Skip to content

Instantly share code, notes, and snippets.

@cshen
Created September 9, 2013 08:26
Show Gist options
  • Save cshen/6492913 to your computer and use it in GitHub Desktop.
Save cshen/6492913 to your computer and use it in GitHub Desktop.
Installing CUDA Toolkit 5.5 on Ubuntu 12.10 Linux
Installing CUDA Toolkit 5.5 on Ubuntu 12.10 Linux
The following explains how to install CUDA Toolkit 5.5 on 64-bit Ubuntu 12.10 Linux. I have tested it on a self-assembled desktop with AMD Phenom II X4 CPU, 4GB RAM, 500GB hard drive, 650W power supply, and NVIDIA GeForce GTX 550 Ti graphics card. The instruction assumes you have the necessary CUDA compatible hardware support. Depending on your system configuration, your mileage may vary.
CUDA Repository
Retrieve the CUDA repository package for Ubuntu 12.10 from the CUDA download site and install it in a terminal.
$ sudo dpkg -i cuda-repo-ubuntu1210_5.5-0_amd64.deb
$ sudo apt-get update
Linux Kernel Header
Then you need to install the necessary Linux kernel headers.
$ sudo apt-get install linux-headers-`uname -r`
Proprietary Video Driver
The built-in nouveau video driver in Ubuntu is incompatible with the CUDA Toolkit, and you have to replace it with the proprietary NVIDIA driver.
$ sudo apt-get remove --purge \
xserver-xorg-video-nouveau
$ sudo apt-get install nvidia-settings \
nvidia-current-dev nvidia-modprobe
You can reboot the system afterwards and verify the driver installation with the nvidia-settings utility.
CUDA Toolkit
Then you can install the CUDA Toolkit using apt-get.
$ sudo apt-get install cuda
Environment Variables
As part of the CUDA environment, you should add the following in the .bashrc file of your home folder.
export CUDA_HOME=/usr/local/cuda-5.5
export LD_LIBRARY_PATH=${CUDA_HOME}/lib64
PATH=${CUDA_HOME}/bin:${PATH}
export PATH
CUDA SDK Samples
Now you can copy the SDK samples into your home directory, and proceed with the build process.
$ cuda-install-samples-5.5.sh ~
$ cd ~/NVIDIA_CUDA-5.5_Samples
$ make
If everything goes well, you should be able to verify your CUDA installation by running the deviceQuery sample in bin/linux/release.
Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment