Minutes, IBIS Quality Task Group

02 October 2018

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

ROLL CALL

eASIC:                              David Banas
Ericsson:                           Anders Ekholm
Intel:                              Michael Mirmak
                                    Eugene Lim
IO Methodology:                   * Lance Wang
Mentor, a Siemens Business          Arpad Muranyi
Signal Integrity Software:        * Mike LaBonte
SPI Sim                           * Wei-hsing Huang
Teraspeed Labs:                   * Bob Ross
Visteon:                            Preetesh Rathod

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 Sep 18 and Sep 25 meetings were reviewed.
Lance Wang moved to approve both. Bob Ross seconded.

ARs:

- Mike LaBonte to check if DLLs are closed once not needed.
  Mike reported this as done. The DLLs are closed at the right time.

- Bob Ross to get permission to do vendor model review for SPI Sim model.
  Bob reported this as done. It was sent to Zuken and APT.

New IBISCHK bugs:

- BUG202
Wei-hsing said the test case has a mismatch between the AMI file and executable, but
that would only affect operation, not checking.  Bob asked if compiling IBISCHK with
newer libraries would help.  Wei-hsing said the model uses GLIBC 2.11, but the problem
goes away with GLIBC 2.14.  Mike noted that IBISCHK has been successfully checking
C++ models for some time, so C++ was not necessarily the issue.  Wei-hsing said it
does not crash in HSPICE.  Mike noted that HSPICE has IBISCHK embedded, but probably
not 6.1.4, which does the DLL checking.  Bob said Synopsys had not purchased recent
IBISCHK source code.  Wei-hsing said HSPICE is compiled with C++, and that may help it.
Bob asked if there is a way to compile models that will not crash.  Wei-hsing said
it does not crash when compiled with newer GLIBC 2.14.  Bob asked if IBISCHK could
be compiled differently to solve the problem.  Mike suggested filing BUG202 but not
requiring a fix in IBSICHK6.1.5.  He also suggested asking Curtis Clark to try the
testcase, if his product uses at least IBISCHK6.1.3 code.  Bob said if the problem is
not in IBISCHK we can't solve.  Mike noted that the problem was between the model and
the system it was running on.  He suggested that compiling IBISCHK with C++ might help.
Wei-hsing pointed out that it does not crash on the first call, suggesting a complex
initialization or latent memory issue.

AR: Mike LaBonte to write BUG202.
AR: Bob Ross to ask IBISCHK source code licensees to test BUG202 testcase.

IBISCHK615 parser development:
Bob Ross said Atul has signed the contract, and we are waiting for his report.

IBISCHK700 parser development:
Bob Ross noted that we still need to use the new contract template for this, but it
can wait.

Bob Ross moved to adjourn. Lance Wang seconded. Without objection the meeting ended.

Meeting ended: 11:37 ET

Next meeting October 09, 2018
