generic-poky/documentation/dev-manual/dev-manual-model.xml

1123 lines
70 KiB
XML
Raw Blame History

This file contains ambiguous Unicode characters

This file contains Unicode characters that might be confused with other characters. If you think that this is intentional, you can safely ignore this warning. Use the Escape button to reveal them.

<!DOCTYPE chapter PUBLIC "-//OASIS//DTD DocBook XML V4.2//EN"
"http://www.oasis-open.org/docbook/xml/4.2/docbookx.dtd"
[<!ENTITY % poky SYSTEM "../poky.ent"> %poky; ] >
<chapter id='dev-manual-model'>
<title>Common Development Models</title>
<para>
Many development models exist for which you can use the Yocto Project.
This chapter overviews the following methods:
<itemizedlist>
<listitem><para><emphasis>System Development:</emphasis>
System Development covers Board Support Package (BSP) development and kernel
modification or configuration.
If you want to examine specific examples of the system development models,
see the "<link linkend='dev-manual-bsp-appendix'>BSP Development Example</link>"
appendix and the
"<link linkend='dev-manual-kernel-appendix'>Kernel Modification Example</link>" appendix.
</para></listitem>
<listitem><para><emphasis>User Application Development:</emphasis>
User Application Development covers development of applications that you intend
to run on some target hardware.
For a user-space application development example that uses the
<trademark class='trade'>Eclipse</trademark> IDE,
see the
<ulink url='&YOCTO_DOCS_ADT_URL;'>
The Yocto Project Application Development Toolkit (ADT) User's Guide</ulink>.
</para></listitem>
<listitem><para><emphasis>Temporary Source Code Modification:</emphasis>
Direct modification of temporary source code is a convenient development model
to quickly iterate and develop towards a solution.
Once the solution has been implemented, you should of course take steps to
get the changes upstream and applied in the affected recipes.</para></listitem>
<listitem><para><emphasis>Image Development using Hob:</emphasis>
You can use the <ulink url='&YOCTO_HOME_URL;/projects/hob'>Hob</ulink> to build
custom operating system images within the build environment.
Hob provides an efficient interface to the OpenEmbedded build system.</para></listitem>
<listitem><para><emphasis>Using a Development Shell:</emphasis>
You can use a <filename>devshell</filename> to efficiently debug commands or simply
edit packages.
Working inside a development shell is a quick way to set up the OpenEmbedded build
environment to work on parts of a project.</para></listitem>
</itemizedlist>
</para>
<section id='system-development-model'>
<title>System Development Workflow</title>
<para>
System development involves modification or creation of an image that you want to run on
a specific hardware target.
Usually, when you want to create an image that runs on embedded hardware, the image does
not require the same number of features that a full-fledged Linux distribution provides.
Thus, you can create a much smaller image that is designed to use only the hardware
features for your particular hardware.
</para>
<para>
To help you understand how system development works in the Yocto Project, this section
covers two types of image development: BSP creation and kernel modification or
configuration.
</para>
<section id='developing-a-board-support-package-bsp'>
<title>Developing a Board Support Package (BSP)</title>
<para>
A BSP is a packageof recipes that, when applied, during a build results in
an image that you can run on a particular board.
Thus, the package, when compiled into the new image, supports the operation of the board.
</para>
<note>
For a brief list of terms used when describing the development process in the Yocto Project,
see the "<link linkend='yocto-project-terms'>Yocto Project Terms</link>" section.
</note>
<para>
The remainder of this section presents the basic steps used to create a BSP
based on an existing BSP that ships with the Yocto Project.
You can reference the "<link linkend='dev-manual-bsp-appendix'>BSP Development Example</link>"
appendix for a detailed example that uses the Crown Bay BSP as a base BSP from which to start.
</para>
<para>
The following illustration and list summarize the BSP creation general workflow.
</para>
<para>
<imagedata fileref="figures/bsp-dev-flow.png" width="6in" depth="7in" align="center" scalefit="1" />
</para>
<para>
<orderedlist>
<listitem><para><emphasis>Set up your host development system to support
development using the Yocto Project</emphasis>: See the
"<ulink url='&YOCTO_DOCS_QS_URL;#the-linux-distro'>The Linux Distributions</ulink>"
and the
"<ulink url='&YOCTO_DOCS_QS_URL;#packages'>The Packages</ulink>" sections both
in the Yocto Project Quick Start for requirements.</para></listitem>
<listitem><para><emphasis>Establish a local copy of the project files on your
system</emphasis>: You need this <link linkend='source-directory'>source
directory</link> available on your host system.
Having these files on your system gives you access to the build
process and to the tools you need.
For information on how to set up the source directory, see the
"<link linkend='getting-setup'>Getting Setup</link>" section.</para></listitem>
<listitem><para><emphasis>Establish a local copy of the base BSP files</emphasis>: Having
the BSP files on your system gives you access to the build
process and to the tools you need for creating a BSP.
For information on how to get these files, see the
"<link linkend='getting-setup'>Getting Setup</link>" section.</para></listitem>
<listitem><para><emphasis>Choose a BSP that is supported by the Yocto Project
as your base BSP</emphasis>:
The Yocto Project ships with several BSPs that support various hardware.
It is best to base your new BSP on an existing BSP rather than create all the
recipes and configuration files from scratch.
While it is possible to create everything from scratch, basing your new BSP
on something that is close is much easier.
Or, at a minimum, leveraging off an existing BSP
gives you some structure with which to start.</para>
<para>At this point you need to understand your target hardware well enough to determine which
existing BSP it most closely matches.
Things to consider are your hardwares on-board features, such as CPU type and graphics support.
You should look at the README files for supported BSPs to get an idea of which one
you could use.
A generic <trademark class='registered'>Intel</trademark>
<trademark class='trade'>Atom</trademark>-based BSP to consider is the
Crown Bay that does not support the <trademark class='registered'>Intel</trademark>
Embedded Media Graphics Driver (EMGD).
The remainder of this example uses that base BSP.</para>
<para>To see the supported BSPs, go to the
<ulink url='&YOCTO_HOME_URL;/download'>Download</ulink> page on the Yocto Project
website and click on “BSP Downloads.”</para></listitem>
<listitem><para><emphasis>Create your own BSP layer</emphasis>: Layers are ideal for
isolating and storing work for a given piece of hardware.
A layer is really just a location or area in which you place the recipes for your BSP.
In fact, a BSP is, in itself, a special type of layer.
</para>
<para>
Another example that illustrates a layer is an application.
Suppose you are creating an application that has library or other dependencies in
order for it to compile and run.
The layer, in this case, would be where all the recipes that define those dependencies
are kept.
The key point for a layer is that it is an isolated area that contains
all the relevant information for the project that the OpenEmbedded build
system knows about.
For more information on layers, see the
"<link linkend='understanding-and-creating-layers'>Understanding and Creating Layers</link>"
section.
For more information on BSP layers, see the
"<ulink url='&YOCTO_DOCS_BSP_URL;#bsp-layers'>BSP Layers</ulink>" section in the
Yocto Project Board Support Package (BSP) Developer's Guide.</para>
<note>Four BSPs exist that are part of the
Yocto Project release: <filename>atom-pc</filename>, <filename>beagleboard</filename>,
<filename>mpc8315e</filename>, and <filename>routerstationpro</filename>.
The recipes and configurations for these four BSPs are located and dispersed
within the <link linkend='source-directory'>source directory</link>.
On the other hand, BSP layers for Crown Bay, Emenlow, Jasper Forest,
N450, Cedar Trail, Fish River, Fish River Island II, Romley, sys940x, tlk,
and Sugar Bay exist in their own separate layers within the larger
<filename>meta-intel</filename> layer.</note>
<para>When you set up a layer for a new BSP, you should follow a standard layout.
This layout is described in the section
"<ulink url='&YOCTO_DOCS_BSP_URL;#bsp-filelayout'>Example Filesystem Layout</ulink>"
section of the Board Support Package (BSP) Development Guide.
In the standard layout, you will notice a suggested structure for recipes and
configuration information.
You can see the standard layout for the Crown Bay BSP in this example by examining the
directory structure of the <filename>meta-crownbay</filename> layer inside the
source directory.</para></listitem>
<listitem><para><emphasis>Make configuration changes to your new BSP
layer</emphasis>: The standard BSP layer structure organizes the files you need
to edit in <filename>conf</filename> and several <filename>recipes-*</filename>
directories within the BSP layer.
Configuration changes identify where your new layer is on the local system
and identify which kernel you are going to use.
</para></listitem>
<listitem><para><emphasis>Make recipe changes to your new BSP layer</emphasis>: Recipe
changes include altering recipes (<filename>.bb</filename> files), removing
recipes you don't use, and adding new recipes that you need to support your hardware.
</para></listitem>
<listitem><para><emphasis>Prepare for the build</emphasis>: Once you have made all the
changes to your BSP layer, there remains a few things
you need to do for the OpenEmbedded build system in order for it to create your image.
You need to get the build environment ready by sourcing an environment setup script
and you need to be sure two key configuration files are configured appropriately.</para>
<para>The entire process for building an image is overviewed in the section
"<ulink url='&YOCTO_DOCS_QS_URL;#building-image'>Building an Image</ulink>" section
of the Yocto Project Quick Start.
You might want to reference this information.</para></listitem>
<listitem><para><emphasis>Build the image</emphasis>: The OpenEmbedded build system
uses the BitBake tool to build images based on the type of image you want to create.
You can find more information on BitBake
<ulink url='&BITBAKE_DOCS_URL;'>here</ulink>.</para>
<para>The build process supports several types of images to satisfy different needs.
See the
"<ulink url='&YOCTO_DOCS_REF_URL;#ref-images'>Images</ulink>" chapter
in the Yocto Project Reference Manual for information on
supported images.</para></listitem>
</orderedlist>
</para>
<para>
You can view a video presentation on "Building Custom Embedded Images with Yocto"
at <ulink url='http://free-electrons.com/blog/elc-2011-videos'>Free Electrons</ulink>.
You can also find supplemental information in
<ulink url='&YOCTO_DOCS_BSP_URL;'>
The Board Support Package (BSP) Development Guide</ulink>.
Finally, there is wiki page write up of the example also located
<ulink url='&YOCTO_WIKI_URL;/wiki/Transcript:_creating_one_generic_Atom_BSP_from_another'>
here</ulink> that you might find helpful.
</para>
</section>
<section id='modifying-the-kernel'>
<title><anchor id='kernel-spot' />Modifying the Kernel</title>
<para>
Kernel modification involves changing the Yocto Project kernel, which could involve changing
configuration options as well as adding new kernel recipes.
Configuration changes can be added in the form of configuration fragments, while recipe
modification comes through the kernel's <filename>recipes-kernel</filename> area
in a kernel layer you create.
</para>
<para>
The remainder of this section presents a high-level overview of the Yocto Project
kernel architecture and the steps to modify the kernel.
For a complete discussion of the kernel, see
<ulink url='&YOCTO_DOCS_KERNEL_URL;'>
The Yocto Project Kernel Architecture and Use Manual</ulink>.
You can reference the appendix
"<link linkend='dev-manual-kernel-appendix'>Kernel Modification Example</link>"
for a detailed example that changes the configuration of a kernel.
</para>
<section id='kernel-overview'>
<title>Kernel Overview</title>
<para>
Traditionally, when one thinks of a patched kernel, they think of a base kernel
source tree and a fixed structure that contains kernel patches.
The Yocto Project, however, employs mechanisms, that in a sense, result in a kernel source
generator.
By the end of this section, this analogy will become clearer.
</para>
<para>
You can find a web interface to the Yocto Project kernel source repositories at
<ulink url='&YOCTO_GIT_URL;'></ulink>.
If you look at the interface, you will see to the left a grouping of
Git repositories titled "Yocto Linux Kernel."
Within this group, you will find several kernels supported by
the Yocto Project:
<itemizedlist>
<listitem><para><emphasis><filename>linux-yocto-2.6.34</filename></emphasis> - The
stable Yocto Project kernel that is based on the Linux 2.6.34 released kernel.</para></listitem>
<listitem><para><emphasis><filename>linux-yocto-2.6.37</filename></emphasis> - The
stable Yocto Project kernel that is based on the Linux 2.6.37 released kernel.</para></listitem>
<listitem><para><emphasis><filename>linux-yocto-3.0</filename></emphasis> - The stable
Yocto Project kernel that is based on the Linux 3.0 released kernel.</para></listitem>
<listitem><para><emphasis><filename>linux-yocto-3.0-1.1.x</filename></emphasis> - The
stable Yocto Project kernel to use with the Yocto Project Release 1.1.x. This kernel
is based on the Linux 3.0 released kernel.</para></listitem>
<listitem><para><emphasis><filename>linux-yocto-3.2</filename></emphasis> - The
stable Yocto Project kernel to use with the Yocto Project Release 1.2. This kernel
is based on the Linux 3.2 released kernel.</para></listitem>
<listitem><para><emphasis><filename>linux-yocto-dev</filename></emphasis> - A development
kernel based on the latest upstream release candidate available.</para></listitem>
</itemizedlist>
</para>
<para>
The kernels are maintained using the Git revision control system
that structures them using the familiar "tree", "branch", and "leaf" scheme.
Branches represent diversions from general code to more specific code, while leaves
represent the end-points for a complete and unique kernel whose source files
when gathered from the root of the tree to the leaf accumulate to create the files
necessary for a specific piece of hardware and its features.
The following figure displays this concept:
<para>
<imagedata fileref="figures/kernel-overview-1.png"
width="6in" depth="6in" align="center" scale="100" />
</para>
<para>
Within the figure, the "Kernel.org Branch Point" represents the point in the tree
where a supported base kernel is modified from the Linux kernel.
For example, this could be the branch point for the <filename>linux-yocto-3.0</filename>
kernel.
Thus, everything further to the right in the structure is based on the
<filename>linux-yocto-3.0</filename> kernel.
Branch points to right in the figure represent where the
<filename>linux-yocto-3.0</filename> kernel is modified for specific hardware
or types of kernels, such as real-time kernels.
Each leaf thus represents the end-point for a kernel designed to run on a specific
targeted device.
</para>
<para>
The overall result is a Git-maintained repository from which all the supported
kernel types can be derived for all the supported devices.
A big advantage to this scheme is the sharing of common features by keeping them in
"larger" branches within the tree.
This practice eliminates redundant storage of similar features shared among kernels.
</para>
<note>
Keep in mind the figure does not take into account all the supported Yocto
Project kernel types, but rather shows a single generic kernel just for conceptual purposes.
Also keep in mind that this structure represents the Yocto Project source repositories
that are either pulled from during the build or established on the host development system
prior to the build by either cloning a particular kernel's Git repository or by
downloading and unpacking a tarball.
</note>
<para>
Storage of all the available kernel source code is one thing, while representing the
code on your host development system is another.
Conceptually, you can think of the kernel source repositories as all the
source files necessary for all the supported kernels.
As a developer, you are just interested in the source files for the kernel on
on which you are working.
And, furthermore, you need them available on your host system.
</para>
<para>
You make kernel source code available on your host development system by using
Git to create a bare clone of the Yocto Project kernel Git repository
in which you are interested.
Then, you use Git again to clone a copy of that bare clone.
This copy represents the directory structure on your host system that is particular
to the kernel you want.
These are the files you actually modify to change the kernel.
See the <link linkend='local-kernel-files'>Yocto Project Kernel</link> item earlier
in this manual for an example of how to set up the kernel source directory
structure on your host system.
</para>
<para>
This next figure illustrates how the kernel source files might be arranged on
your host system.
</para>
<para>
<imagedata fileref="figures/kernel-overview-3-denzil.png"
width="6in" depth="4in" align="center" scale="100" />
</para>
<para>
In the previous figure, the file structure on the left represents the bare clone
set up to track the Yocto Project kernel Git repository.
The structure on the right represents the copy of the bare clone.
When you make modifcations to the kernel source code, this is the area in which
you work.
Once you make corrections, you must use Git to push the committed changes to the
bare clone.
The example in <xref linkend='modifying-the-kernel-source-code'>
Modifying the Kernel Source Code</xref> provides a detailed example.
</para>
<para>
What happens during the build?
When you build the kernel on your development system all files needed for the build
are taken from the source repositories pointed to by the
<filename>SRC_URI</filename> variable and gathered in a temporary work area
where they are subsequently used to create the unique kernel.
Thus, in a sense, the process constructs a local source tree specific to your
kernel to generate the new kernel image - a source generator if you will.
</para>
The following figure shows the temporary file structure
created on your host system when the build occurs.
This build directory contains all the source files used during the build.
</para>
<para>
<imagedata fileref="figures/kernel-overview-2.png"
width="6in" depth="5in" align="center" scale="100" />
</para>
<para>
Again, for a complete discussion of the Yocto Project kernel's architecture and its
branching strategy,
see <ulink url='&YOCTO_DOCS_KERNEL_URL;'>
The Yocto Project Kernel Architecture and Use Manual</ulink>.
You can also reference the
"<link linkend='modifying-the-kernel-source-code'>Modifying the Kernel Source Code</link>"
section for a detailed example that modifies the kernel.
</para>
</section>
<section id='kernel-modification-workflow'>
<title>Kernel Modification Workflow</title>
<para>
This illustration and the following list summarizes the kernel modification general workflow.
</para>
<para>
<imagedata fileref="figures/kernel-dev-flow.png"
width="6in" depth="7.5in" align="center" scalefit="1" />
</para>
<para>
<orderedlist>
<listitem><para><emphasis>Set up your host development system to support
development using the Yocto Project</emphasis>: See
"<ulink url='&YOCTO_DOCS_QS_URL;#the-linux-distro'>The Linux Distributions</ulink>" and
"<ulink url='&YOCTO_DOCS_QS_URL;#packages'>The Packages</ulink>" sections both
in the Yocto Project Quick Start for requirements.</para></listitem>
<listitem><para><emphasis>Establish a local copy of project files on your
system</emphasis>: Having the <link linkend='source-directory'>source
directory</link> on your system gives you access to the build process and tools
you need.
For information on how to get these files, see the bulleted item
"<link linkend='local-yp-release'>Yocto Project Release</link>" earlier in this manual.
</para></listitem>
<listitem><para><emphasis>Set up a local copy of the <filename>poky-extras</filename> Git
repository</emphasis>: This local repository is the area for your configuration
fragments, new kernel recipes, and the kernel <filename>.bbappend</filename>
file used during the build.
It is good practice to set this repository up inside your local
source directory.
For information on how to get these files, see the bulleted item
"<link linkend='poky-extras-repo'>The <filename>poky-extras</filename> Git Repository</link>"
earlier in this manual.
<note>While it is certainly possible to modify the kernel without involving
a local Git repository, the suggested workflow for kernel modification
using the Yocto Project does use a Git repository.</note></para></listitem>
<listitem><para><emphasis>Establish a local copy of the Yocto Project kernel files on your
system</emphasis>: In order to make modifications to the kernel you need two things:
a bare clone of the Yocto Project kernel you are modifying and
a copy of that bare clone.
The bare clone is required by the build process and is the area to which you
push your kernel source changes (pulling does not work with bare clones).
The copy of the bare clone is a local Git repository that contains all the kernel's
source files.
You make your changes to the files in this copy of the bare clone.
For information on how to set these two items up, see the bulleted item
"<link linkend='local-kernel-files'>Yocto Project Kernel</link>"
earlier in this manual.</para></listitem>
<listitem><para><emphasis>Make changes to the kernel source code if
applicable</emphasis>: Modifying the kernel does not always mean directly
changing source files.
However, if you have to do this, you make the changes in the local
Git repository you set up to hold the source files (i.e. the copy of the
bare clone).
Once the changes are made, you need to use Git commands to commit the changes
and then push them to the bare clone.</para></listitem>
<listitem><para><emphasis>Make kernel configuration changes
if applicable</emphasis>:
If your situation calls for changing the kernel's configuration, you can
use <filename>menuconfig</filename>
to enable and disable kernel configurations.
Using <filename>menuconfig</filename> allows you to interactively develop and test the
configuration changes you are making to the kernel.
When saved, changes using <filename>menuconfig</filename> update the kernel's
<filename>.config</filename>.
Try to resist the temptation of directly editing the <filename>.config</filename>
file found in the
<link linkend='build-directory'>build directory</link> at
<filename>tmp/sysroots/&lt;machine-name&gt;/kernel</filename>.
Doing so, can produce unexpected results when the OpenEmbedded build system
regenerates the configuration file.</para>
<para>Once you are satisfied with the configuration changes made using
<filename>menuconfig</filename>, you can directly examine the
<filename>.config</filename> file against a saved original and gather those
changes into a config fragment to be referenced from within the kernel's
<filename>.bbappend</filename> file.</para></listitem>
<listitem><para><emphasis>Add or extend kernel recipes if applicable</emphasis>:
The standard
layer structure organizes recipe files inside the
<filename>meta-kernel-dev</filename> layer that is within the local
<filename>poky-extras</filename> Git repository.
If you need to add new kernel recipes, you add them within this layer.
Also within this area, you will find the <filename>.bbappend</filename>
file that appends information to the kernel's recipe file used during the
build.
</para></listitem>
<listitem><para><emphasis>Prepare for the build</emphasis>: Once you have made all the
changes to your kernel (configurations, source code changes, recipe additions,
or recipe changes), there remains a few things
you need to do in order for the build system to create your image.
If you have not done so, you need to get the build environment ready by sourcing
the environment setup script described earlier.
You also need to be sure two key configuration files
(<filename>local.conf</filename> and <filename>bblayers.conf</filename>)
are configured appropriately.</para>
<para>The entire process for building an image is overviewed in the
"<ulink url='&YOCTO_DOCS_QS_URL;#building-image'>Building an Image</ulink>"
section of the Yocto Project Quick Start.
You might want to reference this information.
Also, you should look at the detailed examples found in the appendices at
at the end of this manual.</para></listitem>
<listitem><para><emphasis>Build the image</emphasis>: The OpenEmbedded
build system uses the BitBake
tool to build images based on the type of image you want to create.
You can find more information on BitBake
<ulink url='&BITBAKE_DOCS_URL;'>here</ulink>.</para>
<para>The build process supports several types of images to satisfy different needs.
See the "<ulink url='&YOCTO_DOCS_REF_URL;#ref-images'>Images</ulink>" chapter in
the Yocto Project Reference Manual for information on supported images.</para></listitem>
<listitem><para><emphasis>Make your configuration changes available
in the kernel layer</emphasis>: Up to this point, all the configuration changes to the
kernel have been done and tested iteratively.
Once they are tested and ready to go, you can move them into the kernel layer,
which allows you to distribute the layer.</para></listitem>
<listitem><para><emphasis>If applicable, share your in-tree changes</emphasis>:
If the changes you made
are suited for all Yocto Project kernel users, you might want to send them on
for inclusion into the upstream kernel's Git repository.
If the changes are accepted, the Yocto Project Maintainer pulls them into
the master branch of the kernel tree.
Doing so makes them available to everyone using the kernel.</para></listitem>
</orderedlist>
</para>
</section>
</section>
</section>
<section id='application-development-workflow'>
<title>Application Development Workflow</title>
<para>
Application development involves creation of an application that you want to be able
to run on your target hardware, which is running a Yocto Project kernel image.
The Yocto Project provides an Application Development Toolkit (ADT) that
facilitates quick development and integration of your application into its run-time environment.
Using the ADT you can employ cross-development toolchains designed for your target hardware
to compile and link your application.
You can then deploy your application to the actual hardware or to the QEMU emulator for testing.
If you are familiar with the popular Eclipse IDE, you can use an Eclipse Yocto Plug-in to
allow you to develop, deploy, and test your application all from within Eclipse.
</para>
<para>
While we strongly suggest using the ADT to develop your application, you might
not want to.
If this is the case, you can still use pieces of the Yocto Project for your development process.
However, because the process can vary greatly, this manual does not provide detail on the process.
</para>
<section id='workflow-using-the-adt-and-eclipse'>
<title>Workflow Using the ADT and <trademark class='trade'>Eclipse</trademark></title>
<para>
To help you understand how application development works using the ADT, this section
provides an overview of the general development process.
If you want to see a detailed example of the process as it is used from within the Eclipse
IDE, see
<ulink url='&YOCTO_DOCS_ADT_URL;'>
The Application Development Toolkit (ADT) User's Manual</ulink>.
</para>
<para>
The following illustration and list summarize the application development general workflow.
</para>
<para>
<imagedata fileref="figures/app-dev-flow.png"
width="7in" depth="8in" align="center" scale="100" />
</para>
<para>
<orderedlist>
<listitem><para><emphasis>Prepare the Host System for the Yocto Project</emphasis>:
See
"<ulink url='&YOCTO_DOCS_QS_URL;#the-linux-distro'>The Linux Distributions</ulink>" and
"<ulink url='&YOCTO_DOCS_QS_URL;#packages'>The Packages</ulink>" sections both
in the Yocto Project Quick Start for requirements.</para></listitem>
<listitem><para><emphasis>Secure the Yocto Project Kernel Target Image</emphasis>:
You must have a target kernel image that has been built using the OpenEmbeded
build system.</para>
<para>Depending on whether the Yocto Project has a pre-built image that matches your target
architecture and where you are going to run the image while you develop your application
(QEMU or real hardware), the area from which you get the image differs.
<itemizedlist>
<listitem><para>Download the image from
<ulink url='&YOCTO_MACHINES_DL_URL;'>
<filename>machines</filename></ulink> if your target architecture is supported
and you are going to develop and test your application on actual hardware.
</para></listitem>
<listitem><para>Download the image from the
<ulink url='&YOCTO_QEMU_DL_URL;'>
<filename>machines/qemu</filename></ulink> if your target architecture is supported
and you are going to develop and test your application using the QEMU
emulator.</para></listitem>
<listitem><para>Build your image if you cannot find a pre-built image that matches
your target architecture.
If your target architecture is similar to a supported architecture, you can
modify the kernel image before you build it.
See the
"<link linkend='kernel-modification-workflow'>Kernel Modification Workflow</link>"
section earlier in this manual for information on how to create a modified
Yocto Project kernel.</para></listitem>
</itemizedlist></para>
<para>For information on pre-built kernel image naming schemes for images
that can run on the QEMU emulator, see the
"<ulink url='&YOCTO_DOCS_QS_URL;#using-pre-built'>Using Pre-Built Binaries and QEMU</ulink>"
section in the Yocto Project Quick Start.</para></listitem>
<listitem><para><emphasis>Install the ADT</emphasis>:
The ADT provides a target-specific cross-development toolchain, the root filesystem,
the QEMU emulator, and other tools that can help you develop your application.
While it is possible to get these pieces separately, the ADT Installer provides an
easy method.
You can get these pieces by running an ADT installer script, which is configurable.
For information on how to install the ADT, see the
"<ulink url='&YOCTO_DOCS_ADT_URL;#using-the-adt-installer'>Using the ADT Installer</ulink>"
section
in the Yocto Project Application Development (ADT) User's Manual.</para></listitem>
<listitem><para><emphasis>If Applicable, Secure the Target Root Filesystem</emphasis>:
If you choose not to install the ADT using the ADT Installer,
you need to find and download the
appropriate root filesystems.
You can find these tarballs in the same areas used for the kernel images.
Depending on the type of image you are running, the root filesystem you need differs.
For example, if you are developing an application that runs on an image that
supports Sato, you need to get root filesystem that supports Sato.
</para></listitem>
<listitem><para><emphasis>Create and Build your Application</emphasis>:
At this point, you need to have source files for your application.
Once you have the files, you can use the Eclipse IDE to import them and build the
project.
If you are not using Eclipse, you need to use the cross-development tools you have
installed to create the image.</para></listitem>
<listitem><para><emphasis>Deploy the Image with the Application</emphasis>:
If you are using the Eclipse IDE, you can deploy your image to the hardware or to
QEMU through the project's preferences.
If you are not using the Eclipse IDE, then you need to deploy the application using
other methods to the hardware.
Or, if you are using QEMU, you need to use that tool and load your image in for testing.
</para></listitem>
<listitem><para><emphasis>Test and Debug the Application</emphasis>:
Once your application is deployed, you need to test it.
Within the Eclipse IDE, you can use the debubbing environment along with the
set of user-space tools installed along with the ADT to debug your application.
Of course, the same user-space tools are available separately to use if you choose
not to use the Eclipse IDE.</para></listitem>
</orderedlist>
</para>
</section>
<section id='workflow-without-adt'>
<title>Workflow Without ADT</title>
<para>
If you want to develop an application outside of the Yocto Project ADT environment, you
can still employ the cross-development toolchain, the QEMU emulator, and a number of supported
target image files.
You just need to follow these general steps:
<orderedlist>
<listitem><para><emphasis>Install the cross-development toolchain for your target hardware:</emphasis>
For information on how to install the toolchain, see the
"<ulink url='&YOCTO_DOCS_ADT_URL;#using-an-existing-toolchain-tarball'>Using a Cross-Toolchain Tarball</ulink>"
section
in the Yocto Project Application Development (ADT) User's Manual.</para></listitem>
<listitem><para><emphasis>Download the Target Image:</emphasis> The Yocto Project supports
several target architectures and has many pre-built kernel images and root filesystem
images.</para>
<para>If you are going to develop your application on hardware, go to the
<ulink url='&YOCTO_MACHINES_DL_URL;'>
<filename>machines</filename></ulink> download area and choose a target machine area
from which to download the kernel image and root filesystem.
This download area could have several files in it that support development using
actual hardware.
For example, the area might contain <filename>.hddimg</filename> files that combine the
kernel image with the filesystem, boot loaders, etc.
Be sure to get the files you need for your particular development process.</para>
<para>If you are going to develop your application and then run and test it using the QEMU
emulator, go to the
<ulink url='&YOCTO_QEMU_DL_URL;'>
<filename>machines/qemu</filename></ulink> download area.
From this area, go down into the directory for your target architecture
(e.g. <filename>qemux86_64</filename> for an
<trademark class='registered'>Intel</trademark>-based 64-bit architecture).
Download kernel, root filesystem, and any other files you need for your process.
<note>In order to use the root filesystem in QEMU, you need to extract it.
See the
"<ulink url='&YOCTO_DOCS_ADT_URL;#extracting-the-root-filesystem'>Extracting the Root Filesystem</ulink>" section for information on how to extract the
root filesystem.</note></para></listitem>
<listitem><para><emphasis>Develop and Test your Application:</emphasis> At this point,
you have the tools to develop your application.
If you need to separately install and use the QEMU emulator, you can go to
<ulink url='http://www.qemu.org'>QEMU Home Page</ulink> to download and learn about the
emulator.</para></listitem>
</orderedlist>
</para>
</section>
</section>
<section id="modifying-temporary-source-code">
<title>Modifying Temporary Source Code</title>
<para>
You might
find it helpful during development to modify the temporary source code used by recipes
to build packages.
For example, suppose you are developing a patch and you need to experiment a bit
to figure out your solution.
After you have initially built the package, you can iteratively tweak the
source code, which is located in the
<link linkend='build-directory'>build directory</link>, and then
you can force a re-compile and quickly test your altered code.
Once you settle on a solution, you can then preserve your changes in the form of
patches.
You can accomplish these steps all within either a
<ulink url='http://savannah.nongnu.org/projects/quilt'>Quilt</ulink> or
<link linkend='git'>Git</link> workflow.
</para>
<section id='finding-the-temporary-source-code'>
<title>Finding the Temporary Source Code</title>
<para>
During a build, the unpacked temporary source code used by recipes
to build packages is available in the build directory as
defined by the
<filename><ulink url='&YOCTO_DOCS_REF_URL;#var-S'>S</ulink></filename> variable.
Below is the default value for the <filename>S</filename> variable as defined in the
<filename>meta/conf/bitbake.conf</filename> configuration file in the
<link linkend='source-directory'>source directory</link>:
<literallayout class='monospaced'>
S = ${WORKDIR}/${BP}
</literallayout>
You should be aware that many recipes override the <filename>S</filename> variable.
For example, recipes that fetch their source from Git usually set
<filename>S</filename> to <filename>${WORKDIR}/git</filename>.
<note>
<filename>BP</filename> represents the "Base Package", which is the base package
name and the package version:
<literallayout class='monospaced'>
BP = ${BPN}-${PV}
</literallayout>
</note>
</para>
<para>
The path to the work directory for the recipe
(<ulink url='&YOCTO_DOCS_REF_URL;#var-WORKDIR'><filename>WORKDIR</filename></ulink>) depends
on the package name and the architecture of the target device.
For example, here is the work directory for packages whose targets are not device-dependent:
<literallayout class='monospaced'>
${TMPDIR}/work/${PACKAGE_ARCH}-poky-${TARGET_OS}/${PN}-${PV}-${PR}
</literallayout>
Let's look at an example without variables.
Assuming a top-level source directory named <filename>poky</filename>
and a default build directory of <filename>poky/build</filename>,
the following is the work directory for the <filename>acl</filename> package:
<literallayout class='monospaced'>
~/poky/build/tmp/work/i586-poky-linux/acl-2.2.51-r3
</literallayout>
</para>
<para>
If your package is dependent on the target device, the work directory varies slightly:
<literallayout class='monospaced'>
${TMPDIR}/work/${MACHINE}-poky-${TARGET_OS}/${PN}-${PV}-${PR}
</literallayout>
Again, assuming top-level source directory named <filename>poky</filename>
and a default build directory of <filename>poky/build</filename>, the
following is the work directory for the <filename>acl</filename> package that is being
built for a MIPS-based device:
<literallayout class='monospaced'>
~/poky/build/tmp/work/mips-poky-linux/acl-2.2.51-r2
</literallayout>
</para>
<note>
To better understand how the OpenEmbedded build system resolves directories during the
build process, see the glossary entries for the
<ulink url='&YOCTO_DOCS_REF_URL;#var-WORKDIR'><filename>WORKDIR</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-TMPDIR'><filename>TMPDIR</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-TOPDIR'><filename>TOPDIR</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-PACKAGE_ARCH'><filename>PACKAGE_ARCH</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-TARGET_OS'><filename>TARGET_OS</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-PN'><filename>PN</filename></ulink>,
<ulink url='&YOCTO_DOCS_REF_URL;#var-PV'><filename>PV</filename></ulink>,
and
<ulink url='&YOCTO_DOCS_REF_URL;#var-PR'><filename>PR</filename></ulink>
variables in the Yocto Project Reference Manual.
</note>
<para>
Now that you know where to locate the directory that has the temporary source code, you can use a
Quilt or Git workflow to make your edits, test the changes, and preserve the
changes in the form of patches.
</para>
</section>
<section id="using-a-quilt-workflow">
<title>Using a Quilt Workflow</title>
<para>
<ulink url='http://savannah.nongnu.org/projects/quilt'>Quilt</ulink>
is a powerful tool that allows you to capture source code changes without having
a clean source tree.
This section outlines the typical workflow you can use to modify temporary source code,
test changes, and then preserve the changes in the form of a patch all using Quilt.
</para>
<para>
Follow these general steps:
<orderedlist>
<listitem><para><emphasis>Find the Source Code:</emphasis>
The temporary source code used by the OpenEmbedded build system is kept in the
build directory.
See the
"<link linkend='finding-the-temporary-source-code'>Finding the Temporary Source Code</link>"
section to learn how to locate the directory that has the temporary source code for a
particular package.</para></listitem>
<listitem><para><emphasis>Change Your Working Directory:</emphasis>
You need to be in the directory that has the temporary source code.
That directory is defined by the
<ulink url='&YOCTO_DOCS_REF_URL;#var-S'>S</ulink>
variable.</para></listitem>
<listitem><para><emphasis>Create a New Patch:</emphasis>
Before modifying source code, you need to create a new patch.
To create a new patch file, use <filename>quilt new</filename> as below:
<literallayout class='monospaced'>
$ quilt new my_changes.patch
</literallayout></para></listitem>
<listitem><para><emphasis>Notify Quilt and Add Files:</emphasis>
After creating the patch, you need to notify Quilt about the files you will
be changing.
Add the files you will be modifying into the patch you just created:
<literallayout class='monospaced'>
$ quilt add file1.c file2.c file3.c
</literallayout></para></listitem>
<listitem><para><emphasis>Edit the Files:</emphasis>
Make the changes to the temporary source code.</para></listitem>
<listitem><para><emphasis>Test Your Changes:</emphasis>
Once you have modified the source code, the easiest way to test your changes
is by calling the <filename>compile</filename> task as shown in the following example:
<literallayout class='monospaced'>
$ bitbake -c compile -f &lt;name_of_package&gt;
</literallayout>
The <filename>-f</filename> or <filename>--force</filename>
option forces re-execution of the specified task.
If you find problems with your code, you can just keep editing and
re-testing iteratively until things work as expected.
<note>All the modifications you make to the temporary source code
disappear once you <filename>-c clean</filename> or
<filename>-c cleanall</filename> with BitBake for the package.
Modifications will also disappear if you use the <filename>rm_work</filename>
feature as described in the
"<ulink url='&YOCTO_DOCS_QS_URL;#building-image'>Building an Image</ulink>"
section of the Yocto Project Quick Start.
</note></para></listitem>
<listitem><para><emphasis>Generate the Patch:</emphasis>
Once your changes work as expected, you need to use Quilt to generate the final patch that
contains all your modifications.
<literallayout class='monospaced'>
$ quilt refresh
</literallayout>
At this point the <filename>my_changes.patch</filename> file has all your edits made
to the <filename>file1.c</filename>, <filename>file2.c</filename>, and
<filename>file3.c</filename> files.</para>
<para>You can find the resulting patch file in the <filename>patches/</filename>
subdirectory of the source (<filename>S</filename>) directory.</para></listitem>
<listitem><para><emphasis>Copy the Patch File:</emphasis>
For simplicity, copy the patch file into a directory named <filename>files</filename>,
which you can create in the same directory as the recipe.
Placing the patch here guarantees that the OpenEmbedded build system will find
the patch.
Next, add the patch into the
<filename><ulink url='&YOCTO_DOCS_REF_URL;#var-SRC_URI'>SRC_URI</ulink></filename>
of the recipe.
Here is an example:
<literallayout class='monospaced'>
SRC_URI += "file://my_changes.patch"
</literallayout></para></listitem>
<listitem><para><emphasis>Increment the Package Revision Number:</emphasis>
Finally, don't forget to 'bump' the
<filename><ulink url='&YOCTO_DOCS_REF_URL;#var-PR'>PR</ulink></filename>
value in the same recipe since the resulting packages have changed.</para></listitem>
</orderedlist>
</para> </section>
<section id='using-a-git-workflow'>
<title>Using a Git Workflow</title>
<para>
Git is an even more powerful tool that allows you to capture source code changes without having
a clean source tree.
This section outlines the typical workflow you can use to modify temporary source code,
test changes, and then preserve the changes in the form of a patch all using Git.
For general information on Git as it is used in the Yocto Project, see the
"<link linkend='git'>Git</link>" section.
</para>
<note>
This workflow uses Git only for its ability to manage local changes to the source code
and produce patches independent of any version control system used with the Yocto Project.
</note>
<para>
Follow these general steps:
<orderedlist>
<listitem><para><emphasis>Find the Source Code:</emphasis>
The temporary source code used by the OpenEmbedded build system is kept in the
build directory.
See the
"<link linkend='finding-the-temporary-source-code'>Finding the Temporary Source Code</link>"
section to learn how to locate the directory that has the temporary source code for a
particular package.</para></listitem>
<listitem><para><emphasis>Change Your Working Directory:</emphasis>
You need to be in the directory that has the temporary source code.
That directory is defined by the
<ulink url='&YOCTO_DOCS_REF_URL;#var-S'>S</ulink>
variable.</para></listitem>
<listitem><para><emphasis>Initialize a Git Repository:</emphasis>
Use the <filename>git init</filename> command to initialize a new local repository
that is based on the work directory:
<literallayout class='monospaced'>
$ git init
</literallayout></para></listitem>
<listitem><para><emphasis>Stage all the files:</emphasis>
Use the <filename>git add *</filename> command to stage all the files in the source
code directory so that they can be committed:
<literallayout class='monospaced'>
$ git add *
</literallayout></para></listitem>
<listitem><para><emphasis>Commit the Source Files:</emphasis>
Use the <filename>git commit</filename> command to initially commit all the files in
the work directory:
<literallayout class='monospaced'>
$ git commit
</literallayout>
At this point, your Git repository is aware of all the source code files.
Any edits you now make to files will be tracked by Git.</para></listitem>
<listitem><para><emphasis>Edit the Files:</emphasis>
Make the changes to the temporary source code.</para></listitem>
<listitem><para><emphasis>Test Your Changes:</emphasis>
Once you have modified the source code, the easiest way to test your changes
is by calling the <filename>compile</filename> task as shown in the following example:
<literallayout class='monospaced'>
$ bitbake -c compile -f &lt;name_of_package&gt;
</literallayout>
The <filename>-f</filename> or <filename>--force</filename>
option forces re-execution of the specified task.
If you find problems with your code, you can just keep editing and
re-testing iteratively until things work as expected.
<note>All the modifications you make to the temporary source code
disappear once you <filename>-c clean</filename> or
<filename>-c cleanall</filename> with BitBake for the package.
Modifications will also disappear if you use the <filename>rm_work</filename>
feature as described in the
"<ulink url='&YOCTO_DOCS_QS_URL;#building-image'>Building an Image</ulink>"
section of the Yocto Project Quick Start.
</note></para></listitem>
<listitem><para><emphasis>See the List of Files You Changed:</emphasis>
Use the <filename>git status</filename> command to see what files you have actually edited.
The ability to have Git track the files you have changed is an advantage that this
workflow has over the Quilt workflow.
Here is the Git command to list your changed files:
<literallayout class='monospaced'>
$ git status
</literallayout></para></listitem>
<listitem><para><emphasis>Stage the Modified Files:</emphasis>
Use the <filename>git add</filename> command to stage the changed files so they
can be committed as follows:
<literallayout class='monospaced'>
$ git add file1.c file2.c file3.c
</literallayout></para></listitem>
<listitem><para><emphasis>Commit the Staged Files and View Your Changes:</emphasis>
Use the <filename>git commit</filename> command to commit the changes to the
local repository.
Once you have committed the files, you can use the <filename>git log</filename>
command to see your changes:
<literallayout class='monospaced'>
$ git commit
$ git log
</literallayout></para></listitem>
<listitem><para><emphasis>Generate the Patch:</emphasis>
Once the changes are committed, use the <filename>git format-patch</filename>
command to generate a patch file:
<literallayout class='monospaced'>
$ git format-patch HEAD~1
</literallayout>
The <filename>HEAD~1</filename> part of the command causes Git to generate the
patch file for the most recent commit.</para>
<para>At this point, the patch file has all your edits made
to the <filename>file1.c</filename>, <filename>file2.c</filename>, and
<filename>file3.c</filename> files.
You can find the resulting patch file in the current directory.
The patch file ends with <filename>.patch</filename>.</para></listitem>
<listitem><para><emphasis>Copy the Patch File:</emphasis>
For simplicity, copy the patch file into a directory named <filename>files</filename>,
which you can create in the same directory as the recipe.
Placing the patch here guarantees that the OpenEmbedded build system will find
the patch.
Next, add the patch into the
<filename><ulink url='&YOCTO_DOCS_REF_URL;#var-SRC_URI'>SRC_URI</ulink></filename>
of the recipe.
Here is an example:
<literallayout class='monospaced'>
SRC_URI += "file://my_changes.patch"
</literallayout></para></listitem>
<listitem><para><emphasis>Increment the Package Revision Number:</emphasis>
Finally, don't forget to 'bump' the
<filename><ulink url='&YOCTO_DOCS_REF_URL;#var-PR'>PR</ulink></filename>
value in the same recipe since the resulting packages have changed.</para></listitem>
</orderedlist>
</para>
</section>
</section>
<section id='image-development-using-hob'>
<title>Image Development Using Hob</title>
<para>
The <ulink url='&YOCTO_HOME_URL;/projects/hob'>Hob</ulink> is a graphical user interface for the
OpenEmbedded build system, which is based on BitBake.
You can use the Hob to build custom operating system images within the Yocto Project build environment.
Hob simply provides a friendly interface over the build system used during system development.
In other words, building images with the Hob lets you take care of common build tasks more easily.
</para>
<para>
For a better understanding of Hob, see the project page at
<ulink url='&YOCTO_HOME_URL;/projects/hob'></ulink> on the Yocto Project website.
The page has a short introductory training video on Hob.
The following lists some features of Hob:
<itemizedlist>
<listitem><para>You can setup and run Hob using these commands:
<literallayout class='monospaced'>
$ source oe-init-build-env
$ hob
</literallayout></para></listitem>
<listitem><para>You can set the
<ulink url='&YOCTO_DOCS_REF_URL;#var-MACHINE'><filename>MACHINE</filename></ulink>
for which you are building the image.</para></listitem>
<listitem><para>You can modify various policy settings such as the package format used to build with,
the parrallelism BitBake uses, whether or not to build an external toolchain, and which host
to build against.</para></listitem>
<listitem><para>You can manage
<link linkend='understanding-and-creating-layers'>layers</link>.</para></listitem>
<listitem><para>You can select a base image and then add extra packages for your custom build.
</para></listitem>
<listitem><para>You can launch and monitor the build from within Hob.</para></listitem>
</itemizedlist>
</para>
</section>
<section id="platdev-appdev-devshell">
<title>Using a Development Shell</title>
<para>
When debugging certain commands or even when just editing packages,
<filename>devshell</filename> can be a useful tool.
When you invoke <filename>devshell</filename>, source files are
extracted into your working directory and patches are applied.
Then, a new terminal is opened and you are placed in the working directory.
In the new terminal, all the OpenEmbedded build-related environment variables are
still defined so you can use commands such as <filename>configure</filename> and
<filename>make</filename>.
The commands execute just as if the OpenEmbedded build system were executing them.
Consequently, working this way can be helpful when debugging a build or preparing
software to be used with the OpenEmbedded build system.
</para>
<para>
Following is an example that uses <filename>devshell</filename> on a target named
<filename>matchbox-desktop</filename>:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop -c devshell
</literallayout>
</para>
<para>
This command opens a terminal with a shell prompt within the OpenEmbedded build environment.
The default shell is xterm.
The following occurs:
<itemizedlist>
<listitem><para>The <filename>PATH</filename> variable includes the
cross-toolchain.</para></listitem>
<listitem><para>The <filename>pkgconfig</filename> variables find the correct
<filename>.pc</filename> files.</para></listitem>
<listitem><para>The <filename>configure</filename> command finds the
Yocto Project site files as well as any other necessary files.</para></listitem>
</itemizedlist>
Within this environment, you can run <filename>configure</filename>
or <filename>compile</filename> commands as if they were being run by
the OpenEmbedded build system itself.
As noted earlier, the working directory also automatically changes to the
source directory (<ulink url='&YOCTO_DOCS_REF_URL;#var-S'><filename>S</filename></ulink>).
</para>
<para>
When you are finished, you just exit the shell or close the terminal window.
</para>
<para>
Because an external shell is launched rather than opening directly into the
original terminal window, it allows easier interaction with BitBake's multiple
threads as well as accomodates a future client/server split.
</para>
<note>
<para>
It is worth remembering that when using <filename>devshell</filename>
you need to use the full compiler name such as <filename>arm-poky-linux-gnueabi-gcc</filename>
instead of just using <filename>gcc</filename>.
The same applies to other applications such as <filename>binutils</filename>,
<filename>libtool</filename> and so forth.
BitBake sets up environment variables such as <filename>CC</filename>
to assist applications, such as <filename>make</filename> to find the correct tools.
</para>
<para>
It is also worth noting that <filename>devshell</filename> still works over
X11 forwarding and similar situations
</para>
</note>
</section>
</chapter>
<!--
vim: expandtab tw=80 ts=4
-->