JavaMemoryModel: Ok, last thing Monday, rather than first thing

From: Bill Pugh (pugh@cs.umd.edu)
Date: Mon Jul 21 2003 - 21:48:59 EDT


Sorry for the delay.

Jeremy and I have finished first drafts of rewrites of the causality
section of the memory model.

In going over it, we came up with two different ways of explaining
the model. Both explanations detail exactly the same model (excepting
any typo's or other such problems in the documents). Which
explanation you prefer is a matter of taste and opinion as to which
explanation is easier to understand. So we would appreciate any
feedback on that issue.

The two explanations are at:
        http://www.cs.umd.edu/~pugh/java/memoryModel/A.pdf
        http://www.cs.umd.edu/~pugh/java/memoryModel/B.pdf

These aren't intended to be complete, and don't explain lots of
details (such as how the happens-before ordering is established). The
A version includes some definitions, but the B version leaves out
most definitions.

As I said, these are drafts for early adopters (Hello Sarita). We
just want to get feedback on these approaches. We'll have a more
polished version later in the week.

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



This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:46 EDT