1853258 - has-complexity times out on opt builds

Status: RESOLVED FIXED

(Core :: CSS Parsing and Computation, defect)

Description

[David Shin[:dshin] (PTO Til 17th)]

Introduced here

Comment 1

[David Shin[:dshin] (PTO Til 17th)]

From local opt build with modified has-complexity.html that runs the highlighted portion on button click, but with 10000 elements: https://share.firefox.dev/3ECI0FM

Right.. so the invalidator is trying to find .subject, and running across all instances of #container > span, which is in the same subtree as .subject from main.

Comment 2

[Thomas Wisniewski [:twisniewski]]

We're also failing the related WPT with a timeout: https://wpt.fyi/results/css/selectors/invalidation/has-complexity.html

Comment 3

[David Shin[:dshin] (PTO Til 17th)]

Attachment: Bug 1853258: Part 1 - Refactor where additional relative selector invalidations are added. r=emilio,#layout-reviewers

This is slightly more efficient as invalidations are added after duplicates have been removed.
We will be running some matching that we definitely do not want to run on duplicates.

Comment 4

[David Shin[:dshin] (PTO Til 17th)]

Attachment: Bug 1853258: Part 2 - Provide which DOM mutation is happening for relative selector invalidation. r=emilio,#layout-reviewers

Required for upcoming optimization.
Also use the added information to move invalidation acceptance logic.

Depends on D191305

Comment 5

[David Shin[:dshin] (PTO Til 17th)]

Attachment: Bug 1853258: Part 3 - Optimize invalidating relative selectors on DOM mutation. r=emilio,#layout-reviewers

There are cases where we can run a selector match to realize that some DOM mutations
will not make a difference in the relative selector match state, which avoids
O(n^2) behaviour under some circumstances.

Depends on D191306

Comment 6

[Pulsebot]

Pushed by [e-mail]:
https://hg.mozilla.org/integration/autoland/rev/43c5a48d4a2f
Part 1 - Refactor where additional relative selector invalidations are added. r=emilio,layout-reviewers
https://hg.mozilla.org/integration/autoland/rev/4866d633239c
Part 2 - Provide which DOM mutation is happening for relative selector invalidation. r=emilio
https://hg.mozilla.org/integration/autoland/rev/5e5e1e48b9bb
Part 3 - Optimize invalidating relative selectors on DOM mutation. r=emilio

Comment 7

[Norisz Fay [:noriszfay]]

https://hg.mozilla.org/mozilla-central/rev/43c5a48d4a2f
https://hg.mozilla.org/mozilla-central/rev/4866d633239c
https://hg.mozilla.org/mozilla-central/rev/5e5e1e48b9bb