[https://nvbugs/5606166][fix] AutoDeploy: use tuples for cudagraph shape lookup#8772
Conversation
|
/bot run |
📝 WalkthroughWalkthroughThis change normalizes shape keys in torch CUDA graph capture to use explicit tuples and updates a test to force execution of the captured graph path instead of the underlying model's forward method. Changes
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes
Pre-merge checks and finishing touches❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
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. Comment |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 0
🧹 Nitpick comments (1)
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py (1)
145-145: LGTM! Explicit tuple conversion matches the pattern from line 127.The explicit
tuple()conversion ensures consistent shape key types for graph lookup, preventing potential mismatches betweentorch.Sizeand plain tuples.Optional: The static analysis tool suggests using iterable unpacking instead of tuple concatenation for a minor style improvement:
rounded_shapes = [ - (self.round_to_cuda_batch_size(input.shape[0]),) + tuple(input.shape[1:]) + (*((self.round_to_cuda_batch_size(input.shape[0]),),), *input.shape[1:]) for input in args_batched ]However, the current form is clear and readable, so this refactor is entirely optional.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py(2 hunks)tests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.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/compile/backends/torch_cudagraph.pytests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.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/compile/backends/torch_cudagraph.pytests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.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/compile/backends/torch_cudagraph.pytests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.py
🧠 Learnings (2)
📚 Learning: 2025-10-20T17:09:21.560Z
Learnt from: nvchenghaoz
PR: NVIDIA/TensorRT-LLM#8469
File: tensorrt_llm/_torch/auto_deploy/transform/library/rms_norm.py:180-182
Timestamp: 2025-10-20T17:09:21.560Z
Learning: In tensorrt_llm/_torch/auto_deploy/transform/library/rms_norm.py, the _gated_rmsnorm_replacement function does not need to cast the output of torch.ops.auto_deploy.torch_rmsnorm_gated back to the input dtype, even though the custom op returns fp32. The dtype handling is managed elsewhere or the fp32 output is acceptable for downstream consumers.
Applied to files:
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py
📚 Learning: 2025-09-04T07:33:10.618Z
Learnt from: MrGeva
PR: NVIDIA/TensorRT-LLM#7219
File: tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py:162-168
Timestamp: 2025-09-04T07:33:10.618Z
Learning: When users explicitly provide cuda_graph_batch_sizes in TorchCudagraphCompiler, respect their choices and only sanitize the values (clamp, dedupe, sort) without forcing additional batch sizes like 1 or max_batch_size. Only add commonly-used batch sizes when falling back to the heuristic.
Applied to files:
tests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.py
🧬 Code graph analysis (1)
tests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.py (1)
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py (1)
forward(130-164)
🪛 Ruff (0.14.2)
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py
145-145: Consider iterable unpacking instead of concatenation
Replace with iterable unpacking
(RUF005)
tests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.py
39-39: Unused static method argument: args
(ARG004)
39-39: Unused static method argument: kwargs
(ARG004)
40-40: Avoid specifying long messages outside the exception class
(TRY003)
⏰ 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 (3)
tensorrt_llm/_torch/auto_deploy/compile/backends/torch_cudagraph.py (1)
127-127: LGTM! Explicit tuple conversion ensures consistent dictionary keys.Converting
input.shape(atorch.Sizeobject) to an explicit tuple ensures reliable hashing and equality checks when usingcombined_shapeas a dictionary key inself.graphs.tests/unittest/_torch/auto_deploy/unit/singlegpu/compile/test_cuda_graph_batch_sizes.py (2)
38-40: LGTM! Test helper correctly forces CUDA graph path.The
_raise_error_for_forwardhelper is well-designed to ensure that the test exercises the captured graph path by preventing fallback to the model's forward method.Note: The static analysis warnings about unused arguments and long error messages are false positives in this test context—the signature must match
forward(*args, **kwargs)for the substitution to work correctly.
204-210: LGTM! Test modification correctly validates CUDA graph usage.The temporary replacement of
captured_graph.model.forwardwith the error-raising helper effectively verifies that the captured CUDA graph path is exercised. The try/finally block ensures proper cleanup.This test change complements the shape key normalization fix in the main file by confirming that graph lookup works correctly for various batch sizes.
|
PR_Github #22940 [ run ] triggered by Bot. Commit: |
|
PR_Github #22940 [ run ] completed with state |
…ape lookup Signed-off-by: Lucas Liebenwein <11156568+lucaslie@users.noreply.github.com>
lucaslie
force-pushed
the
ll/release_shape_tuple_cudagraph
branch
from
8f2b8be to
a024d81
Compare
|
/bot run |
|
PR_Github #23086 [ run ] triggered by Bot. Commit: |
|
NOTE: wait until #8621 is merged |
|
PR_Github #23086 [ run ] completed with state |
4 participants
Cherry-picked to release branch from #8658
Summary by CodeRabbit
Bug Fixes
Tests
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
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 thestage-listparameter 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.mdand the
scripts/test_to_stage_mapping.pyhelper.kill
killKill all running builds associated with pull request.
skip
skip --comment COMMENTSkip 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-pipelineReuse 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.