From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:52795)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1YQo2q-0005PY-CQ
	for qemu-devel@nongnu.org; Wed, 25 Feb 2015 21:19:21 -0500
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <famz@redhat.com>) id 1YQo2p-000074-Al
	for qemu-devel@nongnu.org; Wed, 25 Feb 2015 21:19:20 -0500
Date: Thu, 26 Feb 2015 10:19:10 +0800
From: Fam Zheng <famz@redhat.com>
Message-ID: <20150226021910.GD10137@ad.nay.redhat.com>
References: <1424792164-2130-1-git-send-email-mreitz@redhat.com>
	<1424792164-2130-6-git-send-email-mreitz@redhat.com>
	<20150225075232.GH5293@ad.nay.redhat.com>
	<54EDD846.5030509@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <54EDD846.5030509@redhat.com>
Subject: Re: [Qemu-devel] [PATCH v3 05/11] block: Move BDS close notifiers
 into BB
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: Max Reitz <mreitz@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>, qemu-block@nongnu.org, qemu-devel@nongnu.org, Markus Armbruster <armbru@redhat.com>, Stefan Hajnoczi <stefanha@redhat.com>, Paolo Bonzini <pbonzini@redhat.com>

On Wed, 02/25 09:12, Max Reitz wrote:
> Because I need a Notifier pointer to give to
> blk_add_{insert,remove}_bs_notifier(), and most importantly because the only
> "opaque" object the callbacks will receive is that Notifier pointer (which
> in this case is actually a DataPlaneBlkChangeNotifier pointer). I cannot
> influence the @data parameter, and I cannot (easily) identify the
> VirtIOBlockDataPlane object from the BlockBackend (which is @data) alone.

Probably like this:

    struct VirtIOBlockDataPlane {
        ...
        Notifier insert_notifier, remove_notifier;
        ...
    };

    static void data_plane_blk_insert_notifier(Notifier *n, void *data)
    {
        VirtIOBlockDataPlane *s = container_of(n, VirtIOBlockDataPlane, insert_notifier);
        assert(s->conf->conf.blk == data);
        data_plane_set_up_op_blockers(s);
    }

    void virtio_blk_data_plane_create(VirtIODevice *vdev, VirtIOBlkConf *conf,
                                      VirtIOBlockDataPlane **dataplane,
    {
        ...
        s->insert_notifier.notify = data_plane_blk_insert_notifier;
        blk_add_insert_bs_notifier(conf->conf.blk, &s->insert_notifier);
        ...
    }

?

> Won't virtio_scsi_hotunplug() (which frees s->{insert,remove}_notifier) be
> called before a second device is plugged?
> 

No, multiple SCSIDevice objects (scsi-disk, scsi-generic, etc...) can be
attached to the same virtio-scsi bus.

Fam