[Bug middle-end/23410] [4.1 Regression] FAIL: gcc.c-torture/execute/950612-1.c execution, at -Os and -O3

["sebastian dot pop at cri dot ensmp dot fr" <gcc-bugzilla at gcc dot gnu dot org>]

------- Additional Comments From sebastian dot pop at cri dot ensmp dot fr  2005-08-18 11:51 -------
Reduced testcase is:

unsigned long long
f4 (unsigned long long diff)
{
  return ((unsigned long long) ((signed long long) diff < 0 ? -diff : diff));
}

main ()
{
  int i;
  for (i = 0; i <= 10; i++)
    if (f4 ((long long) -i) != i)
      abort ();
  
  exit (0);
}

And the problem is in VRP:

Visiting statement:
diff.0D.1338_6 = (long long intD.5) diffD.1337_4;

(analyze_scalar_evolution 
  (loop_nb = 1)
  (scalar = diff.0_6)
(get_scalar_evolution 
  (scalar = diff.0_6)
  (scalar_evolution = (long long int) {0, +, ffffffffffffffff}_1))
(set_scalar_evolution 
  (scalar = diff.0_6)
  (scalar_evolution = (long long int) {0, +, ffffffffffffffff}_1))
)
(instantiate_parameters 
  (loop_nb = 1)
  (chrec = (long long int) {0, +, ffffffffffffffff}_1)
  (res = (long long int) {0, +, ffffffffffffffff}_1))
Found new range for diff.0_6: [0, 0]

The computation of the min bound is not correct.  I'm working on a patch.

-- 


http://gcc.gnu.org/bugzilla/show_bug.cgi?id=23410