Building Tiny Linux Systems with Busybox, Part 2: Building the Kernel
For this example I use Linux kernel version 2.2.17. The 2.4.0-test8 kernel that I tried did not size the RAM disk for the root file system properly, leading to a ``not enough memory'' message at boot time. That bug will probably be repaired in the 2.4 series of kernels by the time you read this.
We will build our example to run on an i386-architecture PC-compatible system with PC keyboard and VGA display, booting from a floppy disk and running the root file system entirely in RAM once the system is booted. This example should also boot from IDE disks and from FLASH EEPROM devices that masquerade as IDE disks. It can also be configured to boot from a CD-ROM.
Build a bzImage-style kernel with all of the facilities needed for the application, plus these three:
RAM disk support (in the Block Devices menu)
Initial RAM disk (initrd) support (also in the Block Devices menu)
ROM file system support (in the File Systems menu)
Don't use kernel modules, because this example system doesn't support them. Don't put any facilities in the kernel that you don't need, as they will use up space that you need on the floppy disk. A kernel with the facilities you need should be around half a megabyte in size and should fit easily on a floppy along with the ROM root file system. A kernel with many unnecessary bells and whistles will be a megabyte or more and won't leave sufficient room for your ROM root file system.
If you're not familiar with building and installing kernels on a normal Linux PC, you'll need to study up on that. In short, I placed the kernel sources in /usr/src/linux and ran:
xhost +localhost su make xconfig make dep make bzImage
This created a compiled Linux kernel in /usr/src/linux/arch/i386/boot/bzImage.
In the busybox source directory, edit the Makefile, changing the variable DOSTATIC from false to true. Then run make. That will create a static-linked version of busybox. Confirm that it is static-linked by running this command:
This should print something like:
statically linked (ELF)It's important to get this right; if you install a dynamic-linked version of Busybox, your system won't run because we aren't installing the runtime dynamic linker and its libraries on the floppy disk for this example.
We're going to go through all of the steps for creating a minimal root file system by hand so that you will understand just how little is necessary to boot your system rather than copying all of the files from the root of your Linux distribution and then being afraid to remove anything because you don't know whether it's necessary. You will need to become root (the superuser) to perform the following steps because the mknod command requires superuser privilege.
Create the tiny-linux directory and change directory into it:
mkdir tiny-linux cd tiny-linux
Create the standard directories in it:
mkdir dev etc etc/init.d bin proc mnt tmp var var/shm chmod 755 . dev etc etc/init.d bin proc mnt tmp var var/shmEnter the tiny-linux/dev directory:
cd devCreate the generic terminal devices:
mknod tty c 5 0 mknod console c 5 1 chmod 666 tty console #Allow anyone to open and write terminals.Create the virtual terminal device for the VGA display:
mknod tty0 c 4 0 chmod 666 tty0Create the RAM disk device:
mknod ram0 b 1 0 chmod 600 ram0Create the null device, used to discard unwanted output:
mknod null c 1 3 chmod 666 nullChange directory to tiny-linux/etc/init.d, where startup scripts are stored:
cd ../etc/init.dUse an editor to create this shell script in tiny-linux/etc/init.d/rcS. It will be executed when the system boots:
#! /bin/sh mount -a # Mount the default file systems mentioned in /etc/fstab.Make the script executable:
chmod 744 rcSChange directory to tiny-linux/etc:
cd ..Use an editor to create the file tiny-linux/etc/fstab, which says what file systems should be mounted at boot time:
proc /proc proc defaults 0 0 none /var/shm shm defaults 0 0Set the mode of tiny-linx/etc/fstab:
chmod 644 fstabUse an editor to create the file tiny-linux/etc/inittab, which tells /bin/init, the system startup program, what processes to start:
::sysinit:/etc/init.d/rcS ::askfirst:/bin/shThe above example runs the script /etc/init.d/rcS at boot time and runs an interactive shell on the console device.
Set the modes of tiny-linux/etc/inittab:
chmod 644 inittab
That's everything necessary to create your root file system, except for the installation of the programs. Change directory to tiny-linux/bin:
cd ../binCopy your static-linked version of Busybox from wherever you built it into tiny-linux/bin/busybox with a command similar to this one:
cp ~/busybox-0.46/busybox busyboxAdd another command name ls to Busybox using the ln command:
ln busybox lsRun ls, and the result should look like this:
-rwxr-xr-x 2 root root 580424 Sep 12 15:17 busybox -rwxr-xr-x 2 root root 580424 Sep 12 15:17 lsRepeat the above ln command for all of these names:
[, ar, basename, cat, chgrp, chmod, chown, chroot, chvt, clear, cp, cut, date, dc, dd, deallocvtdf, dirname, dmesg, du, dumpkmap dutmp, echo, false, fbset, fdflush, find, free, freeramdisk, fsck.minix, grep, gunzip, gzip, halt, head, hostid, hostname, id, init, insmod, kill, killall, length, linuxrc, ln, loadacm, loadfont, loadkmap, logger, logname, lsmod, makedevs, md5sum, mkdir, mkfifo, mkfs.minix, mknod, mkswap, mktemp, more, mount, mt, mv, nc, nslookup, ping, poweroff, printf, ps, pwd, reboot, rm, rmdir, rmmod, sed, setkeycodes, sh, sleep, sort, swapoff, swapon, syn, c syslogd, tail, tar, tee, telnet, test, touch, tri, true, tty, umount, uname, uniq, update, uptime, usleep, uudecode, uuencode, wc, which, whoami, yes, zcatAre you tired yet? Well, now is a good time to take a break--you've finished creating your ROM root file system.
|Designing Electronics with Linux||May 22, 2013|
|Dynamic DNS—an Object Lesson in Problem Solving||May 21, 2013|
|Using Salt Stack and Vagrant for Drupal Development||May 20, 2013|
|Making Linux and Android Get Along (It's Not as Hard as It Sounds)||May 16, 2013|
|Drupal Is a Framework: Why Everyone Needs to Understand This||May 15, 2013|
|Home, My Backup Data Center||May 13, 2013|
- Nice article, thanks for the
38 min 34 sec ago
- I once had a better way I
6 hours 24 min ago
- Not only you I too assumed
6 hours 41 min ago
- another very interesting
8 hours 34 min ago
- Reply to comment | Linux Journal
10 hours 28 min ago
- Reply to comment | Linux Journal
17 hours 22 min ago
- Reply to comment | Linux Journal
17 hours 38 min ago
- Favorite (and easily brute-forced) pw's
19 hours 29 min ago
- Have you tried Boxen? It's a
1 day 1 hour ago
- seo services in india
1 day 5 hours ago
Enter to Win an Adafruit Pi Cobbler Breakout Kit for Raspberry Pi
It's Raspberry Pi month at Linux Journal. Each week in May, Adafruit will be giving away a Pi-related prize to a lucky, randomly drawn LJ reader. Winners will be announced weekly.
Fill out the fields below to enter to win this week's prize-- a Pi Cobbler Breakout Kit for Raspberry Pi.
Congratulations to our winners so far:
- 5-8-13, Pi Starter Pack: Jack Davis
- 5-15-13, Pi Model B 512MB RAM: Patrick Dunn
- 5-21-13, Prototyping Pi Plate Kit: Philip Kirby
- Next winner announced on 5-27-13!
Free Webinar: Hadoop
How to Build an Optimal Hadoop Cluster to Store and Maintain Unlimited Amounts of Data Using Microservers
Realizing the promise of Apache® Hadoop® requires the effective deployment of compute, memory, storage and networking to achieve optimal results. With its flexibility and multitude of options, it is easy to over or under provision the server infrastructure, resulting in poor performance and high TCO. Join us for an in depth, technical discussion with industry experts from leading Hadoop and server companies who will provide insights into the key considerations for designing and deploying an optimal Hadoop cluster.
Some of key questions to be discussed are:
- What is the “typical” Hadoop cluster and what should be installed on the different machine types?
- Why should you consider the typical workload patterns when making your hardware decisions?
- Are all microservers created equal for Hadoop deployments?
- How do I plan for expansion if I require more compute, memory, storage or networking?