Minutes, IBIS Quality Task Group

22 June 2021

12:00-13:00 EST (09:00-10:00 PST)

ROLL CALL

Intel Technology                    Michael Mirmak
Micron Technology                 * Randy Wolff
Signal Integrity Software:        * Mike LaBonte
Teraspeed Labs:                   * Bob Ross
Zuken USA:                        * Lance Wang

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

-----------------------MINUTES ---------------------------
Mike LaBonte conducted the meeting.

Call for IBIS related patent disclosures:

- None


Call for opens:

- None


Review of previous meeting minutes:
Minutes from the June 8, 2021 meetings were reviewed.  Randy Wolff moved to accept
the minutes. Lance Wang seconded.  Without objection the minutes were approved.


ARs:
- Mike LaBonte to fix tschk2 source code price on web pages
  - Done


NEW ITEMS:

New parser bug reports:
Bob Ross reported no new bugs.


IBISCHK development:
Mike LaBonte showed an email from Bob Ross.  Bob said the developer had asked if
the existence of the AMI_Impulse DLL function in the AMI DLL should be checked when
BCI_Training_Mode had "Impulse".  Bob had made a table mapping BCI_Training_Mode
combinations to the AMI_Init AMI_GetWave, AMI_Close and AMI_Impulse functions.  He noted
that AMI_Impulse was new.  Randy Wolff said it seemed like a natural extension of
BUG179, which introduced DLL function checking.  Mike asked if the idea was only to
know when to check for AMI_Impulse.  Bob said that was correct.  Mike said AMI_Init
should have an X in the table because it was always required in the DLL.

Mike asked if there was a new "Impulse_Exists" AMI parameter specified in BIRD201.1.
Bob said he believed there was, but it might have had had a different name.  Mike said
four tests would then be required:

1) If GetWave_Exists=True then check for AMI_GetWave in the DLL
2) If Impulse_Exists=True then check for AMI_Impulse in the DLL
3) If BCI_Training_Mode has "GetWave" then GetWave_Exists must be True
4) If BCI_Training_Mode has "Impulse" then GetWave_Exists must be True

Bob showed the IBIS 7.1 draft, at the Init_Returns_Impulse section.  Mike said
if BCI_Training_Mode called for Impulse then we would need to assure that
Init_Returns_Impulse=True.

Mike checked BIRD201.1 and found that there was no new Impulse_Exists AMI parameter specified.
He suggested the checks to perform were then:

1) If GetWave_Exists=True then the DLL must have AMI_GetWave (we already check that)
2) If BCI_Training_Mode has "GetWave" then GetWave_Exists must be True
3) If BCI_Training_Mode has "Impulse" then the DLL must have AMI_Impulse, and Init_Returns_Impulse must True

Mike asked if it was true that AMI_Init would participate in the training process by
returning the first training impulse response.  Randy said that was a question that
needed clarification.  Mike suggested asking the ATM group.

Bob suggested that if BCI_Training_Mode had GetWave then GetWave_Exists would be
implicitly True.  Mike said the same logic might be used for Init_Returns_Impulse, where
it would be implicitly True if BCI_Training_Mode had Impulse.  He suggested deciding
which way to have ibischk check it, and write a new BIRD to cover the rule.  Bob asked
if it would be an error for Init_Returns_Impulse to be False if BCI_Training_Mode
had Impulse.  Randy and Mike agreed that would be an error.  Mike said the same would
go for GetWave_Exists, which could not be False if BCI_Training_Mode had GetWave.


TSCHK201 development:
Randy said the RHEL version had worked well for him.


Tabled topics (no discussion without motion):
  - BIRD181.2
  - IBISCHK security fixes


Randy Wolff moved to adjourn. Bob Ross seconded. Without objection the meeting ended.

Meeting ended: 13:00 ET

Next meeting June 29, 2021