博主装Ubuntu18.04主要是为了用于跑深度学习，所以我们先来搞搞gcc环境
第一步：安装多版本gcc、g++可切换
sudo apt-get install gcc-4.8 gcc-4.8-multilib
sudo apt-get install g++-4.8 g++-4.8-multilib
sudo apt-get install gcc- gcc--multilib
sudo apt-get install g++- g++--multilib
sudo apt-get install gcc- gcc--multilib
sudo apt-get install g++- g++--multilib
sudo apt-get install gcc- gcc--multilib
sudo apt-get install g++- g++--multilib
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-4.8
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-
sudo update-alternatives --install /usr/bin/gcc gcc /usr/bin/gcc-
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-4.8
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-
sudo update-alternatives --install /usr/bin/g++ g++ /usr/bin/g++-
切换版本命令
sudo update-alternatives --config gcc
sudo update-alternatives --config g++
根据自己想要的环境选择
第二步：准备安装显卡驱动和cuda8.0等相关文件
最新cuda8.0 及其补丁
cuda_8.0.61.2_linux.run
cuda_8.0.61_375.26_linux.run
最新支持cuda8.0的CUDNN
libcudnn7_7.1.4.18-1+cuda8.0_amd64.deb
libcudnn7-dev_7.1.4.18-1+cuda8.0_amd64.deb
libcudnn7-doc_7.1.4.18-1+cuda8.0_amd64.deb
cuda8.0 安装包解压文件
/001/InstallUtils.pm（从cuda_8.0.61.2_linux.run中解压出来的文件，后面会讲到）
第三步：安装显卡驱动
第三步：安装显卡驱动
- 1、开机 nomodeset 进入系统
- 开机进引导界面 第一项 按e 进入配置启动
- 在quiet splash - - -后加上 nomodeset
- 按F10 保存 进入系统
quiet splash - - -
quiet splash nomodeset
- 2、禁用系统自带NVIDIA驱动
sudo vim /etc/modprobe.d/blacklist.conf
# 在文件尾加入
blacklist nouveau
options nouveau modeset=
# 保存并退出 执行下面命令 更新引导
sudo update-initramfs –u
- 3、安装 NVIDIA 驱动
# 切换gcc 版本 到gcc- 以上 (使用高版本感觉会好一点)
# 查看支持的驱动版本
ubuntu-drivers devices
# 安装驱动
sudo ubuntu-drivers autoinstall
# 根据查询的版本安装比较保险 例如
sudo apt-get install nvidia-driver-
# 装驱动 需要关闭 安全启动
- 5、重启系统
sudo reboot
# 查看NVIDIA驱动 使用情况
nvidia-smi
- 6、安装cuda8.0
- 安装依赖
-
sudo apt-get install freeglut3-dev build-essential libx11-dev libxmu-dev libxi-dev libgl1-mesa-glx libglu1-mesa libglu1-mesa-dev
- 切换gcc版本到 4.8
-
sudo update-alternatives --config gcc
- 解压cuda8.0
-
sh cuda_8..61_375.26_linux.run --noexec --target
# 将runfile文件解压并且放到001文件夹中
# 将InstalUtil.pm 拷贝到 /etc/perl/
sudo cp InstalUtil.pm /etc/perl/- 安装cuda8.0及补丁
-
# 可选 加运行权限
chmod u+x cuda_8..61_375.26_linux.run
chmod u+x cuda_8.0.61.2_linux.run
# 运行
sudo ./chmod u+x cuda_8..61_375.26_linux.run Do you accept the previously read EULA?
accept/decline/quit: accept You are attempting to install on an unsupported configuration. Do you wish to continue?
(y)es/(n)o [ default is no ]: y Install NVIDIA Accelerated Graphics Driver for Linux-x86_64 375.26?
(y)es/(n)o/(q)uit: n 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 /home/deep ]: # 安装补丁
sudo ./cuda_8.0.61.2_linux.run- 添加环境变量
-
cd
vim .bashrc
# 添加到文件尾部
export PATH=/usr/local/cuda-8.0/bin:$PATH
export LD_LIBRARY_PATH=/usr/local/cuda-8.0/lib64$LD_LIBRARY_PATH
# 保存 退出
sudo su
source .bashrc- 重启系统
sudo reboot
- 安装cudnn
-
sudo dpkg -i libcudnn7_7.1.4.-+cuda8.0_amd64.deb
sudo dpkg -i libcudnn7-dev_7.1.4.-+cuda8.0_amd64.deb
sudo dpkg -i libcudnn7-doc_7.1.4.-+cuda8.0_amd64.deb- 查看cuda版本和cudnn版本
-
# cuda 版本
cat /usr/local/cuda/version.txt
# cudnn 版本
cat /usr/include/x86_64-linux-gnu/cudnn_v7.h | grep CUDNN_MAJOR -A- 编译
-
# 不用编译全部 只编译deviceQuery
cd /home/deep/NVIDIA_CUDA-.0_Samples/1_Utilities/deviceQuery
make- 测试
./deviceQuery # 出现显卡信息
./deviceQuery Starting... CUDA Device Query (Runtime API) version (CUDART static linking) Detected CUDA Capable device(s) Device : "GeForce GTX 1080"
CUDA Driver Version / Runtime Version 9.1 / 8.0
CUDA Capability Major/Minor version number: 6.1
Total amount of global memory: MBytes ( bytes)
() Multiprocessors, () CUDA Cores/MP: CUDA Cores
GPU Max Clock rate: MHz (1.73 GHz)
Memory Clock rate: Mhz
Memory Bus Width: -bit
L2 Cache Size: bytes
Maximum Texture Dimension Size (x,y,z) 1D=(), 2D=(, ), 3D=(, , )
Maximum Layered 1D Texture Size, (num) layers 1D=(), layers
Maximum Layered 2D Texture Size, (num) layers 2D=(, ), layers
Total amount of constant memory: bytes
Total amount of shared memory per block: bytes
Total number of registers available per block:
Warp size:
Maximum number of threads per multiprocessor:
Maximum number of threads per block:
Max dimension size of a thread block (x,y,z): (, , )
Max dimension size of a grid size (x,y,z): (, , )
Maximum memory pitch: bytes
Texture alignment: bytes
Concurrent copy and kernel execution: Yes with copy engine(s)
Run time limit on kernels: Yes
Integrated GPU sharing Host Memory: No
Support host page-locked memory mapping: Yes
Alignment requirement for Surfaces: Yes
Device has ECC support: Disabled
Device supports Unified Addressing (UVA): Yes
Device PCI Domain ID / Bus ID / location ID: / /
Compute Mode:
< Default (multiple host threads can use ::cudaSetDevice() with device simultaneously) > deviceQuery, CUDA Driver = CUDART, CUDA Driver Version = 9.1, CUDA Runtime Version = 8.0, NumDevs = , Device0 = GeForce GTX
Result = PASS
如果出现相应的显卡信息表示安装成功了。