mirror of https://github.com/ROCm/composable_kernel.git synced 2026-06-30 03:37:38 +00:00

Files

Jeff Huang 93699b81f5 Gate FMHA padding features behind build flag

- add CK_TILE_FMHA_ENABLE_SEQLEN_PADDING option in top-level CMake and propagate
  compile definitions and environment export
- wire the flag through fmha example CMake, scripts, and tests so padding
  codepaths are only built/executed when enabled
- guard invoke sites (smoke tests, benchmarks, unit harness) and Kargs
  initialization with the new macro or runtime env checks to avoid touching
  cu_seqlen*/padding fields when disabled

2025-10-15 08:18:48 +08:00

algorithm

refactor: use snake_case naming in ck_tile/core components (#2766 )

2025-09-03 09:34:11 +02:00

arch

update s_barrier's logic in gfx12 architecture (#3003 )

2025-10-14 08:49:34 -07:00

container

[CK_TILE] Add permuteN optimization to remove lds operation in c_shuffle (#2764 )

2025-09-08 22:02:48 -07:00

numeric

[CK_TILE] fix pk_fp4 compilation for non-gfx950 GPUs (#2983 )

2025-10-09 07:43:41 -07:00

tensor

Fix for Add the API to load SGPR (#2913 )

2025-09-25 10:32:42 -07:00

utility

[CK_TILE] Support f32 in FMHA (fwd and bwd) (#2836 )

2025-09-27 18:03:48 +05:00

config.hpp

Gate FMHA padding features behind build flag

2025-10-15 08:18:48 +08:00

README.md

introducing ck_tile! (#1216 )

2024-04-15 19:27:12 -05:00

README.md

ck_tile/core

ck_tile/core contains every basic functions and structures to create a GPU kernel using ck_tile. User should only include ck_tile/core.hpp this single header to use all the functionality. Everything is under ck_tile namespace. The coding style under this folder should be similar to std (snake_case for structure/function, Camel for template types...)

algorithm/
    coordinate transform and some other reusable algorithm
arch/
    contains some basic device building block like mma, buffer addressing, etc...
container/
    contains basic container data structure, array/sequence/tuple/...
numeric/
    data type, and data type related math
tensor/
    tensor descriptors and tile level API
utility/
    other utility function for both host/device