From: Marc Lehmann Subject: Re: [PATCH] f2fs: convert inline_data for f2fs_fiemap Date: Sat, 17 Oct 2015 00:57:13 +0200 Message-ID: <20151016225713.GB4666@schmorp.de> References: <1445011637-16015-1-git-send-email-jaegeuk@kernel.org> Mime-Version: 1.0 Content-Type: text/plain; charset="us-ascii" Content-Transfer-Encoding: 7bit Return-path: Received: from sog-mx-4.v43.ch3.sourceforge.com ([172.29.43.194] helo=mx.sourceforge.net) by sfs-ml-4.v29.ch3.sourceforge.com with esmtp (Exim 4.76) (envelope-from ) id 1ZnDwA-0005u7-Je for linux-f2fs-devel@lists.sourceforge.net; Fri, 16 Oct 2015 22:57:22 +0000 Received: from mail.nethype.de ([5.9.56.24]) by sog-mx-4.v43.ch3.sourceforge.com with esmtps (TLSv1:AES128-SHA:128) (Exim 4.76) id 1ZnDw8-0003kX-AW for linux-f2fs-devel@lists.sourceforge.net; Fri, 16 Oct 2015 22:57:22 +0000 Received: from [10.0.0.5] (helo=doom.schmorp.de) by mail.nethype.de with esmtp (Exim 4.84) (envelope-from ) id 1ZnDw1-000702-Iu for linux-f2fs-devel@lists.sourceforge.net; Fri, 16 Oct 2015 22:57:13 +0000 Received: from [10.0.0.1] (helo=cerebro.laendle) by doom.schmorp.de with esmtp (Exim 4.84) (envelope-from ) id 1ZnDw1-0000kD-Eb for linux-f2fs-devel@lists.sourceforge.net; Fri, 16 Oct 2015 22:57:13 +0000 Received: from root by cerebro.laendle with local (Exim 4.84) (envelope-from ) id 1ZnDw1-0001JM-E3 for linux-f2fs-devel@lists.sourceforge.net; Sat, 17 Oct 2015 00:57:13 +0200 Content-Disposition: inline In-Reply-To: <1445011637-16015-1-git-send-email-jaegeuk@kernel.org> List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , Errors-To: linux-f2fs-devel-bounces@lists.sourceforge.net To: linux-f2fs-devel@lists.sourceforge.net On Fri, Oct 16, 2015 at 09:07:17AM -0700, Jaegeuk Kim wrote: > For fiemap, we should convert inline_data for user. Sorry for not answeriung your other mails, I'm busy, but I just saw this patch: > + /* we don't need to use inline_data strictly */ > + if (f2fs_has_inline_data(inode)) { > + ret = f2fs_convert_inline_inode(inode); If that means that fiemap is no longer read-only but changes the layout, by moving the data out of the inode, it's buggy - the correct way is to flag those extents with FIEMAP_EXTENT_DATA_INLINE (and/or possibly other flags). I didn't investigate the code, so maybe I am completely off here, just guessing, really. If not, fiemap really is there to map the data extents, not to de-optimise the filesystem. It's especially bad as fiemap is used by a lot of tools to map disk layout. If I am wrong, then please ignore my random mail here :) -- The choice of a Deliantra, the free code+content MORPG -----==- _GNU_ http://www.deliantra.net ----==-- _ generation ---==---(_)__ __ ____ __ Marc Lehmann --==---/ / _ \/ // /\ \/ / schmorp@schmorp.de -=====/_/_//_/\_,_/ /_/\_\ ------------------------------------------------------------------------------