From owner-freebsd-mobile Thu Aug 1 12:39:54 1996 Return-Path: owner-mobile Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id MAA03117 for mobile-outgoing; Thu, 1 Aug 1996 12:39:54 -0700 (PDT) Received: from rocky.mt.sri.com (rocky.mt.sri.com [206.127.76.100]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id MAA03112 for ; Thu, 1 Aug 1996 12:39:51 -0700 (PDT) Received: (from nate@localhost) by rocky.mt.sri.com (8.7.5/8.7.3) id NAA08134; Thu, 1 Aug 1996 13:28:39 -0600 (MDT) Date: Thu, 1 Aug 1996 13:28:39 -0600 (MDT) Message-Id: <199608011928.NAA08134@rocky.mt.sri.com> From: Nate Williams To: mobile@freebsd.org, hosokawa@mt.cs.keio.ac.jp Subject: Pccardc diffs Sender: owner-mobile@freebsd.org X-Loop: FreeBSD.org Precedence: bulk The following shar file contains the 'useful' changes from the -current (as of 8/1/96) usr.sbin/pccard/pccardc sources to the 7/31-PAO release sources. Only the functional changes are supplied, white-space and other gratituious changes are not included. I hope to do the same with pccardd as well, but it's changed a bit more and trying to merge in the changes has been alot more work due to a whole bunch of changes I have difficulties accepting. IMHO there are *MUCH* betters ways of solving the problem, but since everyone involved has been so busy communication has been difficult. Note: The diffs provided introduce bugs, but I haven't spent time to look at them. The IRQ's listed by the old code are correct, and the code with these changes in them are wrong. Note also that these diffs will not compile on a FreeBSD-current system and require additional changes to the kernel tree which have not been (and may not be) integrated. In particular, I don't like the 'beep' functions as they currently exist in the Nomad implementation, and in my opinion the 'power' code is a kludge to work-around the non-ability of suspend/resume on the laptops. The primary intention of the below diff is to allow the Nomad's and other developers to 'sync' up their sources with the FreeBSD sources *completely*. They should be able grab a copy of FreeBSD-current sources today and apply these changes to pccardc, at which point in time they will have the *exact* same functional sources as before w/out any non-necessary work for the Nomad's and/or myself when intergration works. My secondary purpose is to have people review the PAO 'functional' diffs and fix the bugs it introduces, especially regarding the printing out of the IRQ's. The original code may have bugs in it, but at least the bugs aren't visible in my testing, so the PAO code may fix bugs and introduce new ones at the same time. If someone else wants to help out and do the pccardd merge I would certainly appreciate the help. My time is *very* limited to work on this, but I don't want any 'current' work to slip through the cracks. Finally, in regards to the new 'kernel' changes, it is my belief that you can completely throw out the PAO APM code since: 1) It doesn't follow the specification w/regards to apm_idel/busy 2) It adds *NO* functional changes but does complicate the readability of the code. I intend to integrate my APM BIOS changes posted a couple weeks earlier into the -current kernel, and then remove more of the complexity from the APM code soon. The patches I sent out for review went out at a bad time right after 2.1.5, so I'll ping the folks and see if I can get a response on them. My goals are: 1) Integrate the APM/BIOS changes posted earlier 2) Finish the usr.sbin/pccard/pccardd merges 3) Review the 'XT_KEYBOARD' change from the PAO release and run it by S'ren. I like the idea, but am unsure if the implementation is acceptable by S'ren. 4) Completely remove /sys/i386/include/laptop.h - It's an un-necessary kludge that I never should have brought in. 5) Integrate in some of the new drivers and upgrade the older ones with the newer code. We're getting closer to having a much closer 'merge' of the PAO sources and FreeBSD-current, and given that everyone's time is limited it will make it easier on both the users and the developers to have the sources more closely resemble one another. Nate