CVS difference for ais/ai-00109.txt

Differences between 1.1 and version 1.2
Log of other versions for file ais/ai-00109.txt

--- ais/ai-00109.txt	1998/09/30 00:17:18	1.1
+++ ais/ai-00109.txt	1999/03/22 23:00:33	1.2
@@ -1,4 +1,4 @@
-!standard 13.03    (55)                               98-03-27  AI95-00109/05
+!standard 13.03    (55)                               99-03-22  AI95-00109/06
 !class binding interpretation 96-04-04
 !status work item 98-04-01
 !status ARG Approved (subject to letter ballot)  7-0-3  97-11-14
@@ -11,11 +11,8 @@
 !difficulty Hard
 !subject Size and Alignment Attributes for Subtypes
 
-!summary 98-03-27
+!summary 99-03-22
 
-???Bob will update this to address the issues in the minutes of the
-April 1998 meeting.
-
 This AI addresses issues related to Size and Alignment attributes for
 subtypes.  For Size and Alignment of objects, see AI95-00051.
 
@@ -30,6 +27,12 @@
     record or array does not fit in a single word, the Size may be
     rounded up, but not past the next word boundary.
 
+    The implementation need not support an Alignment clause for a
+    subtype unless the Alignment evenly divides the size (in storage
+    elements) that the implementation would normally choose for objects
+    of the subtype.  For indefinite subtypes, the smallest such size is
+    used.
+
 Implementation Advice:
 
     If a type obeys the rules for well-defined unchecked conversion
@@ -183,12 +186,12 @@
   54   The recommended level of support for the Size attribute of subtypes is:
 
      55  The Size (if not specified) of a static discrete or fixed point
-	 subtype should be the number of bits needed to represent each
-	 value belonging to the subtype using an unbiased representation,
-	 leaving space for a sign bit only if the subtype contains
-	 negative values.  If such a subtype is a first subtype, then an
-	 implementation should support a specified Size for it that
-	 reflects this representation.
+         subtype should be the number of bits needed to represent each
+         value belonging to the subtype using an unbiased representation,
+         leaving space for a sign bit only if the subtype contains
+         negative values.  If such a subtype is a first subtype, then an
+         implementation should support a specified Size for it that
+         reflects this representation.
 
 Now, consider:
 

Questions? Ask the ACAA Technical Agent