Repository navigation
Conversation
Signed-off-by: Masahiro Tanaka <tanaka.masahiro@gmail.com>
Codex Review SummaryThis comment shows the latest Codex review activity on this pull request.
ℹ️ About Codex in GitHubYour team has set up Codex to review pull requests in this repo. Reviews are triggered when you
Codex reacts with 👀 while any review is running, comments if it has suggestions, and reacts with 👍 once all reviews finish with no findings. |
![chatgpt-codex-connector[bot]](https://avatars.githubusercontent.com/in/1144995?s=60&v=4){kind=small}
There was a problem hiding this comment.
💡 Codex Review
Here are some automated review suggestions for this pull request.
Reviewed commit: e42df3ff27
ℹ️ About Codex in GitHub
Codex has been enabled to automatically review pull requests in this repo. Reviews are triggered when you
- Open a pull request for review
- Mark a draft as ready
- Comment "@codex review".
If Codex has suggestions, it will comment; otherwise it will react with 👍.
When you sign up for Codex through ChatGPT, Codex can also answer questions or update the PR, like "@codex address that feedback".
Signed-off-by: Masahiro Tanaka <tanaka.masahiro@gmail.com>
This branch has not been deployed
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Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be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.
Problem
After successful distributed teardown, DeepSpeed could retain references to destroyed process groups in its group caches. Reinitialization could then reuse a stale TP group. In the CPU/Gloo AutoEP test path, completed engines could also remain in hook cycles until interpreter finalization; native Linux reproduced a SIGABRT while a Gloo worker released Python tensor state during finalization.
This shutdown-lifetime issue is a suspected cause of the CPU/Gloo SIGABRT observed in [this CI run](https://github.com/deepspeedai/DeepSpeed/actions/runs/38838610946/job/116773374488), which motivated this investigation. We later reproduced the same abort signature on native Linux, but without a core dump from the original run, we cannot confirm that both failures share the same root cause.
Approach
Release DeepSpeed-owned group aliases and metadata only after backend teardown succeeds. WORLD teardown clears all owned group state, while explicit subgroup teardown removes only identity-matching entries. Note that backend failures preserve the caches, and NON_GROUP_MEMBER remains a no-op.
The CPU/Gloo test helper runs cyclic collection after a successful callback and global teardown; production teardown does not force garbage collection.