#include <bits/stdc++.h>
using namespace std;

const int MAXN = 1e5;

typedef long long Long;

Long g[MAXN + 10], ff[MAXN + 10];

void init() {
    g[0] = g[1] = 0;
    for(int i = 2; i <= MAXN; i++) {
        g[i] = i-1 + g[(i+1)/2-1] + g[i-(i+1)/2];
    }
    ff[0] = ff[1] = 0;
    for(int i = 2; i <= MAXN; i++) {
        ff[i] = i * 1LL * (i - 1) / 2;
    }
}

void calc(int a, int b, int c) {
    if(a > b) return;
    if(a == b) {
        printf("%d ", a);
        return;
    }
    cerr << "calc(" << a << " " << b << " " << c << ")\n";
    int lo = a, hi = (a+b)/2;
    while(hi-lo > 1) {
        cerr << "[" << lo << " " << hi << "]\n";
        int m = (lo + hi) / 2;
        if(c <= b-a + ff[m-a] + ff[b-m]) lo = m;
        else hi = m;
    }
    if(c <= b-a + ff[hi-a]+ff[b-hi]) lo = hi;
    printf("%d ", lo);
    int x = lo - a, y = b - lo;
    if(c <= g[x]+ff[y]) {
        calc(a, lo-1, g[x]);
        calc(lo+1, b, c-g[x]);
    } else if(c <= ff[x]+g[y]) {
        calc(a, lo-1, c-g[y]);
        calc(lo+1, b, g[y]);
    } else if(ff[x]+g[y] <= c) {
        calc(a, lo-1, ff[x]);
        calc(lo+1, b, c - ff[x]);
    } else {
        calc(a, lo-1, c - ff[y]);
        calc(lo+1, b, ff[y]);
    }
}

void solve(int n, int c) {
    if(c < g[n] || c > ff[n]) {
        printf("-1\n");
    } else {
        calc(1,n, c);
        printf("\n");
    }
}

int main() {
    init();
    int Q; scanf("%d", &Q);
    for(int q = 0; q < Q; q++) {
        int n, c; scanf("%d%d", &n, &c);
        solve(n, c);
    }
}