[<prev] [next>] [<thread-prev] [day] [month] [year] [list]
Message-ID: <20060807192219.GA17521@suse.cz>
Date: Mon, 7 Aug 2006 21:22:19 +0200
From: Vojtech Pavlik <vojtech@...e.cz>
To: Andi Kleen <ak@...e.de>
Cc: Andreas Mohr <andi@...x01.fht-esslingen.de>,
Dmitry Torokhov <dtor@...ightbb.com>,
Rusty Russell <rusty@...tcorp.com.au>,
lkml - Kernel Mailing List <linux-kernel@...r.kernel.org>
Subject: Re: [PATCH] Turn rdmsr, rdtsc into inline functions, clarify names
On Mon, Aug 07, 2006 at 06:12:03PM +0200, Andi Kleen wrote:
> Vojtech Pavlik <vojtech@...e.cz> writes:
>
> > On Mon, Aug 07, 2006 at 05:57:14PM +0200, Andi Kleen wrote:
> > > > That way a driver could use
> > > >
> > > > if (gtod_cpu_cycles_needed <= 500)
> > > > gettimeofday();
> > > > else
> > > > funky_fast_workaround();
> > >
> > > Sounds like overengineering to me. I prefer something simple.
> >
> > I could as well benchmark gettimeofday() in the gameport init.
>
> Except you can't because the only way to benchmark it would
> be to use gettimeofday already and then you don't know
> how much is overhead and what is the real time.
I run it for 1/10th of a second in a loop and see how many iterations
I've done. That's enough to check if its fast enough to use in the
gameport routine. Less than 50k iterations and I can't use it.
--
Vojtech Pavlik
Director SuSE Labs
-
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