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

Files

Andriy Roshchenko 78f8490cb6 Add OCP FP8 support in CK_TILE (#1829 )

* Add OCP FP8 to CK_TILE

* Validate OCP FP8 in FMHA FWD under VALID=1

[ROCm/composable_kernel commit: 35aebe5936]

2025-01-27 11:59:49 -07:00

algorithm

[CK-Tile] Enable vectorized reads on all layouts & improve perf. (#1835 )

2025-01-27 16:37:19 +01:00

arch

CK-Tile Grouped GEMM refactor and post PR fixes (#1756 )

2025-01-21 21:06:10 +01:00

container

[CK-Tile] Enable vectorized reads on all layouts & improve perf. (#1835 )

2025-01-27 16:37:19 +01:00

numeric

Add OCP FP8 support in CK_TILE (#1829 )

2025-01-27 11:59:49 -07:00

tensor

[CK-Tile] Enable vectorized reads on all layouts & improve perf. (#1835 )

2025-01-27 16:37:19 +01:00

utility

add fp8 as dst (#1830 )

2025-01-22 17:34:27 +08:00

config.hpp

Add OCP FP8 support in CK_TILE (#1829 )

2025-01-27 11:59:49 -07:00

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