I've been less than happy with other kernel books I've read. Admittedly,it's a difficult subject: there's a lot to cover, and you really need quite a bit of programming and general Unix knowledge before you could even consider jumping into this area. But I have the background, have even written simple Unix drivers, and yet every other kernel programming book has disappointed me.
It's all so overwhelming: there are conventions and quirks that have developed over time and surely are second nature to the people who have been doing Linux kernels for years, but these things are baffling to the newcomer.
This book tries to get you past that. The authors specifically say that they have tried to cover the things that confused them when they first started looking at the kernel. I'm sure their efforts aren't perfect, but the effort does definitely show.
The authors present several programming projects to help explore the kernel concepts, and every chapter has review questions to help firm up your understanding. The approach is from user space when possible: the assumption is that you are comfortable with application programming and that is used as the base to lead you down into the work done by the kernel for your programs. There's plenty of annotated source code here, both for x86 and PowerPC architectures. The inclusion of PowerPC information was an unexpected bonus; other books I've read have usually ignored that entirely or glossed it over quickly.
Of course you need a background in C, and while this does try to cover general kernel subjects, it wouldn't hurt to have at least some prior reading there. A little familiarity with hardware and light assembly language will help also, although the authors do give some coverage there.
I'm looking forward to spending more time exploring this book.
Tony Lawrence 2005/10/04 Rating:
Order (or just read more about) The Linux Kernel Primer from Amazon.com
Got something to add? Send me email.
More Articles by Tony Lawrence © 2011-04-29 Tony Lawrence
The danger of computers becoming like humans is not as great as the danger of humans becoming like computers. (Konrad Zuse)