@tduncan / Nvidia-Optimization-Stack-VLLM Mirror · read-only / commits / 15063741e3
Fork 0

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 17, 00:30 UTC

[Misc] Missing error message for custom ops import (#5282)

Jie Fu (傅杰) committed Jun 7, 2024 at 11:17 UTC 15063741e30881d7a982c3436c3299a0551327dc
1 file changed +4 -2
vllm/_custom_ops.py
+4 -2
@@ -5,8 +5,10 @@ import torch
5 try:
6 from vllm._C import cache_ops as vllm_cache_ops
7 from vllm._C import ops as vllm_ops
8 -except ImportError:
9 - pass
8 +except ImportError as e:
9 + from vllm.logger import init_logger
10 + logger = init_logger(__name__)
11 + logger.warning("Failed to import from vllm._C with %r", e)
12
13
14 # activation ops