From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:48090)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <pbutsykin@virtuozzo.com>) id 1bhIfw-0000BK-FL
	for qemu-devel@nongnu.org; Tue, 06 Sep 2016 11:52:41 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <pbutsykin@virtuozzo.com>) id 1bhIfs-0002R2-EP
	for qemu-devel@nongnu.org; Tue, 06 Sep 2016 11:52:39 -0400
References: <20160829171021.4902-1-pbutsykin@virtuozzo.com>
	<20160829171021.4902-2-pbutsykin@virtuozzo.com>
	<20160901143137.GC6355@noname.redhat.com>
From: Pavel Butsykin <pbutsykin@virtuozzo.com>
Message-ID: <57CEDEA1.1090700@virtuozzo.com>
Date: Tue, 6 Sep 2016 18:20:01 +0300
MIME-Version: 1.0
In-Reply-To: <20160901143137.GC6355@noname.redhat.com>
Content-Type: text/plain; charset="windows-1252"; format=flowed
Content-Transfer-Encoding: 7bit
Subject: Re: [Qemu-devel] [PATCH RFC v2 01/22] block/pcache: empty pcache
 driver filter
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: qemu-block@nongnu.org, qemu-devel@nongnu.org, mreitz@redhat.com, stefanha@redhat.com, den@openvz.org, jsnow@redhat.com, eblake@redhat.com, famz@redhat.com

On 01.09.2016 17:31, Kevin Wolf wrote:
> Am 29.08.2016 um 19:10 hat Pavel Butsykin geschrieben:
>> The basic version of pcache driver for easy preparation of a patch set.
>>
>> Signed-off-by: Pavel Butsykin <pbutsykin@virtuozzo.com>
>
>> +    .bdrv_aio_readv                     = pcache_aio_readv,
>> +    .bdrv_aio_writev                    = pcache_aio_writev,
>
> Can you please use .bdrv_co_preadv/pwritev instead and make everything
> based on bytes rather than sectors?
>
> Internally you can still spawn AIO requests to achieve the same
> parallelism as you have now (we'll just need new byte-based
> bdrv_co_aio_prw_vector() wrappers, but the functionality is there) and I
> don't think making the outer layer coroutine based would be too hard. In
> fact it might even simplify some code.
>
Of course, I just wrote pcache when new byte-based interfaces have not
yet been implemented.

> Kevin
>