Actual source code: matrix.c
petsc-3.15.0 2021-03-30
1: /*
2: This is where the abstract matrix operations are defined
3: */
5: #include <petsc/private/matimpl.h>
6: #include <petsc/private/isimpl.h>
7: #include <petsc/private/vecimpl.h>
9: /* Logging support */
10: PetscClassId MAT_CLASSID;
11: PetscClassId MAT_COLORING_CLASSID;
12: PetscClassId MAT_FDCOLORING_CLASSID;
13: PetscClassId MAT_TRANSPOSECOLORING_CLASSID;
15: PetscLogEvent MAT_Mult, MAT_Mults, MAT_MultConstrained, MAT_MultAdd, MAT_MultTranspose;
16: PetscLogEvent MAT_MultTransposeConstrained, MAT_MultTransposeAdd, MAT_Solve, MAT_Solves, MAT_SolveAdd, MAT_SolveTranspose, MAT_MatSolve,MAT_MatTrSolve;
17: PetscLogEvent MAT_SolveTransposeAdd, MAT_SOR, MAT_ForwardSolve, MAT_BackwardSolve, MAT_LUFactor, MAT_LUFactorSymbolic;
18: PetscLogEvent MAT_LUFactorNumeric, MAT_CholeskyFactor, MAT_CholeskyFactorSymbolic, MAT_CholeskyFactorNumeric, MAT_ILUFactor;
19: PetscLogEvent MAT_ILUFactorSymbolic, MAT_ICCFactorSymbolic, MAT_Copy, MAT_Convert, MAT_Scale, MAT_AssemblyBegin;
20: PetscLogEvent MAT_QRFactorNumeric, MAT_QRFactorSymbolic, MAT_QRFactor;
21: PetscLogEvent MAT_AssemblyEnd, MAT_SetValues, MAT_GetValues, MAT_GetRow, MAT_GetRowIJ, MAT_CreateSubMats, MAT_GetOrdering, MAT_RedundantMat, MAT_GetSeqNonzeroStructure;
22: PetscLogEvent MAT_IncreaseOverlap, MAT_Partitioning, MAT_PartitioningND, MAT_Coarsen, MAT_ZeroEntries, MAT_Load, MAT_View, MAT_AXPY, MAT_FDColoringCreate;
23: PetscLogEvent MAT_FDColoringSetUp, MAT_FDColoringApply,MAT_Transpose,MAT_FDColoringFunction, MAT_CreateSubMat;
24: PetscLogEvent MAT_TransposeColoringCreate;
25: PetscLogEvent MAT_MatMult, MAT_MatMultSymbolic, MAT_MatMultNumeric;
26: PetscLogEvent MAT_PtAP, MAT_PtAPSymbolic, MAT_PtAPNumeric,MAT_RARt, MAT_RARtSymbolic, MAT_RARtNumeric;
27: PetscLogEvent MAT_MatTransposeMult, MAT_MatTransposeMultSymbolic, MAT_MatTransposeMultNumeric;
28: PetscLogEvent MAT_TransposeMatMult, MAT_TransposeMatMultSymbolic, MAT_TransposeMatMultNumeric;
29: PetscLogEvent MAT_MatMatMult, MAT_MatMatMultSymbolic, MAT_MatMatMultNumeric;
30: PetscLogEvent MAT_MultHermitianTranspose,MAT_MultHermitianTransposeAdd;
31: PetscLogEvent MAT_Getsymtranspose, MAT_Getsymtransreduced, MAT_GetBrowsOfAcols;
32: PetscLogEvent MAT_GetBrowsOfAocols, MAT_Getlocalmat, MAT_Getlocalmatcondensed, MAT_Seqstompi, MAT_Seqstompinum, MAT_Seqstompisym;
33: PetscLogEvent MAT_Applypapt, MAT_Applypapt_numeric, MAT_Applypapt_symbolic, MAT_GetSequentialNonzeroStructure;
34: PetscLogEvent MAT_GetMultiProcBlock;
35: PetscLogEvent MAT_CUSPARSECopyToGPU, MAT_CUSPARSECopyFromGPU, MAT_CUSPARSEGenerateTranspose, MAT_CUSPARSESolveAnalysis;
36: PetscLogEvent MAT_PreallCOO, MAT_SetVCOO;
37: PetscLogEvent MAT_SetValuesBatch;
38: PetscLogEvent MAT_ViennaCLCopyToGPU;
39: PetscLogEvent MAT_DenseCopyToGPU, MAT_DenseCopyFromGPU;
40: PetscLogEvent MAT_Merge,MAT_Residual,MAT_SetRandom;
41: PetscLogEvent MAT_FactorFactS,MAT_FactorInvS;
42: PetscLogEvent MATCOLORING_Apply,MATCOLORING_Comm,MATCOLORING_Local,MATCOLORING_ISCreate,MATCOLORING_SetUp,MATCOLORING_Weights;
44: const char *const MatFactorTypes[] = {"NONE","LU","CHOLESKY","ILU","ICC","ILUDT","QR","MatFactorType","MAT_FACTOR_",NULL};
46: /*@
47: MatSetRandom - Sets all components of a matrix to random numbers. For sparse matrices that have been preallocated but not been assembled it randomly selects appropriate locations,
48: for sparse matrices that already have locations it fills the locations with random numbers
50: Logically Collective on Mat
52: Input Parameters:
53: + x - the matrix
54: - rctx - the random number context, formed by PetscRandomCreate(), or NULL and
55: it will create one internally.
57: Output Parameter:
58: . x - the matrix
60: Example of Usage:
61: .vb
62: PetscRandomCreate(PETSC_COMM_WORLD,&rctx);
63: MatSetRandom(x,rctx);
64: PetscRandomDestroy(rctx);
65: .ve
67: Level: intermediate
70: .seealso: MatZeroEntries(), MatSetValues(), PetscRandomCreate(), PetscRandomDestroy()
71: @*/
72: PetscErrorCode MatSetRandom(Mat x,PetscRandom rctx)
73: {
75: PetscRandom randObj = NULL;
82: if (!x->ops->setrandom) SETERRQ1(PetscObjectComm((PetscObject)x),PETSC_ERR_SUP,"Mat type %s",((PetscObject)x)->type_name);
84: if (!rctx) {
85: MPI_Comm comm;
86: PetscObjectGetComm((PetscObject)x,&comm);
87: PetscRandomCreate(comm,&randObj);
88: PetscRandomSetFromOptions(randObj);
89: rctx = randObj;
90: }
92: PetscLogEventBegin(MAT_SetRandom,x,rctx,0,0);
93: (*x->ops->setrandom)(x,rctx);
94: PetscLogEventEnd(MAT_SetRandom,x,rctx,0,0);
96: MatAssemblyBegin(x, MAT_FINAL_ASSEMBLY);
97: MatAssemblyEnd(x, MAT_FINAL_ASSEMBLY);
98: PetscRandomDestroy(&randObj);
99: return(0);
100: }
102: /*@
103: MatFactorGetErrorZeroPivot - returns the pivot value that was determined to be zero and the row it occurred in
105: Logically Collective on Mat
107: Input Parameters:
108: . mat - the factored matrix
110: Output Parameter:
111: + pivot - the pivot value computed
112: - row - the row that the zero pivot occurred. Note that this row must be interpreted carefully due to row reorderings and which processes
113: the share the matrix
115: Level: advanced
117: Notes:
118: This routine does not work for factorizations done with external packages.
120: This routine should only be called if MatGetFactorError() returns a value of MAT_FACTOR_NUMERIC_ZEROPIVOT
122: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
124: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatLUFactorSymbolic(), MatCholeskyFactorSymbolic(), MatFactorClearError(), MatFactorGetErrorZeroPivot()
125: @*/
126: PetscErrorCode MatFactorGetErrorZeroPivot(Mat mat,PetscReal *pivot,PetscInt *row)
127: {
130: *pivot = mat->factorerror_zeropivot_value;
131: *row = mat->factorerror_zeropivot_row;
132: return(0);
133: }
135: /*@
136: MatFactorGetError - gets the error code from a factorization
138: Logically Collective on Mat
140: Input Parameters:
141: . mat - the factored matrix
143: Output Parameter:
144: . err - the error code
146: Level: advanced
148: Notes:
149: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
151: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatLUFactorSymbolic(), MatCholeskyFactorSymbolic(), MatFactorClearError(), MatFactorGetErrorZeroPivot()
152: @*/
153: PetscErrorCode MatFactorGetError(Mat mat,MatFactorError *err)
154: {
157: *err = mat->factorerrortype;
158: return(0);
159: }
161: /*@
162: MatFactorClearError - clears the error code in a factorization
164: Logically Collective on Mat
166: Input Parameter:
167: . mat - the factored matrix
169: Level: developer
171: Notes:
172: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
174: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatLUFactorSymbolic(), MatCholeskyFactorSymbolic(), MatFactorGetError(), MatFactorGetErrorZeroPivot()
175: @*/
176: PetscErrorCode MatFactorClearError(Mat mat)
177: {
180: mat->factorerrortype = MAT_FACTOR_NOERROR;
181: mat->factorerror_zeropivot_value = 0.0;
182: mat->factorerror_zeropivot_row = 0;
183: return(0);
184: }
186: PETSC_INTERN PetscErrorCode MatFindNonzeroRowsOrCols_Basic(Mat mat,PetscBool cols,PetscReal tol,IS *nonzero)
187: {
188: PetscErrorCode ierr;
189: Vec r,l;
190: const PetscScalar *al;
191: PetscInt i,nz,gnz,N,n;
194: MatCreateVecs(mat,&r,&l);
195: if (!cols) { /* nonzero rows */
196: MatGetSize(mat,&N,NULL);
197: MatGetLocalSize(mat,&n,NULL);
198: VecSet(l,0.0);
199: VecSetRandom(r,NULL);
200: MatMult(mat,r,l);
201: VecGetArrayRead(l,&al);
202: } else { /* nonzero columns */
203: MatGetSize(mat,NULL,&N);
204: MatGetLocalSize(mat,NULL,&n);
205: VecSet(r,0.0);
206: VecSetRandom(l,NULL);
207: MatMultTranspose(mat,l,r);
208: VecGetArrayRead(r,&al);
209: }
210: if (tol <= 0.0) { for (i=0,nz=0;i<n;i++) if (al[i] != 0.0) nz++; }
211: else { for (i=0,nz=0;i<n;i++) if (PetscAbsScalar(al[i]) > tol) nz++; }
212: MPIU_Allreduce(&nz,&gnz,1,MPIU_INT,MPI_SUM,PetscObjectComm((PetscObject)mat));
213: if (gnz != N) {
214: PetscInt *nzr;
215: PetscMalloc1(nz,&nzr);
216: if (nz) {
217: if (tol < 0) { for (i=0,nz=0;i<n;i++) if (al[i] != 0.0) nzr[nz++] = i; }
218: else { for (i=0,nz=0;i<n;i++) if (PetscAbsScalar(al[i]) > tol) nzr[nz++] = i; }
219: }
220: ISCreateGeneral(PetscObjectComm((PetscObject)mat),nz,nzr,PETSC_OWN_POINTER,nonzero);
221: } else *nonzero = NULL;
222: if (!cols) { /* nonzero rows */
223: VecRestoreArrayRead(l,&al);
224: } else {
225: VecRestoreArrayRead(r,&al);
226: }
227: VecDestroy(&l);
228: VecDestroy(&r);
229: return(0);
230: }
232: /*@
233: MatFindNonzeroRows - Locate all rows that are not completely zero in the matrix
235: Input Parameter:
236: . A - the matrix
238: Output Parameter:
239: . keptrows - the rows that are not completely zero
241: Notes:
242: keptrows is set to NULL if all rows are nonzero.
244: Level: intermediate
246: @*/
247: PetscErrorCode MatFindNonzeroRows(Mat mat,IS *keptrows)
248: {
255: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
256: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
257: if (!mat->ops->findnonzerorows) {
258: MatFindNonzeroRowsOrCols_Basic(mat,PETSC_FALSE,0.0,keptrows);
259: } else {
260: (*mat->ops->findnonzerorows)(mat,keptrows);
261: }
262: return(0);
263: }
265: /*@
266: MatFindZeroRows - Locate all rows that are completely zero in the matrix
268: Input Parameter:
269: . A - the matrix
271: Output Parameter:
272: . zerorows - the rows that are completely zero
274: Notes:
275: zerorows is set to NULL if no rows are zero.
277: Level: intermediate
279: @*/
280: PetscErrorCode MatFindZeroRows(Mat mat,IS *zerorows)
281: {
283: IS keptrows;
284: PetscInt m, n;
289: MatFindNonzeroRows(mat, &keptrows);
290: /* MatFindNonzeroRows sets keptrows to NULL if there are no zero rows.
291: In keeping with this convention, we set zerorows to NULL if there are no zero
292: rows. */
293: if (keptrows == NULL) {
294: *zerorows = NULL;
295: } else {
296: MatGetOwnershipRange(mat,&m,&n);
297: ISComplement(keptrows,m,n,zerorows);
298: ISDestroy(&keptrows);
299: }
300: return(0);
301: }
303: /*@
304: MatGetDiagonalBlock - Returns the part of the matrix associated with the on-process coupling
306: Not Collective
308: Input Parameters:
309: . A - the matrix
311: Output Parameters:
312: . a - the diagonal part (which is a SEQUENTIAL matrix)
314: Notes:
315: see the manual page for MatCreateAIJ() for more information on the "diagonal part" of the matrix.
316: Use caution, as the reference count on the returned matrix is not incremented and it is used as
317: part of the containing MPI Mat's normal operation.
319: Level: advanced
321: @*/
322: PetscErrorCode MatGetDiagonalBlock(Mat A,Mat *a)
323: {
330: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
331: if (!A->ops->getdiagonalblock) {
332: PetscMPIInt size;
333: MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);
334: if (size == 1) {
335: *a = A;
336: return(0);
337: } else SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Not coded for matrix type %s",((PetscObject)A)->type_name);
338: }
339: (*A->ops->getdiagonalblock)(A,a);
340: return(0);
341: }
343: /*@
344: MatGetTrace - Gets the trace of a matrix. The sum of the diagonal entries.
346: Collective on Mat
348: Input Parameters:
349: . mat - the matrix
351: Output Parameter:
352: . trace - the sum of the diagonal entries
354: Level: advanced
356: @*/
357: PetscErrorCode MatGetTrace(Mat mat,PetscScalar *trace)
358: {
360: Vec diag;
363: MatCreateVecs(mat,&diag,NULL);
364: MatGetDiagonal(mat,diag);
365: VecSum(diag,trace);
366: VecDestroy(&diag);
367: return(0);
368: }
370: /*@
371: MatRealPart - Zeros out the imaginary part of the matrix
373: Logically Collective on Mat
375: Input Parameters:
376: . mat - the matrix
378: Level: advanced
381: .seealso: MatImaginaryPart()
382: @*/
383: PetscErrorCode MatRealPart(Mat mat)
384: {
390: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
391: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
392: if (!mat->ops->realpart) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
393: MatCheckPreallocated(mat,1);
394: (*mat->ops->realpart)(mat);
395: return(0);
396: }
398: /*@C
399: MatGetGhosts - Get the global index of all ghost nodes defined by the sparse matrix
401: Collective on Mat
403: Input Parameter:
404: . mat - the matrix
406: Output Parameters:
407: + nghosts - number of ghosts (note for BAIJ matrices there is one ghost for each block)
408: - ghosts - the global indices of the ghost points
410: Notes:
411: the nghosts and ghosts are suitable to pass into VecCreateGhost()
413: Level: advanced
415: @*/
416: PetscErrorCode MatGetGhosts(Mat mat,PetscInt *nghosts,const PetscInt *ghosts[])
417: {
423: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
424: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
425: if (!mat->ops->getghosts) {
426: if (nghosts) *nghosts = 0;
427: if (ghosts) *ghosts = NULL;
428: } else {
429: (*mat->ops->getghosts)(mat,nghosts,ghosts);
430: }
431: return(0);
432: }
435: /*@
436: MatImaginaryPart - Moves the imaginary part of the matrix to the real part and zeros the imaginary part
438: Logically Collective on Mat
440: Input Parameters:
441: . mat - the matrix
443: Level: advanced
446: .seealso: MatRealPart()
447: @*/
448: PetscErrorCode MatImaginaryPart(Mat mat)
449: {
455: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
456: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
457: if (!mat->ops->imaginarypart) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
458: MatCheckPreallocated(mat,1);
459: (*mat->ops->imaginarypart)(mat);
460: return(0);
461: }
463: /*@
464: MatMissingDiagonal - Determine if sparse matrix is missing a diagonal entry (or block entry for BAIJ matrices)
466: Not Collective
468: Input Parameter:
469: . mat - the matrix
471: Output Parameters:
472: + missing - is any diagonal missing
473: - dd - first diagonal entry that is missing (optional) on this process
475: Level: advanced
478: .seealso: MatRealPart()
479: @*/
480: PetscErrorCode MatMissingDiagonal(Mat mat,PetscBool *missing,PetscInt *dd)
481: {
488: if (!mat->assembled) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix %s",((PetscObject)mat)->type_name);
489: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
490: if (!mat->ops->missingdiagonal) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
491: (*mat->ops->missingdiagonal)(mat,missing,dd);
492: return(0);
493: }
495: /*@C
496: MatGetRow - Gets a row of a matrix. You MUST call MatRestoreRow()
497: for each row that you get to ensure that your application does
498: not bleed memory.
500: Not Collective
502: Input Parameters:
503: + mat - the matrix
504: - row - the row to get
506: Output Parameters:
507: + ncols - if not NULL, the number of nonzeros in the row
508: . cols - if not NULL, the column numbers
509: - vals - if not NULL, the values
511: Notes:
512: This routine is provided for people who need to have direct access
513: to the structure of a matrix. We hope that we provide enough
514: high-level matrix routines that few users will need it.
516: MatGetRow() always returns 0-based column indices, regardless of
517: whether the internal representation is 0-based (default) or 1-based.
519: For better efficiency, set cols and/or vals to NULL if you do
520: not wish to extract these quantities.
522: The user can only examine the values extracted with MatGetRow();
523: the values cannot be altered. To change the matrix entries, one
524: must use MatSetValues().
526: You can only have one call to MatGetRow() outstanding for a particular
527: matrix at a time, per processor. MatGetRow() can only obtain rows
528: associated with the given processor, it cannot get rows from the
529: other processors; for that we suggest using MatCreateSubMatrices(), then
530: MatGetRow() on the submatrix. The row index passed to MatGetRow()
531: is in the global number of rows.
533: Fortran Notes:
534: The calling sequence from Fortran is
535: .vb
536: MatGetRow(matrix,row,ncols,cols,values,ierr)
537: Mat matrix (input)
538: integer row (input)
539: integer ncols (output)
540: integer cols(maxcols) (output)
541: double precision (or double complex) values(maxcols) output
542: .ve
543: where maxcols >= maximum nonzeros in any row of the matrix.
546: Caution:
547: Do not try to change the contents of the output arrays (cols and vals).
548: In some cases, this may corrupt the matrix.
550: Level: advanced
552: .seealso: MatRestoreRow(), MatSetValues(), MatGetValues(), MatCreateSubMatrices(), MatGetDiagonal()
553: @*/
554: PetscErrorCode MatGetRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
555: {
557: PetscInt incols;
562: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
563: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
564: if (!mat->ops->getrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
565: MatCheckPreallocated(mat,1);
566: if (row < mat->rmap->rstart || row >= mat->rmap->rend) SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Only for local rows, %D not in [%D,%D)",row,mat->rmap->rstart,mat->rmap->rend);
567: PetscLogEventBegin(MAT_GetRow,mat,0,0,0);
568: (*mat->ops->getrow)(mat,row,&incols,(PetscInt**)cols,(PetscScalar**)vals);
569: if (ncols) *ncols = incols;
570: PetscLogEventEnd(MAT_GetRow,mat,0,0,0);
571: return(0);
572: }
574: /*@
575: MatConjugate - replaces the matrix values with their complex conjugates
577: Logically Collective on Mat
579: Input Parameters:
580: . mat - the matrix
582: Level: advanced
584: .seealso: VecConjugate()
585: @*/
586: PetscErrorCode MatConjugate(Mat mat)
587: {
588: #if defined(PETSC_USE_COMPLEX)
593: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
594: if (!mat->ops->conjugate) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Not provided for matrix type %s, send email to petsc-maint@mcs.anl.gov",((PetscObject)mat)->type_name);
595: (*mat->ops->conjugate)(mat);
596: #else
598: #endif
599: return(0);
600: }
602: /*@C
603: MatRestoreRow - Frees any temporary space allocated by MatGetRow().
605: Not Collective
607: Input Parameters:
608: + mat - the matrix
609: . row - the row to get
610: . ncols, cols - the number of nonzeros and their columns
611: - vals - if nonzero the column values
613: Notes:
614: This routine should be called after you have finished examining the entries.
616: This routine zeros out ncols, cols, and vals. This is to prevent accidental
617: us of the array after it has been restored. If you pass NULL, it will
618: not zero the pointers. Use of cols or vals after MatRestoreRow is invalid.
620: Fortran Notes:
621: The calling sequence from Fortran is
622: .vb
623: MatRestoreRow(matrix,row,ncols,cols,values,ierr)
624: Mat matrix (input)
625: integer row (input)
626: integer ncols (output)
627: integer cols(maxcols) (output)
628: double precision (or double complex) values(maxcols) output
629: .ve
630: Where maxcols >= maximum nonzeros in any row of the matrix.
632: In Fortran MatRestoreRow() MUST be called after MatGetRow()
633: before another call to MatGetRow() can be made.
635: Level: advanced
637: .seealso: MatGetRow()
638: @*/
639: PetscErrorCode MatRestoreRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
640: {
646: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
647: if (!mat->ops->restorerow) return(0);
648: (*mat->ops->restorerow)(mat,row,ncols,(PetscInt **)cols,(PetscScalar **)vals);
649: if (ncols) *ncols = 0;
650: if (cols) *cols = NULL;
651: if (vals) *vals = NULL;
652: return(0);
653: }
655: /*@
656: MatGetRowUpperTriangular - Sets a flag to enable calls to MatGetRow() for matrix in MATSBAIJ format.
657: You should call MatRestoreRowUpperTriangular() after calling MatGetRow/MatRestoreRow() to disable the flag.
659: Not Collective
661: Input Parameters:
662: . mat - the matrix
664: Notes:
665: The flag is to ensure that users are aware of MatGetRow() only provides the upper triangular part of the row for the matrices in MATSBAIJ format.
667: Level: advanced
669: .seealso: MatRestoreRowUpperTriangular()
670: @*/
671: PetscErrorCode MatGetRowUpperTriangular(Mat mat)
672: {
678: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
679: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
680: MatCheckPreallocated(mat,1);
681: if (!mat->ops->getrowuppertriangular) return(0);
682: (*mat->ops->getrowuppertriangular)(mat);
683: return(0);
684: }
686: /*@
687: MatRestoreRowUpperTriangular - Disable calls to MatGetRow() for matrix in MATSBAIJ format.
689: Not Collective
691: Input Parameters:
692: . mat - the matrix
694: Notes:
695: This routine should be called after you have finished MatGetRow/MatRestoreRow().
698: Level: advanced
700: .seealso: MatGetRowUpperTriangular()
701: @*/
702: PetscErrorCode MatRestoreRowUpperTriangular(Mat mat)
703: {
709: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
710: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
711: MatCheckPreallocated(mat,1);
712: if (!mat->ops->restorerowuppertriangular) return(0);
713: (*mat->ops->restorerowuppertriangular)(mat);
714: return(0);
715: }
717: /*@C
718: MatSetOptionsPrefix - Sets the prefix used for searching for all
719: Mat options in the database.
721: Logically Collective on Mat
723: Input Parameter:
724: + A - the Mat context
725: - prefix - the prefix to prepend to all option names
727: Notes:
728: A hyphen (-) must NOT be given at the beginning of the prefix name.
729: The first character of all runtime options is AUTOMATICALLY the hyphen.
731: Level: advanced
733: .seealso: MatSetFromOptions()
734: @*/
735: PetscErrorCode MatSetOptionsPrefix(Mat A,const char prefix[])
736: {
741: PetscObjectSetOptionsPrefix((PetscObject)A,prefix);
742: return(0);
743: }
745: /*@C
746: MatAppendOptionsPrefix - Appends to the prefix used for searching for all
747: Mat options in the database.
749: Logically Collective on Mat
751: Input Parameters:
752: + A - the Mat context
753: - prefix - the prefix to prepend to all option names
755: Notes:
756: A hyphen (-) must NOT be given at the beginning of the prefix name.
757: The first character of all runtime options is AUTOMATICALLY the hyphen.
759: Level: advanced
761: .seealso: MatGetOptionsPrefix()
762: @*/
763: PetscErrorCode MatAppendOptionsPrefix(Mat A,const char prefix[])
764: {
769: PetscObjectAppendOptionsPrefix((PetscObject)A,prefix);
770: return(0);
771: }
773: /*@C
774: MatGetOptionsPrefix - Gets the prefix used for searching for all
775: Mat options in the database.
777: Not Collective
779: Input Parameter:
780: . A - the Mat context
782: Output Parameter:
783: . prefix - pointer to the prefix string used
785: Notes:
786: On the fortran side, the user should pass in a string 'prefix' of
787: sufficient length to hold the prefix.
789: Level: advanced
791: .seealso: MatAppendOptionsPrefix()
792: @*/
793: PetscErrorCode MatGetOptionsPrefix(Mat A,const char *prefix[])
794: {
799: PetscObjectGetOptionsPrefix((PetscObject)A,prefix);
800: return(0);
801: }
803: /*@
804: MatResetPreallocation - Reset mat to use the original nonzero pattern provided by users.
806: Collective on Mat
808: Input Parameters:
809: . A - the Mat context
811: Notes:
812: The allocated memory will be shrunk after calling MatAssembly with MAT_FINAL_ASSEMBLY. Users can reset the preallocation to access the original memory.
813: Currently support MPIAIJ and SEQAIJ.
815: Level: beginner
817: .seealso: MatSeqAIJSetPreallocation(), MatMPIAIJSetPreallocation(), MatXAIJSetPreallocation()
818: @*/
819: PetscErrorCode MatResetPreallocation(Mat A)
820: {
826: PetscUseMethod(A,"MatResetPreallocation_C",(Mat),(A));
827: return(0);
828: }
831: /*@
832: MatSetUp - Sets up the internal matrix data structures for later use.
834: Collective on Mat
836: Input Parameters:
837: . A - the Mat context
839: Notes:
840: If the user has not set preallocation for this matrix then a default preallocation that is likely to be inefficient is used.
842: If a suitable preallocation routine is used, this function does not need to be called.
844: See the Performance chapter of the PETSc users manual for how to preallocate matrices
846: Level: beginner
848: .seealso: MatCreate(), MatDestroy()
849: @*/
850: PetscErrorCode MatSetUp(Mat A)
851: {
852: PetscMPIInt size;
857: if (!((PetscObject)A)->type_name) {
858: MPI_Comm_size(PetscObjectComm((PetscObject)A), &size);
859: if (size == 1) {
860: MatSetType(A, MATSEQAIJ);
861: } else {
862: MatSetType(A, MATMPIAIJ);
863: }
864: }
865: if (!A->preallocated && A->ops->setup) {
866: PetscInfo(A,"Warning not preallocating matrix storage\n");
867: (*A->ops->setup)(A);
868: }
869: PetscLayoutSetUp(A->rmap);
870: PetscLayoutSetUp(A->cmap);
871: A->preallocated = PETSC_TRUE;
872: return(0);
873: }
875: #if defined(PETSC_HAVE_SAWS)
876: #include <petscviewersaws.h>
877: #endif
879: /*@C
880: MatViewFromOptions - View from Options
882: Collective on Mat
884: Input Parameters:
885: + A - the Mat context
886: . obj - Optional object
887: - name - command line option
889: Level: intermediate
890: .seealso: Mat, MatView, PetscObjectViewFromOptions(), MatCreate()
891: @*/
892: PetscErrorCode MatViewFromOptions(Mat A,PetscObject obj,const char name[])
893: {
898: PetscObjectViewFromOptions((PetscObject)A,obj,name);
899: return(0);
900: }
902: /*@C
903: MatView - Visualizes a matrix object.
905: Collective on Mat
907: Input Parameters:
908: + mat - the matrix
909: - viewer - visualization context
911: Notes:
912: The available visualization contexts include
913: + PETSC_VIEWER_STDOUT_SELF - for sequential matrices
914: . PETSC_VIEWER_STDOUT_WORLD - for parallel matrices created on PETSC_COMM_WORLD
915: . PETSC_VIEWER_STDOUT_(comm) - for matrices created on MPI communicator comm
916: - PETSC_VIEWER_DRAW_WORLD - graphical display of nonzero structure
918: The user can open alternative visualization contexts with
919: + PetscViewerASCIIOpen() - Outputs matrix to a specified file
920: . PetscViewerBinaryOpen() - Outputs matrix in binary to a
921: specified file; corresponding input uses MatLoad()
922: . PetscViewerDrawOpen() - Outputs nonzero matrix structure to
923: an X window display
924: - PetscViewerSocketOpen() - Outputs matrix to Socket viewer.
925: Currently only the sequential dense and AIJ
926: matrix types support the Socket viewer.
928: The user can call PetscViewerPushFormat() to specify the output
929: format of ASCII printed objects (when using PETSC_VIEWER_STDOUT_SELF,
930: PETSC_VIEWER_STDOUT_WORLD and PetscViewerASCIIOpen). Available formats include
931: + PETSC_VIEWER_DEFAULT - default, prints matrix contents
932: . PETSC_VIEWER_ASCII_MATLAB - prints matrix contents in Matlab format
933: . PETSC_VIEWER_ASCII_DENSE - prints entire matrix including zeros
934: . PETSC_VIEWER_ASCII_COMMON - prints matrix contents, using a sparse
935: format common among all matrix types
936: . PETSC_VIEWER_ASCII_IMPL - prints matrix contents, using an implementation-specific
937: format (which is in many cases the same as the default)
938: . PETSC_VIEWER_ASCII_INFO - prints basic information about the matrix
939: size and structure (not the matrix entries)
940: - PETSC_VIEWER_ASCII_INFO_DETAIL - prints more detailed information about
941: the matrix structure
943: Options Database Keys:
944: + -mat_view ::ascii_info - Prints info on matrix at conclusion of MatAssemblyEnd()
945: . -mat_view ::ascii_info_detail - Prints more detailed info
946: . -mat_view - Prints matrix in ASCII format
947: . -mat_view ::ascii_matlab - Prints matrix in Matlab format
948: . -mat_view draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
949: . -display <name> - Sets display name (default is host)
950: . -draw_pause <sec> - Sets number of seconds to pause after display
951: . -mat_view socket - Sends matrix to socket, can be accessed from Matlab (see Users-Manual: ch_matlab for details)
952: . -viewer_socket_machine <machine> -
953: . -viewer_socket_port <port> -
954: . -mat_view binary - save matrix to file in binary format
955: - -viewer_binary_filename <name> -
956: Level: beginner
958: Notes:
959: The ASCII viewers are only recommended for small matrices on at most a moderate number of processes,
960: the program will seemingly hang and take hours for larger matrices, for larger matrices one should use the binary format.
962: In the debugger you can do "call MatView(mat,0)" to display the matrix. (The same holds for any PETSc object viewer).
964: See the manual page for MatLoad() for the exact format of the binary file when the binary
965: viewer is used.
967: See share/petsc/matlab/PetscBinaryRead.m for a Matlab code that can read in the binary file when the binary
968: viewer is used and lib/petsc/bin/PetscBinaryIO.py for loading them into Python.
970: One can use '-mat_view draw -draw_pause -1' to pause the graphical display of matrix nonzero structure,
971: and then use the following mouse functions.
972: + left mouse: zoom in
973: . middle mouse: zoom out
974: - right mouse: continue with the simulation
976: .seealso: PetscViewerPushFormat(), PetscViewerASCIIOpen(), PetscViewerDrawOpen(),
977: PetscViewerSocketOpen(), PetscViewerBinaryOpen(), MatLoad()
978: @*/
979: PetscErrorCode MatView(Mat mat,PetscViewer viewer)
980: {
981: PetscErrorCode ierr;
982: PetscInt rows,cols,rbs,cbs;
983: PetscBool isascii,isstring,issaws;
984: PetscViewerFormat format;
985: PetscMPIInt size;
990: if (!viewer) {PetscViewerASCIIGetStdout(PetscObjectComm((PetscObject)mat),&viewer);}
993: MatCheckPreallocated(mat,1);
995: PetscViewerGetFormat(viewer,&format);
996: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
997: if (size == 1 && format == PETSC_VIEWER_LOAD_BALANCE) return(0);
999: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSTRING,&isstring);
1000: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&isascii);
1001: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSAWS,&issaws);
1002: if ((!isascii || (format != PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL)) && mat->factortype) {
1003: SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"No viewers for factored matrix except ASCII info or info_detail");
1004: }
1006: PetscLogEventBegin(MAT_View,mat,viewer,0,0);
1007: if (isascii) {
1008: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ORDER,"Must call MatAssemblyBegin/End() before viewing matrix");
1009: PetscObjectPrintClassNamePrefixType((PetscObject)mat,viewer);
1010: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
1011: MatNullSpace nullsp,transnullsp;
1013: PetscViewerASCIIPushTab(viewer);
1014: MatGetSize(mat,&rows,&cols);
1015: MatGetBlockSizes(mat,&rbs,&cbs);
1016: if (rbs != 1 || cbs != 1) {
1017: if (rbs != cbs) {PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D, rbs=%D, cbs=%D\n",rows,cols,rbs,cbs);}
1018: else {PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D, bs=%D\n",rows,cols,rbs);}
1019: } else {
1020: PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D\n",rows,cols);
1021: }
1022: if (mat->factortype) {
1023: MatSolverType solver;
1024: MatFactorGetSolverType(mat,&solver);
1025: PetscViewerASCIIPrintf(viewer,"package used to perform factorization: %s\n",solver);
1026: }
1027: if (mat->ops->getinfo) {
1028: MatInfo info;
1029: MatGetInfo(mat,MAT_GLOBAL_SUM,&info);
1030: PetscViewerASCIIPrintf(viewer,"total: nonzeros=%.f, allocated nonzeros=%.f\n",info.nz_used,info.nz_allocated);
1031: if (!mat->factortype) {
1032: PetscViewerASCIIPrintf(viewer,"total number of mallocs used during MatSetValues calls=%D\n",(PetscInt)info.mallocs);
1033: }
1034: }
1035: MatGetNullSpace(mat,&nullsp);
1036: MatGetTransposeNullSpace(mat,&transnullsp);
1037: if (nullsp) {PetscViewerASCIIPrintf(viewer," has attached null space\n");}
1038: if (transnullsp && transnullsp != nullsp) {PetscViewerASCIIPrintf(viewer," has attached transposed null space\n");}
1039: MatGetNearNullSpace(mat,&nullsp);
1040: if (nullsp) {PetscViewerASCIIPrintf(viewer," has attached near null space\n");}
1041: PetscViewerASCIIPushTab(viewer);
1042: MatProductView(mat,viewer);
1043: PetscViewerASCIIPopTab(viewer);
1044: }
1045: } else if (issaws) {
1046: #if defined(PETSC_HAVE_SAWS)
1047: PetscMPIInt rank;
1049: PetscObjectName((PetscObject)mat);
1050: MPI_Comm_rank(PETSC_COMM_WORLD,&rank);
1051: if (!((PetscObject)mat)->amsmem && !rank) {
1052: PetscObjectViewSAWs((PetscObject)mat,viewer);
1053: }
1054: #endif
1055: } else if (isstring) {
1056: const char *type;
1057: MatGetType(mat,&type);
1058: PetscViewerStringSPrintf(viewer," MatType: %-7.7s",type);
1059: if (mat->ops->view) {(*mat->ops->view)(mat,viewer);}
1060: }
1061: if ((format == PETSC_VIEWER_NATIVE || format == PETSC_VIEWER_LOAD_BALANCE) && mat->ops->viewnative) {
1062: PetscViewerASCIIPushTab(viewer);
1063: (*mat->ops->viewnative)(mat,viewer);
1064: PetscViewerASCIIPopTab(viewer);
1065: } else if (mat->ops->view) {
1066: PetscViewerASCIIPushTab(viewer);
1067: (*mat->ops->view)(mat,viewer);
1068: PetscViewerASCIIPopTab(viewer);
1069: }
1070: if (isascii) {
1071: PetscViewerGetFormat(viewer,&format);
1072: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
1073: PetscViewerASCIIPopTab(viewer);
1074: }
1075: }
1076: PetscLogEventEnd(MAT_View,mat,viewer,0,0);
1077: return(0);
1078: }
1080: #if defined(PETSC_USE_DEBUG)
1081: #include <../src/sys/totalview/tv_data_display.h>
1082: PETSC_UNUSED static int TV_display_type(const struct _p_Mat *mat)
1083: {
1084: TV_add_row("Local rows", "int", &mat->rmap->n);
1085: TV_add_row("Local columns", "int", &mat->cmap->n);
1086: TV_add_row("Global rows", "int", &mat->rmap->N);
1087: TV_add_row("Global columns", "int", &mat->cmap->N);
1088: TV_add_row("Typename", TV_ascii_string_type, ((PetscObject)mat)->type_name);
1089: return TV_format_OK;
1090: }
1091: #endif
1093: /*@C
1094: MatLoad - Loads a matrix that has been stored in binary/HDF5 format
1095: with MatView(). The matrix format is determined from the options database.
1096: Generates a parallel MPI matrix if the communicator has more than one
1097: processor. The default matrix type is AIJ.
1099: Collective on PetscViewer
1101: Input Parameters:
1102: + mat - the newly loaded matrix, this needs to have been created with MatCreate()
1103: or some related function before a call to MatLoad()
1104: - viewer - binary/HDF5 file viewer
1106: Options Database Keys:
1107: Used with block matrix formats (MATSEQBAIJ, ...) to specify
1108: block size
1109: . -matload_block_size <bs>
1111: Level: beginner
1113: Notes:
1114: If the Mat type has not yet been given then MATAIJ is used, call MatSetFromOptions() on the
1115: Mat before calling this routine if you wish to set it from the options database.
1117: MatLoad() automatically loads into the options database any options
1118: given in the file filename.info where filename is the name of the file
1119: that was passed to the PetscViewerBinaryOpen(). The options in the info
1120: file will be ignored if you use the -viewer_binary_skip_info option.
1122: If the type or size of mat is not set before a call to MatLoad, PETSc
1123: sets the default matrix type AIJ and sets the local and global sizes.
1124: If type and/or size is already set, then the same are used.
1126: In parallel, each processor can load a subset of rows (or the
1127: entire matrix). This routine is especially useful when a large
1128: matrix is stored on disk and only part of it is desired on each
1129: processor. For example, a parallel solver may access only some of
1130: the rows from each processor. The algorithm used here reads
1131: relatively small blocks of data rather than reading the entire
1132: matrix and then subsetting it.
1134: Viewer's PetscViewerType must be either PETSCVIEWERBINARY or PETSCVIEWERHDF5.
1135: Such viewer can be created using PetscViewerBinaryOpen()/PetscViewerHDF5Open(),
1136: or the sequence like
1137: $ PetscViewer v;
1138: $ PetscViewerCreate(PETSC_COMM_WORLD,&v);
1139: $ PetscViewerSetType(v,PETSCVIEWERBINARY);
1140: $ PetscViewerSetFromOptions(v);
1141: $ PetscViewerFileSetMode(v,FILE_MODE_READ);
1142: $ PetscViewerFileSetName(v,"datafile");
1143: The optional PetscViewerSetFromOptions() call allows to override PetscViewerSetType() using option
1144: $ -viewer_type {binary,hdf5}
1146: See the example src/ksp/ksp/tutorials/ex27.c with the first approach,
1147: and src/mat/tutorials/ex10.c with the second approach.
1149: Notes about the PETSc binary format:
1150: In case of PETSCVIEWERBINARY, a native PETSc binary format is used. Each of the blocks
1151: is read onto rank 0 and then shipped to its destination rank, one after another.
1152: Multiple objects, both matrices and vectors, can be stored within the same file.
1153: Their PetscObject name is ignored; they are loaded in the order of their storage.
1155: Most users should not need to know the details of the binary storage
1156: format, since MatLoad() and MatView() completely hide these details.
1157: But for anyone who's interested, the standard binary matrix storage
1158: format is
1160: $ PetscInt MAT_FILE_CLASSID
1161: $ PetscInt number of rows
1162: $ PetscInt number of columns
1163: $ PetscInt total number of nonzeros
1164: $ PetscInt *number nonzeros in each row
1165: $ PetscInt *column indices of all nonzeros (starting index is zero)
1166: $ PetscScalar *values of all nonzeros
1168: PETSc automatically does the byte swapping for
1169: machines that store the bytes reversed, e.g. DEC alpha, freebsd,
1170: linux, Windows and the paragon; thus if you write your own binary
1171: read/write routines you have to swap the bytes; see PetscBinaryRead()
1172: and PetscBinaryWrite() to see how this may be done.
1174: Notes about the HDF5 (MATLAB MAT-File Version 7.3) format:
1175: In case of PETSCVIEWERHDF5, a parallel HDF5 reader is used.
1176: Each processor's chunk is loaded independently by its owning rank.
1177: Multiple objects, both matrices and vectors, can be stored within the same file.
1178: They are looked up by their PetscObject name.
1180: As the MATLAB MAT-File Version 7.3 format is also a HDF5 flavor, we decided to use
1181: by default the same structure and naming of the AIJ arrays and column count
1182: within the HDF5 file. This means that a MAT file saved with -v7.3 flag, e.g.
1183: $ save example.mat A b -v7.3
1184: can be directly read by this routine (see Reference 1 for details).
1185: Note that depending on your MATLAB version, this format might be a default,
1186: otherwise you can set it as default in Preferences.
1188: Unless -nocompression flag is used to save the file in MATLAB,
1189: PETSc must be configured with ZLIB package.
1191: See also examples src/mat/tutorials/ex10.c and src/ksp/ksp/tutorials/ex27.c
1193: Current HDF5 (MAT-File) limitations:
1194: This reader currently supports only real MATSEQAIJ, MATMPIAIJ, MATSEQDENSE and MATMPIDENSE matrices.
1196: Corresponding MatView() is not yet implemented.
1198: The loaded matrix is actually a transpose of the original one in MATLAB,
1199: unless you push PETSC_VIEWER_HDF5_MAT format (see examples above).
1200: With this format, matrix is automatically transposed by PETSc,
1201: unless the matrix is marked as SPD or symmetric
1202: (see MatSetOption(), MAT_SPD, MAT_SYMMETRIC).
1204: References:
1205: 1. MATLAB(R) Documentation, manual page of save(), https://www.mathworks.com/help/matlab/ref/save.html#btox10b-1-version
1207: .seealso: PetscViewerBinaryOpen(), PetscViewerSetType(), MatView(), VecLoad()
1209: @*/
1210: PetscErrorCode MatLoad(Mat mat,PetscViewer viewer)
1211: {
1213: PetscBool flg;
1219: if (!((PetscObject)mat)->type_name) {
1220: MatSetType(mat,MATAIJ);
1221: }
1223: flg = PETSC_FALSE;
1224: PetscOptionsGetBool(((PetscObject)mat)->options,((PetscObject)mat)->prefix,"-matload_symmetric",&flg,NULL);
1225: if (flg) {
1226: MatSetOption(mat,MAT_SYMMETRIC,PETSC_TRUE);
1227: MatSetOption(mat,MAT_SYMMETRY_ETERNAL,PETSC_TRUE);
1228: }
1229: flg = PETSC_FALSE;
1230: PetscOptionsGetBool(((PetscObject)mat)->options,((PetscObject)mat)->prefix,"-matload_spd",&flg,NULL);
1231: if (flg) {
1232: MatSetOption(mat,MAT_SPD,PETSC_TRUE);
1233: }
1235: if (!mat->ops->load) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatLoad is not supported for type %s",((PetscObject)mat)->type_name);
1236: PetscLogEventBegin(MAT_Load,mat,viewer,0,0);
1237: (*mat->ops->load)(mat,viewer);
1238: PetscLogEventEnd(MAT_Load,mat,viewer,0,0);
1239: return(0);
1240: }
1242: static PetscErrorCode MatDestroy_Redundant(Mat_Redundant **redundant)
1243: {
1245: Mat_Redundant *redund = *redundant;
1246: PetscInt i;
1249: if (redund){
1250: if (redund->matseq) { /* via MatCreateSubMatrices() */
1251: ISDestroy(&redund->isrow);
1252: ISDestroy(&redund->iscol);
1253: MatDestroySubMatrices(1,&redund->matseq);
1254: } else {
1255: PetscFree2(redund->send_rank,redund->recv_rank);
1256: PetscFree(redund->sbuf_j);
1257: PetscFree(redund->sbuf_a);
1258: for (i=0; i<redund->nrecvs; i++) {
1259: PetscFree(redund->rbuf_j[i]);
1260: PetscFree(redund->rbuf_a[i]);
1261: }
1262: PetscFree4(redund->sbuf_nz,redund->rbuf_nz,redund->rbuf_j,redund->rbuf_a);
1263: }
1265: if (redund->subcomm) {
1266: PetscCommDestroy(&redund->subcomm);
1267: }
1268: PetscFree(redund);
1269: }
1270: return(0);
1271: }
1273: /*@C
1274: MatDestroy - Frees space taken by a matrix.
1276: Collective on Mat
1278: Input Parameter:
1279: . A - the matrix
1281: Level: beginner
1283: @*/
1284: PetscErrorCode MatDestroy(Mat *A)
1285: {
1289: if (!*A) return(0);
1291: if (--((PetscObject)(*A))->refct > 0) {*A = NULL; return(0);}
1293: /* if memory was published with SAWs then destroy it */
1294: PetscObjectSAWsViewOff((PetscObject)*A);
1295: if ((*A)->ops->destroy) {
1296: (*(*A)->ops->destroy)(*A);
1297: }
1299: PetscFree((*A)->defaultvectype);
1300: PetscFree((*A)->bsizes);
1301: PetscFree((*A)->solvertype);
1302: MatDestroy_Redundant(&(*A)->redundant);
1303: MatProductClear(*A);
1304: MatNullSpaceDestroy(&(*A)->nullsp);
1305: MatNullSpaceDestroy(&(*A)->transnullsp);
1306: MatNullSpaceDestroy(&(*A)->nearnullsp);
1307: MatDestroy(&(*A)->schur);
1308: PetscLayoutDestroy(&(*A)->rmap);
1309: PetscLayoutDestroy(&(*A)->cmap);
1310: PetscHeaderDestroy(A);
1311: return(0);
1312: }
1314: /*@C
1315: MatSetValues - Inserts or adds a block of values into a matrix.
1316: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
1317: MUST be called after all calls to MatSetValues() have been completed.
1319: Not Collective
1321: Input Parameters:
1322: + mat - the matrix
1323: . v - a logically two-dimensional array of values
1324: . m, idxm - the number of rows and their global indices
1325: . n, idxn - the number of columns and their global indices
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
1330: Notes:
1331: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
1332: MatSetUp() before using this routine
1334: By default the values, v, are row-oriented. See MatSetOption() for other options.
1336: Calls to MatSetValues() with the INSERT_VALUES and ADD_VALUES
1337: options cannot be mixed without intervening calls to the assembly
1338: routines.
1340: MatSetValues() uses 0-based row and column numbers in Fortran
1341: as well as in C.
1343: Negative indices may be passed in idxm and idxn, these rows and columns are
1344: simply ignored. This allows easily inserting element stiffness matrices
1345: with homogeneous Dirchlet boundary conditions that you don't want represented
1346: in the matrix.
1348: Efficiency Alert:
1349: The routine MatSetValuesBlocked() may offer much better efficiency
1350: for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1352: Level: beginner
1354: Developer Notes:
1355: This is labeled with C so does not automatically generate Fortran stubs and interfaces
1356: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
1358: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1359: InsertMode, INSERT_VALUES, ADD_VALUES
1360: @*/
1361: PetscErrorCode MatSetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1362: {
1368: if (!m || !n) return(0); /* no values to insert */
1371: MatCheckPreallocated(mat,1);
1373: if (mat->insertmode == NOT_SET_VALUES) {
1374: mat->insertmode = addv;
1375: } else if (PetscUnlikely(mat->insertmode != addv)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1376: if (PetscDefined(USE_DEBUG)) {
1377: PetscInt i,j;
1379: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1380: if (!mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1382: for (i=0; i<m; i++) {
1383: for (j=0; j<n; j++) {
1384: if (mat->erroriffailure && PetscIsInfOrNanScalar(v[i*n+j]))
1385: #if defined(PETSC_USE_COMPLEX)
1386: SETERRQ4(PETSC_COMM_SELF,PETSC_ERR_FP,"Inserting %g+ig at matrix entry (%D,%D)",(double)PetscRealPart(v[i*n+j]),(double)PetscImaginaryPart(v[i*n+j]),idxm[i],idxn[j]);
1387: #else
1388: SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_FP,"Inserting %g at matrix entry (%D,%D)",(double)v[i*n+j],idxm[i],idxn[j]);
1389: #endif
1390: }
1391: }
1392: for (i=0; i<m; i++) if (idxm[i] >= mat->rmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Cannot insert in row %D, maximum is %D",idxm[i],mat->rmap->N-1);
1393: for (i=0; i<n; i++) if (idxn[i] >= mat->cmap->N) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Cannot insert in column %D, maximum is %D",idxn[i],mat->cmap->N-1);
1394: }
1396: if (mat->assembled) {
1397: mat->was_assembled = PETSC_TRUE;
1398: mat->assembled = PETSC_FALSE;
1399: }
1400: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1401: (*mat->ops->setvalues)(mat,m,idxm,n,idxn,v,addv);
1402: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1403: return(0);
1404: }
1407: /*@
1408: MatSetValuesRowLocal - Inserts a row (block row for BAIJ matrices) of nonzero
1409: values into a matrix
1411: Not Collective
1413: Input Parameters:
1414: + mat - the matrix
1415: . row - the (block) row to set
1416: - v - a logically two-dimensional array of values
1418: Notes:
1419: By the values, v, are column-oriented (for the block version) and sorted
1421: All the nonzeros in the row must be provided
1423: The matrix must have previously had its column indices set
1425: The row must belong to this process
1427: Level: intermediate
1429: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1430: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues(), MatSetValuesRow(), MatSetLocalToGlobalMapping()
1431: @*/
1432: PetscErrorCode MatSetValuesRowLocal(Mat mat,PetscInt row,const PetscScalar v[])
1433: {
1435: PetscInt globalrow;
1441: ISLocalToGlobalMappingApply(mat->rmap->mapping,1,&row,&globalrow);
1442: MatSetValuesRow(mat,globalrow,v);
1443: return(0);
1444: }
1446: /*@
1447: MatSetValuesRow - Inserts a row (block row for BAIJ matrices) of nonzero
1448: values into a matrix
1450: Not Collective
1452: Input Parameters:
1453: + mat - the matrix
1454: . row - the (block) row to set
1455: - v - a logically two-dimensional (column major) array of values for block matrices with blocksize larger than one, otherwise a one dimensional array of values
1457: Notes:
1458: The values, v, are column-oriented for the block version.
1460: All the nonzeros in the row must be provided
1462: THE MATRIX MUST HAVE PREVIOUSLY HAD ITS COLUMN INDICES SET. IT IS RARE THAT THIS ROUTINE IS USED, usually MatSetValues() is used.
1464: The row must belong to this process
1466: Level: advanced
1468: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1469: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues()
1470: @*/
1471: PetscErrorCode MatSetValuesRow(Mat mat,PetscInt row,const PetscScalar v[])
1472: {
1478: MatCheckPreallocated(mat,1);
1480: if (PetscUnlikely(mat->insertmode == ADD_VALUES)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add and insert values");
1481: if (PetscUnlikely(mat->factortype)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1482: mat->insertmode = INSERT_VALUES;
1484: if (mat->assembled) {
1485: mat->was_assembled = PETSC_TRUE;
1486: mat->assembled = PETSC_FALSE;
1487: }
1488: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1489: if (!mat->ops->setvaluesrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1490: (*mat->ops->setvaluesrow)(mat,row,v);
1491: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1492: return(0);
1493: }
1495: /*@
1496: MatSetValuesStencil - Inserts or adds a block of values into a matrix.
1497: Using structured grid indexing
1499: Not Collective
1501: Input Parameters:
1502: + mat - the matrix
1503: . m - number of rows being entered
1504: . idxm - grid coordinates (and component number when dof > 1) for matrix rows being entered
1505: . n - number of columns being entered
1506: . idxn - grid coordinates (and component number when dof > 1) for matrix columns being entered
1507: . v - a logically two-dimensional array of values
1508: - addv - either ADD_VALUES or INSERT_VALUES, where
1509: ADD_VALUES adds values to any existing entries, and
1510: INSERT_VALUES replaces existing entries with new values
1512: Notes:
1513: By default the values, v, are row-oriented. See MatSetOption() for other options.
1515: Calls to MatSetValuesStencil() with the INSERT_VALUES and ADD_VALUES
1516: options cannot be mixed without intervening calls to the assembly
1517: routines.
1519: The grid coordinates are across the entire grid, not just the local portion
1521: MatSetValuesStencil() uses 0-based row and column numbers in Fortran
1522: as well as in C.
1524: For setting/accessing vector values via array coordinates you can use the DMDAVecGetArray() routine
1526: In order to use this routine you must either obtain the matrix with DMCreateMatrix()
1527: or call MatSetLocalToGlobalMapping() and MatSetStencil() first.
1529: The columns and rows in the stencil passed in MUST be contained within the
1530: ghost region of the given process as set with DMDACreateXXX() or MatSetStencil(). For example,
1531: if you create a DMDA with an overlap of one grid level and on a particular process its first
1532: local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1533: first i index you can use in your column and row indices in MatSetStencil() is 5.
1535: In Fortran idxm and idxn should be declared as
1536: $ MatStencil idxm(4,m),idxn(4,n)
1537: and the values inserted using
1538: $ idxm(MatStencil_i,1) = i
1539: $ idxm(MatStencil_j,1) = j
1540: $ idxm(MatStencil_k,1) = k
1541: $ idxm(MatStencil_c,1) = c
1542: etc
1544: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
1545: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
1546: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
1547: DM_BOUNDARY_PERIODIC boundary type.
1549: 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
1550: a single value per point) you can skip filling those indices.
1552: Inspired by the structured grid interface to the HYPRE package
1553: (https://computation.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods)
1555: Efficiency Alert:
1556: The routine MatSetValuesBlockedStencil() may offer much better efficiency
1557: for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1559: Level: beginner
1561: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1562: MatSetValues(), MatSetValuesBlockedStencil(), MatSetStencil(), DMCreateMatrix(), DMDAVecGetArray(), MatStencil
1563: @*/
1564: PetscErrorCode MatSetValuesStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1565: {
1567: PetscInt buf[8192],*bufm=NULL,*bufn=NULL,*jdxm,*jdxn;
1568: PetscInt j,i,dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1569: PetscInt *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1572: if (!m || !n) return(0); /* no values to insert */
1578: if ((m+n) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1579: jdxm = buf; jdxn = buf+m;
1580: } else {
1581: PetscMalloc2(m,&bufm,n,&bufn);
1582: jdxm = bufm; jdxn = bufn;
1583: }
1584: for (i=0; i<m; i++) {
1585: for (j=0; j<3-sdim; j++) dxm++;
1586: tmp = *dxm++ - starts[0];
1587: for (j=0; j<dim-1; j++) {
1588: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1589: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1590: }
1591: if (mat->stencil.noc) dxm++;
1592: jdxm[i] = tmp;
1593: }
1594: for (i=0; i<n; i++) {
1595: for (j=0; j<3-sdim; j++) dxn++;
1596: tmp = *dxn++ - starts[0];
1597: for (j=0; j<dim-1; j++) {
1598: if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1599: else tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1600: }
1601: if (mat->stencil.noc) dxn++;
1602: jdxn[i] = tmp;
1603: }
1604: MatSetValuesLocal(mat,m,jdxm,n,jdxn,v,addv);
1605: PetscFree2(bufm,bufn);
1606: return(0);
1607: }
1609: /*@
1610: MatSetValuesBlockedStencil - Inserts or adds a block of values into a matrix.
1611: Using structured grid indexing
1613: Not Collective
1615: Input Parameters:
1616: + mat - the matrix
1617: . m - number of rows being entered
1618: . idxm - grid coordinates for matrix rows being entered
1619: . n - number of columns being entered
1620: . idxn - grid coordinates for matrix columns being entered
1621: . v - a logically two-dimensional array of values
1622: - addv - either ADD_VALUES or INSERT_VALUES, where
1623: ADD_VALUES adds values to any existing entries, and
1624: INSERT_VALUES replaces existing entries with new values
1626: Notes:
1627: By default the values, v, are row-oriented and unsorted.
1628: See MatSetOption() for other options.
1630: Calls to MatSetValuesBlockedStencil() with the INSERT_VALUES and ADD_VALUES
1631: options cannot be mixed without intervening calls to the assembly
1632: routines.
1634: The grid coordinates are across the entire grid, not just the local portion
1636: MatSetValuesBlockedStencil() uses 0-based row and column numbers in Fortran
1637: as well as in C.
1639: For setting/accessing vector values via array coordinates you can use the DMDAVecGetArray() routine
1641: In order to use this routine you must either obtain the matrix with DMCreateMatrix()
1642: or call MatSetBlockSize(), MatSetLocalToGlobalMapping() and MatSetStencil() first.
1644: The columns and rows in the stencil passed in MUST be contained within the
1645: ghost region of the given process as set with DMDACreateXXX() or MatSetStencil(). For example,
1646: if you create a DMDA with an overlap of one grid level and on a particular process its first
1647: local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1648: first i index you can use in your column and row indices in MatSetStencil() is 5.
1650: In Fortran idxm and idxn should be declared as
1651: $ MatStencil idxm(4,m),idxn(4,n)
1652: and the values inserted using
1653: $ idxm(MatStencil_i,1) = i
1654: $ idxm(MatStencil_j,1) = j
1655: $ idxm(MatStencil_k,1) = k
1656: etc
1658: Negative indices may be passed in idxm and idxn, these rows and columns are
1659: simply ignored. This allows easily inserting element stiffness matrices
1660: with homogeneous Dirchlet boundary conditions that you don't want represented
1661: in the matrix.
1663: Inspired by the structured grid interface to the HYPRE package
1664: (https://computation.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods)
1666: Level: beginner
1668: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1669: MatSetValues(), MatSetValuesStencil(), MatSetStencil(), DMCreateMatrix(), DMDAVecGetArray(), MatStencil,
1670: MatSetBlockSize(), MatSetLocalToGlobalMapping()
1671: @*/
1672: PetscErrorCode MatSetValuesBlockedStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1673: {
1675: PetscInt buf[8192],*bufm=NULL,*bufn=NULL,*jdxm,*jdxn;
1676: PetscInt j,i,dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1677: PetscInt *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1680: if (!m || !n) return(0); /* no values to insert */
1687: if ((m+n) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1688: jdxm = buf; jdxn = buf+m;
1689: } else {
1690: PetscMalloc2(m,&bufm,n,&bufn);
1691: jdxm = bufm; jdxn = bufn;
1692: }
1693: for (i=0; i<m; i++) {
1694: for (j=0; j<3-sdim; j++) dxm++;
1695: tmp = *dxm++ - starts[0];
1696: for (j=0; j<sdim-1; j++) {
1697: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1698: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1699: }
1700: dxm++;
1701: jdxm[i] = tmp;
1702: }
1703: for (i=0; i<n; i++) {
1704: for (j=0; j<3-sdim; j++) dxn++;
1705: tmp = *dxn++ - starts[0];
1706: for (j=0; j<sdim-1; j++) {
1707: if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1708: else tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1709: }
1710: dxn++;
1711: jdxn[i] = tmp;
1712: }
1713: MatSetValuesBlockedLocal(mat,m,jdxm,n,jdxn,v,addv);
1714: PetscFree2(bufm,bufn);
1715: return(0);
1716: }
1718: /*@
1719: MatSetStencil - Sets the grid information for setting values into a matrix via
1720: MatSetValuesStencil()
1722: Not Collective
1724: Input Parameters:
1725: + mat - the matrix
1726: . dim - dimension of the grid 1, 2, or 3
1727: . dims - number of grid points in x, y, and z direction, including ghost points on your processor
1728: . starts - starting point of ghost nodes on your processor in x, y, and z direction
1729: - dof - number of degrees of freedom per node
1732: Inspired by the structured grid interface to the HYPRE package
1733: (www.llnl.gov/CASC/hyper)
1735: For matrices generated with DMCreateMatrix() this routine is automatically called and so not needed by the
1736: user.
1738: Level: beginner
1740: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1741: MatSetValues(), MatSetValuesBlockedStencil(), MatSetValuesStencil()
1742: @*/
1743: PetscErrorCode MatSetStencil(Mat mat,PetscInt dim,const PetscInt dims[],const PetscInt starts[],PetscInt dof)
1744: {
1745: PetscInt i;
1752: mat->stencil.dim = dim + (dof > 1);
1753: for (i=0; i<dim; i++) {
1754: mat->stencil.dims[i] = dims[dim-i-1]; /* copy the values in backwards */
1755: mat->stencil.starts[i] = starts[dim-i-1];
1756: }
1757: mat->stencil.dims[dim] = dof;
1758: mat->stencil.starts[dim] = 0;
1759: mat->stencil.noc = (PetscBool)(dof == 1);
1760: return(0);
1761: }
1763: /*@C
1764: MatSetValuesBlocked - Inserts or adds a block of values into a matrix.
1766: Not Collective
1768: Input Parameters:
1769: + mat - the matrix
1770: . v - a logically two-dimensional array of values
1771: . m, idxm - the number of block rows and their global block indices
1772: . n, idxn - the number of block columns and their global block indices
1773: - addv - either ADD_VALUES or INSERT_VALUES, where
1774: ADD_VALUES adds values to any existing entries, and
1775: INSERT_VALUES replaces existing entries with new values
1777: Notes:
1778: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call
1779: MatXXXXSetPreallocation() or MatSetUp() before using this routine.
1781: The m and n count the NUMBER of blocks in the row direction and column direction,
1782: NOT the total number of rows/columns; for example, if the block size is 2 and
1783: you are passing in values for rows 2,3,4,5 then m would be 2 (not 4).
1784: The values in idxm would be 1 2; that is the first index for each block divided by
1785: the block size.
1787: Note that you must call MatSetBlockSize() when constructing this matrix (before
1788: preallocating it).
1790: By default the values, v, are row-oriented, so the layout of
1791: v is the same as for MatSetValues(). See MatSetOption() for other options.
1793: Calls to MatSetValuesBlocked() with the INSERT_VALUES and ADD_VALUES
1794: options cannot be mixed without intervening calls to the assembly
1795: routines.
1797: MatSetValuesBlocked() uses 0-based row and column numbers in Fortran
1798: as well as in C.
1800: Negative indices may be passed in idxm and idxn, these rows and columns are
1801: simply ignored. This allows easily inserting element stiffness matrices
1802: with homogeneous Dirchlet boundary conditions that you don't want represented
1803: in the matrix.
1805: Each time an entry is set within a sparse matrix via MatSetValues(),
1806: internal searching must be done to determine where to place the
1807: data in the matrix storage space. By instead inserting blocks of
1808: entries via MatSetValuesBlocked(), the overhead of matrix assembly is
1809: reduced.
1811: Example:
1812: $ Suppose m=n=2 and block size(bs) = 2 The array is
1813: $
1814: $ 1 2 | 3 4
1815: $ 5 6 | 7 8
1816: $ - - - | - - -
1817: $ 9 10 | 11 12
1818: $ 13 14 | 15 16
1819: $
1820: $ v[] should be passed in like
1821: $ v[] = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16]
1822: $
1823: $ If you are not using row oriented storage of v (that is you called MatSetOption(mat,MAT_ROW_ORIENTED,PETSC_FALSE)) then
1824: $ v[] = [1,5,9,13,2,6,10,14,3,7,11,15,4,8,12,16]
1826: Level: intermediate
1828: .seealso: MatSetBlockSize(), MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesBlockedLocal()
1829: @*/
1830: PetscErrorCode MatSetValuesBlocked(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1831: {
1837: if (!m || !n) return(0); /* no values to insert */
1841: MatCheckPreallocated(mat,1);
1842: if (mat->insertmode == NOT_SET_VALUES) {
1843: mat->insertmode = addv;
1844: } else if (PetscUnlikely(mat->insertmode != addv)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1845: if (PetscDefined(USE_DEBUG)) {
1846: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1847: if (!mat->ops->setvaluesblocked && !mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1848: }
1849: if (PetscDefined(USE_DEBUG)) {
1850: PetscInt rbs,cbs,M,N,i;
1851: MatGetBlockSizes(mat,&rbs,&cbs);
1852: MatGetSize(mat,&M,&N);
1853: for (i=0; i<m; i++) {
1854: if (idxm[i]*rbs >= M) SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Row block index %D (index %D) greater than row length %D",i,idxm[i],M);
1855: }
1856: for (i=0; i<n; i++) {
1857: if (idxn[i]*cbs >= N) SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Column block index %D (index %D) great than column length %D",i,idxn[i],N);
1858: }
1859: }
1860: if (mat->assembled) {
1861: mat->was_assembled = PETSC_TRUE;
1862: mat->assembled = PETSC_FALSE;
1863: }
1864: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1865: if (mat->ops->setvaluesblocked) {
1866: (*mat->ops->setvaluesblocked)(mat,m,idxm,n,idxn,v,addv);
1867: } else {
1868: PetscInt buf[8192],*bufr=NULL,*bufc=NULL,*iidxm,*iidxn;
1869: PetscInt i,j,bs,cbs;
1870: MatGetBlockSizes(mat,&bs,&cbs);
1871: if (m*bs+n*cbs <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1872: iidxm = buf; iidxn = buf + m*bs;
1873: } else {
1874: PetscMalloc2(m*bs,&bufr,n*cbs,&bufc);
1875: iidxm = bufr; iidxn = bufc;
1876: }
1877: for (i=0; i<m; i++) {
1878: for (j=0; j<bs; j++) {
1879: iidxm[i*bs+j] = bs*idxm[i] + j;
1880: }
1881: }
1882: for (i=0; i<n; i++) {
1883: for (j=0; j<cbs; j++) {
1884: iidxn[i*cbs+j] = cbs*idxn[i] + j;
1885: }
1886: }
1887: MatSetValues(mat,m*bs,iidxm,n*cbs,iidxn,v,addv);
1888: PetscFree2(bufr,bufc);
1889: }
1890: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1891: return(0);
1892: }
1894: /*@C
1895: MatGetValues - Gets a block of values from a matrix.
1897: Not Collective; can only return values that are owned by the give process
1899: Input Parameters:
1900: + mat - the matrix
1901: . v - a logically two-dimensional array for storing the values
1902: . m, idxm - the number of rows and their global indices
1903: - n, idxn - the number of columns and their global indices
1905: Notes:
1906: The user must allocate space (m*n PetscScalars) for the values, v.
1907: The values, v, are then returned in a row-oriented format,
1908: analogous to that used by default in MatSetValues().
1910: MatGetValues() uses 0-based row and column numbers in
1911: Fortran as well as in C.
1913: MatGetValues() requires that the matrix has been assembled
1914: with MatAssemblyBegin()/MatAssemblyEnd(). Thus, calls to
1915: MatSetValues() and MatGetValues() CANNOT be made in succession
1916: without intermediate matrix assembly.
1918: Negative row or column indices will be ignored and those locations in v[] will be
1919: left unchanged.
1921: For the standard row-based matrix formats, idxm[] can only contain rows owned by the requesting MPI rank.
1922: That is, rows with global index greater than or equal to restart and less than rend where restart and rend are obtainable
1923: from MatGetOwnershipRange(mat,&rstart,&rend).
1925: Level: advanced
1927: .seealso: MatGetRow(), MatCreateSubMatrices(), MatSetValues(), MatGetOwnershipRange(), MatGetValuesLocal()
1928: @*/
1929: PetscErrorCode MatGetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[])
1930: {
1936: if (!m || !n) return(0);
1940: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1941: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1942: if (!mat->ops->getvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1943: MatCheckPreallocated(mat,1);
1945: PetscLogEventBegin(MAT_GetValues,mat,0,0,0);
1946: (*mat->ops->getvalues)(mat,m,idxm,n,idxn,v);
1947: PetscLogEventEnd(MAT_GetValues,mat,0,0,0);
1948: return(0);
1949: }
1951: /*@C
1952: MatGetValuesLocal - retrieves values from certain locations in a matrix using the local numbering of the indices
1953: defined previously by MatSetLocalToGlobalMapping()
1955: Not Collective
1957: Input Parameters:
1958: + mat - the matrix
1959: . nrow, irow - number of rows and their local indices
1960: - ncol, icol - number of columns and their local indices
1962: Output Parameter:
1963: . y - a logically two-dimensional array of values
1965: Notes:
1966: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatSetLocalToGlobalMapping() before using this routine.
1968: This routine can only return values that are owned by the requesting MPI rank. That is, for standard matrix formats, rows that, in the global numbering,
1969: are greater than or equal to restart and less than rend where restart and rend are obtainable from MatGetOwnershipRange(mat,&rstart,&rend). One can
1970: determine if the resulting global row associated with the local row r is owned by the requesting MPI rank by applying the ISLocalToGlobalMapping set
1971: with MatSetLocalToGlobalMapping().
1973: Developer Notes:
1974: This is labelled with C so does not automatically generate Fortran stubs and interfaces
1975: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
1977: Level: advanced
1979: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetLocalToGlobalMapping(),
1980: MatSetValuesLocal(), MatGetValues()
1981: @*/
1982: PetscErrorCode MatGetValuesLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],PetscScalar y[])
1983: {
1989: MatCheckPreallocated(mat,1);
1990: if (!nrow || !ncol) return(0); /* no values to retrieve */
1993: if (PetscDefined(USE_DEBUG)) {
1994: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1995: if (!mat->ops->getvalueslocal && !mat->ops->getvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1996: }
1997: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1998: PetscLogEventBegin(MAT_GetValues,mat,0,0,0);
1999: if (mat->ops->getvalueslocal) {
2000: (*mat->ops->getvalueslocal)(mat,nrow,irow,ncol,icol,y);
2001: } else {
2002: PetscInt buf[8192],*bufr=NULL,*bufc=NULL,*irowm,*icolm;
2003: if ((nrow+ncol) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
2004: irowm = buf; icolm = buf+nrow;
2005: } else {
2006: PetscMalloc2(nrow,&bufr,ncol,&bufc);
2007: irowm = bufr; icolm = bufc;
2008: }
2009: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MatGetValuesLocal() cannot proceed without local-to-global row mapping (See MatSetLocalToGlobalMapping()).");
2010: if (!mat->cmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MatGetValuesLocal() cannot proceed without local-to-global column mapping (See MatSetLocalToGlobalMapping()).");
2011: ISLocalToGlobalMappingApply(mat->rmap->mapping,nrow,irow,irowm);
2012: ISLocalToGlobalMappingApply(mat->cmap->mapping,ncol,icol,icolm);
2013: MatGetValues(mat,nrow,irowm,ncol,icolm,y);
2014: PetscFree2(bufr,bufc);
2015: }
2016: PetscLogEventEnd(MAT_GetValues,mat,0,0,0);
2017: return(0);
2018: }
2020: /*@
2021: MatSetValuesBatch - Adds (ADD_VALUES) many blocks of values into a matrix at once. The blocks must all be square and
2022: the same size. Currently, this can only be called once and creates the given matrix.
2024: Not Collective
2026: Input Parameters:
2027: + mat - the matrix
2028: . nb - the number of blocks
2029: . bs - the number of rows (and columns) in each block
2030: . rows - a concatenation of the rows for each block
2031: - v - a concatenation of logically two-dimensional arrays of values
2033: Notes:
2034: In the future, we will extend this routine to handle rectangular blocks, and to allow multiple calls for a given matrix.
2036: Level: advanced
2038: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
2039: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues()
2040: @*/
2041: PetscErrorCode MatSetValuesBatch(Mat mat, PetscInt nb, PetscInt bs, PetscInt rows[], const PetscScalar v[])
2042: {
2050: if (PetscUnlikelyDebug(mat->factortype)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2052: PetscLogEventBegin(MAT_SetValuesBatch,mat,0,0,0);
2053: if (mat->ops->setvaluesbatch) {
2054: (*mat->ops->setvaluesbatch)(mat,nb,bs,rows,v);
2055: } else {
2056: PetscInt b;
2057: for (b = 0; b < nb; ++b) {
2058: MatSetValues(mat, bs, &rows[b*bs], bs, &rows[b*bs], &v[b*bs*bs], ADD_VALUES);
2059: }
2060: }
2061: PetscLogEventEnd(MAT_SetValuesBatch,mat,0,0,0);
2062: return(0);
2063: }
2065: /*@
2066: MatSetLocalToGlobalMapping - Sets a local-to-global numbering for use by
2067: the routine MatSetValuesLocal() to allow users to insert matrix entries
2068: using a local (per-processor) numbering.
2070: Not Collective
2072: Input Parameters:
2073: + x - the matrix
2074: . rmapping - row mapping created with ISLocalToGlobalMappingCreate() or ISLocalToGlobalMappingCreateIS()
2075: - cmapping - column mapping
2077: Level: intermediate
2080: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesLocal(), MatGetValuesLocal()
2081: @*/
2082: PetscErrorCode MatSetLocalToGlobalMapping(Mat x,ISLocalToGlobalMapping rmapping,ISLocalToGlobalMapping cmapping)
2083: {
2092: if (x->ops->setlocaltoglobalmapping) {
2093: (*x->ops->setlocaltoglobalmapping)(x,rmapping,cmapping);
2094: } else {
2095: PetscLayoutSetISLocalToGlobalMapping(x->rmap,rmapping);
2096: PetscLayoutSetISLocalToGlobalMapping(x->cmap,cmapping);
2097: }
2098: return(0);
2099: }
2102: /*@
2103: MatGetLocalToGlobalMapping - Gets the local-to-global numbering set by MatSetLocalToGlobalMapping()
2105: Not Collective
2107: Input Parameters:
2108: . A - the matrix
2110: Output Parameters:
2111: + rmapping - row mapping
2112: - cmapping - column mapping
2114: Level: advanced
2117: .seealso: MatSetValuesLocal()
2118: @*/
2119: PetscErrorCode MatGetLocalToGlobalMapping(Mat A,ISLocalToGlobalMapping *rmapping,ISLocalToGlobalMapping *cmapping)
2120: {
2126: if (rmapping) *rmapping = A->rmap->mapping;
2127: if (cmapping) *cmapping = A->cmap->mapping;
2128: return(0);
2129: }
2131: /*@
2132: MatSetLayouts - Sets the PetscLayout objects for rows and columns of a matrix
2134: Logically Collective on A
2136: Input Parameters:
2137: + A - the matrix
2138: . rmap - row layout
2139: - cmap - column layout
2141: Level: advanced
2143: .seealso: MatCreateVecs(), MatGetLocalToGlobalMapping(), MatGetLayouts()
2144: @*/
2145: PetscErrorCode MatSetLayouts(Mat A,PetscLayout rmap,PetscLayout cmap)
2146: {
2152: PetscLayoutReference(rmap,&A->rmap);
2153: PetscLayoutReference(cmap,&A->cmap);
2154: return(0);
2155: }
2157: /*@
2158: MatGetLayouts - Gets the PetscLayout objects for rows and columns
2160: Not Collective
2162: Input Parameters:
2163: . A - the matrix
2165: Output Parameters:
2166: + rmap - row layout
2167: - cmap - column layout
2169: Level: advanced
2171: .seealso: MatCreateVecs(), MatGetLocalToGlobalMapping(), MatSetLayouts()
2172: @*/
2173: PetscErrorCode MatGetLayouts(Mat A,PetscLayout *rmap,PetscLayout *cmap)
2174: {
2180: if (rmap) *rmap = A->rmap;
2181: if (cmap) *cmap = A->cmap;
2182: return(0);
2183: }
2185: /*@C
2186: MatSetValuesLocal - Inserts or adds values into certain locations of a matrix,
2187: using a local numbering of the nodes.
2189: Not Collective
2191: Input Parameters:
2192: + mat - the matrix
2193: . nrow, irow - number of rows and their local indices
2194: . ncol, icol - number of columns and their local indices
2195: . y - a logically two-dimensional array of values
2196: - addv - either INSERT_VALUES or ADD_VALUES, where
2197: ADD_VALUES adds values to any existing entries, and
2198: INSERT_VALUES replaces existing entries with new values
2200: Notes:
2201: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
2202: MatSetUp() before using this routine
2204: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatSetLocalToGlobalMapping() before using this routine
2206: Calls to MatSetValuesLocal() with the INSERT_VALUES and ADD_VALUES
2207: options cannot be mixed without intervening calls to the assembly
2208: routines.
2210: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
2211: MUST be called after all calls to MatSetValuesLocal() have been completed.
2213: Level: intermediate
2215: Developer Notes:
2216: This is labeled with C so does not automatically generate Fortran stubs and interfaces
2217: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
2219: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetLocalToGlobalMapping(),
2220: MatSetValueLocal(), MatGetValuesLocal()
2221: @*/
2222: PetscErrorCode MatSetValuesLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
2223: {
2229: MatCheckPreallocated(mat,1);
2230: if (!nrow || !ncol) return(0); /* no values to insert */
2233: if (mat->insertmode == NOT_SET_VALUES) {
2234: mat->insertmode = addv;
2235: }
2236: else if (PetscUnlikely(mat->insertmode != addv)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
2237: if (PetscDefined(USE_DEBUG)) {
2238: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2239: if (!mat->ops->setvalueslocal && !mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2240: }
2242: if (mat->assembled) {
2243: mat->was_assembled = PETSC_TRUE;
2244: mat->assembled = PETSC_FALSE;
2245: }
2246: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
2247: if (mat->ops->setvalueslocal) {
2248: (*mat->ops->setvalueslocal)(mat,nrow,irow,ncol,icol,y,addv);
2249: } else {
2250: PetscInt buf[8192],*bufr=NULL,*bufc=NULL,*irowm,*icolm;
2251: if ((nrow+ncol) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
2252: irowm = buf; icolm = buf+nrow;
2253: } else {
2254: PetscMalloc2(nrow,&bufr,ncol,&bufc);
2255: irowm = bufr; icolm = bufc;
2256: }
2257: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MatSetValuesLocal() cannot proceed without local-to-global row mapping (See MatSetLocalToGlobalMapping()).");
2258: if (!mat->cmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MatSetValuesLocal() cannot proceed without local-to-global column mapping (See MatSetLocalToGlobalMapping()).");
2259: ISLocalToGlobalMappingApply(mat->rmap->mapping,nrow,irow,irowm);
2260: ISLocalToGlobalMappingApply(mat->cmap->mapping,ncol,icol,icolm);
2261: MatSetValues(mat,nrow,irowm,ncol,icolm,y,addv);
2262: PetscFree2(bufr,bufc);
2263: }
2264: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
2265: return(0);
2266: }
2268: /*@C
2269: MatSetValuesBlockedLocal - Inserts or adds values into certain locations of a matrix,
2270: using a local ordering of the nodes a block at a time.
2272: Not Collective
2274: Input Parameters:
2275: + x - the matrix
2276: . nrow, irow - number of rows and their local indices
2277: . ncol, icol - number of columns and their local indices
2278: . y - a logically two-dimensional array of values
2279: - addv - either INSERT_VALUES or ADD_VALUES, where
2280: ADD_VALUES adds values to any existing entries, and
2281: INSERT_VALUES replaces existing entries with new values
2283: Notes:
2284: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
2285: MatSetUp() before using this routine
2287: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatSetBlockSize() and MatSetLocalToGlobalMapping()
2288: before using this routineBefore calling MatSetValuesLocal(), the user must first set the
2290: Calls to MatSetValuesBlockedLocal() with the INSERT_VALUES and ADD_VALUES
2291: options cannot be mixed without intervening calls to the assembly
2292: routines.
2294: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
2295: MUST be called after all calls to MatSetValuesBlockedLocal() have been completed.
2297: Level: intermediate
2299: Developer Notes:
2300: This is labeled with C so does not automatically generate Fortran stubs and interfaces
2301: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
2303: .seealso: MatSetBlockSize(), MatSetLocalToGlobalMapping(), MatAssemblyBegin(), MatAssemblyEnd(),
2304: MatSetValuesLocal(), MatSetValuesBlocked()
2305: @*/
2306: PetscErrorCode MatSetValuesBlockedLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
2307: {
2313: MatCheckPreallocated(mat,1);
2314: if (!nrow || !ncol) return(0); /* no values to insert */
2318: if (mat->insertmode == NOT_SET_VALUES) {
2319: mat->insertmode = addv;
2320: } else if (PetscUnlikely(mat->insertmode != addv)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
2321: if (PetscDefined(USE_DEBUG)) {
2322: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2323: if (!mat->ops->setvaluesblockedlocal && !mat->ops->setvaluesblocked && !mat->ops->setvalueslocal && !mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2324: }
2326: if (mat->assembled) {
2327: mat->was_assembled = PETSC_TRUE;
2328: mat->assembled = PETSC_FALSE;
2329: }
2330: if (PetscUnlikelyDebug(mat->rmap->mapping)) { /* Condition on the mapping existing, because MatSetValuesBlockedLocal_IS does not require it to be set. */
2331: PetscInt irbs, rbs;
2332: MatGetBlockSizes(mat, &rbs, NULL);
2333: ISLocalToGlobalMappingGetBlockSize(mat->rmap->mapping,&irbs);
2334: if (rbs != irbs) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Different row block sizes! mat %D, row l2g map %D",rbs,irbs);
2335: }
2336: if (PetscUnlikelyDebug(mat->cmap->mapping)) {
2337: PetscInt icbs, cbs;
2338: MatGetBlockSizes(mat,NULL,&cbs);
2339: ISLocalToGlobalMappingGetBlockSize(mat->cmap->mapping,&icbs);
2340: if (cbs != icbs) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Different col block sizes! mat %D, col l2g map %D",cbs,icbs);
2341: }
2342: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
2343: if (mat->ops->setvaluesblockedlocal) {
2344: (*mat->ops->setvaluesblockedlocal)(mat,nrow,irow,ncol,icol,y,addv);
2345: } else {
2346: PetscInt buf[8192],*bufr=NULL,*bufc=NULL,*irowm,*icolm;
2347: if ((nrow+ncol) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
2348: irowm = buf; icolm = buf + nrow;
2349: } else {
2350: PetscMalloc2(nrow,&bufr,ncol,&bufc);
2351: irowm = bufr; icolm = bufc;
2352: }
2353: ISLocalToGlobalMappingApplyBlock(mat->rmap->mapping,nrow,irow,irowm);
2354: ISLocalToGlobalMappingApplyBlock(mat->cmap->mapping,ncol,icol,icolm);
2355: MatSetValuesBlocked(mat,nrow,irowm,ncol,icolm,y,addv);
2356: PetscFree2(bufr,bufc);
2357: }
2358: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
2359: return(0);
2360: }
2362: /*@
2363: MatMultDiagonalBlock - Computes the matrix-vector product, y = Dx. Where D is defined by the inode or block structure of the diagonal
2365: Collective on Mat
2367: Input Parameters:
2368: + mat - the matrix
2369: - x - the vector to be multiplied
2371: Output Parameters:
2372: . y - the result
2374: Notes:
2375: The vectors x and y cannot be the same. I.e., one cannot
2376: call MatMult(A,y,y).
2378: Level: developer
2380: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2381: @*/
2382: PetscErrorCode MatMultDiagonalBlock(Mat mat,Vec x,Vec y)
2383: {
2392: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2393: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2394: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2395: MatCheckPreallocated(mat,1);
2397: if (!mat->ops->multdiagonalblock) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply defined",((PetscObject)mat)->type_name);
2398: (*mat->ops->multdiagonalblock)(mat,x,y);
2399: PetscObjectStateIncrease((PetscObject)y);
2400: return(0);
2401: }
2403: /* --------------------------------------------------------*/
2404: /*@
2405: MatMult - Computes the matrix-vector product, y = Ax.
2407: Neighbor-wise Collective on Mat
2409: Input Parameters:
2410: + mat - the matrix
2411: - x - the vector to be multiplied
2413: Output Parameters:
2414: . y - the result
2416: Notes:
2417: The vectors x and y cannot be the same. I.e., one cannot
2418: call MatMult(A,y,y).
2420: Level: beginner
2422: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2423: @*/
2424: PetscErrorCode MatMult(Mat mat,Vec x,Vec y)
2425: {
2433: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2434: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2435: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2436: #if !defined(PETSC_HAVE_CONSTRAINTS)
2437: 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);
2438: 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);
2439: 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);
2440: #endif
2441: VecSetErrorIfLocked(y,3);
2442: if (mat->erroriffailure) {VecValidValues(x,2,PETSC_TRUE);}
2443: MatCheckPreallocated(mat,1);
2445: VecLockReadPush(x);
2446: if (!mat->ops->mult) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply defined",((PetscObject)mat)->type_name);
2447: PetscLogEventBegin(MAT_Mult,mat,x,y,0);
2448: (*mat->ops->mult)(mat,x,y);
2449: PetscLogEventEnd(MAT_Mult,mat,x,y,0);
2450: if (mat->erroriffailure) {VecValidValues(y,3,PETSC_FALSE);}
2451: VecLockReadPop(x);
2452: return(0);
2453: }
2455: /*@
2456: MatMultTranspose - Computes matrix transpose times a vector y = A^T * x.
2458: Neighbor-wise Collective on Mat
2460: Input Parameters:
2461: + mat - the matrix
2462: - x - the vector to be multiplied
2464: Output Parameters:
2465: . y - the result
2467: Notes:
2468: The vectors x and y cannot be the same. I.e., one cannot
2469: call MatMultTranspose(A,y,y).
2471: For complex numbers this does NOT compute the Hermitian (complex conjugate) transpose multiple,
2472: use MatMultHermitianTranspose()
2474: Level: beginner
2476: .seealso: MatMult(), MatMultAdd(), MatMultTransposeAdd(), MatMultHermitianTranspose(), MatTranspose()
2477: @*/
2478: PetscErrorCode MatMultTranspose(Mat mat,Vec x,Vec y)
2479: {
2480: PetscErrorCode (*op)(Mat,Vec,Vec)=NULL,ierr;
2488: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2489: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2490: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2491: #if !defined(PETSC_HAVE_CONSTRAINTS)
2492: 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);
2493: 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);
2494: #endif
2495: if (mat->erroriffailure) {VecValidValues(x,2,PETSC_TRUE);}
2496: MatCheckPreallocated(mat,1);
2498: if (!mat->ops->multtranspose) {
2499: if (mat->symmetric && mat->ops->mult) op = mat->ops->mult;
2500: if (!op) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply transpose defined or is symmetric and does not have a multiply defined",((PetscObject)mat)->type_name);
2501: } else op = mat->ops->multtranspose;
2502: PetscLogEventBegin(MAT_MultTranspose,mat,x,y,0);
2503: VecLockReadPush(x);
2504: (*op)(mat,x,y);
2505: VecLockReadPop(x);
2506: PetscLogEventEnd(MAT_MultTranspose,mat,x,y,0);
2507: PetscObjectStateIncrease((PetscObject)y);
2508: if (mat->erroriffailure) {VecValidValues(y,3,PETSC_FALSE);}
2509: return(0);
2510: }
2512: /*@
2513: MatMultHermitianTranspose - Computes matrix Hermitian transpose times a vector.
2515: Neighbor-wise Collective on Mat
2517: Input Parameters:
2518: + mat - the matrix
2519: - x - the vector to be multilplied
2521: Output Parameters:
2522: . y - the result
2524: Notes:
2525: The vectors x and y cannot be the same. I.e., one cannot
2526: call MatMultHermitianTranspose(A,y,y).
2528: Also called the conjugate transpose, complex conjugate transpose, or adjoint.
2530: For real numbers MatMultTranspose() and MatMultHermitianTranspose() are identical.
2532: Level: beginner
2534: .seealso: MatMult(), MatMultAdd(), MatMultHermitianTransposeAdd(), MatMultTranspose()
2535: @*/
2536: PetscErrorCode MatMultHermitianTranspose(Mat mat,Vec x,Vec y)
2537: {
2546: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2547: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2548: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2549: #if !defined(PETSC_HAVE_CONSTRAINTS)
2550: 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);
2551: 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);
2552: #endif
2553: MatCheckPreallocated(mat,1);
2555: PetscLogEventBegin(MAT_MultHermitianTranspose,mat,x,y,0);
2556: #if defined(PETSC_USE_COMPLEX)
2557: if (mat->ops->multhermitiantranspose || (mat->hermitian && mat->ops->mult)) {
2558: VecLockReadPush(x);
2559: if (mat->ops->multhermitiantranspose) {
2560: (*mat->ops->multhermitiantranspose)(mat,x,y);
2561: } else {
2562: (*mat->ops->mult)(mat,x,y);
2563: }
2564: VecLockReadPop(x);
2565: } else {
2566: Vec w;
2567: VecDuplicate(x,&w);
2568: VecCopy(x,w);
2569: VecConjugate(w);
2570: MatMultTranspose(mat,w,y);
2571: VecDestroy(&w);
2572: VecConjugate(y);
2573: }
2574: PetscObjectStateIncrease((PetscObject)y);
2575: #else
2576: MatMultTranspose(mat,x,y);
2577: #endif
2578: PetscLogEventEnd(MAT_MultHermitianTranspose,mat,x,y,0);
2579: return(0);
2580: }
2582: /*@
2583: MatMultAdd - Computes v3 = v2 + A * v1.
2585: Neighbor-wise Collective on Mat
2587: Input Parameters:
2588: + mat - the matrix
2589: - v1, v2 - the vectors
2591: Output Parameters:
2592: . v3 - the result
2594: Notes:
2595: The vectors v1 and v3 cannot be the same. I.e., one cannot
2596: call MatMultAdd(A,v1,v2,v1).
2598: Level: beginner
2600: .seealso: MatMultTranspose(), MatMult(), MatMultTransposeAdd()
2601: @*/
2602: PetscErrorCode MatMultAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2603: {
2613: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2614: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2615: if (mat->cmap->N != v1->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->cmap->N,v1->map->N);
2616: /* 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);
2617: 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); */
2618: 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);
2619: 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);
2620: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2621: MatCheckPreallocated(mat,1);
2623: if (!mat->ops->multadd) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"No MatMultAdd() for matrix type %s",((PetscObject)mat)->type_name);
2624: PetscLogEventBegin(MAT_MultAdd,mat,v1,v2,v3);
2625: VecLockReadPush(v1);
2626: (*mat->ops->multadd)(mat,v1,v2,v3);
2627: VecLockReadPop(v1);
2628: PetscLogEventEnd(MAT_MultAdd,mat,v1,v2,v3);
2629: PetscObjectStateIncrease((PetscObject)v3);
2630: return(0);
2631: }
2633: /*@
2634: MatMultTransposeAdd - Computes v3 = v2 + A' * v1.
2636: Neighbor-wise Collective on Mat
2638: Input Parameters:
2639: + mat - the matrix
2640: - v1, v2 - the vectors
2642: Output Parameters:
2643: . v3 - the result
2645: Notes:
2646: The vectors v1 and v3 cannot be the same. I.e., one cannot
2647: call MatMultTransposeAdd(A,v1,v2,v1).
2649: Level: beginner
2651: .seealso: MatMultTranspose(), MatMultAdd(), MatMult()
2652: @*/
2653: PetscErrorCode MatMultTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2654: {
2664: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2665: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2666: if (!mat->ops->multtransposeadd) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2667: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2668: if (mat->rmap->N != v1->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->rmap->N,v1->map->N);
2669: if (mat->cmap->N != v2->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: global dim %D %D",mat->cmap->N,v2->map->N);
2670: if (mat->cmap->N != v3->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: global dim %D %D",mat->cmap->N,v3->map->N);
2671: MatCheckPreallocated(mat,1);
2673: PetscLogEventBegin(MAT_MultTransposeAdd,mat,v1,v2,v3);
2674: VecLockReadPush(v1);
2675: (*mat->ops->multtransposeadd)(mat,v1,v2,v3);
2676: VecLockReadPop(v1);
2677: PetscLogEventEnd(MAT_MultTransposeAdd,mat,v1,v2,v3);
2678: PetscObjectStateIncrease((PetscObject)v3);
2679: return(0);
2680: }
2682: /*@
2683: MatMultHermitianTransposeAdd - Computes v3 = v2 + A^H * v1.
2685: Neighbor-wise Collective on Mat
2687: Input Parameters:
2688: + mat - the matrix
2689: - v1, v2 - the vectors
2691: Output Parameters:
2692: . v3 - the result
2694: Notes:
2695: The vectors v1 and v3 cannot be the same. I.e., one cannot
2696: call MatMultHermitianTransposeAdd(A,v1,v2,v1).
2698: Level: beginner
2700: .seealso: MatMultHermitianTranspose(), MatMultTranspose(), MatMultAdd(), MatMult()
2701: @*/
2702: PetscErrorCode MatMultHermitianTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2703: {
2713: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2714: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2715: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2716: if (mat->rmap->N != v1->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v1: global dim %D %D",mat->rmap->N,v1->map->N);
2717: if (mat->cmap->N != v2->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v2: global dim %D %D",mat->cmap->N,v2->map->N);
2718: if (mat->cmap->N != v3->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec v3: global dim %D %D",mat->cmap->N,v3->map->N);
2719: MatCheckPreallocated(mat,1);
2721: PetscLogEventBegin(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);
2722: VecLockReadPush(v1);
2723: if (mat->ops->multhermitiantransposeadd) {
2724: (*mat->ops->multhermitiantransposeadd)(mat,v1,v2,v3);
2725: } else {
2726: Vec w,z;
2727: VecDuplicate(v1,&w);
2728: VecCopy(v1,w);
2729: VecConjugate(w);
2730: VecDuplicate(v3,&z);
2731: MatMultTranspose(mat,w,z);
2732: VecDestroy(&w);
2733: VecConjugate(z);
2734: if (v2 != v3) {
2735: VecWAXPY(v3,1.0,v2,z);
2736: } else {
2737: VecAXPY(v3,1.0,z);
2738: }
2739: VecDestroy(&z);
2740: }
2741: VecLockReadPop(v1);
2742: PetscLogEventEnd(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);
2743: PetscObjectStateIncrease((PetscObject)v3);
2744: return(0);
2745: }
2747: /*@
2748: MatMultConstrained - The inner multiplication routine for a
2749: constrained matrix P^T A P.
2751: Neighbor-wise Collective on Mat
2753: Input Parameters:
2754: + mat - the matrix
2755: - x - the vector to be multilplied
2757: Output Parameters:
2758: . y - the result
2760: Notes:
2761: The vectors x and y cannot be the same. I.e., one cannot
2762: call MatMult(A,y,y).
2764: Level: beginner
2766: .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2767: @*/
2768: PetscErrorCode MatMultConstrained(Mat mat,Vec x,Vec y)
2769: {
2776: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2777: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2778: if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2779: 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);
2780: 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);
2781: 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);
2783: PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);
2784: VecLockReadPush(x);
2785: (*mat->ops->multconstrained)(mat,x,y);
2786: VecLockReadPop(x);
2787: PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);
2788: PetscObjectStateIncrease((PetscObject)y);
2789: return(0);
2790: }
2792: /*@
2793: MatMultTransposeConstrained - The inner multiplication routine for a
2794: constrained matrix P^T A^T P.
2796: Neighbor-wise Collective on Mat
2798: Input Parameters:
2799: + mat - the matrix
2800: - x - the vector to be multilplied
2802: Output Parameters:
2803: . y - the result
2805: Notes:
2806: The vectors x and y cannot be the same. I.e., one cannot
2807: call MatMult(A,y,y).
2809: Level: beginner
2811: .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2812: @*/
2813: PetscErrorCode MatMultTransposeConstrained(Mat mat,Vec x,Vec y)
2814: {
2821: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2822: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2823: if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2824: 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);
2825: 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);
2827: PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);
2828: (*mat->ops->multtransposeconstrained)(mat,x,y);
2829: PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);
2830: PetscObjectStateIncrease((PetscObject)y);
2831: return(0);
2832: }
2834: /*@C
2835: MatGetFactorType - gets the type of factorization it is
2837: Not Collective
2839: Input Parameters:
2840: . mat - the matrix
2842: Output Parameters:
2843: . t - the type, one of MAT_FACTOR_NONE, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ILU, MAT_FACTOR_ICC,MAT_FACTOR_ILUDT
2845: Level: intermediate
2847: .seealso: MatFactorType, MatGetFactor(), MatSetFactorType()
2848: @*/
2849: PetscErrorCode MatGetFactorType(Mat mat,MatFactorType *t)
2850: {
2855: *t = mat->factortype;
2856: return(0);
2857: }
2859: /*@C
2860: MatSetFactorType - sets the type of factorization it is
2862: Logically Collective on Mat
2864: Input Parameters:
2865: + mat - the matrix
2866: - t - the type, one of MAT_FACTOR_NONE, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ILU, MAT_FACTOR_ICC,MAT_FACTOR_ILUDT
2868: Level: intermediate
2870: .seealso: MatFactorType, MatGetFactor(), MatGetFactorType()
2871: @*/
2872: PetscErrorCode MatSetFactorType(Mat mat, MatFactorType t)
2873: {
2877: mat->factortype = t;
2878: return(0);
2879: }
2881: /* ------------------------------------------------------------*/
2882: /*@C
2883: MatGetInfo - Returns information about matrix storage (number of
2884: nonzeros, memory, etc.).
2886: Collective on Mat if MAT_GLOBAL_MAX or MAT_GLOBAL_SUM is used as the flag
2888: Input Parameters:
2889: . mat - the matrix
2891: Output Parameters:
2892: + flag - flag indicating the type of parameters to be returned
2893: (MAT_LOCAL - local matrix, MAT_GLOBAL_MAX - maximum over all processors,
2894: MAT_GLOBAL_SUM - sum over all processors)
2895: - info - matrix information context
2897: Notes:
2898: The MatInfo context contains a variety of matrix data, including
2899: number of nonzeros allocated and used, number of mallocs during
2900: matrix assembly, etc. Additional information for factored matrices
2901: is provided (such as the fill ratio, number of mallocs during
2902: factorization, etc.). Much of this info is printed to PETSC_STDOUT
2903: when using the runtime options
2904: $ -info -mat_view ::ascii_info
2906: Example for C/C++ Users:
2907: See the file ${PETSC_DIR}/include/petscmat.h for a complete list of
2908: data within the MatInfo context. For example,
2909: .vb
2910: MatInfo info;
2911: Mat A;
2912: double mal, nz_a, nz_u;
2914: MatGetInfo(A,MAT_LOCAL,&info);
2915: mal = info.mallocs;
2916: nz_a = info.nz_allocated;
2917: .ve
2919: Example for Fortran Users:
2920: Fortran users should declare info as a double precision
2921: array of dimension MAT_INFO_SIZE, and then extract the parameters
2922: of interest. See the file ${PETSC_DIR}/include/petsc/finclude/petscmat.h
2923: a complete list of parameter names.
2924: .vb
2925: double precision info(MAT_INFO_SIZE)
2926: double precision mal, nz_a
2927: Mat A
2928: integer ierr
2930: call MatGetInfo(A,MAT_LOCAL,info,ierr)
2931: mal = info(MAT_INFO_MALLOCS)
2932: nz_a = info(MAT_INFO_NZ_ALLOCATED)
2933: .ve
2935: Level: intermediate
2937: Developer Note: fortran interface is not autogenerated as the f90
2938: interface defintion cannot be generated correctly [due to MatInfo]
2940: .seealso: MatStashGetInfo()
2942: @*/
2943: PetscErrorCode MatGetInfo(Mat mat,MatInfoType flag,MatInfo *info)
2944: {
2951: if (!mat->ops->getinfo) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2952: MatCheckPreallocated(mat,1);
2953: (*mat->ops->getinfo)(mat,flag,info);
2954: return(0);
2955: }
2957: /*
2958: This is used by external packages where it is not easy to get the info from the actual
2959: matrix factorization.
2960: */
2961: PetscErrorCode MatGetInfo_External(Mat A,MatInfoType flag,MatInfo *info)
2962: {
2966: PetscMemzero(info,sizeof(MatInfo));
2967: return(0);
2968: }
2970: /* ----------------------------------------------------------*/
2972: /*@C
2973: MatLUFactor - Performs in-place LU factorization of matrix.
2975: Collective on Mat
2977: Input Parameters:
2978: + mat - the matrix
2979: . row - row permutation
2980: . col - column permutation
2981: - info - options for factorization, includes
2982: $ fill - expected fill as ratio of original fill.
2983: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2984: $ Run with the option -info to determine an optimal value to use
2986: Notes:
2987: Most users should employ the simplified KSP interface for linear solvers
2988: instead of working directly with matrix algebra routines such as this.
2989: See, e.g., KSPCreate().
2991: This changes the state of the matrix to a factored matrix; it cannot be used
2992: for example with MatSetValues() unless one first calls MatSetUnfactored().
2994: Level: developer
2996: .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(),
2997: MatGetOrdering(), MatSetUnfactored(), MatFactorInfo, MatGetFactor()
2999: Developer Note: fortran interface is not autogenerated as the f90
3000: interface defintion cannot be generated correctly [due to MatFactorInfo]
3002: @*/
3003: PetscErrorCode MatLUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
3004: {
3006: MatFactorInfo tinfo;
3014: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3015: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3016: if (!mat->ops->lufactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3017: MatCheckPreallocated(mat,1);
3018: if (!info) {
3019: MatFactorInfoInitialize(&tinfo);
3020: info = &tinfo;
3021: }
3023: PetscLogEventBegin(MAT_LUFactor,mat,row,col,0);
3024: (*mat->ops->lufactor)(mat,row,col,info);
3025: PetscLogEventEnd(MAT_LUFactor,mat,row,col,0);
3026: PetscObjectStateIncrease((PetscObject)mat);
3027: return(0);
3028: }
3030: /*@C
3031: MatILUFactor - Performs in-place ILU factorization of matrix.
3033: Collective on Mat
3035: Input Parameters:
3036: + mat - the matrix
3037: . row - row permutation
3038: . col - column permutation
3039: - info - structure containing
3040: $ levels - number of levels of fill.
3041: $ expected fill - as ratio of original fill.
3042: $ 1 or 0 - indicating force fill on diagonal (improves robustness for matrices
3043: missing diagonal entries)
3045: Notes:
3046: Probably really in-place only when level of fill is zero, otherwise allocates
3047: new space to store factored matrix and deletes previous memory.
3049: Most users should employ the simplified KSP interface for linear solvers
3050: instead of working directly with matrix algebra routines such as this.
3051: See, e.g., KSPCreate().
3053: Level: developer
3055: .seealso: MatILUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
3057: Developer Note: fortran interface is not autogenerated as the f90
3058: interface defintion cannot be generated correctly [due to MatFactorInfo]
3060: @*/
3061: PetscErrorCode MatILUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
3062: {
3071: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"matrix must be square");
3072: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3073: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3074: if (!mat->ops->ilufactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3075: MatCheckPreallocated(mat,1);
3077: PetscLogEventBegin(MAT_ILUFactor,mat,row,col,0);
3078: (*mat->ops->ilufactor)(mat,row,col,info);
3079: PetscLogEventEnd(MAT_ILUFactor,mat,row,col,0);
3080: PetscObjectStateIncrease((PetscObject)mat);
3081: return(0);
3082: }
3084: /*@C
3085: MatLUFactorSymbolic - Performs symbolic LU factorization of matrix.
3086: Call this routine before calling MatLUFactorNumeric().
3088: Collective on Mat
3090: Input Parameters:
3091: + fact - the factor matrix obtained with MatGetFactor()
3092: . mat - the matrix
3093: . row, col - row and column permutations
3094: - info - options for factorization, includes
3095: $ fill - expected fill as ratio of original fill.
3096: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
3097: $ Run with the option -info to determine an optimal value to use
3100: Notes:
3101: See Users-Manual: ch_mat for additional information about choosing the fill factor for better efficiency.
3103: Most users should employ the simplified KSP interface for linear solvers
3104: instead of working directly with matrix algebra routines such as this.
3105: See, e.g., KSPCreate().
3107: Level: developer
3109: .seealso: MatLUFactor(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo, MatFactorInfoInitialize()
3111: Developer Note: fortran interface is not autogenerated as the f90
3112: interface defintion cannot be generated correctly [due to MatFactorInfo]
3114: @*/
3115: PetscErrorCode MatLUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
3116: {
3118: MatFactorInfo tinfo;
3127: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3128: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3129: if (!(fact)->ops->lufactorsymbolic) {
3130: MatSolverType stype;
3131: MatFactorGetSolverType(fact,&stype);
3132: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic LU using solver package %s",((PetscObject)mat)->type_name,stype);
3133: }
3134: MatCheckPreallocated(mat,2);
3135: if (!info) {
3136: MatFactorInfoInitialize(&tinfo);
3137: info = &tinfo;
3138: }
3140: PetscLogEventBegin(MAT_LUFactorSymbolic,mat,row,col,0);
3141: (fact->ops->lufactorsymbolic)(fact,mat,row,col,info);
3142: PetscLogEventEnd(MAT_LUFactorSymbolic,mat,row,col,0);
3143: PetscObjectStateIncrease((PetscObject)fact);
3144: return(0);
3145: }
3147: /*@C
3148: MatLUFactorNumeric - Performs numeric LU factorization of a matrix.
3149: Call this routine after first calling MatLUFactorSymbolic().
3151: Collective on Mat
3153: Input Parameters:
3154: + fact - the factor matrix obtained with MatGetFactor()
3155: . mat - the matrix
3156: - info - options for factorization
3158: Notes:
3159: See MatLUFactor() for in-place factorization. See
3160: MatCholeskyFactorNumeric() for the symmetric, positive definite case.
3162: Most users should employ the simplified KSP interface for linear solvers
3163: instead of working directly with matrix algebra routines such as this.
3164: See, e.g., KSPCreate().
3166: Level: developer
3168: .seealso: MatLUFactorSymbolic(), MatLUFactor(), MatCholeskyFactor()
3170: Developer Note: fortran interface is not autogenerated as the f90
3171: interface defintion cannot be generated correctly [due to MatFactorInfo]
3173: @*/
3174: PetscErrorCode MatLUFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
3175: {
3176: MatFactorInfo tinfo;
3184: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3185: if (mat->rmap->N != (fact)->rmap->N || mat->cmap->N != (fact)->cmap->N) SETERRQ4(PetscObjectComm((PetscObject)mat),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);
3187: if (!(fact)->ops->lufactornumeric) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s numeric LU",((PetscObject)mat)->type_name);
3188: MatCheckPreallocated(mat,2);
3189: if (!info) {
3190: MatFactorInfoInitialize(&tinfo);
3191: info = &tinfo;
3192: }
3194: PetscLogEventBegin(MAT_LUFactorNumeric,mat,fact,0,0);
3195: (fact->ops->lufactornumeric)(fact,mat,info);
3196: PetscLogEventEnd(MAT_LUFactorNumeric,mat,fact,0,0);
3197: MatViewFromOptions(fact,NULL,"-mat_factor_view");
3198: PetscObjectStateIncrease((PetscObject)fact);
3199: return(0);
3200: }
3202: /*@C
3203: MatCholeskyFactor - Performs in-place Cholesky factorization of a
3204: symmetric matrix.
3206: Collective on Mat
3208: Input Parameters:
3209: + mat - the matrix
3210: . perm - row and column permutations
3211: - f - expected fill as ratio of original fill
3213: Notes:
3214: See MatLUFactor() for the nonsymmetric case. See also
3215: MatCholeskyFactorSymbolic(), and MatCholeskyFactorNumeric().
3217: Most users should employ the simplified KSP interface for linear solvers
3218: instead of working directly with matrix algebra routines such as this.
3219: See, e.g., KSPCreate().
3221: Level: developer
3223: .seealso: MatLUFactor(), MatCholeskyFactorSymbolic(), MatCholeskyFactorNumeric()
3224: MatGetOrdering()
3226: Developer Note: fortran interface is not autogenerated as the f90
3227: interface defintion cannot be generated correctly [due to MatFactorInfo]
3229: @*/
3230: PetscErrorCode MatCholeskyFactor(Mat mat,IS perm,const MatFactorInfo *info)
3231: {
3233: MatFactorInfo tinfo;
3240: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"Matrix must be square");
3241: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3242: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3243: if (!mat->ops->choleskyfactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"In-place factorization for Mat type %s is not supported, try out-of-place factorization. See MatCholeskyFactorSymbolic/Numeric",((PetscObject)mat)->type_name);
3244: MatCheckPreallocated(mat,1);
3245: if (!info) {
3246: MatFactorInfoInitialize(&tinfo);
3247: info = &tinfo;
3248: }
3250: PetscLogEventBegin(MAT_CholeskyFactor,mat,perm,0,0);
3251: (*mat->ops->choleskyfactor)(mat,perm,info);
3252: PetscLogEventEnd(MAT_CholeskyFactor,mat,perm,0,0);
3253: PetscObjectStateIncrease((PetscObject)mat);
3254: return(0);
3255: }
3257: /*@C
3258: MatCholeskyFactorSymbolic - Performs symbolic Cholesky factorization
3259: of a symmetric matrix.
3261: Collective on Mat
3263: Input Parameters:
3264: + fact - the factor matrix obtained with MatGetFactor()
3265: . mat - the matrix
3266: . perm - row and column permutations
3267: - info - options for factorization, includes
3268: $ fill - expected fill as ratio of original fill.
3269: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
3270: $ Run with the option -info to determine an optimal value to use
3272: Notes:
3273: See MatLUFactorSymbolic() for the nonsymmetric case. See also
3274: MatCholeskyFactor() and MatCholeskyFactorNumeric().
3276: Most users should employ the simplified KSP interface for linear solvers
3277: instead of working directly with matrix algebra routines such as this.
3278: See, e.g., KSPCreate().
3280: Level: developer
3282: .seealso: MatLUFactorSymbolic(), MatCholeskyFactor(), MatCholeskyFactorNumeric()
3283: MatGetOrdering()
3285: Developer Note: fortran interface is not autogenerated as the f90
3286: interface defintion cannot be generated correctly [due to MatFactorInfo]
3288: @*/
3289: PetscErrorCode MatCholeskyFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
3290: {
3292: MatFactorInfo tinfo;
3300: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"Matrix must be square");
3301: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3302: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3303: if (!(fact)->ops->choleskyfactorsymbolic) {
3304: MatSolverType stype;
3305: MatFactorGetSolverType(fact,&stype);
3306: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s symbolic factor Cholesky using solver package %s",((PetscObject)mat)->type_name,stype);
3307: }
3308: MatCheckPreallocated(mat,2);
3309: if (!info) {
3310: MatFactorInfoInitialize(&tinfo);
3311: info = &tinfo;
3312: }
3314: PetscLogEventBegin(MAT_CholeskyFactorSymbolic,mat,perm,0,0);
3315: (fact->ops->choleskyfactorsymbolic)(fact,mat,perm,info);
3316: PetscLogEventEnd(MAT_CholeskyFactorSymbolic,mat,perm,0,0);
3317: PetscObjectStateIncrease((PetscObject)fact);
3318: return(0);
3319: }
3321: /*@C
3322: MatCholeskyFactorNumeric - Performs numeric Cholesky factorization
3323: of a symmetric matrix. Call this routine after first calling
3324: MatCholeskyFactorSymbolic().
3326: Collective on Mat
3328: Input Parameters:
3329: + fact - the factor matrix obtained with MatGetFactor()
3330: . mat - the initial matrix
3331: . info - options for factorization
3332: - fact - the symbolic factor of mat
3335: Notes:
3336: Most users should employ the simplified KSP interface for linear solvers
3337: instead of working directly with matrix algebra routines such as this.
3338: See, e.g., KSPCreate().
3340: Level: developer
3342: .seealso: MatCholeskyFactorSymbolic(), MatCholeskyFactor(), MatLUFactorNumeric()
3344: Developer Note: fortran interface is not autogenerated as the f90
3345: interface defintion cannot be generated correctly [due to MatFactorInfo]
3347: @*/
3348: PetscErrorCode MatCholeskyFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
3349: {
3350: MatFactorInfo tinfo;
3358: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3359: if (!(fact)->ops->choleskyfactornumeric) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s numeric factor Cholesky",((PetscObject)mat)->type_name);
3360: if (mat->rmap->N != (fact)->rmap->N || mat->cmap->N != (fact)->cmap->N) SETERRQ4(PetscObjectComm((PetscObject)mat),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);
3361: MatCheckPreallocated(mat,2);
3362: if (!info) {
3363: MatFactorInfoInitialize(&tinfo);
3364: info = &tinfo;
3365: }
3367: PetscLogEventBegin(MAT_CholeskyFactorNumeric,mat,fact,0,0);
3368: (fact->ops->choleskyfactornumeric)(fact,mat,info);
3369: PetscLogEventEnd(MAT_CholeskyFactorNumeric,mat,fact,0,0);
3370: MatViewFromOptions(fact,NULL,"-mat_factor_view");
3371: PetscObjectStateIncrease((PetscObject)fact);
3372: return(0);
3373: }
3375: /*@C
3376: MatQRFactor - Performs in-place QR factorization of matrix.
3378: Collective on Mat
3380: Input Parameters:
3381: + mat - the matrix
3382: . col - column permutation
3383: - info - options for factorization, includes
3384: $ fill - expected fill as ratio of original fill.
3385: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
3386: $ Run with the option -info to determine an optimal value to use
3388: Notes:
3389: Most users should employ the simplified KSP interface for linear solvers
3390: instead of working directly with matrix algebra routines such as this.
3391: See, e.g., KSPCreate().
3393: This changes the state of the matrix to a factored matrix; it cannot be used
3394: for example with MatSetValues() unless one first calls MatSetUnfactored().
3396: Level: developer
3398: .seealso: MatQRFactorSymbolic(), MatQRFactorNumeric(), MatLUFactor(),
3399: MatSetUnfactored(), MatFactorInfo, MatGetFactor()
3401: Developer Note: fortran interface is not autogenerated as the f90
3402: interface defintion cannot be generated correctly [due to MatFactorInfo]
3404: @*/
3405: PetscErrorCode MatQRFactor(Mat mat, IS col, const MatFactorInfo *info)
3406: {
3414: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3415: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3416: MatCheckPreallocated(mat,1);
3417: PetscLogEventBegin(MAT_QRFactor,mat,col,0,0);
3418: PetscUseMethod(mat,"MatQRFactor_C", (Mat,IS,const MatFactorInfo*), (mat, col, info));
3419: PetscLogEventEnd(MAT_QRFactor,mat,col,0,0);
3420: PetscObjectStateIncrease((PetscObject)mat);
3421: return(0);
3422: }
3424: /*@C
3425: MatQRFactorSymbolic - Performs symbolic QR factorization of matrix.
3426: Call this routine before calling MatQRFactorNumeric().
3428: Collective on Mat
3430: Input Parameters:
3431: + fact - the factor matrix obtained with MatGetFactor()
3432: . mat - the matrix
3433: . col - column permutation
3434: - info - options for factorization, includes
3435: $ fill - expected fill as ratio of original fill.
3436: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
3437: $ Run with the option -info to determine an optimal value to use
3439: Most users should employ the simplified KSP interface for linear solvers
3440: instead of working directly with matrix algebra routines such as this.
3441: See, e.g., KSPCreate().
3443: Level: developer
3445: .seealso: MatQRFactor(), MatQRFactorNumeric(), MatLUFactor(), MatFactorInfo, MatFactorInfoInitialize()
3447: Developer Note: fortran interface is not autogenerated as the f90
3448: interface defintion cannot be generated correctly [due to MatFactorInfo]
3450: @*/
3451: PetscErrorCode MatQRFactorSymbolic(Mat fact,Mat mat,IS col,const MatFactorInfo *info)
3452: {
3454: MatFactorInfo tinfo;
3462: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3463: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3464: MatCheckPreallocated(mat,2);
3465: if (!info) {
3466: MatFactorInfoInitialize(&tinfo);
3467: info = &tinfo;
3468: }
3470: PetscLogEventBegin(MAT_QRFactorSymbolic,fact,mat,col,0);
3471: PetscUseMethod(fact,"MatQRFactorSymbolic_C", (Mat,Mat,IS,const MatFactorInfo*), (fact, mat, col, info));
3472: PetscLogEventEnd(MAT_QRFactorSymbolic,fact,mat,col,0);
3473: PetscObjectStateIncrease((PetscObject)fact);
3474: return(0);
3475: }
3477: /*@C
3478: MatQRFactorNumeric - Performs numeric QR factorization of a matrix.
3479: Call this routine after first calling MatQRFactorSymbolic().
3481: Collective on Mat
3483: Input Parameters:
3484: + fact - the factor matrix obtained with MatGetFactor()
3485: . mat - the matrix
3486: - info - options for factorization
3488: Notes:
3489: See MatQRFactor() for in-place factorization.
3491: Most users should employ the simplified KSP interface for linear solvers
3492: instead of working directly with matrix algebra routines such as this.
3493: See, e.g., KSPCreate().
3495: Level: developer
3497: .seealso: MatQRFactorSymbolic(), MatLUFactor()
3499: Developer Note: fortran interface is not autogenerated as the f90
3500: interface defintion cannot be generated correctly [due to MatFactorInfo]
3502: @*/
3503: PetscErrorCode MatQRFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
3504: {
3505: MatFactorInfo tinfo;
3513: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3514: if (mat->rmap->N != (fact)->rmap->N || mat->cmap->N != (fact)->cmap->N) SETERRQ4(PetscObjectComm((PetscObject)mat),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);
3516: MatCheckPreallocated(mat,2);
3517: if (!info) {
3518: MatFactorInfoInitialize(&tinfo);
3519: info = &tinfo;
3520: }
3522: PetscLogEventBegin(MAT_QRFactorNumeric,mat,fact,0,0);
3523: PetscUseMethod(fact,"MatQRFactorNumeric_C", (Mat,Mat,const MatFactorInfo*), (fact, mat, info));
3524: PetscLogEventEnd(MAT_QRFactorNumeric,mat,fact,0,0);
3525: MatViewFromOptions(fact,NULL,"-mat_factor_view");
3526: PetscObjectStateIncrease((PetscObject)fact);
3527: return(0);
3528: }
3530: /* ----------------------------------------------------------------*/
3531: /*@
3532: MatSolve - Solves A x = b, given a factored matrix.
3534: Neighbor-wise Collective on Mat
3536: Input Parameters:
3537: + mat - the factored matrix
3538: - b - the right-hand-side vector
3540: Output Parameter:
3541: . x - the result vector
3543: Notes:
3544: The vectors b and x cannot be the same. I.e., one cannot
3545: call MatSolve(A,x,x).
3547: Notes:
3548: Most users should employ the simplified KSP interface for linear solvers
3549: instead of working directly with matrix algebra routines such as this.
3550: See, e.g., KSPCreate().
3552: Level: developer
3554: .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd()
3555: @*/
3556: PetscErrorCode MatSolve(Mat mat,Vec b,Vec x)
3557: {
3567: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3568: if (mat->cmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3569: if (mat->rmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3570: 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);
3571: if (!mat->rmap->N && !mat->cmap->N) return(0);
3572: MatCheckPreallocated(mat,1);
3574: PetscLogEventBegin(MAT_Solve,mat,b,x,0);
3575: if (mat->factorerrortype) {
3576: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
3577: VecSetInf(x);
3578: } else {
3579: if (!mat->ops->solve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3580: (*mat->ops->solve)(mat,b,x);
3581: }
3582: PetscLogEventEnd(MAT_Solve,mat,b,x,0);
3583: PetscObjectStateIncrease((PetscObject)x);
3584: return(0);
3585: }
3587: static PetscErrorCode MatMatSolve_Basic(Mat A,Mat B,Mat X,PetscBool trans)
3588: {
3590: Vec b,x;
3591: PetscInt m,N,i;
3592: PetscScalar *bb,*xx;
3593: PetscErrorCode (*f)(Mat,Vec,Vec);
3596: if (A->factorerrortype) {
3597: PetscInfo1(A,"MatFactorError %D\n",A->factorerrortype);
3598: MatSetInf(X);
3599: return(0);
3600: }
3601: f = trans ? A->ops->solvetranspose : A->ops->solve;
3602: if (!f) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
3604: MatDenseGetArrayRead(B,(const PetscScalar**)&bb);
3605: MatDenseGetArray(X,&xx);
3606: MatGetLocalSize(B,&m,NULL); /* number local rows */
3607: MatGetSize(B,NULL,&N); /* total columns in dense matrix */
3608: MatCreateVecs(A,&x,&b);
3609: for (i=0; i<N; i++) {
3610: VecPlaceArray(b,bb + i*m);
3611: VecPlaceArray(x,xx + i*m);
3612: (*f)(A,b,x);
3613: VecResetArray(x);
3614: VecResetArray(b);
3615: }
3616: VecDestroy(&b);
3617: VecDestroy(&x);
3618: MatDenseRestoreArrayRead(B,(const PetscScalar**)&bb);
3619: MatDenseRestoreArray(X,&xx);
3620: return(0);
3621: }
3623: /*@
3624: MatMatSolve - Solves A X = B, given a factored matrix.
3626: Neighbor-wise Collective on Mat
3628: Input Parameters:
3629: + A - the factored matrix
3630: - B - the right-hand-side matrix MATDENSE (or sparse -- when using MUMPS)
3632: Output Parameter:
3633: . X - the result matrix (dense matrix)
3635: Notes:
3636: If B is a MATDENSE matrix then one can call MatMatSolve(A,B,B) except with MKL_CPARDISO;
3637: otherwise, B and X cannot be the same.
3639: Notes:
3640: Most users should usually employ the simplified KSP interface for linear solvers
3641: instead of working directly with matrix algebra routines such as this.
3642: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3643: at a time.
3645: Level: developer
3647: .seealso: MatMatSolveTranspose(), MatLUFactor(), MatCholeskyFactor()
3648: @*/
3649: PetscErrorCode MatMatSolve(Mat A,Mat B,Mat X)
3650: {
3660: if (A->cmap->N != X->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat X: global dim %D %D",A->cmap->N,X->rmap->N);
3661: if (A->rmap->N != B->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat B: global dim %D %D",A->rmap->N,B->rmap->N);
3662: if (X->cmap->N != B->cmap->N) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Solution matrix must have same number of columns as rhs matrix");
3663: if (!A->rmap->N && !A->cmap->N) return(0);
3664: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3665: MatCheckPreallocated(A,1);
3667: PetscLogEventBegin(MAT_MatSolve,A,B,X,0);
3668: if (!A->ops->matsolve) {
3669: PetscInfo1(A,"Mat type %s using basic MatMatSolve\n",((PetscObject)A)->type_name);
3670: MatMatSolve_Basic(A,B,X,PETSC_FALSE);
3671: } else {
3672: (*A->ops->matsolve)(A,B,X);
3673: }
3674: PetscLogEventEnd(MAT_MatSolve,A,B,X,0);
3675: PetscObjectStateIncrease((PetscObject)X);
3676: return(0);
3677: }
3679: /*@
3680: MatMatSolveTranspose - Solves A^T X = B, given a factored matrix.
3682: Neighbor-wise Collective on Mat
3684: Input Parameters:
3685: + A - the factored matrix
3686: - B - the right-hand-side matrix (dense matrix)
3688: Output Parameter:
3689: . X - the result matrix (dense matrix)
3691: Notes:
3692: The matrices B and X cannot be the same. I.e., one cannot
3693: call MatMatSolveTranspose(A,X,X).
3695: Notes:
3696: Most users should usually employ the simplified KSP interface for linear solvers
3697: instead of working directly with matrix algebra routines such as this.
3698: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3699: at a time.
3701: When using SuperLU_Dist or MUMPS as a parallel solver, PETSc will use their functionality to solve multiple right hand sides simultaneously.
3703: Level: developer
3705: .seealso: MatMatSolve(), MatLUFactor(), MatCholeskyFactor()
3706: @*/
3707: PetscErrorCode MatMatSolveTranspose(Mat A,Mat B,Mat X)
3708: {
3718: if (X == B) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_IDN,"X and B must be different matrices");
3719: if (A->cmap->N != X->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat X: global dim %D %D",A->cmap->N,X->rmap->N);
3720: if (A->rmap->N != B->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat B: global dim %D %D",A->rmap->N,B->rmap->N);
3721: 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);
3722: if (X->cmap->N < B->cmap->N) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Solution matrix must have same number of columns as rhs matrix");
3723: if (!A->rmap->N && !A->cmap->N) return(0);
3724: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3725: MatCheckPreallocated(A,1);
3727: PetscLogEventBegin(MAT_MatSolve,A,B,X,0);
3728: if (!A->ops->matsolvetranspose) {
3729: PetscInfo1(A,"Mat type %s using basic MatMatSolveTranspose\n",((PetscObject)A)->type_name);
3730: MatMatSolve_Basic(A,B,X,PETSC_TRUE);
3731: } else {
3732: (*A->ops->matsolvetranspose)(A,B,X);
3733: }
3734: PetscLogEventEnd(MAT_MatSolve,A,B,X,0);
3735: PetscObjectStateIncrease((PetscObject)X);
3736: return(0);
3737: }
3739: /*@
3740: MatMatTransposeSolve - Solves A X = B^T, given a factored matrix.
3742: Neighbor-wise Collective on Mat
3744: Input Parameters:
3745: + A - the factored matrix
3746: - Bt - the transpose of right-hand-side matrix
3748: Output Parameter:
3749: . X - the result matrix (dense matrix)
3751: Notes:
3752: Most users should usually employ the simplified KSP interface for linear solvers
3753: instead of working directly with matrix algebra routines such as this.
3754: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3755: at a time.
3757: For MUMPS, it only supports centralized sparse compressed column format on the host processor for right hand side matrix. User must create B^T in sparse compressed row format on the host processor and call MatMatTransposeSolve() to implement MUMPS' MatMatSolve().
3759: Level: developer
3761: .seealso: MatMatSolve(), MatMatSolveTranspose(), MatLUFactor(), MatCholeskyFactor()
3762: @*/
3763: PetscErrorCode MatMatTransposeSolve(Mat A,Mat Bt,Mat X)
3764: {
3775: if (X == Bt) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_IDN,"X and B must be different matrices");
3776: if (A->cmap->N != X->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat X: global dim %D %D",A->cmap->N,X->rmap->N);
3777: if (A->rmap->N != Bt->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Mat A,Mat Bt: global dim %D %D",A->rmap->N,Bt->cmap->N);
3778: if (X->cmap->N < Bt->rmap->N) SETERRQ(PetscObjectComm((PetscObject)X),PETSC_ERR_ARG_SIZ,"Solution matrix must have same number of columns as row number of the rhs matrix");
3779: if (!A->rmap->N && !A->cmap->N) return(0);
3780: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3781: MatCheckPreallocated(A,1);
3783: if (!A->ops->mattransposesolve) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
3784: PetscLogEventBegin(MAT_MatTrSolve,A,Bt,X,0);
3785: (*A->ops->mattransposesolve)(A,Bt,X);
3786: PetscLogEventEnd(MAT_MatTrSolve,A,Bt,X,0);
3787: PetscObjectStateIncrease((PetscObject)X);
3788: return(0);
3789: }
3791: /*@
3792: MatForwardSolve - Solves L x = b, given a factored matrix, A = LU, or
3793: U^T*D^(1/2) x = b, given a factored symmetric matrix, A = U^T*D*U,
3795: Neighbor-wise Collective on Mat
3797: Input Parameters:
3798: + mat - the factored matrix
3799: - b - the right-hand-side vector
3801: Output Parameter:
3802: . x - the result vector
3804: Notes:
3805: MatSolve() should be used for most applications, as it performs
3806: a forward solve followed by a backward solve.
3808: The vectors b and x cannot be the same, i.e., one cannot
3809: call MatForwardSolve(A,x,x).
3811: For matrix in seqsbaij format with block size larger than 1,
3812: the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3813: MatForwardSolve() solves U^T*D y = b, and
3814: MatBackwardSolve() solves U x = y.
3815: Thus they do not provide a symmetric preconditioner.
3817: Most users should employ the simplified KSP interface for linear solvers
3818: instead of working directly with matrix algebra routines such as this.
3819: See, e.g., KSPCreate().
3821: Level: developer
3823: .seealso: MatSolve(), MatBackwardSolve()
3824: @*/
3825: PetscErrorCode MatForwardSolve(Mat mat,Vec b,Vec x)
3826: {
3836: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3837: if (mat->cmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3838: if (mat->rmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3839: 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);
3840: if (!mat->rmap->N && !mat->cmap->N) return(0);
3841: MatCheckPreallocated(mat,1);
3843: if (!mat->ops->forwardsolve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3844: PetscLogEventBegin(MAT_ForwardSolve,mat,b,x,0);
3845: (*mat->ops->forwardsolve)(mat,b,x);
3846: PetscLogEventEnd(MAT_ForwardSolve,mat,b,x,0);
3847: PetscObjectStateIncrease((PetscObject)x);
3848: return(0);
3849: }
3851: /*@
3852: MatBackwardSolve - Solves U x = b, given a factored matrix, A = LU.
3853: D^(1/2) U x = b, given a factored symmetric matrix, A = U^T*D*U,
3855: Neighbor-wise Collective on Mat
3857: Input Parameters:
3858: + mat - the factored matrix
3859: - b - the right-hand-side vector
3861: Output Parameter:
3862: . x - the result vector
3864: Notes:
3865: MatSolve() should be used for most applications, as it performs
3866: a forward solve followed by a backward solve.
3868: The vectors b and x cannot be the same. I.e., one cannot
3869: call MatBackwardSolve(A,x,x).
3871: For matrix in seqsbaij format with block size larger than 1,
3872: the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3873: MatForwardSolve() solves U^T*D y = b, and
3874: MatBackwardSolve() solves U x = y.
3875: Thus they do not provide a symmetric preconditioner.
3877: Most users should employ the simplified KSP interface for linear solvers
3878: instead of working directly with matrix algebra routines such as this.
3879: See, e.g., KSPCreate().
3881: Level: developer
3883: .seealso: MatSolve(), MatForwardSolve()
3884: @*/
3885: PetscErrorCode MatBackwardSolve(Mat mat,Vec b,Vec x)
3886: {
3896: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3897: if (mat->cmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3898: if (mat->rmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3899: 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);
3900: if (!mat->rmap->N && !mat->cmap->N) return(0);
3901: MatCheckPreallocated(mat,1);
3903: if (!mat->ops->backwardsolve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3904: PetscLogEventBegin(MAT_BackwardSolve,mat,b,x,0);
3905: (*mat->ops->backwardsolve)(mat,b,x);
3906: PetscLogEventEnd(MAT_BackwardSolve,mat,b,x,0);
3907: PetscObjectStateIncrease((PetscObject)x);
3908: return(0);
3909: }
3911: /*@
3912: MatSolveAdd - Computes x = y + inv(A)*b, given a factored matrix.
3914: Neighbor-wise Collective on Mat
3916: Input Parameters:
3917: + mat - the factored matrix
3918: . b - the right-hand-side vector
3919: - y - the vector to be added to
3921: Output Parameter:
3922: . x - the result vector
3924: Notes:
3925: The vectors b and x cannot be the same. I.e., one cannot
3926: call MatSolveAdd(A,x,y,x).
3928: Most users should employ the simplified KSP interface for linear solvers
3929: instead of working directly with matrix algebra routines such as this.
3930: See, e.g., KSPCreate().
3932: Level: developer
3934: .seealso: MatSolve(), MatSolveTranspose(), MatSolveTransposeAdd()
3935: @*/
3936: PetscErrorCode MatSolveAdd(Mat mat,Vec b,Vec y,Vec x)
3937: {
3938: PetscScalar one = 1.0;
3939: Vec tmp;
3951: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3952: if (mat->cmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
3953: if (mat->rmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
3954: if (mat->rmap->N != y->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->rmap->N,y->map->N);
3955: 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);
3956: 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);
3957: if (!mat->rmap->N && !mat->cmap->N) return(0);
3958: MatCheckPreallocated(mat,1);
3960: PetscLogEventBegin(MAT_SolveAdd,mat,b,x,y);
3961: if (mat->factorerrortype) {
3962: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
3963: VecSetInf(x);
3964: } else if (mat->ops->solveadd) {
3965: (*mat->ops->solveadd)(mat,b,y,x);
3966: } else {
3967: /* do the solve then the add manually */
3968: if (x != y) {
3969: MatSolve(mat,b,x);
3970: VecAXPY(x,one,y);
3971: } else {
3972: VecDuplicate(x,&tmp);
3973: PetscLogObjectParent((PetscObject)mat,(PetscObject)tmp);
3974: VecCopy(x,tmp);
3975: MatSolve(mat,b,x);
3976: VecAXPY(x,one,tmp);
3977: VecDestroy(&tmp);
3978: }
3979: }
3980: PetscLogEventEnd(MAT_SolveAdd,mat,b,x,y);
3981: PetscObjectStateIncrease((PetscObject)x);
3982: return(0);
3983: }
3985: /*@
3986: MatSolveTranspose - Solves A' x = b, given a factored matrix.
3988: Neighbor-wise Collective on Mat
3990: Input Parameters:
3991: + mat - the factored matrix
3992: - b - the right-hand-side vector
3994: Output Parameter:
3995: . x - the result vector
3997: Notes:
3998: The vectors b and x cannot be the same. I.e., one cannot
3999: call MatSolveTranspose(A,x,x).
4001: Most users should employ the simplified KSP interface for linear solvers
4002: instead of working directly with matrix algebra routines such as this.
4003: See, e.g., KSPCreate().
4005: Level: developer
4007: .seealso: MatSolve(), MatSolveAdd(), MatSolveTransposeAdd()
4008: @*/
4009: PetscErrorCode MatSolveTranspose(Mat mat,Vec b,Vec x)
4010: {
4020: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
4021: if (mat->rmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
4022: if (mat->cmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->cmap->N,b->map->N);
4023: if (!mat->rmap->N && !mat->cmap->N) return(0);
4024: MatCheckPreallocated(mat,1);
4025: PetscLogEventBegin(MAT_SolveTranspose,mat,b,x,0);
4026: if (mat->factorerrortype) {
4027: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
4028: VecSetInf(x);
4029: } else {
4030: if (!mat->ops->solvetranspose) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s",((PetscObject)mat)->type_name);
4031: (*mat->ops->solvetranspose)(mat,b,x);
4032: }
4033: PetscLogEventEnd(MAT_SolveTranspose,mat,b,x,0);
4034: PetscObjectStateIncrease((PetscObject)x);
4035: return(0);
4036: }
4038: /*@
4039: MatSolveTransposeAdd - Computes x = y + inv(Transpose(A)) b, given a
4040: factored matrix.
4042: Neighbor-wise Collective on Mat
4044: Input Parameters:
4045: + mat - the factored matrix
4046: . b - the right-hand-side vector
4047: - y - the vector to be added to
4049: Output Parameter:
4050: . x - the result vector
4052: Notes:
4053: The vectors b and x cannot be the same. I.e., one cannot
4054: call MatSolveTransposeAdd(A,x,y,x).
4056: Most users should employ the simplified KSP interface for linear solvers
4057: instead of working directly with matrix algebra routines such as this.
4058: See, e.g., KSPCreate().
4060: Level: developer
4062: .seealso: MatSolve(), MatSolveAdd(), MatSolveTranspose()
4063: @*/
4064: PetscErrorCode MatSolveTransposeAdd(Mat mat,Vec b,Vec y,Vec x)
4065: {
4066: PetscScalar one = 1.0;
4068: Vec tmp;
4079: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
4080: if (mat->rmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->rmap->N,x->map->N);
4081: if (mat->cmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->cmap->N,b->map->N);
4082: if (mat->cmap->N != y->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec y: global dim %D %D",mat->cmap->N,y->map->N);
4083: 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);
4084: if (!mat->rmap->N && !mat->cmap->N) return(0);
4085: MatCheckPreallocated(mat,1);
4087: PetscLogEventBegin(MAT_SolveTransposeAdd,mat,b,x,y);
4088: if (mat->factorerrortype) {
4089: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
4090: VecSetInf(x);
4091: } else if (mat->ops->solvetransposeadd){
4092: (*mat->ops->solvetransposeadd)(mat,b,y,x);
4093: } else {
4094: /* do the solve then the add manually */
4095: if (x != y) {
4096: MatSolveTranspose(mat,b,x);
4097: VecAXPY(x,one,y);
4098: } else {
4099: VecDuplicate(x,&tmp);
4100: PetscLogObjectParent((PetscObject)mat,(PetscObject)tmp);
4101: VecCopy(x,tmp);
4102: MatSolveTranspose(mat,b,x);
4103: VecAXPY(x,one,tmp);
4104: VecDestroy(&tmp);
4105: }
4106: }
4107: PetscLogEventEnd(MAT_SolveTransposeAdd,mat,b,x,y);
4108: PetscObjectStateIncrease((PetscObject)x);
4109: return(0);
4110: }
4111: /* ----------------------------------------------------------------*/
4113: /*@
4114: MatSOR - Computes relaxation (SOR, Gauss-Seidel) sweeps.
4116: Neighbor-wise Collective on Mat
4118: Input Parameters:
4119: + mat - the matrix
4120: . b - the right hand side
4121: . omega - the relaxation factor
4122: . flag - flag indicating the type of SOR (see below)
4123: . shift - diagonal shift
4124: . its - the number of iterations
4125: - lits - the number of local iterations
4127: Output Parameters:
4128: . x - the solution (can contain an initial guess, use option SOR_ZERO_INITIAL_GUESS to indicate no guess)
4130: SOR Flags:
4131: + SOR_FORWARD_SWEEP - forward SOR
4132: . SOR_BACKWARD_SWEEP - backward SOR
4133: . SOR_SYMMETRIC_SWEEP - SSOR (symmetric SOR)
4134: . SOR_LOCAL_FORWARD_SWEEP - local forward SOR
4135: . SOR_LOCAL_BACKWARD_SWEEP - local forward SOR
4136: . SOR_LOCAL_SYMMETRIC_SWEEP - local SSOR
4137: . SOR_APPLY_UPPER, SOR_APPLY_LOWER - applies
4138: upper/lower triangular part of matrix to
4139: vector (with omega)
4140: - SOR_ZERO_INITIAL_GUESS - zero initial guess
4142: Notes:
4143: SOR_LOCAL_FORWARD_SWEEP, SOR_LOCAL_BACKWARD_SWEEP, and
4144: SOR_LOCAL_SYMMETRIC_SWEEP perform separate independent smoothings
4145: on each processor.
4147: Application programmers will not generally use MatSOR() directly,
4148: but instead will employ the KSP/PC interface.
4150: Notes:
4151: for BAIJ, SBAIJ, and AIJ matrices with Inodes this does a block SOR smoothing, otherwise it does a pointwise smoothing
4153: Notes for Advanced Users:
4154: The flags are implemented as bitwise inclusive or operations.
4155: For example, use (SOR_ZERO_INITIAL_GUESS | SOR_SYMMETRIC_SWEEP)
4156: to specify a zero initial guess for SSOR.
4158: Most users should employ the simplified KSP interface for linear solvers
4159: instead of working directly with matrix algebra routines such as this.
4160: See, e.g., KSPCreate().
4162: Vectors x and b CANNOT be the same
4164: Developer Note: We should add block SOR support for AIJ matrices with block size set to great than one and no inodes
4166: Level: developer
4168: @*/
4169: PetscErrorCode MatSOR(Mat mat,Vec b,PetscReal omega,MatSORType flag,PetscReal shift,PetscInt its,PetscInt lits,Vec x)
4170: {
4180: if (!mat->ops->sor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4181: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4182: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4183: if (mat->cmap->N != x->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec x: global dim %D %D",mat->cmap->N,x->map->N);
4184: if (mat->rmap->N != b->map->N) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_SIZ,"Mat mat,Vec b: global dim %D %D",mat->rmap->N,b->map->N);
4185: 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);
4186: if (its <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D positive",its);
4187: if (lits <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires local its %D positive",lits);
4188: if (b == x) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_IDN,"b and x vector cannot be the same");
4190: MatCheckPreallocated(mat,1);
4191: PetscLogEventBegin(MAT_SOR,mat,b,x,0);
4192: ierr =(*mat->ops->sor)(mat,b,omega,flag,shift,its,lits,x);
4193: PetscLogEventEnd(MAT_SOR,mat,b,x,0);
4194: PetscObjectStateIncrease((PetscObject)x);
4195: return(0);
4196: }
4198: /*
4199: Default matrix copy routine.
4200: */
4201: PetscErrorCode MatCopy_Basic(Mat A,Mat B,MatStructure str)
4202: {
4203: PetscErrorCode ierr;
4204: PetscInt i,rstart = 0,rend = 0,nz;
4205: const PetscInt *cwork;
4206: const PetscScalar *vwork;
4209: if (B->assembled) {
4210: MatZeroEntries(B);
4211: }
4212: if (str == SAME_NONZERO_PATTERN) {
4213: MatGetOwnershipRange(A,&rstart,&rend);
4214: for (i=rstart; i<rend; i++) {
4215: MatGetRow(A,i,&nz,&cwork,&vwork);
4216: MatSetValues(B,1,&i,nz,cwork,vwork,INSERT_VALUES);
4217: MatRestoreRow(A,i,&nz,&cwork,&vwork);
4218: }
4219: } else {
4220: MatAYPX(B,0.0,A,str);
4221: }
4222: MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);
4223: MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);
4224: return(0);
4225: }
4227: /*@
4228: MatCopy - Copies a matrix to another matrix.
4230: Collective on Mat
4232: Input Parameters:
4233: + A - the matrix
4234: - str - SAME_NONZERO_PATTERN or DIFFERENT_NONZERO_PATTERN
4236: Output Parameter:
4237: . B - where the copy is put
4239: Notes:
4240: If you use SAME_NONZERO_PATTERN then the two matrices had better have the
4241: same nonzero pattern or the routine will crash.
4243: MatCopy() copies the matrix entries of a matrix to another existing
4244: matrix (after first zeroing the second matrix). A related routine is
4245: MatConvert(), which first creates a new matrix and then copies the data.
4247: Level: intermediate
4249: .seealso: MatConvert(), MatDuplicate()
4251: @*/
4252: PetscErrorCode MatCopy(Mat A,Mat B,MatStructure str)
4253: {
4255: PetscInt i;
4263: MatCheckPreallocated(B,2);
4264: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4265: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4266: if (A->rmap->N != B->rmap->N || A->cmap->N != B->cmap->N) SETERRQ4(PetscObjectComm((PetscObject)A),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);
4267: MatCheckPreallocated(A,1);
4268: if (A == B) return(0);
4270: PetscLogEventBegin(MAT_Copy,A,B,0,0);
4271: if (A->ops->copy) {
4272: (*A->ops->copy)(A,B,str);
4273: } else { /* generic conversion */
4274: MatCopy_Basic(A,B,str);
4275: }
4277: B->stencil.dim = A->stencil.dim;
4278: B->stencil.noc = A->stencil.noc;
4279: for (i=0; i<=A->stencil.dim; i++) {
4280: B->stencil.dims[i] = A->stencil.dims[i];
4281: B->stencil.starts[i] = A->stencil.starts[i];
4282: }
4284: PetscLogEventEnd(MAT_Copy,A,B,0,0);
4285: PetscObjectStateIncrease((PetscObject)B);
4286: return(0);
4287: }
4289: /*@C
4290: MatConvert - Converts a matrix to another matrix, either of the same
4291: or different type.
4293: Collective on Mat
4295: Input Parameters:
4296: + mat - the matrix
4297: . newtype - new matrix type. Use MATSAME to create a new matrix of the
4298: same type as the original matrix.
4299: - reuse - denotes if the destination matrix is to be created or reused.
4300: Use MAT_INPLACE_MATRIX for inplace conversion (that is when you want the input mat to be changed to contain the matrix in the new format), otherwise use
4301: MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX (can only be used after the first call was made with MAT_INITIAL_MATRIX, causes the matrix space in M to be reused).
4303: Output Parameter:
4304: . M - pointer to place new matrix
4306: Notes:
4307: MatConvert() first creates a new matrix and then copies the data from
4308: the first matrix. A related routine is MatCopy(), which copies the matrix
4309: entries of one matrix to another already existing matrix context.
4311: Cannot be used to convert a sequential matrix to parallel or parallel to sequential,
4312: the MPI communicator of the generated matrix is always the same as the communicator
4313: of the input matrix.
4315: Level: intermediate
4317: .seealso: MatCopy(), MatDuplicate()
4318: @*/
4319: PetscErrorCode MatConvert(Mat mat, MatType newtype,MatReuse reuse,Mat *M)
4320: {
4322: PetscBool sametype,issame,flg,issymmetric,ishermitian;
4323: char convname[256],mtype[256];
4324: Mat B;
4330: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4331: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4332: MatCheckPreallocated(mat,1);
4334: PetscOptionsGetString(((PetscObject)mat)->options,((PetscObject)mat)->prefix,"-matconvert_type",mtype,sizeof(mtype),&flg);
4335: if (flg) newtype = mtype;
4337: PetscObjectTypeCompare((PetscObject)mat,newtype,&sametype);
4338: PetscStrcmp(newtype,"same",&issame);
4339: if ((reuse == MAT_INPLACE_MATRIX) && (mat != *M)) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"MAT_INPLACE_MATRIX requires same input and output matrix");
4340: if ((reuse == MAT_REUSE_MATRIX) && (mat == *M)) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"MAT_REUSE_MATRIX means reuse matrix in final argument, perhaps you mean MAT_INPLACE_MATRIX");
4342: if ((reuse == MAT_INPLACE_MATRIX) && (issame || sametype)) {
4343: PetscInfo3(mat,"Early return for inplace %s %d %d\n",((PetscObject)mat)->type_name,sametype,issame);
4344: return(0);
4345: }
4347: /* Cache Mat options because some converter use MatHeaderReplace */
4348: issymmetric = mat->symmetric;
4349: ishermitian = mat->hermitian;
4351: if ((sametype || issame) && (reuse==MAT_INITIAL_MATRIX) && mat->ops->duplicate) {
4352: PetscInfo3(mat,"Calling duplicate for initial matrix %s %d %d\n",((PetscObject)mat)->type_name,sametype,issame);
4353: (*mat->ops->duplicate)(mat,MAT_COPY_VALUES,M);
4354: } else {
4355: PetscErrorCode (*conv)(Mat, MatType,MatReuse,Mat*)=NULL;
4356: const char *prefix[3] = {"seq","mpi",""};
4357: PetscInt i;
4358: /*
4359: Order of precedence:
4360: 0) See if newtype is a superclass of the current matrix.
4361: 1) See if a specialized converter is known to the current matrix.
4362: 2) See if a specialized converter is known to the desired matrix class.
4363: 3) See if a good general converter is registered for the desired class
4364: (as of 6/27/03 only MATMPIADJ falls into this category).
4365: 4) See if a good general converter is known for the current matrix.
4366: 5) Use a really basic converter.
4367: */
4369: /* 0) See if newtype is a superclass of the current matrix.
4370: i.e mat is mpiaij and newtype is aij */
4371: for (i=0; i<2; i++) {
4372: PetscStrncpy(convname,prefix[i],sizeof(convname));
4373: PetscStrlcat(convname,newtype,sizeof(convname));
4374: PetscStrcmp(convname,((PetscObject)mat)->type_name,&flg);
4375: PetscInfo3(mat,"Check superclass %s %s -> %d\n",convname,((PetscObject)mat)->type_name,flg);
4376: if (flg) {
4377: if (reuse == MAT_INPLACE_MATRIX) {
4378: PetscInfo(mat,"Early return\n");
4379: return(0);
4380: } else if (reuse == MAT_INITIAL_MATRIX && mat->ops->duplicate) {
4381: PetscInfo(mat,"Calling MatDuplicate\n");
4382: (*mat->ops->duplicate)(mat,MAT_COPY_VALUES,M);
4383: return(0);
4384: } else if (reuse == MAT_REUSE_MATRIX && mat->ops->copy) {
4385: PetscInfo(mat,"Calling MatCopy\n");
4386: MatCopy(mat,*M,SAME_NONZERO_PATTERN);
4387: return(0);
4388: }
4389: }
4390: }
4391: /* 1) See if a specialized converter is known to the current matrix and the desired class */
4392: for (i=0; i<3; i++) {
4393: PetscStrncpy(convname,"MatConvert_",sizeof(convname));
4394: PetscStrlcat(convname,((PetscObject)mat)->type_name,sizeof(convname));
4395: PetscStrlcat(convname,"_",sizeof(convname));
4396: PetscStrlcat(convname,prefix[i],sizeof(convname));
4397: PetscStrlcat(convname,issame ? ((PetscObject)mat)->type_name : newtype,sizeof(convname));
4398: PetscStrlcat(convname,"_C",sizeof(convname));
4399: PetscObjectQueryFunction((PetscObject)mat,convname,&conv);
4400: PetscInfo3(mat,"Check specialized (1) %s (%s) -> %d\n",convname,((PetscObject)mat)->type_name,!!conv);
4401: if (conv) goto foundconv;
4402: }
4404: /* 2) See if a specialized converter is known to the desired matrix class. */
4405: MatCreate(PetscObjectComm((PetscObject)mat),&B);
4406: MatSetSizes(B,mat->rmap->n,mat->cmap->n,mat->rmap->N,mat->cmap->N);
4407: MatSetType(B,newtype);
4408: for (i=0; i<3; i++) {
4409: PetscStrncpy(convname,"MatConvert_",sizeof(convname));
4410: PetscStrlcat(convname,((PetscObject)mat)->type_name,sizeof(convname));
4411: PetscStrlcat(convname,"_",sizeof(convname));
4412: PetscStrlcat(convname,prefix[i],sizeof(convname));
4413: PetscStrlcat(convname,newtype,sizeof(convname));
4414: PetscStrlcat(convname,"_C",sizeof(convname));
4415: PetscObjectQueryFunction((PetscObject)B,convname,&conv);
4416: PetscInfo3(mat,"Check specialized (2) %s (%s) -> %d\n",convname,((PetscObject)B)->type_name,!!conv);
4417: if (conv) {
4418: MatDestroy(&B);
4419: goto foundconv;
4420: }
4421: }
4423: /* 3) See if a good general converter is registered for the desired class */
4424: conv = B->ops->convertfrom;
4425: PetscInfo2(mat,"Check convertfrom (%s) -> %d\n",((PetscObject)B)->type_name,!!conv);
4426: MatDestroy(&B);
4427: if (conv) goto foundconv;
4429: /* 4) See if a good general converter is known for the current matrix */
4430: if (mat->ops->convert) conv = mat->ops->convert;
4432: PetscInfo2(mat,"Check general convert (%s) -> %d\n",((PetscObject)mat)->type_name,!!conv);
4433: if (conv) goto foundconv;
4435: /* 5) Use a really basic converter. */
4436: PetscInfo(mat,"Using MatConvert_Basic\n");
4437: conv = MatConvert_Basic;
4439: foundconv:
4440: PetscLogEventBegin(MAT_Convert,mat,0,0,0);
4441: (*conv)(mat,newtype,reuse,M);
4442: if (mat->rmap->mapping && mat->cmap->mapping && !(*M)->rmap->mapping && !(*M)->cmap->mapping) {
4443: /* the block sizes must be same if the mappings are copied over */
4444: (*M)->rmap->bs = mat->rmap->bs;
4445: (*M)->cmap->bs = mat->cmap->bs;
4446: PetscObjectReference((PetscObject)mat->rmap->mapping);
4447: PetscObjectReference((PetscObject)mat->cmap->mapping);
4448: (*M)->rmap->mapping = mat->rmap->mapping;
4449: (*M)->cmap->mapping = mat->cmap->mapping;
4450: }
4451: (*M)->stencil.dim = mat->stencil.dim;
4452: (*M)->stencil.noc = mat->stencil.noc;
4453: for (i=0; i<=mat->stencil.dim; i++) {
4454: (*M)->stencil.dims[i] = mat->stencil.dims[i];
4455: (*M)->stencil.starts[i] = mat->stencil.starts[i];
4456: }
4457: PetscLogEventEnd(MAT_Convert,mat,0,0,0);
4458: }
4459: PetscObjectStateIncrease((PetscObject)*M);
4461: /* Copy Mat options */
4462: if (issymmetric) {
4463: MatSetOption(*M,MAT_SYMMETRIC,PETSC_TRUE);
4464: }
4465: if (ishermitian) {
4466: MatSetOption(*M,MAT_HERMITIAN,PETSC_TRUE);
4467: }
4468: return(0);
4469: }
4471: /*@C
4472: MatFactorGetSolverType - Returns name of the package providing the factorization routines
4474: Not Collective
4476: Input Parameter:
4477: . mat - the matrix, must be a factored matrix
4479: Output Parameter:
4480: . type - the string name of the package (do not free this string)
4482: Notes:
4483: In Fortran you pass in a empty string and the package name will be copied into it.
4484: (Make sure the string is long enough)
4486: Level: intermediate
4488: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatGetFactor()
4489: @*/
4490: PetscErrorCode MatFactorGetSolverType(Mat mat, MatSolverType *type)
4491: {
4492: PetscErrorCode ierr, (*conv)(Mat,MatSolverType*);
4497: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Only for factored matrix");
4498: PetscObjectQueryFunction((PetscObject)mat,"MatFactorGetSolverType_C",&conv);
4499: if (!conv) {
4500: *type = MATSOLVERPETSC;
4501: } else {
4502: (*conv)(mat,type);
4503: }
4504: return(0);
4505: }
4507: typedef struct _MatSolverTypeForSpecifcType* MatSolverTypeForSpecifcType;
4508: struct _MatSolverTypeForSpecifcType {
4509: MatType mtype;
4510: PetscErrorCode (*createfactor[MAT_FACTOR_NUM_TYPES])(Mat,MatFactorType,Mat*);
4511: MatSolverTypeForSpecifcType next;
4512: };
4514: typedef struct _MatSolverTypeHolder* MatSolverTypeHolder;
4515: struct _MatSolverTypeHolder {
4516: char *name;
4517: MatSolverTypeForSpecifcType handlers;
4518: MatSolverTypeHolder next;
4519: };
4521: static MatSolverTypeHolder MatSolverTypeHolders = NULL;
4523: /*@C
4524: MatSolverTypeRegister - Registers a MatSolverType that works for a particular matrix type
4526: Input Parameters:
4527: + package - name of the package, for example petsc or superlu
4528: . mtype - the matrix type that works with this package
4529: . ftype - the type of factorization supported by the package
4530: - createfactor - routine that will create the factored matrix ready to be used
4532: Level: intermediate
4534: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatGetFactor()
4535: @*/
4536: PetscErrorCode MatSolverTypeRegister(MatSolverType package,MatType mtype,MatFactorType ftype,PetscErrorCode (*createfactor)(Mat,MatFactorType,Mat*))
4537: {
4538: PetscErrorCode ierr;
4539: MatSolverTypeHolder next = MatSolverTypeHolders,prev = NULL;
4540: PetscBool flg;
4541: MatSolverTypeForSpecifcType inext,iprev = NULL;
4544: MatInitializePackage();
4545: if (!next) {
4546: PetscNew(&MatSolverTypeHolders);
4547: PetscStrallocpy(package,&MatSolverTypeHolders->name);
4548: PetscNew(&MatSolverTypeHolders->handlers);
4549: PetscStrallocpy(mtype,(char **)&MatSolverTypeHolders->handlers->mtype);
4550: MatSolverTypeHolders->handlers->createfactor[(int)ftype-1] = createfactor;
4551: return(0);
4552: }
4553: while (next) {
4554: PetscStrcasecmp(package,next->name,&flg);
4555: if (flg) {
4556: if (!next->handlers) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"MatSolverTypeHolder is missing handlers");
4557: inext = next->handlers;
4558: while (inext) {
4559: PetscStrcasecmp(mtype,inext->mtype,&flg);
4560: if (flg) {
4561: inext->createfactor[(int)ftype-1] = createfactor;
4562: return(0);
4563: }
4564: iprev = inext;
4565: inext = inext->next;
4566: }
4567: PetscNew(&iprev->next);
4568: PetscStrallocpy(mtype,(char **)&iprev->next->mtype);
4569: iprev->next->createfactor[(int)ftype-1] = createfactor;
4570: return(0);
4571: }
4572: prev = next;
4573: next = next->next;
4574: }
4575: PetscNew(&prev->next);
4576: PetscStrallocpy(package,&prev->next->name);
4577: PetscNew(&prev->next->handlers);
4578: PetscStrallocpy(mtype,(char **)&prev->next->handlers->mtype);
4579: prev->next->handlers->createfactor[(int)ftype-1] = createfactor;
4580: return(0);
4581: }
4583: /*@C
4584: MatSolveTypeGet - Gets the function that creates the factor matrix if it exist
4586: Input Parameters:
4587: + type - name of the package, for example petsc or superlu
4588: . ftype - the type of factorization supported by the type
4589: - mtype - the matrix type that works with this type
4591: Output Parameters:
4592: + foundtype - PETSC_TRUE if the type was registered
4593: . foundmtype - PETSC_TRUE if the type supports the requested mtype
4594: - createfactor - routine that will create the factored matrix ready to be used or NULL if not found
4596: Level: intermediate
4598: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatSolvePackageRegister), MatGetFactor()
4599: @*/
4600: PetscErrorCode MatSolverTypeGet(MatSolverType type,MatType mtype,MatFactorType ftype,PetscBool *foundtype,PetscBool *foundmtype,PetscErrorCode (**createfactor)(Mat,MatFactorType,Mat*))
4601: {
4602: PetscErrorCode ierr;
4603: MatSolverTypeHolder next = MatSolverTypeHolders;
4604: PetscBool flg;
4605: MatSolverTypeForSpecifcType inext;
4608: if (foundtype) *foundtype = PETSC_FALSE;
4609: if (foundmtype) *foundmtype = PETSC_FALSE;
4610: if (createfactor) *createfactor = NULL;
4612: if (type) {
4613: while (next) {
4614: PetscStrcasecmp(type,next->name,&flg);
4615: if (flg) {
4616: if (foundtype) *foundtype = PETSC_TRUE;
4617: inext = next->handlers;
4618: while (inext) {
4619: PetscStrbeginswith(mtype,inext->mtype,&flg);
4620: if (flg) {
4621: if (foundmtype) *foundmtype = PETSC_TRUE;
4622: if (createfactor) *createfactor = inext->createfactor[(int)ftype-1];
4623: return(0);
4624: }
4625: inext = inext->next;
4626: }
4627: }
4628: next = next->next;
4629: }
4630: } else {
4631: while (next) {
4632: inext = next->handlers;
4633: while (inext) {
4634: PetscStrcmp(mtype,inext->mtype,&flg);
4635: if (flg && inext->createfactor[(int)ftype-1]) {
4636: if (foundtype) *foundtype = PETSC_TRUE;
4637: if (foundmtype) *foundmtype = PETSC_TRUE;
4638: if (createfactor) *createfactor = inext->createfactor[(int)ftype-1];
4639: return(0);
4640: }
4641: inext = inext->next;
4642: }
4643: next = next->next;
4644: }
4645: /* try with base classes inext->mtype */
4646: next = MatSolverTypeHolders;
4647: while (next) {
4648: inext = next->handlers;
4649: while (inext) {
4650: PetscStrbeginswith(mtype,inext->mtype,&flg);
4651: if (flg && inext->createfactor[(int)ftype-1]) {
4652: if (foundtype) *foundtype = PETSC_TRUE;
4653: if (foundmtype) *foundmtype = PETSC_TRUE;
4654: if (createfactor) *createfactor = inext->createfactor[(int)ftype-1];
4655: return(0);
4656: }
4657: inext = inext->next;
4658: }
4659: next = next->next;
4660: }
4661: }
4662: return(0);
4663: }
4665: PetscErrorCode MatSolverTypeDestroy(void)
4666: {
4667: PetscErrorCode ierr;
4668: MatSolverTypeHolder next = MatSolverTypeHolders,prev;
4669: MatSolverTypeForSpecifcType inext,iprev;
4672: while (next) {
4673: PetscFree(next->name);
4674: inext = next->handlers;
4675: while (inext) {
4676: PetscFree(inext->mtype);
4677: iprev = inext;
4678: inext = inext->next;
4679: PetscFree(iprev);
4680: }
4681: prev = next;
4682: next = next->next;
4683: PetscFree(prev);
4684: }
4685: MatSolverTypeHolders = NULL;
4686: return(0);
4687: }
4689: /*@C
4690: MatFactorGetUseOrdering - Indicates if the factorization uses the ordering provided in MatLUFactorSymbolic(), MatCholeskyFactorSymbolic()
4692: Logically Collective on Mat
4694: Input Parameters:
4695: . mat - the matrix
4697: Output Parameters:
4698: . flg - PETSC_TRUE if uses the ordering
4700: Notes:
4701: Most internal PETSc factorizations use the ordering past to the factorization routine but external
4702: packages do no, thus we want to skip the ordering when it is not needed.
4704: Level: developer
4706: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatGetFactor(), MatLUFactorSymbolic(), MatCholeskyFactorSymbolic()
4707: @*/
4708: PetscErrorCode MatFactorGetUseOrdering(Mat mat, PetscBool *flg)
4709: {
4711: *flg = mat->useordering;
4712: return(0);
4713: }
4715: /*@C
4716: MatGetFactor - Returns a matrix suitable to calls to MatXXFactorSymbolic()
4718: Collective on Mat
4720: Input Parameters:
4721: + mat - the matrix
4722: . type - name of solver type, for example, superlu, petsc (to use PETSc's default)
4723: - ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
4725: Output Parameters:
4726: . f - the factor matrix used with MatXXFactorSymbolic() calls
4728: Notes:
4729: Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
4730: such as pastix, superlu, mumps etc.
4732: PETSc must have been ./configure to use the external solver, using the option --download-package
4734: Developer Notes:
4735: This should actually be called MatCreateFactor() since it creates a new factor object
4737: Level: intermediate
4739: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatFactorGetUseOrdering(), MatSolverTypeRegister()
4740: @*/
4741: PetscErrorCode MatGetFactor(Mat mat, MatSolverType type,MatFactorType ftype,Mat *f)
4742: {
4743: PetscErrorCode ierr,(*conv)(Mat,MatFactorType,Mat*);
4744: PetscBool foundtype,foundmtype;
4750: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4751: MatCheckPreallocated(mat,1);
4753: MatSolverTypeGet(type,((PetscObject)mat)->type_name,ftype,&foundtype,&foundmtype,&conv);
4754: if (!foundtype) {
4755: if (type) {
4756: SETERRQ4(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"Could not locate solver type %s for factorization type %s and matrix type %s. Perhaps you must ./configure with --download-%s",type,MatFactorTypes[ftype],((PetscObject)mat)->type_name,type);
4757: } else {
4758: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"Could not locate a solver type for factorization type %s and matrix type %s.",MatFactorTypes[ftype],((PetscObject)mat)->type_name);
4759: }
4760: }
4761: if (!foundmtype) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"MatSolverType %s does not support matrix type %s",type,((PetscObject)mat)->type_name);
4762: if (!conv) SETERRQ3(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"MatSolverType %s does not support factorization type %s for matrix type %s",type,MatFactorTypes[ftype],((PetscObject)mat)->type_name);
4764: (*conv)(mat,ftype,f);
4765: return(0);
4766: }
4768: /*@C
4769: MatGetFactorAvailable - Returns a a flag if matrix supports particular type and factor type
4771: Not Collective
4773: Input Parameters:
4774: + mat - the matrix
4775: . type - name of solver type, for example, superlu, petsc (to use PETSc's default)
4776: - ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
4778: Output Parameter:
4779: . flg - PETSC_TRUE if the factorization is available
4781: Notes:
4782: Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
4783: such as pastix, superlu, mumps etc.
4785: PETSc must have been ./configure to use the external solver, using the option --download-package
4787: Developer Notes:
4788: This should actually be called MatCreateFactorAvailable() since MatGetFactor() creates a new factor object
4790: Level: intermediate
4792: .seealso: MatCopy(), MatDuplicate(), MatGetFactor(), MatSolverTypeRegister()
4793: @*/
4794: PetscErrorCode MatGetFactorAvailable(Mat mat, MatSolverType type,MatFactorType ftype,PetscBool *flg)
4795: {
4796: PetscErrorCode ierr, (*gconv)(Mat,MatFactorType,Mat*);
4802: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4803: MatCheckPreallocated(mat,1);
4805: *flg = PETSC_FALSE;
4806: MatSolverTypeGet(type,((PetscObject)mat)->type_name,ftype,NULL,NULL,&gconv);
4807: if (gconv) {
4808: *flg = PETSC_TRUE;
4809: }
4810: return(0);
4811: }
4813: #include <petscdmtypes.h>
4815: /*@
4816: MatDuplicate - Duplicates a matrix including the non-zero structure.
4818: Collective on Mat
4820: Input Parameters:
4821: + mat - the matrix
4822: - op - One of MAT_DO_NOT_COPY_VALUES, MAT_COPY_VALUES, or MAT_SHARE_NONZERO_PATTERN.
4823: See the manual page for MatDuplicateOption for an explanation of these options.
4825: Output Parameter:
4826: . M - pointer to place new matrix
4828: Level: intermediate
4830: Notes:
4831: You cannot change the nonzero pattern for the parent or child matrix if you use MAT_SHARE_NONZERO_PATTERN.
4832: When original mat is a product of matrix operation, e.g., an output of MatMatMult() or MatCreateSubMatrix(), only the simple matrix data structure of mat is duplicated and the internal data structures created for the reuse of previous matrix operations are not duplicated. User should not use MatDuplicate() to create new matrix M if M is intended to be reused as the product of matrix operation.
4834: .seealso: MatCopy(), MatConvert(), MatDuplicateOption
4835: @*/
4836: PetscErrorCode MatDuplicate(Mat mat,MatDuplicateOption op,Mat *M)
4837: {
4839: Mat B;
4840: PetscInt i;
4841: DM dm;
4842: void (*viewf)(void);
4848: if (op == MAT_COPY_VALUES && !mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MAT_COPY_VALUES not allowed for unassembled matrix");
4849: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4850: MatCheckPreallocated(mat,1);
4852: *M = NULL;
4853: if (!mat->ops->duplicate) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Not written for matrix type %s\n",((PetscObject)mat)->type_name);
4854: PetscLogEventBegin(MAT_Convert,mat,0,0,0);
4855: (*mat->ops->duplicate)(mat,op,M);
4856: B = *M;
4858: MatGetOperation(mat,MATOP_VIEW,&viewf);
4859: if (viewf) {
4860: MatSetOperation(B,MATOP_VIEW,viewf);
4861: }
4863: B->stencil.dim = mat->stencil.dim;
4864: B->stencil.noc = mat->stencil.noc;
4865: for (i=0; i<=mat->stencil.dim; i++) {
4866: B->stencil.dims[i] = mat->stencil.dims[i];
4867: B->stencil.starts[i] = mat->stencil.starts[i];
4868: }
4870: B->nooffproczerorows = mat->nooffproczerorows;
4871: B->nooffprocentries = mat->nooffprocentries;
4873: PetscObjectQuery((PetscObject) mat, "__PETSc_dm", (PetscObject*) &dm);
4874: if (dm) {
4875: PetscObjectCompose((PetscObject) B, "__PETSc_dm", (PetscObject) dm);
4876: }
4877: PetscLogEventEnd(MAT_Convert,mat,0,0,0);
4878: PetscObjectStateIncrease((PetscObject)B);
4879: return(0);
4880: }
4882: /*@
4883: MatGetDiagonal - Gets the diagonal of a matrix.
4885: Logically Collective on Mat
4887: Input Parameters:
4888: + mat - the matrix
4889: - v - the vector for storing the diagonal
4891: Output Parameter:
4892: . v - the diagonal of the matrix
4894: Level: intermediate
4896: Note:
4897: Currently only correct in parallel for square matrices.
4899: .seealso: MatGetRow(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs()
4900: @*/
4901: PetscErrorCode MatGetDiagonal(Mat mat,Vec v)
4902: {
4909: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4910: if (!mat->ops->getdiagonal) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4911: MatCheckPreallocated(mat,1);
4913: (*mat->ops->getdiagonal)(mat,v);
4914: PetscObjectStateIncrease((PetscObject)v);
4915: return(0);
4916: }
4918: /*@C
4919: MatGetRowMin - Gets the minimum value (of the real part) of each
4920: row of the matrix
4922: Logically Collective on Mat
4924: Input Parameters:
4925: . mat - the matrix
4927: Output Parameter:
4928: + v - the vector for storing the maximums
4929: - idx - the indices of the column found for each row (optional)
4931: Level: intermediate
4933: Notes:
4934: The result of this call are the same as if one converted the matrix to dense format
4935: and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
4937: This code is only implemented for a couple of matrix formats.
4939: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs(),
4940: MatGetRowMax()
4941: @*/
4942: PetscErrorCode MatGetRowMin(Mat mat,Vec v,PetscInt idx[])
4943: {
4950: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4952: if (!mat->cmap->N) {
4953: VecSet(v,PETSC_MAX_REAL);
4954: if (idx) {
4955: PetscInt i,m = mat->rmap->n;
4956: for (i=0; i<m; i++) idx[i] = -1;
4957: }
4958: } else {
4959: if (!mat->ops->getrowmin) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4960: MatCheckPreallocated(mat,1);
4961: }
4962: (*mat->ops->getrowmin)(mat,v,idx);
4963: PetscObjectStateIncrease((PetscObject)v);
4964: return(0);
4965: }
4967: /*@C
4968: MatGetRowMinAbs - Gets the minimum value (in absolute value) of each
4969: row of the matrix
4971: Logically Collective on Mat
4973: Input Parameters:
4974: . mat - the matrix
4976: Output Parameter:
4977: + v - the vector for storing the minimums
4978: - idx - the indices of the column found for each row (or NULL if not needed)
4980: Level: intermediate
4982: Notes:
4983: if a row is completely empty or has only 0.0 values then the idx[] value for that
4984: row is 0 (the first column).
4986: This code is only implemented for a couple of matrix formats.
4988: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMaxAbs(), MatGetRowMin()
4989: @*/
4990: PetscErrorCode MatGetRowMinAbs(Mat mat,Vec v,PetscInt idx[])
4991: {
4998: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4999: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5001: if (!mat->cmap->N) {
5002: VecSet(v,0.0);
5003: if (idx) {
5004: PetscInt i,m = mat->rmap->n;
5005: for (i=0; i<m; i++) idx[i] = -1;
5006: }
5007: } else {
5008: if (!mat->ops->getrowminabs) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5009: MatCheckPreallocated(mat,1);
5010: if (idx) {PetscArrayzero(idx,mat->rmap->n);}
5011: (*mat->ops->getrowminabs)(mat,v,idx);
5012: }
5013: PetscObjectStateIncrease((PetscObject)v);
5014: return(0);
5015: }
5017: /*@C
5018: MatGetRowMax - Gets the maximum value (of the real part) of each
5019: row of the matrix
5021: Logically Collective on Mat
5023: Input Parameters:
5024: . mat - the matrix
5026: Output Parameter:
5027: + v - the vector for storing the maximums
5028: - idx - the indices of the column found for each row (optional)
5030: Level: intermediate
5032: Notes:
5033: The result of this call are the same as if one converted the matrix to dense format
5034: and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
5036: This code is only implemented for a couple of matrix formats.
5038: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs(), MatGetRowMin()
5039: @*/
5040: PetscErrorCode MatGetRowMax(Mat mat,Vec v,PetscInt idx[])
5041: {
5048: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5050: if (!mat->cmap->N) {
5051: VecSet(v,PETSC_MIN_REAL);
5052: if (idx) {
5053: PetscInt i,m = mat->rmap->n;
5054: for (i=0; i<m; i++) idx[i] = -1;
5055: }
5056: } else {
5057: if (!mat->ops->getrowmax) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5058: MatCheckPreallocated(mat,1);
5059: (*mat->ops->getrowmax)(mat,v,idx);
5060: }
5061: PetscObjectStateIncrease((PetscObject)v);
5062: return(0);
5063: }
5065: /*@C
5066: MatGetRowMaxAbs - Gets the maximum value (in absolute value) of each
5067: row of the matrix
5069: Logically Collective on Mat
5071: Input Parameters:
5072: . mat - the matrix
5074: Output Parameter:
5075: + v - the vector for storing the maximums
5076: - idx - the indices of the column found for each row (or NULL if not needed)
5078: Level: intermediate
5080: Notes:
5081: if a row is completely empty or has only 0.0 values then the idx[] value for that
5082: row is 0 (the first column).
5084: This code is only implemented for a couple of matrix formats.
5086: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMin()
5087: @*/
5088: PetscErrorCode MatGetRowMaxAbs(Mat mat,Vec v,PetscInt idx[])
5089: {
5096: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5098: if (!mat->cmap->N) {
5099: VecSet(v,0.0);
5100: if (idx) {
5101: PetscInt i,m = mat->rmap->n;
5102: for (i=0; i<m; i++) idx[i] = -1;
5103: }
5104: } else {
5105: if (!mat->ops->getrowmaxabs) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5106: MatCheckPreallocated(mat,1);
5107: if (idx) {PetscArrayzero(idx,mat->rmap->n);}
5108: (*mat->ops->getrowmaxabs)(mat,v,idx);
5109: }
5110: PetscObjectStateIncrease((PetscObject)v);
5111: return(0);
5112: }
5114: /*@
5115: MatGetRowSum - Gets the sum of each row of the matrix
5117: Logically or Neighborhood Collective on Mat
5119: Input Parameters:
5120: . mat - the matrix
5122: Output Parameter:
5123: . v - the vector for storing the sum of rows
5125: Level: intermediate
5127: Notes:
5128: This code is slow since it is not currently specialized for different formats
5130: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMin()
5131: @*/
5132: PetscErrorCode MatGetRowSum(Mat mat, Vec v)
5133: {
5134: Vec ones;
5141: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5142: MatCheckPreallocated(mat,1);
5143: MatCreateVecs(mat,&ones,NULL);
5144: VecSet(ones,1.);
5145: MatMult(mat,ones,v);
5146: VecDestroy(&ones);
5147: return(0);
5148: }
5150: /*@
5151: MatTranspose - Computes an in-place or out-of-place transpose of a matrix.
5153: Collective on Mat
5155: Input Parameter:
5156: + mat - the matrix to transpose
5157: - reuse - either MAT_INITIAL_MATRIX, MAT_REUSE_MATRIX, or MAT_INPLACE_MATRIX
5159: Output Parameters:
5160: . B - the transpose
5162: Notes:
5163: If you use MAT_INPLACE_MATRIX then you must pass in &mat for B
5165: MAT_REUSE_MATRIX causes the B matrix from a previous call to this function with MAT_INITIAL_MATRIX to be used
5167: Consider using MatCreateTranspose() instead if you only need a matrix that behaves like the transpose, but don't need the storage to be changed.
5169: Level: intermediate
5171: .seealso: MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
5172: @*/
5173: PetscErrorCode MatTranspose(Mat mat,MatReuse reuse,Mat *B)
5174: {
5180: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5181: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5182: if (!mat->ops->transpose) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5183: if (reuse == MAT_INPLACE_MATRIX && mat != *B) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"MAT_INPLACE_MATRIX requires last matrix to match first");
5184: if (reuse == MAT_REUSE_MATRIX && mat == *B) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Perhaps you mean MAT_INPLACE_MATRIX");
5185: MatCheckPreallocated(mat,1);
5187: PetscLogEventBegin(MAT_Transpose,mat,0,0,0);
5188: (*mat->ops->transpose)(mat,reuse,B);
5189: PetscLogEventEnd(MAT_Transpose,mat,0,0,0);
5190: if (B) {PetscObjectStateIncrease((PetscObject)*B);}
5191: return(0);
5192: }
5194: /*@
5195: MatIsTranspose - Test whether a matrix is another one's transpose,
5196: or its own, in which case it tests symmetry.
5198: Collective on Mat
5200: Input Parameter:
5201: + A - the matrix to test
5202: - B - the matrix to test against, this can equal the first parameter
5204: Output Parameters:
5205: . flg - the result
5207: Notes:
5208: Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
5209: has a running time of the order of the number of nonzeros; the parallel
5210: test involves parallel copies of the block-offdiagonal parts of the matrix.
5212: Level: intermediate
5214: .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian()
5215: @*/
5216: PetscErrorCode MatIsTranspose(Mat A,Mat B,PetscReal tol,PetscBool *flg)
5217: {
5218: PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscBool*),(*g)(Mat,Mat,PetscReal,PetscBool*);
5224: PetscObjectQueryFunction((PetscObject)A,"MatIsTranspose_C",&f);
5225: PetscObjectQueryFunction((PetscObject)B,"MatIsTranspose_C",&g);
5226: *flg = PETSC_FALSE;
5227: if (f && g) {
5228: if (f == g) {
5229: (*f)(A,B,tol,flg);
5230: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for symmetry test");
5231: } else {
5232: MatType mattype;
5233: if (!f) {
5234: MatGetType(A,&mattype);
5235: } else {
5236: MatGetType(B,&mattype);
5237: }
5238: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for transpose",mattype);
5239: }
5240: return(0);
5241: }
5243: /*@
5244: MatHermitianTranspose - Computes an in-place or out-of-place transpose of a matrix in complex conjugate.
5246: Collective on Mat
5248: Input Parameter:
5249: + mat - the matrix to transpose and complex conjugate
5250: - reuse - MAT_INITIAL_MATRIX to create a new matrix, MAT_INPLACE_MATRIX to reuse the first argument to store the transpose
5252: Output Parameters:
5253: . B - the Hermitian
5255: Level: intermediate
5257: .seealso: MatTranspose(), MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
5258: @*/
5259: PetscErrorCode MatHermitianTranspose(Mat mat,MatReuse reuse,Mat *B)
5260: {
5264: MatTranspose(mat,reuse,B);
5265: #if defined(PETSC_USE_COMPLEX)
5266: MatConjugate(*B);
5267: #endif
5268: return(0);
5269: }
5271: /*@
5272: MatIsHermitianTranspose - Test whether a matrix is another one's Hermitian transpose,
5274: Collective on Mat
5276: Input Parameter:
5277: + A - the matrix to test
5278: - B - the matrix to test against, this can equal the first parameter
5280: Output Parameters:
5281: . flg - the result
5283: Notes:
5284: Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
5285: has a running time of the order of the number of nonzeros; the parallel
5286: test involves parallel copies of the block-offdiagonal parts of the matrix.
5288: Level: intermediate
5290: .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian(), MatIsTranspose()
5291: @*/
5292: PetscErrorCode MatIsHermitianTranspose(Mat A,Mat B,PetscReal tol,PetscBool *flg)
5293: {
5294: PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscBool*),(*g)(Mat,Mat,PetscReal,PetscBool*);
5300: PetscObjectQueryFunction((PetscObject)A,"MatIsHermitianTranspose_C",&f);
5301: PetscObjectQueryFunction((PetscObject)B,"MatIsHermitianTranspose_C",&g);
5302: if (f && g) {
5303: if (f==g) {
5304: (*f)(A,B,tol,flg);
5305: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for Hermitian test");
5306: }
5307: return(0);
5308: }
5310: /*@
5311: MatPermute - Creates a new matrix with rows and columns permuted from the
5312: original.
5314: Collective on Mat
5316: Input Parameters:
5317: + mat - the matrix to permute
5318: . row - row permutation, each processor supplies only the permutation for its rows
5319: - col - column permutation, each processor supplies only the permutation for its columns
5321: Output Parameters:
5322: . B - the permuted matrix
5324: Level: advanced
5326: Note:
5327: The index sets map from row/col of permuted matrix to row/col of original matrix.
5328: The index sets should be on the same communicator as Mat and have the same local sizes.
5330: Developer Note:
5331: If you want to implement MatPermute for a matrix type, and your approach doesn't
5332: exploit the fact that row and col are permutations, consider implementing the
5333: more general MatCreateSubMatrix() instead.
5335: .seealso: MatGetOrdering(), ISAllGather()
5337: @*/
5338: PetscErrorCode MatPermute(Mat mat,IS row,IS col,Mat *B)
5339: {
5350: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5351: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5352: if (!mat->ops->permute && !mat->ops->createsubmatrix) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatPermute not available for Mat type %s",((PetscObject)mat)->type_name);
5353: MatCheckPreallocated(mat,1);
5355: if (mat->ops->permute) {
5356: (*mat->ops->permute)(mat,row,col,B);
5357: PetscObjectStateIncrease((PetscObject)*B);
5358: } else {
5359: MatCreateSubMatrix(mat, row, col, MAT_INITIAL_MATRIX, B);
5360: }
5361: return(0);
5362: }
5364: /*@
5365: MatEqual - Compares two matrices.
5367: Collective on Mat
5369: Input Parameters:
5370: + A - the first matrix
5371: - B - the second matrix
5373: Output Parameter:
5374: . flg - PETSC_TRUE if the matrices are equal; PETSC_FALSE otherwise.
5376: Level: intermediate
5378: @*/
5379: PetscErrorCode MatEqual(Mat A,Mat B,PetscBool *flg)
5380: {
5390: MatCheckPreallocated(B,2);
5391: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5392: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5393: if (A->rmap->N != B->rmap->N || A->cmap->N != B->cmap->N) SETERRQ4(PetscObjectComm((PetscObject)A),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);
5394: if (!A->ops->equal) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
5395: if (!B->ops->equal) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)B)->type_name);
5396: if (A->ops->equal != B->ops->equal) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"A is type: %s\nB is type: %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
5397: MatCheckPreallocated(A,1);
5399: (*A->ops->equal)(A,B,flg);
5400: return(0);
5401: }
5403: /*@
5404: MatDiagonalScale - Scales a matrix on the left and right by diagonal
5405: matrices that are stored as vectors. Either of the two scaling
5406: matrices can be NULL.
5408: Collective on Mat
5410: Input Parameters:
5411: + mat - the matrix to be scaled
5412: . l - the left scaling vector (or NULL)
5413: - r - the right scaling vector (or NULL)
5415: Notes:
5416: MatDiagonalScale() computes A = LAR, where
5417: L = a diagonal matrix (stored as a vector), R = a diagonal matrix (stored as a vector)
5418: The L scales the rows of the matrix, the R scales the columns of the matrix.
5420: Level: intermediate
5423: .seealso: MatScale(), MatShift(), MatDiagonalSet()
5424: @*/
5425: PetscErrorCode MatDiagonalScale(Mat mat,Vec l,Vec r)
5426: {
5434: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5435: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5436: MatCheckPreallocated(mat,1);
5437: if (!l && !r) return(0);
5439: if (!mat->ops->diagonalscale) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5440: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
5441: (*mat->ops->diagonalscale)(mat,l,r);
5442: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
5443: PetscObjectStateIncrease((PetscObject)mat);
5444: return(0);
5445: }
5447: /*@
5448: MatScale - Scales all elements of a matrix by a given number.
5450: Logically Collective on Mat
5452: Input Parameters:
5453: + mat - the matrix to be scaled
5454: - a - the scaling value
5456: Output Parameter:
5457: . mat - the scaled matrix
5459: Level: intermediate
5461: .seealso: MatDiagonalScale()
5462: @*/
5463: PetscErrorCode MatScale(Mat mat,PetscScalar a)
5464: {
5470: if (a != (PetscScalar)1.0 && !mat->ops->scale) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5471: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5472: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5474: MatCheckPreallocated(mat,1);
5476: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
5477: if (a != (PetscScalar)1.0) {
5478: (*mat->ops->scale)(mat,a);
5479: PetscObjectStateIncrease((PetscObject)mat);
5480: }
5481: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
5482: return(0);
5483: }
5485: /*@
5486: MatNorm - Calculates various norms of a matrix.
5488: Collective on Mat
5490: Input Parameters:
5491: + mat - the matrix
5492: - type - the type of norm, NORM_1, NORM_FROBENIUS, NORM_INFINITY
5494: Output Parameters:
5495: . nrm - the resulting norm
5497: Level: intermediate
5499: @*/
5500: PetscErrorCode MatNorm(Mat mat,NormType type,PetscReal *nrm)
5501: {
5509: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5510: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5511: if (!mat->ops->norm) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5512: MatCheckPreallocated(mat,1);
5514: (*mat->ops->norm)(mat,type,nrm);
5515: return(0);
5516: }
5518: /*
5519: This variable is used to prevent counting of MatAssemblyBegin() that
5520: are called from within a MatAssemblyEnd().
5521: */
5522: static PetscInt MatAssemblyEnd_InUse = 0;
5523: /*@
5524: MatAssemblyBegin - Begins assembling the matrix. This routine should
5525: be called after completing all calls to MatSetValues().
5527: Collective on Mat
5529: Input Parameters:
5530: + mat - the matrix
5531: - type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
5533: Notes:
5534: MatSetValues() generally caches the values. The matrix is ready to
5535: use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
5536: Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
5537: in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
5538: using the matrix.
5540: ALL processes that share a matrix MUST call MatAssemblyBegin() and MatAssemblyEnd() the SAME NUMBER of times, and each time with the
5541: same flag of MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY for all processes. Thus you CANNOT locally change from ADD_VALUES to INSERT_VALUES, that is
5542: a global collective operation requring all processes that share the matrix.
5544: Space for preallocated nonzeros that is not filled by a call to MatSetValues() or a related routine are compressed
5545: out by assembly. If you intend to use that extra space on a subsequent assembly, be sure to insert explicit zeros
5546: before MAT_FINAL_ASSEMBLY so the space is not compressed out.
5548: Level: beginner
5550: .seealso: MatAssemblyEnd(), MatSetValues(), MatAssembled()
5551: @*/
5552: PetscErrorCode MatAssemblyBegin(Mat mat,MatAssemblyType type)
5553: {
5559: MatCheckPreallocated(mat,1);
5560: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix.\nDid you forget to call MatSetUnfactored()?");
5561: if (mat->assembled) {
5562: mat->was_assembled = PETSC_TRUE;
5563: mat->assembled = PETSC_FALSE;
5564: }
5566: if (!MatAssemblyEnd_InUse) {
5567: PetscLogEventBegin(MAT_AssemblyBegin,mat,0,0,0);
5568: if (mat->ops->assemblybegin) {(*mat->ops->assemblybegin)(mat,type);}
5569: PetscLogEventEnd(MAT_AssemblyBegin,mat,0,0,0);
5570: } else if (mat->ops->assemblybegin) {
5571: (*mat->ops->assemblybegin)(mat,type);
5572: }
5573: return(0);
5574: }
5576: /*@
5577: MatAssembled - Indicates if a matrix has been assembled and is ready for
5578: use; for example, in matrix-vector product.
5580: Not Collective
5582: Input Parameter:
5583: . mat - the matrix
5585: Output Parameter:
5586: . assembled - PETSC_TRUE or PETSC_FALSE
5588: Level: advanced
5590: .seealso: MatAssemblyEnd(), MatSetValues(), MatAssemblyBegin()
5591: @*/
5592: PetscErrorCode MatAssembled(Mat mat,PetscBool *assembled)
5593: {
5597: *assembled = mat->assembled;
5598: return(0);
5599: }
5601: /*@
5602: MatAssemblyEnd - Completes assembling the matrix. This routine should
5603: be called after MatAssemblyBegin().
5605: Collective on Mat
5607: Input Parameters:
5608: + mat - the matrix
5609: - type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
5611: Options Database Keys:
5612: + -mat_view ::ascii_info - Prints info on matrix at conclusion of MatEndAssembly()
5613: . -mat_view ::ascii_info_detail - Prints more detailed info
5614: . -mat_view - Prints matrix in ASCII format
5615: . -mat_view ::ascii_matlab - Prints matrix in Matlab format
5616: . -mat_view draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
5617: . -display <name> - Sets display name (default is host)
5618: . -draw_pause <sec> - Sets number of seconds to pause after display
5619: . -mat_view socket - Sends matrix to socket, can be accessed from Matlab (See Users-Manual: ch_matlab)
5620: . -viewer_socket_machine <machine> - Machine to use for socket
5621: . -viewer_socket_port <port> - Port number to use for socket
5622: - -mat_view binary:filename[:append] - Save matrix to file in binary format
5624: Notes:
5625: MatSetValues() generally caches the values. The matrix is ready to
5626: use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
5627: Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
5628: in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
5629: using the matrix.
5631: Space for preallocated nonzeros that is not filled by a call to MatSetValues() or a related routine are compressed
5632: out by assembly. If you intend to use that extra space on a subsequent assembly, be sure to insert explicit zeros
5633: before MAT_FINAL_ASSEMBLY so the space is not compressed out.
5635: Level: beginner
5637: .seealso: MatAssemblyBegin(), MatSetValues(), PetscDrawOpenX(), PetscDrawCreate(), MatView(), MatAssembled(), PetscViewerSocketOpen()
5638: @*/
5639: PetscErrorCode MatAssemblyEnd(Mat mat,MatAssemblyType type)
5640: {
5641: PetscErrorCode ierr;
5642: static PetscInt inassm = 0;
5643: PetscBool flg = PETSC_FALSE;
5649: inassm++;
5650: MatAssemblyEnd_InUse++;
5651: if (MatAssemblyEnd_InUse == 1) { /* Do the logging only the first time through */
5652: PetscLogEventBegin(MAT_AssemblyEnd,mat,0,0,0);
5653: if (mat->ops->assemblyend) {
5654: (*mat->ops->assemblyend)(mat,type);
5655: }
5656: PetscLogEventEnd(MAT_AssemblyEnd,mat,0,0,0);
5657: } else if (mat->ops->assemblyend) {
5658: (*mat->ops->assemblyend)(mat,type);
5659: }
5661: /* Flush assembly is not a true assembly */
5662: if (type != MAT_FLUSH_ASSEMBLY) {
5663: mat->num_ass++;
5664: mat->assembled = PETSC_TRUE;
5665: mat->ass_nonzerostate = mat->nonzerostate;
5666: }
5668: mat->insertmode = NOT_SET_VALUES;
5669: MatAssemblyEnd_InUse--;
5670: PetscObjectStateIncrease((PetscObject)mat);
5671: if (!mat->symmetric_eternal) {
5672: mat->symmetric_set = PETSC_FALSE;
5673: mat->hermitian_set = PETSC_FALSE;
5674: mat->structurally_symmetric_set = PETSC_FALSE;
5675: }
5676: if (inassm == 1 && type != MAT_FLUSH_ASSEMBLY) {
5677: MatViewFromOptions(mat,NULL,"-mat_view");
5679: if (mat->checksymmetryonassembly) {
5680: MatIsSymmetric(mat,mat->checksymmetrytol,&flg);
5681: if (flg) {
5682: PetscPrintf(PetscObjectComm((PetscObject)mat),"Matrix is symmetric (tolerance %g)\n",(double)mat->checksymmetrytol);
5683: } else {
5684: PetscPrintf(PetscObjectComm((PetscObject)mat),"Matrix is not symmetric (tolerance %g)\n",(double)mat->checksymmetrytol);
5685: }
5686: }
5687: if (mat->nullsp && mat->checknullspaceonassembly) {
5688: MatNullSpaceTest(mat->nullsp,mat,NULL);
5689: }
5690: }
5691: inassm--;
5692: return(0);
5693: }
5695: /*@
5696: MatSetOption - Sets a parameter option for a matrix. Some options
5697: may be specific to certain storage formats. Some options
5698: determine how values will be inserted (or added). Sorted,
5699: row-oriented input will generally assemble the fastest. The default
5700: is row-oriented.
5702: Logically Collective on Mat for certain operations, such as MAT_SPD, not collective for MAT_ROW_ORIENTED, see MatOption
5704: Input Parameters:
5705: + mat - the matrix
5706: . option - the option, one of those listed below (and possibly others),
5707: - flg - turn the option on (PETSC_TRUE) or off (PETSC_FALSE)
5709: Options Describing Matrix Structure:
5710: + MAT_SPD - symmetric positive definite
5711: . MAT_SYMMETRIC - symmetric in terms of both structure and value
5712: . MAT_HERMITIAN - transpose is the complex conjugation
5713: . MAT_STRUCTURALLY_SYMMETRIC - symmetric nonzero structure
5714: - MAT_SYMMETRY_ETERNAL - if you would like the symmetry/Hermitian flag
5715: you set to be kept with all future use of the matrix
5716: including after MatAssemblyBegin/End() which could
5717: potentially change the symmetry structure, i.e. you
5718: KNOW the matrix will ALWAYS have the property you set.
5719: Note that setting this flag alone implies nothing about whether the matrix is symmetric/Hermitian;
5720: the relevant flags must be set independently.
5723: Options For Use with MatSetValues():
5724: Insert a logically dense subblock, which can be
5725: . MAT_ROW_ORIENTED - row-oriented (default)
5727: Note these options reflect the data you pass in with MatSetValues(); it has
5728: nothing to do with how the data is stored internally in the matrix
5729: data structure.
5731: When (re)assembling a matrix, we can restrict the input for
5732: efficiency/debugging purposes. These options include:
5733: + MAT_NEW_NONZERO_LOCATIONS - additional insertions will be allowed if they generate a new nonzero (slow)
5734: . MAT_FORCE_DIAGONAL_ENTRIES - forces diagonal entries to be allocated
5735: . MAT_IGNORE_OFF_PROC_ENTRIES - drops off-processor entries
5736: . MAT_NEW_NONZERO_LOCATION_ERR - generates an error for new matrix entry
5737: . MAT_USE_HASH_TABLE - uses a hash table to speed up matrix assembly
5738: . MAT_NO_OFF_PROC_ENTRIES - you know each process will only set values for its own rows, will generate an error if
5739: any process sets values for another process. This avoids all reductions in the MatAssembly routines and thus improves
5740: performance for very large process counts.
5741: - MAT_SUBSET_OFF_PROC_ENTRIES - you know that the first assembly after setting this flag will set a superset
5742: of the off-process entries required for all subsequent assemblies. This avoids a rendezvous step in the MatAssembly
5743: functions, instead sending only neighbor messages.
5745: Notes:
5746: Except for MAT_UNUSED_NONZERO_LOCATION_ERR and MAT_ROW_ORIENTED all processes that share the matrix must pass the same value in flg!
5748: Some options are relevant only for particular matrix types and
5749: are thus ignored by others. Other options are not supported by
5750: certain matrix types and will generate an error message if set.
5752: If using a Fortran 77 module to compute a matrix, one may need to
5753: use the column-oriented option (or convert to the row-oriented
5754: format).
5756: MAT_NEW_NONZERO_LOCATIONS set to PETSC_FALSE indicates that any add or insertion
5757: that would generate a new entry in the nonzero structure is instead
5758: ignored. Thus, if memory has not alredy been allocated for this particular
5759: data, then the insertion is ignored. For dense matrices, in which
5760: the entire array is allocated, no entries are ever ignored.
5761: Set after the first MatAssemblyEnd(). If this option is set then the MatAssemblyBegin/End() processes has one less global reduction
5763: MAT_NEW_NONZERO_LOCATION_ERR set to PETSC_TRUE indicates that any add or insertion
5764: that would generate a new entry in the nonzero structure instead produces
5765: an error. (Currently supported for AIJ and BAIJ formats only.) If this option is set then the MatAssemblyBegin/End() processes has one less global reduction
5767: MAT_NEW_NONZERO_ALLOCATION_ERR set to PETSC_TRUE indicates that any add or insertion
5768: that would generate a new entry that has not been preallocated will
5769: instead produce an error. (Currently supported for AIJ and BAIJ formats
5770: only.) This is a useful flag when debugging matrix memory preallocation.
5771: If this option is set then the MatAssemblyBegin/End() processes has one less global reduction
5773: MAT_IGNORE_OFF_PROC_ENTRIES set to PETSC_TRUE indicates entries destined for
5774: other processors should be dropped, rather than stashed.
5775: This is useful if you know that the "owning" processor is also
5776: always generating the correct matrix entries, so that PETSc need
5777: not transfer duplicate entries generated on another processor.
5779: MAT_USE_HASH_TABLE indicates that a hash table be used to improve the
5780: searches during matrix assembly. When this flag is set, the hash table
5781: is created during the first Matrix Assembly. This hash table is
5782: used the next time through, during MatSetVaules()/MatSetVaulesBlocked()
5783: to improve the searching of indices. MAT_NEW_NONZERO_LOCATIONS flag
5784: should be used with MAT_USE_HASH_TABLE flag. This option is currently
5785: supported by MATMPIBAIJ format only.
5787: MAT_KEEP_NONZERO_PATTERN indicates when MatZeroRows() is called the zeroed entries
5788: are kept in the nonzero structure
5790: MAT_IGNORE_ZERO_ENTRIES - for AIJ/IS matrices this will stop zero values from creating
5791: a zero location in the matrix
5793: MAT_USE_INODES - indicates using inode version of the code - works with AIJ matrix types
5795: MAT_NO_OFF_PROC_ZERO_ROWS - you know each process will only zero its own rows. This avoids all reductions in the
5796: zero row routines and thus improves performance for very large process counts.
5798: MAT_IGNORE_LOWER_TRIANGULAR - For SBAIJ matrices will ignore any insertions you make in the lower triangular
5799: part of the matrix (since they should match the upper triangular part).
5801: MAT_SORTED_FULL - each process provides exactly its local rows; all column indices for a given row are passed in a
5802: single call to MatSetValues(), preallocation is perfect, row oriented, INSERT_VALUES is used. Common
5803: with finite difference schemes with non-periodic boundary conditions.
5805: Level: intermediate
5807: .seealso: MatOption, Mat
5809: @*/
5810: PetscErrorCode MatSetOption(Mat mat,MatOption op,PetscBool flg)
5811: {
5816: if (op > 0) {
5819: }
5821: if (((int) op) <= MAT_OPTION_MIN || ((int) op) >= MAT_OPTION_MAX) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Options %d is out of range",(int)op);
5823: switch (op) {
5824: case MAT_FORCE_DIAGONAL_ENTRIES:
5825: mat->force_diagonals = flg;
5826: return(0);
5827: case MAT_NO_OFF_PROC_ENTRIES:
5828: mat->nooffprocentries = flg;
5829: return(0);
5830: case MAT_SUBSET_OFF_PROC_ENTRIES:
5831: mat->assembly_subset = flg;
5832: if (!mat->assembly_subset) { /* See the same logic in VecAssembly wrt VEC_SUBSET_OFF_PROC_ENTRIES */
5833: #if !defined(PETSC_HAVE_MPIUNI)
5834: MatStashScatterDestroy_BTS(&mat->stash);
5835: #endif
5836: mat->stash.first_assembly_done = PETSC_FALSE;
5837: }
5838: return(0);
5839: case MAT_NO_OFF_PROC_ZERO_ROWS:
5840: mat->nooffproczerorows = flg;
5841: return(0);
5842: case MAT_SPD:
5843: mat->spd_set = PETSC_TRUE;
5844: mat->spd = flg;
5845: if (flg) {
5846: mat->symmetric = PETSC_TRUE;
5847: mat->structurally_symmetric = PETSC_TRUE;
5848: mat->symmetric_set = PETSC_TRUE;
5849: mat->structurally_symmetric_set = PETSC_TRUE;
5850: }
5851: break;
5852: case MAT_SYMMETRIC:
5853: mat->symmetric = flg;
5854: if (flg) mat->structurally_symmetric = PETSC_TRUE;
5855: mat->symmetric_set = PETSC_TRUE;
5856: mat->structurally_symmetric_set = flg;
5857: #if !defined(PETSC_USE_COMPLEX)
5858: mat->hermitian = flg;
5859: mat->hermitian_set = PETSC_TRUE;
5860: #endif
5861: break;
5862: case MAT_HERMITIAN:
5863: mat->hermitian = flg;
5864: if (flg) mat->structurally_symmetric = PETSC_TRUE;
5865: mat->hermitian_set = PETSC_TRUE;
5866: mat->structurally_symmetric_set = flg;
5867: #if !defined(PETSC_USE_COMPLEX)
5868: mat->symmetric = flg;
5869: mat->symmetric_set = PETSC_TRUE;
5870: #endif
5871: break;
5872: case MAT_STRUCTURALLY_SYMMETRIC:
5873: mat->structurally_symmetric = flg;
5874: mat->structurally_symmetric_set = PETSC_TRUE;
5875: break;
5876: case MAT_SYMMETRY_ETERNAL:
5877: mat->symmetric_eternal = flg;
5878: break;
5879: case MAT_STRUCTURE_ONLY:
5880: mat->structure_only = flg;
5881: break;
5882: case MAT_SORTED_FULL:
5883: mat->sortedfull = flg;
5884: break;
5885: default:
5886: break;
5887: }
5888: if (mat->ops->setoption) {
5889: (*mat->ops->setoption)(mat,op,flg);
5890: }
5891: return(0);
5892: }
5894: /*@
5895: MatGetOption - Gets a parameter option that has been set for a matrix.
5897: Logically Collective on Mat for certain operations, such as MAT_SPD, not collective for MAT_ROW_ORIENTED, see MatOption
5899: Input Parameters:
5900: + mat - the matrix
5901: - option - the option, this only responds to certain options, check the code for which ones
5903: Output Parameter:
5904: . flg - turn the option on (PETSC_TRUE) or off (PETSC_FALSE)
5906: Notes:
5907: Can only be called after MatSetSizes() and MatSetType() have been set.
5909: Level: intermediate
5911: .seealso: MatOption, MatSetOption()
5913: @*/
5914: PetscErrorCode MatGetOption(Mat mat,MatOption op,PetscBool *flg)
5915: {
5920: if (((int) op) <= MAT_OPTION_MIN || ((int) op) >= MAT_OPTION_MAX) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Options %d is out of range",(int)op);
5921: if (!((PetscObject)mat)->type_name) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_TYPENOTSET,"Cannot get options until type and size have been set, see MatSetType() and MatSetSizes()");
5923: switch (op) {
5924: case MAT_NO_OFF_PROC_ENTRIES:
5925: *flg = mat->nooffprocentries;
5926: break;
5927: case MAT_NO_OFF_PROC_ZERO_ROWS:
5928: *flg = mat->nooffproczerorows;
5929: break;
5930: case MAT_SYMMETRIC:
5931: *flg = mat->symmetric;
5932: break;
5933: case MAT_HERMITIAN:
5934: *flg = mat->hermitian;
5935: break;
5936: case MAT_STRUCTURALLY_SYMMETRIC:
5937: *flg = mat->structurally_symmetric;
5938: break;
5939: case MAT_SYMMETRY_ETERNAL:
5940: *flg = mat->symmetric_eternal;
5941: break;
5942: case MAT_SPD:
5943: *flg = mat->spd;
5944: break;
5945: default:
5946: break;
5947: }
5948: return(0);
5949: }
5951: /*@
5952: MatZeroEntries - Zeros all entries of a matrix. For sparse matrices
5953: this routine retains the old nonzero structure.
5955: Logically Collective on Mat
5957: Input Parameters:
5958: . mat - the matrix
5960: Level: intermediate
5962: Notes:
5963: If the matrix was not preallocated then a default, likely poor preallocation will be set in the matrix, so this should be called after the preallocation phase.
5964: See the Performance chapter of the users manual for information on preallocating matrices.
5966: .seealso: MatZeroRows()
5967: @*/
5968: PetscErrorCode MatZeroEntries(Mat mat)
5969: {
5975: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5976: 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");
5977: if (!mat->ops->zeroentries) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5978: MatCheckPreallocated(mat,1);
5980: PetscLogEventBegin(MAT_ZeroEntries,mat,0,0,0);
5981: (*mat->ops->zeroentries)(mat);
5982: PetscLogEventEnd(MAT_ZeroEntries,mat,0,0,0);
5983: PetscObjectStateIncrease((PetscObject)mat);
5984: return(0);
5985: }
5987: /*@
5988: MatZeroRowsColumns - Zeros all entries (except possibly the main diagonal)
5989: of a set of rows and columns of a matrix.
5991: Collective on Mat
5993: Input Parameters:
5994: + mat - the matrix
5995: . numRows - the number of rows to remove
5996: . rows - the global row indices
5997: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5998: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5999: - b - optional vector of right hand side, that will be adjusted by provided solution
6001: Notes:
6002: This does not change the nonzero structure of the matrix, it merely zeros those entries in the matrix.
6004: The user can set a value in the diagonal entry (or for the AIJ and
6005: row formats can optionally remove the main diagonal entry from the
6006: nonzero structure as well, by passing 0.0 as the final argument).
6008: For the parallel case, all processes that share the matrix (i.e.,
6009: those in the communicator used for matrix creation) MUST call this
6010: routine, regardless of whether any rows being zeroed are owned by
6011: them.
6013: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6014: list only rows local to itself).
6016: The option MAT_NO_OFF_PROC_ZERO_ROWS does not apply to this routine.
6018: Level: intermediate
6020: .seealso: MatZeroRowsIS(), MatZeroRows(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6021: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6022: @*/
6023: PetscErrorCode MatZeroRowsColumns(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6024: {
6031: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6032: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6033: if (!mat->ops->zerorowscolumns) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6034: MatCheckPreallocated(mat,1);
6036: (*mat->ops->zerorowscolumns)(mat,numRows,rows,diag,x,b);
6037: MatViewFromOptions(mat,NULL,"-mat_view");
6038: PetscObjectStateIncrease((PetscObject)mat);
6039: return(0);
6040: }
6042: /*@
6043: MatZeroRowsColumnsIS - Zeros all entries (except possibly the main diagonal)
6044: of a set of rows and columns of a matrix.
6046: Collective on Mat
6048: Input Parameters:
6049: + mat - the matrix
6050: . is - the rows to zero
6051: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
6052: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6053: - b - optional vector of right hand side, that will be adjusted by provided solution
6055: Notes:
6056: This does not change the nonzero structure of the matrix, it merely zeros those entries in the matrix.
6058: The user can set a value in the diagonal entry (or for the AIJ and
6059: row formats can optionally remove the main diagonal entry from the
6060: nonzero structure as well, by passing 0.0 as the final argument).
6062: For the parallel case, all processes that share the matrix (i.e.,
6063: those in the communicator used for matrix creation) MUST call this
6064: routine, regardless of whether any rows being zeroed are owned by
6065: them.
6067: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6068: list only rows local to itself).
6070: The option MAT_NO_OFF_PROC_ZERO_ROWS does not apply to this routine.
6072: Level: intermediate
6074: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6075: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRows(), MatZeroRowsColumnsStencil()
6076: @*/
6077: PetscErrorCode MatZeroRowsColumnsIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6078: {
6080: PetscInt numRows;
6081: const PetscInt *rows;
6088: ISGetLocalSize(is,&numRows);
6089: ISGetIndices(is,&rows);
6090: MatZeroRowsColumns(mat,numRows,rows,diag,x,b);
6091: ISRestoreIndices(is,&rows);
6092: return(0);
6093: }
6095: /*@
6096: MatZeroRows - Zeros all entries (except possibly the main diagonal)
6097: of a set of rows of a matrix.
6099: Collective on Mat
6101: Input Parameters:
6102: + mat - the matrix
6103: . numRows - the number of rows to remove
6104: . rows - the global row indices
6105: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
6106: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6107: - b - optional vector of right hand side, that will be adjusted by provided solution
6109: Notes:
6110: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
6111: but does not release memory. For the dense and block diagonal
6112: formats this does not alter the nonzero structure.
6114: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6115: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6116: merely zeroed.
6118: The user can set a value in the diagonal entry (or for the AIJ and
6119: row formats can optionally remove the main diagonal entry from the
6120: nonzero structure as well, by passing 0.0 as the final argument).
6122: For the parallel case, all processes that share the matrix (i.e.,
6123: those in the communicator used for matrix creation) MUST call this
6124: routine, regardless of whether any rows being zeroed are owned by
6125: them.
6127: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6128: list only rows local to itself).
6130: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6131: owns that are to be zeroed. This saves a global synchronization in the implementation.
6133: Level: intermediate
6135: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6136: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6137: @*/
6138: PetscErrorCode MatZeroRows(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6139: {
6146: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6147: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6148: if (!mat->ops->zerorows) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6149: MatCheckPreallocated(mat,1);
6151: (*mat->ops->zerorows)(mat,numRows,rows,diag,x,b);
6152: MatViewFromOptions(mat,NULL,"-mat_view");
6153: PetscObjectStateIncrease((PetscObject)mat);
6154: return(0);
6155: }
6157: /*@
6158: MatZeroRowsIS - Zeros all entries (except possibly the main diagonal)
6159: of a set of rows of a matrix.
6161: Collective on Mat
6163: Input Parameters:
6164: + mat - the matrix
6165: . is - index set of rows to remove
6166: . diag - value put in all diagonals of eliminated rows
6167: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6168: - b - optional vector of right hand side, that will be adjusted by provided solution
6170: Notes:
6171: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
6172: but does not release memory. For the dense and block diagonal
6173: formats this does not alter the nonzero structure.
6175: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6176: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6177: merely zeroed.
6179: The user can set a value in the diagonal entry (or for the AIJ and
6180: row formats can optionally remove the main diagonal entry from the
6181: nonzero structure as well, by passing 0.0 as the final argument).
6183: For the parallel case, all processes that share the matrix (i.e.,
6184: those in the communicator used for matrix creation) MUST call this
6185: routine, regardless of whether any rows being zeroed are owned by
6186: them.
6188: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6189: list only rows local to itself).
6191: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6192: owns that are to be zeroed. This saves a global synchronization in the implementation.
6194: Level: intermediate
6196: .seealso: MatZeroRows(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6197: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6198: @*/
6199: PetscErrorCode MatZeroRowsIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6200: {
6201: PetscInt numRows;
6202: const PetscInt *rows;
6209: ISGetLocalSize(is,&numRows);
6210: ISGetIndices(is,&rows);
6211: MatZeroRows(mat,numRows,rows,diag,x,b);
6212: ISRestoreIndices(is,&rows);
6213: return(0);
6214: }
6216: /*@
6217: MatZeroRowsStencil - Zeros all entries (except possibly the main diagonal)
6218: of a set of rows of a matrix. These rows must be local to the process.
6220: Collective on Mat
6222: Input Parameters:
6223: + mat - the matrix
6224: . numRows - the number of rows to remove
6225: . rows - the grid coordinates (and component number when dof > 1) for matrix rows
6226: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
6227: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6228: - b - optional vector of right hand side, that will be adjusted by provided solution
6230: Notes:
6231: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
6232: but does not release memory. For the dense and block diagonal
6233: formats this does not alter the nonzero structure.
6235: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6236: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6237: merely zeroed.
6239: The user can set a value in the diagonal entry (or for the AIJ and
6240: row formats can optionally remove the main diagonal entry from the
6241: nonzero structure as well, by passing 0.0 as the final argument).
6243: For the parallel case, all processes that share the matrix (i.e.,
6244: those in the communicator used for matrix creation) MUST call this
6245: routine, regardless of whether any rows being zeroed are owned by
6246: them.
6248: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6249: list only rows local to itself).
6251: The grid coordinates are across the entire grid, not just the local portion
6253: In Fortran idxm and idxn should be declared as
6254: $ MatStencil idxm(4,m)
6255: and the values inserted using
6256: $ idxm(MatStencil_i,1) = i
6257: $ idxm(MatStencil_j,1) = j
6258: $ idxm(MatStencil_k,1) = k
6259: $ idxm(MatStencil_c,1) = c
6260: etc
6262: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
6263: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
6264: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
6265: DM_BOUNDARY_PERIODIC boundary type.
6267: 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
6268: a single value per point) you can skip filling those indices.
6270: Level: intermediate
6272: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsl(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6273: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6274: @*/
6275: PetscErrorCode MatZeroRowsStencil(Mat mat,PetscInt numRows,const MatStencil rows[],PetscScalar diag,Vec x,Vec b)
6276: {
6277: PetscInt dim = mat->stencil.dim;
6278: PetscInt sdim = dim - (1 - (PetscInt) mat->stencil.noc);
6279: PetscInt *dims = mat->stencil.dims+1;
6280: PetscInt *starts = mat->stencil.starts;
6281: PetscInt *dxm = (PetscInt*) rows;
6282: PetscInt *jdxm, i, j, tmp, numNewRows = 0;
6290: PetscMalloc1(numRows, &jdxm);
6291: for (i = 0; i < numRows; ++i) {
6292: /* Skip unused dimensions (they are ordered k, j, i, c) */
6293: for (j = 0; j < 3-sdim; ++j) dxm++;
6294: /* Local index in X dir */
6295: tmp = *dxm++ - starts[0];
6296: /* Loop over remaining dimensions */
6297: for (j = 0; j < dim-1; ++j) {
6298: /* If nonlocal, set index to be negative */
6299: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
6300: /* Update local index */
6301: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
6302: }
6303: /* Skip component slot if necessary */
6304: if (mat->stencil.noc) dxm++;
6305: /* Local row number */
6306: if (tmp >= 0) {
6307: jdxm[numNewRows++] = tmp;
6308: }
6309: }
6310: MatZeroRowsLocal(mat,numNewRows,jdxm,diag,x,b);
6311: PetscFree(jdxm);
6312: return(0);
6313: }
6315: /*@
6316: MatZeroRowsColumnsStencil - Zeros all row and column entries (except possibly the main diagonal)
6317: of a set of rows and columns of a matrix.
6319: Collective on Mat
6321: Input Parameters:
6322: + mat - the matrix
6323: . numRows - the number of rows/columns to remove
6324: . rows - the grid coordinates (and component number when dof > 1) for matrix rows
6325: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
6326: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6327: - b - optional vector of right hand side, that will be adjusted by provided solution
6329: Notes:
6330: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
6331: but does not release memory. For the dense and block diagonal
6332: formats this does not alter the nonzero structure.
6334: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6335: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6336: merely zeroed.
6338: The user can set a value in the diagonal entry (or for the AIJ and
6339: row formats can optionally remove the main diagonal entry from the
6340: nonzero structure as well, by passing 0.0 as the final argument).
6342: For the parallel case, all processes that share the matrix (i.e.,
6343: those in the communicator used for matrix creation) MUST call this
6344: routine, regardless of whether any rows being zeroed are owned by
6345: them.
6347: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
6348: list only rows local to itself, but the row/column numbers are given in local numbering).
6350: The grid coordinates are across the entire grid, not just the local portion
6352: In Fortran idxm and idxn should be declared as
6353: $ MatStencil idxm(4,m)
6354: and the values inserted using
6355: $ idxm(MatStencil_i,1) = i
6356: $ idxm(MatStencil_j,1) = j
6357: $ idxm(MatStencil_k,1) = k
6358: $ idxm(MatStencil_c,1) = c
6359: etc
6361: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
6362: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
6363: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
6364: DM_BOUNDARY_PERIODIC boundary type.
6366: 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
6367: a single value per point) you can skip filling those indices.
6369: Level: intermediate
6371: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6372: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRows()
6373: @*/
6374: PetscErrorCode MatZeroRowsColumnsStencil(Mat mat,PetscInt numRows,const MatStencil rows[],PetscScalar diag,Vec x,Vec b)
6375: {
6376: PetscInt dim = mat->stencil.dim;
6377: PetscInt sdim = dim - (1 - (PetscInt) mat->stencil.noc);
6378: PetscInt *dims = mat->stencil.dims+1;
6379: PetscInt *starts = mat->stencil.starts;
6380: PetscInt *dxm = (PetscInt*) rows;
6381: PetscInt *jdxm, i, j, tmp, numNewRows = 0;
6389: PetscMalloc1(numRows, &jdxm);
6390: for (i = 0; i < numRows; ++i) {
6391: /* Skip unused dimensions (they are ordered k, j, i, c) */
6392: for (j = 0; j < 3-sdim; ++j) dxm++;
6393: /* Local index in X dir */
6394: tmp = *dxm++ - starts[0];
6395: /* Loop over remaining dimensions */
6396: for (j = 0; j < dim-1; ++j) {
6397: /* If nonlocal, set index to be negative */
6398: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
6399: /* Update local index */
6400: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
6401: }
6402: /* Skip component slot if necessary */
6403: if (mat->stencil.noc) dxm++;
6404: /* Local row number */
6405: if (tmp >= 0) {
6406: jdxm[numNewRows++] = tmp;
6407: }
6408: }
6409: MatZeroRowsColumnsLocal(mat,numNewRows,jdxm,diag,x,b);
6410: PetscFree(jdxm);
6411: return(0);
6412: }
6414: /*@C
6415: MatZeroRowsLocal - Zeros all entries (except possibly the main diagonal)
6416: of a set of rows of a matrix; using local numbering of rows.
6418: Collective on Mat
6420: Input Parameters:
6421: + mat - the matrix
6422: . numRows - the number of rows to remove
6423: . rows - the global row indices
6424: . diag - value put in all diagonals of eliminated rows
6425: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6426: - b - optional vector of right hand side, that will be adjusted by provided solution
6428: Notes:
6429: Before calling MatZeroRowsLocal(), the user must first set the
6430: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6432: For the AIJ matrix formats this removes the old nonzero structure,
6433: but does not release memory. For the dense and block diagonal
6434: formats this does not alter the nonzero structure.
6436: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6437: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6438: merely zeroed.
6440: The user can set a value in the diagonal entry (or for the AIJ and
6441: row formats can optionally remove the main diagonal entry from the
6442: nonzero structure as well, by passing 0.0 as the final argument).
6444: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6445: owns that are to be zeroed. This saves a global synchronization in the implementation.
6447: Level: intermediate
6449: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRows(), MatSetOption(),
6450: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6451: @*/
6452: PetscErrorCode MatZeroRowsLocal(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6453: {
6460: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6461: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6462: MatCheckPreallocated(mat,1);
6464: if (mat->ops->zerorowslocal) {
6465: (*mat->ops->zerorowslocal)(mat,numRows,rows,diag,x,b);
6466: } else {
6467: IS is, newis;
6468: const PetscInt *newRows;
6470: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Need to provide local to global mapping to matrix first");
6471: ISCreateGeneral(PETSC_COMM_SELF,numRows,rows,PETSC_COPY_VALUES,&is);
6472: ISLocalToGlobalMappingApplyIS(mat->rmap->mapping,is,&newis);
6473: ISGetIndices(newis,&newRows);
6474: (*mat->ops->zerorows)(mat,numRows,newRows,diag,x,b);
6475: ISRestoreIndices(newis,&newRows);
6476: ISDestroy(&newis);
6477: ISDestroy(&is);
6478: }
6479: PetscObjectStateIncrease((PetscObject)mat);
6480: return(0);
6481: }
6483: /*@
6484: MatZeroRowsLocalIS - Zeros all entries (except possibly the main diagonal)
6485: of a set of rows of a matrix; using local numbering of rows.
6487: Collective on Mat
6489: Input Parameters:
6490: + mat - the matrix
6491: . is - index set of rows to remove
6492: . diag - value put in all diagonals of eliminated rows
6493: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6494: - b - optional vector of right hand side, that will be adjusted by provided solution
6496: Notes:
6497: Before calling MatZeroRowsLocalIS(), the user must first set the
6498: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6500: For the AIJ matrix formats this removes the old nonzero structure,
6501: but does not release memory. For the dense and block diagonal
6502: formats this does not alter the nonzero structure.
6504: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6505: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6506: merely zeroed.
6508: The user can set a value in the diagonal entry (or for the AIJ and
6509: row formats can optionally remove the main diagonal entry from the
6510: nonzero structure as well, by passing 0.0 as the final argument).
6512: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6513: owns that are to be zeroed. This saves a global synchronization in the implementation.
6515: Level: intermediate
6517: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRows(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6518: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6519: @*/
6520: PetscErrorCode MatZeroRowsLocalIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6521: {
6523: PetscInt numRows;
6524: const PetscInt *rows;
6530: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6531: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6532: MatCheckPreallocated(mat,1);
6534: ISGetLocalSize(is,&numRows);
6535: ISGetIndices(is,&rows);
6536: MatZeroRowsLocal(mat,numRows,rows,diag,x,b);
6537: ISRestoreIndices(is,&rows);
6538: return(0);
6539: }
6541: /*@
6542: MatZeroRowsColumnsLocal - Zeros all entries (except possibly the main diagonal)
6543: of a set of rows and columns of a matrix; using local numbering of rows.
6545: Collective on Mat
6547: Input Parameters:
6548: + mat - the matrix
6549: . numRows - the number of rows to remove
6550: . rows - the global row indices
6551: . diag - value put in all diagonals of eliminated rows
6552: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6553: - b - optional vector of right hand side, that will be adjusted by provided solution
6555: Notes:
6556: Before calling MatZeroRowsColumnsLocal(), the user must first set the
6557: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6559: The user can set a value in the diagonal entry (or for the AIJ and
6560: row formats can optionally remove the main diagonal entry from the
6561: nonzero structure as well, by passing 0.0 as the final argument).
6563: Level: intermediate
6565: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6566: MatZeroRows(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6567: @*/
6568: PetscErrorCode MatZeroRowsColumnsLocal(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6569: {
6571: IS is, newis;
6572: const PetscInt *newRows;
6578: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6579: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6580: MatCheckPreallocated(mat,1);
6582: if (!mat->cmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Need to provide local to global mapping to matrix first");
6583: ISCreateGeneral(PETSC_COMM_SELF,numRows,rows,PETSC_COPY_VALUES,&is);
6584: ISLocalToGlobalMappingApplyIS(mat->cmap->mapping,is,&newis);
6585: ISGetIndices(newis,&newRows);
6586: (*mat->ops->zerorowscolumns)(mat,numRows,newRows,diag,x,b);
6587: ISRestoreIndices(newis,&newRows);
6588: ISDestroy(&newis);
6589: ISDestroy(&is);
6590: PetscObjectStateIncrease((PetscObject)mat);
6591: return(0);
6592: }
6594: /*@
6595: MatZeroRowsColumnsLocalIS - Zeros all entries (except possibly the main diagonal)
6596: of a set of rows and columns of a matrix; using local numbering of rows.
6598: Collective on Mat
6600: Input Parameters:
6601: + mat - the matrix
6602: . is - index set of rows to remove
6603: . diag - value put in all diagonals of eliminated rows
6604: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6605: - b - optional vector of right hand side, that will be adjusted by provided solution
6607: Notes:
6608: Before calling MatZeroRowsColumnsLocalIS(), the user must first set the
6609: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6611: The user can set a value in the diagonal entry (or for the AIJ and
6612: row formats can optionally remove the main diagonal entry from the
6613: nonzero structure as well, by passing 0.0 as the final argument).
6615: Level: intermediate
6617: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6618: MatZeroRowsColumnsLocal(), MatZeroRows(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6619: @*/
6620: PetscErrorCode MatZeroRowsColumnsLocalIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6621: {
6623: PetscInt numRows;
6624: const PetscInt *rows;
6630: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6631: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6632: MatCheckPreallocated(mat,1);
6634: ISGetLocalSize(is,&numRows);
6635: ISGetIndices(is,&rows);
6636: MatZeroRowsColumnsLocal(mat,numRows,rows,diag,x,b);
6637: ISRestoreIndices(is,&rows);
6638: return(0);
6639: }
6641: /*@C
6642: MatGetSize - Returns the numbers of rows and columns in a matrix.
6644: Not Collective
6646: Input Parameter:
6647: . mat - the matrix
6649: Output Parameters:
6650: + m - the number of global rows
6651: - n - the number of global columns
6653: Note: both output parameters can be NULL on input.
6655: Level: beginner
6657: .seealso: MatGetLocalSize()
6658: @*/
6659: PetscErrorCode MatGetSize(Mat mat,PetscInt *m,PetscInt *n)
6660: {
6663: if (m) *m = mat->rmap->N;
6664: if (n) *n = mat->cmap->N;
6665: return(0);
6666: }
6668: /*@C
6669: MatGetLocalSize - Returns the number of local rows and local columns
6670: of a matrix, that is the local size of the left and right vectors as returned by MatCreateVecs().
6672: Not Collective
6674: Input Parameters:
6675: . mat - the matrix
6677: Output Parameters:
6678: + m - the number of local rows
6679: - n - the number of local columns
6681: Note: both output parameters can be NULL on input.
6683: Level: beginner
6685: .seealso: MatGetSize()
6686: @*/
6687: PetscErrorCode MatGetLocalSize(Mat mat,PetscInt *m,PetscInt *n)
6688: {
6693: if (m) *m = mat->rmap->n;
6694: if (n) *n = mat->cmap->n;
6695: return(0);
6696: }
6698: /*@C
6699: MatGetOwnershipRangeColumn - Returns the range of matrix columns associated with rows of a vector one multiplies by that owned by
6700: this processor. (The columns of the "diagonal block")
6702: Not Collective, unless matrix has not been allocated, then collective on Mat
6704: Input Parameters:
6705: . mat - the matrix
6707: Output Parameters:
6708: + m - the global index of the first local column
6709: - n - one more than the global index of the last local column
6711: Notes:
6712: both output parameters can be NULL on input.
6714: Level: developer
6716: .seealso: MatGetOwnershipRange(), MatGetOwnershipRanges(), MatGetOwnershipRangesColumn()
6718: @*/
6719: PetscErrorCode MatGetOwnershipRangeColumn(Mat mat,PetscInt *m,PetscInt *n)
6720: {
6726: MatCheckPreallocated(mat,1);
6727: if (m) *m = mat->cmap->rstart;
6728: if (n) *n = mat->cmap->rend;
6729: return(0);
6730: }
6732: /*@C
6733: MatGetOwnershipRange - Returns the range of matrix rows owned by
6734: this processor, assuming that the matrix is laid out with the first
6735: n1 rows on the first processor, the next n2 rows on the second, etc.
6736: For certain parallel layouts this range may not be well defined.
6738: Not Collective
6740: Input Parameters:
6741: . mat - the matrix
6743: Output Parameters:
6744: + m - the global index of the first local row
6745: - n - one more than the global index of the last local row
6747: Note: Both output parameters can be NULL on input.
6748: $ This function requires that the matrix be preallocated. If you have not preallocated, consider using
6749: $ PetscSplitOwnership(MPI_Comm comm, PetscInt *n, PetscInt *N)
6750: $ and then MPI_Scan() to calculate prefix sums of the local sizes.
6752: Level: beginner
6754: .seealso: MatGetOwnershipRanges(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn(), PetscSplitOwnership(), PetscSplitOwnershipBlock()
6756: @*/
6757: PetscErrorCode MatGetOwnershipRange(Mat mat,PetscInt *m,PetscInt *n)
6758: {
6764: MatCheckPreallocated(mat,1);
6765: if (m) *m = mat->rmap->rstart;
6766: if (n) *n = mat->rmap->rend;
6767: return(0);
6768: }
6770: /*@C
6771: MatGetOwnershipRanges - Returns the range of matrix rows owned by
6772: each process
6774: Not Collective, unless matrix has not been allocated, then collective on Mat
6776: Input Parameters:
6777: . mat - the matrix
6779: Output Parameters:
6780: . ranges - start of each processors portion plus one more than the total length at the end
6782: Level: beginner
6784: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn()
6786: @*/
6787: PetscErrorCode MatGetOwnershipRanges(Mat mat,const PetscInt **ranges)
6788: {
6794: MatCheckPreallocated(mat,1);
6795: PetscLayoutGetRanges(mat->rmap,ranges);
6796: return(0);
6797: }
6799: /*@C
6800: MatGetOwnershipRangesColumn - Returns the range of matrix columns associated with rows of a vector one multiplies by that owned by
6801: this processor. (The columns of the "diagonal blocks" for each process)
6803: Not Collective, unless matrix has not been allocated, then collective on Mat
6805: Input Parameters:
6806: . mat - the matrix
6808: Output Parameters:
6809: . ranges - start of each processors portion plus one more then the total length at the end
6811: Level: beginner
6813: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRanges()
6815: @*/
6816: PetscErrorCode MatGetOwnershipRangesColumn(Mat mat,const PetscInt **ranges)
6817: {
6823: MatCheckPreallocated(mat,1);
6824: PetscLayoutGetRanges(mat->cmap,ranges);
6825: return(0);
6826: }
6828: /*@C
6829: MatGetOwnershipIS - Get row and column ownership as index sets
6831: Not Collective
6833: Input Arguments:
6834: . A - matrix of type Elemental or ScaLAPACK
6836: Output Arguments:
6837: + rows - rows in which this process owns elements
6838: - cols - columns in which this process owns elements
6840: Level: intermediate
6842: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatSetValues(), MATELEMENTAL
6843: @*/
6844: PetscErrorCode MatGetOwnershipIS(Mat A,IS *rows,IS *cols)
6845: {
6846: PetscErrorCode ierr,(*f)(Mat,IS*,IS*);
6849: MatCheckPreallocated(A,1);
6850: PetscObjectQueryFunction((PetscObject)A,"MatGetOwnershipIS_C",&f);
6851: if (f) {
6852: (*f)(A,rows,cols);
6853: } else { /* Create a standard row-based partition, each process is responsible for ALL columns in their row block */
6854: if (rows) {ISCreateStride(PETSC_COMM_SELF,A->rmap->n,A->rmap->rstart,1,rows);}
6855: if (cols) {ISCreateStride(PETSC_COMM_SELF,A->cmap->N,0,1,cols);}
6856: }
6857: return(0);
6858: }
6860: /*@C
6861: MatILUFactorSymbolic - Performs symbolic ILU factorization of a matrix.
6862: Uses levels of fill only, not drop tolerance. Use MatLUFactorNumeric()
6863: to complete the factorization.
6865: Collective on Mat
6867: Input Parameters:
6868: + mat - the matrix
6869: . row - row permutation
6870: . column - column permutation
6871: - info - structure containing
6872: $ levels - number of levels of fill.
6873: $ expected fill - as ratio of original fill.
6874: $ 1 or 0 - indicating force fill on diagonal (improves robustness for matrices
6875: missing diagonal entries)
6877: Output Parameters:
6878: . fact - new matrix that has been symbolically factored
6880: Notes:
6881: See Users-Manual: ch_mat for additional information about choosing the fill factor for better efficiency.
6883: Most users should employ the simplified KSP interface for linear solvers
6884: instead of working directly with matrix algebra routines such as this.
6885: See, e.g., KSPCreate().
6887: Level: developer
6889: .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
6890: MatGetOrdering(), MatFactorInfo
6892: Note: this uses the definition of level of fill as in Y. Saad, 2003
6894: Developer Note: fortran interface is not autogenerated as the f90
6895: interface defintion cannot be generated correctly [due to MatFactorInfo]
6897: References:
6898: Y. Saad, Iterative methods for sparse linear systems Philadelphia: Society for Industrial and Applied Mathematics, 2003
6899: @*/
6900: PetscErrorCode MatILUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
6901: {
6911: if (info->levels < 0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Levels of fill negative %D",(PetscInt)info->levels);
6912: if (info->fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %g",(double)info->fill);
6913: if (!fact->ops->ilufactorsymbolic) {
6914: MatSolverType stype;
6915: MatFactorGetSolverType(fact,&stype);
6916: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic ILU using solver type %s",((PetscObject)mat)->type_name,stype);
6917: }
6918: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6919: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6920: MatCheckPreallocated(mat,2);
6922: PetscLogEventBegin(MAT_ILUFactorSymbolic,mat,row,col,0);
6923: (fact->ops->ilufactorsymbolic)(fact,mat,row,col,info);
6924: PetscLogEventEnd(MAT_ILUFactorSymbolic,mat,row,col,0);
6925: return(0);
6926: }
6928: /*@C
6929: MatICCFactorSymbolic - Performs symbolic incomplete
6930: Cholesky factorization for a symmetric matrix. Use
6931: MatCholeskyFactorNumeric() to complete the factorization.
6933: Collective on Mat
6935: Input Parameters:
6936: + mat - the matrix
6937: . perm - row and column permutation
6938: - info - structure containing
6939: $ levels - number of levels of fill.
6940: $ expected fill - as ratio of original fill.
6942: Output Parameter:
6943: . fact - the factored matrix
6945: Notes:
6946: Most users should employ the KSP interface for linear solvers
6947: instead of working directly with matrix algebra routines such as this.
6948: See, e.g., KSPCreate().
6950: Level: developer
6952: .seealso: MatCholeskyFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
6954: Note: this uses the definition of level of fill as in Y. Saad, 2003
6956: Developer Note: fortran interface is not autogenerated as the f90
6957: interface defintion cannot be generated correctly [due to MatFactorInfo]
6959: References:
6960: Y. Saad, Iterative methods for sparse linear systems Philadelphia: Society for Industrial and Applied Mathematics, 2003
6961: @*/
6962: PetscErrorCode MatICCFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
6963: {
6972: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6973: if (info->levels < 0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Levels negative %D",(PetscInt) info->levels);
6974: if (info->fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %g",(double)info->fill);
6975: if (!(fact)->ops->iccfactorsymbolic) {
6976: MatSolverType stype;
6977: MatFactorGetSolverType(fact,&stype);
6978: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic ICC using solver type %s",((PetscObject)mat)->type_name,stype);
6979: }
6980: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6981: MatCheckPreallocated(mat,2);
6983: PetscLogEventBegin(MAT_ICCFactorSymbolic,mat,perm,0,0);
6984: (fact->ops->iccfactorsymbolic)(fact,mat,perm,info);
6985: PetscLogEventEnd(MAT_ICCFactorSymbolic,mat,perm,0,0);
6986: return(0);
6987: }
6989: /*@C
6990: MatCreateSubMatrices - Extracts several submatrices from a matrix. If submat
6991: points to an array of valid matrices, they may be reused to store the new
6992: submatrices.
6994: Collective on Mat
6996: Input Parameters:
6997: + mat - the matrix
6998: . n - the number of submatrixes to be extracted (on this processor, may be zero)
6999: . irow, icol - index sets of rows and columns to extract
7000: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7002: Output Parameter:
7003: . submat - the array of submatrices
7005: Notes:
7006: MatCreateSubMatrices() can extract ONLY sequential submatrices
7007: (from both sequential and parallel matrices). Use MatCreateSubMatrix()
7008: to extract a parallel submatrix.
7010: Some matrix types place restrictions on the row and column
7011: indices, such as that they be sorted or that they be equal to each other.
7013: The index sets may not have duplicate entries.
7015: When extracting submatrices from a parallel matrix, each processor can
7016: form a different submatrix by setting the rows and columns of its
7017: individual index sets according to the local submatrix desired.
7019: When finished using the submatrices, the user should destroy
7020: them with MatDestroySubMatrices().
7022: MAT_REUSE_MATRIX can only be used when the nonzero structure of the
7023: original matrix has not changed from that last call to MatCreateSubMatrices().
7025: This routine creates the matrices in submat; you should NOT create them before
7026: calling it. It also allocates the array of matrix pointers submat.
7028: For BAIJ matrices the index sets must respect the block structure, that is if they
7029: request one row/column in a block, they must request all rows/columns that are in
7030: that block. For example, if the block size is 2 you cannot request just row 0 and
7031: column 0.
7033: Fortran Note:
7034: The Fortran interface is slightly different from that given below; it
7035: requires one to pass in as submat a Mat (integer) array of size at least n+1.
7037: Level: advanced
7040: .seealso: MatDestroySubMatrices(), MatCreateSubMatrix(), MatGetRow(), MatGetDiagonal(), MatReuse
7041: @*/
7042: PetscErrorCode MatCreateSubMatrices(Mat mat,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *submat[])
7043: {
7045: PetscInt i;
7046: PetscBool eq;
7051: if (n) {
7056: }
7058: if (n && scall == MAT_REUSE_MATRIX) {
7061: }
7062: if (!mat->ops->createsubmatrices) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7063: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7064: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7065: MatCheckPreallocated(mat,1);
7067: PetscLogEventBegin(MAT_CreateSubMats,mat,0,0,0);
7068: (*mat->ops->createsubmatrices)(mat,n,irow,icol,scall,submat);
7069: PetscLogEventEnd(MAT_CreateSubMats,mat,0,0,0);
7070: for (i=0; i<n; i++) {
7071: (*submat)[i]->factortype = MAT_FACTOR_NONE; /* in case in place factorization was previously done on submatrix */
7072: ISEqualUnsorted(irow[i],icol[i],&eq);
7073: if (eq) {
7074: MatPropagateSymmetryOptions(mat,(*submat)[i]);
7075: }
7076: }
7077: return(0);
7078: }
7080: /*@C
7081: MatCreateSubMatricesMPI - Extracts MPI submatrices across a sub communicator of mat (by pairs of IS that may live on subcomms).
7083: Collective on Mat
7085: Input Parameters:
7086: + mat - the matrix
7087: . n - the number of submatrixes to be extracted
7088: . irow, icol - index sets of rows and columns to extract
7089: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7091: Output Parameter:
7092: . submat - the array of submatrices
7094: Level: advanced
7097: .seealso: MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRow(), MatGetDiagonal(), MatReuse
7098: @*/
7099: PetscErrorCode MatCreateSubMatricesMPI(Mat mat,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *submat[])
7100: {
7102: PetscInt i;
7103: PetscBool eq;
7108: if (n) {
7113: }
7115: if (n && scall == MAT_REUSE_MATRIX) {
7118: }
7119: if (!mat->ops->createsubmatricesmpi) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7120: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7121: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7122: MatCheckPreallocated(mat,1);
7124: PetscLogEventBegin(MAT_CreateSubMats,mat,0,0,0);
7125: (*mat->ops->createsubmatricesmpi)(mat,n,irow,icol,scall,submat);
7126: PetscLogEventEnd(MAT_CreateSubMats,mat,0,0,0);
7127: for (i=0; i<n; i++) {
7128: ISEqualUnsorted(irow[i],icol[i],&eq);
7129: if (eq) {
7130: MatPropagateSymmetryOptions(mat,(*submat)[i]);
7131: }
7132: }
7133: return(0);
7134: }
7136: /*@C
7137: MatDestroyMatrices - Destroys an array of matrices.
7139: Collective on Mat
7141: Input Parameters:
7142: + n - the number of local matrices
7143: - mat - the matrices (note that this is a pointer to the array of matrices)
7145: Level: advanced
7147: Notes:
7148: Frees not only the matrices, but also the array that contains the matrices
7149: In Fortran will not free the array.
7151: .seealso: MatCreateSubMatrices() MatDestroySubMatrices()
7152: @*/
7153: PetscErrorCode MatDestroyMatrices(PetscInt n,Mat *mat[])
7154: {
7156: PetscInt i;
7159: if (!*mat) return(0);
7160: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Trying to destroy negative number of matrices %D",n);
7163: for (i=0; i<n; i++) {
7164: MatDestroy(&(*mat)[i]);
7165: }
7167: /* memory is allocated even if n = 0 */
7168: PetscFree(*mat);
7169: return(0);
7170: }
7172: /*@C
7173: MatDestroySubMatrices - Destroys a set of matrices obtained with MatCreateSubMatrices().
7175: Collective on Mat
7177: Input Parameters:
7178: + n - the number of local matrices
7179: - mat - the matrices (note that this is a pointer to the array of matrices, just to match the calling
7180: sequence of MatCreateSubMatrices())
7182: Level: advanced
7184: Notes:
7185: Frees not only the matrices, but also the array that contains the matrices
7186: In Fortran will not free the array.
7188: .seealso: MatCreateSubMatrices()
7189: @*/
7190: PetscErrorCode MatDestroySubMatrices(PetscInt n,Mat *mat[])
7191: {
7193: Mat mat0;
7196: if (!*mat) return(0);
7197: /* mat[] is an array of length n+1, see MatCreateSubMatrices_xxx() */
7198: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Trying to destroy negative number of matrices %D",n);
7201: mat0 = (*mat)[0];
7202: if (mat0 && mat0->ops->destroysubmatrices) {
7203: (mat0->ops->destroysubmatrices)(n,mat);
7204: } else {
7205: MatDestroyMatrices(n,mat);
7206: }
7207: return(0);
7208: }
7210: /*@C
7211: MatGetSeqNonzeroStructure - Extracts the sequential nonzero structure from a matrix.
7213: Collective on Mat
7215: Input Parameters:
7216: . mat - the matrix
7218: Output Parameter:
7219: . matstruct - the sequential matrix with the nonzero structure of mat
7221: Level: intermediate
7223: .seealso: MatDestroySeqNonzeroStructure(), MatCreateSubMatrices(), MatDestroyMatrices()
7224: @*/
7225: PetscErrorCode MatGetSeqNonzeroStructure(Mat mat,Mat *matstruct)
7226: {
7234: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7235: MatCheckPreallocated(mat,1);
7237: if (!mat->ops->getseqnonzerostructure) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Not for matrix type %s\n",((PetscObject)mat)->type_name);
7238: PetscLogEventBegin(MAT_GetSeqNonzeroStructure,mat,0,0,0);
7239: (*mat->ops->getseqnonzerostructure)(mat,matstruct);
7240: PetscLogEventEnd(MAT_GetSeqNonzeroStructure,mat,0,0,0);
7241: return(0);
7242: }
7244: /*@C
7245: MatDestroySeqNonzeroStructure - Destroys matrix obtained with MatGetSeqNonzeroStructure().
7247: Collective on Mat
7249: Input Parameters:
7250: . mat - the matrix (note that this is a pointer to the array of matrices, just to match the calling
7251: sequence of MatGetSequentialNonzeroStructure())
7253: Level: advanced
7255: Notes:
7256: Frees not only the matrices, but also the array that contains the matrices
7258: .seealso: MatGetSeqNonzeroStructure()
7259: @*/
7260: PetscErrorCode MatDestroySeqNonzeroStructure(Mat *mat)
7261: {
7266: MatDestroy(mat);
7267: return(0);
7268: }
7270: /*@
7271: MatIncreaseOverlap - Given a set of submatrices indicated by index sets,
7272: replaces the index sets by larger ones that represent submatrices with
7273: additional overlap.
7275: Collective on Mat
7277: Input Parameters:
7278: + mat - the matrix
7279: . n - the number of index sets
7280: . is - the array of index sets (these index sets will changed during the call)
7281: - ov - the additional overlap requested
7283: Options Database:
7284: . -mat_increase_overlap_scalable - use a scalable algorithm to compute the overlap (supported by MPIAIJ matrix)
7286: Level: developer
7289: .seealso: MatCreateSubMatrices()
7290: @*/
7291: PetscErrorCode MatIncreaseOverlap(Mat mat,PetscInt n,IS is[],PetscInt ov)
7292: {
7298: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Must have one or more domains, you have %D",n);
7299: if (n) {
7302: }
7303: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7304: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7305: MatCheckPreallocated(mat,1);
7307: if (!ov) return(0);
7308: if (!mat->ops->increaseoverlap) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7309: PetscLogEventBegin(MAT_IncreaseOverlap,mat,0,0,0);
7310: (*mat->ops->increaseoverlap)(mat,n,is,ov);
7311: PetscLogEventEnd(MAT_IncreaseOverlap,mat,0,0,0);
7312: return(0);
7313: }
7316: PetscErrorCode MatIncreaseOverlapSplit_Single(Mat,IS*,PetscInt);
7318: /*@
7319: MatIncreaseOverlapSplit - Given a set of submatrices indicated by index sets across
7320: a sub communicator, replaces the index sets by larger ones that represent submatrices with
7321: additional overlap.
7323: Collective on Mat
7325: Input Parameters:
7326: + mat - the matrix
7327: . n - the number of index sets
7328: . is - the array of index sets (these index sets will changed during the call)
7329: - ov - the additional overlap requested
7331: Options Database:
7332: . -mat_increase_overlap_scalable - use a scalable algorithm to compute the overlap (supported by MPIAIJ matrix)
7334: Level: developer
7337: .seealso: MatCreateSubMatrices()
7338: @*/
7339: PetscErrorCode MatIncreaseOverlapSplit(Mat mat,PetscInt n,IS is[],PetscInt ov)
7340: {
7341: PetscInt i;
7347: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Must have one or more domains, you have %D",n);
7348: if (n) {
7351: }
7352: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
7353: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7354: MatCheckPreallocated(mat,1);
7355: if (!ov) return(0);
7356: PetscLogEventBegin(MAT_IncreaseOverlap,mat,0,0,0);
7357: for (i=0; i<n; i++){
7358: MatIncreaseOverlapSplit_Single(mat,&is[i],ov);
7359: }
7360: PetscLogEventEnd(MAT_IncreaseOverlap,mat,0,0,0);
7361: return(0);
7362: }
7367: /*@
7368: MatGetBlockSize - Returns the matrix block size.
7370: Not Collective
7372: Input Parameter:
7373: . mat - the matrix
7375: Output Parameter:
7376: . bs - block size
7378: Notes:
7379: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7381: If the block size has not been set yet this routine returns 1.
7383: Level: intermediate
7385: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSizes()
7386: @*/
7387: PetscErrorCode MatGetBlockSize(Mat mat,PetscInt *bs)
7388: {
7392: *bs = PetscAbs(mat->rmap->bs);
7393: return(0);
7394: }
7396: /*@
7397: MatGetBlockSizes - Returns the matrix block row and column sizes.
7399: Not Collective
7401: Input Parameter:
7402: . mat - the matrix
7404: Output Parameter:
7405: + rbs - row block size
7406: - cbs - column block size
7408: Notes:
7409: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7410: If you pass a different block size for the columns than the rows, the row block size determines the square block storage.
7412: If a block size has not been set yet this routine returns 1.
7414: Level: intermediate
7416: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSize(), MatSetBlockSizes()
7417: @*/
7418: PetscErrorCode MatGetBlockSizes(Mat mat,PetscInt *rbs, PetscInt *cbs)
7419: {
7424: if (rbs) *rbs = PetscAbs(mat->rmap->bs);
7425: if (cbs) *cbs = PetscAbs(mat->cmap->bs);
7426: return(0);
7427: }
7429: /*@
7430: MatSetBlockSize - Sets the matrix block size.
7432: Logically Collective on Mat
7434: Input Parameters:
7435: + mat - the matrix
7436: - bs - block size
7438: Notes:
7439: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7440: This must be called before MatSetUp() or MatXXXSetPreallocation() (or will default to 1) and the block size cannot be changed later.
7442: For MATMPIAIJ and MATSEQAIJ matrix formats, this function can be called at a later stage, provided that the specified block size
7443: is compatible with the matrix local sizes.
7445: Level: intermediate
7447: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes()
7448: @*/
7449: PetscErrorCode MatSetBlockSize(Mat mat,PetscInt bs)
7450: {
7456: MatSetBlockSizes(mat,bs,bs);
7457: return(0);
7458: }
7460: /*@
7461: MatSetVariableBlockSizes - Sets a diagonal blocks of the matrix that need not be of the same size
7463: Logically Collective on Mat
7465: Input Parameters:
7466: + mat - the matrix
7467: . nblocks - the number of blocks on this process
7468: - bsizes - the block sizes
7470: Notes:
7471: Currently used by PCVPBJACOBI for SeqAIJ matrices
7473: Level: intermediate
7475: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes(), MatGetVariableBlockSizes()
7476: @*/
7477: PetscErrorCode MatSetVariableBlockSizes(Mat mat,PetscInt nblocks,PetscInt *bsizes)
7478: {
7480: PetscInt i,ncnt = 0, nlocal;
7484: if (nblocks < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Number of local blocks must be great than or equal to zero");
7485: MatGetLocalSize(mat,&nlocal,NULL);
7486: for (i=0; i<nblocks; i++) ncnt += bsizes[i];
7487: if (ncnt != nlocal) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Sum of local block sizes %D does not equal local size of matrix %D",ncnt,nlocal);
7488: PetscFree(mat->bsizes);
7489: mat->nblocks = nblocks;
7490: PetscMalloc1(nblocks,&mat->bsizes);
7491: PetscArraycpy(mat->bsizes,bsizes,nblocks);
7492: return(0);
7493: }
7495: /*@C
7496: MatGetVariableBlockSizes - Gets a diagonal blocks of the matrix that need not be of the same size
7498: Logically Collective on Mat
7500: Input Parameters:
7501: . mat - the matrix
7503: Output Parameters:
7504: + nblocks - the number of blocks on this process
7505: - bsizes - the block sizes
7507: Notes: Currently not supported from Fortran
7509: Level: intermediate
7511: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes(), MatSetVariableBlockSizes()
7512: @*/
7513: PetscErrorCode MatGetVariableBlockSizes(Mat mat,PetscInt *nblocks,const PetscInt **bsizes)
7514: {
7517: *nblocks = mat->nblocks;
7518: *bsizes = mat->bsizes;
7519: return(0);
7520: }
7522: /*@
7523: MatSetBlockSizes - Sets the matrix block row and column sizes.
7525: Logically Collective on Mat
7527: Input Parameters:
7528: + mat - the matrix
7529: . rbs - row block size
7530: - cbs - column block size
7532: Notes:
7533: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7534: If you pass a different block size for the columns than the rows, the row block size determines the square block storage.
7535: This must be called before MatSetUp() or MatXXXSetPreallocation() (or will default to 1) and the block size cannot be changed later.
7537: For MATMPIAIJ and MATSEQAIJ matrix formats, this function can be called at a later stage, provided that the specified block sizes
7538: are compatible with the matrix local sizes.
7540: The row and column block size determine the blocksize of the "row" and "column" vectors returned by MatCreateVecs().
7542: Level: intermediate
7544: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSize(), MatGetBlockSizes()
7545: @*/
7546: PetscErrorCode MatSetBlockSizes(Mat mat,PetscInt rbs,PetscInt cbs)
7547: {
7554: if (mat->ops->setblocksizes) {
7555: (*mat->ops->setblocksizes)(mat,rbs,cbs);
7556: }
7557: if (mat->rmap->refcnt) {
7558: ISLocalToGlobalMapping l2g = NULL;
7559: PetscLayout nmap = NULL;
7561: PetscLayoutDuplicate(mat->rmap,&nmap);
7562: if (mat->rmap->mapping) {
7563: ISLocalToGlobalMappingDuplicate(mat->rmap->mapping,&l2g);
7564: }
7565: PetscLayoutDestroy(&mat->rmap);
7566: mat->rmap = nmap;
7567: mat->rmap->mapping = l2g;
7568: }
7569: if (mat->cmap->refcnt) {
7570: ISLocalToGlobalMapping l2g = NULL;
7571: PetscLayout nmap = NULL;
7573: PetscLayoutDuplicate(mat->cmap,&nmap);
7574: if (mat->cmap->mapping) {
7575: ISLocalToGlobalMappingDuplicate(mat->cmap->mapping,&l2g);
7576: }
7577: PetscLayoutDestroy(&mat->cmap);
7578: mat->cmap = nmap;
7579: mat->cmap->mapping = l2g;
7580: }
7581: PetscLayoutSetBlockSize(mat->rmap,rbs);
7582: PetscLayoutSetBlockSize(mat->cmap,cbs);
7583: return(0);
7584: }
7586: /*@
7587: MatSetBlockSizesFromMats - Sets the matrix block row and column sizes to match a pair of matrices
7589: Logically Collective on Mat
7591: Input Parameters:
7592: + mat - the matrix
7593: . fromRow - matrix from which to copy row block size
7594: - fromCol - matrix from which to copy column block size (can be same as fromRow)
7596: Level: developer
7598: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes()
7599: @*/
7600: PetscErrorCode MatSetBlockSizesFromMats(Mat mat,Mat fromRow,Mat fromCol)
7601: {
7608: if (fromRow->rmap->bs > 0) {PetscLayoutSetBlockSize(mat->rmap,fromRow->rmap->bs);}
7609: if (fromCol->cmap->bs > 0) {PetscLayoutSetBlockSize(mat->cmap,fromCol->cmap->bs);}
7610: return(0);
7611: }
7613: /*@
7614: MatResidual - Default routine to calculate the residual.
7616: Collective on Mat
7618: Input Parameters:
7619: + mat - the matrix
7620: . b - the right-hand-side
7621: - x - the approximate solution
7623: Output Parameter:
7624: . r - location to store the residual
7626: Level: developer
7628: .seealso: PCMGSetResidual()
7629: @*/
7630: PetscErrorCode MatResidual(Mat mat,Vec b,Vec x,Vec r)
7631: {
7640: MatCheckPreallocated(mat,1);
7641: PetscLogEventBegin(MAT_Residual,mat,0,0,0);
7642: if (!mat->ops->residual) {
7643: MatMult(mat,x,r);
7644: VecAYPX(r,-1.0,b);
7645: } else {
7646: (*mat->ops->residual)(mat,b,x,r);
7647: }
7648: PetscLogEventEnd(MAT_Residual,mat,0,0,0);
7649: return(0);
7650: }
7652: /*@C
7653: MatGetRowIJ - Returns the compressed row storage i and j indices for sequential matrices.
7655: Collective on Mat
7657: Input Parameters:
7658: + mat - the matrix
7659: . shift - 0 or 1 indicating we want the indices starting at 0 or 1
7660: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be symmetrized
7661: - inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7662: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7663: always used.
7665: Output Parameters:
7666: + n - number of rows in the (possibly compressed) matrix
7667: . ia - the row pointers; that is ia[0] = 0, ia[row] = ia[row-1] + number of elements in that row of the matrix
7668: . ja - the column indices
7669: - done - indicates if the routine actually worked and returned appropriate ia[] and ja[] arrays; callers
7670: are responsible for handling the case when done == PETSC_FALSE and ia and ja are not set
7672: Level: developer
7674: Notes:
7675: You CANNOT change any of the ia[] or ja[] values.
7677: Use MatRestoreRowIJ() when you are finished accessing the ia[] and ja[] values.
7679: Fortran Notes:
7680: In Fortran use
7681: $
7682: $ PetscInt ia(1), ja(1)
7683: $ PetscOffset iia, jja
7684: $ call MatGetRowIJ(mat,shift,symmetric,inodecompressed,n,ia,iia,ja,jja,done,ierr)
7685: $ ! Access the ith and jth entries via ia(iia + i) and ja(jja + j)
7687: or
7688: $
7689: $ PetscInt, pointer :: ia(:),ja(:)
7690: $ call MatGetRowIJF90(mat,shift,symmetric,inodecompressed,n,ia,ja,done,ierr)
7691: $ ! Access the ith and jth entries via ia(i) and ja(j)
7693: .seealso: MatGetColumnIJ(), MatRestoreRowIJ(), MatSeqAIJGetArray()
7694: @*/
7695: PetscErrorCode MatGetRowIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7696: {
7706: MatCheckPreallocated(mat,1);
7707: if (!mat->ops->getrowij) *done = PETSC_FALSE;
7708: else {
7709: *done = PETSC_TRUE;
7710: PetscLogEventBegin(MAT_GetRowIJ,mat,0,0,0);
7711: (*mat->ops->getrowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7712: PetscLogEventEnd(MAT_GetRowIJ,mat,0,0,0);
7713: }
7714: return(0);
7715: }
7717: /*@C
7718: MatGetColumnIJ - Returns the compressed column storage i and j indices for sequential matrices.
7720: Collective on Mat
7722: Input Parameters:
7723: + mat - the matrix
7724: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7725: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7726: symmetrized
7727: . inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7728: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7729: always used.
7730: . n - number of columns in the (possibly compressed) matrix
7731: . ia - the column pointers; that is ia[0] = 0, ia[col] = i[col-1] + number of elements in that col of the matrix
7732: - ja - the row indices
7734: Output Parameters:
7735: . done - PETSC_TRUE or PETSC_FALSE, indicating whether the values have been returned
7737: Level: developer
7739: .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
7740: @*/
7741: PetscErrorCode MatGetColumnIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7742: {
7752: MatCheckPreallocated(mat,1);
7753: if (!mat->ops->getcolumnij) *done = PETSC_FALSE;
7754: else {
7755: *done = PETSC_TRUE;
7756: (*mat->ops->getcolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7757: }
7758: return(0);
7759: }
7761: /*@C
7762: MatRestoreRowIJ - Call after you are completed with the ia,ja indices obtained with
7763: MatGetRowIJ().
7765: Collective on Mat
7767: Input Parameters:
7768: + mat - the matrix
7769: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7770: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7771: symmetrized
7772: . inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7773: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7774: always used.
7775: . n - size of (possibly compressed) matrix
7776: . ia - the row pointers
7777: - ja - the column indices
7779: Output Parameters:
7780: . done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
7782: Note:
7783: This routine zeros out n, ia, and ja. This is to prevent accidental
7784: us of the array after it has been restored. If you pass NULL, it will
7785: not zero the pointers. Use of ia or ja after MatRestoreRowIJ() is invalid.
7787: Level: developer
7789: .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
7790: @*/
7791: PetscErrorCode MatRestoreRowIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7792: {
7801: MatCheckPreallocated(mat,1);
7803: if (!mat->ops->restorerowij) *done = PETSC_FALSE;
7804: else {
7805: *done = PETSC_TRUE;
7806: (*mat->ops->restorerowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7807: if (n) *n = 0;
7808: if (ia) *ia = NULL;
7809: if (ja) *ja = NULL;
7810: }
7811: return(0);
7812: }
7814: /*@C
7815: MatRestoreColumnIJ - Call after you are completed with the ia,ja indices obtained with
7816: MatGetColumnIJ().
7818: Collective on Mat
7820: Input Parameters:
7821: + mat - the matrix
7822: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7823: - symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7824: symmetrized
7825: - inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7826: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7827: always used.
7829: Output Parameters:
7830: + n - size of (possibly compressed) matrix
7831: . ia - the column pointers
7832: . ja - the row indices
7833: - done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
7835: Level: developer
7837: .seealso: MatGetColumnIJ(), MatRestoreRowIJ()
7838: @*/
7839: PetscErrorCode MatRestoreColumnIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7840: {
7849: MatCheckPreallocated(mat,1);
7851: if (!mat->ops->restorecolumnij) *done = PETSC_FALSE;
7852: else {
7853: *done = PETSC_TRUE;
7854: (*mat->ops->restorecolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7855: if (n) *n = 0;
7856: if (ia) *ia = NULL;
7857: if (ja) *ja = NULL;
7858: }
7859: return(0);
7860: }
7862: /*@C
7863: MatColoringPatch -Used inside matrix coloring routines that
7864: use MatGetRowIJ() and/or MatGetColumnIJ().
7866: Collective on Mat
7868: Input Parameters:
7869: + mat - the matrix
7870: . ncolors - max color value
7871: . n - number of entries in colorarray
7872: - colorarray - array indicating color for each column
7874: Output Parameters:
7875: . iscoloring - coloring generated using colorarray information
7877: Level: developer
7879: .seealso: MatGetRowIJ(), MatGetColumnIJ()
7881: @*/
7882: PetscErrorCode MatColoringPatch(Mat mat,PetscInt ncolors,PetscInt n,ISColoringValue colorarray[],ISColoring *iscoloring)
7883: {
7891: MatCheckPreallocated(mat,1);
7893: if (!mat->ops->coloringpatch) {
7894: ISColoringCreate(PetscObjectComm((PetscObject)mat),ncolors,n,colorarray,PETSC_OWN_POINTER,iscoloring);
7895: } else {
7896: (*mat->ops->coloringpatch)(mat,ncolors,n,colorarray,iscoloring);
7897: }
7898: return(0);
7899: }
7902: /*@
7903: MatSetUnfactored - Resets a factored matrix to be treated as unfactored.
7905: Logically Collective on Mat
7907: Input Parameter:
7908: . mat - the factored matrix to be reset
7910: Notes:
7911: This routine should be used only with factored matrices formed by in-place
7912: factorization via ILU(0) (or by in-place LU factorization for the MATSEQDENSE
7913: format). This option can save memory, for example, when solving nonlinear
7914: systems with a matrix-free Newton-Krylov method and a matrix-based, in-place
7915: ILU(0) preconditioner.
7917: Note that one can specify in-place ILU(0) factorization by calling
7918: .vb
7919: PCType(pc,PCILU);
7920: PCFactorSeUseInPlace(pc);
7921: .ve
7922: or by using the options -pc_type ilu -pc_factor_in_place
7924: In-place factorization ILU(0) can also be used as a local
7925: solver for the blocks within the block Jacobi or additive Schwarz
7926: methods (runtime option: -sub_pc_factor_in_place). See Users-Manual: ch_pc
7927: for details on setting local solver options.
7929: Most users should employ the simplified KSP interface for linear solvers
7930: instead of working directly with matrix algebra routines such as this.
7931: See, e.g., KSPCreate().
7933: Level: developer
7935: .seealso: PCFactorSetUseInPlace(), PCFactorGetUseInPlace()
7937: @*/
7938: PetscErrorCode MatSetUnfactored(Mat mat)
7939: {
7945: MatCheckPreallocated(mat,1);
7946: mat->factortype = MAT_FACTOR_NONE;
7947: if (!mat->ops->setunfactored) return(0);
7948: (*mat->ops->setunfactored)(mat);
7949: return(0);
7950: }
7952: /*MC
7953: MatDenseGetArrayF90 - Accesses a matrix array from Fortran90.
7955: Synopsis:
7956: MatDenseGetArrayF90(Mat x,{Scalar, pointer :: xx_v(:,:)},integer ierr)
7958: Not collective
7960: Input Parameter:
7961: . x - matrix
7963: Output Parameters:
7964: + xx_v - the Fortran90 pointer to the array
7965: - ierr - error code
7967: Example of Usage:
7968: .vb
7969: PetscScalar, pointer xx_v(:,:)
7970: ....
7971: call MatDenseGetArrayF90(x,xx_v,ierr)
7972: a = xx_v(3)
7973: call MatDenseRestoreArrayF90(x,xx_v,ierr)
7974: .ve
7976: Level: advanced
7978: .seealso: MatDenseRestoreArrayF90(), MatDenseGetArray(), MatDenseRestoreArray(), MatSeqAIJGetArrayF90()
7980: M*/
7982: /*MC
7983: MatDenseRestoreArrayF90 - Restores a matrix array that has been
7984: accessed with MatDenseGetArrayF90().
7986: Synopsis:
7987: MatDenseRestoreArrayF90(Mat x,{Scalar, pointer :: xx_v(:,:)},integer ierr)
7989: Not collective
7991: Input Parameters:
7992: + x - matrix
7993: - xx_v - the Fortran90 pointer to the array
7995: Output Parameter:
7996: . ierr - error code
7998: Example of Usage:
7999: .vb
8000: PetscScalar, pointer xx_v(:,:)
8001: ....
8002: call MatDenseGetArrayF90(x,xx_v,ierr)
8003: a = xx_v(3)
8004: call MatDenseRestoreArrayF90(x,xx_v,ierr)
8005: .ve
8007: Level: advanced
8009: .seealso: MatDenseGetArrayF90(), MatDenseGetArray(), MatDenseRestoreArray(), MatSeqAIJRestoreArrayF90()
8011: M*/
8014: /*MC
8015: MatSeqAIJGetArrayF90 - Accesses a matrix array from Fortran90.
8017: Synopsis:
8018: MatSeqAIJGetArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
8020: Not collective
8022: Input Parameter:
8023: . x - matrix
8025: Output Parameters:
8026: + xx_v - the Fortran90 pointer to the array
8027: - ierr - error code
8029: Example of Usage:
8030: .vb
8031: PetscScalar, pointer xx_v(:)
8032: ....
8033: call MatSeqAIJGetArrayF90(x,xx_v,ierr)
8034: a = xx_v(3)
8035: call MatSeqAIJRestoreArrayF90(x,xx_v,ierr)
8036: .ve
8038: Level: advanced
8040: .seealso: MatSeqAIJRestoreArrayF90(), MatSeqAIJGetArray(), MatSeqAIJRestoreArray(), MatDenseGetArrayF90()
8042: M*/
8044: /*MC
8045: MatSeqAIJRestoreArrayF90 - Restores a matrix array that has been
8046: accessed with MatSeqAIJGetArrayF90().
8048: Synopsis:
8049: MatSeqAIJRestoreArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
8051: Not collective
8053: Input Parameters:
8054: + x - matrix
8055: - xx_v - the Fortran90 pointer to the array
8057: Output Parameter:
8058: . ierr - error code
8060: Example of Usage:
8061: .vb
8062: PetscScalar, pointer xx_v(:)
8063: ....
8064: call MatSeqAIJGetArrayF90(x,xx_v,ierr)
8065: a = xx_v(3)
8066: call MatSeqAIJRestoreArrayF90(x,xx_v,ierr)
8067: .ve
8069: Level: advanced
8071: .seealso: MatSeqAIJGetArrayF90(), MatSeqAIJGetArray(), MatSeqAIJRestoreArray(), MatDenseRestoreArrayF90()
8073: M*/
8076: /*@
8077: MatCreateSubMatrix - Gets a single submatrix on the same number of processors
8078: as the original matrix.
8080: Collective on Mat
8082: Input Parameters:
8083: + mat - the original matrix
8084: . isrow - parallel IS containing the rows this processor should obtain
8085: . 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.
8086: - cll - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
8088: Output Parameter:
8089: . newmat - the new submatrix, of the same type as the old
8091: Level: advanced
8093: Notes:
8094: The submatrix will be able to be multiplied with vectors using the same layout as iscol.
8096: Some matrix types place restrictions on the row and column indices, such
8097: as that they be sorted or that they be equal to each other.
8099: The index sets may not have duplicate entries.
8101: The first time this is called you should use a cll of MAT_INITIAL_MATRIX,
8102: the MatCreateSubMatrix() routine will create the newmat for you. Any additional calls
8103: to this routine with a mat of the same nonzero structure and with a call of MAT_REUSE_MATRIX
8104: will reuse the matrix generated the first time. You should call MatDestroy() on newmat when
8105: you are finished using it.
8107: The communicator of the newly obtained matrix is ALWAYS the same as the communicator of
8108: the input matrix.
8110: If iscol is NULL then all columns are obtained (not supported in Fortran).
8112: Example usage:
8113: Consider the following 8x8 matrix with 34 non-zero values, that is
8114: assembled across 3 processors. Let's assume that proc0 owns 3 rows,
8115: proc1 owns 3 rows, proc2 owns 2 rows. This division can be shown
8116: as follows:
8118: .vb
8119: 1 2 0 | 0 3 0 | 0 4
8120: Proc0 0 5 6 | 7 0 0 | 8 0
8121: 9 0 10 | 11 0 0 | 12 0
8122: -------------------------------------
8123: 13 0 14 | 15 16 17 | 0 0
8124: Proc1 0 18 0 | 19 20 21 | 0 0
8125: 0 0 0 | 22 23 0 | 24 0
8126: -------------------------------------
8127: Proc2 25 26 27 | 0 0 28 | 29 0
8128: 30 0 0 | 31 32 33 | 0 34
8129: .ve
8131: Suppose isrow = [0 1 | 4 | 6 7] and iscol = [1 2 | 3 4 5 | 6]. The resulting submatrix is
8133: .vb
8134: 2 0 | 0 3 0 | 0
8135: Proc0 5 6 | 7 0 0 | 8
8136: -------------------------------
8137: Proc1 18 0 | 19 20 21 | 0
8138: -------------------------------
8139: Proc2 26 27 | 0 0 28 | 29
8140: 0 0 | 31 32 33 | 0
8141: .ve
8144: .seealso: MatCreateSubMatrices(), MatCreateSubMatricesMPI(), MatCreateSubMatrixVirtual(), MatSubMatrixVirtualUpdate()
8145: @*/
8146: PetscErrorCode MatCreateSubMatrix(Mat mat,IS isrow,IS iscol,MatReuse cll,Mat *newmat)
8147: {
8149: PetscMPIInt size;
8150: Mat *local;
8151: IS iscoltmp;
8152: PetscBool flg;
8161: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
8162: if (cll == MAT_IGNORE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Cannot use MAT_IGNORE_MATRIX");
8164: MatCheckPreallocated(mat,1);
8165: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
8167: if (!iscol || isrow == iscol) {
8168: PetscBool stride;
8169: PetscMPIInt grabentirematrix = 0,grab;
8170: PetscObjectTypeCompare((PetscObject)isrow,ISSTRIDE,&stride);
8171: if (stride) {
8172: PetscInt first,step,n,rstart,rend;
8173: ISStrideGetInfo(isrow,&first,&step);
8174: if (step == 1) {
8175: MatGetOwnershipRange(mat,&rstart,&rend);
8176: if (rstart == first) {
8177: ISGetLocalSize(isrow,&n);
8178: if (n == rend-rstart) {
8179: grabentirematrix = 1;
8180: }
8181: }
8182: }
8183: }
8184: MPIU_Allreduce(&grabentirematrix,&grab,1,MPI_INT,MPI_MIN,PetscObjectComm((PetscObject)mat));
8185: if (grab) {
8186: PetscInfo(mat,"Getting entire matrix as submatrix\n");
8187: if (cll == MAT_INITIAL_MATRIX) {
8188: *newmat = mat;
8189: PetscObjectReference((PetscObject)mat);
8190: }
8191: return(0);
8192: }
8193: }
8195: if (!iscol) {
8196: ISCreateStride(PetscObjectComm((PetscObject)mat),mat->cmap->n,mat->cmap->rstart,1,&iscoltmp);
8197: } else {
8198: iscoltmp = iscol;
8199: }
8201: /* if original matrix is on just one processor then use submatrix generated */
8202: if (mat->ops->createsubmatrices && !mat->ops->createsubmatrix && size == 1 && cll == MAT_REUSE_MATRIX) {
8203: MatCreateSubMatrices(mat,1,&isrow,&iscoltmp,MAT_REUSE_MATRIX,&newmat);
8204: goto setproperties;
8205: } else if (mat->ops->createsubmatrices && !mat->ops->createsubmatrix && size == 1) {
8206: MatCreateSubMatrices(mat,1,&isrow,&iscoltmp,MAT_INITIAL_MATRIX,&local);
8207: *newmat = *local;
8208: PetscFree(local);
8209: goto setproperties;
8210: } else if (!mat->ops->createsubmatrix) {
8211: /* Create a new matrix type that implements the operation using the full matrix */
8212: PetscLogEventBegin(MAT_CreateSubMat,mat,0,0,0);
8213: switch (cll) {
8214: case MAT_INITIAL_MATRIX:
8215: MatCreateSubMatrixVirtual(mat,isrow,iscoltmp,newmat);
8216: break;
8217: case MAT_REUSE_MATRIX:
8218: MatSubMatrixVirtualUpdate(*newmat,mat,isrow,iscoltmp);
8219: break;
8220: default: SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Invalid MatReuse, must be either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX");
8221: }
8222: PetscLogEventEnd(MAT_CreateSubMat,mat,0,0,0);
8223: goto setproperties;
8224: }
8226: if (!mat->ops->createsubmatrix) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8227: PetscLogEventBegin(MAT_CreateSubMat,mat,0,0,0);
8228: (*mat->ops->createsubmatrix)(mat,isrow,iscoltmp,cll,newmat);
8229: PetscLogEventEnd(MAT_CreateSubMat,mat,0,0,0);
8231: setproperties:
8232: ISEqualUnsorted(isrow,iscoltmp,&flg);
8233: if (flg) {
8234: MatPropagateSymmetryOptions(mat,*newmat);
8235: }
8236: if (!iscol) {ISDestroy(&iscoltmp);}
8237: if (*newmat && cll == MAT_INITIAL_MATRIX) {PetscObjectStateIncrease((PetscObject)*newmat);}
8238: return(0);
8239: }
8241: /*@
8242: MatPropagateSymmetryOptions - Propagates symmetry options set on a matrix to another matrix
8244: Not Collective
8246: Input Parameters:
8247: + A - the matrix we wish to propagate options from
8248: - B - the matrix we wish to propagate options to
8250: Level: beginner
8252: Notes: Propagates the options associated to MAT_SYMMETRY_ETERNAL, MAT_STRUCTURALLY_SYMMETRIC, MAT_HERMITIAN, MAT_SPD and MAT_SYMMETRIC
8254: .seealso: MatSetOption()
8255: @*/
8256: PetscErrorCode MatPropagateSymmetryOptions(Mat A, Mat B)
8257: {
8263: if (A->symmetric_eternal) { /* symmetric_eternal does not have a corresponding *set flag */
8264: MatSetOption(B,MAT_SYMMETRY_ETERNAL,A->symmetric_eternal);
8265: }
8266: if (A->structurally_symmetric_set) {
8267: MatSetOption(B,MAT_STRUCTURALLY_SYMMETRIC,A->structurally_symmetric);
8268: }
8269: if (A->hermitian_set) {
8270: MatSetOption(B,MAT_HERMITIAN,A->hermitian);
8271: }
8272: if (A->spd_set) {
8273: MatSetOption(B,MAT_SPD,A->spd);
8274: }
8275: if (A->symmetric_set) {
8276: MatSetOption(B,MAT_SYMMETRIC,A->symmetric);
8277: }
8278: return(0);
8279: }
8281: /*@
8282: MatStashSetInitialSize - sets the sizes of the matrix stash, that is
8283: used during the assembly process to store values that belong to
8284: other processors.
8286: Not Collective
8288: Input Parameters:
8289: + mat - the matrix
8290: . size - the initial size of the stash.
8291: - bsize - the initial size of the block-stash(if used).
8293: Options Database Keys:
8294: + -matstash_initial_size <size> or <size0,size1,...sizep-1>
8295: - -matstash_block_initial_size <bsize> or <bsize0,bsize1,...bsizep-1>
8297: Level: intermediate
8299: Notes:
8300: The block-stash is used for values set with MatSetValuesBlocked() while
8301: the stash is used for values set with MatSetValues()
8303: Run with the option -info and look for output of the form
8304: MatAssemblyBegin_MPIXXX:Stash has MM entries, uses nn mallocs.
8305: to determine the appropriate value, MM, to use for size and
8306: MatAssemblyBegin_MPIXXX:Block-Stash has BMM entries, uses nn mallocs.
8307: to determine the value, BMM to use for bsize
8310: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashGetInfo()
8312: @*/
8313: PetscErrorCode MatStashSetInitialSize(Mat mat,PetscInt size, PetscInt bsize)
8314: {
8320: MatStashSetInitialSize_Private(&mat->stash,size);
8321: MatStashSetInitialSize_Private(&mat->bstash,bsize);
8322: return(0);
8323: }
8325: /*@
8326: MatInterpolateAdd - w = y + A*x or A'*x depending on the shape of
8327: the matrix
8329: Neighbor-wise Collective on Mat
8331: Input Parameters:
8332: + mat - the matrix
8333: . x,y - the vectors
8334: - w - where the result is stored
8336: Level: intermediate
8338: Notes:
8339: w may be the same vector as y.
8341: This allows one to use either the restriction or interpolation (its transpose)
8342: matrix to do the interpolation
8344: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
8346: @*/
8347: PetscErrorCode MatInterpolateAdd(Mat A,Vec x,Vec y,Vec w)
8348: {
8350: PetscInt M,N,Ny;
8357: MatGetSize(A,&M,&N);
8358: VecGetSize(y,&Ny);
8359: if (M == Ny) {
8360: MatMultAdd(A,x,y,w);
8361: } else {
8362: MatMultTransposeAdd(A,x,y,w);
8363: }
8364: return(0);
8365: }
8367: /*@
8368: MatInterpolate - y = A*x or A'*x depending on the shape of
8369: the matrix
8371: Neighbor-wise Collective on Mat
8373: Input Parameters:
8374: + mat - the matrix
8375: - x,y - the vectors
8377: Level: intermediate
8379: Notes:
8380: This allows one to use either the restriction or interpolation (its transpose)
8381: matrix to do the interpolation
8383: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
8385: @*/
8386: PetscErrorCode MatInterpolate(Mat A,Vec x,Vec y)
8387: {
8389: PetscInt M,N,Ny;
8395: MatGetSize(A,&M,&N);
8396: VecGetSize(y,&Ny);
8397: if (M == Ny) {
8398: MatMult(A,x,y);
8399: } else {
8400: MatMultTranspose(A,x,y);
8401: }
8402: return(0);
8403: }
8405: /*@
8406: MatRestrict - y = A*x or A'*x
8408: Neighbor-wise Collective on Mat
8410: Input Parameters:
8411: + mat - the matrix
8412: - x,y - the vectors
8414: Level: intermediate
8416: Notes:
8417: This allows one to use either the restriction or interpolation (its transpose)
8418: matrix to do the restriction
8420: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatInterpolate()
8422: @*/
8423: PetscErrorCode MatRestrict(Mat A,Vec x,Vec y)
8424: {
8426: PetscInt M,N,Ny;
8432: MatGetSize(A,&M,&N);
8433: VecGetSize(y,&Ny);
8434: if (M == Ny) {
8435: MatMult(A,x,y);
8436: } else {
8437: MatMultTranspose(A,x,y);
8438: }
8439: return(0);
8440: }
8442: /*@
8443: MatMatInterpolateAdd - Y = W + A*X or W + A'*X
8445: Neighbor-wise Collective on Mat
8447: Input Parameters:
8448: + mat - the matrix
8449: - w, x - the input dense matrices
8451: Output Parameters:
8452: . y - the output dense matrix
8454: Level: intermediate
8456: Notes:
8457: This allows one to use either the restriction or interpolation (its transpose)
8458: matrix to do the interpolation. y matrix can be reused if already created with the proper sizes,
8459: otherwise it will be recreated. y must be initialized to NULL if not supplied.
8461: .seealso: MatInterpolateAdd(), MatMatInterpolate(), MatMatRestrict()
8463: @*/
8464: PetscErrorCode MatMatInterpolateAdd(Mat A,Mat x,Mat w,Mat *y)
8465: {
8467: PetscInt M,N,Mx,Nx,Mo,My = 0,Ny = 0;
8468: PetscBool trans = PETSC_TRUE;
8469: MatReuse reuse = MAT_INITIAL_MATRIX;
8477: MatGetSize(A,&M,&N);
8478: MatGetSize(x,&Mx,&Nx);
8479: if (N == Mx) trans = PETSC_FALSE;
8480: else if (M != Mx) SETERRQ4(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Size mismatch: A %Dx%D, X %Dx%D",M,N,Mx,Nx);
8481: Mo = trans ? N : M;
8482: if (*y) {
8483: MatGetSize(*y,&My,&Ny);
8484: if (Mo == My && Nx == Ny) { reuse = MAT_REUSE_MATRIX; }
8485: else {
8486: if (w && *y == w) SETERRQ6(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Cannot reuse y and w, size mismatch: A %Dx%D, X %Dx%D, Y %Dx%D",M,N,Mx,Nx,My,Ny);
8487: MatDestroy(y);
8488: }
8489: }
8491: if (w && *y == w) { /* this is to minimize changes in PCMG */
8492: PetscBool flg;
8494: PetscObjectQuery((PetscObject)*y,"__MatMatIntAdd_w",(PetscObject*)&w);
8495: if (w) {
8496: PetscInt My,Ny,Mw,Nw;
8498: PetscObjectTypeCompare((PetscObject)*y,((PetscObject)w)->type_name,&flg);
8499: MatGetSize(*y,&My,&Ny);
8500: MatGetSize(w,&Mw,&Nw);
8501: if (!flg || My != Mw || Ny != Nw) w = NULL;
8502: }
8503: if (!w) {
8504: MatDuplicate(*y,MAT_COPY_VALUES,&w);
8505: PetscObjectCompose((PetscObject)*y,"__MatMatIntAdd_w",(PetscObject)w);
8506: PetscLogObjectParent((PetscObject)*y,(PetscObject)w);
8507: PetscObjectDereference((PetscObject)w);
8508: } else {
8509: MatCopy(*y,w,UNKNOWN_NONZERO_PATTERN);
8510: }
8511: }
8512: if (!trans) {
8513: MatMatMult(A,x,reuse,PETSC_DEFAULT,y);
8514: } else {
8515: MatTransposeMatMult(A,x,reuse,PETSC_DEFAULT,y);
8516: }
8517: if (w) {
8518: MatAXPY(*y,1.0,w,UNKNOWN_NONZERO_PATTERN);
8519: }
8520: return(0);
8521: }
8523: /*@
8524: MatMatInterpolate - Y = A*X or A'*X
8526: Neighbor-wise Collective on Mat
8528: Input Parameters:
8529: + mat - the matrix
8530: - x - the input dense matrix
8532: Output Parameters:
8533: . y - the output dense matrix
8536: Level: intermediate
8538: Notes:
8539: This allows one to use either the restriction or interpolation (its transpose)
8540: matrix to do the interpolation. y matrix can be reused if already created with the proper sizes,
8541: otherwise it will be recreated. y must be initialized to NULL if not supplied.
8543: .seealso: MatInterpolate(), MatRestrict(), MatMatRestrict()
8545: @*/
8546: PetscErrorCode MatMatInterpolate(Mat A,Mat x,Mat *y)
8547: {
8551: MatMatInterpolateAdd(A,x,NULL,y);
8552: return(0);
8553: }
8555: /*@
8556: MatMatRestrict - Y = A*X or A'*X
8558: Neighbor-wise Collective on Mat
8560: Input Parameters:
8561: + mat - the matrix
8562: - x - the input dense matrix
8564: Output Parameters:
8565: . y - the output dense matrix
8568: Level: intermediate
8570: Notes:
8571: This allows one to use either the restriction or interpolation (its transpose)
8572: matrix to do the restriction. y matrix can be reused if already created with the proper sizes,
8573: otherwise it will be recreated. y must be initialized to NULL if not supplied.
8575: .seealso: MatRestrict(), MatInterpolate(), MatMatInterpolate()
8576: @*/
8577: PetscErrorCode MatMatRestrict(Mat A,Mat x,Mat *y)
8578: {
8582: MatMatInterpolateAdd(A,x,NULL,y);
8583: return(0);
8584: }
8586: /*@
8587: MatGetNullSpace - retrieves the null space of a matrix.
8589: Logically Collective on Mat
8591: Input Parameters:
8592: + mat - the matrix
8593: - nullsp - the null space object
8595: Level: developer
8597: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatSetNullSpace()
8598: @*/
8599: PetscErrorCode MatGetNullSpace(Mat mat, MatNullSpace *nullsp)
8600: {
8604: *nullsp = (mat->symmetric_set && mat->symmetric && !mat->nullsp) ? mat->transnullsp : mat->nullsp;
8605: return(0);
8606: }
8608: /*@
8609: MatSetNullSpace - attaches a null space to a matrix.
8611: Logically Collective on Mat
8613: Input Parameters:
8614: + mat - the matrix
8615: - nullsp - the null space object
8617: Level: advanced
8619: Notes:
8620: This null space is used by the linear solvers. Overwrites any previous null space that may have been attached
8622: For inconsistent singular systems (linear systems where the right hand side is not in the range of the operator) you also likely should
8623: call MatSetTransposeNullSpace(). This allows the linear system to be solved in a least squares sense.
8625: You can remove the null space by calling this routine with an nullsp of NULL
8628: The fundamental theorem of linear algebra (Gilbert Strang, Introduction to Applied Mathematics, page 72) states that
8629: the domain of a matrix A (from R^n to R^m (m rows, n columns) R^n = the direct sum of the null space of A, n(A), + the range of A^T, R(A^T).
8630: Similarly R^m = direct sum n(A^T) + R(A). Hence the linear system A x = b has a solution only if b in R(A) (or correspondingly b is orthogonal to
8631: n(A^T)) and if x is a solution then x + alpha n(A) is a solution for any alpha. The minimum norm solution is orthogonal to n(A). For problems without a solution
8632: the solution that minimizes the norm of the residual (the least squares solution) can be obtained by solving A x = \hat{b} where \hat{b} is b orthogonalized to the n(A^T).
8634: Krylov solvers can produce the minimal norm solution to the least squares problem by utilizing MatNullSpaceRemove().
8636: If the matrix is known to be symmetric because it is an SBAIJ matrix or one as called MatSetOption(mat,MAT_SYMMETRIC or MAT_SYMMETRIC_ETERNAL,PETSC_TRUE); this
8637: routine also automatically calls MatSetTransposeNullSpace().
8639: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatGetNullSpace(), MatSetTransposeNullSpace(), MatGetTransposeNullSpace(), MatNullSpaceRemove()
8640: @*/
8641: PetscErrorCode MatSetNullSpace(Mat mat,MatNullSpace nullsp)
8642: {
8648: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8649: MatNullSpaceDestroy(&mat->nullsp);
8650: mat->nullsp = nullsp;
8651: if (mat->symmetric_set && mat->symmetric) {
8652: MatSetTransposeNullSpace(mat,nullsp);
8653: }
8654: return(0);
8655: }
8657: /*@
8658: MatGetTransposeNullSpace - retrieves the null space of the transpose of a matrix.
8660: Logically Collective on Mat
8662: Input Parameters:
8663: + mat - the matrix
8664: - nullsp - the null space object
8666: Level: developer
8668: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatSetTransposeNullSpace(), MatSetNullSpace(), MatGetNullSpace()
8669: @*/
8670: PetscErrorCode MatGetTransposeNullSpace(Mat mat, MatNullSpace *nullsp)
8671: {
8676: *nullsp = (mat->symmetric_set && mat->symmetric && !mat->transnullsp) ? mat->nullsp : mat->transnullsp;
8677: return(0);
8678: }
8680: /*@
8681: MatSetTransposeNullSpace - attaches a null space to a matrix.
8683: Logically Collective on Mat
8685: Input Parameters:
8686: + mat - the matrix
8687: - nullsp - the null space object
8689: Level: advanced
8691: Notes:
8692: For inconsistent singular systems (linear systems where the right hand side is not in the range of the operator) this allows the linear system to be solved in a least squares sense.
8693: You must also call MatSetNullSpace()
8696: The fundamental theorem of linear algebra (Gilbert Strang, Introduction to Applied Mathematics, page 72) states that
8697: the domain of a matrix A (from R^n to R^m (m rows, n columns) R^n = the direct sum of the null space of A, n(A), + the range of A^T, R(A^T).
8698: Similarly R^m = direct sum n(A^T) + R(A). Hence the linear system A x = b has a solution only if b in R(A) (or correspondingly b is orthogonal to
8699: n(A^T)) and if x is a solution then x + alpha n(A) is a solution for any alpha. The minimum norm solution is orthogonal to n(A). For problems without a solution
8700: the solution that minimizes the norm of the residual (the least squares solution) can be obtained by solving A x = \hat{b} where \hat{b} is b orthogonalized to the n(A^T).
8702: Krylov solvers can produce the minimal norm solution to the least squares problem by utilizing MatNullSpaceRemove().
8704: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatGetNullSpace(), MatSetNullSpace(), MatGetTransposeNullSpace(), MatNullSpaceRemove()
8705: @*/
8706: PetscErrorCode MatSetTransposeNullSpace(Mat mat,MatNullSpace nullsp)
8707: {
8713: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8714: MatNullSpaceDestroy(&mat->transnullsp);
8715: mat->transnullsp = nullsp;
8716: return(0);
8717: }
8719: /*@
8720: MatSetNearNullSpace - attaches a null space to a matrix, which is often the null space (rigid body modes) of the operator without boundary conditions
8721: This null space will be used to provide near null space vectors to a multigrid preconditioner built from this matrix.
8723: Logically Collective on Mat
8725: Input Parameters:
8726: + mat - the matrix
8727: - nullsp - the null space object
8729: Level: advanced
8731: Notes:
8732: Overwrites any previous near null space that may have been attached
8734: You can remove the null space by calling this routine with an nullsp of NULL
8736: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNullSpace(), MatNullSpaceCreateRigidBody(), MatGetNearNullSpace()
8737: @*/
8738: PetscErrorCode MatSetNearNullSpace(Mat mat,MatNullSpace nullsp)
8739: {
8746: MatCheckPreallocated(mat,1);
8747: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8748: MatNullSpaceDestroy(&mat->nearnullsp);
8749: mat->nearnullsp = nullsp;
8750: return(0);
8751: }
8753: /*@
8754: MatGetNearNullSpace - Get null space attached with MatSetNearNullSpace()
8756: Not Collective
8758: Input Parameter:
8759: . mat - the matrix
8761: Output Parameter:
8762: . nullsp - the null space object, NULL if not set
8764: Level: developer
8766: .seealso: MatSetNearNullSpace(), MatGetNullSpace(), MatNullSpaceCreate()
8767: @*/
8768: PetscErrorCode MatGetNearNullSpace(Mat mat,MatNullSpace *nullsp)
8769: {
8774: MatCheckPreallocated(mat,1);
8775: *nullsp = mat->nearnullsp;
8776: return(0);
8777: }
8779: /*@C
8780: MatICCFactor - Performs in-place incomplete Cholesky factorization of matrix.
8782: Collective on Mat
8784: Input Parameters:
8785: + mat - the matrix
8786: . row - row/column permutation
8787: . fill - expected fill factor >= 1.0
8788: - level - level of fill, for ICC(k)
8790: Notes:
8791: Probably really in-place only when level of fill is zero, otherwise allocates
8792: new space to store factored matrix and deletes previous memory.
8794: Most users should employ the simplified KSP interface for linear solvers
8795: instead of working directly with matrix algebra routines such as this.
8796: See, e.g., KSPCreate().
8798: Level: developer
8801: .seealso: MatICCFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
8803: Developer Note: fortran interface is not autogenerated as the f90
8804: interface defintion cannot be generated correctly [due to MatFactorInfo]
8806: @*/
8807: PetscErrorCode MatICCFactor(Mat mat,IS row,const MatFactorInfo *info)
8808: {
8816: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"matrix must be square");
8817: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
8818: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
8819: if (!mat->ops->iccfactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8820: MatCheckPreallocated(mat,1);
8821: (*mat->ops->iccfactor)(mat,row,info);
8822: PetscObjectStateIncrease((PetscObject)mat);
8823: return(0);
8824: }
8826: /*@
8827: MatDiagonalScaleLocal - Scales columns of a matrix given the scaling values including the
8828: ghosted ones.
8830: Not Collective
8832: Input Parameters:
8833: + mat - the matrix
8834: - diag = the diagonal values, including ghost ones
8836: Level: developer
8838: Notes:
8839: Works only for MPIAIJ and MPIBAIJ matrices
8841: .seealso: MatDiagonalScale()
8842: @*/
8843: PetscErrorCode MatDiagonalScaleLocal(Mat mat,Vec diag)
8844: {
8846: PetscMPIInt size;
8853: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
8854: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
8855: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
8856: if (size == 1) {
8857: PetscInt n,m;
8858: VecGetSize(diag,&n);
8859: MatGetSize(mat,NULL,&m);
8860: if (m == n) {
8861: MatDiagonalScale(mat,NULL,diag);
8862: } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only supported for sequential matrices when no ghost points/periodic conditions");
8863: } else {
8864: PetscUseMethod(mat,"MatDiagonalScaleLocal_C",(Mat,Vec),(mat,diag));
8865: }
8866: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
8867: PetscObjectStateIncrease((PetscObject)mat);
8868: return(0);
8869: }
8871: /*@
8872: MatGetInertia - Gets the inertia from a factored matrix
8874: Collective on Mat
8876: Input Parameter:
8877: . mat - the matrix
8879: Output Parameters:
8880: + nneg - number of negative eigenvalues
8881: . nzero - number of zero eigenvalues
8882: - npos - number of positive eigenvalues
8884: Level: advanced
8886: Notes:
8887: Matrix must have been factored by MatCholeskyFactor()
8890: @*/
8891: PetscErrorCode MatGetInertia(Mat mat,PetscInt *nneg,PetscInt *nzero,PetscInt *npos)
8892: {
8898: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
8899: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Numeric factor mat is not assembled");
8900: if (!mat->ops->getinertia) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8901: (*mat->ops->getinertia)(mat,nneg,nzero,npos);
8902: return(0);
8903: }
8905: /* ----------------------------------------------------------------*/
8906: /*@C
8907: MatSolves - Solves A x = b, given a factored matrix, for a collection of vectors
8909: Neighbor-wise Collective on Mats
8911: Input Parameters:
8912: + mat - the factored matrix
8913: - b - the right-hand-side vectors
8915: Output Parameter:
8916: . x - the result vectors
8918: Notes:
8919: The vectors b and x cannot be the same. I.e., one cannot
8920: call MatSolves(A,x,x).
8922: Notes:
8923: Most users should employ the simplified KSP interface for linear solvers
8924: instead of working directly with matrix algebra routines such as this.
8925: See, e.g., KSPCreate().
8927: Level: developer
8929: .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd(), MatSolve()
8930: @*/
8931: PetscErrorCode MatSolves(Mat mat,Vecs b,Vecs x)
8932: {
8938: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
8939: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
8940: if (!mat->rmap->N && !mat->cmap->N) return(0);
8942: if (!mat->ops->solves) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8943: MatCheckPreallocated(mat,1);
8944: PetscLogEventBegin(MAT_Solves,mat,0,0,0);
8945: (*mat->ops->solves)(mat,b,x);
8946: PetscLogEventEnd(MAT_Solves,mat,0,0,0);
8947: return(0);
8948: }
8950: /*@
8951: MatIsSymmetric - Test whether a matrix is symmetric
8953: Collective on Mat
8955: Input Parameter:
8956: + A - the matrix to test
8957: - tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact transpose)
8959: Output Parameters:
8960: . flg - the result
8962: Notes:
8963: For real numbers MatIsSymmetric() and MatIsHermitian() return identical results
8965: Level: intermediate
8967: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetricKnown()
8968: @*/
8969: PetscErrorCode MatIsSymmetric(Mat A,PetscReal tol,PetscBool *flg)
8970: {
8977: if (!A->symmetric_set) {
8978: if (!A->ops->issymmetric) {
8979: MatType mattype;
8980: MatGetType(A,&mattype);
8981: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for symmetric",mattype);
8982: }
8983: (*A->ops->issymmetric)(A,tol,flg);
8984: if (!tol) {
8985: MatSetOption(A,MAT_SYMMETRIC,*flg);
8986: }
8987: } else if (A->symmetric) {
8988: *flg = PETSC_TRUE;
8989: } else if (!tol) {
8990: *flg = PETSC_FALSE;
8991: } else {
8992: if (!A->ops->issymmetric) {
8993: MatType mattype;
8994: MatGetType(A,&mattype);
8995: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for symmetric",mattype);
8996: }
8997: (*A->ops->issymmetric)(A,tol,flg);
8998: }
8999: return(0);
9000: }
9002: /*@
9003: MatIsHermitian - Test whether a matrix is Hermitian
9005: Collective on Mat
9007: Input Parameter:
9008: + A - the matrix to test
9009: - tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact Hermitian)
9011: Output Parameters:
9012: . flg - the result
9014: Level: intermediate
9016: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(),
9017: MatIsSymmetricKnown(), MatIsSymmetric()
9018: @*/
9019: PetscErrorCode MatIsHermitian(Mat A,PetscReal tol,PetscBool *flg)
9020: {
9027: if (!A->hermitian_set) {
9028: if (!A->ops->ishermitian) {
9029: MatType mattype;
9030: MatGetType(A,&mattype);
9031: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for hermitian",mattype);
9032: }
9033: (*A->ops->ishermitian)(A,tol,flg);
9034: if (!tol) {
9035: MatSetOption(A,MAT_HERMITIAN,*flg);
9036: }
9037: } else if (A->hermitian) {
9038: *flg = PETSC_TRUE;
9039: } else if (!tol) {
9040: *flg = PETSC_FALSE;
9041: } else {
9042: if (!A->ops->ishermitian) {
9043: MatType mattype;
9044: MatGetType(A,&mattype);
9045: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for hermitian",mattype);
9046: }
9047: (*A->ops->ishermitian)(A,tol,flg);
9048: }
9049: return(0);
9050: }
9052: /*@
9053: MatIsSymmetricKnown - Checks the flag on the matrix to see if it is symmetric.
9055: Not Collective
9057: Input Parameter:
9058: . A - the matrix to check
9060: Output Parameters:
9061: + set - if the symmetric flag is set (this tells you if the next flag is valid)
9062: - flg - the result
9064: Level: advanced
9066: Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsSymmetric()
9067: if you want it explicitly checked
9069: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
9070: @*/
9071: PetscErrorCode MatIsSymmetricKnown(Mat A,PetscBool *set,PetscBool *flg)
9072: {
9077: if (A->symmetric_set) {
9078: *set = PETSC_TRUE;
9079: *flg = A->symmetric;
9080: } else {
9081: *set = PETSC_FALSE;
9082: }
9083: return(0);
9084: }
9086: /*@
9087: MatIsHermitianKnown - Checks the flag on the matrix to see if it is hermitian.
9089: Not Collective
9091: Input Parameter:
9092: . A - the matrix to check
9094: Output Parameters:
9095: + set - if the hermitian flag is set (this tells you if the next flag is valid)
9096: - flg - the result
9098: Level: advanced
9100: Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsHermitian()
9101: if you want it explicitly checked
9103: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
9104: @*/
9105: PetscErrorCode MatIsHermitianKnown(Mat A,PetscBool *set,PetscBool *flg)
9106: {
9111: if (A->hermitian_set) {
9112: *set = PETSC_TRUE;
9113: *flg = A->hermitian;
9114: } else {
9115: *set = PETSC_FALSE;
9116: }
9117: return(0);
9118: }
9120: /*@
9121: MatIsStructurallySymmetric - Test whether a matrix is structurally symmetric
9123: Collective on Mat
9125: Input Parameter:
9126: . A - the matrix to test
9128: Output Parameters:
9129: . flg - the result
9131: Level: intermediate
9133: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsSymmetric(), MatSetOption()
9134: @*/
9135: PetscErrorCode MatIsStructurallySymmetric(Mat A,PetscBool *flg)
9136: {
9142: if (!A->structurally_symmetric_set) {
9143: if (!A->ops->isstructurallysymmetric) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Matrix of type %s does not support checking for structural symmetric",((PetscObject)A)->type_name);
9144: (*A->ops->isstructurallysymmetric)(A,flg);
9145: MatSetOption(A,MAT_STRUCTURALLY_SYMMETRIC,*flg);
9146: } else *flg = A->structurally_symmetric;
9147: return(0);
9148: }
9150: /*@
9151: MatStashGetInfo - Gets how many values are currently in the matrix stash, i.e. need
9152: to be communicated to other processors during the MatAssemblyBegin/End() process
9154: Not collective
9156: Input Parameter:
9157: . vec - the vector
9159: Output Parameters:
9160: + nstash - the size of the stash
9161: . reallocs - the number of additional mallocs incurred.
9162: . bnstash - the size of the block stash
9163: - breallocs - the number of additional mallocs incurred.in the block stash
9165: Level: advanced
9167: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashSetInitialSize()
9169: @*/
9170: PetscErrorCode MatStashGetInfo(Mat mat,PetscInt *nstash,PetscInt *reallocs,PetscInt *bnstash,PetscInt *breallocs)
9171: {
9175: MatStashGetInfo_Private(&mat->stash,nstash,reallocs);
9176: MatStashGetInfo_Private(&mat->bstash,bnstash,breallocs);
9177: return(0);
9178: }
9180: /*@C
9181: MatCreateVecs - Get vector(s) compatible with the matrix, i.e. with the same
9182: parallel layout
9184: Collective on Mat
9186: Input Parameter:
9187: . mat - the matrix
9189: Output Parameter:
9190: + right - (optional) vector that the matrix can be multiplied against
9191: - left - (optional) vector that the matrix vector product can be stored in
9193: Notes:
9194: The blocksize of the returned vectors is determined by the row and column block sizes set with MatSetBlockSizes() or the single blocksize (same for both) set by MatSetBlockSize().
9196: Notes:
9197: These are new vectors which are not owned by the Mat, they should be destroyed in VecDestroy() when no longer needed
9199: Level: advanced
9201: .seealso: MatCreate(), VecDestroy()
9202: @*/
9203: PetscErrorCode MatCreateVecs(Mat mat,Vec *right,Vec *left)
9204: {
9210: if (mat->ops->getvecs) {
9211: (*mat->ops->getvecs)(mat,right,left);
9212: } else {
9213: PetscInt rbs,cbs;
9214: MatGetBlockSizes(mat,&rbs,&cbs);
9215: if (right) {
9216: if (mat->cmap->n < 0) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"PetscLayout for columns not yet setup");
9217: VecCreate(PetscObjectComm((PetscObject)mat),right);
9218: VecSetSizes(*right,mat->cmap->n,PETSC_DETERMINE);
9219: VecSetBlockSize(*right,cbs);
9220: VecSetType(*right,mat->defaultvectype);
9221: PetscLayoutReference(mat->cmap,&(*right)->map);
9222: }
9223: if (left) {
9224: if (mat->rmap->n < 0) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"PetscLayout for rows not yet setup");
9225: VecCreate(PetscObjectComm((PetscObject)mat),left);
9226: VecSetSizes(*left,mat->rmap->n,PETSC_DETERMINE);
9227: VecSetBlockSize(*left,rbs);
9228: VecSetType(*left,mat->defaultvectype);
9229: PetscLayoutReference(mat->rmap,&(*left)->map);
9230: }
9231: }
9232: return(0);
9233: }
9235: /*@C
9236: MatFactorInfoInitialize - Initializes a MatFactorInfo data structure
9237: with default values.
9239: Not Collective
9241: Input Parameters:
9242: . info - the MatFactorInfo data structure
9245: Notes:
9246: The solvers are generally used through the KSP and PC objects, for example
9247: PCLU, PCILU, PCCHOLESKY, PCICC
9249: Level: developer
9251: .seealso: MatFactorInfo
9253: Developer Note: fortran interface is not autogenerated as the f90
9254: interface defintion cannot be generated correctly [due to MatFactorInfo]
9256: @*/
9258: PetscErrorCode MatFactorInfoInitialize(MatFactorInfo *info)
9259: {
9263: PetscMemzero(info,sizeof(MatFactorInfo));
9264: return(0);
9265: }
9267: /*@
9268: MatFactorSetSchurIS - Set indices corresponding to the Schur complement you wish to have computed
9270: Collective on Mat
9272: Input Parameters:
9273: + mat - the factored matrix
9274: - is - the index set defining the Schur indices (0-based)
9276: Notes:
9277: Call MatFactorSolveSchurComplement() or MatFactorSolveSchurComplementTranspose() after this call to solve a Schur complement system.
9279: You can call MatFactorGetSchurComplement() or MatFactorCreateSchurComplement() after this call.
9281: Level: developer
9283: .seealso: MatGetFactor(), MatFactorGetSchurComplement(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSolveSchurComplement(),
9284: MatFactorSolveSchurComplementTranspose(), MatFactorSolveSchurComplement()
9286: @*/
9287: PetscErrorCode MatFactorSetSchurIS(Mat mat,IS is)
9288: {
9289: PetscErrorCode ierr,(*f)(Mat,IS);
9297: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Only for factored matrix");
9298: PetscObjectQueryFunction((PetscObject)mat,"MatFactorSetSchurIS_C",&f);
9299: if (!f) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"The selected MatSolverType does not support Schur complement computation. You should use MATSOLVERMUMPS or MATSOLVERMKL_PARDISO");
9300: MatDestroy(&mat->schur);
9301: (*f)(mat,is);
9302: if (!mat->schur) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_PLIB,"Schur complement has not been created");
9303: return(0);
9304: }
9306: /*@
9307: MatFactorCreateSchurComplement - Create a Schur complement matrix object using Schur data computed during the factorization step
9309: Logically Collective on Mat
9311: Input Parameters:
9312: + F - the factored matrix obtained by calling MatGetFactor() from PETSc-MUMPS interface
9313: . S - location where to return the Schur complement, can be NULL
9314: - status - the status of the Schur complement matrix, can be NULL
9316: Notes:
9317: You must call MatFactorSetSchurIS() before calling this routine.
9319: The routine provides a copy of the Schur matrix stored within the solver data structures.
9320: The caller must destroy the object when it is no longer needed.
9321: If MatFactorInvertSchurComplement() has been called, the routine gets back the inverse.
9323: Use MatFactorGetSchurComplement() to get access to the Schur complement matrix inside the factored matrix instead of making a copy of it (which this function does)
9325: Developer Notes:
9326: The reason this routine exists is because the representation of the Schur complement within the factor matrix may be different than a standard PETSc
9327: matrix representation and we normally do not want to use the time or memory to make a copy as a regular PETSc matrix.
9329: See MatCreateSchurComplement() or MatGetSchurComplement() for ways to create virtual or approximate Schur complements.
9331: Level: advanced
9333: References:
9335: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorGetSchurComplement(), MatFactorSchurStatus
9336: @*/
9337: PetscErrorCode MatFactorCreateSchurComplement(Mat F,Mat* S,MatFactorSchurStatus* status)
9338: {
9345: if (S) {
9346: PetscErrorCode (*f)(Mat,Mat*);
9348: PetscObjectQueryFunction((PetscObject)F,"MatFactorCreateSchurComplement_C",&f);
9349: if (f) {
9350: (*f)(F,S);
9351: } else {
9352: MatDuplicate(F->schur,MAT_COPY_VALUES,S);
9353: }
9354: }
9355: if (status) *status = F->schur_status;
9356: return(0);
9357: }
9359: /*@
9360: MatFactorGetSchurComplement - Gets access to a Schur complement matrix using the current Schur data within a factored matrix
9362: Logically Collective on Mat
9364: Input Parameters:
9365: + F - the factored matrix obtained by calling MatGetFactor()
9366: . *S - location where to return the Schur complement, can be NULL
9367: - status - the status of the Schur complement matrix, can be NULL
9369: Notes:
9370: You must call MatFactorSetSchurIS() before calling this routine.
9372: Schur complement mode is currently implemented for sequential matrices.
9373: The routine returns a the Schur Complement stored within the data strutures of the solver.
9374: If MatFactorInvertSchurComplement() has previously been called, the returned matrix is actually the inverse of the Schur complement.
9375: The returned matrix should not be destroyed; the caller should call MatFactorRestoreSchurComplement() when the object is no longer needed.
9377: Use MatFactorCreateSchurComplement() to create a copy of the Schur complement matrix that is within a factored matrix
9379: See MatCreateSchurComplement() or MatGetSchurComplement() for ways to create virtual or approximate Schur complements.
9381: Level: advanced
9383: References:
9385: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSchurStatus
9386: @*/
9387: PetscErrorCode MatFactorGetSchurComplement(Mat F,Mat* S,MatFactorSchurStatus* status)
9388: {
9393: if (S) *S = F->schur;
9394: if (status) *status = F->schur_status;
9395: return(0);
9396: }
9398: /*@
9399: MatFactorRestoreSchurComplement - Restore the Schur complement matrix object obtained from a call to MatFactorGetSchurComplement
9401: Logically Collective on Mat
9403: Input Parameters:
9404: + F - the factored matrix obtained by calling MatGetFactor()
9405: . *S - location where the Schur complement is stored
9406: - status - the status of the Schur complement matrix (see MatFactorSchurStatus)
9408: Notes:
9410: Level: advanced
9412: References:
9414: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSchurStatus
9415: @*/
9416: PetscErrorCode MatFactorRestoreSchurComplement(Mat F,Mat* S,MatFactorSchurStatus status)
9417: {
9422: if (S) {
9424: *S = NULL;
9425: }
9426: F->schur_status = status;
9427: MatFactorUpdateSchurStatus_Private(F);
9428: return(0);
9429: }
9431: /*@
9432: MatFactorSolveSchurComplementTranspose - Solve the transpose of the Schur complement system computed during the factorization step
9434: Logically Collective on Mat
9436: Input Parameters:
9437: + F - the factored matrix obtained by calling MatGetFactor()
9438: . rhs - location where the right hand side of the Schur complement system is stored
9439: - sol - location where the solution of the Schur complement system has to be returned
9441: Notes:
9442: The sizes of the vectors should match the size of the Schur complement
9444: Must be called after MatFactorSetSchurIS()
9446: Level: advanced
9448: References:
9450: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorSolveSchurComplement()
9451: @*/
9452: PetscErrorCode MatFactorSolveSchurComplementTranspose(Mat F, Vec rhs, Vec sol)
9453: {
9465: MatFactorFactorizeSchurComplement(F);
9466: switch (F->schur_status) {
9467: case MAT_FACTOR_SCHUR_FACTORED:
9468: MatSolveTranspose(F->schur,rhs,sol);
9469: break;
9470: case MAT_FACTOR_SCHUR_INVERTED:
9471: MatMultTranspose(F->schur,rhs,sol);
9472: break;
9473: default:
9474: SETERRQ1(PetscObjectComm((PetscObject)F),PETSC_ERR_SUP,"Unhandled MatFactorSchurStatus %D",F->schur_status);
9475: }
9476: return(0);
9477: }
9479: /*@
9480: MatFactorSolveSchurComplement - Solve the Schur complement system computed during the factorization step
9482: Logically Collective on Mat
9484: Input Parameters:
9485: + F - the factored matrix obtained by calling MatGetFactor()
9486: . rhs - location where the right hand side of the Schur complement system is stored
9487: - sol - location where the solution of the Schur complement system has to be returned
9489: Notes:
9490: The sizes of the vectors should match the size of the Schur complement
9492: Must be called after MatFactorSetSchurIS()
9494: Level: advanced
9496: References:
9498: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorSolveSchurComplementTranspose()
9499: @*/
9500: PetscErrorCode MatFactorSolveSchurComplement(Mat F, Vec rhs, Vec sol)
9501: {
9513: MatFactorFactorizeSchurComplement(F);
9514: switch (F->schur_status) {
9515: case MAT_FACTOR_SCHUR_FACTORED:
9516: MatSolve(F->schur,rhs,sol);
9517: break;
9518: case MAT_FACTOR_SCHUR_INVERTED:
9519: MatMult(F->schur,rhs,sol);
9520: break;
9521: default:
9522: SETERRQ1(PetscObjectComm((PetscObject)F),PETSC_ERR_SUP,"Unhandled MatFactorSchurStatus %D",F->schur_status);
9523: }
9524: return(0);
9525: }
9527: /*@
9528: MatFactorInvertSchurComplement - Invert the Schur complement matrix computed during the factorization step
9530: Logically Collective on Mat
9532: Input Parameters:
9533: . F - the factored matrix obtained by calling MatGetFactor()
9535: Notes:
9536: Must be called after MatFactorSetSchurIS().
9538: Call MatFactorGetSchurComplement() or MatFactorCreateSchurComplement() AFTER this call to actually compute the inverse and get access to it.
9540: Level: advanced
9542: References:
9544: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorGetSchurComplement(), MatFactorCreateSchurComplement()
9545: @*/
9546: PetscErrorCode MatFactorInvertSchurComplement(Mat F)
9547: {
9553: if (F->schur_status == MAT_FACTOR_SCHUR_INVERTED) return(0);
9554: MatFactorFactorizeSchurComplement(F);
9555: MatFactorInvertSchurComplement_Private(F);
9556: F->schur_status = MAT_FACTOR_SCHUR_INVERTED;
9557: return(0);
9558: }
9560: /*@
9561: MatFactorFactorizeSchurComplement - Factorize the Schur complement matrix computed during the factorization step
9563: Logically Collective on Mat
9565: Input Parameters:
9566: . F - the factored matrix obtained by calling MatGetFactor()
9568: Notes:
9569: Must be called after MatFactorSetSchurIS().
9571: Level: advanced
9573: References:
9575: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorInvertSchurComplement()
9576: @*/
9577: PetscErrorCode MatFactorFactorizeSchurComplement(Mat F)
9578: {
9584: if (F->schur_status == MAT_FACTOR_SCHUR_INVERTED || F->schur_status == MAT_FACTOR_SCHUR_FACTORED) return(0);
9585: MatFactorFactorizeSchurComplement_Private(F);
9586: F->schur_status = MAT_FACTOR_SCHUR_FACTORED;
9587: return(0);
9588: }
9590: /*@
9591: MatPtAP - Creates the matrix product C = P^T * A * P
9593: Neighbor-wise Collective on Mat
9595: Input Parameters:
9596: + A - the matrix
9597: . P - the projection matrix
9598: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9599: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(P)), use PETSC_DEFAULT if you do not have a good estimate
9600: if the result is a dense matrix this is irrelevent
9602: Output Parameters:
9603: . C - the product matrix
9605: Notes:
9606: C will be created and must be destroyed by the user with MatDestroy().
9608: For matrix types without special implementation the function fallbacks to MatMatMult() followed by MatTransposeMatMult().
9610: Level: intermediate
9612: .seealso: MatMatMult(), MatRARt()
9613: @*/
9614: PetscErrorCode MatPtAP(Mat A,Mat P,MatReuse scall,PetscReal fill,Mat *C)
9615: {
9619: if (scall == MAT_REUSE_MATRIX) MatCheckProduct(*C,5);
9620: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9622: if (scall == MAT_INITIAL_MATRIX) {
9623: MatProductCreate(A,P,NULL,C);
9624: MatProductSetType(*C,MATPRODUCT_PtAP);
9625: MatProductSetAlgorithm(*C,"default");
9626: MatProductSetFill(*C,fill);
9628: (*C)->product->api_user = PETSC_TRUE;
9629: MatProductSetFromOptions(*C);
9630: if (!(*C)->ops->productsymbolic) SETERRQ3(PetscObjectComm((PetscObject)(*C)),PETSC_ERR_SUP,"MatProduct %s not supported for A %s and P %s",MatProductTypes[MATPRODUCT_PtAP],((PetscObject)A)->type_name,((PetscObject)P)->type_name);
9631: MatProductSymbolic(*C);
9632: } else { /* scall == MAT_REUSE_MATRIX */
9633: MatProductReplaceMats(A,P,NULL,*C);
9634: }
9636: MatProductNumeric(*C);
9637: if (A->symmetric_set && A->symmetric) {
9638: MatSetOption(*C,MAT_SYMMETRIC,PETSC_TRUE);
9639: }
9640: return(0);
9641: }
9643: /*@
9644: MatRARt - Creates the matrix product C = R * A * R^T
9646: Neighbor-wise Collective on Mat
9648: Input Parameters:
9649: + A - the matrix
9650: . R - the projection matrix
9651: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9652: - fill - expected fill as ratio of nnz(C)/nnz(A), use PETSC_DEFAULT if you do not have a good estimate
9653: if the result is a dense matrix this is irrelevent
9655: Output Parameters:
9656: . C - the product matrix
9658: Notes:
9659: C will be created and must be destroyed by the user with MatDestroy().
9661: This routine is currently only implemented for pairs of AIJ matrices and classes
9662: which inherit from AIJ. Due to PETSc sparse matrix block row distribution among processes,
9663: parallel MatRARt is implemented via explicit transpose of R, which could be very expensive.
9664: We recommend using MatPtAP().
9666: Level: intermediate
9668: .seealso: MatMatMult(), MatPtAP()
9669: @*/
9670: PetscErrorCode MatRARt(Mat A,Mat R,MatReuse scall,PetscReal fill,Mat *C)
9671: {
9675: if (scall == MAT_REUSE_MATRIX) MatCheckProduct(*C,5);
9676: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9678: if (scall == MAT_INITIAL_MATRIX) {
9679: MatProductCreate(A,R,NULL,C);
9680: MatProductSetType(*C,MATPRODUCT_RARt);
9681: MatProductSetAlgorithm(*C,"default");
9682: MatProductSetFill(*C,fill);
9684: (*C)->product->api_user = PETSC_TRUE;
9685: MatProductSetFromOptions(*C);
9686: if (!(*C)->ops->productsymbolic) SETERRQ3(PetscObjectComm((PetscObject)(*C)),PETSC_ERR_SUP,"MatProduct %s not supported for A %s and R %s",MatProductTypes[MATPRODUCT_RARt],((PetscObject)A)->type_name,((PetscObject)R)->type_name);
9687: MatProductSymbolic(*C);
9688: } else { /* scall == MAT_REUSE_MATRIX */
9689: MatProductReplaceMats(A,R,NULL,*C);
9690: }
9692: MatProductNumeric(*C);
9693: if (A->symmetric_set && A->symmetric) {
9694: MatSetOption(*C,MAT_SYMMETRIC,PETSC_TRUE);
9695: }
9696: return(0);
9697: }
9700: static PetscErrorCode MatProduct_Private(Mat A,Mat B,MatReuse scall,PetscReal fill,MatProductType ptype, Mat *C)
9701: {
9705: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9707: if (scall == MAT_INITIAL_MATRIX) {
9708: PetscInfo1(A,"Calling MatProduct API with MAT_INITIAL_MATRIX and product type %s\n",MatProductTypes[ptype]);
9709: MatProductCreate(A,B,NULL,C);
9710: MatProductSetType(*C,ptype);
9711: MatProductSetAlgorithm(*C,MATPRODUCTALGORITHM_DEFAULT);
9712: MatProductSetFill(*C,fill);
9714: (*C)->product->api_user = PETSC_TRUE;
9715: MatProductSetFromOptions(*C);
9716: MatProductSymbolic(*C);
9717: } else { /* scall == MAT_REUSE_MATRIX */
9718: Mat_Product *product = (*C)->product;
9720: PetscInfo2(A,"Calling MatProduct API with MAT_REUSE_MATRIX %s product present and product type %s\n",product ? "with" : "without",MatProductTypes[ptype]);
9721: if (!product) {
9722: /* user provide the dense matrix *C without calling MatProductCreate() */
9723: PetscBool isdense;
9725: PetscObjectBaseTypeCompareAny((PetscObject)(*C),&isdense,MATSEQDENSE,MATMPIDENSE,"");
9726: if (isdense) {
9727: /* user wants to reuse an assembled dense matrix */
9728: /* Create product -- see MatCreateProduct() */
9729: MatProductCreate_Private(A,B,NULL,*C);
9730: product = (*C)->product;
9731: product->fill = fill;
9732: product->api_user = PETSC_TRUE;
9733: product->clear = PETSC_TRUE;
9735: MatProductSetType(*C,ptype);
9736: MatProductSetFromOptions(*C);
9737: if (!(*C)->ops->productsymbolic) SETERRQ3(PetscObjectComm((PetscObject)(*C)),PETSC_ERR_SUP,"MatProduct %s not supported for %s and %s",MatProductTypes[ptype],((PetscObject)A)->type_name,((PetscObject)B)->type_name);
9738: MatProductSymbolic(*C);
9739: } else SETERRQ(PetscObjectComm((PetscObject)(*C)),PETSC_ERR_SUP,"Call MatProductCreate() first");
9740: } else { /* user may change input matrices A or B when REUSE */
9741: MatProductReplaceMats(A,B,NULL,*C);
9742: }
9743: }
9744: MatProductNumeric(*C);
9745: return(0);
9746: }
9748: /*@
9749: MatMatMult - Performs Matrix-Matrix Multiplication C=A*B.
9751: Neighbor-wise Collective on Mat
9753: Input Parameters:
9754: + A - the left matrix
9755: . B - the right matrix
9756: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9757: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if you do not have a good estimate
9758: if the result is a dense matrix this is irrelevent
9760: Output Parameters:
9761: . C - the product matrix
9763: Notes:
9764: Unless scall is MAT_REUSE_MATRIX C will be created.
9766: MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call and C was obtained from a previous
9767: call to this function with MAT_INITIAL_MATRIX.
9769: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value actually needed.
9771: If you have many matrices with the same non-zero structure to multiply, you should use MatProductCreate()/MatProductSymbolic(C)/ReplaceMats(), and call MatProductNumeric() repeatedly.
9773: In the special case where matrix B (and hence C) are dense you can create the correctly sized matrix C yourself and then call this routine with MAT_REUSE_MATRIX, rather than first having MatMatMult() create it for you. You can NEVER do this if the matrix C is sparse.
9775: Level: intermediate
9777: .seealso: MatTransposeMatMult(), MatMatTransposeMult(), MatPtAP()
9778: @*/
9779: PetscErrorCode MatMatMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9780: {
9784: MatProduct_Private(A,B,scall,fill,MATPRODUCT_AB,C);
9785: return(0);
9786: }
9788: /*@
9789: MatMatTransposeMult - Performs Matrix-Matrix Multiplication C=A*B^T.
9791: Neighbor-wise Collective on Mat
9793: Input Parameters:
9794: + A - the left matrix
9795: . B - the right matrix
9796: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9797: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if not known
9799: Output Parameters:
9800: . C - the product matrix
9802: Notes:
9803: C will be created if MAT_INITIAL_MATRIX and must be destroyed by the user with MatDestroy().
9805: MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call
9807: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9808: actually needed.
9810: This routine is currently only implemented for pairs of SeqAIJ matrices, for the SeqDense class,
9811: and for pairs of MPIDense matrices.
9813: Options Database Keys:
9814: . -matmattransmult_mpidense_mpidense_via {allgatherv,cyclic} - Choose between algorthims for MPIDense matrices: the
9815: first redundantly copies the transposed B matrix on each process and requiers O(log P) communication complexity;
9816: the second never stores more than one portion of the B matrix at a time by requires O(P) communication complexity.
9818: Level: intermediate
9820: .seealso: MatMatMult(), MatTransposeMatMult() MatPtAP()
9821: @*/
9822: PetscErrorCode MatMatTransposeMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9823: {
9827: MatProduct_Private(A,B,scall,fill,MATPRODUCT_ABt,C);
9828: return(0);
9829: }
9831: /*@
9832: MatTransposeMatMult - Performs Matrix-Matrix Multiplication C=A^T*B.
9834: Neighbor-wise Collective on Mat
9836: Input Parameters:
9837: + A - the left matrix
9838: . B - the right matrix
9839: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9840: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if not known
9842: Output Parameters:
9843: . C - the product matrix
9845: Notes:
9846: C will be created if MAT_INITIAL_MATRIX and must be destroyed by the user with MatDestroy().
9848: MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call.
9850: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9851: actually needed.
9853: This routine is currently implemented for pairs of AIJ matrices and pairs of SeqDense matrices and classes
9854: which inherit from SeqAIJ. C will be of same type as the input matrices.
9856: Level: intermediate
9858: .seealso: MatMatMult(), MatMatTransposeMult(), MatPtAP()
9859: @*/
9860: PetscErrorCode MatTransposeMatMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9861: {
9865: MatProduct_Private(A,B,scall,fill,MATPRODUCT_AtB,C);
9866: return(0);
9867: }
9869: /*@
9870: MatMatMatMult - Performs Matrix-Matrix-Matrix Multiplication D=A*B*C.
9872: Neighbor-wise Collective on Mat
9874: Input Parameters:
9875: + A - the left matrix
9876: . B - the middle matrix
9877: . C - the right matrix
9878: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9879: - fill - expected fill as ratio of nnz(D)/(nnz(A) + nnz(B)+nnz(C)), use PETSC_DEFAULT if you do not have a good estimate
9880: if the result is a dense matrix this is irrelevent
9882: Output Parameters:
9883: . D - the product matrix
9885: Notes:
9886: Unless scall is MAT_REUSE_MATRIX D will be created.
9888: MAT_REUSE_MATRIX can only be used if the matrices A, B and C have the same nonzero pattern as in the previous call
9890: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9891: actually needed.
9893: If you have many matrices with the same non-zero structure to multiply, you
9894: should use MAT_REUSE_MATRIX in all calls but the first or
9896: Level: intermediate
9898: .seealso: MatMatMult, MatPtAP()
9899: @*/
9900: PetscErrorCode MatMatMatMult(Mat A,Mat B,Mat C,MatReuse scall,PetscReal fill,Mat *D)
9901: {
9905: if (scall == MAT_REUSE_MATRIX) MatCheckProduct(*D,6);
9906: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9908: if (scall == MAT_INITIAL_MATRIX) {
9909: MatProductCreate(A,B,C,D);
9910: MatProductSetType(*D,MATPRODUCT_ABC);
9911: MatProductSetAlgorithm(*D,"default");
9912: MatProductSetFill(*D,fill);
9914: (*D)->product->api_user = PETSC_TRUE;
9915: MatProductSetFromOptions(*D);
9916: if (!(*D)->ops->productsymbolic) SETERRQ4(PetscObjectComm((PetscObject)(*D)),PETSC_ERR_SUP,"MatProduct %s not supported for A %s, B %s and C %s",MatProductTypes[MATPRODUCT_ABC],((PetscObject)A)->type_name,((PetscObject)B)->type_name,((PetscObject)C)->type_name);
9917: MatProductSymbolic(*D);
9918: } else { /* user may change input matrices when REUSE */
9919: MatProductReplaceMats(A,B,C,*D);
9920: }
9921: MatProductNumeric(*D);
9922: return(0);
9923: }
9925: /*@
9926: MatCreateRedundantMatrix - Create redundant matrices and put them into processors of subcommunicators.
9928: Collective on Mat
9930: Input Parameters:
9931: + mat - the matrix
9932: . nsubcomm - the number of subcommunicators (= number of redundant parallel or sequential matrices)
9933: . subcomm - MPI communicator split from the communicator where mat resides in (or MPI_COMM_NULL if nsubcomm is used)
9934: - reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9936: Output Parameter:
9937: . matredundant - redundant matrix
9939: Notes:
9940: MAT_REUSE_MATRIX can only be used when the nonzero structure of the
9941: original matrix has not changed from that last call to MatCreateRedundantMatrix().
9943: This routine creates the duplicated matrices in subcommunicators; you should NOT create them before
9944: calling it.
9946: Level: advanced
9949: .seealso: MatDestroy()
9950: @*/
9951: PetscErrorCode MatCreateRedundantMatrix(Mat mat,PetscInt nsubcomm,MPI_Comm subcomm,MatReuse reuse,Mat *matredundant)
9952: {
9954: MPI_Comm comm;
9955: PetscMPIInt size;
9956: PetscInt mloc_sub,nloc_sub,rstart,rend,M=mat->rmap->N,N=mat->cmap->N,bs=mat->rmap->bs;
9957: Mat_Redundant *redund=NULL;
9958: PetscSubcomm psubcomm=NULL;
9959: MPI_Comm subcomm_in=subcomm;
9960: Mat *matseq;
9961: IS isrow,iscol;
9962: PetscBool newsubcomm=PETSC_FALSE;
9966: if (nsubcomm && reuse == MAT_REUSE_MATRIX) {
9969: }
9971: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
9972: if (size == 1 || nsubcomm == 1) {
9973: if (reuse == MAT_INITIAL_MATRIX) {
9974: MatDuplicate(mat,MAT_COPY_VALUES,matredundant);
9975: } else {
9976: if (*matredundant == mat) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"MAT_REUSE_MATRIX means reuse the matrix passed in as the final argument, not the original matrix");
9977: MatCopy(mat,*matredundant,SAME_NONZERO_PATTERN);
9978: }
9979: return(0);
9980: }
9982: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9983: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9984: MatCheckPreallocated(mat,1);
9986: PetscLogEventBegin(MAT_RedundantMat,mat,0,0,0);
9987: if (subcomm_in == MPI_COMM_NULL && reuse == MAT_INITIAL_MATRIX) { /* get subcomm if user does not provide subcomm */
9988: /* create psubcomm, then get subcomm */
9989: PetscObjectGetComm((PetscObject)mat,&comm);
9990: MPI_Comm_size(comm,&size);
9991: if (nsubcomm < 1 || nsubcomm > size) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"nsubcomm must between 1 and %D",size);
9993: PetscSubcommCreate(comm,&psubcomm);
9994: PetscSubcommSetNumber(psubcomm,nsubcomm);
9995: PetscSubcommSetType(psubcomm,PETSC_SUBCOMM_CONTIGUOUS);
9996: PetscSubcommSetFromOptions(psubcomm);
9997: PetscCommDuplicate(PetscSubcommChild(psubcomm),&subcomm,NULL);
9998: newsubcomm = PETSC_TRUE;
9999: PetscSubcommDestroy(&psubcomm);
10000: }
10002: /* get isrow, iscol and a local sequential matrix matseq[0] */
10003: if (reuse == MAT_INITIAL_MATRIX) {
10004: mloc_sub = PETSC_DECIDE;
10005: nloc_sub = PETSC_DECIDE;
10006: if (bs < 1) {
10007: PetscSplitOwnership(subcomm,&mloc_sub,&M);
10008: PetscSplitOwnership(subcomm,&nloc_sub,&N);
10009: } else {
10010: PetscSplitOwnershipBlock(subcomm,bs,&mloc_sub,&M);
10011: PetscSplitOwnershipBlock(subcomm,bs,&nloc_sub,&N);
10012: }
10013: MPI_Scan(&mloc_sub,&rend,1,MPIU_INT,MPI_SUM,subcomm);
10014: rstart = rend - mloc_sub;
10015: ISCreateStride(PETSC_COMM_SELF,mloc_sub,rstart,1,&isrow);
10016: ISCreateStride(PETSC_COMM_SELF,N,0,1,&iscol);
10017: } else { /* reuse == MAT_REUSE_MATRIX */
10018: if (*matredundant == mat) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"MAT_REUSE_MATRIX means reuse the matrix passed in as the final argument, not the original matrix");
10019: /* retrieve subcomm */
10020: PetscObjectGetComm((PetscObject)(*matredundant),&subcomm);
10021: redund = (*matredundant)->redundant;
10022: isrow = redund->isrow;
10023: iscol = redund->iscol;
10024: matseq = redund->matseq;
10025: }
10026: MatCreateSubMatrices(mat,1,&isrow,&iscol,reuse,&matseq);
10028: /* get matredundant over subcomm */
10029: if (reuse == MAT_INITIAL_MATRIX) {
10030: MatCreateMPIMatConcatenateSeqMat(subcomm,matseq[0],nloc_sub,reuse,matredundant);
10032: /* create a supporting struct and attach it to C for reuse */
10033: PetscNewLog(*matredundant,&redund);
10034: (*matredundant)->redundant = redund;
10035: redund->isrow = isrow;
10036: redund->iscol = iscol;
10037: redund->matseq = matseq;
10038: if (newsubcomm) {
10039: redund->subcomm = subcomm;
10040: } else {
10041: redund->subcomm = MPI_COMM_NULL;
10042: }
10043: } else {
10044: MatCreateMPIMatConcatenateSeqMat(subcomm,matseq[0],PETSC_DECIDE,reuse,matredundant);
10045: }
10046: PetscLogEventEnd(MAT_RedundantMat,mat,0,0,0);
10047: return(0);
10048: }
10050: /*@C
10051: MatGetMultiProcBlock - Create multiple [bjacobi] 'parallel submatrices' from
10052: a given 'mat' object. Each submatrix can span multiple procs.
10054: Collective on Mat
10056: Input Parameters:
10057: + mat - the matrix
10058: . subcomm - the subcommunicator obtained by com_split(comm)
10059: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10061: Output Parameter:
10062: . subMat - 'parallel submatrices each spans a given subcomm
10064: Notes:
10065: The submatrix partition across processors is dictated by 'subComm' a
10066: communicator obtained by com_split(comm). The comm_split
10067: is not restriced to be grouped with consecutive original ranks.
10069: Due the comm_split() usage, the parallel layout of the submatrices
10070: map directly to the layout of the original matrix [wrt the local
10071: row,col partitioning]. So the original 'DiagonalMat' naturally maps
10072: into the 'DiagonalMat' of the subMat, hence it is used directly from
10073: the subMat. However the offDiagMat looses some columns - and this is
10074: reconstructed with MatSetValues()
10076: Level: advanced
10079: .seealso: MatCreateSubMatrices()
10080: @*/
10081: PetscErrorCode MatGetMultiProcBlock(Mat mat, MPI_Comm subComm, MatReuse scall,Mat *subMat)
10082: {
10084: PetscMPIInt commsize,subCommSize;
10087: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&commsize);
10088: MPI_Comm_size(subComm,&subCommSize);
10089: if (subCommSize > commsize) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"CommSize %D < SubCommZize %D",commsize,subCommSize);
10091: if (scall == MAT_REUSE_MATRIX && *subMat == mat) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"MAT_REUSE_MATRIX means reuse the matrix passed in as the final argument, not the original matrix");
10092: PetscLogEventBegin(MAT_GetMultiProcBlock,mat,0,0,0);
10093: (*mat->ops->getmultiprocblock)(mat,subComm,scall,subMat);
10094: PetscLogEventEnd(MAT_GetMultiProcBlock,mat,0,0,0);
10095: return(0);
10096: }
10098: /*@
10099: MatGetLocalSubMatrix - Gets a reference to a submatrix specified in local numbering
10101: Not Collective
10103: Input Arguments:
10104: + mat - matrix to extract local submatrix from
10105: . isrow - local row indices for submatrix
10106: - iscol - local column indices for submatrix
10108: Output Arguments:
10109: . submat - the submatrix
10111: Level: intermediate
10113: Notes:
10114: The submat should be returned with MatRestoreLocalSubMatrix().
10116: Depending on the format of mat, the returned submat may not implement MatMult(). Its communicator may be
10117: the same as mat, it may be PETSC_COMM_SELF, or some other subcomm of mat's.
10119: The submat always implements MatSetValuesLocal(). If isrow and iscol have the same block size, then
10120: MatSetValuesBlockedLocal() will also be implemented.
10122: The mat must have had a ISLocalToGlobalMapping provided to it with MatSetLocalToGlobalMapping(). Note that
10123: matrices obtained with DMCreateMatrix() generally already have the local to global mapping provided.
10125: .seealso: MatRestoreLocalSubMatrix(), MatCreateLocalRef(), MatSetLocalToGlobalMapping()
10126: @*/
10127: PetscErrorCode MatGetLocalSubMatrix(Mat mat,IS isrow,IS iscol,Mat *submat)
10128: {
10137: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Matrix must have local to global mapping provided before this call");
10139: if (mat->ops->getlocalsubmatrix) {
10140: (*mat->ops->getlocalsubmatrix)(mat,isrow,iscol,submat);
10141: } else {
10142: MatCreateLocalRef(mat,isrow,iscol,submat);
10143: }
10144: return(0);
10145: }
10147: /*@
10148: MatRestoreLocalSubMatrix - Restores a reference to a submatrix specified in local numbering
10150: Not Collective
10152: Input Arguments:
10153: mat - matrix to extract local submatrix from
10154: isrow - local row indices for submatrix
10155: iscol - local column indices for submatrix
10156: submat - the submatrix
10158: Level: intermediate
10160: .seealso: MatGetLocalSubMatrix()
10161: @*/
10162: PetscErrorCode MatRestoreLocalSubMatrix(Mat mat,IS isrow,IS iscol,Mat *submat)
10163: {
10172: if (*submat) {
10174: }
10176: if (mat->ops->restorelocalsubmatrix) {
10177: (*mat->ops->restorelocalsubmatrix)(mat,isrow,iscol,submat);
10178: } else {
10179: MatDestroy(submat);
10180: }
10181: *submat = NULL;
10182: return(0);
10183: }
10185: /* --------------------------------------------------------*/
10186: /*@
10187: MatFindZeroDiagonals - Finds all the rows of a matrix that have zero or no diagonal entry in the matrix
10189: Collective on Mat
10191: Input Parameter:
10192: . mat - the matrix
10194: Output Parameter:
10195: . is - if any rows have zero diagonals this contains the list of them
10197: Level: developer
10199: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
10200: @*/
10201: PetscErrorCode MatFindZeroDiagonals(Mat mat,IS *is)
10202: {
10208: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10209: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10211: if (!mat->ops->findzerodiagonals) {
10212: Vec diag;
10213: const PetscScalar *a;
10214: PetscInt *rows;
10215: PetscInt rStart, rEnd, r, nrow = 0;
10217: MatCreateVecs(mat, &diag, NULL);
10218: MatGetDiagonal(mat, diag);
10219: MatGetOwnershipRange(mat, &rStart, &rEnd);
10220: VecGetArrayRead(diag, &a);
10221: for (r = 0; r < rEnd-rStart; ++r) if (a[r] == 0.0) ++nrow;
10222: PetscMalloc1(nrow, &rows);
10223: nrow = 0;
10224: for (r = 0; r < rEnd-rStart; ++r) if (a[r] == 0.0) rows[nrow++] = r+rStart;
10225: VecRestoreArrayRead(diag, &a);
10226: VecDestroy(&diag);
10227: ISCreateGeneral(PetscObjectComm((PetscObject) mat), nrow, rows, PETSC_OWN_POINTER, is);
10228: } else {
10229: (*mat->ops->findzerodiagonals)(mat, is);
10230: }
10231: return(0);
10232: }
10234: /*@
10235: MatFindOffBlockDiagonalEntries - Finds all the rows of a matrix that have entries outside of the main diagonal block (defined by the matrix block size)
10237: Collective on Mat
10239: Input Parameter:
10240: . mat - the matrix
10242: Output Parameter:
10243: . is - contains the list of rows with off block diagonal entries
10245: Level: developer
10247: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
10248: @*/
10249: PetscErrorCode MatFindOffBlockDiagonalEntries(Mat mat,IS *is)
10250: {
10256: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10257: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10259: if (!mat->ops->findoffblockdiagonalentries) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a find off block diagonal entries defined",((PetscObject)mat)->type_name);
10260: (*mat->ops->findoffblockdiagonalentries)(mat,is);
10261: return(0);
10262: }
10264: /*@C
10265: MatInvertBlockDiagonal - Inverts the block diagonal entries.
10267: Collective on Mat
10269: Input Parameters:
10270: . mat - the matrix
10272: Output Parameters:
10273: . values - the block inverses in column major order (FORTRAN-like)
10275: Note:
10276: This routine is not available from Fortran.
10278: Level: advanced
10280: .seealso: MatInvertBockDiagonalMat
10281: @*/
10282: PetscErrorCode MatInvertBlockDiagonal(Mat mat,const PetscScalar **values)
10283: {
10288: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10289: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10290: if (!mat->ops->invertblockdiagonal) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for type %s",((PetscObject)mat)->type_name);
10291: (*mat->ops->invertblockdiagonal)(mat,values);
10292: return(0);
10293: }
10295: /*@C
10296: MatInvertVariableBlockDiagonal - Inverts the block diagonal entries.
10298: Collective on Mat
10300: Input Parameters:
10301: + mat - the matrix
10302: . nblocks - the number of blocks
10303: - bsizes - the size of each block
10305: Output Parameters:
10306: . values - the block inverses in column major order (FORTRAN-like)
10308: Note:
10309: This routine is not available from Fortran.
10311: Level: advanced
10313: .seealso: MatInvertBockDiagonal()
10314: @*/
10315: PetscErrorCode MatInvertVariableBlockDiagonal(Mat mat,PetscInt nblocks,const PetscInt *bsizes,PetscScalar *values)
10316: {
10321: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10322: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10323: if (!mat->ops->invertvariableblockdiagonal) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for type",((PetscObject)mat)->type_name);
10324: (*mat->ops->invertvariableblockdiagonal)(mat,nblocks,bsizes,values);
10325: return(0);
10326: }
10328: /*@
10329: MatInvertBlockDiagonalMat - set matrix C to be the inverted block diagonal of matrix A
10331: Collective on Mat
10333: Input Parameters:
10334: . A - the matrix
10336: Output Parameters:
10337: . C - matrix with inverted block diagonal of A. This matrix should be created and may have its type set.
10339: Notes: the blocksize of the matrix is used to determine the blocks on the diagonal of C
10341: Level: advanced
10343: .seealso: MatInvertBockDiagonal()
10344: @*/
10345: PetscErrorCode MatInvertBlockDiagonalMat(Mat A,Mat C)
10346: {
10347: PetscErrorCode ierr;
10348: const PetscScalar *vals;
10349: PetscInt *dnnz;
10350: PetscInt M,N,m,n,rstart,rend,bs,i,j;
10353: MatInvertBlockDiagonal(A,&vals);
10354: MatGetBlockSize(A,&bs);
10355: MatGetSize(A,&M,&N);
10356: MatGetLocalSize(A,&m,&n);
10357: MatSetSizes(C,m,n,M,N);
10358: MatSetBlockSize(C,bs);
10359: PetscMalloc1(m/bs,&dnnz);
10360: for (j = 0; j < m/bs; j++) dnnz[j] = 1;
10361: MatXAIJSetPreallocation(C,bs,dnnz,NULL,NULL,NULL);
10362: PetscFree(dnnz);
10363: MatGetOwnershipRange(C,&rstart,&rend);
10364: MatSetOption(C,MAT_ROW_ORIENTED,PETSC_FALSE);
10365: for (i = rstart/bs; i < rend/bs; i++) {
10366: MatSetValuesBlocked(C,1,&i,1,&i,&vals[(i-rstart/bs)*bs*bs],INSERT_VALUES);
10367: }
10368: MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);
10369: MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);
10370: MatSetOption(C,MAT_ROW_ORIENTED,PETSC_TRUE);
10371: return(0);
10372: }
10374: /*@C
10375: MatTransposeColoringDestroy - Destroys a coloring context for matrix product C=A*B^T that was created
10376: via MatTransposeColoringCreate().
10378: Collective on MatTransposeColoring
10380: Input Parameter:
10381: . c - coloring context
10383: Level: intermediate
10385: .seealso: MatTransposeColoringCreate()
10386: @*/
10387: PetscErrorCode MatTransposeColoringDestroy(MatTransposeColoring *c)
10388: {
10389: PetscErrorCode ierr;
10390: MatTransposeColoring matcolor=*c;
10393: if (!matcolor) return(0);
10394: if (--((PetscObject)matcolor)->refct > 0) {matcolor = NULL; return(0);}
10396: PetscFree3(matcolor->ncolumns,matcolor->nrows,matcolor->colorforrow);
10397: PetscFree(matcolor->rows);
10398: PetscFree(matcolor->den2sp);
10399: PetscFree(matcolor->colorforcol);
10400: PetscFree(matcolor->columns);
10401: if (matcolor->brows>0) {
10402: PetscFree(matcolor->lstart);
10403: }
10404: PetscHeaderDestroy(c);
10405: return(0);
10406: }
10408: /*@C
10409: MatTransColoringApplySpToDen - Given a symbolic matrix product C=A*B^T for which
10410: a MatTransposeColoring context has been created, computes a dense B^T by Apply
10411: MatTransposeColoring to sparse B.
10413: Collective on MatTransposeColoring
10415: Input Parameters:
10416: + B - sparse matrix B
10417: . Btdense - symbolic dense matrix B^T
10418: - coloring - coloring context created with MatTransposeColoringCreate()
10420: Output Parameter:
10421: . Btdense - dense matrix B^T
10423: Level: advanced
10425: Notes:
10426: These are used internally for some implementations of MatRARt()
10428: .seealso: MatTransposeColoringCreate(), MatTransposeColoringDestroy(), MatTransColoringApplyDenToSp()
10430: @*/
10431: PetscErrorCode MatTransColoringApplySpToDen(MatTransposeColoring coloring,Mat B,Mat Btdense)
10432: {
10440: if (!B->ops->transcoloringapplysptoden) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for this matrix type %s",((PetscObject)B)->type_name);
10441: (B->ops->transcoloringapplysptoden)(coloring,B,Btdense);
10442: return(0);
10443: }
10445: /*@C
10446: MatTransColoringApplyDenToSp - Given a symbolic matrix product Csp=A*B^T for which
10447: a MatTransposeColoring context has been created and a dense matrix Cden=A*Btdense
10448: in which Btdens is obtained from MatTransColoringApplySpToDen(), recover sparse matrix
10449: Csp from Cden.
10451: Collective on MatTransposeColoring
10453: Input Parameters:
10454: + coloring - coloring context created with MatTransposeColoringCreate()
10455: - Cden - matrix product of a sparse matrix and a dense matrix Btdense
10457: Output Parameter:
10458: . Csp - sparse matrix
10460: Level: advanced
10462: Notes:
10463: These are used internally for some implementations of MatRARt()
10465: .seealso: MatTransposeColoringCreate(), MatTransposeColoringDestroy(), MatTransColoringApplySpToDen()
10467: @*/
10468: PetscErrorCode MatTransColoringApplyDenToSp(MatTransposeColoring matcoloring,Mat Cden,Mat Csp)
10469: {
10477: if (!Csp->ops->transcoloringapplydentosp) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for this matrix type %s",((PetscObject)Csp)->type_name);
10478: (Csp->ops->transcoloringapplydentosp)(matcoloring,Cden,Csp);
10479: MatAssemblyBegin(Csp,MAT_FINAL_ASSEMBLY);
10480: MatAssemblyEnd(Csp,MAT_FINAL_ASSEMBLY);
10481: return(0);
10482: }
10484: /*@C
10485: MatTransposeColoringCreate - Creates a matrix coloring context for matrix product C=A*B^T.
10487: Collective on Mat
10489: Input Parameters:
10490: + mat - the matrix product C
10491: - iscoloring - the coloring of the matrix; usually obtained with MatColoringCreate() or DMCreateColoring()
10493: Output Parameter:
10494: . color - the new coloring context
10496: Level: intermediate
10498: .seealso: MatTransposeColoringDestroy(), MatTransColoringApplySpToDen(),
10499: MatTransColoringApplyDenToSp()
10500: @*/
10501: PetscErrorCode MatTransposeColoringCreate(Mat mat,ISColoring iscoloring,MatTransposeColoring *color)
10502: {
10503: MatTransposeColoring c;
10504: MPI_Comm comm;
10505: PetscErrorCode ierr;
10508: PetscLogEventBegin(MAT_TransposeColoringCreate,mat,0,0,0);
10509: PetscObjectGetComm((PetscObject)mat,&comm);
10510: PetscHeaderCreate(c,MAT_TRANSPOSECOLORING_CLASSID,"MatTransposeColoring","Matrix product C=A*B^T via coloring","Mat",comm,MatTransposeColoringDestroy,NULL);
10512: c->ctype = iscoloring->ctype;
10513: if (mat->ops->transposecoloringcreate) {
10514: (*mat->ops->transposecoloringcreate)(mat,iscoloring,c);
10515: } else SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Code not yet written for matrix type %s",((PetscObject)mat)->type_name);
10517: *color = c;
10518: PetscLogEventEnd(MAT_TransposeColoringCreate,mat,0,0,0);
10519: return(0);
10520: }
10522: /*@
10523: MatGetNonzeroState - Returns a 64 bit integer representing the current state of nonzeros in the matrix. If the
10524: matrix has had no new nonzero locations added to the matrix since the previous call then the value will be the
10525: same, otherwise it will be larger
10527: Not Collective
10529: Input Parameter:
10530: . A - the matrix
10532: Output Parameter:
10533: . state - the current state
10535: Notes:
10536: You can only compare states from two different calls to the SAME matrix, you cannot compare calls between
10537: different matrices
10539: Level: intermediate
10541: @*/
10542: PetscErrorCode MatGetNonzeroState(Mat mat,PetscObjectState *state)
10543: {
10546: *state = mat->nonzerostate;
10547: return(0);
10548: }
10550: /*@
10551: MatCreateMPIMatConcatenateSeqMat - Creates a single large PETSc matrix by concatenating sequential
10552: matrices from each processor
10554: Collective
10556: Input Parameters:
10557: + comm - the communicators the parallel matrix will live on
10558: . seqmat - the input sequential matrices
10559: . n - number of local columns (or PETSC_DECIDE)
10560: - reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10562: Output Parameter:
10563: . mpimat - the parallel matrix generated
10565: Level: advanced
10567: Notes:
10568: The number of columns of the matrix in EACH processor MUST be the same.
10570: @*/
10571: PetscErrorCode MatCreateMPIMatConcatenateSeqMat(MPI_Comm comm,Mat seqmat,PetscInt n,MatReuse reuse,Mat *mpimat)
10572: {
10576: if (!seqmat->ops->creatempimatconcatenateseqmat) SETERRQ1(PetscObjectComm((PetscObject)seqmat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)seqmat)->type_name);
10577: if (reuse == MAT_REUSE_MATRIX && seqmat == *mpimat) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"MAT_REUSE_MATRIX means reuse the matrix passed in as the final argument, not the original matrix");
10579: PetscLogEventBegin(MAT_Merge,seqmat,0,0,0);
10580: (*seqmat->ops->creatempimatconcatenateseqmat)(comm,seqmat,n,reuse,mpimat);
10581: PetscLogEventEnd(MAT_Merge,seqmat,0,0,0);
10582: return(0);
10583: }
10585: /*@
10586: MatSubdomainsCreateCoalesce - Creates index subdomains by coalescing adjacent
10587: ranks' ownership ranges.
10589: Collective on A
10591: Input Parameters:
10592: + A - the matrix to create subdomains from
10593: - N - requested number of subdomains
10596: Output Parameters:
10597: + n - number of subdomains resulting on this rank
10598: - iss - IS list with indices of subdomains on this rank
10600: Level: advanced
10602: Notes:
10603: number of subdomains must be smaller than the communicator size
10604: @*/
10605: PetscErrorCode MatSubdomainsCreateCoalesce(Mat A,PetscInt N,PetscInt *n,IS *iss[])
10606: {
10607: MPI_Comm comm,subcomm;
10608: PetscMPIInt size,rank,color;
10609: PetscInt rstart,rend,k;
10610: PetscErrorCode ierr;
10613: PetscObjectGetComm((PetscObject)A,&comm);
10614: MPI_Comm_size(comm,&size);
10615: MPI_Comm_rank(comm,&rank);
10616: if (N < 1 || N >= (PetscInt)size) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"number of subdomains must be > 0 and < %D, got N = %D",size,N);
10617: *n = 1;
10618: k = ((PetscInt)size)/N + ((PetscInt)size%N>0); /* There are up to k ranks to a color */
10619: color = rank/k;
10620: MPI_Comm_split(comm,color,rank,&subcomm);
10621: PetscMalloc1(1,iss);
10622: MatGetOwnershipRange(A,&rstart,&rend);
10623: ISCreateStride(subcomm,rend-rstart,rstart,1,iss[0]);
10624: MPI_Comm_free(&subcomm);
10625: return(0);
10626: }
10628: /*@
10629: MatGalerkin - Constructs the coarse grid problem via Galerkin projection.
10631: If the interpolation and restriction operators are the same, uses MatPtAP.
10632: If they are not the same, use MatMatMatMult.
10634: Once the coarse grid problem is constructed, correct for interpolation operators
10635: that are not of full rank, which can legitimately happen in the case of non-nested
10636: geometric multigrid.
10638: Input Parameters:
10639: + restrct - restriction operator
10640: . dA - fine grid matrix
10641: . interpolate - interpolation operator
10642: . reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10643: - fill - expected fill, use PETSC_DEFAULT if you do not have a good estimate
10645: Output Parameters:
10646: . A - the Galerkin coarse matrix
10648: Options Database Key:
10649: . -pc_mg_galerkin <both,pmat,mat,none>
10651: Level: developer
10653: .seealso: MatPtAP(), MatMatMatMult()
10654: @*/
10655: PetscErrorCode MatGalerkin(Mat restrct, Mat dA, Mat interpolate, MatReuse reuse, PetscReal fill, Mat *A)
10656: {
10658: IS zerorows;
10659: Vec diag;
10662: if (reuse == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
10663: /* Construct the coarse grid matrix */
10664: if (interpolate == restrct) {
10665: MatPtAP(dA,interpolate,reuse,fill,A);
10666: } else {
10667: MatMatMatMult(restrct,dA,interpolate,reuse,fill,A);
10668: }
10670: /* If the interpolation matrix is not of full rank, A will have zero rows.
10671: This can legitimately happen in the case of non-nested geometric multigrid.
10672: In that event, we set the rows of the matrix to the rows of the identity,
10673: ignoring the equations (as the RHS will also be zero). */
10675: MatFindZeroRows(*A, &zerorows);
10677: if (zerorows != NULL) { /* if there are any zero rows */
10678: MatCreateVecs(*A, &diag, NULL);
10679: MatGetDiagonal(*A, diag);
10680: VecISSet(diag, zerorows, 1.0);
10681: MatDiagonalSet(*A, diag, INSERT_VALUES);
10682: VecDestroy(&diag);
10683: ISDestroy(&zerorows);
10684: }
10685: return(0);
10686: }
10688: /*@C
10689: MatSetOperation - Allows user to set a matrix operation for any matrix type
10691: Logically Collective on Mat
10693: Input Parameters:
10694: + mat - the matrix
10695: . op - the name of the operation
10696: - f - the function that provides the operation
10698: Level: developer
10700: Usage:
10701: $ extern PetscErrorCode usermult(Mat,Vec,Vec);
10702: $ MatCreateXXX(comm,...&A);
10703: $ MatSetOperation(A,MATOP_MULT,(void(*)(void))usermult);
10705: Notes:
10706: See the file include/petscmat.h for a complete list of matrix
10707: operations, which all have the form MATOP_<OPERATION>, where
10708: <OPERATION> is the name (in all capital letters) of the
10709: user interface routine (e.g., MatMult() -> MATOP_MULT).
10711: All user-provided functions (except for MATOP_DESTROY) should have the same calling
10712: sequence as the usual matrix interface routines, since they
10713: are intended to be accessed via the usual matrix interface
10714: routines, e.g.,
10715: $ MatMult(Mat,Vec,Vec) -> usermult(Mat,Vec,Vec)
10717: In particular each function MUST return an error code of 0 on success and
10718: nonzero on failure.
10720: This routine is distinct from MatShellSetOperation() in that it can be called on any matrix type.
10722: .seealso: MatGetOperation(), MatCreateShell(), MatShellSetContext(), MatShellSetOperation()
10723: @*/
10724: PetscErrorCode MatSetOperation(Mat mat,MatOperation op,void (*f)(void))
10725: {
10728: if (op == MATOP_VIEW && !mat->ops->viewnative && f != (void (*)(void))(mat->ops->view)) {
10729: mat->ops->viewnative = mat->ops->view;
10730: }
10731: (((void(**)(void))mat->ops)[op]) = f;
10732: return(0);
10733: }
10735: /*@C
10736: MatGetOperation - Gets a matrix operation for any matrix type.
10738: Not Collective
10740: Input Parameters:
10741: + mat - the matrix
10742: - op - the name of the operation
10744: Output Parameter:
10745: . f - the function that provides the operation
10747: Level: developer
10749: Usage:
10750: $ PetscErrorCode (*usermult)(Mat,Vec,Vec);
10751: $ MatGetOperation(A,MATOP_MULT,(void(**)(void))&usermult);
10753: Notes:
10754: See the file include/petscmat.h for a complete list of matrix
10755: operations, which all have the form MATOP_<OPERATION>, where
10756: <OPERATION> is the name (in all capital letters) of the
10757: user interface routine (e.g., MatMult() -> MATOP_MULT).
10759: This routine is distinct from MatShellGetOperation() in that it can be called on any matrix type.
10761: .seealso: MatSetOperation(), MatCreateShell(), MatShellGetContext(), MatShellGetOperation()
10762: @*/
10763: PetscErrorCode MatGetOperation(Mat mat,MatOperation op,void(**f)(void))
10764: {
10767: *f = (((void (**)(void))mat->ops)[op]);
10768: return(0);
10769: }
10771: /*@
10772: MatHasOperation - Determines whether the given matrix supports the particular
10773: operation.
10775: Not Collective
10777: Input Parameters:
10778: + mat - the matrix
10779: - op - the operation, for example, MATOP_GET_DIAGONAL
10781: Output Parameter:
10782: . has - either PETSC_TRUE or PETSC_FALSE
10784: Level: advanced
10786: Notes:
10787: See the file include/petscmat.h for a complete list of matrix
10788: operations, which all have the form MATOP_<OPERATION>, where
10789: <OPERATION> is the name (in all capital letters) of the
10790: user-level routine. E.g., MatNorm() -> MATOP_NORM.
10792: .seealso: MatCreateShell()
10793: @*/
10794: PetscErrorCode MatHasOperation(Mat mat,MatOperation op,PetscBool *has)
10795: {
10800: /* symbolic product can be set before matrix type */
10803: if (mat->ops->hasoperation) {
10804: (*mat->ops->hasoperation)(mat,op,has);
10805: } else {
10806: if (((void**)mat->ops)[op]) *has = PETSC_TRUE;
10807: else {
10808: *has = PETSC_FALSE;
10809: if (op == MATOP_CREATE_SUBMATRIX) {
10810: PetscMPIInt size;
10812: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
10813: if (size == 1) {
10814: MatHasOperation(mat,MATOP_CREATE_SUBMATRICES,has);
10815: }
10816: }
10817: }
10818: }
10819: return(0);
10820: }
10822: /*@
10823: MatHasCongruentLayouts - Determines whether the rows and columns layouts
10824: of the matrix are congruent
10826: Collective on mat
10828: Input Parameters:
10829: . mat - the matrix
10831: Output Parameter:
10832: . cong - either PETSC_TRUE or PETSC_FALSE
10834: Level: beginner
10836: Notes:
10838: .seealso: MatCreate(), MatSetSizes()
10839: @*/
10840: PetscErrorCode MatHasCongruentLayouts(Mat mat,PetscBool *cong)
10841: {
10848: if (!mat->rmap || !mat->cmap) {
10849: *cong = mat->rmap == mat->cmap ? PETSC_TRUE : PETSC_FALSE;
10850: return(0);
10851: }
10852: if (mat->congruentlayouts == PETSC_DECIDE) { /* first time we compare rows and cols layouts */
10853: PetscLayoutCompare(mat->rmap,mat->cmap,cong);
10854: if (*cong) mat->congruentlayouts = 1;
10855: else mat->congruentlayouts = 0;
10856: } else *cong = mat->congruentlayouts ? PETSC_TRUE : PETSC_FALSE;
10857: return(0);
10858: }
10860: PetscErrorCode MatSetInf(Mat A)
10861: {
10865: if (!A->ops->setinf) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"No support for this operation for this matrix type");
10866: (*A->ops->setinf)(A);
10867: return(0);
10868: }