01 Day 
29
00 Month 
87
11 Year 
00

-a Matt Maidment        - Intel   
aa Brad Pierce          - Synopsys 
aa Mark Hartoog         - Synopsys 
aa Dave Rich            - Mentor Graphics 
aa Gordon Vreugdenhil   - Mentor Graphics 
-a Alex Gran            - Mentor Graphics 
aa Heath Chambers       - Consultant/Trainer
-a Tom Alsop            - Intel 
aa Cliff Cummings       - Sunburst Design
aa Shalom Bresticker    - Intel 
aa Don Mills            - LCDM Engineering 
aa Arnab Saha           - Mentor Graphics
-a Peter Flake          - Elda Technology
-- Scott Little         - Freescale
aa Daniel Schostak      - ARM 
aa Kaiming Ho           - Fraunhofer Institute
a- Steven Sharp         - Cadence 
aa Francoise Martinolle - Cadence  
-- David Gates          - AMD
aa Eric Coffin          - Mentor Graphics
-- John Havlicek        - Freescale
-- Rishiyur Nikhil      - BlueSpec
-a Jonathan Bromley     - Verilab
a- Greg Jaxon           - Synopsys

Agenda

+ Review IEEE patent policy
  http://standards.ieee.org/board/pat/pat-slideset.ppt
 
  Reviewed.

+ Previous Meeting Minutes

  http://www.eda.org/sv-bc/minutes/sv-bc_10_07_19.txt
  
  Cliff moves to accept the minutes.
  Dave seconds.
  No opposed.
  Abstain: 
  Motion passes.

+ E-mail Vote

All issues in ballot were approved for closing

  http://www.eda.org/sv-bc/hm/10412.html

+ Mantis 2991

No-mantis 6 from Cliff presentation, requesting enum overloading as in VHDL, 
vs. the current C-like behavior, is really Mantis 2992. The reuse is allowed 
in strongly-typed E, but in SV would require convoluted rules. Said to be 
related to AOP. In VHDL can reuse enum names, because they are effectively 
overloaded functions, adapting to the different types. Not like SV enums. 

AI: Brad to follow up to Mehdi with this information.
  

+ Mantis 696

No progress. Gord says # instantiation syntax is ok. Brad would like to use 
inference from actuals instead. Gord not necessarily opposed if the actuals 
were required to be legal targets for type() operator.

AI: Brad to send proposal


+ Mantis 2310

Cliff - Choice between enhancing the current preprocessor, or adopting an 
existing SV macro language such as m4. But can't anyone already use these 
preprocessors if they want to?

Shalom - Points out the need for portability, sees value in designating a 
standard implementation of an adopted preprocessor.

Gord - Software integration not the role of P1800 standard, maybe an issue 
for SPIRIT.

Eric - Not sure what users want.

Gord - Should clean up errata before adding features.

Greg - Is it even feasible to reconverge?

Brad - Not willing to be the bad guy that breaks user legacy

Shalom - Users willing to adapt to LRM clarifications

Gord expresses skepticism

AI: Eric to start a discussion on the reflector.


+ Mantis 3053

Not even clear what's being asked for, big, diverse collection of issues.

Greg objects to instance specificity of struct types declared in non-virtual 
arrayed interface instances, which are actually homogenous in design subset. 
Gord opposed to a change.

Gord -- Central issue is how statically typed should SV be? Currently the 
universe of types and topology is known by the end of elaboration, and hardly 
any dynamic typing, which is a performance killer.

Steven - Agrees.

Greg -- And dynamic cuts you off from synthesis.


+ Mantis 3055 

AI: Gord and Mark to meet F2F when Gord in Bay Area.

Maybe Gord will be in Boston, too. If so, he'll meet with Francoise and 
Steven, too.


+ Mantis 2991

   Steven -- start with something new instead of building on top of existing
 enumerated types

   Gord -- Issue related to AOP. The E model is totally different, we'd need 
to see entire extension upfront. Maybe the testbench should be carved off as 
independent sublanguage.

Cliff motion to adjourn at 10:58