From mboxrd@z Thu Jan 1 00:00:00 1970 From: Chris Wedgwood Subject: Re: [RFC] support multiple max offset limits for a single superblock Date: Fri, 30 Nov 2007 09:14:14 -0800 Message-ID: <20071130171414.GA5048@puku.stupidest.org> References: <47503BB5.2010501@sandeen.net> Mime-Version: 1.0 Content-Type: text/plain; charset=us-ascii Cc: linux-fsdevel , ext4 development To: Eric Sandeen Return-path: Received: from smtp124.sbc.mail.sp1.yahoo.com ([69.147.64.97]:37515 "HELO smtp124.sbc.mail.sp1.yahoo.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with SMTP id S1762983AbXK3R00 (ORCPT ); Fri, 30 Nov 2007 12:26:26 -0500 Content-Disposition: inline In-Reply-To: <47503BB5.2010501@sandeen.net> Sender: linux-fsdevel-owner@vger.kernel.org List-Id: linux-fsdevel.vger.kernel.org On Fri, Nov 30, 2007 at 10:35:01AM -0600, Eric Sandeen wrote: > I could do the same for ext4, but Chris Mason prodded me to think of something > more generic... this is what I came up with. I think it (for now) should remain in the few file-systems that are affected to avoid making a new ->i_op right now. If it turns out this is a more commonly need in the future we could do this, but it seems a bit heavy handy just yet given at present. Also, it might turn at that it makes sense to to on-the-fly conversion in the filesystem (where applicable) at some later stage, so this wouldn't be useful there. I assume it's probably hard to convert a large bitmap-based ext4 file to an extent-based one (how large can a transaction be?) but that might not always be the case and it's probably fine for smaller files. > (BTW another option would be to convert old-format files when > accessed, but that has its own set of tradeoffs...) I think doing that where possible makes more sense. As I said above I'm not sure how easy that is to do for all files, but then again fsck could take care of those if prodded.