From owner-freebsd-current  Mon Sep  2 01:28:24 1996
Return-Path: owner-current
Received: (from root@localhost)
          by freefall.freebsd.org (8.7.5/8.7.3) id BAA24405
          for current-outgoing; Mon, 2 Sep 1996 01:28:24 -0700 (PDT)
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.19])
          by freefall.freebsd.org (8.7.5/8.7.3) with SMTP id BAA24252
          for <freebsd-current@FreeBSD.org>; Mon, 2 Sep 1996 01:27:33 -0700 (PDT)
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.12/8.6.9) id SAA10477; Mon, 2 Sep 1996 18:24:59 +1000
Date: Mon, 2 Sep 1996 18:24:59 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199609020824.SAA10477@godzilla.zeta.org.au>
To: freebsd-current@FreeBSD.org, j@uriah.heep.sax.de
Subject: Re: ddb now requires sio
Cc: koshy@india.hp.com
Sender: owner-current@FreeBSD.org
X-Loop: FreeBSD.org
Precedence: bulk

As A JOSEPH KOSHY wrote:
> 
> `option DDB' now /requires/ sio0 to be present on the system.
> 
> In `.../conf/files.i386' we have:
> 
>     i386/i386/i386-gdbstub.c        optional        ddb
> 
> and `i386-gdbstub.c' calls many functions from the serial driver.

This should be easy to circumvent by #include'ing "sio.h", and
depending the gdb remote stuff from #if NSIO > 0.

i386-gdbstub.c should call cngetc() instead of siocngetc(), etc.
This probably doesn't work now because the console driver is so
primitive:

1) it does LF -> CRLF translations on output, and there is no way
   to stop this.

2) It only supports _one_ console device and that device can't be
   changed properly after it is decided at boot time.  This is OK
   for /dev/console but it leaves the driver `cn' functions for all
   devices other than the console inaccessible, and for debugging
   it's best if the physical device is different from the one used
   for the console.  The low level (driver) `cn' functions take a
   dev_t arg but all drivers ignore it, while the high level `cn'
   functions don't take a dev_t arg.

Bruce