Name: tb29552 Date: 11/03/2000


/*
java version "1.3.0"
Java(TM) 2 Runtime Environment, Standard Edition (build 1.3.0)
Java HotSpot(TM) Client VM (build 1.3.0, mixed mode)

Strange behavior of JPDA -

long a = 973230999;
long b = a*1000L;

in debuging JVM all fine, but in JVM with debugger I see b is -1711011192 ?
How i can debug, if my debugger lay to me???

My server jvm is running with following params -

/usr/java1.3/bin/java -Xdebug -Xnoagent -Djava.compiler=NONE
-Xrunjdwp:transport=dt_socket,server=y,suspend=n,address=60013
-Xoptimize -Djdbc.drivers=postgresql.Driver

As client jvm with debugger I use jdk130/windows
and as JPDA frontend used Karmira BugSeeker2 and Inprise JBuilder4 Enterprise
both show b as -1711011192 :)

It must be stressed, if server jvm is jdk122, all's fine with remote debugging

Cheers, Alexey
*/
//
class testLocals {
    public long test() {
        long a = 973230999L;
        long b = a*1000L;
        long l = 22;
        float f = 6.66f;
        double d = 7.77;
        System.out.print ("a is: ");
        System.out.println(a);
        System.out.print ("b is: ");
        System.out.println(b);
        System.out.print ("l is: ");
        System.out.println(l);
        System.out.print ("f is: ");
        System.out.println(f);
        System.out.print ("d is: ");
        System.out.println(d);
        return b;
    }

    public static void main(String[] args) {
        testLocals my = new testLocals ();
        long c = my.test();
        System.out.print ("c is: ");
        System.out.println(c);
    }
}

(Review ID: 111781)
======================================================================

Name: egR10015 Date: 05/07/2001


This bug is still reproducible with HS 1.3.1-rc2-b23 on solx86 2.8.
Incorrect values are reported for type double on this platform.
The bug 4419982, closed as duplicate of this one, specifies
the problem in detail.
Note that incorrect double values are never observed with Merlin
on all platforms including solx86, as well as with Ladybird on
all platforms except solx86.

The bug affects the following tests from testbase_nsk:

nsk/jdi/ArrayReference/getValue/getvalue001
nsk/jdi/ArrayReference/getValues/getvalues001
nsk/jdi/ArrayReference/getValues_ii/getvaluesii001
nsk/jdi/ArrayReference/getValues_ii/getvaluesii002
nsk/jdi/ArrayReference/setValue/setvalue001
nsk/jdi/ArrayReference/setValues_ilii/setvaluesilii001
nsk/jdi/ArrayReference/setValues_l/setvaluesl001
nsk/jdi/ArrayReference/setValues_l/setvaluesl002

I mentioned above list of the failed tests for an internal
script of the Novosibirsk Hotspot testing team. It allows
the script to automatically generate list of the knowing bugs.

======================================================================

Name: egR10015 Date: 05/07/2001


The bug also affects the following test from testbase_nsk:

nsk/jdi/ClassType/setValue/setvalue001
======================================================================