[Webkit-unassigned] [Bug 228954] New: iOS 14.6 Regression Accelerated animations freeze when invalidating layout with shadow dom

[bugzilla-daemon at webkit.org]

https://bugs.webkit.org/show_bug.cgi?id=228954

            Bug ID: 228954
           Summary: iOS 14.6 Regression Accelerated animations freeze when
                    invalidating layout with shadow dom
           Product: WebKit
           Version: Safari Technology Preview
          Hardware: Unspecified
                OS: Unspecified
            Status: NEW
          Severity: Normal
          Priority: P2
         Component: Animations
          Assignee: webkit-unassigned at lists.webkit.org
          Reporter: ldebeasi at gmail.com
                CC: dino at apple.com, graouts at apple.com

Created attachment 435256

  --> https://bugs.webkit.org/attachment.cgi?id=435256&action=review

Code reproduction

When invalidating the layout on a component with the Shadow DOM, any accelerated animations running on the component temporarily freeze. This issue was introduced around iOS 14.6. This issue can happen in Ionic apps when a page transition is occurring and developers try to render content on the entering view while the animation is playing (see: https://github.com/ionic-team/ionic-framework/issues/23732).

Steps to reproduce:

1. Open code reproduction on an iOS device running iOS 14.6 or later.
2. Click the "Start Animation" button.
3. Content should translate from the right to the left but should freeze about 100ms in because the layout was invalidated.
4. Repeat steps 1-2 on an iOS device running iOS 14.5 or older. You should notice that the animation does not freeze.

Expected Behavior:

I would expect the animation not to freeze when invalidating the layout.

Actual Behavior:

The animation freezes when invalidating the layout.

Other Information:

- This only happens with accelerated animations.
- This does not happen on Chrome or Firefox.
- This does not reproduce on Safari for macOS on Version 14.1 (16611.1.21.161.6). However, the issue does reproduce in STP 128.
- Possibly related to an older bug that fixed a similar issue: https://bugs.webkit.org/show_bug.cgi?id=201048

-- 
You are receiving this mail because:
You are the assignee for the bug.
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.webkit.org/pipermail/webkit-unassigned/attachments/20210810/f85a3b4d/attachment.htm>