frozenleaves/vllm
1
0
Fork 0
mirror of https://github.com/vllm-project/vllm.git synced 2025-10-20 23:03:52 +08:00
Code Issues Packages Projects Releases Wiki Activity

[V1] Guard Against Main Thread Usage (#14972)

Browse Source 
Signed-off-by: rshaw@neuralmagic.com <robertgshaw2@gmail.com>
This commit is contained in:
Robert Shaw
2025-03-17 16:23:02 -04:00
committed by simon-mo
parent 5ac1a8e6e4
commit 16e9064f84
1 changed files with 6 additions and 0 deletions
Download Patch File Download Diff File Expand all files Collapse all files

6
vllm/engine/arg_utils.py
View File

@ -3,6 +3,7 @@
import argparse import argparse
import dataclasses import dataclasses
import json import json
import threading
from dataclasses import dataclass from dataclasses import dataclass
from typing import (TYPE_CHECKING, Any, Dict, List, Literal, Mapping, Optional, from typing import (TYPE_CHECKING, Any, Dict, List, Literal, Mapping, Optional,
Tuple, Type, Union, cast, get_args) Tuple, Type, Union, cast, get_args)
@ -1576,6 +1577,11 @@ class EngineArgs:
############################################################# #############################################################
# Experimental Features - allow users to opt in. # Experimental Features - allow users to opt in.
# Signal Handlers requires running in main thread.
if (threading.current_thread() != threading.main_thread()
and _warn_or_fallback("Engine in background thread")):
return False
# LoRA is supported on V1, but off by default for now. # LoRA is supported on V1, but off by default for now.
if self.enable_lora and _warn_or_fallback("LORA"): if self.enable_lora and _warn_or_fallback("LORA"):
return False return False