Description
RTX 3090 needs cuda version >= 11.1
but most of the repos now are created before the release of 30 series GPU.
How can I run these repos on RTX 3090?
This issue not only appears in some but most repos now.
A clear and concise description of the bug or issue.
Like this repo:
It needs pytorch=1.0.1=cuda92py36h65efead_0, tensorflow-gpu=1.12.0=h0d30ee6_0 and cudatoolkit=9.2=0.
And I get the following error:
“cudnn error cudnn_status_execution_failed”.
This problem exists in most of the repos now, since most of the repos are not build with cuda >=11.1.
Environment
TensorRT Version:
GPU Type: RTX 3090
Nvidia Driver Version: 460.39
CUDA Version: 11.1
CUDNN Version: 7.6.5
Operating System + Version: Ubuntu 20.04 LTS
Python Version (if applicable): 3.6.13
TensorFlow Version (if applicable): 1.12.0
PyTorch Version (if applicable): 1.0.1
Baremetal or Container (if container which image + tag): Baremetal
Relevant Files
Please attach or include links to any models, data, files, or scripts necessary to reproduce your issue. (Github repo, Google Drive, Dropbox, etc.)
Steps To Reproduce
Execute “python run.py”
It will give error “cudnn error cudnn_status_execution_failed”
Please include:
- Exact steps/commands to build your repro
- Exact steps/commands to run your repro
- Full traceback of errors encountered