First, you need to install the Nvidia binary drivers from RPMFusion. The below commands should handle that:
sudo rpm -Uvh http://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-18.noarch.rpm sudo rpm -Uvh http://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-18.noarch.rpm sudo yum -y install kmod-nvidia xorg-x11-drv-nvidia-devel
Then, you need to download and install the CUDA software from https://developer.nvidia.com/cuda-downloads
Install with the following command:
chmod ugo+x cuda_5.0.35_linux_64_fedora16-1.run sudo ./cuda_5.0.35_linux_64_fedora16-1.run -override compiler
Accept the license, answer n to installing the driver, y to the Toolkit, and y to Samples, I left installation paths at default.
Create a file /etc/ld.so.conf.d/cuda.conf that contains the below two lines:
/usr/local/cuda/lib64 /usr/local/cuda/lib
Download the latest source for cgminer and compile with the following command below:
CFLAGS=-I/usr/local/cuda/include LDFLAGS=-L/usr/lib64/nvidia ./configure --enable-cpumining && make && sudo make install
That should get you to the point of having cgminer compiled for Nvidia on Fedora 18. Feel free to post comments/corrections.
I also had to do `cd /usr/lib64/nvidia && ln -s libOpenCL.so.1 libOpenCL.so`. For some reason, installer didn’t create this one symbolic link.