From mboxrd@z Thu Jan  1 00:00:00 1970
Received: from eggs.gnu.org ([2001:4830:134:3::10]:43667)
	by lists.gnu.org with esmtp (Exim 4.71)
	(envelope-from <qiaonuohan@cn.fujitsu.com>) id 1UuHzg-0006yV-8A
	for qemu-devel@nongnu.org; Wed, 03 Jul 2013 04:00:55 -0400
Received: from Debian-exim by eggs.gnu.org with spam-scanned (Exim 4.71)
	(envelope-from <qiaonuohan@cn.fujitsu.com>) id 1UuHzZ-000729-6n
	for qemu-devel@nongnu.org; Wed, 03 Jul 2013 04:00:52 -0400
Received: from [222.73.24.84] (port=23378 helo=song.cn.fujitsu.com)
	by eggs.gnu.org with esmtp (Exim 4.71)
	(envelope-from <qiaonuohan@cn.fujitsu.com>) id 1UuHzY-0006zU-RL
	for qemu-devel@nongnu.org; Wed, 03 Jul 2013 04:00:45 -0400
Message-ID: <51D3D547.3000803@cn.fujitsu.com>
Date: Wed, 03 Jul 2013 15:39:51 +0800
From: Qiao Nuohan <qiaonuohan@cn.fujitsu.com>
MIME-Version: 1.0
References: <1369709437-24969-1-git-send-email-qiaonuohan@cn.fujitsu.com>
	<20130619134952.GE31475@stefanha-thinkpad.muc.redhat.com>
	<51C2667B.4030600@cn.fujitsu.com>
	<20130620085719.GA15672@stefanha-thinkpad.redhat.com>
	<51CBE58D.8080507@cn.fujitsu.com>
	<20130627085451.GB14351@stefanha-thinkpad.redhat.com>
	<51CCFB98.2020404@cn.fujitsu.com>
	<20130701114526.GA5543@stefanha-thinkpad.redhat.com>
In-Reply-To: <20130701114526.GA5543@stefanha-thinkpad.redhat.com>
Content-Transfer-Encoding: 7bit
Content-Type: text/plain; charset=ISO-8859-1; format=flowed
Subject: Re: [Qemu-devel] [PATCH v4 0/9] Make 'dump-guest-memory' dump in
 kdump-compressed format
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>, lcapitulino@redhat.com
Cc: anderson@redhat.com, kumagai-atsushi@mxc.nes.nec.co.jp, zhangxh@cn.fujitsu.com, qemu-devel@nongnu.org, afaerber@suse.de

On 07/01/2013 07:45 PM, Stefan Hajnoczi wrote:
> I'm now convinced that kdump is worthwhile, thanks for providing data.
>
> It would be nice to see the flattened kdump approach.

Hi Stefan and Luiz,

Thanks for Stefan's review again!

I am starting to implement the function using flatten format. And the idea is:

the process will use "seek" to find the place to write if it possible,
otherwise flatten format will be used to avoid "seek" operation.

I send the mail just to make sure if the idea will be accepted by qemu people
and I am spending time on useful thing. Additionally, since the introspection
is not ready yet, the 5th version of my patches will be sent later just to
see if the process of creating kdump-compressed format will be convinced by
qemu people.

P.S.

In flatten format, data will be write to dumpfile block by block, and uses the
following structure to indicate the offset and size of a data block.

struct makedumpfile_data_header {
         int64_t offset;
         int64_t buf_size;
};

For more information, please refer to makedumpfile


http://sourceforge.net/projects/makedumpfile/

-- 
Regards
Qiao Nuohan