Subject: RE: [sv-ac] SVAC Meeting minutes 3/20/03
From: Bassam Tabbara (bassam@novas.com)
Date: Fri Mar 21 2003 - 10:17:20 PST
Hi All,
Just a sneak peek, please no cross WG comments ..., Jay brought this up
in the syntax WG, as it stands the thinking on recommendation to AC is
as follows, let Jay/Surrendra/Bassam know if you have concern would like
to be brought up in our upcoming meeting. BTW, the semantic bit is just
to explain what they mean of course.
--- There are 3 different semantics for which an abbreviation is proposed: o repeat _operator M [: N] - the expression is repeated M or M:N times (i.e. a, a, a, a) o nth_event_operator M [ : N ] - stop at the specified occurrence of an event M or M:N (a, true, a, a, true, a) o counting_operator M [ : N ] - count the number of occurrences of an expression and return true if the sequence has that number (a, true, a, a, true, a, true, true). o boolean_abbrev ::= repeat_operator | nth_event_operator | counting_operator sequence_abbrev ::= repeat_operator repeat_operator ::= | [* const_range_expression [ : constant_range_expression ] ] counting_operator ::= [*= const_range_expression [ : const_range_expression ] ] OR [= const_range_expression [ : const_range_expression ] ] nth_event_operator ::= [*> const_range_expression [ : const_range_expression ] ] OR [-> const_range_expression [ : const_range_expression ] ]--
I am leaning towards [*? ...] Quick motivation is that "*" would be the repeat, qualified by other things (=, > ...), and leave room for future expansion if any.
-- Dr. Bassam Tabbara Technical Manager, R&D Novas Software, Inc.
http://www.novas.com (408) 467-7893
> -----Original Message----- > From: owner-sv-ac@server.eda.org > [mailto:owner-sv-ac@server.eda.org] On Behalf Of John Havlicek > Sent: Friday, March 21, 2003 7:56 AM > To: erichm@cadence.com > Cc: Stephen.Meier@synopsys.com; sv-ac@server.eda.org > Subject: Re: [sv-ac] SVAC Meeting minutes 3/20/03 > > > All: > > I agree with Erich 100% on this issue. We should not use the syntax > "*=[n]" in SVA for the same operator that is written as "[=n]" in PSL. > > John H. > > > > > > > Steve, > > > > Regarding this item in the minutes: > > > > ========= > > Item 19: Addition of explicit operator for next event > > > > Adam asked if the nonconsecutive repetition ( *=) provided the same > > functionality. Surrendra indicated that it does. With > this fact it > > was agreed to withdraw this item based on fact that there is no > > justification to have two ways to express same thing. ========= > > > > If I understand the minutes correctly (and unfortunately I > was delayed > > yesterday and could not join the meeting until after this was > > discussed), I agree with the decision. But I'd also like > to point out > > to the SV-ACthat this is a case in which the syntax for SVA clashes > > with the syntax for PSL, and we should do something to correct this > > situation. > > > > Specifically, PSL has two "nonconsecutive" repetition operators: > > > > for N, which is either a number or a range: > > > > b[=N] matches any string that has N occurrences of b > anywhere within it > > > > b[->N] matches any string that ENDS in the Nth occurrence of b > > (and b[->] is shorthand for b[->1]) > > > > Note that both operators are very useful. The first operator is > > useful for counting occurrences of some condition in parallel with > > some other behavior (e.g., Seq1 'intersect' Cond[=0:n] ). > The second > > operator is useful as the 'next_event', or more generally the > > 'nth_event' operator (e.g., the sequence A; B[->3]; C means > "look for > > A, then look for the 3rd occurrence of B, then look for C"). > > > > SVA has used syntax similar to the first operator to express the > > semantics of the second operator. This will potentially confuse > > users. > > > > A proposal has been made to the DWG to adopt both operators in SVA, > > and to align their syntax with that of PSL to avoid > confusion (either > > using the same operators as PSL, or using alternative > operators that > > are different from PSL but clearly correspond with the semantically > > equivalent PSL operator). > > > > Regards, > > > > Erich > > > > > > ------------------------------------------- > > Erich Marschner, Cadence Design Systems > > Senior Architect, Advanced Verification > > Phone: +1 410 750 6995 Email: erichm@cadence.com > > Vmail: +1 410 872 4369 Email: erichm@comcast.net > > > > | -----Original Message----- > > | From: Stephen Meier [mailto:Stephen.Meier@synopsys.com] > > | Sent: Friday, March 21, 2003 1:40 AM > > | To: System Verilog Assertion > > | Subject: [sv-ac] SVAC Meeting minutes 3/20/03 > > | > > | > > | Hi: > > | > > | Here are minutes from today. > > | > > | -Steve > > | > > | Steve Meier (stephen.meier@synopsys.com) W: 650-584-4476, > > | Cell: 408-393-8246 > > | >
This archive was generated by hypermail 2b28 : Fri Mar 21 2003 - 10:18:10 PST