On Mon, Jun 15, 2009 at 6:04 PM, Bill Nottingham<notting redhat com> wrote:That sounds about right to me.
>> It would probably be most interesting to perform that test on the
>> x86-only ATOM, since I can see CMOV being a bigger win on an in-order
>> (I can't personally protest: I think all the x86 stuff I have has CMOV).
> Last time I profiled it (using the bits of openbench that actually worked),
> i586 -> i686 was an improvement of 1% (Core2Duo) to 2% (Atom). Athlon64 was
> essentially equal.
> SPECCPU results showed similar.