[None][fix] Fix enable_alltoall passed to CutlassFusedMoE#11016
[None][fix] Fix enable_alltoall passed to CutlassFusedMoE#11016syuoni merged 3 commits intoNVIDIA:mainfrom
Conversation
|
/bot run --disable-fail-fast |
📝 WalkthroughWalkthroughThe changes update benchmark configuration logic and refactor the MOE alltoall usage detection mechanism. The benchmark runner now pre-computes context sequence length and batch size parameters for the prefill path, while MOE modules replace internal alltoall detection with an explicit enable_alltoall flag that can be controlled at invocation time. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~30 minutes 🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ Finishing touches
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tensorrt_llm/_torch/modules/fused_moe/fused_moe_cutlass.py (1)
686-696:forward_chunkshould passenable_alltoall=Truetorun_moewhen using external alltoall distribution.When
self.enable_alltoall=True,forward_chunkperforms alltoall redistribution externally via Python (lines 570-663) before callingrun_moe. However, the kernel call (lines 686-696) does not pass theenable_alltoallparameter, so it defaults toFalse.The C++ kernel uses
enable_alltoallto determine whether to skip filling invalid output tokens with zeros in the finalize step (moe_kernels.cu:1913-1917). When alltoall redistribution is used, the comment states: "If all-to-all comm is enabled, finalizeMoeRouting doesn't need to fill the invalid output tokens with zeros." This logic applies regardless of whether alltoall is handled internally by the kernel or externally by Python.By not passing
enable_alltoall=True, the kernel will unnecessarily attempt to zero invalid outputs even though the data was already properly redistributed externally. Passenable_alltoall=self.enable_alltoallto therun_moecall so the kernel behavior aligns with the actual data distribution method.
|
PR_Github #33664 [ run ] triggered by Bot. Commit: |
|
/bot run --disable-fail-fast |
|
PR_Github #33686 [ run ] triggered by Bot. Commit: |
|
PR_Github #33686 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #33730 [ run ] triggered by Bot. Commit: |
|
PR_Github #33730 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #33794 [ run ] triggered by Bot. Commit: |
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com>
|
PR_Github #33794 [ run ] completed with state
|
syuoni
force-pushed
the
fix-config-moe
branch
from
2e7c0a0 to
ee8fe55
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #33873 [ run ] triggered by Bot. Commit: |
|
PR_Github #33873 [ run ] completed with state
|
|
/bot run --disable-fail-fast |
|
PR_Github #33933 [ run ] triggered by Bot. Commit: |
|
PR_Github #33933 [ run ] completed with state |
5 participants
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
Update tava architecture diagram if there is a significant design change in PR.
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.
Details
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 thestage-listparameter 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.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip 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-pipelineReuse 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.
Summary by CodeRabbit
Release Notes
New Features
Performance Improvements
Refactor
✏️ Tip: You can customize this high-level summary in your review settings.