[Bug c++/49279] [4.5/4.6/4.7 Regression] Optimization incorrectly presuming constant variable inside loop in g++ 4.5 and 4.6 with -O2 and -O3 for x86_64 targets

[rguenth at gcc dot gnu.org]

http://gcc.gnu.org/bugzilla/show_bug.cgi?id=49279
Richard Guenther <rguenth at gcc dot gnu.org> changed:

           What    |Removed                     |Added
----------------------------------------------------------------------------
           Keywords|                            |wrong-code
             Status|NEW                         |ASSIGNED
      Known to work|                            |4.4.6
         AssignedTo|unassigned at gcc dot       |rguenth at gcc dot gnu.org
                   |gnu.org                     |
            Summary|Optimization incorrectly    |[4.5/4.6/4.7 Regression]
                   |presuming constant variable |Optimization incorrectly
                   |inside loop in g++ 4.5 and  |presuming constant variable
                   |4.6 with -O2 and -O3 for    |inside loop in g++ 4.5 and
                   |x86_64 targets              |4.6 with -O2 and -O3 for
                   |                            |x86_64 targets
      Known to fail|                            |4.5.4, 4.6.0, 4.7.0

--- Comment #2 from Richard Guenther <rguenth at gcc dot gnu.org> 2011-06-06 
09:05:21 UTC ---
Passes with -fno-strict-aliasing.  Presumably the alias-improvement branch
merge is the real issue.  Not sure if it isn't an issue with the testcase
which has

    inline const Derived & derived () const
    {
      return *static_cast < const Derived *>(this);
    } inline Derived & derived ()
    {
      return *static_cast < Derived * >(this);
    }

I wasn't able to follow the template instantiation quickly ... ;)

I will have a look.