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, 4 insertions, 2 deletions

diff --git a/mm/page_alloc.c b/mm/page_alloc.c
index a13ded1938f0..572b93ea475c 100644
--- a/mm/page_alloc.c
+++ b/mm/page_alloc.c
@@ -1990,7 +1990,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
         if (!order)
                 return NULL;
 
-        if (compaction_deferred(preferred_zone)) {
+        if (compaction_deferred(preferred_zone, order)) {
                 *deferred_compaction = true;
                 return NULL;
         }
@@ -2012,6 +2012,8 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
                 if (page) {
                         preferred_zone->compact_considered = 0;
                         preferred_zone->compact_defer_shift = 0;
+                        if (order >= preferred_zone->compact_order_failed)
+                                preferred_zone->compact_order_failed = order + 1;
                         count_vm_event(COMPACTSUCCESS);
                         return page;
                 }
@@ -2028,7 +2030,7 @@ __alloc_pages_direct_compact(gfp_t gfp_mask, unsigned int order,
                  * defer if the failure was a sync compaction failure.
                  */
                 if (sync_migration)
-                        defer_compaction(preferred_zone);
+                        defer_compaction(preferred_zone, order);
 
                 cond_resched();
         }