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, 08:15 UTC
[doc] improve debugging code (#10206)
Signed-off-by: youkaichao <youkaichao@gmail.com>
youkaichao committed
Nov 10, 2024 at 17:49 UTC
f0f2e5638ef4858b00b137bea1c3f8312e48efa6
1 file changed
+3
docs/source/getting_started/debugging.rst
+3
@@ -75,6 +75,9 @@ If GPU/CPU communication cannot be established, you can use the following Python
75
76
print("PyTorch GLOO is successful!")
77
78
+ if world_size <= 1:
79
+ exit()
80
+
81
# Test vLLM NCCL, with cuda graph
82
from vllm.distributed.device_communicators.pynccl import PyNcclCommunicator
83