[None][test] Remove closed-bug waive entries for release/1.3 by QiJune · Pull Request #20590 · NVIDIA/TensorRT-LLM

QiJune · 2026-10-19T08:33:14Z

Description

Remove 5 waive entries on release/1.3 whose NVBugs are closed and whose tests passed on the waived platform in pre-merge CI (L0_MergeRequest_PR #342 and #375, both run with post-merge extra stages). The branch is rebased on release/1.3 at 7f1b51c; the 5 removals are unchanged since #375.

NVBug	Entries	Evidence (PASSED in both #342 and #375)
6481323	1	DGX_B300-4_GPUs-PyTorch-Post-Merge-2
6428057	1	DGX_B200 / DGX_B300 4-GPU Post-Merge-1
6762288	1 (triton variant)	DGX_H100-2_GPUs-PyTorch-GptOss-1
6700265	1 (DGX_H100)	DGX_H100-PyTorch-5
6075431	1	A100X / B200_PCIe PackageSanityCheck

Closed bugs handled by other release/1.3 PRs: 6893730 (#20592), the 6762288 cutlass variant and 6672360 (#20632), 6278337 (#20585).

Kept waived (closed bugs, not yet verified)

These closed bugs keep their waives because their tests did not run on the waived platform in #342. They will be unwaived in a follow-up PR with the matching stages.

NVBug	Entries	Why not verified
6435097	1 (B200)	test lives in `l0_dgx_h200`; no H200 stage in the run
6695515	1 (B200)	no result
6812288	1 (GB200)	multi-node; no result
6432818	1 (GB300)	passed only on DGX_B200, not on GB300
6428092	1	`test_llm_multi_gpu_pytorch.py -m "gpu2"`; no result
6432826	1	`test_llm_get_stats_pp2`; no result

CI failures (unrelated to this PR)

How can I build it without docker container? #342: GB300-4_GPUs-PyTorch-Post-Merge-3 test_unittests_v2[unittest/_torch/attention] timed out.
Get error "newSize <= getCapacity()" when call endpoint #375: DGX_B200-4_GPUs-PyTorch-Post-Merge-2 TestQwen3_30B_A3B_Instruct_2507::test_skip_softmax_attention_4gpus[target_sparsity_0.9-fp8kv=False] failed with Executor worker died during initialization: MPI worker exited unexpectedly; fail-fast then aborted the other DGX_B200 / DGX_B300 4-GPU post-merge stages, after the unwaived tests in them had already passed. This Qwen3 case had no failures in release/1.3 post-merge on DGX_B200 over the previous 30 days (29 passed, 5 skipped).

Test Coverage

Test-list change only. Rerun with --extra-stage covering the 5 unwaived entries (DGX_B200 / DGX_B300 4-GPU post-merge, DGX_H100 GptOss, PackageSanityCheck).

PR Checklist

Please check this after reviewing the above items as appropriate for this PR.

🤖 Generated with Claude Code

Dev Engineer Review

The change removes five closed-bug waiver entries from tests/integration/test_lists/waives.txt. No runtime code or public APIs change. The successful CI run supports removing the waivers.

QA Engineer Review

The change removes five integration-test waivers for NVBugs 6481323, 6428057, 6762288 (Triton), 6700265, and 6075431. No test functions changed. Coverage verdict: sufficient, based on the reported successful release CI run.

Per-File QA Perspective

tests/integration/test_lists/waives.txt: Verify the five affected tests on their target release platforms. Confirm that the remaining waivers still match their intended tests and platforms.

QiJune · 2026-10-19T08:36:53Z

/bot run --extra-stage "DGX_B200-4_GPUs-PyTorch-Post-Merge-, DGX_B300-4_GPUs-PyTorch-Post-Merge-, GB300-4_GPUs-PyTorch-Post-Merge-*, DGX_H100-2_GPUs-PyTorch-GptOss-1, PackageSanityCheck"

coderabbitai · 2026-10-19T08:39:54Z

No actionable comments were generated in the recent review. 🎉

ℹ️ Recent review info

⚙️ Run configuration

Configuration used: Repository: NVIDIA/TensorRT-LLM/.coderabbit.yaml
Review profile: CHILL
Plan: Enterprise
Run ID: 3eb154e5-7fb3-428f-b38a-103cf1d23760

📥 Commits

Reviewing files that changed from the base of the PR and between 435a63f and 189d5e8.

📒 Files selected for processing: 1

tests/integration/test_lists/waives.txt

💤 Files with no reviewable changes: 1

tests/integration/test_lists/waives.txt

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

Walkthrough

The integration waiver list removes five skip entries for DeepSeek, GPTOSS Eagle3, DGX_H100 DeepSeekV3Lite, and Eagle3 speculative-decoding tests.

Changes

Integration test waivers

Layer / File(s)	Summary
Update integration test waivers `tests/integration/test_lists/waives.txt`	Removes five skip entries for the listed model configurations and test environments.

Priority: ⬇️ Low

Estimated code review effort: 1 (Trivial) | ~4 minutes

Change: Other

Suggested reviewers: mikeiovine

Merge Risk: ⚪ Minimal · up to 189d5

No actionable merge-blocking issue is established. The waiver update is mergeable after normal checks.

Pre-merge checks: 5

Check name	Status	Explanation
Title check		The title clearly identifies the test-list cleanup and specifies that closed-bug waive entries are being removed for release/1.3.
Description check		The description explains the purpose, lists all five removed waive entries and verification evidence, documents retained waives and unrelated CI failures, and identifies test coverage. The required se…
Docstring Coverage		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		Check skipped because no linked issues were found for this pull request.
Out of Scope Changes check		Check skipped because no linked issues were found for this pull request.

Full details: Description check

Explanation

The description explains the purpose, lists all five removed waive entries and verification evidence, documents retained waives and unrelated CI failures, and identifies test coverage. The required sections are present and the checklist is addressed.

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.

✨ Finishing Touches

🧪 Generate unit tests (beta)

Create a new PR

Autofix · Keep fixing CodeRabbit findings and required CI, and resolving merge conflicts

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

tensorrt-cicd · 2026-10-19T08:44:27Z

PR_Github #78868 [ run ] triggered by Bot. Commit: cdd87c8 Link to invocation

tensorrt-cicd · 2026-10-19T17:34:56Z

PR_Github #78868 [ run ] completed with state SUCCESS. Commit: cdd87c8
/LLM/release-1.3/L0_MergeRequest_PR pipeline #342 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

QiJune · 2026-10-20T00:34:21Z

/bot run --extra-stage "DGX_B200-4_GPUs-PyTorch-Post-Merge-, DGX_B300-4_GPUs-PyTorch-Post-Merge-, DGX_H100-2_GPUs-PyTorch-GptOss-1, PackageSanityCheck"

tensorrt-cicd · 2026-10-20T00:42:18Z

PR_Github #79063 [ run ] triggered by Bot. Commit: 0618246 Link to invocation

tensorrt-cicd · 2026-10-20T06:28:48Z

PR_Github #79063 [ run ] completed with state FAILURE. Commit: 0618246
/LLM/release-1.3/L0_MergeRequest_PR pipeline #375 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

QiJune · 2026-10-20T23:42:13Z

/bot run --extra-stage "DGX_B200-4_GPUs-PyTorch-Post-Merge-, DGX_B300-4_GPUs-PyTorch-Post-Merge-, DGX_H100-2_GPUs-PyTorch-GptOss-1, PackageSanityCheck"

tensorrt-cicd · 2026-10-20T23:54:20Z

PR_Github #79416 [ run ] triggered by Bot. Commit: 435a63f Link to invocation

Remove the remaining 6 waive entries on release/1.3 whose NVBugs are closed. PRs NVIDIA#20585, NVIDIA#20590, NVIDIA#20592 and NVIDIA#20632 handle the other closed-bug entries. - 6428092, 6432826: pipeline-parallel TorchSampler IndexError, fixed by NVIDIA#16163, which is on release/1.3. - 6435097, 6432818: closed as fixed by NVIDIA#16163, but their B200 / GB300 waivers were added later from runs on those GPUs; only QA lists run these tests there. - 6695515, 6812288: closed as not reproducible. The 6428092 entry waives the whole `-m "gpu2"` selection of test_llm_multi_gpu_pytorch.py, which also holds test_llm_rpc_tp2, test_llm_rpc_streaming_tp2 and test_llm_return_logprobs_streaming_tp2. main waives those three under 6859934 because it does not have the fix in NVIDIA#19847 yet; release/1.3 has it. Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>

tensorrt-cicd · 2026-10-21T01:49:49Z

PR_Github #79416 [ run ] completed with state SUCCESS. Commit: 435a63f
/LLM/release-1.3/L0_MergeRequest_PR pipeline #427 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

QiJune · 2026-10-21T03:13:20Z

/bot run

tensorrt-cicd · 2026-10-21T03:21:17Z

PR_Github #79496 [ run ] triggered by Bot. Commit: 435a63f Link to invocation

tensorrt-cicd · 2026-10-21T13:33:18Z

PR_Github #79496 [ run ] completed with state SUCCESS. Commit: 435a63f
/LLM/release-1.3/L0_MergeRequest_PR pipeline #441 completed with status: 'SUCCESS'

CI Report

Link to invocation

Remove 5 waive entries on release/1.3 whose NVBugs are closed and whose tests passed on the waived platform in pre-merge CI (L0_MergeRequest_PR 6481323, 6428057, 6762288 (triton variant), 6700265, 6075431. Handled by other release/1.3 PRs: 6893730 (NVIDIA#20592), the 6762288 cutlass variant and 6672360 (NVIDIA#20632), 6278337 (NVIDIA#20585). Waives for the other closed bugs (6435097, 6695515, 6812288, 6432818, 6428092, 6432826) are kept: their tests did not run on the waived platform in that CI run, so the unwaive is not verified yet. Co-Authored-By: Claude Opus 5.5 <noreply@anthropic.com> Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>

QiJune · 2026-10-22T00:23:42Z

/bot skip --comment "Rebased onto release/1.3 to resolve a waives.txt conflict only; the diff is still the same 5 waive deletions. L0_MergeRequest_PR #441 passed on the previous commit 435a63f."

tensorrt-cicd · 2026-10-22T00:31:22Z

PR_Github #79723 [ skip ] triggered by Bot. Commit: 189d5e8 Link to invocation

tensorrt-cicd · 2026-10-22T00:44:27Z

PR_Github #79723 [ skip ] completed with state SUCCESS. Commit: 189d5e8
Skipping testing for commit 189d5e8

Link to invocation

github-actions Bot assigned QiJune Oct 19, 2026

QiJune added the ci: full pre-merge approved label Oct 19, 2026

QiJune force-pushed the release-1.3-unwaive-closed-bugs branch from cdd87c8 to 0618246 Compare October 20, 2026 00:34

QiJune force-pushed the release-1.3-unwaive-closed-bugs branch from ec16637 to 435a63f Compare October 20, 2026 23:39

QiJune mentioned this pull request Oct 20, 2026

[None][test] Unwaive test_llm_multi_gpu_pytorch gpu2 tests for release/1.3 #20716

Merged

1 task done

mikeiovine approved these changes Oct 21, 2026

View reviewed changes

QiJune enabled auto-merge (squash) October 21, 2026 03:17

QiJune force-pushed the release-1.3-unwaive-closed-bugs branch from 435a63f to 189d5e8 Compare October 22, 2026 00:22

QiJune disabled auto-merge October 22, 2026 00:47

QiJune merged commit c5f24ec into NVIDIA:release/1.3 Oct 22, 2026

Conversation

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

Uh oh!

Description

Kept waived (closed bugs, not yet verified)

CI failures (unrelated to this PR)

Test Coverage

PR Checklist

Dev Engineer Review

QA Engineer Review

Per-File QA Perspective

Uh oh!

QiJune commented Oct 19, 2026

Uh oh!

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

Uh oh!

Walkthrough

Changes

Uh oh!

tensorrt-cicd commented Oct 19, 2026

Uh oh!

tensorrt-cicd commented Oct 19, 2026

Uh oh!

QiJune commented Oct 20, 2026

Uh oh!

tensorrt-cicd commented Oct 20, 2026

Uh oh!

tensorrt-cicd commented Oct 20, 2026

Uh oh!

QiJune commented Oct 20, 2026

Uh oh!

tensorrt-cicd commented Oct 20, 2026

Uh oh!

tensorrt-cicd commented Oct 21, 2026

Uh oh!

QiJune commented Oct 21, 2026

Uh oh!

tensorrt-cicd commented Oct 21, 2026

Uh oh!

tensorrt-cicd commented Oct 21, 2026

Uh oh!

QiJune commented Oct 22, 2026

Uh oh!

tensorrt-cicd commented Oct 22, 2026

Uh oh!

tensorrt-cicd commented Oct 22, 2026

Uh oh!

Reviewers

Uh oh!

Assignees

Labels

Projects

Milestone

Development

Uh oh!

3 participants

QiJune commented Oct 19, 2026 •

edited by coderabbitai Bot

Loading

coderabbitai Bot commented Oct 19, 2026 •

edited

Loading