[patch,gfortran] PR 17740

[Erik Edelmann]

tobias.schlueter@physik.uni-muenchen.de wrote:
> Erik Edelmann wrote:
> > There is however, one thing that I am still a bit unsure about (I
> > hope this question doesn't sound too stupid); I learned (the hard
> > way) that expr2->value.function.esym is NULL sometimes (which is
> > why the patch checks that first).  Can I trust
> > expr2->value.function.esym to allways be non-NULL for elemental
> > functions?  I browsed the gfortran source code, but found no
> > definite answer to that question.
> 
> It can apparently, see resolve.c:767.
>   if (sym->attr.generic)
>     {
>       s =
> 	gfc_search_interface (sym->generic, 0, &expr->value.function.actual);
>       if (s != NULL)
> 	{
> 	  expr->value.function.name = s->name;
> 	  expr->value.function.esym = s;
> 	  expr->ts = s->ts;
> 	  if (s->as != NULL)
> 	    expr->rank = s->as->rank;
> 	  return MATCH_YES;
> 	}
> 
>       /* TODO: Need to search for elemental references in generic interface */
>     }
> I assume what's missing is that array arguments are matched against elemental
> function in interfaces, but I'm not sure.
> 
> Furthermore, trans-expr.c:1550 says this:
>   /* expr.value.function.esym is the resolved (specific) function symbol for
>      most functions.  However this isn't set for dummy procedures.  */
>   sym = expr->value.function.esym;
>   if (!sym)
>     sym = expr->symtree->n.sym;
>   gfc_conv_function_call (se, sym, expr->value.function.actual);
> 
> It looks like your patch is correct, but I'd rather we understand when esym
> should be set, first.

I've browsed some more code, and did some testing.  First,
value.function.esym was non-NULL in all test I did with dummy
procedures, so I don't quite understand that comment from
trans-expr.c:1550 above.  Instead, value.function.esym seems to
be NULL for, and only for, intrinsics.  My patch work with
elemental intrinsics because elemental intrinsics are cought by a
previous if-statement:

if (expr2->value.function.isym && !gfc_is_intrinsic_libcall (expr2))
    return NULL;

If we still want to be really safe from elemental intrinsics, we
could of course check for them explicitely; modified patch
attached.  (This by the way how we check for elemental functions
in resolve.c/resolve_function().)


        Erik
-------------- next part --------------
Index: gcc/fortran/trans-expr.c
===================================================================
RCS file: /cvsroot/gcc/gcc/gcc/fortran/trans-expr.c,v
retrieving revision 1.57
diff -u -p -r1.57 trans-expr.c
--- gcc/fortran/trans-expr.c	16 Aug 2005 12:58:46 -0000	1.57
+++ gcc/fortran/trans-expr.c	22 Aug 2005 21:23:48 -0000
@@ -2165,7 +2165,10 @@ gfc_trans_arrayfunc_assign (gfc_expr * e
     return NULL;
 
   /* Elemental functions don't need a temporary anyway.  */
-  if (expr2->symtree->n.sym->attr.elemental)
+  if ((expr2->value.function.esym != NULL
+       &&  expr2->value.function.esym->attr.elemental)
+      || (expr2->value.function.isym != NULL
+       &&  expr2->value.function.isym->elemental))
     return NULL;
 
   /* Check for a dependency.  */