Bartłomiej Kocot
5705af0aad
[CK TILE] GEMM with packed i4 ( #1885 )
...
* [CK TILE] GEMM with packed i4
* Fixes
* fixes
* fixes
* fixes
[ROCm/composable_kernel commit: 4d9973ec8e
2025-02-20 09:59:49 +01:00
carlushuang
21264b4e60
[CK_TILE] Fix mock token id, support g1u1/g1u0 through same inline code block ( #1808 )
...
* fix mock token id
* prepare host for g1u1
* reformat inline-asm
* restructure uk_0
* restructure gate_up
* done
* change default to init=1
* update readme
* fix a bug in interleave pipeline
* rcp for silu
[ROCm/composable_kernel commit: 1ff50e78c6
2025-01-16 17:51:10 +08:00
carlushuang
74b0db75f7
[CK_TILE] fused-moe first version ( #1634 )
...
* moe pipeline
* update code
* compile OK
* update
* update cpu reference
* update pipeline_gemm0
* compiler ok
* update pipeline
* rename to ex pipeline
* block-asm
* update
* update
* update first gemm ok
* compute correct
* update file structure
* update README
* update
* update
* update code
* update API
* return unsupport case
* add comment
* update readme
* update
* uncomment
* update
* fix build err
---------
Co-authored-by: valarLip <340077269@qq.com >
[ROCm/composable_kernel commit: 440e28b08f
2024-11-26 11:14:56 +08:00
carlushuang
ea3af1dfbc
topk_softmax ( #1592 )
...
* topk_softmax
* remove some file
* fix atomix linear_offset
* address various comment, and change sfc get_index api to static(tuple)
[ROCm/composable_kernel commit: b098b71b05
2024-10-26 23:52:49 +08:00
