From owner-freebsd-stable@FreeBSD.ORG Wed Nov 17 22:18:52 2004 Return-Path: Delivered-To: freebsd-stable@freebsd.org Received: from mx1.FreeBSD.org (mx1.freebsd.org [216.136.204.125]) by hub.freebsd.org (Postfix) with ESMTP id EF37616A4D1 for ; Wed, 17 Nov 2004 22:18:52 +0000 (GMT) Received: from mout2.freenet.de (mout2.freenet.de [194.97.50.155]) by mx1.FreeBSD.org (Postfix) with ESMTP id 2500943D1F for ; Wed, 17 Nov 2004 22:18:52 +0000 (GMT) (envelope-from dhke@freenet.de) Received: from [194.97.50.135] (helo=mx2.freenet.de) by mout2.freenet.de with esmtpa (Exim 4.43) id 1CUY8Z-0001R9-Ct for stable@freebsd.org; Wed, 17 Nov 2004 23:18:51 +0100 Received: from p5091da87.dip.t-dialin.net ([80.145.218.135] helo=peter.home.wul) by mx2.freenet.de with esmtpa (ID dhke@freenet.de) (Exim 4.43 #13) id 1CUY8Y-0000SB-7H for stable@freebsd.org; Wed, 17 Nov 2004 23:18:51 +0100 Received: (qmail 771 invoked by uid 1001); 17 Nov 2004 22:18:48 -0000 Date: Wed, 17 Nov 2004 23:18:48 +0100 From: Peter Wullinger To: stable@freebsd.org Message-ID: <20041117221848.GA668@peter.home.wul> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-15 Content-Disposition: inline Content-Transfer-Encoding: 8bit User-Agent: Mutt/1.4.2.1i cc: Thomas Quinot cc: =?iso-8859-15?Q?S=F8ren?= Schmidt Subject: atapicam & Promise [maybe PR/73675] X-BeenThere: freebsd-stable@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Production branch of FreeBSD source code List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Wed, 17 Nov 2004 22:18:53 -0000 Hello, This is a strange, but non-critical one, and since I have no idea, where to start, I should probably pass this on to folks who have better knowledge of the code and hardware concerned: The normal (disfunctional) configuration is: atapci0 M \- ad0 S \- ad1 atapci1 (Via) M \- acd0 (dvdrom) S \- afd0 (zip) atapci2 (Promise) M \- acd1 (burner) When enabling atapicam and having my cd-burner attached to the third ata channel, the kernel is unable to boot and hangs with an interrupt storm. This happens on a GENERIC kernel as well as on my custom, stripped-down kernel. Increasing the detection delay for interrupt storms to hw.intr_storm_threshold=15000 does not help. The occurence of the problem is also independent from the fact, if the particular interrupt is shared between the Promise controller and other hardware. Controller (2.20.0.15) and burner (1.04) firmware are newest available. Switching to atapci0 M \- ad0 S \- ad1 atapci1 M \- acd1 (burner) S \- acd0 (dvdrom) atapci2 M \- afd0 (zip) however strangely remedies this problem. Cut-down dmesg output from a non-working kernel is attached below. This problem has surfaced in -CURRENT at around May, has persisted since and is still present in 5_3_RELENG. I'm willing to provide additional debugging information (if somebody tells me how to get some), test patches and the usual stuff, if anybody feels inclined to fix the problem. Cheers, Peter [cc to søren and thomas assuming they are the most reasonable targets] Motherboard: Asus A7M266 == dmesg == atapci0: port 0xd800-0xd80f,0x376,0x170-0x177,0x3f6,0x1f0-0x1f7 at device 4.1 on pci0 atapci1: port 0x8800-0x880f,0x9000-0x9003,0x9400-0x9407,0x9800-0x9803,0xa000-0xa007 mem 0xe4800000-0xe4803fff irq 10 at device 11.0 on pci0 ad0: 117800MB [239340/16/63] at ata0-master UDMA100 ad1: 117800MB [239340/16/63] at ata0-slave UDMA100 acd0: DVDROM at ata1-master UDMA33 afd0: REMOVABLE at ata1-slave BIOSPIO acd1: CDRW at ata2-master UDMA33 GEOM_VINUM: subdisk home.p1.s0 is up GEOM_VINUM: subdisk home.p0.s0 is up [...] interrupt storm == dmesg == -- Mut ist keine Vorbedingung für den Verlust des Lebens. -- Sir Peter Ustinov