pytorch / pytorch/TensorRT

🐛 [Bug] Remove prepare_inputs and passing inputs during compilation.

Open
#3,384 0 comments 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

bug story: Dynamo Frontend & Partitioning
Dominant language
Python
Stars
3k
Forks
410
Avg merge
3d 18h
Merged PRs (30d)
78

Description

Bug Description

Currently, we pass trt_arg_inputs and trt_kwarg_inputs to compile_module https://github.com/pytorch/TensorRT/blob/main/py/torch_tensorrt/dynamo/_compiler.py#L682. These are actually not being used. The prepare inputs call also fails sometimes during graph parsing for dry run. Since we read all the input info from graph metadata now, we can consider removing user inputs being passed around internally.

To Reproduce

Steps to reproduce the behavior:

Expected behavior

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

Contributor guide

Open the contributing guide

First steps

  1. Read the whole issue, then the project's contributing guide.
  2. Comment on the issue to say you are picking it up — it saves two people doing the same work.
  3. Fork the repository and make your change on a branch.
  4. Open a pull request that references the issue number.

Research direction

Start in py/torch_tensorrt/dynamo/_compiler.py around line 682 and trace how prepare_inputs, trt_arg_inputs, and trt_kwarg_inputs reach compile_module. Confirm that graph metadata supplies the input information, then remove the unused input passing and prepare_inputs path while preserving dry-run graph parsing. Done means compilation no longer depends on those user inputs.

Written by the indexing model from the issue text.

Assessment

Tech stack
python
Domain
backend, compilers
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Mostly clear
Newbie friendliness
35/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.