================================================================================ 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 June 3, 2020 Meeting (* means attended at least using audio) ANSYS Curtis Clark* Intel Corp. Michael Mirmak* Ifiok Umoh Eric Edwards Keysight Technologies Radek Biernacki Mentor, A Siemens Business Arpad Muranyi* Micron Technology Justin Butterfield* Randy Wolff* SiSoft Walter Katz Mike LaBonte Teraspeed Labs Bob Ross* Zuken USA Lance Wang* 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 May 27, 2020 meeting. Arpad moved to approve the minutes. Randy seconded. The minutes were approved without objection. Review of ARs: - Michael to send out EMD draft12. - Michael reported this was done. Opens: - None. EMD Draft Review: Bob noted the draft title was listed as 10. Michael noted we had left off on connection rule 2 regarding Rail connections. There was a question if rail terminals can exist without IO terminals and if the converse of this is true as well. Michael noted the rule in 2.a.1 existed, and we added equivalent reverse rules for 1.a.i and 1.b.i. Michael noted the phrase "shall be connected" was changed to "shall be considered connected". He asked if the other rules should be changed to match this. Randy agreed this would be a good idea. Bob noted, in rule 1, they are considered connected, but in rule 2, they are considered shorted. Michael asked if they are considered the same net. Bob noted when we have bus_label they are shorted. When the same signal_name is used for the IOs, they are not shorted. Michael changed 2.a.ii and 2.a.iii to say "shall be considered shorted". Bob noted we are referring to a single interface with these rules. Michael noted rule 2.a.iv deals with the overlap cases. He asked if the reverse of all these are true. Bob stated rule number 2.a.v deals with overlapping pin_names. Randy asked if we could look at an example to discuss this, such as the example at the top of page 35. Michael noted these are all Pin_Rail with bus_label and signal_name. Bob commented the power pin U3.W could be specified as one terminal and the VDD could be specified by another terminal and overlapping these would be illegal. Randy commented the pin or pin_name cannot be overlapping with the bus_label. Randy asked about the word "overlap" and if "subset" would be better. Michael asked which would be the subset. He asked if "overlap" could be changed to "appear" or "contained". Bob stated every pin has a signal_name, you can have a bus_label as a subset of the signal_name. We do the same check with the Interconnect Model syntax. Michael asked if we want to add a Venn diagram to the specification to describe these cases. Bob replied it could be helpful to have a Venn diagram example. Michael added this to the bin list. Michael asked if the phrase "shall be connected" in rule 2.a.v is still correct. Randy suggested to use "shall be shorted". Randy asked about the sentence not being complete. Michael noted the sentence is completed below and suggested to add a colon. Bob suggested to change "any interface" to "any one interface". Bob noted rule 2.b is the global pin rule, which shorts all connections for all designator interfaces, and the emphasis is on the all interfaces rather than any one interface. This rule means everything gets shorted to a single signal_name terminal. Michael asked what "global" means in this case. Randy noted, in the last example, the *.VDD is used, which is a completely merged mode for the EMD pins and Designator pins. This merges all these signal_names to one terminal. Randy noted we cannot remove the examples, since they show these rules. Bob noted we could have more pathological examples to highlight rules. Michael asked if these rules cross EMD Model, Set, and Group boundaries. He asked about the header of 2.b which is simple compared to 2.a. He suggested to make the header similar to 2.a. Michael suggested to say "Pin_Rail connections shall be consider shorted across Designators". Michael asked about the word "global". Bob replied you do not short the EMD signal_name with the Designator signal_name, so this rule applies to only the Designators. Randy suggested to add a parenthetical stating "(not the EMD interface)". Bob noted it is optional to use this shortcut. Arpad suggested to say "merged into one node". Randy suggested to use the word "terminal". Arpad asked if we should use this wording elsewhere. Bob noted we need to be careful about logically connected and shorted differences. He noted this shortcut is very close to using the connection as one terminal on the board. Bob asked if we should delete the rules marked for deletion following this section. All of this should be already covered in other sections. Michael will send out EMD draft13 [AR]. He asked everyone to review rule 2.b. Next Meeting: The next meeting will be June 10. Arpad moved to adjourn. Randy seconded. The meeting adjourned without objection. ================================================================================ Bin List: EMD Comments to be Resolved: (See BIRD202.1 tracking spreadsheet) IBIS-ISS Parser: - IBIS-ISS parser scope document