Minutes, IBIS Quality Task Group 25 January 2022 12:00-13:00 EST (09:00-10:00 PST) ROLL CALL ANSYS Curtis Clark Intel Technology Michael Mirmak Micron Technology * Randy Wolff Siemens EDA Weston Beal Signal Integrity Software: * Mike LaBonte Teraspeed Labs: * Bob Ross Zuken USA: * Lance Wang Everyone in attendance marked by * NOTE: "AR" = Action Required. -----------------------MINUTES --------------------------- Mike LaBonte conducted the meeting. Call for IBIS related patent disclosures: - None Call for opens: - None Review of previous meeting minutes: Minutes from the January 18, 2021 meeting were reviewed. Randy Wolff moved to accept the minutes. Bob Ross seconded the motion. Without objection, the minutes were approved. ARs: - AR: Mike LaBonte to write ibischk bug report to eliminate system calls Done. BUG229. - AR: Mike LaBonte to investigate why E0730 is printed only with -emd Done. - AR: Weston Beal to prepare presentation on IBIS Quality Specification and checklist No report. NEW ITEMS: IBISCHK710 development: Bob Ross said the updated message spreadsheet had been received from the developer. Mike LaBonte showed the IBISCHK7 User Guide WIP3 he had drafted using that spreadsheet. He noticed that the copyright year needed to be updated. Also EMS file types were not covered in a few places where they should have been. Bob said now that the message tables were in the document, the count of error messages needed an update. He said the last line of the spreadsheet would help with that. Mike noted that the first two lines in the message table were "W,E6413" and "E,B6801". Those were not in correct sort order. Mike asked how check 6801 could be either B or E. Randy suggested having separate lines for those. Mike and Bob agreed. Mike noted that the E,B6801 CMODEL_ERR_1 message was about a memory failure, which was normal for "B" bug messages. He would update the document. Bob suggested sending out WIP4 for review in advance. AR: Mike LaBonte to send updated draft IBISCHK7.1.0 User Guide for review AR: Mike LaBonte to investigate CMODEL_ERR_1 E,B confusion Parser bug reports: BUG228: Mike LaBonte said we had previously discussed making the check retroactive to IBIS file versions prior to 7.0. Randy Wolff said that should be done specifically when the .ibs file is referenced from EMD. Directly checking the .ibs would not be retroactive, it would be checked for its own [IBIS Ver]. Bob Ross said we would get two messages or one, depending on the .ibs file version. Mike said we should fix the problem where the geIbisVersion variable was being used to select checks for .ibis files, instead of the geFileVersion variable. Bob said that may not be a bug. He felt the behavior was closer to what we wanted. Mike felt it was unintentional, but he said we might let it go for now. Randy asked what the difference was between the E0730 and E5808 checks. The messages seemed very similar. Mike suggested asking the developer. Randy said the .ibs file in the Micron testcase was a clear problem for EMD. He asked if all applicable IBISCHK checks should be checked for IBIS 7.1 compliance under EMD, or just E5808 and E0730. We agreed that all IBIS 7.1 checks should be applied when the file was used as part of an EMD file set. A new message was crafted to explain why the file was being checked for compliance with a newer specification: Checking bug228.ibs for IBIS 7.1 Compatibility. The file has [IBIS Ver] 5.1. However, IBIS 7.1 compatibility is required for IBIS files referenced from EMD. Bob noted that the file would result in two error messages. Mike asked if it should always be an IBIS 7.1 check, the release in which EMD appeared, or should it be whatever IBIS version the EMD file used. Randy felt it should be the EMD version, which someday could be 8.0, with new requirements. AR: Bob Ross to update BUG228 BUG229: Mike LaBonte showed a draft of BUG229, "External package model fails on Windows with UNC working directory". He described the bug. Mike said that the testcase would always pass on Linux, and it would pass on Windows as long as the .ibs file was not in a directory using a UNC path. Bob suggested adding a warning to alert users. Mike said there was a simple solution using standard POSIX functions to read the directory, avoiding the system calls. Randy Wolff asked it the bug should be introduced in the Friday IBIS Open Forum meeting. We agreed to wait, because BUG228 was expected to consume significant meeting time. IBIS quality checklist: No discussion. Tabled topics (no discussion without motion): - BIRD181.2o - IBISCHK security fixes Randy Wolff moved moved to adjourn. Bob Ross seconded. Without objection the meeting ended. Meeting ended: 13:04 ET Next meeting February 1, 2022