From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([208.118.235.92]:59729)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <kwolf@redhat.com>) id 1UOCEe-0003tc-NV
	for qemu-devel@nongnu.org; Fri, 05 Apr 2013 15:23:43 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <kwolf@redhat.com>) id 1UOCEc-0007Ym-Fj
	for qemu-devel@nongnu.org; Fri, 05 Apr 2013 15:23:40 -0400
Received: from mx1.redhat.com ([209.132.183.28]:40353)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <kwolf@redhat.com>) id 1UOCEc-0007YY-8B
	for qemu-devel@nongnu.org; Fri, 05 Apr 2013 15:23:38 -0400
Date: Fri, 5 Apr 2013 21:23:27 +0200
From: Kevin Wolf <kwolf@redhat.com>
Message-ID: <20130405192327.GG2351@dhcp-200-207.str.redhat.com>
References: <1364291919-19563-1-git-send-email-pl@kamp.de>
	<1364291919-19563-7-git-send-email-pl@kamp.de>
MIME-Version: 1.0
Content-Type: text/plain; charset=us-ascii
Content-Disposition: inline
In-Reply-To: <1364291919-19563-7-git-send-email-pl@kamp.de>
Subject: Re: [Qemu-devel] [PATCHv5 06/10] migration: search for zero instead
 of dup pages
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: Peter Lieven <pl@kamp.de>
Cc: Anthony Liguori <aliguori@us.ibm.com>, quintela@redhat.com, Stefan Hajnoczi <stefanha@gmail.com>, qemu-devel@nongnu.org, Orit Wasserman <owasserm@redhat.com>, Paolo Bonzini <pbonzini@redhat.com>

Am 26.03.2013 um 10:58 hat Peter Lieven geschrieben:
> virtually all dup pages are zero pages. remove
> the special is_dup_page() function and use the
> optimized buffer_find_nonzero_offset() function
> instead.
> 
> here buffer_find_nonzero_offset() is used directly
> to avoid the unnecssary additional checks in
> buffer_is_zero().
> 
> raw performace gain checking 1 GByte zeroed memory
> over is_dup_page() is approx. 10-12% with SSE2
> and 8-10% with unsigned long arithmedtic.
> 
> Signed-off-by: Peter Lieven <pl@kamp.de>
> Reviewed-by: Orit Wasserman <owasserm@redhat.com>
> Reviewed-by: Eric Blake <eblake@redhat.com>

Okay, so I bisected again and this is the second patch that is involved
in the slowness of qemu-iotests case 007.

The problem seems to be that the RAM of a guest is in fact _not_ zeroed
during initialisation. It hits my test case reliably because I'm running
with MALLOC_PERTURB_. Now I'm wondering if in practice this happens only
under such test conditions, or if real guests could be affected as well
and we should make sure to get zeroed memory for RAM.

Kevin