//==============================================================================
int Ifpack_ML::
ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const
{
if (IsComputed() == false)
IFPACK_CHK_ERR(-1);
int numVectors = X.NumVectors();
if (numVectors != Y.NumVectors())
IFPACK_CHK_ERR(-1); // wrong input
Time_->ResetStartTime();
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
Teuchos::RefCountPtr<const Epetra_MultiVector> Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0])
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
else
Xcopy = Teuchos::rcp( &X, false );
for (int i=0; i<numVectors; i++) {
IFPACK_CHK_ERR(MLPrec_->ApplyInverse(*Xcopy,Y));
}
++NumApplyInverse_;
ApplyInverseTime_ += Time_->ElapsedTime();
return(0);
}
//=======================================================
int EpetraExt_HypreIJMatrix::Multiply(bool TransA,
const Epetra_MultiVector& X,
Epetra_MultiVector& Y) const
{
//printf("Proc[%d], Row start: %d, Row End: %d\n", Comm().MyPID(), MyRowStart_, MyRowEnd_);
bool SameVectors = false;
int NumVectors = X.NumVectors();
if (NumVectors != Y.NumVectors()) return -1; // X and Y must have same number of vectors
if(X.Pointers() == Y.Pointers()){
SameVectors = true;
}
for(int VecNum = 0; VecNum < NumVectors; VecNum++) {
//Get values for current vector in multivector.
double * x_values;
double * y_values;
EPETRA_CHK_ERR((*X(VecNum)).ExtractView(&x_values));
double *x_temp = x_local->data;
double *y_temp = y_local->data;
if(!SameVectors){
EPETRA_CHK_ERR((*Y(VecNum)).ExtractView(&y_values));
} else {
y_values = new double[X.MyLength()];
}
y_local->data = y_values;
EPETRA_CHK_ERR(HYPRE_ParVectorSetConstantValues(par_y,0.0));
// Temporarily make a pointer to data in Hypre for end
// Replace data in Hypre vectors with epetra values
x_local->data = x_values;
// Do actual computation.
if(TransA) {
// Use transpose of A in multiply
EPETRA_CHK_ERR(HYPRE_ParCSRMatrixMatvecT(1.0, ParMatrix_, par_x, 1.0, par_y));
} else {
EPETRA_CHK_ERR(HYPRE_ParCSRMatrixMatvec(1.0, ParMatrix_, par_x, 1.0, par_y));
}
if(SameVectors){
int NumEntries = Y.MyLength();
std::vector<double> new_values; new_values.resize(NumEntries);
std::vector<int> new_indices; new_indices.resize(NumEntries);
for(int i = 0; i < NumEntries; i++){
new_values[i] = y_values[i];
new_indices[i] = i;
}
EPETRA_CHK_ERR((*Y(VecNum)).ReplaceMyValues(NumEntries, &new_values[0], &new_indices[0]));
delete[] y_values;
}
x_local->data = x_temp;
y_local->data = y_temp;
}
double flops = (double) NumVectors * (double) NumGlobalNonzeros();
UpdateFlops(flops);
return 0;
} //Multiply()
//==============================================================================
int Ifpack_Hypre::Multiply(bool TransA, const Epetra_MultiVector& X, Epetra_MultiVector& Y) const{
if(IsInitialized() == false){
IFPACK_CHK_ERR(-1);
}
bool SameVectors = false;
int NumVectors = X.NumVectors();
if (NumVectors != Y.NumVectors()) IFPACK_CHK_ERR(-1); // X and Y must have same number of vectors
if(X.Pointers() == Y.Pointers()){
SameVectors = true;
}
for(int VecNum = 0; VecNum < NumVectors; VecNum++) {
//Get values for current vector in multivector.
double * XValues;
double * YValues;
IFPACK_CHK_ERR((*X(VecNum)).ExtractView(&XValues));
double *XTemp = XLocal_->data;
double *YTemp = YLocal_->data;
if(!SameVectors){
IFPACK_CHK_ERR((*Y(VecNum)).ExtractView(&YValues));
} else {
YValues = new double[X.MyLength()];
}
YLocal_->data = YValues;
IFPACK_CHK_ERR(HYPRE_ParVectorSetConstantValues(ParY_,0.0));
// Temporarily make a pointer to data in Hypre for end
// Replace data in Hypre vectors with epetra values
XLocal_->data = XValues;
// Do actual computation.
if(TransA) {
// Use transpose of A in multiply
IFPACK_CHK_ERR(HYPRE_ParCSRMatrixMatvecT(1.0, ParMatrix_, ParX_, 1.0, ParY_));
} else {
IFPACK_CHK_ERR(HYPRE_ParCSRMatrixMatvec(1.0, ParMatrix_, ParX_, 1.0, ParY_));
}
if(SameVectors){
int NumEntries = Y.MyLength();
std::vector<double> new_values; new_values.resize(NumEntries);
std::vector<int> new_indices; new_indices.resize(NumEntries);
for(int i = 0; i < NumEntries; i++){
new_values[i] = YValues[i];
new_indices[i] = i;
}
IFPACK_CHK_ERR((*Y(VecNum)).ReplaceMyValues(NumEntries, &new_values[0], &new_indices[0]));
delete[] YValues;
}
XLocal_->data = XTemp;
YLocal_->data = YTemp;
}
return 0;
} //Multiply()
int Ifpack_IHSS::ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const{
if(!IsComputed_) return -1;
Time_.ResetStartTime();
bool initial_guess_is_zero=false;
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
Teuchos::RefCountPtr<const Epetra_MultiVector> Xcopy;
Epetra_MultiVector Temp(X);
if (X.Pointers()[0] == Y.Pointers()[0]){
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
// Since the user didn't give us anything better, our initial guess is zero.
Y.Scale(0.0);
initial_guess_is_zero=true;
}
else
Xcopy = Teuchos::rcp( &X, false );
Epetra_MultiVector T1(Y),T2(*Xcopy);
// Note: Since Aherm and Askew are actually (aI+H) and (aI+S) accordingly (to save memory),
// the application thereof needs to be a little different.
// The published algorithm is:
// temp = (aI+H)^{-1} [ (aI-S) y + x ]
// y = (aI+S)^{-1} [ (aI-H) temp + x ]
//
// But we're doing:
// temp = (aI+H)^{-1} [ 2a y - Shat y + x ]
// y = (aI+S)^{-1} [ 2 a temp - Hhat temp + x ]
for(int i=0;i<NumSweeps_;i++){
// temp = (aI+H)^{-1} [ 2a y - Shat y + x ]
if(!initial_guess_is_zero || i >0 ){
Askew_->Apply(Y,T1);
T2.Update(2*Alpha_,Y,-1,T1,1);
}
Pherm_->ApplyInverse(T2,Y);
// y = (aI+S)^{-1} [ 2 a temp - Hhat temp + x ]
Aherm_->Apply(Y,T1);
T2.Scale(1.0,*Xcopy);
T2.Update(2*Alpha_,Y,-1,T1,1.0);
Pskew_->ApplyInverse(T2,Y);
}
// Counter update
NumApplyInverse_++;
ApplyInverseTime_ += Time_.ElapsedTime();
return 0;
}
//==============================================================================
// Note that Ifpack_PointRelaxation and Jacobi is much faster than
// Ifpack_AdditiveSchwarz<Ifpack_PointRelaxation> (because of the
// way the matrix-vector produce is performed).
//
// Another ML-related observation is that the starting solution (in Y)
// is NOT supposed to be zero. This may slow down the application of just
// one sweep of Jacobi.
//
int Ifpack_PointRelaxation::
ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const
{
if (!IsComputed())
IFPACK_CHK_ERR(-3);
if (X.NumVectors() != Y.NumVectors())
IFPACK_CHK_ERR(-2);
Time_->ResetStartTime();
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
Teuchos::RefCountPtr< const Epetra_MultiVector > Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0])
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
else
Xcopy = Teuchos::rcp( &X, false );
if (ZeroStartingSolution_)
Y.PutScalar(0.0);
// Flops are updated in each of the following.
switch (PrecType_) {
case IFPACK_JACOBI:
IFPACK_CHK_ERR(ApplyInverseJacobi(*Xcopy,Y));
break;
case IFPACK_GS:
IFPACK_CHK_ERR(ApplyInverseGS(*Xcopy,Y));
break;
case IFPACK_SGS:
IFPACK_CHK_ERR(ApplyInverseSGS(*Xcopy,Y));
break;
default:
IFPACK_CHK_ERR(-1); // something wrong
}
++NumApplyInverse_;
ApplyInverseTime_ += Time_->ElapsedTime();
return(0);
}
//=============================================================================
// This function finds Y such that LDU Y = X or U(trans) D L(trans) Y = X for multiple RHS
int Ifpack_IC::ApplyInverse(const Epetra_MultiVector& X,
Epetra_MultiVector& Y) const
{
if (!IsComputed())
IFPACK_CHK_ERR(-3); // compute preconditioner first
if (X.NumVectors() != Y.NumVectors())
IFPACK_CHK_ERR(-2); // Return error: X and Y not the same size
Time_.ResetStartTime();
bool Upper = true;
bool UnitDiagonal = true;
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
RefCountPtr< const Epetra_MultiVector > Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0])
Xcopy = rcp( new Epetra_MultiVector(X) );
else
Xcopy = rcp( &X, false );
U_->Solve(Upper, true, UnitDiagonal, *Xcopy, Y);
Y.Multiply(1.0, *D_, Y, 0.0); // y = D*y (D_ has inverse of diagonal)
U_->Solve(Upper, false, UnitDiagonal, Y, Y); // Solve Uy = y
#ifdef IFPACK_FLOPCOUNTERS
ApplyInverseFlops_ += 4.0 * U_->NumGlobalNonzeros();
ApplyInverseFlops_ += D_->GlobalLength();
#endif
++NumApplyInverse_;
ApplyInverseTime_ += Time_.ElapsedTime();
return(0);
}
//=============================================================================
int Ifpack_ICT::ApplyInverse(const Epetra_MultiVector& X,
Epetra_MultiVector& Y) const
{
if (!IsComputed())
IFPACK_CHK_ERR(-3); // compute preconditioner first
if (X.NumVectors() != Y.NumVectors())
IFPACK_CHK_ERR(-2); // Return error: X and Y not the same size
Time_.ResetStartTime();
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
Teuchos::RefCountPtr<const Epetra_MultiVector> Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0])
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
else
Xcopy = Teuchos::rcp( &X, false );
// NOTE: H_ is based on SerialMap_, while Xcopy is based
// on A.Map()... which are in general different. However, Solve()
// does not seem to care... which is fine with me.
//
EPETRA_CHK_ERR(H_->Solve(false,false,false,*Xcopy,Y));
EPETRA_CHK_ERR(H_->Solve(false,true,false,Y,Y));
#ifdef IFPACK_FLOPCOUNTERS
// these are global flop count
ApplyInverseFlops_ += 4.0 * GlobalNonzeros_;
#endif
++NumApplyInverse_;
ApplyInverseTime_ += Time_.ElapsedTime();
return(0);
}
//==============================================================================
int Ifpack_Hypre::ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const{
if(IsComputed() == false){
IFPACK_CHK_ERR(-1);
}
// These are hypre requirements
// hypre needs A, X, and Y to have the same contiguous distribution
// NOTE: Maps are only considered to be contiguous if they were generated using a
// particular constructor. Otherwise, LinearMap() will not detect whether they are
// actually contiguous.
if(!X.Map().LinearMap() || !Y.Map().LinearMap()) {
std::cerr << "ERROR: X and Y must have contiguous maps.\n";
IFPACK_CHK_ERR(-1);
}
if(!X.Map().PointSameAs(*MySimpleMap_) ||
!Y.Map().PointSameAs(*MySimpleMap_)) {
std::cerr << "ERROR: X, Y, and A must have the same distribution.\n";
IFPACK_CHK_ERR(-1);
}
Time_.ResetStartTime();
bool SameVectors = false;
int NumVectors = X.NumVectors();
if (NumVectors != Y.NumVectors()) IFPACK_CHK_ERR(-1); // X and Y must have same number of vectors
if(X.Pointers() == Y.Pointers()){
SameVectors = true;
}
for(int VecNum = 0; VecNum < NumVectors; VecNum++) {
//Get values for current vector in multivector.
// FIXME amk Nov 23, 2015: This will not work for funky data layouts
double * XValues;
IFPACK_CHK_ERR((*X(VecNum)).ExtractView(&XValues));
double * YValues;
if(!SameVectors){
IFPACK_CHK_ERR((*Y(VecNum)).ExtractView(&YValues));
} else {
YValues = new double[X.MyLength()];
}
// Temporarily make a pointer to data in Hypre for end
double *XTemp = XLocal_->data;
// Replace data in Hypre vectors with epetra values
XLocal_->data = XValues;
double *YTemp = YLocal_->data;
YLocal_->data = YValues;
IFPACK_CHK_ERR(HYPRE_ParVectorSetConstantValues(ParY_, 0.0));
if(SolveOrPrec_ == Solver){
// Use the solver methods
IFPACK_CHK_ERR(SolverSolvePtr_(Solver_, ParMatrix_, ParX_, ParY_));
} else {
// Apply the preconditioner
IFPACK_CHK_ERR(PrecondSolvePtr_(Preconditioner_, ParMatrix_, ParX_, ParY_));
}
if(SameVectors){
int NumEntries = Y.MyLength();
std::vector<double> new_values; new_values.resize(NumEntries);
std::vector<int> new_indices; new_indices.resize(NumEntries);
for(int i = 0; i < NumEntries; i++){
new_values[i] = YValues[i];
new_indices[i] = i;
}
IFPACK_CHK_ERR((*Y(VecNum)).ReplaceMyValues(NumEntries, &new_values[0], &new_indices[0]));
delete[] YValues;
}
XLocal_->data = XTemp;
YLocal_->data = YTemp;
}
NumApplyInverse_ = NumApplyInverse_ + 1;
ApplyInverseTime_ = ApplyInverseTime_ + Time_.ElapsedTime();
return 0;
} //ApplyInverse()
int Ifpack_SORa::ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const{
if(!IsComputed_) return -1;
Time_.ResetStartTime();
bool initial_guess_is_zero=false;
const int lclNumRows = W_->NumMyRows();
const int NumVectors = X.NumVectors();
Epetra_MultiVector Temp(A_->RowMatrixRowMap(),NumVectors);
double omega=GetOmega();
// need to create an auxiliary vector, Xcopy
Teuchos::RCP<const Epetra_MultiVector> Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0]){
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
// Since the user didn't give us anything better, our initial guess is zero.
Y.Scale(0.0);
initial_guess_is_zero=true;
}
else
Xcopy = Teuchos::rcp( &X, false );
Teuchos::RCP< Epetra_MultiVector > T2;
// Note: Assuming that the matrix has an importer. Ifpack_PointRelaxation doesn't do this, but given that
// I have a CrsMatrix, I'm probably OK.
// Note: This is the lazy man's version sacrificing a few extra flops for avoiding if statements to determine
// if things are on or off processor.
// Note: T2 must be zero'd out
if (IsParallel_ && W_->Importer()) T2 = Teuchos::rcp( new Epetra_MultiVector(W_->Importer()->TargetMap(),NumVectors,true));
else T2 = Teuchos::rcp( new Epetra_MultiVector(A_->RowMatrixRowMap(),NumVectors,true));
// Pointer grabs
int* rowptr,*colind;
double *values;
double **t_ptr,** y_ptr, ** t2_ptr, **x_ptr,*d_ptr;
T2->ExtractView(&t2_ptr);
Y.ExtractView(&y_ptr);
Temp.ExtractView(&t_ptr);
Xcopy->ExtractView(&x_ptr);
Wdiag_->ExtractView(&d_ptr);
IFPACK_CHK_ERR(W_->ExtractCrsDataPointers(rowptr,colind,values));
for(int i=0; i<NumSweeps_; i++){
// Calculate b-Ax
if(!initial_guess_is_zero || i > 0) {
A_->Apply(Y,Temp);
Temp.Update(1.0,*Xcopy,-1.0);
}
else
Temp.Update(1.0,*Xcopy,0.0);
// Note: The off-processor entries of T2 never get touched (they're always zero) and the other entries are updated
// in this sweep before they are used, so we don't need to reset T2 to zero here.
// Do backsolve & update
// x = x + W^{-1} (b - A x)
for(int j=0; j<lclNumRows; j++){
double diag=d_ptr[j];
for (int m=0 ; m<NumVectors; m++) {
double dtmp=0.0;
// Note: Since the diagonal is in the matrix, we need to zero that entry of T2 here to make sure it doesn't contribute.
t2_ptr[m][j]=0.0;
for(int k=rowptr[j];k<rowptr[j+1];k++){
dtmp+= values[k]*t2_ptr[m][colind[k]];
}
// Yes, we need to update both of these.
t2_ptr[m][j] = (t_ptr[m][j]- dtmp)/diag;
y_ptr[m][j] += omega*t2_ptr[m][j];
}
}
}
// Counter update
NumApplyInverse_++;
ApplyInverseTime_ += Time_.ElapsedTime();
return 0;
}
// ================================================ ====== ==== ==== == =
//! Apply the preconditioner to an Epetra_MultiVector X, puts the result in Y
int ML_Epetra::FaceMatrixFreePreconditioner::ApplyInverse(const Epetra_MultiVector& B_, Epetra_MultiVector& X) const{
const Epetra_MultiVector *B;
Epetra_MultiVector *Bcopy=0;
/* Sanity Checks */
int NumVectors=B_.NumVectors();
if (!B_.Map().SameAs(*FaceDomainMap_)) ML_CHK_ERR(-1);
if (NumVectors != X.NumVectors()) ML_CHK_ERR(-1);
Epetra_MultiVector r_edge(*FaceDomainMap_,NumVectors,false);
Epetra_MultiVector e_edge(*FaceDomainMap_,NumVectors,false);
Epetra_MultiVector e_node(*CoarseMap_,NumVectors,false);
Epetra_MultiVector r_node(*CoarseMap_,NumVectors,false);
/* Deal with the B==X case */
if (B_.Pointers()[0] == X.Pointers()[0]){
Bcopy=new Epetra_MultiVector(B_);
B=Bcopy;
X.PutScalar(0.0);
}
else B=&B_;
for(int i=0;i<num_cycles;i++){
/* Pre-smoothing */
#ifdef HAVE_ML_IFPACK
if(Smoother_) ML_CHK_ERR(Smoother_->ApplyInverse(*B,X));
#endif
if(MaxLevels > 0){
if(i != 0
#ifdef HAVE_ML_IFPACK
|| Smoother_
#endif
){
/* Calculate Residual (r_e = b - (S+M+Addon) * x) */
ML_CHK_ERR(Operator_->Apply(X,r_edge));
ML_CHK_ERR(r_edge.Update(1.0,*B,-1.0));
/* Xfer to coarse grid (r_n = P' * r_e) */
ML_CHK_ERR(Prolongator_->Multiply(true,r_edge,r_node));
}
else{
/* Xfer to coarse grid (r_n = P' * r_e) */
ML_CHK_ERR(Prolongator_->Multiply(true,*B,r_node));
}
/* AMG on coarse grid (e_n = (CoarseMatrix)^{-1} r_n) */
ML_CHK_ERR(CoarsePC->ApplyInverse(r_node,e_node));
/* Xfer back to fine grid (e_e = P * e_n) */
ML_CHK_ERR(Prolongator_->Multiply(false,e_node,e_edge));
/* Add in correction (x = x + e_e) */
ML_CHK_ERR(X.Update(1.0,e_edge,1.0));
}/*end if*/
/* Post-Smoothing */
#ifdef HAVE_ML_IFPACK
if(Smoother_) ML_CHK_ERR(Smoother_->ApplyInverse(*B,X));
#endif
}/*end for*/
/* Cleanup */
if(Bcopy) delete Bcopy;
return 0;
}/*end ApplyInverse*/
//=============================================================================
int Epetra_FastCrsMatrix::Solve(bool Upper, bool Trans, bool UnitDiagonal, const Epetra_MultiVector& X, Epetra_MultiVector& Y) const {
//
// This function find Y such that LY = X or UY = X or the transpose cases.
//
if (X.NumVectors()==1 && Y.NumVectors()==1) {
double * xp = (double *) X[0];
double * yp = (double *) Y[0];
Epetra_Vector x(View, X.Map(), xp);
Epetra_Vector y(View, Y.Map(), yp);
return(Solve(Upper, Trans, UnitDiagonal, x, y));
}
if (!Filled()) EPETRA_CHK_ERR(-1); // Matrix must be filled.
if ((Upper) && (!UpperTriangular())) EPETRA_CHK_ERR(-2);
if ((!Upper) && (!LowerTriangular())) EPETRA_CHK_ERR(-3);
if ((!UnitDiagonal) && (NoDiagonal())) EPETRA_CHK_ERR(-4); // If matrix has no diagonal, we must use UnitDiagonal
if ((!UnitDiagonal) && (NumMyDiagonals()<NumMyRows_)) EPETRA_CHK_ERR(-5); // Need each row to have a diagonal
int i, j, j0, k;
int * NumEntriesPerRow = NumEntriesPerRow_;
int ** Indices = Indices_;
double ** Values = Values_;
double diag;
// If upper, point to last row
if ((Upper && !Trans) || (!Upper && Trans)) {
NumEntriesPerRow += NumMyRows_-1;
Indices += NumMyRows_-1;
Values += NumMyRows_-1;
}
double **Xp = (double**)X.Pointers();
double **Yp = (double**)Y.Pointers();
int NumVectors = X.NumVectors();
if (!Trans) {
if (Upper) {
j0 = 1;
if (NoDiagonal()) j0--; // Include first term if no diagonal
for (i=NumMyRows_-1; i >=0; i--) {
int NumEntries = *NumEntriesPerRow--;
int * RowIndices = *Indices--;
double * RowValues = *Values--;
if (!UnitDiagonal) diag = 1.0/RowValues[0]; // Take inverse of diagonal once for later use
for (k=0; k<NumVectors; k++) {
double sum = 0.0;
for (j=j0; j < NumEntries; j++) sum += RowValues[j] * Yp[k][RowIndices[j]];
if (UnitDiagonal) Yp[k][i] = Xp[k][i] - sum;
else Yp[k][i] = (Xp[k][i] - sum)*diag;
}
}
}
else {
j0 = 1;
if (NoDiagonal()) j0--; // Include first term if no diagonal
for (i=0; i < NumMyRows_; i++) {
int NumEntries = *NumEntriesPerRow++ - j0;
int * RowIndices = *Indices++;
double * RowValues = *Values++;
if (!UnitDiagonal) diag = 1.0/RowValues[NumEntries]; // Take inverse of diagonal once for later use
for (k=0; k<NumVectors; k++) {
double sum = 0.0;
for (j=0; j < NumEntries; j++) sum += RowValues[j] * Yp[k][RowIndices[j]];
if (UnitDiagonal) Yp[k][i] = Xp[k][i] - sum;
else Yp[k][i] = (Xp[k][i] - sum)*diag;
}
}
}
}
// *********** Transpose case *******************************
else {
for (k=0; k<NumVectors; k++)
if (Yp[k]!=Xp[k]) for (i=0; i < NumMyRows_; i++) Yp[k][i] = Xp[k][i]; // Initialize y for transpose multiply
if (Upper) {
j0 = 1;
if (NoDiagonal()) j0--; // Include first term if no diagonal
for (i=0; i < NumMyRows_; i++) {
int NumEntries = *NumEntriesPerRow++;
int * RowIndices = *Indices++;
double * RowValues = *Values++;
if (!UnitDiagonal) diag = 1.0/RowValues[j0]; // Take inverse of diagonal once for later use
for (k=0; k<NumVectors; k++) {
if (!UnitDiagonal) Yp[k][i] = Yp[k][i]*diag;
for (j=j0; j < NumEntries; j++) Yp[k][RowIndices[j]] -= RowValues[j] * Yp[k][i];
}
}
}
else {
j0 = 1;
if (NoDiagonal()) j0--; // Include first term if no diagonal
for (i=NumMyRows_-1; i>=0; i--) {
int NumEntries = *NumEntriesPerRow-- - j0;
int * RowIndices = *Indices--;
double * RowValues = *Values--;
for (k=0; k<NumVectors; k++) {
if (!UnitDiagonal) Yp[k][i] = Yp[k][i]/Xp[k][i];
for (j=0; j < NumEntries; j++) Yp[k][RowIndices[j]] -= RowValues[j] * Yp[k][i];
}
}
}
}
UpdateFlops(2*NumVectors*NumGlobalNonzeros64());
return(0);
}
//=============================================================================
int Epetra_FastCrsMatrix::Multiply(bool TransA, const Epetra_MultiVector& X, Epetra_MultiVector& Y) const {
//
// This function forms the product Y = A * Y or Y = A' * X
//
if (X.NumVectors()==1 && Y.NumVectors()==1) {
double * xp = (double *) X[0];
double * yp = (double *) Y[0];
Epetra_Vector x(View, X.Map(), xp);
Epetra_Vector y(View, Y.Map(), yp);
return(Multiply(TransA, x, y));
}
if (!Filled()) EPETRA_CHK_ERR(-1); // Matrix must be filled.
int i, j, k;
int * NumEntriesPerRow = NumEntriesPerRow_;
int ** Indices = Indices_;
double ** Values = Values_;
double **Xp = (double**)X.Pointers();
double **Yp = (double**)Y.Pointers();
int NumVectors = X.NumVectors();
int NumMyCols_ = NumMyCols();
// Need to better manage the Import and Export vectors:
// - Need accessor functions
// - Need to make the NumVector match (use a View to do this)
// - Need to look at RightScale and ColSum routines too.
if (!TransA) {
// If we have a non-trivial importer, we must import elements that are permuted or are on other processors
if (Importer()!=0) {
if (ImportVector_!=0) {
if (ImportVector_->NumVectors()!=NumVectors) { delete ImportVector_; ImportVector_= 0;}
}
if (ImportVector_==0) ImportVector_ = new Epetra_MultiVector(ColMap(),NumVectors); // Create import vector if needed
ImportVector_->Import(X, *Importer(), Insert);
Xp = (double**)ImportVector_->Pointers();
}
// If we have a non-trivial exporter, we must export elements that are permuted or belong to other processors
if (Exporter()!=0) {
if (ExportVector_!=0) {
if (ExportVector_->NumVectors()!=NumVectors) { delete ExportVector_; ExportVector_= 0;}
}
if (ExportVector_==0) ExportVector_ = new Epetra_MultiVector(RowMap(),NumVectors); // Create Export vector if needed
Yp = (double**)ExportVector_->Pointers();
}
// Do actual computation
for (i=0; i < NumMyRows_; i++) {
int NumEntries = *NumEntriesPerRow++;
int * RowIndices = *Indices++;
double * RowValues = *Values++;
for (k=0; k<NumVectors; k++) {
double sum = 0.0;
for (j=0; j < NumEntries; j++) sum += RowValues[j] * Xp[k][RowIndices[j]];
Yp[k][i] = sum;
}
}
if (Exporter()!=0) Y.Export(*ExportVector_, *Exporter(), Add); // Fill Y with Values from export vector
}
else { // Transpose operation
// If we have a non-trivial exporter, we must import elements that are permuted or are on other processors
if (Exporter()!=0) {
if (ExportVector_!=0) {
if (ExportVector_->NumVectors()!=NumVectors) { delete ExportVector_; ExportVector_= 0;}
}
if (ExportVector_==0) ExportVector_ = new Epetra_MultiVector(RowMap(),NumVectors); // Create Export vector if needed
ExportVector_->Import(X, *Exporter(), Insert);
Xp = (double**)ExportVector_->Pointers();
}
// If we have a non-trivial importer, we must export elements that are permuted or belong to other processors
if (Importer()!=0) {
if (ImportVector_!=0) {
if (ImportVector_->NumVectors()!=NumVectors) { delete ImportVector_; ImportVector_= 0;}
}
if (ImportVector_==0) ImportVector_ = new Epetra_MultiVector(ColMap(),NumVectors); // Create import vector if needed
Yp = (double**)ImportVector_->Pointers();
}
// Do actual computation
for (k=0; k<NumVectors; k++)
for (i=0; i < NumMyCols_; i++) Yp[k][i] = 0.0; // Initialize y for transpose multiply
for (i=0; i < NumMyRows_; i++) {
int NumEntries = *NumEntriesPerRow++;
int * RowIndices = *Indices++;
double * RowValues = *Values++;
for (k=0; k<NumVectors; k++) {
for (j=0; j < NumEntries; j++) Yp[k][RowIndices[j]] += RowValues[j] * Xp[k][i];
}
}
if (Importer()!=0) Y.Export(*ImportVector_, *Importer(), Add); // Fill Y with Values from export vector
}
UpdateFlops(2*NumVectors*NumGlobalNonzeros64());
return(0);
}
//==============================================================================
int Ifpack_Chebyshev::
ApplyInverse(const Epetra_MultiVector& X, Epetra_MultiVector& Y) const
{
if (!IsComputed())
IFPACK_CHK_ERR(-3);
if (PolyDegree_ == 0)
return 0;
int nVec = X.NumVectors();
int len = X.MyLength();
if (nVec != Y.NumVectors())
IFPACK_CHK_ERR(-2);
Time_->ResetStartTime();
// AztecOO gives X and Y pointing to the same memory location,
// need to create an auxiliary vector, Xcopy
Teuchos::RefCountPtr<const Epetra_MultiVector> Xcopy;
if (X.Pointers()[0] == Y.Pointers()[0])
Xcopy = Teuchos::rcp( new Epetra_MultiVector(X) );
else
Xcopy = Teuchos::rcp( &X, false );
double **xPtr = 0, **yPtr = 0;
Xcopy->ExtractView(&xPtr);
Y.ExtractView(&yPtr);
#ifdef HAVE_IFPACK_EPETRAEXT
EpetraExt_PointToBlockDiagPermute* IBD=0;
if (UseBlockMode_) IBD=&*InvBlockDiagonal_;
#endif
//--- Do a quick solve when the matrix is identity
double *invDiag=0;
if(!UseBlockMode_) invDiag=InvDiagonal_->Values();
if ((LambdaMin_ == 1.0) && (LambdaMax_ == LambdaMin_)) {
#ifdef HAVE_IFPACK_EPETRAEXT
if(UseBlockMode_) IBD->ApplyInverse(*Xcopy,Y);
else
#endif
if (nVec == 1) {
double *yPointer = yPtr[0], *xPointer = xPtr[0];
for (int i = 0; i < len; ++i)
yPointer[i] = xPointer[i]*invDiag[i];
}
else {
int i, k;
for (i = 0; i < len; ++i) {
double coeff = invDiag[i];
for (k = 0; k < nVec; ++k)
yPtr[k][i] = xPtr[k][i] * coeff;
}
} // if (nVec == 1)
return 0;
} // if ((LambdaMin_ == 1.0) && (LambdaMax_ == LambdaMin_))
//--- Initialize coefficients
// Note that delta stores the inverse of ML_Cheby::delta
double alpha = LambdaMax_ / EigRatio_;
double beta = 1.1 * LambdaMax_;
double delta = 2.0 / (beta - alpha);
double theta = 0.5 * (beta + alpha);
double s1 = theta * delta;
//--- Define vectors
// In ML_Cheby, V corresponds to pAux and W to dk
Epetra_MultiVector V(X);
Epetra_MultiVector W(X);
#ifdef HAVE_IFPACK_EPETRAEXT
Epetra_MultiVector Temp(X);
#endif
double *vPointer = V.Values(), *wPointer = W.Values();
double oneOverTheta = 1.0/theta;
int i, j, k;
//--- If solving normal equations, multiply RHS by A^T
if(SolveNormalEquations_){
Apply_Transpose(Operator_,Y,V);
Y=V;
}
// Do the smoothing when block scaling is turned OFF
// --- Treat the initial guess
if (ZeroStartingSolution_ == false) {
Operator_->Apply(Y, V);
// Compute W = invDiag * ( X - V )/ Theta
#ifdef HAVE_IFPACK_EPETRAEXT
if(UseBlockMode_) {
Temp.Update(oneOverTheta,X,-oneOverTheta,V,0.0);
IBD->ApplyInverse(Temp,W);
// Perform additional matvecs for normal equations
// CMS: Testing this only in block mode FOR NOW
if(SolveNormalEquations_){
IBD->ApplyInverse(W,Temp);
Apply_Transpose(Operator_,Temp,W);
}
}
else
#endif
if (nVec == 1) {
double *xPointer = xPtr[0];
for (i = 0; i < len; ++i)
wPointer[i] = invDiag[i] * (xPointer[i] - vPointer[i]) * oneOverTheta;
}
else {
for (i = 0; i < len; ++i) {
double coeff = invDiag[i]*oneOverTheta;
double *wi = wPointer + i, *vi = vPointer + i;
for (k = 0; k < nVec; ++k) {
*wi = (xPtr[k][i] - (*vi)) * coeff;
wi = wi + len; vi = vi + len;
}
}
} // if (nVec == 1)
// Update the vector Y
Y.Update(1.0, W, 1.0);
}
else {
// Compute W = invDiag * X / Theta
#ifdef HAVE_IFPACK_EPETRAEXT
if(UseBlockMode_) {
IBD->ApplyInverse(X,W);
// Perform additional matvecs for normal equations
// CMS: Testing this only in block mode FOR NOW
if(SolveNormalEquations_){
IBD->ApplyInverse(W,Temp);
Apply_Transpose(Operator_,Temp,W);
}
W.Scale(oneOverTheta);
Y.Update(1.0, W, 0.0);
}
else
#endif
if (nVec == 1) {
double *xPointer = xPtr[0];
for (i = 0; i < len; ++i){
wPointer[i] = invDiag[i] * xPointer[i] * oneOverTheta;
}
memcpy(yPtr[0], wPointer, len*sizeof(double));
}
else {
for (i = 0; i < len; ++i) {
double coeff = invDiag[i]*oneOverTheta;
double *wi = wPointer + i;
for (k = 0; k < nVec; ++k) {
*wi = xPtr[k][i] * coeff;
wi = wi + len;
}
}
for (k = 0; k < nVec; ++k)
memcpy(yPtr[k], wPointer + k*len, len*sizeof(double));
} // if (nVec == 1)
} // if (ZeroStartingSolution_ == false)
//--- Apply the polynomial
double rhok = 1.0/s1, rhokp1;
double dtemp1, dtemp2;
int degreeMinusOne = PolyDegree_ - 1;
if (nVec == 1) {
double *xPointer = xPtr[0];
for (k = 0; k < degreeMinusOne; ++k) {
Operator_->Apply(Y, V);
rhokp1 = 1.0 / (2.0*s1 - rhok);
dtemp1 = rhokp1 * rhok;
dtemp2 = 2.0 * rhokp1 * delta;
rhok = rhokp1;
// Compute W = dtemp1 * W
W.Scale(dtemp1);
// Compute W = W + dtemp2 * invDiag * ( X - V )
#ifdef HAVE_IFPACK_EPETRAEXT
if(UseBlockMode_) {
//NTS: We can clobber V since it will be reset in the Apply
V.Update(dtemp2,X,-dtemp2);
IBD->ApplyInverse(V,Temp);
// Perform additional matvecs for normal equations
// CMS: Testing this only in block mode FOR NOW
if(SolveNormalEquations_){
IBD->ApplyInverse(V,Temp);
Apply_Transpose(Operator_,Temp,V);
}
W.Update(1.0,Temp,1.0);
}
else{
#endif
for (i = 0; i < len; ++i)
wPointer[i] += dtemp2* invDiag[i] * (xPointer[i] - vPointer[i]);
#ifdef HAVE_IFPACK_EPETRAEXT
}
#endif
// Update the vector Y
Y.Update(1.0, W, 1.0);
} // for (k = 0; k < degreeMinusOne; ++k)
}
else {
for (k = 0; k < degreeMinusOne; ++k) {
Operator_->Apply(Y, V);
rhokp1 = 1.0 / (2.0*s1 - rhok);
dtemp1 = rhokp1 * rhok;
dtemp2 = 2.0 * rhokp1 * delta;
rhok = rhokp1;
// Compute W = dtemp1 * W
W.Scale(dtemp1);
// Compute W = W + dtemp2 * invDiag * ( X - V )
#ifdef HAVE_IFPACK_EPETRAEXT
if(UseBlockMode_) {
//We can clobber V since it will be reset in the Apply
V.Update(dtemp2,X,-dtemp2);
IBD->ApplyInverse(V,Temp);
// Perform additional matvecs for normal equations
// CMS: Testing this only in block mode FOR NOW
if(SolveNormalEquations_){
IBD->ApplyInverse(V,Temp);
Apply_Transpose(Operator_,Temp,V);
}
W.Update(1.0,Temp,1.0);
}
else{
#endif
for (i = 0; i < len; ++i) {
double coeff = invDiag[i]*dtemp2;
double *wi = wPointer + i, *vi = vPointer + i;
for (j = 0; j < nVec; ++j) {
*wi += (xPtr[j][i] - (*vi)) * coeff;
wi = wi + len; vi = vi + len;
}
}
#ifdef HAVE_IFPACK_EPETRAEXT
}
#endif
// Update the vector Y
Y.Update(1.0, W, 1.0);
} // for (k = 0; k < degreeMinusOne; ++k)
} // if (nVec == 1)
// Flops are updated in each of the following.
++NumApplyInverse_;
ApplyInverseTime_ += Time_->ElapsedTime();
return(0);
}
//=======================================================
int EpetraExt_HypreIJMatrix::Solve(bool Upper, bool transpose, bool UnitDiagonal, const Epetra_MultiVector & X, Epetra_MultiVector & Y) const {
bool SameVectors = false;
int NumVectors = X.NumVectors();
if (NumVectors != Y.NumVectors()) return -1; // X and Y must have same number of vectors
if(X.Pointers() == Y.Pointers()){
SameVectors = true;
}
if(SolveOrPrec_ == Solver){
if(IsSolverSetup_[0] == false){
SetupSolver();
}
} else {
if(IsPrecondSetup_[0] == false){
SetupPrecond();
}
}
for(int VecNum = 0; VecNum < NumVectors; VecNum++) {
//Get values for current vector in multivector.
double * x_values;
EPETRA_CHK_ERR((*X(VecNum)).ExtractView(&x_values));
double * y_values;
if(!SameVectors){
EPETRA_CHK_ERR((*Y(VecNum)).ExtractView(&y_values));
} else {
y_values = new double[X.MyLength()];
}
// Temporarily make a pointer to data in Hypre for end
double *x_temp = x_local->data;
// Replace data in Hypre vectors with epetra values
x_local->data = x_values;
double *y_temp = y_local->data;
y_local->data = y_values;
EPETRA_CHK_ERR(HYPRE_ParVectorSetConstantValues(par_y, 0.0));
if(transpose && !TransposeSolve_){
// User requested a transpose solve, but the solver selected doesn't provide one
EPETRA_CHK_ERR(-1);
}
if(SolveOrPrec_ == Solver){
// Use the solver methods
EPETRA_CHK_ERR(SolverSolvePtr_(Solver_, ParMatrix_, par_x, par_y));
} else {
// Apply the preconditioner
EPETRA_CHK_ERR(PrecondSolvePtr_(Preconditioner_, ParMatrix_, par_x, par_y));
}
if(SameVectors){
int NumEntries = Y.MyLength();
std::vector<double> new_values; new_values.resize(NumEntries);
std::vector<int> new_indices; new_indices.resize(NumEntries);
for(int i = 0; i < NumEntries; i++){
new_values[i] = y_values[i];
new_indices[i] = i;
}
EPETRA_CHK_ERR((*Y(VecNum)).ReplaceMyValues(NumEntries, &new_values[0], &new_indices[0]));
delete[] y_values;
}
x_local->data = x_temp;
y_local->data = y_temp;
}
double flops = (double) NumVectors * (double) NumGlobalNonzeros();
UpdateFlops(flops);
return 0;
} //Solve()
int CopyMultiVector(double** matlabApr, const Epetra_MultiVector& A) {
Epetra_BlockMap bmap = A.Map();
const Epetra_Comm & comm = bmap.Comm();
int numProc = comm.NumProc();
if (numProc==1)
DoCopyMultiVector(matlabApr, A);
else {
// In the case of more than one column in the multivector, and writing to MatrixMarket
// format, we call this function recursively, passing each vector of the multivector
// individually so that we can get all of it written to file before going on to the next
// multivector
if (A.NumVectors() > 1) {
for (int i=0; i < A.NumVectors(); i++)
if (CopyMultiVector(matlabApr, *(A(i)))) return(-1);
return(0);
}
Epetra_Map map(-1, bmap.NumMyPoints(), 0, comm);
// Create a veiw of this multivector using a map (instead of block map)
Epetra_MultiVector A1(View, map, A.Pointers(), A.NumVectors());
int numRows = map.NumMyElements();
Epetra_Map allGidsMap(-1, numRows, 0,comm);
Epetra_IntVector allGids(allGidsMap);
for (int i=0; i<numRows; i++) allGids[i] = map.GID(i);
// Now construct a MultiVector on PE 0 by strip-mining the rows of the input matrix A.
int numChunks = numProc;
int stripSize = allGids.GlobalLength()/numChunks;
int remainder = allGids.GlobalLength()%numChunks;
int curStart = 0;
int curStripSize = 0;
Epetra_IntSerialDenseVector importGidList;
int numImportGids = 0;
if (comm.MyPID()==0)
importGidList.Size(stripSize+1); // Set size of vector to max needed
for (int i=0; i<numChunks; i++) {
if (comm.MyPID()==0) { // Only PE 0 does this part
curStripSize = stripSize;
if (i<remainder) curStripSize++; // handle leftovers
for (int j=0; j<curStripSize; j++) importGidList[j] = j + curStart;
curStart += curStripSize;
}
// The following import map will be non-trivial only on PE 0.
Epetra_Map importGidMap(-1, curStripSize, importGidList.Values(), 0, comm);
Epetra_Import gidImporter(importGidMap, allGidsMap);
Epetra_IntVector importGids(importGidMap);
if (importGids.Import(allGids, gidImporter, Insert)) return(-1);
// importGids now has a list of GIDs for the current strip of matrix rows.
// Use these values to build another importer that will get rows of the matrix.
// The following import map will be non-trivial only on PE 0.
Epetra_Map importMap(-1, importGids.MyLength(), importGids.Values(), 0, comm);
Epetra_Import importer(importMap, map);
Epetra_MultiVector importA(importMap, A1.NumVectors());
if (importA.Import(A1, importer, Insert)) return(-1);
// Finally we are ready to write this strip of the matrix to ostream
if (DoCopyMultiVector(matlabApr, importA)) return(-1);
}
}
return(0);
}
