Po Yen Chen cf2d635ea2 [CK_TILE] Fix incorrect computation of group mode PagedAttention (#1688) ...

* Allow getting batch size from splitkv tile partitioner

* Fix wrong paged-kvcache impl for group mode

* Fix wrong example code for page-kvcache

* Undo changes in fmha_fwd.cpp

* Always use 2D block table

* Add is_gappy kernel argument for paged-kvcache

The is_gappy argument is used for differentiating seqstart_k_ptr usage
in flash-attention & xformers

* Remove out-of-date comments

* Remove no-longer used method

* Fix wrong # page-block calculation

* Fix wrong comment

---------

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

2024-11-26 20:37:54 +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

[CK_TILE] Add PagedAttention kernels (#1387)

2024-08-28 20:50:43 +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

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

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

fmha_fwd_splitkv_combine_kernel.hpp

[CK_TILE] Change output accum tensor layout of fmha fwd split-kv & combine kernels (#1527)

2024-10-01 22:13:52 +08:00

fmha_fwd_splitkv_combine_tile_partitioner.hpp

[CK_TILE] Change output accum tensor layout of fmha fwd split-kv & combine kernels (#1527)

2024-10-01 22:13:52 +08:00

fmha_fwd_splitkv_kernel.hpp

[CK_TILE] Fix incorrect computation of group mode PagedAttention (#1688)

2024-11-26 20:37:54 +08:00

fmha_fwd_splitkv_tile_partitioner.hpp

[CK_TILE] Fix incorrect computation of group mode PagedAttention (#1688)

2024-11-26 20:37:54 +08:00

fmha_fwd_tile_partitioner.hpp

CK Tile FA Training kernels (#1286)

2024-06-04 13:12:45 -05:00