ROCm/composable_kernel
1
0
Fork 0
mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-24 23:05:54 +00:00
Code Issues Packages Projects Releases Wiki Activity
3,013 Commits 754 Branches 38 Tags
409a7d8edb7270d786687f6692eb777cb7889ee7
Commit Graph

15 Commits

Author SHA1 Message Date
Sami Remes
0033748c62 revert custom ldstile, should be able to use the regular ones 2026-01-28 10:37:13 -05:00
Sami Remes
30d4c25d5a use PackedSize in slicing 2026-01-27 13:01:54 -05:00
Sami Remes
f62cc5415f current state of pipeline 2026-01-27 12:56:24 -05:00
Sami Remes
70c7fcda43 WIP: debugging... 2026-01-26 11:33:45 -05:00
Sami Remes
d2a7c2f041 compiles again using get_y_sliced_thread_data in warpgemm loop 2026-01-23 11:01:43 -05:00
Sami Remes
f09e10936d fixed vector load siz for fp4 2026-01-16 12:04:34 -05:00
Sami Remes
16ca5cb532 WIP 2026-01-16 08:22:11 -05:00
Sami Remes
f6f9931541 WIP 2026-01-14 12:07:26 -05:00
Sami Remes
edd11c9852 Extend comp async pipeline with scales 2026-01-13 06:46:28 -05:00
Sami Remes
f944bc03fa Extend comp async pipeline with scales 2026-01-13 05:47:55 -05:00
Sami Remes
ec1a069a60 Use simpler layout for scales. 2026-01-12 11:03:27 -05:00
Sami Remes
10fb184812 WIP: fixing loading logic 2025-12-19 12:38:32 -05:00
Sami Remes
86cc59e754 fix settings for example, fix some things in pipeline 2025-12-19 12:35:03 -05:00
Sami Remes
0faed29885 refactor the mx pipeline, backup the modified flatmm pipeline 2025-12-18 12:34:08 -05:00
Sami Remes
4985afb03c adap gemm_mx_kernel.hpp from flatmm, comment changes needed to mx pipeline from flatmm 2025-12-18 04:06:04 -05:00