elliptic Namespace Reference

Items related to composing nonlinear elliptic solver executables. More...

Namespaces
namespace	analytic_data
	Items related to pointwise analytic data for elliptic solves, such as initial guesses, analytic solutions, and background quantities in elliptic PDEs.
namespace	BoundaryConditions
	Boundary conditions for elliptic systems.
namespace	dg
	Functionality related to discontinuous Galerkin discretizations of elliptic equations.
namespace	OptionTags
	Option tags for nonlinear elliptic solver executables.
namespace	protocols
	Protocols related to elliptic systems
namespace	subdomain_preconditioners
	Linear solvers that approximately invert the elliptic::dg::subdomain_operator::SubdomainOperator to make the Schwarz subdomain solver converge faster.
namespace	Triggers
	Triggers for elliptic executables.

Classes
struct	DefaultElementsAllocator
	A Parallel::protocols::ArrayElementsAllocator that creates array elements to cover the initial computational domain. More...
struct	DgElementArray
	The parallel component responsible for managing the DG elements that compose the computational domain. More...
struct	Solver
	A complete nonlinear elliptic solver stack. Use to compose an executable. More...

Typedefs
template<typename System , bool Linearized>
using	get_sources_computer = tmpl::conditional_t< Linearized, typename detail::sources_computer_linearized< System >::type, typename System::sources_computer >
	The System::sources_computer or the System::sources_computer_linearized, depending on the Linearized parameter. If the system has no sources_computer_linearized alias it is assumed to be linear, so the System::sources_computer is returned either way.
template<typename System , bool Linearized>
using	get_sources_argument_tags = typename tmpl::conditional_t< std::is_same_v< get_sources_computer< System, Linearized >, void >, detail::NoSourcesComputer, get_sources_computer< System, Linearized > >::argument_tags
	The argument_tags of either the System::sources_computer or the System::sources_computer_linearized, depending on the Linearized parameter, or an empty list if the sources computer is void.

Enumerations
enum class	BoundaryConditionType { Dirichlet , Neumann }
	Identify types of boundary conditions for elliptic equations. More...

Functions
template<bool Linearized, typename ArgsTransform = void, typename BoundaryConditionClasses = tmpl::list<>, size_t Dim, typename DbTagsList , typename MapKeys , typename... FieldsAndFluxes>
void	apply_boundary_condition (const elliptic::BoundaryConditions::BoundaryCondition< Dim > &boundary_condition, const db::DataBox< DbTagsList > &box, const MapKeys &map_keys_to_direction, FieldsAndFluxes &&... fields_and_fluxes)
	Apply the boundary_condition to the fields_and_fluxes with arguments from interface tags in the DataBox. More...
std::ostream &	operator<< (std::ostream &os, BoundaryConditionType boundary_condition_type)

Detailed Description

Items related to composing nonlinear elliptic solver executables.

Functionality related to solving elliptic partial differential equations.

Enumeration Type Documentation

BoundaryConditionType

enum class elliptic::BoundaryConditionType

strong

Identify types of boundary conditions for elliptic equations.

Enumerator
Dirichlet	Dirichlet boundary conditions like \(u(x_0)=u_0\).
Neumann	Neumann boundary conditions like \(n^i\partial_i u(x_0)=v_0\), where \(\boldsymbol{n}\) is the normal to the domain boundary.

Function Documentation

apply_boundary_condition()

template<bool Linearized, typename ArgsTransform = void, typename BoundaryConditionClasses = tmpl::list<>, size_t Dim, typename DbTagsList , typename MapKeys , typename... FieldsAndFluxes>

void elliptic::apply_boundary_condition	(	const elliptic::BoundaryConditions::BoundaryCondition< Dim > &	boundary_condition,
		const db::DataBox< DbTagsList > &	box,
		const MapKeys &	map_keys_to_direction,
		FieldsAndFluxes &&...	fields_and_fluxes
	)

Apply the boundary_condition to the fields_and_fluxes with arguments from interface tags in the DataBox.

This functions assumes the arguments for the boundary_condition are stored in the DataBox in tags domain::Tags::Faces<Dim, Tag>. This may turn out not to be the most efficient setup, so code that uses the boundary conditions doesn't have to use this function but can procure the arguments differently. For example, future optimizations may involve storing a subset of arguments that don't change during an elliptic solve in direction-maps in the DataBox, and slicing other arguments to the interface every time the boundary conditions are applied.

The ArgsTransform template parameter can be used to transform the set of argument tags for the boundary conditions further. It must be compatible with tmpl::transform. For example, it may wrap the tags in another prefix. Set it to void (default) to apply no transformation.

The BoundaryConditionClasses can be used to list a set of classes derived from elliptic::BoundaryConditions::BoundaryCondition that are iterated to determine the concrete type of boundary_condition. It can be tmpl::list<> (default) to use the classes listed in Metavariables::factory_creation instead.