-- --------------------------------------------------------------------- Neil Korpusik Tel: 408-276-6385 Frontend Technologies (FTAP) Fax: 408-276-5092 Sun Microsystems email: neil.korpusik@sun.com --------------------------------------------------------------------- -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. Results of the last email vote: 1789 - passed 11 yes (Cliff's vote was on the earlier proposal) 0 no 0 abstain 5 didn't vote 1857 - failed 2 yes 7 no 2 abstain 5 didn't vote 1371 - failed 1 yes 8 no 2 abstain 5 didn't vote Please note that the operating guidelines are: - One (1) week to respond (7pm PST June 22, 2007) - An issue passes if there are zero ** NO ** votes and at least half of the eligible voters respond with a YES vote. - Any NO vote must be accompanied by a reason. This issue will then be up for discussion at the next conference call. 1789 ___ Yes ___ No - new proposal was posted during the vote (f -old one) Y Arturo Salz Yf Cliff Cummings - Dave Rich Y Francoise Martinolle Typo in the last row: Str.nethod (...) ^ Y Neil Korpusik - Ray Ryan Y Gordon Vreugdenhil Y Steven Sharp Y Stu Sutherland - Heath Chambers - Don Mills Y Jonathan Bromley If possible, I'd like to squeeze in a friendly amendment to replace the phrase "assigned to an integral type" (or similar) with something like "assigned to a variable of integral type". Such wording appears in at least six places. The proposal is reasonably clear without this amendment, but the confusion of "type" and "variable" is unfortunate. [From Geoffrey] I'm open to the friendly amendment, but I'm not sure it's correct to say "variable" because I believe one could also use string literals as default values for integer type parameters, along the lines of parameter [31:0] input = "AAAA"; The current text has this sentence: A string literal can be assigned to a string type or an integral type. and I tried to stick with the "convention." [From Shalom] The most correct term might be 'data object'. Y Mark Hartoog Y Mike Mintz Y Geoffrey Coram <--- the author - David Scott 1857 ___ Yes ___ No N Arturo Salz This should be discussed in the committee. N Cliff Cummings I believe I favor the proposal but I am not completely following the second example and the issue is complex enough that I would like Gord to personally explain the issue and how the second example works in our regular EC conference call on 6/25. - Dave Rich Y Francoise Martinolle N Neil Korpusik There were a lot of differences of opinions being expressed through the email discussions of this item on the reflector. - Ray Ryan Y Gordon Vreugdenhil A Steven Sharp N Stu Sutherland I vote no 1857 and 1371 because the e-mail traffic and voting on these items indicates that there are still concerns to be resolved. - Heath Chambers - Don Mills A Jonathan Bromley I am in favour of the proposal, but I lack the expertise in compiler theory to be 100% sure that it doesn't have any unpleasant side effects. N Mark Hartoog This proposal needs more discussion. I think there are better ways of doing this N Mike Mintz We need to discuss this some more. N Geoffrey Coram I agree with Cliff that the material for 8.24 needs to be explained. - David Scott 1371 ___ Yes ___ No N Arturo Salz The verbiage needs work. There's some under-specification. N Cliff Cummings I agree with Jonathan about the extra sentence fragment, but I also believe the addition to the first paragraph is incomplete. Since the first paragraph is explicit about what happens if there is at least one initial block in at least one program, it should also explain what happens when there are not initial blocks in any programs. I believe when there are one or more programs with no initial blocks, such as when the programs might only contain declarations, task and function definitions (for whatever reason), then the programs do not call explicit $exit commands. If this is true, it should be properly worded and added to the first paragraph of the proposal. Also, the proposal is font inconsistent, especially in the second paragraph of the proposal (friendly amendment update). Descendent is spelled three different ways in two paragraphs, including "decedent" the legal term for a dead person :-) Are "descendent threads" described elsewhere in the LRM? Also minor typo: "If there is least one" should be "If there is at least one ..." The proposal still is not explicit about what happens when there are NO initial blocks in any of the programs. I assume something like: If there are no initial blocks in any programs, the simulation shall not execute an implicit $exit command. - Dave Rich A Francoise Martinolle (originally voted yes, but then changed it to No) No reason for the no vote --> doesn't count as a no vote. Y Neil Korpusik - Ray Ryan N Gordon Vreugdenhil Agree with others that there needs to be a bit more fine tuning. A Steven Sharp N Stu Sutherland I vote no 1857 and 1371 because the e-mail traffic and voting on these items indicates that there are still concerns to be resolved. - Heath Chambers - Don Mills N Jonathan Bromley My vote becomes Yes if the spurious incomplete sentence When all initial constructs in a program have completed is removed from the middle of the last paragraph of the proposal. I hope this can be regarded as a friendly amendment. N Mark Hartoog Proposal needs more work to cover missing cases and improve wording as others have indicated. N Mike Mintz We need to discuss this some more. N Geoffrey Coram The fragment needs to be removed; Cliff is right that the text should say something about what happens when there are no initial blocks (what about only final blocks?). - David ScottReceived on Sat Jun 23 15:41:26 2007
This archive was generated by hypermail 2.1.8 : Sat Jun 23 2007 - 15:41:41 PDT