gecko-dev/testing/web-platform/tests/hr-time
Noam Rosenthal 8eed3f9fe1 Bug 1862329 [wpt PR 42815] - Retain an AnimationClock per-document instead of per-page, a=testonly
Automatic update from web-platform-tests
Retain an AnimationClock per-document instead of per-page

Since each document coarsens timestamps differently, having a clock
for each makes more sense than having one per the all page and then
coarsening when retrieving the time.

Coarsening on the getter was causing a performance regression, so
coarsening at the start of the frame when setting the clock fixes it,
but requires a separate clock per document.

We still keep the page clock for things like SVG images and intersection
stability, though that can be removed in the future.

This also matches the spec more closely, where time is coarsened
when updating the rendering rather than in DocumentTimeline.currentTime.

Bug: 1496136
Change-Id: Ifbd4ca03471232598585db9ee38208c00bcead99
Reviewed-on: https://chromium-review.googlesource.com/c/chromium/src/+/4981913
Reviewed-by: Robert Flack <flackr@chromium.org>
Commit-Queue: Noam Rosenthal <nrosenthal@chromium.org>
Cr-Commit-Position: refs/heads/main@{#1217900}

--

wpt-commits: f3f612aca8a961713d0fc5633d061a0fe3cc3201
wpt-pr: 42815
2023-11-19 15:19:22 +00:00
..
resources
META.yml
basic.any.js
clamped-time-origin-isolated.https.html
clamped-time-origin-isolated.https.html.headers
clamped-time-origin.html
cross-origin-isolated-timing-attack.https.html
cross-origin-isolated-timing-attack.https.html.headers
idlharness-shadowrealm.window.js
idlharness.any.js
monotonic-clock.any.js
navigation-start-post-before-unload.html
performance-tojson.html
raf-coarsened-time.https.html
raf-coarsened-time.https.html.headers
test_cross_frame_start.html
timeOrigin.html
timing-attack.html
unload-manual.html
window-worker-timeOrigin.window.js