Minutes of the August 4, 2003 SV-BC Meeting (Design Modeling Committee) --a Tom Kiley - Mentor aaa Matt Maidment - Intel aaa Brad Pierce - Synopsis a-a Karen Pieper - Synopsis -aa Johny Srouji - Intel aaa Dan Jacobi - Intel aaa Dave Rich - Synopsys aaa Francoise Martinolle- Cadence a-a Jay Lawrence - Cadence aa- Dennis Brophy - Mentor -a- Vassilios Gerousis - Infineon a-- Cliff Cummings - Sunburst Design a-- Mark Hartoog - Synopsys a-- Don Mills - LCDM Engineering Meeting Minutes: ================ Review of our meeting minutes: ------------------------------ Karen was unclear as to which issue (following #7 from the minutes that the following comments applied (from the last minutes:): * Item #7: Dave is not the owner. postponed * a verbage in V2k to take care of this. Dave moves that we close this item. Brad seconds. No oppose. Passes This was in section 12.4 of 1364-2001 * the closure of this item is pending a clarification w/ Stuwart Karen moves that we modify "a verbage in V2k to take care of this" by prepending Item #11. She also proposes that we modify "the closure of this item is pending a clarification w/ Stuwart" by prepending Item #12. We also need to change "Stuwart" to "Stu". Karen moves that we accept the minutes from the last meeting with the above mentioned modifications. Dennis seconds. No opposed. No abstain. Passes. Review of open Action Items: ---------------------------- ==> Johny will submit a modified version of the operating guidelines to sv-bc Done. ==> Everyone to read the latest version of our operating guidelines before our next tele-call on August 4th, as we will vote on it then. Done. ==> Dave Rich to find an owner for a donation on issue #4: "VCD dumping for all types" SV-CC will take this one as an api issue. ==> Brad to contact Stu regarding issue #12 (whether it needs his approval) Stu did not respond. Brad will ping him one more time. If there is no response to that mail, we will close the issue. ==> Dan to check on syntax issues wrt port declaration syntax (issue #18) Done. A proposal now exists in SV-BC-18 and SV-BC-46. ==> Dave will include further clarification issues that are required for issue #8 in a list and get them clarified, after which we will vote on this proposal. No action this time. ==> Vote on the "longest prefix issue" by next tele-call, which is scheduled for Monday August 4th See below. ==> Dan to contact Stephen Boyd on why dpi_import was dropped When this was first introduced into the draft, the keyword was not in the donation as far as we can remember. Francoise indicated that it was in some of the drafts. If the SV-CC wants this, they can vote on it and add it. This should close this issue. ==> Karen: include the old issues list on the WEB Done except for the two issues we numbered today. Vote on Operating Guidelines ---------------------------- They were attached to Johny's announcement for this meeting. Cliff: "All proposals must be based upon a prototype." Cliff is concerned that this language will limit fixing issues found in use of the tool. We agreed that fixing errata and usage issues was not covered by the prototype language. Dennis moves we accept the operating guidelines. Cliff seconds. No objections. Francoise abstains. Passes. Open Issues List ---------------- Vote on SV-BC-47 -- Static prefix proposal Cliff was interested in Matt's opinion. Matt indicated that he thinks its ok, but that the corner cases are hard. Dave indicated that this was the same as VHDL. Francoise asked what 1364 does. Matt indicated that it hasn't been resolved there. Cliff indicated that he would fix it here and then take it to 1364. Dave points that this covers more because it includes aggregate expressions. Francoise can an always be sensitive to a subarray because of limits on the indexing? Karen says no because static prefix is a function of the variable reference not on limits to its index. Dave moves we accept this proposal. Dennis seconds. Francoise opposes. No abstain. Passes. Open Action Items ----------------- Item #6: Karen moves that we close this issue. Cliff seconds. No opposed. No abstain. Passes. Item #18: Dan moves that we accept this proposal. Brad seconds. No opposed. No abstain. Passes. Item #46: Dan moves that we accept this proposal. Brad seconds. No opposed. No abstain. Passes. Item #20: Dan moves that we close this issue. Brad seconds. No opposed. No abstain. Passes. Item #30: Brad moves we close this. Dave seconds. No opposed. No abstain. Passes. Item #40: Dave moves that we accept the proposal. Cliff seconds. No opposed. No abstain. Passes. Item #43: Karen moves that we close the issues. Dan seconds. No opposed. No abstain. Passes. Item #50: Dan moves that we accept the proposal. Brad seconds. No opposed. No abstain. Passes. New issues discussion: ---------------------- Synopsys has commited to making proposals for: 16, 48, 49, 51, 52, 53 Cliff wants to add an issue for whether or not ref ports should be the default for variables in an interface. He also commits to make a proposal. Matt indicated that there are a few places where attributes are missing: Modport ports, for example. Assertions as well. Issue discussion: Item #4: Wait for CC to accept to close the issue. Karen will move it to proposal state. Item #7: A proposal exists in issue #29. In reviewing it, we decided that the result of the comparison should be x. For case statements if either are out of range, then the comparison will fail. Cliff indicates that we need examples for the corner cases. 2) Cliff needs an example. We commented that once a bad value is in an enum, simple assignments can propagate it. 3) Karen will take the action item to discuss who owns this issue with David Smith. Karen will move the proposal to issue 7. Item #14: Jay and Arturo think you cannot typedef hierarchical references to types. We are looking at introducing them through namespaces. Dave moves that we close the issue. Matt is concerned that the standard doesn't clearly state that. We need a short paragraph that says that these are not allowed to make it clear. Item #19: Interfaces in modules aren't allowed because they would only be useful among sub-modules of the module. That is more compilication than is necessary. Brad moves that we close this one. Dave seconds. No opposed. No abstain. Passes. Item #32: What is the difference between 'unsigned 'signed and $unsigned and $signed. ' is an operator that takes type or type modifiers (signed, unsigned). Dave moves that we close the issue. Brad seconds. No opposed. No abstain. Passes. Item #35: Straw pole: If Cliff proposes to move logic to ulogic and bit to ubit will you support this? Dave: No. Karen: No. Cliff: Yes. Francoise: No. Mark: No. Brad: No. Jay: No. Cliff would like to delay discussion on this until he can strengthen his argument. Open Action Items for the Next Meeting -------------------------------------- ==> Karen to confirm that SV-CC will take on issue #4: "VCD dumping for all types" SV-CC will take this one as an api issue. ==> Brad to contact Stu regarding issue #12 (whether it needs his approval) Stu did not respond. Brad will ping him one more time. If there is no response to that mail, we will close the issue. ==> Dave will include further clarification issues that are required for issue #8 in a list and get them clarified, after which we will vote on this proposal. ==> Dan to contact Stephen Boyd on why dpi_import was dropped When this was first introduced into the draft, the keyword was not in the donation as far as we can remember. Francoise indicated that it was in some of the drafts. If the SV-CC wants this, they can vote on it and add it. This should close this issue. ==> Karen: update the web to reflect today's changes. ==> Karen will discuss with David who should own issue #7. Our next meeting is 8/18/03 at 9am Pacific.