SvScMeetingMinutes20080603 < P1800

P1800 Web>SystemVerilogSpecialCommittee>SvScMeetingMinutes>SvScMeetingMinutes20080603 (2008-06-06, ErikSeligman) (raw view)EditAttach
<verbatim>
Minutes of the sv-sc sub-committee meeting, June 3, 2008

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

    020220 Day
    306817
    000000 Month
    655444
    000000 Year
    888888

--[-----a]   Arturo Salz          - Synopsys
vv[aa-aa-]   Abigail Morehouse    - Mentor
--[-----a]   Bassam Tabbara       - Synopsys
--[-----a]   Brad Pierce          - Synopsys
-v[-a-aaa]   Cliff Cummings       - Sunburst Design
vv[-aaaaa]   Dave Rich            - Mentor Graphics
vv[aaa-aa]   Dmitry Korchemny     - Intel
-v[-a-aa-]   Don Mills            -
vv[-aaaaa]   Eduard Cerny         - Synopsys
tt[aaaaaa]   Erik Seligman        - Intel (chair)
vv[aaaaaa]   Francoise Martinolle - Cadence
vv[aaaaaa]   Gordon Vreugdenhil   - Mentor Graphics
vv[aaaaa-]   Jin Yang             - Intel
vv[aaaaaa]   John Havlicek        - Freescale
--[-----a]   Jonathan Bromley     - Doulas
--[a--a-a]   Karen Pieper         - Accellera
vv[aaaaaa]   Lisa Piper           - Cadence
vv[aaaaaa]   Manisha Kulshrestha  - Mentor Graphics
vv[aaaaaa]   Mark Hartoog         - Synopsys
vv[aaaaaa]   Mehdi Mohtashemi     - Synopsys
vv[aaa...]   Michael Burns        - Freescale
vv[aaaaaa]   Mirek Forczek        - Aldec
vv[aaaaaa]   Neil Korpusik        - Sun Microsystems
--[----a-]   Ray Ryan             - Mentor
--[----aa]   Shalom Bresticker    - Intel
vv[aaaaaa]   Steven Sharp         - Cadence
--[---aaa]   Stu Sutherland       - Sutherland HDL
--[----aa]   Surrendra Dudani     - Synopsys
vv[aaaaaa]   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/SvScMeetingMinutes20080520
    3. Action Item Review
           * Reviews of existing proposals:
	 
http://www.eda.org/twiki/bin/view.cgi/P1800/CurrentProposalFixesNeeded
                 o Update on the ones still TBA
                 o Discussion: Amend proposals, or just include fixes in new proposals?
           * Mike, Jin: Comments on reviews of code- will self-determined types hurt?
    4. Proposal Writing- Status, proposed ETAs, new action items?
           * 2398: New semantics for concurrent asserts in procedural code (Erik, Gord): 6/10 to write, 6/24 to pass
           * 1728: Let proposal (Dmitry): Can we assign reviewers, target 6/17 to pass
                 o Need CC expert (Abi) to help on VPI aspects
           * 1900: Checkers (Dmitry, Erik, Mike, ...): 6/24 to write,
7/8 to pass
           * 2088: Covergroups in Checkers (Tom?): 6/24 to write, 7/15 to pass
           * 2089: Final in Checkers (Tom?) : 6/24 to write, 7/15 to pass
           * 2182: VPI in Checkers (owner?): 7/1 to write, 7/15 to pass
           * 2370: $past in procedural code (owner?) : 7/1 to write,
7/15 to pass
           * New proposal for self-determined types (Gord): 6/24 to write, 7/8 to pass
           * New proposal for "@edge clk" (owner?): 6/24 to write, 7/8 to pass
           * How to communicate superseded proposals: 1995, 2110?
    5. Technical topics
           * Opens?


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

1.  Patent Policy:
     Move:  Neil  Consider the patent policy read
     Second:  Gordon
     Approved unanimously
     Ref:  http://standards.ieee.org/board/pat/pat-slideset.ppt


2.  Approve minutes.
     Move:  Gordon:  Approve the minutes from the last meeting.
     second:  Dmitry:
     Approved unanimously

3.  Meeting time next week
     Next week is DAC.  Many people will be at DAC.  Skip next week's meeting.
     Next meeting will be Tuesday, June 17.

4.  Action item review

     A.  Review of current Mantis items to determine impact of self-determined
     	type change and change for assertions in procedural code
	Erik	- Most people have reviewed their assigned proposals for
     		  effects of the self-determined types changes.
		- Haven't heard from Dave:  Mantis items 1901, 1800
		- Neil:  1757, 1737:  Will get them done soon
	Erik 	- Rather than amend every proposal, create new proposal with
     		  all amendments
	Neil	- Draft 6 will be available wednesday
     		  New proposals should be written relateive to draft 6.

     B.  Review of existing SVA code for impact of self-determined types change.
	Mike	- Unable to find any exposure: They use typed arguments in almost
		  all their SVA code.
	Jin.	Unable to find any exposure in Intel code.
	Tom:    - Sun just transitioning to SVA. Most code written is my own.
		  Very little use of parameters to properties & sequences.
		  Exposure should be minimal.


5.  Proposal Writing:
     Erik	- Deadline for SC committe work is end of July.  Need to get
     		  proposals started.

     Proposal status:
     2398	- Erik has written.  Gord should free up next week.
		  Will do what he can.  Target: draft 6/10, pass 6/24
     1728:  Let	- Dmitry has written.  Some discussion going on.
		  Some questions.  Need a VPI expert for VPI section
		  Should we allow let definitions within sequences & properties?
		  Should text remain in assertions chapter?
		    k
	Gord	- Could just keep text in assertions chapter.
	Neil	- Would be in assertions chapter, but would be general
	Mark	- Precedent for this:  bind was originally in assertions chapter,
		  later moved out.

	Reviewers:  Gord has provided comments:
	    Have everyone review it.  The proposoal is posted on Mantis

     1900:  Dmitry has main part of checker proposal
	Mike	- has piece on randome simulation of free variables
	Erik 	- Will cover checkers in procedural code.
	Tom	- Will take first part of checker proposal

     2088 & 2089:  Tom has these.
	Dmitry 	- We can just roll them into new 1900 proposal.

     2182 VPI in checkers.  No owner
	Abby	- There is a CC meeting tomorrow.  Can ask for help on this.
     2370  $past in procedural code
	Jin	- will take a look at it.
     New proposal for self-determined types:
	Gord	- Still thinking about how to actually write it.
     New proposal for "@ edge clk"
	Jin	- Will take this one.


     Superceded proposals 1995 & 2110:  Might be in Draft 6:
	Erik	- Will new proposals need to cross out 1995 changes?
	Neil	- 2110 is in feedback state:  It won't be in Draft 6.
		  We will just need to vote to mark as duplicate.
	Neil	- Put 1995 in feedback state.  It will also not appear in
		  Draft 6.


5.  Technical topics

     Discussions on Mike Burns' initial proposal on random variables in 
Checkers
     Mike	- Proposal would allow limited subset of assume properties
     		  to constrain the random values generated for random variables.
     Tom		- Would assume properties referring to combinational or
		  sequential derivatives affect assignments ot free vars?
     Mike	- Random variables are constrained only if free var name appears
     		  lexically in the assume property.
     Gord:  In example
     Mike	- Example shows a bus.  Half of bits clocked on one clock, half
		  clocked on a different clock.
		- Assume that some bits solved for, some bits constant.
     Gord	- Could you use randmode to determine if a variable is updated
		  at a given time?
     Mike	- Are all bits updated at a time?  or do variables remain
		  constant if they don't appear in an assume property clocked
		  on the current clock?
     Gord	- If you are not solving for it?  Then it's arbitrary.

     John	- Concern about when solving will occur.  May need to have
		  setup time so that value will be available for other
		  assignments/assertions.
     Mike	- Now thinking about state machine semantics:  The sequential
		  update always reads old value.
     John	- Doesn't see real need for variable split over two clocks.
     Gord	- Current rand does not allow this
     Gord	- Unpacked arrays are fine.  Only bits of packed array must
		  be solved together.
		- Can independently set randmode for different members of
		  unpacked array.
     Mike	- If defining assume property for stim gen. Fundamental propblem
		  If property defined in terms of state & free vars.
		  Free var values wouldn't show up until they are assigned
		  Can't turn assume to assert and have it pass.
     Dmitry	- Why is this a problem.
     John	- Doesn't like th is because he doesn't want to write separate
		  assertions for assume-guarantee.

     Mike	- How can we do this?  We can't determine what value ought to
		  be, then stick it back into previous time-step so that other
		  assertions can read it.


Meeeting adjourned.
Next meeting will be Tuesday, June 17.
</verbatim>



-- Main.ErikSeligman - 06 Jun 2008
Topic revision: r1 - 2008-06-06 - 19:41:39 - ErikSeligman
P1800