Chris Hegarty wrote:
java/lang/Thread/ThreadStateTest.java can fail with when
checkThreadState finds an unexpected state.
Exception in thread "main" java.lang.RuntimeException: MyThread
expected to have TERMINATED but got RUNNABLE
at ThreadStateTest.checkThreadState(ThreadStateTest.java:119)
at ThreadStateTest.main(ThreadStateTest.java:96)
There is a race between the thread being put in a specific state and
the thread testing for that state. The test should retry the thread
state check a number of times before failing. Also, some minor cleanup
and update to use a more recent j.u.c reusable synchronization barrier.
http://cr.openjdk.java.net/~chegar/7021010/jdk8.webrev.00/webrev/
-Chris.
The retry loop in checkThreadState make sense. Is the 100ms sleep a bit
excessive? The thread will likely get to the expected state in a
fraction of that time.
One question on the TERMINATED state. Given that the check is now moved
to after the join then could this be a simple getState check rather than
using checkThreadState?
The clean-up to use Phaser looks good to me.
-Alan.
