Talks 2025

The project of CMRX was started as an experiment if hassle-free approach to memory isolation on MMU-less microcontrollers is technically possible and feasible.

Motivation that fueled this experiment was me witnessing how much hassle it is to maintain a reasonably cybersecure project running on a microcontroller using contemporary modern tech stack. It turned out that memory isolation can both be hassle-free and still perform fast even on $1 microcontroller. It just works without consuming 80% of your CPU. That’s a huge win.

But technical feasibility is just one of the aspects. Embedded industry is extremely conservative and short-sighted. You’ll find people arguing that true pro embedded work should ever only be done using assembly. Thus, this year I decided to spread the word about CMRX with conference talks. These talks had two main purposes:

• To validate the idea with community of developers and collect valuable feedback on use cases that might not be supported. You’d be surprised how hard it is to get ANY feedback these days.
• To let the community know that memory isolation does not need to be hard and the suffering often connected with managing memory partitioning.

The tour-de-conferences had 4 stops, five if I include EurOpen conference where I’ve been invited to talk on the topic of memory isolation in general:

• FOSDEM ‘25 in Brussels - talk in Microkernel and component-based devroom to the community of microkernel developers as CMRX is a microkernel too
• Bratislava OpenCamp - contribution to the embedded portion of the conference with the audience of embedded developers
• OpenAlt z.s. in Brno - one of the only two talks that failed to be recorded :)
• ESE Kongress in Sindelfingen - here the audience was mixed safety and non-safety and provided some valuable input

In these talks I’ve been asking one question: How many of you did work with memory isolation? Only few hands raised. Which means that the industry still lives in oblivion and denial. We don’t need this, we don’t want this and it is complicated.

That’s why I will continue in year 2026 as long as there will be bandwidth to give talks. See you at FOSDEM 2026!

Presentation slides used in ESE Kongress talk are available for download: