LKML Archive on lore.kernel.org
help / color / mirror / Atom feed
From: Mel Gorman <mel@csn.ul.ie>
To: akpm@linux-foundation.org
Cc: Mel Gorman <mel@csn.ul.ie>,
	linux-kernel@vger.kernel.org, linux-mm@kvack.org
Subject: [PATCH 12/12] Be more agressive about stealing when MIGRATE_RECLAIMABLE allocations fallback
Date: Thu,  1 Mar 2007 10:06:30 +0000 (GMT)	[thread overview]
Message-ID: <20070301100630.29753.64743.sendpatchset@skynet.skynet.ie> (raw)
In-Reply-To: <20070301100229.29753.86342.sendpatchset@skynet.skynet.ie>


MIGRATE_RECLAIMABLE allocations tend to be very bursty in nature like
when updatedb starts. It is likely this will occur in situations where
MAX_ORDER blocks of pages are not free. This means that updatedb can scatter
MIGRATE_RECLAIMABLE pages throughout the address space. This patch is more
agressive about stealing blocks of pages for MIGRATE_RECLAIMABLE.

Signed-off-by: Mel Gorman <mel@csn.ul.ie>
---

 page_alloc.c |   18 +++++++++++++++---
 1 files changed, 15 insertions(+), 3 deletions(-)

diff -rup -X /usr/src/patchset-0.6/bin//dontdiff linux-2.6.20-mm2-011_biasplacement/mm/page_alloc.c linux-2.6.20-mm2-012_grabbyreclaim/mm/page_alloc.c
--- linux-2.6.20-mm2-011_biasplacement/mm/page_alloc.c	2007-02-20 18:52:18.000000000 +0000
+++ linux-2.6.20-mm2-012_grabbyreclaim/mm/page_alloc.c	2007-02-20 18:54:35.000000000 +0000
@@ -806,11 +806,23 @@ retry:
 
 			/*
 			 * If breaking a large block of pages, move all free
-			 * pages to the preferred allocation list
+			 * pages to the preferred allocation list. If falling
+			 * back for a reclaimable kernel allocation, be more
+			 * agressive about taking ownership of free pages
 			 */
-			if (unlikely(current_order >= MAX_ORDER / 2)) {
+			if (unlikely(current_order >= MAX_ORDER / 2) ||
+					start_migratetype == MIGRATE_RECLAIMABLE) {
+				unsigned long pages;
+				pages = move_freepages_block(zone, page,
+								start_migratetype);
+
+				/* Claim the whole block if over half of it is free */
+				if ((pages << current_order) >= (1 << (MAX_ORDER-2)) &&
+						migratetype != MIGRATE_HIGHATOMIC)
+					set_pageblock_migratetype(page,
+								start_migratetype);
+
 				migratetype = start_migratetype;
-				move_freepages_block(zone, page, migratetype);
 			}
 
 			/* Remove the page from the freelists */

      parent reply	other threads:[~2007-03-01 10:06 UTC|newest]

Thread overview: 13+ messages / expand[flat|nested]  mbox.gz  Atom feed  top
2007-03-01 10:02 [PATCH 0/12] Group pages of related mobility together to reduce external fragmentation v28 Mel Gorman
2007-03-01 10:02 ` [PATCH 1/12] Add a bitmap that is used to track flags affecting a block of pages Mel Gorman
2007-03-01 10:03 ` [PATCH 2/12] Add __GFP_MOVABLE for callers to flag allocations from high memory that may be migrated Mel Gorman
2007-03-01 10:03 ` [PATCH 3/12] Add __GFP_MOVABLE for callers to flag allocations from low " Mel Gorman
2007-03-01 10:03 ` [PATCH 4/12] Split the free lists for movable and unmovable allocations Mel Gorman
2007-03-01 10:04 ` [PATCH 5/12] Choose pages from the per-cpu list based on migration type Mel Gorman
2007-03-01 10:04 ` [PATCH 6/12] Add a configure option to group pages by mobility Mel Gorman
2007-03-01 10:04 ` [PATCH 7/12] Drain per-cpu lists when high-order allocations fail Mel Gorman
2007-03-01 10:05 ` [PATCH 8/12] Move free pages between lists on steal Mel Gorman
2007-03-01 10:05 ` [PATCH 9/12] Group short-lived and reclaimable kernel allocations Mel Gorman
2007-03-01 10:05 ` [PATCH 10/12] Group high-order atomic allocations Mel Gorman
2007-03-01 10:06 ` [PATCH 11/12] Bias the placement of kernel pages at lower PFNs Mel Gorman
2007-03-01 10:06 ` Mel Gorman [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=20070301100630.29753.64743.sendpatchset@skynet.skynet.ie \
    --to=mel@csn.ul.ie \
    --cc=akpm@linux-foundation.org \
    --cc=linux-kernel@vger.kernel.org \
    --cc=linux-mm@kvack.org \
    --subject='Re: [PATCH 12/12] Be more agressive about stealing when MIGRATE_RECLAIMABLE allocations fallback' \
    /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

This is a public inbox, see mirroring instructions
for how to clone and mirror all data and code used for this inbox;
as well as URLs for NNTP newsgroup(s).