Index: head/en_US.ISO8859-1/articles/committers-guide/article.xml =================================================================== --- head/en_US.ISO8859-1/articles/committers-guide/article.xml (revision 50803) +++ head/en_US.ISO8859-1/articles/committers-guide/article.xml (revision 50804) @@ -1,5301 +1,5301 @@ ]>
Committer's Guide The &os; Documentation Project 1999 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 2016 2017 The &os; Documentation Project &tm-attrib.freebsd; &tm-attrib.coverity; &tm-attrib.ibm; &tm-attrib.intel; &tm-attrib.sparc; &tm-attrib.general; $FreeBSD$ $FreeBSD$ This document provides information for the &os; 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. Almost all &os; 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 for more information. This document may also be of interest to members of the &os; community who want to learn more about how the project works. Administrative Details Login Methods &man.ssh.1;, protocol 2 only Main Shell Host freefall.FreeBSD.org src/ Subversion Root svn+ssh://repo.FreeBSD.org/base (see also ). doc/ Subversion Root svn+ssh://repo.FreeBSD.org/doc (see also ). ports/ Subversion Root svn+ssh://repo.FreeBSD.org/ports (see also ). Internal Mailing Lists 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 can be found in the files /local/mail/repository-name-developers-archive and /local/mail/repository-name-committers-archive on the FreeBSD.org cluster.) Core Team monthly reports /home/core/public/monthly-reports on the FreeBSD.org cluster. Ports Management Team monthly reports /home/portmgr/public/monthly-reports on the FreeBSD.org cluster. Noteworthy src/ SVN Branches stable/8 (8.X-STABLE), stable/9 (9.X-STABLE), stable/10 (10.X-STABLE), head (-CURRENT) &man.ssh.1; is required to connect to the project hosts. For more information, see . Useful links: &os; Project Internal Pages &os; Project Hosts &os; Project Administrative Groups Open<acronym>PGP</acronym> Keys for &os; Cryptographic keys conforming to the OpenPGP (Pretty Good Privacy) standard are used by the &os; project to authenticate committers. Messages carrying important information like public SSH keys can be signed with the OpenPGP key to prove that they are really from the committer. See PGP & GPG: Email for the Practical Paranoid by Michael Lucas and for more information. Creating a Key Existing keys can be used, but should be checked with doc/head/share/pgpkeys/checkkey.sh first. For those who do not yet have an OpenPGP key, or need a new key to meet &os; security requirements, here we show how to generate one. Install security/gnupg. Enter these lines in ~/.gnupg/gpg.conf to set minimum acceptable defaults: fixed-list-mode keyid-format 0xlong personal-digest-preferences SHA512 SHA384 SHA256 SHA224 default-preference-list SHA512 SHA384 SHA256 SHA224 AES256 AES192 AES CAST5 BZIP2 ZLIB ZIP Uncompressed use-agent verify-options show-uid-validity list-options show-uid-validity sig-notation issuer-fpr@notations.openpgp.fifthhorseman.net=%g cert-digest-algo SHA512 Generate a key: &prompt.user; gpg --full-gen-key gpg (GnuPG) 2.1.8; Copyright (C) 2015 Free Software Foundation, Inc. This is free software: you are free to change and redistribute it. There is NO WARRANTY, to the extent permitted by law. Warning: using insecure memory! Please select what kind of key you want: (1) RSA and RSA (default) (2) DSA and Elgamal (3) DSA (sign only) (4) RSA (sign only) Your selection? 1 RSA keys may be between 1024 and 4096 bits long. What keysize do you want? (2048) 2048 Requested keysize is 2048 bits Please specify how long the key should be valid. 0 = key does not expire <n> = key expires in n days <n>w = key expires in n weeks <n>m = key expires in n months <n>y = key expires in n years Key is valid for? (0) 3y Key expires at Wed Nov 4 17:20:20 2015 MST Is this correct? (y/N) y GnuPG needs to construct a user ID to identify your key. Real name: Chucky Daemon Email address: notreal@example.com Comment: You selected this USER-ID: "Chucky Daemon <notreal@example.com>" Change (N)ame, (C)omment, (E)mail or (O)kay/(Q)uit? o You need a Passphrase to protect your secret key. 2048-bit keys with a three-year expiration provide adequate protection at present (2013-12). describes the situation in more detail. A three year key lifespan is short enough to obsolete keys weakened by advancing computer power, but long enough to reduce key management problems. Use your real name here, preferably matching that shown on government-issued ID to make it easier for others to verify your identity. Text that may help others identify you can be entered in the Comment section. After the email address is entered, a passphrase is requested. Methods of creating a secure passphrase are contentious. Rather than suggest a single way, here are some links to sites that describe various methods: , , , . Protect the private key and passphrase. If either the private key or passphrase may have been compromised or disclosed, immediately notify accounts@FreeBSD.org and revoke the key. Committing the new key is shown in . Kerberos and LDAP web Password for &os; Cluster The &os; cluster requires a Kerberos password to access certain services. The Kerberos password also serves as the LDAP web password, since LDAP is proxying to Kerberos in the cluster. Some of the services which require this include: Bugzilla Jenkins To create a new Kerberos account in the &os; cluster, or to reset a Kerberos password for an existing account using a random password generator: &prompt.user; ssh kpasswd.freebsd.org This must be done from a machine outside of the &os;.org cluster. A Kerberos password can also be set manually by logging into freefall.FreeBSD.org and running: &prompt.user; kpasswd Unless the Kerberos-authenticated services of the &os;.org cluster have been used previously, Client unknown will be shown. This error means that the ssh kpasswd.freebsd.org method shown above must be used first to initialize the Kerberos account. Commit Bit Types The &os; 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 &os; 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. Committer Type Responsible Tree Components src core@ src/, doc/ subject to appropriate review doc doceng@ doc/, ports/, src/ documentation ports portmgr@ ports/ 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. 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. Policy for Committer Activity in Other Trees All committers may modify base/head/share/misc/committers-*.dot, base/head/usr.bin/calendar/calendars/calendar.freebsd, and ports/head/astro/xearth/files. 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. Any committer may make changes to any other tree with an "Approved by" from a non-mentored committer with the appropriate bit. Committers can aquire an additional bit by the usual process of finding a mentor who will propose them to core, doceng, or portmgr, as appropriate. When approved, they will be added to 'access' and the normal mentoring period will ensue, which will involve a continuing of Approved by for some period. "Approved by" is only acceptable from non-mentored src committers -- mentored committers can provide a "Reviewed by" but not an "Approved by". Subversion Primer New committers are assumed to already be familiar with the basic operation of Subversion. If not, start by reading the Subversion Book. Introduction The &os; source repository switched from CVS to Subversion on May 31st, 2008. The first real SVN commit is r179447. The &os; doc/www repository switched from CVS to Subversion on May 19th, 2012. The first real SVN commit is r38821. The &os; ports repository switched from CVS to Subversion on July 14th, 2012. The first real SVN commit is r300894. Subversion can be installed from the &os; Ports Collection by issuing these commands: &prompt.root; pkg install subversion Getting Started There are a few ways to obtain a working copy of the tree from Subversion. This section will explain them. Direct Checkout The first is to check out directly from the main repository. For the src tree, use: &prompt.user; svn checkout svn+ssh://repo.freebsd.org/base/head /usr/src For the doc tree, use: &prompt.user; svn checkout svn+ssh://repo.freebsd.org/doc/head /usr/doc For the ports tree, use: &prompt.user; svn checkout svn+ssh://repo.freebsd.org/ports/head /usr/ports Though the remaining examples in this document are written with the workflow of working with the src tree in mind, the underlying concepts are the same for working with the doc and the ports tree. Ports related Subversion operations are listed in . The above command will check out a CURRENT source tree as /usr/src/, 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 head, but that can be renamed safely. svn+ssh means the SVN protocol tunnelled over SSH. The name of the server is repo.freebsd.org, base is the path to the repository, and head is the subdirectory within the repository. If your &os; login name is different from the login name used on the local machine, either include it in the URL (for example svn+ssh://jarjar@repo.freebsd.org/base/head), or add an entry to ~/.ssh/config in the form: Host repo.freebsd.org User jarjar This is the simplest method, but it is hard to tell just yet how much load it will place on the repository. The svn diff does not require access to the server as SVN stores a reference copy of every file in the working copy. This, however, means that Subversion working copies are very large in size. Checkout from a Mirror Check out a working copy from a mirror by substituting the mirror's URL for svn+ssh://repo.freebsd.org/base. This can be an official mirror or a mirror maintained by using svnsync. There is a serious disadvantage to this method: every time something is to be committed, a svn relocate to the master repository has to be done, remembering to svn relocate back to the mirror after the commit. Also, since svn relocate 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. The hassle of a local svnsync 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. <literal>RELENG_*</literal> Branches and General Layout In svn+ssh://repo.freebsd.org/base, base refers to the source tree. Similarly, ports refers to the ports tree, and so on. These are separate repositories with their own change number sequences, access controls and commit mail. For the base repository, HEAD refers to the -CURRENT tree. For example, head/bin/ls is what would go into /usr/src/bin/ls in a release. Some key locations are: /head/ which corresponds to HEAD, also known as -CURRENT. /stable/n which corresponds to RELENG_n. /releng/n.n which corresponds to RELENG_n_n. /release/n.n.n which corresponds to RELENG_n_n_n_RELEASE. /vendor* is the vendor branch import work area. This directory itself does not contain branches, however its subdirectories do. This contrasts with the stable, releng and release directories. /projects and /user feature a branch work area, like in Perforce. As above, the /user directory does not contain branches itself. &os; Documentation Project Branches and Layout In svn+ssh://repo.freebsd.org/doc, doc refers to the repository root of the source tree. In general, most &os; Documentation Project work will be done within the head/ branch of the documentation source tree. &os; documentation is written and/or translated to various languages, each in a separate directory in the head/ branch. Each translation set contains several subdirectories for the various parts of the &os; Documentation Project. A few noteworthy directories are: /articles/ contains the source code for articles written by various &os; contributors. /books/ contains the source code for the different books, such as the &os; Handbook. /htdocs/ contains the source code for the &os; website. &os; Ports Tree Branches and Layout In svn+ssh://repo.freebsd.org/ports, ports refers to the repository root of the ports tree. In general, most &os; port work will be done within the head/ branch of the ports tree which is the actual ports tree used to install software. Some other key locations are: /branches/RELENG_n_n_n which corresponds to RELENG_n_n_n is used to merge back security updates in preparation for a release. /tags/RELEASE_n_n_n which corresponds to RELEASE_n_n_n represents a release tag of the ports tree. /tags/RELEASE_n_EOL represents the end of life tag of a specific &os; branch. Daily Use This section will explain how to perform common day-to-day operations with Subversion. Help SVN has built in help documentation. It can be accessed by typing: &prompt.user; svn help Additional information can be found in the Subversion Book. Checkout As seen earlier, to check out the &os; head branch: &prompt.user; svn checkout svn+ssh://repo.freebsd.org/base/head /usr/src At some point, more than just HEAD 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). To do this, first check out the root of the repository: &prompt.user; svn checkout --depth=immediates svn+ssh://repo.freebsd.org/base This will give base with all the files it contains (at the time of writing, just ROADMAP.txt) and empty subdirectories for head, stable, vendor and so on. Expanding the working copy is possible. Just change the depth of the various subdirectories: &prompt.user; svn up --set-depth=infinity base/head &prompt.user; svn up --set-depth=immediates base/release base/releng base/stable The above command will pull down a full copy of head, plus empty copies of every release tag, every releng branch, and every stable branch. If at a later date merging to 7-STABLE is required, expand the working copy: &prompt.user; svn up --set-depth=infinity base/stable/7 Subtrees do not have to be expanded completely. For instance, expanding only stable/7/sys and then later expand the rest of stable/7: &prompt.user; svn up --set-depth=infinity base/stable/7/sys &prompt.user; svn up --set-depth=infinity base/stable/7 Updating the tree with svn update will only update what was previously asked for (in this case, head and stable/7; it will not pull down the whole tree. Decreasing the depth of a working copy is not possible. Anonymous Checkout 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 back to the main repository. To do this, use: &prompt.user; svn co https://svn.FreeBSD.org/base/head /usr/src More details on using Subversion this way can be found in Using Subversion. Updating the Tree To update a working copy to either the latest revision, or a specific revision: &prompt.user; svn update &prompt.user; svn update -r12345 Status To view the local changes that have been made to the working copy: &prompt.user; svn status To show local changes and files that are out-of-date do: &prompt.user; svn status --show-updates Editing and Committing Unlike Perforce, SVN does not need to be told in advance about file editing. To commit all changes in the current directory and all subdirectories: &prompt.user; svn commit To commit all changes in, for example, lib/libfetch/ and usr/bin/fetch/ in a single operation: &prompt.user; svn commit lib/libfetch usr/bin/fetch There is also a commit wrapper for the ports tree to handle the properties and sanity checking the changes: &prompt.user; /usr/ports/Tools/scripts/psvn commit Adding and Removing Files Before adding files, get a copy of auto-props.txt (there is also a ports tree specific version) and add it to ~/.subversion/config according to the instructions in the file. If you added something before reading this, use svn rm --keep-local 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 svn help. Files are added to a SVN repository with svn add. To add a file named foo, edit it, then: &prompt.user; svn add foo Most new source files should include a $&os;$ string near the start of the file. On commit, svn will expand the $&os;$ string, adding the file path, revision number, date and time of commit, and the username of the committer. Files which cannot be modified may be committed without the $&os;$ string. Files can be removed with svn remove: &prompt.user; svn remove foo Subversion does not require deleting the file before using svn rm, and indeed complains if that happens. It is possible to add directories with svn add: &prompt.user; mkdir bar &prompt.user; svn add bar Although svn mkdir makes this easier by combining the creation of the directory and the adding of it: &prompt.user; svn mkdir bar Like files, directories are removed with svn rm. There is no separate command specifically for removing directories. &prompt.user; svn rm bar Copying and Moving Files This command creates a copy of foo.c named bar.c, with the new file also under version control: &prompt.user; svn copy foo.c bar.c The example above is equivalent to: &prompt.user; cp foo.c bar.c &prompt.user; svn add bar.c To move and rename a file: &prompt.user; svn move foo.c bar.c Log and Annotate svn log shows revisions and commit messages, most recent first, for files or directories. When used on a directory, all revisions that affected the directory and files within that directory are shown. svn annotate, or equally svn praise or svn blame, shows the most recent revision number and who committed that revision for each line of a file. Diffs svn diff displays changes to the working copy. Diffs generated by SVN are unified and include new files by default in the diff output. svn diff can show the changes between two revisions of the same file: &prompt.user; svn diff -r179453:179454 ROADMAP.txt It can also show all changes for a specific changeset. This command shows what changes were made to the current directory and all subdirectories in changeset 179454: &prompt.user; svn diff -c179454 . Reverting Local changes (including additions and deletions) can be reverted using svn revert. It does not update out-of-date files, but just replaces them with pristine copies of the original version. Conflicts If an svn update 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: &prompt.user; svn resolved foo However, the preferred procedure is: &prompt.user; svn resolve --accept=working foo The two examples are equivalent. Possible values for --accept are: working: use the version in your working directory (which one presumes has been edited to resolve the conflicts). base: use a pristine copy of the version you had before svn update, discarding your own changes, the conflicting changes, and possibly other intervening changes as well. mine-full: use what you had before svn update, including your own changes, but discarding the conflicting changes, and possibly other intervening changes as well. theirs-full: use the version that was retrieved when you did svn update, discarding your own changes. Advanced Use Sparse Checkouts SVN allows sparse, or partial checkouts of a directory by adding to a svn checkout. Valid arguments to are: empty: the directory itself without any of its contents. files: the directory and any files it contains. immediates: the directory and any files and directories it contains, but none of the subdirectories' contents. infinity: anything. The --depth option applies to many other commands, including svn commit, svn revert, and svn diff. Since --depth is sticky, there is a --set-depth option for svn update that will change the selected depth. Thus, given the working copy produced by the previous example: &prompt.user; cd ~/freebsd &prompt.user; svn update --set-depth=immediates . The above command will populate the working copy in ~/freebsd with ROADMAP.txt and empty subdirectories, and nothing will happen when svn update is executed on the subdirectories. However, this command will set the depth for head (in this case) to infinity, and fully populate it: &prompt.user; svn update --set-depth=infinity head Direct Operation 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: log, diff mkdir remove, copy, rename propset, propedit, propdel merge Branching is very fast. This command would be used to branch RELENG_8: &prompt.user; svn copy svn+ssh://repo.freebsd.org/base/head svn+ssh://repo.freebsd.org/base/stable/8 This is equivalent to these commands which take minutes and hours as opposed to seconds, depending on your network connection: &prompt.user; svn checkout --depth=immediates svn+ssh://repo.freebsd.org/base &prompt.user; cd base &prompt.user; svn update --set-depth=infinity head &prompt.user; svn copy head stable/8 &prompt.user; svn commit stable/8 Merging with <acronym>SVN</acronym> This section deals with merging code from one branch to another (typically, from head to a stable branch). In all examples below, $FSVN refers to the location of the &os; Subversion repository, svn+ssh://repo.freebsd.org/base/. About Merge Tracking From the user's perspective, merge tracking information (or mergeinfo) is stored in a property called svn:mergeinfo, 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 svn:mergeinfo. It is not inherited. For instance, stable/6/contrib/openpam/ does not implicitly inherit mergeinfo from stable/6/, or stable/6/contrib/. Doing so would make partial checkouts very hard to manage. Instead, mergeinfo is explicitly propagated down the tree. For merging something into branch/foo/bar/, these rules apply: If branch/foo/bar/ does not already have a mergeinfo record, but a direct ancestor (for instance, branch/foo/) does, then that record will be propagated down to branch/foo/bar/ before information about the current merge is recorded. Information about the current merge will not be propagated back up that ancestor. If a direct descendant of branch/foo/bar/ (for instance, branch/foo/bar/baz/) already has a mergeinfo record, information about the current merge will be propagated down to it. If you consider the case where a revision changes several separate parts of the tree (for example, branch/foo/bar/ and branch/foo/quux/), but you only want to merge some of it (for example, branch/foo/bar/), you will see that these rules make sense. If mergeinfo was propagated up, it would seem like that revision had also been merged to branch/foo/quux/, when in fact it had not been. Selecting the Source and Target Branch When Merging Merging to stable/ branches should originate from head/. For example: &prompt.user; svn merge -c r123456 ^/head/ stable/11 &prompt.user; svn commit stable/11 Note the sections below which outline changes to the target location of the stable/ branch starting with stable/10. Merges to releng/ branches should always originate from the corresponding stable/ branch. For example: &prompt.user; svn merge -c r123456 ^/stable/11 releng/11.0 &prompt.user; svn commit releng/11.0 Committers are only permitted to commit to the releng/ branches during a release cycle after receiving approval from the Release Engineering Team, after which only the Security Officer may commit to a releng/ branch for a Security Advisory or Errata Notice. Selecting the Source and Target for <literal>stable/10</literal> and Newer Starting with the stable/10 branch, all merges are merged to and committed from the root of the branch. All merges look like: &prompt.user; svn merge -c r123456 ^/head/ checkout &prompt.user; svn commit checkout Note that checkout must be a complete checkout of the branch to which the merge occurs. Merges to releng/ branches must always originate from the corresponding stable/ branch. For example: &prompt.user; svn merge -c r123456 ^/stable/10 releng/10.0 Selecting the Source and Target for <literal>stable/9</literal> and Older For stable/9 and earlier, a different strategy was used, distributing mergeinfo around the tree so that merges could be performed without a complete checkout. This procedure proved extremely error-prone, with the convenience of partial checkouts for merges significantly outweighed by the complexity of picking mergeinfo targets. The procedure below describes this now-obsoleted process, which should be used only for merges prior to stable/10. Because of mergeinfo propagation, it is important to choose the source and target for the merge carefully to minimise property changes on unrelated directories. The rules for selecting the merge target (the directory where the changes are being merged to) can be summarized: Never merge directly to a file. Never, ever merge directly to a file. Never, ever, ever merge directly to a file. Changes to kernel code are merged to sys/. For instance, a change to the &man.ichwd.4; driver is merged to sys/, not sys/dev/ichwd/. Likewise, a change to the TCP/IP stack is merged to sys/, not sys/netinet/. Changes to code under etc/ is merged at etc/, not below it. Changes to vendor code (code in contrib/, crypto/ and so on) are merged to the directory where vendor imports happen. For instance, a change to crypto/openssl/util/ is merged to crypto/openssl/. This is rarely an issue, however, since changes to vendor code are usually merged wholesale. 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 usr.bin/xlint/arch/i386/ is merged to usr.bin/xlint/. 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 lib/libc/gen/ should be merged to lib/libc/. There may be cases where it makes sense to deviate from the rules for userland programs and libraries. For instance, everything under lib/libpam/ is merged to lib/libpam/, even though the library itself and all of the modules each have their own Makefile. Changes to manual pages are merged to share/man/manN/, for the appropriate value of N. Other changes to share/ are merged to the appropriate subdirectory and not to share/ directly. Changes to a top-level file in the source tree such as UPDATING or Makefile.inc1 are merged directly to that file rather than to the root of the whole tree. Yes, this is an exception to the first three rules. When in doubt, ask. If a merge changes 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 merging a revision that changed a kernel API and updated all the userland bits that used that API, merge the kernel change to sys, and the userland bits to the appropriate userland directories, then commit all of these in one go. The source will almost invariably be the same as the target. For instance, always merge stable/7/lib/libc/ from head/lib/libc/. 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 bin/pkill/ in head to usr.bin/pkill/ in stable/7. Preparing the Merge Target Because of the mergeinfo propagation issues described earlier, it is very important to never merge changes into a sparse working copy. Always use a full checkout of the branch being merged into. For instance, when merging from HEAD to 7, use a full checkout of stable/7: &prompt.user; cd stable/7 &prompt.user; svn up --set-depth=infinity The target directory must also be up-to-date and must not contain any uncommitted changes or stray files. Identifying Revisions Identifying revisions to be merged is a must. If the target already has complete mergeinfo, ask SVN for a list: &prompt.user; cd stable/6/contrib/openpam &prompt.user; svn mergeinfo --show-revs=eligible $FSVN/head/contrib/openpam If the target does not have complete mergeinfo, check the log for the merge source. Merging Now, let us start merging! The Principles For example, To merge: revision $R in directory $target in stable branch $B from directory $source in head $FSVN is svn+ssh://repo.freebsd.org/base Assuming that revisions $P and $Q have already been merged, and that the current directory is an up-to-date working copy of stable/$B, the existing mergeinfo looks like this: &prompt.user; svn propget svn:mergeinfo -R $target $target - /head/$source:$P,$Q Merging is done like so: &prompt.user; svn merge -c$R $FSVN/head/$source $target Checking the results of this is possible with svn diff. The svn:mergeinfo now looks like: &prompt.user; svn propget svn:mergeinfo -R $target $target - head/$source:$P,$Q,$R 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. Practical Example As a practical example, consider this scenario. The changes to netmap.4 in r238987 are to be merged from CURRENT to 9-STABLE. The file resides in head/share/man/man4. According to , 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 . The first step is to inspect the existing mergeinfo. &prompt.user; svn propget svn:mergeinfo -R stable/9/share/man/man4 Take a quick note of how it looks before moving on to the next step; doing the actual merge: &prompt.user; svn merge -c r238987 svn+ssh://repo.freebsd.org/base/head/share/man/man4 stable/9/share/man/man4 --- 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 Check that the revision number of the merged revision has been added. Once this is verified, the only thing left is the actual commit. &prompt.user; svn commit stable/9/share/man/man4 Merging into the Kernel (<filename>sys/</filename>) 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 (sys/). Once svn merge has been executed, svn diff 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 Merge r226222 from head, or similar. Precautions Before Committing As always, build world (or appropriate parts of it). Check the changes with svn diff and svn stat. Make sure all the files that should have been added or deleted were in fact added or deleted. Take a closer look at any property change (marked by a M in the second column of svn stat). Normally, no svn:mergeinfo properties should be anywhere except the target directory (or directories). If something looks fishy, ask for help. Committing 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. Vendor Imports with <acronym>SVN</acronym> Please read this entire section before starting a vendor import. Patches to vendor code fall into two categories: 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 cannot wait until the next vendor release. &os; patches: these are patches that modify the vendor code to address &os;-specific issues. The nature of a patch dictates where it should be committed: Vendor patches must 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 must not 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. &os; patches are committed directly to head. Preparing the Tree 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. Flattening During the conversion from CVS to Subversion, vendor branches were imported with the same layout as the main tree. This means that the pf vendor sources ended up in vendor/pf/dist/contrib/pf. The vendor source is best directly in vendor/pf/dist. To flatten the pf tree: &prompt.user; cd vendor/pf/dist/contrib/pf &prompt.user; svn mv $(svn list) ../.. &prompt.user; cd ../.. &prompt.user; svn rm contrib &prompt.user; svn propdel -R svn:mergeinfo . &prompt.user; svn commit The propdel bit is necessary because starting with 1.5, Subversion will automatically add svn:mergeinfo 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. Tags may be flattened as well (3, 4, 3.5 etc.); the procedure is exactly the same, only changing dist to 3.5 or similar, and putting the svn commit off until the end of the process. Cleaning Up The dist 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. OpenSSH, for example, includes two files that originated with &os; and still contain the original version tags. To do this: &prompt.user; svn propdel svn:keywords -R . &prompt.user; svn commit Bootstrapping Merge History If importing for the first time after the switch to Subversion, bootstrap svn:mergeinfo 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: &prompt.user; cd head/contrib/pf &prompt.user; svn merge --record-only svn+ssh://repo.freebsd.org/base/vendor/pf/dist@180876 . &prompt.user; svn commit Importing New Sources With two commits—one for the import itself and one for the tag—this step can optionally be repeated for every upstream release between the last import and the current import. Preparing the Vendor Sources Unlike in CVS 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. A svn add is required to add any files that were added since the last vendor import, and svn rm 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. &prompt.user; cd vendor/pf/dist &prompt.user; svn list -R | grep -v '/$' | sort >../old &prompt.user; cd ../pf-4.3 &prompt.user; find . -type f | cut -c 3- | sort >../new With these two files, comm -23 ../old ../new will list removed files (files only in old), while comm -13 ../old ../new will list added files only in new. Importing into the Vendor Tree Now, the sources must be copied into dist and the svn add and svn rm commands are used as needed: &prompt.user; cd vendor/pf/pf-4.3 &prompt.user; tar cf - . | tar xf - -C ../dist &prompt.user; cd ../dist &prompt.user; comm -23 ../old ../new | xargs svn rm &prompt.user; comm -13 ../old ../new | xargs svn --parents add If any directories were removed, they will have to be svn rmed manually. Nothing will break if they are not, but they will remain in the tree. Check properties on any new files. All text files should have svn:eol-style set to native. All binary files should have svn:mime-type set to application/octet-stream unless there is a more appropriate media type. Executable files should have svn:executable set to *. No other properties should exist on any file in the tree. Committing is now possible. However, it is good practice to make sure that everything is okay by using the svn stat and svn diff commands. Tagging Once committed, vendor releases are tagged for future reference. The best and quickest way to do this is directly in the repository: &prompt.user; svn cp svn+ssh://repo.freebsd.org/base/vendor/pf/dist svn+ssh://repo.freebsd.org/base/vendor/pf/4.3 Once that is complete, svn up the working copy of vendor/pf to get the new tag, although this is rarely needed. If creating the tag in the working copy of the tree, svn:mergeinfo results must be removed: &prompt.user; cd vendor/pf &prompt.user; svn cp dist 4.3 &prompt.user; svn propdel svn:mergeinfo -R 4.3 Merging to Head &prompt.user; cd head/contrib/pf &prompt.user; svn up &prompt.user; svn merge --accept=postpone svn+ssh://repo.freebsd.org/base/vendor/pf/dist . The --accept=postpone tells Subversion not to complain about merge conflicts as they will be handled manually. The cvs2svn changeover occurred on June 3, 2008. When performing vendor merges for packages which were already present and converted by the cvs2svn process, the command used to merge /vendor/package_name/dist to /head/package_location (for example, head/contrib/sendmail) must use to indicate the revision to merge from the /vendor tree. For example: &prompt.user; svn checkout svn+ssh://repo.freebsd.org/base/head/contrib/sendmail &prompt.user; cd sendmail &prompt.user; svn merge -c r261190 ^/vendor/sendmail/dist . ^ is an alias for the repository path. If using the Zsh shell, the ^ must be escaped with \. This means ^/head should be \^/head. It is necessary to resolve any merge conflicts. 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: &prompt.user; svn diff --no-diff-deleted --old=svn+ssh://repo.freebsd.org/base/vendor/pf/dist --new=. The --no-diff-deleted tells Subversion not to complain about files that are in the vendor tree but not in the main tree. Things that would have previously been removed before the vendor import, like the vendor's makefiles and configure scripts. Using CVS, 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. 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. Committing the Vendor Import 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. From Scratch Importing into the Vendor Tree This section is an example of importing and tagging byacc into head. First, prepare the directory in vendor: &prompt.user; svn co --depth immediates $FSVN/vendor &prompt.user; cd vendor &prompt.user; svn mkdir byacc &prompt.user; svn mkdir byacc/dist Now, import the sources into the dist directory. Once the files are in place, svn add the new ones, then svn commit and tag the imported version. To save time and bandwidth, direct remote committing and tagging is possible: &prompt.user; svn cp -m "Tag byacc 20120115" $FSVN/vendor/byacc/dist $FSVN/vendor/byacc/20120115 Merging to <literal>head</literal> Due to this being a new file, copy it for the merge: &prompt.user; svn cp -m "Import byacc to contrib" $FSVN/vendor/byacc/dist $FSVN/head/contrib/byacc Working normally on newly imported sources is still possible. Reverting a Commit Reverting a commit to a previous version is fairly easy: &prompt.user; svn merge -r179454:179453 ROADMAP.txt &prompt.user; svn commit Change number syntax, with negative meaning a reverse change, can also be used: &prompt.user; svn merge -c -179454 ROADMAP.txt &prompt.user; svn commit This can also be done directly in the repository: &prompt.user; svn merge -r179454:179453 svn+ssh://repo.freebsd.org/base/ROADMAP.txt It is important to ensure that the mergeinfo is correct when reverting a file to permit svn mergeinfo --eligible to work as expected. 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: &prompt.user; svn copy svn+ssh://repo.freebsd.org/base/ROADMAP.txt@179454 &prompt.user; svn commit or, equally: &prompt.user; svn copy svn+ssh://repo.freebsd.org/base/ROADMAP.txt@179454 svn+ssh://repo.freebsd.org/base Do not simply recreate the file manually and svn add it—this will cause history to be lost. Fixing Mistakes 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 svn status and svn diff before committing. Mistakes will happen but, they can generally be fixed without disruption. Take a case of adding a file in the wrong location. The right thing to do is to svn move 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. The wrong thing to do is to delete the file and then svn add 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. Setting up a <application>svnsync</application> Mirror Avoid setting up a svnsync mirror unless there is a very good reason for it. Such reasons might be to support multiple local read-only client machines, or if the 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 international links are involved, expect this to take four to ten times longer. 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. There are several ways to do this: &prompt.user; rsync -va --partial --progress freefall:/home/peter/svnmirror-base-r179637.tbz2 . &prompt.user; rsync -va --partial --progress rsync://repoman.freebsd.org:50873/svnseed/svnmirror-base-r215629.tar.xz . &prompt.user; fetch ftp://ftp.freebsd.org/pub/FreeBSD/development/subversion/svnmirror-base-r221445.tar.xz Extract the file to somewhere like home/svnmirror/base/. Then, update it, so that it fetches changes since the last revision in the archive: &prompt.user; svnsync sync file:///home/svnmirror/base Now, set that up to run from &man.cron.8;, do checkouts locally, set up a svnserve server for local machines to talk to, etc. The seed mirror is set to fetch from svn://svn.freebsd.org/base. The configuration for the mirror is stored in revprop 0 on the local mirror. To see the configuration, try: &prompt.user; svn proplist -v --revprop -r 0 file:///home/svnmirror/base Use propset to change things. Committing High-<acronym>ASCII</acronym> Data Files that have high-ASCII bits are considered binary files in SVN, so the pre-commit checks fail and indicate that the mime-type property should be set to application/octet-stream. However, the use of this is discouraged, so please do not set it. The best way is always avoiding high-ASCII 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 fbsd:notbinary property with propset: &prompt.user; svn propset fbsd:notbinary yes foo.data Maintaining a Project Branch A project branch is one that is synced to head (or another branch) is used to develop a project then commit it back to head. In SVN, dolphin branching is used for this. A dolphin 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 the branch is commited back at the end, it is dead (although a new branch with the same name can be created after the dead one was deleted). As per http://people.freebsd.org/~peter/svn_notes.txt, work that is intended to be merged back into HEAD should be in base/projects/. If the work is beneficial to the &os; community in some way but not intended to be merged directly back into HEAD then the proper location is base/user/username/. This page contains further details. To create a project branch: &prompt.user; svn copy svn+ssh://repo.freebsd.org/base/head svn+ssh://repo.freebsd.org/base/projects/spif To merge changes from HEAD back into the project branch: &prompt.user; cd copy_of_spif &prompt.user; svn merge svn+ssh://repo.freebsd.org/base/head &prompt.user; svn commit It is important to resolve any merge conflicts before committing. Some Tips In commit logs etc., rev 179872 is spelled r179872 as per convention. Speeding up svn is possible by adding these entries to ~/.ssh/config: Host * ControlPath ~/.ssh/sockets/master-%l-%r@%h:%p ControlMaster auto ControlPersist yes and then typing mkdir ~/.ssh/sockets Checking out a working copy with a stock Subversion client without &os;-specific patches (OPTIONS_SET=FREEBSD_TEMPLATE) will mean that $FreeBSD$ tags will not be expanded. Once the correct version has been installed, trick Subversion into expanding them like so: &prompt.user; svn propdel -R svn:keywords . &prompt.user; svn revert -R . This will wipe out uncommitted patches. It is possible to automatically fill the "Sponsored by" and "MFC after" commit log fields by setting "freebsd-sponsored-by" and "freebsd-mfc-after" fields in the "[miscellany]" section of the ~/.subversion/config configuration file. For example: freebsd-sponsored-by = The FreeBSD Foundation freebsd-mfc-after = 2 weeks Setup, Conventions, and Traditions There are a number of things to do as a new developer. The first set of steps is specific to committers only. These steps must be done by a mentor for those who are not committers. For New Committers Those who have been given commit rights to the &os; repositories must follow these steps. Get mentor approval before committing each of these changes! The .ent and .xml files mentioned below exist in the &os; Documentation Project SVN repository at svn+ssh://repo.FreeBSD.org/doc/. New files that do not have the FreeBSD=%H svn:keywords property will be rejected when attempting to commit them to the repository. Be sure to read regarding adding and removing files. Verify that ~/.subversion/config contains the necessary auto-props entries from auto-props.txt mentioned there. All src commits go to &os.current; first before being merged to &os.stable;. The &os.stable; branch must maintain ABI and API compatibility with earlier versions of that branch. Do not merge changes that break this compatibility. Steps for New Committers Add an Author Entity doc/head/share/xml/authors.ent — Add an author entity. Later steps depend on this entity, and missing this step will cause the doc/ build to fail. This is a relatively easy task, but remains a good first test of version control skills. Update the List of Developers and Contributors doc/head/en_US.ISO8859-1/articles/contributors/contrib.committers.xml — Add an entry to the Developers section of the Contributors List. Entries are sorted by last name. doc/head/en_US.ISO8859-1/articles/contributors/contrib.additional.xml — Remove the entry from the Additional Contributors section. Entries are sorted by first name. Add a News Item doc/head/share/xml/news.xml — Add an entry. Look for the other entries that announce new committers and follow the format. Use the date from the commit bit approval email from core@FreeBSD.org. Add a <acronym>PGP</acronym> Key doc/head/share/pgpkeys/pgpkeys.ent and doc/head/share/pgpkeys/pgpkeys-developers.xml - Add your PGP or GnuPG key. Those who do not yet have a key should see . &a.des.email; has written a shell script (doc/head/share/pgpkeys/addkey.sh) to make this easier. See the README file for more information. Use doc/head/share/pgpkeys/checkkey.sh to verify that keys meet minimal best-practices standards. After adding and checking a key, add both updated files to source control and then commit them. Entries in this file are sorted by last name. It is very important to have a current PGP/GnuPG key in the repository. The key may be required for positive identification of a committer. For example, the &a.admins; might need it for account recovery. A complete keyring of FreeBSD.org users is available for download from http://www.FreeBSD.org/doc/pgpkeyring.txt. Update Mentor and Mentee Information base/head/share/misc/committers-repository.dot — Add an entry to the current committers section, where repository is doc, ports, or src, depending on the commit privileges granted. Add an entry for each additional mentor/mentee relationship in the bottom section. Generate a <application>Kerberos</application> Password See to generate or set a Kerberos for use with other &os; services like the bug tracking database. Optional: Enable Wiki Account &os; Wiki Account — A wiki account allows sharing projects and ideas. Those who do not yet have an account can contact clusteradm@FreeBSD.org to obtain one. Optional: Update Wiki Information Wiki Information - After gaining access to the wiki, some people add entries to the How We Got Here, Irc Nicks, and Dogs of FreeBSD pages. Optional: Update Ports with Personal Information ports/astro/xearth/files/freebsd.committers.markers and src/usr.bin/calendar/calendars/calendar.freebsd - Some people add entries for themselves to these files to show where they are located or the date of their birthday. Optional: Prevent Duplicate Mailings Subscribers to &a.svn-src-all.name;, &a.svn-ports-all.name; or &a.svn-doc-all.name; might wish to unsubscribe to avoid receiving duplicate copies of commit messages and followups. For Everyone Introduce yourself to the other developers, otherwise no one will have any idea who you are or what you are working on. The introduction need not be a comprehensive biography, just write a paragraph or two about who you are, what you plan to be working on as a developer in &os;, and who will be your mentor. Email this to the &a.developers; and you will be on your way! Log into freefall.FreeBSD.org and create a /var/forward/user (where user is your username) file containing the e-mail address where you want mail addressed to yourusername@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 freefall may get truncated without warning if space needs to be freed, so forward it or save it elsewhere. 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. To have these checks turned off for your email, create a file named ~/.spam_lover on freefall.FreeBSD.org. Those who are developers but not committers will not be subscribed to the committers or developers mailing lists. The subscriptions are derived from the access rights. Mentors All new developers have a mentor assigned to them for the first few months. A mentor is responsible for teaching the mentee the rules and conventions of the project and guiding their first steps in the developer community. The mentor is also personally responsible for the mentee's actions during this initial period. For committers: do not commit anything without first getting mentor approval. Document that approval with an Approved by: line in the commit message. When the mentor decides that a mentee has learned the ropes and is ready to commit on their own, the mentor announces it with a commit to conf/mentors. This file is in the svnadmin branch of each repository: src base/svnadmin/conf/mentors doc doc/svnadmin/conf/mentors ports ports/svnadmin/conf/mentors Commit Log Messages This section contains some suggestions and traditions for how commit logs are formatted. As well as including an informative message with each commit, some additional information may be needed. This information consists of one or more lines containing the key word or phrase, a colon, tabs for formatting, and then the additional information. The key words or phrases are: PR: The problem report (if any) which is affected (typically, by being closed) by this commit. Multiple PRs may be specified on one line, separated by commas or spaces. Submitted by: The name and e-mail address of the person that submitted the fix; for developers, just the username on the &os; cluster. If the submitter is the maintainer of the port being committed, include "(maintainer)" after the email address. Avoid obfuscating the email address of the submitter as this adds additional work when searching logs. Reviewed by: The name and e-mail address of the person or people that reviewed the change; for developers, just the username on the &os; cluster. If a patch was submitted to a mailing list for review, and the review was favorable, then just include the list name. Approved by: The name and e-mail address of the person or people that approved the change; for developers, just the username on the &os; 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 must be approved by the release engineering team. While under mentorship, get mentor approval before the commit. Enter the mentor's username in this field, and note that they are a mentor: Approved by: username-of-mentor (mentor) If a team approved these commits then include the team name followed by the username of the approver in parentheses. For example: Approved by: re (username) Obtained from: The name of the project (if any) from which the code was obtained. Do not use this line for the name of an individual person. MFC after: If you wish to receive an e-mail reminder to MFC at a later date, specify the number of days, weeks, or months after which an MFC is planned. MFC to: If the commit should be merged to a subset of stable branches, specify the branch names. MFC with: If the commit should be merged together with a previous one in a single MFC commit (for example, where this commit corrects a bug in the previous change), specify the corresponding revision number. Relnotes: If the change is a candidate for inclusion in the release notes for the next release from the branch, set to yes. Security: If the change is related to a security vulnerability or security exposure, include one or more references or a description of the issue. If possible, include a VuXML URL or a CVE ID. Differential Revision: The full URL of the Phabricator review. This line must be the last line. For example: https://reviews.freebsd.org/D1708. Commit Log for a Commit Based on a PR The commit is based on a patch from a PR submitted by John Smith. The commit message PR and Submitted by fields are filled.. ... PR: 12345 Submitted by: John Smith <John.Smith@example.com> Commit Log for a Commit Needing Review The virtual memory system is being changed. After posting patches to the appropriate mailing list (in this case, freebsd-arch) and the changes have been approved. ... Reviewed by: -arch Commit Log for a Commit Needing Approval Commit a port, after working with the listed MAINTAINER, who said to go ahead and commit. ... Approved by: abc (maintainer) Where abc is the account name of the person who approved. Commit Log for a Commit Bringing in Code from OpenBSD Commiting some code based on work done in the OpenBSD project. ... Obtained from: OpenBSD Commit Log for a Change to &os.current; with a Planned Commit to &os.stable; to Follow at a Later Date. Committing some code which will be merged from &os.current; into the &os.stable; branch after two weeks. ... MFC after: 2 weeks Where 2 is the number of days, weeks, or months after which an MFC is planned. The weeks option may be day, days, week, weeks, month, months. It is often necessary to combine these. Consider the situation where a user has submitted a PR containing code from the NetBSD project. Looking at the PR, the developer sees it is not an area of the tree they normally work in, so they have the change reviewed by the arch mailing list. Since the change is complex, the developer opts to MFC after one month to allow adequate testing. The extra information to include in the commit would look something like Example Combined Commit Log PR: 54321 Submitted by: John Smith <John.Smith@example.com> Reviewed by: -arch Obtained from: NetBSD MFC after: 1 month Relnotes: yes Preferred License for New Files The &os; Project suggests and uses this text as the preferred license scheme: /*- * 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] */ 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. The &os; project discourages completely new licenses and variations on the standard licenses. New licenses require the approval of the &a.core; 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. 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. 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. Keeping Track of Licenses Granted to the &os; Project Various software or data exist in the repositories where the &os; project has been granted a special licence to be able to use them. A case in point are the Terminus fonts for use with &man.vt.4;. Here the author Dimitar Zhekov has allowed us to use the "Terminus BSD Console" font under a 2-clause BSD license rather than the regular Open Font License he normally uses. It is clearly sensible to keep a record of any such license grants. To that end, the &a.core; has decided to keep an archive of them. Whenever the &os; project is granted a special license we require the &a.core; to be notified. Any developers involved in arranging such a license grant, please send details to the &a.core; including: Contact details for people or organizations granting the special license. What files, directories etc. in the repositories are covered by the license grant including the revision numbers where any specially licensed material was committed. The date the license comes into effect from. Unless otherwise agreed, this will be the date the license was issued by the authors of the software in question. The license text. A note of any restrictions, limitations or exceptions that apply specifically to &os;'s usage of the licensed material. Any other relevant information. Once the &a.core; is satisfied that all the necessary details have been gathered and are correct, the secretary will send a PGP-signed acknowledgement of receipt including the license details. This receipt will be persistently archived and serve as our permanent record of the license grant. The license archive should contain only details of license grants; this is not the place for any discussions around licensing or other subjects. Access to data within the license archive will be available on request to the &a.core;. Developer Relations When 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. Working on 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. Trying to modify something which is clearly being actively maintained by someone else (and it is only by watching the repository-committers mailing list that a developer can really get a feel for just what is and is not) then consider sending the change to them instead, just as a developer would have before becoming a committer. For ports, contact the listed MAINTAINER in the Makefile. For other parts of the repository, if it is not clear who the active maintainer is, it may help to scan the revision history to see who has committed changes in the past. An example script that lists each person who has committed to a given file along with the number of commits each person has made can be found at on freefall at ~eadler/bin/whodid. If queries go unanswered or the committer otherwise indicates a lack of interest in the area affected, go ahead and commit it. Avoid sending private emails to maintainers. Other people might be interested in the conversation, not just the final output. If there is any doubt about a commit for any reason at all, have it reviewed by -hackers before committing. Better to have it flamed then and there rather than when it is part of the repository. If a commit does results in controversy erupting, it may be advisable to consider backing the change out again until the matter is settled. Remember, with a version control system we can always change it back. Do not impugn the intentions of others. If they see a different solution to a problem, or even a different problem, it is probably not because they are stupid, because they have questionable parentage, or because they are trying to destroy hard work, personal image, or &os;, but basically because they have a different outlook on the world. Different is good. 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. 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. 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. If in Doubt... When unsure about something, whether it be a technical issue or a project convention be sure to ask. If you stay silent you will never make progress. If it relates to a technical issue ask on the public mailing lists. Avoid the temptation to email the individual person that knows the answer. This way everyone will be able to learn from the question and the answer. For project specific or administrative questions ask, in order: Your mentor or former mentor. An experienced committer on IRC, email, etc. Any team with a "hat", as they can give you a definitive answer. If still not sure, ask on &a.developers;. Once your question is answered, if no one pointed you to documentation that spelled out the answer to your question, document it, as others will have the same question. Bugzilla The &os; Project utilizes Bugzilla for tracking bugs and change requests. Be sure that if you commit a fix or suggestion found in the PR database to close it. It is also considered nice if you take time to close any PRs associated with your commits, if appropriate. Committers with non-&os;.org Bugzilla accounts can have the old account merged with the &os;.org account by entering a new bug. Choose Supporting Services as the Product, and Bug Tracker as the Component. You can find out more about Bugzilla at: &os; Problem Report Handling Guidelines http://www.FreeBSD.org/support.html Phabricator The &os; Project utilizes Phabricator for code review requests. See the CodeReview wiki page for details. Who's Who Besides the repository meisters, there are other &os; 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: &a.doceng; doceng is the group responsible for the documentation build infrastructure, approving new documentation committers, and ensuring that the &os; website and documentation on the FTP site is up to date with respect to the subversion 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 charter. Committers interested in contributing to the documentation should familiarize themselves with the Documentation Project Primer. &a.bde.email; 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 &os;. &a.re.members.email; 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. Hiroki is also the keeper of the release documentation (src/release/doc/*). 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. &a.so.email; &a.so; is the &os; Security Officer and oversees the &a.security-officer;. &a.wollman.email; 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 &os;. &a.committers; &a.svn-src-all.name;, &a.svn-ports-all.name; and &a.svn-doc-all.name; are the mailing lists that the version control system uses to send commit messages to. Never send email directly to these lists. Only send replies to this list when they are short and are directly related to a commit. &a.developers; All committers are subscribed to -developers. This list was created to be a forum for the committers community issues. Examples are Core voting, announcements, etc. The &a.developers; is for the exclusive use of &os; committers. To develop &os;, 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 &os;. All &os; committers are expected not to not publish or forward messages from the &a.developers; outside the list membership without permission of all of the authors. 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. This list is not intended as a place for code reviews or for any technical discussion. In fact using it as such hurts the &os; Project as it gives a sense of a closed list where general decisions affecting all of the &os; using community are made without being open. Last, but not least never, never ever, email the &a.developers; and CC:/BCC: another &os; list. Never, ever email another &os; email list and CC:/BCC: the &a.developers;. Doing so can greatly diminish the benefits of this list. SSH Quick-Start Guide If you do not wish to type your password in every time you use &man.ssh.1;, and you use 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 .xsession or .xinitrc. See &man.ssh-agent.1; for details. Generate a key pair using &man.ssh-keygen.1;. The key pair will wind up in your $HOME/.ssh/ directory. Only ECDSA, Ed25519 or RSA keys are supported. Send your public key ($HOME/.ssh/id_ecdsa.pub, $HOME/.ssh/id_ed25519.pub, or $HOME/.ssh/id_rsa.pub) to the person setting you up as a committer so it can be put into yourlogin in /etc/ssh-keys/ on freefall. Now &man.ssh-add.1; can be used for authentication once per session. It prompts for the private key's pass phrase, and then stores it in the authentication agent (&man.ssh-agent.1;). Use ssh-add -d to remove keys stored in the agent. Test with a simple remote command: ssh freefall.FreeBSD.org ls /usr. For more information, see security/openssh, &man.ssh.1;, &man.ssh-add.1;, &man.ssh-agent.1;, &man.ssh-keygen.1;, and &man.scp.1;. For information on adding, changing, or removing &man.ssh.1; keys, see this article. &coverity; Availability for &os; Committers All &os; developers can obtain access to Coverity analysis results of all &os; Project software. All who are interested in obtaining access to the analysis results of the automated Coverity runs, can sign up at Coverity Scan. The &os; wiki includes a mini-guide for developers who are interested in working with the &coverity; analysis reports: http://wiki.freebsd.org/CoverityPrevent. 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. Finally, all &os; developers who are going to use &coverity; are always encouraged to ask for more details and usage information, by posting any questions to the mailing list of the &os; developers. The &os; Committers' Big List of Rules Everyone involved with the &os; project is expected to abide by the Code of Conduct available from http://www.FreeBSD.org/internal/code-of-conduct.html. As committers, you form the public face of the project, and how you behave has a vital impact on the public perception of it. This guide expands on the parts of the Code of Conduct specific to committers. Respect other committers. Respect other contributors. Discuss any significant change before committing. Respect existing maintainers (if listed in the MAINTAINER field in Makefile or in MAINTAINER in the top-level directory). 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. 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. Do not fight in public with other committers; it looks bad. Respect all code freezes and read the committers and developers mailing lists in a timely manner so you know when a code freeze is in effect. When in doubt on any procedure, ask first! Test your changes before committing them. Do not commit to anything under the src/contrib, src/crypto, or src/sys/contrib trees without explicit approval from the respective maintainers. 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 emergency (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 hearing 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 strictly informal 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. In all other aspects of project operation, core is a subset of committers and is bound by the same rules. 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 special powers 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. Details Respect other committers. 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 treat other committers with respect at all times, on public forums and in private email. 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. 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 persons 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 energy economics, 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. 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. Respect other contributors. 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. Consider the points raised under and apply them also to contributors. Discuss any significant change before committing. The repository is not where changes are initially submitted for correctness or argued over, that happens first in the mailing lists or by use of the Phabricator service. The commit will only happen once something resembling consensus has been reached. This does not mean that permission is required before correcting every obvious syntax error or manual page misspelling, just that it is good 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 surprised by those changes. The very best way of making sure that things are on the right track is to have code reviewed by one or more other committers. When in doubt, ask for review! Respect existing maintainers if listed. Many parts of &os; are not owned in the sense that any specific individual will jump up and yell if you commit a change to their area, but it still pays to check first. One convention we use is to put a maintainer line in the Makefile for any package or subtree which is being actively maintained by one or more people; see http://www.FreeBSD.org/doc/en_US.ISO8859-1/books/developers-handbook/policies.html 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 maintainer-ship of something is not clear, look at the repository logs for the files in question and see if someone has been working recently or predominantly in that area. Other areas of &os; fall under the control of someone who manages an overall category of &os; evolution, such as internationalization or networking. See http://www.FreeBSD.org/administration.html for more information on this. 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. 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 very 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. 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. This is another do not argue about it 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. Changes to the security branches (for example, releng/9.3) must be approved by a member of the &a.security-officer;, or in some cases, by a member of the &a.re;. Do not fight in public with other committers; it looks bad. 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. Do not air angry words in public and do not forward private correspondence or other private communications to public mailing lists, mail aliases, instant messaging channels or social media sites. 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 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 third party to resolve the dispute. All parties involved must then agree to be bound by the decision reached by this third party. Respect all code freezes and read the committers and developers mailing list on a timely basis so you know when a code freeze is in effect. 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 &os; Happy Reeducation Camp we run in Greenland. When in doubt on any procedure, ask first! 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 how in the heck do I do this? We already know you are an intelligent person; otherwise, you would not be a committer. Test your changes before committing them. 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 &os; Internal Page for a list of available resources. As other architectures are added to the &os; supported platforms list, the appropriate shared testing resources will be made available. Do not commit to anything under the src/contrib, src/crypto, and src/sys/contrib trees without explicit approval from the respective maintainers. 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 explicit approval from the maintainer (or you are the maintainer), do not commit there! 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, submit your patches to the vendor. If your changes are &os;-specific, talk to the maintainer; they may be willing to apply them locally. But whatever you do, do not commit there by yourself! Contact the &a.core; if you wish to take up maintainership of an unmaintained part of the tree. Policy on Multiple Architectures &os; 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 &os; portable across the platforms we support, core has developed this mandate:
Our 32-bit reference platform is &arch.i386;, and our 64-bit reference platform is &arch.amd64;. 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.
The &arch.i386; and &arch.amd64; platforms were chosen due to being more readily available to developers and as representatives of more diverse processor and system designs - big versus little endian, register file versus register stack, different DMA and cache implementations, hardware page tables versus software TLB management etc. We will continue to re-evaluate this policy as cost and availability of the 64-bit platforms change. 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. Other Suggestions When committing documentation changes, use a spell checker before committing. For all XML docs, verify that the formatting directives are correct by running make lint and textproc/igor. For manual pages, run sysutils/manck and textproc/igor 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 MLINKs installed. 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 doc/ . 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. Deprecating Features When it is necessary to remove functionality from software in the base system, follow these guidelines whenever possible: 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. The option, utility, or interface is preserved until the next major (point zero) release. 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. Privacy and Confidentiality Most &os; business is done in public. &os; is an open project. Which means that not only can anyone use the source code, but that most of the development process is open to public scrutiny. Certain sensitive matters must remain private or held under embargo. There unfortunately cannot be complete transparency. As a &os; developer you will have a certain degree of privileged access to information. Consequently you are expected to respect certain requirements for confidentiality. Sometimes the need for confidentiality comes from external collaborators or has a specific time limit. Mostly though, it is a matter of not releasing private communications. The Security Officer has sole control over the release of security advisories. Where there are security problems that affect many different operating systems, &os; frequently depends on early access to be able to prepare advisories for coordinated release. Unless &os; developers can be trusted to maintain security, such early access will not be made available. The Security Officer is responsible for controlling pre-release access to information about vulnerabilities, and for timing the release of all advisories. He may request help under condition of confidentiality from any developer with relevant knowledge to prepare security fixes. Communications with Core are kept confidential for as long as necessary. Communications to core will initially be treated as confidential. Eventually however, most of Core's business will be summarized into the monthly or quarterly core reports. Care will be taken to avoid publicising any sensitive details. Records of some particularly sensitive subjects may not be reported on at all and will be retained only in Core's private archives. Non-disclosure Agreements may be required for access to certain commercially sensitive data. Access to certain commercially sensitive data may only be available under a Non-Disclosure Agreement. The FreeBSD Foundation legal staff must be consulted before any binding agreements are entered into. Private communications must not be made public without permission. Beyond the specific requirements above there is a general expectation not to publish private communications between developers without the consent of all parties involved. Ask permission before forwarding a message onto a public mailing list, or posting it to a forum or website that can be accessed by other than the original correspondents. Communications on project-only or restricted access channels must be kept private. Similarly to personal communications, certain internal communications channels, including &os; Committer only mailing lists and restricted access IRC channels are considered private communications. Permission is required to publish material from these sources. Core may approve publication. Where it is impractical to obtain permission due to the number of correspondents or where permission to publish is unreasonably withheld, Core may approve release of such private matters that merit more general publication. Support for Multiple Architectures &os; 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 &os; 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 &os; target audience. Statement of General Intent The &os; 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 &os; 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 &os; 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. The &os; 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. Tier 1: Fully Supported Architectures 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 &os; cluster for actual hardware. Tier 1 architectures are expected to be Production Quality with respects to all aspects of the &os; operating system, including installation and development environments. 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. 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. 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. 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. Current Tier 1 platforms are &arch.i386; and &arch.amd64;. Tier 2: Developmental Architectures 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 maintainers are 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 &os;-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 &os; should be feasible to implement on these platforms, but an implementation is not required before the feature may be added to the &os; 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 &os; 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 &os; 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. 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. Tier 2 architectures have basic 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. Tier 2 architectures can be integrated into the &os; 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 be relatively recent. It should be integrated into the &os; documentation. Current Tier 2 platforms are &arch.arm;, &arch.arm64;, &arch.ia64; (through &os; 10), &arch.pc98;, &arch.powerpc;, and &arch.sparc64;. Tier 3: Experimental Architectures Tier 3 platforms are not supported by the security officer and release engineering teams. At the discretion of the toolchain maintainers, 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. Initial support for Tier 3 platforms is worked on in external SCM repositories. The transition to &os;'s subversion takes place after the platform boots multi-user on hardware; sharing via subversion is needed for wider exposure; and multiple developers are actively working on the platform. Platforms that transition to Tier 3 status may be removed from the tree if they are no longer actively supported by the &os; developer community at the discretion of the release engineer. Tier 3 platforms may have ports support, either integrated or external, but do not require it. 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 &os; tree. Current Tier 3 platforms are &arch.mips;, and &arch.riscv;. Tier 4: Unsupported Architectures Tier 4 systems are not supported in any form by the project. All systems not otherwise classified into a support tier are Tier 4 systems. The &arch.ia64; platform is transitioning to Tier 4 status in &os; 11. Policy on Changing the Tier of an Architecture Systems may only be moved from one tier to another by approval of the &os; Core Team, which shall make that decision in collaboration with the Security Officer, Release Engineering, and toolchain maintenance teams. Ports Specific FAQ Adding a New Port How do I add a new port? First, please read the section about repository copies. The easiest way to add a new port is the addport script located in the ports/Tools/scripts directory. It adds a port from the directory specified, determining the category automatically from the port Makefile. It also adds an entry to the port's category Makefile. It was written by &a.mharo.email;, &a.will.email;, and &a.garga.email;. When sending questions about this script to the &a.ports;, please also CC &a.crees.email;, the current maintainer. Any other things I need to know when I add a new port? Check the port, preferably to make sure it compiles and packages correctly. This is the recommended sequence: &prompt.root; make install - &prompt.root; make package - &prompt.root; make deinstall - &prompt.root; pkg add package you built above - &prompt.root; make deinstall - &prompt.root; make reinstall - &prompt.root; make package +&prompt.root; make package +&prompt.root; make deinstall +&prompt.root; pkg add package you built above +&prompt.root; make deinstall +&prompt.root; make reinstall +&prompt.root; make package The Porters Handbook contains more detailed instructions. 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. If the port came from a submitter who has not contributed to the Project before, add that person's name to the Additional Contributors section of the &os; Contributors List. Close the PR if the port came in as a PR. To close a PR, change the state to Issue Resolved and the resolution as Fixed. Removing an Existing Port How do I remove an existing port? 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. Make sure there is no dependency on the port in the ports collection: The port's PKGNAME appears in exactly one line in a recent INDEX file. No other ports contains any reference to the port's directory or PKGNAME in their Makefiles When using Git, consider using git grep, it is much faster than grep -r. Then, remove the port: Remove the port's files and directory with svn remove. Remove the SUBDIR listing of the port in the parent directory Makefile. Add an entry to ports/MOVED. Search for entries in ports/security/vuxml/vuln.xml and adjust them accordingly. In particular, check for previous packages with the new name which version could include the new port. Remove the port from ports/LEGAL if it is there. Alternatively, you can use the rmport script, from ports/Tools/scripts. This script was written by &a.vd.email;. When sending questions about this script to the &a.ports;, please also CC &a.crees.email;, the current maintainer. Re-adding a Deleted Port How do I re-add a deleted port? This is essentially the reverse of deleting a port. Do not use svn add to add the port. Follow these steps. If they are unclear, or are not working, ask for help, do not just svn add the port. Figure out when the port was removed. Use this list, or look for the port on freshports, and then copy the last living revision of the port: &prompt.user; cd /usr/ports/category &prompt.user; svn cp 'svn+ssh://repo.freebsd.org/ports/head/category/portname/@XXXXXX' portname Pick the revision that is just before the removal. For example, if the revision where it was removed is 269874, use 269873. It is also possible to specify a date. In that case, pick a date that is before the removal but after the last commit to the port. &prompt.user; cd /usr/ports/category &prompt.user; svn cp 'svn+ssh://repo.freebsd.org/ports/head/category/portname/@{YYYY-MM-DD}' portname Make the changes necessary to get the port working again. If it was deleted because the distfiles are no longer available, either volunteer to host the distfiles, or find someone else to do so. If some files have been added, or were removed during the resurrection process, use svn add or svn remove to make sure all the files in the port will be committed. Restore the SUBDIR listing of the port in the parent directory Makefile, keeping the entries sorted. Delete the port entry from ports/MOVED. If the port had an entry in ports/LEGAL, restore it. svn commit these changes, preferably in one step. The addport script mentioned in now detects when the port to add has previously existed, and attempts to handle all except the ports/LEGAL step automatically. Repository Copies When do we need a repository copy? 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 related means it is a different version or a slightly modified version. Examples are print/ghostscript* (different versions) and x11-wm/windowmaker* (English-only and internationalized version). Another example is when a port is moved from one subdirectory to another, or when the name of a directory must be changed because the authors renamed their software even though it is a descendant of a port already in a tree. What do I need to do? With Subversion, a repo copy can be done by any committer: Doing a repo copy: Verify that the target directory does not exist. Use svn up to make certain the original files, directories, and checkout information is current. Use svn move or svn copy to do the repo copy. Upgrade the copied port to the new version. Remember to add or change the PKGNAMEPREFIX or PKGNAMESUFFIX so there are no duplicate ports with the same name. In some rare cases it may be necessary to change the PORTNAME instead of adding PKGNAMEPREFIX or PKGNAMESUFFIX, but this is only done when it is really needed — for example, 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 textproc/libxml to textproc/libxml2. In most cases, adding or changing PKGNAMEPREFIX or PKGNAMESUFFIX suffices. Add the new subdirectory to the SUBDIR listing in the parent directory Makefile. You can run make checksubdirs in the parent directory to check this. If the port changed categories, modify the CATEGORIES line of the port's Makefile accordingly Add an entry to ports/MOVED, if you remove the original port. Commit all changes on one commit. When removing a port: Perform a thorough check of the ports collection for any dependencies on the old port location/name, and update them. Running grep on INDEX is not enough because some ports have dependencies enabled by compile-time options. A full grep -r of the ports collection is recommended. Remove the old port and the old SUBDIR entry. Add an entry to ports/MOVED. After repo moves (rename operations where a port is copied and the old location is removed): 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. Ports Freeze What is a ports freeze? A ports freeze was a restricted state the ports tree was put in before a release. It was used to ensure a higher quality for the packages shipped with a release. It usually lasted a couple of weeks. During that time, build problems were fixed, and the release packages were built. This practice is no longer used, as the packages for the releases are built from the current stable, quarterly branch. For more information on how to merge commits to the quarterly branch, see . Quarterly Branches What is the procedure to request authorization for merging a commit to the quarterly branch? When doing the commit, add the branch name to the MFH: line, for example: MFH: 2014Q1 It will automatically notify the &a.ports-secteam; and the &a.portmgr;. They will then decide if the commit can be merged and answer with the procedure. If the commit has already been made, send an email to the &a.ports-secteam; and the &a.portmgr; with the revision number and a small description of why the commit needs to be merged. Are there any changes that can be committed without approval? The following blanket approvals are in effect: Fixes that do not result in a change in contents of the resulting package. For example: pkg-descr: WWW: URL updates (existing 404, moved or incorrect) Build, runtime or packaging fixes, if the quarterly branch version is currently broken. These fixes must be tested on the quarterly branch. Missing dependencies (detected, linked against but not registered via *_DEPENDS). Fixing shebangs, stripping installed libraries and binaries, and plist fixes. Backport of security and reliability fixes which only result in PORTREVISION bumps and no changes to enabled features. for example, adding a patch fixing a buffer overflow. No unauthorized commits can ever be made without approval of either &a.ports-secteam; or &a.portmgr;. What is the procedure for merging commits to the quarterly branch? A script is provided to automate merging a specific commit: ports/Tools/scripts/mfh. It is used as follows: &prompt.user; /usr/ports/Tools/scripts/mfh 380362 U 2015Q1 Checked out revision 380443. A 2015Q1/security Updating '2015Q1/security/rubygem-sshkit': A 2015Q1/security/rubygem-sshkit A 2015Q1/security/rubygem-sshkit/Makefile A 2015Q1/security/rubygem-sshkit/distinfo A 2015Q1/security/rubygem-sshkit/pkg-descr Updated to revision 380443. --- Merging r380362 into '2015Q1': U 2015Q1/security/rubygem-sshkit/Makefile U 2015Q1/security/rubygem-sshkit/distinfo --- Recording mergeinfo for merge of r380362 into '2015Q1': U 2015Q1 --- Recording mergeinfo for merge of r380362 into '2015Q1/security': G 2015Q1/security --- Eliding mergeinfo from '2015Q1/security': U 2015Q1/security --- Recording mergeinfo for merge of r380362 into '2015Q1/security/rubygem-sshkit': G 2015Q1/security/rubygem-sshkit --- Eliding mergeinfo from '2015Q1/security/rubygem-sshkit': U 2015Q1/security/rubygem-sshkit M 2015Q1 M 2015Q1/security/rubygem-sshkit/Makefile M 2015Q1/security/rubygem-sshkit/distinfo Index: 2015Q1/security/rubygem-sshkit/Makefile =================================================================== --- 2015Q1/security/rubygem-sshkit/Makefile (revision 380443) +++ 2015Q1/security/rubygem-sshkit/Makefile (working copy) @@ -2,7 +2,7 @@ # $FreeBSD$ PORTNAME= sshkit -PORTVERSION= 1.6.1 +PORTVERSION= 1.7.0 CATEGORIES= security rubygems MASTER_SITES= RG Index: 2015Q1/security/rubygem-sshkit/distinfo =================================================================== --- 2015Q1/security/rubygem-sshkit/distinfo (revision 380443) +++ 2015Q1/security/rubygem-sshkit/distinfo (working copy) @@ -1,2 +1,2 @@ -SHA256 (rubygem/sshkit-1.6.1.gem) = 8ca67e46bb4ea50fdb0553cda77552f3e41b17a5aa919877d93875dfa22c03a7 -SIZE (rubygem/sshkit-1.6.1.gem) = 135680 +SHA256 (rubygem/sshkit-1.7.0.gem) = 90effd1813363bae7355f4a45ebc8335a8ca74acc8d0933ba6ee6d40f281a2cf +SIZE (rubygem/sshkit-1.7.0.gem) = 136192 Index: 2015Q1 =================================================================== --- 2015Q1 (revision 380443) +++ 2015Q1 (working copy) Property changes on: 2015Q1 ___________________________________________________________________ Modified: svn:mergeinfo Merged /head:r380362 Do you want to commit? (no = start a shell) [y/n] At that point, the script will either open a shell for you to fix things, or open your text editor with the commit message all prepared and then commit the merge. The script assumes that you can connect to repo.FreeBSD.org with SSH directly, so if your local login name is different than your &os; cluster account, you need a few lines in your ~/.ssh/config: Host repo.freebsd.org # Can be *.freebsd.org User freebsd-login The script is also able to merge more than one revision at a time. If there have been other updates to the port since the branch was created that have not been merged because they were not security related. Add the different revisions in the order they were committed on the mfh command line. The new commit log message will contain the combined log messages from all the original commits. These messages must be edited to show what is actually being done with the new commit. &prompt.user; /usr/ports/Tools/scripts/mfh r407208 r407713 r407722 r408567 r408943 r410728 The mfh script can also take an optional first argument, the branch where the merge is being done. Only the latest quarterly branch is supported, so specifying the branch is discouraged. To be safe, the script will give a warning if the quarterly branch is not the latest: &prompt.user; /usr/ports/Tools/scripts/mfh 2016Q1 r407208 r407713 /!\ The latest branch is 2016Q2, do you really want to commit to 2016Q1? [y/n] Creating a New Category What is the procedure for creating a new category? Please see Proposing a New Category in the Porter's Handbook. Once that procedure has been followed and the PR has been assigned to the &a.portmgr;, it is their decision whether or not to approve it. If they do, it is their responsibility to: Perform any needed moves. (This only applies to physical categories.) Update the VALID_CATEGORIES definition in ports/Mk/bsd.port.mk. Assign the PR back to you. What do I need to do to implement a new physical category? Upgrade each moved port's Makefile. Do not connect the new category to the build yet. To do this, you will need to: Change the port's CATEGORIES (this was the point of the exercise, remember?) The new category is listed first. This will help to ensure that the PKGORIGIN is correct. Run a make describe. Since the top-level make index that you will be running in a few steps is an iteration of make describe over the entire ports hierarchy, catching any errors here will save you having to re-run that step later on. If you want to be really thorough, now might be a good time to run &man.portlint.1;. Check that the PKGORIGINs are correct. The ports system uses each port's CATEGORIES entry to create its PKGORIGIN, 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. To do this, use the chkorigin.sh tool: env PORTSDIR=/path/to/ports sh -e /path/to/ports/Tools/scripts/chkorigin.sh. This will check every port in the ports tree, even those not connected to the build, so you can run it directly after the move operation. Hint: do not forget to look at the PKGORIGINs of any slave ports of the ports you just moved! On your own local system, test the proposed changes: first, comment out the SUBDIR entries in the old ports' categories' Makefiles; then enable building the new category in ports/Makefile. Run make checksubdirs in the affected category directories to check the SUBDIR entries. Next, in the ports/ directory, run make index. This can take over 40 minutes on even modern systems; however, it is a necessary step to prevent problems for other people. Once this is done, you can commit the updated ports/Makefile to connect the new category to the build and also commit the Makefile changes for the old category or categories. Add appropriate entries to ports/MOVED. Update the documentation by modifying: the list of categories in the Porter's Handbook doc/en_US.ISO8859-1/htdocs/ports. Note that these are now displayed by sub-groups, as specified in doc/en_US.ISO8859-1/htdocs/ports/categories.descriptions. (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. 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. It is not necessary to manually update the ports web pages to reflect the new category. This is done automatically via the change to en_US.ISO8859-1/htdocs/ports/categories and the automated rebuild of INDEX. What do I need to do to implement a new virtual category? This is much simpler than a physical category. Only a few modifications are needed: the list of categories in the Porter's Handbook en_US.ISO8859-1/htdocs/ports/categories Miscellaneous Questions How do I know if my port is building correctly or not? The packages are built multiple times each week. If a port fails, the maintainer will receive an email from pkg-fallout@FreeBSD.org. Reports for all the package builds (official, experimental, and non-regression) are aggregated at pkg-status.FreeBSD.org. I added a new port. Do I need to add it to the INDEX? No. The file can either be generated by running make index, or a pre-generated version can be downloaded with make fetchindex. Are there any other files I am not allowed to touch? Any file directly under ports/, or any file under a subdirectory that starts with an uppercase letter (Mk/, Tools/, etc.). In particular, the &a.portmgr; is very protective of ports/Mk/bsd.port*.mk so do not commit changes to those files unless you want to face their wra(i)th. What is the proper procedure for updating the checksum for a port's distfile when the file changes without a version change? When the checksum for a distribution file is updated due to the author updating the file without changing the port's revision, the commit message includes 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. How can an experimental test build of the ports tree (exp-run) be requested? An exp-run must be completed before patches with a significant ports impact are committed. The patch can be against the ports tree or the base system. Full package builds will be done with the patches provided by the submitter, and the submitter is required to fix detected problems (fallout) before commit. Go to the Bugzilla new PR page. Select the product your patch is about. Fill in the bug report as normal. Remember to attach the patch. If at the top it says Show Advanced Fields click on it. It will now say Hide Advanced Fields. Many new fields will be available. If it already says Hide Advanced Fields, no need to do anything. In the Flags section, set the exp-run one to ?. As for all other fields, hovering the mouse over any field shows more details. Submit. When the &a.portmgr; replies, fix the fallout. The fallout might be a bug in the original patch or other ports that need to be fixed. Both need to be addressed. Update the patch to repair all the fallout and repeat. Issues Specific to Developers Who Are Not Committers A few people who have access to the &os; machines do not have commit bits. 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: Administrative Details Conventions Get your mentor to add you to the Additional Contributors (doc/en_US.ISO8859-1/articles/contributors/contrib.additional.xml), if you are not already listed there. Developer Relations SSH Quick-Start Guide The &os; Committers' Big List of Rules Information About &ga; As of December 12, 2012, &ga; was enabled on the &os; Project website to collect anonymized usage statistics regarding usage of the site. The information collected is valuable to the &os; Documentation Project, to identify various problems on the &os; website. &ga; General Policy The &os; Project takes visitor privacy very seriously. As such, the &os; Project website honors the Do Not Track header before fetching the tracking code from Google. For more information, please see the &os; Privacy Policy. &ga; access is not arbitrarily allowed — access must be requested, voted on by the &a.doceng;, and explicitly granted. Requests for &ga; data must include a specific purpose. For example, a valid reason for requesting access would be to see the most frequently used web browsers when viewing &os; web pages to ensure page rendering speeds are acceptable. Conversely, to see what web browsers are most frequently used (without stating why) would be rejected. All requests must include the timeframe for which the data would be required. For example, it must be explicitly stated if the requested data would be needed for a timeframe covering a span of 3 weeks, or if the request would be one-time only. Any request for &ga; data without a clear, reasonable reason beneficial to the &os; Project will be rejected. Data Available Through &ga; A few examples of the types of &ga; data available include: Commonly used web browsers Page load times Site access by language Miscellaneous Questions Why are trivial or cosmetic changes to files on a vendor branch a bad idea? From now on, every new vendor release of that file will need to have patches merged in by hand. From now on, every new vendor release of that file will need to have patches verified by hand. How do I add a new file to a branch? 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 doc and ports trees. The src tree uses SVN and requires more care because of the mergeinfo properties. See the Subversion Primer for details on how to perform an MFC. How do I access people.FreeBSD.org to put up personal or project information? people.FreeBSD.org is the same as freefall.FreeBSD.org. Just create a public_html directory. Anything you place in that directory will automatically be visible under http://people.FreeBSD.org/. Where are the mailing list archives stored? The mailing lists are archived under /local/mail on freefall.FreeBSD.org. I would like to mentor a new committer. What process do I need to follow? See the New Account Creation Procedure document on the internal pages. Benefits and Perks for &os; Comitters Recognition Recognition as a competent software engineer is the longest lasting value. In addition, getting a chance to work with some of the best people that every engineer would dream of meeting is a great perk! FreeBSD Mall &os; committers can get a free 4-CD or DVD set at conferences from &os; Mall, Inc.. <acronym>IRC</acronym> In addition, developers may request a cloaked hostmask for their account on the Freenode IRC network in the form of freebsd/developer/freefall name or freebsd/developer/NickServ name. To request a cloak, send an email to &a.irc.email; with your requested hostmask and NickServ account name. <systemitem class="domainname">Gandi.net</systemitem> Gandi provides website hosting, cloud computing, domain registration, and X.509 certificate services. Gandi offers an E-rate discount to all &os; developers. Send mail to non-profit@gandi.net using your @freebsd.org mail address, and indicate your Gandi handle.
Index: head/en_US.ISO8859-1/articles/vinum/article.xml =================================================================== --- head/en_US.ISO8859-1/articles/vinum/article.xml (revision 50803) +++ head/en_US.ISO8859-1/articles/vinum/article.xml (revision 50804) @@ -1,1285 +1,1285 @@
The <filename>vinum</filename> Volume Manager Greg Lehey Originally written by Synopsis No matter the type of disks, there are always potential problems. The disks can be too small, too slow, or too unreliable to meet the system's requirements. While disks are getting bigger, so are data storage requirements. Often a file system is needed that is bigger than a disk's capacity. Various solutions to these problems have been proposed and implemented. One method is through the use of multiple, and sometimes redundant, disks. In addition to supporting various cards and controllers for hardware Redundant Array of Independent Disks RAID systems, the base &os; system includes the vinum volume manager, a block device driver that implements virtual disk drives and addresses these three problems. vinum provides more flexibility, performance, and reliability than traditional disk storage and implements RAID-0, RAID-1, and RAID-5 models, both individually and in combination. This chapter provides an overview of potential problems with traditional disk storage, and an introduction to the vinum volume manager. Starting with &os; 5, vinum has been rewritten in order to fit into the GEOM architecture, while retaining the original ideas, terminology, and on-disk metadata. This rewrite is called gvinum (for GEOM vinum). While this chapter uses the term vinum, any command invocations should be performed with gvinum. The name of the kernel module has changed from the original vinum.ko to geom_vinum.ko, and all device nodes reside under /dev/gvinum instead of /dev/vinum. As of &os; 6, the original vinum implementation is no longer available in the code base. Access Bottlenecks Modern systems frequently need to access data in a highly concurrent manner. For example, large FTP or HTTP servers can maintain thousands of concurrent sessions and have multiple 100 Mbit/s connections to the outside world, well beyond the sustained transfer rate of most disks. Current disk drives can transfer data sequentially at up to 70 MB/s, but this value is of little importance in an environment where many independent processes access a drive, and where they may achieve only a fraction of these values. In such cases, it is more interesting to view the problem from the viewpoint of the disk subsystem. The important parameter is the load that a transfer places on the subsystem, or the time for which a transfer occupies the drives involved in the transfer. In any disk transfer, the drive must first position the heads, wait for the first sector to pass under the read head, and then perform the transfer. These actions can be considered to be atomic as it does not make any sense to interrupt them. Consider a typical transfer of about 10 kB: the current generation of high-performance disks can position the heads in an average of 3.5 ms. The fastest drives spin at 15,000 rpm, so the average rotational latency (half a revolution) is 2 ms. At 70 MB/s, the transfer itself takes about 150 μs, almost nothing compared to the positioning time. In such a case, the effective transfer rate drops to a little over 1 MB/s and is clearly highly dependent on the transfer size. The traditional and obvious solution to this bottleneck is more spindles: rather than using one large disk, use several smaller disks with the same aggregate storage space. Each disk is capable of positioning and transferring independently, so the effective throughput increases by a factor close to the number of disks used. The actual throughput improvement is smaller than the number of disks involved. Although each drive is capable of transferring in parallel, there is no way to ensure that the requests are evenly distributed across the drives. Inevitably the load on one drive will be higher than on another. disk concatenation Vinum concatenation The evenness of the load on the disks is strongly dependent on the way the data is shared across the drives. In the following discussion, it is convenient to think of the disk storage as a large number of data sectors which are addressable by number, rather like the pages in a book. The most obvious method is to divide the virtual disk into groups of consecutive sectors the size of the individual physical disks and store them in this manner, rather like taking a large book and tearing it into smaller sections. This method is called concatenation and has the advantage that the disks are not required to have any specific size relationships. It works well when the access to the virtual disk is spread evenly about its address space. When access is concentrated on a smaller area, the improvement is less marked. illustrates the sequence in which storage units are allocated in a concatenated organization.
Concatenated Organization
disk striping Vinum striping RAID An alternative mapping is to divide the address space into smaller, equal-sized components and store them sequentially on different devices. For example, the first 256 sectors may be stored on the first disk, the next 256 sectors on the next disk and so on. After filling the last disk, the process repeats until the disks are full. This mapping is called striping or RAID-0. RAID offers various forms of fault tolerance, though RAID-0 is somewhat misleading as it provides no redundancy. Striping requires somewhat more effort to locate the data, and it can cause additional I/O load where a transfer is spread over multiple disks, but it can also provide a more constant load across the disks. illustrates the sequence in which storage units are allocated in a striped organization.
Striped Organization
Data Integrity The final problem with disks is that they are unreliable. Although reliability has increased tremendously over the last few years, disk drives are still the most likely core component of a server to fail. When they do, the results can be catastrophic and replacing a failed disk drive and restoring data can result in server downtime. disk mirroring vinum mirroring RAID-1 One approach to this problem is mirroring, or RAID-1, which keeps two copies of the data on different physical hardware. Any write to the volume writes to both disks; a read can be satisfied from either, so if one drive fails, the data is still available on the other drive. Mirroring has two problems: It requires twice as much disk storage as a non-redundant solution. Writes must be performed to both drives, so they take up twice the bandwidth of a non-mirrored volume. Reads do not suffer from a performance penalty and can even be faster. RAID-5 An alternative solution is parity, implemented in RAID levels 2, 3, 4 and 5. Of these, RAID-5 is the most interesting. As implemented in vinum, it is a variant on a striped organization which dedicates one block of each stripe to parity one of the other blocks. As implemented by vinum, a RAID-5 plex is similar to a striped plex, except that it implements RAID-5 by including a parity block in each stripe. As required by RAID-5, the location of this parity block changes from one stripe to the next. The numbers in the data blocks indicate the relative block numbers.
<acronym>RAID</acronym>-5 Organization
Compared to mirroring, RAID-5 has the advantage of requiring significantly less storage space. Read access is similar to that of striped organizations, but write access is significantly slower, approximately 25% of the read performance. If one drive fails, the array can continue to operate in degraded mode where a read from one of the remaining accessible drives continues normally, but a read from the failed drive is recalculated from the corresponding block from all the remaining drives. <filename>vinum</filename> Objects In order to address these problems, vinum implements a four-level hierarchy of objects: The most visible object is the virtual disk, called a volume. Volumes have essentially the same properties as a &unix; disk drive, though there are some minor differences. For one, they have no size limitations. Volumes are composed of plexes, each of which represent the total address space of a volume. This level in the hierarchy provides redundancy. Think of plexes as individual disks in a mirrored array, each containing the same data. Since vinum exists within the &unix; disk storage framework, it would be possible to use &unix; partitions as the building block for multi-disk plexes. In fact, this turns out to be too inflexible as &unix; disks can have only a limited number of partitions. Instead, vinum subdivides a single &unix; partition, the drive, into contiguous areas called subdisks, which are used as building blocks for plexes. Subdisks reside on vinum drives, currently &unix; partitions. vinum drives can contain any number of subdisks. With the exception of a small area at the beginning of the drive, which is used for storing configuration and state information, the entire drive is available for data storage. The following sections describe the way these objects provide the functionality required of vinum. Volume Size Considerations Plexes can include multiple subdisks spread over all drives in the vinum configuration. As a result, the size of an individual drive does not limit the size of a plex or a volume. Redundant Data Storage vinum implements mirroring by attaching multiple plexes to a volume. Each plex is a representation of the data in a volume. A volume may contain between one and eight plexes. Although a plex represents the complete data of a volume, it is possible for parts of the representation to be physically missing, either by design (by not defining a subdisk for parts of the plex) or by accident (as a result of the failure of a drive). As long as at least one plex can provide the data for the complete address range of the volume, the volume is fully functional. Which Plex Organization? vinum implements both concatenation and striping at the plex level: A concatenated plex uses the address space of each subdisk in turn. Concatenated plexes are the most flexible as they can contain any number of subdisks, and the subdisks may be of different length. The plex may be extended by adding additional subdisks. They require less CPU time than striped plexes, though the difference in CPU overhead is not measurable. On the other hand, they are most susceptible to hot spots, where one disk is very active and others are idle. A striped plex stripes the data across each subdisk. The subdisks must all be the same size and there must be at least two subdisks in order to distinguish it from a concatenated plex. The greatest advantage of striped plexes is that they reduce hot spots. By choosing an optimum sized stripe, about 256 kB, the load can be evened out on the component drives. Extending a plex by adding new subdisks is so complicated that vinum does not implement it. summarizes the advantages and disadvantages of each plex organization. <filename>vinum</filename> Plex Organizations Plex type Minimum subdisks Can add subdisks Must be equal size Application concatenated 1 yes no Large data storage with maximum placement flexibility and moderate performance striped 2 no yes High performance in combination with highly concurrent access
Some Examples vinum maintains a configuration database which describes the objects known to an individual system. Initially, the user creates the configuration database from one or more configuration files using &man.gvinum.8;. vinum stores a copy of its configuration database on each disk device under its control. This database is updated on each state change, so that a restart accurately restores the state of each vinum object. The Configuration File The configuration file describes individual vinum objects. The definition of a simple volume might be: drive a device /dev/da3h volume myvol plex org concat sd length 512m drive a This file describes four vinum objects: The drive line describes a disk partition (drive) and its location relative to the underlying hardware. It is given the symbolic name a. This separation of symbolic names from device names allows disks to be moved from one location to another without confusion. The volume line describes a volume. The only required attribute is the name, in this case myvol. The plex line defines a plex. The only required parameter is the organization, in this case concat. No name is necessary as the system automatically generates a name from the volume name by adding the suffix .px, where x is the number of the plex in the volume. Thus this plex will be called myvol.p0. The sd line describes a subdisk. The minimum specifications are the name of a drive on which to store it, and the length of the subdisk. No name is necessary as the system automatically assigns names derived from the plex name by adding the suffix .sx, where x is the number of the subdisk in the plex. Thus vinum gives this subdisk the name myvol.p0.s0. After processing this file, &man.gvinum.8; produces the following output: - &prompt.root; gvinum -> create config1 - Configuration summary - Drives: 1 (4 configured) - Volumes: 1 (4 configured) - Plexes: 1 (8 configured) - Subdisks: 1 (16 configured) +&prompt.root; gvinum -> create config1 +Configuration summary +Drives: 1 (4 configured) +Volumes: 1 (4 configured) +Plexes: 1 (8 configured) +Subdisks: 1 (16 configured) - D a State: up Device /dev/da3h Avail: 2061/2573 MB (80%) + D a State: up Device /dev/da3h Avail: 2061/2573 MB (80%) - V myvol State: up Plexes: 1 Size: 512 MB + V myvol State: up Plexes: 1 Size: 512 MB - P myvol.p0 C State: up Subdisks: 1 Size: 512 MB + P myvol.p0 C State: up Subdisks: 1 Size: 512 MB - S myvol.p0.s0 State: up PO: 0 B Size: 512 MB + S myvol.p0.s0 State: up PO: 0 B Size: 512 MB This output shows the brief listing format of &man.gvinum.8;. It is represented graphically in .
A Simple <filename>vinum</filename> Volume
This figure, and the ones which follow, represent a volume, which contains the plexes, which in turn contains the subdisks. In this example, the volume contains one plex, and the plex contains one subdisk. This particular volume has no specific advantage over a conventional disk partition. It contains a single plex, so it is not redundant. The plex contains a single subdisk, so there is no difference in storage allocation from a conventional disk partition. The following sections illustrate various more interesting configuration methods. Increased Resilience: Mirroring The resilience of a volume can be increased by mirroring. When laying out a mirrored volume, it is important to ensure that the subdisks of each plex are on different drives, so that a drive failure will not take down both plexes. The following configuration mirrors a volume: drive b device /dev/da4h volume mirror plex org concat sd length 512m drive a plex org concat sd length 512m drive b In this example, it was not necessary to specify a definition of drive a again, since vinum keeps track of all objects in its configuration database. After processing this definition, the configuration looks like: Drives: 2 (4 configured) Volumes: 2 (4 configured) Plexes: 3 (8 configured) Subdisks: 3 (16 configured) D a State: up Device /dev/da3h Avail: 1549/2573 MB (60%) D b State: up Device /dev/da4h Avail: 2061/2573 MB (80%) V myvol State: up Plexes: 1 Size: 512 MB V mirror State: up Plexes: 2 Size: 512 MB P myvol.p0 C State: up Subdisks: 1 Size: 512 MB P mirror.p0 C State: up Subdisks: 1 Size: 512 MB P mirror.p1 C State: initializing Subdisks: 1 Size: 512 MB S myvol.p0.s0 State: up PO: 0 B Size: 512 MB S mirror.p0.s0 State: up PO: 0 B Size: 512 MB S mirror.p1.s0 State: empty PO: 0 B Size: 512 MB shows the structure graphically.
A Mirrored <filename>vinum</filename> Volume
In this example, each plex contains the full 512 MB of address space. As in the previous example, each plex contains only a single subdisk. Optimizing Performance The mirrored volume in the previous example is more resistant to failure than an unmirrored volume, but its performance is less as each write to the volume requires a write to both drives, using up a greater proportion of the total disk bandwidth. Performance considerations demand a different approach: instead of mirroring, the data is striped across as many disk drives as possible. The following configuration shows a volume with a plex striped across four disk drives: drive c device /dev/da5h drive d device /dev/da6h volume stripe plex org striped 512k sd length 128m drive a sd length 128m drive b sd length 128m drive c sd length 128m drive d As before, it is not necessary to define the drives which are already known to vinum. After processing this definition, the configuration looks like: Drives: 4 (4 configured) Volumes: 3 (4 configured) Plexes: 4 (8 configured) Subdisks: 7 (16 configured) D a State: up Device /dev/da3h Avail: 1421/2573 MB (55%) D b State: up Device /dev/da4h Avail: 1933/2573 MB (75%) D c State: up Device /dev/da5h Avail: 2445/2573 MB (95%) D d State: up Device /dev/da6h Avail: 2445/2573 MB (95%) V myvol State: up Plexes: 1 Size: 512 MB V mirror State: up Plexes: 2 Size: 512 MB V striped State: up Plexes: 1 Size: 512 MB P myvol.p0 C State: up Subdisks: 1 Size: 512 MB P mirror.p0 C State: up Subdisks: 1 Size: 512 MB P mirror.p1 C State: initializing Subdisks: 1 Size: 512 MB P striped.p1 State: up Subdisks: 1 Size: 512 MB S myvol.p0.s0 State: up PO: 0 B Size: 512 MB S mirror.p0.s0 State: up PO: 0 B Size: 512 MB S mirror.p1.s0 State: empty PO: 0 B Size: 512 MB S striped.p0.s0 State: up PO: 0 B Size: 128 MB S striped.p0.s1 State: up PO: 512 kB Size: 128 MB S striped.p0.s2 State: up PO: 1024 kB Size: 128 MB S striped.p0.s3 State: up PO: 1536 kB Size: 128 MB
A Striped <filename>vinum</filename> Volume
This volume is represented in . The darkness of the stripes indicates the position within the plex address space, where the lightest stripes come first and the darkest last. Resilience and Performance With sufficient hardware, it is possible to build volumes which show both increased resilience and increased performance compared to standard &unix; partitions. A typical configuration file might be: volume raid10 plex org striped 512k sd length 102480k drive a sd length 102480k drive b sd length 102480k drive c sd length 102480k drive d sd length 102480k drive e plex org striped 512k sd length 102480k drive c sd length 102480k drive d sd length 102480k drive e sd length 102480k drive a sd length 102480k drive b The subdisks of the second plex are offset by two drives from those of the first plex. This helps to ensure that writes do not go to the same subdisks even if a transfer goes over two drives. represents the structure of this volume.
A Mirrored, Striped <filename>vinum</filename> Volume
Object Naming vinum assigns default names to plexes and subdisks, although they may be overridden. Overriding the default names is not recommended as it does not bring a significant advantage and it can cause confusion. Names may contain any non-blank character, but it is recommended to restrict them to letters, digits and the underscore characters. The names of volumes, plexes, and subdisks may be up to 64 characters long, and the names of drives may be up to 32 characters long. vinum objects are assigned device nodes in the hierarchy /dev/gvinum. The configuration shown above would cause vinum to create the following device nodes: Device entries for each volume. These are the main devices used by vinum. The configuration above would include the devices /dev/gvinum/myvol, /dev/gvinum/mirror, /dev/gvinum/striped, /dev/gvinum/raid5 and /dev/gvinum/raid10. All volumes get direct entries under /dev/gvinum/. The directories /dev/gvinum/plex, and /dev/gvinum/sd, which contain device nodes for each plex and for each subdisk, respectively. For example, consider the following configuration file: drive drive1 device /dev/sd1h drive drive2 device /dev/sd2h drive drive3 device /dev/sd3h drive drive4 device /dev/sd4h volume s64 setupstate plex org striped 64k sd length 100m drive drive1 sd length 100m drive drive2 sd length 100m drive drive3 sd length 100m drive drive4 After processing this file, &man.gvinum.8; creates the following structure in /dev/gvinum: drwxr-xr-x 2 root wheel 512 Apr 13 16:46 plex crwxr-xr-- 1 root wheel 91, 2 Apr 13 16:46 s64 drwxr-xr-x 2 root wheel 512 Apr 13 16:46 sd /dev/vinum/plex: total 0 crwxr-xr-- 1 root wheel 25, 0x10000002 Apr 13 16:46 s64.p0 /dev/vinum/sd: total 0 crwxr-xr-- 1 root wheel 91, 0x20000002 Apr 13 16:46 s64.p0.s0 crwxr-xr-- 1 root wheel 91, 0x20100002 Apr 13 16:46 s64.p0.s1 crwxr-xr-- 1 root wheel 91, 0x20200002 Apr 13 16:46 s64.p0.s2 crwxr-xr-- 1 root wheel 91, 0x20300002 Apr 13 16:46 s64.p0.s3 Although it is recommended that plexes and subdisks should not be allocated specific names, vinum drives must be named. This makes it possible to move a drive to a different location and still recognize it automatically. Drive names may be up to 32 characters long. Creating File Systems Volumes appear to the system to be identical to disks, with one exception. Unlike &unix; drives, vinum does not partition volumes, which thus do not contain a partition table. This has required modification to some disk utilities, notably &man.newfs.8;, so that it does not try to interpret the last letter of a vinum volume name as a partition identifier. For example, a disk drive may have a name like /dev/ad0a or /dev/da2h. These names represent the first partition (a) on the first (0) IDE disk (ad) and the eighth partition (h) on the third (2) SCSI disk (da) respectively. By contrast, a vinum volume might be called /dev/gvinum/concat, which has no relationship with a partition name. In order to create a file system on this volume, use &man.newfs.8;: &prompt.root; newfs /dev/gvinum/concat Configuring <filename>vinum</filename> The GENERIC kernel does not contain vinum. It is possible to build a custom kernel which includes vinum, but this is not recommended. The standard way to start vinum is as a kernel module. &man.kldload.8; is not needed because when &man.gvinum.8; starts, it checks whether the module has been loaded, and if it is not, it loads it automatically. Startup vinum stores configuration information on the disk slices in essentially the same form as in the configuration files. When reading from the configuration database, vinum recognizes a number of keywords which are not allowed in the configuration files. For example, a disk configuration might contain the following text: volume myvol state up volume bigraid state down plex name myvol.p0 state up org concat vol myvol plex name myvol.p1 state up org concat vol myvol plex name myvol.p2 state init org striped 512b vol myvol plex name bigraid.p0 state initializing org raid5 512b vol bigraid sd name myvol.p0.s0 drive a plex myvol.p0 state up len 1048576b driveoffset 265b plexoffset 0b sd name myvol.p0.s1 drive b plex myvol.p0 state up len 1048576b driveoffset 265b plexoffset 1048576b sd name myvol.p1.s0 drive c plex myvol.p1 state up len 1048576b driveoffset 265b plexoffset 0b sd name myvol.p1.s1 drive d plex myvol.p1 state up len 1048576b driveoffset 265b plexoffset 1048576b sd name myvol.p2.s0 drive a plex myvol.p2 state init len 524288b driveoffset 1048841b plexoffset 0b sd name myvol.p2.s1 drive b plex myvol.p2 state init len 524288b driveoffset 1048841b plexoffset 524288b sd name myvol.p2.s2 drive c plex myvol.p2 state init len 524288b driveoffset 1048841b plexoffset 1048576b sd name myvol.p2.s3 drive d plex myvol.p2 state init len 524288b driveoffset 1048841b plexoffset 1572864b sd name bigraid.p0.s0 drive a plex bigraid.p0 state initializing len 4194304b driveoff set 1573129b plexoffset 0b sd name bigraid.p0.s1 drive b plex bigraid.p0 state initializing len 4194304b driveoff set 1573129b plexoffset 4194304b sd name bigraid.p0.s2 drive c plex bigraid.p0 state initializing len 4194304b driveoff set 1573129b plexoffset 8388608b sd name bigraid.p0.s3 drive d plex bigraid.p0 state initializing len 4194304b driveoff set 1573129b plexoffset 12582912b sd name bigraid.p0.s4 drive e plex bigraid.p0 state initializing len 4194304b driveoff set 1573129b plexoffset 16777216b The obvious differences here are the presence of explicit location information and naming, both of which are allowed but discouraged, and the information on the states. vinum does not store information about drives in the configuration information. It finds the drives by scanning the configured disk drives for partitions with a vinum label. This enables vinum to identify drives correctly even if they have been assigned different &unix; drive IDs. Automatic Startup Gvinum always features an automatic startup once the kernel module is loaded, via &man.loader.conf.5;. To load the Gvinum module at boot time, add geom_vinum_load="YES" to /boot/loader.conf. When vinum is started with gvinum start, vinum reads the configuration database from one of the vinum drives. Under normal circumstances, each drive contains an identical copy of the configuration database, so it does not matter which drive is read. After a crash, however, vinum must determine which drive was updated most recently and read the configuration from this drive. It then updates the configuration, if necessary, from progressively older drives. Using <filename>vinum</filename> for the Root File System For a machine that has fully-mirrored file systems using vinum, it is desirable to also mirror the root file system. Setting up such a configuration is less trivial than mirroring an arbitrary file system because: The root file system must be available very early during the boot process, so the vinum infrastructure must already be available at this time. The volume containing the root file system also contains the system bootstrap and the kernel. These must be read using the host system's native utilities, such as the BIOS, which often cannot be taught about the details of vinum. In the following sections, the term root volume is generally used to describe the vinum volume that contains the root file system. Starting up <filename>vinum</filename> Early Enough for the Root File System vinum must be available early in the system boot as &man.loader.8; must be able to load the vinum kernel module before starting the kernel. This can be accomplished by putting this line in /boot/loader.conf: geom_vinum_load="YES" Making a <filename>vinum</filename>-based Root Volume Accessible to the Bootstrap The current &os; bootstrap is only 7.5 KB of code and does not understand the internal vinum structures. This means that it cannot parse the vinum configuration data or figure out the elements of a boot volume. Thus, some workarounds are necessary to provide the bootstrap code with the illusion of a standard a partition that contains the root file system. For this to be possible, the following requirements must be met for the root volume: The root volume must not be a stripe or RAID-5. The root volume must not contain more than one concatenated subdisk per plex. Note that it is desirable and possible to use multiple plexes, each containing one replica of the root file system. The bootstrap process will only use one replica for finding the bootstrap and all boot files, until the kernel mounts the root file system. Each single subdisk within these plexes needs its own a partition illusion, for the respective device to be bootable. It is not strictly needed that each of these faked a partitions is located at the same offset within its device, compared with other devices containing plexes of the root volume. However, it is probably a good idea to create the vinum volumes that way so the resulting mirrored devices are symmetric, to avoid confusion. In order to set up these a partitions for each device containing part of the root volume, the following is required: The location, offset from the beginning of the device, and size of this device's subdisk that is part of the root volume needs to be examined, using the command: &prompt.root; gvinum l -rv root vinum offsets and sizes are measured in bytes. They must be divided by 512 in order to obtain the block numbers that are to be used by bsdlabel. Run this command for each device that participates in the root volume: &prompt.root; bsdlabel -e devname devname must be either the name of the disk, like da0 for disks without a slice table, or the name of the slice, like ad0s1. If there is already an a partition on the device from a pre-vinum root file system, it should be renamed to something else so that it remains accessible (just in case), but will no longer be used by default to bootstrap the system. A currently mounted root file system cannot be renamed, so this must be executed either when being booted from a Fixit media, or in a two-step process where, in a mirror, the disk that is not been currently booted is manipulated first. The offset of the vinum partition on this device (if any) must be added to the offset of the respective root volume subdisk on this device. The resulting value will become the offset value for the new a partition. The size value for this partition can be taken verbatim from the calculation above. The fstype should be 4.2BSD. The fsize, bsize, and cpg values should be chosen to match the actual file system, though they are fairly unimportant within this context. That way, a new a partition will be established that overlaps the vinum partition on this device. bsdlabel will only allow for this overlap if the vinum partition has properly been marked using the vinum fstype. A faked a partition now exists on each device that has one replica of the root volume. It is highly recommendable to verify the result using a command like: &prompt.root; fsck -n /dev/devnamea It should be remembered that all files containing control information must be relative to the root file system in the vinum volume which, when setting up a new vinum root volume, might not match the root file system that is currently active. So in particular, /etc/fstab and /boot/loader.conf need to be taken care of. At next reboot, the bootstrap should figure out the appropriate control information from the new vinum-based root file system, and act accordingly. At the end of the kernel initialization process, after all devices have been announced, the prominent notice that shows the success of this setup is a message like: Mounting root from ufs:/dev/gvinum/root Example of a <filename>vinum</filename>-based Root Setup After the vinum root volume has been set up, the output of gvinum l -rv root could look like: ... Subdisk root.p0.s0: Size: 125829120 bytes (120 MB) State: up Plex root.p0 at offset 0 (0 B) Drive disk0 (/dev/da0h) at offset 135680 (132 kB) Subdisk root.p1.s0: Size: 125829120 bytes (120 MB) State: up Plex root.p1 at offset 0 (0 B) Drive disk1 (/dev/da1h) at offset 135680 (132 kB) The values to note are 135680 for the offset, relative to partition /dev/da0h. This translates to 265 512-byte disk blocks in bsdlabel's terms. Likewise, the size of this root volume is 245760 512-byte blocks. /dev/da1h, containing the second replica of this root volume, has a symmetric setup. The bsdlabel for these devices might look like: ... 8 partitions: # size offset fstype [fsize bsize bps/cpg] a: 245760 281 4.2BSD 2048 16384 0 # (Cyl. 0*- 15*) c: 71771688 0 unused 0 0 # (Cyl. 0 - 4467*) h: 71771672 16 vinum # (Cyl. 0*- 4467*) It can be observed that the size parameter for the faked a partition matches the value outlined above, while the offset parameter is the sum of the offset within the vinum partition h, and the offset of this partition within the device or slice. This is a typical setup that is necessary to avoid the problem described in . The entire a partition is completely within the h partition containing all the vinum data for this device. In the above example, the entire device is dedicated to vinum and there is no leftover pre-vinum root partition. Troubleshooting The following list contains a few known pitfalls and solutions. System Bootstrap Loads, but System Does Not Boot If for any reason the system does not continue to boot, the bootstrap can be interrupted by pressing space at the 10-seconds warning. The loader variable vinum.autostart can be examined by typing show and manipulated using set or unset. If the vinum kernel module was not yet in the list of modules to load automatically, type load geom_vinum. When ready, the boot process can be continued by typing boot -as which requests the kernel to ask for the root file system to mount () and make the boot process stop in single-user mode (), where the root file system is mounted read-only. That way, even if only one plex of a multi-plex volume has been mounted, no data inconsistency between plexes is being risked. At the prompt asking for a root file system to mount, any device that contains a valid root file system can be entered. If /etc/fstab is set up correctly, the default should be something like ufs:/dev/gvinum/root. A typical alternate choice would be something like ufs:da0d which could be a hypothetical partition containing the pre-vinum root file system. Care should be taken if one of the alias a partitions is entered here, that it actually references the subdisks of the vinum root device, because in a mirrored setup, this would only mount one piece of a mirrored root device. If this file system is to be mounted read-write later on, it is necessary to remove the other plex(es) of the vinum root volume since these plexes would otherwise carry inconsistent data. Only Primary Bootstrap Loads If /boot/loader fails to load, but the primary bootstrap still loads (visible by a single dash in the left column of the screen right after the boot process starts), an attempt can be made to interrupt the primary bootstrap by pressing space. This will make the bootstrap stop in stage two. An attempt can be made here to boot off an alternate partition, like the partition containing the previous root file system that has been moved away from a. Nothing Boots, the Bootstrap Panics This situation will happen if the bootstrap had been destroyed by the vinum installation. Unfortunately, vinum accidentally leaves only 4 KB at the beginning of its partition free before starting to write its vinum header information. However, the stage one and two bootstraps plus the bsdlabel require 8 KB. So if a vinum partition was started at offset 0 within a slice or disk that was meant to be bootable, the vinum setup will trash the bootstrap. Similarly, if the above situation has been recovered, by booting from a Fixit media, and the bootstrap has been re-installed using bsdlabel -B as described in , the bootstrap will trash the vinum header, and vinum will no longer find its disk(s). Though no actual vinum configuration data or data in vinum volumes will be trashed, and it would be possible to recover all the data by entering exactly the same vinum configuration data again, the situation is hard to fix. It is necessary to move the entire vinum partition by at least 4 KB, in order to have the vinum header and the system bootstrap no longer collide.
Index: head/en_US.ISO8859-1/books/porters-handbook/upgrading/chapter.xml =================================================================== --- head/en_US.ISO8859-1/books/porters-handbook/upgrading/chapter.xml (revision 50803) +++ head/en_US.ISO8859-1/books/porters-handbook/upgrading/chapter.xml (revision 50804) @@ -1,390 +1,390 @@ Upgrading a Port When a port is not the most recent version available from the authors, update the local working copy of /usr/ports. The port might have already been updated to the new version. When working with more than a few ports, it will probably be easier to use Subversion to keep the whole ports collection up-to-date, as described in the Handbook. This will have the added benefit of tracking all the port's dependencies. The next step is to see if there is an update already pending. To do this, there are two options. There is a searchable interface to the FreeBSD Problem Report (PR) or bug database. Select Ports & Packages in the Product multiple select menu, and enter the name of the port in the Summary field. However, sometimes people forget to put the name of the port into the Summary field in an unambiguous fashion. In that case, try searching in the Comment field in the Detailled Bug Information section, or try the &os; Ports Monitoring System (also known as portsmon). This system attempts to classify port PRs by portname. To search for PRs about a particular port, use the Overview of One Port. If there is no pending PR, the next step is to send an email to the port's maintainer, as shown by make maintainer. That person may already be working on an upgrade, or have a reason to not upgrade the port right now (because of, for example, stability problems of the new version), and there is no need to duplicate their work. Note that unmaintained ports are listed with a maintainer of ports@FreeBSD.org, which is just the general ports mailing list, so sending mail there probably will not help in this case. If the maintainer asks you to do the upgrade or there is no maintainer, then help out &os; by preparing the update! Please do this by using the &man.diff.1; command in the base system. To create a suitable diff for a single patch, copy the file that needs patching to something.orig, save the changes to something and then create the patch: &prompt.user; diff -u something.orig something > something.diff Otherwise, either use the svn diff method () or copy the contents of the port to an entire different directory and use the result of the recursive &man.diff.1; output of the new and old ports directories (for example, if the modified port directory is called superedit and the original is in our tree as superedit.bak, then save the result of diff -ruN superedit.bak superedit). Either unified or context diff is fine, but port committers generally prefer unified diffs. Note the use of the -N option—this is the accepted way to force diff to properly deal with the case of new files being added or old files being deleted. Before sending us the diff, please examine the output to make sure all the changes make sense. (In particular, make sure to first clean out the work directories with make clean). If some files have been added, copied, moved, or removed, add this information to the problem report so that the committer picking up the patch will know what &man.svn.1; commands to run. To simplify common operations with patch files, use make makepatch as described in . Other tools exists, like /usr/ports/Tools/scripts/patchtool.py. Before using it, please read /usr/ports/Tools/scripts/README.patchtool. If the port is unmaintained, and you are actively using it, please consider volunteering to become its maintainer. &os; has over 4000 ports without maintainers, and this is an area where more volunteers are always needed. (For a detailed description of the responsibilities of maintainers, refer to the section in the Developer's Handbook.) To submit the diff, use the bug submit form (product Ports & Packages, component Individual Port(s)). If the submitter is also maintaining the port, be sure to put [MAINTAINER] at the beginning of the Summary line. Always include the category with the port name, followed by colon, and brief descripton of the issue. For example: category/portname: add FOO option, or if maintaining the port, [MAINTAINER] category/portname: Update to X.Y. Please mention any added or deleted files in the message, as they have to be explicitly specified to &man.svn.1; when doing a commit. Do not compress or encode the diff. Before submitting the bug, review the Writing the problem report section in the Problem Reports article. It contains far more information about how to write useful problem reports. If the upgrade is motivated by security concerns or a serious fault in the currently committed port, please notify the &a.portmgr; to request immediate rebuilding and redistribution of the port's package. Unsuspecting users of pkg will otherwise continue to install the old version via pkg install for several weeks. Please use &man.diff.1; or svn diff to create updates to existing ports. Other formats include the whole file and make it impossible to see just what has changed. When diffs are not included, the entire update might be ignored. Now that all of that is done, read about how to keep up-to-date in . Using <application>Subversion</application> to Make Patches When possible, please submit a &man.svn.1; diff. They are easier to handle than diffs between new and old directories. It is easier to see what has changed, and to update the diff if something was modified in the Ports Collection since the work on it began, or if the committer asks for something to be fixed. Also, a patch generated with svn diff can be easily applied with svn patch and will save some time to the committer. &prompt.user; cd ~/my_wrkdir &prompt.user; svn co https://svn.FreeBSD.org/ports/head/dns/pdnsd &prompt.user; cd ~/my_wrkdir/pdnsd This can be anywhere, of course. Building ports is not limited to within /usr/ports/. svn.FreeBSD.org is the &os; public Subversion server. See Subversion mirror sites for more information. While in the port directory, make any changes that are needed. If adding, copying, moving, or removing a file, use svn to track these changes: &prompt.user; svn add new_file &prompt.user; svn copy some_file file_copy &prompt.user; svn move old_name new_name &prompt.user; svn remove deleted_file Make sure to check the port using the checklist in and . &prompt.user; svn status &prompt.user; svn update This will attempt to merge the differences between the patch and current repository version. Watch the output carefully. The letter in front of each file name indicates what was done with it. See for a complete list. <application>Subversion</application> Update File Prefixes U The file was updated without problems. G The file was updated without problems (only when working against a remote repository). M The file had been modified, and was merged without conflicts. C The file had been modified, and was merged with conflicts.
If C is displayed as a result of svn update, it means something changed in the Subversion repository and &man.svn.1; was not able to merge the local changes with those from the repository. It is always a good idea to inspect the changes anyway, since &man.svn.1; does not know anything about the structure of a port, so it might (and probably will) merge things that do not make sense. The last step is to make a unified &man.diff.1; of the changes: &prompt.user; svn diff > ../`make -VPKGNAME`.diff If files have been added, copied, moved, or removed, include the &man.svn.1; add, copy, move, and remove commands that were used. svn move or svn copy must be run before the patch can be applied. svn add or svn remove must be run after the patch is applied. Send the patch following the problem report submission guidelines. The patch can be automatically generated and the PR pre-filled with the contact information by using port submit. See for more details. <filename>UPDATING</filename> and <filename>MOVED</filename> <filename>/usr/ports/UPDATING</filename> If upgrading the port requires special steps like changing configuration files or running a specific program, it must be documented in this file. The format of an entry in this file is: YYYYMMDD: AFFECTS: users of portcategory/portname AUTHOR: Your name <Your email address> Special instructions When including exact portmaster, portupgrade, and/or pkg instructions, please make sure to get the shell escaping right. For example, do not use: &prompt.root; pkg delete -g -f docbook-xml* docbook-sk* docbook[2345]??-* docbook-4* As shown, the command will only work with bourne shells. Instead, use the form shown below, which will work with both bourne shell and c-shell: &prompt.root; pkg delete -g -f docbook-xml\* docbook-sk\* docbook\[2345\]\?\?-\* docbook-4\* It is recommended that the AFFECTS line contains a glob matching all the ports affected by the entry so that automated tools can parse it as easily as possible. If an update concerns all the existing BIND 9 versions the AFFECTS content must be users of dns/bind9*, it must not be users of BIND 9 <filename>/usr/ports/MOVED</filename> This file is used to list moved or removed ports. Each line in the file is made up of the name of the port, where the port was moved, when, and why. If the port was removed, the section detailing where it was moved can be left blank. Each section must be separated by the | (pipe) character, like so: old name|new name (blank for deleted)|date of move|reason The date must be entered in the form YYYY-MM-DD. New entries are added to the end of the list to keep it in chronological order, with the oldest entry at the top of the list. If a port was removed but has since been restored, delete the line in this file that states that it was removed. If a port was renamed and then renamed back to its original name, add a new one with the intermediate name to the old name, and remove the old entry as to not create a loop. Any changes must be validated with Tools/scripts/MOVEDlint.awk. If using a ports directory other than /usr/ports, use: &prompt.user; cd /home/user/ports - &prompt.user; env PORTSDIR=$PWD Tools/scripts/MOVEDlint.awk +&prompt.user; env PORTSDIR=$PWD Tools/scripts/MOVEDlint.awk