From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([140.186.70.92]:56839)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <alevy@redhat.com>) id 1R0WjY-0007fC-Bp
	for qemu-devel@nongnu.org; Mon, 05 Sep 2011 06:49:00 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <alevy@redhat.com>) id 1R0WjU-00034E-A2
	for qemu-devel@nongnu.org; Mon, 05 Sep 2011 06:48:56 -0400
Received: from mx1.redhat.com ([209.132.183.28]:36804)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <alevy@redhat.com>) id 1R0WjT-00034A-VW
	for qemu-devel@nongnu.org; Mon, 05 Sep 2011 06:48:52 -0400
Date: Mon, 5 Sep 2011 13:47:00 +0300
From: Alon Levy <alevy@redhat.com>
Message-ID: <20110905104700.GN27006@bow>
References: <1314976794-31931-1-git-send-email-kraxel@redhat.com>
	<4E633A37.40208@redhat.com> <4E649033.2040105@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <4E649033.2040105@redhat.com>
Subject: Re: [Qemu-devel] [Spice-devel] [PATCH] server: don't call
 reds_stream_free from worker thread context
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: Gerd Hoffmann <kraxel@redhat.com>
Cc: Yonit Halperin <yhalperi@redhat.com>, qemu-devel@nongnu.org, spice-devel@freedesktop.org

On Mon, Sep 05, 2011 at 11:02:43AM +0200, Gerd Hoffmann wrote:
>   Hi,
> 
> >Hi,
> >RED_WORKER_MESSAGE_DISPLAY_DISCONNECT is not the only place that
> >triggers red_disconnect_channel (and as a result,
> >reds_stream_free(dispatcher->stream)). red_disconnect_channel is called
> >also when there is an error upon receive/send and also when timeouts
> >related to the client occur (e.g., in flush_display_commands).
> 
> Ok.
> 
> >We probably better make the dispatcher bi-directional, i.e., not only
> >push messages to the worker, but also listen.
> 
> That sounds like a non-trivial thing.
> 
> What does the master branch here btw?  I had a brief look and saw
> that the code looks quite different here (probably due to the
> multiclient work).
> 

I verified it still calls reds_stream_free from the worker thread, only
now the call itself is done in red_channel.c (via red_channel_disconnect
or something like that), which is called from red_worker.c

> cheers,
>   Gerd
> _______________________________________________
> Spice-devel mailing list
> Spice-devel@lists.freedesktop.org
> http://lists.freedesktop.org/mailman/listinfo/spice-devel