From a8b0026847b8c43445c921ad2c85521c92eb175f Mon Sep 17 00:00:00 2001 From: Al Viro Date: Mon, 20 Nov 2023 20:02:11 -0500 Subject: rename(): avoid a deadlock in the case of parents having no common ancestor ... and fix the directory locking documentation and proof of correctness. Holding ->s_vfs_rename_mutex *almost* prevents ->d_parent changes; the case where we really don't want it is splicing the root of disconnected tree to somewhere. In other words, ->s_vfs_rename_mutex is sufficient to stabilize "X is an ancestor of Y" only if X and Y are already in the same tree. Otherwise it can go from false to true, and one can construct a deadlock on that. Make lock_two_directories() report an error in such case and update the callers of lock_rename()/lock_rename_child() to handle such errors. And yes, such conditions are not impossible to create ;-/ Reviewed-by: Jan Kara Signed-off-by: Al Viro --- fs/nfsd/vfs.c | 4 ++++ 1 file changed, 4 insertions(+) (limited to 'fs/nfsd') diff --git a/fs/nfsd/vfs.c b/fs/nfsd/vfs.c index fbbea7498f02..a99260c3f9bc 100644 --- a/fs/nfsd/vfs.c +++ b/fs/nfsd/vfs.c @@ -1813,6 +1813,10 @@ retry: } trap = lock_rename(tdentry, fdentry); + if (IS_ERR(trap)) { + err = (rqstp->rq_vers == 2) ? nfserr_acces : nfserr_xdev; + goto out; + } err = fh_fill_pre_attrs(ffhp); if (err != nfs_ok) goto out_unlock; -- cgit v1.2.3