[None][fix] Fix the error where checkpoint_dir is assigned as NONE wh… #8198
Conversation
…en model_format is TLLM_ENGINE, causing an infinite loop of URL requests
📝 WalkthroughWalkthroughIntroduces a guard in from_pretrained to handle a None checkpoint_dir: skips AutoConfig.from_pretrained, sets pretrained_config to None, model_dir to an empty path, and logs a warning; otherwise, retains existing behavior. Changes
Sequence Diagram(s)sequenceDiagram
autonumber
participant Caller
participant ModelConfig as ModelConfig.from_pretrained
participant HF as transformers.AutoConfig
Caller->>ModelConfig: from_pretrained(checkpoint_dir)
alt checkpoint_dir is not None
ModelConfig->>HF: AutoConfig.from_pretrained(checkpoint_dir)
HF-->>ModelConfig: pretrained_config
ModelConfig-->>Caller: model_config (with pretrained_config)
else checkpoint_dir is None
Note over ModelConfig: Log warning about missing checkpoint_dir
ModelConfig-->>Caller: model_config (pretrained_config=None, model_dir="")
end
Estimated code review effort🎯 2 (Simple) | ⏱️ ~10 minutes Pre-merge checks and finishing touches❌ Failed checks (2 warnings)
✅ Passed checks (1 passed)
✨ 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: 1
🧹 Nitpick comments (1)
tensorrt_llm/_torch/model_config.py (1)
409-413: Improve warning message andmodel_dirinitialization.If you decide to allow
Nonecheckpoint_dir instead of raising an exception:
Path("")is semantically unclear - consider usingPath()or a more explicit placeholder likePath("/invalid")with a comment explaining its purpose.- The warning message should be more actionable - explain the context (e.g., "When using TLLM_ENGINE format, checkpoint_dir may be None") and clarify the impact.
Example improvement:
else: pretrained_config = None - model_dir = Path("") + model_dir = Path() # Empty path for TLLM_ENGINE format where HF config is not needed logger.warning( - "checkpoint_dir is None, please check your checkpoint directory path.") + "checkpoint_dir is None (expected for TLLM_ENGINE format). " + "Skipping HuggingFace config loading. Ensure this ModelConfig is only used " + "in contexts that don't require pretrained_config attributes.")
📜 Review details
Configuration used: Path: .coderabbit.yaml
Review profile: CHILL
Plan: Pro
📒 Files selected for processing (1)
tensorrt_llm/_torch/model_config.py(1 hunks)
🧰 Additional context used
📓 Path-based instructions (3)
**/*.{h,hpp,hh,hxx,cpp,cxx,cc,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Use only spaces, no tabs; indent with 4 spaces.
Files:
tensorrt_llm/_torch/model_config.py
**/*.py
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
**/*.py: Python code must target Python 3.8+.
Indent Python code with 4 spaces; do not use tabs.
Maintain module namespace when importing; prefer 'from package.subpackage import foo' then 'foo.SomeClass()' instead of importing the class directly.
Python filenames should be snake_case (e.g., some_file.py).
Python classes use PascalCase names.
Functions and methods use snake_case names.
Local variables use snake_case; prefix 'k' for variables that start with a number (e.g., k_99th_percentile).
Global variables use upper SNAKE_CASE prefixed with 'G' (e.g., G_MY_GLOBAL).
Constants use upper SNAKE_CASE (e.g., MY_CONSTANT).
Avoid shadowing variables from an outer scope.
Initialize all externally visible members of a class in the constructor.
Prefer docstrings for interfaces that may be used outside a file; comments for in-function or file-local interfaces.
Use Google-style docstrings for classes and functions (Sphinx-parsable).
Document attributes and variables inline so they render under the class/function docstring.
Avoid reflection when a simpler, explicit approach suffices (e.g., avoid dict(**locals()) patterns).
In try/except, catch the most specific exceptions possible.
For duck-typing try/except, keep the try body minimal and use else for the main logic.
Files:
tensorrt_llm/_torch/model_config.py
**/*.{cpp,cxx,cc,h,hpp,hh,hxx,cu,cuh,py}
📄 CodeRabbit inference engine (CODING_GUIDELINES.md)
Prepend the NVIDIA Apache-2.0 copyright header with current year to the top of all source files (e.g., .cpp, .h, .cu, .py).
Files:
tensorrt_llm/_torch/model_config.py
🧬 Code graph analysis (1)
tensorrt_llm/_torch/model_config.py (5)
tensorrt_llm/models/automodel.py (1)
AutoConfig(10-49)tensorrt_llm/llmapi/tokenizer.py (1)
from_pretrained(75-78)tests/unittest/llmapi/test_llm.py (1)
from_pretrained(296-299)tensorrt_llm/llmapi/llm_args.py (2)
model_dir(1171-1173)model_dir(1176-1180)tensorrt_llm/logger.py (1)
warning(131-132)
| # When handling the case where model_format is TLLM_ENGINE | ||
| # send cyclic requests to the NONE URL. | ||
| if checkpoint_dir is not None: | ||
| pretrained_config = transformers.AutoConfig.from_pretrained( | ||
| checkpoint_dir, | ||
| trust_remote_code=trust_remote_code, | ||
| ) | ||
|
|
||
| # Find the cache path by looking for the config.json file which should be in all | ||
| # huggingface models | ||
| model_dir = Path( | ||
| transformers.utils.hub.cached_file(checkpoint_dir, | ||
| 'config.json')).parent | ||
| # Find the cache path by looking for the config.json file which should be in all | ||
| # huggingface models | ||
| model_dir = Path( | ||
| transformers.utils.hub.cached_file(checkpoint_dir, | ||
| 'config.json')).parent | ||
| else: | ||
| pretrained_config = None | ||
| model_dir = Path("") | ||
| logger.warning( | ||
| "checkpoint_dir is None, please check your checkpoint directory path.") |
There was a problem hiding this comment.
🧩 Analysis chain
Critical: Downstream code not safe for None pretrained_config.
The guard prevents the invalid URL requests (good), but introducing None for pretrained_config creates runtime errors in properties and methods that assume it's non-null:
- Line 183-188: The
torch_dtypeproperty will raiseAttributeErrorwhen accessingself.pretrained_config.torch_dtypeifpretrained_configisNone. - Lines 439-535:
get_bindings_model_config()extensively accessespretrained_configattributes (num_attention_heads,hidden_size,vocab_size, etc.) withoutNonechecks. - Lines 554-572:
get_layer_types()andget_num_attention_layers()also accesspretrained_configattributes.
Recommended solutions:
- Raise an exception instead of logging a warning if
checkpoint_dirisNone, since the config cannot function properly without a validpretrained_config. - Add
Noneguards to all properties/methods that accesspretrained_configattributes (ifNoneis truly a valid state). - Fix the root cause so that
checkpoint_diris neverNonewhenfrom_pretrainedis called.
Apply this diff to raise an exception instead:
- else:
- pretrained_config = None
- model_dir = Path("")
- logger.warning(
- "checkpoint_dir is None, please check your checkpoint directory path.")
+ else:
+ raise ValueError(
+ "checkpoint_dir is None. Cannot load model config without a valid checkpoint directory."
+ )Run the following script to verify all usages of pretrained_config in this file:
🏁 Script executed:
#!/bin/bash
# Description: Find all accesses to pretrained_config attributes that lack None checks
# Search for attribute access patterns on pretrained_config
rg -nP 'self\.pretrained_config\.\w+' tensorrt_llm/_torch/model_config.pyLength of output: 1514
Enforce non-null pretrained_config or guard all accesses
- Unconditional accesses in methods/properties (
torch_dtypeat L188;get_bindings_model_config()at L457-470;get_layer_types()/get_num_attention_layers()at L544-573; plus others) will crash ifpretrained_configisNone. - Replace the
elsebranch infrom_pretrainedto raise instead of assigningNone, or add explicitNonechecks around every access.
- else:
- pretrained_config = None
- model_dir = Path("")
- logger.warning(
- "checkpoint_dir is None, please check your checkpoint directory path.")
+ else:
+ raise ValueError(
+ "checkpoint_dir is None. Cannot load model config without a valid checkpoint directory."
+ )📝 Committable suggestion
‼️ IMPORTANT
Carefully review the code before committing. Ensure that it accurately replaces the highlighted code, contains no missing lines, and has no issues with indentation. Thoroughly test & benchmark the code to ensure it meets the requirements.
| # When handling the case where model_format is TLLM_ENGINE | |
| # send cyclic requests to the NONE URL. | |
| if checkpoint_dir is not None: | |
| pretrained_config = transformers.AutoConfig.from_pretrained( | |
| checkpoint_dir, | |
| trust_remote_code=trust_remote_code, | |
| ) | |
| # Find the cache path by looking for the config.json file which should be in all | |
| # huggingface models | |
| model_dir = Path( | |
| transformers.utils.hub.cached_file(checkpoint_dir, | |
| 'config.json')).parent | |
| # Find the cache path by looking for the config.json file which should be in all | |
| # huggingface models | |
| model_dir = Path( | |
| transformers.utils.hub.cached_file(checkpoint_dir, | |
| 'config.json')).parent | |
| else: | |
| pretrained_config = None | |
| model_dir = Path("") | |
| logger.warning( | |
| "checkpoint_dir is None, please check your checkpoint directory path.") | |
| # When handling the case where model_format is TLLM_ENGINE | |
| # send cyclic requests to the NONE URL. | |
| if checkpoint_dir is not None: | |
| pretrained_config = transformers.AutoConfig.from_pretrained( | |
| checkpoint_dir, | |
| trust_remote_code=trust_remote_code, | |
| ) | |
| # Find the cache path by looking for the config.json file which should be in all | |
| # huggingface models | |
| model_dir = Path( | |
| transformers.utils.hub.cached_file(checkpoint_dir, | |
| 'config.json')).parent | |
| else: | |
| raise ValueError( | |
| "checkpoint_dir is None. Cannot load model config without a valid checkpoint directory." | |
| ) |
🤖 Prompt for AI Agents
In tensorrt_llm/_torch/model_config.py around lines 396 to 413, the current else
branch sets pretrained_config to None which will cause unconditional accesses
later to crash; change the else branch to raise a clear ValueError (or similar)
indicating checkpoint_dir is required, and remove the model_dir = Path("")
fallback so callers never see an invalid model_config; alternatively, if you
prefer nullable, add explicit None checks and raise when any method/property is
invoked without pretrained_config, but pick one approach and apply it
consistently to ensure pretrained_config is never accessed when None.
svc-trtllm-gh-bot
added
the
Community want to contribute
To address the issue described above, an if check is added to improve code runtime efficiency.
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.