composable_kernel

mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-14 18:17:44 +00:00

Files

Po Yen Chen 8cc793d26a [CK_TILE] Optimize fmha splitkv & splitkv combine kernels (#1577 )

* Use smaller width for lse_accum dist tensor

* Update pipeline comment

* Fix wrong distribution for lse_accum

* Remove duplicate dim in lse_accum dist encoding

* Decide fmha splitkv combine kernel kBlockSize by kM0

* Remove assumption of MPerThread=1

* Add log<4> & log<8> specialization

* Enlarge occupancy array

* Fix vector size for small tile

* Add support for kMaxSplits=8

* Re-format gemm.hpp

* Use 16x16x16 warp gemm for fwd_splitkv

* Centralize policy code changes

* Leave fp8/bf8 tile settings unchanged

[ROCm/composable_kernel commit: 95e722a3b3]

2024-10-21 10:52:11 +08:00

01_fmha

[CK_TILE] Optimize fmha splitkv & splitkv combine kernels (#1577 )

2024-10-21 10:52:11 +08:00

02_layernorm2d

[CK_TILE] Update example README files & fix script compatibility issue (#1548 )

2024-10-08 10:45:12 +08:00

03_gemm

[CK_TILE] Add block universal gemm pipeline policy (#1557 )

2024-10-15 13:53:41 +02:00

04_img2col

[CK_TILE] Update example README files & fix script compatibility issue (#1548 )