官术网_书友最值得收藏!

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:

  1. 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
    
  2. 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'
    
  3. 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
    
  4. 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
    
  5. Run the following command to make the variables take effect:
    source ~/.bashrc
    ldconfig
    
  6. Install pyrit dependencies:
    apt-get install libssl-dev python-dev python-scapy
    
  7. 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
    
  8. 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
主站蜘蛛池模板: 楚雄市| 余姚市| 如皋市| 浙江省| 永年县| 亚东县| 金乡县| 枣强县| 东方市| 如东县| 张家界市| 页游| 营山县| 天镇县| 昌乐县| 洞口县| 维西| 团风县| 静乐县| 仙桃市| 古丈县| 水富县| 衡水市| 峨边| 杭锦后旗| 阜新| 通道| 永州市| 永福县| 赣州市| 赤壁市| 南平市| 临海市| 县级市| 英吉沙县| 屏东县| 化隆| 竹山县| 渑池县| 嵊州市| 土默特左旗|