Pipelined non-blocking two level cache system with inherent transaction collision-avoidance

摘要

A cache subsystem in a data processing system is structured to place the L1 cache RAMs after the L2 cache RAMs in the pipeline for processing both CPU write transactions and L1 line-fill transactions. In this manner the lines loaded into the L1 cache are updated by all CPU write transactions without having to perform any explicit checks and thus read-miss/write-miss conflicts are avoided. The present invention also places the L1 tag RAM before the L1 data RAM for both CPU write transactions and L1 line-fill transactions, such that CPU write transactions may check that a line is in the L1 cache before updating it. L1 line-fill transactions can then check that the line to be transferred from the L2 cache to the L1 cache is not already in the L1 cache. If the line is already present the cache line-fill transaction is cancelled, thus avoiding multiple allocations in the L1 cache. <IMAGE>