*BSD News Article 18783


Return to BSD News archive

Newsgroups: comp.os.386bsd.bugs
Path: sserve!newshost.anu.edu.au!munnari.oz.au!news.Hawaii.Edu!ames!agate!howland.reston.ans.net!xlink.net!gmd.de!borneo!veit
From: veit@borneo.gmd.de (Holger Veit)
Subject: Re: NetBSD-current
Message-ID: <1993Jul23.163226.9610@gmd.de>
Sender: news@gmd.de (USENET News)
Nntp-Posting-Host: borneo
Organization: GMD - German National Research Center for Computer Science
X-Newsreader: Tin 1.1 PL5
References: <22p0d4INNc68@gap.caltech.edu>
Date: Fri, 23 Jul 1993 16:32:26 GMT
Lines: 39

>I am having a problem with my keyboard.  When my i486/66 computer boots up and
>gets to the getty prompt, I am unable to type anything.  However if I hit a 
>few keys during the rc boot up stage, I am able to type on the keyboard.  
>Or if I login remotely and kill the getty, and a new getty is started, I am
>also able then to log in.  This problem only occurs on my i486/66 VLB.  My
>i386/40 works just fine.  I am using the latest NetBSD-current sources.
>Does anyone have similar problems or a solution? 

>-Grge Tanaka
>glt@cco.caltech.edu
>glt@ugcs.caltech.edu

This is likely one of the problems "not-IBM compatible" or "keyboard reconnected"
that exist with different brands of 8042 controllers.
"not-IBM-compatible" means: there are some keyboards that obviously
do no longer (at least) use the bit mapping for the 8042 control register
as specified in the IBM technical reference manual. This also hits
my codrv driver in some cases, so I am interested to fix this.
It seems that a number of newer boards have this type of "cheap"
controllers in, unfortunately a lot of local bus machines are among them.
"keyboard reconnected" means that the cold reset of the keyboard in
pccons does not work correctly, either by slow timing, or by incompatible
response to the reset command. 
As it was reported, these controllers sometimes do not seem to accept
the gatea20 option any longer as well; the gatea20 is then accessible 
by a "FAST A20" option in the BIOS which falls back to some bit in
a register of the board chipset.

There are no definite solutions or fixes that work well with all existing
controllers (the patchkit fix for pccons didn't fix this completely,
apparently), but some experimenting is going on.

--
         Dr. Holger Veit                   | INTERNET: Holger.Veit@gmd.de
|  |   / GMD-SET German National Research  | Phone: (+49) 2241 14 2448
|__|  /  Center for Computer Science       | Fax:   (+49) 2241 14 2342
|  | /   P.O. Box 13 16                    | Had a nightmare yesterday:
|  |/    Schloss Birlinghoven              | My system started up with
         53731 St. Augustin, Germany       | ... Booting vmunix.el ...