[Bug tree-optimization/54855] Unnecessary duplication when performing scalar operation on vector element

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=54855

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |NEW
   Last reconfirmed|                            |2012-10-08
     Ever Confirmed|0                           |1

--- Comment #1 from Richard Guenther <rguenth at gcc dot gnu.org> 2012-10-08 14:19:27 UTC ---
Confirmed.  Does not work for + though, as -0.0 + 0.0 is 0.0.  At least
if I remember the signed-zero mess correctly ;)

On the tree level we see in-memory v because of the component modification:

  _7 = BIT_FIELD_REF <v, 64, 0>;
  _8 = _7 - 1.0e+0;
  BIT_FIELD_REF <v, 64, 0> = _8;
  v.0_10 = v;
  v.1_11 = v.0_10 * { 2.0e+0, 2.0e+0 };
  v = v.1_11;

so either lowering this differently in the first place or detecting
this kind of pattern would fix it.

Similar trick may be used for multiplication and division.