ByMusti

__device__ int global_thread_idx_x()
{
	return blockIdx.x*blockDim.x + threadIdx.x;
}

int ceil_div(int const numerator, int const denominator)
{
	auto res = std::div(numerator, denominator);
	return res.rem ? (res.quot + 1) : res.quot;
}


__host__ __device__ __forceinline__
float norm(float3 const & p, float3 const & q) {
	float xDist = ((p.x - q.x)*(p.x - q.x));
	float yDist = ((p.y - q.y)*(p.y - q.y));
	float zDist = ((p.z - q.z)*(p.z - q.z));

	return (xDist + yDist + zDist); // wir verzichten auf sqrt, da es für uns irrelevant ist
}

__host__ __device__ __forceinline__
dim3 blocksPerGrid(dim3 const & block, int3 const & size)
{
	return dim3(ceil_div(size.x, block.x), 1, 1);
}