PetscErrorCode PCSetUp_ML(PC pc)
{
PetscErrorCode ierr;
PetscMPIInt size;
FineGridCtx *PetscMLdata;
ML *ml_object;
ML_Aggregate *agg_object;
ML_Operator *mlmat;
PetscInt nlocal_allcols,Nlevels,mllevel,level,level1,m,fine_level,bs;
Mat A,Aloc;
GridCtx *gridctx;
PC_MG *mg = (PC_MG*)pc->data;
PC_ML *pc_ml = (PC_ML*)mg->innerctx;
PetscBool isSeq, isMPI;
KSP smoother;
PC subpc;
PetscInt mesh_level, old_mesh_level;
PetscFunctionBegin;
A = pc->pmat;
ierr = MPI_Comm_size(((PetscObject)A)->comm,&size);CHKERRQ(ierr);
if (pc->setupcalled) {
if (pc->flag == SAME_NONZERO_PATTERN && pc_ml->reuse_interpolation) {
/*
Reuse interpolaton instead of recomputing aggregates and updating the whole hierarchy. This is less expensive for
multiple solves in which the matrix is not changing too quickly.
*/
ml_object = pc_ml->ml_object;
gridctx = pc_ml->gridctx;
Nlevels = pc_ml->Nlevels;
fine_level = Nlevels - 1;
gridctx[fine_level].A = A;
ierr = PetscObjectTypeCompare((PetscObject) A, MATSEQAIJ, &isSeq);CHKERRQ(ierr);
ierr = PetscObjectTypeCompare((PetscObject) A, MATMPIAIJ, &isMPI);CHKERRQ(ierr);
if (isMPI){
ierr = MatConvert_MPIAIJ_ML(A,PETSC_NULL,MAT_INITIAL_MATRIX,&Aloc);CHKERRQ(ierr);
} else if (isSeq) {
Aloc = A;
ierr = PetscObjectReference((PetscObject)Aloc);CHKERRQ(ierr);
} else SETERRQ1(((PetscObject)pc)->comm,PETSC_ERR_ARG_WRONG, "Matrix type '%s' cannot be used with ML. ML can only handle AIJ matrices.",((PetscObject)A)->type_name);
ierr = MatGetSize(Aloc,&m,&nlocal_allcols);CHKERRQ(ierr);
PetscMLdata = pc_ml->PetscMLdata;
ierr = MatDestroy(&PetscMLdata->Aloc);CHKERRQ(ierr);
PetscMLdata->A = A;
PetscMLdata->Aloc = Aloc;
ML_Init_Amatrix(ml_object,0,m,m,PetscMLdata);
ML_Set_Amatrix_Matvec(ml_object,0,PetscML_matvec);
mesh_level = ml_object->ML_finest_level;
while (ml_object->SingleLevel[mesh_level].Rmat->to) {
old_mesh_level = mesh_level;
mesh_level = ml_object->SingleLevel[mesh_level].Rmat->to->levelnum;
/* clean and regenerate A */
mlmat = &(ml_object->Amat[mesh_level]);
ML_Operator_Clean(mlmat);
ML_Operator_Init(mlmat,ml_object->comm);
ML_Gen_AmatrixRAP(ml_object, old_mesh_level, mesh_level);
}
level = fine_level - 1;
if (size == 1) { /* convert ML P, R and A into seqaij format */
for (mllevel=1; mllevel<Nlevels; mllevel++){
mlmat = &(ml_object->Amat[mllevel]);
ierr = MatWrapML_SeqAIJ(mlmat,MAT_REUSE_MATRIX,&gridctx[level].A);CHKERRQ(ierr);
level--;
}
} else { /* convert ML P and R into shell format, ML A into mpiaij format */
for (mllevel=1; mllevel<Nlevels; mllevel++){
mlmat = &(ml_object->Amat[mllevel]);
ierr = MatWrapML_MPIAIJ(mlmat,MAT_REUSE_MATRIX,&gridctx[level].A);CHKERRQ(ierr);
level--;
}
}
for (level=0; level<fine_level; level++) {
if (level > 0){
ierr = PCMGSetResidual(pc,level,PCMGDefaultResidual,gridctx[level].A);CHKERRQ(ierr);
}
ierr = KSPSetOperators(gridctx[level].ksp,gridctx[level].A,gridctx[level].A,SAME_NONZERO_PATTERN);CHKERRQ(ierr);
}
ierr = PCMGSetResidual(pc,fine_level,PCMGDefaultResidual,gridctx[fine_level].A);CHKERRQ(ierr);
ierr = KSPSetOperators(gridctx[fine_level].ksp,gridctx[level].A,gridctx[fine_level].A,SAME_NONZERO_PATTERN);CHKERRQ(ierr);
ierr = PCSetUp_MG(pc);CHKERRQ(ierr);
PetscFunctionReturn(0);
} else {
/* since ML can change the size of vectors/matrices at any level we must destroy everything */
ierr = PCReset_ML(pc);CHKERRQ(ierr);
ierr = PCReset_MG(pc);CHKERRQ(ierr);
}
}
/* setup special features of PCML */
/*--------------------------------*/
/* covert A to Aloc to be used by ML at fine grid */
pc_ml->size = size;
ierr = PetscObjectTypeCompare((PetscObject) A, MATSEQAIJ, &isSeq);CHKERRQ(ierr);
ierr = PetscObjectTypeCompare((PetscObject) A, MATMPIAIJ, &isMPI);CHKERRQ(ierr);
if (isMPI){
ierr = MatConvert_MPIAIJ_ML(A,PETSC_NULL,MAT_INITIAL_MATRIX,&Aloc);CHKERRQ(ierr);
} else if (isSeq) {
Aloc = A;
ierr = PetscObjectReference((PetscObject)Aloc);CHKERRQ(ierr);
} else SETERRQ1(((PetscObject)pc)->comm,PETSC_ERR_ARG_WRONG, "Matrix type '%s' cannot be used with ML. ML can only handle AIJ matrices.",((PetscObject)A)->type_name);
/* create and initialize struct 'PetscMLdata' */
ierr = PetscNewLog(pc,FineGridCtx,&PetscMLdata);CHKERRQ(ierr);
pc_ml->PetscMLdata = PetscMLdata;
ierr = PetscMalloc((Aloc->cmap->n+1)*sizeof(PetscScalar),&PetscMLdata->pwork);CHKERRQ(ierr);
ierr = VecCreate(PETSC_COMM_SELF,&PetscMLdata->x);CHKERRQ(ierr);
ierr = VecSetSizes(PetscMLdata->x,Aloc->cmap->n,Aloc->cmap->n);CHKERRQ(ierr);
ierr = VecSetType(PetscMLdata->x,VECSEQ);CHKERRQ(ierr);
ierr = VecCreate(PETSC_COMM_SELF,&PetscMLdata->y);CHKERRQ(ierr);
ierr = VecSetSizes(PetscMLdata->y,A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr);
ierr = VecSetType(PetscMLdata->y,VECSEQ);CHKERRQ(ierr);
PetscMLdata->A = A;
PetscMLdata->Aloc = Aloc;
/* create ML discretization matrix at fine grid */
/* ML requires input of fine-grid matrix. It determines nlevels. */
ierr = MatGetSize(Aloc,&m,&nlocal_allcols);CHKERRQ(ierr);
ierr = MatGetBlockSize(A,&bs);CHKERRQ(ierr);
ML_Create(&ml_object,pc_ml->MaxNlevels);
ML_Comm_Set_UsrComm(ml_object->comm,((PetscObject)A)->comm);
pc_ml->ml_object = ml_object;
ML_Init_Amatrix(ml_object,0,m,m,PetscMLdata);
ML_Set_Amatrix_Getrow(ml_object,0,PetscML_getrow,PetscML_comm,nlocal_allcols);
ML_Set_Amatrix_Matvec(ml_object,0,PetscML_matvec);
ML_Set_Symmetrize(ml_object,pc_ml->Symmetrize ? ML_YES : ML_NO);
/* aggregation */
ML_Aggregate_Create(&agg_object);
pc_ml->agg_object = agg_object;
{
MatNullSpace mnull;
ierr = MatGetNearNullSpace(A,&mnull);CHKERRQ(ierr);
if (pc_ml->nulltype == PCML_NULLSPACE_AUTO) {
if (mnull) pc_ml->nulltype = PCML_NULLSPACE_USER;
else if (bs > 1) pc_ml->nulltype = PCML_NULLSPACE_BLOCK;
else pc_ml->nulltype = PCML_NULLSPACE_SCALAR;
}
switch (pc_ml->nulltype) {
case PCML_NULLSPACE_USER: {
PetscScalar *nullvec;
const PetscScalar *v;
PetscBool has_const;
PetscInt i,j,mlocal,nvec,M;
const Vec *vecs;
if (!mnull) SETERRQ(((PetscObject)pc)->comm,PETSC_ERR_USER,"Must provide explicit null space using MatSetNearNullSpace() to use user-specified null space");
ierr = MatGetSize(A,&M,PETSC_NULL);CHKERRQ(ierr);
ierr = MatGetLocalSize(Aloc,&mlocal,PETSC_NULL);CHKERRQ(ierr);
ierr = MatNullSpaceGetVecs(mnull,&has_const,&nvec,&vecs);CHKERRQ(ierr);
ierr = PetscMalloc((nvec+!!has_const)*mlocal*sizeof *nullvec,&nullvec);CHKERRQ(ierr);
if (has_const) for (i=0; i<mlocal; i++) nullvec[i] = 1.0/M;
for (i=0; i<nvec; i++) {
ierr = VecGetArrayRead(vecs[i],&v);CHKERRQ(ierr);
for (j=0; j<mlocal; j++) nullvec[(i+!!has_const)*mlocal + j] = v[j];
ierr = VecRestoreArrayRead(vecs[i],&v);CHKERRQ(ierr);
}
ierr = ML_Aggregate_Set_NullSpace(agg_object,bs,nvec+!!has_const,nullvec,mlocal);CHKERRQ(ierr);
ierr = PetscFree(nullvec);CHKERRQ(ierr);
} break;
case PCML_NULLSPACE_BLOCK:
ierr = ML_Aggregate_Set_NullSpace(agg_object,bs,bs,0,0);CHKERRQ(ierr);
break;
case PCML_NULLSPACE_SCALAR:
break;
default: SETERRQ(((PetscObject)pc)->comm,PETSC_ERR_SUP,"Unknown null space type");
}
}
ML_Aggregate_Set_MaxCoarseSize(agg_object,pc_ml->MaxCoarseSize);
/* set options */
switch (pc_ml->CoarsenScheme) {
case 1:
ML_Aggregate_Set_CoarsenScheme_Coupled(agg_object);break;
case 2:
ML_Aggregate_Set_CoarsenScheme_MIS(agg_object);break;
case 3:
ML_Aggregate_Set_CoarsenScheme_METIS(agg_object);break;
}
ML_Aggregate_Set_Threshold(agg_object,pc_ml->Threshold);
ML_Aggregate_Set_DampingFactor(agg_object,pc_ml->DampingFactor);
if (pc_ml->SpectralNormScheme_Anorm){
ML_Set_SpectralNormScheme_Anorm(ml_object);
}
agg_object->keep_agg_information = (int)pc_ml->KeepAggInfo;
agg_object->keep_P_tentative = (int)pc_ml->Reusable;
agg_object->block_scaled_SA = (int)pc_ml->BlockScaling;
agg_object->minimizing_energy = (int)pc_ml->EnergyMinimization;
agg_object->minimizing_energy_droptol = (double)pc_ml->EnergyMinimizationDropTol;
agg_object->cheap_minimizing_energy = (int)pc_ml->EnergyMinimizationCheap;
if (pc_ml->OldHierarchy) {
Nlevels = ML_Gen_MGHierarchy_UsingAggregation(ml_object,0,ML_INCREASING,agg_object);
} else {
Nlevels = ML_Gen_MultiLevelHierarchy_UsingAggregation(ml_object,0,ML_INCREASING,agg_object);
}
if (Nlevels<=0) SETERRQ1(((PetscObject)pc)->comm,PETSC_ERR_ARG_OUTOFRANGE,"Nlevels %d must > 0",Nlevels);
pc_ml->Nlevels = Nlevels;
fine_level = Nlevels - 1;
ierr = PCMGSetLevels(pc,Nlevels,PETSC_NULL);CHKERRQ(ierr);
/* set default smoothers */
for (level=1; level<=fine_level; level++){
if (size == 1){
ierr = PCMGGetSmoother(pc,level,&smoother);CHKERRQ(ierr);
ierr = KSPSetType(smoother,KSPRICHARDSON);CHKERRQ(ierr);
ierr = KSPGetPC(smoother,&subpc);CHKERRQ(ierr);
ierr = PCSetType(subpc,PCSOR);CHKERRQ(ierr);
} else {
ierr = PCMGGetSmoother(pc,level,&smoother);CHKERRQ(ierr);
ierr = KSPSetType(smoother,KSPRICHARDSON);CHKERRQ(ierr);
ierr = KSPGetPC(smoother,&subpc);CHKERRQ(ierr);
ierr = PCSetType(subpc,PCSOR);CHKERRQ(ierr);
}
}
ierr = PetscObjectOptionsBegin((PetscObject)pc);CHKERRQ(ierr);
ierr = PCSetFromOptions_MG(pc);CHKERRQ(ierr); /* should be called in PCSetFromOptions_ML(), but cannot be called prior to PCMGSetLevels() */
ierr = PetscOptionsEnd();CHKERRQ(ierr);
ierr = PetscMalloc(Nlevels*sizeof(GridCtx),&gridctx);CHKERRQ(ierr);
pc_ml->gridctx = gridctx;
/* wrap ML matrices by PETSc shell matrices at coarsened grids.
Level 0 is the finest grid for ML, but coarsest for PETSc! */
gridctx[fine_level].A = A;
level = fine_level - 1;
if (size == 1){ /* convert ML P, R and A into seqaij format */
for (mllevel=1; mllevel<Nlevels; mllevel++){
mlmat = &(ml_object->Pmat[mllevel]);
ierr = MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].P);CHKERRQ(ierr);
mlmat = &(ml_object->Rmat[mllevel-1]);
ierr = MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].R);CHKERRQ(ierr);
mlmat = &(ml_object->Amat[mllevel]);
ierr = MatWrapML_SeqAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].A);CHKERRQ(ierr);
level--;
}
} else { /* convert ML P and R into shell format, ML A into mpiaij format */
for (mllevel=1; mllevel<Nlevels; mllevel++){
mlmat = &(ml_object->Pmat[mllevel]);
ierr = MatWrapML_SHELL(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].P);CHKERRQ(ierr);
mlmat = &(ml_object->Rmat[mllevel-1]);
ierr = MatWrapML_SHELL(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].R);CHKERRQ(ierr);
mlmat = &(ml_object->Amat[mllevel]);
ierr = MatWrapML_MPIAIJ(mlmat,MAT_INITIAL_MATRIX,&gridctx[level].A);CHKERRQ(ierr);
level--;
}
}
/* create vectors and ksp at all levels */
for (level=0; level<fine_level; level++){
level1 = level + 1;
ierr = VecCreate(((PetscObject)gridctx[level].A)->comm,&gridctx[level].x);CHKERRQ(ierr);
ierr = VecSetSizes(gridctx[level].x,gridctx[level].A->cmap->n,PETSC_DECIDE);CHKERRQ(ierr);
ierr = VecSetType(gridctx[level].x,VECMPI);CHKERRQ(ierr);
ierr = PCMGSetX(pc,level,gridctx[level].x);CHKERRQ(ierr);
ierr = VecCreate(((PetscObject)gridctx[level].A)->comm,&gridctx[level].b);CHKERRQ(ierr);
ierr = VecSetSizes(gridctx[level].b,gridctx[level].A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr);
ierr = VecSetType(gridctx[level].b,VECMPI);CHKERRQ(ierr);
ierr = PCMGSetRhs(pc,level,gridctx[level].b);CHKERRQ(ierr);
ierr = VecCreate(((PetscObject)gridctx[level1].A)->comm,&gridctx[level1].r);CHKERRQ(ierr);
ierr = VecSetSizes(gridctx[level1].r,gridctx[level1].A->rmap->n,PETSC_DECIDE);CHKERRQ(ierr);
ierr = VecSetType(gridctx[level1].r,VECMPI);CHKERRQ(ierr);
ierr = PCMGSetR(pc,level1,gridctx[level1].r);CHKERRQ(ierr);
if (level == 0){
ierr = PCMGGetCoarseSolve(pc,&gridctx[level].ksp);CHKERRQ(ierr);
} else {
ierr = PCMGGetSmoother(pc,level,&gridctx[level].ksp);CHKERRQ(ierr);
}
}
ierr = PCMGGetSmoother(pc,fine_level,&gridctx[fine_level].ksp);CHKERRQ(ierr);
/* create coarse level and the interpolation between the levels */
for (level=0; level<fine_level; level++){
level1 = level + 1;
ierr = PCMGSetInterpolation(pc,level1,gridctx[level].P);CHKERRQ(ierr);
ierr = PCMGSetRestriction(pc,level1,gridctx[level].R);CHKERRQ(ierr);
if (level > 0){
ierr = PCMGSetResidual(pc,level,PCMGDefaultResidual,gridctx[level].A);CHKERRQ(ierr);
}
ierr = KSPSetOperators(gridctx[level].ksp,gridctx[level].A,gridctx[level].A,DIFFERENT_NONZERO_PATTERN);CHKERRQ(ierr);
}
ierr = PCMGSetResidual(pc,fine_level,PCMGDefaultResidual,gridctx[fine_level].A);CHKERRQ(ierr);
ierr = KSPSetOperators(gridctx[fine_level].ksp,gridctx[level].A,gridctx[fine_level].A,DIFFERENT_NONZERO_PATTERN);CHKERRQ(ierr);
/* setupcalled is set to 0 so that MG is setup from scratch */
pc->setupcalled = 0;
ierr = PCSetUp_MG(pc);CHKERRQ(ierr);
PetscFunctionReturn(0);
}
int main(int argc, char *argv[])
{
int num_PDE_eqns=1, N_levels=3, nsmooth=2;
int leng, level, N_grid_pts, coarsest_level;
int leng1,leng2;
/* See Aztec User's Guide for more information on the */
/* variables that follow. */
int proc_config[AZ_PROC_SIZE], options[AZ_OPTIONS_SIZE];
double params[AZ_PARAMS_SIZE], status[AZ_STATUS_SIZE];
/* data structure for matrix corresponding to the fine grid */
double *val = NULL, *xxx, *rhs, solve_time, setup_time, start_time;
AZ_MATRIX *Amat;
AZ_PRECOND *Pmat = NULL;
ML *ml;
FILE *fp;
int i, j, Nrigid, *garbage, nblocks=0, *blocks = NULL, *block_pde=NULL;
struct AZ_SCALING *scaling;
ML_Aggregate *ag;
double *mode, *rigid=NULL, alpha;
char filename[80];
int one = 1;
int proc,nprocs;
char pathfilename[100];
#ifdef ML_MPI
MPI_Init(&argc,&argv);
/* get number of processors and the name of this processor */
AZ_set_proc_config(proc_config, MPI_COMM_WORLD);
proc = proc_config[AZ_node];
nprocs = proc_config[AZ_N_procs];
#else
AZ_set_proc_config(proc_config, AZ_NOT_MPI);
proc = 0;
nprocs = 1;
#endif
if (proc_config[AZ_node] == 0) {
sprintf(pathfilename,"%s/inputfile",argv[1]);
ML_Reader_ReadInput(pathfilename, &context);
}
else context = (struct reader_context *) ML_allocate(sizeof(struct reader_context));
AZ_broadcast((char *) context, sizeof(struct reader_context), proc_config,
AZ_PACK);
AZ_broadcast((char *) NULL , 0 , proc_config, AZ_SEND);
N_levels = context->N_levels;
printf("N_levels %d\n",N_levels);
nsmooth = context->nsmooth;
num_PDE_eqns = context->N_dofPerNode;
printf("num_PDE_eqns %d\n",num_PDE_eqns);
ML_Set_PrintLevel(context->output_level);
/* read in the number of matrix equations */
leng = 0;
if (proc_config[AZ_node] == 0) {
sprintf(pathfilename,"%s/data_matrix.txt",argv[1]);
fp=fopen(pathfilename,"r");
if (fp==NULL) {
printf("**ERR** couldn't open file data_matrix.txt\n");
exit(1);
}
fscanf(fp,"%d",&leng);
fclose(fp);
}
leng = AZ_gsum_int(leng, proc_config);
N_grid_pts=leng/num_PDE_eqns;
/* initialize the list of global indices. NOTE: the list of global */
/* indices must be in ascending order so that subsequent calls to */
/* AZ_find_index() will function properly. */
#if 0
if (proc_config[AZ_N_procs] == 1) i = AZ_linear;
else i = AZ_file;
#endif
i = AZ_linear;
/* cannot use AZ_input_update for variable blocks (forgot why, but debugged through it)*/
/* make a linear distribution of the matrix */
/* if the linear distribution does not align with the blocks, */
/* this is corrected in ML_AZ_Reader_ReadVariableBlocks */
leng1 = leng/nprocs;
leng2 = leng-leng1*nprocs;
if (proc >= leng2)
{
leng2 += (proc*leng1);
}
else
{
leng1++;
leng2 = proc*leng1;
}
N_update = leng1;
update = (int*)AZ_allocate((N_update+1)*sizeof(int));
if (update==NULL)
{
(void) fprintf (stderr, "Not enough space to allocate 'update'\n");
fflush(stderr); exit(EXIT_FAILURE);
}
for (i=0; i<N_update; i++) update[i] = i+leng2;
#if 0 /* debug */
printf("proc %d N_update %d\n",proc_config[AZ_node],N_update);
fflush(stdout);
#endif
sprintf(pathfilename,"%s/data_vblocks.txt",argv[1]);
ML_AZ_Reader_ReadVariableBlocks(pathfilename,&nblocks,&blocks,&block_pde,
&N_update,&update,proc_config);
#if 0 /* debug */
printf("proc %d N_update %d\n",proc_config[AZ_node],N_update);
fflush(stdout);
#endif
sprintf(pathfilename,"%s/data_matrix.txt",argv[1]);
AZ_input_msr_matrix(pathfilename,update, &val, &bindx, N_update, proc_config);
/* This code is to fix things up so that we are sure we have */
/* all blocks (including the ghost nodes) the same size. */
/* not sure, whether this is a good idea with variable blocks */
/* the examples inpufiles (see top of this file) don't need it */
/* anyway */
/*
AZ_block_MSR(&bindx, &val, N_update, num_PDE_eqns, update);
*/
AZ_transform_norowreordering(proc_config, &external, bindx, val, update, &update_index,
&extern_index, &data_org, N_update, 0, 0, 0, &cpntr,
AZ_MSR_MATRIX);
Amat = AZ_matrix_create( leng );
AZ_set_MSR(Amat, bindx, val, data_org, 0, NULL, AZ_LOCAL);
Amat->matrix_type = data_org[AZ_matrix_type];
data_org[AZ_N_rows] = data_org[AZ_N_internal] + data_org[AZ_N_border];
start_time = AZ_second();
options[AZ_scaling] = AZ_none;
ML_Create(&ml, N_levels);
/* set up discretization matrix and matrix vector function */
AZ_ML_Set_Amat(ml, 0, N_update, N_update, Amat, proc_config);
ML_Set_ResidualOutputFrequency(ml, context->output);
ML_Set_Tolerance(ml, context->tol);
ML_Aggregate_Create( &ag );
if (ML_strcmp(context->agg_coarsen_scheme,"Mis") == 0) {
ML_Aggregate_Set_CoarsenScheme_MIS(ag);
}
else if (ML_strcmp(context->agg_coarsen_scheme,"Uncoupled") == 0) {
ML_Aggregate_Set_CoarsenScheme_Uncoupled(ag);
}
else if (ML_strcmp(context->agg_coarsen_scheme,"Coupled") == 0) {
ML_Aggregate_Set_CoarsenScheme_Coupled(ag);
}
else if (ML_strcmp(context->agg_coarsen_scheme,"Metis") == 0) {
ML_Aggregate_Set_CoarsenScheme_METIS(ag);
for (i=0; i<N_levels; i++)
ML_Aggregate_Set_NodesPerAggr(ml,ag,i,9);
}
else if (ML_strcmp(context->agg_coarsen_scheme,"VBMetis") == 0) {
/* when no blocks read, use standard metis assuming constant block sizes */
if (!blocks)
ML_Aggregate_Set_CoarsenScheme_METIS(ag);
else {
ML_Aggregate_Set_CoarsenScheme_VBMETIS(ag);
ML_Aggregate_Set_Vblocks_CoarsenScheme_VBMETIS(ag,0,N_levels,nblocks,
blocks,block_pde,N_update);
}
for (i=0; i<N_levels; i++)
ML_Aggregate_Set_NodesPerAggr(ml,ag,i,9);
}
else {
printf("**ERR** ML: Unknown aggregation scheme %s\n",context->agg_coarsen_scheme);
exit(-1);
}
ML_Aggregate_Set_DampingFactor(ag, context->agg_damping);
ML_Aggregate_Set_MaxCoarseSize( ag, context->maxcoarsesize);
ML_Aggregate_Set_Threshold(ag, context->agg_thresh);
if (ML_strcmp(context->agg_spectral_norm,"Calc") == 0) {
ML_Set_SpectralNormScheme_Calc(ml);
}
else if (ML_strcmp(context->agg_spectral_norm,"Anorm") == 0) {
ML_Set_SpectralNormScheme_Anorm(ml);
}
else {
printf("**WRN** ML: Unknown spectral norm scheme %s\n",context->agg_spectral_norm);
}
/* read in the rigid body modes */
Nrigid = 0;
if (proc_config[AZ_node] == 0) {
sprintf(filename,"data_nullsp%d.txt",Nrigid);
sprintf(pathfilename,"%s/%s",argv[1],filename);
while( (fp = fopen(pathfilename,"r")) != NULL) {
fclose(fp);
Nrigid++;
sprintf(filename,"data_nullsp%d.txt",Nrigid);
sprintf(pathfilename,"%s/%s",argv[1],filename);
}
}
Nrigid = AZ_gsum_int(Nrigid,proc_config);
if (Nrigid != 0) {
rigid = (double *) ML_allocate( sizeof(double)*Nrigid*(N_update+1) );
if (rigid == NULL) {
printf("Error: Not enough space for rigid body modes\n");
}
}
/* Set rhs */
sprintf(pathfilename,"%s/data_rhs.txt",argv[1]);
fp = fopen(pathfilename,"r");
if (fp == NULL) {
rhs=(double *)ML_allocate(leng*sizeof(double));
if (proc_config[AZ_node] == 0) printf("taking linear vector for rhs\n");
for (i = 0; i < N_update; i++) rhs[i] = (double) update[i];
}
else {
fclose(fp);
if (proc_config[AZ_node] == 0) printf("reading rhs from a file\n");
AZ_input_msr_matrix(pathfilename, update, &rhs, &garbage, N_update,
proc_config);
}
AZ_reorder_vec(rhs, data_org, update_index, NULL);
for (i = 0; i < Nrigid; i++) {
sprintf(filename,"data_nullsp%d.txt",i);
sprintf(pathfilename,"%s/%s",argv[1],filename);
AZ_input_msr_matrix(pathfilename, update, &mode, &garbage, N_update,
proc_config);
AZ_reorder_vec(mode, data_org, update_index, NULL);
#if 0 /* test the given rigid body mode, output-vector should be ~0 */
Amat->matvec(mode, rigid, Amat, proc_config);
for (j = 0; j < N_update; j++) printf("this is %d %e\n",j,rigid[j]);
#endif
for (j = 0; j < i; j++) {
alpha = -AZ_gdot(N_update, mode, &(rigid[j*N_update]), proc_config)/
AZ_gdot(N_update, &(rigid[j*N_update]), &(rigid[j*N_update]),
proc_config);
DAXPY_F77(&N_update, &alpha, &(rigid[j*N_update]), &one, mode, &one);
}
/* rhs orthogonalization */
alpha = -AZ_gdot(N_update, mode, rhs, proc_config)/
AZ_gdot(N_update, mode, mode, proc_config);
DAXPY_F77(&N_update, &alpha, mode, &one, rhs, &one);
for (j = 0; j < N_update; j++) rigid[i*N_update+j] = mode[j];
free(mode);
free(garbage);
}
for (j = 0; j < Nrigid; j++) {
alpha = -AZ_gdot(N_update, rhs, &(rigid[j*N_update]), proc_config)/
AZ_gdot(N_update, &(rigid[j*N_update]), &(rigid[j*N_update]),
proc_config);
DAXPY_F77(&N_update, &alpha, &(rigid[j*N_update]), &one, rhs, &one);
}
#if 0 /* for testing the default nullsp */
ML_Aggregate_Set_NullSpace(ag, num_PDE_eqns, 6, NULL, N_update);
#else
if (Nrigid != 0) {
ML_Aggregate_Set_NullSpace(ag, num_PDE_eqns, Nrigid, rigid, N_update);
}
#endif
if (rigid) ML_free(rigid);
ag->keep_agg_information = 1;
coarsest_level = ML_Gen_MGHierarchy_UsingAggregation(ml, 0,
ML_INCREASING, ag);
coarsest_level--;
if ( proc_config[AZ_node] == 0 )
printf("Coarse level = %d \n", coarsest_level);
#if 0
/* set up smoothers */
if (!blocks)
blocks = (int *) ML_allocate(sizeof(int)*N_update);
#endif
for (level = 0; level < coarsest_level; level++) {
num_PDE_eqns = ml->Amat[level].num_PDEs;
/* Sparse approximate inverse smoother that acutally does both */
/* pre and post smoothing. */
if (ML_strcmp(context->smoother,"Parasails") == 0) {
ML_Gen_Smoother_ParaSails(ml , level, ML_PRESMOOTHER, nsmooth,
parasails_sym, parasails_thresh,
parasails_nlevels, parasails_filter,
(int) parasails_loadbal, parasails_factorized);
}
/* This is the symmetric Gauss-Seidel smoothing that we usually use. */
/* In parallel, it is not a true Gauss-Seidel in that each processor */
/* does a Gauss-Seidel on its local submatrix independent of the */
/* other processors. */
else if (ML_strcmp(context->smoother,"GaussSeidel") == 0) {
ML_Gen_Smoother_GaussSeidel(ml , level, ML_BOTH, nsmooth,1.);
}
else if (ML_strcmp(context->smoother,"SymGaussSeidel") == 0) {
ML_Gen_Smoother_SymGaussSeidel(ml , level, ML_BOTH, nsmooth,1.);
}
else if (ML_strcmp(context->smoother,"Poly") == 0) {
ML_Gen_Smoother_Cheby(ml, level, ML_BOTH, 30., nsmooth);
}
else if (ML_strcmp(context->smoother,"BlockGaussSeidel") == 0) {
ML_Gen_Smoother_BlockGaussSeidel(ml , level, ML_BOTH, nsmooth,1.,
num_PDE_eqns);
}
else if (ML_strcmp(context->smoother,"VBSymGaussSeidel") == 0) {
if (blocks) ML_free(blocks);
if (block_pde) ML_free(block_pde);
blocks = NULL;
block_pde = NULL;
nblocks = 0;
ML_Aggregate_Get_Vblocks_CoarsenScheme_VBMETIS(ag,level,N_levels,&nblocks,
&blocks,&block_pde);
if (blocks==NULL) ML_Gen_Blocks_Aggregates(ag, level, &nblocks, &blocks);
ML_Gen_Smoother_VBlockSymGaussSeidel(ml , level, ML_BOTH, nsmooth,1.,
nblocks, blocks);
}
/* This is a true Gauss Seidel in parallel. This seems to work for */
/* elasticity problems. However, I don't believe that this is very */
/* efficient in parallel. */
/*
nblocks = ml->Amat[level].invec_leng;
for (i =0; i < nblocks; i++) blocks[i] = i;
ML_Gen_Smoother_VBlockSymGaussSeidelSequential(ml , level, ML_PRESMOOTHER,
nsmooth, 1., nblocks, blocks);
ML_Gen_Smoother_VBlockSymGaussSeidelSequential(ml, level, ML_POSTSMOOTHER,
nsmooth, 1., nblocks, blocks);
*/
/* Jacobi Smoothing */
else if (ML_strcmp(context->smoother,"Jacobi") == 0) {
ML_Gen_Smoother_Jacobi(ml , level, ML_PRESMOOTHER, nsmooth,.4);
ML_Gen_Smoother_Jacobi(ml , level, ML_POSTSMOOTHER, nsmooth,.4);
}
/* This does a block Gauss-Seidel (not true GS in parallel) */
/* where each processor has 'nblocks' blocks. */
/* */
else if (ML_strcmp(context->smoother,"Metis") == 0) {
if (blocks) ML_free(blocks);
if (block_pde) ML_free(block_pde);
nblocks = 250;
ML_Gen_Blocks_Metis(ml, level, &nblocks, &blocks);
ML_Gen_Smoother_VBlockSymGaussSeidel(ml , level, ML_BOTH, nsmooth,1.,
nblocks, blocks);
}
else {
printf("unknown smoother %s\n",context->smoother);
exit(1);
}
}
/* set coarse level solver */
nsmooth = context->coarse_its;
/* Sparse approximate inverse smoother that acutally does both */
/* pre and post smoothing. */
if (ML_strcmp(context->coarse_solve,"Parasails") == 0) {
ML_Gen_Smoother_ParaSails(ml , coarsest_level, ML_PRESMOOTHER, nsmooth,
parasails_sym, parasails_thresh,
parasails_nlevels, parasails_filter,
(int) parasails_loadbal, parasails_factorized);
}
else if (ML_strcmp(context->coarse_solve,"GaussSeidel") == 0) {
ML_Gen_Smoother_GaussSeidel(ml , coarsest_level, ML_BOTH, nsmooth,1.);
}
else if (ML_strcmp(context->coarse_solve,"Poly") == 0) {
ML_Gen_Smoother_Cheby(ml, coarsest_level, ML_BOTH, 30., nsmooth);
}
else if (ML_strcmp(context->coarse_solve,"SymGaussSeidel") == 0) {
ML_Gen_Smoother_SymGaussSeidel(ml , coarsest_level, ML_BOTH, nsmooth,1.);
}
else if (ML_strcmp(context->coarse_solve,"BlockGaussSeidel") == 0) {
ML_Gen_Smoother_BlockGaussSeidel(ml, coarsest_level, ML_BOTH, nsmooth,1.,
num_PDE_eqns);
}
else if (ML_strcmp(context->coarse_solve,"Aggregate") == 0) {
if (blocks) ML_free(blocks);
if (block_pde) ML_free(block_pde);
ML_Gen_Blocks_Aggregates(ag, coarsest_level, &nblocks, &blocks);
ML_Gen_Smoother_VBlockSymGaussSeidel(ml , coarsest_level, ML_BOTH,
nsmooth,1., nblocks, blocks);
}
else if (ML_strcmp(context->coarse_solve,"Jacobi") == 0) {
ML_Gen_Smoother_Jacobi(ml , coarsest_level, ML_BOTH, nsmooth,.5);
}
else if (ML_strcmp(context->coarse_solve,"Metis") == 0) {
if (blocks) ML_free(blocks);
if (block_pde) ML_free(block_pde);
nblocks = 250;
ML_Gen_Blocks_Metis(ml, coarsest_level, &nblocks, &blocks);
ML_Gen_Smoother_VBlockSymGaussSeidel(ml , coarsest_level, ML_BOTH,
nsmooth,1., nblocks, blocks);
}
else if (ML_strcmp(context->coarse_solve,"SuperLU") == 0) {
ML_Gen_CoarseSolverSuperLU( ml, coarsest_level);
}
else if (ML_strcmp(context->coarse_solve,"Amesos") == 0) {
ML_Gen_Smoother_Amesos(ml,coarsest_level,ML_AMESOS_KLU,-1, 0.0);
}
else {
printf("unknown coarse grid solver %s\n",context->coarse_solve);
exit(1);
}
ML_Gen_Solver(ml, ML_MGV, 0, coarsest_level);
AZ_defaults(options, params);
if (ML_strcmp(context->krylov,"Cg") == 0) {
options[AZ_solver] = AZ_cg;
}
else if (ML_strcmp(context->krylov,"Bicgstab") == 0) {
options[AZ_solver] = AZ_bicgstab;
}
else if (ML_strcmp(context->krylov,"Tfqmr") == 0) {
options[AZ_solver] = AZ_tfqmr;
}
else if (ML_strcmp(context->krylov,"Gmres") == 0) {
options[AZ_solver] = AZ_gmres;
}
else {
printf("unknown krylov method %s\n",context->krylov);
}
if (blocks) ML_free(blocks);
if (block_pde) ML_free(block_pde);
options[AZ_scaling] = AZ_none;
options[AZ_precond] = AZ_user_precond;
options[AZ_conv] = AZ_r0;
options[AZ_output] = 1;
options[AZ_max_iter] = context->max_outer_its;
options[AZ_poly_ord] = 5;
options[AZ_kspace] = 130;
params[AZ_tol] = context->tol;
options[AZ_output] = context->output;
ML_free(context);
AZ_set_ML_preconditioner(&Pmat, Amat, ml, options);
setup_time = AZ_second() - start_time;
xxx = (double *) malloc( leng*sizeof(double));
for (iii = 0; iii < leng; iii++) xxx[iii] = 0.0;
/* Set x */
/*
there is no initguess supplied with these examples for the moment....
*/
fp = fopen("initguessfile","r");
if (fp != NULL) {
fclose(fp);
if (proc_config[AZ_node]== 0) printf("reading initial guess from file\n");
AZ_input_msr_matrix("data_initguess.txt", update, &xxx, &garbage, N_update,
proc_config);
options[AZ_conv] = AZ_expected_values;
}
else if (proc_config[AZ_node]== 0) printf("taking 0 initial guess \n");
AZ_reorder_vec(xxx, data_org, update_index, NULL);
/* if Dirichlet BC ... put the answer in */
for (i = 0; i < data_org[AZ_N_internal]+data_org[AZ_N_border]; i++) {
if ( (val[i] > .99999999) && (val[i] < 1.0000001))
xxx[i] = rhs[i];
}
fp = fopen("AZ_no_multilevel.dat","r");
scaling = AZ_scaling_create();
start_time = AZ_second();
if (fp != NULL) {
fclose(fp);
options[AZ_precond] = AZ_none;
options[AZ_scaling] = AZ_sym_diag;
options[AZ_ignore_scaling] = AZ_TRUE;
options[AZ_keep_info] = 1;
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, NULL, scaling);
/*
options[AZ_pre_calc] = AZ_reuse;
options[AZ_conv] = AZ_expected_values;
if (proc_config[AZ_node] == 0)
printf("\n-------- Second solve with improved convergence test -----\n");
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, NULL, scaling);
if (proc_config[AZ_node] == 0)
printf("\n-------- Third solve with improved convergence test -----\n");
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, NULL, scaling);
*/
}
else {
options[AZ_keep_info] = 1;
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, Pmat, scaling);
options[AZ_pre_calc] = AZ_reuse;
options[AZ_conv] = AZ_expected_values;
/*
if (proc_config[AZ_node] == 0)
printf("\n-------- Second solve with improved convergence test -----\n");
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, Pmat, scaling);
if (proc_config[AZ_node] == 0)
printf("\n-------- Third solve with improved convergence test -----\n");
AZ_iterate(xxx, rhs, options, params, status, proc_config, Amat, Pmat, scaling);
*/
}
solve_time = AZ_second() - start_time;
if (proc_config[AZ_node] == 0)
printf("Solve time = %e, MG Setup time = %e\n", solve_time, setup_time);
if (proc_config[AZ_node] == 0)
printf("Printing out a few entries of the solution ...\n");
for (j=0;j<Amat->data_org[AZ_N_internal]+ Amat->data_org[AZ_N_border];j++)
if (update[j] == 7) {printf("solution(gid = %d) = %10.4e\n",
update[j],xxx[update_index[j]]); fflush(stdout);}
j = AZ_gsum_int(7, proc_config); /* sync processors */
for (j=0;j<Amat->data_org[AZ_N_internal]+ Amat->data_org[AZ_N_border];j++)
if (update[j] == 23) {printf("solution(gid = %d) = %10.4e\n",
update[j],xxx[update_index[j]]); fflush(stdout);}
j = AZ_gsum_int(7, proc_config); /* sync processors */
for (j=0;j<Amat->data_org[AZ_N_internal]+ Amat->data_org[AZ_N_border];j++)
if (update[j] == 47) {printf("solution(gid = %d) = %10.4e\n",
update[j],xxx[update_index[j]]); fflush(stdout);}
j = AZ_gsum_int(7, proc_config); /* sync processors */
for (j=0;j<Amat->data_org[AZ_N_internal]+ Amat->data_org[AZ_N_border];j++)
if (update[j] == 101) {printf("solution(gid = %d) = %10.4e\n",
update[j],xxx[update_index[j]]); fflush(stdout);}
j = AZ_gsum_int(7, proc_config); /* sync processors */
for (j=0;j<Amat->data_org[AZ_N_internal]+ Amat->data_org[AZ_N_border];j++)
if (update[j] == 171) {printf("solution(gid = %d) = %10.4e\n",
update[j],xxx[update_index[j]]); fflush(stdout);}
ML_Aggregate_Destroy(&ag);
ML_Destroy(&ml);
AZ_free((void *) Amat->data_org);
AZ_free((void *) Amat->val);
AZ_free((void *) Amat->bindx);
AZ_free((void *) update);
AZ_free((void *) external);
AZ_free((void *) extern_index);
AZ_free((void *) update_index);
AZ_scaling_destroy(&scaling);
if (Amat != NULL) AZ_matrix_destroy(&Amat);
if (Pmat != NULL) AZ_precond_destroy(&Pmat);
free(xxx);
free(rhs);
#ifdef ML_MPI
MPI_Finalize();
#endif
return 0;
}
