Installing and configuring nVidia video card drivers
In this recipe, we will embrace Compute Unified Device Architecture (CUDA), the nVidia parallel computing architecture. The first step will be the installation of the nVidia developer display driver followed by the installation of the CUDA toolkit. This will give us dramatic increases in computer performance with the power of the GPU which will be used in scenarios like password cracking.
Note
For more information about CUDA, please visit their website at http://www.nvidia.com/object/cuda_home_new.html.
Getting ready
An Internet connection is required to complete this recipe.
The preparation of kernel headers is needed before starting this task, which is explained in the Preparing kernel headers recipe at the beginning of this chapter.
In order to accomplish the installation of the nVidia driver, the X session needs to be shut down.
How to do it...
Let's begin the process of installing and configuring the nVidia video card drivers:
- Download the nVidia developer display driver according to your CPU architecture:
cd /tmp/ wget http://developer.download.nvidia.com/compute/cuda/4_1/rel/drivers/NVIDIA-Linux-x86_64-285.05.33.run
- Install the driver:
chmod +x NVIDIA-Linux-x86_64-285.05.33.run ./NVIDIA-Linux-x86_64-285.05.33.run –kernel-source-path='/usr/src/linux'
- Download the CUDA toolkit:
wget http://developer.download.nvidia.com/compute/cuda/4_1/rel/toolkit/cudatoolkit_4.1.28_linux_64_ubuntu11.04.run - Install the CUDA toolkit to
/opt:chmod +x cudatoolkit_4.1.28_linux_64_ubuntu11.04.run ./cudatoolkit_4.1.28_linux_64_ubuntu11.04.runConfigure the environment variables required for nvcc to work: echo PATH=$PATH:/opt/cuda/bin >> ~/.bashrc echo LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/opt/cuda/lib >> ~/.bashrc echo export PATH >> ~/.bashrc echo export LD_LIBRARY_PATH >> ~/.bashrc
- Run the following command to make the variables take effect:
source ~/.bashrc ldconfig
- Install pyrit dependencies:
apt-get install libssl-dev python-dev python-scapy - Download and install the GPU powered tool, pyrit:
svn co http://pyrit.googlecode.com/svn/trunk/ pyrit_src cd pyrit_src/pyrit python setup.py build python setup.py install
- Finally, add the nVidia GPU module to pyrit:
cd /tmp/pyrit_src/cpyrit_cuda python setup.py build python setup.py install
Note
To verify if nvcc is installed correctly, we issue the following command:
nvcc –V
To perform a benchmark, we simply type the following command:
pyrit benchmark