[https://nvbugs/6422432][fix] Unwaive testcase by YihuiLu512 · Pull Request #17947 · NVIDIA/TensorRT-LLM

YihuiLu512 · 2026-08-19T03:54:36Z

Dev Engineer Review

Updated tests/integration/test_lists/waives.txt.
Removed obsolete DeepSeek R1 latency and DeepSeekV3Lite NVFP4 waivers.
Added waivers for GLM52 NVFP4, Qwen softmax attention, LTX2 CUDA-graph TRTLLM, DGX B200 performance alignment, Exaone attention backends, Qwen3 LoRA, standalone packaging, Kimi K3 generation performance, and related performance scenarios.
No API, performance, or error-handling code changed.
The available diff output did not show the complete entry-level changes. Verify waiver format, bug references, test paths, scope, and duplicate entries.

QA Engineer Review

Modified file: tests/integration/test_lists/waives.txt.
Added waivers for the GLM52, Qwen, LTX2, DGX B200, Exaone, Qwen3, standalone packaging, Kimi K3, and related performance scenarios.
Removed DeepSeek R1 latency and DeepSeekV3Lite NVFP4 waivers.
No test-db/ or qa/ files were modified.
Verdict: needs follow-up because CBTS coverage data is unavailable.

Description

Test Coverage

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2"

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.

YihuiLu512 · 2026-08-19T03:54:52Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2"

coderabbitai · 2026-08-19T03:59:21Z

Note

Reviews paused

It looks like this branch is under active development. To avoid overwhelming you with review comments due to an influx of new commits, CodeRabbit has automatically paused this review. You can configure this behavior by changing the reviews.auto_review.auto_pause_after_reviewed_commits setting.

Use the following commands to manage reviews:

@coderabbitai resume to resume automatic reviews.
@coderabbitai review to trigger a single review.

Use the checkboxes below for quick actions:

▶️ Resume reviews
🔍 Trigger review

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: aed7f500-5694-450d-9278-eea8bd454441

📥 Commits

Reviewing files that changed from the base of the PR and between 02d0f85 and 1496608.

📒 Files selected for processing (1)

tests/integration/test_lists/waives.txt

Included review availability: Your plan provides up to 12 included reviews per hour; 11 remain after this review.

Walkthrough

The change updates tests/integration/test_lists/waives.txt with current model, hardware, packaging, and performance waivers. It also removes obsolete DeepSeek waiver entries.

Changes

Integration test waiver maintenance

Layer / File(s)	Summary
Model and serving waivers `tests/integration/test_lists/waives.txt`	Adds waivers for GLM52, Qwen3 softmax attention, Exaone attention backends, and Qwen3 FP8 LoRA tests. Removes obsolete DeepSeek waivers.
Visual-generation and hardware waivers `tests/integration/test_lists/waives.txt`	Adds waivers for LTX2 CUDA-graph TRTLLM tests and DGX B200 layer-wise benchmark performance.
Performance and packaging waivers `tests/integration/test_lists/waives.txt`	Adds waivers for GB300 DeepSeek scenarios, standalone-package unit tests, and Kimi K3 generation performance.

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

Merge Risk: 🔵 Low · up to 14966

The PR re-enables a previously waived testcase. It is mergeable with owner awareness, but the affected GB200, DGX B300, and RTXPro6000D post-merge stages should be confirmed to ensure the waiver removal applies to the intended scope.

Possibly related PRs

NVIDIA/TensorRT-LLM#15195: Both PRs update tests/integration/test_lists/waives.txt with temporary test waivers.
NVIDIA/TensorRT-LLM#15585: Both PRs adjust integration test waiver entries in tests/integration/test_lists/waives.txt.

Suggested reviewers: bowenfu, tensorrt-cicd, trtllm-agent

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The description includes the required template and a test command, but the Description section does not explain the issue or solution. The test coverage section also does not identify the relevant tes…	Add a short Description that states which testcase is being unwaived and why. Expand Test Coverage with the relevant test or waiver changes and explain how the listed CI stages validate the change.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title follows the required NVBugs and type format and clearly identifies the change as a testcase unwaive fix.
Docstring Coverage	✅ Passed	No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0…
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.

Full details: Description check

Explanation

The description includes the required template and a test command, but the Description section does not explain the issue or solution. The test coverage section also does not identify the relevant tests or expected validation.

Full details: Docstring Coverage

Explanation

No functions found in the changed files to evaluate docstring coverage. Skipping docstring coverage check. Docstring coverage is scoped to functions touched by this diff. Analyzed 0 functions across 0 files.

Fix all pre-merge checks with AI

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create PR with unit tests

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

tensorrt-cicd · 2026-08-19T04:00:59Z

PR_Github #67268 [ run ] triggered by Bot. Commit: a8031c3 Link to invocation

tensorrt-cicd · 2026-08-19T08:36:40Z

PR_Github #67268 [ run ] completed with state SUCCESS. Commit: a8031c3
/LLM/main/L0_MergeRequest_PR pipeline #54793 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

YihuiLu512 · 2026-08-19T13:37:52Z

@QiJune Please add the following label so that I can verify whether the bug is reproducible: ci: post-merge approved.

YihuiLu512 · 2026-08-19T13:38:19Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-19T13:44:38Z

PR_Github #67453 [ run ] triggered by Bot. Commit: a8031c3 Link to invocation

tensorrt-cicd · 2026-08-19T21:33:06Z

PR_Github #67453 [ run ] completed with state SUCCESS. Commit: a8031c3
/LLM/main/L0_MergeRequest_PR pipeline #54954 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

github-actions · 2026-08-20T03:02:40Z

Removed the "ci: post-merge approved" label because @YihuiLu512 could not be verified as an active member of NVIDIA/trt-llm-ci-approvers. Ask a member of that team to apply it.

YihuiLu512 · 2026-08-20T03:02:45Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

YihuiLu512 · 2026-08-20T03:04:15Z

/bot kill

tensorrt-cicd · 2026-08-20T03:13:22Z

PR_Github #67635 [ kill ] triggered by Bot. Commit: a8031c3 Link to invocation

tensorrt-cicd · 2026-08-20T03:13:25Z

PR_Github #67635 [ kill ] completed with state SUCCESS. Commit: a8031c3
Successfully killed previous jobs for commit a8031c3

Link to invocation

YihuiLu512 · 2026-08-20T04:18:09Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-20T04:23:27Z

PR_Github #67653 [ run ] triggered by Bot. Commit: a8031c3 Link to invocation

tensorrt-cicd · 2026-08-20T05:12:57Z

PR_Github #67653 [ run ] completed with state FAILURE. Commit: a8031c3
/LLM/main/L0_MergeRequest_PR pipeline #55140 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

YihuiLu512 · 2026-08-20T07:38:00Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-20T07:44:40Z

PR_Github #67748 [ run ] triggered by Bot. Commit: a8031c3 Link to invocation

tensorrt-cicd · 2026-08-20T08:35:19Z

PR_Github #67748 [ run ] completed with state FAILURE. Commit: a8031c3
/LLM/main/L0_MergeRequest_PR pipeline #55227 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

YihuiLu512 · 2026-08-21T01:57:36Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-21T02:04:55Z

PR_Github #68062 [ run ] triggered by Bot. Commit: a8031c3 Link to invocation

YihuiLu512 · 2026-08-21T13:26:35Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-24T03:36:21Z

PR_Github #68646 [ run ] completed with state ABORTED. Commit: a8031c3

Link to invocation

tensorrt-cicd · 2026-08-24T06:46:03Z

PR_Github #68666 [ run ] completed with state FAILURE. Commit: a022956
/LLM/main/L0_MergeRequest_PR pipeline #56070 completed with status: 'UNSTABLE'

CI Report

⚠️ Multi-GPU Label Required:
Multi-GPU tests require the ci: full pre-merge approved label on this PR. Ask a member of NVIDIA/trt-llm-ci-approvers to add the label, then re-trigger CI with the same bot command (no rebase needed).

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

Link to invocation

YihuiLu512 · 2026-08-24T07:55:49Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-24T08:01:22Z

PR_Github #68748 [ run ] triggered by Bot. Commit: a022956 Link to invocation

tensorrt-cicd · 2026-08-24T16:27:16Z

PR_Github #68748 [ run ] completed with state SUCCESS. Commit: a022956
/LLM/main/L0_MergeRequest_PR pipeline #56147 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

YihuiLu512 · 2026-08-25T03:36:26Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-1, RTXPro6000D-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-25T03:42:08Z

PR_Github #68989 [ run ] triggered by Bot. Commit: 3aa54cc Link to invocation

YihuiLu512 · 2026-08-25T04:19:45Z

/bot run --extra-stage "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-25T04:20:59Z

PR_Github #68989 [ run ] completed with state FAILURE. Commit: 3aa54cc
/LLM/main/L0_MergeRequest_PR pipeline #56365 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

tensorrt-cicd · 2026-08-25T04:25:24Z

PR_Github #68996 [ run ] triggered by Bot. Commit: 3aa54cc Link to invocation

tensorrt-cicd · 2026-08-25T10:52:59Z

PR_Github #68996 [ run ] completed with state FAILURE. Commit: 3aa54cc
/LLM/main/L0_MergeRequest_PR pipeline #56372 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

YihuiLu512 · 2026-08-26T01:06:58Z

/bot run --extra-stage "DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-26T01:12:58Z

PR_Github #69254 [ run ] triggered by Bot. Commit: 02d0f85 Link to invocation

YihuiLu512 · 2026-08-26T05:29:09Z

/bot kill

tensorrt-cicd · 2026-08-26T05:35:38Z

PR_Github #69324 [ kill ] triggered by Bot. Commit: 02d0f85 Link to invocation

tensorrt-cicd · 2026-08-26T05:35:42Z

PR_Github #69254 [ run ] completed with state ABORTED. Commit: 02d0f85
/LLM/main/L0_MergeRequest_PR pipeline #56612 completed with status: 'FAILURE'

CI Report

⚠️ Action Required:

Please check the failed tests and fix your PR
If you cannot view the failures, ask the CI triggerer to share details
Once fixed, request an NVIDIA team member to trigger CI again

CI Agent Failure Analysis

Link to invocation

tensorrt-cicd · 2026-08-26T05:35:52Z

PR_Github #69324 [ kill ] completed with state SUCCESS. Commit: 02d0f85
Successfully killed previous jobs for commit 02d0f85

Link to invocation

YihuiLu512 · 2026-08-26T05:36:57Z

/bot run --stage-list "GB200-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-1, DGX_B300-4_GPUs-PyTorch-Post-Merge-2" --disable-fail-fast

tensorrt-cicd · 2026-08-26T05:43:21Z

PR_Github #69330 [ run ] triggered by Bot. Commit: 02d0f85 Link to invocation

Signed-off-by: Yihui Lu <269394165+YihuiLu512@users.noreply.github.com>

YihuiLu512 · 2026-08-26T08:52:34Z

/bot run --disable-fail-fast

tensorrt-cicd · 2026-08-26T08:58:48Z

PR_Github #69399 [ run ] triggered by Bot. Commit: 1496608 Link to invocation

tensorrt-cicd · 2026-08-26T09:00:45Z

PR_Github #69330 [ run ] completed with state ABORTED. Commit: 02d0f85

Link to invocation

tensorrt-cicd · 2026-08-26T11:31:37Z

PR_Github #69399 [ run ] completed with state SUCCESS. Commit: 1496608
/LLM/main/L0_MergeRequest_PR pipeline #56736 completed with status: 'SUCCESS'

CI Report

Link to invocation

github-actions Bot assigned YihuiLu512 Aug 19, 2026

YihuiLu512 enabled auto-merge (squash) August 19, 2026 03:55

coderabbitai Bot mentioned this pull request Aug 19, 2026

[https://nvbugs/6602927][test] Unwaive fixed KV cache V2 deadlock case #17962

Merged

1 task

BowenFu approved these changes Aug 19, 2026

View reviewed changes

coderabbitai Bot mentioned this pull request Aug 19, 2026

[https://nvbugs/6631019][fix] Add a torch_compiling(enable) contextmanager to _torch/utils.py and lower… #17949

Merged

2 tasks

YihuiLu512 added the ci: post-merge approved Approved by TRT-LLM CI approvers for broad post-merge CI requests label Aug 20, 2026

github-actions Bot removed the ci: post-merge approved Approved by TRT-LLM CI approvers for broad post-merge CI requests label Aug 20, 2026

QiJune added the ci: post-merge approved Approved by TRT-LLM CI approvers for broad post-merge CI requests label Aug 20, 2026

yuxianq added the ci: full pre-merge approved label Aug 24, 2026

YihuiLu512 force-pushed the dev-fix-15-J14108-N6422432 branch from a022956 to 3aa54cc Compare August 25, 2026 03:22

YihuiLu512 force-pushed the dev-fix-15-J14108-N6422432 branch from 3aa54cc to 02d0f85 Compare August 26, 2026 01:05

BowenFu approved these changes Aug 26, 2026

View reviewed changes

[https://nvbugs/6422432][fix] Unwaive testcase

1496608

Signed-off-by: Yihui Lu <269394165+YihuiLu512@users.noreply.github.com>

YihuiLu512 force-pushed the dev-fix-15-J14108-N6422432 branch from 02d0f85 to 1496608 Compare August 26, 2026 08:52

YihuiLu512 merged commit 8bde010 into NVIDIA:main Aug 26, 2026
10 checks passed

Uh oh!

Conversation

YihuiLu512 commented Aug 19, 2026 • edited by coderabbitai Bot Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Dev Engineer Review

QA Engineer Review

Description

Test Coverage

PR Checklist

GitHub Bot Help

Uh oh!

YihuiLu512 commented Aug 19, 2026

Uh oh!

coderabbitai Bot commented Aug 19, 2026 • edited Loading Uh oh! There was an error while loading. Please reload this page.

Uh oh!

Reviews paused

Walkthrough

Changes

❌ Failed checks (1 warning)

Uh oh!

tensorrt-cicd commented Aug 19, 2026

Uh oh!

tensorrt-cicd commented Aug 19, 2026

Uh oh!

YihuiLu512 commented Aug 19, 2026

Uh oh!

YihuiLu512 commented Aug 19, 2026

Uh oh!

tensorrt-cicd commented Aug 19, 2026

Uh oh!

tensorrt-cicd commented Aug 19, 2026

Uh oh!

github-actions Bot commented Aug 20, 2026

Uh oh!

YihuiLu512 commented Aug 20, 2026

Uh oh!

YihuiLu512 commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

YihuiLu512 commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

YihuiLu512 commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

tensorrt-cicd commented Aug 20, 2026

Uh oh!

YihuiLu512 commented Aug 21, 2026

Uh oh!

tensorrt-cicd commented Aug 21, 2026

Uh oh!

YihuiLu512 commented Aug 21, 2026

Uh oh!

tensorrt-cicd commented Aug 24, 2026

Uh oh!

tensorrt-cicd commented Aug 24, 2026

Uh oh!

YihuiLu512 commented Aug 24, 2026

Uh oh!

tensorrt-cicd commented Aug 24, 2026

Uh oh!

tensorrt-cicd commented Aug 24, 2026

Uh oh!

YihuiLu512 commented Aug 25, 2026

Uh oh!

tensorrt-cicd commented Aug 25, 2026

Uh oh!

YihuiLu512 commented Aug 25, 2026

Uh oh!

tensorrt-cicd commented Aug 25, 2026

Uh oh!

tensorrt-cicd commented Aug 25, 2026

Uh oh!

YihuiLu512 commented Aug 19, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Aug 19, 2026 •

edited

Loading