NVIDIA / NVIDIA/TensorRT

undefined symbol: OrtGetApiBase, version VERS_1.15.1

Open
#3,753 1 comment 0 reactions 0 assignees View on GitHub

Nobody has claimed this yet.

triaged
Dominant language
C++
Stars
13.4k
Forks
2.4k
Avg merge
5d 3h
Merged PRs (30d)
2

Description

Hello! I have a DINO model converted to an onnx model using mmdeploy with a custom op. I was trying to use the nvcr.io/nvidia/tensorrt:23.01-py3 image to use trtexec to convert this model to an trt plan while specifying the plugin to be used. But i have been seeing this weird error when during the plugin I am getting this error

trtexec: symbol lookup error: /workspace/project/libs/libmmdeploy_onnxruntime_ops.so: undefined symbol: OrtGetApiBase, version VERS_1.15.1

I made sure to add the libonnxruntime 1.15.1 and libonnxruntime.so files to the LD_PRELOAD and LD_LIBRARY_PATH but i still get the same error. Any idea how to potentially fix this or debug this further?

TensorRT version: 8.5.2

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 by reproducing the failure with trtexec in the nvcr.io/nvidia/tensorrt:23.01-py3 image and inspect the loading of /workspace/project/libs/libmmdeploy_onnxruntime_ops.so. Check how OrtGetApiBase and the ONNX Runtime 1.15.1 libraries are resolved with LD_PRELOAD and LD_LIBRARY_PATH; done means identifying the incompatible or missing library configuration and documenting a working conversion setup.

Written by the indexing model from the issue text.

Assessment

Tech stack
cpp
Domain
machine-learning
Issue type
Bug
Difficulty
4/5
Estimated time
3-5 days
Activity status
Stale
Clarity
Needs clarification
Newbie friendliness
20/100

Get new issues in your inbox

A short digest of beginner-friendly GitHub issues.