Skip to content

[Bugfix] Avoid to warmup when world size is 1 #5442

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Closed
wants to merge 6 commits into from
Closed

[Bugfix] Avoid to warmup when world size is 1 #5442

Changes from all commits
Commits
Show all changes
6 commits
Select commit Hold shift + click to select a range
ed4f36b
Unify the var
kerthcet Jun 12, 2024
babdc81
Avoid to warmup once world size is 1
kerthcet Jun 12, 2024
f45bc68
address comments
kerthcet Jun 12, 2024
47b28bd
Revert "Unify the var"
kerthcet Jun 12, 2024
3fba472
address comments
kerthcet Jun 12, 2024
1ae0335
address comments
kerthcet Jun 12, 2024
File filter

Filter by extension

Filter by extension
Conversations
Failed to load comments.
Loading
Jump to
Jump to file
Failed to load files.
Loading
Diff view
Diff view
6 changes: 6 additions & 0 deletions vllm/distributed/parallel_state.py
View file
Open in desktop
Original file line number Diff line number Diff line change
Expand Up @@ -115,6 +115,12 @@ def init_distributed_environment(
local_rank = rank
global _LOCAL_RANK
_LOCAL_RANK = local_rank

if world_size == 1:
# Avoid to warmup if world size is 1, to be compatible with vGPU.
# See https://github.com/vllm-project/vllm/issues/4587 for details.
return

# A small all_reduce for warmup.
data = torch.zeros(1)
if torch.cuda.is_available():
Expand Down