#pragma GCC optimize("Ofast")
#pragma GCC optimize("unroll-loops")
#pragma GCC target("avx,avx2,fma")
#include <bits/stdc++.h>
using namespace std;
typedef long long ll;
typedef long double ld;
int N, K, a[100001];
ll pref[100001] = {0};
ll dp[2][100001] = {0};
int last_split[202][100001];
int cht[100001], fr = 0, bk = -1;
#define m(j) (pref[j])
#define c(j) (dp[0][j]-pref[j]*pref[N])
inline ld isect(const int &a, const int &b)
{
return (ld)(c(b) - c(a)) / (ld)(m(a) - m(b));
}
inline bool bad(const int &x, const int &y, const int &z)
{
return isect(z, y) <= isect(z, x);
}
inline ll eval(const int &i, const ll &x)
{
return m(i) * x + c(i);
}
int main()
{
cin.tie(0)->sync_with_stdio(0);
cin >> N >> K;
for (int i = 1; i <= N; i++)
{
cin >> a[i];
pref[i] = pref[i-1] + a[i];
}
for (int i = 0; i <= N; i++)
dp[0][i] = pref[i] * (pref[N] - pref[i]);
for (int k = 2; k <= K + 1; k++)
{
fr = 0, bk = -1;
for (int i = k; i <= N; i++)
{
while (fr < bk && bad(cht[bk-1], cht[bk], i-1))
bk--;
cht[++bk] = i-1;
while (fr < bk && eval(cht[fr], pref[i]) <= eval(cht[fr+1], pref[i]))
fr++;
dp[1][i] = eval(cht[fr], pref[i]) + pref[i] * (pref[N] - pref[i]);
last_split[k][i] = cht[fr];
}
swap(dp[0], dp[1]);
}
cout << dp[0][N] << '\n';
int i = N;
do
{
i = last_split[K+1][i];
cout << i << ' ';
}
while (--K);
cout << '\n';
return 0;
}
