*BSD News Article 30827


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!harbinger.cc.monash.edu.au!msuinfo!agate!ihnp4.ucsd.edu!news.service.uci.edu!bob
From: bob@beamlab.ps.uci.edu (bob prohaska)
Newsgroups: comp.os.386bsd.questions
Subject: Re: HELP: Error 11 when compiling kernel
Date: 23 May 1994 06:54:03 GMT
Organization: University of California, Irvine
Lines: 19
Message-ID: <2rpjub$6nc@news.service.uci.edu>
References: <allenm.769403889@wizard.ucs.sfu.ca> <JKH.94May22060811@whisker.hubbard.ie>
Reply-To: rprohask@uci.edu
NNTP-Posting-Host: nemesis.ps.uci.edu
X-Newsreader: TIN [version 1.2 PL2]

Just for the record, I'd suggest swapping the memory. 


I've been seeing the same problem, and it finally went away when my
16 megs of 70ns memory was replaced with 8 megs of apparently slightly
faster memory. The 70 ns memory tested ok under all circumstances, and
even ran OS/2 without trouble. It wouldn't run Windows NT, which was
the event that induced me to swap it with 16 megs of 60 ns memory from
a mac ci. Now both machines appear to work well.

The really curious part is that no amount of fiddling with the motherboard
speed, turbo switch and wait states made any difference. The only way
to identify the problem was physical replacement of the memory. Power on
self tests were always ok, and whatever memory test software I tried was
always happy. In fact, the "70 ns" memory tested at about 55 ns, probably
about the same as the 60 ns parts (which, because they seem to work, have
never been subject to any serious scrutiny).

bob.