to get the offset I do the log2 of the number of bytes per data. then the index is the log2 of the total number of raws. What about the number of bytes for a word selection. How do you get those bits required for a word addressing/selection.
This is he I have got How does direct mapped cache work?
Wikipedia does not really talk about word selection since it talks about the cache being devided into 3 colunms, whiles with word selection we have 4 colunms.
Let us assume I have a frame of 16 words and a word is 64 bits. The address length is 32 bits. How do I calculate the number of bits required to address a byte within a word for instance.