xfs
[Top] [All Lists]

[PATCH] xfs: fix stale inode flush avoidance

To: xfs@xxxxxxxxxxx
Subject: [PATCH] xfs: fix stale inode flush avoidance
From: Dave Chinner <david@xxxxxxxxxxxxx>
Date: Mon, 11 Jan 2010 22:45:21 +1100
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@xxxxxxxxxxxxx>
---
 fs/xfs/xfs_inode.c |   21 +++++++++++++++------
 1 files changed, 15 insertions(+), 6 deletions(-)

diff --git a/fs/xfs/xfs_inode.c b/fs/xfs/xfs_inode.c
index 391d36b..ef77fd8 100644
--- 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;
        }
@@ -2872,6 +2868,19 @@ 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
         * to disk, because the log record didn't make it to disk!
-- 
1.6.5

<Prev in Thread] Current Thread [Next in Thread>