On Mon, Jun 15, 2009 at 5:08 PM, Gregory Maxwell <gmaxwell@xxxxxxxxx> wrote:
On Mon, Jun 15, 2009 at 6:04 PM, Bill Nottingham<notting@xxxxxxxxxx> 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
>> CPU.
>> (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.
My problem with this is that I have a friend that uses Fedora almost exclusively for his systems. He uses Pentium III, Pentium II, and Athlon XP systems. So, the only real issue is SSE2. No Athlon chip prior to AMD Athlon 64 supports SSE2. So if we move to i686, SSE2 would have to be cut out.
-- fedora-devel-list mailing list fedora-devel-list@xxxxxxxxxx https://www.redhat.com/mailman/listinfo/fedora-devel-list
