[None][fix] Fix KV cache clearing with KV Connector API #8750
Conversation
Signed-off-by: jthomson04 <[email protected]>
|
/bot run |
📝 WalkthroughWalkthroughModified request termination loop in the KV connector executor to process all finished requests instead of breaking after the first one, and updated the test suite with refactored LLM constructor logic and a new multi-request connector test. Changes
Sequence Diagram(s)sequenceDiagram
rect rgb(240, 248, 255)
note over py_executor: Before: Early Exit
py_executor->>py_executor: Process finished_request[0]
py_executor->>py_executor: break (exit loop)
note over py_executor: Remaining requests unprocessed
end
rect rgb(245, 245, 220)
note over py_executor: After: Process All
py_executor->>py_executor: Process finished_request[0]
py_executor->>py_executor: continue (next iteration)
py_executor->>py_executor: Process finished_request[1]
py_executor->>py_executor: continue (next iteration)
py_executor->>py_executor: ... (all requests processed)
end
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes
Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 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 (2)
tests/integration/defs/llmapi/test_llm_api_connector.py (2)
405-426: Good test coverage for multi-request KV cache cleanup.This test effectively validates the PR fix by exercising the scenario where multiple requests must finish in the same iteration and have their KV cache freed before a subsequent large request can proceed.
Test logic:
- First batch: 2 requests consuming ~103 tokens total
- Third request: 110 tokens, would fail without proper cleanup (max_tokens=120)
The mock setup with
side_effectechoing finished requests is a clean approach to simulate async save behavior.Consider adding explicit assertions to strengthen the regression test:
# The KV cache of both prior requests should be freed, allowing the third request to run. model.generate([2] * 110, sampling_params=sampling_params) + + # Verify both prior requests were marked as finished + assert scheduler.request_finished.call_count == 3
1-1: Consider updating copyright year to include 2025.The copyright header currently shows "2022-2024" but modifications are being made in 2025. Per coding guidelines, consider updating to "2022-2025".
-# SPDX-FileCopyrightText: Copyright (c) 2022-2024 NVIDIA CORPORATION & AFFILIATES. All rights reserved. +# SPDX-FileCopyrightText: Copyright (c) 2022-2025 NVIDIA CORPORATION & AFFILIATES. All rights reserved.
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
tensorrt_llm/_torch/pyexecutor/py_executor.py(0 hunks)tests/integration/defs/llmapi/test_llm_api_connector.py(2 hunks)
💤 Files with no reviewable changes (1)
- tensorrt_llm/_torch/pyexecutor/py_executor.py
🧰 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:
tests/integration/defs/llmapi/test_llm_api_connector.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:
tests/integration/defs/llmapi/test_llm_api_connector.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:
tests/integration/defs/llmapi/test_llm_api_connector.py
🧬 Code graph analysis (1)
tests/integration/defs/llmapi/test_llm_api_connector.py (3)
tensorrt_llm/llmapi/llm_args.py (1)
KvCacheConfig(1246-1390)tensorrt_llm/llmapi/llm.py (2)
LLM(1103-1119)generate(259-341)tensorrt_llm/_torch/pyexecutor/kv_cache_connector.py (6)
get_num_new_matched_tokens(196-210)get_num_new_matched_tokens(412-436)request_finished(213-224)request_finished(483-507)get_finished(158-174)get_finished(509-558)
🪛 Ruff (0.14.2)
tests/integration/defs/llmapi/test_llm_api_connector.py
406-406: Unused function argument: enforce_single_worker
(ARG001)
⏰ 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)
tests/integration/defs/llmapi/test_llm_api_connector.py (2)
44-54: LGTM! Clean refactoring to centralize test defaults.The refactored
model_fnnow uses adefault_kwargsdictionary merged with user-provided overrides, making the test fixture more maintainable and explicit about default parameters. The merge pattern{**default_kwargs, **kwargs}correctly gives precedence to user-provided values.
406-406: Static analysis false positive: fixture has side effects.The
enforce_single_workerparameter is a pytest fixture that sets the environment variableTLLM_WORKER_USE_SINGLE_PROCESSviamonkeypatch(defined at line 60). Fixtures with side effects don't require explicit references in the function body.
|
PR_Github #22839 [ run ] triggered by Bot. Commit: |
|
PR_Github #22839 [ run ] completed with state |
|
/bot run |
|
PR_Github #22912 [ run ] triggered by Bot. Commit: |
|
PR_Github #22912 [ run ] completed with state |
|
/bot run |
|
PR_Github #23078 [ run ] triggered by Bot. Commit: |
|
PR_Github #23078 [ run ] completed with state |
|
/bot run |
|
PR_Github #23081 [ run ] triggered by Bot. Commit: |
|
PR_Github #23081 [ run ] completed with state |
|
/bot run |
|
PR_Github #23209 [ run ] triggered by Bot. Commit: |
|
PR_Github #23209 [ run ] completed with state |
jthomson04
force-pushed
the
jthomson04/fix-connector-multi-request
branch
from
0abc0ce to
cffba0c
Compare
|
/bot run |
|
PR_Github #23327 [ run ] triggered by Bot. Commit: |
|
PR_Github #23327 [ run ] completed with state |
|
/bot run |
|
PR_Github #23409 [ run ] triggered by Bot. Commit: |
|
PR_Github #23409 [ run ] completed with state |
|
/bot run |
|
PR_Github #23411 [ run ] triggered by Bot. Commit: |
|
PR_Github #23528 [ run ] completed with state |
|
PR_Github #23532 [ kill ] triggered by Bot. Commit: |
|
PR_Github #23532 [ kill ] completed with state |
jthomson04
force-pushed
the
jthomson04/fix-connector-multi-request
branch
from
d47ef3b to
97573b8
Compare
|
/bot run |
|
PR_Github #23537 [ run ] triggered by Bot. Commit: |
|
PR_Github #23537 [ run ] completed with state |
|
/bot run |
|
PR_Github #23586 [ run ] triggered by Bot. Commit: |
|
PR_Github #23586 [ run ] completed with state |
|
/bot run |
|
PR_Github #23667 [ run ] triggered by Bot. Commit: |
|
PR_Github #23667 [ run ] completed with state |
|
/bot run |
|
PR_Github #23679 [ run ] triggered by Bot. Commit: |
|
PR_Github #23679 [ run ] completed with state |
|
/bot run |
|
PR_Github #23693 [ run ] triggered by Bot. Commit: |
|
PR_Github #23693 [ run ] completed with state |
|
/bot run |
|
PR_Github #23711 [ run ] triggered by Bot. Commit: |
|
PR_Github #23711 [ run ] completed with state |
3 participants
Fixes two separate issues:
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.
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.