From: Mark Fasheh Subject: Re: [PATCH 1/5] vfs: vfs-level fiemap interface Date: Wed, 28 May 2008 13:12:40 -0700 Message-ID: <20080528201240.GE8325@wotan.suse.de> References: <20080525000157.GK8325@wotan.suse.de> <20080528194215.GI7263@webber.adilger.int> <20080528195429.GC23090@unused.rdu.redhat.com> Reply-To: Mark Fasheh Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: Andreas Dilger , linux-fsdevel@vger.kernel.org To: Josef Bacik Return-path: Received: from mx1.suse.de ([195.135.220.2]:57780 "EHLO mx1.suse.de" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1752655AbYE1UMl (ORCPT ); Wed, 28 May 2008 16:12:41 -0400 Content-Disposition: inline In-Reply-To: <20080528195429.GC23090@unused.rdu.redhat.com> Sender: linux-fsdevel-owner@vger.kernel.org List-ID: On Wed, May 28, 2008 at 03:54:29PM -0400, Josef Bacik wrote: > > > Btw, how many block-based file systems that don't return multiple blocks via > > > get_blocks() are there that we actually care about enough to write this > > > code? > > > > That I have no clue about. Joseph? > > > > > > Well I can't comment on the "care enough about" part, but looking at it I don't > think reiserfs does, it just maps the one block we ask for. Just cursory > looking through stuff it looks like most other stuff does map as much as it can, > even FAT does. Thanks much, Josef, thanks for looking. Isn't reiserfs extent based anyway? So really, it should (eventually) get a real ->fiemap callback. --Mark -- Mark Fasheh