====== Install CUDA Development Toolkit + SDK on Linux (Fedora 12+) ======

[[http://developer.download.nvidia.com/compute/cuda/3_2_prod/docs/Getting_Started_Linux.pdf|nVidia Linux Documentation]]
[[http://developer.nvidia.com/object/cuda_3_2_downloads.html#Linux|NVidia Developer Download page]]

  * download the packages
<code>
wget http://developer.download.nvidia.com/compute/cuda/3_2_prod/drivers/devdriver_3.2_linux_32_260.19.26.run
wget http://www.nvidia.com/object/thankyou.html?url=/compute/cuda/3_2_prod/toolkit/cudatoolkit_3.2.16_linux_32_fedora13.run
wget http://developer.download.nvidia.com/compute/cuda/3_2_prod/sdk/gpucomputingsdk_3.2.16_linux.run
</code>
  * install needed libraries
<code>
yum install -y  kernel-PAE-devel libXmu libXmu-devel freeglut-devel glut 
</code>
  * Remove the nvidia standard driver and install the dev version 
<code>
yum remove xorg-x11-drv-nvidia xorg-x11-drv-nvidia-libs xorg-x11-drv-nvidia-devel
init 3
sh devdriver_3.2_linux_32_260.19.26.run -a
init 5
</code>

  * install nVidia toolkit and sdk as root
<code>
sh cudatoolkit_3.2.16_linux_32_fedora13.run.sh -- auto
cat > /usr/local/etc/profile.d/cuda.sh << 'EOF'
export PATH=/usr/local/cuda/bin/:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda/lib/:$LD_LIBRARY_PATH
EOF
. /usr/local/etc/profile.d/cuda.sh
echo "/usr/local/cuda/lib/" > /etc/ld.so.conf.d/cuda.conf
ldconfig
</code>

  * the ''gpucomputingsdk'' thing have to be installed by each user on his own directory:
<code>
wget http://developer.download.nvidia.com/compute/cuda/3_2_prod/sdk/gpucomputingsdk_3.2.16_linux.run
sh gpucomputingsdk_3.2.16_linux.run 
</code>