[Bug middle-end/65855] missing optimization: triangular numbers

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=65855

Richard Biener <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |ASSIGNED
   Last reconfirmed|                            |2015-04-23
           Assignee|unassigned at gcc dot gnu.org      |rguenth at gcc dot gnu.org
     Ever confirmed|0                           |1

--- Comment #1 from Richard Biener <rguenth at gcc dot gnu.org> ---
For some reason we can't compute the number of iterations of this loop which is
because we have to assume that

 (long unsigned int) n_3(D) != 18446744073709551615

and for some reason we couldn't simplify that to true!?

  if (!exit_must_be_taken && !POINTER_TYPE_P (type))
    {
      if (integer_nonzerop (iv0->step))
        assumption = fold_build2 (NE_EXPR, boolean_type_node,
                                  iv1->base, TYPE_MAX_VALUE (type));

if we fix that we fail in the following way:

(chrec_apply
  (varying_loop = 1
)
  (chrec = {1, +, {2, +, 1}_1}_1)
  (x = (long unsigned int) n_3(D) + 18446744073709551615)
  (res = scev_not_known))
not replacing:
  t_4 = PHI <t_5(3)>

because we can't compute the final result expression.