US 12,657,024 B2 · Storage of transformed tensor in a cache

Storage of transformed tensor in a cache

AssigneeNVIDIA Corporation

InventorsHarold Carter Edwards, Stephen Anthony Bernard Jones, Alexander Lev Minkin, Olivier Giroux, Gokul Ramaswamy Hirisave Chandra Shekhara, Aditya Avinash Atluri, Apoorv Parle, Ronny Meir Krashinsky, Alan Kaatz, Andrew Robert Kerr, Jack H. Choquette

Life of this patent

Estimating the term…

Abstract

Apparatuses, systems, and techniques to perform a tensor prefetch instruction to cause one or more tensors to be transformed and stored into one or more caches. In at least one embodiment, one or more circuits of a GPU are to perform a tensor prefetch instruction to cause one or more tensors to be transformed and stored into one or more GPU caches.

Claims20

Dependent claims

A graphics processing unit (GPU), comprising:

one or more circuits to perform a tensor prefetch instruction to cause one or more tensors to be transformed based, at least in part, on a parameter indicating a portion of a tensor which one or more image-to-column transformations are to be applied and stored into one or more GPU caches.

Dependencies are inferred from the claim text.

Facts

Application

18/086,484Filed 21 Dec 2022

Priority

IN 20221106574216 Nov 2022

Published

US 2024/0169472 A123 May 2024

Granted

16 Jun 2026

Term ends

Not estimatedThe source records don’t support an estimate.

Classification23 codes

G06F 9/30047
To perform operations on memory: prefetch instructions; cache control instructions
G06F 3/0646
Making use of a particular technique: horizontal data movement in storage systems, i.e. moving data in between storage devices or systems
G06F 3/0659
Vertical data movement, i.e. input-output transfer; data movement between one or more hosts and one or more storage devices: command handling…
G06F 3/0673
In-line storage system: single storage device