Linux Foundation Wiki

project collaboration site

User Tools

Site Tools


Build the CIP System and Artifacts (4/5)

This set of instructions assumes you have already Setup and Configured the B@D Virtual Machine. If that is not the case please go back to the Beaglebone Black Setup & Configuration.

Creating an initramfs with BusyBox for the Beaglebone Black

BusyBox combines tiny versions of many common UNIX utilities into a single small executable. It provides replacements for most of the utilities you usually find in GNU fileutils, shellutils, etc. BusyBox provides a fairly complete environment from which to boot any embedded system.


You must have the CIP Board-at-Desk Single-Developer (B@D) v0.9 Virtual Machine set up with make, gcc and the arm cross compilers.

1. Connect to the B@D Virtual Machine.

user@host:~/location-of-B@D-Vagrant-files$ vagrant ssh

2. Install the library for curses which is used by the make menuconfig command

vagrant@guest:~$ sudo apt-get install libncurses5-dev

Create a directory to work in

3. Make the necessary directories in which to work and where the output will be stored so it is accessible from the website.

vagrant@guest:~$ mkdir -p work/initramfs

vagrant@guest:~$ cd work

vagrant@guest:~/work$ sudo mkdir /var/www/images/kernel-ci/initramfs

4. Download the busybox source code and unzip it.

vagrant@guest:~/work$ wget

vagrant@guest:~/work$ tar -xvf busybox-1.26.2.tar.bz2

vagrant@guest:~/work$ cd busybox-1.26.2

5. Create the default configuration

vagrant@guest:~/work/busybox-1.26.2$ make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- defconfig

6. Configure and make the initramfs

vagrant@guest:~/work/busybox-1.26.2$ make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- menuconfig

7. Select the following settings: Busybox Settings → Build Options → Build Busybox as a static binary (no shared libs)

8. Press Y or the space bar to select that option.

9. Press the Tab key to move to the buttons at the bottom of the window and select the < Exit > button twice.

10. On the dialog box that asks “Do you wish to save your new configuration?” Select Yes and press Enter.

11. Execute the following commands for building the initramfs

vagrant@guest:~/work/busybox-1.26.2$ make ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- CONFIG_PREFIX=/home/vagrant/work/initramfs /home/vagrant/work/initramfs install

vagrant@guest:~/work/busybox-1.26.2$ cd ..

vagrant@guest:~/work$ cd initramfs


After the successful completion of the above commands, you can see 3 directories (bin, sbin, usr) and one file, linuxrc created in your initramfs directory.

The /dev Directory:

12. Create the /dev directory and create some special files under it.

vagrant@guest:~/work/initramfs$ mkdir dev

vagrant@guest:~/work/initramfs$ sudo mknod dev/console c 5 1

vagrant@guest:~/work/initramfs$ sudo mknod dev/null c 1 3

vagrant@guest:~/work/initramfs$ sudo mknod dev/zero c 1 5


- /dev/console, is the place which the kernel uses for interaction.

- /dev/null behaves like an empty file and is used to suppress the output from commands/programs.

- /dev/zero contains sequence of zeros used to fill up the memory regions.

The /lib and /usr/lib Directories:

13. For the static libraries, copy from the ARM cross compiler toolchain path.

vagrant@guest:~/work/initramfs$ mkdir lib usr/lib

vagrant@guest:~/work/initramfs$ rsync -a /usr/arm-linux-gnueabihf/lib/ ./lib/

vagrant@guest:~/work/initramfs$ rsync -a /usr/arm-linux-gnueabihf/lib/ ./usr/lib/

The /proc, /sys, and /root Directories:

14. Create directories for mounting the virtual filesystems procfs, sysfs and root directory.

vagrant@guest:~/work/initramfs$ mkdir proc sys root

The /etc Directory:

15. Create /etc and then, create additional files inside this directory.

vagrant@guest:~/work/initramfs$ mkdir etc

vagrant@guest:~/work/initramfs$ nano etc/inittab

null::sysinit:/bin/mount -a
null::sysinit:/bin/hostname -F /etc/hostname
null::respawn:/bin/cttyhack /bin/login root

16. Create another file called fstab and populate it. This file will mount the virtual file systems.

vagrant@guest:~/work/initramfs$ nano etc/fstab

proc  /proc proc  defaults  0 0
sysfs /sys  sysfs defaults  0 0

17. Create the hostname file.

vagrant@guest:~/work/initramfs$ nano etc/hostname


18. Create the passwd file.

vagrant@guest:~/work/initramfs$ nano etc/passwd


19. Create an init file. vagrant@guest:~/work/initramfs$ touch init

vagrant@guest:~/work/initramfs$ chmod +x init

Edit init giving it the following contents:

20. Create the initramfs.cpio.gz

vagrant@guest:~/work/initramfs$ find . -depth -print | cpio -ocvB | gzip -c > ../initramfs.cpio.gz

Test it on the Beaglebone Black:

21. Copy the initramfs.cpio.gz to the initramfs directory where it can be retrieved by LAVA.

vagrant@guest:~/work/initramfs$ sudo cp ../initramfs.cpio.gz /var/www/images/kernel-ci/initramfs

22. In the LAVA web-based frontend, click on SchedulerSubmit Job

23. Copy the Beaglebone Black Health Check Job and paste it into the Job Definition box.

24. Change the link for the initramfs.cpio.gz file to the VM's IP Address and the path to the file:

 url: http://localhost:8010/initramfs/initramfs.cpio.gz
 compression: gz

Building the CIP Kernel with Kernel CI


  • A working CIP Board-at-Desk Single-Developer (B@D) v0.9 Virtual Machine


1. Connect to the B@D Virtual Machine.

user@host:~/location-of-B@D-Vagrant-files$ vagrant ssh


2. Change to the git-repos directory

vagrant@guest:~$ cd git-repos

Note that in the case of the pre-provisioned Vagrant instance it already has a clone of the linux-cip repository, you only need a git pull from within the linux-cip directory to ensure that it is up to date and proceed to step 4.

3. Clone the CIP Linux Kernel

vagrant@guest:~/git-repos$ git clone git://

4. Find the branch of the kernel version you want (We will use '4.4.69-cip4' for this example)

vagrant@guest:~/git-repos$ cd linux-cip

vagrant@guest:~/git-repos/linux-cip$ git tag -l | grep cip


5. Create a new branch using the latest CIP tag.

vagrant@guest:~/git-repos/linux-cip$ git checkout -b cip_v4.4.69 v4.4.69-cip4

6. Set the environment variables. You can create a tree name that describes your project. Select the Architecture of the target device (i.e. arm, arm64, mips, i386, amd64, etc.). Choose the cross compiler you need for that architecture.

vagrant@guest:~/git-repos/linux-cip$ export TREE_NAME=cip-example

vagrant@guest:~/git-repos/linux-cip$ export ARCH=arm

vagrant@guest:~/git-repos/linux-cip$ export CROSS_COMPILE=arm-linux-gnueabihf-

Note: Don't forget the dash (-) at the end of the CROSS_COMPILE line!

7. Execute the command, passing in the Beaglebone Black configuration.

vagrant@guest:~/git-repos/linux-cip$ ~/kernelci-build/ -c omap2plus_defconfig -p CIP-KernelCI

 make -j4 -k -s ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- O=build-arm omap2plus_defconfig
 make -j4 -k -s ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- O=build-arm
 make -j4 -k -s ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- O=build-arm modules
 make -j4 -k -s ARCH=arm CROSS_COMPILE=arm-linux-gnueabihf- O=build-arm modules_install
 INFO: published artifacts
 INFO: triggering build

8. Start the Kernel CI web server

vagrant@guest:/vagrant/scripts/ &

9. On your host machine, open a web browser and enter the following in the address box:

http://localhost:5000 Note that if you are using Windows 10 as a host you will need to use either Chrome or Firefox as a browser as the Windows browsers do not work with the KernelCI webserver.

10. You will see the KernelCI Website home page from which, you can navigate to the different builds and Trees that you've created (see the Example Builds section below).

11. Click on the Jobs button at the top of the page and you will see all of the available Trees

12. Click on the Tree name (cip-example) and you will see the list of available builds, or kernel versions for that tree.

13. Click on the v4.4.69 kernel and you will see the list of build configurations (omap2plus).

14. The files that resulted from the build are available in the KernelCI website by navigating to the Tree Name, Kernel version, and Configuration. They are stored on the hard drive at:


Example Builds

15. For instance, for the build with the following parameters:

  • TREE_NAME: cip-example
  • KERNEL_VERSION: v4.4.69-cip4
  • BRANCH: cip_v4.4.69 (This came from the name we used in the git checkout step.)
  • ARCH: arm
  • CONFIG: omap2plus

The files will be located at:


16. Kernel CI generates the following files. Make a note of the ones in bold. They will be needed for testing the CIP Kernel:

  • build.log shows the output of the build process to help track down any issues in compilation.
  • kernel.config holds the configuration settings used to build the kernel.
  • is the kernel's symbol table that is used to debug kernel runtime errors.
  • zImage is the compressed kernel image to be installed on target device.
  • modules is a directory that holds all of the kernel modules needed for the CIP Kernel.
  • dtbs directory which holds all of the generated Device Tree Binary's.

17. Check the reference test cases and try them as reference examples.

civilinfrastructureplatform/cipsystembuildhowto.txt · Last modified: 2017/07/19 09:03 by rajmarshall