[None][infra] Waive 13 failed cases for main in post-merge 2876

[trtllm-agent]

Auto-generated Waive PR

Created by: TensorRT LLM CI Report (requested by @mzweilz)
Target branch: main
Bug(s): 6550099, 6550126, 6550127, 6550130, 6550132, 6550133

Waive entries added

accuracy/test_llm_api_pytorch.py::TestNemotronV3Super::test_fp8_4gpus[attention_dp_off-python_mamba_cache] SKIP (https://nvbugs/6550132)
accuracy/test_llm_api_pytorch_multimodal.py::TestGemma4_26B_A4B::test_nvfp4 SKIP (https://nvbugs/6550127)
cache_transceiver_precheck[ctx_0] SKIP (https://nvbugs/6550130)
cache_transceiver_precheck[gen_0] SKIP (https://nvbugs/6550130)
examples/visual_gen/test_visual_gen.py::test_qwenimage_cuda_graph_lpips_against_golden SKIP (https://nvbugs/6550126)
examples/visual_gen/test_visual_gen.py::test_qwenimage_lpips_against_golden SKIP (https://nvbugs/6550126)
perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-v32-fp4_32k4k_con2048_ctx1_dep4_gen1_dep32_eplb288_mtp1_ccb-NIXL] SKIP (https://nvbugs/6550133)
perf/test_perf_sanity.py::test_e2e[disagg_upload-e2e-gb200_deepseek-v32-fp4_8k1k_con4096_ctx1_dep4_gen1_dep32_eplb256_mtp0_ccb-NIXL] SKIP (https://nvbugs/6550133)
perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-r1-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6550133)
perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_deepseek-v32-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb256_mtp3_ccb-NIXL] SKIP (https://nvbugs/6550133)
perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_kimi-k25-thinking-fp4_8k1k_con1024_ctx1_dep4_gen1_dep32_eplb416_mtp3_ccb-NIXL] SKIP (https://nvbugs/6550133)
perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb200_kimi-k25-thinking-fp4_8k1k_con4096_ctx1_dep4_gen1_dep16_eplb0_mtp0_ccb-NIXL] SKIP (https://nvbugs/6550133)
unittest/_torch/speculative/hw_agnostic/test_advanced_sampling_mode.py::test_no_topk_matches_full[0.9] SKIP (https://nvbugs/6550099)

Already waived (skipped)

• accuracy/test_llm_api_pytorch.py::TestNemotronV3Super::test_fp8_4gpus[attention_dp_on-python_mamba_cache]

---

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

Summary

• Added 13 skip waivers to tests/integration/test_lists/waives.txt.
• Covered failing or unavailable tests in PyTorch accuracy, multimodal, cache transceiver, visual generation, performance sanity, and speculative sampling suites.
• Added NVBUG references for bugs 6550099, 6550126, 6550127, 6550130, 6550132, and 6550133.
• No public API changes.

Dev Engineer Review

• The change is limited to test waiver configuration.
• The change does not affect production code, performance, or error handling.
• Review must confirm that each test path, waiver format, bug reference, and scope matches the related failures.
• Review must confirm that no duplicate waiver exists. One related test was already waived.

QA Engineer Review

• Modified file: tests/integration/test_lists/waives.txt.
• Added 13 skip entries.
• No test-db or qa files were modified.
• CBTS coverage data is unavailable.
• Verdict: needs follow-up.

[mzweilz]

/bot skip --comment "waive only"

[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: 1f33c3e1-d2ef-4849-9865-ab0f6166fdf1

📥 Commits

Reviewing files that changed from the base of the PR and between 0d77dbd and 7a06163.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

---

Walkthrough

Added 13 skip entries to the integration waiver list for newly tracked failing or unavailable tests across multiple suites.

Changes

Integration test waivers

Layer / File(s)	Summary
Add integration test waivers tests/integration/test_lists/waives.txt	Added NVBUG-linked skips for Nemotron, Gemma, cache-transceiver, QwenImage, GB200 performance, and advanced sampling tests.

Estimated code review effort: 1 (Trivial) | ~5 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#15195: Adds integration test skip waivers in the same waiver list.
• NVIDIA/TensorRT-LLM#15585: Adds or adjusts test skip waivers in the same waiver list.

Suggested reviewers: brnguyen2, qijune

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly identifies the infrastructure change and the 13 waived post-merge cases on main.
Description check	✅ Passed	The description explains the waiver purpose, lists all affected tests and bug IDs, and identifies the target branch and requester.
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.}

[tensorrt-cicd]

PR_Github #63341 [ skip ] triggered by Bot. Commit: 7a06163 Link to invocation

[tensorrt-cicd]

PR_Github #63341 [ skip ] completed with state SUCCESS. Commit: 7a06163
Skipping testing for commit 7a06163

Link to invocation

[BowenFu]

Heads-up: the two entries this added at waives.txt:106-107 are breaking Check Test List for every PR on main.

cache_transceiver_precheck[ctx_0] SKIP (https://nvbugs/6550130)
cache_transceiver_precheck[gen_0] SKIP (https://nvbugs/6550130)

Non-existent test name in l0 or qa list found in waives.txt: cache_transceiver_precheck[ctx_0]
Non-existent test name in l0 or qa list found in waives.txt: cache_transceiver_precheck[gen_0]

These are synthetic JUnit case names emitted by jenkins/scripts/perf/disaggregated/slurm_ct_precheck_gate.sh for the perf-sanity disagg gate — they are not pytest IDs in any test-db/l0_*.yml or qa/* list, so the validator rejects them. The real pytest targets are tests/unittest/others/test_cache_transceiver_precheck_{config,run}.py and tests/unittest/disaggregated/test_cache_transceiver_precheck_e2e.py.

Confirmed red on #16432 (build 51356), #17121 (51365), #17175 (51337) and #17179 (51350) — all different authors, all since this merged at 03:15Z. Suggest dropping the two lines (or waiving the real pytest IDs instead) so pre-merge can go green again.