Minutes, IBIS Quality Task Group

24 March 2020

11:00-12:00 EST (08:00-09:00 PST)

ROLL CALL

Intel Corp.                         Michael Mirmak
Mentor, a Siemens Business          Arpad Muranyi
Micron Technology                 * Randy Wolff
Signal Integrity Software:        * Mike LaBonte
SPISim                              Wei-hsing Huang
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 March 17 meeting were reviewed.  Randy Wolff moved to accept.
Lance Wang seconded. Without objection the minutes were approved.


ARs:

- Bob Ross to review the updated TSCHK2 license document and send to SAE ITC for review
  - Done. Bob said it had already been updated in the past.
- Mike LaBonte to continue editing IBISCHK "orphan" messages
  - Done


IBISCHK7 source code license:
Bob Ross reported no change from the last meeting.


TSCHK2 license agreement:
Mike showed the existing TSCHK2 license agreement, which was an older SAE International
document.  Bob Ross felt a more up to date version should be found somewhere, he would
look for it.  No response had been yet received from SAE ITC.

AR: Bob Ross to find more up to date TSCHK2 source code license document


Better IBISCHK error messages for BUG210:
Mike LaBonte showed before and after versions of the "orphan" IBISCHK error messages.
We reviewed all six sections.

Randy Wolff said "-scoped" was missing from the message for SUBMDL_ERR_2 in section 1.

It was agreed that section 2 was good with changes as shown.

Section 3 changed messages such as "Orphan Data line" to "Data line not allowed here",
but it was felt that that would not be helpful enough.  Mike said it had been difficult
to assess the true meaning of the error by code inspection, but he could investigate
further.  Randy said the section 4 messages have have cited [Pin Numbers] and [Begin
Board Description] incorrectly.

Mike said the section 4 messages were more specific because the keywords were
recognized, but found to be out of place even within a top level keyword.  Bob Ross
said we had to follow specific ordering rules for [Define Package Model] and [Begin
Electrical Board Model].  Randy said EBD R/L/C might be required to be found under [Path
Description]. Also 'NC' might be required to be found under [Pin List].  He suggested
changing "found not under" to "not found under" in section 4.

It was agreed that section 5 was good with changes as shown.

Bob said adding line numbers in section 6 might be a problem. Randy suggested inserting
"found" before "with" in each message. He asked about the list of [End *] keywords
for which no message had been found. Mike searched the source code and found that
[End Algorithmic Model] uses MDL_ERR_74, "Unknown Line Among Model Data". Bob felt
that was OK as long as the line number was given with it. Mike noted that only the
last 2 messages in section 6 had "orphan". The first 6 were other [End *] keyword
messages that should be consistent, yet not all such messages had been found because
they did not all contain "[End ". We decided to fix only the last two and leave all
other [End *] issues for a separate bug.

AR: Mike LaBonte to update IBISCHK "orphan" message sections 3, 4, and 6.


New IBISCHK bugs:
We reviewed a bug report sent by Michael Mirmak on March 3.  Bob Ross felt the testcase
was unnecessarily large.  The bug involved a model with [Voltage Range] 1.035V 1.0V
1.07V that was giving an excessive current warning for -1.5V.  Bob said the point had
been extrapolated, and that was valid.  Lance Wang said there would always be points
outside the [Voltage Range] for clamps, in real life, but the model was required to
have points only down to -VCC.  Randy Wolff felt this was a real bug. IBISCHK should
not extrapolate down to voltages outside -VCC to +2VCC.

AR: Bob Ross to reduce over-extrapolation bug testcase size and post it


IBIS-ISS parser:
No discussion.


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

Meeting ended: 12:17 ET

Next meeting March 31, 2020