[petsc-users] Ksp Initial residual norm

[Matthew Knepley]

On Sun, Feb 3, 2019 at 2:24 PM Smith, Barry F. via petsc-users <
petsc-users at mcs.anl.gov> wrote:

>
>
> > On Feb 3, 2019, at 1:16 PM, Edoardo alinovi <edoardo.alinovi at gmail.com>
> wrote:
> >
> > Thank you very much Barry for the suggestion.
> >
> > Unfortunately, I am using Fortran and not C++ . Do you have an
> equivalent trick in this case? :)
>
>    You would need to provide this function as a C function (in its own
> little .c file) and then call it from your Fortran code.
>

I think you can also use:


https://www.mcs.anl.gov/petsc/petsc-current/docs/manualpages/KSP/KSPGetResidualHistory.html

    Matt


>    Barry
>
>    There is no way to access the data structure directly from Fortran.
>
> > ------
> >
> > Edoardo Alinovi, Ph.D.
> >
> > DICCA, Scuola Politecnica,
> > Universita' degli Studi di Genova,
> > 1, via Montallegro,
> > 16145 Genova, Italy
> >
> > Email: edoardo.alinovi at dicca.unige.it
> > Tel: +39 010 353 2540
> >
> >
> >
> >
> >
> > Il giorno dom 3 feb 2019 alle ore 20:03 Smith, Barry F. <
> bsmith at mcs.anl.gov> ha scritto:
> >
> >    If you use the default KSP convergence tests (which almost everyone
> does) the initial residual norm is stored in ksp->rnorm0. Thus to access it
> you need to include the private header file that defines the KSP object
> kspimpl.h so something like
> >
> > #include <petsc/private/kspimpl.h>
> >
> >    PetscErrorCode KSPGetInitialResidualNorm(KSP ksp,PetscReal *norm0)
> >   {
> >    *norm0 = ksp->rnorm0;
> >    return 0;
> >   }
> >
> > should work.
> >
> >    Barry
> >
> >
> >
> >
> >
> >
> > > On Feb 3, 2019, at 8:27 AM, Edoardo alinovi via petsc-users <
> petsc-users at mcs.anl.gov> wrote:
> > >
> > > Dear all,
> > >
> > > I have a very simple question for you. How can I get the initial
> residual from a ksp solver? I can easily extract the final (preconditioned)
> residual norm  but  when I try to use to do the same with the initial
> residual using:
> > >
> > > kspInitialResidual()
> > >
> > > PETSc tells me that this feature is not supported. I am actually doing
> it applying the definition of residual "by hand", but I am wondering if
> somenthing already built-in is available or not.
> > >
> > > Thank you very much!
> > > ------
> > >
> > > Edoardo Alinovi, Ph.D.
> > >
> > > DICCA, Scuola Politecnica,
> > > Universita' degli Studi di Genova,
> > > 1, via Montallegro,
> > > 16145 Genova, Italy
> > >
> > > Email: edoardo.alinovi at dicca.unige.it
> > > Tel: +39 010 353 2540
> > >
> > >
> > >
> >
>
>

-- 
What most experimenters take for granted before they begin their
experiments is infinitely more interesting than any results to which their
experiments lead.
-- Norbert Wiener

https://www.cse.buffalo.edu/~knepley/ <http://www.cse.buffalo.edu/~knepley/>
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-users/attachments/20190203/0a68fca2/attachment.html>