[https://nvbugs/5888410][fix] Enable warmup for Helix CP

[brb-nv]

Description

This MR enables warmup for Helix CP. Without this, there's a significant discrepancy in the Median TPOT reported by benchmarking with cudagraph because of the first iteration including time taken for cudagraph warmup.

Test Coverage

Existing tests should suffice.

$ pytest tests/integration/defs/accuracy/test_disaggregated_serving.py::TestDeepSeekV3Lite::test_auto_dtype_with_helix -s -v

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)

• 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.

GitHub Bot Help

/bot [-h] ['run', 'kill', 'skip', 'reuse-pipeline'] ...

Provide a user friendly way for developers to interact with a Jenkins server.

Run /bot [-h|--help] to print this help message.

See details below for each supported subcommand.

Details

run [--reuse-test (optional)pipeline-id --disable-fail-fast --skip-test --stage-list "A10-PyTorch-1, xxx" --gpu-type "A30, H100_PCIe" --test-backend "pytorch, cpp" --add-multi-gpu-test --only-multi-gpu-test --disable-multi-gpu-test --post-merge --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" --detailed-log --debug(experimental)]

Launch build/test pipelines. All previously running jobs will be killed.

--reuse-test (optional)pipeline-id (OPTIONAL) : Allow the new pipeline to reuse build artifacts and skip successful test stages from a specified pipeline or the last pipeline if no pipeline-id is indicated. If the Git commit ID has changed, this option will be always ignored. The DEFAULT behavior of the bot is to reuse build artifacts and successful test results from the last pipeline.

--disable-reuse-test (OPTIONAL) : Explicitly prevent the pipeline from reusing build artifacts and skipping successful test stages from a previous pipeline. Ensure that all builds and tests are run regardless of previous successes.

--disable-fail-fast (OPTIONAL) : Disable fail fast on build/tests/infra failures.

--skip-test (OPTIONAL) : Skip all test stages, but still run build stages, package stages and sanity check stages. Note: Does NOT update GitHub check status.

--stage-list "A10-PyTorch-1, xxx" (OPTIONAL) : Only run the specified test stages. Examples: "A10-PyTorch-1, xxx". Note: Does NOT update GitHub check status.

--gpu-type "A30, H100_PCIe" (OPTIONAL) : Only run the test stages on the specified GPU types. Examples: "A30, H100_PCIe". Note: Does NOT update GitHub check status.

--test-backend "pytorch, cpp" (OPTIONAL) : Skip test stages which don't match the specified backends. Only support [pytorch, cpp, tensorrt, triton]. Examples: "pytorch, cpp" (does not run test stages with tensorrt or triton backend). Note: Does NOT update GitHub pipeline status.

--only-multi-gpu-test (OPTIONAL) : Only run the multi-GPU tests. Note: Does NOT update GitHub check status.

--disable-multi-gpu-test (OPTIONAL) : Disable the multi-GPU tests. Note: Does NOT update GitHub check status.

--add-multi-gpu-test (OPTIONAL) : Force run the multi-GPU tests in addition to running L0 pre-merge pipeline.

--post-merge (OPTIONAL) : Run the L0 post-merge pipeline instead of the ordinary L0 pre-merge pipeline.

--extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx" (OPTIONAL) : Run the ordinary L0 pre-merge pipeline and specified test stages. Examples: --extra-stage "H100_PCIe-TensorRT-Post-Merge-1, xxx".

--detailed-log (OPTIONAL) : Enable flushing out all logs to the Jenkins console. This will significantly increase the log volume and may slow down the job.

--debug (OPTIONAL) : Experimental feature. Enable access to the CI container for debugging purpose. Note: Specify exactly one stage in the stage-list parameter to access the appropriate container environment. Note: Does NOT update GitHub check status.

For guidance on mapping tests to stage names, see docs/source/reference/ci-overview.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip testing for latest commit on pull request. --comment "Reason for skipping build/test" is required. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

reuse-pipeline

reuse-pipeline

Reuse a previous pipeline to validate current commit. This action will also kill all currently running builds associated with the pull request. IMPORTANT NOTE: This is dangerous since lack of user care and validation can cause top of tree to break.

Summary by CodeRabbit

• Bug Fixes
  • Fixed warmup behavior for Helix checkpoint parallel configurations to allow proper initialization of parallel execution paths.
  • Improved dummy request generation for Helix checkpoint parallelism by ensuring proper token count and decoding iteration initialization across all ranks.

[brb-nv]

/bot run --disable-fail-fast

[coderabbitai]

📝 Walkthrough

Walkthrough

These changes enable Helix collective parallelism to use warmup functionality by modifying control flow conditions and removing a previous constraint. Dummy request initialization is updated to ensure proper token and decoding iteration state for Helix CP scenarios.

Changes

Cohort / File(s)	Summary
Warmup and Assertion Handling tensorrt_llm/_torch/pyexecutor/model_engine.py	Modified warmup skip condition from cp_size > 1 to also check cp_type is not CpType.HELIX, allowing Helix type to proceed with warmup. Removed assertion in _prepare_tp_inputs that prevented warmup for helix parallelism.
Dummy Request Initialization tensorrt_llm/_torch/pyexecutor/resource_manager.py	In add_dummy_requests, clamped token_num to minimum of 2 for Helix CP to prevent negative past_seen_token_num. Set req.py_decoding_iter = 1 for both active and inactive helix ranks to ensure decoding iteration state is initialized.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~12 minutes

🚥 Pre-merge checks | ✅ 2 | ❌ 1

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Docstring Coverage	⚠️ Warning	Docstring coverage is 66.67% which is insufficient. The required threshold is 80.00%.	Write docstrings for the functions missing them to satisfy the coverage threshold.

✅ Passed checks (2 passed)

Check name	Status	Explanation
Title check	✅ Passed	The PR title clearly and specifically describes the main change: enabling warmup functionality for Helix CP, which is the primary focus of the changeset.
Description check	✅ Passed	The PR description includes the required sections: a clear explanation of the change and its motivation (why warmup is needed for Helix CP to fix TPOT discrepancy), test coverage details with an example test command, and a completed checklist.

_{✏️ Tip: You can configure your own custom pre-merge checks in the settings.}

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

Tip

Issue Planner is now in beta. Read the docs and try it out! Share your feedback on Discord.

---

Thanks for using CodeRabbit! It's free for OSS, and your support helps us grow. If you like it, consider giving us a shout-out.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

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

[coderabbitai]

Actionable comments posted: 1

🤖 Fix all issues with AI agents

In `@tensorrt_llm/_torch/pyexecutor/resource_manager.py`:
- Around line 609-613: The current clamp that sets token_num = max(token_num, 2)
when self.mapping.has_cp_helix() can push token_num beyond KV-cache capacity
(e.g., tokens_per_block == 1); update the logic around token_num (using
variables token_nums, max_num_draft_tokens, and mapping.has_cp_helix()) to first
check available KV capacity (derive from the same capacity/config values used
elsewhere, e.g., tokens_per_block or KV slots) and only apply the Helix minimum
clamp if there is room; if capacity is insufficient, skip the Helix warmup path
(do not raise token_num) and proceed without the clamp to avoid overfilling the
cache.

[tensorrt-cicd]

PR_Github #35678 [ run ] triggered by Bot. Commit: ee2baf9

[brb-nv]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #35684 [ run ] triggered by Bot. Commit: bce7692

[brb-nv]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #35720 [ run ] triggered by Bot. Commit: 3c1dd3e

[tensorrt-cicd]

PR_Github #35720 [ run ] completed with state SUCCESS. Commit: 3c1dd3e
/LLM/main/L0_MergeRequest_PR pipeline #27589 completed with status: 'FAILURE'

⚠️ 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

[brb-nv]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #35816 [ run ] triggered by Bot. Commit: 3c1dd3e

[tensorrt-cicd]

PR_Github #35816 [ run ] completed with state SUCCESS. Commit: 3c1dd3e
/LLM/main/L0_MergeRequest_PR pipeline #27665 completed with status: 'SUCCESS'