ubuntu18.04中cuda11.0和cuda9.0多版本安装与切换

ubuntu18.04中cuda11.0和cuda9.0多版本安装与切换

服务器中已经存在cuda11.0，现在需要用到cuda9.0
查看当前cuda版本：

~$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2020 NVIDIA Corporation
Built on Thu_Jun_11_22:26:38_PDT_2020
Cuda compilation tools, release 11.0, V11.0.194
Build cuda_11.0_bu.TC445_37.28540450_0

也可以通过nvidia-smi查看

下载cuda9.0的安装包

这里是历史版本的下载地址

https://developer.nvidia.com/cuda-toolkit-archive

选择runfile类型的安装文件

安装

下载完毕，得到文件cuda_9.0.176_384.81_linux.run
到对应目录下执行命令：

#先执行下面的命令安装相关依赖，
#否则会出现`Missing recommended library`错误
sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev 
 
 
sudo sh cuda_9.0.176_384.81_linux.run #开始安装

#按q退出协议说明.
accept/decline/quit: accept  #接受协议
 
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 375.26? 
y)es/(n)o/(q)uit: n  #是否显卡驱动包，由于已经安装显卡驱动，选择n
 
Install the CUDA 9.0 Toolkit?
(y)es/(n)o/(q)uit: y #是否安装工具包，选择y
 
Enter Toolkit Location
[ default is /usr/local/cuda-9.0 ]: #工具包安装地址
 
Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y #添加软链接，第一次安装cuda选yes，如果已经有cuda了就选no
 
Install the CUDA 9.0 Samples?
(y)es/(n)o/(q)uit: y #安装样例
 
Enter CUDA Samples Location
 [ default is /root ]:  #样例安装地址默认即可

如果出现“You are attempting to install on an unsupported configuration. Do you wish to continue?”，意味着可能是gcc的版本过高，后续安装会报错：

Error: unsupported compiler: 7.5.0. Use --override to override this check.
Error: cannot find Toolkit in /usr/local/cuda-9.0

===========
= Summary =
===========

Driver:   Not Selected
Toolkit:  Installation Failed. Using unsupported Compiler.
Samples:  Cannot find Toolkit in /usr/local/cuda-9.0

问题出在gcc的版本过高，查看gcc的版本信息：

~$ gcc --version
gcc (Ubuntu 7.5.0-3ubuntu1~18.04) 7.5.0
Copyright (C) 2017 Free Software Foundation, Inc.
This is free software; see the source for copying conditions.  There is NO
warranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.

这时就需要安装更低版本的gcc(6.0)，因为cuda9.0不支持gcc7

gcc降级

查看gcc文件夹中是否存在gcc6：

~$ ls /usr/bin/gcc*
/usr/bin/gcc    /usr/bin/gcc-ar    /usr/bin/gcc-nm    /usr/bin/gcc-ranlib
/usr/bin/gcc-7  /usr/bin/gcc-ar-7  /usr/bin/gcc-nm-7  /usr/bin/gcc-ranlib-7

如果不存在，则需要安装gcc6：

sudo apt-get install gcc-6

安装后的gcc文件夹：

~$ ls /usr/bin/gcc*
/usr/bin/gcc    /usr/bin/gcc-ar    /usr/bin/gcc-nm    /usr/bin/gcc-ranlib
/usr/bin/gcc-6  /usr/bin/gcc-ar-6  /usr/bin/gcc-nm-6  /usr/bin/gcc-ranlib-6
/usr/bin/gcc-7  /usr/bin/gcc-ar-7  /usr/bin/gcc-nm-7  /usr/bin/gcc-ranlib-7

gcc版本切换

修改版本优先级：

sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-6 100

上述命令将gcc-6的优先级设定为100
查看并调整优先级：

sudo update-alternatives --config gcc

继续安装cuda

调整gcc版本后继续进行安装，步骤与之前相同，最后安装成功：

===========
= Summary =
===========

Driver:   Not Selected
Toolkit:  Installed in /usr/local/cuda-9.0
Samples:  Installed in /home/songhui

Please make sure that
 -   PATH includes /usr/local/cuda-9.0/bin
 -   LD_LIBRARY_PATH includes /usr/local/cuda-9.0/lib64, or, add /usr/local/cuda-9.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-9.0/bin

Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-9.0/doc/pdf for detailed information on setting up CUDA.

***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 384.00 is required for CUDA 9.0 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

cuda版本切换

安装完后可以在/usr/local/目录下看到有多个版本的cuda存在：

~$ ls /usr/local/
bin  cuda  cuda-11.0  cuda-9.0  etc  games  include  lib  man  sbin  share  src

正在使用的cuda-11.0需要切换成cuda-9.0，而cuda是一个软链接，目前指向了cuda-11.0，所以只要将cuda指向cuda-9.0就完成了切换。
用stat命令查看cuda文件夹属性：

~$ stat /usr/local/cuda
  File: /usr/local/cuda -> /usr/local/cuda-11.0/
  Size: 21              Blocks: 0          IO Block: 4096   symbolic link
Device: fd00h/64768d    Inode: 8916387     Links: 1
Access: (0777/lrwxrwxrwx)  Uid: (    0/    root)   Gid: (    0/    root)
Access: 2021-09-22 14:40:15.444379761 +0800
Modify: 2020-11-19 10:37:24.010222584 +0800
Change: 2020-11-19 10:37:24.010222584 +0800
 Birth: -

可以看到文件类型是symbolic link软链接，指向了/usr/local/cuda-11.0/
所以把这个cuda文件夹删掉再新建一个指向/usr/local/cuda-9.0/的软链接就好了：

$ sudo rm -rf cuda
$ sudo ln -s /usr/local/cuda-9.0 /usr/local/cuda

如果没有修改成功，查看.bashrc文件配置：

vim ~/.bashrc

在最后添加：

export PATH="/usr/local/cuda/bin:$PATH"
export LD_LIBRARY_PATH=/usr/local/cuda/lib64:$LD_LIBRARY_PATH