[None][chore] Add placement test for ray executor #9122
Merged
Conversation
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
hchings
changed the title
Contributor
📝 WalkthroughWalkthroughThis pull request introduces RPC (Remote Procedure Call) mode support for Ray-based execution in TensorRT-LLM. A new utility function detects RPC mode via environment variable, and refactored executor and worker classes now support both traditional Ray queue-based and RPC-based submission through mixin abstractions. Changes
Sequence Diagram(s)sequenceDiagram
actor User
participant RayExecutor
participant RpcExecutorMixin
participant RpcClient
participant RayGPUWorker
participant RpcWorkerMixin
participant RPCServer
User->>RayExecutor: submit(request)
alt RPC Mode Enabled
RayExecutor->>RpcExecutorMixin: submit(request)
RpcExecutorMixin->>RpcClient: submit(request).remote(need_response=False)
RpcClient->>RPCServer: submit RPC call
RPCServer->>RayGPUWorker: process request
RayGPUWorker->>RpcWorkerMixin: (engine execution)
RpcWorkerMixin->>RPCServer: store response
RpcClient-->>RpcExecutorMixin: return GenerationResult
RpcExecutorMixin-->>RayExecutor: return GenerationResult
else Ray Queue Mode
RayExecutor->>RayGPUWorker: enqueue to Ray workers
RayGPUWorker->>RayGPUWorker: process via queue
end
RayExecutor-->>User: GenerationResult
sequenceDiagram
participant RayExecutor
participant RayGPUWorker
participant Environment
Environment->>Environment: check TLLM_RAY_USE_RPC="1"
RayExecutor->>RayExecutor: init_rpc_executor()<br/>(RPC mode)
RayExecutor->>RayGPUWorker: __init__(..., rpc_addr)
alt ray_use_rpc() == True
RayGPUWorker->>RayGPUWorker: init_rpc_worker(rank, rpc_addr)
RayGPUWorker->>RayGPUWorker: start_rpc_server() on rank 0
else ray_use_rpc() == False
RayGPUWorker->>RayGPUWorker: setup_engine()
end
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45 minutes Areas requiring extra attention:
Pre-merge checks and finishing touches❌ Failed checks (3 warnings)
✨ 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}
Contributor
There was a problem hiding this comment.
Actionable comments posted: 4
🧹 Nitpick comments (1)
tensorrt_llm/executor/rpc/rpc_client.py (1)
424-434: Consider more robust loop startup synchronization.The
time.sleep(0.1)approach works but is not deterministic. For more reliable startup detection, consider using athreading.Eventto signal when the loop is fully running.Optional enhancement:
def _ensure_event_loop(self): """Ensure we have a running event loop in a background thread.""" if self._loop is None or not self._loop.is_running(): self._loop = asyncio.new_event_loop() # ... exception handler setup ... loop_started = threading.Event() def run_loop(): asyncio.set_event_loop(self._loop) self._loop.call_soon(loop_started.set) self._loop.run_forever() self._loop_thread = threading.Thread(target=run_loop, daemon=True, name="rpc_client_loop") self._loop_thread.start() # Wait for loop to actually start (with timeout) loop_started.wait(timeout=1.0)
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (10)
tensorrt_llm/_utils.py(1 hunks)tensorrt_llm/executor/ray_executor.py(4 hunks)tensorrt_llm/executor/ray_gpu_worker.py(7 hunks)tensorrt_llm/executor/result.py(6 hunks)tensorrt_llm/executor/rpc/rpc_client.py(2 hunks)tensorrt_llm/executor/rpc_proxy.py(5 hunks)tensorrt_llm/executor/rpc_worker.py(5 hunks)tests/integration/defs/examples/test_ray.py(1 hunks)tests/integration/test_lists/test-db/l0_h100.yml(1 hunks)tests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.py(2 hunks)
🧰 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:
tensorrt_llm/_utils.pytensorrt_llm/executor/rpc/rpc_client.pytests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.pytensorrt_llm/executor/rpc_proxy.pytensorrt_llm/executor/result.pytests/integration/defs/examples/test_ray.pytensorrt_llm/executor/ray_executor.pytensorrt_llm/executor/rpc_worker.pytensorrt_llm/executor/ray_gpu_worker.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/_utils.pytensorrt_llm/executor/rpc/rpc_client.pytests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.pytensorrt_llm/executor/rpc_proxy.pytensorrt_llm/executor/result.pytests/integration/defs/examples/test_ray.pytensorrt_llm/executor/ray_executor.pytensorrt_llm/executor/rpc_worker.pytensorrt_llm/executor/ray_gpu_worker.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:
tensorrt_llm/_utils.pytensorrt_llm/executor/rpc/rpc_client.pytests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.pytensorrt_llm/executor/rpc_proxy.pytensorrt_llm/executor/result.pytests/integration/defs/examples/test_ray.pytensorrt_llm/executor/ray_executor.pytensorrt_llm/executor/rpc_worker.pytensorrt_llm/executor/ray_gpu_worker.py
🧠 Learnings (8)
📓 Common learnings
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7520
File: tensorrt_llm/_torch/pyexecutor/resource_manager.py:605-613
Timestamp: 2025-09-24T03:31:28.908Z
Learning: In TensorRT-LLM Ray orchestrator mode, ProcessGroups are initialized with both Gloo and NCCL backends (e.g., "cuda:nccl,cpu:gloo"), allowing PyTorch distributed to automatically route CPU tensors through Gloo and GPU tensors through NCCL. This eliminates the need for manual device placement when performing allreduce operations on base types.
📚 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:
tests/integration/test_lists/test-db/l0_h100.yml
📚 Learning: 2025-09-17T02:48:52.732Z
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7781
File: tests/integration/test_lists/waives.txt:313-313
Timestamp: 2025-09-17T02:48:52.732Z
Learning: In TensorRT-LLM, `tests/integration/test_lists/waives.txt` is specifically for waiving/skipping tests, while other test list files like those in `test-db/` and `qa/` directories are for different test execution contexts (pre-merge, post-merge, QA tests). The same test appearing in both waives.txt and execution list files is intentional - the test is part of test suites but will be skipped due to the waiver.
Applied to files:
tests/integration/test_lists/test-db/l0_h100.yml
📚 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:
tests/integration/test_lists/test-db/l0_h100.yml
📚 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:
tests/integration/test_lists/test-db/l0_h100.yml
📚 Learning: 2025-08-29T14:07:45.863Z
Learnt from: EmmaQiaoCh
Repo: NVIDIA/TensorRT-LLM PR: 7370
File: tests/unittest/trt/model_api/test_model_quantization.py:24-27
Timestamp: 2025-08-29T14:07:45.863Z
Learning: In TensorRT-LLM's CI infrastructure, pytest skip markers (pytest.mark.skip) are properly honored even when test files have __main__ blocks that call test functions directly. The testing system correctly skips tests without requiring modifications to the __main__ block execution pattern.
Applied to files:
tests/integration/test_lists/test-db/l0_h100.ymltests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.py
📚 Learning: 2025-09-24T03:31:28.908Z
Learnt from: tongyuantongyu
Repo: NVIDIA/TensorRT-LLM PR: 7520
File: tensorrt_llm/_torch/pyexecutor/resource_manager.py:605-613
Timestamp: 2025-09-24T03:31:28.908Z
Learning: In TensorRT-LLM Ray orchestrator mode, ProcessGroups are initialized with both Gloo and NCCL backends (e.g., "cuda:nccl,cpu:gloo"), allowing PyTorch distributed to automatically route CPU tensors through Gloo and GPU tensors through NCCL. This eliminates the need for manual device placement when performing allreduce operations on base types.
Applied to files:
tests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.pytensorrt_llm/executor/result.pytensorrt_llm/executor/ray_executor.pytensorrt_llm/executor/ray_gpu_worker.py
📚 Learning: 2025-08-19T12:45:11.997Z
Learnt from: amitz-nv
Repo: NVIDIA/TensorRT-LLM PR: 7033
File: tensorrt_llm/_torch/pyexecutor/model_engine.py:0-0
Timestamp: 2025-08-19T12:45:11.997Z
Learning: In tensorrt_llm/_torch/pyexecutor/model_engine.py, DoRA (Delta Orthogonal Rank Adaptation) functionality was removed from the PyTorch flow to eliminate issues with inverted DoRA detection logic. The original is_dora condition was checking if scaling_vec_pointer == 0, which was potentially incorrect.
Applied to files:
tensorrt_llm/executor/ray_gpu_worker.py
🧬 Code graph analysis (8)
tensorrt_llm/executor/rpc/rpc_client.py (2)
tensorrt_llm/llmapi/utils.py (3)
loop(492-493)get(415-445)get(498-515)tensorrt_llm/logger.py (1)
debug(144-145)
tests/unittest/_torch/ray_orchestrator/multi_gpu/test_executor.py (2)
tensorrt_llm/llmapi/llm_args.py (1)
KvCacheConfig(1261-1405)tensorrt_llm/llmapi/llm.py (1)
_collective_rpc(1014-1039)
tensorrt_llm/executor/rpc_proxy.py (8)
tensorrt_llm/executor/rpc/rpc_common.py (1)
get_unique_ipc_addr(9-16)tensorrt_llm/executor/rpc/rpc_client.py (3)
RPCClient(78-578)remote(50-55)remote_streaming(71-75)tensorrt_llm/executor/ray_executor.py (3)
start(229-230)shutdown(250-317)submit(192-227)tensorrt_llm/executor/rpc_worker.py (5)
start(222-223)shutdown(279-284)submit(60-65)RpcWorker(169-291)main_task(226-277)tensorrt_llm/executor/result.py (4)
register(178-180)register(223-226)GenerationResult(773-948)result(877-888)tensorrt_llm/executor/executor.py (5)
shutdown(298-299)submit(110-111)_get_next_client_id(218-221)_get_logprob_params(223-239)_handle_background_error(255-292)tensorrt_llm/executor/proxy.py (2)
shutdown(368-415)submit(417-442)tensorrt_llm/executor/postproc_worker.py (1)
PostprocWorkerConfig(42-49)
tensorrt_llm/executor/result.py (3)
tensorrt_llm/_utils.py (2)
mpi_disabled(522-524)ray_use_rpc(527-531)tensorrt_llm/executor/ray_executor.py (1)
use_ray_queue(241-242)tensorrt_llm/executor/executor.py (1)
use_ray_queue(106-107)
tests/integration/defs/examples/test_ray.py (1)
tests/integration/defs/conftest.py (1)
llm_venv(702-719)
tensorrt_llm/executor/ray_executor.py (6)
tensorrt_llm/_utils.py (3)
get_free_port(476-479)nvtx_range_debug(901-925)ray_use_rpc(527-531)tensorrt_llm/llmapi/utils.py (1)
logger_debug(105-118)tensorrt_llm/executor/ray_gpu_worker.py (7)
RayGPUWorker(158-305)RayWorkerWrapper(42-155)submit(77-78)start(244-245)abort_request(85-86)shutdown(106-107)shutdown(247-285)tensorrt_llm/executor/result.py (8)
result(877-888)GenerationResult(773-948)RayAsyncQueue(170-206)RaySyncQueue(214-251)warmup(189-193)warmup(235-239)request_id(822-823)get(245-251)tensorrt_llm/executor/rpc_proxy.py (8)
RpcExecutorMixin(23-274)init_rpc_executor(35-43)setup_engine_remote(348-349)setup_mainloop(45-84)_fetch_responses_loop_async(185-189)submit(86-104)abort_request(355-356)shutdown(358-393)tensorrt_llm/executor/rpc/rpc_client.py (2)
remote(50-55)close(126-176)
tensorrt_llm/executor/rpc_worker.py (3)
tensorrt_llm/executor/base_worker.py (7)
set_result_queue(281-284)start(331-333)fetch_stats(266-273)fetch_kv_cache_events(275-279)_stats_serializer(617-629)_kv_cache_events_serializer(633-635)shutdown(605-613)tensorrt_llm/executor/rpc/rpc_server.py (4)
RPCServer(19-517)bind(84-97)start(488-517)shutdown(99-158)tensorrt_llm/_utils.py (1)
mpi_rank(534-541)
tensorrt_llm/executor/ray_gpu_worker.py (3)
tensorrt_llm/_torch/utils.py (1)
get_device_uuid(322-327)tensorrt_llm/_utils.py (1)
ray_use_rpc(527-531)tensorrt_llm/executor/rpc_worker.py (5)
RpcWorkerMixin(25-166)shutdown(279-284)init_rpc_worker(40-51)start_rpc_server(53-58)start(222-223)
🪛 Ruff (0.14.4)
tensorrt_llm/executor/rpc_proxy.py
326-326: f-string without any placeholders
Remove extraneous f prefix
(F541)
tensorrt_llm/executor/ray_executor.py
266-266: Do not catch blind exception: Exception
(BLE001)
286-286: Do not catch blind exception: Exception
(BLE001)
292-292: Do not catch blind exception: Exception
(BLE001)
388-388: Avoid equality comparisons to False; use not ret: for false checks
Replace with not ret
(E712)
tensorrt_llm/executor/rpc_worker.py
42-43: Avoid specifying long messages outside the exception class
(TRY003)
114-114: Unused method argument: timeout
(ARG002)
119-119: Unused method argument: timeout
(ARG002)
tensorrt_llm/executor/ray_gpu_worker.py
194-195: Avoid specifying long messages outside the exception class
(TRY003)
263-263: Do not catch blind exception: Exception
(BLE001)
⏰ 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 (1)
tensorrt_llm/executor/rpc/rpc_client.py (1)
4-4: LGTM: Global import improves code clarity.Moving the
timeimport to module level follows Python best practices and makes dependencies explicit.
Collaborator
Author
|
Only top commit ( |
hchings
changed the title
hchings
changed the title
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #24536 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #24536 [ run ] completed with state |
Signed-off-by: Erin Ho <[email protected]>
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #24635 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #24635 [ run ] completed with state |
zheyuf
pushed a commit
to zheyuf/TensorRT-LLM
that referenced
this pull request
Nov 19, 2025
Signed-off-by: Erin Ho <[email protected]>
greg-kwasniewski1
pushed a commit
to nv-auto-deploy/TensorRT-LLM
that referenced
this pull request
Nov 20, 2025
Signed-off-by: Erin Ho <[email protected]>
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
3 participants
Add this suggestion to a batch that can be applied as a single commit.
This suggestion is invalid because no changes were made to the code.
Suggestions cannot be applied while the pull request is closed.
Suggestions cannot be applied while viewing a subset of changes.
Only one suggestion per line can be applied in a batch.
Add this suggestion to a batch that can be applied as a single commit.
Applying suggestions on deleted lines is not supported.
You must change the existing code in this line in order to create a valid suggestion.
Outdated suggestions cannot be applied.
This suggestion has been applied or marked resolved.
Suggestions cannot be applied from pending reviews.
Suggestions cannot be applied on multi-line comments.
Suggestions cannot be applied while the pull request is queued to merge.
Suggestion cannot be applied right now. Please check back later.
Summary by CodeRabbit
Release Notes
New Features
TLLM_RAY_USE_RPCenvironment variable.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
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.