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

[cuda][misc] remove error_on_invalid_device_count_status (#7069)

Browse Source
This commit is contained in:
youkaichao
2024-08-02 00:14:21 -07:00
committed by GitHub
parent cf2a1a4d9d
commit 660dea1235
3 changed files with 3 additions and 32 deletions
Download Patch File Download Diff File Expand all files Collapse all files

3
vllm/executor/multiproc_gpu_executor.py
View File

@@ -17,7 +17,6 @@ from vllm.logger import init_logger
from vllm.sequence import ExecuteModelRequest, SamplerOutput
from vllm.triton_utils import maybe_set_triton_cache_manager
from vllm.utils import (_run_task_with_lock, cuda_device_count_stateless,
error_on_invalid_device_count_status,
get_distributed_init_method, get_open_port,
get_vllm_instance_id, make_async,
update_environment_variables)
@@ -79,8 +78,6 @@ class MultiprocessingGPUExecutor(DistributedGPUExecutor):
f"please ensure that world_size ({world_size}) "
f"is less than than max local gpu count ({cuda_device_count})")
error_on_invalid_device_count_status()
# 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.