To restate what I think Bill is saying, in order to guarantee
that
volatile x = 0;
while (x == 0) Thread.yield();
...
does what any programmer would expect, we need to do one of the following:
1. Define memory semantics for yield()
2. Guarantee that an infinite number of yield()s causes at least one
actual yield.
Can anyone think of any alternatives?
I think that Bill is right that (2) is a better path.
-Doug
-------------------------------
JavaMemoryModel mailing list - http://www.cs.umd.edu/~pugh/java/memoryModel
This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:36 EDT