const int N = *;

const int threadsPerBlock = ;

__global__ void dot(float *a, float *b, float *c)

{

    __shared__  float  cache[threadsPerBlock];

    int tid = threadIdx.x + blockId.x*blockDim.x;

    int cacheIndex = threadIdx.x;

    float temp = ;

    while(tid<N){

        temp += a[tid]*b[tid];

        tid += blockDim.x * gridDim.x;

    }

    cache[cacheIndex] = temp;

    __syncthreads();

    int i = blockDim.x/;

    while(i != ){

        if(cacheIndex < i)

            cache[cacheIndex] += cache[cacheIndex + i];

        __syncthreads();

        i /= ;

    }

    if(cacheIndex == )

        c[blockIdex.x] = cache[];

}