Unified memory systems and methods
AssigneeNVIDIA Corporation
InventorsAmit Rao, Ashish Srivastava, Yogesh Kini
Life of this patent
Estimating the term…
Abstract
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.).
Claims20
A processor, comprising:
- one or more circuits to store one or more references to one or more graphics processing unit (GPU) physical addresses to be referenced by one or more central processing unit (CPU) virtual addresses.
A method, comprising:
- storing one or more graphics processing unit (GPU) physical addresses to be referenced by one or more central processing unit (CPU) virtual addresses.
A system, comprising:
- one or more circuits to store one or more graphics processing unit (GPU) physical addresses to be referenced by one or more central processing unit (CPU) virtual addresses.
Dependencies are inferred from the claim text.