diff options
| author | Alexey Bataev <a.bataev@hotmail.com> | 2018-10-09 14:49:00 +0000 |
|---|---|---|
| committer | Alexey Bataev <a.bataev@hotmail.com> | 2018-10-09 14:49:00 +0000 |
| commit | 9ea3c385973c395c4798145f71ff3ae34ccaabeb (patch) | |
| tree | 0a2f511ca7fb094143065d6ee32b2976e7079a80 /clang/test/OpenMP/nvptx_data_sharing.cpp | |
| parent | f5fac1826a867d5e65310456670cc1d5a495ed02 (diff) | |
| download | bcm5719-llvm-9ea3c385973c395c4798145f71ff3ae34ccaabeb.tar.gz bcm5719-llvm-9ea3c385973c395c4798145f71ff3ae34ccaabeb.zip | |
[OPENMP][NVPTX] Support memory coalescing for globalized variables.
Added support for memory coalescing for better performance for
globalized variables. From now on all the globalized variables are
represented as arrays of 32 elements and each thread accesses these
elements using `tid & 31` as index.
llvm-svn: 344049
Diffstat (limited to 'clang/test/OpenMP/nvptx_data_sharing.cpp')
| -rw-r--r-- | clang/test/OpenMP/nvptx_data_sharing.cpp | 12 |
1 files changed, 9 insertions, 3 deletions
diff --git a/clang/test/OpenMP/nvptx_data_sharing.cpp b/clang/test/OpenMP/nvptx_data_sharing.cpp index 0acb1199152..83206769243 100644 --- a/clang/test/OpenMP/nvptx_data_sharing.cpp +++ b/clang/test/OpenMP/nvptx_data_sharing.cpp @@ -39,10 +39,16 @@ void test_ds(){ // CK1: [[SHAREDARGS2:%.+]] = alloca i8** // CK1: call void @__kmpc_kernel_init // CK1: call void @__kmpc_data_sharing_init_stack -// CK1: [[GLOBALSTACK:%.+]] = call i8* @__kmpc_data_sharing_push_stack(i64 8, i16 0) +// CK1: [[GLOBALSTACK:%.+]] = call i8* @__kmpc_data_sharing_push_stack(i64 256, i16 0) // CK1: [[GLOBALSTACK2:%.+]] = bitcast i8* [[GLOBALSTACK]] to %struct._globalized_locals_ty* -// CK1: [[A:%.+]] = getelementptr inbounds %struct._globalized_locals_ty, %struct._globalized_locals_ty* [[GLOBALSTACK2]], i32 0, i32 0 -// CK1: [[B:%.+]] = getelementptr inbounds %struct._globalized_locals_ty, %struct._globalized_locals_ty* [[GLOBALSTACK2]], i32 0, i32 1 +// CK1: [[A_ARR:%.+]] = getelementptr inbounds %struct._globalized_locals_ty, %struct._globalized_locals_ty* [[GLOBALSTACK2]], i32 0, i32 0 +// CK1: [[TID:%.+]] = call i32 @llvm.nvvm.read.ptx.sreg.tid.x() +// CK1: [[LID:%.+]] = and i32 [[TID]], 31 +// CK1: [[A:%.+]] = getelementptr inbounds [32 x i32], [32 x i32]* [[A_ARR]], i32 0, i32 [[LID]] +// CK1: [[B_ARR:%.+]] = getelementptr inbounds %struct._globalized_locals_ty, %struct._globalized_locals_ty* [[GLOBALSTACK2]], i32 0, i32 1 +// CK1: [[TID:%.+]] = call i32 @llvm.nvvm.read.ptx.sreg.tid.x() +// CK1: [[LID:%.+]] = and i32 [[TID]], 31 +// CK1: [[B:%.+]] = getelementptr inbounds [32 x i32], [32 x i32]* [[B_ARR]], i32 0, i32 [[LID]] // CK1: store i32 10, i32* [[A]] // CK1: call void @__kmpc_kernel_prepare_parallel({{.*}}, i16 1) // CK1: call void @__kmpc_begin_sharing_variables(i8*** [[SHAREDARGS1]], i64 1) |
