Instantly share code, notes, and snippets.

Embed
What would you like to do?
Install NVIDIA Driver and CUDA on Ubuntu / CentOS / Fedora Linux OS

In this article, I will share some of my experience on installing NVIDIA driver and CUDA on Linux OS. Here I mainly use Ubuntu as example. Comments for CentOS/Fedora are also provided as much as I can.

Table of Contents

Table of contents generated with markdown-toc

Install NVIDIA Graphics Driver via apt-get

In Ubuntu systems, drivers for NVIDIA Graphics Cards are already provided in the official repository. Installation is as simple as one command.

For ubuntu 14.04.5 LTS, the latest version is 352. To install the driver, excute sudo apt-get nvidia-352 nvidia-modprobe, and then reboot the machine.

For ubuntu 16.04.3 LTS, the latest version is 375. To install the driver, excute sudo apt-get nvidia-375 nvidia-modprobe, and then reboot the machine.

The nvidia-modprobe utility is used to load NVIDIA kernel modules and create NVIDIA character device files automatically everytime your machine boots up.

It is recommended for new users to install the driver via this way because it is simple. However, it has some drawbacks:

  1. The driver included in official Ubuntu repository is usually not the latest.
  2. There would be some naming conflicts when other repositories (e.g. ones from CUDA) are added to the system.
  3. One has to reinstall the driver after Linux kernel are updated.

Install NVIDIA Graphics Driver via runfile

For advanced user who wants to get the latest version of the driver, get rid of the reinstallation issue caused bby dkms, or using Linux distributions that do not have nvidia drivers provided in the repositories, installing from runfile is recommended.

Remove Previous Installations (Important)

One might have installed the driver via apt-get. So before reinstall the driver from runfile, uninstalling previous installations is required. Executing the following scripts carefully one by one.

sudo apt-get purge nvidia*

# Note this might remove your cuda installation as well
sudo apt-get autoremove 

# Recommended if .deb files from NVIDIA were installed
# Change 1404 to the exact system version or use tab autocompletion
# After executing this file, /etc/apt/sources.list.d should contain no files related to nvidia or cuda
sudo dpkg -P cuda-repo-ubuntu1404

Download the Driver

The latest NVIDIA driver for Linux OS can be fetched from NVIDIA's official website. The first one in the list, i.e. Latest Long Lived Branch version for Linux x86_64/AMD64/EM64T, is suitable for most case.

If you want to down load the driver directly in a Linux shell, the script below would be useful.

cd ~
wget http://us.download.nvidia.com/XFree86/Linux-x86_64/384.69/NVIDIA-Linux-x86_64-384.69.run

Detailed installation instruction can be found in the download page via a README hyperlink in the ADDITIONAL INFORMATION tab. I have also summarized key steps below.

Install Dependencies

Software required for the runfile are officially listed here. But this page seems to be stale and not easy to follow.

For Ubuntu, installing the following dependencies is enough.

  1. build-essential -- For building the driver
  2. (Optional) gcc-multilib -- For providing 32-bit support
  3. dkms -- For providing dkms support
  4. (Optional) xorg and xorg-dev. On a workstation with GUI, this is require but usually have already been installed, because you have already got the graphic display. On headless servers without GUI, this is not a must.

As a summary, excuting sudo apt-get install build-essential gcc-multilib dkms to install all dependencies.

Required packages for CentOS are epel-release dkms libstdc++.i686. Execute yum install epel-release dkms libstdc++.i686.

Required packages for Fedora are dkms libstdc++.i686 kernel-devel. Execute dnf install dkms libstdc++.i686 kernel-devel.

Creat Blacklist for Nouveau Driver

Create a file at /etc/modprobe.d/blacklist-nouveau.conf with the following contents:

blacklist nouveau
options nouveau modeset=0

Note: It is also possible for the NVIDIA installation runfile to creat this blacklist file automatically. Excute the runfile and follow instructions when an error realted Nouveau appears.

Then,

  1. for Ubuntu 14.04 LTS, reboot the computer;
  2. for Ubuntu 16.04 LTS, excute sudo update-initramfs -u and reboot the computer;
  3. for CentOS/Fedora, excute sudo dracut --force and reboot the computer.

Stop lightdm/gdm/kdm

After the computer is rebooted. We need to stop the desktop manager before excuting the runfile to install the driver. lightdm is the default desktop manager in Ubuntu. If GNOME or KDE desktop environment is used, installed desktop manager will then be gdm or kdm.

  1. For Ubuntu 14.04 / 16.04, excuting sudo service lightdm stop (or use gdm or kdm instead of lightdm)
  2. For Ubuntu 16.04 / Fedora / CentOS, excuting sudo systemctl stop lightdm (or use gdm or kdm instead of lightdm)

Excuting the Runfile

After above batch of preparition, we can eventually start excuting the runfile. So this is why I, from the very begining, recommend new users to install the driver via apt-get.

cd ~
chmod +x NVIDIA-Linux-x86_64-384.69.run
sudo ./NVIDIA-Linux-x86_64-384.69.run --dkms -s

Note:

  1. option --dkms is used for register dkms module into the kernel so that update of the kernel will not require a reinstallation of the driver. This option should be turned on by default.
  2. option -s is used for silent installation which should used for batch installation. For installation on a single computer, this option should be turned off for more installtion information.
  3. option --no-opengl-files can also be added if non-NVIDIA (AMD or Intel) graphics are used for display while NVIDIA graphics are used for display.
  4. The installer may prompt warning on a system without X.Org installed. It is safe to ignore that based on my experience.
WARNING: nvidia-installer was forced to guess the X library path '/usr/lib' and X module path '/usr/lib/xorg/modules'; these paths were not queryable from the system.  If X fails to find the NVIDIA X driver module, please install the `pkg-config` utility and the X.Org SDK/development package for your distribution and reinstall the driver.

Check the Installation

After a succesful installation, nvidia-smi command will report all your CUDA-capable devices in the system.

Common Errors and Solutions

  1. ERROR: Unable to load the 'nvidia-drm' kernel module.
  • One probable reason is that the system is boot from UEFI but Secure Boot option is turned on in the BIOS setting. Turn it off and the problem will be solved.

Additional Notes

nvidia-smi -pm 1 can enable the persistent mode, which will save some time from loading the driver. It will have significant effect on machines with more than 4 GPUs.

nvidia-smi -e 0 can disable ECC on TESLA products, which will provide about 1/15 more video memory. Reboot is reqired for taking effect. nvidia-smi -e 1 can be used to enable ECC again.

nvidia-smi -pl <some power value> can be used for increasing or decrasing the TDP limit of the GPU. Increasing will encourage higher GPU Boost frequency, but is somehow DANGEROUS and HARMFUL to the GPU. Decreasing will help to same some power, which is useful for machines that does not have enough power supply and will shutdown unintendedly when pull all GPU to their maximum load.

-i <GPUID> can be added after above commands to specify individual GPU.

These commands can be added to /etc/rc.local for excuting at system boot.

Install CUDA

Installing CUDA from runfile is much simpler and smoother than installing the NVIDIA driver. It just involves copying files to system directories and has nothing to do with the system kernel or online compilation. Removing CUDA is simply removing the installation directory. So I personally does not recommend adding NVIDIA's repositories and install CUDA via apt-get or other package managers as it will not reduce the complexity of installation or uninstallation but increase the risk of messing up the configurations for repositories.

The CUDA runfile installer can be downloaded from NVIDIA's websie. But what you download is a package the following three components:

  1. an NVIDIA driver installer, but usually of stale version;
  2. the actual CUDA installer;
  3. the CUDA samples installer;

To extract above three components, one can execute the runfile installer with --extract option. Then, executing the second one will finish the CUDA installation. Installation of the samples are also recommended because useful tool such as deviceQuery and p2pBandwidthLatencyTest are provided.

Scripts for installing CUDA Toolkit are summarized below.

cd ~
wget http://developer.download.nvidia.com/compute/cuda/7.5/Prod/local_installers/cuda_7.5.18_linux.run
chmod +x cuda_7.5.18_linux.run
./cuda_7.5.18_linux.run --extract=$HOME
sudo ./cuda-linux64-rel-7.5.18-19867135.run

After the installation finishes, configure runtime library.

sudo bash -c "echo /usr/local/cuda/lib64/ > /etc/ld.so.conf.d/cuda.conf"
sudo ldconfig

It is also recommended for Ubuntu users to append string /usr/local/cuda/bin to system file /etc/environments so that nvcc will be included in $PATH. This will take effect after reboot.

Install cuDNN

The recommended way for installing cuDNN is to first copy the tgz file to /usr/local and then extract it, and then remove the tgz file if necessary. This method will preserve symbolic links. At last, execute sudo ldconfig to update the shared library cache.

@parth126

This comment has been minimized.

parth126 commented Sep 26, 2016

I can not thank you enough for this information. Finally I can use my GPU after weeks of trying various solutions and reinstalling ubuntu a few times. This solution worked perfectly and without a single glitch.

@shravankumar147

This comment has been minimized.

shravankumar147 commented Oct 25, 2016

I was tried and installed successfully but my desktop has gone, it only shown just a blinking cursor at top left of the screen. I've tried many ways, like stopping and starting gdm, lightdm. None has worked for me, at last I installed current nvidia through sudo apt-get install nvidia-current, after successful installation and reboot, I get back my desktop. However this time I could not use gpu. when I try nvidia-smi:
nvidia-smi: command not found error occurring

any help to fix the issue greatly appreciated. Thank you

@wangruohui

This comment has been minimized.

Owner

wangruohui commented Nov 4, 2016

@shravankumar147

I am not quite sure about what happens on your desktop. But I suggest you try apt-get install nvidia-361 --reinstall (ubuntu 1604) or apt-get install nvidia-352 --reinstall (Ubuntu 1404). The nvidia-current seems to be pointed to a considerably old version.

@wangruohui

This comment has been minimized.

Owner

wangruohui commented Nov 4, 2016

@shravankumar147

and apt-get install nvidia-modprobe as I write in part Install NVIDIA Graphics Driver via apt-get

@cseeker

This comment has been minimized.

cseeker commented Nov 24, 2016

I'm using ASUS mainboard and ubuntu 14.04.
If someone meet "ERROR: Unable to load the 'nvidia-drm' kernel module."
Check uefi secure boot.
Maybe your bios setting of it is for "windows".
Just change it to for "other OS"
Then, save and exit.
You can see the installed driver after reboot.

@BarTsouri

This comment has been minimized.

BarTsouri commented Mar 5, 2017

thank you so much

@yoelfme

This comment has been minimized.

yoelfme commented Apr 5, 2017

Thanks you so much @wangruohui, atfter more than 20 attempts with other guides, yours has saved my life.

@luango3110

This comment has been minimized.

luango3110 commented Apr 12, 2017

Thank you so much. You really save my life after I struggled with many errors

@yongbeima

This comment has been minimized.

yongbeima commented May 1, 2017

working well,thanks..

@Kif11

This comment has been minimized.

Kif11 commented May 5, 2017

Very good comprehensive manual. Thanks.

@visha-l

This comment has been minimized.

visha-l commented Jun 2, 2017

I followed all the instruction but getting this error when i run make command in darknet.

ubuntu@ip-10-0-0-226:~/darknet$ make
nvcc  -gencode arch=compute_20,code=[sm_20,sm_21] -gencode arch=compute_30,code=sm_30 -gencode arch=compute_35,code=sm_35 -gencode arch=compute_50,code=[sm_50,compute_50] -gencode arch=compute_52,code=[sm_52,compute_52]  -DGPU -I/usr/local/cuda-7.0/include/ --compiler-options "-Wall -Wfatal-errors  -Ofast -DGPU" -c ./src/convolutional_kernels.cu -o obj/convolutional_kernels.o
/bin/sh: 1: nvcc: not found
make: *** [obj/convolutional_kernels.o] Error 127

@vimac001

This comment has been minimized.

vimac001 commented Jun 22, 2017

Done. The second version. And it worked, but now i am stuck in a login loop and can't find a solution. I am using Ubuntu 17.04

@wangruohui

This comment has been minimized.

Owner

wangruohui commented Aug 29, 2017

@visha-l This is because /usr/local/cuda/bin is not in your path. You can try export PATH=$PATH:/usr/local/cuda/bin to solve your problem.

@srihas619

This comment has been minimized.

srihas619 commented Sep 10, 2017

@vimac001 I am also stuck there. I hope someone can help us (without uninstalling nvidia)

@nathantsoi

This comment has been minimized.

nathantsoi commented Sep 13, 2017

this is awesome, thx @wangruohui

@sli888

This comment has been minimized.

sli888 commented Sep 21, 2017

I have to say this is the best instruction out there for Nvidia Linux Driver installation! Thank you.

@wa1618i

This comment has been minimized.

wa1618i commented Sep 30, 2017

THANKS

@mlushpenko

This comment has been minimized.

mlushpenko commented Oct 26, 2017

The first option that actually worked, thanks! Before this, my drivers were installed to /usr/lib/nvidia for some reason and adding those folders to PATH wouldn't help.

I do have one problem still - only one card out of 5 is visible, nothing changed to hardware setup. My friend told me it's possible that some default config was used on Ubuntu to enable only 1 card instead of all available, but I don't have enough knowledge about that. Any ideas what may be the issue?

@wangmengbo

This comment has been minimized.

wangmengbo commented Nov 12, 2017

In the section Install NVIDIA Graphics Driver via apt-get, the command missing 'install', which should be sudo apt-get install nvidia-xxx nvidia-modprobe.
Not a big deal but save time.
But thx a lot for saving my life. I spent days on configuring what was going wrong with my Ubuntu.

@arnaghosh

This comment has been minimized.

arnaghosh commented Nov 13, 2017

Hi
Thanks for the detailed steps. I tried installing nvidia drivers using the 1st step (apt-get). I was installing nvidia-384 on Ubuntu 16.04 and I have Geforce GTX 960. When I tried downloading the official driver from the Nvidia website, I got the runfile corresponding to 384.98. However I see that apt-get nvidia-384 installs 384.90. I don't know if this is an issue :/
When I try the command nvidia-smi, it gives me the error:-
NVIDIA-SMI has failed because it couldn't communicate with the NVIDIA driver. Make sure that the latest NVIDIA driver is installed and running.
I have purged all nvidia drivers right now. Could you suggest what I should do?

@robmoggach

This comment has been minimized.

robmoggach commented Nov 16, 2017

Thanks. Great work.

@pablorpalafox

This comment has been minimized.

pablorpalafox commented Nov 20, 2017

@pablorpalafox

This comment has been minimized.

pablorpalafox commented Nov 20, 2017

@Salzie

This comment has been minimized.

Salzie commented Nov 23, 2017

This is the best and complete guide. After weeks of trying and getting errors and shit, this is the one stop solution to all problems. Can't thank you enough for this.

@htcai

This comment has been minimized.

htcai commented Nov 27, 2017

This post is amazingly helpful!

A minor note: I am wondering whether it has been pointed out somewhere else. In my Latest Fedora 27 Workstation, at the step of Excuting the Runfile, I experienced freezing of the desktop after I entered the command to execute the runfile. I guess it is due to the stopping of gdm. My solution is to switch to the virtual console by pressing Ctrl + Alt + 2 before stopping gdm, which worked perfectly following the description in the post.

@IvyGongoogle

This comment has been minimized.

IvyGongoogle commented Dec 7, 2017

thanks a million

@kunth

This comment has been minimized.

kunth commented Dec 11, 2017

Great work. Thanks.

@ankanheesun

This comment has been minimized.

ankanheesun commented Dec 13, 2017

Your post is awesome !!!

Thanks for explaining almost of what I haven't know before 🥇

@Cysu

This comment has been minimized.

Cysu commented Dec 16, 2017

@wangruohui, Ruohui, thank you very much for the step-by-step guidelines. Eventually I have to use Ubuntu =]

@bearpaw

This comment has been minimized.

bearpaw commented Dec 18, 2017

@Cysu I happened to get here today and found it's written by Bro Hui 23333

@derek-pappas

This comment has been minimized.

derek-pappas commented Jan 14, 2018

I do the Nvidia driver installation using ssh from another machine in case things go south...

@Mohith22

This comment has been minimized.

Mohith22 commented Jan 22, 2018

I have successfully installed Nvidia drivers. But after installation the command - "nvidia-smi" is not working. I used apt-get to install the drivers. Can you help me out?

@euisuny

This comment has been minimized.

euisuny commented Jan 24, 2018

Thank you so much for this! Just wanted to point out a small typo - the header "Creat Blacklist for Nouveau Driver" seems to be missing an e.

@KarthikVundela

This comment has been minimized.

KarthikVundela commented Feb 5, 2018

Awesome post!

In the note for executing the run file,
3.option --no-opengl-files can also be added if non-NVIDIA (AMD or Intel) graphics are used for display while NVIDIA graphics are used for display.

I assume that you there is some typo at the end of the sentence. May I know what were you trying to say?
I have Ubuntu 16.04 and created the blacklist file. Is the blacklist file sufficient to avoid conflicts between default Intel graphics and Nvidia graphics or is this flag also required during installation with run file. I am a novice and do not understand the difference between the blacklist file and this flag.
Thanks.

@martedva

This comment has been minimized.

martedva commented Feb 13, 2018

When trying to run the nvidia-smi command to test everything works, i get:

"nvidia-smi: command not found"

I'm using the second option, with CUDA, on my "Bash on Ubuntu on Windows" installation.

I have not done the following step, as I am uncertain how to do this - could this be the reason for the missing command?

"It is also recommended for Ubuntu users to append string /usr/local/cuda/bin to system file /etc/environments so that nvcc will be included in $PATH. This will take effect after reboot."

@jfpuget

This comment has been minimized.

jfpuget commented Feb 13, 2018

HI, thanks for this, it helped me a lot. A minor update: the latest driver (390) runfile checks for X running before testing for nouveau, hence your pre install steps aren't correct anymore. Better do the nouveau override and reboot before trying the installer IMHO.

@atran

This comment has been minimized.

atran commented Feb 13, 2018

Had to do sudo apt-get install instead of just sudo apt-get here and edited /etc/environment instead of /etc/environments here.

@hxsnow10

This comment has been minimized.

hxsnow10 commented Feb 26, 2018

thaks a lot. After i make server shut down, finally install tensorflow 1.5 this time!

@wesruv

This comment has been minimized.

wesruv commented Mar 17, 2018

Thanks so much, one note, when I disabled lightdm I completely lost my screen, maybe smarter people than me know that that'd happen, but it may be worth mentioning?

I googled around to find some solution where I could get a CLI, but quickly switched to SSH'ing into the machine from another machine on the network.

Worked great, thanks so much!

@LauraRuis

This comment has been minimized.

LauraRuis commented Mar 19, 2018

Thanks a lot for this!

People who are installing with the .run file and are having trouble with the message "The distribution-provided pre-install script failed! Are you sure you want to continue?", for me the fix was disabling UEFI secure boot in the bios.

Also: to not get stuck in the login loop add the option --no-opengl-files when executing the run file for the NVIDIA driver

@rameshka

This comment has been minimized.

rameshka commented Mar 20, 2018

Thanks a lot for the tips..
As @LauraRuis mentioned, adding option --no-opengl-files to execute the run file prevent login loop.

@Madhivarman

This comment has been minimized.

Madhivarman commented Mar 24, 2018

I followed the instructions and I stuck in the login loop. After few hours struggle, I am back to the home desktop. I came to know that some weird Nvidia driver installation caused this problem. How do I choose right driver for my laptop?

lspci

00:00.0 Host bridge: Intel Corporation Haswell-ULT DRAM Controller (rev 0b)
00:02.0 VGA compatible controller: Intel Corporation Haswell-ULT Integrated Graphics Controller (rev 0b)
00:03.0 Audio device: Intel Corporation Haswell-ULT HD Audio Controller (rev 0b)
00:14.0 USB controller: Intel Corporation 8 Series USB xHCI HC (rev 04)
00:16.0 Communication controller: Intel Corporation 8 Series HECI #0 (rev 04)
00:1b.0 Audio device: Intel Corporation 8 Series HD Audio Controller (rev 04)
00:1c.0 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 1 (rev e4)
00:1c.2 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 3 (rev e4)
00:1c.3 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 4 (rev e4)
00:1c.4 PCI bridge: Intel Corporation 8 Series PCI Express Root Port 5 (rev e4)
00:1d.0 USB controller: Intel Corporation 8 Series USB EHCI #1 (rev 04)
00:1f.0 ISA bridge: Intel Corporation 8 Series LPC Controller (rev 04)
00:1f.2 SATA controller: Intel Corporation 8 Series SATA Controller 1 [AHCI mode] (rev 04)
00:1f.3 SMBus: Intel Corporation 8 Series SMBus Controller (rev 04)
06:00.0 Network controller: Qualcomm Atheros QCA9565 / AR9565 Wireless Network Adapter (rev 01)
07:00.0 Ethernet controller: Realtek Semiconductor Co., Ltd. RTL8101/2/6E PCI Express Fast/Gigabit Ethernet controller (rev 07)
08:00.0 3D controller: NVIDIA Corporation GF117M [GeForce 610M/710M/810M/820M / GT 620M/625M/630M/720M] (rev ff)

Any suggestions?

@robertpmatson

This comment has been minimized.

robertpmatson commented Apr 8, 2018

Many thanks for writing this excellent article, it really helped me. I followed the NVIDIA instructions to the letter and had issues, this article should be the official guide.

@bhavikngala

This comment has been minimized.

bhavikngala commented Apr 9, 2018

Excellent guide. I had the login loop issue. Turned out that, my display card is nonNVIDIA. The --no-opengl-files can flag along with the installation steps above solved the issue.

@ny2292000

This comment has been minimized.

ny2292000 commented Apr 10, 2018

I am new to this issue. I have an nvidia motherboard with an internal nvidia video driver with NVIDIA® 780a SLI Chipset. I also added a nvidia K620 video for cuda computing. I started with a fresh install of Ubuntu 16.04. After I installed the video
sudo apt-get nvidia-375 nvidia-modprobe

I wasn't able to login again. I suspect that the nvidia-375 doesn't work with the motherboard video and that breaks something. In the past I was able to get the ubuntu to work by Alt-CTL-F1 and purge nvidia-*, but that defeats the purpose.

Any guidance would be extremely helpful. Thanks

@eatsleepraverepeat

This comment has been minimized.

eatsleepraverepeat commented Apr 15, 2018

Well, it's clear that this excellent guide saved a lot of time of messing around. Mentioned driver version is outdated, but the idea is still the same. Works like a charm! Big thanks for your work.

@earcz

This comment has been minimized.

earcz commented Apr 19, 2018

I would like to know that why additional softwares in system settings shows " No proprietary drivers are in use" ?
By the way, amazing solution, thank you.

@neelkadia

This comment has been minimized.

neelkadia commented Apr 27, 2018

THANK YOU BILLION TIMES!! Just installed driver and CUDA without crashing my system as I haven't backed up some very important patent pending code!

@bhw7

This comment has been minimized.

bhw7 commented May 8, 2018

I've just created a bash script that does all but the CUDA install. It's free to use! GIST

@partnercloudsupport

This comment has been minimized.

partnercloudsupport commented Jun 12, 2018

This is the most valuable document regarding CUDA install and drivers ever created.
I tried for 2 weeks trying different things to get this to work. May you be forever blessed for sharing your findings. Thank you.

@chaneyw

This comment has been minimized.

chaneyw commented Jul 16, 2018

If anyone gets the blinking cursor in the upper left corner on reboot, and no other response, a solution is to Ctrl-Alt-F2 on Ubuntu. This will bring up a menu in which you can choose "Login with command line only" or something similar, I believe it is the last option of the 4 options presented. Then you will have a command line from which you can continue to follow the instructions. Reboot again and you're golden.

@fybaft2012

This comment has been minimized.

fybaft2012 commented Jul 25, 2018

Dear @wangruohui, I met a problem of my Ubuntu 16.04, I really hope you can help me. I installed Nvidia Titan X and drivers, Cuda, everything worked well half a year ago. But yesterday I restart my server, it showed the error "your system is running in low graphics mode...", then I try to removed the Nvidia GPU and use the default graphic card, everything is okay. But I need use Nvidia GPU for deep learning anyway.
Could you tell me what is wrong with my server and how to solve it?
Thanks! @wangruohui

@ArayCHN

This comment has been minimized.

ArayCHN commented Jul 31, 2018

Thanks @chaneyw you saved my day!! Can you @wangruihui please also add his advice to the tutorial? I actually got stuck there and was unable to input via keyboard on stopping lightdm. Thanks to y'all again.

@katnoria

This comment has been minimized.

katnoria commented Sep 10, 2018

Thanks @wangruohi, your guide helped me resolve ImportError: libnvidia-fatbinaryloader.so.396.54, I couldn't find that file anywhere under /usr/lib/nvidia-xx folder. A clean install using your guide along with NVidia accelerated library helped me use my GPU again :)

Thank you

@vinaykumar2491

This comment has been minimized.

vinaykumar2491 commented Oct 11, 2018

Hi,
I'm trying to install nvidia drivers on a KVM instance (base_OS=ubuntu16.04, vm_OS= ubuntu 16.04). I have 2 GPUs, One for base OS and other for virtualization.
I'm able to read NVIDIA devices in VM instance with lspci but I'm unable to install the drivers.

I'm facing the following issues:

  1. If I do sudo apt-get install nvidia-390 then all drivers are installed in my VM but, when I reboot my VM I get stuck in the login screen.
  2. Else, If I do driver installation via .run file (downloaded from the nvidia website), then as per the steps described in the solution above, I need to stop my X-server in the VM, so I do sudo service lightdm stop and my VM goes into a black screen. After I reboot my VM, and try to install drivers by running the .run file I still get the error to stop my X-server.

What should I do?
Thanks

@Mozhdeh-d

This comment has been minimized.

Mozhdeh-d commented Oct 19, 2018

Hi,
I installed Nvidia driver 390.77 on my system successfully via the Software Center > Additional software. Now when I try to install CUDA, the following error occurs:
Warning: Incomplete installation! This installation did not install the CUDA driver. A driver of version at least 384.00 is required for CUDA 9.1 functionality to work.

And I'm sure that the version of installed Nvidia driver is greater than 384.00!! Because when I use nvidia-smi command, it shows that the version of my my Nvidia driver is 390.77

Any suggestion? :(
Thanks in advance

Ubuntu: 18.04
Geforce GTX 750Ti
Nvidia driver: 390.77
CUDA: 9.1.85

Sign up for free to join this conversation on GitHub. Already have an account? Sign in to comment