#include <bits/stdc++.h>

using namespace std;

using ull = unsigned long long;
using L = __uint128_t;

int mod;
void add(int &x, int y) { if ((x += y - mod) < 0) x += mod; }
void sub(int &x, int y) { if ((x -= y) < 0) x += mod; }
int neg(int x) { return x ? mod - x : 0; }
struct FastMod {
  ull b, m;
  FastMod(ull b) : b(b), m(ull((L(1) << 64) / b)) {}
  ull operator()(ull a) {
    ull q = (ull)((L(m) * a) >> 64);
    ull r = a - q * b; // can be proven that 0 <= r < 2*b
    return r >= b ? r - b : r;
  }
} R(2);
int mpow(int a, int b) {
  int ret = 1;
  for (; b; b >>= 1) {
    if (b & 1) ret = R((ull)ret * a);
    a = R((ull)a * a);
  }
  return ret;
}

const int N = 1e7;
const int BRUTE_N2_LIMIT = 50;

int n, q, qinv;
int pw[N + 5], qk[N + 5], qfac[N + 5], qifac[N + 5];
int f[N + 5];
int bsPw[32769], gsPw[32768];
int query(int n) { return R((ull)bsPw[n & 32767] * gsPw[n >> 15]); }
int ans;

int main() {
  scanf("%d%d%d", &n, &q, &mod), R = FastMod(mod), qinv = mpow(q, mod - 2);
  bsPw[0] = gsPw[0] = 1;
  for (int i = 1; i <= (1 << 15); ++i) add(bsPw[i] = bsPw[i - 1], bsPw[i - 1]), add(bsPw[i], bsPw[i - 1]);
  for (int i = 1; i < (1 << 15); ++i) gsPw[i] = R((ull)gsPw[i - 1] * bsPw[1 << 15]);
  pw[0] = 1;
  for (int i = 1; i <= n; ++i) pw[i] = R((ull)pw[i - 1] * q);
  qfac[0] = 1;
  for (int i = 1; i <= n; ++i) qfac[i] = R((ull)qfac[i - 1] * (qk[i] = pw[i] - 1));
  qifac[n] = mpow(qfac[n], mod - 2);
  for (int i = n; i; --i) qifac[i - 1] = R((ull)qifac[i] * qk[i]);
  f[0] = 1;
  for (int i = 1; i <= min(BRUTE_N2_LIMIT, n); ++i)
    f[i] = R((ull)f[i - 1] * (pw[n] - pw[n - i] + mod));
  for (int i = 1; i <= min(BRUTE_N2_LIMIT, n); ++i)
    for (int j = 0; j < i; ++j) f[i] = R(f[i] + (mod - f[j]) * R(R((ull)qfac[i] * qifac[j]) * qifac[i - j]));
  for (int i = BRUTE_N2_LIMIT + 1; i <= n; ++i) {
    int coe1 = qinv;
    add(coe1, pw[n]), sub(coe1, pw[i - 1]), sub(coe1, pw[i - 2]), sub(coe1, pw[n - i]);
    int coe2 = R((ull)(pw[n - 1] - pw[i - 3] + mod) * qk[i - 1]);
    f[i] = R((ull)coe1 * f[i - 1] + (ull)coe2 * f[i - 2]);
  }
  int t = 1;
  for (int i = 1; i <= n; ++i) t = (ull)t * q % (mod - 1);
  for (int i = 0, c = 1; i <= n; ++i)
    ans = R(ans + R(query(c) * R(qifac[i] * R((ull)qifac[n - i] * f[n - i])))),
    c = (ull)c * t % (mod - 1);
  ans = R((ull)ans * qfac[n]);
  printf("%d\n", ans);
}