From owner-freebsd-current  Sun Sep  3 23:27:31 1995
Return-Path: current-owner
Received: (from majordom@localhost)
          by freefall.FreeBSD.org (8.6.11/8.6.6) id XAA18125
          for current-outgoing; Sun, 3 Sep 1995 23:27:31 -0700
Received: from godzilla.zeta.org.au (godzilla.zeta.org.au [203.2.228.34])
          by freefall.FreeBSD.org (8.6.11/8.6.6) with ESMTP id XAA18099
          for <freebsd-current@freebsd.org>; Sun, 3 Sep 1995 23:27:12 -0700
Received: (from bde@localhost) by godzilla.zeta.org.au (8.6.9/8.6.9) id QAA10078; Mon, 4 Sep 1995 16:22:21 +1000
Date: Mon, 4 Sep 1995 16:22:21 +1000
From: Bruce Evans <bde@zeta.org.au>
Message-Id: <199509040622.QAA10078@godzilla.zeta.org.au>
To: deischen@iworks.InterWorks.org, freebsd-current@freebsd.org
Subject: Re: Getting around conflicts for a driver w/out base address
Sender: current-owner@freebsd.org
Precedence: bulk

>Isn't there already a size field - id_alive?  At least that's how it looks
>like it is being used in isa.c.  It is set to the value returned by a driver's
>probe function.

id_alive is the driver-alive flag corrupted to hold the i/o size.  It isn't
possible to set it at config time.  Actually, we need a list of i/o port
ranges because some drivers use scattered ports.

>How about making the drivers set this field like they do for others (id_maddr
>and id_msize for example)?  Then the probe routine can actually return a
>success/fail status.

They should set a new id_iosize field.

>Or is it more complicated than this when considering config?

No, but introducing a new field is a bit more complicated, especially for
a field that should be settable using config.

Bruce