xref: /petsc/src/mat/interface/matrix.c (revision 3f9fe4453ac6fcef10788d326c676dfc3fb403b0)
1 #define PETSCMAT_DLL
2 
3 /*
4    This is where the abstract matrix operations are defined
5 */
6 
7 #include "private/matimpl.h"        /*I "petscmat.h" I*/
8 #include "private/vecimpl.h"
9 
10 /* Logging support */
11 PetscClassId PETSCMAT_DLLEXPORT MAT_CLASSID;
12 PetscClassId PETSCMAT_DLLEXPORT MAT_FDCOLORING_CLASSID;
13 
14 PetscLogEvent  MAT_Mult, MAT_Mults, MAT_MultConstrained, MAT_MultAdd, MAT_MultTranspose;
15 PetscLogEvent  MAT_MultTransposeConstrained, MAT_MultTransposeAdd, MAT_Solve, MAT_Solves, MAT_SolveAdd, MAT_SolveTranspose, MAT_MatSolve;
16 PetscLogEvent  MAT_SolveTransposeAdd, MAT_SOR, MAT_ForwardSolve, MAT_BackwardSolve, MAT_LUFactor, MAT_LUFactorSymbolic;
17 PetscLogEvent  MAT_LUFactorNumeric, MAT_CholeskyFactor, MAT_CholeskyFactorSymbolic, MAT_CholeskyFactorNumeric, MAT_ILUFactor;
18 PetscLogEvent  MAT_ILUFactorSymbolic, MAT_ICCFactorSymbolic, MAT_Copy, MAT_Convert, MAT_Scale, MAT_AssemblyBegin;
19 PetscLogEvent  MAT_AssemblyEnd, MAT_SetValues, MAT_GetValues, MAT_GetRow, MAT_GetRowIJ, MAT_GetSubMatrices, MAT_GetColoring, MAT_GetOrdering, MAT_GetRedundantMatrix, MAT_GetSeqNonzeroStructure;
20 PetscLogEvent  MAT_IncreaseOverlap, MAT_Partitioning, MAT_ZeroEntries, MAT_Load, MAT_View, MAT_AXPY, MAT_FDColoringCreate;
21 PetscLogEvent  MAT_FDColoringApply,MAT_Transpose,MAT_FDColoringFunction;
22 PetscLogEvent  MAT_MatMult, MAT_MatMultSymbolic, MAT_MatMultNumeric;
23 PetscLogEvent  MAT_PtAP, MAT_PtAPSymbolic, MAT_PtAPNumeric;
24 PetscLogEvent  MAT_MatMultTranspose, MAT_MatMultTransposeSymbolic, MAT_MatMultTransposeNumeric;
25 PetscLogEvent  MAT_MultHermitianTranspose,MAT_MultHermitianTransposeAdd;
26 PetscLogEvent  MAT_Getsymtranspose, MAT_Getsymtransreduced, MAT_Transpose_SeqAIJ, MAT_GetBrowsOfAcols;
27 PetscLogEvent  MAT_GetBrowsOfAocols, MAT_Getlocalmat, MAT_Getlocalmatcondensed, MAT_Seqstompi, MAT_Seqstompinum, MAT_Seqstompisym;
28 PetscLogEvent  MAT_Applypapt, MAT_Applypapt_numeric, MAT_Applypapt_symbolic, MAT_GetSequentialNonzeroStructure;
29 
30 /* nasty global values for MatSetValue() */
31 PetscInt    PETSCMAT_DLLEXPORT MatSetValue_Row = 0;
32 PetscInt    PETSCMAT_DLLEXPORT MatSetValue_Column = 0;
33 PetscScalar PETSCMAT_DLLEXPORT MatSetValue_Value = 0.0;
34 
35 const char *const MatFactorTypes[] = {"NONE","LU","CHOLESKY","ILU","ICC","ILUDT","MatFactorType","MAT_FACTOR_",0};
36 
37 #undef __FUNCT__
38 #define __FUNCT__ "MatGetDiagonalBlock"
39 /*@
40    MatGetDiagonalBlock - Returns the part of the matrix associated with the on-process coupling
41 
42    Not Collective
43 
44    Input Parameters:
45 +  mat - the matrix
46 -  reuse - indicates you are passing in the a matrix and want it reused
47 
48    Output Parameters:
49 +   iscopy - indicates a copy of the diagonal matrix was created and you should use MatDestroy() on it
50 -   a - the diagonal part (which is a SEQUENTIAL matrix)
51 
52    Notes: see the manual page for MatCreateMPIAIJ() for more information on the "diagonal part" of the matrix
53 
54    Level: advanced
55 
56 @*/
57 PetscErrorCode PETSCMAT_DLLEXPORT MatGetDiagonalBlock(Mat A,PetscTruth *iscopy,MatReuse reuse,Mat *a)
58 {
59   PetscErrorCode ierr,(*f)(Mat,PetscTruth*,MatReuse,Mat*);
60   PetscMPIInt    size;
61 
62   PetscFunctionBegin;
63   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
64   PetscValidType(A,1);
65   PetscValidPointer(iscopy,2);
66   PetscValidPointer(a,3);
67   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
68   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
69   ierr = MPI_Comm_size(((PetscObject)A)->comm,&size);CHKERRQ(ierr);
70   ierr = PetscObjectQueryFunction((PetscObject)A,"MatGetDiagonalBlock_C",(void (**)(void))&f);CHKERRQ(ierr);
71   if (f) {
72     ierr = (*f)(A,iscopy,reuse,a);CHKERRQ(ierr);
73   } else if (size == 1) {
74     *a = A;
75     *iscopy = PETSC_FALSE;
76   } else SETERRQ(((PetscObject)A)->comm,PETSC_ERR_SUP,"Cannot get diagonal part for this matrix");
77   PetscFunctionReturn(0);
78 }
79 
80 #undef __FUNCT__
81 #define __FUNCT__ "MatGetTrace"
82 /*@
83    MatGetTrace - Gets the trace of a matrix. The sum of the diagonal entries.
84 
85    Collective on Mat
86 
87    Input Parameters:
88 .  mat - the matrix
89 
90    Output Parameter:
91 .   trace - the sum of the diagonal entries
92 
93    Level: advanced
94 
95 @*/
96 PetscErrorCode  MatGetTrace(Mat mat,PetscScalar *trace)
97 {
98    PetscErrorCode ierr;
99    Vec            diag;
100 
101    PetscFunctionBegin;
102    ierr = MatGetVecs(mat,&diag,PETSC_NULL);CHKERRQ(ierr);
103    ierr = MatGetDiagonal(mat,diag);CHKERRQ(ierr);
104    ierr = VecSum(diag,trace);CHKERRQ(ierr);
105    ierr = VecDestroy(diag);CHKERRQ(ierr);
106    PetscFunctionReturn(0);
107 }
108 
109 #undef __FUNCT__
110 #define __FUNCT__ "MatRealPart"
111 /*@
112    MatRealPart - Zeros out the imaginary part of the matrix
113 
114    Logically Collective on Mat
115 
116    Input Parameters:
117 .  mat - the matrix
118 
119    Level: advanced
120 
121 
122 .seealso: MatImaginaryPart()
123 @*/
124 PetscErrorCode PETSCMAT_DLLEXPORT MatRealPart(Mat mat)
125 {
126   PetscErrorCode ierr;
127 
128   PetscFunctionBegin;
129   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
130   PetscValidType(mat,1);
131   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
132   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
133   if (!mat->ops->realpart) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
134   ierr = MatPreallocated(mat);CHKERRQ(ierr);
135   ierr = (*mat->ops->realpart)(mat);CHKERRQ(ierr);
136   PetscFunctionReturn(0);
137 }
138 
139 #undef __FUNCT__
140 #define __FUNCT__ "MatGetGhosts"
141 /*@C
142    MatGetGhosts - Get the global index of all ghost nodes defined by the sparse matrix
143 
144    Collective on Mat
145 
146    Input Parameter:
147 .  mat - the matrix
148 
149    Output Parameters:
150 +   nghosts - number of ghosts (note for BAIJ matrices there is one ghost for each block)
151 -   ghosts - the global indices of the ghost points
152 
153    Notes: the nghosts and ghosts are suitable to pass into VecCreateGhost()
154 
155    Level: advanced
156 
157 @*/
158 PetscErrorCode PETSCMAT_DLLEXPORT MatGetGhosts(Mat mat,PetscInt *nghosts,const PetscInt *ghosts[])
159 {
160   PetscErrorCode ierr;
161 
162   PetscFunctionBegin;
163   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
164   PetscValidType(mat,1);
165   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
166   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
167   if (!mat->ops->getghosts) {
168     if (nghosts) *nghosts = 0;
169     if (ghosts) *ghosts = 0;
170   } else {
171     ierr = (*mat->ops->getghosts)(mat,nghosts,ghosts);CHKERRQ(ierr);
172   }
173   PetscFunctionReturn(0);
174 }
175 
176 
177 #undef __FUNCT__
178 #define __FUNCT__ "MatImaginaryPart"
179 /*@
180    MatImaginaryPart - Moves the imaginary part of the matrix to the real part and zeros the imaginary part
181 
182    Logically Collective on Mat
183 
184    Input Parameters:
185 .  mat - the matrix
186 
187    Level: advanced
188 
189 
190 .seealso: MatRealPart()
191 @*/
192 PetscErrorCode PETSCMAT_DLLEXPORT MatImaginaryPart(Mat mat)
193 {
194   PetscErrorCode ierr;
195 
196   PetscFunctionBegin;
197   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
198   PetscValidType(mat,1);
199   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
200   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
201   if (!mat->ops->imaginarypart) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
202   ierr = MatPreallocated(mat);CHKERRQ(ierr);
203   ierr = (*mat->ops->imaginarypart)(mat);CHKERRQ(ierr);
204   PetscFunctionReturn(0);
205 }
206 
207 #undef __FUNCT__
208 #define __FUNCT__ "MatMissingDiagonal"
209 /*@
210    MatMissingDiagonal - Determine if sparse matrix is missing a diagonal entry (or block entry for BAIJ matrices)
211 
212    Collective on Mat
213 
214    Input Parameter:
215 .  mat - the matrix
216 
217    Output Parameters:
218 +  missing - is any diagonal missing
219 -  dd - first diagonal entry that is missing (optional)
220 
221    Level: advanced
222 
223 
224 .seealso: MatRealPart()
225 @*/
226 PetscErrorCode PETSCMAT_DLLEXPORT MatMissingDiagonal(Mat mat,PetscTruth *missing,PetscInt *dd)
227 {
228   PetscErrorCode ierr;
229 
230   PetscFunctionBegin;
231   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
232   PetscValidType(mat,1);
233   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
234   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
235   if (!mat->ops->missingdiagonal) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
236   ierr = (*mat->ops->missingdiagonal)(mat,missing,dd);CHKERRQ(ierr);
237   PetscFunctionReturn(0);
238 }
239 
240 #undef __FUNCT__
241 #define __FUNCT__ "MatGetRow"
242 /*@C
243    MatGetRow - Gets a row of a matrix.  You MUST call MatRestoreRow()
244    for each row that you get to ensure that your application does
245    not bleed memory.
246 
247    Not Collective
248 
249    Input Parameters:
250 +  mat - the matrix
251 -  row - the row to get
252 
253    Output Parameters:
254 +  ncols -  if not NULL, the number of nonzeros in the row
255 .  cols - if not NULL, the column numbers
256 -  vals - if not NULL, the values
257 
258    Notes:
259    This routine is provided for people who need to have direct access
260    to the structure of a matrix.  We hope that we provide enough
261    high-level matrix routines that few users will need it.
262 
263    MatGetRow() always returns 0-based column indices, regardless of
264    whether the internal representation is 0-based (default) or 1-based.
265 
266    For better efficiency, set cols and/or vals to PETSC_NULL if you do
267    not wish to extract these quantities.
268 
269    The user can only examine the values extracted with MatGetRow();
270    the values cannot be altered.  To change the matrix entries, one
271    must use MatSetValues().
272 
273    You can only have one call to MatGetRow() outstanding for a particular
274    matrix at a time, per processor. MatGetRow() can only obtain rows
275    associated with the given processor, it cannot get rows from the
276    other processors; for that we suggest using MatGetSubMatrices(), then
277    MatGetRow() on the submatrix. The row indix passed to MatGetRows()
278    is in the global number of rows.
279 
280    Fortran Notes:
281    The calling sequence from Fortran is
282 .vb
283    MatGetRow(matrix,row,ncols,cols,values,ierr)
284          Mat     matrix (input)
285          integer row    (input)
286          integer ncols  (output)
287          integer cols(maxcols) (output)
288          double precision (or double complex) values(maxcols) output
289 .ve
290    where maxcols >= maximum nonzeros in any row of the matrix.
291 
292 
293    Caution:
294    Do not try to change the contents of the output arrays (cols and vals).
295    In some cases, this may corrupt the matrix.
296 
297    Level: advanced
298 
299    Concepts: matrices^row access
300 
301 .seealso: MatRestoreRow(), MatSetValues(), MatGetValues(), MatGetSubMatrices(), MatGetDiagonal()
302 @*/
303 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
304 {
305   PetscErrorCode ierr;
306   PetscInt       incols;
307 
308   PetscFunctionBegin;
309   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
310   PetscValidType(mat,1);
311   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
312   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
313   if (!mat->ops->getrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
314   ierr = MatPreallocated(mat);CHKERRQ(ierr);
315   ierr = PetscLogEventBegin(MAT_GetRow,mat,0,0,0);CHKERRQ(ierr);
316   ierr = (*mat->ops->getrow)(mat,row,&incols,(PetscInt **)cols,(PetscScalar **)vals);CHKERRQ(ierr);
317   if (ncols) *ncols = incols;
318   ierr = PetscLogEventEnd(MAT_GetRow,mat,0,0,0);CHKERRQ(ierr);
319   PetscFunctionReturn(0);
320 }
321 
322 #undef __FUNCT__
323 #define __FUNCT__ "MatConjugate"
324 /*@
325    MatConjugate - replaces the matrix values with their complex conjugates
326 
327    Logically Collective on Mat
328 
329    Input Parameters:
330 .  mat - the matrix
331 
332    Level: advanced
333 
334 .seealso:  VecConjugate()
335 @*/
336 PetscErrorCode PETSCMAT_DLLEXPORT MatConjugate(Mat mat)
337 {
338   PetscErrorCode ierr;
339 
340   PetscFunctionBegin;
341   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
342   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
343   if (!mat->ops->conjugate) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Not provided for this matrix format, send email to petsc-maint@mcs.anl.gov");
344   ierr = (*mat->ops->conjugate)(mat);CHKERRQ(ierr);
345   PetscFunctionReturn(0);
346 }
347 
348 #undef __FUNCT__
349 #define __FUNCT__ "MatRestoreRow"
350 /*@C
351    MatRestoreRow - Frees any temporary space allocated by MatGetRow().
352 
353    Not Collective
354 
355    Input Parameters:
356 +  mat - the matrix
357 .  row - the row to get
358 .  ncols, cols - the number of nonzeros and their columns
359 -  vals - if nonzero the column values
360 
361    Notes:
362    This routine should be called after you have finished examining the entries.
363 
364    Fortran Notes:
365    The calling sequence from Fortran is
366 .vb
367    MatRestoreRow(matrix,row,ncols,cols,values,ierr)
368       Mat     matrix (input)
369       integer row    (input)
370       integer ncols  (output)
371       integer cols(maxcols) (output)
372       double precision (or double complex) values(maxcols) output
373 .ve
374    Where maxcols >= maximum nonzeros in any row of the matrix.
375 
376    In Fortran MatRestoreRow() MUST be called after MatGetRow()
377    before another call to MatGetRow() can be made.
378 
379    Level: advanced
380 
381 .seealso:  MatGetRow()
382 @*/
383 PetscErrorCode PETSCMAT_DLLEXPORT MatRestoreRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
384 {
385   PetscErrorCode ierr;
386 
387   PetscFunctionBegin;
388   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
389   PetscValidIntPointer(ncols,3);
390   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
391   if (!mat->ops->restorerow) PetscFunctionReturn(0);
392   ierr = (*mat->ops->restorerow)(mat,row,ncols,(PetscInt **)cols,(PetscScalar **)vals);CHKERRQ(ierr);
393   PetscFunctionReturn(0);
394 }
395 
396 #undef __FUNCT__
397 #define __FUNCT__ "MatGetRowUpperTriangular"
398 /*@
399    MatGetRowUpperTriangular - Sets a flag to enable calls to MatGetRow() for matrix in MATSBAIJ format.
400    You should call MatRestoreRowUpperTriangular() after calling MatGetRow/MatRestoreRow() to disable the flag.
401 
402    Not Collective
403 
404    Input Parameters:
405 +  mat - the matrix
406 
407    Notes:
408    The flag is to ensure that users are aware of MatGetRow() only provides the upper trianglular part of the row for the matrices in MATSBAIJ format.
409 
410    Level: advanced
411 
412    Concepts: matrices^row access
413 
414 .seealso: MatRestoreRowRowUpperTriangular()
415 @*/
416 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowUpperTriangular(Mat mat)
417 {
418   PetscErrorCode ierr;
419 
420   PetscFunctionBegin;
421   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
422   PetscValidType(mat,1);
423   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
424   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
425   if (!mat->ops->getrowuppertriangular) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
426   ierr = MatPreallocated(mat);CHKERRQ(ierr);
427   ierr = (*mat->ops->getrowuppertriangular)(mat);CHKERRQ(ierr);
428   PetscFunctionReturn(0);
429 }
430 
431 #undef __FUNCT__
432 #define __FUNCT__ "MatRestoreRowUpperTriangular"
433 /*@
434    MatRestoreRowUpperTriangular - Disable calls to MatGetRow() for matrix in MATSBAIJ format.
435 
436    Not Collective
437 
438    Input Parameters:
439 +  mat - the matrix
440 
441    Notes:
442    This routine should be called after you have finished MatGetRow/MatRestoreRow().
443 
444 
445    Level: advanced
446 
447 .seealso:  MatGetRowUpperTriangular()
448 @*/
449 PetscErrorCode PETSCMAT_DLLEXPORT MatRestoreRowUpperTriangular(Mat mat)
450 {
451   PetscErrorCode ierr;
452 
453   PetscFunctionBegin;
454   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
455   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
456   if (!mat->ops->restorerowuppertriangular) PetscFunctionReturn(0);
457   ierr = (*mat->ops->restorerowuppertriangular)(mat);CHKERRQ(ierr);
458   PetscFunctionReturn(0);
459 }
460 
461 #undef __FUNCT__
462 #define __FUNCT__ "MatSetOptionsPrefix"
463 /*@C
464    MatSetOptionsPrefix - Sets the prefix used for searching for all
465    Mat options in the database.
466 
467    Logically Collective on Mat
468 
469    Input Parameter:
470 +  A - the Mat context
471 -  prefix - the prefix to prepend to all option names
472 
473    Notes:
474    A hyphen (-) must NOT be given at the beginning of the prefix name.
475    The first character of all runtime options is AUTOMATICALLY the hyphen.
476 
477    Level: advanced
478 
479 .keywords: Mat, set, options, prefix, database
480 
481 .seealso: MatSetFromOptions()
482 @*/
483 PetscErrorCode PETSCMAT_DLLEXPORT MatSetOptionsPrefix(Mat A,const char prefix[])
484 {
485   PetscErrorCode ierr;
486 
487   PetscFunctionBegin;
488   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
489   ierr = PetscObjectSetOptionsPrefix((PetscObject)A,prefix);CHKERRQ(ierr);
490   PetscFunctionReturn(0);
491 }
492 
493 #undef __FUNCT__
494 #define __FUNCT__ "MatAppendOptionsPrefix"
495 /*@C
496    MatAppendOptionsPrefix - Appends to the prefix used for searching for all
497    Mat options in the database.
498 
499    Logically Collective on Mat
500 
501    Input Parameters:
502 +  A - the Mat context
503 -  prefix - the prefix to prepend to all option names
504 
505    Notes:
506    A hyphen (-) must NOT be given at the beginning of the prefix name.
507    The first character of all runtime options is AUTOMATICALLY the hyphen.
508 
509    Level: advanced
510 
511 .keywords: Mat, append, options, prefix, database
512 
513 .seealso: MatGetOptionsPrefix()
514 @*/
515 PetscErrorCode PETSCMAT_DLLEXPORT MatAppendOptionsPrefix(Mat A,const char prefix[])
516 {
517   PetscErrorCode ierr;
518 
519   PetscFunctionBegin;
520   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
521   ierr = PetscObjectAppendOptionsPrefix((PetscObject)A,prefix);CHKERRQ(ierr);
522   PetscFunctionReturn(0);
523 }
524 
525 #undef __FUNCT__
526 #define __FUNCT__ "MatGetOptionsPrefix"
527 /*@C
528    MatGetOptionsPrefix - Sets the prefix used for searching for all
529    Mat options in the database.
530 
531    Not Collective
532 
533    Input Parameter:
534 .  A - the Mat context
535 
536    Output Parameter:
537 .  prefix - pointer to the prefix string used
538 
539    Notes: On the fortran side, the user should pass in a string 'prefix' of
540    sufficient length to hold the prefix.
541 
542    Level: advanced
543 
544 .keywords: Mat, get, options, prefix, database
545 
546 .seealso: MatAppendOptionsPrefix()
547 @*/
548 PetscErrorCode PETSCMAT_DLLEXPORT MatGetOptionsPrefix(Mat A,const char *prefix[])
549 {
550   PetscErrorCode ierr;
551 
552   PetscFunctionBegin;
553   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
554   ierr = PetscObjectGetOptionsPrefix((PetscObject)A,prefix);CHKERRQ(ierr);
555   PetscFunctionReturn(0);
556 }
557 
558 #undef __FUNCT__
559 #define __FUNCT__ "MatSetUp"
560 /*@
561    MatSetUp - Sets up the internal matrix data structures for the later use.
562 
563    Collective on Mat
564 
565    Input Parameters:
566 .  A - the Mat context
567 
568    Notes:
569    For basic use of the Mat classes the user need not explicitly call
570    MatSetUp(), since these actions will happen automatically.
571 
572    Level: advanced
573 
574 .keywords: Mat, setup
575 
576 .seealso: MatCreate(), MatDestroy()
577 @*/
578 PetscErrorCode PETSCMAT_DLLEXPORT MatSetUp(Mat A)
579 {
580   PetscMPIInt    size;
581   PetscErrorCode ierr;
582 
583   PetscFunctionBegin;
584   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
585   if (!((PetscObject)A)->type_name) {
586     ierr = MPI_Comm_size(((PetscObject)A)->comm, &size);CHKERRQ(ierr);
587     if (size == 1) {
588       ierr = MatSetType(A, MATSEQAIJ);CHKERRQ(ierr);
589     } else {
590       ierr = MatSetType(A, MATMPIAIJ);CHKERRQ(ierr);
591     }
592   }
593   ierr = MatSetUpPreallocation(A);CHKERRQ(ierr);
594   PetscFunctionReturn(0);
595 }
596 
597 #undef __FUNCT__
598 #define __FUNCT__ "MatView"
599 /*@C
600    MatView - Visualizes a matrix object.
601 
602    Collective on Mat
603 
604    Input Parameters:
605 +  mat - the matrix
606 -  viewer - visualization context
607 
608   Notes:
609   The available visualization contexts include
610 +    PETSC_VIEWER_STDOUT_SELF - standard output (default)
611 .    PETSC_VIEWER_STDOUT_WORLD - synchronized standard
612         output where only the first processor opens
613         the file.  All other processors send their
614         data to the first processor to print.
615 -     PETSC_VIEWER_DRAW_WORLD - graphical display of nonzero structure
616 
617    The user can open alternative visualization contexts with
618 +    PetscViewerASCIIOpen() - Outputs matrix to a specified file
619 .    PetscViewerBinaryOpen() - Outputs matrix in binary to a
620          specified file; corresponding input uses MatLoad()
621 .    PetscViewerDrawOpen() - Outputs nonzero matrix structure to
622          an X window display
623 -    PetscViewerSocketOpen() - Outputs matrix to Socket viewer.
624          Currently only the sequential dense and AIJ
625          matrix types support the Socket viewer.
626 
627    The user can call PetscViewerSetFormat() to specify the output
628    format of ASCII printed objects (when using PETSC_VIEWER_STDOUT_SELF,
629    PETSC_VIEWER_STDOUT_WORLD and PetscViewerASCIIOpen).  Available formats include
630 +    PETSC_VIEWER_DEFAULT - default, prints matrix contents
631 .    PETSC_VIEWER_ASCII_MATLAB - prints matrix contents in Matlab format
632 .    PETSC_VIEWER_ASCII_DENSE - prints entire matrix including zeros
633 .    PETSC_VIEWER_ASCII_COMMON - prints matrix contents, using a sparse
634          format common among all matrix types
635 .    PETSC_VIEWER_ASCII_IMPL - prints matrix contents, using an implementation-specific
636          format (which is in many cases the same as the default)
637 .    PETSC_VIEWER_ASCII_INFO - prints basic information about the matrix
638          size and structure (not the matrix entries)
639 .    PETSC_VIEWER_ASCII_INFO_DETAIL - prints more detailed information about
640          the matrix structure
641 
642    Options Database Keys:
643 +  -mat_view_info - Prints info on matrix at conclusion of MatEndAssembly()
644 .  -mat_view_info_detailed - Prints more detailed info
645 .  -mat_view - Prints matrix in ASCII format
646 .  -mat_view_matlab - Prints matrix in Matlab format
647 .  -mat_view_draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
648 .  -display <name> - Sets display name (default is host)
649 .  -draw_pause <sec> - Sets number of seconds to pause after display
650 .  -mat_view_socket - Sends matrix to socket, can be accessed from Matlab (see users manual)
651 .  -viewer_socket_machine <machine>
652 .  -viewer_socket_port <port>
653 .  -mat_view_binary - save matrix to file in binary format
654 -  -viewer_binary_filename <name>
655    Level: beginner
656 
657    Notes: see the manual page for MatLoad() for the exact format of the binary file when the binary
658       viewer is used.
659 
660       See bin/matlab/PetscBinaryRead.m for a Matlab code that can read in the binary file when the binary
661       viewer is used.
662 
663    Concepts: matrices^viewing
664    Concepts: matrices^plotting
665    Concepts: matrices^printing
666 
667 .seealso: PetscViewerSetFormat(), PetscViewerASCIIOpen(), PetscViewerDrawOpen(),
668           PetscViewerSocketOpen(), PetscViewerBinaryOpen(), MatLoad()
669 @*/
670 PetscErrorCode PETSCMAT_DLLEXPORT MatView(Mat mat,PetscViewer viewer)
671 {
672   PetscErrorCode    ierr;
673   PetscInt          rows,cols;
674   PetscTruth        iascii;
675   const MatType     cstr;
676   PetscViewerFormat format;
677 
678   PetscFunctionBegin;
679   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
680   PetscValidType(mat,1);
681   if (!viewer) {
682     ierr = PetscViewerASCIIGetStdout(((PetscObject)mat)->comm,&viewer);CHKERRQ(ierr);
683   }
684   PetscValidHeaderSpecific(viewer,PETSC_VIEWER_CLASSID,2);
685   PetscCheckSameComm(mat,1,viewer,2);
686   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ORDER,"Must call MatAssemblyBegin/End() before viewing matrix");
687   ierr = MatPreallocated(mat);CHKERRQ(ierr);
688 
689   ierr = PetscLogEventBegin(MAT_View,mat,viewer,0,0);CHKERRQ(ierr);
690   ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);CHKERRQ(ierr);
691   if (iascii) {
692     ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr);
693     if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
694       if (((PetscObject)mat)->prefix) {
695         ierr = PetscViewerASCIIPrintf(viewer,"Matrix Object:(%s)\n",((PetscObject)mat)->prefix);CHKERRQ(ierr);
696       } else {
697         ierr = PetscViewerASCIIPrintf(viewer,"Matrix Object:\n");CHKERRQ(ierr);
698       }
699       ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr);
700       ierr = MatGetType(mat,&cstr);CHKERRQ(ierr);
701       ierr = MatGetSize(mat,&rows,&cols);CHKERRQ(ierr);
702       ierr = PetscViewerASCIIPrintf(viewer,"type=%s, rows=%D, cols=%D\n",cstr,rows,cols);CHKERRQ(ierr);
703       if (mat->factortype) {
704         const MatSolverPackage solver;
705         ierr = MatFactorGetSolverPackage(mat,&solver);CHKERRQ(ierr);
706         ierr = PetscViewerASCIIPrintf(viewer,"package used to perform factorization: %s\n",solver);CHKERRQ(ierr);
707       }
708       if (mat->ops->getinfo) {
709         MatInfo info;
710         ierr = MatGetInfo(mat,MAT_GLOBAL_SUM,&info);CHKERRQ(ierr);
711         ierr = PetscViewerASCIIPrintf(viewer,"total: nonzeros=%D, allocated nonzeros=%D\n",(PetscInt)info.nz_used,(PetscInt)info.nz_allocated);CHKERRQ(ierr);
712         ierr = PetscViewerASCIIPrintf(viewer,"total number of mallocs used during MatSetValues calls =%D\n",(PetscInt)info.mallocs);CHKERRQ(ierr);
713       }
714     }
715   }
716   if (mat->ops->view) {
717     ierr = PetscViewerASCIIPushTab(viewer);CHKERRQ(ierr);
718     ierr = (*mat->ops->view)(mat,viewer);CHKERRQ(ierr);
719     ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr);
720   } else if (!iascii) {
721     SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Viewer type %s not supported",((PetscObject)viewer)->type_name);
722   }
723   if (iascii) {
724     ierr = PetscViewerGetFormat(viewer,&format);CHKERRQ(ierr);
725     if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
726       ierr = PetscViewerASCIIPopTab(viewer);CHKERRQ(ierr);
727     }
728   }
729   ierr = PetscLogEventEnd(MAT_View,mat,viewer,0,0);CHKERRQ(ierr);
730   PetscFunctionReturn(0);
731 }
732 
733 #if defined(PETSC_USE_DEBUG)
734 #include "../src/sys/totalview/tv_data_display.h"
735 PETSC_UNUSED static int TV_display_type(const struct _p_Mat *mat)
736 {
737   TV_add_row("Local rows", "int", &mat->rmap->n);
738   TV_add_row("Local columns", "int", &mat->cmap->n);
739   TV_add_row("Global rows", "int", &mat->rmap->N);
740   TV_add_row("Global columns", "int", &mat->cmap->N);
741   TV_add_row("Typename", TV_ascii_string_type, ((PetscObject)mat)->type_name);
742   return TV_format_OK;
743 }
744 #endif
745 
746 #undef __FUNCT__
747 #define __FUNCT__ "MatLoad"
748 /*@C
749    MatLoad - Loads a matrix that has been stored in binary format
750    with MatView().  The matrix format is determined from the options database.
751    Generates a parallel MPI matrix if the communicator has more than one
752    processor.  The default matrix type is AIJ.
753 
754    Collective on PetscViewer
755 
756    Input Parameters:
757 +  viewer - binary file viewer, created with PetscViewerBinaryOpen()
758 -  outtype - type of matrix desired, for example MATSEQAIJ,
759               MATMPISBAIJ etc.  See types in petsc/include/petscmat.h.
760 
761    Output Parameters:
762 .  newmat - new matrix
763 
764    Basic Options Database Keys:
765 +    -mat_type seqaij   - AIJ type
766 .    -mat_type mpiaij   - parallel AIJ type
767 .    -mat_type seqbaij  - block AIJ type
768 .    -mat_type mpibaij  - parallel block AIJ type
769 .    -mat_type seqsbaij - block symmetric AIJ type
770 .    -mat_type mpisbaij - parallel block symmetric AIJ type
771 .    -mat_type seqdense - dense type
772 .    -mat_type mpidense - parallel dense type
773 .    -mat_type blockmat - sequential blockmat type
774 .    -matload_symmetric - matrix in file is symmetric
775 -    -matload_spd       - matrix in file is symmetric positive definite
776 
777    More Options Database Keys:
778    Used with block matrix formats (MATSEQBAIJ,  ...) to specify
779    block size
780 .    -matload_block_size <bs>
781 
782    Level: beginner
783 
784    Notes:
785    MatLoad() automatically loads into the options database any options
786    given in the file filename.info where filename is the name of the file
787    that was passed to the PetscViewerBinaryOpen(). The options in the info
788    file will be ignored if you use the -viewer_binary_skip_info option.
789 
790    If the type or size of newmat is not set before a call to MatLoad, PETSc
791    sets the default matrix type AIJ and sets the local and global sizes.
792    If type and/or size is already set, then the same are used.
793 
794    In parallel, each processor can load a subset of rows (or the
795    entire matrix).  This routine is especially useful when a large
796    matrix is stored on disk and only part of it is desired on each
797    processor.  For example, a parallel solver may access only some of
798    the rows from each processor.  The algorithm used here reads
799    relatively small blocks of data rather than reading the entire
800    matrix and then subsetting it.
801 
802    Notes for advanced users:
803    Most users should not need to know the details of the binary storage
804    format, since MatLoad() and MatView() completely hide these details.
805    But for anyone who's interested, the standard binary matrix storage
806    format is
807 
808 $    int    MAT_FILE_CLASSID
809 $    int    number of rows
810 $    int    number of columns
811 $    int    total number of nonzeros
812 $    int    *number nonzeros in each row
813 $    int    *column indices of all nonzeros (starting index is zero)
814 $    PetscScalar *values of all nonzeros
815 
816    PETSc automatically does the byte swapping for
817 machines that store the bytes reversed, e.g.  DEC alpha, freebsd,
818 linux, Windows and the paragon; thus if you write your own binary
819 read/write routines you have to swap the bytes; see PetscBinaryRead()
820 and PetscBinaryWrite() to see how this may be done.
821 
822 .keywords: matrix, load, binary, input
823 
824 .seealso: PetscViewerBinaryOpen(), MatView(), VecLoad()
825 
826  @*/
827 PetscErrorCode PETSCMAT_DLLEXPORT MatLoad(PetscViewer viewer, Mat newmat)
828 {
829   PetscErrorCode ierr;
830   PetscTruth     isbinary,flg;
831   const MatType  outtype=0;
832   const char     *prefix;
833 
834   PetscFunctionBegin;
835   PetscValidHeaderSpecific(viewer,PETSC_VIEWER_CLASSID,1);
836   PetscValidPointer(newmat,3);
837   ierr = PetscTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&isbinary);CHKERRQ(ierr);
838   if (!isbinary) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Invalid viewer; open viewer with PetscViewerBinaryOpen()");
839 
840 
841   if (((PetscObject)newmat)->type_name) outtype = ((PetscObject)newmat)->type_name;
842   if (!outtype) {
843     ierr = MatSetFromOptions(newmat);CHKERRQ(ierr);
844   }
845 
846   if (!newmat->ops->load) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatLoad is not supported for type: %s",outtype);
847 
848   ierr = PetscLogEventBegin(MAT_Load,viewer,0,0,0);CHKERRQ(ierr);
849   ierr = (*newmat->ops->load)(viewer,newmat);CHKERRQ(ierr);
850   ierr = PetscLogEventEnd(MAT_Load,viewer,0,0,0);CHKERRQ(ierr);
851 
852   flg  = PETSC_FALSE;
853   ierr = PetscOptionsGetTruth(prefix,"-matload_symmetric",&flg,PETSC_NULL);CHKERRQ(ierr);
854   if (flg) {
855     ierr = MatSetOption(newmat,MAT_SYMMETRIC,PETSC_TRUE);CHKERRQ(ierr);
856     ierr = MatSetOption(newmat,MAT_SYMMETRY_ETERNAL,PETSC_TRUE);CHKERRQ(ierr);
857   }
858   flg  = PETSC_FALSE;
859   ierr = PetscOptionsGetTruth(prefix,"-matload_spd",&flg,PETSC_NULL);CHKERRQ(ierr);
860   if (flg) {
861     ierr = MatSetOption(newmat,MAT_SPD,PETSC_TRUE);CHKERRQ(ierr);
862   }
863   PetscFunctionReturn(0);
864 }
865 
866 #undef __FUNCT__
867 #define __FUNCT__ "MatScaleSystem"
868 /*@
869    MatScaleSystem - Scale a vector solution and right hand side to
870    match the scaling of a scaled matrix.
871 
872    Collective on Mat
873 
874    Input Parameter:
875 +  mat - the matrix
876 .  b - right hand side vector (or PETSC_NULL)
877 -  x - solution vector (or PETSC_NULL)
878 
879 
880    Notes:
881    For AIJ, and BAIJ matrix formats, the matrices are not
882    internally scaled, so this does nothing.
883 
884    The KSP methods automatically call this routine when required
885    (via PCPreSolve()) so it is rarely used directly.
886 
887    Level: Developer
888 
889    Concepts: matrices^scaling
890 
891 .seealso: MatUseScaledForm(), MatUnScaleSystem()
892 @*/
893 PetscErrorCode PETSCMAT_DLLEXPORT MatScaleSystem(Mat mat,Vec b,Vec x)
894 {
895   PetscErrorCode ierr;
896 
897   PetscFunctionBegin;
898   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
899   PetscValidType(mat,1);
900   ierr = MatPreallocated(mat);CHKERRQ(ierr);
901   if (x) {PetscValidHeaderSpecific(x,VEC_CLASSID,3);PetscCheckSameComm(mat,1,x,3);}
902   if (b) {PetscValidHeaderSpecific(b,VEC_CLASSID,2);PetscCheckSameComm(mat,1,b,2);}
903 
904   if (mat->ops->scalesystem) {
905     ierr = (*mat->ops->scalesystem)(mat,b,x);CHKERRQ(ierr);
906   }
907   PetscFunctionReturn(0);
908 }
909 
910 #undef __FUNCT__
911 #define __FUNCT__ "MatUnScaleSystem"
912 /*@
913    MatUnScaleSystem - Unscales a vector solution and right hand side to
914    match the original scaling of a scaled matrix.
915 
916    Collective on Mat
917 
918    Input Parameter:
919 +  mat - the matrix
920 .  b - right hand side vector (or PETSC_NULL)
921 -  x - solution vector (or PETSC_NULL)
922 
923 
924    Notes:
925    For AIJ and BAIJ matrix formats, the matrices are not
926    internally scaled, so this does nothing.
927 
928    The KSP methods automatically call this routine when required
929    (via PCPreSolve()) so it is rarely used directly.
930 
931    Level: Developer
932 
933 .seealso: MatUseScaledForm(), MatScaleSystem()
934 @*/
935 PetscErrorCode PETSCMAT_DLLEXPORT MatUnScaleSystem(Mat mat,Vec b,Vec x)
936 {
937   PetscErrorCode ierr;
938 
939   PetscFunctionBegin;
940   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
941   PetscValidType(mat,1);
942   ierr = MatPreallocated(mat);CHKERRQ(ierr);
943   if (x) {PetscValidHeaderSpecific(x,VEC_CLASSID,3);PetscCheckSameComm(mat,1,x,3);}
944   if (b) {PetscValidHeaderSpecific(b,VEC_CLASSID,2);PetscCheckSameComm(mat,1,b,2);}
945   if (mat->ops->unscalesystem) {
946     ierr = (*mat->ops->unscalesystem)(mat,b,x);CHKERRQ(ierr);
947   }
948   PetscFunctionReturn(0);
949 }
950 
951 #undef __FUNCT__
952 #define __FUNCT__ "MatUseScaledForm"
953 /*@
954    MatUseScaledForm - For matrix storage formats that scale the
955    matrix indicates matrix operations (MatMult() etc) are
956    applied using the scaled matrix.
957 
958    Logically Collective on Mat
959 
960    Input Parameter:
961 +  mat - the matrix
962 -  scaled - PETSC_TRUE for applying the scaled, PETSC_FALSE for
963             applying the original matrix
964 
965    Notes:
966    For scaled matrix formats, applying the original, unscaled matrix
967    will be slightly more expensive
968 
969    Level: Developer
970 
971 .seealso: MatScaleSystem(), MatUnScaleSystem()
972 @*/
973 PetscErrorCode PETSCMAT_DLLEXPORT MatUseScaledForm(Mat mat,PetscTruth scaled)
974 {
975   PetscErrorCode ierr;
976 
977   PetscFunctionBegin;
978   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
979   PetscValidType(mat,1);
980   ierr = MatPreallocated(mat);CHKERRQ(ierr);
981   if (mat->ops->usescaledform) {
982     ierr = (*mat->ops->usescaledform)(mat,scaled);CHKERRQ(ierr);
983   }
984   PetscFunctionReturn(0);
985 }
986 
987 #undef __FUNCT__
988 #define __FUNCT__ "MatDestroy"
989 /*@
990    MatDestroy - Frees space taken by a matrix.
991 
992    Collective on Mat
993 
994    Input Parameter:
995 .  A - the matrix
996 
997    Level: beginner
998 
999 @*/
1000 PetscErrorCode PETSCMAT_DLLEXPORT MatDestroy(Mat A)
1001 {
1002   PetscErrorCode ierr;
1003   PetscFunctionBegin;
1004   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
1005   if (--((PetscObject)A)->refct > 0) PetscFunctionReturn(0);
1006   ierr = MatPreallocated(A);CHKERRQ(ierr);
1007   /* if memory was published with AMS then destroy it */
1008   ierr = PetscObjectDepublish(A);CHKERRQ(ierr);
1009   if (A->ops->destroy) {
1010     ierr = (*A->ops->destroy)(A);CHKERRQ(ierr);
1011   }
1012   if (A->mapping) {
1013     ierr = ISLocalToGlobalMappingDestroy(A->mapping);CHKERRQ(ierr);
1014   }
1015   if (A->bmapping) {
1016     ierr = ISLocalToGlobalMappingDestroy(A->bmapping);CHKERRQ(ierr);
1017   }
1018 
1019   if (A->spptr){ierr = PetscFree(A->spptr);CHKERRQ(ierr);}
1020   ierr = PetscLayoutDestroy(A->rmap);CHKERRQ(ierr);
1021   ierr = PetscLayoutDestroy(A->cmap);CHKERRQ(ierr);
1022   ierr = PetscHeaderDestroy(A);CHKERRQ(ierr);
1023   PetscFunctionReturn(0);
1024 }
1025 
1026 #undef __FUNCT__
1027 #define __FUNCT__ "MatSetValues"
1028 /*@
1029    MatSetValues - Inserts or adds a block of values into a matrix.
1030    These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
1031    MUST be called after all calls to MatSetValues() have been completed.
1032 
1033    Not Collective
1034 
1035    Input Parameters:
1036 +  mat - the matrix
1037 .  v - a logically two-dimensional array of values
1038 .  m, idxm - the number of rows and their global indices
1039 .  n, idxn - the number of columns and their global indices
1040 -  addv - either ADD_VALUES or INSERT_VALUES, where
1041    ADD_VALUES adds values to any existing entries, and
1042    INSERT_VALUES replaces existing entries with new values
1043 
1044    Notes:
1045    By default the values, v, are row-oriented. See MatSetOption() for other options.
1046 
1047    Calls to MatSetValues() with the INSERT_VALUES and ADD_VALUES
1048    options cannot be mixed without intervening calls to the assembly
1049    routines.
1050 
1051    MatSetValues() uses 0-based row and column numbers in Fortran
1052    as well as in C.
1053 
1054    Negative indices may be passed in idxm and idxn, these rows and columns are
1055    simply ignored. This allows easily inserting element stiffness matrices
1056    with homogeneous Dirchlet boundary conditions that you don't want represented
1057    in the matrix.
1058 
1059    Efficiency Alert:
1060    The routine MatSetValuesBlocked() may offer much better efficiency
1061    for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1062 
1063    Level: beginner
1064 
1065    Concepts: matrices^putting entries in
1066 
1067 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1068           InsertMode, INSERT_VALUES, ADD_VALUES
1069 @*/
1070 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1071 {
1072   PetscErrorCode ierr;
1073 
1074   PetscFunctionBegin;
1075   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1076   PetscValidType(mat,1);
1077   if (!m || !n) PetscFunctionReturn(0); /* no values to insert */
1078   PetscValidIntPointer(idxm,3);
1079   PetscValidIntPointer(idxn,5);
1080   if (v) PetscValidDoublePointer(v,6);
1081   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1082   if (mat->insertmode == NOT_SET_VALUES) {
1083     mat->insertmode = addv;
1084   }
1085 #if defined(PETSC_USE_DEBUG)
1086   else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1087   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1088 #endif
1089 
1090   if (mat->assembled) {
1091     mat->was_assembled = PETSC_TRUE;
1092     mat->assembled     = PETSC_FALSE;
1093   }
1094   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1095   if (!mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1096   ierr = (*mat->ops->setvalues)(mat,m,idxm,n,idxn,v,addv);CHKERRQ(ierr);
1097   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1098   PetscFunctionReturn(0);
1099 }
1100 
1101 
1102 #undef __FUNCT__
1103 #define __FUNCT__ "MatSetValuesRowLocal"
1104 /*@
1105    MatSetValuesRowLocal - Inserts a row (block row for BAIJ matrices) of nonzero
1106         values into a matrix
1107 
1108    Not Collective
1109 
1110    Input Parameters:
1111 +  mat - the matrix
1112 .  row - the (block) row to set
1113 -  v - a logically two-dimensional array of values
1114 
1115    Notes:
1116    By the values, v, are column-oriented (for the block version) and sorted
1117 
1118    All the nonzeros in the row must be provided
1119 
1120    The matrix must have previously had its column indices set
1121 
1122    The row must belong to this process
1123 
1124    Level: intermediate
1125 
1126    Concepts: matrices^putting entries in
1127 
1128 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1129           InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues(), MatSetValuesRow(), MatSetLocalToGlobalMapping()
1130 @*/
1131 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesRowLocal(Mat mat,PetscInt row,const PetscScalar v[])
1132 {
1133   PetscErrorCode ierr;
1134 
1135   PetscFunctionBegin;
1136   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1137   PetscValidType(mat,1);
1138   PetscValidScalarPointer(v,2);
1139   ierr = MatSetValuesRow(mat, mat->mapping->indices[row],v);CHKERRQ(ierr);
1140   PetscFunctionReturn(0);
1141 }
1142 
1143 #undef __FUNCT__
1144 #define __FUNCT__ "MatSetValuesRow"
1145 /*@
1146    MatSetValuesRow - Inserts a row (block row for BAIJ matrices) of nonzero
1147         values into a matrix
1148 
1149    Not Collective
1150 
1151    Input Parameters:
1152 +  mat - the matrix
1153 .  row - the (block) row to set
1154 -  v - a logically two-dimensional array of values
1155 
1156    Notes:
1157    The values, v, are column-oriented for the block version.
1158 
1159    All the nonzeros in the row must be provided
1160 
1161    THE MATRIX MUSAT HAVE PREVIOUSLY HAD ITS COLUMN INDICES SET. IT IS RARE THAT THIS ROUTINE IS USED, usually MatSetValues() is used.
1162 
1163    The row must belong to this process
1164 
1165    Level: advanced
1166 
1167    Concepts: matrices^putting entries in
1168 
1169 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1170           InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues()
1171 @*/
1172 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesRow(Mat mat,PetscInt row,const PetscScalar v[])
1173 {
1174   PetscErrorCode ierr;
1175 
1176   PetscFunctionBegin;
1177   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1178   PetscValidType(mat,1);
1179   PetscValidScalarPointer(v,2);
1180 #if defined(PETSC_USE_DEBUG)
1181   if (mat->insertmode == ADD_VALUES) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add and insert values");
1182   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1183 #endif
1184   mat->insertmode = INSERT_VALUES;
1185 
1186   if (mat->assembled) {
1187     mat->was_assembled = PETSC_TRUE;
1188     mat->assembled     = PETSC_FALSE;
1189   }
1190   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1191   if (!mat->ops->setvaluesrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1192   ierr = (*mat->ops->setvaluesrow)(mat,row,v);CHKERRQ(ierr);
1193   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1194   PetscFunctionReturn(0);
1195 }
1196 
1197 #undef __FUNCT__
1198 #define __FUNCT__ "MatSetValuesStencil"
1199 /*@
1200    MatSetValuesStencil - Inserts or adds a block of values into a matrix.
1201      Using structured grid indexing
1202 
1203    Not Collective
1204 
1205    Input Parameters:
1206 +  mat - the matrix
1207 .  m - number of rows being entered
1208 .  idxm - grid coordinates (and component number when dof > 1) for matrix rows being entered
1209 .  n - number of columns being entered
1210 .  idxn - grid coordinates (and component number when dof > 1) for matrix columns being entered
1211 .  v - a logically two-dimensional array of values
1212 -  addv - either ADD_VALUES or INSERT_VALUES, where
1213    ADD_VALUES adds values to any existing entries, and
1214    INSERT_VALUES replaces existing entries with new values
1215 
1216    Notes:
1217    By default the values, v, are row-oriented.  See MatSetOption() for other options.
1218 
1219    Calls to MatSetValuesStencil() with the INSERT_VALUES and ADD_VALUES
1220    options cannot be mixed without intervening calls to the assembly
1221    routines.
1222 
1223    The grid coordinates are across the entire grid, not just the local portion
1224 
1225    MatSetValuesStencil() uses 0-based row and column numbers in Fortran
1226    as well as in C.
1227 
1228    For setting/accessing vector values via array coordinates you can use the DAVecGetArray() routine
1229 
1230    In order to use this routine you must either obtain the matrix with DAGetMatrix()
1231    or call MatSetLocalToGlobalMapping() and MatSetStencil() first.
1232 
1233    The columns and rows in the stencil passed in MUST be contained within the
1234    ghost region of the given process as set with DACreateXXX() or MatSetStencil(). For example,
1235    if you create a DA with an overlap of one grid level and on a particular process its first
1236    local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1237    first i index you can use in your column and row indices in MatSetStencil() is 5.
1238 
1239    In Fortran idxm and idxn should be declared as
1240 $     MatStencil idxm(4,m),idxn(4,n)
1241    and the values inserted using
1242 $    idxm(MatStencil_i,1) = i
1243 $    idxm(MatStencil_j,1) = j
1244 $    idxm(MatStencil_k,1) = k
1245 $    idxm(MatStencil_c,1) = c
1246    etc
1247 
1248    For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
1249    obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
1250    etc to obtain values that obtained by wrapping the values from the left edge. This does not work for the DA_NONPERIODIC
1251    wrap.
1252 
1253    For indices that don't mean anything for your case (like the k index when working in 2d) or the c index when you have
1254    a single value per point) you can skip filling those indices.
1255 
1256    Inspired by the structured grid interface to the HYPRE package
1257    (http://www.llnl.gov/CASC/hypre)
1258 
1259    Efficiency Alert:
1260    The routine MatSetValuesBlockedStencil() may offer much better efficiency
1261    for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1262 
1263    Level: beginner
1264 
1265    Concepts: matrices^putting entries in
1266 
1267 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1268           MatSetValues(), MatSetValuesBlockedStencil(), MatSetStencil(), DAGetMatrix(), DAVecGetArray(), MatStencil
1269 @*/
1270 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1271 {
1272   PetscErrorCode ierr;
1273   PetscInt       j,i,jdxm[128],jdxn[256],dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1274   PetscInt       *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1275 
1276   PetscFunctionBegin;
1277   if (!m || !n) PetscFunctionReturn(0); /* no values to insert */
1278   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1279   PetscValidType(mat,1);
1280   PetscValidIntPointer(idxm,3);
1281   PetscValidIntPointer(idxn,5);
1282   PetscValidScalarPointer(v,6);
1283 
1284   if (m > 128) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only set 128 rows at a time; trying to set %D",m);
1285   if (n > 256) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only set 256 columns at a time; trying to set %D",n);
1286 
1287   for (i=0; i<m; i++) {
1288     for (j=0; j<3-sdim; j++) dxm++;
1289     tmp = *dxm++ - starts[0];
1290     for (j=0; j<dim-1; j++) {
1291       if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
1292       else                                       tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1293     }
1294     if (mat->stencil.noc) dxm++;
1295     jdxm[i] = tmp;
1296   }
1297   for (i=0; i<n; i++) {
1298     for (j=0; j<3-sdim; j++) dxn++;
1299     tmp = *dxn++ - starts[0];
1300     for (j=0; j<dim-1; j++) {
1301       if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
1302       else                                       tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1303     }
1304     if (mat->stencil.noc) dxn++;
1305     jdxn[i] = tmp;
1306   }
1307   ierr = MatSetValuesLocal(mat,m,jdxm,n,jdxn,v,addv);CHKERRQ(ierr);
1308   PetscFunctionReturn(0);
1309 }
1310 
1311 #undef __FUNCT__
1312 #define __FUNCT__ "MatSetValuesBlockedStencil"
1313 /*@C
1314    MatSetValuesBlockedStencil - Inserts or adds a block of values into a matrix.
1315      Using structured grid indexing
1316 
1317    Not Collective
1318 
1319    Input Parameters:
1320 +  mat - the matrix
1321 .  m - number of rows being entered
1322 .  idxm - grid coordinates for matrix rows being entered
1323 .  n - number of columns being entered
1324 .  idxn - grid coordinates for matrix columns being entered
1325 .  v - a logically two-dimensional array of values
1326 -  addv - either ADD_VALUES or INSERT_VALUES, where
1327    ADD_VALUES adds values to any existing entries, and
1328    INSERT_VALUES replaces existing entries with new values
1329 
1330    Notes:
1331    By default the values, v, are row-oriented and unsorted.
1332    See MatSetOption() for other options.
1333 
1334    Calls to MatSetValuesBlockedStencil() with the INSERT_VALUES and ADD_VALUES
1335    options cannot be mixed without intervening calls to the assembly
1336    routines.
1337 
1338    The grid coordinates are across the entire grid, not just the local portion
1339 
1340    MatSetValuesBlockedStencil() uses 0-based row and column numbers in Fortran
1341    as well as in C.
1342 
1343    For setting/accessing vector values via array coordinates you can use the DAVecGetArray() routine
1344 
1345    In order to use this routine you must either obtain the matrix with DAGetMatrix()
1346    or call MatSetBlockSize(), MatSetLocalToGlobalMapping() and MatSetStencil() first.
1347 
1348    The columns and rows in the stencil passed in MUST be contained within the
1349    ghost region of the given process as set with DACreateXXX() or MatSetStencil(). For example,
1350    if you create a DA with an overlap of one grid level and on a particular process its first
1351    local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1352    first i index you can use in your column and row indices in MatSetStencil() is 5.
1353 
1354    In Fortran idxm and idxn should be declared as
1355 $     MatStencil idxm(4,m),idxn(4,n)
1356    and the values inserted using
1357 $    idxm(MatStencil_i,1) = i
1358 $    idxm(MatStencil_j,1) = j
1359 $    idxm(MatStencil_k,1) = k
1360    etc
1361 
1362    Negative indices may be passed in idxm and idxn, these rows and columns are
1363    simply ignored. This allows easily inserting element stiffness matrices
1364    with homogeneous Dirchlet boundary conditions that you don't want represented
1365    in the matrix.
1366 
1367    Inspired by the structured grid interface to the HYPRE package
1368    (http://www.llnl.gov/CASC/hypre)
1369 
1370    Level: beginner
1371 
1372    Concepts: matrices^putting entries in
1373 
1374 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1375           MatSetValues(), MatSetValuesStencil(), MatSetStencil(), DAGetMatrix(), DAVecGetArray(), MatStencil,
1376           MatSetBlockSize(), MatSetLocalToGlobalMapping()
1377 @*/
1378 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesBlockedStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1379 {
1380   PetscErrorCode ierr;
1381   PetscInt       j,i,jdxm[128],jdxn[256],dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1382   PetscInt       *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1383 
1384   PetscFunctionBegin;
1385   if (!m || !n) PetscFunctionReturn(0); /* no values to insert */
1386   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1387   PetscValidType(mat,1);
1388   PetscValidIntPointer(idxm,3);
1389   PetscValidIntPointer(idxn,5);
1390   PetscValidScalarPointer(v,6);
1391 
1392   if (m > 128) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only set 128 rows at a time; trying to set %D",m);
1393   if (n > 128) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Can only set 256 columns at a time; trying to set %D",n);
1394 
1395   for (i=0; i<m; i++) {
1396     for (j=0; j<3-sdim; j++) dxm++;
1397     tmp = *dxm++ - starts[0];
1398     for (j=0; j<sdim-1; j++) {
1399       if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
1400       else                                      tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1401     }
1402     dxm++;
1403     jdxm[i] = tmp;
1404   }
1405   for (i=0; i<n; i++) {
1406     for (j=0; j<3-sdim; j++) dxn++;
1407     tmp = *dxn++ - starts[0];
1408     for (j=0; j<sdim-1; j++) {
1409       if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
1410       else                                       tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1411     }
1412     dxn++;
1413     jdxn[i] = tmp;
1414   }
1415   ierr = MatSetValuesBlockedLocal(mat,m,jdxm,n,jdxn,v,addv);CHKERRQ(ierr);
1416   PetscFunctionReturn(0);
1417 }
1418 
1419 #undef __FUNCT__
1420 #define __FUNCT__ "MatSetStencil"
1421 /*@
1422    MatSetStencil - Sets the grid information for setting values into a matrix via
1423         MatSetValuesStencil()
1424 
1425    Not Collective
1426 
1427    Input Parameters:
1428 +  mat - the matrix
1429 .  dim - dimension of the grid 1, 2, or 3
1430 .  dims - number of grid points in x, y, and z direction, including ghost points on your processor
1431 .  starts - starting point of ghost nodes on your processor in x, y, and z direction
1432 -  dof - number of degrees of freedom per node
1433 
1434 
1435    Inspired by the structured grid interface to the HYPRE package
1436    (www.llnl.gov/CASC/hyper)
1437 
1438    For matrices generated with DAGetMatrix() this routine is automatically called and so not needed by the
1439    user.
1440 
1441    Level: beginner
1442 
1443    Concepts: matrices^putting entries in
1444 
1445 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1446           MatSetValues(), MatSetValuesBlockedStencil(), MatSetValuesStencil()
1447 @*/
1448 PetscErrorCode PETSCMAT_DLLEXPORT MatSetStencil(Mat mat,PetscInt dim,const PetscInt dims[],const PetscInt starts[],PetscInt dof)
1449 {
1450   PetscInt i;
1451 
1452   PetscFunctionBegin;
1453   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1454   PetscValidIntPointer(dims,3);
1455   PetscValidIntPointer(starts,4);
1456 
1457   mat->stencil.dim = dim + (dof > 1);
1458   for (i=0; i<dim; i++) {
1459     mat->stencil.dims[i]   = dims[dim-i-1];      /* copy the values in backwards */
1460     mat->stencil.starts[i] = starts[dim-i-1];
1461   }
1462   mat->stencil.dims[dim]   = dof;
1463   mat->stencil.starts[dim] = 0;
1464   mat->stencil.noc         = (PetscTruth)(dof == 1);
1465   PetscFunctionReturn(0);
1466 }
1467 
1468 #undef __FUNCT__
1469 #define __FUNCT__ "MatSetValuesBlocked"
1470 /*@
1471    MatSetValuesBlocked - Inserts or adds a block of values into a matrix.
1472 
1473    Not Collective
1474 
1475    Input Parameters:
1476 +  mat - the matrix
1477 .  v - a logically two-dimensional array of values
1478 .  m, idxm - the number of block rows and their global block indices
1479 .  n, idxn - the number of block columns and their global block indices
1480 -  addv - either ADD_VALUES or INSERT_VALUES, where
1481    ADD_VALUES adds values to any existing entries, and
1482    INSERT_VALUES replaces existing entries with new values
1483 
1484    Notes:
1485    The m and n count the NUMBER of blocks in the row direction and column direction,
1486    NOT the total number of rows/columns; for example, if the block size is 2 and
1487    you are passing in values for rows 2,3,4,5  then m would be 2 (not 4).
1488    The values in idxm would be 1 2; that is the first index for each block divided by
1489    the block size.
1490 
1491    Note that you must call MatSetBlockSize() when constructing this matrix (after
1492    preallocating it).
1493 
1494    By default the values, v, are row-oriented, so the layout of
1495    v is the same as for MatSetValues(). See MatSetOption() for other options.
1496 
1497    Calls to MatSetValuesBlocked() with the INSERT_VALUES and ADD_VALUES
1498    options cannot be mixed without intervening calls to the assembly
1499    routines.
1500 
1501    MatSetValuesBlocked() uses 0-based row and column numbers in Fortran
1502    as well as in C.
1503 
1504    Negative indices may be passed in idxm and idxn, these rows and columns are
1505    simply ignored. This allows easily inserting element stiffness matrices
1506    with homogeneous Dirchlet boundary conditions that you don't want represented
1507    in the matrix.
1508 
1509    Each time an entry is set within a sparse matrix via MatSetValues(),
1510    internal searching must be done to determine where to place the the
1511    data in the matrix storage space.  By instead inserting blocks of
1512    entries via MatSetValuesBlocked(), the overhead of matrix assembly is
1513    reduced.
1514 
1515    Example:
1516 $   Suppose m=n=2 and block size(bs) = 2 The array is
1517 $
1518 $   1  2  | 3  4
1519 $   5  6  | 7  8
1520 $   - - - | - - -
1521 $   9  10 | 11 12
1522 $   13 14 | 15 16
1523 $
1524 $   v[] should be passed in like
1525 $   v[] = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16]
1526 $
1527 $  If you are not using row oriented storage of v (that is you called MatSetOption(mat,MAT_ROW_ORIENTED,PETSC_FALSE)) then
1528 $   v[] = [1,5,9,13,2,6,10,14,3,7,11,15,4,8,12,16]
1529 
1530    Level: intermediate
1531 
1532    Concepts: matrices^putting entries in blocked
1533 
1534 .seealso: MatSetBlockSize(), MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesBlockedLocal()
1535 @*/
1536 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesBlocked(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1537 {
1538   PetscErrorCode ierr;
1539 
1540   PetscFunctionBegin;
1541   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1542   PetscValidType(mat,1);
1543   if (!m || !n) PetscFunctionReturn(0); /* no values to insert */
1544   PetscValidIntPointer(idxm,3);
1545   PetscValidIntPointer(idxn,5);
1546   PetscValidScalarPointer(v,6);
1547   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1548   if (mat->insertmode == NOT_SET_VALUES) {
1549     mat->insertmode = addv;
1550   }
1551 #if defined(PETSC_USE_DEBUG)
1552   else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1553   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1554 #endif
1555 
1556   if (mat->assembled) {
1557     mat->was_assembled = PETSC_TRUE;
1558     mat->assembled     = PETSC_FALSE;
1559   }
1560   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1561   if (mat->ops->setvaluesblocked) {
1562     ierr = (*mat->ops->setvaluesblocked)(mat,m,idxm,n,idxn,v,addv);CHKERRQ(ierr);
1563   } else {
1564     PetscInt buf[4096],*ibufm=0,*ibufn=0;
1565     PetscInt i,j,*iidxm,*iidxn,bs=mat->rmap->bs;
1566     if ((m+n)*bs <= 4096) {
1567       iidxm = buf; iidxn = buf + m*bs;
1568     } else {
1569       ierr = PetscMalloc2(m*bs,PetscInt,&ibufm,n*bs,PetscInt,&ibufn);CHKERRQ(ierr);
1570       iidxm = ibufm; iidxn = ibufn;
1571     }
1572     for (i=0; i<m; i++) {
1573       for (j=0; j<bs; j++) {
1574 	iidxm[i*bs+j] = bs*idxm[i] + j;
1575       }
1576     }
1577     for (i=0; i<n; i++) {
1578       for (j=0; j<bs; j++) {
1579 	iidxn[i*bs+j] = bs*idxn[i] + j;
1580       }
1581     }
1582     ierr = MatSetValues(mat,bs*m,iidxm,bs*n,iidxn,v,addv);CHKERRQ(ierr);
1583     ierr = PetscFree2(ibufm,ibufn);CHKERRQ(ierr);
1584   }
1585   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1586   PetscFunctionReturn(0);
1587 }
1588 
1589 #undef __FUNCT__
1590 #define __FUNCT__ "MatGetValues"
1591 /*@
1592    MatGetValues - Gets a block of values from a matrix.
1593 
1594    Not Collective; currently only returns a local block
1595 
1596    Input Parameters:
1597 +  mat - the matrix
1598 .  v - a logically two-dimensional array for storing the values
1599 .  m, idxm - the number of rows and their global indices
1600 -  n, idxn - the number of columns and their global indices
1601 
1602    Notes:
1603    The user must allocate space (m*n PetscScalars) for the values, v.
1604    The values, v, are then returned in a row-oriented format,
1605    analogous to that used by default in MatSetValues().
1606 
1607    MatGetValues() uses 0-based row and column numbers in
1608    Fortran as well as in C.
1609 
1610    MatGetValues() requires that the matrix has been assembled
1611    with MatAssemblyBegin()/MatAssemblyEnd().  Thus, calls to
1612    MatSetValues() and MatGetValues() CANNOT be made in succession
1613    without intermediate matrix assembly.
1614 
1615    Negative row or column indices will be ignored and those locations in v[] will be
1616    left unchanged.
1617 
1618    Level: advanced
1619 
1620    Concepts: matrices^accessing values
1621 
1622 .seealso: MatGetRow(), MatGetSubMatrices(), MatSetValues()
1623 @*/
1624 PetscErrorCode PETSCMAT_DLLEXPORT MatGetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[])
1625 {
1626   PetscErrorCode ierr;
1627 
1628   PetscFunctionBegin;
1629   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1630   PetscValidType(mat,1);
1631   if (!m || !n) PetscFunctionReturn(0);
1632   PetscValidIntPointer(idxm,3);
1633   PetscValidIntPointer(idxn,5);
1634   PetscValidScalarPointer(v,6);
1635   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1636   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1637   if (!mat->ops->getvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1638   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1639 
1640   ierr = PetscLogEventBegin(MAT_GetValues,mat,0,0,0);CHKERRQ(ierr);
1641   ierr = (*mat->ops->getvalues)(mat,m,idxm,n,idxn,v);CHKERRQ(ierr);
1642   ierr = PetscLogEventEnd(MAT_GetValues,mat,0,0,0);CHKERRQ(ierr);
1643   PetscFunctionReturn(0);
1644 }
1645 
1646 #undef __FUNCT__
1647 #define __FUNCT__ "MatSetLocalToGlobalMapping"
1648 /*@
1649    MatSetLocalToGlobalMapping - Sets a local-to-global numbering for use by
1650    the routine MatSetValuesLocal() to allow users to insert matrix entries
1651    using a local (per-processor) numbering.
1652 
1653    Not Collective
1654 
1655    Input Parameters:
1656 +  x - the matrix
1657 -  mapping - mapping created with ISLocalToGlobalMappingCreate()
1658              or ISLocalToGlobalMappingCreateIS()
1659 
1660    Level: intermediate
1661 
1662    Concepts: matrices^local to global mapping
1663    Concepts: local to global mapping^for matrices
1664 
1665 .seealso:  MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesLocal()
1666 @*/
1667 PetscErrorCode PETSCMAT_DLLEXPORT MatSetLocalToGlobalMapping(Mat x,ISLocalToGlobalMapping mapping)
1668 {
1669   PetscErrorCode ierr;
1670   PetscFunctionBegin;
1671   PetscValidHeaderSpecific(x,MAT_CLASSID,1);
1672   PetscValidType(x,1);
1673   PetscValidHeaderSpecific(mapping,IS_LTOGM_CLASSID,2);
1674   if (x->mapping) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Mapping already set for matrix");
1675   ierr = MatPreallocated(x);CHKERRQ(ierr);
1676 
1677   if (x->ops->setlocaltoglobalmapping) {
1678     ierr = (*x->ops->setlocaltoglobalmapping)(x,mapping);CHKERRQ(ierr);
1679   } else {
1680     ierr = PetscObjectReference((PetscObject)mapping);CHKERRQ(ierr);
1681     if (x->mapping) { ierr = ISLocalToGlobalMappingDestroy(x->mapping);CHKERRQ(ierr); }
1682     x->mapping = mapping;
1683   }
1684   PetscFunctionReturn(0);
1685 }
1686 
1687 #undef __FUNCT__
1688 #define __FUNCT__ "MatSetLocalToGlobalMappingBlock"
1689 /*@
1690    MatSetLocalToGlobalMappingBlock - Sets a local-to-global numbering for use
1691    by the routine MatSetValuesBlockedLocal() to allow users to insert matrix
1692    entries using a local (per-processor) numbering.
1693 
1694    Not Collective
1695 
1696    Input Parameters:
1697 +  x - the matrix
1698 -  mapping - mapping created with ISLocalToGlobalMappingCreate() or
1699              ISLocalToGlobalMappingCreateIS()
1700 
1701    Level: intermediate
1702 
1703    Concepts: matrices^local to global mapping blocked
1704    Concepts: local to global mapping^for matrices, blocked
1705 
1706 .seealso:  MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesBlockedLocal(),
1707            MatSetValuesBlocked(), MatSetValuesLocal()
1708 @*/
1709 PetscErrorCode PETSCMAT_DLLEXPORT MatSetLocalToGlobalMappingBlock(Mat x,ISLocalToGlobalMapping mapping)
1710 {
1711   PetscErrorCode ierr;
1712   PetscFunctionBegin;
1713   PetscValidHeaderSpecific(x,MAT_CLASSID,1);
1714   PetscValidType(x,1);
1715   PetscValidHeaderSpecific(mapping,IS_LTOGM_CLASSID,2);
1716   if (x->bmapping) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Mapping already set for matrix");
1717   ierr = PetscObjectReference((PetscObject)mapping);CHKERRQ(ierr);
1718   if (x->bmapping) { ierr = ISLocalToGlobalMappingDestroy(x->bmapping);CHKERRQ(ierr); }
1719   x->bmapping = mapping;
1720   PetscFunctionReturn(0);
1721 }
1722 
1723 #undef __FUNCT__
1724 #define __FUNCT__ "MatSetValuesLocal"
1725 /*@
1726    MatSetValuesLocal - Inserts or adds values into certain locations of a matrix,
1727    using a local ordering of the nodes.
1728 
1729    Not Collective
1730 
1731    Input Parameters:
1732 +  x - the matrix
1733 .  nrow, irow - number of rows and their local indices
1734 .  ncol, icol - number of columns and their local indices
1735 .  y -  a logically two-dimensional array of values
1736 -  addv - either INSERT_VALUES or ADD_VALUES, where
1737    ADD_VALUES adds values to any existing entries, and
1738    INSERT_VALUES replaces existing entries with new values
1739 
1740    Notes:
1741    Before calling MatSetValuesLocal(), the user must first set the
1742    local-to-global mapping by calling MatSetLocalToGlobalMapping().
1743 
1744    Calls to MatSetValuesLocal() with the INSERT_VALUES and ADD_VALUES
1745    options cannot be mixed without intervening calls to the assembly
1746    routines.
1747 
1748    These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
1749    MUST be called after all calls to MatSetValuesLocal() have been completed.
1750 
1751    Level: intermediate
1752 
1753    Concepts: matrices^putting entries in with local numbering
1754 
1755 .seealso:  MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetLocalToGlobalMapping(),
1756            MatSetValueLocal()
1757 @*/
1758 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
1759 {
1760   PetscErrorCode ierr;
1761   PetscInt       irowm[2048],icolm[2048];
1762 
1763   PetscFunctionBegin;
1764   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1765   PetscValidType(mat,1);
1766   if (!nrow || !ncol) PetscFunctionReturn(0); /* no values to insert */
1767   PetscValidIntPointer(irow,3);
1768   PetscValidIntPointer(icol,5);
1769   PetscValidScalarPointer(y,6);
1770   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1771   if (mat->insertmode == NOT_SET_VALUES) {
1772     mat->insertmode = addv;
1773   }
1774 #if defined(PETSC_USE_DEBUG)
1775   else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1776   if (!mat->ops->setvalueslocal && (nrow > 2048 || ncol > 2048)) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_SUP,"Number column/row indices must be <= 2048: are %D %D",nrow,ncol);
1777   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1778 #endif
1779 
1780   if (mat->assembled) {
1781     mat->was_assembled = PETSC_TRUE;
1782     mat->assembled     = PETSC_FALSE;
1783   }
1784   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1785   if (!mat->ops->setvalueslocal) {
1786     ierr = ISLocalToGlobalMappingApply(mat->mapping,nrow,irow,irowm);CHKERRQ(ierr);
1787     ierr = ISLocalToGlobalMappingApply(mat->mapping,ncol,icol,icolm);CHKERRQ(ierr);
1788     ierr = (*mat->ops->setvalues)(mat,nrow,irowm,ncol,icolm,y,addv);CHKERRQ(ierr);
1789   } else {
1790     ierr = (*mat->ops->setvalueslocal)(mat,nrow,irow,ncol,icol,y,addv);CHKERRQ(ierr);
1791   }
1792   mat->same_nonzero = PETSC_FALSE;
1793   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1794   PetscFunctionReturn(0);
1795 }
1796 
1797 #undef __FUNCT__
1798 #define __FUNCT__ "MatSetValuesBlockedLocal"
1799 /*@
1800    MatSetValuesBlockedLocal - Inserts or adds values into certain locations of a matrix,
1801    using a local ordering of the nodes a block at a time.
1802 
1803    Not Collective
1804 
1805    Input Parameters:
1806 +  x - the matrix
1807 .  nrow, irow - number of rows and their local indices
1808 .  ncol, icol - number of columns and their local indices
1809 .  y -  a logically two-dimensional array of values
1810 -  addv - either INSERT_VALUES or ADD_VALUES, where
1811    ADD_VALUES adds values to any existing entries, and
1812    INSERT_VALUES replaces existing entries with new values
1813 
1814    Notes:
1815    Before calling MatSetValuesBlockedLocal(), the user must first set the
1816    block size using MatSetBlockSize(), and the local-to-global mapping by
1817    calling MatSetLocalToGlobalMappingBlock(), where the mapping MUST be
1818    set for matrix blocks, not for matrix elements.
1819 
1820    Calls to MatSetValuesBlockedLocal() with the INSERT_VALUES and ADD_VALUES
1821    options cannot be mixed without intervening calls to the assembly
1822    routines.
1823 
1824    These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
1825    MUST be called after all calls to MatSetValuesBlockedLocal() have been completed.
1826 
1827    Level: intermediate
1828 
1829    Concepts: matrices^putting blocked values in with local numbering
1830 
1831 .seealso:  MatSetBlockSize(), MatSetLocalToGlobalMappingBlock(), MatAssemblyBegin(), MatAssemblyEnd(),
1832            MatSetValuesLocal(), MatSetLocalToGlobalMappingBlock(), MatSetValuesBlocked()
1833 @*/
1834 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesBlockedLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
1835 {
1836   PetscErrorCode ierr;
1837   PetscInt       irowm[2048],icolm[2048];
1838 
1839   PetscFunctionBegin;
1840   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1841   PetscValidType(mat,1);
1842   if (!nrow || !ncol) PetscFunctionReturn(0); /* no values to insert */
1843   PetscValidIntPointer(irow,3);
1844   PetscValidIntPointer(icol,5);
1845   PetscValidScalarPointer(y,6);
1846   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1847   if (mat->insertmode == NOT_SET_VALUES) {
1848     mat->insertmode = addv;
1849   }
1850 #if defined(PETSC_USE_DEBUG)
1851   else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1852   if (!mat->bmapping) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Local to global never set with MatSetLocalToGlobalMappingBlock()");
1853   if (nrow > 2048 || ncol > 2048) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_SUP,"Number column/row indices must be <= 2048: are %D %D",nrow,ncol);
1854   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1855 #endif
1856 
1857   if (mat->assembled) {
1858     mat->was_assembled = PETSC_TRUE;
1859     mat->assembled     = PETSC_FALSE;
1860   }
1861   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1862   ierr = ISLocalToGlobalMappingApply(mat->bmapping,nrow,irow,irowm);CHKERRQ(ierr);
1863   ierr = ISLocalToGlobalMappingApply(mat->bmapping,ncol,icol,icolm);CHKERRQ(ierr);
1864   if (mat->ops->setvaluesblocked) {
1865     ierr = (*mat->ops->setvaluesblocked)(mat,nrow,irowm,ncol,icolm,y,addv);CHKERRQ(ierr);
1866   } else {
1867     PetscInt buf[4096],*ibufm=0,*ibufn=0;
1868     PetscInt i,j,*iirowm,*iicolm,bs=mat->rmap->bs;
1869     if ((nrow+ncol)*bs <= 4096) {
1870       iirowm = buf; iicolm = buf + nrow*bs;
1871     } else {
1872       ierr = PetscMalloc2(nrow*bs,PetscInt,&ibufm,ncol*bs,PetscInt,&ibufn);CHKERRQ(ierr);
1873       iirowm = ibufm; iicolm = ibufn;
1874     }
1875     for (i=0; i<nrow; i++) {
1876       for (j=0; j<bs; j++) {
1877 	iirowm[i*bs+j] = bs*irowm[i] + j;
1878       }
1879     }
1880     for (i=0; i<ncol; i++) {
1881       for (j=0; j<bs; j++) {
1882 	iicolm[i*bs+j] = bs*icolm[i] + j;
1883       }
1884     }
1885     ierr = MatSetValues(mat,bs*nrow,iirowm,bs*ncol,iicolm,y,addv);CHKERRQ(ierr);
1886     ierr = PetscFree2(ibufm,ibufn);CHKERRQ(ierr);
1887   }
1888   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
1889   PetscFunctionReturn(0);
1890 }
1891 
1892 #undef __FUNCT__
1893 #define __FUNCT__ "MatMultDiagonalBlock"
1894 /*@
1895    MatMultDiagonalBlock - Computes the matrix-vector product, y = Dx. Where D is defined by the inode or block structure of the diagonal
1896 
1897    Collective on Mat and Vec
1898 
1899    Input Parameters:
1900 +  mat - the matrix
1901 -  x   - the vector to be multiplied
1902 
1903    Output Parameters:
1904 .  y - the result
1905 
1906    Notes:
1907    The vectors x and y cannot be the same.  I.e., one cannot
1908    call MatMult(A,y,y).
1909 
1910    Level: developer
1911 
1912    Concepts: matrix-vector product
1913 
1914 .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
1915 @*/
1916 PetscErrorCode PETSCMAT_DLLEXPORT MatMultDiagonalBlock(Mat mat,Vec x,Vec y)
1917 {
1918   PetscErrorCode ierr;
1919 
1920   PetscFunctionBegin;
1921   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1922   PetscValidType(mat,1);
1923   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
1924   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
1925 
1926   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1927   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1928   if (x == y) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
1929   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1930 
1931   if (!mat->ops->multdiagonalblock) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"This matrix type does not have a multiply defined");
1932   ierr = (*mat->ops->multdiagonalblock)(mat,x,y);CHKERRQ(ierr);
1933   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
1934   PetscFunctionReturn(0);
1935 }
1936 
1937 /* --------------------------------------------------------*/
1938 #undef __FUNCT__
1939 #define __FUNCT__ "MatMult"
1940 /*@
1941    MatMult - Computes the matrix-vector product, y = Ax.
1942 
1943    Neighbor-wise Collective on Mat and Vec
1944 
1945    Input Parameters:
1946 +  mat - the matrix
1947 -  x   - the vector to be multiplied
1948 
1949    Output Parameters:
1950 .  y - the result
1951 
1952    Notes:
1953    The vectors x and y cannot be the same.  I.e., one cannot
1954    call MatMult(A,y,y).
1955 
1956    Level: beginner
1957 
1958    Concepts: matrix-vector product
1959 
1960 .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
1961 @*/
1962 PetscErrorCode PETSCMAT_DLLEXPORT MatMult(Mat mat,Vec x,Vec y)
1963 {
1964   PetscErrorCode ierr;
1965 
1966   PetscFunctionBegin;
1967   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
1968   PetscValidType(mat,1);
1969   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
1970   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
1971 
1972   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1973   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1974   if (x == y) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
1975 #ifndef PETSC_HAVE_CONSTRAINTS
1976   if (mat->cmap->N != x->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
1977   if (mat->rmap->N != y->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->rmap->N,y->map->N);
1978   if (mat->rmap->n != y->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: local dim %D %D",mat->rmap->n,y->map->n);
1979 #endif
1980   ierr = MatPreallocated(mat);CHKERRQ(ierr);
1981 
1982   if (mat->nullsp) {
1983     ierr = MatNullSpaceRemove(mat->nullsp,x,&x);CHKERRQ(ierr);
1984   }
1985 
1986   if (!mat->ops->mult) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"This matrix type does not have a multiply defined");
1987   ierr = PetscLogEventBegin(MAT_Mult,mat,x,y,0);CHKERRQ(ierr);
1988   ierr = (*mat->ops->mult)(mat,x,y);CHKERRQ(ierr);
1989   ierr = PetscLogEventEnd(MAT_Mult,mat,x,y,0);CHKERRQ(ierr);
1990 
1991   if (mat->nullsp) {
1992     ierr = MatNullSpaceRemove(mat->nullsp,y,PETSC_NULL);CHKERRQ(ierr);
1993   }
1994   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
1995   PetscFunctionReturn(0);
1996 }
1997 
1998 #undef __FUNCT__
1999 #define __FUNCT__ "MatMultTranspose"
2000 /*@
2001    MatMultTranspose - Computes matrix transpose times a vector.
2002 
2003    Neighbor-wise Collective on Mat and Vec
2004 
2005    Input Parameters:
2006 +  mat - the matrix
2007 -  x   - the vector to be multilplied
2008 
2009    Output Parameters:
2010 .  y - the result
2011 
2012    Notes:
2013    The vectors x and y cannot be the same.  I.e., one cannot
2014    call MatMultTranspose(A,y,y).
2015 
2016    Level: beginner
2017 
2018    Concepts: matrix vector product^transpose
2019 
2020 .seealso: MatMult(), MatMultAdd(), MatMultTransposeAdd()
2021 @*/
2022 PetscErrorCode PETSCMAT_DLLEXPORT MatMultTranspose(Mat mat,Vec x,Vec y)
2023 {
2024   PetscErrorCode ierr;
2025 
2026   PetscFunctionBegin;
2027   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2028   PetscValidType(mat,1);
2029   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
2030   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
2031 
2032   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2033   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2034   if (x == y) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2035 #ifndef PETSC_HAVE_CONSTRAINTS
2036   if (mat->rmap->N != x->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
2037   if (mat->cmap->N != y->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->cmap->N,y->map->N);
2038 #endif
2039   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2040 
2041   if (!mat->ops->multtranspose) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"This matrix type does not have a multiply tranpose defined");
2042   ierr = PetscLogEventBegin(MAT_MultTranspose,mat,x,y,0);CHKERRQ(ierr);
2043   ierr = (*mat->ops->multtranspose)(mat,x,y);CHKERRQ(ierr);
2044   ierr = PetscLogEventEnd(MAT_MultTranspose,mat,x,y,0);CHKERRQ(ierr);
2045   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
2046   PetscFunctionReturn(0);
2047 }
2048 
2049 #undef __FUNCT__
2050 #define __FUNCT__ "MatMultHermitianTranspose"
2051 /*@
2052    MatMultHermitianTranspose - Computes matrix Hermitian transpose times a vector.
2053 
2054    Neighbor-wise Collective on Mat and Vec
2055 
2056    Input Parameters:
2057 +  mat - the matrix
2058 -  x   - the vector to be multilplied
2059 
2060    Output Parameters:
2061 .  y - the result
2062 
2063    Notes:
2064    The vectors x and y cannot be the same.  I.e., one cannot
2065    call MatMultHermitianTranspose(A,y,y).
2066 
2067    Level: beginner
2068 
2069    Concepts: matrix vector product^transpose
2070 
2071 .seealso: MatMult(), MatMultAdd(), MatMultHermitianTransposeAdd(), MatMultTranspose()
2072 @*/
2073 PetscErrorCode PETSCMAT_DLLEXPORT MatMultHermitianTranspose(Mat mat,Vec x,Vec y)
2074 {
2075   PetscErrorCode ierr;
2076 
2077   PetscFunctionBegin;
2078   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2079   PetscValidType(mat,1);
2080   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
2081   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
2082 
2083   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2084   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2085   if (x == y) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2086 #ifndef PETSC_HAVE_CONSTRAINTS
2087   if (mat->rmap->N != x->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
2088   if (mat->cmap->N != y->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->cmap->N,y->map->N);
2089 #endif
2090   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2091 
2092   if (!mat->ops->multtranspose) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"This matrix type does not have a multiply tranpose defined");
2093   ierr = PetscLogEventBegin(MAT_MultHermitianTranspose,mat,x,y,0);CHKERRQ(ierr);
2094   ierr = (*mat->ops->multhermitiantranspose)(mat,x,y);CHKERRQ(ierr);
2095   ierr = PetscLogEventEnd(MAT_MultHermitianTranspose,mat,x,y,0);CHKERRQ(ierr);
2096   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
2097   PetscFunctionReturn(0);
2098 }
2099 
2100 #undef __FUNCT__
2101 #define __FUNCT__ "MatMultAdd"
2102 /*@
2103     MatMultAdd -  Computes v3 = v2 + A * v1.
2104 
2105     Neighbor-wise Collective on Mat and Vec
2106 
2107     Input Parameters:
2108 +   mat - the matrix
2109 -   v1, v2 - the vectors
2110 
2111     Output Parameters:
2112 .   v3 - the result
2113 
2114     Notes:
2115     The vectors v1 and v3 cannot be the same.  I.e., one cannot
2116     call MatMultAdd(A,v1,v2,v1).
2117 
2118     Level: beginner
2119 
2120     Concepts: matrix vector product^addition
2121 
2122 .seealso: MatMultTranspose(), MatMult(), MatMultTransposeAdd()
2123 @*/
2124 PetscErrorCode PETSCMAT_DLLEXPORT MatMultAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2125 {
2126   PetscErrorCode ierr;
2127 
2128   PetscFunctionBegin;
2129   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2130   PetscValidType(mat,1);
2131   PetscValidHeaderSpecific(v1,VEC_CLASSID,2);
2132   PetscValidHeaderSpecific(v2,VEC_CLASSID,3);
2133   PetscValidHeaderSpecific(v3,VEC_CLASSID,4);
2134 
2135   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2136   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2137   if (mat->cmap->N != v1->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->cmap->N,v1->map->N);
2138   /* if (mat->rmap->N != v2->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: global dim %D %D",mat->rmap->N,v2->map->N);
2139      if (mat->rmap->N != v3->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: global dim %D %D",mat->rmap->N,v3->map->N); */
2140   if (mat->rmap->n != v3->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: local dim %D %D",mat->rmap->n,v3->map->n);
2141   if (mat->rmap->n != v2->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: local dim %D %D",mat->rmap->n,v2->map->n);
2142   if (v1 == v3) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2143   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2144 
2145   ierr = PetscLogEventBegin(MAT_MultAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2146   ierr = (*mat->ops->multadd)(mat,v1,v2,v3);CHKERRQ(ierr);
2147   ierr = PetscLogEventEnd(MAT_MultAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2148   ierr = PetscObjectStateIncrease((PetscObject)v3);CHKERRQ(ierr);
2149   PetscFunctionReturn(0);
2150 }
2151 
2152 #undef __FUNCT__
2153 #define __FUNCT__ "MatMultTransposeAdd"
2154 /*@
2155    MatMultTransposeAdd - Computes v3 = v2 + A' * v1.
2156 
2157    Neighbor-wise Collective on Mat and Vec
2158 
2159    Input Parameters:
2160 +  mat - the matrix
2161 -  v1, v2 - the vectors
2162 
2163    Output Parameters:
2164 .  v3 - the result
2165 
2166    Notes:
2167    The vectors v1 and v3 cannot be the same.  I.e., one cannot
2168    call MatMultTransposeAdd(A,v1,v2,v1).
2169 
2170    Level: beginner
2171 
2172    Concepts: matrix vector product^transpose and addition
2173 
2174 .seealso: MatMultTranspose(), MatMultAdd(), MatMult()
2175 @*/
2176 PetscErrorCode PETSCMAT_DLLEXPORT MatMultTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2177 {
2178   PetscErrorCode ierr;
2179 
2180   PetscFunctionBegin;
2181   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2182   PetscValidType(mat,1);
2183   PetscValidHeaderSpecific(v1,VEC_CLASSID,2);
2184   PetscValidHeaderSpecific(v2,VEC_CLASSID,3);
2185   PetscValidHeaderSpecific(v3,VEC_CLASSID,4);
2186 
2187   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2188   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2189   if (!mat->ops->multtransposeadd) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2190   if (v1 == v3) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2191   if (mat->rmap->N != v1->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->rmap->N,v1->map->N);
2192   if (mat->cmap->N != v2->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: global dim %D %D",mat->cmap->N,v2->map->N);
2193   if (mat->cmap->N != v3->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: global dim %D %D",mat->cmap->N,v3->map->N);
2194   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2195 
2196   ierr = PetscLogEventBegin(MAT_MultTransposeAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2197   ierr = (*mat->ops->multtransposeadd)(mat,v1,v2,v3);CHKERRQ(ierr);
2198   ierr = PetscLogEventEnd(MAT_MultTransposeAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2199   ierr = PetscObjectStateIncrease((PetscObject)v3);CHKERRQ(ierr);
2200   PetscFunctionReturn(0);
2201 }
2202 
2203 #undef __FUNCT__
2204 #define __FUNCT__ "MatMultHermitianTransposeAdd"
2205 /*@
2206    MatMultHermitianTransposeAdd - Computes v3 = v2 + A^H * v1.
2207 
2208    Neighbor-wise Collective on Mat and Vec
2209 
2210    Input Parameters:
2211 +  mat - the matrix
2212 -  v1, v2 - the vectors
2213 
2214    Output Parameters:
2215 .  v3 - the result
2216 
2217    Notes:
2218    The vectors v1 and v3 cannot be the same.  I.e., one cannot
2219    call MatMultHermitianTransposeAdd(A,v1,v2,v1).
2220 
2221    Level: beginner
2222 
2223    Concepts: matrix vector product^transpose and addition
2224 
2225 .seealso: MatMultHermitianTranspose(), MatMultTranspose(), MatMultAdd(), MatMult()
2226 @*/
2227 PetscErrorCode PETSCMAT_DLLEXPORT MatMultHermitianTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2228 {
2229   PetscErrorCode ierr;
2230 
2231   PetscFunctionBegin;
2232   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2233   PetscValidType(mat,1);
2234   PetscValidHeaderSpecific(v1,VEC_CLASSID,2);
2235   PetscValidHeaderSpecific(v2,VEC_CLASSID,3);
2236   PetscValidHeaderSpecific(v3,VEC_CLASSID,4);
2237 
2238   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2239   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2240   if (!mat->ops->multtransposeadd) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2241   if (v1 == v3) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2242   if (mat->rmap->N != v1->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->rmap->N,v1->map->N);
2243   if (mat->cmap->N != v2->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: global dim %D %D",mat->cmap->N,v2->map->N);
2244   if (mat->cmap->N != v3->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: global dim %D %D",mat->cmap->N,v3->map->N);
2245   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2246 
2247   ierr = PetscLogEventBegin(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2248   ierr = (*mat->ops->multhermitiantransposeadd)(mat,v1,v2,v3);CHKERRQ(ierr);
2249   ierr = PetscLogEventEnd(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);CHKERRQ(ierr);
2250   ierr = PetscObjectStateIncrease((PetscObject)v3);CHKERRQ(ierr);
2251   PetscFunctionReturn(0);
2252 }
2253 
2254 #undef __FUNCT__
2255 #define __FUNCT__ "MatMultConstrained"
2256 /*@
2257    MatMultConstrained - The inner multiplication routine for a
2258    constrained matrix P^T A P.
2259 
2260    Neighbor-wise Collective on Mat and Vec
2261 
2262    Input Parameters:
2263 +  mat - the matrix
2264 -  x   - the vector to be multilplied
2265 
2266    Output Parameters:
2267 .  y - the result
2268 
2269    Notes:
2270    The vectors x and y cannot be the same.  I.e., one cannot
2271    call MatMult(A,y,y).
2272 
2273    Level: beginner
2274 
2275 .keywords: matrix, multiply, matrix-vector product, constraint
2276 .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2277 @*/
2278 PetscErrorCode PETSCMAT_DLLEXPORT MatMultConstrained(Mat mat,Vec x,Vec y)
2279 {
2280   PetscErrorCode ierr;
2281 
2282   PetscFunctionBegin;
2283   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2284   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
2285   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
2286   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2287   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2288   if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2289   if (mat->cmap->N != x->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
2290   if (mat->rmap->N != y->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->rmap->N,y->map->N);
2291   if (mat->rmap->n != y->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: local dim %D %D",mat->rmap->n,y->map->n);
2292 
2293   ierr = PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);CHKERRQ(ierr);
2294   ierr = (*mat->ops->multconstrained)(mat,x,y);CHKERRQ(ierr);
2295   ierr = PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);CHKERRQ(ierr);
2296   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
2297 
2298   PetscFunctionReturn(0);
2299 }
2300 
2301 #undef __FUNCT__
2302 #define __FUNCT__ "MatMultTransposeConstrained"
2303 /*@
2304    MatMultTransposeConstrained - The inner multiplication routine for a
2305    constrained matrix P^T A^T P.
2306 
2307    Neighbor-wise Collective on Mat and Vec
2308 
2309    Input Parameters:
2310 +  mat - the matrix
2311 -  x   - the vector to be multilplied
2312 
2313    Output Parameters:
2314 .  y - the result
2315 
2316    Notes:
2317    The vectors x and y cannot be the same.  I.e., one cannot
2318    call MatMult(A,y,y).
2319 
2320    Level: beginner
2321 
2322 .keywords: matrix, multiply, matrix-vector product, constraint
2323 .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2324 @*/
2325 PetscErrorCode PETSCMAT_DLLEXPORT MatMultTransposeConstrained(Mat mat,Vec x,Vec y)
2326 {
2327   PetscErrorCode ierr;
2328 
2329   PetscFunctionBegin;
2330   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2331   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
2332   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
2333   if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2334   if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2335   if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2336   if (mat->rmap->N != x->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
2337   if (mat->cmap->N != y->map->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->rmap->N,y->map->N);
2338 
2339   ierr = PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);CHKERRQ(ierr);
2340   ierr = (*mat->ops->multtransposeconstrained)(mat,x,y);CHKERRQ(ierr);
2341   ierr = PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);CHKERRQ(ierr);
2342   ierr = PetscObjectStateIncrease((PetscObject)y);CHKERRQ(ierr);
2343 
2344   PetscFunctionReturn(0);
2345 }
2346 
2347 #undef __FUNCT__
2348 #define __FUNCT__ "MatGetFactorType"
2349 /*@C
2350    MatGetFactorType - gets the type of factorization it is
2351 
2352    Note Collective
2353    as the flag
2354 
2355    Input Parameters:
2356 .  mat - the matrix
2357 
2358    Output Parameters:
2359 .  t - the type, one of MAT_FACTOR_NONE, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ILU, MAT_FACTOR_ICC,MAT_FACTOR_ILUDT
2360 
2361     Level: intermediate
2362 
2363 .seealso:    MatFactorType, MatGetFactor()
2364 @*/
2365 PetscErrorCode PETSCMAT_DLLEXPORT MatGetFactorType(Mat mat,MatFactorType *t)
2366 {
2367   PetscFunctionBegin;
2368   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2369   PetscValidType(mat,1);
2370   *t = mat->factortype;
2371   PetscFunctionReturn(0);
2372 }
2373 
2374 /* ------------------------------------------------------------*/
2375 #undef __FUNCT__
2376 #define __FUNCT__ "MatGetInfo"
2377 /*@C
2378    MatGetInfo - Returns information about matrix storage (number of
2379    nonzeros, memory, etc.).
2380 
2381    Collective on Mat if MAT_GLOBAL_MAX or MAT_GLOBAL_SUM is used as the flag
2382 
2383    Input Parameters:
2384 .  mat - the matrix
2385 
2386    Output Parameters:
2387 +  flag - flag indicating the type of parameters to be returned
2388    (MAT_LOCAL - local matrix, MAT_GLOBAL_MAX - maximum over all processors,
2389    MAT_GLOBAL_SUM - sum over all processors)
2390 -  info - matrix information context
2391 
2392    Notes:
2393    The MatInfo context contains a variety of matrix data, including
2394    number of nonzeros allocated and used, number of mallocs during
2395    matrix assembly, etc.  Additional information for factored matrices
2396    is provided (such as the fill ratio, number of mallocs during
2397    factorization, etc.).  Much of this info is printed to PETSC_STDOUT
2398    when using the runtime options
2399 $       -info -mat_view_info
2400 
2401    Example for C/C++ Users:
2402    See the file ${PETSC_DIR}/include/petscmat.h for a complete list of
2403    data within the MatInfo context.  For example,
2404 .vb
2405       MatInfo info;
2406       Mat     A;
2407       double  mal, nz_a, nz_u;
2408 
2409       MatGetInfo(A,MAT_LOCAL,&info);
2410       mal  = info.mallocs;
2411       nz_a = info.nz_allocated;
2412 .ve
2413 
2414    Example for Fortran Users:
2415    Fortran users should declare info as a double precision
2416    array of dimension MAT_INFO_SIZE, and then extract the parameters
2417    of interest.  See the file ${PETSC_DIR}/include/finclude/petscmat.h
2418    a complete list of parameter names.
2419 .vb
2420       double  precision info(MAT_INFO_SIZE)
2421       double  precision mal, nz_a
2422       Mat     A
2423       integer ierr
2424 
2425       call MatGetInfo(A,MAT_LOCAL,info,ierr)
2426       mal = info(MAT_INFO_MALLOCS)
2427       nz_a = info(MAT_INFO_NZ_ALLOCATED)
2428 .ve
2429 
2430     Level: intermediate
2431 
2432     Concepts: matrices^getting information on
2433 
2434     Developer Note: fortran interface is not autogenerated as the f90
2435     interface defintion cannot be generated correctly [due to MatInfo]
2436 
2437 @*/
2438 PetscErrorCode PETSCMAT_DLLEXPORT MatGetInfo(Mat mat,MatInfoType flag,MatInfo *info)
2439 {
2440   PetscErrorCode ierr;
2441 
2442   PetscFunctionBegin;
2443   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2444   PetscValidType(mat,1);
2445   PetscValidPointer(info,3);
2446   if (!mat->ops->getinfo) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2447   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2448   ierr = (*mat->ops->getinfo)(mat,flag,info);CHKERRQ(ierr);
2449   PetscFunctionReturn(0);
2450 }
2451 
2452 /* ----------------------------------------------------------*/
2453 
2454 #undef __FUNCT__
2455 #define __FUNCT__ "MatLUFactor"
2456 /*@C
2457    MatLUFactor - Performs in-place LU factorization of matrix.
2458 
2459    Collective on Mat
2460 
2461    Input Parameters:
2462 +  mat - the matrix
2463 .  row - row permutation
2464 .  col - column permutation
2465 -  info - options for factorization, includes
2466 $          fill - expected fill as ratio of original fill.
2467 $          dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2468 $                   Run with the option -info to determine an optimal value to use
2469 
2470    Notes:
2471    Most users should employ the simplified KSP interface for linear solvers
2472    instead of working directly with matrix algebra routines such as this.
2473    See, e.g., KSPCreate().
2474 
2475    This changes the state of the matrix to a factored matrix; it cannot be used
2476    for example with MatSetValues() unless one first calls MatSetUnfactored().
2477 
2478    Level: developer
2479 
2480    Concepts: matrices^LU factorization
2481 
2482 .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(),
2483           MatGetOrdering(), MatSetUnfactored(), MatFactorInfo
2484 
2485     Developer Note: fortran interface is not autogenerated as the f90
2486     interface defintion cannot be generated correctly [due to MatFactorInfo]
2487 
2488 @*/
2489 PetscErrorCode PETSCMAT_DLLEXPORT MatLUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
2490 {
2491   PetscErrorCode ierr;
2492 
2493   PetscFunctionBegin;
2494   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2495   if (row) PetscValidHeaderSpecific(row,IS_CLASSID,2);
2496   if (col) PetscValidHeaderSpecific(col,IS_CLASSID,3);
2497   PetscValidPointer(info,4);
2498   PetscValidType(mat,1);
2499   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2500   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2501   if (!mat->ops->lufactor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2502   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2503 
2504   ierr = PetscLogEventBegin(MAT_LUFactor,mat,row,col,0);CHKERRQ(ierr);
2505   ierr = (*mat->ops->lufactor)(mat,row,col,info);CHKERRQ(ierr);
2506   ierr = PetscLogEventEnd(MAT_LUFactor,mat,row,col,0);CHKERRQ(ierr);
2507   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
2508   PetscFunctionReturn(0);
2509 }
2510 
2511 #undef __FUNCT__
2512 #define __FUNCT__ "MatILUFactor"
2513 /*@C
2514    MatILUFactor - Performs in-place ILU factorization of matrix.
2515 
2516    Collective on Mat
2517 
2518    Input Parameters:
2519 +  mat - the matrix
2520 .  row - row permutation
2521 .  col - column permutation
2522 -  info - structure containing
2523 $      levels - number of levels of fill.
2524 $      expected fill - as ratio of original fill.
2525 $      1 or 0 - indicating force fill on diagonal (improves robustness for matrices
2526                 missing diagonal entries)
2527 
2528    Notes:
2529    Probably really in-place only when level of fill is zero, otherwise allocates
2530    new space to store factored matrix and deletes previous memory.
2531 
2532    Most users should employ the simplified KSP interface for linear solvers
2533    instead of working directly with matrix algebra routines such as this.
2534    See, e.g., KSPCreate().
2535 
2536    Level: developer
2537 
2538    Concepts: matrices^ILU factorization
2539 
2540 .seealso: MatILUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
2541 
2542     Developer Note: fortran interface is not autogenerated as the f90
2543     interface defintion cannot be generated correctly [due to MatFactorInfo]
2544 
2545 @*/
2546 PetscErrorCode PETSCMAT_DLLEXPORT MatILUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
2547 {
2548   PetscErrorCode ierr;
2549 
2550   PetscFunctionBegin;
2551   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2552   if (row) PetscValidHeaderSpecific(row,IS_CLASSID,2);
2553   if (col) PetscValidHeaderSpecific(col,IS_CLASSID,3);
2554   PetscValidPointer(info,4);
2555   PetscValidType(mat,1);
2556   if (mat->rmap->N != mat->cmap->N) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONG,"matrix must be square");
2557   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2558   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2559   if (!mat->ops->ilufactor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2560   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2561 
2562   ierr = PetscLogEventBegin(MAT_ILUFactor,mat,row,col,0);CHKERRQ(ierr);
2563   ierr = (*mat->ops->ilufactor)(mat,row,col,info);CHKERRQ(ierr);
2564   ierr = PetscLogEventEnd(MAT_ILUFactor,mat,row,col,0);CHKERRQ(ierr);
2565   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
2566   PetscFunctionReturn(0);
2567 }
2568 
2569 #undef __FUNCT__
2570 #define __FUNCT__ "MatLUFactorSymbolic"
2571 /*@C
2572    MatLUFactorSymbolic - Performs symbolic LU factorization of matrix.
2573    Call this routine before calling MatLUFactorNumeric().
2574 
2575    Collective on Mat
2576 
2577    Input Parameters:
2578 +  fact - the factor matrix obtained with MatGetFactor()
2579 .  mat - the matrix
2580 .  row, col - row and column permutations
2581 -  info - options for factorization, includes
2582 $          fill - expected fill as ratio of original fill.
2583 $          dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2584 $                   Run with the option -info to determine an optimal value to use
2585 
2586 
2587    Notes:
2588    See the users manual for additional information about
2589    choosing the fill factor for better efficiency.
2590 
2591    Most users should employ the simplified KSP interface for linear solvers
2592    instead of working directly with matrix algebra routines such as this.
2593    See, e.g., KSPCreate().
2594 
2595    Level: developer
2596 
2597    Concepts: matrices^LU symbolic factorization
2598 
2599 .seealso: MatLUFactor(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
2600 
2601     Developer Note: fortran interface is not autogenerated as the f90
2602     interface defintion cannot be generated correctly [due to MatFactorInfo]
2603 
2604 @*/
2605 PetscErrorCode PETSCMAT_DLLEXPORT MatLUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
2606 {
2607   PetscErrorCode ierr;
2608 
2609   PetscFunctionBegin;
2610   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2611   if (row) PetscValidHeaderSpecific(row,IS_CLASSID,2);
2612   if (col) PetscValidHeaderSpecific(col,IS_CLASSID,3);
2613   PetscValidPointer(info,4);
2614   PetscValidType(mat,1);
2615   PetscValidPointer(fact,5);
2616   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2617   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2618   if (!(fact)->ops->lufactorsymbolic) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Matrix type %s  symbolic LU",((PetscObject)mat)->type_name);
2619   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2620 
2621   ierr = PetscLogEventBegin(MAT_LUFactorSymbolic,mat,row,col,0);CHKERRQ(ierr);
2622   ierr = (fact->ops->lufactorsymbolic)(fact,mat,row,col,info);CHKERRQ(ierr);
2623   ierr = PetscLogEventEnd(MAT_LUFactorSymbolic,mat,row,col,0);CHKERRQ(ierr);
2624   ierr = PetscObjectStateIncrease((PetscObject)fact);CHKERRQ(ierr);
2625   PetscFunctionReturn(0);
2626 }
2627 
2628 #undef __FUNCT__
2629 #define __FUNCT__ "MatLUFactorNumeric"
2630 /*@C
2631    MatLUFactorNumeric - Performs numeric LU factorization of a matrix.
2632    Call this routine after first calling MatLUFactorSymbolic().
2633 
2634    Collective on Mat
2635 
2636    Input Parameters:
2637 +  fact - the factor matrix obtained with MatGetFactor()
2638 .  mat - the matrix
2639 -  info - options for factorization
2640 
2641    Notes:
2642    See MatLUFactor() for in-place factorization.  See
2643    MatCholeskyFactorNumeric() for the symmetric, positive definite case.
2644 
2645    Most users should employ the simplified KSP interface for linear solvers
2646    instead of working directly with matrix algebra routines such as this.
2647    See, e.g., KSPCreate().
2648 
2649    Level: developer
2650 
2651    Concepts: matrices^LU numeric factorization
2652 
2653 .seealso: MatLUFactorSymbolic(), MatLUFactor(), MatCholeskyFactor()
2654 
2655     Developer Note: fortran interface is not autogenerated as the f90
2656     interface defintion cannot be generated correctly [due to MatFactorInfo]
2657 
2658 @*/
2659 PetscErrorCode PETSCMAT_DLLEXPORT MatLUFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
2660 {
2661   PetscErrorCode ierr;
2662 
2663   PetscFunctionBegin;
2664   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2665   PetscValidType(mat,1);
2666   PetscValidPointer(fact,2);
2667   PetscValidHeaderSpecific(fact,MAT_CLASSID,2);
2668   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2669   if (mat->rmap->N != (fact)->rmap->N || mat->cmap->N != (fact)->cmap->N) {
2670     SETERRQ4(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Mat fact: global dimensions are different %D should = %D %D should = %D",mat->rmap->N,(fact)->rmap->N,mat->cmap->N,(fact)->cmap->N);
2671   }
2672   if (!(fact)->ops->lufactornumeric) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2673   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2674   ierr = PetscLogEventBegin(MAT_LUFactorNumeric,mat,fact,0,0);CHKERRQ(ierr);
2675   ierr = (fact->ops->lufactornumeric)(fact,mat,info);CHKERRQ(ierr);
2676   ierr = PetscLogEventEnd(MAT_LUFactorNumeric,mat,fact,0,0);CHKERRQ(ierr);
2677 
2678   ierr = MatView_Private(fact);CHKERRQ(ierr);
2679   ierr = PetscObjectStateIncrease((PetscObject)fact);CHKERRQ(ierr);
2680   PetscFunctionReturn(0);
2681 }
2682 
2683 #undef __FUNCT__
2684 #define __FUNCT__ "MatCholeskyFactor"
2685 /*@C
2686    MatCholeskyFactor - Performs in-place Cholesky factorization of a
2687    symmetric matrix.
2688 
2689    Collective on Mat
2690 
2691    Input Parameters:
2692 +  mat - the matrix
2693 .  perm - row and column permutations
2694 -  f - expected fill as ratio of original fill
2695 
2696    Notes:
2697    See MatLUFactor() for the nonsymmetric case.  See also
2698    MatCholeskyFactorSymbolic(), and MatCholeskyFactorNumeric().
2699 
2700    Most users should employ the simplified KSP interface for linear solvers
2701    instead of working directly with matrix algebra routines such as this.
2702    See, e.g., KSPCreate().
2703 
2704    Level: developer
2705 
2706    Concepts: matrices^Cholesky factorization
2707 
2708 .seealso: MatLUFactor(), MatCholeskyFactorSymbolic(), MatCholeskyFactorNumeric()
2709           MatGetOrdering()
2710 
2711     Developer Note: fortran interface is not autogenerated as the f90
2712     interface defintion cannot be generated correctly [due to MatFactorInfo]
2713 
2714 @*/
2715 PetscErrorCode PETSCMAT_DLLEXPORT MatCholeskyFactor(Mat mat,IS perm,const MatFactorInfo *info)
2716 {
2717   PetscErrorCode ierr;
2718 
2719   PetscFunctionBegin;
2720   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2721   PetscValidType(mat,1);
2722   PetscValidHeaderSpecific(perm,IS_CLASSID,2);
2723   PetscValidPointer(info,3);
2724   if (mat->rmap->N != mat->cmap->N) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONG,"Matrix must be square");
2725   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2726   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2727   if (!mat->ops->choleskyfactor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2728   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2729 
2730   ierr = PetscLogEventBegin(MAT_CholeskyFactor,mat,perm,0,0);CHKERRQ(ierr);
2731   ierr = (*mat->ops->choleskyfactor)(mat,perm,info);CHKERRQ(ierr);
2732   ierr = PetscLogEventEnd(MAT_CholeskyFactor,mat,perm,0,0);CHKERRQ(ierr);
2733   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
2734   PetscFunctionReturn(0);
2735 }
2736 
2737 #undef __FUNCT__
2738 #define __FUNCT__ "MatCholeskyFactorSymbolic"
2739 /*@C
2740    MatCholeskyFactorSymbolic - Performs symbolic Cholesky factorization
2741    of a symmetric matrix.
2742 
2743    Collective on Mat
2744 
2745    Input Parameters:
2746 +  fact - the factor matrix obtained with MatGetFactor()
2747 .  mat - the matrix
2748 .  perm - row and column permutations
2749 -  info - options for factorization, includes
2750 $          fill - expected fill as ratio of original fill.
2751 $          dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2752 $                   Run with the option -info to determine an optimal value to use
2753 
2754    Notes:
2755    See MatLUFactorSymbolic() for the nonsymmetric case.  See also
2756    MatCholeskyFactor() and MatCholeskyFactorNumeric().
2757 
2758    Most users should employ the simplified KSP interface for linear solvers
2759    instead of working directly with matrix algebra routines such as this.
2760    See, e.g., KSPCreate().
2761 
2762    Level: developer
2763 
2764    Concepts: matrices^Cholesky symbolic factorization
2765 
2766 .seealso: MatLUFactorSymbolic(), MatCholeskyFactor(), MatCholeskyFactorNumeric()
2767           MatGetOrdering()
2768 
2769     Developer Note: fortran interface is not autogenerated as the f90
2770     interface defintion cannot be generated correctly [due to MatFactorInfo]
2771 
2772 @*/
2773 PetscErrorCode PETSCMAT_DLLEXPORT MatCholeskyFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
2774 {
2775   PetscErrorCode ierr;
2776 
2777   PetscFunctionBegin;
2778   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2779   PetscValidType(mat,1);
2780   if (perm) PetscValidHeaderSpecific(perm,IS_CLASSID,2);
2781   PetscValidPointer(info,3);
2782   PetscValidPointer(fact,4);
2783   if (mat->rmap->N != mat->cmap->N) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONG,"Matrix must be square");
2784   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2785   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2786   if (!(fact)->ops->choleskyfactorsymbolic) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2787   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2788 
2789   ierr = PetscLogEventBegin(MAT_CholeskyFactorSymbolic,mat,perm,0,0);CHKERRQ(ierr);
2790   ierr = (fact->ops->choleskyfactorsymbolic)(fact,mat,perm,info);CHKERRQ(ierr);
2791   ierr = PetscLogEventEnd(MAT_CholeskyFactorSymbolic,mat,perm,0,0);CHKERRQ(ierr);
2792   ierr = PetscObjectStateIncrease((PetscObject)fact);CHKERRQ(ierr);
2793   PetscFunctionReturn(0);
2794 }
2795 
2796 #undef __FUNCT__
2797 #define __FUNCT__ "MatCholeskyFactorNumeric"
2798 /*@C
2799    MatCholeskyFactorNumeric - Performs numeric Cholesky factorization
2800    of a symmetric matrix. Call this routine after first calling
2801    MatCholeskyFactorSymbolic().
2802 
2803    Collective on Mat
2804 
2805    Input Parameters:
2806 +  fact - the factor matrix obtained with MatGetFactor()
2807 .  mat - the initial matrix
2808 .  info - options for factorization
2809 -  fact - the symbolic factor of mat
2810 
2811 
2812    Notes:
2813    Most users should employ the simplified KSP interface for linear solvers
2814    instead of working directly with matrix algebra routines such as this.
2815    See, e.g., KSPCreate().
2816 
2817    Level: developer
2818 
2819    Concepts: matrices^Cholesky numeric factorization
2820 
2821 .seealso: MatCholeskyFactorSymbolic(), MatCholeskyFactor(), MatLUFactorNumeric()
2822 
2823     Developer Note: fortran interface is not autogenerated as the f90
2824     interface defintion cannot be generated correctly [due to MatFactorInfo]
2825 
2826 @*/
2827 PetscErrorCode PETSCMAT_DLLEXPORT MatCholeskyFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
2828 {
2829   PetscErrorCode ierr;
2830 
2831   PetscFunctionBegin;
2832   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2833   PetscValidType(mat,1);
2834   PetscValidPointer(fact,2);
2835   PetscValidHeaderSpecific(fact,MAT_CLASSID,2);
2836   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2837   if (!(fact)->ops->choleskyfactornumeric) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2838   if (mat->rmap->N != (fact)->rmap->N || mat->cmap->N != (fact)->cmap->N) {
2839     SETERRQ4(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Mat fact: global dim %D should = %D %D should = %D",mat->rmap->N,(fact)->rmap->N,mat->cmap->N,(fact)->cmap->N);
2840   }
2841   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2842 
2843   ierr = PetscLogEventBegin(MAT_CholeskyFactorNumeric,mat,fact,0,0);CHKERRQ(ierr);
2844   ierr = (fact->ops->choleskyfactornumeric)(fact,mat,info);CHKERRQ(ierr);
2845   ierr = PetscLogEventEnd(MAT_CholeskyFactorNumeric,mat,fact,0,0);CHKERRQ(ierr);
2846 
2847   ierr = MatView_Private(fact);CHKERRQ(ierr);
2848   ierr = PetscObjectStateIncrease((PetscObject)fact);CHKERRQ(ierr);
2849   PetscFunctionReturn(0);
2850 }
2851 
2852 /* ----------------------------------------------------------------*/
2853 #undef __FUNCT__
2854 #define __FUNCT__ "MatSolve"
2855 /*@
2856    MatSolve - Solves A x = b, given a factored matrix.
2857 
2858    Neighbor-wise Collective on Mat and Vec
2859 
2860    Input Parameters:
2861 +  mat - the factored matrix
2862 -  b - the right-hand-side vector
2863 
2864    Output Parameter:
2865 .  x - the result vector
2866 
2867    Notes:
2868    The vectors b and x cannot be the same.  I.e., one cannot
2869    call MatSolve(A,x,x).
2870 
2871    Notes:
2872    Most users should employ the simplified KSP interface for linear solvers
2873    instead of working directly with matrix algebra routines such as this.
2874    See, e.g., KSPCreate().
2875 
2876    Level: developer
2877 
2878    Concepts: matrices^triangular solves
2879 
2880 .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd()
2881 @*/
2882 PetscErrorCode PETSCMAT_DLLEXPORT MatSolve(Mat mat,Vec b,Vec x)
2883 {
2884   PetscErrorCode ierr;
2885 
2886   PetscFunctionBegin;
2887   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
2888   PetscValidType(mat,1);
2889   PetscValidHeaderSpecific(b,VEC_CLASSID,2);
2890   PetscValidHeaderSpecific(x,VEC_CLASSID,3);
2891   PetscCheckSameComm(mat,1,b,2);
2892   PetscCheckSameComm(mat,1,x,3);
2893   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
2894   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
2895   if (mat->cmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
2896   if (mat->rmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
2897   if (mat->rmap->n != b->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: local dim %D %D",mat->rmap->n,b->map->n);
2898   if (!mat->rmap->N && !mat->cmap->N) PetscFunctionReturn(0);
2899   if (!mat->ops->solve) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2900   ierr = MatPreallocated(mat);CHKERRQ(ierr);
2901 
2902   ierr = PetscLogEventBegin(MAT_Solve,mat,b,x,0);CHKERRQ(ierr);
2903   ierr = (*mat->ops->solve)(mat,b,x);CHKERRQ(ierr);
2904   ierr = PetscLogEventEnd(MAT_Solve,mat,b,x,0);CHKERRQ(ierr);
2905   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
2906   PetscFunctionReturn(0);
2907 }
2908 
2909 #undef __FUNCT__
2910 #define __FUNCT__ "MatMatSolve_Basic"
2911 PetscErrorCode PETSCMAT_DLLEXPORT MatMatSolve_Basic(Mat A,Mat B,Mat X)
2912 {
2913   PetscErrorCode ierr;
2914   Vec            b,x;
2915   PetscInt       m,N,i;
2916   PetscScalar    *bb,*xx;
2917 
2918   PetscFunctionBegin;
2919   ierr = MatGetArray(B,&bb);CHKERRQ(ierr);
2920   ierr = MatGetArray(X,&xx);CHKERRQ(ierr);
2921   ierr = MatGetLocalSize(B,&m,PETSC_NULL);CHKERRQ(ierr);  /* number local rows */
2922   ierr = MatGetSize(B,PETSC_NULL,&N);CHKERRQ(ierr);       /* total columns in dense matrix */
2923   ierr = MatGetVecs(A,&x,&b);CHKERRQ(ierr);
2924   for (i=0; i<N; i++) {
2925     ierr = VecPlaceArray(b,bb + i*m);CHKERRQ(ierr);
2926     ierr = VecPlaceArray(x,xx + i*m);CHKERRQ(ierr);
2927     ierr = MatSolve(A,b,x);CHKERRQ(ierr);
2928     ierr = VecResetArray(x);CHKERRQ(ierr);
2929     ierr = VecResetArray(b);CHKERRQ(ierr);
2930   }
2931   ierr = VecDestroy(b);CHKERRQ(ierr);
2932   ierr = VecDestroy(x);CHKERRQ(ierr);
2933   ierr = MatRestoreArray(B,&bb);CHKERRQ(ierr);
2934   ierr = MatRestoreArray(X,&xx);CHKERRQ(ierr);
2935   PetscFunctionReturn(0);
2936 }
2937 
2938 #undef __FUNCT__
2939 #define __FUNCT__ "MatMatSolve"
2940 /*@
2941    MatMatSolve - Solves A X = B, given a factored matrix.
2942 
2943    Neighbor-wise Collective on Mat
2944 
2945    Input Parameters:
2946 +  mat - the factored matrix
2947 -  B - the right-hand-side matrix  (dense matrix)
2948 
2949    Output Parameter:
2950 .  X - the result matrix (dense matrix)
2951 
2952    Notes:
2953    The matrices b and x cannot be the same.  I.e., one cannot
2954    call MatMatSolve(A,x,x).
2955 
2956    Notes:
2957    Most users should usually employ the simplified KSP interface for linear solvers
2958    instead of working directly with matrix algebra routines such as this.
2959    See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
2960    at a time.
2961 
2962    Level: developer
2963 
2964    Concepts: matrices^triangular solves
2965 
2966 .seealso: MatMatSolveAdd(), MatMatSolveTranspose(), MatMatSolveTransposeAdd(), MatLUFactor(), MatCholeskyFactor()
2967 @*/
2968 PetscErrorCode PETSCMAT_DLLEXPORT MatMatSolve(Mat A,Mat B,Mat X)
2969 {
2970   PetscErrorCode ierr;
2971 
2972   PetscFunctionBegin;
2973   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
2974   PetscValidType(A,1);
2975   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
2976   PetscValidHeaderSpecific(X,MAT_CLASSID,3);
2977   PetscCheckSameComm(A,1,B,2);
2978   PetscCheckSameComm(A,1,X,3);
2979   if (X == B) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_IDN,"X and B must be different matrices");
2980   if (!A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
2981   if (A->cmap->N != X->rmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Mat A,Mat X: global dim %D %D",A->cmap->N,X->rmap->N);
2982   if (A->rmap->N != B->rmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Mat A,Mat B: global dim %D %D",A->rmap->N,B->rmap->N);
2983   if (A->rmap->n != B->rmap->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat A,Mat B: local dim %D %D",A->rmap->n,B->rmap->n);
2984   if (!A->rmap->N && !A->cmap->N) PetscFunctionReturn(0);
2985   ierr = MatPreallocated(A);CHKERRQ(ierr);
2986 
2987   ierr = PetscLogEventBegin(MAT_MatSolve,A,B,X,0);CHKERRQ(ierr);
2988   if (!A->ops->matsolve) {
2989     ierr = PetscInfo1(A,"Mat type %s using basic MatMatSolve",((PetscObject)A)->type_name);CHKERRQ(ierr);
2990     ierr = MatMatSolve_Basic(A,B,X);CHKERRQ(ierr);
2991   } else {
2992     ierr = (*A->ops->matsolve)(A,B,X);CHKERRQ(ierr);
2993   }
2994   ierr = PetscLogEventEnd(MAT_MatSolve,A,B,X,0);CHKERRQ(ierr);
2995   ierr = PetscObjectStateIncrease((PetscObject)X);CHKERRQ(ierr);
2996   PetscFunctionReturn(0);
2997 }
2998 
2999 
3000 #undef __FUNCT__
3001 #define __FUNCT__ "MatForwardSolve"
3002 /*@
3003    MatForwardSolve - Solves L x = b, given a factored matrix, A = LU, or
3004                             U^T*D^(1/2) x = b, given a factored symmetric matrix, A = U^T*D*U,
3005 
3006    Neighbor-wise Collective on Mat and Vec
3007 
3008    Input Parameters:
3009 +  mat - the factored matrix
3010 -  b - the right-hand-side vector
3011 
3012    Output Parameter:
3013 .  x - the result vector
3014 
3015    Notes:
3016    MatSolve() should be used for most applications, as it performs
3017    a forward solve followed by a backward solve.
3018 
3019    The vectors b and x cannot be the same,  i.e., one cannot
3020    call MatForwardSolve(A,x,x).
3021 
3022    For matrix in seqsbaij format with block size larger than 1,
3023    the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3024    MatForwardSolve() solves U^T*D y = b, and
3025    MatBackwardSolve() solves U x = y.
3026    Thus they do not provide a symmetric preconditioner.
3027 
3028    Most users should employ the simplified KSP interface for linear solvers
3029    instead of working directly with matrix algebra routines such as this.
3030    See, e.g., KSPCreate().
3031 
3032    Level: developer
3033 
3034    Concepts: matrices^forward solves
3035 
3036 .seealso: MatSolve(), MatBackwardSolve()
3037 @*/
3038 PetscErrorCode PETSCMAT_DLLEXPORT MatForwardSolve(Mat mat,Vec b,Vec x)
3039 {
3040   PetscErrorCode ierr;
3041 
3042   PetscFunctionBegin;
3043   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3044   PetscValidType(mat,1);
3045   PetscValidHeaderSpecific(b,VEC_CLASSID,2);
3046   PetscValidHeaderSpecific(x,VEC_CLASSID,3);
3047   PetscCheckSameComm(mat,1,b,2);
3048   PetscCheckSameComm(mat,1,x,3);
3049   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3050   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3051   if (!mat->ops->forwardsolve) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3052   if (mat->cmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3053   if (mat->rmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3054   if (mat->rmap->n != b->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: local dim %D %D",mat->rmap->n,b->map->n);
3055   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3056   ierr = PetscLogEventBegin(MAT_ForwardSolve,mat,b,x,0);CHKERRQ(ierr);
3057   ierr = (*mat->ops->forwardsolve)(mat,b,x);CHKERRQ(ierr);
3058   ierr = PetscLogEventEnd(MAT_ForwardSolve,mat,b,x,0);CHKERRQ(ierr);
3059   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3060   PetscFunctionReturn(0);
3061 }
3062 
3063 #undef __FUNCT__
3064 #define __FUNCT__ "MatBackwardSolve"
3065 /*@
3066    MatBackwardSolve - Solves U x = b, given a factored matrix, A = LU.
3067                              D^(1/2) U x = b, given a factored symmetric matrix, A = U^T*D*U,
3068 
3069    Neighbor-wise Collective on Mat and Vec
3070 
3071    Input Parameters:
3072 +  mat - the factored matrix
3073 -  b - the right-hand-side vector
3074 
3075    Output Parameter:
3076 .  x - the result vector
3077 
3078    Notes:
3079    MatSolve() should be used for most applications, as it performs
3080    a forward solve followed by a backward solve.
3081 
3082    The vectors b and x cannot be the same.  I.e., one cannot
3083    call MatBackwardSolve(A,x,x).
3084 
3085    For matrix in seqsbaij format with block size larger than 1,
3086    the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3087    MatForwardSolve() solves U^T*D y = b, and
3088    MatBackwardSolve() solves U x = y.
3089    Thus they do not provide a symmetric preconditioner.
3090 
3091    Most users should employ the simplified KSP interface for linear solvers
3092    instead of working directly with matrix algebra routines such as this.
3093    See, e.g., KSPCreate().
3094 
3095    Level: developer
3096 
3097    Concepts: matrices^backward solves
3098 
3099 .seealso: MatSolve(), MatForwardSolve()
3100 @*/
3101 PetscErrorCode PETSCMAT_DLLEXPORT MatBackwardSolve(Mat mat,Vec b,Vec x)
3102 {
3103   PetscErrorCode ierr;
3104 
3105   PetscFunctionBegin;
3106   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3107   PetscValidType(mat,1);
3108   PetscValidHeaderSpecific(b,VEC_CLASSID,2);
3109   PetscValidHeaderSpecific(x,VEC_CLASSID,3);
3110   PetscCheckSameComm(mat,1,b,2);
3111   PetscCheckSameComm(mat,1,x,3);
3112   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3113   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3114   if (!mat->ops->backwardsolve) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3115   if (mat->cmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3116   if (mat->rmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3117   if (mat->rmap->n != b->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: local dim %D %D",mat->rmap->n,b->map->n);
3118   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3119 
3120   ierr = PetscLogEventBegin(MAT_BackwardSolve,mat,b,x,0);CHKERRQ(ierr);
3121   ierr = (*mat->ops->backwardsolve)(mat,b,x);CHKERRQ(ierr);
3122   ierr = PetscLogEventEnd(MAT_BackwardSolve,mat,b,x,0);CHKERRQ(ierr);
3123   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3124   PetscFunctionReturn(0);
3125 }
3126 
3127 #undef __FUNCT__
3128 #define __FUNCT__ "MatSolveAdd"
3129 /*@
3130    MatSolveAdd - Computes x = y + inv(A)*b, given a factored matrix.
3131 
3132    Neighbor-wise Collective on Mat and Vec
3133 
3134    Input Parameters:
3135 +  mat - the factored matrix
3136 .  b - the right-hand-side vector
3137 -  y - the vector to be added to
3138 
3139    Output Parameter:
3140 .  x - the result vector
3141 
3142    Notes:
3143    The vectors b and x cannot be the same.  I.e., one cannot
3144    call MatSolveAdd(A,x,y,x).
3145 
3146    Most users should employ the simplified KSP interface for linear solvers
3147    instead of working directly with matrix algebra routines such as this.
3148    See, e.g., KSPCreate().
3149 
3150    Level: developer
3151 
3152    Concepts: matrices^triangular solves
3153 
3154 .seealso: MatSolve(), MatSolveTranspose(), MatSolveTransposeAdd()
3155 @*/
3156 PetscErrorCode PETSCMAT_DLLEXPORT MatSolveAdd(Mat mat,Vec b,Vec y,Vec x)
3157 {
3158   PetscScalar    one = 1.0;
3159   Vec            tmp;
3160   PetscErrorCode ierr;
3161 
3162   PetscFunctionBegin;
3163   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3164   PetscValidType(mat,1);
3165   PetscValidHeaderSpecific(y,VEC_CLASSID,2);
3166   PetscValidHeaderSpecific(b,VEC_CLASSID,3);
3167   PetscValidHeaderSpecific(x,VEC_CLASSID,4);
3168   PetscCheckSameComm(mat,1,b,2);
3169   PetscCheckSameComm(mat,1,y,2);
3170   PetscCheckSameComm(mat,1,x,3);
3171   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3172   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3173   if (mat->cmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3174   if (mat->rmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3175   if (mat->rmap->N != y->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->rmap->N,y->map->N);
3176   if (mat->rmap->n != b->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: local dim %D %D",mat->rmap->n,b->map->n);
3177   if (x->map->n != y->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Vec x,Vec y: local dim %D %D",x->map->n,y->map->n);
3178   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3179 
3180   ierr = PetscLogEventBegin(MAT_SolveAdd,mat,b,x,y);CHKERRQ(ierr);
3181   if (mat->ops->solveadd)  {
3182     ierr = (*mat->ops->solveadd)(mat,b,y,x);CHKERRQ(ierr);
3183   } else {
3184     /* do the solve then the add manually */
3185     if (x != y) {
3186       ierr = MatSolve(mat,b,x);CHKERRQ(ierr);
3187       ierr = VecAXPY(x,one,y);CHKERRQ(ierr);
3188     } else {
3189       ierr = VecDuplicate(x,&tmp);CHKERRQ(ierr);
3190       ierr = PetscLogObjectParent(mat,tmp);CHKERRQ(ierr);
3191       ierr = VecCopy(x,tmp);CHKERRQ(ierr);
3192       ierr = MatSolve(mat,b,x);CHKERRQ(ierr);
3193       ierr = VecAXPY(x,one,tmp);CHKERRQ(ierr);
3194       ierr = VecDestroy(tmp);CHKERRQ(ierr);
3195     }
3196   }
3197   ierr = PetscLogEventEnd(MAT_SolveAdd,mat,b,x,y);CHKERRQ(ierr);
3198   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3199   PetscFunctionReturn(0);
3200 }
3201 
3202 #undef __FUNCT__
3203 #define __FUNCT__ "MatSolveTranspose"
3204 /*@
3205    MatSolveTranspose - Solves A' x = b, given a factored matrix.
3206 
3207    Neighbor-wise Collective on Mat and Vec
3208 
3209    Input Parameters:
3210 +  mat - the factored matrix
3211 -  b - the right-hand-side vector
3212 
3213    Output Parameter:
3214 .  x - the result vector
3215 
3216    Notes:
3217    The vectors b and x cannot be the same.  I.e., one cannot
3218    call MatSolveTranspose(A,x,x).
3219 
3220    Most users should employ the simplified KSP interface for linear solvers
3221    instead of working directly with matrix algebra routines such as this.
3222    See, e.g., KSPCreate().
3223 
3224    Level: developer
3225 
3226    Concepts: matrices^triangular solves
3227 
3228 .seealso: MatSolve(), MatSolveAdd(), MatSolveTransposeAdd()
3229 @*/
3230 PetscErrorCode PETSCMAT_DLLEXPORT MatSolveTranspose(Mat mat,Vec b,Vec x)
3231 {
3232   PetscErrorCode ierr;
3233 
3234   PetscFunctionBegin;
3235   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3236   PetscValidType(mat,1);
3237   PetscValidHeaderSpecific(b,VEC_CLASSID,2);
3238   PetscValidHeaderSpecific(x,VEC_CLASSID,3);
3239   PetscCheckSameComm(mat,1,b,2);
3240   PetscCheckSameComm(mat,1,x,3);
3241   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3242   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3243   if (!mat->ops->solvetranspose) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Matrix type %s",((PetscObject)mat)->type_name);
3244   if (mat->rmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
3245   if (mat->cmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->cmap->N,b->map->N);
3246   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3247   ierr = PetscLogEventBegin(MAT_SolveTranspose,mat,b,x,0);CHKERRQ(ierr);
3248   ierr = (*mat->ops->solvetranspose)(mat,b,x);CHKERRQ(ierr);
3249   ierr = PetscLogEventEnd(MAT_SolveTranspose,mat,b,x,0);CHKERRQ(ierr);
3250   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3251   PetscFunctionReturn(0);
3252 }
3253 
3254 #undef __FUNCT__
3255 #define __FUNCT__ "MatSolveTransposeAdd"
3256 /*@
3257    MatSolveTransposeAdd - Computes x = y + inv(Transpose(A)) b, given a
3258                       factored matrix.
3259 
3260    Neighbor-wise Collective on Mat and Vec
3261 
3262    Input Parameters:
3263 +  mat - the factored matrix
3264 .  b - the right-hand-side vector
3265 -  y - the vector to be added to
3266 
3267    Output Parameter:
3268 .  x - the result vector
3269 
3270    Notes:
3271    The vectors b and x cannot be the same.  I.e., one cannot
3272    call MatSolveTransposeAdd(A,x,y,x).
3273 
3274    Most users should employ the simplified KSP interface for linear solvers
3275    instead of working directly with matrix algebra routines such as this.
3276    See, e.g., KSPCreate().
3277 
3278    Level: developer
3279 
3280    Concepts: matrices^triangular solves
3281 
3282 .seealso: MatSolve(), MatSolveAdd(), MatSolveTranspose()
3283 @*/
3284 PetscErrorCode PETSCMAT_DLLEXPORT MatSolveTransposeAdd(Mat mat,Vec b,Vec y,Vec x)
3285 {
3286   PetscScalar    one = 1.0;
3287   PetscErrorCode ierr;
3288   Vec            tmp;
3289 
3290   PetscFunctionBegin;
3291   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3292   PetscValidType(mat,1);
3293   PetscValidHeaderSpecific(y,VEC_CLASSID,2);
3294   PetscValidHeaderSpecific(b,VEC_CLASSID,3);
3295   PetscValidHeaderSpecific(x,VEC_CLASSID,4);
3296   PetscCheckSameComm(mat,1,b,2);
3297   PetscCheckSameComm(mat,1,y,3);
3298   PetscCheckSameComm(mat,1,x,4);
3299   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3300   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3301   if (mat->rmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
3302   if (mat->cmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->cmap->N,b->map->N);
3303   if (mat->cmap->N != y->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->cmap->N,y->map->N);
3304   if (x->map->n != y->map->n)   SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Vec x,Vec y: local dim %D %D",x->map->n,y->map->n);
3305   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3306 
3307   ierr = PetscLogEventBegin(MAT_SolveTransposeAdd,mat,b,x,y);CHKERRQ(ierr);
3308   if (mat->ops->solvetransposeadd) {
3309     ierr = (*mat->ops->solvetransposeadd)(mat,b,y,x);CHKERRQ(ierr);
3310   } else {
3311     /* do the solve then the add manually */
3312     if (x != y) {
3313       ierr = MatSolveTranspose(mat,b,x);CHKERRQ(ierr);
3314       ierr = VecAXPY(x,one,y);CHKERRQ(ierr);
3315     } else {
3316       ierr = VecDuplicate(x,&tmp);CHKERRQ(ierr);
3317       ierr = PetscLogObjectParent(mat,tmp);CHKERRQ(ierr);
3318       ierr = VecCopy(x,tmp);CHKERRQ(ierr);
3319       ierr = MatSolveTranspose(mat,b,x);CHKERRQ(ierr);
3320       ierr = VecAXPY(x,one,tmp);CHKERRQ(ierr);
3321       ierr = VecDestroy(tmp);CHKERRQ(ierr);
3322     }
3323   }
3324   ierr = PetscLogEventEnd(MAT_SolveTransposeAdd,mat,b,x,y);CHKERRQ(ierr);
3325   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3326   PetscFunctionReturn(0);
3327 }
3328 /* ----------------------------------------------------------------*/
3329 
3330 #undef __FUNCT__
3331 #define __FUNCT__ "MatSOR"
3332 /*@
3333    MatSOR - Computes relaxation (SOR, Gauss-Seidel) sweeps.
3334 
3335    Neighbor-wise Collective on Mat and Vec
3336 
3337    Input Parameters:
3338 +  mat - the matrix
3339 .  b - the right hand side
3340 .  omega - the relaxation factor
3341 .  flag - flag indicating the type of SOR (see below)
3342 .  shift -  diagonal shift
3343 .  its - the number of iterations
3344 -  lits - the number of local iterations
3345 
3346    Output Parameters:
3347 .  x - the solution (can contain an initial guess, use option SOR_ZERO_INITIAL_GUESS to indicate no guess)
3348 
3349    SOR Flags:
3350 .     SOR_FORWARD_SWEEP - forward SOR
3351 .     SOR_BACKWARD_SWEEP - backward SOR
3352 .     SOR_SYMMETRIC_SWEEP - SSOR (symmetric SOR)
3353 .     SOR_LOCAL_FORWARD_SWEEP - local forward SOR
3354 .     SOR_LOCAL_BACKWARD_SWEEP - local forward SOR
3355 .     SOR_LOCAL_SYMMETRIC_SWEEP - local SSOR
3356 .     SOR_APPLY_UPPER, SOR_APPLY_LOWER - applies
3357          upper/lower triangular part of matrix to
3358          vector (with omega)
3359 .     SOR_ZERO_INITIAL_GUESS - zero initial guess
3360 
3361    Notes:
3362    SOR_LOCAL_FORWARD_SWEEP, SOR_LOCAL_BACKWARD_SWEEP, and
3363    SOR_LOCAL_SYMMETRIC_SWEEP perform separate independent smoothings
3364    on each processor.
3365 
3366    Application programmers will not generally use MatSOR() directly,
3367    but instead will employ the KSP/PC interface.
3368 
3369    Notes: for BAIJ, SBAIJ, and AIJ matrices with Inodes this does a block SOR smoothing, otherwise it does a pointwise smoothing
3370 
3371    Notes for Advanced Users:
3372    The flags are implemented as bitwise inclusive or operations.
3373    For example, use (SOR_ZERO_INITIAL_GUESS | SOR_SYMMETRIC_SWEEP)
3374    to specify a zero initial guess for SSOR.
3375 
3376    Most users should employ the simplified KSP interface for linear solvers
3377    instead of working directly with matrix algebra routines such as this.
3378    See, e.g., KSPCreate().
3379 
3380    Vectors x and b CANNOT be the same
3381 
3382    Level: developer
3383 
3384    Concepts: matrices^relaxation
3385    Concepts: matrices^SOR
3386    Concepts: matrices^Gauss-Seidel
3387 
3388 @*/
3389 PetscErrorCode PETSCMAT_DLLEXPORT MatSOR(Mat mat,Vec b,PetscReal omega,MatSORType flag,PetscReal shift,PetscInt its,PetscInt lits,Vec x)
3390 {
3391   PetscErrorCode ierr;
3392 
3393   PetscFunctionBegin;
3394   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3395   PetscValidType(mat,1);
3396   PetscValidHeaderSpecific(b,VEC_CLASSID,2);
3397   PetscValidHeaderSpecific(x,VEC_CLASSID,8);
3398   PetscCheckSameComm(mat,1,b,2);
3399   PetscCheckSameComm(mat,1,x,8);
3400   if (!mat->ops->sor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3401   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3402   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3403   if (mat->cmap->N != x->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3404   if (mat->rmap->N != b->map->N) SETERRQ2(((PetscObject)mat)->comm,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3405   if (mat->rmap->n != b->map->n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: local dim %D %D",mat->rmap->n,b->map->n);
3406   if (its <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D positive",its);
3407   if (lits <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires local its %D positive",lits);
3408   if (b == x) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_IDN,"b and x vector cannot be the same");
3409 
3410   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3411   ierr = PetscLogEventBegin(MAT_SOR,mat,b,x,0);CHKERRQ(ierr);
3412   ierr =(*mat->ops->sor)(mat,b,omega,flag,shift,its,lits,x);CHKERRQ(ierr);
3413   ierr = PetscLogEventEnd(MAT_SOR,mat,b,x,0);CHKERRQ(ierr);
3414   ierr = PetscObjectStateIncrease((PetscObject)x);CHKERRQ(ierr);
3415   PetscFunctionReturn(0);
3416 }
3417 
3418 #undef __FUNCT__
3419 #define __FUNCT__ "MatCopy_Basic"
3420 /*
3421       Default matrix copy routine.
3422 */
3423 PetscErrorCode MatCopy_Basic(Mat A,Mat B,MatStructure str)
3424 {
3425   PetscErrorCode    ierr;
3426   PetscInt          i,rstart = 0,rend = 0,nz;
3427   const PetscInt    *cwork;
3428   const PetscScalar *vwork;
3429 
3430   PetscFunctionBegin;
3431   if (B->assembled) {
3432     ierr = MatZeroEntries(B);CHKERRQ(ierr);
3433   }
3434   ierr = MatGetOwnershipRange(A,&rstart,&rend);CHKERRQ(ierr);
3435   for (i=rstart; i<rend; i++) {
3436     ierr = MatGetRow(A,i,&nz,&cwork,&vwork);CHKERRQ(ierr);
3437     ierr = MatSetValues(B,1,&i,nz,cwork,vwork,INSERT_VALUES);CHKERRQ(ierr);
3438     ierr = MatRestoreRow(A,i,&nz,&cwork,&vwork);CHKERRQ(ierr);
3439   }
3440   ierr = MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr);
3441   ierr = MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);CHKERRQ(ierr);
3442   ierr = PetscObjectStateIncrease((PetscObject)B);CHKERRQ(ierr);
3443   PetscFunctionReturn(0);
3444 }
3445 
3446 #undef __FUNCT__
3447 #define __FUNCT__ "MatCopy"
3448 /*@
3449    MatCopy - Copys a matrix to another matrix.
3450 
3451    Collective on Mat
3452 
3453    Input Parameters:
3454 +  A - the matrix
3455 -  str - SAME_NONZERO_PATTERN or DIFFERENT_NONZERO_PATTERN
3456 
3457    Output Parameter:
3458 .  B - where the copy is put
3459 
3460    Notes:
3461    If you use SAME_NONZERO_PATTERN then the two matrices had better have the
3462    same nonzero pattern or the routine will crash.
3463 
3464    MatCopy() copies the matrix entries of a matrix to another existing
3465    matrix (after first zeroing the second matrix).  A related routine is
3466    MatConvert(), which first creates a new matrix and then copies the data.
3467 
3468    Level: intermediate
3469 
3470    Concepts: matrices^copying
3471 
3472 .seealso: MatConvert(), MatDuplicate()
3473 
3474 @*/
3475 PetscErrorCode PETSCMAT_DLLEXPORT MatCopy(Mat A,Mat B,MatStructure str)
3476 {
3477   PetscErrorCode ierr;
3478   PetscInt       i;
3479 
3480   PetscFunctionBegin;
3481   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
3482   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
3483   PetscValidType(A,1);
3484   PetscValidType(B,2);
3485   PetscCheckSameComm(A,1,B,2);
3486   ierr = MatPreallocated(B);CHKERRQ(ierr);
3487   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3488   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3489   if (A->rmap->N != B->rmap->N || A->cmap->N != B->cmap->N) SETERRQ4(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Mat A,Mat B: global dim (%D,%D) (%D,%D)",A->rmap->N,B->rmap->N,A->cmap->N,B->cmap->N);
3490   ierr = MatPreallocated(A);CHKERRQ(ierr);
3491 
3492   ierr = PetscLogEventBegin(MAT_Copy,A,B,0,0);CHKERRQ(ierr);
3493   if (A->ops->copy) {
3494     ierr = (*A->ops->copy)(A,B,str);CHKERRQ(ierr);
3495   } else { /* generic conversion */
3496     ierr = MatCopy_Basic(A,B,str);CHKERRQ(ierr);
3497   }
3498   if (A->mapping) {
3499     if (B->mapping) {ierr = ISLocalToGlobalMappingDestroy(B->mapping);CHKERRQ(ierr);B->mapping = 0;}
3500     ierr = MatSetLocalToGlobalMapping(B,A->mapping);CHKERRQ(ierr);
3501   }
3502   if (A->bmapping) {
3503     if (B->bmapping) {ierr = ISLocalToGlobalMappingDestroy(B->bmapping);CHKERRQ(ierr);B->bmapping = 0;}
3504     ierr = MatSetLocalToGlobalMappingBlock(B,A->mapping);CHKERRQ(ierr);
3505   }
3506 
3507   B->stencil.dim = A->stencil.dim;
3508   B->stencil.noc = A->stencil.noc;
3509   for (i=0; i<=A->stencil.dim; i++) {
3510     B->stencil.dims[i]   = A->stencil.dims[i];
3511     B->stencil.starts[i] = A->stencil.starts[i];
3512   }
3513 
3514   ierr = PetscLogEventEnd(MAT_Copy,A,B,0,0);CHKERRQ(ierr);
3515   ierr = PetscObjectStateIncrease((PetscObject)B);CHKERRQ(ierr);
3516   PetscFunctionReturn(0);
3517 }
3518 
3519 #undef __FUNCT__
3520 #define __FUNCT__ "MatConvert"
3521 /*@C
3522    MatConvert - Converts a matrix to another matrix, either of the same
3523    or different type.
3524 
3525    Collective on Mat
3526 
3527    Input Parameters:
3528 +  mat - the matrix
3529 .  newtype - new matrix type.  Use MATSAME to create a new matrix of the
3530    same type as the original matrix.
3531 -  reuse - denotes if the destination matrix is to be created or reused.  Currently
3532    MAT_REUSE_MATRIX is only supported for inplace conversion, otherwise use
3533    MAT_INITIAL_MATRIX.
3534 
3535    Output Parameter:
3536 .  M - pointer to place new matrix
3537 
3538    Notes:
3539    MatConvert() first creates a new matrix and then copies the data from
3540    the first matrix.  A related routine is MatCopy(), which copies the matrix
3541    entries of one matrix to another already existing matrix context.
3542 
3543    Cannot be used to convert a sequential matrix to parallel or parallel to sequential,
3544    the MPI communicator of the generated matrix is always the same as the communicator
3545    of the input matrix.
3546 
3547    Level: intermediate
3548 
3549    Concepts: matrices^converting between storage formats
3550 
3551 .seealso: MatCopy(), MatDuplicate()
3552 @*/
3553 PetscErrorCode PETSCMAT_DLLEXPORT MatConvert(Mat mat, const MatType newtype,MatReuse reuse,Mat *M)
3554 {
3555   PetscErrorCode         ierr;
3556   PetscTruth             sametype,issame,flg;
3557   char                   convname[256],mtype[256];
3558   Mat                    B;
3559 
3560   PetscFunctionBegin;
3561   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3562   PetscValidType(mat,1);
3563   PetscValidPointer(M,3);
3564   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3565   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3566   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3567 
3568   ierr = PetscOptionsGetString(((PetscObject)mat)->prefix,"-matconvert_type",mtype,256,&flg);CHKERRQ(ierr);
3569   if (flg) {
3570     newtype = mtype;
3571   }
3572   ierr = PetscTypeCompare((PetscObject)mat,newtype,&sametype);CHKERRQ(ierr);
3573   ierr = PetscStrcmp(newtype,"same",&issame);CHKERRQ(ierr);
3574   if ((reuse == MAT_REUSE_MATRIX) && (mat != *M)) {
3575     SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"MAT_REUSE_MATRIX only supported for in-place conversion currently");
3576   }
3577 
3578   if ((reuse == MAT_REUSE_MATRIX) && (issame || sametype)) PetscFunctionReturn(0);
3579 
3580   if ((sametype || issame) && (reuse==MAT_INITIAL_MATRIX) && mat->ops->duplicate) {
3581     ierr = (*mat->ops->duplicate)(mat,MAT_COPY_VALUES,M);CHKERRQ(ierr);
3582   } else {
3583     PetscErrorCode (*conv)(Mat, const MatType,MatReuse,Mat*)=PETSC_NULL;
3584     const char     *prefix[3] = {"seq","mpi",""};
3585     PetscInt       i;
3586     /*
3587        Order of precedence:
3588        1) See if a specialized converter is known to the current matrix.
3589        2) See if a specialized converter is known to the desired matrix class.
3590        3) See if a good general converter is registered for the desired class
3591           (as of 6/27/03 only MATMPIADJ falls into this category).
3592        4) See if a good general converter is known for the current matrix.
3593        5) Use a really basic converter.
3594     */
3595 
3596     /* 1) See if a specialized converter is known to the current matrix and the desired class */
3597     for (i=0; i<3; i++) {
3598       ierr = PetscStrcpy(convname,"MatConvert_");CHKERRQ(ierr);
3599       ierr = PetscStrcat(convname,((PetscObject)mat)->type_name);CHKERRQ(ierr);
3600       ierr = PetscStrcat(convname,"_");CHKERRQ(ierr);
3601       ierr = PetscStrcat(convname,prefix[i]);CHKERRQ(ierr);
3602       ierr = PetscStrcat(convname,newtype);CHKERRQ(ierr);
3603       ierr = PetscStrcat(convname,"_C");CHKERRQ(ierr);
3604       ierr = PetscObjectQueryFunction((PetscObject)mat,convname,(void (**)(void))&conv);CHKERRQ(ierr);
3605       if (conv) goto foundconv;
3606     }
3607 
3608     /* 2)  See if a specialized converter is known to the desired matrix class. */
3609     ierr = MatCreate(((PetscObject)mat)->comm,&B);CHKERRQ(ierr);
3610     ierr = MatSetSizes(B,mat->rmap->n,mat->cmap->n,mat->rmap->N,mat->cmap->N);CHKERRQ(ierr);
3611     ierr = MatSetType(B,newtype);CHKERRQ(ierr);
3612     for (i=0; i<3; i++) {
3613       ierr = PetscStrcpy(convname,"MatConvert_");CHKERRQ(ierr);
3614       ierr = PetscStrcat(convname,((PetscObject)mat)->type_name);CHKERRQ(ierr);
3615       ierr = PetscStrcat(convname,"_");CHKERRQ(ierr);
3616       ierr = PetscStrcat(convname,prefix[i]);CHKERRQ(ierr);
3617       ierr = PetscStrcat(convname,newtype);CHKERRQ(ierr);
3618       ierr = PetscStrcat(convname,"_C");CHKERRQ(ierr);
3619       ierr = PetscObjectQueryFunction((PetscObject)B,convname,(void (**)(void))&conv);CHKERRQ(ierr);
3620       if (conv) {
3621         ierr = MatDestroy(B);CHKERRQ(ierr);
3622         goto foundconv;
3623       }
3624     }
3625 
3626     /* 3) See if a good general converter is registered for the desired class */
3627     conv = B->ops->convertfrom;
3628     ierr = MatDestroy(B);CHKERRQ(ierr);
3629     if (conv) goto foundconv;
3630 
3631     /* 4) See if a good general converter is known for the current matrix */
3632     if (mat->ops->convert) {
3633       conv = mat->ops->convert;
3634     }
3635     if (conv) goto foundconv;
3636 
3637     /* 5) Use a really basic converter. */
3638     conv = MatConvert_Basic;
3639 
3640     foundconv:
3641     ierr = PetscLogEventBegin(MAT_Convert,mat,0,0,0);CHKERRQ(ierr);
3642     ierr = (*conv)(mat,newtype,reuse,M);CHKERRQ(ierr);
3643     ierr = PetscLogEventEnd(MAT_Convert,mat,0,0,0);CHKERRQ(ierr);
3644   }
3645   ierr = PetscObjectStateIncrease((PetscObject)*M);CHKERRQ(ierr);
3646   PetscFunctionReturn(0);
3647 }
3648 
3649 #undef __FUNCT__
3650 #define __FUNCT__ "MatFactorGetSolverPackage"
3651 /*@C
3652    MatFactorGetSolverPackage - Returns name of the package providing the factorization routines
3653 
3654    Not Collective
3655 
3656    Input Parameter:
3657 .  mat - the matrix, must be a factored matrix
3658 
3659    Output Parameter:
3660 .   type - the string name of the package (do not free this string)
3661 
3662    Notes:
3663       In Fortran you pass in a empty string and the package name will be copied into it.
3664     (Make sure the string is long enough)
3665 
3666    Level: intermediate
3667 
3668 .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatGetFactor()
3669 @*/
3670 PetscErrorCode PETSCMAT_DLLEXPORT MatFactorGetSolverPackage(Mat mat, const MatSolverPackage *type)
3671 {
3672   PetscErrorCode         ierr;
3673   PetscErrorCode         (*conv)(Mat,const MatSolverPackage*);
3674 
3675   PetscFunctionBegin;
3676   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3677   PetscValidType(mat,1);
3678   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Only for factored matrix");
3679   ierr = PetscObjectQueryFunction((PetscObject)mat,"MatFactorGetSolverPackage_C",(void (**)(void))&conv);CHKERRQ(ierr);
3680   if (!conv) {
3681     *type = MATSOLVERPETSC;
3682   } else {
3683     ierr = (*conv)(mat,type);CHKERRQ(ierr);
3684   }
3685   PetscFunctionReturn(0);
3686 }
3687 
3688 #undef __FUNCT__
3689 #define __FUNCT__ "MatGetFactor"
3690 /*@C
3691    MatGetFactor - Returns a matrix suitable to calls to MatXXFactorSymbolic()
3692 
3693    Collective on Mat
3694 
3695    Input Parameters:
3696 +  mat - the matrix
3697 .  type - name of solver type, for example, spooles, superlu, plapack, petsc (to use PETSc's default)
3698 -  ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
3699 
3700    Output Parameters:
3701 .  f - the factor matrix used with MatXXFactorSymbolic() calls
3702 
3703    Notes:
3704       Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
3705      such as pastix, superlu, mumps, spooles etc.
3706 
3707       PETSc must have been ./configure to use the external solver, using the option --download-package
3708 
3709    Level: intermediate
3710 
3711 .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable()
3712 @*/
3713 PetscErrorCode PETSCMAT_DLLEXPORT MatGetFactor(Mat mat, const MatSolverPackage type,MatFactorType ftype,Mat *f)
3714 {
3715   PetscErrorCode  ierr,(*conv)(Mat,MatFactorType,Mat*);
3716   char            convname[256];
3717 
3718   PetscFunctionBegin;
3719   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3720   PetscValidType(mat,1);
3721 
3722   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3723   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3724 
3725   ierr = PetscStrcpy(convname,"MatGetFactor_");CHKERRQ(ierr);
3726   ierr = PetscStrcat(convname,type);CHKERRQ(ierr);
3727   ierr = PetscStrcat(convname,"_C");CHKERRQ(ierr);
3728   ierr = PetscObjectQueryFunction((PetscObject)mat,convname,(void (**)(void))&conv);CHKERRQ(ierr);
3729   if (!conv) {
3730     PetscTruth flag;
3731     MPI_Comm   comm;
3732 
3733     ierr = PetscObjectGetComm((PetscObject)mat,&comm);CHKERRQ(ierr);
3734     ierr = PetscStrcasecmp(MATSOLVERPETSC,type,&flag);CHKERRQ(ierr);
3735     if (flag) {
3736       SETERRQ2(comm,PETSC_ERR_SUP,"Matrix format %s does not have a built-in PETSc %s",((PetscObject)mat)->type_name,MatFactorTypes[ftype]);
3737     } else {
3738       SETERRQ4(comm,PETSC_ERR_SUP,"Matrix format %s does not have a solver package %s for %s. Perhaps you must ./configure with --download-%s",((PetscObject)mat)->type_name,type,MatFactorTypes[ftype],type);
3739     }
3740   }
3741   ierr = (*conv)(mat,ftype,f);CHKERRQ(ierr);
3742   PetscFunctionReturn(0);
3743 }
3744 
3745 #undef __FUNCT__
3746 #define __FUNCT__ "MatGetFactorAvailable"
3747 /*@C
3748    MatGetFactorAvailable - Returns a a flag if matrix supports particular package and factor type
3749 
3750    Not Collective
3751 
3752    Input Parameters:
3753 +  mat - the matrix
3754 .  type - name of solver type, for example, spooles, superlu, plapack, petsc (to use PETSc's default)
3755 -  ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
3756 
3757    Output Parameter:
3758 .    flg - PETSC_TRUE if the factorization is available
3759 
3760    Notes:
3761       Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
3762      such as pastix, superlu, mumps, spooles etc.
3763 
3764       PETSc must have been ./configure to use the external solver, using the option --download-package
3765 
3766    Level: intermediate
3767 
3768 .seealso: MatCopy(), MatDuplicate(), MatGetFactor()
3769 @*/
3770 PetscErrorCode PETSCMAT_DLLEXPORT MatGetFactorAvailable(Mat mat, const MatSolverPackage type,MatFactorType ftype,PetscTruth *flg)
3771 {
3772   PetscErrorCode         ierr;
3773   char                   convname[256];
3774   PetscErrorCode         (*conv)(Mat,MatFactorType,PetscTruth*);
3775 
3776   PetscFunctionBegin;
3777   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3778   PetscValidType(mat,1);
3779 
3780   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3781   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3782 
3783   ierr = PetscStrcpy(convname,"MatGetFactorAvailable_");CHKERRQ(ierr);
3784   ierr = PetscStrcat(convname,type);CHKERRQ(ierr);
3785   ierr = PetscStrcat(convname,"_C");CHKERRQ(ierr);
3786   ierr = PetscObjectQueryFunction((PetscObject)mat,convname,(void (**)(void))&conv);CHKERRQ(ierr);
3787   if (!conv) {
3788     *flg = PETSC_FALSE;
3789   } else {
3790     ierr = (*conv)(mat,ftype,flg);CHKERRQ(ierr);
3791   }
3792   PetscFunctionReturn(0);
3793 }
3794 
3795 
3796 #undef __FUNCT__
3797 #define __FUNCT__ "MatDuplicate"
3798 /*@
3799    MatDuplicate - Duplicates a matrix including the non-zero structure.
3800 
3801    Collective on Mat
3802 
3803    Input Parameters:
3804 +  mat - the matrix
3805 -  op - either MAT_DO_NOT_COPY_VALUES or MAT_COPY_VALUES, cause it to copy the numerical values in the matrix
3806         MAT_SHARE_NONZERO_PATTERN to share the nonzero patterns with the previous matrix and not copy them.
3807 
3808    Output Parameter:
3809 .  M - pointer to place new matrix
3810 
3811    Level: intermediate
3812 
3813    Concepts: matrices^duplicating
3814 
3815     Notes: You cannot change the nonzero pattern for the parent or child matrix if you use MAT_SHARE_NONZERO_PATTERN.
3816 
3817 .seealso: MatCopy(), MatConvert()
3818 @*/
3819 PetscErrorCode PETSCMAT_DLLEXPORT MatDuplicate(Mat mat,MatDuplicateOption op,Mat *M)
3820 {
3821   PetscErrorCode ierr;
3822   Mat            B;
3823   PetscInt       i;
3824 
3825   PetscFunctionBegin;
3826   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3827   PetscValidType(mat,1);
3828   PetscValidPointer(M,3);
3829   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3830   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3831   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3832 
3833   *M  = 0;
3834   if (!mat->ops->duplicate) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Not written for this matrix type");
3835   ierr = PetscLogEventBegin(MAT_Convert,mat,0,0,0);CHKERRQ(ierr);
3836   ierr = (*mat->ops->duplicate)(mat,op,M);CHKERRQ(ierr);
3837   B = *M;
3838   if (mat->mapping) {
3839     ierr = MatSetLocalToGlobalMapping(B,mat->mapping);CHKERRQ(ierr);
3840   }
3841   if (mat->bmapping) {
3842     ierr = MatSetLocalToGlobalMappingBlock(B,mat->bmapping);CHKERRQ(ierr);
3843   }
3844   ierr = PetscLayoutCopy(mat->rmap,&B->rmap);CHKERRQ(ierr);
3845   ierr = PetscLayoutCopy(mat->cmap,&B->cmap);CHKERRQ(ierr);
3846 
3847   B->stencil.dim = mat->stencil.dim;
3848   B->stencil.noc = mat->stencil.noc;
3849   for (i=0; i<=mat->stencil.dim; i++) {
3850     B->stencil.dims[i]   = mat->stencil.dims[i];
3851     B->stencil.starts[i] = mat->stencil.starts[i];
3852   }
3853 
3854   ierr = PetscLogEventEnd(MAT_Convert,mat,0,0,0);CHKERRQ(ierr);
3855   ierr = PetscObjectStateIncrease((PetscObject)B);CHKERRQ(ierr);
3856   PetscFunctionReturn(0);
3857 }
3858 
3859 #undef __FUNCT__
3860 #define __FUNCT__ "MatGetDiagonal"
3861 /*@
3862    MatGetDiagonal - Gets the diagonal of a matrix.
3863 
3864    Logically Collective on Mat and Vec
3865 
3866    Input Parameters:
3867 +  mat - the matrix
3868 -  v - the vector for storing the diagonal
3869 
3870    Output Parameter:
3871 .  v - the diagonal of the matrix
3872 
3873    Level: intermediate
3874 
3875    Concepts: matrices^accessing diagonals
3876 
3877 .seealso: MatGetRow(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMaxAbs()
3878 @*/
3879 PetscErrorCode PETSCMAT_DLLEXPORT MatGetDiagonal(Mat mat,Vec v)
3880 {
3881   PetscErrorCode ierr;
3882 
3883   PetscFunctionBegin;
3884   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3885   PetscValidType(mat,1);
3886   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
3887   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3888   if (!mat->ops->getdiagonal) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3889   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3890 
3891   ierr = (*mat->ops->getdiagonal)(mat,v);CHKERRQ(ierr);
3892   ierr = PetscObjectStateIncrease((PetscObject)v);CHKERRQ(ierr);
3893   PetscFunctionReturn(0);
3894 }
3895 
3896 #undef __FUNCT__
3897 #define __FUNCT__ "MatGetRowMin"
3898 /*@
3899    MatGetRowMin - Gets the minimum value (of the real part) of each
3900         row of the matrix
3901 
3902    Logically Collective on Mat and Vec
3903 
3904    Input Parameters:
3905 .  mat - the matrix
3906 
3907    Output Parameter:
3908 +  v - the vector for storing the maximums
3909 -  idx - the indices of the column found for each row (optional)
3910 
3911    Level: intermediate
3912 
3913    Notes: The result of this call are the same as if one converted the matrix to dense format
3914       and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
3915 
3916     This code is only implemented for a couple of matrix formats.
3917 
3918    Concepts: matrices^getting row maximums
3919 
3920 .seealso: MatGetDiagonal(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMaxAbs(),
3921           MatGetRowMax()
3922 @*/
3923 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowMin(Mat mat,Vec v,PetscInt idx[])
3924 {
3925   PetscErrorCode ierr;
3926 
3927   PetscFunctionBegin;
3928   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3929   PetscValidType(mat,1);
3930   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
3931   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3932   if (!mat->ops->getrowmax) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3933   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3934 
3935   ierr = (*mat->ops->getrowmin)(mat,v,idx);CHKERRQ(ierr);
3936   ierr = PetscObjectStateIncrease((PetscObject)v);CHKERRQ(ierr);
3937   PetscFunctionReturn(0);
3938 }
3939 
3940 #undef __FUNCT__
3941 #define __FUNCT__ "MatGetRowMinAbs"
3942 /*@
3943    MatGetRowMinAbs - Gets the minimum value (in absolute value) of each
3944         row of the matrix
3945 
3946    Logically Collective on Mat and Vec
3947 
3948    Input Parameters:
3949 .  mat - the matrix
3950 
3951    Output Parameter:
3952 +  v - the vector for storing the minimums
3953 -  idx - the indices of the column found for each row (optional)
3954 
3955    Level: intermediate
3956 
3957    Notes: if a row is completely empty or has only 0.0 values then the idx[] value for that
3958     row is 0 (the first column).
3959 
3960     This code is only implemented for a couple of matrix formats.
3961 
3962    Concepts: matrices^getting row maximums
3963 
3964 .seealso: MatGetDiagonal(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMax(), MatGetRowMaxAbs(), MatGetRowMin()
3965 @*/
3966 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowMinAbs(Mat mat,Vec v,PetscInt idx[])
3967 {
3968   PetscErrorCode ierr;
3969 
3970   PetscFunctionBegin;
3971   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
3972   PetscValidType(mat,1);
3973   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
3974   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3975   if (!mat->ops->getrowminabs) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3976   ierr = MatPreallocated(mat);CHKERRQ(ierr);
3977   if (idx) {ierr = PetscMemzero(idx,mat->rmap->n*sizeof(PetscInt));CHKERRQ(ierr);}
3978 
3979   ierr = (*mat->ops->getrowminabs)(mat,v,idx);CHKERRQ(ierr);
3980   ierr = PetscObjectStateIncrease((PetscObject)v);CHKERRQ(ierr);
3981   PetscFunctionReturn(0);
3982 }
3983 
3984 #undef __FUNCT__
3985 #define __FUNCT__ "MatGetRowMax"
3986 /*@
3987    MatGetRowMax - Gets the maximum value (of the real part) of each
3988         row of the matrix
3989 
3990    Logically Collective on Mat and Vec
3991 
3992    Input Parameters:
3993 .  mat - the matrix
3994 
3995    Output Parameter:
3996 +  v - the vector for storing the maximums
3997 -  idx - the indices of the column found for each row (optional)
3998 
3999    Level: intermediate
4000 
4001    Notes: The result of this call are the same as if one converted the matrix to dense format
4002       and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
4003 
4004     This code is only implemented for a couple of matrix formats.
4005 
4006    Concepts: matrices^getting row maximums
4007 
4008 .seealso: MatGetDiagonal(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMaxAbs(), MatGetRowMin()
4009 @*/
4010 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowMax(Mat mat,Vec v,PetscInt idx[])
4011 {
4012   PetscErrorCode ierr;
4013 
4014   PetscFunctionBegin;
4015   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4016   PetscValidType(mat,1);
4017   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
4018   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4019   if (!mat->ops->getrowmax) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4020   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4021 
4022   ierr = (*mat->ops->getrowmax)(mat,v,idx);CHKERRQ(ierr);
4023   ierr = PetscObjectStateIncrease((PetscObject)v);CHKERRQ(ierr);
4024   PetscFunctionReturn(0);
4025 }
4026 
4027 #undef __FUNCT__
4028 #define __FUNCT__ "MatGetRowMaxAbs"
4029 /*@
4030    MatGetRowMaxAbs - Gets the maximum value (in absolute value) of each
4031         row of the matrix
4032 
4033    Logically Collective on Mat and Vec
4034 
4035    Input Parameters:
4036 .  mat - the matrix
4037 
4038    Output Parameter:
4039 +  v - the vector for storing the maximums
4040 -  idx - the indices of the column found for each row (optional)
4041 
4042    Level: intermediate
4043 
4044    Notes: if a row is completely empty or has only 0.0 values then the idx[] value for that
4045     row is 0 (the first column).
4046 
4047     This code is only implemented for a couple of matrix formats.
4048 
4049    Concepts: matrices^getting row maximums
4050 
4051 .seealso: MatGetDiagonal(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMax(), MatGetRowMin()
4052 @*/
4053 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowMaxAbs(Mat mat,Vec v,PetscInt idx[])
4054 {
4055   PetscErrorCode ierr;
4056 
4057   PetscFunctionBegin;
4058   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4059   PetscValidType(mat,1);
4060   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
4061   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4062   if (!mat->ops->getrowmaxabs) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4063   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4064   if (idx) {ierr = PetscMemzero(idx,mat->rmap->n*sizeof(PetscInt));CHKERRQ(ierr);}
4065 
4066   ierr = (*mat->ops->getrowmaxabs)(mat,v,idx);CHKERRQ(ierr);
4067   ierr = PetscObjectStateIncrease((PetscObject)v);CHKERRQ(ierr);
4068   PetscFunctionReturn(0);
4069 }
4070 
4071 #undef __FUNCT__
4072 #define __FUNCT__ "MatGetRowSum"
4073 /*@
4074    MatGetRowSum - Gets the sum of each row of the matrix
4075 
4076    Logically Collective on Mat and Vec
4077 
4078    Input Parameters:
4079 .  mat - the matrix
4080 
4081    Output Parameter:
4082 .  v - the vector for storing the sum of rows
4083 
4084    Level: intermediate
4085 
4086    Notes: This code is slow since it is not currently specialized for different formats
4087 
4088    Concepts: matrices^getting row sums
4089 
4090 .seealso: MatGetDiagonal(), MatGetSubMatrices(), MatGetSubmatrix(), MatGetRowMax(), MatGetRowMin()
4091 @*/
4092 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowSum(Mat mat, Vec v)
4093 {
4094   PetscInt       start = 0, end = 0, row;
4095   PetscScalar   *array;
4096   PetscErrorCode ierr;
4097 
4098   PetscFunctionBegin;
4099   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4100   PetscValidType(mat,1);
4101   PetscValidHeaderSpecific(v,VEC_CLASSID,2);
4102   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4103   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4104   ierr = MatGetOwnershipRange(mat, &start, &end);CHKERRQ(ierr);
4105   ierr = VecGetArray(v, &array);CHKERRQ(ierr);
4106   for(row = start; row < end; ++row) {
4107     PetscInt           ncols, col;
4108     const PetscInt    *cols;
4109     const PetscScalar *vals;
4110 
4111     array[row - start] = 0.0;
4112     ierr = MatGetRow(mat, row, &ncols, &cols, &vals);CHKERRQ(ierr);
4113     for(col = 0; col < ncols; col++) {
4114       array[row - start] += vals[col];
4115     }
4116     ierr = MatRestoreRow(mat, row, &ncols, &cols, &vals);CHKERRQ(ierr);
4117   }
4118   ierr = VecRestoreArray(v, &array);CHKERRQ(ierr);
4119   ierr = PetscObjectStateIncrease((PetscObject) v);CHKERRQ(ierr);
4120   PetscFunctionReturn(0);
4121 }
4122 
4123 #undef __FUNCT__
4124 #define __FUNCT__ "MatTranspose"
4125 /*@
4126    MatTranspose - Computes an in-place or out-of-place transpose of a matrix.
4127 
4128    Collective on Mat
4129 
4130    Input Parameter:
4131 +  mat - the matrix to transpose
4132 -  reuse - store the transpose matrix in the provided B
4133 
4134    Output Parameters:
4135 .  B - the transpose
4136 
4137    Notes:
4138      If you  pass in &mat for B the transpose will be done in place
4139 
4140    Level: intermediate
4141 
4142    Concepts: matrices^transposing
4143 
4144 .seealso: MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
4145 @*/
4146 PetscErrorCode PETSCMAT_DLLEXPORT MatTranspose(Mat mat,MatReuse reuse,Mat *B)
4147 {
4148   PetscErrorCode ierr;
4149 
4150   PetscFunctionBegin;
4151   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4152   PetscValidType(mat,1);
4153   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4154   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4155   if (!mat->ops->transpose) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4156   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4157 
4158   ierr = PetscLogEventBegin(MAT_Transpose,mat,0,0,0);CHKERRQ(ierr);
4159   ierr = (*mat->ops->transpose)(mat,reuse,B);CHKERRQ(ierr);
4160   ierr = PetscLogEventEnd(MAT_Transpose,mat,0,0,0);CHKERRQ(ierr);
4161   if (B) {ierr = PetscObjectStateIncrease((PetscObject)*B);CHKERRQ(ierr);}
4162   PetscFunctionReturn(0);
4163 }
4164 
4165 #undef __FUNCT__
4166 #define __FUNCT__ "MatIsTranspose"
4167 /*@
4168    MatIsTranspose - Test whether a matrix is another one's transpose,
4169         or its own, in which case it tests symmetry.
4170 
4171    Collective on Mat
4172 
4173    Input Parameter:
4174 +  A - the matrix to test
4175 -  B - the matrix to test against, this can equal the first parameter
4176 
4177    Output Parameters:
4178 .  flg - the result
4179 
4180    Notes:
4181    Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
4182    has a running time of the order of the number of nonzeros; the parallel
4183    test involves parallel copies of the block-offdiagonal parts of the matrix.
4184 
4185    Level: intermediate
4186 
4187    Concepts: matrices^transposing, matrix^symmetry
4188 
4189 .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian()
4190 @*/
4191 PetscErrorCode PETSCMAT_DLLEXPORT MatIsTranspose(Mat A,Mat B,PetscReal tol,PetscTruth *flg)
4192 {
4193   PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscTruth*),(*g)(Mat,Mat,PetscReal,PetscTruth*);
4194 
4195   PetscFunctionBegin;
4196   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
4197   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
4198   PetscValidPointer(flg,3);
4199   ierr = PetscObjectQueryFunction((PetscObject)A,"MatIsTranspose_C",(void (**)(void))&f);CHKERRQ(ierr);
4200   ierr = PetscObjectQueryFunction((PetscObject)B,"MatIsTranspose_C",(void (**)(void))&g);CHKERRQ(ierr);
4201   if (f && g) {
4202     if (f==g) {
4203       ierr = (*f)(A,B,tol,flg);CHKERRQ(ierr);
4204     } else {
4205       SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for symmetry test");
4206     }
4207   }
4208   PetscFunctionReturn(0);
4209 }
4210 
4211 #undef __FUNCT__
4212 #define __FUNCT__ "MatHermitianTranspose"
4213 /*@
4214    MatHermitianTranspose - Computes an in-place or out-of-place transpose of a matrix in complex conjugate.
4215 
4216    Collective on Mat
4217 
4218    Input Parameter:
4219 +  mat - the matrix to transpose and complex conjugate
4220 -  reuse - store the transpose matrix in the provided B
4221 
4222    Output Parameters:
4223 .  B - the Hermitian
4224 
4225    Notes:
4226      If you  pass in &mat for B the Hermitian will be done in place
4227 
4228    Level: intermediate
4229 
4230    Concepts: matrices^transposing, complex conjugatex
4231 
4232 .seealso: MatTranspose(), MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
4233 @*/
4234 PetscErrorCode PETSCMAT_DLLEXPORT MatHermitianTranspose(Mat mat,MatReuse reuse,Mat *B)
4235 {
4236   PetscErrorCode ierr;
4237 
4238   PetscFunctionBegin;
4239   ierr = MatTranspose(mat,reuse,B);CHKERRQ(ierr);
4240 #if defined(PETSC_USE_COMPLEX)
4241   ierr = MatConjugate(*B);CHKERRQ(ierr);
4242 #endif
4243   PetscFunctionReturn(0);
4244 }
4245 
4246 #undef __FUNCT__
4247 #define __FUNCT__ "MatIsHermitianTranspose"
4248 /*@
4249    MatIsHermitianTranspose - Test whether a matrix is another one's Hermitian transpose,
4250 
4251    Collective on Mat
4252 
4253    Input Parameter:
4254 +  A - the matrix to test
4255 -  B - the matrix to test against, this can equal the first parameter
4256 
4257    Output Parameters:
4258 .  flg - the result
4259 
4260    Notes:
4261    Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
4262    has a running time of the order of the number of nonzeros; the parallel
4263    test involves parallel copies of the block-offdiagonal parts of the matrix.
4264 
4265    Level: intermediate
4266 
4267    Concepts: matrices^transposing, matrix^symmetry
4268 
4269 .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian(), MatIsTranspose()
4270 @*/
4271 PetscErrorCode PETSCMAT_DLLEXPORT MatIsHermitianTranspose(Mat A,Mat B,PetscReal tol,PetscTruth *flg)
4272 {
4273   PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscTruth*),(*g)(Mat,Mat,PetscReal,PetscTruth*);
4274 
4275   PetscFunctionBegin;
4276   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
4277   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
4278   PetscValidPointer(flg,3);
4279   ierr = PetscObjectQueryFunction((PetscObject)A,"MatIsHermitianTranspose_C",(void (**)(void))&f);CHKERRQ(ierr);
4280   ierr = PetscObjectQueryFunction((PetscObject)B,"MatIsHermitianTranspose_C",(void (**)(void))&g);CHKERRQ(ierr);
4281   if (f && g) {
4282     if (f==g) {
4283       ierr = (*f)(A,B,tol,flg);CHKERRQ(ierr);
4284     } else {
4285       SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for Hermitian test");
4286     }
4287   }
4288   PetscFunctionReturn(0);
4289 }
4290 
4291 #undef __FUNCT__
4292 #define __FUNCT__ "MatPermute"
4293 /*@
4294    MatPermute - Creates a new matrix with rows and columns permuted from the
4295    original.
4296 
4297    Collective on Mat
4298 
4299    Input Parameters:
4300 +  mat - the matrix to permute
4301 .  row - row permutation, each processor supplies only the permutation for its rows
4302 -  col - column permutation, each processor needs the entire column permutation, that is
4303          this is the same size as the total number of columns in the matrix. It can often
4304          be obtained with ISAllGather() on the row permutation
4305 
4306    Output Parameters:
4307 .  B - the permuted matrix
4308 
4309    Level: advanced
4310 
4311    Concepts: matrices^permuting
4312 
4313 .seealso: MatGetOrdering(), ISAllGather()
4314 
4315 @*/
4316 PetscErrorCode PETSCMAT_DLLEXPORT MatPermute(Mat mat,IS row,IS col,Mat *B)
4317 {
4318   PetscErrorCode ierr;
4319 
4320   PetscFunctionBegin;
4321   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4322   PetscValidType(mat,1);
4323   PetscValidHeaderSpecific(row,IS_CLASSID,2);
4324   PetscValidHeaderSpecific(col,IS_CLASSID,3);
4325   PetscValidPointer(B,4);
4326   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4327   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4328   if (!mat->ops->permute) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatPermute not available for Mat type %s",((PetscObject)mat)->type_name);
4329   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4330 
4331   ierr = (*mat->ops->permute)(mat,row,col,B);CHKERRQ(ierr);
4332   ierr = PetscObjectStateIncrease((PetscObject)*B);CHKERRQ(ierr);
4333   PetscFunctionReturn(0);
4334 }
4335 
4336 #undef __FUNCT__
4337 #define __FUNCT__ "MatPermuteSparsify"
4338 /*@
4339   MatPermuteSparsify - Creates a new matrix with rows and columns permuted from the
4340   original and sparsified to the prescribed tolerance.
4341 
4342   Collective on Mat
4343 
4344   Input Parameters:
4345 + A    - The matrix to permute
4346 . band - The half-bandwidth of the sparsified matrix, or PETSC_DECIDE
4347 . frac - The half-bandwidth as a fraction of the total size, or 0.0
4348 . tol  - The drop tolerance
4349 . rowp - The row permutation
4350 - colp - The column permutation
4351 
4352   Output Parameter:
4353 . B    - The permuted, sparsified matrix
4354 
4355   Level: advanced
4356 
4357   Note:
4358   The default behavior (band = PETSC_DECIDE and frac = 0.0) is to
4359   restrict the half-bandwidth of the resulting matrix to 5% of the
4360   total matrix size.
4361 
4362 .keywords: matrix, permute, sparsify
4363 
4364 .seealso: MatGetOrdering(), MatPermute()
4365 @*/
4366 PetscErrorCode PETSCMAT_DLLEXPORT MatPermuteSparsify(Mat A, PetscInt band, PetscReal frac, PetscReal tol, IS rowp, IS colp, Mat *B)
4367 {
4368   IS                irowp, icolp;
4369   const PetscInt    *rows, *cols;
4370   PetscInt          M, N, locRowStart = 0, locRowEnd = 0;
4371   PetscInt          nz, newNz;
4372   const PetscInt    *cwork;
4373   PetscInt          *cnew;
4374   const PetscScalar *vwork;
4375   PetscScalar       *vnew;
4376   PetscInt          bw, issize;
4377   PetscInt          row, locRow, newRow, col, newCol;
4378   PetscErrorCode    ierr;
4379 
4380   PetscFunctionBegin;
4381   PetscValidHeaderSpecific(A,    MAT_CLASSID,1);
4382   PetscValidHeaderSpecific(rowp, IS_CLASSID,5);
4383   PetscValidHeaderSpecific(colp, IS_CLASSID,6);
4384   PetscValidPointer(B,7);
4385   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE, "Not for unassembled matrix");
4386   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE, "Not for factored matrix");
4387   if (!A->ops->permutesparsify) {
4388     ierr = MatGetSize(A, &M, &N);CHKERRQ(ierr);
4389     ierr = MatGetOwnershipRange(A, &locRowStart, &locRowEnd);CHKERRQ(ierr);
4390     ierr = ISGetSize(rowp, &issize);CHKERRQ(ierr);
4391     if (issize != M) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG, "Wrong size %D for row permutation, should be %D", issize, M);
4392     ierr = ISGetSize(colp, &issize);CHKERRQ(ierr);
4393     if (issize != N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG, "Wrong size %D for column permutation, should be %D", issize, N);
4394     ierr = ISInvertPermutation(rowp, 0, &irowp);CHKERRQ(ierr);
4395     ierr = ISGetIndices(irowp, &rows);CHKERRQ(ierr);
4396     ierr = ISInvertPermutation(colp, 0, &icolp);CHKERRQ(ierr);
4397     ierr = ISGetIndices(icolp, &cols);CHKERRQ(ierr);
4398     ierr = PetscMalloc(N*sizeof(PetscInt),&cnew);CHKERRQ(ierr);
4399     ierr = PetscMalloc(N*sizeof(PetscScalar),&vnew);CHKERRQ(ierr);
4400 
4401     /* Setup bandwidth to include */
4402     if (band == PETSC_DECIDE) {
4403       if (frac <= 0.0)
4404         bw = (PetscInt) (M * 0.05);
4405       else
4406         bw = (PetscInt) (M * frac);
4407     } else {
4408       if (band <= 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG, "Bandwidth must be a positive integer");
4409       bw = band;
4410     }
4411 
4412     /* Put values into new matrix */
4413     ierr = MatDuplicate(A, MAT_DO_NOT_COPY_VALUES, B);CHKERRQ(ierr);
4414     for(row = locRowStart, locRow = 0; row < locRowEnd; row++, locRow++) {
4415       ierr = MatGetRow(A, row, &nz, &cwork, &vwork);CHKERRQ(ierr);
4416       newRow   = rows[locRow]+locRowStart;
4417       for(col = 0, newNz = 0; col < nz; col++) {
4418         newCol = cols[cwork[col]];
4419         if ((newCol >= newRow - bw) && (newCol < newRow + bw) && (PetscAbsScalar(vwork[col]) >= tol)) {
4420           cnew[newNz] = newCol;
4421           vnew[newNz] = vwork[col];
4422           newNz++;
4423         }
4424       }
4425       ierr = MatSetValues(*B, 1, &newRow, newNz, cnew, vnew, INSERT_VALUES);CHKERRQ(ierr);
4426       ierr = MatRestoreRow(A, row, &nz, &cwork, &vwork);CHKERRQ(ierr);
4427     }
4428     ierr = PetscFree(cnew);CHKERRQ(ierr);
4429     ierr = PetscFree(vnew);CHKERRQ(ierr);
4430     ierr = MatAssemblyBegin(*B, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr);
4431     ierr = MatAssemblyEnd(*B, MAT_FINAL_ASSEMBLY);CHKERRQ(ierr);
4432     ierr = ISRestoreIndices(irowp, &rows);CHKERRQ(ierr);
4433     ierr = ISRestoreIndices(icolp, &cols);CHKERRQ(ierr);
4434     ierr = ISDestroy(irowp);CHKERRQ(ierr);
4435     ierr = ISDestroy(icolp);CHKERRQ(ierr);
4436   } else {
4437     ierr = (*A->ops->permutesparsify)(A, band, frac, tol, rowp, colp, B);CHKERRQ(ierr);
4438   }
4439   ierr = PetscObjectStateIncrease((PetscObject)*B);CHKERRQ(ierr);
4440   PetscFunctionReturn(0);
4441 }
4442 
4443 #undef __FUNCT__
4444 #define __FUNCT__ "MatEqual"
4445 /*@
4446    MatEqual - Compares two matrices.
4447 
4448    Collective on Mat
4449 
4450    Input Parameters:
4451 +  A - the first matrix
4452 -  B - the second matrix
4453 
4454    Output Parameter:
4455 .  flg - PETSC_TRUE if the matrices are equal; PETSC_FALSE otherwise.
4456 
4457    Level: intermediate
4458 
4459    Concepts: matrices^equality between
4460 @*/
4461 PetscErrorCode PETSCMAT_DLLEXPORT MatEqual(Mat A,Mat B,PetscTruth *flg)
4462 {
4463   PetscErrorCode ierr;
4464 
4465   PetscFunctionBegin;
4466   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
4467   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
4468   PetscValidType(A,1);
4469   PetscValidType(B,2);
4470   PetscValidIntPointer(flg,3);
4471   PetscCheckSameComm(A,1,B,2);
4472   ierr = MatPreallocated(B);CHKERRQ(ierr);
4473   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4474   if (!B->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4475   if (A->rmap->N != B->rmap->N || A->cmap->N != B->cmap->N) SETERRQ4(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Mat A,Mat B: global dim %D %D %D %D",A->rmap->N,B->rmap->N,A->cmap->N,B->cmap->N);
4476   if (!A->ops->equal) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
4477   if (!B->ops->equal) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)B)->type_name);
4478   if (A->ops->equal != B->ops->equal) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_INCOMP,"A is type: %s\nB is type: %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
4479   ierr = MatPreallocated(A);CHKERRQ(ierr);
4480 
4481   ierr = (*A->ops->equal)(A,B,flg);CHKERRQ(ierr);
4482   PetscFunctionReturn(0);
4483 }
4484 
4485 #undef __FUNCT__
4486 #define __FUNCT__ "MatDiagonalScale"
4487 /*@
4488    MatDiagonalScale - Scales a matrix on the left and right by diagonal
4489    matrices that are stored as vectors.  Either of the two scaling
4490    matrices can be PETSC_NULL.
4491 
4492    Collective on Mat
4493 
4494    Input Parameters:
4495 +  mat - the matrix to be scaled
4496 .  l - the left scaling vector (or PETSC_NULL)
4497 -  r - the right scaling vector (or PETSC_NULL)
4498 
4499    Notes:
4500    MatDiagonalScale() computes A = LAR, where
4501    L = a diagonal matrix (stored as a vector), R = a diagonal matrix (stored as a vector)
4502    The L scales the rows of the matrix, the R scales the columns of the matrix.
4503 
4504    Level: intermediate
4505 
4506    Concepts: matrices^diagonal scaling
4507    Concepts: diagonal scaling of matrices
4508 
4509 .seealso: MatScale()
4510 @*/
4511 PetscErrorCode PETSCMAT_DLLEXPORT MatDiagonalScale(Mat mat,Vec l,Vec r)
4512 {
4513   PetscErrorCode ierr;
4514 
4515   PetscFunctionBegin;
4516   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4517   PetscValidType(mat,1);
4518   if (!mat->ops->diagonalscale) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4519   if (l) {PetscValidHeaderSpecific(l,VEC_CLASSID,2);PetscCheckSameComm(mat,1,l,2);}
4520   if (r) {PetscValidHeaderSpecific(r,VEC_CLASSID,3);PetscCheckSameComm(mat,1,r,3);}
4521   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4522   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4523   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4524 
4525   ierr = PetscLogEventBegin(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
4526   ierr = (*mat->ops->diagonalscale)(mat,l,r);CHKERRQ(ierr);
4527   ierr = PetscLogEventEnd(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
4528   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
4529   PetscFunctionReturn(0);
4530 }
4531 
4532 #undef __FUNCT__
4533 #define __FUNCT__ "MatScale"
4534 /*@
4535     MatScale - Scales all elements of a matrix by a given number.
4536 
4537     Logically Collective on Mat
4538 
4539     Input Parameters:
4540 +   mat - the matrix to be scaled
4541 -   a  - the scaling value
4542 
4543     Output Parameter:
4544 .   mat - the scaled matrix
4545 
4546     Level: intermediate
4547 
4548     Concepts: matrices^scaling all entries
4549 
4550 .seealso: MatDiagonalScale()
4551 @*/
4552 PetscErrorCode PETSCMAT_DLLEXPORT MatScale(Mat mat,PetscScalar a)
4553 {
4554   PetscErrorCode ierr;
4555 
4556   PetscFunctionBegin;
4557   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4558   PetscValidType(mat,1);
4559   if (a != 1.0 && !mat->ops->scale) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4560   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4561   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4562   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4563 
4564   ierr = PetscLogEventBegin(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
4565   if (a != 1.0) {
4566     ierr = (*mat->ops->scale)(mat,a);CHKERRQ(ierr);
4567     ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
4568   }
4569   ierr = PetscLogEventEnd(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
4570   PetscFunctionReturn(0);
4571 }
4572 
4573 #undef __FUNCT__
4574 #define __FUNCT__ "MatNorm"
4575 /*@
4576    MatNorm - Calculates various norms of a matrix.
4577 
4578    Collective on Mat
4579 
4580    Input Parameters:
4581 +  mat - the matrix
4582 -  type - the type of norm, NORM_1, NORM_FROBENIUS, NORM_INFINITY
4583 
4584    Output Parameters:
4585 .  nrm - the resulting norm
4586 
4587    Level: intermediate
4588 
4589    Concepts: matrices^norm
4590    Concepts: norm^of matrix
4591 @*/
4592 PetscErrorCode PETSCMAT_DLLEXPORT MatNorm(Mat mat,NormType type,PetscReal *nrm)
4593 {
4594   PetscErrorCode ierr;
4595 
4596   PetscFunctionBegin;
4597   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4598   PetscValidType(mat,1);
4599   PetscValidScalarPointer(nrm,3);
4600 
4601   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4602   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4603   if (!mat->ops->norm) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4604   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4605 
4606   ierr = (*mat->ops->norm)(mat,type,nrm);CHKERRQ(ierr);
4607   PetscFunctionReturn(0);
4608 }
4609 
4610 /*
4611      This variable is used to prevent counting of MatAssemblyBegin() that
4612    are called from within a MatAssemblyEnd().
4613 */
4614 static PetscInt MatAssemblyEnd_InUse = 0;
4615 #undef __FUNCT__
4616 #define __FUNCT__ "MatAssemblyBegin"
4617 /*@
4618    MatAssemblyBegin - Begins assembling the matrix.  This routine should
4619    be called after completing all calls to MatSetValues().
4620 
4621    Collective on Mat
4622 
4623    Input Parameters:
4624 +  mat - the matrix
4625 -  type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
4626 
4627    Notes:
4628    MatSetValues() generally caches the values.  The matrix is ready to
4629    use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
4630    Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
4631    in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
4632    using the matrix.
4633 
4634    Level: beginner
4635 
4636    Concepts: matrices^assembling
4637 
4638 .seealso: MatAssemblyEnd(), MatSetValues(), MatAssembled()
4639 @*/
4640 PetscErrorCode PETSCMAT_DLLEXPORT MatAssemblyBegin(Mat mat,MatAssemblyType type)
4641 {
4642   PetscErrorCode ierr;
4643 
4644   PetscFunctionBegin;
4645   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4646   PetscValidType(mat,1);
4647   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4648   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix.\nDid you forget to call MatSetUnfactored()?");
4649   if (mat->assembled) {
4650     mat->was_assembled = PETSC_TRUE;
4651     mat->assembled     = PETSC_FALSE;
4652   }
4653   if (!MatAssemblyEnd_InUse) {
4654     ierr = PetscLogEventBegin(MAT_AssemblyBegin,mat,0,0,0);CHKERRQ(ierr);
4655     if (mat->ops->assemblybegin){ierr = (*mat->ops->assemblybegin)(mat,type);CHKERRQ(ierr);}
4656     ierr = PetscLogEventEnd(MAT_AssemblyBegin,mat,0,0,0);CHKERRQ(ierr);
4657   } else {
4658     if (mat->ops->assemblybegin){ierr = (*mat->ops->assemblybegin)(mat,type);CHKERRQ(ierr);}
4659   }
4660   PetscFunctionReturn(0);
4661 }
4662 
4663 #undef __FUNCT__
4664 #define __FUNCT__ "MatAssembled"
4665 /*@
4666    MatAssembled - Indicates if a matrix has been assembled and is ready for
4667      use; for example, in matrix-vector product.
4668 
4669    Not Collective
4670 
4671    Input Parameter:
4672 .  mat - the matrix
4673 
4674    Output Parameter:
4675 .  assembled - PETSC_TRUE or PETSC_FALSE
4676 
4677    Level: advanced
4678 
4679    Concepts: matrices^assembled?
4680 
4681 .seealso: MatAssemblyEnd(), MatSetValues(), MatAssemblyBegin()
4682 @*/
4683 PetscErrorCode PETSCMAT_DLLEXPORT MatAssembled(Mat mat,PetscTruth *assembled)
4684 {
4685   PetscFunctionBegin;
4686   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4687   PetscValidType(mat,1);
4688   PetscValidPointer(assembled,2);
4689   *assembled = mat->assembled;
4690   PetscFunctionReturn(0);
4691 }
4692 
4693 #undef __FUNCT__
4694 #define __FUNCT__ "MatView_Private"
4695 /*
4696     Processes command line options to determine if/how a matrix
4697   is to be viewed. Called by MatAssemblyEnd() and MatLoad().
4698 */
4699 PetscErrorCode MatView_Private(Mat mat)
4700 {
4701   PetscErrorCode    ierr;
4702   PetscTruth        flg1 = PETSC_FALSE,flg2 = PETSC_FALSE,flg3 = PETSC_FALSE,flg4 = PETSC_FALSE,flg6 = PETSC_FALSE,flg7 = PETSC_FALSE,flg8 = PETSC_FALSE;
4703   static PetscTruth incall = PETSC_FALSE;
4704 #if defined(PETSC_USE_SOCKET_VIEWER)
4705   PetscTruth        flg5 = PETSC_FALSE;
4706 #endif
4707 
4708   PetscFunctionBegin;
4709   if (incall) PetscFunctionReturn(0);
4710   incall = PETSC_TRUE;
4711   ierr = PetscOptionsBegin(((PetscObject)mat)->comm,((PetscObject)mat)->prefix,"Matrix Options","Mat");CHKERRQ(ierr);
4712     ierr = PetscOptionsTruth("-mat_view_info","Information on matrix size","MatView",flg1,&flg1,PETSC_NULL);CHKERRQ(ierr);
4713     ierr = PetscOptionsTruth("-mat_view_info_detailed","Nonzeros in the matrix","MatView",flg2,&flg2,PETSC_NULL);CHKERRQ(ierr);
4714     ierr = PetscOptionsTruth("-mat_view","Print matrix to stdout","MatView",flg3,&flg3,PETSC_NULL);CHKERRQ(ierr);
4715     ierr = PetscOptionsTruth("-mat_view_matlab","Print matrix to stdout in a format Matlab can read","MatView",flg4,&flg4,PETSC_NULL);CHKERRQ(ierr);
4716 #if defined(PETSC_USE_SOCKET_VIEWER)
4717     ierr = PetscOptionsTruth("-mat_view_socket","Send matrix to socket (can be read from matlab)","MatView",flg5,&flg5,PETSC_NULL);CHKERRQ(ierr);
4718 #endif
4719     ierr = PetscOptionsTruth("-mat_view_binary","Save matrix to file in binary format","MatView",flg6,&flg6,PETSC_NULL);CHKERRQ(ierr);
4720     ierr = PetscOptionsTruth("-mat_view_draw","Draw the matrix nonzero structure","MatView",flg7,&flg7,PETSC_NULL);CHKERRQ(ierr);
4721   ierr = PetscOptionsEnd();CHKERRQ(ierr);
4722 
4723   if (flg1) {
4724     PetscViewer viewer;
4725 
4726     ierr = PetscViewerASCIIGetStdout(((PetscObject)mat)->comm,&viewer);CHKERRQ(ierr);
4727     ierr = PetscViewerPushFormat(viewer,PETSC_VIEWER_ASCII_INFO);CHKERRQ(ierr);
4728     ierr = MatView(mat,viewer);CHKERRQ(ierr);
4729     ierr = PetscViewerPopFormat(viewer);CHKERRQ(ierr);
4730   }
4731   if (flg2) {
4732     PetscViewer viewer;
4733 
4734     ierr = PetscViewerASCIIGetStdout(((PetscObject)mat)->comm,&viewer);CHKERRQ(ierr);
4735     ierr = PetscViewerPushFormat(viewer,PETSC_VIEWER_ASCII_INFO_DETAIL);CHKERRQ(ierr);
4736     ierr = MatView(mat,viewer);CHKERRQ(ierr);
4737     ierr = PetscViewerPopFormat(viewer);CHKERRQ(ierr);
4738   }
4739   if (flg3) {
4740     PetscViewer viewer;
4741 
4742     ierr = PetscViewerASCIIGetStdout(((PetscObject)mat)->comm,&viewer);CHKERRQ(ierr);
4743     ierr = MatView(mat,viewer);CHKERRQ(ierr);
4744   }
4745   if (flg4) {
4746     PetscViewer viewer;
4747 
4748     ierr = PetscViewerASCIIGetStdout(((PetscObject)mat)->comm,&viewer);CHKERRQ(ierr);
4749     ierr = PetscViewerPushFormat(viewer,PETSC_VIEWER_ASCII_MATLAB);CHKERRQ(ierr);
4750     ierr = MatView(mat,viewer);CHKERRQ(ierr);
4751     ierr = PetscViewerPopFormat(viewer);CHKERRQ(ierr);
4752   }
4753 #if defined(PETSC_USE_SOCKET_VIEWER)
4754   if (flg5) {
4755     ierr = MatView(mat,PETSC_VIEWER_SOCKET_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4756     ierr = PetscViewerFlush(PETSC_VIEWER_SOCKET_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4757   }
4758 #endif
4759   if (flg6) {
4760     ierr = MatView(mat,PETSC_VIEWER_BINARY_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4761     ierr = PetscViewerFlush(PETSC_VIEWER_BINARY_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4762   }
4763   if (flg7) {
4764     ierr = PetscOptionsGetTruth(((PetscObject)mat)->prefix,"-mat_view_contour",&flg8,PETSC_NULL);CHKERRQ(ierr);
4765     if (flg8) {
4766       PetscViewerPushFormat(PETSC_VIEWER_DRAW_(((PetscObject)mat)->comm),PETSC_VIEWER_DRAW_CONTOUR);CHKERRQ(ierr);
4767     }
4768     ierr = MatView(mat,PETSC_VIEWER_DRAW_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4769     ierr = PetscViewerFlush(PETSC_VIEWER_DRAW_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4770     if (flg8) {
4771       PetscViewerPopFormat(PETSC_VIEWER_DRAW_(((PetscObject)mat)->comm));CHKERRQ(ierr);
4772     }
4773   }
4774   incall = PETSC_FALSE;
4775   PetscFunctionReturn(0);
4776 }
4777 
4778 #undef __FUNCT__
4779 #define __FUNCT__ "MatAssemblyEnd"
4780 /*@
4781    MatAssemblyEnd - Completes assembling the matrix.  This routine should
4782    be called after MatAssemblyBegin().
4783 
4784    Collective on Mat
4785 
4786    Input Parameters:
4787 +  mat - the matrix
4788 -  type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
4789 
4790    Options Database Keys:
4791 +  -mat_view_info - Prints info on matrix at conclusion of MatEndAssembly()
4792 .  -mat_view_info_detailed - Prints more detailed info
4793 .  -mat_view - Prints matrix in ASCII format
4794 .  -mat_view_matlab - Prints matrix in Matlab format
4795 .  -mat_view_draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
4796 .  -display <name> - Sets display name (default is host)
4797 .  -draw_pause <sec> - Sets number of seconds to pause after display
4798 .  -mat_view_socket - Sends matrix to socket, can be accessed from Matlab (see users manual)
4799 .  -viewer_socket_machine <machine>
4800 .  -viewer_socket_port <port>
4801 .  -mat_view_binary - save matrix to file in binary format
4802 -  -viewer_binary_filename <name>
4803 
4804    Notes:
4805    MatSetValues() generally caches the values.  The matrix is ready to
4806    use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
4807    Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
4808    in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
4809    using the matrix.
4810 
4811    Level: beginner
4812 
4813 .seealso: MatAssemblyBegin(), MatSetValues(), PetscDrawOpenX(), MatView(), MatAssembled(), PetscViewerSocketOpen()
4814 @*/
4815 PetscErrorCode PETSCMAT_DLLEXPORT MatAssemblyEnd(Mat mat,MatAssemblyType type)
4816 {
4817   PetscErrorCode  ierr;
4818   static PetscInt inassm = 0;
4819   PetscTruth      flg = PETSC_FALSE;
4820 
4821   PetscFunctionBegin;
4822   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4823   PetscValidType(mat,1);
4824 
4825   inassm++;
4826   MatAssemblyEnd_InUse++;
4827   if (MatAssemblyEnd_InUse == 1) { /* Do the logging only the first time through */
4828     ierr = PetscLogEventBegin(MAT_AssemblyEnd,mat,0,0,0);CHKERRQ(ierr);
4829     if (mat->ops->assemblyend) {
4830       ierr = (*mat->ops->assemblyend)(mat,type);CHKERRQ(ierr);
4831     }
4832     ierr = PetscLogEventEnd(MAT_AssemblyEnd,mat,0,0,0);CHKERRQ(ierr);
4833   } else {
4834     if (mat->ops->assemblyend) {
4835       ierr = (*mat->ops->assemblyend)(mat,type);CHKERRQ(ierr);
4836     }
4837   }
4838 
4839   /* Flush assembly is not a true assembly */
4840   if (type != MAT_FLUSH_ASSEMBLY) {
4841     mat->assembled  = PETSC_TRUE; mat->num_ass++;
4842   }
4843   mat->insertmode = NOT_SET_VALUES;
4844   MatAssemblyEnd_InUse--;
4845   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
4846   if (!mat->symmetric_eternal) {
4847     mat->symmetric_set              = PETSC_FALSE;
4848     mat->hermitian_set              = PETSC_FALSE;
4849     mat->structurally_symmetric_set = PETSC_FALSE;
4850   }
4851   if (inassm == 1 && type != MAT_FLUSH_ASSEMBLY) {
4852     ierr = MatView_Private(mat);CHKERRQ(ierr);
4853     ierr = PetscOptionsGetTruth(((PetscObject)mat)->prefix,"-mat_is_symmetric",&flg,PETSC_NULL);CHKERRQ(ierr);
4854     if (flg) {
4855       PetscReal tol = 0.0;
4856       ierr = PetscOptionsGetReal(((PetscObject)mat)->prefix,"-mat_is_symmetric",&tol,PETSC_NULL);CHKERRQ(ierr);
4857       ierr = MatIsSymmetric(mat,tol,&flg);CHKERRQ(ierr);
4858       if (flg) {
4859         ierr = PetscPrintf(((PetscObject)mat)->comm,"Matrix is symmetric (tolerance %G)\n",tol);CHKERRQ(ierr);
4860       } else {
4861         ierr = PetscPrintf(((PetscObject)mat)->comm,"Matrix is not symmetric (tolerance %G)\n",tol);CHKERRQ(ierr);
4862       }
4863     }
4864   }
4865   inassm--;
4866   PetscFunctionReturn(0);
4867 }
4868 
4869 #undef __FUNCT__
4870 #define __FUNCT__ "MatSetOption"
4871 /*@
4872    MatSetOption - Sets a parameter option for a matrix. Some options
4873    may be specific to certain storage formats.  Some options
4874    determine how values will be inserted (or added). Sorted,
4875    row-oriented input will generally assemble the fastest. The default
4876    is row-oriented, nonsorted input.
4877 
4878    Logically Collective on Mat
4879 
4880    Input Parameters:
4881 +  mat - the matrix
4882 .  option - the option, one of those listed below (and possibly others),
4883 -  flg - turn the option on (PETSC_TRUE) or off (PETSC_FALSE)
4884 
4885   Options Describing Matrix Structure:
4886 +    MAT_SPD - symmetric positive definite
4887 -    MAT_SYMMETRIC - symmetric in terms of both structure and value
4888 .    MAT_HERMITIAN - transpose is the complex conjugation
4889 .    MAT_STRUCTURALLY_SYMMETRIC - symmetric nonzero structure
4890 -    MAT_SYMMETRY_ETERNAL - if you would like the symmetry/Hermitian flag
4891                             you set to be kept with all future use of the matrix
4892                             including after MatAssemblyBegin/End() which could
4893                             potentially change the symmetry structure, i.e. you
4894                             KNOW the matrix will ALWAYS have the property you set.
4895 
4896 
4897    Options For Use with MatSetValues():
4898    Insert a logically dense subblock, which can be
4899 .    MAT_ROW_ORIENTED - row-oriented (default)
4900 
4901    Note these options reflect the data you pass in with MatSetValues(); it has
4902    nothing to do with how the data is stored internally in the matrix
4903    data structure.
4904 
4905    When (re)assembling a matrix, we can restrict the input for
4906    efficiency/debugging purposes.  These options include
4907 +    MAT_NEW_NONZERO_LOCATIONS - additional insertions will be
4908         allowed if they generate a new nonzero
4909 .    MAT_NEW_DIAGONALS - new diagonals will be allowed (for block diagonal format only)
4910 .    MAT_IGNORE_OFF_PROC_ENTRIES - drops off-processor entries
4911 .    MAT_NEW_NONZERO_LOCATION_ERR - generates an error for new matrix entry
4912 -    MAT_USE_HASH_TABLE - uses a hash table to speed up matrix assembly
4913 
4914    Notes:
4915    Some options are relevant only for particular matrix types and
4916    are thus ignored by others.  Other options are not supported by
4917    certain matrix types and will generate an error message if set.
4918 
4919    If using a Fortran 77 module to compute a matrix, one may need to
4920    use the column-oriented option (or convert to the row-oriented
4921    format).
4922 
4923    MAT_NEW_NONZERO_LOCATIONS set to PETSC_FALSE indicates that any add or insertion
4924    that would generate a new entry in the nonzero structure is instead
4925    ignored.  Thus, if memory has not alredy been allocated for this particular
4926    data, then the insertion is ignored. For dense matrices, in which
4927    the entire array is allocated, no entries are ever ignored.
4928    Set after the first MatAssemblyEnd()
4929 
4930    MAT_NEW_NONZERO_LOCATION_ERR indicates that any add or insertion
4931    that would generate a new entry in the nonzero structure instead produces
4932    an error. (Currently supported for AIJ and BAIJ formats only.)
4933    This is a useful flag when using SAME_NONZERO_PATTERN in calling
4934    KSPSetOperators() to ensure that the nonzero pattern truely does
4935    remain unchanged. Set after the first MatAssemblyEnd()
4936 
4937    MAT_NEW_NONZERO_ALLOCATION_ERR indicates that any add or insertion
4938    that would generate a new entry that has not been preallocated will
4939    instead produce an error. (Currently supported for AIJ and BAIJ formats
4940    only.) This is a useful flag when debugging matrix memory preallocation.
4941 
4942    MAT_IGNORE_OFF_PROC_ENTRIES indicates entries destined for
4943    other processors should be dropped, rather than stashed.
4944    This is useful if you know that the "owning" processor is also
4945    always generating the correct matrix entries, so that PETSc need
4946    not transfer duplicate entries generated on another processor.
4947 
4948    MAT_USE_HASH_TABLE indicates that a hash table be used to improve the
4949    searches during matrix assembly. When this flag is set, the hash table
4950    is created during the first Matrix Assembly. This hash table is
4951    used the next time through, during MatSetVaules()/MatSetVaulesBlocked()
4952    to improve the searching of indices. MAT_NEW_NONZERO_LOCATIONS flag
4953    should be used with MAT_USE_HASH_TABLE flag. This option is currently
4954    supported by MATMPIBAIJ format only.
4955 
4956    MAT_KEEP_NONZERO_PATTERN indicates when MatZeroRows() is called the zeroed entries
4957    are kept in the nonzero structure
4958 
4959    MAT_IGNORE_ZERO_ENTRIES - for AIJ/IS matrices this will stop zero values from creating
4960    a zero location in the matrix
4961 
4962    MAT_USE_INODES - indicates using inode version of the code - works with AIJ and
4963    ROWBS matrix types
4964 
4965    Level: intermediate
4966 
4967    Concepts: matrices^setting options
4968 
4969 @*/
4970 PetscErrorCode PETSCMAT_DLLEXPORT MatSetOption(Mat mat,MatOption op,PetscTruth flg)
4971 {
4972   PetscErrorCode ierr;
4973 
4974   PetscFunctionBegin;
4975   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
4976   PetscValidType(mat,1);
4977   if (((int) op) < 0 || ((int) op) >= NUM_MAT_OPTIONS) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Options %d is out of range",(int)op);
4978   if (!((PetscObject)mat)->type_name) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_TYPENOTSET,"Cannot set options until type and size have been set, see MatSetType() and MatSetSizes()");
4979   ierr = MatPreallocated(mat);CHKERRQ(ierr);
4980   switch (op) {
4981   case MAT_SPD:
4982     mat->spd_set                         = PETSC_TRUE;
4983     mat->spd                             = flg;
4984     if (flg) {
4985       mat->symmetric                     = PETSC_TRUE;
4986       mat->structurally_symmetric        = PETSC_TRUE;
4987       mat->symmetric_set                 = PETSC_TRUE;
4988       mat->structurally_symmetric_set    = PETSC_TRUE;
4989     }
4990     break;
4991   case MAT_SYMMETRIC:
4992     mat->symmetric                       = flg;
4993     if (flg) mat->structurally_symmetric = PETSC_TRUE;
4994     mat->symmetric_set                   = PETSC_TRUE;
4995     mat->structurally_symmetric_set      = flg;
4996     break;
4997   case MAT_HERMITIAN:
4998     mat->hermitian                       = flg;
4999     if (flg) mat->structurally_symmetric = PETSC_TRUE;
5000     mat->hermitian_set                   = PETSC_TRUE;
5001     mat->structurally_symmetric_set      = flg;
5002     break;
5003   case MAT_STRUCTURALLY_SYMMETRIC:
5004     mat->structurally_symmetric          = flg;
5005     mat->structurally_symmetric_set      = PETSC_TRUE;
5006     break;
5007   case MAT_SYMMETRY_ETERNAL:
5008     mat->symmetric_eternal               = flg;
5009     break;
5010   default:
5011     break;
5012   }
5013   if (mat->ops->setoption) {
5014     ierr = (*mat->ops->setoption)(mat,op,flg);CHKERRQ(ierr);
5015   }
5016   PetscFunctionReturn(0);
5017 }
5018 
5019 #undef __FUNCT__
5020 #define __FUNCT__ "MatZeroEntries"
5021 /*@
5022    MatZeroEntries - Zeros all entries of a matrix.  For sparse matrices
5023    this routine retains the old nonzero structure.
5024 
5025    Logically Collective on Mat
5026 
5027    Input Parameters:
5028 .  mat - the matrix
5029 
5030    Level: intermediate
5031 
5032    Concepts: matrices^zeroing
5033 
5034 .seealso: MatZeroRows()
5035 @*/
5036 PetscErrorCode PETSCMAT_DLLEXPORT MatZeroEntries(Mat mat)
5037 {
5038   PetscErrorCode ierr;
5039 
5040   PetscFunctionBegin;
5041   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5042   PetscValidType(mat,1);
5043   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5044   if (mat->insertmode != NOT_SET_VALUES) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for matrices where you have set values but not yet assembled");
5045   if (!mat->ops->zeroentries) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5046   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5047 
5048   ierr = PetscLogEventBegin(MAT_ZeroEntries,mat,0,0,0);CHKERRQ(ierr);
5049   ierr = (*mat->ops->zeroentries)(mat);CHKERRQ(ierr);
5050   ierr = PetscLogEventEnd(MAT_ZeroEntries,mat,0,0,0);CHKERRQ(ierr);
5051   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
5052   PetscFunctionReturn(0);
5053 }
5054 
5055 #undef __FUNCT__
5056 #define __FUNCT__ "MatZeroRows"
5057 /*@C
5058    MatZeroRows - Zeros all entries (except possibly the main diagonal)
5059    of a set of rows of a matrix.
5060 
5061    Collective on Mat
5062 
5063    Input Parameters:
5064 +  mat - the matrix
5065 .  numRows - the number of rows to remove
5066 .  rows - the global row indices
5067 -  diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5068 
5069    Notes:
5070    For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5071    but does not release memory.  For the dense and block diagonal
5072    formats this does not alter the nonzero structure.
5073 
5074    If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5075    of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5076    merely zeroed.
5077 
5078    The user can set a value in the diagonal entry (or for the AIJ and
5079    row formats can optionally remove the main diagonal entry from the
5080    nonzero structure as well, by passing 0.0 as the final argument).
5081 
5082    For the parallel case, all processes that share the matrix (i.e.,
5083    those in the communicator used for matrix creation) MUST call this
5084    routine, regardless of whether any rows being zeroed are owned by
5085    them.
5086 
5087    Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5088    list only rows local to itself).
5089 
5090    Level: intermediate
5091 
5092    Concepts: matrices^zeroing rows
5093 
5094 .seealso: MatZeroRowsIS(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption()
5095 @*/
5096 PetscErrorCode PETSCMAT_DLLEXPORT MatZeroRows(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag)
5097 {
5098   PetscErrorCode ierr;
5099 
5100   PetscFunctionBegin;
5101   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5102   PetscValidType(mat,1);
5103   if (numRows) PetscValidIntPointer(rows,3);
5104   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5105   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5106   if (!mat->ops->zerorows) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5107   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5108 
5109   ierr = (*mat->ops->zerorows)(mat,numRows,rows,diag);CHKERRQ(ierr);
5110   ierr = MatView_Private(mat);CHKERRQ(ierr);
5111   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
5112   PetscFunctionReturn(0);
5113 }
5114 
5115 #undef __FUNCT__
5116 #define __FUNCT__ "MatZeroRowsIS"
5117 /*@C
5118    MatZeroRowsIS - Zeros all entries (except possibly the main diagonal)
5119    of a set of rows of a matrix.
5120 
5121    Collective on Mat
5122 
5123    Input Parameters:
5124 +  mat - the matrix
5125 .  is - index set of rows to remove
5126 -  diag - value put in all diagonals of eliminated rows
5127 
5128    Notes:
5129    For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5130    but does not release memory.  For the dense and block diagonal
5131    formats this does not alter the nonzero structure.
5132 
5133    If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5134    of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5135    merely zeroed.
5136 
5137    The user can set a value in the diagonal entry (or for the AIJ and
5138    row formats can optionally remove the main diagonal entry from the
5139    nonzero structure as well, by passing 0.0 as the final argument).
5140 
5141    For the parallel case, all processes that share the matrix (i.e.,
5142    those in the communicator used for matrix creation) MUST call this
5143    routine, regardless of whether any rows being zeroed are owned by
5144    them.
5145 
5146    Each processor should list the rows that IT wants zeroed
5147 
5148    Level: intermediate
5149 
5150    Concepts: matrices^zeroing rows
5151 
5152 .seealso: MatZeroRows(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption()
5153 @*/
5154 PetscErrorCode PETSCMAT_DLLEXPORT MatZeroRowsIS(Mat mat,IS is,PetscScalar diag)
5155 {
5156   PetscInt       numRows;
5157   const PetscInt *rows;
5158   PetscErrorCode ierr;
5159 
5160   PetscFunctionBegin;
5161   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5162   PetscValidType(mat,1);
5163   PetscValidHeaderSpecific(is,IS_CLASSID,2);
5164   ierr = ISGetLocalSize(is,&numRows);CHKERRQ(ierr);
5165   ierr = ISGetIndices(is,&rows);CHKERRQ(ierr);
5166   ierr = MatZeroRows(mat,numRows,rows,diag);CHKERRQ(ierr);
5167   ierr = ISRestoreIndices(is,&rows);CHKERRQ(ierr);
5168   PetscFunctionReturn(0);
5169 }
5170 
5171 #undef __FUNCT__
5172 #define __FUNCT__ "MatZeroRowsLocal"
5173 /*@C
5174    MatZeroRowsLocal - Zeros all entries (except possibly the main diagonal)
5175    of a set of rows of a matrix; using local numbering of rows.
5176 
5177    Logically Collective on Mat
5178 
5179    Input Parameters:
5180 +  mat - the matrix
5181 .  numRows - the number of rows to remove
5182 .  rows - the global row indices
5183 -  diag - value put in all diagonals of eliminated rows
5184 
5185    Notes:
5186    Before calling MatZeroRowsLocal(), the user must first set the
5187    local-to-global mapping by calling MatSetLocalToGlobalMapping().
5188 
5189    For the AIJ matrix formats this removes the old nonzero structure,
5190    but does not release memory.  For the dense and block diagonal
5191    formats this does not alter the nonzero structure.
5192 
5193    If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5194    of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5195    merely zeroed.
5196 
5197    The user can set a value in the diagonal entry (or for the AIJ and
5198    row formats can optionally remove the main diagonal entry from the
5199    nonzero structure as well, by passing 0.0 as the final argument).
5200 
5201    Level: intermediate
5202 
5203    Concepts: matrices^zeroing
5204 
5205 .seealso: MatZeroRows(), MatZeroRowsLocalIS(), MatZeroEntries(), MatZeroRows(), MatSetLocalToGlobalMapping
5206 @*/
5207 PetscErrorCode PETSCMAT_DLLEXPORT MatZeroRowsLocal(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag)
5208 {
5209   PetscErrorCode ierr;
5210 
5211   PetscFunctionBegin;
5212   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5213   PetscValidType(mat,1);
5214   if (numRows) PetscValidIntPointer(rows,3);
5215   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5216   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5217   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5218 
5219   if (mat->ops->zerorowslocal) {
5220     ierr = (*mat->ops->zerorowslocal)(mat,numRows,rows,diag);CHKERRQ(ierr);
5221   } else {
5222     IS             is, newis;
5223     const PetscInt *newRows;
5224 
5225     if (!mat->mapping) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Need to provide local to global mapping to matrix first");
5226     ierr = ISCreateGeneral(PETSC_COMM_SELF,numRows,rows,&is);CHKERRQ(ierr);
5227     ierr = ISLocalToGlobalMappingApplyIS(mat->mapping,is,&newis);CHKERRQ(ierr);
5228     ierr = ISGetIndices(newis,&newRows);CHKERRQ(ierr);
5229     ierr = (*mat->ops->zerorows)(mat,numRows,newRows,diag);CHKERRQ(ierr);
5230     ierr = ISRestoreIndices(newis,&newRows);CHKERRQ(ierr);
5231     ierr = ISDestroy(newis);CHKERRQ(ierr);
5232     ierr = ISDestroy(is);CHKERRQ(ierr);
5233   }
5234   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
5235   PetscFunctionReturn(0);
5236 }
5237 
5238 #undef __FUNCT__
5239 #define __FUNCT__ "MatZeroRowsLocalIS"
5240 /*@C
5241    MatZeroRowsLocalIS - Zeros all entries (except possibly the main diagonal)
5242    of a set of rows of a matrix; using local numbering of rows.
5243 
5244    Logically Collective on Mat
5245 
5246    Input Parameters:
5247 +  mat - the matrix
5248 .  is - index set of rows to remove
5249 -  diag - value put in all diagonals of eliminated rows
5250 
5251    Notes:
5252    Before calling MatZeroRowsLocalIS(), the user must first set the
5253    local-to-global mapping by calling MatSetLocalToGlobalMapping().
5254 
5255    For the AIJ matrix formats this removes the old nonzero structure,
5256    but does not release memory.  For the dense and block diagonal
5257    formats this does not alter the nonzero structure.
5258 
5259    If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5260    of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5261    merely zeroed.
5262 
5263    The user can set a value in the diagonal entry (or for the AIJ and
5264    row formats can optionally remove the main diagonal entry from the
5265    nonzero structure as well, by passing 0.0 as the final argument).
5266 
5267    Level: intermediate
5268 
5269    Concepts: matrices^zeroing
5270 
5271 .seealso: MatZeroRows(), MatZeroRowsLocal(), MatZeroEntries(), MatZeroRows(), MatSetLocalToGlobalMapping
5272 @*/
5273 PetscErrorCode PETSCMAT_DLLEXPORT MatZeroRowsLocalIS(Mat mat,IS is,PetscScalar diag)
5274 {
5275   PetscErrorCode ierr;
5276   PetscInt       numRows;
5277   const PetscInt *rows;
5278 
5279   PetscFunctionBegin;
5280   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5281   PetscValidType(mat,1);
5282   PetscValidHeaderSpecific(is,IS_CLASSID,2);
5283   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5284   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5285   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5286 
5287   ierr = ISGetLocalSize(is,&numRows);CHKERRQ(ierr);
5288   ierr = ISGetIndices(is,&rows);CHKERRQ(ierr);
5289   ierr = MatZeroRowsLocal(mat,numRows,rows,diag);CHKERRQ(ierr);
5290   ierr = ISRestoreIndices(is,&rows);CHKERRQ(ierr);
5291   PetscFunctionReturn(0);
5292 }
5293 
5294 #undef __FUNCT__
5295 #define __FUNCT__ "MatGetSize"
5296 /*@
5297    MatGetSize - Returns the numbers of rows and columns in a matrix.
5298 
5299    Not Collective
5300 
5301    Input Parameter:
5302 .  mat - the matrix
5303 
5304    Output Parameters:
5305 +  m - the number of global rows
5306 -  n - the number of global columns
5307 
5308    Note: both output parameters can be PETSC_NULL on input.
5309 
5310    Level: beginner
5311 
5312    Concepts: matrices^size
5313 
5314 .seealso: MatGetLocalSize()
5315 @*/
5316 PetscErrorCode PETSCMAT_DLLEXPORT MatGetSize(Mat mat,PetscInt *m,PetscInt* n)
5317 {
5318   PetscFunctionBegin;
5319   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5320   if (m) *m = mat->rmap->N;
5321   if (n) *n = mat->cmap->N;
5322   PetscFunctionReturn(0);
5323 }
5324 
5325 #undef __FUNCT__
5326 #define __FUNCT__ "MatGetLocalSize"
5327 /*@
5328    MatGetLocalSize - Returns the number of rows and columns in a matrix
5329    stored locally.  This information may be implementation dependent, so
5330    use with care.
5331 
5332    Not Collective
5333 
5334    Input Parameters:
5335 .  mat - the matrix
5336 
5337    Output Parameters:
5338 +  m - the number of local rows
5339 -  n - the number of local columns
5340 
5341    Note: both output parameters can be PETSC_NULL on input.
5342 
5343    Level: beginner
5344 
5345    Concepts: matrices^local size
5346 
5347 .seealso: MatGetSize()
5348 @*/
5349 PetscErrorCode PETSCMAT_DLLEXPORT MatGetLocalSize(Mat mat,PetscInt *m,PetscInt* n)
5350 {
5351   PetscFunctionBegin;
5352   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5353   if (m) PetscValidIntPointer(m,2);
5354   if (n) PetscValidIntPointer(n,3);
5355   if (m) *m = mat->rmap->n;
5356   if (n) *n = mat->cmap->n;
5357   PetscFunctionReturn(0);
5358 }
5359 
5360 #undef __FUNCT__
5361 #define __FUNCT__ "MatGetOwnershipRangeColumn"
5362 /*@
5363    MatGetOwnershipRangeColumn - Returns the range of matrix columns owned by
5364    this processor.
5365 
5366    Not Collective, unless matrix has not been allocated, then collective on Mat
5367 
5368    Input Parameters:
5369 .  mat - the matrix
5370 
5371    Output Parameters:
5372 +  m - the global index of the first local column
5373 -  n - one more than the global index of the last local column
5374 
5375    Notes: both output parameters can be PETSC_NULL on input.
5376 
5377    Level: developer
5378 
5379    Concepts: matrices^column ownership
5380 
5381 .seealso:  MatGetOwnershipRange(), MatGetOwnershipRanges(), MatGetOwnershipRangesColumn()
5382 
5383 @*/
5384 PetscErrorCode PETSCMAT_DLLEXPORT MatGetOwnershipRangeColumn(Mat mat,PetscInt *m,PetscInt* n)
5385 {
5386   PetscErrorCode ierr;
5387 
5388   PetscFunctionBegin;
5389   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5390   PetscValidType(mat,1);
5391   if (m) PetscValidIntPointer(m,2);
5392   if (n) PetscValidIntPointer(n,3);
5393   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5394   if (m) *m = mat->cmap->rstart;
5395   if (n) *n = mat->cmap->rend;
5396   PetscFunctionReturn(0);
5397 }
5398 
5399 #undef __FUNCT__
5400 #define __FUNCT__ "MatGetOwnershipRange"
5401 /*@
5402    MatGetOwnershipRange - Returns the range of matrix rows owned by
5403    this processor, assuming that the matrix is laid out with the first
5404    n1 rows on the first processor, the next n2 rows on the second, etc.
5405    For certain parallel layouts this range may not be well defined.
5406 
5407    Not Collective, unless matrix has not been allocated, then collective on Mat
5408 
5409    Input Parameters:
5410 .  mat - the matrix
5411 
5412    Output Parameters:
5413 +  m - the global index of the first local row
5414 -  n - one more than the global index of the last local row
5415 
5416    Note: both output parameters can be PETSC_NULL on input.
5417 
5418    Level: beginner
5419 
5420    Concepts: matrices^row ownership
5421 
5422 .seealso:   MatGetOwnershipRanges(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn()
5423 
5424 @*/
5425 PetscErrorCode PETSCMAT_DLLEXPORT MatGetOwnershipRange(Mat mat,PetscInt *m,PetscInt* n)
5426 {
5427   PetscErrorCode ierr;
5428 
5429   PetscFunctionBegin;
5430   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5431   PetscValidType(mat,1);
5432   if (m) PetscValidIntPointer(m,2);
5433   if (n) PetscValidIntPointer(n,3);
5434   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5435   if (m) *m = mat->rmap->rstart;
5436   if (n) *n = mat->rmap->rend;
5437   PetscFunctionReturn(0);
5438 }
5439 
5440 #undef __FUNCT__
5441 #define __FUNCT__ "MatGetOwnershipRanges"
5442 /*@C
5443    MatGetOwnershipRanges - Returns the range of matrix rows owned by
5444    each process
5445 
5446    Not Collective, unless matrix has not been allocated, then collective on Mat
5447 
5448    Input Parameters:
5449 .  mat - the matrix
5450 
5451    Output Parameters:
5452 .  ranges - start of each processors portion plus one more then the total length at the end
5453 
5454    Level: beginner
5455 
5456    Concepts: matrices^row ownership
5457 
5458 .seealso:   MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn()
5459 
5460 @*/
5461 PetscErrorCode PETSCMAT_DLLEXPORT MatGetOwnershipRanges(Mat mat,const PetscInt **ranges)
5462 {
5463   PetscErrorCode ierr;
5464 
5465   PetscFunctionBegin;
5466   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5467   PetscValidType(mat,1);
5468   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5469   ierr = PetscLayoutGetRanges(mat->rmap,ranges);CHKERRQ(ierr);
5470   PetscFunctionReturn(0);
5471 }
5472 
5473 #undef __FUNCT__
5474 #define __FUNCT__ "MatGetOwnershipRangesColumn"
5475 /*@C
5476    MatGetOwnershipRangesColumn - Returns the range of local columns for each process
5477 
5478    Not Collective, unless matrix has not been allocated, then collective on Mat
5479 
5480    Input Parameters:
5481 .  mat - the matrix
5482 
5483    Output Parameters:
5484 .  ranges - start of each processors portion plus one more then the total length at the end
5485 
5486    Level: beginner
5487 
5488    Concepts: matrices^column ownership
5489 
5490 .seealso:   MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRanges()
5491 
5492 @*/
5493 PetscErrorCode PETSCMAT_DLLEXPORT MatGetOwnershipRangesColumn(Mat mat,const PetscInt **ranges)
5494 {
5495   PetscErrorCode ierr;
5496 
5497   PetscFunctionBegin;
5498   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5499   PetscValidType(mat,1);
5500   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5501   ierr = PetscLayoutGetRanges(mat->cmap,ranges);CHKERRQ(ierr);
5502   PetscFunctionReturn(0);
5503 }
5504 
5505 #undef __FUNCT__
5506 #define __FUNCT__ "MatILUFactorSymbolic"
5507 /*@C
5508    MatILUFactorSymbolic - Performs symbolic ILU factorization of a matrix.
5509    Uses levels of fill only, not drop tolerance. Use MatLUFactorNumeric()
5510    to complete the factorization.
5511 
5512    Collective on Mat
5513 
5514    Input Parameters:
5515 +  mat - the matrix
5516 .  row - row permutation
5517 .  column - column permutation
5518 -  info - structure containing
5519 $      levels - number of levels of fill.
5520 $      expected fill - as ratio of original fill.
5521 $      1 or 0 - indicating force fill on diagonal (improves robustness for matrices
5522                 missing diagonal entries)
5523 
5524    Output Parameters:
5525 .  fact - new matrix that has been symbolically factored
5526 
5527    Notes:
5528    See the users manual for additional information about
5529    choosing the fill factor for better efficiency.
5530 
5531    Most users should employ the simplified KSP interface for linear solvers
5532    instead of working directly with matrix algebra routines such as this.
5533    See, e.g., KSPCreate().
5534 
5535    Level: developer
5536 
5537   Concepts: matrices^symbolic LU factorization
5538   Concepts: matrices^factorization
5539   Concepts: LU^symbolic factorization
5540 
5541 .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
5542           MatGetOrdering(), MatFactorInfo
5543 
5544     Developer Note: fortran interface is not autogenerated as the f90
5545     interface defintion cannot be generated correctly [due to MatFactorInfo]
5546 
5547 @*/
5548 PetscErrorCode PETSCMAT_DLLEXPORT MatILUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
5549 {
5550   PetscErrorCode ierr;
5551 
5552   PetscFunctionBegin;
5553   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5554   PetscValidType(mat,1);
5555   PetscValidHeaderSpecific(row,IS_CLASSID,2);
5556   PetscValidHeaderSpecific(col,IS_CLASSID,3);
5557   PetscValidPointer(info,4);
5558   PetscValidPointer(fact,5);
5559   if (info->levels < 0) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Levels of fill negative %D",(PetscInt)info->levels);
5560   if (info->fill < 1.0) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %G",info->fill);
5561   if (!(fact)->ops->ilufactorsymbolic) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Matrix type %s  symbolic ILU",((PetscObject)mat)->type_name);
5562   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5563   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5564   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5565 
5566   ierr = PetscLogEventBegin(MAT_ILUFactorSymbolic,mat,row,col,0);CHKERRQ(ierr);
5567   ierr = (fact->ops->ilufactorsymbolic)(fact,mat,row,col,info);CHKERRQ(ierr);
5568   ierr = PetscLogEventEnd(MAT_ILUFactorSymbolic,mat,row,col,0);CHKERRQ(ierr);
5569   PetscFunctionReturn(0);
5570 }
5571 
5572 #undef __FUNCT__
5573 #define __FUNCT__ "MatICCFactorSymbolic"
5574 /*@C
5575    MatICCFactorSymbolic - Performs symbolic incomplete
5576    Cholesky factorization for a symmetric matrix.  Use
5577    MatCholeskyFactorNumeric() to complete the factorization.
5578 
5579    Collective on Mat
5580 
5581    Input Parameters:
5582 +  mat - the matrix
5583 .  perm - row and column permutation
5584 -  info - structure containing
5585 $      levels - number of levels of fill.
5586 $      expected fill - as ratio of original fill.
5587 
5588    Output Parameter:
5589 .  fact - the factored matrix
5590 
5591    Notes:
5592    Most users should employ the KSP interface for linear solvers
5593    instead of working directly with matrix algebra routines such as this.
5594    See, e.g., KSPCreate().
5595 
5596    Level: developer
5597 
5598   Concepts: matrices^symbolic incomplete Cholesky factorization
5599   Concepts: matrices^factorization
5600   Concepts: Cholsky^symbolic factorization
5601 
5602 .seealso: MatCholeskyFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
5603 
5604     Developer Note: fortran interface is not autogenerated as the f90
5605     interface defintion cannot be generated correctly [due to MatFactorInfo]
5606 
5607 @*/
5608 PetscErrorCode PETSCMAT_DLLEXPORT MatICCFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
5609 {
5610   PetscErrorCode ierr;
5611 
5612   PetscFunctionBegin;
5613   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5614   PetscValidType(mat,1);
5615   PetscValidHeaderSpecific(perm,IS_CLASSID,2);
5616   PetscValidPointer(info,3);
5617   PetscValidPointer(fact,4);
5618   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5619   if (info->levels < 0) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Levels negative %D",(PetscInt) info->levels);
5620   if (info->fill < 1.0) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %G",info->fill);
5621   if (!(fact)->ops->iccfactorsymbolic) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Matrix type %s  symbolic ICC",((PetscObject)mat)->type_name);
5622   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5623   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5624 
5625   ierr = PetscLogEventBegin(MAT_ICCFactorSymbolic,mat,perm,0,0);CHKERRQ(ierr);
5626   ierr = (fact->ops->iccfactorsymbolic)(fact,mat,perm,info);CHKERRQ(ierr);
5627   ierr = PetscLogEventEnd(MAT_ICCFactorSymbolic,mat,perm,0,0);CHKERRQ(ierr);
5628   PetscFunctionReturn(0);
5629 }
5630 
5631 #undef __FUNCT__
5632 #define __FUNCT__ "MatGetArray"
5633 /*@C
5634    MatGetArray - Returns a pointer to the element values in the matrix.
5635    The result of this routine is dependent on the underlying matrix data
5636    structure, and may not even work for certain matrix types.  You MUST
5637    call MatRestoreArray() when you no longer need to access the array.
5638 
5639    Not Collective
5640 
5641    Input Parameter:
5642 .  mat - the matrix
5643 
5644    Output Parameter:
5645 .  v - the location of the values
5646 
5647 
5648    Fortran Note:
5649    This routine is used differently from Fortran, e.g.,
5650 .vb
5651         Mat         mat
5652         PetscScalar mat_array(1)
5653         PetscOffset i_mat
5654         PetscErrorCode ierr
5655         call MatGetArray(mat,mat_array,i_mat,ierr)
5656 
5657   C  Access first local entry in matrix; note that array is
5658   C  treated as one dimensional
5659         value = mat_array(i_mat + 1)
5660 
5661         [... other code ...]
5662         call MatRestoreArray(mat,mat_array,i_mat,ierr)
5663 .ve
5664 
5665    See the Fortran chapter of the users manual and
5666    petsc/src/mat/examples/tests for details.
5667 
5668    Level: advanced
5669 
5670    Concepts: matrices^access array
5671 
5672 .seealso: MatRestoreArray(), MatGetArrayF90(), MatGetRowIJ()
5673 @*/
5674 PetscErrorCode PETSCMAT_DLLEXPORT MatGetArray(Mat mat,PetscScalar *v[])
5675 {
5676   PetscErrorCode ierr;
5677 
5678   PetscFunctionBegin;
5679   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5680   PetscValidType(mat,1);
5681   PetscValidPointer(v,2);
5682   if (!mat->ops->getarray) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5683   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5684   ierr = (*mat->ops->getarray)(mat,v);CHKERRQ(ierr);
5685   CHKMEMQ;
5686   PetscFunctionReturn(0);
5687 }
5688 
5689 #undef __FUNCT__
5690 #define __FUNCT__ "MatRestoreArray"
5691 /*@C
5692    MatRestoreArray - Restores the matrix after MatGetArray() has been called.
5693 
5694    Not Collective
5695 
5696    Input Parameter:
5697 +  mat - the matrix
5698 -  v - the location of the values
5699 
5700    Fortran Note:
5701    This routine is used differently from Fortran, e.g.,
5702 .vb
5703         Mat         mat
5704         PetscScalar mat_array(1)
5705         PetscOffset i_mat
5706         PetscErrorCode ierr
5707         call MatGetArray(mat,mat_array,i_mat,ierr)
5708 
5709   C  Access first local entry in matrix; note that array is
5710   C  treated as one dimensional
5711         value = mat_array(i_mat + 1)
5712 
5713         [... other code ...]
5714         call MatRestoreArray(mat,mat_array,i_mat,ierr)
5715 .ve
5716 
5717    See the Fortran chapter of the users manual and
5718    petsc/src/mat/examples/tests for details
5719 
5720    Level: advanced
5721 
5722 .seealso: MatGetArray(), MatRestoreArrayF90()
5723 @*/
5724 PetscErrorCode PETSCMAT_DLLEXPORT MatRestoreArray(Mat mat,PetscScalar *v[])
5725 {
5726   PetscErrorCode ierr;
5727 
5728   PetscFunctionBegin;
5729   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5730   PetscValidType(mat,1);
5731   PetscValidPointer(v,2);
5732 #if defined(PETSC_USE_DEBUG)
5733   CHKMEMQ;
5734 #endif
5735   if (!mat->ops->restorearray) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5736   ierr = (*mat->ops->restorearray)(mat,v);CHKERRQ(ierr);
5737   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
5738   PetscFunctionReturn(0);
5739 }
5740 
5741 #undef __FUNCT__
5742 #define __FUNCT__ "MatGetSubMatrices"
5743 /*@C
5744    MatGetSubMatrices - Extracts several submatrices from a matrix. If submat
5745    points to an array of valid matrices, they may be reused to store the new
5746    submatrices.
5747 
5748    Collective on Mat
5749 
5750    Input Parameters:
5751 +  mat - the matrix
5752 .  n   - the number of submatrixes to be extracted (on this processor, may be zero)
5753 .  irow, icol - index sets of rows and columns to extract
5754 -  scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
5755 
5756    Output Parameter:
5757 .  submat - the array of submatrices
5758 
5759    Notes:
5760    MatGetSubMatrices() can extract ONLY sequential submatrices
5761    (from both sequential and parallel matrices). Use MatGetSubMatrix()
5762    to extract a parallel submatrix.
5763 
5764    When extracting submatrices from a parallel matrix, each processor can
5765    form a different submatrix by setting the rows and columns of its
5766    individual index sets according to the local submatrix desired.
5767 
5768    When finished using the submatrices, the user should destroy
5769    them with MatDestroyMatrices().
5770 
5771    MAT_REUSE_MATRIX can only be used when the nonzero structure of the
5772    original matrix has not changed from that last call to MatGetSubMatrices().
5773 
5774    This routine creates the matrices in submat; you should NOT create them before
5775    calling it. It also allocates the array of matrix pointers submat.
5776 
5777    For BAIJ matrices the index sets must respect the block structure, that is if they
5778    request one row/column in a block, they must request all rows/columns that are in
5779    that block. For example, if the block size is 2 you cannot request just row 0 and
5780    column 0.
5781 
5782    Fortran Note:
5783    The Fortran interface is slightly different from that given below; it
5784    requires one to pass in  as submat a Mat (integer) array of size at least m.
5785 
5786    Level: advanced
5787 
5788    Concepts: matrices^accessing submatrices
5789    Concepts: submatrices
5790 
5791 .seealso: MatDestroyMatrices(), MatGetSubMatrix(), MatGetRow(), MatGetDiagonal(), MatReuse
5792 @*/
5793 PetscErrorCode PETSCMAT_DLLEXPORT MatGetSubMatrices(Mat mat,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *submat[])
5794 {
5795   PetscErrorCode ierr;
5796   PetscInt        i;
5797   PetscTruth      eq;
5798 
5799   PetscFunctionBegin;
5800   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5801   PetscValidType(mat,1);
5802   if (n) {
5803     PetscValidPointer(irow,3);
5804     PetscValidHeaderSpecific(*irow,IS_CLASSID,3);
5805     PetscValidPointer(icol,4);
5806     PetscValidHeaderSpecific(*icol,IS_CLASSID,4);
5807   }
5808   PetscValidPointer(submat,6);
5809   if (n && scall == MAT_REUSE_MATRIX) {
5810     PetscValidPointer(*submat,6);
5811     PetscValidHeaderSpecific(**submat,MAT_CLASSID,6);
5812   }
5813   if (!mat->ops->getsubmatrices) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5814   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5815   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5816   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5817 
5818   ierr = PetscLogEventBegin(MAT_GetSubMatrices,mat,0,0,0);CHKERRQ(ierr);
5819   ierr = (*mat->ops->getsubmatrices)(mat,n,irow,icol,scall,submat);CHKERRQ(ierr);
5820   ierr = PetscLogEventEnd(MAT_GetSubMatrices,mat,0,0,0);CHKERRQ(ierr);
5821   for (i=0; i<n; i++) {
5822     if (mat->symmetric || mat->structurally_symmetric || mat->hermitian) {
5823       ierr = ISEqual(irow[i],icol[i],&eq);CHKERRQ(ierr);
5824       if (eq) {
5825 	if (mat->symmetric){
5826 	  ierr = MatSetOption((*submat)[i],MAT_SYMMETRIC,PETSC_TRUE);CHKERRQ(ierr);
5827 	} else if (mat->hermitian) {
5828 	  ierr = MatSetOption((*submat)[i],MAT_HERMITIAN,PETSC_TRUE);CHKERRQ(ierr);
5829 	} else if (mat->structurally_symmetric) {
5830 	  ierr = MatSetOption((*submat)[i],MAT_STRUCTURALLY_SYMMETRIC,PETSC_TRUE);CHKERRQ(ierr);
5831 	}
5832       }
5833     }
5834   }
5835   PetscFunctionReturn(0);
5836 }
5837 
5838 #undef __FUNCT__
5839 #define __FUNCT__ "MatDestroyMatrices"
5840 /*@C
5841    MatDestroyMatrices - Destroys a set of matrices obtained with MatGetSubMatrices().
5842 
5843    Collective on Mat
5844 
5845    Input Parameters:
5846 +  n - the number of local matrices
5847 -  mat - the matrices (note that this is a pointer to the array of matrices, just to match the calling
5848                        sequence of MatGetSubMatrices())
5849 
5850    Level: advanced
5851 
5852     Notes: Frees not only the matrices, but also the array that contains the matrices
5853            In Fortran will not free the array.
5854 
5855 .seealso: MatGetSubMatrices()
5856 @*/
5857 PetscErrorCode PETSCMAT_DLLEXPORT MatDestroyMatrices(PetscInt n,Mat *mat[])
5858 {
5859   PetscErrorCode ierr;
5860   PetscInt       i;
5861 
5862   PetscFunctionBegin;
5863   if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Trying to destroy negative number of matrices %D",n);
5864   PetscValidPointer(mat,2);
5865   for (i=0; i<n; i++) {
5866     ierr = MatDestroy((*mat)[i]);CHKERRQ(ierr);
5867   }
5868   /* memory is allocated even if n = 0 */
5869   ierr = PetscFree(*mat);CHKERRQ(ierr);
5870   PetscFunctionReturn(0);
5871 }
5872 
5873 #undef __FUNCT__
5874 #define __FUNCT__ "MatGetSeqNonzeroStructure"
5875 /*@C
5876    MatGetSeqNonzeroStructure - Extracts the sequential nonzero structure from a matrix.
5877 
5878    Collective on Mat
5879 
5880    Input Parameters:
5881 .  mat - the matrix
5882 
5883    Output Parameter:
5884 .  matstruct - the sequential matrix with the nonzero structure of mat
5885 
5886   Level: intermediate
5887 
5888 .seealso: MatDestroySeqNonzeroStructure(), MatGetSubMatrices(), MatDestroyMatrices()
5889 @*/
5890 PetscErrorCode PETSCMAT_DLLEXPORT MatGetSeqNonzeroStructure(Mat mat,Mat *matstruct)
5891 {
5892   PetscErrorCode ierr;
5893 
5894   PetscFunctionBegin;
5895   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5896   PetscValidPointer(matstruct,2);
5897 
5898   PetscValidType(mat,1);
5899   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5900   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5901 
5902   if (!mat->ops->getseqnonzerostructure) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Not for matrix type %s\n",((PetscObject)mat)->type_name);
5903   ierr = PetscLogEventBegin(MAT_GetSeqNonzeroStructure,mat,0,0,0);CHKERRQ(ierr);
5904   ierr = (*mat->ops->getseqnonzerostructure)(mat,matstruct);CHKERRQ(ierr);
5905   ierr = PetscLogEventEnd(MAT_GetSeqNonzeroStructure,mat,0,0,0);CHKERRQ(ierr);
5906   PetscFunctionReturn(0);
5907 }
5908 
5909 #undef __FUNCT__
5910 #define __FUNCT__ "MatDestroySeqNonzeroStructure"
5911 /*@C
5912    MatDestroySeqNonzeroStructure - Destroys matrix obtained with MatGetSeqNonzeroStructure().
5913 
5914    Collective on Mat
5915 
5916    Input Parameters:
5917 .  mat - the matrix (note that this is a pointer to the array of matrices, just to match the calling
5918                        sequence of MatGetSequentialNonzeroStructure())
5919 
5920    Level: advanced
5921 
5922     Notes: Frees not only the matrices, but also the array that contains the matrices
5923 
5924 .seealso: MatGetSeqNonzeroStructure()
5925 @*/
5926 PetscErrorCode PETSCMAT_DLLEXPORT MatDestroySeqNonzeroStructure(Mat *mat)
5927 {
5928   PetscErrorCode ierr;
5929 
5930   PetscFunctionBegin;
5931   PetscValidPointer(mat,1);
5932   ierr = MatDestroy(*mat);CHKERRQ(ierr);
5933   PetscFunctionReturn(0);
5934 }
5935 
5936 #undef __FUNCT__
5937 #define __FUNCT__ "MatIncreaseOverlap"
5938 /*@
5939    MatIncreaseOverlap - Given a set of submatrices indicated by index sets,
5940    replaces the index sets by larger ones that represent submatrices with
5941    additional overlap.
5942 
5943    Collective on Mat
5944 
5945    Input Parameters:
5946 +  mat - the matrix
5947 .  n   - the number of index sets
5948 .  is  - the array of index sets (these index sets will changed during the call)
5949 -  ov  - the additional overlap requested
5950 
5951    Level: developer
5952 
5953    Concepts: overlap
5954    Concepts: ASM^computing overlap
5955 
5956 .seealso: MatGetSubMatrices()
5957 @*/
5958 PetscErrorCode PETSCMAT_DLLEXPORT MatIncreaseOverlap(Mat mat,PetscInt n,IS is[],PetscInt ov)
5959 {
5960   PetscErrorCode ierr;
5961 
5962   PetscFunctionBegin;
5963   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
5964   PetscValidType(mat,1);
5965   if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Must have one or more domains, you have %D",n);
5966   if (n) {
5967     PetscValidPointer(is,3);
5968     PetscValidHeaderSpecific(*is,IS_CLASSID,3);
5969   }
5970   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5971   if (mat->factortype)     SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5972   ierr = MatPreallocated(mat);CHKERRQ(ierr);
5973 
5974   if (!ov) PetscFunctionReturn(0);
5975   if (!mat->ops->increaseoverlap) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5976   ierr = PetscLogEventBegin(MAT_IncreaseOverlap,mat,0,0,0);CHKERRQ(ierr);
5977   ierr = (*mat->ops->increaseoverlap)(mat,n,is,ov);CHKERRQ(ierr);
5978   ierr = PetscLogEventEnd(MAT_IncreaseOverlap,mat,0,0,0);CHKERRQ(ierr);
5979   PetscFunctionReturn(0);
5980 }
5981 
5982 #undef __FUNCT__
5983 #define __FUNCT__ "MatGetBlockSize"
5984 /*@
5985    MatGetBlockSize - Returns the matrix block size; useful especially for the
5986    block row and block diagonal formats.
5987 
5988    Not Collective
5989 
5990    Input Parameter:
5991 .  mat - the matrix
5992 
5993    Output Parameter:
5994 .  bs - block size
5995 
5996    Notes:
5997    Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ
5998 
5999    Level: intermediate
6000 
6001    Concepts: matrices^block size
6002 
6003 .seealso: MatCreateSeqBAIJ(), MatCreateMPIBAIJ()
6004 @*/
6005 PetscErrorCode PETSCMAT_DLLEXPORT MatGetBlockSize(Mat mat,PetscInt *bs)
6006 {
6007   PetscErrorCode ierr;
6008 
6009   PetscFunctionBegin;
6010   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6011   PetscValidType(mat,1);
6012   PetscValidIntPointer(bs,2);
6013   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6014   *bs = mat->rmap->bs;
6015   PetscFunctionReturn(0);
6016 }
6017 
6018 #undef __FUNCT__
6019 #define __FUNCT__ "MatSetBlockSize"
6020 /*@
6021    MatSetBlockSize - Sets the matrix block size; for many matrix types you
6022      cannot use this and MUST set the blocksize when you preallocate the matrix
6023 
6024    Logically Collective on Mat
6025 
6026    Input Parameters:
6027 +  mat - the matrix
6028 -  bs - block size
6029 
6030    Notes:
6031      For BAIJ matrices, this just checks that the block size agrees with the BAIJ size,
6032      it is not possible to change BAIJ block sizes after preallocation.
6033 
6034    Level: intermediate
6035 
6036    Concepts: matrices^block size
6037 
6038 .seealso: MatCreateSeqBAIJ(), MatCreateMPIBAIJ(), MatGetBlockSize()
6039 @*/
6040 PetscErrorCode PETSCMAT_DLLEXPORT MatSetBlockSize(Mat mat,PetscInt bs)
6041 {
6042   PetscErrorCode ierr;
6043 
6044   PetscFunctionBegin;
6045   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6046   PetscValidType(mat,1);
6047   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6048   if (bs < 1) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Block size %d, must be positive",bs);
6049   if (mat->ops->setblocksize) {
6050     ierr = (*mat->ops->setblocksize)(mat,bs);CHKERRQ(ierr);
6051   } else {
6052     SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_INCOMP,"Cannot set the blocksize for matrix type %s",((PetscObject)mat)->type_name);
6053   }
6054   PetscFunctionReturn(0);
6055 }
6056 
6057 #undef __FUNCT__
6058 #define __FUNCT__ "MatGetRowIJ"
6059 /*@C
6060     MatGetRowIJ - Returns the compressed row storage i and j indices for sequential matrices.
6061 
6062    Collective on Mat
6063 
6064     Input Parameters:
6065 +   mat - the matrix
6066 .   shift -  0 or 1 indicating we want the indices starting at 0 or 1
6067 .   symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
6068                 symmetrized
6069 -   inodecompressed - PETSC_TRUE or PETSC_FALSE  indicating if the nonzero structure of the
6070                  inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
6071                  always used.
6072 
6073     Output Parameters:
6074 +   n - number of rows in the (possibly compressed) matrix
6075 .   ia - the row pointers [of length n+1]
6076 .   ja - the column indices
6077 -   done - indicates if the routine actually worked and returned appropriate ia[] and ja[] arrays; callers
6078            are responsible for handling the case when done == PETSC_FALSE and ia and ja are not set
6079 
6080     Level: developer
6081 
6082     Notes: You CANNOT change any of the ia[] or ja[] values.
6083 
6084            Use MatRestoreRowIJ() when you are finished accessing the ia[] and ja[] values
6085 
6086     Fortran Node
6087 
6088            In Fortran use
6089 $           PetscInt ia(1), ja(1)
6090 $           PetscOffset iia, jja
6091 $      call MatGetRowIJ(mat,shift,symmetric,inodecompressed,n,ia,iia,ja,jja,done,ierr)
6092 $
6093 $          or
6094 $
6095 $           PetscScalar, pointer :: xx_v(:)
6096 $    call  MatGetRowIJF90(mat,shift,symmetric,inodecompressed,n,ia,ja,done,ierr)
6097 
6098 
6099        Acess the ith and jth entries via ia(iia + i) and ja(jja + j)
6100 
6101 .seealso: MatGetColumnIJ(), MatRestoreRowIJ(), MatGetArray()
6102 @*/
6103 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRowIJ(Mat mat,PetscInt shift,PetscTruth symmetric,PetscTruth inodecompressed,PetscInt *n,PetscInt *ia[],PetscInt* ja[],PetscTruth *done)
6104 {
6105   PetscErrorCode ierr;
6106 
6107   PetscFunctionBegin;
6108   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6109   PetscValidType(mat,1);
6110   PetscValidIntPointer(n,4);
6111   if (ia) PetscValidIntPointer(ia,5);
6112   if (ja) PetscValidIntPointer(ja,6);
6113   PetscValidIntPointer(done,7);
6114   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6115   if (!mat->ops->getrowij) *done = PETSC_FALSE;
6116   else {
6117     *done = PETSC_TRUE;
6118     ierr = PetscLogEventBegin(MAT_GetRowIJ,mat,0,0,0);CHKERRQ(ierr);
6119     ierr  = (*mat->ops->getrowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);CHKERRQ(ierr);
6120     ierr = PetscLogEventEnd(MAT_GetRowIJ,mat,0,0,0);CHKERRQ(ierr);
6121   }
6122   PetscFunctionReturn(0);
6123 }
6124 
6125 #undef __FUNCT__
6126 #define __FUNCT__ "MatGetColumnIJ"
6127 /*@C
6128     MatGetColumnIJ - Returns the compressed column storage i and j indices for sequential matrices.
6129 
6130     Collective on Mat
6131 
6132     Input Parameters:
6133 +   mat - the matrix
6134 .   shift - 1 or zero indicating we want the indices starting at 0 or 1
6135 .   symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
6136                 symmetrized
6137 -   inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
6138                  inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
6139                  always used.
6140 
6141     Output Parameters:
6142 +   n - number of columns in the (possibly compressed) matrix
6143 .   ia - the column pointers
6144 .   ja - the row indices
6145 -   done - PETSC_TRUE or PETSC_FALSE, indicating whether the values have been returned
6146 
6147     Level: developer
6148 
6149 .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
6150 @*/
6151 PetscErrorCode PETSCMAT_DLLEXPORT MatGetColumnIJ(Mat mat,PetscInt shift,PetscTruth symmetric,PetscTruth inodecompressed,PetscInt *n,PetscInt *ia[],PetscInt* ja[],PetscTruth *done)
6152 {
6153   PetscErrorCode ierr;
6154 
6155   PetscFunctionBegin;
6156   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6157   PetscValidType(mat,1);
6158   PetscValidIntPointer(n,4);
6159   if (ia) PetscValidIntPointer(ia,5);
6160   if (ja) PetscValidIntPointer(ja,6);
6161   PetscValidIntPointer(done,7);
6162   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6163   if (!mat->ops->getcolumnij) *done = PETSC_FALSE;
6164   else {
6165     *done = PETSC_TRUE;
6166     ierr  = (*mat->ops->getcolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);CHKERRQ(ierr);
6167   }
6168   PetscFunctionReturn(0);
6169 }
6170 
6171 #undef __FUNCT__
6172 #define __FUNCT__ "MatRestoreRowIJ"
6173 /*@C
6174     MatRestoreRowIJ - Call after you are completed with the ia,ja indices obtained with
6175     MatGetRowIJ().
6176 
6177     Collective on Mat
6178 
6179     Input Parameters:
6180 +   mat - the matrix
6181 .   shift - 1 or zero indicating we want the indices starting at 0 or 1
6182 .   symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
6183                 symmetrized
6184 -   inodecompressed -  PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
6185                  inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
6186                  always used.
6187 
6188     Output Parameters:
6189 +   n - size of (possibly compressed) matrix
6190 .   ia - the row pointers
6191 .   ja - the column indices
6192 -   done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
6193 
6194     Level: developer
6195 
6196 .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
6197 @*/
6198 PetscErrorCode PETSCMAT_DLLEXPORT MatRestoreRowIJ(Mat mat,PetscInt shift,PetscTruth symmetric,PetscTruth inodecompressed,PetscInt *n,PetscInt *ia[],PetscInt* ja[],PetscTruth *done)
6199 {
6200   PetscErrorCode ierr;
6201 
6202   PetscFunctionBegin;
6203   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6204   PetscValidType(mat,1);
6205   if (ia) PetscValidIntPointer(ia,5);
6206   if (ja) PetscValidIntPointer(ja,6);
6207   PetscValidIntPointer(done,7);
6208   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6209 
6210   if (!mat->ops->restorerowij) *done = PETSC_FALSE;
6211   else {
6212     *done = PETSC_TRUE;
6213     ierr  = (*mat->ops->restorerowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);CHKERRQ(ierr);
6214   }
6215   PetscFunctionReturn(0);
6216 }
6217 
6218 #undef __FUNCT__
6219 #define __FUNCT__ "MatRestoreColumnIJ"
6220 /*@C
6221     MatRestoreColumnIJ - Call after you are completed with the ia,ja indices obtained with
6222     MatGetColumnIJ().
6223 
6224     Collective on Mat
6225 
6226     Input Parameters:
6227 +   mat - the matrix
6228 .   shift - 1 or zero indicating we want the indices starting at 0 or 1
6229 -   symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
6230                 symmetrized
6231 -   inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
6232                  inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
6233                  always used.
6234 
6235     Output Parameters:
6236 +   n - size of (possibly compressed) matrix
6237 .   ia - the column pointers
6238 .   ja - the row indices
6239 -   done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
6240 
6241     Level: developer
6242 
6243 .seealso: MatGetColumnIJ(), MatRestoreRowIJ()
6244 @*/
6245 PetscErrorCode PETSCMAT_DLLEXPORT MatRestoreColumnIJ(Mat mat,PetscInt shift,PetscTruth symmetric,PetscTruth inodecompressed,PetscInt *n,PetscInt *ia[],PetscInt* ja[],PetscTruth *done)
6246 {
6247   PetscErrorCode ierr;
6248 
6249   PetscFunctionBegin;
6250   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6251   PetscValidType(mat,1);
6252   if (ia) PetscValidIntPointer(ia,5);
6253   if (ja) PetscValidIntPointer(ja,6);
6254   PetscValidIntPointer(done,7);
6255   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6256 
6257   if (!mat->ops->restorecolumnij) *done = PETSC_FALSE;
6258   else {
6259     *done = PETSC_TRUE;
6260     ierr  = (*mat->ops->restorecolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);CHKERRQ(ierr);
6261   }
6262   PetscFunctionReturn(0);
6263 }
6264 
6265 #undef __FUNCT__
6266 #define __FUNCT__ "MatColoringPatch"
6267 /*@C
6268     MatColoringPatch -Used inside matrix coloring routines that
6269     use MatGetRowIJ() and/or MatGetColumnIJ().
6270 
6271     Collective on Mat
6272 
6273     Input Parameters:
6274 +   mat - the matrix
6275 .   ncolors - max color value
6276 .   n   - number of entries in colorarray
6277 -   colorarray - array indicating color for each column
6278 
6279     Output Parameters:
6280 .   iscoloring - coloring generated using colorarray information
6281 
6282     Level: developer
6283 
6284 .seealso: MatGetRowIJ(), MatGetColumnIJ()
6285 
6286 @*/
6287 PetscErrorCode PETSCMAT_DLLEXPORT MatColoringPatch(Mat mat,PetscInt ncolors,PetscInt n,ISColoringValue colorarray[],ISColoring *iscoloring)
6288 {
6289   PetscErrorCode ierr;
6290 
6291   PetscFunctionBegin;
6292   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6293   PetscValidType(mat,1);
6294   PetscValidIntPointer(colorarray,4);
6295   PetscValidPointer(iscoloring,5);
6296   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6297 
6298   if (!mat->ops->coloringpatch){
6299     ierr = ISColoringCreate(((PetscObject)mat)->comm,ncolors,n,colorarray,iscoloring);CHKERRQ(ierr);
6300   } else {
6301     ierr = (*mat->ops->coloringpatch)(mat,ncolors,n,colorarray,iscoloring);CHKERRQ(ierr);
6302   }
6303   PetscFunctionReturn(0);
6304 }
6305 
6306 
6307 #undef __FUNCT__
6308 #define __FUNCT__ "MatSetUnfactored"
6309 /*@
6310    MatSetUnfactored - Resets a factored matrix to be treated as unfactored.
6311 
6312    Logically Collective on Mat
6313 
6314    Input Parameter:
6315 .  mat - the factored matrix to be reset
6316 
6317    Notes:
6318    This routine should be used only with factored matrices formed by in-place
6319    factorization via ILU(0) (or by in-place LU factorization for the MATSEQDENSE
6320    format).  This option can save memory, for example, when solving nonlinear
6321    systems with a matrix-free Newton-Krylov method and a matrix-based, in-place
6322    ILU(0) preconditioner.
6323 
6324    Note that one can specify in-place ILU(0) factorization by calling
6325 .vb
6326      PCType(pc,PCILU);
6327      PCFactorSeUseInPlace(pc);
6328 .ve
6329    or by using the options -pc_type ilu -pc_factor_in_place
6330 
6331    In-place factorization ILU(0) can also be used as a local
6332    solver for the blocks within the block Jacobi or additive Schwarz
6333    methods (runtime option: -sub_pc_factor_in_place).  See the discussion
6334    of these preconditioners in the users manual for details on setting
6335    local solver options.
6336 
6337    Most users should employ the simplified KSP interface for linear solvers
6338    instead of working directly with matrix algebra routines such as this.
6339    See, e.g., KSPCreate().
6340 
6341    Level: developer
6342 
6343 .seealso: PCFactorSetUseInPlace()
6344 
6345    Concepts: matrices^unfactored
6346 
6347 @*/
6348 PetscErrorCode PETSCMAT_DLLEXPORT MatSetUnfactored(Mat mat)
6349 {
6350   PetscErrorCode ierr;
6351 
6352   PetscFunctionBegin;
6353   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6354   PetscValidType(mat,1);
6355   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6356   mat->factortype = MAT_FACTOR_NONE;
6357   if (!mat->ops->setunfactored) PetscFunctionReturn(0);
6358   ierr = (*mat->ops->setunfactored)(mat);CHKERRQ(ierr);
6359   PetscFunctionReturn(0);
6360 }
6361 
6362 /*MC
6363     MatGetArrayF90 - Accesses a matrix array from Fortran90.
6364 
6365     Synopsis:
6366     MatGetArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
6367 
6368     Not collective
6369 
6370     Input Parameter:
6371 .   x - matrix
6372 
6373     Output Parameters:
6374 +   xx_v - the Fortran90 pointer to the array
6375 -   ierr - error code
6376 
6377     Example of Usage:
6378 .vb
6379       PetscScalar, pointer xx_v(:)
6380       ....
6381       call MatGetArrayF90(x,xx_v,ierr)
6382       a = xx_v(3)
6383       call MatRestoreArrayF90(x,xx_v,ierr)
6384 .ve
6385 
6386     Notes:
6387     Not yet supported for all F90 compilers
6388 
6389     Level: advanced
6390 
6391 .seealso:  MatRestoreArrayF90(), MatGetArray(), MatRestoreArray()
6392 
6393     Concepts: matrices^accessing array
6394 
6395 M*/
6396 
6397 /*MC
6398     MatRestoreArrayF90 - Restores a matrix array that has been
6399     accessed with MatGetArrayF90().
6400 
6401     Synopsis:
6402     MatRestoreArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
6403 
6404     Not collective
6405 
6406     Input Parameters:
6407 +   x - matrix
6408 -   xx_v - the Fortran90 pointer to the array
6409 
6410     Output Parameter:
6411 .   ierr - error code
6412 
6413     Example of Usage:
6414 .vb
6415        PetscScalar, pointer xx_v(:)
6416        ....
6417        call MatGetArrayF90(x,xx_v,ierr)
6418        a = xx_v(3)
6419        call MatRestoreArrayF90(x,xx_v,ierr)
6420 .ve
6421 
6422     Notes:
6423     Not yet supported for all F90 compilers
6424 
6425     Level: advanced
6426 
6427 .seealso:  MatGetArrayF90(), MatGetArray(), MatRestoreArray()
6428 
6429 M*/
6430 
6431 
6432 #undef __FUNCT__
6433 #define __FUNCT__ "MatGetSubMatrix"
6434 /*@
6435     MatGetSubMatrix - Gets a single submatrix on the same number of processors
6436                       as the original matrix.
6437 
6438     Collective on Mat
6439 
6440     Input Parameters:
6441 +   mat - the original matrix
6442 .   isrow - parallel IS containing the rows this processor should obtain
6443 .   iscol - parallel IS containing all columns you wish to keep. Each process should list the columns that will be in IT's "diagonal part" in the new matrix.
6444 -   cll - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
6445 
6446     Output Parameter:
6447 .   newmat - the new submatrix, of the same type as the old
6448 
6449     Level: advanced
6450 
6451     Notes:
6452     The submatrix will be able to be multiplied with vectors using the same layout as iscol.
6453 
6454     The rows in isrow will be sorted into the same order as the original matrix on each process.
6455 
6456       The first time this is called you should use a cll of MAT_INITIAL_MATRIX,
6457    the MatGetSubMatrix() routine will create the newmat for you. Any additional calls
6458    to this routine with a mat of the same nonzero structure and with a call of MAT_REUSE_MATRIX
6459    will reuse the matrix generated the first time.  You should call MatDestroy() on newmat when
6460    you are finished using it.
6461 
6462     The communicator of the newly obtained matrix is ALWAYS the same as the communicator of
6463     the input matrix.
6464 
6465     If iscol is PETSC_NULL then all columns are obtained (not supported in Fortran).
6466 
6467    Example usage:
6468    Consider the following 8x8 matrix with 34 non-zero values, that is
6469    assembled across 3 processors. Lets assume that proc0 owns 3 rows,
6470    proc1 owns 3 rows, proc2 owns 2 rows. This division can be shown
6471    as follows:
6472 
6473 .vb
6474             1  2  0  |  0  3  0  |  0  4
6475     Proc0   0  5  6  |  7  0  0  |  8  0
6476             9  0 10  | 11  0  0  | 12  0
6477     -------------------------------------
6478            13  0 14  | 15 16 17  |  0  0
6479     Proc1   0 18  0  | 19 20 21  |  0  0
6480             0  0  0  | 22 23  0  | 24  0
6481     -------------------------------------
6482     Proc2  25 26 27  |  0  0 28  | 29  0
6483            30  0  0  | 31 32 33  |  0 34
6484 .ve
6485 
6486     Suppose isrow = [0 1 | 4 | 6 7] and iscol = [1 2 | 3 4 5 | 6].  The resulting submatrix is
6487 
6488 .vb
6489             2  0  |  0  3  0  |  0
6490     Proc0   5  6  |  7  0  0  |  8
6491     -------------------------------
6492     Proc1  18  0  | 19 20 21  |  0
6493     -------------------------------
6494     Proc2  26 27  |  0  0 28  | 29
6495             0  0  | 31 32 33  |  0
6496 .ve
6497 
6498 
6499     Concepts: matrices^submatrices
6500 
6501 .seealso: MatGetSubMatrices()
6502 @*/
6503 PetscErrorCode PETSCMAT_DLLEXPORT MatGetSubMatrix(Mat mat,IS isrow,IS iscol,MatReuse cll,Mat *newmat)
6504 {
6505   PetscErrorCode ierr;
6506   PetscMPIInt    size;
6507   Mat            *local;
6508   IS             iscoltmp;
6509 
6510   PetscFunctionBegin;
6511   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6512   PetscValidHeaderSpecific(isrow,IS_CLASSID,2);
6513   if (iscol) PetscValidHeaderSpecific(iscol,IS_CLASSID,3);
6514   PetscValidPointer(newmat,5);
6515   if (cll == MAT_REUSE_MATRIX) PetscValidHeaderSpecific(*newmat,MAT_CLASSID,5);
6516   PetscValidType(mat,1);
6517   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6518   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6519   ierr = MPI_Comm_size(((PetscObject)mat)->comm,&size);CHKERRQ(ierr);
6520 
6521   if (!iscol) {
6522     ierr = ISCreateStride(((PetscObject)mat)->comm,mat->cmap->n,mat->cmap->rstart,1,&iscoltmp);CHKERRQ(ierr);
6523   } else {
6524     iscoltmp = iscol;
6525   }
6526 
6527   /* if original matrix is on just one processor then use submatrix generated */
6528   if (mat->ops->getsubmatrices && !mat->ops->getsubmatrix && size == 1 && cll == MAT_REUSE_MATRIX) {
6529     ierr = MatGetSubMatrices(mat,1,&isrow,&iscoltmp,MAT_REUSE_MATRIX,&newmat);CHKERRQ(ierr);
6530     if (!iscol) {ierr = ISDestroy(iscoltmp);CHKERRQ(ierr);}
6531     PetscFunctionReturn(0);
6532   } else if (mat->ops->getsubmatrices && !mat->ops->getsubmatrix && size == 1) {
6533     ierr    = MatGetSubMatrices(mat,1,&isrow,&iscoltmp,MAT_INITIAL_MATRIX,&local);CHKERRQ(ierr);
6534     *newmat = *local;
6535     ierr    = PetscFree(local);CHKERRQ(ierr);
6536     if (!iscol) {ierr = ISDestroy(iscoltmp);CHKERRQ(ierr);}
6537     PetscFunctionReturn(0);
6538   } else if (!mat->ops->getsubmatrix) {
6539     /* Create a new matrix type that implements the operation using the full matrix */
6540     switch (cll) {
6541       case MAT_INITIAL_MATRIX:
6542         ierr = MatCreateSubMatrix(mat,isrow,iscoltmp,newmat);CHKERRQ(ierr);
6543         break;
6544       case MAT_REUSE_MATRIX:
6545         ierr = MatSubMatrixUpdate(*newmat,mat,isrow,iscoltmp);CHKERRQ(ierr);
6546         break;
6547       default: SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Invalid MatReuse, must be either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX");
6548     }
6549     if (!iscol) {ierr = ISDestroy(iscoltmp);CHKERRQ(ierr);}
6550     PetscFunctionReturn(0);
6551   }
6552 
6553   if (!mat->ops->getsubmatrix) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6554   ierr = (*mat->ops->getsubmatrix)(mat,isrow,iscoltmp,cll,newmat);CHKERRQ(ierr);
6555   if (!iscol) {ierr = ISDestroy(iscoltmp);CHKERRQ(ierr);}
6556   ierr = PetscObjectStateIncrease((PetscObject)*newmat);CHKERRQ(ierr);
6557   PetscFunctionReturn(0);
6558 }
6559 
6560 #undef __FUNCT__
6561 #define __FUNCT__ "MatStashSetInitialSize"
6562 /*@
6563    MatStashSetInitialSize - sets the sizes of the matrix stash, that is
6564    used during the assembly process to store values that belong to
6565    other processors.
6566 
6567    Not Collective
6568 
6569    Input Parameters:
6570 +  mat   - the matrix
6571 .  size  - the initial size of the stash.
6572 -  bsize - the initial size of the block-stash(if used).
6573 
6574    Options Database Keys:
6575 +   -matstash_initial_size <size> or <size0,size1,...sizep-1>
6576 -   -matstash_block_initial_size <bsize>  or <bsize0,bsize1,...bsizep-1>
6577 
6578    Level: intermediate
6579 
6580    Notes:
6581      The block-stash is used for values set with MatSetValuesBlocked() while
6582      the stash is used for values set with MatSetValues()
6583 
6584      Run with the option -info and look for output of the form
6585      MatAssemblyBegin_MPIXXX:Stash has MM entries, uses nn mallocs.
6586      to determine the appropriate value, MM, to use for size and
6587      MatAssemblyBegin_MPIXXX:Block-Stash has BMM entries, uses nn mallocs.
6588      to determine the value, BMM to use for bsize
6589 
6590    Concepts: stash^setting matrix size
6591    Concepts: matrices^stash
6592 
6593 .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashGetInfo()
6594 
6595 @*/
6596 PetscErrorCode PETSCMAT_DLLEXPORT MatStashSetInitialSize(Mat mat,PetscInt size, PetscInt bsize)
6597 {
6598   PetscErrorCode ierr;
6599 
6600   PetscFunctionBegin;
6601   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6602   PetscValidType(mat,1);
6603   ierr = MatStashSetInitialSize_Private(&mat->stash,size);CHKERRQ(ierr);
6604   ierr = MatStashSetInitialSize_Private(&mat->bstash,bsize);CHKERRQ(ierr);
6605   PetscFunctionReturn(0);
6606 }
6607 
6608 #undef __FUNCT__
6609 #define __FUNCT__ "MatInterpolateAdd"
6610 /*@
6611    MatInterpolateAdd - w = y + A*x or A'*x depending on the shape of
6612      the matrix
6613 
6614    Neighbor-wise Collective on Mat
6615 
6616    Input Parameters:
6617 +  mat   - the matrix
6618 .  x,y - the vectors
6619 -  w - where the result is stored
6620 
6621    Level: intermediate
6622 
6623    Notes:
6624     w may be the same vector as y.
6625 
6626     This allows one to use either the restriction or interpolation (its transpose)
6627     matrix to do the interpolation
6628 
6629     Concepts: interpolation
6630 
6631 .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
6632 
6633 @*/
6634 PetscErrorCode PETSCMAT_DLLEXPORT MatInterpolateAdd(Mat A,Vec x,Vec y,Vec w)
6635 {
6636   PetscErrorCode ierr;
6637   PetscInt       M,N;
6638 
6639   PetscFunctionBegin;
6640   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
6641   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
6642   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
6643   PetscValidHeaderSpecific(w,VEC_CLASSID,4);
6644   PetscValidType(A,1);
6645   ierr = MatPreallocated(A);CHKERRQ(ierr);
6646   ierr = MatGetSize(A,&M,&N);CHKERRQ(ierr);
6647   if (N > M) {
6648     ierr = MatMultTransposeAdd(A,x,y,w);CHKERRQ(ierr);
6649   } else {
6650     ierr = MatMultAdd(A,x,y,w);CHKERRQ(ierr);
6651   }
6652   PetscFunctionReturn(0);
6653 }
6654 
6655 #undef __FUNCT__
6656 #define __FUNCT__ "MatInterpolate"
6657 /*@
6658    MatInterpolate - y = A*x or A'*x depending on the shape of
6659      the matrix
6660 
6661    Neighbor-wise Collective on Mat
6662 
6663    Input Parameters:
6664 +  mat   - the matrix
6665 -  x,y - the vectors
6666 
6667    Level: intermediate
6668 
6669    Notes:
6670     This allows one to use either the restriction or interpolation (its transpose)
6671     matrix to do the interpolation
6672 
6673    Concepts: matrices^interpolation
6674 
6675 .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
6676 
6677 @*/
6678 PetscErrorCode PETSCMAT_DLLEXPORT MatInterpolate(Mat A,Vec x,Vec y)
6679 {
6680   PetscErrorCode ierr;
6681   PetscInt       M,N;
6682 
6683   PetscFunctionBegin;
6684   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
6685   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
6686   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
6687   PetscValidType(A,1);
6688   ierr = MatPreallocated(A);CHKERRQ(ierr);
6689   ierr = MatGetSize(A,&M,&N);CHKERRQ(ierr);
6690   if (N > M) {
6691     ierr = MatMultTranspose(A,x,y);CHKERRQ(ierr);
6692   } else {
6693     ierr = MatMult(A,x,y);CHKERRQ(ierr);
6694   }
6695   PetscFunctionReturn(0);
6696 }
6697 
6698 #undef __FUNCT__
6699 #define __FUNCT__ "MatRestrict"
6700 /*@
6701    MatRestrict - y = A*x or A'*x
6702 
6703    Neighbor-wise Collective on Mat
6704 
6705    Input Parameters:
6706 +  mat   - the matrix
6707 -  x,y - the vectors
6708 
6709    Level: intermediate
6710 
6711    Notes:
6712     This allows one to use either the restriction or interpolation (its transpose)
6713     matrix to do the restriction
6714 
6715    Concepts: matrices^restriction
6716 
6717 .seealso: MatMultAdd(), MatMultTransposeAdd(), MatInterpolate()
6718 
6719 @*/
6720 PetscErrorCode PETSCMAT_DLLEXPORT MatRestrict(Mat A,Vec x,Vec y)
6721 {
6722   PetscErrorCode ierr;
6723   PetscInt       M,N;
6724 
6725   PetscFunctionBegin;
6726   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
6727   PetscValidHeaderSpecific(x,VEC_CLASSID,2);
6728   PetscValidHeaderSpecific(y,VEC_CLASSID,3);
6729   PetscValidType(A,1);
6730   ierr = MatPreallocated(A);CHKERRQ(ierr);
6731 
6732   ierr = MatGetSize(A,&M,&N);CHKERRQ(ierr);
6733   if (N > M) {
6734     ierr = MatMult(A,x,y);CHKERRQ(ierr);
6735   } else {
6736     ierr = MatMultTranspose(A,x,y);CHKERRQ(ierr);
6737   }
6738   PetscFunctionReturn(0);
6739 }
6740 
6741 #undef __FUNCT__
6742 #define __FUNCT__ "MatNullSpaceAttach"
6743 /*@
6744    MatNullSpaceAttach - attaches a null space to a matrix.
6745         This null space will be removed from the resulting vector whenever
6746         MatMult() is called
6747 
6748    Logically Collective on Mat and MatNullSpace
6749 
6750    Input Parameters:
6751 +  mat - the matrix
6752 -  nullsp - the null space object
6753 
6754    Level: developer
6755 
6756    Notes:
6757       Overwrites any previous null space that may have been attached
6758 
6759    Concepts: null space^attaching to matrix
6760 
6761 .seealso: MatCreate(), MatNullSpaceCreate()
6762 @*/
6763 PetscErrorCode PETSCMAT_DLLEXPORT MatNullSpaceAttach(Mat mat,MatNullSpace nullsp)
6764 {
6765   PetscErrorCode ierr;
6766 
6767   PetscFunctionBegin;
6768   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6769   PetscValidType(mat,1);
6770   PetscValidHeaderSpecific(nullsp,MAT_NULLSPACE_CLASSID,2);
6771   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6772   ierr = PetscObjectReference((PetscObject)nullsp);CHKERRQ(ierr);
6773   if (mat->nullsp) { ierr = MatNullSpaceDestroy(mat->nullsp);CHKERRQ(ierr); }
6774   mat->nullsp = nullsp;
6775   PetscFunctionReturn(0);
6776 }
6777 
6778 #undef __FUNCT__
6779 #define __FUNCT__ "MatICCFactor"
6780 /*@C
6781    MatICCFactor - Performs in-place incomplete Cholesky factorization of matrix.
6782 
6783    Collective on Mat
6784 
6785    Input Parameters:
6786 +  mat - the matrix
6787 .  row - row/column permutation
6788 .  fill - expected fill factor >= 1.0
6789 -  level - level of fill, for ICC(k)
6790 
6791    Notes:
6792    Probably really in-place only when level of fill is zero, otherwise allocates
6793    new space to store factored matrix and deletes previous memory.
6794 
6795    Most users should employ the simplified KSP interface for linear solvers
6796    instead of working directly with matrix algebra routines such as this.
6797    See, e.g., KSPCreate().
6798 
6799    Level: developer
6800 
6801    Concepts: matrices^incomplete Cholesky factorization
6802    Concepts: Cholesky factorization
6803 
6804 .seealso: MatICCFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
6805 
6806     Developer Note: fortran interface is not autogenerated as the f90
6807     interface defintion cannot be generated correctly [due to MatFactorInfo]
6808 
6809 @*/
6810 PetscErrorCode PETSCMAT_DLLEXPORT MatICCFactor(Mat mat,IS row,const MatFactorInfo* info)
6811 {
6812   PetscErrorCode ierr;
6813 
6814   PetscFunctionBegin;
6815   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6816   PetscValidType(mat,1);
6817   if (row) PetscValidHeaderSpecific(row,IS_CLASSID,2);
6818   PetscValidPointer(info,3);
6819   if (mat->rmap->N != mat->cmap->N) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONG,"matrix must be square");
6820   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6821   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6822   if (!mat->ops->iccfactor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6823   ierr = MatPreallocated(mat);CHKERRQ(ierr);
6824   ierr = (*mat->ops->iccfactor)(mat,row,info);CHKERRQ(ierr);
6825   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
6826   PetscFunctionReturn(0);
6827 }
6828 
6829 #undef __FUNCT__
6830 #define __FUNCT__ "MatSetValuesAdic"
6831 /*@
6832    MatSetValuesAdic - Sets values computed with ADIC automatic differentiation into a matrix.
6833 
6834    Not Collective
6835 
6836    Input Parameters:
6837 +  mat - the matrix
6838 -  v - the values compute with ADIC
6839 
6840    Level: developer
6841 
6842    Notes:
6843      Must call MatSetColoring() before using this routine. Also this matrix must already
6844      have its nonzero pattern determined.
6845 
6846 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
6847           MatSetValues(), MatSetColoring(), MatSetValuesAdifor()
6848 @*/
6849 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesAdic(Mat mat,void *v)
6850 {
6851   PetscErrorCode ierr;
6852 
6853   PetscFunctionBegin;
6854   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6855   PetscValidType(mat,1);
6856   PetscValidPointer(mat,2);
6857 
6858   if (!mat->assembled) {
6859     SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
6860   }
6861   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
6862   if (!mat->ops->setvaluesadic) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6863   ierr = (*mat->ops->setvaluesadic)(mat,v);CHKERRQ(ierr);
6864   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
6865   ierr = MatView_Private(mat);CHKERRQ(ierr);
6866   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
6867   PetscFunctionReturn(0);
6868 }
6869 
6870 
6871 #undef __FUNCT__
6872 #define __FUNCT__ "MatSetColoring"
6873 /*@
6874    MatSetColoring - Sets a coloring used by calls to MatSetValuesAdic()
6875 
6876    Not Collective
6877 
6878    Input Parameters:
6879 +  mat - the matrix
6880 -  coloring - the coloring
6881 
6882    Level: developer
6883 
6884 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
6885           MatSetValues(), MatSetValuesAdic()
6886 @*/
6887 PetscErrorCode PETSCMAT_DLLEXPORT MatSetColoring(Mat mat,ISColoring coloring)
6888 {
6889   PetscErrorCode ierr;
6890 
6891   PetscFunctionBegin;
6892   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6893   PetscValidType(mat,1);
6894   PetscValidPointer(coloring,2);
6895 
6896   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
6897   if (!mat->ops->setcoloring) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6898   ierr = (*mat->ops->setcoloring)(mat,coloring);CHKERRQ(ierr);
6899   PetscFunctionReturn(0);
6900 }
6901 
6902 #undef __FUNCT__
6903 #define __FUNCT__ "MatSetValuesAdifor"
6904 /*@
6905    MatSetValuesAdifor - Sets values computed with automatic differentiation into a matrix.
6906 
6907    Not Collective
6908 
6909    Input Parameters:
6910 +  mat - the matrix
6911 .  nl - leading dimension of v
6912 -  v - the values compute with ADIFOR
6913 
6914    Level: developer
6915 
6916    Notes:
6917      Must call MatSetColoring() before using this routine. Also this matrix must already
6918      have its nonzero pattern determined.
6919 
6920 .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
6921           MatSetValues(), MatSetColoring()
6922 @*/
6923 PetscErrorCode PETSCMAT_DLLEXPORT MatSetValuesAdifor(Mat mat,PetscInt nl,void *v)
6924 {
6925   PetscErrorCode ierr;
6926 
6927   PetscFunctionBegin;
6928   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6929   PetscValidType(mat,1);
6930   PetscValidPointer(v,3);
6931 
6932   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
6933   ierr = PetscLogEventBegin(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
6934   if (!mat->ops->setvaluesadifor) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6935   ierr = (*mat->ops->setvaluesadifor)(mat,nl,v);CHKERRQ(ierr);
6936   ierr = PetscLogEventEnd(MAT_SetValues,mat,0,0,0);CHKERRQ(ierr);
6937   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
6938   PetscFunctionReturn(0);
6939 }
6940 
6941 #undef __FUNCT__
6942 #define __FUNCT__ "MatDiagonalScaleLocal"
6943 /*@
6944    MatDiagonalScaleLocal - Scales columns of a matrix given the scaling values including the
6945          ghosted ones.
6946 
6947    Not Collective
6948 
6949    Input Parameters:
6950 +  mat - the matrix
6951 -  diag = the diagonal values, including ghost ones
6952 
6953    Level: developer
6954 
6955    Notes: Works only for MPIAIJ and MPIBAIJ matrices
6956 
6957 .seealso: MatDiagonalScale()
6958 @*/
6959 PetscErrorCode PETSCMAT_DLLEXPORT MatDiagonalScaleLocal(Mat mat,Vec diag)
6960 {
6961   PetscErrorCode ierr;
6962   PetscMPIInt    size;
6963 
6964   PetscFunctionBegin;
6965   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
6966   PetscValidHeaderSpecific(diag,VEC_CLASSID,2);
6967   PetscValidType(mat,1);
6968 
6969   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
6970   ierr = PetscLogEventBegin(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
6971   ierr = MPI_Comm_size(((PetscObject)mat)->comm,&size);CHKERRQ(ierr);
6972   if (size == 1) {
6973     PetscInt n,m;
6974     ierr = VecGetSize(diag,&n);CHKERRQ(ierr);
6975     ierr = MatGetSize(mat,0,&m);CHKERRQ(ierr);
6976     if (m == n) {
6977       ierr = MatDiagonalScale(mat,0,diag);CHKERRQ(ierr);
6978     } else {
6979       SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only supported for sequential matrices when no ghost points/periodic conditions");
6980     }
6981   } else {
6982     PetscErrorCode (*f)(Mat,Vec);
6983     ierr = PetscObjectQueryFunction((PetscObject)mat,"MatDiagonalScaleLocal_C",(void (**)(void))&f);CHKERRQ(ierr);
6984     if (f) {
6985       ierr = (*f)(mat,diag);CHKERRQ(ierr);
6986     } else {
6987       SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only supported for MPIAIJ and MPIBAIJ parallel matrices");
6988     }
6989   }
6990   ierr = PetscLogEventEnd(MAT_Scale,mat,0,0,0);CHKERRQ(ierr);
6991   ierr = PetscObjectStateIncrease((PetscObject)mat);CHKERRQ(ierr);
6992   PetscFunctionReturn(0);
6993 }
6994 
6995 #undef __FUNCT__
6996 #define __FUNCT__ "MatGetInertia"
6997 /*@
6998    MatGetInertia - Gets the inertia from a factored matrix
6999 
7000    Collective on Mat
7001 
7002    Input Parameter:
7003 .  mat - the matrix
7004 
7005    Output Parameters:
7006 +   nneg - number of negative eigenvalues
7007 .   nzero - number of zero eigenvalues
7008 -   npos - number of positive eigenvalues
7009 
7010    Level: advanced
7011 
7012    Notes: Matrix must have been factored by MatCholeskyFactor()
7013 
7014 
7015 @*/
7016 PetscErrorCode PETSCMAT_DLLEXPORT MatGetInertia(Mat mat,PetscInt *nneg,PetscInt *nzero,PetscInt *npos)
7017 {
7018   PetscErrorCode ierr;
7019 
7020   PetscFunctionBegin;
7021   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
7022   PetscValidType(mat,1);
7023   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
7024   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Numeric factor mat is not assembled");
7025   if (!mat->ops->getinertia) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7026   ierr = (*mat->ops->getinertia)(mat,nneg,nzero,npos);CHKERRQ(ierr);
7027   PetscFunctionReturn(0);
7028 }
7029 
7030 /* ----------------------------------------------------------------*/
7031 #undef __FUNCT__
7032 #define __FUNCT__ "MatSolves"
7033 /*@C
7034    MatSolves - Solves A x = b, given a factored matrix, for a collection of vectors
7035 
7036    Neighbor-wise Collective on Mat and Vecs
7037 
7038    Input Parameters:
7039 +  mat - the factored matrix
7040 -  b - the right-hand-side vectors
7041 
7042    Output Parameter:
7043 .  x - the result vectors
7044 
7045    Notes:
7046    The vectors b and x cannot be the same.  I.e., one cannot
7047    call MatSolves(A,x,x).
7048 
7049    Notes:
7050    Most users should employ the simplified KSP interface for linear solvers
7051    instead of working directly with matrix algebra routines such as this.
7052    See, e.g., KSPCreate().
7053 
7054    Level: developer
7055 
7056    Concepts: matrices^triangular solves
7057 
7058 .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd(), MatSolve()
7059 @*/
7060 PetscErrorCode PETSCMAT_DLLEXPORT MatSolves(Mat mat,Vecs b,Vecs x)
7061 {
7062   PetscErrorCode ierr;
7063 
7064   PetscFunctionBegin;
7065   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
7066   PetscValidType(mat,1);
7067   if (x == b) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_IDN,"x and b must be different vectors");
7068   if (!mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
7069   if (!mat->rmap->N && !mat->cmap->N) PetscFunctionReturn(0);
7070 
7071   if (!mat->ops->solves) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7072   ierr = MatPreallocated(mat);CHKERRQ(ierr);
7073   ierr = PetscLogEventBegin(MAT_Solves,mat,0,0,0);CHKERRQ(ierr);
7074   ierr = (*mat->ops->solves)(mat,b,x);CHKERRQ(ierr);
7075   ierr = PetscLogEventEnd(MAT_Solves,mat,0,0,0);CHKERRQ(ierr);
7076   PetscFunctionReturn(0);
7077 }
7078 
7079 #undef __FUNCT__
7080 #define __FUNCT__ "MatIsSymmetric"
7081 /*@
7082    MatIsSymmetric - Test whether a matrix is symmetric
7083 
7084    Collective on Mat
7085 
7086    Input Parameter:
7087 +  A - the matrix to test
7088 -  tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact transpose)
7089 
7090    Output Parameters:
7091 .  flg - the result
7092 
7093    Level: intermediate
7094 
7095    Concepts: matrix^symmetry
7096 
7097 .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetricKnown()
7098 @*/
7099 PetscErrorCode PETSCMAT_DLLEXPORT MatIsSymmetric(Mat A,PetscReal tol,PetscTruth *flg)
7100 {
7101   PetscErrorCode ierr;
7102 
7103   PetscFunctionBegin;
7104   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7105   PetscValidPointer(flg,2);
7106 
7107   if (!A->symmetric_set) {
7108     if (!A->ops->issymmetric) {
7109       const MatType mattype;
7110       ierr = MatGetType(A,&mattype);CHKERRQ(ierr);
7111       SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type <%s> does not support checking for symmetric",mattype);
7112     }
7113     ierr = (*A->ops->issymmetric)(A,tol,flg);CHKERRQ(ierr);
7114     if (!tol) {
7115       A->symmetric_set = PETSC_TRUE;
7116       A->symmetric = *flg;
7117       if (A->symmetric) {
7118 	A->structurally_symmetric_set = PETSC_TRUE;
7119 	A->structurally_symmetric     = PETSC_TRUE;
7120       }
7121     }
7122   } else if (A->symmetric) {
7123     *flg = PETSC_TRUE;
7124   } else if (!tol) {
7125     *flg = PETSC_FALSE;
7126   } else {
7127     if (!A->ops->issymmetric) {
7128       const MatType mattype;
7129       ierr = MatGetType(A,&mattype);CHKERRQ(ierr);
7130       SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type <%s> does not support checking for symmetric",mattype);
7131     }
7132     ierr = (*A->ops->issymmetric)(A,tol,flg);CHKERRQ(ierr);
7133   }
7134   PetscFunctionReturn(0);
7135 }
7136 
7137 #undef __FUNCT__
7138 #define __FUNCT__ "MatIsHermitian"
7139 /*@
7140    MatIsHermitian - Test whether a matrix is Hermitian
7141 
7142    Collective on Mat
7143 
7144    Input Parameter:
7145 +  A - the matrix to test
7146 -  tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact Hermitian)
7147 
7148    Output Parameters:
7149 .  flg - the result
7150 
7151    Level: intermediate
7152 
7153    Concepts: matrix^symmetry
7154 
7155 .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetricKnown()
7156 @*/
7157 PetscErrorCode PETSCMAT_DLLEXPORT MatIsHermitian(Mat A,PetscReal tol,PetscTruth *flg)
7158 {
7159   PetscErrorCode ierr;
7160 
7161   PetscFunctionBegin;
7162   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7163   PetscValidPointer(flg,2);
7164 
7165   if (!A->hermitian_set) {
7166     if (!A->ops->ishermitian) {
7167       const MatType mattype;
7168       ierr = MatGetType(A,&mattype);CHKERRQ(ierr);
7169       SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type <%s> does not support checking for hermitian",mattype);
7170     }
7171     ierr = (*A->ops->ishermitian)(A,tol,flg);CHKERRQ(ierr);
7172     if (!tol) {
7173       A->hermitian_set = PETSC_TRUE;
7174       A->hermitian = *flg;
7175       if (A->hermitian) {
7176 	A->structurally_symmetric_set = PETSC_TRUE;
7177 	A->structurally_symmetric     = PETSC_TRUE;
7178       }
7179     }
7180   } else if (A->hermitian) {
7181     *flg = PETSC_TRUE;
7182   } else if (!tol) {
7183     *flg = PETSC_FALSE;
7184   } else {
7185     if (!A->ops->ishermitian) {
7186       const MatType mattype;
7187       ierr = MatGetType(A,&mattype);CHKERRQ(ierr);
7188       SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type <%s> does not support checking for hermitian",mattype);
7189     }
7190     ierr = (*A->ops->ishermitian)(A,tol,flg);CHKERRQ(ierr);
7191   }
7192   PetscFunctionReturn(0);
7193 }
7194 
7195 #undef __FUNCT__
7196 #define __FUNCT__ "MatIsSymmetricKnown"
7197 /*@
7198    MatIsSymmetricKnown - Checks the flag on the matrix to see if it is symmetric.
7199 
7200    Not Collective
7201 
7202    Input Parameter:
7203 .  A - the matrix to check
7204 
7205    Output Parameters:
7206 +  set - if the symmetric flag is set (this tells you if the next flag is valid)
7207 -  flg - the result
7208 
7209    Level: advanced
7210 
7211    Concepts: matrix^symmetry
7212 
7213    Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsSymmetric()
7214          if you want it explicitly checked
7215 
7216 .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
7217 @*/
7218 PetscErrorCode PETSCMAT_DLLEXPORT MatIsSymmetricKnown(Mat A,PetscTruth *set,PetscTruth *flg)
7219 {
7220   PetscFunctionBegin;
7221   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7222   PetscValidPointer(set,2);
7223   PetscValidPointer(flg,3);
7224   if (A->symmetric_set) {
7225     *set = PETSC_TRUE;
7226     *flg = A->symmetric;
7227   } else {
7228     *set = PETSC_FALSE;
7229   }
7230   PetscFunctionReturn(0);
7231 }
7232 
7233 #undef __FUNCT__
7234 #define __FUNCT__ "MatIsHermitianKnown"
7235 /*@
7236    MatIsHermitianKnown - Checks the flag on the matrix to see if it is hermitian.
7237 
7238    Not Collective
7239 
7240    Input Parameter:
7241 .  A - the matrix to check
7242 
7243    Output Parameters:
7244 +  set - if the hermitian flag is set (this tells you if the next flag is valid)
7245 -  flg - the result
7246 
7247    Level: advanced
7248 
7249    Concepts: matrix^symmetry
7250 
7251    Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsHermitian()
7252          if you want it explicitly checked
7253 
7254 .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
7255 @*/
7256 PetscErrorCode PETSCMAT_DLLEXPORT MatIsHermitianKnown(Mat A,PetscTruth *set,PetscTruth *flg)
7257 {
7258   PetscFunctionBegin;
7259   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7260   PetscValidPointer(set,2);
7261   PetscValidPointer(flg,3);
7262   if (A->hermitian_set) {
7263     *set = PETSC_TRUE;
7264     *flg = A->hermitian;
7265   } else {
7266     *set = PETSC_FALSE;
7267   }
7268   PetscFunctionReturn(0);
7269 }
7270 
7271 #undef __FUNCT__
7272 #define __FUNCT__ "MatIsStructurallySymmetric"
7273 /*@
7274    MatIsStructurallySymmetric - Test whether a matrix is structurally symmetric
7275 
7276    Collective on Mat
7277 
7278    Input Parameter:
7279 .  A - the matrix to test
7280 
7281    Output Parameters:
7282 .  flg - the result
7283 
7284    Level: intermediate
7285 
7286    Concepts: matrix^symmetry
7287 
7288 .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsSymmetric(), MatSetOption()
7289 @*/
7290 PetscErrorCode PETSCMAT_DLLEXPORT MatIsStructurallySymmetric(Mat A,PetscTruth *flg)
7291 {
7292   PetscErrorCode ierr;
7293 
7294   PetscFunctionBegin;
7295   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7296   PetscValidPointer(flg,2);
7297   if (!A->structurally_symmetric_set) {
7298     if (!A->ops->isstructurallysymmetric) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_SUP,"Matrix does not support checking for structural symmetric");
7299     ierr = (*A->ops->isstructurallysymmetric)(A,&A->structurally_symmetric);CHKERRQ(ierr);
7300     A->structurally_symmetric_set = PETSC_TRUE;
7301   }
7302   *flg = A->structurally_symmetric;
7303   PetscFunctionReturn(0);
7304 }
7305 
7306 #undef __FUNCT__
7307 #define __FUNCT__ "MatStashGetInfo"
7308 extern PetscErrorCode MatStashGetInfo_Private(MatStash*,PetscInt*,PetscInt*);
7309 /*@
7310    MatStashGetInfo - Gets how many values are currently in the vector stash, i.e. need
7311        to be communicated to other processors during the MatAssemblyBegin/End() process
7312 
7313     Not collective
7314 
7315    Input Parameter:
7316 .   vec - the vector
7317 
7318    Output Parameters:
7319 +   nstash   - the size of the stash
7320 .   reallocs - the number of additional mallocs incurred.
7321 .   bnstash   - the size of the block stash
7322 -   breallocs - the number of additional mallocs incurred.in the block stash
7323 
7324    Level: advanced
7325 
7326 .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashSetInitialSize()
7327 
7328 @*/
7329 PetscErrorCode PETSCMAT_DLLEXPORT MatStashGetInfo(Mat mat,PetscInt *nstash,PetscInt *reallocs,PetscInt *bnstash,PetscInt *breallocs)
7330 {
7331   PetscErrorCode ierr;
7332   PetscFunctionBegin;
7333   ierr = MatStashGetInfo_Private(&mat->stash,nstash,reallocs);CHKERRQ(ierr);
7334   ierr = MatStashGetInfo_Private(&mat->bstash,bnstash,breallocs);CHKERRQ(ierr);
7335   PetscFunctionReturn(0);
7336 }
7337 
7338 #undef __FUNCT__
7339 #define __FUNCT__ "MatGetVecs"
7340 /*@C
7341    MatGetVecs - Get vector(s) compatible with the matrix, i.e. with the same
7342      parallel layout
7343 
7344    Collective on Mat
7345 
7346    Input Parameter:
7347 .  mat - the matrix
7348 
7349    Output Parameter:
7350 +   right - (optional) vector that the matrix can be multiplied against
7351 -   left - (optional) vector that the matrix vector product can be stored in
7352 
7353   Level: advanced
7354 
7355 .seealso: MatCreate()
7356 @*/
7357 PetscErrorCode PETSCMAT_DLLEXPORT MatGetVecs(Mat mat,Vec *right,Vec *left)
7358 {
7359   PetscErrorCode ierr;
7360 
7361   PetscFunctionBegin;
7362   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
7363   PetscValidType(mat,1);
7364   ierr = MatPreallocated(mat);CHKERRQ(ierr);
7365   if (mat->ops->getvecs) {
7366     ierr = (*mat->ops->getvecs)(mat,right,left);CHKERRQ(ierr);
7367   } else {
7368     PetscMPIInt size;
7369     ierr = MPI_Comm_size(((PetscObject)mat)->comm, &size);CHKERRQ(ierr);
7370     if (right) {
7371       ierr = VecCreate(((PetscObject)mat)->comm,right);CHKERRQ(ierr);
7372       ierr = VecSetSizes(*right,mat->cmap->n,PETSC_DETERMINE);CHKERRQ(ierr);
7373       ierr = VecSetBlockSize(*right,mat->rmap->bs);CHKERRQ(ierr);
7374       if (size > 1) {
7375         /* New vectors uses Mat cmap and does not create a new one */
7376 	ierr = PetscLayoutDestroy((*right)->map);CHKERRQ(ierr);
7377 	(*right)->map = mat->cmap;
7378 	mat->cmap->refcnt++;
7379 
7380         ierr = VecSetType(*right,VECMPI);CHKERRQ(ierr);
7381       } else {ierr = VecSetType(*right,VECSEQ);CHKERRQ(ierr);}
7382     }
7383     if (left) {
7384       ierr = VecCreate(((PetscObject)mat)->comm,left);CHKERRQ(ierr);
7385       ierr = VecSetSizes(*left,mat->rmap->n,PETSC_DETERMINE);CHKERRQ(ierr);
7386       ierr = VecSetBlockSize(*left,mat->rmap->bs);CHKERRQ(ierr);
7387       if (size > 1) {
7388         /* New vectors uses Mat rmap and does not create a new one */
7389 	ierr = PetscLayoutDestroy((*left)->map);CHKERRQ(ierr);
7390 	(*left)->map = mat->rmap;
7391 	mat->rmap->refcnt++;
7392 
7393         ierr = VecSetType(*left,VECMPI);CHKERRQ(ierr);
7394       } else {ierr = VecSetType(*left,VECSEQ);CHKERRQ(ierr);}
7395     }
7396   }
7397   if (mat->mapping) {
7398     if (right) {ierr = VecSetLocalToGlobalMapping(*right,mat->mapping);CHKERRQ(ierr);}
7399     if (left) {ierr = VecSetLocalToGlobalMapping(*left,mat->mapping);CHKERRQ(ierr);}
7400   }
7401   if (mat->bmapping) {
7402     if (right) {ierr = VecSetLocalToGlobalMappingBlock(*right,mat->bmapping);CHKERRQ(ierr);}
7403     if (left) {ierr = VecSetLocalToGlobalMappingBlock(*left,mat->bmapping);CHKERRQ(ierr);}
7404   }
7405   PetscFunctionReturn(0);
7406 }
7407 
7408 #undef __FUNCT__
7409 #define __FUNCT__ "MatFactorInfoInitialize"
7410 /*@C
7411    MatFactorInfoInitialize - Initializes a MatFactorInfo data structure
7412      with default values.
7413 
7414    Not Collective
7415 
7416    Input Parameters:
7417 .    info - the MatFactorInfo data structure
7418 
7419 
7420    Notes: The solvers are generally used through the KSP and PC objects, for example
7421           PCLU, PCILU, PCCHOLESKY, PCICC
7422 
7423    Level: developer
7424 
7425 .seealso: MatFactorInfo
7426 
7427     Developer Note: fortran interface is not autogenerated as the f90
7428     interface defintion cannot be generated correctly [due to MatFactorInfo]
7429 
7430 @*/
7431 
7432 PetscErrorCode PETSCMAT_DLLEXPORT MatFactorInfoInitialize(MatFactorInfo *info)
7433 {
7434   PetscErrorCode ierr;
7435 
7436   PetscFunctionBegin;
7437   ierr = PetscMemzero(info,sizeof(MatFactorInfo));CHKERRQ(ierr);
7438   PetscFunctionReturn(0);
7439 }
7440 
7441 #undef __FUNCT__
7442 #define __FUNCT__ "MatPtAP"
7443 /*@
7444    MatPtAP - Creates the matrix product C = P^T * A * P
7445 
7446    Neighbor-wise Collective on Mat
7447 
7448    Input Parameters:
7449 +  A - the matrix
7450 .  P - the projection matrix
7451 .  scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7452 -  fill - expected fill as ratio of nnz(C)/nnz(A)
7453 
7454    Output Parameters:
7455 .  C - the product matrix
7456 
7457    Notes:
7458    C will be created and must be destroyed by the user with MatDestroy().
7459 
7460    This routine is currently only implemented for pairs of AIJ matrices and classes
7461    which inherit from AIJ.
7462 
7463    Level: intermediate
7464 
7465 .seealso: MatPtAPSymbolic(), MatPtAPNumeric(), MatMatMult()
7466 @*/
7467 PetscErrorCode PETSCMAT_DLLEXPORT MatPtAP(Mat A,Mat P,MatReuse scall,PetscReal fill,Mat *C)
7468 {
7469   PetscErrorCode ierr;
7470 
7471   PetscFunctionBegin;
7472   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7473   PetscValidType(A,1);
7474   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7475   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7476   PetscValidHeaderSpecific(P,MAT_CLASSID,2);
7477   PetscValidType(P,2);
7478   ierr = MatPreallocated(P);CHKERRQ(ierr);
7479   if (!P->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7480   if (P->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7481   PetscValidPointer(C,3);
7482   if (P->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
7483   if (fill < 1.0) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Expected fill=%G must be >= 1.0",fill);
7484   ierr = MatPreallocated(A);CHKERRQ(ierr);
7485 
7486   if (!A->ops->ptap) {
7487     const MatType mattype;
7488     ierr = MatGetType(A,&mattype);CHKERRQ(ierr);
7489     SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"Matrix of type <%s> does not support PtAP",mattype);
7490   }
7491   ierr = PetscLogEventBegin(MAT_PtAP,A,P,0,0);CHKERRQ(ierr);
7492   ierr = (*A->ops->ptap)(A,P,scall,fill,C);CHKERRQ(ierr);
7493   ierr = PetscLogEventEnd(MAT_PtAP,A,P,0,0);CHKERRQ(ierr);
7494 
7495   PetscFunctionReturn(0);
7496 }
7497 
7498 #undef __FUNCT__
7499 #define __FUNCT__ "MatPtAPNumeric"
7500 /*@
7501    MatPtAPNumeric - Computes the matrix product C = P^T * A * P
7502 
7503    Neighbor-wise Collective on Mat
7504 
7505    Input Parameters:
7506 +  A - the matrix
7507 -  P - the projection matrix
7508 
7509    Output Parameters:
7510 .  C - the product matrix
7511 
7512    Notes:
7513    C must have been created by calling MatPtAPSymbolic and must be destroyed by
7514    the user using MatDeatroy().
7515 
7516    This routine is currently only implemented for pairs of AIJ matrices and classes
7517    which inherit from AIJ.  C will be of type MATAIJ.
7518 
7519    Level: intermediate
7520 
7521 .seealso: MatPtAP(), MatPtAPSymbolic(), MatMatMultNumeric()
7522 @*/
7523 PetscErrorCode PETSCMAT_DLLEXPORT MatPtAPNumeric(Mat A,Mat P,Mat C)
7524 {
7525   PetscErrorCode ierr;
7526 
7527   PetscFunctionBegin;
7528   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7529   PetscValidType(A,1);
7530   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7531   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7532   PetscValidHeaderSpecific(P,MAT_CLASSID,2);
7533   PetscValidType(P,2);
7534   ierr = MatPreallocated(P);CHKERRQ(ierr);
7535   if (!P->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7536   if (P->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7537   PetscValidHeaderSpecific(C,MAT_CLASSID,3);
7538   PetscValidType(C,3);
7539   ierr = MatPreallocated(C);CHKERRQ(ierr);
7540   if (C->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7541   if (P->cmap->N!=C->rmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->cmap->N,C->rmap->N);
7542   if (P->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
7543   if (A->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
7544   if (P->cmap->N!=C->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->cmap->N,C->cmap->N);
7545   ierr = MatPreallocated(A);CHKERRQ(ierr);
7546 
7547   ierr = PetscLogEventBegin(MAT_PtAPNumeric,A,P,0,0);CHKERRQ(ierr);
7548   ierr = (*A->ops->ptapnumeric)(A,P,C);CHKERRQ(ierr);
7549   ierr = PetscLogEventEnd(MAT_PtAPNumeric,A,P,0,0);CHKERRQ(ierr);
7550   PetscFunctionReturn(0);
7551 }
7552 
7553 #undef __FUNCT__
7554 #define __FUNCT__ "MatPtAPSymbolic"
7555 /*@
7556    MatPtAPSymbolic - Creates the (i,j) structure of the matrix product C = P^T * A * P
7557 
7558    Neighbor-wise Collective on Mat
7559 
7560    Input Parameters:
7561 +  A - the matrix
7562 -  P - the projection matrix
7563 
7564    Output Parameters:
7565 .  C - the (i,j) structure of the product matrix
7566 
7567    Notes:
7568    C will be created and must be destroyed by the user with MatDestroy().
7569 
7570    This routine is currently only implemented for pairs of SeqAIJ matrices and classes
7571    which inherit from SeqAIJ.  C will be of type MATSEQAIJ.  The product is computed using
7572    this (i,j) structure by calling MatPtAPNumeric().
7573 
7574    Level: intermediate
7575 
7576 .seealso: MatPtAP(), MatPtAPNumeric(), MatMatMultSymbolic()
7577 @*/
7578 PetscErrorCode PETSCMAT_DLLEXPORT MatPtAPSymbolic(Mat A,Mat P,PetscReal fill,Mat *C)
7579 {
7580   PetscErrorCode ierr;
7581 
7582   PetscFunctionBegin;
7583   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7584   PetscValidType(A,1);
7585   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7586   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7587   if (fill <1.0) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Expected fill=%G must be >= 1.0",fill);
7588   PetscValidHeaderSpecific(P,MAT_CLASSID,2);
7589   PetscValidType(P,2);
7590   ierr = MatPreallocated(P);CHKERRQ(ierr);
7591   if (!P->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7592   if (P->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7593   PetscValidPointer(C,3);
7594 
7595   if (P->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
7596   if (A->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
7597   ierr = MatPreallocated(A);CHKERRQ(ierr);
7598   ierr = PetscLogEventBegin(MAT_PtAPSymbolic,A,P,0,0);CHKERRQ(ierr);
7599   ierr = (*A->ops->ptapsymbolic)(A,P,fill,C);CHKERRQ(ierr);
7600   ierr = PetscLogEventEnd(MAT_PtAPSymbolic,A,P,0,0);CHKERRQ(ierr);
7601 
7602   ierr = MatSetBlockSize(*C,A->rmap->bs);CHKERRQ(ierr);
7603 
7604   PetscFunctionReturn(0);
7605 }
7606 
7607 #undef __FUNCT__
7608 #define __FUNCT__ "MatMatMult"
7609 /*@
7610    MatMatMult - Performs Matrix-Matrix Multiplication C=A*B.
7611 
7612    Neighbor-wise Collective on Mat
7613 
7614    Input Parameters:
7615 +  A - the left matrix
7616 .  B - the right matrix
7617 .  scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7618 -  fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if you do not have a good estimate
7619           if the result is a dense matrix this is irrelevent
7620 
7621    Output Parameters:
7622 .  C - the product matrix
7623 
7624    Notes:
7625    Unless scall is MAT_REUSE_MATRIX C will be created.
7626 
7627    MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call
7628 
7629    To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
7630    actually needed.
7631 
7632    If you have many matrices with the same non-zero structure to multiply, you
7633    should either
7634 $   1) use MAT_REUSE_MATRIX in all calls but the first or
7635 $   2) call MatMatMultSymbolic() once and then MatMatMultNumeric() for each product needed
7636 
7637    Level: intermediate
7638 
7639 .seealso: MatMatMultSymbolic(), MatMatMultNumeric(), MatPtAP()
7640 @*/
7641 PetscErrorCode PETSCMAT_DLLEXPORT MatMatMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
7642 {
7643   PetscErrorCode ierr;
7644   PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
7645   PetscErrorCode (*fB)(Mat,Mat,MatReuse,PetscReal,Mat*);
7646   PetscErrorCode (*mult)(Mat,Mat,MatReuse,PetscReal,Mat *)=PETSC_NULL;
7647 
7648   PetscFunctionBegin;
7649   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7650   PetscValidType(A,1);
7651   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7652   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7653   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
7654   PetscValidType(B,2);
7655   ierr = MatPreallocated(B);CHKERRQ(ierr);
7656   if (!B->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7657   if (B->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7658   PetscValidPointer(C,3);
7659   if (B->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
7660   if (scall == MAT_REUSE_MATRIX){
7661     PetscValidPointer(*C,5);
7662     PetscValidHeaderSpecific(*C,MAT_CLASSID,5);
7663   }
7664   if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
7665   if (fill < 1.0) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Expected fill=%G must be >= 1.0",fill);
7666   ierr = MatPreallocated(A);CHKERRQ(ierr);
7667 
7668   fA = A->ops->matmult;
7669   fB = B->ops->matmult;
7670   if (fB == fA) {
7671     if (!fB) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"MatMatMult not supported for B of type %s",((PetscObject)B)->type_name);
7672     mult = fB;
7673   } else {
7674     /* dispatch based on the type of A and B */
7675     char  multname[256];
7676     ierr = PetscStrcpy(multname,"MatMatMult_");CHKERRQ(ierr);
7677     ierr = PetscStrcat(multname,((PetscObject)A)->type_name);CHKERRQ(ierr);
7678     ierr = PetscStrcat(multname,"_");CHKERRQ(ierr);
7679     ierr = PetscStrcat(multname,((PetscObject)B)->type_name);CHKERRQ(ierr);
7680     ierr = PetscStrcat(multname,"_C");CHKERRQ(ierr); /* e.g., multname = "MatMatMult_seqdense_seqaij_C" */
7681     ierr = PetscObjectQueryFunction((PetscObject)B,multname,(void (**)(void))&mult);CHKERRQ(ierr);
7682     if (!mult) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_INCOMP,"MatMatMult requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
7683   }
7684   ierr = PetscLogEventBegin(MAT_MatMult,A,B,0,0);CHKERRQ(ierr);
7685   ierr = (*mult)(A,B,scall,fill,C);CHKERRQ(ierr);
7686   ierr = PetscLogEventEnd(MAT_MatMult,A,B,0,0);CHKERRQ(ierr);
7687   PetscFunctionReturn(0);
7688 }
7689 
7690 #undef __FUNCT__
7691 #define __FUNCT__ "MatMatMultSymbolic"
7692 /*@
7693    MatMatMultSymbolic - Performs construction, preallocation, and computes the ij structure
7694    of the matrix-matrix product C=A*B.  Call this routine before calling MatMatMultNumeric().
7695 
7696    Neighbor-wise Collective on Mat
7697 
7698    Input Parameters:
7699 +  A - the left matrix
7700 .  B - the right matrix
7701 -  fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if you do not have a good estimate,
7702       if C is a dense matrix this is irrelevent
7703 
7704    Output Parameters:
7705 .  C - the product matrix
7706 
7707    Notes:
7708    Unless scall is MAT_REUSE_MATRIX C will be created.
7709 
7710    To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
7711    actually needed.
7712 
7713    This routine is currently implemented for
7714     - pairs of AIJ matrices and classes which inherit from AIJ, C will be of type AIJ
7715     - pairs of AIJ (A) and Dense (B) matrix, C will be of type Dense.
7716     - pairs of Dense (A) and AIJ (B) matrix, C will be of type Dense.
7717 
7718    Level: intermediate
7719 
7720    Developers Note: There are ways to estimate the number of nonzeros in the resulting product, see for example, http://arxiv.org/abs/1006.4173
7721      We should incorporate them into PETSc.
7722 
7723 .seealso: MatMatMult(), MatMatMultNumeric()
7724 @*/
7725 PetscErrorCode PETSCMAT_DLLEXPORT MatMatMultSymbolic(Mat A,Mat B,PetscReal fill,Mat *C)
7726 {
7727   PetscErrorCode ierr;
7728   PetscErrorCode (*Asymbolic)(Mat,Mat,PetscReal,Mat *);
7729   PetscErrorCode (*Bsymbolic)(Mat,Mat,PetscReal,Mat *);
7730   PetscErrorCode (*symbolic)(Mat,Mat,PetscReal,Mat *)=PETSC_NULL;
7731 
7732   PetscFunctionBegin;
7733   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7734   PetscValidType(A,1);
7735   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7736   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7737 
7738   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
7739   PetscValidType(B,2);
7740   ierr = MatPreallocated(B);CHKERRQ(ierr);
7741   if (!B->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7742   if (B->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7743   PetscValidPointer(C,3);
7744 
7745   if (B->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
7746   if (fill == PETSC_DEFAULT) fill = 2.0;
7747   if (fill < 1.0) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Expected fill=%G must be > 1.0",fill);
7748   ierr = MatPreallocated(A);CHKERRQ(ierr);
7749 
7750   Asymbolic = A->ops->matmultsymbolic;
7751   Bsymbolic = B->ops->matmultsymbolic;
7752   if (Asymbolic == Bsymbolic){
7753     if (!Bsymbolic) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"C=A*B not implemented for B of type %s",((PetscObject)B)->type_name);
7754     symbolic = Bsymbolic;
7755   } else { /* dispatch based on the type of A and B */
7756     char  symbolicname[256];
7757     ierr = PetscStrcpy(symbolicname,"MatMatMultSymbolic_");CHKERRQ(ierr);
7758     ierr = PetscStrcat(symbolicname,((PetscObject)A)->type_name);CHKERRQ(ierr);
7759     ierr = PetscStrcat(symbolicname,"_");CHKERRQ(ierr);
7760     ierr = PetscStrcat(symbolicname,((PetscObject)B)->type_name);CHKERRQ(ierr);
7761     ierr = PetscStrcat(symbolicname,"_C");CHKERRQ(ierr);
7762     ierr = PetscObjectQueryFunction((PetscObject)B,symbolicname,(void (**)(void))&symbolic);CHKERRQ(ierr);
7763     if (!symbolic) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_INCOMP,"MatMatMultSymbolic requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
7764   }
7765   ierr = PetscLogEventBegin(MAT_MatMultSymbolic,A,B,0,0);CHKERRQ(ierr);
7766   ierr = (*symbolic)(A,B,fill,C);CHKERRQ(ierr);
7767   ierr = PetscLogEventEnd(MAT_MatMultSymbolic,A,B,0,0);CHKERRQ(ierr);
7768   PetscFunctionReturn(0);
7769 }
7770 
7771 #undef __FUNCT__
7772 #define __FUNCT__ "MatMatMultNumeric"
7773 /*@
7774    MatMatMultNumeric - Performs the numeric matrix-matrix product.
7775    Call this routine after first calling MatMatMultSymbolic().
7776 
7777    Neighbor-wise Collective on Mat
7778 
7779    Input Parameters:
7780 +  A - the left matrix
7781 -  B - the right matrix
7782 
7783    Output Parameters:
7784 .  C - the product matrix, which was created by from MatMatMultSymbolic() or a call to MatMatMult().
7785 
7786    Notes:
7787    C must have been created with MatMatMultSymbolic().
7788 
7789    This routine is currently implemented for
7790     - pairs of AIJ matrices and classes which inherit from AIJ, C will be of type MATAIJ.
7791     - pairs of AIJ (A) and Dense (B) matrix, C will be of type Dense.
7792     - pairs of Dense (A) and AIJ (B) matrix, C will be of type Dense.
7793 
7794    Level: intermediate
7795 
7796 .seealso: MatMatMult(), MatMatMultSymbolic()
7797 @*/
7798 PetscErrorCode PETSCMAT_DLLEXPORT MatMatMultNumeric(Mat A,Mat B,Mat C)
7799 {
7800   PetscErrorCode ierr;
7801   PetscErrorCode (*Anumeric)(Mat,Mat,Mat);
7802   PetscErrorCode (*Bnumeric)(Mat,Mat,Mat);
7803   PetscErrorCode (*numeric)(Mat,Mat,Mat)=PETSC_NULL;
7804 
7805   PetscFunctionBegin;
7806   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7807   PetscValidType(A,1);
7808   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7809   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7810 
7811   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
7812   PetscValidType(B,2);
7813   ierr = MatPreallocated(B);CHKERRQ(ierr);
7814   if (!B->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7815   if (B->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7816 
7817   PetscValidHeaderSpecific(C,MAT_CLASSID,3);
7818   PetscValidType(C,3);
7819   ierr = MatPreallocated(C);CHKERRQ(ierr);
7820   if (!C->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7821   if (C->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7822 
7823   if (B->cmap->N!=C->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->cmap->N,C->cmap->N);
7824   if (B->rmap->N!=A->cmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
7825   if (A->rmap->N!=C->rmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",A->rmap->N,C->rmap->N);
7826   ierr = MatPreallocated(A);CHKERRQ(ierr);
7827 
7828   Anumeric = A->ops->matmultnumeric;
7829   Bnumeric = B->ops->matmultnumeric;
7830   if (Anumeric == Bnumeric){
7831     if (!Bnumeric) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"MatMatMultNumeric not supported for B of type %s",((PetscObject)B)->type_name);
7832     numeric = Bnumeric;
7833   } else {
7834     char  numericname[256];
7835     ierr = PetscStrcpy(numericname,"MatMatMultNumeric_");CHKERRQ(ierr);
7836     ierr = PetscStrcat(numericname,((PetscObject)A)->type_name);CHKERRQ(ierr);
7837     ierr = PetscStrcat(numericname,"_");CHKERRQ(ierr);
7838     ierr = PetscStrcat(numericname,((PetscObject)B)->type_name);CHKERRQ(ierr);
7839     ierr = PetscStrcat(numericname,"_C");CHKERRQ(ierr);
7840     ierr = PetscObjectQueryFunction((PetscObject)B,numericname,(void (**)(void))&numeric);CHKERRQ(ierr);
7841     if (!numeric)
7842       SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_INCOMP,"MatMatMultNumeric requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
7843   }
7844   ierr = PetscLogEventBegin(MAT_MatMultNumeric,A,B,0,0);CHKERRQ(ierr);
7845   ierr = (*numeric)(A,B,C);CHKERRQ(ierr);
7846   ierr = PetscLogEventEnd(MAT_MatMultNumeric,A,B,0,0);CHKERRQ(ierr);
7847   PetscFunctionReturn(0);
7848 }
7849 
7850 #undef __FUNCT__
7851 #define __FUNCT__ "MatMatMultTranspose"
7852 /*@
7853    MatMatMultTranspose - Performs Matrix-Matrix Multiplication C=A^T*B.
7854 
7855    Neighbor-wise Collective on Mat
7856 
7857    Input Parameters:
7858 +  A - the left matrix
7859 .  B - the right matrix
7860 .  scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7861 -  fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if not known
7862 
7863    Output Parameters:
7864 .  C - the product matrix
7865 
7866    Notes:
7867    C will be created if MAT_INITIAL_MATRIX and must be destroyed by the user with MatDestroy().
7868 
7869    MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call
7870 
7871   To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
7872    actually needed.
7873 
7874    This routine is currently only implemented for pairs of SeqAIJ matrices and pairs of SeqDense matrices and classes
7875    which inherit from SeqAIJ.  C will be of type MATSEQAIJ.
7876 
7877    Level: intermediate
7878 
7879 .seealso: MatMatMultTransposeSymbolic(), MatMatMultTransposeNumeric(), MatPtAP()
7880 @*/
7881 PetscErrorCode PETSCMAT_DLLEXPORT MatMatMultTranspose(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
7882 {
7883   PetscErrorCode ierr;
7884   PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
7885   PetscErrorCode (*fB)(Mat,Mat,MatReuse,PetscReal,Mat*);
7886 
7887   PetscFunctionBegin;
7888   PetscValidHeaderSpecific(A,MAT_CLASSID,1);
7889   PetscValidType(A,1);
7890   if (!A->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7891   if (A->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7892   PetscValidHeaderSpecific(B,MAT_CLASSID,2);
7893   PetscValidType(B,2);
7894   ierr = MatPreallocated(B);CHKERRQ(ierr);
7895   if (!B->assembled) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7896   if (B->factortype) SETERRQ(((PetscObject)A)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7897   PetscValidPointer(C,3);
7898   if (B->rmap->N!=A->rmap->N) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->rmap->N);
7899   if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
7900   if (fill < 1.0) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_ARG_SIZ,"Expected fill=%G must be > 1.0",fill);
7901   ierr = MatPreallocated(A);CHKERRQ(ierr);
7902 
7903   fA = A->ops->matmulttranspose;
7904   if (!fA) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"MatMatMultTranspose not supported for A of type %s",((PetscObject)A)->type_name);
7905   fB = B->ops->matmulttranspose;
7906   if (!fB) SETERRQ1(((PetscObject)A)->comm,PETSC_ERR_SUP,"MatMatMultTranspose not supported for B of type %s",((PetscObject)B)->type_name);
7907   if (fB!=fA) SETERRQ2(((PetscObject)A)->comm,PETSC_ERR_ARG_INCOMP,"MatMatMultTranspose requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
7908 
7909   ierr = PetscLogEventBegin(MAT_MatMultTranspose,A,B,0,0);CHKERRQ(ierr);
7910   ierr = (*A->ops->matmulttranspose)(A,B,scall,fill,C);CHKERRQ(ierr);
7911   ierr = PetscLogEventEnd(MAT_MatMultTranspose,A,B,0,0);CHKERRQ(ierr);
7912 
7913   PetscFunctionReturn(0);
7914 }
7915 
7916 #undef __FUNCT__
7917 #define __FUNCT__ "MatGetRedundantMatrix"
7918 /*@C
7919    MatGetRedundantMatrix - Create redundant matrices and put them into processors of subcommunicators.
7920 
7921    Collective on Mat
7922 
7923    Input Parameters:
7924 +  mat - the matrix
7925 .  nsubcomm - the number of subcommunicators (= number of redundant pareallel or sequential matrices)
7926 .  subcomm - MPI communicator split from the communicator where mat resides in
7927 .  mlocal_red - number of local rows of the redundant matrix
7928 -  reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7929 
7930    Output Parameter:
7931 .  matredundant - redundant matrix
7932 
7933    Notes:
7934    MAT_REUSE_MATRIX can only be used when the nonzero structure of the
7935    original matrix has not changed from that last call to MatGetRedundantMatrix().
7936 
7937    This routine creates the duplicated matrices in subcommunicators; you should NOT create them before
7938    calling it.
7939 
7940    Only MPIAIJ matrix is supported.
7941 
7942    Level: advanced
7943 
7944    Concepts: subcommunicator
7945    Concepts: duplicate matrix
7946 
7947 .seealso: MatDestroy()
7948 @*/
7949 PetscErrorCode PETSCMAT_DLLEXPORT MatGetRedundantMatrix(Mat mat,PetscInt nsubcomm,MPI_Comm subcomm,PetscInt mlocal_red,MatReuse reuse,Mat *matredundant)
7950 {
7951   PetscErrorCode ierr;
7952 
7953   PetscFunctionBegin;
7954   PetscValidHeaderSpecific(mat,MAT_CLASSID,1);
7955   if (nsubcomm && reuse == MAT_REUSE_MATRIX) {
7956     PetscValidPointer(*matredundant,6);
7957     PetscValidHeaderSpecific(*matredundant,MAT_CLASSID,6);
7958   }
7959   if (!mat->ops->getredundantmatrix) SETERRQ1(((PetscObject)mat)->comm,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7960   if (!mat->assembled) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7961   if (mat->factortype) SETERRQ(((PetscObject)mat)->comm,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7962   ierr = MatPreallocated(mat);CHKERRQ(ierr);
7963 
7964   ierr = PetscLogEventBegin(MAT_GetRedundantMatrix,mat,0,0,0);CHKERRQ(ierr);
7965   ierr = (*mat->ops->getredundantmatrix)(mat,nsubcomm,subcomm,mlocal_red,reuse,matredundant);CHKERRQ(ierr);
7966   ierr = PetscLogEventEnd(MAT_GetRedundantMatrix,mat,0,0,0);CHKERRQ(ierr);
7967   PetscFunctionReturn(0);
7968 }
7969