Python: Uniform Naming for units->unit Arguments #3615
Workflow file for this run
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: 🐧 CUDA | |
on: [push, pull_request] | |
concurrency: | |
group: ${{ github.ref }}-${{ github.head_ref }}-cuda | |
cancel-in-progress: true | |
jobs: | |
# Ref.: | |
# https://gitlab.com/nvidia/container-images/cuda/-/blob/master/dist/ubuntu18.04/10.1/base/Dockerfile | |
# https://github.com/ComputationalRadiationPhysics/picongpu/blob/0.5.0/share/picongpu/dockerfiles/ubuntu-1604/Dockerfile | |
# https://developer.download.nvidia.com/compute/cuda/repos/ubuntu1804/x86_64/ | |
build_nvcc: | |
name: NVCC 11.3.1 SP | |
runs-on: ubuntu-20.04 | |
if: github.event.pull_request.draft == false | |
env: | |
CXXFLAGS: "-Werror" | |
CMAKE_GENERATOR: Ninja | |
steps: | |
- uses: actions/checkout@v4 | |
- name: install dependencies | |
run: | | |
.github/workflows/dependencies/nvcc11-openmpi.sh | |
- name: CCache Cache | |
uses: actions/cache@v4 | |
# - once stored under a key, they become immutable (even if local cache path content changes) | |
# - for a refresh the key has to change, e.g., hash of a tracked file in the key | |
with: | |
path: | | |
~/.ccache | |
~/.cache/ccache | |
key: ccache-cuda-nvcc-${{ hashFiles('.github/workflows/cuda.yml') }}-${{ hashFiles('cmake/dependencies/ABLASTR.cmake') }} | |
restore-keys: | | |
ccache-cuda-nvcc-${{ hashFiles('.github/workflows/cuda.yml') }}- | |
ccache-cuda-nvcc- | |
- name: build ImpactX | |
run: | | |
export PATH=/usr/local/nvidia/bin:/usr/local/cuda/bin:${PATH} | |
export LD_LIBRARY_PATH=/usr/local/nvidia/lib:/usr/local/nvidia/lib64:/usr/local/cuda/lib64:${LD_LIBRARY_PATH} | |
which nvcc || echo "nvcc not in PATH!" | |
cmake -S . -B build \ | |
-DCMAKE_VERBOSE_MAKEFILE=ON \ | |
-DImpactX_COMPUTE=CUDA \ | |
-DImpactX_PYTHON=ON \ | |
-DAMReX_CUDA_ARCH=6.0 \ | |
-DImpactX_PRECISION=SINGLE \ | |
-DAMReX_CUDA_ERROR_CROSS_EXECUTION_SPACE_CALL=ON \ | |
-DAMReX_CUDA_ERROR_CAPTURE_THIS=ON | |
cmake --build build -j 4 |