Is it possible to get the source code for the Linux kernel that MathWorks uses to create the images for the Zynq support packages?

I am using one of the Zynq support packages that comes with a MathWorks Linux image, but I need to either use my own Linux kernel or modify the one that is downloaded to the SD during the Hardware Setup of that support package. When I try to integrate the generated code into our own kernel, I get errors that I believe are due to using different drivers. Can we get the Linux kernel code that MathWorks uses?

Answer 1

[MathWorks Support Team]

The MathWorks buildroot repository can be found on GitHub and contains the source code for all Linux images:

https://github.com/mathworks/buildroot

The buildroot framework specifically for each support package can be found here:

• Embedded Coder Support Package for Xilinx Zynq Platform: https://github.com/mathworks/buildroot/tree/mathworks_zynq_R21.1.0
• Communications Toolbox Support Package for Xilinx Zynq-Based Radio https://github.com/mathworks/buildroot/tree/mathworks_zynq_iiosdr_R20.1.0
• Vision HDL Toolbox Support Package for Xilinx Zynq-Based Hardware: https://github.com/mathworks/buildroot/tree/mathworks_zynq_vision_R21.1.0

Starting with the MATLAB R2024a, the features of the Xilinx Zynq Support from Vision HDL Toolbox will be available through Xilinx Zynq SoC Support from SoC Blockset, following is the link: 

https://www.mathworks.com/hardware-support/xilinx-soc.html

For releases prior to MATLAB R2024a, unlike SoC Blockset, the above support packages do not explicitly support adding custom boards or Linux image customization, but the above repository tag should give you a starting point to investigate your own solution.