[sv-cc] SV-CC Meeting minutes for 04/15/2005

From: Charles Dawson <chas_at_.....>
Date: Fri Apr 15 2005 - 12:52:39 PDT
Minutes of 04/15/2005 SV-CC Meeting.

ATTENDEES
0000000000000000000000000000
5555555555544444444444444444
0000000000011111111111110000
4444222111122222111100009999
1100200211021100211022102211
5185392692526581470360369250
xxxxxxxxxxxx--xxxxxxxxxxxxxx Charles Dawson
xxx-xxxxxxxxxxx-xxxxxxxxxxxx Francoise Martinolle
xxxx-xxxxxxxxxxxxxxxxxxxxxxx Doug Warmke
xxx-xxxxxxx-xxxxxxxxxxxxxxx- Ralph Duncan
xxxxxxx-xxxx-x-xxxxxxxxx-x-- Sachchidananda Patel
xxxxxxxxxxxx-xxxxxxxxxxxxx-- Michael Rohleder
xxxxxx-xx-xxxxxxxxxxxxxxx-x- Jim Garnett
xxxxxxxxxxxxxxxxxxxxxxxx--x- Steven Dovich
xxxxxxxxxxx--xxxxxxxxxxx-xxx Jim Vellenga
xxxx------------------------ Rohit Rana
-xx----xx--x-xxxxxxxxxxx-xxx Bassam Tabbara
xxx---x--x-xx-xxx----------- Stu Sutherland
-xx------------------------- Tapati Basu
-x-------------------------- Ashish Naik
------xxxx---xxx-xxxxxxxxxxx Andrzej Litwiniuk
-----------x-xx--xxxxx-xxx-- John Stickley
-------xxxxx---x-xxxx-x-xxxx Ghassan Khoory
--------x--xxxx--xxxxxxxxxxx Joao Geada
-------------x-x-xxxxxx--x-- Rob Slater
---------------------------x Swapnajit Mitra
---------------------------x Karen Pieper
-------------------------x-- Angshuman Saha
---------------------x------ Kevin Cameron
-------------------x-------- Amit Kohli
------------------x--------- Surrendra Dudani

1.  Reviewed Patent information.

   - Charles Dawson reviewed the patent information.


2.  Reviewed minutes of the 04/08/2005 Meeting.

   - JimV/Steve ACCEPTED


3.  Liaisons

   - Nothing else reported.
   - The P1800 meeting is to meet on Tuesday.  Some of the language
     constructs that have changed will require (structure member type
     initialization) VPI enhancements. Reported by Francoise.
   - Chas reported on the status he gave Johny after our last meeting.
   - Chas reported that he asked Johny for some period of time after the other
     committees have completed their work so that the SV-CC can react to their
     changes.  Chas will make sure this ends up on the P1800 agenda.


4.  New business

   - var bit vs net bit (email from Francoise)?
     Not and issue.  Can be dropped from the agenda for next time.

   - Item 617 and the related SV-CC Item 668
     Francoise/JimV move to pass 668.  PASSED (unanimous)

   - Item 610 and the definition of vpiFullName
     Francoise filed a SV-BC item on %m (Item 672)
     Chas to add mantis item for vpiFullName to have a better definition in
     the next version of the spec.
     Stu made friendly amendment to change "will have" to "has".
     Stu made friendly amendment to get rid of "But".
     Change wording in note 7 to use "shall"
     Francoise/Rohit.  PASSED (unanimous) as amended.

   - We will meet next at our regularly scheduled meeting on Weds 4/20 at
     12pm EDT.  Chas will report on the P1800 meeting.


5.  Reviewed of items with proposals.

   - Item 426
     Postponed due to 425.

   - Item 425
     Postponed again.  Does this routine work on an entire assertion or on
     a step of an assertion or both?  If the former, then this okay.  If one
     of the later two then the routine might be useful.  Further input from
     Bassam needed.

   - Item 433
     Friendly amendment to change "Coverage API and Control" to
     "Coverage Control and API".
     Steve/Stu.  PASSED (unanimous) as amended.

   - Item 434
     Michael/Stu.  PASSED (unanimous)

   - Item 586
     Friendly amendment from Ralph to remove "effectively" and from ";" on.
     Ralph/Michael.  PASSED (unanimous) as amended.

   - Item 585
     Michael/Steve.  PASSED (unanimous)

     Motion to reconsider (steve/Stu).
     Friendly amendment to change "an object vpiHandle" to "the object's handle".
     PASSED (unanimous) as amended.

   - Item 455
     Skipped (Francoise and Chas to still work on it).

   - Item 489
     Proposal does not look complete.  Francoise to look at it again.

   - Item 447
     Values should be masks.  Properties return values need to be taken
     out of the diagram.  Note that they can be or'ed together.
     Chas to rework.

   - Item 526
     Problem is that they can out live the ability to access them in the HDL.
     Francoise's suggestion was that you would need to reacquire handles to
     dynamic objects every time control is returned to VPI.

     Make the property that if it is valid, then return TRUE, or FALSE in
     all other situations (like when it is too hard to figure out).

     Michael thinks that it really needs to be part of the specification
     to prevent non portable applications.  Chas agrees.

     Ralph wanted more time to talk with others who know more.  Postponed.

   - Item 372
     Issue with the preponed phase description.  Steve suggests using VHPI
     terminology.  Chas described ideal solution.  More work here later.
     This proposal is good enough for now.
     JimV/Stu.  PASSED (unanimous)

   - Item 417
     JimV/Sachi.  PASSED (unanimous)

   - Item 606
     Steve/Stu.   PASSED (unanimous)

   - Item 463
     Skipped (chas has an action to fix)

   - Item 345
     Stu had an issue with multiple uses of the phrase "unless otherwise
     specified".  Need to add description of get/put value for function
     return values and function arguments.  What about ports?  If you can get
     the value then you need the size too.  Steve will try to refine further.

   - Item 533
     Sachi made a friendly amendment to change vpiPublic to vpiPublicVis.
     Sachi/Ralph.  PASSED (unanimous)

   - Item 534
     JimV/Rohit. PASSED (Francoise abstains, cannot remember why it was added,
     and Tapati might remember)

   - Item 352
     The issue is in 32.9 now (references in the description were prior to
     Draft 4).  Francoise thinks that the original text is correct.  Francoise
     will write up the explanation and add as a bugnote.

     JimV moves that we close this as not a bug with the explanation that
     Francoise added.  Ralph seconds.  PASSED (unanimous)

   - Item 528
     In 32.16 the iteration to param assign needs to be one to many.
     Section 32.20 should be 32.21.  Section 32.21 should be 32.22.
     Long discussion on the 32.21 diagram.
     Scope in 32.21 should be dotted and italicized.  Don't re-do changes
     in note 3 of 32.22.
     Need parens after vpi_get_value() and vpi_put_value().  These routines
     should put/get handles.
     Francoise will work on it some more.

Meeting ended at 2:00pm EDT.


6.  Old Business


7.  Action items

   SV-CC action items:
   - Francoise to ask Peter Ashenden what was done to improve
     printing from Rational Rose.
   - Francoise to inquire about the feasibility of third parties
     shipping the UML for the diagrams.
   - JimV to enter a new SV-CC item for adding tables for return values of
     the properties.
   - Stu to file new SV-CC item that will address bigger issue he pointed
     out in Item 321.
   - Francoise and Chas to investigate what to do for systemVerilog and generates
   - Chas/Francoise to review/change proposal for Item 455.
   - Chas to redo proposal for Item 463.
   - Francoise to clarify definition of frames for Item 465.
   - Sachi to rework proposal for Item 482.
   - Bassam to work on new proposal for Item 431.
   - Chas to add mantis item for vpiFullName to have a better definition in
     the next version of the spec.
   - Bassam to answer question on Item 425.
   - Francoise to look at proposal for Item 489.
   - Chas to rework proposal for Item 447.
   - Steve to continue work on Item 345.
   - Francoise to continue work on Item 528.

   PTF action items:
   - Steve to compare BNF with the access available
     for attributes to see if they match
   - Stu to enter new PTF item for save/restart/reset issue.
   - Francoise to lookup wording for PTF 524 in VHPI.
   - Francoise will open a new PTF issue to look for situations like 25.6.15,
     where multiple methods are used access the same object enclosure
   - Chas to reword proposal for PTF 525.
   - Francoise to file a PTF item that asks to specify the order that iteration
     occur in, when the order is important.
   - Steve to add ETF item for Annex C to remove the Informative label, but
     still allow the contents to be optional.
   - Chas to file an Item on the usage of the term "tags" at the beginning of
     clause 27.


8.  Items for consideration at the next meeting (they already have proposals):

   - Item 465 32.19 Please clarify vpiWaitingProcesses iteration
   - Item 589 31.10 cannot set the user_data field
   - Item 482 32.27 Don't need vpiBuiltIn property
   - Item 606 32.30 Problems with data model for clocking blocks
   - Item 604 Annex E: routines are not always declared in instantiable scopes
   - Item 425 29.3.2.1 violates a principle design feature of VPI
   - Item 487 32.34 Order of operands should be in 32.39
   - Item 431 29.5 pollutes the VPI name space

--
Charles Dawson
Senior Engineering Manager
NC-Verilog Team
Cadence Design Systems, Inc.
270 Billerica Road
Chelmsford, MA  01824
(978) 262 - 6273
chas@cadence.com
Received on Fri Apr 15 12:52:50 2005

This archive was generated by hypermail 2.1.8 : Fri Apr 15 2005 - 12:53:46 PDT