Date: Mon, 11 Mar 2002 09:29:25 -0500 From: Brian T.Schellenberger <bts@babbleon.org> To: Kris Kennaway <kris@obsecurity.org> Cc: Kris Kennaway <kris@obsecurity.org>, BOUWSMA Beery <freebsd-user@netscum.dyndns.dk>, hackers@FreeBSD.ORG Subject: Re: Performance of FreeBSD vs NetBSD (was: Re: Performance of -current vs -stable) Message-ID: <20020311142926.3079BBA05@i8k.babbleon.org> In-Reply-To: <20020311055415.A17397@xor.obsecurity.org> References: <200203102005.g2AK5SN99334@beerswilling.netscum.dyndns.dk> <20020311132340.105D6BA05@i8k.babbleon.org> <20020311055415.A17397@xor.obsecurity.org>
next in thread | previous in thread | raw e-mail | index | archive | help
On Monday 11 March 2002 08:54 am, Kris Kennaway wrote: | On Mon, Mar 11, 2002 at 08:23:38AM -0500, Brian T.Schellenberger wrote: | > On Monday 11 March 2002 03:25 am, Kris Kennaway wrote: | > | On Sun, Mar 10, 2002 at 09:05:28PM +0100, BOUWSMA Beery wrote: | > | > I built both a WITNESS and a WITNESSless kernel with more recent | > | > k0deZ, and in the case of playing an mp3 file with `mpg123', I | > | > saw practically no difference between the two, based on %cpu as | > | > shown by `top' (like I say, completely unscientific and inaccurate) | > | | > | As you are no doubt aware there are significant infrastructural | > | changes in -current relating to SMP scalability. It's in a very | > | interim state at the moment, and one of the downsides is increased | > | interrupt latency and lock contention for certain operations (yes, | > | audio playback is one of them). | > | | > | Basically, it's a known issue. | > | > At -stable as well as -current or at -current only? | | What I'm talking about is a -current issue only. I don't recall | reading the earlier thread. Well, the original article in *this* thread stated: In fact, the %cpu needed by `mpg123' seemed identical between -current, both with and without W[I]TNESS, and -stable. Look: [55% CPU to decode for all three cases] In both -current and -stable, the audio is usually smooth but periodically has a hiccup or two and loops briefly. But the very same hardware, booted into NetBSD off the same disk, running a NetBSD-native binary of mpg123 on NetBSD-current shows this: [37% CPU to decode] So I don't think that your explanation covers the observed behavior. Though the *system* time and *interrupt* time seem to track as one might expect for this issue, there still seems to be something *else* going on per the mpg123 time and idle time. Ref: http://docs.freebsd.org/cgi/getmsg.cgi?fetch=13691+0+current/freebsd-hackers | | Kris -- Brian T. Schellenberger . . . . . . . bts@wnt.sas.com (work) Brian, the man from Babble-On . . . . bts@babbleon.org (personal) ME --> http://www.babbleon.org http://www.eff.org <-- GOOD GUYS --> http://www.programming-freedom.org To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-hackers" in the body of the message
Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?20020311142926.3079BBA05>