From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:43730)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1ZEbBZ-0003Pv-JT
	for qemu-devel@nongnu.org; Mon, 13 Jul 2015 06:42:10 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1ZEbBU-0006f8-G8
	for qemu-devel@nongnu.org; Mon, 13 Jul 2015 06:42:09 -0400
Received: from mx1.redhat.com ([209.132.183.28]:55635)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <amit.shah@redhat.com>) id 1ZEbBU-0006f3-BU
	for qemu-devel@nongnu.org; Mon, 13 Jul 2015 06:42:04 -0400
Date: Mon, 13 Jul 2015 16:11:49 +0530
From: Amit Shah <amit.shah@redhat.com>
Message-ID: <20150713104149.GB11699@grmbl.mre>
References: <1436775182-67309-1-git-send-email-pyssling@ludd.ltu.se>
	<55A38FBA.50505@redhat.com>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <55A38FBA.50505@redhat.com>
Subject: Re: [Qemu-devel] [PATCH] qemu-char: Fix missed data on unix socket
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: Paolo Bonzini <pbonzini@redhat.com>
Cc: liangshunbin@jovaunn.com, qemu-devel@nongnu.org, pyssling@ludd.ltu.se

On (Mon) 13 Jul 2015 [12:15:22], Paolo Bonzini wrote:
> 
> 
> On 13/07/2015 10:13, pyssling@ludd.ltu.se wrote:
> > Commit 812c1057 introduced HUP detection on unix and tcp sockets prior
> > to a read in tcp_chr_read. This unfortunately broke CloudStack 4.2
> > which relied on the old behaviour where data on a socket was readable
> > even if a HUP was present.
> > 
> > On Linux a working solution seems to be to simply check the HUP after
> > reading available data, while keeping the original behaviour for windows.
> > 
> > There is then a divergence in behaviour for the two platforms, but this
> > seems better than breaking a whole software stack.
> 
> There is no need to do something special on Windows, I think.  You can
> unconditionally check G_IO_HUP after reading.  One nit:
> 
> > -    if (size == 0) {
> > +    if (size == 0 || (size < 0 && cond & G_IO_HUP)) {
> 
> Please put (cond & G_IO_HUP) within parentheses.

Also, returning TRUE there isn't right - if the connection ends, we
should return FALSE.


		Amit