next up previous contents
Next: Contents

The Linux Kernel Hackers' Guide

Michael K. Johnson

Copyright © 1992, 1993, 1994, 1995
Michael K. Johnson
201 Howell Street, Apt. 1C,
Chapel Hill, North Carolina
27514-4818
johnsonm@nigel.vnet.net

The Linux Kernel Hackers' Guide may be reproduced and distributed in whole or in part, subject to the following conditions:

  1. The copyright notice above and this permission notice must be preserved complete on all complete or partial copies.
  2. Any translation or derivative work of The Linux Kernel Hackers' Guide must be approved by the author in writing before distribution.
  3. If you distribute The Linux Kernel Hackers' Guide in part, instructions for obtaining the complete version of The Linux Kernel Hackers' Guide must be included, and a means for obtaining a complete version provided.
  4. Small portions may be reproduced as illustrations for reviews or quotes in other works without this permission notice, if proper citation is given.
  5. The GNU General Public License referenced below may be reproduced under the conditions given within it.
  6. Several sections of this document are held under separate copyright. When these sections are covered by a different copyright, the seperate copyright is noted. If you distribute The Linux Kernel Hackers' Guide in part, and that part is, in whole, held under a seperate copyright, the conditions of that copyright apply.
Exceptions to these rules may be granted for academic purposes: Write to Michael K. Johnson, at the above address, or email johnsonm@nigel.vnet.net, and ask. These restrictions are here to protect the authors, not to restrict you as educators and learners. All source code in The Linux Kernel Hackers' Guide is placed under the GNU General Public License. See Appendix #gnulicense#24> for a copy of the GNU ``GPL.'' Source code for all full example programs is available on-line as tsx-11.mit.edu:/pub/linux/docs/hacker-source.tar.Z and a copy of the GPL is available in that file as COPYING. [O.K., so it will be available when there is some source to distribute...]

UNIX is a trademark of X/Open
MS-DOS is a trademark of Microsoft Corporation.
Linux is not a trademark, and has no connection to UNIX or X/Open.

If any trademarks have been unintentionally unacknowledged, please inform the editor, Michael K. Johnson, 201 Howell Street, Apt. 1C, Chapel Hill, North Carolina 27514-4818, email johnsonm@nigel.vnet.net.

Introduction The The Linux Kernel Hackers' Guide is inspired by all of us ``kernel hacker wannabees'' who just did not know enough about unix systems to hack the Linux kernel when it first came out, and had to learn slowly. This guide is designed to help you get up to speed on the concepts that are not intuitively obvious, and to document the internal structures of Linux so that you don't have to read the whole kernel source to figure out what is happening with one variable, or to discover the purpose of one function call. Why Linux? Well, Linux is the first free unix clone for the 386 to be freely available. It is a complete re-write, and has been kept small, so it does not have a lot of the time-honored baggage that other free operating systems (like 386BSD) carry, and so is easier to understand and modify. Unix has been around for over twenty years, but only in the last few years have microcomputers become powerful enough to run a modern protected, multiuser, multitasking operating system. Furthermore, unix implementations have not been free. Because of this, very little free documentation has been written, at least for the kernel internals. Unix, though simple at first, has grown more and more appendages, and has become a very complex system, which only ``wizards'' understand. With , however, we have a chance to change this, for a few reasons:

It is our hope that this book will help the nascent kernel hacker learn how to hack the Linux kernel, by giving an understanding of how the kernel is structured.

Thanks to...

Linus Torvalds,
of course, for starting this whole time sink, and for gently providing explanations whenever necessary. He has done a wonderful job of keeping the kernel source code understandable and neat. I can't imagine having learned so much in the past few years without .
Krishna Balasubramanian and Douglas Johnson,
for writing much of the section on memory management, and helping with the rest.
Stanley Scalsky,
for helping document the system call interface.
Rik Faith,
for writing the section on how to write a SCSI device driver.
Robert Baruch,
for the review of Writing UNIX Device Drivers and for his help with the section on writing device drivers.
Linux Journal,
for providing me with a Linux-related job, and for allowing me to do work on the KHG on their time.
Kim Johnson, my wife,
for tolerating and encouraging me even when I spend my time on crazy stuff like Linux.
Copyright Acknowledgements:
Linux Memory Management:
The original version of this document is copyright © 1993 Krishna Balasubramanian. Some changes copyright © 1993 Michael K. Johnson and Douglas R. Johnson.
How System Calls Work:
The original version of this document is copyright © 1993 Stanley Scalsky. Some changes copyright © 1993 Michael K. Johnson
Writing a SCSI Device Driver
The original version of this document is copyright © 1993 Rickard E. Faith. Some modifications are copyright © 1993 Michael K. Johnson. The author has approved the inclusion of this material, despite the slightly more restrictive copyright on this whole document. The original copyright restrictions, which still apply to any work derived solely from this work, are:
Copyright © 1993 Rickard E. Faith (faith@cs.unc.edu). All rights reserved. Permission is granted to make and distribute verbatim copies of this paper provided the copyright notice and this permission notice are preserved on all copies.
If you wish to make a derived work, please start from the original document. To do so, please contact Rickard E. Faith, faith@cs.unc.edu. The original is available for anonymous ftp as ftp.cs.unc.edu:/pub/faith/papers/scsi.paper.tar.gz.