mm, numa: avoid setting zone_reclaim_mode unless a node is sufficiently distant
authorDavid Rientjes <rientjes@google.com>
Thu, 25 Oct 2012 20:38:08 +0000 (13:38 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 25 Oct 2012 21:37:53 +0000 (14:37 -0700)
Commit 957f822a0ab9 ("mm, numa: reclaim from all nodes within reclaim
distance") caused zone_reclaim_mode to be set for all systems where two
nodes are within RECLAIM_DISTANCE of each other.  This is the opposite
of what we actually want: zone_reclaim_mode should be set if two nodes
are sufficiently distant.

Signed-off-by: David Rientjes <rientjes@google.com>
Reported-by: Julian Wollrath <jwollrath@web.de>
Tested-by: Julian Wollrath <jwollrath@web.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: Patrik Kullman <patrik.kullman@gmail.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_alloc.c

index b0012ab..5b74de6 100644 (file)
@@ -1809,10 +1809,10 @@ static void __paginginit init_zone_allows_reclaim(int nid)
        int i;
 
        for_each_online_node(i)
-               if (node_distance(nid, i) <= RECLAIM_DISTANCE) {
+               if (node_distance(nid, i) <= RECLAIM_DISTANCE)
                        node_set(i, NODE_DATA(nid)->reclaim_nodes);
+               else
                        zone_reclaim_mode = 1;
-               }
 }
 
 #else  /* CONFIG_NUMA */