La traduction de cette page n'est pas à jour. Cliquez ici pour voir la dernière version en anglais.

Cibles de processeurs multicœurs

Modéliser l'exécution concurrente de systèmes conçus pour être déployés dans un environnement multicœur ou FPGA

La programmation multicœur ou la modélisation pour une exécution concurrente permet de créer des systèmes concurrents pouvant être déployer dans des environnements à processeur multicœur ou multiprocesseurs, par exemple des systèmes de traitement du signal ou de contrôle. Les techniques de partitionnement et de mapping de Simulink^® aident à surmonter les problèmes couramment associés au design de systèmes pour une exécution concurrente.

La figure présente un exemple de système comportant plusieurs fonctions conçues pour s'exécuter dans un environnement multiprocesseur CPU et FPGA. Ce système est partitionné en différents composants mappés sur le planificateur de tâches du CPU ou sur le FPGA.

Pour découvrir les fondamentaux de la programmation multicœur, consultez la rubrique Concepts in Multicore Programming. Pour plus d'informations sur le design de systèmes pour une exécution concurrente dans Simulink, consultez la rubrique Multicore Programming with Simulink.

Fonctions

`Simulink.architecture.config`	Create or convert configuration for concurrent execution
`Simulink.architecture.add`	Add tasks or triggers to selected architecture of model
`Simulink.architecture.delete`	Delete triggers and tasks from selected architecture of model
`Simulink.architecture.find_system`	Find objects under architecture object
`Simulink.architecture.get_param`	Get configuration parameters of architecture objects
`Simulink.architecture.importAndSelect`	Import and select target architecture for concurrent execution environment for model
`Simulink.architecture.profile`	Generate profile report for model configured for concurrent execution
`Simulink.architecture.register`	Add custom target architecture to concurrent execution target architecture selector
`Simulink.architecture.set_param`	Set architecture object properties

Objets

Simulink.GlobalDataTransfer Configure concurrent execution data transfers

Rubriques

Comprendre la programmation multicœur dans Simulink

Concepts in Multicore Programming
Theory relevant to modeling for concurrent execution.
Multicore Programming with Simulink
Modeling for concurrent execution using Simulink.
Implicit and Explicit Partitioning of Models
Learn about the key differences between implicit and explicit partitioning.
Supported Targets for Multicore Programming
Deploy concurrent execution models to supported multicore targets.
Limitations with Multicore Programming in Simulink
Limitations and considerations when partitioning a model for concurrent execution.

Configurer votre modèle

Configure Your Model for Concurrent Execution

Learn how to configure your Simulink model to take advantage of concurrent execution.
- Concurrent Execution Tool
- Data Transfer Options for Concurrent Execution
Specify a Target Architecture
Choose or define a target architecture for a model configured for concurrent execution.
Partition Your Model Using Explicit Partitioning
Add tasks, create partitions, and map individual tasks to partitions using explicit partitioning.
Configure Data Transfer Settings Between Concurrent Tasks
Specify options for handling data transfers between concurrently executing partitions.

Implémenter le parallélisme dans Simulink

Implement Data Parallelism in Simulink
This example shows how to implement data parallelism for a system in a Simulink model.
Implement Task Parallelism in Simulink
This example shows how to implement task parallelism for a system represented in a Simulink® model.
Implement Pipelining in Simulink
This example shows how to implement pipelining for a system represented in a Simulink® model.

Génération de code

Optimize and Deploy on a Multicore Target
Configure a model for concurrent execution using explicit partitioning and deploy it to a target.
Assign Tasks to Cores for Multicore Programming
This example shows how to take advantage of executing code on a multicore processor by graphical partitioning.

Informations connexes

Multicore Programming with Simulink

Exemples présentés

Assign Tasks to Cores for Multicore Programming

Take advantage of executing code on a multicore processor by graphical partitioning. This example requires Simulink® Coder™ software to generate multithreaded code.

Ouvrir l'exemple

Implement an FFT on a Multicore Processor and an FPGA

Take advantage of a multicore processor target with FPGA acceleration by graphically partitioning a model. This example requires Simulink Coder™ to generate multi-threaded code and HDL Coder™ to generate HDL code. You cannot generate HDL code on Macintosh systems.

Ouvrir le modèle

Multicore Programming of a Field-Oriented Control on Zynq

Demonstrates how to implement a control algorithm that contains multiple rates on Zynq®. To take advantage of both the cores and the FPGA hardware, the example uses a graphical partitioning approach such that code from different partitions is distributed across the cores and the hardware.

Ouvrir le modèle

Multicore Deployment of a Plant Model

Illustrates how to take advantage of executing multithreaded code on a multicore processor using graphical partitioning. This example requires Simulink® Coder™ to generate multithreaded code.

Ouvrir le modèle

Modeling Objects with Identical Dynamics Using For Each Subsystem

Model multiple objects with identical dynamics using the For Each subsystem. The number of objects is parameterized by the length of the input signal.

Ouvrir le modèle

Vectorizing a Scalar Algorithm with a For-Each Subsystem

Use the for-each subsystem. In this example, the operations are performed on a vector for simplicity.

Ouvrir le modèle

Tiled Processing of 2-D Signals with For-Each Subsystem

Use a for-each subsystem. In this example, the operations are performed on matrices.

Ouvrir le modèle

Multithreaded Simulation Using For Each Subsystem

Speed up execution of a model on multiple cores using a For Each Subsystem in Rapid Accelerator simulation.

Ouvrir le modèle