|
|
EDA365欢迎您登录!
您需要 登录 才可以下载或查看,没有帐号?注册
x
Since the earliest days of microprocessors, system designers have been plagued by a problem in which the
! a* ]$ s1 C+ M: [: [/ Gspeed of the CPU's operation exceeded the bandwidth of the memory subsystem to which it was connected.
: u$ ?/ l4 Q* _: a1 @/ I& mTo avoid wasting CPU cycles while waiting for the memory to fetch the requested data, the universally
n* {, _& G/ A6 j/ gadopted solution was to use an area of faster (and thus more expensive) memory to cache main memory data.3 @. O7 K0 Z7 D) m
This solution allowed the CPU to operate at its natural speed as long as the data it required was available in$ \+ {$ n A# J! h1 h5 s7 Z
the cache. |
|