ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 18:17:44 +00:00
Code Issues Packages Projects Releases Wiki Activity

Merge commit 'cfe5e448dbf2d60ee22358e3d047600aca004090' into develop

Browse Source
This commit is contained in:
assistant-librarian[bot]
2025-08-27 11:11:35 +00:00
parent 3e2a8fca83
commit cca9297466
2 changed files with 2 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_bwd_weight_two_stage_xdl_cshuffle.hpp
View File

@@ -641,7 +641,7 @@ struct DeviceGroupedConvBwdWeightTwoStage_Xdl_CShuffle
// Ensure that k_batch_ does not exceed the maximum value
// for the GEMM pipeline.
const auto k_batch_max = static_cast<index_t>((gemmK - 1) / KPerBlock);
k_batch_ = std::min(k_batch_, k_batch_max);
k_batch_ = std::max(std::min(k_batch_, k_batch_max), 1);
if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
{

2
include/ck/tensor_operation/gpu/device/impl/device_grouped_conv_bwd_weight_xdl_cshuffle_v3.hpp
View File

@@ -506,7 +506,7 @@ struct DeviceGroupedConvBwdWeight_Xdl_CShuffleV3
// Ensure that k_batch_ does not exceed the maximum value
// for the GEMM pipeline.
const auto k_batch_max = static_cast<index_t>((gemmK - 1) / K0PerBlock);
k_batch_ = std::min(k_batch_, k_batch_max);
k_batch_ = std::max(std::min(k_batch_, k_batch_max), 1);
if(ck::EnvIsEnabled(CK_ENV(CK_LOGGING)))
{