diff options
Diffstat (limited to 'test/OpenMP/nvptx_teams_codegen.cpp')
| -rw-r--r-- | test/OpenMP/nvptx_teams_codegen.cpp | 44 |
1 files changed, 38 insertions, 6 deletions
diff --git a/test/OpenMP/nvptx_teams_codegen.cpp b/test/OpenMP/nvptx_teams_codegen.cpp index 2c50f037b2280..2d1ab9f063624 100644 --- a/test/OpenMP/nvptx_teams_codegen.cpp +++ b/test/OpenMP/nvptx_teams_codegen.cpp @@ -27,6 +27,14 @@ int main (int argc, char **argv) { return tmain(argv); } +// CK1: [[MEM_TY:%.+]] = type { [128 x i8] } +// CK1-DAG: [[SHARED_GLOBAL_RD:@.+]] = common addrspace(3) global [[MEM_TY]] zeroinitializer +// CK1-DAG: [[KERNEL_PTR:@.+]] = internal addrspace(3) global i8* null +// CK1-DAG: [[KERNEL_SIZE1:@.+]] = internal unnamed_addr constant i{{64|32}} 4 +// CK1-DAG: [[KERNEL_SIZE2:@.+]] = internal unnamed_addr constant i{{64|32}} {{8|4}} +// CK1-DAG: [[KERNEL_SHARED1:@.+]] = internal unnamed_addr constant i16 1 +// CK1-DAG: [[KERNEL_SHARED2:@.+]] = internal unnamed_addr constant i16 1 + // only nvptx side: do not outline teams region and do not call fork_teams // CK1: define {{.*}}void @{{[^,]+}}(i{{[0-9]+}} [[ARGC:%.+]]) // CK1: {{.+}} = alloca i{{[0-9]+}}*, @@ -36,7 +44,11 @@ int main (int argc, char **argv) { // CK1: store {{.+}} 0, {{.+}}, // CK1: store i{{[0-9]+}} [[ARGC]], i{{[0-9]+}}* [[ARGCADDR]], // CK1-64: [[CONV:%.+]] = bitcast i{{[0-9]+}}* [[ARGCADDR]] to i{{[0-9]+}}* -// CK1: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} 4, i16 0) +// CK1: [[IS_SHARED:%.+]] = load i16, i16* [[KERNEL_SHARED1]], +// CK1: [[SIZE:%.+]] = load i{{64|32}}, i{{64|32}}* [[KERNEL_SIZE1]], +// CK1: call void @__kmpc_get_team_static_memory(i16 0, i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds ([[MEM_TY]], [[MEM_TY]] addrspace(3)* [[SHARED_GLOBAL_RD]], i32 0, i32 0, i32 0) to i8*), i{{64|32}} [[SIZE]], i16 [[IS_SHARED]], i8** addrspacecast (i8* addrspace(3)* [[KERNEL_PTR]] to i8**)) +// CK1: [[KERNEL_RD:%.+]] = load i8*, i8* addrspace(3)* [[KERNEL_PTR]], +// CK1: [[GLOBALSTACK:%.+]] = getelementptr inbounds i8, i8* [[KERNEL_RD]], i{{64|32}} 0 // CK1-64: [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[CONV]] // CK1-32: [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[ARGCADDR]] // CK1: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0 @@ -53,7 +65,11 @@ int main (int argc, char **argv) { // CK1: [[ARGCADDR_PTR:%.+]] = alloca i{{.+}}***, // CK1: [[ARGCADDR:%.+]] = alloca i{{.+}}**, // CK1: store i{{.+}}** [[ARGC]], i{{.+}}*** [[ARGCADDR]] -// CK1: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} {{4|8}}, i16 0) +// CK1: [[IS_SHARED:%.+]] = load i16, i16* [[KERNEL_SHARED2]], +// CK1: [[SIZE:%.+]] = load i{{64|32}}, i{{64|32}}* [[KERNEL_SIZE2]], +// CK1: call void @__kmpc_get_team_static_memory(i16 0, i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds ([[MEM_TY]], [[MEM_TY]] addrspace(3)* [[SHARED_GLOBAL_RD]], i32 0, i32 0, i32 0) to i8*), i{{64|32}} [[SIZE]], i16 [[IS_SHARED]], i8** addrspacecast (i8* addrspace(3)* [[KERNEL_PTR]] to i8**)) +// CK1: [[KERNEL_RD:%.+]] = load i8*, i8* addrspace(3)* [[KERNEL_PTR]], +// CK1: [[GLOBALSTACK:%.+]] = getelementptr inbounds i8, i8* [[KERNEL_RD]], i{{64|32}} 0 // CK1: [[ARG:%.+]] = load i{{[0-9]+}}**, i{{[0-9]+}}*** [[ARGCADDR]] // CK1: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0 // CK1: store i{{[0-9]+}}** [[ARG]], i{{[0-9]+}}*** [[ARGCADDR]], @@ -98,6 +114,14 @@ int main (int argc, char **argv) { return tmain(argv); } +// CK2: [[MEM_TY:%.+]] = type { [128 x i8] } +// CK2-DAG: [[SHARED_GLOBAL_RD:@.+]] = common addrspace(3) global [[MEM_TY]] zeroinitializer +// CK2-DAG: [[KERNEL_PTR:@.+]] = internal addrspace(3) global i8* null +// CK2-DAG: [[KERNEL_SIZE1:@.+]] = internal unnamed_addr constant i{{64|32}} 4 +// CK2-DAG: [[KERNEL_SIZE2:@.+]] = internal unnamed_addr constant i{{64|32}} {{8|4}} +// CK2-DAG: [[KERNEL_SHARED1:@.+]] = internal unnamed_addr constant i16 1 +// CK2-DAG: [[KERNEL_SHARED2:@.+]] = internal unnamed_addr constant i16 1 + // CK2: define {{.*}}void @{{[^,]+}}(i{{[0-9]+}} [[A_IN:%.+]], i{{[0-9]+}} [[B_IN:%.+]], i{{[0-9]+}} [[ARGC_IN:.+]]) // CK2: {{.}} = alloca i{{[0-9]+}}*, // CK2: {{.}} = alloca i{{[0-9]+}}*, @@ -105,18 +129,22 @@ int main (int argc, char **argv) { // CK2: [[AADDR:%.+]] = alloca i{{[0-9]+}}, // CK2: [[BADDR:%.+]] = alloca i{{[0-9]+}}, // CK2: [[ARGCADDR:%.+]] = alloca i{{[0-9]+}}, -// CK2: {{%.+}} = call i32 @__kmpc_global_thread_num( // CK2: store i{{[0-9]+}} [[A_IN]], i{{[0-9]+}}* [[AADDR]], // CK2: store i{{[0-9]+}} [[B_IN]], i{{[0-9]+}}* [[BADDR]], // CK2: store i{{[0-9]+}} [[ARGC_IN]], i{{[0-9]+}}* [[ARGCADDR]], // CK2-64: [[ACONV:%.+]] = bitcast i64* [[AADDR]] to i32* // CK2-64: [[BCONV:%.+]] = bitcast i64* [[BADDR]] to i32* // CK2-64: [[CONV:%.+]] = bitcast i64* [[ARGCADDR]] to i32* -// CK2: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} 4, i16 0) +// CK2: [[IS_SHARED:%.+]] = load i16, i16* [[KERNEL_SHARED1]], +// CK2: [[SIZE:%.+]] = load i{{64|32}}, i{{64|32}}* [[KERNEL_SIZE1]], +// CK2: call void @__kmpc_get_team_static_memory(i16 0, i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds ([[MEM_TY]], [[MEM_TY]] addrspace(3)* [[SHARED_GLOBAL_RD]], i32 0, i32 0, i32 0) to i8*), i{{64|32}} [[SIZE]], i16 [[IS_SHARED]], i8** addrspacecast (i8* addrspace(3)* [[KERNEL_PTR]] to i8**)) +// CK2: [[KERNEL_RD:%.+]] = load i8*, i8* addrspace(3)* [[KERNEL_PTR]], +// CK2: [[GLOBALSTACK:%.+]] = getelementptr inbounds i8, i8* [[KERNEL_RD]], i{{64|32}} 0 // CK2-64: [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[CONV]] // CK2-32: [[ARG:%.+]] = load i{{[0-9]+}}, i{{[0-9]+}}* [[ARGCADDR]] // CK2: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0 // CK2: store i{{[0-9]+}} [[ARG]], i{{[0-9]+}}* [[ARGCADDR]], +// CK2: {{%.+}} = call i32 @__kmpc_global_thread_num( // CK2: store i{{[0-9]+}}* [[ARGCADDR]], i{{[0-9]+}}** [[ARGCADDR_PTR]], // CK2: [[ARGCADDR_PTR_REF:%.+]] = load i{{[0-9]+}}*, i{{[0-9]+}}** [[ARGCADDR_PTR]], // CK2: store i{{[0-9]+}} 0, i{{[0-9]+}}* [[ARGCADDR_PTR_REF]], @@ -129,14 +157,18 @@ int main (int argc, char **argv) { // CK2: [[AADDR:%.+]] = alloca i{{[0-9]+}}, // CK2: [[BADDR:%.+]] = alloca i{{[0-9]+}}, // CK2: [[ARGCADDR:%.+]] = alloca i{{[0-9]+}}**, -// CK2: {{%.+}} = call i32 @__kmpc_global_thread_num( // CK2: store i{{[0-9]+}} [[A_IN]], i{{[0-9]+}}* [[AADDR]], // CK2: store i{{[0-9]+}} [[B_IN]], i{{[0-9]+}}* [[BADDR]], // CK2: store i{{[0-9]+}}** [[ARGC]], i{{[0-9]+}}*** [[ARGCADDR]], -// CK2: call i8* @__kmpc_data_sharing_push_stack(i{{[0-9]+}} {{4|8}}, i16 0) +// CK2: [[IS_SHARED:%.+]] = load i16, i16* [[KERNEL_SHARED2]], +// CK2: [[SIZE:%.+]] = load i{{64|32}}, i{{64|32}}* [[KERNEL_SIZE2]], +// CK2: call void @__kmpc_get_team_static_memory(i16 0, i8* addrspacecast (i8 addrspace(3)* getelementptr inbounds ([[MEM_TY]], [[MEM_TY]] addrspace(3)* [[SHARED_GLOBAL_RD]], i32 0, i32 0, i32 0) to i8*), i{{64|32}} [[SIZE]], i16 [[IS_SHARED]], i8** addrspacecast (i8* addrspace(3)* [[KERNEL_PTR]] to i8**)) +// CK2: [[KERNEL_RD:%.+]] = load i8*, i8* addrspace(3)* [[KERNEL_PTR]], +// CK2: [[GLOBALSTACK:%.+]] = getelementptr inbounds i8, i8* [[KERNEL_RD]], i{{64|32}} 0 // CK2: [[ARG:%.+]] = load i{{[0-9]+}}**, i{{[0-9]+}}*** [[ARGCADDR]] // CK2: [[ARGCADDR:%.+]] = getelementptr inbounds %struct.{{.*}}, %struct.{{.*}}* %{{.*}}, i{{[0-9]+}} 0, i{{[0-9]+}} 0 // CK2: store i{{[0-9]+}}** [[ARG]], i{{[0-9]+}}*** [[ARGCADDR]], +// CK2: {{%.+}} = call i32 @__kmpc_global_thread_num( // CK2: store i{{[0-9]+}}*** [[ARGCADDR]], i{{[0-9]+}}**** [[ARGCADDR_PTR]], // CK2: [[ARGCADDR_PTR_REF:%.+]] = load i{{[0-9]+}}***, i{{[0-9]+}}**** [[ARGCADDR_PTR]], // CK2: store i{{[0-9]+}}** null, i{{[0-9]+}}*** [[ARGCADDR_PTR_REF]], |