Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

perf_test/blas/blas3: Disable simd verify for cuda/10.2.2 #1093

Merged
merged 1 commit into from
Sep 2, 2021
Merged

perf_test/blas/blas3: Disable simd verify for cuda/10.2.2 #1093

Changes from all commits
Commits
Show all changes
1 commit
Select commit
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
10 changes: 10 additions & 0 deletions perf_test/blas/blas3/KokkosBlas3_gemm_perf_test.hpp
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -1504,6 +1504,12 @@ template <class dstViewType>
static inline void __gemm_copy_simd_view_to_3d_view(gemm_simd_args_t src,
dstViewType dst,
options_t options) {
// clang-format off
// Related issue: https://github.com/kokkos/kokkos-kernels/issues/998
// CUDA VERSION 10.2.2 generates a compiler error:
// KokkosBlas3_gemm_perf_test.hpp: error: ‘h_subview_type_2d’ was not declared in this scope
// clang-format on
#if (CUDA_VERSION != 10020)
using dst_scalar_type = typename dstViewType::value_type;
using src_scalar_type = typename view_type_5d::value_type;
size_t remainder, vector_batch_size, simd_batch_size, last_batch;
Expand Down Expand Up @@ -1631,6 +1637,10 @@ static inline void __gemm_copy_simd_view_to_3d_view(gemm_simd_args_t src,
out:
Kokkos::deep_copy(dst, h_dst);
Kokkos::fence();
#else
Kokkos::abort(
"Cannot perform simd verification with cuda/10.2.2, rerun with -v 0");
#endif // #if (CUDA_VERSION != 10020)
}

/**
Expand Down