From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:48369)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2Gnc-0001sO-FM
	for qemu-devel@nongnu.org; Wed, 02 Jul 2014 05:26:03 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2GnT-0005Br-Eb
	for qemu-devel@nongnu.org; Wed, 02 Jul 2014 05:25:56 -0400
Received: from mx1.redhat.com ([209.132.183.28]:14641)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbonzini@redhat.com>) id 1X2GnT-0005Bj-5f
	for qemu-devel@nongnu.org; Wed, 02 Jul 2014 05:25:47 -0400
Message-ID: <53B3D011.9000200@redhat.com>
Date: Wed, 02 Jul 2014 11:25:37 +0200
From: Paolo Bonzini <pbonzini@redhat.com>
MIME-Version: 1.0
References: <1404291017-7456-1-git-send-email-arei.gonglei@huawei.com>
	<53B3CB04.5040909@redhat.com> <53B3CFB8.5000800@huawei.com>
In-Reply-To: <53B3CFB8.5000800@huawei.com>
Content-Type: text/plain; charset=ISO-8859-15; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH] ide: fix double free
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: ChenLiang <chenliang88@huawei.com>
Cc: kwolf@redhat.com, arei.gonglei@huawei.com, weidong.huang@huawei.com, qemu-devel@nongnu.org, stefanha@redhat.com

Il 02/07/2014 11:24, ChenLiang ha scritto:
> On 2014/7/2 17:04, Paolo Bonzini wrote:
>
>> This is definitely a heavyweight solution, and in fact the bug should
>> not be there in the first place.  See dma_complete:
>>
>> static void dma_complete(DMAAIOCB *dbs, int ret)
>> {
>>     trace_dma_complete(dbs, ret, dbs->common.cb);
>>
>>     dma_bdrv_unmap(dbs);
>>     if (dbs->common.cb) {
>>         dbs->common.cb(dbs->common.opaque, ret);
>>     }
>>     qemu_iovec_destroy(&dbs->iov);
>>     if (dbs->bh) {
>>         qemu_bh_delete(dbs->bh);
>>         dbs->bh = NULL;
>>     }
>>     if (!dbs->in_cancel) {
>>         /* Requests may complete while dma_aio_cancel is in progress.  In
>>          * this case, the AIOCB should not be released because it is still
>>          * referenced by dma_aio_cancel.  */
>>         qemu_aio_release(dbs);
>>     }
>> }
>>
>> Perhaps something like this?
>>
>> diff --git a/dma-helpers.c b/dma-helpers.c
>> index 53cbe92..21b70d12 100644
>> --- a/dma-helpers.c
>> +++ b/dma-helpers.c
>> @@ -181,15 +181,15 @@ static void dma_aio_cancel(BlockDriverAIOCB *acb)
>>
>>      trace_dma_aio_cancel(dbs);
>>
>> +    dbs->in_cancel = true;
>>      if (dbs->acb) {
>>          BlockDriverAIOCB *acb = dbs->acb;
>>          dbs->acb = NULL;
>> -        dbs->in_cancel = true;
>>          bdrv_aio_cancel(acb);
>> -        dbs->in_cancel = false;
>>      }
>>      dbs->common.cb = NULL;
>>      dma_complete(dbs, 0);
>> +    qemu_aio_release(dbs);
>>  }
>>
>>  static const AIOCBInfo dma_aiocb_info = {
>>
>>
>> .
>>
>
> acked, thanks

Did you test this? :)  What is the testcase?

Paolo