[None][Feat] WIP DO NOT REVIEW Nemotron moe fp8 - #8598
Closed
nvchenghaoz wants to merge 16 commits into
Closed
Conversation
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: junq <22017000+QiJune@users.noreply.github.com>
Signed-off-by: Lucas Liebenwein <11156568+lucaslie@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
… --> dtype Signed-off-by: Lucas Liebenwein <11156568+lucaslie@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Signed-off-by: Chenghao Zhang <211069071+nvchenghaoz@users.noreply.github.com>
Contributor
|
Caution Review failedThe pull request is closed. 📝 WalkthroughWalkthroughThis pull request introduces fused Mixture-of-Experts (MoE) infrastructure with CUDA and Triton kernels, adds gated RMSNorm support, refactors attention pattern matching, improves KV-cache tracking, and simplifies executor APIs by removing lora_config and kv_connector_config from public signatures. Changes
Sequence Diagram(s)sequenceDiagram
participant App as Application
participant Executor as Executor API
participant Worker as Worker
participant LLMArgs as LLMArgs
rect rgb(200, 220, 255)
Note over Executor,Worker: Old API (Removed)
App->>Executor: create(..., lora_config, kv_connector_config)
Executor->>Worker: init(..., lora_config, kv_connector_config)
Worker->>Worker: store lora_config, kv_connector_config
end
rect rgb(200, 255, 200)
Note over Executor,LLMArgs: New API (Updated)
App->>Executor: create(..., llm_args)
Executor->>Worker: init(..., llm_args)
Worker->>LLMArgs: extract lora_config, kv_connector_config from llm_args
LLMArgs->>Worker: return configs from llm_args
end
sequenceDiagram
participant Input as Token Input
participant Router as Router
participant MoEAlign as MoE Align Kernel
participant Triton as Triton Fused MoE
participant Output as Output
Input->>Router: topk_ids, routing_weights
Router->>MoEAlign: topk_ids, num_experts, block_size
MoEAlign->>MoEAlign: compute per-expert counts
MoEAlign->>MoEAlign: prefix-sum for token placement
MoEAlign-->>Triton: sorted_token_ids, expert_ids, num_tokens_post_pad
Triton->>Triton: select path (mlp_style: gated_mlp vs mlp)
alt gated_mlp: W1, W3, W2 with gate
Triton->>Triton: act(W1*x) * W3*x
Triton->>Triton: W2 * (gated output)
else mlp: W_up, W_down with activation
Triton->>Triton: act(W_up * x)
Triton->>Triton: W_down * (activated)
end
Triton->>Output: fused MoE output
Estimated code review effort🎯 4 (Complex) | ⏱️ ~60 minutes The diff introduces substantial new infrastructure (CUDA kernel, Triton kernels, gated RMSNorm) requiring careful verification of numerical correctness and performance assumptions. The MoE alignment logic, Triton fused operations, and attention pattern splitting are logic-dense. Executor API simplification is repetitive across many files but must be verified for consistency. Cross-cutting concerns (KV cache tracking, chunked prefill wiring) and multiple interconnected features (configuration handling, test parameterization) increase cognitive load despite some pattern repetition. Possibly related PRs
Suggested labels
Suggested reviewers
✨ Finishing touches
🧪 Generate unit tests (beta)
📜 Recent review detailsConfiguration used: Path: .coderabbit.yaml Review profile: CHILL Plan: Pro 📒 Files selected for processing (42)
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 |
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
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
New Features
Improvements
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.
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.