A high-throughput and memory-efficient inference and serving engine for LLMs
Mirrored from https://github.com/tduncan-lum/Nvidia-Optimization-Stack---VLLM.git
Synced Nov 15, 08:30 UTC
[Deprecation] Remove unused sync methods in `async_timeout` (#18792)
Signed-off-by: DarkLight1337 <tlleungac@connect.ust.hk>
Cyrus Leung committed
May 28, 2025 at 15:08 UTC
0f0926b43fb5445e2468034eb0bc8076cdd6af86
1 file changed
-19
vllm/engine/async_timeout.py
-19
@@ -8,7 +8,6 @@
8
import asyncio
9
import enum
10
import sys
11
-import warnings
11
from types import TracebackType
12
from typing import Any, Optional, Type
13
@@ -66,24 +65,6 @@ else:
65
else:
66
self.update(deadline)
67
69
- def __enter__(self) -> "Timeout":
70
- warnings.warn(
71
- "with timeout() is deprecated, use async with timeout()",
72
- DeprecationWarning,
73
- stacklevel=2,
74
- )
75
- self._do_enter()
76
- return self
77
-
78
- def __exit__(
79
- self,
80
- exc_type: Optional[Type[BaseException]],
81
- exc_val: Optional[BaseException],
82
- exc_tb: Optional[TracebackType],
83
- ) -> Optional[bool]:
84
- self._do_exit(exc_type)
85
- return None
86
-
68
async def __aenter__(self) -> "Timeout":
69
self._do_enter()
70
return self