ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-02 20:51:23 +00:00
Code Issues Packages Projects Releases Wiki Activity

Fixed f8_gemm NaN (#975)

Browse Source 
* workaround nan problem by changing output to fp16

* enable f8/bf8 gemm tests on MI200

* workaround f16 to f8 conversion

---------

Co-authored-by: Jing Zhang <jizha@amd.com>
This commit is contained in:
zjing14
2023-10-10 10:30:26 -05:00
committed by GitHub
parent 5913609168
commit ac9595a9f1
5 changed files with 18 additions and 19 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
include/ck/tensor_operation/gpu/element/unary_element_wise_operation.hpp
View File

@@ -173,8 +173,7 @@ struct PassThrough
template <>
__host__ __device__ void operator()<bf8_t, half_t>(bf8_t& y, const half_t& x) const
{
// to-do: fix half_t to bf8_t convert
y = ck::type_convert<bf8_t>(ck::type_convert<float>(x));
y = ck::type_convert<bf8_t>(x);
}
#endif
};