From mboxrd@z Thu Jan 1 00:00:00 1970 Return-Path: Received: from cn.fujitsu.com ([59.151.112.132]:52038 "EHLO heian.cn.fujitsu.com" rhost-flags-OK-FAIL-OK-FAIL) by vger.kernel.org with ESMTP id S1750719AbbJNE2h convert rfc822-to-8bit (ORCPT ); Wed, 14 Oct 2015 00:28:37 -0400 From: Zhao Lei To: CC: References: <20151013161913.GL27761@twin.jikos.cz> In-Reply-To: <20151013161913.GL27761@twin.jikos.cz> Subject: RE: [PATCH 0/3] btrfs-progs: mkfs: Fix different mixed type by argument sequence Date: Wed, 14 Oct 2015 12:28:03 +0800 Message-ID: <00a101d10638$b763c7c0$262b5740$@cn.fujitsu.com> MIME-Version: 1.0 Content-Type: text/plain; charset="UTF-8" Sender: linux-btrfs-owner@vger.kernel.org List-ID: Hi, David Sterba > -----Original Message----- > From: David Sterba [mailto:dsterba@suse.cz] > Sent: Wednesday, October 14, 2015 12:19 AM > To: Zhao Lei > Cc: linux-btrfs@vger.kernel.org > Subject: Re: [PATCH 0/3] btrfs-progs: mkfs: Fix different mixed type by > argument sequence > > On Tue, Oct 13, 2015 at 08:52:16PM +0800, Zhao Lei wrote: > > Given a 200G vdd1 and 1G vdd2: > > > > In current code: > > mkfs.btrfs -f /dev/vdd1 /dev/vdd2 > > and > > mkfs.btrfs -f /dev/vdd2 /dev/vdd1 > > will create different "mixed" type. > > I think combining large and small devices was not intended use for the > mixed-bg, nevertheless current behaviour is not right. > > Chandan is working on dropping the forced mixed-bg completely. We've > discussed that on IRC, I'm ok with it but this needs more testing. So far it looks > fine, small filesystems get created and usable, though some tuning might be > needed. > > My intentions for 4.3 is to take Chandan's work provided that we test it enough. > There are like 3 weeks left. In case of problems, I'll take this patchset so at > least we get the inconsisten behaviour fixed. Thanks for explanation. If 4.3 released with dropping mixed-bg, the PATCH 1/3 maybe still necessary, I'll rebase after 4.3. And another problem: This time the vdh1 is changed to 100M: [root@kerneldev progs]# ./mkfs.btrfs -f /dev/vdh1 /dev/vdh2 btrfs-progs v4.2.2-25-gc6b29b6-dirty See http://btrfs.wiki.kernel.org for more information. Label: (null) UUID: 4465aed7-d33c-4b76-9723-a76df74547fb Node size: 16384 Sector size: 4096 Filesystem size: 1.20GiB Block group profiles: Data+Metadata: RAID0 72.00MiB System: RAID1 12.00MiB SSD detected: no Incompat features: mixed-bg, extref, skinny-metadata Number of devices: 2 Devices: ID SIZE PATH 1 120.55MiB /dev/vdh1 2 1.08GiB /dev/vdh2 [root@kerneldev progs]# ./mkfs.btrfs -f /dev/vdh2 /dev/vdh1 btrfs-progs v4.2.2-25-gc6b29b6-dirty See http://btrfs.wiki.kernel.org for more information. not enough free space [root@kerneldev progs]# Can be fixed by checking raid support only based on "large disk count". This fix maybe conflict with Chandan's work, so I'll begin work after he finished. Thanks Zhaolei