🐛 [Bug] RuntimeError: Failed to extract symbolic shape expressions from source FX graph partition

[zewenli98]

Bug Description

Running supported models via run_llm.py gets the error:

Traceback (most recent call last):
  File "/home/zewenl/Documents/pytorch/TensorRT/tools/llm/run_llm.py", line 333, in <module>
    trt_model = compile_torchtrt(model, input_ids, args)
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/zewenl/Documents/pytorch/TensorRT/tools/llm/run_llm.py", line 134, in compile_torchtrt
    trt_model = torch_tensorrt.dynamo.compile(
                ^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/zewenl/Documents/pytorch/TensorRT/py/torch_tensorrt/dynamo/_compiler.py", line 798, in compile
    trt_gm = compile_module(
             ^^^^^^^^^^^^^^^
  File "/home/zewenl/Documents/pytorch/TensorRT/py/torch_tensorrt/dynamo/_compiler.py", line 1044, in compile_module
    trt_module = convert_module(
                 ^^^^^^^^^^^^^^^
  File "/home/zewenl/Documents/pytorch/TensorRT/py/torch_tensorrt/dynamo/conversion/_conversion.py", line 343, in convert_module
    serialized_interpreter_result = interpret_module_to_result(
                                    ^^^^^^^^^^^^^^^^^^^^^^^^^^^
  File "/home/zewenl/Documents/pytorch/TensorRT/py/torch_tensorrt/dynamo/conversion/_conversion.py", line 215, in interpret_module_to_result
    raise RuntimeError(
RuntimeError: Failed to extract symbolic shape expressions from source FX graph partition

To Reproduce

python run_llm.py --model Qwen/Qwen2.5-0.5B-Instruct --prompt "What is parallel programming?" --model_precision FP16 --num_tokens 128 --cache static_v2 --benchmark
or 
python run_llm.py --model gpt2 --prompt "What is parallel programming?" --model_precision FP16 --num_tokens 128 --cache static_v2 --benchmark

Expected behavior

correct outputs

Environment

Build information about Torch-TensorRT can be found by turning on debug messages

• Torch-TensorRT Version (e.g. 1.0.0):
• PyTorch Version (e.g. 1.0):
• CPU Architecture:
• OS (e.g., Linux):
• How you installed PyTorch (conda, pip, libtorch, source):
• Build command you used (if compiling from source):
• Are you using local sources or building from archives:
• Python version:
• CUDA version:
• GPU models and configuration:
• Any other relevant information:

Additional context

[narendasan]

@zewenli98 what build / commit?

[zewenli98]

@zewenli98 what build / commit?

c4a57a7