#include<bits/stdc++.h>
namespace {
#pragma GCC diagnostic ignored "-Wunused-function"
#include<atcoder/all>
#pragma GCC diagnostic warning "-Wunused-function"
using namespace std;
using namespace atcoder;
#define rep(i,n) for(int i = 0; i < (int)(n); i++)
#define rrep(i,n) for(int i = (int)(n) - 1; i >= 0; i--)
#define all(x) begin(x), end(x)
#define rall(x) rbegin(x), rend(x)
template<class T> bool chmax(T& a, const T& b) { if (a < b) { a = b; return true; } else return false; }
template<class T> bool chmin(T& a, const T& b) { if (b < a) { a = b; return true; } else return false; }
using ll = long long;
using P = pair<int,int>;
using VI = vector<int>;
using VVI = vector<VI>;
using VL = vector<ll>;
using VVL = vector<VL>;
using mint = modint998244353;
using dmint = modint;

int p;
constexpr int FACT_SIZE = 1000000;
dmint Fact[FACT_SIZE + 1];
dmint iFact[FACT_SIZE + 1];
const auto fact_init = [] {
  int FACT_SIZE = p - 1;
    Fact[0] = dmint::raw(1);
    for(int i = 1; i <= FACT_SIZE; ++i) {
        Fact[i] = Fact[i-1] * i;
    }
    iFact[FACT_SIZE] = Fact[FACT_SIZE].inv();
    for(int i = FACT_SIZE; i; --i) {
        iFact[i-1] = iFact[i] * i;
    }
    return false;
};

dmint comb(int n, int k) {
    if (k == 0) return dmint::raw(1);
    assert(n >= 0 && k >= 0);
    if (k > n) return dmint::raw(0);
    return Fact[n] * iFact[n - k] * iFact[k];
}

dmint icomb(int n, int k) {
    return iFact[n] * Fact[n - k] * Fact[k];
}

dmint fact(int n) {return Fact[n];}
dmint perm(int n, int k) {
    assert(0 <= n);
    return Fact[n] * iFact[n - k];
}

} int main() {
  ios::sync_with_stdio(false);
  cin.tie(0);
  ll n;
  cin >> n >> p;
  dmint::set_mod(p);
  fact_init();
  int r = internal::primitive_root_constexpr(p);
  VI val2e(p), e2val(p);
  {
    int v = 1;
    for (int e = 0; e < p - 1; e++) {
      val2e[v] = e;
      e2val[e] = v;
      v = ll(v * r) % p;
    }
    assert(v == 1);
  }
  vector<mint> f(p - 1);
  f[0] = 1;
  while (n) {
    int m = n % p; n /= p;
    vector<mint> g(p - 1);
    rep(i, m + 1) g[val2e[comb(m, i).val()]]++;
    f = convolution(f, g);
    while (ssize(f) > p - 1) {
      f[ssize(f) - 1 - (p - 1)] += f[ssize(f) - 1];
      f.pop_back();
    }
  }
  mint ans;
  rep(e, p - 1) ans += e2val[e] * f[e];
  cout << ans.val() << '\n';
}