Merge pull request #72 from DarthAffe/build

Improved builds to be closer to original sd.cpp builds; Fixed windows…

.github/workflows/backends.yml

@@ -31,7 +31,7 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
@@ -58,41 +58,19 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
           submodules: recursive
 
-      - name:  Install Cuda Toolkit 12.8
-        id: cuda-12
-        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
-          xcopy "C:\Program Files\NVIDIA GPU Computing Toolkit\CUDA\v12.8\visual_studio_integration\MSBuildExtensions\*" "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\MSBuild\Microsoft\VC\v170\BuildCustomizations" /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: Install cuda-toolkit
+        id: cuda-toolkit
+        uses: Jimver/cuda-toolkit@v0.2.22
+        with:
+          cuda: "12.8.1"
+          method: "network"
+          sub-packages: '["nvcc", "cudart", "cublas", "cublas_dev", "thrust", "visual_studio_integration"]'
 
       - name: Build
         id: cmake_build
@@ -101,7 +79,7 @@ jobs:
           mkdir build
           cd build
           call "C:\Program Files\Microsoft Visual Studio\2022\Enterprise\VC\Auxiliary\Build\vcvars64.bat"
-          cmake .. -DSD_CUDA=ON -DCMAKE_CUDA_ARCHITECTURES="61;75;86;89;100" -DSD_BUILD_SHARED_LIBS=ON
+          cmake .. -DSD_CUDA=ON -DCMAKE_CUDA_ARCHITECTURES="61;70;75;80;86;89;90;100;120" -DSD_BUILD_SHARED_LIBS=ON
           cmake --build . --config Release
       
       - name: Upload artifact
@@ -114,18 +92,53 @@ jobs:
   windows-hip:
     runs-on: windows-latest
 
-    steps:
-      - name: Install
-        id: depends
+    env:
+      HIPSDK_INSTALLER_VERSION: "25.Q3"
+      GPU_TARGETS: "gfx1151;gfx1200;gfx1201;gfx1100;gfx1101;gfx1102;gfx1030;gfx1031;gfx1032"
+      
+    steps:          
+      - name: Checkout
+        id: checkout
+        uses: actions/checkout@v6
+        with:
+          repository: 'leejet/stable-diffusion.cpp'
+          ref: '${{ github.event.inputs.commit }}'
+          submodules: recursive  
+
+      - name: Cache ROCm Installation
+        id: cache-rocm
+        uses: actions/cache@v4
+        with:
+          path: C:\Program Files\AMD\ROCm
+          key: rocm-${{ env.HIPSDK_INSTALLER_VERSION }}-${{ runner.os }}
+
+      - name: ccache
+        uses: ggml-org/ccache-action@v1.2.16
+        with:
+          key: windows-latest-cmake-hip-${{ env.HIPSDK_INSTALLER_VERSION }}-x64
+          evict-old-files: 1d
+
+      - name: Install ROCm
+        if: steps.cache-rocm.outputs.cache-hit != 'true'
         run: |
           $ErrorActionPreference = "Stop"
           write-host "Downloading AMD HIP SDK Installer"
-          Invoke-WebRequest -Uri "https://download.amd.com/developer/eula/rocm-hub/AMD-Software-PRO-Edition-25.Q3-WinSvr2022-For-HIP.exe" -OutFile "${env:RUNNER_TEMP}\rocm-install.exe"
+          Invoke-WebRequest -Uri "https://download.amd.com/developer/eula/rocm-hub/AMD-Software-PRO-Edition-${{ env.HIPSDK_INSTALLER_VERSION }}-WinSvr2022-For-HIP.exe" -OutFile "${env:RUNNER_TEMP}\rocm-install.exe"
           write-host "Installing AMD HIP SDK"
-          Start-Process "${env:RUNNER_TEMP}\rocm-install.exe" -ArgumentList '-install' -NoNewWindow -Wait
+          $proc = Start-Process "${env:RUNNER_TEMP}\rocm-install.exe" -ArgumentList '-install' -NoNewWindow -PassThru
+          $completed = $proc.WaitForExit(600000)
+          if (-not $completed) {
+              Write-Error "ROCm installation timed out after 10 minutes. Killing the process"
+              $proc.Kill()
+              exit 1
+          }
+          if ($proc.ExitCode -ne 0) {
+              Write-Error "ROCm installation failed with exit code $($proc.ExitCode)"
+              exit 1
+          }
           write-host "Completed AMD HIP SDK installation"
+
       - name: Verify ROCm
-        id: verify
         run: |
           # Find and test ROCm installation
           $clangPath = Get-ChildItem 'C:\Program Files\AMD\ROCm\*\bin\clang.exe' | Select-Object -First 1
@@ -134,24 +147,24 @@ jobs:
             exit 1
           }
           & $clangPath.FullName --version
-          
-      - name: Checkout
-        id: checkout
-        uses: actions/checkout@v4.2.2
-        with:
-          repository: 'leejet/stable-diffusion.cpp'
-          ref: '${{ github.event.inputs.commit }}'
-          submodules: recursive  
+          # Set HIP_PATH environment variable for later steps
+          echo "HIP_PATH=$(Resolve-Path 'C:\Program Files\AMD\ROCm\*\bin\clang.exe' | split-path | split-path)" >> $env:GITHUB_ENV
 
       - name: Build
-        id: cmake_build
         run: |
-          $env:HIP_PATH=$(Resolve-Path 'C:\Program Files\AMD\ROCm\*\bin\clang.exe' | split-path | split-path)
-          $env:CMAKE_PREFIX_PATH="${env:HIP_PATH}"
           mkdir build
           cd build
-          cmake -G "Unix Makefiles" .. -DCMAKE_C_COMPILER="${env:HIP_PATH}\bin\clang.exe" -DCMAKE_CXX_COMPILER="${env:HIP_PATH}\bin\clang++.exe" -DGGML_HIP=ON -DCMAKE_BUILD_TYPE=Release -DAMDGPU_TARGETS="gfx1100;gfx1101;gfx1030" -DGGML_RPC=ON -DSD_BUILD_SHARED_LIBS=ON
-          cmake --build . --config Release
+          $env:CMAKE_PREFIX_PATH="${env:HIP_PATH}"
+          cmake .. `
+            -G "Unix Makefiles" `
+            -DSD_HIPBLAS=ON `
+            -DSD_BUILD_SHARED_LIBS=ON `
+            -DGGML_NATIVE=OFF `
+            -DCMAKE_C_COMPILER=clang `
+            -DCMAKE_CXX_COMPILER=clang++ `
+            -DCMAKE_BUILD_TYPE=Release `
+            -DGPU_TARGETS="${{ env.GPU_TARGETS }}"
+          cmake --build . --config Release --parallel ${env:NUMBER_OF_PROCESSORS}
               
       - name: Upload artifact
         id: upload_artifact
@@ -169,7 +182,7 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
@@ -216,12 +229,18 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
           submodules: recursive
 
+      - name: Dependencies
+        id: depends
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential
+
       - name: Build
         id: cmake_build
         run: |
@@ -237,6 +256,39 @@ jobs:
           name: linux-${{ matrix.build }}
           path: ./build/bin/libstable-diffusion.so
           
+  linux-vulkan:
+    runs-on: ubuntu-latest
+            
+    steps:
+      - name: Checkout
+        id: checkout
+        uses: actions/checkout@v6
+        with:
+          repository: 'leejet/stable-diffusion.cpp'
+          ref: '${{ github.event.inputs.commit }}'
+          submodules: recursive
+
+      - name: Dependencies
+        id: depends
+        run: |
+          sudo apt-get update
+          sudo apt-get install build-essential libvulkan-dev glslc
+          
+      - name: Build
+        id: cmake_build
+        run: |
+          mkdir build
+          cd build
+          cmake .. -DSD_BUILD_SHARED_LIBS=ON -DSD_VULKAN=ON
+          cmake --build . --config Release
+                  
+      - name: Upload artifact
+        id: upload_artifact
+        uses: actions/upload-artifact@v4.6.1
+        with:
+          name: linux-vulkan
+          path: ./build/bin/libstable-diffusion.so
+          
   linux-cuda12:
     runs-on: ubuntu-latest
     container: nvidia/cuda:12.8.0-devel-ubuntu24.04
@@ -251,7 +303,7 @@ jobs:
     
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
@@ -278,7 +330,7 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
@@ -313,31 +365,99 @@ jobs:
           path: ./build/bin/libstable-diffusion.so
           
   linux-hip:
-    runs-on: ubuntu-22.04
-    container: rocm/dev-ubuntu-22.04:6.4.4
+    runs-on: ubuntu-latest
+    container: rocm/dev-ubuntu-24.04:7.2
+
+    env:
+      ROCM_VERSION: "7.2"
+      UBUNTU_VERSION: "24.04"
+      GPU_TARGETS: "gfx1151;gfx1150;gfx1100;gfx1101;gfx1102;gfx1200;gfx1201"
    
     steps:
-      - name: Dependencies
-        id: depends
-        run: |
-          sudo apt-get update
-          sudo apt-get install -y build-essential git cmake rocblas-dev hipblas-dev
-          hipconfig
-          
-      - name: Clone
+      - run: apt-get update && apt-get install -y git
+      
+      - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
           submodules: recursive
-                    
+
+      - name: Free disk space
+        run: |
+          # Remove preinstalled SDKs and caches not needed for this job
+          sudo rm -rf /usr/share/dotnet || true
+          sudo rm -rf /usr/local/lib/android || true
+          sudo rm -rf /opt/ghc || true
+          sudo rm -rf /usr/local/.ghcup || true
+          sudo rm -rf /opt/hostedtoolcache || true
+
+          # Remove old package lists and caches
+          sudo rm -rf /var/lib/apt/lists/* || true
+          sudo apt clean
+
+      - name: Dependencies
+        id: depends
+        run: |
+          sudo apt-get update
+          sudo apt install -y \
+            cmake \
+            hip-dev \
+            hipblas-dev \
+            ninja-build \
+            rocm-dev \
+            zip
+          # Clean apt caches to recover disk space
+          sudo apt clean
+          sudo rm -rf /var/lib/apt/lists/* || true
+
+      - name: Setup ROCm Environment
+        run: |
+          # Add ROCm to PATH for current session
+          echo "/opt/rocm/bin" >> $GITHUB_PATH
+
+          # Build regex pattern from ${{ env.GPU_TARGETS }} (match target as substring)
+          TARGET_REGEX="($(printf '%s' "${{ env.GPU_TARGETS }}" | sed 's/;/|/g'))"
+
+          # Remove library files for architectures we're not building for to save disk space
+          echo "Cleaning up unneeded architecture files..."
+          cd /opt/rocm/lib/rocblas/library
+          # Keep only our target architectures
+          for file in *; do
+            if printf '%s' "$file" | grep -q 'gfx'; then
+              if ! printf '%s' "$file" | grep -Eq "$TARGET_REGEX"; then
+                echo "Removing $file" &&
+                sudo rm -f "$file";
+              fi
+            fi
+          done
+
+          cd /opt/rocm/lib/hipblaslt/library
+          for file in *; do
+            if printf '%s' "$file" | grep -q 'gfx'; then
+              if ! printf '%s' "$file" | grep -Eq "$TARGET_REGEX"; then
+                echo "Removing $file" &&
+                sudo rm -f "$file";
+              fi
+            fi
+          done
+
       - name: Build
         id: cmake_build
         run: |
           mkdir build
           cd build
-          cmake .. -DCMAKE_HIP_COMPILER="$(hipconfig -l)/clang" -DGGML_HIP=ON -DSD_BUILD_SHARED_LIBS=ON
+          cmake .. -G Ninja \
+            -DCMAKE_CXX_COMPILER=amdclang++ \
+            -DCMAKE_C_COMPILER=amdclang \
+            -DCMAKE_BUILD_TYPE=Release \
+            -DSD_HIPBLAS=ON \
+            -DGPU_TARGETS="${{ env.GPU_TARGETS }}" \
+            -DAMDGPU_TARGETS="${{ env.GPU_TARGETS }}" \
+            -DCMAKE_BUILD_WITH_INSTALL_RPATH=ON \
+            -DCMAKE_POSITION_INDEPENDENT_CODE=ON \
+            -DSD_BUILD_SHARED_LIBS=ON
           cmake --build . --config Release
                
       - name: Upload artifact Rocm
@@ -365,7 +485,7 @@ jobs:
     steps:          
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
         with:
           repository: 'leejet/stable-diffusion.cpp'
           ref: '${{ github.event.inputs.commit }}'
@@ -396,6 +516,7 @@ jobs:
       - windows-hip
       - windows-vulkan
       - linux-cpu
+      - linux-vulkan
       - linux-cuda12
       - linux-sycl
       - linux-hip
@@ -404,7 +525,7 @@ jobs:
     steps:
       - name: Checkout
         id: checkout
-        uses: actions/checkout@v4.2.2
+        uses: actions/checkout@v6
       
       - name: Download artifacts
         id: download_artifacts

Backends/StableDiffusion.NET.Backend.Vulkan.nuspec

@@ -19,6 +19,7 @@
     <file src="StableDiffusion.NET.Backend.props" target="build/net8.0/StableDiffusion.NET.Backend.Vulkan.props" />
 
     <file src="windows-vulkan/stable-diffusion.dll" target="runtimes\win-x64\native\vulkan\stable-diffusion.dll" />
+    <file src="linux-vulkan/libstable-diffusion.so" target="runtimes\linux-x64\native\vulkan\libstable-diffusion.so" />
     
     <file src="sd_net_vulkan.png" target="sd_net_vulkan.png" />
     <file src="readme.md" target="readme.md" />

StableDiffusion.NET/Backends/VulkanBackend.cs

@@ -13,7 +13,7 @@ public class VulkanBackend : IBackend
 
     public int Priority { get; set; } = 5;
 
-    public bool IsAvailable => RuntimeInformation.IsOSPlatform(OSPlatform.Windows)
+    public bool IsAvailable => (RuntimeInformation.IsOSPlatform(OSPlatform.Windows) || RuntimeInformation.IsOSPlatform(OSPlatform.Linux))
                             && (RuntimeInformation.OSArchitecture == Architecture.X64);
 
     public string PathPart => "vulkan";