From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from aserp1040.oracle.com ([141.146.126.69]:43612 "EHLO aserp1040.oracle.com" rhost-flags-OK-OK-OK-OK) by vger.kernel.org with ESMTP id S1751025AbcJBTBm (ORCPT ); Sun, 2 Oct 2016 15:01:42 -0400 Date: Sun, 2 Oct 2016 12:01:07 -0700 From: "Darrick J. Wong" Subject: XFS reflink trees updated for 4.9 Message-ID: <20161002190106.GC21076@birch.djwong.org> MIME-Version: 1.0 Content-Type: text/plain; charset=us-ascii Content-Disposition: inline Sender: linux-xfs-owner@vger.kernel.org List-ID: List-Id: xfs To: Christoph Hellwig , Dave Chinner Cc: linux-xfs@vger.kernel.org, Brian Foster , amir73il@gmail.com Seeing as 4.8 is probably going out today, here's the latest for-4.9 branch: https://github.com/djwong/linux/tree/for-dave-for-4.9-9 https://github.com/djwong/xfsprogs/tree/for-dave-for-4.9-9 https://github.com/djwong/xfstests/tree/djwong-devel I've rolled up all the comment fixes that have come in this week. Thanks to everyone who's reviewed the patches (Dave, Christoph, and Brian come to mind)! I've also sped up stale CoW staging extent recovery by stashing all those records in the right side of the refcount btree per a suggestion from Dave on Friday afternoon. It was a relatively small change, so I didn't bother holding off, and everything still passes the same tests they did before, AFAICT. --- Dave: the two xfs_repair patches I posted yesterday are for xfsprogs 4.8. One fixes some garbled rmap code that crashed when we overestimated the freespace btree space requirements in phase5, and needs to go in before the release. The second patch re-adds the lost blocks to the freesp btree so we get rid of those "sorry, lost XXX blocks" messages. That one isn't so urgent. --D