Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id S932127AbbLNSAU (ORCPT ); Mon, 14 Dec 2015 13:00:20 -0500 Received: from a.ns.miles-group.at ([95.130.255.143]:11949 "EHLO radon.swed.at" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1753125AbbLNSAS (ORCPT ); Mon, 14 Dec 2015 13:00:18 -0500 Subject: Re: [PATCH 1/1] fs:ubifs:recovery:fixup UBIFS cannot recover master node issue To: =?UTF-8?B?QmVhbiBIdW8g6ZyN5paM5paMIChiZWFuaHVvKQ==?= , Artem Bityutskiy , Adrian Hunter , Brian Norris References: <566A9378.4070900@nod.at> Cc: "linux-mtd@lists.infradead.org" , "linux-kernel@vger.kernel.org" , Boris Brezillon From: Richard Weinberger X-Enigmail-Draft-Status: N1110 Message-ID: <566F03AC.90305@nod.at> Date: Mon, 14 Dec 2015 19:00:12 +0100 User-Agent: Mozilla/5.0 (X11; Linux x86_64; rv:38.0) Gecko/20100101 Thunderbird/38.4.0 MIME-Version: 1.0 In-Reply-To: Content-Type: text/plain; charset=gbk Content-Transfer-Encoding: 8bit Sender: linux-kernel-owner@vger.kernel.org List-ID: X-Mailing-List: linux-kernel@vger.kernel.org Content-Length: 2095 Lines: 49 Bean, Am 14.12.2015 um 04:55 schrieb Bean Huo ?????? (beanhuo): > Dear Richard > >> Bean, >> >> Am 11.12.2015 um 09:26 schrieb Bean Huo ?????? (beanhuo): >>> For MLC NAND, paired page issue is now a common known issue. >>> This patch is just for master node cannot be recovered while there >>> will two pages be damaged in one single master node block. >>> As for this patch, if there are more than one page data in master node >>> block being damaged, and as long as exist one uncorrupted master node >>> block, master node will be recovered. >> >> So, this patch is part if a larger patch series to make UBIFS MLC aware? > > No, this is not one part of my path series, just a single and dedicated to > Master node. [...] > Currently, we get more feedbacks from our customers who are using MLC NAND, > They more like UBIFS more reliable, Even can tolerate to discard some user > Data after next power on. Means that they don't want to UBIFS mount failed just > Because of power loss, If to discard the data for the stability of the system, they > prefer to choose the latter. MLC is currently simply not supported. If your hardware does not have a mechanism do temper power-loss the paired page issue will damage UBI and UBIFS. Please correct me if I'm wrong but this patch just papers over one symptom of that. > For UBIFS master node on MLC NAND, I often found that one of master node block is OK, > But because of second master node block exist two pages damaged data, recovery always > Fails. Not matter SLC or MLC, as long as there is a good master node, recovery must be > Successful. This needs a much more detailed explanation. In which scenarios on SLC NAND can you get such an unmountable UBIFS? Maybe UBIFS is too strict and NAND behaves differently than UBIFS expects but we need to understand it in depth. Thanks, //richard -- To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/