I was taught that when a reference is mapped to a cache block, X, for the first time, the word is stored in the cache block, bearing a tag and index that helps identify it for future reads. Then, when a later reference is deemed to have the same tag and index for cache block X, the value inside the cache block is returned as the read value. Is it the same process for multi-word caches?
I was given a direct-mapped cache: Four 4-word blocks, 32 bits per word, and these word address references...
Ref Tag Index Hit/Miss 134 8 1 Miss 212 13 1 Miss 135 8 1 Hit 213 13 1 Hit
...were what I determine to be the hit/miss for the references because the tags and the index match, even if the value is wrong. But the answer key tells me the hit/miss for the references are:
Ref Tag Index Hit/Miss 134 8 1 Miss 212 13 1 Miss 135 8 1 Miss 213 13 1 Miss
...so there's more involved when it comes to multi-word caches? ...or the answer key is wrong. Can anyone clarify?