From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: (majordomo@vger.kernel.org) by vger.kernel.org via listexpand id ; Fri, 17 May 2002 08:29:10 -0400 Received: (majordomo@vger.kernel.org) by vger.kernel.org id ; Fri, 17 May 2002 08:27:50 -0400 Received: from pc-62-31-66-178-ed.blueyonder.co.uk ([62.31.66.178]:32641 "EHLO sisko.scot.redhat.com") by vger.kernel.org with ESMTP id ; Fri, 17 May 2002 08:26:32 -0400 Date: Fri, 17 May 2002 13:26:25 +0100 From: "Stephen C. Tweedie" To: Tomas Szepe Cc: "Stephen C. Tweedie" , ext3-users@redhat.com, ext2-devel@lists.sourceforge.net, linux-fsdevel@vger.kernel.org, linux-kernel@vger.kernel.org, Andrew Morton , Andreas Dilger Subject: Re: [Ext2-devel] Re: Ext3-0.9.18 available Message-ID: <20020517132625.I2693@redhat.com> In-Reply-To: <20020516175637.A21624@redhat.com> <20020517121746.GA6613@louise.pinerecords.com> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline User-Agent: Mutt/1.2.5.1i Sender: linux-kernel-owner@vger.kernel.org X-Mailing-List: linux-kernel@vger.kernel.org Hi, On Fri, May 17, 2002 at 02:17:47PM +0200, Tomas Szepe wrote: > > ext3-0.9.18 is now available for 2.4.19-pre8. Some of the fixes in > > this release are already in the 2.4.19-pre8, but there are some > > important new fixes in the patch and users are encouraged to upgrade. > > This release fixes all known outstanding bug reports. > > Is there at least a remote possibility of the fixes getting ported > to be included in the linux-2.2 ext3 patch (the latest being 0.0.7a?)? No, there's no need. The changes come in several groups: Performance tweaks: * Speed up MS_SYNC writes * speed up fsyncs in non-journaled data modes a little Config/cosmetic tweaks: * Set up kjournald to be parented under init properly * config: ext3 is no longer experimental Fix bugs introduced at various stages during the 2.4 port: * don't consider ENOSPC a fatal error when allocating an inode * fix over-zealous ext3 complaint about locked buffers * fix O_SYNC * fix tiny race where a buffer could be written to disk too soon Fix bugs arising from changes made elsewhere during 2.4: * fix i_blocks getting inconsistent after disk full * fix "dump corrupts filesystems" core VFS bug * fix very rare buffer leak Fix a bug in the LVM interaction (code isn't present in 2.2): * fix LVM snapshot deadlock None of the bug-fixes apply to the 2.2 version of the code. Cheers, Stephen