介紹

CentOS下安裝nvidia+cuda+cudnn

NVIDIA驅(qū)動

1. 去NVIDIA官網(wǎng)下載合適版本驅(qū)動，

2. 安裝lspci娩井，使用下面命令逗嫡，找尋lspci，發(fā)現(xiàn)在pciutils中崖瞭，故安裝pciutils

   yum whatprovides */lspci
   yum install pciutils 
   

3. 檢查是否安裝了NVIDIA的GPU（硬件層面）：lspci | grep -i nvidia

4. 安裝kernel-devel和kernel-headers

   sudo yum install kernel-devel
   sudo yum install kernel-headers
   

5. 賦予運行權(quán)限chmod a+x NVIDIA-Linux-x86_64-410.78.run

6. 禁用nouveau

   # 打開配置文件：
   vi /usr/lib/modprobe.d/dist-blacklist.conf
   # 加上或修改 兩行
   blacklist nouveau
   options nouveau modeset=0
   查看nouveau是否禁用, 如果沒有輸出代表成功
   lsmod | grep nouveau 
   

7. 可選

   備份原來的 initramfs nouveau image鏡像
   mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img  
   創(chuàng)建新的 initramfs image鏡像
   dracut /boot/initramfs-$(uname -r).img  $(uname -r)  
   

8. 安裝

   運行命令
   sudo ./NVIDIA-Linux-x86_64-410.78.run
   如果報錯，則使用
   sudo ./Nvidia*.sh --kernel-source-path=/usr/src/kernels/按TAB補全
   

CUDA

1. 去這里選擇合適版本下載

2. 賦予運行權(quán)限chmod a+x cuda_10.0.130_410.48_linux.run

3. 安裝sudo ./cuda_10.0.130_410.48_linux.run

   1. 會先有個閱讀聲明撑毛，一直按D书聚，然后accept。 
   2. 很多選項
       Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 410.48?
       (y)es/(n)o/(q)uit: n
       Install the CUDA 10.0 Toolkit?
       (y)es/(n)o/(q)uit: y
       Enter Toolkit Location
       [ default is /usr/local/cuda-10.0 ]: 
       Do you want to install a symbolic link at /usr/local/cuda?
       (y)es/(n)o/(q)uit: y
       Install the CUDA 10.0 Samples?
       (y)es/(n)o/(q)uit: n
   
       選項install the OpenGL libraries,如果雙顯卡(集顯+獨顯)選擇n藻雌，如果只有獨顯可以選擇y雌续，如果雙顯卡選擇y的話，會出現(xiàn)黑屏或者循環(huán)登錄的問題胯杭，如果加了上面的參數(shù)就不會出現(xiàn)這個選項了驯杜。 
   
   3. 安裝過程結(jié)束后會有以下信息：
       ===========
       = Summary =
       ===========
       Driver:   Not Selected
       Toolkit:  Installed in /usr/local/cuda-10.0
       Samples:  Not Selected
   
       Please make sure that
       -   PATH includes /usr/local/cuda-10.0/bin
       -   LD_LIBRARY_PATH includes /usr/local/cuda-10.0/lib64, or, add /usr/local/cuda-10.0/lib64 to /etc/ld.so.conf and run ldconfig as root
   
       To uninstall the CUDA Toolkit, run the uninstall script in /usr/local/cuda-10.0/bin
   
       Please see CUDA_Installation_Guide_Linux.pdf in /usr/local/cuda-10.0/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 384.00 is required for CUDA 10.0 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_11482.log
   

4. 將cuda的bin文件和lib導出到系統(tǒng)環(huán)境中，版本不一樣則更換其中cuda-x.x

   export PATH="/usr/local/cuda-10.0/bin:$PATH" 
   export LD_LIBRARY_PATH="/usr/local/cuda-10.0/lib64:$LD_LIBRARY_PATH"
   或者
   vi ~/.bashrc
   export PATH="/usr/local/cuda-10.0/bin:$PATH" 
   export LD_LIBRARY_PATH="/usr/local/cuda-10.0/lib64:$LD_LIBRARY_PATH"
   source ~/.bashrc
   

5. 測試：如果下面測試的最后結(jié)果都是Result = PASS做个，說明CUDA安裝成功啦鸽心。

   1. 如果成功會輸出版本信息nvcc –V

   2. 編譯并測試設(shè)備 deviceQuery：

      cd /usr/local/cuda-9.2/samples/1_Utilities/deviceQuery
      sudo make
      ./deviceQuery
      

   3. 編譯并測試帶寬 bandwidthTest：

      cd ../bandwidthTest
      sudo make
      ./bandwidthTest
      

6. 其他
   所需的libcudart.so.8.0如果正確安裝的話，以下兩種方法同理：

   1. sudo ldconfig /usr/local/cuda-8.0/lib64
   2. export LD_LIBRARY_PATH=$LD_LIBRARY_PATH:/usr/local/cuda-8.0/lib64
   3. 如果仍然不行居暖，再嘗試執(zhí)行:

   export PATH=\$PATH:/usr/local/cuda-8.0/bin 
   export LIBRARY_PATH=$LIBRARY_PATH:/usr/local/cuda-8.0/lib64 
   source /etc/profile 
   

   1. 此時會顯示/sbin/ldconfig.real: /usr/local/cuda-8.0/lib64/libcudnn.so.6 不是符號連接顽频。不用擔心，這時已經(jīng)解決問題了太闺。

安裝cudnn

參考：https://docs.nvidia.com/deeplearning/sdk/cudnn-install/index.html

1. 去這里下載相應版本

2. 解壓
   tar -xzvf cudnn-10.0-linux-x64-v7.tgz

3. 復制

   cp include/cudnn.h /usr/local/cuda-10.0/include/
   cp lib64/libcudnn* /usr/local/cuda-10.0/lib64/
   

4. 授權(quán)
   sudo chmod a+r /usr/local/cuda-10.0/include/cudnn.h /usr/local/cuda-10.0/lib64/libcudnn*