Skip to content

[None][fix] Fix topk outIndices when using vectorized_process #9404

New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

Sign up for GitHub

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Jump to bottom
Merged
chang-l merged 1 commit into from
Nov 24, 2025
Merged

[None][fix] Fix topk outIndices when using vectorized_process #9404

chang-l merged 1 commit into from
Nov 24, 2025

Conversation

@yweng0828
Copy link
Collaborator

@yweng0828 yweng0828 commented Nov 24, 2025 •
edited
Loading

@coderabbitai summary

Description

Fix the write-back value when using vectorized_process in customer topk to fix the issue of negative indexes.

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 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.

@coderabbitai
Copy link
Contributor

coderabbitai bot commented Nov 24, 2025

📝 Walkthrough

Walkthrough

A conditional logic adjustment is made to the final storage loop in the TopK kernel. When stride1 equals 1, the output indices are calculated differently to align with the vectorized execution path, ensuring consistent global-to-local index mapping across both code paths.

Changes

Cohort / File(s) Summary
TopK kernel index mapping fix
cpp/tensorrt_llm/kernels/indexerTopK.cu		 Modified final storage loop with conditional logic: when stride1 == 1, outIndices directly uses smemOutput values without rowStart subtraction; otherwise retains previous behavior. Aligns indexing between vectorized and non-vectorized paths.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

  • Verify that the stride1 == 1 condition correctly identifies the vectorized path scenario
  • Confirm smemOutput[i] contains properly pre-computed indices and doesn't require offset adjustment
  • Validate the index mapping produces correct output across both conditional branches

Pre-merge checks and finishing touches

❌ Failed checks (2 warnings)
Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description is mostly the template with placeholders unfilled; Description, Test Coverage, and PR Checklist sections contain only comments without actual content. Fill in the Description section explaining the issue and solution, add relevant test coverage details in Test Coverage section, and check the PR Checklist items before submitting.
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.
✅ Passed checks (1 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly summarizes the fix: correcting topk outIndices behavior when using vectorized_process, and follows the required format with [None][fix] prefix.
✨ Finishing touches
  • 📝 Generate docstrings
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Post copyable unit tests in a comment
📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 2c869f2 and fd64266.

📒 Files selected for processing (1)
  • cpp/tensorrt_llm/kernels/indexerTopK.cu (1 hunks)
🧰 Additional context used
🧠 Learnings (2)
📚 Learning: 2025-08-09T20:57:04.084Z 
Learnt from: sklevtsov-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 3294
File: cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_gemm_tma_warp_specialized_input.cu:118-127
Timestamp: 2025-08-09T20:57:04.084Z
Learning: In the CUTLASS MoE finalize fusion implementation (cpp/tensorrt_llm/kernels/cutlass_kernels/moe_gemm/moe_gemm_tma_warp_specialized_input.cu), when setting `fused_finalize_epilogue.stride_final_output` with shape `(hidden_size, num_output_tokens, 1)`, the `num_rows_in_final_output` should be set to `num_output_tokens` (not `hidden_size`) because of a swap+transpose operation that maps rows of the output tensor to `hidden_size` and columns to `num_output_tokens`.

Applied to files:

  • cpp/tensorrt_llm/kernels/indexerTopK.cu
📚 Learning: 2025-08-08T05:10:38.906Z 
Learnt from: sklevtsov-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 3294
File: cpp/tensorrt_llm/cutlass_extensions/include/cutlass_extensions/epilogue/fusion/sm90_visitor_scatter.hpp:0-0
Timestamp: 2025-08-08T05:10:38.906Z
Learning: The ScaledAccPerRowBiasPerColScaleScatter fusion in CUTLASS extensions (cpp/tensorrt_llm/cutlass_extensions/include/cutlass_extensions/epilogue/fusion/sm90_visitor_scatter.hpp) is specifically designed for per-column scaling factors only, so it uses a fixed Stride<_0,_1,int64_t> rather than conditional stride logic.

Applied to files:

  • cpp/tensorrt_llm/kernels/indexerTopK.cu
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
  • GitHub Check: Pre-commit Check
🔇 Additional comments (1)
cpp/tensorrt_llm/kernels/indexerTopK.cu (1)

573-581: Index adjustment now correctly reflects vectorized vs. strided indexing

When stride1 == 1, processBins is only reached via vectorized_process(logits + rowStart, rowEnd - rowStart, ...), so the idx it sees (and thus the values written into smemOutput) are already row-local in [0, rowEnd - rowStart). Dropping the - rowStart in this case fixes the previous shift for contiguous layouts.

For stride1 != 1, idx still ranges over [rowStart, rowEnd) in the non‑vectorized loop, so keeping smemOutput[i] - rowStart preserves the intended row‑local semantics. The mergeBlocks path uses rowStart == 0, so behavior there remains unchanged.

This looks like the right minimal fix for the vectorized_process path without impacting the other configurations.

Tip

📝 Customizable high-level summaries are now available in beta!

You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.

  • Provide your own instructions using the high_level_summary_instructions setting.
  • Format the summary however you like (bullet lists, tables, multi-section layouts, contributor stats, etc.).
  • Use high_level_summary_in_walkthrough to move the summary from the description to the walkthrough section.

Example instruction:

"Divide the high-level summary into five sections:

  1. 📝 Description — Summarize the main change in 50–60 words, explaining what was done.
  2. 📓 References — List relevant issues, discussions, documentation, or related PRs.
  3. 📦 Dependencies & Requirements — Mention any new/updated dependencies, environment variable changes, or configuration updates.
  4. 📊 Contributor Summary — Include a Markdown table showing contributions:
    | Contributor | Lines Added | Lines Removed | Files Changed |
  5. ✔️ Additional Notes — Add any extra reviewer context.
    Keep each section concise (under 200 words) and use bullet or numbered lists for clarity."

Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later.

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.

@yweng0828
Copy link
Collaborator Author

yweng0828 commented Nov 24, 2025

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

tensorrt-cicd commented Nov 24, 2025

PR_Github #25547 [ run ] triggered by Bot. Commit: fd64266

dcampora
dcampora approved these changes Nov 24, 2025
View reviewed changes
Copy link
Collaborator

@dcampora dcampora left a comment

Choose a reason for hiding this comment

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

LGTM.

@tensorrt-cicd
Copy link
Collaborator

tensorrt-cicd commented Nov 24, 2025

PR_Github #25547 [ run ] completed with state SUCCESS. Commit: fd64266
/LLM/main/L0_MergeRequest_PR pipeline #19346 completed with status: 'SUCCESS'
Pipeline passed with automatic retried tests. Check the rerun report for details.

@chang-l chang-l merged commit 336593c into NVIDIA:main Nov 24, 2025
8 of 11 checks passed
codego7250 pushed a commit to codego7250/TensorRT-LLM that referenced this pull request Dec 11, 2025
@yweng0828 @codego7250
[None][fix] Fix topk outIndices when using vectorized_process (NVIDIA…
5d9a4b0 
…#9404)

Signed-off-by: Yue Weng <[email protected]>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@dcampora dcampora dcampora approved these changes

@ChristinaZ ChristinaZ Awaiting requested review from ChristinaZ

@chang-l chang-l Awaiting requested review from chang-l

Assignees

No one assigned

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

4 participants

@yweng0828 @tensorrt-cicd @dcampora @chang-l