Thanks. I see it is OK to just exclude the tests.
On the other hand the failure looks not simply from the tests because
different testcases break in different platforms (but all Uncaught exception
in similar categories). I just wonder if it exposes some issues in exception
handling/testing.
public void threadShouldThrow() {
threadFailed = true;
fail("should throw exception");
}
public void run() {
try {
lock.lockInterruptibly();
threadShouldThrow();
} catch(InterruptedException success){}
}
Sorry or the confusion.
On 9/5/08, Nathan Beyer <[EMAIL PROTECTED]> wrote:
>
> I mentioned it in another thread, so it may have gone unnoticed.
>
> On Thu, Sep 4, 2008 at 10:38 AM, Tim Ellison <[EMAIL PROTECTED]>
> wrote:
> > chunrong lai wrote:
> >> hi, Nathan:
> >> I see the commit breaks some classlib test, include
> ReentrantLockTest,
> >> AbstractQueuedSynchronizerTest, ExecutorsTest and ScheduledExecutorTest,
> as
> >> reported in the integrity testing. Do we need a check?
> >> One can reproduce the error in the directory of working_classlib: ant
> >> -Dtest.jre.home=...\drlvm\...\jdk\jre -Dtest.case=ReentrantLockTest
> test.
> >> Thanks.
> >
> > In Nathan's defense, he didn't break anything but simply enabled tests
> > that we should have been running. Maybe these need JIRAs and adding to
> > the exclude list until somebody steps forward to address them.
> >
> > Regards,
> > Tim
> >
> >
>
