--- a/src/ChangeLog	Tue Jul 28 23:19:27 2009 +0900
+++ b/src/ChangeLog	Tue Jul 28 23:23:47 2009 +0900
@@ -1,3 +1,10 @@
+2009-07-28  Stephen Turnbull  <stephen@xemacs.org>
+
+	* faces.c (ensure_face_cachel_contains_charset):
+	(add_face_cachel):
+	(update_face_cachel_data):
+	Fix typos in comments and otherwise improve them.
+
 2009-07-27  Stephen J. Turnbull  <stephen@xemacs.org>
 
 	* eval.c (debug-on-error): Document interaction with process filters.

--- a/src/faces.c	Tue Jul 28 23:19:27 2009 +0900
+++ b/src/faces.c	Tue Jul 28 23:23:47 2009 +0900
@@ -1171,7 +1171,11 @@
     /* Lookup the face again, this time allowing the fallback. If this
        succeeds, it'll give a font intended for the script in question,
        which is preferable to translating to ISO10646-1 and using the
-       fixed-width fallback.  */
+       fixed-width fallback.
+
+       #### This is questionable.  The problem is that unusual scripts
+       will typically fallback to the hard-coded values as the user is
+       unlikely to have specified them herself, a common complaint. */
     new_val = face_property_matching_instance (face, Qfont,
 					       charset, domain,
 					       ERROR_ME_DEBUG_WARN, 0,
@@ -1353,7 +1357,7 @@
   Dynarr_add (w->face_cachels, new_cachel);
 
   /* The face's background pixmap have not yet been frobbed (see comment
-     int update_face_cachel_data), so we have to do it now */
+     in update_face_cachel_data), so we have to do it now */
   if (must_finish_frobbing)
     {
       int default_face = EQ (face, Vdefault_face);
@@ -1394,7 +1398,7 @@
 	 which in turn might require that the cache we're building be up to
 	 date, hence a crash. Here's a typical scenario of this:
 
-	 - a new window is created and it's face cache elements are
+	 - a new window is created and its face cache elements are
 	 initialized through a call to reset_face_cachels[1]. At that point,
 	 the cache for the default and modeline faces (normaly taken care of
 	 by redisplay itself) are null.