[None][fix] Add refresh blocks to Python executor

[Tabrizian]

Summary by CodeRabbit

• Improvements
  • Enhanced KV cache block synchronization during token generation for more reliable memory resource management.

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 the stage-list parameter 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.md
and the scripts/test_to_stage_mapping.py helper.

kill

kill

Kill all running builds associated with pull request.

skip

skip --comment COMMENT

Skip 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-pipeline

Reuse 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.

[coderabbitai]

📝 Walkthrough

Walkthrough

The changes expose a refresh_blocks() method to the Python API for BaseKVCacheManager through both nanobind and pybind bindings, mapping to the existing C++ refreshBlocks() function. The method is subsequently invoked in the resource manager's batch processing workflow to refresh KV cache blocks after updating cache state.

Changes

Cohort / File(s)	Summary
Python binding additions cpp/tensorrt_llm/nanobind/batch_manager/kvCacheManager.cpp, cpp/tensorrt_llm/pybind/batch_manager/kvCacheManager.cpp	Added refresh_blocks() method binding for BaseKVCacheManager, exposing the underlying C++ refreshBlocks() function through both nanobind and pybind interfaces. Nanobind version includes gil_scoped_release call guard.
Resource manager integration tensorrt_llm/_torch/pyexecutor/resource_manager.py	Inserted call to self.impl.refresh_blocks() in KVCacheManager.prepare_resources() following KV cache state updates via KvCacheConnectorManager, introducing a synchronization step before processing generation batches.

Estimated code review effort

🎯 1 (Trivial) | ⏱️ ~5 minutes

• Binding files: Verify nanobind and pybind bindings follow existing patterns and correctly expose the refreshBlocks() method with appropriate call guards.
• Resource manager integration: Confirm the placement and timing of the refresh_blocks() call is appropriate within the batch processing workflow.

Pre-merge checks and finishing touches

❌ Failed checks (1 warning)

Check name	Status	Explanation	Resolution
Description check	⚠️ Warning	The PR description is empty; it contains only the repository template structure without any explanation of the changes, rationale, or test coverage.	Fill in the Description section explaining what the refresh_blocks functionality does and why it's needed. Add Test Coverage section detailing which tests validate these changes.

✅ Passed checks (1 passed)

Check name	Status	Explanation
Title check	✅ Passed	The title clearly describes the main change: adding refresh_blocks functionality to the Python executor for KV cache management.

✨ Finishing touches

• 📝 Generate docstrings

🧪 Generate unit tests (beta)

• Create PR with unit tests
• Post copyable unit tests in a comment

---

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.

❤️ Share

• X
• Mastodon
• Reddit
• LinkedIn

_{Comment @coderabbitai help to get the list of available commands and usage tips.}

[Tabrizian]

/bot run --disable-fail-fast

[tensorrt-cicd]

PR_Github #23421 [ run ] triggered by Bot. Commit: edab75a

[thorjohnsen]

LGTM. This is a critical bug fix. Without this fix, block offloading, onboarding and partial reuse will not work correctly. C++ executor already does this.

[tensorrt-cicd]

PR_Github #23421 [ run ] completed with state SUCCESS. Commit: edab75a
/LLM/main/L0_MergeRequest_PR pipeline #17637 completed with status: 'FAILURE'

[Tabrizian]

/bot run --disable-fail-fast

[dcampora]

LGTM

[tensorrt-cicd]

PR_Github #23477 [ run ] triggered by Bot. Commit: edab75a

[tensorrt-cicd]

PR_Github #23477 [ run ] completed with state SUCCESS. Commit: edab75a
/LLM/main/L0_MergeRequest_PR pipeline #17673 completed with status: 'FAILURE'

[Tabrizian]

/bot run

[tensorrt-cicd]

PR_Github #23530 [ run ] triggered by Bot. Commit: edab75a

[tensorrt-cicd]

PR_Github #23530 [ run ] completed with state SUCCESS. Commit: edab75a
/LLM/main/L0_MergeRequest_PR pipeline #17710 completed with status: 'FAILURE'

[Tabrizian]

/bot run --disable-fail-fast

[eopXD]

Should the refresh happen after processing the generation batch?

[eopXD]

(Sorry this is not an aggressive ping. Repeating the above comment since I cannot have an empty "Request to change" action)

Should the refresh happen after processing the generation batch?