[petsc-dev] new Fortran error

[Mark Adams]

>
>
> I suspect you have upgraded gfortran and now have gfortran version 10,


09:35 knepley/feature-swarm-fortran= ~/Codes/petsc$ mpif90 --version
GNU Fortran (Homebrew GCC 10.1.0) 10.1.0


> which is pickier about argument matching.
>
> If you add -fallow-argument-mismatch to the fortran flags, do things go
> away?
>
>
Well, it turns 6 errors into 2 warnings:

          FC arch-macosx-gnu-g/tests/dm/impls/swarm/tutorials/ex1f90.o
/Users/markadams/Codes/petsc/src/dm/impls/swarm/tutorials/ex1f90.F90:39:30:

   34 |       call PetscObjectSetName(dm, 'Potential Grid',
ierr);CHKERRA(ierr)
      |                              2
......
   39 |       call PetscObjectSetName(fe, 'fe', ierr);CHKERRA(ierr)
      |                              1
Warning: Type mismatch between actual argument at (1) and actual argument
at (2) (INTEGER(8)/TYPE(tdm)).
     FLINKER arch-macosx-gnu-g/tests/dm/impls/swarm/tutorials/ex1f90
        TEST
arch-macosx-gnu-g/tests/counts/dm_impls_swarm_tutorials-ex1f90_0.counts
 ok dm_impls_swarm_tutorials-ex1f90_0
 ok diff-dm_impls_swarm_tutorials-ex1f90_0

Oh well, I guess we need to add this param if we want to set names and take
it out of regression tests.

I wonder why it seems to say a PetsFE is an INTEGER(8) ...

Thanks,
Mark


> Lawrenmce
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.mcs.anl.gov/pipermail/petsc-dev/attachments/20200710/c19210f1/attachment.html>