CUDA + NVIDIA drivers - Revision history

DrEdWilliams: /* Debian 10 */

2021-08-22T20:11:05Z

Debian 10

← Older revision		Revision as of 20:11, 22 August 2021
Line 28:		Line 28:
	yum -y install yum-plugin-nvidia		yum -y install yum-plugin-nvidia

	=== Debian 10 ===		=== Debian 10/11 ===
	Install the repo and then deploy the full package:		Install the repo and then deploy the full package:
	sudo apt-get install -y software-properties-common		sudo apt-get install -y software-properties-common

DrEdWilliams: /* Debian 10 */

2021-08-22T20:07:02Z

Debian 10

← Older revision		Revision as of 20:07, 22 August 2021
Line 36:		Line 36:
	sudo apt-get update		sudo apt-get update
	sudo apt-get -y install cuda		sudo apt-get -y install cuda

			Alternatively, if the full CUDA distribution is not wanted/needed:
			sudo apt -y install nvidia-driver nvidia-smi

			After the install completes, a reboot is required to activate the drivers.

	== Updating drivers ==		== Updating drivers ==

DrEdWilliams: /* Official Instructions) */

2021-01-20T01:14:20Z

Official Instructions)

← Older revision		Revision as of 01:14, 20 January 2021
Line 3:		Line 3:

	=== Fedora 33 ===		=== Fedora 33 ===
	~~==== Official Instructions) ====~~
	Install the repo and then deploy the full package:		Install the repo and then deploy the full package:
	sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora33.repo		sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora33.repo

DrEdWilliams at 01:14, 20 January 2021

2021-01-20T01:14:03Z

← Older revision		Revision as of 01:14, 20 January 2021
Line 2:		Line 2:
	Download repo rpm from http://developer.nvidia.com/cuda-downloads (or copy the repo file over from another similar system). For further information, see the [https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html Installation Guide for Linux] and the [https://docs.nvidia.com/cuda/cuda-quick-start-guide/index.html CUDA Quick Start Guide].		Download repo rpm from http://developer.nvidia.com/cuda-downloads (or copy the repo file over from another similar system). For further information, see the [https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html Installation Guide for Linux] and the [https://docs.nvidia.com/cuda/cuda-quick-start-guide/index.html CUDA Quick Start Guide].

	=== Fedora ~~32/~~33 ===		=== Fedora 33 ===
	==== Official Instructions) ====		==== Official Instructions) ====
	Install the repo and then deploy the full package:		Install the repo and then deploy the full package:
	sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-~~fedora32~~.repo		sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora33.repo
	sudo dnf clean all		sudo dnf clean all
	sudo dnf -y module install nvidia-driver:latest-dkms		sudo dnf -y module install nvidia-driver:latest-dkms
	sudo dnf -y install cuda		sudo dnf -y install cuda

	~~==== Using RPMFusion ====~~
	~~While using the NVIDIA repo would be preferable ... it doesn't seem to work. This does ... YMMV~~
	~~sudo dnf -y install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm~~
	~~sudo dnf -y install https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm~~
	~~sudo dnf -y install akmod-nvidia~~

	~~Afterward, you can still install the rest of the CUDA packages from the main NVIDIA repo ...~~
	~~sudo dnf -y install cuda~~

	~~Ref:~~
	* https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-fedora-32
	* https://www.if-not-true-then-false.com/2015/fedora-nvidia-guide/#before-installation

	=== CentOS 7 ===		=== CentOS 7 ===

DrEdWilliams: /* Debian 10 */

2020-12-19T01:29:34Z

Debian 10

← Older revision		Revision as of 01:29, 19 December 2020
Line 44:		Line 44:
	=== Debian 10 ===		=== Debian 10 ===
	Install the repo and then deploy the full package:		Install the repo and then deploy the full package:
			sudo apt-get install -y software-properties-common
	sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/7fa2af80.pub		sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/7fa2af80.pub
	sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/ /"		sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/ /"

10.244.0.27: /* Using RPMFusion */

2020-11-29T19:44:27Z

Using RPMFusion

← Older revision		Revision as of 19:44, 29 November 2020
Line 19:		Line 19:
	sudo dnf -y install cuda		sudo dnf -y install cuda

	Ref: https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-fedora-32, https://www.if-not-true-then-false.com/2015/fedora-nvidia-guide/#before-installation		Ref:
			* https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-fedora-32
			* https://www.if-not-true-then-false.com/2015/fedora-nvidia-guide/#before-installation

	=== CentOS 7 ===		=== CentOS 7 ===

DrEdWilliams at 18:50, 29 November 2020

2020-11-29T18:50:18Z

← Older revision		Revision as of 18:50, 29 November 2020
Line 18:		Line 18:
	Afterward, you can still install the rest of the CUDA packages from the main NVIDIA repo ...		Afterward, you can still install the rest of the CUDA packages from the main NVIDIA repo ...
	sudo dnf -y install cuda		sudo dnf -y install cuda

			Ref: https://linuxconfig.org/how-to-install-the-nvidia-drivers-on-fedora-32, https://www.if-not-true-then-false.com/2015/fedora-nvidia-guide/#before-installation

	=== CentOS 7 ===		=== CentOS 7 ===

DrEdWilliams: /* Fedora 32/33 */

2020-11-29T18:08:09Z

Fedora 32/33

← Older revision		Revision as of 18:08, 29 November 2020
Line 3:		Line 3:

	=== Fedora 32/33 ===		=== Fedora 32/33 ===
			==== Official Instructions) ====
	Install the repo and then deploy the full package:		Install the repo and then deploy the full package:
	sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora32.repo		sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora32.repo
	sudo dnf clean all		sudo dnf clean all
	sudo dnf -y module install nvidia-driver:latest-dkms		sudo dnf -y module install nvidia-driver:latest-dkms
			sudo dnf -y install cuda

			==== Using RPMFusion ====
			While using the NVIDIA repo would be preferable ... it doesn't seem to work. This does ... YMMV
			sudo dnf -y install https://download1.rpmfusion.org/free/fedora/rpmfusion-free-release-$(rpm -E %fedora).noarch.rpm
			sudo dnf -y install https://download1.rpmfusion.org/nonfree/fedora/rpmfusion-nonfree-release-$(rpm -E %fedora).noarch.rpm
			sudo dnf -y install akmod-nvidia

			Afterward, you can still install the rest of the CUDA packages from the main NVIDIA repo ...
	sudo dnf -y install cuda		sudo dnf -y install cuda

DrEdWilliams at 16:39, 29 November 2020

2020-11-29T16:39:58Z

← Older revision		Revision as of 16:39, 29 November 2020
Line 1:		Line 1:
	== Installation ==		== Installation ==
	Download repo rpm from http://developer.nvidia.com/cuda-downloads (or copy the repo file over from another system)		Download repo rpm from http://developer.nvidia.com/cuda-downloads (or copy the repo file over from another similar system). For further information, see the [https://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html Installation Guide for Linux] and the [https://docs.nvidia.com/cuda/cuda-quick-start-guide/index.html CUDA Quick Start Guide].

			=== Fedora 32/33 ===
			Install the repo and then deploy the full package:
			sudo dnf config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/fedora32/x86_64/cuda-fedora32.repo
			sudo dnf clean all
			sudo dnf -y module install nvidia-driver:latest-dkms
			sudo dnf -y install cuda

			=== CentOS 7 ===
	It seems that the dependency for the linux kernel devel package has been lost ... and the driver install just silently fails		It seems that the dependency for the linux kernel devel package has been lost ... and the driver install just silently fails
	yum install kernel-devel		yum install kernel-devel

			Now install the repository
			sudo yum-config-manager --add-repo https://developer.download.nvidia.com/compute/cuda/repos/rhel7/x86_64/cuda-rhel7.repo
			sudo yum clean all

	On regular compute nodes where the full CUDA libraries aren't needed:		On regular compute nodes where the full CUDA libraries aren't needed:
	yum install nvidia-driver-~~cuda~~		yum install nvidia-driver-latest-dkms

	On a full workstation or where CUDA is needed:		On a full workstation or where CUDA is needed:
	yum install cuda		sudo yum -y install nvidia-driver-latest-dkms cuda
			sudo yum -y install cuda-drivers

	There is a yum plugin ~~(for CentOS, anyway)~~ that facilitates the installation and management of NVIDIA kernel modules:		There is a yum plugin that facilitates the installation and management of NVIDIA kernel modules:

	yum -y install yum-plugin-nvidia		yum -y install yum-plugin-nvidia

	~~Read more at~~: ~~http~~://~~docs~~.nvidia.com/cuda/cuda-~~installation~~-~~guide~~-~~linux/index.html#ixzz4zMULzF9o~~		=== Debian 10 ===
			Install the repo and then deploy the full package:
			sudo apt-key adv --fetch-keys https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/7fa2af80.pub
			sudo add-apt-repository "deb https://developer.download.nvidia.com/compute/cuda/repos/debian10/x86_64/ /"
			sudo add-apt-repository contrib
			sudo apt-get update
			sudo apt-get -y install cuda

	== Updating drivers ==		== Updating drivers ==
	The above process installs the cuda repository for yum ... and while most of the time updates will happen automatically, there are some things to consider when updating.		The above process installs the cuda repository for yum ... and while most of the time updates will happen automatically, there are some things to consider when updating.

			The command examples are for CentOS ... use the command and package set appropriate for the OS in use.

	=== Reinstalling ===		=== Reinstalling ===

DrEdWilliams at 15:19, 29 November 2020

2020-11-29T15:19:54Z

@@ Line 1: / Line 1: @@
 Download repo rpm from http://developer.nvidia.com/cuda-downloads (or copy the repo file over from another system)
@@ Line 9: / Line 10: @@
 On a full workstation or where CUDA is needed:
   yum install cuda
 Read more at: http://docs.nvidia.com/cuda/cuda-installation-guide-linux/index.html#ixzz4zMULzF9o