From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:37405)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1ZZZqi-0002Qy-S0
	for qemu-devel@nongnu.org; Wed, 09 Sep 2015 03:31:21 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <paolo.bonzini@gmail.com>) id 1ZZZqi-0000as-14
	for qemu-devel@nongnu.org; Wed, 09 Sep 2015 03:31:20 -0400
Sender: Paolo Bonzini <paolo.bonzini@gmail.com>
References: <1438868176-20364-1-git-send-email-pbonzini@redhat.com>
	<1438868176-20364-2-git-send-email-pbonzini@redhat.com>
	<20150909060619.GB21413@ad.nay.redhat.com>
From: Paolo Bonzini <pbonzini@redhat.com>
Message-ID: <55EFE03E.10804@redhat.com>
Date: Wed, 9 Sep 2015 09:31:10 +0200
MIME-Version: 1.0
In-Reply-To: <20150909060619.GB21413@ad.nay.redhat.com>
Content-Type: text/plain; charset=windows-1252
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH 01/18] iothread: release iothread around
	aio_poll
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: Fam Zheng <famz@redhat.com>
Cc: kwolf@redhat.com, qemu-block@nongnu.org, qemu-devel@nongnu.org, stefanha@redhat.com



On 09/09/2015 08:06, Fam Zheng wrote:
> This patch will probably go in before aio_disable_clients,

Actually I think it's blocked by aio_disable_clients.

> if any, but I'm not
> quite confident about the interface yet: listing a precise set of clients from
> monitor is an ugly coupling between modules:
> 
>     aio_disable_clients(bdrv_get_aio_context(bs),
>                         AIO_CLIENT_NBD | AIO_CLIENT_IOEVENTFD | AIO_CLIENT_FOO);
> 
>     ...
> 
>     aio_enble_clients(bdrv_get_aio_context(bs),
>                       AIO_CLIENT_NBD | AIO_CLIENT_IOEVENTFD | AIO_CLIENT_FOO);
> 
> I prefer at least an abstraction:
> 
>     bdrv_quiesce_begin(bs);
> 
>     ...
> 
>     bdrv_quiesce_end(bs);

That's fine.

Paolo

> My point is maybe we should leave documenting this to whichever series that
> introduces it?