Merge branch 'piercefreeman-feature/demo-wheels'

* piercefreeman-feature/demo-wheels: (25 commits)
  Install standard non-wheel package
  Remove release creation
  Build wheel on each push
  Isolate 2.0.0 & cuda12
  Clean setup.py imports
  Remove builder project
  Bump version
  Add notes to github action workflow
  Add torch dependency to final build
  Exclude cuda erroring builds
  Exclude additional disallowed matrix params
  Full version matrix
  Add CUDA 11.7
  Release is actually unsupported
  echo OS version
  Temp disable deploy
  OS version build numbers
  Restore full build matrix
  Refactor and clean of setup.py
  Strip cuda name from torch version
  ...

.github/workflows/cuda/cu102-Linux.sh

@@ -1,6 +1,8 @@
 #!/bin/bash
 
-OS=ubuntu1804
+# Strip the periods from the version number
+OS_VERSION=$(echo $(lsb_release -sr) | tr -d .)
+OS=ubuntu${OS_VERSION}
 
 wget -nv https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-${OS}.pin
 sudo mv cuda-${OS}.pin /etc/apt/preferences.d/cuda-repository-pin-600

.github/workflows/cuda/cu113-Linux.sh

@@ -1,11 +1,17 @@
 #!/bin/bash
 
-OS=ubuntu1804
+# Strip the periods from the version number
+OS_VERSION=$(echo $(lsb_release -sr) | tr -d .)
+OS=ubuntu${OS_VERSION}
 
 wget -nv https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-${OS}.pin
 sudo mv cuda-${OS}.pin /etc/apt/preferences.d/cuda-repository-pin-600
 wget -nv https://developer.download.nvidia.com/compute/cuda/11.3.0/local_installers/cuda-repo-${OS}-11-3-local_11.3.0-465.19.01-1_amd64.deb
 sudo dpkg -i cuda-repo-${OS}-11-3-local_11.3.0-465.19.01-1_amd64.deb
+
+# TODO: If on version < 22.04, install via signal-desktop-keyring
+# For future versions it's deprecated and should be moved into the trusted folder
+# sudo mv /var/cuda-repo-${OS}-11-3-local/7fa2af80.pub /etc/apt/trusted.gpg.d/
 sudo apt-key add /var/cuda-repo-${OS}-11-3-local/7fa2af80.pub
 
 sudo apt-get -qq update

.github/workflows/cuda/cu116-Linux.sh

@@ -1,10 +1,13 @@
 #!/bin/bash
 
-OS=ubuntu1804
+# Strip the periods from the version number
+OS_VERSION=$(echo $(lsb_release -sr) | tr -d .)
+OS=ubuntu${OS_VERSION}
 
 wget -nv https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-${OS}.pin
 sudo mv cuda-${OS}.pin /etc/apt/preferences.d/cuda-repository-pin-600
 wget -nv https://developer.download.nvidia.com/compute/cuda/11.6.2/local_installers/cuda-repo-${OS}-11-6-local_11.6.2-510.47.03-1_amd64.deb
+
 sudo dpkg -i cuda-repo-${OS}-11-6-local_11.6.2-510.47.03-1_amd64.deb
 sudo apt-key add /var/cuda-repo-${OS}-11-6-local/7fa2af80.pub
 

.github/workflows/cuda/cu117-Linux-env.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+CUDA_HOME=/usr/local/cuda-11.7
+LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
+PATH=${CUDA_HOME}/bin:${PATH}
+
+export FORCE_CUDA=1
+export TORCH_CUDA_ARCH_LIST="3.5;5.0+PTX;6.0;7.0;7.5;8.0;8.6"
+export CUDA_HOME=/usr/local/cuda-11.7

.github/workflows/cuda/cu117-Linux.sh

@@ -0,0 +1,18 @@
+#!/bin/bash
+
+# Strip the periods from the version number
+OS_VERSION=$(echo $(lsb_release -sr) | tr -d .)
+OS=ubuntu${OS_VERSION}
+
+wget -nv https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-${OS}.pin
+sudo mv cuda-${OS}.pin /etc/apt/preferences.d/cuda-repository-pin-600
+wget -nv https://developer.download.nvidia.com/compute/cuda/11.7.0/local_installers/cuda-repo-${OS}-11-7-local_11.7.0-515.43.04-1_amd64.deb
+
+sudo dpkg -i cuda-repo-${OS}-11-7-local_11.7.0-515.43.04-1_amd64.deb
+sudo cp /var/cuda-repo-${OS}-11-7-local/cuda-*-keyring.gpg /usr/share/keyrings/
+
+sudo apt-get -qq update
+sudo apt install cuda cuda-nvcc-11-7 cuda-libraries-dev-11-7
+sudo apt clean
+
+rm -f https://developer.download.nvidia.com/compute/cuda/11.7.0/local_installers/cuda-repo-${OS}-11-7-local_11.7.0-515.43.04-1_amd64.deb

.github/workflows/cuda/cu120-Linux-env.sh

@@ -0,0 +1,9 @@
+#!/bin/bash
+
+CUDA_HOME=/usr/local/cuda-12.0
+LD_LIBRARY_PATH=${CUDA_HOME}/lib64:${LD_LIBRARY_PATH}
+PATH=${CUDA_HOME}/bin:${PATH}
+
+export FORCE_CUDA=1
+export TORCH_CUDA_ARCH_LIST="3.5;5.0+PTX;6.0;7.0;7.5;8.0;8.6"
+export CUDA_HOME=/usr/local/cuda-12.0

.github/workflows/cuda/cu120-Linux.sh

@@ -0,0 +1,18 @@
+#!/bin/bash
+
+# Strip the periods from the version number
+OS_VERSION=$(echo $(lsb_release -sr) | tr -d .)
+OS=ubuntu${OS_VERSION}
+
+wget -nv https://developer.download.nvidia.com/compute/cuda/repos/${OS}/x86_64/cuda-${OS}.pin
+sudo mv cuda-${OS}.pin /etc/apt/preferences.d/cuda-repository-pin-600
+wget -nv https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda-repo-${OS}-12-0-local_12.0.0-525.60.13-1_amd64.deb
+
+sudo dpkg -i cuda-repo-${OS}-12-0-local_12.0.0-525.60.13-1_amd64.deb
+sudo cp /var/cuda-repo-${OS}-12-0-local/cuda-*-keyring.gpg /usr/share/keyrings/
+
+sudo apt-get -qq update
+sudo apt install cuda cuda-nvcc-12-0 cuda-libraries-dev-12-0
+sudo apt clean
+
+rm -f https://developer.download.nvidia.com/compute/cuda/12.0.0/local_installers/cuda-repo-${OS}-12-0-local_12.0.0-525.60.13-1_amd64.deb

.github/workflows/publish.yml

@@ -1,49 +1,83 @@
-# This workflow will upload a Python Package to Release asset
+# This workflow will:
+# - Create a new Github release
+# - Build wheels for supported architectures
+# - Deploy the wheels to the Github release
+# - Release the static code to PyPi
 # For more information see: https://help.github.com/en/actions/language-and-framework-guides/using-python-with-github-actions#publishing-to-package-registries
 
+name: Build wheels and deploy
 
-name: Python Package
-
+#on:
+#  create:
+#    tags:
+#      - '**'
 on:
-  create:
-    tags:
-      - '**'
+  push
 
 jobs:
-  release:
-    name: Create Release
-    runs-on: ubuntu-latest
-    steps:
-      - name: Get the tag version
-        id: extract_branch
-        run: echo ::set-output name=branch::${GITHUB_REF#refs/tags/}
-        shell: bash
+  # setup_release:
+  #   name: Create Release
+  #   runs-on: ubuntu-latest
+  #   steps:
+  #     - name: Get the tag version
+  #       id: extract_branch
+  #       run: echo ::set-output name=branch::${GITHUB_REF#refs/tags/}
+  #       shell: bash
 
-      - name: Create Release
-        id: create_release
-        uses: actions/create-release@v1
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        with:
-          tag_name: ${{ steps.extract_branch.outputs.branch }}
-          release_name: ${{ steps.extract_branch.outputs.branch }}
-      
-  wheel:
+  #     - name: Create Release
+  #       id: create_release
+  #       uses: actions/create-release@v1
+  #       env:
+  #         GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+  #       with:
+  #         tag_name: ${{ steps.extract_branch.outputs.branch }}
+  #         release_name: ${{ steps.extract_branch.outputs.branch }}
+
+  build_wheels:
     name: Build Wheel
     runs-on: ${{ matrix.os }}
-    needs: release
-    
+    #needs: setup_release
+
     strategy:
       fail-fast: false
       matrix:
-          # os: [ubuntu-20.04]
-          os: [ubuntu-18.04]
-          python-version: ['3.7', '3.8', '3.9', '3.10']
-          torch-version: [1.11.0, 1.12.0, 1.12.1]
-          cuda-version: ['113', '116']
+          os: [ubuntu-20.04, ubuntu-22.04]
+          #python-version: ['3.7', '3.8', '3.9', '3.10']
+          #torch-version: ['1.11.0', '1.12.0', '1.13.0', '2.0.1']
+          #cuda-version: ['113', '116', '117', '120']
+          python-version: ['3.10']
+          torch-version: ['2.0.1']
+          cuda-version: ['120']
           exclude:
-            - torch-version: 1.11.0
+            # Nvidia only supports 11.7+ for ubuntu-22.04
+            - os: ubuntu-22.04
               cuda-version: '116'
+            - os: ubuntu-22.04
+              cuda-version: '113'
+            # Torch only builds cuda 117 for 1.13.0+
+            - cuda-version: '117'
+              torch-version: '1.11.0'
+            - cuda-version: '117'
+              torch-version: '1.12.0'
+            # Torch only builds cuda 116 for 1.12.0+
+            - cuda-version: '116'
+              torch-version: '1.11.0'
+            # Torch only builds cuda 120 for 2.0.1+
+            - cuda-version: '120'
+              torch-version: '1.11.0'
+            - cuda-version: '120'
+              torch-version: '1.12.0'
+            - cuda-version: '120'
+              torch-version: '1.13.0'
+            # 1.13.0 drops support for cuda 11.3
+            - cuda-version: '113'
+              torch-version: '1.13.0'
+            - cuda-version: '113'
+              torch-version: '2.0.1'
+            # Fails with "Validation Error" on artifact upload
+            - cuda-version: '117'
+              torch-version: '1.13.0'
+              os: ubuntu-20.04
 
     steps:
       - name: Checkout
@@ -82,13 +116,24 @@ jobs:
       - name: Install PyTorch ${{ matrix.torch-version }}+cu${{ matrix.cuda-version }}
         run: |
           pip install numpy pyyaml scipy ipython mkl mkl-include ninja cython typing pandas typing-extensions dataclasses && conda clean -ya
-          pip install --no-index --no-cache-dir torch==${{ matrix.torch-version }} -f https://download.pytorch.org/whl/cu${{ matrix.cuda-version }}/torch_stable.html
+          pip install --no-cache-dir torch==${{ matrix.torch-version }}
           python --version
           python -c "import torch; print('PyTorch:', torch.__version__)"
           python -c "import torch; print('CUDA:', torch.version.cuda)"
           python -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"
         shell:
           bash
+
+      # - name: Install PyTorch ${{ matrix.torch-version }}+cu${{ matrix.cuda-version }}
+      #   run: |
+      #     pip install numpy pyyaml scipy ipython mkl mkl-include ninja cython typing pandas typing-extensions dataclasses && conda clean -ya
+      #     pip install --no-index --no-cache-dir torch==${{ matrix.torch-version }} -f https://download.pytorch.org/whl/cu${{ matrix.cuda-version }}/torch_stable.html
+      #     python --version
+      #     python -c "import torch; print('PyTorch:', torch.__version__)"
+      #     python -c "import torch; print('CUDA:', torch.version.cuda)"
+      #     python -c "from torch.utils import cpp_extension; print (cpp_extension.CUDA_HOME)"
+      #   shell:
+      #     bash
 
       - name: Get the tag version
         id: extract_branch
@@ -104,24 +149,60 @@ jobs:
 
       - name: Build wheel
         run: |
+          export FLASH_ATTENTION_FORCE_BUILD="TRUE"
           export FORCE_CUDA="1"
           export PATH=/usr/local/nvidia/bin:/usr/local/nvidia/lib64:$PATH
           export LD_LIBRARY_PATH=/usr/local/nvidia/lib64:/usr/local/cuda/lib64:$LD_LIBRARY_PATH
           export CUDA_INSTALL_DIR=/usr/local/cuda-11.3$CUDA_INSTALL_DIR
-          pip install wheel
+          pip install ninja packaging setuptools wheel
           python setup.py bdist_wheel --dist-dir=dist
           tmpname=cu${{ matrix.cuda-version }}torch${{ matrix.torch-version }}
           wheel_name=$(ls dist/*whl | xargs -n 1 basename | sed "s/-/+$tmpname-/2")
-          ls dist/*whl |xargs -I {} mv {} ${wheel_name}
+          ls dist/*whl |xargs -I {} mv {} dist/${wheel_name}
           echo "wheel_name=${wheel_name}" >> $GITHUB_ENV
-      
-      - name: Upload Release Asset
-        id: upload_release_asset 
-        uses: actions/upload-release-asset@v1
-        env:
-          GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
-        with:
-          upload_url: ${{ steps.get_current_release.outputs.upload_url }}
-          asset_path: ./${{env.wheel_name}}
-          asset_name: ${{env.wheel_name}}
-          asset_content_type: application/*
+
+      - name: Log Built Wheels
+        run: |
+          ls dist
+
+      # - name: Upload Release Asset
+      #   id: upload_release_asset 
+      #   uses: actions/upload-release-asset@v1
+      #   env:
+      #     GITHUB_TOKEN: ${{ secrets.GITHUB_TOKEN }}
+      #   with:
+      #     upload_url: ${{ steps.get_current_release.outputs.upload_url }}
+      #     asset_path: ./dist/${{env.wheel_name}}
+      #     asset_name: ${{env.wheel_name}}
+      #     asset_content_type: application/*
+
+  # publish_package:
+  #   name: Publish package
+  #   needs: [build_wheels]
+
+  #   runs-on: ubuntu-latest
+
+  #   steps:
+  #     - uses: actions/checkout@v3
+
+  #     - uses: actions/setup-python@v4
+  #       with:
+  #         python-version: '3.10'
+
+  #     - name: Install dependencies
+  #       run: |
+  #         pip install ninja packaging setuptools wheel twine
+  #         pip install torch
+
+  #     - name: Build core package
+  #       env:
+  #         FLASH_ATTENTION_SKIP_CUDA_BUILD: "TRUE"
+  #       run: |
+  #         python setup.py sdist --dist-dir=dist
+
+  #     - name: Deploy
+  #       env:
+  #         TWINE_USERNAME: ${{ secrets.PYPI_USERNAME }}
+  #         TWINE_PASSWORD: ${{ secrets.PYPI_TOKEN }}
+  #       run: |
+  #         python -m twine upload dist/*