Re: [PATCH v4 00/11] ocfs2: wire up {clone,copy,dedupe}_range

From: Al Viro <viro@ZenIV.linux.org.uk>
To: "Darrick J. Wong" <darrick.wong@oracle.com>
Cc: mfasheh@versity.com, akpm@linux-foundation.org,
	jlbec@evilplan.org, linux-fsdevel@vger.kernel.org,
	hch@infradead.org, ocfs2-devel@oss.oracle.com
Subject: Re: [PATCH v4 00/11] ocfs2: wire up {clone,copy,dedupe}_range
Date: Sat, 10 Dec 2016 23:43:25 +0000	[thread overview]
Message-ID: <20161210234325.GJ1555@ZenIV.linux.org.uk> (raw)
In-Reply-To: <148140257493.24628.6724911560073347399.stgit@birch.djwong.org>

On Sat, Dec 10, 2016 at 12:42:55PM -0800, Darrick J. Wong wrote:
> Hi all,
> 
> This is the fourth submission of a series of patches that wire up the
> existing ocfs2 reflinking capabilities to the new(ish) VFS
> {copy,clone,dedupe}_range interface.  The first few patches hoist some
> of the reflink/dedupe prep functions into the VFS since the code is
> common to XFS and ocfs2.  Then there are fixes for some ocfs2 bugs that
> I found; and finally the last kernel patch contains the new code.
> 
> A few minor fixes to xfstests are needed to make more of the tests run;
> those have been sent separately to the fstests lists.  Some of the
> xfstests will fail because ocfs2's design does not allow sharing of
> blocks between disjoint sets of files with shared blocks.  That is to
> say, if A shares with B and C shares with D, B and C cannot be made to
> share blocks.

Merged, will be in tonight's push.