IBIS Macromodel Task Group Meeting date: 30 August 2016 Members (asterisk for those attending): ANSYS: * Dan Dvorscak Curtis Clark Broadcom (Avago): Xingdong Dai * Bob Miller Cadence Design Systems: * Ambrish Varma Brad Brim Kumar Keshavan Ken Willis Cisco: Seungyong (Brian) Baek eASIC: David Banas Marc Kowalski Ericsson: Anders Ekholm GlobalFoundries: Steve Parker IBM * Luis Armenta Intel: * Michael Mirmak Keysight Technologies: Fangyi Rao * Radek Biernacki * Ming Yan Maxim Integrated Products: Hassan Rafat Mentor Graphics: John Angulo * Arpad Muranyi Micron Technology: Randy Wolff Justin Butterfield QLogic Corp.: James Zhou Andy Joy SiSoft: * Walter Katz Todd Westerhoff * Mike LaBonte Synopsys: Rita Horner Kevin Li Teraspeed Consulting Group: Scott McMorrow Teraspeed Labs: * Bob Ross TI: Alfred Chong The meeting was led by Arpad Muranyi. -------------------------------------------------------------------------------- Opens: - Arpad: We have a new agenda item 7 for IBIS 6.2 BIRDs ------------- Review of ARs: - Walter send [Pin Mapping] change proposal to Mike for posting. - Done. - Bob Miller submit Back-channel BIRD to Open Forum - Done, but there is a numbering issue. ------------------------- No patents were declared. ------------------------- Review of Meeting Minutes: - Arpad: Does anyone have any comments or corrections? [none] - Walter: Motion to approve the minutes. - Dan: Second. - Arpad: Anyone opposed? [none] ------------- New Discussion: New Lightweight Backchannel Proposal: - Bob: The word DLL is used throughout, but it should be "executable model", except for the "DLL_Id" parameter. - Bob M.: I'm OK with that change. - Arpad: Should the BIRD be rewritten? - Bob R.: That would be best. - It should be a new BIRD too, not an update to 147. - Part of this should be a in a new IBIS section 10.9, not in 10.8. - Bob M showed the BIRD. - Arpad: We decided last week that discussion on this was finished. - Bob R: The only stated protocol is "basic". - Should there be definitions for Fibre Channel, etc.? - Bob M: Basic describes a hypothetical protocol. - Walter: Basic is a placeholder, we should not imply that it exists. - Bob R: I want to send change suggestions to Bob M. - Radek: We should discuss how protocols will be established. - How will we avoid name conflicts? - Arpad: BCI_ID might help make it unique. - Bob M: Two vendors might use the same name for different protocols. - Walter: An approval process is implied. - The Open Forum can handle that. - We need just a name registry. - Radek: It should be in the BIRD explicitly that private protocols are prohibited. - Or they might start with underscores, for example. - We need to establish IBIS approved protocols. - Walter: That could be in this BIRD or separate from it. - Radek: I would prefer it in the BIRD. - Bob M: We can highlight this in the notes. - Standard protocols might always begin with "IBIS". - Bob R: Private protocols might start with a company name. - Bob M: We don't know who will do this outside of IBIS. - Arpad: Could other organizations approve protocols? - Bob M: Any company could publish their protocol. - Ambrish: PCI-SIG might want to create a new name but IBIS would own it. - Bob M: A protocol could be private or published? - This condenses three options to two. - Radek: We should at least set a naming convention. - Bob R: Prepend with IBIS for official protocols. - If PCI-SIG copyrights PCI-Gen4 we can't publish it. - Walter: We should hope to have that problem a year from now. - The IBIS prefix should be all we need for now. - Arpad: How do we prevent someone from using a protocol name before approval. - Ambrish: We have had this discussion before. - Arpad: Do we discuss this next week? - Walter moved to table the issue. - Bob R seconded. - There were no objections. [Pin Mapping] modifications proposal: - Walter: I'm ready to submit this to the Open Forum. - Walter showed the BIRD. - Bob: I favor submitting this. - Arpad: Time check, are there other IBIS 6.2 BIRDs to discuss here? - Mike: Some 6.2 BIRDs are editorial, some are submitted. - Walter described changes to the Pin Mapping BIRD. - Arpad: Has this been submitted? - Mike: Usually we have a motion to send to Open Forum. - Walter: We usually recommend whether to approve. - Arpad: That is for submitted BIRDs that come back here. - Walter: I will submit it. Model_name and Signal_name Restriction for POWER and GND Pins BIRD draft: - Bob R showed draft 3 of the BIRD. - Bob R described the changes. - Arpad: If the third column has POWER for two pins and the signals are VCC3 and VCC4, does the tool short them together? - Bob R: No. - I plan to submit this. - It also has a few editorial changes. - Arpad: Do CIRCUITCALL and NC fall in this category? - Mike: CIRCUITCALL connectivity is established externally. - Walter: I originally suggested same signal name must have same model name. - If two pins are signal A but have different model names should that be allowed? - We certainly don't want POWER and GND mixed up - Radek: It is OK to limit this to POWER and GND. - Bob R: It would be a violation for pins with the same signal to use POWER and NC model names. - Walter: If four pins use POWER, GND, CIRCUITCALL, and NC some are in violation, but others aren't. - Arpad: We have POWER and GND model names, not pin names. - Bob R: For pins with same signal name all should have the same model name. - Bob R showed [Model Data] Matrix Sub-parameter Terminology Correction BIRD draft 1: - Bob R described the changes. - Michael M and Radek agreed with the changes. - Bob will submit the BIRD to the Open Forum. - Walter: Motion to adjourn. - Michael M: Second. - Arpad: Thank you all for joining. ------------- Next meeting: 06 September 2016 12:00pm PT ------------- IBIS Interconnect SPICE Wish List: 1) Simulator directives