[TRTLLM-9302][chore] Move build config from BaseLlmArgs to TrtLlmArgs #9249
+206
−218
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
Signed-off-by: junq <[email protected]>
Signed-off-by: junq <[email protected]>
Collaborator
Author
|
/bot run |
Contributor
📝 WalkthroughWalkthroughThe pull request refactors and extends Changes
Sequence DiagramsequenceDiagram
participant Client
participant BaseLlmArgs
participant TrtLlmArgs
participant Validators
participant BuildConfig
Client->>TrtLlmArgs: Create instance with args
TrtLlmArgs->>Validators: validate_model_format_misc()
Validators->>Validators: Set model_format=HF (default)
Validators->>Validators: Check speculative_config type
alt Speculative Config Present
Validators->>Validators: Map config type to decoding_mode
Validators->>BuildConfig: Sync decoding settings
Validators->>BuildConfig: Adjust max_draft_len if needed
end
TrtLlmArgs->>Validators: validate_lora_config_consistency()
Validators->>Validators: Check lora_dir && lora_config
TrtLlmArgs->>Validators: validate_peft_cache_config()
Validators->>Validators: Reject lora_prefetch_dir if PEFT enabled
TrtLlmArgs->>Validators: validate_runtime_args()
Validators->>Validators: Check max_batch_size vs max_num_tokens
TrtLlmArgs->>TrtLlmArgs: init_build_config()
TrtLlmArgs->>BuildConfig: Auto-create if missing
TrtLlmArgs->>BuildConfig: Incorporate runtime knobs
Client->>TrtLlmArgs: get_runtime_sizes()
TrtLlmArgs-->>Client: Return (max_beam_width, max_num_tokens, max_seq_len, max_batch_size)
Estimated code review effort🎯 4 (Complex) | ⏱️ ~45–75 minutes
Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ Finishing touches
🧪 Generate unit tests (beta)
Tip 📝 Customizable high-level summaries are now available in beta!You can now customize how CodeRabbit generates the high-level summary in your pull requests — including its content, structure, tone, and formatting.
Example instruction:
Note: This feature is currently in beta for Pro-tier users, and pricing will be announced later. 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: 1
🧹 Nitpick comments (2)
tensorrt_llm/llmapi/llm_args.py (2)
1994-2021: Clarify the LoRA config validation logic.The validator handles multiple scenarios but the interaction between
enable_loraandlora_configcould be clearer:
- Lines 2003-2008: You warn when
enable_lorais set alongsidelora_configfor PyTorch backends, suggestingenable_lorais ignored. However, this check only applies when both are truthy.- Lines 2010-2020: Auto-populating
lora_target_moduleswhen bothlora_dirandlora_target_modulesare empty is helpful, but consider whether this default should apply in all cases or only when LoRA is explicitly enabled.Consider adding a comment explaining the precedence rules:
+ # Precedence: lora_config > enable_lora if self.enable_lora and self.lora_config is not None and self.backend in [ 'pytorch', '_autodeploy' ]:
2115-2153: Document the overwriting behavior.The validator silently overwrites
max_batch_sizeandmax_num_tokensif they exceedbuild_configvalues (lines 2125-2136), which could be surprising to users. While warnings are logged, consider whether these should fail fast instead.Consider making this behavior more explicit in the field documentation or raising an error instead of silently capping:
if self.max_batch_size is not None: if self.max_batch_size > self.build_config.max_batch_size: + # Option 1: Raise an error + raise ValueError( + f"max_batch_size [{self.max_batch_size}] exceeds build_config.max_batch_size [{self.build_config.max_batch_size}]" + ) + # Option 2: Keep current behavior but improve logging self.max_batch_size = self.build_config.max_batch_size logger.warning( f"max_batch_size [{self.max_batch_size}] is overridden by build_config.max_batch_size [{self.build_config.max_batch_size}] in build_config" )
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tensorrt_llm/llmapi/llm_args.py(4 hunks)
🧰 Additional context used
🧠 Learnings (11)
📓 Common learnings
Learnt from: venkywonka
Repo: NVIDIA/TensorRT-LLM PR: 6029
File: .github/pull_request_template.md:45-53
Timestamp: 2025-08-27T17:50:13.264Z
Learning: For PR templates in TensorRT-LLM, avoid suggesting changes that would increase developer overhead, such as converting plain bullets to mandatory checkboxes. The team prefers guidance-style bullets that don't require explicit interaction to reduce friction in the PR creation process.
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.
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.
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which can contain default `cuda_graph_config` values, so `llm_args` may already have this config before the extra options processing.
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.
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: shaharmor98
Repo: NVIDIA/TensorRT-LLM PR: 7231
File: tensorrt_llm/_torch/pyexecutor/_util.py:504-509
Timestamp: 2025-08-26T06:07:02.166Z
Learning: In tensorrt_llm/_torch/pyexecutor/_util.py, when calling model_engine.set_lora_model_config(), pass model_binding_config.mlp_hidden_size directly without multiplying by mapping.tp_size, as the mlp_hidden_size from get_bindings_model_config() is already the per-TP rank value needed for LoRA weight packaging.
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.
📚 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:
tensorrt_llm/llmapi/llm_args.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 with asserts for total size and TP divisibility.
Applied to files:
tensorrt_llm/llmapi/llm_args.py
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which can contain default `cuda_graph_config` values, so `llm_args` may already have this config before the extra options processing.
Applied to files:
tensorrt_llm/llmapi/llm_args.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:
tensorrt_llm/llmapi/llm_args.py
📚 Learning: 2025-08-14T15:38:01.771Z
Learnt from: MatthiasKohl
Repo: NVIDIA/TensorRT-LLM PR: 6904
File: cpp/tensorrt_llm/pybind/thop/bindings.cpp:55-57
Timestamp: 2025-08-14T15:38:01.771Z
Learning: In TensorRT-LLM Python bindings, tensor parameter collections like mla_tensor_params and spec_decoding_tensor_params are kept as required parameters without defaults to maintain API consistency, even when it might affect backward compatibility.
Applied to files:
tensorrt_llm/llmapi/llm_args.py
📚 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:
tensorrt_llm/llmapi/llm_args.py
📚 Learning: 2025-08-26T06:07:02.166Z
Learnt from: shaharmor98
Repo: NVIDIA/TensorRT-LLM PR: 7231
File: tensorrt_llm/_torch/pyexecutor/_util.py:504-509
Timestamp: 2025-08-26T06:07:02.166Z
Learning: In tensorrt_llm/_torch/pyexecutor/_util.py, when calling model_engine.set_lora_model_config(), pass model_binding_config.mlp_hidden_size directly without multiplying by mapping.tp_size, as the mlp_hidden_size from get_bindings_model_config() is already the per-TP rank value needed for LoRA weight packaging.
Applied to files:
tensorrt_llm/llmapi/llm_args.py
📚 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:
tensorrt_llm/llmapi/llm_args.py
📚 Learning: 2025-08-21T00:16:56.457Z
Learnt from: farshadghodsian
Repo: NVIDIA/TensorRT-LLM PR: 7101
File: docs/source/blogs/tech_blog/blog9_Deploying_GPT_OSS_on_TRTLLM.md:36-36
Timestamp: 2025-08-21T00:16:56.457Z
Learning: TensorRT-LLM container release tags in documentation should only reference published NGC container images. The README badge version may be ahead of the actual published container versions.
Applied to files:
tensorrt_llm/llmapi/llm_args.py
📚 Learning: 2025-08-26T09:37:10.463Z
Learnt from: jiaganc
Repo: NVIDIA/TensorRT-LLM PR: 7031
File: tensorrt_llm/bench/dataclasses/configuration.py:90-104
Timestamp: 2025-08-26T09:37:10.463Z
Learning: In TensorRT-LLM's bench configuration, the `get_pytorch_perf_config()` method returns `self.pytorch_config` which is a Dict[str, Any] that can contain default values including `cuda_graph_config`, making the fallback `llm_args["cuda_graph_config"]` safe to use.
Applied to files:
tensorrt_llm/llmapi/llm_args.py
🧬 Code graph analysis (1)
tensorrt_llm/llmapi/llm_args.py (5)
tensorrt_llm/logger.py (1)
warning(132-133)tensorrt_llm/lora_helper.py (1)
get_default_trtllm_modules_to_hf_modules(42-58)tensorrt_llm/builder.py (2)
default(45-50)BuildConfig(453-611)tensorrt_llm/llmapi/llm_utils.py (1)
model_format(286-287)tensorrt_llm/llmapi/utils.py (1)
get_type_repr(662-667)
🪛 Ruff (0.14.5)
tensorrt_llm/llmapi/llm_args.py
2026-2028: Avoid specifying long messages outside the exception class
(TRY003)
2698-2700: Avoid specifying long messages outside the exception class
(TRY003)
2729-2731: Avoid specifying long messages outside the exception class
(TRY003)
⏰ 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 (8)
tensorrt_llm/llmapi/llm_args.py (8)
1986-1992: LGTM - Runtime consistency check looks good.The validator appropriately warns when
max_batch_sizeexceedsmax_num_tokens, which helps catch potential configuration issues without blocking execution.
2023-2029: LGTM - Appropriate validation for unsupported feature.The validator correctly prevents the use of
lora_prefetch_dirwhen PEFT cache is configured, as this combination is not currently supported.
2040-2093: LGTM - Well-structured new TrtLlmArgs class.The new
TrtLlmArgsclass properly extendsBaseLlmArgswith TensorRT-specific configuration options. Field definitions are well-documented and have sensible defaults.
2094-2113: LGTM - Sensible default BuildConfig creation.The validator appropriately creates a default
BuildConfigwhen none is provided, using runtime parameters as defaults. This ensuresbuild_configis always initialized.
2219-2227: LGTM - Eagle config instantiation looks correct.The Eagle configuration is properly constructed with all required parameters from
speculative_config.
2306-2343: LGTM - Proper model format detection and configuration loading.The validator correctly:
- Detects model format (TLLM_ENGINE, TLLM_CKPT, or HF)
- Loads appropriate configurations from engine or checkpoint directories
- Warns when
build_configis provided but ignored for TLLM_ENGINE format- Fills KV cache config from runtime defaults when available
2688-2692: LGTM - Appropriate model format for PyTorch backend.Setting
_model_formattoHFis correct for the PyTorch backend, which works with HuggingFace models.
2694-2744: LGTM - Comprehensive speculative config validation.The validator properly handles all speculative decoding config types with appropriate assertions and configuration adjustments:
- Eagle, NGram, DraftTarget: Validates required fields
- MTP: Sets
max_draft_lenfromnum_nextn_predict_layers- SaveHiddenStates: Appropriately constrains batch size and disables incompatible features
- Extracts and validates speculative model directory and format
The logic is thorough and handles all supported speculation types correctly.
Collaborator
|
PR_Github #24855 [ run ] triggered by Bot. Commit: |
Signed-off-by: junq <[email protected]>
Collaborator
Author
|
/bot run |
Collaborator
|
PR_Github #24860 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #24855 [ run ] completed with state |
Signed-off-by: junq <[email protected]>
Collaborator
Author
|
/bot run --disable-fail-fast |
Collaborator
|
PR_Github #24883 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #24860 [ run ] completed with state |
Signed-off-by: junq <[email protected]>
Collaborator
Author
|
/bot run --stage-list "H100_PCIe-PyTorch-1,H100_PCIe-PyTorch-2,H100_PCIe-PyTorch-3,H100_PCIe-PyTorch-4" |
Collaborator
|
PR_Github #24906 [ run ] triggered by Bot. Commit: |
Collaborator
|
PR_Github #24883 [ run ] completed with state |
Collaborator
|
PR_Github #24906 [ run ] completed with state |
Collaborator
Author
|
/bot run --disable-fail-fast |
Collaborator
|
PR_Github #24956 [ run ] triggered by Bot. Commit: |
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
2 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
Bug Fixes
Refactor
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.
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.