#ifndef ROUND_H_
#define ROUND_H_

#include "ascon.h"
#include "printstate.h"

static inline uint32_t ROR32(uint32_t x, int n) {
  return x >> n | x << (-n & 31);
}

static inline uint64_t ROR(uint64_t x, int n) {
  uint32_t al = (uint32_t)x;
  uint32_t ah = x >> 32;
  uint32_t bl, bh;
  bl = (n % 2) ? ROR32(ah, (n - 1) / 2) : ROR32(al, n / 2);
  bh = (n % 2) ? ROR32(al, (n + 1) / 2) : ROR32(ah, n / 2);
  return (uint64_t)bh << 32 | bl;
}

static inline void ROUND(state_t* s, uint64_t C) {
  state_t t;
  /* addition of round constant */
  s->x[2] ^= C;
  /* printstate(" round constant", s); */
  /* substitution layer */
  s->x[0] ^= s->x[4];
  s->x[4] ^= s->x[3];
  s->x[2] ^= s->x[1];
  /* start of keccak s-box */
  t.x[0] = s->x[0] ^ (~s->x[1] & s->x[2]);
  t.x[1] = s->x[1] ^ (~s->x[2] & s->x[3]);
  t.x[2] = s->x[2] ^ (~s->x[3] & s->x[4]);
  t.x[3] = s->x[3] ^ (~s->x[4] & s->x[0]);
  t.x[4] = s->x[4] ^ (~s->x[0] & s->x[1]);
  /* end of keccak s-box */
  t.x[1] ^= t.x[0];
  t.x[0] ^= t.x[4];
  t.x[3] ^= t.x[2];
  t.x[2] = ~t.x[2];
  /* printstate(" substitution layer", &t); */
  /* linear diffusion layer */
  s->x[0] = t.x[0] ^ ROR(t.x[0], 19) ^ ROR(t.x[0], 28);
  s->x[1] = t.x[1] ^ ROR(t.x[1], 61) ^ ROR(t.x[1], 39);
  s->x[2] = t.x[2] ^ ROR(t.x[2], 1) ^ ROR(t.x[2], 6);
  s->x[3] = t.x[3] ^ ROR(t.x[3], 10) ^ ROR(t.x[3], 17);
  s->x[4] = t.x[4] ^ ROR(t.x[4], 7) ^ ROR(t.x[4], 41);
  printstate(" round output", s);
}

#endif /* ROUND_H_ */