Linux Kernel - Introduction

Linux Torvalds developed the first version of Linux in 1991 as an operating system for computers powered by the intel microprocessor.  Started as a terminal emulator Linux evolved over a period.

What is an operating system ?

The operating system is considered as a part of the system responsible for basic use and administration. This includes the kernel and device drivers, boot loaders, command shell or other user interface and basic file and system utilities. The term system, in turn refers to the operating system  and all the application running on top of it.

The user interface is the outermost portion of the operating system, the kernel is the innermost. The kernel sometimes referred as the supervisor, core or internals of the operating system. Typical components if the kernel are interrupt handlers to service interrupt requests, a scheduler to share the processor time among multiple processes, a memory management system to manage process address spaces and system services such as networking and inter process communication.

On modern system with the protected memory management units the kernel typically resides in an elevated system state compared to  normal user applications. This includes a protected memory space and full access to the hardware. The system state and memory space is collectively referred as kernel space. Conversely user applications execute in a user space.

Applications running on the system communicate with the kernel via system calls. An application typically calls  he functions in a library for example, the C library that in turn depend in the system call interface to instruct the kernel to carry out certain tasks on the applications behalf. When an application executes a system call, we say that the kernel is executing on behalf of an application, furthermore the application is said to be executing a system call in a kernel space and the kernel is running in process-context. This relationship that applications call into the kernel via the system call interface, is the fundamental manner in which the applications get work done.

The kernel also manages the systems hardware. When the hardware wants to communicate with the system it will generate an interrupt, which will interrupt the processor and in turn it interrupts the kernel. Each interrupts will have a interrupt number and respective interrupt handler for it. We will discuss about interrupts in out further posts.

Monolithic and Microkernel designs

Monolithic Kernels are designed as entirely single process running in a single address space. They are implemented as single static binaries.

Microkernels are not implemented as a single process, instead the functionality of the kernel is broken down into separate processes, called as servers. Since they are running as separate processes, the communication within  kernel happens via inter process communication such as message passing.

Linux is a monolithic kernel, which means it is built as a single static binary, and runs in a single address space entirely in kernel mode. However Linux borrows many things from the microkernel, like capability to preempt itself, support for kernel threads and the capability to dynamically load kernel binaries ; called as kernel modules.