source: chapter05/chapter05.xml@ cfabeed

<chapter id="chapter05" xreflabel="Chapter 5">
<title>Constructing a temporary system</title>
<?dbhtml filename="chapter05.html" dir="chapter05"?>


<sect1 id="ch05-introduction">
<title>Introduction</title>
<?dbhtml filename="introduction.html" dir="chapter05"?>

<para>In this chapter we will compile and install a minimal
Linux system. This system will contain just enough tools to be able
to start constructing the final LFS system in the next chapter.</para>

<para>The building of this minimal system is done in two steps: first we
build a brand-new and host-independent toolchain (compiler, assembler,
linker and libraries), and then use this to build all the other essential
tools.</para>

<para>The files compiled in this chapter will be installed under the
<filename class="directory">$LFS/tools</filename> directory
to keep them separate from the files installed in the next chapter.
Since the packages compiled here are merely temporary, we don't want
them to pollute the soon-to-be LFS system.</para>

<para>The key to learning what makes a Linux system work is to know
what each package is used for and why the user or the system needs it.
For this purpose a short summary of the content of each package is given
before the actual installation instructions. For a short description of
each program in a package, please refer to the corresponding section in
<xref linkend="appendixa"/>.</para>

<para>The build instructions assume that you are using the bash shell. There
is also a general expectation that you have already unpacked the sources for a
package and have performed a <userinput>cd</userinput> into the unpacked source
directory before issuing the build commands.</para>

<para>Several of the packages are patched before compilation, but only when
the patch is needed to circumvent a problem. Often the patch is needed in
both this and the next chapter, but sometimes in only one of them. Therefore,
don't worry when instructions for a downloaded patch seem to be missing.</para>

<para>During the installation of most packages you will
see all kinds of compiler warnings scroll by on your screen. These are
normal and can be safely ignored. They are just what they say they are:
warnings -- mostly about deprecated, but not invalid, use of the C or C++
syntax. It's just that C standards have changed rather often and some
packages still use the older standard, which is not really a problem.</para>

<para><emphasis>Unless</emphasis> told not to, you should normally delete the
source and build directories after installing each package -- for cleanness
sake and to save space.</para>

<para>Before continuing, make sure the LFS environment variable is set up
properly by executing the following:</para>

<screen><userinput>echo $LFS</userinput></screen>

<para>Make sure the output shows the path to your LFS partition's mount
point, which is <filename class="directory">/mnt/lfs</filename> if you
followed our example.</para>

</sect1>


<sect1 id="ch05-toolchaintechnotes">
<title>Toolchain technical notes</title>
<?dbhtml filename="toolchaintechnotes.html" dir="chapter05"?>

<para>This section attempts to explain some of the rationale and technical
details behind the overall build method. It's not essential that you understand
everything here immediately. Most of it will make sense once you have performed
an actual build. Feel free to refer back here at any time.</para>

<para>The overall goal of <xref linkend="chapter05"/> is to provide a sane,
temporary environment that we can chroot into, and from which we can produce a
clean, trouble-free build of the target LFS system in
<xref linkend="chapter06"/>. Along the way, we attempt to divorce ourselves
from the host system as much as possible, and in so doing build a
self-contained and self-hosted toolchain. It should be noted that the
build process has been designed in such a way so as to minimize the risks for
new readers and provide maximum educational value at the same time. In other
words, more advanced techniques could be used to build the system.</para>

<important>
<para>Before continuing, you really should be aware of the name of your working
platform, often also referred to as the <emphasis>target triplet</emphasis>. For
many folks the target triplet will be, for example:
<emphasis>i686-pc-linux-gnu</emphasis>. A simple way to determine your target
triplet is to run the <filename>config.guess</filename> script that comes with
the source for many packages. Unpack the Binutils sources and run the script:
<userinput>./config.guess</userinput> and note the output.</para>

<para>You'll also need to be aware of the name of your platform's
<emphasis>dynamic linker</emphasis>, often also referred to as the
<emphasis>dynamic loader</emphasis>, not to be confused with the standard linker
<emphasis>ld</emphasis> that is part of Binutils. The dynamic linker is provided
by Glibc and has the job of finding and loading the shared libraries needed by a
program, preparing the program to run and then running it. For most folks, the
name of the dynamic linker will be <emphasis>ld-linux.so.2</emphasis>. On
platforms that are less prevalent, the name might be
<emphasis>ld.so.1</emphasis> and newer 64 bit platforms might even have
something completely different. You should be able to determine the name
of your platform's dynamic linker by looking in the
<filename class="directory">/lib</filename> directory on your host system. A
surefire way is to inspect a random binary from your host system by running:
<userinput>'readelf -l &lt;name of binary&gt; | grep interpreter'</userinput>
and noting the output. The authoritative reference covering all platforms is in
the <filename>shlib-versions</filename> file in the root of the Glibc source
tree.</para>
</important>

<para>Some key technical points of how the <xref linkend="chapter05"/> build
method works:</para>

<itemizedlist>
<listitem><para>Similar in principle to cross compiling whereby tools installed
into the same prefix work in cooperation and thus utilize a little GNU
"magic".</para></listitem>

<listitem><para>Careful manipulation of the standard linker's library search
path to ensure programs are linked only against libraries we
choose.</para></listitem>

<listitem><para>Careful manipulation of <userinput>gcc</userinput>'s
<emphasis>specs</emphasis> file to tell the compiler which target dynamic
linker will be used.</para></listitem>
</itemizedlist>

<para>Binutils is installed first because both GCC and Glibc perform various
feature tests on the assembler and linker during their respective runs of
<userinput>./configure</userinput> to determine which software features to enable
or disable. This is more important than one might first realize. An incorrectly
configured GCC or Glibc can result in a subtly broken toolchain where the impact
of such breakage might not show up until near the end of the build of a whole
distribution. Thankfully, a test suite failure will usually alert us before too
much time is wasted.</para>

<para>Binutils installs its assembler and linker into two locations,
<filename class="directory">/tools/bin</filename> and
<filename class="directory">/tools/$TARGET_TRIPLET/bin</filename>. In reality,
the tools in one location are hard linked to the other. An important facet of
the linker is its library search order. Detailed information can be obtained
from <userinput>ld</userinput> by passing it the <emphasis>--verbose</emphasis>
flag. For example: <userinput>'ld --verbose | grep SEARCH'</userinput> will
show you the current search paths and their order. You can see what files are
actually linked by <userinput>ld</userinput> by compiling a dummy program and
passing the <emphasis>--verbose</emphasis> switch. For example:
<userinput>'gcc dummy.c -Wl,--verbose 2>&amp;1 | grep succeeded'</userinput>
will show you all the files successfully opened during the link.</para>

<para>The next package installed is GCC and during its run of
<userinput>./configure</userinput> you'll see, for example:</para>

<blockquote><screen>checking what assembler to use... /tools/i686-pc-linux-gnu/bin/as
checking what linker to use... /tools/i686-pc-linux-gnu/bin/ld</screen></blockquote>

<para>This is important for the reasons mentioned above. It also demonstrates
that GCC's configure script does not search the $PATH directories to find which
tools to use. However, during the actual operation of <userinput>gcc</userinput>
itself, the same search paths are not necessarily used. You can find out which
standard linker <userinput>gcc</userinput> will use by running:
<userinput>'gcc -print-prog-name=ld'</userinput>.
Detailed information can be obtained from <userinput>gcc</userinput> by passing
it the <emphasis>-v</emphasis> flag while compiling a dummy program. For
example: <userinput>'gcc -v dummy.c'</userinput> will show you detailed
information about the preprocessor, compilation and assembly stages, including
<userinput>gcc</userinput>'s include search paths and their order.</para>
 
<para>The next package installed is Glibc. The most important considerations for
building Glibc are the compiler, binary tools and kernel headers. The compiler
is generally no problem as Glibc will always use the <userinput>gcc</userinput>
found in a $PATH directory. The binary tools and kernel headers can be a little
more troublesome. Therefore we take no risks and use the available configure
switches to enforce the correct selections. After the run of
<userinput>./configure</userinput> you can check the contents of the
<filename>config.make</filename> file in the
<filename class="directory">glibc-build</filename> directory for all the
important details. You'll note some interesting items like the use of
<userinput>CC="gcc -B/tools/bin/"</userinput> to control which binary tools are
used, and also the use of the <emphasis>-nostdinc</emphasis> and
<emphasis>-isystem</emphasis> flags to control the compiler's include search
path. These items help to highlight an important aspect of the Glibc package:
it is very self-sufficient in terms of its build machinery and generally does
not rely on toolchain defaults.</para>

<para>After the Glibc installation, we make some adjustments to ensure that
searching and linking take place only within our <filename>/tools</filename>
prefix. We install an adjusted <userinput>ld</userinput>, which has a hard-wired
search path limited to <filename class="directory">/tools/lib</filename>. Then
we amend <userinput>gcc</userinput>'s specs file to point to our new dynamic
linker in <filename class="directory">/tools/lib</filename>. This last step is
<emphasis>vital</emphasis> to the whole process. As mentioned above, a
hard-wired path to a dynamic linker is embedded into every ELF shared
executable. You can inspect this by running:
<userinput>'readelf -l &lt;name of binary&gt; | grep interpreter'</userinput>.
By amending <userinput>gcc</userinput>'s specs file, we are ensuring that every
program compiled from here through the end of <xref linkend="chapter05"/> will
use our new dynamic linker in
<filename class="directory">/tools/lib</filename>.</para>

<para>The need to use the new dynamic linker is also the reason why we apply the
Specs patch for the second pass of GCC. Failure to do so will result in the GCC
programs themselves having the name of the dynamic linker from the host system's
<filename class="directory">/lib</filename> directory embedded into them, which
would defeat our goal of getting away from the host.</para>

<para>During the second pass of Binutils, we are able to utilize the
<emphasis>--with-lib-path</emphasis> configure switch to control
<userinput>ld</userinput>'s library search path. From this point onwards, the
core toolchain is self-contained and self-hosted. The remainder of the
<xref linkend="chapter05"/> packages all build against the new Glibc in
<filename class="directory">/tools</filename> and all is well.</para>

<para>Upon entering the chroot environment in <xref linkend="chapter06"/>, the
first major package we install is Glibc, due to its self-sufficient nature that
we mentioned above. Once this Glibc is installed into
<filename class="directory">/usr</filename>, we perform a quick changeover of
the toolchain defaults, then proceed for real in building the rest of the
target <xref linkend="chapter06"/> LFS system.</para>

<sect2>
<title>Notes on static linking</title>

<para>Most programs have to perform, beside their specific task, many rather
common and sometimes trivial operations. These include allocating memory,
searching directories, reading and writing files, string handling, pattern
matching, arithmetic and many other tasks. Instead of obliging each program to
reinvent the wheel, the GNU system provides all these basic functions in
ready-made libraries. The major library on any Linux system is
<emphasis>Glibc</emphasis>.</para>

<para>There are two primary ways of linking the functions from a library to a
program that uses them: statically or dynamically. When a program is linked
statically, the code of the used functions is included in the executable,
resulting in a rather bulky program. When a program is dynamically linked, what
is included is a reference to the dynamic linker, the name of the library, and
the name of the function, resulting in a much smaller executable. (A third way
is to use the programming interface of the dynamic linker. See the
<emphasis>dlopen</emphasis> man page for more information.)</para>

<para>Dynamic linking is the default on Linux and has three major advantages
over static linking. First, you need only one copy of the executable library
code on your hard disk, instead of having many copies of the same code included
into a whole bunch of programs -- thus saving disk space. Second, when several
programs use the same library function at the same time, only one copy of the
function's code is required in core -- thus saving memory space. Third, when a
library function gets a bug fixed or is otherwise improved, you only need to
recompile this one library, instead of having to recompile all the programs that
make use of the improved function.</para>

<para>If dynamic linking has several advantages, why then do we statically link
the first two packages in this chapter? The reasons are threefold: historical,
educational, and technical. Historical, because earlier versions of LFS
statically linked every program in this chapter. Educational, because knowing
the difference is useful. Technical, because we gain an element of independence
from the host in doing so, meaning that those programs can be used
independently of the host system. However, it's worth noting that an overall
successful LFS build can still be achieved when the first two packages are
built dynamically.</para>

</sect2>

</sect1>


<sect1 id="ch05-creatingtoolsdir">
<title>Creating the $LFS/tools directory</title>
<?dbhtml filename="creatingtoolsdir.html" dir="chapter05"?>

<para>All programs compiled in this chapter will be installed under <filename
class="directory">$LFS/tools</filename> to keep them separate from the
programs compiled in the next chapter. The programs compiled here are only
temporary tools and won't be a part of the final LFS system and by keeping them
in a separate directory, we can later easily throw them away.</para>

<para>If later you wish to search through the binaries of your system to see
what files they make use of or link against, then to make this searching easier
you may want to choose a unique name. Instead of the simple "tools" you could
use something like "tools-for-lfs".</para>

<para>Create the required directory by running the following:</para>

<screen><userinput>mkdir $LFS/tools</userinput></screen>

<para>The next step is to create a <filename>/tools</filename> symlink on
your host system. It will point to the directory we just created on the LFS
partition:</para>

<screen><userinput>ln -s $LFS/tools /</userinput></screen>

<para>This symlink enables us to compile our toolchain so that it always
refers to <filename>/tools</filename>, meaning that the compiler, assembler
and linker will work both in this chapter (when we are still using some tools
from the host) <emphasis>and</emphasis> in the next (when we are chrooted to
the LFS partition).</para>

<note><para>Study the above command closely. It can be confusing at first
glance. The <userinput>ln</userinput> command has several syntax variations,
so be sure to check the ln man page before reporting what you may think is an
error.</para></note>

</sect1>


<sect1 id="ch05-addinguser">
<title>Adding the user lfs</title>
<?dbhtml filename="addinguser.html" dir="chapter05"?>

<para>When logged in as <emphasis>root</emphasis>, making a single mistake
can damage or even wreck your system. Therefore we recommend that you
build the packages in this chapter as an unprivileged user. You could
of course use your own user name,  but to make it easier to set up a clean
work environment we'll create a new user <emphasis>lfs</emphasis> and
use this one during the installation process. As <emphasis>root</emphasis>,
issue the following commands to add the new user:</para>

<screen><userinput>useradd -s /bin/bash -m lfs
passwd lfs</userinput></screen>

<para>Now grant this new user <emphasis>lfs</emphasis> full access to
<filename class="directory">$LFS/tools</filename> by giving it ownership
of the directory:</para>

<screen><userinput>chown lfs $LFS/tools</userinput></screen>

<para>If you made a separate working directory as suggested, give user
<emphasis>lfs</emphasis> ownership of this directory too:</para>

<screen><userinput>chown lfs $LFS/sources</userinput></screen>

<para>Next, login as user <emphasis>lfs</emphasis>. This can be done via a
virtual console, through a display manager, or with the following substitute
user command:</para>

<screen><userinput>su - lfs</userinput></screen>

<para>The "<userinput>-</userinput>" instructs <userinput>su</userinput> to
start a new, clean shell.</para>

</sect1>


<sect1 id="ch05-settingenviron">
<title>Setting up the environment</title>
<?dbhtml filename="settingenvironment.html" dir="chapter05"?>

<para>While logged in as user <emphasis>lfs</emphasis>, issue the
following commands to set up a good work environment:</para>

<screen><userinput>cat &gt; ~/.bash_profile &lt;&lt; "EOF"</userinput>
set +h
umask 022
LFS=/mnt/lfs
LC_ALL=POSIX
PATH=/tools/bin:$PATH
export LFS LC_ALL PATH
unset CC CXX CPP LD_LIBRARY_PATH LD_PRELOAD
<userinput>EOF

source ~/.bash_profile</userinput></screen>

<para>The <userinput>set +h</userinput> command turns off
<userinput>bash</userinput>'s hash function. Normally hashing is a useful
feature: <userinput>bash</userinput> uses a hash table to remember the
full pathnames of executable files to avoid searching the PATH time and time
again to find the same executable. However, we'd like the new tools to be
used as soon as they are installed.  By switching off the hash function, our
"interactive" commands (<userinput>make</userinput>,
<userinput>patch</userinput>, <userinput>sed</userinput>,
<userinput>cp</userinput> and so forth) will always use
the newest available version during the build process.</para>

<para>Setting the user file-creation mask to 022 ensures that newly created
files and directories are only writable for their owner, but readable and
executable for anyone.</para>
 
<para>The LFS variable should of course be set to the mount point you
chose.</para>

<para>The LC_ALL variable controls the localization of certain programs,
making their messages follow the conventions of a specified country. If your
host system uses a version of Glibc older than 2.2.4,
having LC_ALL set to something other than "POSIX" or "C" during this chapter
may cause trouble if you exit the chroot environment and wish to return later.
By setting LC_ALL to "POSIX" (or "C", the two are equivalent)  we ensure that
everything will work as expected in the chroot environment.</para>

<para>We prepend <filename>/tools/bin</filename> to the standard PATH so
that, as we move along through this chapter, the tools we build will get used
during the rest of the building process.</para>

<para>The CC, CXX, CPP, LD_LIBRARY_PATH and LD_PRELOAD environment variables all
have the potential to cause havoc with our Chapter 5 toolchain. We therefore
unset them to prevent any chance of this happening.</para>

<para>Now, after sourcing the just-created profile, we're all set to begin
building the temporary tools that will support us in later chapters.</para>

</sect1>


&c5-binutils-pass1;
&c5-gcc-pass1;
&c5-kernelheaders;
&c5-glibc;


<sect1 id="ch05-locking-glibc">
<title>"Locking in" Glibc</title>
<?dbhtml filename="lockingglibc.html" dir="chapter05"?>

<para>Now that the temporary C libraries have been installed, we want all
the tools compiled in the rest of this chapter to be linked against these
libraries. To accomplish this, we need to adjust the linker and the compiler's
specs file.</para>

<para>First install the adjusted linker by running the following from within
the <filename class="directory">binutils-build</filename> directory:</para>

<screen><userinput>make -C ld install</userinput></screen>

<para>The linker was adjusted a little while back, at the end of the first
pass of Binutils. From this point onwards everything will link <emphasis>only
</emphasis> against the libraries in <filename>/tools/lib</filename>.</para>

<note><para>If you somehow missed the earlier warning to retain the Binutils
source and build directories from the first pass or otherwise accidentally
deleted them or just don't have access to them, don't worry, all is not lost.
Just ignore the above command. The result is a small chance of subsequent
programs linking against libraries on the host. This is not ideal, however,
it's not a major problem. The situation is corrected when we install the
second pass of Binutils later on.</para></note>

<para>Now that the adjusted linker is installed, you have to remove the
Binutils build and source directories.</para>

<para>The next thing to do is to amend our GCC specs file so that it points
to the new dynamic linker. A simple sed will accomplish this:</para>

<!-- Ampersands are needed to allow cut and paste -->
   
<screen><userinput>SPECFILE=/tools/lib/gcc-lib/*/*/specs &amp;&amp;
sed -e 's@ /lib/ld-linux.so.2@ /tools/lib/ld-linux.so.2@g' \
&nbsp;&nbsp;&nbsp;&nbsp;$SPECFILE &gt; tempspecfile &amp;&amp;
mv -f tempspecfile $SPECFILE &amp;&amp;
unset SPECFILE</userinput></screen>

<para>We recommend that you cut-and-paste the above rather than try and type it
all in. Or you can edit the specs file by hand if you want to: just replace any
occurrence of "/lib/ld-linux.so.2" with "/tools/lib/ld-linux.so.2".</para>

<important><para>If you are working on a platform where the name of the dynamic
linker is something other than <filename>ld-linux.so.2</filename>, you
<emphasis>must</emphasis> substitute <filename>ld-linux.so.2</filename> with the
name of your platform's dynamic linker in the above commands. Refer back to
<xref linkend="ch05-toolchaintechnotes"/> if necessary.</para></important>

<para>Lastly, there is a possibility that some include files from the host
system have found their way into GCC's private include dir. This can happen
because of GCC's "fixincludes" process which runs as part of the GCC build.
We'll explain more about this further on in this chapter.  For now, run the
following commands to eliminate this possibility:</para>

<screen><userinput>rm -f /tools/lib/gcc-lib/*/*/include/{pthread.h,bits/sigthread.h}</userinput></screen>

<!-- HACK - Force some whitespace to appease tidy -->
<literallayout></literallayout>

<caution><para>It is imperative at this point to stop and ensure that the basic
functions (compiling and linking) of the new toolchain are working as expected.
For this we are going to perform a simple sanity check:</para>

<screen><userinput>echo 'main(){}' &gt; dummy.c
gcc dummy.c
readelf -l a.out | grep ': /tools'</userinput></screen>

<para>If everything is working correctly, there should be no errors, and the
output of the last command will be:</para>

<blockquote><screen>[Requesting program interpreter: /tools/lib/ld-linux.so.2]</screen></blockquote>

<para>If you did not receive the output as shown above, or received no output at
all, then something is seriously wrong. You will need to investigate and retrace
your steps to find out where the problem is and correct it. There is no point in
continuing until this is done. Most likely something went wrong with the specs
file amendment above. Note especially that <filename>/tools/lib</filename>
appears as the prefix of our dynamic linker. Of course, if you are working on a
platform where the name of the dynamic linker is something other than
<filename>ld-linux.so.2</filename>, then the output will be slightly
different.</para>

<para>Once you are satisfied that all is well, clean up the test files:</para>

<screen><userinput>rm dummy.c a.out</userinput></screen>
</caution>

<!-- HACK - Force some whitespace to appease tidy -->
<literallayout></literallayout>

<para>This completes the installation of the self-contained toolchain, and it
can now be used to build the rest of the temporary tools.</para>

</sect1>


&c5-tcl;
&c5-expect;
&c5-dejagnu;
&c5-gcc-pass2;
&c5-binutils-pass2;

&c5-gawk;
&c5-coreutils;
&c5-bzip2;
&c5-gzip;
&c5-diffutils;
&c5-findutils;
&c5-make;
&c5-grep;
&c5-sed;
&c5-gettext;
&c5-ncurses;
&c5-patch;
&c5-tar;
&c5-texinfo;
&c5-bash;
&c5-utillinux;
&c5-perl;


<sect1 id="ch05-stripping">
<title>Stripping</title>
<?dbhtml filename="stripping.html" dir="chapter05"?>

<para>The steps in this section are optional. If your LFS partition is rather
small, you will be glad to learn that you can throw away some unnecessary
things. The executables and libraries you have built so far contain about 130 MB
of unneeded debugging symbols. Remove those symbols like this:</para>

<screen><userinput>strip --strip-unneeded /tools/{,s}bin/*
strip --strip-debug /tools/lib/*</userinput></screen>

<para>The first of the above commands will skip some twenty files, reporting
that it doesn't recognize their file format. Most of them are scripts instead
of binaries.</para>

<para>Take care <emphasis>not</emphasis> to use
<userinput>--strip-unneeded</userinput> on the libraries -- they would be
destroyed and you would have to build Glibc all over again.</para>

<para>To save another couple of megabytes, you can throw away all the
documentation:</para>

<screen><userinput>rm -rf /tools/{,share/}{doc,info,man}</userinput></screen>

<para>You will now need to have at least 850 MB of free space on your LFS
filesystem to be able to build and install Glibc in the next phase. If you can
build and install Glibc, you can build and install the rest too.</para>

</sect1>

</chapter>