mm: when migrate_pages returns 0, all pages must have been released
authorAndrea Arcangeli <aarcange@redhat.com>
Tue, 1 Feb 2011 23:52:32 +0000 (15:52 -0800)
committerLinus Torvalds <torvalds@linux-foundation.org>
Thu, 3 Feb 2011 00:03:18 +0000 (16:03 -0800)
In some cases migrate_pages could return zero while still leaving a few
pages in the pagelist (and some caller wouldn't notice it has to call
putback_lru_pages after commit cf608ac19c9 ("mm: compaction: fix
COMPACTPAGEFAILED counting")).

Add one missing putback_lru_pages not added by commit cf608ac19c95 ("mm:
compaction: fix COMPACTPAGEFAILED counting").

Signed-off-by: Andrea Arcangeli <aarcange@redhat.com>
Signed-off-by: Minchan Kim <minchan.kim@gmail.com>
Reviewed-by: Minchan Kim <minchan.kim@gmail.com>
Cc: Christoph Lameter <cl@linux.com>
Acked-by: Mel Gorman <mel@csn.ul.ie>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/memory-failure.c
mm/migrate.c

index 548fbd70f026bfbec6c578630ce1bcd496d7cc59..75398b0bfedeb98ea2d463f0e071ec3c37e89c1d 100644 (file)
@@ -1419,6 +1419,7 @@ int soft_offline_page(struct page *page, int flags)
                ret = migrate_pages(&pagelist, new_page, MPOL_MF_MOVE_ALL,
                                                                0, true);
                if (ret) {
+                       putback_lru_pages(&pagelist);
                        pr_info("soft offline: %#lx: migration failed %d, type %lx\n",
                                pfn, ret, page->flags);
                        if (ret > 0)
index 9f29a3b7aac2242c511b8a783226c247af301993..155a2e9a8059c13249d93437589342349432e875 100644 (file)
@@ -772,6 +772,7 @@ uncharge:
 unlock:
        unlock_page(page);
 
+move_newpage:
        if (rc != -EAGAIN) {
                /*
                 * A page that has been migrated has all references
@@ -785,8 +786,6 @@ unlock:
                putback_lru_page(page);
        }
 
-move_newpage:
-
        /*
         * Move the new page to the LRU. If migration was not successful
         * then this will free the page.