[None][chore] Weekly mass integration of release/1.1 #9594
Conversation
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
9aa3592 to
bb19e3d
Compare
|
/bot run --disable-fail-fast |
📝 WalkthroughWalkthroughThis PR adds GIL-release guards to Python bindings for cache transceiver methods, updates many docs (including new guided-decoding and KV-cache-connector pages and multiple path/config edits), expands and clarifies example scripts, and removes a few TODO comments; one pybind binding chain termination was adjusted (semicolon) in the bindings file. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~25 minutes
Pre-merge checks and finishing touches✅ Passed checks (3 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: 7
🧹 Nitpick comments (6)
examples/llm-api/llm_runtime.py (1)
4-47: Consider using triple double quotes for the module docstring.The module-level docstring uses triple single quotes (
'''), while function docstrings in this file (lines 55-63, 95) use triple double quotes ("""). For consistency and to follow Python conventions (PEP 257), prefer triple double quotes for all docstrings.-''' +""" This script demonstrates various runtime configuration options in TensorRT-LLM, including KV cache management and CUDA graph optimizations. ... -''' +"""docs/source/features/guided-decoding.md (3)
9-9: Consider rephrasing for concision.The phrase "a variety of advanced serving scenarios" could be tightened to something like "several advanced serving scenarios" or "diverse advanced serving scenarios" for better readability. This is an optional style improvement.
68-70: Reword to reduce successive sentence repetition.Three consecutive sentences begin with "Description:". Consider restructuring to vary the sentence structure or combine related items, improving readability without changing content. Example: collapse the description into inline documentation or use different introductory phrases.
75-80: Clarify phrasing in example implementation overview.Lines 75–80 contain the phrase "file-system based KV cache." Consider: "file-system-based KV cache" (with hyphen between "file" and "system") to improve grammatical clarity per English style guidelines for compound adjectives.
docs/source/features/kv-cache-connector.md (2)
9-9: Consider rephrasing for concision.The phrase "a variety of advanced serving scenarios" could be tightened to "several advanced serving scenarios" or similar wording for slightly better readability. This is an optional style improvement.
77-79: Minor grammar improvement for compound adjectives.Line 77 uses "file-system based KV cache." Consider: "file-system-based KV cache" (adding hyphen between "file" and "system") to follow English style guidelines for compound adjectives. This is a minor polish suggestion.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (20)
cpp/include/tensorrt_llm/runtime/iTensor.h(0 hunks)cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cpp(1 hunks)cpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp(1 hunks)docs/source/commands/trtllm-serve/trtllm-serve.rst(4 hunks)docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md(0 hunks)docs/source/features/guided-decoding.md(1 hunks)docs/source/features/kv-cache-connector.md(1 hunks)docs/source/features/parallel-strategy.md(1 hunks)docs/source/features/sampling.md(1 hunks)docs/source/index.rst(1 hunks)docs/source/installation/linux.md(1 hunks)docs/source/overview.md(1 hunks)docs/source/quick-start-guide.md(1 hunks)examples/llm-api/llm_mgmn_llm_distributed.sh(2 hunks)examples/llm-api/llm_mgmn_trtllm_bench.sh(2 hunks)examples/llm-api/llm_mgmn_trtllm_serve.sh(2 hunks)examples/llm-api/llm_runtime.py(2 hunks)examples/llm-api/llm_sampling.py(3 hunks)examples/serve/deepseek_r1_reasoning_parser.sh(1 hunks)tensorrt_llm/functional.py(0 hunks)
💤 Files with no reviewable changes (3)
- cpp/include/tensorrt_llm/runtime/iTensor.h
- docs/source/deployment-guide/deployment-guide-for-gpt-oss-on-trtllm.md
- tensorrt_llm/functional.py
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{cpp,h,cu}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{cpp,h,cu}: Closing braces of namespaces should have a comment saying the namespace it closes (e.g.,} // namespace foo)
Preferconstorconstexprvariables over#definewhenever possible, as the latter are not visible to the compiler
A variable that is not modified after its initialization should be declared asconst
Except0(only used in comparison for checking signness/existence/emptiness) andnullptr,true,false, all other literals should only be used for variable initialization and should be replaced with named constants
Use Allman indentation style for braces in C++
Put the semicolon for an emptyfororwhileloop in a new line
The statement forming the body of aswitch,while,do .. whileorforstatement shall be a compound statement (use brace-delimited statements)
Ifandelseshould 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.,thisIsASubDirandthisIsAFilename.cpp)
All filenames involved in compilation of a compilation target must have case-insensitive unique filenames
All types (including class names) should use camel case with uppercase first letter (e.g.,FooBarClass)
Local variables, methods and namespaces should 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 should 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 should use camel case prefixed by a lower case 's' (e.g.,sMutableStaticGlobal)
Locally visible static variables should use camel case with lowercase prefix 's' as the first letter of the name (e.g.,static std::once_flag sFlag;)
Public, private and protected class member variables should use camel case prefixed with 'm' (e.g.,mNbFooValues), though the 'm' pre...
Files:
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cppcpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp
**/*.{cpp,h,cu,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code files should contain an NVIDIA copyright header that includes the current year at the top
Files:
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cppexamples/llm-api/llm_sampling.pyexamples/llm-api/llm_runtime.pycpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp
**/*.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 (e.g., usefrom package.subpackage import fooand thenfoo.SomeClass()instead offrom package.subpackage.foo import SomeClass)
Python filenames should use snake_case (e.g.,some_file.py)
Python class names should use PascalCase (e.g.,class SomeClass)
Python function and method names should use snake_case (e.g.,def my_awesome_function():)
Python local variable names should use snake_case, with prefixkfor variable names that start with a number (e.g.,k_99th_percentile = ...)
Python global variables should use upper snake_case with prefixG(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 Python interfaces that may be used outside a file, prefer docstrings over comments
Python comments 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
Python attributes and variables can be documented inline with type and description (e.g.,self.x = 5followed by"""<type>: Description of 'x'""")
Avoid using reflection in Python when functionality can be easily achieved without reflection
When using try-except blocks in Python, limit the except clause to the smallest set of specific errors possible instead of catching all exceptions
When using try-except blocks in Python to handle multiple possible variable types (duck-typing), keep the body of the try as small as possible and use the else block to implement the logic
Files:
examples/llm-api/llm_sampling.pyexamples/llm-api/llm_runtime.py
🧠 Learnings (25)
📚 Learning: 2025-08-27T17:50:13.264Z
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/features/parallel-strategy.mdexamples/llm-api/llm_mgmn_llm_distributed.shdocs/source/quick-start-guide.md
📚 Learning: 2025-07-28T17:06:08.621Z
Learnt from: moraxu
Repo: NVIDIA/TensorRT-LLM PR: 6303
File: tests/integration/test_lists/qa/examples_test_list.txt:494-494
Timestamp: 2025-07-28T17:06:08.621Z
Learning: In TensorRT-LLM testing, it's common to have both CLI flow tests (test_cli_flow.py) and PyTorch API tests (test_llm_api_pytorch.py) for the same model. These serve different purposes: CLI flow tests validate the traditional command-line workflow, while PyTorch API tests validate the newer LLM API backend. Both are legitimate and should coexist.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/features/parallel-strategy.mdexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.shdocs/source/overview.mddocs/source/quick-start-guide.md
📚 Learning: 2025-09-02T13:42:44.885Z
Learnt from: pcastonguay
Repo: NVIDIA/TensorRT-LLM PR: 7455
File: tensorrt_llm/_torch/pyexecutor/py_executor.py:1852-1860
Timestamp: 2025-09-02T13:42:44.885Z
Learning: In MPI communication within TensorRT-LLM pipeline parallelism, different communication types (tokens, logits, termination sync) must use disjoint tag namespaces to avoid message routing collisions when using the same source/destination patterns.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.sh
📚 Learning: 2025-09-23T14:58:05.372Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/kernels/nccl_device/config.cu:42-49
Timestamp: 2025-09-23T14:58:05.372Z
Learning: In TensorRT-LLM NCCL device kernels (cpp/tensorrt_llm/kernels/nccl_device/), the token partitioning intentionally uses ceil-like distribution (same token_per_rank for all ranks) to ensure all ranks launch the same number of blocks. This is required for optimal NCCL device API barrier performance, even though it may launch extra blocks for non-existent tokens on later ranks. Runtime bounds checking in the kernel (blockID validation) handles the overshoot cases.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.sh
📚 Learning: 2025-09-09T09:40:45.658Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 7645
File: tests/integration/test_lists/qa/llm_function_core.txt:648-648
Timestamp: 2025-09-09T09:40:45.658Z
Learning: In TensorRT-LLM test lists, it's common and intentional for the same test to appear in multiple test list files when they serve different purposes (e.g., llm_function_core.txt for comprehensive core functionality testing and llm_function_core_sanity.txt for quick sanity checks). This duplication allows tests to be run in different testing contexts.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/features/parallel-strategy.mdexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.shdocs/source/overview.md
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
Repo: NVIDIA/TensorRT-LLM PR: 6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/features/parallel-strategy.mdexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.sh
📚 Learning: 2025-08-01T15:14:45.673Z
Learnt from: yibinl-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 6506
File: examples/models/core/mixtral/requirements.txt:3-3
Timestamp: 2025-08-01T15:14:45.673Z
Learning: In TensorRT-LLM, examples directory can have different dependency versions than the root requirements.txt file. Version conflicts between root and examples dependencies are acceptable because examples are designed to be standalone and self-contained.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/overview.mdexamples/llm-api/llm_runtime.py
📚 Learning: 2025-08-20T07:43:36.447Z
Learnt from: ChristinaZ
Repo: NVIDIA/TensorRT-LLM PR: 7068
File: cpp/tensorrt_llm/kernels/moeTopKFuncs.cuh:169-172
Timestamp: 2025-08-20T07:43:36.447Z
Learning: In TensorRT-LLM MOE kernels, when processing up to 128 experts across 32 threads, each thread handles at most 4 experts (N < 5 constraint), where N represents candidates per thread rather than total system capacity.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.sh
📚 Learning: 2025-09-16T09:30:09.716Z
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7763
File: cpp/tensorrt_llm/CMakeLists.txt:297-301
Timestamp: 2025-09-16T09:30:09.716Z
Learning: In the TensorRT-LLM project, NCCL libraries are loaded earlier by PyTorch libraries or the bindings library, so the main shared library doesn't need NCCL paths in its RPATH - the libraries will already be available in the process address space when needed.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.sh
📚 Learning: 2025-08-26T09:49:04.956Z
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7192
File: tests/integration/test_lists/test-db/l0_dgx_b200.yml:56-72
Timestamp: 2025-08-26T09:49:04.956Z
Learning: In TensorRT-LLM test configuration files, the test scheduling system handles wildcard matching with special rules that prevent duplicate test execution even when the same tests appear in multiple yaml files with overlapping GPU wildcards (e.g., "*b200*" and "*gb200*").
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.sh
📚 Learning: 2025-11-27T09:23:18.742Z
Learnt from: fredricz-20070104
Repo: NVIDIA/TensorRT-LLM PR: 9511
File: tests/integration/defs/examples/serve/test_serve.py:136-186
Timestamp: 2025-11-27T09:23:18.742Z
Learning: In TensorRT-LLM testing, when adding test cases based on RCCA commands, the command format should be copied exactly as it appears in the RCCA case, even if it differs from existing tests. For example, some RCCA commands for trtllm-serve may omit the "serve" subcommand while others include it.
Applied to files:
examples/llm-api/llm_mgmn_trtllm_bench.shdocs/source/features/parallel-strategy.mdexamples/serve/deepseek_r1_reasoning_parser.shexamples/llm-api/llm_mgmn_llm_distributed.shexamples/llm-api/llm_mgmn_trtllm_serve.shdocs/source/commands/trtllm-serve/trtllm-serve.rstdocs/source/quick-start-guide.md
📚 Learning: 2025-08-21T00:16:56.457Z
Learnt from: farshadghodsian
Repo: NVIDIA/TensorRT-LLM PR: 7101
File: docs/source/blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.md:36-36
Timestamp: 2025-08-21T00:16:56.457Z
Learning: TensorRT-LLM container release tags in documentation should only reference published NGC container images. The README badge version may be ahead of the actual published container versions.
Applied to files:
docs/source/features/parallel-strategy.mdexamples/llm-api/llm_mgmn_llm_distributed.shdocs/source/overview.md
📚 Learning: 2025-08-11T20:09:24.389Z
Learnt from: achartier
Repo: NVIDIA/TensorRT-LLM PR: 6763
File: tests/integration/defs/triton_server/conftest.py:16-22
Timestamp: 2025-08-11T20:09:24.389Z
Learning: In the TensorRT-LLM test infrastructure, the team prefers simple, direct solutions (like hard-coding directory traversal counts) over more complex but robust approaches when dealing with stable directory structures. They accept the maintenance cost of updating tests if the layout changes.
Applied to files:
examples/llm-api/llm_mgmn_llm_distributed.shdocs/source/overview.md
📚 Learning: 2025-08-14T06:36:40.701Z
Learnt from: timlee0212
Repo: NVIDIA/TensorRT-LLM PR: 6886
File: tensorrt_llm/_torch/models/modeling_deepseekv3.py:0-0
Timestamp: 2025-08-14T06:36:40.701Z
Learning: In DeepSeek V3 model (tensorrt_llm/_torch/models/modeling_deepseekv3.py), the disagreement between AllReduce.__init__ guard and _compute_mlp_tp_size logic for MNNVL usage is expected by design. The AllReduce component and MLP TP-size computation intentionally use different criteria for MNNVL availability decisions.
Applied to files:
examples/llm-api/llm_mgmn_llm_distributed.sh
📚 Learning: 2025-09-18T05:41:45.847Z
Learnt from: pengbowang-nv
Repo: NVIDIA/TensorRT-LLM PR: 7120
File: tensorrt_llm/llmapi/llm.py:690-697
Timestamp: 2025-09-18T05:41:45.847Z
Learning: Kimi model support is currently focused on the PyTorch backend path, with TRT path support potentially coming later.
Applied to files:
docs/source/overview.md
📚 Learning: 2025-09-23T15:12:38.312Z
Learnt from: nv-lschneider
Repo: NVIDIA/TensorRT-LLM PR: 7910
File: cpp/tensorrt_llm/thop/allreduceOp.cpp:352-446
Timestamp: 2025-09-23T15:12:38.312Z
Learning: In TensorRT-LLM NCCL device implementation, NCCL version 2.28+ requirements are handled at runtime in the nccl_device/config layer rather than with compile-time guards. This allows the allreduceOp to remain version-agnostic and delegates version compatibility validation to the appropriate lower-level components that can gracefully handle unsupported configurations.
Applied to files:
docs/source/overview.md
📚 Learning: 2025-08-14T15:43:23.107Z
Learnt from: MatthiasKohl
Repo: NVIDIA/TensorRT-LLM PR: 6904
File: tensorrt_llm/_torch/attention_backend/trtllm.py:259-262
Timestamp: 2025-08-14T15:43:23.107Z
Learning: In TensorRT-LLM's attention backend, tensor parameters in the plan() method are assigned directly without validation (dtype, device, contiguity checks). This maintains consistency across all tensor inputs and follows the pattern of trusting callers to provide correctly formatted tensors.
Applied to files:
docs/source/overview.md
📚 Learning: 2025-08-27T14:23:55.566Z
Learnt from: ixlmar
Repo: NVIDIA/TensorRT-LLM PR: 7294
File: tensorrt_llm/_torch/modules/rms_norm.py:17-17
Timestamp: 2025-08-27T14:23:55.566Z
Learning: The TensorRT-LLM project requires Python 3.10+ as evidenced by the use of TypeAlias from typing module, match/case statements, and union type | syntax throughout the codebase, despite some documentation still mentioning Python 3.8+.
Applied to files:
docs/source/overview.md
📚 Learning: 2025-08-08T04:10:19.038Z
Learnt from: djns99
Repo: NVIDIA/TensorRT-LLM PR: 6728
File: cpp/tensorrt_llm/plugins/mixtureOfExperts/mixtureOfExpertsPlugin.cpp:966-966
Timestamp: 2025-08-08T04:10:19.038Z
Learning: TensorRT plugins currently don't support padding functionality, and TensorRT is not getting new features (in maintenance mode). This means that duplicating parameters like mExpertHiddenSize in function calls, even with TODO comments, can be acceptable as pragmatic solutions within these constraints.
Applied to files:
docs/source/overview.md
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which can contain default `cuda_graph_config` values, so `llm_args` may already have this config before the extra options processing.
Applied to files:
docs/source/commands/trtllm-serve/trtllm-serve.rstexamples/llm-api/llm_runtime.py
📚 Learning: 2025-08-21T09:41:49.347Z
Learnt from: eopXD
Repo: NVIDIA/TensorRT-LLM PR: 6768
File: cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp:2010-2045
Timestamp: 2025-08-21T09:41:49.347Z
Learning: In cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp, updateSequenceCacheBlockOffsets is specifically for updating bookkeeping when blocks are added during the context phase, not for refreshing offsets after detach operations. During detach operations, GenerationRequest::removeFrontBlock handles the necessary cache block bookkeeping internally.
Applied to files:
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cppcpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp
📚 Learning: 2025-08-20T06:48:45.368Z
Learnt from: eopXD
Repo: NVIDIA/TensorRT-LLM PR: 6768
File: cpp/include/tensorrt_llm/batch_manager/kvCacheManager.h:0-0
Timestamp: 2025-08-20T06:48:45.368Z
Learning: In cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp, updateSequenceCacheBlockOffsets is only called when adding a sequence, not during detach operations. During detach, the cache block bookkeeping is handled by GenerationRequest::removeFrontBlock.
Applied to files:
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cppcpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp
📚 Learning: 2025-08-06T08:18:28.669Z
Learnt from: zhengd-nv
Repo: NVIDIA/TensorRT-LLM PR: 6633
File: cpp/tensorrt_llm/batch_manager/dataTransceiverImpl.cpp:145-155
Timestamp: 2025-08-06T08:18:28.669Z
Learning: In cpp/tensorrt_llm/batch_manager/dataTransceiverImpl.cpp, the existing `mMtxForMap` mutex in DataSenderImpl is sufficient to synchronize measurement file operations in the `release` method, as all file operations occur within the same critical section that protects the `mRequestToSession` map access.
Applied to files:
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cppcpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp
📚 Learning: 2025-10-17T13:21:31.724Z
Learnt from: ixlmar
Repo: NVIDIA/TensorRT-LLM PR: 8398
File: tensorrt_llm/_torch/pyexecutor/sampling_utils.py:237-272
Timestamp: 2025-10-17T13:21:31.724Z
Learning: The setup.py file in TensorRT-LLM explicitly requires Python 3.10+ via `python_requires=">=3.10, <4"`, making match/case statements and other Python 3.10+ features appropriate throughout the codebase.
Applied to files:
docs/source/installation/linux.md
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM's bench configuration, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which is a Dict[str, Any] that can contain default values including `cuda_graph_config`, making the fallback `llm_args["cuda_graph_config"]` safe to use.
Applied to files:
examples/llm-api/llm_runtime.py
🧬 Code graph analysis (3)
cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cpp (1)
cpp/tensorrt_llm/batch_manager/cacheTransceiver.cpp (6)
checkContextTransferStatus(419-522)checkContextTransferStatus(419-419)checkGenTransferStatus(524-678)checkGenTransferStatus(524-524)checkGenTransferComplete(680-683)checkGenTransferComplete(680-680)
examples/llm-api/llm_sampling.py (3)
tensorrt_llm/_torch/pyexecutor/sampler.py (1)
beam_width(134-137)tests/unittest/_torch/modeling/test_modeling_out_of_tree.py (1)
sampling_params(58-59)tensorrt_llm/_torch/pyexecutor/model_engine.py (1)
use_beam_search(522-523)
cpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp (1)
cpp/tensorrt_llm/batch_manager/cacheTransceiver.cpp (6)
checkContextTransferStatus(419-522)checkContextTransferStatus(419-419)checkGenTransferStatus(524-678)checkGenTransferStatus(524-524)checkGenTransferComplete(680-683)checkGenTransferComplete(680-680)
🪛 LanguageTool
docs/source/features/kv-cache-connector.md
[style] ~9-~9: Consider using a more concise synonym.
Context: ... Cache Connector is designed to support a variety of advanced serving scenarios: 1. **KV Ca...
(A_VARIETY_OF)
[style] ~68-~68: Three successive sentences begin with the same word. Consider rewording the sentence or use a thesaurus to find a synonym.
Context: ...-> tuple[list[int], list[int]]`** * Description: Polled by the runtime to check the s...
(ENGLISH_WORD_REPEAT_BEGINNING_RULE)
[grammar] ~77-~77: Use a hyphen to join words.
Context: ...w This example implements a file-system based KV cache. 1.Save: When a reque...
(QB_NEW_EN_HYPHEN)
⏰ 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 (16)
examples/llm-api/llm_sampling.py (2)
194-195: LGTM!The beam search demonstration is correctly integrated into the showcase flow.
239-240: LGTM!The CLI handling for the beam search demo is correctly implemented and follows the established pattern.
docs/source/commands/trtllm-serve/trtllm-serve.rst (2)
257-273: New documentation section adds clear YAML configuration guidance.The new "Configuring with YAML Files" section is well-structured and provides practical examples for both top-level and nested argument configuration. The explanation of the YAML file hierarchy and the code examples are clear and helpful for users.
44-44: Fix API reference mismatch between display text and link target.The documentation displays
tensorrt_llm.llmapi.LlmArgsbut links totensorrt_llm.llmapi.TorchLlmArgs. These are distinct classes serving different backends (TensorRT vs PyTorch). Ensure the display text matches the target class being documented, or update the link to align with the intended API being referenced.Additionally, verify that the updated literalinclude paths (
../../../../examples/...for lines 44, 50, 59, 65, 100, 106) correctly resolve from the RST file location atdocs/source/commands/trtllm-serve/trtllm-serve.rst.docs/source/installation/linux.md (1)
12-16: LGTM!The CUDA Toolkit version update to 13.0 and the addition of the
cuda-compat-13-0package note with the forward compatibility link are accurate and align with the PyTorch CUDA 13.0 package installation on line 20.examples/llm-api/llm_runtime.py (1)
94-107: LGTM!The expanded
example_kv_cache_configfunction with inline comments documentingKvCacheConfigparameters improves readability and serves as a good example for users.docs/source/features/sampling.md (1)
2-2: LGTM!The removal of "guided decoding" from this page is consistent with the PR's documentation reorganization, where guided decoding is now documented in its own dedicated page (
docs/source/features/guided-decoding.md).examples/serve/deepseek_r1_reasoning_parser.sh (1)
3-20: Verify the batch size configuration consistency.The YAML config sets
cuda_graph_config.max_batch_size: 512(line 6), while the CLI flag--max_batch_size 1024(line 20) specifies a larger value. This may be intentional (CUDA graphs only capture up to batch size 512 for memory efficiency, while the runtime can handle up to 1024), but please verify this is the desired behavior.cpp/tensorrt_llm/nanobind/batch_manager/cacheTransceiver.cpp (1)
91-94: LGTM on GIL release guards.Adding
nb::call_guard<nb::gil_scoped_release>()forcheck_context_transfer_statusandcheck_gen_transfer_statusis appropriate since these methods wait on transfer futures and may block. Releasing the GIL allows other Python threads to continue execution during the wait.cpp/tensorrt_llm/pybind/batch_manager/cacheTransceiver.cpp (1)
87-90: LGTM on GIL release guards.The
py::call_guard<py::gil_scoped_release>()additions forcheck_context_transfer_statusandcheck_gen_transfer_statusappropriately release the GIL during potentially blocking transfer status checks, consistent with the nanobind implementation.docs/source/features/parallel-strategy.md (1)
83-83: LGTM!The added guidance clearly instructs users how to apply the YAML configuration when deploying with
trtllm-serveortrtllm-bench. The placement follows the configuration example logically and maintains consistency with existing documentation style.docs/source/quick-start-guide.md (1)
34-34: LGTM!The added reference guides users to the pre-quantized model collection, creating a natural pathway to explore optimization options beyond the quick-start examples. The external link to the NVIDIA inference-optimized models collection aligns well with the deployment context.
docs/source/index.rst (1)
74-74: LGTM!The two new feature documentation pages are correctly integrated into the Features toctree. Placements are consistent with the existing navigation structure and properly reference the new files introduced in this PR.
Also applies to: 80-80
docs/source/features/guided-decoding.md (1)
1-583: Comprehensive feature documentation with strong coverage.The guided decoding documentation is well-structured with clear examples for both online (
trtllm-serve) and offline (LLM API) usage paths. Code snippets are syntactically correct and demonstrate practical use cases across all supported grammar types.docs/source/features/kv-cache-connector.md (1)
1-105: Well-structured API and architecture documentation.The KV Cache Connector documentation provides clear API reference material with well-organized scheduler and worker interfaces. The example implementation section effectively illustrates design patterns and includes honest limitations assessment, helping developers understand production requirements.
docs/source/overview.md (1)
41-52: LGTM! Navigation restructuring improves usability.Converting inline feature descriptions to cross-referenced links enhances documentation navigation and reduces redundancy. The relative paths are consistent and properly formatted. Links to newly introduced feature pages (guided-decoding.md) integrate naturally with the existing feature structure.
Verify that all linked feature files exist and that relative paths resolve correctly in the Sphinx documentation build. All target files appear to be present in the PR, but confirming path resolution would be helpful.
|
PR_Github #26454 [ run ] triggered by Bot. Commit: |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
bb19e3d to
9476886
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #26459 [ run ] triggered by Bot. Commit: |
|
PR_Github #26454 [ run ] completed with state |
|
PR_Github #26459 [ run ] completed with state |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
9476886 to
49a1cbf
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #26638 [ run ] triggered by Bot. Commit: |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
49a1cbf to
6699ed7
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #26660 [ run ] triggered by Bot. Commit: |
|
PR_Github #26638 [ run ] completed with state |
|
PR_Github #26660 [ run ] completed with state |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
2 times, most recently
from
3262e2e to
731fa6d
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #26811 [ run ] triggered by Bot. Commit: |
|
PR_Github #26811 [ run ] completed with state |
|
/bot run --disable-fail-fast |
|
PR_Github #26988 [ run ] triggered by Bot. Commit: |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
731fa6d to
2e3e7d2
Compare
|
/bot run --disable-fail-fast |
|
PR_Github #26994 [ run ] triggered by Bot. Commit: |
|
PR_Github #26988 [ run ] completed with state |
|
PR_Github #26994 [ run ] completed with state |
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
2e3e7d2 to
eaa6d1b
Compare
|
/bot skip --comment "Failed tests are flakey, filed #9749 to waive" (permission to skip CI was obtained from @schetlur-nv) |
|
PR_Github #27147 [ skip ] triggered by Bot. Commit: |
|
PR_Github #27147 [ skip ] completed with state |
|
Failed tests have been waived by #9749 |
Signed-off-by: Iman Tabrizian <10105175+tabrizian@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: jthomson04 <jwillthomson19@gmail.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Yan Chunwei <328693+Superjomn@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Pengyun Lin <81065165+LinPoly@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Enwei Zhu <21126786+syuoni@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Yiqing Yan <yiqingy@nvidia.com> Co-authored-by: Yanchao Lu <yanchaol@nvidia.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: Yan Chunwei <328693+Superjomn@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
…VIDIA#9571) Signed-off-by: junq <22017000+QiJune@users.noreply.github.com> Signed-off-by: Mike Iovine <6158008+mikeiovine@users.noreply.github.com> Signed-off-by: Mike Iovine <miovine@nvidia.com>
mikeiovine
force-pushed
the
mass-integrate-1.1
branch
from
eaa6d1b to
1a3942e
Compare
|
/bot skip --comment "see above" |
|
PR_Github #27162 [ skip ] triggered by Bot. Commit: |
|
PR_Github #27162 [ skip ] completed with state |
10 participants
Description
PRs explicitly excluded in this round:
Test Coverage
Existing tests.
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 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.
Summary by CodeRabbit
New Features
Documentation
Performance
Chores
✏️ Tip: You can customize this high-level summary in your review settings.