Minutes, IBIS Quality Committee

27 May 2008

11-12 AM EST (8-9 AM PST)

ROLL CALL
  Adam Tambone
* Anders Ekholm, Ericsson
  Barry Katz, SiSoft
  Benny Lazer
  Benjamin P Silva
  Bob Cox, Micron
* Bob Ross, Teraspeed Consulting Group
  Brian Arsenault
* David Banas, Xilinx
  Eckhard Lenski, Nokia Siemens Networks
  Eric Brock
* Guan Tao, Huawei Technologies
  Gregory R Edlund
  Hazem Hegazy
  Huang Chunxing, Huawei Technologies
  John Figueroa
  John Angulo, Mentor Graphics
  Katja Koller, Nokia Siemens Networks
  Kevin Fisher
  Kim Helliwell, LSI Logic
* Lance Wang, IOMethodology
  Lynne Green
* Mike LaBonte, Cisco Systems
  Mike Mayer, SiSoft
  Moshiul Haque, Micron Technology
  Peter LaFlamme
  Randy Wolff, Micron Technology
  Radovan Vuletic, Qimonda
  Robert Haller, Enterasys
  Roy Leventhal, Leventhal Design & Communications
  Sherif Hammad, Mentor Graphics
  Todd Westerhoff, SiSoft
  Tom Dagostino, Teraspeed Consulting Group
  Kazuyoshi Shoji, Hitachi
  Sadahiro Nonoyama

Everyone in attendance marked by *

NOTE: "AR" = Action Required.

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

Call for patent disclosure:

- No one declared a patent.
  - David asked if declarations must be made before filing.
  - This is an EIA requirement.

AR Review:

- Bob look into test load measurement points BIRD

New items:

- BIRD 105 may be voted next IBIS open forum meeting
  - Bob: We need a fresh look at this, instead of of adding more keywords.
  - Anders: [Test Load] needs more flexibility.
  - Mike: We still need an example IBIS file with golden waveforms
    - Anders: We could create one using a part with publicly available
      IBIS and SPICE models.

AR: Anders create example IBIS file with golden waveforms

- Continued review of the IQ Specification:

5.2.9.	{LEVEL 3}  [Model Spec] Pulse subparameters complete
- David: This models the inertia of an input.
- Bob: This is rarely found in data sheets.
- Mike: Every input has this characteristic.
- David: This testing is generally not done.
  - Minimum clock periods preclude these glitches
  - Mike: Negative reflection from a long stub could create this effect.
  - David: Design parameters are set to preclude this
- David: Is Pulse_high higher than Vinh or lower?
  - IBIS spec shows pulse immunity after crossing 1st threshold, not 2nd
  - Vinl/Vinh gray region denotes possibility of switching.
  - Pulse_time eliminates that possibility if pulse is short.
- We changed this check to clarify it.
- Bob: Maybe this check is asking for too much.
  - If these are absent designers have to keep glitches out.
- David: In general we need a way to discuss requirement relaxations.
  - Mike: Sometimes relaxations are important:
    - Strictly adhering to some limits leaves "margin on the table".
- David: We should reintroduce the "OPTIONAL" keyword
  - Mike: We used to have:
    - CONDITIONAL - Check required if the part has some characteristic
    - OPTIONAL    - Check not required, but suggested.
- We decided to make this check OPTIONAL

AR: Mike check IBISCHK code for 5.1.10-11
AR: Mike add explanation of OPTIONAL and make 5.2.9 OPTIONAL

5.2.10.	{LEVEL 3}  [Model Spec] Pulse_high greater than Vinh
5.2.11.	{LEVEL 3}  [Model Spec] Pulse_low less than Vinl
- We decided to delete these.

AR: Mike mark 5.2.10 and 5.2.11 for deletion

5.2.12.	{LEVEL 3}  [Model Spec] Pulse_time reasonable
- We should have some kind of sanity check for this
  - What if Pulse_time is 1 second?
- Rise/fall time of probable input signals is not a good metric to use.
- We will revisit this after seeing what IBISCHK does with it

AR: Mike check IBISCHK code for Pulse_time checks

Next meeting:

03 Jun 2008 11-12 AM EST (8-9 AM PST)

Meeting ended at 01:01 PM Eastern Time.
