Actual source code: lapack.c

slepc-3.6.1 2015-09-03
Report Typos and Errors
  1: /*
  2:    This file implements a wrapper to the LAPACK eigenvalue subroutines.
  3:    Generalized problems are transformed to standard ones only if necessary.

  5:    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
  6:    SLEPc - Scalable Library for Eigenvalue Problem Computations
  7:    Copyright (c) 2002-2015, Universitat Politecnica de Valencia, Spain

  9:    This file is part of SLEPc.

 11:    SLEPc is free software: you can redistribute it and/or modify it under  the
 12:    terms of version 3 of the GNU Lesser General Public License as published by
 13:    the Free Software Foundation.

 15:    SLEPc  is  distributed in the hope that it will be useful, but WITHOUT  ANY
 16:    WARRANTY;  without even the implied warranty of MERCHANTABILITY or  FITNESS
 17:    FOR  A  PARTICULAR PURPOSE. See the GNU Lesser General Public  License  for
 18:    more details.

 20:    You  should have received a copy of the GNU Lesser General  Public  License
 21:    along with SLEPc. If not, see <http://www.gnu.org/licenses/>.
 22:    - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
 23: */

 25: #include <slepc/private/epsimpl.h>

 29: PetscErrorCode EPSSetUp_LAPACK(EPS eps)
 30: {
 31:   PetscErrorCode ierr,ierra,ierrb;
 32:   PetscBool      isshift,denseok=PETSC_FALSE;
 33:   Mat            A,B,OP,Adense=NULL,Bdense=NULL;
 34:   PetscScalar    shift,*Ap,*Bp;
 35:   PetscInt       i,ld,nmat;
 36:   KSP            ksp;
 37:   PC             pc;
 38:   Vec            v;

 41:   eps->ncv = eps->n;
 42:   if (eps->mpd) { PetscInfo(eps,"Warning: parameter mpd ignored\n"); }
 43:   if (!eps->which) { EPSSetWhichEigenpairs_Default(eps); }
 44:   if (eps->balance!=EPS_BALANCE_NONE) { PetscInfo(eps,"Warning: balancing ignored\n"); }
 45:   if (eps->extraction) { PetscInfo(eps,"Warning: extraction type ignored\n"); }
 46:   EPSAllocateSolution(eps,0);

 48:   /* attempt to get dense representations of A and B separately */
 49:   PetscObjectTypeCompare((PetscObject)eps->st,STSHIFT,&isshift);
 50:   if (isshift) {
 51:     STGetNumMatrices(eps->st,&nmat);
 52:     STGetOperators(eps->st,0,&A);
 53:     if (nmat>1) { STGetOperators(eps->st,1,&B); }
 54:     PetscPushErrorHandler(PetscIgnoreErrorHandler,NULL);
 55:     ierra = SlepcMatConvertSeqDense(A,&Adense);
 56:     if (eps->isgeneralized) {
 57:       ierrb = SlepcMatConvertSeqDense(B,&Bdense);
 58:     } else {
 59:       ierrb = 0;
 60:     }
 61:     PetscPopErrorHandler();
 62:     denseok = (ierra == 0 && ierrb == 0)? PETSC_TRUE: PETSC_FALSE;
 63:   }

 65:   /* setup DS */
 66:   if (denseok) {
 67:     if (eps->isgeneralized) {
 68:       if (eps->ishermitian) {
 69:         if (eps->ispositive) {
 70:           DSSetType(eps->ds,DSGHEP);
 71:         } else {
 72:           DSSetType(eps->ds,DSGNHEP); /* TODO: should be DSGHIEP */
 73:         }
 74:       } else {
 75:         DSSetType(eps->ds,DSGNHEP);
 76:       }
 77:     } else {
 78:       if (eps->ishermitian) {
 79:         DSSetType(eps->ds,DSHEP);
 80:       } else {
 81:         DSSetType(eps->ds,DSNHEP);
 82:       }
 83:     }
 84:   } else {
 85:     DSSetType(eps->ds,DSNHEP);
 86:   }
 87:   DSAllocate(eps->ds,eps->ncv);
 88:   DSGetLeadingDimension(eps->ds,&ld);
 89:   DSSetDimensions(eps->ds,eps->ncv,0,0,0);

 91:   if (denseok) {
 92:     STGetShift(eps->st,&shift);
 93:     if (shift != 0.0) {
 94:       MatShift(Adense,shift);
 95:     }
 96:     /* use dummy pc and ksp to avoid problems when B is not positive definite */
 97:     STGetKSP(eps->st,&ksp);
 98:     KSPSetType(ksp,KSPPREONLY);
 99:     KSPGetPC(ksp,&pc);
100:     PCSetType(pc,PCNONE);
101:   } else {
102:     PetscInfo(eps,"Using slow explicit operator\n");
103:     STComputeExplicitOperator(eps->st,&OP);
104:     MatDestroy(&Adense);
105:     SlepcMatConvertSeqDense(OP,&Adense);
106:   }

108:   /* fill DS matrices */
109:   VecCreateSeqWithArray(PETSC_COMM_SELF,1,ld,NULL,&v);
110:   DSGetArray(eps->ds,DS_MAT_A,&Ap);
111:   for (i=0;i<ld;i++) {
112:     VecPlaceArray(v,Ap+i*ld);
113:     MatGetColumnVector(Adense,v,i);
114:     VecResetArray(v);
115:   }
116:   DSRestoreArray(eps->ds,DS_MAT_A,&Ap);
117:   if (denseok && eps->isgeneralized) {
118:     DSGetArray(eps->ds,DS_MAT_B,&Bp);
119:     for (i=0;i<ld;i++) {
120:       VecPlaceArray(v,Bp+i*ld);
121:       MatGetColumnVector(Bdense,v,i);
122:       VecResetArray(v);
123:     }
124:     DSRestoreArray(eps->ds,DS_MAT_B,&Bp);
125:   }
126:   VecDestroy(&v);
127:   MatDestroy(&Adense);
128:   if (!denseok) { MatDestroy(&OP); }
129:   if (denseok && eps->isgeneralized) { MatDestroy(&Bdense); }
130:   return(0);
131: }

135: PetscErrorCode EPSSolve_LAPACK(EPS eps)
136: {
138:   PetscInt       n=eps->n,i,low,high;
139:   PetscScalar    *array,*pX;
140:   Vec            v;

143:   DSSolve(eps->ds,eps->eigr,eps->eigi);
144:   DSSort(eps->ds,eps->eigr,eps->eigi,NULL,NULL,NULL);

146:   /* right eigenvectors */
147:   DSVectors(eps->ds,DS_MAT_X,NULL,NULL);
148:   DSGetArray(eps->ds,DS_MAT_X,&pX);
149:   for (i=0;i<eps->ncv;i++) {
150:     BVGetColumn(eps->V,i,&v);
151:     VecGetOwnershipRange(v,&low,&high);
152:     VecGetArray(v,&array);
153:     PetscMemcpy(array,pX+i*n+low,(high-low)*sizeof(PetscScalar));
154:     VecRestoreArray(v,&array);
155:     BVRestoreColumn(eps->V,i,&v);
156:   }
157:   DSRestoreArray(eps->ds,DS_MAT_X,&pX);

159:   eps->nconv  = eps->ncv;
160:   eps->its    = 1;
161:   eps->reason = EPS_CONVERGED_TOL;
162:   return(0);
163: }

167: PETSC_EXTERN PetscErrorCode EPSCreate_LAPACK(EPS eps)
168: {
170:   eps->ops->solve                = EPSSolve_LAPACK;
171:   eps->ops->setup                = EPSSetUp_LAPACK;
172:   eps->ops->backtransform        = EPSBackTransform_Default;
173:   return(0);
174: }