Closed Bug 1838045 Opened 2 years ago Closed 2 years ago

Categories

(Core :: CSS Parsing and Computation, defect)

defect

Tracking

()

VERIFIED FIXED
116 Branch
Tracking Status
firefox-esr102 --- unaffected
firefox114 --- unaffected
firefox115 + verified
firefox116 --- verified

People

(Reporter: twisniewski, Assigned: emilio)

References

(Blocks 2 open bugs, Regressed 1 open bug, Regression, )

Details

(Keywords: regression, webcompat:platform-bug)

Attachments

(2 files, 1 obsolete file)

As reported on webcompat.com, the site https://www.ing.com.au/securebanking/ no longer loads on pre-release builds, but does on Firefox 114 release (it can take a few seconds for the page to load). Mozregression pointed me to bug 1834081:

Bug 1834081 - Cache inline stylesheets regardless of shadow dom. r=mstange

Some pages (like Reddit) would have duplicate multi-kilobyte stylesheets
even at the top level.

Differential Revision: https://phabricator.services.mozilla.com/D178548

2023-06-12T12:32:26.230000: DEBUG : Did not find a branch, checking all integration branches
2023-06-12T12:32:26.233000: INFO : The bisection is done.
2023-06-12T12:32:26.234000: INFO : Stopped

I'm not sure how it would break the page, so I thought I would ask before diving into a deep diagnosis, in case it's a red herring.

Set release status flags based on info from the regressing bug 1834081

:emilio, since you are the author of the regressor, bug 1834081, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

[Tracking Requested - why for this release]: We should avoid shipping this regression.

Tracking for Fx115
:emilio we can look at back the regressor out of beta if you don't have a fix. Could you triage the severity/priority?

The bug is marked as tracked for firefox115 (beta). However, the bug still isn't assigned.

:fgriffith, could you please find an assignee for this tracked bug? Given that it is a regression and we know the cause, we could also simply backout the regressor. If you disagree with the tracking decision, please talk with the release managers.

For more information, please visit BugBot documentation.

Flags: needinfo?(fgriffith)

Yikes, shut up BugBot.

Assignee: nobody → emilio
Flags: needinfo?(fgriffith)

Thank you.

Turns out we do fire load events on these (I thought they were only for
sheets with imports).

Simplify a bit some related code for links. Callers / observers need to
deal with SheetComplete being called sync anyways because of this:

https://searchfox.org/mozilla-central/rev/962a843f6d96283c45162c788dc72bf217fca7df/layout/style/Loader.cpp#1873

So this removes the need for SheetLoadData to be a runnable etc.

Yeah, let's disable the regressing bug on beta.

Flags: needinfo?(emilio)

Comment on attachment 9338906 [details]
Bug 1838045 - Beta patch, disable regressing bug.

Beta/Release Uplift Approval Request

  • User impact if declined: comment 0, possibly other sites affected too.
  • Is this code covered by automated tests?: Yes
  • Has the fix been verified in Nightly?: Yes
  • Needs manual test from QE?: Yes
  • If yes, steps to reproduce: comment 0
  • List of other uplifts needed: none
  • Risk to taking this patch: Low
  • Why is the change risky/not risky? (and alternatives if risky): Toggles back the pref introduced in the regressing bug to restore to pre-regression behavior.
  • String changes made/needed: none
  • Is Android affected?: Yes
Attachment #9338906 - Flags: approval-mozilla-beta?
Flags: qe-verify+

Comment on attachment 9338906 [details]
Bug 1838045 - Beta patch, disable regressing bug.

Approved for 115.0b5.

Attachment #9338906 - Flags: approval-mozilla-beta? → approval-mozilla-beta+

Okay let's backout the regressor directly instead I guess :)

Flags: needinfo?(dmeehan)
Attachment #9338906 - Attachment is obsolete: true

I can back out the regressor (Bug 1834081) and a regression fix that landed afterwards (Bug 1834786) in the next beta push

Flags: needinfo?(dmeehan)
Blocks: 1838346

The spec asks to post a task, see html.spec.whatwg.org/#update-a-style-block

What we were doing is quite crazy. Unifying the scheduling between
<style> and <link rel=stylesheet> on the patch above causes failures on
some mochitests.

Some of them aren't really related to these and are a pre-existing bug,
see bug 1838346.

Depends on D180840

Depends on: 1838404

Setting Fx115 to Fixed, the regressor bug 1834081 was backed out of beta

Pushed by ealvarez@mozilla.com: https://hg.mozilla.org/integration/autoland/rev/10fc6f33d4ca Fix load event on cached inline style sheets. r=mstange https://hg.mozilla.org/integration/autoland/rev/775a6521f77b Make <link rel=stylesheet> load event scheduling saner. r=smaug
Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/40553 for changes under testing/web-platform/tests
Regressions: 1838544
QA Whiteboard: [qa-triaged]

Page https://www.ing.com.au/securebanking/ is still not loading on Win10x64 and MAC 12.6 using Firefox build 115.0b5 (20230613195413), should I wait for 115.0b6? Thank you.

Flags: needinfo?(emilio)

Not sure, Donal?

Flags: needinfo?(emilio) → needinfo?(dmeehan)
Status: NEW → RESOLVED
Closed: 2 years ago
Resolution: --- → FIXED
Target Milestone: --- → 116 Branch

(In reply to Monica Chiorean from comment #20)

Page https://www.ing.com.au/securebanking/ is still not loading on Win10x64 and MAC 12.6 using Firefox build 115.0b5 (20230613195413), should I wait for 115.0b6? Thank you.

Yes, you should wait for 115.0b6 or test from the latest Treeherder builds.
The backout was pushed after the 115.0b5 build. 115.0b6 builds later today.

Flags: needinfo?(dmeehan)
Upstream PR merged by moz-wptsync-bot

Verified issue as fixed on Win10/Mac 12.6/Ubuntu 20.04 using FF builds 115.0b6(20230615175802) and 116.0a1(20230616094829).

Status: RESOLVED → VERIFIED
QA Whiteboard: [qa-triaged]
Flags: qe-verify+
Regressions: 1874766
Blocks: 1899242
Regressions: 1940025
You need to log in before you can comment on or make changes to this bug.

Attachment

General

Created:
Updated:
Size: