Accelerating Embedded Device Development With Eclipse ThreadX

When Microsoft contributed the Azure RTOS to the Eclipse Foundation in November 2023, it was an exciting development. The move, which created the Eclipse ThreadX project, represented the inception of the first and only open source real-time operating system certified for safety-critical applications. Not only was this a first for the open source world, but it was built on a versatile and mature technology that has been at the forefront of embedded systems development since 1997. 

Since the initial migration, the project has looked quiet from the outside. But a lot of work has been done behind the scenes to build the infrastructure needed to help the ThreadX project continue to move forward, integrate with other projects, and provide value in the context of open source. The very uniqueness of ThreadX that makes it such an exciting and potentially valuable project also posed some challenges to ensuring it thrived in the open source environment. 

There have been several major milestones in the last year that have set the stage for the ThreadX project to get started on some exciting developments. Two of the most notable are the creation of the ThreadX Alliance and the launch of the ThreadX Special Interest Group (SIG). 

ThreadX Alliance Provides Certification Infrastructure

The first release from the ThreadX project at Eclipse was version 6.4.1, made available on 28 February 2024. This completed the projet’s transition to open source. There were only a handful of technical changes from the previous version, 6.4.0. 

One essential characteristic of ThreadX, and crucial to its future, is safety certification. As a real-time operating system, ThreadX targets several safety-critical embedded use cases. In-vehicle deployment is a good example. Unsurprisingly, in this case, the software must be certified to a very high standard, specifically the IEC 61508 and ISO 26262 standards. 

Obtaining these certifications is expensive, and they need to be maintained as releases continue to be published. Not every release of ThreadX will need to be certified, but certification is going to be a recurring cost that the project must bear. Open source projects usually tackle this with sponsorhips while open source Foundations usually leverage memberships. But the embedded developers who build commercial solutions are not used to this kind of arrangement: they’re more used to purchasing licenses and support contracts, for example.

The ThreadX Alliance was created to provide a middle ground. Organisations interested in using ThreadX can pay for a membership, and then license the safety artifacts they need. This creates a recurring revenue stream to support ThreadX’s certification efforts, while providing embedded developers a business model they are familiar with.

Special Interest Group Ensures SDV Voice Is Heard

Unsurprisingly, one of the most exciting potential use cases for ThreadX is in software-defined vehicles (SDVs). That’s why the Eclipse Software Defined Vehicle Working Group (SDV WG) created the ThreadX SIG to ensure that productive conversations could happen between the project itself, as well as the interested members of the SDV WG. 

To be clear, the SIG doesn’t have technical decision-making power. That, as always, remains with the committers driving the project forward. The existence of the SIG is really meant to ensure that the committers have an idea of the feature set and technology direction most valuable and interesting to the SDV community. 

Introduction to Open Source Helped Project Get Underway

Because of the project’s somewhat unique position in the broader open source ecosystem, it includes support and contribution from a wide variety of contributors, not all of whom are familiar with the open source process. That’s why the Eclipse Foundation decided I should join the project as project lead. 

It’s not common for Eclipse Foundation staff to join projects in such a formal role. As a rule, we prefer to avoid being in positions of influence in projects, as our role is really as custodians, not as agenda setters. But in this case, it was decided it would be helpful if someone could come in and help the team understand the open source production model, the Eclipse Foundation processes, and so on. 

It's been a big learning experience for me. As a result, we’ve been able to get the project focused and moving in the right direction. And one thing that’s become clear to me as I’ve gotten more involved is that there’s tons of potential for ThreadX in the market, and it’s very exciting to be a part of that. We progressed a lot as a team, and are now ready to put new ideas into practice.

Work Underway to Expand ThreadX’s Scope and Accessibility 

Obviously, as with any project, there’s work currently underway, but also both plans and possibilities for the future. Some immediate work is preparing the release of  ThreadX v6.4.2, which will be a small service release updating a subset of the core components. That release, scheduled for the end of Q1 2025, is mainly focused on fixing issues brought to our attention by the community. 

Another immediate goal was to release an AsciiDoc version of the project documentation and publish it on our website. It’s in fact already available on GitHub since early February. This move will enable us to easily produce an HTML version from the source of the documentation. 

In the next few months, we will definitely implement support for additional microcontrollers and platforms. Unlike in the PC world, where there are only three big processor makers, the embedded ecosystem is incredibly diverse, so support for all those different platforms and architectures is necessary for industry adoption.

Opportunities for Inroads and Improvements Abound

One really exciting development and opportunity for the ThreadX project is the creation of the OpenHW Foundation. OpenHW joined the Eclipse Foundation last year, bringing its expertise in the RISV-V open hardware architecture into the software domain. This represents a huge opportunity for ThreadX. Our existing intersections with the automotive and IoT industries mean there’s an opportunity to build a comprehensive RISC-V based ecosystem, from the level of chips all the way up to supporting platforms in the cloud. That story will actually form the core of our presence at the Embedded World this year. 

Another potential development I would really like to see is making the GUIX Studio, a tool used for building user interfaces, be made cross-platform. Unfortunately, because it’s written in Windows-specific technologies, this will likely involve rewriting the entire thing. My dream scenario would be to have a set of extensions that could be deployed in our Eclipse Theia IDE. 

One last wish I have is that, by the end of the year, we’ve established a predictable release cadence and grown our contributor and committer base. We’ve had a lot of growth already, which is fantastic, but now that a lot of the groundwork has been laid, it’s an excellent time to get involved. 

Involvement can take many forms. You can get directly involved with the project itself, sign up for the general mailing list or the developer mailing list, or contact us about joining the ThreadX Alliance.

So, please join us and let’s make Eclipse ThreadX a success together!