Actual source code: matrix.c
petsc-3.12.3 2020-01-03
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_AssemblyEnd, MAT_SetValues, MAT_GetValues, MAT_GetRow, MAT_GetRowIJ, MAT_CreateSubMats, MAT_GetOrdering, MAT_RedundantMat, MAT_GetSeqNonzeroStructure;
21: PetscLogEvent MAT_IncreaseOverlap, MAT_Partitioning, MAT_PartitioningND, MAT_Coarsen, MAT_ZeroEntries, MAT_Load, MAT_View, MAT_AXPY, MAT_FDColoringCreate;
22: PetscLogEvent MAT_FDColoringSetUp, MAT_FDColoringApply,MAT_Transpose,MAT_FDColoringFunction, MAT_CreateSubMat;
23: PetscLogEvent MAT_TransposeColoringCreate;
24: PetscLogEvent MAT_MatMult, MAT_MatMultSymbolic, MAT_MatMultNumeric;
25: PetscLogEvent MAT_PtAP, MAT_PtAPSymbolic, MAT_PtAPNumeric,MAT_RARt, MAT_RARtSymbolic, MAT_RARtNumeric;
26: PetscLogEvent MAT_MatTransposeMult, MAT_MatTransposeMultSymbolic, MAT_MatTransposeMultNumeric;
27: PetscLogEvent MAT_TransposeMatMult, MAT_TransposeMatMultSymbolic, MAT_TransposeMatMultNumeric;
28: PetscLogEvent MAT_MatMatMult, MAT_MatMatMultSymbolic, MAT_MatMatMultNumeric;
29: PetscLogEvent MAT_MultHermitianTranspose,MAT_MultHermitianTransposeAdd;
30: PetscLogEvent MAT_Getsymtranspose, MAT_Getsymtransreduced, MAT_GetBrowsOfAcols;
31: PetscLogEvent MAT_GetBrowsOfAocols, MAT_Getlocalmat, MAT_Getlocalmatcondensed, MAT_Seqstompi, MAT_Seqstompinum, MAT_Seqstompisym;
32: PetscLogEvent MAT_Applypapt, MAT_Applypapt_numeric, MAT_Applypapt_symbolic, MAT_GetSequentialNonzeroStructure;
33: PetscLogEvent MAT_GetMultiProcBlock;
34: PetscLogEvent MAT_CUSPARSECopyToGPU, MAT_SetValuesBatch;
35: PetscLogEvent MAT_ViennaCLCopyToGPU;
36: PetscLogEvent MAT_DenseCopyToGPU, MAT_DenseCopyFromGPU;
37: PetscLogEvent MAT_Merge,MAT_Residual,MAT_SetRandom;
38: PetscLogEvent MAT_FactorFactS,MAT_FactorInvS;
39: PetscLogEvent MATCOLORING_Apply,MATCOLORING_Comm,MATCOLORING_Local,MATCOLORING_ISCreate,MATCOLORING_SetUp,MATCOLORING_Weights;
41: const char *const MatFactorTypes[] = {"NONE","LU","CHOLESKY","ILU","ICC","ILUDT","MatFactorType","MAT_FACTOR_",0};
43: /*@
44: 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,
45: for sparse matrices that already have locations it fills the locations with random numbers
47: Logically Collective on Mat
49: Input Parameters:
50: + x - the matrix
51: - rctx - the random number context, formed by PetscRandomCreate(), or NULL and
52: it will create one internally.
54: Output Parameter:
55: . x - the matrix
57: Example of Usage:
58: .vb
59: PetscRandomCreate(PETSC_COMM_WORLD,&rctx);
60: MatSetRandom(x,rctx);
61: PetscRandomDestroy(rctx);
62: .ve
64: Level: intermediate
67: .seealso: MatZeroEntries(), MatSetValues(), PetscRandomCreate(), PetscRandomDestroy()
68: @*/
69: PetscErrorCode MatSetRandom(Mat x,PetscRandom rctx)
70: {
72: PetscRandom randObj = NULL;
79: if (!x->ops->setrandom) SETERRQ1(PetscObjectComm((PetscObject)x),PETSC_ERR_SUP,"Mat type %s",((PetscObject)x)->type_name);
81: if (!rctx) {
82: MPI_Comm comm;
83: PetscObjectGetComm((PetscObject)x,&comm);
84: PetscRandomCreate(comm,&randObj);
85: PetscRandomSetFromOptions(randObj);
86: rctx = randObj;
87: }
89: PetscLogEventBegin(MAT_SetRandom,x,rctx,0,0);
90: (*x->ops->setrandom)(x,rctx);
91: PetscLogEventEnd(MAT_SetRandom,x,rctx,0,0);
93: MatAssemblyBegin(x, MAT_FINAL_ASSEMBLY);
94: MatAssemblyEnd(x, MAT_FINAL_ASSEMBLY);
95: PetscRandomDestroy(&randObj);
96: return(0);
97: }
99: /*@
100: MatFactorGetErrorZeroPivot - returns the pivot value that was determined to be zero and the row it occurred in
102: Logically Collective on Mat
104: Input Parameters:
105: . mat - the factored matrix
107: Output Parameter:
108: + pivot - the pivot value computed
109: - row - the row that the zero pivot occurred. Note that this row must be interpreted carefully due to row reorderings and which processes
110: the share the matrix
112: Level: advanced
114: Notes:
115: This routine does not work for factorizations done with external packages.
116: This routine should only be called if MatGetFactorError() returns a value of MAT_FACTOR_NUMERIC_ZEROPIVOT
118: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
120: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatFactorSymbolic(), MatFactorClearError(), MatFactorGetErrorZeroPivot()
121: @*/
122: PetscErrorCode MatFactorGetErrorZeroPivot(Mat mat,PetscReal *pivot,PetscInt *row)
123: {
126: *pivot = mat->factorerror_zeropivot_value;
127: *row = mat->factorerror_zeropivot_row;
128: return(0);
129: }
131: /*@
132: MatFactorGetError - gets the error code from a factorization
134: Logically Collective on Mat
136: Input Parameters:
137: . mat - the factored matrix
139: Output Parameter:
140: . err - the error code
142: Level: advanced
144: Notes:
145: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
147: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatFactorSymbolic(), MatFactorClearError(), MatFactorGetErrorZeroPivot()
148: @*/
149: PetscErrorCode MatFactorGetError(Mat mat,MatFactorError *err)
150: {
153: *err = mat->factorerrortype;
154: return(0);
155: }
157: /*@
158: MatFactorClearError - clears the error code in a factorization
160: Logically Collective on Mat
162: Input Parameter:
163: . mat - the factored matrix
165: Level: developer
167: Notes:
168: This can be called on non-factored matrices that come from, for example, matrices used in SOR.
170: .seealso: MatZeroEntries(), MatFactor(), MatGetFactor(), MatFactorSymbolic(), MatFactorGetError(), MatFactorGetErrorZeroPivot()
171: @*/
172: PetscErrorCode MatFactorClearError(Mat mat)
173: {
176: mat->factorerrortype = MAT_FACTOR_NOERROR;
177: mat->factorerror_zeropivot_value = 0.0;
178: mat->factorerror_zeropivot_row = 0;
179: return(0);
180: }
182: PETSC_INTERN PetscErrorCode MatFindNonzeroRowsOrCols_Basic(Mat mat,PetscBool cols,PetscReal tol,IS *nonzero)
183: {
184: PetscErrorCode ierr;
185: Vec r,l;
186: const PetscScalar *al;
187: PetscInt i,nz,gnz,N,n;
190: MatCreateVecs(mat,&r,&l);
191: if (!cols) { /* nonzero rows */
192: MatGetSize(mat,&N,NULL);
193: MatGetLocalSize(mat,&n,NULL);
194: VecSet(l,0.0);
195: VecSetRandom(r,NULL);
196: MatMult(mat,r,l);
197: VecGetArrayRead(l,&al);
198: } else { /* nonzero columns */
199: MatGetSize(mat,NULL,&N);
200: MatGetLocalSize(mat,NULL,&n);
201: VecSet(r,0.0);
202: VecSetRandom(l,NULL);
203: MatMultTranspose(mat,l,r);
204: VecGetArrayRead(r,&al);
205: }
206: if (tol <= 0.0) { for (i=0,nz=0;i<n;i++) if (al[i] != 0.0) nz++; }
207: else { for (i=0,nz=0;i<n;i++) if (PetscAbsScalar(al[i]) > tol) nz++; }
208: MPIU_Allreduce(&nz,&gnz,1,MPIU_INT,MPI_SUM,PetscObjectComm((PetscObject)mat));
209: if (gnz != N) {
210: PetscInt *nzr;
211: PetscMalloc1(nz,&nzr);
212: if (nz) {
213: if (tol < 0) { for (i=0,nz=0;i<n;i++) if (al[i] != 0.0) nzr[nz++] = i; }
214: else { for (i=0,nz=0;i<n;i++) if (PetscAbsScalar(al[i]) > tol) nzr[nz++] = i; }
215: }
216: ISCreateGeneral(PetscObjectComm((PetscObject)mat),nz,nzr,PETSC_OWN_POINTER,nonzero);
217: } else *nonzero = NULL;
218: if (!cols) { /* nonzero rows */
219: VecRestoreArrayRead(l,&al);
220: } else {
221: VecRestoreArrayRead(r,&al);
222: }
223: VecDestroy(&l);
224: VecDestroy(&r);
225: return(0);
226: }
228: /*@
229: MatFindNonzeroRows - Locate all rows that are not completely zero in the matrix
231: Input Parameter:
232: . A - the matrix
234: Output Parameter:
235: . keptrows - the rows that are not completely zero
237: Notes:
238: keptrows is set to NULL if all rows are nonzero.
240: Level: intermediate
242: @*/
243: PetscErrorCode MatFindNonzeroRows(Mat mat,IS *keptrows)
244: {
251: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
252: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
253: if (!mat->ops->findnonzerorows) {
254: MatFindNonzeroRowsOrCols_Basic(mat,PETSC_FALSE,0.0,keptrows);
255: } else {
256: (*mat->ops->findnonzerorows)(mat,keptrows);
257: }
258: return(0);
259: }
261: /*@
262: MatFindZeroRows - Locate all rows that are completely zero in the matrix
264: Input Parameter:
265: . A - the matrix
267: Output Parameter:
268: . zerorows - the rows that are completely zero
270: Notes:
271: zerorows is set to NULL if no rows are zero.
273: Level: intermediate
275: @*/
276: PetscErrorCode MatFindZeroRows(Mat mat,IS *zerorows)
277: {
279: IS keptrows;
280: PetscInt m, n;
285: MatFindNonzeroRows(mat, &keptrows);
286: /* MatFindNonzeroRows sets keptrows to NULL if there are no zero rows.
287: In keeping with this convention, we set zerorows to NULL if there are no zero
288: rows. */
289: if (keptrows == NULL) {
290: *zerorows = NULL;
291: } else {
292: MatGetOwnershipRange(mat,&m,&n);
293: ISComplement(keptrows,m,n,zerorows);
294: ISDestroy(&keptrows);
295: }
296: return(0);
297: }
299: /*@
300: MatGetDiagonalBlock - Returns the part of the matrix associated with the on-process coupling
302: Not Collective
304: Input Parameters:
305: . A - the matrix
307: Output Parameters:
308: . a - the diagonal part (which is a SEQUENTIAL matrix)
310: Notes:
311: see the manual page for MatCreateAIJ() for more information on the "diagonal part" of the matrix.
312: Use caution, as the reference count on the returned matrix is not incremented and it is used as
313: part of the containing MPI Mat's normal operation.
315: Level: advanced
317: @*/
318: PetscErrorCode MatGetDiagonalBlock(Mat A,Mat *a)
319: {
326: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
327: if (!A->ops->getdiagonalblock) {
328: PetscMPIInt size;
329: MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);
330: if (size == 1) {
331: *a = A;
332: return(0);
333: } else SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Not coded for matrix type %s",((PetscObject)A)->type_name);
334: }
335: (*A->ops->getdiagonalblock)(A,a);
336: return(0);
337: }
339: /*@
340: MatGetTrace - Gets the trace of a matrix. The sum of the diagonal entries.
342: Collective on Mat
344: Input Parameters:
345: . mat - the matrix
347: Output Parameter:
348: . trace - the sum of the diagonal entries
350: Level: advanced
352: @*/
353: PetscErrorCode MatGetTrace(Mat mat,PetscScalar *trace)
354: {
356: Vec diag;
359: MatCreateVecs(mat,&diag,NULL);
360: MatGetDiagonal(mat,diag);
361: VecSum(diag,trace);
362: VecDestroy(&diag);
363: return(0);
364: }
366: /*@
367: MatRealPart - Zeros out the imaginary part of the matrix
369: Logically Collective on Mat
371: Input Parameters:
372: . mat - the matrix
374: Level: advanced
377: .seealso: MatImaginaryPart()
378: @*/
379: PetscErrorCode MatRealPart(Mat mat)
380: {
386: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
387: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
388: if (!mat->ops->realpart) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
389: MatCheckPreallocated(mat,1);
390: (*mat->ops->realpart)(mat);
391: return(0);
392: }
394: /*@C
395: MatGetGhosts - Get the global index of all ghost nodes defined by the sparse matrix
397: Collective on Mat
399: Input Parameter:
400: . mat - the matrix
402: Output Parameters:
403: + nghosts - number of ghosts (note for BAIJ matrices there is one ghost for each block)
404: - ghosts - the global indices of the ghost points
406: Notes:
407: the nghosts and ghosts are suitable to pass into VecCreateGhost()
409: Level: advanced
411: @*/
412: PetscErrorCode MatGetGhosts(Mat mat,PetscInt *nghosts,const PetscInt *ghosts[])
413: {
419: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
420: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
421: if (!mat->ops->getghosts) {
422: if (nghosts) *nghosts = 0;
423: if (ghosts) *ghosts = 0;
424: } else {
425: (*mat->ops->getghosts)(mat,nghosts,ghosts);
426: }
427: return(0);
428: }
431: /*@
432: MatImaginaryPart - Moves the imaginary part of the matrix to the real part and zeros the imaginary part
434: Logically Collective on Mat
436: Input Parameters:
437: . mat - the matrix
439: Level: advanced
442: .seealso: MatRealPart()
443: @*/
444: PetscErrorCode MatImaginaryPart(Mat mat)
445: {
451: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
452: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
453: if (!mat->ops->imaginarypart) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
454: MatCheckPreallocated(mat,1);
455: (*mat->ops->imaginarypart)(mat);
456: return(0);
457: }
459: /*@
460: MatMissingDiagonal - Determine if sparse matrix is missing a diagonal entry (or block entry for BAIJ matrices)
462: Not Collective
464: Input Parameter:
465: . mat - the matrix
467: Output Parameters:
468: + missing - is any diagonal missing
469: - dd - first diagonal entry that is missing (optional) on this process
471: Level: advanced
474: .seealso: MatRealPart()
475: @*/
476: PetscErrorCode MatMissingDiagonal(Mat mat,PetscBool *missing,PetscInt *dd)
477: {
484: if (!mat->assembled) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix %s",((PetscObject)mat)->type_name);
485: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
486: if (!mat->ops->missingdiagonal) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
487: (*mat->ops->missingdiagonal)(mat,missing,dd);
488: return(0);
489: }
491: /*@C
492: MatGetRow - Gets a row of a matrix. You MUST call MatRestoreRow()
493: for each row that you get to ensure that your application does
494: not bleed memory.
496: Not Collective
498: Input Parameters:
499: + mat - the matrix
500: - row - the row to get
502: Output Parameters:
503: + ncols - if not NULL, the number of nonzeros in the row
504: . cols - if not NULL, the column numbers
505: - vals - if not NULL, the values
507: Notes:
508: This routine is provided for people who need to have direct access
509: to the structure of a matrix. We hope that we provide enough
510: high-level matrix routines that few users will need it.
512: MatGetRow() always returns 0-based column indices, regardless of
513: whether the internal representation is 0-based (default) or 1-based.
515: For better efficiency, set cols and/or vals to NULL if you do
516: not wish to extract these quantities.
518: The user can only examine the values extracted with MatGetRow();
519: the values cannot be altered. To change the matrix entries, one
520: must use MatSetValues().
522: You can only have one call to MatGetRow() outstanding for a particular
523: matrix at a time, per processor. MatGetRow() can only obtain rows
524: associated with the given processor, it cannot get rows from the
525: other processors; for that we suggest using MatCreateSubMatrices(), then
526: MatGetRow() on the submatrix. The row index passed to MatGetRow()
527: is in the global number of rows.
529: Fortran Notes:
530: The calling sequence from Fortran is
531: .vb
532: MatGetRow(matrix,row,ncols,cols,values,ierr)
533: Mat matrix (input)
534: integer row (input)
535: integer ncols (output)
536: integer cols(maxcols) (output)
537: double precision (or double complex) values(maxcols) output
538: .ve
539: where maxcols >= maximum nonzeros in any row of the matrix.
542: Caution:
543: Do not try to change the contents of the output arrays (cols and vals).
544: In some cases, this may corrupt the matrix.
546: Level: advanced
548: .seealso: MatRestoreRow(), MatSetValues(), MatGetValues(), MatCreateSubMatrices(), MatGetDiagonal()
549: @*/
550: PetscErrorCode MatGetRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
551: {
553: PetscInt incols;
558: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
559: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
560: if (!mat->ops->getrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
561: MatCheckPreallocated(mat,1);
562: PetscLogEventBegin(MAT_GetRow,mat,0,0,0);
563: (*mat->ops->getrow)(mat,row,&incols,(PetscInt**)cols,(PetscScalar**)vals);
564: if (ncols) *ncols = incols;
565: PetscLogEventEnd(MAT_GetRow,mat,0,0,0);
566: return(0);
567: }
569: /*@
570: MatConjugate - replaces the matrix values with their complex conjugates
572: Logically Collective on Mat
574: Input Parameters:
575: . mat - the matrix
577: Level: advanced
579: .seealso: VecConjugate()
580: @*/
581: PetscErrorCode MatConjugate(Mat mat)
582: {
583: #if defined(PETSC_USE_COMPLEX)
588: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
589: 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);
590: (*mat->ops->conjugate)(mat);
591: #else
593: #endif
594: return(0);
595: }
597: /*@C
598: MatRestoreRow - Frees any temporary space allocated by MatGetRow().
600: Not Collective
602: Input Parameters:
603: + mat - the matrix
604: . row - the row to get
605: . ncols, cols - the number of nonzeros and their columns
606: - vals - if nonzero the column values
608: Notes:
609: This routine should be called after you have finished examining the entries.
611: This routine zeros out ncols, cols, and vals. This is to prevent accidental
612: us of the array after it has been restored. If you pass NULL, it will
613: not zero the pointers. Use of cols or vals after MatRestoreRow is invalid.
615: Fortran Notes:
616: The calling sequence from Fortran is
617: .vb
618: MatRestoreRow(matrix,row,ncols,cols,values,ierr)
619: Mat matrix (input)
620: integer row (input)
621: integer ncols (output)
622: integer cols(maxcols) (output)
623: double precision (or double complex) values(maxcols) output
624: .ve
625: Where maxcols >= maximum nonzeros in any row of the matrix.
627: In Fortran MatRestoreRow() MUST be called after MatGetRow()
628: before another call to MatGetRow() can be made.
630: Level: advanced
632: .seealso: MatGetRow()
633: @*/
634: PetscErrorCode MatRestoreRow(Mat mat,PetscInt row,PetscInt *ncols,const PetscInt *cols[],const PetscScalar *vals[])
635: {
641: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
642: if (!mat->ops->restorerow) return(0);
643: (*mat->ops->restorerow)(mat,row,ncols,(PetscInt **)cols,(PetscScalar **)vals);
644: if (ncols) *ncols = 0;
645: if (cols) *cols = NULL;
646: if (vals) *vals = NULL;
647: return(0);
648: }
650: /*@
651: MatGetRowUpperTriangular - Sets a flag to enable calls to MatGetRow() for matrix in MATSBAIJ format.
652: You should call MatRestoreRowUpperTriangular() after calling MatGetRow/MatRestoreRow() to disable the flag.
654: Not Collective
656: Input Parameters:
657: . mat - the matrix
659: Notes:
660: 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.
662: Level: advanced
664: .seealso: MatRestoreRowUpperTriangular()
665: @*/
666: PetscErrorCode MatGetRowUpperTriangular(Mat mat)
667: {
673: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
674: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
675: MatCheckPreallocated(mat,1);
676: if (!mat->ops->getrowuppertriangular) return(0);
677: (*mat->ops->getrowuppertriangular)(mat);
678: return(0);
679: }
681: /*@
682: MatRestoreRowUpperTriangular - Disable calls to MatGetRow() for matrix in MATSBAIJ format.
684: Not Collective
686: Input Parameters:
687: . mat - the matrix
689: Notes:
690: This routine should be called after you have finished MatGetRow/MatRestoreRow().
693: Level: advanced
695: .seealso: MatGetRowUpperTriangular()
696: @*/
697: PetscErrorCode MatRestoreRowUpperTriangular(Mat mat)
698: {
704: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
705: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
706: MatCheckPreallocated(mat,1);
707: if (!mat->ops->restorerowuppertriangular) return(0);
708: (*mat->ops->restorerowuppertriangular)(mat);
709: return(0);
710: }
712: /*@C
713: MatSetOptionsPrefix - Sets the prefix used for searching for all
714: Mat options in the database.
716: Logically Collective on Mat
718: Input Parameter:
719: + A - the Mat context
720: - prefix - the prefix to prepend to all option names
722: Notes:
723: A hyphen (-) must NOT be given at the beginning of the prefix name.
724: The first character of all runtime options is AUTOMATICALLY the hyphen.
726: Level: advanced
728: .seealso: MatSetFromOptions()
729: @*/
730: PetscErrorCode MatSetOptionsPrefix(Mat A,const char prefix[])
731: {
736: PetscObjectSetOptionsPrefix((PetscObject)A,prefix);
737: return(0);
738: }
740: /*@C
741: MatAppendOptionsPrefix - Appends to the prefix used for searching for all
742: Mat options in the database.
744: Logically Collective on Mat
746: Input Parameters:
747: + A - the Mat context
748: - prefix - the prefix to prepend to all option names
750: Notes:
751: A hyphen (-) must NOT be given at the beginning of the prefix name.
752: The first character of all runtime options is AUTOMATICALLY the hyphen.
754: Level: advanced
756: .seealso: MatGetOptionsPrefix()
757: @*/
758: PetscErrorCode MatAppendOptionsPrefix(Mat A,const char prefix[])
759: {
764: PetscObjectAppendOptionsPrefix((PetscObject)A,prefix);
765: return(0);
766: }
768: /*@C
769: MatGetOptionsPrefix - Sets the prefix used for searching for all
770: Mat options in the database.
772: Not Collective
774: Input Parameter:
775: . A - the Mat context
777: Output Parameter:
778: . prefix - pointer to the prefix string used
780: Notes:
781: On the fortran side, the user should pass in a string 'prefix' of
782: sufficient length to hold the prefix.
784: Level: advanced
786: .seealso: MatAppendOptionsPrefix()
787: @*/
788: PetscErrorCode MatGetOptionsPrefix(Mat A,const char *prefix[])
789: {
794: PetscObjectGetOptionsPrefix((PetscObject)A,prefix);
795: return(0);
796: }
798: /*@
799: MatResetPreallocation - Reset mat to use the original nonzero pattern provided by users.
801: Collective on Mat
803: Input Parameters:
804: . A - the Mat context
806: Notes:
807: The allocated memory will be shrunk after calling MatAssembly with MAT_FINAL_ASSEMBLY. Users can reset the preallocation to access the original memory.
808: Currently support MPIAIJ and SEQAIJ.
810: Level: beginner
812: .seealso: MatSeqAIJSetPreallocation(), MatMPIAIJSetPreallocation(), MatXAIJSetPreallocation()
813: @*/
814: PetscErrorCode MatResetPreallocation(Mat A)
815: {
821: PetscUseMethod(A,"MatResetPreallocation_C",(Mat),(A));
822: return(0);
823: }
826: /*@
827: MatSetUp - Sets up the internal matrix data structures for the later use.
829: Collective on Mat
831: Input Parameters:
832: . A - the Mat context
834: Notes:
835: If the user has not set preallocation for this matrix then a default preallocation that is likely to be inefficient is used.
837: If a suitable preallocation routine is used, this function does not need to be called.
839: See the Performance chapter of the PETSc users manual for how to preallocate matrices
841: Level: beginner
843: .seealso: MatCreate(), MatDestroy()
844: @*/
845: PetscErrorCode MatSetUp(Mat A)
846: {
847: PetscMPIInt size;
852: if (!((PetscObject)A)->type_name) {
853: MPI_Comm_size(PetscObjectComm((PetscObject)A), &size);
854: if (size == 1) {
855: MatSetType(A, MATSEQAIJ);
856: } else {
857: MatSetType(A, MATMPIAIJ);
858: }
859: }
860: if (!A->preallocated && A->ops->setup) {
861: PetscInfo(A,"Warning not preallocating matrix storage\n");
862: (*A->ops->setup)(A);
863: }
864: PetscLayoutSetUp(A->rmap);
865: PetscLayoutSetUp(A->cmap);
866: A->preallocated = PETSC_TRUE;
867: return(0);
868: }
870: #if defined(PETSC_HAVE_SAWS)
871: #include <petscviewersaws.h>
872: #endif
873: /*@C
874: MatView - Visualizes a matrix object.
876: Collective on Mat
878: Input Parameters:
879: + mat - the matrix
880: - viewer - visualization context
882: Notes:
883: The available visualization contexts include
884: + PETSC_VIEWER_STDOUT_SELF - for sequential matrices
885: . PETSC_VIEWER_STDOUT_WORLD - for parallel matrices created on PETSC_COMM_WORLD
886: . PETSC_VIEWER_STDOUT_(comm) - for matrices created on MPI communicator comm
887: - PETSC_VIEWER_DRAW_WORLD - graphical display of nonzero structure
889: The user can open alternative visualization contexts with
890: + PetscViewerASCIIOpen() - Outputs matrix to a specified file
891: . PetscViewerBinaryOpen() - Outputs matrix in binary to a
892: specified file; corresponding input uses MatLoad()
893: . PetscViewerDrawOpen() - Outputs nonzero matrix structure to
894: an X window display
895: - PetscViewerSocketOpen() - Outputs matrix to Socket viewer.
896: Currently only the sequential dense and AIJ
897: matrix types support the Socket viewer.
899: The user can call PetscViewerPushFormat() to specify the output
900: format of ASCII printed objects (when using PETSC_VIEWER_STDOUT_SELF,
901: PETSC_VIEWER_STDOUT_WORLD and PetscViewerASCIIOpen). Available formats include
902: + PETSC_VIEWER_DEFAULT - default, prints matrix contents
903: . PETSC_VIEWER_ASCII_MATLAB - prints matrix contents in Matlab format
904: . PETSC_VIEWER_ASCII_DENSE - prints entire matrix including zeros
905: . PETSC_VIEWER_ASCII_COMMON - prints matrix contents, using a sparse
906: format common among all matrix types
907: . PETSC_VIEWER_ASCII_IMPL - prints matrix contents, using an implementation-specific
908: format (which is in many cases the same as the default)
909: . PETSC_VIEWER_ASCII_INFO - prints basic information about the matrix
910: size and structure (not the matrix entries)
911: - PETSC_VIEWER_ASCII_INFO_DETAIL - prints more detailed information about
912: the matrix structure
914: Options Database Keys:
915: + -mat_view ::ascii_info - Prints info on matrix at conclusion of MatAssemblyEnd()
916: . -mat_view ::ascii_info_detail - Prints more detailed info
917: . -mat_view - Prints matrix in ASCII format
918: . -mat_view ::ascii_matlab - Prints matrix in Matlab format
919: . -mat_view draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
920: . -display <name> - Sets display name (default is host)
921: . -draw_pause <sec> - Sets number of seconds to pause after display
922: . -mat_view socket - Sends matrix to socket, can be accessed from Matlab (see Users-Manual: Chapter 12 Using MATLAB with PETSc for details)
923: . -viewer_socket_machine <machine> -
924: . -viewer_socket_port <port> -
925: . -mat_view binary - save matrix to file in binary format
926: - -viewer_binary_filename <name> -
927: Level: beginner
929: Notes:
930: The ASCII viewers are only recommended for small matrices on at most a moderate number of processes,
931: the program will seemingly hang and take hours for larger matrices, for larger matrices one should use the binary format.
933: See the manual page for MatLoad() for the exact format of the binary file when the binary
934: viewer is used.
936: See share/petsc/matlab/PetscBinaryRead.m for a Matlab code that can read in the binary file when the binary
937: viewer is used.
939: One can use '-mat_view draw -draw_pause -1' to pause the graphical display of matrix nonzero structure,
940: and then use the following mouse functions.
941: + left mouse: zoom in
942: . middle mouse: zoom out
943: - right mouse: continue with the simulation
945: .seealso: PetscViewerPushFormat(), PetscViewerASCIIOpen(), PetscViewerDrawOpen(),
946: PetscViewerSocketOpen(), PetscViewerBinaryOpen(), MatLoad()
947: @*/
948: PetscErrorCode MatView(Mat mat,PetscViewer viewer)
949: {
950: PetscErrorCode ierr;
951: PetscInt rows,cols,rbs,cbs;
952: PetscBool iascii,ibinary,isstring;
953: PetscViewerFormat format;
954: PetscMPIInt size;
955: #if defined(PETSC_HAVE_SAWS)
956: PetscBool issaws;
957: #endif
962: if (!viewer) {
963: PetscViewerASCIIGetStdout(PetscObjectComm((PetscObject)mat),&viewer);
964: }
967: MatCheckPreallocated(mat,1);
968: PetscViewerGetFormat(viewer,&format);
969: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
970: if (size == 1 && format == PETSC_VIEWER_LOAD_BALANCE) return(0);
971: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERBINARY,&ibinary);
972: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSTRING,&isstring);
973: if (ibinary) {
974: PetscBool mpiio;
975: PetscViewerBinaryGetUseMPIIO(viewer,&mpiio);
976: if (mpiio) SETERRQ(PetscObjectComm((PetscObject)viewer),PETSC_ERR_SUP,"PETSc matrix viewers do not support using MPI-IO, turn off that flag");
977: }
979: PetscLogEventBegin(MAT_View,mat,viewer,0,0);
980: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERASCII,&iascii);
981: if ((!iascii || (format != PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL)) && mat->factortype) {
982: SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"No viewers for factored matrix except ASCII info or info_detailed");
983: }
985: #if defined(PETSC_HAVE_SAWS)
986: PetscObjectTypeCompare((PetscObject)viewer,PETSCVIEWERSAWS,&issaws);
987: #endif
988: if (iascii) {
989: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ORDER,"Must call MatAssemblyBegin/End() before viewing matrix");
990: PetscObjectPrintClassNamePrefixType((PetscObject)mat,viewer);
991: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
992: MatNullSpace nullsp,transnullsp;
994: PetscViewerASCIIPushTab(viewer);
995: MatGetSize(mat,&rows,&cols);
996: MatGetBlockSizes(mat,&rbs,&cbs);
997: if (rbs != 1 || cbs != 1) {
998: if (rbs != cbs) {PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D, rbs=%D, cbs=%D\n",rows,cols,rbs,cbs);}
999: else {PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D, bs=%D\n",rows,cols,rbs);}
1000: } else {
1001: PetscViewerASCIIPrintf(viewer,"rows=%D, cols=%D\n",rows,cols);
1002: }
1003: if (mat->factortype) {
1004: MatSolverType solver;
1005: MatFactorGetSolverType(mat,&solver);
1006: PetscViewerASCIIPrintf(viewer,"package used to perform factorization: %s\n",solver);
1007: }
1008: if (mat->ops->getinfo) {
1009: MatInfo info;
1010: MatGetInfo(mat,MAT_GLOBAL_SUM,&info);
1011: PetscViewerASCIIPrintf(viewer,"total: nonzeros=%.f, allocated nonzeros=%.f\n",info.nz_used,info.nz_allocated);
1012: PetscViewerASCIIPrintf(viewer,"total number of mallocs used during MatSetValues calls=%D\n",(PetscInt)info.mallocs);
1013: }
1014: MatGetNullSpace(mat,&nullsp);
1015: MatGetTransposeNullSpace(mat,&transnullsp);
1016: if (nullsp) {PetscViewerASCIIPrintf(viewer," has attached null space\n");}
1017: if (transnullsp && transnullsp != nullsp) {PetscViewerASCIIPrintf(viewer," has attached transposed null space\n");}
1018: MatGetNearNullSpace(mat,&nullsp);
1019: if (nullsp) {PetscViewerASCIIPrintf(viewer," has attached near null space\n");}
1020: }
1021: #if defined(PETSC_HAVE_SAWS)
1022: } else if (issaws) {
1023: PetscMPIInt rank;
1025: PetscObjectName((PetscObject)mat);
1026: MPI_Comm_rank(PETSC_COMM_WORLD,&rank);
1027: if (!((PetscObject)mat)->amsmem && !rank) {
1028: PetscObjectViewSAWs((PetscObject)mat,viewer);
1029: }
1030: #endif
1031: } else if (isstring) {
1032: const char *type;
1033: MatGetType(mat,&type);
1034: PetscViewerStringSPrintf(viewer," MatType: %-7.7s",type);
1035: if (mat->ops->view) {(*mat->ops->view)(mat,viewer);}
1036: }
1037: if ((format == PETSC_VIEWER_NATIVE || format == PETSC_VIEWER_LOAD_BALANCE) && mat->ops->viewnative) {
1038: PetscViewerASCIIPushTab(viewer);
1039: (*mat->ops->viewnative)(mat,viewer);
1040: PetscViewerASCIIPopTab(viewer);
1041: } else if (mat->ops->view) {
1042: PetscViewerASCIIPushTab(viewer);
1043: (*mat->ops->view)(mat,viewer);
1044: PetscViewerASCIIPopTab(viewer);
1045: }
1046: if (iascii) {
1047: PetscViewerGetFormat(viewer,&format);
1048: if (format == PETSC_VIEWER_ASCII_INFO || format == PETSC_VIEWER_ASCII_INFO_DETAIL) {
1049: PetscViewerASCIIPopTab(viewer);
1050: }
1051: }
1052: PetscLogEventEnd(MAT_View,mat,viewer,0,0);
1053: return(0);
1054: }
1056: #if defined(PETSC_USE_DEBUG)
1057: #include <../src/sys/totalview/tv_data_display.h>
1058: PETSC_UNUSED static int TV_display_type(const struct _p_Mat *mat)
1059: {
1060: TV_add_row("Local rows", "int", &mat->rmap->n);
1061: TV_add_row("Local columns", "int", &mat->cmap->n);
1062: TV_add_row("Global rows", "int", &mat->rmap->N);
1063: TV_add_row("Global columns", "int", &mat->cmap->N);
1064: TV_add_row("Typename", TV_ascii_string_type, ((PetscObject)mat)->type_name);
1065: return TV_format_OK;
1066: }
1067: #endif
1069: /*@C
1070: MatLoad - Loads a matrix that has been stored in binary/HDF5 format
1071: with MatView(). The matrix format is determined from the options database.
1072: Generates a parallel MPI matrix if the communicator has more than one
1073: processor. The default matrix type is AIJ.
1075: Collective on PetscViewer
1077: Input Parameters:
1078: + newmat - the newly loaded matrix, this needs to have been created with MatCreate()
1079: or some related function before a call to MatLoad()
1080: - viewer - binary/HDF5 file viewer
1082: Options Database Keys:
1083: Used with block matrix formats (MATSEQBAIJ, ...) to specify
1084: block size
1085: . -matload_block_size <bs>
1087: Level: beginner
1089: Notes:
1090: If the Mat type has not yet been given then MATAIJ is used, call MatSetFromOptions() on the
1091: Mat before calling this routine if you wish to set it from the options database.
1093: MatLoad() automatically loads into the options database any options
1094: given in the file filename.info where filename is the name of the file
1095: that was passed to the PetscViewerBinaryOpen(). The options in the info
1096: file will be ignored if you use the -viewer_binary_skip_info option.
1098: If the type or size of newmat is not set before a call to MatLoad, PETSc
1099: sets the default matrix type AIJ and sets the local and global sizes.
1100: If type and/or size is already set, then the same are used.
1102: In parallel, each processor can load a subset of rows (or the
1103: entire matrix). This routine is especially useful when a large
1104: matrix is stored on disk and only part of it is desired on each
1105: processor. For example, a parallel solver may access only some of
1106: the rows from each processor. The algorithm used here reads
1107: relatively small blocks of data rather than reading the entire
1108: matrix and then subsetting it.
1110: Viewer's PetscViewerType must be either PETSCVIEWERBINARY or PETSCVIEWERHDF5.
1111: Such viewer can be created using PetscViewerBinaryOpen()/PetscViewerHDF5Open(),
1112: or the sequence like
1113: $ PetscViewer v;
1114: $ PetscViewerCreate(PETSC_COMM_WORLD,&v);
1115: $ PetscViewerSetType(v,PETSCVIEWERBINARY);
1116: $ PetscViewerSetFromOptions(v);
1117: $ PetscViewerFileSetMode(v,FILE_MODE_READ);
1118: $ PetscViewerFileSetName(v,"datafile");
1119: The optional PetscViewerSetFromOptions() call allows to override PetscViewerSetType() using option
1120: $ -viewer_type {binary,hdf5}
1122: See the example src/ksp/ksp/examples/tutorials/ex27.c with the first approach,
1123: and src/mat/examples/tutorials/ex10.c with the second approach.
1125: Notes about the PETSc binary format:
1126: In case of PETSCVIEWERBINARY, a native PETSc binary format is used. Each of the blocks
1127: is read onto rank 0 and then shipped to its destination rank, one after another.
1128: Multiple objects, both matrices and vectors, can be stored within the same file.
1129: Their PetscObject name is ignored; they are loaded in the order of their storage.
1131: Most users should not need to know the details of the binary storage
1132: format, since MatLoad() and MatView() completely hide these details.
1133: But for anyone who's interested, the standard binary matrix storage
1134: format is
1136: $ PetscInt MAT_FILE_CLASSID
1137: $ PetscInt number of rows
1138: $ PetscInt number of columns
1139: $ PetscInt total number of nonzeros
1140: $ PetscInt *number nonzeros in each row
1141: $ PetscInt *column indices of all nonzeros (starting index is zero)
1142: $ PetscScalar *values of all nonzeros
1144: PETSc automatically does the byte swapping for
1145: machines that store the bytes reversed, e.g. DEC alpha, freebsd,
1146: linux, Windows and the paragon; thus if you write your own binary
1147: read/write routines you have to swap the bytes; see PetscBinaryRead()
1148: and PetscBinaryWrite() to see how this may be done.
1150: Notes about the HDF5 (MATLAB MAT-File Version 7.3) format:
1151: In case of PETSCVIEWERHDF5, a parallel HDF5 reader is used.
1152: Each processor's chunk is loaded independently by its owning rank.
1153: Multiple objects, both matrices and vectors, can be stored within the same file.
1154: They are looked up by their PetscObject name.
1156: As the MATLAB MAT-File Version 7.3 format is also a HDF5 flavor, we decided to use
1157: by default the same structure and naming of the AIJ arrays and column count
1158: within the HDF5 file. This means that a MAT file saved with -v7.3 flag, e.g.
1159: $ save example.mat A b -v7.3
1160: can be directly read by this routine (see Reference 1 for details).
1161: Note that depending on your MATLAB version, this format might be a default,
1162: otherwise you can set it as default in Preferences.
1164: Unless -nocompression flag is used to save the file in MATLAB,
1165: PETSc must be configured with ZLIB package.
1167: See also examples src/mat/examples/tutorials/ex10.c and src/ksp/ksp/examples/tutorials/ex27.c
1169: Current HDF5 (MAT-File) limitations:
1170: This reader currently supports only real MATSEQAIJ, MATMPIAIJ, MATSEQDENSE and MATMPIDENSE matrices.
1172: Corresponding MatView() is not yet implemented.
1174: The loaded matrix is actually a transpose of the original one in MATLAB,
1175: unless you push PETSC_VIEWER_HDF5_MAT format (see examples above).
1176: With this format, matrix is automatically transposed by PETSc,
1177: unless the matrix is marked as SPD or symmetric
1178: (see MatSetOption(), MAT_SPD, MAT_SYMMETRIC).
1180: References:
1181: 1. MATLAB(R) Documentation, manual page of save(), https://www.mathworks.com/help/matlab/ref/save.html#btox10b-1-version
1183: .seealso: PetscViewerBinaryOpen(), PetscViewerSetType(), MatView(), VecLoad()
1185: @*/
1186: PetscErrorCode MatLoad(Mat newmat,PetscViewer viewer)
1187: {
1189: PetscBool flg;
1195: if (!((PetscObject)newmat)->type_name) {
1196: MatSetType(newmat,MATAIJ);
1197: }
1199: flg = PETSC_FALSE;
1200: PetscOptionsGetBool(((PetscObject)newmat)->options,((PetscObject)newmat)->prefix,"-matload_symmetric",&flg,NULL);
1201: if (flg) {
1202: MatSetOption(newmat,MAT_SYMMETRIC,PETSC_TRUE);
1203: MatSetOption(newmat,MAT_SYMMETRY_ETERNAL,PETSC_TRUE);
1204: }
1205: flg = PETSC_FALSE;
1206: PetscOptionsGetBool(((PetscObject)newmat)->options,((PetscObject)newmat)->prefix,"-matload_spd",&flg,NULL);
1207: if (flg) {
1208: MatSetOption(newmat,MAT_SPD,PETSC_TRUE);
1209: }
1211: if (!newmat->ops->load) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatLoad is not supported for type %s",((PetscObject)newmat)->type_name);
1212: PetscLogEventBegin(MAT_Load,viewer,0,0,0);
1213: (*newmat->ops->load)(newmat,viewer);
1214: PetscLogEventEnd(MAT_Load,viewer,0,0,0);
1215: return(0);
1216: }
1218: PetscErrorCode MatDestroy_Redundant(Mat_Redundant **redundant)
1219: {
1221: Mat_Redundant *redund = *redundant;
1222: PetscInt i;
1225: if (redund){
1226: if (redund->matseq) { /* via MatCreateSubMatrices() */
1227: ISDestroy(&redund->isrow);
1228: ISDestroy(&redund->iscol);
1229: MatDestroySubMatrices(1,&redund->matseq);
1230: } else {
1231: PetscFree2(redund->send_rank,redund->recv_rank);
1232: PetscFree(redund->sbuf_j);
1233: PetscFree(redund->sbuf_a);
1234: for (i=0; i<redund->nrecvs; i++) {
1235: PetscFree(redund->rbuf_j[i]);
1236: PetscFree(redund->rbuf_a[i]);
1237: }
1238: PetscFree4(redund->sbuf_nz,redund->rbuf_nz,redund->rbuf_j,redund->rbuf_a);
1239: }
1241: if (redund->subcomm) {
1242: PetscCommDestroy(&redund->subcomm);
1243: }
1244: PetscFree(redund);
1245: }
1246: return(0);
1247: }
1249: /*@
1250: MatDestroy - Frees space taken by a matrix.
1252: Collective on Mat
1254: Input Parameter:
1255: . A - the matrix
1257: Level: beginner
1259: @*/
1260: PetscErrorCode MatDestroy(Mat *A)
1261: {
1265: if (!*A) return(0);
1267: if (--((PetscObject)(*A))->refct > 0) {*A = NULL; return(0);}
1269: /* if memory was published with SAWs then destroy it */
1270: PetscObjectSAWsViewOff((PetscObject)*A);
1271: if ((*A)->ops->destroy) {
1272: (*(*A)->ops->destroy)(*A);
1273: }
1275: PetscFree((*A)->defaultvectype);
1276: PetscFree((*A)->bsizes);
1277: PetscFree((*A)->solvertype);
1278: MatDestroy_Redundant(&(*A)->redundant);
1279: MatNullSpaceDestroy(&(*A)->nullsp);
1280: MatNullSpaceDestroy(&(*A)->transnullsp);
1281: MatNullSpaceDestroy(&(*A)->nearnullsp);
1282: MatDestroy(&(*A)->schur);
1283: PetscLayoutDestroy(&(*A)->rmap);
1284: PetscLayoutDestroy(&(*A)->cmap);
1285: PetscHeaderDestroy(A);
1286: return(0);
1287: }
1289: /*@C
1290: MatSetValues - Inserts or adds a block of values into a matrix.
1291: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
1292: MUST be called after all calls to MatSetValues() have been completed.
1294: Not Collective
1296: Input Parameters:
1297: + mat - the matrix
1298: . v - a logically two-dimensional array of values
1299: . m, idxm - the number of rows and their global indices
1300: . n, idxn - the number of columns and their global indices
1301: - addv - either ADD_VALUES or INSERT_VALUES, where
1302: ADD_VALUES adds values to any existing entries, and
1303: INSERT_VALUES replaces existing entries with new values
1305: Notes:
1306: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
1307: MatSetUp() before using this routine
1309: By default the values, v, are row-oriented. See MatSetOption() for other options.
1311: Calls to MatSetValues() with the INSERT_VALUES and ADD_VALUES
1312: options cannot be mixed without intervening calls to the assembly
1313: routines.
1315: MatSetValues() uses 0-based row and column numbers in Fortran
1316: as well as in C.
1318: Negative indices may be passed in idxm and idxn, these rows and columns are
1319: simply ignored. This allows easily inserting element stiffness matrices
1320: with homogeneous Dirchlet boundary conditions that you don't want represented
1321: in the matrix.
1323: Efficiency Alert:
1324: The routine MatSetValuesBlocked() may offer much better efficiency
1325: for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1327: Level: beginner
1329: Developer Notes:
1330: This is labeled with C so does not automatically generate Fortran stubs and interfaces
1331: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
1333: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1334: InsertMode, INSERT_VALUES, ADD_VALUES
1335: @*/
1336: PetscErrorCode MatSetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1337: {
1339: #if defined(PETSC_USE_DEBUG)
1340: PetscInt i,j;
1341: #endif
1346: if (!m || !n) return(0); /* no values to insert */
1349: MatCheckPreallocated(mat,1);
1351: if (mat->insertmode == NOT_SET_VALUES) {
1352: mat->insertmode = addv;
1353: }
1354: #if defined(PETSC_USE_DEBUG)
1355: else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1356: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1357: if (!mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1359: for (i=0; i<m; i++) {
1360: for (j=0; j<n; j++) {
1361: if (mat->erroriffailure && PetscIsInfOrNanScalar(v[i*n+j]))
1362: #if defined(PETSC_USE_COMPLEX)
1363: 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]);
1364: #else
1365: SETERRQ3(PETSC_COMM_SELF,PETSC_ERR_FP,"Inserting %g at matrix entry (%D,%D)",(double)v[i*n+j],idxm[i],idxn[j]);
1366: #endif
1367: }
1368: }
1369: #endif
1371: if (mat->assembled) {
1372: mat->was_assembled = PETSC_TRUE;
1373: mat->assembled = PETSC_FALSE;
1374: }
1375: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1376: (*mat->ops->setvalues)(mat,m,idxm,n,idxn,v,addv);
1377: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1378: return(0);
1379: }
1382: /*@
1383: MatSetValuesRowLocal - Inserts a row (block row for BAIJ matrices) of nonzero
1384: values into a matrix
1386: Not Collective
1388: Input Parameters:
1389: + mat - the matrix
1390: . row - the (block) row to set
1391: - v - a logically two-dimensional array of values
1393: Notes:
1394: By the values, v, are column-oriented (for the block version) and sorted
1396: All the nonzeros in the row must be provided
1398: The matrix must have previously had its column indices set
1400: The row must belong to this process
1402: Level: intermediate
1404: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1405: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues(), MatSetValuesRow(), MatSetLocalToGlobalMapping()
1406: @*/
1407: PetscErrorCode MatSetValuesRowLocal(Mat mat,PetscInt row,const PetscScalar v[])
1408: {
1410: PetscInt globalrow;
1416: ISLocalToGlobalMappingApply(mat->rmap->mapping,1,&row,&globalrow);
1417: MatSetValuesRow(mat,globalrow,v);
1418: return(0);
1419: }
1421: /*@
1422: MatSetValuesRow - Inserts a row (block row for BAIJ matrices) of nonzero
1423: values into a matrix
1425: Not Collective
1427: Input Parameters:
1428: + mat - the matrix
1429: . row - the (block) row to set
1430: - 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
1432: Notes:
1433: The values, v, are column-oriented for the block version.
1435: All the nonzeros in the row must be provided
1437: THE MATRIX MUST HAVE PREVIOUSLY HAD ITS COLUMN INDICES SET. IT IS RARE THAT THIS ROUTINE IS USED, usually MatSetValues() is used.
1439: The row must belong to this process
1441: Level: advanced
1443: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1444: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues()
1445: @*/
1446: PetscErrorCode MatSetValuesRow(Mat mat,PetscInt row,const PetscScalar v[])
1447: {
1453: MatCheckPreallocated(mat,1);
1455: #if defined(PETSC_USE_DEBUG)
1456: if (mat->insertmode == ADD_VALUES) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add and insert values");
1457: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1458: #endif
1459: mat->insertmode = INSERT_VALUES;
1461: if (mat->assembled) {
1462: mat->was_assembled = PETSC_TRUE;
1463: mat->assembled = PETSC_FALSE;
1464: }
1465: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1466: if (!mat->ops->setvaluesrow) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1467: (*mat->ops->setvaluesrow)(mat,row,v);
1468: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1469: return(0);
1470: }
1472: /*@
1473: MatSetValuesStencil - Inserts or adds a block of values into a matrix.
1474: Using structured grid indexing
1476: Not Collective
1478: Input Parameters:
1479: + mat - the matrix
1480: . m - number of rows being entered
1481: . idxm - grid coordinates (and component number when dof > 1) for matrix rows being entered
1482: . n - number of columns being entered
1483: . idxn - grid coordinates (and component number when dof > 1) for matrix columns being entered
1484: . v - a logically two-dimensional array of values
1485: - addv - either ADD_VALUES or INSERT_VALUES, where
1486: ADD_VALUES adds values to any existing entries, and
1487: INSERT_VALUES replaces existing entries with new values
1489: Notes:
1490: By default the values, v, are row-oriented. See MatSetOption() for other options.
1492: Calls to MatSetValuesStencil() with the INSERT_VALUES and ADD_VALUES
1493: options cannot be mixed without intervening calls to the assembly
1494: routines.
1496: The grid coordinates are across the entire grid, not just the local portion
1498: MatSetValuesStencil() uses 0-based row and column numbers in Fortran
1499: as well as in C.
1501: For setting/accessing vector values via array coordinates you can use the DMDAVecGetArray() routine
1503: In order to use this routine you must either obtain the matrix with DMCreateMatrix()
1504: or call MatSetLocalToGlobalMapping() and MatSetStencil() first.
1506: The columns and rows in the stencil passed in MUST be contained within the
1507: ghost region of the given process as set with DMDACreateXXX() or MatSetStencil(). For example,
1508: if you create a DMDA with an overlap of one grid level and on a particular process its first
1509: local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1510: first i index you can use in your column and row indices in MatSetStencil() is 5.
1512: In Fortran idxm and idxn should be declared as
1513: $ MatStencil idxm(4,m),idxn(4,n)
1514: and the values inserted using
1515: $ idxm(MatStencil_i,1) = i
1516: $ idxm(MatStencil_j,1) = j
1517: $ idxm(MatStencil_k,1) = k
1518: $ idxm(MatStencil_c,1) = c
1519: etc
1521: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
1522: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
1523: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
1524: DM_BOUNDARY_PERIODIC boundary type.
1526: 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
1527: a single value per point) you can skip filling those indices.
1529: Inspired by the structured grid interface to the HYPRE package
1530: (https://computation.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods)
1532: Efficiency Alert:
1533: The routine MatSetValuesBlockedStencil() may offer much better efficiency
1534: for users of block sparse formats (MATSEQBAIJ and MATMPIBAIJ).
1536: Level: beginner
1538: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1539: MatSetValues(), MatSetValuesBlockedStencil(), MatSetStencil(), DMCreateMatrix(), DMDAVecGetArray(), MatStencil
1540: @*/
1541: PetscErrorCode MatSetValuesStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1542: {
1544: PetscInt buf[8192],*bufm=0,*bufn=0,*jdxm,*jdxn;
1545: PetscInt j,i,dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1546: PetscInt *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1549: if (!m || !n) return(0); /* no values to insert */
1555: if ((m+n) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1556: jdxm = buf; jdxn = buf+m;
1557: } else {
1558: PetscMalloc2(m,&bufm,n,&bufn);
1559: jdxm = bufm; jdxn = bufn;
1560: }
1561: for (i=0; i<m; i++) {
1562: for (j=0; j<3-sdim; j++) dxm++;
1563: tmp = *dxm++ - starts[0];
1564: for (j=0; j<dim-1; j++) {
1565: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1566: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1567: }
1568: if (mat->stencil.noc) dxm++;
1569: jdxm[i] = tmp;
1570: }
1571: for (i=0; i<n; i++) {
1572: for (j=0; j<3-sdim; j++) dxn++;
1573: tmp = *dxn++ - starts[0];
1574: for (j=0; j<dim-1; j++) {
1575: if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1576: else tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1577: }
1578: if (mat->stencil.noc) dxn++;
1579: jdxn[i] = tmp;
1580: }
1581: MatSetValuesLocal(mat,m,jdxm,n,jdxn,v,addv);
1582: PetscFree2(bufm,bufn);
1583: return(0);
1584: }
1586: /*@
1587: MatSetValuesBlockedStencil - Inserts or adds a block of values into a matrix.
1588: Using structured grid indexing
1590: Not Collective
1592: Input Parameters:
1593: + mat - the matrix
1594: . m - number of rows being entered
1595: . idxm - grid coordinates for matrix rows being entered
1596: . n - number of columns being entered
1597: . idxn - grid coordinates for matrix columns being entered
1598: . v - a logically two-dimensional array of values
1599: - addv - either ADD_VALUES or INSERT_VALUES, where
1600: ADD_VALUES adds values to any existing entries, and
1601: INSERT_VALUES replaces existing entries with new values
1603: Notes:
1604: By default the values, v, are row-oriented and unsorted.
1605: See MatSetOption() for other options.
1607: Calls to MatSetValuesBlockedStencil() with the INSERT_VALUES and ADD_VALUES
1608: options cannot be mixed without intervening calls to the assembly
1609: routines.
1611: The grid coordinates are across the entire grid, not just the local portion
1613: MatSetValuesBlockedStencil() uses 0-based row and column numbers in Fortran
1614: as well as in C.
1616: For setting/accessing vector values via array coordinates you can use the DMDAVecGetArray() routine
1618: In order to use this routine you must either obtain the matrix with DMCreateMatrix()
1619: or call MatSetBlockSize(), MatSetLocalToGlobalMapping() and MatSetStencil() first.
1621: The columns and rows in the stencil passed in MUST be contained within the
1622: ghost region of the given process as set with DMDACreateXXX() or MatSetStencil(). For example,
1623: if you create a DMDA with an overlap of one grid level and on a particular process its first
1624: local nonghost x logical coordinate is 6 (so its first ghost x logical coordinate is 5) the
1625: first i index you can use in your column and row indices in MatSetStencil() is 5.
1627: In Fortran idxm and idxn should be declared as
1628: $ MatStencil idxm(4,m),idxn(4,n)
1629: and the values inserted using
1630: $ idxm(MatStencil_i,1) = i
1631: $ idxm(MatStencil_j,1) = j
1632: $ idxm(MatStencil_k,1) = k
1633: etc
1635: Negative indices may be passed in idxm and idxn, these rows and columns are
1636: simply ignored. This allows easily inserting element stiffness matrices
1637: with homogeneous Dirchlet boundary conditions that you don't want represented
1638: in the matrix.
1640: Inspired by the structured grid interface to the HYPRE package
1641: (https://computation.llnl.gov/projects/hypre-scalable-linear-solvers-multigrid-methods)
1643: Level: beginner
1645: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1646: MatSetValues(), MatSetValuesStencil(), MatSetStencil(), DMCreateMatrix(), DMDAVecGetArray(), MatStencil,
1647: MatSetBlockSize(), MatSetLocalToGlobalMapping()
1648: @*/
1649: PetscErrorCode MatSetValuesBlockedStencil(Mat mat,PetscInt m,const MatStencil idxm[],PetscInt n,const MatStencil idxn[],const PetscScalar v[],InsertMode addv)
1650: {
1652: PetscInt buf[8192],*bufm=0,*bufn=0,*jdxm,*jdxn;
1653: PetscInt j,i,dim = mat->stencil.dim,*dims = mat->stencil.dims+1,tmp;
1654: PetscInt *starts = mat->stencil.starts,*dxm = (PetscInt*)idxm,*dxn = (PetscInt*)idxn,sdim = dim - (1 - (PetscInt)mat->stencil.noc);
1657: if (!m || !n) return(0); /* no values to insert */
1664: if ((m+n) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1665: jdxm = buf; jdxn = buf+m;
1666: } else {
1667: PetscMalloc2(m,&bufm,n,&bufn);
1668: jdxm = bufm; jdxn = bufn;
1669: }
1670: for (i=0; i<m; i++) {
1671: for (j=0; j<3-sdim; j++) dxm++;
1672: tmp = *dxm++ - starts[0];
1673: for (j=0; j<sdim-1; j++) {
1674: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1675: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
1676: }
1677: dxm++;
1678: jdxm[i] = tmp;
1679: }
1680: for (i=0; i<n; i++) {
1681: for (j=0; j<3-sdim; j++) dxn++;
1682: tmp = *dxn++ - starts[0];
1683: for (j=0; j<sdim-1; j++) {
1684: if ((*dxn++ - starts[j+1]) < 0 || tmp < 0) tmp = -1;
1685: else tmp = tmp*dims[j] + *(dxn-1) - starts[j+1];
1686: }
1687: dxn++;
1688: jdxn[i] = tmp;
1689: }
1690: MatSetValuesBlockedLocal(mat,m,jdxm,n,jdxn,v,addv);
1691: PetscFree2(bufm,bufn);
1692: return(0);
1693: }
1695: /*@
1696: MatSetStencil - Sets the grid information for setting values into a matrix via
1697: MatSetValuesStencil()
1699: Not Collective
1701: Input Parameters:
1702: + mat - the matrix
1703: . dim - dimension of the grid 1, 2, or 3
1704: . dims - number of grid points in x, y, and z direction, including ghost points on your processor
1705: . starts - starting point of ghost nodes on your processor in x, y, and z direction
1706: - dof - number of degrees of freedom per node
1709: Inspired by the structured grid interface to the HYPRE package
1710: (www.llnl.gov/CASC/hyper)
1712: For matrices generated with DMCreateMatrix() this routine is automatically called and so not needed by the
1713: user.
1715: Level: beginner
1717: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal()
1718: MatSetValues(), MatSetValuesBlockedStencil(), MatSetValuesStencil()
1719: @*/
1720: PetscErrorCode MatSetStencil(Mat mat,PetscInt dim,const PetscInt dims[],const PetscInt starts[],PetscInt dof)
1721: {
1722: PetscInt i;
1729: mat->stencil.dim = dim + (dof > 1);
1730: for (i=0; i<dim; i++) {
1731: mat->stencil.dims[i] = dims[dim-i-1]; /* copy the values in backwards */
1732: mat->stencil.starts[i] = starts[dim-i-1];
1733: }
1734: mat->stencil.dims[dim] = dof;
1735: mat->stencil.starts[dim] = 0;
1736: mat->stencil.noc = (PetscBool)(dof == 1);
1737: return(0);
1738: }
1740: /*@C
1741: MatSetValuesBlocked - Inserts or adds a block of values into a matrix.
1743: Not Collective
1745: Input Parameters:
1746: + mat - the matrix
1747: . v - a logically two-dimensional array of values
1748: . m, idxm - the number of block rows and their global block indices
1749: . n, idxn - the number of block columns and their global block indices
1750: - addv - either ADD_VALUES or INSERT_VALUES, where
1751: ADD_VALUES adds values to any existing entries, and
1752: INSERT_VALUES replaces existing entries with new values
1754: Notes:
1755: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call
1756: MatXXXXSetPreallocation() or MatSetUp() before using this routine.
1758: The m and n count the NUMBER of blocks in the row direction and column direction,
1759: NOT the total number of rows/columns; for example, if the block size is 2 and
1760: you are passing in values for rows 2,3,4,5 then m would be 2 (not 4).
1761: The values in idxm would be 1 2; that is the first index for each block divided by
1762: the block size.
1764: Note that you must call MatSetBlockSize() when constructing this matrix (before
1765: preallocating it).
1767: By default the values, v, are row-oriented, so the layout of
1768: v is the same as for MatSetValues(). See MatSetOption() for other options.
1770: Calls to MatSetValuesBlocked() with the INSERT_VALUES and ADD_VALUES
1771: options cannot be mixed without intervening calls to the assembly
1772: routines.
1774: MatSetValuesBlocked() uses 0-based row and column numbers in Fortran
1775: as well as in C.
1777: Negative indices may be passed in idxm and idxn, these rows and columns are
1778: simply ignored. This allows easily inserting element stiffness matrices
1779: with homogeneous Dirchlet boundary conditions that you don't want represented
1780: in the matrix.
1782: Each time an entry is set within a sparse matrix via MatSetValues(),
1783: internal searching must be done to determine where to place the
1784: data in the matrix storage space. By instead inserting blocks of
1785: entries via MatSetValuesBlocked(), the overhead of matrix assembly is
1786: reduced.
1788: Example:
1789: $ Suppose m=n=2 and block size(bs) = 2 The array is
1790: $
1791: $ 1 2 | 3 4
1792: $ 5 6 | 7 8
1793: $ - - - | - - -
1794: $ 9 10 | 11 12
1795: $ 13 14 | 15 16
1796: $
1797: $ v[] should be passed in like
1798: $ v[] = [1,2,3,4,5,6,7,8,9,10,11,12,13,14,15,16]
1799: $
1800: $ If you are not using row oriented storage of v (that is you called MatSetOption(mat,MAT_ROW_ORIENTED,PETSC_FALSE)) then
1801: $ v[] = [1,5,9,13,2,6,10,14,3,7,11,15,4,8,12,16]
1803: Level: intermediate
1805: .seealso: MatSetBlockSize(), MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesBlockedLocal()
1806: @*/
1807: PetscErrorCode MatSetValuesBlocked(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],const PetscScalar v[],InsertMode addv)
1808: {
1814: if (!m || !n) return(0); /* no values to insert */
1818: MatCheckPreallocated(mat,1);
1819: if (mat->insertmode == NOT_SET_VALUES) {
1820: mat->insertmode = addv;
1821: }
1822: #if defined(PETSC_USE_DEBUG)
1823: else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
1824: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1825: if (!mat->ops->setvaluesblocked && !mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1826: #endif
1828: if (mat->assembled) {
1829: mat->was_assembled = PETSC_TRUE;
1830: mat->assembled = PETSC_FALSE;
1831: }
1832: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
1833: if (mat->ops->setvaluesblocked) {
1834: (*mat->ops->setvaluesblocked)(mat,m,idxm,n,idxn,v,addv);
1835: } else {
1836: PetscInt buf[8192],*bufr=0,*bufc=0,*iidxm,*iidxn;
1837: PetscInt i,j,bs,cbs;
1838: MatGetBlockSizes(mat,&bs,&cbs);
1839: if (m*bs+n*cbs <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
1840: iidxm = buf; iidxn = buf + m*bs;
1841: } else {
1842: PetscMalloc2(m*bs,&bufr,n*cbs,&bufc);
1843: iidxm = bufr; iidxn = bufc;
1844: }
1845: for (i=0; i<m; i++) {
1846: for (j=0; j<bs; j++) {
1847: iidxm[i*bs+j] = bs*idxm[i] + j;
1848: }
1849: }
1850: for (i=0; i<n; i++) {
1851: for (j=0; j<cbs; j++) {
1852: iidxn[i*cbs+j] = cbs*idxn[i] + j;
1853: }
1854: }
1855: MatSetValues(mat,m*bs,iidxm,n*cbs,iidxn,v,addv);
1856: PetscFree2(bufr,bufc);
1857: }
1858: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
1859: return(0);
1860: }
1862: /*@
1863: MatGetValues - Gets a block of values from a matrix.
1865: Not Collective; currently only returns a local block
1867: Input Parameters:
1868: + mat - the matrix
1869: . v - a logically two-dimensional array for storing the values
1870: . m, idxm - the number of rows and their global indices
1871: - n, idxn - the number of columns and their global indices
1873: Notes:
1874: The user must allocate space (m*n PetscScalars) for the values, v.
1875: The values, v, are then returned in a row-oriented format,
1876: analogous to that used by default in MatSetValues().
1878: MatGetValues() uses 0-based row and column numbers in
1879: Fortran as well as in C.
1881: MatGetValues() requires that the matrix has been assembled
1882: with MatAssemblyBegin()/MatAssemblyEnd(). Thus, calls to
1883: MatSetValues() and MatGetValues() CANNOT be made in succession
1884: without intermediate matrix assembly.
1886: Negative row or column indices will be ignored and those locations in v[] will be
1887: left unchanged.
1889: Level: advanced
1891: .seealso: MatGetRow(), MatCreateSubMatrices(), MatSetValues()
1892: @*/
1893: PetscErrorCode MatGetValues(Mat mat,PetscInt m,const PetscInt idxm[],PetscInt n,const PetscInt idxn[],PetscScalar v[])
1894: {
1900: if (!m || !n) return(0);
1904: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
1905: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1906: if (!mat->ops->getvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
1907: MatCheckPreallocated(mat,1);
1909: PetscLogEventBegin(MAT_GetValues,mat,0,0,0);
1910: (*mat->ops->getvalues)(mat,m,idxm,n,idxn,v);
1911: PetscLogEventEnd(MAT_GetValues,mat,0,0,0);
1912: return(0);
1913: }
1915: /*@
1916: MatSetValuesBatch - Adds (ADD_VALUES) many blocks of values into a matrix at once. The blocks must all be square and
1917: the same size. Currently, this can only be called once and creates the given matrix.
1919: Not Collective
1921: Input Parameters:
1922: + mat - the matrix
1923: . nb - the number of blocks
1924: . bs - the number of rows (and columns) in each block
1925: . rows - a concatenation of the rows for each block
1926: - v - a concatenation of logically two-dimensional arrays of values
1928: Notes:
1929: In the future, we will extend this routine to handle rectangular blocks, and to allow multiple calls for a given matrix.
1931: Level: advanced
1933: .seealso: MatSetOption(), MatAssemblyBegin(), MatAssemblyEnd(), MatSetValuesBlocked(), MatSetValuesLocal(),
1934: InsertMode, INSERT_VALUES, ADD_VALUES, MatSetValues()
1935: @*/
1936: PetscErrorCode MatSetValuesBatch(Mat mat, PetscInt nb, PetscInt bs, PetscInt rows[], const PetscScalar v[])
1937: {
1945: #if defined(PETSC_USE_DEBUG)
1946: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
1947: #endif
1949: PetscLogEventBegin(MAT_SetValuesBatch,mat,0,0,0);
1950: if (mat->ops->setvaluesbatch) {
1951: (*mat->ops->setvaluesbatch)(mat,nb,bs,rows,v);
1952: } else {
1953: PetscInt b;
1954: for (b = 0; b < nb; ++b) {
1955: MatSetValues(mat, bs, &rows[b*bs], bs, &rows[b*bs], &v[b*bs*bs], ADD_VALUES);
1956: }
1957: }
1958: PetscLogEventEnd(MAT_SetValuesBatch,mat,0,0,0);
1959: return(0);
1960: }
1962: /*@
1963: MatSetLocalToGlobalMapping - Sets a local-to-global numbering for use by
1964: the routine MatSetValuesLocal() to allow users to insert matrix entries
1965: using a local (per-processor) numbering.
1967: Not Collective
1969: Input Parameters:
1970: + x - the matrix
1971: . rmapping - row mapping created with ISLocalToGlobalMappingCreate() or ISLocalToGlobalMappingCreateIS()
1972: - cmapping - column mapping
1974: Level: intermediate
1977: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetValuesLocal()
1978: @*/
1979: PetscErrorCode MatSetLocalToGlobalMapping(Mat x,ISLocalToGlobalMapping rmapping,ISLocalToGlobalMapping cmapping)
1980: {
1989: if (x->ops->setlocaltoglobalmapping) {
1990: (*x->ops->setlocaltoglobalmapping)(x,rmapping,cmapping);
1991: } else {
1992: PetscLayoutSetISLocalToGlobalMapping(x->rmap,rmapping);
1993: PetscLayoutSetISLocalToGlobalMapping(x->cmap,cmapping);
1994: }
1995: return(0);
1996: }
1999: /*@
2000: MatGetLocalToGlobalMapping - Gets the local-to-global numbering set by MatSetLocalToGlobalMapping()
2002: Not Collective
2004: Input Parameters:
2005: . A - the matrix
2007: Output Parameters:
2008: + rmapping - row mapping
2009: - cmapping - column mapping
2011: Level: advanced
2014: .seealso: MatSetValuesLocal()
2015: @*/
2016: PetscErrorCode MatGetLocalToGlobalMapping(Mat A,ISLocalToGlobalMapping *rmapping,ISLocalToGlobalMapping *cmapping)
2017: {
2023: if (rmapping) *rmapping = A->rmap->mapping;
2024: if (cmapping) *cmapping = A->cmap->mapping;
2025: return(0);
2026: }
2028: /*@
2029: MatGetLayouts - Gets the PetscLayout objects for rows and columns
2031: Not Collective
2033: Input Parameters:
2034: . A - the matrix
2036: Output Parameters:
2037: + rmap - row layout
2038: - cmap - column layout
2040: Level: advanced
2042: .seealso: MatCreateVecs(), MatGetLocalToGlobalMapping()
2043: @*/
2044: PetscErrorCode MatGetLayouts(Mat A,PetscLayout *rmap,PetscLayout *cmap)
2045: {
2051: if (rmap) *rmap = A->rmap;
2052: if (cmap) *cmap = A->cmap;
2053: return(0);
2054: }
2056: /*@C
2057: MatSetValuesLocal - Inserts or adds values into certain locations of a matrix,
2058: using a local ordering of the nodes.
2060: Not Collective
2062: Input Parameters:
2063: + mat - the matrix
2064: . nrow, irow - number of rows and their local indices
2065: . ncol, icol - number of columns and their local indices
2066: . y - a logically two-dimensional array of values
2067: - addv - either INSERT_VALUES or ADD_VALUES, where
2068: ADD_VALUES adds values to any existing entries, and
2069: INSERT_VALUES replaces existing entries with new values
2071: Notes:
2072: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
2073: MatSetUp() before using this routine
2075: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatSetLocalToGlobalMapping() before using this routine
2077: Calls to MatSetValuesLocal() with the INSERT_VALUES and ADD_VALUES
2078: options cannot be mixed without intervening calls to the assembly
2079: routines.
2081: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
2082: MUST be called after all calls to MatSetValuesLocal() have been completed.
2084: Level: intermediate
2086: Developer Notes:
2087: This is labeled with C so does not automatically generate Fortran stubs and interfaces
2088: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
2090: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), MatSetValues(), MatSetLocalToGlobalMapping(),
2091: MatSetValueLocal()
2092: @*/
2093: PetscErrorCode MatSetValuesLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
2094: {
2100: MatCheckPreallocated(mat,1);
2101: if (!nrow || !ncol) return(0); /* no values to insert */
2104: if (mat->insertmode == NOT_SET_VALUES) {
2105: mat->insertmode = addv;
2106: }
2107: #if defined(PETSC_USE_DEBUG)
2108: else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
2109: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2110: if (!mat->ops->setvalueslocal && !mat->ops->setvalues) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2111: #endif
2113: if (mat->assembled) {
2114: mat->was_assembled = PETSC_TRUE;
2115: mat->assembled = PETSC_FALSE;
2116: }
2117: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
2118: if (mat->ops->setvalueslocal) {
2119: (*mat->ops->setvalueslocal)(mat,nrow,irow,ncol,icol,y,addv);
2120: } else {
2121: PetscInt buf[8192],*bufr=0,*bufc=0,*irowm,*icolm;
2122: if ((nrow+ncol) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
2123: irowm = buf; icolm = buf+nrow;
2124: } else {
2125: PetscMalloc2(nrow,&bufr,ncol,&bufc);
2126: irowm = bufr; icolm = bufc;
2127: }
2128: ISLocalToGlobalMappingApply(mat->rmap->mapping,nrow,irow,irowm);
2129: ISLocalToGlobalMappingApply(mat->cmap->mapping,ncol,icol,icolm);
2130: MatSetValues(mat,nrow,irowm,ncol,icolm,y,addv);
2131: PetscFree2(bufr,bufc);
2132: }
2133: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
2134: return(0);
2135: }
2137: /*@C
2138: MatSetValuesBlockedLocal - Inserts or adds values into certain locations of a matrix,
2139: using a local ordering of the nodes a block at a time.
2141: Not Collective
2143: Input Parameters:
2144: + x - the matrix
2145: . nrow, irow - number of rows and their local indices
2146: . ncol, icol - number of columns and their local indices
2147: . y - a logically two-dimensional array of values
2148: - addv - either INSERT_VALUES or ADD_VALUES, where
2149: ADD_VALUES adds values to any existing entries, and
2150: INSERT_VALUES replaces existing entries with new values
2152: Notes:
2153: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatXXXXSetPreallocation() or
2154: MatSetUp() before using this routine
2156: If you create the matrix yourself (that is not with a call to DMCreateMatrix()) then you MUST call MatSetBlockSize() and MatSetLocalToGlobalMapping()
2157: before using this routineBefore calling MatSetValuesLocal(), the user must first set the
2159: Calls to MatSetValuesBlockedLocal() with the INSERT_VALUES and ADD_VALUES
2160: options cannot be mixed without intervening calls to the assembly
2161: routines.
2163: These values may be cached, so MatAssemblyBegin() and MatAssemblyEnd()
2164: MUST be called after all calls to MatSetValuesBlockedLocal() have been completed.
2166: Level: intermediate
2168: Developer Notes:
2169: This is labeled with C so does not automatically generate Fortran stubs and interfaces
2170: because it requires multiple Fortran interfaces depending on which arguments are scalar or arrays.
2172: .seealso: MatSetBlockSize(), MatSetLocalToGlobalMapping(), MatAssemblyBegin(), MatAssemblyEnd(),
2173: MatSetValuesLocal(), MatSetValuesBlocked()
2174: @*/
2175: PetscErrorCode MatSetValuesBlockedLocal(Mat mat,PetscInt nrow,const PetscInt irow[],PetscInt ncol,const PetscInt icol[],const PetscScalar y[],InsertMode addv)
2176: {
2182: MatCheckPreallocated(mat,1);
2183: if (!nrow || !ncol) return(0); /* no values to insert */
2187: if (mat->insertmode == NOT_SET_VALUES) {
2188: mat->insertmode = addv;
2189: }
2190: #if defined(PETSC_USE_DEBUG)
2191: else if (mat->insertmode != addv) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Cannot mix add values and insert values");
2192: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2193: 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);
2194: #endif
2196: if (mat->assembled) {
2197: mat->was_assembled = PETSC_TRUE;
2198: mat->assembled = PETSC_FALSE;
2199: }
2200: #if defined(PETSC_USE_DEBUG)
2201: /* Condition on the mapping existing, because MatSetValuesBlockedLocal_IS does not require it to be set. */
2202: if (mat->rmap->mapping) {
2203: PetscInt irbs, rbs;
2204: MatGetBlockSizes(mat, &rbs, NULL);
2205: ISLocalToGlobalMappingGetBlockSize(mat->rmap->mapping,&irbs);
2206: if (rbs != irbs) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Different row block sizes! mat %D, row l2g map %D",rbs,irbs);
2207: }
2208: if (mat->cmap->mapping) {
2209: PetscInt icbs, cbs;
2210: MatGetBlockSizes(mat,NULL,&cbs);
2211: ISLocalToGlobalMappingGetBlockSize(mat->cmap->mapping,&icbs);
2212: if (cbs != icbs) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Different col block sizes! mat %D, col l2g map %D",cbs,icbs);
2213: }
2214: #endif
2215: PetscLogEventBegin(MAT_SetValues,mat,0,0,0);
2216: if (mat->ops->setvaluesblockedlocal) {
2217: (*mat->ops->setvaluesblockedlocal)(mat,nrow,irow,ncol,icol,y,addv);
2218: } else {
2219: PetscInt buf[8192],*bufr=0,*bufc=0,*irowm,*icolm;
2220: if ((nrow+ncol) <= (PetscInt)(sizeof(buf)/sizeof(PetscInt))) {
2221: irowm = buf; icolm = buf + nrow;
2222: } else {
2223: PetscMalloc2(nrow,&bufr,ncol,&bufc);
2224: irowm = bufr; icolm = bufc;
2225: }
2226: ISLocalToGlobalMappingApplyBlock(mat->rmap->mapping,nrow,irow,irowm);
2227: ISLocalToGlobalMappingApplyBlock(mat->cmap->mapping,ncol,icol,icolm);
2228: MatSetValuesBlocked(mat,nrow,irowm,ncol,icolm,y,addv);
2229: PetscFree2(bufr,bufc);
2230: }
2231: PetscLogEventEnd(MAT_SetValues,mat,0,0,0);
2232: return(0);
2233: }
2235: /*@
2236: MatMultDiagonalBlock - Computes the matrix-vector product, y = Dx. Where D is defined by the inode or block structure of the diagonal
2238: Collective on Mat
2240: Input Parameters:
2241: + mat - the matrix
2242: - x - the vector to be multiplied
2244: Output Parameters:
2245: . y - the result
2247: Notes:
2248: The vectors x and y cannot be the same. I.e., one cannot
2249: call MatMult(A,y,y).
2251: Level: developer
2253: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2254: @*/
2255: PetscErrorCode MatMultDiagonalBlock(Mat mat,Vec x,Vec y)
2256: {
2265: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2266: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2267: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2268: MatCheckPreallocated(mat,1);
2270: if (!mat->ops->multdiagonalblock) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply defined",((PetscObject)mat)->type_name);
2271: (*mat->ops->multdiagonalblock)(mat,x,y);
2272: PetscObjectStateIncrease((PetscObject)y);
2273: return(0);
2274: }
2276: /* --------------------------------------------------------*/
2277: /*@
2278: MatMult - Computes the matrix-vector product, y = Ax.
2280: Neighbor-wise Collective on Mat
2282: Input Parameters:
2283: + mat - the matrix
2284: - x - the vector to be multiplied
2286: Output Parameters:
2287: . y - the result
2289: Notes:
2290: The vectors x and y cannot be the same. I.e., one cannot
2291: call MatMult(A,y,y).
2293: Level: beginner
2295: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2296: @*/
2297: PetscErrorCode MatMult(Mat mat,Vec x,Vec y)
2298: {
2306: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2307: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2308: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2309: #if !defined(PETSC_HAVE_CONSTRAINTS)
2310: 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);
2311: 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);
2312: 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);
2313: #endif
2314: VecSetErrorIfLocked(y,3);
2315: if (mat->erroriffailure) {VecValidValues(x,2,PETSC_TRUE);}
2316: MatCheckPreallocated(mat,1);
2318: VecLockReadPush(x);
2319: if (!mat->ops->mult) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply defined",((PetscObject)mat)->type_name);
2320: PetscLogEventBegin(MAT_Mult,mat,x,y,0);
2321: (*mat->ops->mult)(mat,x,y);
2322: PetscLogEventEnd(MAT_Mult,mat,x,y,0);
2323: if (mat->erroriffailure) {VecValidValues(y,3,PETSC_FALSE);}
2324: VecLockReadPop(x);
2325: return(0);
2326: }
2328: /*@
2329: MatMultTranspose - Computes matrix transpose times a vector y = A^T * x.
2331: Neighbor-wise Collective on Mat
2333: Input Parameters:
2334: + mat - the matrix
2335: - x - the vector to be multiplied
2337: Output Parameters:
2338: . y - the result
2340: Notes:
2341: The vectors x and y cannot be the same. I.e., one cannot
2342: call MatMultTranspose(A,y,y).
2344: For complex numbers this does NOT compute the Hermitian (complex conjugate) transpose multiple,
2345: use MatMultHermitianTranspose()
2347: Level: beginner
2349: .seealso: MatMult(), MatMultAdd(), MatMultTransposeAdd(), MatMultHermitianTranspose(), MatTranspose()
2350: @*/
2351: PetscErrorCode MatMultTranspose(Mat mat,Vec x,Vec y)
2352: {
2361: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2362: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2363: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2364: #if !defined(PETSC_HAVE_CONSTRAINTS)
2365: 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);
2366: 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);
2367: #endif
2368: if (mat->erroriffailure) {VecValidValues(x,2,PETSC_TRUE);}
2369: MatCheckPreallocated(mat,1);
2371: if (!mat->ops->multtranspose) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s does not have a multiply transpose defined",((PetscObject)mat)->type_name);
2372: PetscLogEventBegin(MAT_MultTranspose,mat,x,y,0);
2373: VecLockReadPush(x);
2374: (*mat->ops->multtranspose)(mat,x,y);
2375: VecLockReadPop(x);
2376: PetscLogEventEnd(MAT_MultTranspose,mat,x,y,0);
2377: PetscObjectStateIncrease((PetscObject)y);
2378: if (mat->erroriffailure) {VecValidValues(y,3,PETSC_FALSE);}
2379: return(0);
2380: }
2382: /*@
2383: MatMultHermitianTranspose - Computes matrix Hermitian transpose times a vector.
2385: Neighbor-wise Collective on Mat
2387: Input Parameters:
2388: + mat - the matrix
2389: - x - the vector to be multilplied
2391: Output Parameters:
2392: . y - the result
2394: Notes:
2395: The vectors x and y cannot be the same. I.e., one cannot
2396: call MatMultHermitianTranspose(A,y,y).
2398: Also called the conjugate transpose, complex conjugate transpose, or adjoint.
2400: For real numbers MatMultTranspose() and MatMultHermitianTranspose() are identical.
2402: Level: beginner
2404: .seealso: MatMult(), MatMultAdd(), MatMultHermitianTransposeAdd(), MatMultTranspose()
2405: @*/
2406: PetscErrorCode MatMultHermitianTranspose(Mat mat,Vec x,Vec y)
2407: {
2409: Vec w;
2417: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2418: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2419: if (x == y) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2420: #if !defined(PETSC_HAVE_CONSTRAINTS)
2421: 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);
2422: 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);
2423: #endif
2424: MatCheckPreallocated(mat,1);
2426: PetscLogEventBegin(MAT_MultHermitianTranspose,mat,x,y,0);
2427: if (mat->ops->multhermitiantranspose) {
2428: VecLockReadPush(x);
2429: (*mat->ops->multhermitiantranspose)(mat,x,y);
2430: VecLockReadPop(x);
2431: } else {
2432: VecDuplicate(x,&w);
2433: VecCopy(x,w);
2434: VecConjugate(w);
2435: MatMultTranspose(mat,w,y);
2436: VecDestroy(&w);
2437: VecConjugate(y);
2438: }
2439: PetscLogEventEnd(MAT_MultHermitianTranspose,mat,x,y,0);
2440: PetscObjectStateIncrease((PetscObject)y);
2441: return(0);
2442: }
2444: /*@
2445: MatMultAdd - Computes v3 = v2 + A * v1.
2447: Neighbor-wise Collective on Mat
2449: Input Parameters:
2450: + mat - the matrix
2451: - v1, v2 - the vectors
2453: Output Parameters:
2454: . v3 - the result
2456: Notes:
2457: The vectors v1 and v3 cannot be the same. I.e., one cannot
2458: call MatMultAdd(A,v1,v2,v1).
2460: Level: beginner
2462: .seealso: MatMultTranspose(), MatMult(), MatMultTransposeAdd()
2463: @*/
2464: PetscErrorCode MatMultAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2465: {
2475: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2476: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2477: 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);
2478: /* 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);
2479: 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); */
2480: 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);
2481: 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);
2482: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2483: MatCheckPreallocated(mat,1);
2485: if (!mat->ops->multadd) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"No MatMultAdd() for matrix type %s",((PetscObject)mat)->type_name);
2486: PetscLogEventBegin(MAT_MultAdd,mat,v1,v2,v3);
2487: VecLockReadPush(v1);
2488: (*mat->ops->multadd)(mat,v1,v2,v3);
2489: VecLockReadPop(v1);
2490: PetscLogEventEnd(MAT_MultAdd,mat,v1,v2,v3);
2491: PetscObjectStateIncrease((PetscObject)v3);
2492: return(0);
2493: }
2495: /*@
2496: MatMultTransposeAdd - Computes v3 = v2 + A' * v1.
2498: Neighbor-wise Collective on Mat
2500: Input Parameters:
2501: + mat - the matrix
2502: - v1, v2 - the vectors
2504: Output Parameters:
2505: . v3 - the result
2507: Notes:
2508: The vectors v1 and v3 cannot be the same. I.e., one cannot
2509: call MatMultTransposeAdd(A,v1,v2,v1).
2511: Level: beginner
2513: .seealso: MatMultTranspose(), MatMultAdd(), MatMult()
2514: @*/
2515: PetscErrorCode MatMultTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2516: {
2526: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2527: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2528: if (!mat->ops->multtransposeadd) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2529: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2530: 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);
2531: 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);
2532: 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);
2533: MatCheckPreallocated(mat,1);
2535: PetscLogEventBegin(MAT_MultTransposeAdd,mat,v1,v2,v3);
2536: VecLockReadPush(v1);
2537: (*mat->ops->multtransposeadd)(mat,v1,v2,v3);
2538: VecLockReadPop(v1);
2539: PetscLogEventEnd(MAT_MultTransposeAdd,mat,v1,v2,v3);
2540: PetscObjectStateIncrease((PetscObject)v3);
2541: return(0);
2542: }
2544: /*@
2545: MatMultHermitianTransposeAdd - Computes v3 = v2 + A^H * v1.
2547: Neighbor-wise Collective on Mat
2549: Input Parameters:
2550: + mat - the matrix
2551: - v1, v2 - the vectors
2553: Output Parameters:
2554: . v3 - the result
2556: Notes:
2557: The vectors v1 and v3 cannot be the same. I.e., one cannot
2558: call MatMultHermitianTransposeAdd(A,v1,v2,v1).
2560: Level: beginner
2562: .seealso: MatMultHermitianTranspose(), MatMultTranspose(), MatMultAdd(), MatMult()
2563: @*/
2564: PetscErrorCode MatMultHermitianTransposeAdd(Mat mat,Vec v1,Vec v2,Vec v3)
2565: {
2575: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2576: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2577: if (v1 == v3) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"v1 and v3 must be different vectors");
2578: 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);
2579: 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);
2580: 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);
2581: MatCheckPreallocated(mat,1);
2583: PetscLogEventBegin(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);
2584: VecLockReadPush(v1);
2585: if (mat->ops->multhermitiantransposeadd) {
2586: (*mat->ops->multhermitiantransposeadd)(mat,v1,v2,v3);
2587: } else {
2588: Vec w,z;
2589: VecDuplicate(v1,&w);
2590: VecCopy(v1,w);
2591: VecConjugate(w);
2592: VecDuplicate(v3,&z);
2593: MatMultTranspose(mat,w,z);
2594: VecDestroy(&w);
2595: VecConjugate(z);
2596: if (v2 != v3) {
2597: VecWAXPY(v3,1.0,v2,z);
2598: } else {
2599: VecAXPY(v3,1.0,z);
2600: }
2601: VecDestroy(&z);
2602: }
2603: VecLockReadPop(v1);
2604: PetscLogEventEnd(MAT_MultHermitianTransposeAdd,mat,v1,v2,v3);
2605: PetscObjectStateIncrease((PetscObject)v3);
2606: return(0);
2607: }
2609: /*@
2610: MatMultConstrained - The inner multiplication routine for a
2611: constrained matrix P^T A P.
2613: Neighbor-wise Collective on Mat
2615: Input Parameters:
2616: + mat - the matrix
2617: - x - the vector to be multilplied
2619: Output Parameters:
2620: . y - the result
2622: Notes:
2623: The vectors x and y cannot be the same. I.e., one cannot
2624: call MatMult(A,y,y).
2626: Level: beginner
2628: .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2629: @*/
2630: PetscErrorCode MatMultConstrained(Mat mat,Vec x,Vec y)
2631: {
2638: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2639: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2640: if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2641: 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);
2642: 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);
2643: 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);
2645: PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);
2646: VecLockReadPush(x);
2647: (*mat->ops->multconstrained)(mat,x,y);
2648: VecLockReadPop(x);
2649: PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);
2650: PetscObjectStateIncrease((PetscObject)y);
2651: return(0);
2652: }
2654: /*@
2655: MatMultTransposeConstrained - The inner multiplication routine for a
2656: constrained matrix P^T A^T P.
2658: Neighbor-wise Collective on Mat
2660: Input Parameters:
2661: + mat - the matrix
2662: - x - the vector to be multilplied
2664: Output Parameters:
2665: . y - the result
2667: Notes:
2668: The vectors x and y cannot be the same. I.e., one cannot
2669: call MatMult(A,y,y).
2671: Level: beginner
2673: .seealso: MatMult(), MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
2674: @*/
2675: PetscErrorCode MatMultTransposeConstrained(Mat mat,Vec x,Vec y)
2676: {
2683: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2684: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2685: if (x == y) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"x and y must be different vectors");
2686: 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);
2687: 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);
2689: PetscLogEventBegin(MAT_MultConstrained,mat,x,y,0);
2690: (*mat->ops->multtransposeconstrained)(mat,x,y);
2691: PetscLogEventEnd(MAT_MultConstrained,mat,x,y,0);
2692: PetscObjectStateIncrease((PetscObject)y);
2693: return(0);
2694: }
2696: /*@C
2697: MatGetFactorType - gets the type of factorization it is
2699: Not Collective
2701: Input Parameters:
2702: . mat - the matrix
2704: Output Parameters:
2705: . t - the type, one of MAT_FACTOR_NONE, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ILU, MAT_FACTOR_ICC,MAT_FACTOR_ILUDT
2707: Level: intermediate
2709: .seealso: MatFactorType, MatGetFactor(), MatSetFactorType()
2710: @*/
2711: PetscErrorCode MatGetFactorType(Mat mat,MatFactorType *t)
2712: {
2717: *t = mat->factortype;
2718: return(0);
2719: }
2721: /*@C
2722: MatSetFactorType - sets the type of factorization it is
2724: Logically Collective on Mat
2726: Input Parameters:
2727: + mat - the matrix
2728: - t - the type, one of MAT_FACTOR_NONE, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ILU, MAT_FACTOR_ICC,MAT_FACTOR_ILUDT
2730: Level: intermediate
2732: .seealso: MatFactorType, MatGetFactor(), MatGetFactorType()
2733: @*/
2734: PetscErrorCode MatSetFactorType(Mat mat, MatFactorType t)
2735: {
2739: mat->factortype = t;
2740: return(0);
2741: }
2743: /* ------------------------------------------------------------*/
2744: /*@C
2745: MatGetInfo - Returns information about matrix storage (number of
2746: nonzeros, memory, etc.).
2748: Collective on Mat if MAT_GLOBAL_MAX or MAT_GLOBAL_SUM is used as the flag
2750: Input Parameters:
2751: . mat - the matrix
2753: Output Parameters:
2754: + flag - flag indicating the type of parameters to be returned
2755: (MAT_LOCAL - local matrix, MAT_GLOBAL_MAX - maximum over all processors,
2756: MAT_GLOBAL_SUM - sum over all processors)
2757: - info - matrix information context
2759: Notes:
2760: The MatInfo context contains a variety of matrix data, including
2761: number of nonzeros allocated and used, number of mallocs during
2762: matrix assembly, etc. Additional information for factored matrices
2763: is provided (such as the fill ratio, number of mallocs during
2764: factorization, etc.). Much of this info is printed to PETSC_STDOUT
2765: when using the runtime options
2766: $ -info -mat_view ::ascii_info
2768: Example for C/C++ Users:
2769: See the file ${PETSC_DIR}/include/petscmat.h for a complete list of
2770: data within the MatInfo context. For example,
2771: .vb
2772: MatInfo info;
2773: Mat A;
2774: double mal, nz_a, nz_u;
2776: MatGetInfo(A,MAT_LOCAL,&info);
2777: mal = info.mallocs;
2778: nz_a = info.nz_allocated;
2779: .ve
2781: Example for Fortran Users:
2782: Fortran users should declare info as a double precision
2783: array of dimension MAT_INFO_SIZE, and then extract the parameters
2784: of interest. See the file ${PETSC_DIR}/include/petsc/finclude/petscmat.h
2785: a complete list of parameter names.
2786: .vb
2787: double precision info(MAT_INFO_SIZE)
2788: double precision mal, nz_a
2789: Mat A
2790: integer ierr
2792: call MatGetInfo(A,MAT_LOCAL,info,ierr)
2793: mal = info(MAT_INFO_MALLOCS)
2794: nz_a = info(MAT_INFO_NZ_ALLOCATED)
2795: .ve
2797: Level: intermediate
2799: Developer Note: fortran interface is not autogenerated as the f90
2800: interface defintion cannot be generated correctly [due to MatInfo]
2802: .seealso: MatStashGetInfo()
2804: @*/
2805: PetscErrorCode MatGetInfo(Mat mat,MatInfoType flag,MatInfo *info)
2806: {
2813: if (!mat->ops->getinfo) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2814: MatCheckPreallocated(mat,1);
2815: (*mat->ops->getinfo)(mat,flag,info);
2816: return(0);
2817: }
2819: /*
2820: This is used by external packages where it is not easy to get the info from the actual
2821: matrix factorization.
2822: */
2823: PetscErrorCode MatGetInfo_External(Mat A,MatInfoType flag,MatInfo *info)
2824: {
2828: PetscMemzero(info,sizeof(MatInfo));
2829: return(0);
2830: }
2832: /* ----------------------------------------------------------*/
2834: /*@C
2835: MatLUFactor - Performs in-place LU factorization of matrix.
2837: Collective on Mat
2839: Input Parameters:
2840: + mat - the matrix
2841: . row - row permutation
2842: . col - column permutation
2843: - info - options for factorization, includes
2844: $ fill - expected fill as ratio of original fill.
2845: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2846: $ Run with the option -info to determine an optimal value to use
2848: Notes:
2849: Most users should employ the simplified KSP interface for linear solvers
2850: instead of working directly with matrix algebra routines such as this.
2851: See, e.g., KSPCreate().
2853: This changes the state of the matrix to a factored matrix; it cannot be used
2854: for example with MatSetValues() unless one first calls MatSetUnfactored().
2856: Level: developer
2858: .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(),
2859: MatGetOrdering(), MatSetUnfactored(), MatFactorInfo, MatGetFactor()
2861: Developer Note: fortran interface is not autogenerated as the f90
2862: interface defintion cannot be generated correctly [due to MatFactorInfo]
2864: @*/
2865: PetscErrorCode MatLUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
2866: {
2868: MatFactorInfo tinfo;
2876: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2877: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2878: if (!mat->ops->lufactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2879: MatCheckPreallocated(mat,1);
2880: if (!info) {
2881: MatFactorInfoInitialize(&tinfo);
2882: info = &tinfo;
2883: }
2885: PetscLogEventBegin(MAT_LUFactor,mat,row,col,0);
2886: (*mat->ops->lufactor)(mat,row,col,info);
2887: PetscLogEventEnd(MAT_LUFactor,mat,row,col,0);
2888: PetscObjectStateIncrease((PetscObject)mat);
2889: return(0);
2890: }
2892: /*@C
2893: MatILUFactor - Performs in-place ILU factorization of matrix.
2895: Collective on Mat
2897: Input Parameters:
2898: + mat - the matrix
2899: . row - row permutation
2900: . col - column permutation
2901: - info - structure containing
2902: $ levels - number of levels of fill.
2903: $ expected fill - as ratio of original fill.
2904: $ 1 or 0 - indicating force fill on diagonal (improves robustness for matrices
2905: missing diagonal entries)
2907: Notes:
2908: Probably really in-place only when level of fill is zero, otherwise allocates
2909: new space to store factored matrix and deletes previous memory.
2911: Most users should employ the simplified KSP interface for linear solvers
2912: instead of working directly with matrix algebra routines such as this.
2913: See, e.g., KSPCreate().
2915: Level: developer
2917: .seealso: MatILUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
2919: Developer Note: fortran interface is not autogenerated as the f90
2920: interface defintion cannot be generated correctly [due to MatFactorInfo]
2922: @*/
2923: PetscErrorCode MatILUFactor(Mat mat,IS row,IS col,const MatFactorInfo *info)
2924: {
2933: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"matrix must be square");
2934: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2935: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2936: if (!mat->ops->ilufactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
2937: MatCheckPreallocated(mat,1);
2939: PetscLogEventBegin(MAT_ILUFactor,mat,row,col,0);
2940: (*mat->ops->ilufactor)(mat,row,col,info);
2941: PetscLogEventEnd(MAT_ILUFactor,mat,row,col,0);
2942: PetscObjectStateIncrease((PetscObject)mat);
2943: return(0);
2944: }
2946: /*@C
2947: MatLUFactorSymbolic - Performs symbolic LU factorization of matrix.
2948: Call this routine before calling MatLUFactorNumeric().
2950: Collective on Mat
2952: Input Parameters:
2953: + fact - the factor matrix obtained with MatGetFactor()
2954: . mat - the matrix
2955: . row, col - row and column permutations
2956: - info - options for factorization, includes
2957: $ fill - expected fill as ratio of original fill.
2958: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
2959: $ Run with the option -info to determine an optimal value to use
2962: Notes:
2963: See Users-Manual: ch_mat for additional information about choosing the fill factor for better efficiency.
2965: Most users should employ the simplified KSP interface for linear solvers
2966: instead of working directly with matrix algebra routines such as this.
2967: See, e.g., KSPCreate().
2969: Level: developer
2971: .seealso: MatLUFactor(), MatLUFactorNumeric(), MatCholeskyFactor(), MatFactorInfo, MatFactorInfoInitialize()
2973: Developer Note: fortran interface is not autogenerated as the f90
2974: interface defintion cannot be generated correctly [due to MatFactorInfo]
2976: @*/
2977: PetscErrorCode MatLUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
2978: {
2980: MatFactorInfo tinfo;
2989: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
2990: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
2991: if (!(fact)->ops->lufactorsymbolic) {
2992: MatSolverType spackage;
2993: MatFactorGetSolverType(fact,&spackage);
2994: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic LU using solver package %s",((PetscObject)mat)->type_name,spackage);
2995: }
2996: MatCheckPreallocated(mat,2);
2997: if (!info) {
2998: MatFactorInfoInitialize(&tinfo);
2999: info = &tinfo;
3000: }
3002: PetscLogEventBegin(MAT_LUFactorSymbolic,mat,row,col,0);
3003: (fact->ops->lufactorsymbolic)(fact,mat,row,col,info);
3004: PetscLogEventEnd(MAT_LUFactorSymbolic,mat,row,col,0);
3005: PetscObjectStateIncrease((PetscObject)fact);
3006: return(0);
3007: }
3009: /*@C
3010: MatLUFactorNumeric - Performs numeric LU factorization of a matrix.
3011: Call this routine after first calling MatLUFactorSymbolic().
3013: Collective on Mat
3015: Input Parameters:
3016: + fact - the factor matrix obtained with MatGetFactor()
3017: . mat - the matrix
3018: - info - options for factorization
3020: Notes:
3021: See MatLUFactor() for in-place factorization. See
3022: MatCholeskyFactorNumeric() for the symmetric, positive definite case.
3024: Most users should employ the simplified KSP interface for linear solvers
3025: instead of working directly with matrix algebra routines such as this.
3026: See, e.g., KSPCreate().
3028: Level: developer
3030: .seealso: MatLUFactorSymbolic(), MatLUFactor(), MatCholeskyFactor()
3032: Developer Note: fortran interface is not autogenerated as the f90
3033: interface defintion cannot be generated correctly [due to MatFactorInfo]
3035: @*/
3036: PetscErrorCode MatLUFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
3037: {
3038: MatFactorInfo tinfo;
3046: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3047: 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);
3049: if (!(fact)->ops->lufactornumeric) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s numeric LU",((PetscObject)mat)->type_name);
3050: MatCheckPreallocated(mat,2);
3051: if (!info) {
3052: MatFactorInfoInitialize(&tinfo);
3053: info = &tinfo;
3054: }
3056: PetscLogEventBegin(MAT_LUFactorNumeric,mat,fact,0,0);
3057: (fact->ops->lufactornumeric)(fact,mat,info);
3058: PetscLogEventEnd(MAT_LUFactorNumeric,mat,fact,0,0);
3059: MatViewFromOptions(fact,NULL,"-mat_factor_view");
3060: PetscObjectStateIncrease((PetscObject)fact);
3061: return(0);
3062: }
3064: /*@C
3065: MatCholeskyFactor - Performs in-place Cholesky factorization of a
3066: symmetric matrix.
3068: Collective on Mat
3070: Input Parameters:
3071: + mat - the matrix
3072: . perm - row and column permutations
3073: - f - expected fill as ratio of original fill
3075: Notes:
3076: See MatLUFactor() for the nonsymmetric case. See also
3077: MatCholeskyFactorSymbolic(), and MatCholeskyFactorNumeric().
3079: Most users should employ the simplified KSP interface for linear solvers
3080: instead of working directly with matrix algebra routines such as this.
3081: See, e.g., KSPCreate().
3083: Level: developer
3085: .seealso: MatLUFactor(), MatCholeskyFactorSymbolic(), MatCholeskyFactorNumeric()
3086: MatGetOrdering()
3088: Developer Note: fortran interface is not autogenerated as the f90
3089: interface defintion cannot be generated correctly [due to MatFactorInfo]
3091: @*/
3092: PetscErrorCode MatCholeskyFactor(Mat mat,IS perm,const MatFactorInfo *info)
3093: {
3095: MatFactorInfo tinfo;
3102: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"Matrix must be square");
3103: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3104: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3105: 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);
3106: MatCheckPreallocated(mat,1);
3107: if (!info) {
3108: MatFactorInfoInitialize(&tinfo);
3109: info = &tinfo;
3110: }
3112: PetscLogEventBegin(MAT_CholeskyFactor,mat,perm,0,0);
3113: (*mat->ops->choleskyfactor)(mat,perm,info);
3114: PetscLogEventEnd(MAT_CholeskyFactor,mat,perm,0,0);
3115: PetscObjectStateIncrease((PetscObject)mat);
3116: return(0);
3117: }
3119: /*@C
3120: MatCholeskyFactorSymbolic - Performs symbolic Cholesky factorization
3121: of a symmetric matrix.
3123: Collective on Mat
3125: Input Parameters:
3126: + fact - the factor matrix obtained with MatGetFactor()
3127: . mat - the matrix
3128: . perm - row and column permutations
3129: - info - options for factorization, includes
3130: $ fill - expected fill as ratio of original fill.
3131: $ dtcol - pivot tolerance (0 no pivot, 1 full column pivoting)
3132: $ Run with the option -info to determine an optimal value to use
3134: Notes:
3135: See MatLUFactorSymbolic() for the nonsymmetric case. See also
3136: MatCholeskyFactor() and MatCholeskyFactorNumeric().
3138: Most users should employ the simplified KSP interface for linear solvers
3139: instead of working directly with matrix algebra routines such as this.
3140: See, e.g., KSPCreate().
3142: Level: developer
3144: .seealso: MatLUFactorSymbolic(), MatCholeskyFactor(), MatCholeskyFactorNumeric()
3145: MatGetOrdering()
3147: Developer Note: fortran interface is not autogenerated as the f90
3148: interface defintion cannot be generated correctly [due to MatFactorInfo]
3150: @*/
3151: PetscErrorCode MatCholeskyFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
3152: {
3154: MatFactorInfo tinfo;
3162: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"Matrix must be square");
3163: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3164: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3165: if (!(fact)->ops->choleskyfactorsymbolic) {
3166: MatSolverType spackage;
3167: MatFactorGetSolverType(fact,&spackage);
3168: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s symbolic factor Cholesky using solver package %s",((PetscObject)mat)->type_name,spackage);
3169: }
3170: MatCheckPreallocated(mat,2);
3171: if (!info) {
3172: MatFactorInfoInitialize(&tinfo);
3173: info = &tinfo;
3174: }
3176: PetscLogEventBegin(MAT_CholeskyFactorSymbolic,mat,perm,0,0);
3177: (fact->ops->choleskyfactorsymbolic)(fact,mat,perm,info);
3178: PetscLogEventEnd(MAT_CholeskyFactorSymbolic,mat,perm,0,0);
3179: PetscObjectStateIncrease((PetscObject)fact);
3180: return(0);
3181: }
3183: /*@C
3184: MatCholeskyFactorNumeric - Performs numeric Cholesky factorization
3185: of a symmetric matrix. Call this routine after first calling
3186: MatCholeskyFactorSymbolic().
3188: Collective on Mat
3190: Input Parameters:
3191: + fact - the factor matrix obtained with MatGetFactor()
3192: . mat - the initial matrix
3193: . info - options for factorization
3194: - fact - the symbolic factor of mat
3197: Notes:
3198: Most users should employ the simplified KSP interface for linear solvers
3199: instead of working directly with matrix algebra routines such as this.
3200: See, e.g., KSPCreate().
3202: Level: developer
3204: .seealso: MatCholeskyFactorSymbolic(), MatCholeskyFactor(), MatLUFactorNumeric()
3206: Developer Note: fortran interface is not autogenerated as the f90
3207: interface defintion cannot be generated correctly [due to MatFactorInfo]
3209: @*/
3210: PetscErrorCode MatCholeskyFactorNumeric(Mat fact,Mat mat,const MatFactorInfo *info)
3211: {
3212: MatFactorInfo tinfo;
3220: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3221: if (!(fact)->ops->choleskyfactornumeric) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s numeric factor Cholesky",((PetscObject)mat)->type_name);
3222: 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);
3223: MatCheckPreallocated(mat,2);
3224: if (!info) {
3225: MatFactorInfoInitialize(&tinfo);
3226: info = &tinfo;
3227: }
3229: PetscLogEventBegin(MAT_CholeskyFactorNumeric,mat,fact,0,0);
3230: (fact->ops->choleskyfactornumeric)(fact,mat,info);
3231: PetscLogEventEnd(MAT_CholeskyFactorNumeric,mat,fact,0,0);
3232: MatViewFromOptions(fact,NULL,"-mat_factor_view");
3233: PetscObjectStateIncrease((PetscObject)fact);
3234: return(0);
3235: }
3237: /* ----------------------------------------------------------------*/
3238: /*@
3239: MatSolve - Solves A x = b, given a factored matrix.
3241: Neighbor-wise Collective on Mat
3243: Input Parameters:
3244: + mat - the factored matrix
3245: - b - the right-hand-side vector
3247: Output Parameter:
3248: . x - the result vector
3250: Notes:
3251: The vectors b and x cannot be the same. I.e., one cannot
3252: call MatSolve(A,x,x).
3254: Notes:
3255: Most users should employ the simplified KSP interface for linear solvers
3256: instead of working directly with matrix algebra routines such as this.
3257: See, e.g., KSPCreate().
3259: Level: developer
3261: .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd()
3262: @*/
3263: PetscErrorCode MatSolve(Mat mat,Vec b,Vec x)
3264: {
3274: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3275: 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);
3276: 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);
3277: 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);
3278: if (!mat->rmap->N && !mat->cmap->N) return(0);
3279: if (!mat->ops->solve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3280: MatCheckPreallocated(mat,1);
3282: PetscLogEventBegin(MAT_Solve,mat,b,x,0);
3283: if (mat->factorerrortype) {
3284: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
3285: VecSetInf(x);
3286: } else {
3287: if (!mat->ops->solve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3288: (*mat->ops->solve)(mat,b,x);
3289: }
3290: PetscLogEventEnd(MAT_Solve,mat,b,x,0);
3291: PetscObjectStateIncrease((PetscObject)x);
3292: return(0);
3293: }
3295: static PetscErrorCode MatMatSolve_Basic(Mat A,Mat B,Mat X,PetscBool trans)
3296: {
3298: Vec b,x;
3299: PetscInt m,N,i;
3300: PetscScalar *bb,*xx;
3303: MatDenseGetArrayRead(B,(const PetscScalar**)&bb);
3304: MatDenseGetArray(X,&xx);
3305: MatGetLocalSize(B,&m,NULL); /* number local rows */
3306: MatGetSize(B,NULL,&N); /* total columns in dense matrix */
3307: MatCreateVecs(A,&x,&b);
3308: for (i=0; i<N; i++) {
3309: VecPlaceArray(b,bb + i*m);
3310: VecPlaceArray(x,xx + i*m);
3311: if (trans) {
3312: MatSolveTranspose(A,b,x);
3313: } else {
3314: MatSolve(A,b,x);
3315: }
3316: VecResetArray(x);
3317: VecResetArray(b);
3318: }
3319: VecDestroy(&b);
3320: VecDestroy(&x);
3321: MatDenseRestoreArrayRead(B,(const PetscScalar**)&bb);
3322: MatDenseRestoreArray(X,&xx);
3323: return(0);
3324: }
3326: /*@
3327: MatMatSolve - Solves A X = B, given a factored matrix.
3329: Neighbor-wise Collective on Mat
3331: Input Parameters:
3332: + A - the factored matrix
3333: - B - the right-hand-side matrix (dense matrix)
3335: Output Parameter:
3336: . X - the result matrix (dense matrix)
3338: Notes:
3339: The matrices b and x cannot be the same. I.e., one cannot
3340: call MatMatSolve(A,x,x).
3342: Notes:
3343: Most users should usually employ the simplified KSP interface for linear solvers
3344: instead of working directly with matrix algebra routines such as this.
3345: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3346: at a time.
3348: When using SuperLU_Dist as a parallel solver PETSc will use the SuperLU_Dist functionality to solve multiple right hand sides simultaneously. For MUMPS
3349: it calls a separate solve for each right hand side since MUMPS does not yet support distributed right hand sides.
3351: Since the resulting matrix X must always be dense we do not support sparse representation of the matrix B.
3353: Level: developer
3355: .seealso: MatMatSolveTranspose(), MatLUFactor(), MatCholeskyFactor()
3356: @*/
3357: PetscErrorCode MatMatSolve(Mat A,Mat B,Mat X)
3358: {
3368: if (X == B) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_IDN,"X and B must be different matrices");
3369: 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);
3370: 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);
3371: 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");
3372: if (!A->rmap->N && !A->cmap->N) return(0);
3373: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3374: MatCheckPreallocated(A,1);
3376: PetscLogEventBegin(MAT_MatSolve,A,B,X,0);
3377: if (!A->ops->matsolve) {
3378: PetscInfo1(A,"Mat type %s using basic MatMatSolve\n",((PetscObject)A)->type_name);
3379: MatMatSolve_Basic(A,B,X,PETSC_FALSE);
3380: } else {
3381: (*A->ops->matsolve)(A,B,X);
3382: }
3383: PetscLogEventEnd(MAT_MatSolve,A,B,X,0);
3384: PetscObjectStateIncrease((PetscObject)X);
3385: return(0);
3386: }
3388: /*@
3389: MatMatSolveTranspose - Solves A^T X = B, given a factored matrix.
3391: Neighbor-wise Collective on Mat
3393: Input Parameters:
3394: + A - the factored matrix
3395: - B - the right-hand-side matrix (dense matrix)
3397: Output Parameter:
3398: . X - the result matrix (dense matrix)
3400: Notes:
3401: The matrices B and X cannot be the same. I.e., one cannot
3402: call MatMatSolveTranspose(A,X,X).
3404: Notes:
3405: Most users should usually employ the simplified KSP interface for linear solvers
3406: instead of working directly with matrix algebra routines such as this.
3407: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3408: at a time.
3410: When using SuperLU_Dist or MUMPS as a parallel solver, PETSc will use their functionality to solve multiple right hand sides simultaneously.
3412: Level: developer
3414: .seealso: MatMatSolve(), MatLUFactor(), MatCholeskyFactor()
3415: @*/
3416: PetscErrorCode MatMatSolveTranspose(Mat A,Mat B,Mat X)
3417: {
3427: if (X == B) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_IDN,"X and B must be different matrices");
3428: 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);
3429: 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);
3430: 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);
3431: 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");
3432: if (!A->rmap->N && !A->cmap->N) return(0);
3433: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3434: MatCheckPreallocated(A,1);
3436: PetscLogEventBegin(MAT_MatSolve,A,B,X,0);
3437: if (!A->ops->matsolvetranspose) {
3438: PetscInfo1(A,"Mat type %s using basic MatMatSolveTranspose\n",((PetscObject)A)->type_name);
3439: MatMatSolve_Basic(A,B,X,PETSC_TRUE);
3440: } else {
3441: (*A->ops->matsolvetranspose)(A,B,X);
3442: }
3443: PetscLogEventEnd(MAT_MatSolve,A,B,X,0);
3444: PetscObjectStateIncrease((PetscObject)X);
3445: return(0);
3446: }
3448: /*@
3449: MatMatTransposeSolve - Solves A X = B^T, given a factored matrix.
3451: Neighbor-wise Collective on Mat
3453: Input Parameters:
3454: + A - the factored matrix
3455: - Bt - the transpose of right-hand-side matrix
3457: Output Parameter:
3458: . X - the result matrix (dense matrix)
3460: Notes:
3461: Most users should usually employ the simplified KSP interface for linear solvers
3462: instead of working directly with matrix algebra routines such as this.
3463: See, e.g., KSPCreate(). However KSP can only solve for one vector (column of X)
3464: at a time.
3466: 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().
3468: Level: developer
3470: .seealso: MatMatSolve(), MatMatSolveTranspose(), MatLUFactor(), MatCholeskyFactor()
3471: @*/
3472: PetscErrorCode MatMatTransposeSolve(Mat A,Mat Bt,Mat X)
3473: {
3484: if (X == Bt) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_IDN,"X and B must be different matrices");
3485: 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);
3486: 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);
3487: 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");
3488: if (!A->rmap->N && !A->cmap->N) return(0);
3489: if (!A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3490: MatCheckPreallocated(A,1);
3492: if (!A->ops->mattransposesolve) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
3493: PetscLogEventBegin(MAT_MatTrSolve,A,Bt,X,0);
3494: (*A->ops->mattransposesolve)(A,Bt,X);
3495: PetscLogEventEnd(MAT_MatTrSolve,A,Bt,X,0);
3496: PetscObjectStateIncrease((PetscObject)X);
3497: return(0);
3498: }
3500: /*@
3501: MatForwardSolve - Solves L x = b, given a factored matrix, A = LU, or
3502: U^T*D^(1/2) x = b, given a factored symmetric matrix, A = U^T*D*U,
3504: Neighbor-wise Collective on Mat
3506: Input Parameters:
3507: + mat - the factored matrix
3508: - b - the right-hand-side vector
3510: Output Parameter:
3511: . x - the result vector
3513: Notes:
3514: MatSolve() should be used for most applications, as it performs
3515: a forward solve followed by a backward solve.
3517: The vectors b and x cannot be the same, i.e., one cannot
3518: call MatForwardSolve(A,x,x).
3520: For matrix in seqsbaij format with block size larger than 1,
3521: the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3522: MatForwardSolve() solves U^T*D y = b, and
3523: MatBackwardSolve() solves U x = y.
3524: Thus they do not provide a symmetric preconditioner.
3526: Most users should employ the simplified KSP interface for linear solvers
3527: instead of working directly with matrix algebra routines such as this.
3528: See, e.g., KSPCreate().
3530: Level: developer
3532: .seealso: MatSolve(), MatBackwardSolve()
3533: @*/
3534: PetscErrorCode MatForwardSolve(Mat mat,Vec b,Vec x)
3535: {
3545: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3546: 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);
3547: 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);
3548: 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);
3549: if (!mat->rmap->N && !mat->cmap->N) return(0);
3550: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3551: MatCheckPreallocated(mat,1);
3553: if (!mat->ops->forwardsolve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3554: PetscLogEventBegin(MAT_ForwardSolve,mat,b,x,0);
3555: (*mat->ops->forwardsolve)(mat,b,x);
3556: PetscLogEventEnd(MAT_ForwardSolve,mat,b,x,0);
3557: PetscObjectStateIncrease((PetscObject)x);
3558: return(0);
3559: }
3561: /*@
3562: MatBackwardSolve - Solves U x = b, given a factored matrix, A = LU.
3563: D^(1/2) U x = b, given a factored symmetric matrix, A = U^T*D*U,
3565: Neighbor-wise Collective on Mat
3567: Input Parameters:
3568: + mat - the factored matrix
3569: - b - the right-hand-side vector
3571: Output Parameter:
3572: . x - the result vector
3574: Notes:
3575: MatSolve() should be used for most applications, as it performs
3576: a forward solve followed by a backward solve.
3578: The vectors b and x cannot be the same. I.e., one cannot
3579: call MatBackwardSolve(A,x,x).
3581: For matrix in seqsbaij format with block size larger than 1,
3582: the diagonal blocks are not implemented as D = D^(1/2) * D^(1/2) yet.
3583: MatForwardSolve() solves U^T*D y = b, and
3584: MatBackwardSolve() solves U x = y.
3585: Thus they do not provide a symmetric preconditioner.
3587: Most users should employ the simplified KSP interface for linear solvers
3588: instead of working directly with matrix algebra routines such as this.
3589: See, e.g., KSPCreate().
3591: Level: developer
3593: .seealso: MatSolve(), MatForwardSolve()
3594: @*/
3595: PetscErrorCode MatBackwardSolve(Mat mat,Vec b,Vec x)
3596: {
3606: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3607: 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);
3608: 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);
3609: 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);
3610: if (!mat->rmap->N && !mat->cmap->N) return(0);
3611: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3612: MatCheckPreallocated(mat,1);
3614: if (!mat->ops->backwardsolve) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3615: PetscLogEventBegin(MAT_BackwardSolve,mat,b,x,0);
3616: (*mat->ops->backwardsolve)(mat,b,x);
3617: PetscLogEventEnd(MAT_BackwardSolve,mat,b,x,0);
3618: PetscObjectStateIncrease((PetscObject)x);
3619: return(0);
3620: }
3622: /*@
3623: MatSolveAdd - Computes x = y + inv(A)*b, given a factored matrix.
3625: Neighbor-wise Collective on Mat
3627: Input Parameters:
3628: + mat - the factored matrix
3629: . b - the right-hand-side vector
3630: - y - the vector to be added to
3632: Output Parameter:
3633: . x - the result vector
3635: Notes:
3636: The vectors b and x cannot be the same. I.e., one cannot
3637: call MatSolveAdd(A,x,y,x).
3639: Most users should employ the simplified KSP interface for linear solvers
3640: instead of working directly with matrix algebra routines such as this.
3641: See, e.g., KSPCreate().
3643: Level: developer
3645: .seealso: MatSolve(), MatSolveTranspose(), MatSolveTransposeAdd()
3646: @*/
3647: PetscErrorCode MatSolveAdd(Mat mat,Vec b,Vec y,Vec x)
3648: {
3649: PetscScalar one = 1.0;
3650: Vec tmp;
3662: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3663: 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);
3664: 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);
3665: 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);
3666: 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);
3667: 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);
3668: if (!mat->rmap->N && !mat->cmap->N) return(0);
3669: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3670: MatCheckPreallocated(mat,1);
3672: PetscLogEventBegin(MAT_SolveAdd,mat,b,x,y);
3673: if (mat->ops->solveadd) {
3674: (*mat->ops->solveadd)(mat,b,y,x);
3675: } else {
3676: /* do the solve then the add manually */
3677: if (x != y) {
3678: MatSolve(mat,b,x);
3679: VecAXPY(x,one,y);
3680: } else {
3681: VecDuplicate(x,&tmp);
3682: PetscLogObjectParent((PetscObject)mat,(PetscObject)tmp);
3683: VecCopy(x,tmp);
3684: MatSolve(mat,b,x);
3685: VecAXPY(x,one,tmp);
3686: VecDestroy(&tmp);
3687: }
3688: }
3689: PetscLogEventEnd(MAT_SolveAdd,mat,b,x,y);
3690: PetscObjectStateIncrease((PetscObject)x);
3691: return(0);
3692: }
3694: /*@
3695: MatSolveTranspose - Solves A' x = b, given a factored matrix.
3697: Neighbor-wise Collective on Mat
3699: Input Parameters:
3700: + mat - the factored matrix
3701: - b - the right-hand-side vector
3703: Output Parameter:
3704: . x - the result vector
3706: Notes:
3707: The vectors b and x cannot be the same. I.e., one cannot
3708: call MatSolveTranspose(A,x,x).
3710: Most users should employ the simplified KSP interface for linear solvers
3711: instead of working directly with matrix algebra routines such as this.
3712: See, e.g., KSPCreate().
3714: Level: developer
3716: .seealso: MatSolve(), MatSolveAdd(), MatSolveTransposeAdd()
3717: @*/
3718: PetscErrorCode MatSolveTranspose(Mat mat,Vec b,Vec x)
3719: {
3729: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3730: 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);
3731: 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);
3732: if (!mat->rmap->N && !mat->cmap->N) return(0);
3733: MatCheckPreallocated(mat,1);
3734: PetscLogEventBegin(MAT_SolveTranspose,mat,b,x,0);
3735: if (mat->factorerrortype) {
3736: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
3737: VecSetInf(x);
3738: } else {
3739: if (!mat->ops->solvetranspose) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s",((PetscObject)mat)->type_name);
3740: (*mat->ops->solvetranspose)(mat,b,x);
3741: }
3742: PetscLogEventEnd(MAT_SolveTranspose,mat,b,x,0);
3743: PetscObjectStateIncrease((PetscObject)x);
3744: return(0);
3745: }
3747: /*@
3748: MatSolveTransposeAdd - Computes x = y + inv(Transpose(A)) b, given a
3749: factored matrix.
3751: Neighbor-wise Collective on Mat
3753: Input Parameters:
3754: + mat - the factored matrix
3755: . b - the right-hand-side vector
3756: - y - the vector to be added to
3758: Output Parameter:
3759: . x - the result vector
3761: Notes:
3762: The vectors b and x cannot be the same. I.e., one cannot
3763: call MatSolveTransposeAdd(A,x,y,x).
3765: Most users should employ the simplified KSP interface for linear solvers
3766: instead of working directly with matrix algebra routines such as this.
3767: See, e.g., KSPCreate().
3769: Level: developer
3771: .seealso: MatSolve(), MatSolveAdd(), MatSolveTranspose()
3772: @*/
3773: PetscErrorCode MatSolveTransposeAdd(Mat mat,Vec b,Vec y,Vec x)
3774: {
3775: PetscScalar one = 1.0;
3777: Vec tmp;
3788: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
3789: 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);
3790: 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);
3791: 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);
3792: 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);
3793: if (!mat->rmap->N && !mat->cmap->N) return(0);
3794: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
3795: MatCheckPreallocated(mat,1);
3797: PetscLogEventBegin(MAT_SolveTransposeAdd,mat,b,x,y);
3798: if (mat->ops->solvetransposeadd) {
3799: if (mat->factorerrortype) {
3800: PetscInfo1(mat,"MatFactorError %D\n",mat->factorerrortype);
3801: VecSetInf(x);
3802: } else {
3803: (*mat->ops->solvetransposeadd)(mat,b,y,x);
3804: }
3805: } else {
3806: /* do the solve then the add manually */
3807: if (x != y) {
3808: MatSolveTranspose(mat,b,x);
3809: VecAXPY(x,one,y);
3810: } else {
3811: VecDuplicate(x,&tmp);
3812: PetscLogObjectParent((PetscObject)mat,(PetscObject)tmp);
3813: VecCopy(x,tmp);
3814: MatSolveTranspose(mat,b,x);
3815: VecAXPY(x,one,tmp);
3816: VecDestroy(&tmp);
3817: }
3818: }
3819: PetscLogEventEnd(MAT_SolveTransposeAdd,mat,b,x,y);
3820: PetscObjectStateIncrease((PetscObject)x);
3821: return(0);
3822: }
3823: /* ----------------------------------------------------------------*/
3825: /*@
3826: MatSOR - Computes relaxation (SOR, Gauss-Seidel) sweeps.
3828: Neighbor-wise Collective on Mat
3830: Input Parameters:
3831: + mat - the matrix
3832: . b - the right hand side
3833: . omega - the relaxation factor
3834: . flag - flag indicating the type of SOR (see below)
3835: . shift - diagonal shift
3836: . its - the number of iterations
3837: - lits - the number of local iterations
3839: Output Parameters:
3840: . x - the solution (can contain an initial guess, use option SOR_ZERO_INITIAL_GUESS to indicate no guess)
3842: SOR Flags:
3843: + SOR_FORWARD_SWEEP - forward SOR
3844: . SOR_BACKWARD_SWEEP - backward SOR
3845: . SOR_SYMMETRIC_SWEEP - SSOR (symmetric SOR)
3846: . SOR_LOCAL_FORWARD_SWEEP - local forward SOR
3847: . SOR_LOCAL_BACKWARD_SWEEP - local forward SOR
3848: . SOR_LOCAL_SYMMETRIC_SWEEP - local SSOR
3849: . SOR_APPLY_UPPER, SOR_APPLY_LOWER - applies
3850: upper/lower triangular part of matrix to
3851: vector (with omega)
3852: - SOR_ZERO_INITIAL_GUESS - zero initial guess
3854: Notes:
3855: SOR_LOCAL_FORWARD_SWEEP, SOR_LOCAL_BACKWARD_SWEEP, and
3856: SOR_LOCAL_SYMMETRIC_SWEEP perform separate independent smoothings
3857: on each processor.
3859: Application programmers will not generally use MatSOR() directly,
3860: but instead will employ the KSP/PC interface.
3862: Notes:
3863: for BAIJ, SBAIJ, and AIJ matrices with Inodes this does a block SOR smoothing, otherwise it does a pointwise smoothing
3865: Notes for Advanced Users:
3866: The flags are implemented as bitwise inclusive or operations.
3867: For example, use (SOR_ZERO_INITIAL_GUESS | SOR_SYMMETRIC_SWEEP)
3868: to specify a zero initial guess for SSOR.
3870: Most users should employ the simplified KSP interface for linear solvers
3871: instead of working directly with matrix algebra routines such as this.
3872: See, e.g., KSPCreate().
3874: Vectors x and b CANNOT be the same
3876: Developer Note: We should add block SOR support for AIJ matrices with block size set to great than one and no inodes
3878: Level: developer
3880: @*/
3881: PetscErrorCode MatSOR(Mat mat,Vec b,PetscReal omega,MatSORType flag,PetscReal shift,PetscInt its,PetscInt lits,Vec x)
3882: {
3892: if (!mat->ops->sor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
3893: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3894: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3895: 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);
3896: 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);
3897: 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);
3898: if (its <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires global its %D positive",its);
3899: if (lits <= 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"Relaxation requires local its %D positive",lits);
3900: if (b == x) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_IDN,"b and x vector cannot be the same");
3902: MatCheckPreallocated(mat,1);
3903: PetscLogEventBegin(MAT_SOR,mat,b,x,0);
3904: ierr =(*mat->ops->sor)(mat,b,omega,flag,shift,its,lits,x);
3905: PetscLogEventEnd(MAT_SOR,mat,b,x,0);
3906: PetscObjectStateIncrease((PetscObject)x);
3907: return(0);
3908: }
3910: /*
3911: Default matrix copy routine.
3912: */
3913: PetscErrorCode MatCopy_Basic(Mat A,Mat B,MatStructure str)
3914: {
3915: PetscErrorCode ierr;
3916: PetscInt i,rstart = 0,rend = 0,nz;
3917: const PetscInt *cwork;
3918: const PetscScalar *vwork;
3921: if (B->assembled) {
3922: MatZeroEntries(B);
3923: }
3924: if (str == SAME_NONZERO_PATTERN) {
3925: MatGetOwnershipRange(A,&rstart,&rend);
3926: for (i=rstart; i<rend; i++) {
3927: MatGetRow(A,i,&nz,&cwork,&vwork);
3928: MatSetValues(B,1,&i,nz,cwork,vwork,INSERT_VALUES);
3929: MatRestoreRow(A,i,&nz,&cwork,&vwork);
3930: }
3931: } else {
3932: MatAYPX(B,0.0,A,str);
3933: }
3934: MatAssemblyBegin(B,MAT_FINAL_ASSEMBLY);
3935: MatAssemblyEnd(B,MAT_FINAL_ASSEMBLY);
3936: return(0);
3937: }
3939: /*@
3940: MatCopy - Copies a matrix to another matrix.
3942: Collective on Mat
3944: Input Parameters:
3945: + A - the matrix
3946: - str - SAME_NONZERO_PATTERN or DIFFERENT_NONZERO_PATTERN
3948: Output Parameter:
3949: . B - where the copy is put
3951: Notes:
3952: If you use SAME_NONZERO_PATTERN then the two matrices had better have the
3953: same nonzero pattern or the routine will crash.
3955: MatCopy() copies the matrix entries of a matrix to another existing
3956: matrix (after first zeroing the second matrix). A related routine is
3957: MatConvert(), which first creates a new matrix and then copies the data.
3959: Level: intermediate
3961: .seealso: MatConvert(), MatDuplicate()
3963: @*/
3964: PetscErrorCode MatCopy(Mat A,Mat B,MatStructure str)
3965: {
3967: PetscInt i;
3975: MatCheckPreallocated(B,2);
3976: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
3977: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
3978: 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);
3979: MatCheckPreallocated(A,1);
3980: if (A == B) return(0);
3982: PetscLogEventBegin(MAT_Copy,A,B,0,0);
3983: if (A->ops->copy) {
3984: (*A->ops->copy)(A,B,str);
3985: } else { /* generic conversion */
3986: MatCopy_Basic(A,B,str);
3987: }
3989: B->stencil.dim = A->stencil.dim;
3990: B->stencil.noc = A->stencil.noc;
3991: for (i=0; i<=A->stencil.dim; i++) {
3992: B->stencil.dims[i] = A->stencil.dims[i];
3993: B->stencil.starts[i] = A->stencil.starts[i];
3994: }
3996: PetscLogEventEnd(MAT_Copy,A,B,0,0);
3997: PetscObjectStateIncrease((PetscObject)B);
3998: return(0);
3999: }
4001: /*@C
4002: MatConvert - Converts a matrix to another matrix, either of the same
4003: or different type.
4005: Collective on Mat
4007: Input Parameters:
4008: + mat - the matrix
4009: . newtype - new matrix type. Use MATSAME to create a new matrix of the
4010: same type as the original matrix.
4011: - reuse - denotes if the destination matrix is to be created or reused.
4012: 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
4013: 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).
4015: Output Parameter:
4016: . M - pointer to place new matrix
4018: Notes:
4019: MatConvert() first creates a new matrix and then copies the data from
4020: the first matrix. A related routine is MatCopy(), which copies the matrix
4021: entries of one matrix to another already existing matrix context.
4023: Cannot be used to convert a sequential matrix to parallel or parallel to sequential,
4024: the MPI communicator of the generated matrix is always the same as the communicator
4025: of the input matrix.
4027: Level: intermediate
4029: .seealso: MatCopy(), MatDuplicate()
4030: @*/
4031: PetscErrorCode MatConvert(Mat mat, MatType newtype,MatReuse reuse,Mat *M)
4032: {
4034: PetscBool sametype,issame,flg;
4035: char convname[256],mtype[256];
4036: Mat B;
4042: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4043: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4044: MatCheckPreallocated(mat,1);
4046: PetscOptionsGetString(((PetscObject)mat)->options,((PetscObject)mat)->prefix,"-matconvert_type",mtype,256,&flg);
4047: if (flg) newtype = mtype;
4049: PetscObjectTypeCompare((PetscObject)mat,newtype,&sametype);
4050: PetscStrcmp(newtype,"same",&issame);
4051: if ((reuse == MAT_INPLACE_MATRIX) && (mat != *M)) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"MAT_INPLACE_MATRIX requires same input and output matrix");
4052: 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");
4054: if ((reuse == MAT_INPLACE_MATRIX) && (issame || sametype)) {
4055: PetscInfo3(mat,"Early return for inplace %s %d %d\n",((PetscObject)mat)->type_name,sametype,issame);
4056: return(0);
4057: }
4059: if ((sametype || issame) && (reuse==MAT_INITIAL_MATRIX) && mat->ops->duplicate) {
4060: PetscInfo3(mat,"Calling duplicate for initial matrix %s %d %d\n",((PetscObject)mat)->type_name,sametype,issame);
4061: (*mat->ops->duplicate)(mat,MAT_COPY_VALUES,M);
4062: } else {
4063: PetscErrorCode (*conv)(Mat, MatType,MatReuse,Mat*)=NULL;
4064: const char *prefix[3] = {"seq","mpi",""};
4065: PetscInt i;
4066: /*
4067: Order of precedence:
4068: 0) See if newtype is a superclass of the current matrix.
4069: 1) See if a specialized converter is known to the current matrix.
4070: 2) See if a specialized converter is known to the desired matrix class.
4071: 3) See if a good general converter is registered for the desired class
4072: (as of 6/27/03 only MATMPIADJ falls into this category).
4073: 4) See if a good general converter is known for the current matrix.
4074: 5) Use a really basic converter.
4075: */
4077: /* 0) See if newtype is a superclass of the current matrix.
4078: i.e mat is mpiaij and newtype is aij */
4079: for (i=0; i<2; i++) {
4080: PetscStrncpy(convname,prefix[i],sizeof(convname));
4081: PetscStrlcat(convname,newtype,sizeof(convname));
4082: PetscStrcmp(convname,((PetscObject)mat)->type_name,&flg);
4083: PetscInfo3(mat,"Check superclass %s %s -> %d\n",convname,((PetscObject)mat)->type_name,flg);
4084: if (flg) {
4085: if (reuse == MAT_INPLACE_MATRIX) {
4086: return(0);
4087: } else if (reuse == MAT_INITIAL_MATRIX && mat->ops->duplicate) {
4088: (*mat->ops->duplicate)(mat,MAT_COPY_VALUES,M);
4089: return(0);
4090: } else if (reuse == MAT_REUSE_MATRIX && mat->ops->copy) {
4091: MatCopy(mat,*M,SAME_NONZERO_PATTERN);
4092: return(0);
4093: }
4094: }
4095: }
4096: /* 1) See if a specialized converter is known to the current matrix and the desired class */
4097: for (i=0; i<3; i++) {
4098: PetscStrncpy(convname,"MatConvert_",sizeof(convname));
4099: PetscStrlcat(convname,((PetscObject)mat)->type_name,sizeof(convname));
4100: PetscStrlcat(convname,"_",sizeof(convname));
4101: PetscStrlcat(convname,prefix[i],sizeof(convname));
4102: PetscStrlcat(convname,issame ? ((PetscObject)mat)->type_name : newtype,sizeof(convname));
4103: PetscStrlcat(convname,"_C",sizeof(convname));
4104: PetscObjectQueryFunction((PetscObject)mat,convname,&conv);
4105: PetscInfo3(mat,"Check specialized (1) %s (%s) -> %d\n",convname,((PetscObject)mat)->type_name,!!conv);
4106: if (conv) goto foundconv;
4107: }
4109: /* 2) See if a specialized converter is known to the desired matrix class. */
4110: MatCreate(PetscObjectComm((PetscObject)mat),&B);
4111: MatSetSizes(B,mat->rmap->n,mat->cmap->n,mat->rmap->N,mat->cmap->N);
4112: MatSetType(B,newtype);
4113: for (i=0; i<3; i++) {
4114: PetscStrncpy(convname,"MatConvert_",sizeof(convname));
4115: PetscStrlcat(convname,((PetscObject)mat)->type_name,sizeof(convname));
4116: PetscStrlcat(convname,"_",sizeof(convname));
4117: PetscStrlcat(convname,prefix[i],sizeof(convname));
4118: PetscStrlcat(convname,newtype,sizeof(convname));
4119: PetscStrlcat(convname,"_C",sizeof(convname));
4120: PetscObjectQueryFunction((PetscObject)B,convname,&conv);
4121: PetscInfo3(mat,"Check specialized (2) %s (%s) -> %d\n",convname,((PetscObject)B)->type_name,!!conv);
4122: if (conv) {
4123: MatDestroy(&B);
4124: goto foundconv;
4125: }
4126: }
4128: /* 3) See if a good general converter is registered for the desired class */
4129: conv = B->ops->convertfrom;
4130: PetscInfo2(mat,"Check convertfrom (%s) -> %d\n",((PetscObject)B)->type_name,!!conv);
4131: MatDestroy(&B);
4132: if (conv) goto foundconv;
4134: /* 4) See if a good general converter is known for the current matrix */
4135: if (mat->ops->convert) conv = mat->ops->convert;
4137: PetscInfo2(mat,"Check general convert (%s) -> %d\n",((PetscObject)mat)->type_name,!!conv);
4138: if (conv) goto foundconv;
4140: /* 5) Use a really basic converter. */
4141: PetscInfo(mat,"Using MatConvert_Basic\n");
4142: conv = MatConvert_Basic;
4144: foundconv:
4145: PetscLogEventBegin(MAT_Convert,mat,0,0,0);
4146: (*conv)(mat,newtype,reuse,M);
4147: if (mat->rmap->mapping && mat->cmap->mapping && !(*M)->rmap->mapping && !(*M)->cmap->mapping) {
4148: /* the block sizes must be same if the mappings are copied over */
4149: (*M)->rmap->bs = mat->rmap->bs;
4150: (*M)->cmap->bs = mat->cmap->bs;
4151: PetscObjectReference((PetscObject)mat->rmap->mapping);
4152: PetscObjectReference((PetscObject)mat->cmap->mapping);
4153: (*M)->rmap->mapping = mat->rmap->mapping;
4154: (*M)->cmap->mapping = mat->cmap->mapping;
4155: }
4156: (*M)->stencil.dim = mat->stencil.dim;
4157: (*M)->stencil.noc = mat->stencil.noc;
4158: for (i=0; i<=mat->stencil.dim; i++) {
4159: (*M)->stencil.dims[i] = mat->stencil.dims[i];
4160: (*M)->stencil.starts[i] = mat->stencil.starts[i];
4161: }
4162: PetscLogEventEnd(MAT_Convert,mat,0,0,0);
4163: }
4164: PetscObjectStateIncrease((PetscObject)*M);
4166: /* Copy Mat options */
4167: if (mat->symmetric) {MatSetOption(*M,MAT_SYMMETRIC,PETSC_TRUE);}
4168: if (mat->hermitian) {MatSetOption(*M,MAT_HERMITIAN,PETSC_TRUE);}
4169: return(0);
4170: }
4172: /*@C
4173: MatFactorGetSolverType - Returns name of the package providing the factorization routines
4175: Not Collective
4177: Input Parameter:
4178: . mat - the matrix, must be a factored matrix
4180: Output Parameter:
4181: . type - the string name of the package (do not free this string)
4183: Notes:
4184: In Fortran you pass in a empty string and the package name will be copied into it.
4185: (Make sure the string is long enough)
4187: Level: intermediate
4189: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable(), MatGetFactor()
4190: @*/
4191: PetscErrorCode MatFactorGetSolverType(Mat mat, MatSolverType *type)
4192: {
4193: PetscErrorCode ierr, (*conv)(Mat,MatSolverType*);
4198: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Only for factored matrix");
4199: PetscObjectQueryFunction((PetscObject)mat,"MatFactorGetSolverType_C",&conv);
4200: if (!conv) {
4201: *type = MATSOLVERPETSC;
4202: } else {
4203: (*conv)(mat,type);
4204: }
4205: return(0);
4206: }
4208: typedef struct _MatSolverTypeForSpecifcType* MatSolverTypeForSpecifcType;
4209: struct _MatSolverTypeForSpecifcType {
4210: MatType mtype;
4211: PetscErrorCode (*getfactor[4])(Mat,MatFactorType,Mat*);
4212: MatSolverTypeForSpecifcType next;
4213: };
4215: typedef struct _MatSolverTypeHolder* MatSolverTypeHolder;
4216: struct _MatSolverTypeHolder {
4217: char *name;
4218: MatSolverTypeForSpecifcType handlers;
4219: MatSolverTypeHolder next;
4220: };
4222: static MatSolverTypeHolder MatSolverTypeHolders = NULL;
4224: /*@C
4225: MatSolvePackageRegister - Registers a MatSolverType that works for a particular matrix type
4227: Input Parameters:
4228: + package - name of the package, for example petsc or superlu
4229: . mtype - the matrix type that works with this package
4230: . ftype - the type of factorization supported by the package
4231: - getfactor - routine that will create the factored matrix ready to be used
4233: Level: intermediate
4235: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable()
4236: @*/
4237: PetscErrorCode MatSolverTypeRegister(MatSolverType package,MatType mtype,MatFactorType ftype,PetscErrorCode (*getfactor)(Mat,MatFactorType,Mat*))
4238: {
4239: PetscErrorCode ierr;
4240: MatSolverTypeHolder next = MatSolverTypeHolders,prev = NULL;
4241: PetscBool flg;
4242: MatSolverTypeForSpecifcType inext,iprev = NULL;
4245: MatInitializePackage();
4246: if (!next) {
4247: PetscNew(&MatSolverTypeHolders);
4248: PetscStrallocpy(package,&MatSolverTypeHolders->name);
4249: PetscNew(&MatSolverTypeHolders->handlers);
4250: PetscStrallocpy(mtype,(char **)&MatSolverTypeHolders->handlers->mtype);
4251: MatSolverTypeHolders->handlers->getfactor[(int)ftype-1] = getfactor;
4252: return(0);
4253: }
4254: while (next) {
4255: PetscStrcasecmp(package,next->name,&flg);
4256: if (flg) {
4257: if (!next->handlers) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_PLIB,"MatSolverTypeHolder is missing handlers");
4258: inext = next->handlers;
4259: while (inext) {
4260: PetscStrcasecmp(mtype,inext->mtype,&flg);
4261: if (flg) {
4262: inext->getfactor[(int)ftype-1] = getfactor;
4263: return(0);
4264: }
4265: iprev = inext;
4266: inext = inext->next;
4267: }
4268: PetscNew(&iprev->next);
4269: PetscStrallocpy(mtype,(char **)&iprev->next->mtype);
4270: iprev->next->getfactor[(int)ftype-1] = getfactor;
4271: return(0);
4272: }
4273: prev = next;
4274: next = next->next;
4275: }
4276: PetscNew(&prev->next);
4277: PetscStrallocpy(package,&prev->next->name);
4278: PetscNew(&prev->next->handlers);
4279: PetscStrallocpy(mtype,(char **)&prev->next->handlers->mtype);
4280: prev->next->handlers->getfactor[(int)ftype-1] = getfactor;
4281: return(0);
4282: }
4284: /*@C
4285: MatSolvePackageGet - Get's the function that creates the factor matrix if it exist
4287: Input Parameters:
4288: + package - name of the package, for example petsc or superlu
4289: . ftype - the type of factorization supported by the package
4290: - mtype - the matrix type that works with this package
4292: Output Parameters:
4293: + foundpackage - PETSC_TRUE if the package was registered
4294: . foundmtype - PETSC_TRUE if the package supports the requested mtype
4295: - getfactor - routine that will create the factored matrix ready to be used or NULL if not found
4297: Level: intermediate
4299: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable()
4300: @*/
4301: PetscErrorCode MatSolverTypeGet(MatSolverType package,MatType mtype,MatFactorType ftype,PetscBool *foundpackage,PetscBool *foundmtype,PetscErrorCode (**getfactor)(Mat,MatFactorType,Mat*))
4302: {
4303: PetscErrorCode ierr;
4304: MatSolverTypeHolder next = MatSolverTypeHolders;
4305: PetscBool flg;
4306: MatSolverTypeForSpecifcType inext;
4309: if (foundpackage) *foundpackage = PETSC_FALSE;
4310: if (foundmtype) *foundmtype = PETSC_FALSE;
4311: if (getfactor) *getfactor = NULL;
4313: if (package) {
4314: while (next) {
4315: PetscStrcasecmp(package,next->name,&flg);
4316: if (flg) {
4317: if (foundpackage) *foundpackage = PETSC_TRUE;
4318: inext = next->handlers;
4319: while (inext) {
4320: PetscStrbeginswith(mtype,inext->mtype,&flg);
4321: if (flg) {
4322: if (foundmtype) *foundmtype = PETSC_TRUE;
4323: if (getfactor) *getfactor = inext->getfactor[(int)ftype-1];
4324: return(0);
4325: }
4326: inext = inext->next;
4327: }
4328: }
4329: next = next->next;
4330: }
4331: } else {
4332: while (next) {
4333: inext = next->handlers;
4334: while (inext) {
4335: PetscStrbeginswith(mtype,inext->mtype,&flg);
4336: if (flg && inext->getfactor[(int)ftype-1]) {
4337: if (foundpackage) *foundpackage = PETSC_TRUE;
4338: if (foundmtype) *foundmtype = PETSC_TRUE;
4339: if (getfactor) *getfactor = inext->getfactor[(int)ftype-1];
4340: return(0);
4341: }
4342: inext = inext->next;
4343: }
4344: next = next->next;
4345: }
4346: }
4347: return(0);
4348: }
4350: PetscErrorCode MatSolverTypeDestroy(void)
4351: {
4352: PetscErrorCode ierr;
4353: MatSolverTypeHolder next = MatSolverTypeHolders,prev;
4354: MatSolverTypeForSpecifcType inext,iprev;
4357: while (next) {
4358: PetscFree(next->name);
4359: inext = next->handlers;
4360: while (inext) {
4361: PetscFree(inext->mtype);
4362: iprev = inext;
4363: inext = inext->next;
4364: PetscFree(iprev);
4365: }
4366: prev = next;
4367: next = next->next;
4368: PetscFree(prev);
4369: }
4370: MatSolverTypeHolders = NULL;
4371: return(0);
4372: }
4374: /*@C
4375: MatGetFactor - Returns a matrix suitable to calls to MatXXFactorSymbolic()
4377: Collective on Mat
4379: Input Parameters:
4380: + mat - the matrix
4381: . type - name of solver type, for example, superlu, petsc (to use PETSc's default)
4382: - ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
4384: Output Parameters:
4385: . f - the factor matrix used with MatXXFactorSymbolic() calls
4387: Notes:
4388: Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
4389: such as pastix, superlu, mumps etc.
4391: PETSc must have been ./configure to use the external solver, using the option --download-package
4393: Level: intermediate
4395: .seealso: MatCopy(), MatDuplicate(), MatGetFactorAvailable()
4396: @*/
4397: PetscErrorCode MatGetFactor(Mat mat, MatSolverType type,MatFactorType ftype,Mat *f)
4398: {
4399: PetscErrorCode ierr,(*conv)(Mat,MatFactorType,Mat*);
4400: PetscBool foundpackage,foundmtype;
4406: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4407: MatCheckPreallocated(mat,1);
4409: MatSolverTypeGet(type,((PetscObject)mat)->type_name,ftype,&foundpackage,&foundmtype,&conv);
4410: if (!foundpackage) {
4411: if (type) {
4412: SETERRQ4(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"Could not locate solver package %s for factorization type %s and matrix type %s. Perhaps you must ./configure with --download-%s",type,MatFactorTypes[ftype],((PetscObject)mat)->type_name,type);
4413: } else {
4414: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"Could not locate a solver package for factorization type %s and matrix type %s.",MatFactorTypes[ftype],((PetscObject)mat)->type_name);
4415: }
4416: }
4417: if (!foundmtype) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_MISSING_FACTOR,"MatSolverType %s does not support matrix type %s",type,((PetscObject)mat)->type_name);
4418: 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);
4420: #if defined(PETSC_USE_COMPLEX)
4421: if (mat->hermitian && !mat->symmetric && (ftype == MAT_FACTOR_CHOLESKY||ftype == MAT_FACTOR_ICC)) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Hermitian CHOLESKY or ICC Factor is not supported");
4422: #endif
4424: (*conv)(mat,ftype,f);
4425: return(0);
4426: }
4428: /*@C
4429: MatGetFactorAvailable - Returns a a flag if matrix supports particular package and factor type
4431: Not Collective
4433: Input Parameters:
4434: + mat - the matrix
4435: . type - name of solver type, for example, superlu, petsc (to use PETSc's default)
4436: - ftype - factor type, MAT_FACTOR_LU, MAT_FACTOR_CHOLESKY, MAT_FACTOR_ICC, MAT_FACTOR_ILU,
4438: Output Parameter:
4439: . flg - PETSC_TRUE if the factorization is available
4441: Notes:
4442: Some PETSc matrix formats have alternative solvers available that are contained in alternative packages
4443: such as pastix, superlu, mumps etc.
4445: PETSc must have been ./configure to use the external solver, using the option --download-package
4447: Level: intermediate
4449: .seealso: MatCopy(), MatDuplicate(), MatGetFactor()
4450: @*/
4451: PetscErrorCode MatGetFactorAvailable(Mat mat, MatSolverType type,MatFactorType ftype,PetscBool *flg)
4452: {
4453: PetscErrorCode ierr, (*gconv)(Mat,MatFactorType,Mat*);
4459: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4460: MatCheckPreallocated(mat,1);
4462: *flg = PETSC_FALSE;
4463: MatSolverTypeGet(type,((PetscObject)mat)->type_name,ftype,NULL,NULL,&gconv);
4464: if (gconv) {
4465: *flg = PETSC_TRUE;
4466: }
4467: return(0);
4468: }
4470: #include <petscdmtypes.h>
4472: /*@
4473: MatDuplicate - Duplicates a matrix including the non-zero structure.
4475: Collective on Mat
4477: Input Parameters:
4478: + mat - the matrix
4479: - op - One of MAT_DO_NOT_COPY_VALUES, MAT_COPY_VALUES, or MAT_SHARE_NONZERO_PATTERN.
4480: See the manual page for MatDuplicateOption for an explanation of these options.
4482: Output Parameter:
4483: . M - pointer to place new matrix
4485: Level: intermediate
4487: Notes:
4488: You cannot change the nonzero pattern for the parent or child matrix if you use MAT_SHARE_NONZERO_PATTERN.
4489: 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.
4491: .seealso: MatCopy(), MatConvert(), MatDuplicateOption
4492: @*/
4493: PetscErrorCode MatDuplicate(Mat mat,MatDuplicateOption op,Mat *M)
4494: {
4496: Mat B;
4497: PetscInt i;
4498: DM dm;
4499: void (*viewf)(void);
4505: if (op == MAT_COPY_VALUES && !mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"MAT_COPY_VALUES not allowed for unassembled matrix");
4506: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4507: MatCheckPreallocated(mat,1);
4509: *M = 0;
4510: if (!mat->ops->duplicate) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Not written for matrix type %s\n",((PetscObject)mat)->type_name);
4511: PetscLogEventBegin(MAT_Convert,mat,0,0,0);
4512: (*mat->ops->duplicate)(mat,op,M);
4513: B = *M;
4515: MatGetOperation(mat,MATOP_VIEW,&viewf);
4516: if (viewf) {
4517: MatSetOperation(B,MATOP_VIEW,viewf);
4518: }
4520: B->stencil.dim = mat->stencil.dim;
4521: B->stencil.noc = mat->stencil.noc;
4522: for (i=0; i<=mat->stencil.dim; i++) {
4523: B->stencil.dims[i] = mat->stencil.dims[i];
4524: B->stencil.starts[i] = mat->stencil.starts[i];
4525: }
4527: B->nooffproczerorows = mat->nooffproczerorows;
4528: B->nooffprocentries = mat->nooffprocentries;
4530: PetscObjectQuery((PetscObject) mat, "__PETSc_dm", (PetscObject*) &dm);
4531: if (dm) {
4532: PetscObjectCompose((PetscObject) B, "__PETSc_dm", (PetscObject) dm);
4533: }
4534: PetscLogEventEnd(MAT_Convert,mat,0,0,0);
4535: PetscObjectStateIncrease((PetscObject)B);
4536: return(0);
4537: }
4539: /*@
4540: MatGetDiagonal - Gets the diagonal of a matrix.
4542: Logically Collective on Mat
4544: Input Parameters:
4545: + mat - the matrix
4546: - v - the vector for storing the diagonal
4548: Output Parameter:
4549: . v - the diagonal of the matrix
4551: Level: intermediate
4553: Note:
4554: Currently only correct in parallel for square matrices.
4556: .seealso: MatGetRow(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs()
4557: @*/
4558: PetscErrorCode MatGetDiagonal(Mat mat,Vec v)
4559: {
4566: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4567: if (!mat->ops->getdiagonal) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4568: MatCheckPreallocated(mat,1);
4570: (*mat->ops->getdiagonal)(mat,v);
4571: PetscObjectStateIncrease((PetscObject)v);
4572: return(0);
4573: }
4575: /*@C
4576: MatGetRowMin - Gets the minimum value (of the real part) of each
4577: row of the matrix
4579: Logically Collective on Mat
4581: Input Parameters:
4582: . mat - the matrix
4584: Output Parameter:
4585: + v - the vector for storing the maximums
4586: - idx - the indices of the column found for each row (optional)
4588: Level: intermediate
4590: Notes:
4591: The result of this call are the same as if one converted the matrix to dense format
4592: and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
4594: This code is only implemented for a couple of matrix formats.
4596: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs(),
4597: MatGetRowMax()
4598: @*/
4599: PetscErrorCode MatGetRowMin(Mat mat,Vec v,PetscInt idx[])
4600: {
4607: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4608: if (!mat->ops->getrowmax) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4609: MatCheckPreallocated(mat,1);
4611: (*mat->ops->getrowmin)(mat,v,idx);
4612: PetscObjectStateIncrease((PetscObject)v);
4613: return(0);
4614: }
4616: /*@C
4617: MatGetRowMinAbs - Gets the minimum value (in absolute value) of each
4618: row of the matrix
4620: Logically Collective on Mat
4622: Input Parameters:
4623: . mat - the matrix
4625: Output Parameter:
4626: + v - the vector for storing the minimums
4627: - idx - the indices of the column found for each row (or NULL if not needed)
4629: Level: intermediate
4631: Notes:
4632: if a row is completely empty or has only 0.0 values then the idx[] value for that
4633: row is 0 (the first column).
4635: This code is only implemented for a couple of matrix formats.
4637: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMaxAbs(), MatGetRowMin()
4638: @*/
4639: PetscErrorCode MatGetRowMinAbs(Mat mat,Vec v,PetscInt idx[])
4640: {
4647: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4648: if (!mat->ops->getrowminabs) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4649: MatCheckPreallocated(mat,1);
4650: if (idx) {PetscArrayzero(idx,mat->rmap->n);}
4652: (*mat->ops->getrowminabs)(mat,v,idx);
4653: PetscObjectStateIncrease((PetscObject)v);
4654: return(0);
4655: }
4657: /*@C
4658: MatGetRowMax - Gets the maximum value (of the real part) of each
4659: row of the matrix
4661: Logically Collective on Mat
4663: Input Parameters:
4664: . mat - the matrix
4666: Output Parameter:
4667: + v - the vector for storing the maximums
4668: - idx - the indices of the column found for each row (optional)
4670: Level: intermediate
4672: Notes:
4673: The result of this call are the same as if one converted the matrix to dense format
4674: and found the minimum value in each row (i.e. the implicit zeros are counted as zeros).
4676: This code is only implemented for a couple of matrix formats.
4678: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMaxAbs(), MatGetRowMin()
4679: @*/
4680: PetscErrorCode MatGetRowMax(Mat mat,Vec v,PetscInt idx[])
4681: {
4688: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4689: if (!mat->ops->getrowmax) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4690: MatCheckPreallocated(mat,1);
4692: (*mat->ops->getrowmax)(mat,v,idx);
4693: PetscObjectStateIncrease((PetscObject)v);
4694: return(0);
4695: }
4697: /*@C
4698: MatGetRowMaxAbs - Gets the maximum value (in absolute value) of each
4699: row of the matrix
4701: Logically Collective on Mat
4703: Input Parameters:
4704: . mat - the matrix
4706: Output Parameter:
4707: + v - the vector for storing the maximums
4708: - idx - the indices of the column found for each row (or NULL if not needed)
4710: Level: intermediate
4712: Notes:
4713: if a row is completely empty or has only 0.0 values then the idx[] value for that
4714: row is 0 (the first column).
4716: This code is only implemented for a couple of matrix formats.
4718: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMin()
4719: @*/
4720: PetscErrorCode MatGetRowMaxAbs(Mat mat,Vec v,PetscInt idx[])
4721: {
4728: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4729: if (!mat->ops->getrowmaxabs) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4730: MatCheckPreallocated(mat,1);
4731: if (idx) {PetscArrayzero(idx,mat->rmap->n);}
4733: (*mat->ops->getrowmaxabs)(mat,v,idx);
4734: PetscObjectStateIncrease((PetscObject)v);
4735: return(0);
4736: }
4738: /*@
4739: MatGetRowSum - Gets the sum of each row of the matrix
4741: Logically or Neighborhood Collective on Mat
4743: Input Parameters:
4744: . mat - the matrix
4746: Output Parameter:
4747: . v - the vector for storing the sum of rows
4749: Level: intermediate
4751: Notes:
4752: This code is slow since it is not currently specialized for different formats
4754: .seealso: MatGetDiagonal(), MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRowMax(), MatGetRowMin()
4755: @*/
4756: PetscErrorCode MatGetRowSum(Mat mat, Vec v)
4757: {
4758: Vec ones;
4765: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4766: MatCheckPreallocated(mat,1);
4767: MatCreateVecs(mat,&ones,NULL);
4768: VecSet(ones,1.);
4769: MatMult(mat,ones,v);
4770: VecDestroy(&ones);
4771: return(0);
4772: }
4774: /*@
4775: MatTranspose - Computes an in-place or out-of-place transpose of a matrix.
4777: Collective on Mat
4779: Input Parameter:
4780: + mat - the matrix to transpose
4781: - reuse - either MAT_INITIAL_MATRIX, MAT_REUSE_MATRIX, or MAT_INPLACE_MATRIX
4783: Output Parameters:
4784: . B - the transpose
4786: Notes:
4787: If you use MAT_INPLACE_MATRIX then you must pass in &mat for B
4789: MAT_REUSE_MATRIX causes the B matrix from a previous call to this function with MAT_INITIAL_MATRIX to be used
4791: Consider using MatCreateTranspose() instead if you only need a matrix that behaves like the transpose, but don't need the storage to be changed.
4793: Level: intermediate
4795: .seealso: MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
4796: @*/
4797: PetscErrorCode MatTranspose(Mat mat,MatReuse reuse,Mat *B)
4798: {
4804: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4805: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4806: if (!mat->ops->transpose) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
4807: if (reuse == MAT_INPLACE_MATRIX && mat != *B) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"MAT_INPLACE_MATRIX requires last matrix to match first");
4808: if (reuse == MAT_REUSE_MATRIX && mat == *B) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Perhaps you mean MAT_INPLACE_MATRIX");
4809: MatCheckPreallocated(mat,1);
4811: PetscLogEventBegin(MAT_Transpose,mat,0,0,0);
4812: (*mat->ops->transpose)(mat,reuse,B);
4813: PetscLogEventEnd(MAT_Transpose,mat,0,0,0);
4814: if (B) {PetscObjectStateIncrease((PetscObject)*B);}
4815: return(0);
4816: }
4818: /*@
4819: MatIsTranspose - Test whether a matrix is another one's transpose,
4820: or its own, in which case it tests symmetry.
4822: Collective on Mat
4824: Input Parameter:
4825: + A - the matrix to test
4826: - B - the matrix to test against, this can equal the first parameter
4828: Output Parameters:
4829: . flg - the result
4831: Notes:
4832: Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
4833: has a running time of the order of the number of nonzeros; the parallel
4834: test involves parallel copies of the block-offdiagonal parts of the matrix.
4836: Level: intermediate
4838: .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian()
4839: @*/
4840: PetscErrorCode MatIsTranspose(Mat A,Mat B,PetscReal tol,PetscBool *flg)
4841: {
4842: PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscBool*),(*g)(Mat,Mat,PetscReal,PetscBool*);
4848: PetscObjectQueryFunction((PetscObject)A,"MatIsTranspose_C",&f);
4849: PetscObjectQueryFunction((PetscObject)B,"MatIsTranspose_C",&g);
4850: *flg = PETSC_FALSE;
4851: if (f && g) {
4852: if (f == g) {
4853: (*f)(A,B,tol,flg);
4854: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for symmetry test");
4855: } else {
4856: MatType mattype;
4857: if (!f) {
4858: MatGetType(A,&mattype);
4859: } else {
4860: MatGetType(B,&mattype);
4861: }
4862: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for transpose",mattype);
4863: }
4864: return(0);
4865: }
4867: /*@
4868: MatHermitianTranspose - Computes an in-place or out-of-place transpose of a matrix in complex conjugate.
4870: Collective on Mat
4872: Input Parameter:
4873: + mat - the matrix to transpose and complex conjugate
4874: - reuse - MAT_INITIAL_MATRIX to create a new matrix, MAT_INPLACE_MATRIX to reuse the first argument to store the transpose
4876: Output Parameters:
4877: . B - the Hermitian
4879: Level: intermediate
4881: .seealso: MatTranspose(), MatMultTranspose(), MatMultTransposeAdd(), MatIsTranspose(), MatReuse
4882: @*/
4883: PetscErrorCode MatHermitianTranspose(Mat mat,MatReuse reuse,Mat *B)
4884: {
4888: MatTranspose(mat,reuse,B);
4889: #if defined(PETSC_USE_COMPLEX)
4890: MatConjugate(*B);
4891: #endif
4892: return(0);
4893: }
4895: /*@
4896: MatIsHermitianTranspose - Test whether a matrix is another one's Hermitian transpose,
4898: Collective on Mat
4900: Input Parameter:
4901: + A - the matrix to test
4902: - B - the matrix to test against, this can equal the first parameter
4904: Output Parameters:
4905: . flg - the result
4907: Notes:
4908: Only available for SeqAIJ/MPIAIJ matrices. The sequential algorithm
4909: has a running time of the order of the number of nonzeros; the parallel
4910: test involves parallel copies of the block-offdiagonal parts of the matrix.
4912: Level: intermediate
4914: .seealso: MatTranspose(), MatIsSymmetric(), MatIsHermitian(), MatIsTranspose()
4915: @*/
4916: PetscErrorCode MatIsHermitianTranspose(Mat A,Mat B,PetscReal tol,PetscBool *flg)
4917: {
4918: PetscErrorCode ierr,(*f)(Mat,Mat,PetscReal,PetscBool*),(*g)(Mat,Mat,PetscReal,PetscBool*);
4924: PetscObjectQueryFunction((PetscObject)A,"MatIsHermitianTranspose_C",&f);
4925: PetscObjectQueryFunction((PetscObject)B,"MatIsHermitianTranspose_C",&g);
4926: if (f && g) {
4927: if (f==g) {
4928: (*f)(A,B,tol,flg);
4929: } else SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_NOTSAMETYPE,"Matrices do not have the same comparator for Hermitian test");
4930: }
4931: return(0);
4932: }
4934: /*@
4935: MatPermute - Creates a new matrix with rows and columns permuted from the
4936: original.
4938: Collective on Mat
4940: Input Parameters:
4941: + mat - the matrix to permute
4942: . row - row permutation, each processor supplies only the permutation for its rows
4943: - col - column permutation, each processor supplies only the permutation for its columns
4945: Output Parameters:
4946: . B - the permuted matrix
4948: Level: advanced
4950: Note:
4951: The index sets map from row/col of permuted matrix to row/col of original matrix.
4952: The index sets should be on the same communicator as Mat and have the same local sizes.
4954: .seealso: MatGetOrdering(), ISAllGather()
4956: @*/
4957: PetscErrorCode MatPermute(Mat mat,IS row,IS col,Mat *B)
4958: {
4967: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
4968: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
4969: if (!mat->ops->permute) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"MatPermute not available for Mat type %s",((PetscObject)mat)->type_name);
4970: MatCheckPreallocated(mat,1);
4972: (*mat->ops->permute)(mat,row,col,B);
4973: PetscObjectStateIncrease((PetscObject)*B);
4974: return(0);
4975: }
4977: /*@
4978: MatEqual - Compares two matrices.
4980: Collective on Mat
4982: Input Parameters:
4983: + A - the first matrix
4984: - B - the second matrix
4986: Output Parameter:
4987: . flg - PETSC_TRUE if the matrices are equal; PETSC_FALSE otherwise.
4989: Level: intermediate
4991: @*/
4992: PetscErrorCode MatEqual(Mat A,Mat B,PetscBool *flg)
4993: {
5003: MatCheckPreallocated(B,2);
5004: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5005: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5006: 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);
5007: if (!A->ops->equal) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)A)->type_name);
5008: if (!B->ops->equal) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Mat type %s",((PetscObject)B)->type_name);
5009: 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);
5010: MatCheckPreallocated(A,1);
5012: (*A->ops->equal)(A,B,flg);
5013: return(0);
5014: }
5016: /*@
5017: MatDiagonalScale - Scales a matrix on the left and right by diagonal
5018: matrices that are stored as vectors. Either of the two scaling
5019: matrices can be NULL.
5021: Collective on Mat
5023: Input Parameters:
5024: + mat - the matrix to be scaled
5025: . l - the left scaling vector (or NULL)
5026: - r - the right scaling vector (or NULL)
5028: Notes:
5029: MatDiagonalScale() computes A = LAR, where
5030: L = a diagonal matrix (stored as a vector), R = a diagonal matrix (stored as a vector)
5031: The L scales the rows of the matrix, the R scales the columns of the matrix.
5033: Level: intermediate
5036: .seealso: MatScale(), MatShift(), MatDiagonalSet()
5037: @*/
5038: PetscErrorCode MatDiagonalScale(Mat mat,Vec l,Vec r)
5039: {
5045: if (!mat->ops->diagonalscale) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5048: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5049: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5050: MatCheckPreallocated(mat,1);
5052: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
5053: (*mat->ops->diagonalscale)(mat,l,r);
5054: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
5055: PetscObjectStateIncrease((PetscObject)mat);
5056: return(0);
5057: }
5059: /*@
5060: MatScale - Scales all elements of a matrix by a given number.
5062: Logically Collective on Mat
5064: Input Parameters:
5065: + mat - the matrix to be scaled
5066: - a - the scaling value
5068: Output Parameter:
5069: . mat - the scaled matrix
5071: Level: intermediate
5073: .seealso: MatDiagonalScale()
5074: @*/
5075: PetscErrorCode MatScale(Mat mat,PetscScalar a)
5076: {
5082: if (a != (PetscScalar)1.0 && !mat->ops->scale) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5083: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5084: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5086: MatCheckPreallocated(mat,1);
5088: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
5089: if (a != (PetscScalar)1.0) {
5090: (*mat->ops->scale)(mat,a);
5091: PetscObjectStateIncrease((PetscObject)mat);
5092: }
5093: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
5094: return(0);
5095: }
5097: /*@
5098: MatNorm - Calculates various norms of a matrix.
5100: Collective on Mat
5102: Input Parameters:
5103: + mat - the matrix
5104: - type - the type of norm, NORM_1, NORM_FROBENIUS, NORM_INFINITY
5106: Output Parameters:
5107: . nrm - the resulting norm
5109: Level: intermediate
5111: @*/
5112: PetscErrorCode MatNorm(Mat mat,NormType type,PetscReal *nrm)
5113: {
5121: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5122: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5123: if (!mat->ops->norm) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5124: MatCheckPreallocated(mat,1);
5126: (*mat->ops->norm)(mat,type,nrm);
5127: return(0);
5128: }
5130: /*
5131: This variable is used to prevent counting of MatAssemblyBegin() that
5132: are called from within a MatAssemblyEnd().
5133: */
5134: static PetscInt MatAssemblyEnd_InUse = 0;
5135: /*@
5136: MatAssemblyBegin - Begins assembling the matrix. This routine should
5137: be called after completing all calls to MatSetValues().
5139: Collective on Mat
5141: Input Parameters:
5142: + mat - the matrix
5143: - type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
5145: Notes:
5146: MatSetValues() generally caches the values. The matrix is ready to
5147: use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
5148: Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
5149: in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
5150: using the matrix.
5152: ALL processes that share a matrix MUST call MatAssemblyBegin() and MatAssemblyEnd() the SAME NUMBER of times, and each time with the
5153: 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
5154: a global collective operation requring all processes that share the matrix.
5156: Space for preallocated nonzeros that is not filled by a call to MatSetValues() or a related routine are compressed
5157: out by assembly. If you intend to use that extra space on a subsequent assembly, be sure to insert explicit zeros
5158: before MAT_FINAL_ASSEMBLY so the space is not compressed out.
5160: Level: beginner
5162: .seealso: MatAssemblyEnd(), MatSetValues(), MatAssembled()
5163: @*/
5164: PetscErrorCode MatAssemblyBegin(Mat mat,MatAssemblyType type)
5165: {
5171: MatCheckPreallocated(mat,1);
5172: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix.\nDid you forget to call MatSetUnfactored()?");
5173: if (mat->assembled) {
5174: mat->was_assembled = PETSC_TRUE;
5175: mat->assembled = PETSC_FALSE;
5176: }
5178: if (!MatAssemblyEnd_InUse) {
5179: PetscLogEventBegin(MAT_AssemblyBegin,mat,0,0,0);
5180: if (mat->ops->assemblybegin) {(*mat->ops->assemblybegin)(mat,type);}
5181: PetscLogEventEnd(MAT_AssemblyBegin,mat,0,0,0);
5182: } else if (mat->ops->assemblybegin) {
5183: (*mat->ops->assemblybegin)(mat,type);
5184: }
5185: return(0);
5186: }
5188: /*@
5189: MatAssembled - Indicates if a matrix has been assembled and is ready for
5190: use; for example, in matrix-vector product.
5192: Not Collective
5194: Input Parameter:
5195: . mat - the matrix
5197: Output Parameter:
5198: . assembled - PETSC_TRUE or PETSC_FALSE
5200: Level: advanced
5202: .seealso: MatAssemblyEnd(), MatSetValues(), MatAssemblyBegin()
5203: @*/
5204: PetscErrorCode MatAssembled(Mat mat,PetscBool *assembled)
5205: {
5209: *assembled = mat->assembled;
5210: return(0);
5211: }
5213: /*@
5214: MatAssemblyEnd - Completes assembling the matrix. This routine should
5215: be called after MatAssemblyBegin().
5217: Collective on Mat
5219: Input Parameters:
5220: + mat - the matrix
5221: - type - type of assembly, either MAT_FLUSH_ASSEMBLY or MAT_FINAL_ASSEMBLY
5223: Options Database Keys:
5224: + -mat_view ::ascii_info - Prints info on matrix at conclusion of MatEndAssembly()
5225: . -mat_view ::ascii_info_detail - Prints more detailed info
5226: . -mat_view - Prints matrix in ASCII format
5227: . -mat_view ::ascii_matlab - Prints matrix in Matlab format
5228: . -mat_view draw - PetscDraws nonzero structure of matrix, using MatView() and PetscDrawOpenX().
5229: . -display <name> - Sets display name (default is host)
5230: . -draw_pause <sec> - Sets number of seconds to pause after display
5231: . -mat_view socket - Sends matrix to socket, can be accessed from Matlab (See Users-Manual: Chapter 12 Using MATLAB with PETSc )
5232: . -viewer_socket_machine <machine> - Machine to use for socket
5233: . -viewer_socket_port <port> - Port number to use for socket
5234: - -mat_view binary:filename[:append] - Save matrix to file in binary format
5236: Notes:
5237: MatSetValues() generally caches the values. The matrix is ready to
5238: use only after MatAssemblyBegin() and MatAssemblyEnd() have been called.
5239: Use MAT_FLUSH_ASSEMBLY when switching between ADD_VALUES and INSERT_VALUES
5240: in MatSetValues(); use MAT_FINAL_ASSEMBLY for the final assembly before
5241: using the matrix.
5243: Space for preallocated nonzeros that is not filled by a call to MatSetValues() or a related routine are compressed
5244: out by assembly. If you intend to use that extra space on a subsequent assembly, be sure to insert explicit zeros
5245: before MAT_FINAL_ASSEMBLY so the space is not compressed out.
5247: Level: beginner
5249: .seealso: MatAssemblyBegin(), MatSetValues(), PetscDrawOpenX(), PetscDrawCreate(), MatView(), MatAssembled(), PetscViewerSocketOpen()
5250: @*/
5251: PetscErrorCode MatAssemblyEnd(Mat mat,MatAssemblyType type)
5252: {
5253: PetscErrorCode ierr;
5254: static PetscInt inassm = 0;
5255: PetscBool flg = PETSC_FALSE;
5261: inassm++;
5262: MatAssemblyEnd_InUse++;
5263: if (MatAssemblyEnd_InUse == 1) { /* Do the logging only the first time through */
5264: PetscLogEventBegin(MAT_AssemblyEnd,mat,0,0,0);
5265: if (mat->ops->assemblyend) {
5266: (*mat->ops->assemblyend)(mat,type);
5267: }
5268: PetscLogEventEnd(MAT_AssemblyEnd,mat,0,0,0);
5269: } else if (mat->ops->assemblyend) {
5270: (*mat->ops->assemblyend)(mat,type);
5271: }
5273: /* Flush assembly is not a true assembly */
5274: if (type != MAT_FLUSH_ASSEMBLY) {
5275: mat->num_ass++;
5276: mat->assembled = PETSC_TRUE;
5277: mat->ass_nonzerostate = mat->nonzerostate;
5278: }
5280: mat->insertmode = NOT_SET_VALUES;
5281: MatAssemblyEnd_InUse--;
5282: PetscObjectStateIncrease((PetscObject)mat);
5283: if (!mat->symmetric_eternal) {
5284: mat->symmetric_set = PETSC_FALSE;
5285: mat->hermitian_set = PETSC_FALSE;
5286: mat->structurally_symmetric_set = PETSC_FALSE;
5287: }
5288: if (inassm == 1 && type != MAT_FLUSH_ASSEMBLY) {
5289: MatViewFromOptions(mat,NULL,"-mat_view");
5291: if (mat->checksymmetryonassembly) {
5292: MatIsSymmetric(mat,mat->checksymmetrytol,&flg);
5293: if (flg) {
5294: PetscPrintf(PetscObjectComm((PetscObject)mat),"Matrix is symmetric (tolerance %g)\n",(double)mat->checksymmetrytol);
5295: } else {
5296: PetscPrintf(PetscObjectComm((PetscObject)mat),"Matrix is not symmetric (tolerance %g)\n",(double)mat->checksymmetrytol);
5297: }
5298: }
5299: if (mat->nullsp && mat->checknullspaceonassembly) {
5300: MatNullSpaceTest(mat->nullsp,mat,NULL);
5301: }
5302: }
5303: inassm--;
5304: return(0);
5305: }
5307: /*@
5308: MatSetOption - Sets a parameter option for a matrix. Some options
5309: may be specific to certain storage formats. Some options
5310: determine how values will be inserted (or added). Sorted,
5311: row-oriented input will generally assemble the fastest. The default
5312: is row-oriented.
5314: Logically Collective on Mat for certain operations, such as MAT_SPD, not collective for MAT_ROW_ORIENTED, see MatOption
5316: Input Parameters:
5317: + mat - the matrix
5318: . option - the option, one of those listed below (and possibly others),
5319: - flg - turn the option on (PETSC_TRUE) or off (PETSC_FALSE)
5321: Options Describing Matrix Structure:
5322: + MAT_SPD - symmetric positive definite
5323: . MAT_SYMMETRIC - symmetric in terms of both structure and value
5324: . MAT_HERMITIAN - transpose is the complex conjugation
5325: . MAT_STRUCTURALLY_SYMMETRIC - symmetric nonzero structure
5326: - MAT_SYMMETRY_ETERNAL - if you would like the symmetry/Hermitian flag
5327: you set to be kept with all future use of the matrix
5328: including after MatAssemblyBegin/End() which could
5329: potentially change the symmetry structure, i.e. you
5330: KNOW the matrix will ALWAYS have the property you set.
5333: Options For Use with MatSetValues():
5334: Insert a logically dense subblock, which can be
5335: . MAT_ROW_ORIENTED - row-oriented (default)
5337: Note these options reflect the data you pass in with MatSetValues(); it has
5338: nothing to do with how the data is stored internally in the matrix
5339: data structure.
5341: When (re)assembling a matrix, we can restrict the input for
5342: efficiency/debugging purposes. These options include:
5343: + MAT_NEW_NONZERO_LOCATIONS - additional insertions will be allowed if they generate a new nonzero (slow)
5344: . MAT_NEW_DIAGONALS - new diagonals will be allowed (for block diagonal format only)
5345: . MAT_IGNORE_OFF_PROC_ENTRIES - drops off-processor entries
5346: . MAT_NEW_NONZERO_LOCATION_ERR - generates an error for new matrix entry
5347: . MAT_USE_HASH_TABLE - uses a hash table to speed up matrix assembly
5348: . MAT_NO_OFF_PROC_ENTRIES - you know each process will only set values for its own rows, will generate an error if
5349: any process sets values for another process. This avoids all reductions in the MatAssembly routines and thus improves
5350: performance for very large process counts.
5351: - MAT_SUBSET_OFF_PROC_ENTRIES - you know that the first assembly after setting this flag will set a superset
5352: of the off-process entries required for all subsequent assemblies. This avoids a rendezvous step in the MatAssembly
5353: functions, instead sending only neighbor messages.
5355: Notes:
5356: Except for MAT_UNUSED_NONZERO_LOCATION_ERR and MAT_ROW_ORIENTED all processes that share the matrix must pass the same value in flg!
5358: Some options are relevant only for particular matrix types and
5359: are thus ignored by others. Other options are not supported by
5360: certain matrix types and will generate an error message if set.
5362: If using a Fortran 77 module to compute a matrix, one may need to
5363: use the column-oriented option (or convert to the row-oriented
5364: format).
5366: MAT_NEW_NONZERO_LOCATIONS set to PETSC_FALSE indicates that any add or insertion
5367: that would generate a new entry in the nonzero structure is instead
5368: ignored. Thus, if memory has not alredy been allocated for this particular
5369: data, then the insertion is ignored. For dense matrices, in which
5370: the entire array is allocated, no entries are ever ignored.
5371: Set after the first MatAssemblyEnd(). If this option is set then the MatAssemblyBegin/End() processes has one less global reduction
5373: MAT_NEW_NONZERO_LOCATION_ERR set to PETSC_TRUE indicates that any add or insertion
5374: that would generate a new entry in the nonzero structure instead produces
5375: 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
5377: MAT_NEW_NONZERO_ALLOCATION_ERR set to PETSC_TRUE indicates that any add or insertion
5378: that would generate a new entry that has not been preallocated will
5379: instead produce an error. (Currently supported for AIJ and BAIJ formats
5380: only.) This is a useful flag when debugging matrix memory preallocation.
5381: If this option is set then the MatAssemblyBegin/End() processes has one less global reduction
5383: MAT_IGNORE_OFF_PROC_ENTRIES set to PETSC_TRUE indicates entries destined for
5384: other processors should be dropped, rather than stashed.
5385: This is useful if you know that the "owning" processor is also
5386: always generating the correct matrix entries, so that PETSc need
5387: not transfer duplicate entries generated on another processor.
5389: MAT_USE_HASH_TABLE indicates that a hash table be used to improve the
5390: searches during matrix assembly. When this flag is set, the hash table
5391: is created during the first Matrix Assembly. This hash table is
5392: used the next time through, during MatSetVaules()/MatSetVaulesBlocked()
5393: to improve the searching of indices. MAT_NEW_NONZERO_LOCATIONS flag
5394: should be used with MAT_USE_HASH_TABLE flag. This option is currently
5395: supported by MATMPIBAIJ format only.
5397: MAT_KEEP_NONZERO_PATTERN indicates when MatZeroRows() is called the zeroed entries
5398: are kept in the nonzero structure
5400: MAT_IGNORE_ZERO_ENTRIES - for AIJ/IS matrices this will stop zero values from creating
5401: a zero location in the matrix
5403: MAT_USE_INODES - indicates using inode version of the code - works with AIJ matrix types
5405: MAT_NO_OFF_PROC_ZERO_ROWS - you know each process will only zero its own rows. This avoids all reductions in the
5406: zero row routines and thus improves performance for very large process counts.
5408: MAT_IGNORE_LOWER_TRIANGULAR - For SBAIJ matrices will ignore any insertions you make in the lower triangular
5409: part of the matrix (since they should match the upper triangular part).
5411: MAT_SORTED_FULL - each process provides exactly its local rows; all column indices for a given row are passed in a
5412: single call to MatSetValues(), preallocation is perfect, row oriented, INSERT_VALUES is used. Common
5413: with finite difference schemes with non-periodic boundary conditions.
5414: Notes:
5415: Can only be called after MatSetSizes() and MatSetType() have been set.
5417: Level: intermediate
5419: .seealso: MatOption, Mat
5421: @*/
5422: PetscErrorCode MatSetOption(Mat mat,MatOption op,PetscBool flg)
5423: {
5429: if (op > 0) {
5432: }
5434: 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);
5435: if (!((PetscObject)mat)->type_name) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_TYPENOTSET,"Cannot set options until type and size have been set, see MatSetType() and MatSetSizes()");
5437: switch (op) {
5438: case MAT_NO_OFF_PROC_ENTRIES:
5439: mat->nooffprocentries = flg;
5440: return(0);
5441: break;
5442: case MAT_SUBSET_OFF_PROC_ENTRIES:
5443: mat->assembly_subset = flg;
5444: if (!mat->assembly_subset) { /* See the same logic in VecAssembly wrt VEC_SUBSET_OFF_PROC_ENTRIES */
5445: #if !defined(PETSC_HAVE_MPIUNI)
5446: MatStashScatterDestroy_BTS(&mat->stash);
5447: #endif
5448: mat->stash.first_assembly_done = PETSC_FALSE;
5449: }
5450: return(0);
5451: case MAT_NO_OFF_PROC_ZERO_ROWS:
5452: mat->nooffproczerorows = flg;
5453: return(0);
5454: break;
5455: case MAT_SPD:
5456: mat->spd_set = PETSC_TRUE;
5457: mat->spd = flg;
5458: if (flg) {
5459: mat->symmetric = PETSC_TRUE;
5460: mat->structurally_symmetric = PETSC_TRUE;
5461: mat->symmetric_set = PETSC_TRUE;
5462: mat->structurally_symmetric_set = PETSC_TRUE;
5463: }
5464: break;
5465: case MAT_SYMMETRIC:
5466: mat->symmetric = flg;
5467: if (flg) mat->structurally_symmetric = PETSC_TRUE;
5468: mat->symmetric_set = PETSC_TRUE;
5469: mat->structurally_symmetric_set = flg;
5470: #if !defined(PETSC_USE_COMPLEX)
5471: mat->hermitian = flg;
5472: mat->hermitian_set = PETSC_TRUE;
5473: #endif
5474: break;
5475: case MAT_HERMITIAN:
5476: mat->hermitian = flg;
5477: if (flg) mat->structurally_symmetric = PETSC_TRUE;
5478: mat->hermitian_set = PETSC_TRUE;
5479: mat->structurally_symmetric_set = flg;
5480: #if !defined(PETSC_USE_COMPLEX)
5481: mat->symmetric = flg;
5482: mat->symmetric_set = PETSC_TRUE;
5483: #endif
5484: break;
5485: case MAT_STRUCTURALLY_SYMMETRIC:
5486: mat->structurally_symmetric = flg;
5487: mat->structurally_symmetric_set = PETSC_TRUE;
5488: break;
5489: case MAT_SYMMETRY_ETERNAL:
5490: mat->symmetric_eternal = flg;
5491: break;
5492: case MAT_STRUCTURE_ONLY:
5493: mat->structure_only = flg;
5494: break;
5495: case MAT_SORTED_FULL:
5496: mat->sortedfull = flg;
5497: break;
5498: default:
5499: break;
5500: }
5501: if (mat->ops->setoption) {
5502: (*mat->ops->setoption)(mat,op,flg);
5503: }
5504: return(0);
5505: }
5507: /*@
5508: MatGetOption - Gets a parameter option that has been set for a matrix.
5510: Logically Collective on Mat for certain operations, such as MAT_SPD, not collective for MAT_ROW_ORIENTED, see MatOption
5512: Input Parameters:
5513: + mat - the matrix
5514: - option - the option, this only responds to certain options, check the code for which ones
5516: Output Parameter:
5517: . flg - turn the option on (PETSC_TRUE) or off (PETSC_FALSE)
5519: Notes:
5520: Can only be called after MatSetSizes() and MatSetType() have been set.
5522: Level: intermediate
5524: .seealso: MatOption, MatSetOption()
5526: @*/
5527: PetscErrorCode MatGetOption(Mat mat,MatOption op,PetscBool *flg)
5528: {
5533: 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);
5534: 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()");
5536: switch (op) {
5537: case MAT_NO_OFF_PROC_ENTRIES:
5538: *flg = mat->nooffprocentries;
5539: break;
5540: case MAT_NO_OFF_PROC_ZERO_ROWS:
5541: *flg = mat->nooffproczerorows;
5542: break;
5543: case MAT_SYMMETRIC:
5544: *flg = mat->symmetric;
5545: break;
5546: case MAT_HERMITIAN:
5547: *flg = mat->hermitian;
5548: break;
5549: case MAT_STRUCTURALLY_SYMMETRIC:
5550: *flg = mat->structurally_symmetric;
5551: break;
5552: case MAT_SYMMETRY_ETERNAL:
5553: *flg = mat->symmetric_eternal;
5554: break;
5555: case MAT_SPD:
5556: *flg = mat->spd;
5557: break;
5558: default:
5559: break;
5560: }
5561: return(0);
5562: }
5564: /*@
5565: MatZeroEntries - Zeros all entries of a matrix. For sparse matrices
5566: this routine retains the old nonzero structure.
5568: Logically Collective on Mat
5570: Input Parameters:
5571: . mat - the matrix
5573: Level: intermediate
5575: Notes:
5576: 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.
5577: See the Performance chapter of the users manual for information on preallocating matrices.
5579: .seealso: MatZeroRows()
5580: @*/
5581: PetscErrorCode MatZeroEntries(Mat mat)
5582: {
5588: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5589: 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");
5590: if (!mat->ops->zeroentries) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5591: MatCheckPreallocated(mat,1);
5593: PetscLogEventBegin(MAT_ZeroEntries,mat,0,0,0);
5594: (*mat->ops->zeroentries)(mat);
5595: PetscLogEventEnd(MAT_ZeroEntries,mat,0,0,0);
5596: PetscObjectStateIncrease((PetscObject)mat);
5597: return(0);
5598: }
5600: /*@
5601: MatZeroRowsColumns - Zeros all entries (except possibly the main diagonal)
5602: of a set of rows and columns of a matrix.
5604: Collective on Mat
5606: Input Parameters:
5607: + mat - the matrix
5608: . numRows - the number of rows to remove
5609: . rows - the global row indices
5610: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5611: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5612: - b - optional vector of right hand side, that will be adjusted by provided solution
5614: Notes:
5615: This does not change the nonzero structure of the matrix, it merely zeros those entries in the matrix.
5617: The user can set a value in the diagonal entry (or for the AIJ and
5618: row formats can optionally remove the main diagonal entry from the
5619: nonzero structure as well, by passing 0.0 as the final argument).
5621: For the parallel case, all processes that share the matrix (i.e.,
5622: those in the communicator used for matrix creation) MUST call this
5623: routine, regardless of whether any rows being zeroed are owned by
5624: them.
5626: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5627: list only rows local to itself).
5629: The option MAT_NO_OFF_PROC_ZERO_ROWS does not apply to this routine.
5631: Level: intermediate
5633: .seealso: MatZeroRowsIS(), MatZeroRows(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5634: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
5635: @*/
5636: PetscErrorCode MatZeroRowsColumns(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
5637: {
5644: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5645: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5646: if (!mat->ops->zerorowscolumns) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5647: MatCheckPreallocated(mat,1);
5649: (*mat->ops->zerorowscolumns)(mat,numRows,rows,diag,x,b);
5650: MatViewFromOptions(mat,NULL,"-mat_view");
5651: PetscObjectStateIncrease((PetscObject)mat);
5652: return(0);
5653: }
5655: /*@
5656: MatZeroRowsColumnsIS - Zeros all entries (except possibly the main diagonal)
5657: of a set of rows and columns of a matrix.
5659: Collective on Mat
5661: Input Parameters:
5662: + mat - the matrix
5663: . is - the rows to zero
5664: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5665: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5666: - b - optional vector of right hand side, that will be adjusted by provided solution
5668: Notes:
5669: This does not change the nonzero structure of the matrix, it merely zeros those entries in the matrix.
5671: The user can set a value in the diagonal entry (or for the AIJ and
5672: row formats can optionally remove the main diagonal entry from the
5673: nonzero structure as well, by passing 0.0 as the final argument).
5675: For the parallel case, all processes that share the matrix (i.e.,
5676: those in the communicator used for matrix creation) MUST call this
5677: routine, regardless of whether any rows being zeroed are owned by
5678: them.
5680: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5681: list only rows local to itself).
5683: The option MAT_NO_OFF_PROC_ZERO_ROWS does not apply to this routine.
5685: Level: intermediate
5687: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5688: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRows(), MatZeroRowsColumnsStencil()
5689: @*/
5690: PetscErrorCode MatZeroRowsColumnsIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
5691: {
5693: PetscInt numRows;
5694: const PetscInt *rows;
5701: ISGetLocalSize(is,&numRows);
5702: ISGetIndices(is,&rows);
5703: MatZeroRowsColumns(mat,numRows,rows,diag,x,b);
5704: ISRestoreIndices(is,&rows);
5705: return(0);
5706: }
5708: /*@
5709: MatZeroRows - Zeros all entries (except possibly the main diagonal)
5710: of a set of rows of a matrix.
5712: Collective on Mat
5714: Input Parameters:
5715: + mat - the matrix
5716: . numRows - the number of rows to remove
5717: . rows - the global row indices
5718: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5719: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5720: - b - optional vector of right hand side, that will be adjusted by provided solution
5722: Notes:
5723: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5724: but does not release memory. For the dense and block diagonal
5725: formats this does not alter the nonzero structure.
5727: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5728: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5729: merely zeroed.
5731: The user can set a value in the diagonal entry (or for the AIJ and
5732: row formats can optionally remove the main diagonal entry from the
5733: nonzero structure as well, by passing 0.0 as the final argument).
5735: For the parallel case, all processes that share the matrix (i.e.,
5736: those in the communicator used for matrix creation) MUST call this
5737: routine, regardless of whether any rows being zeroed are owned by
5738: them.
5740: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5741: list only rows local to itself).
5743: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
5744: owns that are to be zeroed. This saves a global synchronization in the implementation.
5746: Level: intermediate
5748: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5749: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
5750: @*/
5751: PetscErrorCode MatZeroRows(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
5752: {
5759: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
5760: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
5761: if (!mat->ops->zerorows) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
5762: MatCheckPreallocated(mat,1);
5764: (*mat->ops->zerorows)(mat,numRows,rows,diag,x,b);
5765: MatViewFromOptions(mat,NULL,"-mat_view");
5766: PetscObjectStateIncrease((PetscObject)mat);
5767: return(0);
5768: }
5770: /*@
5771: MatZeroRowsIS - Zeros all entries (except possibly the main diagonal)
5772: of a set of rows of a matrix.
5774: Collective on Mat
5776: Input Parameters:
5777: + mat - the matrix
5778: . is - index set of rows to remove
5779: . diag - value put in all diagonals of eliminated rows
5780: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5781: - b - optional vector of right hand side, that will be adjusted by provided solution
5783: Notes:
5784: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5785: but does not release memory. For the dense and block diagonal
5786: formats this does not alter the nonzero structure.
5788: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5789: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5790: merely zeroed.
5792: The user can set a value in the diagonal entry (or for the AIJ and
5793: row formats can optionally remove the main diagonal entry from the
5794: nonzero structure as well, by passing 0.0 as the final argument).
5796: For the parallel case, all processes that share the matrix (i.e.,
5797: those in the communicator used for matrix creation) MUST call this
5798: routine, regardless of whether any rows being zeroed are owned by
5799: them.
5801: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5802: list only rows local to itself).
5804: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
5805: owns that are to be zeroed. This saves a global synchronization in the implementation.
5807: Level: intermediate
5809: .seealso: MatZeroRows(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5810: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
5811: @*/
5812: PetscErrorCode MatZeroRowsIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
5813: {
5814: PetscInt numRows;
5815: const PetscInt *rows;
5822: ISGetLocalSize(is,&numRows);
5823: ISGetIndices(is,&rows);
5824: MatZeroRows(mat,numRows,rows,diag,x,b);
5825: ISRestoreIndices(is,&rows);
5826: return(0);
5827: }
5829: /*@
5830: MatZeroRowsStencil - Zeros all entries (except possibly the main diagonal)
5831: of a set of rows of a matrix. These rows must be local to the process.
5833: Collective on Mat
5835: Input Parameters:
5836: + mat - the matrix
5837: . numRows - the number of rows to remove
5838: . rows - the grid coordinates (and component number when dof > 1) for matrix rows
5839: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5840: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5841: - b - optional vector of right hand side, that will be adjusted by provided solution
5843: Notes:
5844: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5845: but does not release memory. For the dense and block diagonal
5846: formats this does not alter the nonzero structure.
5848: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5849: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5850: merely zeroed.
5852: The user can set a value in the diagonal entry (or for the AIJ and
5853: row formats can optionally remove the main diagonal entry from the
5854: nonzero structure as well, by passing 0.0 as the final argument).
5856: For the parallel case, all processes that share the matrix (i.e.,
5857: those in the communicator used for matrix creation) MUST call this
5858: routine, regardless of whether any rows being zeroed are owned by
5859: them.
5861: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5862: list only rows local to itself).
5864: The grid coordinates are across the entire grid, not just the local portion
5866: In Fortran idxm and idxn should be declared as
5867: $ MatStencil idxm(4,m)
5868: and the values inserted using
5869: $ idxm(MatStencil_i,1) = i
5870: $ idxm(MatStencil_j,1) = j
5871: $ idxm(MatStencil_k,1) = k
5872: $ idxm(MatStencil_c,1) = c
5873: etc
5875: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
5876: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
5877: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
5878: DM_BOUNDARY_PERIODIC boundary type.
5880: 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
5881: a single value per point) you can skip filling those indices.
5883: Level: intermediate
5885: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsl(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5886: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
5887: @*/
5888: PetscErrorCode MatZeroRowsStencil(Mat mat,PetscInt numRows,const MatStencil rows[],PetscScalar diag,Vec x,Vec b)
5889: {
5890: PetscInt dim = mat->stencil.dim;
5891: PetscInt sdim = dim - (1 - (PetscInt) mat->stencil.noc);
5892: PetscInt *dims = mat->stencil.dims+1;
5893: PetscInt *starts = mat->stencil.starts;
5894: PetscInt *dxm = (PetscInt*) rows;
5895: PetscInt *jdxm, i, j, tmp, numNewRows = 0;
5903: PetscMalloc1(numRows, &jdxm);
5904: for (i = 0; i < numRows; ++i) {
5905: /* Skip unused dimensions (they are ordered k, j, i, c) */
5906: for (j = 0; j < 3-sdim; ++j) dxm++;
5907: /* Local index in X dir */
5908: tmp = *dxm++ - starts[0];
5909: /* Loop over remaining dimensions */
5910: for (j = 0; j < dim-1; ++j) {
5911: /* If nonlocal, set index to be negative */
5912: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
5913: /* Update local index */
5914: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
5915: }
5916: /* Skip component slot if necessary */
5917: if (mat->stencil.noc) dxm++;
5918: /* Local row number */
5919: if (tmp >= 0) {
5920: jdxm[numNewRows++] = tmp;
5921: }
5922: }
5923: MatZeroRowsLocal(mat,numNewRows,jdxm,diag,x,b);
5924: PetscFree(jdxm);
5925: return(0);
5926: }
5928: /*@
5929: MatZeroRowsColumnsStencil - Zeros all row and column entries (except possibly the main diagonal)
5930: of a set of rows and columns of a matrix.
5932: Collective on Mat
5934: Input Parameters:
5935: + mat - the matrix
5936: . numRows - the number of rows/columns to remove
5937: . rows - the grid coordinates (and component number when dof > 1) for matrix rows
5938: . diag - value put in all diagonals of eliminated rows (0.0 will even eliminate diagonal entry)
5939: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
5940: - b - optional vector of right hand side, that will be adjusted by provided solution
5942: Notes:
5943: For the AIJ and BAIJ matrix formats this removes the old nonzero structure,
5944: but does not release memory. For the dense and block diagonal
5945: formats this does not alter the nonzero structure.
5947: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
5948: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
5949: merely zeroed.
5951: The user can set a value in the diagonal entry (or for the AIJ and
5952: row formats can optionally remove the main diagonal entry from the
5953: nonzero structure as well, by passing 0.0 as the final argument).
5955: For the parallel case, all processes that share the matrix (i.e.,
5956: those in the communicator used for matrix creation) MUST call this
5957: routine, regardless of whether any rows being zeroed are owned by
5958: them.
5960: Each processor can indicate any rows in the entire matrix to be zeroed (i.e. each process does NOT have to
5961: list only rows local to itself, but the row/column numbers are given in local numbering).
5963: The grid coordinates are across the entire grid, not just the local portion
5965: In Fortran idxm and idxn should be declared as
5966: $ MatStencil idxm(4,m)
5967: and the values inserted using
5968: $ idxm(MatStencil_i,1) = i
5969: $ idxm(MatStencil_j,1) = j
5970: $ idxm(MatStencil_k,1) = k
5971: $ idxm(MatStencil_c,1) = c
5972: etc
5974: For periodic boundary conditions use negative indices for values to the left (below 0; that are to be
5975: obtained by wrapping values from right edge). For values to the right of the last entry using that index plus one
5976: etc to obtain values that obtained by wrapping the values from the left edge. This does not work for anything but the
5977: DM_BOUNDARY_PERIODIC boundary type.
5979: 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
5980: a single value per point) you can skip filling those indices.
5982: Level: intermediate
5984: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
5985: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRows()
5986: @*/
5987: PetscErrorCode MatZeroRowsColumnsStencil(Mat mat,PetscInt numRows,const MatStencil rows[],PetscScalar diag,Vec x,Vec b)
5988: {
5989: PetscInt dim = mat->stencil.dim;
5990: PetscInt sdim = dim - (1 - (PetscInt) mat->stencil.noc);
5991: PetscInt *dims = mat->stencil.dims+1;
5992: PetscInt *starts = mat->stencil.starts;
5993: PetscInt *dxm = (PetscInt*) rows;
5994: PetscInt *jdxm, i, j, tmp, numNewRows = 0;
6002: PetscMalloc1(numRows, &jdxm);
6003: for (i = 0; i < numRows; ++i) {
6004: /* Skip unused dimensions (they are ordered k, j, i, c) */
6005: for (j = 0; j < 3-sdim; ++j) dxm++;
6006: /* Local index in X dir */
6007: tmp = *dxm++ - starts[0];
6008: /* Loop over remaining dimensions */
6009: for (j = 0; j < dim-1; ++j) {
6010: /* If nonlocal, set index to be negative */
6011: if ((*dxm++ - starts[j+1]) < 0 || tmp < 0) tmp = PETSC_MIN_INT;
6012: /* Update local index */
6013: else tmp = tmp*dims[j] + *(dxm-1) - starts[j+1];
6014: }
6015: /* Skip component slot if necessary */
6016: if (mat->stencil.noc) dxm++;
6017: /* Local row number */
6018: if (tmp >= 0) {
6019: jdxm[numNewRows++] = tmp;
6020: }
6021: }
6022: MatZeroRowsColumnsLocal(mat,numNewRows,jdxm,diag,x,b);
6023: PetscFree(jdxm);
6024: return(0);
6025: }
6027: /*@C
6028: MatZeroRowsLocal - Zeros all entries (except possibly the main diagonal)
6029: of a set of rows of a matrix; using local numbering of rows.
6031: Collective on Mat
6033: Input Parameters:
6034: + mat - the matrix
6035: . numRows - the number of rows to remove
6036: . rows - the global row indices
6037: . diag - value put in all diagonals of eliminated rows
6038: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6039: - b - optional vector of right hand side, that will be adjusted by provided solution
6041: Notes:
6042: Before calling MatZeroRowsLocal(), the user must first set the
6043: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6045: For the AIJ matrix formats this removes the old nonzero structure,
6046: but does not release memory. For the dense and block diagonal
6047: formats this does not alter the nonzero structure.
6049: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6050: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6051: merely zeroed.
6053: The user can set a value in the diagonal entry (or for the AIJ and
6054: row formats can optionally remove the main diagonal entry from the
6055: nonzero structure as well, by passing 0.0 as the final argument).
6057: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6058: owns that are to be zeroed. This saves a global synchronization in the implementation.
6060: Level: intermediate
6062: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRows(), MatSetOption(),
6063: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6064: @*/
6065: PetscErrorCode MatZeroRowsLocal(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6066: {
6073: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6074: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6075: MatCheckPreallocated(mat,1);
6077: if (mat->ops->zerorowslocal) {
6078: (*mat->ops->zerorowslocal)(mat,numRows,rows,diag,x,b);
6079: } else {
6080: IS is, newis;
6081: const PetscInt *newRows;
6083: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Need to provide local to global mapping to matrix first");
6084: ISCreateGeneral(PETSC_COMM_SELF,numRows,rows,PETSC_COPY_VALUES,&is);
6085: ISLocalToGlobalMappingApplyIS(mat->rmap->mapping,is,&newis);
6086: ISGetIndices(newis,&newRows);
6087: (*mat->ops->zerorows)(mat,numRows,newRows,diag,x,b);
6088: ISRestoreIndices(newis,&newRows);
6089: ISDestroy(&newis);
6090: ISDestroy(&is);
6091: }
6092: PetscObjectStateIncrease((PetscObject)mat);
6093: return(0);
6094: }
6096: /*@
6097: MatZeroRowsLocalIS - Zeros all entries (except possibly the main diagonal)
6098: of a set of rows of a matrix; using local numbering of rows.
6100: Collective on Mat
6102: Input Parameters:
6103: + mat - the matrix
6104: . is - index set of rows to remove
6105: . diag - value put in all diagonals of eliminated rows
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: Before calling MatZeroRowsLocalIS(), the user must first set the
6111: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6113: For the AIJ matrix formats this removes the old nonzero structure,
6114: but does not release memory. For the dense and block diagonal
6115: formats this does not alter the nonzero structure.
6117: If the option MatSetOption(mat,MAT_KEEP_NONZERO_PATTERN,PETSC_TRUE) the nonzero structure
6118: of the matrix is not changed (even for AIJ and BAIJ matrices) the values are
6119: merely zeroed.
6121: The user can set a value in the diagonal entry (or for the AIJ and
6122: row formats can optionally remove the main diagonal entry from the
6123: nonzero structure as well, by passing 0.0 as the final argument).
6125: You can call MatSetOption(mat,MAT_NO_OFF_PROC_ZERO_ROWS,PETSC_TRUE) if each process indicates only rows it
6126: owns that are to be zeroed. This saves a global synchronization in the implementation.
6128: Level: intermediate
6130: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRows(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6131: MatZeroRowsColumnsLocal(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6132: @*/
6133: PetscErrorCode MatZeroRowsLocalIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6134: {
6136: PetscInt numRows;
6137: const PetscInt *rows;
6143: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6144: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6145: MatCheckPreallocated(mat,1);
6147: ISGetLocalSize(is,&numRows);
6148: ISGetIndices(is,&rows);
6149: MatZeroRowsLocal(mat,numRows,rows,diag,x,b);
6150: ISRestoreIndices(is,&rows);
6151: return(0);
6152: }
6154: /*@
6155: MatZeroRowsColumnsLocal - Zeros all entries (except possibly the main diagonal)
6156: of a set of rows and columns of a matrix; using local numbering of rows.
6158: Collective on Mat
6160: Input Parameters:
6161: + mat - the matrix
6162: . numRows - the number of rows to remove
6163: . rows - the global row indices
6164: . diag - value put in all diagonals of eliminated rows
6165: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6166: - b - optional vector of right hand side, that will be adjusted by provided solution
6168: Notes:
6169: Before calling MatZeroRowsColumnsLocal(), the user must first set the
6170: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6172: The user can set a value in the diagonal entry (or for the AIJ and
6173: row formats can optionally remove the main diagonal entry from the
6174: nonzero structure as well, by passing 0.0 as the final argument).
6176: Level: intermediate
6178: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6179: MatZeroRows(), MatZeroRowsColumnsLocalIS(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6180: @*/
6181: PetscErrorCode MatZeroRowsColumnsLocal(Mat mat,PetscInt numRows,const PetscInt rows[],PetscScalar diag,Vec x,Vec b)
6182: {
6184: IS is, newis;
6185: const PetscInt *newRows;
6191: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6192: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6193: MatCheckPreallocated(mat,1);
6195: if (!mat->cmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Need to provide local to global mapping to matrix first");
6196: ISCreateGeneral(PETSC_COMM_SELF,numRows,rows,PETSC_COPY_VALUES,&is);
6197: ISLocalToGlobalMappingApplyIS(mat->cmap->mapping,is,&newis);
6198: ISGetIndices(newis,&newRows);
6199: (*mat->ops->zerorowscolumns)(mat,numRows,newRows,diag,x,b);
6200: ISRestoreIndices(newis,&newRows);
6201: ISDestroy(&newis);
6202: ISDestroy(&is);
6203: PetscObjectStateIncrease((PetscObject)mat);
6204: return(0);
6205: }
6207: /*@
6208: MatZeroRowsColumnsLocalIS - Zeros all entries (except possibly the main diagonal)
6209: of a set of rows and columns of a matrix; using local numbering of rows.
6211: Collective on Mat
6213: Input Parameters:
6214: + mat - the matrix
6215: . is - index set of rows to remove
6216: . diag - value put in all diagonals of eliminated rows
6217: . x - optional vector of solutions for zeroed rows (other entries in vector are not used)
6218: - b - optional vector of right hand side, that will be adjusted by provided solution
6220: Notes:
6221: Before calling MatZeroRowsColumnsLocalIS(), the user must first set the
6222: local-to-global mapping by calling MatSetLocalToGlobalMapping().
6224: The user can set a value in the diagonal entry (or for the AIJ and
6225: row formats can optionally remove the main diagonal entry from the
6226: nonzero structure as well, by passing 0.0 as the final argument).
6228: Level: intermediate
6230: .seealso: MatZeroRowsIS(), MatZeroRowsColumns(), MatZeroRowsLocalIS(), MatZeroRowsStencil(), MatZeroEntries(), MatZeroRowsLocal(), MatSetOption(),
6231: MatZeroRowsColumnsLocal(), MatZeroRows(), MatZeroRowsColumnsIS(), MatZeroRowsColumnsStencil()
6232: @*/
6233: PetscErrorCode MatZeroRowsColumnsLocalIS(Mat mat,IS is,PetscScalar diag,Vec x,Vec b)
6234: {
6236: PetscInt numRows;
6237: const PetscInt *rows;
6243: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6244: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6245: MatCheckPreallocated(mat,1);
6247: ISGetLocalSize(is,&numRows);
6248: ISGetIndices(is,&rows);
6249: MatZeroRowsColumnsLocal(mat,numRows,rows,diag,x,b);
6250: ISRestoreIndices(is,&rows);
6251: return(0);
6252: }
6254: /*@C
6255: MatGetSize - Returns the numbers of rows and columns in a matrix.
6257: Not Collective
6259: Input Parameter:
6260: . mat - the matrix
6262: Output Parameters:
6263: + m - the number of global rows
6264: - n - the number of global columns
6266: Note: both output parameters can be NULL on input.
6268: Level: beginner
6270: .seealso: MatGetLocalSize()
6271: @*/
6272: PetscErrorCode MatGetSize(Mat mat,PetscInt *m,PetscInt *n)
6273: {
6276: if (m) *m = mat->rmap->N;
6277: if (n) *n = mat->cmap->N;
6278: return(0);
6279: }
6281: /*@C
6282: MatGetLocalSize - Returns the number of rows and columns in a matrix
6283: stored locally. This information may be implementation dependent, so
6284: use with care.
6286: Not Collective
6288: Input Parameters:
6289: . mat - the matrix
6291: Output Parameters:
6292: + m - the number of local rows
6293: - n - the number of local columns
6295: Note: both output parameters can be NULL on input.
6297: Level: beginner
6299: .seealso: MatGetSize()
6300: @*/
6301: PetscErrorCode MatGetLocalSize(Mat mat,PetscInt *m,PetscInt *n)
6302: {
6307: if (m) *m = mat->rmap->n;
6308: if (n) *n = mat->cmap->n;
6309: return(0);
6310: }
6312: /*@C
6313: MatGetOwnershipRangeColumn - Returns the range of matrix columns associated with rows of a vector one multiplies by that owned by
6314: this processor. (The columns of the "diagonal block")
6316: Not Collective, unless matrix has not been allocated, then collective on Mat
6318: Input Parameters:
6319: . mat - the matrix
6321: Output Parameters:
6322: + m - the global index of the first local column
6323: - n - one more than the global index of the last local column
6325: Notes:
6326: both output parameters can be NULL on input.
6328: Level: developer
6330: .seealso: MatGetOwnershipRange(), MatGetOwnershipRanges(), MatGetOwnershipRangesColumn()
6332: @*/
6333: PetscErrorCode MatGetOwnershipRangeColumn(Mat mat,PetscInt *m,PetscInt *n)
6334: {
6340: MatCheckPreallocated(mat,1);
6341: if (m) *m = mat->cmap->rstart;
6342: if (n) *n = mat->cmap->rend;
6343: return(0);
6344: }
6346: /*@C
6347: MatGetOwnershipRange - Returns the range of matrix rows owned by
6348: this processor, assuming that the matrix is laid out with the first
6349: n1 rows on the first processor, the next n2 rows on the second, etc.
6350: For certain parallel layouts this range may not be well defined.
6352: Not Collective
6354: Input Parameters:
6355: . mat - the matrix
6357: Output Parameters:
6358: + m - the global index of the first local row
6359: - n - one more than the global index of the last local row
6361: Note: Both output parameters can be NULL on input.
6362: $ This function requires that the matrix be preallocated. If you have not preallocated, consider using
6363: $ PetscSplitOwnership(MPI_Comm comm, PetscInt *n, PetscInt *N)
6364: $ and then MPI_Scan() to calculate prefix sums of the local sizes.
6366: Level: beginner
6368: .seealso: MatGetOwnershipRanges(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn(), PetscSplitOwnership(), PetscSplitOwnershipBlock()
6370: @*/
6371: PetscErrorCode MatGetOwnershipRange(Mat mat,PetscInt *m,PetscInt *n)
6372: {
6378: MatCheckPreallocated(mat,1);
6379: if (m) *m = mat->rmap->rstart;
6380: if (n) *n = mat->rmap->rend;
6381: return(0);
6382: }
6384: /*@C
6385: MatGetOwnershipRanges - Returns the range of matrix rows owned by
6386: each process
6388: Not Collective, unless matrix has not been allocated, then collective on Mat
6390: Input Parameters:
6391: . mat - the matrix
6393: Output Parameters:
6394: . ranges - start of each processors portion plus one more than the total length at the end
6396: Level: beginner
6398: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRangesColumn()
6400: @*/
6401: PetscErrorCode MatGetOwnershipRanges(Mat mat,const PetscInt **ranges)
6402: {
6408: MatCheckPreallocated(mat,1);
6409: PetscLayoutGetRanges(mat->rmap,ranges);
6410: return(0);
6411: }
6413: /*@C
6414: MatGetOwnershipRangesColumn - Returns the range of matrix columns associated with rows of a vector one multiplies by that owned by
6415: this processor. (The columns of the "diagonal blocks" for each process)
6417: Not Collective, unless matrix has not been allocated, then collective on Mat
6419: Input Parameters:
6420: . mat - the matrix
6422: Output Parameters:
6423: . ranges - start of each processors portion plus one more then the total length at the end
6425: Level: beginner
6427: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatGetOwnershipRanges()
6429: @*/
6430: PetscErrorCode MatGetOwnershipRangesColumn(Mat mat,const PetscInt **ranges)
6431: {
6437: MatCheckPreallocated(mat,1);
6438: PetscLayoutGetRanges(mat->cmap,ranges);
6439: return(0);
6440: }
6442: /*@C
6443: MatGetOwnershipIS - Get row and column ownership as index sets
6445: Not Collective
6447: Input Arguments:
6448: . A - matrix of type Elemental
6450: Output Arguments:
6451: + rows - rows in which this process owns elements
6452: - cols - columns in which this process owns elements
6454: Level: intermediate
6456: .seealso: MatGetOwnershipRange(), MatGetOwnershipRangeColumn(), MatSetValues(), MATELEMENTAL
6457: @*/
6458: PetscErrorCode MatGetOwnershipIS(Mat A,IS *rows,IS *cols)
6459: {
6460: PetscErrorCode ierr,(*f)(Mat,IS*,IS*);
6463: MatCheckPreallocated(A,1);
6464: PetscObjectQueryFunction((PetscObject)A,"MatGetOwnershipIS_C",&f);
6465: if (f) {
6466: (*f)(A,rows,cols);
6467: } else { /* Create a standard row-based partition, each process is responsible for ALL columns in their row block */
6468: if (rows) {ISCreateStride(PETSC_COMM_SELF,A->rmap->n,A->rmap->rstart,1,rows);}
6469: if (cols) {ISCreateStride(PETSC_COMM_SELF,A->cmap->N,0,1,cols);}
6470: }
6471: return(0);
6472: }
6474: /*@C
6475: MatILUFactorSymbolic - Performs symbolic ILU factorization of a matrix.
6476: Uses levels of fill only, not drop tolerance. Use MatLUFactorNumeric()
6477: to complete the factorization.
6479: Collective on Mat
6481: Input Parameters:
6482: + mat - the matrix
6483: . row - row permutation
6484: . column - column permutation
6485: - info - structure containing
6486: $ levels - number of levels of fill.
6487: $ expected fill - as ratio of original fill.
6488: $ 1 or 0 - indicating force fill on diagonal (improves robustness for matrices
6489: missing diagonal entries)
6491: Output Parameters:
6492: . fact - new matrix that has been symbolically factored
6494: Notes:
6495: See Users-Manual: ch_mat for additional information about choosing the fill factor for better efficiency.
6497: Most users should employ the simplified KSP interface for linear solvers
6498: instead of working directly with matrix algebra routines such as this.
6499: See, e.g., KSPCreate().
6501: Level: developer
6503: .seealso: MatLUFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
6504: MatGetOrdering(), MatFactorInfo
6506: Note: this uses the definition of level of fill as in Y. Saad, 2003
6508: Developer Note: fortran interface is not autogenerated as the f90
6509: interface defintion cannot be generated correctly [due to MatFactorInfo]
6511: References:
6512: Y. Saad, Iterative methods for sparse linear systems Philadelphia: Society for Industrial and Applied Mathematics, 2003
6513: @*/
6514: PetscErrorCode MatILUFactorSymbolic(Mat fact,Mat mat,IS row,IS col,const MatFactorInfo *info)
6515: {
6525: if (info->levels < 0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Levels of fill negative %D",(PetscInt)info->levels);
6526: if (info->fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %g",(double)info->fill);
6527: if (!(fact)->ops->ilufactorsymbolic) {
6528: MatSolverType spackage;
6529: MatFactorGetSolverType(fact,&spackage);
6530: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic ILU using solver package %s",((PetscObject)mat)->type_name,spackage);
6531: }
6532: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6533: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6534: MatCheckPreallocated(mat,2);
6536: PetscLogEventBegin(MAT_ILUFactorSymbolic,mat,row,col,0);
6537: (fact->ops->ilufactorsymbolic)(fact,mat,row,col,info);
6538: PetscLogEventEnd(MAT_ILUFactorSymbolic,mat,row,col,0);
6539: return(0);
6540: }
6542: /*@C
6543: MatICCFactorSymbolic - Performs symbolic incomplete
6544: Cholesky factorization for a symmetric matrix. Use
6545: MatCholeskyFactorNumeric() to complete the factorization.
6547: Collective on Mat
6549: Input Parameters:
6550: + mat - the matrix
6551: . perm - row and column permutation
6552: - info - structure containing
6553: $ levels - number of levels of fill.
6554: $ expected fill - as ratio of original fill.
6556: Output Parameter:
6557: . fact - the factored matrix
6559: Notes:
6560: Most users should employ the KSP interface for linear solvers
6561: instead of working directly with matrix algebra routines such as this.
6562: See, e.g., KSPCreate().
6564: Level: developer
6566: .seealso: MatCholeskyFactorNumeric(), MatCholeskyFactor(), MatFactorInfo
6568: Note: this uses the definition of level of fill as in Y. Saad, 2003
6570: Developer Note: fortran interface is not autogenerated as the f90
6571: interface defintion cannot be generated correctly [due to MatFactorInfo]
6573: References:
6574: Y. Saad, Iterative methods for sparse linear systems Philadelphia: Society for Industrial and Applied Mathematics, 2003
6575: @*/
6576: PetscErrorCode MatICCFactorSymbolic(Mat fact,Mat mat,IS perm,const MatFactorInfo *info)
6577: {
6586: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6587: if (info->levels < 0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Levels negative %D",(PetscInt) info->levels);
6588: if (info->fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Expected fill less than 1.0 %g",(double)info->fill);
6589: if (!(fact)->ops->iccfactorsymbolic) {
6590: MatSolverType spackage;
6591: MatFactorGetSolverType(fact,&spackage);
6592: SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Matrix type %s symbolic ICC using solver package %s",((PetscObject)mat)->type_name,spackage);
6593: }
6594: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6595: MatCheckPreallocated(mat,2);
6597: PetscLogEventBegin(MAT_ICCFactorSymbolic,mat,perm,0,0);
6598: (fact->ops->iccfactorsymbolic)(fact,mat,perm,info);
6599: PetscLogEventEnd(MAT_ICCFactorSymbolic,mat,perm,0,0);
6600: return(0);
6601: }
6603: /*@C
6604: MatCreateSubMatrices - Extracts several submatrices from a matrix. If submat
6605: points to an array of valid matrices, they may be reused to store the new
6606: submatrices.
6608: Collective on Mat
6610: Input Parameters:
6611: + mat - the matrix
6612: . n - the number of submatrixes to be extracted (on this processor, may be zero)
6613: . irow, icol - index sets of rows and columns to extract
6614: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
6616: Output Parameter:
6617: . submat - the array of submatrices
6619: Notes:
6620: MatCreateSubMatrices() can extract ONLY sequential submatrices
6621: (from both sequential and parallel matrices). Use MatCreateSubMatrix()
6622: to extract a parallel submatrix.
6624: Some matrix types place restrictions on the row and column
6625: indices, such as that they be sorted or that they be equal to each other.
6627: The index sets may not have duplicate entries.
6629: When extracting submatrices from a parallel matrix, each processor can
6630: form a different submatrix by setting the rows and columns of its
6631: individual index sets according to the local submatrix desired.
6633: When finished using the submatrices, the user should destroy
6634: them with MatDestroySubMatrices().
6636: MAT_REUSE_MATRIX can only be used when the nonzero structure of the
6637: original matrix has not changed from that last call to MatCreateSubMatrices().
6639: This routine creates the matrices in submat; you should NOT create them before
6640: calling it. It also allocates the array of matrix pointers submat.
6642: For BAIJ matrices the index sets must respect the block structure, that is if they
6643: request one row/column in a block, they must request all rows/columns that are in
6644: that block. For example, if the block size is 2 you cannot request just row 0 and
6645: column 0.
6647: Fortran Note:
6648: The Fortran interface is slightly different from that given below; it
6649: requires one to pass in as submat a Mat (integer) array of size at least n+1.
6651: Level: advanced
6654: .seealso: MatDestroySubMatrices(), MatCreateSubMatrix(), MatGetRow(), MatGetDiagonal(), MatReuse
6655: @*/
6656: PetscErrorCode MatCreateSubMatrices(Mat mat,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *submat[])
6657: {
6659: PetscInt i;
6660: PetscBool eq;
6665: if (n) {
6670: }
6672: if (n && scall == MAT_REUSE_MATRIX) {
6675: }
6676: if (!mat->ops->createsubmatrices) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6677: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6678: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6679: MatCheckPreallocated(mat,1);
6681: PetscLogEventBegin(MAT_CreateSubMats,mat,0,0,0);
6682: (*mat->ops->createsubmatrices)(mat,n,irow,icol,scall,submat);
6683: PetscLogEventEnd(MAT_CreateSubMats,mat,0,0,0);
6684: for (i=0; i<n; i++) {
6685: (*submat)[i]->factortype = MAT_FACTOR_NONE; /* in case in place factorization was previously done on submatrix */
6686: if (mat->symmetric || mat->structurally_symmetric || mat->hermitian) {
6687: ISEqual(irow[i],icol[i],&eq);
6688: if (eq) {
6689: if (mat->symmetric) {
6690: MatSetOption((*submat)[i],MAT_SYMMETRIC,PETSC_TRUE);
6691: } else if (mat->hermitian) {
6692: MatSetOption((*submat)[i],MAT_HERMITIAN,PETSC_TRUE);
6693: } else if (mat->structurally_symmetric) {
6694: MatSetOption((*submat)[i],MAT_STRUCTURALLY_SYMMETRIC,PETSC_TRUE);
6695: }
6696: }
6697: }
6698: }
6699: return(0);
6700: }
6702: /*@C
6703: MatCreateSubMatricesMPI - Extracts MPI submatrices across a sub communicator of mat (by pairs of IS that may live on subcomms).
6705: Collective on Mat
6707: Input Parameters:
6708: + mat - the matrix
6709: . n - the number of submatrixes to be extracted
6710: . irow, icol - index sets of rows and columns to extract
6711: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
6713: Output Parameter:
6714: . submat - the array of submatrices
6716: Level: advanced
6719: .seealso: MatCreateSubMatrices(), MatCreateSubMatrix(), MatGetRow(), MatGetDiagonal(), MatReuse
6720: @*/
6721: PetscErrorCode MatCreateSubMatricesMPI(Mat mat,PetscInt n,const IS irow[],const IS icol[],MatReuse scall,Mat *submat[])
6722: {
6724: PetscInt i;
6725: PetscBool eq;
6730: if (n) {
6735: }
6737: if (n && scall == MAT_REUSE_MATRIX) {
6740: }
6741: if (!mat->ops->createsubmatricesmpi) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6742: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6743: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6744: MatCheckPreallocated(mat,1);
6746: PetscLogEventBegin(MAT_CreateSubMats,mat,0,0,0);
6747: (*mat->ops->createsubmatricesmpi)(mat,n,irow,icol,scall,submat);
6748: PetscLogEventEnd(MAT_CreateSubMats,mat,0,0,0);
6749: for (i=0; i<n; i++) {
6750: if (mat->symmetric || mat->structurally_symmetric || mat->hermitian) {
6751: ISEqual(irow[i],icol[i],&eq);
6752: if (eq) {
6753: if (mat->symmetric) {
6754: MatSetOption((*submat)[i],MAT_SYMMETRIC,PETSC_TRUE);
6755: } else if (mat->hermitian) {
6756: MatSetOption((*submat)[i],MAT_HERMITIAN,PETSC_TRUE);
6757: } else if (mat->structurally_symmetric) {
6758: MatSetOption((*submat)[i],MAT_STRUCTURALLY_SYMMETRIC,PETSC_TRUE);
6759: }
6760: }
6761: }
6762: }
6763: return(0);
6764: }
6766: /*@C
6767: MatDestroyMatrices - Destroys an array of matrices.
6769: Collective on Mat
6771: Input Parameters:
6772: + n - the number of local matrices
6773: - mat - the matrices (note that this is a pointer to the array of matrices)
6775: Level: advanced
6777: Notes:
6778: Frees not only the matrices, but also the array that contains the matrices
6779: In Fortran will not free the array.
6781: .seealso: MatCreateSubMatrices() MatDestroySubMatrices()
6782: @*/
6783: PetscErrorCode MatDestroyMatrices(PetscInt n,Mat *mat[])
6784: {
6786: PetscInt i;
6789: if (!*mat) return(0);
6790: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Trying to destroy negative number of matrices %D",n);
6793: for (i=0; i<n; i++) {
6794: MatDestroy(&(*mat)[i]);
6795: }
6797: /* memory is allocated even if n = 0 */
6798: PetscFree(*mat);
6799: return(0);
6800: }
6802: /*@C
6803: MatDestroySubMatrices - Destroys a set of matrices obtained with MatCreateSubMatrices().
6805: Collective on Mat
6807: Input Parameters:
6808: + n - the number of local matrices
6809: - mat - the matrices (note that this is a pointer to the array of matrices, just to match the calling
6810: sequence of MatCreateSubMatrices())
6812: Level: advanced
6814: Notes:
6815: Frees not only the matrices, but also the array that contains the matrices
6816: In Fortran will not free the array.
6818: .seealso: MatCreateSubMatrices()
6819: @*/
6820: PetscErrorCode MatDestroySubMatrices(PetscInt n,Mat *mat[])
6821: {
6823: Mat mat0;
6826: if (!*mat) return(0);
6827: /* mat[] is an array of length n+1, see MatCreateSubMatrices_xxx() */
6828: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Trying to destroy negative number of matrices %D",n);
6831: mat0 = (*mat)[0];
6832: if (mat0 && mat0->ops->destroysubmatrices) {
6833: (mat0->ops->destroysubmatrices)(n,mat);
6834: } else {
6835: MatDestroyMatrices(n,mat);
6836: }
6837: return(0);
6838: }
6840: /*@C
6841: MatGetSeqNonzeroStructure - Extracts the sequential nonzero structure from a matrix.
6843: Collective on Mat
6845: Input Parameters:
6846: . mat - the matrix
6848: Output Parameter:
6849: . matstruct - the sequential matrix with the nonzero structure of mat
6851: Level: intermediate
6853: .seealso: MatDestroySeqNonzeroStructure(), MatCreateSubMatrices(), MatDestroyMatrices()
6854: @*/
6855: PetscErrorCode MatGetSeqNonzeroStructure(Mat mat,Mat *matstruct)
6856: {
6864: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6865: MatCheckPreallocated(mat,1);
6867: if (!mat->ops->getseqnonzerostructure) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Not for matrix type %s\n",((PetscObject)mat)->type_name);
6868: PetscLogEventBegin(MAT_GetSeqNonzeroStructure,mat,0,0,0);
6869: (*mat->ops->getseqnonzerostructure)(mat,matstruct);
6870: PetscLogEventEnd(MAT_GetSeqNonzeroStructure,mat,0,0,0);
6871: return(0);
6872: }
6874: /*@C
6875: MatDestroySeqNonzeroStructure - Destroys matrix obtained with MatGetSeqNonzeroStructure().
6877: Collective on Mat
6879: Input Parameters:
6880: . mat - the matrix (note that this is a pointer to the array of matrices, just to match the calling
6881: sequence of MatGetSequentialNonzeroStructure())
6883: Level: advanced
6885: Notes:
6886: Frees not only the matrices, but also the array that contains the matrices
6888: .seealso: MatGetSeqNonzeroStructure()
6889: @*/
6890: PetscErrorCode MatDestroySeqNonzeroStructure(Mat *mat)
6891: {
6896: MatDestroy(mat);
6897: return(0);
6898: }
6900: /*@
6901: MatIncreaseOverlap - Given a set of submatrices indicated by index sets,
6902: replaces the index sets by larger ones that represent submatrices with
6903: additional overlap.
6905: Collective on Mat
6907: Input Parameters:
6908: + mat - the matrix
6909: . n - the number of index sets
6910: . is - the array of index sets (these index sets will changed during the call)
6911: - ov - the additional overlap requested
6913: Options Database:
6914: . -mat_increase_overlap_scalable - use a scalable algorithm to compute the overlap (supported by MPIAIJ matrix)
6916: Level: developer
6919: .seealso: MatCreateSubMatrices()
6920: @*/
6921: PetscErrorCode MatIncreaseOverlap(Mat mat,PetscInt n,IS is[],PetscInt ov)
6922: {
6928: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Must have one or more domains, you have %D",n);
6929: if (n) {
6932: }
6933: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6934: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6935: MatCheckPreallocated(mat,1);
6937: if (!ov) return(0);
6938: if (!mat->ops->increaseoverlap) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
6939: PetscLogEventBegin(MAT_IncreaseOverlap,mat,0,0,0);
6940: (*mat->ops->increaseoverlap)(mat,n,is,ov);
6941: PetscLogEventEnd(MAT_IncreaseOverlap,mat,0,0,0);
6942: return(0);
6943: }
6946: PetscErrorCode MatIncreaseOverlapSplit_Single(Mat,IS*,PetscInt);
6948: /*@
6949: MatIncreaseOverlapSplit - Given a set of submatrices indicated by index sets across
6950: a sub communicator, replaces the index sets by larger ones that represent submatrices with
6951: additional overlap.
6953: Collective on Mat
6955: Input Parameters:
6956: + mat - the matrix
6957: . n - the number of index sets
6958: . is - the array of index sets (these index sets will changed during the call)
6959: - ov - the additional overlap requested
6961: Options Database:
6962: . -mat_increase_overlap_scalable - use a scalable algorithm to compute the overlap (supported by MPIAIJ matrix)
6964: Level: developer
6967: .seealso: MatCreateSubMatrices()
6968: @*/
6969: PetscErrorCode MatIncreaseOverlapSplit(Mat mat,PetscInt n,IS is[],PetscInt ov)
6970: {
6971: PetscInt i;
6977: if (n < 0) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"Must have one or more domains, you have %D",n);
6978: if (n) {
6981: }
6982: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
6983: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
6984: MatCheckPreallocated(mat,1);
6985: if (!ov) return(0);
6986: PetscLogEventBegin(MAT_IncreaseOverlap,mat,0,0,0);
6987: for(i=0; i<n; i++){
6988: MatIncreaseOverlapSplit_Single(mat,&is[i],ov);
6989: }
6990: PetscLogEventEnd(MAT_IncreaseOverlap,mat,0,0,0);
6991: return(0);
6992: }
6997: /*@
6998: MatGetBlockSize - Returns the matrix block size.
7000: Not Collective
7002: Input Parameter:
7003: . mat - the matrix
7005: Output Parameter:
7006: . bs - block size
7008: Notes:
7009: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7011: If the block size has not been set yet this routine returns 1.
7013: Level: intermediate
7015: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSizes()
7016: @*/
7017: PetscErrorCode MatGetBlockSize(Mat mat,PetscInt *bs)
7018: {
7022: *bs = PetscAbs(mat->rmap->bs);
7023: return(0);
7024: }
7026: /*@
7027: MatGetBlockSizes - Returns the matrix block row and column sizes.
7029: Not Collective
7031: Input Parameter:
7032: . mat - the matrix
7034: Output Parameter:
7035: + rbs - row block size
7036: - cbs - column block size
7038: Notes:
7039: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7040: If you pass a different block size for the columns than the rows, the row block size determines the square block storage.
7042: If a block size has not been set yet this routine returns 1.
7044: Level: intermediate
7046: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSize(), MatSetBlockSizes()
7047: @*/
7048: PetscErrorCode MatGetBlockSizes(Mat mat,PetscInt *rbs, PetscInt *cbs)
7049: {
7054: if (rbs) *rbs = PetscAbs(mat->rmap->bs);
7055: if (cbs) *cbs = PetscAbs(mat->cmap->bs);
7056: return(0);
7057: }
7059: /*@
7060: MatSetBlockSize - Sets the matrix block size.
7062: Logically Collective on Mat
7064: Input Parameters:
7065: + mat - the matrix
7066: - bs - block size
7068: Notes:
7069: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7070: This must be called before MatSetUp() or MatXXXSetPreallocation() (or will default to 1) and the block size cannot be changed later.
7072: For MATMPIAIJ and MATSEQAIJ matrix formats, this function can be called at a later stage, provided that the specified block size
7073: is compatible with the matrix local sizes.
7075: Level: intermediate
7077: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes()
7078: @*/
7079: PetscErrorCode MatSetBlockSize(Mat mat,PetscInt bs)
7080: {
7086: MatSetBlockSizes(mat,bs,bs);
7087: return(0);
7088: }
7090: /*@
7091: MatSetVariableBlockSizes - Sets a diagonal blocks of the matrix that need not be of the same size
7093: Logically Collective on Mat
7095: Input Parameters:
7096: + mat - the matrix
7097: . nblocks - the number of blocks on this process
7098: - bsizes - the block sizes
7100: Notes:
7101: Currently used by PCVPBJACOBI for SeqAIJ matrices
7103: Level: intermediate
7105: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes(), MatGetVariableBlockSizes()
7106: @*/
7107: PetscErrorCode MatSetVariableBlockSizes(Mat mat,PetscInt nblocks,PetscInt *bsizes)
7108: {
7110: PetscInt i,ncnt = 0, nlocal;
7114: if (nblocks < 0) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"Number of local blocks must be great than or equal to zero");
7115: MatGetLocalSize(mat,&nlocal,NULL);
7116: for (i=0; i<nblocks; i++) ncnt += bsizes[i];
7117: 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);
7118: PetscFree(mat->bsizes);
7119: mat->nblocks = nblocks;
7120: PetscMalloc1(nblocks,&mat->bsizes);
7121: PetscArraycpy(mat->bsizes,bsizes,nblocks);
7122: return(0);
7123: }
7125: /*@C
7126: MatGetVariableBlockSizes - Gets a diagonal blocks of the matrix that need not be of the same size
7128: Logically Collective on Mat
7130: Input Parameters:
7131: . mat - the matrix
7133: Output Parameters:
7134: + nblocks - the number of blocks on this process
7135: - bsizes - the block sizes
7137: Notes: Currently not supported from Fortran
7139: Level: intermediate
7141: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes(), MatGetBlockSizes(), MatSetVariableBlockSizes()
7142: @*/
7143: PetscErrorCode MatGetVariableBlockSizes(Mat mat,PetscInt *nblocks,const PetscInt **bsizes)
7144: {
7147: *nblocks = mat->nblocks;
7148: *bsizes = mat->bsizes;
7149: return(0);
7150: }
7152: /*@
7153: MatSetBlockSizes - Sets the matrix block row and column sizes.
7155: Logically Collective on Mat
7157: Input Parameters:
7158: + mat - the matrix
7159: - rbs - row block size
7160: - cbs - column block size
7162: Notes:
7163: Block row formats are MATSEQBAIJ, MATMPIBAIJ, MATSEQSBAIJ, MATMPISBAIJ. These formats ALWAYS have square block storage in the matrix.
7164: If you pass a different block size for the columns than the rows, the row block size determines the square block storage.
7165: This must be called before MatSetUp() or MatXXXSetPreallocation() (or will default to 1) and the block size cannot be changed later
7167: For MATMPIAIJ and MATSEQAIJ matrix formats, this function can be called at a later stage, provided that the specified block sizes
7168: are compatible with the matrix local sizes.
7170: The row and column block size determine the blocksize of the "row" and "column" vectors returned by MatCreateVecs().
7172: Level: intermediate
7174: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSize(), MatGetBlockSizes()
7175: @*/
7176: PetscErrorCode MatSetBlockSizes(Mat mat,PetscInt rbs,PetscInt cbs)
7177: {
7184: if (mat->ops->setblocksizes) {
7185: (*mat->ops->setblocksizes)(mat,rbs,cbs);
7186: }
7187: if (mat->rmap->refcnt) {
7188: ISLocalToGlobalMapping l2g = NULL;
7189: PetscLayout nmap = NULL;
7191: PetscLayoutDuplicate(mat->rmap,&nmap);
7192: if (mat->rmap->mapping) {
7193: ISLocalToGlobalMappingDuplicate(mat->rmap->mapping,&l2g);
7194: }
7195: PetscLayoutDestroy(&mat->rmap);
7196: mat->rmap = nmap;
7197: mat->rmap->mapping = l2g;
7198: }
7199: if (mat->cmap->refcnt) {
7200: ISLocalToGlobalMapping l2g = NULL;
7201: PetscLayout nmap = NULL;
7203: PetscLayoutDuplicate(mat->cmap,&nmap);
7204: if (mat->cmap->mapping) {
7205: ISLocalToGlobalMappingDuplicate(mat->cmap->mapping,&l2g);
7206: }
7207: PetscLayoutDestroy(&mat->cmap);
7208: mat->cmap = nmap;
7209: mat->cmap->mapping = l2g;
7210: }
7211: PetscLayoutSetBlockSize(mat->rmap,rbs);
7212: PetscLayoutSetBlockSize(mat->cmap,cbs);
7213: return(0);
7214: }
7216: /*@
7217: MatSetBlockSizesFromMats - Sets the matrix block row and column sizes to match a pair of matrices
7219: Logically Collective on Mat
7221: Input Parameters:
7222: + mat - the matrix
7223: . fromRow - matrix from which to copy row block size
7224: - fromCol - matrix from which to copy column block size (can be same as fromRow)
7226: Level: developer
7228: .seealso: MatCreateSeqBAIJ(), MatCreateBAIJ(), MatGetBlockSize(), MatSetBlockSizes()
7229: @*/
7230: PetscErrorCode MatSetBlockSizesFromMats(Mat mat,Mat fromRow,Mat fromCol)
7231: {
7238: if (fromRow->rmap->bs > 0) {PetscLayoutSetBlockSize(mat->rmap,fromRow->rmap->bs);}
7239: if (fromCol->cmap->bs > 0) {PetscLayoutSetBlockSize(mat->cmap,fromCol->cmap->bs);}
7240: return(0);
7241: }
7243: /*@
7244: MatResidual - Default routine to calculate the residual.
7246: Collective on Mat
7248: Input Parameters:
7249: + mat - the matrix
7250: . b - the right-hand-side
7251: - x - the approximate solution
7253: Output Parameter:
7254: . r - location to store the residual
7256: Level: developer
7258: .seealso: PCMGSetResidual()
7259: @*/
7260: PetscErrorCode MatResidual(Mat mat,Vec b,Vec x,Vec r)
7261: {
7270: MatCheckPreallocated(mat,1);
7271: PetscLogEventBegin(MAT_Residual,mat,0,0,0);
7272: if (!mat->ops->residual) {
7273: MatMult(mat,x,r);
7274: VecAYPX(r,-1.0,b);
7275: } else {
7276: (*mat->ops->residual)(mat,b,x,r);
7277: }
7278: PetscLogEventEnd(MAT_Residual,mat,0,0,0);
7279: return(0);
7280: }
7282: /*@C
7283: MatGetRowIJ - Returns the compressed row storage i and j indices for sequential matrices.
7285: Collective on Mat
7287: Input Parameters:
7288: + mat - the matrix
7289: . shift - 0 or 1 indicating we want the indices starting at 0 or 1
7290: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be symmetrized
7291: - inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7292: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7293: always used.
7295: Output Parameters:
7296: + n - number of rows in the (possibly compressed) matrix
7297: . ia - the row pointers; that is ia[0] = 0, ia[row] = ia[row-1] + number of elements in that row of the matrix
7298: . ja - the column indices
7299: - done - indicates if the routine actually worked and returned appropriate ia[] and ja[] arrays; callers
7300: are responsible for handling the case when done == PETSC_FALSE and ia and ja are not set
7302: Level: developer
7304: Notes:
7305: You CANNOT change any of the ia[] or ja[] values.
7307: Use MatRestoreRowIJ() when you are finished accessing the ia[] and ja[] values.
7309: Fortran Notes:
7310: In Fortran use
7311: $
7312: $ PetscInt ia(1), ja(1)
7313: $ PetscOffset iia, jja
7314: $ call MatGetRowIJ(mat,shift,symmetric,inodecompressed,n,ia,iia,ja,jja,done,ierr)
7315: $ ! Access the ith and jth entries via ia(iia + i) and ja(jja + j)
7317: or
7318: $
7319: $ PetscInt, pointer :: ia(:),ja(:)
7320: $ call MatGetRowIJF90(mat,shift,symmetric,inodecompressed,n,ia,ja,done,ierr)
7321: $ ! Access the ith and jth entries via ia(i) and ja(j)
7323: .seealso: MatGetColumnIJ(), MatRestoreRowIJ(), MatSeqAIJGetArray()
7324: @*/
7325: PetscErrorCode MatGetRowIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7326: {
7336: MatCheckPreallocated(mat,1);
7337: if (!mat->ops->getrowij) *done = PETSC_FALSE;
7338: else {
7339: *done = PETSC_TRUE;
7340: PetscLogEventBegin(MAT_GetRowIJ,mat,0,0,0);
7341: (*mat->ops->getrowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7342: PetscLogEventEnd(MAT_GetRowIJ,mat,0,0,0);
7343: }
7344: return(0);
7345: }
7347: /*@C
7348: MatGetColumnIJ - Returns the compressed column storage i and j indices for sequential matrices.
7350: Collective on Mat
7352: Input Parameters:
7353: + mat - the matrix
7354: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7355: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7356: symmetrized
7357: . inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7358: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7359: always used.
7360: . n - number of columns in the (possibly compressed) matrix
7361: . ia - the column pointers; that is ia[0] = 0, ia[col] = i[col-1] + number of elements in that col of the matrix
7362: - ja - the row indices
7364: Output Parameters:
7365: . done - PETSC_TRUE or PETSC_FALSE, indicating whether the values have been returned
7367: Level: developer
7369: .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
7370: @*/
7371: PetscErrorCode MatGetColumnIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7372: {
7382: MatCheckPreallocated(mat,1);
7383: if (!mat->ops->getcolumnij) *done = PETSC_FALSE;
7384: else {
7385: *done = PETSC_TRUE;
7386: (*mat->ops->getcolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7387: }
7388: return(0);
7389: }
7391: /*@C
7392: MatRestoreRowIJ - Call after you are completed with the ia,ja indices obtained with
7393: MatGetRowIJ().
7395: Collective on Mat
7397: Input Parameters:
7398: + mat - the matrix
7399: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7400: . symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7401: symmetrized
7402: . inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7403: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7404: always used.
7405: . n - size of (possibly compressed) matrix
7406: . ia - the row pointers
7407: - ja - the column indices
7409: Output Parameters:
7410: . done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
7412: Note:
7413: This routine zeros out n, ia, and ja. This is to prevent accidental
7414: us of the array after it has been restored. If you pass NULL, it will
7415: not zero the pointers. Use of ia or ja after MatRestoreRowIJ() is invalid.
7417: Level: developer
7419: .seealso: MatGetRowIJ(), MatRestoreColumnIJ()
7420: @*/
7421: PetscErrorCode MatRestoreRowIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7422: {
7431: MatCheckPreallocated(mat,1);
7433: if (!mat->ops->restorerowij) *done = PETSC_FALSE;
7434: else {
7435: *done = PETSC_TRUE;
7436: (*mat->ops->restorerowij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7437: if (n) *n = 0;
7438: if (ia) *ia = NULL;
7439: if (ja) *ja = NULL;
7440: }
7441: return(0);
7442: }
7444: /*@C
7445: MatRestoreColumnIJ - Call after you are completed with the ia,ja indices obtained with
7446: MatGetColumnIJ().
7448: Collective on Mat
7450: Input Parameters:
7451: + mat - the matrix
7452: . shift - 1 or zero indicating we want the indices starting at 0 or 1
7453: - symmetric - PETSC_TRUE or PETSC_FALSE indicating the matrix data structure should be
7454: symmetrized
7455: - inodecompressed - PETSC_TRUE or PETSC_FALSE indicating if the nonzero structure of the
7456: inodes or the nonzero elements is wanted. For BAIJ matrices the compressed version is
7457: always used.
7459: Output Parameters:
7460: + n - size of (possibly compressed) matrix
7461: . ia - the column pointers
7462: . ja - the row indices
7463: - done - PETSC_TRUE or PETSC_FALSE indicated that the values have been returned
7465: Level: developer
7467: .seealso: MatGetColumnIJ(), MatRestoreRowIJ()
7468: @*/
7469: PetscErrorCode MatRestoreColumnIJ(Mat mat,PetscInt shift,PetscBool symmetric,PetscBool inodecompressed,PetscInt *n,const PetscInt *ia[],const PetscInt *ja[],PetscBool *done)
7470: {
7479: MatCheckPreallocated(mat,1);
7481: if (!mat->ops->restorecolumnij) *done = PETSC_FALSE;
7482: else {
7483: *done = PETSC_TRUE;
7484: (*mat->ops->restorecolumnij)(mat,shift,symmetric,inodecompressed,n,ia,ja,done);
7485: if (n) *n = 0;
7486: if (ia) *ia = NULL;
7487: if (ja) *ja = NULL;
7488: }
7489: return(0);
7490: }
7492: /*@C
7493: MatColoringPatch -Used inside matrix coloring routines that
7494: use MatGetRowIJ() and/or MatGetColumnIJ().
7496: Collective on Mat
7498: Input Parameters:
7499: + mat - the matrix
7500: . ncolors - max color value
7501: . n - number of entries in colorarray
7502: - colorarray - array indicating color for each column
7504: Output Parameters:
7505: . iscoloring - coloring generated using colorarray information
7507: Level: developer
7509: .seealso: MatGetRowIJ(), MatGetColumnIJ()
7511: @*/
7512: PetscErrorCode MatColoringPatch(Mat mat,PetscInt ncolors,PetscInt n,ISColoringValue colorarray[],ISColoring *iscoloring)
7513: {
7521: MatCheckPreallocated(mat,1);
7523: if (!mat->ops->coloringpatch) {
7524: ISColoringCreate(PetscObjectComm((PetscObject)mat),ncolors,n,colorarray,PETSC_OWN_POINTER,iscoloring);
7525: } else {
7526: (*mat->ops->coloringpatch)(mat,ncolors,n,colorarray,iscoloring);
7527: }
7528: return(0);
7529: }
7532: /*@
7533: MatSetUnfactored - Resets a factored matrix to be treated as unfactored.
7535: Logically Collective on Mat
7537: Input Parameter:
7538: . mat - the factored matrix to be reset
7540: Notes:
7541: This routine should be used only with factored matrices formed by in-place
7542: factorization via ILU(0) (or by in-place LU factorization for the MATSEQDENSE
7543: format). This option can save memory, for example, when solving nonlinear
7544: systems with a matrix-free Newton-Krylov method and a matrix-based, in-place
7545: ILU(0) preconditioner.
7547: Note that one can specify in-place ILU(0) factorization by calling
7548: .vb
7549: PCType(pc,PCILU);
7550: PCFactorSeUseInPlace(pc);
7551: .ve
7552: or by using the options -pc_type ilu -pc_factor_in_place
7554: In-place factorization ILU(0) can also be used as a local
7555: solver for the blocks within the block Jacobi or additive Schwarz
7556: methods (runtime option: -sub_pc_factor_in_place). See Users-Manual: ch_pc
7557: for details on setting local solver options.
7559: Most users should employ the simplified KSP interface for linear solvers
7560: instead of working directly with matrix algebra routines such as this.
7561: See, e.g., KSPCreate().
7563: Level: developer
7565: .seealso: PCFactorSetUseInPlace(), PCFactorGetUseInPlace()
7567: @*/
7568: PetscErrorCode MatSetUnfactored(Mat mat)
7569: {
7575: MatCheckPreallocated(mat,1);
7576: mat->factortype = MAT_FACTOR_NONE;
7577: if (!mat->ops->setunfactored) return(0);
7578: (*mat->ops->setunfactored)(mat);
7579: return(0);
7580: }
7582: /*MC
7583: MatDenseGetArrayF90 - Accesses a matrix array from Fortran90.
7585: Synopsis:
7586: MatDenseGetArrayF90(Mat x,{Scalar, pointer :: xx_v(:,:)},integer ierr)
7588: Not collective
7590: Input Parameter:
7591: . x - matrix
7593: Output Parameters:
7594: + xx_v - the Fortran90 pointer to the array
7595: - ierr - error code
7597: Example of Usage:
7598: .vb
7599: PetscScalar, pointer xx_v(:,:)
7600: ....
7601: call MatDenseGetArrayF90(x,xx_v,ierr)
7602: a = xx_v(3)
7603: call MatDenseRestoreArrayF90(x,xx_v,ierr)
7604: .ve
7606: Level: advanced
7608: .seealso: MatDenseRestoreArrayF90(), MatDenseGetArray(), MatDenseRestoreArray(), MatSeqAIJGetArrayF90()
7610: M*/
7612: /*MC
7613: MatDenseRestoreArrayF90 - Restores a matrix array that has been
7614: accessed with MatDenseGetArrayF90().
7616: Synopsis:
7617: MatDenseRestoreArrayF90(Mat x,{Scalar, pointer :: xx_v(:,:)},integer ierr)
7619: Not collective
7621: Input Parameters:
7622: + x - matrix
7623: - xx_v - the Fortran90 pointer to the array
7625: Output Parameter:
7626: . ierr - error code
7628: Example of Usage:
7629: .vb
7630: PetscScalar, pointer xx_v(:,:)
7631: ....
7632: call MatDenseGetArrayF90(x,xx_v,ierr)
7633: a = xx_v(3)
7634: call MatDenseRestoreArrayF90(x,xx_v,ierr)
7635: .ve
7637: Level: advanced
7639: .seealso: MatDenseGetArrayF90(), MatDenseGetArray(), MatDenseRestoreArray(), MatSeqAIJRestoreArrayF90()
7641: M*/
7644: /*MC
7645: MatSeqAIJGetArrayF90 - Accesses a matrix array from Fortran90.
7647: Synopsis:
7648: MatSeqAIJGetArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
7650: Not collective
7652: Input Parameter:
7653: . x - matrix
7655: Output Parameters:
7656: + xx_v - the Fortran90 pointer to the array
7657: - ierr - error code
7659: Example of Usage:
7660: .vb
7661: PetscScalar, pointer xx_v(:)
7662: ....
7663: call MatSeqAIJGetArrayF90(x,xx_v,ierr)
7664: a = xx_v(3)
7665: call MatSeqAIJRestoreArrayF90(x,xx_v,ierr)
7666: .ve
7668: Level: advanced
7670: .seealso: MatSeqAIJRestoreArrayF90(), MatSeqAIJGetArray(), MatSeqAIJRestoreArray(), MatDenseGetArrayF90()
7672: M*/
7674: /*MC
7675: MatSeqAIJRestoreArrayF90 - Restores a matrix array that has been
7676: accessed with MatSeqAIJGetArrayF90().
7678: Synopsis:
7679: MatSeqAIJRestoreArrayF90(Mat x,{Scalar, pointer :: xx_v(:)},integer ierr)
7681: Not collective
7683: Input Parameters:
7684: + x - matrix
7685: - xx_v - the Fortran90 pointer to the array
7687: Output Parameter:
7688: . ierr - error code
7690: Example of Usage:
7691: .vb
7692: PetscScalar, pointer xx_v(:)
7693: ....
7694: call MatSeqAIJGetArrayF90(x,xx_v,ierr)
7695: a = xx_v(3)
7696: call MatSeqAIJRestoreArrayF90(x,xx_v,ierr)
7697: .ve
7699: Level: advanced
7701: .seealso: MatSeqAIJGetArrayF90(), MatSeqAIJGetArray(), MatSeqAIJRestoreArray(), MatDenseRestoreArrayF90()
7703: M*/
7706: /*@
7707: MatCreateSubMatrix - Gets a single submatrix on the same number of processors
7708: as the original matrix.
7710: Collective on Mat
7712: Input Parameters:
7713: + mat - the original matrix
7714: . isrow - parallel IS containing the rows this processor should obtain
7715: . 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.
7716: - cll - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
7718: Output Parameter:
7719: . newmat - the new submatrix, of the same type as the old
7721: Level: advanced
7723: Notes:
7724: The submatrix will be able to be multiplied with vectors using the same layout as iscol.
7726: Some matrix types place restrictions on the row and column indices, such
7727: as that they be sorted or that they be equal to each other.
7729: The index sets may not have duplicate entries.
7731: The first time this is called you should use a cll of MAT_INITIAL_MATRIX,
7732: the MatCreateSubMatrix() routine will create the newmat for you. Any additional calls
7733: to this routine with a mat of the same nonzero structure and with a call of MAT_REUSE_MATRIX
7734: will reuse the matrix generated the first time. You should call MatDestroy() on newmat when
7735: you are finished using it.
7737: The communicator of the newly obtained matrix is ALWAYS the same as the communicator of
7738: the input matrix.
7740: If iscol is NULL then all columns are obtained (not supported in Fortran).
7742: Example usage:
7743: Consider the following 8x8 matrix with 34 non-zero values, that is
7744: assembled across 3 processors. Let's assume that proc0 owns 3 rows,
7745: proc1 owns 3 rows, proc2 owns 2 rows. This division can be shown
7746: as follows:
7748: .vb
7749: 1 2 0 | 0 3 0 | 0 4
7750: Proc0 0 5 6 | 7 0 0 | 8 0
7751: 9 0 10 | 11 0 0 | 12 0
7752: -------------------------------------
7753: 13 0 14 | 15 16 17 | 0 0
7754: Proc1 0 18 0 | 19 20 21 | 0 0
7755: 0 0 0 | 22 23 0 | 24 0
7756: -------------------------------------
7757: Proc2 25 26 27 | 0 0 28 | 29 0
7758: 30 0 0 | 31 32 33 | 0 34
7759: .ve
7761: Suppose isrow = [0 1 | 4 | 6 7] and iscol = [1 2 | 3 4 5 | 6]. The resulting submatrix is
7763: .vb
7764: 2 0 | 0 3 0 | 0
7765: Proc0 5 6 | 7 0 0 | 8
7766: -------------------------------
7767: Proc1 18 0 | 19 20 21 | 0
7768: -------------------------------
7769: Proc2 26 27 | 0 0 28 | 29
7770: 0 0 | 31 32 33 | 0
7771: .ve
7774: .seealso: MatCreateSubMatrices(), MatCreateSubMatricesMPI(), MatCreateSubMatrixVirtual(), MatSubMatrixVirtualUpdate()
7775: @*/
7776: PetscErrorCode MatCreateSubMatrix(Mat mat,IS isrow,IS iscol,MatReuse cll,Mat *newmat)
7777: {
7779: PetscMPIInt size;
7780: Mat *local;
7781: IS iscoltmp;
7790: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
7791: if (cll == MAT_IGNORE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Cannot use MAT_IGNORE_MATRIX");
7793: MatCheckPreallocated(mat,1);
7794: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
7796: if (!iscol || isrow == iscol) {
7797: PetscBool stride;
7798: PetscMPIInt grabentirematrix = 0,grab;
7799: PetscObjectTypeCompare((PetscObject)isrow,ISSTRIDE,&stride);
7800: if (stride) {
7801: PetscInt first,step,n,rstart,rend;
7802: ISStrideGetInfo(isrow,&first,&step);
7803: if (step == 1) {
7804: MatGetOwnershipRange(mat,&rstart,&rend);
7805: if (rstart == first) {
7806: ISGetLocalSize(isrow,&n);
7807: if (n == rend-rstart) {
7808: grabentirematrix = 1;
7809: }
7810: }
7811: }
7812: }
7813: MPIU_Allreduce(&grabentirematrix,&grab,1,MPI_INT,MPI_MIN,PetscObjectComm((PetscObject)mat));
7814: if (grab) {
7815: PetscInfo(mat,"Getting entire matrix as submatrix\n");
7816: if (cll == MAT_INITIAL_MATRIX) {
7817: *newmat = mat;
7818: PetscObjectReference((PetscObject)mat);
7819: }
7820: return(0);
7821: }
7822: }
7824: if (!iscol) {
7825: ISCreateStride(PetscObjectComm((PetscObject)mat),mat->cmap->n,mat->cmap->rstart,1,&iscoltmp);
7826: } else {
7827: iscoltmp = iscol;
7828: }
7830: /* if original matrix is on just one processor then use submatrix generated */
7831: if (mat->ops->createsubmatrices && !mat->ops->createsubmatrix && size == 1 && cll == MAT_REUSE_MATRIX) {
7832: MatCreateSubMatrices(mat,1,&isrow,&iscoltmp,MAT_REUSE_MATRIX,&newmat);
7833: goto setproperties;
7834: } else if (mat->ops->createsubmatrices && !mat->ops->createsubmatrix && size == 1) {
7835: MatCreateSubMatrices(mat,1,&isrow,&iscoltmp,MAT_INITIAL_MATRIX,&local);
7836: *newmat = *local;
7837: PetscFree(local);
7838: goto setproperties;
7839: } else if (!mat->ops->createsubmatrix) {
7840: /* Create a new matrix type that implements the operation using the full matrix */
7841: PetscLogEventBegin(MAT_CreateSubMat,mat,0,0,0);
7842: switch (cll) {
7843: case MAT_INITIAL_MATRIX:
7844: MatCreateSubMatrixVirtual(mat,isrow,iscoltmp,newmat);
7845: break;
7846: case MAT_REUSE_MATRIX:
7847: MatSubMatrixVirtualUpdate(*newmat,mat,isrow,iscoltmp);
7848: break;
7849: default: SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"Invalid MatReuse, must be either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX");
7850: }
7851: PetscLogEventEnd(MAT_CreateSubMat,mat,0,0,0);
7852: goto setproperties;
7853: }
7855: if (!mat->ops->createsubmatrix) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
7856: PetscLogEventBegin(MAT_CreateSubMat,mat,0,0,0);
7857: (*mat->ops->createsubmatrix)(mat,isrow,iscoltmp,cll,newmat);
7858: PetscLogEventEnd(MAT_CreateSubMat,mat,0,0,0);
7860: /* Propagate symmetry information for diagonal blocks */
7861: setproperties:
7862: if (isrow == iscoltmp) {
7863: if (mat->symmetric_set && mat->symmetric) {
7864: MatSetOption(*newmat,MAT_SYMMETRIC,PETSC_TRUE);
7865: }
7866: if (mat->structurally_symmetric_set && mat->structurally_symmetric) {
7867: MatSetOption(*newmat,MAT_STRUCTURALLY_SYMMETRIC,PETSC_TRUE);
7868: }
7869: if (mat->hermitian_set && mat->hermitian) {
7870: MatSetOption(*newmat,MAT_HERMITIAN,PETSC_TRUE);
7871: }
7872: if (mat->spd_set && mat->spd) {
7873: MatSetOption(*newmat,MAT_SPD,PETSC_TRUE);
7874: }
7875: }
7877: if (!iscol) {ISDestroy(&iscoltmp);}
7878: if (*newmat && cll == MAT_INITIAL_MATRIX) {PetscObjectStateIncrease((PetscObject)*newmat);}
7879: return(0);
7880: }
7882: /*@
7883: MatStashSetInitialSize - sets the sizes of the matrix stash, that is
7884: used during the assembly process to store values that belong to
7885: other processors.
7887: Not Collective
7889: Input Parameters:
7890: + mat - the matrix
7891: . size - the initial size of the stash.
7892: - bsize - the initial size of the block-stash(if used).
7894: Options Database Keys:
7895: + -matstash_initial_size <size> or <size0,size1,...sizep-1>
7896: - -matstash_block_initial_size <bsize> or <bsize0,bsize1,...bsizep-1>
7898: Level: intermediate
7900: Notes:
7901: The block-stash is used for values set with MatSetValuesBlocked() while
7902: the stash is used for values set with MatSetValues()
7904: Run with the option -info and look for output of the form
7905: MatAssemblyBegin_MPIXXX:Stash has MM entries, uses nn mallocs.
7906: to determine the appropriate value, MM, to use for size and
7907: MatAssemblyBegin_MPIXXX:Block-Stash has BMM entries, uses nn mallocs.
7908: to determine the value, BMM to use for bsize
7911: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashGetInfo()
7913: @*/
7914: PetscErrorCode MatStashSetInitialSize(Mat mat,PetscInt size, PetscInt bsize)
7915: {
7921: MatStashSetInitialSize_Private(&mat->stash,size);
7922: MatStashSetInitialSize_Private(&mat->bstash,bsize);
7923: return(0);
7924: }
7926: /*@
7927: MatInterpolateAdd - w = y + A*x or A'*x depending on the shape of
7928: the matrix
7930: Neighbor-wise Collective on Mat
7932: Input Parameters:
7933: + mat - the matrix
7934: . x,y - the vectors
7935: - w - where the result is stored
7937: Level: intermediate
7939: Notes:
7940: w may be the same vector as y.
7942: This allows one to use either the restriction or interpolation (its transpose)
7943: matrix to do the interpolation
7945: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
7947: @*/
7948: PetscErrorCode MatInterpolateAdd(Mat A,Vec x,Vec y,Vec w)
7949: {
7951: PetscInt M,N,Ny;
7959: MatCheckPreallocated(A,1);
7960: MatGetSize(A,&M,&N);
7961: VecGetSize(y,&Ny);
7962: if (M == Ny) {
7963: MatMultAdd(A,x,y,w);
7964: } else {
7965: MatMultTransposeAdd(A,x,y,w);
7966: }
7967: return(0);
7968: }
7970: /*@
7971: MatInterpolate - y = A*x or A'*x depending on the shape of
7972: the matrix
7974: Neighbor-wise Collective on Mat
7976: Input Parameters:
7977: + mat - the matrix
7978: - x,y - the vectors
7980: Level: intermediate
7982: Notes:
7983: This allows one to use either the restriction or interpolation (its transpose)
7984: matrix to do the interpolation
7986: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatRestrict()
7988: @*/
7989: PetscErrorCode MatInterpolate(Mat A,Vec x,Vec y)
7990: {
7992: PetscInt M,N,Ny;
7999: MatCheckPreallocated(A,1);
8000: MatGetSize(A,&M,&N);
8001: VecGetSize(y,&Ny);
8002: if (M == Ny) {
8003: MatMult(A,x,y);
8004: } else {
8005: MatMultTranspose(A,x,y);
8006: }
8007: return(0);
8008: }
8010: /*@
8011: MatRestrict - y = A*x or A'*x
8013: Neighbor-wise Collective on Mat
8015: Input Parameters:
8016: + mat - the matrix
8017: - x,y - the vectors
8019: Level: intermediate
8021: Notes:
8022: This allows one to use either the restriction or interpolation (its transpose)
8023: matrix to do the restriction
8025: .seealso: MatMultAdd(), MatMultTransposeAdd(), MatInterpolate()
8027: @*/
8028: PetscErrorCode MatRestrict(Mat A,Vec x,Vec y)
8029: {
8031: PetscInt M,N,Ny;
8038: MatCheckPreallocated(A,1);
8040: MatGetSize(A,&M,&N);
8041: VecGetSize(y,&Ny);
8042: if (M == Ny) {
8043: MatMult(A,x,y);
8044: } else {
8045: MatMultTranspose(A,x,y);
8046: }
8047: return(0);
8048: }
8050: /*@
8051: MatGetNullSpace - retrieves the null space of a matrix.
8053: Logically Collective on Mat
8055: Input Parameters:
8056: + mat - the matrix
8057: - nullsp - the null space object
8059: Level: developer
8061: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatSetNullSpace()
8062: @*/
8063: PetscErrorCode MatGetNullSpace(Mat mat, MatNullSpace *nullsp)
8064: {
8068: *nullsp = (mat->symmetric_set && mat->symmetric && !mat->nullsp) ? mat->transnullsp : mat->nullsp;
8069: return(0);
8070: }
8072: /*@
8073: MatSetNullSpace - attaches a null space to a matrix.
8075: Logically Collective on Mat
8077: Input Parameters:
8078: + mat - the matrix
8079: - nullsp - the null space object
8081: Level: advanced
8083: Notes:
8084: This null space is used by the linear solvers. Overwrites any previous null space that may have been attached
8086: For inconsistent singular systems (linear systems where the right hand side is not in the range of the operator) you also likely should
8087: call MatSetTransposeNullSpace(). This allows the linear system to be solved in a least squares sense.
8089: You can remove the null space by calling this routine with an nullsp of NULL
8092: The fundamental theorem of linear algebra (Gilbert Strang, Introduction to Applied Mathematics, page 72) states that
8093: 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).
8094: 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
8095: 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
8096: 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).
8098: Krylov solvers can produce the minimal norm solution to the least squares problem by utilizing MatNullSpaceRemove().
8100: 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
8101: routine also automatically calls MatSetTransposeNullSpace().
8103: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatGetNullSpace(), MatSetTransposeNullSpace(), MatGetTransposeNullSpace(), MatNullSpaceRemove()
8104: @*/
8105: PetscErrorCode MatSetNullSpace(Mat mat,MatNullSpace nullsp)
8106: {
8112: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8113: MatNullSpaceDestroy(&mat->nullsp);
8114: mat->nullsp = nullsp;
8115: if (mat->symmetric_set && mat->symmetric) {
8116: MatSetTransposeNullSpace(mat,nullsp);
8117: }
8118: return(0);
8119: }
8121: /*@
8122: MatGetTransposeNullSpace - retrieves the null space of the transpose of a matrix.
8124: Logically Collective on Mat
8126: Input Parameters:
8127: + mat - the matrix
8128: - nullsp - the null space object
8130: Level: developer
8132: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatSetTransposeNullSpace(), MatSetNullSpace(), MatGetNullSpace()
8133: @*/
8134: PetscErrorCode MatGetTransposeNullSpace(Mat mat, MatNullSpace *nullsp)
8135: {
8140: *nullsp = (mat->symmetric_set && mat->symmetric && !mat->transnullsp) ? mat->nullsp : mat->transnullsp;
8141: return(0);
8142: }
8144: /*@
8145: MatSetTransposeNullSpace - attaches a null space to a matrix.
8147: Logically Collective on Mat
8149: Input Parameters:
8150: + mat - the matrix
8151: - nullsp - the null space object
8153: Level: advanced
8155: Notes:
8156: 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.
8157: You must also call MatSetNullSpace()
8160: The fundamental theorem of linear algebra (Gilbert Strang, Introduction to Applied Mathematics, page 72) states that
8161: 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).
8162: 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
8163: 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
8164: 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).
8166: Krylov solvers can produce the minimal norm solution to the least squares problem by utilizing MatNullSpaceRemove().
8168: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNearNullSpace(), MatGetNullSpace(), MatSetNullSpace(), MatGetTransposeNullSpace(), MatNullSpaceRemove()
8169: @*/
8170: PetscErrorCode MatSetTransposeNullSpace(Mat mat,MatNullSpace nullsp)
8171: {
8177: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8178: MatNullSpaceDestroy(&mat->transnullsp);
8179: mat->transnullsp = nullsp;
8180: return(0);
8181: }
8183: /*@
8184: MatSetNearNullSpace - attaches a null space to a matrix, which is often the null space (rigid body modes) of the operator without boundary conditions
8185: This null space will be used to provide near null space vectors to a multigrid preconditioner built from this matrix.
8187: Logically Collective on Mat
8189: Input Parameters:
8190: + mat - the matrix
8191: - nullsp - the null space object
8193: Level: advanced
8195: Notes:
8196: Overwrites any previous near null space that may have been attached
8198: You can remove the null space by calling this routine with an nullsp of NULL
8200: .seealso: MatCreate(), MatNullSpaceCreate(), MatSetNullSpace(), MatNullSpaceCreateRigidBody(), MatGetNearNullSpace()
8201: @*/
8202: PetscErrorCode MatSetNearNullSpace(Mat mat,MatNullSpace nullsp)
8203: {
8210: MatCheckPreallocated(mat,1);
8211: if (nullsp) {PetscObjectReference((PetscObject)nullsp);}
8212: MatNullSpaceDestroy(&mat->nearnullsp);
8213: mat->nearnullsp = nullsp;
8214: return(0);
8215: }
8217: /*@
8218: MatGetNearNullSpace -Get null space attached with MatSetNearNullSpace()
8220: Not Collective
8222: Input Parameters:
8223: . mat - the matrix
8225: Output Parameters:
8226: . nullsp - the null space object, NULL if not set
8228: Level: developer
8230: .seealso: MatSetNearNullSpace(), MatGetNullSpace(), MatNullSpaceCreate()
8231: @*/
8232: PetscErrorCode MatGetNearNullSpace(Mat mat,MatNullSpace *nullsp)
8233: {
8238: MatCheckPreallocated(mat,1);
8239: *nullsp = mat->nearnullsp;
8240: return(0);
8241: }
8243: /*@C
8244: MatICCFactor - Performs in-place incomplete Cholesky factorization of matrix.
8246: Collective on Mat
8248: Input Parameters:
8249: + mat - the matrix
8250: . row - row/column permutation
8251: . fill - expected fill factor >= 1.0
8252: - level - level of fill, for ICC(k)
8254: Notes:
8255: Probably really in-place only when level of fill is zero, otherwise allocates
8256: new space to store factored matrix and deletes previous memory.
8258: Most users should employ the simplified KSP interface for linear solvers
8259: instead of working directly with matrix algebra routines such as this.
8260: See, e.g., KSPCreate().
8262: Level: developer
8265: .seealso: MatICCFactorSymbolic(), MatLUFactorNumeric(), MatCholeskyFactor()
8267: Developer Note: fortran interface is not autogenerated as the f90
8268: interface defintion cannot be generated correctly [due to MatFactorInfo]
8270: @*/
8271: PetscErrorCode MatICCFactor(Mat mat,IS row,const MatFactorInfo *info)
8272: {
8280: if (mat->rmap->N != mat->cmap->N) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONG,"matrix must be square");
8281: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
8282: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
8283: if (!mat->ops->iccfactor) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8284: MatCheckPreallocated(mat,1);
8285: (*mat->ops->iccfactor)(mat,row,info);
8286: PetscObjectStateIncrease((PetscObject)mat);
8287: return(0);
8288: }
8290: /*@
8291: MatDiagonalScaleLocal - Scales columns of a matrix given the scaling values including the
8292: ghosted ones.
8294: Not Collective
8296: Input Parameters:
8297: + mat - the matrix
8298: - diag = the diagonal values, including ghost ones
8300: Level: developer
8302: Notes:
8303: Works only for MPIAIJ and MPIBAIJ matrices
8305: .seealso: MatDiagonalScale()
8306: @*/
8307: PetscErrorCode MatDiagonalScaleLocal(Mat mat,Vec diag)
8308: {
8310: PetscMPIInt size;
8317: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Matrix must be already assembled");
8318: PetscLogEventBegin(MAT_Scale,mat,0,0,0);
8319: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
8320: if (size == 1) {
8321: PetscInt n,m;
8322: VecGetSize(diag,&n);
8323: MatGetSize(mat,0,&m);
8324: if (m == n) {
8325: MatDiagonalScale(mat,0,diag);
8326: } else SETERRQ(PETSC_COMM_SELF,PETSC_ERR_SUP,"Only supported for sequential matrices when no ghost points/periodic conditions");
8327: } else {
8328: PetscUseMethod(mat,"MatDiagonalScaleLocal_C",(Mat,Vec),(mat,diag));
8329: }
8330: PetscLogEventEnd(MAT_Scale,mat,0,0,0);
8331: PetscObjectStateIncrease((PetscObject)mat);
8332: return(0);
8333: }
8335: /*@
8336: MatGetInertia - Gets the inertia from a factored matrix
8338: Collective on Mat
8340: Input Parameter:
8341: . mat - the matrix
8343: Output Parameters:
8344: + nneg - number of negative eigenvalues
8345: . nzero - number of zero eigenvalues
8346: - npos - number of positive eigenvalues
8348: Level: advanced
8350: Notes:
8351: Matrix must have been factored by MatCholeskyFactor()
8354: @*/
8355: PetscErrorCode MatGetInertia(Mat mat,PetscInt *nneg,PetscInt *nzero,PetscInt *npos)
8356: {
8362: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
8363: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Numeric factor mat is not assembled");
8364: if (!mat->ops->getinertia) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8365: (*mat->ops->getinertia)(mat,nneg,nzero,npos);
8366: return(0);
8367: }
8369: /* ----------------------------------------------------------------*/
8370: /*@C
8371: MatSolves - Solves A x = b, given a factored matrix, for a collection of vectors
8373: Neighbor-wise Collective on Mats
8375: Input Parameters:
8376: + mat - the factored matrix
8377: - b - the right-hand-side vectors
8379: Output Parameter:
8380: . x - the result vectors
8382: Notes:
8383: The vectors b and x cannot be the same. I.e., one cannot
8384: call MatSolves(A,x,x).
8386: Notes:
8387: Most users should employ the simplified KSP interface for linear solvers
8388: instead of working directly with matrix algebra routines such as this.
8389: See, e.g., KSPCreate().
8391: Level: developer
8393: .seealso: MatSolveAdd(), MatSolveTranspose(), MatSolveTransposeAdd(), MatSolve()
8394: @*/
8395: PetscErrorCode MatSolves(Mat mat,Vecs b,Vecs x)
8396: {
8402: if (x == b) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_IDN,"x and b must be different vectors");
8403: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Unfactored matrix");
8404: if (!mat->rmap->N && !mat->cmap->N) return(0);
8406: if (!mat->ops->solves) SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)mat)->type_name);
8407: MatCheckPreallocated(mat,1);
8408: PetscLogEventBegin(MAT_Solves,mat,0,0,0);
8409: (*mat->ops->solves)(mat,b,x);
8410: PetscLogEventEnd(MAT_Solves,mat,0,0,0);
8411: return(0);
8412: }
8414: /*@
8415: MatIsSymmetric - Test whether a matrix is symmetric
8417: Collective on Mat
8419: Input Parameter:
8420: + A - the matrix to test
8421: - tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact transpose)
8423: Output Parameters:
8424: . flg - the result
8426: Notes:
8427: For real numbers MatIsSymmetric() and MatIsHermitian() return identical results
8429: Level: intermediate
8431: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetricKnown()
8432: @*/
8433: PetscErrorCode MatIsSymmetric(Mat A,PetscReal tol,PetscBool *flg)
8434: {
8441: if (!A->symmetric_set) {
8442: if (!A->ops->issymmetric) {
8443: MatType mattype;
8444: MatGetType(A,&mattype);
8445: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for symmetric",mattype);
8446: }
8447: (*A->ops->issymmetric)(A,tol,flg);
8448: if (!tol) {
8449: A->symmetric_set = PETSC_TRUE;
8450: A->symmetric = *flg;
8451: if (A->symmetric) {
8452: A->structurally_symmetric_set = PETSC_TRUE;
8453: A->structurally_symmetric = PETSC_TRUE;
8454: }
8455: }
8456: } else if (A->symmetric) {
8457: *flg = PETSC_TRUE;
8458: } else if (!tol) {
8459: *flg = PETSC_FALSE;
8460: } else {
8461: if (!A->ops->issymmetric) {
8462: MatType mattype;
8463: MatGetType(A,&mattype);
8464: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for symmetric",mattype);
8465: }
8466: (*A->ops->issymmetric)(A,tol,flg);
8467: }
8468: return(0);
8469: }
8471: /*@
8472: MatIsHermitian - Test whether a matrix is Hermitian
8474: Collective on Mat
8476: Input Parameter:
8477: + A - the matrix to test
8478: - tol - difference between value and its transpose less than this amount counts as equal (use 0.0 for exact Hermitian)
8480: Output Parameters:
8481: . flg - the result
8483: Level: intermediate
8485: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(),
8486: MatIsSymmetricKnown(), MatIsSymmetric()
8487: @*/
8488: PetscErrorCode MatIsHermitian(Mat A,PetscReal tol,PetscBool *flg)
8489: {
8496: if (!A->hermitian_set) {
8497: if (!A->ops->ishermitian) {
8498: MatType mattype;
8499: MatGetType(A,&mattype);
8500: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for hermitian",mattype);
8501: }
8502: (*A->ops->ishermitian)(A,tol,flg);
8503: if (!tol) {
8504: A->hermitian_set = PETSC_TRUE;
8505: A->hermitian = *flg;
8506: if (A->hermitian) {
8507: A->structurally_symmetric_set = PETSC_TRUE;
8508: A->structurally_symmetric = PETSC_TRUE;
8509: }
8510: }
8511: } else if (A->hermitian) {
8512: *flg = PETSC_TRUE;
8513: } else if (!tol) {
8514: *flg = PETSC_FALSE;
8515: } else {
8516: if (!A->ops->ishermitian) {
8517: MatType mattype;
8518: MatGetType(A,&mattype);
8519: SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Matrix of type %s does not support checking for hermitian",mattype);
8520: }
8521: (*A->ops->ishermitian)(A,tol,flg);
8522: }
8523: return(0);
8524: }
8526: /*@
8527: MatIsSymmetricKnown - Checks the flag on the matrix to see if it is symmetric.
8529: Not Collective
8531: Input Parameter:
8532: . A - the matrix to check
8534: Output Parameters:
8535: + set - if the symmetric flag is set (this tells you if the next flag is valid)
8536: - flg - the result
8538: Level: advanced
8540: Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsSymmetric()
8541: if you want it explicitly checked
8543: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
8544: @*/
8545: PetscErrorCode MatIsSymmetricKnown(Mat A,PetscBool *set,PetscBool *flg)
8546: {
8551: if (A->symmetric_set) {
8552: *set = PETSC_TRUE;
8553: *flg = A->symmetric;
8554: } else {
8555: *set = PETSC_FALSE;
8556: }
8557: return(0);
8558: }
8560: /*@
8561: MatIsHermitianKnown - Checks the flag on the matrix to see if it is hermitian.
8563: Not Collective
8565: Input Parameter:
8566: . A - the matrix to check
8568: Output Parameters:
8569: + set - if the hermitian flag is set (this tells you if the next flag is valid)
8570: - flg - the result
8572: Level: advanced
8574: Note: Does not check the matrix values directly, so this may return unknown (set = PETSC_FALSE). Use MatIsHermitian()
8575: if you want it explicitly checked
8577: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsStructurallySymmetric(), MatSetOption(), MatIsSymmetric()
8578: @*/
8579: PetscErrorCode MatIsHermitianKnown(Mat A,PetscBool *set,PetscBool *flg)
8580: {
8585: if (A->hermitian_set) {
8586: *set = PETSC_TRUE;
8587: *flg = A->hermitian;
8588: } else {
8589: *set = PETSC_FALSE;
8590: }
8591: return(0);
8592: }
8594: /*@
8595: MatIsStructurallySymmetric - Test whether a matrix is structurally symmetric
8597: Collective on Mat
8599: Input Parameter:
8600: . A - the matrix to test
8602: Output Parameters:
8603: . flg - the result
8605: Level: intermediate
8607: .seealso: MatTranspose(), MatIsTranspose(), MatIsHermitian(), MatIsSymmetric(), MatSetOption()
8608: @*/
8609: PetscErrorCode MatIsStructurallySymmetric(Mat A,PetscBool *flg)
8610: {
8616: if (!A->structurally_symmetric_set) {
8617: 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);
8618: (*A->ops->isstructurallysymmetric)(A,&A->structurally_symmetric);
8620: A->structurally_symmetric_set = PETSC_TRUE;
8621: }
8622: *flg = A->structurally_symmetric;
8623: return(0);
8624: }
8626: /*@
8627: MatStashGetInfo - Gets how many values are currently in the matrix stash, i.e. need
8628: to be communicated to other processors during the MatAssemblyBegin/End() process
8630: Not collective
8632: Input Parameter:
8633: . vec - the vector
8635: Output Parameters:
8636: + nstash - the size of the stash
8637: . reallocs - the number of additional mallocs incurred.
8638: . bnstash - the size of the block stash
8639: - breallocs - the number of additional mallocs incurred.in the block stash
8641: Level: advanced
8643: .seealso: MatAssemblyBegin(), MatAssemblyEnd(), Mat, MatStashSetInitialSize()
8645: @*/
8646: PetscErrorCode MatStashGetInfo(Mat mat,PetscInt *nstash,PetscInt *reallocs,PetscInt *bnstash,PetscInt *breallocs)
8647: {
8651: MatStashGetInfo_Private(&mat->stash,nstash,reallocs);
8652: MatStashGetInfo_Private(&mat->bstash,bnstash,breallocs);
8653: return(0);
8654: }
8656: /*@C
8657: MatCreateVecs - Get vector(s) compatible with the matrix, i.e. with the same
8658: parallel layout
8660: Collective on Mat
8662: Input Parameter:
8663: . mat - the matrix
8665: Output Parameter:
8666: + right - (optional) vector that the matrix can be multiplied against
8667: - left - (optional) vector that the matrix vector product can be stored in
8669: Notes:
8670: 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().
8672: Notes:
8673: These are new vectors which are not owned by the Mat, they should be destroyed in VecDestroy() when no longer needed
8675: Level: advanced
8677: .seealso: MatCreate(), VecDestroy()
8678: @*/
8679: PetscErrorCode MatCreateVecs(Mat mat,Vec *right,Vec *left)
8680: {
8686: if (mat->ops->getvecs) {
8687: (*mat->ops->getvecs)(mat,right,left);
8688: } else {
8689: PetscInt rbs,cbs;
8690: MatGetBlockSizes(mat,&rbs,&cbs);
8691: if (right) {
8692: if (mat->cmap->n < 0) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"PetscLayout for columns not yet setup");
8693: VecCreate(PetscObjectComm((PetscObject)mat),right);
8694: VecSetSizes(*right,mat->cmap->n,PETSC_DETERMINE);
8695: VecSetBlockSize(*right,cbs);
8696: VecSetType(*right,mat->defaultvectype);
8697: PetscLayoutReference(mat->cmap,&(*right)->map);
8698: }
8699: if (left) {
8700: if (mat->rmap->n < 0) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"PetscLayout for rows not yet setup");
8701: VecCreate(PetscObjectComm((PetscObject)mat),left);
8702: VecSetSizes(*left,mat->rmap->n,PETSC_DETERMINE);
8703: VecSetBlockSize(*left,rbs);
8704: VecSetType(*left,mat->defaultvectype);
8705: PetscLayoutReference(mat->rmap,&(*left)->map);
8706: }
8707: }
8708: return(0);
8709: }
8711: /*@C
8712: MatFactorInfoInitialize - Initializes a MatFactorInfo data structure
8713: with default values.
8715: Not Collective
8717: Input Parameters:
8718: . info - the MatFactorInfo data structure
8721: Notes:
8722: The solvers are generally used through the KSP and PC objects, for example
8723: PCLU, PCILU, PCCHOLESKY, PCICC
8725: Level: developer
8727: .seealso: MatFactorInfo
8729: Developer Note: fortran interface is not autogenerated as the f90
8730: interface defintion cannot be generated correctly [due to MatFactorInfo]
8732: @*/
8734: PetscErrorCode MatFactorInfoInitialize(MatFactorInfo *info)
8735: {
8739: PetscMemzero(info,sizeof(MatFactorInfo));
8740: return(0);
8741: }
8743: /*@
8744: MatFactorSetSchurIS - Set indices corresponding to the Schur complement you wish to have computed
8746: Collective on Mat
8748: Input Parameters:
8749: + mat - the factored matrix
8750: - is - the index set defining the Schur indices (0-based)
8752: Notes:
8753: Call MatFactorSolveSchurComplement() or MatFactorSolveSchurComplementTranspose() after this call to solve a Schur complement system.
8755: You can call MatFactorGetSchurComplement() or MatFactorCreateSchurComplement() after this call.
8757: Level: developer
8759: .seealso: MatGetFactor(), MatFactorGetSchurComplement(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSolveSchurComplement(),
8760: MatFactorSolveSchurComplementTranspose(), MatFactorSolveSchurComplement()
8762: @*/
8763: PetscErrorCode MatFactorSetSchurIS(Mat mat,IS is)
8764: {
8765: PetscErrorCode ierr,(*f)(Mat,IS);
8773: if (!mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Only for factored matrix");
8774: PetscObjectQueryFunction((PetscObject)mat,"MatFactorSetSchurIS_C",&f);
8775: 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");
8776: MatDestroy(&mat->schur);
8777: (*f)(mat,is);
8778: if (!mat->schur) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_PLIB,"Schur complement has not been created");
8779: return(0);
8780: }
8782: /*@
8783: MatFactorCreateSchurComplement - Create a Schur complement matrix object using Schur data computed during the factorization step
8785: Logically Collective on Mat
8787: Input Parameters:
8788: + F - the factored matrix obtained by calling MatGetFactor() from PETSc-MUMPS interface
8789: . S - location where to return the Schur complement, can be NULL
8790: - status - the status of the Schur complement matrix, can be NULL
8792: Notes:
8793: You must call MatFactorSetSchurIS() before calling this routine.
8795: The routine provides a copy of the Schur matrix stored within the solver data structures.
8796: The caller must destroy the object when it is no longer needed.
8797: If MatFactorInvertSchurComplement() has been called, the routine gets back the inverse.
8799: 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)
8801: Developer Notes:
8802: The reason this routine exists is because the representation of the Schur complement within the factor matrix may be different than a standard PETSc
8803: matrix representation and we normally do not want to use the time or memory to make a copy as a regular PETSc matrix.
8805: See MatCreateSchurComplement() or MatGetSchurComplement() for ways to create virtual or approximate Schur complements.
8807: Level: advanced
8809: References:
8811: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorGetSchurComplement(), MatFactorSchurStatus
8812: @*/
8813: PetscErrorCode MatFactorCreateSchurComplement(Mat F,Mat* S,MatFactorSchurStatus* status)
8814: {
8821: if (S) {
8822: PetscErrorCode (*f)(Mat,Mat*);
8824: PetscObjectQueryFunction((PetscObject)F,"MatFactorCreateSchurComplement_C",&f);
8825: if (f) {
8826: (*f)(F,S);
8827: } else {
8828: MatDuplicate(F->schur,MAT_COPY_VALUES,S);
8829: }
8830: }
8831: if (status) *status = F->schur_status;
8832: return(0);
8833: }
8835: /*@
8836: MatFactorGetSchurComplement - Gets access to a Schur complement matrix using the current Schur data within a factored matrix
8838: Logically Collective on Mat
8840: Input Parameters:
8841: + F - the factored matrix obtained by calling MatGetFactor()
8842: . *S - location where to return the Schur complement, can be NULL
8843: - status - the status of the Schur complement matrix, can be NULL
8845: Notes:
8846: You must call MatFactorSetSchurIS() before calling this routine.
8848: Schur complement mode is currently implemented for sequential matrices.
8849: The routine returns a the Schur Complement stored within the data strutures of the solver.
8850: If MatFactorInvertSchurComplement() has previously been called, the returned matrix is actually the inverse of the Schur complement.
8851: The returned matrix should not be destroyed; the caller should call MatFactorRestoreSchurComplement() when the object is no longer needed.
8853: Use MatFactorCreateSchurComplement() to create a copy of the Schur complement matrix that is within a factored matrix
8855: See MatCreateSchurComplement() or MatGetSchurComplement() for ways to create virtual or approximate Schur complements.
8857: Level: advanced
8859: References:
8861: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSchurStatus
8862: @*/
8863: PetscErrorCode MatFactorGetSchurComplement(Mat F,Mat* S,MatFactorSchurStatus* status)
8864: {
8869: if (S) *S = F->schur;
8870: if (status) *status = F->schur_status;
8871: return(0);
8872: }
8874: /*@
8875: MatFactorRestoreSchurComplement - Restore the Schur complement matrix object obtained from a call to MatFactorGetSchurComplement
8877: Logically Collective on Mat
8879: Input Parameters:
8880: + F - the factored matrix obtained by calling MatGetFactor()
8881: . *S - location where the Schur complement is stored
8882: - status - the status of the Schur complement matrix (see MatFactorSchurStatus)
8884: Notes:
8886: Level: advanced
8888: References:
8890: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorRestoreSchurComplement(), MatFactorCreateSchurComplement(), MatFactorSchurStatus
8891: @*/
8892: PetscErrorCode MatFactorRestoreSchurComplement(Mat F,Mat* S,MatFactorSchurStatus status)
8893: {
8898: if (S) {
8900: *S = NULL;
8901: }
8902: F->schur_status = status;
8903: MatFactorUpdateSchurStatus_Private(F);
8904: return(0);
8905: }
8907: /*@
8908: MatFactorSolveSchurComplementTranspose - Solve the transpose of the Schur complement system computed during the factorization step
8910: Logically Collective on Mat
8912: Input Parameters:
8913: + F - the factored matrix obtained by calling MatGetFactor()
8914: . rhs - location where the right hand side of the Schur complement system is stored
8915: - sol - location where the solution of the Schur complement system has to be returned
8917: Notes:
8918: The sizes of the vectors should match the size of the Schur complement
8920: Must be called after MatFactorSetSchurIS()
8922: Level: advanced
8924: References:
8926: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorSolveSchurComplement()
8927: @*/
8928: PetscErrorCode MatFactorSolveSchurComplementTranspose(Mat F, Vec rhs, Vec sol)
8929: {
8941: MatFactorFactorizeSchurComplement(F);
8942: switch (F->schur_status) {
8943: case MAT_FACTOR_SCHUR_FACTORED:
8944: MatSolveTranspose(F->schur,rhs,sol);
8945: break;
8946: case MAT_FACTOR_SCHUR_INVERTED:
8947: MatMultTranspose(F->schur,rhs,sol);
8948: break;
8949: default:
8950: SETERRQ1(PetscObjectComm((PetscObject)F),PETSC_ERR_SUP,"Unhandled MatFactorSchurStatus %D",F->schur_status);
8951: break;
8952: }
8953: return(0);
8954: }
8956: /*@
8957: MatFactorSolveSchurComplement - Solve the Schur complement system computed during the factorization step
8959: Logically Collective on Mat
8961: Input Parameters:
8962: + F - the factored matrix obtained by calling MatGetFactor()
8963: . rhs - location where the right hand side of the Schur complement system is stored
8964: - sol - location where the solution of the Schur complement system has to be returned
8966: Notes:
8967: The sizes of the vectors should match the size of the Schur complement
8969: Must be called after MatFactorSetSchurIS()
8971: Level: advanced
8973: References:
8975: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorSolveSchurComplementTranspose()
8976: @*/
8977: PetscErrorCode MatFactorSolveSchurComplement(Mat F, Vec rhs, Vec sol)
8978: {
8990: MatFactorFactorizeSchurComplement(F);
8991: switch (F->schur_status) {
8992: case MAT_FACTOR_SCHUR_FACTORED:
8993: MatSolve(F->schur,rhs,sol);
8994: break;
8995: case MAT_FACTOR_SCHUR_INVERTED:
8996: MatMult(F->schur,rhs,sol);
8997: break;
8998: default:
8999: SETERRQ1(PetscObjectComm((PetscObject)F),PETSC_ERR_SUP,"Unhandled MatFactorSchurStatus %D",F->schur_status);
9000: break;
9001: }
9002: return(0);
9003: }
9005: /*@
9006: MatFactorInvertSchurComplement - Invert the Schur complement matrix computed during the factorization step
9008: Logically Collective on Mat
9010: Input Parameters:
9011: . F - the factored matrix obtained by calling MatGetFactor()
9013: Notes:
9014: Must be called after MatFactorSetSchurIS().
9016: Call MatFactorGetSchurComplement() or MatFactorCreateSchurComplement() AFTER this call to actually compute the inverse and get access to it.
9018: Level: advanced
9020: References:
9022: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorGetSchurComplement(), MatFactorCreateSchurComplement()
9023: @*/
9024: PetscErrorCode MatFactorInvertSchurComplement(Mat F)
9025: {
9031: if (F->schur_status == MAT_FACTOR_SCHUR_INVERTED) return(0);
9032: MatFactorFactorizeSchurComplement(F);
9033: MatFactorInvertSchurComplement_Private(F);
9034: F->schur_status = MAT_FACTOR_SCHUR_INVERTED;
9035: return(0);
9036: }
9038: /*@
9039: MatFactorFactorizeSchurComplement - Factorize the Schur complement matrix computed during the factorization step
9041: Logically Collective on Mat
9043: Input Parameters:
9044: . F - the factored matrix obtained by calling MatGetFactor()
9046: Notes:
9047: Must be called after MatFactorSetSchurIS().
9049: Level: advanced
9051: References:
9053: .seealso: MatGetFactor(), MatFactorSetSchurIS(), MatFactorInvertSchurComplement()
9054: @*/
9055: PetscErrorCode MatFactorFactorizeSchurComplement(Mat F)
9056: {
9062: if (F->schur_status == MAT_FACTOR_SCHUR_INVERTED || F->schur_status == MAT_FACTOR_SCHUR_FACTORED) return(0);
9063: MatFactorFactorizeSchurComplement_Private(F);
9064: F->schur_status = MAT_FACTOR_SCHUR_FACTORED;
9065: return(0);
9066: }
9068: PetscErrorCode MatPtAP_Basic(Mat A,Mat P,MatReuse scall,PetscReal fill,Mat *C)
9069: {
9070: Mat AP;
9074: PetscInfo2(A,"Mat types %s and %s using basic PtAP\n",((PetscObject)A)->type_name,((PetscObject)P)->type_name);
9075: MatMatMult(A,P,MAT_INITIAL_MATRIX,PETSC_DEFAULT,&AP);
9076: MatTransposeMatMult(P,AP,scall,fill,C);
9077: MatDestroy(&AP);
9078: return(0);
9079: }
9081: /*@
9082: MatPtAP - Creates the matrix product C = P^T * A * P
9084: Neighbor-wise Collective on Mat
9086: Input Parameters:
9087: + A - the matrix
9088: . P - the projection matrix
9089: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9090: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(P)), use PETSC_DEFAULT if you do not have a good estimate
9091: if the result is a dense matrix this is irrelevent
9093: Output Parameters:
9094: . C - the product matrix
9096: Notes:
9097: C will be created and must be destroyed by the user with MatDestroy().
9099: For matrix types without special implementation the function fallbacks to MatMatMult() followed by MatTransposeMatMult().
9101: Level: intermediate
9103: .seealso: MatPtAPSymbolic(), MatPtAPNumeric(), MatMatMult(), MatRARt()
9104: @*/
9105: PetscErrorCode MatPtAP(Mat A,Mat P,MatReuse scall,PetscReal fill,Mat *C)
9106: {
9108: PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
9109: PetscErrorCode (*fP)(Mat,Mat,MatReuse,PetscReal,Mat*);
9110: PetscErrorCode (*ptap)(Mat,Mat,MatReuse,PetscReal,Mat*)=NULL;
9111: PetscBool sametype;
9116: MatCheckPreallocated(A,1);
9117: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9118: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9119: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9122: MatCheckPreallocated(P,2);
9123: if (!P->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9124: if (P->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9126: if (A->rmap->N != A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix A must be square, %D != %D",A->rmap->N,A->cmap->N);
9127: if (P->rmap->N != A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
9128: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9129: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9131: if (scall == MAT_REUSE_MATRIX) {
9135: PetscLogEventBegin(MAT_PtAP,A,P,0,0);
9136: PetscLogEventBegin(MAT_PtAPNumeric,A,P,0,0);
9137: if ((*C)->ops->ptapnumeric) {
9138: (*(*C)->ops->ptapnumeric)(A,P,*C);
9139: } else {
9140: MatPtAP_Basic(A,P,scall,fill,C);
9141: }
9142: PetscLogEventEnd(MAT_PtAPNumeric,A,P,0,0);
9143: PetscLogEventEnd(MAT_PtAP,A,P,0,0);
9144: return(0);
9145: }
9147: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9148: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9150: fA = A->ops->ptap;
9151: fP = P->ops->ptap;
9152: PetscStrcmp(((PetscObject)A)->type_name,((PetscObject)P)->type_name,&sametype);
9153: if (fP == fA && sametype) {
9154: ptap = fA;
9155: } else {
9156: /* dispatch based on the type of A and P from their PetscObject's PetscFunctionLists. */
9157: char ptapname[256];
9158: PetscStrncpy(ptapname,"MatPtAP_",sizeof(ptapname));
9159: PetscStrlcat(ptapname,((PetscObject)A)->type_name,sizeof(ptapname));
9160: PetscStrlcat(ptapname,"_",sizeof(ptapname));
9161: PetscStrlcat(ptapname,((PetscObject)P)->type_name,sizeof(ptapname));
9162: PetscStrlcat(ptapname,"_C",sizeof(ptapname)); /* e.g., ptapname = "MatPtAP_seqdense_seqaij_C" */
9163: PetscObjectQueryFunction((PetscObject)P,ptapname,&ptap);
9164: }
9166: if (!ptap) ptap = MatPtAP_Basic;
9167: PetscLogEventBegin(MAT_PtAP,A,P,0,0);
9168: (*ptap)(A,P,scall,fill,C);
9169: PetscLogEventEnd(MAT_PtAP,A,P,0,0);
9170: if (A->symmetric_set && A->symmetric) {
9171: MatSetOption(*C,MAT_SYMMETRIC,PETSC_TRUE);
9172: }
9173: return(0);
9174: }
9176: /*@
9177: MatPtAPNumeric - Computes the matrix product C = P^T * A * P
9179: Neighbor-wise Collective on Mat
9181: Input Parameters:
9182: + A - the matrix
9183: - P - the projection matrix
9185: Output Parameters:
9186: . C - the product matrix
9188: Notes:
9189: C must have been created by calling MatPtAPSymbolic and must be destroyed by
9190: the user using MatDeatroy().
9192: This routine is currently only implemented for pairs of AIJ matrices and classes
9193: which inherit from AIJ. C will be of type MATAIJ.
9195: Level: intermediate
9197: .seealso: MatPtAP(), MatPtAPSymbolic(), MatMatMultNumeric()
9198: @*/
9199: PetscErrorCode MatPtAPNumeric(Mat A,Mat P,Mat C)
9200: {
9206: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9207: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9210: MatCheckPreallocated(P,2);
9211: if (!P->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9212: if (P->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9215: MatCheckPreallocated(C,3);
9216: if (C->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9217: if (P->cmap->N!=C->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->cmap->N,C->rmap->N);
9218: if (P->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
9219: if (A->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
9220: if (P->cmap->N!=C->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->cmap->N,C->cmap->N);
9221: MatCheckPreallocated(A,1);
9223: if (!C->ops->ptapnumeric) SETERRQ(PetscObjectComm((PetscObject)C),PETSC_ERR_ARG_WRONGSTATE,"MatPtAPNumeric implementation is missing. You should call MatPtAPSymbolic first");
9224: PetscLogEventBegin(MAT_PtAPNumeric,A,P,0,0);
9225: (*C->ops->ptapnumeric)(A,P,C);
9226: PetscLogEventEnd(MAT_PtAPNumeric,A,P,0,0);
9227: return(0);
9228: }
9230: /*@
9231: MatPtAPSymbolic - Creates the (i,j) structure of the matrix product C = P^T * A * P
9233: Neighbor-wise Collective on Mat
9235: Input Parameters:
9236: + A - the matrix
9237: - P - the projection matrix
9239: Output Parameters:
9240: . C - the (i,j) structure of the product matrix
9242: Notes:
9243: C will be created and must be destroyed by the user with MatDestroy().
9245: This routine is currently only implemented for pairs of SeqAIJ matrices and classes
9246: which inherit from SeqAIJ. C will be of type MATSEQAIJ. The product is computed using
9247: this (i,j) structure by calling MatPtAPNumeric().
9249: Level: intermediate
9251: .seealso: MatPtAP(), MatPtAPNumeric(), MatMatMultSymbolic()
9252: @*/
9253: PetscErrorCode MatPtAPSymbolic(Mat A,Mat P,PetscReal fill,Mat *C)
9254: {
9260: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9261: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9262: if (fill <1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9265: MatCheckPreallocated(P,2);
9266: if (!P->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9267: if (P->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9270: if (P->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",P->rmap->N,A->cmap->N);
9271: if (A->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
9272: MatCheckPreallocated(A,1);
9274: if (!A->ops->ptapsymbolic) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"MatType %s",((PetscObject)A)->type_name);
9275: PetscLogEventBegin(MAT_PtAPSymbolic,A,P,0,0);
9276: (*A->ops->ptapsymbolic)(A,P,fill,C);
9277: PetscLogEventEnd(MAT_PtAPSymbolic,A,P,0,0);
9279: /* MatSetBlockSize(*C,A->rmap->bs); NO! this is not always true -ma */
9280: return(0);
9281: }
9283: /*@
9284: MatRARt - Creates the matrix product C = R * A * R^T
9286: Neighbor-wise Collective on Mat
9288: Input Parameters:
9289: + A - the matrix
9290: . R - the projection matrix
9291: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9292: - fill - expected fill as ratio of nnz(C)/nnz(A), use PETSC_DEFAULT if you do not have a good estimate
9293: if the result is a dense matrix this is irrelevent
9295: Output Parameters:
9296: . C - the product matrix
9298: Notes:
9299: C will be created and must be destroyed by the user with MatDestroy().
9301: This routine is currently only implemented for pairs of AIJ matrices and classes
9302: which inherit from AIJ. Due to PETSc sparse matrix block row distribution among processes,
9303: parallel MatRARt is implemented via explicit transpose of R, which could be very expensive.
9304: We recommend using MatPtAP().
9306: Level: intermediate
9308: .seealso: MatRARtSymbolic(), MatRARtNumeric(), MatMatMult(), MatPtAP()
9309: @*/
9310: PetscErrorCode MatRARt(Mat A,Mat R,MatReuse scall,PetscReal fill,Mat *C)
9311: {
9317: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9318: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9319: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9322: MatCheckPreallocated(R,2);
9323: if (!R->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9324: if (R->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9326: if (R->cmap->N!=A->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)R),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",R->cmap->N,A->rmap->N);
9328: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9329: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9330: MatCheckPreallocated(A,1);
9332: if (!A->ops->rart) {
9333: Mat Rt;
9334: MatTranspose(R,MAT_INITIAL_MATRIX,&Rt);
9335: MatMatMatMult(R,A,Rt,scall,fill,C);
9336: MatDestroy(&Rt);
9337: return(0);
9338: }
9339: PetscLogEventBegin(MAT_RARt,A,R,0,0);
9340: (*A->ops->rart)(A,R,scall,fill,C);
9341: PetscLogEventEnd(MAT_RARt,A,R,0,0);
9342: return(0);
9343: }
9345: /*@
9346: MatRARtNumeric - Computes the matrix product C = R * A * R^T
9348: Neighbor-wise Collective on Mat
9350: Input Parameters:
9351: + A - the matrix
9352: - R - the projection matrix
9354: Output Parameters:
9355: . C - the product matrix
9357: Notes:
9358: C must have been created by calling MatRARtSymbolic and must be destroyed by
9359: the user using MatDestroy().
9361: This routine is currently only implemented for pairs of AIJ matrices and classes
9362: which inherit from AIJ. C will be of type MATAIJ.
9364: Level: intermediate
9366: .seealso: MatRARt(), MatRARtSymbolic(), MatMatMultNumeric()
9367: @*/
9368: PetscErrorCode MatRARtNumeric(Mat A,Mat R,Mat C)
9369: {
9375: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9376: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9379: MatCheckPreallocated(R,2);
9380: if (!R->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9381: if (R->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9384: MatCheckPreallocated(C,3);
9385: if (C->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9386: if (R->rmap->N!=C->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",R->rmap->N,C->rmap->N);
9387: if (R->cmap->N!=A->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",R->cmap->N,A->rmap->N);
9388: if (A->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
9389: if (R->rmap->N!=C->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",R->rmap->N,C->cmap->N);
9390: MatCheckPreallocated(A,1);
9392: PetscLogEventBegin(MAT_RARtNumeric,A,R,0,0);
9393: (*A->ops->rartnumeric)(A,R,C);
9394: PetscLogEventEnd(MAT_RARtNumeric,A,R,0,0);
9395: return(0);
9396: }
9398: /*@
9399: MatRARtSymbolic - Creates the (i,j) structure of the matrix product C = R * A * R^T
9401: Neighbor-wise Collective on Mat
9403: Input Parameters:
9404: + A - the matrix
9405: - R - the projection matrix
9407: Output Parameters:
9408: . C - the (i,j) structure of the product matrix
9410: Notes:
9411: C will be created and must be destroyed by the user with MatDestroy().
9413: This routine is currently only implemented for pairs of SeqAIJ matrices and classes
9414: which inherit from SeqAIJ. C will be of type MATSEQAIJ. The product is computed using
9415: this (i,j) structure by calling MatRARtNumeric().
9417: Level: intermediate
9419: .seealso: MatRARt(), MatRARtNumeric(), MatMatMultSymbolic()
9420: @*/
9421: PetscErrorCode MatRARtSymbolic(Mat A,Mat R,PetscReal fill,Mat *C)
9422: {
9428: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9429: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9430: if (fill <1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9433: MatCheckPreallocated(R,2);
9434: if (!R->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9435: if (R->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9438: if (R->cmap->N!=A->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",R->cmap->N,A->rmap->N);
9439: if (A->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix 'A' must be square, %D != %D",A->rmap->N,A->cmap->N);
9440: MatCheckPreallocated(A,1);
9441: PetscLogEventBegin(MAT_RARtSymbolic,A,R,0,0);
9442: (*A->ops->rartsymbolic)(A,R,fill,C);
9443: PetscLogEventEnd(MAT_RARtSymbolic,A,R,0,0);
9445: MatSetBlockSizes(*C,PetscAbs(R->rmap->bs),PetscAbs(R->rmap->bs));
9446: return(0);
9447: }
9449: /*@
9450: MatMatMult - Performs Matrix-Matrix Multiplication C=A*B.
9452: Neighbor-wise Collective on Mat
9454: Input Parameters:
9455: + A - the left matrix
9456: . B - the right matrix
9457: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9458: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if you do not have a good estimate
9459: if the result is a dense matrix this is irrelevent
9461: Output Parameters:
9462: . C - the product matrix
9464: Notes:
9465: Unless scall is MAT_REUSE_MATRIX C will be created.
9467: 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
9468: call to this function with either MAT_INITIAL_MATRIX or MatMatMultSymbolic()
9470: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9471: actually needed.
9473: If you have many matrices with the same non-zero structure to multiply, you
9474: should either
9475: $ 1) use MAT_REUSE_MATRIX in all calls but the first or
9476: $ 2) call MatMatMultSymbolic() once and then MatMatMultNumeric() for each product needed
9477: 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
9478: with MAT_REUSE_MATRIX, rather than first having MatMatMult() create it for you. You can NEVER do this if the matrix C is sparse.
9480: Level: intermediate
9482: .seealso: MatMatMultSymbolic(), MatMatMultNumeric(), MatTransposeMatMult(), MatMatTransposeMult(), MatPtAP()
9483: @*/
9484: PetscErrorCode MatMatMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9485: {
9487: PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
9488: PetscErrorCode (*fB)(Mat,Mat,MatReuse,PetscReal,Mat*);
9489: PetscErrorCode (*mult)(Mat,Mat,MatReuse,PetscReal,Mat*)=NULL;
9490: Mat T;
9491: PetscBool istrans;
9496: MatCheckPreallocated(A,1);
9497: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9498: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9501: MatCheckPreallocated(B,2);
9502: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9503: if (B->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9505: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9506: if (B->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
9507: PetscObjectTypeCompare((PetscObject)A,MATTRANSPOSEMAT,&istrans);
9508: if (istrans) {
9509: MatTransposeGetMat(A,&T);
9510: MatTransposeMatMult(T,B,scall,fill,C);
9511: return(0);
9512: } else {
9513: PetscObjectTypeCompare((PetscObject)B,MATTRANSPOSEMAT,&istrans);
9514: if (istrans) {
9515: MatTransposeGetMat(B,&T);
9516: MatMatTransposeMult(A,T,scall,fill,C);
9517: return(0);
9518: }
9519: }
9520: if (scall == MAT_REUSE_MATRIX) {
9523: PetscLogEventBegin(MAT_MatMult,A,B,0,0);
9524: PetscLogEventBegin(MAT_MatMultNumeric,A,B,0,0);
9525: (*(*C)->ops->matmultnumeric)(A,B,*C);
9526: PetscLogEventEnd(MAT_MatMultNumeric,A,B,0,0);
9527: PetscLogEventEnd(MAT_MatMult,A,B,0,0);
9528: return(0);
9529: }
9530: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9531: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9533: fA = A->ops->matmult;
9534: fB = B->ops->matmult;
9535: if (fB == fA && fB) mult = fB;
9536: else {
9537: /* dispatch based on the type of A and B from their PetscObject's PetscFunctionLists. */
9538: char multname[256];
9539: PetscStrncpy(multname,"MatMatMult_",sizeof(multname));
9540: PetscStrlcat(multname,((PetscObject)A)->type_name,sizeof(multname));
9541: PetscStrlcat(multname,"_",sizeof(multname));
9542: PetscStrlcat(multname,((PetscObject)B)->type_name,sizeof(multname));
9543: PetscStrlcat(multname,"_C",sizeof(multname)); /* e.g., multname = "MatMatMult_seqdense_seqaij_C" */
9544: PetscObjectQueryFunction((PetscObject)B,multname,&mult);
9545: if (!mult) {
9546: PetscObjectQueryFunction((PetscObject)A,multname,&mult);
9547: }
9548: if (!mult) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"MatMatMult requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
9549: }
9550: PetscLogEventBegin(MAT_MatMult,A,B,0,0);
9551: (*mult)(A,B,scall,fill,C);
9552: PetscLogEventEnd(MAT_MatMult,A,B,0,0);
9553: return(0);
9554: }
9556: /*@
9557: MatMatMultSymbolic - Performs construction, preallocation, and computes the ij structure
9558: of the matrix-matrix product C=A*B. Call this routine before calling MatMatMultNumeric().
9560: Neighbor-wise Collective on Mat
9562: Input Parameters:
9563: + A - the left matrix
9564: . B - the right matrix
9565: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if you do not have a good estimate,
9566: if C is a dense matrix this is irrelevent
9568: Output Parameters:
9569: . C - the product matrix
9571: Notes:
9572: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9573: actually needed.
9575: This routine is currently implemented for
9576: - pairs of AIJ matrices and classes which inherit from AIJ, C will be of type AIJ
9577: - pairs of AIJ (A) and Dense (B) matrix, C will be of type Dense.
9578: - pairs of Dense (A) and AIJ (B) matrix, C will be of type Dense.
9580: Level: intermediate
9582: Developers Note: There are ways to estimate the number of nonzeros in the resulting product, see for example, https://arxiv.org/abs/1006.4173
9583: We should incorporate them into PETSc.
9585: .seealso: MatMatMult(), MatMatMultNumeric()
9586: @*/
9587: PetscErrorCode MatMatMultSymbolic(Mat A,Mat B,PetscReal fill,Mat *C)
9588: {
9590: PetscErrorCode (*Asymbolic)(Mat,Mat,PetscReal,Mat*);
9591: PetscErrorCode (*Bsymbolic)(Mat,Mat,PetscReal,Mat*);
9592: PetscErrorCode (*symbolic)(Mat,Mat,PetscReal,Mat*)=NULL;
9597: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9598: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9602: MatCheckPreallocated(B,2);
9603: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9604: if (B->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9607: if (B->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
9608: if (fill == PETSC_DEFAULT) fill = 2.0;
9609: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be > 1.0",(double)fill);
9610: MatCheckPreallocated(A,1);
9612: Asymbolic = A->ops->matmultsymbolic;
9613: Bsymbolic = B->ops->matmultsymbolic;
9614: if (Asymbolic == Bsymbolic) {
9615: if (!Bsymbolic) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"C=A*B not implemented for B of type %s",((PetscObject)B)->type_name);
9616: symbolic = Bsymbolic;
9617: } else { /* dispatch based on the type of A and B */
9618: char symbolicname[256];
9619: PetscStrncpy(symbolicname,"MatMatMultSymbolic_",sizeof(symbolicname));
9620: PetscStrlcat(symbolicname,((PetscObject)A)->type_name,sizeof(symbolicname));
9621: PetscStrlcat(symbolicname,"_",sizeof(symbolicname));
9622: PetscStrlcat(symbolicname,((PetscObject)B)->type_name,sizeof(symbolicname));
9623: PetscStrlcat(symbolicname,"_C",sizeof(symbolicname));
9624: PetscObjectQueryFunction((PetscObject)B,symbolicname,&symbolic);
9625: if (!symbolic) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"MatMatMultSymbolic requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
9626: }
9627: PetscLogEventBegin(MAT_MatMultSymbolic,A,B,0,0);
9628: *C = NULL;
9629: (*symbolic)(A,B,fill,C);
9630: PetscLogEventEnd(MAT_MatMultSymbolic,A,B,0,0);
9631: return(0);
9632: }
9634: /*@
9635: MatMatMultNumeric - Performs the numeric matrix-matrix product.
9636: Call this routine after first calling MatMatMultSymbolic().
9638: Neighbor-wise Collective on Mat
9640: Input Parameters:
9641: + A - the left matrix
9642: - B - the right matrix
9644: Output Parameters:
9645: . C - the product matrix, which was created by from MatMatMultSymbolic() or a call to MatMatMult().
9647: Notes:
9648: C must have been created with MatMatMultSymbolic().
9650: This routine is currently implemented for
9651: - pairs of AIJ matrices and classes which inherit from AIJ, C will be of type MATAIJ.
9652: - pairs of AIJ (A) and Dense (B) matrix, C will be of type Dense.
9653: - pairs of Dense (A) and AIJ (B) matrix, C will be of type Dense.
9655: Level: intermediate
9657: .seealso: MatMatMult(), MatMatMultSymbolic()
9658: @*/
9659: PetscErrorCode MatMatMultNumeric(Mat A,Mat B,Mat C)
9660: {
9664: MatMatMult(A,B,MAT_REUSE_MATRIX,0.0,&C);
9665: return(0);
9666: }
9668: /*@
9669: MatMatTransposeMult - Performs Matrix-Matrix Multiplication C=A*B^T.
9671: Neighbor-wise Collective on Mat
9673: Input Parameters:
9674: + A - the left matrix
9675: . B - the right matrix
9676: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9677: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if not known
9679: Output Parameters:
9680: . C - the product matrix
9682: Notes:
9683: C will be created if MAT_INITIAL_MATRIX and must be destroyed by the user with MatDestroy().
9685: MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call
9687: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9688: actually needed.
9690: This routine is currently only implemented for pairs of SeqAIJ matrices, for the SeqDense class,
9691: and for pairs of MPIDense matrices.
9693: Options Database Keys:
9694: . -matmattransmult_mpidense_mpidense_via {allgatherv,cyclic} - Choose between algorthims for MPIDense matrices: the
9695: first redundantly copies the transposed B matrix on each process and requiers O(log P) communication complexity;
9696: the second never stores more than one portion of the B matrix at a time by requires O(P) communication complexity.
9698: Level: intermediate
9700: .seealso: MatMatTransposeMultSymbolic(), MatMatTransposeMultNumeric(), MatMatMult(), MatTransposeMatMult() MatPtAP()
9701: @*/
9702: PetscErrorCode MatMatTransposeMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9703: {
9705: PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
9706: PetscErrorCode (*fB)(Mat,Mat,MatReuse,PetscReal,Mat*);
9707: Mat T;
9708: PetscBool istrans;
9713: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9714: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9715: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9718: MatCheckPreallocated(B,2);
9719: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9720: if (B->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9722: if (B->cmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, AN %D != BN %D",A->cmap->N,B->cmap->N);
9723: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9724: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be > 1.0",(double)fill);
9725: MatCheckPreallocated(A,1);
9727: PetscObjectTypeCompare((PetscObject)B,MATTRANSPOSEMAT,&istrans);
9728: if (istrans) {
9729: MatTransposeGetMat(B,&T);
9730: MatMatMult(A,T,scall,fill,C);
9731: return(0);
9732: }
9733: fA = A->ops->mattransposemult;
9734: if (!fA) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"MatMatTransposeMult not supported for A of type %s",((PetscObject)A)->type_name);
9735: fB = B->ops->mattransposemult;
9736: if (!fB) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"MatMatTransposeMult not supported for B of type %s",((PetscObject)B)->type_name);
9737: if (fB!=fA) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"MatMatTransposeMult requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
9739: PetscLogEventBegin(MAT_MatTransposeMult,A,B,0,0);
9740: if (scall == MAT_INITIAL_MATRIX) {
9741: PetscLogEventBegin(MAT_MatTransposeMultSymbolic,A,B,0,0);
9742: (*A->ops->mattransposemultsymbolic)(A,B,fill,C);
9743: PetscLogEventEnd(MAT_MatTransposeMultSymbolic,A,B,0,0);
9744: }
9745: PetscLogEventBegin(MAT_MatTransposeMultNumeric,A,B,0,0);
9746: (*A->ops->mattransposemultnumeric)(A,B,*C);
9747: PetscLogEventEnd(MAT_MatTransposeMultNumeric,A,B,0,0);
9748: PetscLogEventEnd(MAT_MatTransposeMult,A,B,0,0);
9749: return(0);
9750: }
9752: /*@
9753: MatTransposeMatMult - Performs Matrix-Matrix Multiplication C=A^T*B.
9755: Neighbor-wise Collective on Mat
9757: Input Parameters:
9758: + A - the left matrix
9759: . B - the right matrix
9760: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9761: - fill - expected fill as ratio of nnz(C)/(nnz(A) + nnz(B)), use PETSC_DEFAULT if not known
9763: Output Parameters:
9764: . C - the product matrix
9766: Notes:
9767: C will be created if MAT_INITIAL_MATRIX and must be destroyed by the user with MatDestroy().
9769: MAT_REUSE_MATRIX can only be used if the matrices A and B have the same nonzero pattern as in the previous call
9771: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9772: actually needed.
9774: This routine is currently implemented for pairs of AIJ matrices and pairs of SeqDense matrices and classes
9775: which inherit from SeqAIJ. C will be of same type as the input matrices.
9777: Level: intermediate
9779: .seealso: MatTransposeMatMultSymbolic(), MatTransposeMatMultNumeric(), MatMatMult(), MatMatTransposeMult(), MatPtAP()
9780: @*/
9781: PetscErrorCode MatTransposeMatMult(Mat A,Mat B,MatReuse scall,PetscReal fill,Mat *C)
9782: {
9784: PetscErrorCode (*fA)(Mat,Mat,MatReuse,PetscReal,Mat*);
9785: PetscErrorCode (*fB)(Mat,Mat,MatReuse,PetscReal,Mat*);
9786: PetscErrorCode (*transposematmult)(Mat,Mat,MatReuse,PetscReal,Mat*) = NULL;
9787: Mat T;
9788: PetscBool istrans;
9793: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9794: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9795: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9798: MatCheckPreallocated(B,2);
9799: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9800: if (B->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9802: if (B->rmap->N!=A->rmap->N) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->rmap->N);
9803: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9804: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be > 1.0",(double)fill);
9805: MatCheckPreallocated(A,1);
9807: PetscObjectTypeCompare((PetscObject)A,MATTRANSPOSEMAT,&istrans);
9808: if (istrans) {
9809: MatTransposeGetMat(A,&T);
9810: MatMatMult(T,B,scall,fill,C);
9811: return(0);
9812: }
9813: fA = A->ops->transposematmult;
9814: fB = B->ops->transposematmult;
9815: if (fB == fA && fA) transposematmult = fA;
9816: else {
9817: /* dispatch based on the type of A and B from their PetscObject's PetscFunctionLists. */
9818: char multname[256];
9819: PetscStrncpy(multname,"MatTransposeMatMult_",sizeof(multname));
9820: PetscStrlcat(multname,((PetscObject)A)->type_name,sizeof(multname));
9821: PetscStrlcat(multname,"_",sizeof(multname));
9822: PetscStrlcat(multname,((PetscObject)B)->type_name,sizeof(multname));
9823: PetscStrlcat(multname,"_C",sizeof(multname)); /* e.g., multname = "MatMatMult_seqdense_seqaij_C" */
9824: PetscObjectQueryFunction((PetscObject)B,multname,&transposematmult);
9825: if (!transposematmult) {
9826: PetscObjectQueryFunction((PetscObject)A,multname,&transposematmult);
9827: }
9828: if (!transposematmult) SETERRQ2(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"MatTransposeMatMult requires A, %s, to be compatible with B, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name);
9829: }
9830: PetscLogEventBegin(MAT_TransposeMatMult,A,B,0,0);
9831: (*transposematmult)(A,B,scall,fill,C);
9832: PetscLogEventEnd(MAT_TransposeMatMult,A,B,0,0);
9833: return(0);
9834: }
9836: /*@
9837: MatMatMatMult - Performs Matrix-Matrix-Matrix Multiplication D=A*B*C.
9839: Neighbor-wise Collective on Mat
9841: Input Parameters:
9842: + A - the left matrix
9843: . B - the middle matrix
9844: . C - the right matrix
9845: . scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9846: - 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
9847: if the result is a dense matrix this is irrelevent
9849: Output Parameters:
9850: . D - the product matrix
9852: Notes:
9853: Unless scall is MAT_REUSE_MATRIX D will be created.
9855: MAT_REUSE_MATRIX can only be used if the matrices A, B and C have the same nonzero pattern as in the previous call
9857: To determine the correct fill value, run with -info and search for the string "Fill ratio" to see the value
9858: actually needed.
9860: If you have many matrices with the same non-zero structure to multiply, you
9861: should use MAT_REUSE_MATRIX in all calls but the first or
9863: Level: intermediate
9865: .seealso: MatMatMult, MatPtAP()
9866: @*/
9867: PetscErrorCode MatMatMatMult(Mat A,Mat B,Mat C,MatReuse scall,PetscReal fill,Mat *D)
9868: {
9870: PetscErrorCode (*fA)(Mat,Mat,Mat,MatReuse,PetscReal,Mat*);
9871: PetscErrorCode (*fB)(Mat,Mat,Mat,MatReuse,PetscReal,Mat*);
9872: PetscErrorCode (*fC)(Mat,Mat,Mat,MatReuse,PetscReal,Mat*);
9873: PetscErrorCode (*mult)(Mat,Mat,Mat,MatReuse,PetscReal,Mat*)=NULL;
9878: MatCheckPreallocated(A,1);
9879: if (scall == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
9880: if (!A->assembled) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9881: if (A->factortype) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9884: MatCheckPreallocated(B,2);
9885: if (!B->assembled) SETERRQ(PetscObjectComm((PetscObject)B),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9886: if (B->factortype) SETERRQ(PetscObjectComm((PetscObject)B),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9889: MatCheckPreallocated(C,3);
9890: if (!C->assembled) SETERRQ(PetscObjectComm((PetscObject)C),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9891: if (C->factortype) SETERRQ(PetscObjectComm((PetscObject)C),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9892: if (B->rmap->N!=A->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)B),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",B->rmap->N,A->cmap->N);
9893: if (C->rmap->N!=B->cmap->N) SETERRQ2(PetscObjectComm((PetscObject)C),PETSC_ERR_ARG_SIZ,"Matrix dimensions are incompatible, %D != %D",C->rmap->N,B->cmap->N);
9894: if (scall == MAT_REUSE_MATRIX) {
9897: PetscLogEventBegin(MAT_MatMatMult,A,B,0,0);
9898: (*(*D)->ops->matmatmult)(A,B,C,scall,fill,D);
9899: PetscLogEventEnd(MAT_MatMatMult,A,B,0,0);
9900: return(0);
9901: }
9902: if (fill == PETSC_DEFAULT || fill == PETSC_DECIDE) fill = 2.0;
9903: if (fill < 1.0) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_SIZ,"Expected fill=%g must be >= 1.0",(double)fill);
9905: fA = A->ops->matmatmult;
9906: fB = B->ops->matmatmult;
9907: fC = C->ops->matmatmult;
9908: if (fA == fB && fA == fC) {
9909: if (!fA) SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"MatMatMatMult not supported for A of type %s",((PetscObject)A)->type_name);
9910: mult = fA;
9911: } else {
9912: /* dispatch based on the type of A, B and C from their PetscObject's PetscFunctionLists. */
9913: char multname[256];
9914: PetscStrncpy(multname,"MatMatMatMult_",sizeof(multname));
9915: PetscStrlcat(multname,((PetscObject)A)->type_name,sizeof(multname));
9916: PetscStrlcat(multname,"_",sizeof(multname));
9917: PetscStrlcat(multname,((PetscObject)B)->type_name,sizeof(multname));
9918: PetscStrlcat(multname,"_",sizeof(multname));
9919: PetscStrlcat(multname,((PetscObject)C)->type_name,sizeof(multname));
9920: PetscStrlcat(multname,"_C",sizeof(multname));
9921: PetscObjectQueryFunction((PetscObject)B,multname,&mult);
9922: if (!mult) SETERRQ3(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_INCOMP,"MatMatMatMult requires A, %s, to be compatible with B, %s, C, %s",((PetscObject)A)->type_name,((PetscObject)B)->type_name,((PetscObject)C)->type_name);
9923: }
9924: PetscLogEventBegin(MAT_MatMatMult,A,B,0,0);
9925: (*mult)(A,B,C,scall,fill,D);
9926: PetscLogEventEnd(MAT_MatMatMult,A,B,0,0);
9927: return(0);
9928: }
9930: /*@
9931: MatCreateRedundantMatrix - Create redundant matrices and put them into processors of subcommunicators.
9933: Collective on Mat
9935: Input Parameters:
9936: + mat - the matrix
9937: . nsubcomm - the number of subcommunicators (= number of redundant parallel or sequential matrices)
9938: . subcomm - MPI communicator split from the communicator where mat resides in (or MPI_COMM_NULL if nsubcomm is used)
9939: - reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
9941: Output Parameter:
9942: . matredundant - redundant matrix
9944: Notes:
9945: MAT_REUSE_MATRIX can only be used when the nonzero structure of the
9946: original matrix has not changed from that last call to MatCreateRedundantMatrix().
9948: This routine creates the duplicated matrices in subcommunicators; you should NOT create them before
9949: calling it.
9951: Level: advanced
9954: .seealso: MatDestroy()
9955: @*/
9956: PetscErrorCode MatCreateRedundantMatrix(Mat mat,PetscInt nsubcomm,MPI_Comm subcomm,MatReuse reuse,Mat *matredundant)
9957: {
9959: MPI_Comm comm;
9960: PetscMPIInt size;
9961: PetscInt mloc_sub,nloc_sub,rstart,rend,M=mat->rmap->N,N=mat->cmap->N,bs=mat->rmap->bs;
9962: Mat_Redundant *redund=NULL;
9963: PetscSubcomm psubcomm=NULL;
9964: MPI_Comm subcomm_in=subcomm;
9965: Mat *matseq;
9966: IS isrow,iscol;
9967: PetscBool newsubcomm=PETSC_FALSE;
9971: if (nsubcomm && reuse == MAT_REUSE_MATRIX) {
9974: }
9976: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
9977: if (size == 1 || nsubcomm == 1) {
9978: if (reuse == MAT_INITIAL_MATRIX) {
9979: MatDuplicate(mat,MAT_COPY_VALUES,matredundant);
9980: } else {
9981: 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");
9982: MatCopy(mat,*matredundant,SAME_NONZERO_PATTERN);
9983: }
9984: return(0);
9985: }
9987: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
9988: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
9989: MatCheckPreallocated(mat,1);
9991: PetscLogEventBegin(MAT_RedundantMat,mat,0,0,0);
9992: if (subcomm_in == MPI_COMM_NULL && reuse == MAT_INITIAL_MATRIX) { /* get subcomm if user does not provide subcomm */
9993: /* create psubcomm, then get subcomm */
9994: PetscObjectGetComm((PetscObject)mat,&comm);
9995: MPI_Comm_size(comm,&size);
9996: if (nsubcomm < 1 || nsubcomm > size) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_ARG_SIZ,"nsubcomm must between 1 and %D",size);
9998: PetscSubcommCreate(comm,&psubcomm);
9999: PetscSubcommSetNumber(psubcomm,nsubcomm);
10000: PetscSubcommSetType(psubcomm,PETSC_SUBCOMM_CONTIGUOUS);
10001: PetscSubcommSetFromOptions(psubcomm);
10002: PetscCommDuplicate(PetscSubcommChild(psubcomm),&subcomm,NULL);
10003: newsubcomm = PETSC_TRUE;
10004: PetscSubcommDestroy(&psubcomm);
10005: }
10007: /* get isrow, iscol and a local sequential matrix matseq[0] */
10008: if (reuse == MAT_INITIAL_MATRIX) {
10009: mloc_sub = PETSC_DECIDE;
10010: nloc_sub = PETSC_DECIDE;
10011: if (bs < 1) {
10012: PetscSplitOwnership(subcomm,&mloc_sub,&M);
10013: PetscSplitOwnership(subcomm,&nloc_sub,&N);
10014: } else {
10015: PetscSplitOwnershipBlock(subcomm,bs,&mloc_sub,&M);
10016: PetscSplitOwnershipBlock(subcomm,bs,&nloc_sub,&N);
10017: }
10018: MPI_Scan(&mloc_sub,&rend,1,MPIU_INT,MPI_SUM,subcomm);
10019: rstart = rend - mloc_sub;
10020: ISCreateStride(PETSC_COMM_SELF,mloc_sub,rstart,1,&isrow);
10021: ISCreateStride(PETSC_COMM_SELF,N,0,1,&iscol);
10022: } else { /* reuse == MAT_REUSE_MATRIX */
10023: 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");
10024: /* retrieve subcomm */
10025: PetscObjectGetComm((PetscObject)(*matredundant),&subcomm);
10026: redund = (*matredundant)->redundant;
10027: isrow = redund->isrow;
10028: iscol = redund->iscol;
10029: matseq = redund->matseq;
10030: }
10031: MatCreateSubMatrices(mat,1,&isrow,&iscol,reuse,&matseq);
10033: /* get matredundant over subcomm */
10034: if (reuse == MAT_INITIAL_MATRIX) {
10035: MatCreateMPIMatConcatenateSeqMat(subcomm,matseq[0],nloc_sub,reuse,matredundant);
10037: /* create a supporting struct and attach it to C for reuse */
10038: PetscNewLog(*matredundant,&redund);
10039: (*matredundant)->redundant = redund;
10040: redund->isrow = isrow;
10041: redund->iscol = iscol;
10042: redund->matseq = matseq;
10043: if (newsubcomm) {
10044: redund->subcomm = subcomm;
10045: } else {
10046: redund->subcomm = MPI_COMM_NULL;
10047: }
10048: } else {
10049: MatCreateMPIMatConcatenateSeqMat(subcomm,matseq[0],PETSC_DECIDE,reuse,matredundant);
10050: }
10051: PetscLogEventEnd(MAT_RedundantMat,mat,0,0,0);
10052: return(0);
10053: }
10055: /*@C
10056: MatGetMultiProcBlock - Create multiple [bjacobi] 'parallel submatrices' from
10057: a given 'mat' object. Each submatrix can span multiple procs.
10059: Collective on Mat
10061: Input Parameters:
10062: + mat - the matrix
10063: . subcomm - the subcommunicator obtained by com_split(comm)
10064: - scall - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10066: Output Parameter:
10067: . subMat - 'parallel submatrices each spans a given subcomm
10069: Notes:
10070: The submatrix partition across processors is dictated by 'subComm' a
10071: communicator obtained by com_split(comm). The comm_split
10072: is not restriced to be grouped with consecutive original ranks.
10074: Due the comm_split() usage, the parallel layout of the submatrices
10075: map directly to the layout of the original matrix [wrt the local
10076: row,col partitioning]. So the original 'DiagonalMat' naturally maps
10077: into the 'DiagonalMat' of the subMat, hence it is used directly from
10078: the subMat. However the offDiagMat looses some columns - and this is
10079: reconstructed with MatSetValues()
10081: Level: advanced
10084: .seealso: MatCreateSubMatrices()
10085: @*/
10086: PetscErrorCode MatGetMultiProcBlock(Mat mat, MPI_Comm subComm, MatReuse scall,Mat *subMat)
10087: {
10089: PetscMPIInt commsize,subCommSize;
10092: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&commsize);
10093: MPI_Comm_size(subComm,&subCommSize);
10094: if (subCommSize > commsize) SETERRQ2(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_OUTOFRANGE,"CommSize %D < SubCommZize %D",commsize,subCommSize);
10096: 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");
10097: PetscLogEventBegin(MAT_GetMultiProcBlock,mat,0,0,0);
10098: (*mat->ops->getmultiprocblock)(mat,subComm,scall,subMat);
10099: PetscLogEventEnd(MAT_GetMultiProcBlock,mat,0,0,0);
10100: return(0);
10101: }
10103: /*@
10104: MatGetLocalSubMatrix - Gets a reference to a submatrix specified in local numbering
10106: Not Collective
10108: Input Arguments:
10109: + mat - matrix to extract local submatrix from
10110: . isrow - local row indices for submatrix
10111: - iscol - local column indices for submatrix
10113: Output Arguments:
10114: . submat - the submatrix
10116: Level: intermediate
10118: Notes:
10119: The submat should be returned with MatRestoreLocalSubMatrix().
10121: Depending on the format of mat, the returned submat may not implement MatMult(). Its communicator may be
10122: the same as mat, it may be PETSC_COMM_SELF, or some other subcomm of mat's.
10124: The submat always implements MatSetValuesLocal(). If isrow and iscol have the same block size, then
10125: MatSetValuesBlockedLocal() will also be implemented.
10127: The mat must have had a ISLocalToGlobalMapping provided to it with MatSetLocalToGlobalMapping(). Note that
10128: matrices obtained with DMCreateMatrix() generally already have the local to global mapping provided.
10130: .seealso: MatRestoreLocalSubMatrix(), MatCreateLocalRef(), MatSetLocalToGlobalMapping()
10131: @*/
10132: PetscErrorCode MatGetLocalSubMatrix(Mat mat,IS isrow,IS iscol,Mat *submat)
10133: {
10142: if (!mat->rmap->mapping) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Matrix must have local to global mapping provided before this call");
10144: if (mat->ops->getlocalsubmatrix) {
10145: (*mat->ops->getlocalsubmatrix)(mat,isrow,iscol,submat);
10146: } else {
10147: MatCreateLocalRef(mat,isrow,iscol,submat);
10148: }
10149: return(0);
10150: }
10152: /*@
10153: MatRestoreLocalSubMatrix - Restores a reference to a submatrix specified in local numbering
10155: Not Collective
10157: Input Arguments:
10158: mat - matrix to extract local submatrix from
10159: isrow - local row indices for submatrix
10160: iscol - local column indices for submatrix
10161: submat - the submatrix
10163: Level: intermediate
10165: .seealso: MatGetLocalSubMatrix()
10166: @*/
10167: PetscErrorCode MatRestoreLocalSubMatrix(Mat mat,IS isrow,IS iscol,Mat *submat)
10168: {
10177: if (*submat) {
10179: }
10181: if (mat->ops->restorelocalsubmatrix) {
10182: (*mat->ops->restorelocalsubmatrix)(mat,isrow,iscol,submat);
10183: } else {
10184: MatDestroy(submat);
10185: }
10186: *submat = NULL;
10187: return(0);
10188: }
10190: /* --------------------------------------------------------*/
10191: /*@
10192: MatFindZeroDiagonals - Finds all the rows of a matrix that have zero or no diagonal entry in the matrix
10194: Collective on Mat
10196: Input Parameter:
10197: . mat - the matrix
10199: Output Parameter:
10200: . is - if any rows have zero diagonals this contains the list of them
10202: Level: developer
10204: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
10205: @*/
10206: PetscErrorCode MatFindZeroDiagonals(Mat mat,IS *is)
10207: {
10213: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10214: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10216: if (!mat->ops->findzerodiagonals) {
10217: Vec diag;
10218: const PetscScalar *a;
10219: PetscInt *rows;
10220: PetscInt rStart, rEnd, r, nrow = 0;
10222: MatCreateVecs(mat, &diag, NULL);
10223: MatGetDiagonal(mat, diag);
10224: MatGetOwnershipRange(mat, &rStart, &rEnd);
10225: VecGetArrayRead(diag, &a);
10226: for (r = 0; r < rEnd-rStart; ++r) if (a[r] == 0.0) ++nrow;
10227: PetscMalloc1(nrow, &rows);
10228: nrow = 0;
10229: for (r = 0; r < rEnd-rStart; ++r) if (a[r] == 0.0) rows[nrow++] = r+rStart;
10230: VecRestoreArrayRead(diag, &a);
10231: VecDestroy(&diag);
10232: ISCreateGeneral(PetscObjectComm((PetscObject) mat), nrow, rows, PETSC_OWN_POINTER, is);
10233: } else {
10234: (*mat->ops->findzerodiagonals)(mat, is);
10235: }
10236: return(0);
10237: }
10239: /*@
10240: MatFindOffBlockDiagonalEntries - Finds all the rows of a matrix that have entries outside of the main diagonal block (defined by the matrix block size)
10242: Collective on Mat
10244: Input Parameter:
10245: . mat - the matrix
10247: Output Parameter:
10248: . is - contains the list of rows with off block diagonal entries
10250: Level: developer
10252: .seealso: MatMultTranspose(), MatMultAdd(), MatMultTransposeAdd()
10253: @*/
10254: PetscErrorCode MatFindOffBlockDiagonalEntries(Mat mat,IS *is)
10255: {
10261: if (!mat->assembled) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10262: if (mat->factortype) SETERRQ(PetscObjectComm((PetscObject)mat),PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10264: 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);
10265: (*mat->ops->findoffblockdiagonalentries)(mat,is);
10266: return(0);
10267: }
10269: /*@C
10270: MatInvertBlockDiagonal - Inverts the block diagonal entries.
10272: Collective on Mat
10274: Input Parameters:
10275: . mat - the matrix
10277: Output Parameters:
10278: . values - the block inverses in column major order (FORTRAN-like)
10280: Note:
10281: This routine is not available from Fortran.
10283: Level: advanced
10285: .seealso: MatInvertBockDiagonalMat
10286: @*/
10287: PetscErrorCode MatInvertBlockDiagonal(Mat mat,const PetscScalar **values)
10288: {
10293: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10294: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10295: if (!mat->ops->invertblockdiagonal) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for type %s",((PetscObject)mat)->type_name);
10296: (*mat->ops->invertblockdiagonal)(mat,values);
10297: return(0);
10298: }
10300: /*@C
10301: MatInvertVariableBlockDiagonal - Inverts the block diagonal entries.
10303: Collective on Mat
10305: Input Parameters:
10306: + mat - the matrix
10307: . nblocks - the number of blocks
10308: - bsizes - the size of each block
10310: Output Parameters:
10311: . values - the block inverses in column major order (FORTRAN-like)
10313: Note:
10314: This routine is not available from Fortran.
10316: Level: advanced
10318: .seealso: MatInvertBockDiagonal()
10319: @*/
10320: PetscErrorCode MatInvertVariableBlockDiagonal(Mat mat,PetscInt nblocks,const PetscInt *bsizes,PetscScalar *values)
10321: {
10326: if (!mat->assembled) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for unassembled matrix");
10327: if (mat->factortype) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONGSTATE,"Not for factored matrix");
10328: if (!mat->ops->invertvariableblockdiagonal) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for type",((PetscObject)mat)->type_name);
10329: (*mat->ops->invertvariableblockdiagonal)(mat,nblocks,bsizes,values);
10330: return(0);
10331: }
10333: /*@
10334: MatInvertBlockDiagonalMat - set matrix C to be the inverted block diagonal of matrix A
10336: Collective on Mat
10338: Input Parameters:
10339: . A - the matrix
10341: Output Parameters:
10342: . C - matrix with inverted block diagonal of A. This matrix should be created and may have its type set.
10344: Notes: the blocksize of the matrix is used to determine the blocks on the diagonal of C
10346: Level: advanced
10348: .seealso: MatInvertBockDiagonal()
10349: @*/
10350: PetscErrorCode MatInvertBlockDiagonalMat(Mat A,Mat C)
10351: {
10352: PetscErrorCode ierr;
10353: const PetscScalar *vals;
10354: PetscInt *dnnz;
10355: PetscInt M,N,m,n,rstart,rend,bs,i,j;
10358: MatInvertBlockDiagonal(A,&vals);
10359: MatGetBlockSize(A,&bs);
10360: MatGetSize(A,&M,&N);
10361: MatGetLocalSize(A,&m,&n);
10362: MatSetSizes(C,m,n,M,N);
10363: MatSetBlockSize(C,bs);
10364: PetscMalloc1(m/bs,&dnnz);
10365: for (j = 0; j < m/bs; j++) dnnz[j] = 1;
10366: MatXAIJSetPreallocation(C,bs,dnnz,NULL,NULL,NULL);
10367: PetscFree(dnnz);
10368: MatGetOwnershipRange(C,&rstart,&rend);
10369: MatSetOption(C,MAT_ROW_ORIENTED,PETSC_FALSE);
10370: for (i = rstart/bs; i < rend/bs; i++) {
10371: MatSetValuesBlocked(C,1,&i,1,&i,&vals[(i-rstart/bs)*bs*bs],INSERT_VALUES);
10372: }
10373: MatAssemblyBegin(C,MAT_FINAL_ASSEMBLY);
10374: MatAssemblyEnd(C,MAT_FINAL_ASSEMBLY);
10375: MatSetOption(C,MAT_ROW_ORIENTED,PETSC_TRUE);
10376: return(0);
10377: }
10379: /*@C
10380: MatTransposeColoringDestroy - Destroys a coloring context for matrix product C=A*B^T that was created
10381: via MatTransposeColoringCreate().
10383: Collective on MatTransposeColoring
10385: Input Parameter:
10386: . c - coloring context
10388: Level: intermediate
10390: .seealso: MatTransposeColoringCreate()
10391: @*/
10392: PetscErrorCode MatTransposeColoringDestroy(MatTransposeColoring *c)
10393: {
10394: PetscErrorCode ierr;
10395: MatTransposeColoring matcolor=*c;
10398: if (!matcolor) return(0);
10399: if (--((PetscObject)matcolor)->refct > 0) {matcolor = 0; return(0);}
10401: PetscFree3(matcolor->ncolumns,matcolor->nrows,matcolor->colorforrow);
10402: PetscFree(matcolor->rows);
10403: PetscFree(matcolor->den2sp);
10404: PetscFree(matcolor->colorforcol);
10405: PetscFree(matcolor->columns);
10406: if (matcolor->brows>0) {
10407: PetscFree(matcolor->lstart);
10408: }
10409: PetscHeaderDestroy(c);
10410: return(0);
10411: }
10413: /*@C
10414: MatTransColoringApplySpToDen - Given a symbolic matrix product C=A*B^T for which
10415: a MatTransposeColoring context has been created, computes a dense B^T by Apply
10416: MatTransposeColoring to sparse B.
10418: Collective on MatTransposeColoring
10420: Input Parameters:
10421: + B - sparse matrix B
10422: . Btdense - symbolic dense matrix B^T
10423: - coloring - coloring context created with MatTransposeColoringCreate()
10425: Output Parameter:
10426: . Btdense - dense matrix B^T
10428: Level: advanced
10430: Notes:
10431: These are used internally for some implementations of MatRARt()
10433: .seealso: MatTransposeColoringCreate(), MatTransposeColoringDestroy(), MatTransColoringApplyDenToSp()
10435: @*/
10436: PetscErrorCode MatTransColoringApplySpToDen(MatTransposeColoring coloring,Mat B,Mat Btdense)
10437: {
10445: if (!B->ops->transcoloringapplysptoden) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for this matrix type %s",((PetscObject)B)->type_name);
10446: (B->ops->transcoloringapplysptoden)(coloring,B,Btdense);
10447: return(0);
10448: }
10450: /*@C
10451: MatTransColoringApplyDenToSp - Given a symbolic matrix product Csp=A*B^T for which
10452: a MatTransposeColoring context has been created and a dense matrix Cden=A*Btdense
10453: in which Btdens is obtained from MatTransColoringApplySpToDen(), recover sparse matrix
10454: Csp from Cden.
10456: Collective on MatTransposeColoring
10458: Input Parameters:
10459: + coloring - coloring context created with MatTransposeColoringCreate()
10460: - Cden - matrix product of a sparse matrix and a dense matrix Btdense
10462: Output Parameter:
10463: . Csp - sparse matrix
10465: Level: advanced
10467: Notes:
10468: These are used internally for some implementations of MatRARt()
10470: .seealso: MatTransposeColoringCreate(), MatTransposeColoringDestroy(), MatTransColoringApplySpToDen()
10472: @*/
10473: PetscErrorCode MatTransColoringApplyDenToSp(MatTransposeColoring matcoloring,Mat Cden,Mat Csp)
10474: {
10482: if (!Csp->ops->transcoloringapplydentosp) SETERRQ1(PETSC_COMM_SELF,PETSC_ERR_SUP,"Not supported for this matrix type %s",((PetscObject)Csp)->type_name);
10483: (Csp->ops->transcoloringapplydentosp)(matcoloring,Cden,Csp);
10484: return(0);
10485: }
10487: /*@C
10488: MatTransposeColoringCreate - Creates a matrix coloring context for matrix product C=A*B^T.
10490: Collective on Mat
10492: Input Parameters:
10493: + mat - the matrix product C
10494: - iscoloring - the coloring of the matrix; usually obtained with MatColoringCreate() or DMCreateColoring()
10496: Output Parameter:
10497: . color - the new coloring context
10499: Level: intermediate
10501: .seealso: MatTransposeColoringDestroy(), MatTransColoringApplySpToDen(),
10502: MatTransColoringApplyDenToSp()
10503: @*/
10504: PetscErrorCode MatTransposeColoringCreate(Mat mat,ISColoring iscoloring,MatTransposeColoring *color)
10505: {
10506: MatTransposeColoring c;
10507: MPI_Comm comm;
10508: PetscErrorCode ierr;
10511: PetscLogEventBegin(MAT_TransposeColoringCreate,mat,0,0,0);
10512: PetscObjectGetComm((PetscObject)mat,&comm);
10513: PetscHeaderCreate(c,MAT_TRANSPOSECOLORING_CLASSID,"MatTransposeColoring","Matrix product C=A*B^T via coloring","Mat",comm,MatTransposeColoringDestroy,NULL);
10515: c->ctype = iscoloring->ctype;
10516: if (mat->ops->transposecoloringcreate) {
10517: (*mat->ops->transposecoloringcreate)(mat,iscoloring,c);
10518: } else SETERRQ1(PetscObjectComm((PetscObject)mat),PETSC_ERR_SUP,"Code not yet written for matrix type %s",((PetscObject)mat)->type_name);
10520: *color = c;
10521: PetscLogEventEnd(MAT_TransposeColoringCreate,mat,0,0,0);
10522: return(0);
10523: }
10525: /*@
10526: MatGetNonzeroState - Returns a 64 bit integer representing the current state of nonzeros in the matrix. If the
10527: matrix has had no new nonzero locations added to the matrix since the previous call then the value will be the
10528: same, otherwise it will be larger
10530: Not Collective
10532: Input Parameter:
10533: . A - the matrix
10535: Output Parameter:
10536: . state - the current state
10538: Notes:
10539: You can only compare states from two different calls to the SAME matrix, you cannot compare calls between
10540: different matrices
10542: Level: intermediate
10544: @*/
10545: PetscErrorCode MatGetNonzeroState(Mat mat,PetscObjectState *state)
10546: {
10549: *state = mat->nonzerostate;
10550: return(0);
10551: }
10553: /*@
10554: MatCreateMPIMatConcatenateSeqMat - Creates a single large PETSc matrix by concatenating sequential
10555: matrices from each processor
10557: Collective
10559: Input Parameters:
10560: + comm - the communicators the parallel matrix will live on
10561: . seqmat - the input sequential matrices
10562: . n - number of local columns (or PETSC_DECIDE)
10563: - reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10565: Output Parameter:
10566: . mpimat - the parallel matrix generated
10568: Level: advanced
10570: Notes:
10571: The number of columns of the matrix in EACH processor MUST be the same.
10573: @*/
10574: PetscErrorCode MatCreateMPIMatConcatenateSeqMat(MPI_Comm comm,Mat seqmat,PetscInt n,MatReuse reuse,Mat *mpimat)
10575: {
10579: if (!seqmat->ops->creatempimatconcatenateseqmat) SETERRQ1(PetscObjectComm((PetscObject)seqmat),PETSC_ERR_SUP,"Mat type %s",((PetscObject)seqmat)->type_name);
10580: 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");
10582: PetscLogEventBegin(MAT_Merge,seqmat,0,0,0);
10583: (*seqmat->ops->creatempimatconcatenateseqmat)(comm,seqmat,n,reuse,mpimat);
10584: PetscLogEventEnd(MAT_Merge,seqmat,0,0,0);
10585: return(0);
10586: }
10588: /*@
10589: MatSubdomainsCreateCoalesce - Creates index subdomains by coalescing adjacent
10590: ranks' ownership ranges.
10592: Collective on A
10594: Input Parameters:
10595: + A - the matrix to create subdomains from
10596: - N - requested number of subdomains
10599: Output Parameters:
10600: + n - number of subdomains resulting on this rank
10601: - iss - IS list with indices of subdomains on this rank
10603: Level: advanced
10605: Notes:
10606: number of subdomains must be smaller than the communicator size
10607: @*/
10608: PetscErrorCode MatSubdomainsCreateCoalesce(Mat A,PetscInt N,PetscInt *n,IS *iss[])
10609: {
10610: MPI_Comm comm,subcomm;
10611: PetscMPIInt size,rank,color;
10612: PetscInt rstart,rend,k;
10613: PetscErrorCode ierr;
10616: PetscObjectGetComm((PetscObject)A,&comm);
10617: MPI_Comm_size(comm,&size);
10618: MPI_Comm_rank(comm,&rank);
10619: 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);
10620: *n = 1;
10621: k = ((PetscInt)size)/N + ((PetscInt)size%N>0); /* There are up to k ranks to a color */
10622: color = rank/k;
10623: MPI_Comm_split(comm,color,rank,&subcomm);
10624: PetscMalloc1(1,iss);
10625: MatGetOwnershipRange(A,&rstart,&rend);
10626: ISCreateStride(subcomm,rend-rstart,rstart,1,iss[0]);
10627: MPI_Comm_free(&subcomm);
10628: return(0);
10629: }
10631: /*@
10632: MatGalerkin - Constructs the coarse grid problem via Galerkin projection.
10634: If the interpolation and restriction operators are the same, uses MatPtAP.
10635: If they are not the same, use MatMatMatMult.
10637: Once the coarse grid problem is constructed, correct for interpolation operators
10638: that are not of full rank, which can legitimately happen in the case of non-nested
10639: geometric multigrid.
10641: Input Parameters:
10642: + restrct - restriction operator
10643: . dA - fine grid matrix
10644: . interpolate - interpolation operator
10645: . reuse - either MAT_INITIAL_MATRIX or MAT_REUSE_MATRIX
10646: - fill - expected fill, use PETSC_DEFAULT if you do not have a good estimate
10648: Output Parameters:
10649: . A - the Galerkin coarse matrix
10651: Options Database Key:
10652: . -pc_mg_galerkin <both,pmat,mat,none>
10654: Level: developer
10656: .seealso: MatPtAP(), MatMatMatMult()
10657: @*/
10658: PetscErrorCode MatGalerkin(Mat restrct, Mat dA, Mat interpolate, MatReuse reuse, PetscReal fill, Mat *A)
10659: {
10661: IS zerorows;
10662: Vec diag;
10665: if (reuse == MAT_INPLACE_MATRIX) SETERRQ(PetscObjectComm((PetscObject)A),PETSC_ERR_SUP,"Inplace product not supported");
10666: /* Construct the coarse grid matrix */
10667: if (interpolate == restrct) {
10668: MatPtAP(dA,interpolate,reuse,fill,A);
10669: } else {
10670: MatMatMatMult(restrct,dA,interpolate,reuse,fill,A);
10671: }
10673: /* If the interpolation matrix is not of full rank, A will have zero rows.
10674: This can legitimately happen in the case of non-nested geometric multigrid.
10675: In that event, we set the rows of the matrix to the rows of the identity,
10676: ignoring the equations (as the RHS will also be zero). */
10678: MatFindZeroRows(*A, &zerorows);
10680: if (zerorows != NULL) { /* if there are any zero rows */
10681: MatCreateVecs(*A, &diag, NULL);
10682: MatGetDiagonal(*A, diag);
10683: VecISSet(diag, zerorows, 1.0);
10684: MatDiagonalSet(*A, diag, INSERT_VALUES);
10685: VecDestroy(&diag);
10686: ISDestroy(&zerorows);
10687: }
10688: return(0);
10689: }
10691: /*@C
10692: MatSetOperation - Allows user to set a matrix operation for any matrix type
10694: Logically Collective on Mat
10696: Input Parameters:
10697: + mat - the matrix
10698: . op - the name of the operation
10699: - f - the function that provides the operation
10701: Level: developer
10703: Usage:
10704: $ extern PetscErrorCode usermult(Mat,Vec,Vec);
10705: $ MatCreateXXX(comm,...&A);
10706: $ MatSetOperation(A,MATOP_MULT,(void(*)(void))usermult);
10708: Notes:
10709: See the file include/petscmat.h for a complete list of matrix
10710: operations, which all have the form MATOP_<OPERATION>, where
10711: <OPERATION> is the name (in all capital letters) of the
10712: user interface routine (e.g., MatMult() -> MATOP_MULT).
10714: All user-provided functions (except for MATOP_DESTROY) should have the same calling
10715: sequence as the usual matrix interface routines, since they
10716: are intended to be accessed via the usual matrix interface
10717: routines, e.g.,
10718: $ MatMult(Mat,Vec,Vec) -> usermult(Mat,Vec,Vec)
10720: In particular each function MUST return an error code of 0 on success and
10721: nonzero on failure.
10723: This routine is distinct from MatShellSetOperation() in that it can be called on any matrix type.
10725: .seealso: MatGetOperation(), MatCreateShell(), MatShellSetContext(), MatShellSetOperation()
10726: @*/
10727: PetscErrorCode MatSetOperation(Mat mat,MatOperation op,void (*f)(void))
10728: {
10731: if (op == MATOP_VIEW && !mat->ops->viewnative && f != (void (*)(void))(mat->ops->view)) {
10732: mat->ops->viewnative = mat->ops->view;
10733: }
10734: (((void(**)(void))mat->ops)[op]) = f;
10735: return(0);
10736: }
10738: /*@C
10739: MatGetOperation - Gets a matrix operation for any matrix type.
10741: Not Collective
10743: Input Parameters:
10744: + mat - the matrix
10745: - op - the name of the operation
10747: Output Parameter:
10748: . f - the function that provides the operation
10750: Level: developer
10752: Usage:
10753: $ PetscErrorCode (*usermult)(Mat,Vec,Vec);
10754: $ MatGetOperation(A,MATOP_MULT,(void(**)(void))&usermult);
10756: Notes:
10757: See the file include/petscmat.h for a complete list of matrix
10758: operations, which all have the form MATOP_<OPERATION>, where
10759: <OPERATION> is the name (in all capital letters) of the
10760: user interface routine (e.g., MatMult() -> MATOP_MULT).
10762: This routine is distinct from MatShellGetOperation() in that it can be called on any matrix type.
10764: .seealso: MatSetOperation(), MatCreateShell(), MatShellGetContext(), MatShellGetOperation()
10765: @*/
10766: PetscErrorCode MatGetOperation(Mat mat,MatOperation op,void(**f)(void))
10767: {
10770: *f = (((void (**)(void))mat->ops)[op]);
10771: return(0);
10772: }
10774: /*@
10775: MatHasOperation - Determines whether the given matrix supports the particular
10776: operation.
10778: Not Collective
10780: Input Parameters:
10781: + mat - the matrix
10782: - op - the operation, for example, MATOP_GET_DIAGONAL
10784: Output Parameter:
10785: . has - either PETSC_TRUE or PETSC_FALSE
10787: Level: advanced
10789: Notes:
10790: See the file include/petscmat.h for a complete list of matrix
10791: operations, which all have the form MATOP_<OPERATION>, where
10792: <OPERATION> is the name (in all capital letters) of the
10793: user-level routine. E.g., MatNorm() -> MATOP_NORM.
10795: .seealso: MatCreateShell()
10796: @*/
10797: PetscErrorCode MatHasOperation(Mat mat,MatOperation op,PetscBool *has)
10798: {
10805: if (mat->ops->hasoperation) {
10806: (*mat->ops->hasoperation)(mat,op,has);
10807: } else {
10808: if (((void**)mat->ops)[op]) *has = PETSC_TRUE;
10809: else {
10810: *has = PETSC_FALSE;
10811: if (op == MATOP_CREATE_SUBMATRIX) {
10812: PetscMPIInt size;
10814: MPI_Comm_size(PetscObjectComm((PetscObject)mat),&size);
10815: if (size == 1) {
10816: MatHasOperation(mat,MATOP_CREATE_SUBMATRICES,has);
10817: }
10818: }
10819: }
10820: }
10821: return(0);
10822: }
10824: /*@
10825: MatHasCongruentLayouts - Determines whether the rows and columns layouts
10826: of the matrix are congruent
10828: Collective on mat
10830: Input Parameters:
10831: . mat - the matrix
10833: Output Parameter:
10834: . cong - either PETSC_TRUE or PETSC_FALSE
10836: Level: beginner
10838: Notes:
10840: .seealso: MatCreate(), MatSetSizes()
10841: @*/
10842: PetscErrorCode MatHasCongruentLayouts(Mat mat,PetscBool *cong)
10843: {
10850: if (!mat->rmap || !mat->cmap) {
10851: *cong = mat->rmap == mat->cmap ? PETSC_TRUE : PETSC_FALSE;
10852: return(0);
10853: }
10854: if (mat->congruentlayouts == PETSC_DECIDE) { /* first time we compare rows and cols layouts */
10855: PetscLayoutCompare(mat->rmap,mat->cmap,cong);
10856: if (*cong) mat->congruentlayouts = 1;
10857: else mat->congruentlayouts = 0;
10858: } else *cong = mat->congruentlayouts ? PETSC_TRUE : PETSC_FALSE;
10859: return(0);
10860: }
10862: /*@
10863: MatFreeIntermediateDataStructures - Free intermediate data structures created for reuse,
10864: e.g., matrx product of MatPtAP.
10866: Collective on mat
10868: Input Parameters:
10869: . mat - the matrix
10871: Output Parameter:
10872: . mat - the matrix with intermediate data structures released
10874: Level: advanced
10876: Notes:
10878: .seealso: MatPtAP(), MatMatMult()
10879: @*/
10880: PetscErrorCode MatFreeIntermediateDataStructures(Mat mat)
10881: {
10887: if (mat->ops->freeintermediatedatastructures) {
10888: (*mat->ops->freeintermediatedatastructures)(mat);
10889: }
10890: return(0);
10891: }