mm, mempolicy: migrate_to_node should only migrate to node

migrate_to_node() is intended to migrate a page from one source node to
a target node.

Today, migrate_to_node() could end up migrating to any node, not only
the target node.  This is because the page migration allocator,
new_node_page() does not pass __GFP_THISNODE to
alloc_pages_exact_node().  This causes the target node to be preferred
but allows fallback to any other node in order of affinity.

Prevent this by allocating with __GFP_THISNODE.  If memory is not
available, -ENOMEM will be returned as appropriate.

Signed-off-by: David Rientjes <[email protected]>
Reviewed-by: Naoya Horiguchi <[email protected]>
Cc: Mel Gorman <[email protected]>
Signed-off-by: Andrew Morton <[email protected]>
Signed-off-by: Linus Torvalds <[email protected]>

diff --git a/mm/mempolicy.c b/mm/mempolicy.c
index 4721046a134a6b8c6ab177c92cebbc67da02dc9d..69d05acfa18c83799920a143193511917b730c92 100644 (file)
--- a/mm/mempolicy.c
+++ b/mm/mempolicy.c
@@ -945,7 +945,8 @@ static struct page *new_node_page(struct page *page, unsigned long node, int **x
                return alloc_huge_page_node(page_hstate(compound_head(page)),
                                        node);
        else
-               return alloc_pages_exact_node(node, GFP_HIGHUSER_MOVABLE, 0);
+               return alloc_pages_exact_node(node, GFP_HIGHUSER_MOVABLE |
+                                                   __GFP_THISNODE, 0);
 }
 
 /*