RISC-V Processor With FPGA Hardware Acceleration

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current RISC-V computation devices face challenges in supporting user-defined instruction sets for high-speed special-purpose computations, requiring complex ASIC development and limiting flexibility and performance, especially in fields needing rapid modification or addition of functions.

Innovation Solution

A hardware high-speed computation device combining RISC-V processors with FPGAs, allowing user-defined instruction sets to be executed within a single chip, enabling flexible configuration and modification of user-defined instructions without changing the basic ISA, and allowing external IC control through an interlocking configuration with an FPGA unit.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Productivity

If a custom-made chip (ASIC) is used for special-purpose computation, then processing speed and performance are improved, but development time and cost increase significantly

Engineering Contradiction:
Improveprocessing speedVSAvoiddevelopment time
Core Design Contradiction:
ProductivityVSLoss of time

Solution Approach 1:

The patent applies dynamics by making the computation device reconfigurable through FPGA technology. The special-purpose computation logic can be dynamically changed by loading different binary files into the FPGA unit, allowing the same hardware to adapt to different computational tasks without physical remanufacturing. This resolves the contradiction by providing ASIC-like performance for specific tasks while maintaining the flexibility to change functions through software updates rather than hardware redesign.

Inventive Principle:
Principle #15Dynamics

2Adaptability or versatility

If a separate FPGA chip is interlocked with a general-purpose processor, then flexibility for modification is improved, but system complexity and interface complexity increase

Engineering Contradiction:
Improveflexibility for modificationVSAvoidsystem complexity
Core Design Contradiction:
Adaptability or versatilityVSDevice complexity

Solution Approach 1:

The patent merges the general-purpose processor and FPGA into a single integrated chip. The FPGA unit is directly coupled to the RISC-V processor core on the same substrate, eliminating the need for external bus interfaces and separate chip communication protocols. This integration reduces system complexity while maintaining the flexibility benefits of FPGA technology, as the special-purpose computation can be seamlessly coordinated with the processor through internal connections rather than external bus structures.

Inventive Principle:
Principle #5Merging (Combining)

3Productivity

If an extended ISA structure is designed and fixed in hardware, then high-performance computation is achieved, but adaptability for future changes is reduced

Engineering Contradiction:
Improvecomputation performanceVSAvoidadaptability for changes
Core Design Contradiction:
ProductivityVSAdaptability or versatility

Solution Approach 1:

The patent implements dynamics by making the instruction set extension configurable through software rather than fixed in hardware. The FPGA unit can be reconfigured by loading different binary files that define different extended instruction sets and corresponding computation logic. This allows the system to maintain high-performance computation for current tasks while adapting to future requirements by simply updating the FPGA configuration, eliminating the need for hardware redesign.

Inventive Principle:
Principle #15Dynamics

4Speed

If a high-speed serial interface protocol is used for communication, then data transmission speed is improved, but application complexity and control difficulty increase

Engineering Contradiction:
Improvedata transmission speedVSAvoidcontrol simplicity
Core Design Contradiction:
SpeedVSEase of operation

Solution Approach 1:

The patent eliminates the need for high-speed serial interface protocols by integrating the FPGA unit directly with the RISC-V processor on the same chip. Data transmission between the processor and FPGA occurs through internal chip connections rather than external bus structures, removing the complexity of protocol implementation while maintaining high-speed communication through direct internal pathways.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11409529B2RISC-V implemented processor with hardware acceleration supporting user defined instruction set and method thereof
Publication Date: 2022.08.09 ZARAMTECH
  • US11409529B2 patent drawing
  • US11409529B2 patent drawing
  • US11409529B2 patent drawing

AI summary

The present invention relates to a hardware high-speed computation combined RISC-V based computation device for supporting a user-defined instruction set and a method thereof which configures a hardware high-speed computation unit executing a user-defined function through a field programmable gate array (FPGA) in a single chip together with a RISC-V based computation device, executes general computation and user-defined computation in an instruction level, not a separate bus connection configuration, through a program using a RISC-V based instruction set including a user-defined instruction set, and provides flexibility capable of optionally changing the user-defined instruction set and a corresponding function and a method thereof.