================================================================================ 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 May 16, 2018 Meeting (* means attended at least using audio) ANSYS Curtis Clark Cadence Design Systems Bradley Brim Intel Corp. Michael Mirmak* 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* 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 11 meeting. Mike LaBonte moved to approve the minutes. Bob Ross seconded. The minutes were approved without objection. Review of ARs: - Mike to send out the BIRD189.6_draft2_v2. - Mike reported this is done. Opens: - None. BIRD189.6_draft3_v1 Review: Michael shared a new version of the draft, which he has removed the strike out text. He noted that we still need to resolve the highlighted text and asked Bob if we have received any wording changes to the text. Bob asked if this draft has the latest changes, as he sees the word "subset" and the paragraph from Walter Katz is not included. Michael stated we should work with the draft that has those changes. On pages 34 and 35 of BIRD189.6_draft2_v2, Michael removed the strike out text. Bob agreed that this text is no longer needed. Michael noted that this version does have Walter's new paragraph and the word "subset" has been replaced. Michael asked if the highlighted text needs to be resolved or is new. Bob commented the phrase "two or more" need to be removed. This would violate the one-to-one rule for I/O pins. Michael asked if we should add any text related to this. Bob thought the text is okay, as we make the one-to-one rule clear elsewhere. Bob suggested to remove the highlighting on the second group of text. Michael asked about the other highlighted sections on page 34 and if changes are needed to these. He thought the example regarding "10 VDD POWER" would need some editorial changes. Bob suggested to break out the example text, add the [Pin] keyword followed by ellipsis, and to make the font courier to be more consistent with other examples. Michael made these changes and asked if quotes are needed for the Terminal_type line. Bob agreed we should add the quotes. Michael noted we still may need to make some slight changes in the editorial work to this section. Michael noted one editorial change to the BIRD189.6 revision history sentence on page 5, where a period was missing. On page 17, Randy Wolff asked about the GND and POWER terms and if we should say these are model_types rather than pins. Michael commented we are not consistent with these in the IBIS specification. Bob stated since these are reserved names it is okay. Michael noted we could log this as an issue with the entire IBIS specification. Bob thought that this might not be worth the effort to clean up. Michael, on page 25, changed "die pads" to "die pad" in the list of what an [Interconnect Model] may contain. Randy asked if "pin name" should be "pin_name" in the preceding paragraph. Michael agreed. Michael noted on page 34 there is a comment that can be resolved. Randy suggested on page 47 to change "Example 12 and 13" to "Examples 12 and 13". Bob noted the new Example 13 is a decoupling capacitor model. Mike suggested to change "decap" to "decoupling" for the name of the Interconnect Model. Bob commented "decap" is a common abbreviation used in industry. Mike thought to at least change "De_Cap" to "Decap" for the Interconnect Model names. Michael asked about how the change revision would work for the red text. Mike noted he would generate comparison between document between BIRD189.5 and BIRD189.6 to make the changes clear in the final posted version. He asked if the intention is to make the text red in the actual IBIS specification. Bob agreed we should remove the red text as his intention was to show changes relating to the rail rule relaxation. Bob noted that there is a statement about the changes being shown in red, and we should delete this statement. Michael deleted this statement on page 6. Arpad Muranyi asked if we need to add an additional rule for PDNs with missing rail terminals based on the email discussion over the reflector. Randy commented the relate text is on page 10. Arpad commented that he noticed we do not have a third bullet relating to PDNs, as we do for I/Os. Walter had sent an email stating this was an oversight. Arpad proposed some text over email and Randy had agreed with this text. Arpad suggested to add this as a third bullet under the PDN section. Randy stated we are trying to describe the case when there is no PDN model defined with BIRD189 syntax. Bob commented we can use [Voltage Range] if a PDN model is not included. Arpad noted we do not want to mandate what to do if there is no PDN model from BIRD189. We do not want to force the EDA tool to use the legacy package model. Bob there could be technical problems, and we need to have [Pin Mapping] to ensure power rails connect appropriately. The EDA tool would have to make these connections. Arpad noted that this is also true for I/Os, and he asked if we should delete the similar note for I/Os. Mike asked about the phrase "in this document" and if the intent is IBIS 7.0. Arpad replied that the intent is IBIS 7.0. He wanted to avoid using the word "legacy". Arpad noted the intent is that a model maker can still use the legacy package model formats. Bob commented that there could be issues with tools handling this in different ways. Arpad moved to add his proposed text as a new bullet item to the document. Randy seconded the motion. There were no objections. Bob moved to submit the cleaned up draft version as BIRD189.6 to the IBIS Open Forum. Walter seconded the motion. Arpad asked what the cleaned up version means. Michael clarified that this version will have the red text changed to black and the markup will be based on a comparison of BIRD189.5 and BIRD189.6. Michael to send the BIRD189.6_draft3_v1 version to Mike for posting and submission to the IBIS Open Forum [AR]. Next Meeting: The next meeting is TBD. Mike moved to adjourn. Arpad seconded. The meeting adjourned without objection.