mirror of https://github.com/ROCm/composable_kernel.git synced 2026-07-17 00:58:44 +00:00

Files

Qianfeng 8632221814 [CK_TILE] Add fmha fwd headdim96 support (#1608 )

* Add ceil_to_qualified_tile_length()

* Rename kK0BlockLength to kQKHeaddim

* Add kSubQKHeaddim concept to support headdim96

* Fix in math.hpp to avoid using __half interfaces

* Add LdsBufferSequence instance for headdim96

* Update in fmha_fwd/fmha_fwd_splitkv codegen to support hd96 testing

* Disable hd96 instance generation in codegen fmha_fwd and fmha_fwd_splitkv to save compiling time

* Reformat one file

* Fix text alignment in fmha_fwd_splitkv.py

---------

Co-authored-by: Po Yen Chen <PoYen.Chen@amd.com>

2024-10-30 14:03:16 +08:00

algorithm

[CK_TILE] add scatter_gather (#1609 )

2024-10-29 18:19:29 +08:00

arch

topk_softmax (#1592 )

2024-10-26 23:52:49 +08:00

container

topk_softmax (#1592 )

2024-10-26 23:52:49 +08:00

numeric

[CK_TILE] Add fmha fwd headdim96 support (#1608 )

2024-10-30 14:03:16 +08:00

tensor

topk_softmax (#1592 )

2024-10-26 23:52:49 +08:00

utility

topk_softmax (#1592 )

2024-10-26 23:52:49 +08:00

config.hpp

topk_softmax (#1592 )

2024-10-26 23:52:49 +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