[TRTLLM-10362][feat] Added Mamba and MLA layers to the sharding tests #10364
Conversation
Signed-off-by: greg-kwasniewski1 <[email protected]>
greg-kwasniewski1
requested review from
lucaslie,
mlefeb01,
tcherckez-nvidia and
zeroepoch
|
/bot run |
greg-kwasniewski1
changed the title
📝 WalkthroughWalkthroughThe PR adds test coverage for Mamba2 and MLA model layers in a tensor parallelism sharding test suite and updates the CI configuration to trigger multi-GPU tests when relevant auto_deploy source files or integration test files change. Changes
Estimated code review effort🎯 3 (Moderate) | ⏱️ ~20 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (3 passed)
✨ Finishing touches
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: 1
🧹 Nitpick comments (2)
jenkins/L0_MergeRequest.groovy (1)
722-722: Consider adding the entire auto_deploy directory to align with PR objectives.Issue #8595 aims to ensure multi-GPU tests run when diffs include source files under
tensorrt_llm/_torch/auto_deploy. Currently, only a specific file (sharding.py) is added. Given that other entries in this list use directory patterns (e.g.,"tensorrt_llm/_torch/distributed/"at line 715,"tensorrt_llm/_torch/modules/fused_moe/"at line 718), consider adding the entire directory pattern instead:🔎 Suggested change
- "tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py", + "tensorrt_llm/_torch/auto_deploy/",This would capture all changes under the auto_deploy directory, not just the sharding file, providing more comprehensive coverage aligned with the issue objective.
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py (1)
391-414: Consider extracting Mamba2 config creation to reduce duplication.The
SimpleNamespaceconfiguration forNemotronHMamba2Mixeris duplicated between_run_sharding_execution_job(lines 244-264) and_run_pattern_detection_job(lines 393-413). A helper function could reduce this duplication.🔎 Example helper function
def _create_mamba2_config(num_features: int, num_heads: int, bias: bool) -> SimpleNamespace: """Create config for NemotronHMamba2Mixer testing.""" return SimpleNamespace( hidden_size=num_features, ssm_state_size=16, mamba_num_heads=num_heads, mamba_head_dim=num_features // num_heads, n_groups=1, chunk_size=256, conv_kernel=4, use_conv_bias=bias, use_bias=bias, mamba_hidden_act="silu", layer_norm_epsilon=1e-5, time_step_limit=(0.0, float("inf")), time_step_min=0.001, time_step_max=0.1, time_step_floor=1e-4, initializer_range=0.02, rescale_prenorm_residual=False, residual_in_fp32=False, num_hidden_layers=1, )
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (2)
jenkins/L0_MergeRequest.groovytests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
🧰 Additional context used
📓 Path-based instructions (2)
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Code developed for TensorRT-LLM should conform to Python 3.8+
Indent Python code with 4 spaces. Do not use tabs
Always maintain the namespace when importing in Python, even if only one class or function from a module is used
Python files should use snake_case naming:some_file.py
Python classes should use PascalCase naming:class SomeClass
Python functions and methods should use snake_case naming:def my_awesome_function():
Python local variables should use snake_case naming:my_variable = ...
Python variable names that start with a number should be prefixed with 'k':k_99th_percentile = ...
Python global variables should use upper snake_case with prefix 'G':G_MY_GLOBAL = ...
Python constants should use upper snake_case naming:MY_CONSTANT = ...
Avoid shadowing variables declared in an outer scope in Python
Initialize all externally visible members of a Python class in the constructor
For Python interfaces that may be used outside a file, prefer docstrings over comments
Python comments should be reserved for code within a function, or interfaces that are local to a file
Use Google style docstrings in Python for classes and functions, which can be parsed by Sphinx
Python attributes and variables can be documented inline with type and description
Avoid using reflection in Python when functionality can be easily achieved without reflection
When using try-except blocks in Python, limit the except to the smallest set of errors possible
When using try-except blocks in Python to handle multiple possible variable types (duck-typing), keep the body of the try as small as possible, using the else block for logic
Files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
**/*.{cpp,h,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
All TensorRT-LLM Open Source Software code should contain an NVIDIA copyright header that includes the year of its latest meaningful modification
Files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
🧠 Learnings (10)
📓 Common learnings
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.
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.
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*").
📚 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:
jenkins/L0_MergeRequest.groovy
📚 Learning: 2025-08-06T13:58:07.506Z
Learnt from: galagam
Repo: NVIDIA/TensorRT-LLM PR: 6487
File: tests/unittest/_torch/auto_deploy/unit/singlegpu/test_ad_trtllm_bench.py:1-12
Timestamp: 2025-08-06T13:58:07.506Z
Learning: In TensorRT-LLM, test files (files under tests/ directories) do not require NVIDIA copyright headers, unlike production source code files. Test files typically start directly with imports, docstrings, or code.
Applied to files:
jenkins/L0_MergeRequest.groovy
📚 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:
jenkins/L0_MergeRequest.groovy
📚 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:
jenkins/L0_MergeRequest.groovy
📚 Learning: 2025-08-01T15:14:45.673Z
Learnt from: yibinl-nvidia
Repo: NVIDIA/TensorRT-LLM PR: 6506
File: examples/models/core/mixtral/requirements.txt:3-3
Timestamp: 2025-08-01T15:14:45.673Z
Learning: In TensorRT-LLM, examples directory can have different dependency versions than the root requirements.txt file. Version conflicts between root and examples dependencies are acceptable because examples are designed to be standalone and self-contained.
Applied to files:
jenkins/L0_MergeRequest.groovy
📚 Learning: 2025-09-29T15:14:28.503Z
Learnt from: amitz-nv
Repo: NVIDIA/TensorRT-LLM PR: 8063
File: tensorrt_llm/lora_manager.py:1080-1112
Timestamp: 2025-09-29T15:14:28.503Z
Learning: In tensorrt_llm/lora_manager.py, when calculating part_sizes for attn_qkv fused LoRA modules, the sizes are correctly multiplied by tp_size because model_config.num_heads and model_config.num_kv_heads are already divided by tp_size (per-TP-rank values), so multiplication is needed to get the original full concatenated dimension size. The interleave_fused_lora_weights_for_tp function provides proper validation with asserts for total size and TP divisibility.
Applied to files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
📚 Learning: 2025-09-29T15:14:28.503Z
Learnt from: amitz-nv
Repo: NVIDIA/TensorRT-LLM PR: 8063
File: tensorrt_llm/lora_manager.py:1080-1112
Timestamp: 2025-09-29T15:14:28.503Z
Learning: In tensorrt_llm/lora_manager.py, when calculating part_sizes for attn_qkv fused LoRA modules, the sizes are correctly multiplied by tp_size because model_config.num_heads and model_config.num_kv_heads are already divided by tp_size (per-TP-rank values), so multiplication is needed to get the original full concatenated dimension size. The interleave_fused_lora_weights_for_tp function provides proper validation.
Applied to files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
📚 Learning: 2025-10-20T17:07:18.745Z
Learnt from: nvchenghaoz
Repo: NVIDIA/TensorRT-LLM PR: 8469
File: tensorrt_llm/_torch/auto_deploy/models/patches/nemotron_h.py:98-116
Timestamp: 2025-10-20T17:07:18.745Z
Learning: In NemotronH models (tensorrt_llm/_torch/auto_deploy/models/patches/nemotron_h.py), the gate (self.gate) returns topk_indices and topk_weights that are already in the correct shape to be passed directly to torch_ops.auto_deploy.torch_moe without needing to reshape them when hidden_states is flattened.
Applied to files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
📚 Learning: 2025-10-20T16:54:09.824Z
Learnt from: nvchenghaoz
Repo: NVIDIA/TensorRT-LLM PR: 8469
File: tensorrt_llm/_torch/auto_deploy/custom_ops/rms_norm.py:6-6
Timestamp: 2025-10-20T16:54:09.824Z
Learning: In tensorrt_llm/_torch/auto_deploy/custom_ops/rms_norm.py, the import `from ...modules.mamba.layernorm_gated import _layer_norm_fwd` is correct and should not be changed to modules.fla.layernorm_gated. The _layer_norm_fwd function exists in both modules/mamba/layernorm_gated.py and modules/fla/layernorm_gated.py, but the mamba version is the intended implementation for this use case.
Applied to files:
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
🧬 Code graph analysis (1)
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py (3)
tensorrt_llm/_torch/auto_deploy/models/custom/modeling_nemotron_h.py (1)
NemotronHMamba2Mixer(58-198)tensorrt_llm/_torch/auto_deploy/custom_ops/torch_attention.py (1)
torch_attention(96-212)tensorrt_llm/_torch/auto_deploy/transform/library/sharding.py (2)
SplitDimension(88-96)WeightShardingInfo(262-322)
⏰ 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 (7)
jenkins/L0_MergeRequest.groovy (1)
744-744: LGTM!Adding the integration test file for auto_deploy to the multi-GPU trigger list is appropriate and aligns with expanding test coverage for auto_deploy workflows.
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py (6)
40-47: LGTM on the new TP plan entries.The Mamba2 and MLA projection mappings are correctly added. The "gather" designation for
q_a_projandkv_a_proj_with_mqaindicates these should not be sharded (replicated across ranks), whileq_b_projandkv_b_projuse column-wise sharding.
137-214: Well-structured MLA block implementation for sharding tests.The simplified MLA implementation correctly models the key architectural components:
- KV compression/decompression path with
kv_a_proj_with_mqaandkv_b_proj- Query path with layernorm between
q_a_projandq_b_proj- Attention using only the nope (non-positional) components is sufficient for validating sharding patterns
The comment at line 164 ("KV compression path (not sharded - gather)") aligns with the
base_model_tp_planconfiguration.
241-265: LGTM on NemotronHMamba2Mixer configuration.The configuration correctly provides all required attributes for
NemotronHMamba2Mixerinitialization, with appropriately scaled-down values for unit testing.
266-287: LGTM on MLA_Block configuration for execution tests.Using production-like parameter ratios (kv_lora_rank=256, scaled head dimensions) is appropriate for validating the execution path correctness.
316-320: Parameter accounting for replicated LayerNorm is correct.The formula
2 * kv_lora_rank * (world_size - 1) // world_sizecorrectly accounts for the additional parameters fromq_a_layernormthat are replicated (not sharded) across ranks. When base calculation divides all params by world_size, this adjustment adds back the "extra" replicated portion.
601-636: Good expansion of test coverage for Mamba2 and MLA layers.The test parametrization correctly includes
NemotronHMamba2MixerandMLA_Blockwith the expectedtorch_dist_all_reduceoperation, matching their row-wise output projections.
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
Show resolved
Hide resolved
|
PR_Github #30277 [ run ] triggered by Bot. Commit: |
tests/unittest/_torch/auto_deploy/unit/multigpu/transformations/library/test_tp_sharding.py
Outdated
Show resolved
Hide resolved
Signed-off-by: greg-kwasniewski1 <[email protected]>
Signed-off-by: greg-kwasniewski1 <[email protected]>
Signed-off-by: greg-kwasniewski1 <[email protected]>
|
/bot run |
|
PR_Github #30282 [ run ] triggered by Bot. Commit: |
|
PR_Github #30282 [ run ] completed with state
|
|
/bot run |
|
PR_Github #30295 [ run ] triggered by Bot. Commit: |
Signed-off-by: greg-kwasniewski1 <[email protected]>
|
/bot run |
|
PR_Github #30297 [ run ] triggered by Bot. Commit: |
|
PR_Github #30297 [ run ] completed with state
|
Signed-off-by: greg-kwasniewski1 <[email protected]>
|
/bot run |
|
PR_Github #30337 [ run ] triggered by Bot. Commit: |
|
PR_Github #30337 [ run ] completed with state
|
|
/bot run |
|
PR_Github #30345 [ run ] triggered by Bot. Commit: |
|
PR_Github #30345 [ run ] completed with state
|
3 participants
Fixes #10362
Fixes #8595
Summary by CodeRabbit
Release Notes
Tests
Chores
✏️ Tip: You can customize this high-level summary in your review settings.
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.