From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp1040.oracle.com ([141.146.126.69]:18822 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1750998AbdHaUCz (ORCPT ); Thu, 31 Aug 2017 16:02:55 -0400 Date: Thu, 31 Aug 2017 13:02:44 -0700 From: "Darrick J. Wong" Subject: Re: [PATCH] xfs: remove experimental tag for reflinks Message-ID: <20170831200244.GN3775@magnolia> References: <20170830145400.3681-1-hch@lst.de> <20170831064033.GF3775@magnolia> <20170831124321.GA21939@bfoster.bfoster> <20170831133019.GA5468@lst.de> <20170831153148.GJ3775@magnolia> <4b7dda9e-f470-65b6-08d0-e862dc967cd4@redhat.com> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline In-Reply-To: <4b7dda9e-f470-65b6-08d0-e862dc967cd4@redhat.com> Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Eric Sandeen Cc: Christoph Hellwig , Brian Foster , linux-xfs@vger.kernel.org, Dave Chinner On Thu, Aug 31, 2017 at 10:55:31AM -0500, Eric Sandeen wrote: > On 8/31/17 10:31 AM, Darrick J. Wong wrote: > > On Thu, Aug 31, 2017 at 03:30:19PM +0200, Christoph Hellwig wrote: > >> On Thu, Aug 31, 2017 at 08:43:21AM -0400, Brian Foster wrote: > >>> FWIW, I don't really have a strong opinion. To me, removing experimental > >>> means we feel the code has stabilized long enough in principle, there > >>> are no significant problems (i.e., corruption/crash vectors) that we are > >>> aware of and the feature is complete (full userspace tool support, etc). > >>> The in-core extent list thing seems like more of a general problem to me > >> > >> Agreed so far. > > > > Dave? Eric? Any perspective you'd like to offer? :) > > This is a bit of a naiive question, but how many applications are out > there that can be used with reflink right now? This would obviously Anything that uses cp --reflink, which could be any number of deploy scripts to VM "snapshot" tools. ISTR that cifs & nfs now both support reflink too, and now that ocfs2 also supports the ficlone* ioctls (sort of) we can expect wider use. dupremove (offline dedupe tool for btrfs) can also talk to xfs/ocfs2. > delay things a bit, but I had considered writing something up for > LWN or Fedora Planet or $WHATEVER describing these new xfs features, > how they can be used, and encourage some early-adopter testing. Try > to get some buzz going and some real-world use. I know there's already some dedicated experimenters using reflink... > It's always a catch 22; nobody uses it until it's marked stable, > but we never know if it's really stable until people outside the > development community use it. ;) > > As for the allocation issues w/ the in core extent list, yeah, that > worries me. Yeah. On the one hand it's a pre-existing condition that's orthogonal to reflink (and hey, we even put in cowextsize to mitigate the problems on VMs) but OTOH it probably gets worse with cow. --D > > -Eric > -- > To unsubscribe from this list: send the line "unsubscribe linux-xfs" in > the body of a message to majordomo@vger.kernel.org > More majordomo info at http://vger.kernel.org/majordomo-info.html