Return to BSD News archive
Path: sserve!newshost.anu.edu.au!munnari.oz.au!spool.mu.edu!howland.reston.ans.net!europa.eng.gtefsd.com!uunet!mcsun!sun4nl!hacktic!not-for-mail
From: ptuomola@hacktic.nl (Petri Tuomola)
Newsgroups: comp.os.386bsd.questions
Subject: Re: The reason for stray interrupts
Date: 27 Oct 1993 09:04:43 +0100
Organization: Hack-Tic, networking for the masses
Lines: 30
Message-ID: <2ala39INNcr0@xs4all.hacktic.nl>
References: <2ais9gINN2t8@xs4all.hacktic.nl> <wilko.751660652@spoetnix.idca.tds.philips.nl> <2ajv84$9r2@pdq.coe.montana.edu>
NNTP-Posting-Host: xs4all.hacktic.nl
osyjm@cs.montana.edu (Jaye Mathisen) writes:
>DEC 386-16 and 386-20 sx's don't have the problem, I get 0 stray intr
>messages on those machines.
>On the other hand, the 486-24 that DEC made that was produced by Tandy does
>have the problem.
It might be (although I'm not sure) that this has something to do with speed.
This machine I'm running NetBSD on is a 386/33, and it does get stray
interrupts. If I turn off the turbo mode (read: from 33MHz to 8MHz), stray
interrupts disappear. And your experience seems to show the same - on slow
systems stray interrupts don't occur.
But that should, in fact, be obvious... if a stray interrupt is caused by
someone raising a interrupt line and 8259 having not enough time to priorize
it, when running at lower speed it has more time and priorizes it etc...
Oh well.
But, I have also ran both Mach386 and Interactive SysV on this hardware
at 33MHz without stray interrupts. Or don't they just show them?
Petri
--
----------------------------------------------------------------------------
Petri Tuomola (root@echelon.hacktic.nl) (ptuomola@hacktic.nl)
"Get stoned - eat wet concrete" HAM: OH2LJY