NFS: LOOKUP_DIRECTORY is also ok with symlinks

Commit ac795161c936 (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: ac795161c936 ("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>

diff --git a/fs/nfs/dir.c b/fs/nfs/dir.c
index 7bc7cf6..75cb1cb 100644 (file)
--- a/fs/nfs/dir.c
+++ b/fs/nfs/dir.c
@@ -2010,14 +2010,14 @@ no_open:
        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)) {