[https://nvbugs/6581048][test] Unwaive TorchSampler beam-search e2e case

[zhaoyangwang-nvidia]

Description

test_beam_search_e2e[multi_process-TorchSampler-no_cuda_graph_and_overlap-stop_token_ids0-1-1-True-True-True]
was waived under https://nvbugs/6581048. The failure is no longer reproducible on
current main, so this PR removes the waiver and puts the case back into pre-merge.

Verification was done on an A30 -- the same GPU l0_a30.yml schedules this file on --
with main built locally at bac9a5cc58 (1.3.0rc25, -a 80-real):

Scope	Result
Target case, 20 consecutive runs (fresh pytest process each time)	20 passed / 0 failed
Target case, initial single run	1 passed
Whole tests/unittest/_torch/sampler/test_beam_search.py	345 passed, 244 skipped, 0 failed

Each of the 20 iterations ran in its own pytest process, so the LLM object and the
MPI worker pool were rebuilt every time rather than reused.

Because this parametrization runs with multi_process, MPI worker spawning was checked
before the test (MPI_Init plus an MPIPoolExecutor spawn round-trip) to make sure a
degraded single-process fallback could not turn into a false pass.

The 244 skips are the test's own pytest.skip guards for TRTLLMSampler combinations
(return_log_probs with multiple prompts, repeated log_probs calls, stop_token_ids),
not environment-related.

Test Coverage

• unittest/_torch/sampler/test_beam_search.py::test_beam_search_e2e[multi_process-TorchSampler-no_cuda_graph_and_overlap-stop_token_ids0-1-1-True-True-True]

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.

Dev Engineer Review

• Removed the obsolete TorchSampler beam-search waiver from tests/integration/test_lists/waives.txt.
• The waiver entry format and test scope remain consistent.
• No code, API, or configuration changes were made.
• Current-main verification reported 21 consecutive passes for the target case and 345 full-file passes.

QA Engineer Review

• No test-db/ or qa/ files were modified.
• Removed one waiver entry for the multi-process TorchSampler beam-search end-to-end test.
• The test returned to pre-merge testing.
• Verdict: needs follow-up because CBTS coverage data is unavailable.

[coderabbitai]

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: 06b291e8-9f6a-43a0-9850-54ddbc72da7b

📥 Commits

Reviewing files that changed from the base of the PR and between be27329 and 78377df.

📒 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: Your plan includes up to 12 reviews per rolling hour; 11 remain after this review.

---

Walkthrough

The change removes a skip entry for the multi-process TorchSampler beam-search integration test without CUDA graphs or overlap.

Changes

TorchSampler test execution

Layer / File(s)	Summary
Remove TorchSampler test waiver tests/integration/test_lists/waives.txt	The skip configuration was removed for the beam-search test with stop token IDs.

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

Merge Risk: ⚪ Minimal · up to 78377

This change restores one previously waived beam-search test to pre-merge coverage after repeated successful runs; no actionable merge-blocking risk remains beyond normal checks and review.

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Description check	✅ Passed	The description explains the removed waiver, the reason, the verification environment, test results, and relevant test coverage.
Title check	✅ Passed	The title clearly identifies the NVBugs issue, test change, and removal of the TorchSampler beam-search end-to-end waiver.
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.

✨ Finishing Touches

🧪 Generate unit tests (beta)

• Create PR with unit tests

---

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

[zhaoyangwang-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #66766 [ ] completed with state FAILURE. Commit: ``

Link to invocation

[zhaoyangwang-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #66774 [ run ] triggered by Bot. Commit: 78377df Link to invocation

[tensorrt-cicd]

PR_Github #66774 [ run ] completed with state FAILURE. Commit: 78377df
/LLM/main/L0_MergeRequest_PR pipeline #54369 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

[zhaoyangwang-nvidia]

/bot run

[tensorrt-cicd]

PR_Github #66904 [ run ] triggered by Bot. Commit: 78377df Link to invocation

[tensorrt-cicd]

PR_Github #66904 [ run ] completed with state SUCCESS. Commit: 78377df
/LLM/main/L0_MergeRequest_PR pipeline #54460 completed with status: 'SUCCESS'

CI Report

Link to invocation