Skip to content

Navigation Menu

Sign in
Appearance settings

Search code, repositories, users, issues, pull requests...

Provide feedback

We read every piece of feedback, and take your input very seriously.

Saved searches

Use saved searches to filter your results more quickly
Appearance settings

[https://nvbugs/5916151][fix] Unwaive test_fused_moe_w4a8_nvfp4_fp8[TRTLLM]#12400

Merged
EmmaQiaoCh merged 1 commit intoNVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom
xxi-nv:unwaive-nvbug-5916151xxi-nv/TensorRT-LLM:unwaive-nvbug-5916151Copy head branch name to clipboard
Mar 23, 2026
Merged

[https://nvbugs/5916151][fix] Unwaive test_fused_moe_w4a8_nvfp4_fp8[TRTLLM]#12400
EmmaQiaoCh merged 1 commit intoNVIDIA:mainNVIDIA/TensorRT-LLM:mainfrom
xxi-nv:unwaive-nvbug-5916151xxi-nv/TensorRT-LLM:unwaive-nvbug-5916151Copy head branch name to clipboard

Conversation

@xxi-nv
Copy link
Copy Markdown
Collaborator

@xxi-nv xxi-nv commented Mar 20, 2026 •
edited
Loading

Summary

Test plan

  • Verified on B300 (umb-b300-003, 4x NVIDIA B300 SXM6 AC): 7/7 passes (1 initial + 1 CUTLASS + 5 repeat runs)
  • CI should pass with the test unwaived

Summary by CodeRabbit

  • Tests
    • Re-enabled a previously waived test case for fused MOE (mixed expert operations) with specific numerical precision configurations.

@xxi-nv
[https://nvbugs/5916151][fix] Unwaive test_fused_moe_w4a8_nvfp4_fp8[T…
f5bfe7a 
…RTLLM]

The illegal memory access bug (NVBug 5916151) was fixed by PR NVIDIA#11502
which corrected the .cuda() call ordering in the test. Verified on B300
with 7/7 passes (1 initial + 1 CUTLASS + 5 repeat runs).

Signed-off-by: xxi <xxi@nvidia.com>
@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai Bot commented Mar 20, 2026 •
edited
Loading

📝 Walkthrough

Walkthrough

One integration test waiver entry is removed from the test waiver list, allowing the previously skipped test case unittest/_torch/modules/test_fused_moe.py::test_fused_moe_w4a8_nvfp4_fp8[TRTLLM] to run in integration testing.

Changes

Cohort / File(s) Summary
Test Waiver Removal
tests/integration/test_lists/waives.txt		 Removed one test waiver entry for test_fused_moe_w4a8_nvfp4_fp8[TRTLLM] that was previously marked as SKIP, enabling this test case to execute.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~10 minutes

🚥 Pre-merge checks | ✅ 3
✅ Passed checks (3 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly identifies the main change: removing a waiver for a specific test case related to NVBug 5916151, which aligns with the changeset.
Docstring Coverage ✅ Passed No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check.
Description check ✅ Passed The PR description is clear and well-structured, explaining the issue, the fix, and verification details.

✏️ Tip: You can configure your own custom pre-merge checks in the settings.

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests
📝 Coding Plan
  • Generate coding plan for human review comments

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

Tip

CodeRabbit can suggest fixes for GitHub Check annotations.

Configure the reviews.tools.github-checks setting to adjust the time to wait for GitHub Checks to complete.

@xxi-nv
Copy link
Copy Markdown
Collaborator Author

xxi-nv commented Mar 20, 2026

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 20, 2026

PR_Github #39707 [ run ] triggered by Bot. Commit: f5bfe7a Link to invocation

@xxi-nv
Copy link
Copy Markdown
Collaborator Author

xxi-nv commented Mar 20, 2026

/bot run --stage-list "DGX_B300-4_GPUs-PyTorch-Post-Merge-2"

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 20, 2026

PR_Github #39710 [ run ] triggered by Bot. Commit: f5bfe7a Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 20, 2026

PR_Github #39710 [ run ] completed with state SUCCESS. Commit: f5bfe7a
/LLM/main/L0_MergeRequest_PR pipeline #30908 (Partly Tested) completed with status: 'SUCCESS'

CI Report

Link to invocation

@xxi-nv
Copy link
Copy Markdown
Collaborator Author

xxi-nv commented Mar 20, 2026

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 20, 2026

PR_Github #39734 [ run ] triggered by Bot. Commit: f5bfe7a Link to invocation

@xxi-nv xxi-nv requested a review from EmmaQiaoCh March 20, 2026 11:59
@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 20, 2026

PR_Github #39734 [ run ] completed with state SUCCESS. Commit: f5bfe7a
/LLM/main/L0_MergeRequest_PR pipeline #30929 completed with status: 'SUCCESS'

CI Report

Link to invocation

@xxi-nv
Copy link
Copy Markdown
Collaborator Author

xxi-nv commented Mar 22, 2026

/bot run --disable-fail-fast

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 22, 2026

PR_Github #39847 [ run ] triggered by Bot. Commit: f5bfe7a Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Mar 23, 2026

PR_Github #39847 [ run ] completed with state SUCCESS. Commit: f5bfe7a
/LLM/main/L0_MergeRequest_PR pipeline #31022 completed with status: 'SUCCESS'

CI Report

Link to invocation

@EmmaQiaoCh EmmaQiaoCh merged commit 53dc38e into NVIDIA:main Mar 23, 2026
8 of 10 checks passed
longcheng-nv pushed a commit to longcheng-nv/TensorRT-LLM that referenced this pull request Mar 31, 2026
@xxi-nv @longcheng-nv
[https://nvbugs/5916151][fix] Unwaive test_fused_moe_w4a8_nvfp4_fp8[T…
0cddaf2 
…RTLLM] (NVIDIA#12400)

Signed-off-by: xxi <xxi@nvidia.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@EmmaQiaoCh EmmaQiaoCh EmmaQiaoCh approved these changes

Assignees

@xxi-nv xxi-nv

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@xxi-nv @tensorrt-cicd @EmmaQiaoCh
Morty Proxy This is a proxified and sanitized view of the page, visit original site.