Towards convergence ? If we want to add causality to this model how about
this. At the read discontinuity we identify the set of all previous writes
from this thread and the set of all future writes. Previous is well
defined since we are SC. For causality we need to ensure that the value we
read doesn't depend on any of the future writes by this thread.
Martin Trotter
-------------------------------
JavaMemoryModel mailing list - http://www.cs.umd.edu/~pugh/java/memoryModel
This archive was generated by hypermail 2b29 : Thu Oct 13 2005 - 07:00:47 EDT