From owner-freebsd-current Tue Dec 3 6:22:10 2002 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 7478F37B401 for ; Tue, 3 Dec 2002 06:22:08 -0800 (PST) Received: from grassmarket.ucs.ed.ac.uk (grassmarket.ucs.ed.ac.uk [129.215.166.64]) by mx1.FreeBSD.org (Postfix) with ESMTP id 6912943EBE for ; Tue, 3 Dec 2002 06:21:57 -0800 (PST) (envelope-from bruce@cran.org.uk) Received: from fourtytwo.gamesoc (12266209.resnet.ed.ac.uk [10.6.0.100] (may be forged)) by grassmarket.ucs.ed.ac.uk (8.11.6/8.11.6) with ESMTP id gB3ELdp27624 for ; Tue, 3 Dec 2002 14:21:49 GMT Received: from fourtytwo.gamesoc (localhost.gamesoc [127.0.0.1]) by fourtytwo.gamesoc (8.12.6/8.12.6) with ESMTP id gB3EMpLU000239 for ; Tue, 3 Dec 2002 14:22:51 GMT (envelope-from bruce@fourtytwo.gamesoc) Received: (from bruce@localhost) by fourtytwo.gamesoc (8.12.6/8.12.6/Submit) id gB3EMpWV000238 for current@freebsd.org; Tue, 3 Dec 2002 14:22:51 GMT Date: Tue, 3 Dec 2002 14:22:51 +0000 From: Bruce Cran To: current@freebsd.org Subject: ATA/ATAPI related panic Message-ID: <20021203142251.GA156@fourtytwo.gamesoc> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.5.1i 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 A few days ago I installed DP2, and also tried the 20021129 snapshot. I found a few problems with -current on my computer. I've had a problem with my DVD and CDRW drives under FreeBSD from 4.5 onwards. In released prior to 4.7 I used to get panics, I think when the system had heavy I/O loads, such as when building world - I thought this may have been due to the VIA controller. I installed DP2, and on one of the boots, got: acd0: REQUEST_SENSE command timeout - resetting ata1: resetting devices Fatal trap 12: page fault while in kernel mode fault trap address = 0x0 fault code = Supervisor read, page not present instruction pointer = 0x8:0xc0151ca2 stack pointer = 0x10:0xd68e3c5c frame pointer = 0x10:0xd68e3c70 code segment = base 0x0, limit 0xfffff, type = 0x16 = DPL 0, pres 1, def32 l, gran 1 processor eflags = interrupt enable resume, IOPL=0 current process = 13 (swi6:clock) trap number = 12 panic : page fault The DVD drive is a UDMA33 LG CRD-8080B on secondary master and the CDRW drive is a LG CED-8080B on secondary slave. It's connected to a Asus A7V 266-E motherboard which has a VIA 8233 ATA controller. Usually I get either REQUEST_SENSE or READ_CAPACITY, sometimes followed by a 'data read overrun' message. This happens both on -stable and -current. Usually there's a pause for a few seconds before the system boots as normal. The drives are initially configured for PIO4,PIO4, although I can manually set them to UDMA33,WDMA2. The strange thing is that the problem happens at random - sometimes the system will boot normally with no errors, other times it will stop for a minute with a few of those timeout/resetting messages. When I disconnected all the drives from the secondary controller, booting stopped for about 5 seconds before detecting ata1 io and irq. If I disconnect the DVD drive the problem disappears, although the CDRW is still only configured for PIO4 mode. I had hoped that -current had sorted this problem, but apparently not. Are the LG drives one of the ones which don't implement the ATAPI protocol correctly? I've got limited time at the moment, but can do a few more tests if needed, since the -current installation is just for testing. Another problem I've come across is that the bktr driver in 20021129 appears to be broken - specifically, in bktr_core.c it complains that IICBUS is not a member of the structure. Another possible bug I've found is in df. I compiled a kernel then, before overwriting the old backup, I tried running cp -ivR kernel.old kernel.old.orig, forgetting that /boot wouldn't have enough free space. When I next ran 'df -h' /boot was reporting -6MB free. I deleted /boot/kernel.old.orig and the free space was correctly reported again, but is this a bug in df? My filesytem is UFS1. I have also tried compiling a custom kernel with viapm, iicbus and smbus support, but on booting I get a message saying fsck exited on signal 8. There is some disk activity for a few seconds, but nothing else happens. I can press Ctrl-Alt-Delete and the messages appear saying the computer's rebooting, but no login prompt ever appears - the only option is to reboot. Another kernel-related query I have is about CPU throttling. I see a message saying my Athlon XP can be throttled in 16 steps from 100% to 6.2%, but how would I go about throttling it? Are there any IOCTLs, or is this feature only for laptops where they automatically get throttled when running on batteries? I know it probably doesn't make sense on a desktop computer, but I'm interested - ACPI support seems brilliant, and although hibernation doesn't seem to work, all the other features work perfectly. -- Bruce Cran To Unsubscribe: send mail to majordomo@FreeBSD.org with "unsubscribe freebsd-current" in the body of the message