#include<bits/stdc++.h>
#define N 2000005
#define M 10000006
#define il inline
#pragma GCC optimize(3,"Ofast","inline")
using namespace std;

#define getchar() (p1 == p2 && (p2 = (p1 = buf1) + fread(buf1, 1, 1 << 21, stdin), p1 == p2) ? EOF : *p1++)
char buf1[1 << 23], *p1 = buf1, *p2 = buf1, ubuf[1 << 23], *u = ubuf;
namespace IO
{
	template<typename T>
	void read(T &_x){_x=0;int _f=1;char ch=getchar();while(!isdigit(ch)) _f=(ch=='-'?-1:_f),ch=getchar();while(isdigit(ch)) _x=_x*10+(ch^48),ch=getchar();_x*=_f;}
	template<typename T,typename... Args>
	void read(T &_x,Args&...others){Read(_x);Read(others...);}
	const int BUF=20000000;char buf[BUF],to,stk[32];int plen;
	#define pc(x) buf[plen++]=x
	#define flush(); fwrite(buf,1,plen,stdout),plen=0;
	template<typename T>inline void print(T x){if(!x){pc(48);return;}if(x<0) x=-x,pc('-');for(;x;x/=10) stk[++to]=48+x%10;while(to) pc(stk[to--]);}
}
using namespace IO;

int n, m, B1, B2, root;
vector<int>p[N];
int dep[N], ans[N], dfn[N], k[N], len, tot, n1, n2;
struct RMQ_LCA { //O(n log n) - O(1) LCA
    int z[N], r[N], lg[N], tot;
    int f[N][22];
    void dfs(int x, int fa) {
        dep[x] = dep[fa] + 1;
        r[++tot] = x;
        dfn[x] = ++len;
        z[x] = tot;
        for(auto y : p[x]) {
            dfs(y, x);
            r[++tot] = x;
        }
    }
    int Min(int x, int y) {
        return dep[x] < dep[y] ? x : y;
    }
    il int LCA(int x, int y) {
        x = z[x], y = z[y];
        if(x > y) swap(x, y);
        int h = lg[y - x + 1];
        return Min(f[x][h], f[y - (1 << h) + 1][h]);
    }
    void init() {
        dfs(root, 0);
        for(int i = 2; i <= tot; i++) lg[i] = lg[i / 2] + 1;
        for(int i = 1; i <= tot; i++) f[i][0] = r[i];
        for(int j = 1; j <= 20; j++) 
        for(int i = 1; i + (1 << j) - 1 <= tot; i++) f[i][j] = Min(f[i][j - 1], f[i + (1 << (j - 1))][j - 1]);
    }
}ljm;

int tim1, tim2;
struct DS { //分块：O(1)修改, O(sqrt(n))查询
    int v1[N], v2[N];
    int X[M], Y[M];
    il void update(int x, int y) {
        X[++tim1] = x; Y[tim1] = y;
        v1[k[x]] += y;
        v2[x] += y;
    }
    il int query(int L, int R) {
    	if(L > R) return 0;
    	int res = 0;
    	if(k[L] == k[R]) {
    		for(int i = L; i <= R; i++) res += k[i];
    		return res;
		}
        int rr = B2 * k[L];
        for(int i = L; i <= rr; i++) res += v2[i];
        for(int i = k[L] + 1; i <= k[R]; i++) res += v1[i];
        return res;
    }
    il void Go(int x) { //撤销回到时间轴为x时
        while(tim1 > x) {
            v1[k[X[tim1]]] -= Y[tim1];
            v2[X[tim1]] -= Y[tim1];
            tim1--;
        }
    }
}ds;

struct List {
    int Pre, Nxt;
}t[N];

struct node {
    int l, r, x, id;
}z[N], h[N];
bool cmp1(node A, node B) {
    return A.l < B.l;
}
bool cmp2(node A, node B) {
    return A.r > B.r;
}

int v[N];
bool cmp(int x, int y) {
    return dfn[x] < dfn[y];
}
vector<int>r[N];
void Prepare() {
    for(int i = 1; i <= n; i++) v[i] = i;
    sort(v + 1, v + n + 1, cmp);
    for(int i = 1; i <= n; i++) r[dep[v[i]]].push_back(v[i]);
    for(int d = 1; d <= n; d++) {
        for(int i = 0; i < r[d].size(); i++) {
            if(i == 0) t[r[d][i]].Pre = -1;
            else t[r[d][i]].Pre = r[d][i - 1];
            if(i == r[d].size() - 1) t[r[d][i]].Nxt = -1;
            else {
                t[r[d][i]].Nxt = r[d][i + 1];
                ds.update(d - dep[ljm.LCA(r[d][i], r[d][i + 1])], 1);
            }
        }
        if(r[d].size()) ds.update(d, 1); //d深度只要有一个,贡献就要加1
    }
    n1 = tim1, n2 = tim2;
}


int X[M], Y[M], opt[M];
il void Go(int x) {
    while(tim2 > x) {
        if(opt[tim2] == 1) t[X[tim2]].Pre = Y[tim2];
        else t[X[tim2]].Nxt = Y[tim2];
        tim2--;
    }
}
il void Del(int x) {
    if(t[x].Pre != -1) {
        ds.update(dep[x] - dep[ljm.LCA(x, t[x].Pre)], -1);
        tim2++; X[tim2] = t[x].Pre, Y[tim2] = t[t[x].Pre].Nxt, opt[tim2] = 2;
        t[t[x].Pre].Nxt = t[x].Nxt; 
    }
    if(t[x].Nxt != -1) {
        ds.update(dep[x] - dep[ljm.LCA(x, t[x].Nxt)], -1);
        tim2++; X[tim2] = t[x].Nxt, Y[tim2] = t[t[x].Nxt].Pre, opt[tim2] = 1;
        t[t[x].Nxt].Pre = t[x].Pre; 
    }
    if(t[x].Pre != -1 && t[x].Nxt != -1) 
        ds.update(dep[x] - dep[ljm.LCA(t[x].Pre, t[x].Nxt)], 1);
    if(t[x].Pre == -1 && t[x].Nxt == -1)  //这个深度删掉没有点了
        ds.update(dep[x], -1);
}

il void Sol(int L, int R) { //处理左端点在[L,R]的询问
    sort(h + 1, h + tot + 1, cmp2);
    int Num = 0;
    for(int i = 1; i < L; i++) Del(i), Num++;
    int curL = L, curR = n;
    for(int i = 1; i <= tot; i++) {
        while(curR > h[i].r) Del(curR--), Num++;
        int lst1 = tim1, lst2 = tim2;
        while(curL < h[i].l) Del(curL++), Num++;
        ans[h[i].id] = ds.query(h[i].x + 1, n);
        ds.Go(lst1); Go(lst2);
        curL = L;
    }
    //cerr << Num << endl;
    ds.Go(n1); Go(n2);
}

signed main() {
	//freopen("x.in", "r", stdin);
	//freopen("x.out", "w", stdout);
    ios::sync_with_stdio(false);
    cin.tie(0); cout.tie(0);

    read(n),read(m);
    int vv = sqrt(m);
	B1 = max(1, n / vv); B2 = sqrt(n);
	for(int i = 1;i <= n;i++) k[i] = (i - 1) / B2 + 1;
    for(int i = 1, fa; i <= n; i++) {
        read(fa);
        if(fa) p[fa].push_back(i);
        else root = i;
    }
    ljm.init(); Prepare();
    for(int i = 1; i <= m; i++) read(z[i].l),read(z[i].r),read(z[i].x),z[i].id = i;
    sort(z + 1, z + m + 1, cmp1);
    int j = 0;
    for(int i = 1; i <= n; i += B1) {
        tot = 0;
        while(j + 1 <= m && z[j + 1].l <= i + B1 - 1) j++, h[++tot] = z[j];
        Sol(i, min(n, i + B1 - 1)); 
    }
    for(int i = 1; i <= m; i++) print(ans[i]),pc('\n');
    flush();
    return 0;
}

//start : 2025/1/24 19:03