摘要
提出一种新的适用于256比特以下的整数乘法的软件实现方式.用软件实现大整数乘法时,一般采用所谓"纸笔运算"的方式,这种方式要求在内存中开辟一个区域来存放运算的中间结果.新的实现方式调整了乘法运算的步骤,充分地利用了寄存器组,几乎不需要用内存来存放中间结果,有效地减少了对内存访问的次数,从而提高了速度.
The author proposes a new implementation of multiplication. This implementation is suitable for multiple two integers whose bit-length is less than 256 bits. Traditional way to implement multiple precession multiplication is called'pencil and paper', which need a large memory area to save the interim result. New implementation efficiently uses the registers and almost need no memory for interim result saving. The memory access times is greatly reduced, so the speed is increased.
出处
《广州大学学报（自然科学版）》
CAS
2004年第2期111-113,共3页
Journal of Guangzhou University:Natural Science Edition
关键词
多精度整数乘法
内存访问
循环
递归
multiple precision multiplication
memory access
cycle
recursion
