Skip to content

[None][test] Remove low priority QA perf test cases#13374

Merged
yufeiwu-nv merged 2 commits into
NVIDIA:mainfrom
yufeiwu-nv:fix_RTX
Apr 24, 2026
Merged

[None][test] Remove low priority QA perf test cases#13374
yufeiwu-nv merged 2 commits into
NVIDIA:mainfrom
yufeiwu-nv:fix_RTX

Conversation

@yufeiwu-nv
Copy link
Copy Markdown
Collaborator

@yufeiwu-nv yufeiwu-nv commented Apr 23, 2026 •
edited
Loading

Signed-off-by: yufeiwu-nv [email protected]

Summary by CodeRabbit

  • Tests
    • Removed MoE backend configuration overrides from model pattern configurations for DeepSeek, GPT-OSS 120B, and Qwen3 models.

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

  • 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.

@yufeiwu-nv yufeiwu-nv requested a review from a team as a code owner April 23, 2026 08:56
@coderabbitai
Copy link
Copy Markdown
Contributor

coderabbitai Bot commented Apr 23, 2026 •
edited
Loading

📝 Walkthrough

Walkthrough

Removes explicit MoE backend configuration overrides (moe_config.backend: 'TRTLLM') from multiple model-pattern configurations in the PyTorch performance test suite. Models matching these patterns will no longer force the MoE backend to TRTLLM via this config mapping.

Changes

Cohort / File(s) Summary
MoE Backend Configuration Overrides
tests/integration/defs/perf/pytorch_model_config.py		 Removes explicit moe_config.backend: 'TRTLLM' overrides from DeepSeek fp4/max-tokens, GPT-OSS 120B max-throughput/min-latency, and Qwen3 fp4-on-B200 model patterns. Also removes enable_attention_dp disable flag from Qwen3 pattern.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description is incomplete and does not provide the required information. The Description and Test Coverage sections are entirely empty, leaving no explanation of the changes, rationale, or testing approach. Add a clear Description explaining what MoE backend configurations were removed and why, and document the relevant test cases that validate these changes.
✅ Passed checks (4 passed)
Check name Status Explanation
Title check ✅ Passed The title accurately reflects the main change: removing explicit MoE backend configuration overrides from performance test model patterns, which aligns with removing low-priority QA test configurations.
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.

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

✨ Finishing Touches
🧪 Generate unit tests (beta)
  • Create PR with unit tests

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

@yufeiwu-nv
Copy link
Copy Markdown
Collaborator Author

yufeiwu-nv commented Apr 23, 2026

/bot skip --comment "only test list modify"

@yufeiwu-nv yufeiwu-nv enabled auto-merge (squash) April 23, 2026 12:42
@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 23, 2026

PR_Github #45184 [ skip ] triggered by Bot. Commit: 4d1739e Link to invocation

@yufeiwu-nv yufeiwu-nv requested a review from a team as a code owner April 23, 2026 12:51
@yufeiwu-nv
Copy link
Copy Markdown
Collaborator Author

yufeiwu-nv commented Apr 23, 2026

/bot skip --comment "only test list modify"

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 23, 2026

PR_Github #45186 [ skip ] triggered by Bot. Commit: c11b060 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 23, 2026

PR_Github #45184 [ skip ] completed with state ABORTED. Commit: 4d1739e

Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 23, 2026

PR_Github #45186 [ skip ] completed with state SUCCESS. Commit: c11b060
Skipping testing for commit c11b060

Link to invocation

ruodil
ruodil reviewed Apr 24, 2026
View reviewed changes
Comment thread tests/integration/test_lists/qa/llm_perf_sanity.yml
@yufeiwu-nv
fix
664d13a 
Signed-off-by: yufeiwu-nv <[email protected]>
@yufeiwu-nv
fix according to reviewer
3d74005 
Signed-off-by: yufeiwu-nv <[email protected]>
@yufeiwu-nv
Copy link
Copy Markdown
Collaborator Author

yufeiwu-nv commented Apr 24, 2026

/bot skip --comment "only test list modify"

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 24, 2026

PR_Github #45326 [ skip ] triggered by Bot. Commit: 3d74005 Link to invocation

@tensorrt-cicd
Copy link
Copy Markdown
Collaborator

tensorrt-cicd commented Apr 24, 2026

PR_Github #45326 [ skip ] completed with state SUCCESS. Commit: 3d74005
Skipping testing for commit 3d74005

Link to invocation

@yufeiwu-nv yufeiwu-nv merged commit bcf7dac into NVIDIA:main Apr 24, 2026
5 checks passed
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Reviewers

@ruodil ruodil ruodil approved these changes

Assignees

@yufeiwu-nv yufeiwu-nv

Labels

None yet

Projects

None yet

Milestone

No milestone

Development

Successfully merging this pull request may close these issues.

3 participants

@yufeiwu-nv @tensorrt-cicd @ruodil