From owner-freebsd-current Sat Nov 23 14:00:54 1996 Return-Path: owner-current Received: (from root@localhost) by freefall.freebsd.org (8.7.5/8.7.3) id OAA15427 for current-outgoing; Sat, 23 Nov 1996 14:00:54 -0800 (PST) Received: from rf900.physics.usyd.edu.au (rf900.physics.usyd.edu.au [129.78.129.109]) by freefall.freebsd.org (8.7.5/8.7.3) with ESMTP id OAA15411; Sat, 23 Nov 1996 14:00:42 -0800 (PST) Received: (from dawes@localhost) by rf900.physics.usyd.edu.au (8.8.2/8.8.2) id JAA05404; Sun, 24 Nov 1996 09:00:38 +1100 (EST) From: David Dawes Message-Id: <199611232200.JAA05404@rf900.physics.usyd.edu.au> Subject: Re: Four problems that really bug me with CURRENT In-Reply-To: <199611231301.OAA17983@x14.mi.uni-koeln.de> from Stefan Esser at "Nov 23, 96 02:00:30 pm" To: se@FreeBSD.org (Stefan Esser) Date: Sun, 24 Nov 1996 09:00:38 +1100 (EST) Cc: current@FreeBSD.org X-Mailer: ELM [version 2.4ME+ PL28 (25)] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Sender: owner-current@FreeBSD.org X-Loop: FreeBSD.org Precedence: bulk >> >PCI bus probing seems not to be a general problem. >> >Please send more details about your system (chip set, >> >peripherals, ...). The most important information is >> >the VERBOSE boot message log. (Boot with "-v" and send >> >me the output of "dmesg", or cut the messages out of the >> >"/var/log/messages" file ...) >> >> The recent XFree86 release uses PCI code that is very closely based on > >There is PCI code in XFree86 ??? Yes. It was originally used for passive probing, but for some video cards, we also need to write to the PCI config space. It would be better if most OSs provided an interface for doing this though. >> that in 2.2-current from about two months ago, and I know of some >> cases where the PCI bus probe seems to fail. The common factor is >> that they are PCs made by HP, and at least some are PPro machines. > >Well, but how do you think I'm going to fix this, >if I have no idea what actually fails. Send me the >information I asked for (verbose boot messages and >the information about the chip set, CPU and BIOS >you can easily get at), and I'll make it work. The machines I'm referring to are not running FreeBSD (usually they are running Linux). At least one of the PPro machines is using a fairly old revision of the Orion chipset. I'll see if I can get some more information about this. As another data point, the scanpci program distributed with XFree86 (which uses differnet PCI code) does detect the PCI bus on these machines, and after that has been run, the PCI code in the XFree86 servers works again. David