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