From owner-freebsd-hackers  Mon Sep 11 01:22:48 1995
Return-Path: hackers-owner
Received: (from majordom@localhost)
          by freefall.freebsd.org (8.6.12/8.6.6) id BAA14398
          for hackers-outgoing; Mon, 11 Sep 1995 01:22:48 -0700
Received: from GndRsh.aac.dev.com (GndRsh.aac.dev.com [198.145.92.241])
          by freefall.freebsd.org (8.6.12/8.6.6) with ESMTP id BAA14392
          ; Mon, 11 Sep 1995 01:22:43 -0700
Received: (from rgrimes@localhost) by GndRsh.aac.dev.com (8.6.12/8.6.12) id BAA01146; Mon, 11 Sep 1995 01:22:39 -0700
From: "Rodney W. Grimes" <rgrimes@GndRsh.aac.dev.com>
Message-Id: <199509110822.BAA01146@GndRsh.aac.dev.com>
Subject: Re: cvs commit: src/sys/sys conf.h
To: sos@FreeBSD.org
Date: Mon, 11 Sep 1995 01:22:38 -0700 (PDT)
Cc: FreeBSD-hackers@FreeBSD.org (FreeBSD hackers)
In-Reply-To: <199509110745.AAA12527@freefall.freebsd.org> from "sos@FreeBSD.org" at Sep 11, 95 00:45:57 am
X-Mailer: ELM [version 2.4 PL24]
MIME-Version: 1.0
Content-Type: text/plain; charset=US-ASCII
Content-Transfer-Encoding: 7bit
Content-Length: 2518      
Sender: hackers-owner@FreeBSD.org
Precedence: bulk

[I have moved this thread to -hackers, it is no longer a commit
related thread.  NOTE:  I am not on hackers at this time, so don't
kill this note or drop me from an cc: replies :-)]

> > > nate:
> > > Laptops are becoming more and more common, and IBM still is the #1
> > > supplier, though unfortunately still uses the brain-dead keyboard
> > > controller which doesn't work with sycons.
> > 
> > rgrimes:
> > I would not call the PS/2 keyboard controller, aka the type 2 keyboard
> > controller.  IBM fixed several bugs in this design for the MCA class
> > machines, the detection (Gilluwe page 274 is simple) and differences
> > (same, page 273) to work these controllers is _trivial_, Terry Lambert
> > has said it a dozen times, and now I am going to say it, syscons is
> > plain broken if it does not detect and deal with this trivial delta.
> > 
> > It is all of 18 lines of assembler :-).  I'll fax the two pages to
> > any one in the world who _PROMISES_ to fix the code in syscons to
> > deal with this.
> 
> Hmm, does the 18 lines of assembler include the translation nessesary
> for X to work also ?? If so fax me the pages !

No, the 18 lines of assembler tell you if you have a Type 1 or Type 2
controller.  Only the type 1 controller allows you to wiggle the
command byte translate bit to a state of 1 (bit 0x40).  Probably the
big snag that is hanging syscons is (comments my rgrimes added within
the []'s) ``Type 1 controller (described below [in the book, not here])
must wait at least 7 microseconds after bit 0 [of port 0x64] transitions
from 0 to 1, before reading the data from port 0x60.  Code to do this is shown
in the Keyboard_read subroutine.

I suggest since you are the syscons maintainer that you go buy the Gilluwe
book as all of chapter 8 is applicable to you (``The Keyboard system'').
It is 58 pages of _VERY_ good info including your translation stuff I 
think as it shows all the tables.

> 
> 	Fax: +45 9815 7586  

Some how I don't think I am going to get a 14.4 fax connection :-), going
out as soon as I can get to a flat bed photo copier, it is very hard to
feed a 2" book through the fax machine :-).

> please write Att: Soren Schmidt on it..

Okay!  Probably some time tonight, I'll be over at the copy center today
and can photo copy the pages.  You should get the book, ISBN 0-201-62277-7,
it was $45.00 for my copy.

-- 
Rod Grimes                                      rgrimes@gndrsh.aac.dev.com
Accurate Automation Company                 Reliable computers for FreeBSD