RE: SDL-News: Process types

Subject: RE: SDL-News: Process types
From: Erik Mats (Erik.Mats#telelogic.com)
Date: Thu May 22 2003 - 16:53:42 GMT

• Messages sorted by: [ date ] [ thread ] [ subject ] [ author ]
• Next message: Scott Shen: "SDL-News: calling a procedure defined in other service"
• Previous message: KEMoss6#aol.com: "SDL-News: Process types"
• Maybe in reply to: KEMoss6#aol.com: "SDL-News: Process types"
• Maybe reply: Erik Mats: "RE: SDL-News: Process types"

Hi Keith,
 
you may want to consider using the "SENDER" expression to determine where a
particular signal originated. However, if you only compile one particular
process type in your test executable there is a great chance that "sender"
will always be "env:1".
 
If you are using the Tau SDL Suite Simulator there is a "trick" you can do
which involves doing "output-internal" instead of "output-to" or
"output-via". This will allow you to specify the sender of signals you
input.
 
If you are using the Tau SDL Suite, you could also implement an "enabling
condition" corresponding to an "input <Signal> FROM <PId>", although this
will be non-standard and a little technical. If I may shamelessly "plug"
myself a bit, I would recommend looking up my article "Setting Your Signal
Priorities Straight" (on how to implement several levels of priority inputs)
from http://www.telelogic.com/resources/
<http://www.telelogic.com/resources/> . The implementation given in this
article should be - somewhat easily - modifiable for an "input from"
construct, where the enabling condition may determine which process sent a
particular signal instance, before accepting an input. I am not sure whether
this would be useful for you.
 
I hope this helps. Let me know if I misunderstood your question, or if I can
clarify anything.
 
Best regards
/Erik
------------------------------------------------------------------------
Tau Generation2 - Development Visualized. Productivity Realized.
Learn more about Telelogic's new system and software development solution
at:
< <http://www.taug2.com/> http://www.taug2.com>
------------------------------------------------------------------------
Telelogic Tau Support
1-800-577-8449 x 1
 <mailto:TauSupport.US#Telelogic.com> mailto:TauSupport.US#Telelogic.com

Erik Mats
Senior Technical Support Engineer, Team Lead, Tau Support
Telelogic North America Inc,
400 Valley Road Ste 200, Mount Arlington NJ 07856
http://www.telelogic.com <http://www.telelogic.com/>
mailto:erik.mats#telelogic.com <mailto:erik.mats#telelogic.com>
Phone: +1-973-770-5830
Fax: +1-973-770-6401
Mobile phone: +1-973-615-6906

Telelogic - Automating the development lifecycle!
------------------------------------------------------------------------

 -----Original Message-----
From: KEMoss6#aol.com [mailto:KEMoss6#aol.com]
Sent: Thursday, May 22, 2003 12:16 PM
To: sdlnews#sdl-forum.org
Subject: SDL-news: Process types

Dear All

This is a question that has me tied up in knots

I have a process type which has a number of inputs which are similar in that
the signal names are the same. In the grand scheme of things they come from
instances of another process type. I want to test the first process type on
its own.

How do I know which input it is receiving a signal when the signal name is
the same and from a number of possible sources? I name the gates and paths.
There are VIA or TO statements in output but no FROM in input. Please is
there simple way around this or must I give up trying to test a process type
under these conditions.

Keith Moss

Dear All

This is a question that has me tied up in knots

I have a process type which has a number of inputs which are similar in that the signal names are the same. In the grand scheme of things they come from instances of another process type. I want to test the first process type on its own.

How do I know which input it is receiving a signal when the signal name is the same and from a number of possible sources? I name the gates and paths. There are VIA or TO statements in output but no FROM in input. Please is there simple way around this or must I give up trying to test a process type under these conditions.

Keith Moss