From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from ptb-relay03.plus.net ([212.159.14.214]) by canuck.infradead.org with esmtps (Exim 4.42 #1 (Red Hat Linux)) id 1CgR57-0001Me-4U for linux-mtd@lists.infradead.org; Mon, 20 Dec 2004 12:12:26 -0500 From: "Gareth Bult (Encryptec)" To: Josh Boyer In-Reply-To: <1103560493.25853.31.camel@weaponx.rchland.ibm.com> References: <1103232128.15927.70.camel@squizzey.bult.co.uk> <41C1FDFF.3020308@us.ibm.com> <1103233669.15929.81.camel@squizzey.bult.co.uk> <1103288058.3018.4.camel@weaponx.rchland.ibm.com> <1103297611.15917.184.camel@squizzey.bult.co.uk> <41C3030D.8040607@us.ibm.com> <1103302002.15927.198.camel@squizzey.bult.co.uk> <41C31307.70909@us.ibm.com> <1103304360.15914.210.camel@squizzey.bult.co.uk> <41C318C8.4030702@us.ibm.com> <20041218160255.GA1083@wohnheim.fh-wedel.de> <1103560493.25853.31.camel@weaponx.rchland.ibm.com> Content-Type: text/plain; charset=iso-8859-7 Date: Mon, 20 Dec 2004 17:12:28 +0000 Message-Id: <1103562748.10800.62.camel@squizzey.bult.co.uk> Mime-Version: 1.0 Content-Transfer-Encoding: quoted-printable Cc: Linux MTD Subject: Re: JFFS2 mount time List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Mon, 2004-12-20 at 10:34 -0600, Josh Boyer wrote: > Ok, that's why I said "most". And if your El Cheapo hardware has built > in wear leveling, then doing wear leveling on top of that is always > questionable. It might not make things worse, but it's probably not > very efficient. Mmm, if the USB flash micro-controllers have build-in wear levelling, they'd be quite something .. (!)=20 > Now if you have El Super Cheapo hardware that explicitly states you nee= d > to do wear leveling, that's a different story ;). Here's a quote from a "Kingston" advert; "Controllers automatically lock out bad memory cells and move the data to avoid corruption. Controllers also automatically distribute write cycles across the flash cells to extend the life of the flash memory card" What this means exactly in real terms .. [?] (for example, do you lose some of your key to a bad block table, if so how much ???) Here's a bit from "Verbatim"; "For ultra-reliability, Verbatim=A2s new Store =A1n=A2 Go Pro drives feat= ure an on-board 32 bit ARM-7 microprocessor that manages I/O operations and many of the drive=A2s technical features. An advanced wear-leveling algorithm is used to distribute writes evenly among flash storage cells. By evenly distributing the writing, Store =A1n=A2 Go Pro achieves an unprecedented reliability rating of over one million write/erase cycles. Advanced Error Detection Code/Error Correction Code furthers the Store =A1n=A2 Go Pro=A2s immunity to failure." Interesting articles / links; http://www.digitalmediadesigner.com/articles/viewarticle.jsp?id=3D27727 http://www.techworld.com/features/index.cfm?fuseaction=3Ddisplayfeature&F= eatureID=3D413 In summary: lots of manufacturers seem to be quoting virtual / physical block mapping on the fly with integrated wear levelling ... (!) So maybe wear levelling is out of date at filesystem level ? The techworld link to the M-Systems chips looks interesting .. Can anyone with detailed flash experience comment ? Gareth. > That's because it was the only removable media for PCs for a long time.= =20 > At least to the non-geeks. I can remember buying spare hard drives and > carrying those from machine to machine if I needed to do big transfers > ;). >=20 > josh >=20