[Bug middle-end/47033] loop distribution has problems on sane testcases

[pinskia at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=47033

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

           What    |Removed                     |Added
----------------------------------------------------------------------------
             Status|NEW                         |RESOLVED
         Resolution|---                         |FIXED
           See Also|                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=43884,
                   |                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=45948,
                   |                            |https://gcc.gnu.org/bugzill
                   |                            |a/show_bug.cgi?id=44334
   Target Milestone|---                         |4.6.0

--- Comment #5 from Andrew Pinski <pinskia at gcc dot gnu.org> ---
(In reply to Jan Hubicka from comment #0)
> The following patch:

Which was committed as r0-106204 with the change of the testcase too but not
the number of memset's so I assume the problem reported here was fixed.


Note r8-18 removed the undefined behavior from the testcase and updated the
testcase, as PRE is able to remove one of the stores even.