IBIS Macromodel Task Group

Meeting date: 19 August 2014

Members (asterisk for those attending):
Altera:                     * David Banas
ANSYS:                      * Dan Dvorscak
                              Curtis Clark
Avago (LSI)                   Xingdong Dai
Cadence Design Systems:     * Ambrish Varma
                              Brad Brim
                              Kumar Keshavan
                              Ken Willis
Ericsson:                     Anders Ekholm
Intel:                        Michael Mirmak
Keysight Technologies:        Fangyi Rao
                            * Radek Biernacki
Maxim Integrated Products:    Hassan Rafat
Mentor Graphics:            * John Angulo
                            * Arpad Muranyi
Micron Technology:          * Randy Wolff
                              Justin Butterfield
QLogic Corp.                  James Zhou
                              Andy Joy
eASIC                       * Marc Kowalski
SiSoft:                     * Walter Katz
                            * Todd Westerhoff
                            * Mike LaBonte
Synopsys                      Rita Horner
Teraspeed Consulting Group:   Scott McMorrow
                            * Bob Ross

(Note: Agilent has changed to Keysight)

The meeting was led by Arpad Muranyi.

------------------------------------------------------------------------
Opens:

- Randy: Would like to discuss BIRD 173 updates.

--------------------------
Call for patent disclosure:

- None

-------------
Review of ARs:

- Ambrish investigate other ways to communicate tap coefficients
  - In progress

- Arpad to review IBIS spec for min max issues.
  - In progress

- Brad check within Cadence about overclocking approaches
  - Radek: This came up at DesignCon, we probably could remove this item.

- Arpad - Work with Randy to update BIRD 173.
  - Done

-------------
New Discussion:

Introduction of Marc Kowalski:
- ASIC signal integrity at eASIC.
- Mainly SerDes.


BIRD 173:
- Randy showed the updated BIRD.
- Randy: New language (from spec) is in blue.
- Radek: It needs to say "positive semi-definite", which allows zero.
- Randy: Eigenvalues need to be real and non-negative.
  - We need a better source for M-matrix definition.
  - We are adding a diagonal dominance condition.
  - Vladimir has said it is not enough to satisfy passivity.
  - An additional sentence is needed.
  - This should satisfy all passive rules.
- Bob: "larger" should be "greater than the magnitude"
- Ambrish: Will this be checked in the Touchstone parser?
- Arpad: The goal here is to fix matrix definitions.
  - We don't support s-parameter directly in ISS.
  - There could be a cookbook.


BIRD 128/147:
- Walter: In the BIRD 147 time domain flow the tap manipulation allowed is increment/decrement.
  - The RX sends this and the TX decides if it can do it.
  - If approved the BIRD should support IBIS approved BCI for known standards, true?
- Ambrish: Correct
- Walter: It does not support PCIeG3 backchannel training.
- Ambrish: It is a general feedback mechanism, but it needs work for PCIeG3.
- Walter: PCIeG3 only sets coefficients, it does not increment/decrement.
- Ambrish: We want to propose something to address that, but got sidetracked.
- Walter motioned to table discussion of BIRD 147 until PCIeG3 support is added.
- Arpad: We already did that.
- Ambrish: We need to discuss this.
- Arpad: I thought Ambrish was working on this.
- Ambrish: I am working on it.
- Todd: Should we discuss this?
- Walter withdrew his motion.

- Arpad: Is anything new on BIRD 128?
- Ambrish: 128.1 still stands.


Min/Max data:
- Arpad: No update today.


Redriver BIRD:
- Walter: We need Fangyi to be part of the discussion.
  - With Init-only models the proposal changes.
  - Mixtures of models get complicated.
- Radek: How much does this matter?
- Walter: We have seen it, not sure how much it matters.


BUG 150:
- Arpad: We do not have Michael M today.


Quote characters in IBIS files:
- Arpad motioned to untable this topic.
- Walter seconded.
- Walter: Case sensitivity in IBIS is a problem.
  - We should allow uppercase file names.
  - IBIS rarely used double quotes describing text fields.
  - For AMI we enclosed strings in quotes.
  - It wasn't written very clearly though.
  - This was originally defined for DLL strings.
  - Then we required it in AMI files.
  - In the IBIS file strings are only sometimes double quoted.
  - This should be required only when there are spaces.
  - It should be allowed in AMI files only under certain conditions.
- Arpad: Double quotes have caused problems with the parser.
  - Some text editors change the second quote to non-ASCII back-quote characters.
  - We should spell that out.
- John: IBIS is limited to ASCII but AMI is not.
- Mike: The spec calls for characters 0x20 to 0x7F.
- Arpad: Do we need a BIRD on this?
- Bob: We should allow them in AMI files, for String.
- Arpad: Allow all extended ASCII, or just certain characters.
- Arpad showed the IBIS 6.0 specification.
- Arpad: 0x7E is not allowed, I wonder why.
- Todd: 0x7E is tilde and 0x7F is delete.
- Walter: We should have the same requirements for all of our file formats.
  - But then some simulators will have trouble.
- Arpad showed page IBIS 6.0 specification 187.
- Walter: In an AMI file we only have flexibility in String contents.
  - This is a strong limitation.
- Mike: IBISCHK could be more informative when back-quotes are detected.
- Radek: We should clarify the language here.
- Walter: It should be clarified higher up in the specification.
- Arpad: We should keep it separate.

-------------
Next meeting: 26 August 2014 12:00pm PT

-------------
IBIS Interconnect SPICE Wish List:

1) Simulator directives
