RE: JavaMemoryModel: A new box for the Java Memory model

From: David Holmes (dholmes@ics.mq.edu.au)
Date: Tue Nov 09 1999 - 16:19:03 EST


> -----Original Message-----
> From: Doug Lea [mailto:dl@altair.cs.oswego.edu]
> I agree that language changes are a serious matter. But I think that
> David's point (which I agree with) is that if we are relying on having
> people use volatile and final as standard mechanisms to obtain
> conformance to model in situations not otherwise requiring locks, then
> the inability to apply these to array elements is a serious problem --
> perhaps even a fatal flaw.

Thanks Doug - that was exactly my point. I would argue that the only reason
this wasn't originally in the language was because they couldn't figure out
a good syntax ;-) I'm afraid I don't see any alternative other than a C/C++
style approach - after all they were faced with the same problem.
Unfortunately Java already has the additional complication that two
different array syntax forms are allowed.

In any case extending the semantics to cover array elements would be for
completeness only. The new semantics of volatile and final can be defined
without any reference to arrays.

Cheers,
David

-------------------------------
JavaMemoryModel mailing list - http://www.cs.umd.edu/~pugh/java/memoryModel



This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:22 EDT