Ubuntu 14.04 LTS is out, loads of new features have been added. Here are some procedures I followed to install CUDA 6.0 on my DELL Inspiron.

First of all, Ubuntu need to be installed successfully, and the
necessary libs are also need to installed:

sudo apt-get install build-essential gcc-4.4 g++-4.4 libxi-dev libxmu-dev freeglut3-dev

Things you need to have before start the installation
process:
1. Latest NVIDIA graphic driver
(NVIDIA-Linux-x86_64-331.49.run)
2. CUDA 6.0 installation file (cuda_6.0.37_linux_64.run)

Now let's install the graphic driver

If you’re not sure your system type, press Ctrl+Alt+T to open
terminal and run:

uname -m

1. Find out your graphics card model

Use the lspci command to find out the model of your graphics
card

$ lspci -vnn | grep -i VGA -A 12
01:00.0 VGA compatible controller [0300]: NVIDIA Corporation GT218 [GeForce 210] [10de:0a65] (rev a2) (prog-if 00 [VGA controller])
Subsystem: ASUSTeK Computer Inc. Device [1043:8416]

2. Find out the right driver version for your graphics card

Visit http://www.nvidia.com/Download/index.aspx

Fill in the details about your graphics card and system and then
click Search. On the next page, it should tell you the correct
driver version with a download link and additional information.

For the above GeForce 210 card, it showed 331.67 as the correct
driver which can be downloaded from the website. However we shall
install the drivers from ppa to make things easier.

3. Setup the xorg-edgers ppa

The xorg-edgers ppa provides the very latest nvidia drivers. Run
the following commands to set it up.

$ sudo add-apt-repository ppa:xorg-edgers/ppa -y
$ sudo apt-get update

Now the ppa is setup and the package information is also updated.

 

4. Install the driver

Either you can install the driver directly by installing a single package containing "nvidia" and the major version number ( 173, 304, 310, 313, 319, 331, 334 or 337).

# 331 driver 
$ sudo apt-get install nvidia-331 
 # 334 driver 
$ sudo apt-get install nvidia-334 
# install the latest version 
$ sudo apt-get install nvidia-current

Or you can enable it from the "Additional Drivers" section. This is different on different Ubuntu flavors.

 

5. Verify the installation

The last thing to do is verify that the nvidia drivers are loaded
and working. Run the lspci command again and this time the kernel
driver should show nvidia

$ lspci -vnn | grep -i VGA -A 12

Check the last line which says "kernel driver in use: nvidia". This
shows that nvidia drivers are now in action. Also check hardware
acceleration with the glxinfo command

$ glxinfo | grep OpenGL | grep renderer
OpenGL renderer string: GeForce 210/PCIe/SSE2

The OpenGL renderer string should be anything other than "MESA". Then it indicates that the hardware drivers are being used for hardware acceleration.

6. Nvidia settings tool

Nvidia would install a gui tool called "Nvidia X Server Settings" somewhere in the menu. It can also be launched from the command line using the command "nvidia-settings". The tool shows miscellaneous information about the graphics card and the monitor connected, and also allows to configure various options.

7. Removing the drivers

Incase anything goes wrong after the installation, like you are not
able to boot Ubuntu, then try removing the Nvidia drivers.

Boot into the recovery console from the grub menu and then issue
the following commands

# remount root file system as writable 
$ mount -o remount,rw / 
 # remove all nvidia packages
 $ apt-get purge nvidia*

Additional Notes

Many tutorials out there talk about blacklisting the nouveau driver. This is no longer necessary, since the nvidia driver would blacklist nouveau itself. This can be verified by checking the contents of nvidia driver files in the the modprobe.d directory.

$ grep 'nouveau' /etc/modprobe.d/* | grep nvidia 
/etc/modprobe.d/nvidia-331_hybrid.conf:blacklist nouveau
/etc/modprobe.d/nvidia-331_hybrid.conf:blacklist lbm-nouveau
/etc/modprobe.d/nvidia-331_hybrid.conf:alias nouveau off
/etc/modprobe.d/nvidia-331_hybrid.conf:alias lbm-nouveau off
/etc/modprobe.d/nvidia-graphics-drivers.conf:blacklist nouveau
/etc/modprobe.d/nvidia-graphics-drivers.conf:blacklist lbm-nouveau
/etc/modprobe.d/nvidia-graphics-drivers.conf:alias nouveau off
/etc/modprobe.d/nvidia-graphics-drivers.conf:alias lbm-nouveau off

Note that the files "nvidia-331_hybrid.conf" and "nvidia-graphics-drivers.conf" have blacklisted nouveau.

To check information about the nvidia driver module, use the commands lsmod, modprobe and modinfo

# check that nvidia kernel module is loaded or not 
$ lsmod | grep nvidia 
nvidia              10699336  49
drm 302817 2 nvidia # find the real name of the nvidia module
$ modprobe -R nvidia
nvidia_331 # details about the nvidia_331 module
$ modinfo nvidia_331
filename: /lib/modules/3.13.0-24-generic/updates/dkms/nvidia_331.ko
alias: char-major-195-*
version: 331.67
supported: external
license: NVIDIA
.....

The kernel module file for the nvidia driver is located at "/lib/modules/3.13.0-24-generic/updates/dkms/nvidia_331.ko".

Note that it is a "dkms" module which means, its loaded dynamically. Due to this the grub screen, the Ubuntu/Kubuntu splash screens would have a low resolution since at that time the nvidia drivers are not in effect, and whatever resolution is available via the VESA extensions, are used.

8. Another Option

$ sudo add-apt-repository ppa:xorg-edgers/ppa

$ sudo apt-get update

$ sudo apt-get install nvidia-340

 
Then remove it
$ sudo apt-get remove nvidia-340
 

9. Install CUDA

$ sudo apt install
nvidia-cuda-toolkit
 
 
One more tip:
Highlight CUDA syntax in Sublime
cd ~/.config/sublime-text-2/Packages/User
git clone
git://github.com/harrism/sublimetext-cuda-cpp.git

Install CUDA 6.0 on Ubuntu 14.04 LTS的更多相关文章

  1. How to Install LibreOffice 6.0 on Ubuntu 16.04 LTS +

    1. Add the LibreOffice 6.0 PPA The LibreOffice Fresh PPA is maintained by LibreOffice. It provides l ...

  2. Zabbix 3.0 for Ubuntu 14.04 LTS 安装

    准备工作 apt-get install gettextapt-get install unzipapt-get install rar一.安装主程序 代码: 全选wget http://repo.z ...

  3. 【Android 系统开发】CyanogenMod 13.0 源码下载 编译 ROM 制作 ( 手机平台 : 小米4 | 编译平台 : Ubuntu 14.04 LTS 虚拟机)

                 分类: Android 系统开发(5)                                              作者同类文章X 版权声明:本文为博主原创文章 ...

  4. Install Cocos2d-x v3.3 on Ubuntu 14.04 & Ubuntu 14.10(转)

    Install Cocos2d-x v3.3 on Ubuntu 14.04 & Ubuntu 14.10 1 get the source code sudo apt-get install ...

  5. Ubuntu 14.04 LTS 安装Docker

    Docker官方是有很详细的安装文档(https://docs.docker.com/engine/installation/ubuntulinux/),这里做了一个Ubuntu 14.04 LTS中 ...

  6. 完整卸载 kubuntu-desktop from Ubuntu 14.04 LTS

    系统 ubuntu 14.04 LTS 64Bit 目的:卸载kubuntu-desktop 方法一: sudo apt-get remove libkde3support4 k3b-data ntr ...

  7. Ubuntu 14.04 LTS 下 android 2.3.5 源码编译过程

    Ubuntu 14.04 LTS 下 android 2.3.5 源码编译过程   在新的Ubuntu 64位系统下去编译早期的安卓源码是会出现很多问题的,因为64位系统在安装完成后,很多32位的兼容 ...

  8. Ubuntu 14.04 LTS 64bit 编译SDL的问题

    http://blog.csdn.net/jhting/article/details/38523945 Ubuntu 14.04 LTS 64bit 编译SDL的问题 分类: C/C++2014-0 ...

  9. 【转】基于Ubuntu 14.04 LTS编译Android4.4.2源代码

    原文网址:http://blog.csdn.net/gobitan/article/details/24367439 基于Ubuntu 14.04 LTS编译Android4.4.2源代码       ...

随机推荐

  1. 飞行姿态角度表示: heading pitch roll

    //创建初始化摄像机视图 var initialPosition=new Cesium.Cartesian3.fromDegrees(-73.998114468289017509, 40.674512 ...

  2. Truffle测试框架

    Truffle测试框架 2018年06月08日 19:01:19 tianlongtc 阅读数 1000   Truffle 有一个标准的自动化测试框架,让你可以非常方便地测试您的合约.这个框架允许您 ...

  3. kotlin委托属性

    fun main(arg: Array<String>) { val myClass1 = myClass1() myClass1.name="mycalsss1" v ...

  4. Qt编写安防视频监控系统7-全屏切换

    一.前言 全屏切换这个功能点属于简单的,一般会做到右键菜单中,也提供了快捷键比如alt+enter来触发,恢复全屏则按esc即可,全屏处理基本上都是隐藏通道面板以外的窗体,保持最大化展示,由于采用了模 ...

  5. 【计算机视觉】关于OpenCV中GPU配置编译的相关事项

    [计算机视觉]关于OpenCV中GPU配置编译的相关事项 标签(空格分隔): [计算机视觉] 前一段发现了OpenCV中关于GPU以及opencl的相关知识,打算升级一下对OpenCV的使用,但是发现 ...

  6. Java学习笔记-包,classpath,import,jar

    这里介绍Java的包,classpath,import和jar 包(package) 对类文件进行分类管理 给类提供多层命名空间 写在程序文件的第一行 类名的全称的是 包名.类名 包也是一种封装形式 ...

  7. 使用Dreamweaver制作简单网站(二)

    继续上周没完成的 一.新建iframe.css 1.点击文件-选择新建-css 2.ctrl+s保存为iframe.css 在style文件夹下. 3.回到main.html 右键选择-附加样式表,选 ...

  8. 内存泄漏之malloc替换方法

    //内存泄漏之malloc替换方法 //内存泄漏之malloc替换方法#include "stdio.h"#include "stdlib.h" /*文件路径名 ...

  9. Python+requests 发送简单请求--》获取响应状态--》获取请求响应数据

    Python+requests 发送简单请求-->获取响应状态-->获取请求响应数据 1.环境:安装了Python和vscode编译器(Python自带的编译器也ok).fiddler抓包 ...

  10. 使用pycharm开发web——django2.1.5(四)视图和模板相关

    刘老师说这块很重要..... 应该是很重要,大概看了一下,这里面关于views中函数作用,大概看来可能就是相应请求,传入数据和跳转,基本功能上貌似这些框架都差不多吧(其实我并没用过3个框架以上.... ...