hgbook
view es/mq-collab.tex @ 485:039ed6f5935b
translated a section
| author | Javier Rojas <jerojasro@devnull.li> |
|---|---|
| date | Mon Jan 05 23:55:35 2009 -0500 (2009-01-05) |
| parents | 6cf30b3ed48f |
| children | b1ae672fd92b |
line source
1 \chapter{Usos avanzados de las Colas de Mercurial}
2 \label{chap:mq-collab}
4 Auunque es fácil aprender los usos más directos de las Colas de
5 Mercurial, tener algo de disciplina junto con algunas de las
6 capacidadees menos usadas de MQ hace posible trabajar en entornos de
7 desarrollo complejos.
9 En este capítulo, usaré como ejemplo una técnica que he usado para
10 administrar el desarrollo de un controlador de dispositivo Infiniband
11 para el kernel de Linux. El controlador en cuestión es grande
12 (al menos en lo que se refiere a controladores), con 25,000 líneas de
13 código esparcidas en 35 ficheros fuente. Es mantenido por un equipo
14 pequeño de desarrolladores.
16 Aunque mucho del material en este capítulo es específico de Linux, los
17 mismos principios aplican a cualquier base de código de la que usted
18 no sea el propietario principal, y sobre la que usted necesita hacer
19 un montón de desarrollo.
21 \section{El problema de múltiples objetivos}
23 El kernel de Linux cambia con rapidez, y nunca ha sido estable
24 internamente; los desarrolladores hacen cambios drásticos entre
25 %TODO no encontré una traducción adecuada para "release". Por eso el
26 %cambio
27 versiones frecuentemente. Esto significa que una versión del
28 controlador que funciona bien con una versión particular del kernel ni
29 siquiera \emph{compilará} correctamente contra, típicamente, cualquier
30 otra versión.
32 Para mantener un controlador, debemos tener en cuenta una buena
33 cantidad de versiones de Linux en mente.
34 \begin{itemize}
35 \item Un objetivo es el árbol de desarrollo principal del kernel de
36 Linux. En este caso el mantenimiento del código es compartido
37 parcialmente por otros desarrolladores en la comunidad del kernel,
38 %TODO drive-by.
39 quienes hacen modificaciones ``de-afán'' al controlador a medida que
40 desarrollan y refinan subsistemas en el kernel.
41 %TODO backport
42 \item También mantenemos algunos ``backports'' para versiones antiguas
43 del kernel de Linux, para dar soporte a las necesidades de los
44 clientes que están corriendo versiones antiguas de Linux que no
45 incorporan nuestros controladores. (Hacer el \emph{backport} de un
46 pedazo de código es modificarlo para que trabaje en una versión
47 de su entorno objetivo anterior a aquella para la cual fue escrito.)
48 \item Finalmente, nosotros liberamos nuestro software de acuerdo a un
49 cronograma que no necesariamente está alineado con el que usan los
50 distribuidores de Linux y los desarrolladores del kernel, así que
51 podemos entregar nuevas características a los clientes sin forzarlos
52 a actualizar kernels completos o distribuciones.
53 \end{itemize}
55 \subsection{Aproximaciones tentadoras que no funcionan adecuadamente}
57 Hay dos maneras estándar de mantener una porción de software que debe
58 funcionar en muchos entornos diferentes.
60 La primera es mantener varias ramas, cada una pensada para un único
61 entorno. El problema de esta aproximación es que usted debe tener una
62 disciplina férrea con el flujo de cambios entre repositorios. Una
63 nueva característica o un arreglo de fallo deben empezar su vida en un
64 repositorio ``prístino'', y luego propagarse a cada repositorio de
65 backport. Los cambios para backports están más limitados respecto a
66 las ramas a las que deberían propagarse; un cambio para backport que
67 es aplicado a una rama en la que no corresponde probablemente hará que
68 el controlador no compile.
70 La segunda es mantener un único árbol de código fuente lleno de
71 declaraciones que activen o desactiven secciones de código dependiendo
72 del entorno objetivo. Ya que estos ``ifdefs'' no están permitidos en
73 el árbol del kernel de Linux, debe seguirse algún proceso manual o
74 automático para eliminarlos y producir un árbol limpio. Una base de
75 código mantenida de esta manera se convierte rápidamente en un nido de
76 ratas de bloques condicionales que son difíciles de entender y
77 mantener.
79 %TODO canónica?
80 Ninguno de estos enfoques es adecuado para situaciones en las que
81 usted no es ``dueño'' de la copia canónica de un árbol de fuentes. En
82 el caso de un controlador de Linux que es distribuido con el kernel
83 estándar, el árbol de Linux contiene la copia del código que será
84 considerada por el mundo como la canónica. La versión oficial de
85 ``mi'' controlador puede ser modificada por gente que no conozco, sin
86 que yo siquiera me entere de ello hasta después de que los cambios
87 aparecen en el árbol de Linus.
89 Estos enfoques tienen la debilidad adicional de dificultar la
90 %TODO upstream. no no es río arriba
91 generación de parches bien formados para enviarlos a la versión
92 oficial.
94 En principio, las Colas de Mercurial parecen ser un buen candidato
95 para administrar un escenario de desarrollo como el de arriba. Aunque
96 este es de hecho el caso, MQ tiene unas cuantas características
97 adicionales que hacen el trabajo más agradable.
99 \section{Conditionally applying patches with
100 guards}
102 Perhaps the best way to maintain sanity with so many targets is to be
103 able to choose specific patches to apply for a given situation. MQ
104 provides a feature called ``guards'' (which originates with quilt's
105 \texttt{guards} command) that does just this. To start off, let's
106 create a simple repository for experimenting in.
107 \interaction{mq.guards.init}
108 This gives us a tiny repository that contains two patches that don't
109 have any dependencies on each other, because they touch different files.
111 The idea behind conditional application is that you can ``tag'' a
112 patch with a \emph{guard}, which is simply a text string of your
113 choosing, then tell MQ to select specific guards to use when applying
114 patches. MQ will then either apply, or skip over, a guarded patch,
115 depending on the guards that you have selected.
117 A patch can have an arbitrary number of guards;
118 each one is \emph{positive} (``apply this patch if this guard is
119 selected'') or \emph{negative} (``skip this patch if this guard is
120 selected''). A patch with no guards is always applied.
122 \section{Controlling the guards on a patch}
124 The \hgxcmd{mq}{qguard} command lets you determine which guards should
125 apply to a patch, or display the guards that are already in effect.
126 Without any arguments, it displays the guards on the current topmost
127 patch.
128 \interaction{mq.guards.qguard}
129 To set a positive guard on a patch, prefix the name of the guard with
130 a ``\texttt{+}''.
131 \interaction{mq.guards.qguard.pos}
132 To set a negative guard on a patch, prefix the name of the guard with
133 a ``\texttt{-}''.
134 \interaction{mq.guards.qguard.neg}
136 \begin{note}
137 The \hgxcmd{mq}{qguard} command \emph{sets} the guards on a patch; it
138 doesn't \emph{modify} them. What this means is that if you run
139 \hgcmdargs{qguard}{+a +b} on a patch, then \hgcmdargs{qguard}{+c} on
140 the same patch, the \emph{only} guard that will be set on it
141 afterwards is \texttt{+c}.
142 \end{note}
144 Mercurial stores guards in the \sfilename{series} file; the form in
145 which they are stored is easy both to understand and to edit by hand.
146 (In other words, you don't have to use the \hgxcmd{mq}{qguard} command if
147 you don't want to; it's okay to simply edit the \sfilename{series}
148 file.)
149 \interaction{mq.guards.series}
151 \section{Selecting the guards to use}
153 The \hgxcmd{mq}{qselect} command determines which guards are active at a
154 given time. The effect of this is to determine which patches MQ will
155 apply the next time you run \hgxcmd{mq}{qpush}. It has no other effect; in
156 particular, it doesn't do anything to patches that are already
157 applied.
159 With no arguments, the \hgxcmd{mq}{qselect} command lists the guards
160 currently in effect, one per line of output. Each argument is treated
161 as the name of a guard to apply.
162 \interaction{mq.guards.qselect.foo}
163 In case you're interested, the currently selected guards are stored in
164 the \sfilename{guards} file.
165 \interaction{mq.guards.qselect.cat}
166 We can see the effect the selected guards have when we run
167 \hgxcmd{mq}{qpush}.
168 \interaction{mq.guards.qselect.qpush}
170 A guard cannot start with a ``\texttt{+}'' or ``\texttt{-}''
171 character. The name of a guard must not contain white space, but most
172 other characters are acceptable. If you try to use a guard with an
173 invalid name, MQ will complain:
174 \interaction{mq.guards.qselect.error}
175 Changing the selected guards changes the patches that are applied.
176 \interaction{mq.guards.qselect.quux}
177 You can see in the example below that negative guards take precedence
178 over positive guards.
179 \interaction{mq.guards.qselect.foobar}
181 \section{MQ's rules for applying patches}
183 The rules that MQ uses when deciding whether to apply a patch
184 are as follows.
185 \begin{itemize}
186 \item A patch that has no guards is always applied.
187 \item If the patch has any negative guard that matches any currently
188 selected guard, the patch is skipped.
189 \item If the patch has any positive guard that matches any currently
190 selected guard, the patch is applied.
191 \item If the patch has positive or negative guards, but none matches
192 any currently selected guard, the patch is skipped.
193 \end{itemize}
195 \section{Trimming the work environment}
197 In working on the device driver I mentioned earlier, I don't apply the
198 patches to a normal Linux kernel tree. Instead, I use a repository
199 that contains only a snapshot of the source files and headers that are
200 relevant to Infiniband development. This repository is~1\% the size
201 of a kernel repository, so it's easier to work with.
203 I then choose a ``base'' version on top of which the patches are
204 applied. This is a snapshot of the Linux kernel tree as of a revision
205 of my choosing. When I take the snapshot, I record the changeset ID
206 from the kernel repository in the commit message. Since the snapshot
207 preserves the ``shape'' and content of the relevant parts of the
208 kernel tree, I can apply my patches on top of either my tiny
209 repository or a normal kernel tree.
211 Normally, the base tree atop which the patches apply should be a
212 snapshot of a very recent upstream tree. This best facilitates the
213 development of patches that can easily be submitted upstream with few
214 or no modifications.
216 \section{Dividing up the \sfilename{series} file}
218 I categorise the patches in the \sfilename{series} file into a number
219 of logical groups. Each section of like patches begins with a block
220 of comments that describes the purpose of the patches that follow.
222 The sequence of patch groups that I maintain follows. The ordering of
223 these groups is important; I'll describe why after I introduce the
224 groups.
225 \begin{itemize}
226 \item The ``accepted'' group. Patches that the development team has
227 submitted to the maintainer of the Infiniband subsystem, and which
228 he has accepted, but which are not present in the snapshot that the
229 tiny repository is based on. These are ``read only'' patches,
230 present only to transform the tree into a similar state as it is in
231 the upstream maintainer's repository.
232 \item The ``rework'' group. Patches that I have submitted, but that
233 the upstream maintainer has requested modifications to before he
234 will accept them.
235 \item The ``pending'' group. Patches that I have not yet submitted to
236 the upstream maintainer, but which we have finished working on.
237 These will be ``read only'' for a while. If the upstream maintainer
238 accepts them upon submission, I'll move them to the end of the
239 ``accepted'' group. If he requests that I modify any, I'll move
240 them to the beginning of the ``rework'' group.
241 \item The ``in progress'' group. Patches that are actively being
242 developed, and should not be submitted anywhere yet.
243 \item The ``backport'' group. Patches that adapt the source tree to
244 older versions of the kernel tree.
245 \item The ``do not ship'' group. Patches that for some reason should
246 never be submitted upstream. For example, one such patch might
247 change embedded driver identification strings to make it easier to
248 distinguish, in the field, between an out-of-tree version of the
249 driver and a version shipped by a distribution vendor.
250 \end{itemize}
252 Now to return to the reasons for ordering groups of patches in this
253 way. We would like the lowest patches in the stack to be as stable as
254 possible, so that we will not need to rework higher patches due to
255 changes in context. Putting patches that will never be changed first
256 in the \sfilename{series} file serves this purpose.
258 We would also like the patches that we know we'll need to modify to be
259 applied on top of a source tree that resembles the upstream tree as
260 closely as possible. This is why we keep accepted patches around for
261 a while.
263 The ``backport'' and ``do not ship'' patches float at the end of the
264 \sfilename{series} file. The backport patches must be applied on top
265 of all other patches, and the ``do not ship'' patches might as well
266 stay out of harm's way.
268 \section{Maintaining the patch series}
270 In my work, I use a number of guards to control which patches are to
271 be applied.
273 \begin{itemize}
274 \item ``Accepted'' patches are guarded with \texttt{accepted}. I
275 enable this guard most of the time. When I'm applying the patches
276 on top of a tree where the patches are already present, I can turn
277 this patch off, and the patches that follow it will apply cleanly.
278 \item Patches that are ``finished'', but not yet submitted, have no
279 guards. If I'm applying the patch stack to a copy of the upstream
280 tree, I don't need to enable any guards in order to get a reasonably
281 safe source tree.
282 \item Those patches that need reworking before being resubmitted are
283 guarded with \texttt{rework}.
284 \item For those patches that are still under development, I use
285 \texttt{devel}.
286 \item A backport patch may have several guards, one for each version
287 of the kernel to which it applies. For example, a patch that
288 backports a piece of code to~2.6.9 will have a~\texttt{2.6.9} guard.
289 \end{itemize}
290 This variety of guards gives me considerable flexibility in
291 qdetermining what kind of source tree I want to end up with. For most
292 situations, the selection of appropriate guards is automated during
293 the build process, but I can manually tune the guards to use for less
294 common circumstances.
296 \subsection{The art of writing backport patches}
298 Using MQ, writing a backport patch is a simple process. All such a
299 patch has to do is modify a piece of code that uses a kernel feature
300 not present in the older version of the kernel, so that the driver
301 continues to work correctly under that older version.
303 A useful goal when writing a good backport patch is to make your code
304 look as if it was written for the older version of the kernel you're
305 targeting. The less obtrusive the patch, the easier it will be to
306 understand and maintain. If you're writing a collection of backport
307 patches to avoid the ``rat's nest'' effect of lots of
308 \texttt{\#ifdef}s (hunks of source code that are only used
309 conditionally) in your code, don't introduce version-dependent
310 \texttt{\#ifdef}s into the patches. Instead, write several patches,
311 each of which makes unconditional changes, and control their
312 application using guards.
314 There are two reasons to divide backport patches into a distinct
315 group, away from the ``regular'' patches whose effects they modify.
316 The first is that intermingling the two makes it more difficult to use
317 a tool like the \hgext{patchbomb} extension to automate the process of
318 submitting the patches to an upstream maintainer. The second is that
319 a backport patch could perturb the context in which a subsequent
320 regular patch is applied, making it impossible to apply the regular
321 patch cleanly \emph{without} the earlier backport patch already being
322 applied.
324 \section{Useful tips for developing with MQ}
326 \subsection{Organising patches in directories}
328 If you're working on a substantial project with MQ, it's not difficult
329 to accumulate a large number of patches. For example, I have one
330 patch repository that contains over 250 patches.
332 If you can group these patches into separate logical categories, you
333 can if you like store them in different directories; MQ has no
334 problems with patch names that contain path separators.
336 \subsection{Viewing the history of a patch}
337 \label{mq-collab:tips:interdiff}
339 If you're developing a set of patches over a long time, it's a good
340 idea to maintain them in a repository, as discussed in
341 section~\ref{sec:mq:repo}. If you do so, you'll quickly discover that
342 using the \hgcmd{diff} command to look at the history of changes to a
343 patch is unworkable. This is in part because you're looking at the
344 second derivative of the real code (a diff of a diff), but also
345 because MQ adds noise to the process by modifying time stamps and
346 directory names when it updates a patch.
348 However, you can use the \hgext{extdiff} extension, which is bundled
349 with Mercurial, to turn a diff of two versions of a patch into
350 something readable. To do this, you will need a third-party package
351 called \package{patchutils}~\cite{web:patchutils}. This provides a
352 command named \command{interdiff}, which shows the differences between
353 two diffs as a diff. Used on two versions of the same diff, it
354 generates a diff that represents the diff from the first to the second
355 version.
357 You can enable the \hgext{extdiff} extension in the usual way, by
358 adding a line to the \rcsection{extensions} section of your \hgrc.
359 \begin{codesample2}
360 [extensions]
361 extdiff =
362 \end{codesample2}
363 The \command{interdiff} command expects to be passed the names of two
364 files, but the \hgext{extdiff} extension passes the program it runs a
365 pair of directories, each of which can contain an arbitrary number of
366 files. We thus need a small program that will run \command{interdiff}
367 on each pair of files in these two directories. This program is
368 available as \sfilename{hg-interdiff} in the \dirname{examples}
369 directory of the source code repository that accompanies this book.
370 \excode{hg-interdiff}
372 With the \sfilename{hg-interdiff} program in your shell's search path,
373 you can run it as follows, from inside an MQ patch directory:
374 \begin{codesample2}
375 hg extdiff -p hg-interdiff -r A:B my-change.patch
376 \end{codesample2}
377 Since you'll probably want to use this long-winded command a lot, you
378 can get \hgext{hgext} to make it available as a normal Mercurial
379 command, again by editing your \hgrc.
380 \begin{codesample2}
381 [extdiff]
382 cmd.interdiff = hg-interdiff
383 \end{codesample2}
384 This directs \hgext{hgext} to make an \texttt{interdiff} command
385 available, so you can now shorten the previous invocation of
386 \hgxcmd{extdiff}{extdiff} to something a little more wieldy.
387 \begin{codesample2}
388 hg interdiff -r A:B my-change.patch
389 \end{codesample2}
391 \begin{note}
392 The \command{interdiff} command works well only if the underlying
393 files against which versions of a patch are generated remain the
394 same. If you create a patch, modify the underlying files, and then
395 regenerate the patch, \command{interdiff} may not produce useful
396 output.
397 \end{note}
399 The \hgext{extdiff} extension is useful for more than merely improving
400 the presentation of MQ~patches. To read more about it, go to
401 section~\ref{sec:hgext:extdiff}.
403 %%% Local Variables:
404 %%% mode: latex
405 %%% TeX-master: "00book"
406 %%% End: