Bruno, Thank you. Is there any particular reason that these raw pointers are not exposed outside ? It would make it much easier for people who would like to solve using their own solvers on gpu's and hybrid systems much easier, right ? without depending on Trilinos or Petsc ?
Best, Pratik. On Thursday, August 23, 2018 at 4:10:02 PM UTC+2, Bruno Turcksin wrote: > > Pratik, > > Le jeu. 23 août 2018 à 09:34, <[email protected] <javascript:>> a écrit : > >> >> Thank you. So, I can get the matrix values by &system_matrix.begin() and >> the column indices by doing &sparsity_pattern.begin() but I am a bit >> confused as to how I get the row pointers. Or did I misunderstand something >> ? >> > Take a look here > <https://github.com/dealii/dealii/blob/master/source/lac/sparse_direct.cc#L196> > > for the method I have explained. Note that you do not get access to a > pointer to the rows and columns but to an iterator to a structure which you > can ask for the row and column indices. If you need to have access to the > raw pointers, you need to modify deal.II by making your matrix class a > friend of SparseMatrix (see here > <https://github.com/dealii/dealii/blob/master/include/deal.II/lac/sparse_matrix.h#L1710> > ) > > Best, > > Bruno > -- The deal.II project is located at http://www.dealii.org/ For mailing list/forum options, see https://groups.google.com/d/forum/dealii?hl=en --- You received this message because you are subscribed to the Google Groups "deal.II User Group" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. For more options, visit https://groups.google.com/d/optout.
