Processor Soft Error Detection via Simulation-Based Error Injection

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Designing computer processors to effectively detect and handle soft errors, such as bit-flips caused by radiation events, is challenging, especially in large logic state-space designs where error detection and handling logic behaves differently for various instructions and data, making simulation-based testing complex.

Innovation Solution

A method involving monitoring a baseline simulation to identify a target processing cycle for error injection, performing test simulations with injected errors, and determining if the error is detected by the processor's error-checking logic, using a software model constructed in a hardware description language like VHDL to assess soft error handling capabilities.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Reliability

If soft error detection and handling logic is incorporated into computer processor designs, then reliability against soft errors is improved, but device complexity increases

Engineering Contradiction:
Improvesoft error detection capabilityVSAvoidprocessor design complexity
Core Design Contradiction:
ReliabilityVSDevice complexity

Solution Approach 1:

The patent uses simulation models (copies) of the processor design to test soft error handling without modifying the actual processor architecture. By creating virtual replicas in a software simulation environment, the invention enables thorough error detection testing while avoiding the complexity overhead of implementing physical error detection hardware in the processor itself.

Inventive Principle:
Principle #26Copying

2Measurement precision

If comprehensive soft error testing is performed across all instructions and data, then measurement precision of error detection is improved, but productivity of the testing process deteriorates

Engineering Contradiction:
Improveerror detection accuracyVSAvoidtesting efficiency
Core Design Contradiction:
Measurement precisionVSProductivity

Solution Approach 1:

The patent performs preliminary identification of target processing cycles and error injection targets through baseline simulation before conducting actual error injection tests. This preliminary action filters and prepares the test data, so that when comprehensive testing is performed, it focuses on pre-identified critical paths and instructions, maintaining high measurement precision while improving testing productivity by avoiding brute-force testing of all possible states.

Inventive Principle:
Principle #10Preliminary action

Data Source

PatentUS10324816B2Checking a computer processor design for soft error handling
Publication Date: 2019.06.18 INTERNATIONAL BUSINESS MACHINE CORPORATION
  • US10324816B2 patent drawing
  • US10324816B2 patent drawing
  • US10324816B2 patent drawing

AI summary

Checking a computer processor design for soft error handling. A baseline simulation of a computer processor design is monitored to identify a target processing cycle of the baseline simulation during which a predefined event occurs during the baseline simulation. The baseline simulation is performed in accordance with a software model of the computer processor design, and the event is associated with processing an instruction that directly involves a predefined error injection target. A test simulation of the computer processor design is performed in accordance with the software model of the computer processor design. An error is injected into the predefined error injection target during a target processing cycle of the test simulation. A determination is made as to whether the error is detected by error-checking logic of the computer processor design.