Minutes of Weekly Meeting, 2010-10-11

Meeting called to order: 10:36 AM EDT

1. Roll Call

Eric Cormack
Carl Walker
Ian McIntosh
Patrick Au
Brad Van Treuren
Peter Horwood (joined 10:48)

Tim Pender
Heiko Ehrenberg
Brian Erickson

2. Review and approve previous minutes:

10/04/2010 minutes:

  • Draft circulated on 10/04/2010.
  • No corrections noted.
  • Insufficient attendees to approve.

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 to consider what data items are missing from Data Elements diagram
  • 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/Brad: Draft "straw man" Volume 4 for review - Ongoing
  • All: Review "Role of Languages" in White Paper Volume 4 - Ongoing
  • All: Review 'straw man' virtual systems and notes on forums:
    http://forums.sjtag.org/viewtopic.php?f=29&t=109. - Ongoing
  • All: Add to, or comment on, the bullet point list of architecture drivers. - Ongoing.
  • All: Provide forum comment on the graphics used during the meeting; suggest "building blocks" that may be used in future:
    http://forums.sjtag.org/viewtopic.php?f=29&p=257#p257 - Ongoing.
  • ALL: review / comment in preparation for upcoming meetings. - Ongoing
  • Tim: Draft matrix of SJTAG features against evolving solution options. - Ongoing.
  • All: Post suggestions for key SJTAG gateway features on the forum (Ian will create topic) - Ongoing
  • All: Post suggested draft texts for survey comments in existing 2009 Survey thread (http://forums.sjtag.org/viewtopic.php?f=32&t=83) - Ongoing
  • Carl: Contact Zoe about what material we could obtain for our use. - Ongoing
  • Ian/Brad: Condense gateway comments and queries into a concise set of questions. - Ongoing

4. Discussion Topics

  1. Texts for Survey Result web pages
    • [Ian] Since last week, I've linked in all the images so this is a bit easier to work on without all the page flipping. We got as far as 3.3 last time.
    • {Forum post "Follow-up User Survey (2009) Results" shared, http://forums.sjtag.org/viewtopic.php?f=33&t=124}
    • 3.4
    • [Brad] What doesn't come out is the diversity within a single design; that within a single design there may be a requirement for multiple control interfaces.
    • [Ian] OK, so that's something we should add.
    • [Brad] In that last paragraph, we should be emphasising that embedded control is included within the group of people using a mixture of methods, and that this adds to the complexity of defining the structure.
    • [Ian] I guess this is one of the questions that we could have been smarter over what options we offered.
    • [Brad] It shows that the need for SJTAG is going to come sooner than later.
    • 3.5
    • {No comments}
    • 3.6
    • {No comments}
    • 3.7
    • [Brad] I like the cross-reference to the IJTAG concerns.
    • [Ian] It has to be mentioned somewhere!
    • [Brad] As designs shrink, VHDL or Verilog will start to surface, as people look at the verifications that have to be accomplished.
    • 3.8
    • [Ian] This is where we were concerned about the claims of using CAD to describe the system.
    • [Brad] I was thinking that this maybe more of an education issue; people perceive that they can use CAD, but haven't really crossed the boundary yet.
    • [Ian] Yes, that was what I had been thinking. They can see that have the backplane in CAD so think that they have all the bits of the puzzle, and have missed the issue of where the boards plug into the backplane.
    • [Brad] I think we should add some commentary that we need to do more education in this area. The White Paper is insufficient in revealing the complexity of the problem.
    • 3.9
    • {No comments}
    • 4.1
    • [Ian] I think what was important is that no-one is getting turned off from SJTAG.
    • [Brad] That's fine for a description of the results.
    • 4.2
    • [Ian] This has a forward reference to Section 8.
    • [Brad] That's a smart thing to do.
    • 4.3
    • [Ian] I think the absence of response is important - no-one thinks there's no benefit from SJTAG.
    • 5.1
    • [Ian] We don't have a breakdown of the Use Cases being considered.
    • [Eric] Perhaps this is a discussion point for ITC.
    • [Ian] That's a good idea.
    • [Brad] We can put the diagram on the Poster for discussion.
    • 5.2
    • [Ian] Another forward reference, to Section 7 on gateways.
    • 5.3
    • [Brad] The concurrency results were much higher than I expected.
    • [Ian] Yes, I was surprised as I think there's still a tendency to think of testing as a linear sequence of events.
    • [Brad] Mainly because that's what the tools have driven you to do.
    • 5.4
    • {No comments}
    • 5.5
    • [Brad] Another area of education for the new White Paper.
    • 5.6
    • [Ian] People aren't so interested in vector level control.
    • [Brad] Or the perception is that they don't want to deal with anything other than high level control.
    • 5.7
    • {No comments}
    • 5.8
    • [Brad] As an aside, Ken Posse was indicating that there may be a public draft of P1687 just before ITC.
    • 5.9
    • [Brad] Obviously those in the 16% haven't tried to tune SERDES over a backplane!
    • [Ian] It may be that we could find a correlation between those people and those who declared little or no experience of using JTAG in systems.
    • [Brad] That was my thought too.
    • 5.10
    • [Ian] Two camps: This may be another ITC discussion point.
    • [Brad] Yes. When you get into root cause analysis you begin to see the value. I think Cisco Systems found quite a big bang by kicking off the resolution process maybe two weeks early, before the units get back to the factory.
    • [Carl] That is very true.
    • [Ian] We add a note here?
    • [Brad] Add a note on the differences and how we can bring the two camps together.
    • 6.1
    • {No comments}
    • 6.2
    • [Ian] We had a diagram attached to the question but that doesn't seem to have helped. Maybe it needed more explanation.
    • [Brad] I expect the software section of the White Paper will help to clarify this.
    • 6.3
    • [Brad] I like what you have here.
    • 6.4
    • {No comments}
    • 6.5
    • [Brad] I don't believe you can say any more. While people using HSDL may recognise the OO aspects, VHDL designers may not realise that they're using OO.
    • [Carl] Typically they do not.
    • 6.6
    • [Ian] Am I taking the wrong conclusions from this? I'm not convinced.
    • [Brad] Well, once you start to think of things on customer sites, maybe behind firewalls, you can't rely on connectivity to the outside world. My problem is the diversity of systems I work on, as I probably have all of these situations. It's not as clear cut as the question might imply.
    • [Brad] Some systems might be able to afford a dedicated Test Management processor, other may have something in a FPGA. We need to support the full spectrum.
    • 6.7
    • [Ian] This is the one where we had a big gap in the middle of the memory ranges, between 8 MBytes and 4 GBytes.
    • [Patrick] But why? 8 MBytes is not that big nowadays.
    • [Ian] It might be different types of system: Those with CPUs with mass storage attached at one end and those with FPGAs and limited address space and on-board memory at the other.
    • [Brad] Maybe there's a difference in interpretation too: Is is per board or per system?
    • [Brad] A strategy where the FRU maintains it's own data might fit into the smaller category while something using a central repository might go to the 4 GByte end.
    • [Ian] This may be another one for ITC.
    • [Brad] We need to understand the Use Space.
  2. ITC - Fringe Meeting, Poster
    • [Ian] I put together a draft message to send out as an invite for the Fringe Meeting.
    • {2010_Oct_Newsflash.htm shared}
    • [Ian] Are there any comments on this?
    • [Brad] I liked it.
    • [Patrick] Looks pretty good.
    • [Peter] You got my email at the weekend?
    • [Ian] Yes, I did thanks, Peter.
    • [Ian] Any changes?
    • [Carl] Looks good.
    • Eric moved to approve for issue, seconded by Brad, no objections or abstentions
    • [Ian] On the poster, I contacted Bill Eklow to ask about the logistics, as the early suggestion was having two groups to a table, so I wanted to know if that was still the case and whether it meant we got a full poster board, half a board, etc.
    • [Ian] Bill apologised for the lack of information but he should be sending out information in the next few days.
    • [Brad] What happened in the past was that you got two tables together with a cork board between them.
    • [Ian] Yeah, I think because it's a joint poster they're maybe going to try to economise on space by sharing tables.
    • [Brad] I think if you plan for something around 1m wide by 1.5m high you should be OK.
  3. White Paper Volume 3 - Key Gateway Features:
    • {Not discussed due to lack of time}

5. Schedule next meeting

October 18th

Schedule for October 2010:
  • [Ian] I know we don't usually think of next month's schedule this early, but I was considering that it might be a good idea to skip Nov 1st, as there will be a few people either setting up at ITC or travelling to ITC that day.
  • [Brad] That might be good idea.
  • [Ian] We'll have the Fringe Meeting on the Thursday anyway. And possibly the following Monday will be too soon after the Fringe Meeting.
  • [Ian] We don't need to decide now, I just wanted to get the thought out for consideration.

6. Any other business

The motion on signal naming conventions from the meeting of May 3rd remains tabled.

7. Review new action items

  • Ian: Send out Newsflash announcement of ITC Fringe Meeting and Poster.

8. Adjourn

Eric moved to adjourn at 11:41 AM EDT, seconded by Patrick.>/p

Respectfully submitted,
Ian McIntosh