serac::GradientAssemblyLookupTables Struct Reference

this object figures out the sparsity pattern associated with a finite element discretization of the given test and trial function spaces, and records which nonzero each element "stiffness" matrix maps to, to facilitate assembling the element matrices into the global sparse matrix. e.g. More...

#include <dof_numbering.hpp>

Classes
struct	Entry
	a type for representing a nonzero entry in a sparse matrix More...

Public Member Functions
	GradientAssemblyLookupTables (const serac::BlockElementRestriction &block_test_dofs, const serac::BlockElementRestriction &block_trial_dofs)
	create lookup tables describing which degrees of freedom correspond to each domain/boundary element More...
uint32_t	operator() (int i, int j) const
	return the index (into the nonzero entries) corresponding to entry (i,j) More...

Public Attributes
uint32_t	nnz
	how many nonzero entries appear in the sparse matrix
std::vector< int >	row_ptr
	array holding the offsets for a given row of the sparse matrix i.e. row r corresponds to the indices [row_ptr[r], row_ptr[r+1])
std::vector< int >	col_ind
	array holding the column associated with each nonzero entry
std::unordered_map< Entry, uint32_t, Entry::Hasher >	nz_LUT
	nz_LUT returns the index of the col_ind / value CSR arrays corresponding to the (i,j) entry

Detailed Description

this object figures out the sparsity pattern associated with a finite element discretization of the given test and trial function spaces, and records which nonzero each element "stiffness" matrix maps to, to facilitate assembling the element matrices into the global sparse matrix. e.g.

element_nonzero_LUT(e, i, j) says where (in the global sparse matrix) to put the (i,j) component of the matrix associated with element element matrix e

Note: due to an internal inconsistency between mfem::FiniteElementSpace and mfem::FaceRestriction, we choose to use the Restriction operator as the "source of truth", since we are also using its convention for quadrature point numbering.

Definition at line 253 of file dof_numbering.hpp.

Constructor & Destructor Documentation

GradientAssemblyLookupTables()

serac::GradientAssemblyLookupTables::GradientAssemblyLookupTables ( const serac::BlockElementRestriction &  block_test_dofs, const serac::BlockElementRestriction &  block_trial_dofs  )

inline

create lookup tables describing which degrees of freedom correspond to each domain/boundary element

Parameters

block_test_dofs	object containing information about dofs for the test space
block_trial_dofs	object containing information about dofs for the trial space

Definition at line 287 of file dof_numbering.hpp.

Member Function Documentation

operator()()

uint32_t serac::GradientAssemblyLookupTables::operator() ( int  i, int  j  ) const

inline

return the index (into the nonzero entries) corresponding to entry (i,j)

Parameters

i	the row
j	the column

Definition at line 352 of file dof_numbering.hpp.

---

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

• src/serac/numerics/functional/dof_numbering.hpp