SV-EC Meeting Minutes 11 November 2002 Active (At one or more of last 4 meetins) (a-aaaa-aaa-) v Cliff Cummings (Sunburst) (aaaaa-aaaa-) v David Smith (Synopsys) (-aaa-a-aaaa) v Dennis Brophy (ModelTech) (---aaa-a-aa) v Francoise Martinolle (Cadence) (aaaaaaaaaaa) v Karen Pieper (Synopsys) (aaa-aaaaaaa) v Kevin Cameron (National) (---a-aa----) n Kurt Takara (0-in) (--a--a-a---) n Michael McNamara (Verisity) (aaaapaaaaaa) v Mehdi Mohtashemi (Synopsys) (-aaaaaaaa-a) v Neil Korpusik (Sun) (-aaaaaaa-aa) v Stefen Boyd (Boyd Technology, Inc.) (-aaaaa--a--) n Tom Fitzpatrick (Co-Design) (-----a-a-a-) n Tim Corcoran (WHDL) (---------aa) n Don Mills (LCDM Engineering) (-----aa-aaa) v Brad Pierce (Synopsys) (--------aa-) n Stu Sutherland (-----a----a) n Arturo Salz (Synopsys) Inactive (Missed last 4 meetings) (--a--------) n Adam Krolnik (LSI Logic) (---a-a-----) n Alec Stanculescu (Fintronic) (-aaa-------) n Dave Kelf (Co-Design) (aa-aa------) n Heath Chambers (HMC) (aaa--------) n Paul Graham (Cadence) (aapaa------) n Peter Flake (Co-Design) (a----------) n Roy Armoni (Intel) (aapa-a-----) n Simon Davidmann (Co-Design) (aa---a-----) n Steven Sharp (Cadence) (-----a-----) n Stephen Meier (Synopsys) (-----a-----) n Zeev Kirshenbaum (Verisity) a => Attended p => Attended by proxy - => Missed v => Eligible to vote at next meeting (based on attendance only) n => Not eligible to vote at next meeting (based on attendance only) Action Items: 1) Mehdi to see if a meeting can be setup right after the BC all day meeting which ends at 3:00, to discuss the REV-17, along with channels, mailboxes. [Will discuss with Chair and send out email as soon as possible]. Otherwise, setup a meeting with Kevin to discuss the mbox/ channels in more detail. 2) ALL: reveiw all proposals on the reflector, channel,references, etc. to be prepared. 3) Mehdi: find out when the reconciled testbench donation can be placed on the web and place it. Mehdi Mohtashemi will take notes. 1. Review minutes from the October 28 2002 meeting and approve Mehdi M. Proposed acceptance Kevin second passed / Neil K. abstain [since was not at the meeting] 2. Review Action items ( Kevin had outstanding item) 2a) Web site is up and running, no one made any claims otherwise. 2b) Kevin: Update proposal for Alias with committee feedback. The latest update was October 31, 2002. Not much major changed, module enhanced. 3. Review Alias proposal. 3a) The examples in the notes that were illegal were modified and rolled in. Added alias can be anywhere as module instantiated. undeclared nets are as scalar and example multiple aliases. 3b) **** Vote to Accept the Alias proposal *** Mehdi: Put a vote for acceptance in the october 31st proposal posted to the website sv-ec. Any second for the proposal: Kevin Passed / one abstain: Francoise 4. Review Open Issues from testbench review. Start with REV-16. REV-16 - $sync/$trigger Kevin: just a cleanup, suggested adding the to the event operation. Mehdi: We are reviewing it further making adjustment to the definition, possibly removing some aspects and features. Kevin: will get to review it with the doc and LRM when presented REV-17 - alloc Kevin: This was suggested as to make the declaration permanent, and make the mailboxes synthesizable. Neil: Do you envision this [mailbox] as a class. Kevin: It is dynamic thing right now in Vera, it would be better to do static so that we can have synthesizable type. Therefore also suggested an alternative syntax. Mehdi: Also, Channels are different than Mailboxes, Kevin: doing same type of things, mailbox higher level, channel lower level. Neil: if similar to class, then why not method such as a class method. Kevin: blocking/non-blocking assign, some symmetry with other proposals, with sempahore/mailboxes. Arturo: the latest changes, we are looking at removing the alloc function, using new, get, put and retrieval. Kevin: if rewritten then can roll it with mailboxes, and setting any arbitrary data type. Arturo: but no data checking on it Kevin: looking at references, it has arbitrary data type and you have to use it and then place it to channel. Arturo: dynamic, also looking at parametrization of classes at some point. Kevin: discussion on references, why can we not mailbox new. in a module need to use it. Synthesizable channels are the goal., and not put any distinction. Arturo: Run time type-checking how would you want to make it synthesizable. Kevin: you should be able to combine the proposals together. Arturo: you can not have it both dynamic and static, no references in the donation. Kevin: maybe easier to bring the references and make this happen. Arturo: placing this on the stack, some memory allocation, declaring on the stack. Arturo: removing the funny keywords, making it more object method usage. Kevin: just for reference, the reference proposal went out on sep27th, more for c interfacing. can we use that to clean the syntax. Kevin: large refrences proposal, is to use some portion of it for now, handle are sub-optimal, you should be able to say what the data type is. reference is not a pointer. Arturo: you allow assigning to a reference, aliasing problem for compiler, optimization problems. the var is the dnoated for references. too many issues with full references. Kevin: With the full proposal, is what we want eventually. Mehdi: lets get back to REV-17 and discuss the the alloc and the methods on it. we are using new, and methods, put and get, part of access methods. Arturo: the discussion on the mailboxes what would happen if you want to pass it to a function as a parameter, and the assignemnet that Kevin is proposing. Kevin: rolling the channel proposal with mailboxes is needed, Mehdi: we need discussion on this, synthesizablity, rehashing the issues. Also look at methods vs. operators on the functionality of mailboxes and semaphores. Kevin: Reference proposal with respect to var. Mehdi: will re-create a new item for the above combined discussion. Mehdi: I will send out a discussion time with Kevin, locally discuss the channel/mailboxes/issues. making an action item to look at REV-11 and REV-10 as well. Neil: REV-11 related to mailbox impelementation is identified as closed. REV-18 clocking-domain Mehdi: clocking-domain name, and change of syntax Kevin: no issue with domain name. use plain name for the clock, you can not de-reference the name, Mehdi: the CLOCK syntax has been changed from the clocking domain, will mark closed. REV-19:REV24: have been marked closed. REV-25 using system tasks Kevin: you can use operators, rather than using the system tasks. Mehdi: some of this will be part of discussion on references and handles, mailboxes, etc. 5. Review LRM status and progress Mehdi: Had discussionw with Stu and Arturo and David, getting timeline from Stu on merging the testbench donation within the sv LRM. Mehdi: change bars will be in place 3.0 and 3.1 to show what is added and modified. Francois: currently testbench donation and aliase? Mehdi: At this point those two are slated to be in LRM, the 3.1, the testbench donation reconciliation and the aliases which was voted on. Francois: Is document on the web? Mehdi: Waiting on getting a timeline and then placed on the website. I will find out and send out info. 6. On the discussion with channels Kevin: anyone has any reviews please send it to the committe. Francois: can we do the meeting after 3:00 pm for the basic committe meeting. Action item for Mehdi to see if this can be done and send email. --------------------------------------------------------------------