EoS RTOS Roadmap 2026: Tickless Idle, RT-IPC, Formal Verification
Three large RTOS research targets for 2026: a tickless scheduler targeting sub-microsecond wake latency, RT-IPC primitives sharing memory across security domains, and a formally verified context-switch path.
1. Tickless idle
The current EoS scheduler uses a 1 ms tick interrupt. This prevents the CPU from entering deep sleep for more than 1 ms, wasting power on battery devices. The 2026 tickless scheduler programs the RTC to wake the CPU only when the next task deadline arrives. On a typical IoT workload (1 Hz sensor read, 10 Hz display update), this reduces idle power from 8 mA to 0.4 mA — a 20× improvement.
2. RT-IPC across security domains
Today, eIPC uses copy-based message passing across MPU domain boundaries. For high-bandwidth use cases (neural data, video), copy overhead is prohibitive. RT-IPC introduces a shared-memory window with hardware-enforced read/write permissions: the producer domain has write access, the consumer domain has read access, and neither can access the other's private memory. This enables zero-copy neural data transfer at 61 MB/s.
3. Formal verification of the context switch
The context switch is the most security-critical code in any RTOS. A bug here can corrupt task state, leak secrets across security domains, or enable privilege escalation. We are using TLA+ to specify the context-switch state machine and Coq to prove that the implementation matches the spec. The verified context switch will ship in EoS 2.0.
What's not on the list
We are explicitly not adding a POSIX compatibility layer, a dynamic linker, or a general-purpose memory allocator in 2026. These features would increase the kernel's attack surface and binary size. EoS remains a purpose-built embedded RTOS, not a general-purpose OS.
About these articles
How these articles are published
Every article on this site renders through this one page, from the Foundation's shared content registry. The title, date, badge, tags, and body all come from the same data the News page lists — which is why the two can never disagree about what an article says or when it appeared.
The registry replaced eight separate article pages whose metadata had drifted apart: seven of them showed a different publication date than the news listing. One source of truth is less interesting than eight bespoke pages, and considerably more honest.
What an article is
Articles are the long form: explanations of how something works, technical writing about the stack, and reflections on the work. News carries the timely announcement; the article carries the full treatment — the reasoning, the context, and the detail that would not fit in a listing.
Each article states its assumptions and links its evidence where evidence exists. The source material behind the technical claims — the code, the measurements, the filings — is public in the Foundation's repositories.
Dates and addresses
The date on an article is its publication date. The address is /article/ followed by the article's slug, and the older /article-xxx addresses from the previous site still resolve to the same pages — nothing already linked or indexed breaks.
If an article is corrected after publication, the correction is noted rather than silently made — the same standard our press releases are held to. The page you are reading is the current version, and the record of what changed stays with it.