N, M = map(int,input().split())
A = []
for i in range(N):
    A.append(sum(list(map(int, input().split()))))

dp = [[0]*N for _ in range(2)]
dp[1][0] = A[0]
for i in range(1,N):
    dp[0][i] = max(dp[0][i-1], dp[1][i-1] - A[i])
    dp[1][i] = max(dp[0][i-1] + A[i], dp[1][i-1])
print(max(dp[0][-1], dp[1][-1]))