ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-18 20:09:25 +00:00
Code Issues Packages Projects Releases Wiki Activity

ck: add CK_USE_GFX950 macro (#3636)

Browse Source
This commit is contained in:
yinglu
2026-01-27 03:38:45 +08:00
committed by GitHub
parent b8751e505d
commit 8942a19d5e
4 changed files with 7 additions and 9 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_bwd_data/device_grouped_conv_bwd_data_xdl_instance.hpp
View File

@@ -376,7 +376,7 @@ using device_grouped_conv_bwd_data_xdl_f32_optimized_loads_instances =
// clang-format on
>;
#if defined(__gfx950__)
#if defined(CK_USE_GFX950)
constexpr auto _k_per_block = 32;
#else
constexpr auto _k_per_block = 16;

2
library/include/ck/library/tensor_operation_instance/gpu/grouped_conv_fwd/device_grouped_conv_fwd_xdl_merged_groups_instance.hpp
View File

@@ -147,7 +147,7 @@ using device_grouped_conv_fwd_xdl_merged_groups_f32_instances = std::tuple<
// clang-format on
>;
#if defined(__gfx950__)
#if defined(CK_USE_GFX950)
constexpr auto _k_per_block = 32;
#else
constexpr auto _k_per_block = 16;