I am trying to writ the CUDA C++ equivalent code for the following C++ code. I have tried several ways but I can’t find a way to implement this code in parallel.

void func(result, parents, parentslength, matchparents, matchlength) {
  int j = 0;
  for (int i = 0;  i < parentslength;  i++) {
    if (j < matchlength  &&  parents[i] == matchparents[j]) {
      result[i] = j;
      ++j;
    }
    else {
      result[i] = -1;
    }
  }
}

This is an example of the inputs and expected output:

parentslength = 12,
parents = [0, 0, 0, 0, 1, 1, 1, 2, 2, 2, 3, 3],
matchlength = 6,
matchparents = [0, 0, 0, 2, 2, 3],
result = [0, 1, 2, -1, -1, -1, -1, 3, 4, -1, 5, -1]