前言

博主由于視覺開發(fā)需求, 配置nvidia驅動并映射到docker中運行, 在本文中記錄過程及遇到的問題

硬件及軟件環(huán)境

   Static hostname: debian
         Icon name: computer-desktop
           Chassis: desktop
  Operating System: Debian GNU/Linux 11 (bullseye)
            Kernel: Linux 5.10.0-19-amd64
      Architecture: x86-64
     
  CPU: 12th Gen Intel(R) Core(TM) i7-12700F
  GPU: Nvidia Quadro M2000

驅動

下載驅動

根據(jù)自己的顯卡型號去官網(wǎng)搜索對應的驅動程序. 本機選擇470.161…03版本驅動.
NVIDIA 驅動程序下載 官方高級驅動搜索

cuda對應驅動版本要求對照表:
NVIDIA CUDA Toolkit Release Notes

注意!

1. 直接使用apt-get install nvidia-driver時不可運行(can not communicate with nvidia driver 類似報錯)
2. 下載最新驅動525時不可運行(can not communicate with nvidia driver 類似報錯)
3. 安裝時需屏蔽x server及nouveau¹

安裝

禁用nouveau

sudo bash -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf"
sudo bash -c "echo options nouveau modeset=0 >> /etc/modprobe.d/blacklist-nvidia-nouveau.conf"

禁用后重啟計算機

sudo reboot

安裝依賴

后續(xù)編譯使用的依賴軟件²

sudo apt-get install gcc g++ cmake pkg-config libglvnd-dev
sudo apt-get install linux-headers-$(uname -r|sed 's/[^-]*-[^-]*-//')

禁用xserver

sudo service gdm3 stop

輸入該行指令后會進入命令行狀態(tài), 此時只有一個光標, 通過按Ctrl + Alt + F1和Ctrl + Alt + F2即可跳出輸入用戶名密碼的指令行.

設置可執(zhí)行并運行

chmod +x ~/Downloads/NVIDIA-Linux-x86_64-470.161.03.run

# 需要以管理員權限運行
sudo ~/Downloads/NVIDIA-Linux-x86_64-470.161.03.run 

中間彈窗可以按照以下幾個選項：

Are you sure you want to continue? ->                  CONTINUE INSTALLATION
Would you like to run the nvidia-xconfig utility? ->             YES

安裝完成后, 重啟計算機并刪除禁用nouveau時創(chuàng)建的blacklist文件

檢查安裝結果

nvidia-smi

# 輸出
Thu Mar  9 14:22:29 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.161.03   Driver Version: 470.161.03   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  Quadro M2000        Off  | 00000000:01:00.0  On |                  N/A |
| 63%   59C    P0    38W /  75W |    769MiB /  4041MiB |     30%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|    0   N/A  N/A      1835      G   /usr/lib/xorg/Xorg                282MiB |
|    0   N/A  N/A      1982      G   /usr/bin/gnome-shell              110MiB |
|    0   N/A  N/A     30799      G   gnome-control-center               39MiB |
+-----------------------------------------------------------------------------+

Docker配置

Docker安裝

安裝可以參考此文如何建立并使用docker

nvidia-container-runtime安裝3

命令

nano nvidia-container-runtime-script.sh

腳本內容

curl -s -L https://nvidia.github.io/nvidia-container-runtime/gpgkey | \
  sudo apt-key add -
distribution=$(. /etc/os-release;echo $ID$VERSION_ID)
curl -s -L https://nvidia.github.io/nvidia-container-runtime/$distribution/nvidia-container-runtime.list | \
  sudo tee /etc/apt/sources.list.d/nvidia-container-runtime.list
sudo apt-get update

執(zhí)行腳本

nvidia-container-runtime-script.sh

安裝 nvidia-container-runtime

sudo apt-get install nvidia-container-runtime
sudo systemctl restart docker # 重啟docker

檢測

which nvidia-container-runtime-hook 
/usr/bin/nvidia-container-runtime-hook

Docker gpu 驗證

docker pull nvidia/cuda:11.3.1-base-ubuntu20.04
docker run --gpus all --rm -it nvidia/cuda:11.3.1-base-ubuntu20.04 bash

nvidia-smi
#輸出如下, 說明運行成功:
root@8a57ae3075d7:/# nvidia-smi
Thu Mar  9 06:42:20 2023       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 470.161.03   Driver Version: 470.161.03   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  Quadro M2000        Off  | 00000000:01:00.0  On |                  N/A |
| 62%   53C    P0    28W /  75W |    761MiB /  4041MiB |     34%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
+-----------------------------------------------------------------------------+

卸載指令

卸載安裝的驅動可以使用⁴:

sudo apt-get --purge remove "*nvidia*"
sudo /usr/bin/nvidia-uninstall

總結

本文用來記錄Debian11在安裝nvidia驅動和docker運行時遇到的一些問題, 由于是事后補寫可能中間有些異常處理略有缺漏, 各位同學有問題可以留言交流.

異常處理參考鏈接

顯卡驅動報錯：NVIDIA-SMI has failed because it couldn’t communicate with the NVIDIA driver.
固件錯誤Possible missing firmware解決辦法

到了這里，關于debian11安裝Nvidia驅動及Docker運行的文章就介紹完了。如果您還想了解更多內容，請在右上角搜索TOY模板網(wǎng)以前的文章或繼續(xù)瀏覽下面的相關文章，希望大家以后多多支持TOY模板網(wǎng)！