ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-02 12:41:26 +00:00
Code Issues Packages Projects Releases Wiki Activity

Flatmm merge (#2168)

Browse Source 
* sync with function interface of cshuffleepiloge,fix flatmm build fail

* move code from solin/flatmm which add mfma16*16*32fp8 and optimize flatmm

---------

Co-authored-by: solin <bingzhou@amd.com>
This commit is contained in:
BingYuan.Zhou
2025-05-08 12:59:57 +08:00
committed by GitHub
parent c7b8e86e34
commit 6a3960c1e1
11 changed files with 552 additions and 192 deletions
Download Patch File Download Diff File Expand all files Collapse all files

8
include/ck_tile/ops/gemm/warp/warp_gemm.hpp
View File

@@ -193,6 +193,14 @@ using WarpGemmMfmaBf16Bf16F32M64N4K16 = WarpGemmImpl<WarpGemmAtrributeMfmaIterat
using WarpGemmMfma_f32_32x32x16_fp8_fp8 = WarpGemmImpl<
WarpGemmAtrributeMfma<WarpGemmAttributeMfmaImpl_f32_32x32x16_fp8_fp8<WGAttrCtlEnum::Default_>>>;
using WarpGemmMfma_f32_32x32x32_fp8_fp8 = WarpGemmImpl<WarpGemmAtrributeMfmaIterateK<
WarpGemmAttributeMfmaImpl_f32_32x32x16_fp8_fp8<WGAttrCtlEnum::Default_>,
2>>;
using WarpGemmMfma_f32_32x32x32_bf8_bf8 = WarpGemmImpl<WarpGemmAtrributeMfmaIterateK<
WarpGemmAttributeMfmaImpl_f32_32x32x16_bf8_bf8<WGAttrCtlEnum::Default_>,
2>>;
using WarpGemmMfma_f32_32x32x16_fp8_bf8 = WarpGemmImpl<
WarpGemmAtrributeMfma<WarpGemmAttributeMfmaImpl_f32_32x32x16_fp8_bf8<WGAttrCtlEnum::Default_>>>;