#include<iostream>
#include<vector>
#include<algorithm>
using namespace std;
using ll=long long;
#define all(v) v.begin(),v.end()
#define rall(v) v.rbegin(),v.rend()
template<class T> bool chmax(T &a, T b){if (a < b){a = b;return true;} else return false;}
template<class T> bool chmin(T &a, T b){if (a > b){a = b;return true;} else return false;}

const int MAX=2e6+10;
const ll MOD=998244353;
vector<ll>fac(MAX),finv(MAX),inv(MAX);

void set_fac(){
    fac[0]=fac[1]=1;
    finv[0]=finv[1]=1;
    inv[1]=1;
    for (int i=2;i<MAX;i++){
        fac[i]=fac[i-1]*i%MOD;
        inv[i]=MOD-inv[MOD%i]*(MOD/i)%MOD;
        finv[i]=finv[i-1]*inv[i]%MOD;
    }
}

ll cmb(int n,int r){
    if (r<0||n<r)return 0;
    return fac[n]*(finv[r]*finv[n-r]%MOD)%MOD;
}

ll power(ll x,ll r){
    ll re=1,k=x;
    for(int i=0;i<60;i++){
        if(r>>i&1){
            re=(re*k)%MOD;
        }
        k=k*k%MOD;
    }
    return re;
}

ll modinv(ll x){
    return power(x, MOD-2);
}

int main(){
    ios::sync_with_stdio(false);
    cin.tie(nullptr);
    set_fac();
    ll N,P;
    cin>>N>>P;
    ll ans=fac[N]-1;
    ll K=1,L=1;
    for(int i=P;i<=N;i+=P){
        K*=fac[P-1];
        K%=MOD;
        L*=finv[P-1]*modinv(max(1LL,i-P))%MOD;
        L%=MOD;
        ans-=cmb(N,i)*fac[i-1]%MOD*K%MOD*L%MOD;
        ans%=MOD;
        if(ans<0)ans+=MOD;
    }
    cout<<ans<<"\n";
}