swap: redirty page if page write fails on swap file
authorJerome Marchand <jmarchan@redhat.com>
Mon, 29 Apr 2013 22:08:47 +0000 (15:08 -0700)
committerLinus Torvalds <torvalds@linux-foundation.org>
Mon, 29 Apr 2013 22:54:39 +0000 (15:54 -0700)
Since commit 62c230bc1790 ("mm: add support for a filesystem to activate
swap files and use direct_IO for writing swap pages"), swap_writepage()
calls direct_IO on swap files.  However, in that case the page isn't
redirtied if I/O fails, and is therefore handled afterwards as if it has
been successfully written to the swap file, leading to memory corruption
when the page is eventually swapped back in.

This patch sets the page dirty when direct_IO() fails.  It fixes a
memory corruption that happened while using swap-over-NFS.

Signed-off-by: Jerome Marchand <jmarchan@redhat.com>
Acked-by: Johannes Weiner <hannes@cmpxchg.org>
Acked-by: Mel Gorman <mgorman@suse.de>
Cc: Hugh Dickins <hughd@google.com>
Cc: <stable@vger.kernel.org> [3.6+]
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Signed-off-by: Linus Torvalds <torvalds@linux-foundation.org>
mm/page_io.c

index 8e0e5c0..eb3300f 100644 (file)
@@ -231,6 +231,8 @@ int __swap_writepage(struct page *page, struct writeback_control *wbc,
                if (ret == PAGE_SIZE) {
                        count_vm_event(PSWPOUT);
                        ret = 0;
+               } else {
+                       set_page_dirty(page);
                }
                return ret;
        }