*BSD News Article 22871


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!ariel.ucs.unimelb.EDU.AU!werple.apana.org.au!news
From: andrew@werple.apana.org.au (Andrew Herbert)
Newsgroups: comp.os.386bsd.questions
Subject: Re: The reason for stray interrupts
Date: 27 Oct 1993 22:31:45 +1000
Organization: werple public-access unix, Melbourne
Lines: 36
Message-ID: <2alpnh$1cm@werple.apana.org.au>
References: <2ais9gINN2t8@xs4all.hacktic.nl>
NNTP-Posting-Host: werple.apana.org.au

ptuomola@hacktic.nl (Petri Tuomola) writes:

>Many people have been asking why their machines display "ISA strayintr 7"
>messages. This is an extract from /sys/arch/i386/isa/isa.c:
...
>BTW. My experience is that they cause no harm, except the very small extra
>system load caused by stray interrupts getting handled (you can't notice the
>difference). I get stray interrupts with a "steady rate":

>interrupt         count     rate
>stray irq          2985        1 <--
>com2 irq9          2985        1

Notice the coincidence?

On my system they were also caused by a serial board or two.  I suspect the
four-porter was doing it, actually.

Studying the fas serial driver (recently posted to one of the source
newsgroups) will show the great lengths required to prevent (or at least
minimise) these nasties.

Or you can get a Cyclom-8Y serial board and use my NetBSD cy driver. :-)

ttyp2 / 81 > vmstat -i
interrupt         count     rate
clk irq0       13498315       99
fdc0 irq6             1        0
pc0 irq1         141237        1
cy0 irq9        6047652       44
mms0 irq5        143525        1
ahb0 irq11      1933774       14
we0 irq10       1351548       10
Total          23116052      171

Andrew