Skip to content

[Dev] fix cg missing wgrad hook#2999

Merged
Wohox merged 18 commits into
NVIDIA:devfrom
Wohox:pingtian/fix_cg_missing_wgrad_hook
Jan 29, 2026
Merged

[Dev] fix cg missing wgrad hook#2999
Wohox merged 18 commits into
NVIDIA:devfrom
Wohox:pingtian/fix_cg_missing_wgrad_hook

Conversation

@Wohox

@Wohox Wohox commented Jan 19, 2026 •
edited
Loading

Copy link
Copy Markdown
Contributor

What does this PR do ?

This PR fixes the missing backward post hook bug, this bug happens when delay_wgrad_compute and cuda_graph are both enabled, the parameters within delayed wgrad computation scope will skip the backward post hook.

The fix is in TE, this PR only adds version guard and several flag assertions.
Related TE PR: NVIDIA/TransformerEngine#2614
PR for main: #3074

Contribution process

flowchart LR
    A[Pre-checks] --> B[PR Tests]
    subgraph Code Review/Approval
        C1[Expert Review] --> C2[Final Review]
    end
    B --> C1
    C2 --> D[Merge]
Loading

Pre-checks

  • I want this PR in a versioned release and have added the appropriate Milestone (e.g., Core 0.8)
  • I have added relevant unit tests
  • I have added relevant functional tests
  • I have added proper typing to my code Typing guidelines
  • I have added relevant documentation
  • I have run the autoformatter.sh on my PR

Code review

The following process is enforced via the CODEOWNERS file for changes into megatron/core. For changes outside of megatron/core, it is up to the PR author whether or not to tag the Final Reviewer team.

For MRs into `main` branch

Feel free to message or comment the @mcore-oncall to help accelerate your merge into main. The less complex your PR is, the faster it will be approved and merged!

(Step 1): Add PR label Expert Review

(Step 2): Collect the expert reviewers reviews

  1. Attach the Expert Review label when your PR is ready for review.
  2. GitHub auto-assigns expert reviewers based on your changes. They will get notified and pick up your PR soon.

⚠️ Only proceed to the next step once all reviewers have approved, merge-conflict are resolved and the CI is passing.
Final Review might get declined if these requirements are not fulfilled.

(Step 3): Final Review

  1. Add Final Review label
  2. GitHub auto-assigns final reviewers based on your changes. They will get notified and pick up your PR soon.

(Optional Step 4): Cherry-pick into release branch

If this PR also needs to be merged into core_r* release branches, after this PR has been merged, select Cherry-pick to open a new PR into the release branch.

For MRs into `dev` branch The proposed review process for `dev` branch is under active discussion.

MRs are mergable after one approval by either eharper@nvidia.com or zijiey@nvidia.com.

Merging your PR

Any member of core-adlr and core-nemo will be able to merge your PR.

@copy-pr-bot

copy-pr-bot Bot commented Jan 19, 2026

Copy link
Copy Markdown

This pull request requires additional validation before any workflows can run on NVIDIA's runners.

Pull request vetters can view their responsibilities here.

Contributors can view more details about this message here.

@Wohox Wohox requested review from buptzyb and lhb8125 January 20, 2026 04:00
buptzyb
buptzyb reviewed Jan 22, 2026
View reviewed changes
Comment thread megatron/core/extensions/transformer_engine.py Outdated
Comment thread megatron/core/transformer/transformer_config.py Outdated
Comment thread megatron/core/transformer/module.py Outdated
@Wohox Wohox mentioned this pull request Jan 26, 2026
Merged
6 tasks
@Wohox Wohox added dev branch Dev branch related issues and development bug Something isn't working labels Jan 26, 2026
@Wohox

Wohox commented Jan 26, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test f7209d9

@ko3n1g ko3n1g added this to the Core 0.16 milestone Jan 26, 2026
@Wohox Wohox requested review from a team as code owners January 26, 2026 07:39
@Wohox Wohox requested a review from BestJuly January 26, 2026 07:41
BestJuly
BestJuly approved these changes Jan 26, 2026
View reviewed changes

@BestJuly BestJuly left a comment

Copy link
Copy Markdown
Contributor

Choose a reason for hiding this comment

The reason will be displayed to describe this comment to others. Learn more.

LGTM

@Wohox Wohox changed the title (Draft)[Dev] fix cg missing wgrad hook [Dev] fix cg missing wgrad hook Jan 27, 2026
@Wohox Wohox enabled auto-merge January 27, 2026 02:13
@Wohox

Wohox commented Jan 27, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test 6233d37

@Wohox

Wohox commented Jan 28, 2026

Copy link
Copy Markdown
Contributor Author

/ok to test 524af67

@Wohox Wohox added this pull request to the merge queue Jan 28, 2026
github-merge-queue Bot pushed a commit that referenced this pull request Jan 28, 2026
@Wohox
[Dev] fix cg missing wgrad hook (#2999)
45c1d3c
@github-merge-queue github-merge-queue Bot removed this pull request from the merge queue due to failed status checks Jan 28, 2026
@Wohox Wohox added this pull request to the merge queue Jan 29, 2026
Merged via the queue into NVIDIA:dev with commit d587dd1 Jan 29, 2026
46 checks passed
@Wohox Wohox deleted the pingtian/fix_cg_missing_wgrad_hook branch January 29, 2026 07:21
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@BestJuly BestJuly BestJuly approved these changes

@buptzyb buptzyb buptzyb approved these changes

@lhb8125 lhb8125 Awaiting requested review from lhb8125

Assignees

No one assigned

Labels

bug Something isn't working dev branch Dev branch related issues and development

Projects

None yet

Milestone

Core 0.16

Development

Successfully merging this pull request may close these issues.

4 participants

@Wohox @BestJuly @buptzyb @ko3n1g