Phone Conference P1450.1 Working Doc Subgroup Thurs Jan 3, 10:00 am PST --------------------------------------------- Attendees: ---------- Jason Doege Jim Teisher Peter Wohl Tony Taylor Greg Maston Doug Sprague Agenda ------ Review the p1450.1 spec, draft D13, Dec. 3, available on the dot1 website, with respect to the internal-review-resolutions doc also found on the web, latest dated Dec 26. Brief review of our process; updating resolution document as we review and finalize the decisions, but working the same draft until we complete one pass of the issues. Issues ------ resuming at: JC1: accepted no issue JC2: passing parameters in parameters and macros, particularly into Shift blocks. Quoted signal expressions and mixed %# need to be reviewed. The limitation that macros, procedures, etc. are not allowed in a Shift should be a dot0 clarification. [AI1] to Tony and Greg to review the issues with supporting extended dot1 constructs under the Shift block; [AI11] for Greg to review dot0 Shift statements and determine need for additional clarification (or if not, identify where it is clearly stated). DS1: general concern; Working Group will look for simplification opportunities as they continue refining the dot1 constructs. DS2: concern about ambiguity of handling of MergedScan with procedures that have multiple Shift blocks defined. [AI2] Tony will add elaboration to If-Else to allow Shift blocks in separate If-Else clauses to contain the same set of signals; see Annex 0 for an example of how this happens. [AI3] Greg will distribute current clarification doc for dot0 for review next time. DS3: add Constants to the references cited. DS4: [AI4] Greg will take last paragraph of section 6.1 and rewite and send to Tony. DS5: agreed. DS6: inconsistent reference to commas or space-separated integer lists. Tony will clarify that integer lists are NOT space-separated; require an operator (note spaces are allowed but not between two integers without an operator (either comma or dot-dot) present). DS7: requested clarification to all the types of expressions. [AI5] to Doug to provide something. DS8: agreed. DS9: question about whether the value of variables is "sticky" or context/scope sensitive... Working Group agreed that assignment of values to variables is done during execution and variables retain the last assignment irrespective of location (ie variable assignment is to a single global definition, that is, "sticky", and that there are no local values under a procedure call). Variable value maintains the execution path of variable assignment and retains the last assigned value always. [AI6] to Tony and Greg to determine where to put this clarification. DS10: Conflict between elaboration of Fixed and the syntax that provides the function. Tony will review and add to 12.1 syntax. DS11: spelling corrected (or will be). DS12: grammer will be fixed. DS13: corrected. DS14: ambiguity of X statements location inside procedures/macros during pattern execution; need additional data to provide sufficient location. Doug wrote an email on this topic; will include email from Doug and [AI12] Group will discuss the resolution of this issue next time. DS15: inconsistent reference to file_type and file_format - fields that are not defined at this moment. Need to resolve how we want to handle these fields. [AI7] Greg will distribute the current recommendation doc for review by this group and decide how we want to formalize this information. DS16: change made. DS16-2: question about the need for the statement "extend c", since the next execution starts immediately at the end of pattern c... Tony identified the purpose of the extend statement is to define behavior of the signals in 'c' when 'c' has completed. Doug then asked what would happen if there was no 'extend' statement present. Tony indicated that there might be a difference in behavior with ending pattern 'c' with a V statement or a BreakPoint... Need the Extend statement to be present if a BreakPoint is NOT present. Greg asked if the grey areas of fig. 8 could have arrows or something to indicate that signal values are provided, and from what statements caused those values to be present. DS17: [AI8] Tony ship specified Annex to Doug for corrections. Doug asked about scan-hierarchy constructs and moving constructs to dot6. This was part of DM1. Doug identified a desire to support hierarchical scan definitions as part of dot1, rather than requiring dot6 to provide this mechanism. Tony identified this issue as not closed yet because we have not reviewed the dot6 hierarchy constructs. [AI9] Tony will provide access to this group of the current dot6 in order for this group to close on DM1; [AI10] is defined to make sure Working Group resolves the scan-hierarchy support questions. Next Steps ---------- Next meeting in two weeks, Jan 17, 2002, same time and phone#. Tony will update standards document and distribute, and update the resolution document with decisions made in this meeting. Next meeting to close on remaining open issues (note [AI10], [AI12] in particular). AIs --- [AI1] Tony and Greg review JC2 issues and clarify requirements. [AI2] Tony to add claritication about Shift signal-sets under If-Else [AI3] Greg to distribute current working clarification doc for dot0. [AI4] Greg to rewite last paragraph of section 6.1 [AI5] Doug to provide expression-type-clarification diagram or text. [AI6] Tony and Greg to clarify behavior of variable assignment. [AI7] Greg to distribute current dot1 recommendation doc. [AI8] Tony ship specified Annex to Doug for editing. [AI9] Tony to provide access to dot-6 (repeat AI from prior meeting) [AI10] Group close on DM1 and what to do with scan hierarchy constructs. [AI11] Greg review dot0 Shift statements and identify need for more clarif. [AI12] Group to review Doug's proposal on X-stmnt-location and resolve.