Skip to content
This repository was archived by the owner on Aug 7, 2025. It is now read-only.

Set vllm multiproc method to spawn #3310

Merged
merged 2 commits into from
Sep 13, 2024
Merged

Set vllm multiproc method to spawn #3310

merged 2 commits into from
Sep 13, 2024

Conversation

mreso
Copy link
Collaborator

@mreso mreso commented Sep 11, 2024

Description

This PR changes the vllm method to create their workers from fork to spawn to avoid an issue where vllm tries to reinitialize the CUDA context

Type of change

Please delete options that are not relevant.

  • Bug fix (non-breaking change which fixes an issue)

Feature/Issue validation/testing

Please describe the Unit or Integration tests that you ran to verify your changes and relevant result summary. Provide instructions so it can be reproduced.
Please also list any relevant details for your test configuration.

  • Test A
    Logs for Test A

  • Test B
    Logs for Test B

Checklist:

  • Did you have fun?

@mreso mreso requested a review from agunapal September 11, 2024 23:02
@mreso mreso force-pushed the fix/vllm_spawn_method branch from 00f7c81 to 67e30b0 Compare September 11, 2024 23:04
@mreso mreso force-pushed the fix/vllm_spawn_method branch from 67e30b0 to 7c61e5a Compare September 11, 2024 23:11
@mreso mreso added this pull request to the merge queue Sep 13, 2024
Merged via the queue into master with commit 048aa53 Sep 13, 2024
14 checks passed
Sign up for free to subscribe to this conversation on GitHub. Already have an account? Sign in.
Labels
None yet
Projects
None yet
Development

Successfully merging this pull request may close these issues.

2 participants