Re: [$ieee-1850] IEEE 1850 PSL LRM draft 5....

Hi Harry,

Below please find my comments for chapters 1-3.

1) 
Clause 1.1 Scope (p.1): the name of the language developed at
Accellera appears once as "PSL": (in the 2nd parenthesis) and later as
"Accellera PSL" (at the 4th line). Since the name of the IEEE-1850
logic is also "PSL" (first line, first parenthesis), I think it would
be clearer if the short name of the name of the Accellera language
would read "Accellera PSL" (in the 2nd parenthesis of this paragraph).

2)

1.3.1 Functional Specification (p. 2), starting at the 10th line of
the 3rd paragraph reading "They can be connected using the temporal
operators always and next to get the property…". There are a few minor
problems:
 a. The sentence is inaccurate since the given property uses also 
     the suffix implication operators (old style {}())
 b. It seems more appropriate to use new style suffix implication
 c. The sentence is not well phrases: what does it mean to connect 
     expressions using temporal operators? The phrasing should be 
     "temporal operators are applied to expressions". 
I thus suggest rephrasing this sentence and the examples as follows:
"The following property, obtained by applying the temporal operators
\mt{always} and |=> (\mt{non-overlapping suffix implication}) to these
expressions,
          always {req;ack;!cancel}|=> (ena || enb)
means that following any sequence of {req;ack;!cancel} (i.e., always),
either ena or enb is asserted one cycles later. Adding the directive
assert as follows:
          assert always {req;ack;!cancel}|=> (ena || enb);
completes the specification, indicating…."

[I guess this comment was due for v1.1 as well. I gave it since I
believe it is an easy non-controversial fix].

3) 
1.3.2.1 Simulation, Figure 1: I think this comment was also made by
Erich. The figure looks very bad. I am willing to redraw it, if you
can use one of the following formats: gif,gpg,png,bmp,ppt.

4) 
1.4 Contents of this standard (p.5): last item has a redundant "/" at the end.

5) 
3.1.48 strong operator (p. 11): In accordance with the change we made
to the section on weak vs. strong operators (4.4.6.1), I think we
should add the word "usually" to this definition as well. The new
definition should thus read: "A temporal operator, the non-negated use
of which usually creates a liveness property."

6) 
3.2 Acronyms and abbreviations (p. 12): OSCI is not Optional Branching
Extension.

Best regards,
Dana.


On Wed, 16 Mar 2005 19:23:23 -0800, Harry D. Foster
<harrydfoster@comcast.net> wrote:
> Hi Group,
> 
> You can now download the latest draft for the IEEE 1850 LRM at:
> 
>        http://www.eda.org/ieee-1850/docs/IEEE-1850-PSL-DRAFT-Mar-16.pdf
> 
> The latest draft  removes 'endpoints' from the language.   There were
> other minor edits that occurred.  Note that change bars are turned on
> for all edits.   Also note that I still see a few minor spacing
> problems I want to fix (like before Box 47).   If you see other spacing
> problems--please let me know.
> 
> Also, I would appreciate a good review of Annex A, the BNF.  I'm
> worried about its accuracy.
> 
> Best regards,
> 
> -Harry
> ---
> Harry D. Foster
> Chief Methodologist           
> Jasper Design Automation
> http://www.jasper-da.com
> Tel: +1 650.966.0239     Cell: +1 650.804.5000
> 
>