NFS: LOOKUP_DIRECTORY is also ok with symlinks

commit e0caaf7 upstream.

Commit ac79516 (NFSv4: Handle case where the lookup of a directory
fails) [1], part of Linux since 5.17-rc2, introduced a regression, where
a symbolic link on an NFS mount to a directory on another NFS does not
resolve(?) the first time it is accessed:

Reported-by: Paul Menzel <pmenzel@molgen.mpg.de>
Fixes: ac79516 ("NFSv4: Handle case where the lookup of a directory fails")
Signed-off-by: Trond Myklebust <trond.myklebust@hammerspace.com>
Tested-by: Donald Buczek <buczek@molgen.mpg.de>
Signed-off-by: Anna Schumaker <Anna.Schumaker@Netapp.com>
Signed-off-by: Greg Kroah-Hartman <gregkh@linuxfoundation.org>

fs/nfs/dir.c

@@ -1982,14 +1982,14 @@ int nfs_atomic_open(struct inode *dir, struct dentry *dentry,
 	if (!res) {
 		inode = d_inode(dentry);
 		if ((lookup_flags & LOOKUP_DIRECTORY) && inode &&
-		    !S_ISDIR(inode->i_mode))
+		    !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode)))
 			res = ERR_PTR(-ENOTDIR);
 		else if (inode && S_ISREG(inode->i_mode))
 			res = ERR_PTR(-EOPENSTALE);
 	} else if (!IS_ERR(res)) {
 		inode = d_inode(res);
 		if ((lookup_flags & LOOKUP_DIRECTORY) && inode &&
-		    !S_ISDIR(inode->i_mode)) {
+		    !(S_ISDIR(inode->i_mode) || S_ISLNK(inode->i_mode))) {
 			dput(res);
 			res = ERR_PTR(-ENOTDIR);
 		} else if (inode && S_ISREG(inode->i_mode)) {