KEMBAR78
[None][fix] Enable FP8 ContextMLA on GB300 by longlee0622 · Pull Request #8080 · NVIDIA/TensorRT-LLM · GitHub
Skip to content

Conversation

@longlee0622
Copy link
Collaborator

@longlee0622 longlee0622 commented Sep 30, 2025

Summary by CodeRabbit

  • New Features
    • Expanded FP8 MLA support to GPUs with SM 103, alongside existing architectures, broadening hardware compatibility for accelerated attention workloads.
    • FP8 MLA now activates only when FP8 KV cache quantization is enabled, ensuring consistent behavior across supported devices.
    • Users on compatible GPUs may see improved throughput for attention operations when FP8 modes are enabled and KV cache quantization is configured accordingly.

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.

@longlee0622 longlee0622 requested a review from yuxianq September 30, 2025 01:02
@coderabbitai
Copy link
Contributor

coderabbitai bot commented Sep 30, 2025

📝 Walkthrough

Walkthrough

Extends the condition enabling FP8 MLA context to include SM version 103, alongside 90, 100, and 120, and ensures it is gated by op->mKVCacheQuantMode.hasFp8KvCache(). No other logic or interfaces are modified.

Changes

Cohort / File(s) Summary
FP8 MLA gating update
cpp/tensorrt_llm/thop/attentionOp.cpp
Updated condition for setting op->mFP8ContextMLA to require SM in {90, 100, 103, 120} and op->mKVCacheQuantMode.hasFp8KvCache()

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 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 just the template scaffolding and lacks a filled-in title, summary of what was changed and why, and details on test coverage, leaving all required sections empty. Please update the PR description by inserting a proper title following the “[JIRA ticket/…][type] Summary” format, provide a concise “Description” of the issue and the solution, list relevant tests under “Test Coverage,” and ensure the checklist reflects the actual changes.
✅ Passed checks (1 passed)
Check name Status Explanation
Title Check ✅ Passed The title succinctly identifies the primary update—enabling FP8 ContextMLA on GB300—while following the project’s bracketed prefix convention, providing clear and focused context without extraneous detail.
✨ Finishing touches
  • 📝 Generate Docstrings
🧪 Generate unit tests
  • 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
🧪 Early access (Sonnet 4.5): enabled

We are currently testing the Sonnet 4.5 model, which is expected to improve code review quality. However, this model may lead to increased noise levels in the review comments. Please disable the early access features if the noise level causes any inconvenience.

Note:

  • Public repositories are always opted into early access features.
  • You can enable or disable early access features from the CodeRabbit UI or by updating the CodeRabbit configuration file.

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

@longlee0622 longlee0622 changed the title [None][Fix] Enable FP8 ContextMLA on GB300 [None][fix] Enable FP8 ContextMLA on GB300 Sep 30, 2025
@longlee0622
Copy link
Collaborator Author

/bot run

@longlee0622 longlee0622 self-assigned this Sep 30, 2025
@tensorrt-cicd
Copy link
Collaborator

PR_Github #20311 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20311 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15319 completed with status: 'FAILURE'

@longlee0622
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20334 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20334 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15338 completed with status: 'FAILURE'

@longlee0622
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20340 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20340 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15344 completed with status: 'FAILURE'

@longlee0622 longlee0622 force-pushed the enable-gb300-fp8-mha branch from e422f6a to 1c32f6c Compare October 9, 2025 01:36
@longlee0622
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20833 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20833 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15753 completed with status: 'FAILURE'

@longlee0622
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20854 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20854 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15771 completed with status: 'FAILURE'

Signed-off-by: Jonas Li <6110159+longlee0622@users.noreply.github.com>
@longlee0622 longlee0622 force-pushed the enable-gb300-fp8-mha branch from 1c32f6c to 882973e Compare October 9, 2025 09:49
@longlee0622
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20881 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20881 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15793 completed with status: 'FAILURE'

@longlee0622
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20930 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20930 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15831 completed with status: 'SUCCESS'

@longlee0622 longlee0622 merged commit 76a47c7 into NVIDIA:main Oct 10, 2025
5 checks passed
@longlee0622 longlee0622 deleted the enable-gb300-fp8-mha branch October 10, 2025 02:21
ZhanruiSunCh pushed a commit to ZhanruiSunCh/TensorRT-LLM that referenced this pull request Oct 10, 2025
Signed-off-by: Jonas Li <6110159+longlee0622@users.noreply.github.com>
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.

5 participants