1834567 - High freq Assertion failure: !StylistNeedsUpdate(), at /builds/worker/checkouts/gecko/layout/style/ServoStyleSet.cpp:609

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect, P5)

Description

[Treeherder Bug Filer]

Filed by: nerli [at] mozilla.com
Parsed log: https://treeherder.mozilla.org/logviewer?job_id=416731257&repo=autoland
Full log: https://firefox-ci-tc.services.mozilla.com/api/queue/v1/task/SpqfCmcXSUyab6wDXa0u-Q/runs/0/artifacts/public/logs/live_backing.log

---

[task 2023-05-23T15:13:51.417Z] 15:13:51     INFO - TEST-START | browser/components/preferences/tests/browser_localSearchShortcuts.js
[task 2023-05-23T15:13:51.434Z] 15:13:51     INFO - GECKO(1590) | [Child 1786: Main Thread]: I/DocShellAndDOMWindowLeak ++DOCSHELL 7f1434733c00 == 1 [pid = 1786] [id = 11]
[task 2023-05-23T15:13:51.436Z] 15:13:51     INFO - GECKO(1590) | [Child 1786: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 1 (7f143986d470) [pid = 1786] [serial = 23] [outer = 0]
[task 2023-05-23T15:13:51.439Z] 15:13:51     INFO - GECKO(1590) | [Child 1786: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 2 (7f1434735000) [pid = 1786] [serial = 24] [outer = 7f143986d470]
[task 2023-05-23T15:13:51.593Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOCSHELL 7fc008a09c00 == 17 [pid = 1590] [id = 358]
[task 2023-05-23T15:13:51.596Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 56 (7fc0015d4940) [pid = 1590] [serial = 1162] [outer = 0]
[task 2023-05-23T15:13:51.599Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 57 (7fc00b6d1400) [pid = 1590] [serial = 1163] [outer = 7fc0015d4940]
[task 2023-05-23T15:13:51.614Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590, Main Thread] WARNING: '!sTextInputHandlingWidget', file /builds/worker/checkouts/gecko/dom/events/IMEStateManager.cpp:1015
[task 2023-05-23T15:13:51.630Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 58 (7fc007212800) [pid = 1590] [serial = 1164] [outer = 7fc0015d4940]
[task 2023-05-23T15:13:51.675Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590, Main Thread] WARNING: 'aOwner->IsDiscarded()', file /builds/worker/workspace/obj-build/dist/include/mozilla/dom/SyncedContextInlines.h:95
[task 2023-05-23T15:13:51.973Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOCSHELL 7fc00c7b6000 == 18 [pid = 1590] [id = 359]
[task 2023-05-23T15:13:51.975Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 59 (7fc00d0704e0) [pid = 1590] [serial = 1165] [outer = 0]
[task 2023-05-23T15:13:51.976Z] 15:13:51     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 60 (7fc00c7b6800) [pid = 1590] [serial = 1166] [outer = 7fc00d0704e0]
[task 2023-05-23T15:13:52.007Z] 15:13:52     INFO - GECKO(1590) | [Child 1782: Main Thread]: I/DocShellAndDOMWindowLeak --DOCSHELL 7fdb90036c00 == 0 [pid = 1782] [id = 40] [url = about:blank]
[task 2023-05-23T15:13:52.383Z] 15:13:52     INFO - GECKO(1590) | [Parent 1590: Main Thread]: I/DocShellAndDOMWindowLeak ++DOMWINDOW == 61 (7fc00c9b8800) [pid = 1590] [serial = 1167] [outer = 7fc00d0704e0]
[task 2023-05-23T15:13:52.741Z] 15:13:52     INFO - GECKO(1590) | [Parent 1590, Main Thread] WARNING: NS_ENSURE_SUCCESS(rv, rv) failed with result 0x80004005 (NS_ERROR_FAILURE): file /builds/worker/checkouts/gecko/toolkit/xre/nsXREDirProvider.cpp:475
[task 2023-05-23T15:13:52.836Z] 15:13:52     INFO - GECKO(1590) | [fluent-dom] While translating an element with fluent ID "translations-manage-download-button" a child element of type "hbox" was removed. Either the fluent message does not contain markup, or it does not contain markup of this type.
[task 2023-05-23T15:13:52.839Z] 15:13:52     INFO - GECKO(1590) | [fluent-dom] While translating an element with fluent ID "translations-manage-delete-button" a child element of type "hbox" was removed. Either the fluent message does not contain markup, or it does not contain markup of this type.
[task 2023-05-23T15:13:52.889Z] 15:13:52     INFO - GECKO(1590) | Assertion failure: !StylistNeedsUpdate(), at /builds/worker/checkouts/gecko/layout/style/ServoStyleSet.cpp:609
[task 2023-05-23T15:13:52.899Z] 15:13:52     INFO -  Initializing stack-fixing for the first stack frame, this may take a while...

Comment 1

[Iulian Moraru]

First occurrence, so far: backfill range and retriggers.

Comment 2

[Intermittent Failures Robot]

9 failures in 3946 pushes (0.002 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• autoland: 9

Platform and build breakdown:

• linux1804-64-qr: 2
  • debug: 2
• macosx1015-64-qr: 1
  • debug: 1
• windows11-64-2009-qr: 4
  • debug: 4
• windows11-32-2009-qr: 2
  • debug: 2

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-05-22&endday=2023-05-28&tree=all

Comment 3

[Intermittent Failures Robot]

19 failures in 854 pushes (0.022 failures/push) were associated with this bug yesterday.

Repository breakdown:

• autoland: 19

Platform and build breakdown:

• macosx1015-64-qr: 10
  • debug: 10
• linux1804-64-qr: 2
  • debug: 2
• windows11-32-2009-qr: 2
  • debug: 2
• windows11-64-2009-qr: 5
  • debug: 5

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-05-31&endday=2023-05-31&tree=all

Comment 4

[Intermittent Failures Robot]

41 failures in 657 pushes (0.062 failures/push) were associated with this bug yesterday.

Repository breakdown:

• autoland: 29
• mozilla-central: 5
• try: 7

Platform and build breakdown:

• macosx1015-64-qr: 8
  • debug: 8
• windows11-64-2009-qr: 23
  • debug: 23
• linux1804-64-qr: 4
  • debug: 4
• windows11-32-2009-qr: 6
  • debug: 6

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-06-01&endday=2023-06-01&tree=all

Comment 5

[Natalia Csoregi [:nataliaCs]]

Update

There have been 54 failures within the last 7 days:

• 6 failures on Linux 18.04 x64 WebRender debug
• 18 failures on OS X 10.15 WebRender debug
• 8 failures on Windows 11 x86 22H2 WebRender debug
• 22 failures on Windows 11 x64 22H2 WebRender debug

Recent failure log: https://treeherder.mozilla.org/logviewer?job_id=417800508&repo=autoland&lineNumber=17059

Emilio, can you help us assign this to someone?
Thank you.

Comment 6

[Emilio Cobos Álvarez (:emilio)]

Do we have an idea of what caused this to start popping up?

Comment 7

[Emilio Cobos Álvarez (:emilio)]

Also is it always when running browser/components/preferences/tests/browser_localSearchShortcuts.js? Or are there other tests that show this?

Comment 8

[Noemi Erli[:noemi_erli]]

Our investigation reveals that this might be caused by https://bugzilla.mozilla.org/show_bug.cgi?id=1831259.
This was initially landed on 23rd of May and was backed out a few pushes later. Another change was then landed on the 30th and was backed out, followed by another land on 31st.
The last land was requested to be backed out on 1st of June for causing 2 other regressions. Since then, the fails stopped.
All the above can be seen in orange factor.

Found other two tests that show this:
browser/components/preferences/tests/browser_searchRestoreDefaults.js
browser/components/preferences/tests/browser_searchFindMoreLink.js

Comment 9

[BugBot [:suhaib / :marco/ :calixte]]

:Gijs, since you are the author of the regressor, bug 1831259, could you take a look?

For more information, please visit BugBot documentation.

Comment 10

[:Gijs (out for now; he/him)]

Looks like the fluent errors for the language download stuff are maybe a red herring? But either way that's being fixed in https://phabricator.services.mozilla.com/D179718 .

I don't really know what this assertion means - the message is not very helpful for people not familiar with the code - or why it would be triggered as a result of the patches in bug 1831259. Broadly, those patches are trying to make initialization of the preference panes more sensible. On current central, that initialization manually translates all the fluent while the nodes for not-yet-initialized panes are not in the DOM, and waits for that to be complete before inserting the content. After my patch, the content is just inserted into the DOM and we depend on "normal" fluent mutation-observer based l10n to do the translation work.

Emilio, does that help shed light on why this assertion might trigger? Is there something else I can do to help?

Comment 11

[Intermittent Failures Robot]

61 failures in 3471 pushes (0.018 failures/push) were associated with this bug in the last 7 days.

This is the #11 most frequent failure this week.

** This failure happened more than 30 times this week! Resolving this bug is a high priority. **

** Try to resolve this bug as soon as possible. If unresolved for 2 weeks, the affected test(s) may be disabled. **

Repository breakdown:

• autoland: 49
• mozilla-central: 5
• try: 7

Platform and build breakdown:

• macosx1015-64-qr: 19
  • debug: 19
• linux1804-64-qr: 6
  • debug: 6
• windows11-32-2009-qr: 8
  • debug: 8
• windows11-64-2009-qr: 28
  • debug: 28

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-05-29&endday=2023-06-04&tree=all

Comment 12

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1834567 - Downgrade an assertion in ResolveXULTreePseudoStyle. r=#style,#layout

If this happens the caller could potentially get an slightly outdated
style, but I couldn't repro this locally and realistically it's not a
big deal in this caller, and it's not observable to the web, so...

Comment 13

[Emilio Cobos Álvarez (:emilio)]

So what that assertion means is that somebody injected stylesheets (might be a shadow tree, maybe something else), but we haven't flushed before calling ResolveTreePseudoStyle. I couldn't repro this when I tried yesterday and I don't think I want to spend a bunch of time investigating this, worst case we just get a one-frame delay in getting the cursor of a XUL tree right...

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d0e9eed0248a
Downgrade an assertion in ResolveXULTreePseudoStyle. r=jwatt

Comment 15

[Cosmin Sabou [:CosminS]]

https://hg.mozilla.org/mozilla-central/rev/d0e9eed0248a

Comment 16

[BugBot [:suhaib / :marco/ :calixte]]

Set release status flags based on info from the regressing bug 1831259

Comment 17

[Intermittent Failures Robot]

1 failures in 3885 pushes (0.0 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• try: 1

Platform and build breakdown:

• macosx1015-64-qr: 1
  • debug: 1

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-06-05&endday=2023-06-11&tree=all

Comment 18

[Intermittent Failures Robot]

4 failures in 3850 pushes (0.001 failures/push) were associated with this bug in the last 7 days.

Repository breakdown:

• try: 4

Platform and build breakdown:

• linux1804-64-qr: 4
  • debug: 4

For more details, see:
https://treeherder.mozilla.org/intermittent-failures/bugdetails?bug=1834567&startday=2023-09-18&endday=2023-09-24&tree=all