Skip to content

Conversation

@Fridah-nv
Copy link
Collaborator

@Fridah-nv Fridah-nv commented Nov 10, 2025

NOTE: The update can slow down run_shape_prop a bit. e.g. 1.183s -> 1.208s for llama3 8B.

closes #8924

Summary by CodeRabbit

  • Refactor
    • Updated internal shape propagation execution to use Python dispatcher context for improved dispatch behavior handling.

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.

@Fridah-nv Fridah-nv self-assigned this Nov 10, 2025
@Fridah-nv Fridah-nv requested a review from a team as a code owner November 10, 2025 16:54
@Fridah-nv
Copy link
Collaborator Author

/bot run

@coderabbitai
Copy link
Contributor

coderabbitai bot commented Nov 10, 2025

📝 Walkthrough

Walkthrough

Import of enable_python_dispatcher from torch's dispatch module added to handle torch 2.9 compatibility. Shape propagation in _run_shape_prop_single_gm now executes within a Python dispatcher context when all inputs are available, modifying the execution dispatch behavior without altering external interfaces.

Changes

Cohort / File(s) Summary
Torch 2.9 dispatcher compatibility
tensorrt_llm/_torch/auto_deploy/utils/_graph.py
Added import of enable_python_dispatcher from torch._dispatch.python. Wrapped FakeTensorProp shape propagation call in _run_shape_prop_single_gm with enable_python_dispatcher() context when all inputs available.

Estimated code review effort

🎯 2 (Simple) | ⏱️ ~8 minutes

  • Verify that enable_python_dispatcher context is necessary for torch 2.9 compatibility and correctly preserves shape propagation semantics
  • Confirm no unintended side effects from dispatcher context on downstream shape propagation consumers
  • Validate the conditional logic (when all inputs are available) remains intact

Pre-merge checks and finishing touches

❌ Failed checks (2 warnings)
Check name Status Explanation Resolution
Description check ⚠️ Warning The PR description is incomplete. Key sections like 'Description' and 'Test Coverage' are empty templates, with no actual explanation of the issue, solution, or test details provided. Fill in the Description section explaining the FakeTensorProp issue and the context manager solution. Add Test Coverage section listing relevant tests. Provide implementation details in the template.
Docstring Coverage ⚠️ Warning Docstring coverage is 0.00% which is insufficient. The required threshold is 80.00%. You can run @coderabbitai generate docstrings to improve docstring coverage.
✅ Passed checks (3 passed)
Check name Status Explanation
Title check ✅ Passed The title clearly identifies the main change: adding a context manager to fix FakeTensorProp issue, with proper NVBugs reference and fix type.
Linked Issues check ✅ Passed The PR implementation aligns with issue #8924's objective to address torch 2.9 upgrade bugs in AutoDeploy by fixing FakeTensorProp with the enable_python_dispatcher context manager.
Out of Scope Changes check ✅ Passed All changes are scoped to fix FakeTensorProp in AutoDeploy, directly addressing the linked issue #8924 objective of resolving torch 2.9 upgrade issues.
✨ Finishing touches
  • 📝 Generate docstrings
🧪 Generate unit tests (beta)
  • Create PR with unit tests
  • Post copyable unit tests in a comment

📜 Recent review details

Configuration used: Path: .coderabbit.yaml

Review profile: CHILL

Plan: Pro

📥 Commits

Reviewing files that changed from the base of the PR and between 50c4863 and c8451b7.

📒 Files selected for processing (1)
  • tensorrt_llm/_torch/auto_deploy/utils/_graph.py (2 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}

📄 CodeRabbit inference engine (CODING_GUIDELINES.md)

Use only spaces, no tabs; indent with 4 spaces.

Files:

  • tensorrt_llm/_torch/auto_deploy/utils/_graph.py
**/*.py

📄 CodeRabbit inference engine (CODING_GUIDELINES.md)

**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.

Files:

  • tensorrt_llm/_torch/auto_deploy/utils/_graph.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}

📄 CodeRabbit inference engine (CODING_GUIDELINES.md)

Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).

Files:

  • tensorrt_llm/_torch/auto_deploy/utils/_graph.py
⏰ Context from checks skipped due to timeout of 90000ms. You can increase the timeout in your CodeRabbit configuration to a maximum of 15 minutes (900000ms). (1)
  • GitHub Check: Pre-commit Check
🔇 Additional comments (2)
tensorrt_llm/_torch/auto_deploy/utils/_graph.py (2)

214-215: Context manager implementation looks correct.

The enable_python_dispatcher() context wraps the FakeTensorProp propagation as intended. The implementation is clean and localized. Note that while this modifies the dispatch behavior internally, the function's external interface remains unchanged.

The PR description mentions a slight performance impact (~2% slowdown in the example). This tradeoff appears acceptable for torch 2.9 compatibility.


10-10: The import and usage pattern are confirmed as the recommended PyTorch 2.9 solution.

The enable_python_dispatcher() from torch._dispatch.python is the documented recommended approach for handling FakeTensorProp with dynamic shapes in PyTorch 2.9+, and this pattern aligns with PyTorch's internal implementation for AOT/autograd code. Your proposed changes follow the correct and supported pattern for torch 2.9 compatibility.


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

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

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24033 [ run ] triggered by Bot. Commit: c8451b7

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24033 [ run ] completed with state FAILURE. Commit: c8451b7
/LLM/main/L0_MergeRequest_PR pipeline #18107 completed with status: 'FAILURE'

@Fridah-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24038 [ run ] triggered by Bot. Commit: 198cc81

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24038 [ run ] completed with state SUCCESS. Commit: 198cc81
/LLM/main/L0_MergeRequest_PR pipeline #18112 completed with status: 'FAILURE'

@Fridah-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24046 [ run ] triggered by Bot. Commit: 198cc81

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24046 [ run ] completed with state SUCCESS. Commit: 198cc81
/LLM/main/L0_MergeRequest_PR pipeline #18120 completed with status: 'FAILURE'

Signed-off-by: Fridah-nv <201670829+Fridah-nv@users.noreply.github.com>
Signed-off-by: Fridah-nv <201670829+Fridah-nv@users.noreply.github.com>
@Fridah-nv Fridah-nv force-pushed the user/fridah/dispatcher branch from 198cc81 to fe7b876 Compare November 10, 2025 20:29
@Fridah-nv
Copy link
Collaborator Author

/bot run

@tensorrt-cicd
Copy link
Collaborator

PR_Github #24048 [ run ] triggered by Bot. Commit: fe7b876

@lucaslie lucaslie moved this from Backlog to In review in AutoDeploy Board Nov 10, 2025
@Fridah-nv Fridah-nv enabled auto-merge (squash) November 10, 2025 23:29
@tensorrt-cicd
Copy link
Collaborator

PR_Github #24048 [ run ] completed with state SUCCESS. Commit: fe7b876
/LLM/main/L0_MergeRequest_PR pipeline #18122 completed with status: 'SUCCESS'

@Fridah-nv Fridah-nv merged commit f40e1f7 into NVIDIA:main Nov 11, 2025
5 checks passed
@github-project-automation github-project-automation bot moved this from In review to Done in AutoDeploy Board Nov 11, 2025
suyoggupta pushed a commit to nv-auto-deploy/TensorRT-LLM that referenced this pull request Nov 12, 2025
…op (NVIDIA#9047)

Signed-off-by: Fridah-nv <201670829+Fridah-nv@users.noreply.github.com>
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment

Labels

None yet

Projects

Archived in project

Development

Successfully merging this pull request may close these issues.

[Feature]: AutoDeploy: torch 2.9 upgrade

3 participants