Semiconductor Sub-Block Latency Reduction via Flip-Flop Elimination

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

In semiconductor devices, the use of multiple flip-flops in synchronous data communication paths leads to increased latency due to the need for data to be held and latched at each clock cycle, resulting in longer data transfer times and clock latency.

Innovation Solution

Eliminating the flip-flop on the input of a second functional block when the first block already has a flip-flop on its output, allowing direct data transfer without the additional latency of holding and latching data at each clock cycle.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If multiple flip-flops are used in synchronous data communication paths, then data can be properly synchronized and held at each clock cycle, but latency increases due to the need to latch data at each clock cycle

Engineering Contradiction:
Improvedata synchronizationVSAvoidlatency
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent removes redundant flip-flops from the data communication path between functional blocks. Specifically, when a first functional block has an output flip-flop and a second functional block has an input flip-flop, the input flip-flop of the second block is eliminated since the output flip-flop of the first block already provides the necessary synchronization and data holding function.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The patent merges the synchronization function into a single flip-flop (the output flip-flop of the first functional block) that serves both blocks. This consolidation eliminates the need for separate input flip-flops in receiving blocks, reducing total flip-flop count while maintaining proper data synchronization across clock cycles.

Inventive Principle:
Principle #5Merging (Combining)

2Reliability

If flip-flops are used at both input and output of functional blocks, then data synchronization is ensured, but the number of flip-flops increases leading to higher device complexity

Engineering Contradiction:
Improvedata synchronizationVSAvoidnumber of flip-flops
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent extracts and removes redundant input flip-flops from second functional blocks that receive data from first functional blocks with output flip-flops. This elimination reduces the total number of flip-flops in the device while maintaining adequate synchronization through the remaining output flip-flops of the first blocks.

Inventive Principle:
Principle #2Taking out (Extraction)

Solution Approach 2:

The synchronization function is merged into the output flip-flops of transmitting functional blocks, which serve dual purposes: data buffering for the transmitting block and synchronized data source for the receiving block. This merging reduces overall device complexity by eliminating duplicate synchronization elements.

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS10310743B2Latency improvements between sub-blocks
Publication Date: 2019.06.04 MICRON TECHNOLOGY INC
  • US10310743B2 patent drawing
  • US10310743B2 patent drawing
  • US10310743B2 patent drawing

AI summary

A semiconductor device includes a first functional block. The first functional block includes one or more input buffers configured to receive signals, and one or more flip-flops configured to receive the signals from the one or more input buffers and output the received signals from the first functional block. The semiconductor device also includes a second functional block coupled to the first functional block. The second functional block includes a decode logic configured to directly receive the output signals from the one or more flip-flops of the first functional block.