[TRTLLM-1234][feat] Fixed sharding for shared embedding projections#11348
[TRTLLM-1234][feat] Fixed sharding for shared embedding projections#11348greg-kwasniewski1 wants to merge 5 commits intoNVIDIA:mainfrom
Conversation
|
/bot run --add-multi-gpu-test |
📝 WalkthroughWalkthroughThis change modifies weight-to-consumer traversal logic and shape handling within node mapping utilities. The traversal now continues through auxiliary operations, stopping only when no unvisited auxiliary nodes remain (removing an early exit condition). Shape information is explicitly set to None when unavailable, and boundary condition checks require non-None shape before performing dimension comparisons. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes 🚥 Pre-merge checks | ✅ 2 | ❌ 1❌ Failed checks (1 warning)
✅ Passed checks (2 passed)
✏️ Tip: You can configure your own custom pre-merge checks in the settings. ✨ 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
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/auto_deploy/utils/node_utils.py (1)
1066-1089:⚠️ Potential issue | 🟠 Major
filter_conditionis missing the sameNoneguard added toboundary_condition, risking aTypeError.
boundary_condition(line 1068) now correctly checksnode.meta["lin_node_shape"] is not Nonebefore indexing. However,filter_condition(line 1086) accessesnode.meta["lin_node_shape"][dim]without this guard.When a lin_op node has
lin_node_shape = None(e.g., no weight found),boundary_conditionreturnsFalse, so BFS does not stop there — the node ends up inside the subgraph. Thenfilter_conditionis applied to it and crashes withTypeError: 'NoneType' object is not subscriptable.🐛 Proposed fix — add the same None guard to filter_condition
def filter_condition(node: Node, dim: int) -> bool: if match_on_shapes: - if is_any_lin_op(node): + if is_any_lin_op(node) and node.meta["lin_node_shape"] is not None: return node.meta["lin_node_shape"][dim] == embd return False else:
|
PR_Github #35112 [ run ] triggered by Bot. Commit: |
greg-kwasniewski1
requested review from
tcherckez-nvidia
and removed request for
bmarimuthu-nv
|
PR_Github #35112 [ run ] completed with state
|
|
/bot run |
|
PR_Github #35128 [ run ] triggered by Bot. Commit: |
|
PR_Github #35128 [ run ] completed with state |
|
/bot run --add-multi-gpu-test |
|
PR_Github #35146 [ run ] triggered by Bot. Commit: |
|
PR_Github #35146 [ run ] completed with state
|
|
/bot run --add-multi-gpu-test |
|
PR_Github #35196 [ run ] triggered by Bot. Commit: |
|
PR_Github #35196 [ run ] completed with state
|
|
/bot run --add-multi-gpu-test |
|
PR_Github #35203 [ run ] triggered by Bot. Commit: |
|
PR_Github #35203 [ run ] completed with state
|
|
/bot run --add-multi-gpu-test |
|
PR_Github #35216 [ run ] triggered by Bot. Commit: |
|
PR_Github #35216 [ run ] completed with state
|
|
/bot run --reuse-test |
|
PR_Github #35234 [ run ] triggered by Bot. Commit: |
|
PR_Github #35234 [ run ] completed with state
|
|
/bot run --reuse-test |
|
PR_Github #35240 [ run ] triggered by Bot. Commit: |
|
PR_Github #35240 [ run ] completed with state |
3 participants
Fixes #11353
Some models (e.g., Qwen, Starcoder...) use the same embedding weight for both the first and the final linear layers. Weight-to-node mapping failed for these cases, and the last linear node doesn't have its weight assigned.
This PR adds the support for one-to-many weight-to-node assignment.
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
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.