cuda_helpers.h

#pragma once

#define CUDA_1D_KERNEL_LOOP(i, n)                                \
  for (int i = (blockIdx.x * blockDim.x) + threadIdx.x; i < (n); \
       i += (blockDim.x * gridDim.x))

template <typename integer>
constexpr __host__ __device__ inline integer ceil_div(integer n, integer m) {
  return (n + m - 1) / m;
}