[None][feat] Support EPLB in Qwen3 MoE #7443
Conversation
Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com>
lucifer1004
requested review from
byshiue,
jinyangyuan-nvidia and
symphonylyh
and removed request for
byshiue and
symphonylyh
lucifer1004
changed the title
📝 WalkthroughWalkthroughAdds a second CUDA stream for MoE balancing in Qwen3MoE by extending aux_stream_dict passed to create_moe: retains AuxStreamType.MoeChunkingOverlap with the provided aux_stream and adds AuxStreamType.MoeBalancer with a new torch.cuda.Stream(). No public APIs or signatures changed. Changes
Sequence Diagram(s)sequenceDiagram
actor User
participant Qwen3MoE as Qwen3MoE
participant MoE as create_moe
participant Stream1 as CUDA Stream: ChunkingOverlap
participant Stream2 as CUDA Stream: Balancer
User->>Qwen3MoE: instantiate(model, aux_stream)
activate Qwen3MoE
Qwen3MoE->>Stream1: use provided aux_stream
Qwen3MoE->>Stream2: create new torch.cuda.Stream()
Qwen3MoE->>MoE: create_moe(aux_stream_dict={ChunkingOverlap: Stream1,<br/>Balancer: Stream2})
deactivate Qwen3MoE
note over MoE,Stream2: MoE may schedule balancing on Stream2 while chunking overlap uses Stream1
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Suggested labels
Suggested reviewers
✨ 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. 🪧 TipsChatThere are 3 ways to chat with CodeRabbit:
SupportNeed help? Create a ticket on our support page for assistance with any issues or questions. CodeRabbit Commands (Invoked using PR/Issue comments)Type Other keywords and placeholders
Status, Documentation and Community
|
![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/models/modeling_qwen3_moe.py (1)
1-1: Add NVIDIA 2025 copyright header (and SPDX) per coding guidelines.This file is missing the required header.
+# Copyright (c) 2025, NVIDIA CORPORATION. All rights reserved. +# SPDX-License-Identifier: Apache-2.0 + import os
🧹 Nitpick comments (1)
tensorrt_llm/_torch/models/modeling_qwen3_moe.py (1)
111-114: Consider reusing a per-model balancer stream instead of per-layer.You already share aux_stream from Qwen3MoEModel across layers; doing the same for MoeBalancer avoids N_streams = N_layers and reduces overhead without sacrificing concurrency.
Example (outside this hunk):
# In Qwen3MoEModel.__init__ self.aux_stream = torch.cuda.Stream() self.moe_balancer_stream = torch.cuda.Stream(device=self.aux_stream.device) # In Qwen3MoEDecoderLayer.__init__ self.mlp = Qwen3MoE(model_config, aux_stream, balancer_stream=model_config_parent.moe_balancer_stream, layer_idx=layer_idx) # In Qwen3MoE.__init__ signature def __init__(..., aux_stream: torch.cuda.Stream, balancer_stream: Optional[torch.cuda.Stream] = None, ...): ... aux_stream_dict = { AuxStreamType.MoeChunkingOverlap: aux_stream, AuxStreamType.MoeBalancer: balancer_stream or torch.cuda.Stream(device=aux_stream.device), }
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
💡 Knowledge Base configuration:
- MCP integration is disabled by default for public repositories
- Jira integration is disabled by default for public repositories
- Linear integration is disabled by default for public repositories
You can enable these sources in your CodeRabbit configuration.
📒 Files selected for processing (1)
tensorrt_llm/_torch/models/modeling_qwen3_moe.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (2)
**/*.{cpp,cc,cxx,cu,h,hpp,hh,hxx,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.{cpp,cc,cxx,cu,h,hpp,hh,hxx,cuh,py}: Use spaces only; no tabs; indent with 4 spaces
Prepend NVIDIA copyright header (current year) to all source files (.cpp, .h, .cu, .py, etc.)
Files:
tensorrt_llm/_torch/models/modeling_qwen3_moe.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+
Indent Python with 4 spaces; no tabs
Preserve module namespaces when importing: from package.subpackage import foo; then call foo.SomeClass() instead of importing the class directly
Python naming: files snake_case; classes PascalCase; functions/methods snake_case; locals snake_case (prefix k_ when starting with a number); globals UPPER_SNAKE_CASE with G_ prefix; constants UPPER_SNAKE_CASE
Avoid shadowing outer-scope variables; initialize all externally visible members in init
Prefer docstrings for interfaces used outside a file; limit comments to function-internal or file-local interfaces
Use Google-style docstrings for classes and functions; document attributes/variables inline so Sphinx can render them
Avoid reflection when simpler alternatives exist; prefer explicit parameters and return dicts over locals()/dynamic tricks
In try/except, catch the narrowest exceptions possible; keep try bodies minimal and use else for the main logic when doing duck-typing checks
Files:
tensorrt_llm/_torch/models/modeling_qwen3_moe.py
⏰ 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). (2)
- GitHub Check: Check PR Checklist Resolution
- GitHub Check: Pre-commit Check
🔇 Additional comments (1)
tensorrt_llm/_torch/models/modeling_qwen3_moe.py (1)
111-114: AuxStreamType.MoeBalancer is defined and create_moe correctly forwards aux_stream_dict to all backends (fused_moe_wide_ep consumes the MoeBalancer stream as intended).
Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com> Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com>
Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com>
|
/bot run |
|
PR_Github #18682 [ run ] triggered by Bot |
|
PR_Github #18682 [ run ] completed with state |
|
/bot run |
|
PR_Github #19253 [ run ] triggered by Bot |
|
PR_Github #19253 [ run ] completed with state |
Head branch was pushed to by a user without write access
|
/bot run |
|
PR_Github #19274 [ run ] triggered by Bot |
|
PR_Github #19274 [ run ] completed with state |
|
/bot run |
|
PR_Github #19276 [ run ] triggered by Bot |
|
PR_Github #19276 [ run ] completed with state |
Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
Signed-off-by: Gabriel Wu <13583761+lucifer1004@users.noreply.github.com> Co-authored-by: coderabbitai[bot] <136622811+coderabbitai[bot]@users.noreply.github.com>
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.