Po Yen Chen a1c07e8d91 [CK_TILE] Change output accum tensor layout of fmha fwd split-kv & combine kernels (#1527) ...

* Use same layout for o_acc and o tensor

* Use better param names in partitioner

* Remove redundant kargs 'max_seqlen_q'

* Use better param names in splitkv kernel

* Add comment for additional kernel arguments

* Sync empty loop early return logics between pipelines

* Pass more arguments to cmake in scripts

* Align backslashes

* Fix wrong o_acc tensor view strides

* Change o_acc layout if o_perm=0

* Handle whole row masked via attn_bias

* Use use vector width = 1 for o_acc

* Use more even split sizes

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

..

block_attention_bias_enum.hpp

[CK_TILE] support alibi (#1269)

2024-05-07 22:32:54 +08:00

block_dropout.hpp

[CK_TILE] FA bwd kernels optimization (#1397)

2024-08-16 13:40:10 -07:00

block_masking.hpp

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

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

block_position_encoding.hpp

[CK_TILE] Add PagedAttention kernels (#1387)

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

block_rotary_embedding.hpp

[CK_TILE] Add PagedAttention kernels (#1387)

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

page_block_navigator.hpp

[CK_TILE] Add PagedAttention kernels (#1387)

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