regressions?

[Dominique Dhumieres]

> Something is definitely broken on your system!  I think we
> can agree that _gfortran_iargc() isn't _gfortran_fraction_r4
> or _gfortran_exponent_r4.  What does "svn diff iresolve.c"
> show.

880,902d879
< gfc_resolve_index_func (gfc_expr * f, gfc_expr * str,
<                       ATTRIBUTE_UNUSED gfc_expr * sub_str, gfc_expr * back)
< {
<   gfc_typespec ts;
< 
<   f->ts.type = BT_INTEGER;
<   f->ts.kind = gfc_default_integer_kind;
< 
<   if (back && back->ts.kind != gfc_default_integer_kind)
<     {
<       ts.type = BT_LOGICAL;
<       ts.kind = gfc_default_integer_kind;
<       ts.derived = NULL;
<       ts.cl = NULL;
<       gfc_convert_type (back, &ts, 2);
<     }
< 
<   f->value.function.name =
<     gfc_get_string ("__index_%d_i%d", str->ts.kind, f->ts.kind);
< }
< 
< 
< void
1048,1049c1025
<   f->value.function.name = gfc_get_string ("__len_%d_i%d", string->ts.kind,
<                                          gfc_default_integer_kind);
---
>   f->value.function.name = gfc_get_string ("__len_%d", string->ts.kind);

i.e., the changes corresponding to the FX Coudert's patch

http://gcc.gnu.org/ml/fortran/2006-10/msg00022/intrinsics.diff

I don't this I have made a mistake in this file. If this is the cause of the
problem, I'll need a closer look to what I have done elsewhere.

Dominique