Method for realizing virtual execution environment of central processing unit (CPU)/graphics processing unit (GPU) heterogeneous platform

Abstract

The invention relates to a method for realizing a virtual execution environment of a central processing unit (CPU)/graphics processing unit (GPU) heterogeneous platform, which belongs to the technical field of telecommunication. In the method, programs can be run on an X86CPU and NVIDIA GPU hybrid architecture by a dynamic binary translation technique; static information and dynamic information of the programs are acquired by the dynamic binary translation technique; program execution nested loops, a dependency relationship among the loops and data streams of inlet and outlet hot blocks are searched through the information; and the execution is implemented by two steps of: 1, acquiring information, optimizing the hot blocks and storing the hot blocks into files; 2, and generating a hybridprogram comprising a serial instruction and a parallel instruction and executing the program comprising the hybrid instructions. The method has the advantages that: the traditional serial programs can be run a CPU/GPU heterogeneous multi-core architecture by the dynamic binary translation technique, without modifying program source codes; and the execution of the programs can be accelerated by the GPU.

Description

technical field [0001] The invention relates to a method in the technical field of telecommunications, in particular to a method for realizing a virtual execution environment of a CPU / GPU heterogeneous platform. Background technique [0002] Multi-core technology is the current trend of processor development, but there is a certain limit to improve the parallel processing performance of the processor by simply increasing the number of CPU cores. Therefore, in order to meet the growing computing needs, processing cores for certain types of computing are introduced to form a heterogeneous multi-core architecture. For specific fields, heterogeneous cores can make up for the shortcomings of traditional CPU computing, while CPU / GPU is a heterogeneous multi-core architecture that has developed rapidly in recent years. GPU (Graphics Acceleration Unit) is suitable for processing a large number of computationally intensive tasks, such as image and multimedia processing. Due to its...

AI Technical Summary

Problems solved by technology

[0003] 1. The problem of code rewriting cost: Since the program on the CPU cannot directly run on the GPU, the source code needs to be modified or rewritten according to the parallel model. At present, a large number of computationally intensive programs are developed with general-purpose C / C++. The program on the CPU is a time-consuming and heavy workload task

And the source code of many programs is difficult to obtain

[0004] 2. Binary compatibility issues: Specific programming environments will bring about compatibility issues. Due to the differences in the storage systems of different hardware and the number of processor cores, the compiled programs run on different GPU products, and some even run on the same GPU. It is also not compatible on different generations of the

In addition, it does not solve the compatibility problem of binary programs