From owner-freebsd-hackers Fri Jun 9 16:30:24 1995 Return-Path: hackers-owner Received: (from majordom@localhost) by freefall.cdrom.com (8.6.10/8.6.6) id QAA09474 for hackers-outgoing; Fri, 9 Jun 1995 16:30:24 -0700 Received: from gndrsh.aac.dev.com (gndrsh.aac.dev.com [198.145.92.241]) by freefall.cdrom.com (8.6.10/8.6.6) with ESMTP id QAA09468 for ; Fri, 9 Jun 1995 16:30:21 -0700 Received: (from rgrimes@localhost) by gndrsh.aac.dev.com (8.6.11/8.6.9) id QAA06712; Fri, 9 Jun 1995 16:22:34 -0700 From: "Rodney W. Grimes" Message-Id: <199506092322.QAA06712@gndrsh.aac.dev.com> Subject: Re: problems with Compaq Prosignia 300 To: terry@cs.weber.edu (Terry Lambert) Date: Fri, 9 Jun 1995 16:22:33 -0700 (PDT) Cc: esser@zpr.uni-koeln.de, sakr@itp.ac.ru, hackers@freebsd.org In-Reply-To: <9506091838.AA24845@cs.weber.edu> from "Terry Lambert" at Jun 9, 95 12:38:48 pm X-Mailer: ELM [version 2.4 PL24] MIME-Version: 1.0 Content-Type: text/plain; charset=US-ASCII Content-Transfer-Encoding: 7bit Content-Length: 1502 Sender: hackers-owner@freebsd.org Precedence: bulk > > [ ... Compaq PCI box ... ] > > > The NCR and generic PCI drivers in FreeBSD > > were written by a friend and me, so we > > should be able to work out a solution. > > > > If FreeBSD doesn't see the NCR controller, > > then there probably are some boot messages, > > that might help understand the problem. > > > > I need a complete log of all messages that > > deal with PCI (i.e. start with "pci0"). > > > > You'll probably have to write down these > > messages, I'm afraid. (There are other > > possibilities, but they probably aren't > > worth the effort or they would require > > more hardware.) > > According to someone who had this problem as well and discovered what > it was and posted to the list, the problem is that the Compaq is in > 2.0 mode and FreeBSD only understands 1.x PCI. > > Apparently, you can force it to 1.x mode, and this is what NetBSD has > that FreeBSD does not in this area. Gee Terry, as one who is pushing VM86 bios calls all the time I am surprized you did not come up with the real correct solution and that is to call the PCI BIOS32 interface (yes, *all* PCI spec 1.x and 2.x compliant machines *must* implement this). We don't even need VM86 to do it since these are protected mode safe BIOS calls. We do have to make sure we save the BIOS data areas though, but that is pretty easy to fix. -- Rod Grimes rgrimes@gndrsh.aac.dev.com Accurate Automation Company Custom computers for FreeBSD