IEEE P1076.1 Subcommittee on VHPI-AMS - June 24, 2010 8am PDT


Attendees

  • David Smith, champion
  • Peter Ashenton - absent
  • Yaseen Zaidi, note taker
  • Ernst Christen

Agenda

  • Call to order
  • Approval of agenda
  • Review and approve minutes from May 20 meeting
  • Review action items and any content provided from them
  • Discussion
  • Next Meeting
  • AOB
  • Adjourn

1. Call to order

David opened the meeting and started presenting the slides

2. Approval of agenda

The agenda was discussed and there were no changes.

3. Review and approve minutes from May 20 meeting

Minutes were approved with no objections.

4. Review action items and any content provided from them

  • David - pursue contact with Francois to see if there is a donation
    • Cadence has a donation and they are working to pull it together and pass it by their lawyers.
  • David - update scope with feedback from meeting and post on site
    • See discussion for details. New scope is posted on TWiki in the same place the minutes are posted.
  • Peter - develop initial analysis of objects that need to be added to existing VHP
    • Peter has been occupied and not able to make progress.

5. Discussion

  • David briefly highlighted the main points of his slides from last telco for Ernst who had missed it
  • David stressed on a clear and concise scope.
  • The sense of the working group is we should extend VHPI to VHPI-AMS within the bounds of VHPI, meaning that concepts like meta-simulation, foreign models, cosimulation, backplane interface and debugger should not be directly addressed by VHPI-AMS, although these applications are the ultimate goal of VHPI and VHPI-AMS. Because VHPI does not address the applications, VHPI-AMS should not either. The use cases, though, should give sufficient insight to the user how VHPI-AMS extensions can contribute in building analog simulation applications.
  • Discussion headed into the ability of writing to quantities e.g. adjusting the tuning frequency of an oscillator during simulation. Two thoughts prevailed.

  • Writing to quantities in circuit simulation is analogous to network topology change - insertion and removal of sources, shorts and opens as suggested in Martin Vlach's paper. However, the working group agrees that any change to the information model during simulation that implies reelaboration is beyond VHPI-AMS. The VHPI-AMS should not provide any functionality that results in new topological configuration, therefore VHPI-AMS should not be viewed as a means of simulation control as this capability in VHPI itself is rather limited. David mentioned we should shoot for some debugger capability in terms of start/restart, which may be further extended if later there are requirements.
  • Writing to quantities without inducing reelaboration. David said that SPICE allows this, in theory VHPI-AMS should handle this task. Ernst said that in SPICE such resetting of objects is usually done in DC operations and not in transient analysis. Ernst mentioned that VHPI-AMS may provide writing ability within the context of language semantics e.g. VHDL-AMS allows modeling of discontinuous behavior using BREAK statement which may be used for initial conditions or reinitialization, and GENERIC for static objects. However, the quantities are the unknowns in the ODEs, and resetting them during iterative numerical calculations is unusual. Ernst suggested we need to get a better understanding of what actually writing to quantities means for the solver and simulation cycle.
  • We looked at Ernst's presentation from Jan 2010 on Verilog-AMS VPI for ideas.
  • We discussed other possible actions for the future:
    • Review of objects for extension
    • Review of functions for extension
    • Evaluate PutValue for analog

5. Next Meeting

Thursday July 22, 2010, 8am PDT.

Action Items:

  • David - Write Peter about a time line for object analysis.
  • David - Do analysis of VHPI-AMS functions.
  • David - Keep in touch with Cadence on possible donation.
  • Peter - develop initial analysis of objects that need to be added to existing VHPI.
  • Ernst - Check with IEEE about the donation rights form and requirements involved in the donation process. Also, continue pursuing a possible donation.
  • Ernst - Investigate technical feasibility of writing to quantities and requirement to related use cases - look at PutValue.
  • Ernst - Move all presentations at an easy access location in TWiki - break presentations from minutes.
  • Yaseen - Do an analysis of VPI for Verilog-AMS specific tasks.

6. AOB

7. Adjourn

Meeting was adjourned at 9am PDT.

-- DavidSmith - 2010-07-22

Topic attachments
I Attachment Action Size Date Who Comment
PowerPointpptx VHPI_SC_-24_June_2010.pptx manage 42.9 K 2010-07-22 - 00:41 DavidSmith  
Topic revision: r2 - 2016-01-22 - 22:15:34 - ErnstChristen
 
Copyright © 2008-2019 by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding TWiki? Send feedback