From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from mailman by lists.gnu.org with tmda-scanned (Exim 4.43)
	id 1O73GY-0008E3-Ds
	for qemu-devel@nongnu.org; Wed, 28 Apr 2010 05:09:10 -0400
Received: from [140.186.70.92] (port=60046 helo=eggs.gnu.org)
	by lists.gnu.org with esmtp (Exim 4.43) id 1O73GW-0008Dv-4u
	for qemu-devel@nongnu.org; Wed, 28 Apr 2010 05:09:09 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.69)
	(envelope-from <kwolf@redhat.com>) id 1O73GU-0006Bz-I0
	for qemu-devel@nongnu.org; Wed, 28 Apr 2010 05:09:07 -0400
Received: from mx1.redhat.com ([209.132.183.28]:53452)
	by eggs.gnu.org with esmtp (Exim 4.69)
	(envelope-from <kwolf@redhat.com>) id 1O73GU-0006Bf-Ah
	for qemu-devel@nongnu.org; Wed, 28 Apr 2010 05:09:06 -0400
Message-ID: <4BD7FAF2.1000900@redhat.com>
Date: Wed, 28 Apr 2010 11:08:02 +0200
From: Kevin Wolf <kwolf@redhat.com>
MIME-Version: 1.0
Subject: Re: [Qemu-devel] thread synchronization in qcow2.c and qcow2-cluster.c
References: <s2g50c1b8771004270906v69ad619cycc188f212fb0440a@mail.gmail.com>	<w2rfbd9d3991004271249sd088419diba35b1a5c0a20458@mail.gmail.com>
	<k2v50c1b8771004271339zf55a7abej8251f18169e5d3a2@mail.gmail.com>
In-Reply-To: <k2v50c1b8771004271339zf55a7abej8251f18169e5d3a2@mail.gmail.com>
Content-Type: text/plain; charset=ISO-8859-1
Content-Transfer-Encoding: 7bit
List-Id: qemu-devel.nongnu.org
List-Unsubscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=unsubscribe>
List-Archive: <http://lists.gnu.org/pipermail/qemu-devel>
List-Post: <mailto:qemu-devel@nongnu.org>
List-Help: <mailto:qemu-devel-request@nongnu.org?subject=help>
List-Subscribe: <http://lists.nongnu.org/mailman/listinfo/qemu-devel>,
	<mailto:qemu-devel-request@nongnu.org?subject=subscribe>
To: "Chunqiang (CQ) Tang" <tangchq@gmail.com>
Cc: Stefan Hajnoczi <stefanha@gmail.com>, qemu-devel@nongnu.org, Avi Kivity <avi@redhat.com>

Am 27.04.2010 22:39, schrieb Chunqiang (CQ) Tang:
>> kvm-all.c:kvm_cpu_exec:
>>        qemu_mutex_unlock_iothread();
>>        ret = kvm_vcpu_ioctl(env, KVM_RUN, 0);
>>        qemu_mutex_lock_iothread();
> 
> Thank you for the information. I also suspected that
> qemu_mutex_lock_iothread() does the synchronization. However, my
> profiling showed that qemu-kvm.c:kvm_cpu_exec() in is actually
> executed, instead of kvm-all.c:kvm_cpu_exec(). Also I previously
> profiled all executions of qemu_mutex_lock_iothread(), and found that
> it only protects the vl.c:main_loop_wai() thread but does NOT protect
> the qemu-kvm.c:kvm_cpu_exec() thread. Did I miss something or is this
> a defect? I did extensive profiling but still don't know the source
> code well enough to confidently draw a conclusion.

That code path exists only in qemu-kvm, but yes, looks wrong to me. The
block drivers are definitely not prepared to run in parallel in multiple
threads.

Avi, something missing there in qemu-kvm?

Kevin