vbatts/llama.cpp
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

cmake : make LLAMA_NATIVE flag actually use the instructions supported by the processor (#3273)

Browse source 
* fix LLAMA_NATIVE

* syntax

* alternate implementation

* my eyes must be getting bad...

* set cmake LLAMA_NATIVE=ON by default

* march=native doesn't work for ios/tvos, so disable for those targets. also see what happens if we use it on msvc

* revert 8283237 and only allow LLAMA_NATIVE on x86 like the Makefile

* remove -DLLAMA_MPI=ON

---------

Co-authored-by: netrunnereve <netrunnereve@users.noreply.github.com>
This commit is contained in:
Eve 2023-10-03 16:53:15 +00:00 • committed by GitHub
parent ff5a3f0c09
commit 017efe899d
No known key found for this signature in database
GPG key ID: 4AEE18F83AFDEB23
3 changed files with 26 additions and 20 deletions
Download patch file Download diff file Expand all files Collapse all files

2
flake.nix
View file

@ -62,7 +62,7 @@
mkdir -p $out/include
cp ${src}/llama.h $out/include/
'';
cmakeFlags = [ "-DLLAMA_BUILD_SERVER=ON" "-DLLAMA_MPI=ON" "-DBUILD_SHARED_LIBS=ON" "-DCMAKE_SKIP_BUILD_RPATH=ON" ];
cmakeFlags = [ "-DLLAMA_NATIVE=OFF" "-DLLAMA_BUILD_SERVER=ON" "-DBUILD_SHARED_LIBS=ON" "-DCMAKE_SKIP_BUILD_RPATH=ON" ];
in
{
packages.default = pkgs.stdenv.mkDerivation {