System and Method for Page-Conscious GPU Instruction

摘要

Embodiments disclose a system and method for reducing virtual address translation latency in a wide execution engine that implements virtual memory. One example method describes a method comprising receiving a wavefront, classifying the wavefront into a subset based on classification criteria selected to reduce virtual address translation latency associated with a memory support structure, and scheduling the wavefront for processing based on the classifying.

主权项

1. A system comprising:
a wavefront classifier configured to classify a wavefront into a subset based on classification criteria selected to reduce virtual address translation latency associated with a memory support structure; and a wavefront scheduler configured to schedule the wavefront for processing by a plurality of processing lanes based on the classification.