diff --git a/doc/ref_transform.rst b/doc/ref_transform.rst
index 5609fc25328ed5b0a243eaa8d33c4d1aa6b3eddc..0ce59b6300f3dcd2b198da1064b554bc4598b4fe 100644
--- a/doc/ref_transform.rst
+++ b/doc/ref_transform.rst
@@ -106,6 +106,8 @@ Creating Batches of Operations
 Finishing up
 ------------
 
+.. automodule:: loopy.transform.save
+
 .. currentmodule:: loopy
 
 .. autofunction:: preprocess_kernel
diff --git a/loopy/transform/save.py b/loopy/transform/save.py
index 5ae3669056129b2cc4c499220be9c14283757b83..603f6c4c70dcabbd9598d8107eabc5d3aa49576b 100644
--- a/loopy/transform/save.py
+++ b/loopy/transform/save.py
@@ -39,6 +39,11 @@ import logging
 logger = logging.getLogger(__name__)
 
 
+__doc__ = """
+.. autofunction:: save_and_reload
+"""
+
+
 # {{{ liveness analysis
 
 class LivenessResult(dict):
@@ -512,18 +517,18 @@ class TemporarySaver(object):
 
 # {{{ auto save and reload across kernel calls
 
-def save_and_reload(knl, **kwargs):
+def save_and_reload(knl):
     """
     Add instructions to save and reload temporary variables that are live
     across kernel calls.
 
-    The basic code transformation turns schedule segments:
+    The basic code transformation turns schedule segments::
 
         t = <...>
         <return followed by call>
         <...> = t
 
-    into this code:
+    into this code::
 
         t = <...>
         t_save_slot = t
@@ -533,8 +538,6 @@ def save_and_reload(knl, **kwargs):
 
     where `t_save_slot` is a newly-created global temporary variable.
 
-    :arg knl:
-    :arg barriers:
     :returns:
     """
     liveness = LivenessAnalysis(knl)