Giters

feyil / Linux-Mint-Nvidia-Driver-Cuda

Set of instructions to successful installation of NVIDIA official driver and CUDA Toolkit

Geek Repo:Geek Repo

Github PK Tool:Github PK Tool

Linux Mint 19.2 Cinnamon NVIDIA 940M Driver Installation & CUDA 10.2 Toolkit Installation Instructions

  • This repo includes set of steps to successfully install NVIDIA Driver and CUDA 10.2 in Mint 19.2 Cinnamon

Official Driver Installation for NVIDIA 940M

  1. First make the environment ready for driver installation by running below commands:
$ sudo dpkg --add-architecture i386
$ sudo apt update
$ sudo apt install build-essential libc6:i386

  1. Second dowload offical NVIDIA driver from NVIDIA driver websites. It is a .run file which includes necessary scripts to install driver.

  2. Disable Nouveau Nvidia driver

$ sudo bash -c "echo blacklist nouveau > /etc/modprobe.d/blacklist-nvidia-nouveau.conf"
$ sudo bash -c "echo options nouveau modeset=0 >> /etc/modprobe.d/blacklist-nvidia-nouveau.conf"
  1. Confirm Updated files has following contents
$ cat /etc/modprobe.d/blacklist-nvidia-nouveau.conf
blacklist nouveau
options nouveau modeset=0
  1. Enter the following linux command to regenerate initramfs:
$ sudo update-initramfs -u
  1. Reboot the system when the computer boots up Nouveau Driver has been disabled.
$ sudo reboot
  1. Stop Desktop Manager. After executing the following linux command the display server will stop, therefore make sure you save all your current work ( if any ) before you proceed:
$ sudo telinit 3

  1. Hit CTRL+ALT+F1 and login with your username and password to open a new TTY1 session or perform the Nvidia driver installation via SSH bash.

  2. Install Nvidia Driver

$ sudo bash NVIDIA-Linux-x86_64-xxx.xx.run
- Accept License
- Would you like to register the kernel module sources with DKMS? This will allow DKMS to automatically build a new module, if you install a different kernel later. -> **YES**
- Install NVIDIA's 32-bit compatibility libraries? -> **YES**
- The distribution-provided pre-install script failed! Are you sure you want to continue? -> CONTINUE INSTALLATION
An incomplete installation of libglvnd was found. Do you want to install a full copy of libglvnd? This will overwrite any existing libglvnd libraries. -> **Install and overwrite existing files**
- Would you like to run the nvidia-xconfig utility? -> **YES**
  1. NVIDIA Driver is now installed. Reboot the system
$ sudo reboot

  1. After NVIDIA Driver installation when the splash screen pop-up. Screen will go black hit CTRL+ALT+F1 and login with username and password

  2. Initialize a new X session for the system. GUI Interface will open.

$ startx
  1. So, we are back to the GUI now we know that our GPU is NVIDIA Optimus capable so will install nvidia-prime
$ sudo apt-get install nvidia-prime

  1. Reboot the system when the reboot succeed GUI will appear without typing startx or following step 12

  2. Thats it we successfully installed latest Nvidia offical driver to the system. You can configure NVIDIA settings from NVIDIA X Server

CUDA 10.2 Installation

  1. Go to the NVIDIA Offical website and dowload CUDA 10.2 Toolkit with following instruction for .deb installation method
$ 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 http://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
  1. After the succesfull completation of cuda reboot the system
$ sudo reboot
  1. If nothing goes wrong we have succesfully installed. Lets check our compiler version
$ cd /etc/usr/local/cuda-10-2/bin
$ ./nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89
  1. Lets test our compiler with NVDIA's examples:
$ cd /etc/usr/local/cuda/samples/0_Simple/matrixMul
$ make
$ ./matrxiMul
[Matrix Multiply Using CUDA] - Starting...
GPU Device 0: "Maxwell" with compute capability 5.0

MatrixA(320,320), MatrixB(640,320)
Computing result using CUDA Kernel...
done
Performance= 105.24 GFlop/s, Time= 1.245 msec, Size= 131072000 Ops, WorkgroupSize= 1024 threads/block
Checking computed result for correctness: Result = PASS

NOTE: The CUDA Samples are not meant for performancemeasurements. Results may vary when GPU Boost is enabled.
  1. Lets configure our environment variables to access CUDA compiler in the system wide. Add followings to the end of .bashrc file.
export PATH=/usr/local/cuda-10.2/bin${PATH:+:${PATH}}$
export LD_LIBRARY_PATH=/usr/local/cuda-10.2/lib64${LD_LIBRARY_PATH:+:${LD_LIBRARY_PATH}}
  1. Open a new terminal window and type:
$ nvcc --version
nvcc: NVIDIA (R) Cuda compiler driver
Copyright (c) 2005-2019 NVIDIA Corporation
Built on Wed_Oct_23_19:24:38_PDT_2019
Cuda compilation tools, release 10.2, V10.2.89
  1. That's it we successfully installed CUDA 10.2 Toolkit we can compile our programs with nvcc command.

References

How to install Nvidia drivers on Linux Mint

How to disable Nouveau nvidia driver on Ubuntu 18.04 Bionic Beaver Linux

How to install CUDA 9.2 on Ubuntu 18.04

nvcc error, “no command” verifying installation of CUDA 10 on Debian 9 Stretch

About

Set of instructions to successful installation of NVIDIA official driver and CUDA Toolkit