====================================================================== IBIS INTERCONNECT TASK GROUP http://www.ibis.org/interconnect_wip/ Mailing list: ibis-interconnect@freelists.org Archives at http://www.freelists.org/archive/ibis-interconn/ ====================================================================== Attendees from September 1 Meeting (* means attended at least using audio) ANSYS Curtis Clark Cadence Design Systems Bradley Brim Cisco David Siadat Intel Corp. Michael Mirmak* Keysight Technologies Radek Biernacki, Ming Yan Mentor, A Siemens Business Arpad Muranyi* Micron Technology Justin Butterfield*, Randy Wolff SAE ITC Maureen Lemankiewicz, Logen Johnson Signal Integrity Software Walter Katz*, Mike LaBonte* Teraspeed Labs Bob Ross* University of Aveiro in Portugal Wael Dghais Michael Mirmak convened the meeting. No patents were declared. Justin Butterfield took minutes. Review of Minutes: - Michael called for review of the minutes from the August 30 meeting. Mike LaBonte noted that the minutes say he convened the meeting, while in fact Michael ran the meeting. Arpad Muranyi moved to approve the minutes. Mike seconded. The minutes were approved without objection. Review of ARs: - Walter Katz and Arpad to look into if the current syntax will support using node 0 as the reference and propose language if it does not. - Arpad reported that he and Walter did discuss the issue, and they found that this could be addressed if interconnect models can be referenced to node 0. Their idea is to add a new Terminal_type called A_gnd which would connect a Terminal line to node 0. Although, he wants to wait to distribute this idea until we have a clean draft of the BIRD to work from. Walter commented that we do not want to have unused terminals for ISS, while for Touchstone we can have unused terminals which can be terminated. The issue is the N+1 terminal which must be associated with a power or ground reference. The problem is when there are no references defined in the [Pin] list. The only thing the EDA can do in that case is to connect the reference to node 0. The question is how to handle this in IBIS. The options are to handle it internally in the EDA tool or to specify node 0 as A_gnd, for instance. Arpad stated that for ISS if all the terminals are connected, there is no need to identify a reference. We should remove the term reference from the ISS examples. For Touchstone in the case there are no pins to connect the reference, we need a syntax to connect to node 0 calling it Ts_ref or A_gnd. Bob commented that if you have a model with only power supplies and no grounds supplies, you would still have the option to use one of the power pins as one of the references. For ISS models, he agrees that we need to bring out all the supplies rather than referring to one as the reference. Michael asked if we are adding an order of precedence that if a reference is not supplied, then we default to node 0. Walter said yes, but the question is how to handle the case when you do not have a rail to connect the reference to. Walter added that another option would be to give an error if there is no power or ground rail to connect the reference to. Arpad suggested to wait on further discussion until he and Walter have officially made their proposal. - Bob to draft a rule in regards to how [Pin Mapping] interacts with BIRD189.5 in terms of bus_label. - Bob reported that this still needs to be discussed. Opens: - None BIRD189.5 draft 6: Arpad thought we made a decision to remove the Unused_port_termination parameter, but we need a clean draft to edit. Mike commented that he has a version of draft 6 which has the Unused_port_termination parameter removed. He did send out a version called bird189.5_draft6_WIP1.docx that includes this change. Bob wanted to make sure the latest draft has the unused port rule in the Terminal line rules as a separate paragraph. Mike showed the latest bird189.5 draft6 WIP1. In the terminal line rules, Arpad had made a motion over e-mail to change this section and that is reflected in this copy including adding the word "definition". Mike asked if we needed the phrase "by the EDA tool". Bob stated that Radek prefers to have no termination of unused ports. Mike proposed to keep the text as is for now. Arpad stated that Radek also had an issue with the word "may" in regards to adding the termination for unused ports and prefers to force them to an open. Arpad stated that he still has some follow up questions, as he does not agree with forcing an open. Bob suggested to add the unused port sentences as separate paragraphs in the Terminal line rules section. Michael suggested to add some comments saying there are issues to be resolved with this section and move on. Our main goal right now is to get a single latest draft. Arpad moved to use this draft (bird189.5_draft6_WIP1.docx) with the discussed changes as the latest working draft. Walter seconded. There were no objections. Walter commented that he has discussed with Brad Brim, and Brad does not want any restrictions on how the EDA tool terminates unused ports. Walter stated that we may get two alternate versions of the unused port rules and vote on those. Arpad requested that Radek's comment on terminating with the port reference impedance be added to the BIRD draft comments. Michael stated that the missing die pad to pin model issue and subsequent order of precedence is the next issue. Arpad commented that the issue is understood, but language needs to be added to the BIRD. Arpad and Walter will take an AR to draft text for this. Bob asked if the die pad to pin path is missing, how do we do know that the path does not exist elsewhere in the [Interconnect Model Set Selector]. Walter said that Arpad and his proposal to change to an [Interconnect Model Set Group], would solve the issue. The group could have sets that are grouped together and work together. There is no decision that the EDA has to make, as only one group would be used at a time. Bob commented that this would be a lot of changes to the BIRD. Walter stated he would like to get a clean draft, and he will make the changes and we can review and decide how to proceed. Bob asked if the [Interconnect Model Set Group] can have only one set. Walter replied, yes, it could have one or more sets. Mike will post draft 6 and email it out to the group. Walter moved to adjourn. Mike seconded. The meeting adjourned without objection.