SequentialDofMapping< IS, GI > Class Template Reference

Fallback for ParallelDofMapping in case there is only one mpi core. More...

#include <DOFMapping.hpp>

Public Types
using	size_type = std::size_t
using	DofIndex = size_type
using	LocalIndex = size_type
using	GlobalIndex = GI

Public Member Functions
template<class Communication >
	SequentialDofMapping (Communication &c)
size_type	localSize () const
	How many DOFs are owned by my processor?
std::array< size_type, 1 >	localSizes () const
	Return the sequence of number of local indices for all processors.
size_type	globalSize () const
	The total number of global DOFs.
std::array< GlobalIndex, 1 >	globalStarts () const
	Return the sequence of starting points of the global indices for all processors.
std::vector< GlobalIndex > const &	globalIndices () const
	Return the vector of global indices.
GlobalIndex	ghostSize () const
	Return number of ghost indices.
std::array< GlobalIndex, 0 >	ghostIndices () const
	Return the vector of ghost indices.
LocalIndex	globalToGhost (GlobalIndex const &n) const
	Map global index to local ghost index.
LocalIndex	dofToGhost (DofIndex const &n) const
	Map DOF index to local ghost index.
GlobalIndex	global (LocalIndex const &n) const
	Global index of local index n.
LocalIndex	globalToLocal (GlobalIndex const &n) const
	Map global index to consecutive local owner index.
LocalIndex	dofToLocal (DofIndex const &n) const
	Map DOF index to consecutive local owner index.
bool	owner (DofIndex const &n) const
	DOF index n is owned by this processor.
bool	globalOwner (GlobalIndex const &n) const
	Global index n is owned by this processor.
bool	globalOwner (int p, GlobalIndex const &n) const
	Global index n is owned by processor p.
template<class Communication >
void	update (Communication &c)
	Update the local to global mapping. Must be called before mapping local to global.
void	debug () const

Detailed Description

template<class IS, class GI = std::size_t>
class AMDiS::SequentialDofMapping< IS, GI >

Fallback for ParallelDofMapping in case there is only one mpi core.

---

The documentation for this class was generated from the following file:

• /home/docs/checkouts/readthedocs.org/user_builds/amdis-test/checkouts/develop/amdis/linearalgebra/DOFMapping.hpp