mm: vmscan: fix endless loop in kswapd balancing
authorJohannes Weiner <hannes@cmpxchg.org>
Thu, 29 Nov 2012 21:54:23 +0000 (13:54 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Fri, 30 Nov 2012 16:51:17 +0000 (08:51 -0800)
commit60cefed485a02bd99b6299dad70666fe49245da7
tree4392ca5fa1cb409d3badc786bf5e2d0fed34ab1c
parentae64ffcac35de0db628ba9631edf8ff34c5cd7ac
mm: vmscan: fix endless loop in kswapd balancing

Kswapd does not in all places have the same criteria for a balanced
zone.  Zones are only being reclaimed when their high watermark is
breached, but compaction checks loop over the zonelist again when the
zone does not meet the low watermark plus two times the size of the
allocation.  This gets kswapd stuck in an endless loop over a small
zone, like the DMA zone, where the high watermark is smaller than the
compaction requirement.

Add a function, zone_balanced(), that checks the watermark, and, for
higher order allocations, if compaction has enough free memory.  Then
use it uniformly to check for balanced zones.

This makes sure that when the compaction watermark is not met, at least
reclaim happens and progress is made - or the zone is declared
unreclaimable at some point and skipped entirely.

Signed-off-by: Johannes Weiner <hannes@cmpxchg.org>
Reported-by: George Spelvin <linux@horizon.com>
Reported-by: Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>
Reported-by: Tomas Racek <tracek@redhat.com>
Tested-by: Johannes Hirte <johannes.hirte@fem.tu-ilmenau.de>
Reviewed-by: Rik van Riel <riel@redhat.com>
Cc: Mel Gorman <mel@csn.ul.ie>
Cc: <stable@vger.kernel.org>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/vmscan.c