Minutes of the sv-sc sub-committee meeting, July 08, 2008

=================Attendance=============================

   0021020220 Day
   8147306817
   0000000000 Month
   7766655444
   0000000000 Year
   8888888888

--[---------a]   Arturo Salz          - Synopsys
vv[a-aaaa-aa-]   Abigail Morehouse    - Mentor
--[---------a]   Bassam Tabbara       - Synopsys
--[---------a]   Brad Pierce          - Synopsys
--[-----a-aaa]   Cliff Cummings       - Sunburst Design
--[--a--aaaaa]   Dave Rich            - Mentor Graphics
vv[aaaaaaa-aa]   Dmitry Korchemny     - Intel
--[-----a-aa-]   Don Mills            -
--[-----aaaaa]   Eduard Cerny         - Synopsys
tt[aaaaaaaaaa]   Erik Seligman        - Intel (chair)
vv[a-aaaaaaaa]   Francoise Martinolle - Cadence
vv[aaaaaaaaaa]   Gordon Vreugdenhil   - Mentor Graphics
vv[aaa-aaaaa-]   Jin Yang             - Intel
--[----aaaaaa]   John Havlicek        - Freescale
--[---------a]   Jonathan Bromley     - Doulas
--[----a--a-a]   Karen Pieper         - Accellera
vv[-aaaaaaaaa]   Lisa Piper           - Cadence
vv[aa-aaaaaaa]   Manisha Kulshrestha  - Mentor Graphics
vv[aaaaaaaaaa]   Mark Hartoog         - Synopsys
--[----aaaaaa]   Mehdi Mohtashemi     - Synopsys
vv[aaaaaaa...]   Michael Burns        - Freescale
--[a---aaaaaa]   Mirek Forczek        - Aldec
vv[aa--aaaaaa]   Neil Korpusik        - Sun Microsystems
--[--------a-]   Ray Ryan             - Mentor
--[--------aa]   Shalom Bresticker    - Intel
vv[a-aaaaaaaa]   Steven Sharp         - Cadence
--[-------aaa]   Stu Sutherland       - Sutherland HDL
--[--------aa]   Surrendra Dudani     - Synopsys
vv[aaaaaaaaaa]   Tom Thatcher         - Sun Microsystems (co-chair)
||
||---- Voting eligibility for current meeting
|------Voting eligibility for next meeting


====================Agenda==============================

Agenda:

   1.  Review the patent policy
   2.  Approve the minutes from the last meeting, available at
       http://www.eda.org/twiki/bin/view.cgi/P1800/SvScMeetingMinutes20080701
   3.  Review of Mantis items:
          * 1728: Let statements: Vote in progress. Don't forget to vote!
       Also, may need voice vote on last-minute update posted by Dmitry.
          * 2415: Ended/triggered fixes: deprecate ended, use triggered
       everywhere. Vote in progress. Don't forget to vote!
          * 2413: Inferred functions in let (Dmitry): Was going to have some
       additional review from Manisha, Dmitry after last week.
       Ready to vote?
          * 2398: Concurrent Asserts in Procedural Code: New semantics:
       (Erik/Gord). Erik's updates based on feedback to be posted before
       meeting. Ready to vote?
          * 2370: $past in Procedural Code (Jin). Jin/Lisa were going to review
       that issues are truly complete. Voice vote on closing as duplicate
       of 1698.
          * 1900: Checkers: (Dmitry/Erik/Mike/Tom): Any missing proposal
       sections? Are we ready to vote once Dmitry assembles full proposal
       & incorporates changes from this week's reviews?
          * 2182: VPI Diagrams for Checkers (Chuck Berking): TBA.
          * 2434: 1549 update: self-determined types (Gord): Assign reviewers,
       plan for vote next week.
          * 2396: New "@edge clk" proposal (Jin): Assign reviewers, plan for
       vote next week.
          * 2414: New proposal for 'let' VPI (Abi): Status? 
   4. Technical topics
          * Opens? 
====================Summary==============================
1.  Votes in progress on 1728 and 2415
2.  Preparing for vote on 1900 (part 1 only), 2398, 2370
3.  Mantis 2413 (Inferred functions in let) will be dropped

====================Notes==============================

1 Patent policy
   Move:  Neil :  Accept patent policy 
   Second: Michael
   Unanimously approved.


2.  Approve Minutes
   Move:  Neil:  Approve minutes from last meeting
   Second: Tom
   Unanimously approved

3.  Review of Mantis Items
  * 1728:  Vote in progress
       Dmitry had uploaded a change based on Mirek's comments.
       Only Tom had voted before this change.
       Tom will review change, and re-vote on the changed proposal.
  * 2415:  Vote in progress
  * 2413:  Inferred functions in let.
    Dmitry   - Drop this proposal for this PAR because of potential non-intuitive 
        behavior.
  * 2398:  Concurrent assertions in procedural code
    Erik   - Uploaded new proposal yesterday.
          - Readyto vote?
      - Will put in corrections and call a vote
  * 2370:  Past in Procedural code
    Jin      - Some style issues:  Not a major issue.
          - Could consider as a duplicate of 1698
    Erik   Will call an E-mail vote to consider as a duplicate of 1698.
    Jin      - Will put comments on 2370 to explain the issues.

  * 1900:
    Michael   - Did intend that temporal assumptions would be used for
            randomization.
      - Constraints are solved combinationally, using past values.
        for current cycle only.
    Gord   - This is a quality of results problem.
    Erik   - For example, A |-> ##5 0  Will fail.
    Gord   - In Testbench area, it is possible to determine if randomiazation
        fails.   Here we can't tell if it fails.  We just have an
        assume property fire.
    Dmitry   - What about const free variables
    Michael   - May want to leave it open as to when they change
    Gord   - If they change they change at same time other free vars are
            changing
    MIchael   - change only once per timestep
    Gord   - once per entry into observed
    MIchael   - const free var:  Could they be constrained with assumes?
            When would we solve them?
    Dmitry   - Solve at beginning of simulation.
          - Solve using assumptions that use only constant values
    MIchael   - Won't this require new syntax/semantics??
    Michael   - Like the idea of ignoring assumes for solving const free var
    Dmitry   - Formulate it as "may ignore"
    Michael   - But there is no impossible way to use assumptions
            without having unclocked assumptions

    Gord   - Question about assume set  What is "next" clock 
    Michael   - 
    Gord   - So when any of the clocks of any of the assumptions fire,
            you re-solve?
    Michael   - has to look at it.
    Michael   - If there are two assumptions, with two different clocks,
        that constrain two different free vars, a change in one clock
        would force solving of both variables . . .
    Gord   - Current wording requires this.
    Michael   - When do variables change, when are they stable.

    Michael   - If you have fast clock and slow clock, free vars constrained
            by assumptions on slow clock not guaranteed to be stable at
        fast clock ticks.

    Erik   - Call for a vote on part 1
    Dmitry   - Wanted change to BNF

  * 2182:  VPI diagrams for checkers
  * 2434:  Self-determined types
    Erik   - Call for reviewers
        Fransoise: 

  * 2396:  @edge clk
    Erik   - Call for reviewers
          Tom
          Gord:  Will also look at it
  * 2414:  New proposal for let VPI
    Erik   - Reviewers?
    Abi      - I reviewed it already
          - Wasn't there something similar in 1503?
      - Will take another look.


4.  Techinal Topics
  * Covergroups
    Tom      - Covergroups are an important feature.
            They are needed to create libraries of checker with
        functional coverage.
    Michael   - Lots of complications.
    Tom      - Will work to resolve them.

Next meeting will be  July 15

-- ErikSeligman - 09 Jul 2008