composable_kernel

mirror of https://github.com/ROCm/composable_kernel.git synced 2026-04-19 22:39:03 +00:00

Files

Thomas Ning b159841a06 Fix the gfx950 numerical errors (#2911 )

* Update grouped_gemm example and pipeline

* find the root cause error in did not enable the transpose in gfx950 correctly

* Fix v3 pipeline, row and col major

* Disable f8 datatype tests, it fails on gfx950

* fix the abd test by clear the runtime argument unsupported

---------

Co-authored-by: AviralGoelAMD <aviral.goel@amd.com>
Co-authored-by: Mateusz Ozga <mateusz.ozga@amd.com>

2025-09-23 22:54:52 -07:00

batched_gemm_kernel.hpp

Revert "[CK-Tile] Add the API to load SGPR (#2878 )" (#2904 )

2025-09-23 14:33:51 -07:00

gemm_kernel.hpp

[CK_TILE] Multiple-ABD GEMM example (#2788 )

2025-09-19 01:14:11 +02:00

gemm_multi_abd_kernel.hpp

Fix the gfx950 numerical errors (#2911 )

2025-09-23 22:54:52 -07:00

gemm_multi_d_kernel.hpp

[CK_TILE] Multiple-ABD GEMM example (#2788 )