Re: [Swarm-Support] strange error with java layer on 64 bit, 4 proc, xeon, ubuntu

[glen e. p. ropella]

Thus spake Marcus G. Daniels circa 15/01/09 01:33 PM:
> The usual problem with x86_64 and Java is the scenario where the Swarm
> scheduler sends a message to a Java object in the `usual' Objective C
> way (i.e. it does not send it directly through an FCall intended for a
> Java destination).

That's great info.  Thanks.  However, the VM abort occurs even with
StartMousetrap.java:

vvvvvvvvvvvvvvvvvvvvvv StartMousetrap.java vvvvvvvvvvvvvvvvvvvvvv

import swarm.Globals;
import swarm.defobj.ZoneImpl;
public class StartMousetrap {
  public static void main (String[] args) {
    Globals.env.initSwarm ("jmousetrap", "2.1", "address@hidden",
args);
    MousetrapBatchSwarm topLevelSwarm =
      new MousetrapBatchSwarm (Globals.env.globalZone);
  }
}
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

And a gutted MousetrapBatchSwarm that looks like:

vvvvvvvvvvvvvvvvvvvvvvvv MousetrapBatchSwarm vvvvvvvvvvvvvvvvvvvvv

import swarm.objectbase.SwarmImpl;
import swarm.defobj.Zone;
public class MousetrapBatchSwarm extends SwarmImpl {
  public MousetrapBatchSwarm (Zone aZone) {
  }
}
^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^^

So, it's not a call via the scheduler that's causing this.  Can  you
think of anything in the construction path that might invoke the same
mframe routines that mangle addresses?

Also, I forgot to mention in the previous e-mail that the simulations
will run and execute normally ~2 out of 30 trials.

-- 
glen e. p. ropella, 971-222-9095, http://agent-based-modeling.com