The function `fill-region-as-paragraph' leaves a temporal marker which
isn't cleared in the buffer.

Here is the structure of `fill-region-as-paragraph':
----------------------------------------------------------------------
(defun fill-region-as-paragraph (from to &optional justify
				      nosqueeze squeeze-after)
[...]
    (setq to (copy-marker (point) t))
[...]
  (if (not (> to (point)))
      nil ;; There is no paragraph, only whitespace: exit now.
[...]
      (goto-char to)
      (unless (eobp) (forward-char 1))
      ;; Return the fill-prefix we used
      fill-prefix)))
----------------------------------------------------------------------
With this code, temporal marker bound to local variable `to' is left in
the buffer. According to elisp reference `(elisp) Overview of Markers',
it is recommended to clear such stale marker with suitable `(set-marker
MARKER nil)':
,----
|    Insertion and deletion in a buffer must check all the markers and
| relocate them if necessary.  This slows processing in a buffer with a
| large number of markers.  For this reason, it is a good idea to make a
| marker point nowhere if you are sure you don’t need it any more.
`----
I'd propose to apply the attached patch to accomplish that.

Regards,
Ikumi Keita