starred/ollama-ollama
1
0
Fork 0
mirror of https://github.com/ollama/ollama.git synced 2026-04-18 00:03:27 +02:00
Code Issues Packages Projects Releases Wiki Activity

MLX: add header vendoring and remove go build tag (#14642)

Browse Source 
* prefer rocm v6 on windows

Avoid building with v7 - more changes are needed

* MLX: add header vendoring and remove go build tag

This switches to using a vendoring approach for the mlx-c headers so that Go
can build without requiring a cmake first.  This enables building the new MLX
based code by default.  Every time cmake runs, the headers are refreshed, so we
can easily keep them in sync when we bump mlx versions.  Basic Windows
and Linux support are verified.

* ci: harden for flaky choco repo servers

CI sometimes fails due to choco not actually installing cache.  Since it just speeds up the build, we can proceed without.

* review comments
This commit is contained in:
Daniel Hiltgen
2026-03-09 17:24:45 -07:00
committed by GitHub
parent 3e06bde643
commit 10e51c5177
142 changed files with 5350 additions and 1064 deletions
Download Patch File Download Diff File Expand all files Collapse all files

64
.github/workflows/test.yaml vendored
View File

@@ -37,7 +37,7 @@ jobs:
| xargs python3 -c "import sys; from pathlib import Path; print(any(Path(x).match(glob) for x in sys.argv[1:] for glob in '$*'.split(' ')))"
}
echo changed=$(changed 'llama/llama.cpp/**/*' 'ml/backend/ggml/ggml/**/*') | tee -a $GITHUB_OUTPUT
echo changed=$(changed 'llama/llama.cpp/**/*' 'ml/backend/ggml/ggml/**/*' '.github/**/*') | tee -a $GITHUB_OUTPUT
echo vendorsha=$(make -f Makefile.sync print-base) | tee -a $GITHUB_OUTPUT
linux:
@@ -60,6 +60,10 @@ jobs:
mesa-vulkan-drivers vulkan-tools
libvulkan1 libvulkan-dev
vulkan-sdk cmake ccache g++ make
- preset: 'MLX CUDA 13'
container: nvidia/cuda:13.0.0-devel-ubuntu22.04
extra-packages: libcudnn9-dev-cuda-13 libopenblas-dev liblapack-dev liblapacke-dev git curl
flags: '-DCMAKE_CUDA_ARCHITECTURES=87 -DBLAS_INCLUDE_DIRS=/usr/include/x86_64-linux-gnu -DLAPACK_INCLUDE_DIRS=/usr/include/x86_64-linux-gnu'
runs-on: linux
container: ${{ matrix.container }}
steps:
@@ -76,6 +80,10 @@ jobs:
$sudo apt-get update
fi
$sudo apt-get install -y cmake ccache ${{ matrix.extra-packages }}
# MLX requires CMake 3.25+, install from official releases
if [ "${{ matrix.preset }}" = "MLX CUDA 13" ]; then
curl -fsSL https://github.com/Kitware/CMake/releases/download/v3.31.2/cmake-3.31.2-linux-$(uname -m).tar.gz | $sudo tar xz -C /usr/local --strip-components 1
fi
# Export VULKAN_SDK if provided by LunarG package (defensive)
if [ -d "/usr/lib/x86_64-linux-gnu/vulkan" ] && [ "${{ matrix.preset }}" = "Vulkan" ]; then
echo "VULKAN_SDK=/usr" >> $GITHUB_ENV
@@ -87,8 +95,8 @@ jobs:
path: /github/home/.cache/ccache
key: ccache-${{ runner.os }}-${{ runner.arch }}-${{ matrix.preset }}-${{ needs.changes.outputs.vendorsha }}
- run: |
cmake --preset ${{ matrix.preset }} ${{ matrix.flags }}
cmake --build --preset ${{ matrix.preset }} --parallel
cmake --preset "${{ matrix.preset }}" ${{ matrix.flags }}
cmake --build --preset "${{ matrix.preset }}" --parallel
windows:
needs: [changes]
@@ -114,12 +122,31 @@ jobs:
flags: '-DAMDGPU_TARGETS=gfx1010 -DCMAKE_C_COMPILER=clang -DCMAKE_CXX_COMPILER=clang++ -DCMAKE_C_FLAGS="-parallel-jobs=4 -Wno-ignored-attributes -Wno-deprecated-pragma" -DCMAKE_CXX_FLAGS="-parallel-jobs=4 -Wno-ignored-attributes -Wno-deprecated-pragma"'
- preset: Vulkan
install: https://sdk.lunarg.com/sdk/download/1.4.321.1/windows/vulkansdk-windows-X64-1.4.321.1.exe
- preset: 'MLX CUDA 13'
install: https://developer.download.nvidia.com/compute/cuda/13.0.0/local_installers/cuda_13.0.0_windows.exe
cudnn-install: https://developer.download.nvidia.com/compute/cudnn/redist/cudnn/windows-x86_64/cudnn-windows-x86_64-9.18.1.3_cuda13-archive.zip
flags: '-DCMAKE_CUDA_ARCHITECTURES=80'
cuda-components:
- '"cudart"'
- '"nvcc"'
- '"cublas"'
- '"cublas_dev"'
- '"cufft"'
- '"cufft_dev"'
- '"nvrtc"'
- '"nvrtc_dev"'
- '"crt"'
- '"nvvm"'
- '"nvptxcompiler"'
cuda-version: '13.0'
runs-on: windows
steps:
- run: |
choco install -y --no-progress ccache ninja
ccache -o cache_dir=${{ github.workspace }}\.ccache
- if: matrix.preset == 'CUDA' || matrix.preset == 'ROCm' || matrix.preset == 'Vulkan'
if (Get-Command ccache -ErrorAction SilentlyContinue) {
ccache -o cache_dir=${{ github.workspace }}\.ccache
}
- if: matrix.preset == 'CUDA' || matrix.preset == 'ROCm' || matrix.preset == 'Vulkan' || matrix.preset == 'MLX CUDA 13'
id: cache-install
uses: actions/cache/restore@v4
with:
@@ -127,8 +154,9 @@ jobs:
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA
C:\Program Files\AMD\ROCm
C:\VulkanSDK
key: ${{ matrix.install }}
- if: matrix.preset == 'CUDA'
C:\Program Files\NVIDIA\CUDNN
key: ${{ matrix.install }}-${{ matrix.cudnn-install }}
- if: matrix.preset == 'CUDA' || matrix.preset == 'MLX CUDA 13'
name: Install CUDA ${{ matrix.cuda-version }}
run: |
$ErrorActionPreference = "Stop"
@@ -164,10 +192,27 @@ jobs:
Invoke-WebRequest -Uri "${{ matrix.install }}" -OutFile "install.exe"
Start-Process -FilePath .\install.exe -ArgumentList "-c","--am","--al","in" -NoNewWindow -Wait
}
$vulkanPath = (Resolve-Path "C:\VulkanSDK\*").path
echo "$vulkanPath\bin" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
echo "VULKAN_SDK=$vulkanPath" >> $env:GITHUB_ENV
- if: matrix.preset == 'MLX CUDA 13'
name: Install cuDNN for MLX
run: |
$ErrorActionPreference = "Stop"
$cudnnRoot = "C:\Program Files\NVIDIA\CUDNN"
if ("${{ steps.cache-install.outputs.cache-hit }}" -ne 'true') {
Invoke-WebRequest -Uri "${{ matrix.cudnn-install }}" -OutFile "cudnn.zip"
Expand-Archive -Path cudnn.zip -DestinationPath cudnn-extracted
$cudnnDir = (Get-ChildItem -Path cudnn-extracted -Directory)[0].FullName
New-Item -ItemType Directory -Force -Path $cudnnRoot
Copy-Item -Path "$cudnnDir\*" -Destination "$cudnnRoot\" -Recurse
}
echo "CUDNN_ROOT_DIR=$cudnnRoot" | Out-File -FilePath $env:GITHUB_ENV -Append
echo "CUDNN_INCLUDE_PATH=$cudnnRoot\include" | Out-File -FilePath $env:GITHUB_ENV -Append
echo "CUDNN_LIBRARY_PATH=$cudnnRoot\lib\x64" | Out-File -FilePath $env:GITHUB_ENV -Append
echo "$cudnnRoot\bin\x64" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append
- if: ${{ !cancelled() && steps.cache-install.outputs.cache-hit != 'true' }}
uses: actions/cache/save@v4
with:
@@ -175,7 +220,8 @@ jobs:
C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA
C:\Program Files\AMD\ROCm
C:\VulkanSDK
key: ${{ matrix.install }}
C:\Program Files\NVIDIA\CUDNN
key: ${{ matrix.install }}-${{ matrix.cudnn-install }}
- uses: actions/checkout@v4
- uses: actions/cache@v4
with: