Dynamic Critical Path Detector for IC Timing

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Semiconductor integrated circuit chips designed to handle worst-case process variations often result in performance reductions, power wastage, and increased design time due to the need for excessive timing margins and power consumption, with existing monitoring methods providing coarse measurements and not continuously monitoring critical paths.

Innovation Solution

A method involving latches and a comparator to monitor critical paths in real-time, generating delayed data signals to detect timing failures and execute corrective measures, allowing for precise power management and reduced power consumption by operating closer to nominal conditions.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If semiconductor integrated circuit chips are designed to handle worst-case process variations, then reliability is improved, but performance is reduced and power consumption increases

Engineering Contradiction:
Improvechip reliability under worst-case conditionsVSAvoidpower consumption
Core Design Contradiction:
ReliabilityVSUse of energy by moving object

Solution Approach 1:

The patent implements dynamic voltage and frequency scaling based on real-time monitoring of actual process conditions. The system transitions from static worst-case design to dynamic adaptation, adjusting operating parameters according to measured temperature, voltage, and process variations, thereby reducing power consumption while maintaining reliability under actual operating conditions

Inventive Principle:
Principle #15Dynamics

Solution Approach 2:

The system continuously monitors process parameters (temperature, voltage, timing margins) and dynamically adjusts operating parameters (clock frequency, voltage levels) to match actual conditions. This replaces the fixed worst-case parameter settings with adaptive parameter optimization, reducing power consumption while maintaining reliability

Inventive Principle:
Principle #35Parameter changes

2Reliability

If timing margins are increased to ensure operation under worst-case conditions, then reliability is improved, but design time increases and chip die area increases

Engineering Contradiction:
Improvetiming closure under worst-case conditionsVSAvoiddesign time
Core Design Contradiction:
ReliabilityVSLoss of time

Solution Approach 1:

The patent implements self-monitoring and self-adjusting mechanisms where the chip automatically detects timing margin utilization and adjusts its operation accordingly. The system includes on-chip sensors and control logic that continuously measure actual timing margins and autonomously optimize performance without requiring external intervention or conservative design margins, thereby reducing design time while maintaining reliability

Inventive Principle:
Principle #25Self-service

3Use of energy by moving object

If performance scan-ring oscillators are used to monitor chip performance, then power management is enabled, but measurement precision is reduced due to coarse monitoring

Engineering Contradiction:
Improvepower management capabilityVSAvoidperformance measurement accuracy
Core Design Contradiction:
Use of energy by moving objectVSMeasurement precision

Solution Approach 1:

The patent divides the monitoring function into multiple independent sensors distributed throughout the chip, each monitoring specific critical paths or regions. This segmented approach provides fine-grained, high-precision measurements of local conditions, replacing the single coarse measurement point of PSROs while enabling detailed power management based on actual performance data from different chip regions

Inventive Principle:
Principle #1Segmentation

4Reliability

If chips are designed for worst-case extremes, then reliability under extreme conditions is improved, but area increases and time to market increases

Engineering Contradiction:
Improveoperation reliability at worst-case extremesVSAvoidchip die area
Core Design Contradiction:
ReliabilityVSArea of stationary object

Solution Approach 1:

The system incorporates self-diagnosis and self-adjustment capabilities that allow the chip to identify and compensate for process variations in real-time. By implementing on-chip sensors, control logic, and adaptive mechanisms, the chip achieves worst-case reliability without requiring oversized design margins, thereby reducing die area while maintaining reliability under extreme conditions

Inventive Principle:
Principle #25Self-service

Data Source

PatentUS8132136B2Dynamic critical path detector for digital logic circuit paths
Publication Date: 2012.03.06 MARVELL ASIA PTE LTD
  • US8132136B2 patent drawing
  • US8132136B2 patent drawing
  • US8132136B2 patent drawing

AI summary

Method for correcting timing failures in an integrated circuit and device for monitoring an integrated circuit. The method includes placing a first and second latch near a critical path. The first latch has an input comprising a data value on the critical path. The method further includes generating a delayed data value from the data value, latching the delayed data value in the second latch, comparing the data value with the delayed data value to determine whether the critical path comprises a timing failure condition, and executing a predetermined corrective measure for the critical path. The invention is also directed to a design structure on which a circuit resides.