当前位置：首页 > news >正文

ubuntu20.04地平线OE3.2.0 GPU Docker使用

news 2025/10/11 6:50:58

1. docker安装

参考博文 ubuntu20.04 安装离线版docker-20.10.0

(1) 下载docker文件

开发PC为x86处理器，从 https://download.docker.com/linux/static/stable/x86_64/下载docker-20.10.0.tgz。

(2) 解压OE包里GPU镜像

解压 OE包里的GPU镜像

gzip -d docker_open_explorer_ubuntu_22_j6_gpu_v3.2.0.tar.gz

(3) 拷贝文件到/usr/bin/目录下

sudo cp docker/* /usr/bin/

若出现如下错误：

cp: cannot create regular file '/usr/bin/dockerd': Text file busy

可用如下指令，强制覆盖目标文件。

sudo cp -f docker/* /usr/bin/

(4) 将docker注册成系统服务

sudo gedit /etc/systemd/system/docker.service

文件中写人如下内容：

[Unit]
Description=Docker Application Container Engine
Documentation=https://docs.docker.com
After=network-online.target firewalld.service
Wants=network-online.target[Service]
Type=notify
ExecStart=/usr/bin/dockerd
ExecReload=/bin/kill -s HUP $MAINPID
LimitNOFILE=infinity
LimitNPROC=infinity
TimeoutStartSec=0
Delegate=yes
KillMode=process
Restart=on-failure
StartLimitBurst=3
StartLimitInterval=60s[Install]
WantedBy=multi-user.target

(5) 文件增加可执行权限

sudo chmod +x /etc/systemd/system/docker.service
systemctl daemon-reload

(6) 启动及测试

systemctl start docker
docker --version

输出如下类似：

Docker version 20.10.0, build 7287ab3

代表安装完成。

(7) 卸载

# 停止docker
sudo systemctl stop docker
# 移除开机自启动
systemctl disable docker.service
# 删除service服务
rm -f /etc/systemd/system/docker.service
# 删除docker相关命令
rm -f /usr/bin/docker*
rm -f /usr/bin/containerd*
rm -f /usr/bin/ctr
rm -f /usr/bin/runc
# 删除docker目录和容器相关文件
sudo rm -rf /var/lib/docker
sudo rm -rf /var/lib/containerd
# 测试
docker --version

2. 安装NVIDIA Container Toolkit

# 配置源
curl -fsSL https://nvidia.github.io/libnvidia-container/gpgkey | sudo gpg --dearmor -o /usr/share/keyrings/nvidia-container-toolkit-keyring.gpg \&& curl -s -L https://nvidia.github.io/libnvidia-container/stable/deb/nvidia-container-toolkit.list | \sed 's#deb https://#deb [signed-by=/usr/share/keyrings/nvidia-container-toolkit-keyring.gpg] https://#g' | \sudo tee /etc/apt/sources.list.d/nvidia-container-toolkit.listsudo apt-get update
# 安装NVIDIA Container Toolkit包
export NVIDIA_CONTAINER_TOOLKIT_VERSION=1.17.8-1
sudo apt-get install -y       nvidia-container-toolkit=${NVIDIA_CONTAINER_TOOLKIT_VERSION}       nvidia-container-toolkit-base=${NVIDIA_CONTAINER_TOOLKIT_VERSION}       libnvidia-container-tools=${NVIDIA_CONTAINER_TOOLKIT_VERSION}       libnvidia-container1=${NVIDIA_CONTAINER_TOOLKIT_VERSION}#重启docker
sudo systemctl restart docker

3. 加载镜像

# 加载tar镜像
docker load -i docker_open_explorer_ubuntu_22_j6_gpu_v3.2.0.tar
# 查看镜像id
docker images

输出如下：

REPOSITORY                                   TAG       IMAGE ID       CREATED        SIZE
openexplorer/ai_toolchain_ubuntu_22_j6_gpu   v3.2.0    1d79ca1300ec   3 months ago   28.4GB

运行：

sudo docker run -it --gpus all 1d79ca1300ec

容器内验证可使用宿主机资源

kai@kai:~/Downloads/OE_3.2.0$ sudo docker run -it --gpus all 1d79ca1300ec
root@ca8a5a317809:/open_explorer# nvidia-smi 
Fri Oct 10 16:38:41 2025       
+---------------------------------------------------------------------------------------+
| NVIDIA-SMI 535.183.01             Driver Version: 535.183.01   CUDA Version: 12.2     |
|-----------------------------------------+----------------------+----------------------+
| GPU  Name                 Persistence-M | Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp   Perf          Pwr:Usage/Cap |         Memory-Usage | GPU-Util  Compute M. |
|                                         |                      |               MIG M. |
|=========================================+======================+======================|
|   0  Quadro P2200                   Off | 00000000:65:00.0  On |                  N/A |
| 52%   46C    P8               8W /  75W |    569MiB /  5120MiB |      0%      Default |
|                                         |                      |                  N/A |
+-----------------------------------------+----------------------+----------------------++---------------------------------------------------------------------------------------+
| Processes:                                                                            |
|  GPU   GI   CI        PID   Type   Process name                            GPU Memory |
|        ID   ID                                                             Usage      |
|=======================================================================================|
+---------------------------------------------------------------------------------------+

查看全文

http://www.dtcms.com/a/465375.html