vmscan: only defer compaction for failed order and higher

Currently a failed order-9 (transparent hugepage) compaction can lead to
memory compaction being temporarily disabled for a memory zone.  Even if
we only need compaction for an order 2 allocation, eg.  for jumbo frames
networking.

The fix is relatively straightforward: keep track of the highest order at
which compaction is succeeding, and only defer compaction for orders at
which compaction is failing.

Signed-off-by: Rik van Riel <riel@redhat.com>
Cc: Andrea Arcangeli <aarcange@redhat.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Cc: Johannes Weiner <hannes@cmpxchg.org>
Cc: Minchan Kim <minchan.kim@gmail.com>
Cc: KOSAKI Motohiro <kosaki.motohiro@jp.fujitsu.com>
Cc: Hillf Danton <dhillf@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>

1 files changed, 11 insertions, 1 deletions

diff --git a/mm/compaction.c b/mm/compaction.c
index 36f0f61f4a24..c4b344a95032 100644
--- a/mm/compaction.c
+++ b/mm/compaction.c
@@ -695,9 +695,19 @@ static int __compact_pgdat(pg_data_t *pgdat, struct compact_control *cc)
                 INIT_LIST_HEAD(&cc->freepages);
                 INIT_LIST_HEAD(&cc->migratepages);
 
-                if (cc->order < 0 || !compaction_deferred(zone))
+                if (cc->order < 0 || !compaction_deferred(zone, cc->order))
                         compact_zone(zone, cc);
 
+                if (cc->order > 0) {
+                        int ok = zone_watermark_ok(zone, cc->order,
+                                                low_wmark_pages(zone), 0, 0);
+                        if (ok && cc->order > zone->compact_order_failed)
+                                zone->compact_order_failed = cc->order + 1;
+                        /* Currently async compaction is never deferred. */
+                        else if (!ok && cc->sync)
+                                defer_compaction(zone, cc->order);
+                }
+
                 VM_BUG_ON(!list_empty(&cc->freepages));
                 VM_BUG_ON(!list_empty(&cc->migratepages));
         }