#include <bits/stdc++.h>
using namespace std;
using ll = long long;

const int N = 1e6 + 10;
const ll MOD = 998244353;

struct BIT {
    int n;
    vector <int> bit;
    void init(int _n) {n = _n, bit.resize(n + 1);}
    void add(int x, int val) {
        for (; x <= n; x += (x & -x)) bit[x] += val;
    }
    int query(int x) {
        int re = 0;
        for (; x; x -= (x & -x)) re += bit[x];
        return re;
    }
    int queryRange(int l, int r) {
        return query(r) - query(l - 1);
    }
}tot;

ll pw[N];

int main() {
    ios::sync_with_stdio(0), cin.tie(0);
    int L, B;
    cin >> L >> B;
    vector<int>P(L + 1), rnk(L + 1);
    vector<vector<int> > V(L + 1);

    tot.init(L);

    pw[0] = 1;
    for (int i = 0; i < L; ++ i, pw[i] = pw[i - 1] * B % MOD) cin >> P[i], rnk[P[i]] = i, tot.add(i + 1, 1);
    for (int i = 0; i < L; ++ i) {
        V[i].resize(B + 10);
        for (int j = 0; j < B; ++ j) cin >> V[i][j];
    }

    ll ans = 0, mul = 1;


    for (int i = L - 1; ~i; -- i) {
        int pos = rnk[i] + 1, bit;

        tot.add(pos, -1);
        bit = tot.queryRange(1, pos);

        // cerr << bit << " ";

        BIT inv;
        inv.init(B);

        for (int j = 0; j < B; ++ j) {
            int val = V[pos - 1][j] + 1;
            (ans += inv.queryRange(val, B) * pw[bit] % MOD * pw[bit] % MOD * pw[i - bit] % MOD * mul) %= MOD;
            inv.add(val, 1);
        }

        // cerr << ans << " ";

        (ans += (pw[i - bit] * (pw[i - bit] - 1) >> 1) % MOD * (((B * (B - 1)) >> 1) % MOD) * (pw[bit] * pw[bit] % MOD) % MOD * mul % MOD) %= MOD;

        // cerr << ans << "\n";

        mul = mul * B % MOD;
    }
    cout << ans;
}