[Bug fortran/36366] [4.3/4.4 Regression] ICE in gfc_conv_component_ref

["pault at gcc dot gnu dot org" <gcc-bugzilla at gcc dot gnu dot org>]

------- Comment #3 from pault at gcc dot gnu dot org  2008-06-07 09:05 -------
Index: gcc/fortran/resolve.c
> ===================================================================
> *** gcc/fortran/resolve.c       (revision 136412)
> --- gcc/fortran/resolve.c       (working copy)
> *************** resolve_fl_derived (gfc_symbol *sym)
> *** 7637,7642 ****
> --- 7637,7647 ----
>           return FAILURE;
>         }
> 
> +       /* Ensure that all the derived type components are put on the
> +        derived type list; even in formal namespaces.  */
> +       if (c->ts.type == BT_DERIVED && c->ts.derived)
> +       resolve_fl_derived (c->ts.derived);
> + 
>         if (c->pointer || c->allocatable ||  c->as == NULL)
>         continue;

This patch broke some tests because it gets into a loop with pointer components
whose type is the same as the 'host' derived type.  This was easily fixed.  I
could not apply the patch because my tree update went pear-shaped.  I'll fix it
all in a weeks time, when I'm back home.

Paul


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=36366