xref: /dragonfly/share/man/man7/committer.7 (revision 1465342b)
1.\" Copyright (c) 2003,2004 The DragonFly Project.  All rights reserved.
2.\"
3.\" This code is derived from software contributed to The DragonFly Project
4.\" by Matthew Dillon <dillon@backplane.com>
5.\"
6.\" Redistribution and use in source and binary forms, with or without
7.\" modification, are permitted provided that the following conditions
8.\" are met:
9.\"
10.\" 1. Redistributions of source code must retain the above copyright
11.\"    notice, this list of conditions and the following disclaimer.
12.\" 2. Redistributions in binary form must reproduce the above copyright
13.\"    notice, this list of conditions and the following disclaimer in
14.\"    the documentation and/or other materials provided with the
15.\"    distribution.
16.\" 3. Neither the name of The DragonFly Project nor the names of its
17.\"    contributors may be used to endorse or promote products derived
18.\"    from this software without specific, prior written permission.
19.\"
20.\" THIS SOFTWARE IS PROVIDED BY THE COPYRIGHT HOLDERS AND CONTRIBUTORS
21.\" ``AS IS'' AND ANY EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT
22.\" LIMITED TO, THE IMPLIED WARRANTIES OF MERCHANTABILITY AND FITNESS
23.\" FOR A PARTICULAR PURPOSE ARE DISCLAIMED.  IN NO EVENT SHALL THE
24.\" COPYRIGHT HOLDERS OR CONTRIBUTORS BE LIABLE FOR ANY DIRECT, INDIRECT,
25.\" INCIDENTAL, SPECIAL, EXEMPLARY OR CONSEQUENTIAL DAMAGES (INCLUDING,
26.\" BUT NOT LIMITED TO, PROCUREMENT OF SUBSTITUTE GOODS OR SERVICES;
27.\" LOSS OF USE, DATA, OR PROFITS; OR BUSINESS INTERRUPTION) HOWEVER CAUSED
28.\" AND ON ANY THEORY OF LIABILITY, WHETHER IN CONTRACT, STRICT LIABILITY,
29.\" OR TORT (INCLUDING NEGLIGENCE OR OTHERWISE) ARISING IN ANY WAY OUT
30.\" OF THE USE OF THIS SOFTWARE, EVEN IF ADVISED OF THE POSSIBILITY OF
31.\" SUCH DAMAGE.
32.\"
33.\" $DragonFly: src/share/man/man7/committer.7,v 1.9 2007/12/23 15:31:29 swildner Exp $
34.\"
35.Dd January 1, 2007
36.Dt COMMITTER 7
37.Os
38.Sh NAME
39.Nm committer
40.Nd rules for DragonFly committers
41.Sh CVS REPOSITORY ON YOUR LOCAL MACHINE
42Use
43.Xr cvsup 1
44to mirror the
45.Dx
46CVS repository itself onto your local box
47(if you haven't already).  See the file
48.Pa /usr/share/examples/cvsup/DragonFly-cvs-supfile .
49.Pp
50You basically want to do all CVS operations not related to commits
51via your local repository, and do commit operations directly to
52the master site.
53.Pp
54We strongly recommend that you set up a
55.Xr cron 8
56job to automatically
57cvsup at least once a day, so your local CVS repository is always
58in sync.
59.Sh CHECKING OUT THE SOURCES
60Your
61.Pa ~/.cvsrc
62should contain:
63.Bd -literal -offset indent
64cvs -q
65diff -u
66update -Pd
67checkout -P
68.Ed
69.Pp
70Checking out and updating a checked out source tree:
71.Bd -literal -offset indent
72cd /usr
73cvs -d /home/dcvs checkout src
74cvs -d /home/dcvs update src
75.Ed
76.Pp
77We do
78.Em NOT
79recommend that you use a cron job to
80.Nm cvs Cm update
81your
82checked out source.  Automatic updates can interfere with
83any work-in-progress that you have.
84.Sh SSH DSA KEYS:
85The CVS repository machine is
86.Pa crater.dragonflybsd.org ,
87and the
88.Dx
89developer machine is
90.Pa leaf.dragonflybsd.org .
91We create
92an account for you on both machines and install your public SSH
93key to give you access.
94.Pp
95Your
96.Pa crater
97account is set up for CVS repository only. It can
98only operate as a CVS slave and cannot be logged into.  That is,
99.Pa crater.dragonflybsd.org
100is only used as part of
101.Nm cvs Fl d Ar ...
102operations.
103.Pp
104Your
105.Pa leaf
106account is a general developer account.  Any
107.Dx
108developer can have a
109.Pa leaf
110account, whether a committer or not.
111It can be useful as a developer rendezvous,
112however.  For example, people upload kernel cores to
113.Pa leaf
114so other
115developers can look at them.  You log into your
116.Pa leaf
117account with:
118.Pp
119.Li ssh you@leaf.dragonflybsd.org
120.Pp
121The rules for account use are in
122.Pa leaf Ap s
123MOTD.
124It is very important that you never install a password or create a SSH
125key pair on
126.Pa leaf
127to use to access other machines.
128Because non-committers can have
129.Pa leaf
130accounts, since
131.Pa leaf
132is not considered
133a secure machine.
134.Sh TESTING COMMIT ACCESS
135There is a directory called
136.Pa /usr/src/test/test .
137To test your commit
138access, try making a modification and committing a file in this
139directory.
140.Pp
141.Bd -literal -offset indent
142cd /usr/src/test/test
143(edit something)
144cvs -d you@crater.dragonflybsd.org:/cvs commit file_you_edited
145.Ed
146.Sh COMMITTING REAL WORK
147Make modifications as needed.  For example, edit files.  If adding
148new files make CVS aware of them like this.  Files can just be
149added locally, but directories have to run through
150.Pa crater .
151These
152operations do not actually affect the repository (except directories
153being added are mkdir'd in the repository).  Instead they are
154stored in the checkout source's
155.Pa CVS/
156subdirectory and then
157synchronized to the repository when you
158.Nm cvs Cm commit .
159.Pp
160.Bd -literal -offset indent
161cvs add filename
162cvs -d you@crater.dragonflybsd.org:/cvs add directory
163.Ed
164.Pp
165To commit to the repository, use:
166.Bd -literal -offset indent
167cvs -d you@crater.dragonflybsd.org:/cvs commit files_or_directories
168.Ed
169.Pp
170Do not set
171.Ev CVSROOT
172to
173.Pa you@crater.dragonflybsd.org:/cvs .
174The reason is that you want the default
175.Ev CVSROOT
176in your checked out
177sources to point at your local CVS repository, not at
178.Pa crater .
179This reduces instances where accidental commits or repository
180operations are made on
181.Pa crater .
182.Pp
183It is best to avoid
184.Nm cvs Cm update Ap ing
185directly from the repository.
186e.g. try to avoid doing
187.Nm cvs Fl d Ar ... Cm update
188directly from
189.Pa crater .
190Instead, use
191.Xr cvsup 1
192to resync your local copy of the repository
193and use
194.Nm cvs Cm update
195or
196.Nm cvs Fl d Ar /home/dcvs Cm update
197to update from
198your local copy of the repository.
199.Pp
200The idea here is to try to avoid having CVS set its
201.Pa CVS/Root
202file in any given checked out CVS directory to point at
203.Pa crater .
204You really want it to just point at your local copy of the CVS
205repository.
206.Pp
207Never do
208.Nm cvs Cm tag
209or
210.Cm rtag
211operations.  Such operations can be
212very dangerous and only highly experienced CVS admins should
213do them.  That's basically just two or three people (Matt and Joerg
214primarily).
215.Pp
216The best way to resynchronize your local CVS repository after
217making a commit is to cvsup again.
218.Sh DISCUSSING COMMITTABLE WORK BEFORE HAND
219Discussion prior to commit usually occurs on the kernel@, submit@, or bugs@
220mailing lists.  It depends on the work involved.  Simple and obvious work,
221such as documentation edits or additions, don't really need a head's up.
222.Pp
223Simple and obvious bug fixes don't need a head's up, other than to
224say that you will (or just have) committed the fix, so you don't
225race other committers trying to do the same thing.  Usually the
226developer most active in a discussion about a bug commits the
227fix, but it isn't considered a big deal.
228.Pp
229More complex issues are usually discussed on the lists first.
230Non-trivial but straight forward bug fixes usually go through
231a testing period, where you say something like:
232.Do
233Here is a patch
234to driver BLAH that fixes A, B, and C, please test it.  If there
235are no objections I will commit it next Tuesday.
236.Dc
237(usually a week,
238or more depending on the complexity of the patch).
239.Pp
240New drivers or utilities are usually discussed.  Committers will
241often commit new work
242.Em without
243hooking it into the buildworld or
244buildkernel infrastructure in order to be able to continue
245development on it in piecemeal without having to worry about it
246breaking buildworld or buildkernel, and then they hook it in as a
247last step after they've stabilized it.  Examples of this include
248new versions of GCC, updates to vendor packages such as bind,
249sendmail, etc.
250.Sh DEVELOPER LOCKS
251Areas within the CVS repository are never explicitly locked.
252Often situations will arise where one developer commits work and
253another developer finds an issue with it that needs to be corrected.
254.Pp
255All committed work becomes community property.  No developer has a
256.Sq lock
257on any part of the source tree.  However, if a developer is
258actively working on a portion of the source tree and you find a bug
259or other issue, courtesy dictates that you post to kernel@ and/or
260email the developer.
261.Pp
262This means that, generally, if you do not see a commit to an area
263of the source tree in the last few weeks, it isn't considered active and
264you don't really need to confer with the developer that made the
265commit, though you should still post to the kernel@ mailing list
266and, of course, confer with developers when their expertise is
267needed.
268.Pp
269One exception to this rule is documentation.  If any developer commits
270new work, the documentation guys have free reign to go in and
271correct
272.Xr mdoc 7
273errors.  This is really a convenience as most developers
274are not
275.Xr mdoc 7
276gurus and it's a waste of time for the doc guys to post
277to kernel@ for all the little corrections they make.
278.Sh CONFLICTS
279On the occasion that a major code conflict occurs, for example if two
280people are doing major work in the same area of the source tree and forgot
281to collaborate with each other, the project leader will be responsible for
282resolving the conflict.  Again, the repository is considered community
283property and it must be acceptable for any developer to be able to work on
284any area of the tree that he or she has an interest in.
285.Sh MAJOR ARCHITECTURAL CHANGES
286This is generally Matt Dillon's area of expertise.  All major architectural
287changes must be discussed on the kernel@ mailing list and he retains
288veto power.
289.Pp
290This isn't usually an issue with any work.  At best if something
291doesn't look right architecturally he'll chip in with adjustments to
292make it fit in.  Nothing ever really gets vetoed.
293.Sh SEE ALSO
294.Xr cvs 1 ,
295.Xr cvsup 1 ,
296.Xr development 7
297