Minutes of Weekly Meeting, 2012-06-11

Meeting called to order: 11:06 AM EDT

1. Roll Call

Ian McIntosh
Carl Walker
Eric Cormack
Harrison Miles
Peter Horwood
Adam Ley (left 11:36)
Brian Erickson
Brad Van Treuren (joined 11:11)
Heiko Ehrenberg (joined 11:20)

Patrick Au
Tim Pender

2. Review and approve previous minutes:

05/21/2012 minutes:

  • Updated draft circulated on 05/31/2012.
  • No further corrections noted.
  • Adam moved to approve, seconded by Carl. 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
  • Heiko to prepare overview of proposed updates for 1149.1 - Ongoing.
  • Brad/Ian to prepare proposed revision to text of paragraph 4.2 of the Group Procedures for review at next meeting - COMPLETE.

4. Discussion Topics

  1. Review/approval of updates to Group Operating Procedures
    • {Ian shared draft text that had previously been circulated to the group}
    • Section 2.6 was not part of the action from the previous meeting, but a need for clarification of the status of proxy attendance and excused absence in respect of voting rights and core membership had been identified. There were no comments from the group regarding the proposed revision to this section.
    • {Brian moved to accept the proposed revision, seconded by Eric, with no objections or abstentions}
    • During review of Section 4.2 at the previous meeting it was felt that there were two type of face to face meeting: 'Public' meetings that were not to be considered part of the Group's normal business, and regular group meetings that took place face to face rather than via teleconference. However, in preparing the draft text, Brad and Ian considered that there was a third meeting type where a subgroup may meet to address a specific issue (e.g. a 'Tiger Team'): While such a meeting was part of the Group's overall business it was not involving the whole group, so it would be inappropriate to count attendance at these for the purposes of voting or core membership. Brad further explained that some other groups had found difficulty due to having no constitution for such meetings.
    • There were no comments from the group regarding the proposed revision to this section.
    • {Adam moved to accept the proposed revision, seconded by Eric, with no objections or abstentions}
    • Ian will update the Procedures posted on the website in due course.
  2. What is the current 'state of the art' for SJTAG applications?
    - What do current applications achieve and how do they do it?
    - Conclude the revisit to the SJTAG Proof of Concept demonstration and identify practical case studies from within the group.
    • {SJTAG-Demonstrations-v2.ppt shared}
    • Resuming with slide 9, Peter noted that this was showing the return of test data back into the originating tooling for diagnostics, and then setting out to develop further scenarios where this model could be applied. Adam added that this was demonstrating an early and limited capability for the round trip, with the prospect of addressing other Use Cases. If there was a desire to extend the demonstration in that way then this was still possible.
    • Harrison wondered how introducing instruments would affect the model - would it become simpler or more complex? Brad noted that would require more than just vectors (e.g. STAPL, SVF) while Ian felt it would demand totally different software tooling. Adam commented that the example looked at one single closed loop with a particular mix of tooling, although the presentation referred to expanding that to consider a different mix of tools and different use cases. Adam also suggested that any possible expansion should stay within what is possible with existing silicon, since introducing hardware that is not yet available would make it less of a 'demonstration' and more of a theoretical exercise.
    • Brad observed that demonstration showed the Test Manager to be coupled, even though it is not embedded, and that there is a hardware delineation between the Test Manager, the Test Controller or test execution engine, etc., that is independent of the architecture, and concluded that the diagram correctly shows the control elements. From a P1687 perspective, this should still hold although there may be primitives below the layer of the device.
    • Harrison was interested to discover what the Use Cases might be, given that manufacturing is generally becoming more distributed, and referenced what he termed 'the hardware confidence test' as an initial stage. Ian remarked that in his environment once the hardware confidence test was done it was likely that testing would step out of the JTAG domain and use the BIT in the system firmware and software. Harrison argued that software was typically oriented towards the 'application' rather than diagnostics. Ian agreed but noted that in most cases he was used to fault detection with location to 'box' level was all that a service technician needed on the aircraft, as the priority was to get the aircraft back in service; detailed fault diagnosis could come later. However, Ian also recognized that Brad had previously described a different scenario when early fault diagnosis might be an expectation of the customer. Brad picked up that this was significant point: While the Use Case might be the same, the difference in scenario changed the expectations for the Use Case. The need to maintain uptime on telecomm equipment meant that early determination of fault trends was vital while those trends may prove more difficult to spot on Mil-Aero sector equipment where usage and deployment can be highly variable.
    • Slide 10 suggested the possibility using alternative tooling with the demonstration, either Goepel or JTAG Technologies. Brad though that Goepel had done a demonstration although JTAG Technologies possibly had not. Slide 11 summarized further demonstrations planned for conferences during 2006.
    • Ian asked if other real world examples were available from within the group that were shareable, possibly addressing other Use Cases, or highlighting specific issues. Brad wondered if the paper where he initially presented the Test Flow and Control Language (from around 2002) was available on the SJTAG website. Ian doubted it as the earliest material collected from Ben Bennetts was from 2005. Brad would try to make this available for the next meeting.
    • Brad asked the group to consider what P1687 can provide, suggesting that there may a conflict due to a tendency to look at P1687 instruments in the same way that memory mapped functional instruments might be viewed, although trying to get and use the information may not be as simple. Harrison restated the point that current P1687 ('P1687.0') instruments were really only conceived to validate the silicon.

5. Key Takeaway for today's meeting

  • Understanding that there is a delineation between the Test Manager, the Test Controller and diagnostics.
  • In a 1687 environment, the Test Manager may present data that operates at a lower level than STAPL, SVF, etc.
  • Use Cases also have a scenario aspect that differentiates requirements.

6. Schedule next meeting

Next Meeting:
June 18.

June schedule:

7. Any other business


8. Review new action items


9. Adjourn

Eric moved to adjourn at 12:07 PM EDT, seconded by Peter.

Respectfully submitted,
Ian McIntosh