linux-mm.kvack.org archive mirror
 help / color / mirror / Atom feed
From: Hiroyuki KAMEZAWA <kamezawa.hiroyu@jp.fujitsu.com>
To: Hiroyuki KAMEZAWA <kamezawa.hiroyu@jp.fujitsu.com>
Cc: Linux Kernel ML <linux-kernel@vger.kernel.org>,
	linux-mm <linux-mm@kvack.org>,
	LHMS <lhms-devel@lists.sourceforge.net>
Subject: Re: [Lhms-devel] [RFC] buddy allocator without bitmap(3) [1/3]
Date: Fri, 03 Sep 2004 08:54:31 +0900	[thread overview]
Message-ID: <4137B2B7.8080109@jp.fujitsu.com> (raw)
In-Reply-To: <4136D318.9060102@jp.fujitsu.com>

> New function: calculate_aligned_end()
> 
> calculate_aligned_end() removes some pages from system for removing invalid
> mem_map access from __free_pages_bulk() main loop.(This is in 4th patch)
> 
This is an illustration of the effects of calculate_aligned_end().

Examples for MAX_ORDER=4 is here.
In this case, an alignment of memmap is (1 << (4-1))=8

[unaligned end address case]

Consider contiguous mem_map from index 0 to index 19.
mem_map[16-19] is unaligned.

pfn     0           4           8          12            16 17 18 19
         -------------------------------------------------------------
order 0 |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  -- out of range --
         -------------------------------------------------------------
order 1 |     |     |     |     |     |     |     |     |     |     |
         -------------------------------------------------------------
order 2 |           |           |           |           |           |
         ------------------------------------------------------------
order 3 |                       |                       |
         -------------------------------------------------
         <----------------------> <---------------------> <---------??? ---->
In this case, invalid mem_map access will occur during

(1) coalescing page 16 with page 20 in order=2. <- this means memory access to page 20.

calculate_aligned_end() removes page 19.

  pfn     0           4           8          12            16 17 18 19
         -------------------------------------------------------------
order 0 |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  | X|   -- out of range --
         -------------------------------------------------------------
order 1 |     |     |     |     |     |     |     |     |     |
         -------------------------------------------------------
order 2 |           |           |           |           |
         -------------------------------------------------
order 3 |                       |                       |
         -------------------------------------------------
         <----------------------> <--------------------->

         page 19 is removed.
         -> page 18 and page 19 cannot be coalesced.
         -> page 16 - page 19 cannot be coalesced.
         -> accessing invalid page 20 will not occur.


[unaligned start address case]
Consider a mem_map begins from index 2.

pfn     0     2     4           8          12           16
               -------------------------------------------------------------------
order 0       |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
               -------------------------------------------------------------------
order 1       |     |     |     |     |     |     |     |     |     |     |     |
               --------------------------------------------------------------------
order 2             |           |           |           |           |           |
                     -------------------------------------------------------------
order 3                         |                       |                       |
                                 -------------------------------------------------

In this case, invalid mem_map access will occur during
	(1) coalescing page 2 and page 0 in order=1
	(2) coalescing page 4 and page 0 in order=2

calculate_aligned_end() removes page 2 and 4.

pfn     0     2     4           8          12           16
               -------------------------------------------------------------------
order 0       |x |  |x |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |  |
               -------------------------------------------------------------------
order 1                   |     |     |     |     |     |     |     |     |     |
                           --------------------------------------------------------
order 2                         |           |           |           |           |
                                 -------------------------------------------------
order 3                         |                       |                       |
                                 -------------------------------------------------

	page 2 is removed.
	-> page 2 and page 3 cannot be coalesced in order=0
	-> accessing invalid page 0 in order=1 will not occur.
	
	page 4 is removed.
	-> page 4 and page 5 cannot be coalesced in order=0.
	-> page 4 and page 6 cannot be coalesced in order=1.
	-> accessing invalid page 0 in order=2 will not occur.

Thanks.
--Kame
-- 
--the clue is these footmarks leading to the door.--
KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>

--
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:"aart@kvack.org"> aart@kvack.org </a>

      reply	other threads:[~2004-09-02 23:49 UTC|newest]

Thread overview: 2+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2004-09-02  8:00 Hiroyuki KAMEZAWA
2004-09-02 23:54 ` Hiroyuki KAMEZAWA [this message]

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=4137B2B7.8080109@jp.fujitsu.com \
    --to=kamezawa.hiroyu@jp.fujitsu.com \
    --cc=lhms-devel@lists.sourceforge.net \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.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 a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox