UNIFIED MEMORY SYSTEMS AND METHODS

摘要

The present invention facilitates efficient and effective utilization of unified virtual addresses across multiple components. In one embodiment, the presented new approach or solution uses Operating System (OS) allocation on the central processing unit (CPU) combined with graphics processing unit (GPU) driver mappings to provide a unified virtual address (VA) across both GPU and CPU. The new approach helps ensure that a GPU VA pointer does not collide with a CPU pointer provided by OS CPU allocation (e.g., like one returned by “malloc” C runtime API, etc.).

主权项

1. An address allocation process comprising:
allocating physical memory pages to a second component; allocating a virtual address associated with a first component; and mapping the allocated physical pages to the virtual address allocated in association with the first component, wherein the mapping of the virtual address allocated in association with the first component is performed by an element associated with the second component.