SV-EC Meeting Minutes 21 July 2003 11:00 am. Monday Voting Members (3/4 or > 75%) (rr) (aa) Arturo Salz (Synopsys) (-a) Brad Pierce (Synopsys) (aa) David Smith (Synopsys) (--) Francoise Martinolle (Cadence) (aa) Jay Lawrence (Cadence) (--) Jeff Freedman (ModelTech) (aa) Michael Burns (Motorola) (-a) Mehdi Mohtashemi (Synopsys) (aa) Neil Korpusik (Sun) Non-Voting Members (attendance based) (--) Chris Spear (Synopsys) (-a) Dennis Brophy (ModelTech) (aa) Dave Rich (Synopsys) Guests (non-voting) (--) Cliff Cummings (IEEE 1364) (--) Stefen Boyd (IEEE 1364) (-a) Stu Sutherland (IEEE 1364) (-a) Kevin Cameron (National) Inactive Members (Missed last 4 meetings) r => Regular meeting x => Extra meeting (Presence counts for attendance, absence does not) a => Attended p => Attended by proxy - => Missed Action Items: [identified with AI (#) in this text, # refers to AI number] This week: AI-5: Arturo: Respond to Cliff's issues - add to web page. AI-6: David: Pursue with chairs how the Separate Compilation committees results will be integrated into the LRM and whether it belongs to a single committee or is it similar to the Scheduling Semantics where it must be presented and approved by all committees. Minutes 7/21/03 taken by David Smith 0. Before the meeting formally started Jay and David held a discussion concerning the email from Cliff Cummings concerning the Scheduling Semantics. We discussed where the response to Cliff should come from. The options discussed where SV-EC, the scheduling semantics group, or ask someone specific to respond. We agreed to ask Arturo or Phil to respond. 1. Review of meeting minutes from 7 July 2003 meeting Motion: Accept the minutes from 7 July 2003 meeting Moved: Michael Second: Arturo Abstain: Brad Against: None Approved 2. Review of action items AI-2: Discussed with Karen about EXT-5 disposition and resolved that there is no implication on SV-EC from the SV-BC action. We will continue to include it. AI-3: Review of operating guidelines complete. AI-4: Received response from Arturo. See below. 3. Vote on the operating guidelines We discussed the issues raised by Neil in his email and resolved all issues. Jay again raised his concern about some of the content of the operating guidelines. Motion: Accept the operating guidelines as ammended on the web site. Moved: Michael (Motorola) Second: Brad (Synopsys) Abstain: Jay (Cadence) Against: None Approved 4. Review and discussion of the proposed list of extensions Jay raised the issue that he needed more detail on some of the items before he was comfortable with prioritizing them. The committee went through each proposed extension, discussed the intent, and then gave priorities based on 1- strong desire, 2 - don't care, 3 - not important. EXT-1: Jay, Arturo - 3 Arturo - describe this issue as adding additional methods to strings for manipulation. Arturo observed that much of this functionality can be provided through the DPI. EXT-2: Mehdi, Neil, Michael, Arturo, Jay - 1 Michael - described the need to be able to have finer control over process execution especially the ability to kill a thread. Jay - raised the question of this was going to be a resubmittal of one of the failed proposals. The response was that we need to focus on the desire for the functionality and wait for any proposal to see the contents. EXT-3: Neil, Mehid, Arturo - 1, Jay - 2 Neil - described the use of virtual ports within Vera and that something lack this is required within SystemVerilog. EXT-4: Arturo, Jay - 2 Arturo - described this item as being both adding a system call to the underlying operating system (such as the "system" call) and other routines related to the new types. EXT-5: Arturo, Michael, Neil - 2, Jay - 3 The problem was described as being related to the increase of the number of attributes for routines making it difficult to remember the explicit order required. EXT-6: Arturo - 2, Jay, Neil - 1 Add suport for finding size of types and in enhance array queries for new types. Jay - Do we allow $bits in constant expression? Dave - Original Superlog donation did. Stu - There is an existing 1364 request to do this. Jay - Good idea but some nasty implementation details. EXT-7: Neil, Mehdi, Arturo - 1, Jay - 2 Arturo - Actions in C API exist but need to be added to language (for example always@property). David - may be done in SV-AC since it is related to assertions. Jay - not necessarily so, where it goes in language belongs in EC. Need to limit what in the C API gets put into the language since some of it is there only for GUI support. Arturo - agreed. EXT-8: Arturo - 2 Arturo - describe functionality. EXT-9: Arturo - 1, Jay, Neil - 2 Jay - Important capablity and may need to coordinate with SV-AC. Difficult problem so come to session co-chaired by me and Arturo at ICCAD. EXT-10: Arturo, Neil - 1, Jay, Michael - 2 Michael - may be able to be handle with vendor tools and may not be required in language. EXT-11: Arturo, Michael - 1, Neil - 2, Jay - 3 EXT-12: Arturo - 1, Neil, Jay, Mehdi - 2 EXT-13: Arturo, Jay, Dennis, Michael, Neil - 1 Jay - Will give an easier mechanism for defining thing. EXT-14: Arturo, Mehdi, Jay, Neil - 1, 5. Start review of Errata ERR-1: Motion: Change Table 11-1 to indicate that there is no garbage collection for chandles. Move: David Second: Mehdi Abstain: none Object: none ERR-3: Motion: Correct Section 12.4.6 to change the word constraints to constrain in "the value of mode constraints the value of len". Move: David Second: Mehdi Abstain: none Object: none ERR-2/ERR-4: Discussed these errata and indicated that they should be converged and handled by Arturo. 6. Next meeting: 4 August 2003 at 11:00am 7. Close of meeting