[sv-ac] SV-AC meeting notes 2006-12-12

All:

Please find the meeting notes attached.  Let me know
if corrections are required.

J.H.


Minutes of IEEE P1800 SV-AC meeting #2006-27: 2006-12-12
Written by: John Havlicek

Code for the conference call:

        Domestic:         888-635-9997
        International:    763-315-6815
        Participant code: 2638198

Attendance Record
	Legend:
		x = attended
		- = missed
		r = represented
		. = not yet a member
		v = valid voter (2 out of last 3)
		n = not valid voter
                t = chair eligible to vote only to make or break a tie

New PAR, attendance re-initialized on 2006-08-22:

nn[-----xxx------] Shalom Bresticker (Intel)
vv[xxxxxxxxxxx-xx] Doron Bustan (Freescale)
vv[xxxxxxxxxxxx-x] Eduard Cerny (Synopsys)	
nv[-x-x-xxx-x---x] Surrendra Dudani (Synopsys)
vv[xxx-xxx-------] Yaniv Fais (Freescale)
nn[-x------x-----] Faisal Haque (Cisco)
tt[xxxxxxxxxxxxxx] John Havlicek (Freescale - provisional Chair)
vv[xxxxxxxxxx-xxx] Dmitry Korchemny (Intel - provisional Co-Chair)
nn[-------xx-xxxx] Manisha Kulshrestha (Mentor Graphics)
nn[-------x-xx-x-] Jiang Long (Mentor Graphics)
vv[-xx-xxxxxxx-x-] Hillel Miller (Freescale)
nn[---xx-xxx-----] Don Mills (Microchip)
vv[xxxx-xxxxxxxxx] Lisa Piper (Cadence)
vv[x-xx----------] Tej Singh (Mentor Graphics)
nn[----xx-x------] Stu Sutherland (Sutherland HDL)
vv[xxxxxxxxxxxxxx] Bassam Tabbara (Synopsys)
   |-------------- attendance on 2006-12-12
 |---------------- voting eligibility on 2006-12-12
|----------------- new voting eligibility

Agenda:
-------

- Review IEEE working group rules
- 1550:  $sampled definition and sampled-value functions
- 1601:  implicit type
- 1549:  Missing formal types
- 1668:  Local var initializers
- 1641:  Severity specification for error msgs (should be discussed together with SV-BC)
- 1667:  Local variable args for seq and prop
- 1591:  $past syntax not precise

Minutes:
--------

- 1550:  $sampled definition and sampled-value functions
  EC:  Not a proposal yet.  Depends on work of other committees.
  TS:  Use in covergroup needs to be clear, say covergroup sensitive
    to a clock and $past or $stable used in coverpoints.
  EC:  Depends on how the covergroup is sampled.  The way that
    strobe = 1 behaves may be changing in other committees.
  JH:  Are there any situations where assertion expressions would
    use the reactive region definition?
  EC:  Maybe this is possible with expect.
  TS:  A single $past could be called from two different regions, e.g.
    if it is in a task that is called both from a module and from a 
    program.
  EC:  Execution context defines the region used for the definition
    of $past.
  DK & JH:  Let's split this proposal into $sampled and the rest.  We
    can make progress on $sampled, but we are dependent for the others.
    EC to enter new mantis item for the other sampled value functions.

- 1601:  implicit type
  LP:  Objection to new keywords.  "void", "virtual", "super" are possible
    keywords.  Also, use ";" in the formal argument list.
  DK:  Problem with void is that if we want to define a function with 
    implicit type, void will collide.  This may not be worth worrying about.
    E.g., what if you want to pass a function to a sequence

     sequence (f); f(1) ##1 f(2); endsequence

  EC:  There is a keyword "wildcard".
  BT:  void implies there is no type, and then it seems like we are type
    casting.
  EC:  We might push back with "implicit" and try to show that it is useful
    elsewhere.
  BT:  Should research wildcard.
  JH:  Let's push back on implicit -- it is the word we want.
  DK:  I am for implicit, and for sequences and properties it is the same 
    as no type.
  EC:  We could send "implicit" to SV-EC committee and say we would like to
    have functions that return unknown type and we need to be able to 
    distinguish it from void.  JH can create the mail and work with BT to
    send it to SV-EC.

  DK & JH:  Let's form a subgroup to study the semantics of argument passing 
    to sequences and properties.  DB, LP, TS, BT volunteered.  They will
    organize themselves.  DK & JH will monitor the subgroup.

  DK:  Please send feedback on the comment sent to the reflector on 1591.
  TS:  Please comment on questions sent to the reflector.

Items we did not get to:

- 1549:  Missing formal types
- 1668:  Local var initializers
- 1641:  Severity specification for error msgs (should be discussed together with SV-BC)
- 1667:  Local variable args for seq and prop
- 1591:  $past syntax not precise

Next meeting: 

2007-01-09 at 17:00 UTC (11:00 CST)