[None][feat] Use list instead of torch tensor for new tokens in update requests #7730
Conversation
|
/bot run |
📝 WalkthroughWalkthroughBindings in C++ (nanobind and pybind) change add_new_tokens_to_requests to accept per-request token lists. Python sampler introduces token caching to defer mutations and commits them in batches via the updated binding. Control flow in drafting/rejection paths routes token additions through the cache and commits at cycle end. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Sampler
participant Cache as Per-request Token Cache
participant Binding as add_new_tokens_to_requests
participant LlmReq as LlmRequest
rect rgb(245,245,255)
note over Sampler: New flow (cached, batched commit)
Sampler->>Cache: add_token(request, token, beam, step, cache)
Sampler->>Cache: ... repeat per path (draft/reject)
Sampler->>Binding: add_new_tokens_to_requests(requests, tokens_per_request, beam)
Binding->>LlmReq: for each request: push each token
end
sequenceDiagram
autonumber
participant Sampler
participant Binding as add_new_token (old usage)
participant LlmReq as LlmRequest
rect rgb(255,245,245)
note over Sampler: Previous flow (immediate mutation)
Sampler->>Binding: add_new_token(request, token, beam)
Binding->>LlmReq: push token
Sampler->>Binding: ... repeated per step
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests
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 |
|
PR_Github #18629 [ run ] triggered by Bot |
![coderabbitai[bot]](https://avatars.githubusercontent.com/in/347564?s=60&v=4){kind=small}
There was a problem hiding this comment.
Actionable comments posted: 1
Caution
Some comments are outside the diff and can’t be posted inline due to platform limitations.
⚠️ Outside diff range comments (1)
tensorrt_llm/_torch/pyexecutor/sampler.py (1)
1-1: Add NVIDIA Apache-2.0 header.Project guideline requires the license header at the top of all Python files.
+# SPDX-FileCopyrightText: Copyright (c) 2025 NVIDIA CORPORATION & AFFILIATES. +# SPDX-License-Identifier: Apache-2.0
🧹 Nitpick comments (4)
cpp/tensorrt_llm/pybind/batch_manager/bindings.cpp (1)
438-454: Make the binding more efficient and robust (const&, size_t loop, release GIL).
- Take requests by const& (you mutate the pointees, not the vector).
- Use size_t for the index to match requests.size().
- Release the GIL while pushing tokens to reduce contention on large batches.
- [](std::vector<std::shared_ptr<tb::LlmRequest>>& requests, - std::vector<std::vector<tb::LlmRequest::TokenIdType>> const& tokens, int beam_idx) + [](std::vector<std::shared_ptr<tb::LlmRequest>> const& requests, + std::vector<std::vector<tb::LlmRequest::TokenIdType>> const& tokens, int beam_idx) { TLLM_CHECK_WITH_INFO( requests.size() == tokens.size(), "Expected the same number of requests and token containers."); - for (int i = 0; i < requests.size(); ++i) + for (size_t i = 0; i < requests.size(); ++i) { for (const auto& token : tokens[i]) { requests[i]->addNewToken(token, beam_idx); } } }, - py::arg("requests"), py::arg("tokens"), py::arg("beam_idx"), + py::arg("requests"), py::arg("tokens"), py::arg("beam_idx"), + py::call_guard<py::gil_scoped_release>(), "Add new tokens to multiple LLM requests. The tokens vector should contain tokens for beam beam_idx of all " "requests in order.");cpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp (1)
431-447: Apply the same const&, size_t loop, and GIL release in nanobind binding.Keeps both bindings consistent and reduces overhead under contention.
- [](std::vector<std::shared_ptr<tb::LlmRequest>>& requests, - std::vector<std::vector<tb::LlmRequest::TokenIdType>> const& tokens, int beam_idx) + [](std::vector<std::shared_ptr<tb::LlmRequest>> const& requests, + std::vector<std::vector<tb::LlmRequest::TokenIdType>> const& tokens, int beam_idx) { TLLM_CHECK_WITH_INFO( requests.size() == tokens.size(), "Expected the same number of requests and token containers."); - for (int i = 0; i < requests.size(); ++i) + for (size_t i = 0; i < requests.size(); ++i) { for (const auto& token : tokens[i]) { requests[i]->addNewToken(token, beam_idx); } } }, - nb::arg("requests"), nb::arg("tokens"), nb::arg("beam_idx"), + nb::arg("requests"), nb::arg("tokens"), nb::arg("beam_idx"), + nb::call_guard<nb::gil_scoped_release>(), "Add new tokens to multiple LLM requests. The tokens vector should contain tokens for beam beam_idx of all " "requests in order.");tensorrt_llm/_torch/pyexecutor/sampler.py (2)
350-352: Fix type hint: avoid implicit Optional (RUF013).Annotate cache as Optional to satisfy typing and linters.
-def add_token(request: LlmRequest, - new_tokens: torch.Tensor, - *, - beam: int, - step: int = 0, - cache: dict[LlmRequest, List[int]] = None) -> int: +def add_token(request: LlmRequest, + new_tokens: torch.Tensor, + *, + beam: int, + step: int = 0, + cache: Optional[dict[LlmRequest, List[int]]] = None) -> int:
594-597: Stabilize request/token pairing on commit.keys() and values() preserve order in CPython, but zip(items) avoids any mismatch risk and reads clearer.
- # Commit adding tokens - add_new_tokens_to_requests(list(self._cached_tokens.keys()), - list(self._cached_tokens.values()), BEAM_0) + # Commit adding tokens + items = list(self._cached_tokens.items()) + if items: + reqs, toks = zip(*items) + add_new_tokens_to_requests(list(reqs), list(toks), BEAM_0)
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (3)
cpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp(1 hunks)cpp/tensorrt_llm/pybind/batch_manager/bindings.cpp(1 hunks)tensorrt_llm/_torch/pyexecutor/sampler.py(7 hunks)
🧰 Additional context used
📓 Path-based instructions (6)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh}: Namespace closing braces must include a trailing comment with the namespace name (e.g., '} // namespace foo').
Prefer const or constexpr variables over #define for constants.
Declare variables that are not modified after initialization as const.
Avoid magic literals in code; except for 0, nullptr, true, false. Use named constants for comparisons and logic.
Use Allman brace style for formatting.
Place the semicolon of an empty for/while loop on a new line.
Bodies of switch/while/do-while/for must be compound statements (brace-delimited), and if/else must always be followed by brace-delimited statements.
Type names (e.g., classes) must be CamelCase starting with an uppercase letter (e.g., FooBar).
Local variables, methods, and namespaces use lowerCamelCase (e.g., localFooBar).
Non-magic-number global variables that are non-static and not in an anonymous namespace must be lowerCamelCase prefixed with 'g' (e.g., gDontUseGlobalFoos).
Non-magic-number globals that are static or in an anonymous namespace use lowerCamelCase prefixed with 's' (e.g., sMutableStaticGlobal).
Locally visible static variables use lowerCamelCase with 's' prefix (e.g., static std::once_flag sFlag).
Private/protected member variables use 'm' prefix with CamelCase (e.g., mNbFooValues). Public members may omit, but 'm' is encouraged for clarity.
Constants (enums, global constants, static constants, and function-scope magic/literal constants) use uppercase SNAKE_CASE with 'k' prefix (e.g., kDIGIT_NUM).
Function-scope constants that are not magic numbers or literals are named like non-constant variables (e.g., bool const pass = a && b).
If macros are necessary, name them in UPPER_SNAKE_CASE (e.g., FOO_VERSION) and prefer constants over #define.
Use LLVM clang-format; wrap lines at a maximum of 120 columns; use '// clang-format off/on' sparingly with justification.
Use smart pointers for heap allocations; prefer unique_ptr for sole ownership, shared_ptr for shared...
Files:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp
**/*.{cpp,cxx,cc,cu,h,hpp,hh,hxx,cuh}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
C++ filenames should be lowerCamelCase (first letter lowercase) and must be case-insensitive unique within a compilation target.
Files:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp
**/*.{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:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpptensorrt_llm/_torch/pyexecutor/sampler.py
**/*.{h,hpp,hh,hxx,cpp,cxx,cc}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc}: Prefer anonymous namespaces over 'static' for internal linkage of functions.
All templates (class/function/member/static) must be instantiated at least once; non-POD classes should have private data members.
Files:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp
**/*.{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:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpptensorrt_llm/_torch/pyexecutor/sampler.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/pyexecutor/sampler.py
🧠 Learnings (3)
📓 Common learnings
Learnt from: eopXD
PR: NVIDIA/TensorRT-LLM#6767
File: cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp:0-0
Timestamp: 2025-08-15T06:46:54.897Z
Learning: In cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp addToken function, newly allocated blocks are unshared by design. The beam search path in addToken (when sequence.getNumTokens() > windowSize) is currently broken/non-functional with SWA, so the block allocation doesn't follow a shared-then-unshared pattern.
📚 Learning: 2025-08-15T06:46:54.897Z
Learnt from: eopXD
PR: NVIDIA/TensorRT-LLM#6767
File: cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp:0-0
Timestamp: 2025-08-15T06:46:54.897Z
Learning: In cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp addToken function, newly allocated blocks are unshared by design. The beam search path in addToken (when sequence.getNumTokens() > windowSize) is currently broken/non-functional with SWA, so the block allocation doesn't follow a shared-then-unshared pattern.
Applied to files:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpptensorrt_llm/_torch/pyexecutor/sampler.py
📚 Learning: 2025-08-14T21:04:50.248Z
Learnt from: thorjohnsen
PR: NVIDIA/TensorRT-LLM#6910
File: cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp:0-0
Timestamp: 2025-08-14T21:04:50.248Z
Learning: In KV cache onboarding logic during prefill in cpp/tensorrt_llm/batch_manager/kvCacheManager.cpp, when calculating which blocks fall within the attention window, use getTokensPerBlock() to advance token indices rather than block->getUniqueTokens().size(), because the calculation needs to consider the post-prefill state where blocks will be filled to capacity, not their current token count.
Applied to files:
cpp/tensorrt_llm/pybind/batch_manager/bindings.cppcpp/tensorrt_llm/nanobind/batch_manager/bindings.cpp
🧬 Code graph analysis (1)
tensorrt_llm/_torch/pyexecutor/sampler.py (2)
tensorrt_llm/_torch/pyexecutor/llm_request.py (1)
LlmRequest(284-426)tensorrt_llm/_torch/pyexecutor/sampler_utils.py (1)
handle_stop_single_beam(44-61)
🪛 Ruff (0.12.2)
tensorrt_llm/_torch/pyexecutor/sampler.py
351-351: PEP 484 prohibits implicit Optional
Convert to Optional[T]
(RUF013)
⏰ 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
|
PR_Github #18629 [ run ] completed with state |
Signed-off-by: Daniel Campora <961215+dcampora@users.noreply.github.com>
dcampora
force-pushed
the
user/dcampora/optimize_update_requests
branch
from
ed65c45 to
7a2ebae
Compare
dcampora
changed the title
|
/bot run |
|
PR_Github #19495 [ run ] triggered by Bot |
|
PR_Github #19495 [ run ] completed with state |
4 participants
Summary by CodeRabbit
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.
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.