Mirror Register Buffer Bypasses Internal Bus for Data Transfer

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Current semiconductor devices face inefficiencies in data transfer and data organization due to the need for CPU involvement in specifying transfer addresses and the subsequent longer processing times, especially when using DMA/DTC controllers require CPU intervention for data reduction.

Innovation Solution

A semiconductor device configuration that includes a processor, memory, external interface, register, mirror register buffer, and internal buses, where the mirror register buffer assigns consecutive addresses to data from the register, allowing direct transfer to memory without using the internal bus, thereby reducing data transfer and organization time.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Adaptability or versatility

If CPU is used for data transfer, then data transfer can be performed with flexible address specification, but data transfer time becomes longer and CPU cannot perform other processing

Engineering Contradiction:
Improveaddress specification flexibilityVSAvoiddata transfer time
Core Design Contradiction:
Adaptability or versatilityVSLoss of time

Solution Approach 1:

The system segments the data transfer function by separating address specification (performed by CPU) from actual data transfer (performed by dedicated transfer circuit). This allows the CPU to specify addresses flexibly while the transfer circuit handles the actual data movement independently, resolving the contradiction between flexible address specification and fast data transfer.

Inventive Principle:
Principle #1Segmentation

2Speed

If DMA/DTC controller is used for data transfer, then data transfer speed is improved, but CPU involvement is still required for data reduction increasing total processing time

Engineering Contradiction:
Improvedata transfer speedVSAvoidtotal processing time
Core Design Contradiction:
SpeedVSLoss of time

Solution Approach 1:

The transfer circuit performs data transfer in advance and independently of CPU intervention. By pre-configuring transfer parameters and executing transfers autonomously, the system eliminates the need for CPU involvement in data reduction operations, thereby reducing total processing time while maintaining high transfer speeds.

Inventive Principle:
Principle #10Preliminary action

3Device complexity

If data transfer goes through internal bus, then system architecture is simple, but data transfer time increases

Engineering Contradiction:
Improvesystem architecture complexityVSAvoiddata transfer time
Core Design Contradiction:
Device complexityVSLoss of time

Solution Approach 1:

A dedicated transfer circuit is introduced as an intermediary component between the data source and destination. This specialized mediator handles data transfer operations independently of the internal bus, providing a faster transfer path while maintaining overall system architectural simplicity through modular design.

Inventive Principle:
Principle #24Intermediary (Mediator)

Data Source

PatentEP4099176B1Semiconductor device
Publication Date: 2024.10.02 RENESAS ELECTRONICS CORP
  • EP4099176B1 patent drawingFigure 1
  • EP4099176B1 patent drawingFigure 2
  • EP4099176B1 patent drawingFigure 3A

AI summary

A semiconductor device capable of shortening a time required for data transfer and data organizing is provided. The solid state device includes a processor (10), a memory (30), an external interface (40), registers (51, 52, 53) for storing data received by the external interface (40), a mirror register buffer (60), a processor (10), a memory (30), an external interface (40), registers (51, 52, 53), and an internal bus (80) connected to the mirror register buffer (60). Registers (51, 52, 53) output data to the mirror register buffer (60) without going through the internal bus (80). Mirror register buffer (60) gives the data input from the registers (51, 52, 53) an address in a mirror register buffer (60) different from the address allocated to the register (51, 52, 53), and transfers the data to the memory (30) without passing through the internal bus (80).