$sphinx.addnodesdocument)}( rawsourcechildren]( translations LanguagesNode)}(hhh](h pending_xref)}(hhh]docutils.nodesTextChinese (Simplified)}parenthsba attributes}(ids]classes]names]dupnames]backrefs] refdomainstdreftypedoc reftarget0/translations/zh_CN/process/handling-regressionsmodnameN classnameN refexplicitutagnamehhh ubh)}(hhh]hChinese (Traditional)}hh2sbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/zh_TW/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hItalian}hhFsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/it_IT/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hJapanese}hhZsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/ja_JP/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hKorean}hhnsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/ko_KR/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hPortuguese (Brazilian)}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/pt_BR/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubh)}(hhh]hSpanish}hhsbah}(h]h ]h"]h$]h&] refdomainh)reftypeh+ reftarget0/translations/sp_SP/process/handling-regressionsmodnameN classnameN refexplicituh1hhh ubeh}(h]h ]h"]h$]h&]current_languageEnglishuh1h hh _documenthsourceNlineNubhcomment)}(h0SPDX-License-Identifier: (GPL-2.0+ OR CC-BY-4.0)h]h0SPDX-License-Identifier: (GPL-2.0+ OR CC-BY-4.0)}hhsbah}(h]h ]h"]h$]h&] xml:spacepreserveuh1hhhhhhJ/var/lib/git/docbuild/linux/Documentation/process/handling-regressions.rsthKubh)}(hFSee the bottom of this file for additional redistribution information.h]hFSee the bottom of this file for additional redistribution information.}hhsbah}(h]h ]h"]h$]h&]hhuh1hhhhhhhhKubhsection)}(hhh](htitle)}(hHandling regressionsh]hHandling regressions}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhhhhhKubh paragraph)}(hXV*We don't cause regressions* -- this document describes what this "first rule of Linux kernel development" means in practice for developers. It complements Documentation/admin-guide/reporting-regressions.rst, which covers the topic from a user's point of view; if you never read that text, go and at least skim over it before continuing here.h](hemphasis)}(h*We don't cause regressions*h]hWe don’t cause regressions}(hhhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhubhX@ -- this document describes what this “first rule of Linux kernel development” means in practice for developers. It complements Documentation/admin-guide/reporting-regressions.rst, which covers the topic from a user’s point of view; if you never read that text, go and at least skim over it before continuing here.}(hhhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhhhhubh)}(hhh](h)}(h$The important bits (aka "The TL;DR")h]h(The important bits (aka “The TL;DR”)}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhKubhenumerated_list)}(hhh](h list_item)}(hXEnsure subscribers of the `regression mailing list `_ (regressions@lists.linux.dev) quickly become aware of any new regression report: * When receiving a mailed report that did not CC the list, bring it into the loop by immediately sending at least a brief "Reply-all" with the list CCed. * Forward or bounce any reports submitted in bug trackers to the list. h](h)}(hEnsure subscribers of the `regression mailing list `_ (regressions@lists.linux.dev) quickly become aware of any new regression report:h](hEnsure subscribers of the }(hj'hhhNhNubh reference)}(hA`regression mailing list `_h]hregression mailing list}(hj1hhhNhNubah}(h]h ]h"]h$]h&]nameregression mailing listrefuri$https://lore.kernel.org/regressions/uh1j/hj'ubhtarget)}(h' h]h}(h]regression-mailing-listah ]h"]regression mailing listah$]h&]refurijBuh1jC referencedKhj'ubh (}(hj'hhhNhNubj0)}(hregressions@lists.linux.devh]hregressions@lists.linux.dev}(hjWhhhNhNubah}(h]h ]h"]h$]h&]refuri"mailto:regressions@lists.linux.devuh1j/hj'ubh4) quickly become aware of any new regression report:}(hj'hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhj#ubh block_quote)}(h* When receiving a mailed report that did not CC the list, bring it into the loop by immediately sending at least a brief "Reply-all" with the list CCed. * Forward or bounce any reports submitted in bug trackers to the list. h]h bullet_list)}(hhh](j")}(hWhen receiving a mailed report that did not CC the list, bring it into the loop by immediately sending at least a brief "Reply-all" with the list CCed. h]h)}(hWhen receiving a mailed report that did not CC the list, bring it into the loop by immediately sending at least a brief "Reply-all" with the list CCed.h]hWhen receiving a mailed report that did not CC the list, bring it into the loop by immediately sending at least a brief “Reply-all” with the list CCed.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj|ubah}(h]h ]h"]h$]h&]uh1j!hjyubj")}(hEForward or bounce any reports submitted in bug trackers to the list. h]h)}(hDForward or bounce any reports submitted in bug trackers to the list.h]hDForward or bounce any reports submitted in bug trackers to the list.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjyubeh}(h]h ]h"]h$]h&]bullet*uh1jwhhhKhjsubah}(h]h ]h"]h$]h&]uh1jqhhhKhj#ubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXMake the Linux kernel regression tracking bot "regzbot" track the issue (this is optional, but recommended): * For mailed reports, check if the reporter included a line like ``#regzbot introduced: v5.13..v5.14-rc1``. If not, send a reply (with the regressions list in CC) containing a paragraph like the following, which tells regzbot when the issue started to happen:: #regzbot ^introduced: 1f2e3d4c5b6a * When forwarding reports from a bug tracker to the regressions list (see above), include a paragraph like the following:: #regzbot introduced: v5.13..v5.14-rc1 #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789 h](h)}(hlMake the Linux kernel regression tracking bot "regzbot" track the issue (this is optional, but recommended):h]hpMake the Linux kernel regression tracking bot “regzbot” track the issue (this is optional, but recommended):}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubjr)}(hXc* For mailed reports, check if the reporter included a line like ``#regzbot introduced: v5.13..v5.14-rc1``. If not, send a reply (with the regressions list in CC) containing a paragraph like the following, which tells regzbot when the issue started to happen:: #regzbot ^introduced: 1f2e3d4c5b6a * When forwarding reports from a bug tracker to the regressions list (see above), include a paragraph like the following:: #regzbot introduced: v5.13..v5.14-rc1 #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789 h]jx)}(hhh](j")}(hX(For mailed reports, check if the reporter included a line like ``#regzbot introduced: v5.13..v5.14-rc1``. If not, send a reply (with the regressions list in CC) containing a paragraph like the following, which tells regzbot when the issue started to happen:: #regzbot ^introduced: 1f2e3d4c5b6a h](h)}(hXFor mailed reports, check if the reporter included a line like ``#regzbot introduced: v5.13..v5.14-rc1``. If not, send a reply (with the regressions list in CC) containing a paragraph like the following, which tells regzbot when the issue started to happen::h](h?For mailed reports, check if the reporter included a line like }(hjhhhNhNubhliteral)}(h)``#regzbot introduced: v5.13..v5.14-rc1``h]h%#regzbot introduced: v5.13..v5.14-rc1}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh. If not, send a reply (with the regressions list in CC) containing a paragraph like the following, which tells regzbot when the issue started to happen:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjubh literal_block)}(h"#regzbot ^introduced: 1f2e3d4c5b6ah]h"#regzbot ^introduced: 1f2e3d4c5b6a}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhK"hjubeh}(h]h ]h"]h$]h&]uh1j!hjubj")}(hX&When forwarding reports from a bug tracker to the regressions list (see above), include a paragraph like the following:: #regzbot introduced: v5.13..v5.14-rc1 #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789 h](h)}(hxWhen forwarding reports from a bug tracker to the regressions list (see above), include a paragraph like the following::h]hwWhen forwarding reports from a bug tracker to the regressions list (see above), include a paragraph like the following:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK$hjubj)}(h#regzbot introduced: v5.13..v5.14-rc1 #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789h]h#regzbot introduced: v5.13..v5.14-rc1 #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789}hj'sbah}(h]h ]h"]h$]h&]hhuh1jhhhK'hjubeh}(h]h ]h"]h$]h&]uh1j!hjubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhjubah}(h]h ]h"]h$]h&]uh1jqhhhKhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXWhen submitting fixes for regressions, add "Closes:" tags to the patch description pointing to all places where the issue was reported, as mandated by Documentation/process/submitting-patches.rst and :ref:`Documentation/process/5.Posting.rst `. If you are only fixing part of the issue that caused the regression, you may use "Link:" tags instead. regzbot currently makes no distinction between the two. h]h)}(hXWhen submitting fixes for regressions, add "Closes:" tags to the patch description pointing to all places where the issue was reported, as mandated by Documentation/process/submitting-patches.rst and :ref:`Documentation/process/5.Posting.rst `. If you are only fixing part of the issue that caused the regression, you may use "Link:" tags instead. regzbot currently makes no distinction between the two.h](hWhen submitting fixes for regressions, add “Closes:” tags to the patch description pointing to all places where the issue was reported, as mandated by Documentation/process/submitting-patches.rst and }(hjQhhhNhNubh)}(h@:ref:`Documentation/process/5.Posting.rst `h]hinline)}(hj[h]h#Documentation/process/5.Posting.rst}(hj_hhhNhNubah}(h]h ](xrefstdstd-refeh"]h$]h&]uh1j]hjYubah}(h]h ]h"]h$]h&]refdocprocess/handling-regressions refdomainjjreftyperef refexplicitrefwarn reftargetdevelopment_postinguh1hhhhK+hjQubh. If you are only fixing part of the issue that caused the regression, you may use “Link:” tags instead. regzbot currently makes no distinction between the two.}(hjQhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhK+hjMubah}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hTry to fix regressions quickly once the culprit has been identified; fixes for most regressions should be merged within two weeks, but some need to be resolved within two or three days. h]h)}(hTry to fix regressions quickly once the culprit has been identified; fixes for most regressions should be merged within two weeks, but some need to be resolved within two or three days.h]hTry to fix regressions quickly once the culprit has been identified; fixes for most regressions should be merged within two weeks, but some need to be resolved within two or three days.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK3hjubah}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]enumtypearabicprefixhsuffix.uh1jhj hhhhhKubeh}(h] the-important-bits-aka-the-tl-drah ]h"]$the important bits (aka "the tl;dr")ah$]h&]uh1hhhhhhhhKubh)}(hhh](h)}(hCAll the details on Linux kernel regressions relevant for developersh]hCAll the details on Linux kernel regressions relevant for developers}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhK9ubh)}(hhh](h)}(h#The important basics in more detailh]h#The important basics in more detail}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhK=ubh)}(hhh](h)}(h,What to do when receiving regression reportsh]h,What to do when receiving regression reports}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhKAubh)}(hEnsure the Linux kernel's regression tracker and others subscribers of the `regression mailing list `_ (regressions@lists.linux.dev) become aware of any newly reported regression:h](hMEnsure the Linux kernel’s regression tracker and others subscribers of the }(hjhhhNhNubj0)}(hA`regression mailing list `_h]hregression mailing list}(hjhhhNhNubah}(h]h ]h"]h$]h&]nameregression mailing listjA$https://lore.kernel.org/regressions/uh1j/hjubjD)}(h' h]h}(h]id1ah ]h"]h$]regression mailing listah&]refurijuh1jCjRKhjubh (}(hjhhhNhNubj0)}(hregressions@lists.linux.devh]hregressions@lists.linux.dev}(hjhhhNhNubah}(h]h ]h"]h$]h&]refuri"mailto:regressions@lists.linux.devuh1j/hjubh0) become aware of any newly reported regression:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKChjhhubjr)}(hX* When you receive a report by mail that did not CC the list, immediately bring it into the loop by sending at least a brief "Reply-all" with the list CCed; try to ensure it gets CCed again in case you reply to a reply that omitted the list. * If a report submitted in a bug tracker hits your Inbox, forward or bounce it to the list. Consider checking the list archives beforehand, if the reporter already forwarded the report as instructed by Documentation/admin-guide/reporting-issues.rst. h]jx)}(hhh](j")}(hWhen you receive a report by mail that did not CC the list, immediately bring it into the loop by sending at least a brief "Reply-all" with the list CCed; try to ensure it gets CCed again in case you reply to a reply that omitted the list. h]h)}(hWhen you receive a report by mail that did not CC the list, immediately bring it into the loop by sending at least a brief "Reply-all" with the list CCed; try to ensure it gets CCed again in case you reply to a reply that omitted the list.h]hWhen you receive a report by mail that did not CC the list, immediately bring it into the loop by sending at least a brief “Reply-all” with the list CCed; try to ensure it gets CCed again in case you reply to a reply that omitted the list.}(hj;hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKGhj7ubah}(h]h ]h"]h$]h&]uh1j!hj4ubj")}(hIf a report submitted in a bug tracker hits your Inbox, forward or bounce it to the list. Consider checking the list archives beforehand, if the reporter already forwarded the report as instructed by Documentation/admin-guide/reporting-issues.rst. h]h)}(hIf a report submitted in a bug tracker hits your Inbox, forward or bounce it to the list. Consider checking the list archives beforehand, if the reporter already forwarded the report as instructed by Documentation/admin-guide/reporting-issues.rst.h]hIf a report submitted in a bug tracker hits your Inbox, forward or bounce it to the list. Consider checking the list archives beforehand, if the reporter already forwarded the report as instructed by Documentation/admin-guide/reporting-issues.rst.}(hjShhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKLhjOubah}(h]h ]h"]h$]h&]uh1j!hj4ubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKGhj0ubah}(h]h ]h"]h$]h&]uh1jqhhhKGhjhhubh)}(h{When doing either, consider making the Linux kernel regression tracking bot "regzbot" immediately start tracking the issue:h]hWhen doing either, consider making the Linux kernel regression tracking bot “regzbot” immediately start tracking the issue:}(hjshhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKQhjhhubjr)}(hX* For mailed reports, check if the reporter included a "regzbot command" like ``#regzbot introduced: 1f2e3d4c5b6a``. If not, send a reply (with the regressions list in CC) with a paragraph like the following::: #regzbot ^introduced: v5.13..v5.14-rc1 This tells regzbot the version range in which the issue started to happen; you can specify a range using commit-ids as well or state a single commit-id in case the reporter bisected the culprit. Note the caret (^) before the "introduced": it tells regzbot to treat the parent mail (the one you reply to) as the initial report for the regression you want to see tracked; that's important, as regzbot will later look out for patches with "Closes:" tags pointing to the report in the archives on lore.kernel.org. * When forwarding a regression reported to a bug tracker, include a paragraph with these regzbot commands:: #regzbot introduced: 1f2e3d4c5b6a #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789 Regzbot will then automatically associate patches with the report that contain "Closes:" tags pointing to your mail or the mentioned ticket. h]jx)}(hhh](j")}(hXFor mailed reports, check if the reporter included a "regzbot command" like ``#regzbot introduced: 1f2e3d4c5b6a``. If not, send a reply (with the regressions list in CC) with a paragraph like the following::: #regzbot ^introduced: v5.13..v5.14-rc1 This tells regzbot the version range in which the issue started to happen; you can specify a range using commit-ids as well or state a single commit-id in case the reporter bisected the culprit. Note the caret (^) before the "introduced": it tells regzbot to treat the parent mail (the one you reply to) as the initial report for the regression you want to see tracked; that's important, as regzbot will later look out for patches with "Closes:" tags pointing to the report in the archives on lore.kernel.org. h](h)}(hFor mailed reports, check if the reporter included a "regzbot command" like ``#regzbot introduced: 1f2e3d4c5b6a``. If not, send a reply (with the regressions list in CC) with a paragraph like the following:::h](hPFor mailed reports, check if the reporter included a “regzbot command” like }(hjhhhNhNubj)}(h%``#regzbot introduced: 1f2e3d4c5b6a``h]h!#regzbot introduced: 1f2e3d4c5b6a}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh^. If not, send a reply (with the regressions list in CC) with a paragraph like the following::}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKThjubj)}(h&#regzbot ^introduced: v5.13..v5.14-rc1h]h&#regzbot ^introduced: v5.13..v5.14-rc1}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhKXhjubh)}(hThis tells regzbot the version range in which the issue started to happen; you can specify a range using commit-ids as well or state a single commit-id in case the reporter bisected the culprit.h]hThis tells regzbot the version range in which the issue started to happen; you can specify a range using commit-ids as well or state a single commit-id in case the reporter bisected the culprit.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKZhjubh)}(hX:Note the caret (^) before the "introduced": it tells regzbot to treat the parent mail (the one you reply to) as the initial report for the regression you want to see tracked; that's important, as regzbot will later look out for patches with "Closes:" tags pointing to the report in the archives on lore.kernel.org.h]hXDNote the caret (^) before the “introduced”: it tells regzbot to treat the parent mail (the one you reply to) as the initial report for the regression you want to see tracked; that’s important, as regzbot will later look out for patches with “Closes:” tags pointing to the report in the archives on lore.kernel.org.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK^hjubeh}(h]h ]h"]h$]h&]uh1j!hjubj")}(hXWhen forwarding a regression reported to a bug tracker, include a paragraph with these regzbot commands:: #regzbot introduced: 1f2e3d4c5b6a #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789 Regzbot will then automatically associate patches with the report that contain "Closes:" tags pointing to your mail or the mentioned ticket. h](h)}(hiWhen forwarding a regression reported to a bug tracker, include a paragraph with these regzbot commands::h]hhWhen forwarding a regression reported to a bug tracker, include a paragraph with these regzbot commands:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKdhjubj)}(h#regzbot introduced: 1f2e3d4c5b6a #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789h]h#regzbot introduced: 1f2e3d4c5b6a #regzbot from: Some N. Ice Human #regzbot monitor: http://some.bugtracker.example.com/ticket?id=123456789}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhKghjubh)}(hRegzbot will then automatically associate patches with the report that contain "Closes:" tags pointing to your mail or the mentioned ticket.h]hRegzbot will then automatically associate patches with the report that contain “Closes:” tags pointing to your mail or the mentioned ticket.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKkhjubeh}(h]h ]h"]h$]h&]uh1j!hjubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKThjubah}(h]h ]h"]h$]h&]uh1jqhhhKThjhhubeh}(h],what-to-do-when-receiving-regression-reportsah ]h"],what to do when receiving regression reportsah$]h&]uh1hhjhhhhhKAubh)}(hhh](h)}(h(What's important when fixing regressionsh]h*What’s important when fixing regressions}(hj'hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj$hhhhhKoubh)}(hX&You don't need to do anything special when submitting fixes for regression, just remember to do what Documentation/process/submitting-patches.rst, :ref:`Documentation/process/5.Posting.rst `, and Documentation/process/stable-kernel-rules.rst already explain in more detail:h](hYou don’t need to do anything special when submitting fixes for regression, just remember to do what Documentation/process/submitting-patches.rst, }(hj5hhhNhNubh)}(h@:ref:`Documentation/process/5.Posting.rst `h]j^)}(hj?h]h#Documentation/process/5.Posting.rst}(hjAhhhNhNubah}(h]h ](jistdstd-refeh"]h$]h&]uh1j]hj=ubah}(h]h ]h"]h$]h&]refdocjv refdomainjKreftyperef refexplicitrefwarnj|development_postinguh1hhhhKqhj5ubhS, and Documentation/process/stable-kernel-rules.rst already explain in more detail:}(hj5hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKqhj$hhubjr)}(hX* Point to all places where the issue was reported using "Closes:" tags:: Closes: https://lore.kernel.org/r/30th.anniversary.repost@klaava.Helsinki.FI/ Closes: https://bugzilla.kernel.org/show_bug.cgi?id=1234567890 If you are only fixing part of the issue, you may use "Link:" instead as described in the first document mentioned above. regzbot currently treats both of these equivalently and considers the linked reports as resolved. * Add a "Fixes:" tag to specify the commit causing the regression. * If the culprit was merged in an earlier development cycle, explicitly mark the fix for backporting using the ``Cc: stable@vger.kernel.org`` tag. h]jx)}(hhh](j")}(hXPoint to all places where the issue was reported using "Closes:" tags:: Closes: https://lore.kernel.org/r/30th.anniversary.repost@klaava.Helsinki.FI/ Closes: https://bugzilla.kernel.org/show_bug.cgi?id=1234567890 If you are only fixing part of the issue, you may use "Link:" instead as described in the first document mentioned above. regzbot currently treats both of these equivalently and considers the linked reports as resolved. h](h)}(hGPoint to all places where the issue was reported using "Closes:" tags::h]hJPoint to all places where the issue was reported using “Closes:” tags:}(hjrhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKvhjnubj)}(hCloses: https://lore.kernel.org/r/30th.anniversary.repost@klaava.Helsinki.FI/ Closes: https://bugzilla.kernel.org/show_bug.cgi?id=1234567890h]hCloses: https://lore.kernel.org/r/30th.anniversary.repost@klaava.Helsinki.FI/ Closes: https://bugzilla.kernel.org/show_bug.cgi?id=1234567890}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhKxhjnubh)}(hIf you are only fixing part of the issue, you may use "Link:" instead as described in the first document mentioned above. regzbot currently treats both of these equivalently and considers the linked reports as resolved.h]hIf you are only fixing part of the issue, you may use “Link:” instead as described in the first document mentioned above. regzbot currently treats both of these equivalently and considers the linked reports as resolved.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhK{hjnubeh}(h]h ]h"]h$]h&]uh1j!hjkubj")}(hAAdd a "Fixes:" tag to specify the commit causing the regression. h]h)}(h@Add a "Fixes:" tag to specify the commit causing the regression.h]hDAdd a “Fixes:” tag to specify the commit causing the regression.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjkubj")}(hIf the culprit was merged in an earlier development cycle, explicitly mark the fix for backporting using the ``Cc: stable@vger.kernel.org`` tag. h]h)}(hIf the culprit was merged in an earlier development cycle, explicitly mark the fix for backporting using the ``Cc: stable@vger.kernel.org`` tag.h](hmIf the culprit was merged in an earlier development cycle, explicitly mark the fix for backporting using the }(hjhhhNhNubj)}(h``Cc: stable@vger.kernel.org``h]hCc: stable@vger.kernel.org}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1jhjubh tag.}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjkubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKvhjgubah}(h]h ]h"]h$]h&]uh1jqhhhKvhj$hhubh)}(hXAll this is expected from you and important when it comes to regression, as these tags are of great value for everyone (you included) that might be looking into the issue weeks, months, or years later. These tags are also crucial for tools and scripts used by other kernel developers or Linux distributions; one of these tools is regzbot, which heavily relies on the "Closes:" tags to associate reports for regression with changes resolving them.h]hXAll this is expected from you and important when it comes to regression, as these tags are of great value for everyone (you included) that might be looking into the issue weeks, months, or years later. These tags are also crucial for tools and scripts used by other kernel developers or Linux distributions; one of these tools is regzbot, which heavily relies on the “Closes:” tags to associate reports for regression with changes resolving them.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj$hhubeh}(h](what-s-important-when-fixing-regressionsah ]h"](what's important when fixing regressionsah$]h&]uh1hhjhhhhhKoubh)}(hhh](h)}(h6Expectations and best practices for fixing regressionsh]h6Expectations and best practices for fixing regressions}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhKubh)}(hAs a Linux kernel developer, you are expected to give your best to prevent situations where a regression caused by a recent change of yours leaves users only these options:h]hAs a Linux kernel developer, you are expected to give your best to prevent situations where a regression caused by a recent change of yours leaves users only these options:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hX* Run a kernel with a regression that impacts usage. * Switch to an older or newer kernel series. * Continue running an outdated and thus potentially insecure kernel for more than three weeks after the regression's culprit was identified. Ideally it should be less than two. And it ought to be just a few days, if the issue is severe or affects many users -- either in general or in prevalent environments. h]jx)}(hhh](j")}(h3Run a kernel with a regression that impacts usage. h]h)}(h2Run a kernel with a regression that impacts usage.h]h2Run a kernel with a regression that impacts usage.}(hj0hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj,ubah}(h]h ]h"]h$]h&]uh1j!hj)ubj")}(h+Switch to an older or newer kernel series. h]h)}(h*Switch to an older or newer kernel series.h]h*Switch to an older or newer kernel series.}(hjHhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjDubah}(h]h ]h"]h$]h&]uh1j!hj)ubj")}(hX3Continue running an outdated and thus potentially insecure kernel for more than three weeks after the regression's culprit was identified. Ideally it should be less than two. And it ought to be just a few days, if the issue is severe or affects many users -- either in general or in prevalent environments. h]h)}(hX2Continue running an outdated and thus potentially insecure kernel for more than three weeks after the regression's culprit was identified. Ideally it should be less than two. And it ought to be just a few days, if the issue is severe or affects many users -- either in general or in prevalent environments.h]hX4Continue running an outdated and thus potentially insecure kernel for more than three weeks after the regression’s culprit was identified. Ideally it should be less than two. And it ought to be just a few days, if the issue is severe or affects many users -- either in general or in prevalent environments.}(hj`hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj\ubah}(h]h ]h"]h$]h&]uh1j!hj)ubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhj%ubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubh)}(hhHow to realize that in practice depends on various factors. Use the following rules of thumb as a guide.h]hhHow to realize that in practice depends on various factors. Use the following rules of thumb as a guide.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubh)}(h In general:h]h In general:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hX* Prioritize work on regressions over all other Linux kernel work, unless the latter concerns a severe issue (e.g. acute security vulnerability, data loss, bricked hardware, ...). * Expedite fixing mainline regressions that recently made it into a proper mainline, stable, or longterm release (either directly or via backport). * Do not consider regressions from the current cycle as something that can wait till the end of the cycle, as the issue might discourage or prevent users and CI systems from testing mainline now or generally. * Work with the required care to avoid additional or bigger damage, even if resolving an issue then might take longer than outlined below. h]jx)}(hhh](j")}(hPrioritize work on regressions over all other Linux kernel work, unless the latter concerns a severe issue (e.g. acute security vulnerability, data loss, bricked hardware, ...). h]h)}(hPrioritize work on regressions over all other Linux kernel work, unless the latter concerns a severe issue (e.g. acute security vulnerability, data loss, bricked hardware, ...).h]hPrioritize work on regressions over all other Linux kernel work, unless the latter concerns a severe issue (e.g. acute security vulnerability, data loss, bricked hardware, ...).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hExpedite fixing mainline regressions that recently made it into a proper mainline, stable, or longterm release (either directly or via backport). h]h)}(hExpedite fixing mainline regressions that recently made it into a proper mainline, stable, or longterm release (either directly or via backport).h]hExpedite fixing mainline regressions that recently made it into a proper mainline, stable, or longterm release (either directly or via backport).}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hDo not consider regressions from the current cycle as something that can wait till the end of the cycle, as the issue might discourage or prevent users and CI systems from testing mainline now or generally. h]h)}(hDo not consider regressions from the current cycle as something that can wait till the end of the cycle, as the issue might discourage or prevent users and CI systems from testing mainline now or generally.h]hDo not consider regressions from the current cycle as something that can wait till the end of the cycle, as the issue might discourage or prevent users and CI systems from testing mainline now or generally.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hWork with the required care to avoid additional or bigger damage, even if resolving an issue then might take longer than outlined below. h]h)}(hWork with the required care to avoid additional or bigger damage, even if resolving an issue then might take longer than outlined below.h]hWork with the required care to avoid additional or bigger damage, even if resolving an issue then might take longer than outlined below.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhjubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubh)}(h4On timing once the culprit of a regression is known:h]h4On timing once the culprit of a regression is known:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hX* Aim to mainline a fix within two or three days, if the issue is severe or bothering many users -- either in general or in prevalent conditions like a particular hardware environment, distribution, or stable/longterm series. * Aim to mainline a fix by Sunday after the next, if the culprit made it into a recent mainline, stable, or longterm release (either directly or via backport); if the culprit became known early during a week and is simple to resolve, try to mainline the fix within the same week. * For other regressions, aim to mainline fixes before the hindmost Sunday within the next three weeks. One or two Sundays later are acceptable, if the regression is something people can live with easily for a while -- like a mild performance regression. * It's strongly discouraged to delay mainlining regression fixes till the next merge window, except when the fix is extraordinarily risky or when the culprit was mainlined more than a year ago. h]jx)}(hhh](j")}(hAim to mainline a fix within two or three days, if the issue is severe or bothering many users -- either in general or in prevalent conditions like a particular hardware environment, distribution, or stable/longterm series. h]h)}(hAim to mainline a fix within two or three days, if the issue is severe or bothering many users -- either in general or in prevalent conditions like a particular hardware environment, distribution, or stable/longterm series.h]hAim to mainline a fix within two or three days, if the issue is severe or bothering many users -- either in general or in prevalent conditions like a particular hardware environment, distribution, or stable/longterm series.}(hj(hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj$ubah}(h]h ]h"]h$]h&]uh1j!hj!ubj")}(hXAim to mainline a fix by Sunday after the next, if the culprit made it into a recent mainline, stable, or longterm release (either directly or via backport); if the culprit became known early during a week and is simple to resolve, try to mainline the fix within the same week. h]h)}(hXAim to mainline a fix by Sunday after the next, if the culprit made it into a recent mainline, stable, or longterm release (either directly or via backport); if the culprit became known early during a week and is simple to resolve, try to mainline the fix within the same week.h]hXAim to mainline a fix by Sunday after the next, if the culprit made it into a recent mainline, stable, or longterm release (either directly or via backport); if the culprit became known early during a week and is simple to resolve, try to mainline the fix within the same week.}(hj@hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj<ubah}(h]h ]h"]h$]h&]uh1j!hj!ubj")}(hFor other regressions, aim to mainline fixes before the hindmost Sunday within the next three weeks. One or two Sundays later are acceptable, if the regression is something people can live with easily for a while -- like a mild performance regression. h]h)}(hFor other regressions, aim to mainline fixes before the hindmost Sunday within the next three weeks. One or two Sundays later are acceptable, if the regression is something people can live with easily for a while -- like a mild performance regression.h]hFor other regressions, aim to mainline fixes before the hindmost Sunday within the next three weeks. One or two Sundays later are acceptable, if the regression is something people can live with easily for a while -- like a mild performance regression.}(hjXhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjTubah}(h]h ]h"]h$]h&]uh1j!hj!ubj")}(hIt's strongly discouraged to delay mainlining regression fixes till the next merge window, except when the fix is extraordinarily risky or when the culprit was mainlined more than a year ago. h]h)}(hIt's strongly discouraged to delay mainlining regression fixes till the next merge window, except when the fix is extraordinarily risky or when the culprit was mainlined more than a year ago.h]hIt’s strongly discouraged to delay mainlining regression fixes till the next merge window, except when the fix is extraordinarily risky or when the culprit was mainlined more than a year ago.}(hjphhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjlubah}(h]h ]h"]h$]h&]uh1j!hj!ubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhjubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubh)}(h On procedure:h]h On procedure:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hX* Always consider reverting the culprit, as it's often the quickest and least dangerous way to fix a regression. Don't worry about mainlining a fixed variant later: that should be straight-forward, as most of the code went through review once already. * Try to resolve any regressions introduced in mainline during the past twelve months before the current development cycle ends: Linus wants such regressions to be handled like those from the current cycle, unless fixing bears unusual risks. * Consider CCing Linus on discussions or patch review, if a regression seems tangly. Do the same in precarious or urgent cases -- especially if the subsystem maintainer might be unavailable. Also CC the stable team, when you know such a regression made it into a mainline, stable, or longterm release. * For urgent regressions, consider asking Linus to pick up the fix straight from the mailing list: he is totally fine with that for uncontroversial fixes. Ideally though such requests should happen in accordance with the subsystem maintainers or come directly from them. * In case you are unsure if a fix is worth the risk applying just days before a new mainline release, send Linus a mail with the usual lists and people in CC; in it, summarize the situation while asking him to consider picking up the fix straight from the list. He then himself can make the call and when needed even postpone the release. Such requests again should ideally happen in accordance with the subsystem maintainers or come directly from them. h]jx)}(hhh](j")}(hAlways consider reverting the culprit, as it's often the quickest and least dangerous way to fix a regression. Don't worry about mainlining a fixed variant later: that should be straight-forward, as most of the code went through review once already. h]h)}(hAlways consider reverting the culprit, as it's often the quickest and least dangerous way to fix a regression. Don't worry about mainlining a fixed variant later: that should be straight-forward, as most of the code went through review once already.h]hAlways consider reverting the culprit, as it’s often the quickest and least dangerous way to fix a regression. Don’t worry about mainlining a fixed variant later: that should be straight-forward, as most of the code went through review once already.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hTry to resolve any regressions introduced in mainline during the past twelve months before the current development cycle ends: Linus wants such regressions to be handled like those from the current cycle, unless fixing bears unusual risks. h]h)}(hTry to resolve any regressions introduced in mainline during the past twelve months before the current development cycle ends: Linus wants such regressions to be handled like those from the current cycle, unless fixing bears unusual risks.h]hTry to resolve any regressions introduced in mainline during the past twelve months before the current development cycle ends: Linus wants such regressions to be handled like those from the current cycle, unless fixing bears unusual risks.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hX,Consider CCing Linus on discussions or patch review, if a regression seems tangly. Do the same in precarious or urgent cases -- especially if the subsystem maintainer might be unavailable. Also CC the stable team, when you know such a regression made it into a mainline, stable, or longterm release. h]h)}(hX+Consider CCing Linus on discussions or patch review, if a regression seems tangly. Do the same in precarious or urgent cases -- especially if the subsystem maintainer might be unavailable. Also CC the stable team, when you know such a regression made it into a mainline, stable, or longterm release.h]hX+Consider CCing Linus on discussions or patch review, if a regression seems tangly. Do the same in precarious or urgent cases -- especially if the subsystem maintainer might be unavailable. Also CC the stable team, when you know such a regression made it into a mainline, stable, or longterm release.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hX For urgent regressions, consider asking Linus to pick up the fix straight from the mailing list: he is totally fine with that for uncontroversial fixes. Ideally though such requests should happen in accordance with the subsystem maintainers or come directly from them. h]h)}(hX For urgent regressions, consider asking Linus to pick up the fix straight from the mailing list: he is totally fine with that for uncontroversial fixes. Ideally though such requests should happen in accordance with the subsystem maintainers or come directly from them.h]hX For urgent regressions, consider asking Linus to pick up the fix straight from the mailing list: he is totally fine with that for uncontroversial fixes. Ideally though such requests should happen in accordance with the subsystem maintainers or come directly from them.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hXIn case you are unsure if a fix is worth the risk applying just days before a new mainline release, send Linus a mail with the usual lists and people in CC; in it, summarize the situation while asking him to consider picking up the fix straight from the list. He then himself can make the call and when needed even postpone the release. Such requests again should ideally happen in accordance with the subsystem maintainers or come directly from them. h]h)}(hXIn case you are unsure if a fix is worth the risk applying just days before a new mainline release, send Linus a mail with the usual lists and people in CC; in it, summarize the situation while asking him to consider picking up the fix straight from the list. He then himself can make the call and when needed even postpone the release. Such requests again should ideally happen in accordance with the subsystem maintainers or come directly from them.h]hXIn case you are unsure if a fix is worth the risk applying just days before a new mainline release, send Linus a mail with the usual lists and people in CC; in it, summarize the situation while asking him to consider picking up the fix straight from the list. He then himself can make the call and when needed even postpone the release. Such requests again should ideally happen in accordance with the subsystem maintainers or come directly from them.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhjubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubh)}(h&Regarding stable and longterm kernels:h]h&Regarding stable and longterm kernels:}(hj)hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hXf* You are free to leave regressions to the stable team, if they at no point in time occurred with mainline or were fixed there already. * If a regression made it into a proper mainline release during the past twelve months, ensure to tag the fix with "Cc: stable@vger.kernel.org", as a "Fixes:" tag alone does not guarantee a backport. Please add the same tag, in case you know the culprit was backported to stable or longterm kernels. * When receiving reports about regressions in recent stable or longterm kernel series, please evaluate at least briefly if the issue might happen in current mainline as well -- and if that seems likely, take hold of the report. If in doubt, ask the reporter to check mainline. * Whenever you want to swiftly resolve a regression that recently also made it into a proper mainline, stable, or longterm release, fix it quickly in mainline; when appropriate thus involve Linus to fast-track the fix (see above). That's because the stable team normally does neither revert nor fix any changes that cause the same problems in mainline. * In case of urgent regression fixes you might want to ensure prompt backporting by dropping the stable team a note once the fix was mainlined; this is especially advisable during merge windows and shortly thereafter, as the fix otherwise might land at the end of a huge patch queue. h]jx)}(hhh](j")}(hYou are free to leave regressions to the stable team, if they at no point in time occurred with mainline or were fixed there already. h]h)}(hYou are free to leave regressions to the stable team, if they at no point in time occurred with mainline or were fixed there already.h]hYou are free to leave regressions to the stable team, if they at no point in time occurred with mainline or were fixed there already.}(hjBhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhj>ubah}(h]h ]h"]h$]h&]uh1j!hj;ubj")}(hX*If a regression made it into a proper mainline release during the past twelve months, ensure to tag the fix with "Cc: stable@vger.kernel.org", as a "Fixes:" tag alone does not guarantee a backport. Please add the same tag, in case you know the culprit was backported to stable or longterm kernels. h]h)}(hX)If a regression made it into a proper mainline release during the past twelve months, ensure to tag the fix with "Cc: stable@vger.kernel.org", as a "Fixes:" tag alone does not guarantee a backport. Please add the same tag, in case you know the culprit was backported to stable or longterm kernels.h](hxIf a regression made it into a proper mainline release during the past twelve months, ensure to tag the fix with “Cc: }(hjZhhhNhNubj0)}(hstable@vger.kernel.orgh]hstable@vger.kernel.org}(hjbhhhNhNubah}(h]h ]h"]h$]h&]refurimailto:stable@vger.kernel.orguh1j/hjZubh”, as a “Fixes:” tag alone does not guarantee a backport. Please add the same tag, in case you know the culprit was backported to stable or longterm kernels.}(hjZhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhKhjVubah}(h]h ]h"]h$]h&]uh1j!hj;ubj")}(hXWhen receiving reports about regressions in recent stable or longterm kernel series, please evaluate at least briefly if the issue might happen in current mainline as well -- and if that seems likely, take hold of the report. If in doubt, ask the reporter to check mainline. h]h)}(hXWhen receiving reports about regressions in recent stable or longterm kernel series, please evaluate at least briefly if the issue might happen in current mainline as well -- and if that seems likely, take hold of the report. If in doubt, ask the reporter to check mainline.h]hXWhen receiving reports about regressions in recent stable or longterm kernel series, please evaluate at least briefly if the issue might happen in current mainline as well -- and if that seems likely, take hold of the report. If in doubt, ask the reporter to check mainline.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hj;ubj")}(hX_Whenever you want to swiftly resolve a regression that recently also made it into a proper mainline, stable, or longterm release, fix it quickly in mainline; when appropriate thus involve Linus to fast-track the fix (see above). That's because the stable team normally does neither revert nor fix any changes that cause the same problems in mainline. h]h)}(hX^Whenever you want to swiftly resolve a regression that recently also made it into a proper mainline, stable, or longterm release, fix it quickly in mainline; when appropriate thus involve Linus to fast-track the fix (see above). That's because the stable team normally does neither revert nor fix any changes that cause the same problems in mainline.h]hX`Whenever you want to swiftly resolve a regression that recently also made it into a proper mainline, stable, or longterm release, fix it quickly in mainline; when appropriate thus involve Linus to fast-track the fix (see above). That’s because the stable team normally does neither revert nor fix any changes that cause the same problems in mainline.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hj;ubj")}(hXIn case of urgent regression fixes you might want to ensure prompt backporting by dropping the stable team a note once the fix was mainlined; this is especially advisable during merge windows and shortly thereafter, as the fix otherwise might land at the end of a huge patch queue. h]h)}(hXIn case of urgent regression fixes you might want to ensure prompt backporting by dropping the stable team a note once the fix was mainlined; this is especially advisable during merge windows and shortly thereafter, as the fix otherwise might land at the end of a huge patch queue.h]hXIn case of urgent regression fixes you might want to ensure prompt backporting by dropping the stable team a note once the fix was mainlined; this is especially advisable during merge windows and shortly thereafter, as the fix otherwise might land at the end of a huge patch queue.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hj;ubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhj7ubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubh)}(hOn patch flow:h]hOn patch flow:}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjhhubjr)}(hXD* Developers, when trying to reach the time periods mentioned above, remember to account for the time it takes to get fixes tested, reviewed, and merged by Linus, ideally with them being in linux-next at least briefly. Hence, if a fix is urgent, make it obvious to ensure others handle it appropriately. * Reviewers, you are kindly asked to assist developers in reaching the time periods mentioned above by reviewing regression fixes in a timely manner. * Subsystem maintainers, you likewise are encouraged to expedite the handling of regression fixes. Thus evaluate if skipping linux-next is an option for the particular fix. Also consider sending git pull requests more often than usual when needed. And try to avoid holding onto regression fixes over weekends -- especially when the fix is marked for backporting. h]jx)}(hhh](j")}(hX.Developers, when trying to reach the time periods mentioned above, remember to account for the time it takes to get fixes tested, reviewed, and merged by Linus, ideally with them being in linux-next at least briefly. Hence, if a fix is urgent, make it obvious to ensure others handle it appropriately. h]h)}(hX-Developers, when trying to reach the time periods mentioned above, remember to account for the time it takes to get fixes tested, reviewed, and merged by Linus, ideally with them being in linux-next at least briefly. Hence, if a fix is urgent, make it obvious to ensure others handle it appropriately.h]hX-Developers, when trying to reach the time periods mentioned above, remember to account for the time it takes to get fixes tested, reviewed, and merged by Linus, ideally with them being in linux-next at least briefly. Hence, if a fix is urgent, make it obvious to ensure others handle it appropriately.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhKhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hReviewers, you are kindly asked to assist developers in reaching the time periods mentioned above by reviewing regression fixes in a timely manner. h]h)}(hReviewers, you are kindly asked to assist developers in reaching the time periods mentioned above by reviewing regression fixes in a timely manner.h]hReviewers, you are kindly asked to assist developers in reaching the time periods mentioned above by reviewing regression fixes in a timely manner.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1j!hjubj")}(hXjSubsystem maintainers, you likewise are encouraged to expedite the handling of regression fixes. Thus evaluate if skipping linux-next is an option for the particular fix. Also consider sending git pull requests more often than usual when needed. And try to avoid holding onto regression fixes over weekends -- especially when the fix is marked for backporting. h]h)}(hXhSubsystem maintainers, you likewise are encouraged to expedite the handling of regression fixes. Thus evaluate if skipping linux-next is an option for the particular fix. Also consider sending git pull requests more often than usual when needed. And try to avoid holding onto regression fixes over weekends -- especially when the fix is marked for backporting.h]hXhSubsystem maintainers, you likewise are encouraged to expedite the handling of regression fixes. Thus evaluate if skipping linux-next is an option for the particular fix. Also consider sending git pull requests more often than usual when needed. And try to avoid holding onto regression fixes over weekends -- especially when the fix is marked for backporting.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjubah}(h]h ]h"]h$]h&]uh1j!hjubeh}(h]h ]h"]h$]h&]jjuh1jwhhhKhjubah}(h]h ]h"]h$]h&]uh1jqhhhKhjhhubeh}(h]6expectations-and-best-practices-for-fixing-regressionsah ]h"]6expectations and best practices for fixing regressionsah$]h&]uh1hhjhhhhhKubeh}(h]#the-important-basics-in-more-detailah ]h"]#the important basics in more detailah$]h&]uh1hhjhhhhhK=ubh)}(hhh](h)}(h@More aspects regarding regressions developers should be aware ofh]h@More aspects regarding regressions developers should be aware of}(hjRhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjOhhhhhM ubh)}(hhh](h)}(hhj hhubh)}(hXNEarlier attempts to manually track regressions have shown it's an exhausting and frustrating work, which is why they were abandoned after a while. To prevent this from happening again, Thorsten developed regzbot to facilitate the work, with the long term goal to automate regression tracking as much as possible for everyone involved.h]hXPEarlier attempts to manually track regressions have shown it’s an exhausting and frustrating work, which is why they were abandoned after a while. To prevent this from happening again, Thorsten developed regzbot to facilitate the work, with the long term goal to automate regression tracking as much as possible for everyone involved.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMEhj hhubeh}(h]Ewhy-the-linux-kernel-has-a-regression-tracker-and-why-is-regzbot-usedah ]h"]Gwhy the linux kernel has a regression tracker, and why is regzbot used?ah$]h&]uh1hhj hhhhhM<ubh)}(hhh](h)}(h/How does regression tracking work with regzbot?h]h/How does regression tracking work with regzbot?}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMLubh)}(hX7The bot watches for replies to reports of tracked regressions. Additionally, it's looking out for posted or committed patches referencing such reports with "Closes:" tags; replies to such patch postings are tracked as well. Combined this data provides good insights into the current state of the fixing process.h]hX=The bot watches for replies to reports of tracked regressions. Additionally, it’s looking out for posted or committed patches referencing such reports with “Closes:” tags; replies to such patch postings are tracked as well. Combined this data provides good insights into the current state of the fixing process.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMNhj hhubh)}(hXeRegzbot tries to do its job with as little overhead as possible for both reporters and developers. In fact, only reporters are burdened with an extra duty: they need to tell regzbot about the regression report using the ``#regzbot introduced`` command outlined above; if they don't do that, someone else can take care of that using ``#regzbot ^introduced``.h](hRegzbot tries to do its job with as little overhead as possible for both reporters and developers. In fact, only reporters are burdened with an extra duty: they need to tell regzbot about the regression report using the }(hj hhhNhNubj)}(h``#regzbot introduced``h]h#regzbot introduced}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj ubh[ command outlined above; if they don’t do that, someone else can take care of that using }(hj hhhNhNubj)}(h``#regzbot ^introduced``h]h#regzbot ^introduced}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj ubh.}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMThj hhubh)}(hFor developers there normally is no extra work involved, they just need to make sure to do something that was expected long before regzbot came to light: add links to the patch description pointing to all reports about the issue fixed.h]hFor developers there normally is no extra work involved, they just need to make sure to do something that was expected long before regzbot came to light: add links to the patch description pointing to all reports about the issue fixed.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMZhj hhubeh}(h].how-does-regression-tracking-work-with-regzbotah ]h"]/how does regression tracking work with regzbot?ah$]h&]uh1hhj hhhhhMLubh)}(hhh](h)}(hDo I have to use regzbot?h]hDo I have to use regzbot?}(hj6 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj3 hhhhhM_ubh)}(hXbIt's in the interest of everyone if you do, as kernel maintainers like Linus Torvalds partly rely on regzbot's tracking in their work -- for example when deciding to release a new version or extend the development phase. For this they need to be aware of all unfixed regression; to do that, Linus is known to look into the weekly reports sent by regzbot.h]hXfIt’s in the interest of everyone if you do, as kernel maintainers like Linus Torvalds partly rely on regzbot’s tracking in their work -- for example when deciding to release a new version or extend the development phase. For this they need to be aware of all unfixed regression; to do that, Linus is known to look into the weekly reports sent by regzbot.}(hjD hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMahj3 hhubeh}(h]do-i-have-to-use-regzbotah ]h"]do i have to use regzbot?ah$]h&]uh1hhj hhhhhM_ubh)}(hhh](h)}(h@Do I have to tell regzbot about every regression I stumble upon?h]h@Do I have to tell regzbot about every regression I stumble upon?}(hj] hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjZ hhhhhMhubh)}(hXIdeally yes: we are all humans and easily forget problems when something more important unexpectedly comes up -- for example a bigger problem in the Linux kernel or something in real life that's keeping us away from keyboards for a while. Hence, it's best to tell regzbot about every regression, except when you immediately write a fix and commit it to a tree regularly merged to the affected kernel series.h]hXIdeally yes: we are all humans and easily forget problems when something more important unexpectedly comes up -- for example a bigger problem in the Linux kernel or something in real life that’s keeping us away from keyboards for a while. Hence, it’s best to tell regzbot about every regression, except when you immediately write a fix and commit it to a tree regularly merged to the affected kernel series.}(hjk hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMjhjZ hhubeh}(h]?do-i-have-to-tell-regzbot-about-every-regression-i-stumble-uponah ]h"]@do i have to tell regzbot about every regression i stumble upon?ah$]h&]uh1hhj hhhhhMhubh)}(hhh](h)}(h6How to see which regressions regzbot tracks currently?h]h6How to see which regressions regzbot tracks currently?}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMrubh)}(hX~Check `regzbot's web-interface `_ for the latest info; alternatively, `search for the latest regression report `_, which regzbot normally sends out once a week on Sunday evening (UTC), which is a few hours before Linus usually publishes new (pre-)releases.h](hCheck }(hj hhhNhNubj0)}(hM`regzbot's web-interface `_h]hregzbot’s web-interface}(hj hhhNhNubah}(h]h ]h"]h$]h&]nameregzbot's web-interfacejA0https://linux-regtracking.leemhuis.info/regzbot/uh1j/hj ubjD)}(h3 h]h}(h]regzbot-s-web-interfaceah ]h"]regzbot's web-interfaceah$]h&]refurij uh1jCjRKhj ubh% for the latest info; alternatively, }(hj hhhNhNubj0)}(hw`search for the latest regression report `_h]h'search for the latest regression report}(hj hhhNhNubah}(h]h ]h"]h$]h&]name'search for the latest regression reportjAJhttps://lore.kernel.org/lkml/?q=%22Linux+regressions+report%22+f%3Aregzbotuh1j/hj ubjD)}(hM h]h}(h]'search-for-the-latest-regression-reportah ]h"]'search for the latest regression reportah$]h&]refurij uh1jCjRKhj ubh, which regzbot normally sends out once a week on Sunday evening (UTC), which is a few hours before Linus usually publishes new (pre-)releases.}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMthj hhubeh}(h]5how-to-see-which-regressions-regzbot-tracks-currentlyah ]h"]6how to see which regressions regzbot tracks currently?ah$]h&]uh1hhj hhhhhMrubh)}(hhh](h)}(h"What places is regzbot monitoring?h]h"What places is regzbot monitoring?}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhM{ubh)}(hRegzbot is watching the most important Linux mailing lists as well as the git repositories of linux-next, mainline, and stable/longterm.h]hRegzbot is watching the most important Linux mailing lists as well as the git repositories of linux-next, mainline, and stable/longterm.}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhM}hj hhubeh}(h]!what-places-is-regzbot-monitoringah ]h"]"what places is regzbot monitoring?ah$]h&]uh1hhj hhhhhM{ubh)}(hhh](h)}(h:What kind of issues are supposed to be tracked by regzbot?h]h:What kind of issues are supposed to be tracked by regzbot?}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMubh)}(hX!The bot is meant to track regressions, hence please don't involve regzbot for regular issues. But it's okay for the Linux kernel's regression tracker if you use regzbot to track severe issues, like reports about hangs, corrupted data, or internal errors (Panic, Oops, BUG(), warning, ...).h]hX'The bot is meant to track regressions, hence please don’t involve regzbot for regular issues. But it’s okay for the Linux kernel’s regression tracker if you use regzbot to track severe issues, like reports about hangs, corrupted data, or internal errors (Panic, Oops, BUG(), warning, ...).}(hj$ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj hhubeh}(h]9what-kind-of-issues-are-supposed-to-be-tracked-by-regzbotah ]h"]:what kind of issues are supposed to be tracked by regzbot?ah$]h&]uh1hhj hhhhhMubh)}(hhh](h)}(h@Can I add regressions found by CI systems to regzbot's tracking?h]hBCan I add regressions found by CI systems to regzbot’s tracking?}(hj= hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj: hhhhhMubh)}(hFeel free to do so, if the particular regression likely has impact on practical use cases and thus might be noticed by users; hence, please don't involve regzbot for theoretical regressions unlikely to show themselves in real world usage.h]hFeel free to do so, if the particular regression likely has impact on practical use cases and thus might be noticed by users; hence, please don’t involve regzbot for theoretical regressions unlikely to show themselves in real world usage.}(hjK hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj: hhubeh}(h]?can-i-add-regressions-found-by-ci-systems-to-regzbot-s-trackingah ]h"]@can i add regressions found by ci systems to regzbot's tracking?ah$]h&]uh1hhj hhhhhMubh)}(hhh](h)}(hHow to interact with regzbot?h]hHow to interact with regzbot?}(hjd hhhNhNubah}(h]h ]h"]h$]h&]uh1hhja hhhhhMubh)}(hBy using a 'regzbot command' in a direct or indirect reply to the mail with the regression report. These commands need to be in their own paragraph (IOW: they need to be separated from the rest of the mail using blank lines).h]hBy using a ‘regzbot command’ in a direct or indirect reply to the mail with the regression report. These commands need to be in their own paragraph (IOW: they need to be separated from the rest of the mail using blank lines).}(hjr hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhja hhubh)}(hX^One such command is ``#regzbot introduced: ``, which makes regzbot consider your mail as a regressions report added to the tracking, as already described above; ``#regzbot ^introduced: `` is another such command, which makes regzbot consider the parent mail as a report for a regression which it starts to track.h](hOne such command is }(hj hhhNhNubj)}(h,``#regzbot introduced: ``h]h(#regzbot introduced: }(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj ubht, which makes regzbot consider your mail as a regressions report added to the tracking, as already described above; }(hj hhhNhNubj)}(h-``#regzbot ^introduced: ``h]h)#regzbot ^introduced: }(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1jhj ubh} is another such command, which makes regzbot consider the parent mail as a report for a regression which it starts to track.}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhja hhubh)}(hXOnce one of those two commands has been utilized, other regzbot commands can be used in direct or indirect replies to the report. You can write them below one of the `introduced` commands or in replies to the mail that used one of them or itself is a reply to that mail:h](hOnce one of those two commands has been utilized, other regzbot commands can be used in direct or indirect replies to the report. You can write them below one of the }(hj hhhNhNubhtitle_reference)}(h `introduced`h]h introduced}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1j hj ubh\ commands or in replies to the mail that used one of them or itself is a reply to that mail:}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhja hhubjr)}(hX}* Set or update the title:: #regzbot title: foo * Monitor a discussion or bugzilla.kernel.org ticket where additions aspects of the issue or a fix are discussed -- for example the posting of a patch fixing the regression:: #regzbot monitor: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/ Monitoring only works for lore.kernel.org and bugzilla.kernel.org; regzbot will consider all messages in that thread or ticket as related to the fixing process. * Point to a place with further details of interest, like a mailing list post or a ticket in a bug tracker that are slightly related, but about a different topic:: #regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=123456789 * Mark a regression as fixed by a commit that is heading upstream or already landed:: #regzbot fix: 1f2e3d4c5d * Mark a regression as a duplicate of another one already tracked by regzbot:: #regzbot dup-of: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/ * Mark a regression as invalid:: #regzbot invalid: wasn't a regression, problem has always existed h]jx)}(hhh](j")}(h3Set or update the title:: #regzbot title: foo h](h)}(hSet or update the title::h]hSet or update the title:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(h#regzbot title: fooh]h#regzbot title: foo}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj ubj")}(hXMonitor a discussion or bugzilla.kernel.org ticket where additions aspects of the issue or a fix are discussed -- for example the posting of a patch fixing the regression:: #regzbot monitor: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/ Monitoring only works for lore.kernel.org and bugzilla.kernel.org; regzbot will consider all messages in that thread or ticket as related to the fixing process. h](h)}(hMonitor a discussion or bugzilla.kernel.org ticket where additions aspects of the issue or a fix are discussed -- for example the posting of a patch fixing the regression::h]hMonitor a discussion or bugzilla.kernel.org ticket where additions aspects of the issue or a fix are discussed -- for example the posting of a patch fixing the regression:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(hY#regzbot monitor: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/h]hY#regzbot monitor: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubh)}(hMonitoring only works for lore.kernel.org and bugzilla.kernel.org; regzbot will consider all messages in that thread or ticket as related to the fixing process.h]hMonitoring only works for lore.kernel.org and bugzilla.kernel.org; regzbot will consider all messages in that thread or ticket as related to the fixing process.}(hj! hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj ubj")}(hPoint to a place with further details of interest, like a mailing list post or a ticket in a bug tracker that are slightly related, but about a different topic:: #regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=123456789 h](h)}(hPoint to a place with further details of interest, like a mailing list post or a ticket in a bug tracker that are slightly related, but about a different topic::h]hPoint to a place with further details of interest, like a mailing list post or a ticket in a bug tracker that are slightly related, but about a different topic:}(hj9 hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj5 ubj)}(hD#regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=123456789h]hD#regzbot link: https://bugzilla.kernel.org/show_bug.cgi?id=123456789}hjG sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj5 ubeh}(h]h ]h"]h$]h&]uh1j!hj ubj")}(hrMark a regression as fixed by a commit that is heading upstream or already landed:: #regzbot fix: 1f2e3d4c5d h](h)}(hSMark a regression as fixed by a commit that is heading upstream or already landed::h]hRMark a regression as fixed by a commit that is heading upstream or already landed:}(hj_ hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj[ ubj)}(h#regzbot fix: 1f2e3d4c5dh]h#regzbot fix: 1f2e3d4c5d}hjm sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj[ ubeh}(h]h ]h"]h$]h&]uh1j!hj ubj")}(hMark a regression as a duplicate of another one already tracked by regzbot:: #regzbot dup-of: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/ h](h)}(hLMark a regression as a duplicate of another one already tracked by regzbot::h]hKMark a regression as a duplicate of another one already tracked by regzbot:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(hX#regzbot dup-of: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/h]hX#regzbot dup-of: https://lore.kernel.org/all/30th.anniversary.repost@klaava.Helsinki.FI/}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj ubj")}(hfMark a regression as invalid:: #regzbot invalid: wasn't a regression, problem has always existed h](h)}(hMark a regression as invalid::h]hMark a regression as invalid:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(hA#regzbot invalid: wasn't a regression, problem has always existedh]hA#regzbot invalid: wasn't a regression, problem has always existed}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj ubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhj ubah}(h]h ]h"]h$]h&]uh1jqhhhMhja hhubeh}(h]how-to-interact-with-regzbotah ]h"]how to interact with regzbot?ah$]h&]uh1hhj hhhhhMubh)}(hhh](h)}(h5Is there more to tell about regzbot and its commands?h]h5Is there more to tell about regzbot and its commands?}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMubh)}(hXMore detailed and up-to-date information about the Linux kernel's regression tracking bot can be found on its `project page `_, which among others contains a `getting started guide `_ and `reference documentation `_ which both cover more details than the above section.h](hpMore detailed and up-to-date information about the Linux kernel’s regression tracking bot can be found on its }(hj hhhNhNubj0)}(h4`project page `_h]h project page}(hj hhhNhNubah}(h]h ]h"]h$]h&]name project pagejA"https://gitlab.com/knurd42/regzbotuh1j/hj ubjD)}(h% h]h}(h] project-pageah ]h"] project pageah$]h&]refurij uh1jCjRKhj ubh , which among others contains a }(hj hhhNhNubj0)}(ha`getting started guide `_h]hgetting started guide}(hj hhhNhNubah}(h]h ]h"]h$]h&]namegetting started guidejAFhttps://gitlab.com/knurd42/regzbot/-/blob/main/docs/getting_started.mduh1j/hj ubjD)}(hI h]h}(h]getting-started-guideah ]h"]getting started guideah$]h&]refurij, uh1jCjRKhj ubh and }(hj hhhNhNubj0)}(h]`reference documentation `_h]hreference documentation}(hj> hhhNhNubah}(h]h ]h"]h$]h&]namereference documentationjA@https://gitlab.com/knurd42/regzbot/-/blob/main/docs/reference.mduh1j/hj ubjD)}(hC h]h}(h]reference-documentationah ]h"]reference documentationah$]h&]refurijN uh1jCjRKhj ubh6 which both cover more details than the above section.}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj hhubeh}(h]4is-there-more-to-tell-about-regzbot-and-its-commandsah ]h"]5is there more to tell about regzbot and its commands?ah$]h&]uh1hhj hhhhhMubeh}(h]*more-about-regression-tracking-and-regzbotah ]h"]*more about regression tracking and regzbotah$]h&]uh1hhjhhhhhM8ubh)}(hhh](h)}(h"Quotes from Linus about regressionh]h"Quotes from Linus about regression}(hjy hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjv hhhhhMubh)}(hThe following statements from Linus Torvalds provide some insight into Linux "no regressions" rule and how he expects regressions to be handled:h]hThe following statements from Linus Torvalds provide some insight into Linux “no regressions” rule and how he expects regressions to be handled:}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhhhMhjv hhubh)}(hhh](h)}(h*On how quickly regressions should be fixedh]h*On how quickly regressions should be fixed}(hj hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj hhhhhMubjx)}(hhh](j")}(hXFrom `2026-01-22 `_:: But a user complaining should basically result in an immediate fix - possibly a "revert and rethink". With a later clarification on `2026-01-28 `_:: It's also worth noting that "immediate" obviously doesn't mean "right this *second* when the problem has been reported". But if it's a regression with a known commit that caused it, I think the rule of thumb should generally be "within a week", preferably before the next rc. h](h)}(hvFrom `2026-01-22 `_::h](hFrom }(hj hhhNhNubj0)}(ho`2026-01-22 `_h]h 2026-01-22}(hj hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-22jA_https://lore.kernel.org/all/CAHk-=wheQNiW_WtHGO7bKkT7Uib-p+ai2JP9M+z+FYcZ6CAxYA@mail.gmail.com/uh1j/hj ubjD)}(hb h]h}(h]id2ah ]h"] 2026-01-22ah$]h&]refurij uh1jCjRKhj ubh:}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(heBut a user complaining should basically result in an immediate fix - possibly a "revert and rethink".h]heBut a user complaining should basically result in an immediate fix - possibly a "revert and rethink".}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubh)}(hWith a later clarification on `2026-01-28 `_::h](hWith a later clarification on }(hj hhhNhNubj0)}(hu`2026-01-28 `_h]h 2026-01-28}(hj hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-28jAehttps://lore.kernel.org/all/CAHk-%3Dwi86AosXs66-yi54%2BmpQjPu0upxB8ZAfG%2BLsMyJmcuMSA@mail.gmail.com/uh1j/hj ubjD)}(hh h]h}(h]id3ah ]h"] 2026-01-28ah$]h&]refurijuh1jCjRKhj ubh:}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(hXIt's also worth noting that "immediate" obviously doesn't mean "right this *second* when the problem has been reported". But if it's a regression with a known commit that caused it, I think the rule of thumb should generally be "within a week", preferably before the next rc.h]hXIt's also worth noting that "immediate" obviously doesn't mean "right this *second* when the problem has been reported". But if it's a regression with a known commit that caused it, I think the rule of thumb should generally be "within a week", preferably before the next rc.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubj")}(hFrom `2023-04-21 `_:: Known-broken commits either (a) get a timely fix that doesn't have other questions or (b) get reverted h](h)}(hvFrom `2023-04-21 `_::h](hFrom }(hj3hhhNhNubj0)}(ho`2023-04-21 `_h]h 2023-04-21}(hj;hhhNhNubah}(h]h ]h"]h$]h&]name 2023-04-21jA_https://lore.kernel.org/all/CAHk-=wgD98pmSK3ZyHk_d9kZ2bhgN6DuNZMAJaV0WTtbkf=RDw@mail.gmail.com/uh1j/hj3ubjD)}(hb h]h}(h]id4ah ]h"] 2023-04-21ah$]h&]refurijKuh1jCjRKhj3ubh:}(hj3hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj/ubj)}(hhKnown-broken commits either (a) get a timely fix that doesn't have other questions or (b) get revertedh]hhKnown-broken commits either (a) get a timely fix that doesn't have other questions or (b) get reverted}hjcsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj/ubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubj")}(hXFrom `2021-09-20(2) `_:: [...] review shouldn't hold up reported regressions of existing code. That's just basic _testing_ - either the fix should be applied, or - if the fix is too invasive or too ugly - the problematic source of the regression should be reverted. Review should be about new code, it shouldn't be holding up "there's a bug report, here's the obvious fix". h](h)}(hyFrom `2021-09-20(2) `_::h](hFrom }(hj{hhhNhNubj0)}(hr`2021-09-20(2) `_h]h 2021-09-20(2)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2021-09-20(2)jA_https://lore.kernel.org/all/CAHk-=wgOvmtRw1TNbMC1rn5YqyTKyn0hz+sc4k0DGNn++u9aYw@mail.gmail.com/uh1j/hj{ubjD)}(hb h]h}(h]id5ah ]h"] 2021-09-20(2)ah$]h&]refurijuh1jCjRKhj{ubh:}(hj{hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjwubj)}(hX][...] review shouldn't hold up reported regressions of existing code. That's just basic _testing_ - either the fix should be applied, or - if the fix is too invasive or too ugly - the problematic source of the regression should be reverted. Review should be about new code, it shouldn't be holding up "there's a bug report, here's the obvious fix".h]hX][...] review shouldn't hold up reported regressions of existing code. That's just basic _testing_ - either the fix should be applied, or - if the fix is too invasive or too ugly - the problematic source of the regression should be reverted. Review should be about new code, it shouldn't be holding up "there's a bug report, here's the obvious fix".}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjwubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubj")}(hFrom `2023-05-08 `_:: If something doesn't even build, it should damn well be fixed ASAP. h](h)}(hvFrom `2023-05-08 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2023-05-08 `_h]h 2023-05-08}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2023-05-08jA_https://lore.kernel.org/all/CAHk-=wgzU8_dGn0Yg+DyX7ammTkDUCyEJ4C=NvnHRhxKWC7Wpw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id6ah ]h"] 2023-05-08ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hCIf something doesn't even build, it should damn well be fixed ASAP.h]hCIf something doesn't even build, it should damn well be fixed ASAP.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhj hhubeh}(h]*on-how-quickly-regressions-should-be-fixedah ]h"]*on how quickly regressions should be fixedah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(hJOn how fixing regressions with reverts can help prevent maintainer burnouth]hJOn how fixing regressions with reverts can help prevent maintainer burnout}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh]j")}(hXFrom `2026-01-28 `_:: > So how can I/we make "immediate fixes" happen more often without > contributing to maintainer burnout? [...] the "revert and rethink" model [...] often a good idea in general [...] Exactly so that maintainers don't get stressed out over having a pending problem report that people keep pestering them about. I think people are sometimes a bit too bought into whatever changes they made, and reverting is seen as "too drastic", but I think it's often the quick and easy solution for when there isn't some obvious response to a regression report. h](h)}(h|From `2026-01-28 `_::h](hFrom }(hj-hhhNhNubj0)}(hu`2026-01-28 `_h]h 2026-01-28}(hj5hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-28jAehttps://lore.kernel.org/all/CAHk-%3Dwi86AosXs66-yi54%2BmpQjPu0upxB8ZAfG%2BLsMyJmcuMSA@mail.gmail.com/uh1j/hj-ubjD)}(hh h]h}(h]id7ah ]h"]h$] 2026-01-28ah&]refurijEuh1jCjRKhj-ubh:}(hj-hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj)ubj)}(hX%> So how can I/we make "immediate fixes" happen more often without > contributing to maintainer burnout? [...] the "revert and rethink" model [...] often a good idea in general [...] Exactly so that maintainers don't get stressed out over having a pending problem report that people keep pestering them about. I think people are sometimes a bit too bought into whatever changes they made, and reverting is seen as "too drastic", but I think it's often the quick and easy solution for when there isn't some obvious response to a regression report.h]hX%> So how can I/we make "immediate fixes" happen more often without > contributing to maintainer burnout? [...] the "revert and rethink" model [...] often a good idea in general [...] Exactly so that maintainers don't get stressed out over having a pending problem report that people keep pestering them about. I think people are sometimes a bit too bought into whatever changes they made, and reverting is seen as "too drastic", but I think it's often the quick and easy solution for when there isn't some obvious response to a regression report.}hj]sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj)ubeh}(h]h ]h"]h$]h&]uh1j!hj&hhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]Jon-how-fixing-regressions-with-reverts-can-help-prevent-maintainer-burnoutah ]h"]Jon how fixing regressions with reverts can help prevent maintainer burnoutah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h?On mainlining fixes when the last -rc or a new release is closeh]h?On mainlining fixes when the last -rc or a new release is close}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM ubjx)}(hhh](j")}(hX[From `2026-02-01 `_:: So I think I'd rather see them hit rc8 (later today) and have a week of testing in my tree and be reverted if they cause problems, than have them go in after rc8 and then cause problems in the 6.19 release instead. h](h)}(hzFrom `2026-02-01 `_::h](hFrom }(hjhhhNhNubj0)}(hs`2026-02-01 `_h]h 2026-02-01}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2026-02-01jAchttps://lore.kernel.org/all/CAHk-%3DwhXTw1oPsa%2BTLuY1Rc9D1OAiPVOdR_-R2xG45kwDObKdA@mail.gmail.com/uh1j/hjubjD)}(hf h]h}(h]id8ah ]h"] 2026-02-01ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hSo I think I'd rather see them hit rc8 (later today) and have a week of testing in my tree and be reverted if they cause problems, than have them go in after rc8 and then cause problems in the 6.19 release instead.h]hSo I think I'd rather see them hit rc8 (later today) and have a week of testing in my tree and be reverted if they cause problems, than have them go in after rc8 and then cause problems in the 6.19 release instead.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXFrom `2023-04-20 `_:: But something like this, where the regression was in the previous release and it's just a clear fix with no semantic subtlety, I consider to be just a regular regression that should be expedited - partly to make it into stable, and partly to avoid having to put the fix into _another_ stable kernel. h](h)}(hvFrom `2023-04-20 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2023-04-20 `_h]h 2023-04-20}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2023-04-20jA_https://lore.kernel.org/all/CAHk-=wis_qQy4oDNynNKi5b7Qhosmxtoj1jxo5wmB6SRUwQUBQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id9ah ]h"] 2023-04-20ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hX+But something like this, where the regression was in the previous release and it's just a clear fix with no semantic subtlety, I consider to be just a regular regression that should be expedited - partly to make it into stable, and partly to avoid having to put the fix into _another_ stable kernel.h]hX+But something like this, where the regression was in the previous release and it's just a clear fix with no semantic subtlety, I consider to be just a regular regression that should be expedited - partly to make it into stable, and partly to avoid having to put the fix into _another_ stable kernel.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]>on-mainlining-fixes-when-the-last-rc-or-a-new-release-is-closeah ]h"]?on mainlining fixes when the last -rc or a new release is closeah$]h&]uh1hhjv hhhhhM ubh)}(hhh](h)}(h+On sending merge requests with just one fixh]h+On sending merge requests with just one fix}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj1hhhhhMubjx)}(hhh](j")}(hXFrom `2024-04-24 `_:: If the issue is just that there's nothing else happening, I think people should just point me to the patch and say "can you apply this single fix?" h](h)}(hvFrom `2024-04-24 `_::h](hFrom }(hjIhhhNhNubj0)}(ho`2024-04-24 `_h]h 2024-04-24}(hjQhhhNhNubah}(h]h ]h"]h$]h&]name 2024-04-24jA_https://lore.kernel.org/all/CAHk-=wjy_ph9URuFt-pq+2AJ__p7gFDx=yzVSCsx16xAYvNw9g@mail.gmail.com/uh1j/hjIubjD)}(hb h]h}(h]id10ah ]h"] 2024-04-24ah$]h&]refurijauh1jCjRKhjIubh:}(hjIhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjEubj)}(hIf the issue is just that there's nothing else happening, I think people should just point me to the patch and say "can you apply this single fix?"h]hIf the issue is just that there's nothing else happening, I think people should just point me to the patch and say "can you apply this single fix?"}hjysbah}(h]h ]h"]h$]h&]hhuh1jhhhM!hjEubeh}(h]h ]h"]h$]h&]uh1j!hjBhhhhhNubj")}(hXFrom `2023-04-20 `_:: I'm always open to direct fixes when there is no controversy about the fix. No problem. I still happily deal with individual patches. h](h)}(hvFrom `2023-04-20 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2023-04-20 `_h]h 2023-04-20}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2023-04-20jA_https://lore.kernel.org/all/CAHk-=wis_qQy4oDNynNKi5b7Qhosmxtoj1jxo5wmB6SRUwQUBQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id11ah ]h"]h$] 2023-04-20ah&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM$hjubj)}(hI'm always open to direct fixes when there is no controversy about the fix. No problem. I still happily deal with individual patches.h]hI'm always open to direct fixes when there is no controversy about the fix. No problem. I still happily deal with individual patches.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM&hjubeh}(h]h ]h"]h$]h&]uh1j!hjBhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhj1hhubeh}(h]+on-sending-merge-requests-with-just-one-fixah ]h"]+on sending merge requests with just one fixah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(hEOn the importance of pointing to bug reports using Link:/Closes: tagsh]hEOn the importance of pointing to bug reports using Link:/Closes: tags}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM*ubjx)}(hhh](j")}(hX0From `2025-07-29(1) `_:: [...] revert like this, it really would be good to link to the problems, so that when people try to re-enable it, they have the history for why it didn't work the first time. h](h)}(hyFrom `2025-07-29(1) `_::h](hFrom }(hjhhhNhNubj0)}(hr`2025-07-29(1) `_h]h 2025-07-29(1)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2025-07-29(1)jA_https://lore.kernel.org/all/CAHk-=wj2kJRPWx8B09AAtzj+_g+T6UBX11TP0ebs1WJdTtv=WQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id12ah ]h"] 2025-07-29(1)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM,hjubj)}(h[...] revert like this, it really would be good to link to the problems, so that when people try to re-enable it, they have the history for why it didn't work the first time.h]h[...] revert like this, it really would be good to link to the problems, so that when people try to re-enable it, they have the history for why it didn't work the first time.}hj+sbah}(h]h ]h"]h$]h&]hhuh1jhhhM.hjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hFrom `2022-05-08 `_:: So I have to once more complain [...] [...] There's no link to the actual problem the patch fixes. h](h)}(hvFrom `2022-05-08 `_::h](hFrom }(hjChhhNhNubj0)}(ho`2022-05-08 `_h]h 2022-05-08}(hjKhhhNhNubah}(h]h ]h"]h$]h&]name 2022-05-08jA_https://lore.kernel.org/all/CAHk-=wjMmSZzMJ3Xnskdg4+GGz=5p5p+GSYyFBTh0f-DgvdBWg@mail.gmail.com/uh1j/hjCubjD)}(hb h]h}(h]id13ah ]h"] 2022-05-08ah$]h&]refurij[uh1jCjRKhjCubh:}(hjChhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM2hj?ubj)}(hcSo I have to once more complain [...] [...] There's no link to the actual problem the patch fixes.h]hcSo I have to once more complain [...] [...] There's no link to the actual problem the patch fixes.}hjssbah}(h]h ]h"]h$]h&]hhuh1jhhhM4hj?ubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hFrom `2022-06-22 `_:: See, *that* link [to the report] would have been useful in the commit. h](h)}(hvFrom `2022-06-22 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2022-06-22 `_h]h 2022-06-22}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2022-06-22jA_https://lore.kernel.org/all/CAHk-=wjxzafG-=J8oT30s7upn4RhBs6TX-uVFZ5rME+L5_DoJA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id14ah ]h"] 2022-06-22ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM8hjubj)}(hFSee, *that* link [to the report] would have been useful in the commit.h]hFSee, *that* link [to the report] would have been useful in the commit.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM:hjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhM,hjhhubeh}(h]Con-the-importance-of-pointing-to-bug-reports-using-link-closes-tagsah ]h"]Eon the importance of pointing to bug reports using link:/closes: tagsah$]h&]uh1hhjv hhhhhM*ubh)}(hhh](h)}(h'On why the "no regressions" rule existsh]h+On why the “no regressions” rule exists}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM=ubjx)}(hhh](j")}(hX$From `2026-01-22 `_:: But the basic rule is: be so good about backwards compatibility that users never have to worry about upgrading. They should absolutely feel confident that any kernel-reported problem will either be solved, or have an easy solution that is appropriate for *them* (ie a non-technical user shouldn't be expected to be able to do a lot). Because the last thing we want is people holding back from trying new kernels. h](h)}(hvFrom `2026-01-22 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2026-01-22 `_h]h 2026-01-22}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-22jA_https://lore.kernel.org/all/CAHk-=wheQNiW_WtHGO7bKkT7Uib-p+ai2JP9M+z+FYcZ6CAxYA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id15ah ]h"]h$] 2026-01-22ah&]refurij uh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM?hjubj)}(hXBut the basic rule is: be so good about backwards compatibility that users never have to worry about upgrading. They should absolutely feel confident that any kernel-reported problem will either be solved, or have an easy solution that is appropriate for *them* (ie a non-technical user shouldn't be expected to be able to do a lot). Because the last thing we want is people holding back from trying new kernels.h]hXBut the basic rule is: be so good about backwards compatibility that users never have to worry about upgrading. They should absolutely feel confident that any kernel-reported problem will either be solved, or have an easy solution that is appropriate for *them* (ie a non-technical user shouldn't be expected to be able to do a lot). Because the last thing we want is people holding back from trying new kernels.}hj%sbah}(h]h ]h"]h$]h&]hhuh1jhhhMAhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hX@From `2024-05-28 `_:: I introduced that "no regressions" rule something like two decades ago, because people need to be able to update their kernel without fear of something they relied on suddenly stopping to work. h](h)}(hvFrom `2024-05-28 `_::h](hFrom }(hj=hhhNhNubj0)}(ho`2024-05-28 `_h]h 2024-05-28}(hjEhhhNhNubah}(h]h ]h"]h$]h&]name 2024-05-28jA_https://lore.kernel.org/all/CAHk-=wgtb7y-bEh7tPDvDWru7ZKQ8-KMjZ53Tsk37zsPPdwXbA@mail.gmail.com/uh1j/hj=ubjD)}(hb h]h}(h]id16ah ]h"] 2024-05-28ah$]h&]refurijUuh1jCjRKhj=ubh:}(hj=hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMJhj9ubj)}(hI introduced that "no regressions" rule something like two decades ago, because people need to be able to update their kernel without fear of something they relied on suddenly stopping to work.h]hI introduced that "no regressions" rule something like two decades ago, because people need to be able to update their kernel without fear of something they relied on suddenly stopping to work.}hjmsbah}(h]h ]h"]h$]h&]hhuh1jhhhMLhj9ubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hX+From `2018-08-03 `_:: The whole point of "we do not regress" is so that people can upgrade the kernel and never have to worry about it. [...] Because the only thing that matters IS THE USER. h](h)}(hvFrom `2018-08-03 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2018-08-03 `_h]h 2018-08-03}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2018-08-03jA_https://lore.kernel.org/all/CA+55aFwWZX=CXmWDTkDGb36kf12XmTehmQjbiMPCqCRG2hi9kw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id17ah ]h"] 2018-08-03ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMPhjubj)}(hThe whole point of "we do not regress" is so that people can upgrade the kernel and never have to worry about it. [...] Because the only thing that matters IS THE USER.h]hThe whole point of "we do not regress" is so that people can upgrade the kernel and never have to worry about it. [...] Because the only thing that matters IS THE USER.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMRhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXFrom `2017-10-26(1) `_:: If the kernel used to work for you, the rule is that it continues to work for you. [...] People should basically always feel like they can update their kernel and simply not have to worry about it. I refuse to introduce "you can only update the kernel if you also update that other program" kind of limitations. If the kernel used to work for you, the rule is that it continues to work for you. h](h)}(hzFrom `2017-10-26(1) `_::h](hFrom }(hjhhhNhNubj0)}(hs`2017-10-26(1) `_h]h 2017-10-26(1)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(1)jA`https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/uh1j/hjubjD)}(hc h]h}(h]id18ah ]h"] 2017-10-26(1)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMYhjubj)}(hXIf the kernel used to work for you, the rule is that it continues to work for you. [...] People should basically always feel like they can update their kernel and simply not have to worry about it. I refuse to introduce "you can only update the kernel if you also update that other program" kind of limitations. If the kernel used to work for you, the rule is that it continues to work for you.h]hXIf the kernel used to work for you, the rule is that it continues to work for you. [...] People should basically always feel like they can update their kernel and simply not have to worry about it. I refuse to introduce "you can only update the kernel if you also update that other program" kind of limitations. If the kernel used to work for you, the rule is that it continues to work for you.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM[hjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhM?hjhhubeh}(h]%on-why-the-no-regressions-rule-existsah ]h"]'on why the "no regressions" rule existsah$]h&]uh1hhjv hhhhhM=ubh)}(hhh](h)}(h*On exceptions to the "no regressions" ruleh]h.On exceptions to the “no regressions” rule}(hj"hhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMhubjx)}(hhh](j")}(hX+From `2026-01-22 `_:: There are _very_ few exceptions to that rule, the main one being "the problem was a fundamental huge and gaping security issue and we *had* to make that change, and we couldn't even make your limited use-case just continue to work". The other exception is "the problem was reported years after it was introduced, and now most people rely on the new behavior". [...] Now, if it's one or two users and you can just get them to recompile, that's one thing. Niche hardware and odd use-cases can sometimes be solved that way, and regressions can sometimes be fixed by handholding every single reporter if the reporter is willing and able to change his or her workflow. h](h)}(hvFrom `2026-01-22 `_::h](hFrom }(hj7hhhNhNubj0)}(ho`2026-01-22 `_h]h 2026-01-22}(hj?hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-22jA_https://lore.kernel.org/all/CAHk-=wheQNiW_WtHGO7bKkT7Uib-p+ai2JP9M+z+FYcZ6CAxYA@mail.gmail.com/uh1j/hj7ubjD)}(hb h]h}(h]id19ah ]h"]h$] 2026-01-22ah&]refurijOuh1jCjRKhj7ubh:}(hj7hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMjhj3ubj)}(hXThere are _very_ few exceptions to that rule, the main one being "the problem was a fundamental huge and gaping security issue and we *had* to make that change, and we couldn't even make your limited use-case just continue to work". The other exception is "the problem was reported years after it was introduced, and now most people rely on the new behavior". [...] Now, if it's one or two users and you can just get them to recompile, that's one thing. Niche hardware and odd use-cases can sometimes be solved that way, and regressions can sometimes be fixed by handholding every single reporter if the reporter is willing and able to change his or her workflow.h]hXThere are _very_ few exceptions to that rule, the main one being "the problem was a fundamental huge and gaping security issue and we *had* to make that change, and we couldn't even make your limited use-case just continue to work". The other exception is "the problem was reported years after it was introduced, and now most people rely on the new behavior". [...] Now, if it's one or two users and you can just get them to recompile, that's one thing. Niche hardware and odd use-cases can sometimes be solved that way, and regressions can sometimes be fixed by handholding every single reporter if the reporter is willing and able to change his or her workflow.}hjgsbah}(h]h ]h"]h$]h&]hhuh1jhhhMlhj3ubeh}(h]h ]h"]h$]h&]uh1j!hj0hhhhhNubj")}(hX3From `2023-04-20 `_:: And yes, I do consider "regression in an earlier release" to be a regression that needs fixing. There's obviously a time limit: if that "regression in an earlier release" was a year or more ago, and just took forever for people to notice, and it had semantic changes that now mean that fixing the regression could cause a _new_ regression, then that can cause me to go "Oh, now the new semantics are what we have to live with". h](h)}(hvFrom `2023-04-20 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2023-04-20 `_h]h 2023-04-20}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2023-04-20jA_https://lore.kernel.org/all/CAHk-=wis_qQy4oDNynNKi5b7Qhosmxtoj1jxo5wmB6SRUwQUBQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id20ah ]h"]h$] 2023-04-20ah&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM|hj{ubj)}(hXAnd yes, I do consider "regression in an earlier release" to be a regression that needs fixing. There's obviously a time limit: if that "regression in an earlier release" was a year or more ago, and just took forever for people to notice, and it had semantic changes that now mean that fixing the regression could cause a _new_ regression, then that can cause me to go "Oh, now the new semantics are what we have to live with".h]hXAnd yes, I do consider "regression in an earlier release" to be a regression that needs fixing. There's obviously a time limit: if that "regression in an earlier release" was a year or more ago, and just took forever for people to notice, and it had semantic changes that now mean that fixing the regression could cause a _new_ regression, then that can cause me to go "Oh, now the new semantics are what we have to live with".}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM~hj{ubeh}(h]h ]h"]h$]h&]uh1j!hj0hhhhhNubj")}(hXFrom `2017-10-26(2) `_:: There have been exceptions, but they are few and far between, and they generally have some major and fundamental reasons for having happened, that were basically entirely unavoidable, and people _tried_hard_ to avoid them. Maybe we can't practically support the hardware any more after it is decades old and nobody uses it with modern kernels any more. Maybe there's a serious security issue with how we did things, and people actually depended on that fundamentally broken model. Maybe there was some fundamental other breakage that just _had_ to have a flag day for very core and fundamental reasons. h](h)}(hzFrom `2017-10-26(2) `_::h](hFrom }(hjhhhNhNubj0)}(hs`2017-10-26(2) `_h]h 2017-10-26(2)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(2)jA`https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/uh1j/hjubjD)}(hc h]h}(h]id21ah ]h"] 2017-10-26(2)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hXZThere have been exceptions, but they are few and far between, and they generally have some major and fundamental reasons for having happened, that were basically entirely unavoidable, and people _tried_hard_ to avoid them. Maybe we can't practically support the hardware any more after it is decades old and nobody uses it with modern kernels any more. Maybe there's a serious security issue with how we did things, and people actually depended on that fundamentally broken model. Maybe there was some fundamental other breakage that just _had_ to have a flag day for very core and fundamental reasons.h]hXZThere have been exceptions, but they are few and far between, and they generally have some major and fundamental reasons for having happened, that were basically entirely unavoidable, and people _tried_hard_ to avoid them. Maybe we can't practically support the hardware any more after it is decades old and nobody uses it with modern kernels any more. Maybe there's a serious security issue with how we did things, and people actually depended on that fundamentally broken model. Maybe there was some fundamental other breakage that just _had_ to have a flag day for very core and fundamental reasons.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj0hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMjhjhhubeh}(h](on-exceptions-to-the-no-regressions-ruleah ]h"]*on exceptions to the "no regressions" ruleah$]h&]uh1hhjv hhhhhMhubh)}(hhh](h)}(hKOn situations where updating something in userspace can resolve regressionsh]hKOn situations where updating something in userspace can resolve regressions}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh](j")}(hXFrom `2018-08-03 `_:: And dammit, we upgrade the kernel ALL THE TIME without upgrading any other programs at all. It is absolutely required, because flag-days and dependencies are horribly bad. And it is also required simply because I as a kernel developer do not upgrade random other tools that I don't even care about as I develop the kernel, and I want any of my users to feel safe doing the same time. h](h)}(hvFrom `2018-08-03 `_::h](hFrom }(hj1hhhNhNubj0)}(ho`2018-08-03 `_h]h 2018-08-03}(hj9hhhNhNubah}(h]h ]h"]h$]h&]name 2018-08-03jA_https://lore.kernel.org/all/CA+55aFwWZX=CXmWDTkDGb36kf12XmTehmQjbiMPCqCRG2hi9kw@mail.gmail.com/uh1j/hj1ubjD)}(hb h]h}(h]id22ah ]h"]h$] 2018-08-03ah&]refurijIuh1jCjRKhj1ubh:}(hj1hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj-ubj)}(hXAnd dammit, we upgrade the kernel ALL THE TIME without upgrading any other programs at all. It is absolutely required, because flag-days and dependencies are horribly bad. And it is also required simply because I as a kernel developer do not upgrade random other tools that I don't even care about as I develop the kernel, and I want any of my users to feel safe doing the same time.h]hXAnd dammit, we upgrade the kernel ALL THE TIME without upgrading any other programs at all. It is absolutely required, because flag-days and dependencies are horribly bad. And it is also required simply because I as a kernel developer do not upgrade random other tools that I don't even care about as I develop the kernel, and I want any of my users to feel safe doing the same time.}hjasbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj-ubeh}(h]h ]h"]h$]h&]uh1j!hj*hhhhhNubj")}(hXFrom `2017-10-26(3) `_:: But if something actually breaks, then the change must get fixed or reverted. And it gets fixed in the *kernel*. Not by saying "well, fix your user space then". It was a kernel change that exposed the problem, it needs to be the kernel that corrects for it, because we have a "upgrade in place" model. We don't have a "upgrade with new user space". And I seriously will refuse to take code from people who do not understand and honor this very simple rule. This rule is also not going to change. And yes, I realize that the kernel is "special" in this respect. I'm proud of it. h](h)}(hzFrom `2017-10-26(3) `_::h](hFrom }(hjyhhhNhNubj0)}(hs`2017-10-26(3) `_h]h 2017-10-26(3)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(3)jA`https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/uh1j/hjyubjD)}(hc h]h}(h]id23ah ]h"] 2017-10-26(3)ah$]h&]refurijuh1jCjRKhjyubh:}(hjyhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjuubj)}(hXDBut if something actually breaks, then the change must get fixed or reverted. And it gets fixed in the *kernel*. Not by saying "well, fix your user space then". It was a kernel change that exposed the problem, it needs to be the kernel that corrects for it, because we have a "upgrade in place" model. We don't have a "upgrade with new user space". And I seriously will refuse to take code from people who do not understand and honor this very simple rule. This rule is also not going to change. And yes, I realize that the kernel is "special" in this respect. I'm proud of it.h]hXDBut if something actually breaks, then the change must get fixed or reverted. And it gets fixed in the *kernel*. Not by saying "well, fix your user space then". It was a kernel change that exposed the problem, it needs to be the kernel that corrects for it, because we have a "upgrade in place" model. We don't have a "upgrade with new user space". And I seriously will refuse to take code from people who do not understand and honor this very simple rule. This rule is also not going to change. And yes, I realize that the kernel is "special" in this respect. I'm proud of it.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjuubeh}(h]h ]h"]h$]h&]uh1j!hj*hhhhhNubj")}(hX From `2017-10-26(4) `_:: If you break existing user space setups THAT IS A REGRESSION. It's not ok to say "but we'll fix the user space setup". Really. NOT OK. h](h)}(hyFrom `2017-10-26(4) `_::h](hFrom }(hjhhhNhNubj0)}(hr`2017-10-26(4) `_h]h 2017-10-26(4)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(4)jA_https://lore.kernel.org/all/CA+55aFwiiQYJ+YoLKCXjN_beDVfu38mg=Ggg5LFOcqHE8Qi7Zw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id24ah ]h"] 2017-10-26(4)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hIf you break existing user space setups THAT IS A REGRESSION. It's not ok to say "but we'll fix the user space setup". Really. NOT OK.h]hIf you break existing user space setups THAT IS A REGRESSION. It's not ok to say "but we'll fix the user space setup". Really. NOT OK.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj*hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]Kon-situations-where-updating-something-in-userspace-can-resolve-regressionsah ]h"]Kon situations where updating something in userspace can resolve regressionsah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(hOOn what qualifies as userspace interface, ABI, API, documented interfaces, etc.h]hOOn what qualifies as userspace interface, ABI, API, documented interfaces, etc.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh](j")}(hXFrom `2026-01-20 `_:: So I absolutely detest the whole notion of "ABI changes". It's a meaningless concept, and I hate it with a passion, [...] The Linux rule for regressions is basically based on the philosophical question of "If a tree falls in the forest, and nobody is around to hear it, does it make a sound?". So the only thing that matters is if something breaks user-*conscious* behavior. And when that happens, the distinction between "bug fix" and "new feature" and "ABI change" matters not one whit, and the change needs to be done differently. [...] I just wanted to point out that the argument about whether it's an ABI change or not is irrelevant. If it turns out that some program - not a test script, but something with relevance to conscious user expectations ~ depended on the old broken behavior, then it needs to be done some other way. h](h)}(hvFrom `2026-01-20 `_::h](hFrom }(hj+hhhNhNubj0)}(ho`2026-01-20 `_h]h 2026-01-20}(hj3hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-20jA_https://lore.kernel.org/all/CAHk-=wga8Qu0-OSE9VZbviq9GuqwhPhLUXeAt-S7_9+fMCLkKg@mail.gmail.com/uh1j/hj+ubjD)}(hb h]h}(h]id25ah ]h"] 2026-01-20ah$]h&]refurijCuh1jCjRKhj+ubh:}(hj+hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj'ubj)}(hXGSo I absolutely detest the whole notion of "ABI changes". It's a meaningless concept, and I hate it with a passion, [...] The Linux rule for regressions is basically based on the philosophical question of "If a tree falls in the forest, and nobody is around to hear it, does it make a sound?". So the only thing that matters is if something breaks user-*conscious* behavior. And when that happens, the distinction between "bug fix" and "new feature" and "ABI change" matters not one whit, and the change needs to be done differently. [...] I just wanted to point out that the argument about whether it's an ABI change or not is irrelevant. If it turns out that some program - not a test script, but something with relevance to conscious user expectations ~ depended on the old broken behavior, then it needs to be done some other way.h]hXGSo I absolutely detest the whole notion of "ABI changes". It's a meaningless concept, and I hate it with a passion, [...] The Linux rule for regressions is basically based on the philosophical question of "If a tree falls in the forest, and nobody is around to hear it, does it make a sound?". So the only thing that matters is if something breaks user-*conscious* behavior. And when that happens, the distinction between "bug fix" and "new feature" and "ABI change" matters not one whit, and the change needs to be done differently. [...] I just wanted to point out that the argument about whether it's an ABI change or not is irrelevant. If it turns out that some program - not a test script, but something with relevance to conscious user expectations ~ depended on the old broken behavior, then it needs to be done some other way.}hj[sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj'ubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXFrom `2026-02-13 `_:: > [...] this should not fall under the don't break user space rule [...] Note that the rule is about breaking *users*, not breaking user space per se. [...] If some user setup breaks, things need fixing. [...] but I want to make it very clear that there are no excuses about "user space applications". h](h)}(hvFrom `2026-02-13 `_::h](hFrom }(hjshhhNhNubj0)}(ho`2026-02-13 `_h]h 2026-02-13}(hj{hhhNhNubah}(h]h ]h"]h$]h&]name 2026-02-13jA_https://lore.kernel.org/all/CAHk-=whY-N8kjm8kiFUV5Ei-8AuYw--EPGD-AR3Pd+5GTx2sAQ@mail.gmail.com/uh1j/hjsubjD)}(hb h]h}(h]id26ah ]h"] 2026-02-13ah$]h&]refurijuh1jCjRKhjsubh:}(hjshhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjoubj)}(hX0> [...] this should not fall under the don't break user space rule [...] Note that the rule is about breaking *users*, not breaking user space per se. [...] If some user setup breaks, things need fixing. [...] but I want to make it very clear that there are no excuses about "user space applications".h]hX0> [...] this should not fall under the don't break user space rule [...] Note that the rule is about breaking *users*, not breaking user space per se. [...] If some user setup breaks, things need fixing. [...] but I want to make it very clear that there are no excuses about "user space applications".}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjoubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXDFrom `2021-09-20(4) `_:: [...] a regression is a bit like Schrödinger's cat - if nobody is around to notice it and it doesn't actually affect any real workload, then you can treat the regression as if it doesn't exist. h](h)}(hyFrom `2021-09-20(4) `_::h](hFrom }(hjhhhNhNubj0)}(hr`2021-09-20(4) `_h]h 2021-09-20(4)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2021-09-20(4)jA_https://lore.kernel.org/all/CAHk-=wi7DB2SJ-wngVvsJ7Ak2cM556Q8437sOXo4EJt2BWPdEg@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id27ah ]h"] 2021-09-20(4)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(h[...] a regression is a bit like Schrödinger's cat - if nobody is around to notice it and it doesn't actually affect any real workload, then you can treat the regression as if it doesn't exist.h]h[...] a regression is a bit like Schrödinger's cat - if nobody is around to notice it and it doesn't actually affect any real workload, then you can treat the regression as if it doesn't exist.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXfFrom `2020-05-21 `_:: The rules about regressions have never been about any kind of documented behavior, or where the code lives. The rules about regressions are always about "breaks user workflow". Users are literally the _only_ thing that matters. h](h)}(hvFrom `2020-05-21 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2020-05-21 `_h]h 2020-05-21}(hj hhhNhNubah}(h]h ]h"]h$]h&]name 2020-05-21jA_https://lore.kernel.org/all/CAHk-=wiVi7mSrsMP=fLXQrXK_UimybW=ziLOwSzFTtoXUacWVQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id28ah ]h"] 2020-05-21ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hThe rules about regressions have never been about any kind of documented behavior, or where the code lives. The rules about regressions are always about "breaks user workflow". Users are literally the _only_ thing that matters.h]hThe rules about regressions have never been about any kind of documented behavior, or where the code lives. The rules about regressions are always about "breaks user workflow". Users are literally the _only_ thing that matters.}hj3sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXFFrom `2019-09-15 `_:: One _particularly_ last-minute revert is the top-most commit (ignoring the version change itself) done just before the release, and while it's very annoying, it's perhaps also instructive. What's instructive about it is that I reverted a commit that wasn't actually buggy. In fact, it was doing exactly what it set out to do, and did it very well. In fact it did it _so_ well that the much improved IO patterns it caused then ended up revealing a user-visible regression due to a real bug in a completely unrelated area. The actual details of that regression are not the reason I point that revert out as instructive, though. It's more that it's an instructive example of what counts as a regression, and what the whole "no regressions" kernel rule means. [...] The reverted commit didn't change any API's, and it didn't introduce any new bugs. But it ended up exposing another problem, and as such caused a kernel upgrade to fail for a user. So it got reverted. The point here being that we revert based on user-reported _behavior_, not based on some "it changes the ABI" or "it caused a bug" concept. The problem was really pre-existing, and it just didn't happen to trigger before. [...] Take-away from the whole thing: it's not about whether you change the kernel-userspace ABI, or fix a bug, or about whether the old code "should never have worked in the first place". It's about whether something breaks existing users' workflow. h](h)}(hwFrom `2019-09-15 `_::h](hFrom }(hjKhhhNhNubj0)}(hp`2019-09-15 `_h]h 2019-09-15}(hjShhhNhNubah}(h]h ]h"]h$]h&]name 2019-09-15jA`https://lore.kernel.org/lkml/CAHk-=wiP4K8DRJWsCo=20hn_6054xBamGKF2kPgUzpB5aMaofA@mail.gmail.com/uh1j/hjKubjD)}(hc h]h}(h]id29ah ]h"] 2019-09-15ah$]h&]refurijcuh1jCjRKhjKubh:}(hjKhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjGubj)}(hXOne _particularly_ last-minute revert is the top-most commit (ignoring the version change itself) done just before the release, and while it's very annoying, it's perhaps also instructive. What's instructive about it is that I reverted a commit that wasn't actually buggy. In fact, it was doing exactly what it set out to do, and did it very well. In fact it did it _so_ well that the much improved IO patterns it caused then ended up revealing a user-visible regression due to a real bug in a completely unrelated area. The actual details of that regression are not the reason I point that revert out as instructive, though. It's more that it's an instructive example of what counts as a regression, and what the whole "no regressions" kernel rule means. [...] The reverted commit didn't change any API's, and it didn't introduce any new bugs. But it ended up exposing another problem, and as such caused a kernel upgrade to fail for a user. So it got reverted. The point here being that we revert based on user-reported _behavior_, not based on some "it changes the ABI" or "it caused a bug" concept. The problem was really pre-existing, and it just didn't happen to trigger before. [...] Take-away from the whole thing: it's not about whether you change the kernel-userspace ABI, or fix a bug, or about whether the old code "should never have worked in the first place". It's about whether something breaks existing users' workflow.h]hXOne _particularly_ last-minute revert is the top-most commit (ignoring the version change itself) done just before the release, and while it's very annoying, it's perhaps also instructive. What's instructive about it is that I reverted a commit that wasn't actually buggy. In fact, it was doing exactly what it set out to do, and did it very well. In fact it did it _so_ well that the much improved IO patterns it caused then ended up revealing a user-visible regression due to a real bug in a completely unrelated area. The actual details of that regression are not the reason I point that revert out as instructive, though. It's more that it's an instructive example of what counts as a regression, and what the whole "no regressions" kernel rule means. [...] The reverted commit didn't change any API's, and it didn't introduce any new bugs. But it ended up exposing another problem, and as such caused a kernel upgrade to fail for a user. So it got reverted. The point here being that we revert based on user-reported _behavior_, not based on some "it changes the ABI" or "it caused a bug" concept. The problem was really pre-existing, and it just didn't happen to trigger before. [...] Take-away from the whole thing: it's not about whether you change the kernel-userspace ABI, or fix a bug, or about whether the old code "should never have worked in the first place". It's about whether something breaks existing users' workflow.}hj{sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjGubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hFrom `2017-11-05 `_:: And our regression rule has never been "behavior doesn't change". That would mean that we could never make any changes at all. h](h)}(hvFrom `2017-11-05 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2017-11-05 `_h]h 2017-11-05}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-11-05jA_https://lore.kernel.org/all/CA+55aFzUvbGjD8nQ-+3oiMBx14c_6zOj2n7KLN3UsJ-qsd4Dcw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id30ah ]h"] 2017-11-05ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM hjubj)}(h~And our regression rule has never been "behavior doesn't change". That would mean that we could never make any changes at all.h]h~And our regression rule has never been "behavior doesn't change". That would mean that we could never make any changes at all.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hX9From `2020-05-21 `_:: No amount of "you shouldn't have used this" or "that behavior was undefined, it's your own fault your app broke" or "that used to work simply because of a kernel bug" is at all relevant. h](h)}(hvFrom `2020-05-21 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2020-05-21 `_h]h 2020-05-21}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2020-05-21jA_https://lore.kernel.org/all/CAHk-=wiVi7mSrsMP=fLXQrXK_UimybW=ziLOwSzFTtoXUacWVQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id31ah ]h"]h$] 2020-05-21ah&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hNo amount of "you shouldn't have used this" or "that behavior was undefined, it's your own fault your app broke" or "that used to work simply because of a kernel bug" is at all relevant.h]hNo amount of "you shouldn't have used this" or "that behavior was undefined, it's your own fault your app broke" or "that used to work simply because of a kernel bug" is at all relevant.}hj sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXFrom `2021-05-21 `_:: But no, "that was documented to be broken" (whether it's because the code was in staging or because the man-page said something else) is irrelevant. If staging code is so useful that people end up using it, that means that it's basically regular kernel code with a flag saying "please clean this up". [...] The other side of the coin is that people who talk about "API stability" are entirely wrong. API's don't matter either. You can make any changes to an API you like - as long as nobody notices. Again, the regression rule is not about documentation, not about API's, and not about the phase of the moon. h](h)}(hvFrom `2021-05-21 `_::h](hFrom }(hj#hhhNhNubj0)}(ho`2021-05-21 `_h]h 2021-05-21}(hj+hhhNhNubah}(h]h ]h"]h$]h&]name 2021-05-21jA_https://lore.kernel.org/all/CAHk-=wiVi7mSrsMP=fLXQrXK_UimybW=ziLOwSzFTtoXUacWVQ@mail.gmail.com/uh1j/hj#ubjD)}(hb h]h}(h]id32ah ]h"] 2021-05-21ah$]h&]refurij;uh1jCjRKhj#ubh:}(hj#hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hXcBut no, "that was documented to be broken" (whether it's because the code was in staging or because the man-page said something else) is irrelevant. If staging code is so useful that people end up using it, that means that it's basically regular kernel code with a flag saying "please clean this up". [...] The other side of the coin is that people who talk about "API stability" are entirely wrong. API's don't matter either. You can make any changes to an API you like - as long as nobody notices. Again, the regression rule is not about documentation, not about API's, and not about the phase of the moon.h]hXcBut no, "that was documented to be broken" (whether it's because the code was in staging or because the man-page said something else) is irrelevant. If staging code is so useful that people end up using it, that means that it's basically regular kernel code with a flag saying "please clean this up". [...] The other side of the coin is that people who talk about "API stability" are entirely wrong. API's don't matter either. You can make any changes to an API you like - as long as nobody notices. Again, the regression rule is not about documentation, not about API's, and not about the phase of the moon.}hjSsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hXJFrom `2012-07-06 `_:: > Now this got me wondering if Debian _unstable_ actually qualifies as a > standard distro userspace. Oh, if the kernel breaks some standard user space, that counts. Tons of people run Debian unstable h](h)}(hvFrom `2012-07-06 `_::h](hFrom }(hjkhhhNhNubj0)}(ho`2012-07-06 `_h]h 2012-07-06}(hjshhhNhNubah}(h]h ]h"]h$]h&]name 2012-07-06jA_https://lore.kernel.org/all/CA+55aFwnLJ+0sjx92EGREGTWOx84wwKaraSzpTNJwPVV8edw8g@mail.gmail.com/uh1j/hjkubjD)}(hb h]h}(h]id33ah ]h"] 2012-07-06ah$]h&]refurijuh1jCjRKhjkubh:}(hjkhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM(hjgubj)}(h> Now this got me wondering if Debian _unstable_ actually qualifies as a > standard distro userspace. Oh, if the kernel breaks some standard user space, that counts. Tons of people run Debian unstableh]h> Now this got me wondering if Debian _unstable_ actually qualifies as a > standard distro userspace. Oh, if the kernel breaks some standard user space, that counts. Tons of people run Debian unstable}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM*hjgubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubj")}(hFrom `2011-05-06 `_:: It's clearly NOT an internal tracepoint. By definition. It's being used by powertop. h](h)}(hfFrom `2011-05-06 `_::h](hFrom }(hjhhhNhNubj0)}(h_`2011-05-06 `_h]h 2011-05-06}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2011-05-06jAOhttps://lore.kernel.org/all/BANLkTi=KVXjKR82sqsz4gwjr+E0vtqCmvA@mail.gmail.com/uh1j/hjubjD)}(hR h]h}(h]id34ah ]h"] 2011-05-06ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM0hjubj)}(hTIt's clearly NOT an internal tracepoint. By definition. It's being used by powertop.h]hTIt's clearly NOT an internal tracepoint. By definition. It's being used by powertop.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhM2hjubeh}(h]h ]h"]h$]h&]uh1j!hj$hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]Jon-what-qualifies-as-userspace-interface-abi-api-documented-interfaces-etcah ]h"]Oon what qualifies as userspace interface, abi, api, documented interfaces, etc.ah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h2On regressions noticed by users or test-suites/CIsh]h2On regressions noticed by users or test-suites/CIs}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhM6ubjx)}(hhh](j")}(hXFrom `2026-01-22 `_:: Users complaining is the only real line in the end. [...] a test-suite complaining is then often a *very* good indication that maybe users will hit some problem, and test suite issues should be taken very seriously [...] But a test-suite error isn't necessarily where you have to draw the line - it's a big red flag [...] h](h)}(hvFrom `2026-01-22 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2026-01-22 `_h]h 2026-01-22}(hj%hhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-22jA_https://lore.kernel.org/all/CAHk-=wheQNiW_WtHGO7bKkT7Uib-p+ai2JP9M+z+FYcZ6CAxYA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id35ah ]h"]h$] 2026-01-22ah&]refurij5uh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM8hjubj)}(hXCUsers complaining is the only real line in the end. [...] a test-suite complaining is then often a *very* good indication that maybe users will hit some problem, and test suite issues should be taken very seriously [...] But a test-suite error isn't necessarily where you have to draw the line - it's a big red flag [...]h]hXCUsers complaining is the only real line in the end. [...] a test-suite complaining is then often a *very* good indication that maybe users will hit some problem, and test suite issues should be taken very seriously [...] But a test-suite error isn't necessarily where you have to draw the line - it's a big red flag [...]}hjMsbah}(h]h ]h"]h$]h&]hhuh1jhhhM:hjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXoFrom `2024-29-01 `_:: The "no regressions" rule is not about made-up "if I do this, behavior changes". The "no regressions" rule is about *users*. If you have an actual user that has been doing insane things, and we change something, and now the insane thing no longer works, at that point it's a regression, and we'll sigh, and go "Users are insane" and have to fix it. But if you have some random test that now behaves differently, it's not a regression. It's a *warning* sign, sure: tests are useful. h](h)}(hvFrom `2024-29-01 `_::h](hFrom }(hjehhhNhNubj0)}(ho`2024-29-01 `_h]h 2024-29-01}(hjmhhhNhNubah}(h]h ]h"]h$]h&]name 2024-29-01jA_https://lore.kernel.org/all/CAHk-=wg8BrZEzjJ5kUyZzHPZmFqH6ooMN1gRBCofxxCfucgjaw@mail.gmail.com/uh1j/hjeubjD)}(hb h]h}(h]id36ah ]h"] 2024-29-01ah$]h&]refurij}uh1jCjRKhjeubh:}(hjehhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMChjaubj)}(hXThe "no regressions" rule is not about made-up "if I do this, behavior changes". The "no regressions" rule is about *users*. If you have an actual user that has been doing insane things, and we change something, and now the insane thing no longer works, at that point it's a regression, and we'll sigh, and go "Users are insane" and have to fix it. But if you have some random test that now behaves differently, it's not a regression. It's a *warning* sign, sure: tests are useful.h]hXThe "no regressions" rule is not about made-up "if I do this, behavior changes". The "no regressions" rule is about *users*. If you have an actual user that has been doing insane things, and we change something, and now the insane thing no longer works, at that point it's a regression, and we'll sigh, and go "Users are insane" and have to fix it. But if you have some random test that now behaves differently, it's not a regression. It's a *warning* sign, sure: tests are useful.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMEhjaubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhM8hjhhubeh}(h]2on-regressions-noticed-by-users-or-test-suites-cisah ]h"]2on regressions noticed by users or test-suites/cisah$]h&]uh1hhjv hhhhhM6ubh)}(hhh](h)}(h'On accepting when a regression occurredh]h'On accepting when a regression occurred}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMSubjx)}(hhh](j")}(hX From `2026-01-22 `_:: But starting to argue about users reporting breaking changes is basically the final line for me. I have a couple of people that I have in my spam block-list and refuse to have anything to do with, and they have generally been about exactly that. Note how it's not about making mistakes and _causing_ the regression. That's normal. That's development. But then arguing about it is a no-no. h](h)}(hvFrom `2026-01-22 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2026-01-22 `_h]h 2026-01-22}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2026-01-22jA_https://lore.kernel.org/all/CAHk-=wheQNiW_WtHGO7bKkT7Uib-p+ai2JP9M+z+FYcZ6CAxYA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id37ah ]h"]h$] 2026-01-22ah&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMUhjubj)}(hXBut starting to argue about users reporting breaking changes is basically the final line for me. I have a couple of people that I have in my spam block-list and refuse to have anything to do with, and they have generally been about exactly that. Note how it's not about making mistakes and _causing_ the regression. That's normal. That's development. But then arguing about it is a no-no.h]hXBut starting to argue about users reporting breaking changes is basically the final line for me. I have a couple of people that I have in my spam block-list and refuse to have anything to do with, and they have generally been about exactly that. Note how it's not about making mistakes and _causing_ the regression. That's normal. That's development. But then arguing about it is a no-no.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMWhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXWFrom `2024-06-23 `_:: We don't introduce regressions and then blame others. There's a very clear rule in kernel development: things that break other things ARE NOT FIXES. EVER. They get reverted, or the thing they broke gets fixed. h](h)}(hvFrom `2024-06-23 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2024-06-23 `_h]h 2024-06-23}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2024-06-23jA_https://lore.kernel.org/all/CAHk-=wi_KMO_rJ6OCr8mAWBRg-irziM=T9wxGC+J1VVoQb39gw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id38ah ]h"] 2024-06-23ah$]h&]refurij/uh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhM`hjubj)}(hWe don't introduce regressions and then blame others. There's a very clear rule in kernel development: things that break other things ARE NOT FIXES. EVER. They get reverted, or the thing they broke gets fixed.h]hWe don't introduce regressions and then blame others. There's a very clear rule in kernel development: things that break other things ARE NOT FIXES. EVER. They get reverted, or the thing they broke gets fixed.}hjGsbah}(h]h ]h"]h$]h&]hhuh1jhhhMbhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXtFrom `2021-06-05 `_:: THERE ARE NO VALID ARGUMENTS FOR REGRESSIONS. Honestly, security people need to understand that "not working" is not a success case of security. It's a failure case. Yes, "not working" may be secure. But security in that case is *pointless*. h](h)}(hvFrom `2021-06-05 `_::h](hFrom }(hj_hhhNhNubj0)}(ho`2021-06-05 `_h]h 2021-06-05}(hjghhhNhNubah}(h]h ]h"]h$]h&]name 2021-06-05jA_https://lore.kernel.org/all/CAHk-=wiUVqHN76YUwhkjZzwTdjMMJf_zN4+u7vEJjmEGh3recw@mail.gmail.com/uh1j/hj_ubjD)}(hb h]h}(h]id39ah ]h"] 2021-06-05ah$]h&]refurijwuh1jCjRKhj_ubh:}(hj_hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMkhj[ubj)}(hTHERE ARE NO VALID ARGUMENTS FOR REGRESSIONS. Honestly, security people need to understand that "not working" is not a success case of security. It's a failure case. Yes, "not working" may be secure. But security in that case is *pointless*.h]hTHERE ARE NO VALID ARGUMENTS FOR REGRESSIONS. Honestly, security people need to understand that "not working" is not a success case of security. It's a failure case. Yes, "not working" may be secure. But security in that case is *pointless*.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMmhj[ubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubj")}(hXFrom `2017-10-26(5) `_:: [...] when regressions *do* occur, we admit to them and fix them, instead of blaming user space. The fact that you have apparently been denying the regression now for three weeks means that I will revert, and I will stop pulling apparmor requests until the people involved understand how kernel development is done. h](h)}(hzFrom `2017-10-26(5) `_::h](hFrom }(hjhhhNhNubj0)}(hs`2017-10-26(5) `_h]h 2017-10-26(5)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(5)jA`https://lore.kernel.org/lkml/CA+55aFwiiQYJ+YoLKCXjN_beDVfu38mg=Ggg5LFOcqHE8Qi7Zw@mail.gmail.com/uh1j/hjubjD)}(hc h]h}(h]id40ah ]h"] 2017-10-26(5)ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMthjubj)}(hX<[...] when regressions *do* occur, we admit to them and fix them, instead of blaming user space. The fact that you have apparently been denying the regression now for three weeks means that I will revert, and I will stop pulling apparmor requests until the people involved understand how kernel development is done.h]hX<[...] when regressions *do* occur, we admit to them and fix them, instead of blaming user space. The fact that you have apparently been denying the regression now for three weeks means that I will revert, and I will stop pulling apparmor requests until the people involved understand how kernel development is done.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMvhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMUhjhhubeh}(h]'on-accepting-when-a-regression-occurredah ]h"]'on accepting when a regression occurredah$]h&]uh1hhjv hhhhhMSubh)}(hhh](h)}(hOn back-and-forthh]hOn back-and-forth}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh](j")}(hX_From `2024-05-28 `_:: The "no regressions" rule is that we do not introduce NEW bugs. It *literally* came about because we had an endless dance of "fix two bugs, introduce one new one", and that then resulted in a system that you cannot TRUST. h](h)}(hvFrom `2024-05-28 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2024-05-28 `_h]h 2024-05-28}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2024-05-28jA_https://lore.kernel.org/all/CAHk-=wgtb7y-bEh7tPDvDWru7ZKQ8-KMjZ53Tsk37zsPPdwXbA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id41ah ]h"]h$] 2024-05-28ah&]refurij)uh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhj ubj)}(hThe "no regressions" rule is that we do not introduce NEW bugs. It *literally* came about because we had an endless dance of "fix two bugs, introduce one new one", and that then resulted in a system that you cannot TRUST.h]hThe "no regressions" rule is that we do not introduce NEW bugs. It *literally* came about because we had an endless dance of "fix two bugs, introduce one new one", and that then resulted in a system that you cannot TRUST.}hjAsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhj ubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubj")}(hXFrom `2021-09-20(1) `_:: And the thing that makes regressions special is that back when I wasn't so strict about these things, we'd end up in endless "seesaw situations" where somebody would fix something, it would break something else, then that something else would break, and it would never actually converge on anything reliable at all. h](h)}(hyFrom `2021-09-20(1) `_::h](hFrom }(hjYhhhNhNubj0)}(hr`2021-09-20(1) `_h]h 2021-09-20(1)}(hjahhhNhNubah}(h]h ]h"]h$]h&]name 2021-09-20(1)jA_https://lore.kernel.org/all/CAHk-=wi7DB2SJ-wngVvsJ7Ak2cM556Q8437sOXo4EJt2BWPdEg@mail.gmail.com/uh1j/hjYubjD)}(hb h]h}(h]id42ah ]h"] 2021-09-20(1)ah$]h&]refurijquh1jCjRKhjYubh:}(hjYhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjUubj)}(hX;And the thing that makes regressions special is that back when I wasn't so strict about these things, we'd end up in endless "seesaw situations" where somebody would fix something, it would break something else, then that something else would break, and it would never actually converge on anything reliable at all.h]hX;And the thing that makes regressions special is that back when I wasn't so strict about these things, we'd end up in endless "seesaw situations" where somebody would fix something, it would break something else, then that something else would break, and it would never actually converge on anything reliable at all.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjUubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubj")}(hXFrom `2015-08-13 `_:: The strict policy of no regressions actually originally started mainly wrt suspend/resume issues, where the "fix one machine, break another" kind of back-and-forth caused endless problems, and meant that we didn't actually necessarily make any forward progress, just moving a problem around. h](h)}(hvFrom `2015-08-13 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2015-08-13 `_h]h 2015-08-13}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2015-08-13jA_https://lore.kernel.org/all/CA+55aFxk8-BsiKwr_S-c+4G6wihKPQVMLE34H9wOZpeua6W9+Q@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id43ah ]h"] 2015-08-13ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hX#The strict policy of no regressions actually originally started mainly wrt suspend/resume issues, where the "fix one machine, break another" kind of back-and-forth caused endless problems, and meant that we didn't actually necessarily make any forward progress, just moving a problem around.h]hX#The strict policy of no regressions actually originally started mainly wrt suspend/resume issues, where the "fix one machine, break another" kind of back-and-forth caused endless problems, and meant that we didn't actually necessarily make any forward progress, just moving a problem around.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hj hhhhhNubeh}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]on-back-and-forthah ]h"]on back-and-forthah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h-On changes with a risk of causing regressionsh]h-On changes with a risk of causing regressions}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh]j")}(hX~From `2023-06-02 `_:: So what I think you should do is to fix the bug right, with a clean patch, and no crazy hacks. That is something we can then apply and test. All the while knowing full well that "uhhuh, this is a visible change, we may have to revert it". If then some *real* load ends up showing a regression, we may just be screwed. Our current behavior may be buggy, but we have the rule that once user space depends on kernel bugs, they become features pretty much by definition, however much we might dislike it. h](h)}(hvFrom `2023-06-02 `_::h](hFrom }(hj hhhNhNubj0)}(ho`2023-06-02 `_h]h 2023-06-02}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2023-06-02jA_https://lore.kernel.org/all/CAHk-=wgyAGUMHmQM-5Eb556z5xiHZB7cF05qjrtUH4F7P-1rSA@mail.gmail.com/uh1j/hj ubjD)}(hb h]h}(h]id44ah ]h"] 2023-06-02ah$]h&]refurij#uh1jCjRKhj ubh:}(hj hhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hXSo what I think you should do is to fix the bug right, with a clean patch, and no crazy hacks. That is something we can then apply and test. All the while knowing full well that "uhhuh, this is a visible change, we may have to revert it". If then some *real* load ends up showing a regression, we may just be screwed. Our current behavior may be buggy, but we have the rule that once user space depends on kernel bugs, they become features pretty much by definition, however much we might dislike it.h]hXSo what I think you should do is to fix the bug right, with a clean patch, and no crazy hacks. That is something we can then apply and test. All the while knowing full well that "uhhuh, this is a visible change, we may have to revert it". If then some *real* load ends up showing a regression, we may just be screwed. Our current behavior may be buggy, but we have the rule that once user space depends on kernel bugs, they become features pretty much by definition, however much we might dislike it.}hj;sbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]-on-changes-with-a-risk-of-causing-regressionsah ]h"]-on changes with a risk of causing regressionsah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h-On in-kernel workarounds to avoid regressionsh]h-On in-kernel workarounds to avoid regressions}(hj`hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj]hhhhhMubjx)}(hhh]j")}(hXFrom `2017-10-26(6) `_:: Behavioral changes happen, and maybe we don't even support some feature any more. There's a number of fields in /proc//stat that are printed out as zeroes, simply because they don't even *exist* in the kernel any more, or because showing them was a mistake (typically an information leak). But the numbers got replaced by zeroes, so that the code that used to parse the fields still works. The user might not see everything they used to see, and so behavior is clearly different, but things still _work_, even if they might no longer show sensitive (or no longer relevant) information. h](h)}(hzFrom `2017-10-26(6) `_::h](hFrom }(hjuhhhNhNubj0)}(hs`2017-10-26(6) `_h]h 2017-10-26(6)}(hj}hhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(6)jA`https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/uh1j/hjuubjD)}(hc h]h}(h]id45ah ]h"] 2017-10-26(6)ah$]h&]refurijuh1jCjRKhjuubh:}(hjuhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjqubj)}(hXNBehavioral changes happen, and maybe we don't even support some feature any more. There's a number of fields in /proc//stat that are printed out as zeroes, simply because they don't even *exist* in the kernel any more, or because showing them was a mistake (typically an information leak). But the numbers got replaced by zeroes, so that the code that used to parse the fields still works. The user might not see everything they used to see, and so behavior is clearly different, but things still _work_, even if they might no longer show sensitive (or no longer relevant) information.h]hXNBehavioral changes happen, and maybe we don't even support some feature any more. There's a number of fields in /proc//stat that are printed out as zeroes, simply because they don't even *exist* in the kernel any more, or because showing them was a mistake (typically an information leak). But the numbers got replaced by zeroes, so that the code that used to parse the fields still works. The user might not see everything they used to see, and so behavior is clearly different, but things still _work_, even if they might no longer show sensitive (or no longer relevant) information.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjqubeh}(h]h ]h"]h$]h&]uh1j!hjnhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhj]hhubeh}(h]-on-in-kernel-workarounds-to-avoid-regressionsah ]h"]-on in-kernel workarounds to avoid regressionsah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h!On regressions caused by bugfixesh]h!On regressions caused by bugfixes}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh]j")}(hXFrom `2018-08-03 `_:: > Kernel had a bug which has been fixed That is *ENTIRELY* immaterial. Guys, whether something was buggy or not DOES NOT MATTER. [...] It's basically saying "I took something that worked, and I broke it, but now it's better". Do you not see how f*cking insane that statement is? h](h)}(hvFrom `2018-08-03 `_::h](hFrom }(hjhhhNhNubj0)}(ho`2018-08-03 `_h]h 2018-08-03}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2018-08-03jA_https://lore.kernel.org/all/CA+55aFwWZX=CXmWDTkDGb36kf12XmTehmQjbiMPCqCRG2hi9kw@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id46ah ]h"]h$] 2018-08-03ah&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hX> Kernel had a bug which has been fixed That is *ENTIRELY* immaterial. Guys, whether something was buggy or not DOES NOT MATTER. [...] It's basically saying "I took something that worked, and I broke it, but now it's better". Do you not see how f*cking insane that statement is?h]hX> Kernel had a bug which has been fixed That is *ENTIRELY* immaterial. Guys, whether something was buggy or not DOES NOT MATTER. [...] It's basically saying "I took something that worked, and I broke it, but now it's better". Do you not see how f*cking insane that statement is?}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]!on-regressions-caused-by-bugfixesah ]h"]!on regressions caused by bugfixesah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(hOn internal API changesh]hOn internal API changes}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj1hhhhhMubjx)}(hhh]j")}(hX From `2017-10-26(7) `_:: We do API breakage _inside_ the kernel all the time. We will fix internal problems by saying "you now need to do XYZ", but then it's about internal kernel API's, and the people who do that then also obviously have to fix up all the in-kernel users of that API. Nobody can say "I now broke the API you used, and now _you_ need to fix it up". Whoever broke something gets to fix it too. h](h)}(hzFrom `2017-10-26(7) `_::h](hFrom }(hjIhhhNhNubj0)}(hs`2017-10-26(7) `_h]h 2017-10-26(7)}(hjQhhhNhNubah}(h]h ]h"]h$]h&]name 2017-10-26(7)jA`https://lore.kernel.org/lkml/CA+55aFxW7NMAMvYhkvz1UPbUTUJewRt6Yb51QAx5RtrWOwjebg@mail.gmail.com/uh1j/hjIubjD)}(hc h]h}(h]id47ah ]h"] 2017-10-26(7)ah$]h&]refurijauh1jCjRKhjIubh:}(hjIhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjEubj)}(hXWe do API breakage _inside_ the kernel all the time. We will fix internal problems by saying "you now need to do XYZ", but then it's about internal kernel API's, and the people who do that then also obviously have to fix up all the in-kernel users of that API. Nobody can say "I now broke the API you used, and now _you_ need to fix it up". Whoever broke something gets to fix it too.h]hXWe do API breakage _inside_ the kernel all the time. We will fix internal problems by saying "you now need to do XYZ", but then it's about internal kernel API's, and the people who do that then also obviously have to fix up all the in-kernel users of that API. Nobody can say "I now broke the API you used, and now _you_ need to fix it up". Whoever broke something gets to fix it too.}hjysbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjEubeh}(h]h ]h"]h$]h&]uh1j!hjBhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhj1hhubeh}(h]on-internal-api-changesah ]h"]on internal api changesah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h+On regressions only found after a long timeh]h+On regressions only found after a long time}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh]j")}(hXCFrom `2024-03-28 `_:: I'm definitely not reverting a patch from almost a decade ago as a regression. If it took that long to find, it can't be that critical of a regression. So yes, let's treat it as a regular bug. h](h)}(hvFrom `2024-03-28 `_::h](hFrom }(hjhAhhNhNubj0)}(ho`2024-03-28 `_h]h 2024-03-28}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2024-03-28jA_https://lore.kernel.org/all/CAHk-=wgFuoHpMk_Z_R3qMXVDgq0N1592+bABkyGjwwSL4zBtHA@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id48ah ]h"] 2024-03-28ah$]h&]refurijuh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hI'm definitely not reverting a patch from almost a decade ago as a regression. If it took that long to find, it can't be that critical of a regression. So yes, let's treat it as a regular bug.h]hI'm definitely not reverting a patch from almost a decade ago as a regression. If it took that long to find, it can't be that critical of a regression. So yes, let's treat it as a regular bug.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]+on-regressions-only-found-after-a-long-timeah ]h"]+on regressions only found after a long timeah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h*On testing regressions fixes in linux-nexth]h*On testing regressions fixes in linux-next}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjhhhhhMubjx)}(hhh]j")}(hOn `maintainers summit 2024 `_:: So running fixes though linux-next is just a waste of time. h](h)}(hBOn `maintainers summit 2024 `_::h](hOn }(hjhhhNhNubj0)}(h=`maintainers summit 2024 `_h]hmaintainers summit 2024}(hj%hhhNhNubah}(h]h ]h"]h$]h&]namemaintainers summit 2024jA https://lwn.net/Articles/990599/uh1j/hjubjD)}(h# h]h}(h]maintainers-summit-2024ah ]h"]maintainers summit 2024ah$]h&]refurij5uh1jCjRKhjubh:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(h;So running fixes though linux-next is just a waste of time.h]h;So running fixes though linux-next is just a waste of time.}hjMsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjhhubeh}(h]*on-testing-regressions-fixes-in-linux-nextah ]h"]*on testing regressions fixes in linux-nextah$]h&]uh1hhjv hhhhhMubh)}(hhh](h)}(h-On a few other aspects related to regressionsh]h-On a few other aspects related to regressions}(hjrhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjohhhhhMubjx)}(hhh]j")}(hXFrom `2025-07-29(2) `_ [which `is not quite a regression, but a huge inconvenience `_]:: I no longer have sound. I also suspect that it's purely because "make oldconfig" doesn't work, and probably turned off my old Intel HDA settings. Or something. Renaming config parameters is *bad*. I've harped on the Kconfig phase of the kernel build probably being our nastiest point, and a real pain point to people getting involved with development simply because building your own kernel can be so daunting with hundreds of fairly esoteric questions. h](h)}(hXFrom `2025-07-29(2) `_ [which `is not quite a regression, but a huge inconvenience `_]::h](hFrom }(hjhhhNhNubj0)}(hr`2025-07-29(2) `_h]h 2025-07-29(2)}(hjhhhNhNubah}(h]h ]h"]h$]h&]name 2025-07-29(2)jA_https://lore.kernel.org/all/CAHk-=wjj9DvOZtmTkoLtyfHmy5mNKy6q_96d9=4FUEDXre=cww@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]id49ah ]h"] 2025-07-29(2)ah$]h&]refurijuh1jCjRKhjubh [which }(hjhhhNhNubj0)}(h`is not quite a regression, but a huge inconvenience `_h]h3is not quite a regression, but a huge inconvenience}(hjhhhNhNubah}(h]h ]h"]h$]h&]name3is not quite a regression, but a huge inconveniencejA_https://lore.kernel.org/all/CAHk-=wgO0Rx2LcYT4f75Xs46orbJ4JxO2jbAFQnVKDYAjV5HeQ@mail.gmail.com/uh1j/hjubjD)}(hb h]h}(h]2is-not-quite-a-regression-but-a-huge-inconvenienceah ]h"]3is not quite a regression, but a huge inconvenienceah$]h&]refurijuh1jCjRKhjubh]:}(hjhhhNhNubeh}(h]h ]h"]h$]h&]uh1hhhhMhjubj)}(hXI no longer have sound. I also suspect that it's purely because "make oldconfig" doesn't work, and probably turned off my old Intel HDA settings. Or something. Renaming config parameters is *bad*. I've harped on the Kconfig phase of the kernel build probably being our nastiest point, and a real pain point to people getting involved with development simply because building your own kernel can be so daunting with hundreds of fairly esoteric questions.h]hXI no longer have sound. I also suspect that it's purely because "make oldconfig" doesn't work, and probably turned off my old Intel HDA settings. Or something. Renaming config parameters is *bad*. I've harped on the Kconfig phase of the kernel build probably being our nastiest point, and a real pain point to people getting involved with development simply because building your own kernel can be so daunting with hundreds of fairly esoteric questions.}hjsbah}(h]h ]h"]h$]h&]hhuh1jhhhMhjubeh}(h]h ]h"]h$]h&]uh1j!hjhhhhhNubah}(h]h ]h"]h$]h&]jjuh1jwhhhMhjohhubh)}(hend-of-contenth]hend-of-content}hjsbah}(h]h ]h"]h$]h&]hhuh1hhjohhhhhMubh)}(hX`This text is available under GPL-2.0+ or CC-BY-4.0, as stated at the top of the file. If you want to distribute this text under CC-BY-4.0 only, please use "The Linux kernel developers" for author attribution and link this as source: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/plain/Documentation/process/handling-regressions.rsth]hX`This text is available under GPL-2.0+ or CC-BY-4.0, as stated at the top of the file. If you want to distribute this text under CC-BY-4.0 only, please use "The Linux kernel developers" for author attribution and link this as source: https://git.kernel.org/pub/scm/linux/kernel/git/torvalds/linux.git/plain/Documentation/process/handling-regressions.rst}hjsbah}(h]h ]h"]h$]h&]hhuh1hhjohhhhhMubh)}(hXNote: Only the content of this RST file as found in the Linux kernel sources is available under CC-BY-4.0, as versions of this text that were processed (for example by the kernel's build system) might contain content taken from files which use a more restrictive license.h]hXNote: Only the content of this RST file as found in the Linux kernel sources is available under CC-BY-4.0, as versions of this text that were processed (for example by the kernel's build system) might contain content taken from files which use a more restrictive license.}hjsbah}(h]h ]h"]h$]h&]hhuh1hhjohhhhhMubeh}(h]-on-a-few-other-aspects-related-to-regressionsah ]h"]-on a few other aspects related to regressionsah$]h&]uh1hhjv hhhhhMubeh}(h]"quotes-from-linus-about-regressionah ]h"]"quotes from linus about regressionah$]h&]uh1hhjhhhhhMubeh}(h]Call-the-details-on-linux-kernel-regressions-relevant-for-developersah ]h"]Call the details on linux kernel regressions relevant for developersah$]h&]uh1hhhhhhhhK9ubeh}(h]handling-regressionsah ]h"]handling regressionsah$]h&]uh1hhhhhhhhKubeh}(h]h ]h"]h$]h&]sourcehuh1hcurrent_sourceN current_lineNsettingsdocutils.frontendValues)}(hN generatorN datestampN source_linkN source_urlN toc_backlinksentryfootnote_backlinksK sectnum_xformKstrip_commentsNstrip_elements_with_classesN strip_classesN report_levelK halt_levelKexit_status_levelKdebugNwarning_streamN tracebackinput_encoding utf-8-siginput_encoding_error_handlerstrictoutput_encodingutf-8output_encoding_error_handlerj`error_encodingutf-8error_encoding_error_handlerbackslashreplace language_codeenrecord_dependenciesNconfigN id_prefixhauto_id_prefixid dump_settingsNdump_internalsNdump_transformsNdump_pseudo_xmlNexpose_internalsNstrict_visitorN_disable_configN_sourcehnj _destinationN _config_files]7/var/lib/git/docbuild/linux/Documentation/docutils.confafile_insertion_enabled raw_enabledKline_length_limitM'pep_referencesN pep_base_urlhttps://peps.python.org/pep_file_url_templatepep-%04drfc_referencesN rfc_base_url&https://datatracker.ietf.org/doc/html/ tab_widthKtrim_footnote_reference_spacesyntax_highlightlong smart_quotessmartquotes_locales]character_level_inline_markupdoctitle_xform docinfo_xformKsectsubtitle_xform image_loadinglinkembed_stylesheetcloak_email_addressessection_self_linkenvNubreporterNindirect_targets]substitution_defs}substitution_names}refnames}refids}nameids}(j:j7jjjNjKj2j/jLjIj!jjjjDjAj j jjj, j) j{ jx js jp j j j0 j- jW jT j~ j{ j j j j j j j j j7 j4 j^ j[ j j jk jh j j j6 j3 jX jU j*j'jjj j j j jUjRjjjjj|jyj.j+jjjjjjjkjhjjjjjejbjjjjj_j\jjjjjjjjjj jjjjjjjMjJjjjjj%j"jmjjjjjEjBjjjjjjjjjjj9j6jj~jjjjj{jxjjjZjWj-j*jjjjj.j+jjjkjhjjjjjljij?j<j"jjjjju nametypes}(j:jjNj2jLj!jjDj jj, j{ js j j0 jW j~ j j j j j7 j^ j jk j j6 jX j*jj j jUjjj|j.jjjjkjjjejjj_jjjjjjjjjMjjj%jmjjEjjjjjj9jjjj{jjZj-jjj.jjkjjjlj?j"jjuh}(j7hjj jKjEj/jjIjjjj jjj$jAjj jOjj`j) jjx j/ jp j j j j- j jT j3 j{ jZ j j j j j j j j j4 j j[ j: j ja jh j j j j3 j- jU jO j'jv jj j j j jjRjLjjjjjyjjLjFj+jjjjjjj1jhjbjjjjjjjbj\jjjjjjj\jVjjjjjjjVjPjjjjj jjPjJjjjjjjjJjDjjjjj"jjjjdjjjjjBj<jjjjjjj<j6jj~jjjjj6j0j~jxjjjjj0j*jxjrjjjWjj*j$jj]jjj+jjjjj1jhjbjjjjjijj<j6jjojjjju footnote_refs} citation_refs} autofootnotes]autofootnote_refs]symbol_footnotes]symbol_footnote_refs] footnotes] citations]autofootnote_startKsymbol_footnote_startK id_counter collectionsCounter}jnK1sRparse_messages](hsystem_message)}(hhh]h)}(h:Duplicate explicit target name: "regression mailing list".h]h>Duplicate explicit target name: “regression mailing list”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]j alevelKtypeINFOsourcehnjlineKuh1jhjhhhhhKEubj)}(hhh]h)}(h-Duplicate explicit target name: "2026-01-28".h]h1Duplicate explicit target name: “2026-01-28”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jLalevelKtypejsourcehnjlineKuh1jhj)ubj)}(hhh]h)}(h-Duplicate explicit target name: "2023-04-20".h]h1Duplicate explicit target name: “2023-04-20”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhjubj)}(hhh]h)}(h-Duplicate explicit target name: "2026-01-22".h]h1Duplicate explicit target name: “2026-01-22”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhjubj)}(hhh]h)}(h-Duplicate explicit target name: "2026-01-22".h]h1Duplicate explicit target name: “2026-01-22”.}(hj4hhhNhNubah}(h]h ]h"]h$]h&]uh1hhj1ubah}(h]h ]h"]h$]h&]jValevelKtypejsourcehnjlineKuh1jhj3ubj)}(hhh]h)}(h-Duplicate explicit target name: "2023-04-20".h]h1Duplicate explicit target name: “2023-04-20”.}(hjOhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjLubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhj{ubj)}(hhh]h)}(h-Duplicate explicit target name: "2018-08-03".h]h1Duplicate explicit target name: “2018-08-03”.}(hjjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjgubah}(h]h ]h"]h$]h&]jPalevelKtypejsourcehnjlineKuh1jhj-ubj)}(hhh]h)}(h-Duplicate explicit target name: "2020-05-21".h]h1Duplicate explicit target name: “2020-05-21”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhjubj)}(hhh]h)}(h-Duplicate explicit target name: "2026-01-22".h]h1Duplicate explicit target name: “2026-01-22”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]j<alevelKtypejsourcehnjlineKuh1jhjubj)}(hhh]h)}(h-Duplicate explicit target name: "2026-01-22".h]h1Duplicate explicit target name: “2026-01-22”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhjubj)}(hhh]h)}(h-Duplicate explicit target name: "2024-05-28".h]h1Duplicate explicit target name: “2024-05-28”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]j0alevelKtypejsourcehnjlineKuh1jhj ubj)}(hhh]h)}(h-Duplicate explicit target name: "2018-08-03".h]h1Duplicate explicit target name: “2018-08-03”.}(hjhhhNhNubah}(h]h ]h"]h$]h&]uh1hhjubah}(h]h ]h"]h$]h&]jalevelKtypejsourcehnjlineKuh1jhjubetransform_messages] transformerN include_log] decorationNhhub.