Conversation
Let run_alignment_test.sh own environment setup and propagate setup failures. Signed-off-by: Zhan Rongrui <me@zrr.dev>
Signed-off-by: Zhan Rongrui <me@zrr.dev>
zrr1999
changed the title
Signed-off-by: Zhan Rongrui <me@zrr.dev>
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
PR type
PR information
Repair the CI setup failures reported on #12. Alignment setup currently cannot resolve
paddle-nvidia-nvshmem-cu13; select the CUDA 12.9 image and matching ops wheel, following PFCCLab/Megatron-LM#19. Keep the earlier removal of duplicate setup sorun_alignment_test.showns installation and propagates its failure.The unittest workflow currently invokes
sudo chowneven when the runner already owns the checkout. Check ownership first, and use noninteractive sudo only when repair is needed. A denied required ownership repair still fails the job. Normalize the existing whitespace and quote violations found by the full-repository lint job; Python ASTs are unchanged.Experiment results
Full-repository pre-commit passes. Executed the actual workspace preparation shell for an already-owned checkout, a checkout requiring repair, and a denied repair; the owned case no longer calls sudo, and denied repair remains a failure. Workflow YAML and changed shell syntax pass. The unchanged single-setup behavior retains its earlier success/failure validation.
These are CI setup and formatting changes. No model algorithm, accuracy switch, workflow trigger, or test threshold is changed. GPU training and a real privileged ownership repair were not run for this infrastructure patch.