#include <bits/stdc++.h>
using namespace std;
using i64 = long long;
using pii = pair<int, int>;
constexpr int maxn = 200000 + 10;
int t[maxn], a[maxn], d[maxn];
vector<int> e[maxn];
int f[maxn];
i64 g[maxn];
vector<int> p[maxn];
int id[maxn], pd[maxn];
int dep[maxn], pre[maxn], len[maxn];
int lst[maxn];
int du[maxn];
int n, k;
int now, cyc;
void dfs(int u)
{
    du[dep[u]] = u;
    p[t[u]].emplace_back(dep[u]);
    if (!d[u])
    {
        if (p[t[u]].size() >= k) f[u] = du[p[t[u]][p[t[u]].size() - k]], g[u] = dep[u] - p[t[u]][p[t[u]].size() - k];
        else if (lst[t[u]])
        {
            f[u] = pre[f[lst[t[u]]]];
            g[u] = g[lst[t[u]]] - len[f[lst[t[u]]]];
            if (d[lst[t[u]]]) g[u] += (id[lst[t[u]]] - now + cyc) % cyc + dep[u];
            else g[u] += dep[u] - dep[lst[t[u]]];
        }
        else f[u] = 0, g[u] = 0x3f3f3f3f3f3f3f3f;
    }
    int x = lst[t[u]];
    lst[t[u]] = u;
    for (int v : e[u])
    {
        dep[v] = dep[u] + 1;
        dfs(v);
    }
    lst[t[u]] = x;
    p[t[u]].pop_back();
}
void dwn(int u)
{
    for (int v : e[u])
    {
        if (g[u] + 1 < g[v]) g[v] = g[u] + 1, f[v] = f[u];
        dwn(v);
    }
}
void solve()
{
    cin >> n >> k;
    for (int i=1;i<=n;++i) e[i].clear();
    for (int i=1;i<=n;++i) cin >> t[i]; 
    for (int i=1;i<=n;++i) cin >> a[i];
    memset(d, 0, sizeof(d[0]) * (n + 5));
    for (int i=1;i<=n;++i) ++d[a[i]];
    queue<int> q;
    for (int i=1;i<=n;++i) if (!d[i]) q.push(i);
    while (q.size())
    {
        int u = q.front(); q.pop();
        if (!--d[a[u]]) q.push(a[u]);
    }
    memset(f, 0, sizeof(f[0]) * (n + 5));
    for (int i=1;i<=n;++i) if (!d[i]) e[a[i]].emplace_back(i);
    for (int u=1;u<=n;++u)
        if (!f[u] && d[u])
        {
            vector<int> c;
            int v = u;
            do
            {
                c.emplace_back(v);
                v = a[v];
            } while (v != u);
            int x = 0;
            for (int v : c)
            {
                pd[v] = p[t[v]].size();
                p[t[v]].emplace_back(x);
                id[v] = x++;
            }
            for (int v : c)
            {
                int u = p[t[v]][(pd[v] + k - 1) % p[t[v]].size()];
                f[v] = c[u];
                g[v] = (i64) (k - 1) / p[t[v]].size() * x;
                g[v] += (u - id[v] + x) % x;
                u = p[t[v]][(pd[v] + p[t[v]].size() - 1) % p[t[v]].size()];
                pre[v] = c[u];
                len[v] = (id[v] - u + x) % x;
                if (!len[v]) len[v] = x;
            }
            vector<int> d = c;
            reverse(d.begin(), d.end());
            for (int v : d) lst[t[v]] = v;
            for (int v : d) p[t[v]].clear();
            cyc = x;
            for (int v : d) now = id[v], dep[v] = 0, dfs(v), lst[t[v]] = v;
            int lst = d.back();
            for (int t=0;t<2;++t) for (int v : d)
            {
                if (g[lst] + 1 < g[v]) g[v] = g[lst] + 1, f[v] = f[lst];
                lst = v;
            }
            for (int v : c) dwn(v);
            for (int v : c) ::lst[t[v]] = 0;
        }
    i64 ans = 0;
    // for (int i=1;i<=n;++i)
    // {
    //     map<int, int> mp;
    //     int u = i;
    //     for (;;)
    //     {
    //         ++mp[t[u]];
    //         if (mp[t[u]] == k) break;
    //         u = a[u];
    //     }
    //     assert(u == f[i]);
    // }
    for (int i=1;i<=n;++i) ans += (i64) i * t[f[i]];
    cout << ans << '\n';
    // for (int i=1;i<=n;++i) cout << t[f[i]] << ' ';
    // cout << '\n';
}
int main()
{
    ios::sync_with_stdio(0);
    cin.tie(0);
    int t;
    cin >> t;
    while (t--) solve();
    return 0;
}