Minutes of Weekly Meeting, 2010-04-12
Meeting called to order at 10:35 AM EST
1. Roll Call
Carl Walker
Adam Ley
Brad Van Treuren
Eric Cormack
Michele Portolan
Peter Horwood
Brian Erickson
Heiko Ehrenberg (joined 10:43am)
Patrick Au (joined 10:52am)
Ian McIntosh (joined 11:02am)
Excused absence:
Tim Pender
2. Review and approve previous minutes:
03/22/2010 minutes:
- Motion to approve: Carl
- Second: Heiko
- No objections or abstentions
-> minutes approved
03/29/2010 minutes:
- There was a typo in the preamble
- Motion to approve: Heiko
- Second: Patrick
- No objections or abstentions
-> minutes approved
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
(http://files.sjtag.org/Ericsson-Nov2006/STAPL-Ideas.pdf)
- 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
- Ian/Brad: Consolidate PowerPoint diagrams used during recent meetings.
- COMPLETE
4. Discussion Topics
- White Paper Volume 3 Review - Discussion of system description diagrams
- - Consolidated slides
- - New slide from Ian:
- [Brad] Showing path selection on a board at an abstract level, for
choosing one chain or another;
- - Consolidated slides:
- [Brad] What I've done is I've used Ian's graphics for the devices with a
BScan register (slide 7)
- [Brad] On slide 8 I left the BScan cells showing in the BScan device
- [Brad] On slide 9 I then started removing the BScan cells, showing just
the IC's as boxes, indicating the scan path between the devices
- [Brad] Slide 10 shows two separate chains, as a precursor to introducing
scan path switching (dynamic paths)
- [Brad] On slide 11 I show a new box for Dynamic Path Selection, linking
the two separate scan chains (static paths) to one TAP on the backplane; I'm
showing two options:
- selecting one of the scan chains
- bypassing the whole chain
- [Brad] Slide 12 shows an external path scheme; I left the graphic as it
was and added a question "The scan chain breaks if the mezzanine unplugs" Is
there a way to resolve this? (Dynamic Path)
- [Brad] Slide 13 then shows the Dynamic Path Selection as an option to
bypass the external scan chain (mezzanine)
- [Brad] How do you guys feel about the flow of the slides now
- [Heiko] I'd say it is an improvement
- [Patrick] I agree
- [Brad] Are there things missing so far, or are things simplified too
much
- [Patrick] I don't think so
- [Brad] Starting with slide 15 we show the handling of multiple scan
chains
- {Ian joined}
- [Brad] Slide 15 shows individual chains being daisy-chained, the
simplest configuration at the board level;
- [Brad] Slide 16 introduces Dynamic Path Selection Logic to select
individual scan chains on a board and an external scan path on a mezzanine
card; this may be a new concept for some people doing Boundary Scan
- [Brad] Slide 17 shows a similar graphic, but now the TMS is shared and
the TCK is separate/independent for each chain
- [Brad] Slide 18 introduces radial / star topology, where each static
path is wired independently to the Dynamic Path Selection Logic
- [Brad] Next I was considering about introducing IEEE 1149.7 scan chains;
but then I thought this may not be the right place to introduce this
- [Ian] I agree that at this point in the discussion of the architectures
we should stick with the standard 1149.1 configurations and then introduce
1149.7 and other newer technologies at a later time
- [Adam] I'm fine with that
- [Brad] Does anybody object to removing slides 19 and 20?
- {no objections voiced}
- [Brad] Then we'll remove slide 19 and slide 20 from this slide deck;
and this is what I had so far or scan chain architectures
- [Ian] It seems to me that we'll have to redraw most of the diagrams
shown in the original white paper; the original diagrams show a lot of
detail that may be too distracting (e.g. FPGA, 1149.6, etc.) and may not be
necessary at this level;
- [Carl] I agree
- [Brad] And our goal here is to show that the system-level path
configurations are kind of just an extension of what we already know about
board-level path configurations
- [Brad] What I wanted to achieve with this section is to be able to show
just a single TAP at the board level, where everything below it can
collapse, but I don't think I quite got there yet, there is still too much
detail
- [Ian] We don't want to use up resources/pins at the TAP Interface so we
need to use path selection as part of your solution for that.
- [Ian] Going from board level to system level is the same problem as
managing multiple scan chains at the board level;
- [Brad] I still need to work on these graphics a bit
- [Ian] The elements from slide 11 could used for a simplified version too
- [Ian] We almost have a color scheme here now, so we could omit the text
from inside the blocks
- [Peter] A color key might prove helpful, especially if we change the way
we represent the scan chain selection (details on one slide, general blocks
on another slide)
- [Brad] That's all the slides I have to show so far
- [Ian] I can take an action to come up with some more reducible graphics
to represent boards
5. Schedule next meeting
Monday April 26, 2010, 10:30 AM EDT
May 3, 2010; 10:30am EDT
May 10, 2010; 10:30am EDT
May 17, 2010; 10:30am EDT; Heiko may not make this call
May 24, 2010; 10:30am EDT; Heiko won't make it to this call
6. Any other business
No other business.
7. Review new action items
- Ian to come up with some board representation graphics
8. Adjourn
Brad moved to adjourn at 11:41 AM EST, seconded by Peter.
Heiko