Re: [sv-ac] call to vote on Mantis 1361

Hi Jonathan:

I like the idea of clarification with a diagram.  Maybe we can 
even put it in the LRM.

I'm not sure that your diagram is faithful to the current
written descriptions of the behaviors of the assertion action 
control tasks, although some small adjustments to the text
should be able to fix this.

The proposal currently says

  $assertvacuouson shall enable execution of pass action of all 
  the specified assertions on vacuous success.

To me, this says that there ought to be a fourth state in your
diagram, "action on vacuous passes", and there should be a transition
from the "no actions" state to the "action on vacuous passes" state
under $assertvacuouson.  The transitions from "action on vacuous 
passes" should be

   action on vacuous passes -- $assertpassoff -----> no actions
   action on vacuous passes -- $assertvacuousoff --> no actions
   action on vacuous passes -- $assertpasson ------> action on all passes
   action on vacuous passes -- $assertvacuouson ---> action on all passes

Of course, we don't expect the fourth state to be useful.  Doron
told me he preferred that it be unreachable.  I don't have such 
a strong opinion (at least not yet).

If we want to have the behaviors as shown in your diagram,
then the only detractions I see are that there is no direct
transition

   no actions -- ??? --> action on nonvacuous passes

[you have said that this is not an issue for you] and that the 
transition

   no actions -- $assertvacuouson --> no actions

clashes with the intuition suggested by the name "$assertvacuouson".

I am interested to hear what others think.  We can try to resolve
this through friendly amendments, or someone can vote "no" (including,
of course, an explanation of the rationale) and we can call another 
vote after we have converged on a resolution.

J.H.

> 
> This is a multi-part message in MIME format.
> 
> ------_=_NextPart_001_01C771DF.7BA565D2
> Content-Type: text/plain;
> 	charset="iso-8859-1"
> Content-Transfer-Encoding: 7bit
> X-Former-Content-Transfer-Encoding: quoted-printable
> 
> We now have *two* existence proofs from Doron - one in his
> state transition diagram, one in an email - for how hard it
> is to spell "$assertnonvacuouson" correctly :-)
> 
> I agree with Doron's position but I don't think there is any
> need to change the task names or intended behaviours at this
> stage.  I merely expressed a concern about clarity.  Users
> will not call these tasks a large number of times,
> in practical code; so I have absolutely no problem with the 
> need to make two calls in succession to achieve the desired
> effect.  But we *must* know what happens.  A state diagram
> like Doron's is ideal.  Attached is my attempt at the same
> diagram, for the tasks currently proposed by 1361. 
> 
> Just a thought: Since the behaviour of these tasks is 
> stateful, has anyone considered the possibility of inquiry
> functions to find out what the current state is of any
> given assertion?  You could easily do that by allowing
> $assertpasson and $assertvacuouson to be used either as 
> tasks or as functions returning a bit; when used as
> functions they could indicate the current state.
> --
> Jonathan Bromley, Consultant
> 
> DOULOS - Developing Design Know-how
> VHDL * Verilog * SystemC * e * Perl * Tcl/Tk * Project Services
> 
> Doulos Ltd. Church Hatch, 22 Market Place, Ringwood, Hampshire, BH24 1AW, UK
> Tel: +44 (0)1425 471223                   Email: jonathan.bromley@doulos.com
> Fax: +44 (0)1425 471573                           Web: http://www.doulos.com
> 
> The contents of this message may contain personal views which
> are not the views of Doulos Ltd., unless specifically stated.
> 
> -- 
> This message has been scanned for viruses and
> dangerous content by MailScanner, and is
> believed to be clean.
> 
> 
> ------_=_NextPart_001_01C771DF.7BA565D2
> Content-Type: text/plain; charset=us-ascii
> X-Former-Content-Type: application/octet-stream;
> 	name="passcontrol.pdf"
> Content-Transfer-Encoding: 7bit
> X-Former-Content-Transfer-Encoding: base64
> Content-Description: passcontrol.pdf
> X-Former-Content-Disposition: attachment;
> 	filename="passcontrol.pdf"
> 
> [file:/home/r8aaau/work/abv/ieee/1800/mantis/1361/detached/passcontrol.pdf]
> 
> ------_=_NextPart_001_01C771DF.7BA565D2--

-- 
This message has been scanned for viruses and
dangerous content by MailScanner, and is
believed to be clean.