ocfs2: restore -ERANGE check in ocfs2_xattr_tree_list_index_block()

ocfs2_xattr_list_entry() returns -ERANGE when the buffer handed in by the
caller is too small to hold the whole xattr name list.  That is the normal
listxattr(2) protocol, userspace simply retries with a larger buffer, so
it is not an ocfs2 error and must not be logged as one.

Commit a46fa684fc ("ocfs2: Don't printk the error when listing too
many xattrs.") already knew this and silenced two sites, quoting

  (27738,0):ocfs2_iterate_xattr_buckets:3158 ERROR: status = -34
  (27738,0):ocfs2_xattr_tree_list_index_block:3264 ERROR: status = -34

The refactoring in commit 47bca4950b ("ocfs2: Abstract ocfs2 xattr
tree extend rec iteration process.") then rewrote
ocfs2_xattr_tree_list_index_block() around the new
ocfs2_iterate_xattr_index_block() helper and dropped that guard, so the
second message has been emitted ever since:

  ocfs2_xattr_tree_list_index_block:4513 ERROR: status = -34

The exemption in ocfs2_iterate_xattr_buckets() survived because that
function was left alone, and the new helper does check for -ERANGE itself,
which is why only the outermost caller still logs it.

Restore the missing check.  Only the log line changes: -ERANGE is still
returned to the VFS untouched, so listxattr(2) behaves exactly as before.

Link: https://lore.kernel.org/20260922094252.971631-1-joseph.qi@linux.alibaba.com
Signed-off-by: Joseph Qi <joseph.qi@linux.alibaba.com>
Signed-off-by: Andrew Morton <akpm@linux-foundation.org>
Cc: Heming Zhao <heming.zhao@suse.com>
Cc: Mark Fasheh <mark@fasheh.com>
Cc: Joel Becker <jlbec@evilplan.org>
Cc: Junxiao Bi <junxiao.bi@oracle.com>

fs/ocfs2/xattr.c

@@ -4526,7 +4526,8 @@ static int ocfs2_xattr_tree_list_index_block(struct inode *inode,
 	ret = ocfs2_iterate_xattr_index_block(inode, blk_bh,
 					      ocfs2_list_xattr_tree_rec, &xl);
 	if (ret) {
-		mlog_errno(ret);
+		if (ret != -ERANGE)
+			mlog_errno(ret);
 		goto out;
 	}