From owner-freebsd-current Fri Jan 31 5: 0:51 2003 Delivered-To: freebsd-current@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id 013C437B401 for ; Fri, 31 Jan 2003 05:00:49 -0800 (PST) Received: from shrek.physik.ruhr-uni-bochum.de (shrek.physik.ruhr-uni-bochum.de [134.147.86.22]) by mx1.FreeBSD.org (Postfix) with ESMTP id B521743F43 for ; Fri, 31 Jan 2003 05:00:47 -0800 (PST) (envelope-from strattbo@stud.physik.ruhr-uni-bochum.de) Received: from m42.physik.ruhr-uni-bochum.de ([134.147.86.42]) by shrek.physik.ruhr-uni-bochum.de with esmtp (Exim 3.35 #1 (Debian)) id 18eamk-00031S-00 for ; Fri, 31 Jan 2003 14:00:46 +0100 Received: from strattbo by m42.physik.ruhr-uni-bochum.de with local (Exim 3.35 #1 (Debian)) id 18eamk-0002Oe-00 for ; Fri, 31 Jan 2003 14:00:46 +0100 Date: Fri, 31 Jan 2003 14:00:46 +0100 To: current@freebsd.org Subject: ACPI kernel: no ata0 / no ide disks Message-ID: <20030131130046.GA9186@m42.physik> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.4i From: Thomas Stratmann Sender: owner-freebsd-current@FreeBSD.ORG Precedence: bulk List-ID: List-Archive: (Web Archive) List-Help: (List Instructions) List-Subscribe: List-Unsubscribe: X-Loop: FreeBSD.ORG Hi everyone, I finally got 5.0 of last week on my Tosh. Libretto 110ct running. Took me several days of headache and bootstrap problems, nfs mount and chroot finally solved my problems. (My laptop is a multiboot box with a 4.7 FBSD as well so I could mount the filesystem dedicated to the 5.0 system. I cast a naked 5.0 install onto my other box to do the chrooted kernel/world compiling) Currently I still have two serious problems with my 5.0 kernel, which is in one case GENERIC itself and in the other basically GENERIC again, except I commented out SCSI and USB stuff, uncommented several debugging opts (I knew I would be running into trouble some time...) and only cpu I586_CPU of the cpu lines. I state both problems here since I think they could be related (but probably are not). Problem 1) occurred with the GENERIC kernel, and as well with the new kernel when I >unset acpi_load in the final boot stage (the aprop. module probably just didn't exist prior installation of the custom kernel for some reason, so this all is not about switching the kernel). The symptom is that booting seems to hang (it turned out to be some kind of timeout of 5mins or so) at device probing time right after vga0: at port 0x3c0-0x3df iomem 0xa0000-0xbffff on isa0 Booting -v showed that this delay happens in function isa_probe_pnp_children (could as well be isa_pnp_probe_children or so). When probing continues, the next five lines are of the kind unknown: can't assign resources (port) where one line has (irq) instead of (port). Sorry I have no means of appending dmesg at this moment. So, it works, but booting means making coffee in between. Here some device topology of this setup that might be interesting for problem 2: ad0 at ata0-master, ata0 on isa0 on motherboard. pci0 on pcib0 at pcibus 0 on motherboard. Problem 2) occurred when the acpi module was loaded: the kernel found no disks! The words 'isa' and 'ata' do not appear on the screen, but now my pci topology is pci0 on pcib0 on acpi0 on motherboard. acpi0 is (spaces exactly as on screen) I'm new to acpi and not very good in hardware, so this is really confusing to me. I cannot tell in what way this topology is normal or abnormal, but I would like to use both acpi and my disk! Another thing confuses me, maybe someone can give me some enlightenment: Rebooting without acpi, I noticed that syslogd took long to start up, together with lots of disk activity. dmesg showed the boot messages from all previous boots (which probably syslogd spooled from somewhere, which would explain the long startup and disk light), including the failed boots with acpi support. Now, in these latter cases my kernel saw no disk, so where did the information go that magically appeared later in my dmesg? This is a riddle to me! I would appreciate if someone could shed some light on this. If you need more info (dmesg's out e.g.) drop me a mail at thomas.stratmann@ruhr-uni-bochum.de and I will post it to -current. Cheers, Thomas To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message