Revision: 878
http://javapathfinder.svn.sourceforge.net/javapathfinder/?rev=878&view=rev
Author: pcmehlitz
Date: 2008-06-23 15:25:43 -0700 (Mon, 23 Jun 2008)
Log Message:
-----------
* this is a re-iteration of Thread.start(). The thinking was that it would be more understandable if
we have SchedulingPoints when we start a new thread, but that causes state bloat with pure starter
threads that just return after starting their workers. In this case, we get transition breaks
on both the Thread.start() calls and the RETURN insn of the starter, which can be more confusing
than not seeing a reschedule. In any way, the new threads should be RUNNABLE, after the tart call
(if they are not blocked), so any following re-schedule in the starter should see them
Modified Paths:
--------------
trunk/src/gov/nasa/jpf/jvm/DefaultSchedulerFactory.java
This was sent by the SourceForge.net collaborative development platform, the world's largest Open Source development site.
|
