Written by: Dmitry Korchemny

Date: 2010-04-27

Time: 16:00 UTC (9:00 PDT)

Dial-in information:

Toll number: +1 916-356-2663

Toll free number (US): 888-875-9370 (U.S. toll-free)

Bridge: 3 Passcode: 3065854

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 2010-04-13:

v[xxx] Laurence Bisht (Intel)

v[xxx] Eduard Cerny (Synopsys)

v[xxx] Ben Cohen

v[x-x] Surrendra Dudani (Synopsys)

v[xxx] Dana Fisman (Synopsys)

v[xxx] John Havlicek (Freescale)

v[xxx] Tapan Kapoor (Cadence)

t[xxx] Dmitry Korchemny (Intel – Chair)

v[xx.] Scott Little (Freescale)

v[xxx] Manisha Kulshrestha (Mentor Graphics)

v[x..] Anupam Prabhakar (Mentor Graphics)

v[xxx] Erik Seligman (Intel)

v[xxx] Tom Thatcher (Sun Microsystems – Co-Chair)

|- attendance on 2010-04-27

|--- voting eligibility on 2010-04-27

Agenda:

- Reminder of IEEE patent policy.

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

- Minutes approval

Ben moves to approve the meeting minutes from 2000-04-22.

Erik seconds.

The motion passed: 0n/0a/12y.

- Issue prioritization

All existing enhancements and effort-consuming errata and clarifications have been collected in a spreadsheet in

https://spreadsheets.google.com/ccc?key=0AgfncMbEn369dFl3cXRndWl3TDRwTXF3SkJvWklseHc&hl=en

Action item (all): By the next meeting to fill in individual priority estimation.

The priority should be between 1 and 25 (including), and if several issues are assigned the same priority, these issues should belong to the same category. In addition one may optionally indicate the issue importance using the following scale: must/important/nice/unessential/drop.

The WG prioritization should be taken into account:

* Cleanup and Ambiguity Resolution

* AOP

* Connections to Analog

* Assertions & Checkers

* Interfaces

* Covergroups

* Deprecating Features

* Links to SystemC

Some of the above issues may have a marginal relation to SV-AC only.

Manisha: It is preferable to have proposals first to make the issue intent clear.

Dmitry: It is the WG requirement to prioritize the issues before we start working on them. Marking an issue as important indicates that its resolution is important, there is no need to agree with a specific resolution at this point.

Specific issue discussion:

Ben: It is dangerous to introduce output arguments in checkers, otherwise it may become possible to inadvertently modify DUT signals. A checker is an observer in its nature. Hierarchical references in checkers should also be disallowed. Having nested checkers is not usable, and it is cleaner to write standalone checkers. Checkers are usually small, and allow procedural instantiation. If checkers are instantiated statically, they may be replaced with modules.

Ed: Modules cannot infer their instantiation context.

John: Checkers can be enhanced to represent environment, both stimuli and FV. Having output arguments is important.

Ben: Why not to use modules for this purpose?

John: E.g., one can take advance of checker variable randomization.

Tom: Allow output arguments only when a checker is instantiated in another checker.

Ed: Then you lose the ability to generate stimuli from a checker.

Dmitry: Output checker arguments are required to implement OVL on top of checkers, since OVL checkers generate an assertion status bit. Also, it is useful to prune a DUT from a checker for FV.

- Mantis item resolution and discussion

Tom: Our main charter is the issue prioritization, and not the issue resolution.

Dmitry: We focus on the prioritization, and work on issue resolution as time permits.

2858: Clarify the rules for assigning a value to a non-checker variable from within a checker

Dmitry: This issue is not the trivial according to the email discussion. Suggest skipping for now.

1551: Make disable iff sampled.

Ben: Should be dropped.

Erik: In checkers everything is sampled, and also disable iff. In modules it is non-sampled, and we have an inconsistent behavior.

Tom: disable iff is non-sampled to make the simulation efficient.

Ed: LRM does not even specify from where the disable iff value is taken.

Ben: It is possible to introduce a sampled version of disable iff.

Dmitry: Suggest to consider this Mantis together with 3035, where checker argument sampling is discussed.

Tom: Skip for now.

2484: deferred cover should have only statement_or_null, not a full action block

Ed: Already implemented. Move to resolve as "no change required".

Tom seconds.

Motion passed: 0n/0a/12y

2480: Bug note 7169 not implemented in 1668.

Erik: Already implemented. Move to resolve as "no change required".

Tom seconds.

Motion passed: 0n/0a/12y

2916: Wrong font in 16.4.2

Laurence: Already implemented.

Tom moves to resolve as "no change required".

Laurence seconds.

Motion passed: 0n/0a/12y

2955: Checker example is wrong

Dmitry: Instead of "generate if (coverage_level = cover_none)" it should be "generate if (clevel = cover_none)"

Tapan agreed to write a proposal. We will conduct a voice vote next time.

2252: Several symbols in Annex F are in green

Ed, John - no green symbols anymore

John moves to resolve as "no change required".

Ed seconds.

Motion passed: 0n/0a/12y

3020: Recursive property Restriction 4 is not consistent between Clause 16.13.17 and Annex F.7

John will write a proposal. We will conduct a voice vote next time.

2804: Need to clarify rule (b) in 16.15.6 to allow inferred clock when expression appears in procedural assertion

Erik agreed to own this issue and to suggest an appropriate language for an accurate description.

- Opens.

Next meeting:

Date: 2010-05-11

Time: 16:00 UTC (9:00 PDT)

Duration: 2 hours

Dial-in information:

Toll number: +1 916-356-2663

Toll free number (US): 888-875-9370 (U.S. toll-free)

Bridge: 4 Passcode: 1859839

-- ErikSeligman - 2010-04-30