<?xml version="1.0" encoding="UTF-8"?>
<chapter id="architecture"><?dbhtml filename="arch.html"?>
<title>Architecture overview</title>
<section>
<title>Scheme</title>
<para>
<mediaobject id="arch1">
<imageobject role="html">
<imagedata format="PNG" fileref="images/arch1.png"/>
</imageobject>
<imageobject role="fop">
<imagedata format="SVG" fileref="images.vector/arch1.svg" />
</imageobject>
</mediaobject>
</para>
</section>
<section>
<title>Kernel primitives</title>
<para>
<termdef><glossterm>Thread</glossterm> is the basic execution primitive.</termdef>
</para>
<para>
<termdef><glossterm>Thread context</glossterm> represents state of the <emphasis>thread</emphasis>.
Thread context is built of the context registers contents, FPU state and the stack.</termdef>
</para>
<para>
<termdef>
<glossterm>Task</glossterm> is a multi-purpose entity, serving to
<itemizedlist>
<listitem>incorporate set if its threads</listitem>
<listitem>provide common address space to its threads</listitem>
<listitem>be an end-point in IPC</listitem>
</itemizedlist>
</termdef>
</para>
<para>
<termdef>
<glossterm>Address space area</glossterm> is a mutually disjunctive range of memory with the code, stack and data.
</termdef>
</para>
<para>
<termdef>
<glossterm>Address space</glossterm> is a aggregating entity for address space areas, connecting them to the task.
</termdef>
</para>
</section>
<section>
<title>Monolithic microkernel</title>
<para>
Though HelenOS was initially planned as a microkernel, we were trying to avoid several issues, connected
with microkernels, such as much higher overhead during memory management and hardware operations. For this reason
some of the subsystems, that are to be implemented as servers in classic microkernel design, were implemented
as a part of kernel, thus minimizing this overhead.
</para>
<formalpara>
<title>Memory management</title>
<para>
Unlike the classic microkernel, HelenOS has all its memory management functionality in the kernel, available to the memory
management server via the set of syscalls.
</para>
</formalpara>
<formalpara>
<title>Kernel device drivers</title>
<para>
HelenOS kernel has some of the very basic device drivers
<itemizedlist>
<listitem>ACPI</listitem>
<listitem>APIC</listitem>
<listitem>SMP configuration</listitem>
<listitem>System clock</listitem>
<listitem>Interrupt controllers</listitem>
<listitem>Console</listitem>
<listitem>VESA & frame buffer</listitem>
</itemizedlist>
</para>
</formalpara>
</section>
<section>
<title>IPC</title>
<para>HelenOS IPC is designed in analogy with telephone communication.
Each task has an <emphasis>answerbox</emphasis> and a set of <emphasis>phones</emphasis> to call another tasks' answerboxes.
</para>
<para>Communication
is possible after the connection is established, and can be either <emphasis>asynchronious</emphasis> or <emphasis>synchronious</emphasis>.
</para>
</section>
<section>
<title>Functionality model</title>
<para>
As you know, microkernel design is very simple, just enough to provide communication facility for tasks. Most of the OS functionality
is performed by server tasks, that are running in userspace.
Thus most of the system calls in monolithic kernels, are the IPC calls on server tasks in microkernels.
</para>
<para>
Moreover, problems experience the device drivers. Running in the user space, device driver still needs to recieve interrupts
and access hardware directly.
</para>
<para>
This raises two major problems in microkernels:
<orderedlist numeration="loweralpha">
<listitem>What is the recipient address of the server (e.g. "memory manager" or a specific device driver) ?</listitem>
<listitem>How this server task is going to access hardware or kernel while running in the user mode?</listitem>
</orderedlist>
</para>
<formalpara id="intro_ns">
<title>Name server</title>
<para>As every microkernel, HelenOS has a "Name server" task with "well known" IPC address, that connects user task to any server
just by the string service indentification.</para>
</formalpara>
<formalpara id="intro_ddi">
<title>Device driver interface</title>
<para>Device drivers use special syscalls to map physical memory areas into their address space, to map port regions (mostly ia32).
Interrupts are delivered to the device driver task by the standard IPC means.
</para>
</formalpara>
</section>
</chapter>