Actual source code: ml.c
petsc-dev 2014-02-02
2: /*
3: Provides an interface to the ML smoothed Aggregation
4: Note: Something non-obvious breaks -pc_mg_type ADDITIVE for parallel runs
5: Jed Brown, see [PETSC #18321, #18449].
6: */
7: #include <petsc-private/pcimpl.h> /*I "petscpc.h" I*/
8: #include <../src/ksp/pc/impls/mg/mgimpl.h> /*I "petscksp.h" I*/
9: #include <../src/mat/impls/aij/seq/aij.h>
10: #include <../src/mat/impls/aij/mpi/mpiaij.h>
11: #include <petscdm.h> /* for DMDestroy(&pc->mg) hack */
13: EXTERN_C_BEGIN
14: /* HAVE_CONFIG_H flag is required by ML include files */
15: #if !defined(HAVE_CONFIG_H)
16: #define HAVE_CONFIG_H
17: #endif
18: #include <ml_include.h>
19: #include <ml_viz_stats.h>
20: EXTERN_C_END
22: typedef enum {PCML_NULLSPACE_AUTO,PCML_NULLSPACE_USER,PCML_NULLSPACE_BLOCK,PCML_NULLSPACE_SCALAR} PCMLNullSpaceType;
23: static const char *const PCMLNullSpaceTypes[] = {"AUTO","USER","BLOCK","SCALAR","PCMLNullSpaceType","PCML_NULLSPACE_",0};
25: /* The context (data structure) at each grid level */
26: typedef struct {
27: Vec x,b,r; /* global vectors */
28: Mat A,P,R;
29: KSP ksp;
30: Vec coords; /* projected by ML, if PCSetCoordinates is called; values packed by node */
31: } GridCtx;
33: /* The context used to input PETSc matrix into ML at fine grid */
34: typedef struct {
35: Mat A; /* Petsc matrix in aij format */
36: Mat Aloc; /* local portion of A to be used by ML */
37: Vec x,y;
38: ML_Operator *mlmat;
39: PetscScalar *pwork; /* tmp array used by PetscML_comm() */
40: } FineGridCtx;
42: /* The context associates a ML matrix with a PETSc shell matrix */
43: typedef struct {
44: Mat A; /* PETSc shell matrix associated with mlmat */
45: ML_Operator *mlmat; /* ML matrix assorciated with A */
46: Vec y, work;
47: } Mat_MLShell;
49: /* Private context for the ML preconditioner */
50: typedef struct {
51: ML *ml_object;
52: ML_Aggregate *agg_object;
53: GridCtx *gridctx;
54: FineGridCtx *PetscMLdata;
55: PetscInt Nlevels,MaxNlevels,MaxCoarseSize,CoarsenScheme,EnergyMinimization,MinPerProc,PutOnSingleProc,RepartitionType,ZoltanScheme;
56: PetscReal Threshold,DampingFactor,EnergyMinimizationDropTol,MaxMinRatio,AuxThreshold;
57: PetscBool SpectralNormScheme_Anorm,BlockScaling,EnergyMinimizationCheap,Symmetrize,OldHierarchy,KeepAggInfo,Reusable,Repartition,Aux;
58: PetscBool reuse_interpolation;
59: PCMLNullSpaceType nulltype;
60: PetscMPIInt size; /* size of communicator for pc->pmat */
61: PetscInt dim; /* data from PCSetCoordinates(_ML) */
62: PetscInt nloc;
63: PetscReal *coords; /* ML has a grid object for each level: the finest grid will point into coords */
64: } PC_ML;
68: static int PetscML_getrow(ML_Operator *ML_data, int N_requested_rows, int requested_rows[],int allocated_space, int columns[], double values[], int row_lengths[])
69: {
71: PetscInt m,i,j,k=0,row,*aj;
72: PetscScalar *aa;
73: FineGridCtx *ml=(FineGridCtx*)ML_Get_MyGetrowData(ML_data);
74: Mat_SeqAIJ *a = (Mat_SeqAIJ*)ml->Aloc->data;
76: MatGetSize(ml->Aloc,&m,NULL); if (ierr) return(0);
77: for (i = 0; i<N_requested_rows; i++) {
78: row = requested_rows[i];
79: row_lengths[i] = a->ilen[row];
80: if (allocated_space < k+row_lengths[i]) return(0);
81: if ((row >= 0) || (row <= (m-1))) {
82: aj = a->j + a->i[row];
83: aa = a->a + a->i[row];
84: for (j=0; j<row_lengths[i]; j++) {
85: columns[k] = aj[j];
86: values[k++] = aa[j];
87: }
88: }
89: }
90: return(1);
91: }
95: static PetscErrorCode PetscML_comm(double p[],void *ML_data)
96: {
98: FineGridCtx *ml = (FineGridCtx*)ML_data;
99: Mat A = ml->A;
100: Mat_MPIAIJ *a = (Mat_MPIAIJ*)A->data;
101: PetscMPIInt size;
102: PetscInt i,in_length=A->rmap->n,out_length=ml->Aloc->cmap->n;
103: PetscScalar *array;
106: MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);
107: if (size == 1) return 0;
109: VecPlaceArray(ml->y,p);
110: VecScatterBegin(a->Mvctx,ml->y,a->lvec,INSERT_VALUES,SCATTER_FORWARD);
111: VecScatterEnd(a->Mvctx,ml->y,a->lvec,INSERT_VALUES,SCATTER_FORWARD);
112: VecResetArray(ml->y);
113: VecGetArray(a->lvec,&array);
114: for (i=in_length; i<out_length; i++) p[i] = array[i-in_length];
115: VecRestoreArray(a->lvec,&array);
116: return(0);
117: }
121: static int PetscML_matvec(ML_Operator *ML_data,int in_length,double p[],int out_length,double ap[])
122: {
124: FineGridCtx *ml = (FineGridCtx*)ML_Get_MyMatvecData(ML_data);
125: Mat A = ml->A, Aloc=ml->Aloc;
126: PetscMPIInt size;
127: PetscScalar *pwork=ml->pwork;
128: PetscInt i;
131: MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);
132: if (size == 1) {
133: VecPlaceArray(ml->x,p);
134: } else {
135: for (i=0; i<in_length; i++) pwork[i] = p[i];
136: PetscML_comm(pwork,ml);
137: VecPlaceArray(ml->x,pwork);
138: }
139: VecPlaceArray(ml->y,ap);
140: MatMult(Aloc,ml->x,ml->y);
141: VecResetArray(ml->x);
142: VecResetArray(ml->y);
143: return(0);
144: }
148: static PetscErrorCode MatMult_ML(Mat A,Vec x,Vec y)
149: {
151: Mat_MLShell *shell;
152: PetscScalar *xarray,*yarray;
153: PetscInt x_length,y_length;
156: MatShellGetContext(A,(void**)&shell);
157: VecGetArray(x,&xarray);
158: VecGetArray(y,&yarray);
159: x_length = shell->mlmat->invec_leng;
160: y_length = shell->mlmat->outvec_leng;
161: PetscStackCall("ML_Operator_Apply",ML_Operator_Apply(shell->mlmat,x_length,xarray,y_length,yarray));
162: VecRestoreArray(x,&xarray);
163: VecRestoreArray(y,&yarray);
164: return(0);
165: }
169: /* Computes y = w + A * x
170: It is possible that w == y, but not x == y
171: */
172: static PetscErrorCode MatMultAdd_ML(Mat A,Vec x,Vec w,Vec y)
173: {
174: Mat_MLShell *shell;
175: PetscScalar *xarray,*yarray;
176: PetscInt x_length,y_length;
180: MatShellGetContext(A, (void**) &shell);
181: if (y == w) {
182: if (!shell->work) {
183: VecDuplicate(y, &shell->work);
184: }
185: VecGetArray(x, &xarray);
186: VecGetArray(shell->work, &yarray);
187: x_length = shell->mlmat->invec_leng;
188: y_length = shell->mlmat->outvec_leng;
189: PetscStackCall("ML_Operator_Apply",ML_Operator_Apply(shell->mlmat, x_length, xarray, y_length, yarray));
190: VecRestoreArray(x, &xarray);
191: VecRestoreArray(shell->work, &yarray);
192: VecAXPY(y, 1.0, shell->work);
193: } else {
194: VecGetArray(x, &xarray);
195: VecGetArray(y, &yarray);
196: x_length = shell->mlmat->invec_leng;
197: y_length = shell->mlmat->outvec_leng;
198: PetscStackCall("ML_Operator_Apply",ML_Operator_Apply(shell->mlmat, x_length, xarray, y_length, yarray));
199: VecRestoreArray(x, &xarray);
200: VecRestoreArray(y, &yarray);
201: VecAXPY(y, 1.0, w);
202: }
203: return(0);
204: }
206: /* newtype is ignored since only handles one case */
209: static PetscErrorCode MatConvert_MPIAIJ_ML(Mat A,MatType newtype,MatReuse scall,Mat *Aloc)
210: {
212: Mat_MPIAIJ *mpimat=(Mat_MPIAIJ*)A->data;
213: Mat_SeqAIJ *mat,*a=(Mat_SeqAIJ*)(mpimat->A)->data,*b=(Mat_SeqAIJ*)(mpimat->B)->data;
214: PetscInt *ai=a->i,*aj=a->j,*bi=b->i,*bj=b->j;
215: PetscScalar *aa=a->a,*ba=b->a,*ca;
216: PetscInt am =A->rmap->n,an=A->cmap->n,i,j,k;
217: PetscInt *ci,*cj,ncols;
220: if (am != an) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"A must have a square diagonal portion, am: %d != an: %d",am,an);
222: if (scall == MAT_INITIAL_MATRIX) {
223: PetscMalloc1((1+am),&ci);
224: ci[0] = 0;
225: for (i=0; i<am; i++) ci[i+1] = ci[i] + (ai[i+1] - ai[i]) + (bi[i+1] - bi[i]);
226: PetscMalloc1((1+ci[am]),&cj);
227: PetscMalloc1((1+ci[am]),&ca);
229: k = 0;
230: for (i=0; i<am; i++) {
231: /* diagonal portion of A */
232: ncols = ai[i+1] - ai[i];
233: for (j=0; j<ncols; j++) {
234: cj[k] = *aj++;
235: ca[k++] = *aa++;
236: }
237: /* off-diagonal portion of A */
238: ncols = bi[i+1] - bi[i];
239: for (j=0; j<ncols; j++) {
240: cj[k] = an + (*bj); bj++;
241: ca[k++] = *ba++;
242: }
243: }
244: if (k != ci[am]) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_WRONG,"k: %d != ci[am]: %d",k,ci[am]);
246: /* put together the new matrix */
247: an = mpimat->A->cmap->n+mpimat->B->cmap->n;
248: MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,am,an,ci,cj,ca,Aloc);
250: /* MatCreateSeqAIJWithArrays flags matrix so PETSc doesn't free the user's arrays. */
251: /* Since these are PETSc arrays, change flags to free them as necessary. */
252: mat = (Mat_SeqAIJ*)(*Aloc)->data;
253: mat->free_a = PETSC_TRUE;
254: mat->free_ij = PETSC_TRUE;
256: mat->nonew = 0;
257: } else if (scall == MAT_REUSE_MATRIX) {
258: mat=(Mat_SeqAIJ*)(*Aloc)->data;
259: ci = mat->i; cj = mat->j; ca = mat->a;
260: for (i=0; i<am; i++) {
261: /* diagonal portion of A */
262: ncols = ai[i+1] - ai[i];
263: for (j=0; j<ncols; j++) *ca++ = *aa++;
264: /* off-diagonal portion of A */
265: ncols = bi[i+1] - bi[i];
266: for (j=0; j<ncols; j++) *ca++ = *ba++;
267: }
268: } else SETERRQ1(PetscObjectComm((PetscObject)A),PETSC_ERR_ARG_WRONG,"Invalid MatReuse %d",(int)scall);
269: return(0);
270: }
272: extern PetscErrorCode MatDestroy_Shell(Mat);
275: static PetscErrorCode MatDestroy_ML(Mat A)
276: {
278: Mat_MLShell *shell;
281: MatShellGetContext(A,(void**)&shell);
282: VecDestroy(&shell->y);
283: if (shell->work) {VecDestroy(&shell->work);}
284: PetscFree(shell);
285: MatDestroy_Shell(A);
286: PetscObjectChangeTypeName((PetscObject)A,0);
287: return(0);
288: }
292: static PetscErrorCode MatWrapML_SeqAIJ(ML_Operator *mlmat,MatReuse reuse,Mat *newmat)
293: {
294: struct ML_CSR_MSRdata *matdata = (struct ML_CSR_MSRdata*)mlmat->data;
295: PetscErrorCode ierr;
296: PetscInt m =mlmat->outvec_leng,n=mlmat->invec_leng,*nnz = NULL,nz_max;
297: PetscInt *ml_cols=matdata->columns,*ml_rowptr=matdata->rowptr,*aj,i;
298: PetscScalar *ml_vals=matdata->values,*aa;
301: if (!mlmat->getrow) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_NULL,"mlmat->getrow = NULL");
302: if (m != n) { /* ML Pmat and Rmat are in CSR format. Pass array pointers into SeqAIJ matrix */
303: if (reuse) {
304: Mat_SeqAIJ *aij= (Mat_SeqAIJ*)(*newmat)->data;
305: aij->i = ml_rowptr;
306: aij->j = ml_cols;
307: aij->a = ml_vals;
308: } else {
309: /* sort ml_cols and ml_vals */
310: PetscMalloc1((m+1),&nnz);
311: for (i=0; i<m; i++) nnz[i] = ml_rowptr[i+1] - ml_rowptr[i];
312: aj = ml_cols; aa = ml_vals;
313: for (i=0; i<m; i++) {
314: PetscSortIntWithScalarArray(nnz[i],aj,aa);
315: aj += nnz[i]; aa += nnz[i];
316: }
317: MatCreateSeqAIJWithArrays(PETSC_COMM_SELF,m,n,ml_rowptr,ml_cols,ml_vals,newmat);
318: PetscFree(nnz);
319: }
320: return(0);
321: }
323: nz_max = PetscMax(1,mlmat->max_nz_per_row);
324: PetscMalloc2(nz_max,&aa,nz_max,&aj);
325: if (!reuse) {
326: MatCreate(PETSC_COMM_SELF,newmat);
327: MatSetSizes(*newmat,m,n,PETSC_DECIDE,PETSC_DECIDE);
328: MatSetType(*newmat,MATSEQAIJ);
329: /* keep track of block size for A matrices */
330: MatSetBlockSize (*newmat, mlmat->num_PDEs);
332: PetscMalloc1(m,&nnz);
333: for (i=0; i<m; i++) {
334: PetscStackCall("ML_Operator_Getrow",ML_Operator_Getrow(mlmat,1,&i,nz_max,aj,aa,&nnz[i]));
335: }
336: MatSeqAIJSetPreallocation(*newmat,0,nnz);
337: }
338: for (i=0; i<m; i++) {
339: PetscInt ncols;
341: PetscStackCall("ML_Operator_Getrow",ML_Operator_Getrow(mlmat,1,&i,nz_max,aj,aa,&ncols));
342: MatSetValues(*newmat,1,&i,ncols,aj,aa,INSERT_VALUES);
343: }
344: MatAssemblyBegin(*newmat,MAT_FINAL_ASSEMBLY);
345: MatAssemblyEnd(*newmat,MAT_FINAL_ASSEMBLY);
347: PetscFree2(aa,aj);
348: PetscFree(nnz);
349: return(0);
350: }
354: static PetscErrorCode MatWrapML_SHELL(ML_Operator *mlmat,MatReuse reuse,Mat *newmat)
355: {
357: PetscInt m,n;
358: ML_Comm *MLcomm;
359: Mat_MLShell *shellctx;
362: m = mlmat->outvec_leng;
363: n = mlmat->invec_leng;
365: if (reuse) {
366: MatShellGetContext(*newmat,(void**)&shellctx);
367: shellctx->mlmat = mlmat;
368: return(0);
369: }
371: MLcomm = mlmat->comm;
373: PetscNew(&shellctx);
374: MatCreateShell(MLcomm->USR_comm,m,n,PETSC_DETERMINE,PETSC_DETERMINE,shellctx,newmat);
375: MatShellSetOperation(*newmat,MATOP_MULT,(void(*)(void))MatMult_ML);
376: MatShellSetOperation(*newmat,MATOP_MULT_ADD,(void(*)(void))MatMultAdd_ML);
378: shellctx->A = *newmat;
379: shellctx->mlmat = mlmat;
380: shellctx->work = NULL;
382: VecCreate(MLcomm->USR_comm,&shellctx->y);
383: VecSetSizes(shellctx->y,m,PETSC_DECIDE);
384: VecSetType(shellctx->y,VECSTANDARD);
386: (*newmat)->ops->destroy = MatDestroy_ML;
387: return(0);
388: }
392: static PetscErrorCode MatWrapML_MPIAIJ(ML_Operator *mlmat,MatReuse reuse,Mat *newmat)
393: {
394: PetscInt *aj;
395: PetscScalar *aa;
397: PetscInt i,j,*gordering;
398: PetscInt m=mlmat->outvec_leng,n,nz_max,row;
399: Mat A;
402: if (!mlmat->getrow) SETERRQ(PETSC_COMM_SELF,PETSC_ERR_ARG_NULL,"mlmat->getrow = NULL");
403: n = mlmat->invec_leng;
404: if (m != n) SETERRQ2(PETSC_COMM_SELF,PETSC_ERR_ARG_OUTOFRANGE,"m %d must equal to n %d",m,n);
406: nz_max = PetscMax(1,mlmat->max_nz_per_row);
407: PetscMalloc2(nz_max,&aa,nz_max,&aj);
408: if (reuse) A = *newmat;
409: else {
410: PetscInt *nnzA,*nnzB,*nnz;
411: MatCreate(mlmat->comm->USR_comm,&A);
412: MatSetSizes(A,m,n,PETSC_DECIDE,PETSC_DECIDE);
413: MatSetType(A,MATMPIAIJ);
414: /* keep track of block size for A matrices */
415: MatSetBlockSize (A,mlmat->num_PDEs);
416: PetscMalloc3(m,&nnzA,m,&nnzB,m,&nnz);
418: for (i=0; i<m; i++) {
419: PetscStackCall("ML_Operator_Getrow",ML_Operator_Getrow(mlmat,1,&i,nz_max,aj,aa,&nnz[i]));
420: nnzA[i] = 0;
421: for (j=0; j<nnz[i]; j++) {
422: if (aj[j] < m) nnzA[i]++;
423: }
424: nnzB[i] = nnz[i] - nnzA[i];
425: }
426: MatMPIAIJSetPreallocation(A,0,nnzA,0,nnzB);
427: PetscFree3(nnzA,nnzB,nnz);
428: }
429: /* create global row numbering for a ML_Operator */
430: PetscStackCall("ML_build_global_numbering",ML_build_global_numbering(mlmat,&gordering,"rows"));
431: for (i=0; i<m; i++) {
432: PetscInt ncols;
433: row = gordering[i];
435: PetscStackCall(",ML_Operator_Getrow",ML_Operator_Getrow(mlmat,1,&i,nz_max,aj,aa,&ncols));
436: for (j = 0; j < ncols; j++) aj[j] = gordering[aj[j]];
437: MatSetValues(A,1,&row,ncols,aj,aa,INSERT_VALUES);
438: }
439: PetscStackCall("ML_Operator_Getrow",ML_free(gordering));
440: MatAssemblyBegin(A,MAT_FINAL_ASSEMBLY);
441: MatAssemblyEnd(A,MAT_FINAL_ASSEMBLY);
442: *newmat = A;
444: PetscFree2(aa,aj);
445: return(0);
446: }
448: /* -------------------------------------------------------------------------- */
449: /*
450: PCSetCoordinates_ML
452: Input Parameter:
453: . pc - the preconditioner context
454: */
457: static PetscErrorCode PCSetCoordinates_ML(PC pc, PetscInt ndm, PetscInt a_nloc, PetscReal *coords)
458: {
459: PC_MG *mg = (PC_MG*)pc->data;
460: PC_ML *pc_ml = (PC_ML*)mg->innerctx;
462: PetscInt arrsz,oldarrsz,bs,my0,kk,ii,nloc,Iend;
463: Mat Amat = pc->pmat;
465: /* this function copied and modified from PCSetCoordinates_GEO -TGI */
468: MatGetBlockSize(Amat, &bs);
470: MatGetOwnershipRange(Amat, &my0, &Iend);
471: nloc = (Iend-my0)/bs;
473: if (nloc!=a_nloc) SETERRQ2(PetscObjectComm((PetscObject)Amat),PETSC_ERR_ARG_WRONG, "Number of local blocks must locations = %d %d.",a_nloc,nloc);
474: if ((Iend-my0)%bs!=0) SETERRQ1(PetscObjectComm((PetscObject)Amat),PETSC_ERR_ARG_WRONG, "Bad local size %d.",nloc);
476: oldarrsz = pc_ml->dim * pc_ml->nloc;
477: pc_ml->dim = ndm;
478: pc_ml->nloc = a_nloc;
479: arrsz = ndm * a_nloc;
481: /* create data - syntactic sugar that should be refactored at some point */
482: if (pc_ml->coords==0 || (oldarrsz != arrsz)) {
483: PetscFree(pc_ml->coords);
484: PetscMalloc1((arrsz), &pc_ml->coords);
485: }
486: for (kk=0; kk<arrsz; kk++) pc_ml->coords[kk] = -999.;
487: /* copy data in - column oriented */
488: for (kk = 0; kk < nloc; kk++) {
489: for (ii = 0; ii < ndm; ii++) {
490: pc_ml->coords[ii*nloc + kk] = coords[kk*ndm + ii];
491: }
492: }
493: return(0);
494: }
496: /* -----------------------------------------------------------------------------*/
499: PetscErrorCode PCReset_ML(PC pc)
500: {
502: PC_MG *mg = (PC_MG*)pc->data;
503: PC_ML *pc_ml = (PC_ML*)mg->innerctx;
504: PetscInt level,fine_level=pc_ml->Nlevels-1,dim=pc_ml->dim;
507: if (dim) {
508: ML_Aggregate_Viz_Stats * grid_info = (ML_Aggregate_Viz_Stats*) pc_ml->ml_object->Grid[0].Grid;
510: for (level=0; level<=fine_level; level++) {
511: VecDestroy(&pc_ml->gridctx[level].coords);
512: }
514: grid_info->x = 0; /* do this so ML doesn't try to free coordinates */
515: grid_info->y = 0;
516: grid_info->z = 0;
518: PetscStackCall("ML_Operator_Getrow",ML_Aggregate_VizAndStats_Clean(pc_ml->ml_object));
519: }
520: PetscStackCall("ML_Aggregate_Destroy",ML_Aggregate_Destroy(&pc_ml->agg_object));
521: PetscStackCall("ML_Aggregate_Destroy",ML_Destroy(&pc_ml->ml_object));
523: if (pc_ml->PetscMLdata) {
524: PetscFree(pc_ml->PetscMLdata->pwork);
525: MatDestroy(&pc_ml->PetscMLdata->Aloc);
526: VecDestroy(&pc_ml->PetscMLdata->x);
527: VecDestroy(&pc_ml->PetscMLdata->y);
528: }
529: PetscFree(pc_ml->PetscMLdata);
531: if (pc_ml->gridctx) {
532: for (level=0; level<fine_level; level++) {
533: if (pc_ml->gridctx[level].A) {MatDestroy(&pc_ml->gridctx[level].A);}
534: if (pc_ml->gridctx[level].P) {MatDestroy(&pc_ml->gridctx[level].P);}
535: if (pc_ml->gridctx[level].R) {MatDestroy(&pc_ml->gridctx[level].R);}
536: if (pc_ml->gridctx[level].x) {VecDestroy(&pc_ml->gridctx[level].x);}
537: if (pc_ml->gridctx[level].b) {VecDestroy(&pc_ml->gridctx[level].b);}
538: if (pc_ml->gridctx[level+1].r) {VecDestroy(&pc_ml->gridctx[level+1].r);}
539: }
540: }
541: PetscFree(pc_ml->gridctx);
542: PetscFree(pc_ml->coords);
544: pc_ml->dim = 0;
545: pc_ml->nloc = 0;
546: return(0);
547: }
548: /* -------------------------------------------------------------------------- */
549: /*
550: PCSetUp_ML - Prepares for the use of the ML preconditioner
551: by setting data structures and options.
553: Input Parameter:
554: . pc - the preconditioner context
556: Application Interface Routine: PCSetUp()
558: Notes:
559: The interface routine PCSetUp() is not usually called directly by
560: the user, but instead is called by PCApply() if necessary.
561: */
562: extern PetscErrorCode PCSetFromOptions_MG(PC);
563: extern PetscErrorCode PCReset_MG(PC);
567: PetscErrorCode PCSetUp_ML(PC pc)
568: {
569: PetscErrorCode ierr;
570: PetscMPIInt size;
571: FineGridCtx *PetscMLdata;
572: ML *ml_object;
573: ML_Aggregate *agg_object;
574: ML_Operator *mlmat;
575: PetscInt nlocal_allcols,Nlevels,mllevel,level,level1,m,fine_level,bs;
576: Mat A,Aloc;
577: GridCtx *gridctx;
578: PC_MG *mg = (PC_MG*)pc->data;
579: PC_ML *pc_ml = (PC_ML*)mg->innerctx;
580: PetscBool isSeq, isMPI;
581: KSP smoother;
582: PC subpc;
583: PetscInt mesh_level, old_mesh_level;
584: MatInfo info;
585: static PetscBool cite = PETSC_FALSE;
588: PetscCitationsRegister("@TechReport{ml_users_guide,\n author = {M. Sala and J.J. Hu and R.S. Tuminaro},\n title = {{ML}3.1 {S}moothed {A}ggregation {U}ser's {G}uide},\n institution = {Sandia National Laboratories},\n number = {SAND2004-4821},\n year = 2004\n}\n",&cite);
589: A = pc->pmat;
590: MPI_Comm_size(PetscObjectComm((PetscObject)A),&size);
592: if (pc->setupcalled) {
593: if (pc->flag == SAME_NONZERO_PATTERN && pc_ml->reuse_interpolation) {
594: /*
595: Reuse interpolaton instead of recomputing aggregates and updating the whole hierarchy. This is less expensive for
596: multiple solves in which the matrix is not changing too quickly.
597: */
598: ml_object = pc_ml->ml_object;
599: gridctx = pc_ml->gridctx;
600: Nlevels = pc_ml->Nlevels;
601: fine_level = Nlevels - 1;
602: gridctx[fine_level].A = A;
604: PetscObjectTypeCompare((PetscObject) A, MATSEQAIJ, &isSeq);
605: PetscObjectTypeCompare((PetscObject) A, MATMPIAIJ, &isMPI);
606: if (isMPI) {
607: MatConvert_MPIAIJ_ML(A,NULL,MAT_INITIAL_MATRIX,&Aloc);
608: } else if (isSeq) {
609: Aloc = A;
610: PetscObjectReference((PetscObject)Aloc);
611: } else SETERRQ1(PetscObjectComm((PetscObject)pc),PETSC_ERR_ARG_WRONG, "Matrix type '%s' cannot be used with ML. ML can only handle AIJ matrices.",((PetscObject)A)->type_name);
613: MatGetSize(Aloc,&m,&nlocal_allcols);
614: PetscMLdata = pc_ml->PetscMLdata;
615: MatDestroy(&PetscMLdata->Aloc);
616: PetscMLdata->A = A;
617: PetscMLdata->Aloc = Aloc;
618: PetscStackCall("ML_Aggregate_Destroy",ML_Init_Amatrix(ml_object,0,m,m,PetscMLdata));
619: PetscStackCall("ML_Set_Amatrix_Matvec",ML_Set_Amatrix_Matvec(ml_object,0,PetscML_matvec));
621: mesh_level = ml_object->ML_finest_level;
622: while (ml_object->SingleLevel[mesh_level].Rmat->to) {
623: old_mesh_level = mesh_level;
624: mesh_level = ml_object->SingleLevel[mesh_level].Rmat->to->levelnum;
626: /* clean and regenerate A */
627: mlmat = &(ml_object->Amat[mesh_level]);
628: PetscStackCall("ML_Operator_Clean",ML_Operator_Clean(mlmat));
629: PetscStackCall("ML_Operator_Init",ML_Operator_Init(mlmat,ml_object->comm));
630: PetscStackCall("ML_Gen_AmatrixRAP",ML_Gen_AmatrixRAP(ml_object, old_mesh_level, mesh_level));
631: }
633: level = fine_level - 1;
634: if (size == 1) { /* convert ML P, R and A into seqaij format */
635: for (mllevel=1; mllevel<Nlevels; mllevel++) {
636: mlmat = &(ml_object->Amat[mllevel]);
637: MatWrapML_SeqAIJ(mlmat,MAT_REUSE_MATRIX,&gridctx[level].A);
638: level--;
639: }
640: } else { /* convert ML P and R into shell format, ML A into mpiaij format */
641: for (mllevel=1; mllevel<Nlevels; mllevel++) {
642: mlmat = &(ml_object->Amat[mllevel]);
643: MatWrapML_MPIAIJ(mlmat,MAT_REUSE_MATRIX,&gridctx[level].A);
644: level--;
645: }
646: }
648: for (level=0; level<fine_level; level++) {
649: if (level > 0) {
650: PCMGSetResidual(pc,level,PCMGResidualDefault,gridctx[level].A);
651: }
652: KSPSetOperators(gridctx[level].ksp,gridctx[level].A,gridctx[level].A,SAME_NONZERO_PATTERN);
653: }
654: PCMGSetResidual(pc,fine_level,PCMGResidualDefault,gridctx[fine_level].A);
655: KSPSetOperators(gridctx[fine_level].ksp,gridctx[level].A,gridctx[fine_level].A,SAME_NONZERO_PATTERN);
657: PCSetUp_MG(pc);
658: return(0);
659: } else {
660: /* since ML can change the size of vectors/matrices at any level we must destroy everything */
661: PCReset_ML(pc);
662: PCReset_MG(pc);
663: }
664: }
666: /* setup special features of PCML */
667: /*--------------------------------*/
668: /* covert A to Aloc to be used by ML at fine grid */
669: pc_ml->size = size;
670: PetscObjectTypeCompare((PetscObject) A, MATSEQAIJ, &isSeq);
671: PetscObjectTypeCompare((PetscObject) A, MATMPIAIJ, &isMPI);
672: if (isMPI) {
673: MatConvert_MPIAIJ_ML(A,NULL,MAT_INITIAL_MATRIX,&Aloc);
674: } else if (isSeq) {
675: Aloc = A;
676: PetscObjectReference((PetscObject)Aloc);
677: } else SETERRQ1(PetscObjectComm((PetscObject)pc),PETSC_ERR_ARG_WRONG, "Matrix type '%s' cannot be used with ML. ML can only handle AIJ matrices.",((PetscObject)A)->type_name);
679: /* create and initialize struct 'PetscMLdata' */
680: PetscNewLog(pc,&PetscMLdata);
681: pc_ml->PetscMLdata = PetscMLdata;
682: PetscMalloc1((Aloc->cmap->n+1),&PetscMLdata->pwork);
684: VecCreate(PETSC_COMM_SELF,&PetscMLdata->x);
685: VecSetSizes(PetscMLdata->x,Aloc->cmap->n,Aloc->cmap->n);
686: VecSetType(PetscMLdata->x,VECSEQ);
688: VecCreate(PETSC_COMM_SELF,&PetscMLdata->y);
689: VecSetSizes(PetscMLdata->y,A->rmap->n,PETSC_DECIDE);
690: VecSetType(PetscMLdata->y,VECSEQ);
691: PetscMLdata->A = A;
692: PetscMLdata->Aloc = Aloc;
693: if (pc_ml->dim) { /* create vecs around the coordinate data given */
694: PetscInt i,j,dim=pc_ml->dim;
695: PetscInt nloc = pc_ml->nloc,nlocghost;
696: PetscReal *ghostedcoords;
698: MatGetBlockSize(A,&bs);
699: nlocghost = Aloc->cmap->n / bs;
700: PetscMalloc1(dim*nlocghost,&ghostedcoords);
701: for (i = 0; i < dim; i++) {
702: /* copy coordinate values into first component of pwork */
703: for (j = 0; j < nloc; j++) {
704: PetscMLdata->pwork[bs * j] = pc_ml->coords[nloc * i + j];
705: }
706: /* get the ghost values */
707: PetscML_comm(PetscMLdata->pwork,PetscMLdata);
708: /* write into the vector */
709: for (j = 0; j < nlocghost; j++) {
710: ghostedcoords[i * nlocghost + j] = PetscMLdata->pwork[bs * j];
711: }
712: }
713: /* replace the original coords with the ghosted coords, because these are
714: * what ML needs */
715: PetscFree(pc_ml->coords);
716: pc_ml->coords = ghostedcoords;
717: }
719: /* create ML discretization matrix at fine grid */
720: /* ML requires input of fine-grid matrix. It determines nlevels. */
721: MatGetSize(Aloc,&m,&nlocal_allcols);
722: MatGetBlockSize(A,&bs);
723: PetscStackCall("ML_Create",ML_Create(&ml_object,pc_ml->MaxNlevels));
724: PetscStackCall("ML_Comm_Set_UsrComm",ML_Comm_Set_UsrComm(ml_object->comm,PetscObjectComm((PetscObject)A)));
725: pc_ml->ml_object = ml_object;
726: PetscStackCall("ML_Init_Amatrix",ML_Init_Amatrix(ml_object,0,m,m,PetscMLdata));
727: PetscStackCall("ML_Set_Amatrix_Getrow",ML_Set_Amatrix_Getrow(ml_object,0,PetscML_getrow,PetscML_comm,nlocal_allcols));
728: PetscStackCall("ML_Set_Amatrix_Matvec",ML_Set_Amatrix_Matvec(ml_object,0,PetscML_matvec));
730: PetscStackCall("ML_Set_Symmetrize",ML_Set_Symmetrize(ml_object,pc_ml->Symmetrize ? ML_YES : ML_NO));
732: /* aggregation */
733: PetscStackCall("ML_Aggregate_Create",ML_Aggregate_Create(&agg_object));
734: pc_ml->agg_object = agg_object;
736: {
737: MatNullSpace mnull;
738: MatGetNearNullSpace(A,&mnull);
739: if (pc_ml->nulltype == PCML_NULLSPACE_AUTO) {
740: if (mnull) pc_ml->nulltype = PCML_NULLSPACE_USER;
741: else if (bs > 1) pc_ml->nulltype = PCML_NULLSPACE_BLOCK;
742: else pc_ml->nulltype = PCML_NULLSPACE_SCALAR;
743: }
744: switch (pc_ml->nulltype) {
745: case PCML_NULLSPACE_USER: {
746: PetscScalar *nullvec;
747: const PetscScalar *v;
748: PetscBool has_const;
749: PetscInt i,j,mlocal,nvec,M;
750: const Vec *vecs;
752: if (!mnull) SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_USER,"Must provide explicit null space using MatSetNearNullSpace() to use user-specified null space");
753: MatGetSize(A,&M,NULL);
754: MatGetLocalSize(Aloc,&mlocal,NULL);
755: MatNullSpaceGetVecs(mnull,&has_const,&nvec,&vecs);
756: PetscMalloc1((nvec+!!has_const)*mlocal,&nullvec);
757: if (has_const) for (i=0; i<mlocal; i++) nullvec[i] = 1.0/M;
758: for (i=0; i<nvec; i++) {
759: VecGetArrayRead(vecs[i],&v);
760: for (j=0; j<mlocal; j++) nullvec[(i+!!has_const)*mlocal + j] = v[j];
761: VecRestoreArrayRead(vecs[i],&v);
762: }
763: PetscStackCall("ML_Aggregate_Create",ML_Aggregate_Set_NullSpace(agg_object,bs,nvec+!!has_const,nullvec,mlocal);CHKERRQ(ierr));
764: PetscFree(nullvec);
765: } break;
766: case PCML_NULLSPACE_BLOCK:
767: PetscStackCall("ML_Aggregate_Set_NullSpace",ML_Aggregate_Set_NullSpace(agg_object,bs,bs,0,0);CHKERRQ(ierr));
768: break;
769: case PCML_NULLSPACE_SCALAR:
770: break;
771: default: SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_SUP,"Unknown null space type");
772: }
773: }
774: PetscStackCall("ML_Aggregate_Set_MaxCoarseSize",ML_Aggregate_Set_MaxCoarseSize(agg_object,pc_ml->MaxCoarseSize));
775: /* set options */
776: switch (pc_ml->CoarsenScheme) {
777: case 1:
778: PetscStackCall("ML_Aggregate_Set_CoarsenScheme_Coupled",ML_Aggregate_Set_CoarsenScheme_Coupled(agg_object));break;
779: case 2:
780: PetscStackCall("ML_Aggregate_Set_CoarsenScheme_MIS",ML_Aggregate_Set_CoarsenScheme_MIS(agg_object));break;
781: case 3:
782: PetscStackCall("ML_Aggregate_Set_CoarsenScheme_METIS",ML_Aggregate_Set_CoarsenScheme_METIS(agg_object));break;
783: }
784: PetscStackCall("ML_Aggregate_Set_Threshold",ML_Aggregate_Set_Threshold(agg_object,pc_ml->Threshold));
785: PetscStackCall("ML_Aggregate_Set_DampingFactor",ML_Aggregate_Set_DampingFactor(agg_object,pc_ml->DampingFactor));
786: if (pc_ml->SpectralNormScheme_Anorm) {
787: PetscStackCall("ML_Set_SpectralNormScheme_Anorm",ML_Set_SpectralNormScheme_Anorm(ml_object));
788: }
789: agg_object->keep_agg_information = (int)pc_ml->KeepAggInfo;
790: agg_object->keep_P_tentative = (int)pc_ml->Reusable;
791: agg_object->block_scaled_SA = (int)pc_ml->BlockScaling;
792: agg_object->minimizing_energy = (int)pc_ml->EnergyMinimization;
793: agg_object->minimizing_energy_droptol = (double)pc_ml->EnergyMinimizationDropTol;
794: agg_object->cheap_minimizing_energy = (int)pc_ml->EnergyMinimizationCheap;
796: if (pc_ml->Aux) {
797: if (!pc_ml->dim) SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_USER,"Auxiliary matrix requires coordinates");
798: ml_object->Amat[0].aux_data->threshold = pc_ml->AuxThreshold;
799: ml_object->Amat[0].aux_data->enable = 1;
800: ml_object->Amat[0].aux_data->max_level = 10;
801: ml_object->Amat[0].num_PDEs = bs;
802: }
804: MatGetInfo(A,MAT_LOCAL,&info);
805: ml_object->Amat[0].N_nonzeros = (int) info.nz_used;
807: if (pc_ml->dim) {
808: PetscInt i,dim = pc_ml->dim;
809: ML_Aggregate_Viz_Stats *grid_info;
810: PetscInt nlocghost;
812: MatGetBlockSize(A,&bs);
813: nlocghost = Aloc->cmap->n / bs;
815: PetscStackCall("ML_Aggregate_VizAndStats_Setup(",ML_Aggregate_VizAndStats_Setup(ml_object)); /* create ml info for coords */
816: grid_info = (ML_Aggregate_Viz_Stats*) ml_object->Grid[0].Grid;
817: for (i = 0; i < dim; i++) {
818: /* set the finest level coordinates to point to the column-order array
819: * in pc_ml */
820: /* NOTE: must point away before VizAndStats_Clean so ML doesn't free */
821: switch (i) {
822: case 0: grid_info->x = pc_ml->coords + nlocghost * i; break;
823: case 1: grid_info->y = pc_ml->coords + nlocghost * i; break;
824: case 2: grid_info->z = pc_ml->coords + nlocghost * i; break;
825: default: SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_ARG_SIZ,"PCML coordinate dimension must be <= 3");
826: }
827: }
828: grid_info->Ndim = dim;
829: }
831: /* repartitioning */
832: if (pc_ml->Repartition) {
833: PetscStackCall("ML_Repartition_Activate",ML_Repartition_Activate(ml_object));
834: PetscStackCall("ML_Repartition_Set_LargestMinMaxRatio",ML_Repartition_Set_LargestMinMaxRatio(ml_object,pc_ml->MaxMinRatio));
835: PetscStackCall("ML_Repartition_Set_MinPerProc",ML_Repartition_Set_MinPerProc(ml_object,pc_ml->MinPerProc));
836: PetscStackCall("ML_Repartition_Set_PutOnSingleProc",ML_Repartition_Set_PutOnSingleProc(ml_object,pc_ml->PutOnSingleProc));
837: #if 0 /* Function not yet defined in ml-6.2 */
838: /* I'm not sure what compatibility issues might crop up if we partitioned
839: * on the finest level, so to be safe repartition starts on the next
840: * finest level (reflection default behavior in
841: * ml_MultiLevelPreconditioner) */
842: PetscStackCall("ML_Repartition_Set_StartLevel",ML_Repartition_Set_StartLevel(ml_object,1));
843: #endif
845: if (!pc_ml->RepartitionType) {
846: PetscInt i;
848: if (!pc_ml->dim) SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_USER,"ML Zoltan repartitioning requires coordinates");
849: PetscStackCall("ML_Repartition_Set_Partitioner",ML_Repartition_Set_Partitioner(ml_object,ML_USEZOLTAN));
850: PetscStackCall("ML_Aggregate_Set_Dimensions",ML_Aggregate_Set_Dimensions(agg_object, pc_ml->dim));
852: for (i = 0; i < ml_object->ML_num_levels; i++) {
853: ML_Aggregate_Viz_Stats *grid_info = (ML_Aggregate_Viz_Stats*)ml_object->Grid[i].Grid;
854: grid_info->zoltan_type = pc_ml->ZoltanScheme + 1; /* ml numbers options 1, 2, 3 */
855: /* defaults from ml_agg_info.c */
856: grid_info->zoltan_estimated_its = 40; /* only relevant to hypergraph / fast hypergraph */
857: grid_info->zoltan_timers = 0;
858: grid_info->smoothing_steps = 4; /* only relevant to hypergraph / fast hypergraph */
859: }
860: } else {
861: PetscStackCall("ML_Repartition_Set_Partitioner",ML_Repartition_Set_Partitioner(ml_object,ML_USEPARMETIS));
862: }
863: }
865: if (pc_ml->OldHierarchy) {
866: PetscStackCall("ML_Gen_MGHierarchy_UsingAggregation",Nlevels = ML_Gen_MGHierarchy_UsingAggregation(ml_object,0,ML_INCREASING,agg_object));
867: } else {
868: PetscStackCall("ML_Gen_MultiLevelHierarchy_UsingAggregation",Nlevels = ML_Gen_MultiLevelHierarchy_UsingAggregation(ml_object,0,ML_INCREASING,agg_object));
869: }
870: if (Nlevels<=0) SETERRQ1(PetscObjectComm((PetscObject)pc),PETSC_ERR_ARG_OUTOFRANGE,"Nlevels %d must > 0",Nlevels);
871: pc_ml->Nlevels = Nlevels;
872: fine_level = Nlevels - 1;
874: PCMGSetLevels(pc,Nlevels,NULL);
875: /* set default smoothers */
876: for (level=1; level<=fine_level; level++) {
877: PCMGGetSmoother(pc,level,&smoother);
878: KSPSetType(smoother,KSPRICHARDSON);
879: KSPGetPC(smoother,&subpc);
880: PCSetType(subpc,PCSOR);
881: }
882: PetscObjectOptionsBegin((PetscObject)pc);
883: PCSetFromOptions_MG(pc); /* should be called in PCSetFromOptions_ML(), but cannot be called prior to PCMGSetLevels() */
884: PetscOptionsEnd();
886: PetscMalloc1(Nlevels,&gridctx);
888: pc_ml->gridctx = gridctx;
890: /* wrap ML matrices by PETSc shell matrices at coarsened grids.
891: Level 0 is the finest grid for ML, but coarsest for PETSc! */
892: gridctx[fine_level].A = A;
894: level = fine_level - 1;
895: if (size == 1) { /* convert ML P, R and A into seqaij format */
896: for (mllevel=1; mllevel<Nlevels; mllevel++) {
897: mlmat = &(ml_object->Pmat[mllevel]);
898: MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].P);
899: mlmat = &(ml_object->Rmat[mllevel-1]);
900: MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].R);
902: mlmat = &(ml_object->Amat[mllevel]);
903: MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].A);
904: level--;
905: }
906: } else { /* convert ML P and R into shell format, ML A into mpiaij format */
907: for (mllevel=1; mllevel<Nlevels; mllevel++) {
908: mlmat = &(ml_object->Pmat[mllevel]);
909: MatWrapML_SHELL(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].P);
910: mlmat = &(ml_object->Rmat[mllevel-1]);
911: MatWrapML_SHELL(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].R);
913: mlmat = &(ml_object->Amat[mllevel]);
914: MatWrapML_MPIAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].A);
915: level--;
916: }
917: }
919: /* create vectors and ksp at all levels */
920: for (level=0; level<fine_level; level++) {
921: level1 = level + 1;
922: VecCreate(((PetscObject)gridctx[level].A)->comm,&gridctx[level].x);
923: VecSetSizes(gridctx[level].x,gridctx[level].A->cmap->n,PETSC_DECIDE);
924: VecSetType(gridctx[level].x,VECMPI);
925: PCMGSetX(pc,level,gridctx[level].x);
927: VecCreate(((PetscObject)gridctx[level].A)->comm,&gridctx[level].b);
928: VecSetSizes(gridctx[level].b,gridctx[level].A->rmap->n,PETSC_DECIDE);
929: VecSetType(gridctx[level].b,VECMPI);
930: PCMGSetRhs(pc,level,gridctx[level].b);
932: VecCreate(((PetscObject)gridctx[level1].A)->comm,&gridctx[level1].r);
933: VecSetSizes(gridctx[level1].r,gridctx[level1].A->rmap->n,PETSC_DECIDE);
934: VecSetType(gridctx[level1].r,VECMPI);
935: PCMGSetR(pc,level1,gridctx[level1].r);
937: if (level == 0) {
938: PCMGGetCoarseSolve(pc,&gridctx[level].ksp);
939: } else {
940: PCMGGetSmoother(pc,level,&gridctx[level].ksp);
941: }
942: }
943: PCMGGetSmoother(pc,fine_level,&gridctx[fine_level].ksp);
945: /* create coarse level and the interpolation between the levels */
946: for (level=0; level<fine_level; level++) {
947: level1 = level + 1;
948: PCMGSetInterpolation(pc,level1,gridctx[level].P);
949: PCMGSetRestriction(pc,level1,gridctx[level].R);
950: if (level > 0) {
951: PCMGSetResidual(pc,level,PCMGResidualDefault,gridctx[level].A);
952: }
953: KSPSetOperators(gridctx[level].ksp,gridctx[level].A,gridctx[level].A,DIFFERENT_NONZERO_PATTERN);
954: }
955: PCMGSetResidual(pc,fine_level,PCMGResidualDefault,gridctx[fine_level].A);
956: KSPSetOperators(gridctx[fine_level].ksp,gridctx[level].A,gridctx[fine_level].A,DIFFERENT_NONZERO_PATTERN);
958: /* put coordinate info in levels */
959: if (pc_ml->dim) {
960: PetscInt i,j,dim = pc_ml->dim;
961: PetscInt bs, nloc;
962: PC subpc;
963: PetscReal *array;
965: level = fine_level;
966: for (mllevel = 0; mllevel < Nlevels; mllevel++) {
967: ML_Aggregate_Viz_Stats *grid_info = (ML_Aggregate_Viz_Stats*)ml_object->Amat[mllevel].to->Grid->Grid;
968: MPI_Comm comm = ((PetscObject)gridctx[level].A)->comm;
970: MatGetBlockSize (gridctx[level].A, &bs);
971: MatGetLocalSize (gridctx[level].A, NULL, &nloc);
972: nloc /= bs; /* number of local nodes */
974: VecCreate(comm,&gridctx[level].coords);
975: VecSetSizes(gridctx[level].coords,dim * nloc,PETSC_DECIDE);
976: VecSetType(gridctx[level].coords,VECMPI);
977: VecGetArray(gridctx[level].coords,&array);
978: for (j = 0; j < nloc; j++) {
979: for (i = 0; i < dim; i++) {
980: switch (i) {
981: case 0: array[dim * j + i] = grid_info->x[j]; break;
982: case 1: array[dim * j + i] = grid_info->y[j]; break;
983: case 2: array[dim * j + i] = grid_info->z[j]; break;
984: default: SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_ARG_SIZ,"PCML coordinate dimension must be <= 3");
985: }
986: }
987: }
989: /* passing coordinates to smoothers/coarse solver, should they need them */
990: KSPGetPC(gridctx[level].ksp,&subpc);
991: PCSetCoordinates(subpc,dim,nloc,array);
992: VecRestoreArray(gridctx[level].coords,&array);
993: level--;
994: }
995: }
997: /* setupcalled is set to 0 so that MG is setup from scratch */
998: pc->setupcalled = 0;
999: PCSetUp_MG(pc);
1000: return(0);
1001: }
1003: /* -------------------------------------------------------------------------- */
1004: /*
1005: PCDestroy_ML - Destroys the private context for the ML preconditioner
1006: that was created with PCCreate_ML().
1008: Input Parameter:
1009: . pc - the preconditioner context
1011: Application Interface Routine: PCDestroy()
1012: */
1015: PetscErrorCode PCDestroy_ML(PC pc)
1016: {
1018: PC_MG *mg = (PC_MG*)pc->data;
1019: PC_ML *pc_ml= (PC_ML*)mg->innerctx;
1022: PCReset_ML(pc);
1023: PetscFree(pc_ml);
1024: PCDestroy_MG(pc);
1025: PetscObjectComposeFunction((PetscObject)pc,"PCSetCoordinates_C",NULL);
1026: return(0);
1027: }
1031: PetscErrorCode PCSetFromOptions_ML(PC pc)
1032: {
1034: PetscInt indx,PrintLevel,partindx;
1035: const char *scheme[] = {"Uncoupled","Coupled","MIS","METIS"};
1036: const char *part[] = {"Zoltan","ParMETIS"};
1037: #if defined(HAVE_ML_ZOLTAN)
1038: PetscInt zidx;
1039: const char *zscheme[] = {"RCB","hypergraph","fast_hypergraph"};
1040: #endif
1041: PC_MG *mg = (PC_MG*)pc->data;
1042: PC_ML *pc_ml = (PC_ML*)mg->innerctx;
1043: PetscMPIInt size;
1044: MPI_Comm comm;
1047: PetscObjectGetComm((PetscObject)pc,&comm);
1048: MPI_Comm_size(comm,&size);
1049: PetscOptionsHead("ML options");
1051: PrintLevel = 0;
1052: indx = 0;
1053: partindx = 0;
1055: PetscOptionsInt("-pc_ml_PrintLevel","Print level","ML_Set_PrintLevel",PrintLevel,&PrintLevel,NULL);
1056: PetscStackCall("ML_Set_PrintLeve",ML_Set_PrintLevel(PrintLevel));
1057: PetscOptionsInt("-pc_ml_maxNlevels","Maximum number of levels","None",pc_ml->MaxNlevels,&pc_ml->MaxNlevels,NULL);
1058: PetscOptionsInt("-pc_ml_maxCoarseSize","Maximum coarsest mesh size","ML_Aggregate_Set_MaxCoarseSize",pc_ml->MaxCoarseSize,&pc_ml->MaxCoarseSize,NULL);
1059: PetscOptionsEList("-pc_ml_CoarsenScheme","Aggregate Coarsen Scheme","ML_Aggregate_Set_CoarsenScheme_*",scheme,4,scheme[0],&indx,NULL);
1061: pc_ml->CoarsenScheme = indx;
1063: PetscOptionsReal("-pc_ml_DampingFactor","P damping factor","ML_Aggregate_Set_DampingFactor",pc_ml->DampingFactor,&pc_ml->DampingFactor,NULL);
1064: PetscOptionsReal("-pc_ml_Threshold","Smoother drop tol","ML_Aggregate_Set_Threshold",pc_ml->Threshold,&pc_ml->Threshold,NULL);
1065: PetscOptionsBool("-pc_ml_SpectralNormScheme_Anorm","Method used for estimating spectral radius","ML_Set_SpectralNormScheme_Anorm",pc_ml->SpectralNormScheme_Anorm,&pc_ml->SpectralNormScheme_Anorm,NULL);
1066: PetscOptionsBool("-pc_ml_Symmetrize","Symmetrize aggregation","ML_Set_Symmetrize",pc_ml->Symmetrize,&pc_ml->Symmetrize,NULL);
1067: PetscOptionsBool("-pc_ml_BlockScaling","Scale all dofs at each node together","None",pc_ml->BlockScaling,&pc_ml->BlockScaling,NULL);
1068: PetscOptionsEnum("-pc_ml_nullspace","Which type of null space information to use","None",PCMLNullSpaceTypes,(PetscEnum)pc_ml->nulltype,(PetscEnum*)&pc_ml->nulltype,NULL);
1069: PetscOptionsInt("-pc_ml_EnergyMinimization","Energy minimization norm type (0=no minimization; see ML manual for 1,2,3; -1 and 4 undocumented)","None",pc_ml->EnergyMinimization,&pc_ml->EnergyMinimization,NULL);
1070: PetscOptionsBool("-pc_ml_reuse_interpolation","Reuse the interpolation operators when possible (cheaper, weaker when matrix entries change a lot)","None",pc_ml->reuse_interpolation,&pc_ml->reuse_interpolation,NULL);
1071: /*
1072: The following checks a number of conditions. If we let this stuff slip by, then ML's error handling will take over.
1073: This is suboptimal because it amounts to calling exit(1) so we check for the most common conditions.
1075: We also try to set some sane defaults when energy minimization is activated, otherwise it's hard to find a working
1076: combination of options and ML's exit(1) explanations don't help matters.
1077: */
1078: if (pc_ml->EnergyMinimization < -1 || pc_ml->EnergyMinimization > 4) SETERRQ(comm,PETSC_ERR_ARG_OUTOFRANGE,"EnergyMinimization must be in range -1..4");
1079: if (pc_ml->EnergyMinimization == 4 && size > 1) SETERRQ(comm,PETSC_ERR_SUP,"Energy minimization type 4 does not work in parallel");
1080: if (pc_ml->EnergyMinimization == 4) {PetscInfo(pc,"Mandel's energy minimization scheme is experimental and broken in ML-6.2");}
1081: if (pc_ml->EnergyMinimization) {
1082: PetscOptionsReal("-pc_ml_EnergyMinimizationDropTol","Energy minimization drop tolerance","None",pc_ml->EnergyMinimizationDropTol,&pc_ml->EnergyMinimizationDropTol,NULL);
1083: }
1084: if (pc_ml->EnergyMinimization == 2) {
1085: /* According to ml_MultiLevelPreconditioner.cpp, this option is only meaningful for norm type (2) */
1086: PetscOptionsBool("-pc_ml_EnergyMinimizationCheap","Use cheaper variant of norm type 2","None",pc_ml->EnergyMinimizationCheap,&pc_ml->EnergyMinimizationCheap,NULL);
1087: }
1088: /* energy minimization sometimes breaks if this is turned off, the more classical stuff should be okay without it */
1089: if (pc_ml->EnergyMinimization) pc_ml->KeepAggInfo = PETSC_TRUE;
1090: PetscOptionsBool("-pc_ml_KeepAggInfo","Allows the preconditioner to be reused, or auxilliary matrices to be generated","None",pc_ml->KeepAggInfo,&pc_ml->KeepAggInfo,NULL);
1091: /* Option (-1) doesn't work at all (calls exit(1)) if the tentative restriction operator isn't stored. */
1092: if (pc_ml->EnergyMinimization == -1) pc_ml->Reusable = PETSC_TRUE;
1093: PetscOptionsBool("-pc_ml_Reusable","Store intermedaiate data structures so that the multilevel hierarchy is reusable","None",pc_ml->Reusable,&pc_ml->Reusable,NULL);
1094: /*
1095: ML's C API is severely underdocumented and lacks significant functionality. The C++ API calls
1096: ML_Gen_MultiLevelHierarchy_UsingAggregation() which is a modified copy (!?) of the documented function
1097: ML_Gen_MGHierarchy_UsingAggregation(). This modification, however, does not provide a strict superset of the
1098: functionality in the old function, so some users may still want to use it. Note that many options are ignored in
1099: this context, but ML doesn't provide a way to find out which ones.
1100: */
1101: PetscOptionsBool("-pc_ml_OldHierarchy","Use old routine to generate hierarchy","None",pc_ml->OldHierarchy,&pc_ml->OldHierarchy,NULL);
1102: PetscOptionsBool("-pc_ml_repartition", "Allow ML to repartition levels of the heirarchy","ML_Repartition_Activate",pc_ml->Repartition,&pc_ml->Repartition,NULL);
1103: if (pc_ml->Repartition) {
1104: PetscOptionsReal("-pc_ml_repartitionMaxMinRatio", "Acceptable ratio of repartitioned sizes","ML_Repartition_Set_LargestMinMaxRatio",pc_ml->MaxMinRatio,&pc_ml->MaxMinRatio,NULL);
1105: PetscOptionsInt("-pc_ml_repartitionMinPerProc", "Smallest repartitioned size","ML_Repartition_Set_MinPerProc",pc_ml->MinPerProc,&pc_ml->MinPerProc,NULL);
1106: PetscOptionsInt("-pc_ml_repartitionPutOnSingleProc", "Problem size automatically repartitioned to one processor","ML_Repartition_Set_PutOnSingleProc",pc_ml->PutOnSingleProc,&pc_ml->PutOnSingleProc,NULL);
1107: #if defined(HAVE_ML_ZOLTAN)
1108: partindx = 0;
1109: PetscOptionsEList("-pc_ml_repartitionType", "Repartitioning library to use","ML_Repartition_Set_Partitioner",part,2,part[0],&partindx,NULL);
1111: pc_ml->RepartitionType = partindx;
1112: if (!partindx) {
1113: PetscInt zindx = 0;
1115: PetscOptionsEList("-pc_ml_repartitionZoltanScheme", "Repartitioning scheme to use","None",zscheme,3,zscheme[0],&zindx,NULL);
1117: pc_ml->ZoltanScheme = zindx;
1118: }
1119: #else
1120: partindx = 1;
1121: PetscOptionsEList("-pc_ml_repartitionType", "Repartitioning library to use","ML_Repartition_Set_Partitioner",part,2,part[1],&partindx,NULL);
1122: if (!partindx) SETERRQ(PetscObjectComm((PetscObject)pc),PETSC_ERR_SUP_SYS,"ML not compiled with Zoltan");
1123: #endif
1124: PetscOptionsBool("-pc_ml_Aux","Aggregate using auxiliary coordinate-based laplacian","None",pc_ml->Aux,&pc_ml->Aux,NULL);
1125: PetscOptionsReal("-pc_ml_AuxThreshold","Auxiliary smoother drop tol","None",pc_ml->AuxThreshold,&pc_ml->AuxThreshold,NULL);
1126: }
1127: PetscOptionsTail();
1128: return(0);
1129: }
1131: /* -------------------------------------------------------------------------- */
1132: /*
1133: PCCreate_ML - Creates a ML preconditioner context, PC_ML,
1134: and sets this as the private data within the generic preconditioning
1135: context, PC, that was created within PCCreate().
1137: Input Parameter:
1138: . pc - the preconditioner context
1140: Application Interface Routine: PCCreate()
1141: */
1143: /*MC
1144: PCML - Use algebraic multigrid preconditioning. This preconditioner requires you provide
1145: fine grid discretization matrix. The coarser grid matrices and restriction/interpolation
1146: operators are computed by ML, with the matrices coverted to PETSc matrices in aij format
1147: and the restriction/interpolation operators wrapped as PETSc shell matrices.
1149: Options Database Key:
1150: Multigrid options(inherited)
1151: + -pc_mg_cycles <1>: 1 for V cycle, 2 for W-cycle (MGSetCycles)
1152: . -pc_mg_smoothup <1>: Number of post-smoothing steps (MGSetNumberSmoothUp)
1153: . -pc_mg_smoothdown <1>: Number of pre-smoothing steps (MGSetNumberSmoothDown)
1154: -pc_mg_type <multiplicative>: (one of) additive multiplicative full kascade
1155: ML options:
1156: . -pc_ml_PrintLevel <0>: Print level (ML_Set_PrintLevel)
1157: . -pc_ml_maxNlevels <10>: Maximum number of levels (None)
1158: . -pc_ml_maxCoarseSize <1>: Maximum coarsest mesh size (ML_Aggregate_Set_MaxCoarseSize)
1159: . -pc_ml_CoarsenScheme <Uncoupled>: (one of) Uncoupled Coupled MIS METIS
1160: . -pc_ml_DampingFactor <1.33333>: P damping factor (ML_Aggregate_Set_DampingFactor)
1161: . -pc_ml_Threshold <0>: Smoother drop tol (ML_Aggregate_Set_Threshold)
1162: . -pc_ml_SpectralNormScheme_Anorm <false>: Method used for estimating spectral radius (ML_Set_SpectralNormScheme_Anorm)
1163: . -pc_ml_repartition <false>: Allow ML to repartition levels of the heirarchy (ML_Repartition_Activate)
1164: . -pc_ml_repartitionMaxMinRatio <1.3>: Acceptable ratio of repartitioned sizes (ML_Repartition_Set_LargestMinMaxRatio)
1165: . -pc_ml_repartitionMinPerProc <512>: Smallest repartitioned size (ML_Repartition_Set_MinPerProc)
1166: . -pc_ml_repartitionPutOnSingleProc <5000>: Problem size automatically repartitioned to one processor (ML_Repartition_Set_PutOnSingleProc)
1167: . -pc_ml_repartitionType <Zoltan>: Repartitioning library to use (ML_Repartition_Set_Partitioner)
1168: . -pc_ml_repartitionZoltanScheme <RCB>: Repartitioning scheme to use (None)
1169: . -pc_ml_Aux <false>: Aggregate using auxiliary coordinate-based laplacian (None)
1170: - -pc_ml_AuxThreshold <0.0>: Auxiliary smoother drop tol (None)
1172: Level: intermediate
1174: Concepts: multigrid
1176: .seealso: PCCreate(), PCSetType(), PCType (for list of available types), PC, PCMGType,
1177: PCMGSetLevels(), PCMGGetLevels(), PCMGSetType(), MPSetCycles(), PCMGSetNumberSmoothDown(),
1178: PCMGSetNumberSmoothUp(), PCMGGetCoarseSolve(), PCMGSetResidual(), PCMGSetInterpolation(),
1179: PCMGSetRestriction(), PCMGGetSmoother(), PCMGGetSmootherUp(), PCMGGetSmootherDown(),
1180: PCMGSetCyclesOnLevel(), PCMGSetRhs(), PCMGSetX(), PCMGSetR()
1181: M*/
1185: PETSC_EXTERN PetscErrorCode PCCreate_ML(PC pc)
1186: {
1188: PC_ML *pc_ml;
1189: PC_MG *mg;
1192: /* PCML is an inherited class of PCMG. Initialize pc as PCMG */
1193: PCSetType(pc,PCMG); /* calls PCCreate_MG() and MGCreate_Private() */
1194: PetscObjectChangeTypeName((PetscObject)pc,PCML);
1195: /* Since PCMG tries to use DM assocated with PC must delete it */
1196: DMDestroy(&pc->dm);
1197: mg = (PC_MG*)pc->data;
1198: mg->galerkin = 2; /* Use Galerkin, but it is computed externally */
1200: /* create a supporting struct and attach it to pc */
1201: PetscNewLog(pc,&pc_ml);
1202: mg->innerctx = pc_ml;
1204: pc_ml->ml_object = 0;
1205: pc_ml->agg_object = 0;
1206: pc_ml->gridctx = 0;
1207: pc_ml->PetscMLdata = 0;
1208: pc_ml->Nlevels = -1;
1209: pc_ml->MaxNlevels = 10;
1210: pc_ml->MaxCoarseSize = 1;
1211: pc_ml->CoarsenScheme = 1;
1212: pc_ml->Threshold = 0.0;
1213: pc_ml->DampingFactor = 4.0/3.0;
1214: pc_ml->SpectralNormScheme_Anorm = PETSC_FALSE;
1215: pc_ml->size = 0;
1216: pc_ml->dim = 0;
1217: pc_ml->nloc = 0;
1218: pc_ml->coords = 0;
1219: pc_ml->Repartition = PETSC_FALSE;
1220: pc_ml->MaxMinRatio = 1.3;
1221: pc_ml->MinPerProc = 512;
1222: pc_ml->PutOnSingleProc = 5000;
1223: pc_ml->RepartitionType = 0;
1224: pc_ml->ZoltanScheme = 0;
1225: pc_ml->Aux = PETSC_FALSE;
1226: pc_ml->AuxThreshold = 0.0;
1228: /* allow for coordinates to be passed */
1229: PetscObjectComposeFunction((PetscObject)pc,"PCSetCoordinates_C",PCSetCoordinates_ML);
1231: /* overwrite the pointers of PCMG by the functions of PCML */
1232: pc->ops->setfromoptions = PCSetFromOptions_ML;
1233: pc->ops->setup = PCSetUp_ML;
1234: pc->ops->reset = PCReset_ML;
1235: pc->ops->destroy = PCDestroy_ML;
1236: return(0);
1237: }