From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Mel Gorman <mel@csn.ul.ie>
Cc: linux-mm@kvack.org, mingo@elte.hu, linux-kernel@vger.kernel.org,
nickpiggin@yahoo.com.au, lhms-devel@lists.sourceforge.net
Subject: Re: [PATCH 3/5] Light Fragmentation Avoidance V20: 003_fragcore
Date: Wed, 16 Nov 2005 11:35:17 +0900 [thread overview]
Message-ID: <437A9AE5.8070001@jp.fujitsu.com> (raw)
In-Reply-To: <20051115165002.21980.14423.sendpatchset@skynet.csn.ul.ie>
Hi,
> +/* Remove an element from the buddy allocator from the fallback list */
> +static struct page *__rmqueue_fallback(struct zone *zone, int order,
> + int alloctype)
Should we avoid this fallback as much as possible ?
I think this is a weak point of this approach.
> + /*
> + * If breaking a large block of pages, place the buddies
> + * on the preferred allocation list
> + */
> + if (unlikely(current_order >= MAX_ORDER / 2)) {
> + alloctype = !alloctype;
> + change_pageblock_type(zone, page);
> + area = &zone->free_area_lists[alloctype][current_order];
> + }
Changing RCLM_NORCLM to RLCM_EASY is okay ??
If so, I think adding similar code to free_pages_bulk() is better.
-- Kame
WARNING: multiple messages have this Message-ID (diff)
From: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
To: Mel Gorman <mel@csn.ul.ie>
Cc: linux-mm@kvack.org, mingo@elte.hu, linux-kernel@vger.kernel.org,
nickpiggin@yahoo.com.au, lhms-devel@lists.sourceforge.net
Subject: Re: [PATCH 3/5] Light Fragmentation Avoidance V20: 003_fragcore
Date: Wed, 16 Nov 2005 11:35:17 +0900 [thread overview]
Message-ID: <437A9AE5.8070001@jp.fujitsu.com> (raw)
In-Reply-To: <20051115165002.21980.14423.sendpatchset@skynet.csn.ul.ie>
Hi,
> +/* Remove an element from the buddy allocator from the fallback list */
> +static struct page *__rmqueue_fallback(struct zone *zone, int order,
> + int alloctype)
Should we avoid this fallback as much as possible ?
I think this is a weak point of this approach.
> + /*
> + * If breaking a large block of pages, place the buddies
> + * on the preferred allocation list
> + */
> + if (unlikely(current_order >= MAX_ORDER / 2)) {
> + alloctype = !alloctype;
> + change_pageblock_type(zone, page);
> + area = &zone->free_area_lists[alloctype][current_order];
> + }
Changing RCLM_NORCLM to RLCM_EASY is okay ??
If so, I think adding similar code to free_pages_bulk() is better.
-- Kame
--
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:[~2005-11-16 2:35 UTC|newest]
Thread overview: 62+ messages / expand[flat|nested] mbox.gz Atom feed top
2005-11-15 16:49 [PATCH 0/5] Light Fragmentation Avoidance V20 Mel Gorman
2005-11-15 16:49 ` Mel Gorman
2005-11-15 16:49 ` [PATCH 1/5] Light Fragmentation Avoidance V20: 001_antidefrag_flags Mel Gorman
2005-11-15 16:49 ` Mel Gorman
2005-11-15 23:00 ` Paul Jackson
2005-11-15 23:00 ` Paul Jackson
2005-11-15 23:04 ` Randy.Dunlap
2005-11-15 23:04 ` Randy.Dunlap
2005-11-16 1:36 ` Mel Gorman
2005-11-16 1:36 ` Mel Gorman
2005-11-20 14:45 ` [Lhms-devel] " Paul Jackson
2005-11-20 14:45 ` Paul Jackson
2005-11-15 16:49 ` [PATCH 2/5] Light Fragmentation Avoidance V20: 002_usemap Mel Gorman
2005-11-15 16:49 ` Mel Gorman
2005-11-15 23:36 ` Andi Kleen
2005-11-15 23:36 ` Andi Kleen
2005-11-16 1:43 ` Mel Gorman
2005-11-16 1:43 ` Mel Gorman
2005-11-16 1:52 ` Andi Kleen
2005-11-16 1:52 ` Andi Kleen
2005-11-16 2:07 ` Mel Gorman
2005-11-16 2:07 ` Mel Gorman
2005-11-22 10:13 ` Andy Whitcroft
2005-11-22 10:13 ` Andy Whitcroft
2005-11-22 10:19 ` Mel Gorman
2005-11-22 10:19 ` Mel Gorman
2005-11-22 10:22 ` Andi Kleen
2005-11-22 10:22 ` Andi Kleen
2005-11-22 10:35 ` Mel Gorman
2005-11-22 10:35 ` Mel Gorman
2005-11-22 10:48 ` KAMEZAWA Hiroyuki
2005-11-22 10:48 ` KAMEZAWA Hiroyuki
2005-11-22 19:40 ` Mel Gorman
2005-11-22 19:40 ` Mel Gorman
2005-11-22 10:54 ` KAMEZAWA Hiroyuki
2005-11-22 10:54 ` KAMEZAWA Hiroyuki
2005-11-22 11:10 ` Mel Gorman
2005-11-22 11:10 ` Mel Gorman
2005-11-22 11:35 ` Mel Gorman
2005-11-22 11:35 ` Mel Gorman
2005-11-15 16:50 ` [PATCH 3/5] Light Fragmentation Avoidance V20: 003_fragcore Mel Gorman
2005-11-15 16:50 ` Mel Gorman
2005-11-16 2:35 ` KAMEZAWA Hiroyuki [this message]
2005-11-16 2:35 ` KAMEZAWA Hiroyuki
2005-11-16 10:42 ` Mel Gorman
2005-11-16 10:42 ` Mel Gorman
2005-11-15 16:50 ` [PATCH 4/5] Light Fragmentation Avoidance V20: 004_percpu Mel Gorman
2005-11-15 16:50 ` Mel Gorman
2005-11-15 23:24 ` Paul Jackson
2005-11-15 23:24 ` Paul Jackson
2005-11-16 1:37 ` Mel Gorman
2005-11-16 1:37 ` Mel Gorman
2005-11-15 16:50 ` [PATCH 5/5] Light Fragmentation Avoidance V20: 005_configurable Mel Gorman
2005-11-15 16:50 ` Mel Gorman
2005-11-15 23:39 ` Andi Kleen
2005-11-15 23:39 ` Andi Kleen
2005-11-16 1:47 ` Mel Gorman
2005-11-16 1:47 ` Mel Gorman
2005-11-15 22:54 ` [PATCH 0/5] Light Fragmentation Avoidance V20 Paul Jackson
2005-11-15 22:54 ` Paul Jackson
2005-11-16 1:34 ` Mel Gorman
2005-11-16 1:34 ` Mel Gorman
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=437A9AE5.8070001@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 \
--cc=mel@csn.ul.ie \
--cc=mingo@elte.hu \
--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 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.