Coherent attached processor proxy having hybrid directory
View Patent ↗A coherent attached processor proxy (CAPP) includes transport logic having a first interface configured to support communication with a system fabric of a primary coherent system and a second interface configured to support communication with an attached processor (AP) that is external to the primary coherent system and that includes a cache memory that holds copies of memory blocks belonging to a coherent address space of the primary coherent system. The CAPP further includes one or more master machines that initiate memory access requests on the system fabric of the primary coherent system on behalf of the AP, one or more snoop machines that service requests snooped on the system fabric, and a CAPP directory having a precise directory having a plurality of entries each associated with a smaller data granule and a coarse directory having a plurality of entries each associated with a larger data granule.
1. A method of data processing, comprising:
at a coherent attached processor proxy (CAPP) of a primary coherent system that serves as a participant in coherent communication within the primary coherent system on behalf of an attached processor (AP) external to the primary coherent system, maintaining, in a CAPP directory, entries corresponding to cachelines of interest to the AP that belong to a coherent address space of the primary coherent system, wherein the maintaining includes maintaining in a precise directory a plurality of entries each associated with a respective one of a plurality of smaller data granules and maintaining in a coarse directory a plurality of entries each associated with a respective one of a plurality of larger data granules;
the CAPP participating in coherent communication on a system fabric of the primary coherent system on behalf of the AP by reference to the coarse directory and the precise directory; and
the CAPP installing an entry in the coarse directory in a data-valid coherence state in advance of receipt of all of the larger data granule by the AP.
2. The method of claim 1 , wherein:
the smaller data granule comprises a single cacheline of data; and
the larger data granule comprises a plurality of contiguous cachelines.
3. The method of claim 1 , and further comprising:
in response to at least one AP command, the CAPP decomposes an entry in the coarse directory into a plurality of entries in the precise directory.
4. A method of data processing, comprising:
at a coherent attached processor proxy (CAPP) of a primary coherent system that serves as a participant in coherent communication within the primary coherent system on behalf of an attached processor (AP) external to the primary coherent system, maintaining, in a CAPP directory, entries corresponding to cachelines of interest to the AP that belong to a coherent address space of the primary coherent system, wherein the maintaining includes maintaining in a precise directory a plurality of entries each associated with a respective one of a plurality of smaller data granules and maintaining in a coarse directory a plurality of entries each associated with a respective one of a plurality of larger data granules;
the CAPP participating in coherent communication on a system fabric of the primary coherent system on behalf of the AP by reference to the coarse directory and the precise directory; and
the CAPP determining a composite coherence state for a target address by prioritizing a coherence state indicated by the precise directory over a coherence state indicated by the coarse directory.
5. A method of data processing, comprising:
at a coherent attached processor proxy (CAPP) of a primary coherent system that serves as a participant in coherent communication within the primary coherent system on behalf of an attached processor (AP) external to the primary coherent system, maintaining, in a CAPP directory, entries corresponding to cachelines of interest to the AP that belong to a coherent address space of the primary coherent system, wherein the maintaining includes maintaining in a precise directory a plurality of entries each associated with a respective one of a plurality of smaller data granules and maintaining in a coarse directory a plurality of entries each associated with a respective one of a plurality of larger data granules;
the CAPP participating in coherent communication on a system fabric of the primary coherent system on behalf of the AP by reference to the coarse directory and the precise directory; and
in response to a target address of a snooped memory access request missing in the precise directory and hitting in the coarse directory, the CAPP providing a Retry response and installing an entry corresponding to the target address in the precise directory.