1 // Test target codegen - host bc file has to be created first.
2 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-cuda-mode -x c++ -triple powerpc64le-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm-bc %s -o %t-ppc-host.bc
3 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-cuda-mode -x c++ -triple nvptx64-unknown-unknown -fopenmp-targets=nvptx64-nvidia-cuda -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-ppc-host.bc -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-64
4 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-cuda-mode -x c++ -triple i386-unknown-unknown -fopenmp-targets=nvptx-nvidia-cuda -emit-llvm-bc %s -o %t-x86-host.bc
5 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-cuda-mode -x c++ -triple nvptx-unknown-unknown -fopenmp-targets=nvptx-nvidia-cuda -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
6 // RUN: %clang_cc1 -verify -fopenmp -fopenmp-version=45 -fopenmp-cuda-mode -fexceptions -fcxx-exceptions -x c++ -triple nvptx-unknown-unknown -fopenmp-targets=nvptx-nvidia-cuda -emit-llvm %s -fopenmp-is-device -fopenmp-host-ir-file-path %t-x86-host.bc -o - | FileCheck %s --check-prefix CHECK --check-prefix CHECK-32
7 // expected-no-diagnostics
8 #ifndef HEADER
9 #define HEADER
10
11 // Check that the execution mode of all 2 target regions on the gpu is set to NonSPMD Mode.
12 // CHECK-DAG: {{@__omp_offloading_.+l25}}_exec_mode = weak constant i8 0
13 // CHECK-DAG: {{@__omp_offloading_.+l30}}_exec_mode = weak constant i8 0
14 // CHECK-DAG: {{@__omp_offloading_.+l35}}_exec_mode = weak constant i8 0
15 // CHECK-DAG: {{@__omp_offloading_.+l40}}_exec_mode = weak constant i8 0
16
17 #define N 1000
18
19 template<typename tx>
ftemplate(int n)20 tx ftemplate(int n) {
21 tx a[N];
22 short aa[N];
23 tx b[10];
24
25 #pragma omp target simd
26 for(int i = 0; i < n; i++) {
27 a[i] = 1;
28 }
29
30 #pragma omp target simd
31 for (int i = 0; i < n; i++) {
32 aa[i] += 1;
33 }
34
35 #pragma omp target simd
36 for(int i = 0; i < 10; i++) {
37 b[i] += 1;
38 }
39
40 #pragma omp target simd reduction(+:n)
41 for(int i = 0; i < 10; i++) {
42 b[i] += 1;
43 }
44
45 return a[0];
46 }
47
bar(int n)48 int bar(int n){
49 int a = 0;
50
51 a += ftemplate<int>(n);
52
53 return a;
54 }
55
56 // CHECK-LABEL: define {{.*}}void {{@__omp_offloading_.+l25}}(
57 // CHECK: call void @__kmpc_spmd_kernel_init(i32 %{{.+}}, i16 0, i16 0)
58 // CHECK-NOT: call void @__kmpc_for_static_init
59 // CHECK-NOT: call void @__kmpc_for_static_fini
60 // CHECK: call void @__kmpc_spmd_kernel_deinit_v2(i16 0)
61 // CHECK: ret void
62
63 // CHECK-LABEL: define {{.*}}void {{@__omp_offloading_.+l30}}(
64 // CHECK: call void @__kmpc_spmd_kernel_init(i32 %{{.+}}, i16 0, i16 0)
65 // CHECK-NOT: call void @__kmpc_for_static_init
66 // CHECK-NOT: call void @__kmpc_for_static_fini
67 // CHECK: call void @__kmpc_spmd_kernel_deinit_v2(i16 0)
68 // CHECK: ret void
69
70 // CHECK-LABEL: define {{.*}}void {{@__omp_offloading_.+l35}}(
71 // CHECK: call void @__kmpc_spmd_kernel_init(i32 %{{.+}}, i16 0, i16 0)
72 // CHECK-NOT: call void @__kmpc_for_static_init
73 // CHECK-NOT: call void @__kmpc_for_static_fini
74 // CHECK: call void @__kmpc_spmd_kernel_deinit_v2(i16 0)
75 // CHECK: ret void
76
77 // CHECK-LABEL: define {{.*}}void {{@__omp_offloading_.+l40}}(
78 // CHECK: call void @__kmpc_spmd_kernel_init(i32 %{{.+}}, i16 0, i16 0)
79 // CHECK-NOT: call void @__kmpc_for_static_init
80 // CHECK-NOT: call void @__kmpc_for_static_fini
81 // CHECK-NOT: call i32 @__kmpc_nvptx_simd_reduce_nowait(
82 // CHECK-NOT: call void @__kmpc_nvptx_end_reduce_nowait(
83 // CHECK: call void @__kmpc_spmd_kernel_deinit_v2(i16 0)
84 // CHECK: ret void
85
86
87 #endif
88