composable_kernel

ROCm/composable_kernel

Fork 0

mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-03 13:48:30 +00:00

Commit Graph

Author	SHA1	Message	Date
Enrico Degregori	d559ec00a8	[rocm-libraries] ROCm/rocm-libraries#8554 (commit be9af54) refactor(ck): mx gemm kernel unification ## Motivation CK tile currently has two separate MX GEMM kernels for gfx950 and gfx1250. This pull request refactors and modernizes the MX GEMM kernel and example to use new scale tensor handling, improved kernel argument structures, and updated pipeline and kernel APIs. The changes simplify the interface and improve type safety. JIRA ID ROCM-26313 ## Technical Details - Add support for gfx950 in MX GEMM kernel for gfx1250 and remove unused kernel - Unify comp async pipeline for GEMM and MX GEMM - Unify eight waves pipeline for GEMM and MX GEMM - Move preshuffle MX GEMM pipeline to gemm ops and remove gemm_mx ops - Unify testing framework for MX GEMM - Add gfx950 tests for grouped MX GEMM ## Test Plan - `test_mx_gemm_async.cpp` for MX GEMM on gfx950 - `test_mx_grouped_gemm_comp_async.cpp` for grouped MX GEMM on gfx950 ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.	2026-07-01 08:21:02 +00:00
Enrico Degregori	e75076c826	[rocm-libraries] ROCm/rocm-libraries#8310 (commit 003bc6b) [CK Tile] Fix assert usage MX GEMM ## Motivation See issue https://github.com/ROCm/rocm-libraries/issues/8223 ## Technical Details - Use `std::runtime_error` in `mx_processing.hpp` - Use `static_assert` in `tensor_shuffle_utils.hpp` ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.	2026-06-12 11:42:38 +00:00
Enrico Degregori	bdd7a8333d	[rocm-libraries] ROCm/rocm-libraries#6672 (commit bda3f97) [CK Tile] PermuteN support MX GEMM ## Motivation Add PermuteN support to preshuffle MX GEMM ## Technical Details - Modify `shuffle_b_permuteN` to support MX preshuffled layout - Add `preShuffleScalePermuteN` with same functionality of `preShuffleScale` but layout consistent with PermuteN - Include MX pre-processing functions in the library ## Test Plan Add test configuration for permuteN with preshuffle (both FP4 and FP8) ## Submission Checklist - [x] Look over the contributing guidelines at https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests. Co-authored-by: Cong Ma <congma13@amd.com>	2026-06-05 03:04:43 +00:00

Author

SHA1

Message

Date

Enrico Degregori

d559ec00a8

[rocm-libraries] ROCm/rocm-libraries#8554 (commit be9af54)

refactor(ck): mx gemm kernel unification

## Motivation

CK tile currently has two separate MX GEMM kernels for gfx950 and
gfx1250. This pull request refactors and modernizes the MX GEMM kernel
and example to use new scale tensor handling, improved kernel argument
structures, and updated pipeline and kernel APIs. The changes simplify
the interface and improve type safety.

JIRA ID ROCM-26313

## Technical Details

- Add support for gfx950 in MX GEMM kernel for gfx1250 and remove unused
kernel
 - Unify comp async pipeline for GEMM and MX GEMM
 - Unify eight waves pipeline for GEMM and MX GEMM
 - Move preshuffle MX GEMM pipeline to gemm ops and remove gemm_mx ops
 - Unify testing framework for MX GEMM
 - Add gfx950 tests for grouped MX GEMM

## Test Plan

 - `test_mx_gemm_async.cpp` for MX GEMM on gfx950
 - `test_mx_grouped_gemm_comp_async.cpp` for grouped MX GEMM on gfx950

## Submission Checklist

- [x] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

2026-07-01 08:21:02 +00:00

Enrico Degregori

e75076c826

[rocm-libraries] ROCm/rocm-libraries#8310 (commit 003bc6b)

[CK Tile] Fix assert usage MX GEMM

## Motivation

See issue https://github.com/ROCm/rocm-libraries/issues/8223

## Technical Details

 - Use `std::runtime_error` in `mx_processing.hpp`
 - Use `static_assert` in `tensor_shuffle_utils.hpp`

## Submission Checklist

- [x] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

2026-06-12 11:42:38 +00:00

Enrico Degregori

bdd7a8333d

[rocm-libraries] ROCm/rocm-libraries#6672 (commit bda3f97)

[CK Tile] PermuteN support MX GEMM

## Motivation

Add PermuteN support to preshuffle MX GEMM

## Technical Details

 - Modify `shuffle_b_permuteN` to support MX preshuffled layout
- Add `preShuffleScalePermuteN` with same functionality of
`preShuffleScale` but layout consistent with PermuteN
 - Include MX pre-processing functions in the library

## Test Plan

Add test configuration for permuteN with preshuffle (both FP4 and FP8)

## Submission Checklist

- [x] Look over the contributing guidelines at
https://github.com/ROCm/ROCm/blob/develop/CONTRIBUTING.md#pull-requests.

Co-authored-by: Cong Ma <congma13@amd.com>

2026-06-05 03:04:43 +00:00

3 Commits