IBIS Macromodel Task Group Meeting date: 20 Dec 2011 Members (asterisk for those attending): Agilent: * Fangyi Rao Radek Biernacki Altera: * David Banas Ansys: Samuel Mertens * Dan Dvorscak * Curtis Clark Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg * Ambrish Varma Feras Al-Hawari Celsionix: Kellee Crisafulli Cisco Systems: Ashwin Vasudevan Syed Huq Ericsson: Anders Ekholm IBM: Greg Edlund Intel: Michael Mirmak LSI Logic: Wenyi Jin Maxim Integrated Products: Mahbubul Bari Mentor Graphics: * John Angulo Zhen Mu * Arpad Muranyi * Vladimir Dmitriev-Zdorov Micron Technology: Randy Wolff NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: * Eckhard Lenski QLogic Corp. James Zhou Sigrity: Brad Brim Kumar Keshavan * Ken Willis SiSoft: * Walter Katz Todd Westerhoff Doug Burns * Mike LaBonte Snowbush IP: Marcus Van Ierssel ST Micro: Syed Sadeghi Teraspeed Consulting Group: Scott McMorrow * Bob Ross TI: Casey Morrison Alfred Chong Vitesse Semiconductor: Eric Sweetman Xilinx: Mustansir Fanaswalla The meeting was lead by Arpad Muranyi ------------------------------------------------------------------------ Opens: - Arpad: There will be no meeting next week - The next meeting will be 3 Jan 2012 -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Walter update BIRD 123 - Done, posted in the work archive - Walter create jitter derivation document - Done - Ken propose reduced set of jitter parameters - Decision to support dual models causes a problem - Dual sets of jitter params are required - This AR is done - David write mathematical description of combined Dj proposal - Done, language is in BIRD 123 ------------- New Discussion: BIRD 123.3 draft 5: - Walter: There have been extensive discussions with David, James and Fangyi - Ken: I have suggested deprecating the old jitter parameters - Walter: They still are useful where a PDF table is given - Walter: Some changes: - Usage Out has been removed - EDA tools should use last returned value from each Getwave - Don't understand about "how EDA tool can input the clocks" - Arpad: That may be an old idea, no longer valid - Ken: TX Jitter is redundant but RX is not? - Walter: TX Jitter supports PDF, which the new parameters do not support - Ken What if TX Jitter is defined in both old and new ways? - Walter: Combine them - Ken: We should have guidance for developers - Fangyi: That is a good idea - Walter: It would be best to discuss that in email - Arpad: There are a few typos in the BIRD - Arpad made some corrections which will be sent to Walter - The BIRD draft will be voted for submission next meeting - David: Is there any doubt about the intended use of the RX ... parameters? - Walter: It is well documented - Vladimir: What is the final decision on TX jitter deprecation? - Walter: We should keep them but TX DCD can no longer be Out - Ken: I would prefer to take the old ones out and enhance the new ones - Arpad: We should keep the overlap for one release - Walter: Deprecation is usually a warning of future removal of support AR: Arpad send BIRD 123 changes from today to Walter Arpad showed a presentation on Usage Out parameters: - Slide 2: - Arpad: The phrase "Output only" might be confused - It should mean the value always comes from the DLL - Walter: Correct - Arpad: "All parameters must be ..." requires Out parameters to have a value in the AMI file - How would these values be used? - Walter: It should be ignored - Ambrish: It could restrict the expected return values - The EDA tool can check the returns - Walter: What action does the tool take if the DLL returns a random string? - We can't specify all possible return values - Arpad: Is Default supposed to initialize something? - Walter: No - It is the default for a selection list in a GUI - The default may be between the smallest and largest value - But there should be an implicit default otherwise - Bob: Default replaces Typ - BIRD 140.2 species this - Walter: The Default must be an item in the list - Arpad: BIRD 140 specifies implicit choices when Default is missing - Arpad showed BIRD 127.4 - This restricts Default values to the allowable set - Arpad showed BIRD 140.2 - This specifies that Default overrides Typ for certain parameter types - Slide 3: - Arpad: This diagram shows an Out parameter passed to the EDA tool engine - Ambrish: It can be used for error checking - Arpad: Then it should be an Info - David: The model might provide a refined value - Walter: Then it should be an InOut - Walter: A 4 x 5 table is returned as 20 numbers - The AMI info is needed to know the row column arrangement - This is the only time that info is needed - Arpad: The spec should say that - Slide 4: - Arpad: How is Default used here? - Walter: Default is effectively Value here - Slide 5: - Arpad: If no Default, the first item of a List is taken - Slide 6: - Arpad: Default is different here for an In parameter - Slide 7: - Arpad: Does the user make a List selection for an Out? - Ambrish: The DLL chooses - Arpad: Is the List used to check the model returns? - Ambrish: Yes - Walter: Disagree - Slide 8: - Arpad: The DLL does not know Corner, how will it choose? - Walter: Create a MyCorner Model Specific to tell the DLL - Ambrish: We need to use these to check valid values - Walter: The benefit of that is small - Slide 9: - Arpad: Please email suggestion for further refinements - Ambrish: We should add a new Format to allow returning anything - Arpad: I will not address this now - Bob: BIRD 140.2 might help but we have not voted on it yet - Walter: Format should be required for Out only for Table - The tool should ignore Format otherwise - Bob: Disagree - Arpad: Model makers will not know to specify all possible outputs Meeting ended. ------------- Next meeting: 03 Jan 2011 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives