For multi processor coherency how should we decide the local cache size or space of the individual masters and how the shared memory is related to the local caches?
For multi processor coherency how should we decide the local cache size or space of the individual masters and how the shared memory is related to the local caches?