[#10688][fix] AutoDeploy Fix CUDA graph batch sizes exceeding max_batch_size

[MrGeva]

This MR fixes a bug where the CUDA graph batch sizes heuristic generated batch sizes that exceeded max_batch_size, causing buffer overflow errors during model initialization.

When cuda_graph_batch_sizes was not explicitly set, the heuristic in update_cuda_graph_batch_sizes would generate batch sizes like {1, 17, 33, 49, 65, 81, 97, 113, ...} regardless of max_batch_size.

Summary by CodeRabbit

• Bug Fixes

  • Fixed cuda_graph_batch_sizes heuristic to respect max_batch_size limits, ensuring generated batch sizes don't exceed the specified maximum.

• Tests

  • Added comprehensive test suite validating cuda_graph_batch_sizes behavior across various max_batch_size configurations.

_{✏️ Tip: You can customize this high-level summary in your review settings.}

Description

Test Coverage

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.

[coderabbitai]

📝 Walkthrough

Walkthrough

Modified the cuda_graph_batch_sizes heuristic to cap batch sizes at max_batch_size instead of always generating sizes up to 128. When cuda_graph_batch_sizes is None, additional sizes are now only added if they do not exceed max_batch_size. Corresponding test suite added to validate the updated behavior.

Changes

Cohort / File(s)	Summary
Production Code tensorrt_llm/_torch/auto_deploy/llm_args.py	Modified update_cuda_graph_batch_sizes method to guard heuristic batch size augmentation. Only adds sizes (1, 17, 33, ..., 127 stepping by 16) when they remain within max_batch_size bounds, preventing unnecessary batch sizes exceeding configured limits.
Test Suite tests/unittest/_torch/auto_deploy/unit/singlegpu/shim/test_llm_config.py	Added TestCudaGraphBatchSizesHeuristic with four test cases: small max_batch_size capping (4), medium max_batch_size validation (64), large max_batch_size inclusion (256), explicit size filtering (16), and descending sort order verification.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~15 minutes

🚥 Pre-merge checks | ✅ 3

✅ Passed checks (3 passed)

Check name	Status	Explanation
Docstring Coverage	✅ Passed	Docstring coverage is 85.71% which is sufficient. The required threshold is 80.00%.
Title check	✅ Passed	The title clearly and specifically describes the bug being fixed: CUDA graph batch sizes exceeding max_batch_size, directly aligned with the primary change.
Description check	✅ Passed	The PR description explains the issue and solution, and the author checked the PR checklist. However, the description is minimal and lacks explicit details about test coverage for the heuristic changes.

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

✨ Finishing touches

• 📝 Generate docstrings

---

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.}

[MrGeva]

/bot run

[tensorrt-cicd]

PR_Github #32091 [ run ] triggered by Bot. Commit: af302d0

[tensorrt-cicd]

PR_Github #32091 [ run ] completed with state SUCCESS. Commit: af302d0
/LLM/main/L0_MergeRequest_PR pipeline #24871 completed with status: 'SUCCESS'

[MrGeva]

/bot run

[tensorrt-cicd]

PR_Github #32138 [ run ] triggered by Bot. Commit: a5eb4cd

[tensorrt-cicd]

PR_Github #32138 [ run ] completed with state SUCCESS. Commit: a5eb4cd
/LLM/main/L0_MergeRequest_PR pipeline #24916 completed with status: 'SUCCESS'

[MrGeva]

/bot run

[MrGeva]

/bot run

[tensorrt-cicd]

PR_Github #32452 [ run ] triggered by Bot. Commit: 6a974f1

[tensorrt-cicd]

PR_Github #32452 [ run ] completed with state SUCCESS. Commit: 6a974f1
/LLM/main/L0_MergeRequest_PR pipeline #25139 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

[MrGeva]

/bot run

[tensorrt-cicd]

PR_Github #32457 [ run ] triggered by Bot. Commit: 6a974f1

[tensorrt-cicd]

PR_Github #32457 [ run ] completed with state SUCCESS. Commit: 6a974f1
/LLM/main/L0_MergeRequest_PR pipeline #25144 completed with status: 'SUCCESS'