Minutes, IBIS Quality Task Group 16 February 2021 11:00-12:00 EST (08:00-09:00 PST) ROLL CALL Intel Technology Michael Mirmak Micron Technology * Randy Wolff 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 February 9, 2021 meeting were reviewed. Lance Wang moved to accept the minutes. Randy Wolff seconded. Without object the minutes were accepted. ARs: - Mike LaBonte to update TSCHK2 for bugs 3 and 4 - In progress. - Randy Wolff to request full static code analysis report from Synopsys - Done. It is integrated into their own code, that prevents sending a report. They might run analysis on just our code. NEW ITEMS: New parser bug reports: None. TSCHK 2.0.1 development: Mike LaBonte described changes made to the TSCHK2 source code. To produce smaller files had changed it to calculate the maximum precision of value read in, and use that to determine the output column width and precision. That was done separately for frequency values and other data values. Mike noted that TSCHK was printing all values in the most compact format, which for frequency usually was scientific notation, even when the input file had integers. He asked if frequency should be integer or scientific? Randy Wolff suggested it should produce whatever produced the smallest files, noting that fractional precision had to be preserved. Mike suggested delivering only 32 bit executables. Those wishing to have 64 bit could compile from source code. AR: Mike LaBonte to produce TSCK2.0.1_alpha3 executables with bugs 3 and 4 fixed IBISCHK development: IBISCHK 7.1.0 contract: Bob Ross reported no update on the contract draft. IBISCHK7 source code feedback: No new feedback had been received. Mike LaBonte said he was waiting for feedback from Cadence. IBISCHK security fixes: Mike summarized the status, noting that this was related to the source code feedback topic. Three static code analysis tools had been used, but we had not yet converged on a policy regarding what we would fix. We decided to wait for further information from Synopsys before proceeding. Interconnect model file checking Bob Ross said he would like Touchstone interface checking as well as IBIS-ISS. Mike LaBonte listed 4 approaches for that: 1) IBISCHK7 could call the TSCHK2 executable, if its path were given on the IBISCHK7 command line. 2) TSCHK2 could be directly linked into IBISCHK7. 3) New IBISCHK code could be written to read Touchstone files with very limited parsing. 4) We could decide not to check Touchstone files. Regarding #2, Mike noted that TSCHK2 was written in C++ but work would be required to set up IBISCHK7 to use C++. He said an "extern C" function might be added to TSCHK.a, that IBISCHK7 could link to without C++ support. Bob Ross suggested we might have it convert Touchstone1 to Touchstone2 to get the number of ports, a form of #1. Mike noted that the code for that would give us the algorithm, should we choose approach #3. Revisiting the TSCHK2 development topic, Mike said he had fixed the eda.org URL in the documentation files. Also Mike said he was continuing to leave a line of code commented out, to prevent errors converting Touchstone 2 files with [Reference] to Touchstone 1. Mike felt the check to verify that the input file was Touchstone 1 made no sense when converting from Touchstone 2 to Touchstone 1. He said the check was found in reference normalization, which was called when [Reference] was present, but could never have been called before Mike's code fix. Mike was not completely sure how and when normalization worked. He felt it only made sense when converting a Touchstone 1 file to Touchstone 2, but it was being called in the other direction. Bob Ross was not sure if normalization was just for displaying the data, or if it was part of the internal data representation. Mike said it was called only when writing output files, not as part of parsing. Mike asked if [Interconnect Model] calls to Touchstone files were valid only for S-parameters, or if other formats would apply. Randy Wolff said we would not be looking at the S or other elements inside IBIS-ISS files. Bob said the Touchstone files could have other than S-parameters for IBIS-ISS. Randy said only the port count check could be done then. Mike noted we would need an IBISCHK bug report calling for Touchstone file interface checking. AR: Bob Ross to write IBISCHK bug for Touchstone checking Tabled topics (no discussion without motion): - BIRD181.2 Randy noted there was no movement on this, and it presumably would not be included in the next IBIS specification. Randy Wolff moved to adjourn. Lance Wang seconded. Without objection the meeting ended. Meeting ended: 11:54 ET Next meeting February 23, 2021