From owner-freebsd-current@FreeBSD.ORG Mon Oct 13 02:28:30 2003 Return-Path: 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 67DF116A4B3 for ; Mon, 13 Oct 2003 02:28:30 -0700 (PDT) Received: from spider.deepcore.dk (cpe.atm2-0-56339.0x50c6aa0a.abnxx2.customer.tele.dk [80.198.170.10]) by mx1.FreeBSD.org (Postfix) with ESMTP id 3EDF343FA3 for ; Mon, 13 Oct 2003 02:28:28 -0700 (PDT) (envelope-from sos@spider.deepcore.dk) Received: from spider.deepcore.dk (localhost [127.0.0.1]) by spider.deepcore.dk (8.12.10/8.12.10) with ESMTP id h9D9Rgsi009734; Mon, 13 Oct 2003 11:27:42 +0200 (CEST) (envelope-from sos@spider.deepcore.dk) Received: (from sos@localhost) by spider.deepcore.dk (8.12.10/8.12.10/Submit) id h9D9RfcJ009733; Mon, 13 Oct 2003 11:27:41 +0200 (CEST) (envelope-from sos) From: Soren Schmidt Message-Id: <200310130927.h9D9RfcJ009733@spider.deepcore.dk> In-Reply-To: <20031012214149.GA14931@dan.emsphone.com> To: Dan Nelson Date: Mon, 13 Oct 2003 11:27:41 +0200 (CEST) X-Mailer: ELM [version 2.4ME+ PL99f (25)] MIME-Version: 1.0 Content-Transfer-Encoding: 8bit Content-Type: text/plain; charset=ISO-8859-1 X-mail-scanned: by DeepCore Virus & Spam killer v1.3 cc: freebsd-current@FreeBSD.ORG cc: Robert Ferguson Subject: Re: boot hang: "ata1: resetting devices .. done" (5.1-CURRENT, IBM T30) X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.1 Precedence: list List-Id: Discussions about the use of FreeBSD-current List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , X-List-Received-Date: Mon, 13 Oct 2003 09:28:30 -0000 X-List-Received-Date: Mon, 13 Oct 2003 09:28:30 -0000 It seems Dan Nelson wrote: > In the last episode (Oct 08), Robert Ferguson said: > > I see this problem as well. I'm running on a T40, with a DVD/CDRW in > > the ultrabay, and -CURRENT as of this morning. At boot, it hangs > > immediately after > > > > ad0: 35174MB [71465/16/63] at ata0-master UDMA100 > > ata1: resetting devices .. > > done > > > > Backing out the most recent checkin to sys/dev/ata/ata-queue.c (i.e. > > reverting to version 1.6) makes the problem go away. > > I upgraded from an Oct 1 -> Oct 12 kernel and saw the same hang. > Backing out r1.6 fixed it for me too. Hmm, that probably because I disabled the code that tries to deal with lost interrupts (its no good and can easily panic your box, thats why). We seem to be suffering from lost interrupts in -current for some time. I can on at least one system here confirm that the interrupt does get delivered from the device, but the ATA interrupt handler is newer called. Where the interrupt gets lost and why is still beyond me, anyone that can share some light on this is most welcome... -Søren