Main Content

PIL Execution at the Command Line Using MATLAB Coder

You can use processor-in-the-loop (PIL) executions to verify generated code that you deploy to hardware board by using a MATLAB® Coder™ procedure. You can profile algorithm performance and speed of your generated code on the hardware board. To verify generated code, you must have MATLAB Coder and an Embedded Coder® license.

This example shows how to set up a PIL execution to verify generated code from a MATLAB function using MATLAB Coder at the command line.

This PIL execution is also available with these hardware support packages.

  • Embedded Coder Support Package for BeagleBone® Black Hardware

  • Embedded Coder Support Package for ARM® Cortex®-A Processors

  • Embedded Coder Support Package for STMicroelectronics® Discovery Boards

  1. In the Command Window, select the hardware for PIL execution.

    hw = coder.hardware('ARM Cortex-M3 (QEMU)')
    
    hw = 
    
      Hardware with properties:
    
                Name: 'ARM Cortex-M3 (QEMU)'
        CPUClockRate: 1000
  2. Add the hardware to the MATLAB Coder configuration object.

    cfg = coder.config('lib','ecoder',true); 
    cfg.VerificationMode = 'PIL'; 
    cfg.Hardware = hw; 
    
  3. The memory of the hardware board is limited, and a default value of 20000 is beyond the stack size available in the hardware. We recommend you to set the StackUsageMax value to 512. However, specify an appropriate stack usage based on your algorithm.

    cfg.StackUsageMax = 512;
  4. Generate PIL code for a function, averaging_filter.

    codegen -config cfg averaging_filter -args {zeros(1,16)}

For more information on the averaging_filter function, see the example Generate C/C++ Code from a MATLAB Function.

To install the Embedded Coder Support Package for ARM Cortex-M Processors, see Embedded Coder Supported Hardware.