From mboxrd@z Thu Jan 1 00:00:00 1970 From: Willy Tarreau Subject: Re: [PATCH] LogFS take three Date: Wed, 16 May 2007 07:22:54 +0200 Message-ID: <20070516052253.GI943@1wt.eu> References: <20070515151919.GA32510@lazybastard.org> <20070515133759.9ee434a2.akpm@linux-foundation.org> <20070516000630.GD1220@lazybastard.org> Mime-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Transfer-Encoding: QUOTED-PRINTABLE Cc: Andrew Morton , linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, linux-mtd@lists.infradead.org, Albert Cahalan , Thomas Gleixner , Jan Engelhardt , Evgeniy Polyakov , Pekka Enberg , Greg KH , Ingo Oeser To: =?iso-8859-1?Q?J=F6rn?= Engel Return-path: Content-Disposition: inline In-Reply-To: <20070516000630.GD1220@lazybastard.org> Sender: linux-kernel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Wed, May 16, 2007 at 02:06:31AM +0200, J=F6rn Engel wrote: > On Tue, 15 May 2007 13:37:59 -0700, Andrew Morton wrote: > > It's strange and a bit regrettable that an fs would have dependency= on MTD, > > really. >=20 > It is and changing this wouldn't be too hard. All device access goes > through five functions (read, write, erase, is_bad and mark_bad). As > soon as someone seriously cares I will add a struct logfs_device_ops = and > have a second set of these functions for block devices. >=20 > On hard disks it shouldn't make too much sense. The filesystem will > fragment like a splinter bomb and be just as popular. On hard disks, yes, but as you suggested, there are lots of other flash devices interfaced as block devices. CompactFlash comes to mind, USB keys too. And on these ones, the most important is to reduce the number of writes and to support large sizes. I already see LogFS as an interes= ting alternative to JFFS2 on such devices, eventhough it does not (yet?) sup= port compression. Regards, Willy