Unlock instant, AI-driven research and patent intelligence for your innovation.

Shared front-end assembly line structure based on monolithic multiprocessor system

A multi-processor system and pipeline technology, applied in machine execution devices, concurrent instruction execution, etc., can solve problems such as less flexibility in resource utilization than SMT, poor performance, and inability to run efficiently, so as to improve processing performance and reduce the number of hardware. , the effect of alleviating the bottleneck of the storage wall

Inactive Publication Date: 2011-04-20
NAT UNIV OF DEFENSE TECH
View PDF2 Cites 6 Cited by
  • Summary
  • Abstract
  • Description
  • Claims
  • Application Information

AI Technical Summary

Problems solved by technology

But the existing CMP still has many deficiencies, which are mainly reflected in: the current general-purpose programs oriented to microprocessors based on sequential structures can run efficiently, but oriented to scientific computing, graphics and image processing, etc. The main program cannot run efficiently on CMP; in addition, some technologies of CMP are not yet mature: the flexibility of resource utilization is not as good as SMT, and the performance may be poor for some applications

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
  • Shared front-end assembly line structure based on monolithic multiprocessor system
  • Shared front-end assembly line structure based on monolithic multiprocessor system
  • Shared front-end assembly line structure based on monolithic multiprocessor system

Examples

Experimental program
Comparison scheme
Effect test

Embodiment Construction

[0015] The present invention will be further described in detail below in conjunction with the accompanying drawings and specific embodiments.

[0016] In the classic 5-segment pipeline, it contains the following 5 segments: fetching segment, decoding segment, execution segment, memory access segment, and write-back segment. The same is true in the device, in the present invention, aiming at the program that CMP can't efficiently run parallel operations such as scientific computing, graphics and image processing, etc. and CMP is not as flexible as SMT in resource utilization, it may be possible for some applications Insufficiently poor, a shared front-end pipeline based on a single-chip multiprocessor system is designed.

[0017] like figure 1 and figure 2 As shown, the present invention is based on a shared front-end pipeline structure of a single-chip multi-processor system, including an instruction fetch unit, a decoding unit and a prefetch unit arranged before the execu...

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 shared front-end assembly line structure based on a monolithic multiprocessor system, comprising an instruction fetch unit, a decoding unit and a prefetching unit, wherein the instruction fetch unit is arranged in front of the execution unit of more than two flow cores and is shared by all the flow cores; after the instruction fetch unit finishes instruction fetch operation, the decoding unit sends an instruction to the execution unit of the flow cores; different flow cores are used for processing different threads of the same instruction; the prefetching unit is used for carrying out intermittent instruction prefetching and forming an instruction queue when the instruction is blocked in the execution unit of the flow cores, and the instruction queue in the prefetching unit is successively sent to the execution unit after the blocking state of the execution unit ends. The shared front-end assembly line structure has the advantages of simple and compact structure, low cost, deep parallelism and the like, and can effectively alleviate the bottleneck of memory wall, and greatly improve the processing performances of applications, such as media processing, data intensive scientific calculation and the like.

Description

technical field [0001] The invention mainly relates to the field of single-chip multi-processors, in particular to a shared front-end pipeline structure based on single-chip multi-processor systems. Background technique [0002] A multi-core processor is also called a chip multi-processor (CMP), or a single-chip multi-processor. Since Stanford University in the United States first proposed the idea of ​​on-chip multiprocessor (CMP) and the first multi-core structure prototype in 1996, IBM launched the first commercial multi-core processor POWER4 in 2001, and then Intel and AMD multi-core processors in 2005. applications, and finally multi-core has become the mainstream of the market, and multi-core processors have experienced more than ten years of development. In this process, the application range of multi-core processors has covered many fields such as multimedia computing, embedded devices, personal computers, commercial servers, and high-performance computers. Multi-co...

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
IPC IPC(8): G06F9/38
Inventor 邢座程陈小保付桂涛马安国黄平汤先拓何锐王庆林晏小波李方圆邱建雄蔡放闵银皮梅家祥孟晓冬赵齐王宏燕
Owner NAT UNIV OF DEFENSE TECH