[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

Solid CSP and Peter's Chef

Dear all

I sent a message named "Solid CSP on top of weak synchronized
semantics" (below) to java-threads. I'm afraid it didn't reach 
the list, as I didn't get a copy.

Chef, Customer, Waiter:

Peter Welch's "Chef" and part of the rest of his team (Chef's)
seem to have made it into "Java 1.2 Dev. Handbook", Heller & 
Roberts ISBN 0-7821-2179-9 p.300-302! The example seems to
be enough modified to let go of the reference..

Solid CSP on top of weak synchronized semantics

Is this problem solved in JCSP and CJT?

If it has been solved, I'd like to know how. 

If the basic mechanism in the bottom is weak (=not portable or 
plain wrong on half of the implementations..(?)), how can a
paradigm built on top be any better (any more portable)?

The ReentrantMutex [1] is a proposal to get rid of the weak
semantics, as far as I can see. Will this be implemented
on top of old synchronized, does anybody know?

In "Real-time Execution Semantics" [2] Sun proposes to tighten
the semantics of synchronized for "VMs that implement the
real-time extension". Would you agree in my statement [3]
that this is not wise?






 Oyvind Teig                              oyvind.teig@xxxxxxxxxxxx
 Navia Maritime AS, division Autronica    oyvind.teig@xxxxxxxxxxxx
 7005 Trondhem                               Tel.: +47 73 58 12 68
 Norway                                      Fax.: +47 73 91 93 20
Opening Theme
Non-preemptive Occam in DSP RT system
Author of choir-page