[Bug middle-end/17128] [3.5 Regression] ICE in tree-inline

[jh at suse dot cz]

------- Additional Comments From jh at suse dot cz  2004-09-09 09:06 -------
Subject: Re:  [3.5 Regression] ICE in tree-inline


static __inline__ int
one_utf8_to_utf16 () { }

static __inline__ unsigned char
conversion_loop (int (*const one_conversion)())
{
return one_conversion ();
}
static unsigned char
convert_utf8_utf16 ()
{
  return conversion_loop (one_utf8_to_utf16);
}

2004-09-09  Jan Hubicka  <jh@suse.cz>
	middle-end/17128
	* tree-inline.c (expand_call_inline): Make overactive sanity check happy.
Index: tree-inline.c
===================================================================
RCS file: /cvs/gcc/gcc/gcc/tree-inline.c,v
retrieving revision 1.137
diff -c -3 -p -r1.137 tree-inline.c
*** tree-inline.c	5 Sep 2004 16:05:00 -0000	1.137
--- tree-inline.c	9 Sep 2004 09:02:47 -0000
*************** expand_call_inline (tree *tp, int *walk_
*** 1481,1487 ****
           where previous inlining turned indirect call into direct call by
           constant propagating arguments.  In all other cases we hit a bug
           (incorrect node sharing is most common reason for missing edges.  */
!       if (!dest->needed)
  	abort ();
        cgraph_create_edge (id->node, dest, t)->inline_failed
  	= N_("originally indirect function call not considered for inlining");
--- 1481,1487 ----
           where previous inlining turned indirect call into direct call by
           constant propagating arguments.  In all other cases we hit a bug
           (incorrect node sharing is most common reason for missing edges.  */
!       if (!dest->needed && flag_unit_at_a_time)
  	abort ();
        cgraph_create_edge (id->node, dest, t)->inline_failed
  	= N_("originally indirect function call not considered for inlining");


-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=17128