CUDA: Difference between revisions

NixOS supports using NVIDIA GPUs for pure computing purposes, not just for graphics. For example, many users rely on NixOS for machine learning both locally and on cloud instances. These use cases are supported by the [https://github.com/orgs/NixOS/teams/cuda-maintainers @NixOS/cuda-maintainers team] on GitHub ([https://github.com/orgs/NixOS/projects/27 project board]). If you have an issue using your NVIDIA GPU for computing purposes [https://github.com/NixOS/nixpkgs/issues/new/choose open an issue] on GitHub and tag <code>@NixOS/cuda-maintainers</code>.

{{tip|1='''Data center GPUs''': Note that you may need to adjust your driver version to use "data center" GPUs like V100/A100s. See [https://discourse.nixos.org/t/how-to-use-nvidia-v100-a100-gpus/17754 this thread] for more info.}}

== <code>cudatoolkit</code>, <code>cudnn</code>, and related packages ==

The CUDA toolkit is available in a [https://search.nixos.org/packages?channel=unstable&from=0&size=50&buckets=%7B%22package_attr_set%22%3A%5B%22cudaPackages%22%5D%2C%22package_license_set%22%3A%5B%5D%2C%22package_maintainers_set%22%3A%5B%5D%2C%22package_platforms%22%3A%5B%5D%7D&sort=relevance&type=packages&query=cudatoolkit number of different versions]. Please use the latest major version. You can see where they're defined in nixpkgs [https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/cuda-modules/cudatoolkit/releases.nix here].

* cuDNN is packaged [https://github.com/NixOS/nixpkgs/tree/master/pkgs/development/cuda-modules/cudnn here].

* cuTENSOR is packaged [https://github.com/NixOS/nixpkgs/tree/master/pkgs/development/cuda-modules/cutensor here].

There are some possible ways to setup a development environment using CUDA on NixOS. This can be accomplished in the following ways:

* By making a FHS user env

# Run with `nix-shell cuda-fhs.nix`

   name = "cuda-env";

   targetPkgs = pkgs: with pkgs; [  

     unzip

     cudatoolkit

     libGLU libGL

     xorg.libXi xorg.libXmu freeglut

   profile = ''

     export CUDA_PATH=${pkgs.cudatoolkit}

     export EXTRA_CCFLAGS="-I/usr/include"

   '';

}).env

* By making a nix-shell

{ pkgs ? import </nowiki><nixpkgs><nowiki> {} }:

pkgs.mkShell {

   name = "cuda-env-shell";

     git gitRepo gnupg autoconf curl

     procps gnumake util-linux m4 gperf unzip

     libGLU libGL

     xorg.libXi xorg.libXmu freeglut

   shellHook = ''

       export CUDA_PATH=${pkgs.cudatoolkit}

       export EXTRA_CCFLAGS="-I/usr/include"

   '';           

}

{

   description = "CUDA development environment";

       config.cudaVersion = "12";

     };

   in {

     # alejandra is a nix formatter with a beautiful output

         cudaPackages.cuda_cudart

         cudatoolkit

         cudaPackages.cudnn

         libGLU

         xorg.libXrandr

         zlib

         stdenv.cc

         binutils

       shellHook = ''

         export CUDA_PATH=${pkgs.cudatoolkit}

         export EXTRA_CCFLAGS="-I/usr/include"

         export CMAKE_PREFIX_PATH="${pkgs.fmt.dev}:$CMAKE_PREFIX_PATH"

     };

   };

== Setting up CUDA Binary Cache ==

For more information, refer to the [[Binary Cache#Using a binary cache Using a binary cache|Using a binary cache]] page.

Add the cache to <code>substituters</code> and <code>trusted-public-keys</code> inside your system configuration:

nix.settings = {

   substituters = [

   ];

   trusted-public-keys = [

   ];

};

=== Non-NixOS ===

trusted-users = root @wheel

If your user is in <code>trusted-users</code>, you can also add the cache in your home directory:

== Some things to keep in mind when setting up CUDA in NixOS ==