template <typename T>
__global__ void cmtest(const T * __restrict__ a, const T * __restrict__ b, T * __restrict__ c, int num_arr, int len, int stride_a, int stride_b, int stride_c){
    int idx=threadIdx.x+blockDim.x*blockIdx.x;
    while (idx < num_arr){
        smerge(a+idx, b+idx, c+idx, len, len, stride_a, stride_b, stride_c);
        idx += blockDim.x*gridDim.x;
    }
}