运行环境：

基础镜像：ubuntu:20.04
驱动版本：rock-4.5.2-5.11.39-V01.5.run
DTK:DTK-22.10.1-Ubuntu20.04.1-x86_64.tar.gz
飞浆：paddlepaddle-2.3.2_dtk2210_git0195561-cp38-cp38-manylinux2014_x86_64.whl
用户组信息：
root@d5981efc58e7:/opt/ocr# id  ocr
uid=1001(ocr) gid=1001(ocr) groups=1001(ocr),39(irc),44(video)

使用官方提供镜像测试，发现也是存在这个问题

#启动容器
docker run --rm -it -u 1001:1001 --name paddle-dev --device=/dev/kfd --device=/dev/dri --group-add video --group-add 39 image.sourcefind.cn:5000/dcu/admin/base/paddlepaddle:2.3.2-ubuntu20.04-dtk-22.10-py37-latest /bin/bash
#运行测试代码，
python3 -c "import paddle; paddle.utils.run_check()"
#会出现如下警告：
usr/local/lib/python3.7/site-packages/paddle/fluid/framework.py:597: UserWarning: You are using GPU version Paddle, but your CUDA device is not set properly. CPU device will be used by default.
  "You are using GPU version Paddle, but your CUDA device is not set properly. CPU device will be used by default."
....
PaddlePaddle works well on 2 CPUs.
#正常情况下应该是没有警告输出
PaddlePaddle works well on 2 GPUs.