-
Notifications
You must be signed in to change notification settings - Fork 3.4k
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.
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
[CI] fix horovod tests #14382
[CI] fix horovod tests #14382
Conversation
For completeness sake of what broke - Horovod tests were broken due to NCCL, which from 2.12.7 broke when linked against dynamic The solution is to forcefully downgrade to the last working version Added couple small improvements to how GPU tests are being run (and the environment being prepared) |
This will be ready to merge once Docker images are built and pushed to dockerhub |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
🎉
This reverts commit 1f1e69e.
This reverts commit f8649c6.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Are we aware that the GPU tests are skipped?
https://dev.azure.com/Lightning-AI/lightning/_build/results?buildId=92435&view=logs&jobId=3f274fac-2e11-54ca-487e-194c91f3ae9f
What does this PR do?
Horovod tests are failing with mysterious errors
Does your PR introduce any breaking changes? If yes, please list them.
Before submitting
PR review
Anyone in the community is welcome to review the PR.
Before you start reviewing, make sure you have read the review guidelines. In short, see the following bullet-list:
Did you have fun?
Make sure you had fun coding 🙃
cc @tchaton @rohitgr7 @carmocca @akihironitta @Borda @awaelchli @justusschock @kaushikb11