From mboxrd@z Thu Jan 1 00:00:00 1970 Message-ID: <4A7E57B0.6080602@gmail.com> Date: Sun, 09 Aug 2009 07:59:28 +0300 From: Artem Bityutskiy MIME-Version: 1.0 To: news Subject: Re: UBIFS Corrupt during power failure References: <1239979018.3390.298.camel@localhost.localdomain> <200905150916.54091.sr@denx.de> <1244016510.5847.38.camel@localhost.localdomain> <1244369819.5847.321.camel@localhost.localdomain> <4A6EE895.9050008@gmx.de> In-Reply-To: <4A6EE895.9050008@gmx.de> Content-Type: text/plain; charset=UTF-8; format=flowed Content-Transfer-Encoding: 8bit Cc: Eric Holmberg , dedekind@infradead.org, Jamie Lokier , linux-mtd@lists.infradead.org, Stefan Roese , Adrian Hunter List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On 07/28/2009 03:01 PM, news wrote: > We have similar problems with a SPANSION falsh (S29GL01GP). > I think the reason of the problem is a feature of the chip. > > I reduced the problem to the MTD access (without ubi/ubifs). > We noticed toggle flash-bit(s) after power off during a write cycle. > The toggle flash-bit(s) may occure after power of during an sector-erase > too. > > Simple testsequence: > * flash_erase ... > * cp testfile /dev/mtd0 > - automatic or manuel power off during the cp > Check the flash after reboot (e.g md5sum /dev/mtd0 helps). > > We used the default settings from the CFI (MaxBufWriteSize=6 == 64 byte > buffer). This was fixed. See the latest UBI/UBIFS git trees. The fixes are also in the back-port trees: http://www.linux-mtd.infradead.org/doc/ubifs.html#L_source -- Best Regards, Artem Bityutskiy (Артём Битюцкий)