mirror of
https://github.com/ggerganov/llama.cpp.git
synced 2024-12-24 02:14:35 +00:00
flake : update to support metal on m1/m2 (#1724)
This commit is contained in:
parent
4dc62c545d
commit
5b57a5b726
30
flake.lock
30
flake.lock
@ -1,12 +1,15 @@
|
||||
{
|
||||
"nodes": {
|
||||
"flake-utils": {
|
||||
"inputs": {
|
||||
"systems": "systems"
|
||||
},
|
||||
"locked": {
|
||||
"lastModified": 1676283394,
|
||||
"narHash": "sha256-XX2f9c3iySLCw54rJ/CZs+ZK6IQy7GXNY4nSOyu2QG4=",
|
||||
"lastModified": 1685518550,
|
||||
"narHash": "sha256-o2d0KcvaXzTrPRIo0kOLV0/QXHhDQ5DTi+OxcjO8xqY=",
|
||||
"owner": "numtide",
|
||||
"repo": "flake-utils",
|
||||
"rev": "3db36a8b464d0c4532ba1c7dda728f4576d6d073",
|
||||
"rev": "a1720a10a6cfe8234c0e93907ffe81be440f4cef",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
@ -17,11 +20,11 @@
|
||||
},
|
||||
"nixpkgs": {
|
||||
"locked": {
|
||||
"lastModified": 1678470307,
|
||||
"narHash": "sha256-OEeMUr3ueLIXyW/OaFUX5jUdimyQwMg/7e+/Q0gC/QE=",
|
||||
"lastModified": 1685931219,
|
||||
"narHash": "sha256-8EWeOZ6LKQfgAjB/USffUSELPRjw88A+xTcXnOUvO5M=",
|
||||
"owner": "NixOS",
|
||||
"repo": "nixpkgs",
|
||||
"rev": "0c4800d579af4ed98ecc47d464a5e7b0870c4b1f",
|
||||
"rev": "7409480d5c8584a1a83c422530419efe4afb0d19",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
@ -36,6 +39,21 @@
|
||||
"flake-utils": "flake-utils",
|
||||
"nixpkgs": "nixpkgs"
|
||||
}
|
||||
},
|
||||
"systems": {
|
||||
"locked": {
|
||||
"lastModified": 1681028828,
|
||||
"narHash": "sha256-Vy1rq5AaRuLzOxct8nz4T6wlgyUR7zLU309k9mBC768=",
|
||||
"owner": "nix-systems",
|
||||
"repo": "default",
|
||||
"rev": "da67096a3b9bf56a91d16901293e51ba5b49a27e",
|
||||
"type": "github"
|
||||
},
|
||||
"original": {
|
||||
"owner": "nix-systems",
|
||||
"repo": "default",
|
||||
"type": "github"
|
||||
}
|
||||
}
|
||||
},
|
||||
"root": "root",
|
||||
|
26
flake.nix
26
flake.nix
@ -6,6 +6,13 @@
|
||||
outputs = { self, nixpkgs, flake-utils }:
|
||||
flake-utils.lib.eachDefaultSystem (system:
|
||||
let
|
||||
inherit (pkgs.stdenv) isAarch64 isDarwin;
|
||||
inherit (pkgs.lib) optionals;
|
||||
isM1 = isAarch64 && isDarwin;
|
||||
osSpecific =
|
||||
if isM1 then with pkgs.darwin.apple_sdk_11_0.frameworks; [ Accelerate MetalKit MetalPerformanceShaders MetalPerformanceShadersGraph ]
|
||||
else if isDarwin then with pkgs.darwin.apple_sdk.frameworks; [ Accelerate CoreGraphics CoreVideo ]
|
||||
else [ ];
|
||||
pkgs = import nixpkgs {
|
||||
inherit system;
|
||||
};
|
||||
@ -18,17 +25,22 @@
|
||||
packages.default = pkgs.stdenv.mkDerivation {
|
||||
name = "llama.cpp";
|
||||
src = ./.;
|
||||
postPatch =
|
||||
if isM1 then ''
|
||||
substituteInPlace ./ggml-metal.m \
|
||||
--replace '[[NSBundle mainBundle] pathForResource:@"ggml-metal" ofType:@"metal"];' "@\"$out/ggml-metal.metal\";"
|
||||
'' else "";
|
||||
nativeBuildInputs = with pkgs; [ cmake ];
|
||||
buildInputs = with pkgs; lib.optionals stdenv.isDarwin [
|
||||
darwin.apple_sdk.frameworks.Accelerate
|
||||
];
|
||||
cmakeFlags = with pkgs; lib.optionals (system == "aarch64-darwin") [
|
||||
buildInputs = osSpecific;
|
||||
cmakeFlags = [ "-DLLAMA_BUILD_SERVER=ON" ] ++ (optionals isM1 [
|
||||
"-DCMAKE_C_FLAGS=-D__ARM_FEATURE_DOTPROD=1"
|
||||
];
|
||||
"-DLLAMA_METAL=ON"
|
||||
]);
|
||||
installPhase = ''
|
||||
mkdir -p $out/bin
|
||||
mv bin/* $out/bin/
|
||||
mv $out/bin/main $out/bin/llama
|
||||
mv $out/bin/server $out/bin/llama-server
|
||||
|
||||
echo "#!${llama-python}/bin/python" > $out/bin/convert.py
|
||||
cat ${./convert.py} >> $out/bin/convert.py
|
||||
@ -40,9 +52,7 @@
|
||||
packages = with pkgs; [
|
||||
cmake
|
||||
llama-python
|
||||
] ++ lib.optionals stdenv.isDarwin [
|
||||
darwin.apple_sdk.frameworks.Accelerate
|
||||
];
|
||||
] ++ osSpecific;
|
||||
};
|
||||
}
|
||||
);
|
||||
|
Loading…
Reference in New Issue
Block a user