import sys
from collections import defaultdict, Counter, deque
from itertools import permutations, combinations, product, combinations_with_replacement, groupby, accumulate
import operator
from math import sqrt, gcd, factorial
#from math import isqrt, prod,comb  # python3.8用(notpypy)
from bisect import bisect_left,bisect_right
#from functools import lru_cache,reduce
#from heapq import heappush,heappop,heapify,heappushpop,heapreplace
#import numpy as np
#import networkx as nx
#from networkx.utils import UnionFind
#from numba import njit, b1, i1, i4, i8, f8
#from scipy.sparse import csr_matrix
#from scipy.sparse.csgraph import shortest_path, floyd_warshall, dijkstra, bellman_ford, johnson, NegativeCycleError
#numba例 @njit(i1(i4[:], i8[:, :]),cache=True) 引数i4配列、i8 2次元配列,戻り値i1
def input(): return sys.stdin.readline().rstrip()
def divceil(n, k): return 1+(n-1)//k  # n/kの切り上げを返す
def yn(hantei, yes='Yes', no='No'): print(yes if hantei else no)

def rankdata(arr):#座標圧縮
    unique=list(set(arr))
    unique.sort()
    rank={e: i for i, e in enumerate(unique)}
    return [rank[i] for i in arr]

def main():
    mod = 10**9+7
    mod2 = 998244353
    n=int(input())
    A=list(map(int, input().split()))
    A=rankdata(A)
    dic=defaultdict(list)
    for i in range(n):
        dic[A[i]].append(i)
    ans=0
    count=1
    hantei=n
    suimen=0
    for i in range(max(A)+1)[::-1]:
        tmp=dic[i]
        bis=bisect_left(tmp, hantei) #n未満の個数を数える
        tmp=tmp[bis:]+tmp[:bis]
        #print(i,tmp)
        for tmps in tmp[::-1]:
            if n-count==tmps:
                count+=1
            ans+=1
            #print(tmps,hantei,count)
            if tmps==hantei-1 or (hantei==suimen and tmps==n-count+1):
                ans-=1
            if hantei==suimen:
                suimen+=1
            hantei=tmps
    print(ans)

if __name__ == '__main__':
    main()