Cannot identify device '/dev/v4l2-nvenc' on arm in container with Tesla4

Accelerated Computing Intelligent Video Analytics DeepStream SDK
,
1

My condition:
system: Ubuntu22
architecture: arm64
GPU: Telsa 4

I’m working on a container from image “nvcr.io/nvidia/deepstream:7.1-triton-arm-sbsa”,I pulled it on a arm64 server,
and I have installed nvidia-docker , in the container I can use “nvidia-smi” and I can get the correct info about GPU.
Below is my start command:
“sudo docker run -it --rm --runtime=nvidia --network=host -e NVIDIA_DRIVER_CAPABILITIES=compute,utility,video,graphics --gpus all --privileged -e DISPLAY=:0 -v /tmp/.X11-unix:/tmp/.X11-unix -v /etc/X11:/etc/X11 nvcr.io/nvidia/deepstream:7.1-triton-arm-sbsa”

and then I pull the deepstream_python_app from github，and write a python file based on the “deepstream_test1_rtsp_in_rtsp_out.py” ,but I get this error:

Error: gst-resource-error-quark: Cannot identify device ‘/dev/v4l2-nvenc’. (3): /dvs/git/dirty/git-master_linux/3rdparty/gst/gst-v4l2/gst-v4l2/v4l2_calls.c(657): gst_v4l2_open (): /GstPipeline:pipeline0/nvv4l2h264enc:encoder:
system error: No such file or directory

I searched about this and most topics are about jetson device ，but I am using this container on a normal arm64 server , plz help me with it , thx !

3

Which version of GPU driver do you use, and what model of arm64 CPU do you have? We have only tested this docker image on nvidia’s arm64 CPU

This seems to be a driver problem. Please try to change to another version of the driver.

Here are similar topics

4

Thank you for helping me !
Here are the info of driver and cpu :
Driver Version: 570.124.06
model：Kunpeng-920

when I download driver from the website, I choose the " linux arrch64 " version instead of the “linux arm 64bit ubuntu 22.04” , maybe this would be the issue ?

and I am attempting to use x264enc instead of nvv4l2h264enc; is this approach feasible?

5

You can try it, but we cannot guarantee that nvinfer/nvstreammux and other plugins will work properly because we have not tested the device.