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, 12:15 UTC
[Bugifx] Remove TritonPlaceholder from sys.modules (#17317)
Signed-off-by: Isotr0py <2037008807@qq.com>
Isotr0py committed
May 2, 2025 at 15:45 UTC
9e2de9b9e9f9b850312200534f11e18f008a2d29
1 file changed
+1
-7
vllm/triton_utils/importing.py
+1
-7
@@ -1,6 +1,5 @@
1
# SPDX-License-Identifier: Apache-2.0
2
3
-import sys
3
import types
4
from importlib.util import find_spec
5
@@ -45,9 +44,4 @@ if not HAS_TRITON:
44
super().__init__("triton.language")
45
self.constexpr = None
46
self.dtype = None
48
-
49
- sys.modules['triton'] = TritonPlaceholder()
50
- sys.modules['triton.language'] = TritonLanguagePlaceholder()
51
-
52
-if 'triton' in sys.modules:
53
- logger.info("Triton module has been replaced with a placeholder.")
47
+ self.int64 = None