Parallel Emulation Processor Configuration via Segmented Registers

Resolve Bottlenecks,
Find Innovative Solutions
Generate Solutions

Solution Overview

Problem

Conventional systems are inefficient in testing and validating complex computing hardware due to high costs and time requirements, as they cannot effectively and efficiently configure emulation processors to ensure correct and predictable operation.

Innovation Solution

A hardware-accelerated configuration of multiple emulation processors that allows for rapid self-configuration by modifying registers with executable instructions, enabling parallel and distributed processing to reduce configuration time through efficient bulk data transfer and processor-level communication.

Engineering Contradictions & Design Principles

VSEngineering Contradiction Analysis

1Loss of time

If conventional systems are used to configure emulation processors, then configuration can be performed with existing tools, but the time required to configure and test computing hardware increases rapidly

Engineering Contradiction:
Improveconfiguration timeVSAvoidtesting and validation efficiency
Core Design Contradiction:
Loss of timeVSProductivity

Solution Approach 1:

The configuration process is divided into multiple independent configuration records, each containing specific update operations for emulation processors. These segmented configuration records can be processed in parallel, allowing simultaneous configuration of multiple processors rather than sequential configuration, thus reducing total configuration time while maintaining testing efficiency

Inventive Principle:
Principle #1Segmentation

Solution Approach 2:

Configuration records are prepared and staged in advance before actual emulation processor configuration occurs. The host system generates and queues multiple configuration records with all necessary update operations predetermined, allowing the emulation processors to execute pre-planned configuration sequences in parallel without waiting for real-time instruction generation, thereby reducing configuration time while preserving testing productivity

Inventive Principle:
Principle #10Preliminary action

2Speed

If sequential configuration of emulation processors is used, then system complexity is reduced, but the speed of configuration decreases

Engineering Contradiction:
Improveconfiguration speedVSAvoidconfiguration system complexity
Core Design Contradiction:
SpeedVSDevice complexity

Solution Approach 1:

A host system acts as an intermediary between the configuration source and multiple emulation processors. The host generates configuration records, manages their queuing, and distributes them to multiple emulation processors for parallel execution. This intermediary coordinates the parallel configuration process, enabling high-speed configuration of multiple processors while the host manages the overall system complexity rather than requiring complex point-to-point control logic in each processor

Inventive Principle:
Principle #24Intermediary (Mediator)

3Productivity

If parallel configuration of multiple emulation processors is implemented, then configuration time is reduced, but the complexity of data transfer and coordination increases

Engineering Contradiction:
Improveconfiguration throughputVSAvoiddata transfer mechanism complexity
Core Design Contradiction:
ProductivityVSDevice complexity

Solution Approach 1:

Multiple configuration operations for different emulation processors are merged into a single bulk data transfer operation. The host system packages multiple configuration records containing update operations for multiple processors into one consolidated data transfer, which is then distributed to the respective processors. This merging approach achieves high configuration throughput while simplifying the data transfer mechanism compared to multiple separate transfer operations

Inventive Principle:
Principle #5Merging (Combining)

Data Source

PatentUS11868786B1Systems and methods for distributed and parallelized emulation processor configuration
Publication Date: 2024.01.09 CADENCE DESIGN SYST INC
  • US11868786B1 patent drawing
  • US11868786B1 patent drawing
  • US11868786B1 patent drawing

AI summary

Implementations may include a method of accelerated modification of an emulation processor system, by loading, by a first emulation processor, a first portion of processor instructions into one or more registers of the first emulation processor, in response to a selection of a first programming mode associated with the first emulation processor, and loading, by a second emulation processor operatively coupled with the first emulation processor, a second portion of the processor instructions into one or more registers of the second emulation processor, in response to a selection of a first programming mode associated with the second emulation processor.