本文来源同事贾老板，本人负责编辑

1. 准备工作

• 准备安装包
  NVIDIA-Linux-x86_64-381.22.run # 最新显卡驱动
  cuda_8.0.61_375.26_linux.run # 最新CUDA安装包
  cudnn-8.0-linux-x64-v6.0.tgz # cudnn库v6.0
  cudnn-8.0-linux-x64-v5.1.tgz # cudnn库v5.1 (备用)

• 安装基础环境

# 检查显卡
$ lspci | grep -i vga 04:00.0 VGA compatible controller: NVIDIA Corporation Device 1b00 (rev a1) 
# 检查系统版本，确保系统支持(需要Linux-64bit系统) 
$ uname -m && cat /etc/*release x86_64 CentOS Linux release 7.2.1511 (Core)
# 安装GCC 
$ yum install gcc gcc-c++ 
# 安装Kernel Headers Packages 
$ yum install kernel-devel-$(uname -r) kernel-headers-$(uname -r)

2. 安装显卡驱动

$ sh NVIDIA-Linux-x86_64-381.22.run

• 开始安装

  
  

• Accept

  
  

• Building kernerl modules 安装

  
  

• 32bit兼容包选择, 这里要注意选择 No,不然后面就会出错。

  
  

• X- configuration 的选择页为 Yes

  
  

• 后面的都选择默认即可

3. 安装CUDA

• 开始安装

$ sh cuda_8.0.61_375.26_linux.run
# accept
-------------------------------------------------------------
  Do you accept the previously read EULA?
accept/decline/quit: accept
# no
Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 375.26?
(y)es/(n)o/(q)uit: n
-------------------------------------------------------------
# 后面的就都选yes或者default
Do you want to install the OpenGL libraries?
(y)es/(n)o/(q)uit [ default is yes ]: 
Do you want to run nvidia-xconfig?
This will update the system X configuration file so that the NVIDIA X driver
is used. The pre-existing X configuration file will be backed up.
This option should not be used on systems that require a custom
X configuration, such as systems with multiple GPU vendors.
(y)es/(n)o/(q)uit [ default is no ]: y
Install the CUDA 8.0 Toolkit?
(y)es/(n)o/(q)uit: y

Enter Toolkit Location
 [ default is /usr/local/cuda-8.0 ]: 

Do you want to install a symbolic link at /usr/local/cuda?
(y)es/(n)o/(q)uit: y

Install the CUDA 8.0 Samples?
(y)es/(n)o/(q)uit: y

Enter CUDA Samples Location
 [ default is /root ]: 

Installing the NVIDIA display driver...
The driver installation has failed due to an unknown error. Please consult the driver installation log located at /var/log/nvidia-installer.log.

===========
= Summary =
===========

Driver:   Not Selected
Toolkit:  Installed in /usr/local/cuda-8.0
Samples:  Installed in /root, but missing recommended libraries

Please make sure that
 -   PATH includes /usr/local/cuda-8.0/bin
 -   LD_LIBRARY_PATH includes /usr/local/cuda-8.0/lib64, or, add /usr/local/cuda-8.0/lib64 to /etc/ld.so.conf and run ldconfig as root

To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-8.0/bin

Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-8.0/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 361.00 is required for CUDA 8.0 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

Logfile is /tmp/cuda_install_192.log

• 验证安装结果

# 添加环境变量
# 在 ~/.bashrc的最后面添加下面两行
export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64:/usr/local/cuda-8.0/extras/CUPTI/lib64:$LD_LIBRARY_PATH
# 使生效
$ source ~/.bashrc
# 验证安装结果
$ nvcc -V
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2016 NVIDIA Corporation
Built on Tue_Jan_10_13:22:03_CST_2017
Cuda compilation tools, release 8.0, V8.0.61
$ nvidia-smi
+-----------------------------------------------------------------------------+
| NVIDIA-SMI 381.22                 Driver Version: 381.22                    |
|-------------------------------+----------------------+----------------------+
| GPU  Name        Persistence-M| Bus-Id        Disp.A | Volatile Uncorr. ECC |
| Fan  Temp  Perf  Pwr:Usage/Cap|         Memory-Usage | GPU-Util  Compute M. |
|===============================+======================+======================|
|   0  Graphics Device     Off  | 0000:02:00.0     Off |                  N/A |
| 21%   50C    P8    33W / 265W |      8MiB / 11172MiB |      0%      Default |
+-------------------------------+----------------------+----------------------+
                                                                               
+-----------------------------------------------------------------------------+
| Processes:                                                       GPU Memory |
|  GPU       PID  Type  Process name                               Usage      |
|=============================================================================|
|  No running processes found                                                 |
+-----------------------------------------------------------------------------+

4. 安装 cuDNN 库

$ tar -xvzf cudnn-8.0-linux-x64-v6.0.tgz
$ cp -P cuda/include/cudnn.h /usr/local/cuda-8.0/include
$ cp -P cuda/lib64/libcudnn* /usr/local/cuda-8.0/lib64
$ chmod a+r /usr/local/cuda-8.0/include/cudnn.h /usr/local/cuda-8.0/lib64/libcudnn*

5. 在 Docker 中使用 cuda

下面介绍了如何在docker中使用cuda，主要使用了nvidia-docker

• 安装docker

$ yum install docker
# 启动 Docker 服务，并将其设置为开机启动
$ systemctl start docker.service 
$ systemctl enable docker.service 

• 安装 nvidia-docker

# 1. Install nvidia-docker and nvidia-docker-plugin
$ wget -P /tmp https://github.com/NVIDIA/nvidia-docker/releases/download/v1.0.1/nvidia-docker-1.0.1-1.x86_64.rpm
$ sudo rpm -i /tmp/nvidia-docker*.rpm && rm /tmp/nvidia-docker*.rpm
$ sudo systemctl start nvidia-docker
# 2. 使用nvidia-docker启动容器
$ nvidia-docker run -it --name=CONTAINER_NAME -d DOCKER_IMAGE_NAME /bin/bash 
# 3. 进入容器
$ docker attach CONTAINER_NAME

注意

2 使用nvidia-docker启动容器

这里需要对image进行重新编译，添加nvidia-docker需要的Label，否则运行起来的容器会不能使用GPU。

6. TensorFlow 的 GPU 使用

下载安装GPU版本的TensorFlow，运行以下代码即可测试，无报错说明cuda安装成功

import tensorflow as tf

# 新建一个 graph.
a = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[2, 3], name='a')
b = tf.constant([1.0, 2.0, 3.0, 4.0, 5.0, 6.0], shape=[3, 2], name='b')
c = tf.matmul(a, b)
# 新建session with log_device_placement并设置为True.
sess = tf.Session(config=tf.ConfigProto(log_device_placement=True))
# 运行这个 op.
print sess.run(c)

7. Openface 的 GPU 使用

# 启动编译好的 openface 镜像 chinapnr/openface:0.3
$ nvidia-docker run -it -v /local-folder/:/root/openface_file/ --name=openface -d chinapnr/openface:0.3 /bin/bash
# 进入容器
$ docker exec -it openface /bin/bash
# 训练模型
$ cd /root/openface/training
$ ./main.lua -data /root/cuda-base/openface/CASPEAL-align-folder/ -device 1 -nGPU 1 -testing -alpha 0.2 -nEpochs 100 -epochSize 160 -peoplePerBatch 35 -imagesPerPerson 20 -retrain /root/openface/models/openface/nn4.small2.v1.t7 -modelDef /root/openface/models/openface/nn4.small2.def.lua -cache ../../cuda-base/openface/work