SV-EC Committee Meeting Tuesday February 4 2008 11:00am - 1:00pm PST [ http://www.eda.org/sv-ec/Minutes/SV-EC_Meeting_February_4_2008_Minutes.txt ] With the new calculations for voting rights below (rounded)... 3/4 rule = 0.75 * 37 = 27.75 Meeting number: ------------------------------------------------ 000000000000000000000000000000000000000 000000000111111111122222222223333333333 123456789012345678901234567890123344567 Meeting Days: ------------------------------------------------ (121202020102020101311202020101212011020) Day (481593604882505956041593606715926307724) (000011111100000000000000000011111111000) Month (889900112211223344456677889900011222112) (000000000000000000000000000000000000000) Year (666666666677777777777777777777777777888) ------ Attendees ---------------------------- (-AAAAAAAAAAAAAAAAA-AAAA-A--AAAAA*AA*AAA) Arturo Salz 32 (--AAA-AAAAAAA-AAAAAAAAAA--A-A-AAA*A*AAA) Cliff Cummings 29 (AAAAAAA-AAAAAAAAAAAAAAAAAAAAAAAAA*A*AAA) Dave Rich 36 (AA-A-AAA-AAAAAAA---AAAAAAAAAAAAAA*A*AAA) Francoise Martinolle 31 (-AAAAAAAAAAAAAAAAAAA-AAAAAAAAA-AA*A*AAA) Mehdi Mohtashemi 34 (AAAAAAAAAAAAAAAAAAAAAA-AAAAAAAAAA*A*AAA) Neil Korpusik 36 (AAAAAAAAAA-AAAAAAAAAAAAAAAAAAAAAA*A*AAA) Ray Ryan 36 (AAAAAAAAAAAA-AAA---AAA-AAAAAAAAAA*A*A-A) Gordon Vreugdenhil 31 (AAAAAA--AAAAA-A--AAAAAAAAA-AAAAAA**AAAA) Steven Sharp 31 (--AAAA-A-------------------------*-*---) Phil Moorby 05 - Non-voting (---AA-AAA-AAAA-AA-A--------------*-*---) Doug Warmke 12 - Non-voting (AAAAAAA---AA-A-AAAAAAA---AAAAAAAA**AA--) Stu Sutherland 27 - Non-voting (-AAAA--AAAA-A-AAAAA-AAAA-AAAAAAAA*A*-AA) Heath Chambers 29 (-AAAAAA-A----AAAAAAAAA--AAAAAAA-A*A*AAA) Don Mills 28 (--AA--A---A-AAA--A-AAAA-A-A--A--A*-*AA-) Jonathan Bromley 18 - (2 of last 3) (--A------------------------------*-*---) Logie Ramachandran 01 - Non-voting (----AAA--------------------------*-*---) Melvin Cardoza 03 - Non-votings (-----A-AAAAAA-AAAAAAAAAAAAAAAAAAA*A*AAA) Mark Hartoog 30 (-------A-------------A-----------*-*---) Satia (from Intel) 02 - Non-voting (--------AAA----------------------*-*---) Rob Slater 03 - Non-votings (-------------A-------------------*-*---) Alex Gran - Mentor 01 - Non-voting (---------------A-AAA-AAAAA--A-AA-*A*A--) Mike Mintz 14 - (2 of last 3) (------------------AAAAAAAAAAAA-A-*-*---) Geoffrey Coram 13 - Non-voting (-------------------AAAAAAAAAA-AAA*A*AAA) David Scott - Mentor 17 - (2 of last 3) (------------------------A--------*-*---) Benjamin Chen - Cisco 01 - Non-votings (---------------------------AAAAAA*A*-AA) Mike Burns - Freescale 09 (2 of last 3) (----------------------------------*A---) Harry King - Cisco 01 - Non-voting (--------------------------------------A) Karen Pieper 01 - non-voting on February/4/2008 [for next meeting] 14 people (other than the chair) currently have voting rights ** Minutes taken by Neil Korpusik and Mehdi Mohtashemi ////////////////// February 4, 2008 ///////////////////////// Agenda: ------- 1. Review IEEE patent policy ------------------------- ref: http://standards.ieee.org/board/pat/pat-slideset.ppt Move: Cliff - Assume that the patent policy was read Second: Heath Abstain: None Opposed: None Passed unanimously 2. Review minutes of previous meetings ------------------------------------------ http://www.eda.org/sv-ec/Minutes/SV-EC_Meeting_January_22_2008_Minutes.txt Move: Heath - approve the meeting minutes of January 22, 2008 Second: Neil Abstain: Gord (was not present) Opposed: None Approved 3. Updates from p1800WG January 31 2008 meeting and Champions discussions -------------------------------------------------- Neil: From Working Group meeting of 1/31/08 - There was agreement to have Stu produce an additional draft. Everything approved up to now will be added to this draft. - SV-EC is allowed to work on Mantis 2229, 2233, 2183 - SV-CC schedule is extended to March 31 SV-AC VPI issues can be addressed until March 31 - SV-CC and SV-AC to evaluate SV-AC impact on SV-CC - Karen will send out a revised schedule - SV-EC 1556, 1645 - both were approved - Sending email to SV-CC - will need to contain appropriate wording. From Champions email vote just finished, 2/4/2008: 7 mantis items, Mantis items 2227, 958, 520, 2003 - passed Mantis item 1447, 1858 - failed Mantis items 2181 - not enough particiaption by Champions to vote on this. The Next champion meeting is on Feb 14 2008. The next p1800 WG meeting will be on Feb 28, 2008. 4. Immediate issues to review and 1/31/2008 sv-ec deadline ----------------------------------------------------------- Mantis items: [p1800WG granted permission to work on the following] 2183 2229 2233 Mantis 2183: Only simple identifiers allowed in solve-before constraint Ray: syntax restricts only simple identifier, useful to place indexed expression, for array. proposal is to allow this. Steve - we need to be careful of the word variable, it came up in the bc - "-- Each expression in an 'expression_list' must simply identify a random variable (ie only simple name, indexed name or selected name)." Ray - section 17.4 is referred to as "random variables" - it would never be a select into a packed struct. Gord - this refers to the same issue, what constitutes a variable? - identified to be randomized by the randomizer. - can't declare something rand which is part of a packed structure Steve - agrees that this area is less likely to be confused, but not clear that it is completely safe. - continuous assigns to part of a variable was the issue in sv-bc. Ray - the LRM is using the term "variable" in several different ways - does it just mean a variable declaration? - it becomes very confusing to what is and isn't a variable. - we could just get rid of the parenthetical comment in the proposal. Neil - not all of the strike-outs are showing up in the proposal for me. (a .pdf file was uploaded which now displays properly) Move: Ray -- Approve the updated proposal for Mantis item 2183 Second: Arturo Abstain: None Opposed: None Passed unanimously Mantis 2229: Clarify summary description for "inactive" random variables. Francoise: we looked at these last time. Neil: We did not have permission at that time, now we have and we are reviewing them. Ray: repeats the sentence in 17.10 for ignored, proposal removes it - " Inactive variables are not randomized by the randomize() method, and their values are treated as state variables by the solver." Gord: It is not necessarily to add random after inactive. Move: Ray -- Approve proposal for Mantis 2229 Second: Gord Abstain: None Opposed: None Passed unanimously Mantis 2233: Allowed types for randc Ray: clarifying what are the allowed types for randc to see. Neil - why not also make that change for rand? Ray - thinks that somewhere rand is already described as being only for 2-state values. Arturo - the proposal seems strange, declaring a 4-state random value and then ignoring x,z. Ray - rand - won't solve for x or z either. Arturo - section 17.3, states the restriction of only allowing 0,1 values "Constraints support only 2-state values". Page 283 - Friendly amendment: use the word shall in the proposal. Ray - suggested getting rid of the second sentence that was being added. - The proposal was updated with these changes during the meeting. Move: Ray -- approve the proposal for Mantis item 2233 Second: Arturo Abstain: None Opposed: None Passed unanimously Mantis 1447: This mantis failed in Champion's email vote Neil: email vote had 2 no votes. if we can do anything to make it easier to pass. Dave - would like one part to be returned to the original text. - the use of the word variable was not being used properly. Mike - made the changes during the meeting. - updated the text in 7.4.2 - returned part of the text to the original text Dave - Mantis 2092 - is related to this 2092. Move: Dave -- approve the proposal for Mantis item 1447, 1447-D4-20080204.pdf Second: Mike B Abstain: None Opposed: None Passed unanimously re-approval. Mantis 1858: local:: This also failed in Champion's email vote. Neil: one no vote, champion prefers to see the committee get a better consensus. This will be on the agenda the next Champions conference call. Mantis 1897: David - has a fairly serious issue with this one. - this would be a balloting issue he believes. Gord - we could request that the Champions take another look at it. David - coverage computation requirement - the language isn't clear in one case. - not implementable in the current case. Arturo - it is just what the default is, not a question of implementable. David - changes to this mantis are not likely to affect other groups. Gord - it is only a question of how to do the calculations Karen - the Working Group is more willing to be flexible. Arturo - the only issue is what the default should be. David - see the text of the already approved Mantis 1897 - there are two particular questions - The default is false (not saved in coverage db) - new type option for merging (the default is false) when false, the type coverage is the weighted average of the instance coverages. - since by default, coverage of instances, by default we are throwing away coverage - so that seems odd, so it can't be correct - could you save coverage without the instances? - we lose the ability to merge across runs. Arturo - how do we merge coverage for dynamically created objects? Doesn't expect to be able to specify this in the LRM. The whole load and save database description isn't as thorough as it could be. David - why specify coverage db information at all? Arturo - a backward compatibility issue at this point. It would have been best to have left it out of the LRM altogether. - the per instance portion is useful for limiting what is saved. - the loading and saving portion is simplistic. There would need to be a lot more sophistication to address David's concerns. David - inclined to just ignore the per instance option. They have other database technology in place today. Tossing the coverage makes no sense to him. "when true, coverage is saved in the coverage database..." If implement 1897 there will be a backward compatibility for Mentor. Gord - thinks the LRM may have gone too far as far as coverage db is concerned FM - where is the backward compatibility issue? David - can't be compatible with 1897 and what they are already doing. Gord - normalize what implementations do. Compatible with stuff that implementations have already been providing. Neil - you want to get rid of all references to save to db? Gord - that is probably too simplistic. - if it would be not possible to implement not useful in LRM. David - restoring a db will most likely be different for all implementations. It isn't clear what it means to restore a db. Arturo - per instance was in the 2005 LRM. So there is a backward compatibility issue. David - there isn't enough information to allow for vendor compatibility. Gord - It might be reasonable to say: these options exist and leave it more nebulous. Leave room for vendor choice. - may not have enough experience as a community on how to do this. David - thinks that he, Gord and Arturo can come up with something. 2242 - David - get_cvg, get_inst_cvg return top and bottom portions of coverage numbers. there was another Mantis recently filed saying there is a problem The top might actually be real - they may need to round them off. Maybe not a real issue - but it is wrong. 2244 - David - probably "out of scope" - how to restore coverage 2243 - David - per instance stuff is in here - minor change possible The SV-EC Wants permission from the Working Group to work on: 1897 2243 - correction to 1897 2242 5. Next meetings in 2008 ---------------------- [Tuesday February 19 2008 ? [President's day holiday on Monday Feb 18 2008] Give a break and hold the next meeting on -- No meeting] Monday March 3 2008