Gyselalib++ Documentation

Gyselalib++ is a collection of C++ components for writing gyrokinetic semi-lagrangian codes and similar as well as a collection of such codes. It is based on DDC. We strongly encourage new developers to begin by reading our documentation about Using DDC in Gyselalibxx.

Set-up

In order to set up Gyselalib++ on a new machine, simply run:

git clone --recurse-submodules git@gitlab.maisondelasimulation.fr:gysela-developpers/gyselalibxx.git gyselalibxx
cd gyselalibxx
./bin/install-hooks

or

git clone --recurse-submodules https://gitlab.maisondelasimulation.fr/gysela-developpers/gyselalibxx.git gyselalibxx
cd gyselalibxx
./bin/install-hooks

on a machine for which Gyselalib++ is already used an environment script may be available to set up the necessary modules etc.

Please check the folder toolchains to find the existing configurations. See the documentation about Pre-made build settings for more information on the provided files.

For example in order to set up the environment on the Adastra supercalculator simply run:

source toolchains/mi250.hipcc.adastra.spack/prepare.sh
source toolchains/mi250.hipcc.adastra.spack/environment.sh

Compilation

To compile Gyselalib++, navigate to the folder where the library was cloned to and run:

mkdir build
cd build
cmake -DCMAKE_TOOLCHAIN_FILE=<TOOLCHAIN_FILE> ..
make

The toolchains are found in the folder toolchains/. Each toolchain is designed for specific hardware with the exception of those in the folder docker.gyselalibxx_env/. These toolchains can be used with the docker environment on most hardware.

For more details about toolchains, see the documentation about Pre-made build settings.

Execution

to run the tests:

ctest --output-on-failure

Then, just have a look at tests/geometryXVx/landau/fft/growthrate_t0.0to45.0.png:

and tests/geometryXVx/landau/fft/frequency_t0.0to45.0.png:

Dependencies

To install dependencies through spack, first follow the the 3 first steps of https://github.com/pdidev/spack

Then execute the following:

spack env create gyselalibxx spack.yaml
spack env activate gyselalibxx
spack concretize --reuse
spack install

For example, you can find a Dockerfile installing these dependencies on ubuntu in docker/gyselalibxx_env/Dockerfile.