[sv-cc] SV-CC Meeting minutes for 12/06/2006

Minutes of 12/06/2006 SV-CC Meeting.

ATTENDEES
00000000
66666666
11110000
21009988
00212131
68517306
xxxxxxxx Charles Dawson
x-x-xxxx Ralph Duncan
xxxxxxxx Jim Vellenga
xx-x-xxx Andrzej Litwiniuk
-xxxxxxx Abigail Moorhouse
xxxxxx-x Michael Rohleder
xxxxxxxx Chuck Berking
xx-xxxxx Bassam Tabbara
-xxxxxxx Francoise Martinolle
---xx-xx Amit Kohli
xxxxxxx- Ghassan Khoory


1.  Reviewed Patent information.

   - Chas reviewed the patent information.


2.  Reviewed minutes of the 11/08/2006 Meeting.
     Chuck noted that the minutes do not reflect that the first
     old business item could be closed out.  Today's agenda
     reflects that, but it was not explicit in the minutes.
   - Ralph/Chuck.  ACCEPTED (unanimous)


3.  Liaisons

   - Nothing reported.


4.  New business

   - Packed-arrays and related improvements
     Chuck was wondering who had a chance to review.   Some of
     the definitions for "essential terms" could be changed, but he
     chose them to clarify what he proposed.  Sub-part was one that
     he was open to alternatives.  Others, such as slice and element
     he likes or have already been agreed upon.  Ralph thinks it
     is okay to leave them for now, but when we get to changing the
     LRM, we will have to decide.  Make element the smallest unit
     down to the level where the data type is still consistent.  When
     you get to smaller units, you get to bits which are a different data
     type.  Ralph agrees with end results, but wants a better definition
     than "a datatype that is consistent with the definition".  Seems like
     everyone was on the same page, but that the definition needs to
     be more formal.

     Jim brought up Francoise's comments.  Her concept of a element is
     different.  Chuck needs to talk with her about it directly.

     Slice was discussed at the face to face.  Chuck has tried to
     generalize it to a part select of an array that is contiguous.
     We should ask the language teams to make their definition consistent
     with ours, if we don't like their already existing definition.

     Would need a new packed array var, packed array net and corresponding
     typespecs.  Would have to clarify elem typespec - is a shortcut to
     the typespec for the element.

     Described properties.

     Chuck also thinks the terminology should apply to unpacked arrays.
     Jim was concerned about changing the definition of elem typespec.
     Spec is unclear, but some vendors could have implemented
     it the way Francoise thinks.  Ralph doesn't think that the LRM
     uses the term element consistently.  Chuck agrees, but needs to do
     further research.  Ralph suggested that we get the other committees
     to further clarify this terminology.

     Chas suggested keeping this as a single proposal instead of multiple
     ones, and JimV suggested attaching this as a proposal for Item 1230.

   - Question on "rand" qualifier on struct elements

     This was raised as a question of whether this should be allowed in DPI.
     Chuck was under the impression that the rand qualifier should only be
     allowed in class definitions.  Dave Rich suggests that in terms of
     structs, it would be a qualifier that would be ignored.  The implication
     for Chuck is that this would mean that you could have a randomize
     method on the struct, which is not possible.

     Might want to use a definition inside and outside a class... they should
     be compatible.  Are the semantics meaningful for DPI?  Ralph thinks that
     way that the struct got the value is not important to C code, so long as
     the user recognizes that it has no effect on the C side.

     Everyone agrees that randomization should not happen on the C side.

JimV/Andrzej.  Meeting ended at 1:02pm.

5.  Reviewed items with proposals.


6.  Reviewed SV-CC items with proposals (Straw poll only).


7.  Old Business


8.  Action items
   - Ongoing review of Michael and Abi's compatibility proposal.
   - Francoise and Bassam to continue work on assignment patterns.
   - Francoise to champion adding support for typed parameters to the
     typespec diagrams.
   - Abi to champion adding support for parameterized classes.
   - Abi/JimV to champion improving the ability to compare objects.
   - Steve Dovich to determine best way to deal with issues between versions
     of the IEEE specs.

9.  Items for consideration at the next meeting (they already have proposals):

   - Item 1684 vpiParent clarification needed for complex var/net objects
   - Item 1614 P1800-2005: 27.52 no expr for disable fork objects
   - Item 1631 P1800-2005 27.14 note k has an error
   - Item 1664 IEEE 1800-2005 27.7 Note 'e': First sentence should be rewritten
   - Item 1669 P1800-2005 Sections 27.34 and 27.36 commas are inconsistent
   - Item 1632 P1800-2005 sections 27.14 note k and 27.22 note f are incompatible

10. Next meeting
   The next SV-CC meeting will be on 12/20/2006.
   The next P1800 meeting will be on 01/04/2007.

--
Charles Dawson
Senior Engineering Manager
NC-Verilog Team
Cadence Design Systems, Inc.
270 Billerica Road
Chelmsford, MA  01824
(978) 262 - 6273
chas@cadence.com