This algorithm has already been posted, but the following implementation is very compact and should optimize into branch-free code.
確定! 回上一頁