結果
| 問題 | No.2705 L to R Graph (Another ver.) |
| ユーザー |
 maspy
|
| 提出日時 | 2024-03-29 22:32:37 |
| 言語 | C++23 (gcc 12.3.0 + boost 1.83.0) |
| 結果 |
AC
|
| 実行時間 | 1,562 ms / 3,000 ms |
| コード長 | 29,920 bytes |
| コンパイル時間 | 5,941 ms |
| コンパイル使用メモリ | 327,736 KB |
| 実行使用メモリ | 7,004 KB |
| 最終ジャッジ日時 | 2024-09-30 16:28:46 |
| 合計ジャッジ時間 | 61,490 ms |
|
ジャッジサーバーID （参考情報） |
judge5 / judge1 |
（要ログイン）
テストケース
テストケース表示| 入力 | 結果 | 実行時間 実行使用メモリ |
|---|---|---|
| testcase_00 | AC | 2 ms
5,248 KB |
| testcase_01 | AC | 2 ms
5,248 KB |
| testcase_02 | AC | 2 ms
5,248 KB |
| testcase_03 | AC | 2 ms
5,248 KB |
| testcase_04 | AC | 2 ms
5,248 KB |
| testcase_05 | AC | 2 ms
5,248 KB |
| testcase_06 | AC | 2 ms
5,248 KB |
| testcase_07 | AC | 2 ms
5,248 KB |
| testcase_08 | AC | 1,064 ms
6,384 KB |
| testcase_09 | AC | 1,306 ms
6,704 KB |
| testcase_10 | AC | 799 ms
5,836 KB |
| testcase_11 | AC | 967 ms
6,152 KB |
| testcase_12 | AC | 608 ms
5,376 KB |
| testcase_13 | AC | 296 ms
5,248 KB |
| testcase_14 | AC | 1,232 ms
6,568 KB |
| testcase_15 | AC | 70 ms
5,248 KB |
| testcase_16 | AC | 612 ms
5,504 KB |
| testcase_17 | AC | 191 ms
5,248 KB |
| testcase_18 | AC | 1,148 ms
6,416 KB |
| testcase_19 | AC | 456 ms
5,248 KB |
| testcase_20 | AC | 241 ms
5,248 KB |
| testcase_21 | AC | 910 ms
5,956 KB |
| testcase_22 | AC | 892 ms
5,944 KB |
| testcase_23 | AC | 1,458 ms
6,856 KB |
| testcase_24 | AC | 1,460 ms
6,860 KB |
| testcase_25 | AC | 1,427 ms
6,852 KB |
| testcase_26 | AC | 1,474 ms
6,860 KB |
| testcase_27 | AC | 1,372 ms
6,844 KB |
| testcase_28 | AC | 1,359 ms
6,716 KB |
| testcase_29 | AC | 1,322 ms
6,580 KB |
| testcase_30 | AC | 1,360 ms
6,716 KB |
| testcase_31 | AC | 1,397 ms
6,704 KB |
| testcase_32 | AC | 1,370 ms
6,844 KB |
| testcase_33 | AC | 1,352 ms
6,712 KB |
| testcase_34 | AC | 1,463 ms
6,856 KB |
| testcase_35 | AC | 1,339 ms
6,708 KB |
| testcase_36 | AC | 1,457 ms
6,732 KB |
| testcase_37 | AC | 1,507 ms
6,868 KB |
| testcase_38 | AC | 1,342 ms
6,704 KB |
| testcase_39 | AC | 1,466 ms
6,864 KB |
| testcase_40 | AC | 1,524 ms
6,852 KB |
| testcase_41 | AC | 1,532 ms
7,000 KB |
| testcase_42 | AC | 1,452 ms
6,852 KB |
| testcase_43 | AC | 1,538 ms
7,000 KB |
| testcase_44 | AC | 1,535 ms
6,968 KB |
| testcase_45 | AC | 1,545 ms
7,000 KB |
| testcase_46 | AC | 1,536 ms
6,988 KB |
| testcase_47 | AC | 1,540 ms
6,996 KB |
| testcase_48 | AC | 1,537 ms
6,872 KB |
| testcase_49 | AC | 1,536 ms
7,000 KB |
| testcase_50 | AC | 1,536 ms
7,004 KB |
| testcase_51 | AC | 1,533 ms
6,876 KB |
| testcase_52 | AC | 1,562 ms
7,000 KB |
ソースコード
#line 1 "/home/maspy/compro/library/my_template.hpp"
#if defined(LOCAL)
#include <my_template_compiled.hpp>
#else
// https://codeforces.com/blog/entry/96344
#pragma GCC optimize("Ofast,unroll-loops")
// いまの CF だとこれ入れると動かない？
// #pragma GCC target("avx2,popcnt")
#include <bits/stdc++.h>
using namespace std;
using ll = long long;
using u32 = unsigned int;
using u64 = unsigned long long;
using i128 = __int128;
using u128 = unsigned __int128;
using f128 = __float128;
template <class T>
constexpr T infty = 0;
template <>
constexpr int infty<int> = 1'000'000'000;
template <>
constexpr ll infty<ll> = ll(infty<int>) * infty<int> * 2;
template <>
constexpr u32 infty<u32> = infty<int>;
template <>
constexpr u64 infty<u64> = infty<ll>;
template <>
constexpr i128 infty<i128> = i128(infty<ll>) * infty<ll>;
template <>
constexpr double infty<double> = infty<ll>;
template <>
constexpr long double infty<long double> = infty<ll>;
using pi = pair<ll, ll>;
using vi = vector<ll>;
template <class T>
using vc = vector<T>;
template <class T>
using vvc = vector<vc<T>>;
template <class T>
using vvvc = vector<vvc<T>>;
template <class T>
using vvvvc = vector<vvvc<T>>;
template <class T>
using vvvvvc = vector<vvvvc<T>>;
template <class T>
using pq = priority_queue<T>;
template <class T>
using pqg = priority_queue<T, vector<T>, greater<T>>;
#define vv(type, name, h, ...) \
vector<vector<type>> name(h, vector<type>(__VA_ARGS__))
#define vvv(type, name, h, w, ...) \
vector<vector<vector<type>>> name( \
h, vector<vector<type>>(w, vector<type>(__VA_ARGS__)))
#define vvvv(type, name, a, b, c, ...) \
vector<vector<vector<vector<type>>>> name( \
a, vector<vector<vector<type>>>( \
b, vector<vector<type>>(c, vector<type>(__VA_ARGS__))))
// https://trap.jp/post/1224/
#define FOR1(a) for (ll _ = 0; _ < ll(a); ++_)
#define FOR2(i, a) for (ll i = 0; i < ll(a); ++i)
#define FOR3(i, a, b) for (ll i = a; i < ll(b); ++i)
#define FOR4(i, a, b, c) for (ll i = a; i < ll(b); i += (c))
#define FOR1_R(a) for (ll i = (a)-1; i >= ll(0); --i)
#define FOR2_R(i, a) for (ll i = (a)-1; i >= ll(0); --i)
#define FOR3_R(i, a, b) for (ll i = (b)-1; i >= ll(a); --i)
#define overload4(a, b, c, d, e, ...) e
#define overload3(a, b, c, d, ...) d
#define FOR(...) overload4(__VA_ARGS__, FOR4, FOR3, FOR2, FOR1)(__VA_ARGS__)
#define FOR_R(...) overload3(__VA_ARGS__, FOR3_R, FOR2_R, FOR1_R)(__VA_ARGS__)
#define FOR_subset(t, s) \
for (ll t = (s); t >= 0; t = (t == 0 ? -1 : (t - 1) & (s)))
#define all(x) x.begin(), x.end()
#define len(x) ll(x.size())
#define elif else if
#define eb emplace_back
#define mp make_pair
#define mt make_tuple
#define fi first
#define se second
#define stoi stoll
int popcnt(int x) { return __builtin_popcount(x); }
int popcnt(u32 x) { return __builtin_popcount(x); }
int popcnt(ll x) { return __builtin_popcountll(x); }
int popcnt(u64 x) { return __builtin_popcountll(x); }
int popcnt_mod_2(int x) { return __builtin_parity(x); }
int popcnt_mod_2(u32 x) { return __builtin_parity(x); }
int popcnt_mod_2(ll x) { return __builtin_parityll(x); }
int popcnt_mod_2(u64 x) { return __builtin_parityll(x); }
// (0, 1, 2, 3, 4) -> (-1, 0, 1, 1, 2)
int topbit(int x) { return (x == 0 ? -1 : 31 - __builtin_clz(x)); }
int topbit(u32 x) { return (x == 0 ? -1 : 31 - __builtin_clz(x)); }
int topbit(ll x) { return (x == 0 ? -1 : 63 - __builtin_clzll(x)); }
int topbit(u64 x) { return (x == 0 ? -1 : 63 - __builtin_clzll(x)); }
// (0, 1, 2, 3, 4) -> (-1, 0, 1, 0, 2)
int lowbit(int x) { return (x == 0 ? -1 : __builtin_ctz(x)); }
int lowbit(u32 x) { return (x == 0 ? -1 : __builtin_ctz(x)); }
int lowbit(ll x) { return (x == 0 ? -1 : __builtin_ctzll(x)); }
int lowbit(u64 x) { return (x == 0 ? -1 : __builtin_ctzll(x)); }
template <typename T>
T floor(T a, T b) {
return a / b - (a % b && (a ^ b) < 0);
}
template <typename T>
T ceil(T x, T y) {
return floor(x + y - 1, y);
}
template <typename T>
T bmod(T x, T y) {
return x - y * floor(x, y);
}
template <typename T>
pair<T, T> divmod(T x, T y) {
T q = floor(x, y);
return {q, x - q * y};
}
template <typename T, typename U>
T SUM(const vector<U> &A) {
T sm = 0;
for (auto &&a: A) sm += a;
return sm;
}
#define MIN(v) *min_element(all(v))
#define MAX(v) *max_element(all(v))
#define LB(c, x) distance((c).begin(), lower_bound(all(c), (x)))
#define UB(c, x) distance((c).begin(), upper_bound(all(c), (x)))
#define UNIQUE(x) \
sort(all(x)), x.erase(unique(all(x)), x.end()), x.shrink_to_fit()
template <typename T>
T POP(deque<T> &que) {
T a = que.front();
que.pop_front();
return a;
}
template <typename T>
T POP(pq<T> &que) {
T a = que.top();
que.pop();
return a;
}
template <typename T>
T POP(pqg<T> &que) {
T a = que.top();
que.pop();
return a;
}
template <typename T>
T POP(vc<T> &que) {
T a = que.back();
que.pop_back();
return a;
}
template <typename F>
ll binary_search(F check, ll ok, ll ng, bool check_ok = true) {
if (check_ok) assert(check(ok));
while (abs(ok - ng) > 1) {
auto x = (ng + ok) / 2;
(check(x) ? ok : ng) = x;
}
return ok;
}
template <typename F>
double binary_search_real(F check, double ok, double ng, int iter = 100) {
FOR(iter) {
double x = (ok + ng) / 2;
(check(x) ? ok : ng) = x;
}
return (ok + ng) / 2;
}
template <class T, class S>
inline bool chmax(T &a, const S &b) {
return (a < b ? a = b, 1 : 0);
}
template <class T, class S>
inline bool chmin(T &a, const S &b) {
return (a > b ? a = b, 1 : 0);
}
// ? は -1
vc<int> s_to_vi(const string &S, char first_char) {
vc<int> A(S.size());
FOR(i, S.size()) { A[i] = (S[i] != '?' ? S[i] - first_char : -1); }
return A;
}
template <typename T, typename U>
vector<T> cumsum(vector<U> &A, int off = 1) {
int N = A.size();
vector<T> B(N + 1);
FOR(i, N) { B[i + 1] = B[i] + A[i]; }
if (off == 0) B.erase(B.begin());
return B;
}
// stable sort
template <typename T>
vector<int> argsort(const vector<T> &A) {
vector<int> ids(len(A));
iota(all(ids), 0);
sort(all(ids),
[&](int i, int j) { return (A[i] == A[j] ? i < j : A[i] < A[j]); });
return ids;
}
// A[I[0]], A[I[1]], ...
template <typename T>
vc<T> rearrange(const vc<T> &A, const vc<int> &I) {
vc<T> B(len(I));
FOR(i, len(I)) B[i] = A[I[i]];
return B;
}
#endif
#line 1 "/home/maspy/compro/library/other/io.hpp"
#define FASTIO
#include <unistd.h>
// https://judge.yosupo.jp/submission/21623
namespace fastio {
static constexpr uint32_t SZ = 1 << 17;
char ibuf[SZ];
char obuf[SZ];
char out[100];
// pointer of ibuf, obuf
uint32_t pil = 0, pir = 0, por = 0;
struct Pre {
char num[10000][4];
constexpr Pre() : num() {
for (int i = 0; i < 10000; i++) {
int n = i;
for (int j = 3; j >= 0; j--) {
num[i][j] = n % 10 | '0';
n /= 10;
}
}
}
} constexpr pre;
inline void load() {
memcpy(ibuf, ibuf + pil, pir - pil);
pir = pir - pil + fread(ibuf + pir - pil, 1, SZ - pir + pil, stdin);
pil = 0;
if (pir < SZ) ibuf[pir++] = '\n';
}
inline void flush() {
fwrite(obuf, 1, por, stdout);
por = 0;
}
void rd(char &c) {
do {
if (pil + 1 > pir) load();
c = ibuf[pil++];
} while (isspace(c));
}
void rd(string &x) {
x.clear();
char c;
do {
if (pil + 1 > pir) load();
c = ibuf[pil++];
} while (isspace(c));
do {
x += c;
if (pil == pir) load();
c = ibuf[pil++];
} while (!isspace(c));
}
template <typename T>
void rd_real(T &x) {
string s;
rd(s);
x = stod(s);
}
template <typename T>
void rd_integer(T &x) {
if (pil + 100 > pir) load();
char c;
do
c = ibuf[pil++];
while (c < '-');
bool minus = 0;
if constexpr (is_signed<T>::value || is_same_v<T, i128>) {
if (c == '-') { minus = 1, c = ibuf[pil++]; }
}
x = 0;
while ('0' <= c) { x = x * 10 + (c & 15), c = ibuf[pil++]; }
if constexpr (is_signed<T>::value || is_same_v<T, i128>) {
if (minus) x = -x;
}
}
void rd(int &x) { rd_integer(x); }
void rd(ll &x) { rd_integer(x); }
void rd(i128 &x) { rd_integer(x); }
void rd(u32 &x) { rd_integer(x); }
void rd(u64 &x) { rd_integer(x); }
void rd(u128 &x) { rd_integer(x); }
void rd(double &x) { rd_real(x); }
void rd(long double &x) { rd_real(x); }
void rd(f128 &x) { rd_real(x); }
template <class T, class U>
void rd(pair<T, U> &p) {
return rd(p.first), rd(p.second);
}
template <size_t N = 0, typename T>
void rd_tuple(T &t) {
if constexpr (N < std::tuple_size<T>::value) {
auto &x = std::get<N>(t);
rd(x);
rd_tuple<N + 1>(t);
}
}
template <class... T>
void rd(tuple<T...> &tpl) {
rd_tuple(tpl);
}
template <size_t N = 0, typename T>
void rd(array<T, N> &x) {
for (auto &d: x) rd(d);
}
template <class T>
void rd(vc<T> &x) {
for (auto &d: x) rd(d);
}
void read() {}
template <class H, class... T>
void read(H &h, T &... t) {
rd(h), read(t...);
}
void wt(const char c) {
if (por == SZ) flush();
obuf[por++] = c;
}
void wt(const string s) {
for (char c: s) wt(c);
}
void wt(const char *s) {
size_t len = strlen(s);
for (size_t i = 0; i < len; i++) wt(s[i]);
}
template <typename T>
void wt_integer(T x) {
if (por > SZ - 100) flush();
if (x < 0) { obuf[por++] = '-', x = -x; }
int outi;
for (outi = 96; x >= 10000; outi -= 4) {
memcpy(out + outi, pre.num[x % 10000], 4);
x /= 10000;
}
if (x >= 1000) {
memcpy(obuf + por, pre.num[x], 4);
por += 4;
} else if (x >= 100) {
memcpy(obuf + por, pre.num[x] + 1, 3);
por += 3;
} else if (x >= 10) {
int q = (x * 103) >> 10;
obuf[por] = q | '0';
obuf[por + 1] = (x - q * 10) | '0';
por += 2;
} else
obuf[por++] = x | '0';
memcpy(obuf + por, out + outi + 4, 96 - outi);
por += 96 - outi;
}
template <typename T>
void wt_real(T x) {
ostringstream oss;
oss << fixed << setprecision(15) << double(x);
string s = oss.str();
wt(s);
}
void wt(int x) { wt_integer(x); }
void wt(ll x) { wt_integer(x); }
void wt(i128 x) { wt_integer(x); }
void wt(u32 x) { wt_integer(x); }
void wt(u64 x) { wt_integer(x); }
void wt(u128 x) { wt_integer(x); }
void wt(double x) { wt_real(x); }
void wt(long double x) { wt_real(x); }
void wt(f128 x) { wt_real(x); }
template <class T, class U>
void wt(const pair<T, U> val) {
wt(val.first);
wt(' ');
wt(val.second);
}
template <size_t N = 0, typename T>
void wt_tuple(const T t) {
if constexpr (N < std::tuple_size<T>::value) {
if constexpr (N > 0) { wt(' '); }
const auto x = std::get<N>(t);
wt(x);
wt_tuple<N + 1>(t);
}
}
template <class... T>
void wt(tuple<T...> tpl) {
wt_tuple(tpl);
}
template <class T, size_t S>
void wt(const array<T, S> val) {
auto n = val.size();
for (size_t i = 0; i < n; i++) {
if (i) wt(' ');
wt(val[i]);
}
}
template <class T>
void wt(const vector<T> val) {
auto n = val.size();
for (size_t i = 0; i < n; i++) {
if (i) wt(' ');
wt(val[i]);
}
}
void print() { wt('\n'); }
template <class Head, class... Tail>
void print(Head &&head, Tail &&... tail) {
wt(head);
if (sizeof...(Tail)) wt(' ');
print(forward<Tail>(tail)...);
}
// gcc expansion. called automaticall after main.
void __attribute__((destructor)) _d() { flush(); }
} // namespace fastio
using fastio::read;
using fastio::print;
using fastio::flush;
#define INT(...) \
int __VA_ARGS__; \
read(__VA_ARGS__)
#define LL(...) \
ll __VA_ARGS__; \
read(__VA_ARGS__)
#define U32(...) \
u32 __VA_ARGS__; \
read(__VA_ARGS__)
#define U64(...) \
u64 __VA_ARGS__; \
read(__VA_ARGS__)
#define STR(...) \
string __VA_ARGS__; \
read(__VA_ARGS__)
#define CHAR(...) \
char __VA_ARGS__; \
read(__VA_ARGS__)
#define DBL(...) \
double __VA_ARGS__; \
read(__VA_ARGS__)
#define VEC(type, name, size) \
vector<type> name(size); \
read(name)
#define VV(type, name, h, w) \
vector<vector<type>> name(h, vector<type>(w)); \
read(name)
void YES(bool t = 1) { print(t ? "YES" : "NO"); }
void NO(bool t = 1) { YES(!t); }
void Yes(bool t = 1) { print(t ? "Yes" : "No"); }
void No(bool t = 1) { Yes(!t); }
void yes(bool t = 1) { print(t ? "yes" : "no"); }
void no(bool t = 1) { yes(!t); }
#line 3 "main.cpp"
#line 2 "/home/maspy/compro/library/mod/dynamic_modint.hpp"
#line 2 "/home/maspy/compro/library/mod/modint_common.hpp"
struct has_mod_impl {
template <class T>
static auto check(T &&x) -> decltype(x.get_mod(), std::true_type{});
template <class T>
static auto check(...) -> std::false_type;
};
template <class T>
class has_mod : public decltype(has_mod_impl::check<T>(std::declval<T>())) {};
template <typename mint>
mint inv(int n) {
static const int mod = mint::get_mod();
static vector<mint> dat = {0, 1};
assert(0 <= n);
if (n >= mod) n %= mod;
while (len(dat) <= n) {
int k = len(dat);
int q = (mod + k - 1) / k;
dat.eb(dat[k * q - mod] * mint::raw(q));
}
return dat[n];
}
template <typename mint>
mint fact(int n) {
static const int mod = mint::get_mod();
assert(0 <= n && n < mod);
static vector<mint> dat = {1, 1};
while (len(dat) <= n) dat.eb(dat[len(dat) - 1] * mint::raw(len(dat)));
return dat[n];
}
template <typename mint>
mint fact_inv(int n) {
static vector<mint> dat = {1, 1};
if (n < 0) return mint(0);
while (len(dat) <= n) dat.eb(dat[len(dat) - 1] * inv<mint>(len(dat)));
return dat[n];
}
template <class mint, class... Ts>
mint fact_invs(Ts... xs) {
return (mint(1) * ... * fact_inv<mint>(xs));
}
template <typename mint, class Head, class... Tail>
mint multinomial(Head &&head, Tail &&... tail) {
return fact<mint>(head) * fact_invs<mint>(std::forward<Tail>(tail)...);
}
template <typename mint>
mint C_dense(int n, int k) {
static vvc<mint> C;
static int H = 0, W = 0;
auto calc = [&](int i, int j) -> mint {
if (i == 0) return (j == 0 ? mint(1) : mint(0));
return C[i - 1][j] + (j ? C[i - 1][j - 1] : 0);
};
if (W <= k) {
FOR(i, H) {
C[i].resize(k + 1);
FOR(j, W, k + 1) { C[i][j] = calc(i, j); }
}
W = k + 1;
}
if (H <= n) {
C.resize(n + 1);
FOR(i, H, n + 1) {
C[i].resize(W);
FOR(j, W) { C[i][j] = calc(i, j); }
}
H = n + 1;
}
return C[n][k];
}
template <typename mint, bool large = false, bool dense = false>
mint C(ll n, ll k) {
assert(n >= 0);
if (k < 0 || n < k) return 0;
if constexpr (dense) return C_dense<mint>(n, k);
if constexpr (!large) return multinomial<mint>(n, k, n - k);
k = min(k, n - k);
mint x(1);
FOR(i, k) x *= mint(n - i);
return x * fact_inv<mint>(k);
}
template <typename mint, bool large = false>
mint C_inv(ll n, ll k) {
assert(n >= 0);
assert(0 <= k && k <= n);
if (!large) return fact_inv<mint>(n) * fact<mint>(k) * fact<mint>(n - k);
return mint(1) / C<mint, 1>(n, k);
}
// [x^d](1-x)^{-n}
template <typename mint, bool large = false, bool dense = false>
mint C_negative(ll n, ll d) {
assert(n >= 0);
if (d < 0) return mint(0);
if (n == 0) { return (d == 0 ? mint(1) : mint(0)); }
return C<mint, large, dense>(n + d - 1, d);
}
#line 2 "/home/maspy/compro/library/mod/primitive_root.hpp"
#line 2 "/home/maspy/compro/library/nt/factor.hpp"
#line 2 "/home/maspy/compro/library/random/base.hpp"
u64 RNG_64() {
static uint64_t x_
= uint64_t(chrono::duration_cast<chrono::nanoseconds>(
chrono::high_resolution_clock::now().time_since_epoch())
.count())
* 10150724397891781847ULL;
x_ ^= x_ << 7;
return x_ ^= x_ >> 9;
}
u64 RNG(u64 lim) { return RNG_64() % lim; }
ll RNG(ll l, ll r) { return l + RNG_64() % (r - l); }
#line 2 "/home/maspy/compro/library/mod/mongomery_modint.hpp"
// odd mod.
// x の代わりに rx を持つ
template <int id, typename U1, typename U2>
struct Mongomery_modint {
using mint = Mongomery_modint;
inline static U1 m, r, n2;
static constexpr int W = numeric_limits<U1>::digits;
static void set_mod(U1 mod) {
assert(mod & 1 && mod <= U1(1) << (W - 2));
m = mod, n2 = -U2(m) % m, r = m;
FOR(5) r *= 2 - m * r;
r = -r;
assert(r * m == U1(-1));
}
static U1 reduce(U2 b) { return (b + U2(U1(b) * r) * m) >> W; }
U1 x;
Mongomery_modint() : x(0) {}
Mongomery_modint(U1 x) : x(reduce(U2(x) * n2)){};
U1 val() const {
U1 y = reduce(x);
return y >= m ? y - m : y;
}
mint &operator+=(mint y) {
x = ((x += y.x) >= m ? x - m : x);
return *this;
}
mint &operator-=(mint y) {
x -= (x >= y.x ? y.x : y.x - m);
return *this;
}
mint &operator*=(mint y) {
x = reduce(U2(x) * y.x);
return *this;
}
mint operator+(mint y) const { return mint(*this) += y; }
mint operator-(mint y) const { return mint(*this) -= y; }
mint operator*(mint y) const { return mint(*this) *= y; }
bool operator==(mint y) const {
return (x >= m ? x - m : x) == (y.x >= m ? y.x - m : y.x);
}
bool operator!=(mint y) const { return not operator==(y); }
mint pow(ll n) const {
assert(n >= 0);
mint y = 1, z = *this;
for (; n; n >>= 1, z *= z)
if (n & 1) y *= z;
return y;
}
};
template <int id>
using Mongomery_modint_32 = Mongomery_modint<id, u32, u64>;
template <int id>
using Mongomery_modint_64 = Mongomery_modint<id, u64, u128>;
#line 3 "/home/maspy/compro/library/nt/primetest.hpp"
bool primetest(const u64 x) {
assert(x < u64(1) << 62);
if (x == 2 or x == 3 or x == 5 or x == 7) return true;
if (x % 2 == 0 or x % 3 == 0 or x % 5 == 0 or x % 7 == 0) return false;
if (x < 121) return x > 1;
const u64 d = (x - 1) >> lowbit(x - 1);
using mint = Mongomery_modint_64<202311020>;
mint::set_mod(x);
const mint one(u64(1)), minus_one(x - 1);
auto ok = [&](u64 a) -> bool {
auto y = mint(a).pow(d);
u64 t = d;
while (y != one && y != minus_one && t != x - 1) y *= y, t <<= 1;
if (y != minus_one && t % 2 == 0) return false;
return true;
};
if (x < (u64(1) << 32)) {
for (u64 a: {2, 7, 61})
if (!ok(a)) return false;
} else {
for (u64 a: {2, 325, 9375, 28178, 450775, 9780504, 1795265022}) {
if (!ok(a)) return false;
}
}
return true;
}
#line 5 "/home/maspy/compro/library/nt/factor.hpp"
template <typename mint>
ll rho(ll n, ll c) {
assert(n > 1);
const mint cc(c);
auto f = [&](mint x) { return x * x + cc; };
mint x = 1, y = 2, z = 1, q = 1;
ll g = 1;
const ll m = 1LL << (__lg(n) / 5);
for (ll r = 1; g == 1; r <<= 1) {
x = y;
FOR(r) y = f(y);
for (ll k = 0; k < r && g == 1; k += m) {
z = y;
FOR(min(m, r - k)) y = f(y), q *= x - y;
g = gcd(q.val(), n);
}
}
if (g == n) do {
z = f(z);
g = gcd((x - z).val(), n);
} while (g == 1);
return g;
}
ll find_prime_factor(ll n) {
assert(n > 1);
if (primetest(n)) return n;
FOR(100) {
ll m = 0;
if (n < (1 << 30)) {
using mint = Mongomery_modint_32<20231025>;
mint::set_mod(n);
m = rho<mint>(n, RNG(0, n));
} else {
using mint = Mongomery_modint_64<20231025>;
mint::set_mod(n);
m = rho<mint>(n, RNG(0, n));
}
if (primetest(m)) return m;
n = m;
}
assert(0);
return -1;
}
// ソートしてくれる
vc<pair<ll, int>> factor(ll n) {
assert(n >= 1);
vc<pair<ll, int>> pf;
FOR(p, 2, 100) {
if (p * p > n) break;
if (n % p == 0) {
ll e = 0;
do { n /= p, e += 1; } while (n % p == 0);
pf.eb(p, e);
}
}
while (n > 1) {
ll p = find_prime_factor(n);
ll e = 0;
do { n /= p, e += 1; } while (n % p == 0);
pf.eb(p, e);
}
sort(all(pf));
return pf;
}
vc<pair<ll, int>> factor_by_lpf(ll n, vc<int>& lpf) {
vc<pair<ll, int>> res;
while (n > 1) {
int p = lpf[n];
int e = 0;
while (n % p == 0) {
n /= p;
++e;
}
res.eb(p, e);
}
return res;
}
#line 2 "/home/maspy/compro/library/mod/mod_pow.hpp"
#line 2 "/home/maspy/compro/library/mod/barrett.hpp"
// https://github.com/atcoder/ac-library/blob/master/atcoder/internal_math.hpp
struct Barrett {
u32 m;
u64 im;
explicit Barrett(u32 m = 1) : m(m), im(u64(-1) / m + 1) {}
u32 umod() const { return m; }
u32 modulo(u64 z) {
if (m == 1) return 0;
u64 x = (u64)(((unsigned __int128)(z)*im) >> 64);
u64 y = x * m;
return (z - y + (z < y ? m : 0));
}
u64 floor(u64 z) {
if (m == 1) return z;
u64 x = (u64)(((unsigned __int128)(z)*im) >> 64);
u64 y = x * m;
return (z < y ? x - 1 : x);
}
pair<u64, u32> divmod(u64 z) {
if (m == 1) return {z, 0};
u64 x = (u64)(((unsigned __int128)(z)*im) >> 64);
u64 y = x * m;
if (z < y) return {x - 1, z - y + m};
return {x, z - y};
}
u32 mul(u32 a, u32 b) { return modulo(u64(a) * b); }
};
struct Barrett_64 {
u128 mod, mh, ml;
explicit Barrett_64(u64 mod = 1) : mod(mod) {
u128 m = u128(-1) / mod;
if (m * mod + mod == u128(0)) ++m;
mh = m >> 64;
ml = m & u64(-1);
}
u64 umod() const { return mod; }
u64 modulo(u128 x) {
u128 z = (x & u64(-1)) * ml;
z = (x & u64(-1)) * mh + (x >> 64) * ml + (z >> 64);
z = (x >> 64) * mh + (z >> 64);
x -= z * mod;
return x < mod ? x : x - mod;
}
u64 mul(u64 a, u64 b) { return modulo(u128(a) * b); }
};
#line 5 "/home/maspy/compro/library/mod/mod_pow.hpp"
u32 mod_pow(int a, ll n, int mod) {
assert(n >= 0);
a = ((a %= mod) < 0 ? a + mod : a);
if ((mod & 1) && (mod < (1 << 30))) {
using mint = Mongomery_modint_32<202311021>;
mint::set_mod(mod);
return mint(a).pow(n).val();
}
Barrett bt(mod);
int r = 1;
while (n) {
if (n & 1) r = bt.mul(r, a);
a = bt.mul(a, a), n >>= 1;
}
return r;
}
u64 mod_pow_64(ll a, ll n, u64 mod) {
assert(n >= 0);
a = ((a %= mod) < 0 ? a + mod : a);
if ((mod & 1) && (mod < (u64(1) << 62))) {
using mint = Mongomery_modint_64<202311021>;
mint::set_mod(mod);
return mint(a).pow(n).val();
}
Barrett_64 bt(mod);
ll r = 1;
while (n) {
if (n & 1) r = bt.mul(r, a);
a = bt.mul(a, a), n >>= 1;
}
return r;
}
#line 6 "/home/maspy/compro/library/mod/primitive_root.hpp"
// int
int primitive_root(int p) {
auto pf = factor(p - 1);
auto is_ok = [&](int g) -> bool {
for (auto&& [q, e]: pf)
if (mod_pow(g, (p - 1) / q, p) == 1) return false;
return true;
};
while (1) {
int x = RNG(1, p);
if (is_ok(x)) return x;
}
return -1;
}
ll primitive_root_64(ll p) {
auto pf = factor(p - 1);
auto is_ok = [&](ll g) -> bool {
for (auto&& [q, e]: pf)
if (mod_pow_64(g, (p - 1) / q, p) == 1) return false;
return true;
};
while (1) {
ll x = RNG(1, p);
if (is_ok(x)) return x;
}
return -1;
}
#line 6 "/home/maspy/compro/library/mod/dynamic_modint.hpp"
template <int id>
struct Dynamic_Modint {
static constexpr bool is_modint = true;
using mint = Dynamic_Modint;
u32 val;
static Barrett bt;
static u32 umod() { return bt.umod(); }
static int get_mod() { return (int)(bt.umod()); }
static void set_mod(int m) {
assert(1 <= m);
bt = Barrett(m);
}
static Dynamic_Modint raw(u32 v) {
Dynamic_Modint x;
x.val = v;
return x;
}
Dynamic_Modint() : val(0) {}
Dynamic_Modint(u32 x) : val(bt.modulo(x)) {}
Dynamic_Modint(u64 x) : val(bt.modulo(x)) {}
Dynamic_Modint(int x) : val((x %= get_mod()) < 0 ? x + get_mod() : x) {}
Dynamic_Modint(ll x) : val((x %= get_mod()) < 0 ? x + get_mod() : x) {}
mint& operator+=(const mint& rhs) {
val = (val += rhs.val) < umod() ? val : val - umod();
return *this;
}
mint& operator-=(const mint& rhs) {
val = (val += umod() - rhs.val) < umod() ? val : val - umod();
return *this;
}
mint& operator*=(const mint& rhs) {
val = bt.mul(val, rhs.val);
return *this;
}
mint& operator/=(const mint& rhs) { return *this = *this * rhs.inverse(); }
mint operator-() const { return mint() - *this; }
mint pow(ll n) const {
assert(0 <= n);
mint x = *this, r = 1;
while (n) {
if (n & 1) r *= x;
x *= x, n >>= 1;
}
return r;
}
mint inverse() const {
int x = val, mod = get_mod();
int a = x, b = mod, u = 1, v = 0, t;
while (b > 0) {
t = a / b;
swap(a -= t * b, b), swap(u -= t * v, v);
}
if (u < 0) u += mod;
return u;
}
friend mint operator+(const mint& lhs, const mint& rhs) {
return mint(lhs) += rhs;
}
friend mint operator-(const mint& lhs, const mint& rhs) {
return mint(lhs) -= rhs;
}
friend mint operator*(const mint& lhs, const mint& rhs) {
return mint(lhs) *= rhs;
}
friend mint operator/(const mint& lhs, const mint& rhs) {
return mint(lhs) /= rhs;
}
friend bool operator==(const mint& lhs, const mint& rhs) {
return lhs.val == rhs.val;
}
friend bool operator!=(const mint& lhs, const mint& rhs) {
return lhs.val != rhs.val;
}
static pair<int, int>& get_ntt() {
static pair<int, int> p = {-1, -1};
return p;
}
static void set_ntt_info() {
int mod = get_mod();
int k = lowbit(mod - 1);
int r = primitive_root(mod);
r = mod_pow(r, (mod - 1) >> k, mod);
get_ntt() = {k, r};
}
static pair<int, int> ntt_info() { return get_ntt(); }
static bool can_ntt() { return ntt_info().fi != -1; }
};
#ifdef FASTIO
template <int id>
void rd(Dynamic_Modint<id>& x) {
fastio::rd(x.val);
x.val %= Dynamic_Modint<id>::umod();
}
template <int id>
void wt(Dynamic_Modint<id> x) {
fastio::wt(x.val);
}
#endif
using dmint = Dynamic_Modint<-1>;
template <int id>
Barrett Dynamic_Modint<id>::bt;
#line 1 "/home/maspy/compro/library/enumerate/floor_range.hpp"
// 商が q の区間 [l,r) を q について昇順
template <typename F>
void floor_range(u64 N, F f) {
assert(N <= (u64(1) << 50));
u64 sq = sqrtl(N);
u32 n = (sq * sq + sq <= N ? sq : sq - 1);
u64 prev = N + 1;
for (u32 q = 1; q <= n; ++q) {
u64 x = double(N) / (q + 1) + 1;
f(q, x, prev), prev = x;
}
for (u32 l = sq; l >= 1; --l) { f(u64(double(N) / l), l, l + 1); }
}
#line 2 "/home/maspy/compro/library/nt/primetable.hpp"
template <typename T = int>
vc<T> primetable(int LIM) {
++LIM;
const int S = 32768;
static int done = 2;
static vc<T> primes = {2}, sieve(S + 1);
if (done < LIM) {
done = LIM;
primes = {2}, sieve.assign(S + 1, 0);
const int R = LIM / 2;
primes.reserve(int(LIM / log(LIM) * 1.1));
vc<pair<int, int>> cp;
for (int i = 3; i <= S; i += 2) {
if (!sieve[i]) {
cp.eb(i, i * i / 2);
for (int j = i * i; j <= S; j += 2 * i) sieve[j] = 1;
}
}
for (int L = 1; L <= R; L += S) {
array<bool, S> block{};
for (auto& [p, idx]: cp)
for (int i = idx; i < S + L; idx = (i += p)) block[i - L] = 1;
FOR(i, min(S, R - L)) if (!block[i]) primes.eb((L + i) * 2 + 1);
}
}
int k = LB(primes, LIM + 1);
return {primes.begin(), primes.begin() + k};
}
#line 3 "/home/maspy/compro/library/nt/zeta.hpp"
template <typename T>
void divisor_zeta(vc<T>& A) {
assert(A[0] == 0);
int N = len(A) - 1;
auto P = primetable(N);
for (auto&& p: P) { FOR3(x, 1, N / p + 1) A[p * x] += A[x]; }
}
template <typename T>
void divisor_mobius(vc<T>& A) {
assert(A[0] == 0);
int N = len(A) - 1;
auto P = primetable(N);
for (auto&& p: P) { FOR3_R(x, 1, N / p + 1) A[p * x] -= A[x]; }
}
template <typename T>
void multiplier_zeta(vc<T>& A) {
assert(A[0] == 0);
int N = len(A) - 1;
auto P = primetable(N);
for (auto&& p: P) { FOR3_R(x, 1, N / p + 1) A[x] += A[p * x]; }
}
template <typename T>
void multiplier_mobius(vc<T>& A) {
assert(A[0] == 0);
int N = len(A) - 1;
auto P = primetable(N);
for (auto&& p: P) { FOR3(x, 1, N / p + 1) A[x] -= A[p * x]; }
}
#line 7 "main.cpp"
using mint = dmint;
// path a, cycle b
// cycle の中：L<R ならば全部到達可能
// L=R ならば、1/gcd
// S = 1 として数える、最後に N 倍
void solve() {
LL(N, P);
mint::set_mod(P);
mint ANS = 0;
// 長さ → 可能な LR
vc<mint> A(N);
FOR(d, 1, N) {
A[d] = N * (N + 1) / 2;
// ceil(d,R)=k+1 and floor(d,L)=k のとき引き算発生
// k を固定して
auto F = [&](int k, int l, int r) -> void {
// ceil(d,R)==k+1
ll mx = (k == 0 ? N : floor<ll>(d - 1, k));
// FOR(L, l, r) { A[d] -= (mx - L + 1); }
A[d] -= (mx + 1) * (r - l);
A[d] += ll(l + r - 1) * (r - l) / 2;
};
floor_range(d, F);
F(0, d + 1, N + 1);
}
A.resize(N);
A = cumsum<mint>(A, 0);
// サイクル長に対してかける係数
vc<mint> B(N + 1);
// FOR(b, 1, N + 1) { FOR(i, 1, N + 1) B[b] += mint(b / gcd(b, i)); }
vc<mint> F(N + 1);
FOR(i, 1, N + 1) { F[i] = inv<mint>(i); }
divisor_mobius<mint>(F);
FOR(d, 1, N + 1) {
FOR(i, 1, N / d + 1) {
ll b = d * i;
B[b] += F[d] * mint(N / d);
}
}
FOR(b, 1, N + 1) B[b] *= mint(b);
// S が path 上, (path,path)
{
vc<mint> X(N);
FOR(i, 1, N) {
X[i] = fact<mint>(N - 1) * fact_inv<mint>(N - 1 - i)
* mint(N).pow(N - 1 - i);
}
X = cumsum<mint>(X);
FOR(a, N) {
ANS += A[a] * (X[N] - X[a + 1]);
// FOR(b, 1, N - a) { ANS += A[a] * X[a + b]; }
}
}
// S が path 上, (path,cycle)
{
vc<mint> X(N);
FOR(i, 1, N) {
X[i] = fact<mint>(N - 1) * fact_inv<mint>(N - 1 - i)
* mint(N).pow(N - 1 - i);
}
X = cumsum<mint>(X);
FOR(b, 1, N + 1) {
mint v = mint(N * (N - 1) / 2) * b + B[b];
ANS += v * (X[N] - X[b]);
// FOR(a, N - b) { ANS += v * X[a + b]; }
}
}
// S が cycle 上
FOR(b, 1, N + 1) {
mint x = multinomial<mint>(N - 1, b - 1, N - b) * fact<mint>(b - 1);
x *= mint(N).pow(N - b);
// L<R なら全部行ける
ANS += mint(N * (N - 1) / 2) * x * (b - 1);
// L=R の場合
ANS += x * B[b];
ANS -= mint(N) * x;
}
// S
ANS *= mint(N);
print(ANS);
}
signed main() {
int T = 1;
// INT(T);
FOR(T) solve();
return 0;
}