Minutes of Weekly Meeting, 2007-11-12


Peter Horwood
Carl Nielsen
Al Holliday
Adam Ley
Ian MacIntosh
Jim Webster
Brad Van Treuren

Meeting was called to order at 8:06am EST

Review of meeting minutes for 11/05/2007;

Approved as is (moved by Al, second by Ian)

Discussion of open 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?
      • still ongoing
  • Establish whether TRST needs to be addressed as requirements in the ATCA specification if it is not going to be managed globally (All)
    • still to be resolved
  • Provide feedback of more use cases not yet identified to Brad (All)
    • awaiting feedback ...
  • Review tables (Goals vs. use case matrix) on slides 38-41; (All)
    • awaiting feedback ...

Overview of new SJTAG Web Forum by Ian McIntosh

  • Ian distributed a slide set that was sent out with the meeting agenda
  • Must register to create login to forum
  • Suggest the usage of user names as close as possible to real names
  • Use company names and country for location in registration
  • Set up your locality based on your local time zone
  • Forum headings can be restricted to our own group and can support General SJTAG group as public => Do we need private user groups?
  • Need moderators for each forum
  • Links to other files need to be set up as URL links
  • Private messaging is available
  • The File Manager requires a special login
  • The browse button in the File Manager is used to select and upload a file from your local computer => Information from Ben's SJTAG site has been transferred to the new location => All need to check and add any missing Doc's to the site

Brad has sent email list for new SJTAG Use Case reflector to Carl Walker for Carl to post on new Cisco based email reflector

Discussion on retaining the fault diagnostics in the system

  • Ian McIntosh :- Used specialized buses for this function.
    Then discussed adding of some form of memory recording ie flash etc.
  • Carl Nielsen :- Has done some instances of storing to non volatile, a drive in some market place
  • Brad;- looking for promoting JTAG at system level means also promoting that SJTAG is not just for test
  • Jim Webster:- Is this testing and returning whole systems or board returns?
  • Jim :- what is being sent back to repair depot?, if from multi-card module, when split and then sent back to depot and no-fault found.
  • Ian :- Comments on having cascaded fault memory to ensure that fault information is recorded accurately
  • Jim :- In many cases, fault log has been stored in system fault logs and not on the board. How does one obtain the fault log information in this case?
  • Brad :- Need to scope the fault log to the scope of the test. You need to have some form of board storage for failure diagnostics at the board test level.
  • Brad :- Some failures are due to external factors that are off boards. No Fault founds in system could be due to tolerance of matching 2 cards. ie swap 1 unit send back to depot and no fault will be found.
  • Ian :- Don't want to have too broad of a scope. Deal with board information now and let vendor deal with extended information later.
  • Jim :- We need to focus on what SJTAG needs to do and not solve general board test issues.
  • Al Holliday :- We need to compile questions for ATCA and pose all options that are viable and what can be achieved.
  • Jim :- We may have something pop out that can be used in ATCA and elsewhere though our discussions.
  • Jim :- Different vendors may do test differently: Some may scope it at the system level and some may scope it at a board level. Each board may also be tested with tests developed by different tools (name your favorite PC based BScan tool here)
  • Jim :- Some systems will test at PCB level with one vendor and at system level diagnostic by another. Info must be transferable between vendors
  • Brad :- This is the same issue with ICT tools at manufacturing.
  • Brad :- feels that SVF, STAPL are more vector lang rather than system level
  • Jim :- Indicates that test manager should write the failing device back into the PCB rather than storing the vectors
  • Jim :- We need an independent SJTAG Test Manager that is responsible for formatting the diagnostic and test data in a unified manner that other tools may access
  • Brad :- I feel we need to also address the types of diagnostic data required for each test type. I presented a paper at BTW2006 on this very subject proposing a diagnostic look up table strategy for interconnect tests. The tool vendors present claimed they already have this information in some form that their tools already use for interconnect test.
  • Jim :- SJTAG Test Manager takes information and tabulates it into data and vectors which is able to correspond to diagnostics
  • Jim :- Suggests that the fault info is written into a user code field in a device
  • Brad :- By just writing data into user code will it provide sufficient data to enable the diagnostics. What will happen when there is not fixturing or a different topology (system vs. board scope)?
  • Time was running out and Jim suggested he write up his idea to use as a discussion point for next meeting

Next meeting:

Scheduled for 11/19/2007 at 8:00am EST

Review new action items

  • Jim will write up his thoughts on SJTAG Test Manager and diagnostics for discussion point of next meeting
  • Register on new SJTAG web site (http://www.mcintoshuk.plus.com/sjtag/) (All)
  • All need to check and add any missing Doc's to the site (All)

Meeting adjourned at 9:05am EST

Many thanks to Peter and Adam for supplying their notes to assist in recording the meeting.

Respectfully submitted,