Re: JavaMemoryModel: JLS3 contains glitch concerning volatiles?

From: Bill Pugh (pugh@cs.umd.edu)
Date: Wed Aug 17 2005 - 21:11:26 EDT


On Aug 17, 2005, at 8:05 PM, Bart Jacobs wrote:

> Thanks. What I take away from your answer is the corrected
> definition of
> "correctly synchronized". Hopefully it will make it into the JLS at
> some
> point. I don't really get the message from the rest of your answer.
>
> Thanks,-
> Bart

The point is that nothing in the semantics depend upon the definition of
correctly synchronized.

Rather, you can prove that the model provides certain properties for
executions
of correctly synchronized programs (appropriately defined).

So if the JLS doesn't include the words "non-volatile" when it describes
correctly synchronized programs, it doesn't effect the semantics at all.

Bill

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



This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:01:10 EDT