Note that there were a few action items from this meeting. Neil -- This message has been scanned for viruses and dangerous content by MailScanner, and is believed to be clean. Champions meeting minutes of October 25, 2007 7am - 9am PST Attendees: ---------- * Shalom * Brad * Dave * Neil * Surrendra * Francoise * John Stu 1. Review IEEE patent policy Motion: Dave - assume we read it Second: John http://standards.ieee.org/board/pat/pat-slideset.ppt 2. Review the list of Mantis items in the Resolved state List of Mantis items reviewed before, but failed in an email vote: 1681 SV-AC 2007-10-03 Introduce global clocking http://www.eda-stds.org/svdb/view.php?id=1681 John - 2 no votes in an email vote Dave - couldn't find the svac meeting minutes posted for the 9/24/07 meeting. John - Eric is taking care of the minutes. They go to the reflector From the Sept 25 minutes. - Voice votes on friendly amendments to recently approved proposals. 1681, JM suggested rewording of the sentence in TT's friendly amendment Voice vote: 8y/0n/0a. John - On Sept 13 - the friendly amendment passed. Dave - Thinks it should be a tool option, but will go along with it. Move: Brad - approve the proposal for Mantis 1681 Second: John Abstain: FM - thinks it should be a tool option Dave - same reason 4y, 2a (approved adding Shalom's yes vote later in the meeting) Opposed: None List of Mantis items that have not yet been through the Champions: 1217 SV-BC 2007-09-06 Missing commentary in Table 19-1 Move: Dave - approve the proposal for Mantis 1217 Second: Brad Passed unanimously 1963 SV-BC 2007-09-06 use of 'bit' (a keyword) as generate loop block name Move: Brad - approve the proposal for Mantis 1963 Second: Dave Passed unanimously 1958 SV-BC 2007-09-06 6.8.1: bad xref and mixup between nets and data types Move: Brad - approve the proposal for Mantis 1958 Second: Dave Passed unanimously 1955 SV-BC 2007-09-06 Follow-up to 1257: delete module path restriction that destination may have only one driver Move: Brad - approve the proposal for Mantis 1955 Second: Dave Passed unanimously 1651 SV-BC 2007-09-06 $psprintf Dave - doesn't think 1651 is a duplicate of 1589. In 1589 the function has a different name $psprintf() in 1651 $sformatf() in 1589 Dave - tools have already implemented the form, now a backward compatibility issue. - psprintf has been in wide use for quite some time. - Dave wants the same function using the name already in use - would rather not close this one. - was on vacation when this happened. - psprintf was never in the LRM FM - why was the name changed in 1589? Shalom - to make it consistent with SV naming convention for existing system functions - display, write, etc. is the typical way to name for SV. - consistent with other languates but not verilog (1589) - didn't remember 1589 when filed 1651 Dave - the original intent of filing doesn't matter. - people see sformatf and then ask about psprintf - thinks it makes the committee look foolish, by not having it. Brad - an existing user base from vera uses it. John - why not add it? FM - sformatf already exists - thinks it is a duplicate, same funcionality Move: Dave - reject the svbc proposal to close 1651 as a duplicate because the name used in 1589 is not psprintf Second: Brad Oppose: Francoise - Shalom - the svbc made a concious decision (not verilog like) Passed <--------------------< The next set was approved as a group >---------------> 1941 SV-BC 2007-09-06 10.3.4 omits uwires 1938 SV-BC 2007-09-06 6.5 error and duplication 1665 SV-BC 2007-09-06 sections on aggregate literals and assignment patterns overlap 1485 SV-BC 2007-09-06 1.6 minor errors 1061 SV-BC 2007-09-06 6.5: port connections of signed nets 0910 SV-BC 2007-09-06 Syntax 3-1 should excerpt all literal-related nonterminals John - items in annex A. The ordering of items in the syntax box were switched Usually the items appear in same order in syntax box and annex A - Doesn't object, but thought it was an unintentional issue. Dave - it is a good goal to have the same order Brad - it shouldn't be an issue Shalom - he thought understanding order was more appropiate Brad - would rather change annex A than the syntax box 5-2 - The ordering in the syntax box is more logical AI/John - add a note about the ordering issue. 0979 SV-BC 2007-08-24 difference in local_parameter_declaration between 1364 and 1800 0917 SV-BC 2007-08-23 Assigning same value to two enum names (4.10) John - doesn't think incrementation is a word AI/John - add a bug note on this one. 0916 SV-BC 2007-08-23 "Significant digits" and enumerations -- 4.10 0354 SV-BC 2007-08-23 Add command line switch to control keyword compatibility of input 0331 SV-BC 2007-08-08 Update $display and other file I/O system tasks for aggregate data types AI/John - font problem, note with the editor 1209 SV-BC 2007-08-07 arrays of parameters - clarify 1090 SV-BC 2007-08-07 add `undefineall functionality Move: Dave - approve the proposals for Mantis items 1941 through 1090 in this list. Second: Brad Passed unanimously 1696 SV-CC 2007-09-20 What is vpiRange for a dynamic array, assoc. array, queue? Move: Dave - approve the proposal for Mantis item 1696 Second: Brad Passed unanimously 1865 SV-CC 2007-09-17 Inconsistent field definitions for s_vpi_vecval in svdpi.h and vpi_user.h Move: Dave - approve the proposal for Mantis item 1865 Second: Brad Passed unanimously 1947 SV-CC 2007-09-12 vpiTask, vpiFunction for built-in method calls Move: Francoise - approve the proposal for Mantis item 1947 Second: Dave Passed unanimously 1970 SV-CC 2007-09-12 What is vpiIndexTypespec for an array with a wildcard index type? Move: Dave - approve the proposal for Mantis item 1970 Second: Francoise Passed unanimously 1837 SV-CC 2007-09-12 Wrong outline for net in VPI generate diagram FM - there appears to be an issue reg, with a double arrow, why labelled vpiReg? - Only need a name if there would be different ones leading to it. The only other names are for memories, which is correct. For others use name of targe object. The change to net is correct. Dave - thinks some of the changes are due to backward compat issues Fm - that is why they made the reg array change - maybe it should be logic var Brad - why is it vpiRegArray and not vpiArrayVar? Dave - the minutes had a note about changing it to logic var - that change was not implemented List of issues: -------------- 1. Shalom - should be section 36.75 2. Dave - friendly ammendment mentioned in minutes not made yet 3. Francoise - label issue, on reg (should be taken care of by the friendly) Move: Dave - send 1837 back to the svcc Second: Shalom Passed unanimously 1726 SV-CC 2007-09-12 Clarify meaning of vpiConstantSelect Move: Dave - approve the proposal for Mantis item 1726 Second: Francoise Passed unanimously 1946 SV-CC 2007-09-12 Clarify vpiSize for string vars and other SystemVerilog variable types Shalom - what is meant by a variable array? FM - this returns the number of entries in an array. Shalom - what about a multidimensional array? FM - that isn't specified - can specify an index for each dimension - it isn't clear here. - Doesn't think it is clear anywhere else either. Shalom - the change here is orthogonal to that point. FM - vpisize existed in 1364 when there were MDAs already thinks it needs to remain the same. Move: Francoise - approve the proposal for Mantis item 1946 Second: Dave Passed unanimously 1556 SV-EC 2007-10-02 in-line static variable initialization - require keyword static? Move: Dave - approve the proposal for Mantis item 1556 Second: Shalom Passed unanimously 1980 SV-EC 2007-10-01 make dynamic_array new consistent as new operator John - last paragraph in 7.5.1 Brad - what is the reason for the first sentence? Of course it does. AI/Neil - open a new mantis item for this point (see mantis 2176) Move: Dave - approve the proposal for Mantis item 1980 Second: Brad Passed unanimously 1707 SV-EC 2007-10-01 streaming operator should specify which direction slices are taken from Surrendra - is bitsream a type? Dave - it is a categroy of type Brad - there are bitstream types, The intermediate that you get is not really defined Shalom - 6.24.3 Brad - there are also places where it refers to a bit stream Neil - yes, it refers to an intermediate representation Shalom - annex q has some info on bit stream Brad - you can only see the result of streaming in pack/unpack - similar to casting description where it mentions the concept of an intermediate type. Shalom - clause and sub-clause words - not lrm style. Just give the clause numbers. AI/Neil - note to the editor (in bug notes) Move: Dave - approve the proposal for Mantis item 1707 Second: Brad Abstain: John - not able to spend enough time on it Shalom - didn't review in detail Passed with 2 abstains 1623 SV-EC 2007-09-06 alternate timeunit syntax Move: Shalom - approve the proposal for Mantis item 1623 Second: Francoise Passed unanimously 1119 V-1364 2007-08-07 19.3: Preprocessor macros within strings Brad - There was an argument about backward compatibility. Eventually it was unanimously approved by the svbc. Shalom - Verilogxl was the basis for the original standard Move: Dave - approve the proposal for Mantis item 1119 Second: Shalom Passed unanimously <--------------------< The next set was approved as a group >---------------> 1025 V-1364 2007-09-06 13.1.1: library_cell is defined but not used 1693 V-1364 2007-09-06 $sformat, $swrite syntax 1285 V-1364 2007-09-06 2005/D6, 15.3.3: $fullskew bugs 1203 V-1364 2007-09-06 Add standard way to define functional coverage points no change required 1155 V-1364 2007-09-06 x-pessimism for if statements no change required 1141 V-1364 2007-09-06 Add "bidirectional skew" timing check no change required 1140 V-1364 2007-09-06 SDF should support $timeskew and $fullskew no change required 1118 V-1364 2007-09-06 15.5.1-15.5.4: neg timing checks only? 1031 V-1364 2007-09-06 Move 17.9.3 to Annex 0995 V-1364 2007-09-06 17.2.4.3, 17.2.5: returning EOF 0983 V-1364 2007-08-24 add lists for equality operators 0661 V-1364 2007-08-07 consistent error/warning 1193 V-1364 2006-11-23 Extend index to cover 1364-2005 enhancements 1192 V-1364 2006-11-23 extend index to complete 1364-2001 Move: Shalom - approve the proposals for Mantis items 1025 through 1192 Second: Dave Passed unanimously 3. Additional discussion In the last Champions email vote, 21 passed 13 or so didn't pass (no action on them from WG) 1340 is an example from bc Neil will go through all of them to make sure they are all accounted for. 4. Next Meeting November 8th, 8-10am PSTReceived on Wed Nov 7 18:10:13 2007
This archive was generated by hypermail 2.1.8 : Wed Nov 07 2007 - 18:10:15 PST