Rocky Linux 8.5 环境 NVIDIA460.84、cuda_11.2.0、cudnn8.1.1.33 安装

环境 Rocky Linux 8.5，在官网下载镜像制作启动盘进行系统安装即可。(承接上一篇文章：磁盘挂载与gcc9.3安装 cat /etc/redhat-release)

目录

一、NVIDIA460.84驱动安装

1、禁用nouveau驱动

2、安装显卡驱动

二、CUDA安装

1、安装cuda_11.2.0

2、安装cudnn8.1.1.33

三、参考

---

一、NVIDIA460.84驱动安装

1、禁用nouveau驱动

输入以下 命令进行查看，应该是有回显出现的。如果没有回显出现，那么你可以省略此步骤。

lsmod | grep nouveau

在/etc/modprobe.d/blacklist.conf 中添加nouveau 到黑名单。

vim /etc/modprobe.d/blacklist.conf

在里面添加：

blacklist nouveau

options nouveau modeset=0

保存退出

dracut --force //Linux更新内核

或者备份并更新内核

//重新建立initramfs image文件（生成新的内核，这个内核在开机的时候不会加载nouveau驱动程序）
mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r).img.bak
mv /boot/initramfs-$(uname -r).img.bak /home/initramfs-$(uname -r).img.bak
dracut /boot/initramfs-$(uname -r).img $(uname -r)

//切换到命令行界面
systemctl set-default multi-user.target  
//切回图像界面(systemctl用法:开机启动服务 systemctl enable ***.service)
//systemctl set-default graphical.target

修改后需要重启系统。确认下Nouveau是已经被你干掉，使用命令： lsmod | grep nouveau

2、安装显卡驱动

查看显卡型号

lshw -c video
yum install epel-release #安装epel源
yum -y install gcc kernel-devel dkms
//yum -y install gcc kernel-devel "kernel-devel-uname-r == $(uname -r)" dkms
yum install libglvnd-devel.x86_64

已安装:
  dkms-3.0.3-1.el8.noarch                                                       
  elfutils-libelf-devel-0.185-1.el8.x86_64                                      
  kernel-devel-4.18.0-348.20.1.el8_5.x86_64

安装显卡驱动 :

./NVIDIA-Linux-x86_64-460.84.run

1. Would you like to register the kernel module souces with DKMS? This will allow DKMS to automatically build a new module, if you install a different kernel later? 【No】
2. Nvidia’s 32-bit compatibility libraries? 【No】
3. Would you like to run the nvidia-xconfigutility to automatically update your x configuration so that the NVIDIA x driver will be used when you restart x? Any pre-existing x confile will be backed up. 【Yes】

验证是否安装成功

nvidia-smi

Mon Apr 18 15:19:50 2022       
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.84       Driver Version: 460.84       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 106...  Off  | 00000000:03:00.0 Off |                  N/A |
| 28%   34C    P0    28W / 120W |      0MiB /  6078MiB |      0%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
|   1  GeForce GTX 106...  Off  | 00000000:82:00.0 Off |                  N/A |
| 39%   39C    P0    30W / 120W |      0MiB /  6078MiB |      1%      Default |
|                               |                      |                  N/A |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                                  |
|  GPU   GI   CI        PID   Type   Process name                  GPU Memory |
|        ID   ID                                                   Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

二、CUDA安装

1、安装cuda_11.2.0

./cuda_11.2.0_460.27.04_linux.run

系统安装时，是UEFI模式启动的，则在BIOS中需禁用Security BOOT选项。

踩坑之神：安装失败，查看失败原因

cat /var/log/cuda-installer.log
[INFO]: Driver not installed.
[INFO]: Checking compiler version...
[INFO]: gcc location: /usr/local/bin/gcc

[INFO]: gcc version: gcc 版本 9.3.0 (GCC) 

[INFO]: Initializing menu
[INFO]: Setup complete
[INFO]: Components to install: 
[INFO]: Driver
[INFO]: 460.27.04
[INFO]: Executing NVIDIA-Linux-x86_64-460.27.04.run --ui=none --no-questions --accept-license --disable-nouveau --no-cc-version-check --install-libglvnd  2>&1
[INFO]: Finished with code: 256
[ERROR]: Install of driver component failed.
[ERROR]: Install of 460.27.04 failed, quitting

cat /var/log/nvidia-installer.log
ERROR: The nvidia-drm kernel module was not created.
ERROR: The nvidia-drm kernel module failed to build. This kernel module is required for the proper operation of DRM-KMS. If you do not need to use DRM-KMS, you can try to install this driver package again with the '--no-drm' option.
ERROR: Installation has failed.  Please see the file '/var/log/nvidia-installer.log' for details.  You may find suggestions on fixing installation problems in the README available on the Linux driver download page at www.nvidia.com.

uname -r
ll /usr/src/kernels/

如查询的结果不一致，则解决办法，升级内核
yum -y update

问题解决：安装时去掉Driver选项，因为刚才已经单独安装过了。

./cuda_11.2.0_460.27.04_linux.run 
===========
= Summary =
===========

Driver:   Not Selected
Toolkit:  Installed in /usr/local/cuda-11.2/
Samples:  Installed in /home/hhs-face/, but missing recommended libraries

Please make sure that
 -   PATH includes /usr/local/cuda-11.2/bin
 -   LD_LIBRARY_PATH includes /usr/local/cuda-11.2/lib64, or, add /usr/local/cuda-11.2/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run cuda-uninstaller in /usr/local/cuda-11.2/bin
***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 460.00 is required for CUDA 11.2 functionality to work.
To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:
    sudo <CudaInstaller>.run --silent --driver

配置环境变量

vim /etc/profile
//在末尾添加：
export PATH=/usr/local/cuda-11.2/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda-11.2/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}

//保存退出，立即生效
source /etc/profile

重启、查看版本，验证安装成功

reboot

nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Mon_Nov_30_19:08:53_PST_2020
Cuda compilation tools, release 11.2, V11.2.67
Build cuda_11.2.r11.2/compiler.29373293_0

2、安装cudnn8.1.1.33

chmod +x cudnn-11.2-linux-x64-v8.1.1.33.tgz
tar -zxvf cudnn-11.2-linux-x64-v8.1.1.33.tgz

//将解压后得到的文件夹，分别复制到cuda安装路径下与cuda的bin ，include 和lib文件夹合并。
cp cuda/include/cudnn.h /usr/local/cuda-11.2/include
cp cuda/include/cudnn_version.h /usr/local/cuda-11.2/include
cp cuda/lib64/libcudnn* /usr/local/cuda-11.2/lib64
chmod a+r /usr/local/cuda-11.2/include/cudnn.h /usr/local/cuda-11.2/lib64/libcudnn*

//检验证CUDNN是否安装成功, 检查CUDNN版本，这里的版本是8.1.1。
cat /usr/local/cuda-11.2/include/cudnn_version.h | grep CUDNN_MAJOR -A 2
#define CUDNN_MAJOR 8
#define CUDNN_MINOR 1
#define CUDNN_PATCHLEVEL 1
--
#define CUDNN_VERSION (CUDNN_MAJOR * 1000 + CUDNN_MINOR * 100 + CUDNN_PATCHLEVEL)

三、参考

CUDA和CUDNN的关系