[sv-ac] notes from SV-AC meeting 2007-09-04

From: John Havlicek <john.havlicek_at_.....>
Date: Wed Sep 05 2007 - 04:26:09 PDT
Hi Folks:

My notes from our meeting yesterday are attached.

Let me know if changes are required.

J.H.

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.


Minutes of IEEE P1800 SV-AC meeting #2007-19
Written by: John Havlicek

Date:  2007-09-04
Time:  16:00 UTC (11:00 CDT) 

Dialin information:
-------------------

Country                 Number

AUSTRALIA               1800009128
AUSTRIA                 0800291873
BELGIUM                 080077334
CANADA                  8008671147
CHINA TELECOM (CT)      108001400732
CHINA NETCOM (CNC)      108007140759
DENMARK                 80703159
FINLAND                 0800770233
FRANCE                  0800941695
GERMANY                 08001014519
GREECE                  0080016122039738
HONG KONG               800933578
HUNGARY                 0680017180
INDIA                   0008001006032
INDONESIA               008800105607
IRELAND                 1800944116
ISRAEL                  1809459738
ITALY                   800782388
JAPAN                   00531160427
LUXEMBOURG              80023985
MALAYSIA                1800808386
MONACO                  80093186
NETHERLANDS             08002658223
NEW ZEALAND             0800443736
NORWAY                  80057409
POLAND                  008001114672
PORTUGAL                800819106
RUSSIA                  81080022801012
SINGAPORE               8001011470
SOUTH AFRICA            0800992835
SOUTH KOREA             00308140540
SPAIN                   900967020
SWEDEN                  0201400559
SWITZERLAND             0800563054
TAIWAN                  00801126585
THAILAND                0018001562039684
UNITED KINGDOM          08005280546
UNITED STATES           8008671147

Access Code:  7375405


Attendance Record:
------------------
        Legend:
                x = attended
                - = missed
                r = represented
                . = not yet a member
                v = valid voter (2 out of last 3 or 3/4 overall)
                n = not a valid voter
                t = chair eligible to vote only to make or break a tie

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

vv[xxxxx-xxxxxxxxxxxxxxxxxxxxxxxx-xx] Doron Bustan (Intel)
vv[xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx-x] Eduard Cerny (Synopsys)     
nn[------x-xxx---------x-x-xxx-x---x] Surrendra Dudani (Synopsys)
vv[xxxxxxxxx-xx-xxxxx-xxx-xxx-------] Yaniv Fais (Freescale)
tt[xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx] John Havlicek (Freescale - Chair)
vv[xxxxxxxxxxxxxxxrxxxxxxxxxxxxx-xxx] Dmitry Korchemny (Intel - Co-Chair)
vv[xxx-x--xx--xxxxx----------xx-xxxx] Manisha Kulshrestha (Mentor Graphics)
nn[--------------xxxxx-------x-xx-x-] Jiang Long (Mentor Graphics)
nn[------x--xxx.....................] Joseph Lu (Altera)
vv[xxx..............................] Johan Martensson (Jasper)
nn[-----------x--x-xx--xx-xxxxxxx-x-] Hillel Miller (Freescale)
vv[xxxxxxxxxxxxxx-xxxxxxxx-xxxxxxxxx] Lisa Piper (Cadence)
vv[xxxxx-x-xxxxx-x..................] Erik Seligman (Intel)
nn[-------xxxx-----xxxx-xx----------] Tej Singh (Mentor Graphics)
vv[-xxxxxxxx-xxxxxxxxxxxxxxxxxxxxxxx] Bassam Tabbara (Synopsys)
vv[xxxxxxx-xxxxxxxxxx...............] Tom Thatcher (Sun Microsystems)
   |--------------------------------- attendance on 2007-09-04
 |----------------------------------- voting eligibility on 2007-09-04
|------------------------------------ new voting eligibility


Agenda:
-------

- Reminder of IEEE patent policy.
- Feedback from the champions.
  . Next steps for 1601
  . Next steps for 1648
  . Next steps for 1681
  . Next steps for 1729
- 1549:  Argument passing [DB, LP, JH]
- 1932:  LTL operators [DK, DB]
- 1757:  accept_on, reject_on
- 1900:  Checkers [ES]
- 1682:  Future value functions [DK]
- 1756:  Control of assertions in initial blocks [EC]
- 1758:  Boolean operators ->, <-> [EC]
- 1995:  Assertions and checkers in for loops [ES]
- 2005:  Glitches with immediate assertions [ES]
- 1668:  Local var initializers [JH]
- 1728:  Let proposal [EC, DK]
- Other items


Notes:
------

- Reminder of IEEE patent policy.
- Feedback from the champions.

  Items approved:

  . 1704 passed.

  . 1550 passed with friendly amendments, not yet addressed.
       1) All references to "preponed" should be "Preponed".  This is to 
          make the capitalization of event region names consistent with 
          a change made by the SV-EC in another Mantis item, already 
          incorporated in draft 3a.
       2) The changes to clause 16.8.3 add in two places new text 
          describing that $sampled returns default uninitialized values 
          under certain conditions.
          I recommend that these two new descriptions end with a cross 
          reference: "(see 6.7, Table 6-1)". [clause and table numbers 
          per draft 3a] 

    DK and EC will revise and send out for review. 
    TT will review.  

  . 1567 passed with friendly amendment (make .pdf), already addressed.

  . 1591 passed with friendly amendment (make .pdf), already addressed.  The syntax
    for $changed in 1677 is still the old, restrictive form.
   
    DK will update 1677 and we will revote on it.

  . 1722 passed with friendly amendments ("bind instantiation" --> "bind_instantiation",
    "interface instantiation" --> "interface_instantiation"), already addressed.

  . 1768 passed with friendly amendment: 
       8)  Change 'Type' of 1768 in svdb from Clarification to Enhancement
    Already addressed.

  . 1466 passed with friendly amendment: 
      10) The editor, when implementing 1466, add + and * to 
          cycle_delay_const_range_expression instead of adding ##[+] 
          and ##[*] to cycle_delay_range.

    LP will address this.
    In subsequent e-mail, Brad Pierce retracted the friendly amendment because it 
    introduces +,* into constructs beyond assertions.

  Items failed:
  . Next steps for 1601.
       no - Dave
       [Dave]
       I still believe this enhancement needs a thorough review across all
       committees. It adds additional complexity for what may be an 
       interim solution. There still may be time for this in the 2008 LRM,
       but if not, the penalty for not approving this seems to be one of 
       convenience.
       [Shalom] friendly ammendments suggested
       1) From: "This semantics is described in Subclause 16.7." 
          To:   "This semantics is described in 16.7."
       2) In "The formal arguments w and y of foo2 are untyped, while the 
          formal argument x has data type bit," the word "bit" should be 
          bold.

    LP will update to address Shalom's comments.
    Voice vote on Shalom's amendments all 9y/0n/0a.  The chair was not eligible.
    If champions do not accept on 2007-09-05, we will send revised proposal to
    other committees for one week feedback.

  . Next steps for 1729
       no - Shalom 
       [Shalom] 
       1) The following text is unclear as to what behavior is only
          optional/recommended and what is required:

          "The immediate cover statement is used to detect the occurrence 
          of specific signal values in the procedural code. The tools can 
          collect such information in a database and then report the 
          results at the end of simulation. The reporting should include 
          the number of times the cover statement expression was true and 
          in that sense it is equivalent to recording the success of an 
          assert statement on the same expression.  cover statements can 
          also be used as search targets in formal tools.

          The results of a cover statement shall contain the following:
            - Number of times succeeded
            - Number of times failed"
       2) Most of the text says "can", "should", but the last sentence 
          says "shall".
       3) Also, the following is confusing to the reader as it refers to 
          the BNF without saying so: "In addition, statement_or_null is 
          executed every time expression is true." And if so, 
          'expression' should be in a different font than the rest of 
          the sentence as well as 'statement_or_null'.

    EC will go back through to address Shalom's feedback.
    LP pointed out that 1768 eliminates the requirement for a fail count.
    EC will eliminate the requirement for fail count in alignment with 1768.

  Items under consideration
  . Next steps for 1648
    - DK:  Other committees may find this useful.  Other proposals depend on 
      this one.
    - DK will change the colors.  Then JH will send the revised proposal to 
      all the committees and ask them to consider it.

  . Next steps for 1681
    - This proposal has been revised.
    - We reviewed the responses to the champions.  
    - TT suggested not disturbing the proposal from its current form prior to
      the champions meeting on 2007-09-05.
    - We will hold an e-mail vote to close at 06:00 CDT on 2007-09-05.

- 1549:  Argument passing [DB, LP, JH]
  . JH updated the Annex F part.  Minor changes only.
  . LP has checked the changes from last meeting.  We will wait to see the
    champions say about 1601 tomorrow.  If 1601 fails again, then the 
    non-controversial parts of 1601 will be moved to 1549.

- 1932:  LTL operators [DK, DB]
  . We discussed the clocking of until as inherited, also nexttime.
  . JH will send an example to illustrate the difference between clock flow
    and the recursive application of the clock rewrite rules.

- 1757:  accept_on, reject_on
  . There are still some revisions that need to be made from the previous feedback.

- 1900:  Checkers [ES]
  . Dmitry was working on a major revision of this proposal.  Dmitry will try to 
    finish for the next 
  . DK:  TT said it is important to allow checkers to have output arguments.
    There are several other things that may need to be in a separate enhancement
    proposal.  DK suggests to make a separate proposal.

- 1682:  Future value functions [DK]
  . DB will review again and send feedback, and at that point we will call 
    for a vote if it looks ready.

- 1756:  Control of assertions in initial blocks [EC]
  . There was an e-mail exchange with Manisha.  There is a technical tradeoff
    between backward compatibility and making it easier for the user to get
    the effect of a single assertion attempt after the initial occurrence of
    its leading clocking event.
    
We did not get to the following items.

- 1758:  Boolean operators ->, <-> [EC]
- 1995:  Assertions and checkers in for loops [ES]
- 2005:  Glitches with immediate assertions [ES]
- 1668:  Local var initializers [JH]
- 1728:  Let proposal [EC, DK]
- Other items


Next meeting: 
-------------

2007-09-11 at 16:00 UTC (11:00 CDT), 2 hour slot.
Received on Wed Sep 5 05:09:50 2007

This archive was generated by hypermail 2.1.8 : Wed Sep 05 2007 - 05:10:02 PDT