Check patentability & draft patents in minutes with Patsnap Eureka AI!

Standard exceeding amount processor branch prediction method supporting embedded edge calculation

A technology of edge computing and branch prediction, applied in the computer field, can solve problems such as inability to predict correctly, target address error, predictor inability to predict correctly, etc.

Inactive Publication Date: 2019-08-09
核芯互联科技(青岛)有限公司
View PDF1 Cites 5 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

[0010] (3) The predicted jump is actually jumped, but the target address of the jump is wrong
[0017] When encountering the same jump instruction for the second time, in the scene where the jump actually occurs, the status of the BHT item is "01", and the predicted direction is still no jump, and it is still impossible to predict correctly
If the last occurrence does not jump, the predictor cannot predict correctly

Method used

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
View more

Image

Smart Image Click on the blue labels to locate them in the text.
Viewing Examples
Smart Image
  • Standard exceeding amount processor branch prediction method supporting embedded edge calculation
  • Standard exceeding amount processor branch prediction method supporting embedded edge calculation
  • Standard exceeding amount processor branch prediction method supporting embedded edge calculation

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0053] The present invention will be further explained below in conjunction with specific embodiments.

[0054] A superscalar processor branch prediction method supporting embedded edge computing, comprising the following steps:

[0055] S1: Quantitative analysis component of dynamic branch predictor: adopt computer architecture simulation software GEM5, and CoreMark, Dhrystone Benchmark, to analyze the branch direction and branch address of 2bC, GAg, PAg, GAp, PAp, G-share, TAGE and other branch predictors Perform performance simulation and statistical analysis on prediction accuracy to obtain a quantitative relationship between prediction accuracy and hardware implementation complexity;

[0056] S2: Processor architecture performance analysis component based on hardware counters: Processor performance analysis is divided into two stages: pre-silicon and post-silicon;

[0057] The common method of pre-silicon performance analysis is to create a performance model. The abstrac...

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

PUM

No PUM Login to View More

Abstract

The invention discloses a standard exceeding amount processor branch prediction method supporting embedded edge calculation, which is used for carrying out quantitative analysis through a dynamic branch prediction technology, and simulating and quantitatively analyzing the branch direction and branch address prediction accuracy of a branch predictor by adopting computer architecture simulation software GEM5. The invention also solves the problem of evaluation precision caused by high abstraction level of the processor performance model, and provides a mechanism for shortening the training timeof the branch predictor and a branch prediction method for simplifying process switching processing. Not only is the hit rate of the branch predictor improved, but also the processor power consumption is reduced.

Description

technical field [0001] The invention relates to the field of computer technology, in particular to a superscalar processor branch prediction method supporting embedded edge computing. Background technique [0002] Embedded edge computing AI chips are often SoC chips that perform deep learning calculations directly on terminal devices for specific AI applications, and have strong requirements for power consumption and energy efficiency ratio. In addition to the main control processor, video decoding and other modules, deep learning accelerated computing plays an increasingly important role in edge computing SoC chips. The software and hardware collaboration strategy of the main control processor and deep learning accelerator card is a key technology to improve the performance of edge computing SoC chips. [0003] Superscalar, by increasing instruction parallelism (ILP), is an important way to improve processor performance. Superscalar and very long instruction word (VLIW) t...

Claims

the structure of the environmentally friendly knitted fabric provided by the present invention; figure 2 Flow chart of the yarn wrapping machine for environmentally friendly knitted fabrics and storage devices; image 3 Is the parameter map of the yarn covering machine
Login to View More

Application Information

Patent Timeline
no application Login to View More
Patent Type & Authority Applications(China)
IPC IPC(8): G06F9/38
CPCG06F9/3846
Inventor 陈伟杰
Owner 核芯互联科技(青岛)有限公司
Features
  • R&D
  • Intellectual Property
  • Life Sciences
  • Materials
  • Tech Scout
Why Patsnap Eureka
  • Unparalleled Data Quality
  • Higher Quality Content
  • 60% Fewer Hallucinations
Social media
Patsnap Eureka Blog
Learn More

Browse by: Latest US Patents, China's latest patents, Technical Efficacy Thesaurus, Application Domain, Technology Topic, Popular Technical Reports.

© 2025 PatSnap. All rights reserved.Legal|Privacy policy|Modern Slavery Act Transparency Statement|Sitemap|About US| Contact US: help@patsnap.com