You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
Hi, I build docker images by your dockerfile.
Start with command docker run --gpus=all --network=host --rm --name test --privileged --shm-size 32G -it <docker_image_name> bash
However, with nvidia-smi in container, CUDA Version will be Error. Do you have same problem?
Other gpu images does not have problem.
The text was updated successfully, but these errors were encountered:
Hi, the problem is the ln command ln -s /usr/local/cuda/targets/x86_64-linux/lib/stubs/libcuda.so /usr/local/cuda/targets/x86_64-linux/lib/stubs/libcuda.so.1
If use a nvidia image like nvidia/cuda:11.1.1-cudnn8-devel-centos8 as base image and dont ln -s libcuda. The problem will disappear.
Hi, I build docker images by your dockerfile.
Start with command
docker run --gpus=all --network=host --rm --name test --privileged --shm-size 32G -it <docker_image_name> bash
However, with nvidia-smi in container, CUDA Version will be Error. Do you have same problem?
Other gpu images does not have problem.
The text was updated successfully, but these errors were encountered: