Gitiles
Code Review Sign In
gerrit.openbmc.org / stefanberger / openbmc / d7bf8c17eca8f8c89898a7794462c773c449e983 / . / import-layers / yocto-poky / documentation / ref-manual / usingpoky.xml
blob: 13d9ad6ab758bbea2ada9e53caa84ad70328a83f [file] [log] [blame]
<!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='usingpoky'>
<title>Using the Yocto Project</title>
<para>
This chapter describes common usage for the Yocto Project.
The information is introductory in nature as other manuals in the Yocto Project
documentation set provide more details on how to use the Yocto Project.
</para>
<section id='usingpoky-build'>
<title>Running a Build</title>
<para>
This section provides a summary of the build process and provides information
for less obvious aspects of the build process.
For general information on how to build an image using the OpenEmbedded build
system, see the
"<ulink url='&YOCTO_DOCS_QS_URL;#qs-building-images'>Building Images</ulink>"
section of the Yocto Project Quick Start.
</para>
<section id='build-overview'>
<title>Build Overview</title>
<para>
In the development environment you will need to build an image whenever you change hardware
support, add or change system libraries, or add or change services that have dependencies.
</para>
<mediaobject>
<imageobject>
<imagedata fileref="figures/building-an-image.png" format="PNG" align='center' scalefit='1'/>
</imageobject>
<caption>
<para>Building an Image</para>
</caption>
</mediaobject>
<para>
The first thing you need to do is set up the OpenEmbedded build
environment by sourcing the environment setup script
(i.e.
<link linkend='structure-core-script'><filename>&OE_INIT_FILE;</filename></link>).
Here is an example:
<literallayout class='monospaced'>
$ source &OE_INIT_FILE; [<replaceable>build_dir</replaceable>]
</literallayout>
</para>
<para>
The <replaceable>build_dir</replaceable> argument is optional and specifies the directory the
OpenEmbedded build system uses for the build -
the
<link linkend='build-directory'>Build Directory</link>.
If you do not specify a Build Directory, it defaults to a directory
named <filename>build</filename> in your current working directory.
A common practice is to use a different Build Directory for different targets.
For example, <filename>~/build/x86</filename> for a <filename>qemux86</filename>
target, and <filename>~/build/arm</filename> for a <filename>qemuarm</filename> target.
</para>
<para>
Once the build environment is set up, you can build a target using:
<literallayout class='monospaced'>
$ bitbake <replaceable>target</replaceable>
</literallayout>
<note>
<para>
If you experience a build error due to resources
temporarily being unavailable and it appears you
should not be having this issue, it might be due
to the combination of a 4.3+ Linux kernel and
<filename>systemd</filename> version 228+
(i.e. see this
<ulink url='http://unix.stackexchange.com/questions/253903/creating-threads-fails-with-resource-temporarily-unavailable-with-4-3-kernel'>link</ulink>
for information).
</para>
<para>
To work around this issue, you can try either
of the following:
<itemizedlist>
<listitem><para>
Try the build again.
</para></listitem>
<listitem><para>
Modify the "DefaultTasksMax"
<filename>systemd</filename> parameter
by uncommenting it and setting it to
"infinity".
You can find this parameter in the
<filename>system.conf</filename> file
located in
<filename>/etc/systemd</filename>
on most systems.
</para></listitem>
</itemizedlist>
</para>
</note>
</para>
<para>
The <replaceable>target</replaceable> is the name of the recipe you want to build.
Common targets are the images in <filename>meta/recipes-core/images</filename>,
<filename>meta/recipes-sato/images</filename>, etc. all found in the
<link linkend='source-directory'>Source Directory</link>.
Or, the target can be the name of a recipe for a specific piece of software such as
BusyBox.
For more details about the images the OpenEmbedded build system supports, see the
"<link linkend="ref-images">Images</link>" chapter.
</para>
<note>
Building an image without GNU General Public License Version
3 (GPLv3), or similarly licensed, components is supported for
only minimal and base images.
See the "<link linkend='ref-images'>Images</link>" chapter for more information.
</note>
</section>
<section id='building-an-image-using-gpl-components'>
<title>Building an Image Using GPL Components</title>
<para>
When building an image using GPL components, you need to maintain your original
settings and not switch back and forth applying different versions of the GNU
General Public License.
If you rebuild using different versions of GPL, dependency errors might occur
due to some components not being rebuilt.
</para>
</section>
</section>
<section id='usingpoky-install'>
<title>Installing and Using the Result</title>
<para>
Once an image has been built, it often needs to be installed.
The images and kernels built by the OpenEmbedded build system are placed in the
<link linkend='build-directory'>Build Directory</link> in
<filename class="directory">tmp/deploy/images</filename>.
For information on how to run pre-built images such as <filename>qemux86</filename>
and <filename>qemuarm</filename>, see the
<ulink url='&YOCTO_DOCS_SDK_URL;'>Yocto Project Application Development and the Extensible Software Development Kit (eSDK)</ulink>
manual.
For information about how to install these images, see the documentation for your
particular board or machine.
</para>
</section>
<section id='usingpoky-debugging-tools-and-techniques'>
<title>Debugging Tools and Techniques</title>
<para>
The exact method for debugging build failures depends on the nature of
the problem and on the system's area from which the bug originates.
Standard debugging practices such as comparison against the last
known working version with examination of the changes and the
re-application of steps to identify the one causing the problem are
valid for the Yocto Project just as they are for any other system.
Even though it is impossible to detail every possible potential failure,
this section provides some general tips to aid in debugging.
</para>
<para>
A useful feature for debugging is the error reporting tool.
Configuring the Yocto Project to use this tool causes the
OpenEmbedded build system to produce error reporting commands as
part of the console output.
You can enter the commands after the build completes
to log error information
into a common database, that can help you figure out what might be
going wrong.
For information on how to enable and use this feature, see the
"<ulink url='&YOCTO_DOCS_DEV_URL;#using-the-error-reporting-tool'>Using the Error Reporting Tool</ulink>"
section in the Yocto Project Development Tasks Manual.
</para>
<para>
For discussions on debugging, see the
"<ulink url='&YOCTO_DOCS_DEV_URL;#platdev-gdb-remotedebug'>Debugging With the GNU Project Debugger (GDB) Remotely</ulink>" section
in the Yocto Project Development Tasks Manual
and the
"<ulink url='&YOCTO_DOCS_SDK_URL;#adt-eclipse'>Working within Eclipse</ulink>"
section in the Yocto Project Application Development and the
Extensible Software Development Kit (eSDK) manual.
</para>
<note>
The remainder of this section presents many examples of the
<filename>bitbake</filename> command.
You can learn about BitBake by reading the
<ulink url='&YOCTO_DOCS_BB_URL;#bitbake-user-manual'>BitBake User Manual</ulink>.
</note>
<section id='usingpoky-debugging-viewing-logs-from-failed-tasks'>
<title>Viewing Logs from Failed Tasks</title>
<para>
You can find the log for a task in the file
<filename>${</filename><link linkend='var-WORKDIR'><filename>WORKDIR</filename></link><filename>}/temp/log.do_</filename><replaceable>taskname</replaceable>.
For example, the log for the
<link linkend='ref-tasks-compile'><filename>do_compile</filename></link>
task of the QEMU minimal image for the x86 machine
(<filename>qemux86</filename>) might be in
<filename>tmp/work/qemux86-poky-linux/core-image-minimal/1.0-r0/temp/log.do_compile</filename>.
To see the commands
<link linkend='bitbake-term'>BitBake</link> ran
to generate a log, look at the corresponding
<filename>run.do_</filename><replaceable>taskname</replaceable>
file in the same directory.
</para>
<para>
<filename>log.do_</filename><replaceable>taskname</replaceable> and
<filename>run.do_</filename><replaceable>taskname</replaceable>
are actually symbolic links to
<filename>log.do_</filename><replaceable>taskname</replaceable><filename>.</filename><replaceable>pid</replaceable>
and
<filename>log.run_</filename><replaceable>taskname</replaceable><filename>.</filename><replaceable>pid</replaceable>,
where <replaceable>pid</replaceable> is the PID the task had when
it ran.
The symlinks always point to the files corresponding to the most
recent run.
</para>
</section>
<section id='usingpoky-debugging-viewing-variable-values'>
<title>Viewing Variable Values</title>
<para>
BitBake's <filename>-e</filename> option is used to display
variable values after parsing.
The following command displays the variable values after the
configuration files (i.e. <filename>local.conf</filename>,
<filename>bblayers.conf</filename>,
<filename>bitbake.conf</filename> and so forth) have been
parsed:
<literallayout class='monospaced'>
$ bitbake -e
</literallayout>
The following command displays variable values after a specific
recipe has been parsed.
The variables include those from the configuration as well:
<literallayout class='monospaced'>
$ bitbake -e recipename
</literallayout>
<note><para>
Each recipe has its own private set of variables (datastore).
Internally, after parsing the configuration, a copy of the
resulting datastore is made prior to parsing each recipe.
This copying implies that variables set in one recipe will
not be visible to other recipes.</para>
<para>Likewise, each task within a recipe gets a private
datastore based on the recipe datastore, which means that
variables set within one task will not be visible to
other tasks.</para>
</note>
</para>
<para>
In the output of <filename>bitbake -e</filename>, each variable is
preceded by a description of how the variable got its value,
including temporary values that were later overriden.
This description also includes variable flags (varflags) set on
the variable.
The output can be very helpful during debugging.
</para>
<para>
Variables that are exported to the environment are preceded by
<filename>export</filename> in the output of
<filename>bitbake -e</filename>.
See the following example:
<literallayout class='monospaced'>
export CC="i586-poky-linux-gcc -m32 -march=i586 --sysroot=/home/ulf/poky/build/tmp/sysroots/qemux86"
</literallayout>
</para>
<para>
In addition to variable values, the output of the
<filename>bitbake -e</filename> and
<filename>bitbake -e</filename>&nbsp;<replaceable>recipe</replaceable>
commands includes the following information:
<itemizedlist>
<listitem><para>
The output starts with a tree listing all configuration
files and classes included globally, recursively listing
the files they include or inherit in turn.
Much of the behavior of the OpenEmbedded build system
(including the behavior of the
<link linkend='normal-recipe-build-tasks'>normal recipe build tasks</link>)
is implemented in the
<link linkend='ref-classes-base'><filename>base</filename></link>
class and the classes it inherits, rather than being built
into BitBake itself.
</para></listitem>
<listitem><para>
After the variable values, all functions appear in the
output.
For shell functions, variables referenced within the
function body are expanded.
If a function has been modified using overrides or
using override-style operators like
<filename>_append</filename> and
<filename>_prepend</filename>, then the final assembled
function body appears in the output.
</para></listitem>
</itemizedlist>
</para>
</section>
<section id='viewing-package-information-with-oe-pkgdata-util'>
<title>Viewing Package Information with <filename>oe-pkgdata-util</filename></title>
<para>
You can use the <filename>oe-pkgdata-util</filename> command-line
utility to query
<link linkend='var-PKGDATA_DIR'><filename>PKGDATA_DIR</filename></link>
and display various package-related information.
When you use the utility, you must use it to view information
on packages that have already been built.
</para>
<para>
Following are a few of the available
<filename>oe-pkgdata-util</filename> subcommands.
<note>
You can use the standard * and ? globbing wildcards as part of
package names and paths.
</note>
<itemizedlist>
<listitem><para>
<filename>oe-pkgdata-util list-pkgs [</filename><replaceable>pattern</replaceable><filename>]</filename>:
Lists all packages that have been built, optionally
limiting the match to packages that match
<replaceable>pattern</replaceable>.
</para></listitem>
<listitem><para>
<filename>oe-pkgdata-util list-pkg-files&nbsp;</filename><replaceable>package</replaceable><filename>&nbsp;...</filename>:
Lists the files and directories contained in the given
packages.
<note>
<para>
A different way to view the contents of a package is
to look at the
<filename>${</filename><link linkend='var-WORKDIR'><filename>WORKDIR</filename></link><filename>}/packages-split</filename>
directory of the recipe that generates the
package.
This directory is created by the
<link linkend='ref-tasks-package'><filename>do_package</filename></link>
task and has one subdirectory for each package the
recipe generates, which contains the files stored in
that package.</para>
<para>
If you want to inspect the
<filename>${WORKDIR}/packages-split</filename>
directory, make sure that
<link linkend='ref-classes-rm-work'><filename>rm_work</filename></link>
is not enabled when you build the recipe.
</para>
</note>
</para></listitem>
<listitem><para>
<filename>oe-pkgdata-util find-path&nbsp;</filename><replaceable>path</replaceable><filename>&nbsp;...</filename>:
Lists the names of the packages that contain the given
paths.
For example, the following tells us that
<filename>/usr/share/man/man1/make.1</filename>
is contained in the <filename>make-doc</filename>
package:
<literallayout class='monospaced'>
$ oe-pkgdata-util find-path /usr/share/man/man1/make.1
make-doc: /usr/share/man/man1/make.1
</literallayout>
</para></listitem>
<listitem><para>
<filename>oe-pkgdata-util lookup-recipe&nbsp;</filename><replaceable>package</replaceable><filename>&nbsp;...</filename>:
Lists the name of the recipes that
produce the given packages.
</para></listitem>
</itemizedlist>
</para>
<para>
For more information on the <filename>oe-pkgdata-util</filename>
command, use the help facility:
<literallayout class='monospaced'>
$ oe-pkgdata-util &dash;&dash;help
$ oe-pkgdata-util <replaceable>subcommand</replaceable> --help
</literallayout>
</para>
</section>
<section id='usingpoky-viewing-dependencies-between-recipes-and-tasks'>
<title>Viewing Dependencies Between Recipes and Tasks</title>
<para>
Sometimes it can be hard to see why BitBake wants to build other
recipes before the one you have specified.
Dependency information can help you understand why a recipe is
built.
</para>
<para>
To generate dependency information for a recipe, run the following
command:
<literallayout class='monospaced'>
$ bitbake -g <replaceable>recipename</replaceable>
</literallayout>
This command writes the following files in the current directory:
<itemizedlist>
<listitem><para>
<filename>pn-buildlist</filename>: A list of
recipes/targets involved in building
<replaceable>recipename</replaceable>.
"Involved" here means that at least one task from the
recipe needs to run when building
<replaceable>recipename</replaceable> from scratch.
Targets that are in
<link linkend='var-ASSUME_PROVIDED'><filename>ASSUME_PROVIDED</filename></link>
are not listed.
</para></listitem>
<listitem><para>
<filename>task-depends.dot</filename>: A graph showing
dependencies between tasks.
</para></listitem>
</itemizedlist>
</para>
<para>
The graphs are in
<ulink url='https://en.wikipedia.org/wiki/DOT_%28graph_description_language%29'>DOT</ulink>
format and can be converted to images (e.g. using the
<filename>dot</filename> tool from
<ulink url='http://www.graphviz.org/'>Graphviz</ulink>).
<note><title>Notes</title>
<itemizedlist>
<listitem><para>
DOT files use a plain text format.
The graphs generated using the
<filename>bitbake -g</filename> command are often so
large as to be difficult to read without special
pruning (e.g. with Bitbake's
<filename>-I</filename> option) and processing.
Despite the form and size of the graphs, the
corresponding <filename>.dot</filename> files can still
be possible to read and provide useful information.
</para>
<para>As an example, the
<filename>task-depends.dot</filename> file contains
lines such as the following:
<literallayout class='monospaced'>
"libxslt.do_configure" -> "libxml2.do_populate_sysroot"
</literallayout>
The above example line reveals that the
<link linkend='ref-tasks-configure'><filename>do_configure</filename></link>
task in <filename>libxslt</filename> depends on the
<link linkend='ref-tasks-populate_sysroot'><filename>do_populate_sysroot</filename></link>
task in <filename>libxml2</filename>, which is a normal
<link linkend='var-DEPENDS'><filename>DEPENDS</filename></link>
dependency between the two recipes.
</para></listitem>
<listitem><para>
For an example of how <filename>.dot</filename> files
can be processed, see the
<filename>scripts/contrib/graph-tool</filename> Python
script, which finds and displays paths between graph
nodes.
</para></listitem>
</itemizedlist>
</note>
</para>
<para>
You can use a different method to view dependency information
by using the following command:
<literallayout class='monospaced'>
$ bitbake -g -u taskexp <replaceable>recipename</replaceable>
</literallayout>
This command displays a GUI window from which you can view
build-time and runtime dependencies for the recipes involved in
building <replaceable>recipename</replaceable>.
</para>
</section>
<section id='usingpoky-viewing-task-variable-dependencies'>
<title>Viewing Task Variable Dependencies</title>
<para>
As mentioned in the
"<ulink url='&YOCTO_DOCS_BB_URL;#checksums'>Checksums (Signatures)</ulink>"
section of the BitBake User Manual, BitBake tries to automatically
determine what variables a task depends on so that it can rerun
the task if any values of the variables change.
This determination is usually reliable.
However, if you do things like construct variable names at runtime,
then you might have to manually declare dependencies on those
variables using <filename>vardeps</filename> as described in the
"<ulink url='&YOCTO_DOCS_BB_URL;#variable-flags'>Variable Flags</ulink>"
section of the BitBake User Manual.
</para>
<para>
If you are unsure whether a variable dependency is being picked up
automatically for a given task, you can list the variable
dependencies BitBake has determined by doing the following:
<orderedlist>
<listitem><para>
Build the recipe containing the task:
<literallayout class='monospaced'>
$ bitbake <replaceable>recipename</replaceable>
</literallayout>
</para></listitem>
<listitem><para>
Inside the
<link linkend='var-STAMPS_DIR'><filename>STAMPS_DIR</filename></link>
directory, find the signature data
(<filename>sigdata</filename>) file that corresponds to the
task.
The <filename>sigdata</filename> files contain a pickled
Python database of all the metadata that went into creating
the input checksum for the task.
As an example, for the
<link linkend='ref-tasks-fetch'><filename>do_fetch</filename></link>
task of the <filename>db</filename> recipe, the
<filename>sigdata</filename> file might be found in the
following location:
<literallayout class='monospaced'>
${BUILDDIR}/tmp/stamps/i586-poky-linux/db/6.0.30-r1.do_fetch.sigdata.7c048c18222b16ff0bcee2000ef648b1
</literallayout>
For tasks that are accelerated through the shared state
(<link linkend='shared-state-cache'>sstate</link>)
cache, an additional <filename>siginfo</filename> file is
written into
<link linkend='var-SSTATE_DIR'><filename>SSTATE_DIR</filename></link>
along with the cached task output.
The <filename>siginfo</filename> files contain exactly the
same information as <filename>sigdata</filename> files.
</para></listitem>
<listitem><para>
Run <filename>bitbake-dumpsig</filename> on the
<filename>sigdata</filename> or
<filename>siginfo</filename> file.
Here is an example:
<literallayout class='monospaced'>
$ bitbake-dumpsig ${BUILDDIR}/tmp/stamps/i586-poky-linux/db/6.0.30-r1.do_fetch.sigdata.7c048c18222b16ff0bcee2000ef648b1
</literallayout>
In the output of the above command, you will find a line
like the following, which lists all the (inferred) variable
dependencies for the task.
This list also includes indirect dependencies from
variables depending on other variables, recursively.
<literallayout class='monospaced'>
Task dependencies: ['PV', 'SRCREV', 'SRC_URI', 'SRC_URI[md5sum]', 'SRC_URI[sha256sum]', 'base_do_fetch']
</literallayout>
<note>
Functions (e.g. <filename>base_do_fetch</filename>)
also count as variable dependencies.
These functions in turn depend on the variables they
reference.
</note>
The output of <filename>bitbake-dumpsig</filename> also includes
the value each variable had, a list of dependencies for each
variable, and
<ulink url='&YOCTO_DOCS_BB_URL;#var-BB_HASHBASE_WHITELIST'><filename>BB_HASHBASE_WHITELIST</filename></ulink>
information.
</para></listitem>
</orderedlist>
</para>
<para>
There is also a <filename>bitbake-diffsigs</filename> command for
comparing two <filename>siginfo</filename> or
<filename>sigdata</filename> files.
This command can be helpful when trying to figure out what changed
between two versions of a task.
If you call <filename>bitbake-diffsigs</filename> with just one
file, the command behaves like
<filename>bitbake-dumpsig</filename>.
</para>
<para>
You can also use BitBake to dump out the signature construction
information without executing tasks by using either of the
following BitBake command-line options:
<literallayout class='monospaced'>
&dash;&dash;dump-signatures=<replaceable>SIGNATURE_HANDLER</replaceable>
-S <replaceable>SIGNATURE_HANDLER</replaceable>
</literallayout>
<note>
Two common values for
<replaceable>SIGNATURE_HANDLER</replaceable> are "none" and
"printdiff", which dump only the signature or compare the
dumped signature with the cached one, respectively.
</note>
Using BitBake with either of these options causes BitBake to dump
out <filename>sigdata</filename> files in the
<filename>stamps</filename> directory for every task it would have
executed instead of building the specified target package.
</para>
</section>
<section id='usingpoky-debugging-taskrunning'>
<title>Running Specific Tasks</title>
<para>
Any given recipe consists of a set of tasks.
The standard BitBake behavior in most cases is:
<filename>do_fetch</filename>,
<filename>do_unpack</filename>,
<filename>do_patch</filename>, <filename>do_configure</filename>,
<filename>do_compile</filename>, <filename>do_install</filename>,
<filename>do_package</filename>,
<filename>do_package_write_*</filename>, and
<filename>do_build</filename>.
The default task is <filename>do_build</filename> and any tasks
on which it depends build first.
Some tasks, such as <filename>do_devshell</filename>, are not part
of the default build chain.
If you wish to run a task that is not part of the default build
chain, you can use the <filename>-c</filename> option in BitBake.
Here is an example:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop -c devshell
</literallayout>
</para>
<para>
The <filename>-c</filename> option respects task dependencies,
which means that all other tasks (including tasks from other
recipes) that the specified task depends on will be run before the
task.
Even when you manually specify a task to run with
<filename>-c</filename>, BitBake will only run the task if it
considers it "out of date".
See the
"<link linkend='stamp-files-and-the-rerunning-of-tasks'>Stamp Files and the Rerunning of Tasks</link>"
section for how BitBake determines whether a task is "out of date".
</para>
<para>
If you want to force an up-to-date task to be rerun (e.g.
because you made manual modifications to the recipe's
<link linkend='var-WORKDIR'><filename>WORKDIR</filename></link>
that you want to try out), then you can use the
<filename>-f</filename> option.
<note>
The reason <filename>-f</filename> is never required when
running the
<link linkend='ref-tasks-devshell'><filename>do_devshell</filename></link>
task is because the
<filename>[</filename><ulink url='&YOCTO_DOCS_BB_URL;#variable-flags'><filename>nostamp</filename></ulink><filename>]</filename>
variable flag is already set for the task.
</note>
The following example shows one way you can use the
<filename>-f</filename> option:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop
.
.
make some changes to the source code in the work directory
.
.
$ bitbake matchbox-desktop -c compile -f
$ bitbake matchbox-desktop
</literallayout>
</para>
<para>
This sequence first builds and then recompiles
<filename>matchbox-desktop</filename>.
The last command reruns all tasks (basically the packaging tasks)
after the compile.
BitBake recognizes that the <filename>do_compile</filename>
task was rerun and therefore understands that the other tasks
also need to be run again.
</para>
<para>
Another, shorter way to rerun a task and all
<link linkend='normal-recipe-build-tasks'>normal recipe build tasks</link>
that depend on it is to use the <filename>-C</filename>
option.
<note>
This option is upper-cased and is separate from the
<filename>-c</filename> option, which is lower-cased.
</note>
Using this option invalidates the given task and then runs the
<link linkend='ref-tasks-build'><filename>do_build</filename></link>
task, which is the default task if no task is given, and the
tasks on which it depends.
You could replace the final two commands in the previous example
with the following single command:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop -C compile
</literallayout>
Internally, the <filename>-f</filename> and
<filename>-C</filename> options work by tainting (modifying) the
input checksum of the specified task.
This tainting indirectly causes the task and its
dependent tasks to be rerun through the normal task dependency
mechanisms.
<note>
BitBake explicitly keeps track of which tasks have been
tainted in this fashion, and will print warnings such as the
following for builds involving such tasks:
<literallayout class='monospaced'>
WARNING: /home/ulf/poky/meta/recipes-sato/matchbox-desktop/matchbox-desktop_2.1.bb.do_compile is tainted from a forced run
</literallayout>
The purpose of the warning is to let you know that the work
directory and build output might not be in the clean state they
would be in for a "normal" build, depending on what actions
you took.
To get rid of such warnings, you can remove the work directory
and rebuild the recipe, as follows:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop -c clean
$ bitbake matchbox-desktop
</literallayout>
</note>
</para>
<para>
You can view a list of tasks in a given package by running the
<filename>do_listtasks</filename> task as follows:
<literallayout class='monospaced'>
$ bitbake matchbox-desktop -c listtasks
</literallayout>
The results appear as output to the console and are also in the
file <filename>${WORKDIR}/temp/log.do_listtasks</filename>.
</para>
</section>
<section id='usingpoky-debugging-bitbake'>
<title>General BitBake Problems</title>
<para>
You can see debug output from BitBake by using the <filename>-D</filename> option.
The debug output gives more information about what BitBake
is doing and the reason behind it.
Each <filename>-D</filename> option you use increases the logging level.
The most common usage is <filename>-DDD</filename>.
</para>
<para>
The output from <filename>bitbake -DDD -v</filename> <replaceable>targetname</replaceable> can reveal why
BitBake chose a certain version of a package or why BitBake
picked a certain provider.
This command could also help you in a situation where you think BitBake did something
unexpected.
</para>
</section>
<section id='development-host-system-issues'>
<title>Development Host System Issues</title>
<para>
Sometimes issues on the host development system can cause your
build to fail.
Following are known, host-specific problems.
Be sure to always consult the
<ulink url='&YOCTO_RELEASE_NOTES;'>Release Notes</ulink>
for a look at all release-related issues.
<itemizedlist>
<listitem><para><emphasis><filename>glibc-initial</filename> fails to build</emphasis>:
If your development host system has the unpatched
<filename>GNU Make 3.82</filename>,
the
<link linkend='ref-tasks-install'><filename>do_install</filename></link>
task fails for <filename>glibc-initial</filename> during
the build.</para>
<para>Typically, every distribution that ships
<filename>GNU Make 3.82</filename> as
the default already has the patched version.
However, some distributions, such as Debian, have
<filename>GNU Make 3.82</filename> as an option, which
is unpatched.
You will see this error on these types of distributions.
Switch to <filename>GNU Make 3.81</filename> or patch
your <filename>make</filename> to solve the problem.
</para></listitem>
</itemizedlist>
</para>
</section>
<section id='usingpoky-debugging-buildfile'>
<title>Building with No Dependencies</title>
<para>
To build a specific recipe (<filename>.bb</filename> file),
you can use the following command form:
<literallayout class='monospaced'>
$ bitbake -b <replaceable>somepath</replaceable>/<replaceable>somerecipe</replaceable>.bb
</literallayout>
This command form does not check for dependencies.
Consequently, you should use it
only when you know existing dependencies have been met.
<note>
You can also specify fragments of the filename.
In this case, BitBake checks for a unique match.
</note>
</para>
</section>
<section id='recipe-logging-mechanisms'>
<title>Recipe Logging Mechanisms</title>
<para>
The Yocto Project provides several logging functions for producing
debugging output and reporting errors and warnings.
For Python functions, the following logging functions exist.
All of these functions log to
<filename>${T}/log.do_</filename><replaceable>task</replaceable>,
and can also log to standard output (stdout) with the right
settings:
<itemizedlist>
<listitem><para>
<filename>bb.plain(</filename><replaceable>msg</replaceable><filename>)</filename>:
Writes <replaceable>msg</replaceable> as is to the log while
also logging to stdout.
</para></listitem>
<listitem><para>
<filename>bb.note(</filename><replaceable>msg</replaceable><filename>)</filename>:
Writes "NOTE: <replaceable>msg</replaceable>" to the log.
Also logs to stdout if BitBake is called with "-v".
</para></listitem>
<listitem><para>
<filename>bb.debug(</filename><replaceable>level</replaceable><filename>,&nbsp;</filename><replaceable>msg</replaceable><filename>)</filename>:
Writes "DEBUG: <replaceable>msg</replaceable>" to the log.
Also logs to stdout if the log level is greater than or
equal to <replaceable>level</replaceable>.
See the
"<ulink url='&YOCTO_DOCS_BB_URL;#usage-and-syntax'>-D</ulink>"
option in the BitBake User Manual for more information.
</para></listitem>
<listitem><para>
<filename>bb.warn(</filename><replaceable>msg</replaceable><filename>)</filename>:
Writes "WARNING: <replaceable>msg</replaceable>" to the log
while also logging to stdout.
</para></listitem>
<listitem><para>
<filename>bb.error(</filename><replaceable>msg</replaceable><filename>)</filename>:
Writes "ERROR: <replaceable>msg</replaceable>" to the log
while also logging to stdout.
<note>
Calling this function does not cause the task to fail.
</note>
</para></listitem>
<listitem><para>
<filename>bb.fatal(</filename><replaceable>msg</replaceable><filename>)</filename>:
This logging function is similar to
<filename>bb.error(</filename><replaceable>msg</replaceable><filename>)</filename>
but also causes the calling task to fail.
<note>
<filename>bb.fatal()</filename> raises an exception,
which means you do not need to put a "return"
statement after the function.
</note>
</para></listitem>
</itemizedlist>
</para>
<para>
The same logging functions are also available in shell functions,
under the names
<filename>bbplain</filename>, <filename>bbnote</filename>,
<filename>bbdebug</filename>, <filename>bbwarn</filename>,
<filename>bberror</filename>, and <filename>bbfatal</filename>.
The
<link linkend='ref-classes-logging'><filename>logging</filename></link>
class implements these functions.
See that class in the
<filename>meta/classes</filename> folder of the
<link linkend='source-directory'>Source Directory</link>
for information.
</para>
<section id='logging-with-python'>
<title>Logging With Python</title>
<para>
When creating recipes using Python and inserting code that handles build logs,
keep in mind the goal is to have informative logs while keeping the console as
"silent" as possible.
Also, if you want status messages in the log, use the "debug" loglevel.
</para>
<para>
Following is an example written in Python.
The code handles logging for a function that determines the
number of tasks needed to be run.
See the
"<link linkend='ref-tasks-listtasks'><filename>do_listtasks</filename></link>"
section for additional information:
<literallayout class='monospaced'>
python do_listtasks() {
bb.debug(2, "Starting to figure out the task list")
if noteworthy_condition:
bb.note("There are 47 tasks to run")
bb.debug(2, "Got to point xyz")
if warning_trigger:
bb.warn("Detected warning_trigger, this might be a problem later.")
if recoverable_error:
bb.error("Hit recoverable_error, you really need to fix this!")
if fatal_error:
bb.fatal("fatal_error detected, unable to print the task list")
bb.plain("The tasks present are abc")
bb.debug(2, "Finished figuring out the tasklist")
}
</literallayout>
</para>
</section>
<section id='logging-with-bash'>
<title>Logging With Bash</title>
<para>
When creating recipes using Bash and inserting code that handles build
logs, you have the same goals - informative with minimal console output.
The syntax you use for recipes written in Bash is similar to that of
recipes written in Python described in the previous section.
</para>
<para>
Following is an example written in Bash.
The code logs the progress of the <filename>do_my_function</filename> function.
<literallayout class='monospaced'>
do_my_function() {
bbdebug 2 "Running do_my_function"
if [ exceptional_condition ]; then
bbnote "Hit exceptional_condition"
fi
bbdebug 2 "Got to point xyz"
if [ warning_trigger ]; then
bbwarn "Detected warning_trigger, this might cause a problem later."
fi
if [ recoverable_error ]; then
bberror "Hit recoverable_error, correcting"
fi
if [ fatal_error ]; then
bbfatal "fatal_error detected"
fi
bbdebug 2 "Completed do_my_function"
}
</literallayout>
</para>
</section>
</section>
<section id='usingpoky-debugging-others'>
<title>Other Tips</title>
<para>
Here are some other tips that you might find useful:
<itemizedlist>
<listitem><para>
When adding new packages, it is worth watching for
undesirable items making their way into compiler command
lines.
For example, you do not want references to local system
files like
<filename>/usr/lib/</filename> or
<filename>/usr/include/</filename>.
</para></listitem>
<listitem><para>
If you want to remove the <filename>psplash</filename>
boot splashscreen,
add <filename>psplash=false</filename> to the kernel
command line.
Doing so prevents <filename>psplash</filename> from loading
and thus allows you to see the console.
It is also possible to switch out of the splashscreen by
switching the virtual console (e.g. Fn+Left or Fn+Right
on a Zaurus).
</para></listitem>
<listitem><para>
Removing
<link linkend='var-TMPDIR'><filename>TMPDIR</filename></link>
(usually <filename>tmp/</filename>, within the
<link linkend='build-directory'>Build Directory</link>)
can often fix temporary build issues.
Removing <filename>TMPDIR</filename> is usually a
relatively cheap operation, because task output will be
cached in
<link linkend='var-SSTATE_DIR'><filename>SSTATE_DIR</filename></link>
(usually <filename>sstate-cache/</filename>, which is
also in the Build Directory).
<note>
Removing <filename>TMPDIR</filename> might be a
workaround rather than a fix.
Consequently, trying to determine the underlying cause
of an issue before removing the directory is a good
idea.
</note>
</para></listitem>
<listitem><para>
Understanding how a feature is used in practice within
existing recipes can be very helpful.
It is recommended that you configure some method that
allows you to quickly search through files.</para>
<para>Using GNU Grep, you can use the following shell
function to recursively search through common
recipe-related files, skipping binary files,
<filename>.git</filename> directories, and the
Build Directory (assuming its name starts with
"build"):
<literallayout class='monospaced'>
g() {
grep -Ir \
--exclude-dir=.git \
--exclude-dir='build*' \
--include='*.bb*' \
--include='*.inc*' \
--include='*.conf*' \
--include='*.py*' \
"$@"
}
</literallayout>
Following are some usage examples:
<literallayout class='monospaced'>
$ g FOO # Search recursively for "FOO"
$ g -i foo # Search recursively for "foo", ignoring case
$ g -w FOO # Search recursively for "FOO" as a word, ignoring e.g. "FOOBAR"
</literallayout>
If figuring out how some feature works requires a lot of
searching, it might indicate that the documentation should
be extended or improved.
In such cases, consider filing a documentation bug using
the Yocto Project implementation of
<ulink url='https://bugzilla.yoctoproject.org/'>Bugzilla</ulink>.
For general information on how to submit a bug against
the Yocto Project, see the Yocto Project Bugzilla
<ulink url='&YOCTO_WIKI_URL;/wiki/Bugzilla_Configuration_and_Bug_Tracking'>wiki page</ulink>"
or the
<ulink url='&YOCTO_DOCS_DEV_URL;#submitting-a-defect-against-the-yocto-project'>Submitting a Defect Against the Yocto Project</ulink>"
section, which is in the Yocto Project Development Tasks
Manual.
<note>
The manuals might not be the right place to document
variables that are purely internal and have a limited
scope (e.g. internal variables used to implement a
single <filename>.bbclass</filename> file).
</note>
</para></listitem>
</itemizedlist>
</para>
</section>
</section>
<section id='ref-quick-emulator-qemu'>
<title>Quick EMUlator (QEMU)</title>
<para>
The Yocto Project uses an implementation of the Quick EMUlator (QEMU)
Open Source project as part of the Yocto Project development "tool
set".
</para>
<para>
Within the context of the Yocto Project, QEMU is an
emulator and virtualization machine that allows you to run a complete
image you have built using the Yocto Project as just another task
on your build system.
QEMU is useful for running and testing images and applications on
supported Yocto Project architectures without having actual hardware.
Among other things, the Yocto Project uses QEMU to run automated
Quality Assurance (QA) tests on final images shipped with each
release.
<note>
This implementation is not the same as QEMU in general.
</note>
This section provides a brief reference for the Yocto Project
implementation of QEMU.
</para>
<para>
For official information and documentation on QEMU in general, see the
following references:
<itemizedlist>
<listitem><para>
<emphasis><ulink url='http://wiki.qemu.org/Main_Page'>QEMU Website</ulink>:</emphasis>
The official website for the QEMU Open Source project.
</para></listitem>
<listitem><para>
<emphasis><ulink url='http://wiki.qemu.org/Manual'>Documentation</ulink>:</emphasis>
The QEMU user manual.
</para></listitem>
</itemizedlist>
</para>
<para>
For information on how to use the Yocto Project implementation of
QEMU, see the
"<ulink url='&YOCTO_DOCS_DEV_URL;#dev-manual-qemu'>Using the Quick EMUlator (QEMU)</ulink>"
chapter in the Yocto Project Development Tasks Manual.
</para>
<section id='qemu-availability'>
<title>QEMU Availability</title>
<para>
QEMU is made available with the Yocto Project a number of ways.
One method is to install a Software Development Kit (SDK).
For more information on how to make sure you have
QEMU available, see
"<ulink url='&YOCTO_DOCS_SDK_URL;#the-qemu-emulator'>The QEMU Emulator</ulink>"
section in the Yocto Project Application Development and the
Extensible Software Development Kit (eSDK) manual.
</para>
</section>
<section id='qemu-performance'>
<title>QEMU Performance</title>
<para>
Using QEMU to emulate your hardware can result in speed issues
depending on the target and host architecture mix.
For example, using the <filename>qemux86</filename> image in the
emulator on an Intel-based 32-bit (x86) host machine is fast
because the target and host architectures match.
On the other hand, using the <filename>qemuarm</filename> image
on the same Intel-based host can be slower.
But, you still achieve faithful emulation of ARM-specific issues.
</para>
<para>
To speed things up, the QEMU images support using
<filename>distcc</filename> to call a cross-compiler outside the
emulated system.
If you used <filename>runqemu</filename> to start QEMU, and the
<filename>distccd</filename> application is present on the host
system, any BitBake cross-compiling toolchain available from the
build system is automatically used from within QEMU simply by
calling <filename>distcc</filename>.
You can accomplish this by defining the cross-compiler variable
(e.g. <filename>export CC="distcc"</filename>).
Alternatively, if you are using a suitable SDK image or the
appropriate stand-alone toolchain is present, the toolchain is
also automatically used.
</para>
<note>
Several mechanisms exist that let you connect to the system
running on the QEMU emulator:
<itemizedlist>
<listitem><para>
QEMU provides a framebuffer interface that makes standard
consoles available.
</para></listitem>
<listitem><para>
Generally, headless embedded devices have a serial port.
If so, you can configure the operating system of the
running image to use that port to run a console.
The connection uses standard IP networking.
</para></listitem>
<listitem><para>
SSH servers exist in some QEMU images.
The <filename>core-image-sato</filename> QEMU image has a
Dropbear secure shell (SSH) server that runs with the root
password disabled.
The <filename>core-image-full-cmdline</filename> and
<filename>core-image-lsb</filename> QEMU images
have OpenSSH instead of Dropbear.
Including these SSH servers allow you to use standard
<filename>ssh</filename> and <filename>scp</filename>
commands.
The <filename>core-image-minimal</filename> QEMU image,
however, contains no SSH server.
</para></listitem>
<listitem><para>
You can use a provided, user-space NFS server to boot
the QEMU session using a local copy of the root
filesystem on the host.
In order to make this connection, you must extract a
root filesystem tarball by using the
<filename>runqemu-extract-sdk</filename> command.
After running the command, you must then point the
<filename>runqemu</filename>
script to the extracted directory instead of a root
filesystem image file.
See the
"<ulink url='&YOCTO_DOCS_DEV_URL;#qemu-running-under-a-network-file-system-nfs-server'>Running Under a Network File System (NFS) Server</ulink>"
section in the Yocto Project Development Tasks Manual for
more information.
</para></listitem>
</itemizedlist>
</note>
</section>
<section id='qemu-command-line-syntax'>
<title>QEMU Command-Line Syntax</title>
<para>
The basic <filename>runqemu</filename> command syntax is as
follows:
<literallayout class='monospaced'>
$ runqemu [<replaceable>option</replaceable> ] [...]
</literallayout>
Based on what you provide on the command line,
<filename>runqemu</filename> does a good job of figuring out what
you are trying to do.
For example, by default, QEMU looks for the most recently built
image according to the timestamp when it needs to look for an
image.
Minimally, through the use of options, you must provide either
a machine name, a virtual machine image
(<filename>*wic.vmdk</filename>), or a kernel image
(<filename>*.bin</filename>).
</para>
<para>
Following is the command-line help output for the
<filename>runqemu</filename> command:
<literallayout class='monospaced'>
$ runqemu --help
Usage: you can run this script with any valid combination
of the following environment variables (in any order):
KERNEL - the kernel image file to use
ROOTFS - the rootfs image file or nfsroot directory to use
MACHINE - the machine name (optional, autodetected from KERNEL filename if unspecified)
Simplified QEMU command-line options can be passed with:
nographic - disable video console
serial - enable a serial console on /dev/ttyS0
slirp - enable user networking, no root privileges is required
kvm - enable KVM when running x86/x86_64 (VT-capable CPU required)
kvm-vhost - enable KVM with vhost when running x86/x86_64 (VT-capable CPU required)
publicvnc - enable a VNC server open to all hosts
audio - enable audio
[*/]ovmf* - OVMF firmware file or base name for booting with UEFI
tcpserial=&lt;port&gt; - specify tcp serial port number
biosdir=&lt;dir&gt; - specify custom bios dir
biosfilename=&lt;filename&gt; - specify bios filename
qemuparams=&lt;xyz&gt; - specify custom parameters to QEMU
bootparams=&lt;xyz&gt; - specify custom kernel parameters during boot
help, -h, --help: print this text
Examples:
runqemu
runqemu qemuarm
runqemu tmp/deploy/images/qemuarm
runqemu tmp/deploy/images/qemux86/&lt;qemuboot.conf&gt;
runqemu qemux86-64 core-image-sato ext4
runqemu qemux86-64 wic-image-minimal wic
runqemu path/to/bzImage-qemux86.bin path/to/nfsrootdir/ serial
runqemu qemux86 iso/hddimg/wic.vmdk/wic.qcow2/wic.vdi/ramfs/cpio.gz...
runqemu qemux86 qemuparams="-m 256"
runqemu qemux86 bootparams="psplash=false"
runqemu path/to/&lt;image&gt;-&lt;machine&gt;.wic
runqemu path/to/&lt;image&gt;-&lt;machine&gt;.wic.vmdk
</literallayout>
</para>
</section>
<section id='runqemu-command-line-options'>
<title><filename>runqemu</filename> Command-Line Options</title>
<para>
Following is a description of <filename>runqemu</filename>
options you can provide on the command line:
<note><title>Tip</title>
If you do provide some "illegal" option combination or perhaps
you do not provide enough in the way of options,
<filename>runqemu</filename> provides appropriate error
messaging to help you correct the problem.
</note>
<itemizedlist>
<listitem><para>
<replaceable>QEMUARCH</replaceable>:
The QEMU machine architecture, which must be "qemuarm",
"qemuarm64", "qemumips", "qemumips64", "qemuppc",
"qemux86", or "qemux86-64".
</para></listitem>
<listitem><para>
<filename><replaceable>VM</replaceable></filename>:
The virtual machine image, which must be a
<filename>.wic.vmdk</filename> file.
Use this option when you want to boot a
<filename>.wic.vmdk</filename> image.
The image filename you provide must contain one of the
following strings: "qemux86-64", "qemux86", "qemuarm",
"qemumips64", "qemumips", "qemuppc", or "qemush4".
</para></listitem>
<listitem><para>
<replaceable>ROOTFS</replaceable>:
A root filesystem that has one of the following
filetype extensions: "ext2", "ext3", "ext4", "jffs2",
"nfs", or "btrfs".
If the filename you provide for this option uses “nfs”, it
must provide an explicit root filesystem path.
</para></listitem>
<listitem><para>
<replaceable>KERNEL</replaceable>:
A kernel image, which is a <filename>.bin</filename> file.
When you provide a <filename>.bin</filename> file,
<filename>runqemu</filename> detects it and assumes the
file is a kernel image.
</para></listitem>
<listitem><para>
<replaceable>MACHINE</replaceable>:
The architecture of the QEMU machine, which must be one
of the following: "qemux86", "qemux86-64", "qemuarm",
"qemuarm64", "qemumips", “qemumips64", or "qemuppc".
The <replaceable>MACHINE</replaceable> and
<replaceable>QEMUARCH</replaceable> options are basically
identical.
If you do not provide a <replaceable>MACHINE</replaceable>
option, <filename>runqemu</filename> tries to determine
it based on other options.
</para></listitem>
<listitem><para>
<filename>ramfs</filename>:
Indicates you are booting an initial RAM disk (initramfs)
image, which means the <filename>FSTYPE</filename> is
<filename>cpio.gz</filename>.
</para></listitem>
<listitem><para>
<filename>iso</filename>:
Indicates you are booting an ISO image, which means the
<filename>FSTYPE</filename> is
<filename>.iso</filename>.
</para></listitem>
<listitem><para>
<filename>nographic</filename>:
Disables the video console, which sets the console to
"ttys0".
</para></listitem>
<listitem><para>
<filename>serial</filename>:
Enables a serial console on
<filename>/dev/ttyS0</filename>.
</para></listitem>
<listitem><para>
<filename>biosdir</filename>:
Establishes a custom directory for BIOS, VGA BIOS and
keymaps.
</para></listitem>
<listitem><para>
<filename>biosfilename</filename>:
Establishes a custom BIOS name.
</para></listitem>
<listitem><para>
<filename>qemuparams=\"<replaceable>xyz</replaceable>\"</filename>:
Specifies custom QEMU parameters.
Use this option to pass options other than the simple
"kvm" and "serial" options.
</para></listitem>
<listitem><para><filename>bootparams=\"<replaceable>xyz</replaceable>\"</filename>:
Specifies custom boot parameters for the kernel.
</para></listitem>
<listitem><para>
<filename>audio</filename>:
Enables audio in QEMU.
The <replaceable>MACHINE</replaceable> option must be
either "qemux86" or "qemux86-64" in order for audio to be
enabled.
Additionally, the <filename>snd_intel8x0</filename>
or <filename>snd_ens1370</filename> driver must be
installed in linux guest.
</para></listitem>
<listitem><para>
<filename>slirp</filename>:
Enables "slirp" networking, which is a different way
of networking that does not need root access
but also is not as easy to use or comprehensive
as the default.
</para></listitem>
<listitem><para id='kvm-cond'>
<filename>kvm</filename>:
Enables KVM when running "qemux86" or "qemux86-64"
QEMU architectures.
For KVM to work, all the following conditions must be met:
<itemizedlist>
<listitem><para>
Your <replaceable>MACHINE</replaceable> must be either
qemux86" or "qemux86-64".
</para></listitem>
<listitem><para>
Your build host has to have the KVM modules
installed, which are
<filename>/dev/kvm</filename>.
</para></listitem>
<listitem><para>
The build host <filename>/dev/kvm</filename>
directory has to be both writable and readable.
</para></listitem>
</itemizedlist>
</para></listitem>
<listitem><para>
<filename>kvm-vhost</filename>:
Enables KVM with VHOST support when running "qemux86"
or "qemux86-64" QEMU architectures.
For KVM with VHOST to work, the following conditions must
be met:
<itemizedlist>
<listitem><para>
<link linkend='kvm-cond'>kvm</link> option
conditions must be met.
</para></listitem>
<listitem><para>
Your build host has to have virtio net device, which
are <filename>/dev/vhost-net</filename>.
</para></listitem>
<listitem><para>
The build host <filename>/dev/vhost-net</filename>
directory has to be either readable or writable
and “slirp-enabled”.
</para></listitem>
</itemizedlist>
</para></listitem>
<listitem><para>
<filename>publicvnc</filename>:
Enables a VNC server open to all hosts.
</para></listitem>
</itemizedlist>
</para>
</section>
</section>
<section id='maintaining-build-output-quality'>
<title>Maintaining Build Output Quality</title>
<para>
Many factors can influence the quality of a build.
For example, if you upgrade a recipe to use a new version of an upstream software
package or you experiment with some new configuration options, subtle changes
can occur that you might not detect until later.
Consider the case where your recipe is using a newer version of an upstream package.
In this case, a new version of a piece of software might introduce an optional
dependency on another library, which is auto-detected.
If that library has already been built when the software is building,
the software will link to the built library and that library will be pulled
into your image along with the new software even if you did not want the
library.
</para>
<para>
The
<link linkend='ref-classes-buildhistory'><filename>buildhistory</filename></link>
class exists to help you maintain
the quality of your build output.
You can use the class to highlight unexpected and possibly unwanted
changes in the build output.
When you enable build history, it records information about the contents of
each package and image and then commits that information to a local Git
repository where you can examine the information.
</para>
<para>
The remainder of this section describes the following:
<itemizedlist>
<listitem><para>How you can enable and disable
build history</para></listitem>
<listitem><para>How to understand what the build history contains
</para></listitem>
<listitem><para>How to limit the information used for build history
</para></listitem>
<listitem><para>How to examine the build history from both a
command-line and web interface</para></listitem>
</itemizedlist>
</para>
<section id='enabling-and-disabling-build-history'>
<title>Enabling and Disabling Build History</title>
<para>
Build history is disabled by default.
To enable it, add the following <filename>INHERIT</filename>
statement and set the
<link linkend='var-BUILDHISTORY_COMMIT'><filename>BUILDHISTORY_COMMIT</filename></link>
variable to "1" at the end of your
<filename>conf/local.conf</filename> file found in the
<link linkend='build-directory'>Build Directory</link>:
<literallayout class='monospaced'>
INHERIT += "buildhistory"
BUILDHISTORY_COMMIT = "1"
</literallayout>
Enabling build history as previously described causes the
OpenEmbedded build system to collect build output information and
commit it as a single commit to a local
<link linkend='git'>Git</link> repository.
<note>
Enabling build history increases your build times slightly,
particularly for images, and increases the amount of disk
space used during the build.
</note>
</para>
<para>
You can disable build history by removing the previous statements
from your <filename>conf/local.conf</filename> file.
</para>
</section>
<section id='understanding-what-the-build-history-contains'>
<title>Understanding What the Build History Contains</title>
<para>
Build history information is kept in
<filename>${</filename><link linkend='var-TOPDIR'><filename>TOPDIR</filename></link><filename>}/buildhistory</filename>
in the Build Directory as defined by the
<link linkend='var-BUILDHISTORY_DIR'><filename>BUILDHISTORY_DIR</filename></link>
variable.
The following is an example abbreviated listing:
<imagedata fileref="figures/buildhistory.png" align="center" width="6in" depth="4in" />
</para>
<para>
At the top level, there is a <filename>metadata-revs</filename> file
that lists the revisions of the repositories for the layers enabled
when the build was produced.
The rest of the data splits into separate
<filename>packages</filename>, <filename>images</filename> and
<filename>sdk</filename> directories, the contents of which are
described below.
</para>
<section id='build-history-package-information'>
<title>Build History Package Information</title>
<para>
The history for each package contains a text file that has
name-value pairs with information about the package.
For example, <filename>buildhistory/packages/i586-poky-linux/busybox/busybox/latest</filename>
contains the following:
<literallayout class='monospaced'>
PV = 1.22.1
PR = r32
RPROVIDES =
RDEPENDS = glibc (>= 2.20) update-alternatives-opkg
RRECOMMENDS = busybox-syslog busybox-udhcpc update-rc.d
PKGSIZE = 540168
FILES = /usr/bin/* /usr/sbin/* /usr/lib/busybox/* /usr/lib/lib*.so.* \
/etc /com /var /bin/* /sbin/* /lib/*.so.* /lib/udev/rules.d \
/usr/lib/udev/rules.d /usr/share/busybox /usr/lib/busybox/* \
/usr/share/pixmaps /usr/share/applications /usr/share/idl \
/usr/share/omf /usr/share/sounds /usr/lib/bonobo/servers
FILELIST = /bin/busybox /bin/busybox.nosuid /bin/busybox.suid /bin/sh \
/etc/busybox.links.nosuid /etc/busybox.links.suid
</literallayout>
Most of these name-value pairs correspond to variables used
to produce the package.
The exceptions are <filename>FILELIST</filename>, which is the
actual list of files in the package, and
<filename>PKGSIZE</filename>, which is the total size of files
in the package in bytes.
</para>
<para>
There is also a file corresponding to the recipe from which the
package came (e.g.
<filename>buildhistory/packages/i586-poky-linux/busybox/latest</filename>):
<literallayout class='monospaced'>
PV = 1.22.1
PR = r32
DEPENDS = initscripts kern-tools-native update-rc.d-native \
virtual/i586-poky-linux-compilerlibs virtual/i586-poky-linux-gcc \
virtual/libc virtual/update-alternatives
PACKAGES = busybox-ptest busybox-httpd busybox-udhcpd busybox-udhcpc \
busybox-syslog busybox-mdev busybox-hwclock busybox-dbg \
busybox-staticdev busybox-dev busybox-doc busybox-locale busybox
</literallayout>
</para>
<para>
Finally, for those recipes fetched from a version control
system (e.g., Git), a file exists that lists source revisions
that are specified in the recipe and lists the actual revisions
used during the build.
Listed and actual revisions might differ when
<link linkend='var-SRCREV'><filename>SRCREV</filename></link>
is set to
<filename>${<link linkend='var-AUTOREV'>AUTOREV</link>}</filename>.
Here is an example assuming
<filename>buildhistory/packages/qemux86-poky-linux/linux-yocto/latest_srcrev</filename>):
<literallayout class='monospaced'>
# SRCREV_machine = "38cd560d5022ed2dbd1ab0dca9642e47c98a0aa1"
SRCREV_machine = "38cd560d5022ed2dbd1ab0dca9642e47c98a0aa1"
# SRCREV_meta = "a227f20eff056e511d504b2e490f3774ab260d6f"
SRCREV_meta = "a227f20eff056e511d504b2e490f3774ab260d6f"
</literallayout>
You can use the <filename>buildhistory-collect-srcrevs</filename>
command with the <filename>-a</filename> option to
collect the stored <filename>SRCREV</filename> values
from build history and report them in a format suitable for
use in global configuration (e.g.,
<filename>local.conf</filename> or a distro include file) to
override floating <filename>AUTOREV</filename> values to a
fixed set of revisions.
Here is some example output from this command:
<literallayout class='monospaced'>
$ buildhistory-collect-srcrevs -a
# i586-poky-linux
SRCREV_pn-glibc = "b8079dd0d360648e4e8de48656c5c38972621072"
SRCREV_pn-glibc-initial = "b8079dd0d360648e4e8de48656c5c38972621072"
SRCREV_pn-opkg-utils = "53274f087565fd45d8452c5367997ba6a682a37a"
SRCREV_pn-kmod = "fd56638aed3fe147015bfa10ed4a5f7491303cb4"
# x86_64-linux
SRCREV_pn-gtk-doc-stub-native = "1dea266593edb766d6d898c79451ef193eb17cfa"
SRCREV_pn-dtc-native = "65cc4d2748a2c2e6f27f1cf39e07a5dbabd80ebf"
SRCREV_pn-update-rc.d-native = "eca680ddf28d024954895f59a241a622dd575c11"
SRCREV_glibc_pn-cross-localedef-native = "b8079dd0d360648e4e8de48656c5c38972621072"
SRCREV_localedef_pn-cross-localedef-native = "c833367348d39dad7ba018990bfdaffaec8e9ed3"
SRCREV_pn-prelink-native = "faa069deec99bf61418d0bab831c83d7c1b797ca"
SRCREV_pn-opkg-utils-native = "53274f087565fd45d8452c5367997ba6a682a37a"
SRCREV_pn-kern-tools-native = "23345b8846fe4bd167efdf1bd8a1224b2ba9a5ff"
SRCREV_pn-kmod-native = "fd56638aed3fe147015bfa10ed4a5f7491303cb4"
# qemux86-poky-linux
SRCREV_machine_pn-linux-yocto = "38cd560d5022ed2dbd1ab0dca9642e47c98a0aa1"
SRCREV_meta_pn-linux-yocto = "a227f20eff056e511d504b2e490f3774ab260d6f"
# all-poky-linux
SRCREV_pn-update-rc.d = "eca680ddf28d024954895f59a241a622dd575c11"
</literallayout>
<note>
Here are some notes on using the
<filename>buildhistory-collect-srcrevs</filename> command:
<itemizedlist>
<listitem><para>By default, only values where the
<filename>SRCREV</filename> was
not hardcoded (usually when <filename>AUTOREV</filename>
was used) are reported.
Use the <filename>-a</filename> option to see all
<filename>SRCREV</filename> values.
</para></listitem>
<listitem><para>The output statements might not have any effect
if overrides are applied elsewhere in the build system
configuration.
Use the <filename>-f</filename> option to add the
<filename>forcevariable</filename> override to each output line
if you need to work around this restriction.
</para></listitem>
<listitem><para>The script does apply special handling when
building for multiple machines.
However, the script does place a
comment before each set of values that specifies
which triplet to which they belong as shown above
(e.g., <filename>i586-poky-linux</filename>).
</para></listitem>
</itemizedlist>
</note>
</para>
</section>
<section id='build-history-image-information'>
<title>Build History Image Information</title>
<para>
The files produced for each image are as follows:
<itemizedlist>
<listitem><para><filename>image-files:</filename>
A directory containing selected files from the root
filesystem.
The files are defined by
<link linkend='var-BUILDHISTORY_IMAGE_FILES'><filename>BUILDHISTORY_IMAGE_FILES</filename></link>.
</para></listitem>
<listitem><para><filename>build-id.txt:</filename>
Human-readable information about the build configuration
and metadata source revisions.
This file contains the full build header as printed
by BitBake.</para></listitem>
<listitem><para><filename>*.dot:</filename>
Dependency graphs for the image that are
compatible with <filename>graphviz</filename>.
</para></listitem>
<listitem><para><filename>files-in-image.txt:</filename>
A list of files in the image with permissions,
owner, group, size, and symlink information.
</para></listitem>
<listitem><para><filename>image-info.txt:</filename>
A text file containing name-value pairs with information
about the image.
See the following listing example for more information.
</para></listitem>
<listitem><para><filename>installed-package-names.txt:</filename>
A list of installed packages by name only.</para></listitem>
<listitem><para><filename>installed-package-sizes.txt:</filename>
A list of installed packages ordered by size.
</para></listitem>
<listitem><para><filename>installed-packages.txt:</filename>
A list of installed packages with full package
filenames.</para></listitem>
</itemizedlist>
<note>
Installed package information is able to be gathered and
produced even if package management is disabled for the final
image.
</note>
</para>
<para>
Here is an example of <filename>image-info.txt</filename>:
<literallayout class='monospaced'>
DISTRO = poky
DISTRO_VERSION = 1.7
USER_CLASSES = buildstats image-mklibs image-prelink
IMAGE_CLASSES = image_types
IMAGE_FEATURES = debug-tweaks
IMAGE_LINGUAS =
IMAGE_INSTALL = packagegroup-core-boot run-postinsts
BAD_RECOMMENDATIONS =
NO_RECOMMENDATIONS =
PACKAGE_EXCLUDE =
ROOTFS_POSTPROCESS_COMMAND = write_package_manifest; license_create_manifest; \
write_image_manifest ; buildhistory_list_installed_image ; \
buildhistory_get_image_installed ; ssh_allow_empty_password; \
postinst_enable_logging; rootfs_update_timestamp ; ssh_disable_dns_lookup ;
IMAGE_POSTPROCESS_COMMAND = buildhistory_get_imageinfo ;
IMAGESIZE = 6900
</literallayout>
Other than <filename>IMAGESIZE</filename>, which is the
total size of the files in the image in Kbytes, the
name-value pairs are variables that may have influenced the
content of the image.
This information is often useful when you are trying to determine
why a change in the package or file listings has occurred.
</para>
</section>
<section id='using-build-history-to-gather-image-information-only'>
<title>Using Build History to Gather Image Information Only</title>
<para>
As you can see, build history produces image information,
including dependency graphs, so you can see why something
was pulled into the image.
If you are just interested in this information and not
interested in collecting specific package or SDK information,
you can enable writing only image information without
any history by adding the following to your
<filename>conf/local.conf</filename> file found in the
<link linkend='build-directory'>Build Directory</link>:
<literallayout class='monospaced'>
INHERIT += "buildhistory"
BUILDHISTORY_COMMIT = "0"
BUILDHISTORY_FEATURES = "image"
</literallayout>
Here, you set the
<link linkend='var-BUILDHISTORY_FEATURES'><filename>BUILDHISTORY_FEATURES</filename></link>
variable to use the image feature only.
</para>
</section>
<section id='build-history-sdk-information'>
<title>Build History SDK Information</title>
<para>
Build history collects similar information on the contents
of SDKs
(e.g. <filename>bitbake -c populate_sdk imagename</filename>)
as compared to information it collects for images.
Furthermore, this information differs depending on whether an
extensible or standard SDK is being produced.
</para>
<para>
The following list shows the files produced for SDKs:
<itemizedlist>
<listitem><para><filename>files-in-sdk.txt:</filename>
A list of files in the SDK with permissions,
owner, group, size, and symlink information.
This list includes both the host and target parts
of the SDK.
</para></listitem>
<listitem><para><filename>sdk-info.txt:</filename>
A text file containing name-value pairs with information
about the SDK.
See the following listing example for more information.
</para></listitem>
<listitem><para><filename>sstate-task-sizes.txt:</filename>
A text file containing name-value pairs with information
about task group sizes
(e.g. <filename>do_populate_sysroot</filename> tasks
have a total size).
The <filename>sstate-task-sizes.txt</filename> file
exists only when an extensible SDK is created.
</para></listitem>
<listitem><para><filename>sstate-package-sizes.txt:</filename>
A text file containing name-value pairs with information
for the shared-state packages and sizes in the SDK.
The <filename>sstate-package-sizes.txt</filename> file
exists only when an extensible SDK is created.
</para></listitem>
<listitem><para><filename>sdk-files:</filename>
A folder that contains copies of the files mentioned in
<filename>BUILDHISTORY_SDK_FILES</filename> if the
files are present in the output.
Additionally, the default value of
<filename>BUILDHISTORY_SDK_FILES</filename> is specific
to the extensible SDK although you can set it
differently if you would like to pull in specific files
from the standard SDK.</para>
<para>The default files are
<filename>conf/local.conf</filename>,
<filename>conf/bblayers.conf</filename>,
<filename>conf/auto.conf</filename>,
<filename>conf/locked-sigs.inc</filename>, and
<filename>conf/devtool.conf</filename>.
Thus, for an extensible SDK, these files get copied
into the <filename>sdk-files</filename> directory.
</para></listitem>
<listitem><para>The following information appears under
each of the <filename>host</filename>
and <filename>target</filename> directories
for the portions of the SDK that run on the host and
on the target, respectively:
<note>
The following files for the most part are empty
when producing an extensible SDK because this
type of SDK is not constructed from packages as is
the standard SDK.
</note>
<itemizedlist>
<listitem><para><filename>depends.dot:</filename>
Dependency graph for the SDK that is
compatible with <filename>graphviz</filename>.
</para></listitem>
<listitem><para><filename>installed-package-names.txt:</filename>
A list of installed packages by name only.
</para></listitem>
<listitem><para><filename>installed-package-sizes.txt:</filename>
A list of installed packages ordered by size.
</para></listitem>
<listitem><para><filename>installed-packages.txt:</filename>
A list of installed packages with full package
filenames.</para></listitem>
</itemizedlist>
</para></listitem>
</itemizedlist>
</para>
<para>
Here is an example of <filename>sdk-info.txt</filename>:
<literallayout class='monospaced'>
DISTRO = poky
DISTRO_VERSION = 1.3+snapshot-20130327
SDK_NAME = poky-glibc-i686-arm
SDK_VERSION = 1.3+snapshot
SDKMACHINE =
SDKIMAGE_FEATURES = dev-pkgs dbg-pkgs
BAD_RECOMMENDATIONS =
SDKSIZE = 352712
</literallayout>
Other than <filename>SDKSIZE</filename>, which is the
total size of the files in the SDK in Kbytes, the
name-value pairs are variables that might have influenced the
content of the SDK.
This information is often useful when you are trying to
determine why a change in the package or file listings
has occurred.
</para>
</section>
<section id='examining-build-history-information'>
<title>Examining Build History Information</title>
<para>
You can examine build history output from the command line or
from a web interface.
</para>
<para>
To see any changes that have occurred (assuming you have
<link linkend='var-BUILDHISTORY_COMMIT'><filename>BUILDHISTORY_COMMIT = "1"</filename></link>),
you can simply
use any Git command that allows you to view the history of
a repository.
Here is one method:
<literallayout class='monospaced'>
$ git log -p
</literallayout>
You need to realize, however, that this method does show
changes that are not significant (e.g. a package's size
changing by a few bytes).
</para>
<para>
A command-line tool called <filename>buildhistory-diff</filename>
does exist, though, that queries the Git repository and prints just
the differences that might be significant in human-readable form.
Here is an example:
<literallayout class='monospaced'>
$ ~/poky/poky/scripts/buildhistory-diff . HEAD^
Changes to images/qemux86_64/glibc/core-image-minimal (files-in-image.txt):
/etc/anotherpkg.conf was added
/sbin/anotherpkg was added
* (installed-package-names.txt):
* anotherpkg was added
Changes to images/qemux86_64/glibc/core-image-minimal (installed-package-names.txt):
anotherpkg was added
packages/qemux86_64-poky-linux/v86d: PACKAGES: added "v86d-extras"
* PR changed from "r0" to "r1"
* PV changed from "0.1.10" to "0.1.12"
packages/qemux86_64-poky-linux/v86d/v86d: PKGSIZE changed from 110579 to 144381 (+30%)
* PR changed from "r0" to "r1"
* PV changed from "0.1.10" to "0.1.12"
</literallayout>
<note>
The <filename>buildhistory-diff</filename> tool requires
the <filename>GitPython</filename> package.
Be sure to install it using Pip3 as follows:
<literallayout class='monospaced'>
$ pip3 install GitPython --user
</literallayout>
Alternatively, you can install
<filename>python3-git</filename> using the appropriate
distribution package manager (e.g.
<filename>apt-get</filename>, <filename>dnf</filename>, or
<filename>zipper</filename>).
</note>
</para>
<para>
To see changes to the build history using a web interface, follow
the instruction in the <filename>README</filename> file here.
<ulink url='http://git.yoctoproject.org/cgit/cgit.cgi/buildhistory-web/'></ulink>.
</para>
<para>
Here is a sample screenshot of the interface:
<imagedata fileref="figures/buildhistory-web.png" align="center" scalefit="1" width="130%" contentdepth="130%" />
</para>
</section>
</section>
</section>
<section id='speeding-up-the-build'>
<title>Speeding Up the Build</title>
<para>
Build time can be an issue.
By default, the build system uses simple controls to try and maximize
build efficiency.
In general, the default settings for all the following variables
result in the most efficient build times when dealing with single
socket systems (i.e. a single CPU).
If you have multiple CPUs, you might try increasing the default
values to gain more speed.
See the descriptions in the glossary for each variable for more
information:
<itemizedlist>
<listitem><para>
<link linkend='var-BB_NUMBER_THREADS'><filename>BB_NUMBER_THREADS</filename>:</link>
The maximum number of threads BitBake simultaneously executes.
</para></listitem>
<listitem><para>
<ulink url='&YOCTO_DOCS_BB_URL;#var-BB_NUMBER_PARSE_THREADS'><filename>BB_NUMBER_PARSE_THREADS</filename>:</ulink>
The number of threads BitBake uses during parsing.
</para></listitem>
<listitem><para>
<link linkend='var-PARALLEL_MAKE'><filename>PARALLEL_MAKE</filename>:</link>
Extra options passed to the <filename>make</filename> command
during the
<link linkend='ref-tasks-compile'><filename>do_compile</filename></link>
task in order to specify parallel compilation on the
local build host.
</para></listitem>
<listitem><para>
<link linkend='var-PARALLEL_MAKEINST'><filename>PARALLEL_MAKEINST</filename>:</link>
Extra options passed to the <filename>make</filename> command
during the
<link linkend='ref-tasks-install'><filename>do_install</filename></link>
task in order to specify parallel installation on the
local build host.
</para></listitem>
</itemizedlist>
As mentioned, these variables all scale to the number of processor
cores available on the build system.
For single socket systems, this auto-scaling ensures that the build
system fundamentally takes advantage of potential parallel operations
during the build based on the build machine's capabilities.
</para>
<para>
Following are additional factors that can affect build speed:
<itemizedlist>
<listitem><para>
File system type:
The file system type that the build is being performed on can
also influence performance.
Using <filename>ext4</filename> is recommended as compared
to <filename>ext2</filename> and <filename>ext3</filename>
due to <filename>ext4</filename> improved features
such as extents.
</para></listitem>
<listitem><para>
Disabling the updating of access time using
<filename>noatime</filename>:
The <filename>noatime</filename> mount option prevents the
build system from updating file and directory access times.
</para></listitem>
<listitem><para>
Setting a longer commit:
Using the "commit=" mount option increases the interval
in seconds between disk cache writes.
Changing this interval from the five second default to
something longer increases the risk of data loss but decreases
the need to write to the disk, thus increasing the build
performance.
</para></listitem>
<listitem><para>
Choosing the packaging backend:
Of the available packaging backends, IPK is the fastest.
Additionally, selecting a singular packaging backend also
helps.
</para></listitem>
<listitem><para>
Using <filename>tmpfs</filename> for
<link linkend='var-TMPDIR'><filename>TMPDIR</filename></link>
as a temporary file system:
While this can help speed up the build, the benefits are
limited due to the compiler using
<filename>-pipe</filename>.
The build system goes to some lengths to avoid
<filename>sync()</filename> calls into the
file system on the principle that if there was a significant
failure, the
<link linkend='build-directory'>Build Directory</link>
contents could easily be rebuilt.
</para></listitem>
<listitem><para>
Inheriting the
<link linkend='ref-classes-rm-work'><filename>rm_work</filename></link>
class:
Inheriting this class has shown to speed up builds due to
significantly lower amounts of data stored in the data
cache as well as on disk.
Inheriting this class also makes cleanup of
<link linkend='var-TMPDIR'><filename>TMPDIR</filename></link>
faster, at the expense of being easily able to dive into the
source code.
File system maintainers have recommended that the fastest way
to clean up large numbers of files is to reformat partitions
rather than delete files due to the linear nature of partitions.
This, of course, assumes you structure the disk partitions and
file systems in a way that this is practical.
</para></listitem>
</itemizedlist>
Aside from the previous list, you should keep some trade offs in
mind that can help you speed up the build:
<itemizedlist>
<listitem><para>
Remove items from
<link linkend='var-DISTRO_FEATURES'><filename>DISTRO_FEATURES</filename></link>
that you might not need.
</para></listitem>
<listitem><para>
Exclude debug symbols and other debug information:
If you do not need these symbols and other debug information,
disabling the <filename>*-dbg</filename> package generation
can speed up the build.
You can disable this generation by setting the
<link linkend='var-INHIBIT_PACKAGE_DEBUG_SPLIT'><filename>INHIBIT_PACKAGE_DEBUG_SPLIT</filename></link>
variable to "1".
</para></listitem>
<listitem><para>
Disable static library generation for recipes derived from
<filename>autoconf</filename> or <filename>libtool</filename>:
Following is an example showing how to disable static
libraries and still provide an override to handle exceptions:
<literallayout class='monospaced'>
STATICLIBCONF = "--disable-static"
STATICLIBCONF_sqlite3-native = ""
EXTRA_OECONF += "${STATICLIBCONF}"
</literallayout>
<note><title>Notes</title>
<itemizedlist>
<listitem><para>
Some recipes need static libraries in order to work
correctly (e.g. <filename>pseudo-native</filename>
needs <filename>sqlite3-native</filename>).
Overrides, as in the previous example, account for
these kinds of exceptions.
</para></listitem>
<listitem><para>
Some packages have packaging code that assumes the
presence of the static libraries.
If so, you might need to exclude them as well.
</para></listitem>
</itemizedlist>
</note>
</para></listitem>
</itemizedlist>
</para>
</section>
</chapter>
<!--
vim: expandtab tw=80 ts=4
-->