[CI] refactor workflows (#2504)

no longer run third-party tests on every PR

.github/workflows/integration-tests.yml

@@ -34,7 +34,7 @@ jobs:
           fi
 
 
-  Integration-Tests-Nvidia:
+  Integration-Tests:
     needs: Runner-Preparation
 
     runs-on: ${{ matrix.runner }}
@@ -171,140 +171,8 @@ jobs:
           python3 -m pytest -vs . --reruns 10
           sudo nvidia-smi -i 0 -rgc
 
-  Integration-Tests-Shared-Middle-Layer:
-
-    runs-on: ubuntu-latest
-
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v2
-
-      - name: Clear cache
-        run: |
-          rm -rf ~/.triton
-
-      - name: Update PATH
-        run: |
-          echo "PATH=${HOME}/.local/bin:${PATH}" >> "${GITHUB_ENV}"
-
-      - name: Check pre-commit
-        run: |
-          python3 -m pip install --upgrade pre-commit
-          python3 -m pre_commit run --all-files --verbose
-
-      - name: Install Triton
-        run: |
-          export TRITON_CODEGEN_TRITON_SHARED=1
-          git submodule update --init --recursive
-          cd python
-          python3 -m pip install --upgrade pip
-          python3 -m pip install cmake==3.24
-          python3 -m pip install ninja
-          python3 -m pip uninstall -y triton
-          python3 setup.py build
-          python3 -m pip install --no-build-isolation -vvv '.[tests]'
-
-      - name: Run shared middle-layer lit tests
-        run: |
-          python3 -m pip install lit
-          cd python
-          LIT_TEST_DIR="build/$(ls build | grep -i cmake)/third_party/triton_shared/test"
-          if [ ! -d "${LIT_TEST_DIR}" ]; then
-            echo "Coult not find '${LIT_TEST_DIR}'" ; exit -1
-          fi
-          lit -v "${LIT_TEST_DIR}"
-
-
-  Integration-Tests-Third-Party:
-    needs: Runner-Preparation
-    if: false
-
-    runs-on: ${{ matrix.runner }}
-
-    strategy:
-      matrix:
-        runner: ${{fromJson(needs.Runner-Preparation.outputs.matrix-optional)}}
-
-    steps:
-      - name: Checkout
-        uses: actions/checkout@v2
-
-      - name: Set ROCM ENV
-        if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'gfx908')}}
-        run: |
-          echo "BACKEND=ROCM" >> "${GITHUB_ENV}"
-
-      - name: Set XPU ENV
-        if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'arc770')}}
-        run: |
-          echo "BACKEND=XPU" >> "${GITHUB_ENV}"
-
-      - name: Clear cache
-        run: |
-          rm -rf ~/.triton
-
-      - name: Update PATH
-        run: |
-          echo "PATH=${HOME}/.local/bin:${PATH}" >> "${GITHUB_ENV}"
-
-      - name: Check pre-commit
-        if: ${{ matrix.runner != 'macos-10.15' && (matrix.runner[1] != 'arc770') }}
-        run: |
-          python3 -m pip install --upgrade pre-commit
-          python3 -m pre_commit run --all-files --verbose
-
-      - name: Check pre-commit arc770
-        if: ${{ matrix.runner != 'macos-10.15' && (matrix.runner[1] == 'arc770') }}
-        run: |
-          source ${HOME}/triton_vars.sh
-          source ${HOME}/miniconda3/bin/activate
-          conda activate triton-xpu-ci
-          python3 -m pip install --upgrade pre-commit
-          python3 -m pre_commit run --all-files
-
-      - name: Install Triton on ROCM
-        if: ${{ env.BACKEND == 'ROCM'}}
-        run: |
-          git submodule update --init --recursive
-          cd python
-          python3 -m pip install --upgrade pip
-          python3 -m pip install cmake==3.24
-          python3 -m pip install torch==1.13.1 --index-url https://download.pytorch.org/whl/rocm5.2
-          export TRITON_CODEGEN_AMD_HIP_BACKEND=1
-          python3 -m pip install --no-build-isolation -vvv '.[tests]'
-
-      - name: Install Triton on XPU
-        if: ${{ env.BACKEND == 'XPU'}}
-        run: |
-          source ${HOME}/triton_vars.sh
-          source ${HOME}/miniconda3/bin/activate
-          conda activate triton-xpu-ci
-          git submodule update --init --recursive
-          cd python
-          python3 -m pip install --upgrade pip
-          python3 -m pip install cmake==3.24
-          export TRITON_CODEGEN_INTEL_XPU_BACKEND=1
-          python3 -m pip uninstall -y triton
-          python3 setup.py build
-          python3 -m pip install --no-build-isolation -vvv '.[tests]'
-
-      - name: Run python tests on ROCM
-        if: ${{ env.BACKEND == 'ROCM'}}
-        run: |
-          cd python/test/unit/language
-          python3 -m pytest --capture=tee-sys -rfs --verbose "test_core.py"
-
-      - name: Run python tests on XPU
-        if: ${{ env.BACKEND == 'XPU'}}
-        run: |
-          source ${HOME}/triton_vars.sh
-          source ${HOME}/miniconda3/bin/activate
-          conda activate triton-xpu-ci
-          cd python/test/backend/third_party_backends
-          python3 -m pytest --capture=tee-sys -rfs --verbose --backend xpu
-
   Compare-artifacts:
-    needs: Integration-Tests-Nvidia
+    needs: Integration-Tests
 
     runs-on: ubuntu-latest
 
@@ -333,7 +201,7 @@ jobs:
       - name: Download latest main artifacts
         env:
           ARTIFACT_NAME: artifacts A100
-          ARTIFACT_JOB_NAME: Integration-Tests-Nvidia
+          ARTIFACT_JOB_NAME: Integration-Tests
           MAX_NUM_ACTIONS_PAGES: 30
           GH_TOKEN: ${{ secrets.GITHUB_TOKEN }}
         run: |

.github/workflows/llvm-build/Dockerfile

@@ -1,16 +1,17 @@
 FROM centos:7
 ARG llvm_dir=llvm-project
-
 # Add the cache artifacts and the LLVM source tree to the container
 ADD sccache /sccache
 ADD "${llvm_dir}" /source/llvm-project
 ENV SCCACHE_DIR="/sccache"
 ENV SCCACHE_CACHE_SIZE="2G"
 
+RUN echo -e "[llvmtoolset-build]\nname=LLVM Toolset 13.0 - Build\nbaseurl=https://buildlogs.centos.org/c7-llvm-toolset-13.0.x86_64/\ngpgcheck=0\nenabled=1" > /etc/yum.repos.d/llvmtoolset-build.repo
 # Install build dependencies
 RUN yum install --assumeyes centos-release-scl
-RUN yum install --assumeyes devtoolset-9-gcc* python3-devel python3-pip
-SHELL [ "/usr/bin/scl", "enable", "devtoolset-9" ]
+RUN yum install --assumeyes --nogpgcheck llvm-toolset-13.0
+RUN yum install --assumeyes rh-python38-python-devel rh-python38-python-pip
+SHELL [ "/usr/bin/scl", "enable", "llvm-toolset-13.0", "rh-python38" ]
 
 RUN python3 -m pip install --upgrade pip
 RUN python3 -m pip install --upgrade cmake ninja sccache
@@ -21,17 +22,22 @@ RUN python3 -m pip install -r /source/llvm-project/mlir/python/requirements.txt
 # Configure, Build, Test, and Install LLVM
 RUN cmake -GNinja -Bbuild \
   -DCMAKE_BUILD_TYPE=Release \
-  -DCMAKE_C_COMPILER=gcc \
-  -DCMAKE_CXX_COMPILER=g++ \
+  -DCMAKE_C_COMPILER=clang \
+  -DCMAKE_CXX_COMPILER=clang++ \
+  -DCMAKE_ASM_COMPILER=clang \
   -DCMAKE_C_COMPILER_LAUNCHER=sccache \
   -DCMAKE_CXX_COMPILER_LAUNCHER=sccache \
+  -DCMAKE_CXX_FLAGS="-Wno-everything" \
+  -DCMAKE_LINKER=lld \
   -DCMAKE_INSTALL_PREFIX="/install" \
   -DLLVM_BUILD_UTILS=ON \
+  -DLLVM_BUILD_TOOLS=ON \
   -DLLVM_ENABLE_ASSERTIONS=ON \
   -DMLIR_ENABLE_BINDINGS_PYTHON=ON \
   -DLLVM_ENABLE_PROJECTS=mlir \
+  -DLLVM_ENABLE_TERMINFO=OFF \
   -DLLVM_INSTALL_UTILS=ON \
   -DLLVM_TARGETS_TO_BUILD="host;NVPTX;AMDGPU" \
   /source/llvm-project/llvm
 
-RUN ninja -C build check-mlir install
+RUN ninja -C build install

.github/workflows/llvm-build/llvm-build.yml

@@ -18,43 +18,17 @@ permissions:
 jobs:
 
   build:
+    name: Build on ${{ matrix.config.runner }}
+    runs-on: ${{ matrix.config.runs_on }}
 
     strategy:
       fail-fast: true
-
       matrix:
-        platform: [
-          ubuntu-20.04-x64,
-          ubuntu-22.04-x64,
-          centos-7-x64,
-          macos-x64,
-          macos-arm64
-        ]
-
-        include:
-          # Specify OS versions
-          - platform: ubuntu-20.04-x64
-            host-os: ubuntu-20.04
-            target-os: ubuntu
-            arch: x64
-          - platform: ubuntu-22.04-x64
-            host-os: ubuntu-22.04
-            target-os: ubuntu
-            arch: x64
-          - platform: centos-7-x64
-            host-os: ubuntu-22.04
-            target-os: centos
-            arch: x64
-          - platform: macos-x64
-            host-os: macos-12
-            target-os: macos
-            arch: x64
-          - platform: macos-arm64
-            host-os: macos-12
-            target-os: macos
-            arch: arm64
-
-    runs-on: ${{ matrix.host-os }}
+        config:
+        - {runner: 'Ubuntu 20.04', runs_on: 'ubuntu-20.04', target-os: 'ubuntu', arch: 'x64'}
+        - {runner: 'CentOS 7', runs_on: ['self-hosted', 'CPU'], target-os: 'centos', arch: 'x64'}
+        - {runner: 'MacOS X64', runs_on: 'macos-12', target-os: 'macos', arch: 'x64'}
+        - {runner: 'MacOS ARM64', runs_on: 'macos-12', target-os: 'macos', arch: 'arm64'}
 
     steps:
 
@@ -73,7 +47,7 @@ jobs:
         echo "Short LLVM commit hash: ${SHORT_LLVM_COMMIT_HASH}"
         echo "short_llvm_commit_hash=${SHORT_LLVM_COMMIT_HASH}" >> ${GITHUB_ENV}
 
-        INSTALL_DIR="llvm-${SHORT_LLVM_COMMIT_HASH}-${{ matrix.platform }}"
+        INSTALL_DIR="llvm-${SHORT_LLVM_COMMIT_HASH}-${{ matrix.config.target-os }}-${{ matrix.config.arch }}"
         echo "LLVM installation directory name: ${INSTALL_DIR}"
         echo "llvm_install_dir=${INSTALL_DIR}" >> ${GITHUB_ENV}
 
@@ -99,11 +73,11 @@ jobs:
       uses: actions/cache@v3
       with:
         path: ${{ env.SCCACHE_DIR }}
-        key: ${{ matrix.platform }}-${{ env.short_llvm_commit_hash }}
-        restore-keys: ${{ matrix.platform }}-
+        key: ${{ matrix.config.target-os }}-${{ matrix.config.arch }}-${{ env.short_llvm_commit_hash }}
+        restore-keys: ${{ matrix.config.target-os }}-${{ matrix.config.arch }}-
 
     - name: Configure, Build, Test, and Install LLVM (Ubuntu and macOS x64)
-      if: matrix.arch == 'x64' && contains(fromJSON('["ubuntu", "macos"]'), matrix.target-os)
+      if: matrix.config.arch == 'x64' && (matrix.config.target-os == 'ubuntu' || matrix.config.target-os == 'macos')
       run: >
         python3 -m pip install -r llvm-project/mlir/python/requirements.txt
 
@@ -114,11 +88,13 @@ jobs:
         -DCMAKE_INSTALL_PREFIX="${{ env.llvm_install_dir }}"
         -DCMAKE_LINKER=lld
         -DLLVM_BUILD_UTILS=ON
+        -DLLVM_BUILD_TOOLS=ON
         -DLLVM_ENABLE_ASSERTIONS=ON
         -DMLIR_ENABLE_BINDINGS_PYTHON=ON
         -DLLVM_ENABLE_PROJECTS=mlir
         -DLLVM_INSTALL_UTILS=ON
         -DLLVM_TARGETS_TO_BUILD="host;NVPTX;AMDGPU"
+        -DLLVM_ENABLE_TERMINFO=OFF
         llvm-project/llvm
 
         ninja -C llvm-project/build check-mlir install
@@ -126,7 +102,7 @@ jobs:
         tar czf "${{ env.llvm_install_dir }}.tar.gz" "${{ env.llvm_install_dir }}"
 
     - name: Configure, Build, and Install LLVM (macOS arm64)
-      if: matrix.arch == 'arm64' && matrix.target-os == 'macos'
+      if: matrix.config.arch == 'arm64' && matrix.config.target-os == 'macos'
       run: >
         python3 -m pip install -r llvm-project/mlir/python/requirements.txt
 
@@ -138,6 +114,7 @@ jobs:
         -DCMAKE_LINKER=lld
         -DCMAKE_OSX_ARCHITECTURES=arm64
         -DLLVM_BUILD_UTILS=ON
+        -DLLVM_BUILD_TOOLS=ON
         -DLLVM_ENABLE_ASSERTIONS=ON
         -DMLIR_ENABLE_BINDINGS_PYTHON=ON
         -DLLVM_ENABLE_PROJECTS=mlir
@@ -145,6 +122,7 @@ jobs:
         -DLLVM_INSTALL_UTILS=ON
         -DLLVM_TARGETS_TO_BUILD="AArch64"
         -DLLVM_USE_HOST_TOOLS=ON
+        -DLLVM_ENABLE_TERMINFO=OFF
         llvm-project/llvm
 
         ninja -C llvm-project/build install
@@ -152,7 +130,7 @@ jobs:
         tar czf "${{ env.llvm_install_dir }}.tar.gz" "${{ env.llvm_install_dir }}"
 
     - name: Configure, Build, Test, and Install LLVM (CentOS)
-      if: matrix.target-os == 'centos'
+      if: matrix.config.target-os == 'centos'
       run: |
         docker build --tag llvm-build --build-arg llvm_dir=llvm-project \
           -f llvm-build/.github/workflows/Dockerfile .

.github/workflows/third-party/integration-tests.yml

@@ -0,0 +1,167 @@
+name: Integration Tests
+
+on:
+  workflow_dispatch:
+  pull_request:
+    branches: [main]
+  merge_group:
+    branches: [main]
+    types: [checks_requested]
+
+concurrency:
+  group: ${{ github.ref }}
+  cancel-in-progress: ${{ github.ref != 'refs/heads/master' }}
+
+env:
+  TRITON_USE_ASSERT_ENABLED_LLVM: "TRUE"
+
+jobs:
+  Runner-Preparation:
+    runs-on: ubuntu-latest
+    outputs:
+      matrix-required: ${{ steps.set-matrix.outputs.matrix-required }}
+      matrix-optional: ${{ steps.set-matrix.outputs.matrix-optional }}
+    steps:
+      - name: Prepare runner matrix
+        id: set-matrix
+        run: |
+          if [ x"${{ github.repository }}" == x"openai/triton" ]; then
+            echo '::set-output name=matrix-required::[["self-hosted", "A100"], ["self-hosted", "H100"]]'
+            echo '::set-output name=matrix-optional::[["self-hosted", "gfx908"], ["self-hosted", "arc770"]]'
+          else
+            echo '::set-output name=matrix-required::["ubuntu-latest"]'
+            echo '::set-output name=matrix-optional::["ubuntu-latest"]'
+          fi
+
+
+  Integration-Tests-Shared-Middle-Layer:
+
+    runs-on: ubuntu-latest
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+
+      - name: Clear cache
+        run: |
+          rm -rf ~/.triton
+
+      - name: Update PATH
+        run: |
+          echo "PATH=${HOME}/.local/bin:${PATH}" >> "${GITHUB_ENV}"
+
+      - name: Check pre-commit
+        run: |
+          python3 -m pip install --upgrade pre-commit
+          python3 -m pre_commit run --all-files --verbose
+
+      - name: Install Triton
+        run: |
+          export TRITON_CODEGEN_TRITON_SHARED=1
+          git submodule update --init --recursive
+          cd python
+          python3 -m pip install --upgrade pip
+          python3 -m pip install cmake==3.24
+          python3 -m pip install ninja
+          python3 -m pip uninstall -y triton
+          python3 setup.py build
+          python3 -m pip install --no-build-isolation -vvv '.[tests]'
+
+      - name: Run shared middle-layer lit tests
+        run: |
+          python3 -m pip install lit
+          cd python
+          LIT_TEST_DIR="build/$(ls build | grep -i cmake)/third_party/triton_shared/test"
+          if [ ! -d "${LIT_TEST_DIR}" ]; then
+            echo "Coult not find '${LIT_TEST_DIR}'" ; exit -1
+          fi
+          lit -v "${LIT_TEST_DIR}"
+
+
+  Integration-Tests-Third-Party:
+    needs: Runner-Preparation
+    if: false
+
+    runs-on: ${{ matrix.runner }}
+
+    strategy:
+      matrix:
+        runner: ${{fromJson(needs.Runner-Preparation.outputs.matrix-optional)}}
+
+    steps:
+      - name: Checkout
+        uses: actions/checkout@v2
+
+      - name: Set ROCM ENV
+        if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'gfx908')}}
+        run: |
+          echo "BACKEND=ROCM" >> "${GITHUB_ENV}"
+
+      - name: Set XPU ENV
+        if: ${{(matrix.runner[0] == 'self-hosted') && (matrix.runner[1] == 'arc770')}}
+        run: |
+          echo "BACKEND=XPU" >> "${GITHUB_ENV}"
+
+      - name: Clear cache
+        run: |
+          rm -rf ~/.triton
+
+      - name: Update PATH
+        run: |
+          echo "PATH=${HOME}/.local/bin:${PATH}" >> "${GITHUB_ENV}"
+
+      - name: Check pre-commit
+        if: ${{ matrix.runner != 'macos-10.15' && (matrix.runner[1] != 'arc770') }}
+        run: |
+          python3 -m pip install --upgrade pre-commit
+          python3 -m pre_commit run --all-files --verbose
+
+      - name: Check pre-commit arc770
+        if: ${{ matrix.runner != 'macos-10.15' && (matrix.runner[1] == 'arc770') }}
+        run: |
+          source ${HOME}/triton_vars.sh
+          source ${HOME}/miniconda3/bin/activate
+          conda activate triton-xpu-ci
+          python3 -m pip install --upgrade pre-commit
+          python3 -m pre_commit run --all-files
+
+      - name: Install Triton on ROCM
+        if: ${{ env.BACKEND == 'ROCM'}}
+        run: |
+          git submodule update --init --recursive
+          cd python
+          python3 -m pip install --upgrade pip
+          python3 -m pip install cmake==3.24
+          python3 -m pip install torch==1.13.1 --index-url https://download.pytorch.org/whl/rocm5.2
+          export TRITON_CODEGEN_AMD_HIP_BACKEND=1
+          python3 -m pip install --no-build-isolation -vvv '.[tests]'
+
+      - name: Install Triton on XPU
+        if: ${{ env.BACKEND == 'XPU'}}
+        run: |
+          source ${HOME}/triton_vars.sh
+          source ${HOME}/miniconda3/bin/activate
+          conda activate triton-xpu-ci
+          git submodule update --init --recursive
+          cd python
+          python3 -m pip install --upgrade pip
+          python3 -m pip install cmake==3.24
+          export TRITON_CODEGEN_INTEL_XPU_BACKEND=1
+          python3 -m pip uninstall -y triton
+          python3 setup.py build
+          python3 -m pip install --no-build-isolation -vvv '.[tests]'
+
+      - name: Run python tests on ROCM
+        if: ${{ env.BACKEND == 'ROCM'}}
+        run: |
+          cd python/test/unit/language
+          python3 -m pytest --capture=tee-sys -rfs --verbose "test_core.py"
+
+      - name: Run python tests on XPU
+        if: ${{ env.BACKEND == 'XPU'}}
+        run: |
+          source ${HOME}/triton_vars.sh
+          source ${HOME}/miniconda3/bin/activate
+          conda activate triton-xpu-ci
+          cd python/test/backend/third_party_backends
+          python3 -m pytest --capture=tee-sys -rfs --verbose --backend xpu