mm: zone_reclaim: after a successful zone_reclaim check the min watermark

If we're in the fast path and we succeeded zone_reclaim(), it means we
freed enough memory and we can use the min watermark to have some
margin against concurrent allocations from other CPUs or interrupts.

mm/page_alloc.c

@@ -2220,8 +2220,26 @@ get_page_from_freelist(gfp_t gfp_mask, unsigned int order, int alloc_flags,
 			case ZONE_RECLAIM_FULL:
 				/* scanned but unreclaimable */
 				continue;
+			case ZONE_RECLAIM_SUCCESS:
+				/*
+				 * If we successfully reclaimed
+				 * enough, allow allocations up to the
+				 * min watermark (instead of stopping
+				 * at "mark"). This provides some more
+				 * margin against parallel
+				 * allocations. Using the min
+				 * watermark doesn't alter when we
+				 * wakeup kswapd. It also doesn't
+				 * alter the synchronous direct
+				 * reclaim behavior of zone_reclaim()
+				 * that will still be invoked at the
+				 * next pass if we're still below the
+				 * low watermark (even if kswapd isn't
+				 * woken).
+				 */
+				mark = min_wmark_pages(zone);
+				/* Fall through */
 			default:
-				/* did we reclaim enough */
 				if (zone_watermark_ok(zone, order, mark,
 						ac->classzone_idx, alloc_flags))
 					goto try_this_zone;