At 12:56 AM 27/11/2003, Doug Lea wrote:
>This choice is unfortunate in that the prefer-IE policy is always
>better in systems that must deal with timely thread cancellation. If
>it is not allowed, JVMs won't be able to help make systems more
>responsive to interruptions as a quality-of-implementation feature.
I think it's arguable that achieving responsiveness to interrupts has more
to do with the order in which threads are woken (on acquiring the monitor),
than whether they throw IE when they do. Under the prefer notify (or rather
prefer program order) policy, an implementation could favour waking up an
interrupted thread over an uninterrupted one, even though the interrupted
thread would have to return normally with its interrupted flag set. By
contrast, the solutions we have been discussing to allow prefer IE say
nothing about the order in which threads wake up, only that the interrupted
thread throws IE when it does.
JavaMemoryModel mailing list - http://www.cs.umd.edu/~pugh/java/memoryModel
This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:55 EDT