Subject: Re: SDL-News: SDL transitions
From: Jean-Loup TERRAILLON (jeanloup#wm.estec.esa.nl)
Date: Wed Sep 15 1999 - 13:35:29 GMT
The originator of this message is responsible for its content.
-----From Jean-Loup TERRAILLON <jeanloup#wm.estec.esa.nl> to sdlnews -----
I believe that there are two levels:
1) Theorically, the time semantic of SDL says that execution time of a
transition is zero, that all the possible transitions are fired
consequently at the very same time, and that when no transition can be
fired, then the closest timer expire and the time "jumps" to the next
value, triggering a new transition burst.
2) Practically, execution time is never zero. Therefore, the code
generators of the various tools have to implement a practical solution.
One of it is to turn the scheduling model of the theorical SDL into a
practical priority based (or not) pre-emptive (or not) scheduling model.
The important point is to check that the simulator and the code
generator implements the same scheduling model, such that the code has a
behaviour compliant with the model.
Finally, the scheduling model can be associated with an analytical model
that will check that the application is actually schedulable (taking
into account the real execution time), i.e. that each task complete
before its deadline (or that no event is lost, all the transitions
complete, etc.)
Jean-Loup TERRAILLON
European Space Agency
Software Engineering
-----End text from Jean-Loup TERRAILLON <jeanloup#wm.estec.esa.nl> to sdlnews -----
Join http://www.sdl-forum.org/Society/members.htm for extra SDL Forum Society benefits
For help, email "majordomo#sdl-forum.org" with the body of your email as:
help
or (iff this does not answer your question) email: owner-sdlnews#sdl-forum.org
This archive was generated by hypermail 2a23 : Sun Jun 16 2013 - 10:41:41 GMT