From mboxrd@z Thu Jan 1 00:00:00 1970 From: Erik Mouw Subject: Re: How To Recover Files From ext3 Partition?? Date: Mon, 8 May 2006 14:34:08 +0200 Message-ID: <20060508123408.GE1875@harddisk-recovery.com> References: <20060504143814.GF16570@harddisk-recovery.com> <20060505051618.95519.qmail@web37902.mail.mud.yahoo.com> <20060505111807.GD4900@harddisk-recovery.com> <20060505164147.GK6075@schatzie.adilger.int> <1147085495.5331.5.camel@sisko.sctweedie.blueyonder.co.uk> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andreas Dilger , UZAIR LAKHANI , linux-fsdevel@vger.kernel.org Return-path: Received: from dtp.xs4all.nl ([80.126.206.180]:18228 "HELO abra2.bitwizard.nl") by vger.kernel.org with SMTP id S932082AbWEHMeK (ORCPT ); Mon, 8 May 2006 08:34:10 -0400 To: "Stephen C. Tweedie" Content-Disposition: inline In-Reply-To: <1147085495.5331.5.camel@sisko.sctweedie.blueyonder.co.uk> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Mon, May 08, 2006 at 11:51:34AM +0100, Stephen C. Tweedie wrote: > On Fri, 2006-05-05 at 10:41 -0600, Andreas Dilger wrote: > > There is another mechanism ext3 could potentially use, wherein it > > walks the whole inode in advance of the truncate and creates a > > (potentially) very large transaction handle to do the bitmap updates > > in a single shot (and also reducing the amount of IO needed for an > > unlink by 96%), but nobody has ever cared enough about it to work on > > implementing this. > > Trouble is, there's no guarantee that that transaction would actually > fit into the journal. Most of the time it will, but if it doesn't, then > we deadlock or risk data corruption. Is there some way to determine in advance if a transaction fits into the journal? If so, we could unlink using Andreas's proposal if the journal allows, and fall back to the old method if not. Erik -- +-- Erik Mouw -- www.harddisk-recovery.com -- +31 70 370 12 90 -- | Lab address: Delftechpark 26, 2628 XH, Delft, The Netherlands