[[SZ]]* [[CAP:%.+]] to i8* 514 // OMP50: [[FROMBOOL:%.+]] = zext i1 [[TOBOOL]] to i8 515 // OMP50: store i8 [[FROMBOOL]], i8* [[CONV]], 516 // OMP50: [[SIMD_COND:%.+]] = load i[[SZ]], i[[SZ]]* [[CAP]], 517 // OMP50: [[IF:%.+]] = trunc i8 %{{.+}} to i1 518 // CHECK: br i1 [[IF]], label %[[TRY:[^,]+]], label %[[FAIL:[^,]+]] 519 // CHECK: [[TRY]] 520 // We capture 2 VLA sizes in this target region 521 // CHECK: [[CELEMSIZE2:%.+]] = mul nuw i[[SZ]] 2, [[VLA0:%.+]] 522 // CHECK-64: [[CSIZE:%.+]] = mul nuw i64 [[CELEMSIZE2]], 2 523 // CHECK-32: [[CSZSIZE:%.+]] = mul nuw i32 [[CELEMSIZE2]], 2 524 // CHECK-32: [[CSIZE:%.+]] = sext i32 [[CSZSIZE]] to i64 525 526 // OMP45-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 5, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* [[SR:%[^,]+]], i64* getelementptr inbounds ([5 x i64], [5 x i64]* [[MAPT7]], i32 0, i32 0), i8** null, i32 1, i32 1) 527 // OMP50-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 6, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* [[SR:%[^,]+]], i64* getelementptr inbounds ([6 x i64], [6 x i64]* [[MAPT7]], i32 0, i32 0), i8** null, i32 1, i32 1) 528 // OMP45-DAG: [[BPR]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP:%.+]], i32 0, i32 0 529 // OMP45-DAG: [[PR]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P:%.+]], i32 0, i32 0 530 // OMP45-DAG: [[SR]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S:%.+]], i32 0, i32 0 531 // OMP45-DAG: [[SADDR0:%.+]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S]], i32 [[IDX0:[0-9]+]] 532 // OMP45-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX0]] 533 // OMP45-DAG: [[PADDR0:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX0]] 534 // OMP45-DAG: [[SADDR1:%.+]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S]], i32 [[IDX1:[0-9]+]] 535 // OMP45-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX1]] 536 // OMP45-DAG: [[PADDR1:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX1]] 537 // OMP45-DAG: [[SADDR2:%.+]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S]], i32 [[IDX2:[0-9]+]] 538 // OMP45-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX2]] 539 // OMP45-DAG: [[PADDR2:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX2]] 540 // OMP45-DAG: [[SADDR3:%.+]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S]], i32 [[IDX3:[0-9]+]] 541 // OMP45-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX3]] 542 // OMP45-DAG: [[PADDR3:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX3]] 543 // OMP45-DAG: [[SADDR4:%.+]] = getelementptr inbounds [5 x i64], [5 x i64]* [[S]], i32 [[IDX4:[0-9]+]] 544 // OMP45-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[BP]], i32 [[IDX4]] 545 // OMP45-DAG: [[PADDR4:%.+]] = getelementptr inbounds [5 x i8*], [5 x i8*]* [[P]], i32 [[IDX4]] 546 // OMP50-DAG: [[BPR]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP:%.+]], i32 0, i32 0 547 // OMP50-DAG: [[PR]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P:%.+]], i32 0, i32 0 548 // OMP50-DAG: [[SR]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S:%.+]], i32 0, i32 0 549 // OMP50-DAG: [[SADDR0:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX0:[0-9]+]] 550 // OMP50-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX0]] 551 // OMP50-DAG: [[PADDR0:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX0]] 552 // OMP50-DAG: [[SADDR1:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX1:[0-9]+]] 553 // OMP50-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX1]] 554 // OMP50-DAG: [[PADDR1:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX1]] 555 // OMP50-DAG: [[SADDR2:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX2:[0-9]+]] 556 // OMP50-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX2]] 557 // OMP50-DAG: [[PADDR2:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX2]] 558 // OMP50-DAG: [[SADDR3:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX3:[0-9]+]] 559 // OMP50-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX3]] 560 // OMP50-DAG: [[PADDR3:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX3]] 561 // OMP50-DAG: [[SADDR4:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX4:[0-9]+]] 562 // OMP50-DAG: [[BPADDR4:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX4]] 563 // OMP50-DAG: [[PADDR4:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX4]] 564 // OMP50-DAG: [[SADDR5:%.+]] = getelementptr inbounds [6 x i64], [6 x i64]* [[S]], i32 [[IDX5:[0-9]+]] 565 // OMP50-DAG: [[BPADDR5:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[BP]], i32 [[IDX5]] 566 // OMP50-DAG: [[PADDR5:%.+]] = getelementptr inbounds [6 x i8*], [6 x i8*]* [[P]], i32 [[IDX5]] 567 568 // The names below are not necessarily consistent with the names used for the 569 // addresses above as some are repeated. 570 // CHECK-DAG: store [[S1]]* %{{.+}}, [[S1]]** [[CBPADDR0:%.+]], 571 // CHECK-DAG: store [[S1]]* %{{.+}}, [[S1]]** [[CPADDR0:%.+]], 572 // CHECK-DAG: [[CBPADDR0]] = bitcast i8** {{%[^,]+}} to [[S1]]** 573 // CHECK-DAG: [[CPADDR0]] = bitcast i8** {{%[^,]+}} to [[S1]]** 574 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 575 576 // CHECK-DAG: store i[[SZ]] [[B_CVAL]], i[[SZ]]* [[CBPADDR1:%.+]], 577 // CHECK-DAG: store i[[SZ]] [[B_CVAL]], i[[SZ]]* [[CPADDR1:%.+]], 578 // CHECK-DAG: [[CBPADDR1]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 579 // CHECK-DAG: [[CPADDR1]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 580 // CHECK-DAG: store i64 4, i64* {{%[^,]+}} 581 582 // CHECK-DAG: store i[[SZ]] 2, i[[SZ]]* [[CBPADDR2:%.+]], 583 // CHECK-DAG: store i[[SZ]] 2, i[[SZ]]* [[CPADDR2:%.+]], 584 // CHECK-DAG: [[CBPADDR2]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 585 // CHECK-DAG: [[CPADDR2]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 586 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 587 588 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CBPADDR3:%.+]], 589 // CHECK-DAG: store i[[SZ]] %{{.+}}, i[[SZ]]* [[CPADDR3:%.+]], 590 // CHECK-DAG: [[CBPADDR3]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 591 // CHECK-DAG: [[CPADDR3]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 592 // CHECK-DAG: store i64 {{4|8}}, i64* {{%[^,]+}} 593 594 // CHECK-DAG: store i16* %{{.+}}, i16** [[CBPADDR4:%.+]], 595 // CHECK-DAG: store i16* %{{.+}}, i16** [[CPADDR4:%.+]], 596 // CHECK-DAG: [[CBPADDR4]] = bitcast i8** {{%[^,]+}} to i16** 597 // CHECK-DAG: [[CPADDR4]] = bitcast i8** {{%[^,]+}} to i16** 598 // CHECK-DAG: store i64 [[CSIZE]], i64* {{%[^,]+}} 599 600 // OMP50-DAG: store i[[SZ]] [[SIMD_COND]], i[[SZ]]* [[CBPADDR5:%.+]], 601 // OMP50-DAG: store i[[SZ]] [[SIMD_COND]], i[[SZ]]* [[CPADDR5:%.+]], 602 // OMP50-DAG: [[CBPADDR5]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 603 // OMP50-DAG: [[CPADDR5]] = bitcast i8** {{%[^,]+}} to i[[SZ]]* 604 // OMP50-DAG: store i64 1, i64* {{%[^,]+}} 605 606 // CHECK-NEXT: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 607 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:[^,]+]], label %[[END:[^,]+]] 608 609 // CHECK: [[FAIL]] 610 // OMP45: call void [[HVT7:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 611 // OMP50: call void [[HVT7:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 612 // CHECK-NEXT: br label %[[END]] 613 // CHECK: [[END]] 614 615 // 616 // CHECK: define {{.*}}[[FSTATIC]] 617 // 618 // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 50 619 // CHECK: br i1 [[IF]], label %[[IFTHEN:[^,]+]], label %[[IFELSE:[^,]+]] 620 // CHECK: [[IFTHEN]] 621 // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 4, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* getelementptr inbounds ([4 x i64], [4 x i64]* [[SIZET6]], i32 0, i32 0), i64* getelementptr inbounds ([4 x i64], [4 x i64]* [[MAPT6]], i32 0, i32 0), i8** null, i32 1, i32 1) 622 // CHECK-DAG: [[BPR]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[BP:%.+]], i32 0, i32 0 623 // CHECK-DAG: [[PR]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[P:%.+]], i32 0, i32 0 624 625 // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[BP]], i32 0, i32 0 626 // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[P]], i32 0, i32 0 627 // CHECK-DAG: [[CBPADDR0:%.+]] = bitcast i8** [[BPADDR0]] to i[[SZ]]* 628 // CHECK-DAG: [[CPADDR0:%.+]] = bitcast i8** [[PADDR0]] to i[[SZ]]* 629 // CHECK-DAG: store i[[SZ]] [[VAL0:%.+]], i[[SZ]]* [[CBPADDR0]], 630 // CHECK-DAG: store i[[SZ]] [[VAL0]], i[[SZ]]* [[CPADDR0]], 631 632 // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[BP]], i32 0, i32 1 633 // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[P]], i32 0, i32 1 634 // CHECK-DAG: [[CBPADDR1:%.+]] = bitcast i8** [[BPADDR1]] to i[[SZ]]* 635 // CHECK-DAG: [[CPADDR1:%.+]] = bitcast i8** [[PADDR1]] to i[[SZ]]* 636 // CHECK-DAG: store i[[SZ]] [[VAL1:%.+]], i[[SZ]]* [[CBPADDR1]], 637 // CHECK-DAG: store i[[SZ]] [[VAL1]], i[[SZ]]* [[CPADDR1]], 638 639 // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[BP]], i32 0, i32 2 640 // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[P]], i32 0, i32 2 641 // CHECK-DAG: [[CBPADDR2:%.+]] = bitcast i8** [[BPADDR2]] to i[[SZ]]* 642 // CHECK-DAG: [[CPADDR2:%.+]] = bitcast i8** [[PADDR2]] to i[[SZ]]* 643 // CHECK-DAG: store i[[SZ]] [[VAL2:%.+]], i[[SZ]]* [[CBPADDR2]], 644 // CHECK-DAG: store i[[SZ]] [[VAL2]], i[[SZ]]* [[CPADDR2]], 645 646 // CHECK-DAG: [[BPADDR3:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[BP]], i32 0, i32 3 647 // CHECK-DAG: [[PADDR3:%.+]] = getelementptr inbounds [4 x i8*], [4 x i8*]* [[P]], i32 0, i32 3 648 // CHECK-DAG: [[CBPADDR3:%.+]] = bitcast i8** [[BPADDR3]] to [10 x i32]** 649 // CHECK-DAG: [[CPADDR3:%.+]] = bitcast i8** [[PADDR3]] to [10 x i32]** 650 // CHECK-DAG: store [10 x i32]* [[VAL3:%.+]], [10 x i32]** [[CBPADDR3]], 651 // CHECK-DAG: store [10 x i32]* [[VAL3]], [10 x i32]** [[CPADDR3]], 652 653 // CHECK: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 654 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:.+]], label %[[END:[^,]+]] 655 // CHECK: [[FAIL]] 656 // CHECK: call void [[HVT6:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 657 // CHECK-NEXT: br label %[[END]] 658 // CHECK: [[END]] 659 // CHECK-NEXT: br label %[[IFEND:.+]] 660 // CHECK: [[IFELSE]] 661 // CHECK: call void [[HVT6]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 662 // CHECK-NEXT: br label %[[IFEND]] 663 // CHECK: [[IFEND]] 664 665 // 666 // CHECK: define {{.*}}[[FTEMPLATE]] 667 // 668 // CHECK: [[IF:%.+]] = icmp sgt i32 {{[^,]+}}, 40 669 // CHECK: br i1 [[IF]], label %[[IFTHEN:[^,]+]], label %[[IFELSE:[^,]+]] 670 // CHECK: [[IFTHEN]] 671 // CHECK-DAG: [[RET:%.+]] = call i32 @__tgt_target_teams_mapper(i64 -1, i8* @{{[^,]+}}, i32 3, i8** [[BPR:%[^,]+]], i8** [[PR:%[^,]+]], i64* getelementptr inbounds ([3 x i64], [3 x i64]* [[SIZET5]], i32 0, i32 0), i64* getelementptr inbounds ([3 x i64], [3 x i64]* [[MAPT5]], i32 0, i32 0), i8** null, i32 1, i32 1) 672 // CHECK-DAG: [[BPR]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BP:%.+]], i32 0, i32 0 673 // CHECK-DAG: [[PR]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[P:%.+]], i32 0, i32 0 674 675 // CHECK-DAG: [[BPADDR0:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BP]], i32 0, i32 0 676 // CHECK-DAG: [[PADDR0:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[P]], i32 0, i32 0 677 // CHECK-DAG: [[CBPADDR0:%.+]] = bitcast i8** [[BPADDR0]] to i[[SZ]]* 678 // CHECK-DAG: [[CPADDR0:%.+]] = bitcast i8** [[PADDR0]] to i[[SZ]]* 679 // CHECK-DAG: store i[[SZ]] [[VAL0:%.+]], i[[SZ]]* [[CBPADDR0]], 680 // CHECK-DAG: store i[[SZ]] [[VAL0]], i[[SZ]]* [[CPADDR0]], 681 682 // CHECK-DAG: [[BPADDR1:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BP]], i32 0, i32 1 683 // CHECK-DAG: [[PADDR1:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[P]], i32 0, i32 1 684 // CHECK-DAG: [[CBPADDR1:%.+]] = bitcast i8** [[BPADDR1]] to i[[SZ]]* 685 // CHECK-DAG: [[CPADDR1:%.+]] = bitcast i8** [[PADDR1]] to i[[SZ]]* 686 // CHECK-DAG: store i[[SZ]] [[VAL1:%.+]], i[[SZ]]* [[CBPADDR1]], 687 // CHECK-DAG: store i[[SZ]] [[VAL1]], i[[SZ]]* [[CPADDR1]], 688 689 // CHECK-DAG: [[BPADDR2:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[BP]], i32 0, i32 2 690 // CHECK-DAG: [[PADDR2:%.+]] = getelementptr inbounds [3 x i8*], [3 x i8*]* [[P]], i32 0, i32 2 691 // CHECK-DAG: [[CBPADDR2:%.+]] = bitcast i8** [[BPADDR2]] to [10 x i32]** 692 // CHECK-DAG: [[CPADDR2:%.+]] = bitcast i8** [[PADDR2]] to [10 x i32]** 693 // CHECK-DAG: store [10 x i32]* [[VAL2:%.+]], [10 x i32]** [[CBPADDR2]], 694 // CHECK-DAG: store [10 x i32]* [[VAL2]], [10 x i32]** [[CPADDR2]], 695 696 // CHECK: [[ERROR:%.+]] = icmp ne i32 [[RET]], 0 697 // CHECK-NEXT: br i1 [[ERROR]], label %[[FAIL:.+]], label %[[END:[^,]+]] 698 // CHECK: [[FAIL]] 699 // CHECK: call void [[HVT5:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 700 // CHECK-NEXT: br label %[[END]] 701 // CHECK: [[END]] 702 // CHECK-NEXT: br label %[[IFEND:.+]] 703 // CHECK: [[IFELSE]] 704 // CHECK: call void [[HVT:@.+]]({{[^,]+}}, {{[^,]+}}, {{[^,]+}}) 705 // CHECK-NEXT: br label %[[IFEND]] 706 // CHECK: [[IFEND]] 707 708 // Check that the offloading functions are emitted and that the arguments are 709 // correct and loaded correctly for the target regions of the callees of bar(). 710 711 // CHECK: define internal void [[HVT7]] 712 // Create local storage for each capture. 713 // CHECK: [[LOCAL_THIS:%.+]] = alloca [[S1]]* 714 // CHECK: [[LOCAL_B:%.+]] = alloca i[[SZ]] 715 // CHECK: [[LOCAL_VLA1:%.+]] = alloca i[[SZ]] 716 // CHECK: [[LOCAL_VLA2:%.+]] = alloca i[[SZ]] 717 // CHECK: [[LOCAL_C:%.+]] = alloca i16* 718 // OMP50: [[LOCAL_SIMD_COND_CASTED:%.+]] = alloca i[[SZ]], 719 // CHECK-DAG: store [[S1]]* [[ARG_THIS:%.+]], [[S1]]** [[LOCAL_THIS]] 720 // CHECK-DAG: store i[[SZ]] [[ARG_B:%.+]], i[[SZ]]* [[LOCAL_B]] 721 // CHECK-DAG: store i[[SZ]] [[ARG_VLA1:%.+]], i[[SZ]]* [[LOCAL_VLA1]] 722 // CHECK-DAG: store i[[SZ]] [[ARG_VLA2:%.+]], i[[SZ]]* [[LOCAL_VLA2]] 723 // CHECK-DAG: store i16* [[ARG_C:%.+]], i16** [[LOCAL_C]] 724 // Store captures in the context. 725 // CHECK-DAG: [[REF_THIS:%.+]] = load [[S1]]*, [[S1]]** [[LOCAL_THIS]], 726 // CHECK-64-DAG:[[CONV_BP:%.+]] = bitcast i[[SZ]]* [[LOCAL_B]] to i32* 727 // CHECK-DAG: [[VAL_VLA1:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_VLA1]], 728 // CHECK-DAG: [[VAL_VLA2:%.+]] = load i[[SZ]], i[[SZ]]* [[LOCAL_VLA2]], 729 // CHECK-DAG: [[REF_C:%.+]] = load i16*, i16** [[LOCAL_C]], 730 // OMP50-DAG: [[CONV_COND:%.+]] = bitcast i[[SZ]]* [[LOCAL_SIMD_COND_CASTED]] to i8* 731 // OMP50-DAG: [[SIMD_COND:%.+]] = load i8, i8* [[CONV_COND]], 732 // OMP50-DAG: trunc i8 [[SIMD_COND]] to i1 733 // OMP45-NOT: !nontemporal 734 // OMP50: store double {{.*}}!nontemporal 735 // OMP50: load double, {{.*}}!nontemporal 736 // OMP50: store double {{.*}}!nontemporal 737 738 // CHECK: define internal void [[HVT6]] 739 // Create local storage for each capture. 740 // CHECK: [[LOCAL_A:%.+]] = alloca i[[SZ]] 741 // CHECK: [[LOCAL_AA:%.+]] = alloca i[[SZ]] 742 // CHECK: [[LOCAL_AAA:%.+]] = alloca i[[SZ]] 743 // CHECK: [[LOCAL_B:%.+]] = alloca [10 x i32]* 744 // CHECK-DAG: store i[[SZ]] [[ARG_A:%.+]], i[[SZ]]* [[LOCAL_A]] 745 // CHECK-DAG: store i[[SZ]] [[ARG_AA:%.+]], i[[SZ]]* [[LOCAL_AA]] 746 // CHECK-DAG: store i[[SZ]] [[ARG_AAA:%.+]], i[[SZ]]* [[LOCAL_AAA]] 747 // CHECK-DAG: store [10 x i32]* [[ARG_B:%.+]], [10 x i32]** [[LOCAL_B]] 748 // Store captures in the context. 749 // CHECK-64-DAG:[[CONV_AP:%.+]] = bitcast i[[SZ]]* [[LOCAL_A]] to i32* 750 // CHECK-DAG: [[CONV_AAP:%.+]] = bitcast i[[SZ]]* [[LOCAL_AA]] to i16* 751 // CHECK-DAG: [[CONV_AAAP:%.+]] = bitcast i[[SZ]]* [[LOCAL_AAA]] to i8* 752 // CHECK-DAG: [[REF_B:%.+]] = load [10 x i32]*, [10 x i32]** [[LOCAL_B]], 753 754 // CHECK: define internal void [[HVT5]] 755 // Create local storage for each capture. 756 // CHECK: [[LOCAL_A:%.+]] = alloca i[[SZ]] 757 // CHECK: [[LOCAL_AA:%.+]] = alloca i[[SZ]] 758 // CHECK: [[LOCAL_B:%.+]] = alloca [10 x i32]* 759 // CHECK-DAG: store i[[SZ]] [[ARG_A:%.+]], i[[SZ]]* [[LOCAL_A]] 760 // CHECK-DAG: store i[[SZ]] [[ARG_AA:%.+]], i[[SZ]]* [[LOCAL_AA]] 761 // CHECK-DAG: store [10 x i32]* [[ARG_B:%.+]], [10 x i32]** [[LOCAL_B]] 762 // Store captures in the context. 763 // CHECK-64-DAG:[[CONV_AP:%.+]] = bitcast i[[SZ]]* [[LOCAL_A]] to i32* 764 // CHECK-DAG: [[CONV_AAP:%.+]] = bitcast i[[SZ]]* [[LOCAL_AA]] to i16* 765 // CHECK-DAG: [[REF_B:%.+]] = load [10 x i32]*, [10 x i32]** [[LOCAL_B]], 766 767 // OMP45-NOT: !{!"llvm.loop.vectorize.enable", i1 false} 768 // TOMP45-NOT: !{!"llvm.loop.vectorize.enable", i1 false} 769 // OMP50: !{!"llvm.loop.vectorize.enable", i1 false} 770 // TOMP50: !{!"llvm.loop.vectorize.enable", i1 false} 771 772 #endif 773