ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-05 22:22:27 +00:00
Code Issues Packages Projects Releases Wiki Activity

MX GEMM - Add MX BF8 example (#2071)

Browse Source 
* Add MX GEMM example for MX BF8

* Verified MX FP8 with 16x16x128 scale builtin

* Verify MX BF8 GEMM with BF16 output
This commit is contained in:
Andriy Roshchenko
2025-04-16 15:25:02 -06:00
committed by GitHub
parent 3bb62f16cd
commit da54464cce
5 changed files with 139 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
include/ck/tensor_operation/gpu/warp/xdlops_gemm.hpp
View File

@@ -1141,6 +1141,12 @@ struct MfmaSelector
return MfmaInstr::mfma_scale_f32_16x16x128f8f6f4;
}
template <>
constexpr auto GetMfma<bf8_t, 16, 16, bf8_t, false, true>()
{
return MfmaInstr::mfma_scale_f32_16x16x128f8f6f4;
}
template <>
constexpr auto GetMfma<bf8_t, 32, 32>()
{