Minutes from the 13 September ibis-atm meeting are attached.
The following document, which was discussed during the meeting, has been
posted to the work archive.
*DATE* AUTHOR <http://ibis.org/macromodel_wip/archive-author.html>
ORGANIZATION <http://ibis.org/macromodel_wip/archive-org.html> TITLE
<http://ibis.org/macromodel_wip/archive-title.html> FORMATS
13-SEP-2016 Bob Ross Teraspeed Section 3 Reserved Word Guideline Update
BIRD draft 1 (zip
<http://ibis.org/macromodel_wip/archive/20160913/bobross/Section_3_Reserved_Word_Guideline_Update_BIRD_draft_1.zip>
)(docx
<http://ibis.org/macromodel_wip/archive/20160913/bobross/Section%203%20Reserved%20Word%20Guideline%20Update%20BIRD%20draft%201/section3_2_draft1.docx>
)
IBIS Macromodel Task Group
Meeting date: 13 September 2016
Members (asterisk for those attending):
ANSYS: * Dan Dvorscak
* Curtis Clark
Broadcom (Avago): Xingdong Dai
* Bob Miller
Cadence Design Systems: * Ambrish Varma
Brad Brim
Kumar Keshavan
Ken Willis
Cisco: Seungyong (Brian) Baek
eASIC: David Banas
Marc Kowalski
Ericsson: Anders Ekholm
GlobalFoundries: * Steve Parker
IBM * Luis Armenta
Intel: * Michael Mirmak
Keysight Technologies: Fangyi Rao
* Radek Biernacki
* Ming Yan
Maxim Integrated Products: Hassan Rafat
Mentor Graphics: John Angulo
* Arpad Muranyi
Micron Technology: * Randy Wolff
Justin Butterfield
QLogic Corp.: James Zhou
Andy Joy
SiSoft: * Walter Katz
Todd Westerhoff
* Mike LaBonte
Synopsys: Rita Horner
Kevin Li
Teraspeed Consulting Group: Scott McMorrow
Teraspeed Labs: * Bob Ross
TI: Alfred Chong
The meeting was led by Arpad Muranyi.
--------------------------------------------------------------------------------
Opens:
- Arpad noted that he had made some cleanup of the weekly agenda, including:
- Added the new BIRDs 181 through 184.
- Changed the BIRD 147 entry to 147.1
- Removed some outdated TBD items.
-------------
Review of ARs:
- None.
--------------------------
Call for patent disclosure:
- None.
-------------------------
Review of Meeting Minutes:
- Two sets of minutes needed review, as the August 30th minutes were not
prepared and posted in time for the September 6th meeting.
- August 30th minutes:
- Arpad: Does anyone have any comments or corrections? [none]
- Walter: Motion to approve the minutes.
- Ambrish: Second.
- Arpad: Anyone opposed? [none]
- September 6th minutes:
- Arpad: Does anyone have any comments or corrections? [none]
- Walter: Motion to approve the minutes.
- Ambrish: Second.
- Arpad: Anyone opposed? [none]
-------------
New Discussion:
Agenda item #7: New BIRDs from the Editorial Task Group
- Discussion: Bob Ross and Radek noted that they are going to meet on Thursday
to begin discussions aimed at resolving their differences on this item.
Agenda item #6: BIRD 180: Require Unique Pin Names in [Pin]
- Bob R.: [sharing BIRD 180]
- This is already a BIRD submitted to the Open Forum.
- This whole BIRD is based on adding the restriction that pin names cannot be
repeated in the [Pin] keyword.
- In last week's ATM a motion was made to bring this up for discussion now.
- We can vote on this BIRD in the Open Forum or defer until the interconnect
BIRD is voted on.
- This uniqueness rule has been enforced by the parser since day one anyway.
- Whether this BIRD affects BIRD 125 I'm not sure.
- BIRD 125 allows duplicate pin names under [Pin Numbers] for the purpose
of splitting out die pads.
- Arpad: That is true, but isn't [Pin Numbers] a copy of [Pin]?
- BIRD 125 is for using IBIS ISS subcircuits in package modeling.
- Duplicates may be needed for splits and joins.
- That's why I want to vote on them together, though it is true that Walter
thinks the interconnect BIRD is now more advanced and BIRD 125 is not likely
to be adopted anyway.
- I think it might be wiser to wait and vote on both at the same time.
- Curtis: BIRD 125 concerns [Pin Numbers].
- [Pin Numbers] doesn't have to be identical to [Pin], it can be a subset.
- Since [Pin] and [Pin Numbers] don't have to be identical, can't we treat
BIRD 180 and BIRD 125 independently?
- Arpad: [Reviewing BIRD 125] - Yes, it is [Pin Numbers].
- [Reviewing [Pin Numbers] keyword], Usage Rules: "... but it is not required
to include all of the pins listed under the [Pin] keyword."
- Okay, given what we've just reviewed, I have no objections to voting on
BIRD
180 independently.
- Walter: Motion to table discussion of BIRD 180 here and to recommend to the
Open Forum that BIRD 180 be approved.
- Bob R.: Second.
- Arpad: Anyone opposed? [none]
Draft BIRD for IBIS Section 3, Item 2:
- Bob R.: [sharing the Draft BIRD]
- The first two rules under Section 3 have existed since day one.
- Rule #1 is okay.
- Rule #2 Said POWER, GND, NC, NA, CIRCUIT call are reserved words and must
not be used for any other purposes.
- Rule #2 was too restrictive, and the parser never enforced it that way.
- In practice, GND can be a signal name in a data sheet for example.
- Rule #2 is now relaxed, so those words can be signal names, pin names,
bus labels, etc.
- This proposed language is a compromise between proposals Walter and I
created.
- In the analysis section I copied the latest from the two proposals.
- Walter: I think what you've done is fine.
- The intent was to resolve confusion in the Editorial Committee caused by GND
being used all over the place (model name, signal name, even as node 0).
- I think this proposal clarifies what it should.
- For clarity, please add "for pin name" at the end of the phrase:
"(except for CIRCUITCALL, which exceeds the maximum number of characters)"
- Radek: Since there is no change to item #1, I would remove it from the BIRD.
- Bob: You're right, but I left it there for reference because it defines the
case sensitivity rule.
- Radek: In the "Notes" section for Requirement #2 you mention [Pin List], but
you don't need to.
- Bob: No, the [Pin List] section is relevant because it defines POWER and GND
as reserved words. It's just there for information.
- Discussion: The group discussed the following relevant lines from the Usage
Rules section of [Pin List]:
Any pin associated with a signal name that begins with "GND" or "POWER"
will be interpreted as connecting to the boards ground or power plane. In
addition, NC is a legal signal name and indicates that the Pin is a "no
connect". As per the IBIS standard "GND," "POWER," and "NC" are case
insensitive.
Walter noted that he felt the first sentence wasn't a sensible rule, and that
"plane" (singular) didn't make sense [editorial note: "boards" should be
"board's"]. Walter also noted that the 3rd sentence, which refers back to the
"IBIS standard" and defines GND and POWER as case insensitive (reserved) is
really a mistake in the [Pin List] description (since GND and POWER are
signal names in this case). Walter suggested that these issues could be
corrected in a separate BIRD. Radek suggested that we should just remove the
offending line (3rd sentence) as part of this BIRD. Bob said he didn't want
to complicate this BIRD, and preferred to deal with the EBD ([Pin List] in
particular) issues in a separate BIRD. Radek said he was okay with that, but
in that case he suggested we remove the mention of [Pin List] from the Notes
section of Requirement #2 for this BIRD. Radek thought it was okay to leave
the mention of [Pin List] in the Analysis section as information. Bob agreed
to remove the [Pin List] sentence from the Notes for Requirement #2.
- Walter: Motion to submit this to the Open Forum.
- Bob: Submit it as modified per our discussion here.
- Radek: After this discussion, I second it.
- Arpad: Anyone opposed to Bob submitting it after modification? [none]
- Bob R.: Motion to adjourn.
- Curtis: Second.
- Arpad: Thank you all for joining.
-------------
Next meeting: 20 September 2016 12:00pm PT
-------------
IBIS Interconnect SPICE Wish List:
1) Simulator directives
