CUDA: Difference between revisions

(14 intermediate revisions by 8 users not shown)

Line 1:

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/~~projects~~/27 @NixOS/cuda-maintainers team] on GitHub. If you have an issue using your NVIDIA GPU for computing purposes [https://github.com/~~nixos~~/nixpkgs/issues/new open an issue] on GitHub and tag @NixOS/cuda-maintainers.

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>.

'''Cache''': Using the [https://app.cachix.org/cache/~~cuda~~-~~maintainers#pull cuda~~-~~maintainers~~ cache] is recommended! It will save you valuable time and electrons. Getting set up should be as simple as <code>cachix use ~~cuda~~-~~maintainers~~</code>.

{{tip|1='''Cache''': Using the [https://app.cachix.org/cache/nix-community nix-community cache] is recommended! It will save you valuable time and electrons. Getting set up should be as simple as <code>cachix use nix-community</code>. Click [[#Setting up CUDA Binary Cache|here]] for more details.}}

'''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.

{{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 ==

{{outdated|scope=section|date=July 2024|reason=Note that these examples have been updated more recently (as of 2024-07-30). May not be the best solution. A better resource is likely the packaging CUDA sample code [https://github.com/NixOS/nixpkgs/tree/master/pkgs/development/cuda-modules/cutensor here].}}

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].

Several "CUDA-X" libraries are packages as well. In particular,

* cuDNN is packaged [https://github.com/NixOS/nixpkgs/~~blob~~/master/pkgs/development/~~libraries/science/math~~/cudnn~~/generic.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/~~blob~~/master/pkgs/development~~/libraries/science/math/cutensor/generic.nix here].~~

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

'''Note that these examples haven't been updated in a while (as of 2022-03-12). May not be the best solution. A better resource is likely the packaging CUDA sample code [https://github.com/NixOS/nixpkgs/blob/master/pkgs/test/cuda/cuda-~~library-samples~~/~~generic.nix~~ here].~~'''~~

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

Line 19:

Line 17:

* By making a FHS user env

~~{{file|cuda-fhs~~.nix|nix~~|<nowiki>~~

<syntaxhighlight lang="nix" line="1" start="1"># flake.nix, run with `nix develop`

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

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

(pkgs.~~buildFHSUserEnv~~ {

let

# Change according to the driver used: stable, beta

nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;

in

(pkgs.buildFHSEnv {

name = "cuda-env";

targetPkgs = pkgs: with pkgs; [

Line 37:

Line 39:

unzip

cudatoolkit

~~linuxPackages.nvidia_x11~~

nvidiaPackage

libGLU libGL

xorg.libXi xorg.libXmu freeglut

Line 49:

Line 51:

profile = ''

export CUDA_PATH=${pkgs.cudatoolkit}

# export LD_LIBRARY_PATH=${~~pkgs.linuxPackages.nvidia_x11~~}/lib

# export LD_LIBRARY_PATH=${nvidiaPackage}/lib

export EXTRA_LDFLAGS="-L/lib -L${~~pkgs.linuxPackages.nvidia_x11~~}/lib"

export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"

export EXTRA_CCFLAGS="-I/usr/include"

'';

}).env

</~~nowiki~~>}}

</syntaxhighlight>

* By making a nix-shell

~~{{file|cuda-shell~~.nix|nix~~|<nowiki>~~

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

# flake.nix, run with `nix develop`# Run with `nix-shell cuda-shell.nix`

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

let

nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;

in

pkgs.mkShell {

name = "cuda-env-shell";

Line 66:

Line 71:

git gitRepo gnupg autoconf curl

procps gnumake util-linux m4 gperf unzip

cudatoolkit ~~linuxPackages.nvidia_x11~~

cudatoolkit nvidiaPackage

libGLU libGL

xorg.libXi xorg.libXmu freeglut

Line 74:

Line 79:

shellHook = ''

export CUDA_PATH=${pkgs.cudatoolkit}

# export LD_LIBRARY_PATH=${~~pkgs.linuxPackages.nvidia_x11~~}/lib:${pkgs.~~ncurses5~~}/lib

# export LD_LIBRARY_PATH=${nvidiaPackage}/lib:${pkgs.ncurses}/lib

export EXTRA_LDFLAGS="-L/lib -L${~~pkgs.linuxPackages.nvidia_x11~~}/lib"

export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"

export EXTRA_CCFLAGS="-I/usr/include"

'';

}

</syntaxhighlight>

* By making a flake.nix

<syntaxhighlight lang="nix" line="1" start="1"># flake.nix, run with `nix develop`

{

description = "CUDA development environment";

outputs = {

self,

nixpkgs,

}: let

system = "x86_64-linux";

pkgs = import nixpkgs {

inherit system;

config.allowUnfree = true;

config.cudaSupport = true;

config.cudaVersion = "12";

};

# Change according to the driver used: stable, beta

nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;

in {

# alejandra is a nix formatter with a beautiful output

formatter."${system}" = nixpkgs.legacyPackages.${system}.alejandra;

devShells.${system}.default = pkgs.mkShell {

buildInputs = with pkgs; [

ffmpeg

fmt.dev

cudaPackages.cuda_cudart

cudatoolkit

nvidiaPackage

cudaPackages.cudnn

libGLU

libGL

xorg.libXi

xorg.libXmu

freeglut

xorg.libXext

xorg.libX11

xorg.libXv

xorg.libXrandr

zlib

ncurses

stdenv.cc

binutils

uv

];

shellHook = ''

export LD_LIBRARY_PATH="${nvidiaPackage}/lib:$LD_LIBRARY_PATH"

export CUDA_PATH=${pkgs.cudatoolkit}

export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"

export EXTRA_CCFLAGS="-I/usr/include"

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

export PKG_CONFIG_PATH="${pkgs.fmt.dev}/lib/pkgconfig:$PKG_CONFIG_PATH"

'';

};

}</syntaxhighlight>

== Setting up CUDA Binary Cache ==

The [https://nix-community.org/cache/ Nix-community cache] contains pre-built CUDA packages. By adding it to your system, Nix will fetch these packages instead of building them, saving valuable time and processing power.

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

{{warning|1=You need to rebuild your system at least once after adding the cache, before it can be used.}}

=== NixOS ===

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

{{file|/etc/nixos/configuration.nix|nix|<nowiki>

nix.settings = {

substituters = [

"https://nix-community.cachix.org"

];

trusted-public-keys = [

"nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs="

];

};

</nowiki>}}

=== Non-NixOS ===

If you have [https://www.cachix.org cachix] installed and set up, all you need to do is run:

$ cachix use nix-community

</syntaxHighlight>

Else, you have to add <code>substituters</code> and <code>trusted-public-keys</code> to <code>/etc/nix/nix.conf</code>:

{{file|/etc/nix/nix.conf|nix|<nowiki>

trusted-public-keys = nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs=

trusted-substituters = https://nix-community.cachix.org

trusted-users = root @wheel

</nowiki>}}

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

{{file|~/.config/nix/nix.conf|nix|<nowiki>

substituters = https://nix-community.cachix.org

</nowiki>}}

@@ Line 1: / Line 1: @@
-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/projects/27 @NixOS/cuda-maintainers team] on GitHub. If you have an issue using your NVIDIA GPU for computing purposes [https://github.com/nixos/nixpkgs/issues/new open an issue] on GitHub and tag @NixOS/cuda-maintainers.
+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>.
-'''Cache''': Using the [https://app.cachix.org/cache/cuda-maintainers#pull cuda-maintainers cache] is recommended! It will save you valuable time and electrons. Getting set up should be as simple as <code>cachix use cuda-maintainers</code>.
+{{tip|1='''Cache''': Using the [https://app.cachix.org/cache/nix-community nix-community cache] is recommended! It will save you valuable time and electrons. Getting set up should be as simple as <code>cachix use nix-community</code>. Click [[#Setting up CUDA Binary Cache|here]] for more details.}}
-'''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.
+{{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 ==
+{{outdated|scope=section|date=July 2024|reason=Note that these examples have been updated more recently (as of 2024-07-30). May not be the best solution. A better resource is likely the packaging CUDA sample code [https://github.com/NixOS/nixpkgs/tree/master/pkgs/development/cuda-modules/cutensor here].}}
 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].
 Several "CUDA-X" libraries are packages as well. In particular,
-* cuDNN is packaged [https://github.com/NixOS/nixpkgs/blob/master/pkgs/development/libraries/science/math/cudnn/generic.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/blob/master/pkgs/development/libraries/science/math/cutensor/generic.nix here].
+* cuTENSOR is packaged [https://github.com/NixOS/nixpkgs/tree/master/pkgs/development/cuda-modules/cutensor here].
-'''Note that these examples haven't been updated in a while (as of 2022-03-12). May not be the best solution. A better resource is likely the packaging CUDA sample code [https://github.com/NixOS/nixpkgs/blob/master/pkgs/test/cuda/cuda-library-samples/generic.nix here].'''
 There are some possible ways to setup a development environment using CUDA on NixOS. This can be accomplished in the following ways:
@@ Line 19: / Line 17: @@
 * By making a FHS user env
-{{file|cuda-fhs.nix|nix|<nowiki>
+<syntaxhighlight lang="nix" line="1" start="1"># flake.nix, run with `nix develop`
 # Run with `nix-shell cuda-fhs.nix`
 { pkgs ? import </nowiki><nixpkgs><nowiki> {} }:
-(pkgs.buildFHSUserEnv {
+let
+   # Change according to the driver used: stable, beta
+   nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;
+in
+(pkgs.buildFHSEnv {
    name = "cuda-env";
    targetPkgs = pkgs: with pkgs; [
@@ Line 37: / Line 39: @@
      unzip
      cudatoolkit
-     linuxPackages.nvidia_x11
+     nvidiaPackage
      libGLU libGL
      xorg.libXi xorg.libXmu freeglut
@@ Line 49: / Line 51: @@
    profile = ''
      export CUDA_PATH=${pkgs.cudatoolkit}
-     # export LD_LIBRARY_PATH=${pkgs.linuxPackages.nvidia_x11}/lib
+     # export LD_LIBRARY_PATH=${nvidiaPackage}/lib
-     export EXTRA_LDFLAGS="-L/lib -L${pkgs.linuxPackages.nvidia_x11}/lib"
+     export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"
      export EXTRA_CCFLAGS="-I/usr/include"
    '';
 }).env
-</nowiki>}}
+</syntaxhighlight>
 * By making a nix-shell
-{{file|cuda-shell.nix|nix|<nowiki>
+<syntaxhighlight lang="nix" line="1" start="1">
-# Run with `nix-shell cuda-shell.nix`
+# flake.nix, run with `nix develop`# Run with `nix-shell cuda-shell.nix`
 { pkgs ? import </nowiki><nixpkgs><nowiki> {} }:
+let
+   nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;
+in
 pkgs.mkShell {
     name = "cuda-env-shell";
@@ Line 66: / Line 71: @@
       git gitRepo gnupg autoconf curl
       procps gnumake util-linux m4 gperf unzip
-      cudatoolkit linuxPackages.nvidia_x11
+      cudatoolkit nvidiaPackage
       libGLU libGL
       xorg.libXi xorg.libXmu freeglut
@@ Line 74: / Line 79: @@
     shellHook = ''
        export CUDA_PATH=${pkgs.cudatoolkit}
-       # export LD_LIBRARY_PATH=${pkgs.linuxPackages.nvidia_x11}/lib:${pkgs.ncurses5}/lib
+       # export LD_LIBRARY_PATH=${nvidiaPackage}/lib:${pkgs.ncurses}/lib
-       export EXTRA_LDFLAGS="-L/lib -L${pkgs.linuxPackages.nvidia_x11}/lib"
+       export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"
        export EXTRA_CCFLAGS="-I/usr/include"
     '';
 }
+</syntaxhighlight>
+* By making a flake.nix
+<syntaxhighlight lang="nix" line="1" start="1"># flake.nix, run with `nix develop`
+{
+  description = "CUDA development environment";
+  outputs = {
+    self,
+    nixpkgs,
+  }: let
+    system = "x86_64-linux";
+    pkgs = import nixpkgs {
+      inherit system;
+      config.allowUnfree = true;
+      config.cudaSupport = true;
+      config.cudaVersion = "12";
+    };
+    # Change according to the driver used: stable, beta
+    nvidiaPackage = pkgs.linuxPackages.nvidiaPackages.stable;
+  in {
+    # alejandra is a nix formatter with a beautiful output
+    formatter."${system}" = nixpkgs.legacyPackages.${system}.alejandra;
+    devShells.${system}.default = pkgs.mkShell {
+      buildInputs = with pkgs; [
+        ffmpeg
+        fmt.dev
+        cudaPackages.cuda_cudart
+        cudatoolkit
+        nvidiaPackage
+        cudaPackages.cudnn
+        libGLU
+        libGL
+        xorg.libXi
+        xorg.libXmu
+        freeglut
+        xorg.libXext
+        xorg.libX11
+        xorg.libXv
+        xorg.libXrandr
+        zlib
+        ncurses
+        stdenv.cc
+        binutils
+        uv
+      ];
+      shellHook = ''
+        export LD_LIBRARY_PATH="${nvidiaPackage}/lib:$LD_LIBRARY_PATH"
+        export CUDA_PATH=${pkgs.cudatoolkit}
+        export EXTRA_LDFLAGS="-L/lib -L${nvidiaPackage}/lib"
+        export EXTRA_CCFLAGS="-I/usr/include"
+        export CMAKE_PREFIX_PATH="${pkgs.fmt.dev}:$CMAKE_PREFIX_PATH"
+        export PKG_CONFIG_PATH="${pkgs.fmt.dev}/lib/pkgconfig:$PKG_CONFIG_PATH"
+      '';
+    };
+  };
+}</syntaxhighlight>
+== Setting up CUDA Binary Cache ==
+The [https://nix-community.org/cache/ Nix-community cache] contains pre-built CUDA packages. By adding it to your system, Nix will fetch these packages instead of building them, saving valuable time and processing power.
+For more information, refer to the [[Binary Cache#Using a binary cache Using a binary cache|Using a binary cache]] page.
+{{warning|1=You need to rebuild your system at least once after adding the cache, before it can be used.}}
+=== NixOS ===
+Add the cache to <code>substituters</code> and <code>trusted-public-keys</code> inside your system configuration:
+{{file|/etc/nixos/configuration.nix|nix|<nowiki>
+nix.settings = {
+  substituters = [
+    "https://nix-community.cachix.org"
+  ];
+  trusted-public-keys = [
+    "nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs="
+  ];
+};
+</nowiki>}}
+=== Non-NixOS ===
+If you have [https://www.cachix.org cachix] installed and set up, all you need to do is run:
+<syntaxHighlight lang="console">
+$ cachix use nix-community
+</syntaxHighlight>
+Else, you have to add <code>substituters</code> and <code>trusted-public-keys</code> to <code>/etc/nix/nix.conf</code>:
+{{file|/etc/nix/nix.conf|nix|<nowiki>
+trusted-public-keys = nix-community.cachix.org-1:mB9FSh9qf2dCimDSUo8Zy7bkq5CX+/rkCWyvRCYg3Fs=
+trusted-substituters = https://nix-community.cachix.org
+trusted-users = root @wheel
+</nowiki>}}
+If your user is in <code>trusted-users</code>, you can also add the cache in your home directory:
+{{file|~/.config/nix/nix.conf|nix|<nowiki>
+substituters = https://nix-community.cachix.org
 </nowiki>}}