[Bug tree-optimization/64700] Sink common code through PHI

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

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=64700

--- Comment #13 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Andrew Pinski from comment #7)
> I should note store sinking is handled by the sinking pass by
> r11-408-g84935c9822183c .

And now by phiopt in r16-4260-gdecd4277e97114 (even more limited form was done
with r16-1285-g201cd2d3bef8e9).


For the original test we don't commonialize loads still:
```
  <bb 7> [local count: 326018623]:
  _11 = arr1.6_10 + _5;
  _12 = *_11;
  goto <bb 9>; [100.00%]

  <bb 8> [local count: 632859669]:
  _15 = arr2.9_13 + _5;
  _16 = *_15;

  <bb 9> [local count: 958878292]:
  # cstore_14 = PHI <_12(7), _16(8)>
```