关闭

NVIDIA cuda7在centos6.5中的安装

标签： NVIDIAcuda7centos6-5安装

2015-08-03 15:14 123人阅读评论(0)收藏举报

分类：

CentOS（2）

目录(?)[+]

基本环境

cuda_7.0.28_linux.run
CentOS 6.5

安装过程

禁用Nouveau，这个驱动和Nvidia的驱动有冲突，需要先禁用掉，不然会不让安装，具体可参考NVIDIA CUDA GETTING STARTED GUIDE FOR LINUX
1. 创建文件，/etc/modprobe.d/blacklist-nouveau.conf，其内容如下：
  blacklist nouveau
  options nouveau modeset=0
2. 重新生成kernel initramfs，即Regenerate the kernel initramfs，执行如下命令
  sudo dracut –force
  注意：force前面是2个-
设置系统启动为Text Mode，即修改文件：/etc/inittab，将id:5改为id:3，让系统不从界面化启动，安装完成后，再修改过来。注意，不要从界面启动，然后通过按ctrl+alt+f3进入Text界面来进行安装，这个好像会出问题，会不让安装。
重启系统
直接运行cuda_7.0.28_linux.run，该文件需要有可执行权限
如果安装时，提示X server is running，不让进行安装，则执行如下操作：
1. 查询下和X相关的服务，pstree -p | grep X
2. 如果查询知道有服务在运行，就kill掉，比如：
  |-Xvnc(2259)
  |-Xvnc(2299)
  这时候，就依次执行
  sudo kill -9 2259
  sudo kill -9 2299
  将相关服务给杀掉，这时候应该就可以安装了
运行安装文件后，根据提示进行默认安装即可。安装完成后需要配置PATH和LD_LIBRARY_PATH。
将id:3修改回为id:5，重启系统。
如果重启后停留在进度条界面，正常进不了界面化的登录界面，则按ctrl+alt+f3进入Text界面，登录进入系统。然后查看/etc/X11/xorg.conf文件是否存在。一般是因为这个文件不存在导致的。这时候执行如下命令：
nvidia-xconfig
该命令会生成相应的xorg.conf文件

其他命令

在系统启动时，即进度条状态时，按下Esc键，会转换到类似CMD命令似的界面，会显示相关的启动日志，可以查看相关东西。
也可以通过查看/var/log/文件夹下的Xorg.*类文件来查看启动日志，比如
ls /var/log/Xorg.*
会显示相关的文件：
/var/log/Xorg.0.log
/var/log/Xorg.1.log
/var/log/Xorg.2.log
一般Xorg.0.log是最新的日志文件，可以通过时间知道。

重要参考

endly nvidia-xconfig
NVIDIA CUDA GETTING STARTED GUIDE FOR LINUX

之前都是在Centos6下进行CUDA开发，安装和配置CUDA环境请参考另外一篇博客（Linux（redhat）CUDA安装）。因为Centos7有较大变动，所以安装CUDA过程中遇到了一些麻烦，原因初步归结于OpenGL的安装。下面详细介绍安装配置过程

关闭防火墙和selinux和补充包和建立本地yum源

[root@localhost ~]# systemctl stop firewalld.service
[root@localhost ~]# systemctl disable firewalld.service
rm '/etc/systemd/system/dbus-org.fedoraproject.FirewallD1.service'
rm '/etc/systemd/system/basic.target.wants/firewalld.service'
[root@localhost ~]# vi /etc/selinux/config
[root@localhost ~]# cat /etc/selinux/config

# This file controls the state of SELinux on the system.
# SELINUX= can take one of these three values:
# enforcing - SELinux security policy is enforced.
# permissive - SELinux prints warnings instead of enforcing.
# disabled - No SELinux policy is loaded.
SELINUX=disabled
# SELINUXTYPE= can take one of three two values:
# targeted - Targeted processes are protected,
# minimum - Modification of targeted policy. Only selected processes are protected.
# mls - Multi Level Security protection.
SELINUXTYPE=targeted
重新启动继续配置
=============================================================================================
[root@localhost ~]# cd /etc/yum.repos.d/
[root@localhost yum.repos.d]# mkdir -p ./bak
[root@localhost yum.repos.d]# mv *.repo ./bak/
[root@localhost yum.repos.d]# vi local.repo
[root@localhost yum.repos.d]# cat local.repo
[local_server]
name=This is a local repo
baseurl=file:///mnt/iso
enabled=1
gpgcheck=0
[root@localhost yum.repos.d]# yum clean all
[root@localhost yum.repos.d]# mkdir -p /mnt/iso
[root@localhost yum.repos.d]# mount -o loop /cudainstall/CentOS\ 7\ x86_64.iso /mnt/iso/
[root@localhost yum.repos.d]# vi /etc/rc.local
[root@localhost yum.repos.d]# chmod +x /etc/rc.d/rc.local
[root@localhost yum.repos.d]# cat /etc/rc.lcoa
cat: /etc/rc.lcoa: Aucun fichier ou dossier de ce type
[root@localhost yum.repos.d]# cat /etc/rc.local
#!/bin/bash
# THIS FILE IS ADDED FOR COMPATIBILITY PURPOSES
#
# It is highly advisable to create own systemd services or udev rules
# to run scripts during boot instead of using this file.
#
# In contrast to previous versions due to parallel execution during boot
# this script will NOT be run after all other services.
#
# Please note that you must run 'chmod +x /etc/rc.d/rc.local' to ensure
# that this script will be executed during boot.
mount -o loop /cudainstall/CentOS\ 7\ x86_64.iso /mnt/iso/
touch /var/lock/subsys/local

[root@localhost yum.repos.d]# hostnamectl set-hostname cuda02
[root@cuda02 ~]# yum install kernel-devel gcc gcc-c++ yum install libX*

检查nouveau
[root@cuda01 ~]# lsmod |grep nouveau

如果有输出使用modprobe -r nouveau
[root@cuda01 ~]#init 3
==================================================================================================
安装cuda时一定不要安装OpenGL；切记，否则有可能安装完之后，图形化桌面启动不了，过程中会出现如下错误信息
[root@cuda02 cudainstall]# chmod +x cuda_7.0.28_linux.run
[root@cuda02 cudainstall]# ./cuda_7.0.28_linux.run
Do you accept the previously read EULA? (accept/decline/quit): accept
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 346.46? ((y)es/(n)o/(q)uit): y
Do you want to install the OpenGL libraries? ((y)es/(n)o/(q)uit) [ default is yes ]: n
Install the CUDA 7.0 Toolkit? ((y)es/(n)o/(q)uit): y
Enter Toolkit Location [ default is /usr/local/cuda-7.0 ]:
Do you want to install a symbolic link at /usr/local/cuda? ((y)es/(n)o/(q)uit): y
Install the CUDA 7.0 Samples? ((y)es/(n)o/(q)uit): y
Enter CUDA Samples Location [ default is /root ]:
Installing the NVIDIA display driver...
Installing the CUDA Toolkit in /usr/local/cuda-7.0 ...
Missing recommended library: libGLU.so

Installing the CUDA Samples in /root ...
Copying samples to /root/NVIDIA_CUDA-7.0_Samples now...
Finished copying samples.

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

Driver: Installed
Toolkit: Installed in /usr/local/cuda-7.0
Samples: Installed in /root, but missing recommended libraries

Please make sure that
- PATH includes /usr/local/cuda-7.0/bin
- LD_LIBRARY_PATH includes /usr/local/cuda-7.0/lib64, or, add /usr/local/cuda-7.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-7.0/bin
To uninstall the NVIDIA Driver, run nvidia-uninstall

Please see CUDA_Getting_Started_Guide_For_Linux.pdf in /usr/local/cuda-7.0/doc/pdf for detailed information on setting up CUDA.

Logfile is /tmp/cuda_install_2994.log

安装cuda驱动
[root@cuda02 cudainstall]# chmod +x cuda_7.0.28_linux.run
[root@cuda02 cudainstall]# ./cuda_7.0.28_linux.run
Do you accept the previously read EULA? (accept/decline/quit): accept
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 346.46? ((y)es/(n)o/(q)uit): y
Do you want to install the OpenGL libraries? ((y)es/(n)o/(q)uit) [ default is yes ]: n
Install the CUDA 7.0 Toolkit? ((y)es/(n)o/(q)uit): y
Enter Toolkit Location [ default is /usr/local/cuda-7.0 ]:
Do you want to install a symbolic link at /usr/local/cuda? ((y)es/(n)o/(q)uit): y
Install the CUDA 7.0 Samples? ((y)es/(n)o/(q)uit): y
Enter CUDA Samples Location [ default is /root ]:
Installing the NVIDIA display driver...
Installing the CUDA Toolkit in /usr/local/cuda-7.0 ...
Missing recommended library: libGLU.so

Installing the CUDA Samples in /root ...
Copying samples to /root/NVIDIA_CUDA-7.0_Samples now...
Finished copying samples.

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

Driver: Installed
Toolkit: Installed in /usr/local/cuda-7.0
Samples: Installed in /root, but missing recommended libraries

Please make sure that
- PATH includes /usr/local/cuda-7.0/bin
- LD_LIBRARY_PATH includes /usr/local/cuda-7.0/lib64, or, add /usr/local/cuda-7.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-7.0/bin
To uninstall the NVIDIA Driver, run nvidia-uninstall

Please see CUDA_Getting_Started_Guide_For_Linux.pdf in /usr/local/cuda-7.0/doc/pdf for detailed information on setting up CUDA.

Logfile is /tmp/cuda_install_2994.log

广播cuda的路径
[root@cuda02 cudainstall]# vi /etc/profile.d/cuda.sh
[root@cuda02 cudainstall]# cat /etc/profile.d/cuda.sh
export PATH=/usr/local/cuda-7.0/bin:$PATH
export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-7.0/lib64
[root@cuda02 cudainstall]# chmod 755 /etc/profile.d/cuda.sh
[root@cuda02 cudainstall]# source /etc/profile.d/cuda.sh
[root@cuda02 cudainstall]#
=================================================================================================================
恢复图形界面显示
[root@cuda02 cudainstall]# init 5
=================================================================================================================
安装vnc
[root@cuda02 cudainstall]# yum -y install tigervnc-server