From owner-freebsd-mobile Sun Nov 2 11:04:28 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id LAA06651 for mobile-outgoing; Sun, 2 Nov 1997 11:04:28 -0800 (PST) (envelope-from owner-freebsd-mobile) Received: from gvr.gvr.org (root@gvr.gvr.org [194.151.74.97]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id LAA06646 for ; Sun, 2 Nov 1997 11:04:13 -0800 (PST) (envelope-from guido@gvr.org) Received: (from guido@localhost) by gvr.gvr.org (8.8.6/8.8.5) id UAA05422; Sun, 2 Nov 1997 20:03:57 +0100 (MET) From: Guido van Rooij Message-Id: <199711021903.UAA05422@gvr.gvr.org> Subject: Re: card removal problems In-Reply-To: <199711021712.KAA02976@rocky.mt.sri.com> from Nate Williams at "Nov 2, 97 10:12:39 am" To: nate@mt.sri.com (Nate Williams) Date: Sun, 2 Nov 1997 20:03:57 +0100 (MET) Cc: nate@mt.sri.com, freebsd-mobile@freebsd.org X-Mailer: ELM [version 2.4ME+ PL32 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-freebsd-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk Nate Williams wrote: > > > > Just like with the PAO release I experience problems when removing a > > > > card from a slot. Sometimes this results in a total lockup of the > > > > system, while it keeps beeping. > > > > > > There is a race where it's possible that the driver in in use when you > > > remove it that is un-avoidable. But, this seems to happen pretty > > > rarely. > > > > Actually, it seems that it almost always locks up for me once a driver > > has ben allocated. I inserted that modem card I was talking about > > after applying my patch (so pcic would use not irq 3). It was not being > > used but still the machine locked up. > > I'm confused. You say it would lockup the machine *only* after a driver > was allocated, or it didn't matter. If it doesn't matter, then > something is wrong with your PCIC controller, since all we do is apply > power (or not) to your controller. No, this is after I changed pccard.c to go downwards with irq's. So now I can allocate a driver. > > > I am now going to implement the power patches to see if that would > > solve my problem. > > I'd like to think there is a better solution that doesn't require user > interaction. > > So, if there's no entry for the card in question in /etc/pccard.conf, > after inserting it, waiting for a driver to be NOT found, and then > removing the card your box will lockup? If so, then the power patches No. It does certianly not lock up then/ > won't make any difference unless they 'power' down the card differently > than I do in removal, which means that I can power the card down the > same way and you won't need the 'power' patches. > Is there anything I can do, except low level kernel debugging to daignose this? -Guido