From owner-freebsd-mobile Tue Dec 2 15:56:24 1997 Return-Path: Received: (from root@localhost) by hub.freebsd.org (8.8.7/8.8.7) id PAA17089 for mobile-outgoing; Tue, 2 Dec 1997 15:56:24 -0800 (PST) (envelope-from owner-freebsd-mobile) Received: from ns.mt.sri.com (sri-gw.MT.net [206.127.105.141]) by hub.freebsd.org (8.8.7/8.8.7) with ESMTP id PAA17082 for ; Tue, 2 Dec 1997 15:56:20 -0800 (PST) (envelope-from nate@mt.sri.com) Received: from mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by ns.mt.sri.com (8.8.8/8.8.8) with SMTP id QAA02415 for ; Tue, 2 Dec 1997 16:56:18 -0700 (MST) (envelope-from nate@rocky.mt.sri.com) Received: by mt.sri.com (SMI-8.6/SMI-SVR4) id QAA17208; Tue, 2 Dec 1997 16:56:17 -0700 Date: Tue, 2 Dec 1997 16:56:17 -0700 Message-Id: <199712022356.QAA17208@mt.sri.com> From: Nate Williams MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Transfer-Encoding: 7bit To: mobile@FreeBSD.org Subject: PCI/CardBus framework committed to -current X-Mailer: VM 6.29 under 19.15 XEmacs Lucid Sender: owner-freebsd-mobile@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk Thanks to Ted Faber, we now have a framework for supporting additional PCI/CardBus controllers in PCMCIA legacy mode. For now, the only supported controller is the CLPD6832, but I'm hoping that others are easily added. Note, I modified his code to 'recognize' other card controllers in a similar manner to the old code in -current, but it only enables support for the above mentioned controller. I need folks who have this controller to test the code, and folks who have other controllers to look at the code in /sys/pci/pcic_p.[ch], and copy what Ted has done for your controller. Before anyone gets too excited about CardBus support, this is *NOT* that, but allows people with CardBus machines to use the older PCMCIA (16-bit) cards, which are standard. However, newer cards are CardBus (32-bit), and we do *NOT* support them, or the controllers in 32-bit mode. However, this is a change in the right direction, at least for supporting newer laptops which are almost exclusively using CardBus controllers. Finally, I also made the previous option PCIC_NOCLRREGS the default option, since this is apparently needed for it to work on Ted's box. I tested it on my machines with no ill effects, but it could possibly negatively affect others, so beware. If it does, send me email and we'll try and figure out how to work-around this. On other fronts, there is still some suspend/resume problems with the code in the tree, but there are some patches in testing that should help. These problems happen rarely and shouldn't affect most folks, but I wanted you to be aware that resumes can still fail in certain circumstances, and that I'm aware of the problem. Nate ps. I need to see some testing on this code before I feel comfortable bringing it into -stable, so *please* test it and don't wait for me to bring it in, causing problems as previously happened. :(