From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp1040.oracle.com ([141.146.126.69]:16724 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751445AbbKISto (ORCPT ); Mon, 9 Nov 2015 13:49:44 -0500 Date: Mon, 9 Nov 2015 10:49:13 -0800 From: "Darrick J. Wong" To: Christoph Hellwig Cc: david@fromorbit.com, fstests@vger.kernel.org, linux-ext4@vger.kernel.org, Anna.Schumaker@netapp.com, linux-btrfs@vger.kernel.org, xfs@oss.sgi.com Subject: Re: [RFCv3 00/12] xfstests: test the btrfs/xfs reflink/dedupe ioctls Message-ID: <20151109184913.GB3255@birch.djwong.org> References: <20151007051257.3260.73072.stgit@birch.djwong.org> <20151109075926.GB17974@infradead.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii In-Reply-To: <20151109075926.GB17974@infradead.org> Sender: linux-btrfs-owner@vger.kernel.org List-ID: On Sun, Nov 08, 2015 at 11:59:26PM -0800, Christoph Hellwig wrote: > On Tue, Oct 06, 2015 at 10:12:57PM -0700, Darrick J. Wong wrote: > > * I don't have any interesting NFS/CIFS setups for test. :( > > I have a banrch with client and server support for NFSv4.2 CLONE > support: > > http://git.infradead.org/users/hch/pnfs.git/shortlog/refs/heads/reflink+clone > > For now you want to use btrfs on the server, as using reflinks on XFS > seems to be a little unstable over NFS. I found a few more bugs in the kernel-side implementation, which might explain that. I'm about to start working on making CoW less crappy, but I'll push all the patches out to github. (I wasn't planning on patchbombing again until December.) > > If you're going to start using this mess, you probably ought to just > > pull from my github trees for kernel[1], xfsprogs[2], and xfstests[3]. > > They should just work with the btrfs that's in 4.3. > > > > Comments and questions are, as always, welcome. > > Any reason the groups are called clone? I don't really have an opinion > on clone vs reflink but given that the xfs_io command is reflink I'd > rather be consistent. The existing btrfs reflink tests were tagged in the 'clone' group prior to my patchset. > Otherwise I'd say get it merged ASAP, we can still fix up various > details later. I'll merge your patch and repost the whole pile of tests. I'm almost ready to send a pile of updates for the XFS on-disk structure document which add stuff about the v5 format, rmapbt, and reflink. --D