Originally from "multi-bug" 4352424. Creating separate bug for this
problem.


Name: yyT116575 Date: 10/20/2000


Both Windows 2000 running JDK1.3 for Window:
 with Hotspot
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.0-C)
Java HotSpot(TM) Client VM (build 1.3.0-C, mixed mode)

 AND without Hotspot
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.0-C)
Java HotSpot(TM) Client VM (build 1.3.0-C, interpreted mode)


and Redhat Linux 7.0 running JDK 1.2:
Classic VM (build Linux_JDK_1.2_pre-release-v2, native threads, sunwjit)


It worked well in Sparc Solaris environment.


I wrote the following code for a benchmark. It does nothing really
interesting except hanging.

// Speed.java
public class Speed {

static public void main(String[] argv) {
long j=System.currentTimeMillis();

Object[] prev=null;
for (int i=0; i<1000000; i++) {
Object[] o=new Object[2];
o[0]=new Integer(i);
o[1]=prev;
prev=o;
if (i%10000==0) { // for values under 7000,
// it did not hang
System.err.println(i);
prev=null; // allow object to be discarded
}
}
System.out.println(System.currentTimeMillis()-j);
}
}

Compile and run it. You will get a segmentation fault.
(Review ID: 111131)
======================================================================


Name: elR10090 Date: 03/15/2002


Eugene I. Latkin <###@###.###>

I have added this testcase into testbase_nsk, so that
it were executed regularly by the SQE team. The test
will be shipped with "r13" release of testbase_nsk;
and the test name will be:

    nsk/regression/b4404502

The test still causes VM crash against J2SE builds
1.4.0-b92 and 1.4.1-b05; all platforms, but Linux.
======================================================================