Minutes of Weekly Meeting, 2013-10-28

Meeting called to order: 11:07 AM EDT

1. Roll Call 

Ian McIntosh
Carl Walker
Brian Erickson
Brad Van Treuren
Patrick Au
Heiko Ehrenberg (joined 11:19)
Peter Horwood (joined 11:23)

Excused:
Adam Ley
Tim Pender

2. Review and approve previous minutes:

9/30/2013:

  • Email voting: Motion to approve by Adam, seconded by Eric. Six votes of approval, four abstentions: Motion to approve carried.

10/7/2013:

  • Email voting: Motion to approve by Eric, seconded by Heiko. Five votes of approval, Three abstentions: Motion to approve carried.

10/21/2013:

  • Draft circulated 10/21/2013.
  • Voting deferred as would not be possible to get approval.

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)
  • Harrison will attempt to come up with a table of use cases and their associated layer and what can be done at that layer. Ongoing.
  • Ian/All: Look for real world examples of boards that we could take through from board test to a system test implementation as a worked example case. Ongoing.
  • Ian - Add the previously discussed lists to the 'master' template. Ongoing.
    • Some sections need further expansion that may take time to develop.
  • All - Consider material for BTW: Either specific sub-topics related to our Templates or any other alternative subject we could present. Ongoing.
    • Ian thought this may be overtaken by events as a different format was being proposed this year. Brad felt we could still propose a topic - present a position paper and get feedback; open a discussion.
  • Ian - Update loopback templates to note assumption that instruments provide a mechanism to allow lane selection via JTAG. COMPLETE.
    • {Wiki Template shared}
    • Ian had applied the edit but wondered if a similar note was necessary for the Tx and Rx instruments. Brad agreed but unsure how to handle this.  There are several possible implementation options here and there is a temptation to open a new template for each: An instrument on each lane or a single instrument with a lane selector. That could lead to a huge growth in the number of templates.  It probably isn't enough to add a note to "Assumptions" - Brad found that the Software Patterns people dealt with similar questions by adding to the "Consequences" section. "Assumptions" should comment that the Tx and Rx lanes should be selectable via JTAG, while "Consequences" should flesh out the issues that need to be considered.

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.  Updated website - any problems?

  • Ian had set the new site live on Wednesday, October 23rd.  No problems were reported.

b.  Newsletter
c.  BTW - Can we present anything?
d.  Templates - Review what we've accomplished so far.

  • Ian pulled these three agenda items together as they had become linked.
  • {Draft Newsletter shared}
  • Ian had circulated a draft newsletter with a slightly revised layout to account for the change in the website newsfeeds, but had felt that the sidebar item on the website update was too long and had moved most of it to a page on the website with a link in the newsletter item.
  • The meeting approved of the revised layout.
  • It was suggested that the newsletter could comment on BTW.  Ian agreed but felt that if we intended to cover the templates at BTW then it could all be rolled into the one article.
  • Ian could visualize the illustrations we could feature in a template article but struggled with what text to prepare.  Brad commented that we needed to be clear on the goal of the template and where it fits in.  Referring to the iNEMI BA (Board Assisted) BIST diagram using P1687, Brad considered that SJTAG was dealing with the far left side of the diagram, relying on existing templates to flesh out what was between the left side and right side.
  • It might be good to point to a couple of the templates in the newsletter and to point out the exponential growth that might occur if they are not laid out correctly, describing the approach of addressing any implementation specific concerns in the "consequences" section, with a high level note of what needs to be done in the "Assumptions".
  • Ian will draft up some text for Brad to review and expand as necessary.
  • Brad described how IJTAG now found itself not wanting to be constrained to use only 1149.1 as an interface: Need to describe what an instrument needs to do but now how you write to the register.  This was similar to SJTAG's question of how you wrapped hands around the various interfaces for accessing devices.  Ian likened this to "a VISA layer" where you only need to know the commands and data the instrument used and could ignore the interface used.
  • In the BA BIST meeting, Brad had used his presentation from BTW 2006 and in particular the protocol stack, explaining that we had found SJTAG to be more of a software management issue than a hardware issue.  It was likely that the iNEMI BA BIST activity would move into a Phase 4 and might seek some collaboration with SJTAG.
  • The templates were starting to expose a number of issues, many shared with what iNEMI and IJTAG are now discovering, although Ian noted that we probably only started to recognize those once we had a few templates to compare.  Brad felt it was important to understand the complexity of how we talk to the instrument.
  • The main SJTAG issue isn't really the access link, it's about the "endpoint" - we can hide how we get there.  Other solutions with retargeting simply hamper getting to the abstraction.

6. Key Takeaway for today's meeting

Template construction requires care to avoid the need for vast numbers of templates to address implementation specific cases.

7. Schedule next meeting

  • Next Meeting: November 4 - 11AM EST, 4 PM GMT.  Patrick and Ian are both expect to be absent.
  • November schedule: 11, 18, 25. Thanksgiving may restrict attendance on 11/25.

8. Any other business

  • None.

9. Review new action items

  • None.

 10. Adjourn

Peter moved to adjourn at 12:01 PM EDT, seconded by Brian.

Respectfully submitted, Ian McIntosh