[Bug c/48966] gcc 4.5 gets wrong answer at -O2

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=48966

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|UNCONFIRMED                 |RESOLVED
         Resolution|                            |INVALID

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-05-11 
15:30:00 UTC ---
You are violating C aliasing rules here:

  answer = (*(((int *)&tmp)+1) >> 31) & 1;
            ^