From 8fb933f3677cefd8eea0bfc0fd2f2e26576dc27d Mon Sep 17 00:00:00 2001 From: Stan Seibert Date: Thu, 5 Apr 2012 14:56:39 -0400 Subject: Add CUDA driver install directions and add matplotlib dependency --- doc/source/install.rst | 16 ++++++++++++---- 1 file changed, 12 insertions(+), 4 deletions(-) (limited to 'doc') diff --git a/doc/source/install.rst b/doc/source/install.rst index 2129404..571fb47 100644 --- a/doc/source/install.rst +++ b/doc/source/install.rst @@ -119,7 +119,7 @@ it should be executed as one line:: sudo apt-get install build-essential xorg-dev python-dev \ python-virtualenv python-numpy python-pygame libglu1-mesa-dev \ glutg3-dev cmake uuid-dev liblapack-dev mercurial git subversion \ - libboost-all-dev libatlas-base-dev + python-matplotlib libboost-all-dev libatlas-base-dev To be able to generate the documentation, we also need these tools:: @@ -132,9 +132,17 @@ Step 2: CUDA Toolkit and Driver CUDA requires the use of the official NVIDIA graphics driver, rather than the open source Nouveau driver that is included with Ubuntu. The -NVIDIA driver can be installed by - -.. warning:: FIGURE THIS OUT WITH FRESH UBUNTU INSTALL +NVIDIA driver can be installed by going to the `CUDA 4.1 Download Page +`_ and downloading the 64-bit Linux +developer drivers. (Newer drivers than those listed on this page will +also work.) To install the NVIDIA drivers, you will need to switch to a text console (Ctrl-Alt-F1) and shut down the X server:: + + # This next will kill everything running on your graphical desktop! + sudo service gdm stop + chmod +x NVIDIA-Linux-x86_64-285.05.33.run + sudo ./NVIDIA-Linux-x86_64-285.05.33.run + # Accept the license and pick the default option for the other questions + sudo service gdm start After the driver is installed, you need to download the CUDA 4.1 toolkit for Ubuntu Linux 11.04 (probably 64-bit) on `this page -- cgit