From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:51266)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1SUGGR-0003H3-IX
	for qemu-devel@nongnu.org; Tue, 15 May 2012 07:50:08 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1SUGGM-00053A-TP
	for qemu-devel@nongnu.org; Tue, 15 May 2012 07:50:03 -0400
Received: from mx1.redhat.com ([209.132.183.28]:13612)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1SUGGM-000532-LX
	for qemu-devel@nongnu.org; Tue, 15 May 2012 07:49:58 -0400
Message-ID: <4FB242D1.7040705@redhat.com>
Date: Tue, 15 May 2012 13:49:37 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <87vcjx1y7k.fsf@rho.meyering.net> <4FB23F3E.4000700@redhat.com>
In-Reply-To: <4FB23F3E.4000700@redhat.com>
Content-Type: text/plain; charset=ISO-8859-15
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH] posix-aio: don't set aiocb->ret/active
 outside critical section
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: Kevin Wolf <kwolf@redhat.com>
Cc: Frediano Ziglio <freddy77@gmail.com>, Anthony Liguori <aliguori@us.ibm.com>, Jim Meyering <jim@meyering.net>, Stefan Hajnoczi <stefanha@linux.vnet.ibm.com>, qemu-devel@nongnu.org

Il 15/05/2012 13:34, Kevin Wolf ha scritto:
>>  static void qemu_paio_submit(struct qemu_paiocb *aiocb)
>>  {
>> +    mutex_lock(&lock);
>>      aiocb->ret = -EINPROGRESS;
>>      aiocb->active = 0;
>> -    mutex_lock(&lock);
>>      if (idle_threads == 0 && cur_threads < max_threads)
>>          spawn_thread();
>>      QTAILQ_INSERT_TAIL(&request_list, aiocb, node);
>>      mutex_unlock(&lock);
>>      cond_signal(&cond);
>>  }
> 
> This is just silencing coverity, not really fixing a bug, right? aiocb
> is inserted into request_lists only afterwards, and before it is in the
> list, other threads can't find it.

Indeed... I don't like the patch, the access is out of the lock for a
reason.

Paolo