1893127 - Firefox 125 CSS calculation issue

Status: VERIFIED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[christian.josephi]

Attachment: Firefox 125 CSS issue.png

User Agent: Mozilla/5.0 (Windows NT 10.0; Win64; x64; rv:125.0) Gecko/20100101 Firefox/125.0

Steps to reproduce:

Call https://www.cede.ch/en/music/?view=detail&branch_sub=0&branch=1&aid=18446451 and check the position of the buttons 'Add to shopping cart' and 'On the wish list' on a large screen (desktop). Also gradualy change the width of the browser window, to see the behaviour of the 2 buttons

Actual results:

Starting with version 125 the CSS calculation seems to have a problem. Thus the 2 buttons 'Add to shopping cart' and 'On the wish list' on large screens (desktops) are not getting displayed side by side but above each other (depending on the actual size of the browser window). When changing the width of the browser window, the 2 buttons begin to jump around, from side by side to above each other.

Expected results:

The 2 buttons 'Add to shopping cart' and 'On the wish list' should be displayed side by side consistently, also when changing the width of the browser window, down to the point where the layout changes for smaller screens. This is the behaviour in other browsers and in firefox up to and including version 124.

Comment 1

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

The Bugbug bot thinks this bug should belong to the 'Firefox::Messaging System' component, and is moving the bug to that component. Please correct in case you think the bot is wrong.

Comment 2

[6k64x4ma]

Regression window:
https://hg.mozilla.org/integration/autoland/pushloghtml?fromchange=cfa733c700c170913bdfa72ca88759e93bfc0b95&tochange=09a1b473f8dab1609e2ca9ae45ead6a8300d9a78

Comment 3

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

:emilio, since you are the author of the regressor, bug 1881220, could you take a look? Also, could you set the severity field?

For more information, please visit BugBot documentation.

Comment 4

[Emilio Cobos Álvarez (:emilio)]

Attachment: Reduced test-case.

Hmm, so this is generally hard because due to floating precision as soon as we go back to pixels we can get different results at the end... If we do the math in fixed-point then that's also bad because a lot of features rely on using CSS pixels...

As a workaround the site could use round(down, 50% - ..., 1px) to avoid these shenanigans of course, but that's not great...

Comment 6

[Emilio Cobos Álvarez (:emilio)]

Attachment: Bug 1893127 - Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=#layout,#style

As a short term fix this seems acceptable. Mark the test for the
regressing bug as failing.

Comment 8

[Emilio Cobos Álvarez (:emilio)]

Comment on attachment 9398998 [details]
Bug 1893127 - Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=#layout,#style

Beta/Release Uplift Approval Request

• User impact if declined: comment 0 and duplicates
• 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): Reverts to previous behavior, while keeping the tests.
• String changes made/needed: none
• Is Android affected?: Yes

Comment 9

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/d8e23e8fad18
Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=layout-reviewers,firefox-style-system-reviewers,boris

Comment 10

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Created web-platform-tests PR https://github.com/web-platform-tests/wpt/pull/45944 for changes under testing/web-platform/tests

Comment 11

[Pulsebot]

Backout by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/03b997590154
Backed out changeset d8e23e8fad18 for causing failures in calc-rounding-001.html. CLOSED TREE

Comment 12

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR was closed without merging

Comment 13

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/8256dde05335
Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=layout-reviewers,firefox-style-system-reviewers,boris

Comment 14

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/bacbb8f0dc66
calc-rounding-002.html passes in some platforms

Comment 15

[Sandor Molnar[:smolnar]]

https://hg.mozilla.org/mozilla-central/rev/8256dde05335
https://hg.mozilla.org/mozilla-central/rev/bacbb8f0dc66

Comment 16

[Donal Meehan [:dmeehan]]

Comment on attachment 9398998 [details]
Bug 1893127 - Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=#layout,#style

Approved for 126.0b7

Comment 17

[Pulsebot]

https://hg.mozilla.org/releases/mozilla-beta/rev/9a6dadd99589
https://hg.mozilla.org/releases/mozilla-beta/rev/35ea5eaec0d6

Comment 18

[Web Platform Test Sync Bot [:wpt-sync] (Matrix: #interop:mozilla.org)]

Upstream PR merged by moz-wptsync-bot

Comment 19

[Mihai Boldan, Desktop QA [:mboldan]]

I managed to reproduce this issue on Firefox 126.0a1 (2024-04-11), under Windows 11, using the STR from Comment 0.
I confirm that the issue is no longer reproducible on Firefox 127.0a1 (2024-04-28), or on Firefox 126.0b7(treeherder build). Tests were performed on
Windows 11, macOS 10.15 and Ubuntu 22.04.

Comment 20

[Donal Meehan [:dmeehan]]

Comment on attachment 9398998 [details]
Bug 1893127 - Back out bug 1881220 for causing worse regressions than the problem it was fixing. r=#layout,#style

Rejecting release uplift request since we are in Fx126 RC week.
This will ride the train to release with Fx126.