*BSD News Article 21963


Return to BSD News archive

Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!howland.reston.ans.net!sol.ctr.columbia.edu!news.kei.com!bloom-beacon.mit.edu!ai-lab!life.ai.mit.edu!mycroft
From: mycroft@duality.gnu.ai.mit.edu (Charles Hannum)
Newsgroups: comp.os.386bsd.questions
Subject: Re: NetBSD 0.9 w/- Ultrastore 34f = No Work
Date: 06 Oct 1993 10:29:46 GMT
Organization: MIT Artificial Intelligence Lab
Lines: 23
Message-ID: <MYCROFT.93Oct6062946@duality.gnu.ai.mit.edu>
References: <28q581$12ra@thebes.cc.waikato.ac.nz>
NNTP-Posting-Host: duality.gnu.ai.mit.edu
In-reply-to: hamish@thebes.cc.waikato.ac.nz's message of 4 Oct 1993 21:40:17 GMT


In article <28q581$12ra@thebes.cc.waikato.ac.nz>
hamish@thebes.cc.waikato.ac.nz (Hamish Marson) writes:

   Hi all, I have been trying to install NetBSD on a 486 66Mhz DX2
   with an ultrastore 34f, and it doesn't seem to want to work. The
   board is detected, and the correct parameters are read from the
   card, but then the system says that the board isn't responding, and
   stops, saying I should hit reset....

This is probably related to the `hanging with disk light on' problem
that's been reported by some people with both NetBSD and FreeBSD.
I've seen the same problem recently on a machine here with a 34f but
have not had time to debug it.

Basically, for whatever reason the interrupt (apparently) is getting
delayed until after the disk controller has given up and dropped the
interrupt line, and thus it registers as a stray interrupt.

What I did to work around this was to configure DDB and manually `call
uhaintr 0' every time I got a stray interrupt, but this is severely
suboptimal.  B-)