Embedded SML using the MLton compiler

TL;DR

This paper discusses adapting the MLton compiler for embedded and real-time systems by modifying its runtime components and integrating it into lightweight OS kernels, highlighting practical implementation experiences.

Contribution

It presents specific modifications to MLton's runtime and integration techniques for embedded and real-time environments, extending previous multicore scalability work.

Findings

Modified threading subsystem and garbage collector for embedded use

Successful integration of MLton programs into lightweight OS kernels

Insights into challenges and solutions for real-time system support

Abstract

In this extended abstract we present our current work on leveraging Standard ML for developing embedded and real-time systems. Specifically we detail our experiences in modifying MLton, a whole program, optimizing compiler for Standard ML, for use in such contexts. We focus primarily on the language runtime, re-working the threading subsystem and garbage collector, as well as necessary changes for integrating MLton generated programs into a light weight operating system kernel. We compare and contrast these changes to our previous work on extending MLton for multicore systems, which focused around acheiving scalability.