博客
关于我
强烈建议你试试无所不能的chatGPT,快点击我
Fedora 23+CUDA 8.0+ GTX970 安装
阅读量:4549 次
发布时间:2019-06-08

本文共 3230 字,大约阅读时间需要 10 分钟。

 

PRE-INSTALLATION ACTIONS

Some actions must be taken before the CUDA Toolkit and Driver can be installed on

Linux:
‣ Verify the system has a CUDA-capable GPU.
‣ Verify the system is running a supported version of Linux.
‣ Verify the system has gcc installed.
‣ Verify the system has the correct kernel headers and development packages
installed.
‣ Download the NVIDIA CUDA Toolkit.
‣ Handle conflicting installation methods.

lspci | grep -i nvidiauname -m && cat /etc/*releasegcc --versionsudo dnf install kernel-devel-$(uname -r) kernel-headers-$(uname -r)

As follows:

[root@localhost ~]# lspci | grep -i nvidia01:00.0 VGA compatible controller: NVIDIA Corporation GM204 [GeForce GTX 970] (rev a1)01:00.1 Audio device: NVIDIA Corporation GM204 High Definition Audio Controller (rev a1)[root@localhost ~]# uname -m && cat /etc/*releasex86_64Fedora release 23 (Twenty Three)NAME=FedoraVERSION="23 (Workstation Edition)"ID=fedoraVERSION_ID=23PRETTY_NAME="Fedora 23 (Workstation Edition)"ANSI_COLOR="0;34"CPE_NAME="cpe:/o:fedoraproject:fedora:23"HOME_URL="https://fedoraproject.org/"BUG_REPORT_URL="https://bugzilla.redhat.com/"REDHAT_BUGZILLA_PRODUCT="Fedora"REDHAT_BUGZILLA_PRODUCT_VERSION=23REDHAT_SUPPORT_PRODUCT="Fedora"REDHAT_SUPPORT_PRODUCT_VERSION=23PRIVACY_POLICY_URL=https://fedoraproject.org/wiki/Legal:PrivacyPolicyVARIANT="Workstation Edition"VARIANT_ID=workstationFedora release 23 (Twenty Three)Fedora release 23 (Twenty Three)[root@localhost ~]# gcc --versiongcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6)Copyright (C) 2015 Free Software Foundation, Inc.This is free software; see the source for copying conditions.  There is NOwarranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.[root@localhost ~]# uname -r4.2.3-300.fc23.x86_64[root@localhost ~]# gcc --versiongcc (GCC) 5.3.1 20160406 (Red Hat 5.3.1-6)Copyright (C) 2015 Free Software Foundation, Inc.This is free software; see the source for copying conditions.  There is NOwarranty; not even for MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.[root@localhost ~]# sudo dnf install kernel-devel-$(uname -r) kernel-headers-$(uname -r)

 

Disable nouveau  and

[root@localhost ~]# echo "blacklist nouveau" >> /etc/modprobe.d/blacklist.conf[root@localhost ~]# vi /etc/sysconfig/grub[root@localhost ~]# grub2-mkconfig -o /boot/grub2/grub.cfg[root@localhost ~]# vi /etc/dnf/dnf.conf[root@localhost ~]# mv /boot/initramfs-$(uname -r).img /boot/initramfs-$(uname -r)-nouveau.img[root@localhost ~]# dracut /boot/initramfs-$(uname -r).img $(uname -r)

 Reboot to runlevel 3

[root@localhost ~]# systemctl set-default multi-user.targetRemoved symlink /etc/systemd/system/default.target.Created symlink from /etc/systemd/system/default.target to /usr/lib/systemd/system/multi-user.target.[root@localhost ~]# reboot

 

1.开机按Ctrl+Alt+F5 进入字符界面

2.在root 下输入 yum list NVIDIA  查找安装的驱动名称

3.yum remove name

4.reboot

 

systemctl set-default graphical.target

systemctl set-default multi-user.target

telinit 5

lsmod | grep nouveau

sudo sh cuda_8.0.27_linux.run --tmpdir=/opt/temp/ --no-opengl-libs

 

echo 'blacklist nouveau' >> /etc/modprobe.d/disable-nouveau.confecho 'nouveau modeset=0' >> /etc/modprobe.d/disable-nouveau.conf

转载于:https://www.cnblogs.com/huadongw/p/6517091.html

你可能感兴趣的文章
(转)使用 python Matplotlib 库绘图
查看>>
进程/线程切换原则
查看>>
正则表达式语法
查看>>
20165301 2017-2018-2 《Java程序设计》第四周学习总结
查看>>
Vue的简单入门
查看>>
使用最快的方法计算2的16次方是多少?
查看>>
urllib 中的异常处理
查看>>
【SQL Server高可用性】高可用性概述
查看>>
通过SQL Server的扩展事件来跟踪SQL语句在运行时,时间都消耗到哪儿了?
查看>>
SQL优化:重新编译存储过程和表
查看>>
PCB“有铅”工艺将何去何从?
查看>>
Solr环境搭建
查看>>
ASP.NET的URL过滤
查看>>
自己写的Web服务器
查看>>
自定义定时组件
查看>>
2-素数打比表
查看>>
性能测试
查看>>
浅谈 Python 的 with 语句
查看>>
使用koa+angular+mysql 完成了一个企业站
查看>>
SQL使用范例
查看>>