blob: 0c6df1b38025d6e4865ac4560c61322b6eff2ca1 [file] [log] [blame] [view]
brettw40e953e2017-02-08 17:49:281# Code Reviews
2
3Code reviews are a central part of developing high-quality code for Chromium.
4All changes must be reviewed.
5
Daniel Cheng6bffde02020-06-12 19:10:456The general patch, upload, and land process is covered in more detail in the
7[contributing code](contributing.md) page.
brettw40e953e2017-02-08 17:49:288
9# Code review policies
10
11Ideally the reviewer is someone who is familiar with the area of code you are
brettw2019b9e2017-02-09 06:40:2012touching. Any committer can review code, but an owner must provide a review
13for each directory you are touching. If you have doubts, look at the git blame
14for the file and the `OWNERS` files (see below).
brettw40e953e2017-02-08 17:49:2815
Michael Giuffridaaf367052018-03-22 20:22:3416To indicate a positive review, the reviewer provides a "Code-Review +1" in
17Gerrit, also known as an LGTM ("Looks Good To Me"). A score of "-1" indicates
18the change should not be submitted as-is.
brettw40e953e2017-02-08 17:49:2819
Michael Giuffridaaf367052018-03-22 20:22:3420If you have multiple reviewers, provide a message indicating what you expect
21from each reviewer. Otherwise people might assume their input is not required
22or waste time with redundant reviews.
brettw2019b9e2017-02-09 06:40:2023
Annie Sullivand04212e72017-10-19 21:11:3224Please also read [Respectful Changes](cl_respect.md) and
25[Respectful Code Reviews](cr_respect.md).
26
brettw2019b9e2017-02-09 06:40:2027#### Expectations for all reviewers
brettw40e953e2017-02-08 17:49:2828
29 * Aim to provide some kind of actionable response within 24 hours of receipt
Michael Giuffridaaf367052018-03-22 20:22:3430 (not counting weekends and holidays). This doesn't mean you have to do a
31 complete review, but you should be able to give some initial feedback,
32 request more time, or suggest another reviewer.
brettw40e953e2017-02-08 17:49:2833
Michael Giuffridaaf367052018-03-22 20:22:3434 * Use the status field in Gerrit settings to indicate if you're away and when
Mike Frysinger7b15bde2018-05-15 09:28:0535 you'll be back.
brettw40e953e2017-02-08 17:49:2836
37 * Don't generally discourage people from sending you code reviews. This
Michael Giuffridaaf367052018-03-22 20:22:3438 includes using a blanket "slow" in your status field.
brettw40e953e2017-02-08 17:49:2839
40## OWNERS files
41
brettw2019b9e2017-02-09 06:40:2042In various directories there are files named `OWNERS` that list the email
brettw40e953e2017-02-08 17:49:2843addresses of people qualified to review changes in that directory. You must
44get a positive review from an owner of each directory your change touches.
45
brettw2019b9e2017-02-09 06:40:2046Owners files are recursive, so each file also applies to its subdirectories.
47It's generally best to pick more specific owners. People listed in higher-level
thestig9208d8ba2017-06-09 22:05:3248directories may have less experience with the code in question. For example,
49the reviewers in the `//chrome/browser/component_name/OWNERS` file will likely
50be more familiar with code in `//chrome/browser/component_name/sub_component`
51than reviewers in the higher-level `//chrome/OWNERS` file.
52
53More detail on the owners file format is provided in the "More information"
54section below.
brettw40e953e2017-02-08 17:49:2855
brettw2019b9e2017-02-09 06:40:2056*Tip:* The `git cl owners` command can help find owners.
brettw40e953e2017-02-08 17:49:2857
58While owners must approve all patches, any committer can contribute to the
59review. In some directories the owners can be overloaded or there might be
60people not listed as owners who are more familiar with the low-level code in
61question. In these cases it's common to request a low-level review from an
62appropriate person, and then request a high-level owner review once that's
63complete. As always, be clear what you expect of each reviewer to avoid
64duplicated work.
65
brettw2019b9e2017-02-09 06:40:2066Owners do not have to pick other owners for reviews. Since they should already
67be familiar with the code in question, a thorough review from any appropriate
68committer is sufficient.
brettw40e953e2017-02-08 17:49:2869
brettw2019b9e2017-02-09 06:40:2070#### Expectations of owners
71
72The existing owners of a directory approve additions to the list. It is
Wei-Yin Chen (陳威尹)681bc322017-07-20 01:55:1173preferable to have many directories, each with a smaller number of specific
Dirk Pranke4f9740c2018-10-17 03:01:0674owners rather than large directories with many owners. Owners should:
brettw2019b9e2017-02-09 06:40:2075
76 * Demonstrate excellent judgment, teamwork and ability to uphold Chrome
77 development principles.
78
79 * Be already acting as an owner, providing high-quality reviews and design
Dirk Pranke4f9740c2018-10-17 03:01:0680 feedback.
brettw2019b9e2017-02-09 06:40:2081
Dirk Pranke4f9740c2018-10-17 03:01:0682 * Be a Chromium project member with full commit access of at least three
brettw2019b9e2017-02-09 06:40:2083 months tenure.
84
85 * Have submitted a substantial number of non-trivial changes to the affected
brettw40e953e2017-02-08 17:49:2886 directory.
87
brettw2019b9e2017-02-09 06:40:2088 * Have committed or reviewed substantial work to the affected directory
Dirk Pranke4f9740c2018-10-17 03:01:0689 within the last ninety days.
brettw40e953e2017-02-08 17:49:2890
brettw2019b9e2017-02-09 06:40:2091 * Have the bandwidth to contribute to reviews in a timely manner. If the load
92 is unsustainable, work to expand the number of owners. Don't try to
93 discourage people from sending reviews, including writing "slow" or
94 "emeritus" after your name.
95
Dirk Pranke4f9740c2018-10-17 03:01:0696The above are guidelines more than they are hard rules, and exceptions are
97okay as long as there is a consensus by the existing owners for them.
98For example, seldom-updated directories may have exceptions to the
99"substantiality" and "recency" requirements. Directories in `third_party`
100should list those most familiar with the library, regardless of how often
101the code is updated.
brettw40e953e2017-02-08 17:49:28102
brettw2019b9e2017-02-09 06:40:20103### OWNERS file details
104
105Refer to the [source code](https://chromium.googlesource.com/chromium/tools/depot_tools/+/master/owners.py)
thestig9208d8ba2017-06-09 22:05:32106for all details on the file format.
brettw2019b9e2017-02-09 06:40:20107
108This example indicates that two people are owners, in addition to any owners
109from the parent directory. `git cl owners` will list the comment after an
110owner address, so this is a good place to include restrictions or special
111instructions.
112```
113# You can include comments like this.
114a@chromium.org
115b@chromium.org # Only for the frobinator.
116```
117
118A `*` indicates that all committers are owners:
119```
120*
121```
122
brettwd040b0be2017-02-09 19:11:33123The text `set noparent` will stop owner propagation from parent directories.
Jochen Eisingerea8f92d82017-08-02 17:40:14124This should be rarely used. If you want to use `set noparent` except for IPC
125related files, please first reach out to chrome-eng-review@google.com.
126
Jochen Eisinger8f0c8d82019-10-25 18:28:27127You have to use `set noparent` together with a reference to a file that lists
128the owners for the given use case. Approved use cases are listed in
129`//build/OWNERS.setnoparent`. Owners listed in those files are expected to
130execute special governance functions such as eng review or ipc security review.
131Every set of owners should implement their own means of auditing membership. The
132minimum expectation is that membership in those files is reevaluated on
133project, or affiliation changes.
134
135In this example, only the eng reviewers are owners:
brettw2019b9e2017-02-09 06:40:20136```
137set noparent
Jochen Eisinger8f0c8d82019-10-25 18:28:27138file://ENG_REVIEW_OWNERS
brettw2019b9e2017-02-09 06:40:20139```
140
141The `per-file` directive allows owners to be added that apply only to files
Wei-Yin Chen (陳威尹)681bc322017-07-20 01:55:11142matching a pattern. In this example, owners from the parent directory
brettw2019b9e2017-02-09 06:40:20143apply, plus one person for some classes of files, and all committers are
144owners for the readme:
145```
146per-file foo_bar.cc=a@chromium.org
147per-file foo.*=a@chromium.org
148
149per-file readme.txt=*
150```
151
George Burgess IV1ef04932018-01-27 07:04:04152Note that `per-file` directives cannot directly specify subdirectories, e.g:
153```
154per-file foo/bar.cc=a@chromium.org
155```
156
157is not OK; instead, place a `per-file` directive in `foo/OWNERS`.
158
brettw2019b9e2017-02-09 06:40:20159Other `OWNERS` files can be included by reference by listing the path to the
160file with `file://...`. This example indicates that only the people listed in
161`//ipc/SECURITY_OWNERS` can review the messages files:
162```
163per-file *_messages*.h=set noparent
164per-file *_messages*.h=file://ipc/SECURITY_OWNERS
165```
Steve Kobesf885edf2018-09-11 13:41:11166
167## TBR ("To Be Reviewed")
168
169"TBR" is our mechanism for post-commit review. It should be used rarely and
170only in cases where a normal review is unnecessary, as described under
171"When to TBR", below.
172
173TBR does not mean "no review." A reviewer TBR-ed on a change should still
174review the change. If there are comments after landing, the author is obligated
175to address them in a followup patch.
176
177Do not use TBR just because a change is urgent or the reviewer is being slow.
178Contact the reviewer directly or find somebody else to review your change.
179
180### How to TBR
181
182To send a change TBR, annotate the description and send email like normal.
183Otherwise the reviewer won't know to review the patch.
184
185 * Add the reviewer's email address in the code review tool's reviewer field
186 like normal.
187
Lei Zhang3fd577db2020-05-21 21:33:19188 * Add a line "Tbr: <reviewer's email>" to the bottom of the change list
189 description. e.g. `Tbr: reviewer1@chromium.org,reviewer2@chromium.org`
Steve Kobesf885edf2018-09-11 13:41:11190
191 * Type a message so that the owners in the TBR list can understand who is
192 responsible for reviewing what, as part of their post-commit review
193 responsibility. e.g.
194 ```
195 TBRing reviewers:
196 reviewer1: Please review changes to foo/
197 reviewer2: Please review changes to bar/
198 ```
199
200### When to TBR
201
202#### Reverts and relands
203
204The most common use of TBR is to revert patches that broke the build. Clean
205reverts of recent patches may be submitted TBR. However, TBR should not be used
206if the revert required non-trivial conflict resolution, or if the patch being
207reverted is older than a few days.
208
209A developer relanding a patch can TBR the OWNERS for changes which are identical
210to the original (reverted) patch. If the reland patch contains any new changes
211(such as bug fixes) on top of the original, those changes should go through the
212normal review process.
213
214When creating a reland patch, you should first upload an up-to-date patchset
215with the exact content of the original (reverted) patch, and then upload the
216patchset to be relanded. This is important for the reviewers to understand what
217the fix for relanding was.
218
219#### Mechanical changes
220
221You can use TBR with certain mechanical changes that affect many callers in
222different directories. For example, adding a parameter to a common function in
223`//base`, with callers in `//chrome/browser/foo`, `//net/bar`, and many other
224directories. If the updates to the callers is mechanical, you can:
225
Gabriel Charette064574c2018-11-17 01:36:32226 1. Get a normal owner of the lower-level code you're changing (in this
227 example, the function in `//base`) to do a proper review of those changes.
Steve Kobesf885edf2018-09-11 13:41:11228
Gabriel Charette064574c2018-11-17 01:36:32229 2. Get _somebody_ to review the downstream changes made to the callers as a
230 result of the `//base` change. This is often the same person from the
231 previous step but could be somebody else.
Steve Kobesf885edf2018-09-11 13:41:11232
Gabriel Charette064574c2018-11-17 01:36:32233 3. TBR the owner of the lower-level code you're changing (in this example,
234 `//base`), after they've LGTM'ed the API change, to bypass owners review of
235 the API consumers incurring trivial side-effects.
Steve Kobesf885edf2018-09-11 13:41:11236
237This process ensures that all code is reviewed prior to checkin and that the
Gabriel Charette064574c2018-11-17 01:36:32238concept of the change is reviewed by a qualified person, without having to ping
239many owners with little say in the trivial side-effects they incur.
240
241**Note:** The above policy is only viable for strictly mechanical changes. For
242large-scale scripted changes you should:
243
244 1. Have an owner of the core change review the script.
245
246 2. Use `git cl split` to shard the large change into many small CLs with a
247 clear description of what each reviewer is expected to verify
248 ([example](https://chromium-review.googlesource.com/1191225)).
Steve Kobesf885edf2018-09-11 13:41:11249
250#### Documentation updates
251
252You can TBR documentation updates. Documentation means markdown files, text
253documents, and high-level comments in code. At finer levels of detail, comments
254in source files become more like code and should be reviewed normally (not
255using TBR). Non-TBR-able stuff includes things like function contracts and most
256comments inside functions.
257
258 * Use good judgement. If you're changing something very important, tricky,
259 or something you may not be very familiar with, ask for the code review
260 up-front.
261
262 * Don't TBR changes to policy documents like the style guide or this document.
263
264 * Don't mix unrelated documentation updates with code changes.
265
266 * Be sure to actually send out the email for the code review. If you get one,
267 please actually read the changes.
268