Relay-Version: version B 2.10 5/3/83; site utzoo.UUCP
Path: utzoo!mnetor!uunet!seismo!rochester!rutgers!cbmvax!grr
From: grr@cbmvax.UUCP (George Robbins)
Newsgroups: comp.sys.amiga
Subject: Re: Ctrl-Amiga-Amiga: who takes care of it?
Message-ID: <2171@cbmvax.UUCP>
Date: Fri, 31-Jul-87 10:15:29 EDT
Article-I.D.: cbmvax.2171
Posted: Fri Jul 31 10:15:29 1987
Date-Received: Sun, 2-Aug-87 01:42:36 EDT
References: <5162@utcsri.UUCP>
Reply-To: grr@cbmvax.UUCP (George Robbins)
Distribution: world
Organization: Commodore Technology, West Chester, PA
Lines: 29

In article <5162@utcsri.UUCP> flaps@utcsri.UUCP (Alan J Rosenthal) writes:
> 
> I've crashed my Amiga really thoroughly and found Ctrl-Amiga-Amiga still to
> work.  Is it implemented at a hardware level to any extent, or what?  How
> does it work?

In the A1000 and A2000, the Control-Amiga-Amiga reset works as follows:
The keyboard microprocessor (ROM based) senses the key combination.
It sends a "warning" keycode to the CPU...
I don't think AmigaDOS does anythign special...
It jams the keyboard clock line low...
Hardware on the CPU board times out the clock line...
And asserts reset and halt on the 68000...
Thats all she wrote.

In the A500, a logic gate in the keyboard detects the key combination and
asserts reset directly.  They keyboard micro would still like to send its
keycode and whatnot, but never gets a chance.

So anyway, the system does do a hard, unavoidable, reset when you hit
the C-A-A key combination.  There are however some things in the software
that try to recover the contents of memory after a reset that could
conceivably puke and die if some areas of memory looked good, but were
actually corrupt.

-- 
George Robbins - now working for,	uucp: {ihnp4|seismo|rutgers}!cbmvax!grr
but no way officially representing	arpa: cbmvax!grr@seismo.css.GOV
Commodore, Engineering Department	fone: 215-431-9255 (only by moonlite)