IBIS Macromodel Task Group

Meeting date: 12 December 2023

Members (asterisk for those attending):
Achronix Semiconductor:       Hansel Dsilva
Amazon:                       John Yan
ANSYS:                      * Curtis Clark
                              Wei-hsing Huang
Aurora System:              * Dian Yang
Cadence Design Systems:     * Ambrish Varma
                              Jared James
Dassault Systemes:            Longfei Bai                             
Google:                       Hanfeng Wang
                              GaWon Kim
Intel:                      * Michael Mirmak
                            * Kinger Cai
                            * Chi-te Chen
                              Liwei Zhao
                              Alaeddin Aydiner
Keysight Technologies:        Fangyi Rao
                              Majid Ahadi Dolatsara
                              Stephen Slater
                              Ming Yan
                              Rui Yang
Marvell:                      Steve Parker
Mathworks (SiSoft):           Walter Katz
                              Graham Kus
Micron Technology:            Justin Butterfield
Missouri S&T:                 Chulsoon Hwang
                              Yifan Ding
                              Zhiping Yang
Rivos:                        Yansheng Wang
SAE ITC:                      Michael McNair
Siemens EDA (Mentor):       * Arpad Muranyi
                            * Randy Wolff
Teraspeed Labs:             * Bob Ross
Zuken USA:                    Lance Wang

The meeting was led by Arpad Muranyi.  Curtis Clark took the minutes.

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

Arpad reviewed the schedule of upcoming meetings.  Regularly scheduled meetings
will be held, except for the following two meetings:
  December 26, 2023 (cancelled)
  January  02, 2024 (cancelled)
  
-------------
Review of ARs:
         
Michael: Develop a full syntactically complete example demonstrating the
         AMI Test Data proposal.
         - In progress.  Michael said he was ready to review a draft.

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

- None.

-------------------------
Review of Meeting Minutes:

Arpad asked for any comments or corrections to the minutes of the December 5th
meeting.  Curtis moved to approve the minutes.  Michael seconded the motion.
There were no objections.

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

AMI Test Data proposal:
Michael reviewed draft16, which he had sent to the ATM list prior to the
meeting.  He noted the changes it contained:
- The Executable, Executable_Tx, and Executable_Rx subparameters of the
  [Algorithmic Model] keyword appear in the tree diagram.
- The Executable_index subparameter of the [AMI Test Configuration] keyword is
  defined as having a value that is positive (as opposed to non-zero).

Michael said that he may decide to rename some of the examples in the draft
to match the names he's using in the full syntactically complete example he is
developing.

Michael reviewed his progress in creating the syntactically complete example.
He said he had started with a quick investigation of past AMI examples that were
published on the IBIS website.  He had found an early example provided by SiSoft
in 2008 demonstrating .ami file syntax and source code for creating the .dll or
.so.  The example also contained a simple simulator executable to load and run
the .dll.  The simple simulator took a .csv file defining the impulse response
as an additional input.  Michael realized that this provided a good starting
point for his example of [AMI Test Configuration].

Michael reviewed his example, in which he had created examples of the files
referred to by the [AMI Test Configuration] subparameters.  He said the .ibs
file from the original 2008 example was largely unchanged, but he had added the
[AMI Test Configuration] keyword to the [Algorithmic Model].  He had also
modified the original .csv representing the impulse response so that it fit
the format prescribed for the Input_IR_file.  The original example .csv
contained two columns, time and voltage, but the Input_IR_file uses a single
column and provides only the voltage values.

Michael noted that the original example had only been provided in 32-bit form,
and he asked whether we would want to rebuild any new example .dll in 64-bit
form.  He also noted one possible bug in the original example.  The example
model returns the value of an Input parameter, tx_swing, in the
AMI_parameters_out string.

Curtis and Ambrish commented on the format of the data in the file referred to
by the AMI_input_parameters_file subparameter.  They asked whether the values
related to the AMI function calls' formal parameters (e.g., Sample_interval,
Symbol_time, Number_of_rows, and Agressors for AMI_Init()) should be separated
from the actual AMI parameter values.  Curtis said that doing so would reduce
the AMI parameters section to the exact parameters string the simulator should
pass to the model in AMI_parameters_in.  Arpad agreed and suggested a separate
"Simulator_parameters" section for the function call parameters.  Michael said
he would create a draft17 and introduce a "Simulator_parameters" section and
also a new "Model_parameters" section containing the AMI parameters.  He said he
he would update the full syntax example accordingly.

Michael asked whether the example files he had created would be sufficient.
Arpad asked what we needed to do about the fact that the simple simulator code
from the 2008 example expected a .csv file containing the impulse response, and
the format was different from that of the file referred to by the Input_IR_file
subparameter.  Michael said it had been easy to manually convert the
Input_IR_file to the .csv file format the simple simulator code required, just
for testing purposes.  Michael said he didn't necessarily think we needed to
update the old simulator code to create a test bed for this proposal.  Arpad
agreed and said eventually the ibischk parser might provide this functionality.
Michael said it would be up to the Quality task group to decide what the parser
could do to check this functionality, but this proposal was viewed as a way that
the parser might have enough information to execute the AMI functions and
confirm that they at least returned the proper rootname in their
AMI_parameters_out (see parser enhancement BUG227 -
 https://urldefense.proofpoint.com/v2/url?u=https-3A__ibis.org_bugs_ibischk_bug227.txt&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=G5dJ77dHT08t07kVHzshtlXlU2oadFVA6-TtfYgLMYgplps1b9GRYJ27A-Iq_P8P&s=sBObymgl-f5hUzwPP2M92bI7oh2qwFUP_7adJ9M9DF0&e= ).

- Ambrish: Motion to adjourn.
- Curtis: Second.
- Arpad: Thank you all for joining.

New ARs:

Michael: Send out draft17 of the AMI Test Data proposal including more feedback
         from the ATM meeting and update the full syntax example accordingly.

-------------
Next meeting: 19 December 2023 12:00pm PT
-------------

IBIS Interconnect SPICE Wish List:

1) Simulator directives
