We shouldn't print errors returned from vfs_follow_link(). This was causing spurious errors to show up in the logs. Signed-off-by: Mark Fasheh <mark.fasheh@oracle.com>
@@ -158,8 +158,7 @@ static void *ocfs2_follow_link(struct dentry *dentry,
}
status = vfs_follow_link(nd, link);
- if (status && status != -ENOENT)
- mlog_errno(status);
+
bail:
if (page) {
kunmap(page);