From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from smtp.nokia.com ([147.243.128.26] helo=mgw-da02.nokia.com) by canuck.infradead.org with esmtps (Exim 4.72 #1 (Red Hat Linux)) id 1Pruzn-0000Vj-Ej for linux-mtd@lists.infradead.org; Tue, 22 Feb 2011 16:21:52 +0000 Date: Tue, 22 Feb 2011 18:21:43 +0200 From: Jarkko Lavinen To: Artem Bityutskiy Subject: Re: [PATCH 3/3] mtd: Add mtdswap block driver Message-ID: <20110222162143.GA13529@angel.research.nokia.com> References: <1297692971-7399-1-git-send-email-jarkko.lavinen@nokia.com> <1297692971-7399-4-git-send-email-jarkko.lavinen@nokia.com> <1298038470.5623.50.camel@localhost> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <1298038470.5623.50.camel@localhost> Cc: linux-mtd@lists.infradead.org Reply-To: Jarkko Lavinen List-Id: Linux MTD discussion mailing list List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , On Fri, Feb 18, 2011 at 04:14:30PM +0200, ext Artem Bityutskiy wrote: > ... what's the problem with re-trying the erase operation - it > looks like over-design. > Is it please possible to remove this config option? I have a Hynix SLC OneNand spec which says an erase block with an erase failure or a write error should not be re-written or erased again but instead just garbage collected and marked bad. If I switch permanently to the strict erase failure handling, bad block marking could be too sensitive for some devices expecting retries. If I switch permanently to the lenient mode, mtdswap would retry failed erases which would be against chip spec of at least the one OneNand chip I am working with. If I remove the config option it would then mean strict handling only. Jarkko