GPU cuda驱动安装

cuda驱动官方下载地址(根据操作系统下载安装)：https://developer.nvidia.com/cuda-downloads?target_os=Linux&target_arch=x86_64&target_distro=CentOS&target_version=7&target_type=rpmlocal

下载安装cuda仓库环境 wget http://developer.download.nvidia.com/compute/cuda/11.0.1/local_installers/cuda-repo-rhel7-11-0-local-11.0.1_450.36.06-1.x86_64.rpm sudo rpm -i cuda-repo-rhel7-11-0-local-11.0.1_450.36.06-1.x86_64.rpm sudo yum clean all

 1、安装dkms

wget ftp://ftp.pbone.net/mirror/archive.fedoraproject.org/epel/7.2020-04-20/ppc64/Packages/d/dkms-2.6.1-1.el7.noarch.rpm yum install dkms-2.6.1-1.el7.noarch.rpm

2、安装cuda

sudo yum -y install nvidia-driver-latest-dkms cuda sudo yum -y install cuda-drivers

3、配置环境变量

[root@localhost ~]# vim /etc/profile export CUDA_HOME="/usr/local/cuda-11.0" export PATH=$CUDA_HOME/bin:$PATH export LD_LIBRARY_PATH=$CUDA_HOME/lib64:$LD_LIBRARY_PATH

4、启动驱动

450.36.06是NVIDIA的版本号，当你不知道的时候，进入/usr/src目录中，可以看到里面有nvidia文件夹，后缀就是其版本号

dkms install -m nvidia -v 450.36.06

5、检查驱动安装是否正常

如果报错(NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running)，检查是否有gpu，系统日志/var/log/messages的报错

[root@localhost ~]# nvidia-smi Fri Jul 3 15:27:16 2020 +-----------------------------------------------------------------------------+ | NVIDIA-SMI 450.36.06 Driver Version: 450.36.06 CUDA Version: 11.0 | |-------------------------------+----------------------+----------------------+ | GPU Name Persistence-M| Bus-Id Disp.A | Volatile Uncorr. ECC | | Fan Temp Perf Pwr:Usage/Cap| Memory-Usage | GPU-Util Compute M. | | | | MIG M. | |===============================+======================+======================| | 0 Tesla P40 Off | 00000000:04:00.0 Off | Off | | N/A 20C P8 9W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 1 Tesla P40 Off | 00000000:06:00.0 Off | Off | | N/A 25C P8 9W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 2 Tesla P40 Off | 00000000:07:00.0 Off | Off | | N/A 20C P8 9W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 3 Tesla P40 Off | 00000000:0D:00.0 Off | Off | | N/A 18C P8 8W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 4 Tesla P40 Off | 00000000:0E:00.0 Off | Off | | N/A 20C P8 9W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ | 5 Tesla P40 Off | 00000000:0F:00.0 Off | Off | | N/A 20C P8 8W / 250W | 0MiB / 24451MiB | 0% Default | | | | N/A | +-------------------------------+----------------------+----------------------+ +-----------------------------------------------------------------------------+ | Processes: | | GPU GI CI PID Type Process name GPU Memory | | ID ID Usage | |=============================================================================| | No running processes found | +-----------------------------------------------------------------------------+

相关连接：https://blog.csdn.net/hangzuxi8764/article/details/86572093