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

Revert "[PerfFix] Avoid separate thread for MP executor shm spin (#28012)" (#28289)

Browse Source 
Signed-off-by: NickLucche <nlucches@redhat.com>
This commit is contained in:
Nicolò Lucchesi
2025-11-07 16:07:01 +01:00
committed by GitHub
parent 0f872b7977
commit 68a72a5cc1
9 changed files with 135 additions and 147 deletions
Download Patch File Download Diff File Expand all files Collapse all files

2
vllm/v1/worker/gpu_worker.py
View File

@@ -524,7 +524,7 @@ class Worker(WorkerBase):
@torch.inference_mode()
def sample_tokens(
self, grammar_output: "GrammarOutput | None"
self, grammar_output: "GrammarOutput"
) -> ModelRunnerOutput | AsyncModelRunnerOutput:
return self.model_runner.sample_tokens(grammar_output)