Multithreading processor realizing functions of central processing unit and graphics processor and method

Abstract

The invention relates to a multithreading processor realizing the functions of a central processing unit and a graphics processor. The multithreading processor comprises a graphics fixed-function processing module, a multithreading parallel central processing module and a memory module, wherein the graphics fixed-function processing module is used for processing the fixed function of data in graphics processing; the multithreading parallel central processing module implements a central processing function and a programmable processing function in the graphics processing through unified thread scheduling and interacts graphics data after the programmable processing function and the graphics fixed-function processing module by the memory module; and the memory module supplies an unified storage space for the graphics fixed-function processing module and the multithreading parallel central processing module for realizing the storage, buffering or / and interaction of the data. The invention also relates to a data processing method. The multithreading processor and the method realizing the functions of the central processing unit and the graphics processor have the beneficial effect of realizing load balance among a plurality of threading processing engines.

Description

technical field [0001] The present invention relates to an integrated circuit, and more specifically relates to a multi-thread processor and a method for realizing the functions of a central processing unit and a graphic processing unit. Background technique [0002] CPU (central processing unit) and GPU (graphics processing unit) are the two most important integrated circuit chips in most computer systems, including PCs and portable devices. Traditionally, the CPU and GPU are two separate chips connected by a standard bus. More recently, the functions of these two chips have begun to be combined into a single package to reduce overhead. Specifically, by putting the CPU and GPU die in one package or by putting the full cores of the CPU and GPU in the same package. Regardless of which of the above methods is used, the structure and design of the CPU and GPU are still independent of each other. The resources needed to implement its functionality remain separate or separate....

AI Technical Summary

Problems solved by technology

This traditional design and its improved solution will bring a larger silicon chip usage area. At the same time, this solution will also make the load unbalanced between the CPU and GPU. One may be busy while the other is completely idle. state, resulting in a waste of resources

Images