KEMBAR78
[https://nvbugs/5501820][fix] Add requirements for numba-cuda version to WAR mem corruption by pengbowang-nv · Pull Request #7992 · NVIDIA/TensorRT-LLM · GitHub
Skip to content

Conversation

@pengbowang-nv
Copy link
Collaborator

@pengbowang-nv pengbowang-nv commented Sep 25, 2025

Summary by CodeRabbit

  • Chores

    • Added numba-cuda (>=0.19.0) to installation requirements to ensure the needed GPU dependency is installed.
  • Tests

    • Re-enabled the multi-GPU linear test suite by removing a module-level skip, increasing test coverage and execution.

Description

The pynvjitlink dependency, introduced in trtllm workflow by numba-cuda<0.19.0 and used during torch compile, linked to a static nvptxcompiler of 12.9, which will cause mem corruption in CUDA 13.0 env. This PR is a WAR for it.

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.

@pengbowang-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #19936 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #19936 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15009 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20012 [ run ] triggered by Bot

@pengbowang-nv
Copy link
Collaborator Author

/bot run --add-multi-gpu-test --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20056 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20012 [ run ] completed with state ABORTED
LLM/main/L0_MergeRequest_PR #15072 (Blue Ocean) completed with status: ABORTED

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20056 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #15109 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --add-multi-gpu-test --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20081 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20081 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #15129 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --add-multi-gpu-test --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20105 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20105 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15148 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --add-multi-gpu-test

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20216 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20216 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15248 completed with status: 'SUCCESS'

@pengbowang-nv pengbowang-nv marked this pull request as ready for review October 8, 2025 02:09
@coderabbitai
Copy link
Contributor

coderabbitai bot commented Oct 8, 2025

📝 Walkthrough

Walkthrough

Added a new dependency entry to requirements.txt. Removed a module-level skip in a multi-GPU torch linear test file to allow the tests to run.

Changes

Cohort / File(s) Summary
Dependencies
requirements.txt
Added numba-cuda>=0.19.0 with comment referencing nvbugs/5501820.
Tests (Torch Multi-GPU Linear)
tests/unittest/_torch/multi_gpu/test_linear.py
Removed module-level skip to enable running the test module.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)
Check name Status Explanation Resolution
Description Check ⚠️ Warning The PR description uses the template’s Description and checklist sections but leaves the Test Coverage section empty with only placeholder comments instead of listing any relevant tests, so it does not satisfy the requirement to specify test cases safeguarding the change. Please populate the Test Coverage section with actual tests that validate the updated numba-cuda requirement and remove the placeholder comments so that all template sections are fully completed.
✅ Passed checks (2 passed)
Check name Status Explanation
Title Check ✅ Passed The title follows the required format by including the NVBugs ID and fix type, and it concisely summarizes that the PR adds a version constraint for numba-cuda as a workaround for memory corruption. It accurately reflects the primary change without unnecessary detail.
Docstring Coverage ✅ Passed No functions found in the changes. Docstring coverage check skipped.
✨ 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 e30d9ac and a3e8d18.

📒 Files selected for processing (2)
  • requirements.txt (1 hunks)
  • tests/unittest/_torch/multi_gpu/test_linear.py (0 hunks)
💤 Files with no reviewable changes (1)
  • tests/unittest/_torch/multi_gpu/test_linear.py
🔇 Additional comments (1)
requirements.txt (1)

79-79: Keep constraints.txt in sync with the new floor

Because installs are run with -c constraints.txt, we need to make sure that file does not still pin numba-cuda below 0.19.0 (or omit the entry entirely, if you prefer relaxing there). Otherwise resolution will fail and the WAR will not apply. Please update or double-check the constraint accordingly.


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.

@pengbowang-nv
Copy link
Collaborator Author

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

Copy link
Collaborator

@VALLIS-NERIA VALLIS-NERIA 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

PR_Github #20758 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20758 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15688 completed with status: 'FAILURE'

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
@pengbowang-nv pengbowang-nv force-pushed the fix-mem-corruption-at-exit branch from a3e8d18 to 0f943b4 Compare October 8, 2025 08:36
@pengbowang-nv
Copy link
Collaborator Author

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20780 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20780 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15708 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20832 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20832 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15752 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1"

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20850 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20850 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15769 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20869 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20869 [ run ] completed with state FAILURE
/LLM/main/L0_MergeRequest_PR pipeline #15785 completed with status: 'FAILURE'

@pengbowang-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20942 [ run ] triggered by Bot

@tensorrt-cicd
Copy link
Collaborator

PR_Github #20942 [ run ] completed with state SUCCESS
/LLM/main/L0_MergeRequest_PR pipeline #15842 completed with status: 'SUCCESS'

@pengbowang-nv pengbowang-nv merged commit 7da4b05 into NVIDIA:main Oct 10, 2025
5 checks passed
@pengbowang-nv pengbowang-nv deleted the fix-mem-corruption-at-exit branch October 10, 2025 02:21
pengbowang-nv added a commit to pengbowang-nv/TensorRT-LLM that referenced this pull request Oct 16, 2025
… to WAR mem corruption (NVIDIA#7992)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
pengbowang-nv added a commit that referenced this pull request Oct 20, 2025
… to WAR mem corruption (#7992) (#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 20, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 20, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 20, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 20, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 21, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 21, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 22, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 23, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
mikeiovine pushed a commit to mikeiovine/TensorRT-LLM that referenced this pull request Oct 24, 2025
… to WAR mem corruption (NVIDIA#7992) (NVIDIA#8414)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com>
yufeiwu-nv pushed a commit to yufeiwu-nv/TensorRT-LLM that referenced this pull request Oct 24, 2025
… to WAR mem corruption (NVIDIA#7992)

Signed-off-by: Pengbo Wang <221450789+pengbowang-nv@users.noreply.github.com>
Signed-off-by: yufeiwu-nv <230315618+yufeiwu-nv@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.

3 participants