From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:38706)
	by lists.gnu.org with esmtp (Exim 4.71) (envelope-from <pl@kamp.de>)
	id 1Ya3SJ-0001zt-5E
	for qemu-devel@nongnu.org; Mon, 23 Mar 2015 10:35:55 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pl@kamp.de>) id 1Ya3SG-0007D0-3O
	for qemu-devel@nongnu.org; Mon, 23 Mar 2015 10:35:51 -0400
Received: from mx-v6.kamp.de ([2a02:248:0:51::16]:50794 helo=mx01.kamp.de)
	by eggs.gnu.org with esmtp (Exim 4.71) (envelope-from <pl@kamp.de>)
	id 1Ya3SF-0007CR-Pa
	for qemu-devel@nongnu.org; Mon, 23 Mar 2015 10:35:48 -0400
Message-ID: <551024BE.1050506@kamp.de>
Date: Mon, 23 Mar 2015 15:35:42 +0100
From: Peter Lieven <pl@kamp.de>
MIME-Version: 1.0
References: <1427085970-23823-1-git-send-email-famz@redhat.com>	<1427085970-23823-2-git-send-email-famz@redhat.com>
	<20150323141443.GJ9268@stefanha-thinkpad.redhat.com>
In-Reply-To: <20150323141443.GJ9268@stefanha-thinkpad.redhat.com>
Content-Type: text/plain; charset=windows-1252; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [Qemu-stable] [PATCH 1/2] block: Fix unaligned
 zero write
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: Stefan Hajnoczi <stefanha@gmail.com>, Fam Zheng <famz@redhat.com>
Cc: Kevin Wolf <kwolf@redhat.com>, qemu-block@nongnu.org, qemu-devel@nongnu.org, Stefan Hajnoczi <stefanha@redhat.com>, qemu-stable@nongnu.org

Am 23.03.2015 um 15:14 schrieb Stefan Hajnoczi:
> On Mon, Mar 23, 2015 at 12:46:09PM +0800, Fam Zheng wrote:
>> @@ -3435,6 +3446,10 @@ static int coroutine_fn bdrv_co_do_pwritev(BlockDriverState *bs,
>>           bytes = ROUND_UP(bytes, align);
>>       }
>>   
>> +    if (use_local_qiov) {
>> +        /* Local buffer may have non-zero data. */
>> +        flags &= ~BDRV_REQ_ZERO_WRITE;
>> +    }
>>       ret = bdrv_aligned_pwritev(bs, &req, offset, bytes,
>>                                  use_local_qiov ? &local_qiov : qiov,
>>                                  flags);
>> @@ -3475,14 +3490,32 @@ int coroutine_fn bdrv_co_write_zeroes(BlockDriverState *bs,
>>                                         int64_t sector_num, int nb_sectors,
>>                                         BdrvRequestFlags flags)
>>   {
>> +    int ret;
>> +
>>       trace_bdrv_co_write_zeroes(bs, sector_num, nb_sectors, flags);
>>   
>> -    if (!(bs->open_flags & BDRV_O_UNMAP)) {
>> -        flags &= ~BDRV_REQ_MAY_UNMAP;
>> -    }
> Why is it okay to drop this when the request is aligned?

I also think it is not. This is the only point in the code that clears the
MAY_UNMAP flag when we do not open with discard=on

Peter