Actions: Unblock Metal #71
This file contains bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
| name: Build binaries | |
| on: | |
| push: | |
| branches: [main] | |
| jobs: | |
| unix-build: | |
| runs-on: ${{ matrix.os }} | |
| strategy: | |
| matrix: | |
| os: [ubuntu-22.04, macos-15] | |
| device: [cpu, metal, cuda] | |
| exclude: | |
| - os: macos-15 | |
| device: cpu | |
| - os: macos-15 | |
| device: cuda | |
| - os: ubuntu-22.04 | |
| device: metal | |
| container: ${{ matrix.device == 'cuda' && 'nvidia/cuda:12.8.0-devel-ubuntu22.04' || '' }} | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }}-${{ matrix.os }}-${{ matrix.device }} | |
| cancel-in-progress: true | |
| defaults: | |
| run: | |
| shell: bash | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Install System Dependencies | |
| if: ${{ runner.os == 'Linux' && matrix.device == 'cuda' }} | |
| run: | | |
| apt update | |
| apt install -y cmake build-essential zip unzip libgomp1 git | |
| - name: Setup Deno | |
| uses: denoland/setup-deno@v2 | |
| with: | |
| deno-version: v2.x | |
| - name: Setup Ninja | |
| uses: seanmiddleditch/gha-setup-ninja@master | |
| - name: Run sccache-cache | |
| uses: mozilla-actions/sccache-action@v0.0.7 | |
| - name: Configure sccache | |
| id: sccache | |
| run: | | |
| export SCCACHE_DIR="$PWD/bindings/.sccache" | |
| echo "SCCACHE_DIR=$SCCACHE_DIR" >> $GITHUB_ENV | |
| - name: Cache sccache storage | |
| uses: actions/cache@v4 | |
| with: | |
| path: ${{ env.SCCACHE_DIR }} | |
| key: sccache-${{ runner.os }}-${{ matrix.device }}-${{ github.ref_name }}-${{ github.run_id }} | |
| restore-keys: | | |
| sccache-${{ runner.os }}-${{ matrix.device }}-${{ github.ref_name }}- | |
| sccache-${{ runner.os }}-${{ matrix.device }}- | |
| - name: Get short SHA | |
| id: sha | |
| run: | | |
| LONG_SHA=${{ github.sha }} | |
| echo "sha=${LONG_SHA::7}" >> $GITHUB_OUTPUT | |
| - name: Install dependencies | |
| run: deno install | |
| - name: Save SHA to textfile | |
| run: echo ${{ steps.sha.outputs.sha }} > gitSha.txt | |
| - name: Build bindings and compile | |
| run: | | |
| if [ "${{ matrix.device }}" = "cuda" ]; then | |
| export GGML_CUDA=1 | |
| export CMAKE_CUDA_ARCHITECTURES="75;86;89;120" | |
| fi | |
| deno task bindings | |
| deno task compile | |
| chmod +x YALS | |
| - name: Copy CUDA libraries | |
| run: | | |
| # Move native CUDA lib files | |
| cp /usr/local/cuda/lib64/libcudart.so.12.* lib | |
| cp /usr/local/cuda/lib64/libcublas.so.12.* lib | |
| cp /usr/local/cuda/lib64/libcublasLt.so.12.* lib | |
| # Create symlinks | |
| cd lib | |
| mv libcudart.so* libcudart.so.12 | |
| mv libcublas.so* libcublas.so.12 | |
| mv libcublasLt.so* libcublasLt.so.12 | |
| if: ${{ matrix.device == 'cuda' }} | |
| - name: Set folder name | |
| id: folder-name | |
| run: echo "name=YALS-${{ runner.os }}-${{ matrix.device }}-${{ steps.sha.outputs.sha }}" >> $GITHUB_OUTPUT | |
| - name: Move to artifact folder | |
| run: | | |
| mkdir ${{ steps.folder-name.outputs.name }} | |
| mv lib models templates config_sample.yml YALS ${{ steps.folder-name.outputs.name }} | |
| # - name: Upload artifact | |
| # uses: actions/upload-artifact@v4 | |
| # with: | |
| # name: ${{ steps.folder-name.outputs.name }} | |
| # path: ./${{ steps.folder-name.outputs.name }} | |
| # if-no-files-found: error | |
| - name: Zip files | |
| run: | | |
| cd ${{ steps.folder-name.outputs.name }} | |
| zip -r ${{ steps.folder-name.outputs.name }}.zip YALS lib/* models/* templates/* config_sample.yml | |
| mv ${{ steps.folder-name.outputs.name }}.zip ../ | |
| - name: Upload release | |
| uses: svenstaro/upload-release-action@v2 | |
| with: | |
| file: ${{ steps.folder-name.outputs.name }}.zip | |
| tag: ${{ steps.sha.outputs.sha }} | |
| overwrite: true | |
| release_name: YALS-${{ steps.sha.outputs.sha }} | |
| body: | | |
| CI Builds of YALS commit [${{ steps.sha.outputs.sha }}](https://github.com/${{ github.repository }}/commit/${{ steps.sha.outputs.sha }}) | |
| For more details, please refer to the commit history. | |
| Builds: | |
| - **CPU**: Compiled with AVX1 and AVX2 support. | |
| - **CUDA** - Built for CUDA 12, optimized for Turing and newer consumer GPUs. | |
| - **Metal** - Includes both Metal and CPU support for macOS. | |
| windows-build: | |
| runs-on: ${{ matrix.os }} | |
| strategy: | |
| matrix: | |
| os: [windows-2022] | |
| device: [cpu, cuda] | |
| concurrency: | |
| group: ${{ github.workflow }}-${{ github.ref }}-${{ matrix.os }}-${{ matrix.device }} | |
| cancel-in-progress: true | |
| steps: | |
| - uses: actions/checkout@v4 | |
| - name: Install CUDA 12.8 | |
| if: ${{ matrix.device == 'cuda' }} | |
| run: | | |
| mkdir -p "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" | |
| choco install unzip -y | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_cudart/windows-x86_64/cuda_cudart-windows-x86_64-12.8.57-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvcc/windows-x86_64/cuda_nvcc-windows-x86_64-12.8.61-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvrtc/windows-x86_64/cuda_nvrtc-windows-x86_64-12.8.61-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/libcublas/windows-x86_64/libcublas-windows-x86_64-12.8.3.14-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvtx/windows-x86_64/cuda_nvtx-windows-x86_64-12.8.55-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_profiler_api/windows-x86_64/cuda_profiler_api-windows-x86_64-12.8.55-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/visual_studio_integration/windows-x86_64/visual_studio_integration-windows-x86_64-12.8.55-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_nvprof/windows-x86_64/cuda_nvprof-windows-x86_64-12.8.57-archive.zip" | |
| curl -O "https://developer.download.nvidia.com/compute/cuda/redist/cuda_cccl/windows-x86_64/cuda_cccl-windows-x86_64-12.8.55-archive.zip" | |
| unzip '*.zip' -d "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_cudart-windows-x86_64-12.8.57-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_nvcc-windows-x86_64-12.8.61-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_nvrtc-windows-x86_64-12.8.61-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libcublas-windows-x86_64-12.8.3.14-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_nvtx-windows-x86_64-12.8.55-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_profiler_api-windows-x86_64-12.8.55-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\visual_studio_integration-windows-x86_64-12.8.55-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_nvprof-windows-x86_64-12.8.57-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\cuda_cccl-windows-x86_64-12.8.55-archive\*" "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" /E /I /H /Y | |
| echo "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\bin" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append | |
| echo "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\libnvvp" | Out-File -FilePath $env:GITHUB_PATH -Encoding utf8 -Append | |
| echo "CUDA_PATH=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8 | |
| echo "CUDA_PATH_V12_8=C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8" | Out-File -FilePath $env:GITHUB_ENV -Append -Encoding utf8 | |
| - name: Setup Deno | |
| uses: denoland/setup-deno@v2 | |
| with: | |
| deno-version: v2.x | |
| - name: Setup Ninja | |
| uses: seanmiddleditch/gha-setup-ninja@master | |
| - name: Run sccache-cache | |
| uses: mozilla-actions/sccache-action@v0.0.7 | |
| - name: Configure sccache | |
| run: | | |
| $env:SCCACHE_DIR="$PWD/bindings/.sccache" | |
| echo "SCCACHE_DIR=$env:SCCACHE_DIR" >> $env:GITHUB_ENV | |
| - name: Cache sccache storage | |
| uses: actions/cache@v4 | |
| with: | |
| path: ${{ env.SCCACHE_DIR }} | |
| key: sccache-${{ runner.os }}-${{ matrix.device }}-${{ github.ref_name }}-${{ github.run_id }} | |
| restore-keys: | | |
| sccache-${{ runner.os }}-${{ matrix.device }}-${{ github.ref_name }}- | |
| sccache-${{ runner.os }}-${{ matrix.device }}- | |
| - name: Get short SHA | |
| id: sha | |
| run: echo "sha=$("${{ github.sha }}".Substring(0,7))" >> $env:GITHUB_OUTPUT | |
| - name: Install dependencies | |
| run: deno install | |
| - name: Save SHA to textfile | |
| run: Set-Content -Path "gitSha.txt" -Value "${{ steps.sha.outputs.sha }}" | |
| - name: Build bindings and compile | |
| run: | | |
| $OldDir = $PWD | |
| Import-Module 'C:\Program Files\Microsoft Visual Studio\2022\Enterprise\Common7\Tools\Microsoft.VisualStudio.DevShell.dll' | |
| Enter-VsDevShell -VsInstallPath 'C:\Program Files\Microsoft Visual Studio\2022\Enterprise' -DevCmdArguments '-arch=x64 -host_arch=x64' | |
| cd $OldDir | |
| if ('${{ matrix.device }}' -eq 'cuda') { | |
| $env:GGML_CUDA = 1 | |
| $env:CMAKE_CUDA_ARCHITECTURES="75;86;89;120" | |
| } | |
| deno task bindings-win | |
| deno task compile-win | |
| - name: Copy vcruntime | |
| run: | | |
| cp "C:\Windows\System32\vcruntime140.dll" lib | |
| cp "C:\Windows\System32\vcruntime140_1.dll" lib | |
| - name: Copy CUDA libraries | |
| run: | | |
| cp "${env:CUDA_PATH}\bin\cudart64_*.dll" lib | |
| cp "${env:CUDA_PATH}\bin\cublas64_*.dll" lib | |
| cp "${env:CUDA_PATH}\bin\cublasLt64_*.dll" lib | |
| if: ${{ matrix.device == 'cuda' }} | |
| - name: Set folder name | |
| id: folder-name | |
| run: echo "name=YALS-${{ runner.os }}-${{ matrix.device }}-${{ steps.sha.outputs.sha }}" >> $env:GITHUB_OUTPUT | |
| - name: Move to artifact folder | |
| run: | | |
| mkdir ${{ steps.folder-name.outputs.name }} | |
| mv YALS.exe,lib,models,templates,config_sample.yml ${{ steps.folder-name.outputs.name }} | |
| # - name: Upload artifact | |
| # uses: actions/upload-artifact@v4 | |
| # with: | |
| # name: ${{ steps.folder-name.outputs.name }} | |
| # path: ./${{ steps.folder-name.outputs.name }} | |
| # if-no-files-found: error | |
| - name: Zip files | |
| run: | | |
| cd ${{ steps.folder-name.outputs.name }} | |
| Compress-Archive -Path @('.\YALS.exe', '.\lib', '.\models', '.\templates', '.\config_sample.yml') -DestinationPath ${{ steps.folder-name.outputs.name }}.zip | |
| mv ${{ steps.folder-name.outputs.name }}.zip ../ | |
| - name: Upload release | |
| uses: svenstaro/upload-release-action@v2 | |
| with: | |
| file: ${{ steps.folder-name.outputs.name }}.zip | |
| tag: ${{ steps.sha.outputs.sha }} | |
| overwrite: true | |
| release_name: YALS-${{ steps.sha.outputs.sha }} | |
| body: | | |
| CI Builds of YALS commit [${{ steps.sha.outputs.sha }}](https://github.com/${{ github.repository }}/commit/${{ steps.sha.outputs.sha }}) | |
| For more details, please refer to the commit history. | |
| Builds: | |
| - **CPU**: Compiled with AVX1 and AVX2 support. | |
| - **CUDA** - Built for CUDA 12, optimized for Turing and newer consumer GPUs. | |
| - **Metal** - Includes both Metal and CPU support for macOS. |