composable_kernel

mirror of https://github.com/ROCm/composable_kernel.git synced 2026-05-03 05:01:25 +00:00

Files

Po Yen Chen 24b12d04af [CK_TILE] fmha fwd splitkv optimization for decode (seqlen_q=1) (#1789 )

* Update license year

* Add initial code to override decode problem

* Fix splitkv traits/args overriding error

* Reshape and transpose lse for decode

* Remove debug code

* Prettify example code

* Use better function name

* Add kMergeNumHeadGroupsSeqLenQ flag

Kernel user can use this switch to turn on/off optimization for
some problem sizes

* Add missing flag declarations

* Default turn off kMergeNumHeadGroupsSeqLenQ in codegen

* Group similar statements together

* Remove assumption of seqlen_q=1

* Remove kMergeNumHeadGroupsSeqLenQ from splitkv combine kernel

* Support kMergeNumHeadGroupsSeqLenQ=true in fmha splitkv kernel

* Run kMergeNumHeadGroupsSeqLenQ=true kernels when need

* Fix group mode block skip logics

* Undo changes of normal fwd kernel

* Update in GridSize() and using GridSize() for splitkv kernel (#1799)

---------

Co-authored-by: Qianfeng <qianfeng.zhang@amd.com>

2025-01-07 18:49:24 +08:00

fmha_bwd_kernel.hpp

[CK_TILE] Fix fMHA fwd MakeKargs() compilation errors (#1689 )

2024-11-25 15:30:35 +08:00

fmha_fwd_appendkv_kernel.hpp

Remove using partitioner for all fmha kernels (#1778 )

2024-12-29 14:29:56 +08:00

fmha_fwd_appendkv_tile_partitioner.hpp

[CK_TILE] Add PagedAttention kernels (#1387 )

2024-08-28 20:50:43 +08:00

fmha_fwd_kernel.hpp

Remove using partitioner for all fmha kernels (#1778 )