xfs: fix stale inode flush avoidance

When reclaiming stale inodes, we need to guarantee that inodes are
unpinned before returning with a "clean" status. If we don't we can
reclaim inodes that are pinned, leading to use after free in the
transaction subsystem as transactions complete.

Signed-off-by: Dave Chinner <david@fromorbit.com>
Reviewed-by: Christoph Hellwig <hch@lst.de>
Signed-off-by: Alex Elder <aelder@sgi.com>

diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 391d36b0e68c2a33b769dd66540148e94e7da3f2..ef77fd88c8e3f1b5ff766983035de5e209efe591 100644 (file)
--- a/fs/xfs/xfs_inode.c
+++ b/fs/xfs/xfs_inode.c
@@ -2842,13 +2842,9 @@ xfs_iflush(
 
        /*
         * If the inode isn't dirty, then just release the inode flush lock and
-        * do nothing. Treat stale inodes the same; we cannot rely on the
-        * backing buffer remaining stale in cache for the remaining life of
-        * the stale inode and so xfs_itobp() below may give us a buffer that
-        * no longer contains inodes below. Doing this stale check here also
-        * avoids forcing the log on pinned, stale inodes.
+        * do nothing.
         */
-       if (xfs_inode_clean(ip) || xfs_iflags_test(ip, XFS_ISTALE)) {
+       if (xfs_inode_clean(ip)) {
                xfs_ifunlock(ip);
                return 0;
        }
@@ -2871,6 +2867,19 @@ xfs_iflush(
        }
        xfs_iunpin_wait(ip);
 
+       /*
+        * For stale inodes we cannot rely on the backing buffer remaining
+        * stale in cache for the remaining life of the stale inode and so
+        * xfs_itobp() below may give us a buffer that no longer contains
+        * inodes below. We have to check this after ensuring the inode is
+        * unpinned so that it is safe to reclaim the stale inode after the
+        * flush call.
+        */
+       if (xfs_iflags_test(ip, XFS_ISTALE)) {
+               xfs_ifunlock(ip);
+               return 0;
+       }
+
        /*
         * This may have been unpinned because the filesystem is shutting
         * down forcibly. If that's the case we must not write this inode