INF = 10**10
N, D = map(int, input().split())
dp = [[-INF] * 2 for i in range(N+1)]
dp[0][0] = 0
for i in range(N):
    T, K = map(int, input().split())
    dp[i+1][0] = max(dp[i][0] + T, dp[i][1] + T - D)
    dp[i+1][1] = max(dp[i][0] + K - D, dp[i][1] + K) 
    # print(i+1,dp[i+1])
print(max(dp[N]))