Minutes, IBIS Quality Task Group

May 23, 2023

12:00-13:00 Eastern Time (09:00-10:00 Pacific Time)

ROLL CALL

Intel Corporation                   Michael Mirmak 
Micron Technology                   [Randy Wolff]
Siemens EDA                         Randy Wolff *, Weston Beal *,
                                      Arpad Muranyi 
Teraspeed Labs                      Bob Ross *
Zuken USA                           Lance Wang *
Zuken                               Michael Schader 
GE Healthcare Technologies          Balaji Sankarshanan 


Everyone in attendance marked by *

NOTE: "AR" = Action Required.

-------------------------------  MINUTES  -------------------------------------
Bob Ross conducted the meeting.
Weston Beal recorded the minutes.


Call for IBIS related patent disclosures:
 - None


Review of previous meeting minutes:
Minutes from the 16 May 2023 meetings were reviewed. Randy Wolff moved to 
approve the minutes. Weston seconded. Without objection the minutes were 
approved.


Call for opens:
 - discuss IBIS version of IBS file referenced from an EMD.


AR review:
 - Weston will check with Atul on the idea of 2 error or warning messages for 
   loading AMI executables.
   done
 - IBISCHK720 – Test Randy’s 7.2 Files, Randy Wolff and Bob Ross work on two 
   BUG reports related to a Warning for a legal syntax and a Caution message 
   for an [EMD Pin List]
   ongoing

 
New ARs:
 - Bob will ask Atul about the warnings reported by the compilers and possibly 
   create a different bug report.


Discussion:
a. IBISCHK720 – Test Randy’s 7.2 Files, Randy Wolff and Bob Ross work on two 
BUG reports related to a Warning for a legal syntax and a Error message for an 
[EMD Pin List] (continuation)
Bob presented the draft bug reports for these issues. BUG244 outlines the 
issue of NC used as a signal_name. The parser issues a warning when the usage 
is legal. The resolution in the report is to not produce warnings in 2 
instances, line 20 and 35 of the test file. [AR] Randy and Bob will edit the 
bug report to include comments from our discussion last week. BUG245 describes 
the case where a pin of signal_type NC is used in a terminal line of the EMD 
model. This should generate an error message, but the parser does not generate 
the error.

b.	Pending BUG regarding Platform in AMI – Weston Beal will investigate 
further (We are waiting for a response from Atul.)
Bob mentioned a BUG242 for this issue. This issue is still not clear. Weston 
will continue discussion with Atul to get a clear description of what the 
parser does so that we can create a good bug report to fix this problem.

c.	Graham Kus Mathworks Warning message during architecture compile with:
“<file name> warning: variable ‘<variable name>’ set but not used 
[-Wunused-but-set-variable]  Michael Schaeder may provide some feedback 
concerning gcc compilation.  Curtis Clark’s suggestion.
Bob showed a message from Curtis Clark. Curtis commented in email that there 
are multiple instances of this warning and these should be easy to fix. Curtis 
provided 2 reports for Unix and Windows showing different types of warning 
from the compilers. The Unix warnings are for variables set but not used. The 
Windows warnings are for various types of warnings that should be examined 
carefully. [AR] Bob will ask Atul about the warnings reported by the compilers 
and possibly create a different bug report.

--  Discussion of BUG228 referenced by Arpad Muranyi to resolve the confusion 
about the require IBIS version of an IBS file referenced from an EMD file. Bob 
showed the bug report that created this confusion. The problem is that the 
parser checks something that is not explicitely required in the specification. 
The requirement really only applies to certain cases involving signal names 
and bus labels. The discussion from last week concluded that the checking for 
7.0 compatibility should only be reported when the bus_label warnings are 
found. The group will need to create a new bug report or maybe a BIRD to 
resolve this issue. We will discuss it more next week. 

d.	Quality Specification and checklist power aware discussion and [Merged 
Pin] interactions 
Weston reported that there is no need for a requirement in the quality 
specification regarding [Merged Pins]. The parser checks are sufficient. 
Weston will clean up the draft of the document and send to the task group for 
review.


Tabled topics (no discussion without motion):
a.	Secure Development Lifecycle (SDL) investigation (Security fixes 
    continuation), (TSCHK2, BUG6)
b.	TSCHK2 V2.02 BUG5 and BUG6 (implementation on hold)


Weston moved to adjourn. Randy seconded. Without objection the meeting ended.

Meeting ended: 1:00 pm ET

Next meeting 30 MAY 2023