Minutes of Weekly Meeting, 2015-01-05

called to order at 11:05am EST

1. Roll call:

Carl Walker
Eric Cormack
Peter Horwood
Brad Van Treuren
Michele Portolan
Heiko Ehrenberg

excused absence: Bill Eklow, Ian McIntosh, Adam Ley, Tim Pender

2. Review and approve previous minutes:

  • 12/15/2014 minutes (updated draft circulated 12/16/2014)
  • deferred to next meeting

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.
    •  Some sections need further expansion that may take time to develop.
  • Brad: Contact Gunnar for clarification on the scope of his “system”.
    • Done - see discussion

4. Reminders

  • Call for volunteers, or if none, re-affirmation of officers
  • 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) clarification on the scope of Gunnar’s “system” (in reference to Gunnar’s 2005 ITC paper)

  • Brad shared slide from Gunnar (“In-field Test.pdf”, see http://forums.sjtag.org/viewtopic.php?f=13&t=720);
  • Brad noted that this view/scope of a system is very similar to how he [Brad] looks at this topic.
  • A key aspect is the use of agents. There are two use cases discussed: factory test RBS production test (board level) and in-field test, both using the same software.
  • Gunnar's comments:
    • RBS production test (board level):
      • No backplane, only serial links plus a test connector.
      • Slide 10 is fairly representative, except that  we don’t have I2C, but instead HDLC and Ethernet.
      • The JTAG control system is a commercial product (including TAP Master HW), also 1687 capable.
      • SPI is used only for booting the system.
      • HDLC is used to send commands to the Board Power Managager (BPM) on the board. The SW is
        board resident. Evaluation and measurement control is done by the Test Step Controller.
      • For functional test, all low level SW for HW operations is downloaded to the board over Ethernet,
        and then low level commands are sent to the board. Evaluation is done by the Test Step Controller.
      •  There is no sharp distinction between the Test Step Controller and the Test Manager.
    • RBS in-field test:
      • Test can e.g. be triggered by automatic fault escalation or by remote operator control.
      • The enclosed slide illustrate the architecture. The HwtestMgr starts Test Agents driven by a config
        file.
      • Test Agents consist of one or more Test Methods (test cases), which in turn call low level functions
        for HW operations (same or similar to those used in production test).
      •  At failure, the Test Methods write entries to a persistent log.
  • Brad notes that it seems that this limitation of not delineating between test manager and test controller in the factory test setup is caused by the off-the-shelf tools that are available.

b) 1687.2

  • Michele doesn’t have specific details as to what 1687.2 might entail; Michele has spoken with various people in the IJTAG committee but there was no specific reference to topics for a 1687.2;
  • Brad received an email from Al Crouch with a call to find interested parties to join a P1687.1 working group with the purpose to define other serial controllers to drive a 1687 network; But Brad hasn’t heard from anyone in the iJTAG community about a 1687.2; Heiko notes that Ian perhaps referred to this 1687.1 effort;
  • {Post meeting note: Ian did mean to refer to 1687.1 here}
  • Email from Al Crouch:
    I will soon be collecting up interested parties for 1687.1 to begin the effort of defining how to use other serial controllers (i.e. SPI, I2C, etc.) to drive a 1687 network. This means that embedded instruments can then be used on any chip with a serial interface on a board. We expect the effort to include some HW definition, some extensions to ICL to define the new AccessLink, and maybe some new PDL extensions to handle coordinating embedded instruments behind different controller interfaces (there may be different latency with different controllers).

c) Revisit draft PAR statements - continuation.

  • There was good discussion on the email reflector about this topic. Unfortunately, the key contributors aren’t on today’s call, so we’ll defer further discussion until the next meeting.

6. Key Takeaway for today's meeting

  • For production test, Gunnar did not delineate between test manager and test controller, but he did for in-field architecture.

 7. Glossary terms from this meeting

  • none   

8. Schedule next meeting

  • January 12, 2015 @ 11am EST 
  • January schedule - 12, 19 (Brad might be missing), 26

9. Any other business

  • Reminder on re-affirmation of officers.

10. Review new action items

  • post Gunnar’s submission (text and slide) to SJTAG Forum ?

11. Adjourn

  • Eric moves to adjourn, Peter seconds
  • meeting adjourned at 11:43am EST