Meeting Minutes: SCE-API 2-4-03


Subject: Meeting Minutes: SCE-API 2-4-03
From: Stickley, John (john_stickley@mentorg.com)
Date: Tue Feb 04 2003 - 12:57:26 PST


Attendees:
     Duaine Pryor - Mentor Graphics
     Jan Johnson - Mentor Graphics
     Brian Bailey - Mentor Graphics
     Farid Morsi - Aptix
     Richard Newell - Aptix
     Steve Wang - Axis
     Jason Andrews - Axis
     Joe Daniels
     John Stickley - Mentor Graphics

- Duaine stated that the intent was to broaden discussion to decide
   next steps - what donations to consider, how to orchestrate
   follow-on efforts.

- Discussed motion to separate out transactor piece from the rest
   of the API as proposed by Axis.

- Duaine identified need to reach out better to transactor writers.

- Steve added that similar efforts should be made to reach out
   to TestBench writers and Vendors alike.

- Richard identified the need to be able to develop and debug models
   on platforms other than emulator platforms.

- Duaine mentioned that perhaps part of the problem is that is is
   not clear from the specification that the API is designed to work
   just as well in HDL software simulator environments as it does
   in emulator environments. Certainly it is optimized for high
   performance transaction communications which obviously benefits
   emulators but it has also been found to optimize communications
   with industry HDL software simulators as well, as compared to say
   models based on the signal oriented PLI API.

- Duaine also mentioned that there's a binary reference model
   available from Mentor that demonstrates use of SCE-MI over
   any HDL Verilog simulator.

- Axis also raised a concern that they feel there is too much
   awareness of clocks in the API and that such awareness is
   not relevant to defining an API to a C testbench.

- Richard Newell added that the goals should focus more on defining
   an interface and less on awareness of clock and clock defs.

- Brian suggested having a workshop to gather input from others
   in the industry about what is missing and what additional
   requirements need to be addressed by the API.

- Brian suggested that Duaine give a tutorial to a broader group
   about what SCE-MI is as preparation for getting more input about
   additional requirements and what is missing.

- In the audience we need, in particular, more representation from
   transactor model builders and testbench writers. Also HVL language
   providers.

- ACTION: Jan and Steve to work together to organize this workshop.

- Brian said we should target the workshop for the April or May
   timeframe before the effort becomes stale - preferably April.

- Steve said he might be able to procure funding for the workshop
   from the Accellera organization.

- We will do a followup meeting in 2 weeks to discuss plans for workshop.

- Brian updated on status of current specification
   - The spec is being sent out to all committee chairs under
     Accellera for review, feedback, and votes of approval.
   - They will have a month to do this.

-- johnS
                                                              __
                          ______ | \
______________________/ \__ / \
                                   \ H Dome ___/ |
John Stickley E | a __ ___/ / \____
Principal Engineer l | l | \ /
Verification Solutions Group | f | \/ ____
Mentor Graphics Corp. - MED C \ -- / /
17 E. Cedar Place a \ __/ / /
Ramsey, NJ 07446 p | / ___/
                                    | / /
mailto:John_Stickley@mentor.com \ /
Phone: (201)818-2585 \ /
                                      ---------



This archive was generated by hypermail 2b28 : Tue Feb 04 2003 - 13:02:56 PST