[<prev] [next>] [<thread-prev] [thread-next>] [day] [month] [year] [list]
Message-ID: <40101cc30910031504pa9f8536k82dbd7668dfa643d@mail.gmail.com>
Date: Sun, 4 Oct 2009 00:04:03 +0200
From: Matteo Croce <technoboy85@...il.com>
To: Arjan van de Ven <arjan@...radead.org>
Cc: "H. Peter Anvin" <hpa@...or.com>, linux-kernel@...r.kernel.org
Subject: Re: i686 quirk for AMD Geode
On Sat, Oct 3, 2009 at 8:05 PM, Arjan van de Ven <arjan@...radead.org> wrote:
> On Sat, 3 Oct 2009 04:12:47 +0200
> Matteo Croce <technoboy85@...il.com> wrote:
>>
>> I also did this simple test which gives a ~2.5x speed boost:
>> http://pastebin.ca/1590089
>
> the number that is exposed in /proc/cpuinfo does not control
> if cmov is used... so the patch and the benchmark are totally unrelated.
>
> Anything that needs to know if cmov can be used will need to read the
> flags line..
>
> (since technically cmov is also optional for generation 6 and later)
>
> While I can't say anything against your patch, I do need to say that it
> is something purely cosmetic on first sight...
>
> --
> Arjan van de Ven Intel Open Source Technology Centre
> For development, discussion and tips for power savings,
> visit http://www.lesswatts.org
>
yes cmov and i686 are unrelated but some distros (Debian & co.)
compiles libraries for two archs:
i486 and i686-cmov so only i686 machines does benefit from cmov.
This is why I cared about having an i686 aware userspace but it seems
that for a missing instruction I can't do that
--
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@...r.kernel.org
More majordomo info at http://vger.kernel.org/majordomo-info.html
Please read the FAQ at http://www.tux.org/lkml/
Powered by blists - more mailing lists