DOLFINx_CUAS

Authors: Igor A. Baratta, Jørgen S. Dokken, Sarah Roggendorf

Add-on to DOLFINx for assembly of custom kernels. This packages provides a set of specialized kernels that can be used along-side DOLFINx for assembly into PETSc matrices. See: python/tests/test_cpp_kernels.py for examples on how to interface with the Python-layer.

See: cpp/demo/main.cpp for how to interface with the C++ layer.

Affine meshes (triangles and tetrahedra)

Lagrange elements (Scalar and Vector) (degree 1-5)

• ufl.inner(u, v) * ufl.dx (3D)
• ufl.inner(u, v) * ufl.ds (2D and 3D)
• ufl.inner(ufl.grad(u), ufl.grad(v)) * ufl.dx (3D)
• ufl.inner(ufl.grad(u), ufl.grad(v)) * ufl.ds (2D and 3D)
• ufl.inner(ufl.tr(ufl.sym(ufl.grad(u))), ufl.sym(ufl.grad(v))) * ufl.dx (3D)
• 2 * ufl.inner(ufl.sym(ufl.grad(u)), ufl.sym(ufl.grad(v))) * ufl.dx (3D)
• ufl.inner(ufl.sym(ufl.grad(u)), ufl.sym(ufl.grad(v))) * ufl.ds (2D and 3D)
• 2 * ufl.inner(ufl.sym(ufl.grad(u))) * ufl.FacetNormal(mesh), v) * ufl.ds (2D and 3D)
• 1. * v * ufl.dx (2D and 3D)
• 1. * v * ufl.ds (2D and 3D)

Non-affine meshes (quadrilaterals and hexahedra)

Lagrange elements (degree 1-5)

• ufl.inner(u, v) * ufl.ds

Dependencies

DOLFINX_CUAS depends on DOLFINx and in turn its dependencies. For instructions on how to install DOLFINx, see: https://github.com/FEniCS/dolfinx#readme

Installation

DOLFINX_CUAS is currently a header-only library.

Clone repository

    git clone https://github.com/Wells-Group/asimov-custom-assemblers.git

Build C++ layer

    cd asimov-custom-assemblers
    cmake -G Ninja -B build-dir -DCMAKE_BUILD_TYPE=Release cpp/
    ninja -C build-dir install

Compile C++ demos

Navigate to the folder of a C++ demo, then call

    ffcx *.ufl
    cmake -G Ninja -B build-dir -DCMAKE_BUILD_TYPE=Release .
    ninja -C build-dir 

Build Python-interface

    cd python
    pip3 install . -v --upgrade