[None][infra] Waive 23 failed cases for main in post-merge 2826

[trtllm-agent]

Auto-generated Waive PR

Created by: TensorRT LLM CI Report (requested by @mzweilz)
Target branch: main
Bug(s): 6428002, 6428005, 6428008, 6428014, 6428057, 6428063, 6428069, 6428087, 6428089, 6428091, 6428092, 6428094, 6428096, 6428101, 6428113, 6428124, 6428144

Waive entries added

accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_ctx_pp_gen_tp_asymmetric[GSM8K-gen_tp=1-ctx_pp=4] SKIP (https://nvbugs/6428069)
accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_ctx_pp_gen_tp_asymmetric[GSM8K-gen_tp=2-ctx_pp=4] SKIP (https://nvbugs/6428069)
accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_ctx_pp_gen_tp_asymmetric[MMLU-gen_tp=1-ctx_pp=4] SKIP (https://nvbugs/6428069)
accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_ctx_pp_gen_tp_asymmetric[MMLU-gen_tp=2-ctx_pp=4] SKIP (https://nvbugs/6428069)
accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_tp_pp_symmetric[GSM8K-tp2pp2] SKIP (https://nvbugs/6428069)
accuracy/test_disaggregated_serving.py::TestLlama3_1_8BInstruct::test_tp_pp_symmetric[MMLU-tp2pp2] SKIP (https://nvbugs/6428069)
accuracy/test_llm_api_autodeploy.py::TestNemotronSuperV3::test_mtp[nvfp4_ws4_80gb-trtllm] SKIP (https://nvbugs/6428014)
accuracy/test_llm_api_pytorch.py::TestDeepSeekR1::test_fp8_blockscale[throughput_mtp] SKIP (https://nvbugs/6428101)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[pp4-mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=True] SKIP (https://nvbugs/6428057)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[tp2pp2-mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=False] SKIP (https://nvbugs/6428094)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_bfloat16_4gpus[tp2pp2-mtp_nextn=0-attention_dp=False-cuda_graph=False-overlap_scheduler=False-torch_compile=True] SKIP (https://nvbugs/6428096)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTEDSL-mtp_nextn=0-tp2pp2-fp8kv=False-attention_dp=False-cuda_graph=False-overlap_scheduler=False-low_precision_combine=False-torch_compile=True] SKIP (https://nvbugs/6428087)
accuracy/test_llm_api_pytorch.py::TestDeepSeekV3Lite::test_nvfp4_4gpus[moe_backend=CUTLASS-mtp_nextn=0-pp4-fp8kv=True-attention_dp=True-cuda_graph=True-overlap_scheduler=True-low_precision_combine=False-torch_compile=True] SKIP (https://nvbugs/6428063)
accuracy/test_llm_api_pytorch.py::TestLlama3_3_70BInstruct::test_fp4_tp2pp2[torch_compile=False-enable_gemm_allreduce_fusion=False] SKIP (https://nvbugs/6428089)
accuracy/test_llm_api_pytorch.py::TestQwen3_5_35B_A3B::test_bf16[tp1-TRTLLM] SKIP (https://nvbugs/6428008)
disaggregated/test_disaggregated.py::test_disaggregated_ctxpp4_genpp4[TinyLlama-1.1B-Chat-v1.0] SKIP (https://nvbugs/6428069)
disaggregated/test_disaggregated.py::test_disaggregated_ctxtp2pp2_gentp2pp2[TinyLlama-1.1B-Chat-v1.0] SKIP (https://nvbugs/6428069)
kv_cache/test_kv_cache_v2_scheduler.py::TestKVCacheV2Llama::test_chunked_prefill SKIP (https://nvbugs/6428002)
perf/test_perf_sanity.py::test_e2e[disagg_upload-gen_only-gb300_deepseek-r1-fp4_128k8k_con256_ctx1_pp4_gen1_dep8_eplb0_mtp1_ccb-NIXL] SKIP (https://nvbugs/6428144)
unittest/_torch/modules/test_moe_routing.py SKIP (https://nvbugs/6428113)
unittest/_torch/modules/tests_lora_modules/test_nemotron_h_lora_sanity.py::TestNemotronHLoRA::test_lora_pp2_sanity SKIP (https://nvbugs/6428124)
unittest/llmapi/test_additional_model_outputs.py -m "gpu2" SKIP (https://nvbugs/6428091)
unittest/llmapi/test_llm_multi_gpu_pytorch.py -m "gpu2" SKIP (https://nvbugs/6428092)

Already waived (skipped)

• accuracy/test_llm_api_pytorch_multimodal.py::TestNanoV3Omni::test_auto_dtype[nvfp4]

---

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

Summary by CodeRabbit

• Tests
  • Updated test exemptions for several distributed, multi-GPU, and LLM API scenarios.
  • Added coverage exceptions for additional model and performance cases that are currently unstable or unsupported.
  • Refined existing waivers to better match current test behavior and reduce noisy failures.

[mzweilz]

/bot skip --comment "Waive only"

[coderabbitai]

📝 Walkthrough

Walkthrough

This PR updates the CI waive list (tests/integration/test_lists/waives.txt), adding new SKIP entries and removing/replacing prior ones for various failing or unsupported distributed test configurations across accuracy, disaggregated serving, kv_cache, perf sanity, and unittest suites.

Changes

Test waiver list updates

Layer / File(s)	Summary
Accuracy and disaggregated serving waives tests/integration/test_lists/waives.txt	Updated SKIP entries for Llama3.1 8B Instruct disaggregated pp/tp scenarios, NemotronSuperV3/DeepSeekR1/Qwen3.5 mtp and throughput tests, DeepSeekV3Lite bf16/nvfp4 multi-GPU combinations, Llama3_3_70B fp4 tp2pp2, Qwen3_5_35B bf16, and TinyLlama disaggregated ctx/gen configurations.
KV cache, perf sanity, and unittest waives tests/integration/test_lists/waives.txt	Added SKIP entries for chunked prefill KV cache scheduler test, a disaggregated upload perf sanity scenario, MoE routing and NemotronH LoRA sanity unittests, and llmapi multi-GPU/additional-model-output tests.

Estimated code review effort: 2 (Simple) | ~10 minutes

Possibly related PRs

• NVIDIA/TensorRT-LLM#15585: Both PRs modify waives.txt to skip the same unittest/llmapi related tests.
• NVIDIA/TensorRT-LLM#15960: Both PRs update waives.txt with overlapping SKIP entries for the same disaggregated serving and autodeploy test cases.
• NVIDIA/TensorRT-LLM#16093: Both PRs exclusively update waives.txt to add/remove CI waive entries for newly failing model parameterizations.

Suggested reviewers: mzweilz

🚥 Pre-merge checks | ✅ 5

✅ Passed checks (5 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title is concise and accurately summarizes the main change: waiving failed cases for the main branch.
Description check	✅ Passed	The description clearly states the purpose, target branch, bug links, and waived entries, though the template sections are not fully filled.
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: 2

🤖 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 458-460: The current gpu2 waives are too broad because they skip
every pytest.mark.gpu2 test in the affected modules. Update the entries in the
waives list to target only the specific flaky nodeids in
unittest/llmapi/test_additional_model_outputs.py and
unittest/llmapi/test_llm_multi_gpu_pytorch.py, rather than module-wide -m "gpu2"
selectors. Keep the existing isolated test waiver for
test_additional_model_outputs_integration_pp2 unchanged if that is the only
failing case.
- Around line 425-426: The waive is too broad because it skips the entire
unittest/_torch/modules/test_moe_routing.py module instead of only the failing
tests. Update the entries in waives.txt to target the specific pytest node IDs
that are actually failing, using the existing test_moe_routing module name as
the locator, so unaffected cases like test_default_moe_routing and
test_load_balanced_moe_routing keep running.

🪄 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: 46daca07-20cf-4369-86bc-c2289b6f51e6

📥 Commits

Reviewing files that changed from the base of the PR and between 3892f20 and 63f7d50.

📒 Files selected for processing (1)

• tests/integration/test_lists/waives.txt

[tensorrt-cicd]

PR_Github #58187 [ skip ] triggered by Bot. Commit: 63f7d50 Link to invocation

[tensorrt-cicd]

PR_Github #58187 [ skip ] completed with state SUCCESS. Commit: 63f7d50
Skipping testing for commit 63f7d50

Link to invocation

[mzweilz]

/bot skip --comment "Waive only"

[tensorrt-cicd]

PR_Github #58205 [ skip ] triggered by Bot. Commit: 6c0af1d Link to invocation

[tensorrt-cicd]

PR_Github #58205 [ skip ] completed with state SUCCESS. Commit: 6c0af1d
Skipping testing for commit 6c0af1d

Link to invocation