Details:
I am using nvidia-container-toolkit v 1.16.2 and libnvidia-container v 1.16.2. I have integrated in the AGL (YOCTO ) build. The image i sready and deployed. I can see the nvidia-smi and nvidia-container-cli works fine. The issue is when I do
docker run --rm --gpus all --platform linux/arm64 \
--runtime=nvidia
-e NVIDIA_DISABLE_REQUIRE=1
nvidia/cuda:11.8.0-base-ubuntu22.04 nvidia-smi
docker: Error response from daemon: failed to create task for container: failed to create shim task: OCI runtime create failed: runc create failed: unable to start container process: error during container init: error running hook #0: error running hook: exit status 1, stdout: , stderr: Auto-detected mode as 'legacy'
nvidia-container-cli: ldcache error: process /sbin/ldconfig failed with error code: 1: unknown.
please suggest the resolution for the above stated issue