Keyboard shortcuts

Press or to navigate between chapters

Press S or / to search in the book

Press ? to show this help

Press Esc to hide this help

Aether

Licensing

LICENSE: AETHER OPERATING SYSTEM

The Aether project currently uses a singular license GPLv2.0.

The project maintainer(s) reserves the right to change the license in the future without prior notice.

Note that all licenses used by this project are available in the LICENSES/ directory in the source code.


Copyright (c) 2026 Richard Tichý

Used licenses

These are the licenses used by this project.

GPLv2.0

Valid-License-Identifier: GPL-2.0
Valid-License-Identifier: GPL-2.0-only
Valid-License-Identifier: GPL-2.0+
Valid-License-Identifier: GPL-2.0-or-later
SPDX-URL: https://spdx.org/licenses/GPL-2.0.html
Usage-Guide:
  To use this license in source code, put one of the following SPDX
  tag/value pairs into a comment according to the placement
  guidelines in the licensing rules documentation.
  For 'GNU General Public License (GPL) version 2 only' use:
    SPDX-License-Identifier: GPL-2.0
  or
    SPDX-License-Identifier: GPL-2.0-only
  For 'GNU General Public License (GPL) version 2 or any later version' use:
    SPDX-License-Identifier: GPL-2.0+
  or
    SPDX-License-Identifier: GPL-2.0-or-later
License-Text:

		    GNU GENERAL PUBLIC LICENSE
		       Version 2, June 1991

 Copyright (C) 1989, 1991 Free Software Foundation, Inc.,
 <https://fsf.org/>
 Everyone is permitted to copy and distribute verbatim copies
 of this license document, but changing it is not allowed.

			    Preamble

  The licenses for most software are designed to take away your
freedom to share and change it.  By contrast, the GNU General Public
License is intended to guarantee your freedom to share and change free
software--to make sure the software is free for all its users.  This
General Public License applies to most of the Free Software
Foundation's software and to any other program whose authors commit to
using it.  (Some other Free Software Foundation software is covered by
the GNU Library General Public License instead.)  You can apply it to
your programs, too.

  When we speak of free software, we are referring to freedom, not
price.  Our General Public Licenses are designed to make sure that you
have the freedom to distribute copies of free software (and charge for
this service if you wish), that you receive source code or can get it
if you want it, that you can change the software or use pieces of it
in new free programs; and that you know you can do these things.

  To protect your rights, we need to make restrictions that forbid
anyone to deny you these rights or to ask you to surrender the rights.
These restrictions translate to certain responsibilities for you if you
distribute copies of the software, or if you modify it.

  For example, if you distribute copies of such a program, whether
gratis or for a fee, you must give the recipients all the rights that
you have.  You must make sure that they, too, receive or can get the
source code.  And you must show them these terms so they know their
rights.

  We protect your rights with two steps: (1) copyright the software, and
(2) offer you this license which gives you legal permission to copy,
distribute and/or modify the software.

  Also, for each author's protection and ours, we want to make certain
that everyone understands that there is no warranty for this free
software.  If the software is modified by someone else and passed on, we
want its recipients to know that what they have is not the original, so
that any problems introduced by others will not reflect on the original
authors' reputations.

  Finally, any free program is threatened constantly by software
patents.  We wish to avoid the danger that redistributors of a free
program will individually obtain patent licenses, in effect making the
program proprietary.  To prevent this, we have made it clear that any
patent must be licensed for everyone's free use or not licensed at all.

  The precise terms and conditions for copying, distribution and
modification follow.

		    GNU GENERAL PUBLIC LICENSE
   TERMS AND CONDITIONS FOR COPYING, DISTRIBUTION AND MODIFICATION

  0. This License applies to any program or other work which contains
a notice placed by the copyright holder saying it may be distributed
under the terms of this General Public License.  The "Program", below,
refers to any such program or work, and a "work based on the Program"
means either the Program or any derivative work under copyright law:
that is to say, a work containing the Program or a portion of it,
either verbatim or with modifications and/or translated into another
language.  (Hereinafter, translation is included without limitation in
the term "modification".)  Each licensee is addressed as "you".

Activities other than copying, distribution and modification are not
covered by this License; they are outside its scope.  The act of
running the Program is not restricted, and the output from the Program
is covered only if its contents constitute a work based on the
Program (independent of having been made by running the Program).
Whether that is true depends on what the Program does.

  1. You may copy and distribute verbatim copies of the Program's
source code as you receive it, in any medium, provided that you
conspicuously and appropriately publish on each copy an appropriate
copyright notice and disclaimer of warranty; keep intact all the
notices that refer to this License and to the absence of any warranty;
and give any other recipients of the Program a copy of this License
along with the Program.

You may charge a fee for the physical act of transferring a copy, and
you may at your option offer warranty protection in exchange for a fee.

  2. You may modify your copy or copies of the Program or any portion
of it, thus forming a work based on the Program, and copy and
distribute such modifications or work under the terms of Section 1
above, provided that you also meet all of these conditions:

    a) You must cause the modified files to carry prominent notices
    stating that you changed the files and the date of any change.

    b) You must cause any work that you distribute or publish, that in
    whole or in part contains or is derived from the Program or any
    part thereof, to be licensed as a whole at no charge to all third
    parties under the terms of this License.

    c) If the modified program normally reads commands interactively
    when run, you must cause it, when started running for such
    interactive use in the most ordinary way, to print or display an
    announcement including an appropriate copyright notice and a
    notice that there is no warranty (or else, saying that you provide
    a warranty) and that users may redistribute the program under
    these conditions, and telling the user how to view a copy of this
    License.  (Exception: if the Program itself is interactive but
    does not normally print such an announcement, your work based on
    the Program is not required to print an announcement.)

These requirements apply to the modified work as a whole.  If
identifiable sections of that work are not derived from the Program,
and can be reasonably considered independent and separate works in
themselves, then this License, and its terms, do not apply to those
sections when you distribute them as separate works.  But when you
distribute the same sections as part of a whole which is a work based
on the Program, the distribution of the whole must be on the terms of
this License, whose permissions for other licensees extend to the
entire whole, and thus to each and every part regardless of who wrote it.

Thus, it is not the intent of this section to claim rights or contest
your rights to work written entirely by you; rather, the intent is to
exercise the right to control the distribution of derivative or
collective works based on the Program.

In addition, mere aggregation of another work not based on the Program
with the Program (or with a work based on the Program) on a volume of
a storage or distribution medium does not bring the other work under
the scope of this License.

  3. You may copy and distribute the Program (or a work based on it,
under Section 2) in object code or executable form under the terms of
Sections 1 and 2 above provided that you also do one of the following:

    a) Accompany it with the complete corresponding machine-readable
    source code, which must be distributed under the terms of Sections
    1 and 2 above on a medium customarily used for software interchange; or,

    b) Accompany it with a written offer, valid for at least three
    years, to give any third party, for a charge no more than your
    cost of physically performing source distribution, a complete
    machine-readable copy of the corresponding source code, to be
    distributed under the terms of Sections 1 and 2 above on a medium
    customarily used for software interchange; or,

    c) Accompany it with the information you received as to the offer
    to distribute corresponding source code.  (This alternative is
    allowed only for noncommercial distribution and only if you
    received the program in object code or executable form with such
    an offer, in accord with Subsection b above.)

The source code for a work means the preferred form of the work for
making modifications to it.  For an executable work, complete source
code means all the source code for all modules it contains, plus any
associated interface definition files, plus the scripts used to
control compilation and installation of the executable.  However, as a
special exception, the source code distributed need not include
anything that is normally distributed (in either source or binary
form) with the major components (compiler, kernel, and so on) of the
operating system on which the executable runs, unless that component
itself accompanies the executable.

If distribution of executable or object code is made by offering
access to copy from a designated place, then offering equivalent
access to copy the source code from the same place counts as
distribution of the source code, even though third parties are not
compelled to copy the source along with the object code.

  4. You may not copy, modify, sublicense, or distribute the Program
except as expressly provided under this License.  Any attempt
otherwise to copy, modify, sublicense or distribute the Program is
void, and will automatically terminate your rights under this License.
However, parties who have received copies, or rights, from you under
this License will not have their licenses terminated so long as such
parties remain in full compliance.

  5. You are not required to accept this License, since you have not
signed it.  However, nothing else grants you permission to modify or
distribute the Program or its derivative works.  These actions are
prohibited by law if you do not accept this License.  Therefore, by
modifying or distributing the Program (or any work based on the
Program), you indicate your acceptance of this License to do so, and
all its terms and conditions for copying, distributing or modifying
the Program or works based on it.

  6. Each time you redistribute the Program (or any work based on the
Program), the recipient automatically receives a license from the
original licensor to copy, distribute or modify the Program subject to
these terms and conditions.  You may not impose any further
restrictions on the recipients' exercise of the rights granted herein.
You are not responsible for enforcing compliance by third parties to
this License.

  7. If, as a consequence of a court judgment or allegation of patent
infringement or for any other reason (not limited to patent issues),
conditions are imposed on you (whether by court order, agreement or
otherwise) that contradict the conditions of this License, they do not
excuse you from the conditions of this License.  If you cannot
distribute so as to satisfy simultaneously your obligations under this
License and any other pertinent obligations, then as a consequence you
may not distribute the Program at all.  For example, if a patent
license would not permit royalty-free redistribution of the Program by
all those who receive copies directly or indirectly through you, then
the only way you could satisfy both it and this License would be to
refrain entirely from distribution of the Program.

If any portion of this section is held invalid or unenforceable under
any particular circumstance, the balance of the section is intended to
apply and the section as a whole is intended to apply in other
circumstances.

It is not the purpose of this section to induce you to infringe any
patents or other property right claims or to contest validity of any
such claims; this section has the sole purpose of protecting the
integrity of the free software distribution system, which is
implemented by public license practices.  Many people have made
generous contributions to the wide range of software distributed
through that system in reliance on consistent application of that
system; it is up to the author/donor to decide if he or she is willing
to distribute software through any other system and a licensee cannot
impose that choice.

This section is intended to make thoroughly clear what is believed to
be a consequence of the rest of this License.

  8. If the distribution and/or use of the Program is restricted in
certain countries either by patents or by copyrighted interfaces, the
original copyright holder who places the Program under this License
may add an explicit geographical distribution limitation excluding
those countries, so that distribution is permitted only in or among
countries not thus excluded.  In such case, this License incorporates
the limitation as if written in the body of this License.

  9. The Free Software Foundation may publish revised and/or new versions
of the General Public License from time to time.  Such new versions will
be similar in spirit to the present version, but may differ in detail to
address new problems or concerns.

Each version is given a distinguishing version number.  If the Program
specifies a version number of this License which applies to it and "any
later version", you have the option of following the terms and conditions
either of that version or of any later version published by the Free
Software Foundation.  If the Program does not specify a version number of
this License, you may choose any version ever published by the Free Software
Foundation.

  10. If you wish to incorporate parts of the Program into other free
programs whose distribution conditions are different, write to the author
to ask for permission.  For software which is copyrighted by the Free
Software Foundation, write to the Free Software Foundation; we sometimes
make exceptions for this.  Our decision will be guided by the two goals
of preserving the free status of all derivatives of our free software and
of promoting the sharing and reuse of software generally.

			    NO WARRANTY

  11. BECAUSE THE PROGRAM IS LICENSED FREE OF CHARGE, THERE IS NO WARRANTY
FOR THE PROGRAM, TO THE EXTENT PERMITTED BY APPLICABLE LAW.  EXCEPT WHEN
OTHERWISE STATED IN WRITING THE COPYRIGHT HOLDERS AND/OR OTHER PARTIES
PROVIDE THE PROGRAM "AS IS" WITHOUT WARRANTY OF ANY KIND, EITHER EXPRESSED
OR IMPLIED, INCLUDING, BUT NOT LIMITED TO, THE IMPLIED WARRANTIES OF
MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE.  THE ENTIRE RISK AS
TO THE QUALITY AND PERFORMANCE OF THE PROGRAM IS WITH YOU.  SHOULD THE
PROGRAM PROVE DEFECTIVE, YOU ASSUME THE COST OF ALL NECESSARY SERVICING,
REPAIR OR CORRECTION.

  12. IN NO EVENT UNLESS REQUIRED BY APPLICABLE LAW OR AGREED TO IN WRITING
WILL ANY COPYRIGHT HOLDER, OR ANY OTHER PARTY WHO MAY MODIFY AND/OR
REDISTRIBUTE THE PROGRAM AS PERMITTED ABOVE, BE LIABLE TO YOU FOR DAMAGES,
INCLUDING ANY GENERAL, SPECIAL, INCIDENTAL OR CONSEQUENTIAL DAMAGES ARISING
OUT OF THE USE OR INABILITY TO USE THE PROGRAM (INCLUDING BUT NOT LIMITED
TO LOSS OF DATA OR DATA BEING RENDERED INACCURATE OR LOSSES SUSTAINED BY
YOU OR THIRD PARTIES OR A FAILURE OF THE PROGRAM TO OPERATE WITH ANY OTHER
PROGRAMS), EVEN IF SUCH HOLDER OR OTHER PARTY HAS BEEN ADVISED OF THE
POSSIBILITY OF SUCH DAMAGES.

		     END OF TERMS AND CONDITIONS

	    How to Apply These Terms to Your New Programs

  If you develop a new program, and you want it to be of the greatest
possible use to the public, the best way to achieve this is to make it
free software which everyone can redistribute and change under these terms.

  To do so, attach the following notices to the program.  It is safest
to attach them to the start of each source file to most effectively
convey the exclusion of warranty; and each file should have at least
the "copyright" line and a pointer to where the full notice is found.

    <one line to give the program's name and a brief idea of what it does.>
    Copyright (C) <year>  <name of author>

    This program is free software; you can redistribute it and/or modify
    it under the terms of the GNU General Public License as published by
    the Free Software Foundation; either version 2 of the License, or
    (at your option) any later version.

    This program is distributed in the hope that it will be useful,
    but WITHOUT ANY WARRANTY; without even the implied warranty of
    MERCHANTABILITY or FITNESS FOR A PARTICULAR PURPOSE.  See the
    GNU General Public License for more details.

    You should have received a copy of the GNU General Public License along
    with this program; if not, see <https://www.gnu.org/licenses/>.

Also add information on how to contact you by electronic and paper mail.

If the program is interactive, make it output a short notice like this
when it starts in an interactive mode:

    Gnomovision version 69, Copyright (C) year name of author
    Gnomovision comes with ABSOLUTELY NO WARRANTY; for details type `show w'.
    This is free software, and you are welcome to redistribute it
    under certain conditions; type `show c' for details.

The hypothetical commands `show w' and `show c' should show the appropriate
parts of the General Public License.  Of course, the commands you use may
be called something other than `show w' and `show c'; they could even be
mouse-clicks or menu items--whatever suits your program.

You should also get your employer (if you work as a programmer) or your
school, if any, to sign a "copyright disclaimer" for the program, if
necessary.  Here is a sample; alter the names:

  Yoyodyne, Inc., hereby disclaims all copyright interest in the program
  `Gnomovision' (which makes passes at compilers) written by James Hacker.

  <signature of Ty Coon>, 1 April 1989
  Ty Coon, President of Vice

This General Public License does not permit incorporating your program into
proprietary programs.  If your program is a subroutine library, you may
consider it more useful to permit linking proprietary applications with the
library.  If this is what you want to do, use the GNU Library General
Public License instead of this License.

Aether docs

Note that due to the ongoing kernel rewrite, the docs are not up to date right now.

Aether Architecture

Overview

Aether is a microkernel-inspired operating system kernel targeting x86_64 and aarch64. The kernel itself is small: it manages physical and virtual memory, provides a preemptive scheduler, routes hardware interrupts to userspace tasks, and exposes a minimal syscall interface. Everything else — device drivers, program loading, IPC — lives in userspace.

Boot Flow

  1. The Limine bootloader loads the kernel ELF image and an initrd (tar archive) as a module. It also provides a framebuffer, memory map, RSDP, and command-line string to the kernel via the Limine protocol.
  2. kernel/entry — the Rust entry point (start.rs) is called by Limine. It sets up a two-phase allocator (early bump allocator → paged allocator) and calls kernel_core::main(...).
  3. kernel/core main() initializes subsystems in order:
    • Virtual address allocator
    • Physical memory manager
    • Virtual memory manager (HHDM-aware)
    • Early serial console
    • Platform-specific initialization (ACPI parsing, APIC/GIC setup)
    • Framebuffer terminal
    • Interrupt controller (APIC on x86_64, GIC on aarch64)
    • Paged memory allocator (replaces the early bump allocator)
    • Task registry and scheduler
    • Syscall dispatcher
    • Timer (preemption tick at 100 Hz)
  4. The kernel loads the init ELF binary from the initrd using its own ELF loader, creates a new address space, maps boot_info and the initrd into it, and spawns the init task.
  5. The kernel disables its early console and switches to the scheduler. From this point on the kernel runs only in response to interrupts and syscalls.

Component Map

┌────────────────────────────────────────────────────────────────────┐
│                            Workspace                               │
│                                                                    │
│  kernel/entry ──► kernel/core ──► kernel/hal ──► kernel/platform   │
│                        │                                           │
│                        └──► kernel/api  (syscall ABI + boot_info)  │
│                                                                    │
│  init  ──► libs/libsyscall (C stubs)                               │
│         ──► kernel/api/init/boot_info.h                            │
│                                                                    │
│  hello_world ──► libs/aether_rt (Rust runtime)                     │
│               ──► libs/aether_sys (Rust syscall wrappers)          │
│                                                                    │
│  image/ ──► initrd (tar) + ISO/raw disk image                      │
└────────────────────────────────────────────────────────────────────┘

kernel/entry

The bootloader-facing crate (#![no_std], #![no_main]). Registers Limine requests (framebuffer, memory map, HHDM offset, modules, RSDP, command line) and implements the two-phase allocator hand-off:

  • Early allocator — a simple bump allocator backed by a static buffer, used before page-frame allocation is available.
  • Proxy allocator — a #[global_allocator] wrapper that delegates to whichever allocator is currently active and supports an atomic hand-off.

kernel/core

The heart of the kernel. Key modules:

ModuleResponsibility
schedulerRound-robin preemptive scheduler; supports null/idle task, IRQ wake-up, task exit
syscall_handlerDispatches syscalls to per-syscall handlers
task_registryStores and manages all tasks (kernel threads and userspace processes)
allocatorPaged kernel heap (allocation only; deallocation not yet implemented)
elfLoads ELF64 binaries into a new address space
init_processReads the initrd, finds the init binary, and spawns the init process
tarball_parsingMinimal ustar tarball parser used to locate files in the initrd
tickerConfigures the hardware timer to fire at a fixed Hz and drive the scheduler
platform/*Trait-based abstraction layer consumed by kernel_core

The platform/ sub-module contains Rust traits (e.g. EarlyConsole, PhysicalMemoryManager, Interrupts, Syscalls, VirtualMemoryManager) whose implementations live in kernel/hal.

kernel/hal

The Hardware Abstraction Layer. Contains two sub-modules — arch/x86_64 and arch/aarch64 — selected at compile time via cfg(target_arch = …).

Each implementation covers:

  • CPU — core initialization, per-CPU data structures, halting
  • Early console — UART-based serial output used before the terminal is up
  • Emergency console — fallback output used during kernel panics
  • Interrupts — IDT/GDT (x86_64), exception vectors (aarch64), IRQ masking
  • MMU — page table management, address space switching
  • Syscallssyscall/sysret (x86_64), svc (aarch64) entry/exit
  • Tasks — task frame layout, context switching
  • Timer — LAPIC timer (x86_64), ARM Generic Timer (aarch64)

kernel/platform

Low-level C and assembly code that kernel_hal calls into. Includes:

  • ACPI table parsing and APIC/IOAPIC/GIC discovery
  • Page table structures for both architectures
  • Physical memory map processing
  • String and memory utilities

kernel/api

Shared ABI between the kernel and userspace.

  • init/boot_info.h — the boot_info struct passed by the kernel to the init process at startup (initrd pointer and size).
  • syscalls/syscalls.toml — machine-readable syscall definitions (number, return type, argument names and types).
  • syscalls/errors.toml — error codes returned by syscalls.
  • syscalls/syscall_parser.py — shared Python library that parses the TOML definitions and drives code generators.
  • syscalls/syscall_kernel_gen.py (in kernel/syscalls/) — generates the kernel-side dispatch glue.
  • libs/libsyscall/syscall_c_stubs.py — generates syscalls.h for C userspace.
  • libs/aether_sys/syscall_aether_sys_gen.py — generates the Rust aether_sys crate.

init

The first userspace process (written in C). Its responsibilities are:

  1. Initialize the serial port and use it as stdout.
  2. Parse the initrd tarball and locate the bin/hello_world ELF binary.
  3. Load the ELF binary into a new address space via proc_create / proc_mmap / proc_mprot syscalls.
  4. Spawn the loaded process using proc_spawn.
  5. Optionally handle keyboard input (driver in src/drivers/keyboard/).

hello_world

A minimal Rust binary that prints a message via the write syscall and loops. It exercises the aether_rt runtime and aether_sys syscall library.

libs/aether_rt

The Rust userspace runtime crate. Provides:

  • _start entry point — sets up the heap and calls main
  • A heap allocator (linked-list allocator over a static buffer)
  • Panic handler
  • libc-style shims (memcpy, memmove, memset, memcmp, __errno_location)
  • __rust_probestack / _Unwind_Resume stubs required by the Rust compiler

libs/aether_sys

Auto-generated Rust bindings for every syscall and constant defined in syscalls.toml / errors.toml. Regenerated by cargo build—see build.rs.

libs/libsyscall

Auto-generated C header (syscalls.h) and stubs for every syscall. Used by the init program.

Memory Layout

Kernel

The kernel is loaded by Limine into the higher half of the virtual address space. Limine provides a Higher-Half Direct Map (HHDM) offset so the kernel can access all physical memory through a fixed virtual offset.

Userspace

The lower half of the virtual address space is available to userspace processes. Kernel-provided data (boot info, initrd, bootstrap stack) is mapped at the high end of the lower half, growing downward. See kernel/init_contract.md for the exact layout convention used for the init process.

Syscall ABI

All syscalls follow a consistent convention:

  • Each syscall has a unique number (0–255).
  • The kernel returns an error code in one register and an optional return value in another.
  • Using the return value when the error code is non-zero is undefined behavior.
  • Up to 5 arguments are supported.

See syscalls.md for the full reference.

Building Aether

Host Prerequisites

  • GNU Make — Used for orchestrating builds. Doesn’t have to be in PATH.
  • QEMU — Required for running: qemu-system-x86_64 (x86_64) and/or qemu-system-aarch64 (aarch64).
  • Python 3 — Required by code-generation scripts.
  • Clang / LLVM — Used for compiling C/ASM code, for linking, and binutils.
  • Rust — Used for compiling Rust code.

Architecture Configuration

All build and run commands accept an ARCH variable that selects the target architecture:

  • ARCH=x86_64 — 64-bit x86
  • ARCH=aarch64 — 64-bit ARM

Building

Build everything for x86_64:

make all ARCH=x86_64

Build everything for aarch64:

make all ARCH=aarch64

You can also build individual targets, for example:

# Build only the kernel ELF
make pkg/kernel/install ARCH=x86_64

# Build the init binary
make pkg/init/install ARCH=x86_64

# Build the hello_world binary
make pkg/hello_world/install ARCH=x86_64

# Build the bootable ISO image
make dist/aether-x86_64.iso ARCH=x86_64
# or
make dist/aether-aarch64.img ARCH=aarch64

Development Guide

IDE Support

Aether supports two language servers:

  • Rust — Two Cargo workspaces (/ and /kernel/) consumed by rust-analyzer.
  • C/C++compile_commands.json consumed by clangd.

VS Code

The repository includes .vscode/ configuration:

  • Recommended extensions — see .vscode/extensions.json. Install them via Extensions → … → Show Recommended Extensions. The key extensions are rust-lang.rust-analyzer (Rust) and llvm-vs-code-extensions.vscode-clangd (C/C++).
  • Settings (.vscode/settings.json) — pre-configures rust-analyzer to load Cargo workspaces, and clangd with the correct compile-commands directory.
  • Tasks — the default build task (Terminal → Run Build Task, Ctrl+Shift+B) runs the Sync Project (C + Rust) command and prompts you to pick an architecture.

Zed

The repository includes .zed/ configuration:

  • Tasks (.zed/tasks.json) — preconfigured tasks:
    • Sync Project for x86_64 (C + Rust) — regenerates project files
    • Qemu run x86_64 / Qemu run aarch64 — run in QEMU
    • Qemu debug aarch64 — start QEMU with debugger stub
  • Debug (.zed/debug.json) — a Qemu debug configuration using CodeLLDB to attach to 127.0.0.1:1234.

Hardware Interrupts

IRQ

Interrupt Requests (IRQs) are used to identify hardware events. They are represented as uint8_t values. Userspace programs (such as a driver) can use syscalls to wait for an IRQ.

Before waiting for an IRQ, a userspace task must unmask it by calling irq_unmask(irq). Afterwards, irq_wait(irq) blocks the task until the hardware raises the interrupt. See syscalls.md for the full syscall reference.

x86_64

Since APIC is exclusively in use (meaning no legacy PIC), there have been set some conventions for IRQ numbers. For each IOAPIC pin, there is a corresponding IRQ number. The pin’s number is always identical to the IRQ number.

The interrupt vectors in range 0x00-0x1F are used for CPU exceptions. Due to legacy PIC and its conventions, 0x20-0x2F are not used, as it could lead to some undesired behavior. Therefore, 0x30 has been chosen as the vector offset for the IRQs. That means for every IRQ, its interrupt vector is 0x30 + the IRQ number.

aarch64

On aarch64 (that is, 64-bit ARM architecture), IRQs are used to represent shared peripheral interrupts (SPIs). It has been decided that the IRQ numbers correspond to the SPI numbers. Because interrupt ids (INTIDs) of SPIs start from 0x20, as the range 0x00-0x0F is used for software generated interrupts (SGIs), and 0x10-0x1F is used for private peripheral interrupts (PPIs), the IRQ number of an SPI is 0x20 + INTID.

Syscall ABI Reference

Overview

Syscalls are the interface between userspace processes and the kernel. The Aether syscall ABI is defined in TOML files under kernel/api/syscalls/:

  • syscalls.toml — the current list of all defined syscalls.
  • errors.toml — all possible syscall error codes.

These files are used to automatically generate C headers and Rust bindings for userspace consumers.

The list of syscalls and their functionality will change as the project evolves. Refer to syscalls.toml and errors.toml for the current authoritative definition rather than any derived documentation.

Calling Convention

Syscall number and arguments

The syscall number and arguments are passed in CPU registers:

x86_64

  • Syscall number: rax
  • Arguments (in order): rdi, rsi, rdx, r10, r8

aarch64

  • Syscall number: x8
  • Arguments (in order): x0, x1, x2, x3, x4

Up to 5 arguments per syscall are currently supported. This limit may be increased in the future.

Return values

Every syscall returns two values:

  • An error code — always present.
  • A return value — present only if the syscall produces one.

An error code of 0 means success. Using the return value when the error code is non-zero is undefined behavior.

All possible error codes for the current syscall set are defined in kernel/api/syscalls/errors.toml.

Adding a New Syscall

  1. Define the syscall in kernel/api/syscalls/syscalls.toml:

    [syscalls.my_syscall]
    number = 0x0C          # unique number 0–255
    return_type = "uint64" # one of the valid types
    args = [
        { name = "foo", type = "uint32" },
    ]
    
  2. Implement the handler in kernel/core/src/syscall_handler/. Create a new file sys_my_syscall.rs and implement the logic. Register the handler in kernel/core/src/syscall_handler/mod.rs.

  3. Add error codes (if needed) in kernel/api/syscalls/errors.toml.

Aether Kernel and the Init Program Contract

Data provided by the kernel and initial stack

Upon process entry, the kernel provides the following resources mapped into the init address space:

  • The boot_info structure.
  • The initrd (tarball archive).
  • A single page-sized (4096 bytes) bootstrap stack.

Memory Layout Convention

All kernel-provided data shall reside at the high end of the userspace virtual address range (the “Top of the Lower Half”).

  1. Growth Direction: Data blocks are placed contiguously, starting from the highest available canonical address (e.g., 0x00007FFFFFFFF000) and growing downwards.
  2. Bootstrap Stack: The one-page initial stack is placed as the final (lowest address) item in this kernel-provided data block.
  3. Allocation Boundary: The lowest address of the bootstrap stack serves as the Stack Limit. The init program can rely on this address as the hard upper bound for its own self-allocated “real” stack.

Execution State at Entry

  • RSP/SP: Shall point to the top of the bootstrap stack.
  • RDI/X0: Shall contain the virtual address of the boot_info structure.
  • Alignment: The initial stack pointer must be 16-byte aligned to satisfy the SysV ABI before the first function call.