SV-EC Committee Meeting Monday October 23 2006 11:00am - 1:00pm PST [Minutes distributed for approval at next (sv-ec) committee meeting] (121202) Day (481593) (000011) Month (889900) (000000) Year (666666) --------- Attendees ---------- (-AAAAA) Arturo Salz (--AAA-) Cliff Cummings (AAAAAA) Dave Rich (AA-A-A) Francoise Martinolle (-AAAAA) Mehdi Mohtashemi (AAAAAA) Neil Korpusik (AAAAAA) Ray Ryan (AAAAAA) Gordon Vreugdenhil (AAAAAA) Steven Sharp (--AAAA) Phil Moorby (---AA-) Doug Warmke (AAAAAA) Stu Sutherland (-AAAA-) Heath Chambers (-AAAAA) Don Mills (--AA--) Jonathan Bromley (--A---) Logie Ramachandran (----AA) Melvin Cardoza (-----A) Mark Hartoog ** Minutes taken by Neil Korpusik and Mehdi Mohtashemi ////////////////// October 23rd 2006 ///////////////////////// Agenda: 1. IEEE patent policy ref: http://standards.ieee.org/board/pat/pat-slideset.ppt Move: Arturo - Assume that the patent policy was read Second: Neil Abstain: none Opposed: none passed 2. Review Meeting minutes of October 9th 2006 http://www.eda-stds.org/sv-ec/Minutes/SV-EC_Meeting_October_9_2006_Minutes.txt Move: Stu - Approve the minutes of October 9 2006 Second: Neil Abstain: none Opposed: none passed 3. Action items review Action items: October 9, 2006 ------------------------- AI: name-space resolution: - Review the proposal for 1609. import not allowed for class scopes. [DONE] AI: 594 Doug: change label used in one of the clockvar in the new blue example Doug: upload a pdf for mantis 594 (instead of an html file) [DONE] AI: 1545 Ray - upload a pdf with the strike-outs visible for 1545. AI: face-to-face meeting, Mehdi inform Matt of 11/6 sv-ec face to face meeting [came up in the bc meeting today -- 11/6 will be a non-voting discussion for the svbc] Action items: Previous meetings AI: follow-up to mantis 978 Neil - rewrite unique(), the first 2 sentences appear to be contradictory. - Make sure that it is clear that array elements are returned. - Add some examples for unique(). Neil - Array reduction methods can operate on arrays of non-integral values when a with clause is specified, as long as the type of the expression is an integral value. [Will hopefully be done for the next conference call] 1308 format - Dave will do this one (taking it over from Neil) Gord - re-write section 5.15.4 - only the default argument value makes sense. Dave - what order is used for find and find_index (e.g. traversal is left to right of the bounds, associative is min to max) - 5.15.1 - change last sentence in first paragraph. All - can anyone come up with a realistic example? (last example 5.15.4) - not only where you would need it but also what does it mean. 4. Name Resolution sub-group update ------------------------------------ Gord - forward typdefs Matt has concerns for restricting fwd typedefs to just classes He prefers this capability for all data types. Matt will provide an example for what he wants. Wants fwd typedefs to span arbitarily deep into a compilation unit - Several design units into comp unit - order of compile is unknown Wants a fwd typedef early on but not resolved until late. - still has some compilation unit completion issues with this Gord - arbitrary expression a+b if b not defined in design unit - how much of a scope is $unit? - functions are hierarchical references (not forward ref) - lexical binding into $unit, issues of referring to a function not yet defined (not allowed), but a forward typedef allowed for a module? (would be confussing) - how much of a lexical scope is $unit? could end up with some surprising results. - LHS of continuous assign implicit or not would need to wait until done at end of compilation unit Gord - a lot of this is preliminary work that will then allow us to move forward with a complete proposal. Gord - one item is pending - export - need input from the champions Next champions meeting will be this Thursday at 9am. Not on the list for the champions meeting. AI/Neil & Dave - get this on the champions agenda. 5. Review mantis item list. ---------------------------------------- 890 08-31-06 clarifications in program and clocking blocks (Doug) [for face-to-face meeting] 888 10-20-06 foreach identifiers are too restrictive (doug_warmke) Arturo - would have specified a primary (e.g. a function call) Gord - that is a more general question than a foreach question. Gord - final identifier is the array identifier there is only one loop variable foreach a[i].b[j] // j is the only loop var, a[i] the array identifier participating. i must be constant if a hierarchical reference i is invariant as far as the loop is concerned Ray - original bnf didn't allow super.array_name Arturo - can specify a set of dimensions to loop over. Move: Ray - Approve the proposal for mantis item 888 Second: Arturo Abstain: none Opposed: none passed 1615 10-08-06 can processes spawned by functions execute blocking statements? (related to 1336 bc) Steven - why were always blocks disallowed? Dave - nesting of fork-join (chaining) - root thread needs to be an initial block (concept of root thread not defined) Steven - always block can spawn fork/join - now disallwed? variable initialization blocks Dave - ok to restrict it to initial blocks. - what about automatic variables? want to allow variable declaration initializers for automatic variable - change it to "static variable declaration initializers" Gord - just say "For illegal uses an implementation shall issue an error either at compile time or runtime when an implementation has determined an ..." Arturo - the current proposal disallows that... needs to be fixed Dave - NBA always done right away. Steven - not for intra-assignment form of nba Issues with the example: [Steven] - header of function - ref arg not allowed? - ref arg referenced in a fork/join none (use a class handle instead?) - procedural code change while to forever - take out ';' after @a Dave - if want to call a task from within a function (that spawns a thread) rule in 1364 is being over-ridden for this case... rules about function restrictions Gord - "any procedural code allowed in a task is permitted inside the fork/join_none." Arturo - create a new sentence? Dave - there should be no restrictions on nba... get rid of restrictions on event-triggers and nba used in functions. 1336 Steven - working out issues of what is allowed in the fork/join_none that is allowed in a function. Gord - add one word at front of sentence. However non-constant functions --- 3rd line in this paragraph... Steven - constant function calls.... Mark - 12.3.3 - could add a restriction on constant functions here. Arturo - random seeds, etc. other issues arise if function not called from an initial block. Gord - there seems to be agreement on intent, the only open part is how to express this intent. Stu - function that spawn via fork-join_none may have side-effects Steven - possibly say the following "also not allowed where a function that has side-effects can be used." Gord - wants to avoid having a complete list. AI/Arturo - update the proposal. - review before the f2f Move: Steven - Approve the proposal for mantis item 1615, as amended Second: Stu Abstain: none Opposed: none passed 1605 09-25-06 Clarification of mailbox/semaphore constructor (Arturo) Gord - no conditions were defined where null was actually returned. Move: Arturo - Approve the proposal for mantis item 1605 Second: Stu Abstain: none Opposed: none passed 6. Next meetings: A) Proposal: Face-to-Face meeting November 6th 2006 Hosted by Mentor (San Jose) 9:45am-4pm 11/6/06 At Mentor office ========== ADDITION TO the minutes, not part of regular ========= ========== meeting discussions, informational purposes ========= ========== Action Items updated based on October 23 2006 ==== Action items: October 23, 2006 --------------------------------- AI: Name resolution: Neil & Dave: get this on champion agenda. AI: 1615 Arturo: Update the proposal, review befor face-2-face meeting. AI: for face-2-face meeting: Dave send out meeting invite. Neil: participate in forming an agenda with Mehdi