vmscan: respect higher order in zone_reclaim()
authorJohannes Weiner <hannes@cmpxchg.org>
Tue, 31 Mar 2009 22:19:38 +0000 (15:19 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Wed, 1 Apr 2009 15:59:12 +0000 (08:59 -0700)
commitbd2f6199cf9af472aeefa1b642c9f504f19e6008
treeba8cd5e7d940bcac161e28872998f86a2910362b
parentbd775c42ea5f7c766d03a287083837cf05e7e738
vmscan: respect higher order in zone_reclaim()

During page allocation, there are two stages of direct reclaim that are
applied to each zone in the preferred list.  The first stage using
zone_reclaim() reclaims unmapped file backed pages and slab pages if over
defined limits as these are cheaper to reclaim.  The caller specifies the
order of the target allocation but the scan control is not being correctly
initialised.

The impact is that the correct number of pages are being reclaimed but
that lumpy reclaim is not being applied.  This increases the chances of a
full direct reclaim via try_to_free_pages() is required.

This patch initialises the order field of the scan control as requested by
the caller.

[mel@csn.ul.ie: rewrote changelog]
Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Rik van Riel <riel@redhat.com>
Cc: Andy Whitcroft <apw@shadowen.org>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: KAMEZAWA Hiroyuki <kamezawa.hiroyu@jp.fujitsu.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmscan.c