Minutes, IBIS Quality Committee

22 January 2013

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

ROLL CALL

Altera				    David Banas
Cadence Design Systems:             Joy Li
                                    Yingxin Sun
Cisco Systems:                      Tony Penaloza
Ericsson:                         * Anders Ekholm
Green Streak Programs:              Lynne Green
Huawei Technologies:                Guan Tao
IBM:                                Bruce Archambeault
                                    Greg Edlund
Intel                               Michael Mirmak
IOMethodology:                      Lance Wang
Mentor Graphics:                    John Angulo
Micron Technology:                  Moshiul Haque,
                                    Randy Wolff
Nokia Siemens Networks:             Eckhard Lenski
QLogic Corp.:                       James Zhou
Signal Consulting Group:            Tim Coyle
Signal Integrity Software         * Mike LaBonte
Teraspeed Consulting Group:       * Bob Ross
Texas Instruments:                  Pavani Jella

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

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

Call for opens and IBIS related patent disclosures:

- No quorum last meeting, no minutes

AR Review:

- Mike and Bob create IQ report for DesignCon summit
  - Done

- Mike create outline for new IBISCHK specification
  - In progress

New items:

Mike showed the policies for IBISCHK ERROR, WARNING, CAUTION, NOTE categories:
- ERROR:
  - Bob: For ERROR there can be exceptional items
    - For example I-V v.s V-T endpoint table mismatch over 10% is an ERROR even though the spec
      doesn't say that
  - Mike: At the time we only had WARNING and ERROR
    - But we wanted two thresholds, 2% and 10%

WARNING:
  - Bob: WARNING uses thresholds that are simply common
    - Waveform check violations can be due to a too short waveform

- CAUTION:
  - Bob: CAUTION was for pathological cases like Vmeas outside Vinl/Vinh
  - Mike: These only showed when requested
    - Did we feel there would be too many if these were WARNING?
  - Bob: The dV check is made a CAUTION because it is archaic

- NOTE:
  - Bob: This is used only for non-monotonic data
  - Mike: This came before CAUTION?
  - Bob: Not sure
    - There can be large numbers of these
  - Mike: There is no switch to turn these off
  - Bob: It really is a warning

- INTERNAL ERROR:
  - Bob: These are reported as ERROR
  - Mike: These may be from the lexer, before it reaches the parser
    - Surprising that there are 140 of these, it usually is around a dozen
    - Usually these are cryptic
  - Bob: This is due to ambiguity because the parser is just lost

UNUSED:
  - Bob: We probably branch around these
    - They either don't exist in the code or are never reached
    - The error numbers may still be in the code

- Bob: Another class is system errors like memory issues and crashes
  - We can only report these as bugs
- Mike: We could catch these but it probably is not a good idea

- Numbering:
  - Anders: This is useful 
  - Bob: It is 1 letter and 4 digits
  - Mike: Could we potentially change C200 to W200?
  - Anders: I thought numbers could be reused
  - Bob: The same code is used for Typ, Min, Max
    - In some cases the line number is given
    - Sometimes the model name is given
  - Mike: There is no line number for missing things
  - Bob: There are 1200 messages
  - Mike: That would be an appendix to our document

- Bob: We made some errors
  - The CAUTION one should have been an ERROR
  - We never asked for a -1VDD to +2VDD check
    - In bench testing it would blow transistors
    - Even some SPICE models have trouble
  - Mike: Some clamp models have trouble

- Mike: Our document should have a section based on this material
  - We should be frank that some things are just exceptions to the rules
- Bob: We check for omissions and also too many parameters
- Mike: The parser says nothing when [Model Spec] overrides [Model] params
  - Even when they disagree
- Bob: It will be an ERROR if [Package] has more 3 parameters

Mike showed a spreadsheet of IBISCHK versions:
- Mike: We should have a README stating that platform compatibility is not guaranteed
- Bob: I mostly run Windows
- Anders: Same here
- Mike: Vendors who have the source code mostly compile their own versions
  - This almost guarantees it will run on all tool supported platforms

Review of group meeting schedule:

- We agreed the next meeting would be Feb 5

Next meetings:
- The next two meetings will be Feb 5 and Feb 19

- The meeting ended at 12:16 ET
