Scalable coherence for multi-core processors

摘要

An apparatus of an aspect includes a plurality of cores. The plurality of cores are logically grouped into a plurality of clusters. A cluster sharing map-based coherence directory is coupled with the plurality of cores and is to track sharing of data among the plurality of cores. The cluster sharing map-based coherence directory includes a tag array to store corresponding pairs of addresses and cluster identifiers. Each of the addresses is to identify data. Each of the cluster identifiers is to identify one of the clusters. The cluster sharing map-based coherence directory also includes a cluster sharing map array to store cluster sharing maps. Each of the cluster sharing maps corresponds to one of the pairs of addresses and cluster identifiers. Each of the cluster sharing maps is to indicate intra-cluster sharing of data identified by the corresponding address within a cluster identified by the corresponding cluster identifier.

主权项

1. An apparatus comprising:
a plurality of cores, the plurality of cores logically grouped into a plurality of clusters; and a cluster sharing map-based coherence directory coupled with the plurality of cores configured to track sharing of data among the plurality of cores, the cluster sharing map-based coherence directory including: a tag array configured to store corresponding pairs of addresses and cluster identifiers in multiple sets and ways, each of the addresses configured to identify data, each of the cluster identifiers having a plurality of bits configured to identify only a single one of the clusters, wherein each address corresponds to only a single cluster identifier; and a cluster sharing map array configured to store cluster sharing maps, each of the cluster sharing maps corresponding to one of the pairs of addresses and cluster identifiers in a corresponding way of the tag array, each of the cluster sharing maps configured to indicate intra-cluster sharing of data identified by the corresponding address within a cluster identified by the plurality of bits of the corresponding cluster identifier.