From owner-freebsd-current@FreeBSD.ORG Tue May 3 11:25:55 2005 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 9AB6E16A4CE for ; Tue, 3 May 2005 11:25:55 +0000 (GMT) Received: from asterix.rsu.ru (asterix.rsu.ru [195.208.245.250]) by mx1.FreeBSD.org (Postfix) with ESMTP id 995EE43D2D for ; Tue, 3 May 2005 11:25:54 +0000 (GMT) (envelope-from os@rsu.ru) Received: from os.adsl.r61.net (os@os.adsl.r61.net [195.208.243.95]) (authenticated bits=0) by asterix.rsu.ru (8.13.3/8.13.3) with ESMTP id j43BPW0W092507 (version=TLSv1/SSLv3 cipher=DHE-RSA-AES256-SHA bits=256 verify=NOT); Tue, 3 May 2005 15:25:32 +0400 (MSD) (envelope-from os@rsu.ru) Date: Tue, 3 May 2005 15:25:36 +0400 (MSD) From: Oleg Sharoiko To: Jeff Roberson In-Reply-To: <20050501061323.X71837@mail.chesapeake.net> Message-ID: <20050503151131.R787@localhost> References: <20050501034520.N71837@mail.chesapeake.net> <20050501061323.X71837@mail.chesapeake.net> MIME-Version: 1.0 Content-Type: TEXT/PLAIN; charset=US-ASCII X-Virus-Scanned: ClamAV version 0.81, clamav-milter version 0.81b on asterix.rsu.ru X-Virus-Status: Clean X-Spam-Status: No, score=-5.4 required=5.0 tests=ALL_TRUSTED,AWL,BAYES_00 autolearn=ham version=3.0.2 X-Spam-Checker-Version: SpamAssassin 3.0.2 (2004-11-16) on asterix.rsu.ru X-Mailman-Approved-At: Tue, 03 May 2005 11:42:11 +0000 cc: freebsd-current@freebsd.org cc: Kris Kennaway Subject: Re: diskless/unionfs panics 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: Tue, 03 May 2005 11:25:55 -0000 On Sun, 1 May 2005, Jeff Roberson wrote: JR>Yes, unfortunately this is only part of the problem. The v_object can JR>change on a unionfs vnode leading to inconsistent results for mmap/exec. JR>This also causes problems for unionfs mounted on top of unionfs as the JR>lower vp can also change v_object fields. This was one advantage to the JR>VOP based solution. It would propagate all the way down the stack on each JR>access. So do I get it right that the problems can happen if file being execed or mmaped over unionfs will at the same time be changed on ufs (assuming that ufs was used to mount unionfs)? Does this mean that nullfs also have the same problem? But how can this problem be solved? And what is VOP based solution? -- Oleg Sharoiko. Software and Network Engineer Computer Center of Rostov State University.