#include <iostream>
#include <vector>
#include <cstring>
#include <algorithm>
using namespace std;

struct{char buf[1<<20],*l=buf,*r=l;
operator char(){return l==r&&(r=(l=buf)+fread(buf,1,1<<20,stdin),l==r)?-1:*l++;}
template<typename T>operator T(){
    T x=0;char f=0,c;
    do if((c=*this)=='-')f=1;while(c<'0'||c>'9');
    while(c>='0'&&c<='9')x=x*10+(c^48),c=*this;
    return f?-x:x;
}}in;void out(char c){putchar(c);}
template<typename T>void out(T x){
    static signed stk[39],tp;
    if(x<0)out('-'),x=-x;
    do stk[tp++]=x%10;while(x/=10);
    while(tp)putchar(stk[--tp]^48);
}template<typename T,typename...Args>
void out(T x,Args...args){out(x);out(args...);}

typedef long long ll;
const int N = 5005,M = 1000005,mod = 998244353;
vector<int> c[M];
int a[N];
int n,K;

int f[N][N][2];
int g[N],gg[N];
int fac[N];

int main(){
    n = in,K = in;

    fac[0] = 1;
    for(int k=1;k<=n;k++)
        fac[k] = 1ull*fac[k-1]*k%mod;

    for(int k=1;k<=n;k++){
        a[k] = in;
        c[a[k]%K].push_back(a[k]);
    }
    f[1][0][0] = 1;
    for(int k=2;k<=n;k++)
        for(int j=0;j<k;j++){
            f[k][j][0] = (f[k-1][j][0]+f[k-1][j][1])%mod;
            f[k][j+1][1] = (f[k-1][j][0]*2ull+f[k-1][j][1])%mod;
        }
    g[0] = 1;
    auto add = [](int v){
        memset(gg,0,sizeof(int)*n);
        for(int k=0;k<v;k++)
            for(int j=k;j<n;j++)
                gg[j] = (gg[j]+1ull*g[j-k]*(f[v][k][0]+f[v][k][1]))%mod;
        for(int k=0;k<n;k++)
            g[k] = gg[k];
    };
    for(int k=0;k<K;k++){
        if(c[k].empty())
            continue;
        int i = 1;
        for(int j=1;j<c[k].size();j++,i++)
            if(c[k][j-1]+K!=c[k][j]){
                add(i);
                i = 0;
            }
        add(i);
    }
    int ans = 0;
    for(int k=0;k<n;k++)
        ans = (ans+1ull*mod*mod+((k&1)?-1ull:1ull)*g[k]*fac[n-k])%mod;
    out(ans);
    return 0;
}