#include <bits/stdc++.h>
#define x first
#define y second
using namespace std;
using ll=long long;
using pii=pair<int,int>;
using vi=vector<int>;
using vl=vector<ll>;
#define pb push_back
#define all(a) begin(a),end(a)

const int N=300010,MOD=1e9+7;
const char en='\n';
const ll LLINF=1ll<<60;

int t,n,k,h[N],ti[N],bio[N],ind[N],r[N],de[N];
pair<ll,int> an[N];
vi ch[N];
vi ord;

void dfs(int i,int p=-1)
{
	ind[i]=ord.size();
	ord.pb(i);
	for (auto x: ch[i]) if (x!=p)
	{
		de[x]=de[i]+1;
		dfs(x,i);
	}
	r[i]=ord.size();
}

int main()
{
	ios_base::sync_with_stdio(0);
	cin.tie(0);
	cin>>t;
	while (t--)
	{
		cin>>n>>k;
		for (int i=0;i<n;++i) cin>>ti[i],bio[i]=0,an[i]={LLINF,-1},ch[i].clear(),de[i]=0;
		for (int i=0;i<n;++i) cin>>h[i],--h[i],ch[h[i]].pb(i);
		for (int i=0;i<n;++i) if (!bio[i])
		{
			vi cu;
			int z=i;
			while (!bio[z])
			{
				bio[z]=1;
				cu.pb(z);
				z=h[z];
			}
			int cs=cu.size();
			if (bio[z]==2)
			{
				assert(0);
			}
			else
			{
				int po=-1;
				for (int j=0;j<cs;++j) if (cu[j]==z) po=j;
				vi cik;
				for (int j=po;j<cs;++j) cik.pb(cu[j]);
				int cis=cik.size();
				map<int,pair<vi,vi>> oc;
				for (int j=0;j<cis;++j) oc[ti[cik[j]]].x.pb(j);
				ord.clear();
				vi inds;
				for (auto x: cik)
				{
					ch[h[x]].erase(find(all(ch[h[x]]),x));
					inds.pb(ord.size());
					dfs(x);
					for (int j=inds.back()+1;j<(int)ord.size();++j) oc[ti[ord[j]]].y.pb(j);
				}
				
				for (auto x: oc)
				{
					int xyxs=x.y.x.size();
					if (xyxs)
					{
						ll ru=(k-1)/xyxs;
						int off=(k-1)%xyxs;
						for (int j=0;j<xyxs;++j)
						{
							int pc=x.y.x[j],pt=x.y.x[(j+off)%xyxs];
							an[cik[pc]]={ru*cis+(pt-pc+cis)%cis,x.x};
						}
					}
					int xyys=x.y.y.size();
					if (xyys)
					{
						vi sta;
						for (auto ii: x.y.y)
						{
							while (sta.size() && ii>=r[sta.back()]) sta.pop_back();
							sta.pb(ord[ii]);
							if ((int)sta.size()>=k)
							{
								an[ord[ii]]={de[ord[ii]]-de[sta[sta.size()-k]],x.x};
							}
							else if (xyxs)
							{
								int rk=k-sta.size();
								int po=upper_bound(all(inds),ii)-inds.begin()-1;
								auto cpoit=lower_bound(all(x.y.x),po);
								if (cpoit==x.y.x.end()) cpoit=x.y.x.begin();
								int cpo=cpoit-x.y.x.begin();
								ll ru=(rk-1)/xyxs,off=(rk-1)%xyxs;
								int pc=po,pt=x.y.x[(cpo+off)%xyxs];
								an[ord[ii]]={ru*cis+(pt-pc+cis)%cis,x.x};
							}
						}
					}
				}
				for (int j=cis-1;j>=0;--j) an[cik[j]]=min(an[cik[j]],pair<ll,int>(an[h[cik[j]]].x+1,an[h[cik[j]]].y));
				for (int j=cis-1;j>=0;--j) an[cik[j]]=min(an[cik[j]],pair<ll,int>(an[h[cik[j]]].x+1,an[h[cik[j]]].y));
				for (auto x: ord) an[x]=min(an[x],pair<ll,int>(an[h[x]].x+1,an[h[x]].y)),bio[x]=2;
			}
		}
		ll s=0;
		for (int i=0;i<n;++i)
		{
			//cout<<an[i].x<<' '<<an[i].y<<en;
			s+=(i+1)*1ll*an[i].y;
		}
		cout<<s<<en;
	}
}