From: Bartlomiej Zolnierkiewicz <b.zolnierkie@samsung.com>
To: KOSAKI Motohiro <kosaki.motohiro@gmail.com>
Cc: linux-mm@kvack.org, Kyungmin Park <kyungmin.park@samsung.com>,
Matt Mackall <mpm@selenic.com>
Subject: Re: [PATCH 1/3] proc: add /proc/kpageorder interface
Date: Wed, 06 Jun 2012 10:23:13 +0200 [thread overview]
Message-ID: <201206061023.13237.b.zolnierkie@samsung.com> (raw)
In-Reply-To: <4FCD0D0D.9050003@gmail.com>
On Monday 04 June 2012 21:31:25 KOSAKI Motohiro wrote:
> (6/4/12 4:23 AM), Bartlomiej Zolnierkiewicz wrote:
> > On Friday 01 June 2012 22:31:01 KOSAKI Motohiro wrote:
> >> (6/1/12 12:54 PM), Bartlomiej Zolnierkiewicz wrote:
> >>> From: Bartlomiej Zolnierkiewicz<b.zolnierkie@samsung.com>
> >>> Subject: [PATCH] proc: add /proc/kpageorder interface
> >>>
> >>> This makes page order information available to the user-space.
> >>
> >> No usecase new feature always should be NAKed.
> >
> > It is used to get page orders for Buddy pages and help to monitor
> > free/used pages. Sample usage will be posted for inclusion to
> > Pagemap Demo tools (http://selenic.com/repo/pagemap/).
> >
> > The similar situation is with /proc/kpagetype..
>
> NAK then.
>
> First, your explanation didn't describe any usecase. "There is a similar feature"
> is NOT a usecase.
>
> Second, /proc/kpagetype is one of mistaken feature. It was not designed deeply.
> We have no reason to follow the mistake.
Well, my usecase for /proc/kpagetype is to monitor/debug pageblock changes
(i.e. to verify CMA and compaction operations). It is not perfect since
interface gives us only a snapshot of pageblocks state at some random time.
However it is a straightforward method and requires only minimal changes
to the existing code.
Maybe there is a better way to do this which would give a more accurate
data and capture every state change (maybe a one involving tracing?) but
I don't know about it. Do you know such better way to do it?
> Third, pagemap demo doesn't describe YOUR feature's usefull at all.
pagemap demo doesn't include my patches for /proc/kpage[order,type] yet
so it is not surprising at all (it doesn't even work with current kernels
without my other patches).. ;)
> Fourth, pagemap demo is NOT useful at all. It's just toy. Practically, kpagetype
> is only used from pagetype tool.
I don't quite follow it, what pagetype tool are you referring to (kpagetype
is a new interface)?
Best regards,
--
Bartlomiej Zolnierkiewicz
Samsung Poland R&D Center
--
To unsubscribe, send a message with 'unsubscribe linux-mm' in
the body to majordomo@kvack.org. For more info on Linux MM,
see: http://www.linux-mm.org/ .
Fight unfair telecom internet charges in Canada: sign http://stopthemeter.ca/
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2012-06-06 8:27 UTC|newest]
Thread overview: 6+ messages / expand[flat|nested] mbox.gz Atom feed top
2012-06-01 16:54 Bartlomiej Zolnierkiewicz
2012-06-01 20:31 ` KOSAKI Motohiro
2012-06-04 8:23 ` Bartlomiej Zolnierkiewicz
2012-06-04 19:31 ` KOSAKI Motohiro
2012-06-06 8:23 ` Bartlomiej Zolnierkiewicz [this message]
2012-06-08 2:32 ` KOSAKI Motohiro
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=201206061023.13237.b.zolnierkie@samsung.com \
--to=b.zolnierkie@samsung.com \
--cc=kosaki.motohiro@gmail.com \
--cc=kyungmin.park@samsung.com \
--cc=linux-mm@kvack.org \
--cc=mpm@selenic.com \
/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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox