1 .. SPDX-License-Identifier: (GPL-2.0+ OR CC-BY 1 .. SPDX-License-Identifier: (GPL-2.0+ OR CC-BY-4.0)
2 .. See the bottom of this file for additional 2 .. See the bottom of this file for additional redistribution information.
3 3
4 Handling regressions 4 Handling regressions
5 ++++++++++++++++++++ 5 ++++++++++++++++++++
6 6
7 *We don't cause regressions* -- this document 7 *We don't cause regressions* -- this document describes what this "first rule of
8 Linux kernel development" means in practice fo 8 Linux kernel development" means in practice for developers. It complements
9 Documentation/admin-guide/reporting-regression 9 Documentation/admin-guide/reporting-regressions.rst, which covers the topic from a
10 user's point of view; if you never read that t 10 user's point of view; if you never read that text, go and at least skim over it
11 before continuing here. 11 before continuing here.
12 12
13 The important bits (aka "The TL;DR") 13 The important bits (aka "The TL;DR")
14 ==================================== 14 ====================================
15 15
16 #. Ensure subscribers of the `regression maili 16 #. Ensure subscribers of the `regression mailing list <https://lore.kernel.org/regressions/>`_
17 (regressions@lists.linux.dev) quickly becom 17 (regressions@lists.linux.dev) quickly become aware of any new regression
18 report: 18 report:
19 19
20 * When receiving a mailed report that did 20 * When receiving a mailed report that did not CC the list, bring it into the
21 loop by immediately sending at least a b 21 loop by immediately sending at least a brief "Reply-all" with the list
22 CCed. 22 CCed.
23 23
24 * Forward or bounce any reports submitted 24 * Forward or bounce any reports submitted in bug trackers to the list.
25 25
26 #. Make the Linux kernel regression tracking b 26 #. Make the Linux kernel regression tracking bot "regzbot" track the issue (this
27 is optional, but recommended): 27 is optional, but recommended):
28 28
29 * For mailed reports, check if the reporte 29 * For mailed reports, check if the reporter included a line like ``#regzbot
30 introduced: v5.13..v5.14-rc1``. If not, 30 introduced: v5.13..v5.14-rc1``. If not, send a reply (with the regressions
31 list in CC) containing a paragraph like 31 list in CC) containing a paragraph like the following, which tells regzbot
32 when the issue started to happen:: 32 when the issue started to happen::
33 33
34 #regzbot ^introduced: 1f2e3d4c5b6a 34 #regzbot ^introduced: 1f2e3d4c5b6a
35 35
36 * When forwarding reports from a bug track 36 * When forwarding reports from a bug tracker to the regressions list (see
37 above), include a paragraph like the fol 37 above), include a paragraph like the following::
38 38
39 #regzbot introduced: v5.13..v5.14-rc1 39 #regzbot introduced: v5.13..v5.14-rc1
40 #regzbot from: Some N. Ice Human <some.h 40 #regzbot from: Some N. Ice Human <some.human@example.com>
41 #regzbot monitor: http://some.bugtracke 41 #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789
42 42
43 #. When submitting fixes for regressions, add !! 43 #. When submitting fixes for regressions, add "Link:" tags to the patch
44 description pointing to all places where th 44 description pointing to all places where the issue was reported, as
45 mandated by Documentation/process/submittin 45 mandated by Documentation/process/submitting-patches.rst and
46 :ref:`Documentation/process/5.Posting.rst < !! 46 :ref:`Documentation/process/5.Posting.rst <development_posting>`.
47 only fixing part of the issue that caused t <<
48 "Link:" tags instead. regzbot currently mak <<
49 two. <<
50 47
51 #. Try to fix regressions quickly once the cul 48 #. Try to fix regressions quickly once the culprit has been identified; fixes
52 for most regressions should be merged withi 49 for most regressions should be merged within two weeks, but some need to be
53 resolved within two or three days. 50 resolved within two or three days.
54 51
55 52
56 All the details on Linux kernel regressions re 53 All the details on Linux kernel regressions relevant for developers
57 ============================================== 54 ===================================================================
58 55
59 56
60 The important basics in more detail 57 The important basics in more detail
61 ----------------------------------- 58 -----------------------------------
62 59
63 60
64 What to do when receiving regression reports 61 What to do when receiving regression reports
65 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 62 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
66 63
67 Ensure the Linux kernel's regression tracker a 64 Ensure the Linux kernel's regression tracker and others subscribers of the
68 `regression mailing list <https://lore.kernel. 65 `regression mailing list <https://lore.kernel.org/regressions/>`_
69 (regressions@lists.linux.dev) become aware of 66 (regressions@lists.linux.dev) become aware of any newly reported regression:
70 67
71 * When you receive a report by mail that did 68 * When you receive a report by mail that did not CC the list, immediately bring
72 it into the loop by sending at least a brie 69 it into the loop by sending at least a brief "Reply-all" with the list CCed;
73 try to ensure it gets CCed again in case yo 70 try to ensure it gets CCed again in case you reply to a reply that omitted
74 the list. 71 the list.
75 72
76 * If a report submitted in a bug tracker hits 73 * If a report submitted in a bug tracker hits your Inbox, forward or bounce it
77 to the list. Consider checking the list arc 74 to the list. Consider checking the list archives beforehand, if the reporter
78 already forwarded the report as instructed 75 already forwarded the report as instructed by
79 Documentation/admin-guide/reporting-issues. 76 Documentation/admin-guide/reporting-issues.rst.
80 77
81 When doing either, consider making the Linux k 78 When doing either, consider making the Linux kernel regression tracking bot
82 "regzbot" immediately start tracking the issue 79 "regzbot" immediately start tracking the issue:
83 80
84 * For mailed reports, check if the reporter i 81 * For mailed reports, check if the reporter included a "regzbot command" like
85 ``#regzbot introduced: 1f2e3d4c5b6a``. If n 82 ``#regzbot introduced: 1f2e3d4c5b6a``. If not, send a reply (with the
86 regressions list in CC) with a paragraph li 83 regressions list in CC) with a paragraph like the following:::
87 84
88 #regzbot ^introduced: v5.13..v5.14-rc1 85 #regzbot ^introduced: v5.13..v5.14-rc1
89 86
90 This tells regzbot the version range in whi 87 This tells regzbot the version range in which the issue started to happen;
91 you can specify a range using commit-ids as 88 you can specify a range using commit-ids as well or state a single commit-id
92 in case the reporter bisected the culprit. 89 in case the reporter bisected the culprit.
93 90
94 Note the caret (^) before the "introduced": 91 Note the caret (^) before the "introduced": it tells regzbot to treat the
95 parent mail (the one you reply to) as the i 92 parent mail (the one you reply to) as the initial report for the regression
96 you want to see tracked; that's important, 93 you want to see tracked; that's important, as regzbot will later look out
97 for patches with "Closes:" tags pointing to !! 94 for patches with "Link:" tags pointing to the report in the archives on
98 lore.kernel.org. 95 lore.kernel.org.
99 96
100 * When forwarding a regression reported to a !! 97 * When forwarding a regressions reported to a bug tracker, include a paragraph
101 with these regzbot commands:: 98 with these regzbot commands::
102 99
103 #regzbot introduced: 1f2e3d4c5b6a 100 #regzbot introduced: 1f2e3d4c5b6a
104 #regzbot from: Some N. Ice Human <some.h 101 #regzbot from: Some N. Ice Human <some.human@example.com>
105 #regzbot monitor: http://some.bugtracke 102 #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789
106 103
107 Regzbot will then automatically associate p 104 Regzbot will then automatically associate patches with the report that
108 contain "Closes:" tags pointing to your mai !! 105 contain "Link:" tags pointing to your mail or the mentioned ticket.
109 106
110 What's important when fixing regressions 107 What's important when fixing regressions
111 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 108 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
112 109
113 You don't need to do anything special when sub 110 You don't need to do anything special when submitting fixes for regression, just
114 remember to do what Documentation/process/subm 111 remember to do what Documentation/process/submitting-patches.rst,
115 :ref:`Documentation/process/5.Posting.rst <dev 112 :ref:`Documentation/process/5.Posting.rst <development_posting>`, and
116 Documentation/process/stable-kernel-rules.rst 113 Documentation/process/stable-kernel-rules.rst already explain in more detail:
117 114
118 * Point to all places where the issue was rep !! 115 * Point to all places where the issue was reported using "Link:" tags::
119 116
120 Closes: https://lore.kernel.org/r/30th. !! 117 Link: https://lore.kernel.org/r/30th.anniversary.repost@klaava.Helsinki.FI/
121 Closes: https://bugzilla.kernel.org/sho !! 118 Link: https://bugzilla.kernel.org/show_bug.cgi?id=1234567890
122 <<
123 If you are only fixing part of the issue, y <<
124 described in the first document mentioned a <<
125 both of these equivalently and considers th <<
126 119
127 * Add a "Fixes:" tag to specify the commit ca 120 * Add a "Fixes:" tag to specify the commit causing the regression.
128 121
129 * If the culprit was merged in an earlier dev 122 * If the culprit was merged in an earlier development cycle, explicitly mark
130 the fix for backporting using the ``Cc: sta 123 the fix for backporting using the ``Cc: stable@vger.kernel.org`` tag.
131 124
132 All this is expected from you and important wh 125 All this is expected from you and important when it comes to regression, as
133 these tags are of great value for everyone (yo 126 these tags are of great value for everyone (you included) that might be looking
134 into the issue weeks, months, or years later. 127 into the issue weeks, months, or years later. These tags are also crucial for
135 tools and scripts used by other kernel develop 128 tools and scripts used by other kernel developers or Linux distributions; one of
136 these tools is regzbot, which heavily relies o !! 129 these tools is regzbot, which heavily relies on the "Link:" tags to associate
137 reports for regression with changes resolving 130 reports for regression with changes resolving them.
138 131
139 Expectations and best practices for fixing reg 132 Expectations and best practices for fixing regressions
140 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 133 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
141 134
142 As a Linux kernel developer, you are expected 135 As a Linux kernel developer, you are expected to give your best to prevent
143 situations where a regression caused by a rece 136 situations where a regression caused by a recent change of yours leaves users
144 only these options: 137 only these options:
145 138
146 * Run a kernel with a regression that impacts 139 * Run a kernel with a regression that impacts usage.
147 140
148 * Switch to an older or newer kernel series. 141 * Switch to an older or newer kernel series.
149 142
150 * Continue running an outdated and thus poten 143 * Continue running an outdated and thus potentially insecure kernel for more
151 than three weeks after the regression's cul 144 than three weeks after the regression's culprit was identified. Ideally it
152 should be less than two. And it ought to be 145 should be less than two. And it ought to be just a few days, if the issue is
153 severe or affects many users -- either in g 146 severe or affects many users -- either in general or in prevalent
154 environments. 147 environments.
155 148
156 How to realize that in practice depends on var 149 How to realize that in practice depends on various factors. Use the following
157 rules of thumb as a guide. 150 rules of thumb as a guide.
158 151
159 In general: 152 In general:
160 153
161 * Prioritize work on regressions over all oth 154 * Prioritize work on regressions over all other Linux kernel work, unless the
162 latter concerns a severe issue (e.g. acute 155 latter concerns a severe issue (e.g. acute security vulnerability, data loss,
163 bricked hardware, ...). 156 bricked hardware, ...).
164 157
165 * Expedite fixing mainline regressions that r 158 * Expedite fixing mainline regressions that recently made it into a proper
166 mainline, stable, or longterm release (eith 159 mainline, stable, or longterm release (either directly or via backport).
167 160
168 * Do not consider regressions from the curren 161 * Do not consider regressions from the current cycle as something that can wait
169 till the end of the cycle, as the issue mig 162 till the end of the cycle, as the issue might discourage or prevent users and
170 CI systems from testing mainline now or gen 163 CI systems from testing mainline now or generally.
171 164
172 * Work with the required care to avoid additi 165 * Work with the required care to avoid additional or bigger damage, even if
173 resolving an issue then might take longer t 166 resolving an issue then might take longer than outlined below.
174 167
175 On timing once the culprit of a regression is 168 On timing once the culprit of a regression is known:
176 169
177 * Aim to mainline a fix within two or three d 170 * Aim to mainline a fix within two or three days, if the issue is severe or
178 bothering many users -- either in general o 171 bothering many users -- either in general or in prevalent conditions like a
179 particular hardware environment, distributi 172 particular hardware environment, distribution, or stable/longterm series.
180 173
181 * Aim to mainline a fix by Sunday after the n 174 * Aim to mainline a fix by Sunday after the next, if the culprit made it
182 into a recent mainline, stable, or longterm 175 into a recent mainline, stable, or longterm release (either directly or via
183 backport); if the culprit became known earl 176 backport); if the culprit became known early during a week and is simple to
184 resolve, try to mainline the fix within the 177 resolve, try to mainline the fix within the same week.
185 178
186 * For other regressions, aim to mainline fixe 179 * For other regressions, aim to mainline fixes before the hindmost Sunday
187 within the next three weeks. One or two Sun 180 within the next three weeks. One or two Sundays later are acceptable, if the
188 regression is something people can live wit 181 regression is something people can live with easily for a while -- like a
189 mild performance regression. 182 mild performance regression.
190 183
191 * It's strongly discouraged to delay mainlini 184 * It's strongly discouraged to delay mainlining regression fixes till the next
192 merge window, except when the fix is extrao 185 merge window, except when the fix is extraordinarily risky or when the
193 culprit was mainlined more than a year ago. 186 culprit was mainlined more than a year ago.
194 187
195 On procedure: 188 On procedure:
196 189
197 * Always consider reverting the culprit, as i 190 * Always consider reverting the culprit, as it's often the quickest and least
198 dangerous way to fix a regression. Don't wo 191 dangerous way to fix a regression. Don't worry about mainlining a fixed
199 variant later: that should be straight-forw 192 variant later: that should be straight-forward, as most of the code went
200 through review once already. 193 through review once already.
201 194
202 * Try to resolve any regressions introduced i 195 * Try to resolve any regressions introduced in mainline during the past
203 twelve months before the current developmen 196 twelve months before the current development cycle ends: Linus wants such
204 regressions to be handled like those from t 197 regressions to be handled like those from the current cycle, unless fixing
205 bears unusual risks. 198 bears unusual risks.
206 199
207 * Consider CCing Linus on discussions or patc 200 * Consider CCing Linus on discussions or patch review, if a regression seems
208 tangly. Do the same in precarious or urgent 201 tangly. Do the same in precarious or urgent cases -- especially if the
209 subsystem maintainer might be unavailable. 202 subsystem maintainer might be unavailable. Also CC the stable team, when you
210 know such a regression made it into a mainl 203 know such a regression made it into a mainline, stable, or longterm release.
211 204
212 * For urgent regressions, consider asking Lin 205 * For urgent regressions, consider asking Linus to pick up the fix straight
213 from the mailing list: he is totally fine w 206 from the mailing list: he is totally fine with that for uncontroversial
214 fixes. Ideally though such requests should 207 fixes. Ideally though such requests should happen in accordance with the
215 subsystem maintainers or come directly from 208 subsystem maintainers or come directly from them.
216 209
217 * In case you are unsure if a fix is worth th 210 * In case you are unsure if a fix is worth the risk applying just days before
218 a new mainline release, send Linus a mail w 211 a new mainline release, send Linus a mail with the usual lists and people in
219 CC; in it, summarize the situation while as 212 CC; in it, summarize the situation while asking him to consider picking up
220 the fix straight from the list. He then him 213 the fix straight from the list. He then himself can make the call and when
221 needed even postpone the release. Such requ 214 needed even postpone the release. Such requests again should ideally happen
222 in accordance with the subsystem maintainer 215 in accordance with the subsystem maintainers or come directly from them.
223 216
224 Regarding stable and longterm kernels: 217 Regarding stable and longterm kernels:
225 218
226 * You are free to leave regressions to the st 219 * You are free to leave regressions to the stable team, if they at no point in
227 time occurred with mainline or were fixed t 220 time occurred with mainline or were fixed there already.
228 221
229 * If a regression made it into a proper mainl 222 * If a regression made it into a proper mainline release during the past
230 twelve months, ensure to tag the fix with " 223 twelve months, ensure to tag the fix with "Cc: stable@vger.kernel.org", as a
231 "Fixes:" tag alone does not guarantee a bac 224 "Fixes:" tag alone does not guarantee a backport. Please add the same tag,
232 in case you know the culprit was backported 225 in case you know the culprit was backported to stable or longterm kernels.
233 226
234 * When receiving reports about regressions in 227 * When receiving reports about regressions in recent stable or longterm kernel
235 series, please evaluate at least briefly if 228 series, please evaluate at least briefly if the issue might happen in current
236 mainline as well -- and if that seems likel 229 mainline as well -- and if that seems likely, take hold of the report. If in
237 doubt, ask the reporter to check mainline. 230 doubt, ask the reporter to check mainline.
238 231
239 * Whenever you want to swiftly resolve a regr 232 * Whenever you want to swiftly resolve a regression that recently also made it
240 into a proper mainline, stable, or longterm 233 into a proper mainline, stable, or longterm release, fix it quickly in
241 mainline; when appropriate thus involve Lin 234 mainline; when appropriate thus involve Linus to fast-track the fix (see
242 above). That's because the stable team norm 235 above). That's because the stable team normally does neither revert nor fix
243 any changes that cause the same problems in 236 any changes that cause the same problems in mainline.
244 237
245 * In case of urgent regression fixes you migh 238 * In case of urgent regression fixes you might want to ensure prompt
246 backporting by dropping the stable team a n 239 backporting by dropping the stable team a note once the fix was mainlined;
247 this is especially advisable during merge w 240 this is especially advisable during merge windows and shortly thereafter, as
248 the fix otherwise might land at the end of 241 the fix otherwise might land at the end of a huge patch queue.
249 242
250 On patch flow: 243 On patch flow:
251 244
252 * Developers, when trying to reach the time p 245 * Developers, when trying to reach the time periods mentioned above, remember
253 to account for the time it takes to get fix 246 to account for the time it takes to get fixes tested, reviewed, and merged by
254 Linus, ideally with them being in linux-nex 247 Linus, ideally with them being in linux-next at least briefly. Hence, if a
255 fix is urgent, make it obvious to ensure ot 248 fix is urgent, make it obvious to ensure others handle it appropriately.
256 249
257 * Reviewers, you are kindly asked to assist d 250 * Reviewers, you are kindly asked to assist developers in reaching the time
258 periods mentioned above by reviewing regres 251 periods mentioned above by reviewing regression fixes in a timely manner.
259 252
260 * Subsystem maintainers, you likewise are enc 253 * Subsystem maintainers, you likewise are encouraged to expedite the handling
261 of regression fixes. Thus evaluate if skipp 254 of regression fixes. Thus evaluate if skipping linux-next is an option for
262 the particular fix. Also consider sending g 255 the particular fix. Also consider sending git pull requests more often than
263 usual when needed. And try to avoid holding 256 usual when needed. And try to avoid holding onto regression fixes over
264 weekends -- especially when the fix is mark 257 weekends -- especially when the fix is marked for backporting.
265 258
266 259
267 More aspects regarding regressions developers 260 More aspects regarding regressions developers should be aware of
268 ---------------------------------------------- 261 ----------------------------------------------------------------
269 262
270 263
271 How to deal with changes where a risk of regre 264 How to deal with changes where a risk of regression is known
272 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 265 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
273 266
274 Evaluate how big the risk of regressions is, f 267 Evaluate how big the risk of regressions is, for example by performing a code
275 search in Linux distributions and Git forges. 268 search in Linux distributions and Git forges. Also consider asking other
276 developers or projects likely to be affected t 269 developers or projects likely to be affected to evaluate or even test the
277 proposed change; if problems surface, maybe so 270 proposed change; if problems surface, maybe some solution acceptable for all
278 can be found. 271 can be found.
279 272
280 If the risk of regressions in the end seems to 273 If the risk of regressions in the end seems to be relatively small, go ahead
281 with the change, but let all involved parties 274 with the change, but let all involved parties know about the risk. Hence, make
282 sure your patch description makes this aspect 275 sure your patch description makes this aspect obvious. Once the change is
283 merged, tell the Linux kernel's regression tra 276 merged, tell the Linux kernel's regression tracker and the regressions mailing
284 list about the risk, so everyone has the chang 277 list about the risk, so everyone has the change on the radar in case reports
285 trickle in. Depending on the risk, you also mi 278 trickle in. Depending on the risk, you also might want to ask the subsystem
286 maintainer to mention the issue in his mainlin 279 maintainer to mention the issue in his mainline pull request.
287 280
288 What else is there to known about regressions? 281 What else is there to known about regressions?
289 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 282 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
290 283
291 Check out Documentation/admin-guide/reporting- 284 Check out Documentation/admin-guide/reporting-regressions.rst, it covers a lot
292 of other aspects you want might want to be awa 285 of other aspects you want might want to be aware of:
293 286
294 * the purpose of the "no regressions" rule !! 287 * the purpose of the "no regressions rule"
295 288
296 * what issues actually qualify as regression 289 * what issues actually qualify as regression
297 290
298 * who's in charge for finding the root cause 291 * who's in charge for finding the root cause of a regression
299 292
300 * how to handle tricky situations, e.g. when 293 * how to handle tricky situations, e.g. when a regression is caused by a
301 security fix or when fixing a regression mi 294 security fix or when fixing a regression might cause another one
302 295
303 Whom to ask for advice when it comes to regres 296 Whom to ask for advice when it comes to regressions
304 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 297 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
305 298
306 Send a mail to the regressions mailing list (r 299 Send a mail to the regressions mailing list (regressions@lists.linux.dev) while
307 CCing the Linux kernel's regression tracker (r 300 CCing the Linux kernel's regression tracker (regressions@leemhuis.info); if the
308 issue might better be dealt with in private, f 301 issue might better be dealt with in private, feel free to omit the list.
309 302
310 303
311 More about regression tracking and regzbot 304 More about regression tracking and regzbot
312 ------------------------------------------ 305 ------------------------------------------
313 306
314 307
315 Why the Linux kernel has a regression tracker, 308 Why the Linux kernel has a regression tracker, and why is regzbot used?
316 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 309 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
317 310
318 Rules like "no regressions" need someone to en 311 Rules like "no regressions" need someone to ensure they are followed, otherwise
319 they are broken either accidentally or on purp 312 they are broken either accidentally or on purpose. History has shown this to be
320 true for the Linux kernel as well. That's why 313 true for the Linux kernel as well. That's why Thorsten Leemhuis volunteered to
321 keep an eye on things as the Linux kernel's re 314 keep an eye on things as the Linux kernel's regression tracker, who's
322 occasionally helped by other people. Neither o 315 occasionally helped by other people. Neither of them are paid to do this,
323 that's why regression tracking is done on a be 316 that's why regression tracking is done on a best effort basis.
324 317
325 Earlier attempts to manually track regressions 318 Earlier attempts to manually track regressions have shown it's an exhausting and
326 frustrating work, which is why they were aband 319 frustrating work, which is why they were abandoned after a while. To prevent
327 this from happening again, Thorsten developed 320 this from happening again, Thorsten developed regzbot to facilitate the work,
328 with the long term goal to automate regression 321 with the long term goal to automate regression tracking as much as possible for
329 everyone involved. 322 everyone involved.
330 323
331 How does regression tracking work with regzbot 324 How does regression tracking work with regzbot?
332 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 325 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
333 326
334 The bot watches for replies to reports of trac 327 The bot watches for replies to reports of tracked regressions. Additionally,
335 it's looking out for posted or committed patch 328 it's looking out for posted or committed patches referencing such reports
336 with "Closes:" tags; replies to such patch pos !! 329 with "Link:" tags; replies to such patch postings are tracked as well.
337 Combined this data provides good insights into 330 Combined this data provides good insights into the current state of the fixing
338 process. 331 process.
339 332
340 Regzbot tries to do its job with as little ove 333 Regzbot tries to do its job with as little overhead as possible for both
341 reporters and developers. In fact, only report 334 reporters and developers. In fact, only reporters are burdened with an extra
342 duty: they need to tell regzbot about the regr 335 duty: they need to tell regzbot about the regression report using the ``#regzbot
343 introduced`` command outlined above; if they d 336 introduced`` command outlined above; if they don't do that, someone else can
344 take care of that using ``#regzbot ^introduced 337 take care of that using ``#regzbot ^introduced``.
345 338
346 For developers there normally is no extra work 339 For developers there normally is no extra work involved, they just need to make
347 sure to do something that was expected long be 340 sure to do something that was expected long before regzbot came to light: add
348 links to the patch description pointing to all !! 341 "Link:" tags to the patch description pointing to all reports about the issue
>> 342 fixed.
349 343
350 Do I have to use regzbot? 344 Do I have to use regzbot?
351 ~~~~~~~~~~~~~~~~~~~~~~~~~ 345 ~~~~~~~~~~~~~~~~~~~~~~~~~
352 346
353 It's in the interest of everyone if you do, as 347 It's in the interest of everyone if you do, as kernel maintainers like Linus
354 Torvalds partly rely on regzbot's tracking in 348 Torvalds partly rely on regzbot's tracking in their work -- for example when
355 deciding to release a new version or extend th 349 deciding to release a new version or extend the development phase. For this they
356 need to be aware of all unfixed regression; to 350 need to be aware of all unfixed regression; to do that, Linus is known to look
357 into the weekly reports sent by regzbot. 351 into the weekly reports sent by regzbot.
358 352
359 Do I have to tell regzbot about every regressi 353 Do I have to tell regzbot about every regression I stumble upon?
360 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 354 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
361 355
362 Ideally yes: we are all humans and easily forg 356 Ideally yes: we are all humans and easily forget problems when something more
363 important unexpectedly comes up -- for example 357 important unexpectedly comes up -- for example a bigger problem in the Linux
364 kernel or something in real life that's keepin 358 kernel or something in real life that's keeping us away from keyboards for a
365 while. Hence, it's best to tell regzbot about 359 while. Hence, it's best to tell regzbot about every regression, except when you
366 immediately write a fix and commit it to a tre 360 immediately write a fix and commit it to a tree regularly merged to the affected
367 kernel series. 361 kernel series.
368 362
369 How to see which regressions regzbot tracks cu 363 How to see which regressions regzbot tracks currently?
370 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 364 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
371 365
372 Check `regzbot's web-interface <https://linux- 366 Check `regzbot's web-interface <https://linux-regtracking.leemhuis.info/regzbot/>`_
373 for the latest info; alternatively, `search fo 367 for the latest info; alternatively, `search for the latest regression report
374 <https://lore.kernel.org/lkml/?q=%22Linux+regr 368 <https://lore.kernel.org/lkml/?q=%22Linux+regressions+report%22+f%3Aregzbot>`_,
375 which regzbot normally sends out once a week o 369 which regzbot normally sends out once a week on Sunday evening (UTC), which is a
376 few hours before Linus usually publishes new ( 370 few hours before Linus usually publishes new (pre-)releases.
377 371
378 What places is regzbot monitoring? 372 What places is regzbot monitoring?
379 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 373 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
380 374
381 Regzbot is watching the most important Linux m 375 Regzbot is watching the most important Linux mailing lists as well as the git
382 repositories of linux-next, mainline, and stab 376 repositories of linux-next, mainline, and stable/longterm.
383 377
384 What kind of issues are supposed to be tracked 378 What kind of issues are supposed to be tracked by regzbot?
385 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 379 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
386 380
387 The bot is meant to track regressions, hence p 381 The bot is meant to track regressions, hence please don't involve regzbot for
388 regular issues. But it's okay for the Linux ke 382 regular issues. But it's okay for the Linux kernel's regression tracker if you
389 use regzbot to track severe issues, like repor 383 use regzbot to track severe issues, like reports about hangs, corrupted data,
390 or internal errors (Panic, Oops, BUG(), warnin 384 or internal errors (Panic, Oops, BUG(), warning, ...).
391 385
392 Can I add regressions found by CI systems to r 386 Can I add regressions found by CI systems to regzbot's tracking?
393 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 387 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
394 388
395 Feel free to do so, if the particular regressi 389 Feel free to do so, if the particular regression likely has impact on practical
396 use cases and thus might be noticed by users; 390 use cases and thus might be noticed by users; hence, please don't involve
397 regzbot for theoretical regressions unlikely t 391 regzbot for theoretical regressions unlikely to show themselves in real world
398 usage. 392 usage.
399 393
400 How to interact with regzbot? 394 How to interact with regzbot?
401 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 395 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
402 396
403 By using a 'regzbot command' in a direct or in 397 By using a 'regzbot command' in a direct or indirect reply to the mail with the
404 regression report. These commands need to be i 398 regression report. These commands need to be in their own paragraph (IOW: they
405 need to be separated from the rest of the mail 399 need to be separated from the rest of the mail using blank lines).
406 400
407 One such command is ``#regzbot introduced: <ve 401 One such command is ``#regzbot introduced: <version or commit>``, which makes
408 regzbot consider your mail as a regressions re 402 regzbot consider your mail as a regressions report added to the tracking, as
409 already described above; ``#regzbot ^introduce 403 already described above; ``#regzbot ^introduced: <version or commit>`` is another
410 such command, which makes regzbot consider the 404 such command, which makes regzbot consider the parent mail as a report for a
411 regression which it starts to track. 405 regression which it starts to track.
412 406
413 Once one of those two commands has been utiliz 407 Once one of those two commands has been utilized, other regzbot commands can be
414 used in direct or indirect replies to the repo 408 used in direct or indirect replies to the report. You can write them below one
415 of the `introduced` commands or in replies to 409 of the `introduced` commands or in replies to the mail that used one of them
416 or itself is a reply to that mail: 410 or itself is a reply to that mail:
417 411
418 * Set or update the title:: 412 * Set or update the title::
419 413
420 #regzbot title: foo 414 #regzbot title: foo
421 415
422 * Monitor a discussion or bugzilla.kernel.org 416 * Monitor a discussion or bugzilla.kernel.org ticket where additions aspects of
423 the issue or a fix are discussed -- for exa 417 the issue or a fix are discussed -- for example the posting of a patch fixing
424 the regression:: 418 the regression::
425 419
426 #regzbot monitor: https://lore.kernel.o 420 #regzbot monitor: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/
427 421
428 Monitoring only works for lore.kernel.org a 422 Monitoring only works for lore.kernel.org and bugzilla.kernel.org; regzbot
429 will consider all messages in that thread o 423 will consider all messages in that thread or ticket as related to the fixing
430 process. 424 process.
431 425
432 * Point to a place with further details of in 426 * Point to a place with further details of interest, like a mailing list post
433 or a ticket in a bug tracker that are sligh 427 or a ticket in a bug tracker that are slightly related, but about a different
434 topic:: 428 topic::
435 429
436 #regzbot link: https://bugzilla.kernel. 430 #regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=123456789
437 431
438 * Mark a regression as fixed by a commit that 432 * Mark a regression as fixed by a commit that is heading upstream or already
439 landed:: 433 landed::
440 434
441 #regzbot fix: 1f2e3d4c5d 435 #regzbot fix: 1f2e3d4c5d
442 436
443 * Mark a regression as a duplicate of another 437 * Mark a regression as a duplicate of another one already tracked by regzbot::
444 438
445 #regzbot dup-of: https://lore.kernel.or 439 #regzbot dup-of: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/
446 440
447 * Mark a regression as invalid:: 441 * Mark a regression as invalid::
448 442
449 #regzbot invalid: wasn't a regression, 443 #regzbot invalid: wasn't a regression, problem has always existed
450 444
451 Is there more to tell about regzbot and its co 445 Is there more to tell about regzbot and its commands?
452 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~ 446 ~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~~
453 447
454 More detailed and up-to-date information about 448 More detailed and up-to-date information about the Linux
455 kernel's regression tracking bot can be found 449 kernel's regression tracking bot can be found on its
456 `project page <https://gitlab.com/knurd42/regz 450 `project page <https://gitlab.com/knurd42/regzbot>`_, which among others
457 contains a `getting started guide <https://git 451 contains a `getting started guide <https://gitlab.com/knurd42/regzbot/-/blob/main/docs/getting_started.md>`_
458 and `reference documentation <https://gitlab.c 452 and `reference documentation <https://gitlab.com/knurd42/regzbot/-/blob/main/docs/reference.md>`_
459 which both cover more details than the above s 453 which both cover more details than the above section.
460 454
461 Quotes from Linus about regression 455 Quotes from Linus about regression
462 ---------------------------------- 456 ----------------------------------
463 457
464 Find below a few real life examples of how Lin 458 Find below a few real life examples of how Linus Torvalds expects regressions to
465 be handled: 459 be handled:
466 460
467 * From `2017-10-26 (1/2) 461 * From `2017-10-26 (1/2)
468 <https://lore.kernel.org/lkml/CA+55aFwiiQYJ+ 462 <https://lore.kernel.org/lkml/CA+55aFwiiQYJ+YoLKCXjN_beDVfu38mg=Ggg5LFOcqHE8Qi7Zw@mail.gmail.com/">https://lore.kernel.org/lkml/CA+55aFwiiQYJ+YoLKCXjN_beDVfu38mg=Ggg5LFOcqHE8Qi7Zw@mail.gmail.com/>`_::
469 463
470 If you break existing user space setups 464 If you break existing user space setups THAT IS A REGRESSION.
471 465
472 It's not ok to say "but we'll fix the u 466 It's not ok to say "but we'll fix the user space setup".
473 467
474 Really. NOT OK. 468 Really. NOT OK.
475 469
476 [...] 470 [...]
477 471
478 The first rule is: 472 The first rule is:
479 473
480 - we don't cause regressions 474 - we don't cause regressions
481 475
482 and the corollary is that when regressi 476 and the corollary is that when regressions *do* occur, we admit to
483 them and fix them, instead of blaming u 477 them and fix them, instead of blaming user space.
484 478
485 The fact that you have apparently been 479 The fact that you have apparently been denying the regression now for
486 three weeks means that I will revert, a 480 three weeks means that I will revert, and I will stop pulling apparmor
487 requests until the people involved unde 481 requests until the people involved understand how kernel development
488 is done. 482 is done.
489 483
490 * From `2017-10-26 (2/2) 484 * From `2017-10-26 (2/2)
491 <https://lore.kernel.org/lkml/CA+55aFxW7NMAM 485 <https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/">https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/>`_::
492 486
493 People should basically always feel lik 487 People should basically always feel like they can update their kernel
494 and simply not have to worry about it. 488 and simply not have to worry about it.
495 489
496 I refuse to introduce "you can only upd 490 I refuse to introduce "you can only update the kernel if you also
497 update that other program" kind of limi 491 update that other program" kind of limitations. If the kernel used to
498 work for you, the rule is that it conti 492 work for you, the rule is that it continues to work for you.
499 493
500 There have been exceptions, but they ar 494 There have been exceptions, but they are few and far between, and they
501 generally have some major and fundament 495 generally have some major and fundamental reasons for having happened,
502 that were basically entirely unavoidabl 496 that were basically entirely unavoidable, and people _tried_hard_ to
503 avoid them. Maybe we can't practically 497 avoid them. Maybe we can't practically support the hardware any more
504 after it is decades old and nobody uses 498 after it is decades old and nobody uses it with modern kernels any
505 more. Maybe there's a serious security 499 more. Maybe there's a serious security issue with how we did things,
506 and people actually depended on that fu 500 and people actually depended on that fundamentally broken model. Maybe
507 there was some fundamental other breaka 501 there was some fundamental other breakage that just _had_ to have a
508 flag day for very core and fundamental 502 flag day for very core and fundamental reasons.
509 503
510 And notice that this is very much about 504 And notice that this is very much about *breaking* peoples environments.
511 505
512 Behavioral changes happen, and maybe we 506 Behavioral changes happen, and maybe we don't even support some
513 feature any more. There's a number of f 507 feature any more. There's a number of fields in /proc/<pid>/stat that
514 are printed out as zeroes, simply becau 508 are printed out as zeroes, simply because they don't even *exist* in
515 the kernel any more, or because showing 509 the kernel any more, or because showing them was a mistake (typically
516 an information leak). But the numbers g 510 an information leak). But the numbers got replaced by zeroes, so that
517 the code that used to parse the fields 511 the code that used to parse the fields still works. The user might not
518 see everything they used to see, and so 512 see everything they used to see, and so behavior is clearly different,
519 but things still _work_, even if they m 513 but things still _work_, even if they might no longer show sensitive
520 (or no longer relevant) information. 514 (or no longer relevant) information.
521 515
522 But if something actually breaks, then 516 But if something actually breaks, then the change must get fixed or
523 reverted. And it gets fixed in the *ker 517 reverted. And it gets fixed in the *kernel*. Not by saying "well, fix
524 your user space then". It was a kernel 518 your user space then". It was a kernel change that exposed the
525 problem, it needs to be the kernel that 519 problem, it needs to be the kernel that corrects for it, because we
526 have a "upgrade in place" model. We don 520 have a "upgrade in place" model. We don't have a "upgrade with new
527 user space". 521 user space".
528 522
529 And I seriously will refuse to take cod 523 And I seriously will refuse to take code from people who do not
530 understand and honor this very simple r 524 understand and honor this very simple rule.
531 525
532 This rule is also not going to change. 526 This rule is also not going to change.
533 527
534 And yes, I realize that the kernel is " 528 And yes, I realize that the kernel is "special" in this respect. I'm
535 proud of it. 529 proud of it.
536 530
537 I have seen, and can point to, lots of 531 I have seen, and can point to, lots of projects that go "We need to
538 break that use case in order to make pr 532 break that use case in order to make progress" or "you relied on
539 undocumented behavior, it sucks to be y 533 undocumented behavior, it sucks to be you" or "there's a better way to
540 do what you want to do, and you have to 534 do what you want to do, and you have to change to that new better
541 way", and I simply don't think that's a 535 way", and I simply don't think that's acceptable outside of very early
542 alpha releases that have experimental u 536 alpha releases that have experimental users that know what they signed
543 up for. The kernel hasn't been in that 537 up for. The kernel hasn't been in that situation for the last two
544 decades. 538 decades.
545 539
546 We do API breakage _inside_ the kernel 540 We do API breakage _inside_ the kernel all the time. We will fix
547 internal problems by saying "you now ne 541 internal problems by saying "you now need to do XYZ", but then it's
548 about internal kernel API's, and the pe 542 about internal kernel API's, and the people who do that then also
549 obviously have to fix up all the in-ker 543 obviously have to fix up all the in-kernel users of that API. Nobody
550 can say "I now broke the API you used, 544 can say "I now broke the API you used, and now _you_ need to fix it
551 up". Whoever broke something gets to fi 545 up". Whoever broke something gets to fix it too.
552 546
553 And we simply do not break user space. 547 And we simply do not break user space.
554 548
555 * From `2020-05-21 549 * From `2020-05-21
556 <https://lore.kernel.org/all/CAHk-=wiVi7mSrs 550 <https://lore.kernel.org/all/CAHk-=wiVi7mSrsMP=fLXQrXK_UimybW=ziLOwSzFTtoXUacWVQ@mail.gmail.com/">https://lore.kernel.org/all/CAHk-=wiVi7mSrsMP=fLXQrXK_UimybW=ziLOwSzFTtoXUacWVQ@mail.gmail.com/>`_::
557 551
558 The rules about regressions have never 552 The rules about regressions have never been about any kind of
559 documented behavior, or where the code 553 documented behavior, or where the code lives.
560 554
561 The rules about regressions are always 555 The rules about regressions are always about "breaks user workflow".
562 556
563 Users are literally the _only_ thing th 557 Users are literally the _only_ thing that matters.
564 558
565 No amount of "you shouldn't have used t 559 No amount of "you shouldn't have used this" or "that behavior was
566 undefined, it's your own fault your app 560 undefined, it's your own fault your app broke" or "that used to work
567 simply because of a kernel bug" is at a 561 simply because of a kernel bug" is at all relevant.
568 562
569 Now, reality is never entirely black-an 563 Now, reality is never entirely black-and-white. So we've had things
570 like "serious security issue" etc that 564 like "serious security issue" etc that just forces us to make changes
571 that may break user space. But even the 565 that may break user space. But even then the rule is that we don't
572 really have other options that would al 566 really have other options that would allow things to continue.
573 567
574 And obviously, if users take years to e 568 And obviously, if users take years to even notice that something
575 broke, or if we have sane ways to work 569 broke, or if we have sane ways to work around the breakage that
576 doesn't make for too much trouble for u 570 doesn't make for too much trouble for users (ie "ok, there are a
577 handful of users, and they can use a ke 571 handful of users, and they can use a kernel command line to work
578 around it" kind of things) we've also b 572 around it" kind of things) we've also been a bit less strict.
579 573
580 But no, "that was documented to be brok 574 But no, "that was documented to be broken" (whether it's because the
581 code was in staging or because the man- 575 code was in staging or because the man-page said something else) is
582 irrelevant. If staging code is so usefu 576 irrelevant. If staging code is so useful that people end up using it,
583 that means that it's basically regular 577 that means that it's basically regular kernel code with a flag saying
584 "please clean this up". 578 "please clean this up".
585 579
586 The other side of the coin is that peop 580 The other side of the coin is that people who talk about "API
587 stability" are entirely wrong. API's do 581 stability" are entirely wrong. API's don't matter either. You can make
588 any changes to an API you like - as lon 582 any changes to an API you like - as long as nobody notices.
589 583
590 Again, the regression rule is not about 584 Again, the regression rule is not about documentation, not about
591 API's, and not about the phase of the m 585 API's, and not about the phase of the moon.
592 586
593 It's entirely about "we caused problems 587 It's entirely about "we caused problems for user space that used to work".
594 588
595 * From `2017-11-05 589 * From `2017-11-05
596 <https://lore.kernel.org/all/CA+55aFzUvbGjD8 590 <https://lore.kernel.org/all/CA+55aFzUvbGjD8nQ-+3oiMBx14c_6zOj2n7KLN3UsJ-qsd4Dcw@mail.gmail.com/">https://lore.kernel.org/all/CA+55aFzUvbGjD8nQ-+3oiMBx14c_6zOj2n7KLN3UsJ-qsd4Dcw@mail.gmail.com/>`_::
597 591
598 And our regression rule has never been 592 And our regression rule has never been "behavior doesn't change".
599 That would mean that we could never mak 593 That would mean that we could never make any changes at all.
600 594
601 For example, we do things like add new 595 For example, we do things like add new error handling etc all the
602 time, which we then sometimes even add 596 time, which we then sometimes even add tests for in our kselftest
603 directory. 597 directory.
604 598
605 So clearly behavior changes all the tim 599 So clearly behavior changes all the time and we don't consider that a
606 regression per se. 600 regression per se.
607 601
608 The rule for a regression for the kerne 602 The rule for a regression for the kernel is that some real user
609 workflow breaks. Not some test. Not a " 603 workflow breaks. Not some test. Not a "look, I used to be able to do
610 X, now I can't". 604 X, now I can't".
611 605
612 * From `2018-08-03 606 * From `2018-08-03
613 <https://lore.kernel.org/all/CA+55aFwWZX=CXm 607 <https://lore.kernel.org/all/CA+55aFwWZX=CXmWDTkDGb36kf12XmTehmQjbiMPCqCRG2hi9kw@mail.gmail.com/">https://lore.kernel.org/all/CA+55aFwWZX=CXmWDTkDGb36kf12XmTehmQjbiMPCqCRG2hi9kw@mail.gmail.com/>`_::
614 608
615 YOU ARE MISSING THE #1 KERNEL RULE. 609 YOU ARE MISSING THE #1 KERNEL RULE.
616 610
617 We do not regress, and we do not regres 611 We do not regress, and we do not regress exactly because your are 100% wrong.
618 612
619 And the reason you state for your opini 613 And the reason you state for your opinion is in fact exactly *WHY* you
620 are wrong. 614 are wrong.
621 615
622 Your "good reasons" are pure and utter 616 Your "good reasons" are pure and utter garbage.
623 617
624 The whole point of "we do not regress" 618 The whole point of "we do not regress" is so that people can upgrade
625 the kernel and never have to worry abou 619 the kernel and never have to worry about it.
626 620
627 > Kernel had a bug which has been fixed 621 > Kernel had a bug which has been fixed
628 622
629 That is *ENTIRELY* immaterial. 623 That is *ENTIRELY* immaterial.
630 624
631 Guys, whether something was buggy or no 625 Guys, whether something was buggy or not DOES NOT MATTER.
632 626
633 Why? 627 Why?
634 628
635 Bugs happen. That's a fact of life. Arg 629 Bugs happen. That's a fact of life. Arguing that "we had to break
636 something because we were fixing a bug" 630 something because we were fixing a bug" is completely insane. We fix
637 tens of bugs every single day, thinking 631 tens of bugs every single day, thinking that "fixing a bug" means that
638 we can break something is simply NOT TR 632 we can break something is simply NOT TRUE.
639 633
640 So bugs simply aren't even relevant to 634 So bugs simply aren't even relevant to the discussion. They happen,
641 they get found, they get fixed, and it 635 they get found, they get fixed, and it has nothing to do with "we
642 break users". 636 break users".
643 637
644 Because the only thing that matters IS 638 Because the only thing that matters IS THE USER.
645 639
646 How hard is that to understand? 640 How hard is that to understand?
647 641
648 Anybody who uses "but it was buggy" as 642 Anybody who uses "but it was buggy" as an argument is entirely missing
649 the point. As far as the USER was conce 643 the point. As far as the USER was concerned, it wasn't buggy - it
650 worked for him/her. 644 worked for him/her.
651 645
652 Maybe it worked *because* the user had 646 Maybe it worked *because* the user had taken the bug into account,
653 maybe it worked because the user didn't 647 maybe it worked because the user didn't notice - again, it doesn't
654 matter. It worked for the user. 648 matter. It worked for the user.
655 649
656 Breaking a user workflow for a "bug" is 650 Breaking a user workflow for a "bug" is absolutely the WORST reason
657 for breakage you can imagine. 651 for breakage you can imagine.
658 652
659 It's basically saying "I took something 653 It's basically saying "I took something that worked, and I broke it,
660 but now it's better". Do you not see ho 654 but now it's better". Do you not see how f*cking insane that statement
661 is? 655 is?
662 656
663 And without users, your program is not 657 And without users, your program is not a program, it's a pointless
664 piece of code that you might as well th 658 piece of code that you might as well throw away.
665 659
666 Seriously. This is *why* the #1 rule fo 660 Seriously. This is *why* the #1 rule for kernel development is "we
667 don't break users". Because "I fixed a 661 don't break users". Because "I fixed a bug" is absolutely NOT AN
668 ARGUMENT if that bug fix broke a user s 662 ARGUMENT if that bug fix broke a user setup. You actually introduced a
669 MUCH BIGGER bug by "fixing" something t 663 MUCH BIGGER bug by "fixing" something that the user clearly didn't
670 even care about. 664 even care about.
671 665
672 And dammit, we upgrade the kernel ALL T 666 And dammit, we upgrade the kernel ALL THE TIME without upgrading any
673 other programs at all. It is absolutely 667 other programs at all. It is absolutely required, because flag-days
674 and dependencies are horribly bad. 668 and dependencies are horribly bad.
675 669
676 And it is also required simply because 670 And it is also required simply because I as a kernel developer do not
677 upgrade random other tools that I don't 671 upgrade random other tools that I don't even care about as I develop
678 the kernel, and I want any of my users 672 the kernel, and I want any of my users to feel safe doing the same
679 time. 673 time.
680 674
681 So no. Your rule is COMPLETELY wrong. I 675 So no. Your rule is COMPLETELY wrong. If you cannot upgrade a kernel
682 without upgrading some other random bin 676 without upgrading some other random binary, then we have a problem.
683 677
684 * From `2021-06-05 678 * From `2021-06-05
685 <https://lore.kernel.org/all/CAHk-=wiUVqHN76 679 <https://lore.kernel.org/all/CAHk-=wiUVqHN76YUwhkjZzwTdjMMJf_zN4+u7vEJjmEGh3recw@mail.gmail.com/">https://lore.kernel.org/all/CAHk-=wiUVqHN76YUwhkjZzwTdjMMJf_zN4+u7vEJjmEGh3recw@mail.gmail.com/>`_::
686 680
687 THERE ARE NO VALID ARGUMENTS FOR REGRES 681 THERE ARE NO VALID ARGUMENTS FOR REGRESSIONS.
688 682
689 Honestly, security people need to under 683 Honestly, security people need to understand that "not working" is not
690 a success case of security. It's a fail 684 a success case of security. It's a failure case.
691 685
692 Yes, "not working" may be secure. But s 686 Yes, "not working" may be secure. But security in that case is *pointless*.
693 687
694 * From `2011-05-06 (1/3) 688 * From `2011-05-06 (1/3)
695 <https://lore.kernel.org/all/BANLkTim9YvResB 689 <https://lore.kernel.org/all/BANLkTim9YvResB+PwRp7QTK-a5VNg2PvmQ@mail.gmail.com/">https://lore.kernel.org/all/BANLkTim9YvResB+PwRp7QTK-a5VNg2PvmQ@mail.gmail.com/>`_::
696 690
697 Binary compatibility is more important. 691 Binary compatibility is more important.
698 692
699 And if binaries don't use the interface 693 And if binaries don't use the interface to parse the format (or just
700 parse it wrongly - see the fairly recen 694 parse it wrongly - see the fairly recent example of adding uuid's to
701 /proc/self/mountinfo), then it's a regr 695 /proc/self/mountinfo), then it's a regression.
702 696
703 And regressions get reverted, unless th 697 And regressions get reverted, unless there are security issues or
704 similar that makes us go "Oh Gods, we r 698 similar that makes us go "Oh Gods, we really have to break things".
705 699
706 I don't understand why this simple logi 700 I don't understand why this simple logic is so hard for some kernel
707 developers to understand. Reality matte 701 developers to understand. Reality matters. Your personal wishes matter
708 NOT AT ALL. 702 NOT AT ALL.
709 703
710 If you made an interface that can be us 704 If you made an interface that can be used without parsing the
711 interface description, then we're stuck 705 interface description, then we're stuck with the interface. Theory
712 simply doesn't matter. 706 simply doesn't matter.
713 707
714 You could help fix the tools, and try t 708 You could help fix the tools, and try to avoid the compatibility
715 issues that way. There aren't that many 709 issues that way. There aren't that many of them.
716 710
717 From `2011-05-06 (2/3) 711 From `2011-05-06 (2/3)
718 <https://lore.kernel.org/all/BANLkTi=KVXjKR8 712 <https://lore.kernel.org/all/BANLkTi=KVXjKR82sqsz4gwjr+E0vtqCmvA@mail.gmail.com/">https://lore.kernel.org/all/BANLkTi=KVXjKR82sqsz4gwjr+E0vtqCmvA@mail.gmail.com/>`_::
719 713
720 it's clearly NOT an internal tracepoint 714 it's clearly NOT an internal tracepoint. By definition. It's being
721 used by powertop. 715 used by powertop.
722 716
723 From `2011-05-06 (3/3) 717 From `2011-05-06 (3/3)
724 <https://lore.kernel.org/all/BANLkTinazaXRdG 718 <https://lore.kernel.org/all/BANLkTinazaXRdGovYL7rRVp+j6HbJ7pzhg@mail.gmail.com/">https://lore.kernel.org/all/BANLkTinazaXRdGovYL7rRVp+j6HbJ7pzhg@mail.gmail.com/>`_::
725 719
726 We have programs that use that ABI and 720 We have programs that use that ABI and thus it's a regression if they break.
727 721
728 * From `2012-07-06 <https://lore.kernel.org/al 722 * From `2012-07-06 <https://lore.kernel.org/all/CA+55aFwnLJ+0sjx92EGREGTWOx84wwKaraSzpTNJwPVV8edw8g@mail.gmail.com/">https://lore.kernel.org/all/CA+55aFwnLJ+0sjx92EGREGTWOx84wwKaraSzpTNJwPVV8edw8g@mail.gmail.com/>`_::
729 723
730 > Now this got me wondering if Debian _ 724 > Now this got me wondering if Debian _unstable_ actually qualifies as a
731 > standard distro userspace. 725 > standard distro userspace.
732 726
733 Oh, if the kernel breaks some standard 727 Oh, if the kernel breaks some standard user space, that counts. Tons
734 of people run Debian unstable 728 of people run Debian unstable
735 729
736 * From `2019-09-15 730 * From `2019-09-15
737 <https://lore.kernel.org/lkml/CAHk-=wiP4K8DR 731 <https://lore.kernel.org/lkml/CAHk-=wiP4K8DRJWsCo=20hn_6054xBamGKF2kPgUzpB5aMaofA@mail.gmail.com/">https://lore.kernel.org/lkml/CAHk-=wiP4K8DRJWsCo=20hn_6054xBamGKF2kPgUzpB5aMaofA@mail.gmail.com/>`_::
738 732
739 One _particularly_ last-minute revert i 733 One _particularly_ last-minute revert is the top-most commit (ignoring
740 the version change itself) done just be 734 the version change itself) done just before the release, and while
741 it's very annoying, it's perhaps also i 735 it's very annoying, it's perhaps also instructive.
742 736
743 What's instructive about it is that I r 737 What's instructive about it is that I reverted a commit that wasn't
744 actually buggy. In fact, it was doing e 738 actually buggy. In fact, it was doing exactly what it set out to do,
745 and did it very well. In fact it did it 739 and did it very well. In fact it did it _so_ well that the much
746 improved IO patterns it caused then end 740 improved IO patterns it caused then ended up revealing a user-visible
747 regression due to a real bug in a compl 741 regression due to a real bug in a completely unrelated area.
748 742
749 The actual details of that regression a 743 The actual details of that regression are not the reason I point that
750 revert out as instructive, though. It's 744 revert out as instructive, though. It's more that it's an instructive
751 example of what counts as a regression, 745 example of what counts as a regression, and what the whole "no
752 regressions" kernel rule means. The rev 746 regressions" kernel rule means. The reverted commit didn't change any
753 API's, and it didn't introduce any new 747 API's, and it didn't introduce any new bugs. But it ended up exposing
754 another problem, and as such caused a k 748 another problem, and as such caused a kernel upgrade to fail for a
755 user. So it got reverted. 749 user. So it got reverted.
756 750
757 The point here being that we revert bas 751 The point here being that we revert based on user-reported _behavior_,
758 not based on some "it changes the ABI" 752 not based on some "it changes the ABI" or "it caused a bug" concept.
759 The problem was really pre-existing, an 753 The problem was really pre-existing, and it just didn't happen to
760 trigger before. The better IO patterns 754 trigger before. The better IO patterns introduced by the change just
761 happened to expose an old bug, and peop 755 happened to expose an old bug, and people had grown to depend on the
762 previously benign behavior of that old 756 previously benign behavior of that old issue.
763 757
764 And never fear, we'll re-introduce the 758 And never fear, we'll re-introduce the fix that improved on the IO
765 patterns once we've decided just how to 759 patterns once we've decided just how to handle the fact that we had a
766 bad interaction with an interface that 760 bad interaction with an interface that people had then just happened
767 to rely on incidental behavior for befo 761 to rely on incidental behavior for before. It's just that we'll have
768 to hash through how to do that (there a 762 to hash through how to do that (there are no less than three different
769 patches by three different developers b 763 patches by three different developers being discussed, and there might
770 be more coming...). In the meantime, I 764 be more coming...). In the meantime, I reverted the thing that exposed
771 the problem to users for this release, 765 the problem to users for this release, even if I hope it will be
772 re-introduced (perhaps even backported 766 re-introduced (perhaps even backported as a stable patch) once we have
773 consensus about the issue it exposed. 767 consensus about the issue it exposed.
774 768
775 Take-away from the whole thing: it's no 769 Take-away from the whole thing: it's not about whether you change the
776 kernel-userspace ABI, or fix a bug, or 770 kernel-userspace ABI, or fix a bug, or about whether the old code
777 "should never have worked in the first 771 "should never have worked in the first place". It's about whether
778 something breaks existing users' workfl 772 something breaks existing users' workflow.
779 773
780 Anyway, that was my little aside on the 774 Anyway, that was my little aside on the whole regression thing. Since
781 it's that "first rule of kernel program 775 it's that "first rule of kernel programming", I felt it is perhaps
782 worth just bringing it up every once in 776 worth just bringing it up every once in a while
783 777
784 .. 778 ..
785 end-of-content 779 end-of-content
786 .. 780 ..
787 This text is available under GPL-2.0+ or CC 781 This text is available under GPL-2.0+ or CC-BY-4.0, as stated at the top
788 of the file. If you want to distribute this 782 of the file. If you want to distribute this text under CC-BY-4.0 only,
789 please use "The Linux kernel developers" fo 783 please use "The Linux kernel developers" for author attribution and link
790 this as source: 784 this as source:
791 https://git.kernel.org/pub/scm/linux/kernel 785 https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/plain/Documentation/process/handling-regressions.rst
792 .. 786 ..
793 Note: Only the content of this RST file as 787 Note: Only the content of this RST file as found in the Linux kernel sources
794 is available under CC-BY-4.0, as versions o 788 is available under CC-BY-4.0, as versions of this text that were processed
795 (for example by the kernel's build system) 789 (for example by the kernel's build system) might contain content taken from
796 files which use a more restrictive license. 790 files which use a more restrictive license.
Linux® is a registered trademark of Linus Torvalds in the United States and other countries.
TOMOYO® is a registered trademark of NTT DATA CORPORATION.