path: root/en_US.ISO8859-1/articles/committers-guide/article.xml

<?xml version="1.0" encoding="iso-8859-1" standalone="no"?>
<!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>
      <holder>The FreeBSD Documentation Project</holder>
    </copyright>

    <legalnotice id="trademarks" role="trademarks">
      &tm-attrib.freebsd;
      &tm-attrib.coverity;
      &tm-attrib.cvsup;
      &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/, www/, 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;&nbsp;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 repository to the
	<acronym>CVS</acronym> one, so regular users should not notice
	a difference, however developers most certainly will.</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&mdash;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 three 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>You can check out a working copy from a mirror by simply
	  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 you maintain yourself
	  using <command>svnsync</command> or similar.</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> and
	  <acronym>csup</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>
	<title>Checkout from a Local Mirror Using
	  <acronym>SVK</acronym></title>

	<para>The third alternative is to use <acronym>SVK</acronym>
	  to maintain a local mirror.  It is a version control system
	  build on top of Subversion's storage engine.  It is
	  identical to Subversion in most respects, except that it
	  allows for setting up parts of repositories as mirrors of
	  other repositories, and keeping local branches for merging
	  back into the upstream repositories.  There are extensions
	  that allow <acronym>SVK</acronym> to mirror
	  <acronym>CVS</acronym> and Perforce repositories in addition
	  to Subversion ones.</para>

	<para>Like everything, <acronym>SVK</acronym> has its
	  disadvantages, one being that local revision numbers will
	  not match upstream revision numbers.  This makes it
	  difficult to <command>svk log</command>, <command>svk
	    diff</command>, or <command>svk update</command> to an
	  arbitrary upstream revision.</para>

	<para>To set up a mirror of the &os; repository, do:</para>

	<screen>&prompt.user; <userinput>svk mirror svn+ssh://svn.freebsd.org/base //freebsd/base</userinput></screen>

	<para>The local <acronym>SVK</acronym> repository will be
	  stored in <filename
	    class="directory">~/.svk/local/</filename>, but can be
	  moved to an alternate location.  If it is moved,
	  <filename>~/.svk/config</filename> should be amended
	  manually to reflect the move.</para>

	<para>Any path can be used, not just the one in the example
	  above.  A common pattern is to place mirrors under
	  <literal>//mirror</literal>, e.g.,
	  <filename
	    class="directory">//mirror/freebsd/base/</filename>, and
	  local branches under <literal>//local</literal>.</para>

	<para>To pull down the contents of the repository to the
	  mirror:</para>

	<screen>&prompt.user; <userinput>svk sync //freebsd/base</userinput></screen>

	<note>
	  <para><command>svk sync</command> will take a very long
	    time, possibly several days over a slow network
	    connection.  &a.peter; has a tarball that can be used to
	    jumpstart the mirror, but only if one does not exist
	    already.</para>
	</note>

	<para>To use the tarball referenced above:</para>

	<screen>&prompt.user; <userinput>cd ~</userinput>
&prompt.user; <userinput>scp freefall:/home/peter/dot_svk_r179646.tbz2 .</userinput>
&prompt.user; <userinput>tar xf dot_svk_r179646.tbz2</userinput></screen>

	<para>Then edit <filename>~/.svk/config</filename> and replace
	  <filename
	    class="directory">/scratch/tmp/peter/.svk/local/</filename>
	  with the equivalent of <filename
	    class="directory">/home/<replaceable>jarjar</replaceable>/.svk/local/</filename>.</para>

	<para>You can check out files directly from your mirror, once
	  it has been created:</para>

	<screen>&prompt.user; <userinput>svk checkout //freebsd/base/head /usr/src</userinput></screen>

	<para>Unlike <acronym>SVN</acronym>, <acronym>SVK</acronym>
	  does not store metadata or reference copies in the working
	  copy.  All metadata is recorded in
	  <filename>~/.svk/config</filename>; reference copies are not
	  used at all because <acronym>SVK</acronym> always operates
	  on a local repository.</para>

	<para>When committing from a working copy like the one above,
	  <acronym>SVN</acronym> will commit directly to the upstream
	  repository, then synchronise the mirror.</para>

	<para>However, the <quote>killer app</quote> for
	  <acronym>SVK</acronym> is the ability to work without a
	  network connection.  To do that, a local branch must be set
	  up:</para>

	<screen>&prompt.user; <userinput>svk mkdir //local/freebsd</userinput>
&prompt.user; <userinput>svk copy //freebsd/base/head //local/freebsd/head</userinput></screen>

	<para>Once again, any path can be used, it does not have to
	  specifically be the one in the example.</para>

	<para>Before use, the local branch has to be synchronized,
	  like so:</para>

	<screen>&prompt.user; <userinput>svk pull //local/freebsd/head</userinput></screen>

	<para>Then check out from the newly created local
	  branch:</para>

	<screen>&prompt.user; <userinput>svk checkout //local/freebsd/head /usr/src</userinput></screen>

	<para>The point of this exercise is showing that it is
	  possible to commit work-in-progress to a local branch, and
	  only push it to the upstream repository when work is
	  complete.  The easy way to push is with <command>svk
	    push</command>, but there is a serious disadvantage to it:
	  it will push every single commit made to the local branch
	  incrementally instead of lumping them all into a single
	  commit.  Therefore, using <command>svk smerge</command> is
	  preferable.</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 other key locations are:</para>

	<itemizedlist>
	  <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;&nbsp;Handbook.</para>
	  </listitem>
	  <listitem>
	    <para><emphasis>/htdocs/</emphasis> contains the source
	      code for the &os;&nbsp;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>

      <note>
	<para><acronym>SVN</acronym> and <acronym>SVK</acronym>
	  commands that have direct <acronym>CVS</acronym> equivalents
	  usually have the same name and abbreviations.  For example:
	  <emphasis>checkout</emphasis> and <emphasis>co</emphasis>,
	  <emphasis>update</emphasis> and <emphasis>up</emphasis>, and
	  <emphasis>commit</emphasis> and
	  <emphasis>ci</emphasis>.</para>
      </note>

      <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 one of the following commands:</para>

	<screen>&prompt.user; <userinput>svn co svn://svn.freebsd.org/base/head /usr/src</userinput>
&prompt.user; <userinput>svn co http://svn.freebsd.org/base/head /usr/src</userinput></screen>
      </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><acronym>CVS</acronym> has no direct equivalent of this
	  command.  The nearest would be <command>cvs up -N</command>
	  which shows local changes and files that are out-of-date.
	  Doing this in <acronym>SVN</acronym> is possible too,
	  however:</para>

	<screen>&prompt.user; <userinput>svn status --show-updates</userinput></screen>
      </sect3>

      <sect3>
	<title>Editing and Committing</title>

	<para>Like <acronym>CVS</acronym> but 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>As with <acronym>CVS</acronym>, 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>In <acronym>CVS</acronym>, the directory is immediately
	  created in the repository when you <command>cvs
	    add</command> it; this is not the case in Subversion.
	  Furthermore, unlike <acronym>CVS</acronym>, 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>

	<para>Neither of these operations have equivalents in
	  <acronym>CVS</acronym>.</para>
      </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 by default, unlike
	  <acronym>CVS</acronym>, and include new files by default
	  in the diff output.</para>

	<para>As with <acronym>CVS</acronym>, <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>.  Unlike
	  <command>cvs up -C</command>, it does not update out-of-date
	  files&mdash;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>&dollar;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>&dollar;R</literal>.</para>
	      </listitem>
	      <listitem>
		<para>in directory &dollar;target in stable branch
		  &dollar;B.</para>
	      </listitem>
	      <listitem>
		<para>from directory &dollar;source in head.</para>
	      </listitem>
	      <listitem>
		<para>&dollar;FSVN is
		  <literal>svn+ssh://svn.freebsd.org/base</literal>.</para>
	      </listitem>
	    </itemizedlist>

	    <para>Assuming that revisions &dollar;P and &dollar;Q have
	      already been merged, and that the current directory is
	      an up-to-date working copy of stable/&dollar;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&mdash;one for the import itself and
	    one for the tag&mdash;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>

	<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&mdash;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, unlike with
	  <acronym>CVS</acronym>, 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 checkouts and minimising network traffic is
	possible with the following recipe:</para>

      <screen>&prompt.user; <userinput>svn co --depth=empty svn+ssh://svn.freebsd.org/base fbsvn</userinput>
&prompt.user; <userinput>cd fbsvn</userinput>
&prompt.user; <userinput>svn up --depth=empty stable</userinput>
&prompt.user; <userinput>svn up head</userinput>
&prompt.user; <userinput>cd stable</userinput>
&prompt.user; <userinput>cp -r ../head/ 7</userinput>
&prompt.user; <userinput>cd 7</userinput>
&prompt.user; <userinput>svn switch svn+ssh://svn.freebsd.org/base/stable/7</userinput>
&prompt.user; <userinput>cd ..</userinput>
&prompt.user; <userinput>cp -r 7/ 6</userinput>
&prompt.user; <userinput>cd 6</userinput>
&prompt.user; <userinput>svn switch svn+ssh://svn.freebsd.org/base/stable/6</userinput></screen>

      <para>What this bit of evil does is check out head, stable/7 and
	stable/6.  We create the empty checkout directories under
	<acronym>SVN</acronym>'s control.  In <acronym>SVN</acronym>,
	subtrees are self identifying, like in <acronym>CVS</acronym>.
	We check out head and clone it as stable/7.  Except we don't
	want the head version so we <quote>switch</quote> it to the
	7.x tree location.  <acronym>SVN</acronym> downloads diffs to
	convert the <quote>head</quote> files to
	<quote>stable/7</quote> instead of doing a fresh checkout.
	The same goes for stable/6.  This does, however, definitely
	count as abuse of the working copy client code!</para>

      <para>Checking out a working copy with a stock Subversion client
	without &os;-specific patches
	(<makevar>WITH_FREEBSD_TEMPLATE</makevar>) will mean that
	<literal>&dollar;FreeBSD&dollar;</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 is not a good idea if uncommitted patches exist,
	however.</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/en_US.ISO8859-1/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 &quot;auto-props&quot;
	    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>(For committers only:)
	  If you subscribe to &a.svn-src-all.name; or the &a.cvsall;,
	  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 &ndash; 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 in to it using CVSup.  Then you can run GNATS commands
      locally.
      This lets you query the PR database without needing to be connected to
      the Internet.</para>

    <sect2>
      <title>Mirroring the GNATS Tree</title>

      <para>It is possible to mirror the GNATS database by adding this line
	to your <filename>supfile</filename>. Note that since
	GNATS is not under CVS control it has no tag, so if you are adding
	it to your existing <filename>supfile</filename> it should appear
	before any <quote>tag=</quote> entry as these remain active once set.</para>

	<programlisting>gnats release=current prefix=/usr</programlisting>

	<para>This will place the FreeBSD GNATS tree in
	  <filename>/usr/gnats</filename>.  You can use a
	  <emphasis>refuse</emphasis> file to control which categories to
	  receive.  For example, to only receive <literal>docs</literal> PRs,
	  put this line in
	  <filename>/usr/local/etc/cvsup/sup/refuse</filename><footnote>
	    <para>The precise path depends on the <literal>*default
		base</literal> setting in your
	      <filename>supfile</filename>.</para>
	  </footnote>.</para>
    </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>cvs-committers is the entity that the version control system uses to send you all your
	    commit messages.  You should <emphasis>never</emphasis> send email
	    directly to this list.  You should only send replies to this list
	    when they are short and are directly related to a commit.</para>

	  <para>There is a similar list, svn-committers, which has a
	    similar purpose but is a normal list, i.e., you are free to
	    send any suitable message to this list.</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;&nbsp;Foundation obtained a license for
      &coverity.prevent; from &coverity;&nbsp;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;&nbsp;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; and &a.will;, and is currently maintained
	      by &a.garga;, so please send questions/patches about
	      <command>addport</command> to him.</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 has been written by &a.vd;, who is also its current
	      maintainer, so please send questions, patches or suggestions
	      about <command>rmport</command> to him.</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>
	  </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 &mdash; 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 repocopies.  (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>
	    <para>The procedure is a strict superset of the one to
	      repocopy individual ports (see above).</para>

	      <procedure>
		<step>
		  <para>Upgrade each copied 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 repocopy.
		    Hint: do not forget to look at the
		    <makevar>PKGORIGIN</makevar>s of any slave ports of the
		    ports you just repocopied!</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 instructions for &man.cvsup.1;:</para>

		  <itemizedlist>
		    <listitem>
		      <para>
			add the category to
			<filename>distrib/cvsup/sup/README</filename>
		      </para>
		    </listitem>

		    <listitem>
		      <para>
			adding the following files into
			<filename>distrib/cvsup/sup/ports-<replaceable>categoryname</replaceable></filename>:
			<filename>list.cvs</filename> and
			<filename>releases</filename>.</para>
		    </listitem>

		    <listitem>
		      <para>
			add the category to
			<filename>src/share/examples/cvsup/ports-supfile</filename>
		      </para>
		    </listitem>
		  </itemizedlist>

		  <para>
		    (Note: these are
		    in the src, not the ports, repository).  If you
		    are not a src committer, you will need to submit
		    a PR for this.</para>
		</step>

		<step>
		  <para>
		    Update the list of categories used by &man.sysinstall.8;
		    in <filename>src/usr.sbin/sysinstall</filename>.</para>
		</step>

		<step>
		  <para>Update the documentation by modifying the
		    following:</para>

		  <itemizedlist>
		    <listitem>
		      <para>the section of the Handbook that lists the
			<ulink url="&url.books.handbook;/cvsup.html#CVSUP-COLLEC">
			cvsup collections</ulink>.</para>
		    </listitem>

		    <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><filename>src/usr.sbin/sysinstall</filename></para>
	      </listitem>

	      <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/&lt;arch&gt;/&lt;major_version&gt;</filename>
	      so feel free to dig around.  Each architecture and version has
	      the following subdirectories:</para>

<programlisting>errors        error logs from latest &lt;major_version&gt; run on &lt;arch&gt;
logs          all logs from latest &lt;major_version&gt; run on &lt;arch&gt;
packages      packages from latest &lt;major_version&gt; run on &lt;arch&gt;
bak/errors    error logs from last complete &lt;major_version&gt; run on &lt;arch&gt;
bak/logs      all logs from last complete &lt;major_version&gt; run on &lt;arch&gt;
bak/packages  packages from last complete &lt;major_version&gt; run on &lt;arch&gt;</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>Direct access to <hostid>cvsup-master</hostid></term>

	<listitem>
	  <para>As a committer, you may apply to &a.kuriyama; for direct access
	    to <hostid role="fqdn">cvsup-master.FreeBSD.org</hostid>,
	    providing the public key output from <command>cvpasswd
	    <replaceable>yourusername</replaceable>@FreeBSD.org
	    freefall.FreeBSD.org</command>.  Please note: you must
	    specify <hostid>freefall.FreeBSD.org</hostid> on the
	    <command>cvpasswd</command> command line even though the
	    actual server is <hostid>cvsup-master</hostid>.  Access to
	    <hostid>cvsup-master</hostid> should not be overused as it is
	    a busy machine.</para>
	</listitem>
      </varlistentry>

      <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 &lt;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 &lt;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>