NEO-2

Installation

Prerequisites

git, cmake, make, ninja, gcc/g++/gfortran, MPI (OpenMPI or MPICH), BLAS/LAPACK (OpenBLAS or MKL), SuiteSparse, FFTW, GSL, HDF5/NetCDF with Fortran libraries. If SuiteSparse is not available on your system, it will be built automatically by NEO-2. For code coverage, lcov is also required. On Debian or Ubuntu run

sudo apt install git cmake make ninja-build gcc g++ gfortran
sudo apt install openmpi-bin openmpi-common libopenmpi-dev
sudo apt install libopenblas-dev libsuitesparse-dev
sudo apt install libfftw3-dev libgsl-dev libhdf5-dev libnetcdf-dev libnetcdff-dev
sudo apt install lcov  # Optional, for code coverage

Build

Run

make

inside the NEO-2 directory. This will handle the whole build process, including cmake calls.

You obtain a build directory with subdirectories

• NEO-2-QL with binary neo_2_ql.x for the tokamak version.
• NEO-2-PAR with binary neo_2_par.x for the stellarator version.

Testing

To run tests:

make test

To generate code coverage report:

make coverage

This will build with coverage instrumentation, run all tests, and display a coverage summary. Coverage data files are generated in the build directory.

To view detailed coverage report:

# After running make coverage, view the summary
lcov --summary build/coverage_filtered.info

# Generate HTML report for detailed viewing
cd build && genhtml coverage_filtered.info --output-directory coverage_html
# Open build/coverage_html/index.html in a web browser

If you have additional compilers and MPI implementations (e.g. Intel) installed, be sure to set the environment like

export CC=gcc
export CXX=g++
export FC=gfortran
export MPI_HOME=/usr

before the build. You may also need to deactivate Python environments such as Anaconda.

Run

NEO-2 requires Boozer files for MHD equilibria in text format (.bc ending). These can be generated via executables / functions in https://github.com/itpplasma/libneo/

• efit_to_boozer for tokamaks
• vmec_to_boozer for stellarators: https://github.com/itpplasma/libneo/blob/main/python/libneo/boozer.py