CUDA 5 Drawback Dynamic Parallelism Examples

I searched googling and was able to find a trivial example of the new dynamic parallelism in Compute Capability 3.0 in one of their technical summaries linked from here . I know that HPC-specific cards are likely to be unavailable until this time next year (after nat'l labs get theirs), and yes, I understand that the simple example they gave is enough for you, but the more fun .

Are there any other examples that I missed?

To save you the trouble, here is the whole example provided in the technical overview:

__global__ ChildKernel(void* data){
    //Operate on data
}
__global__ ParentKernel(void *data){
    ChildKernel<<<16, 1>>>(data);
}
// In Host Code
ParentKernel<<<256, 64>>(data);

// Recursion is also supported
__global__ RecursiveKernel(void* data){
    if(continueRecursion == true)
        RecursiveKernel<<<64, 16>>>(data);
}

EDIT: GTC talk CUDA parallelism CUDA 5. . , , .