Skip to content

Fix incorrect outputs and improve performance of commonMemSetLargePattern #10659

Fix incorrect outputs and improve performance of commonMemSetLargePattern

Fix incorrect outputs and improve performance of commonMemSetLargePattern #10659

Workflow file for this run

name: Build and test
on: [push, pull_request]
concurrency:
group: ${{ github.workflow }}-${{ github.head_ref || github.run_id }}
cancel-in-progress: true
permissions:
contents: read
pull-requests: write
jobs:
ubuntu-build:
name: Build - Ubuntu
strategy:
matrix:
os: ['ubuntu-20.04', 'ubuntu-22.04']
build_type: [Debug, Release]
compiler: [{c: gcc, cxx: g++}]
libbacktrace: ['-DVAL_USE_LIBBACKTRACE_BACKTRACE=OFF']
pool_tracking: ['-DUMF_ENABLE_POOL_TRACKING=ON', '-DUMF_ENABLE_POOL_TRACKING=OFF']
latency_tracking: ['-DUR_ENABLE_LATENCY_HISTOGRAM=OFF']
include:
- os: 'ubuntu-22.04'
build_type: Release
compiler: {c: clang, cxx: clang++}
libbacktrace: '-DVAL_USE_LIBBACKTRACE_BACKTRACE=OFF'
- os: 'ubuntu-22.04'
build_type: Release
compiler: {c: gcc, cxx: g++}
libbacktrace: '-DVAL_USE_LIBBACKTRACE_BACKTRACE=ON'
- os: 'ubuntu-22.04'
build_type: Release
compiler: {c: clang, cxx: clang++}
libbacktrace: '-DVAL_USE_LIBBACKTRACE_BACKTRACE=ON'
- os: 'ubuntu-20.04'
build_type: Release
compiler: {c: gcc-7, cxx: g++-7}
- os: 'ubuntu-22.04'
build_type: Release
compiler: {c: clang, cxx: clang++}
latency_tracking: '-DUR_ENABLE_LATENCY_HISTOGRAM=ON'
runs-on: ${{ (matrix.os == 'ubuntu-22.04' && github.repository_owner == 'oneapi-src') && 'intel-ubuntu-22.04' || matrix.os }}
steps:
- uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
- name: Install apt packages
run: |
sudo apt-get update
sudo apt-get install -y ${{matrix.compiler.c}} devscripts
- name: Install libhwloc
run: .github/scripts/install_hwloc.sh
- name: Setup PATH
run: echo "$HOME/.local/bin" >> $GITHUB_PATH
- name: Install g++-7
if: matrix.compiler.cxx == 'g++-7'
run: |
sudo apt-get install -y ${{matrix.compiler.cxx}}
- name: Install libbacktrace
if: matrix.libbacktrace == '-DVAL_USE_LIBBACKTRACE_BACKTRACE=ON'
run: |
git clone https://github.com/ianlancetaylor/libbacktrace.git
cd libbacktrace
./configure
make
sudo make install
cd ..
- name: Download DPC++
if: matrix.os == 'ubuntu-22.04'
run: |
sudo apt install libncurses5
wget -O ${{github.workspace}}/dpcpp_compiler.tar.gz https://github.com/intel/llvm/releases/download/nightly-2024-09-27/sycl_linux.tar.gz
mkdir -p ${{github.workspace}}/dpcpp_compiler
tar -xvf ${{github.workspace}}/dpcpp_compiler.tar.gz -C ${{github.workspace}}/dpcpp_compiler
- name: Configure CMake
if: matrix.os == 'ubuntu-22.04'
# WEXTRA: https://github.com/oneapi-src/unified-runtime/issues/2109
run: >
cmake
-B${{github.workspace}}/build
-DCMAKE_C_COMPILER=${{matrix.compiler.c}}
-DCMAKE_CXX_COMPILER=${{matrix.compiler.cxx}}
-DUR_ENABLE_TRACING=ON
-DCMAKE_BUILD_TYPE=${{matrix.build_type}}
-DUR_BUILD_TESTS=ON
-DUR_FORMAT_CPP_STYLE=OFF
-DUR_DEVELOPER_MODE=ON
-DUR_DPCXX=${{github.workspace}}/dpcpp_compiler/bin/clang++
-DUR_CONFORMANCE_TEST_LOADER=OFF
${{matrix.libbacktrace}}
${{matrix.pool_tracking}}
${{matrix.latency_tracking}}
- name: Configure CMake
if: matrix.os == 'ubuntu-20.04'
# WEXTRA: https://github.com/oneapi-src/unified-runtime/issues/2109
# Note: Disable Werror, since 20.04 raises different ones than 22.04
run: >
cmake
-B${{github.workspace}}/build
-DCMAKE_C_COMPILER=${{matrix.compiler.c}}
-DCMAKE_CXX_COMPILER=${{matrix.compiler.cxx}}
-DUR_ENABLE_TRACING=ON
-DCMAKE_BUILD_TYPE=${{matrix.build_type}}
-DUR_BUILD_TESTS=ON
-DUR_FORMAT_CPP_STYLE=OFF
-DUR_DEVELOPER_MODE=OFF
${{matrix.libbacktrace}}
${{matrix.pool_tracking}}
${{matrix.latency_tracking}}
- name: Build
run: cmake --build ${{github.workspace}}/build -j $(nproc)
- name: Verify hardening flags have been set
run: cmake --build ${{github.workspace}}/build --target verify-hardening
# https://github.com/oneapi-src/unified-runtime/issues/2120
if: ${{ matrix.compiler.cxx != 'clang++' && matrix.os != 'ubuntu-20.04' }}
- name: Test
working-directory: ${{github.workspace}}/build
run: ctest -C ${{matrix.build_type}} --output-on-failure -L "umf|loader|validation|tracing|unit|urtrace"
fuzztest:
name: Fuzz tests short
uses: ./.github/workflows/build-fuzz-reusable.yml
with:
test_label: "fuzz-short"
level-zero:
name: Level Zero
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: L0
runner_name: L0
level-zero-v2:
name: Level Zero V2
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: L0_V2
runner_name: L0
level-zero-static:
name: Level Zero static
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: L0
runner_name: L0
static_loader: ON
static_adapter: ON
opencl:
name: OpenCL
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: OPENCL
runner_name: OPENCL
platform: "Intel(R) OpenCL"
cuda:
name: CUDA
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: CUDA
runner_name: CUDA
hip:
name: HIP
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: HIP
runner_name: HIP
native-cpu:
name: Native CPU
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: NATIVE_CPU
runner_name: NATIVE_CPU
# Native CPU jobs are here to force the loader to be used (UR will not use the loader if there is only one target)
combined-opencl-native-cpu:
name: OpenCL + Native CPU (Loader)
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: OPENCL
other_adapter_name: NATIVE_CPU
runner_name: OPENCL
platform: "OPENCL:Intel(R) OpenCL"
combined-level-zero-native-cpu:
name: Level Zero + Native CPU (Loader)
uses: ./.github/workflows/build-hw-reusable.yml
with:
adapter_name: L0
other_adapter_name: NATIVE_CPU
runner_name: L0
e2e-level-zero:
name: E2E L0
permissions:
contents: read
pull-requests: write
needs: [ubuntu-build, level-zero]
uses: ./.github/workflows/e2e_level_zero.yml
e2e-opencl:
name: E2E OpenCL
permissions:
contents: read
pull-requests: write
needs: [ubuntu-build, opencl]
uses: ./.github/workflows/e2e_opencl.yml
e2e-cuda:
name: E2E CUDA
permissions:
contents: read
pull-requests: write
needs: [ubuntu-build, cuda]
uses: ./.github/workflows/e2e_cuda.yml
windows-build:
name: Build - Windows
strategy:
matrix:
os: ['windows-2019', 'windows-2022']
adapter: [
{name: None, var: ''}, {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON'},
{name: None, var: ''}, {name: L0_V2, var: '-DUR_BUILD_ADAPTER_L0_V2=ON'},
{name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON -DUR_STATIC_ADAPTER_L0=ON'}
]
# TODO: building level zero loader on windows-2019 and clang-cl is currently broken
exclude:
- os: 'windows-2019'
adapter: {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON'}
- os: 'windows-2019'
adapter: {name: L0_V2, var: '-DUR_BUILD_ADAPTER_L0_V2=ON'}
- os: 'windows-2019'
adapter: {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON -DUR_STATIC_ADAPTER_L0=ON'}
- adapter: {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON'}
compiler: {c: clang-cl, cxx: clang-cl}
- adapter: {name: L0_V2, var: '-DUR_BUILD_ADAPTER_L0_V2=ON'}
compiler: {c: clang-cl, cxx: clang-cl}
- adapter: {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON -DUR_STATIC_ADAPTER_L0=ON'}
compiler: {c: clang-cl, cxx: clang-cl}
build_type: [Debug, Release]
compiler: [{c: cl, cxx: cl}, {c: clang-cl, cxx: clang-cl}]
include:
- compiler: {c: clang-cl, cxx: clang-cl}
toolset: "-T ClangCL"
- os: 'windows-2022'
adapter: {name: L0, var: '-DUR_BUILD_ADAPTER_L0=ON -DUR_STATIC_ADAPTER_L0=ON'}
build_type: 'Release'
compiler: {c: cl, cxx: cl}
runs-on: ${{matrix.os}}
steps:
- uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
- name: Install hwloc
run: vcpkg install hwloc:x64-windows
- name: Configure CMake
env:
VCPKG_PATH: "C:/vcpkg/packages/hwloc_x64-windows"
run: >
cmake
-B${{github.workspace}}/build
${{matrix.toolset}}
-DCMAKE_PREFIX_PATH="${{env.VCPKG_PATH}}"
-DCMAKE_C_COMPILER=${{matrix.compiler.c}}
-DCMAKE_CXX_COMPILER=${{matrix.compiler.cxx}}
-DCMAKE_POLICY_DEFAULT_CMP0094=NEW
-DUR_ENABLE_TRACING=ON
-DUR_DEVELOPER_MODE=ON
-DUR_BUILD_TESTS=ON
-DUR_FORMAT_CPP_STYLE=OFF
-DUR_CONFORMANCE_TEST_LOADER=OFF
${{matrix.adapter.var}}
- name: Build all
run: cmake --build ${{github.workspace}}/build --config ${{matrix.build_type}} -j $Env:NUMBER_OF_PROCESSORS
- name: Test
working-directory: ${{github.workspace}}/build
run: ctest -C ${{matrix.build_type}} --output-on-failure -L "umf|loader|validation|tracing|unit|urtrace"
macos-build:
name: Build - MacOS
strategy:
matrix:
os: ['macos-13']
runs-on: ${{matrix.os}}
steps:
- uses: actions/checkout@b4ffde65f46336ab88eb53be808477a3936bae11 # v4.1.1
- uses: actions/setup-python@0a5c61591373683505ea898e09a3ea4f39ef2b9c # v5.0.0
with:
python-version: 3.9
- name: Install prerequisites
run: python3 -m pip install -r third_party/requirements.txt
- name: Install hwloc
run: brew install hwloc
- name: Configure CMake
run: >
cmake
-B${{github.workspace}}/build
-DUR_ENABLE_TRACING=ON
-DUR_DEVELOPER_MODE=ON
-DCMAKE_BUILD_TYPE=Release
-DUR_BUILD_TESTS=ON
-DUR_FORMAT_CPP_STYLE=ON
-DUMF_ENABLE_POOL_TRACKING=ON
- name: Build
run: cmake --build ${{github.workspace}}/build -j $(sysctl -n hw.logicalcpu)