Skip to content
New issue

Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.

By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.

Already on GitHub? Sign in to your account

Altough using onnxruntime-19.0 but still can't use cudnn 9.x #22809

Open
wcycqjy opened this issue Nov 12, 2024 · 1 comment
Open

Altough using onnxruntime-19.0 but still can't use cudnn 9.x #22809

wcycqjy opened this issue Nov 12, 2024 · 1 comment
Labels
platform:jetson issues related to the NVIDIA Jetson platform

Comments

@wcycqjy
Copy link

wcycqjy commented Nov 12, 2024

Describe the issue

I try to infer an onnx model in jetson agx orin with Jetpack 6.1. The cuda is 12.6 and the cudnn is 9.3. I find on the website it says onnxruntime-19.0 supports cudnn 9.x but when I run the program the error still shows
"2024-11-12 10:07:56.895559377 [W:onnxruntime:Default, onnxruntime_pybind_state.cc:965 CreateExecutionProviderInstance] Failed to create CUDAExecutionProvider. Require cuDNN 8.* and CUDA 12.*. Please install all dependencies as mentioned in the GPU requirements page (https://onnxruntime.ai/docs/execution-providers/CUDA-ExecutionProvider.html#requirements), make sure they're in the PATH, and that your GPU is supported."

I install the onnxruntime-gpu 1.19.0 by downloading it from Jetson zoo btw.

To reproduce

Using onnxruntime 1.19.0 on jetson agx orin with Jetpack 6.1.

Urgency

No response

Platform

Linux

OS Version

22.04

ONNX Runtime Installation

Released Package

ONNX Runtime Version or Commit ID

1.19.0-jetson

ONNX Runtime API

Python

Architecture

ARM64

Execution Provider

CUDA

Execution Provider Library Version

cuda 12.6, cudnn 9.3, jetpack 6.1

@github-actions github-actions bot added the platform:jetson issues related to the NVIDIA Jetson platform label Nov 12, 2024
@tianleiwu
Copy link
Contributor

tianleiwu commented Nov 14, 2024

That means the package is built with cudnn 8.x.

If you need cudnn 9.x, you can build 1.20 from source using cudnn 9.x. See https://onnxruntime.ai/docs/build/eps.html#nvidia-jetson-tx1tx2nanoxavierorin for more information.

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment
Labels
platform:jetson issues related to the NVIDIA Jetson platform
Projects
None yet
Development

No branches or pull requests

2 participants