[PATCH] page migration: Fix MPOL_INTERLEAVE behavior for migration via mbind()

migrate_pages_to() allocates a list of new pages on the intended target
node or with the intended policy and then uses the list of new pages as
targets for the migration of a list of pages out of place.

When the pages are allocated it is not clear which of the out of place
pages will be moved to the new pages.  So we cannot specify an address as
needed by alloc_page_vma().  This causes problem for MPOL_INTERLEAVE which
will currently allocate the pages on the first node of the set.  If mbind
is used with vma that has the policy of MPOL_INTERLEAVE then the
interleaving of pages may be destroyed.

This patch fixes that by generating a fake address for each alloc_page_vma
which will result is a distribution of pages as prescribed by
MPOL_INTERLEAVE.

Lee also noted that the sequence of nodes for the new pages seems to be
inverted.  So we also invert the way the lists of pages for migration are
build.

Signed-off-by: Christoph Lameter <clameter@sgi.com>
Signed-off-by: Lee Schermerhorn <lee.schermerhorn@hp.com>
Looks-ok-to: Andi Kleen <ak@suse.de>
Signed-off-by: Andrew Morton <akpm@osdl.org>
Signed-off-by: Linus Torvalds <torvalds@osdl.org>

mm/mempolicy.c

@@ -552,7 +552,7 @@ static void migrate_page_add(struct page *page, struct list_head *pagelist,
 	 */
 	if ((flags & MPOL_MF_MOVE_ALL) || page_mapcount(page) == 1) {
 		if (isolate_lru_page(page))
-			list_add(&page->lru, pagelist);
+			list_add_tail(&page->lru, pagelist);
 	}
 }
 
@@ -569,6 +569,7 @@ static int migrate_pages_to(struct list_head *pagelist,
 	LIST_HEAD(moved);
 	LIST_HEAD(failed);
 	int err = 0;
+	unsigned long offset = 0;
 	int nr_pages;
 	struct page *page;
 	struct list_head *p;
@@ -576,8 +577,21 @@ static int migrate_pages_to(struct list_head *pagelist,
 redo:
 	nr_pages = 0;
 	list_for_each(p, pagelist) {
-		if (vma)
-			page = alloc_page_vma(GFP_HIGHUSER, vma, vma->vm_start);
+		if (vma) {
+			/*
+			 * The address passed to alloc_page_vma is used to
+			 * generate the proper interleave behavior. We fake
+			 * the address here by an increasing offset in order
+			 * to get the proper distribution of pages.
+			 *
+			 * No decision has been made as to which page
+			 * a certain old page is moved to so we cannot
+			 * specify the correct address.
+			 */
+			page = alloc_page_vma(GFP_HIGHUSER, vma,
+					offset + vma->vm_start);
+			offset += PAGE_SIZE;
+		}
 		else
 			page = alloc_pages_node(dest, GFP_HIGHUSER, 0);
 
@@ -585,7 +599,7 @@ redo:
 			err = -ENOMEM;
 			goto out;
 		}
-		list_add(&page->lru, &newlist);
+		list_add_tail(&page->lru, &newlist);
 		nr_pages++;
 		if (nr_pages > MIGRATE_CHUNK_SIZE)
 			break;