[TRTQA-3273][test] Waive 4 failed cases for main in QA CI

[trtllm-agent]

Auto-generated Waive PR

Created by: TensorRT LLM CI Report (requested by qa@nvidia.com)
Target branch: main
Bug(s): 6473161

Waive entries added

full:L40S/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8[fp8kv=False-attn_backend=FLASHINFER-torch_compile=True] SKIP (https://nvbugs/6473161)
full:L40S/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=True-attn_backend=FLASHINFER-torch_compile=True] SKIP (https://nvbugs/6473161)
full:L40S/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8[fp8kv=True-attn_backend=FLASHINFER-torch_compile=True] SKIP (https://nvbugs/6473161)
full:L40S/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=False-attn_backend=FLASHINFER-torch_compile=True] SKIP (https://nvbugs/6473161)

---

This PR was auto-generated by TensorRT LLM CI Report. Please review the waive entries before merging.

Summary by CodeRabbit

• Tests
  • Added skip coverage for FP8 and multi-GPU FP8 Llama 3.1 accuracy tests.
  • Documented the associated known issue reference for these test cases.

[xinhe-nv]

/bot skip --comment "skip test on L40S only"

[coderabbitai]

📝 Walkthrough

Walkthrough

Adds four L40S waiver entries for Llama 3.1 8B instruct FP8 accuracy tests, covering single- and four-GPU variants with both fp8kv settings.

Changes

L40S FP8 test waivers

Layer / File(s)	Summary
Add FP8 skip entries tests/integration/test_lists/waives.txt	Skips test_fp8 and test_fp8_4gpus parameterizations using FLASHINFER, TorchCompile, and both fp8kv values, referencing nvbugs/6473161.

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

Possibly related PRs

• NVIDIA/TensorRT-LLM#16560: Adds related FP8 waiver entries with fp8kv variations.
• NVIDIA/TensorRT-LLM#16573: Modifies the same waiver list with FP8-related skips.
• NVIDIA/TensorRT-LLM#16534: Updates waiver entries for related LLM API PyTorch tests.

🚥 Pre-merge checks | ✅ 4 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR includes the bug and waive list, but it omits the required Description, Test Coverage, and PR Checklist sections from the template.	Add the missing template sections with a short issue/solution summary, the relevant tests or note why none apply, and a completed checklist.

✅ Passed checks (4 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and accurately describes the main change: waiving four QA CI failures on main.
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.}

[coderabbitai]

Actionable comments posted: 1

🧹 Nitpick comments (1)

tests/integration/test_lists/waives.txt (1)

288-291: 📐 Maintainability & Code Quality | 🔵 Trivial

Waiver coverage is sufficient for the stated scope.

The four entries cover both test variants, both fp8kv values, and the single-/four-GPU cases. No additional test-list file is needed; remove these waivers in a follow-up once NVBug 6473161 is fixed.

As per path instructions, coverage for this waiver change is sufficient, with restoration tracked outside this PR.

🤖 Prompt for AI Agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

In `@tests/integration/test_lists/waives.txt` around lines 288 - 291, Keep the
four NVBug 6473161 waivers in the existing test list unchanged, covering both
fp8 test variants, both fp8kv values, and single-/four-GPU cases. Do not add
another test-list file; restoration should occur separately after the underlying
issue is fixed.

Source: Path instructions

🤖 Prompt for all review comments with AI agents

Verify each finding against current code. Fix only still-valid issues, skip the
rest with a brief reason, keep changes minimal, and validate.

Inline comments:
In `@tests/integration/test_lists/waives.txt`:
- Around line 288-291: Add the repository-standard NVIDIA copyright header with
the current year at the beginning of the modified waiver list in waives.txt,
before the existing waiver entries; leave the entries unchanged.

---

Nitpick comments:
In `@tests/integration/test_lists/waives.txt`:
- Around line 288-291: Keep the four NVBug 6473161 waivers in the existing test
list unchanged, covering both fp8 test variants, both fp8kv values, and
single-/four-GPU cases. Do not add another test-list file; restoration should
occur separately after the underlying issue is fixed.

🪄 Autofix (Beta)

Fix all unresolved CodeRabbit comments on this PR:

• Push a commit to this branch (recommended)
• Create a new PR with the fixes

---

ℹ️ Review info

⚙️ Run configuration

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Enterprise

Run ID: 6b185a89-52c7-4c72-bd95-ac4c8ff8d7e8

📥 Commits

Reviewing files that changed from the base of the PR and between fa54a19 and d6789b0.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

[tensorrt-cicd]

PR_Github #60223 [ skip ] triggered by Bot. Commit: d6789b0 Link to invocation

[tensorrt-cicd]

PR_Github #60223 [ skip ] completed with state SUCCESS. Commit: d6789b0
Skipping testing for commit d6789b0

Link to invocation