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 16, 00:45 UTC
[Misc] Improve error message for incorrect pynvml (#12809)
Signed-off-by: youkaichao <youkaichao@gmail.com>
youkaichao committed
Feb 6, 2025 at 15:23 UTC
0408efc6d0c17fba17b2be38d0d0f02e96d2bf9d
1 file changed
+5
-1
vllm/platforms/__init__.py
+5
-1
@@ -41,7 +41,11 @@ def cuda_platform_plugin() -> Optional[str]:
41
is_cuda = True
42
finally:
43
pynvml.nvmlShutdown()
44
- except Exception:
44
+ except Exception as e:
45
+ if "nvml" not in e.__class__.__name__.lower():
46
+ # If the error is not related to NVML, re-raise it.
47
+ raise e
48
+
49
# CUDA is supported on Jetson, but NVML may not be.
50
import os
51