From: Balbir Singh <balbir@linux.vnet.ibm.com>
To: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Cc: "linux-mm@kvack.org" <linux-mm@kvack.org>,
"nickpiggin@yahoo.com.au" <nickpiggin@yahoo.com.au>,
Christoph Lameter <clameter@sgi.com>,
"Lee.Schermerhorn@hp.com" <Lee.Schermerhorn@hp.com>,
Andrew Morton <akpm@linux-foundation.org>,
"Martin J. Bligh" <mbligh@google.com>
Subject: Re: [RFC][PATCH] overwride page->mapping [0/3] intro
Date: Wed, 12 Sep 2007 14:12:14 +0530 [thread overview]
Message-ID: <46E7A666.7080409@linux.vnet.ibm.com> (raw)
In-Reply-To: <20070912114322.e4d8a86e.kamezawa.hiroyu@jp.fujitsu.com>
KAMEZAWA Hiroyuki wrote:
> In general, we cannot inclease size of 'struct page'. So, overriding and
> adding prural meanings to page struct's member is done in many situation.
>
Hi, Kamezawa,
We discussed the struct page size issue at VM summit. If I remember
correctly, Linus suggested that we consider using pfn's instead of
pointers for pointer members in struct page.
> But to do some kind of precise VM mamangement, page struct itself seems to be
> too small. This patchset overrides page->mapping and add on-demand page
> information.
>
> like this:
>
> ==
> page->mapping points to address_space or anon_vma or mapping_info
>
Could you elaborate a little here, on what is the basis to decide
what page->mapping should point to?
> mapping_info is strucutured as
>
> struct mapping_info {
> union {
> anon_vma;
> address_space;
> };
> /* Additional Information to this page */
> };
>
> ==
> This works based on "adding page->mapping interface" patch set, I posted.
>
> My main target is move page_container information to this mapping_info.
> By this, we can avoid increasing size of struct page when container is used.
>
I am not against this goal, but wouldn't we end up with too many
dereferences to get to the container?
i.e, page->mapping->page_container->mem_container.
> Maybe other men may have other information they want to remember.
> This patch set implements mlock_counter on mapping_info as *exmaple*.
> (About mlock_counter, overriding page->lru may be able to be used.)
>
>
> This approach will consume some amount of memory. But I believe this *additional
> information* can be tunred off easily if the user doesn't want this.
>
> I'm glad if I can get some comments.
>
I'll review your patchset and respond if I have any comments or
suggestions.
--
Warm Regards,
Balbir Singh
Linux Technology Center
IBM, ISTL
--
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/ .
Don't email: <a href=mailto:"dont@kvack.org"> email@kvack.org </a>
next prev parent reply other threads:[~2007-09-12 8:46 UTC|newest]
Thread overview: 12+ messages / expand[flat|nested] mbox.gz Atom feed top
2007-09-12 2:43 KAMEZAWA Hiroyuki
2007-09-12 2:45 ` [RFC][PATCH] overwride page->mapping [1/3] cleanup KAMEZAWA Hiroyuki
2007-09-12 2:47 ` [RFC][PATCH] overwride page->mapping [2/3] page_mapping_info KAMEZAWA Hiroyuki
2007-09-12 2:48 ` [RFC][PATCH] override page->mapping [3/3] mlock counter per page KAMEZAWA Hiroyuki
2007-09-12 8:42 ` Balbir Singh [this message]
2007-09-12 13:28 ` [RFC][PATCH] overwride page->mapping [0/3] intro KAMEZAWA Hiroyuki
2007-09-12 19:10 ` Christoph Lameter
2007-09-12 19:12 ` Martin Bligh
2007-09-12 19:17 ` Christoph Lameter
2007-09-13 0:32 ` KAMEZAWA Hiroyuki
2007-09-13 1:05 ` Christoph Lameter
2007-09-13 1:27 ` KAMEZAWA Hiroyuki
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=46E7A666.7080409@linux.vnet.ibm.com \
--to=balbir@linux.vnet.ibm.com \
--cc=Lee.Schermerhorn@hp.com \
--cc=akpm@linux-foundation.org \
--cc=clameter@sgi.com \
--cc=kamezawa.hiroyu@jp.fujitsu.com \
--cc=linux-mm@kvack.org \
--cc=mbligh@google.com \
--cc=nickpiggin@yahoo.com.au \
/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