Public Member Functions
integer function	crs_search (N, Array, VALUE)

type(matrix_t) function, pointer	crs_transpose (A)

subroutine	crs_mergematrix (A, B, PermA, PermB)


subroutine	crs_zeromatrix (A)

subroutine	crs_zerorow (A, n)

subroutine	crs_sortmatrix (A, ValuesToo)

subroutine	crs_sortbasicmatrix (A, ValuesToo)

subroutine	crs_makematrixindex (A, i, j)

subroutine	crs_addtomatrixelement (A, i, j, VALUE)

subroutine	crs_setmatrixelement (A, i, j, VALUE)

real(kind=dp) function	crs_getmatrixelement (A, i, j)

real(kind=dp) function	crs_changematrixelement (A, i, j, NewValue)

subroutine	crs_moverow (A, n1, n2, coeff)

subroutine	crs_gluelocalmatrix (A, N, Dofs, Indeces, LocalMatrix)

subroutine	crs_gluelocalsubmatrix (A, row0, col0, Nrow, Ncol, RowInds, ColInds, RowDofs, ColDofs, LocalMatrix)

subroutine	crs_setsymmdirichlet (A, b, n, val)

real(kind=dp) function	crs_rowsum (A, k)

subroutine	crs_rowsuminfo (A, Values)

type(matrix_t) function, pointer	crs_creatematrix (N, Total, RowNonzeros, Ndeg, Reorder, AllocValues)

subroutine	crs_printmatrix (A)

subroutine	crs_printrhs (A)

subroutine	crs_matrixvectormultiply (A, u, v)

subroutine	crs_transposematrixvectormultiply (A, u, v)

subroutine	crs_complexmatrixvectormultiply (A, u, v)

subroutine	crs_applyprojector (PMatrix, u, uperm, v, vperm, Trans)

subroutine	crs_diagprecondition (u, v, ipar)

subroutine	crs_complexdiagprecondition (u, v, ipar)

subroutine	crs_blockdiagonal (A, B, Blocks)

subroutine	crs_removezeros (A)

subroutine	crs_fctloworder (A)

subroutine	crs_copymatrixtopology (A, B)

subroutine	crs_blockmatrixpick (A, B, Blocks, Nrow, Ncol)

subroutine	crs_blockmatrixpick2 (A, B, BlockStruct, Nrow, Ncol)

logical function	crs_copymatrixprec (A, B)

logical function	crs_incompletelu (A, ILUn)

logical function	crs_complexincompletelu (A, ILUn)

logical function	crs_ilut (A, TOL)

logical function	crs_complexilut (A, TOL)

subroutine	crs_luprecondition (u, v, ipar)

subroutine	crs_complexluprecondition (u, v, ipar)

subroutine	crs_lusolve (N, A, b)

subroutine	crs_complexlusolve (N, A, b)

subroutine	crs_matrixvectorprod (u, v, ipar)

subroutine	crs_complexmatrixvectorprod (u, v, ipar)

subroutine	crs_inspectmatrix (A)

Member Function/Subroutine Documentation

subroutine crsmatrix::crs_addtomatrixelement	(	type(matrix_t)	A,
		integer, intent(in)	i,
		integer, intent(in)	j,
		real(kind=dp), intent(in)	VALUE
	)

Add a given value to an element of a CRS format matrix.

Parameters

	a	Structure holding the matrix
[in]	i	row number of the matrix element
[in]	j	column number of the matrix element
[in]	value	value to be added to the matrix element

References crs_search().

Referenced by solverutils::addtomatrixelement(), crs_moverow(), gebhardtfactors(), nodaldisplacementpenalty(), radiationfactors(), and vankacreate().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_applyprojector	(	type(matrix_t)	PMatrix,
		real(kind=dp), dimension(:)	u,
		integer, dimension(:), pointer	uperm,
		real(kind=dp), dimension(:)	v,
		integer, dimension(:), pointer	vperm,
		logical, optional	Trans
	)

Referenced by applyprojector(), and gmgsweep().

Here is the caller graph for this function:

subroutine crsmatrix::crs_blockdiagonal	(	type(matrix_t), intent(in)	A,
		type(matrix_t)	B,
		integer, intent(in)	Blocks
	)

Pics the block diagonal entries from matrix A to build matrix B.

Parameters

[in]	a	The initial matrix
	b	The block diagonal matrix
[in]	blocks	Number of blocks used in the decomposition

Referenced by itersolve::itersolver().

Here is the caller graph for this function:

subroutine crsmatrix::crs_blockmatrixpick	(	type(matrix_t), intent(in)	A,
		type(matrix_t)	B,
		integer, intent(in)	Blocks,
		integer, intent(in)	Nrow,
		integer, intent(in)	Ncol
	)

Pics a block from matrix A to build matrix B. It is assumed that the matrix is split into given number of equally sized blocks.

Parameters

[in]	a	Initial matrix
	b	Submatrix picked from the larger matrix
[in]	blocks	Number of blocks in the initial matrix
[in]	nrow	Row to be picked
[in]	ncol	Column to be picked

References messages::fatal(), and messages::warn().

Referenced by blocksolve::blockpickmatrix().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_blockmatrixpick2	(	type(matrix_t), intent(in)	A,
		type(matrix_t)	B,
		integer, dimension(:), pointer	BlockStruct,
		integer, intent(in)	Nrow,
		integer, intent(in)	Ncol
	)

Pics a block from matrix A to build matrix B. This subroutine enables the use of nontrivial block decompositions.

Parameters

[in]	a	Initial matrix
	b	Submatrix picked from the larger matrix
	blockstruct	Block decomposition structure of the initial matrix
[in]	nrow	Row to be picked
[in]	ncol	Column to be picked

References messages::fatal(), and messages::warn().

Referenced by blocksolve::blockpickmatrix().

Here is the call graph for this function:

Here is the caller graph for this function:

real(kind=dp) function crsmatrix::crs_changematrixelement	(	type(matrix_t), intent(in)	A,
		integer, intent(in)	i,
		integer, intent(in)	j,
		real(kind=dp), intent(in)	NewValue
	)

Get a given matrix entry from CRS format matrix and replace it with a new value.

Parameters

[in]	a	Structure holding the matrix
[in]	i	row number of the matrix element
[in]	j	column number of the matrix element
[in]	newvalue	Value to be set

References crs_search().

Referenced by solverutils::changematrixelement().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_complexdiagprecondition	(	complex(kind=dp), dimension(*), intent(out)	u,
		complex(kind=dp), dimension(*), intent(in)	v,
		integer, dimension(*)	ipar
	)

Diagonal preconditioning of a CRS format matrix for complex valued matrix equations. Matrix is accessed from a global variable GlobalMatrix.

Parameters

[out]	u	Resulting approximate solution after preconditioning
[in]	v	Given right-hand-side
	ipar	structure holding info from (HUTIter-iterative solver package)

References generalutils::sortf().

Referenced by itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

logical function crsmatrix::crs_complexilut	(	type(matrix_t)	A,
		real(kind=dp), intent(in)	TOL
	)

Buids an incomplete (ILUT) factorization for an iterative solver preconditioner. Complex matrix version.

Parameters

	a	Structure holding input matrix, will also hold the factorization on exit.
[in]	tol	Drop toleranece: if ILUT(i,j) <= NORM(A(i,:))*TOL the value is dropped.

References complexcomputeilut(), and messages::info().

Referenced by eigensolve::arpackeigensolvecomplex(), and itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

logical function crsmatrix::crs_complexincompletelu	(	type(matrix_t)	A,
		integer, intent(in)	ILUn
	)

Buids an incomplete (ILU(n)) factorization for an iterative solver preconditioner. Complex matrix version.

Parameters

	a	Strcture holding input matrix, will also hold the factorization on exit.
[in]	ilun	Order of fills allowed 0-9

References messages::fatal(), messages::info(), and initializecomplexilu1().

Referenced by bicgstablouteriteration(), bicgstabouteriteration(), gcrouteriteration(), innerouteriteration(), itersolve::itersolver(), and monolithicsolve().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_complexluprecondition	(	complex(kind=dp), dimension(huti_ndim), intent(out)	u,
		complex(kind=dp), dimension(huti_ndim), intent(in)	v,
		integer, dimension(*), intent(in)	ipar
	)

Incomplete factorization preconditioner solver for a CRS format matrix. Matrix is accessed from a global variable GlobalMatrix. The Complex version.

Parameters

[in]	ipar	structure holding info from (HUTIter-iterative solver package)
[in]	v	Right-hand-side vector
[out]	u	Solution vector

References crs_complexlusolve().

Referenced by itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_complexlusolve	(	integer, intent(in)	N,
		type(matrix_t), intent(in)	A,
		complex(kind=dp), dimension(n)	b
	)

Solve a complex system Ax=b after factorization A=LUD has been done. This routine is meant as a part of a preconditioner for an iterative solver.

Parameters

[in]	a	Structure holding input matrix
[in]	n	Size of the system
	b	on entry the RHS vector, on exit the solution vector.

References complexlusolve().

Referenced by eigensolve::arpackeigensolvecomplex(), ccg(), and crs_complexluprecondition().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_complexmatrixvectormultiply	(	type(matrix_t), intent(in)	A,
		complex(kind=dp), dimension(*), intent(in)	u,
		complex(kind=dp), dimension(*), intent(out)	v
	)

Matrix vector product (v = Au) for a matrix given in CRS format assuming complex valued matrix equation.

Parameters

[in]	u	Vector to be multiplied
[out]	v	Result vector
[in]	a	Structure holding matrix

Referenced by amgsweep(), eigensolve::arpackeigensolvecomplex(), cmgsweep(), and mgcmv().

Here is the caller graph for this function:

subroutine crsmatrix::crs_complexmatrixvectorprod	(	complex(kind=dp), dimension(huti_ndim), intent(in)	u,
		complex(kind=dp), dimension(huti_ndim)	v,
		integer, dimension(*), intent(in)	ipar
	)

Complex matrix vector product (v = Au) for a matrix given in CRS format. The matrix is accessed from a global variable GlobalMatrix.

Parameters

[in]	ipar	Structure holding info HUTIter-iterative solver package
[in]	u	vector to multiply u
	v	result vector

References complexmatvec().

Referenced by itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

logical function crsmatrix::crs_copymatrixprec	(	type(matrix_t), intent(in)	A,
		type(matrix_t)	B
	)

Copies some preconditioning structures from matrix A to B. The intent is to allow saving of memory and CPU time for cases where similar preconditioning could be used for many components.

Parameters

[in]	a	Initial matrix
	b	New matrix with the same preconditioner

References messages::info().

Referenced by blocksolve::blockmatrixprec(), and blocksolve::blockstandarditer().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_copymatrixtopology	(	type(matrix_t), intent(in)	A,
		type(matrix_t)	B
	)

Copies the matrix topology from matrix A to build matrix B. Note that the topology is really reused, so that if matrix A is destroyed also matrix B becomes unusable.

Parameters

[in]	a	Initial matrix
	b	New matrix with same matrix topology

References messages::fatal(), and messages::info().

Referenced by blocksolve::blockpickmatrix(), and blocksolve::blockprecmatrix().

Here is the call graph for this function:

Here is the caller graph for this function:

type(matrix_t) function, pointer crsmatrix::crs_creatematrix	(	integer, intent(in)	N,
		integer, intent(in)	Total,
		integer, dimension(:), intent(in)	RowNonzeros,
		integer, intent(in)	Ndeg,
		integer, dimension(:), intent(in)	Reorder,
		logical, intent(in)	AllocValues
	)

Create the structures required for a CRS format matrix.

Parameters

[in]	n	Number of rows in the matrix
[in]	total	Total number of nonzero entries in the matrix
[in]	ndeg	Negrees of freedom
[in]	rownonzeros	Number of nonzero entries in rows of the matrix
[in]	reorder	Permutation index for bandwidth reduction
[in]	allocvalues	Should the values arrays be allocated ?

References generalutils::allocatematrix(), and messages::fatal().

Referenced by elementutils::creatematrix(), gebhardtfactors(), and radiationfactors().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_diagprecondition	(	real(kind=dp), dimension(*), intent(out)	u,
		real(kind=dp), dimension(*), intent(in)	v,
		integer, dimension(*)	ipar
	)

Diagonal preconditioning of a CRS format matrix. Matrix is accessed from a global variable GlobalMatrix. Note that if the matrix has been scaled so that the diagonal entries are already ones, this subroutine is obsolite.

Parameters

[out]	u	Resulting approximate solution after preconditioning
[in]	v	Given right-hand-side
	ipar	structure holding info from (HUTIter-iterative solver package)

References generalutils::sortf().

Referenced by itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_fctloworder ( type(matrix_t) A)

Makes a algebraic lower order scheme assuming steady state advection-diffusion equation. This can be applied together with flux corrected transport (FCT) scheme. Also creates a lumped mass to MassValuesLumped and saves the original stiffness matrix values to BulkValues.

For more information see, for example, Dmitri Kuzmin (2008): "Explicit and implicit FEM-FCT algorithms with flux linearization"

Parameters

a	Initial higher order matrix

References crs_rowsuminfo(), messages::fatal(), and messages::info().

Referenced by defutils::defaultfinishassembly().

Here is the call graph for this function:

Here is the caller graph for this function:

real(kind=dp) function crsmatrix::crs_getmatrixelement	(	type(matrix_t), intent(in)	A,
		integer, intent(in)	i,
		integer, intent(in)	j
	)

Get a given matrix entry from CRS format matrix.

Parameters

[in]	a	Structure holding the matrix
[in]	i	row number of the matrix element
[in]	j	column number of the matrix element

References crs_search().

Referenced by solverutils::getmatrixelement(), vanka(), and vankacreate().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_gluelocalmatrix	(	type(matrix_t)	A,
		integer, intent(in)	N,
		integer, intent(in)	Dofs,
		integer, dimension(:), intent(in)	Indeces,
		real(kind=dp), dimension(:,:), intent(in)	LocalMatrix
	)

Add a set of values (.i.e. element stiffness matrix) to a CRS format matrix.

Parameters

	a	Structure holding matrix
[in]	localmatrix	A (N x Dofs) x ( N x Dofs) matrix holding the values to be added to the CRS format matrix
[in]	n	Number of nodes in element
[in]	dofs	Number of degrees of freedom for one node
[in]	indeces	Maps element node numbers to global (or partition) node numbers

References localmatrix().

Referenced by modeldescription::getnodalelementsize(), freesurface::poissonsolve(), solverutils::updateglobalequations(), updateglobalpreconditioner(), and solverutils::updatemassmatrix().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_gluelocalsubmatrix	(	type(matrix_t)	A,
		integer, intent(in)	row0,
		integer, intent(in)	col0,
		integer, intent(in)	Nrow,
		integer, intent(in)	Ncol,
		integer, dimension(:), intent(in)	RowInds,
		integer, dimension(:), intent(in)	ColInds,
		integer, intent(in)	RowDofs,
		integer, intent(in)	ColDofs,
		real(kind=dp), dimension(:,:), intent(in)	LocalMatrix
	)

Add a set of values (.i.e. element stiffness matrix) to a CRS format matrix. For this matrix the entries are ordered so that first for one dof you got all nodes, and then for second etc. There may be on offset to the entries making the subroutine suitable for coupled monolithic matrix assembly.

Parameters

[in]	localmatrix	A (Nrow x RowDofs) x ( Ncol x ColDofs) matrix holding the values to be added to the CRS format matrix
	a	Structure holding matrix
[in]	nrow	Number of nodes for the row variable
[in]	ncol	Number of nodes for the column variable
[in]	rowdofs	Number of dofs for row variable
[in]	coldofs	Number of dofs for column variable
[in]	col0	Offset for column variable
[in]	row0	Offset for row variable
[in]	rowinds	Permutation of the row dofs
[in]	colinds	Permutation of the column dofs

References localmatrix().

Referenced by solverutils::gluelocalsubmatrix().

Here is the call graph for this function:

Here is the caller graph for this function:

logical function crsmatrix::crs_ilut	(	type(matrix_t)	A,
		real(kind=dp), intent(in)	TOL
	)

Buids an incomplete (ILUT) factorization for an iterative solver preconditioner. Real matrix version.

Parameters

	a	Structure holding input matrix, will also hold the factorization on exit.
[in]	tol	Drop toleranece: if ILUT(i,j) <= NORM(A(i,:))*TOL the value is dropped.

References computeilut(), and messages::info().

Referenced by multigrid::amgsolve(), eigensolve::arpackeigensolve(), eigensolve::arpackstabeigensolve(), multigrid::cmgsolve(), multigrid::gmgsolve(), itersolve::itersolver(), paralleleigensolve::parallelarpackeigensolve(), and multigrid::pmgsolve().

Here is the call graph for this function:

Here is the caller graph for this function:

logical function crsmatrix::crs_incompletelu	(	type(matrix_t)	A,
		integer, intent(in)	ILUn
	)

Builds an incomplete (ILU(n)) factorization for a iterative solver preconditioner. Real matrix version.

Parameters

	a	Strcture holding input matrix, will also hold the factorization on exit.
[in]	ilun	Order of fills allowed 0-9

References messages::fatal(), messages::info(), initializeilu1(), and messages::warn().

Referenced by multigrid::amgsolve(), eigensolve::arpackdampedeigensolve(), multigrid::cmgsolve(), multigrid::gmgsolve(), itersolve::itersolver(), paralleleigensolve::parallelarpackeigensolve(), and multigrid::pmgsolve().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_inspectmatrix ( type(matrix_t) A)

Check the matrix for correctness.

Parameters

a	Structure holding the matrix

subroutine crsmatrix::crs_luprecondition	(	real(kind=dp), dimension(huti_ndim), intent(out)	u,
		real(kind=dp), dimension(huti_ndim), intent(in)	v,
		integer, dimension(*), intent(in)	ipar
	)

Incomplete factorization preconditioner solver for a CRS format matrix. Matrix is accessed from a global variable GlobalMatrix. The real version.

Parameters

[in]	ipar	structure holding info from (HUTIter-iterative solver package)
[in]	v	Right-hand-side vector
[out]	u	Solution vector

References crs_lusolve().

Referenced by itersolve::itersolver().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_lusolve	(	integer, intent(in)	N,
		type(matrix_t), intent(in)	A,
		double precision, dimension(n)	b
	)

Solve a system (Ax=b) after factorization A=LUD has been done. This routine is meant as a part of a preconditioner for an iterative solver.

Parameters

[in]	a	Structure holding input matrix
[in]	n	Size of the system
	b	on entry the RHS vector, on exit the solution vector.

References lusolve().

Referenced by bicg(), paralleleigensolve::bicgpareigen(), bicgstab(), cg(), paralleleigensolve::cgpareigen(), crs_luprecondition(), and eigensolve::eigenbicg().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_makematrixindex	(	type(matrix_t)	A,
		integer, intent(in)	i,
		integer, intent(in)	j
	)

Fill in the column number to a CRS format matrix (values are not affected in any way).

Parameters

	a	Structure holding matrix
[in]	i	row number of the matrix element
[in]	j	column number of the matrix element

References messages::error().

Referenced by elementutils::creatematrix(), crs_matrixmatrixmultiply(), gebhardtfactors(), and radiationfactors().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_matrixvectormultiply	(	type(matrix_t), intent(in)	A,
		real(kind=dp), dimension(*), intent(in)	u,
		real(kind=dp), dimension(*), intent(out)	v
	)

Matrix vector product (v = Au) for a matrix given in CRS format.

Parameters

[in]	u	Vector to be multiplied
[out]	v	Result vector
[in]	a	Structure holding matrix

Referenced by amgsweep(), eigensolve::arpackeigensolve(), eigensolve::arpackstabeigensolve(), blocksolve::blockmatrixprec(), blocksolve::blockmatrixvectorprod(), blocksolve::blockupdaterhs(), eigensolve::checkresiduals(), cmgsweep(), eigensolve::eigenmgmv1(), eigensolve::eigenmgmv2(), eliminateperiodic(), freesurfacesolver(), solverutils::matrixvectormultiply(), mgmv(), mymv(), pmgsweep(), pmv(), preconditioningiteration(), rigidbody(), stokessolver(), and itersolve::stopc().

Here is the caller graph for this function:

subroutine crsmatrix::crs_matrixvectorprod	(	real(kind=dp), dimension(huti_ndim), intent(in)	u,
		real(kind=dp), dimension(huti_ndim)	v,
		integer, dimension(*), intent(in)	ipar
	)

Matrix vector product (v = Au) for a matrix given in CRS format. The matrix is accessed from a global variable GlobalMatrix.

Parameters

[in]	ipar	Structure holding info HUTIter-iterative solver package
[in]	u	vector to multiply u
	v	result vector

References matvec().

Referenced by itersolve::itersolver(), and vankaprec().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_mergematrix	(	type(matrix_t), pointer	A,
		type(matrix_t), pointer	B,
		integer, dimension(:), optional, pointer	PermA,
		integer, dimension(:), optional, pointer	PermB
	)

Add another matrix B to matrix A, and eliminate B.

Parameters

a	Structure holding the master matrix
b	Structure holding the slave matrix
perma	Permutation of the master dofs
permb	Permutation of the slave dofs

References messages::fatal(), and messages::info().

Referenced by solverutils::mortarrobinsolver().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_moverow	(	type(matrix_t)	A,
		integer, intent(in)	n1,
		integer, intent(in)	n2,
		real(kind=dp), optional	coeff
	)

Add a row together with another row of a CRS matrix, and thereafter zero it.

Parameters

	a	Structure holding the matrix
[in]	n1	Row number to be copied and zerod
[in]	n2	Row number to be added
	coeff	Optional coefficient to multiply the row to be copied with

References crs_addtomatrixelement().

Referenced by solverutils::moverow().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_printmatrix ( type(matrix_t) A)

Prints the values of the CRS matrix to standard output.

Parameters

a	Structure holding matrix

Referenced by defutils::defaultsolve().

Here is the caller graph for this function:

subroutine crsmatrix::crs_printrhs ( type(matrix_t) A)

Prints the values of the right-hand-side vector to standard output.

Parameters

a	Structure holding matrix

Referenced by defutils::defaultsolve().

Here is the caller graph for this function:

subroutine crsmatrix::crs_removezeros ( type(matrix_t) A)

Removes zeros from the matrix structure. This might be done in order to save memory, or to speed up the matrix operations. One must be carefull since the fact the an entry is zero does not always imply that it would be zero throughout the simulation.

Parameters

a	The matrix which will be returned with the non-zeros removed

References messages::info().

Here is the call graph for this function:

real(kind=dp) function crsmatrix::crs_rowsum	(	type(matrix_t), intent(in)	A,
		integer, intent(in)	k
	)

Computes the rowsoum of a given row in a CRS matrix.

Parameters

[in]	a	Structure holding matrix
[in]	k	Row index

subroutine crsmatrix::crs_rowsuminfo	(	type(matrix_t), intent(in)	A,
		real(kind=dp), dimension(:), optional, pointer	Values
	)

Computes information on the matrix rowsums.

References messages::info().

Referenced by crs_fctloworder().

Here is the call graph for this function:

Here is the caller graph for this function:

integer function crsmatrix::crs_search	(	integer	N,
		integer, dimension(:)	Array,
		integer	VALUE
	)

Search CRS matrix for what?

Referenced by crs_addtomatrixelement(), crs_changematrixelement(), crs_getmatrixelement(), crs_setmatrixelement(), and crs_setsymmdirichlet().

Here is the caller graph for this function:

subroutine crsmatrix::crs_setmatrixelement	(	type(matrix_t)	A,
		integer, intent(in)	i,
		integer, intent(in)	j,
		real(kind=dp), intent(in)	VALUE
	)

Set a given value to an element of a CRS format matrix.

Parameters

	a	Structure holding the matrix
[in]	i	row number of the matrix element
[in]	j	column number of the matrix element
[in]	value	new value of the matrix element

References crs_search().

Referenced by freesurface::poissonsolve(), and solverutils::setmatrixelement().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_setsymmdirichlet	(	type(matrix_t)	A,
		real(kind=dp), dimension(:)	b,
		integer, intent(in)	n,
		real(kind=dp), intent(in)	val
	)

When Dirichlet conditions are set by zeroing the row except for setting the diagonal entry to one, the matrix symmetry is broken. This routine maintains the symmetric structure of the matrix equation.

Parameters

	a	Structure holding matrix
[in]	n	Index of the dofs to be fixed
	b	right-hand-side of the matrix equation
[in]	val	Dirichlet value to be set

References crs_search(), and crs_zerorow().

Referenced by bulkassembly(), defutils::defaultdirichletbcs(), fetisolve::feti(), outletcompute(), solverutils::setdirichletpoint(), setdirichletpoints(), magnetodynamicsutils::setdoftovaluec(), magnetodynamicsutils::setdoftovaluer(), setelementvalues(), setlimitervalues(), and setpointvalues().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_sortbasicmatrix	(	type(basicmatrix_t), target	A,
		logical, intent(in), optional	ValuesToo
	)

Sort columns to ascending order for rows of a CRS format matrix- Optionally also sort the corresponging values of the matrix. This operates just on a basic matrix type.

Parameters

	a	Structure holding the matrix
[in]	valuestoo	Should the values be sorted as well?

References sort(), and generalutils::sortf().

Referenced by sparitersolve::combinecrsmatindices().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_sortmatrix	(	type(matrix_t)	A,
		logical, intent(in), optional	ValuesToo
	)

Sort columns to ascending order for rows of a CRS format matrix- Optionally also sort the corresponging values of the matrix.

Parameters

	a	Structure holding the matrix
[in]	valuestoo	Should the values be sorted as well?

References sort(), and generalutils::sortf().

Referenced by boundaryassembly(), bsolver(), elementutils::creatematrix(), crs_matrixmatrixmultiply(), eliminatedirichlet(), gebhardtfactors(), elementutils::initializematrix(), listmatrix::list_tocrs(), listmatrix::list_tocrsmatrix(), magnetodynamics2d(), radiationfactors(), rigidbody(), setperiodicboundariespass1(), and sparitersolve::sparinitsolve().

Here is the call graph for this function:

Here is the caller graph for this function:

type(matrix_t) function, pointer crsmatrix::crs_transpose ( type(matrix_t), pointer A)

Calculate transpose of A in CRS format: B = A^T.

References generalutils::allocatematrix(), messages::error(), and messages::fatal().

Referenced by multigrid::amgsolve().

Here is the call graph for this function:

Here is the caller graph for this function:

subroutine crsmatrix::crs_transposematrixvectormultiply	(	type(matrix_t), intent(in)	A,
		real(kind=dp), dimension(*), intent(in)	u,
		real(kind=dp), dimension(*), intent(out)	v
	)

Matrix vector product (v = A^T u) for a matrix given in CRS format.

Parameters

[in]	u	Vector to be multiplied
[out]	v	Result vector
[in]	a	Structure holding matrix

Referenced by solverutils::transposematrixvectormultiply().

Here is the caller graph for this function:

subroutine crsmatrix::crs_zeromatrix ( type(matrix_t) A)

Zero a CRS format matrix.

Parameters

a	Structure holding the matrix

Referenced by acousticssolver(), gebhardtfactors(), solverutils::initializetozero(), freesurface::poissonsolve(), radiationfactors(), and stokessolver().

Here is the caller graph for this function:

subroutine crsmatrix::crs_zerorow	(	type(matrix_t)	A,
		integer	n
	)

Zero given row from a CRS format matrix.

Parameters

a	Structure holding the matrix
n	Row number to be zeroed

References lists::listgetlogical().

Referenced by crs_setsymmdirichlet(), freesurface::poissonsolve(), and solverutils::zerorow().

Here is the call graph for this function:

Here is the caller graph for this function:

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

CRSMatrix.src

Public Member Functions

Member Function/Subroutine Documentation