mm: compaction: abort compaction loop if lock is contended or run too long

isolate_migratepages_range() might isolate no pages if for example when
zone->lru_lock is contended and running asynchronous compaction. In this
case, we should abort compaction, otherwise, compact_zone will run a
useless loop and make zone->lru_lock is even contended.

An additional check is added to ensure that cc.migratepages and
cc.freepages get properly drained whan compaction is aborted.

[[email protected]: Putback pages isolated for migration if aborting]
[[email protected]: compact_zone_order requires non-NULL arg contended]
[[email protected]: make compact_zone_order() require non-NULL arg `contended']
[[email protected]: Putback pages isolated for migration if aborting]
Signed-off-by: Andrea Arcangeli <[email protected]>
Signed-off-by: Shaohua Li <[email protected]>
Signed-off-by: Mel Gorman <[email protected]>
Acked-by: Minchan Kim <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>
This commit is contained in:
Shaohua Li 2012-10-08 16:32:27 -07:00 • committed by Linus Torvalds
commit e64c5237cf
2 changed files with 13 additions and 6 deletions

View file

@ -130,7 +130,7 @@ struct compact_control {
int order; /* order a direct compactor needs */
int migratetype; /* MOVABLE, RECLAIMABLE etc */
struct zone *zone;
bool *contended; /* True if a lock was contended */
bool contended; /* True if a lock was contended */
struct page **page; /* Page captured of requested size */
};