KEMBAR78
[https://nvbugs/5488576][fix] Propagate disable_finalize_fusion config flag in WIDEEP MoE backend (cherry-pick #8141) by kaiyux · Pull Request #8566 · NVIDIA/TensorRT-LLM · GitHub
Skip to content

Conversation

@kaiyux
Copy link
Member

@kaiyux kaiyux commented Oct 22, 2025

This PR cherry picks #8141.

Summary by CodeRabbit

Release Notes

  • Improvements
    • Enhanced Mixture of Experts (MoE) execution with configurable finalization fusion behavior, providing users with greater control over optimization settings through model configuration.

Description

Test Coverage

PR Checklist

Please review the following before submitting your PR:

  • PR description clearly explains what and why. If using CodeRabbit's summary, please make sure it makes sense.

  • PR Follows TRT-LLM CODING GUIDELINES to the best of your knowledge.

  • Test cases are provided for new code paths (see test instructions)

  • Any new dependencies have been scanned for license and vulnerabilities

  • CODEOWNERS updated if ownership changes

  • Documentation updated as needed

  • The reviewers assigned automatically/manually are appropriate for the PR.

  • Please check this after reviewing the above items as appropriate for this PR.

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

…g flag in WIDEEP MoE backend (NVIDIA#8141)

Signed-off-by: Sergey Klevtsov <sklevtsov@nvidia.com>
@kaiyux kaiyux requested a review from a team as a code owner October 22, 2025 06:12
@kaiyux
Copy link
Member Author

kaiyux commented Oct 22, 2025

/bot run

@coderabbitai
Copy link
Contributor

coderabbitai bot commented Oct 22, 2025

📝 Walkthrough

Walkthrough

Added a configurable use_fused_finalize flag to the WideEPMoE class, initialized as the negation of model_config.moe_disable_finalize_fusion. Updated the forward_chunk call to use this flag instead of a hardcoded True value, enabling dynamic control over fused finalize fusion behavior.

Changes

Cohort / File(s) Change Summary
WideEPMoE fused finalize configuration
tensorrt_llm/_torch/modules/fused_moe/fused_moe_wide_ep.py
Added use_fused_finalize instance variable initialized from model_config.moe_disable_finalize_fusion negation; updated forward_chunk invocation to pass this flag instead of constant True

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

Pre-merge checks and finishing touches

❌ Failed checks (2 warnings)
Check name Status Explanation Resolution
Docstring Coverage ⚠️ Warning Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. You can run @coderabbitai generate docstrings to improve docstring coverage.
Description Check ⚠️ Warning The pull request description is largely incomplete and fails to meet the template requirements. While the PR title is properly formatted with a NVBugs ID and [fix] type, the Description section contains only a single sentence ("This PR cherry picks #8141.") without explaining the underlying issue, the solution implemented, or the rationale for the changes. The Test Coverage section is entirely empty with no tests listed or explained. Most checklist items are unchecked and not addressed, despite the final checkbox being marked. The author has provided the template structure but without substantive content needed for reviewers to understand what is being changed and why, making it difficult to assess the PR's merit during code review. The author should expand the Description section to clearly explain what issue #8141 addresses and how propagating the disable_finalize_fusion config flag in the WIDEEP MoE backend solves it. The Test Coverage section must list specific test cases that validate these changes. Additionally, the remaining PR Checklist items should be reviewed and appropriate checkboxes marked to confirm the PR meets project standards for coding guidelines, test coverage, documentation, and other requirements.
✅ Passed checks (1 passed)
Check name Status Explanation
Title Check ✅ Passed The pull request title follows the required format with a valid NVBugs ID [https://nvbugs/5488576], a valid type indicator [fix], and a clear, specific summary: "Propagate disable_finalize_fusion config flag in WIDEEP MoE backend (cherry-pick #8141)". This title directly corresponds to the main change described in the raw summary—adding a configurable flag to control fused finalize behavior in the MoE backend and updating the forward_chunk call to use it. The title is concise, descriptive, and clearly communicates the primary change without being vague or overly broad.
✨ Finishing touches
  • 📝 Generate docstrings
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Post copyable unit tests in a comment

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

Comment @coderabbitai help to get the list of available commands and usage tips.

@tensorrt-cicd
Copy link
Collaborator

PR_Github #22137 [ run ] triggered by Bot. Commit: 1f7b575

@kaiyux kaiyux enabled auto-merge (squash) October 22, 2025 10:16
@tensorrt-cicd
Copy link
Collaborator

PR_Github #22137 [ run ] completed with state SUCCESS. Commit: 1f7b575
/LLM/release-1.1/L0_MergeRequest_PR pipeline #223 completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

@kaiyux kaiyux merged commit c7b06b1 into NVIDIA:release/1.1 Oct 22, 2025
9 of 11 checks passed
@kaiyux kaiyux deleted the user/kaiyu/cherry-pick-017583a branch October 22, 2025 14:06
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

None yet

Development

Successfully merging this pull request may close these issues.

4 participants