SV-ACMinutes2014_12_15 < P1800

P1800 Web>SystemVerilogAssertionCommittee>SVACMeetingMinutes>SV-ACMinutes2014_12_15 (2014-12-18, ErikSeligman) (raw view)EditAttach
<pre><verbatim>

Minutes from SV-AC Committee Meeting

Date:    2014-12-15

Time:    16.30:00 UTC (8:30 PST)

Duration: 1.5 hour

Agenda

</verbatim><hr /></pre> <verbatim>- Reminder of IEEE patent policy.
See: http://standards.ieee.org/board/pat/pat-slideset.ppt
- Elaborating recommendations for the next P1800 PAR.
- Opens
Attendance Record:
 </verbatim>
---
<verbatim> 
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 2014-11-18:
v[x--] Shalom Bresticker (Intel)
v[x--] Dennis Brophy (Mentor Graphics)
v[xxx] Eduard Cerny (Synopsys)
v[xxx] Ben Cohen (Accellera)
v[xxx] John Havlicek (Cadence)
v[xx] Tapan Kapoor (Cadence)
t[xxx] Dmitry Korchemny (Intel - Chair)
v[xxx] Scott Little (Intel)
v[xx-] Anupam Prabhakar (Mentor Graphics)
v[xxx] Erik Seligman (Intel)
v[xx-] Samik Sengupta (Synopsys)
|- attendance on 2014-12-01
|--- voting eligibility on 2014-12-15
Minutes
 </verbatim>
---
<verbatim> 
1. IEEE patent policy reminder
2. Elaborating recommendations for the next P1800 PAR
3478: Make drivers of inout ports accessible
Because of the simulation performance penalty induced by this proposal (as showed the feedback from SV-BC members), Ben suggested to drop this proposal from the recommendation list.
SV-AC will not recommend to include this proposal to the next PAR.
5068: concurrent assertions in classes
Ben: This proposal is too complex to include it to the next PAR. Drop?
John: Consider forming a group to study this issue.
SV-AC will recommend to form an inter-committee group to study this issue.
5070: support sequence/property open arguments
John: This item makes sense not only for sequences and properties, but also for functions and tasks, and therefore should belong to SV-EC.
Erik: To defer this proposal to the next PAR
John: Agree
Ed: We already can pass arguments as sequence
John: There is a difference. Currently we have to pass a sequence with its arguments instantiated, and not just its name.
SV-AC will not recommend to include this proposal to the next PAR.
5071: support assertion match instance reflection API
The people were in doubt whether this proposal should be addressed in the next PAR (Ben, Erik).
John: The notion is interesting. But haven’t seen the use case.
Ed: Can access through VPI.
SV-AC will not recommend to include this proposal to the next PAR.
2182: Ballot comment 56: Elaborate VPI diagrams for checkers
SV-AC will request SV-CC to address this item
4571: For next 1800: Probing property local variables for sharing across boundaries
Ed: This is hard to define
Dmitry: Need to understand the algorithmic complexity
John: Semantics: assignment to a global variable as a task. In favor of considering this.
SV-AC will recommend working on this proposal in the next PAR.
3120: "expect" construct to refer to virtual interfaces
Ed: Too complex
Ben: Let’s drop
SV-AC will not recommend to include this proposal to the next PAR.
2944: concurrent assertions in tasks/functions
Erik – Looks like a low hanging fruit.
Ed: Not clear how to name assertions.
Ed: Performance is very important. Any additional feature may negatively affect performance
SV-AC will not recommend to include this proposal to the next PAR.
3057: Make local variables a first class language construct in SVA.
Ed: The feature is almost ready. We went out of time last time.
John: Should be straightforward
SV-AC will recommend working on this proposal in the next PAR.

 3. </verbatim>Opens

None

-- Main.ErikSeligman - 2014-12-18
Topic revision: r1 - 2014-12-18 - 22:13:21 - ErikSeligman
P1800