Return to BSD News archive
Newsgroups: comp.unix.bsd
Path: sserve!manuel!munnari.oz.au!uunet!zaphod.mps.ohio-state.edu!darwin.sura.net!rsg1.er.usgs.gov!ornl!jamsun.ic.ornl.gov!mullens
From: mullens@jamsun.ic.ornl.gov (James A. Mullens)
Subject: Re: Some more on NMI problems (some meager advancement)
Message-ID: <1992Sep8.181528.3876@ornl.gov>
Sender: usenet@ornl.gov (News poster)
Organization: Oak Ridge National Laboratory
References: <1992Sep7.014351.946@ponds.uucp> <cproto.715945053@marsh>
Date: Tue, 8 Sep 1992 18:15:28 GMT
Lines: 55
In article <cproto.715945053@marsh>, cproto@marsh.cs.curtin.edu.au (Computer Protocol) writes:
|> rivers@ponds.uucp (Thomas David Rivers) writes:
|>
|>
|> >Well, I thought I would relay my current status with the NMI
|> >investigation.
|>
|> > Right now, I'm thinking it has something to do with an IDE
|> > controller/disk drive, so I have been examining the wd.c driver
|> > trying to divine what it might be; without too much luck. (I
|> > know very little about the IDE/WD disk controllers.)
|>
|> [STUFF DELETED]
|>
|> > - Dave Rivers -
|> > (rivers@ponds.uucp)
|>
|> I'm also experiencing the problem. I disagree with the IDE controller
|> idea. I still think it's a spurious parity error. Note that the NMI
|> interrupt reports the value of I/O port 61. This value has the highest
|> bit set which indicates a parity error. But after running QAPLUS for
|> several days I'm convinced my memory is ok.
|>
|> Could it be some sort of setup problem - e.g. wait states which can
|> be configured in my AMI BIOS.
|>
|> BTW I do have a IDE controller/disk so Dave might be right.
Memory parity errors can also be the fault of the system board/bus
controller. You really need to check for errors while your system
is "highly loaded" -- lots of I/O going on. A memory tester may not
be able to see the problem otherwise...
I once had a completely-reproducible NMI which could not be found by
a tester. It seems that my computer's "bus tended to be unstable
during periods of high load" and the chips I had purchased did not
tolerate this very well. I switched chip manufacturers and the computer
has run w/o problems ever since. This was so long ago that I
can't recommend the manufacturer, NMBS, on this basis. In fact, I
visited our PC repair shop and found a notice that, FOR A CERTAIN MODEL
OF COMPUTER, NMBS and Siemens DRAM "will cause problems". (Assume that
is someone's theory, not an authority's pronouncement! Don't read too
much into this since -- it just indicates that not all motherboards
and DRAM mix well). You might call your computer/system-board manufacturer
and try to get a recommendation.
I have also read a recommendation that 3-chip and 9-chip SIMMs not be
mixed together, but I have no real reason to believe this makes a
difference.
jim mullens
Internet: jcm@ornl.gov (128.219.128.17)
mullens@jamsun.ic.ornl.gov (128.219. 64.31)
I speak for myself, not for the organization I work for.