dspgv function
void
dspgv()
Implementation
void dspgv(
final int ITYPE,
final String JOBZ,
final String UPLO,
final int N,
final Array<double> AP_,
final Array<double> BP_,
final Array<double> W_,
final Matrix<double> Z_,
final int LDZ,
final Array<double> WORK_,
final Box<int> INFO,
) {
final AP = AP_.having();
final BP = BP_.having();
final W = W_.having();
final Z = Z_.having(ld: LDZ);
final WORK = WORK_.having();
bool UPPER, WANTZ;
String TRANS;
int J, NEIG;
// Test the input parameters.
WANTZ = lsame(JOBZ, 'V');
UPPER = lsame(UPLO, 'U');
INFO.value = 0;
if (ITYPE < 1 || ITYPE > 3) {
INFO.value = -1;
} else if (!(WANTZ || lsame(JOBZ, 'N'))) {
INFO.value = -2;
} else if (!(UPPER || lsame(UPLO, 'L'))) {
INFO.value = -3;
} else if (N < 0) {
INFO.value = -4;
} else if (LDZ < 1 || (WANTZ && LDZ < N)) {
INFO.value = -9;
}
if (INFO.value != 0) {
xerbla('DSPGV', -INFO.value);
return;
}
// Quick return if possible
if (N == 0) return;
// Form a Cholesky factorization of B.
dpptrf(UPLO, N, BP, INFO);
if (INFO.value != 0) {
INFO.value = N + INFO.value;
return;
}
// Transform problem to standard eigenvalue problem and solve.
dspgst(ITYPE, UPLO, N, AP, BP, INFO);
dspev(JOBZ, UPLO, N, AP, W, Z, LDZ, WORK, INFO);
if (WANTZ) {
// Backtransform eigenvectors to the original problem.
NEIG = N;
if (INFO.value > 0) NEIG = INFO.value - 1;
if (ITYPE == 1 || ITYPE == 2) {
// For A*x=(lambda)*B*x and A*B*x=(lambda)*x;
// backtransform eigenvectors: x = inv(L)**T*y or inv(U)*y
if (UPPER) {
TRANS = 'N';
} else {
TRANS = 'T';
}
for (J = 1; J <= NEIG; J++) {
dtpsv(UPLO, TRANS, 'Non-unit', N, BP, Z(1, J).asArray(), 1);
}
} else if (ITYPE == 3) {
// For B*A*x=(lambda)*x;
// backtransform eigenvectors: x = L*y or U**T*y
if (UPPER) {
TRANS = 'T';
} else {
TRANS = 'N';
}
for (J = 1; J <= NEIG; J++) {
dtpmv(UPLO, TRANS, 'Non-unit', N, BP, Z(1, J).asArray(), 1);
}
}
}
}