Skip to content

A superscalar and out-of-order instruction pipeline loosely based on the RISC-V ISA implemented in SystemVerilog.

Notifications You must be signed in to change notification settings

amanda-matthes/simple_superscalar_out-of-order_pipeline

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

16 Commits
 
 
 
 
 
 
 
 
 
 

Repository files navigation

A SystemVerilog implementation of a superscalar and out-of-order instruction pipeline

A superscalar and out-of-order instruction pipeline loosely based on the RISC-V ISA implemented in SystemVerilog. This project was part of my BSc thesis.

Abstract

Superscalar out-of-order execution has become the norm in modern processors. Yet there is little information in the literature about its implementation details. This thesis explores what new hardware structures superscalar out-of-order execution requires. It presents a design for a simple processor, implemented in SystemVerilog, that uses register renaming, reservation stations and a reorder buffer to dynamically schedule instructions.
The thesis also contains an essay that explains why power limitations have caused processor clock rates to stagnate in the noughties. This resulted in heightened pressure on computer architects to increase instruction throughput per cycle.

Pipeline structure

See my thesis text for details.

About

A superscalar and out-of-order instruction pipeline loosely based on the RISC-V ISA implemented in SystemVerilog.

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published