From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:39373)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <yur@virtuozzo.com>) id 1Zj6q4-0006gl-TV
	for qemu-devel@nongnu.org; Mon, 05 Oct 2015 10:34:08 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <yur@virtuozzo.com>) id 1Zj6q1-0005TI-ON
	for qemu-devel@nongnu.org; Mon, 05 Oct 2015 10:34:04 -0400
Received: from relay.parallels.com ([195.214.232.42]:57710)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <yur@virtuozzo.com>) id 1Zj6q1-000378-Gr
	for qemu-devel@nongnu.org; Mon, 05 Oct 2015 10:34:01 -0400
References: <1443685083-6242-1-git-send-email-den@openvz.org>
	<1443685083-6242-4-git-send-email-den@openvz.org>
	<20151001225926.32707.11419@loki>
From: Yuri Pudgorodskiy <yur@virtuozzo.com>
Message-ID: <561286BC.2010206@virtuozzo.com>
Date: Mon, 5 Oct 2015 17:18:36 +0300
MIME-Version: 1.0
In-Reply-To: <20151001225926.32707.11419@loki>
Content-Type: text/plain; charset=utf-8; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH 3/5] qga: guest exec functionality
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: Michael Roth <mdroth@linux.vnet.ibm.com>, "Denis V. Lunev" <den@openvz.org>
Cc: qemu-devel@nongnu.org

On 10/2/2015 1:59 AM, Michael Roth wrote:
>> +#
>> +# Since: 2.5
>> +##
>> +{ 'command': 'guest-exec',
>> +  'data':    { 'path': 'str', '*arg': ['str'], '*env': ['str'],
>> +               '*inp-data': 'str', '*capture-output': 'bool' },
>> +  'returns': 'GuestExec' }
> Would it make sense to just add handle (pid) to GuestExecStatus, and
> have this return GuestExecStatus just the same as guest-exec-status
> does? I'm not sure either way personally, just thought I'd mention it.
>

I do not think it's a good idea. GuestExecStatus contains mostly the 
data about
the finished exec. Having GuestExec returns same struct may make wrong
assumption that it can be synchronous - wait for exec to complete and 
return all
data in a single call.

Implementing synchronous GuestExec is not and easy job - either we occupy
host-guest channel for all time until task finished, which is bad or we 
need to
implement multiplexed messages for concurrent qga commands.