From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:51860)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1etkMf-0007xu-TC
	for qemu-devel@nongnu.org; Wed, 07 Mar 2018 20:29:02 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1etkMf-00006g-7H
	for qemu-devel@nongnu.org; Wed, 07 Mar 2018 20:29:01 -0500
Date: Thu, 8 Mar 2018 09:28:42 +0800
From: Fam Zheng <famz@redhat.com>
Message-ID: <20180308012842.GJ13844@lemon.usersys.redhat.com>
References: <c0d7b993-b3b8-a753-21f5-e2257de27caf@kamp.de>
	<20180305114533.GI7910@stefanha-x1.localdomain>
	<0f1ec3c7-1157-e618-62f8-ffd5e08eb342@kamp.de>
	<20180305145215.GM3131@work-vm>
	<20180306160739.GN31045@stefanha-x1.localdomain>
	<824907bb-3a37-27bd-76db-be17d05bdac0@kamp.de>
	<231fe0ce-3df4-a175-2fba-403023cc65e1@kamp.de>
	<99df10d6-6514-56ee-16ce-8b7d1823cc08@kamp.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <99df10d6-6514-56ee-16ce-8b7d1823cc08@kamp.de>
Subject: Re: [Qemu-devel] block migration and dirty bitmap reset
List-Id: <qemu-devel.nongnu.org>
List-Unsubscribe: <https://lists.nongnu.org/mailman/options/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.nongnu.org/archive/html/qemu-devel/>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <https://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: Peter Lieven <pl@kamp.de>
Cc: Stefan Hajnoczi <stefanha@gmail.com>, "Dr. David Alan Gilbert" <dgilbert@redhat.com>, wency@cn.fujitsu.com, qemu block <qemu-block@nongnu.org>, Juan Quintela <quintela@redhat.com>, "qemu-devel@nongnu.org" <qemu-devel@nongnu.org>, Stefan Hajnoczi <stefanha@redhat.com>

On Wed, 03/07 09:06, Peter Lieven wrote:
> Hi,
> 
> while looking at the code I wonder if the blk_aio_preadv and the bdrv_reset_dirty_bitmap order must
> be swapped in mig_save_device_bulk:
> 
>     qemu_mutex_lock_iothread();
>     aio_context_acquire(blk_get_aio_context(bmds->blk));
>     blk->aiocb = blk_aio_preadv(bb, cur_sector * BDRV_SECTOR_SIZE, &blk->qiov,
>                                 0, blk_mig_read_cb, blk);
> 
>     bdrv_reset_dirty_bitmap(bmds->dirty_bitmap, cur_sector * BDRV_SECTOR_SIZE,
>                             nr_sectors * BDRV_SECTOR_SIZE);
>     aio_context_release(blk_get_aio_context(bmds->blk));
>     qemu_mutex_unlock_iothread();
> 
> In mig_save_device_dirty we first reset the dirty bitmap and read then which shoulds like
> a better idea.

Yes, that sounds right to me.

Fam