ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-15 02:27:57 +00:00
Code Issues Packages Projects Releases Wiki Activity

Support fp64 contraction on gfx94x. (#1029)

Browse Source 
* enable contraction fp64 on gfx94*

* fix the logic

[ROCm/composable_kernel commit: ce52621123]
This commit is contained in:
Illia Silin
2023-11-08 15:03:18 -08:00
committed by GitHub
parent be753a8db1
commit b931cd3cee
1 changed files with 3 additions and 1 deletions
Download Patch File Download Diff File Expand all files Collapse all files

4
include/ck/tensor_operation/gpu/device/impl/device_contraction_multiple_d_xdl_cshuffle.hpp
View File

@@ -595,7 +595,9 @@ struct DeviceContractionMultipleD_Xdl_CShuffle
return false;
}
if(ck::get_device_name() != "gfx90a" && std::is_same<ADataType, double>::value)
if(ck::get_device_name() != "gfx90a" && ck::get_device_name() != "gfx940" &&
ck::get_device_name() != "gfx941" && ck::get_device_name() != "gfx942" &&
std::is_same<ADataType, double>::value)
{
return false;
}