biondizzle/vllm
1
0
Fork 0
Code Issues Pull Requests Actions 2 Packages Projects Releases Wiki Activity

[misc][distributed] use 127.0.0.1 for single-node (#5619)

Browse Source
This commit is contained in:
youkaichao
2024-06-19 01:05:00 -07:00
committed by GitHub
parent f758aed0e8
commit 3eea74889f
2 changed files with 15 additions and 2 deletions
Download Patch File Download Diff File Expand all files Collapse all files

7
vllm/executor/multiproc_gpu_executor.py
View File

@@ -10,7 +10,7 @@ from vllm.executor.multiproc_worker_utils import (ProcessWorkerWrapper,
from vllm.logger import init_logger
from vllm.sequence import ExecuteModelRequest, SamplerOutput
from vllm.utils import (cuda_device_count_stateless,
get_distributed_init_method, get_ip, get_open_port,
get_distributed_init_method, get_open_port,
get_vllm_instance_id, make_async)
logger = init_logger(__name__)
@@ -37,8 +37,11 @@ class MultiprocessingGPUExecutor(DistributedGPUExecutor):
assert world_size <= cuda_device_count_stateless(), (
"please set tensor_parallel_size to less than max local gpu count")
# Multiprocessing-based executor does not support multi-node setting.
# Since it only works for single node, we can use the loopback address
# 127.0.0.1 for communication.
distributed_init_method = get_distributed_init_method(
get_ip(), get_open_port())
"127.0.0.1", get_open_port())
if world_size == 1:
self.workers = []