From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:36490)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2fdN-0003kT-RO
	for qemu-devel@nongnu.org; Thu, 03 Jul 2014 07:57:09 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2fdF-00014K-74
	for qemu-devel@nongnu.org; Thu, 03 Jul 2014 07:57:01 -0400
Received: from mx1.redhat.com ([209.132.183.28]:25165)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2fdE-00014D-T6
	for qemu-devel@nongnu.org; Thu, 03 Jul 2014 07:56:53 -0400
Message-ID: <53B544FC.7010900@redhat.com>
Date: Thu, 03 Jul 2014 13:56:44 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <CACVXFVNVwpcEUeBQr0UsW58B2CG3Qghj-2SSfx9GC1rrwbjw0A@mail.gmail.com>	<20140627120129.GO12061@stefanha-thinkpad.muc.redhat.com>	<CACVXFVPYd6Hs9k6x4pVGJpEzLgY2DAJbm02pgc-cOwQH6rhMzg@mail.gmail.com>	<53ADE769.3060903@redhat.com>	<CACVXFVMYQ62tREcPDs0rh834BuKtygYqLJic71cmuTYjFzgCJg@mail.gmail.com>	<20140630080850.GB30969@stefanha-thinkpad.redhat.com>	<CACVXFVMBxO=aMVnGLuFReB8RUA+qyfOkpNyhaiDaf+DNinRKDg@mail.gmail.com>	<CAJSP0QX70kWC87vSCObe56GCj6NKHnVSH3s7PuGrTWZneaOSkA@mail.gmail.com>	<CACVXFVMfD_i-T+R-SiQkcZseEOhBzqDLG2eu0+hZOQawnDa1mw@mail.gmail.com>	<53B2E69A.1090707@redhat.com>	<20140702085453.GI4660@stefanha-thinkpad.redhat.com>	<CACVXFVOQA4qzopU56=R=mOKJOiUUiMj+BEm8CHPo8hSK8g8ZqQ@mail.gmail.com>	<53B43194.80102@redhat.com>	<CACVXFVNOpFBUwmZmZ0NczsKdnAbY=Fwvy3A-ryLUv0k7BqCKCA@mail.gmail.com>	<53B5308F.3030008@redhat.com>
	<CACVXFVOA90CXQcQh_kTAxg7HwR-jQFns1C0dXZiyZLWOvo-eQw@mail.gmail.com>
In-Reply-To: <CACVXFVOA90CXQcQh_kTAxg7HwR-jQFns1C0dXZiyZLWOvo-eQw@mail.gmail.com>
Content-Type: text/plain; charset=UTF-8; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [regression] dataplane: throughout -40% by commit
	580b6b2aa2
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: Ming Lei <tom.leiming@gmail.com>
Cc: Kevin Wolf <kwolf@redhat.com>, Fam Zheng <famz@redhat.com>, "Michael S. Tsirkin" <mst@redhat.com>, Stefan Hajnoczi <stefanha@gmail.com>, qemu-devel <qemu-devel@nongnu.org>, Stefan Hajnoczi <stefanha@redhat.com>

Il 03/07/2014 13:50, Ming Lei ha scritto:
>> Yes, you can just move the atomic_inc/atomic_dec in aio_poll.
>
> If you mean move inc/dec of 'running' in aio_poll, that won't work.
> When aio_notify() sees 'running', it won't set notifier, and may
> trap to ppoll().

I mean move it to aio_poll, around the second invocation of aio_dispatch.

IIRC the first invocation of aio_dispatch is only used when AioContext 
is running as a GSource.  It should never run for dataplane, so it's 
okay if we only trap aio_notify from the second aio_dispatch().

In 2.2, we should rewrite aio_poll() to follow the 
prepare/poll/check/dispatch model of glib.  Then the optimization 
becomes much simpler (you only call aio_dispatch once) and it also works 
for GSource mode.

Stefan convinced me to send the patch myself, so I will do it now.

Paolo