From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S1759269AbaJaK5m (ORCPT ); Fri, 31 Oct 2014 06:57:42 -0400 Received: from a.ns.miles-group.at ([95.130.255.143]:65275 "EHLO radon.swed.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1757338AbaJaK5l (ORCPT ); Fri, 31 Oct 2014 06:57:41 -0400 Message-ID: <54536B22.4000807@nod.at> Date: Fri, 31 Oct 2014 11:57:38 +0100 From: Richard Weinberger User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:31.0) Gecko/20100101 Thunderbird/31.1.0 MIME-Version: 1.0 To: hujianyang CC: dedekind1@gmail.com, dwmw2@infradead.org, computersforpeace@gmail.com, linux-mtd@lists.infradead.org, linux-kernel@vger.kernel.org Subject: Re: [PATCH] UBI: vtbl: Use ubi_eba_atomic_leb_change() References: <1414259021-5691-1-git-send-email-richard@nod.at> <1414659349.23185.27.camel@sauron.fi.intel.com> <54530A29.9070208@huawei.com> <545343CA.7080507@nod.at> <54536834.1020005@huawei.com> In-Reply-To: <54536834.1020005@huawei.com> Content-Type: text/plain; charset=windows-1252 Content-Transfer-Encoding: 7bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Am 31.10.2014 um 11:45 schrieb hujianyang: > This question is basing on your comment for this patch: > > """ > we can guarantee that the first VTBL record is always > correct and we don't really need the second one anymore. > """ > > I think that means one PEB is enough in your considering. > So I want to know if you are sure about this. Because > we use two leb for master_node in ubifs-level. So maybe > VTBL is like super_node, not master_node, right? Yes, technically one PEB is enough if atomic leb change is used. But existing UBI implementations want a second one and a backup VTBL PEB is good for robustness. i.e. if the PEB turns bad we have a backup and do not lose all volume meta information. Thanks, //richard