论文部分内容阅读
前言门级逻辑模拟通常采用表格驱动法,这种算法需要多种静态和动态表格,信息量很大,如果不动用外存储器,很难对规模较大的数字逻辑系统进行模拟。但是,如果按普通的复盖方法,则由于表格驱动法的特点,模拟过程(其实逻辑描述语言的编译也一样)将是随机用盘的,即使使用虚拟存储,用盘次数也必然非常频繁。这不仅会使响应时间拖得很长,而且也必然增加了很多用盘的开销。为了解决大型逻辑电路的逻辑模拟问题,我们在表格驱动法的基础上提出了一种分块模拟算法,使得在每个模拟时刻里用盘次数尽可能地少。在我们已建立的系统中最多不超过2 N—2次。这里N是逻辑电路分块的总块数。由此看出只要有稳定可靠且速度较快的磁盘,
Introduction Gate-level logic simulation usually uses table-driven methods. This algorithm requires a variety of static and dynamic tables with a large amount of information. It is difficult to simulate a larger digital logic system without using external memory. However, if the normal overlay method is used, the simulation process (in fact, the same is also true of the logic description language) will be randomly used due to the characteristics of the table-driven method. Even if the virtual storage is used, the number of disks must be very frequent. Not only will this drag down the response time, but it will inevitably add a lot of disk overhead. In order to solve the problem of logic simulation of large logic circuits, we propose a block-based simulation algorithm based on table-driven method, which makes the number of disks as few as possible in each simulation time. No more than 2 N-2 times in the system we have established. Where N is the total number of blocks in the logic circuit. This shows that as long as there is a stable and reliable and faster disk,