modifying bitonic sort

Accelerated Computing CUDA CUDA Programming and Performance
1

Dear all;
I has downloaded the code pf bitonic sort and ran it
but when I use array of non power of two, for example 10 elements, the array is not sorted correctly
can you please help me in sorting the array?
here is the code

__global__ static void bitonicSort(int * values)
{
	__shared__ int shared[SHARED_SIZE_LIMIT];

	const int tid = threadIdx.x;

	// Copy input to shared mem.
	shared[tid] = values[tid + blockIdx.x * SHARED_SIZE_LIMIT];
	//	shared[(tid)+(SHARED_SIZE_LIMIT)+blockIdx.x *SHARED_SIZE_LIMIT] = values[tid + (SHARED_SIZE_LIMIT)];
	__syncthreads();

	// Parallel bitonic sort.
	for (int k = 2; k <= SHARED_SIZE_LIMIT; k *= 2)
	{
		// Bitonic merge:
		for (int j = k / 2; j>0; j /= 2)
		{
			int ixj = tid ^ j;

			if (ixj > tid)
			{
				if ((tid & k) == 0)
				{
					if (shared[tid] > shared[ixj])
					{
						swap(shared[tid], shared[ixj]);
					}
				}
				else
				{
					if (shared[tid] < shared[ixj])
					{
						swap(shared[tid], shared[ixj]);
					}
				}
			}

			__syncthreads();
		}
	}

	// Write result.
	values[tid + blockIdx.x * SHARED_SIZE_LIMIT] = shared[tid];


}

its downloaded from
http://developer.download.nvidia.com/compute/cuda/1.1-Beta/x86_website/Data-Parallel_Algorithms.html

2

Why not simply pad the array until it is a power of 2? Use some value that will come out at the end of the sorting, such as INFINITY for floating point values, or MAXINT for integers

3

thank you
I will try