Ubuntu20.04 Linux 5.4.0-62 安装Cuda10.2

一、安装系统推荐NVIDIA显卡驱动

方法1. 安装系统推荐NVIDIA显卡驱动(通常为最新版本)

• 1.1 查看显卡名称及系统推荐驱动

sudo add-apt-repository ppa:graphics-drivers   #添加NVIDA显卡驱动库
sudo apt update
ubuntu-drivers devices                         #显示可安装驱动

运行结果如下图示：

root@labLab:/home/user1/work# ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0 ==
modalias : pci:v000010DEd00001E04sv00001458sd000037C0bc03sc00i00
vendor   : NVIDIA Corporation
model    : TU102 [GeForce RTX 2080 Ti]
manual_install: True
driver   : nvidia-driver-440-server - distro non-free
driver   : nvidia-driver-455 - third-party free
driver   : nvidia-driver-418-server - distro non-free
driver   : nvidia-driver-460 - third-party free recommended          # 系统推荐版本
driver   : nvidia-driver-450-server - distro non-free
driver   : xserver-xorg-video-nouveau - distro free builtin

1.2 运行下示命令安装

# 安装系统推荐版本
sudo ubuntu-drivers autoinstall

or
# 手动指定版本安装
sudo apt install nvidia-driver-455

1.3 重启并查看

# 重启系统
sudo reboot

# 查看GPU信息
nvidia-smi

1.4 实例说明：

# 1.我们选择安装系统推荐版本
sudo ubuntu-drivers autoinstall

# 2.重启系统
sudo reboot

# 3. 查看GPU信息
nvidia-smi

运行结果如下：

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 460.32       Driver Version: 460.32       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. |
|===============================+======================+======================|
|   0  GeForce RTX 208...  Off  | 00000000:01:00.0 Off |                  N/A |
|  0%   31C    P0    71W / 250W |      0MiB / 11016MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce RTX 208...  Off  | 00000000:02:00.0 Off |                  N/A |
|  0%   30C    P0    22W / 250W |      0MiB / 11019MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+

方法2. 手动下载驱动安装。

2.1 查看显卡名称

sudo add-apt-repository ppa:graphics-drivers   #添加NVIDA显卡驱动库
sudo apt update
ubuntu-drivers devices                         #显示可安装驱动

运行结果如下图示：

root@labLab:/home/user1/work# ubuntu-drivers devices
== /sys/devices/pci0000:00/0000:00:01.1/0000:02:00.0 ==
modalias : pci:v000010DEd00001E04sv00001458sd000037C0bc03sc00i00
vendor   : NVIDIA Corporation
model    : TU102 [GeForce RTX 2080 Ti]         # 显卡型号
manual_install: True
driver   : nvidia-driver-440-server - distro non-free
driver   : nvidia-driver-455 - third-party free
driver   : nvidia-driver-418-server - distro non-free
driver   : nvidia-driver-460 - third-party free recommended         
driver   : nvidia-driver-450-server - distro non-free
driver   : xserver-xorg-video-nouveau - distro free builtin

2.2 去nvidia官网下载驱动

nvidia官网：https://www.nvidia.com/Download/index.aspx

点击search。

点击download，进行下载页面。

在上图，右键点击download，选择复制链接地址。并通过wget下载驱动。

wget https://us.download.nvidia.com/XFree86/Linux-x86_64/460.32.03/NVIDIA-Linux-x86_64-460.32.03.run

2.3 运行下示命令安装驱动

sh NVIDIA-Linux-x86_64-460.32.03.run

二、安装指定版本(Cuda10.2)驱动

方法3. 安装指定版本的驱动

3.1 进入nvidia官网驱动高级查找：https://www.nvidia.com/Download/Find.aspx?lang=en-us

点击search 找到指定版本的驱动。

3.2 实列说明：

我们想使用CUDA 10.2，查下表后，故选择440.64版本的驱动。按照方法二下载驱动、安装、重启即可。

运行结果如下：

# 查看GPU信息
nvidia-smi

# 运行结果
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.64       Driver Version: 440.64       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce RTX 208...  Off  | 00000000:01:00.0 Off |                  N/A |
|  0%   31C    P0    71W / 250W |      0MiB / 11016MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce RTX 208...  Off  | 00000000:02:00.0 Off |                  N/A |
|  0%   30C    P0    22W / 250W |      0MiB / 11019MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

3.3 安装cuda-toolkit-10-2

sudo apt install cuda-toolkit-10-2

安装全部完成，结果如下图示。

# linux系统信息

Welcome to Ubuntu 20.04.1 LTS (GNU/Linux 5.4.0-62-generic x86_64)

 * Documentation:  https://help.ubuntu.com
 * Management:     https://landscape.canonical.com
 * Support:        https://ubuntu.com/advantage

3 updates can be installed immediately.
0 of these updates are security updates.
To see these additional updates run: apt list --upgradable


1 updates could not be installed automatically. For more details,
see /var/log/unattended-upgrades/unattended-upgrades.log

# nvcc -V
# 运行结果

nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89

# nvidia-smi
# 运行结果

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.64       Driver Version: 440.64       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce RTX 208...  Off  | 00000000:01:00.0 Off |                  N/A |
|  0%   31C    P0    71W / 250W |      0MiB / 11016MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce RTX 208...  Off  | 00000000:02:00.0 Off |                  N/A |
|  0%   30C    P0    22W / 250W |      0MiB / 11019MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

# 运行程序测试

+-----------------------------------------------------------------------------+
| NVIDIA-SMI 440.64       Driver Version: 440.64       CUDA Version: 10.2     |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  GeForce RTX 208...  Off  | 00000000:01:00.0 Off |                  N/A |
|  0%   32C    P0    43W / 250W |      0MiB / 11016MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
|   1  GeForce RTX 208...  Off  | 00000000:02:00.0 Off |                  N/A |
|  0%   38C    P2   233W / 250W |   9580MiB / 11019MiB |     99%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID   Type   Process name                             Usage      |
|=============================================================================|
|    1    188850      C   python                                      9569MiB |
+-----------------------------------------------------------------------------+