From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:46014)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <kwolf@redhat.com>) id 1fTSJm-0007Ou-NB
	for qemu-devel@nongnu.org; Thu, 14 Jun 2018 09:29:39 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <kwolf@redhat.com>) id 1fTSJl-0006LH-Qt
	for qemu-devel@nongnu.org; Thu, 14 Jun 2018 09:29:38 -0400
Date: Thu, 14 Jun 2018 15:29:30 +0200
From: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20180614132930.GH8564@localhost.localdomain>
References: <20180608192027.284601-1-vsementsov@virtuozzo.com>
	<20180608192027.284601-4-vsementsov@virtuozzo.com>
	<20180614131652.GF8564@localhost.localdomain>
	<a1365d50-b830-b342-953f-2ad8a56b1d23@openvz.org>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <a1365d50-b830-b342-953f-2ad8a56b1d23@openvz.org>
Subject: Re: [Qemu-devel] [PATCH 3/3] qcow2: add compress threads
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: "Denis V. Lunev" <den@openvz.org>
Cc: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>, qemu-devel@nongnu.org, qemu-block@nongnu.org, mreitz@redhat.com, stefanha@gmail.com, pl@kamp.de

Am 14.06.2018 um 15:19 hat Denis V. Lunev geschrieben:
> On 06/14/2018 04:16 PM, Kevin Wolf wrote:
> > Am 08.06.2018 um 21:20 hat Vladimir Sementsov-Ogievskiy geschrieben:
> >> Do data compression in separate threads. This significantly improve
> >> performance for qemu-img convert with -W (allow async writes) and -c
> >> (compressed) options.
> >>
> >> Signed-off-by: Vladimir Sementsov-Ogievskiy <vsementsov@virtuozzo.com>
> > Looks correct to me, but why do we introduce a separate
> > MAX_COMPRESS_THREADS? Can't we simply leave the maximum number of
> > threads to the thread poll?
> >
> > I see that you chose a much smaller number here (4 vs. 64), but is there
> > actually a good reason for this?
> >
> > Kevin
> yes. In the other case the guest will suffer much more from this increased
> activity and load on the host.

Ah, your primary motivation is use in a backup block job? I completely
forgot about that one (and qemu-img shouldn't care because there is no
guest), but that makes some sense.

Makes me wonder whether this value should be configurable. But that can
come later.

Kevin