SV-EC Ballot Resolution Committee Meeting Thursday April 28th 2005, 11:00am - 1:00pm [Minutes distributed for approval at next (sv-ec) committee meeting] (01111222) Day (41258158) (00000000) Month (44444444) (00000000) Year (55555555) --------- Attendees ---------- (-AAAAAAA) Arturo Salz (AAAAAAAA) Brad Pierce (AA--AAAA) Cliff Cummings (AAAAAAAA) Dave Rich (AAAAAAAA) Francoise Martinolle (A-----AA) Karen Pieper (AAAAAAAA) Mehdi Mohtashemi (AAAAAAAA) Neil Korpusik (AAAAAAAA) Ray Ryan (AAAAAAA-) Steven Sharp (AA-AAA--) Surrendra Dudani (-AAAAAAA) Gordon Vreugdenhil (-----A-A) Dennis Brophy (-------A) Phil Moorby (------A-) Tom Fitzpatrick ** Minutes taken by Neil Korpusik and Mehdi Mohtashemi Agenda: 1) Review the IEEE patent policy ref: http://standards.ieee.org/board/pat/pat-slideset.ppt Move: Cliff Second: Gordon Abstain: Opposed: 2) Review Meeting minutes, April 25th 2005 Minutes are not yet completed, will be completed by next time. http://www.eda.org/sv-ec/Minutes.html Will keep the two meetings (today and 25th separate. LEAVE for next meeting 3) Miscellaneous P1800 schedule times for next month: (from April 19, 2005 meeting): May 2 Monday Champions meeting May 5 Thursday P1800 meeting 8am to 10am May 9 Monday Champions meeting May 11 Wednesday Champions meeting May 19 Thursday P1800 meeting May 24 The draft of the LRM will be available for final review May 27 Comments on the May 24th draft are due to Stu May 31 The Balloting LRM is ready June 2 P1800 meeting to send the balloting draft back to reballot Next meeting times: Wednesday: May 4th, 2005 11-1:00pm 4) Continue the resolution of ballot issues http://www.eda.org/sv/sv-champions/P1800_Committee_Assignments_05_03_29.xls Issue 240 - mantis 681 Gordon isn't available for meetings next week. Dennis: Mentor has withdrawn issue 240 that needs to addressed for ballot consensus was not increasing with further discussions, spec will stay as is, the other one was in consideration of schedule to get to ballot. Specification as is written, sources of negative ballot as well. Johnny will submit it appropriately to the board p1800 Karen: then we should vote to close item 240/mantis 681 because it is withdrawn by original submitter Move: Gordon to close and add a bug note for issue 240/Mantis 681 Second: Dave Abstain: Opposed: Issue 240 (Mantis 681) is withdrawn, Mantis 681 is closed/resolved Issue 7 (Mantis 344) (anonymous program) Brad: do we need any explanation for this Cliff: can we place any mention of the intended note of it Anonymous program allows you to encapsulate code to be used, package and $unit and $root. Brad: use the same syntax Dave: it should be under program and stick it in there. Ray: anonymous program is just another program? Dave: with the restriction, Arturo: we should put it in the packages section. Gordon: not equivalent to an anonymous block. AI:Dave - write up a normative description for anonymous program, in the program section and place a reference in the pacakage. Issue 266: (Mantis item 695) leave for end of meeting or next thursday AI: everyone to look at all items LEAVE vote next wedensday. Issue 199 (Mantis 607) some more items that came across from Charles. LEAVE for next meeting Issue 8 (Mantis 370) (chapter 13.13 seed/RNG) [[ from previous meeting 4/25/05 ]] [[ remove the sentence before the example ]] [[ remove the two sentences... between the two examples. ]] Ray: bug note was added by Bill Paulsen, Arturo: initializing statics, lead to correlated. we should treat initial blocks and always blocks like fork/join, as implicit fork/join Ray: intent here to get deterministics seed, Gordon: problem with language, causes problem for optimizing compilers, if not treated differently from textual one from declaration. Ray: 3 calls to new, same tag, they will behave the same, Arturo: mostly the users would not put in that context. Arturo: each of the case it gets its own random number generation. Ray: the initial blocks are pseudo initial blocks, needed some way to say there is a determinitistic way they get their seed, Arturo: even though no parent thread, the parent initialization thread get its seed AI: Arturo writeup and Ray to create the proposal for next time Issue 188 - Mantis 595 Brad: not happy with the proposal, any expression on input signal, port connection. Dave: like any other continuous assignment, Brad: you could put anything there. Gordon: for output it has to be lval, i am fine with this language. Brad: port expression is more restrictive than left hand side of assigment. do you want to have left hand side for assignment pattern, left hand side you can have 'assignment, you can not have port assignment. there are lval Gordon: do not think that would be an issue. Brad: you have to be carefull, output or input. You can put in a port expression in a module expression, heirarchical identifiers. Minutes says: the actual the other port formal. port connection in an output port, Arturo: the intent was that actual, Brad: the lvalues are subset of expressions, in continuous assignement, inout is not continuous assignement. in all cases, the minutes should say actual, Ray: it would be easier to say it shall be a port actual, Brad: is it same for program or modules. AI: Brad to update the proposal with wordings. LEAVE it for vote next week. Issue 233, Mantis 317 bit-streaming of class Arturo: the only thing was on the cyclical. add at the end of paragraph, if object heirarchy is cyclical in nature, ther result is undefined and it may issue an error. AI: Arturo add that to the proposal. AI: Arturo and Dave to look in detail LEAVE for next meeting. Issue 10 (Mantis 409) (ref arguments,should be taken care of with item 236 Dave: are we talking about formal or actual? Gordon: it's the formal. Francoise: it says argument, which is a formal. Motion: to approve the current proposal in mantis 409 Move: Gordon Second: Arturo Abstain: Opposed: approved as is it is written. Issue 22 514 ( type for mailbox ) mailbox with ref semantics, equivalence type-mismatch (pass ref), Cliff: how is it different than fifo. Dave: how do you check the type. Arturo: it is usefull, it can be done with methods. If mailbox isn't bound to a type you are allowed to have a dynamic mailbox. Move: Francoise to approve proposal for issue 22, Mantis 514 Second: Arturo Abstain: Opposed: Issue 30 ( Mantis 549) chapter 15 scheduling semantics - several items here Arturo: It discusses certain inaccuracies that need to be explained and corrected. We could go sub-section by sub-section and create Mantis item per each sub-section. *** Issue 30 -- section 1.1 of the text: pre-active region missing in section 15.3.1 algorithm Arturo: it is a writable region, there is no going back the propone is still there, Gordon: this is an additonal line after that line. AI: Francoise to create new mantis out of 1.1 Mantis item (707) AI: Arturo to add the proposal into svdb. *** Issue 30 -- section 1.2 of the text: Iterative regions Phil: you should have feedback loop from pre-active. Without the feedbacks there is no reason to have the region if there is no feedback. The events are unordered within each region. Cliff: pre-active there is no going back before it. post-observed region. Phil: made a mistake not feedback add a backward arrow on the regions, PLI region Francoise: observed region feedback loop. can we do that. Cliff: assertions hitting another assertions. expect from the feedback path from observed region. not changing any variables, just triggering something. Neil: I seem to remember it was brought up in sub-group discussion and not did it. Phil: original concept, within a region, event is un-ordered, if not have a feedback loop in that region, no need to have it. Not having a feedback region in every would not work. Arturo: made the bubbles instead of squares, because they were not accessible from within the language, but through pli types. Neil: why was this not taken Mantis item (708) AI: Arturo to add bug-note for proposal. *** issue 30 Section 2 section 2.1 -- Preponed region Phil: pre-poned region has to be pli callable region.at the previous postponed region is where the sampling actually takes place. #1step implies sampling in the postponed region of the previous time slot. The preponed region was mainly for optimizations. Cliff: here it says the documentation says isn't region for sampling block, Phil: theroretical model for samplin is not pre-poned, is in post-poned, between two time slots, post-poned time slot is equivalent to pre-poned of next time slot. Cliff: number of discussions on pre-poned the way it is described. often talk about sampling in pre-poned region and sampling in postpone Arturo: right now do not allow pli in pre-poned region, Francoise: definition of PLI region, Cliff: does any thing happen in pre-poned region. Phil: from language nothing happens, it is optimization of how it is implemented, previous 1 step. Cliff: is there such a thing as #2step? where is it taking place. Phil: there is no practical reason, just left it as a possiblity. Cliff: if someone puts it in the code, resolution and go 2 steps. pre-poned regions is little more than conceptual idea. Phil: i do not see any reasons why not allowing PLI callbacks, not all of them if you want a pli-callback within than time-slots but not change anything there. Arturo: issue 2.1 has two proposals, one to disallow pli in pre-pone regions, and it should. Gordon: in LRM it does say it, but does not specifically say you can not write. Arturo: we can change 'specifically' Mantis item 709 AI: Arturo to place the proposal in 709. proposal: LRM already allows it, pre-poned region allows pli-callbacks to control it is a read-only region Gordon: it should be illegal to write values in those regions. *** issue 30 section 2.2 -- Observed region Phil: there is no need to have a call-back in the observed region. Arturo: we should not just allow pli-call backs in observed region, it is done in the two surrounding pos-regions or we can add another bubble to post-poned region Gordon: the order of the bubble can have effect. I do not know the separation ** of regions will do. Arturo: there are regions that language use, regions that pli, and there are regions with mix-use. Ray: in the current diagram bubbles are pli only, box are some mix-use Francoise: there are call-backs for pre-poned and post-poned, not sure for post-observed. Cliff: i have been working with the reduced diagram. Arturo: just put a bubble there. 2.2 observed region stays for language regions, pli call-backs are not allowed. AI: Neil -- add a proposal to add clarifying text to say that PLI callbacks are not allowed in the observed region. Section 15.3, page 187. Mantis 710 *** issue 30 section 2.3, post-poned region Arturo: change the 'it specifically for'. to provides for, Ray: is there any statement what can not be there. Arutor: $monitor, $strobe - both occur here. << "Provides for PLI callbacks and control points." >> AI: Mantis 711, AI: Arturo - write a proposal, place it in svdb. *** issue 30 secgtion 2.4 - cbAtEndOfSimTime Definition - in which region Gordon - it needs to be in a read-write region. Arturo - there was a Cadence paper at DVCon on this (~3 years ago) - should there be a post-reactive region? Arturo - pre-postponed region is more appropriate than post-reactive. - maybe the best solution is to add a reactive-NBA region Gordon - if part of reactive region there might be a write-write race. Phil - the algorithm was intended to be flexible, and should allow for additional regions. - must decide: read/write versus read-only Need a feedback loop for each. Mantis 712 AI: Arturo - write a proposal, place it in svdb. ** issue 30 section 3.1 Phil: events posted after the active region need to be executed within the active region. Gordon: one program causing an event to another program will cause the second program to wait for the next reactive region to run. Can't have mutual activity between programs without the NBA getting in the way. This is undesired behavior. Phil: there is complication with nba events in the prg. Arturo: the intent, is that. a reactive nba region would make sense, Gordon: any nba's in prog. region, scheduled in reactive nba regions. Arturo: same counterpart active and nba. Francoise: another feedback loop. Arturo: if nba loop, no user code or pli gets executed there.g Phil: does it now open the call for post-reactive, same as pre-reactive. before we had agreement with cadence, reactive region mirror of active model where you always go back to the reactive region, we used the nba region the complication. i do believe reactive nba region is needed. Gordon: issue that was raised 17.5, scheuling thing in a task called within a reactive region. making a reactive as its own active region. Cliff: reactive nba region can feedback to active region. Gordon: Continuous assign case. Arturo: there is a loophole, continuous assign back into the design, perhaps, an atomic drive back after you execute the testbench Gordon: delay the assingment ot the reactive nba region. continuous assign in a program block back to design. Won't these finish before the reactive region completes? Could modify to schedule these in a reactive-NBA. 6) Next meetings A) Hold a non-voting meeting on Frdiay April 29th, 2005 from 1:00pm-3:00pm PST. Neil will be chair, Mehdi and Brad will not be available. B) Next meeting Wednesday May 4, 2005, 11:00am-1:00pm. ======================================================================================== Action items: (4/28/2005): 1) AI(issue 7, Mantis 344): Dave place normative text for anonymous program, program section. and place a reference in the pacakage. 2) AI(issue 266, Mantis 607): ALL review the latest updates from Cliff, comments 3) AI(issue 8, Mantis 370): Arturo create a writeup, colaborate with Ray for the proposal. 4) AI(issue 188, Mantis 595): Brad to complete the proposal. 5) AI(issue 233, Mantis 317): Arturo add that to the proposal, Dave/Arturo to look into this in more detail 6) FOR ISSUE 30 (Mantis 349): sub-divide to sub sections one Mantis each 6A) AI (section 1.1 of text): Francoise to create new mantis out of 1.1 Mantis item (707) AI: Arturo to place the proposal in 707. 6B) AI (section 1.2 of text): Francoise to create new mantis out of 1.2 Mantis item (708) Arturo: to add a bug-note for proposal. 6C) AI (section 2.1 of text): Francoise to create new mantis out of 1.2 Mantis item (709) AI: Arturo to place the proposal in 709. 6D) AI (section 2.2 of text): Francoise to create new mantis out of 2.1 Mantis 710 AI: Neil -- add a proposal to add clarifying text to say that PLI callbacks are not allowed in the observed region. Section 15.3, page 187. 6E) AI (section 2.3 of text): Arturo: change the 'it specifically for'. to ' provides for', Mantis 711. 6F) AI (section 2.4 of text): Arturo - write a proposal, place it in svdb. Mantis 712 =============================================================================== Action items: (4/25/2005): 1) AI(issue 266,): ALL review the latest updates from Cliff, comments 2) AI(issue 8, 370): Ray remove the sentence before the example remove the two sentences... between the two examples. The rng of all forked threads are determinde before any thread starts, AI; Ray to update the text of proposal. for next time vote AI: arturo to look at this more in detail. 3) AI(Issue 233, Mantis 317) Dave/Arturo to look into this in more details 4) AI(issue 10, 409): Brad (remove the last sentence into the bug) 5) AI(issue 22, 514): Francoise to place the proposal into svdb. 6) AI(issue 240, 681): Gordon,Arturo, more clarifications on how the rules would look into the parametrized world explain the object that crosses both domain, derivation restriction. use model crossing the two domains, extending through multiple domain. ======================================================================================== Action items: (4/21/2005): 1) AI(issue 266,): ALL review the latest updates from Cliff, comments 2) AI(issue 240, Mantis 681) Please review and raise issues and/or other proposals before our face-to-face meeting on Monday. 3) AI(issue 2, Mantis 219): Arturo to send write-up to Steve for completion 4) AI(issue 5, Mantis 319): Ray to place the proposal on svdb 5) AI(issue 8, Mantis 370): Ray to place the proposal on svdb 6) AI(issue 188, Mantis 595): Brad to place the proposal on svdb 7) AI(issue 199, Mantis 607): Mehdi to request clarifications from Charles 8) AI(issue 10, Mantis 409): Steven to place proposal to the svdb. =========================================== == Issues not yet resolved/voted on ======= Issue # Mantis # 7 344 233 317 266 695 Notes issues (from Cliff) 8 370 ( chap 13.13 seed/RNG) 188 595 (invalid example clocking block assignment -- proposal) 13 412 23 517 24 520 30 549 31 551 32 553 94 511 ( type assignments) 95 512 ( object copying) 96 516 ( type compatibility description) 97 518 ( queues, arrays, page 51) 98 519 ( empty array 99 521 ( queues, pattern assignment) 100 522 ( queues concatenation) 101 523 ( queue return type) 122 251 (enhancement -- coverage) 123 253 ( coverage) 171 564 (cross program references) 187 594 ( interface access through clocking block) 190 597 (unnamed clocking blocks in bnf) 200 608 (clocking block value resolution) 201 609 (##cycle meaning in 16.14 ===================================================================== ============== SUMMARY of action items and issue items ============ ==== Approved issues ======================= on April 28, 2005: 10 ( Mantis 409 ) [re-approved on 4/28/05] 22 ( Mantis 514 ) [approved on 4/28/05] 240 ( Mantis 681 ) [Withdrawn, closed/resolved approved on 4/28/05] on April 25, 2005; 2 ( Mantis 219 ) [has merit but not feasilbe at this time] 5 ( Mantis 319 ) [approved on 4/25/05] 199 ( Mantis 607 ) [approved on 4/25/05] [Not a BUG - resolved] 10 ( Mantis 409 ) [approved on 4/25/05] Mantis 384 ) [approved on 4/25/05] (duplicate of 514) on April 21, 2005: 36 ( Mantis 270 ) [has merit but not feasilbe at this time] 162 (Mantis 552 ) [Closed as no-bug] Mantis 411 ) [approved on 4/21/05] duplicate on April 18, 2005: 283 ( Mantis 615 and 662 ) [approved on 4/18/05] 1 ( Mantis 93 ) [approved on 4/18/05] 281 ( Mantis 683 ) [approved on 4/18/05] 238 ( Mantis 671 ) [approved on 4/18/05] on April 15, 2005: 235 ( Mantis- 666) [approved on 4/15/05] 236 ( Mantis- 642) [approved on 4/15/05, re-approved 4/15/05] on April 12, 2005: 232 ( Mantis- 652 ) [approved on 4/12/05] (**236 ( Mantis- 642 ) [approved on 4/12/05] **) 189 ( Mantis- 596 ) [approved on 4/12/05] on April 11, 2005: 250 ( Mantis- 636 ) [approved on 4/5/05, re-approved 4/11/05] 252 ( Mantis- 637 ) [approved on 4/5/05, re-approved 4/11/05] 229 ( Mantis- 644 ) On issue 229 (644): sv-ec vote is still valid [4/15/05] sent back to the champions for decision on this. Champions may want to divide this up to two sections - initialization of struct - randomization 230 ( Mantis- 646 ) 163 ( Mantis- 554 ) [one abstain -- Neil, (not read)] 255 ( Mantis- 641 ) [one abstain -- Neil, (not read)] on April 5, 2005: 11 ( Mantis- 410 ) 109 ( Mantis- 537 ) 161 ( Mantis- 550 ) 237 ( Mantis- 616 ) 239 ( Mantis- 617 ) (**250 ( Mantis- 636 ) **) 251 ( Mantis- 643 ) (**252 ( Mantis- 637 ) **) 253 ( Mantis- 649 ) [combined with 254 into one Mantis (649) 254 ( Mantis- 649 ) [2 abstains: Francoise & Steven, no knowledge of coverage] 270 ( Mantis- 638 ) 271 ( Mantis- 639 ) [one abstain: Steven, not read the proposal] 272 ( Mantis- 640 ) 283 ( Mantis- 615 ) [Votes: 4 yes(Brad,Neil,Dave, Surrendra), & Mehdi 3 abstain:(Ray,Francoise, Steven), and 1 No (cliff)] Issues marked not feasible 204 Issues sent to other committees for review: 125 294 (sv-cc) (vpi question) ==========================================================