Methods and apparatus for controlling a cache memory
AssigneeSony Computer Entertainment America LLC
InventorHidetaka Magoshi
Life of this patent
Estimating the term…
Abstract
Methods and apparatus enable: the partitioning of a main memory into a plurality of blocks, each block being adjacent to at least one of the other blocks, and each block including a plurality of data units containing one or more bits of data; the partitioning of each block of the main memory into a plurality of zones, each zone containing one or more of the data units; the association of at least some of the respective zones of each given block with respective others of the adjacent blocks to the given block; and the pre-fetching of a given one of the other blocks into a cache memory when any one of the data units within any of the associated zones of that block is addressed.
Claims16
An apparatus for controlling a memory, comprising:
- a main memory partitioned into a plurality of two-dimensional blocks, each block including a plurality of data units containing one or more bytes of data, and each given block being partitioned into a plurality of zones, each zone containing one or more of the data units and at least some of the zones being associated with a respective block adjacent to the given block;
- an address decoder operable to receive addresses to the bytes of data in the main memory, and to decode the addresses to determine in which zones the respective bytes of data are located;
- and
- a pre-fetch address generator operable to generate a request to pre-fetch the blocks that are associated with respective zones into a cache memory when the address decoder receives addresses to bytes of data located in the respective zones.
A method, comprising:
- partitioning a main memory into a plurality of two-dimensional blocks, and partitioning each block into a plurality of data units containing one or more bytes of data;
- partitioning each block of the main memory into a plurality of zones, each zone containing one or more of the data units;
- associating at least one of the zones of at least one of the blocks with another of the blocks;
- and
- pre-fetching at least a portion of the other block into a cache memory when one or more of the bytes of data of one of the data units within the at least one of the zones is addressed.
A method, comprising:
- partitioning a main memory into a plurality of two-dimensional blocks, and partitioning each block into a plurality of data units containing one or more bytes of data;
- partitioning each block of the main memory into a plurality of zones, each zone containing one or more of the data units;
- associating at least some of the respective zones of each block with others of the blocks;
- and
- pre-fetching at least a portion of one of the other blocks into a cache memory when any one of the data units within any of the associated zones of that block is addressed.
A method, comprising:
- partitioning a main memory into a plurality of two-dimensional rectangular blocks, each block being adjacent, to at least one of the other blocks along a common boundary, and each block including a plurality of data units containing one or more bytes of data;
- partitioning each block of the main memory into a plurality of zones, each zone containing one or more of the data units;
- associating at least some of the respective zones of each given block with respective others of the adjacent blocks to the given block;
- and
- pre-fetching a given one of the other blocks into a cache memory when any one of the data units within any of the associated zones of that block is addressed.
Dependencies are inferred from the claim text.