From d9e80b7de91db05c1c4d2e5ebbfd70b3b3ba0e0f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Thu, 29 Apr 2010 03:10:43 +0100 Subject: nfs d_revalidate() is too trigger-happy with d_drop() If dentry found stale happens to be a root of disconnected tree, we can't d_drop() it; its d_hash is actually part of s_anon and d_drop() would simply hide it from shrink_dcache_for_umount(), leading to all sorts of fun, including busy inodes on umount and oopsen after that. Bug had been there since at least 2006 (commit c636eb already has it), so it's definitely -stable fodder. Signed-off-by: Al Viro Cc: stable@kernel.org Signed-off-by: Linus Torvalds --- fs/nfs/dir.c | 2 ++ 1 file changed, 2 insertions(+) (limited to 'fs/nfs') diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c index be46f26c9a56..db3ad849a289 100644 --- a/fs/nfs/dir.c +++ b/fs/nfs/dir.c @@ -837,6 +837,8 @@ out_zap_parent: /* If we have submounts, don't unhash ! */ if (have_submounts(dentry)) goto out_valid; + if (dentry->d_flags & DCACHE_DISCONNECTED) + goto out_valid; shrink_dcache_parent(dentry); } d_drop(dentry); -- cgit v1.2.3