1: /*
2: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
3: SLEPc - Scalable Library for Eigenvalue Problem Computations
4: Copyright (c) 2002-2020, Universitat Politecnica de Valencia, Spain
6: This file is part of SLEPc.
7: SLEPc is distributed under a 2-clause BSD license (see LICENSE).
8: - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
9: */
10: /*
11: ST interface routines, callable by users
12: */
14: #include <slepc/private/stimpl.h> /*I "slepcst.h" I*/
16: PetscErrorCode STApply_Generic(ST st,Vec x,Vec y) 17: {
21: if (st->M && st->P) {
22: MatMult(st->M,x,st->work[0]);
23: STMatSolve(st,st->work[0],y);
24: } else if (st->M) {
25: MatMult(st->M,x,y);
26: } else {
27: STMatSolve(st,x,y);
28: }
29: return(0);
30: }
32: /*@
33: STApply - Applies the spectral transformation operator to a vector, for
34: instance (A - sB)^-1 B in the case of the shift-and-invert transformation
35: and generalized eigenproblem.
37: Collective on st
39: Input Parameters:
40: + st - the spectral transformation context
41: - x - input vector
43: Output Parameter:
44: . y - output vector
46: Level: developer
48: .seealso: STApplyTranspose(), STApplyHermitianTranspose()
49: @*/
50: PetscErrorCode STApply(ST st,Vec x,Vec y) 51: {
53: Mat Op;
60: STCheckMatrices(st,1);
61: if (x == y) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_IDN,"x and y must be different vectors");
62: VecSetErrorIfLocked(y,3);
63: if (!st->ops->apply) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_SUP,"ST does not have apply");
64: STGetOperator_Private(st,&Op);
65: MatMult(Op,x,y);
66: return(0);
67: }
69: PetscErrorCode STApplyTranspose_Generic(ST st,Vec x,Vec y) 70: {
74: if (st->M && st->P) {
75: STMatSolveTranspose(st,x,st->work[0]);
76: MatMultTranspose(st->M,st->work[0],y);
77: } else if (st->M) {
78: MatMultTranspose(st->M,x,y);
79: } else {
80: STMatSolveTranspose(st,x,y);
81: }
82: return(0);
83: }
85: /*@
86: STApplyTranspose - Applies the transpose of the operator to a vector, for
87: instance B^T(A - sB)^-T in the case of the shift-and-invert transformation
88: and generalized eigenproblem.
90: Collective on st
92: Input Parameters:
93: + st - the spectral transformation context
94: - x - input vector
96: Output Parameter:
97: . y - output vector
99: Level: developer
101: .seealso: STApply(), STApplyHermitianTranspose()
102: @*/
103: PetscErrorCode STApplyTranspose(ST st,Vec x,Vec y)104: {
106: Mat Op;
113: STCheckMatrices(st,1);
114: if (x == y) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_IDN,"x and y must be different vectors");
115: VecSetErrorIfLocked(y,3);
116: if (!st->ops->applytrans) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_SUP,"ST does not have applytrans");
117: STGetOperator_Private(st,&Op);
118: MatMultTranspose(Op,x,y);
119: return(0);
120: }
122: /*@
123: STApplyHermitianTranspose - Applies the hermitian-transpose of the operator
124: to a vector, for instance B^H(A - sB)^-H in the case of the shift-and-invert
125: transformation and generalized eigenproblem.
127: Collective on st
129: Input Parameters:
130: + st - the spectral transformation context
131: - x - input vector
133: Output Parameter:
134: . y - output vector
136: Note:
137: Currently implemented via STApplyTranspose() with appropriate conjugation.
139: Level: developer
141: .seealso: STApply(), STApplyTranspose()
142: @*/
143: PetscErrorCode STApplyHermitianTranspose(ST st,Vec x,Vec y)144: {
146: Mat Op;
153: STCheckMatrices(st,1);
154: if (x == y) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_IDN,"x and y must be different vectors");
155: VecSetErrorIfLocked(y,3);
156: if (!st->ops->applytrans) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_SUP,"ST does not have applytrans");
157: STGetOperator_Private(st,&Op);
158: MatMultHermitianTranspose(Op,x,y);
159: return(0);
160: }
162: /*@
163: STGetBilinearForm - Returns the matrix used in the bilinear form with a
164: generalized problem with semi-definite B.
166: Not collective, though a parallel Mat may be returned
168: Input Parameters:
169: . st - the spectral transformation context
171: Output Parameter:
172: . B - output matrix
174: Notes:
175: The output matrix B must be destroyed after use. It will be NULL in
176: case of standard eigenproblems.
178: Level: developer
179: @*/
180: PetscErrorCode STGetBilinearForm(ST st,Mat *B)181: {
188: STCheckMatrices(st,1);
189: (*st->ops->getbilinearform)(st,B);
190: return(0);
191: }
193: PetscErrorCode STGetBilinearForm_Default(ST st,Mat *B)194: {
198: if (st->nmat==1) *B = NULL;
199: else {
200: *B = st->A[1];
201: PetscObjectReference((PetscObject)*B);
202: }
203: return(0);
204: }
206: static PetscErrorCode MatMult_STOperator(Mat Op,Vec x,Vec y)207: {
209: ST st;
212: MatShellGetContext(Op,(void**)&st);
213: STSetUp(st);
214: PetscLogEventBegin(ST_Apply,st,x,y,0);
215: if (st->D) { /* with balancing */
216: VecPointwiseDivide(st->wb,x,st->D);
217: (*st->ops->apply)(st,st->wb,y);
218: VecPointwiseMult(y,y,st->D);
219: } else {
220: (*st->ops->apply)(st,x,y);
221: }
222: PetscLogEventEnd(ST_Apply,st,x,y,0);
223: return(0);
224: }
226: static PetscErrorCode MatMultTranspose_STOperator(Mat Op,Vec x,Vec y)227: {
229: ST st;
232: MatShellGetContext(Op,(void**)&st);
233: STSetUp(st);
234: PetscLogEventBegin(ST_ApplyTranspose,st,x,y,0);
235: if (st->D) { /* with balancing */
236: VecPointwiseMult(st->wb,x,st->D);
237: (*st->ops->applytrans)(st,st->wb,y);
238: VecPointwiseDivide(y,y,st->D);
239: } else {
240: (*st->ops->applytrans)(st,x,y);
241: }
242: PetscLogEventEnd(ST_ApplyTranspose,st,x,y,0);
243: return(0);
244: }
246: #if defined(PETSC_USE_COMPLEX)
247: static PetscErrorCode MatMultHermitianTranspose_STOperator(Mat Op,Vec x,Vec y)248: {
250: ST st;
253: MatShellGetContext(Op,(void**)&st);
254: STSetUp(st);
255: PetscLogEventBegin(ST_ApplyTranspose,st,x,y,0);
256: if (!st->wht) {
257: MatCreateVecs(st->A[0],&st->wht,NULL);
258: PetscLogObjectParent((PetscObject)st,(PetscObject)st->wht);
259: }
260: VecCopy(x,st->wht);
261: VecConjugate(st->wht);
262: if (st->D) { /* with balancing */
263: VecPointwiseMult(st->wb,st->wht,st->D);
264: (*st->ops->applytrans)(st,st->wb,y);
265: VecPointwiseDivide(y,y,st->D);
266: } else {
267: (*st->ops->applytrans)(st,st->wht,y);
268: }
269: VecConjugate(y);
270: PetscLogEventEnd(ST_ApplyTranspose,st,x,y,0);
271: return(0);
272: }
273: #endif
275: PetscErrorCode STGetOperator_Private(ST st,Mat *Op)276: {
278: PetscInt m,n,M,N;
280: if (!st->Op) {
281: MatGetLocalSize(st->A[0],&m,&n);
282: MatGetSize(st->A[0],&M,&N);
283: MatCreateShell(PetscObjectComm((PetscObject)st),m,n,M,N,st,&st->Op);
284: MatShellSetOperation(st->Op,MATOP_MULT,(void(*)(void))MatMult_STOperator);
285: MatShellSetOperation(st->Op,MATOP_MULT_TRANSPOSE,(void(*)(void))MatMultTranspose_STOperator);
286: #if defined(PETSC_USE_COMPLEX)
287: MatShellSetOperation(st->Op,MATOP_MULT_HERMITIAN_TRANSPOSE,(void(*)(void))MatMultHermitianTranspose_STOperator);
288: #else
289: MatShellSetOperation(st->Op,MATOP_MULT_HERMITIAN_TRANSPOSE,(void(*)(void))MatMultTranspose_STOperator);
290: STComputeOperator(st);
291: #endif
292: }
293: if (Op) *Op = st->Op;
294: return(0);
295: }
297: /*@
298: STGetOperator - Returns a shell matrix that represents the operator of the
299: spectral transformation.
301: Collective on st
303: Input Parameter:
304: . st - the spectral transformation context
306: Output Parameter:
307: . Op - operator matrix
309: Notes:
310: The operator is defined in linear eigenproblems only, not in polynomial ones,
311: so the call will fail if more than 2 matrices were passed in STSetMatrices().
313: The returned shell matrix is essentially a wrapper to the STApply() and
314: STApplyTranspose() operations. The operator can often be expressed as
316: .vb
317: Op = D*inv(K)*M*inv(D)
318: .ve
320: where D is the balancing matrix, and M and K are two matrices corresponding
321: to the numerator and denominator for spectral transformations that represent
322: a rational matrix function. In the case of STSHELL, the inner part inv(K)*M
323: is replaced by the user-provided operation from STShellSetApply().
325: The preconditioner matrix K typically depends on the value of the shift, and
326: its inverse is handled via an internal KSP object. Normal usage does not
327: require explicitly calling STGetOperator(), but it can be used to force the
328: creation of K and M, and then K is passed to the KSP. This is useful for
329: setting options associated with the PCFactor (to set MUMPS options, for instance).
331: The returned matrix must NOT be destroyed by the user. Instead, when no
332: longer needed it must be returned with STRestoreOperator(). In particular,
333: this is required before modifying the ST matrices or the shift.
335: A NULL pointer can be passed in Op in case the matrix is not required but we
336: want to force its creation. In this case, STRestoreOperator() should not be
337: called.
339: Level: advanced
341: .seealso: STApply(), STApplyTranspose(), STSetBalanceMatrix(), STShellSetApply(),
342: STGetKSP(), STSetShift(), STRestoreOperator(), STSetMatrices()
343: @*/
344: PetscErrorCode STGetOperator(ST st,Mat *Op)345: {
351: STCheckMatrices(st,1);
352: STCheckNotSeized(st,1);
353: if (st->nmat>2) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_WRONGSTATE,"The operator is not defined in polynomial eigenproblems");
354: STGetOperator_Private(st,Op);
355: if (Op) st->opseized = PETSC_TRUE;
356: return(0);
357: }
359: /*@
360: STRestoreOperator - Restore the previously seized operator matrix.
362: Collective on st
364: Input Parameters:
365: + st - the spectral transformation context
366: - Op - operator matrix
368: Notes:
369: The arguments must match the corresponding call to STGetOperator().
371: Level: advanced
373: .seealso: STGetOperator()
374: @*/
375: PetscErrorCode STRestoreOperator(ST st,Mat *Op)376: {
381: if (!st->opseized) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_WRONGSTATE,"Must be called after STGetOperator()");
382: *Op = NULL;
383: st->opseized = PETSC_FALSE;
384: return(0);
385: }
387: /*
388: STComputeOperator - Computes the matrices that constitute the operator
390: Op = D*inv(K)*M*inv(D).
392: K and M are computed here (D is user-provided) from the system matrices
393: and the shift sigma (whenever these are changed, this function recomputes
394: K and M). This is used only in linear eigenproblems (nmat<3).
396: K is the "preconditioner matrix": it is the denominator in rational operators,
397: e.g. (A-sigma*B) in shift-and-invert. In non-rational transformations such
398: as STFILTER, K=NULL which means identity. After computing K, it is passed to
399: the internal KSP object via KSPSetOperators.
401: M is the numerator in rational operators. If unused it is set to NULL (e.g.
402: in STPRECOND).
404: STSHELL does not compute anything here, but sets the flag as if it was ready.
405: */
406: PetscErrorCode STComputeOperator(ST st)407: {
409: PC pc;
414: if (!st->opready && st->ops->computeoperator) {
415: STCheckMatrices(st,1);
416: if (!st->T) {
417: PetscCalloc1(PetscMax(2,st->nmat),&st->T);
418: PetscLogObjectMemory((PetscObject)st,PetscMax(2,st->nmat)*sizeof(Mat));
419: }
420: PetscLogEventBegin(ST_ComputeOperator,st,0,0,0);
421: (*st->ops->computeoperator)(st);
422: PetscLogEventEnd(ST_ComputeOperator,st,0,0,0);
423: if (st->usesksp) {
424: if (!st->ksp) { STGetKSP(st,&st->ksp); }
425: if (st->P) {
426: STSetDefaultKSP(st);
427: STCheckFactorPackage(st);
428: KSPSetOperators(st->ksp,st->P,st->P);
429: } else {
430: /* STPRECOND defaults to PCNONE if st->P is empty */
431: KSPGetPC(st->ksp,&pc);
432: PCSetType(pc,PCNONE);
433: }
434: }
435: }
436: st->opready = PETSC_TRUE;
437: return(0);
438: }
440: /*@
441: STSetUp - Prepares for the use of a spectral transformation.
443: Collective on st
445: Input Parameter:
446: . st - the spectral transformation context
448: Level: advanced
450: .seealso: STCreate(), STApply(), STDestroy()
451: @*/
452: PetscErrorCode STSetUp(ST st)453: {
454: PetscInt i,n,k;
460: STCheckMatrices(st,1);
461: switch (st->state) {
462: case ST_STATE_INITIAL:
463: PetscInfo(st,"Setting up new ST\n");
464: if (!((PetscObject)st)->type_name) {
465: STSetType(st,STSHIFT);
466: }
467: break;
468: case ST_STATE_SETUP:
469: return(0);
470: case ST_STATE_UPDATED:
471: PetscInfo(st,"Setting up updated ST\n");
472: break;
473: }
474: PetscLogEventBegin(ST_SetUp,st,0,0,0);
475: if (st->state!=ST_STATE_UPDATED) {
476: if (!(st->nmat<3 && st->opready)) {
477: if (st->T) {
478: for (i=0;i<PetscMax(2,st->nmat);i++) {
479: MatDestroy(&st->T[i]);
480: }
481: }
482: MatDestroy(&st->P);
483: }
484: }
485: if (st->D) {
486: MatGetLocalSize(st->A[0],NULL,&n);
487: VecGetLocalSize(st->D,&k);
488: if (n != k) SETERRQ2(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_SIZ,"Balance matrix has wrong dimension %D (should be %D)",k,n);
489: if (!st->wb) {
490: VecDuplicate(st->D,&st->wb);
491: PetscLogObjectParent((PetscObject)st,(PetscObject)st->wb);
492: }
493: }
494: if (st->nmat<3 && st->transform) {
495: STComputeOperator(st);
496: } else {
497: if (!st->T) {
498: PetscCalloc1(PetscMax(2,st->nmat),&st->T);
499: PetscLogObjectMemory((PetscObject)st,PetscMax(2,st->nmat)*sizeof(Mat));
500: }
501: }
502: if (st->ops->setup) { (*st->ops->setup)(st); }
503: st->state = ST_STATE_SETUP;
504: PetscLogEventEnd(ST_SetUp,st,0,0,0);
505: return(0);
506: }
508: /*
509: Computes coefficients for the transformed polynomial,
510: and stores the result in argument S.
512: alpha - value of the parameter of the transformed polynomial
513: beta - value of the previous shift (only used in inplace mode)
514: k - index of first matrix included in the computation
515: coeffs - coefficients of the expansion
516: initial - true if this is the first time (only relevant for shell mode)
517: */
518: PetscErrorCode STMatMAXPY_Private(ST st,PetscScalar alpha,PetscScalar beta,PetscInt k,PetscScalar *coeffs,PetscBool initial,Mat *S)519: {
521: PetscInt *matIdx=NULL,nmat,i,ini=-1;
522: PetscScalar t=1.0,ta,gamma;
523: PetscBool nz=PETSC_FALSE;
526: nmat = st->nmat-k;
527: switch (st->matmode) {
528: case ST_MATMODE_INPLACE:
529: if (st->nmat>2) SETERRQ(PetscObjectComm((PetscObject)st),PETSC_ERR_SUP,"ST_MATMODE_INPLACE not supported for polynomial eigenproblems");
530: if (initial) {
531: PetscObjectReference((PetscObject)st->A[0]);
532: *S = st->A[0];
533: gamma = alpha;
534: } else gamma = alpha-beta;
535: if (gamma != 0.0) {
536: if (st->nmat>1) {
537: MatAXPY(*S,gamma,st->A[1],st->str);
538: } else {
539: MatShift(*S,gamma);
540: }
541: }
542: break;
543: case ST_MATMODE_SHELL:
544: if (initial) {
545: if (st->nmat>2) {
546: PetscMalloc1(nmat,&matIdx);
547: for (i=0;i<nmat;i++) matIdx[i] = k+i;
548: }
549: STMatShellCreate(st,alpha,nmat,matIdx,coeffs,S);
550: PetscLogObjectParent((PetscObject)st,(PetscObject)*S);
551: if (st->nmat>2) { PetscFree(matIdx); }
552: } else {
553: STMatShellShift(*S,alpha);
554: }
555: break;
556: case ST_MATMODE_COPY:
557: if (coeffs) {
558: for (i=0;i<nmat && ini==-1;i++) {
559: if (coeffs[i]!=0.0) ini = i;
560: else t *= alpha;
561: }
562: if (coeffs[ini] != 1.0) nz = PETSC_TRUE;
563: for (i=ini+1;i<nmat&&!nz;i++) if (coeffs[i]!=0.0) nz = PETSC_TRUE;
564: } else { nz = PETSC_TRUE; ini = 0; }
565: if ((alpha == 0.0 || !nz) && t==1.0) {
566: PetscObjectReference((PetscObject)st->A[k+ini]);
567: MatDestroy(S);
568: *S = st->A[k+ini];
569: } else {
570: if (*S && *S!=st->A[k+ini]) {
571: MatSetOption(*S,MAT_NEW_NONZERO_ALLOCATION_ERR,PETSC_FALSE);
572: MatCopy(st->A[k+ini],*S,DIFFERENT_NONZERO_PATTERN);
573: } else {
574: MatDestroy(S);
575: MatDuplicate(st->A[k+ini],MAT_COPY_VALUES,S);
576: MatSetOption(*S,MAT_NEW_NONZERO_ALLOCATION_ERR,PETSC_FALSE);
577: PetscLogObjectParent((PetscObject)st,(PetscObject)*S);
578: }
579: if (coeffs && coeffs[ini]!=1.0) {
580: MatScale(*S,coeffs[ini]);
581: }
582: for (i=ini+k+1;i<PetscMax(2,st->nmat);i++) {
583: t *= alpha;
584: ta = t;
585: if (coeffs) ta *= coeffs[i-k];
586: if (ta!=0.0) {
587: if (st->nmat>1) {
588: MatAXPY(*S,ta,st->A[i],st->str);
589: } else {
590: MatShift(*S,ta);
591: }
592: }
593: }
594: }
595: }
596: MatSetOption(*S,MAT_SYMMETRIC,st->asymm);
597: MatSetOption(*S,MAT_HERMITIAN,(PetscImaginaryPart(st->sigma)==0.0)?st->aherm:PETSC_FALSE);
598: return(0);
599: }
601: /*
602: Computes the values of the coefficients required by STMatMAXPY_Private
603: for the case of monomial basis.
604: */
605: PetscErrorCode STCoeffs_Monomial(ST st, PetscScalar *coeffs)606: {
607: PetscInt k,i,ini,inip;
610: /* Compute binomial coefficients */
611: ini = (st->nmat*(st->nmat-1))/2;
612: for (i=0;i<st->nmat;i++) coeffs[ini+i]=1.0;
613: for (k=st->nmat-1;k>=1;k--) {
614: inip = ini+1;
615: ini = (k*(k-1))/2;
616: coeffs[ini] = 1.0;
617: for (i=1;i<k;i++) coeffs[ini+i] = coeffs[ini+i-1]+coeffs[inip+i-1];
618: }
619: return(0);
620: }
622: /*@
623: STPostSolve - Optional post-solve phase, intended for any actions that must
624: be performed on the ST object after the eigensolver has finished.
626: Collective on st
628: Input Parameters:
629: . st - the spectral transformation context
631: Level: developer
633: .seealso: EPSSolve()
634: @*/
635: PetscErrorCode STPostSolve(ST st)636: {
642: if (st->ops->postsolve) {
643: (*st->ops->postsolve)(st);
644: }
645: return(0);
646: }
648: /*@
649: STBackTransform - Back-transformation phase, intended for
650: spectral transformations which require to transform the computed
651: eigenvalues back to the original eigenvalue problem.
653: Not Collective
655: Input Parameters:
656: + st - the spectral transformation context
657: eigr - real part of a computed eigenvalue
658: - eigi - imaginary part of a computed eigenvalue
660: Level: developer
662: .seealso: STIsInjective()
663: @*/
664: PetscErrorCode STBackTransform(ST st,PetscInt n,PetscScalar* eigr,PetscScalar* eigi)665: {
671: if (st->ops->backtransform) {
672: (*st->ops->backtransform)(st,n,eigr,eigi);
673: }
674: return(0);
675: }
677: /*@
678: STIsInjective - Ask if this spectral transformation is injective or not
679: (that is, if it corresponds to a one-to-one mapping). If not, then it
680: does not make sense to call STBackTransform().
682: Not collective
684: Input Parameter:
685: . st - the spectral transformation context
687: Output Parameter:
688: . is - the answer
690: Level: developer
692: .seealso: STBackTransform()
693: @*/
694: PetscErrorCode STIsInjective(ST st,PetscBool* is)695: {
697: PetscBool shell;
704: PetscObjectTypeCompare((PetscObject)st,STSHELL,&shell);
705: if (shell) {
706: STIsInjective_Shell(st,is);
707: } else *is = st->ops->backtransform? PETSC_TRUE: PETSC_FALSE;
708: return(0);
709: }
711: /*@
712: STMatSetUp - Build the preconditioner matrix used in STMatSolve().
714: Collective on st
716: Input Parameters:
717: + st - the spectral transformation context
718: . sigma - the shift
719: - coeffs - the coefficients (may be NULL)
721: Note:
722: This function is not intended to be called by end users, but by SLEPc
723: solvers that use ST. It builds matrix st->P as follows, then calls KSPSetUp().
724: .vb
725: If (coeffs): st->P = Sum_{i=0:nmat-1} coeffs[i]*sigma^i*A_i.
726: else st->P = Sum_{i=0:nmat-1} sigma^i*A_i
727: .ve
729: Level: developer
731: .seealso: STMatSolve()
732: @*/
733: PetscErrorCode STMatSetUp(ST st,PetscScalar sigma,PetscScalar *coeffs)734: {
740: STCheckMatrices(st,1);
742: PetscLogEventBegin(ST_MatSetUp,st,0,0,0);
743: STMatMAXPY_Private(st,sigma,0.0,0,coeffs,PETSC_TRUE,&st->P);
744: if (!st->ksp) { STGetKSP(st,&st->ksp); }
745: STCheckFactorPackage(st);
746: KSPSetOperators(st->ksp,st->P,st->P);
747: KSPSetUp(st->ksp);
748: PetscLogEventEnd(ST_MatSetUp,st,0,0,0);
749: return(0);
750: }
752: /*@
753: STSetWorkVecs - Sets a number of work vectors into the ST object.
755: Collective on st
757: Input Parameters:
758: + st - the spectral transformation context
759: - nw - number of work vectors to allocate
761: Developers Note:
762: This is SLEPC_EXTERN because it may be required by shell STs.
764: Level: developer
765: @*/
766: PetscErrorCode STSetWorkVecs(ST st,PetscInt nw)767: {
769: PetscInt i;
774: if (nw <= 0) SETERRQ1(PetscObjectComm((PetscObject)st),PETSC_ERR_ARG_OUTOFRANGE,"nw must be > 0: nw = %D",nw);
775: if (st->nwork < nw) {
776: VecDestroyVecs(st->nwork,&st->work);
777: st->nwork = nw;
778: PetscMalloc1(nw,&st->work);
779: for (i=0;i<nw;i++) { STMatCreateVecs(st,&st->work[i],NULL); }
780: PetscLogObjectParents(st,nw,st->work);
781: }
782: return(0);
783: }