All of lore.kernel.org
 help / color / mirror / Atom feed
From: Jingbai Ma <jingbai.ma@hp.com>
To: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: kexec@lists.infradead.org, linux-kernel@vger.kernel.org,
	lisa.mitchell@hp.com, mingo@redhat.com,
	kumagai-atsushi@mxc.nes.nec.co.jp, jingbai.ma@hp.com,
	hpa@zytor.com, yinghai@kernel.org, vgoyal@redhat.com,
	ebiederm@xmission.com
Subject: Re: [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process
Date: Mon, 11 Mar 2013 17:02:41 +0800	[thread overview]
Message-ID: <513D9DB1.7040100@hp.com> (raw)
In-Reply-To: <20130309.133118.214580151.d.hatayama@jp.fujitsu.com>

On 03/09/2013 12:31 PM, HATAYAMA Daisuke wrote:
> From: Jingbai Ma<jingbai.ma@hp.com>
> Subject: Re: [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process
> Date: Fri, 8 Mar 2013 18:06:31 +0800
>
>> On 03/07/2013 11:21 PM, Vivek Goyal wrote:
>>> On Thu, Mar 07, 2013 at 10:58:18PM +0800, Jingbai Ma wrote:
> ...
>>> First of all 64MB per TB should not be a huge deal. And makedumpfile
>>> also has this cyclic mode where you process a map, discard it and then
>>> move on to next section. So memory usage remains constant at the
>>> expense
>>> of processing time.
>>
>> Yes, that's true. But in cyclic mode, makedumpfile will have to
>> write/read bitmap from storage, it will also impact the performance.
>> I have measured the penalty for cyclic mode is about 70%
>> slowdown. Maybe could be faster after mmap implemented.
>
> I guess the slowdown came from the issue that enough VMCOREINFO was
> not provided from the kernel, and unnecessary filtering processing for
> free pages is done multiple times.

Thanks for your comments! It would be very helpful.
I will test it on the machine again.

-- 
Jingbai Ma (jingbai.ma@hp.com)

_______________________________________________
kexec mailing list
kexec@lists.infradead.org
http://lists.infradead.org/mailman/listinfo/kexec

WARNING: multiple messages have this Message-ID (diff)
From: Jingbai Ma <jingbai.ma@hp.com>
To: HATAYAMA Daisuke <d.hatayama@jp.fujitsu.com>
Cc: jingbai.ma@hp.com, vgoyal@redhat.com, kexec@lists.infradead.org,
	linux-kernel@vger.kernel.org, lisa.mitchell@hp.com,
	mingo@redhat.com, kumagai-atsushi@mxc.nes.nec.co.jp,
	ebiederm@xmission.com, hpa@zytor.com, yinghai@kernel.org
Subject: Re: [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process
Date: Mon, 11 Mar 2013 17:02:41 +0800	[thread overview]
Message-ID: <513D9DB1.7040100@hp.com> (raw)
In-Reply-To: <20130309.133118.214580151.d.hatayama@jp.fujitsu.com>

On 03/09/2013 12:31 PM, HATAYAMA Daisuke wrote:
> From: Jingbai Ma<jingbai.ma@hp.com>
> Subject: Re: [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process
> Date: Fri, 8 Mar 2013 18:06:31 +0800
>
>> On 03/07/2013 11:21 PM, Vivek Goyal wrote:
>>> On Thu, Mar 07, 2013 at 10:58:18PM +0800, Jingbai Ma wrote:
> ...
>>> First of all 64MB per TB should not be a huge deal. And makedumpfile
>>> also has this cyclic mode where you process a map, discard it and then
>>> move on to next section. So memory usage remains constant at the
>>> expense
>>> of processing time.
>>
>> Yes, that's true. But in cyclic mode, makedumpfile will have to
>> write/read bitmap from storage, it will also impact the performance.
>> I have measured the penalty for cyclic mode is about 70%
>> slowdown. Maybe could be faster after mmap implemented.
>
> I guess the slowdown came from the issue that enough VMCOREINFO was
> not provided from the kernel, and unnecessary filtering processing for
> free pages is done multiple times.

Thanks for your comments! It would be very helpful.
I will test it on the machine again.

-- 
Jingbai Ma (jingbai.ma@hp.com)

  reply	other threads:[~2013-03-11  9:02 UTC|newest]

Thread overview: 54+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2013-03-07 14:58 [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process Jingbai Ma
2013-03-07 14:58 ` Jingbai Ma
2013-03-07 14:58 ` [RFC PATCH 1/5] crash dump bitmap: add a kernel config and help document Jingbai Ma
2013-03-07 14:58   ` Jingbai Ma
2013-03-07 14:58 ` [RFC PATCH 2/5] crash dump bitmap: init crash dump bitmap in kernel booting process Jingbai Ma
2013-03-07 14:58   ` Jingbai Ma
2013-03-07 14:58 ` [RFC PATCH 3/5] crash dump bitmap: scan memory pages in kernel crash process Jingbai Ma
2013-03-07 14:58   ` Jingbai Ma
2013-03-07 14:59 ` [RFC PATCH 4/5] crash dump bitmap: add a proc interface for crash dump bitmap Jingbai Ma
2013-03-07 14:59   ` Jingbai Ma
2013-03-07 14:59 ` [RFC PATCH 5/5] crash dump bitmap: workaround for kernel 3.9-rc1 kdump issue Jingbai Ma
2013-03-07 14:59   ` Jingbai Ma
2013-03-07 15:21 ` [RFC PATCH 0/5] crash dump bitmap: scan memory pages in kernel to speedup kernel dump process Vivek Goyal
2013-03-07 15:21   ` Vivek Goyal
2013-03-07 21:38   ` Yinghai Lu
2013-03-07 21:38     ` Yinghai Lu
2013-03-07 21:54   ` Eric W. Biederman
2013-03-07 21:54     ` Eric W. Biederman
2013-03-08 15:52     ` Vivek Goyal
2013-03-08 15:52       ` Vivek Goyal
2013-03-08 16:19       ` Eric W. Biederman
2013-03-08 16:19         ` Eric W. Biederman
2013-03-11  8:18       ` Jingbai Ma
2013-03-11  8:18         ` Jingbai Ma
2013-03-11  9:42         ` Eric W. Biederman
2013-03-11  9:42           ` Eric W. Biederman
2013-03-12 10:05           ` Jingbai Ma
2013-03-12 10:05             ` Jingbai Ma
2013-03-12 19:48             ` Eric W. Biederman
2013-03-12 19:48               ` Eric W. Biederman
2013-03-08  1:31   ` HATAYAMA Daisuke
2013-03-08  1:31     ` HATAYAMA Daisuke
2013-03-08 10:06   ` Jingbai Ma
2013-03-08 10:06     ` Jingbai Ma
2013-03-08 10:33     ` H. Peter Anvin
2013-03-08 10:33       ` H. Peter Anvin
2013-03-08 13:31       ` Ma, Jingbai (Kingboard)
2013-03-08 13:31         ` Ma, Jingbai (Kingboard)
2013-03-08 16:13         ` Eric W. Biederman
2013-03-08 16:13           ` Eric W. Biederman
2013-03-11  8:31           ` Jingbai Ma
2013-03-11  8:31             ` Jingbai Ma
2013-03-08 16:19     ` Vivek Goyal
2013-03-08 16:19       ` Vivek Goyal
2013-03-11  8:53       ` Jingbai Ma
2013-03-11  8:53         ` Jingbai Ma
2013-03-09  4:31     ` HATAYAMA Daisuke
2013-03-09  4:31       ` HATAYAMA Daisuke
2013-03-11  9:02       ` Jingbai Ma [this message]
2013-03-11  9:02         ` Jingbai Ma
2013-03-08 10:16   ` Jingbai Ma
2013-03-08 10:16     ` Jingbai Ma
  -- strict thread matches above, loose matches on Subject: below --
2013-03-07 14:05 Jingbai Ma
2013-03-07 14:00 Jingbai Ma

Reply instructions:

You may reply publicly to this message via plain-text email
using any one of the following methods:

* Save the following mbox file, import it into your mail client,
  and reply-to-all from there: mbox

  Avoid top-posting and favor interleaved quoting:
  https://en.wikipedia.org/wiki/Posting_style#Interleaved_style

* Reply using the --to, --cc, and --in-reply-to
  switches of git-send-email(1):

  git send-email \
    --in-reply-to=513D9DB1.7040100@hp.com \
    --to=jingbai.ma@hp.com \
    --cc=d.hatayama@jp.fujitsu.com \
    --cc=ebiederm@xmission.com \
    --cc=hpa@zytor.com \
    --cc=kexec@lists.infradead.org \
    --cc=kumagai-atsushi@mxc.nes.nec.co.jp \
    --cc=linux-kernel@vger.kernel.org \
    --cc=lisa.mitchell@hp.com \
    --cc=mingo@redhat.com \
    --cc=vgoyal@redhat.com \
    --cc=yinghai@kernel.org \
    /path/to/YOUR_REPLY

  https://kernel.org/pub/software/scm/git/docs/git-send-email.html

* If your mail client supports setting the In-Reply-To header
  via mailto: links, try the mailto: link
Be sure your reply has a Subject: header at the top and a blank line before the message body. 
This is an external index of several public inboxes,
see mirroring instructions on how to clone and mirror
all data and code used by this external index.