From owner-freebsd-mobile  Sun Nov  2 19:22:30 1997
Return-Path: <owner-freebsd-mobile>
Received: (from root@localhost)
          by hub.freebsd.org (8.8.7/8.8.7) id TAA00913
          for mobile-outgoing; Sun, 2 Nov 1997 19:22:30 -0800 (PST)
          (envelope-from owner-freebsd-mobile)
Received: from word.smith.net.au (vh1.gsoft.com.au [203.38.152.122])
          by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id TAA00905
          for <freebsd-mobile@freebsd.org>; Sun, 2 Nov 1997 19:22:18 -0800 (PST)
          (envelope-from mike@word.smith.net.au)
Received: from word.smith.net.au (localhost.gsoft.com.au [127.0.0.1])
	by word.smith.net.au (8.8.7/8.8.5) with ESMTP id NAA01109;
	Mon, 3 Nov 1997 13:44:40 +1030 (CST)
Message-Id: <199711030314.NAA01109@word.smith.net.au>
X-Mailer: exmh version 2.0zeta 7/24/97
To: Nate Williams <nate@mt.sri.com>
cc: Mike Smith <mike@smith.net.au>, Guido van Rooij <guido@gvr.org>,
        freebsd-mobile@freebsd.org
Subject: Re: problems after PAO -> 2.2.5 stable 
In-reply-to: Your message of "Sun, 02 Nov 1997 18:39:51 PDT."
             <199711030139.SAA04709@rocky.mt.sri.com> 
Mime-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Date: Mon, 03 Nov 1997 13:44:39 +1030
From: Mike Smith <mike@smith.net.au>
Sender: owner-freebsd-mobile@freebsd.org
X-Loop: FreeBSD.org
Precedence: bulk

> > Remember chaps, the IRQ from the CIS is complete rubbish.  PCCARDs
> > only have *one* interrupt pin; it's the routing in the pcic that does
> > the work.
> 
> Yes, but the current FreeBSD code tells the driver/card that it's using
> the specific interrupt, so it doesn't always work correctly.  We should
> tell the driver card that it's using an 'acceptable' interrupt, but use
> whatever one that is free, but we don't.

Uh, I don't think we are on the same wavelength.  Consider:

device          sio1    at isa? port "IO_COM2" tty irq3 vector siointr

# Megahertz XJ4336
card "Megahertz" "XJ4336-CC4336"
        config  0x21 "sio1" 10
        insert  echo Modem arrives...
        remove  echo Modem departs...

Tuple #6, code = 0x1b (Configuration entry), length = 7
    000:  21 08 aa 60 f8 02 07
        Config index = 0x21
        Card decodes 10 address lines, 8 Bit I/O only
                I/O address # 1: block start = 0x2f8 block length = 0x8

This *works*.  To me, it is clear that the IRQ parameter from the 
pccard.conf entry is being propagated to the sio probe/attach.

What am I missing in your example?

mike