1 环境

wsl wsl2

分支 ubuntu20.4

组件 nvidia-docker2

2 安装wsl2

# windown 10 2004及以上（保证windown环境已正确安装驱动）
wsl --install      # 默认安装ubuntu
wsl --update       # 更新到最新版本
wsl -d ubuntu      # 启动并进入ubuntu

3 安装docker

参考ubuntu安装docker官方教程

4 安装nvidia-docker

4.1 安装nvidia库

1
2
3

$curl -s -L https://nvidia.github.io/nvidia-docker/gpgkey | sudo apt-key add -
$curl -s -L https://nvidia.github.io/nvidia-docker/ubuntu18.04/nvidia-docker.list | sudo tee /etc/apt/sources.list.d/nvidia-docker.list
$sudo apt-get update

4.2 安装nvidia-docker2并重新加载docker守护进程

1
2
3

$sudo apt-get install nvidia-docker2
$sudo pkill -SIGHUP dockerd
$docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi

备注：运行上一步若报错

1
2
3

qyh@qyh-mas:~$ sudo docker run --runtime=nvidia --rm nvidia/cuda nvidia-smi
docker: Error response from daemon: Unknown runtime specified nvidia.
See 'docker run --help'.

当显卡驱动安装好的前提下（用指令nvidia-smi来显示显卡驱动），则换成运行下步，指定安装的cuda版本：

root@YW-HUAN-PC:/etc/apt/sources.list.d# nvidia-smi
Fri Jan 28 10:34:52 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.57.01    Driver Version: 471.41       CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| 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  NVIDIA GeForce ...  Off  | 00000000:01:00.0  On |                  N/A |
| 38%   29C    P8     1W /  38W |    520MiB /  2048MiB |    ERR!      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+
root@YW-HUAN-PC:/etc/apt/sources.list.d# nvidia-docker run --rm nvidia/cuda:10.1-devel | nvidia-smi
Fri Jan 28 10:35:24 2022
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.57.01    Driver Version: 471.41       CUDA Version: 11.4     |
|-------------------------------+----------------------+----------------------+
| 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  NVIDIA GeForce ...  Off  | 00000000:01:00.0  On |                  N/A |
| 38%   29C    P8     1W /  38W |    519MiB /  2048MiB |    ERR!      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+

+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

root@YW-HUAN-PC:/etc/apt/sources.list.d#

显示如上表示成功了。