xinyun/vllm
1
0
Fork 0
mirror of https://git.datalinker.icu/vllm-project/vllm.git synced 2025-12-24 03:25:01 +08:00
Code Issues Packages Projects Releases Wiki Activity

[Log] Skip piecewise cudagraph warn when using full cudagraph (#30657)

Browse Source 
Signed-off-by: Boyuan Feng <boyuan@meta.com>
This commit is contained in:
Boyuan Feng 2025-12-14 18:49:45 -08:00 committed by GitHub
parent e2ed238885
commit 917fdae5b2
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 7 additions and 3 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

10
vllm/config/compilation.py
View File

@ -932,9 +932,13 @@ class CompilationConfig:
self.splitting_ops = list(self._attention_ops) self.splitting_ops = list(self._attention_ops)
added_default_splitting_ops = True added_default_splitting_ops = True
elif len(self.splitting_ops) == 0: elif len(self.splitting_ops) == 0:
logger.warning_once( if (
"Using piecewise compilation with empty splitting_ops" self.cudagraph_mode == CUDAGraphMode.PIECEWISE
) or self.cudagraph_mode == CUDAGraphMode.FULL_AND_PIECEWISE
):
logger.warning_once(
"Using piecewise compilation with empty splitting_ops"
)
if self.cudagraph_mode == CUDAGraphMode.PIECEWISE: if self.cudagraph_mode == CUDAGraphMode.PIECEWISE:
logger.warning_once( logger.warning_once(
"Piecewise compilation with empty splitting_ops do not" "Piecewise compilation with empty splitting_ops do not"