https://practice.geeksforgeeks.org/problems/optimal-strategy-for-a-game-1587115620/1/#18

Hi friends, I was solving this question on gfg. But on 100th test case my solution is showing segmentation error.
Even after many hours I am not able to find out what is the problem with my code.
Could anyone please help me here.

Code:

class Solution{
    public:
    #define ll long long int
    #define pi pair<ll,ll>
    long long maximumAmount(int arr[], int n){
        // Your code here
        pi dp[n][n];
        for(ll i=0;i<n;i++){
            dp[i][i]={arr[i],0};
        }
        ll j1=1;
        while(j1<n){
            ll i=0,j=j1;
            while(j<n){
                ll a=arr[i]+dp[i+1][j].second;
                ll b=arr[j]+dp[i][j-1].second;
                if(a>b){
                    dp[i][j]={a,dp[i+1][j].first};
                }
                else{
                    dp[i][j]={b,dp[i][j-1].first};
                }
                
                i++,j++;
            }
            j1++;
        }
        return dp[0][n-1].first;

}
};

Approach used:
I have made 2D dp array and I am storing values in pairs in that array.
The first value in the pair represents the sum if the first turn is made. The second value in a pair represents the sum if the second turn is taken.
By keeping the values of smaller subarrays the final value would be dp[0][n-1].