ItrSmootherCSR.c File Reference

Smoothers for dCSRmat matrices. More...

#include <math.h>
#include "fasp.h"
#include "fasp_functs.h"

Go to the source code of this file.

Functions
void	fasp_smoother_dcsr_jacobi_ff (dvector *x, dCSRmat *A, dvector *b, const INT nsweeps, INT *ordering, const REAL relax)
	Weighted Jacobi method as a smoother only for the fine points. More...
void	fasp_smoother_dcsr_jacobi (dvector *u, const INT i_1, const INT i_n, const INT s, dCSRmat *A, dvector *b, INT L, const REAL w)
	Weighted Jacobi method as a smoother. More...
void	fasp_smoother_dcsr_gs (dvector *u, const INT i_1, const INT i_n, const INT s, dCSRmat *A, dvector *b, INT L)
	Gauss-Seidel method as a smoother. More...
void	fasp_smoother_dcsr_gs_cf (dvector *u, dCSRmat *A, dvector *b, INT L, INT *mark, const INT order)
	Gauss-Seidel smoother with C/F ordering for Au=b. More...
void	fasp_smoother_dcsr_gs_ff (dvector *u, dCSRmat *A, dvector *b, INT L, INT *mark)
	Gauss-Seidel smoother with on F-points only for Au=b. More...
void	fasp_smoother_dcsr_sgs (dvector *u, dCSRmat *A, dvector *b, INT L)
	Symmetric Gauss-Seidel method as a smoother. More...
void	fasp_smoother_dcsr_sor (dvector *u, const INT i_1, const INT i_n, const INT s, dCSRmat *A, dvector *b, INT L, const REAL w)
	SOR method as a smoother. More...
void	fasp_smoother_dcsr_sor_cf (dvector *u, dCSRmat *A, dvector *b, INT L, const REAL w, INT *mark, const INT order)
	SOR smoother with C/F ordering for Au=b. More...
void	fasp_smoother_dcsr_ilu (dCSRmat *A, dvector *b, dvector *x, void *data)
	ILU method as a smoother. More...
void	fasp_smoother_dcsr_kaczmarz (dvector *u, const INT i_1, const INT i_n, const INT s, dCSRmat *A, dvector *b, INT L, const REAL w)
	Kaczmarz method as a smoother. More...
void	fasp_smoother_dcsr_L1diag (dvector *u, const INT i_1, const INT i_n, const INT s, dCSRmat *A, dvector *b, INT L)
	Diagonal scaling (using L1 norm) as a smoother. More...

Detailed Description

Smoothers for dCSRmat matrices.

Note

This file contains Level-2 (Itr) functions. It requires: AuxArray.c, AuxMemory.c, AuxMessage.c, AuxThreads.c, BlaArray.c, and BlaSpmvCSR.c

---

Copyright (C) 2009–Present by the FASP team. All rights reserved.

Released under the terms of the GNU Lesser General Public License 3.0 or later.

Definition in file ItrSmootherCSR.c.

Function Documentation

fasp_smoother_dcsr_gs()

void fasp_smoother_dcsr_gs	(	dvector *	u,
		const INT	i_1,
		const INT	i_n,
		const INT	s,
		dCSRmat *	A,
		dvector *	b,
		INT	L
	)

Gauss-Seidel method as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
i_1	Starting index
i_n	Ending index
s	Increasing step
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations

Author

Xuehai Huang, Chensong Zhang

Date

09/26/2009

Modified by Chunsheng Feng, Zheng Li on 09/01/2012

Definition at line 250 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_gs_cf()

void fasp_smoother_dcsr_gs_cf	(	dvector *	u,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		INT *	mark,
		const INT	order
	)

Gauss-Seidel smoother with C/F ordering for Au=b.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
mark	C/F marker array
order	C/F ordering: -1: F-first; 1: C-first

Author

Zhiyang Zhou

Date

11/12/2010

Modified by Chunsheng Feng, Xiaoqiang Yue on 05/24/2012

Definition at line 431 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_gs_ff()

void fasp_smoother_dcsr_gs_ff	(	dvector *	u,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		INT *	mark
	)

Gauss-Seidel smoother with on F-points only for Au=b.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
mark	C/F marker array

Author

Chensong Zhang

Date

10/11/2022

Definition at line 712 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_ilu()

void fasp_smoother_dcsr_ilu	(	dCSRmat *	A,
		dvector *	b,
		dvector *	x,
		void *	data
	)

ILU method as a smoother.

Parameters

A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
x	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
data	Pointer to user defined data

Author

Shiquan Zhang, Xiaozhe Hu

Date

2010/11/12

form residual zr = b - A x

Definition at line 1279 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_jacobi()

void fasp_smoother_dcsr_jacobi	(	dvector *	u,
		const INT	i_1,
		const INT	i_n,
		const INT	s,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		const REAL	w
	)

Weighted Jacobi method as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
i_1	Starting index
i_n	Ending index
s	Increasing step
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
w	Over-relaxation weight

Author

Xuehai Huang, Chensong Zhang

Date

09/26/2009

Modified by Chunsheng Feng, Zheng Li on 08/29/2012 Modified by Chensong Zhang on 08/24/2017: Pass weight w as a parameter

Definition at line 97 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_jacobi_ff()

void fasp_smoother_dcsr_jacobi_ff	(	dvector *	x,
		dCSRmat *	A,
		dvector *	b,
		const INT	nsweeps,
		INT *	ordering,
		const REAL	relax
	)

Weighted Jacobi method as a smoother only for the fine points.

Author

Xie Yan

Date

2022/12/12

Definition at line 34 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_kaczmarz()

void fasp_smoother_dcsr_kaczmarz	(	dvector *	u,
		const INT	i_1,
		const INT	i_n,
		const INT	s,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		const REAL	w
	)

Kaczmarz method as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
i_1	Starting index
i_n	Ending index
s	Increasing step
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
w	Over-relaxation weight

Author

Xiaozhe Hu

Date

2010/11/12

Modified by Chunsheng Feng, Zheng Li on 2012/09/01

Definition at line 1362 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_L1diag()

void fasp_smoother_dcsr_L1diag	(	dvector *	u,
		const INT	i_1,
		const INT	i_n,
		const INT	s,
		dCSRmat *	A,
		dvector *	b,
		INT	L
	)

Diagonal scaling (using L1 norm) as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
i_1	Starting index
i_n	Ending index
s	Increasing step
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations

Author

Xiaozhe Hu, James Brannick

Date

01/26/2011

Modified by Chunsheng Feng, Zheng Li on 09/01/2012

Definition at line 1508 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_sgs()

void fasp_smoother_dcsr_sgs	(	dvector *	u,
		dCSRmat *	A,
		dvector *	b,
		INT	L
	)

Symmetric Gauss-Seidel method as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations

Author

Xiaozhe Hu

Date

10/26/2010

Modified by Chunsheng Feng, Zheng Li on 09/01/2012

Definition at line 807 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_sor()

void fasp_smoother_dcsr_sor	(	dvector *	u,
		const INT	i_1,
		const INT	i_n,
		const INT	s,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		const REAL	w
	)

SOR method as a smoother.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
i_1	Starting index
i_n	Ending index
s	Increasing step
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
w	Over-relaxation weight

Author

Xiaozhe Hu

Date

10/26/2010

Modified by Chunsheng Feng, Zheng Li on 09/01/2012

Definition at line 931 of file ItrSmootherCSR.c.

fasp_smoother_dcsr_sor_cf()

void fasp_smoother_dcsr_sor_cf	(	dvector *	u,
		dCSRmat *	A,
		dvector *	b,
		INT	L,
		const REAL	w,
		INT *	mark,
		const INT	order
	)

SOR smoother with C/F ordering for Au=b.

Parameters

u	Pointer to dvector: the unknowns (IN: initial, OUT: approximation)
A	Pointer to dBSRmat: the coefficient matrix
b	Pointer to dvector: the right hand side
L	Number of iterations
w	Over-relaxation weight
mark	C/F marker array
order	C/F ordering: -1: F-first; 1: C-first

Author

Zhiyang Zhou

Date

2010/11/12

Modified by Chunsheng Feng, Zheng Li on 08/29/2012

Definition at line 1061 of file ItrSmootherCSR.c.