From owner-freebsd-current@FreeBSD.ORG Fri May 14 18:37:44 2010 Return-Path: Delivered-To: freebsd-current@FreeBSD.org Received: from mx1.freebsd.org (mx1.freebsd.org [IPv6:2001:4f8:fff6::34]) by hub.freebsd.org (Postfix) with ESMTP id AADA5106566B for ; Fri, 14 May 2010 18:37:44 +0000 (UTC) (envelope-from TERRY@tmk.com) Received: from server.tmk.com (server.tmk.com [204.141.35.63]) by mx1.freebsd.org (Postfix) with ESMTP id 835068FC0A for ; Fri, 14 May 2010 18:37:44 +0000 (UTC) Received: from tmk.com by tmk.com (PMDF V6.4 #37010) id <01NN3XLDKJ28006UN1@tmk.com>; Fri, 14 May 2010 14:08:24 -0400 (EDT) Date: Fri, 14 May 2010 14:02:47 -0400 (EDT) From: Terry Kennedy In-reply-to: "Your message dated Fri, 14 May 2010 08:42:44 -0700" <06D5F9F6F655AD4C92E28B662F7F853E021D4D5E@seaxch09.desktop.isilon.com> To: Matthew Fleming Message-id: <01NN3XTJDNMS006UN1@tmk.com> MIME-version: 1.0 Content-type: TEXT/PLAIN; charset=iso-8859-1 References: <01NN32EOXMYC006UN1@tmk.com> <4BED3912.9080509@FreeBSD.org> <01NN3PQCOFHE006UN1@tmk.com> X-Mailman-Approved-At: Fri, 14 May 2010 21:23:40 +0000 Cc: freebsd-current@FreeBSD.org, freebsd-stable@FreeBSD.org, John Baldwin Subject: RE: Crash dump problem - sleeping thread owns a non-sleepable lock during crash dump write X-BeenThere: freebsd-current@freebsd.org X-Mailman-Version: 2.1.5 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: Fri, 14 May 2010 18:37:44 -0000 > Oops, youre right that other CPUs are running. > > The stop_cpus() call is only made if kdb is entered. doadump() is called > out of boot() which comes later. At Isilon weve been running with a patch > that does stop_cpus() pretty close to the front of panic(9). This is interesting, and changing the behavior will probably allow the crash dump for the original problem (repeatable crash in the bce driver) to be analyzed. At the moment, I'm more interested in dealing with the original problem of the crash in bce. Right now, I'm running this vendor's product under Linux compatibility mode. The vendor is hard at work building a native FreeBSD version of their product. One of two things is going to happen here: 1) the crash doesn't happen in native mode due to different code paths being taken, and I lose the ability to reproduce the crash when the box goes into production, or 2) the crash continues to happen and the ven- dor gets the impression FreeBSD is unstable and not worth supporting. I'd like to avoid that. So, any ideas on how to troubleshoot the panic in bce? Thanks, Terry Kennedy http://www.tmk.com terry@tmk.com New York, NY USA