Skip site navigation (1)Skip section navigation (2)
Date:      Mon, 14 Jun 1999 14:34:55 -0700
From:      Aaron Smith <aaron-fbsd@mutex.org>
To:        James FitzGibbon <james@ehlo.com>
Cc:        freebsd-questions@FreeBSD.ORG, freebsd-smp@FreeBSD.ORG
Subject:   Re: CPU states at 0.0% on 3.2-R SMP box ? 
Message-ID:  <199906142134.OAA55831@sigma.veritas.com>
In-Reply-To: Your message of "Mon, 14 Jun 1999 17:21:43 EDT." <19990614172141.A30011@ehlo.com> 

next in thread | previous in thread | raw e-mail | index | archive | help
this is a problem with an rtc interrupt being lost, which apparently wedges
up all further rtc interrupts. i solved it by searching the mailing list
archives and located this post:

http://docs.freebsd.org/cgi/getmsg.cgi?fetch=54158+0+archive/1999/freebsd-smp/19990607.freebsd-smp

(in case the URL doesn't work for some reason it's by bob wilcox from may
20, the patch is supposedly from tor egge(?) and it patches clkintr).

once i applied the patch i never saw any weirdness.

apparently this is a temporary workaround? can somebody explain more fully?
what cost am i paying for a patch named "BROKEN_RTC_KLUDGE"?

Aaron Smith

On Mon, 14 Jun 1999 17:21:43 EDT, James FitzGibbon writes:
>Back in march of this year there was a thread in -questions and -smp about
>SMP boxes showing all CPU states at 0.0% and processes continuing to take up
>CPU time but not do any actual work.  The responses to people's questions
>were usually along the lines of "Try 3.1-RELEASE, it should be fixed there".
>
>I've just installed 3.2-RELEASE on a Dual Xeon-450 ASUS board, and am seeing
>the same problem.  I've tried using an aout version of /sbin/init, as was
>suggested by someone on the old thread, but that didn't solve anything.
>
>Is there a definate answer about this ?  My boss says that when he last
>tried FreeBSD on this box, it was very unstable, and he corelates the wacky
>top behaviour with this instability.  Hence, as long as top acts wacky, I
>can't move this box into production.
>
>I'm trying to find out if this is merely a presentation/lookup problem or if
>the CPU cycles really aren't being sliced up properly between processors. 
>If anyone has more information, I'd really appreciate hearing it. 
>Alternatively, if anyone who knows the SMP code wants to hack around on a
>box that exhibits this problem, I'd be glad to give you access.


To Unsubscribe: send mail to majordomo@FreeBSD.org
with "unsubscribe freebsd-smp" in the body of the message




Want to link to this message? Use this URL: <https://mail-archive.FreeBSD.org/cgi/mid.cgi?199906142134.OAA55831>