Minutes of Weekly Meeting, 2008-08-18

Meeting called to order at 8:15am EDT

1. Roll Call (Participants):

Brad Van Treuren
Ian McIntosh
Carl Walker
Carl Nielsen
Peter Horwood [had to leave at 8:55am EDT]
Eric Cormack
Tim Pender
Heiko Ehrenberg

2. Review and approve minutes

8/11/2008 minutes approved with corrections (Eric moved, Ian second) [Corrections listed below]

  • [Tim or Patrick?] Limiting ourselves to 4 or 5 wire JTAG we need to have some support of parallel ...
    (Section 4d) Tim.
  • Heiko tendative moderator (Ian backup, if Heiko can't make it) (Section 5) Typo - "tentative".
  • Heiko will represent SJTAG at BTW2008 if no one else from the working group attends the workshop ...
    (At end of section 6) Duplication of text 4 lines earlier.

3. 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?
  • Establish whether TRST needs to be addressed as requirements in the ATCA specification if it is not going to be managed globally (All)
  • Adam review ATCA standard document for FRU's states
  • Brad define agenda for 8/27 meeting
  • Brad will send out revised Scope statement - done
  • Ian will work on rewording Purpose statement - is on Wiki [link on main Wiki page; http://wiki.sjtag.org/index.php?title=Scope_and_Purpose]

4. Discussion Topics

    1. Results and Status from SJTAG Survey Activity
      • [Ian] no new activity

 

    1. Status and review of white paper sections
      • Section 1: no additions / changes to Volume 1
      • Section 2: Heiko made some minor edits to Volume 2
      • Section 3: no additions / changes to Volume 3
      • Language Section: in preparation

 

    1. Status of ITC poster session
      • Draft of slide presentation due by Aug 29. Almost complete. Brad working with Ian to finalize slides.

 

  1. Additional comments regarding to proposed scope and purpose
    • uploaded by Ian to Wiki [link on main Wiki page; http://wiki.sjtag.org/index.php?title=Scope_and_Purpose]
    • [Brad] scope has been reworded based on comments from previous meetings
    • [Eric] why would we (want to) have multiple description languages?
    • [Brad] I stated this in the scope as description language(s) because there are many sources of input to a system description. We have BSDL, which is obvious. We also have descriptions relating to 1532 and in the future P1687, and P1581. There are also cluster models and memory model descriptions for devices in the system. All these define what can be done with an 1149.1 accessible element in the system. Board test tools today deal with these elements. SJTAG has be be aware of these elements and how they are described as well. Different standards may specify different languages Additional features (such as non-BScan circuitry, or functional access memory devices, for example) may be described with tool vendor proprietary languages. Different systems may have different requirements for a description language.
    • [Brad] Also, the system may change in configuration over time (even once it is deployed). What is different about a system over a board is elements may be removed (even hot swapped) and added. Thus, the topology of the circuit is not a static graph, but a dynamically changing graph over time. Carrier boards have to deal with the same problem since they are really sub-systems with their mezzanine boards.
    • [Ian] One could envision some description that would be able to define interconnect testing at the system level between boards. But this is difficult if the boards are changing.
    • [Brad] We may have to partition tests into pieces that can be autonomous and those that are not. This is why I presented my paper on what we call Boundary-Scan Plug-n-Play back at ITC 2005. We are able to keep the board level test vectors resident on the UUT and extract the test information from the UUT using the multidrop connection. The data is then applied to the UUT through the multidrop interface. Therefore, the system test software does not have to change to accommodate a new design. For board level scoped operations, I think we can deal with these issues. The system level issues are much harder.
    • [Ian] Not all designs may be able to afford placing the test data on the UUT, so alternatives need to be considered.
       
    • [Ian] As we brought out in the last meeting is the main function of the Purpose was in the last sentence. The other parts were needing to reside in a description section. It basically comes down to a single sentence now.
    • [Ian] Purpose has been shortened and parts of it should be moved into a descriptive section ("Need for Project" section of the PAR)

5. Schedule next meetings:

Wednesday, August 27th, 2008, 8:15am EDT <- NEW DATE
Monday, September 8th, 2008, 8:15am EDT
Monday, September 15th, 2008, 8:15am EDT
Wednesday, September 24th, 2008, 8:15am EDT

6. Any other business

  • PAR Application (Ian)
    • [Ian] Are we still planning to submit the PAR prior to ITC?
    • [Brad] This is the plan, at least by the end of the year.
    • [Ian] We certainly have enough information available and don’t have to submit the form, but can at least fill it out. The important thing is we could get the reference number (not the P number) to work from.
  • VXI Presentation (Eric)
    • [Eric] Autotestcon is in September. Is there any status.
    • [Brad] Nothing available yet.

7. Review of new action items

  • Brad to define agenda for 8/27 meeting

8. adjourned at 9:15am EDT

moved by Ian, second by Eric

Thanks again to Heiko for his assistance.

Respectfully submitted,
Bradford Van Treuren