CodeReview performance string padding

(defun padded-string (padding n &optional from-end)
  (declare (optimize speed)
           (type string padding))
  (with-output-to-string (s)
    (loop :for i fixnum :below n
       :with padding := (if from-end (reverse padding) padding)
       :do (princ (char padding (mod i (length padding))) s))))

(defun pad-with-string (n str &key (from-end nil))
  (declare (optimize speed)
           (type string str))
  (labels ((circular-list (elements)
             (let ((cycle (copy-list elements)))
               (nconc cycle cycle)))
           (maybe-reverse (thing)
             ;; cannot optimize here because `maybe-reverse'
             ;; takes either string or list
             (if from-end (reverse thing) thing)))
    (coerce (the list (maybe-reverse
                       (the list (loop repeat n
                                    for c character in
                                      (circular-list (coerce (the string (maybe-reverse str)) 'list))
                                    collect c))))
            'string)))

(defun test-padded-string ()
  (declare (optimize speed))
  (loop :repeat 100
     :do (padded-string (make-string 1000 :initial-element #\x) 100)))

(defun test-pad-with-string ()
  (declare (optimize speed))
  (loop :repeat 100
     :do (pad-with-string 100 (make-string 1000 :initial-element #\x))))

(defun print-test-statistics ()
  (time (test-padded-string))
  (gc)
  (time (test-pad-with-string)))

CL-USER> (print-test-statistics)
Evaluation took:
  0.002 seconds of real time
  0.002000 seconds of total run time (0.002000 user, 0.000000 system)
  100.00% CPU
  4,661,395 processor cycles
  502,288 bytes consed

Evaluation took:
  0.002 seconds of real time
  0.002000 seconds of total run time (0.002000 user, 0.000000 system)
  100.00% CPU
  4,911,066 processor cycles
  3,788,512 bytes consed

NIL
CL-USER> (print-test-statistics)
Evaluation took:
  0.002 seconds of real time
  0.002000 seconds of total run time (0.002000 user, 0.000000 system)
  100.00% CPU
  4,621,939 processor cycles
  532,752 bytes consed

Evaluation took:
  0.002 seconds of real time
  0.002000 seconds of total run time (0.002000 user, 0.000000 system)
  100.00% CPU
  4,962,756 processor cycles
  3,788,368 bytes consed

NIL