Customizing function behavior based on cache and scheduling parameters of a memory argument

摘要

Disclosed are a method, a system and a computer program product of operating a data processing system that can include or be coupled to multiple processor cores. In one or more embodiments, each of multiple memory objects can be populated with work items and can be associated with attributes that can include information which can be used to describe data of each memory object and/or which can be used to process data of each memory object. The attributes can be used to indicate one or more of a cache policy, a cache size, and a cache line size, among others. In one or more embodiments, the attributes can be used as a history of how each memory object is used. The attributes can be used to indicate cache history statistics (e.g., a hit rate, a miss rate, etc.).

主权项

1. A method of operating a data processing system, the method comprising:
populating a first memory object in a global memory with a first plurality of work items; populating a second memory object in the global memory with a second plurality of work items, wherein the second memory object does not share a common address space with the first memory object; copying a first portion of the first memory object to a node memory; receiving a plurality of attributes associated with the first memory object and the second memory object; determining a first data size from at least a first attribute of the plurality of attributes that is associated with the first memory object; partitioning a memory store of a first processor core of a plurality of processor cores into a plurality of partitions based on the plurality of attributes, wherein each partition stores a plurality of cache lines; and copying, based on the first data size, a first work item from the first portion of the first memory object to a first partition of the plurality of partitions of the memory store of the first processor core.