[PATCH] [MTD] ROMBLOCK: Readonly Block Device Layer Over MTD

Artem Bityutskiy dedekind at infradead.org
Sun Dec 16 07:45:40 EST 2007


On Sun, 2007-12-16 at 12:46 +0100, Gregory CLEMENT wrote:
> I send this mail again as my previous patch was mangled, I hope that now 
> I fixed it.
> 
> Hi,
> I generated this patch for linux 2.6.24.rc5 and tested it with cramfs
> and squashfs file systems.
> As Konstantin Baydarov submit this patch first maybe it would be better
> than he sign this patch, but if he can't or don't want it, I don't mind
> sign it...
> 
> >From 9a06e0c1818e4b27a04a8adac7eee551c8aac115 Mon Sep 17 00:00:00 2001
> From: Gregory CLEMENT <gclement00 at gmail.com>
> Date: Wed, 12 Dec 2007 17:41:23 +0100
> Subject: [PATCH] [MTD] BLOCK_RO: Readonly Block Device Layer Over MTD
> 
> Description:
> The following patch adds readonly block device layer over mtd
> that allows to use any filesystem on this device in RO mode and thus
> gain faster mount times and better throughput rates.
> It allows to put one RO filesystem into NAND using a bad-block aware
> program initially, and mount it.  But be aware, it doesn't handle run-
> time bit flips or errors (even if this event seems pretty rare).
> 
> How it works:
> Blocks translation routine was added to read sector function. Assuming
> that bad block won't appear during MTD reading and BBT is correct, bad
> block is skipped and requested block is lazily mapped to good one.
> Block driver based on the mtd readonly device driver mtdblock_ro.c and
> translation routine was taken from the patch of Pantelis Antoniou
> (which can be found at
> http://lists.infradead.org/pipermail/linux-mtd/2004-May/009672.html).
> This patch was originally submit by Konstantin Baydarov, but never
> included nor rejected ( see
> http://lists.infradead.org/pipermail/linux-mtd/2006-November/016835.html).
> 
> Signed-off-by: Gregory CLEMENT <gclement00 at gmail.com>

For me this amount of mtdblock drivers looks strange. We have 2 already,
and this is the third. 

If you really cannot modify mtdblock_ro, then I'd kindly ask you to also
write a good explanation: what are all those mtdblock/rombolck, what is
the difference, why we need so many of them. Then I would put your text
to the MTD web site.

-- 
Best regards,
Artem Bityutskiy (Битюцкий Артём)




More information about the linux-mtd mailing list