硬件和操作系统环境

intel cpu芯片：

[root@centos7 ~]# cat /proc/cpuinfo | grep name | cut -f2 -d: | uniq -c
     32  Intel(R) Xeon(R) CPU E5-2667 v2 @ 3.30GHz

显卡硬件情况：

[root@centos7 ~]# lspci | grep -i vga
0e:01.0 VGA compatible controller: Matrox Electronics Systems Ltd. MGA G200eW WPCM450 (rev 0a)
84:00.0 VGA compatible controller: NVIDIA Corporation GP104 [GeForce GTX 1070 Ti] (rev a1)

操作系统情况：

[root@centos7 ~]# uname -r
3.10.0-1160.59.1.el7.x86_64
[root@centos7 ~]# uname -a
Linux centos7.9 3.10.0-1160.59.1.el7.x86_64 #1 SMP Wed Feb 23 16:47:03 UTC 2022 x86_64 x86_64 x86_64 GNU/Linux

安装nvidia驱动-屏蔽nouveau模块

解决直接安装时报错问题，需要屏蔽 nouveau 模块

 yum erase kmod-nvidia

rpm --import https://www.elrepo.org/RPM-GPG-KEY-elrepo.org
rpm -Uvh http://www.elrepo.org/elrepo-release-7.0-2.el7.elrepo.noarch.rpm
yum install yum-plugin-fastestmirror
vim /lib/modprobe.d/dist-blacklist.conf

填入：
#blacklist nvidiafb
blacklist nouveau
options nouveau modeset=0

mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r).img.bak
dracut /boot/initramfs-$(uname -r).img $(uname -r)
lspci | grep nouveau

没有输出，说明屏蔽默认带有的nouveau成功。

安装nvidia驱动-显卡检测工具

yum install nvidia-detect   nvidia显卡检测
nvidia-detect -v                       检测结果
yum -y install kmod-nvidia     安装显卡驱动 这一步应该不需要 

[root@centos7 ~]# nvidia-detect -v 
Probing for supported NVIDIA devices...
[102b:0532] Matrox Electronics Systems Ltd. MGA G200eW WPCM450
[10de:1b82] NVIDIA Corporation GP104 [GeForce GTX 1070 Ti]
This device requires the current 510.47.03 NVIDIA driver kmod-nvidia
WARNING: Xorg log file /var/log/Xorg.0.log does not exist
WARNING: Unable to determine Xorg ABI compatibility
WARNING: The driver for this device does not support the current Xorg version

安装nvidia驱动-脚本安装

cd /opt/nvidia
wget https://international.download.nvidia.com/XFree86/Linux-x86_64/430.40/NVIDIA-Linux-x86_64-430.40.run
chmod +x NVIDIA-Linux-x86_64-460.39.run
./NVIDIA-Linux-x86_64-460.39.run --kernel-source-path=/usr/src/kernels/3.10.0-1160.59.1.el7.x86_64
# 下面是移除NVENC同时运行最大数量的限制 restriction on maximum number of simultaneous NVENC
git clone https://github.com/keylase/nvidia-patch
cd nvidia-patch
bash ./patch.sh

安装nvidia驱动-脚本卸载

./NVIDIA-Linux-x86_64-460.39.run --uninstall

安装nvidia驱动-安装Cuda

yum-config-manager --add-repo http://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo
yum clean all
yum -y install nvidia-driver-latest-dkms cuda

[root@centos7 ~]# nvidia-smi
Sat Mar 26 21:05:07 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.39       Driver Version: 460.39       CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| 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  GeForce GTX 107...  Off  | 00000000:84:00.0 Off |                  N/A |
|  0%   34C    P0    33W / 180W |      0MiB /  8119MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

[root@centos7 ~]# docker run -it --rm --name test --gpus all ubuntu:latest nvidia-smi                                                  
Sat Mar 26 13:04:50 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.39       Driver Version: 460.39       CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| 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  GeForce GTX 107...  Off  | 00000000:84:00.0 Off |                  N/A |
|  0%   34C    P0    33W / 180W |      0MiB /  8119MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

[root@centos7 ~]# docker run -it --rm --name test --gpus all centos:latest nvidia-smi      
Sat Mar 26 13:05:39 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.39       Driver Version: 460.39       CUDA Version: 11.2     |
|-------------------------------+----------------------+----------------------+
| 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  GeForce GTX 107...  Off  | 00000000:84:00.0 Off |                  N/A |
|  0%   34C    P0    33W / 180W |      0MiB /  8119MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+