It was indeed a case of misalignment of compiled files (in my case, make
install going to one location, and the code to be compiled picking up from
a different one)

Thanks to all,
Salvatore

On Mon, Oct 17, 2022 at 4:42 PM Andrew Benson <aben...@carnegiescience.edu>
wrote:

> Hi Salvatore,
>
> I've run into similar error messages occasionally. I've never been able to
> fully track down the cause, but it seems to happen if I change the content
> of
> a derived type (e.g. add some new variable to it), recompile the module
> containing it, but don't recompile some other module that USEs that first
> module. It seems like there are then multiple definitions of the derived
> type
> (from different modules) that are out of sync. So, recompiling everything
> resolves this issue for me.
>
> But, as I said, I'm not 100% sure that this is the explanation!
>
> -Andrew
>
> On Monday, October 17, 2022 2:28:33 AM PDT Salvatore Filippone via Fortran
> wrote:
> > Dear all
> > I am getting the following error message, which I find quite confusing
> >
> >    68 |   use amg_prec_mod
> >
> >       |       1
> >
> > Fatal Error: Mismatch in components of derived type
> > '__vtype_psb_c_par_mat_mod_Psb_c_par_mat_type' from 'psb_c_par_mat_mod'
> at
> > (1): expecting 'bldext', but got 'aclsum'
> > compilation terminated.
> >
> > from
> >  mpifort -v
> > mpifort for MPICH version 3.4.3
> > Using built-in specs.
> > COLLECT_GCC=/opt/gnu/12.1.0/bin/gfortran
> >
> COLLECT_LTO_WRAPPER=/opt/gnu/12.1.0/libexec/gcc/x86_64-pc-linux-gnu/12.1.0/l
> > to-wrapper Target: x86_64-pc-linux-gnu
> > Configured with: ../gcc-12.1.0/configure --prefix=/opt/gnu/12.1.0
> > --enable-languages=c,c++,fortran
> > Thread model: posix
> > Supported LTO compression algorithms: zlib
> > gcc version 12.1.0 (GCC)
> >
> >
> > What might  the compiler be  complaining about? Where can I find
> additional
> > info on what to do?
> > Any help greatly appreciated
> >
> > Salvatore
>
>
> --
>
> * Andrew Benson: https://abensonca.github.io
>
> * Galacticus: https://github.com/galacticusorg/galacticus
>
>
>
>

Reply via email to