Subject: Re: MSC-News: thread of control
From: Sjouke Mauw (sjouke#win.tue.nl )
Date: Mon Aug 18 1997 - 13:22:22 GMT
The originator of this message is responsible for its content.
-----From sjouke#win.tue.nl (Sjouke Mauw) to mscnews -----
> It contains a suggestion for a specification of a 'thread of control', i.e. a protected region
> which must not be interleaved by other events (e.g. an SDL state transition).
I have (at least) one problem with the syntax. It looks very much like the
notation for generalized causal orderings (within a column form instance). The
problem is that it can also be considered as the lines constituting a number of
generalized orderings. This will introduce an ambiguity.
> (For a precise semantics definition ask Sjouke Mauw!)
I have asked Sjouke Mauw, but he was not able to give a precise semantics
> In the second MSC ('REQUEST') I have used a task symbol indicating the activity carried
> out during the transition. Anyway, this also refers to the specification of a remote procedure
1) If there were no arrows coming in/going out, this notation would be identical
to the already existing "local action". So, either you use another notation, or
you assume that there is always an incoming or outgoing message arrow.
2) How does the text "check queue and service subscription" refer to any other
> Please send your comments using the ITU (CCITT) classification:
> a) It cannot be improved
> b) It can be improved
> c) It can be easily improved
d) Why do we need this feature?
-----End text from sjouke#win.tue.nl (Sjouke Mauw) to mscnews -----
For help, email "majordomo#sdl-forum.org" with the body of your email as:
or (iff this does not answer your question) email: owner-mscnews#sdl-forum.org
This archive was generated by hypermail 2a23 : Wed Jun 19 2013 - 13:16:37 GMT