CPU (central processing unit) and GPU (graphic processing unit) on-chip cache sharing method and device

Abstract

The invention discloses a CPU (central processing unit) and GPU (graphic processing unit) on-chip cache sharing method and device. The method includes the steps of caching access and storage requests from CPU / GPU in classification; requesting for arbitration for the cashed access and storage requests different in classification; executing the access and storage requests while subjecting read and write data of the access and storage requests to high-speed cache; enabling the read and write data read or written in an external storage through the access and storage requests to bypass the high-speed cache when the access and storage requests of a GPU are executed, operating the external storage directly, and notifying a CPU core to perform cancellation or update private data backup only when writing hit the target the high-speed cache. The device comprises a CPU request queue, a GPU request queue, an arbiter and a high-speed cache flow execution unit. The method and the device have the advantages that different access features of the CPU and the GPU can be taken into account at the same time, performance is high, hardware implementation is simple and cost is low.

Description

technical field [0001] The invention relates to the field of computer microprocessors, in particular to a method and device for sharing an on-chip cache memory between a CPU and a GPU. Background technique [0002] With the rapid development of VLSI and embedded technology, more and more transistor resources are available on a single chip, and system-on-chip (SoC) technology emerges as the times require. A SoC chip often integrates multiple IP cores with different functions and has relatively complete functions. SoC chips used in handheld terminal machines such as mobile phones and PDAs can integrate almost all functions of an embedded information processing system, and then realize information collection, input, storage, processor, output and other functions on a single chip. . Some current embedded systems (such as mobile phones, game consoles) put forward higher requirements on the performance of multimedia processors such as graphics, images, and videos, so the graphic...

AI Technical Summary

Problems solved by technology

However, the limited memory bandwidth on-chip is difficult to meet the high bandwidth requirements of CPU and GPU at the same time, which affects the performance of both CPU and GPU.

In addition, the memory access characteristics of CPU and GPU are also quite different, and different requirements are put forward for the characteristics of on-chip cache.

The CPU's memory access request is delay-sensitive, and it requires its memory access request to be served quickly; while the GPU's memory access request is bandwidth-sensitive, it requires high-bandwidth services, otherwise the GPU cannot process the image to be displayed in real time

To sum up, the shared use of on-chip cache affects the performance of CPU and GPU to a certain extent, making it impossible to meet both the low-latency requirements of the CPU and the high-bandwidth requirements of the GPU.

Images