libvirt/docs/compiling.html.in
Pavel Hrdina f96395e78e build: mandate use of a build dir != src dir
Historically we've allowed builds in the main src dir, but meson does
not support this. Explicitly force separate build dir in autotools to
align with meson. We must re-enable dependency tracking which the RPM
%configure macro turns off. Without this, the build dir doesn't get
the source directory tree mirrored.

Signed-off-by: Daniel P. Berrangé <berrange@redhat.com>
Signed-off-by: Pavel Hrdina <phrdina@redhat.com>
Reviewed-by: Daniel P. Berrangé <berrange@redhat.com>
2019-11-08 17:07:35 +01:00

143 lines
4.6 KiB
XML

<?xml version="1.0" encoding="UTF-8"?>
<!DOCTYPE html>
<html xmlns="http://www.w3.org/1999/xhtml">
<body>
<h1><a id="installation">libvirt Installation</a></h1>
<ul id="toc"></ul>
<h2><a id="compiling">Compiling a release tarball</a></h2>
<p>
libvirt uses the standard configure/make/install steps and mandates
that the build directory is different that the source directory:
</p>
<pre>
$ xz -c libvirt-x.x.x.tar.xz | tar xvf -
$ cd libvirt-x.x.x
$ mkdir build &amp;&amp; cd build
$ ../configure</pre>
<p>
The <i>configure</i> script can be given options to change its default
behaviour.
</p>
<p>
To get the complete list of the options it can take, pass it the
<i>--help</i> option like this:
</p>
<pre>
$ ../configure <i>--help</i></pre>
<p>
When you have determined which options you want to use (if any),
continue the process.
</p>
<p>
Note the use of <b>sudo</b> with the <i>make install</i> command
below. Using sudo is only required when installing to a location your
user does not have write access to. Installing to a system location
is a good example of this.
</p>
<p>
If you are installing to a location that your user <i>does</i> have write
access to, then you can instead run the <i>make install</i> command
without putting <b>sudo</b> before it.
</p>
<pre>
$ ../configure <i>[possible options]</i>
$ make
$ <b>sudo</b> <i>make install</i></pre>
<p>
At this point you <b>may</b> have to run ldconfig or a similar utility
to update your list of installed shared libs.
</p>
<h2><a id="building">Building from a GIT checkout</a></h2>
<p>
The libvirt build process uses GNU autotools, so after obtaining a
checkout it is necessary to generate the configure script and Makefile.in
templates using the <code>autogen.sh</code> command. By default when
the <code>configure</code> script is run from within a GIT checkout, it
will turn on -Werror for builds. This can be disabled with
--disable-werror, but this is not recommended.
</p>
<p>
Libvirt takes advantage of
the <a href="http://www.gnu.org/software/gnulib/">gnulib</a>
project to provide portability to a number of platforms. This
is normally done dynamically via a git submodule in
the <code>.gnulib</code> subdirectory, which is auto-updated as
needed when you do incremental builds. Setting the environment
variable <code>GNULIB_SRCDIR</code> to a local directory
containing a git checkout of gnulib will let you reduce local
disk space requirements and network download time, regardless of
which actual commit you have in that reference directory.
</p>
<p>
However, if you are developing on a platform where git is not
available, or are behind a firewall that does not allow for git
to easily obtain the gnulib submodule, it is possible to instead
use a static mode of operation where you are then responsible
for updating the git submodule yourself. In this mode, you must
track the exact gnulib commit needed by libvirt (usually not the
latest gnulib.git) via alternative means, such as a shared NFS
drive or manual download, and run this any time libvirt.git
updates the commit stored in the .gnulib submodule:</p>
<pre>
$ GNULIB_SRCDIR=/path/to/gnulib ./autogen.sh --no-git
</pre>
<p>To build &amp; install libvirt to your home
directory the following commands can be run:
</p>
<pre>
$ ./autogen.sh --prefix=$HOME/usr
$ make
$ <b>sudo</b> make install</pre>
<p>
Be aware though, that binaries built with a custom prefix will not
interoperate with OS vendor provided binaries, since the UNIX socket
paths will all be different. To produce a build that is compatible
with normal OS vendor prefixes, use
</p>
<pre>
$ ./autogen.sh --system
$ make
</pre>
<p>
When doing this for day-to-day development purposes, it is recommended
not to install over the OS vendor provided binaries. Instead simply
run libvirt directly from the source tree. For example to run
a privileged libvirtd instance
</p>
<pre>
$ su -
# service libvirtd stop (or systemctl stop libvirtd.service)
# /home/to/your/checkout/src/libvirtd
</pre>
<p>
It is also possible to run virsh directly from the source tree
using the ./run script (which sets some environment variables):
</p>
<pre>
$ ./run ./tools/virsh ....
</pre>
</body>
</html>