Re: [Getfem-users] Question on Preconditioners

[Yves Renard]

Le Lundi 9 Octobre 2006 17:19, Bwalya Malama a écrit :
> > Sorry, the correct declaration is
> >
> > typedef gmm::row_matrix< gmm::wsvector<complex<double> > > mymatrixtype;
> > mymatrixtype C(N2,N2);
> > gmm::diagonal_precond<mymatrixtype>  P;
> > // ...
> > // ... fill C
> > // ...
> > P.build_with(C);
> >
> > i.e. no parentheses after the declaration of P.
>
> Hi Yves:
>
> That too did not work. When I compile, I get the following:
>
> /usr/local/include/gmm_vector.h: In member function ‘T
> gmm::wsvector<T>::r(typename std::map<int, T, std::less<int>,
> std::allocator<std::pair<const int, T> > >::size_type) const [with T =
> std::complex<double>]’:
> /usr/local/include/gmm_vector.h:220:   instantiated from ‘T
> gmm::wsvector<T>::operator[](typename std::map<int, T, std::less<int>,
> std::allocator<std::pair<const int, T> > >::size_type) const [with T =
> std::complex<double>]’
> /usr/local/include/gmm_matrix.h:142:   instantiated from ‘typename
> gmm::linalg_traits<V>::value_type
> gmm::row_matrix<VECT>::operator()(gmm::size_type, gmm::size_type) const
> [with V = gmm::wsvector<std::complex<double> >]’
> /usr/local/include/gmm_precond_diagonal.h:51:   instantiated from ‘void
> gmm::diagonal_precond<Matrix>::build_with(const Matrix&) [with Matrix =
> gmm::row_matrix<gmm::wsvector<std::complex<double> > >]’
> flow3d2s.cpp:1042:   instantiated from here
> /usr/local/include/gmm_vector.h:215: error: no matching function for call
> to ‘lower_bound(size_t&)’

On line about 215 of gmm_vector.h replace
lower_bound(c);
by
this->lower_bound(c);
(my version of g++ (3.3.5) is not sensitive to this).

>
> Without the preconditioner, the code works fine. I just need the solution
> a little quicker than presently. But this error seems to persist. I
> expected a simple fix. Are there any other alternatives?

It just depends on what kind of linear systems you want to solve.

Sincerly,

Yves.


-- 
-------------------------------------------------------------------------
  Yves Renard (address@hidden)       tel : (33) 04.72.43.80.11
  Pole de Mathematiques, INSA de Lyon          fax : (33) 04.72.43.85.29
  Departement d'Informatique
  Institut Camille Jordan - CNRS UMR 5208
  20, rue Albert Einstein
  69621 Villeurbanne Cedex, FRANCE
  http://math.univ-lyon1.fr/~renard
-------------------------------------------------------------------------