From: Jan Kara Subject: Re: Help on Implementation of EXT3 type Ordered Mode in EXT4 Date: Tue, 9 Feb 2010 17:05:22 +0100 Message-ID: <20100209160522.GE15318@atrey.karlin.mff.cuni.cz> References: Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-ext4@vger.kernel.org To: Kailas Joshi Return-path: Received: from atrey.karlin.mff.cuni.cz ([195.113.26.193]:60970 "EHLO atrey.karlin.mff.cuni.cz" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1754411Ab0BIQFY (ORCPT ); Tue, 9 Feb 2010 11:05:24 -0500 Content-Disposition: inline In-Reply-To: Sender: linux-ext4-owner@vger.kernel.org List-ID: Hi, > I recently found that in EXT4 with delayed block the Ordered mode does not > bahave same as in EXT3. > I found a patch for this at http://lwn.net/Articles/324023/, but it has some > journal block estimation problem resulting into deadlock. > > I would like to know if it has been solved. > If not, is it possible to solve it? What are the complexities involved? It has not been solved. The problem is that to commit data on transaction commit (which is what data=ordered mode has historically done), you have to allocate space for these blocks. But that allocation needs to modify a filesystem and thus journal more blocks... And that is tricky - we would have to reserve space in the current transaction for allocation of delayed data. So it gets a bit messy... Why exactly do you need the old data=ordered guarantees? Honza -- Jan Kara SuSE CR Labs