Minutes of Weekly Meeting, 2015-03-16

Meeting called to order: 11:10 AM EDT

1. Roll Call

Ian McIntosh
Eric Cormack
Brian Erickson
Carl Walker
Tim Pender
Brad Van Treuren
Peter Horwood (joined at 11:17 EDT)

Excused:
Adam Ley
Michele Portolan
Heiko Ehrenberg
Bill Eklow

2. Review and approve previous minutes:

3/9/2015 minutes (draft circulated 3/9/2015):

  • No corrections noted;
  • Brian moved to approve, seconded by Tim; No objections or abstentions.

3. Review old action items

  • All: do we feel SJTAG is requiring a new test language to obtain the information needed for diagnostics or is STAPL/SVF sufficient? See also Gunnar's presentation, in particular the new information he'd be looking for in a test language (http://files.sjtag.org/Ericsson-Nov2006/STAPL-Ideas.pdf)
  • Ian: Add the previously discussed lists to the 'master' template. Ongoing.
    • Some sections need further expansion that may take time to develop.
  • Ian: Send Bill links to existing material on SJTAG Templates. COMPLETE

4. Reminders

  • Consider Adam's three points (from the action from the first weekly meeting) and suggest what is preventing us from answering those questions:
    • Establish consensus on goals and constraints
    • What are we trying to achieve?
    • What restrictions are we faced with?
  • Forum thread for discussion: http://forums.sjtag.org/viewtopic.php?f=3&t=172

5. Discussion Topics

a. Revisit draft PAR statements - continuation.

  • Ian had edited the "Illustrative SJTAG Infrastructure" diagram following the comments from the previous meeting.
  • {Diagram shared}
  • An I2C sensor, U18, was added, connected to the FPGA to as an example of an indirectly accessible instrument.  The Vector Store, U10, and ID Data Store, U11, were swapped round to allow an additional bus connection to board edge for the IS Store.  This was typical of arrangements Ian saw where an ID/ETI/Fault store was accessible by the board but could also be interrogated off-line by external equipment.  Carl had seen similar examples.
  • Brad agreed that the added features were things that needed to be supported.  Ian wondered if there was ever an explicit SPI or I2C bridge if it was more usually via some other part like a PLD, FPGA or microprocessor.  He also felt there may be better examples of an alternate access interface than the ID Bus he had shown.  Brad noted that was intended as one example and not the only example.
  • Brad was going to suggest that a store for test was something that could be added but realised that this was actually U10.  Tim noted that the diagram was missing a power-up controller or power sequencer, common on almost all modern digital boards.  It was agreed that this was also used for managing off-peak power consumption reduction, Energy Star compliance, and redundant circuit control as well as during conventional power up.
  • Brad commented that in cases like ATCA there may be a more intelligent power management scheme to command when boards should power up - slaves to some higher control entity.
  • Tim noted that in a partially powered down state, there may be ten chains on a board but only two may be operational,so there needs to be a means to bring those other chains up.  Ian added that another alternative may be for the test to adapt to the power state and only use the operational chains.
  • Ian thought he could add a Power Management device in the space between U4 and U5 {ACTION}.  Possibly the chain could be brought straight to the board edge as that was quite common although other arrangements, as noted by Peter, may come via the Gateway and Selector.
  • Brad felt that "Intelligent Instrument Clusters" were likely to become a feature of designs; a group of instruments that are controlled and co-ordinated by some higher device.  Tests may be operating on a circuit rather than a whole board.  After discussion it was felt that U18/U15 maybe represented a very basic example and that using services in the microprocessor, U3, to do at-speed RAM tests was another example, so there may not be a need to include a specific case to illustrate this.
  • Brad had been trying to look for other examples of people trying to address similar problems to ours and came across an article in the ITEA Journal on "Embedded Instrumentation Systems Architecture" by Nikita Visnevski, describing work done at GE (ITEA Journal 2009; 30: 99–110). 
  • {Article shared}
  • Brad had hoped to find something in the Automotive domain looking at their distributed systems but felt this might serve as a substitute.
  • This talks of virtual (synthetic) instruments and invokes Test Proxies, which we have discussed, and has an Access Manager and Data Manager that are concepts we're also familiar with, but introduces some items we have not considered, such as a Configuration Manager and a Run-Time Manager.  Brad thought the Configuration Manager might be of particular interest as this can determine how the Access Manager and Data Manager interact.
  • There was no doubt value in the article but it would require reading outside of the meeting; Brad only ever intended to introduce the article to the group today, not discuss it in detail.  It references a number of IEEE standards, and a lot is similar to things we've been discussing.

6. Key Takeaway for today's meeting

  • Maybe we have been overlooking a Configuration Manager.
  • Intelligent Instrument Clusters may be feature of future designs.

7. Glossary terms from this meeting

  • May need to add Configuration Manager and Intelligent Instrument Clusters (to be determined after further exploration).

8. Schedule next meeting

  • Next meeting March 23. Michele likely to be absent through March.
  • March schedule: 30.

9. Any other business

  • The newsletter is due. Brad has made progress on the Green Paper and hopes to have a draft ready this week.
  • ITC: Nothing to add.

10. Review new action items

  • Ian: Add Power Manager to illustrative system diagram.

11. Adjourn

Brad moved to adjourn at 11:54 AM EDT, seconded by Brian.

Thanks to Heiko for providing the majority of these notes.

Respectfully submitted,
Ian McIntosh