Give three reasons for having a split I- and D-cache at the L1 level.
What will be an ideal response?
The I-cache is for instructions and the D-cache is for data.
I-Cache is meant for instructions, therefore it most cases it doesn't not have to support writes.
The D-cache will only be used occasionally because not all instructions are to load/save.
If I/D-cache were not split, there would no NOPs in the pipeline, because everything must be paused for the cache to be accessed and return required values if there is a miss.
You might also like to view...
The disadvantage of _________ is that the amount of data that can be stored on the long outer tracks is only the same as what can be stored on the short inner tracks.
A. SSD B. CAV C. ROM D. CLV
On a traditional IBM mainframe, the __________ PSW is located in memory.
a. old b. current c. both A and B d. neither A nor B