Update vllm to point to flash-attention commit that builds FA3 with torch stable API. (Retry) (#49599)

Signed-off-by: Chris Leonard <chleonar@redhat.com>
Signed-off-by: Shengqi Chen <harry-chen@outlook.com>
Co-authored-by: Shengqi Chen <harry-chen@outlook.com>

.buildkite/check-torch-abi.py

@@ -12,10 +12,7 @@ from torch_abi_audit.report import ExtensionReport, PackageReport
 
 # Temporary allowlist of extensions not yet on the stable ABI.
 # Shrink and remove over time.
-ALLOWED_UNSTABLE_LIBRARIES: tuple[str, ...] = (
-    "vllm_flash_attn/_vllm_fa3_C.abi3.so",
-    "third_party/deep_gemm/_C*.so",
-)
+ALLOWED_UNSTABLE_LIBRARIES: tuple[str, ...] = ("third_party/deep_gemm/_C*.so",)
 
 
 def _relative_path(lib: ExtensionReport, package_root: Path) -> str:

cmake/external_projects/vllm_flash_attn.cmake

@@ -39,7 +39,7 @@ else()
   FetchContent_Declare(
           vllm-flash-attn
           GIT_REPOSITORY https://github.com/vllm-project/flash-attention.git
-          GIT_TAG 28e862d21806bc3580207aa0ad4e2759151e9827
+          GIT_TAG f3e1a4f74c99145c0717709860bf765de1703779
           GIT_PROGRESS TRUE
           # Don't share the vllm-flash-attn build between build types
           BINARY_DIR ${CMAKE_BINARY_DIR}/vllm-flash-attn