Kernel Configuration - Part I

Kernel Configuration - Part I

As promised, we’re starting another series on kernel configuration this week. Today, we’ll take a brief minute or two to take the 10,000-foot view of this code enigma.

The kernel in Linux is the core code of the operating system. In effect, it’s the mediator between your software and your hardware. It is, in short, what make the system run. It allocates memory to processes, and interprets and executes instructions from running programs. It makes it possible for the users, via software, to talk to the hardware, including hard drives, processors, RAM, and peripherals. In other words, no kernel, no Linux computer.

The kernel in Linux is, at its best, a monolithic kernel. Simply put, that means that all the drivers necessary to run harware and peripherals are compiled directly into the core code itself. For two reasons, recent Linux distributions have moved in the direction of modular driver loading - the ability to load and unload drivers for hardware on demand. This isn’t strictly withing the definition of a monolithic kernel, but it serves the purpose quite well. It also means less configuration and less firsthand knowledge of hardware for new Linux users.

That’s the high-level view of the Linux kernel. Tomorrow, we’ll dig into the details of configuring and compiling a kernel specifically for your Linux computing needs.