Speculative thread particle with restart optimization and restart optimization method thereof

Abstract

The invention discloses a speculative thread particle with restart optimization and a restart optimization method thereof. The speculative thread particle consists of an input parameter identification part, an execution part, and a result storage part, wherein the input parameter identification part consists of N+1 binary digits which are orderly arranged, and is used for identifying whether the speculative thread particle uses an input parameter corresponding to an input parameter identification byte; the execution part consists of a plurality of program statements in speculative thread, and is used for storing executive statements of the speculative thread particle; and the result storage part is used for storing an execution result of the speculative thread particle. The restart optimization method has the following operation steps of: A, detecting disabled speculative thread input parameters; B, searching speculative thread particles which depend on the failed speculative thread input parameters in step A; and C, restarting the speculative thread particles found in step B. The restart optimization method has the advantages that the restart optimization method can quickly judge the speculative thread particles which must be restarted when the speculative fails, and reduce the cost for thread restart.

Description

technical field [0001] The invention relates to the computer field, and proposes a restart-optimized forward-looking thread particle and a restart optimization method thereof. Background technique [0002] The continuous development of computer technology and the extensive and in-depth application fields have higher requirements for processor performance. However, traditional methods for improving processor performance, such as increasing the integration of processor transistors and increasing processor clock frequency, are greatly limited. These limitations include that the processor technology and clock frequency are close to the limit, and its improvement becomes more and more difficult; with the increase of the clock frequency, the power consumption of the processor is increasing, and the heat dissipation problem has also become a problem that affects the stable performance of the processor. The key issue. [0003] At present, the increasing importance of line delay, p...

AI Technical Summary

Problems solved by technology

Parallel technology based on Thread-Level Speculative (TLS) is one of the main ways to solve these problems. The results inferred by the forward-looking technology are not always correct. When the inference of the forward-looking technology fails, the failure must be restarted. thread, but the time and resource overhead of restarting the thread is huge, and even offset the benefits brought by the parallelization of serial programs. Therefore, optimizing thread restart is called one of the key technologies of serial program parallelization, and it is necessary to further refine Thread, reduce the thread granularity, when the thread has a look-ahead failure, you only need to restart some necessary parts of the thread, without restarting the entire thread

Images