IBIS Macromodel Task Group Meeting date: 10 May 2011 Members (asterisk for those attending): Agilent: Fangyi Rao * Radek Biernacki Ansoft: Chris Herrick Danil Kirsanov Ansys: Samuel Mertens * Dan Dvorscak Deepak Ramaswamy Jianhua Gu * Curtis Clark Arrow Electronics: Ian Dodd Cadence Design Systems: Terry Jernberg Ambrish Varma Celsionix: Kellee Crisafulli Cisco Systems: * Mike LaBonte Stephen Scearce Ashwin Vasudevan Ericsson: Anders Ekholm IBM: Greg Edlund Intel: * Michael Mirmak LSI Logic: Wenyi Jin Mentor Graphics: John Angulo Vladimir Dmitriev-Zdorov Zhen Mu * Arpad Muranyi Micron Technology: Randy Wolff NetLogic Microsystems: Ryan Couts Nokia-Siemens Networks: * Eckhard Lenski Sigrity: Brad Brim Kumar Keshavan * Ken Willis SiSoft: * Walter Katz Mike Steinberger * Todd Westerhoff Doug Burns 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: MM would like to plan what should go into 5.1/5.2 -------------------------- Call for patent disclosure: - None ------------- Review of ARs: - Bob write a BIRD on correcting Table 1-3 in the spec. (Row 23). - In progress, on hold ------------- New Discussion: Michael M showed a list of active BIRDs: - Some technical changes may be important enough to include with editorial changes - They may be easier described with advanced graphics - Consensus that editorial plus some critical changes might go into 5.1 - Radek: The AMI_Version param should not be deferred - MM: 5.2 might follow 5.1 closely - Arpad: 5.1 should have AMI_Version to be safe - MM: We also should consider how much the parser would change Arpad: No plan to discuss the Out/InOut BIRD today: - Arpad is still working on it - Ken: Parameters with Usage = IBIS-ISS should really be Model_Specific - Arpad: Yes it could be done either way - With Model_Specific the tool is just passing data to the IBIS-ISS subcircuit, which is another model - If Info is included as not allowed for Model_Specific then we could not pass parameters to IBIS-ISS subcircuits - Ken: Things like corners will dictate what parameters are passed to the IBIS-ISS subcircuit - Todd: Ken is correct - Arpad: Only a single value is passed to the IBIS-ISS subcircuit Arpad showed the Table BIRD from Bob Ross: - Bob is at the European IBIS summit this week - We could schedule this for a vote next week - Entire tables are sent by the model - Arpad: I added the sentence that parentheses are remove after flattening - Multiple calls to GetWave are clarified - Radek: There is an inconsistency between: "A table is sent to/from" "number of rows returned by the executable model" - Arpad: This should be corrected - Bob removed a paragraph because it repeated the spec AR: Arpad update Table BIRD Arpad showed the task list: - Row 51: Init and GetWave are described differently - This should be fixed for 5.1 - Arpad showed the IBIS 5.0 spec section 3.1.2.6 - It doesn't say if parameters are required or optional - 3.2.2.4 says AMI_parameters_out is optional - Radek: It can be a null pointer or an empty string - Arpad: This should be a 5.1 change - Mike L and Radek agreed, no objections - Arpad: Should row 52 be done after 5.1? - Walter: That is OK - Arpad: Should we drop row 53? - Walter: We should keep it - Arpad: Bob's Table BIRD may cover this, but this is about List - No objection to putting this in 5.1 - Row 54: We need to spell out that Corner is reserved - Tools' existing typ/min/max mechanism should be used, not a new UI control - Walter: Agree - Row 55: Can this go into 5.1? - No objection - Row 56: Can this go into 5.1? - No objection - Row 57: Can this go into 5.1? - Default values make no sense for Out parameters - No objection - Row 58: This should go after 5.1 - This is already in the In/InOut BIRD - No objection - Row 59: Can this go into 5.1? - Walter: This changes existing functionality - Radek: We can decide later - It will be after 5.1 - No objection - Row 60: Can this go into 5.1? - No objection - Row 61: Can this go into 5.1? - Walter: This should go into 5.1 - No objection to that Walter showed "Time Domain GetWave Flow": - The jitter BIRD introduces several parameters - BIRD 120 adds clock_times input to CDR - Arpad: This goes with BIRD 120? - Walter: Yes AR: Walter send Time Domain GetWave Flow diagram for posting Walter showed the "AMI Backchannel Response to Sigrity/Gennun" presentation: - There can be a list of backchannel protocols - Reserved params are in red (page 3) - Branches under protocol name (pages 4 and 5): - Simulator_Commands, mostly training information - These are owned by IBIS (reserved, colored in brown) - GetWave_Commands - EDA tool does not need to know about these - This information could be in an external file - Stimulus patterns defined on age 6 - Can use binary, hex, octal - Parameter passing flow example on page 7 - Tools could record these and show to user - Page 9: A protocol registry is needed - There can be unregistered too - Ken: Is this feedback on our BIRD or a replacement? - Walter: Would like to work on a joint BIRD - Both proposals use BIRD 128 AR: Walter send Backchannel proposal for posting ------------- Next meeting: 24 May 2011 12:00pm PT Next agenda: 1) Task list item discussions ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives