[Bug tree-optimization/62031] [4.8 Regression] Different results between O2 and O2 -fpredictive-commoning

[rguenth at gcc dot gnu.org]

https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62031

--- Comment #11 from Richard Biener <rguenth at gcc dot gnu.org> ---
(In reply to rguenther@suse.de from comment #10)
> On Wed, 15 Oct 2014, clyon at gcc dot gnu.org wrote:
> 
> > https://gcc.gnu.org/bugzilla/show_bug.cgi?id=62031
> > 
> > --- Comment #9 from clyon at gcc dot gnu.org ---
> > Forgot to mention that my observation is true on trunk and 4.9 branch.
> > 
> > I didn't notice it earlier because I run validations for every commit on the
> > 4.9 branch, and not as often on trunk.
> 
> Can you please try to investigate?  (does it only fail at -O0?!)

Especially please state if you are cross-testing and one target triplet that
shows the failure.