SV-ACMinutes2010_11_30 < P1800

P1800 Web>SystemVerilogAssertionCommittee>SVACMeetingMinutes>SV-ACMinutes2010_11_30 (2010-12-03, ErikSeligman) (raw view)EditAttach
Minutes from SV-AC Meeting

Date: 2010-11-30

Time: 16:30 UTC (8:30 PST)

Duration: 1.5 hours

Dial-in information:

--------------------

Meeting ID:            38198

Phone Number(s):

1-888-813-5316         Toll Free within North America

Live Meeting: [[https://webjoin.intel.com/?passcode=5745158]]

Agenda:

-------

- Reminder of IEEE patent policy.

See: http://standards.ieee.org/board/pat/pat-slideset.ppt

- Minutes approval

- Email ballot results

2552: Confusing comments regarding nexttime operator

The issue passed: 10y/0n/0a

- New issues

3295: need a way to control only asserts/covers/assume directives

- Issue resolution/discussion

Addressing champions' feedback.

2330: Clarify that number_of_ticks argument to $past must be compile-time

     constant. John volunteered to update the proposal

2387: Layout of 16.11 is inconsistent

2557: Rules for passing automatic variables to sequence subroutines are not

     clear

2804: Need to clarify rule (b) in 16.15.6 to allow inferred clock when

     expression appears in procedural assertion

2927: Precedence between sequence/property operator and normal expression

     operator

2934: Precedence and associativity of case operator is not shown in the

table

3113: Add port_identifier to constant_primary BNF for sequences, properties

     and checkers

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

Attendance re-initialized on 2010-07-06:

  v[xxxxxxxxx-x-xxxxx--xxx] Laurence Bisht (Intel)

  v[xxxxxxx-xxxxxxxxxxxxx-] Eduard Cerny (Synopsys)

  v[x-xxxxxxx-xxxxx-xxxxxx] Ben Cohen

  n[---xx-x-xxx-x--xxxxxxx] Surrendra Dudani (Synopsys)

  n[-x--xx---xxxx---x-xxxx] Dana Fisman (Synopsys)

  v[-xxxxx-xxxx-x-xxxxxxxx] John Havlicek (Freescale)

  v[xxxxx-xxx-xxxxxxxxxxxx] Tapan Kapoor (Cadence)

  t[xxxxx--xxxxxxxxxxxxxxx] Dmitry Korchemny (Intel ¿ Chair)

  v[xxxxx--xxxxxx-xxxxxxxx] Scott Little (Freescale)

  v[xxxx-xxxxxxxxx-xxxxxxx] Manisha Kulshrestha (Mentor Graphics)

  v[xx-xxxxxxxxxxxxxxxxxxx] Anupam Prabhakar (Mentor Graphics)

  v[xx-xxx-xx--xxxxxxx-xxx] Erik Seligman (Intel)

  v[-xxxx--xxxxxx-xxxxxxx.] Samik Sengupta (Synopsys)

  v[xxxx-xxxxxxxxxxxxx-xxx] Tom Thatcher (Oracle ¿ Co-Chair)

  n[x.....................] Srini Venkataramanan (?)

    |- attendance on 2010-11-23

  |--- voting eligibility on 2010-11-30

Minutes:

- Reminder of IEEE patent policy.

See: http://standards.ieee.org/board/pat/pat-slideset.ppt

Participants were reminded of the IEEE patent policy.

- Minutes approval

Erik: Move to approve minutes from last meeting

Scott:     Second

     Vote results 7y, 0n, 0a

- Email ballot results

2552: Confusing comments regarding nexttime operator

The issue passed: 10y/0n/0a

- New issues

3295: need a way to control only asserts/covers/assume directives

Manisha:   Thinking of adding arguments to control the type of

     assertions.

Ed:  Assertion libraries like OVL give a way to turn off

Erik: But it's painful to have to re-compile for this.

Ed:  How do you maintain backward compatibility.

Manisha:   Make it an optional argument

Ed:  But still would not be backward-compatible.

(Ben & Lawrence joined)

(Srini joined)

Erik: Looks like we will need a new more generalized function to do this

Scott:     Then existing functions would be deprecated?

Manisha:   Don't have to deprecate them this time around.

Scott:     We'll probably need working-group permission to work on this.

- Issue resolution/discussion

Addressing champions' feedback.

2330: Clarify that number_of_ticks argument to $past must be compile-time

     constant. John volunteered to update the proposal

Dmitry:    Language has been updated.

Scott:     MOve to approve

Erik: Second

     Vote results: 9y, 0n, 0a

2387: Layout of 16.11 is inconsistent

     Erik has not updated proposal yet.

2557: Rules for passing automatic variables to sequence subroutines are

not clear

        Erik has not updated proposal yet.

Dmitry:    There was an e-mail discussion regarding this item.

2804: Need to clarify rule (b) in 16.15.6 to allow inferred clock when

expression appears in procedural assertion

Erik: Still don't understand John's objection.

Dmitry:    Defer to next week.

2927: Precedence between sequence/property operator and normal

expression operator

Scott:     "Add at bottom of Table 16-3" Does this mean the addition is IN the

     table, or after the bottom of the table. It's not clear.

     Where does the text actually go?

Ed:  Added text should go in the paragraph right before the table.

Tapan:     Should we also add this text near Table 16-1 Operators in 16-1 are

     also included in 16-3?

Ben: After a second look, don't understand the confusion in the first

     place. It seems clear to me now.

Scott:     It is a useful clarification.

Anupam:    Danger of creating confusion by clarifying something that is already

     clear.

Erik: Uploaded the new proposal to Mantis

Erik: Move to approve modified proposal:

Scott:     Second

     Vote results: 8y, 0n, 1a

     Anupam: abstain for reason given above. Concerned about creating

           confusion by clarifying something that is already clear.

3113: Add port_identifier to constant_primary BNF for sequences,

properties and checkers

Lawrence:

Erik: OK, I see that the BNF does forbid this case. but does the added

     footnote solve the problem?

Erik: Why don't we fix BNF?

Ed:  Not sure we need to change anything.

Dmitry     port_identifier is defined as identifier

Ed:  Formal identifier is just an identifier

Manisha    Def of constant_primary missing sequence_port_item.

Ed:  There is a difference between formal arg and actual arg.

Manisha    If we only check constant-ness after rewriting algorithm, then it's

     OK, but there is nothing that says this in LRM

Ed:  LRM: Section F.3.3:

Dmitry:    OK to add a brief discussion in the LRM describing this?

Ed:  P 325: "Rewriting algorithm substitutes actual arguments for formal

     arguments. If the result is illegal, there shall be an error"

     So I don't see what the problem is.

Scott:    P 363 The same wording is found for properties.

Dmitry:    Missing that wording for checkers?

Ed:  "Mechanism for passing arguments to a checker is similar to that for

     a property"

Ed:  Think that we don't have to change anything.

Manisha:   Agree w/ shalom: Footnote is not adding anything.

Dmitry: Out of time : defer to next time

Meeting time:

Srini: Wanted to move meeting to 7:30am

Manisha:   Fine with this time right now.

Tom: Earlier time Would be too early

Will keep current meeting time for now, as there is no consensus to

change it.

Meeting adjourned.
Topic revision: r1 - 2010-12-03 - 20:18:43 - ErikSeligman
P1800