Another Thread.stop() issue
Dalibor Topic
robilad at yahoo.com
Wed May 31 03:29:34 PDT 2000
Hi,
I've got a bit of follow-up on Jason's Thread.stop() message. I've found a
bug in Kaffe's Thread.stop(Thowable object) method. The following code, legal
under Java 1.1, 1.2 and 1.3 exposes it:
---ThreadStop.java---
class Runner extends Thread {
public void run() {
try{
//Sleep for a while
sleep(10000);
}
catch (InterruptedException _) {}
}
}
class ThreadStop {
static public void main(String[] argv) {
Runner u = new Runner();
// start runner thread
u.start();
try {
Thread.currentThread().sleep(1000);
}
catch (InterruptedException e) {
e.printStackTrace();
}
u.stop(new Exception("Success."));
try {
Thread.currentThread().sleep(1000);
}
catch (InterruptedException e) {
e.printStackTrace();
}
System.exit(0);
}
}
---cut!---
r
unning it with JDK 1.1.7 on Linux gives me:
$ java ThreadStop
java.lang.Exception: Success.
running it with lates kaffe from CVS gives me:
$ kaffe ThreadStop
java.lang.ClassCastException: can't cast `java/lang/Exception' to `java/lang/Error'
at java.lang.Thread.waitOn(Thread.java:473)
at java.lang.Thread.sleep(Thread.java:364)
at Runner.run(ThreadStop.java:7)
If you take a look at the code, you'll see death being cast to an Error
object, which can be thrown without requiring special declaration. On the other
hand, the stop(Throwable) method can receive any kind of throwable object, so
it seems to be a bug to cast it to an Error.
I tried to fix the this with some obvoiues fixes (remove death altogether,
allways use stop0) but unfortunately I couldn't get it to work. I kept getting
this:
java.lang.IllegalMonitorStateException
at java.lang.Thread.sleep(Thread.java:364)
at Runner.run(ThreadStop.java:23)
So I decided to leave it up to the great threading hackers to fix the things :)
Have some fun,
Dali
__________________________________________________
Do You Yahoo!?
Talk to your friends online with Yahoo! Messenger.
http://im.yahoo.com
More information about the kaffe
mailing list