NVIDIA / NVIDIA/TensorRT

Suitable trt environment can't serialize model?

Open
#3,911 5 comments 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

Description

I'm trying to deserialize model on device A and serialize it on device B, but failed。Except for the graphics driver, the environment of A and B is the same. Does the graphics driver version affect the model serialization?
ERROR_LOG: CUDA_ERROR_INVALID_IMAGE(200): device kernel image is invalid

Environment

TensorRT Version: 7.2.2.3

NVIDIA GPU: 3080

NVIDIA Driver Version: device A: 536.23(max cuda version: 12.2) device B: 516.94(max cuda version: 11.7)

CUDA Version: 11.1

CUDNN Version: 8.0.4

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

No repository file, test, or entry point is named. Start by reproducing serialization and deserialization with TensorRT 7.2.2.3, CUDA 11.1, cuDNN 8.0.4, an RTX 3080, and the two reported driver versions; done means determining whether the environment difference explains CUDA_ERROR_INVALID_IMAGE and documenting the compatibility requirement.

Written by the indexing model from the issue text.

Assessment

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.