Minutes of the April 04, 2005 SV-BC Ballot Resolution Committee Meeting 00 Day 41 00 Month 44 00 Year 55 aa Matt Maidment - Intel aa Brad Pierce - Synopsys aa Karen Pieper - Synopsys -a Dan Jacobi - Intel aa Dave Rich - Mentor Graphics -a Rishiyur Nikhil - Bluespec aa Cliff Cummings - Sunburst Design aa Steven Sharp - Cadence -a Don Mills - LCDM Engineering aa Logie Ramachandran - Synopsys aa Gord Vreugdenhil - Mentor Graphics a- Francoise Martinolle- Cadence a- Kathy McKinley - Cadence a- Doug Warmke - Mentor Graphics a- Mark Hartoog - Synopsys a- Greg Jaxon - Synopsys a- Stuart Sutherland - Sutherland HDL a- Dennis Brophy - Mentor Graphics -- Tom Fitzpatrick - Mentor Graphics -- Jonathan Bradford - Micronas a- Surrendra Dudani - Synopsys Agenda: +Review IEEE patent policy http://standards.ieee.org/board/pat/pat-slideset.ppt The patent policy was reviewed. + Issue Resolution Karen moves to adopt this updated language for the resolution of issues 205/207: "The committee read and considered this feedback. While it has merit, the committee believes it is either not feasible to implement at this time or not in the scope or goals of the P1800 project as it was defined & agreed upon by the P1800 WG." Brad seconds No opposed. Abstain: Dave Rich (ambivalent) Issue 266 (Cliff) Cliff reviewed some of proposed changes. Initial feedback: In proposed changes to 4.9, change "declarations" to "declaration" In proposed changes to 4.11 change "normal" to "unpacked" Overall, committee agreed with direction. Cliff will continue to elaborate into a complete proposal that committee will vote on April 6. Issue 234 Committee discussed hierarchical references in the context of always_comb. Gord given AI to provide updated proposal based on feedback that hierarchical references should not be completely prohibited from always_comb. Gord's proposal: "Add 2 new clauses in 8.11 (labelled 2&3 and renumber remaining) which read: 2) A hierarchical reference to an object is a static prefix. 3) A package reference to net or variable is a static prefix. Add to 10.2.1: Hierarchical function calls and function calls from packages are analyzed as normal functions. References to class objects and method calls of class objects do not add anything to the sensitivity list of an always_comb." Gord moves to accept the proposal. Brad seconds. No opposed. No abstain. Motion Passes. AI: Matt add issue 234 to SVDB Issue 243 (SVDB 614) Doug moves to accept proposal accompanying issue 614. Gord seconds. No opposed. No abstain. Motion passes. AI: Matt update SVDB 614 to resolved. Issue 244 Gord explained that in the same way that a defparam cannot make an upward reference through a generate or array of instances, an actual of an interface port cannot be a downward hierarchical reference through a generate or array of instances. Committee agreed with direction, but issue needed refined wording. Gord delivered wording to Matt. AI: Matt enter Gord's proposal for 244 into SVDB AI: Matt notify SV-EC about proposal for 244 Issue 278 (SVDB 618) After some discussion on the merits of allowing expressions in array querying functions, it was decided that something more along the lines of hierarchical_identifier_selects would be more appropriate. AI: Brad update proposal for 278(SVDB 618) to use identifier selects 246 AI: Matt send typo to Stu: ebus_i ebus; --> ebus_i ebus(); Committe spent ~1.75 hours discussing the intention of compositional modport expressions as demonstrated by example on page 302. Some hold the opinion that the LRM does not grant this feature. Others hold the opinion that this is the original intent. The committee identified several possible options: - provide more clarification for using modports to define other modports. - eliminate this capability - enable hierachical identifiers in modport expressions - extend modport expressions to tasks, functions and data types The committee agreed to table discussion until April 6 to allow time for further consideration of direction. AI: Matt add issue 246 to April 6 agenda Issue 247 Committee agreed to defer until direction of 246 is set AI: Matt add issue 247 to April 6 agenda pending resolution of 246 Issue 248 Committee agreed to defer until direction of 246 is set AI: Matt add issue 248 to April 6 agenda pending resolution of 246 Issue 203 The committee discussed the feedback for this issue. AI: Dennis to send proposal to resolve issue 203. Issue 202 Issue was discussed briefly. AI: Doug propose solution to first clause of issue 202 AI: Karen ask submitter of issue 202 to provide example demonstrating how reordering bind statements produces different results. Issue 264 AI: Cliff make proposal for issue 264 for April 6 Isse 265 Primary resistance to addressing this issue are: 1. It is a problem all the way back to Verilog-1995 and may introduce backward incompatibility. This is acknowledged by Stu but he feels that does not mean it is without merit. 2. Significant effort given the timeframe. Stu agreed to provide a very specific proposal no later than April 13th for the committee to review and approve. 3. Difficult for all implementations to agree on the appropriate messaging condition. This can only be addressed upon review of a specific proposal. AI: Stu make a specific proposal for Issue 265 no later than April 13 Issue 29 (SVDB 548) Committee discussed current proposal and decided that the language did not clearly convey the intent. AI: Dave to revise proposal for issue 29 for April 6 Issue 226 (SVDB 623) The primary feedback for the proposal is that there need to be clear semantics for assigning each element of the pattern (compatibility rules). AI: Brad to update proposal for issue 226 for April 6 Issue 35 (SVDB 578) Proposal was discussed. General consensus on everything but last two sentences regarding permission of port collapsing of net whose loconns are variables. Striking this may leave performance "on the table" but enabling this may introduce unexpected behavior. AI: Dave talk to Gord about benefits of port collapsing to simulation perf AI: Logie talk to SNPS simulation team about benefits of port collapsing AI: Matt add issue 35 to April 6 agenda Issue 1 (SVDB 93) This feature (not requiring parentheses for calls) was requested by the SV-EC AI: Matt to e-mail Mehdi to transfer issue 1 to the SV-EC Next Meeting: Wednesday, April 6 8:00am-10:00am PDT Action Items Completed 04/01/05 Matt add issue 243 to April 4th agenda 04/01/05 Matt add issue 244 to April 4th agenda 04/01/05 Matt add issue 246 to April 4 agenda 04/01/05 Matt add issue 247 to April 4 agenda 04/01/05 Matt add issue 248 to April 4 agenda 04/01/05 Matt add issue 227 to April 4 agenda 04/01/05 Matt add issue 224/277 to April 4 agenda 04/01/05 Matt add issue 234 to April 4th agenda 04/01/05 Matt add issue 278 to April 4th agenda 04/01/05 Matt bring up Issue 273 for re-vote Monday. 04/01/05 Brad add issue 224/277 to April 4 agenda 04/01/05 Matt Add Issue 245 to SVDB 04/01/05 Matt enter issue 223 into SVDB 04/01/05 Dave add issue 249 to SVDB. 04/01/05 Cliff to review Item 266 and report status at April 4 meeting. 04/01/05 Steven & Brad will review and propose solution for issue 1 at the April 4 meeting 04/01/05 Gord to propose solution to issue 29 04/01/05 Steven to propose solution to issue 35 at the April 4th meeting 04/01/05 Gord bring new wording and clarification of issue 243 to April 4th meeting 04/01/05 Gord to provide example for issue 246 04/01/05 Karen to get more information for issue 285 04/04/05 Matt e-mail Mehdi to transfer issue 1 to the SV-EC 04/04/05 Matt send typo in 20.0 to Stu: ebus_i ebus; --> ebus_i ebus(); Pending 04/01/05 Karen Ask Johny for the purpose of the 'WG Resolution' column 04/01/05 Steven will review and propose solution for issue 6 at the April 4 meeting. 04/01/05 Karen to provide Gord with SVDB account 04/01/05 Brad to assign issue 29 to Gord 04/01/05 Brad to propose solution to issue 91 at the April 4th meeting 04/01/05 Dave to propose solution to issue 216 at the April 4th meeting 04/01/05 Brad enter issue 269 into SVDB 04/04/05 Matt add issue 234 to SVDB 04/04/05 Matt update SVDB 614 to resolved. 04/04/05 Matt enter Gord's proposal for 244 into SVDB 04/04/05 Matt notify SV-EC about proposal for 244 04/04/05 Matt add issue 246 to April 6 agenda 04/04/05 Matt add issue 247 to April 6 agenda pending resolution of 246 04/04/05 Matt add issue 248 to April 6 agenda pending resolution of 246 04/04/05 Doug propose solution to first clause of issue 202 04/04/05 Karen ask submitter of issue 202 to provide example demonstrating how reordering bind statements produces different results. 04/04/05 Cliff make proposal for issue 264 for April 6 04/04/05 Stu make a specific proposal for Issue 265 no later than April 13 04/04/05 Dave to revise proposal for issue 29 for April 6 04/04/05 Brad to update proposal for issue 226 for April 6 04/04/05 Dave talk to Gord about benefits of port collapsing to simulation perf 04/04/05 Logie talk to SNPS simulation team about benefits of port collapsing 04/04/05 Matt add issue 35 to April 6 agenda