Return to BSD News archive
Path: euryale.cc.adfa.oz.au!newshost.carno.net.au!harbinger.cc.monash.edu.au!munnari.OZ.AU!news.ecn.uoknor.edu!news.cis.okstate.edu!newsfeed.ksu.ksu.edu!news.mid.net!mr.net!news.sgi.com!enews.sgi.com!news.mathworks.com!hunter.premier.net!newsfeed.direct.ca!nntp.teleport.com!news.serv.net!news.serv.net!michaelv
From: michaelv@MindBender.serv.net (Michael L. VanLoon)
Newsgroups: comp.unix.bsd.netbsd.misc
Subject: Re: VM_FALTs with AMD486 and VIA chip set.
Date: 27 Aug 1996 05:12:57 GMT
Organization: Michael L. VanLoon
Lines: 45
Message-ID: <MICHAELV.96Aug26221257@MindBender.serv.net>
References: <markus-2608960833010001@evt-pm0-ip3.halcyon.com>
NNTP-Posting-Host: mindbender.serv.net
In-reply-to: markus@halcyon.com's message of Mon, 26 Aug 1996 08:33:01 -0800
In article <markus-2608960833010001@evt-pm0-ip3.halcyon.com> markus@halcyon.com (Mark Hastings) writes:
I have a number of systems running NetBSD 1.1, on a board from TriEMS.
They use the AMD486DX4/100 with the write back cache. We are seeing
spontaneous VM_FALTS. All boards seem to do it sooner or later. Some
systems may crash frequently and others, one or twice a month.
I wrote a simple script that runs about 40 instances of "find". This will
crash almost any system within 1 to 5 minutes, but we can't seem
to figure out if we have a hardware or software problem, or a
combination of both. Anyone have any ideas what to check?
It sounds to me like a hardware problem. It sounds like you have
slightly flaky cache circuitry, which is further supported by the
fact that you say all the boards, with the same chipsets, do the same
thing.
Have you tried with the cache in only write-thru mode, instead of
write-back. To further convince yourself it isn't NetBSD, if that
doesn't do it, try running with the cache disabled completely.
BTW, these boards seem to run just fine with OS/2 or Windows.
Which Windows. Windows 3.1? 95? NT? FWIW, Windows 3.1 would be
almost as worthless as DOS for finding hardware problems like this
since it isn't really a protected-mode demand-paged pre-emptive
multitasking OS. Windows 95 would be part way there, but it even has
some compromises in its design -- I'd put it in the same category as
OS/2. Windows NT I would expect to act more like NetBSD -- if there
is a weak link in your hardware, it will exaggerate it.
Doing what with OS/2? Did you write anything or find anything that
would thrash the system as thoroughly as the multiple find script you
ran on NetBSD? Merely saying they run fine just sitting there with a
copy of Excel open doesn't say much.
--
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -
Michael L. VanLoon michaelv@MindBender.serv.net
--< Free your mind and your machine -- NetBSD free un*x >--
NetBSD working ports: 386+PC, Mac 68k, Amiga, Atari 68k, HP300, Sun3,
Sun4/4c/4m, DEC MIPS, DEC Alpha, PC532, VAX, MVME68k, arm32...
NetBSD ports in progress: PICA, others...
- - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - - -