Re: Verilog-AMS committee meeting agenda - 28 March 2005

Sri -
I went through Chapter 5, Signals, and updated the four syntax
boxes.  This was pretty trivial; the result is here:

http://www.eda.org/verilog-ams/htmlpages/public-docs/signals_2.3_draft.pdf


I am hoping Graham will send me the Frame for the Syntax before
I tackle Chapter 3, which has many more syntax boxes.


Notes on Chapter 5:
1) 5.2 is titled "Signal access for vector branches" but
Syntax 5-1 on pdf page 9 is really for any signal access,
and also 5.2.1 and 5.2.2 are not restricted to vector
branches, so 5.2 should be re-titled.


2) With multiple people editing the Frame source, we need to be
clear on the use of Character and Paragraph tags.  Eg, Ch 5
uses the "keyword" character tag to make things bold; this is
better than using Format>Style>Bold.

Also, what about spaces for readability?  In this item:
   nature_attribute_identifier ( branch_reference )
there are spaces around the parentheses, but when someone
actually uses it, I would expect "V(neta)" not "V ( neta )"
Does this matter to the parser in anyone's simulator?


3) I need to review the Mantis DB for issues related to Chapter 5,
eg, I think there's something in there about switch branches.


4) I have not copied in the semantic notes:

A.6.10:
  (for contribution_statement)
  NOTE: Semantically restrict the use of analog filter function
  calls in the statements RHS expression when the statement is
  conditionally executed during a simulation.

  (for indirect_contribution_statement)
  NOTE: Semantically restrict the use of indirect contribution
  statements in conditionally executed behavior during a
  simulation (this includes sub analysis (ic, tran) of a simulation).

A.8.2
  (for branch_probe_function call and port_probe_function_call)
  NOTE: Limit port_probe_function_call function to use local ports
  only (no hierarchical reference allowed).
  NOTE: Semantically restrict branch_probe's nature potential/flow
  attribute only.
  NOTE: Semantically restrict port_probe's flow nature's access
  attribute only.
  QUESTION: If an analog/digital UDF of the same named as a nature
  accessor is declared. When this nature access is used in an analog
  probe, should the value of the analog net (using the nature) be
  accessed or the should the UDF be called?
  ANSWER: open issue to be discussed further.

A.8.9
  (for nature_attribute_reference)
  NOTE: Semantically restrict the accessing of idt_nature and
  ddt_nature, access identifier attributes since their values are
  not compatible with constant_expression and analog_expression
  primary types.

  NOTE: Semantically restrict nets used in analog_port_reference
  and analog_net_reference to be continuous only.

(Anything I missed?)

-Geoffrey