================================================================================ 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=cFf1Q9UoZA2mPvCn4-PROdwE3IxkMDWESHR7txjvOsg&s=rdbapUWOoG1OjPlcCjsToFvk3Gve-GOJIqCUjF6wLLU&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=cFf1Q9UoZA2mPvCn4-PROdwE3IxkMDWESHR7txjvOsg&s=Gd30hwNO6UpTk1ro75mkFDkr1zFa3ojAeBbdqUgw8MA&e= ================================================================================ Attendees from February 10, 2021 Meeting (* means attended at least using audio) ANSYS Curtis Clark Intel Corp. Michael Mirmak* Mentor, A Siemens Business Arpad Muranyi* Micron Technology Justin Butterfield* Randy Wolff* SiSoft (MathWorks) Walter Katz 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 February 3, 2021 meeting. Michael displayed the minutes. Randy Wolff moved to approve the minutes. Lance Wang seconded. The minutes were approved without objection. Review of ARs: - Randy to add the EMD Parts Selector proposal to the BIRD [AR]. - Randy stated he does not have a new draft yet. He suggested we need to settle on our ideas before issuing a new draft. Opens: - None. EMD Parts Selector Discussion: Randy noted there was some discussions over the email reflector. Arpad Muranyi commented his question was if the solution being discussed would be general enough for everyone. Michael asked how many new keywords are we introducing. Randy replied there is only one new keyword. The sticking point was how to handle the nesting. The proposal would be to call the config_name in the top-level EMD Parts for the nested case. Randy commented this allows for good flexibility at the lower levels, and the configurations can be limited at each higher level. Arpad asked if there was a need for the configuration selector at the lowest levels. He asked if at the higher level only one of those configurations would apply. Randy replied, if someone is creating a nested EMD, they should point to the necessary and valid combinations at the top-level. Arpad asked if the GUI at the top-level could make the final selection. Michael asked if it would be simple for the user to understand the options they should pick for the combinations. Arpad asked about the hard-coding of the options at the top-level and if this limitation is okay for the user. Randy replied that selecting the configuration allows us not have to make the selection explicitly for the lower level files. Arpad asked if this would corner us in other situations. Michael could not think of cases where this limitation would be a problem. Arpad asked if the EDA tool would be expected to only provide the top-level selection for the configuration. Bob Ross suggested we cannot prevent the tool from selecting the configuration at the lower level. Bob was concerned about nesting at different directory levels. Arpad asked if the files can be in different subdirectories. Michael stated we have text about this in the IBIS specification. Bob was concerned we would need to support a selector at any level in any directory. Michael suggested we can add this to the bin list. Bob stated we could have a selector within a selector. Michael thought this was an issue for the rest of the IBIS specification as well. Michael asked what changes are needed in the BIRD document. Randy replied we need to look at how we define the EMD Parts name for the configuration name. The configuration selector would need to have 3 columns to consider the Voltage List. Michael asked if we can have one name to select all the combinations. Randy replied the name would come from the EMD Configuration Selector. We could use a common name for the EMD Parts and Voltage List. The other option is to have multiple columns, but we would need to decide which name to use at the top-level EMD. Bob suggested to use the EMD Parts list name and have an additional column for the Voltage List. Arpad asked how the top-level selection would work if there are multiple EMD Parts lists with the same name. Bob suggested to name the EMD Parts and the Voltage List uniquely. Arpad suggested to have a numerical integer name for the configurations. Bob was concerned that the nesting selection should be possible at any level. The Voltage List and EMD Parts should be selected as a pair. Randy was also concerned about how the EDA tools would do the selection of EMD Groups. Michael asked if the problem is that we do not know how the tool will go through the selection tree. Arpad stated the EDA tool needs to build a netlist. When a selection is made by the user, the design needs to be re-netlisted. Bob suggested we need some examples. Randy could come up with some examples to show the hard-coded configurations. Michael asked if Bob could come up with some pathological examples of how things can go wrong. Bob would have to think about it, and he noted the Group selection could go wrong. Arpad asked, if we had a selection for the configuration, would we need the Group selection. Randy replied the Group selection would be for a crosstalk model and a model without crosstalk. Randy to come up with a practical example to show the hard-coded configurations [AR]. Bob to come up with a pathological example [AR]. Next Meeting: The next meeting will be February 17. Arpad moved to adjourn. Randy seconded. The meeting adjourned without objection. ================================================================================ Bin List: IBIS-ISS Parser: - IBIS-ISS parser scope document