From mboxrd@z Thu Jan 1 00:00:00 1970 Received: from eggs.gnu.org ([2001:4830:134:3::10]:33251) by lists.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WMtqj-0001xT-6U for qemu-devel@nongnu.org; Mon, 10 Mar 2014 02:38:14 -0400 Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71) (envelope-from ) id 1WMtqe-0001Py-BO for qemu-devel@nongnu.org; Mon, 10 Mar 2014 02:38:09 -0400 Received: from mx1.redhat.com ([209.132.183.28]:46151) by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from ) id 1WMtqe-0001Pp-2m for qemu-devel@nongnu.org; Mon, 10 Mar 2014 02:38:04 -0400 Date: Mon, 10 Mar 2014 14:38:12 +0800 From: Fam Zheng Message-ID: <20140310063811.GA8567@T430.nay.redhat.com> References: <1394180456-18139-1-git-send-email-famz@redhat.com> <20140310051355.GA3317@irqsave.net> MIME-Version: 1.0 Content-Type: text/plain; charset=iso-8859-1 Content-Disposition: inline In-Reply-To: <20140310051355.GA3317@irqsave.net> Content-Transfer-Encoding: quoted-printable Subject: Re: [Qemu-devel] [PATCH v16 00/14] Drop in_use from BlockDriverState and enable point-in-time snapshot exporting over NBD List-Id: List-Unsubscribe: , List-Archive: List-Post: List-Help: List-Subscribe: , To: =?iso-8859-1?Q?Beno=EEt?= Canet Cc: kwolf@redhat.com, rjones@redhat.com, jcody@redhat.com, qemu-devel@nongnu.org, armbru@redhat.com, imain@redhat.com, stefanha@redhat.com, pbonzini@redhat.com, ptoscano@redhat.com On Mon, 03/10 06:13, Beno=EEt Canet wrote: > The Friday 07 Mar 2014 =E0 16:20:42 (+0800), Fam Zheng wrote : > > This series adds for point-in-time snapshot NBD exporting based on > > blockdev-backup (variant of drive-backup with existing device as targ= et). > >=20 > > We get a thin point-in-time snapshot by COW mechanism of drive-backup= , and > > export it through built in NBD server. The steps are as below: > >=20 > > 1. (SHELL) qemu-img create -f qcow2 BACKUP.qcow2 > >=20 > > (Alternatively we can use -o backing_file=3DRUNNING-VM.img to omi= t explicitly > > providing the size by ourselves, but it's risky because RUNNING-V= M.qcow2 is > > used r/w by guest. Whether or not setting backing file in the ima= ge file > > doesn't matter, as we are going to override the backing hd in the= next > > step) > >=20 > > 2. (QMP) blockdev-add backing=3Dsource-drive file.driver=3Dfile file= .filename=3DBACKUP.qcow2 id=3Dtarget0 if=3Dnone driver=3Dqcow2 > >=20 > > (where source-drive is the running BlockDriverState name for > > RUNNING-VM.img. This patch implements "backing=3D" option to over= ride > > backing_hd for added drive) > >=20 > > 3. (QMP) blockdev-backup device=3Dsource-drive sync=3Dnone target=3D= target0 > >=20 > > (this is the QMP command introduced by this series, which use a n= amed > > device as target of drive-backup) > >=20 > > 4. (QMP) nbd-server-add device=3Dtarget0 > >=20 > > When image fleecing done: > >=20 > > 1. (QMP) block-job-cancel device=3Dsource-drive > >=20 > > 2. (HMP) drive_del target0 > >=20 > > 3. (SHELL) rm BACKUP.qcow2 > >=20 > > v16: Address review comments from Jeff and Markus (thanks for reviewi= ng!) > > (A side-by-side diff of v15 -> v16: http://goo.gl/1goc6S) > >=20 > > [03/14] block: Replace in_use with operation blocker > > Improve error message. (Markus) > >=20 > > [05/14] block: Add bdrv_set_backing_hd() > > in bdrv_set_backing_hd: > > * Don't call bdrv_ref and bdrv_unref. (Jeff) > > * Add "bs->open_flags &=3D ~BDRV_O_NO_BACKING". (Jeff) > > in bdrv_open_backing_file: > > * Remove unnecessary "bs->backing_hd =3D NULL;". (Jeff) > > * Don't get wrong refcnt on backing_hd. (Jeff) > > in bdrv_append: > > * Remove unnessary backing_file and backing_format copy, = it's taken > > care of in bdrv_set_backing_hd(). (Jeff) > >=20 > > [06/14] block: Add backing_blocker in BlockDriverState > > Add call to bdrv_ref() since bdrv_set_backing_hd() doesn'= t do it > > now. > >=20 > > [08/14] block: Support dropping active in bdrv_drop_intermediate > > Verify active, top, base are in the same backing chain. (= Jeff) > > Remember to change backing file when base !=3D NULL. (Jef= f) > > =20 > > [09/14] stream: Use bdrv_drop_intermediate and drop close_unused_= images > > Drop more unnecessary code. (Jeff) > >=20 > > Thanks, > > Fam > >=20 > >=20 > > Fam Zheng (14): > > block: Add BlockOpType enum > > block: Introduce op_blockers to BlockDriverState > > block: Replace in_use with operation blocker > > block: Move op_blocker check from block_job_create to its caller > > block: Add bdrv_set_backing_hd() > > block: Add backing_blocker in BlockDriverState > > block: Parse "backing" option to reference existing BDS > > block: Support dropping active in bdrv_drop_intermediate > > stream: Use bdrv_drop_intermediate and drop close_unused_images > > qmp: Add command 'blockdev-backup' > > block: Allow backup on referenced named BlockDriverState > > block: Add blockdev-backup to transaction > > qemu-iotests: Test blockdev-backup in 055 > > qemu-iotests: Image fleecing test case 083 > >=20 > > block-migration.c | 7 +- > > block.c | 314 +++++++++++++++++++++++++++---= ---------- > > block/backup.c | 26 ++++ > > block/commit.c | 2 +- > > block/mirror.c | 1 + > > block/stream.c | 42 +----- > > blockdev.c | 122 ++++++++++++++-- > > blockjob.c | 14 +- > > hw/block/dataplane/virtio-blk.c | 19 ++- > > include/block/block.h | 29 +++- > > include/block/block_int.h | 9 +- > > include/block/blockjob.h | 3 + > > qapi-schema.json | 50 +++++++ > > qmp-commands.hx | 44 ++++++ > > tests/qemu-iotests/055 | 275 +++++++++++++++++++++++++++++-= ----- > > tests/qemu-iotests/055.out | 4 +- > > tests/qemu-iotests/083 | 99 +++++++++++++ > > tests/qemu-iotests/083.out | 5 + > > tests/qemu-iotests/group | 1 + > > 19 files changed, 854 insertions(+), 212 deletions(-) > > create mode 100755 tests/qemu-iotests/083 > > create mode 100644 tests/qemu-iotests/083.out > >=20 > > --=20 > > 1.9.0 > >=20 >=20 > Hi Fam, >=20 > I want to rebase my patches on top of yours but I have trouble applying= your > series to kevin/block-next. Is it based on it ? >=20 No, I used qemu.git master. I see some conflicts, I will rebase and send = to list. But I think you meant kevin/block, since kevin/block-next looks ve= ry old, right? Fam