Do we have test coverage for will-change:clip-path without backdrop filter won't create a backdrop root?
if (object_.HasLayer()) {Do we still need a feature gate here?
if (RuntimeEnabledFeatures::
BackdropRootForClipPathWithBackdropFilterEnabled() &&
properties_ && properties_->ClipPathClip() && object_.HasLayer()) {
}This if block is now empty.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Do we have test coverage for will-change:clip-path without backdrop filter won't create a backdrop root?
This is impossible to check with a web test as the only way to visually determine whether a backdrop root exists is by putting a backdrop filter over it. Anything else would be a bug.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
if (object_.HasLayer()) {Wangsong JinDo we still need a feature gate here?
Just a reminder: if `BackdropRootForClipPathWithBackdropFilterEnabled` is no longer needed, please remove it from runtime_enabled_features.json5.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
if (RuntimeEnabledFeatures::
BackdropRootForClipPathWithBackdropFilterEnabled() &&
properties_ && properties_->ClipPathClip() && object_.HasLayer()) {
}This if block is now empty.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
kDirectReasonsForEffectProperty =Shouldn't we have kWillChangeMask and kWillChangeMaskImage here as well? Maybe I misunderstand, but don't we want to set things up so that a mask can change without re-drawing everything, even if there isn't a backdrop filter?
<link rel="author" href="mailto:mas...@chromium.org">| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
kDirectReasonsForEffectProperty =Shouldn't we have kWillChangeMask and kWillChangeMaskImage here as well? Maybe I misunderstand, but don't we want to set things up so that a mask can change without re-drawing everything, even if there isn't a backdrop filter?
Masks don't normally create an effect node, and that would be a pretty big perf change. The point of this CL is to only create effect nodes for them when a backdrop filter is/will be present, to meet spec while minimizing perf impact.
<link rel="author" href="mailto:mas...@chromium.org">Claire Chambers
Ack! Lesson in why you should never blindly copy-paste. Fixed.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Code-Review | +1 |
kDirectReasonsForEffectProperty =Claire ChambersShouldn't we have kWillChangeMask and kWillChangeMaskImage here as well? Maybe I misunderstand, but don't we want to set things up so that a mask can change without re-drawing everything, even if there isn't a backdrop filter?
Masks don't normally create an effect node, and that would be a pretty big perf change. The point of this CL is to only create effect nodes for them when a backdrop filter is/will be present, to meet spec while minimizing perf impact.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
Exportable changes to web-platform-tests were detected in this CL and a pull request in the upstream repo has been made: https://github.com/web-platform-tests/wpt/pull/57550.
When this CL lands, the bot will automatically merge the PR on GitHub if the required GitHub checks pass; otherwise, ecosystem-infra@ team will triage the failures and may contact you.
WPT Export docs:
https://chromium.googlesource.com/chromium/src/+/main/docs/testing/web_platform_tests.md#Automatic-export-process
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
4 is the latest approved patch-set.
The change was submitted with unreviewed changes in the following files:
```
The name of the file: third_party/blink/renderer/platform/graphics/compositing_reasons.h
Insertions: 1, Deletions: 1.
The diff is too large to show. Please review the diff.
```
Add kAuxiliaryReasonsForBackdropRoot, fix bugs for backdrop root
This change fixes two bugs - that mask images did not create backdrop
roots even though the spec demands it - and, that will-change: clip-path
created a backdrop root even when a child backdrop filter was not
present, making will-change: clip-path an anti-optimization for clip-
path.
A test is added to ensure the behavior for masks is correct.
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |
The WPT PR for this CL has been merged upstream! https://github.com/web-platform-tests/wpt/pull/57550
| Inspect html for hidden footers to help with email filtering. To unsubscribe visit settings. |