The System JTAG Working Group

Minutes of SJTAG Fringe Meeting at ITC'08, 2008-10-30

Meeting called to order at 10:35am PDT.

1. Roll Call

Bradford Van Treuren
Heiko Ehrenberg
Adam Ley
Carl Nielsen

Guests:
Michele Portolan (Alcatel-Lucent Bell Labs - Ireland) [Invited by Brad]
Paul Holowko (BAE Systems) [Invited by Heiko]

2. Resignation of Chair

Brad informed the group that his resignation as chairman of SJTAG will become official at the close of ITC. Brad is saddened by the prospects of having to resign and wants to let the group know that he has enjoyed serving as the chair of this group and that he holds the utmost honor to this team. Brad feels this group has been the most cooperative group and talented group he has worked with and wants to thank everyone for the opportunity to serve as their chairman and for each and every one's efforts.

3. Presentation of slide set shown at the ITC Poster session

Brad presented the slide set highlighting the advancements the SJTAG group has made over the past year. This slide set will be posted on the SJTAG web site. There was a good amount of discussion from the audience. In particular, Paul provided some useful contact points and suggested other APIs to look at as a basis for developing our own SJTAG APIs. These are detailed below.

• Paul suggested we look strongly at the CANopen protocol (http://www.canopen.us/) as this is the open Controller Area Network (CAN) which is implemented in hardware in microcontrollers of more than 22 chip manufacturers.
• Paul also suggested we consider using XML files as the interchange medium for our data representation that has to be portable across tooling.
• Paul suggested we should look at what the automotive industry has been doing regarding diagnostic code standardization (the error codes presented at the debug connector) when we look at defining our diagnostic code representations.
• Paul spent some time explaining what information he has to record when performing diagnostics of a weapons tank when there is a failure in the field. He needs to know the date, environmental conditions (including temperature), vibration, type of impact (if any), and many other characteristics. This data helps him determine what was going on when there was a failure reported by the system software. Likewise, we need to be aware of additional information required to deduce system failures when an SJTAG test fails.
• Paul talked about the use of TEAMS software (http://www.teamqsi.com/prods.html) to aid in modeling their systems.
• Paul further stated they do extensive fault testing and failure rate predictions to determine "failure groups of criticality" or a combination of parts that cause each system failure report. With this information, they are able to produce a division of the circuit by function to be able to identify the set of parts that could cause a specific failure condition. With boundary-scan, the granularity of failure diagnostics can be more accurate and focussed.
• Another standard Paul suggested we look at is the Smart Card industry standard. A good starting point is via GEMPLUS (http://www.gemplus.com/smart/cards/basics/).
• Paul also suggested looking at what the Java Card industry has been able to accomplish (http://java.sun.com/javacard/reference/index.jsp)

4. New TTSG policy

A new policy has been discussed at the TTSG standard chairman meeting where IEEE is now requiring each working group to establish a formal procedures and policies document describing how they do business and govern themselves. Only the P1687 group has such a document at this time. The TTSG will be providing a template for a basis for our document. No time frame was discussed as to when compliance with this ruling must be enacted.

5. Status of voting for new chair and vice chair

The group discussed the timing of the vote for a new chairman and vice chairman. All agreed it should be done at the next conference call. Heiko stated he would not be able to attend the next two meetings and requests that we open up the vote to proxy voting for people who cannot attend the next meeting. All agreed this would be a good alternate solution since we need to establish the new leadership now to be able to proceed under formal representation.

6. Recognition of former Chair and Vice Chair

Regarding the former chair and vice chair, Ian and Brad had some discussions during ITC via email since Ian was unable to attend this year. Ian proposes that the former chair (Brad) and former vice chair (Gunnar) be recognized for their valued participation in these roles and suggests the group give them the status of Emeritus Chair and Emeritus Vice Chair respectively in recognition of their value to the group.

The meeting was adjourned due to a scheduling conflict of the room at 11:45am PDT. The group was going to begin to work on identifying the data elements required to conduct the Structural Test use case as the start of analyzing the data elements against the use cases. This work will be deferred until the next meeting.

Respectfully submitted,
Bradford G. Van Treuren