biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

Reapply [Attention][FA3] Update FA3 to include new swizzle optimization (#34043)

Browse Source 
Signed-off-by: Lucas Wilkinson <lwilkins@redhat.com>
This commit is contained in:
Lucas Wilkinson
2026-02-11 08:07:56 -07:00
committed by GitHub
parent 1b8756562e
commit c7914d30f9
6 changed files with 60 additions and 44 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
cmake/external_projects/vllm_flash_attn.cmake
View File

@@ -38,7 +38,7 @@ else()
FetchContent_Declare(
vllm-flash-attn
GIT_REPOSITORY https://github.com/vllm-project/flash-attention.git
GIT_TAG 188be16520ceefdc625fdf71365585d2ee348fe2
GIT_TAG 5824e6e2008271063c3229ab3e7032bd74abbbc6
GIT_PROGRESS TRUE
# Don't share the vllm-flash-attn build between build types
BINARY_DIR ${CMAKE_BINARY_DIR}/vllm-flash-attn