[None][test] Waive 2 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): 6546909, 6580087

Waive entries added

full:H20/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8_4gpus[tp4-fp8kv=False-attn_backend=TRTLLM-torch_compile=True] SKIP (https://nvbugs/6580087)
full:H20/accuracy/test_llm_api_pytorch.py::TestLlama3_1_8BInstruct::test_fp8[fp8kv=False-attn_backend=TRTLLM-torch_compile=True] SKIP (https://nvbugs/6546909)

---

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

Dev Engineer Review

• Added two SKIP waivers in tests/integration/test_lists/waives.txt.
• Covers the H20 Llama 3.1 8B Instruct FP8 tests for one GPU and four GPUs.
• Entries include torch_compile=True, fp8kv=False, and bug references 6546909 and 6580087.
• No public API or production code changes.
• No duplicate or unintended scope changes identified.

QA Engineer Review

• Modified tests/integration/test_lists/waives.txt.
• Added two waiver entries.
• No test-db/ or qa/ files were modified.
• Verdict: needs follow-up because CBTS coverage data is unavailable.

[jieli-matrix]

/bot skip --comment "Only waive tests"

[coderabbitai]

Walkthrough

Added two H20 test waivers for Llama 3.1 8B FP8 TRTLLM configurations. The entries cover single-GPU and four-GPU tests with torch_compile=True and fp8kv=False.

Changes

H20 Llama FP8 waivers

Layer / File(s)	Summary
Add H20 waiver entries tests/integration/test_lists/waives.txt	Adds waivers for single-GPU and four-GPU Llama 3.1 8B FP8 TRTLLM tests. Each entry references an NVBUG issue.

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

Possibly related PRs

• NVIDIA/TensorRT-LLM#17467: Adds related CI waivers for Llama 3.1 FP8 tests.
• NVIDIA/TensorRT-LLM#17384: Adds two waiver entries to the same waiver list.
• NVIDIA/TensorRT-LLM#17458: Adds CI waiver entries to the same waiver list.

Suggested reviewers: qijune, schetlur-nv

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly identifies the test waiver change and the target branch.
Description check	✅ Passed	The description explains the waiver purpose, lists both affected test cases, and links the associated bugs.
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

🤖 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 280-281: Add the repository-standard NVIDIA copyright header at
the beginning of tests/integration/test_lists/waives.txt, using 2026 as the
latest meaningful year, while preserving the existing waiver entries unchanged.

🪄 Autofix

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: a61d63ab-61ec-47f7-bc7a-3fc3554a43fc

📥 Commits

Reviewing files that changed from the base of the PR and between 67dd1b7 and acdfff0.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

[tensorrt-cicd]

PR_Github #65009 [ skip ] triggered by Bot. Commit: acdfff0 Link to invocation

[tensorrt-cicd]

PR_Github #65009 [ skip ] completed with state SUCCESS. Commit: acdfff0
Skipping testing for commit acdfff0

Link to invocation

[brnguyen2]

Test IDs match the parametrize decorators, no duplicate entries, placement in the H20 block is consistent. LGTM.