[petsc-dev] HDF5 Viewer for Fortran 90

[Chris Eldred]

Hey Satish,

Everything seems to be working fine with HDF5 bindings for Fortran now.
I'll let you guys know if I run into any other problems!

-Chris

On Thu, May 30, 2013 at 1:43 PM, Satish Balay <balay at mcs.anl.gov> wrote:

> I pused the fix in the branch balay/hdf5-ftnbinding. But since this is
> off 'maint/3.4' - you can do the following to use it with petsc-dev.
>
> # checkout origin/master in detached mode - so that it can be discarded
> later
> git fetch origin
> git checkout origin/master
> git merge origin/balay/hdf5-ftnbinding
> make allfortranstubs
> make
>
> Let us know if it works.
>
> thanks,
> Satish
>
>  On Thu, 30 May 2013, Satish Balay wrote:
>
> > On Thu, 30 May 2013, Satish Balay wrote:
> >
> > > Looks like we need custom bindings for
> > > PetscErrorCode  PetscViewerHDF5PushGroup(PetscViewer viewer, const
> char *name)
> > > PetscErrorCode  PetscViewerHDF5GetGroup(PetscViewer viewer, const char
> **name)
> > >
> > > Will take a look.
> > >
> > > Fortran bindings don't make sense for these two?
> > > PetscErrorCode  PetscViewerHDF5Open(MPI_Comm comm, const char name[],
> PetscFileMode type, PetscViewer *hdf5v)
> > I see we already have a  custom binding for this one
> >
> > Satish
> >
> > > PetscErrorCode  PetscViewerHDF5GetFileId(PetscViewer viewer, hid_t
> *file_id)
> > >
> > > Satish
> > >
> > > On Wed, 29 May 2013, Barry Smith wrote:
> > >
> > > >
> > > >    Satish,
> > > >
> > > >      Please branch off of maint and get them all into the next patch
> (as well as master) if Jed hasn't done this.
> > > >
> > > >    Barry
> > > >
> > > > On May 29, 2013, at 3:29 PM, Jed Brown <jedbrown at mcs.anl.gov> wrote:
> > > >
> > > > > Chris Eldred <chris.eldred at gmail.com> writes:
> > > > >
> > > > >> I am having issues getting the HDF5 Viewer to work with Fortran
> 90.
> > > > >> Specifically, I get the following types of errors when attempting
> to link
> > > > >> my executable together:
> > > > >>
> > > > >> /home/user/Dropbox/NTM/SIEVETEST/OPERATOR/src/output.f90:664:
> undefined
> > > > >> reference to `petscviewerhdf5settimestep_'
> > > > >> /home/user/Dropbox/NTM/SIEVETEST/OPERATOR/src/output.f90:668:
> undefined
> > > > >> reference to `petscviewerhdf5pushgroup_'
> > > > >> /home/user/Dropbox/NTM/SIEVETEST/OPERATOR/src/output.f90:675:
> undefined
> > > > >> reference to `petscviewerhdf5popgroup_'
> > > > >>
> > > > >> It seems like the Fortran interfaces for the HDF5-specific viewer
> functions
> > > > >> are not present (there are no issues with the other viewer
> functions).
> > > > >
> > > > > Yes, PetscViewerHDF5PushGroup() needs a custom binding.  The
> others can
> > > > > be auto-generated.
> > > >
> > > >
> > >
> > >
> >
> >
>
>


-- 
Chris Eldred
DOE Computational Science Graduate Fellow
Graduate Student, Atmospheric Science, Colorado State University
B.S. Applied Computational Physics, Carnegie Mellon University, 2009
chris.eldred at gmail.com / celdred at atmos.colostate.edu
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20130603/02f94959/attachment.html>