Skip to content
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

Jump to bottom

Skip cuda backend initialization if no nvidia GPUs are visible #22703

Merged
merged 1 commit into from
Jul 29, 2024
Merged

Skip cuda backend initialization if no nvidia GPUs are visible #22703

merged 1 commit into from
Jul 29, 2024

Conversation

Rifur13
Copy link
Collaborator

@Rifur13 Rifur13 commented Jul 29, 2024

Skip cuda backend initialization if no nvidia GPUs are visible.

It currently crashes, but we want jax to fallback to the CPU.

Testing:

  • Tested on a GPU VM with no cuda driver installed.

@Rifur13 Rifur13 requested a review from hawkinsp July 29, 2024 04:46
@google-ml-butler google-ml-butler bot added kokoro:force-run pull ready Ready for copybara import and testing labels Jul 29, 2024
@copybara-service copybara-service bot merged commit f070c06 into jax-ml:main Jul 29, 2024
16 checks passed
@jjyyxx
Copy link
Contributor

jjyyxx commented Aug 1, 2024

@hawkinsp This patch's intention is good, but does break our environment inside container in an obscure way, since only a subset of GPUs are mounted. #22773 (right after v0.4.31 released) fixes this.

Is it possible to release a new patch version soon?

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Reviewers

@hawkinsp hawkinsp hawkinsp approved these changes

Assignees
No one assigned
Labels
pull ready Ready for copybara import and testing
Projects
None yet
Milestone
No milestone
Development

Successfully merging this pull request may close these issues.
4 participants
@Rifur13 @jjyyxx @hawkinsp @kokoro-team