[Bug tree-optimization/100363] gcc generating wider load/store than warranted at -O3

[pinskia at gcc dot gnu.org via Gcc-bugs]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=100363

Andrew Pinski <pinskia at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
          Component|middle-end                  |tree-optimization
           Keywords|                            |wrong-code

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
On the trunk, on aarch64:
There should be an aliasing check 

  sfrom_289 = from_176 + 18446744073709551615;
  _871 = _843 + 18446744073709551615;
  _872 = _871 > 6;
  _873 = prephitmp_803 + 2;
  _874 = from_176 + 3;
  _875 = _873 - _874;
  _876 = (sizetype) _875;
  _877 = _876 > 12;
  _878 = _872 & _877;
  if (_878 != 0)
    goto <bb 116>; [80.00%]
  else
    goto <bb 126>; [20.00%]


_873 is the sout


In GCC 10 branch we get something similar:
  sfrom_289 = from_176 + 18446744073709551615;
  _859 = _823 + 18446744073709551615;
  _860 = _859 > 8;
  _861 = prephitmp_783 + 2;
  _862 = from_176 + 3;
  _863 = _861 - _862;
  _864 = (sizetype) _863;
  _865 = _864 > 12;
  _866 = _860 & _865;
  if (_866 != 0)
    goto <bb 116>; [80.00%]
  else
    goto <bb 126>; [20.00%]

But I Notice 8 vs 6 here.