AikaRTOS is a simple educational RTOS project originally targeting the STM32 NUCLEO-F411RE board, now also tested on the NUCLEO-H753ZI.
It is designed to explore basic task management, task scheduling, and memory allocation on embedded systems.
The project is written in C++20.
Note: This is a purely educational project. It was originally developed using a NUCLEO-F411RE board for experimentation purposes. Recently tested and confirmed to run on NUCLEO-H753ZI as well, though full multi-platform support is still a work in progress.
- Basic task switching
- Multiple scheduler implementations (e.g., Round-Robin, Fixed Priority)
- Simple memory allocator implementations (e.g., bump, free list)
- Simple startup sequence
- Designed for STM32 Cortex-M4/M7 (tested on F411RE and H753ZI)
- Written in C++20
All examples can be found in the
aikartos/src/testsfolder.
| Example | Description |
|---|---|
round_robin.cpp | Demonstrates basic Round-Robin task switching between three simple infinite loops. |
edf.cpp | Demonstrates Earliest Deadline First (EDF) scheduling with tasks having different deadlines. |
fixed_priority.cpp | Demonstrates Fixed Priority scheduling where tasks are executed based on static priorities. |
lottery.cpp | Demonstrates Lottery Scheduling where tasks are chosen randomly based on ticket allocation. |
priority_aging.cpp | Demonstrates Priority Scheduling with Aging to prevent starvation of low-priority tasks. |
weighted_lottery.cpp | Demonstrates Weighted Lottery Scheduling where tasks have different chances of being selected based on weight. |
stack_overflow.cpp | Demonstrates system behavior when a stack overflow occurs in a task. Useful for testing robustness. |
producer_consumer.cpp | Demonstrates a simple Producer-Consumer system using a shared lock-free queue and cooperative task switching. |
coop_preemptive.cpp | Demonstrates hybrid Cooperative-Preemptive scheduling where each task can have its own quantum or run cooperatively. |
sch_cfs_like.cpp | Demonstrates a CFS-like scheduler where tasks are selected based on the smallest virtual runtime to ensure balanced CPU time distribution. |
sch_mlfq.cpp | Demonstrates a Multilevel Feedback Queue scheduler with per-task quantum levels and automatic priority boosting. |
memory_allocator_bump.cpp | Demonstrates a simple bump allocator used to manage memory in a linear fashion. |
memory_allocator_free_list.cpp | Demonstrates a basic free-list memory allocator with support for reuse and fragmentation handling. |
memory_allocator_dlist.cpp | Demonstrates a double-linked free-list allocator with bidirectional coalescing and minimal overhead on allocation. |
memory_allocator_buddy.cpp | Tests an arena-based buddy allocator (buddy::impl::region) that stores all metadata in external memory. |
memory_allocator_buddy_fixed.cpp | Tests a statically bounded buddy allocator (buddy::impl::fixed) with internal free list table. |
memory_allocator_tlsf.cpp | Tests a dynamic TLSF allocator (tlsf::impl::region) that places bucket index tables inside the managed memory block. |
memory_allocator_tlsf_fixed.cpp | Tests a statically sized TLSF allocator (tlsf::impl::fixed) with a fully internalized index table stored in std::array. |
🎮🎮🎮
A tiny cooperative Snake game demo running on AikaRTOS. Implements movement via timer-driven tasks, showcasing clean task switching and input handling.
A small floating-point demo with smooth motion and fading trail. Uses trigonometric FPU math and per-frame task updates to draw a glowing animated path in the console. Demonstrates how one task can safely use the FPU while others do not, under a preemptive round-robin scheduler.
This project is licensed under the MIT License.
To build and run this project, you will need:
- ARM toolchain (e.g.,
arm-none-eabi-gcc,arm-none-eabi-newlib) - STM32Cube Firmware Package for the selected platform (F4 or H7)
This includes the necessary device headers like<stm32f4xx.h>or<stm32h7xx.h>and low-level drivers.
You can download the STM32Cube packages via:
This project uses CMake as its build system and can be built with Ninja, Make, or STM32CubeIDE.
- Clone the repository:
git clone https://.com/newenclave/aikartos.git
cd aikartos- Configure the build (an example):
cmake -G "Ninja" -B build_f411 \
-DCMAKE_TOOLCHAIN_FILE=gcc-arm-none-eabi.cmake \
-DPLATFORM=f411 \
-DCMSIS_DIRECTORY="C:/soft/ST/STM32Cube_FW_F4/STM32Cube_FW_F4_V1.28.0/Drivers/CMSIS"Note: paths must match your environment.
- Build
cd build_f411
ninja- Flash using your preferred tool (e.g., ST-Link, OpenOCD, CubeProgrammer).