xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2026-05-28 12:37:10 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Bugfix] Fix MTP+FlashInfer crash when trtllm kernels are available but disabled (#26361)

Browse Source 
Signed-off-by: Benjamin Chislett <bchislett@nvidia.com>
Signed-off-by: Benjamin Chislett <chislett.ben@gmail.com>
Co-authored-by: gemini-code-assist[bot] <176961590+gemini-code-assist[bot]@users.noreply.github.com>
This commit is contained in:
Benjamin Chislett 2025-10-07 18:12:26 -04:00 committed by GitHub
parent 1b86bd8e18
commit caf8b1c084
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 2 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/utils/flashinfer.py
View File

@ -220,6 +220,8 @@ def force_use_trtllm_attention() -> bool | None:
def can_use_trtllm_attention(num_qo_heads: int, num_kv_heads: int) -> bool: def can_use_trtllm_attention(num_qo_heads: int, num_kv_heads: int) -> bool:
"""Check if the current configuration supports TRTLLM attention.""" """Check if the current configuration supports TRTLLM attention."""
if force_use_trtllm_attention() is False:
return False
has_trtllm = supports_trtllm_attention() has_trtllm = supports_trtllm_attention()
return has_trtllm and (num_qo_heads % num_kv_heads == 0) return has_trtllm and (num_qo_heads % num_kv_heads == 0)