Home agent based cache transfer acceleration scheme

A method comprising:

• performing parallel lookups to a probe filter and an early probe cache responsive to receiving a memory request;
• retrieve an identifier of a first processing node from a first entry of the early probe cache responsive to a lookup of the early probe cache matching the first entry, wherein the first entry identifies the first processing node as an owner of a first region targeted by the memory request;
• and
• send an early probe to the first processing node responsive to determining a confidence indicator of the first entry is greater than a threshold, wherein the early probe is sent prior to a lookup to the probe filter completing.

An apparatus comprising:

• a probe filter configured to track cache lines that are cached by a plurality of processing nodes;
• and
• a coherent slave unit comprising an early probe cache configured to cache recent lookups to the probe filter, wherein the early probe cache stores entries on a region basis, wherein a region comprises a plurality of cache lines, and wherein the coherent slave unit is configured to:
• perform parallel lookups to the probe filter and the early probe cache responsive to receiving a memory request;
• retrieve an identifier of a first processing node from a first entry of the early probe cache responsive to a lookup of the early probe cache matching the first entry, wherein the first entry identifies the first processing node as an owner of a first region targeted by the memory request;
• and
• send an early probe to the first processing node responsive to determining a confidence indicator of the first entry is greater than a threshold, wherein the early probe is sent prior to a lookup to the probe filter completing.