配置 NVIDIA DeepStream SDK on Tesla V2.0

NVIDIA DeepStream SDK on Tesla V2.0的环境要求为：

• Ubuntu 16.04

• CUDA 9.2

  官方安装指导为：https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html

  在命令行中应做如下选择：（由于已经安装了不同版本的cuda，并没有选择直接生成软链接。）

  Do you accept the previously read EULA?

  accept/decline/quit: accept

  
  

  Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 396.26?

  (y)es/(n)o/(q)uit: n

  
  

  Install the CUDA 9.2 Toolkit?

  (y)es/(n)o/(q)uit: y

  
  

  Enter Toolkit Location

   [ default is /usr/local/cuda-9.2 ]: 

  
  

  Do you want to install a symbolic link at /usr/local/cuda?

  (y)es/(n)o/(q)uit: n

  
  

  Install the CUDA 9.2 Samples?

  (y)es/(n)o/(q)uit: y

  
  

  Enter CUDA Samples Location

   [ default is /home/deepstream ]: 

  
  

  Installing the CUDA Toolkit in /usr/local/cuda-9.2 ...

  Missing recommended library: libXmu.so

  
  

  Installing the CUDA Samples in /home/deepstream ...

  Copying samples to /home/deepstream/NVIDIA_CUDA-9.2_Samples now...

  Finished copying samples.

  
  

  ===========

  = Summary =

  ===========

  
  

  Driver:   Not Selected

  Toolkit:  Installed in /usr/local/cuda-9.2

  Samples:  Installed in /home/deepstream, but missing recommended libraries

  
  

  Please make sure that

   -   PATH includes /usr/local/cuda-9.2/bin

   -   LD_LIBRARY_PATH includes /usr/local/cuda-9.2/lib64, or, add /usr/local/cuda-9.2/lib64 to /etc/ld.so.conf and run ldconfig as root

  
  

  To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-9.2/bin

  
  

  Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-9.2/doc/pdf for detailed information on setting up CUDA.

  
  

  ***WARNING: Incomplete installation! This installation did not install the CUDA Driver. A driver of version at least 384.00 is required for CUDA 9.2 functionality to work.

  To install the driver using this installer, run the following command, replacing <CudaInstaller> with the name of this run file:

      sudo <CudaInstaller>.run -silent -driver

  然后sudo vim .bashrc，按照提示：

   -   PATH includes /usr/local/cuda-9.2/bin

   -   LD_LIBRARY_PATH includes /usr/local/cuda-9.2/lib64

  再source .bashrc完成更新。如果nvcc --version则可以看到安装成功：

  nvcc: NVIDIA (R) Cuda compiler driver

  Copyright (c) 2005-2018 NVIDIA Corporation

  Built on Wed_Apr_11_23:16:29_CDT_2018

  Cuda compilation tools, release 9.2, V9.2.88

• TENSORT 4.0

  • 注意需要cuDNN 7.1.3 环境，但是官网上我只找到有7.1.2和7.1.4？那就赌一把7.1.4向下兼容吧

    我选择了：

    Download cuDNN v7.1.4 (May 16, 2018), for CUDA 9.2

    cuDNN v7.1.4 Library for Linux

  在官网https://developer.nvidia.com/nvidia-tensorrt-download 上下载tar包，然后

  tar -xf TensorRT-4.0.1.6.Ubuntu-16.04.4.x86_64-gnu.cuda-9.2.cudnn7.1.tar

  vim ~/.bashrc

  export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:~/TensorRT-4.0.1.6/lib

  source ~/.bashrc

  
  

  ~/TensorRT-4.0.1.6/python$ sudo pip2 install tensorrt-4.0.1.6-cp27-cp27mu-linux_x86_64.whl

  随后安装Python UFF package 

  
  

  ~/TensorRT-4.0.1.6/uff$ sudo pip2 install uff-0.4.0-py2.py3-none-any.whl

  再安装Python graphsurgeon package 

  
  

  ~/TensorRT-4.0.1.6/graphsurgeon$ sudo pip2 install graphsurgeon-0.2.0-py2.py3-none-any.whl

  
  

• OpenCV 3.4.x 

• Gstreamer 1.8.3

  sudo apt-get install libgstreamer1.0-0 gstreamer1.0-plugins-base gstreamer1.0-plugins-good gstreamer1.0-plugins-bad gstreamer1.0-plugins-ugly gstreamer1.0-libav gstreamer1.0-doc gstreamer1.0-tools

  如果出现错误：dpkg: error processing package initramfs-tools (--configure):

   subprocess installed post-installation script returned error exit status 1

  Errors were encountered while processing:

   initramfs-tools的错误，不要慌，执行：

  sudo rm -f /var/lib/dpkg/info/initramfs-tools.post*
  sudo rm -f /var/lib/dpkg/info/initramfs-tools.pre*
  
  sudo rm -f /var/lib/dpkg/info/bcmwl-kernel-source.post*
  sudo rm -f /var/lib/dpkg/info/bcmwl-kernel-source.pre*
  
  sudo dpkg --configure -a

NVIDIA driver396+

sudo apt install nvidia-396

如果出现Failed to initialize NVML: Driver/library version mismatch，不要慌，重启即可。

（未完待续）