Minutes of Weekly Meeting, 2012-10-01

Meeting called to order: 11:10 AM EDT

1. Roll Call

Ian McIntosh
Carl Walker
Brian Erickson
Adam Ley (left 11:59)
Brad Van Treuren
Tim Pender (joined 11:18)
Heiko Ehrenberg (joined 11:22)

Eric Cormack

2. Review and approve previous minutes:

09/17/2012 minutes:

  • One correction noted: in 2 change 'On' to 'One'.
  • Brian moved to approve with the above noted amendment, seconded by Brad. No
  • objections or abstentions.

3. Review old action items

  • Adam proposed we cover the following at the next meeting:
    • Establish consensus on goals and constraints
    • What are we trying to achieve?
    • What restrictions are we faced with?
  • 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
  • Ian: Contact Bill Eklow regarding use of the ITC mailer to promote an SJTAG Fringe Meeting at ITC. - Ongoing
  • Ian: Make Use Case diagram accessible on the website prior to next meeting. - COMPLETE

4. Discussion Topics

  1. ITC Poster - What do we present this year?
    • Heiko had confirmed outside of the meeting that he could present the SJTAG poster and could also arrange for it to be printed. This left the question of what to present. Ian commented that he and Brad had each provided the major input into the posters they presented, as they could then talk into aspects they were most comfortable with, so whatever was proposed would need to something that Heiko was also comfortable with.
    • Brad noted that a previous poster had described the hardware primitives and that he had hoped that the current discussion on Use Case primitives would have provided a topic to present but he felt that this was not yet well enough developed. Ian agreed.
    • {Tim joined}
    • The hardware primitives had been on the 2010 poster. Ian observed that it had been a promising avenue but had become stuck after starting to examine gateways.
    • {Heiko joined}
    • Ian felt that perhaps P1687 had an easier time there as they were able to define the SIB for themselves while SJTAG needed to consider a variety of existing devices with various properties and mechanisms.
    • Heiko asked if the poster was intended to present a status update or invite feedback and suggested that it could show another problem statement. Brad had been looking back through previous minutes and noted that there were a number of issues that weren't being addressed, such as the interoperability of dot7 in a dot1 infrastructure. These were discussions on topics that are coming up but that many people will not be aware of. Ian held the opinion that many users may simply be waiting for SJTAG to provide 'canned answers' to many of these issues and aren't really that interested in being part of the solution. They could be looking for a 'best practice guide' rather than a standard.
    • A second topic Brad identified was concurrency of instrumentation. While P1687 partly addressed this it was maybe not the most robust solution, and we were still left with the type of issue that Gunnar presented back in 2006 with 55 MBIST blocks that couldn't be run in parallel. We still do not have a solution to that in descriptive form.
    • Ian added that P1687 tended to consider only what was within the device but that a lot of system level test may want to use instruments in one devise in conjunction with an instrument in another device. This would require an additional layer of coordination. In some ways he thought it was similar to the issues of dot6 but without the boundaries that made that an easier problem to solve. Brad added that a lot of the solutions being touted were only applicable within a single vendor's product. When multiple vendors and strategies were involved these didn't really work and difficulties with preemphasis and post emphasis were cited as an example.
    • Ian felt that a restatement of the problem space along with descriptions of these emerging issues could be material for the poster. Brad was reminded that part of our role was to educate industry and without a standard we fall back to the education aspect.
    • Ian remarked that a recent presentation he had attended made reference to SJTAG as 'a replacement for dot5' and asked if the group felt that was what SJTAG was aiming to do. Brad noted that dot5 grew outside of simply JTAG and felt that aligning with dot5 might be dangerous. Heiko suggested that it might be acceptable if it was spun as not being as all encompassing as dot5. Brad felt that maintaining the principle of 1149.1 adherence was a major element of SJTAG. Ian noted that he could understand what people might draw a comparison: The missing 1149 overarching standard and dot2 and dot3 weren't specifically about the serial JTAG bus, but in the mind of many people '1149' means the serial test bus, irrespective of any dot notation. Ian then wondered if SJTAG should be distinguishing itself from dot5, but Brad wasn't sure if dot5 was well enough known in the industry for that to be worthwhile.
    • Heiko referred to the points in the action list that Adam had raised at the very first weekly meeting back in 2007, and in particular the latter two of those, asking if these were things we could answer. Ian noted that it would be good if we could answer these after 200+ meetings! Two surveys had tried to resolve those points amongst others but had probably provided more questions than answers. Brad commented that the Use Case analysis we had started was also intended to help.
    • Brad felt that the most promising avenue for poster topics might come from the key takeaways of the 8/13/2012 meeting where the four scenarios where SJTAG might be used were described.
    • {Minutes shared( http://www.sjtag.org/minutes/minutes120813.html)}
    • These takeaways were relevant to the questions that Adam had raised. Ian was happy to leave the poster slightly open-ended and it gave a cue for the viewer to ask questions and engage in discussion. Brad suggested that the group should consider what might now be preventing us from answering Adam's questions {ACTION}.
  2. SJTAG functional primitives: Top-down or bottom-up? (see also forum thread: http://forums.sjtag.org/viewtopic.php?f=13&t=167)
    • {Forum thread shared}
    • Ian noted that he'd posted a suggestion on a top-down decomposition to try to stimulate some discussion but had received no response. This tried to identify the various subclasses of test types that might exist within a 'structural test'. Brad considered that he could see what this was trying do but couldn't see how you could move down to the functional primitives.
    • {Adam left}
    • Ian agreed that it was difficult to do without going to the level of raw reads and writes which then lost any sense of 'intent' behind the test. Loss of intent was a deficiency in transferring to SVF or STAPL, so Brad wondered if this was moving us in the right direction.

5. Key Takeaway for today's meeting


6. Schedule next meeting

Next Meeting:
October 8 - Heiko and Eric will miss, Brad may miss.

October schedule:
8, 15, 22, 29 - Brad will miss one Monday during October.

7. Any other business


8. Review new action items

  • All: 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?

9. Adjourn

Brian moved to adjourn at 12:03 PM EDT, seconded by Heiko.

Respectfully submitted,
Ian McIntosh