Aparapi multiple put/get seems to have no effects

Question

I am facing a problem by using explicit buffer management with Aparapi. The code hereafter shows that I am trying to manage several put/get in a loop to refresh/get-back data from GPU. It seems that the first put and get are done but not the others.

import com.amd.aparapi

// Dummy test to reproduce explicit buffer management 
public class QuickTestExplicit extends Kernel
{
    private static final float DELTA = (float) 1E-5;
    // will be filled, put on GPU in each iterations
    private float[][] values;
    // will be filled with results, put in GPU once but retrieved several times
    private float[] currentRes;

    private void initData()
    {
        values = new float[2000][20];
        currentRes = new float[2000];
    }

    @Override
    public void run()
    {
        int id = getGlobalId();
        long accum = 0;
        // simple sum of elements
        for (int index = 0; index < 20; ++index)
        {
            accum += values[id][index];
        }
        currentRes[id] = accum;
    }

    public void process()
    {
        boolean passed = true;
        initData();

        if (isExplicit())
        {
            put(currentRes);
        }

        for (int row = 0; row < 2000; ++row)
        {
            for (int i = 0; i < values.length; ++i)
            {
                for (int depth = 0; depth < 20; ++depth)
                {
                    values[i][depth] = (float) row;
                }
            }

            if (isExplicit())
            {
                put(values);
            }

            execute(values.length);

            if (isExplicit())
            {
                get(currentRes);
            }

            // just check the success of the operation (for the example)
            passed = true;
            for (int currentIndexRes = 0; currentIndexRes < currentRes.length; ++currentIndexRes)
            {
                passed &= Math.abs(currentRes[currentIndexRes] - (row * 20.0)) < DELTA;
            }
            if (passed)
            {
                System.out.println("ROW " + row + " PASSED");
            }
            else
            {
                System.out.println("ROW " + row + " FAILED");
            }
        }
    }


    public static void main(String[] args)
    {
        QuickTestExplicit kern = new QuickTestExplicit();
        kern.setExecutionMode(EXECUTION_MODE.GPU);
        kern.setExplicit(true);
        kern.process();
    }
}

So my questions are:

• How to force the update of a large buffer already put in GPU memory ?
• Why, when I am running this piece of code with implicit buffer management, a SIGSEV is thrown ?

I don't think that it's a problem relative to GPU memory capacity (2GB memory in my case, and the application just put 2000*20*4 + 2000*4 = 168KB) I'm using a CUDA architecture. FYI, this program passes when running in JTP mode.

Thanks in advance !

EDIT: I forgot to mention that I'm using the "Aparapi_2014_04_29_Linux64" version available in svn/trunk/Downloads".
It seems that the problem occurred when using 2D Java primitive arrays. I rewrite the same algorithm by using 1D Java primitive arrays, and that worked perfectly... Any ideas ?