Skip to content

[https://nvbugs/6403909][fix] unwaive an attention backend test - #15964

Merged
tburt-nv merged 1 commit into
NVIDIA:mainfrom
yuxianq:fix/remove-nvbug-6403909-waive
Jul 7, 2026
Merged

[https://nvbugs/6403909][fix] unwaive an attention backend test#15964
tburt-nv merged 1 commit into
NVIDIA:mainfrom
yuxianq:fix/remove-nvbug-6403909-waive

Conversation

@yuxianq

@yuxianq yuxianq commented Jul 6, 2026

Copy link
Copy Markdown
Collaborator

Summary by CodeRabbit

  • Tests
    • Removed one test waiver, allowing the related attention backend integration test to run again.

Description

Remove the DGX H100 waiver for
test_attention_backend[qwen2_0_5b_gqa_hd64-ctx-bf16-HND-p32-v1], tracked by
NVBug 6403909. The failure did not reproduce during repeated H100 stress testing,
so this test should return to normal CI coverage.

Test Coverage

  • Ran unittest/_torch/attention/test_attention_backends.py::test_attention_backend
    in 100 fresh pytest processes on a single H100 at commit
    f50ca53daedc9957a7f87cb00de205b5f4323061: 932 tests passed per run,
    93,200 total.
  • Parsed all 558 entries in waives.txt successfully after the removal.
  • Pre-commit waiver checks passed, including duplicate detection and AST validation.

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)

  • If PR introduces API changes, an appropriate PR label is added - either api-compatible or api-breaking. For api-breaking, include BREAKING in the PR title.

  • 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

To see a list of available CI bot commands, please comment /bot help.

Signed-off-by: Yuxian Qiu <142763828+yuxianq@users.noreply.github.com>
@yuxianq

yuxianq commented Jul 6, 2026

Copy link
Copy Markdown
Collaborator Author

/bot run

@coderabbitai

coderabbitai Bot commented Jul 6, 2026

Copy link
Copy Markdown
Contributor

Review Change Stack

📝 Walkthrough

Walkthrough

A single SKIP entry referencing nvbugs/6403909 for the test_attention_backend test case in DGX_H100 was removed from the waives.txt test list file.

Changes

Waive list update

Layer / File(s) Summary
Remove obsolete waive entry
tests/integration/test_lists/waives.txt
Removed the SKIP entry for test_attention_backend[...] on DGX_H100 that referenced nvbugs/6403909.

Estimated code review effort: 1 (Trivial) | ~2 minutes

Possibly related PRs

  • NVIDIA/TensorRT-LLM#15898: Both PRs edit the same waives.txt file by adding/removing skip entries tied to nvbugs references.
  • NVIDIA/TensorRT-LLM#15901: Both PRs modify the same waive list file by removing/adding SKIP entries for nearby nvbugs waiver sources.

Suggested reviewers: pcastonguay

🚥 Pre-merge checks | ✅ 5
✅ Passed checks (5 passed)
Check name Status Explanation
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Linked Issues check ✅ Passed Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check ✅ Passed Check skipped because no linked issues were found for this pull request.
Title check ✅ Passed The title matches the main change and follows the required ticket/type format.
Description check ✅ Passed The description explains the change, motivation, test coverage, and includes the checklist.
✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

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

@yuxianq

yuxianq commented Jul 6, 2026

Copy link
Copy Markdown
Collaborator Author

/bot run --disable-fail-fast

@yuxianq yuxianq changed the title [https://nvbugs/6403909][fix] remove attention backend test waiver [https://nvbugs/6403909][fix] unwaive an attention backend test Jul 6, 2026
@yuxianq
yuxianq requested a review from tburt-nv July 6, 2026 07:22
@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57721 [ run ] triggered by Bot. Commit: 3279cff Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57722 [ run ] triggered by Bot. Commit: 3279cff Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57721 [ run ] completed with state ABORTED. Commit: 3279cff

Link to invocation

@tensorrt-cicd

Copy link
Copy Markdown
Collaborator

PR_Github #57722 [ run ] completed with state SUCCESS. Commit: 3279cff
/LLM/main/L0_MergeRequest_PR pipeline #46432 completed with status: 'SUCCESS'

CI Report

Link to invocation

Comment thread tests/integration/test_lists/waives.txt
@tburt-nv
tburt-nv merged commit d0d5069 into NVIDIA:main Jul 7, 2026
14 checks passed
crazydemo pushed a commit to crazydemo/TensorRT-LLM that referenced this pull request Jul 8, 2026
…IA#15964)

Signed-off-by: Yuxian Qiu <142763828+yuxianq@users.noreply.github.com>
BrianLi23 pushed a commit to BrianLi23/TensorRT-LLM that referenced this pull request Jul 9, 2026
…IA#15964)

Signed-off-by: Yuxian Qiu <142763828+yuxianq@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