Subject: [sv-ac] RE: ASWG meeting minutes
From: Jay Lawrence (lawrence@cadence.com)
Date: Wed Apr 02 2003 - 16:24:01 PST
The Cadence vote on the ASWG syntax is NO.
I fully expect the rest of the committee to vote me down quick. I'm
getting used to it.
Since I believe a no vote at this time should be accompanied by a
written explanation...
I believe that the lack of differentiation of variables from other
formal arguments is unacceptable. It allows no checking of lifetime or
type of the variable.
I believe the late addition of templates as a new form of compilation
unit is extremely unwise. We now have module, interface, config, program
block, and template without any good reason to have more than module and
config.
Jay
===================================
Jay Lawrence
Senior Architect
Functional Verification
Cadence Design Systems, Inc.
(978) 262-6294
lawrence@cadence.com
===================================
> -----Original Message-----
> From: Dave Rich [mailto:David.Rich@synopsys.com]
> Sent: Wednesday, April 02, 2003 6:58 PM
> To: Dave Rich
> Cc: Neil Korpusik; Vassilios.Gerousis@infineon.com;
> Surrendra.Dudani@synopsys.com; Joao.Geada@synopsys.com; Jay
> Lawrence; johny.srouji@intel.com; bassam@novas.com; Stephen
> Meier; Arturo Salz
> Subject: Re: ASWG meeting minutes
>
>
> All,
>
> Here are the minutes of the 4/2 meeting. Attached is the
> final BNF which you are requested to vote on with a single
> yes/no answer. In the interest of keeping us out of an
> infinite loop, I will not accept any more changes to the
> document as a condition for your vote. You are welcome to
> make friendly amendments for typos which, if obvious to me as
> chair, I will edit. Remember, there will be another chance to
> make changes to the LRM and BNF in subseqent reviews of the draft.
>
>
> 1 restrictions on assertions in function/tasks
> Listed as an open issue for the sematics group to resolve
>
> 2 property expression in assert
> Adam Krolnik's proposa. Accepted: unnanimously
>
>
>
> 3 sequence parameters
> Left as one list as in the previous proposal. Reject: Jay by
> proxy (Joao did argue your case ;-) ) Accepted: all others. Passes
>
> 4 multiple clock support
> The proposed BNF and the following statement was unanomously accepted:
>
> "The ASWG has tried to syntactically limit where clocked and
> multi-clocked sequence are legal, but there are a few cases
> where semantic checks are required. We believe that because
> sequences may contain embedded named sequences, and for the
> benefit of the user's understanding, semantic rules for
> clocked sequences need to documented in the LRM."
>
>
> 5 sequence_spec
> Minor friendly amendments were accepted.
> "and/or versus &&/||" Reject: Johny Accept: all others
>
> 6 templates
> Syntax reverted back to original SV-AC proposal. The
> following statement was adopted. Reject: Jay by proxy.
> Accept: all others
> "The ASWG did not have sufficient time nor believed we had
> sufficient charter to address problems with the definition of
> templates. Therefore, we have left the syntax of templates as
> they were in SV3.1 Draft3. The BNF presented here for
> templates reflects updated production terminology."
>
> 7 order of local variable and sequence decls
> All agreed that our original decision was that variable
> declarations must come before nested sequence definitions.
> The BNF has been corrected
> --
> Dave Rich
> Principal Engineer, CAE, VTG
> Tel: 650-584-4026
> Cell: 510-589-2625
> DaveR@Synopsys.com
>
>
This archive was generated by hypermail 2b28 : Wed Apr 02 2003 - 16:25:47 PST