feat: Support structural tag in C++ runtime and upgrade xgrammar to 0.1.21 #6408
Conversation
Signed-off-by: Enwei Zhu <[email protected]>
Signed-off-by: Enwei Zhu <[email protected]>
Signed-off-by: Enwei Zhu <[email protected]>
📝 WalkthroughWalkthroughThis update introduces enhancements to guided decoding using xgrammar by improving tokenizer metadata handling, enabling grammar compiler caching, and supporting a new Changes
Sequence Diagram(s)sequenceDiagram
participant User
participant Executor
participant GuidedDecoder
participant GrammarCompiler
participant EnvUtils
User->>Executor: Submit request with guided decoding (may use kSTRUCTURAL_TAG)
Executor->>GuidedDecoder: Initialize with tokenizer metadata
GuidedDecoder->>EnvUtils: getFloatEnv("XGRAMMAR_CACHE_LIMIT_GB")
EnvUtils-->>GuidedDecoder: Return cache size limit (optional)
GuidedDecoder->>GrammarCompiler: Instantiate with tokenizer info and cache config
GuidedDecoder->>GuidedDecoder: Switch on guide type
alt guide type is kSTRUCTURAL_TAG
GuidedDecoder->>GrammarCompiler: Compile matcher for structured tag
else other guide types
GuidedDecoder->>GrammarCompiler: Compile matcher for other types
end
GrammarCompiler-->>GuidedDecoder: Return matcher
GuidedDecoder-->>Executor: Return matcher
Executor-->>User: Processed guided decoding result
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~15–20 minutes Suggested reviewers
Note ⚡️ Unit Test Generation is now available in beta!Learn more here, or try it out under "Finishing Touches" below. ✨ Finishing Touches
🧪 Generate unit tests
🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. Note: Be mindful of the bot's finite context window. It's strongly recommended to break down tasks such as reading entire modules into smaller chunks. For a focused discussion, use review comments to chat about specific files and their changes, instead of using the PR comments. CodeRabbit Commands (Invoked using PR comments)
Other keywords and placeholders
Documentation and Community
|
coderabbitai
bot
requested review from
chzblych,
LarryXFly,
pamelap-nvidia,
yilin-void and
yuxianq
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 2
📜 Review details
Configuration used: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
3rdparty/xgrammar(1 hunks)cpp/tensorrt_llm/batch_manager/CMakeLists.txt(1 hunks)cpp/tensorrt_llm/batch_manager/guidedDecoder.cpp(3 hunks)cpp/tensorrt_llm/common/envUtils.cpp(1 hunks)cpp/tensorrt_llm/common/envUtils.h(1 hunks)cpp/tensorrt_llm/executor/executorImpl.cpp(0 hunks)requirements.txt(1 hunks)tensorrt_llm/_torch/pyexecutor/grammar_matcher.py(1 hunks)tensorrt_llm/llmapi/llm_args.py(1 hunks)tensorrt_llm/sampling_params.py(1 hunks)
💤 Files with no reviewable changes (1)
- cpp/tensorrt_llm/executor/executorImpl.cpp
🧰 Additional context used
📓 Path-based instructions (4)
**/*.{cpp,h,hpp,cc,cxx}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.{cpp,h,hpp,cc,cxx}: Closing braces of namespaces should have a comment saying the namespace it closes (e.g., } // namespace foo).
Prefer const or constexpr variables over #defines whenever possible, as the latter are not visible to the compiler.
A variable that is not modified after its initialization should be declared as const.
Except 0 (only used in comparison for checking signness/existence/emptiness) and nullptr, true, false, all other literals should only be used for variable initialization.
Use the Allman indentation style for braces in C++ code.
Put the semicolon for an empty for or while loop in a new line.
The statement forming the body of a switch, while, do .. while or for statement shall be a compound statement (use brace-delimited statements).
If and else should always be followed by brace-delimited statements, even if empty or a single statement.
C++ filenames should use camel case with first letter lowercase (e.g., thisIsAFilename.cpp), and all files involved in the compilation of a target must have filenames that are case-insensitive unique.
All types (including class names) are camel case with uppercase first letter (e.g., FooBarClass).
Local variables, methods, and namespaces use camel case with first letter lowercase (e.g., localFooBar).
Non-magic-number global variables that are non-static and not defined in anonymous namespace use camel case prefixed by a lower case 'g' (e.g., gDontUseGlobalFoos).
Non-magic-number global variables that are static or defined in an anonymous namespace use camel case prefixed by a lower case 's' (e.g., sMutableStaticGlobal).
Locally visible static variable uses camel case with lowercase prefix 's' as the first letter of the name (e.g., static std::once_flag sFlag;).
Class member variables use camelcase prefixed with an 'm' (e.g., mNbFooValues). Public member variables do not require the 'm' prefix but it is highly encouraged for clarity.
Enumerations, global constants, static constants at class-scope, and f...
Files:
cpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.hcpp/tensorrt_llm/batch_manager/guidedDecoder.cpp
**/*.{cpp,h,hpp,cc,cxx,cu,py}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
Files:
cpp/tensorrt_llm/common/envUtils.cppcpp/tensorrt_llm/common/envUtils.htensorrt_llm/sampling_params.pycpp/tensorrt_llm/batch_manager/guidedDecoder.cpptensorrt_llm/_torch/pyexecutor/grammar_matcher.pytensorrt_llm/llmapi/llm_args.py
**/*.{h,hpp}
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
Use a preprocessor guard in header files. The guard name must have prefix TRTLLM_ followed by the filename, all in caps, and no trailing underscore.
Files:
cpp/tensorrt_llm/common/envUtils.h
**/*.py
📄 CodeRabbit Inference Engine (CODING_GUIDELINES.md)
**/*.py: The code developed for TensorRT-LLM should conform to Python 3.8+.
Indent Python code with 4 spaces. Do not use tabs.
Always maintain the namespace when importing in Python, even if only one class or function from a module is used.
Python filenames should use snake_case (e.g., some_file.py).
Python classes should use PascalCase (e.g., class SomeClass).
Python functions and methods should use snake_case (e.g., def my_awesome_function():).
Python local variables should use snake_case. Prefix k for variable names that start with a number (e.g., k_99th_percentile = ...).
Python global variables should use upper snake_case and prefix G (e.g., G_MY_GLOBAL = ...).
Python constants should use upper snake_case (e.g., MY_CONSTANT = ...).
Avoid shadowing variables declared in an outer scope in Python.
Initialize all externally visible members of a Python class in the constructor.
For interfaces that may be used outside a file, prefer docstrings over comments in Python.
Comments in Python should be reserved for code within a function, or interfaces that are local to a file.
Use Google style docstrings for Python classes and functions, which can be parsed by Sphinx.
Attributes and variables in Python can be documented inline; attribute docstrings will be rendered under the docstring for the class.
Avoid using reflection in Python when functionality can be easily achieved without it.
When using try-except blocks in Python, limit the except to the smallest set of errors possible.
When using try-except blocks to handle multiple possible variable types in Python, keep the body of the try as small as possible, using the else block to implement the logic.
Files:
tensorrt_llm/sampling_params.pytensorrt_llm/_torch/pyexecutor/grammar_matcher.pytensorrt_llm/llmapi/llm_args.py
🧠 Learnings (1)
cpp/tensorrt_llm/batch_manager/CMakeLists.txt (1)
Learnt from: CR
PR: NVIDIA/TensorRT-LLM#0
File: CODING_GUIDELINES.md:0-0
Timestamp: 2025-07-28T09:51:52.530Z
Learning: Applies to **/*.{cpp,h,hpp,cc,cxx,cu,py} : All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the current year. This includes .cpp, .h, .cu, .py, and any other source files which are compiled or interpreted.
🔇 Additional comments (9)
3rdparty/xgrammar (1)
1-1: Confirm compatibility & immutability of the new xgrammar commitThe pointer jump from
b14a281…to774867c…drags in an opaque set of changes.
Please double-check that:
- The commit belongs to the tagged 0.1.21 release (or another immutable tag) so future fetches remain reproducible.
- No public-API or build-system breakages were introduced that would surface only at runtime (e.g., symbol/ABI drift affecting the C++ guided-decoding integration).
- All unit / integration tests that exercise grammar compilation and guided decoding pass on CI across GPU/back-end matrices.
If none of the above was validated yet, consider adding a changelog excerpt to the PR description and re-running the full CI pipeline with
--reusedisabled to catch regressions early.cpp/tensorrt_llm/common/envUtils.cpp (1)
53-62: LGTM! Implementation follows established patterns.The
getFloatEnvfunction correctly follows the same implementation pattern as the existinggetIntEnvandgetUInt64Envfunctions. The use ofstd::stoffor conversion is appropriate, and the function properly returnsstd::nulloptwhen the environment variable is not set.requirements.txt (1)
54-54: xgrammar 0.1.21 verified and secure
- requirements.txt (line 54): xgrammar bumped from 0.1.19 → 0.1.21
- Confirmed 0.1.21 is published on PyPI
- Latest security advisory (DoS via unbounded in-memory cache) affected < 0.1.18 and was fixed in 0.1.18; 0.1.21 is not vulnerable
With version availability and patch status confirmed, this upgrade is safe to merge.
tensorrt_llm/sampling_params.py (1)
22-22: Documentation update correctly reflects the implementation changes.The docstring update accurately reflects that
structural_tagis now supported by the xgrammar backend rather than being limited to PyTorch backend only, which aligns with the broader changes in this PR.cpp/tensorrt_llm/batch_manager/CMakeLists.txt (1)
62-62: Build exclusion pattern correctly updated for xgrammar changes.The exclusion pattern change from
pybindtonanobindcorrectly reflects the xgrammar submodule update. This ensures that Python binding source files under the nanobind directory are properly excluded from the C++ build.cpp/tensorrt_llm/common/envUtils.h (1)
30-30: Function declaration is correct and consistent.The
getFloatEnvfunction declaration properly matches the implementation and follows the established pattern of other environment variable retrieval functions in this header.tensorrt_llm/llmapi/llm_args.py (1)
31-31: LGTM!The import statement formatting change improves readability by consolidating it to a single line, consistent with other imports in this block.
tensorrt_llm/_torch/pyexecutor/grammar_matcher.py (1)
52-66: Good refactoring to eliminate code duplication.The consolidation of
TokenizerInfocreation into a single instance improves maintainability by following the DRY principle. Setting default values upfront and conditionally updating them is cleaner than creating separate instances in each branch.cpp/tensorrt_llm/batch_manager/guidedDecoder.cpp (1)
46-58: Enhanced tokenizer metadata parsing looks good.The improved tokenizer configuration with JSON metadata extraction provides better flexibility for different tokenizer types. Good use of default values for
vocabTypeandaddPrefixSpace.
syuoni
changed the title
|
/bot run |
|
PR_Github #13257 [ run ] triggered by Bot |
|
PR_Github #13257 [ run ] completed with state |
|
/bot run |
|
PR_Github #13368 [ run ] triggered by Bot |
|
PR_Github #13368 [ run ] completed with state |
|
/bot run |
|
PR_Github #13540 [ run ] triggered by Bot |
|
PR_Github #13540 [ run ] completed with state |
….1.21 (NVIDIA#6408) Signed-off-by: Enwei Zhu <[email protected]> Signed-off-by: Lanyu Liao <[email protected]>
….1.21 (NVIDIA#6408) Signed-off-by: Enwei Zhu <[email protected]>
Summary by CodeRabbit
New Features
Bug Fixes
Documentation
Chores
Description
Test Coverage
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.
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.