================================================================================ IBIS INTERCONNECT TASK GROUP https://urldefense.proofpoint.com/v2/url?u=http-3A__www.ibis.org_interconnect-5Fwip_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=aZKSGi0uPDT4RFDkJsWxQosKz3LT9RQ2f8MGmZs1Gvw&s=BvtBVKdoe1P5IVOS35IV6aDeK-AzhLIgAKMKFcyPkdM&e= Mailing list: ibis-interconnect@freelists.org Archives at https://urldefense.proofpoint.com/v2/url?u=http-3A__www.freelists.org_archive_ibis-2Dinterconn_&d=DwIGAg&c=nKjWec2b6R0mOyPaz7xtfQ&r=DcQR-qLpQg5lIreuM6-NYECRIAFXt268PRNS5WO043M&m=aZKSGi0uPDT4RFDkJsWxQosKz3LT9RQ2f8MGmZs1Gvw&s=o-R9jnJY3izDWD_nORXKClM8GV9U7VR148fKYjWm5CI&e= ================================================================================ Attendees from July 7, 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. Opens: - None. EMD Draft16 Review: Randy shared the latest EMD draft. The first section of yellow highlighted text to review is on page 37. There is a comment about keeping some of the statement regarding the EMD models not being used together in simulation. Arpad asked if this is already covered in the rules. Randy commented, in rule 1.b, there is a similar statement. Bob commented the point of the Aggressor_Only rules is to allow duplicate models. You cannot have identical signal_names if they are not set to Aggressor_Only. Arpad asked if we are saying the same thing in different ways. Bob replied two EMD models on the same pin without the Aggressor_Only flag should not exist and would be an overlap case. Arpad did not see this in rule 1.b. Randy noted this is related to rule 1.a.iv. Bob noted rule 1.a.ii also applies. Arpad asked if we can delete the highlighted text. He suggested we could clarify rule 2.b.ii with this statement. Randy suggested to add to rule 2.b.ii the phrase "but will not be used together in simulation." Bob commented the original concept was within one Group we have selections of models, and the simulation pins can be chosen appropriately. If we have an IO pin without an Aggressor_Only, only one model would be selected and both models cannot be used together. This is an extension of that original rule. Randy noted the next highlighted section mentions figures which do not exist. Michael asked if we need to create similar figures to the Interconnect section, or if we can delete this sentence. Randy agreed to delete this. Bob noted there is no equivalent statement in the rules. The difference from Interconnect is that the signal_names are carried through rather than the pin_names. Randy stated the next highlighted section is similar to the rules in 1.b, but has a good description of Aggressor_Only. Bob commented this is broken up in the new rules. Arpad suggested to move this over to rule 1.b and wordsmith the text later. Michael noted some of the wording and terminology are different. Randy suggested to review this later. Arpad suggested to correct the last sentence word from "rails" to "rail". Randy noted the next section relates to the example and asked if it is useful to move this section to above the example. Arpad agreed this would be good to move before the examples. Randy commented we may want to make this an IBIS comment. Randy noted the next section of highlighted text is a statement regarding all non-rail pin_name pins on page 13. Bob commented the context is that a model might exist without rail pins and only a reference pin, and another model may fill in all the rail pins at each interface. One question is if a rail pin is required. Randy replied you should not always have to have a reference. Bob commented the reference might be supplied elsewhere. Randy asked if we should put an explicit statement about the rail pins under the Designator Pin List. Michael and Arpad agreed this would be good. Randy asked if we want to require listing all the non-rail pins for the device. Arpad stated we can check and enforce this based on the IBIS Component or EMD being used at the Designator. Bob noted the unused pins would have to be no connects. Arpad asked if we want to keep this rule. Bob suggested random partiality of the model pin lists would make checking difficult. Randy suggested we could change the statement to say intended rather than required. Arpad noted the parser could flag this an error, but the EDA tool should still be able to simulate. Randy noted NC can be used for unused pins. Arpad suggested we could remove the rule. Bob noted the rule is if the IBIS Component has 10 pins, then there should be 10 pins in the Component. Arpad noted the EMD rule is more prohibitive, since it can be enforced with the parser. Bob noted a warning or error could be issued. Arpad asked if we want to keep this rule. Bob stated yes. Randy asked if we want a similar rule requiring all the rail pins. Bob noted it is a different issue for rails since they are connecting the same pins. Arpad suggested, if the IO pins are all required, the rail pins should also be required. If we require that all pins are listed, then we need to spell out what happens when they are listed but no models are assigned. Bob noted disconnected models are left as opens for EMD, and we cannot issue a default for EMD. Bob suggested to keep the rule for all the pins. The reason is the file may be connected to a database. Arpad suggested to change the language and remove the non-rail qualifier from the rule. Randy stated the paragraph deals extensively with the IOs and suggested to add a new paragraph to deal with the rails. Arpad suggested to add a comment to the draft and come back to this issue. Randy noted this completes the highlighted text review. He noted one remaining issue is a possible example for item number 5 on page 38. Next Meeting: The next meeting will be July 8 at 8:00am PT. 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