CUDA10.2显卡驱动安装
0 ubuntu电脑配置查看命令
0.1 查看GPU 型号
lspci |grep VGA
部分结果显示:
NVIDIA Corporation GP102 [GeForce GTX 1080 Ti] (rev a1)
0.2 查看CPU 型号
cat /proc/cpuinfo
部分结果显示:
Intel(R) Core(TM) i7-6850K CPU @ 3.60GHz
1 nvidia驱动
官网链接:https://www.nvidia.cn/Download/index.aspx?lang=cn
或者https://www.nvidia.cn/geforce/drivers/
2 CUDA下载链接
官方CUDA:https://developer.nvidia.com/cuda-toolkit-archive
CUDA10.2传送门
推荐run安装
3 可直接命令行安装
3.1 run安装
wget https://developer.download.nvidia.com/compute/cuda/10.2/Prod/local_installers/cuda_10.2.89_440.33
.01_linux.run
3.2 deb安装:
wget https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/cuda-ubuntu1804.pin
sudo mv cuda-ubuntu1804.pin /etc/apt/preferences.d/cuda-repository-pin-600
wget https://developer.download.nvidia.com/compute/cuda/10.2/Prod/local_installers/cuda-repo-ubuntu1804-10-2-local-10.2.89-440.33.01_1.0-1_amd64.deb
sudo dpkg -i cuda-repo-ubuntu1804-10-2-local-10.2.89-440.33.01_1.0-1_amd64.deb
sudo apt-key add /var/cuda-repo-10-2-local-10.2.89-440.33.01/7fa2af80.pub
sudo apt-get update
sudo apt-get -y install cuda
4 配置
打开文件
sudo gedit ~/.bashrc
在文件中加入
export PATH="/usr/local/cuda/bin/:$PATH"
export PATH=/usr/local/cuda/bin${PATH:+:${PATH}}
export LD_LIBRARY_PATH=/usr/local/cuda/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
export CUDA_HOME=/usr/local/cuda
刷新环境
source ~/.bashrc
5 ubuntu删除旧版本的CUDA
sudo apt-get --purge remove cuda-*
#重启
reboot
6 error code(1)
cd /var/lib/dpkg
sudo mv info info.baksudo
mkdir info
7 Failed to initialize NVML: Driver/library version mismatch
原因:NVIDIA 内核驱动版本与系统驱动不一致;采用方案:删除系统驱动,安装与内核驱动一致的系统驱动版本。
(1)查看NVIDIA 内核驱动版本和查看系统驱动:
# 1 查看NVIDIA 内核驱动版本
cat /proc/driver/nvidia/version
# 2 查看系统驱动
## cat /var/log/dpkg.log | grep nvidia
(2)卸载电脑驱动
sudo apt-get purge nvidia*
(3)把显卡驱动加入ppa(个人软件包文档,仅支持Ubuntu)
sudo add-apt-repository ppa:graphics-drivers
sudo apt-get update
(4)重装510版本驱动
sudo apt-get install nvidia-510 nvidia-settings nvidia-prime
(5)查看解决方案是否生效
nvidia-smi
(6)为防止再次出现这样的问题,禁用当前版本的本地更新
sudo apt-mark hold nvidia-510
8 更新系统遇到的问题
(1)W: An error occurred during the signature verification. The repository is not updated and the previous index files will be used. GPG error: http://mirrors.ustc.edu.cn/ros/ubuntu bionic InRelease: The following signatures were invalid: EXPKEYSIG F42ED6FBAB17C654 Open Robotics <info@osrfoundation.org>
cd /var/lib/apt
sudo mv lists oldlist
sudo mkdir -p lists/partial
sudo apt-get clean
sudo apt-get update
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys F42ED6FBAB17C654
(2)W: Target Sources (main/source/Sources) is configured multiple times in /etc/apt/sources.list:2 and /etc/apt/sources.list:7
cd /etc/apt/sources.list.d
rm source.list
(3)W: GPG error: https://mirrors.aliyun.com/nvidia-cuda/ubuntu1804/x86_64 InRelease: The following signatures couldn't be verified because the public key is not available: NO_PUBKEY A4B469963BF863CC
sudo apt-key adv --keyserver keyserver.ubuntu.com --recv-keys A4B469963BF863CC
(4)W: Target Sources (main/source/Sources) is configured multiple times in /etc/apt/sources.list:2 and /etc/apt/sources.list:7
deb http://mirrors.aliyun.com/ubuntu/ bionic main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ bionic-security main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ bionic-updates main restricted universe multiverse
deb http://mirrors.aliyun.com/ubuntu/ bionic-backports main restricted universe multiverse
(5) Errors were encountered while processing: rsyslog
cd /var/lib/dpkg
sudo mv info info.bak
sudo mkdir info
sudo apt-get update
9 An NVIDIA kernel module ‘nvidia-drm’ appears to already be loaded in your kernel.
# 1. 获得超级用户权限
sudo -i
# 2. 关闭所有使用 GPU 的进程
sudo systemctl isolate multi-user.target
sudo modprobe -r nvidia-drm
# 3. 重装 NVIDIA 驱动
sh ./NVIDIA-Linux-x86_64-XX.sh
# 4. 打开使用GPU的进程
sudo systemctl set-default multi-user.target
# 5. 重新启动
sudo reboot