马蜂挺好的：

#include<bits/stdc++.h>
#define ll long long
#define lc x<<1
#define rc x<<1|1
using namespace std;
const ll N = 1e5+10;
struct node {
	ll l, r, val;
	ll lazy_add = 0, lazy_time = 1;
} tr[N << 2];
ll n, m, mod, a[N];
void pushup(ll x) {
	tr[x].val = (tr[lc].val + tr[rc].val) % mod;
	return;
}
void pushdown(ll x) {
	if (tr[x].lazy_time != 1) {
		tr[lc].val = ( tr[lc].val * tr[x].lazy_time) % mod;
		tr[rc].val = ( tr[rc].val * tr[x].lazy_time) % mod;
		tr[lc].val %= mod;
		tr[rc].val %= mod;
		tr[lc].lazy_time = tr[rc].lazy_time = tr[x].lazy_time;
		tr[x].lazy_time = 1;
	}
	if (tr[x].lazy_add) {
		tr[lc].val += (tr[x].lazy_add * (tr[lc].r - tr[lc].l + 1)) % mod;
		tr[rc].val += (tr[x].lazy_add * (tr[rc].r - tr[rc].l + 1)) % mod;
		tr[lc].val %= mod;
		tr[rc].val %= mod;
		tr[lc].lazy_add = tr[rc].lazy_add = tr[x].lazy_add;
		tr[x].lazy_add = 0;
	}
	return;
}
void build(ll x, ll l, ll r) {
	tr[x].l = l;
	tr[x].r = r;
	if (l == r) {
		tr[x].val = a[l] % mod;
		return;
	}
	ll m = (l + r) >> 1;
	build(lc, l, m);
	build(rc, m + 1, r);
	pushup(x);
	return;
}
void change_add(ll x, ll l, ll r, ll k) {
	if (l <= tr[x].l && tr[x].r <= r) {
		tr[x].val += (k * (tr[x].r - tr[x].l + 1) % mod) % mod;
		tr[x].val %= mod;
		tr[x].lazy_add += k;
		return;
	}
	pushdown(x);
	ll m = (tr[x].l + tr[x].r) >> 1;
	if (l <= m) change_add(lc, l, r, k);
	if (r > m) change_add(rc, l, r, k);
	pushup(x);
	return;
}
void change_time(ll x, ll l, ll r, ll k) {
	if (l <= tr[x].l && tr[x].r <= r) {
		tr[x].val = (tr[x].val % mod * k % mod) % mod;
		tr[x].lazy_add += k;
		tr[x].lazy_add = 0;
		return;
	}
	pushdown(x);
	ll m = (tr[x].l + tr[x].r) >> 1;
	if (l <= m) change_time(lc, l, r, k);
	if (r > m) change_time(rc, l, r, k);
	pushup(x);
	return;
}
ll fin(ll x, ll l, ll r) {
	if (l <= tr[x].l && tr[x].r <= r) return tr[x].val;
	pushdown(x);
	ll m = (tr[x].l + tr[x].r) >> 1, sum = 0;
	if (l <= m) sum += fin(lc, l, r) % mod;
	if (r > m) sum += fin(rc, l, r) % mod;
	return sum % mod;
}
int main() {
	cin >> n >> m >> mod;
	for (ll i = 1; i <= n; i++) cin >> a[i];
	build(1, 1, n);
	for (ll i = 1; i <= m; i++) {
		ll op;
		cin >> op;
		if (op == 1) {
			ll a, b, c;
			cin >> a >> b >> c;
			change_time(1, a, b, c);
		} else if (op == 2) {
			ll a, b, c;
			cin >> a >> b >> c;
			change_add(1, a, b, c);
		} else {
			ll a, b;
			cin >> a >> b;
			cout << fin(1, a, b) % mod << "\n";
		}
	}
	return 0;
}

我感觉是%的问题，但我都加了的啊