<?xml version="1.0" encoding="iso-8859-1"?>
<!DOCTYPE article PUBLIC "-//FreeBSD//DTD DocBook XML V4.5-Based Extension//EN"
"../../../share/xml/freebsd45.dtd">
<article lang='en'>
<articleinfo>
<title>Committer's Guide</title>
<corpauthor>The &os; Documentation Project</corpauthor>
<copyright>
<year>1999</year>
<year>2000</year>
<year>2001</year>
<year>2002</year>
<year>2003</year>
<year>2004</year>
<year>2005</year>
<year>2006</year>
<year>2007</year>
<year>2008</year>
<year>2009</year>
<year>2010</year>
<year>2011</year>
<year>2012</year>
<year>2013</year>
<holder>The FreeBSD Documentation Project</holder>
</copyright>
<legalnotice id="trademarks" role="trademarks">
&tm-attrib.freebsd;
&tm-attrib.coverity;
&tm-attrib.ibm;
&tm-attrib.intel;
&tm-attrib.sparc;
&tm-attrib.general;
</legalnotice>
<pubdate>$FreeBSD$</pubdate>
<releaseinfo>$FreeBSD$</releaseinfo>
<abstract>
<para>This document provides information for the FreeBSD
committer community. All new committers should read this
document before they start, and existing committers are
strongly encouraged to review it from time to time.</para>
<para>Almost all FreeBSD developers have commit rights to one or
more repositories. However, a few developers do not, and some
of the information here applies to them as well. (For
instance, some people only have rights to work with the
Problem Report database). Please see <xref
linkend="non-committers"/> for more information.</para>
<para>This document may also be of interest to members of the
FreeBSD community who want to learn more about how the project
works.</para>
</abstract>
</articleinfo>
<sect1 id="admin">
<title>Administrative Details</title>
<informaltable frame="none" orient="port" pgwide="1">
<tgroup cols="2">
<colspec colwidth="20*"/>
<colspec colwidth="80*"/>
<tbody>
<row>
<entry><emphasis>Login Methods</emphasis></entry>
<entry>&man.ssh.1;, protocol 2 only</entry>
</row>
<row>
<entry><emphasis>Main Shell Host</emphasis></entry>
<entry><hostid
role="fqdn">freefall.FreeBSD.org</hostid></entry>
</row>
<row>
<entry><emphasis><literal>src/</literal> Subversion
Root</emphasis></entry>
<entry>
<literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/base</filename>
(see also <xref linkend="subversion-primer"/>).</entry>
</row>
<row>
<entry><emphasis><literal>doc/</literal> Subversion
Root</emphasis></entry>
<entry>
<literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/doc</filename>
(see also <xref linkend="subversion-primer"/>).</entry>
</row>
<row>
<entry><emphasis><literal>ports/</literal> Subversion
Root</emphasis></entry>
<entry>
<literal>svn+ssh://</literal><hostid
role="fqdn">svn.FreeBSD.org</hostid><filename>/ports</filename>
(see also <xref linkend="subversion-primer"/>).</entry>
</row>
<row>
<entry><emphasis>Internal Mailing Lists</emphasis></entry>
<entry>developers (technically called all-developers),
doc-developers, doc-committers, ports-developers,
ports-committers, src-developers, src-committers. (Each
project repository has its own -developers and
-committers mailing lists. Archives for these lists may
be found in files
<filename>/home/mail/<replaceable>repository-name</replaceable>-developers-archive</filename>
and
<filename>/home/mail/<replaceable>repository-name</replaceable>-committers-archive</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.)</entry>
</row>
<row>
<entry><emphasis>Core Team monthly
reports</emphasis></entry>
<entry><filename>/home/core/public/monthly-reports</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.
</entry>
</row>
<row>
<entry><emphasis>Ports Management Team monthly
reports</emphasis></entry>
<entry><filename>/home/portmgr/public/monthly-reports</filename>
on the <hostid role="domainname">FreeBSD.org</hostid>
cluster.</entry>
</row>
<row>
<entry><emphasis>Noteworthy <literal>src/</literal> SVN
Branches</emphasis></entry>
<entry>
<literal>stable/7</literal> (7.X-STABLE),
<literal>stable/8</literal> (8.X-STABLE),
<literal>stable/9</literal> (9.X-STABLE),
<literal>head</literal> (-CURRENT)
</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>It is required that you use &man.ssh.1;
to connect to the project hosts.
If you do
not know anything about &man.ssh.1;, please see
<xref linkend="ssh.guide"/>.</para>
<para>Useful links:</para>
<itemizedlist>
<listitem><para><ulink url="&url.base;/internal/">FreeBSD
Project Internal Pages</ulink></para></listitem>
<listitem><para><ulink
url="&url.base;/internal/machines.html">FreeBSD Project
Hosts</ulink></para></listitem>
<listitem><para><ulink
url="&url.base;/administration.html">FreeBSD Project
Administrative Groups</ulink></para></listitem>
</itemizedlist>
</sect1>
<sect1 id="committer.types">
<title>Commit Bit Types</title>
<para>The FreeBSD repository has a number of components which,
when combined, support the basic operating system source,
documentation, third party application ports infrastructure, and
various maintained utilities. When FreeBSD commit bits are
allocated, the areas of the tree where the bit may be used are
specified. Generally, the areas associated with a bit reflect
who authorized the allocation of the commit bit. Additional
areas of authority may be added at a later date: when this
occurs, the committer should follow normal commit bit allocation
procedures for that area of the tree, seeking approval from the
appropriate entity and possibly getting a mentor for that area
for some period of time.</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="3">
<tbody>
<row>
<entry><emphasis>Committer Type</emphasis></entry>
<entry><emphasis>Responsible</emphasis></entry>
<entry><emphasis>Tree Components</emphasis></entry>
</row>
<row>
<entry>src</entry>
<entry>core@</entry>
<entry>src/, doc/ subject to appropriate review</entry>
</row>
<row>
<entry>doc</entry>
<entry>doceng@</entry>
<entry>doc/, src/ documentation</entry>
</row>
<row>
<entry>ports</entry>
<entry>portmgr@</entry>
<entry>ports/</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<para>Commit bits allocated prior to the development of the notion
of areas of authority may be appropriate for use in many parts
of the tree. However, common sense dictates that a committer
who has not previously worked in an area of the tree seek review
prior to committing, seek approval from the appropriate
responsible party, and/or work with a mentor. Since the rules
regarding code maintenance differ by area of the tree, this is
as much for the benefit of the committer working in an area of
less familiarity as it is for others working on the tree.</para>
<para>Committers are encouraged to seek review for their work as
part of the normal development process, regardless of the area
of the tree where the work is occurring.</para>
<sect2>
<title>Policy for <filename>doc/</filename> committer activity
in <filename>src/</filename></title>
<itemizedlist>
<listitem><para>doc committers may commit documentation
changes to src files, such as man pages, READMEs, fortune
databases, calendar files, and comment fixes without
approval from a src committer, subject to the normal care
and tending of commits.</para></listitem>
<listitem><para>doc committers may commit minor src changes
and fixes, such as build fixes, small features, etc, with an
"Approved by" from a src committer.</para></listitem>
<listitem><para>doc committers may seek an upgrade to a src
commit bit by acquiring a mentor, who will propose the doc
committer to core. When approved, they will be added to
'access' and the normal mentoring period will ensue, which
will involve a continuing of <quote>Approved by</quote> for
some period.</para></listitem>
<listitem><para>"Approved by" is only acceptable from
non-mentored src committers -- mentored committers can
provide a "Reviewed by" but not an "Approved
by".</para></listitem>
</itemizedlist>
</sect2>
</sect1>
<sect1 id="subversion-primer">
<title>Subversion Primer</title>
<para>It is assumed that you are already familiar with the basic
operation of the version control systems in use. Traditionally
this was CVS. Subversion is used for the <literal>src</literal>
tree as of May 2008, the <literal>doc/www</literal> tree as of
May 2012 and the <literal>ports</literal> tree as of July 2012.
</para>
<para><ulink url="http://wiki.freebsd.org/SubversionMissing">There
is a list of things missing in Subversion when compared to CVS
</ulink>. The notes at <ulink
url="http://people.freebsd.org/~peter/svn_notes.txt"></ulink>
might also be useful.</para>
<sect2 id="svn-intro">
<title>Introduction</title>
<para>The &os; source repository switched from
<acronym>CVS</acronym> to Subversion on May 31st, 2008. The
first real <acronym>SVN</acronym> commit is
<emphasis>r179447</emphasis>.</para>
<para>The &os; <literal>doc/www</literal> repository switched
from <acronym>CVS</acronym> to Subversion on May 19th, 2012.
The first real <acronym>SVN</acronym> commit is
<emphasis>r38821</emphasis>.</para>
<note>
<para>Part of the <literal>doc/www</literal>
<acronym>CVS</acronym> to <acronym>SVN</acronym> conversion
included an infrastructural change to the build process.
The most notable change is the location of the
&os; website <literal>www</literal> tree, which has
been moved from
<literal>www/<replaceable>lang</replaceable>/</literal> to
<literal>head/<replaceable>lang</replaceable>/htdocs/</literal>.</para>
</note>
<para>The &os; <literal>ports</literal> repository switched
from <acronym>CVS</acronym> to Subversion on July 14th, 2012.
The first real <acronym>SVN</acronym> commit is
<emphasis>r300894</emphasis>.</para>
<para>There are mechanisms in place to automatically merge
changes back from the Subversion <literal>src</literal>
repository to the <acronym>CVS</acronym> repository for
some &os; branches (<literal>releng/6</literal> through
<literal>releng/9</literal>), however this is purely to
support pre-existing end-user installs and should not be
relied upon, recommended or advertised. Future branches
will not be exported to CVS at all. The
<literal>ports</literal> repository was exported to CVS
for a period of time to aid end user migration, but as of
28th February 2013 is no longer exported.</para>
<para>Subversion is not that different from
<acronym>CVS</acronym> when it comes to daily use, but there
are differences. Subversion has a number of features that
should make developers' lives easier. The most important
advantage to Subversion (and the reason why &os; switched) is
that it handles branches and merging much better than CVS
does. Some of the principal differences are:</para>
<itemizedlist>
<listitem>
<para>Commits are atomic.</para>
</listitem>
<listitem>
<para>Revision numbers apply across the repository—all
files that were modified in the same commit have the same
revision number.</para>
</listitem>
<listitem>
<para>Branching and tagging are namespace operations.</para>
</listitem>
<listitem>
<para>Directories are versioned.</para>
</listitem>
<listitem>
<para>Files and directories can have arbitrary, versioned
metadata attached to them.</para>
</listitem>
<listitem>
<para>Files and directories can be copied, with full history
tracking.</para>
</listitem>
<listitem>
<para>No more contortions due to <acronym>CVS</acronym>
weakness such as applying &man.patch.1; files at compile
time in order to avoid touching vendor branch code.</para>
</listitem>
<listitem>
<para>No more repo-copies.</para>
</listitem>
</itemizedlist>
<para>Subversion can be installed from the &os; Ports
Collection, by issuing the following commands:</para>
<screen>&prompt.root; <userinput>cd /usr/ports/devel/subversion</userinput>
&prompt.root; <userinput>make clean install</userinput></screen>
</sect2>
<sect2 id="svn-getting-started">
<title>Getting Started</title>
<para>There are a few ways to obtain a working copy of the tree
from Subversion. This section will explain them.</para>
<sect3>
<title>Direct Checkout</title>
<para>The first is to check out directly from the main
repository. For the <literal>src</literal> tree,
use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/base/head /usr/src</userinput></screen>
<para>For the <literal>doc</literal> tree, use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/doc/head /usr/doc</userinput></screen>
<para>For the <literal>ports</literal> tree, use:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/ports/head /usr/ports</userinput></screen>
<note>
<para>Though the remaining examples in this document are
written with the workflow of working with the
<literal>src</literal> tree in mind, the underlying
concepts are the same for working with the
<literal>doc</literal> and the <literal>ports</literal>
tree.
Ports related Subversion operations are listed in
<xref linkend="ports"/>.</para>
</note>
<para>The above command will check out a
<literal>CURRENT</literal> source tree as <filename
class="directory"><replaceable>/usr/src/</replaceable></filename>,
which can be any target directory on the local filesystem.
Omitting the final argument of that command causes the
working copy, in this case, to be named <quote>head</quote>,
but that can be renamed safely.</para>
<para><literal>svn+ssh</literal> means the
<acronym>SVN</acronym> protocol tunnelled over
<acronym>SSH</acronym>. The name of the server is
<literal>svn.freebsd.org</literal>, <literal>base</literal>
is the path to the repository, and <literal>head</literal>
is the subdirectory within the repository.</para>
<para>If your &os; login name is different from your login
name on your local machine, you must either include it in
the <acronym>URL</acronym> (for example
<literal>svn+ssh://jarjar@svn.freebsd.org/base/head</literal>),
or add an entry to your <filename>~/.ssh/config</filename>
in the form:</para>
<programlisting>Host svn.freebsd.org
User jarjar</programlisting>
<para>This is the simplest method, but it's hard to tell just
yet how much load it will place on the repository.
Subversion is much faster than <acronym>CVS</acronym>,
however.</para>
<note>
<para>The <command>svn diff</command> does not require
access to the server as <acronym>SVN</acronym> stores a
reference copy of every file in the working copy. This,
however, means that Subversion working copies are very
large in size.</para>
</note>
</sect3>
<sect3>
<title>Checkout from a Mirror</title>
<para>Check out a working copy from a mirror by
substituting the mirror's <acronym>URL</acronym> for
<literal>svn+ssh://svn.freebsd.org/base</literal>. This can
be an official mirror or a mirror maintained by
using <command>svnsync</command>.</para>
<para>There is a serious disadvantage to this method: every
time something is to be committed, a <command>svn switch
--relocate</command> to the master repository has to be
done, remembering to <command>svn switch</command> back to
the mirror after the commit. Also, since <command>svn
switch</command> only works between repositories that have
the same UUID, some hacking of the local repository's UUID
has to occur before it is possible to start using it.</para>
<para>Unlike with <acronym>CVS</acronym>,
the hassle of a local
<command>svnsync</command> mirror probably is not worth it
unless the network connectivity situation or other factors
demand it. If it is needed, see the end of this chapter for
information on how to set one up.</para>
</sect3>
<sect3 id="subversion-primer-base-layout">
<title><literal>RELENG_*</literal> Branches and General
Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/base</literal>,
<emphasis>base</emphasis> refers to the source tree.
Similarly, <emphasis>ports</emphasis> refers to the ports
tree, and so on. These are separate repositories with their
own change number sequences, access controls and commit
mail.</para>
<para>For the base repository, HEAD refers to the -CURRENT
tree. For example, <filename>head/bin/ls</filename> is what
would go into <filename>/usr/src/bin/ls</filename> in a
release. Some key locations are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/head/</emphasis>
which corresponds to <literal>HEAD</literal>, also known as
<literal>-CURRENT</literal>.
</para>
</listitem>
<listitem>
<para><emphasis>/stable/<replaceable>n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n</replaceable></literal>.</para>
</listitem>
<listitem>
<para><emphasis>/releng/<replaceable>n.n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n_n</replaceable></literal>.</para>
</listitem>
<listitem>
<para><emphasis>/release/<replaceable>n.n.n</replaceable></emphasis>
which corresponds to
<literal>RELENG_<replaceable>n_n_n</replaceable>_RELEASE</literal>.</para>
</listitem>
<listitem>
<para><emphasis>/vendor*</emphasis> is the vendor branch
import work area. This directory itself does not
contain branches, however its subdirectories do. This
contrasts with the <emphasis>stable</emphasis>,
<emphasis>releng</emphasis> and
<emphasis>release</emphasis> directories.</para>
</listitem>
<listitem>
<para><emphasis>/projects</emphasis> and
<emphasis>/user</emphasis> feature a branch work area,
like in Perforce. As above, the
<emphasis>/user</emphasis> directory does not contain
branches itself.</para>
</listitem>
</itemizedlist>
</sect3>
<sect3>
<title>&os; Documentation Project Branches and
Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/doc</literal>,
<emphasis>doc</emphasis> refers repository root of the
source tree.</para>
<para>In general, most &os; Documentation Project work will be
done within the <filename>head/</filename> branch of the
source tree.</para>
<para>&os; documentation is written and/or translated to
various languages, each of which within a separate
directory within the <filename>head/</filename>
branch.</para>
<para>Each translation set contains several subdirectories for
the various parts of the &os; Documentation Project. A few
noteworthy directories are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/articles/</emphasis> contains the source
code for articles written by various &os;
contributors.</para>
</listitem>
<listitem>
<para><emphasis>/books/</emphasis> contains the source
code for the different books, such as the
&os; Handbook.</para>
</listitem>
<listitem>
<para><emphasis>/htdocs/</emphasis> contains the source
code for the &os; website.</para>
</listitem>
</itemizedlist>
</sect3>
<sect3>
<title>&os; Ports Tree Branches and Layout</title>
<para>In <literal>svn+ssh://svn.freebsd.org/ports</literal>,
<emphasis>ports</emphasis> refers repository root of the
ports tree.</para>
<para>In general, most &os; port work will be done within
the <filename>head/</filename> branch of the ports tree
which is the actual ports tree used to install software.
Some other key locations are:</para>
<itemizedlist>
<listitem>
<para><emphasis>/branches/RELENG_<replaceable>n_n_n
</replaceable></emphasis> which corresponds to
<literal>RELENG_<replaceable>n_n_n</replaceable></literal>
is used to merge back security updates in preparation
for a release.</para>
</listitem>
<listitem>
<para><emphasis>/tags/RELEASE_<replaceable>n_n_n</replaceable></emphasis>
which corresponds to <literal>RELEASE_<replaceable>n_n_n</replaceable></literal>
represents a release tag of the ports tree.</para>
</listitem>
<listitem>
<para><emphasis>/tags/RELEASE_<replaceable>n</replaceable>_EOL</emphasis>
represents the end of life tag of a specific &os;
branch.</para>
</listitem>
</itemizedlist>
</sect3>
</sect2>
<sect2 id="svn-daily-use">
<title>Daily Use</title>
<para>This section will explain how to perform common day-to-day
operations with Subversion. There should be no difference
between <acronym>SVN</acronym> and <acronym>SVK</acronym> in
daily use, except for the revision renumbering mentioned
earlier.</para>
<sect3>
<title>Help</title>
<para>Both <acronym>SVN</acronym> and <acronym>SVK</acronym>
have built in help documentation. It can be accessed by
typing the following command:</para>
<screen>&prompt.user; <userinput>svn help</userinput></screen>
</sect3>
<sect3>
<title>Checkout</title>
<para>As seen earlier, to check out the &os; head
branch:</para>
<screen>&prompt.user; <userinput>svn checkout svn+ssh://svn.freebsd.org/base/head /usr/src</userinput></screen>
<para>At some point, more than just <literal>HEAD</literal>
will probably be useful, for instance when merging changes
to stable/7. Therefore, it may be useful to have a partial
checkout of the complete tree (a full checkout would be very
painful).</para>
<para>To do this, first check out the root of the
repository:</para>
<screen>&prompt.user; <userinput>svn checkout --depth=immediates svn+ssh://svn.freebsd.org/base</userinput></screen>
<para>This will give <literal>base</literal> with all the
files it contains (at the time of writing, just
<filename>ROADMAP.txt</filename>) and empty subdirectories
for <literal>head</literal>, <literal>stable</literal>,
<literal>vendor</literal> and so on.</para>
<para>Expanding the working copy is possible. Just change the
depth of the various subdirectories:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/head</userinput>
&prompt.user; <userinput>svn up --set-depth=immediates base/release base/releng base/stable</userinput></screen>
<para>The above command will pull down a full copy of
<literal>head</literal>, plus empty copies of every
<literal>release</literal> tag, every
<literal>releng</literal> branch, and every
<literal>stable</literal> branch.</para>
<para>If at a later date merging to
<literal>7-STABLE</literal> is required, expand the working
copy:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7</userinput></screen>
<para>Subtrees do not have to be expanded completely. For
instance, expanding only <literal>stable/7/sys</literal> and
then later expand the rest of
<literal>stable/7</literal>:</para>
<screen>&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7/sys</userinput>
&prompt.user; <userinput>svn up --set-depth=infinity base/stable/7</userinput></screen>
<para>Updating the tree with <command>svn update</command>
will only update what was previously asked for (in this
case, <literal>head</literal> and
<literal>stable/7</literal>; it will not pull down the whole
tree.</para>
<note>
<para>Decreasing the depth of a working copy is not
possible.</para>
</note>
</sect3>
<sect3>
<title>Anonymous Checkout</title>
<para>It is possible to anonymously check out the &os;
repository with Subversion. This will give access to a
read-only tree that can be updated, but not committed
to. To do this, use the following command:</para>
<screen>&prompt.user; <userinput>svn co <replaceable>https://svn0.us-west.FreeBSD.org</replaceable>/base/head /usr/src</userinput></screen>
<para>Select the closest mirror and verify the mirror server
certificate from the list of <ulink
url="&url.books.handbook;/svn-mirrors.html">Subversion
mirror sites</ulink>.</para>
</sect3>
<sect3>
<title>Updating the Tree</title>
<para>To update a working copy to either the latest revision,
or a specific revision:</para>
<screen>&prompt.user; <userinput>svn update</userinput>
&prompt.user; <userinput>svn update -<replaceable>r12345</replaceable></userinput></screen>
</sect3>
<sect3>
<title>Status</title>
<para>To view the local changes that have been made to the
working copy:</para>
<screen>&prompt.user; <userinput>svn status</userinput></screen>
<para>To show local changes and files that are out-of-date do:</para>
<screen>&prompt.user; <userinput>svn status --show-updates</userinput></screen>
</sect3>
<sect3>
<title>Editing and Committing</title>
<para>Unlike Perforce,
<acronym>SVN</acronym> and <acronym>SVK</acronym> do not
need to be told in advance about file editing.</para>
<para><command>svn commit</command> works like the equivalent
<acronym>CVS</acronym> command. To commit all changes in
the current directory and all subdirectories:</para>
<screen>&prompt.user; <userinput>svn commit</userinput></screen>
<para>To commit all changes in, for example, <filename
class="directory"><replaceable>lib/libfetch/</replaceable></filename>
and <filename
class="directory"><replaceable>usr/bin/fetch/</replaceable></filename>
in a single operation:</para>
<screen>&prompt.user; <userinput>svn commit <replaceable>lib/libfetch</replaceable> <replaceable>usr/bin/fetch</replaceable></userinput></screen>
<para>There is also a commit wrapper for the ports tree
to handle the properties and sanity checking your
changes:</para>
<screen>&prompt.user; <userinput>/usr/ports/Tools/scripts/psvn commit
</userinput></screen>
</sect3>
<sect3 id="subversion-primer-add-remove">
<title>Adding and Removing Files</title>
<note>
<para>Before adding files, get a copy of <ulink
url="http://people.freebsd.org/~peter/auto-props.txt">auto-props.txt</ulink>
(there is also a <ulink
url="http://people.freebsd.org/~beat/cvs2svn/auto-props.txt">
ports tree specific version</ulink>)
and add it to <filename>~/.subversion/config</filename>
according to the instructions in the file. If you added
something before you've read this, you may use
<command>svn rm --keep-local</command> for just added
files, fix your config file and re-add them again. The
initial config file is created when you first run a svn
command, even something as simple as <command>svn
help</command>.
</para>
</note>
<para>Files are added to a
<acronym>SVN</acronym> repository with <command>svn
add</command>. To add a file named
<emphasis>foo</emphasis>, edit it, then:</para>
<screen>&prompt.user; <userinput>svn add <replaceable>foo</replaceable></userinput></screen>
<para>Files can be removed with <command>svn
remove</command>:</para>
<screen>&prompt.user; <userinput>svn remove <replaceable>foo</replaceable></userinput></screen>
<para>Subversion does not require <command>rm</command>ing the
file before <command>svn rm</command>ing it, and indeed
complains if that happens.</para>
<para>It is possible to add directories with <command>svn
add</command>:</para>
<screen>&prompt.user; <userinput>mkdir <replaceable>bar</replaceable></userinput>
&prompt.user; <userinput>svn add <replaceable>bar</replaceable></userinput></screen>
<para>Although <command>svn mkdir</command> makes this
easier by combining the creation of the directory and the
adding of it:</para>
<screen>&prompt.user; <userinput>svn mkdir <replaceable>bar</replaceable></userinput></screen>
<para>The directory is not immediately
created in the repository when you use <command>svn
mkdir</command>. Subversion
allows directories to be removed using <command>svn
rm</command>, however there is no <command>svn
rmdir</command>:</para>
<screen>&prompt.user; <userinput>svn rm <replaceable>bar</replaceable></userinput></screen>
</sect3>
<sect3>
<title>Copying and Moving Files</title>
<para>The following (obviously) creates a copy of
<filename>foo.c</filename>, named
<filename>bar.c</filename>:</para>
<screen>&prompt.user; <userinput>svn copy <replaceable>foo.c</replaceable> <replaceable>bar.c</replaceable></userinput></screen>
<para>To move and rename a file:</para>
<screen>&prompt.user; <userinput>svn move <replaceable>foo.c</replaceable> <replaceable>bar.c</replaceable></userinput></screen>
<para>The above command is the exact equivalent of:</para>
<screen>&prompt.user; <userinput>svn copy <replaceable>foo.c</replaceable> <replaceable>bar.c</replaceable></userinput>
&prompt.user; <userinput>svn remove <replaceable>foo.c</replaceable></userinput></screen>
</sect3>
<sect3>
<title>Log and Annotate</title>
<para><command>svn log</command> will show all the
revisions that affect a directory and files within that
directory in reverse chronological order, if run on a
directory. This contrasts with <command>cvs log</command>
in that <acronym>CVS</acronym> shows the complete log for
each file in the directory, including duplicate entries for
revisions that affect multiple files.</para>
<para><command>svn annotate</command>, or equally <command>svn
praise</command> or <command>svn blame</command>, is
equivalent to <command>cvs annotate</command> in everything
but output format.</para>
</sect3>
<sect3>
<title>Diffs</title>
<para><command>svn diff</command> displays changes to the
working copy of the repository. Diffs generated by
<acronym>SVN</acronym> are unified
and include new files by default
in the diff output.</para>
<para><command>svn
diff</command> can show the changes between two revisions
of the same file:</para>
<screen>&prompt.user; <userinput>svn diff -r179453:179454 ROADMAP.txt</userinput></screen>
<para>It can also show all changes for a specific changeset.
The following will show what changes were made to the
current directory and all subdirectories in changeset
179454:</para>
<screen>&prompt.user; <userinput>svn diff -c179454 .</userinput></screen>
</sect3>
<sect3>
<title>Reverting</title>
<para>Local changes (including additions and deletions) can be
reverted using <command>svn revert</command>.
It does not update out-of-date
files—it just replaces them with pristine copies of
the original version.</para>
</sect3>
<sect3>
<title>Conflicts</title>
<para>If a <command>svn update</command> resulted in a merge
conflict, Subversion will remember which files have
conflicts and refuse to commit any changes to those files
until explicitly told that the conflicts have been resolved.
The simple, not yet deprecated procedure is the
following:</para>
<screen>&prompt.user; <userinput>svn resolved <replaceable>foo</replaceable></userinput></screen>
<para>However, the preferred procedure is:</para>
<screen>&prompt.user; <userinput>svn resolve --accept=working <replaceable>foo</replaceable></userinput></screen>
<para>The two examples are equivalent. Possible values for
<literal>--accept</literal> are:</para>
<itemizedlist>
<listitem>
<para><literal>working</literal>: use the version in your
working directory (which one presumes has been edited to
resolve the conflicts).</para>
</listitem>
<listitem>
<para><literal>base</literal>: use a pristine copy of the
version you had before <command>svn update</command>,
discarding your own changes, the conflicting changes,
and possibly other intervening changes as well.</para>
</listitem>
<listitem>
<para><literal>mine-full</literal>: use what you had
before <command>svn update</command>, including your own
changes, but discarding the conflicting changes, and
possibly other intervening changes as well.</para>
</listitem>
<listitem>
<para><literal>theirs-full</literal>: use the version that
was retrieved when you did <command>svn
update</command>, discarding your own changes.</para>
</listitem>
</itemizedlist>
</sect3>
</sect2>
<sect2>
<title>Advanced Use</title>
<sect3>
<title>Sparse Checkouts</title>
<para>The equivalent to <command>cvs checkout -l</command>,
which checks out a directory without its subdirectories, is
<command>svn checkout -N</command>. Unlike
<acronym>CVS</acronym>, <acronym>SVN</acronym> remembers the
<literal>-N</literal> so that a <command>svn
update</command> does not end up pulling down the
subdirectories. In Subversion 1.5 and newer,
<literal>-N</literal> has been deprecated in favour of the
<literal>--depth</literal> option which allows for precise
control. Therefore:</para>
<screen>&prompt.user; <userinput>svn checkout -N svn+ssh://svn.freebsd.org/base ~/freebsd</userinput></screen>
<para>is equivalent to:</para>
<screen>&prompt.user; <userinput>svn checkout --depth=empty svn+ssh://svn.freebsd.org/base ~/freebsd</userinput></screen>
<para>Valid arguments to <literal>--depth</literal>
are:</para>
<itemizedlist>
<listitem>
<para><literal>empty</literal>: the directory itself
without any of its contents.</para>
</listitem>
<listitem>
<para><literal>files</literal>: the directory and any
files it contains.</para>
</listitem>
<listitem>
<para><literal>immediates</literal>: the directory and any
files and directories it contains, but none of the
subdirectories' contents.</para>
</listitem>
<listitem>
<para><literal>infinity</literal>: anything.</para>
</listitem>
</itemizedlist>
<para>The <literal>--depth</literal> option applies to many
other commands, including <command>svn commit</command>,
<command>svn revert</command>, and <command>svn
diff</command>.</para>
<para>Since <literal>--depth</literal> is sticky, there is a
<literal>--set-depth</literal> option for <command>svn
update</command> that will change the selected depth.
Thus, given the working copy produced by the previous
example:</para>
<screen>&prompt.user; <userinput>cd <replaceable>~/freebsd</replaceable></userinput>
&prompt.user; <userinput>svn update --set-depth=immediates .</userinput></screen>
<para>The above command will populate the working copy in
<replaceable>~/freebsd</replaceable> with
<filename>ROADMAP.txt</filename> and empty subdirectories,
and nothing will happen when <command>svn update</command>
is executed on the subdirectories. However, the following
command will set the depth for head (in this case) to
infinity, and fully populate it:</para>
<screen>&prompt.user; <userinput>svn update --set-depth=infinity <replaceable>head</replaceable></userinput></screen>
</sect3>
<sect3>
<title>Direct Operation</title>
<para>Certain operations can be performed directly on the
repository, without touching the working copy.
Specifically, this applies to any operation that does not
require editing a file, including:</para>
<itemizedlist>
<listitem>
<para><literal>log</literal>,
<literal>diff</literal>.</para>
</listitem>
<listitem>
<para><literal>mkdir</literal>.</para>
</listitem>
<listitem>
<para><literal>remove</literal>, <literal>copy</literal>,
<literal>rename</literal>.</para>
</listitem>
<listitem>
<para><literal>propset</literal>,
<literal>propedit</literal>,
<literal>propdel</literal>.</para>
</listitem>
<listitem>
<para><literal>merge</literal>.</para>
</listitem>
</itemizedlist>
<para>Branching is very fast. The following command would be
used to branch <literal>RELENG_8</literal>:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/head svn+ssh://svn.freebsd.org/base/stable/8</userinput></screen>
<para>This is equivalent to the following set of
commands which take minutes and hours as opposed to seconds,
depending on your network connection:</para>
<screen>&prompt.user; <userinput>svn checkout --depth=immediates svn+ssh://svn.freebsd.org/base</userinput>
&prompt.user; <userinput>cd base</userinput>
&prompt.user; <userinput>svn update --depth=infinity head</userinput>
&prompt.user; <userinput>svn copy head stable/8</userinput>
&prompt.user; <userinput>svn commit stable/8</userinput></screen>
</sect3>
<sect3 id="subversion-primer-merge">
<title>Merging with <acronym>SVN</acronym></title>
<para>This section deals with merging code from one branch to
another (typically, from head to a stable branch).</para>
<note>
<para>In all examples below, <literal>$FSVN</literal>
refers to the location of the &os; Subversion repository,
<literal>svn+ssh://svn.freebsd.org/base/</literal>.</para>
</note>
<sect4>
<title>About Merge Tracking</title>
<para>From the user's perspective, merge tracking
information (or mergeinfo) is stored in a property called
<literal>svn:mergeinfo</literal>, which is a
comma-separated list of revisions and ranges of revisions
that have been merged. When set on a file, it applies
only to that file. When set on a directory, it applies to
that directory and its descendants (files and directories)
except for those that have their own
<literal>svn:mergeinfo</literal>.</para>
<para>It is <emphasis>not</emphasis> inherited. For
instance, <filename
class="directory">stable/6/contrib/openpam/</filename>
does not implicitly inherit mergeinfo from <filename
class="directory">stable/6/</filename>, or <filename
class="directory">stable/6/contrib/</filename>. Doing
so would make partial checkouts very hard to manage.
Instead, mergeinfo is explicitly propagated down the tree.
For merging something into <filename
class="directory">branch/foo/bar/</filename>, the
following rules apply:</para>
<orderedlist>
<listitem>
<para>If <filename
class="directory">branch/foo/bar/</filename> doesn't
already have a mergeinfo record, but a direct ancestor
(for instance, <filename
class="directory">branch/foo/</filename>) does,
then that record will be propagated down to
<filename class="directory">branch/foo/bar/</filename>
before information
about the current merge is recorded.</para>
</listitem>
<listitem>
<para>Information about the current merge will
<emphasis>not</emphasis> be propagated back up that
ancestor.</para>
</listitem>
<listitem>
<para>If a direct descendant of <filename
class="directory">branch/foo/bar/</filename> (for
instance, <filename
class="directory">branch/foo/bar/baz/</filename>)
already has a mergeinfo record, information about the
current merge will be propagated down to it.</para>
</listitem>
</orderedlist>
<para>If you consider the case where a revision changes
several separate parts of the tree (for example, <filename
class="directory">branch/foo/bar/</filename> and
<filename class="directory">branch/foo/quux/</filename>),
but you only want to merge some of it (for example,
<filename class="directory">branch/foo/bar/</filename>),
you will see that these rules make sense. If mergeinfo
was propagated up, it would seem like that revision had
also been merged to <filename
class="directory">branch/foo/quux/</filename>, when in
fact it had not been.</para>
</sect4>
<sect4>
<title>Selecting the Source and Target</title>
<para>Because of mergeinfo propagation, it is important to
choose the source and target for the merge carefully to
minimise property changes on unrelated directories.</para>
<para>The rules for selecting the merge target (the
directory that you will merge the changes to) can be
summarized as follows:</para>
<orderedlist>
<listitem>
<para>Never merge directly to a file.</para>
</listitem>
<listitem>
<para>Never, ever merge directly to a file.</para>
</listitem>
<listitem>
<para><emphasis>Never, ever, ever</emphasis> merge
directly to a file.</para>
</listitem>
<listitem>
<para>Changes to kernel code should be merged to
<filename class="directory">sys/</filename>. For
instance, a change to the &man.ichwd.4; driver should
be merged to <filename
class="directory">sys/</filename>, not <filename
class="directory">sys/dev/ichwd/</filename>.
Likewise, a change to the TCP/IP stack should be
merged to <filename class="directory">sys/</filename>,
not <filename
class="directory">sys/netinet/</filename>.</para>
</listitem>
<listitem>
<para>Changes to code under <filename
class="directory">etc/</filename> should be merged
at <filename class="directory">etc/</filename>, not
below it.</para>
</listitem>
<listitem>
<para>Changes to vendor code (code in <filename
class="directory">contrib/</filename>, <filename
class="directory">crypto/</filename> and so on)
should be merged to the directory where vendor imports
happen. For instance, a change to <filename
class="directory">crypto/openssl/util/</filename>
should be merged to <filename
class="directory">crypto/openssl/</filename>. This
is rarely an issue, however, since changes to vendor
code are usually merged wholesale.</para>
</listitem>
<listitem>
<para>Changes to userland programs should as a general
rule be merged to the directory that contains the
Makefile for that program. For instance, a change to
<filename
class="directory">usr.bin/xlint/arch/i386/</filename>
should be merged to <filename
class="directory">usr.bin/xlint/</filename>.</para>
</listitem>
<listitem>
<para>Changes to userland libraries should as a general
rule be merged to the directory that contains the
Makefile for that library. For instance, a change to
<filename class="directory">lib/libc/gen/</filename>
should be merged to <filename
class="directory">lib/libc/</filename>.</para>
</listitem>
<listitem>
<para>There may be cases where it makes sense to deviate
from the rules for userland programs and libraries.
For instance, everything under <filename
class="directory">lib/libpam/</filename> is merged
to <filename class="directory">lib/libpam/</filename>,
even though the library itself and all of the modules
each have their own Makefile.</para>
</listitem>
<listitem>
<para>Changes to manual pages should be merged to <filename
class="directory">share/man/man<replaceable>N</replaceable>/</filename>,
for the appropriate value of
<literal>N</literal>.</para>
</listitem>
<listitem>
<para>Other changes to <filename
class="directory">share/</filename> should be merged
to the appropriate subdirectory and not to
<filename class="directory">share/</filename>
directly.</para>
</listitem>
<listitem>
<para>Changes to a top-level file in the source tree
such as <filename>UPDATING</filename> or
<filename>Makefile.inc1</filename> should be merged
directly to that file rather than to the root of the
whole tree. Yes, this is an exception to the first
three rules.</para>
</listitem>
<listitem>
<para>When in doubt, ask.</para>
</listitem>
</orderedlist>
<para>If you need to merge changes to several places at once
(for instance, changing a kernel interface and every
userland program that uses it), merge each target
separately, then commit them together. For instance, if
you merge a revision that changed a kernel
<acronym>API</acronym> and updated all the userland bits
that used that <acronym>API</acronym>, you would merge the
kernel change to sys, and the userland bits to the
appropriate userland directories, then commit all of these
in one go.</para>
<para>The source will almost invariably be the same as the
target. For instance, you will always merge <filename
class="directory">stable/7/lib/libc/</filename> from
<filename class="directory">head/lib/libc/</filename>.
The only exception would be when merging changes to code
that has moved in the source branch but not in the parent
branch. For instance, a change to &man.pkill.1; would be
merged from <filename
class="directory">bin/pkill/</filename> in head to
<filename class="directory">usr.bin/pkill/</filename> in
stable/7.</para>
</sect4>
<sect4>
<title>Preparing the Merge Target</title>
<para>Because of the mergeinfo propagation issues described
earlier, it is very important that you never merge changes
into a sparse working copy. You must always have a full
checkout of the branch you will merge into. For instance,
when merging from HEAD to 7, you must have a full checkout
of stable/7:</para>
<screen>&prompt.user; <userinput>cd stable/7</userinput>
&prompt.user; <userinput>svn up --set-depth=infinity</userinput></screen>
<para>The target directory must also be up-to-date and must
not contain any uncommitted changes or stray files.</para>
</sect4>
<sect4>
<title>Identifying Revisions</title>
<para>Identifying revisions to be merged is a must. If the
target already has complete mergeinfo, ask
<acronym>SVN</acronym> for a list:</para>
<screen>&prompt.user; <userinput>cd stable/6/contrib/openpam</userinput>
&prompt.user; <userinput>svn mergeinfo --show-revs=eligible $FSVN/head/contrib/openpam</userinput></screen>
<para>If the target does not have complete mergeinfo, check
the log for the merge source.</para>
</sect4>
<sect4>
<title>Merging</title>
<para>Now, let's start merging!</para>
<sect5>
<title>The Principles</title>
<para>Say you would like to merge:</para>
<itemizedlist>
<listitem>
<para>revision <literal>$R</literal>.</para>
</listitem>
<listitem>
<para>in directory $target in stable branch
$B.</para>
</listitem>
<listitem>
<para>from directory $source in head.</para>
</listitem>
<listitem>
<para>$FSVN is
<literal>svn+ssh://svn.freebsd.org/base</literal>.</para>
</listitem>
</itemizedlist>
<para>Assuming that revisions $P and $Q have
already been merged, and that the current directory is
an up-to-date working copy of stable/$B, the
existing mergeinfo looks like this:</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R $target</userinput>
$target - /head/$source:$P,$Q</screen>
<para>Merging is done like so:</para>
<screen>&prompt.user; <userinput>svn merge -c$R $FSVN/head/$source $target</userinput></screen>
<para>Checking the results of this is possible with
<command>svn diff</command>.</para>
<para>The svn:mergeinfo now looks like:</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R $target</userinput>
$target - head/$source:$P,$Q,$R</screen>
<para>If the results are not exactly as shown, assistance
may be required before committing as mistakes may have
been made, or there may be something wrong with the
existing mergeinfo, or there may be a bug in
Subversion.</para>
</sect5>
<sect5>
<title>Practical Example</title>
<para>As a practical example, consider the following scenario:
The changes to <filename>netmap.4</filename> in r238987 is
to be merged from CURRENT to 9-STABLE. The file resides in
<filename class="directory">head/share/man/man4</filename> and
according to <xref linkend="subversion-primer-merge"/> this
is also where to do the merge. Note that in this example
all paths are relative to the top of the svn repository.
for more information on the directory layout, see
<xref linkend="subversion-primer-base-layout"/>.</para>
<para>The first step is to inspect the existing mergeinfo.</para>
<screen>&prompt.user; <userinput>svn propget svn:mergeinfo -R stable/9/share/man/man4</userinput></screen>
<para>Take a quick note of how it looks before moving on to the next
step; doing the actual merge:</para>
<screen>&prompt.user; <userinput>svn merge -c r238987 svn+ssh://svn.freebsd.org/base/head/share/man/man4 stable/9/share/man/man4</userinput>
--- Merging r238987 into 'stable/9/share/man/man4':
U stable/9/share/man/man4/netmap.4
--- Recording mergeinfo for merge of r238987 into
'stable/9/share/man/man4':
U stable/9/share/man/man4</screen>
<para>Check that the revision number of the merged revision
has been added. Once this is verified, the only thing left
is the actual commit.</para>
<screen>&prompt.user; <userinput>svn commit stable/9/share/man/man4</userinput></screen>
</sect5>
<sect5>
<title>Merging into the Kernel
(<filename class="directory">sys/</filename>)</title>
<para>As stated above, merging into the kernel is
different from merging in the rest of the tree. In many
ways merging to the kernel is simpler because there is
always the same merge target
(<filename class="directory">sys/</filename>).</para>
<para>Once <command>svn merge</command> has been executed,
<command>svn diff</command> has to be run on the
directory to check the changes. This may show some
unrelated property changes, but these can be ignored.
Next, build and test the kernel, and, once the tests are
complete, commit the code as normal, making sure that
the commit message starts with <quote>Merge
<replaceable>r226222</replaceable> from head</quote>,
or similar.</para>
</sect5>
</sect4>
<sect4>
<title>Precautions Before Committing</title>
<para>As always, build world (or appropriate parts of
it).</para>
<para>Check the changes with <command>svn diff</command> and
<command>svn stat</command>. Make sure all the files that
should have been added or deleted were in fact added or
deleted.</para>
<para>Take a closer look at any property change (marked by a
<literal>M</literal> in the second column of <command>svn
stat</command>). Normally, no svn:mergeinfo properties
should be anywhere except the target directory (or
directories).</para>
<para>If something looks fishy, ask for help.</para>
</sect4>
<sect4>
<title>Committing</title>
<para>Make sure to commit a top level directory to have the
mergeinfo included as well. Do not specify individual
files on the command line. For more information about
committing files in general, see the relevant section of
this primer.</para>
</sect4>
</sect3>
<sect3>
<title>Vendor imports with <acronym>SVN</acronym></title>
<important>
<para>Please read this entire section before starting a vendor
import.</para>
</important>
<note>
<para>Patches to vendor code fall into two categories:</para>
<itemizedlist>
<listitem>
<para>Vendor patches: these are patches that have been
issued by the vendor, or that have been extracted from
the vendor's version control system, which address
issues which in your opinion can't wait until the next
vendor release.</para>
</listitem>
<listitem>
<para>&os; patches: these are patches that modify the
vendor code to address &os;-specific issues.</para>
</listitem>
</itemizedlist>
<para>The nature of a patch dictates where it should be
committed:</para>
<itemizedlist>
<listitem>
<para>Vendor patches should be committed to the vendor
branch, and merged from there to head. If the patch
addresses an issue in a new release that is currently
being imported, it <emphasis>must not</emphasis> be
committed along with the new release: the release must
be imported and tagged first, then the patch can be
applied and committed. There is no need to re-tag the
vendor sources after committing the patch.</para>
</listitem>
<listitem>
<para>&os; patches should be committed directly to
head.</para>
</listitem>
</itemizedlist>
</note>
<sect4>
<title>Preparing the tree</title>
<para>If importing for the first time after the switch to
Subversion, flattening and cleaning up the vendor tree is
necessary, as well as bootstrapping the merge history in
the main tree.</para>
<sect5>
<title>Flattening</title>
<para>During the conversion from <acronym>CVS</acronym> to
Subversion, vendor branches were imported with the same
layout as the main tree. This means that the
<literal>pf</literal> vendor sources ended up in
<filename>vendor/pf/dist/contrib/pf</filename>. The
vendor source is best directly in
<filename>vendor/pf/dist</filename>.</para>
<para>To flatten the <literal>pf</literal> tree:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/dist/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn mv $(svn list) ../..</userinput>
&prompt.user; <userinput>cd ../..</userinput>
&prompt.user; <userinput>svn rm contrib</userinput>
&prompt.user; <userinput>svn propdel -R svn:mergeinfo .</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>The <literal>propdel</literal> bit is necessary
because starting with 1.5, Subversion will automatically
add <literal>svn:mergeinfo</literal> to any directory
that is copied or moved. In this case, as nothing is
being merged from the deleted tree, they just get in the
way.</para>
<para>Tags may be flattened as well (3, 4, 3.5 etc.); the
procedure is exactly the same, only changing
<literal>dist</literal> to <literal>3.5</literal> or
similar, and putting the <command>svn commit</command>
off until the end of the process.</para>
</sect5>
<sect5>
<title>Cleaning up</title>
<para>The <literal>dist</literal> tree can be cleaned up
as necessary. Disabling keyword expansion is
recommended, as it makes no sense on unmodified vendor
code and in some cases it can even be harmful.
<application>OpenSSH</application>, for example, includes
two files that originated with &os; and still contain the
original version tags. To do this:</para>
<screen>&prompt.user; <userinput>svn propdel svn:keywords -R .</userinput>
&prompt.root; <userinput>svn commit</userinput></screen>
</sect5>
<sect5>
<title>Bootstrapping merge history</title>
<para>If importing for the first time after the switch to
Subversion, bootstrap
<literal>svn:mergeinfo</literal> on the target directory
in the main tree to the revision that corresponds
to the last related change to the vendor tree, prior to
importing new sources:</para>
<screen>&prompt.user; <userinput>cd <replaceable>head/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn merge --record-only svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist@180876</replaceable> .</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
</sect5>
</sect4>
<sect4>
<title>Importing new sources</title>
<para>With two commits—one for the import itself and
one for the tag—this step can optionally be repeated
for every upstream release between the last import and the
current import.</para>
<sect5>
<title>Preparing the vendor sources</title>
<para>Unlike in <acronym>CVS</acronym> where only the needed
parts were imported into the vendor tree to avoid bloating
the main tree, Subversion is able to store a full
distribution in the vendor tree. So, import everything,
but merge only what is required.</para>
<para>A <command>svn add</command> is required to add any
files that were added since the last vendor import, and
<command>svn rm</command> is required to remove any that
were removed since. Preparing sorted lists of the
contents of the vendor tree and of the sources that are
about to be imported is recommended, to facilitate the
process.</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/dist</replaceable></userinput>
&prompt.user; <userinput>svn list -R | grep -v '/$' | sort >../old</userinput>
&prompt.user; <userinput>cd <replaceable>../pf-4.3</replaceable></userinput>
&prompt.user; <userinput>find . -type f | cut -c 3- | sort >../new</userinput></screen>
<para>With these two files,
<command>comm -23 ../old ../new</command>
will list removed files (files only in
<filename>old</filename>), while
<command>comm -13 ../old ../new</command>
will list added files only in <filename>new</filename>.</para>
</sect5>
<sect5>
<title>Importing into the vendor tree</title>
<para>Now, the sources must be copied into
<filename><replaceable>dist</replaceable></filename> and
the <command>svn add</command> and
<command>svn rm</command> commands should be used as
needed:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf/pf-4.3</replaceable></userinput>
&prompt.user; <userinput>tar cf - . | tar xf - -C ../dist</userinput>
&prompt.user; <userinput>cd <replaceable>../dist</replaceable></userinput>
&prompt.user; <userinput>comm -23 ../old ../new | xargs svn rm</userinput>
&prompt.user; <userinput>comm -13 ../old ../new | xargs svn --parents add</userinput></screen>
<para>If any directories were removed, they will have to be
<command>svn rm</command>ed manually. Nothing will break
if they are not, but they will remain in the tree.</para>
<para>Check properties on any new files. All text files
should have <literal>svn:eol-style</literal> set to
<literal>native</literal>. All binary files should have
<literal>svn:mime-type</literal> set to
<literal>application/octet-stream</literal> unless there
is a more appropriate media type. Executable files should
have <literal>svn:executable</literal> set to
<literal>*</literal>. No other properties should exist
on any file in the tree.</para>
<para>Committing is now possible, however it is good
practice to make sure that everything is OK by using the
<command>svn stat</command> and
<command>svn diff</command> commands.</para>
</sect5>
<sect5>
<title>Tagging</title>
<para>Once committed, vendor releases should be tagged for
future reference. The best and quickest way to do this
is directly in the repository:</para>
<screen>&prompt.user; <userinput>svn cp svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/4.3</replaceable></userinput></screen>
<para>Once that is complete, <command>svn up</command> the
working copy of
<filename><replaceable>vendor/pf</replaceable></filename>
to get the new tag, although this is rarely
needed.</para>
<para>If creating the tag in the working copy of the tree,
<command>svn:mergeinfo</command> results must be removed:</para>
<screen>&prompt.user; <userinput>cd <replaceable>vendor/pf</replaceable></userinput>
&prompt.user; <userinput>svn cp dist 4.3</userinput>
&prompt.user; <userinput>svn propdel svn:mergeinfo -R 4.3</userinput></screen>
</sect5>
</sect4>
<sect4>
<title>Merging to head</title>
<screen>&prompt.user; <userinput>cd <replaceable>head/contrib/pf</replaceable></userinput>
&prompt.user; <userinput>svn up</userinput>
&prompt.user; <userinput>svn merge --accept=postpone svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> .</userinput></screen>
<para>The <literal>--accept=postpone</literal> tells
Subversion that it shouldn't complain because merge conflicts
will be taken care of manually.</para>
<para>It is necessary to resolve any merge conflicts.
This process is the same in <acronym>SVN</acronym> as in
<acronym>CVS</acronym>.</para>
<para>Make sure that any files that were added or removed in
the vendor tree have been properly added or removed in the
main tree. To check diffs against the vendor branch:</para>
<screen>&prompt.user; <userinput>svn diff --no-diff-deleted --old=svn+ssh://svn.freebsd.org/base/<replaceable>vendor/pf/dist</replaceable> --new=.</userinput></screen>
<para>The <literal>--no-diff-deleted</literal> tells
Subversion not to complain about files that are in the
vendor tree but not in the main tree, i.e., things that
would have previously been removed before the vendor
import, like for example the like the vendor's makefiles
and configure scripts.</para>
<para>Using <acronym>CVS</acronym>, once a file was off the
vendor branch, it was not able to be put back. With
Subversion, there is no concept of on or off the vendor
branch. If a file that previously had local
modifications, to make it not show up in diffs in the
vendor tree, all that has to be done is remove any left-over
cruft like &os; version tags, which is much easier.</para>
<para>If any changes are required for the world to build
with the new sources, make them now, and keep testing
until everything builds and runs perfectly.</para>
</sect4>
<sect4>
<title>Committing the vendor import</title>
<para>Committing is now possible! Everything must be
committed in one go. If done properly, the tree will move
from a consistent state with old code, to a consistent
state with new code.</para>
</sect4>
<sect4>
<title>From scratch</title>
<sect5>
<title>Importing into the vendor tree</title>
<para>This section is an example of importing and tagging
<application>byacc</application> into
<filename class="directory">head</filename>.</para>
<para>First, prepare the directory in
<filename class="directory">vendor</filename>:</para>
<screen>&prompt.user; <userinput>svn co --depth immediates <replaceable>$FSVN/vendor</replaceable></userinput>
&prompt.user; <userinput>cd <replaceable>vendor</replaceable></userinput>
&prompt.user; <userinput>svn mkdir <replaceable>byacc</replaceable></userinput>
&prompt.user; <userinput>svn mkdir <replaceable>byacc/dist</replaceable></userinput></screen>
<para>Now, import the sources into the
<filename class="directory">dist</filename> directory. Once
the files are in place, <command>svn add</command> the new
ones, then <command>svn commit</command> and tag the
imported version. To save time and bandwidth, direct remote
committing and tagging is possible:</para>
<screen>&prompt.user; <userinput>svn cp -m <replaceable>"Tag byacc 20120115"</replaceable> <replaceable>$FSVN/vendor/byacc/dist</replaceable> <replaceable>$FSVN/vendor/byacc/20120115</replaceable></userinput></screen>
</sect5>
<sect5>
<title>Merging to head</title>
<para>Due to this being a new file, copy it for the
merge:</para>
<screen>&prompt.user; <userinput>svn cp -m <replaceable>"Import byacc to contrib"</replaceable> <replaceable>$FSVN/vendor/byacc/dist</replaceable> <replaceable>$FSVN/head/contrib/byacc</replaceable></userinput></screen>
<para>Working normally on newly imported sources is still
possible.</para>
</sect5>
</sect4>
</sect3>
<sect3>
<title>Reverting a Commit</title>
<para>Reverting a commit to a previous version is fairly
easy:</para>
<screen>&prompt.user; <userinput>svn merge -r179454:179453 ROADMAP.txt</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>Change number syntax, with negative meaning a reverse
change, can also be used:</para>
<screen>&prompt.user; <userinput>svn merge -c -179454 ROADMAP.txt</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>This can also be done directly in the repository:</para>
<screen>&prompt.user; <userinput>svn merge -r179454:179453 svn+ssh://svn.freebsd.org/base/ROADMAP.txt</userinput></screen>
<note>
<para>It is important to ensure that the mergeinfo
is correct when reverting a file in order to permit
<command>svn mergeinfo --eligible</command> to work as
expected.</para>
</note>
<para>Reverting the deletion of a file is slightly different.
Copying the version of the file that predates the deletion
is required. For example, to restore a file that was
deleted in revision N, restore version N-1:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/ROADMAP.txt@179454</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>or, equally:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/ROADMAP.txt@179454 svn+ssh://svn.freebsd.org/base</userinput></screen>
<para>Do <emphasis>not</emphasis> simply recreate the file
manually and <command>svn add</command> it—this will
cause history to be lost.</para>
</sect3>
<sect3>
<title>Fixing Mistakes</title>
<para>While we can do surgery in an emergency, do not plan on
having mistakes fixed behind the scenes. Plan on mistakes
remaining in the logs forever. Be sure to check the output
of <command>svn status</command> and <command>svn
diff</command> before committing.</para>
<para>Mistakes will happen but,
they can generally be fixed without
disruption.</para>
<para>Take a case of adding a file in the wrong location. The
right thing to do is to <command>svn move</command> the file
to the correct location and commit. This causes just a
couple of lines of metadata in the repository journal, and
the logs are all linked up correctly.</para>
<para>The wrong thing to do is to delete the file and then
<command>svn add</command> an independent copy in the
correct location. Instead of a couple of lines of text, the
repository journal grows an entire new copy of the file.
This is a waste.</para>
</sect3>
<sect3>
<title>Setting up a <application>svnsync</application>
Mirror</title>
<para>You probably do not want to do this unless there is a
good reason for it. Such reasons might be to support many
multiple local read-only client machines, or if your network
bandwidth is limited. Starting a fresh mirror from empty
would take a very long time. Expect a minimum of 10 hours
for high speed connectivity. If you have international
links, expect this to take 4 to 10 times longer.</para>
<para>A far better option is to grab a seed file. It is large
(~1GB) but will consume less network traffic and take less
time to fetch than a svnsync will. This is possible in one
of the following three ways:</para>
<screen>&prompt.user; <userinput>rsync -va --partial --progress freefall:/home/peter/svnmirror-base-r179637.tbz2 .</userinput></screen>
<screen>&prompt.user; <userinput>rsync -va --partial --progress rsync://repoman.freebsd.org:50873/svnseed/svnmirror-base-r215629.tar.xz .</userinput></screen>
<screen>&prompt.user; <userinput>fetch ftp://ftp.freebsd.org/pub/FreeBSD/development/subversion/svnmirror-base-r221445.tar.xz</userinput></screen>
<para>Once you have the file, extract it to somewhere like
<filename class="directory">home/svnmirror/base/</filename>.
Then, update it, so that it fetches changes since the last
revision in the archive:</para>
<screen>&prompt.user; <userinput>svnsync sync file:///home/svnmirror/base</userinput></screen>
<para>You can then set that up to run from &man.cron.8;, do
checkouts locally, set up a svnserve server for your local
machines to talk to, etc.</para>
<para>The seed mirror is set to fetch from
<literal>svn://svn.freebsd.org/base</literal>. The
configuration for the mirror is stored in <literal>revprop
0</literal> on the local mirror. To see the
configuration, try:</para>
<screen>&prompt.user; <userinput>svn proplist -v --revprop -r 0 file:///home/svnmirror/base</userinput></screen>
<para>Use <literal>propset</literal> to change things.</para>
</sect3>
<sect3>
<title>Committing High-<acronym>ASCII</acronym> Data</title>
<para>Files that have high-<acronym>ASCII</acronym> bits are
considered binary files in <acronym>SVN</acronym>, so the
pre-commit checks fail and indicate that the
<literal>mime-type</literal> property should be set to
<literal>application/octet-stream</literal>. However, the
use of this is discouraged, so please do not set it. The
best way is always avoiding high-<acronym>ASCII</acronym>
data, so that it can be read everywhere with any text editor
but if it is not avoidable, instead of changing the
mime-type, set the <literal>fbsd:notbinary</literal>
property with <literal>propset</literal>:</para>
<screen>&prompt.user; <userinput>svn propset fbsd:notbinary yes foo.data</userinput></screen>
</sect3>
<sect3>
<title>Maintaining a Project Branch</title>
<para>A project branch is one that's synced to head (or
another branch) is used to develop a project then commit it
back to head. In <acronym>SVN</acronym>,
<quote>dolphin</quote> branching is used for this. A
<quote>dolphin</quote> branch is one that diverges for a
while and is finally committed back to the original branch.
During development code migration in one direction (from
head to the branch only). No code is committed back to head
until the end. Once you commit back at the end, the branch
is dead (although you can have a new branch with the same
name after you delete the branch if you want).</para>
<para>As per <ulink
url="http://people.freebsd.org/~peter/svn_notes.txt">http://people.freebsd.org/~peter/svn_notes.txt</ulink>,
work that is intended to be merged back into HEAD should be
in <filename class="directory">base/projects/</filename>.
If you are doing work that is beneficial to the &os;
community in some way but not intended to be merged directly
back into HEAD then the proper location is <filename
class="directory">base/user/<replaceable>your-name</replaceable>/</filename>.
<ulink
url="http://svnweb.freebsd.org/base/projects/GUIDELINES.txt">This
page</ulink> contains further details.</para>
<para>To create a project branch:</para>
<screen>&prompt.user; <userinput>svn copy svn+ssh://svn.freebsd.org/base/head svn+ssh://svn.freebsd.org/base/projects/spif</userinput></screen>
<para>To merge changes from HEAD back into the project
branch:</para>
<screen>&prompt.user; <userinput>cd copy_of_spif</userinput>
&prompt.user; <userinput>svn merge svn+ssh://svn.freebsd.org/base/head</userinput>
&prompt.user; <userinput>svn commit</userinput></screen>
<para>It is important to resolve any merge conflicts before
committing.</para>
<!--
<para>To collapse everything back at the end:</para>
<screen>&prompt.user; <userinput>svn write me</userinput></screen>
-->
</sect3>
</sect2>
<sect2>
<title>Some Tips</title>
<para>In commit logs etc., <quote>rev 179872</quote> should be
spelled <quote>r179872</quote> as per convention.</para>
<para>Don't remove and re-add the same file in a single commit
as this will break the CVS exporter.</para>
<para>Speeding up svn is
possible by adding the following to <filename>~/.ssh/config</filename>:</para>
<screen>Host *
ControlPath ~/.ssh/sockets/master-%l-%r@%h:%p
ControlMaster auto
ControlPersist yes</screen>
<para>and then typing</para>
<screen><userinput>mkdir ~/.ssh/sockets</userinput></screen>
<para>Checking out a working copy with a stock Subversion client
without &os;-specific patches
(<makevar>OPTIONS_SET=FREEBSD_TEMPLATE</makevar>) will mean that
<literal>$FreeBSD$</literal> tags will not be
expanded. Once the correct version has been installed, trick
Subversion into expanding them like so:</para>
<screen>&prompt.user; <userinput>svn propdel -R svn:keywords .</userinput>
&prompt.user; <userinput>svn revert -R .</userinput></screen>
<para>This will wipe out uncommitted patches.</para>
</sect2>
</sect1>
<sect1 id="conventions">
<title>Conventions and Traditions</title>
<para>As a new developer there are a number of things you should do
first. The first set is specific to committers only. (If you are
not a committer, e.g., have GNATS-only access, then your mentor needs
to do these things for you.)</para>
<sect2 id="conventions-committers">
<title>Guidelines For Committers</title>
<note>
<para>The <literal>.ent</literal>, <literal>.xml</literal>,
and <literal>.xml</literal> files listed below exist in the
&os; Documentation Project SVN repository at
<hostid role="fqdn">svn.FreeBSD.org/doc/</hostid>.</para>
</note>
<para>If you have been given commit rights to one or more of the
repositories:</para>
<itemizedlist>
<listitem>
<para>Add your author entity to
<filename>head/share/xml/authors.ent</filename>;
this should be done first since an omission of this commit will
cause the next commits to break the doc/ build.</para>
<para>This is a relatively easy task, but remains a good first test of
your version control skills.</para>
<important>
<para>New files that do not have the
<literal>FreeBSD=%H</literal> <command>svn:keywords</command>
property will be rejected when attempting to commit them to the
repository. Be sure to read <xref
linkend="subversion-primer-add-remove"/>
regarding adding and removing files, in addition
to verifying that <filename>~/.subversion/config</filename>
contains the necessary "auto-props"
entries from <filename>auto-props.txt</filename> mentioned
there.</para>
</important>
<note>
<para>Don't forget to get mentor approval for these patches!</para>
</note>
</listitem>
<listitem>
<para>Also add your author entity to
<filename>head/share/xml/developers.ent</filename>.</para>
</listitem>
<listitem>
<para>Add yourself to the <quote>Developers</quote> section of
the <ulink url="&url.articles.contributors;/index.html">Contributors List</ulink>
(<filename>head/en_US.ISO8859-1/articles/contributors/contrib.committers.xml</filename>) and remove yourself from the <quote>Additional
Contributors</quote> section (<filename>head/en_US.ISO8859-1/articles/contributors/contrib.additional.xml</filename>).
Please note that entries are sorted by last name.</para>
</listitem>
<listitem>
<para>Add an entry for yourself to
<filename>head/share/xml/news.xml</filename>. Look for the other
entries that look like <quote>A new committer</quote> and follow the
format.</para>
</listitem>
<listitem>
<para>You should add your PGP or GnuPG key to
<filename>head/share/pgpkeys</filename> (and if you do not
have a key, you should create one). Do not forget to commit
the updated <filename>head/share/pgpkeys/pgpkeys.ent</filename>
and <filename>head/share/pgpkeys/pgpkeys-developers.xml</filename>.
Please note that entries are sorted by last name.</para>
<para>&a.des; has
written a shell script (<filename>head/share/pgpkeys/addkey.sh</filename>) to make this extremely simple. See the
<ulink
url="http://svnweb.FreeBSD.org/doc/head/share/pgpkeys/README">README</ulink>
file for more information.</para>
<note>
<para>It is important to have an up-to-date PGP/GnuPG key in
the Handbook, since the key may be required for positive
identification of a committer, e.g., by the &a.admins; for
account recovery. A complete keyring of <hostid
role="domainname">FreeBSD.org</hostid> users is available
for download from <ulink
url="&url.base;/doc/pgpkeyring.txt">http://www.FreeBSD.org/doc/pgpkeyring.txt</ulink>.</para>
</note>
</listitem>
<listitem>
<para>Add an entry for yourself to
<filename>src/share/misc/committers-<replaceable>repository</replaceable>.dot</filename>,
where repository is either doc, ports or src, depending on the commit privileges
you obtained.</para>
</listitem>
<listitem>
<para>Some people add an entry for themselves to
<filename>ports/astro/xearth/files/freebsd.committers.markers</filename>.</para>
</listitem>
<listitem>
<para>Some people add an entry for themselves to
<filename>src/usr.bin/calendar/calendars/calendar.freebsd</filename>.</para>
</listitem>
<listitem>
<para>If you already have an account at the <ulink url="http://wiki.freebsd.org">&os; wiki</ulink>,
make sure your mentor moves you from the
<ulink url="http://wiki.freebsd.org/ContributorsGroup">Contributors group</ulink>
to the
<ulink url="http://wiki.freebsd.org/DevelopersGroup">Developers group</ulink>.
Otherwise, consider signing up for an account so you can publish
projects and ideas you are working on.</para>
</listitem>
<listitem>
<para>Once you get access to the wiki, you may
add yourself to the <ulink
url="http://wiki.freebsd.org/HowWeGotHere">How We Got
Here</ulink> and <ulink
url="http://wiki.freebsd.org/IrcNicks">Irc Nicks</ulink>
pages.</para>
</listitem>
<listitem>
<para>If you subscribe to &a.svn-src-all.name;,
&a.svn-ports-all.name; or &a.svn-doc-all.name;,
you will probably want to unsubscribe to avoid receiving duplicate
copies of commit messages and their followups.</para>
</listitem>
</itemizedlist>
<note>
<para>All <filename>src</filename> commits should go to
&os.current; first before being merged to &os.stable;. No major
new features or high-risk modifications should be made to the
&os.stable; branch.</para>
</note>
</sect2>
<sect2 id="conventions-everyone">
<title>Guidelines For Everyone</title>
<para>Whether or not you have commit rights:</para>
<itemizedlist>
<listitem>
<para>Introduce yourself to the other developers, otherwise no one
will have any idea who you are or what you are working on. You do
not have to write a comprehensive biography, just write a paragraph
or two about who you are and what you plan to be working on as a
developer in FreeBSD. (You should also mention who your mentor
will be). Email this to the &a.developers; and you will
be on your way!</para>
</listitem>
<listitem>
<para>Log into <hostid>hub.FreeBSD.org</hostid> and create a
<filename>/var/forward/<replaceable>user</replaceable></filename>
(where <replaceable>user</replaceable> is your username) file
containing the e-mail address where you want mail addressed to
<replaceable>yourusername</replaceable>@FreeBSD.org to be forwarded.
This includes all of the commit messages as well as any other mail
addressed to the &a.committers; and the &a.developers;. Really
large mailboxes which have taken up permanent residence on
<hostid>hub</hostid> often get <quote>accidentally</quote> truncated
without warning, so forward it or read it and you will not lose
it.</para>
<para>Due to the severe load dealing with SPAM places on
the central mail servers that do the mailing list processing
the front-end server does do some basic checks and will
drop some messages based on these checks. At the moment
proper DNS information for the connecting host is the only
check in place but that may change. Some people blame these
checks for bouncing valid email. If you want these checks
turned off for your email you can place a file named
<filename>.spam_lover</filename> in your home directory
on <hostid role="fqdn">freefall.FreeBSD.org</hostid> to
disable the checks for your email.</para>
</listitem>
</itemizedlist>
<note>
<para>If you are a developer but not a committer, you will
not be subscribed to the committers or developers mailing lists;
the subscriptions are derived from the access rights.</para>
</note>
</sect2>
<sect2 id="mentors">
<title>Mentors</title>
<para>All new developers also have a mentor assigned to them for
the first few months. Your mentor is responsible for teaching
you the rules and conventions of the project and guiding your
first steps in the developer community. Your mentor is also
personally responsible for your actions during this initial
period.</para>
<para>For committers: until your
mentor decides (and announces with a forced
commit to <filename>access</filename>) that you have learned the
ropes and are ready to commit on your own, you should not commit
anything without first getting your mentor's review and
approval, and you should document that approval with an
<literal>Approved by:</literal> line in the commit
message.</para>
</sect2>
</sect1>
<sect1 id="pref-license">
<title>Preferred License for New Files</title>
<para>Currently the &os; Project suggests and uses the following
text as the preferred license scheme:</para>
<programlisting>/*-
* Copyright (c) [year] [your name]
* All rights reserved.
*
* Redistribution and use in source and binary forms, with or without
* modification, are permitted provided that the following conditions
* are met:
* 1. Redistributions of source code must retain the above copyright
* notice, this list of conditions and the following disclaimer.
* 2. Redistributions in binary form must reproduce the above copyright
* notice, this list of conditions and the following disclaimer in the
* documentation and/or other materials provided with the distribution.
*
* THIS SOFTWARE IS PROVIDED BY THE AUTHOR AND CONTRIBUTORS ``AS IS'' AND
* ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, THE
* IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS FOR A PARTICULAR PURPOSE
* ARE DISCLAIMED. IN NO EVENT SHALL THE AUTHOR OR CONTRIBUTORS BE LIABLE
* FOR ANY DIRECT, INDIRECT, INCIDENTAL, SPECIAL, EXEMPLARY, OR CONSEQUENTIAL
* DAMAGES (INCLUDING, BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS
* OR SERVICES; LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION)
* HOWEVER CAUSED AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT
* LIABILITY, OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY
* OUT OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
* SUCH DAMAGE.
*
* [id for your version control system, if any]
*/</programlisting>
<para>The &os; project strongly discourages the so-called
"advertising clause" in new code. Due to the large number of
contributors to the &os; project, complying with this clause for
many commercial vendors has become difficult. If you have code
in the tree with the advertising clause, please consider
removing it. In fact, please consider using the above license
for your code.</para>
<para>The &os; project discourages completely new licenses and
variations on the standard licenses. New licenses require the
approval of <email>core@FreeBSD.org</email> to reside in the
main repository. The more different licenses that are used in
the tree, the more problems that this causes to those wishing to
utilize this code, typically from unintended consequences from a
poorly worded license.</para>
<para>Project policy dictates that code under some non-BSD licenses
must be placed only in specific sections of the repository, and
in some cases, compilation must be conditional or even disabled
by default. For example, the GENERIC kernel must be compiled
under only licenses identical to or substantially similar to the
BSD license. GPL, APSL, CDDL, etc, licensed software must not be
compiled into GENERIC.</para>
<para>Developers are reminded that in open source, getting "open"
right is just as important as getting "source" right, as improper
handling of intellectual property has serious consequences. Any
questions or concerns should immediately be brought to the
attention of the core team.</para>
</sect1>
<sect1 id="developer.relations">
<title>Developer Relations</title>
<para>If you are working directly on your own code or on code
which is already well established as your responsibility, then
there is probably little need to check with other committers
before jumping in with a commit. If you see a bug in an area of
the system which is clearly orphaned (and there are a few such
areas, to our shame), the same applies. If, however, you are
about to modify something which is clearly being actively
maintained by someone else (and it is only by watching the
<literal><replaceable>repository</replaceable>-committers</literal> mailing list that you can
really get a feel for just what is and is not) then consider
sending the change to them instead, just as you would have
before becoming a committer. For ports, you should contact the
listed <makevar>MAINTAINER</makevar> in the
<filename>Makefile</filename>. For other parts of the
repository, if you are unsure who the active maintainer might
be, it may help to scan the revision history
to see who has committed changes in the past. &a.fenner; has
written a nice shell script that can help determine who the
active maintainer might be. It lists each person who has
committed to a given file along with the number of commits each
person has made. It can be found on <hostid>freefall</hostid>
at <filename>~fenner/bin/whodid</filename>. If your queries go
unanswered or the committer otherwise indicates a lack of
interest in the area affected, go ahead and commit
it.</para>
<para>If you are unsure about a commit for any reason at
all, have it reviewed by <literal>-hackers</literal>
before committing. Better to have it flamed then and there
rather than when it is part of the repository. If you do
happen to commit something which results in controversy
erupting, you may also wish to consider backing the change out
again until the matter is settled. Remember – with a version control system we
can always change it back.</para>
<para>Do not impugn the intentions of someone you disagree with.
If they see a different solution to a problem than you, or even
a different problem, it is not because they are stupid, because
they have questionable parentage, or because they are trying to
destroy your hard work, personal image, or FreeBSD, but simply
because they have a different outlook on the world. Different
is good.</para>
<para>Disagree honestly. Argue your position from its merits,
be honest about any shortcomings it may have, and be open to
seeing their solution, or even their vision of the problem,
with an open mind.</para>
<para>Accept correction. We are all fallible. When you have made
a mistake, apologize and get on with life. Do not beat up
yourself, and certainly do not beat up others for your mistake.
Do not waste time on embarrassment or recrimination, just fix
the problem and move on.</para>
<para>Ask for help. Seek out (and give) peer reviews. One of
the ways open source software is supposed to excel is in the
number of eyeballs applied to it; this does not apply if nobody
will review code.</para>
</sect1>
<sect1 id="gnats">
<title>GNATS</title>
<para>The FreeBSD Project utilizes
<application>GNATS</application> for tracking bugs and change
requests. Be sure that if you commit a fix or suggestion found
in a <application>GNATS</application> PR, you use
<command>edit-pr <replaceable>pr-number</replaceable></command>
on <hostid>freefall</hostid> to close it. It is also considered
nice if you take time to close any PRs associated with your
commits, if appropriate. You can also make use of
&man.send-pr.1; yourself for proposing any change which you feel
should probably be made, pending a more extensive peer-review
first.</para>
<para>You can find out more about <application>GNATS</application>
at:</para>
<itemizedlist>
<listitem>
<para><ulink url="&url.articles.pr-guidelines;/index.html">FreeBSD Problem Report Handling Guidelines</ulink></para>
</listitem>
<listitem>
<para><ulink url="http://www.cs.utah.edu/csinfo/texinfo/gnats/gnats.html"></ulink></para>
</listitem>
<listitem>
<para><ulink url="&url.base;/support.html">http://www.FreeBSD.org/support.html</ulink></para>
</listitem>
<listitem>
<para>&man.send-pr.1;</para>
</listitem>
</itemizedlist>
<para>You can run a local copy of GNATS, and then integrate the FreeBSD
GNATS tree by creating an <application>rsync</application> mirror.
Then you can run GNATS commands locally, allowing you to query the PR
database without an Internet connection.</para>
<sect2>
<title>Mirroring the GNATS Tree</title>
<para>It is possible to mirror the GNATS database by installing
<filename role="package">net/rsync</filename>, and
executing:</para>
<screen>&prompt.user; <userinput>rsync -va rsync://bit0.us-west.freebsd.org/FreeBSD-bit/gnats .</userinput></screen>
</sect2>
<sect2 id="gnatstools">
<title>Useful Tools</title>
<para>Other than <command>edit-pr</command> there are a
collection of tools in <filename>~gnats/tools/</filename>
on <hostid>freefall</hostid> which can make
working with PRs much easier.</para>
<para><command>open-pr</command>, <command>close-pr</command>,
<command>take-pr</command>, and <command>feedback-pr</command>
take PR numbers as arguments and then ask you to select from a
preexisting list of change reasons or let you type in your
own.</para>
<para><command>change-pr</command> is a multi purpose tool
that lets you make multiple changes at the same time with one
command.</para>
<para>For example, to assign PR 123456 to yourself type
<command>take-pr <replaceable>123456</replaceable></command>.
If you want to set the PR to patched awaiting an MFC at
the same time use:
<command>change-pr -t -p -m "awaiting MFC"
<replaceable>123456</replaceable></command></para>
</sect2>
</sect1>
<sect1 id="people">
<title>Who's Who</title>
<para>Besides the repository
meisters, there are other FreeBSD project members and teams whom you will
probably get to know in your role as a committer. Briefly,
and by no means all-inclusively, these are:</para>
<variablelist>
<varlistentry>
<term>&a.doceng;</term>
<listitem>
<para>doceng is the group responsible for the documentation build
infrastructure, approving new documentation committers, and
ensuring that the FreeBSD website and documentation on the FTP
site is up to date with respect to the CVS tree. It is not a
conflict resolution body. The vast majority of documentation
related discussion takes place on the &a.doc;. More details regarding the doceng team can be found in its <ulink url="http://www.FreeBSD.org/internal/doceng.html">charter</ulink>. Committers
interested in contributing to the documentation should familiarize
themselves with the <ulink
url="&url.books.fdp-primer;/index.html">Documentation Project
Primer</ulink>.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.ru;</term>
<listitem>
<para>Ruslan is Mister &man.mdoc.7;. If you are writing a
manual page and need
some advice on the structure, or the markup, ask Ruslan.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.bde;</term>
<listitem>
<para>Bruce is the Style Police-Meister.
When you do a commit that could have been done better,
Bruce will be there to tell you. Be thankful that someone
is. Bruce is also very knowledgeable on the various
standards applicable to FreeBSD.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&team.re;</term>
<listitem>
<para>These are the members of the &a.re;. This team is
responsible for setting release deadlines and controlling
the release process. During code freezes, the release
engineers have final authority on all changes to the
system for whichever branch is pending release status. If
there is something you want merged from &os.current; to
&os.stable; (whatever values those may have at any given
time), these are the people to talk to about it.</para>
<para>Hiroki is also the keeper of the release documentation
(<filename>src/release/doc/*</filename>). If you commit a
change that you think is worthy of mention in the release notes,
please make sure he knows about it. Better still, send him
a patch with your suggested commentary.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.simon;</term>
<listitem>
<para>Simon is the
<ulink url="&url.base;/security/">FreeBSD Security
Officer</ulink>
and oversees the &a.security-officer;.
</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.wollman;</term>
<listitem>
<para>If you need advice on obscure network internals or
are not sure of some potential change to the networking
subsystem you have in mind, Garrett is someone to talk
to. Garrett is also very knowledgeable on the various
standards applicable to FreeBSD.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.committers;</term>
<listitem>
<para>&a.svn-src-all.name;, &a.svn-ports-all.name; and
&a.svn-doc-all.name; are the mailing lists that the
version control system uses to send commit messages to.
You should <emphasis>never</emphasis> send email directly
to these lists. You should only send replies to this list
when they are short and are directly related to a
commit.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>&a.developers;</term>
<listitem>
<para>All committers are subscribed to -developers. This list was created to be a
forum for the committers <quote>community</quote> issues.
Examples are Core
voting, announcements, etc.</para>
<para>The &a.developers; is for the exclusive use of
FreeBSD committers. In order to develop FreeBSD, committers must
have the ability to openly discuss matters that will be resolved
before they are publicly announced. Frank discussions of work in
progress are not suitable for open publication and may harm FreeBSD.</para>
<para>All FreeBSD committers are reminded to obey the copyright of the
original author(s) of &a.developers; mail. Do not publish or
forward messages from the &a.developers; outside the list
membership without permission of all of the authors.</para>
<para>Copyright violators will be removed from the &a.developers;,
resulting in a suspension of commit privileges. Repeated or
flagrant violations may result in permanent revocation of
commit privileges.</para>
<para>This list is
<emphasis>not</emphasis> intended as a place for code reviews or a
replacement for the &a.arch;. In fact
using it as such hurts the FreeBSD Project as it gives a sense of a
closed list where general decisions affecting all of the FreeBSD
using community are made without being <quote>open</quote>.
Last, but not least <emphasis>never, never ever, email
the &a.developers; and CC:/BCC: another FreeBSD list</emphasis>.
Never, ever email another FreeBSD email list and CC:/BCC:
the &a.developers;. Doing so can greatly diminish the benefits
of this list.</para>
</listitem>
</varlistentry>
</variablelist>
</sect1>
<sect1 id="ssh.guide">
<title>SSH Quick-Start Guide</title>
<procedure>
<step>
<para>If you do not wish to type your password in every
time you use &man.ssh.1;, and you use RSA or DSA keys to
authenticate, &man.ssh-agent.1; is there for your
convenience. If you want to use &man.ssh-agent.1;, make
sure that you run it before running other applications. X
users, for example, usually do this from their
<filename>.xsession</filename> or
<filename>.xinitrc</filename> file. See &man.ssh-agent.1;
for details.</para>
</step>
<step>
<para>Generate a key pair using &man.ssh-keygen.1;. The key
pair will wind up in your
<filename><envar>$HOME</envar>/.ssh/</filename>
directory.</para>
</step>
<step>
<para>Send your public key
(<filename><envar>$HOME</envar>/.ssh/id_dsa.pub</filename>
or <filename><envar>$HOME</envar>/.ssh/id_rsa.pub</filename>)
to the person setting you up as a committer so it can be put
into the <filename><replaceable>yourlogin</replaceable></filename> file in
<filename class="directory">/etc/ssh-keys/</filename> on
<hostid>freefall</hostid>.
</para>
</step>
</procedure>
<para>Now you should be able to use &man.ssh-add.1; for
authentication once per session. This will prompt you for
your private key's pass phrase, and then store it in your
authentication agent (&man.ssh-agent.1;). If you no longer
wish to have your key stored in the agent, issuing
<command>ssh-add -d</command> will remove it.</para>
<para>Test by doing something such as <command>ssh
freefall.FreeBSD.org ls /usr</command>.</para>
<para>For more information, see
<filename role="package">security/openssh</filename>, &man.ssh.1;,
&man.ssh-add.1;, &man.ssh-agent.1;, &man.ssh-keygen.1;, and
&man.scp.1;.</para>
</sect1>
<sect1 id="coverity">
<title>&coverity.prevent; Availability for &os; Committers</title>
<para>In January 2006, the &os; Foundation obtained a license for
&coverity.prevent; from &coverity; Ltd. With this donation, all
&os; developers can obtain access to <application>Coverity
Prevent</application> analysis results of all &os; Project
software.</para>
<para>&os; developers who are interested in obtaining access to the
analysis results of the automated <application>Coverity
Prevent</application> runs, can find out more by logging
into <hostid>freefall</hostid> and reading the relevant bits of the
files:</para>
<variablelist>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_license.txt</filename></term>
<listitem>
<para>The license terms to which the &os; developers will have
to agree in order to use &coverity.prevent; analysis
results.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_announcement.txt</filename></term>
<listitem>
<para>The announcement posted to the developers' mailing list of the
&os; Project. It contains useful information about the &os;
Foundation and &coverity; Ltd., as well as signup information
for registering with the &coverity.prevent; installation of the
&os; Cluster.</para>
<para>After reading and understanding the license terms
of <filename>coverity_license.txt</filename>, all &os; developers
who are interested in using the analysis results of
&coverity.prevent; should read this file.</para>
</listitem>
</varlistentry>
<varlistentry>
<term><filename>/usr/local/coverity/coverity_readme.txt</filename></term>
<listitem>
<para>A short guide about fixes which are committed to the &os;
source tree after being detected by &coverity.prevent; and
analyzed by a &os; developer.</para>
</listitem>
</varlistentry>
</variablelist>
<para>The &os; Wiki includes a mini-guide for developers who are
interested in working with the &coverity.prevent; analysis reports:
<ulink url="http://wiki.freebsd.org/CoverityPrevent"></ulink>. Please
note that this mini-guide is only readable by &os; developers, so if you
cannot access this page, you will have to ask someone to add you to the
appropriate Wiki access list.</para>
<para>Finally, all &os; developers who are going to use &coverity.prevent;
are always encouraged to ask for more details and usage information, by
posting any questions to the mailing list of the &os; developers.</para>
</sect1>
<sect1 id="rules">
<title>The FreeBSD Committers' Big List of Rules</title>
<orderedlist>
<listitem>
<para>Respect other committers.</para>
</listitem>
<listitem>
<para>Respect other contributors.</para>
</listitem>
<listitem>
<para>Discuss any significant change
<emphasis>before</emphasis> committing.</para>
</listitem>
<listitem>
<para>Respect existing maintainers (if listed in the
<makevar>MAINTAINER</makevar> field in
<filename>Makefile</filename> or in the
<filename>MAINTAINER</filename> file in the top-level
directory).</para>
</listitem>
<listitem>
<para>Any disputed change must be backed out pending
resolution of the dispute if requested by a maintainer.
Security related changes may
override a maintainer's wishes at the Security Officer's
discretion.</para>
</listitem>
<listitem>
<para>Changes go to &os.current; before
&os.stable; unless specifically permitted by
the release engineer or unless they are not applicable to
&os.current;. Any non-trivial or non-urgent
change which is applicable should also be allowed to sit in
&os.current; for at least 3 days before
merging so that it can be given sufficient testing. The
release engineer has the same authority over the
&os.stable; branch as outlined for the
maintainer in rule #5.</para>
</listitem>
<listitem>
<para>Do not fight in public with other committers; it looks
bad. If you must <quote>strongly disagree</quote> about
something, do so only in private.</para>
</listitem>
<listitem>
<para>Respect all code freezes and read the
<literal>committers</literal> and <literal>developers</literal>
mailing lists in a timely manner so you know when a code freeze is
in effect.</para>
</listitem>
<listitem>
<para>When in doubt on any procedure, ask first!</para>
</listitem>
<listitem>
<para>Test your changes before committing them.</para>
</listitem>
<listitem>
<para>Do not commit to anything under the
<filename>src/contrib</filename>,
<filename>src/crypto</filename>, or
<filename>src/sys/contrib</filename> trees without
<emphasis>explicit</emphasis> approval from the respective
maintainer(s).</para>
</listitem>
</orderedlist>
<para>As noted, breaking some of these rules can be grounds for
suspension or, upon repeated offense, permanent removal of
commit privileges. Individual members of core
have the power to temporarily suspend commit privileges until
core as a whole has the chance to review the
issue. In case of an <quote>emergency</quote> (a committer
doing damage to the repository), a temporary suspension may also
be done by the repository meisters.
Only a 2/3 majority of core
has the authority to suspend commit privileges for longer
than a week or to remove them permanently.
This rule does not exist to set core up as a bunch
of cruel dictators who can dispose of committers as casually as
empty soda cans, but to give the project a kind of safety fuse.
If someone is out of control, it is important to be
able to deal with this immediately rather than be paralyzed by
debate. In all cases, a committer whose privileges are
suspended or revoked is entitled to a <quote>hearing</quote> by core,
the total duration of the suspension being determined at that
time. A committer whose privileges are suspended may also
request a review of the decision after 30 days and every 30 days
thereafter (unless the total suspension period is less than 30
days). A committer whose privileges have been revoked entirely
may request a review after a period of 6 months has elapsed.
This review policy is <emphasis>strictly informal</emphasis>
and, in all cases, core reserves the right to either act on or
disregard requests for review if they feel their original
decision to be the right one.</para>
<para>In all other aspects of project operation, core is a subset
of committers and is bound by the <emphasis>same
rules</emphasis>. Just because someone is in core this does not mean
that they have special dispensation to step outside any of
the lines painted here; core's <quote>special powers</quote>
only kick in when it acts as a group, not on an individual
basis. As individuals, the core team members are all committers
first and core second.</para>
<sect2>
<title>Details</title>
<orderedlist>
<listitem id="respect">
<para>Respect other committers.</para>
<para>This means that you need to treat other committers as
the peer-group developers that they are. Despite our
occasional attempts to prove the contrary, one does not get
to be a committer by being stupid and nothing rankles more
than being treated that way by one of your peers. Whether
we always feel respect for one another or not (and
everyone has off days), we still have to
<emphasis>treat</emphasis> other committers with respect
at all times, on public forums and in private email.</para>
<para>Being able to work together long term is this project's
greatest asset, one far more important than any set of
changes to the code, and turning arguments about code into
issues that affect our long-term ability to work
harmoniously together is just not worth the trade-off by
any conceivable stretch of the imagination.</para>
<para>To comply with this rule, do not send email when you are
angry or otherwise behave in a manner which is likely to
strike others as needlessly confrontational. First calm
down, then think about how to communicate in the most
effective fashion for convincing the other person(s) that
your side of the argument is correct, do not just blow off
some steam so you can feel better in the short term at the
cost of a long-term flame war. Not only is this very bad
<quote>energy economics</quote>, but repeated displays of
public aggression which impair our ability to work well
together will be dealt with severely by the project
leadership and may result in suspension or termination of
your commit privileges. The project leadership will
take into account both public and private communications
brought before it. It will not seek the disclosure of
private communications, but it will take it into account
if it is volunteered by the committers involved in the
complaint.</para>
<para>All of this is never an option which the
project's leadership enjoys in the slightest, but unity
comes first. No amount of code or good advice is worth
trading that away.</para>
</listitem>
<listitem>
<para>Respect other contributors.</para>
<para>You were not always a committer. At one time you were
a contributor. Remember that at all times. Remember what
it was like trying to get help and attention. Do not forget
that your work as a contributor was very important to
you. Remember what it was like. Do not discourage, belittle,
or demean contributors. Treat them with respect. They are
our committers in waiting. They are every bit as important
to the project as committers. Their contributions are as
valid and as important as your own. After all, you made
many contributions before you became a committer. Always
remember that. </para>
<para>Consider the points raised under <xref linkend="respect"/>
and apply them also to contributors.</para>
</listitem>
<listitem>
<para>Discuss any significant change
<emphasis>before</emphasis> committing.</para>
<para>The repository is not where changes should be
initially submitted for correctness or argued over, that
should happen first in the mailing lists and the commit should
only happen once something resembling consensus has
been reached. This does not mean that you have to ask
permission before correcting every obvious syntax error or
manual page misspelling, simply that you should try to
develop a feel for when a proposed change is not quite such
a no-brainer and requires some feedback first. People
really do not mind sweeping changes if the result is
something clearly better than what they had before, they
just do not like being <emphasis>surprized</emphasis> by
those changes. The very best way of making sure that
you are on the right track is to have your code reviewed by
one or more other committers.</para>
<para>When in doubt, ask for review!</para>
</listitem>
<listitem>
<para>Respect existing maintainers if listed.</para>
<para>Many parts of FreeBSD are not <quote>owned</quote> in
the sense that any specific individual will jump up and
yell if you commit a change to <quote>their</quote> area,
but it still pays to check first. One convention we use
is to put a maintainer line in the
<filename>Makefile</filename> for any package or subtree
which is being actively maintained by one or more people;
see <ulink
url="&url.books.developers-handbook;/policies.html">
http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/developers-handbook/policies.html</ulink>
for documentation on this. Where sections of code have
several maintainers, commits to affected areas by one
maintainer need to be reviewed by at least one other
maintainer. In cases where the
<quote>maintainer-ship</quote> of something is not clear,
you can also look at the repository logs for the file(s) in
question and see if someone has been working recently or
predominantly in that area.</para>
<para>Other areas of FreeBSD fall under the control of
someone who manages an overall category of FreeBSD
evolution, such as internationalization or networking.
See <ulink
url="&url.base;/administration.html">
http://www.FreeBSD.org/administration.html</ulink>
for more information on this.</para>
</listitem>
<listitem>
<para>Any disputed change must be backed out pending
resolution of the dispute if requested by a maintainer.
Security related changes may
override a maintainer's wishes at the Security Officer's
discretion.</para>
<para>This may be hard to swallow in times of conflict (when
each side is convinced that they are in the right, of
course) but a version control system makes it unnecessary to have an ongoing
dispute raging when it is far easier to simply reverse the
disputed change, get everyone calmed down again and then
try to figure out what is the best way to proceed. If the change
turns out to be the best thing after all, it can be easily
brought back. If it turns out not to be, then the users
did not have to live with the bogus change in the tree
while everyone was busily debating its merits. People
<emphasis>very</emphasis> rarely call for back-outs in the repository
since discussion generally exposes bad or controversial
changes before the commit even happens, but on such rare
occasions the back-out should be done without argument so
that we can get immediately on to the topic of figuring
out whether it was bogus or not.</para>
</listitem>
<listitem>
<para>Changes go to &os.current; before
&os.stable; unless specifically permitted
by the release engineer or unless they are not applicable
to &os.current;. Any non-trivial or
non-urgent change which is applicable should also be
allowed to sit in &os.current; for at least
3 days before merging so that it can be given sufficient
testing. The release engineer has the same authority over
the &os.stable; branch as outlined in rule
#5.</para>
<para>This is another <quote>do not argue about it</quote>
issue since it is the release engineer who is ultimately
responsible (and gets beaten up) if a change turns out to
be bad. Please respect this and give the release engineer
your full cooperation when it comes to the
&os.stable; branch. The management of
&os.stable; may frequently seem to be
overly conservative to the casual observer, but also bear
in mind the fact that conservatism is supposed to be the
hallmark of &os.stable; and different rules
apply there than in &os.current;. There is
also really no point in having &os.current;
be a testing ground if changes are merged over to
&os.stable; immediately. Changes need a
chance to be tested by the &os.current;
developers, so allow some time to elapse before merging
unless the &os.stable; fix is critical,
time sensitive or so obvious as to make further testing
unnecessary (spelling fixes to manual pages, obvious bug/typo
fixes, etc.) In other words, apply common sense.</para>
<para>Changes to the security branches
(for example, <literal>RELENG_7_0</literal>) must be
approved by a member of the &a.security-officer;, or in
some cases, by a member of the &a.re;.</para>
</listitem>
<listitem>
<para>Do not fight in public with other committers; it looks
bad. If you must <quote>strongly disagree</quote> about
something, do so only in private.</para>
<para>This project has a public image to uphold and that
image is very important to all of us, especially if we are
to continue to attract new members. There will be
occasions when, despite everyone's very best attempts at
self-control, tempers are lost and angry words are
exchanged. The best thing that can be done in such cases is to minimize
the effects of this until everyone has cooled back down. That
means that you should not air your angry words in public
and you should not forward private correspondence to
public mailing lists or aliases. What people say
one-to-one is often much less sugar-coated than what they
would say in public, and such communications therefore
have no place there - they only serve to inflame an
already bad situation. If the person sending you a
flame-o-gram at least had the grace to send it privately,
then have the grace to keep it private yourself. If you
feel you are being unfairly treated by another developer,
and it is causing you anguish, bring the matter up with
core rather than taking it public. Core will do its best to
play peace makers and get things back to sanity. In cases
where the dispute involves a change to the codebase and
the participants do not appear to be reaching an amicable
agreement, core may appoint a mutually-agreeable 3rd party
to resolve the dispute. All parties involved must then
agree to be bound by the decision reached by this 3rd
party.</para>
</listitem>
<listitem>
<para>Respect all code freezes and read the
<literal>committers</literal> and <literal>developers</literal>
mailing list on a timely basis so you know when a code freeze is
in effect.</para>
<para>Committing unapproved changes during a code freeze is a really
big mistake and committers are expected to keep up-to-date
on what is going on before jumping in after a long absence
and committing 10 megabytes worth of accumulated stuff.
People who abuse this on a regular basis will have their
commit privileges suspended until they get back from the
FreeBSD Happy Reeducation Camp we run in Greenland.</para>
</listitem>
<listitem>
<para>When in doubt on any procedure, ask first!</para>
<para>Many mistakes are made because someone is in a hurry
and just assumes they know the right way of doing
something. If you have not done it before, chances are
good that you do not actually know the way we do things
and really need to ask first or you are going to
completely embarrass yourself in public. There is no shame
in asking <quote>how in the heck do I do this?</quote> We
already know you are an intelligent person; otherwise, you
would not be a committer.</para>
</listitem>
<listitem>
<para>Test your changes before committing them.</para>
<!-- XXX Needs update re sparc64 + pc98
Also, needs more details on which machines are available for testing
-->
<para>This may sound obvious, but if it really were so
obvious then we probably would not see so many cases of
people clearly not doing this. If your changes are to the
kernel, make sure you can still compile both GENERIC and
LINT. If your changes are anywhere else, make sure you
can still make world. If your changes are to a branch,
make sure your testing occurs with a machine which is
running that code. If you have a change which also may
break another architecture, be sure and test on all
supported architectures. Please refer to the <ulink
url="http://www.FreeBSD.org/internal/">FreeBSD Internal
Page</ulink> for a list of available resources. As other
architectures are added to the FreeBSD supported platforms
list, the appropriate shared testing resources will be
made available.</para>
</listitem>
<listitem>
<para>Do not commit to anything under the
<filename>src/contrib</filename>,
<filename>src/crypto</filename>, and
<filename>src/sys/contrib</filename> trees without
<emphasis>explicit</emphasis> approval from the respective
maintainer(s).</para>
<para>The trees mentioned above are for contributed software
usually imported onto a vendor branch. Committing something
there, even if it does not take the file off the vendor branch,
may cause unnecessary headaches for those responsible for
maintaining that particular piece of software. Thus, unless
you have <emphasis>explicit</emphasis> approval from the
maintainer (or you are the maintainer), do
<emphasis>not</emphasis> commit there!</para>
<para>Please note that this does not mean you should not try to
improve the software in question; you are still more than
welcome to do so. Ideally, you should submit your patches to
the vendor. If your changes are FreeBSD-specific, talk to the
maintainer; they may be willing to apply them locally. But
whatever you do, do <emphasis>not</emphasis> commit there by
yourself!</para>
<para>Contact the &a.core; if you wish to take up maintainership
of an unmaintained part of the tree.</para>
</listitem>
</orderedlist>
</sect2>
<sect2>
<title>Policy on Multiple Architectures</title>
<para>FreeBSD has added several new architecture ports during recent
release cycles and is truly no longer an &i386; centric operating
system. In an effort to make it easier to keep FreeBSD portable
across the platforms we support, core has developed the following
mandate:</para>
<blockquote>
<para>Our 32-bit reference platform is &arch.i386;, and our 64-bit
reference platform is &arch.sparc64;. Major design work (including
major API and ABI changes) must prove itself on at least one
32-bit and at least one 64-bit platform, preferably the
primary reference platforms, before it may be committed
to the source tree.</para>
</blockquote>
<para>The &arch.i386; and &arch.sparc64; platforms were chosen due to being more
readily available to developers and as representatives of more
diverse processor and system designs - big vs little endian,
register file vs register stack, different DMA and cache
implementations, hardware page tables vs software TLB management
etc.</para>
<para>The &arch.ia64; platform has many of the same complications that
&arch.sparc64; has, but is still limited in availability to
developers.</para>
<para>We will continue to re-evaluate this policy as cost and
availability of the 64-bit platforms change.</para>
<para>Developers should also be aware of our Tier Policy for
the long term support of hardware architectures. The rules
here are intended to provide guidance during the development
process, and are distinct from the requirements for features
and architectures listed in that section. The Tier rules for
feature support on architectures at release-time are more
strict than the rules for changes during the development
process.</para>
</sect2>
<sect2>
<title>Other Suggestions</title>
<para>When committing documentation changes, use a spell checker
before committing. For all SGML docs, you should also
verify that your formatting directives are correct by running
<command>make lint</command>.</para>
<para>For all on-line manual pages, run <command>manck</command>
(from ports) over the manual page to verify all of the cross
references and file references are correct and that the man
page has all of the appropriate <makevar>MLINK</makevar>s
installed.</para>
<para>Do not mix style fixes with new functionality. A style
fix is any change which does not modify the functionality of
the code. Mixing the changes obfuscates the functionality
change when asking for differences between revisions, which
can hide any new bugs. Do not include whitespace changes with
content changes in commits to <filename>doc/</filename> or
<filename>www/</filename>. The extra clutter in the diffs
makes the translators' job much more difficult. Instead, make
any style or whitespace changes in separate commits that are
clearly labeled as such in the commit message.</para>
</sect2>
<sect2>
<title>Deprecating Features</title>
<para>When it is necessary to remove functionality from software
in the base system the following guidelines should be followed
whenever possible:</para>
<orderedlist>
<listitem>
<para>Mention is made in the manual page and possibly the
release notes that the option, utility, or interface is
deprecated. Use of the deprecated feature generates a
warning.</para>
</listitem>
<listitem>
<para>The option, utility, or interface is preserved until
the next major (point zero) release.</para>
</listitem>
<listitem>
<para>The option, utility, or interface is removed and no
longer documented. It is now obsolete. It is also
generally a good idea to note its removal in the release
notes.</para>
</listitem>
</orderedlist>
</sect2>
</sect1>
<sect1 id="archs">
<title>Support for Multiple Architectures</title>
<para>FreeBSD is a highly portable operating system intended to
function on many different types of hardware architectures.
Maintaining clean separation of Machine Dependent (MD) and Machine
Independent (MI) code, as well as minimizing MD code, is an important
part of our strategy to remain agile with regards to current
hardware trends. Each new hardware architecture supported by
FreeBSD adds substantially to the cost of code maintenance,
toolchain support, and release engineering. It also dramatically
increases the cost of effective testing of kernel changes. As such,
there is strong motivation to differentiate between classes of
support for various architectures while remaining strong in a few
key architectures that are seen as the FreeBSD "target audience".
</para>
<sect2>
<title>Statement of General Intent</title>
<para>The FreeBSD Project targets "production quality commercial
off-the-shelf (COTS) workstation, server, and high-end embedded
systems". By retaining a focus on a narrow set of architectures
of interest in these environments, the FreeBSD Project is able
to maintain high levels of quality, stability, and performance,
as well as minimize the load on various support teams on the
project, such as the ports team, documentation team,
security officer, and release engineering teams. Diversity in
hardware support broadens the options for FreeBSD consumers by
offering new features and usage opportunities (such as support
for 64-bit CPUs, use in embedded environments, etc.), but these
benefits must always be carefully considered in terms of the real-world
maintenance cost associated with additional platform support.
</para>
<para>The FreeBSD Project differentiates platform targets into
four tiers. Each tier includes a specification of the
requirements for an architecture to be in that tier,
as well as specifying the obligations of developers with
regards to the platform. In addition, a policy is defined
regarding the circumstances required to change the tier
of an architecture.</para>
</sect2>
<sect2>
<title>Tier 1: Fully Supported Architectures</title>
<para>Tier 1 platforms are fully supported by the security
officer, release engineering, and toolchain maintenance staff.
New features added to the operating system must be fully
functional across all Tier 1 architectures for every release
(features which are inherently architecture-specific, such as
support for hardware device drivers, may be exempt from this
requirement). In general, all Tier 1 platforms must have build
and Tinderbox support either in the FreeBSD.org cluster, or be
easily available for all developers. Embedded platforms may
substitute an emulator available in the FreeBSD cluster for
actual hardware.</para>
<para>Tier 1 architectures are expected to be Production Quality
with respects to all aspects of the FreeBSD operating system,
including installation and development environments.</para>
<para>Tier 1 architectures are expected to be completely
integrated into the source tree and have all features
necessary to produce an entire system relevant for that target
architecture. Tier 1 architectures generally have at least 6 active
developers.</para>
<para>Tier 1 architectures are expected to be fully supported by
the ports system. All the ports should build on a Tier 1
platform, or have the appropriate filters to prevent the
inappropriate ones from building there. The packaging system
must support all Tier 1 architectures. To ensure an
architecture's Tier 1 status, proponents of that architecture
must show that all relevant packages can be built on that
platform.</para>
<para>Tier 1 embedded architectures must be able to cross-build
packages on at least one other Tier 1 architecture. The
packages must be the most relevant for the platform, but may
be a non-empty subset of those that build natively.</para>
<para>Tier 1 architectures must be fully documented. All basic
operations need to be covered by the handbook or other
documents. All relevant integration documentation must also
be integrated into the tree, or readily available.</para>
<para>Current Tier 1 platforms are &arch.i386; and &arch.amd64;.</para>
</sect2>
<sect2>
<title>Tier 2: Developmental Architectures</title>
<para>Tier 2 platforms are not supported by the security officer
and release engineering teams. Platform maintainers are
responsible for toolchain support in the tree. The toolchain
maintainer is expected to work with the platform maintainers
to refine these changes. Major new toolchain components are
allowed to break support for Tier 2 architectures if the
FreeBSD-local changes have not been incorporated upstream. The
toolchain maintainers are expected to provide prompt review of
any proposed changes and cannot block, through their inaction,
changes going into the tree. New features added to FreeBSD
should be feasible to implement on these platforms, but an
implementation is not required before the feature may be added
to the FreeBSD source tree. New features that may be difficult
to implement on Tier 2 architectures should provide a means of
disabling them on those architectures. The implementation of
a Tier 2 architecture may be committed to the main FreeBSD
tree as long as it does not interfere with production work on
Tier 1 platforms, or substantially with other Tier 2 platforms.
Before a Tier 2 platform can be added to the FreeBSD base
source tree, the platform must be able to boot multi-user on
actual hardware. Generally, there must be at least three active
developers working on the platform.</para>
<para>Tier 2 architectures are usually systems targeted at Tier 1
support, but that are still under development. Architectures
reaching end of life may also be moved from Tier 1 status to Tier
2 status as the availability of resources to continue to maintain
the system in a Production Quality state diminishes. Well supported
niche architectures may also be Tier 2.</para>
<para>Tier 2 architectures may have some support for them
integrated into the ports infrastructure. They may have cross
compilation support added, at the discretion of portmgr. Some
ports must built natively into packages if the package system
supports that architecture. If not integrated into the base
system, some external patches for the architecture for ports
must be available.</para>
<para>Tier 2 architectures can be integrated into the FreeBSD
handbook. The basics for how to get a system running must be
documented, although not necessarily for every single board or
system a Tier 2 architecture supports. The supported hardware
list must exist and should be no more than a couple of months
old. It should be integrated into the FreeBSD
documentation.</para>
<para>Current Tier 2 platforms are &arch.arm;, &arch.ia64;, &arch.pc98;, &arch.powerpc;, and &arch.sparc64;.</para>
</sect2>
<sect2>
<title>Tier 3: Experimental Architectures</title>
<para>Tier 3 platforms are not supported by the security officer
and release engineering teams. At the discretion of the
toolchain maintainer, they may be supported in the toolchain.
Tier 3 platforms are architectures in the early stages of
development, for non-mainstream hardware platforms, or which
are considered legacy systems unlikely to see broad future
use. New Tier 3 systems will not be committed to the base
source tree. Support for Tier 3 systems may be worked on in
the FreeBSD Perforce Repository, providing source control and
easier change integration from the main FreeBSD tree.
Platforms that transition to Tier 3 status may be removed from
the tree if they are no longer actively supported by the
FreeBSD developer community at the discretion of the release
engineer.</para>
<para>Tier 3 platforms may have ports support, either integrated
or external, but do not require it.</para>
<para>Tier 3 platforms must have the basics documented for how
to build a kernel and how to boot it on at least one target
hardware or emulation environment. This documentation need
not be integrated into the FreeBSD tree.</para>
<para>Current Tier 3 platforms are &arch.mips; and &s390;.</para>
</sect2>
<sect2>
<title>Tier 4: Unsupported Architectures</title>
<para>Tier 4 systems are not supported in any form by the project.
</para>
<para>All systems not otherwise classified into a support tier
are Tier 4 systems.</para>
</sect2>
<sect2>
<title>Policy on Changing the Tier of an Architecture</title>
<para>Systems may only be moved from one tier to another by
approval of the FreeBSD Core Team, which shall make that
decision in collaboration with the Security Officer, Release
Engineering, and toolchain maintenance teams.</para>
</sect2>
</sect1>
<sect1 id="ports">
<title>Ports Specific FAQ</title>
<qandaset>
<qandadiv>
<title>Adding a New Port</title>
<qandaentry>
<question>
<para>How do I add a new port?</para>
</question>
<answer>
<para>First, please read the section about repository
copies.</para>
<para>The easiest way to add a new port is to use the
<command>addport</command> script from your machine (located
in the <filename>ports/Tools/scripts</filename> directory).
It will add a port from the
directory you specify, determining the category automatically
from the port <filename>Makefile</filename>.
It will also add an entry to the port's
category <filename>Makefile</filename>. It was
written by &a.mharo;, &a.will;, and &a.garga;. When sending
questions about this script to the &a.ports;, please also CC
&a.crees;, the current maintainer.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Any other things I need to know when I add a new
port?</para>
</question>
<answer>
<para>Check the port, preferably to make sure it compiles
and packages correctly. This is the recommended
sequence:</para>
<screen>&prompt.root; <userinput>make install</userinput>
&prompt.root; <userinput>make package</userinput>
&prompt.root; <userinput>make deinstall</userinput>
&prompt.root; <userinput>pkg_add <replaceable>package you built above</replaceable></userinput>
&prompt.root; <userinput>make deinstall</userinput>
&prompt.root; <userinput>make reinstall</userinput>
&prompt.root; <userinput>make package</userinput>
</screen>
<para>The
<ulink url="&url.books.porters-handbook;/index.html">Porters
Handbook</ulink> contains more detailed
instructions.</para>
<para>Use &man.portlint.1; to check the syntax of the port.
You do not necessarily have to eliminate all warnings but
make sure you have fixed the simple ones.</para>
<para>If the port came from a submitter who has not
contributed to the Project before, add that person's
name to the <ulink
url="&url.articles.contributors;/contrib-additional.html">Additional
Contributors</ulink> section of the FreeBSD Contributors
List.</para>
<para>Close the PR if the port came in as a PR. To close
a PR, just do
<userinput>edit-pr <replaceable>PR#</replaceable></userinput>
on <hostid>freefall</hostid> and change the
<varname>state</varname> from <constant>open</constant>
to <constant>closed</constant>. You will be asked to
enter a log message and then you are done.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Removing an Existing Port</title>
<qandaentry>
<question>
<para>How do I remove an existing port?</para>
</question>
<answer>
<para>First, please read the section about repository
copies. Before you remove the port, you have to verify
there are no other ports depending on it.</para>
<itemizedlist>
<listitem>
<para>Make sure there is no dependency on the port
in the ports collection:</para>
<itemizedlist>
<listitem>
<para>The port's PKGNAME should appear in exactly one
line in a recent INDEX file.</para>
</listitem>
<listitem>
<para>No other ports should contain any reference to
the port's directory or PKGNAME in their
Makefiles</para>
</listitem>
</itemizedlist>
</listitem>
<listitem>
<para>Then, remove the port:</para>
<procedure>
<step>
<para>Remove the port's files via <command>svn remove</command>.</para>
</step>
<step>
<para>Remove the <makevar>SUBDIR</makevar> listing of the port
in the parent directory <filename>Makefile</filename>.</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>Remove the port from
<filename>ports/LEGAL</filename> if it is there.</para>
</step>
</procedure>
</listitem>
</itemizedlist>
<para>Alternatively, you can use the <command>rmport</command>
script, from <filename class="directory">ports/Tools/scripts</filename>.
This script was written by &a.vd;. When sending
questions about this script to the &a.ports;, please also CC
&a.crees;, the current maintainer.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Re-adding a Deleted Port</title>
<qandaentry>
<question>
<para>How do I re-add a deleted port?</para>
</question>
<answer>
<para>This is essentially the reverse of deleting a port.</para>
<procedure>
<step>
<para>Figure out when the port was removed. Use this
<ulink url="http://people.freebsd.org/~crees/removed_ports/index.xml">list</ulink>
and then copy the last living revision of the port:
<screen>&prompt.user; <userinput>cd /usr/ports/<replaceable>category
</replaceable></userinput>
&prompt.user; <userinput>svn cp 'svn+ssh://svn.freebsd.org/ports/<replaceable>category</replaceable>/<replaceable>portname</replaceable>/@{<replaceable>YYYY-MM-DD</replaceable>}' <replaceable>portname</replaceable>
</userinput></screen>
Pick a date that is before the removal but after the last true
commit.</para>
</step>
<step>
<para>Perform whatever changes are necessary to make the port
work again. If it was deleted because the distfiles are
no longer available you will need to volunteer to host them
yourself, or find someone else to do so.</para>
</step>
<step>
<para><command>svn add</command> or <command>svn remove</command>
any appropriate files.</para>
</step>
<step>
<para>Restore the <makevar>SUBDIR</makevar> listing of the port
in the parent directory <filename>Makefile</filename>, and
delete the entry from <filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>If the port had an entry in
<filename>ports/LEGAL</filename>, restore it.</para>
</step>
<step>
<para><command>svn commit</command> these changes, preferably in
one step.</para>
</step>
</procedure>
<tip>
<para><command>addport</command> now detects when the port to
add has previously existed, and should handle all except
the <filename>ports/LEGAL</filename> step automatically.</para>
</tip>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Repository Copies</title>
<qandaentry>
<question>
<para>When do we need a repository copy?</para>
</question>
<answer>
<para>When you want to add a port that is related to
any port that is already in the tree in a separate
directory, you have to do a repository copy.
Here <wordasword>related</wordasword> means
it is a different version or a slightly modified
version. Examples are
<filename>print/ghostscript*</filename> (different
versions) and <filename>x11-wm/windowmaker*</filename>
(English-only and internationalized version).</para>
<para>Another example is when a port is moved from one
subdirectory to another, or when you want to change the
name of a directory because the author(s) renamed their
software even though it is a
descendant of a port already in a tree.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do?</para>
</question>
<answer>
<para>With Subversion, a repo copy can be done by any
committer:</para>
<itemizedlist>
<listitem>
<para>Doing a repo copy:</para>
<procedure>
<step>
<para>First make sure that you were using an up to
date ports tree and the target directory does not
exist.</para>
</step>
<step>
<para>Use <command>svn move</command> or <command>svn
copy</command> to do the repo copy.</para>
</step>
<step>
<para>Upgrade the copied port to the new version.
Remember to change the <makevar>LATEST_LINK</makevar>
so there are no duplicate ports with the same name.
In some rare cases it may be necessary to change the
<makevar>PORTNAME</makevar> instead of
<makevar>LATEST_LINK</makevar>, but this should only
be done when it is really needed — e.g., using
an existing port as the base for a very similar
program with a different name, or upgrading a port to
a new upstream version which actually changes the
distribution name, like the transition from
<filename>textproc/libxml</filename> to
<filename>textproc/libxml2</filename>. In most cases,
changing <makevar>LATEST_LINK</makevar> should
suffice.</para>
</step>
<step>
<para>Add the new subdirectory to the
<makevar>SUBDIR</makevar> listing in the parent
directory <filename>Makefile</filename>. You can run
<command>make checksubdirs</command> in the parent
directory to check this.</para>
</step>
<step>
<para>If the port changed categories, modify the
<makevar>CATEGORIES</makevar> line of the port's
<filename>Makefile</filename> accordingly</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>, if you remove the
original port.</para>
</step>
<step>
<para>Commit all changes on one commit. A forced commit
is no longer needed with Subversion.</para>
</step>
</procedure>
</listitem>
<listitem>
<para>When removing a port:</para>
<procedure>
<step>
<para>Perform a thorough check of the ports collection for
any dependencies on the old port location/name, and
update them. Running <command>grep</command> on
<filename>INDEX</filename> is not enough because some
ports have dependencies enabled by compile-time options.
A full <command>grep -r</command> of the ports
collection is recommended.</para>
</step>
<step>
<para>Remove the old port and the
old <makevar>SUBDIR</makevar> entry.</para>
</step>
<step>
<para>Add an entry to
<filename>ports/MOVED</filename>.</para>
</step>
</procedure>
</listitem>
<listitem>
<para>After repo moves (<quote>rename</quote> operations where
a port is copied and the old location is removed):</para>
<procedure>
<step>
<para>Follow the same steps that are outlined in the
previous two entries, to activate the new location of
the port and remove the old one.</para>
</step>
</procedure>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Ports Freeze</title>
<qandaentry>
<question>
<para>What is a <quote>ports freeze</quote>?</para>
</question>
<answer>
<para>Before a release, it is necessary to restrict
commits to the ports tree for a short period of time
while the packages and the release itself are being
built. This is to ensure consistency among the various
parts of the release, and is called the <quote>ports
freeze</quote>.</para>
<para>For more information on the background and
policies surrounding a ports freeze, see the
<ulink url="&url.base;/portmgr/qa.html">Portmgr
Quality Assurance page</ulink>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What is a <quote>ports slush</quote> or
<quote>feature freeze</quote>?</para>
</question>
<answer>
<para>During a release cycle the ports tree may be in a
<quote>slush</quote> state instead of in a hard freeze.
The goal during a slush is to reach a stable ports tree
to avoid rebuilding large sets of packages for the
release and to tag the tree. During this time
<quote>sweeping changes</quote> are prohibited unless
specifically permitted by portmgr. Complete details
about what qualifies as a sweeping change can be found
on the <ulink
url="&url.base;/portmgr/implementation.html">Portmgr
Implementation page</ulink>.</para>
<para>The benefit of a slush as opposed to a complete
freeze is that it allows maintainers to continue adding
new ports, making routine version updates, and bug fixes
to most existing ports, as long as the number of
affected ports is minimal. For example, updating the
shared library version on a port that many other ports
depend on.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How long is a ports freeze or slush?</para>
</question>
<answer>
<para>A freeze only lasts long enough to tag the tree.
A slush usually lasts a week or two, but may last
longer.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What does it mean to me?</para>
</question>
<answer>
<para>During a ports freeze, you are not allowed to
commit anything to the tree without explicit approval
from the Ports Management Team. <quote>Explicit
approval</quote> here means that you send a patch to
the Ports Management Team for review and get a reply
saying, <quote>Go ahead and commit it.</quote>
</para>
<para>Not everything is allowed to be committed during
a freeze. Please see the <ulink
url="&url.base;/portmgr/qa.html">Portmgr Quality
Assurance page</ulink> for more information.
</para>
<para>Note that you do not have implicit permission to fix
a port during the freeze just because it is
broken.</para>
<para>During a ports slush, you are still allowed to
commit but you must exercise more caution in what you
commit. Furthermore a special note (typically
<quote>Feature Safe: yes</quote>) must be added to the
commit message.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I know when the ports slush starts?</para>
</question>
<answer>
<para>The Ports Management Team will send out warning
messages to the &a.ports; and &a.committers;
announcing the start of the impending release, usually
two or three weeks in advance. The exact starting time
will not be determined until a few days before the
actual release. This is because the ports slush has to
be synchronized with the release, and it is usually not
known until then when exactly the release will be
rolled.</para>
<para>When the slush starts, there will be another
announcement to the &a.ports; and &a.committers;, of course.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I know when the freeze or slush ends?</para>
</question>
<answer>
<para>A few hours after the release, the Ports Management
Team will send out a mail to the &a.ports; and &a.committers;
announcing the end of the ports freeze or slush. Note
that the release being cut does not automatically indicate
the end of the freeze. We have to make sure there will
be no last minute snafus that result in an immediate
re-rolling of the release.</para>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Creating a New Category</title>
<qandaentry>
<question>
<para>What is the procedure for creating a new category?</para>
</question>
<answer>
<para>Please see
<ulink url="&url.books.porters-handbook;/makefile-categories.html#PROPOSING-CATEGORIES">
Proposing a New Category</ulink> in the Porter's Handbook.
Once that procedure has been followed and the PR has been
assigned to &a.portmgr;, it is their decision whether or
not to approve it. If they do, it is their responsibility
to do the following:</para>
<procedure>
<step>
<para>Perform any needed moves. (This only applies
to physical categories.)</para>
</step>
<step>
<para>Update the <makevar>VALID_CATEGORIES</makevar>
definition in <filename>ports/Mk/bsd.port.mk</filename>.
</para>
</step>
<step>
<para>Assign the PR back to you.</para>
</step>
</procedure>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do to implement a new physical
category?</para>
</question>
<answer>
<procedure>
<step>
<para>Upgrade each moved port's
<filename>Makefile</filename>. Do not connect the
new category to the build yet.</para>
<para>To do this, you will need to:</para>
<procedure>
<step>
<para>Change the port's <makevar>CATEGORIES</makevar>
(this was the point of the exercise, remember?)
The new category should be listed
<emphasis>first</emphasis>. This will help to
ensure that the <makevar>PKGORIGIN</makevar>
is correct.</para>
</step>
<step>
<para>Run a <command>make describe</command>. Since
the top-level <command>make index</command> that
you will be running in a few steps is an iteration
of <command>make describe</command> over the entire
ports hierarchy, catching any errors here will
save you having to re-run that step later on.</para>
</step>
<step>
<para>If you want to be really thorough, now might
be a good time to run &man.portlint.1;.</para>
</step>
</procedure>
</step>
<step>
<para>Check that the <makevar>PKGORIGIN</makevar>s are
correct. The ports system uses each port's
<makevar>CATEGORIES</makevar> entry to create
its <makevar>PKGORIGIN</makevar>, which is used to
connect installed packages to the port directory they
were built from. If this entry is wrong, common port
tools like &man.pkg.version.1; and
&man.portupgrade.1; fail.</para>
<para>To do this, use the <filename>chkorigin.sh</filename>
tool, as follows: <command>env
PORTSDIR=<replaceable>/path/to/ports</replaceable>
sh -e <replaceable>/path/to/ports</replaceable>/Tools/scripts/chkorigin.sh
</command>. This will check <emphasis>every</emphasis>
port in the ports tree, even those not connected to the
build, so you can run it directly after the move
operation.
Hint: do not forget to look at the
<makevar>PKGORIGIN</makevar>s of any slave ports of the
ports you just moved!</para>
</step>
<step>
<para>On your own local system, test the proposed
changes: first, comment out the
<makevar>SUBDIR</makevar> entries in the old
ports' categories' <filename>Makefile</filename>s;
then enable building the new category in
<filename>ports/Makefile</filename>.
Run <command>make checksubdirs</command> in the
affected category directories to check the
<makevar>SUBDIR</makevar> entries. Next, in
the <filename class="directory">ports/</filename>
directory, run <command>make index</command>. This
can take over 40 minutes on even modern systems;
however, it is a necessary step to prevent problems
for other people.</para>
</step>
<step>
<para>Once this is done, you can commit the
updated <filename>ports/Makefile</filename> to
connect the new category to the build and also
commit the <filename>Makefile</filename> changes
for the old category or categories.</para>
</step>
<step>
<para>Add appropriate entries to
<filename>ports/MOVED</filename>.</para>
</step>
<step>
<para>Update the documentation by modifying the
following:</para>
<itemizedlist>
<listitem>
<para>the
<ulink url="&url.books.porters-handbook;/makefile-categories.html#PORTING-CATEGORIES">
list of categories</ulink> in the Porter's Handbook</para>
</listitem>
<listitem>
<para>
<filename>www/en/ports/categories</filename>.
Note that these are now displayed by sub-groups,
as specified in
<filename>www/en/ports/categories.descriptions</filename>.
</para>
</listitem>
</itemizedlist>
<para>(Note: these are
in the docs, not the ports, repository). If you
are not a docs committer, you will need to submit
a PR for this.</para>
</step>
<step>
<para>Only once all the above have been done, and
no one is any longer reporting problems with the
new ports, should the old ports be deleted from
their previous locations in the repository.</para>
</step>
</procedure>
<para>It is not necessary to manually update the <ulink
url="&url.base;/ports/index.html">ports web pages</ulink>
to reflect the new category. This is now done automatically
via your change to <filename>www/en/ports/categories</filename>
and the daily automated rebuild of <filename>INDEX</filename>.
</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What do I need to do to implement a new virtual
category?</para>
</question>
<answer>
<para>This is much simpler than a physical category. You
only need to modify the following:</para>
<itemizedlist>
<listitem>
<para>the
<ulink url="&url.books.porters-handbook;/makefile-categories.html#PORTING-CATEGORIES">
list of categories</ulink> in the Porter's Handbook</para>
</listitem>
<listitem>
<para><filename>www/en/ports/categories</filename></para>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
</qandadiv>
<qandadiv>
<title>Miscellaneous Questions</title>
<qandaentry>
<question>
<para>How do I know if my port is building correctly or
not?</para>
</question>
<answer>
<para>First, go check
<ulink url="http://pointyhat.FreeBSD.org/errorlogs/"></ulink>.
There you will find error logs from the latest package
building runs on all supported platforms for the most
recent branches.</para>
<para>However, just because the port does not show up there
does not mean it is building correctly. (One of the
dependencies may have failed, for instance.) The relevant
directories are available on <hostid>pointyhat</hostid> under
<filename class="directory">/a/portbuild/<arch>/<major_version></filename>
so feel free to dig around. Each architecture and version has
the following subdirectories:</para>
<programlisting>errors error logs from latest <major_version> run on <arch>
logs all logs from latest <major_version> run on <arch>
packages packages from latest <major_version> run on <arch>
bak/errors error logs from last complete <major_version> run on <arch>
bak/logs all logs from last complete <major_version> run on <arch>
bak/packages packages from last complete <major_version> run on <arch></programlisting>
<para>Basically, if the port shows up in
<filename>packages</filename>, or it is in
<filename>logs</filename> but not in
<filename>errors</filename>, it built fine. (The
<filename>errors</filename> directories are what you get
from the web page.)</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>I added a new port. Do I need to add it to the
<filename>INDEX</filename>?</para>
</question>
<answer>
<para>No, <filename>INDEX</filename> is no longer stored
in the SVN repository. The file can either be generated
by running <command>make index</command>, or a pre-generated
version can be downloaded with <command>make
fetchindex</command>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Are there any other files I am not allowed to
touch?</para>
</question>
<answer>
<para>Any file directly under <filename>ports/</filename>, or
any file under a subdirectory that starts with an
uppercase letter (<filename>Mk/</filename>,
<filename>Tools/</filename>, etc.). In particular, the
Ports Management Team is very protective of
<filename>ports/Mk/bsd.port*.mk</filename> so do not
commit changes to those files unless you want to face his
wra(i)th.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What is the proper procedure for updating the checksum
for a port's distfile when the file changes without a
version change?</para>
</question>
<answer>
<para>When the checksum for a port's distfile is updated due
to the author updating the file without changing the port's
revision, the commit message should include a summary of
the relevant diffs between the original and new distfile to
ensure that the distfile has not been corrupted or
maliciously altered. If the current version of the port
has been in the ports tree for a while, a copy of the old
distfile will usually be available on the ftp servers;
otherwise the author or maintainer should be contacted to
find out why the distfile has changed.</para>
</answer>
</qandaentry>
</qandadiv>
</qandaset>
</sect1>
<sect1 id="non-committers">
<title>Issues Specific To Developers Who Are Not Committers</title>
<para>A few people who have access to the FreeBSD machines do not
have commit bits. For instance, the project is willing to give
access to the GNATS database to contributors who have shown interest
and dedication in working on Problem Reports.</para>
<para>Almost all of this document will apply to these developers as
well (except things specific to commits and the mailing list
memberships that go with them). In particular, we recommend that
you read:</para>
<itemizedlist>
<listitem>
<para>
<link linkend="admin">Administrative Details</link>
</para>
</listitem>
<listitem>
<para>
<link linkend="conventions-everyone">Conventions</link>
</para>
<note>
<para>You should get your mentor to add you to the
<quote>Additional Contributors</quote>
(<filename>doc/en_US.ISO8859-1/articles/contributors/contrib.additional.xml</filename>),
if you are not already listed there.</para>
</note>
</listitem>
<listitem>
<para>
<link linkend="developer.relations">Developer Relations</link>
</para>
</listitem>
<listitem>
<para>
<link linkend="ssh.guide">SSH Quick-Start Guide</link>
</para>
</listitem>
<listitem>
<para>
<link linkend="rules">The FreeBSD Committers' Big List of Rules</link>
</para>
</listitem>
</itemizedlist>
</sect1>
<sect1 id="perks">
<title>Perks of the Job</title>
<para>Unfortunately, there are not many perks involved with being a
committer. Recognition as a competent software engineer is probably
the only thing that will be of benefit in the long run. However,
there are at least some perks:</para>
<variablelist>
<varlistentry>
<term>Free 4-CD and DVD Sets</term>
<listitem>
<para>&os; committers can get a free 4-CD or DVD set at
conferences from <ulink url="http://www.freebsdmall.com">
&os; Mall, Inc.</ulink>. The sets are no longer available
as a subscription due to the high shipment costs to
countries outside the USA.</para>
</listitem>
</varlistentry>
<varlistentry>
<term>Freenode IRC Cloaks</term>
<listitem>
<para>&os; developers may request a cloaked hostmask for their
account on the Freenode IRC network in the form of
<literal>freebsd/developer/</literal><replaceable>freefall name</replaceable>
or <literal>freebsd/developer/</literal><replaceable>NickServ name</replaceable>.
To request a cloak, send an email to &a.eadler; with your
requested hostmask and NickServ account name.</para>
</listitem>
</varlistentry>
</variablelist>
</sect1>
<sect1 id="misc">
<title>Miscellaneous Questions</title>
<qandaset>
<qandaentry>
<question>
<para>Why are trivial or cosmetic changes to files on a vendor
branch a bad idea?</para>
</question>
<answer>
<itemizedlist>
<listitem>
<para>From now on, every new vendor release of that file will
need to have patches merged in by hand.</para>
</listitem>
<listitem>
<para>From now on, every new vendor release of that file will
need to have patches <emphasis>verified</emphasis> by hand.</para>
</listitem>
<listitem>
<para>The <option>-j</option> option does not work very well.
Ask &a.obrien; for horror stories.</para>
</listitem>
</itemizedlist>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I add a new file to a branch?</para>
</question>
<answer>
<para>To add a file onto a branch, simply checkout or update
to the branch you want to add to and then add the file using
the add operation as you normally would. This works
fine for the <literal>doc</literal> and
<literal>ports</literal> trees. The <literal>src</literal>
tree uses SVN and requires more care because of the
<literal>mergeinfo</literal> properties. See section 1.4.6
of the <ulink url="http://wiki.freebsd.org/SubversionPrimer">
Subversion Primer</ulink> for details. Refer to <ulink
url="http://wiki.freebsd.org/SubversionPrimer/Merging">
SubversionPrimer/Merging</ulink> for details on how to
perform an MFC.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>What <quote>meta</quote> information should I include in a
commit message?</para>
</question>
<answer>
<para>As well as including an informative message with each commit
you may need to include some additional information as
well.</para>
<para>This information consists of one or more lines containing the
key word or phrase, a colon, tabs for formatting, and then the
additional information.</para>
<para>The key words or phrases are:</para>
<informaltable frame="none" pgwide="1">
<tgroup cols="2">
<tbody>
<row>
<entry><literal>PR:</literal></entry>
<entry>The problem report (if any) which is affected
(typically, by being closed) by this commit.</entry>
</row>
<row>
<entry><literal>Submitted by:</literal></entry>
<entry>The name and e-mail address of the person that
submitted the fix; for committers, just the username on
the FreeBSD cluster.</entry>
</row>
<row>
<entry><literal>Reviewed by:</literal></entry>
<entry>The name and e-mail address of the person or people
that reviewed the change; for committers, just the
username on the FreeBSD cluster. If a patch was
submitted to a mailing list for review, and the review
was favorable, then just include the list name.</entry>
</row>
<row>
<entry><literal>Approved by:</literal></entry>
<entry>The name and e-mail address of the person or people
that approved the change; for committers, just the
username on the FreeBSD cluster. It is customary to get
prior approval for a commit if it is to an area of the
tree to which you do not usually commit. In addition,
during the run up to a new release all commits
<emphasis>must</emphasis> be approved by the release
engineering team. If these are your first commits then
you should have passed them past your mentor first, and
you should list your mentor, as in
``<replaceable>username-of-mentor</replaceable>
<literal>(mentor)</literal>''.
</entry>
</row>
<row>
<entry><literal>Obtained from:</literal></entry>
<entry>The name of the project (if any) from which the code
was obtained.</entry>
</row>
<row>
<entry><literal>MFC after:</literal></entry>
<entry>If you wish to receive an e-mail reminder to
<acronym>MFC</acronym> at a later date, specify the
number of days, weeks, or months after which an
<acronym>MFC</acronym> is planned.</entry>
</row>
<row>
<entry><literal>Security:</literal></entry>
<entry>If the change is related to a security
vulnerability or security exposure, include one or
more references or a description of the
issue.</entry>
</row>
</tbody>
</tgroup>
</informaltable>
<example>
<title>Commit log for a commit based on a PR</title>
<para>You want to commit a change based on a PR submitted by John
Smith containing a patch. The end of the commit message should
look something like this.</para>
<programlisting>...
PR: foo/12345
Submitted by: John Smith <John.Smith@example.com></programlisting>
</example>
<example>
<title>Commit log for a commit needing review</title>
<para>You want to change the virtual memory system. You have
posted patches to the appropriate mailing list (in this case,
<literal>freebsd-arch</literal>) and the changes have been
approved.</para>
<programlisting>...
Reviewed by: -arch</programlisting>
</example>
<example>
<title>Commit log for a commit needing approval</title>
<para>You want to commit a change to a section of the tree with a
MAINTAINER assigned. You have collaborated with the listed
MAINTAINER, who has told you to go ahead and commit.</para>
<programlisting>...
Approved by: <replaceable>abc</replaceable></programlisting>
<para>Where <replaceable>abc</replaceable> is the account name of
the person who approved.</para>
</example>
<example>
<title>Commit log for a commit bringing in code from
OpenBSD</title>
<para>You want to commit some code based on work done in the
OpenBSD project.</para>
<programlisting>...
Obtained from: OpenBSD</programlisting>
</example>
<example>
<title>Commit log for a change to &os.current; with a planned
commit to &os.stable; to follow at a later date.</title>
<para>You want to commit some code which will be merged from
&os.current; into the &os.stable; branch after two
weeks.</para>
<programlisting>...
MFC after: <replaceable>2 weeks</replaceable></programlisting>
<para>Where <replaceable>2</replaceable> is the number of days,
weeks, or months after which an <acronym>MFC</acronym> is
planned. The <replaceable>weeks</replaceable> option may be
<literal>day</literal>, <literal>days</literal>,
<literal>week</literal>, <literal>weeks</literal>,
<literal>month</literal>, <literal>months</literal>,
or may be left off (in which case, days will be assumed).</para>
</example>
<para>In some cases you may need to combine some of these.</para>
<para>Consider the situation where a user has submitted a PR
containing code from the NetBSD project. You are looking at the
PR, but it is not an area of the tree you normally work in, so
you have decided to get the change reviewed by the
<literal>arch</literal> mailing list. Since the change is
complex, you opt to <acronym>MFC</acronym> after one month to
allow adequate testing.</para>
<para>The extra information to include in the commit would look
something like</para>
<programlisting>PR: foo/54321
Submitted by: John Smith <John.Smith@example.com>
Reviewed by: -arch
Obtained from: NetBSD
MFC after: 1 month</programlisting>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>How do I access <hostid
role="fqdn">people.FreeBSD.org</hostid> to put up personal
or project information?</para>
</question>
<answer>
<para><hostid role="fqdn">people.FreeBSD.org</hostid> is the
same as <hostid
role="fqdn">freefall.FreeBSD.org</hostid>. Just create a
<filename>public_html</filename> directory. Anything you
place in that directory will automatically be visible
under <ulink url="http://people.FreeBSD.org/"></ulink>.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>Where are the mailing list archives stored?</para>
</question>
<answer>
<para>The mailing lists are archived under <filename>/g/mail</filename>
which will show up as <filename>/hub/g/mail</filename> with &man.pwd.1;.
This location is accessible from any machine on the FreeBSD cluster.</para>
</answer>
</qandaentry>
<qandaentry>
<question>
<para>I would like to mentor a new committer. What process
do I need to follow?</para>
</question>
<answer>
<para>See the <ulink
url="http://www.freebsd.org/internal/new-account.html">New
Account Creation Procedure</ulink> document on the internal
pages.</para>
</answer>
</qandaentry>
</qandaset>
</sect1>
</article>