Closed
Bug 1211487
Opened 9 years ago
Closed 9 years ago
OMTA can affect the visible region of a ContainerLayer, but doesn't adjust it
Categories
(Core :: Graphics: Layers, defect)
Core
Graphics: Layers
Tracking
()
RESOLVED
DUPLICATE
of bug 1208829
Tracking | Status | |
---|---|---|
firefox44 | --- | affected |
People
(Reporter: mstange, Unassigned)
References
Details
(Whiteboard: [gfx-noted])
Attachments
(1 file)
740 bytes,
text/html
|
Details |
Steps to reproduce:
1. Load the testcase.
2. Move your mouse over the black rectangle.
3. Watch the blue circle be clipped away when it moves downwards.
This clipping happens when the ContainerLayer for the opacity draws its contents to an intermediate surface (which is required for group opacity): The intermediate surface is sized to the visible region of the ContainerLayer, and that visible region was calculated on the main thread by FrameLayerBuilder with the pre-animation child layer transforms in mind.
APZ has a similar problem.
I think we need to add a pass to LayerManagerComposite that throws away the visible region of each ContainerLayer and recomputes it, respecting the current transforms that have been updated by OMTA and APZ.
Reporter | ||
Comment 1•9 years ago
|
||
Whiteboard: [gfx-noted]
Comment 2•9 years ago
|
||
This is fixed by the patches in bug 1208829.
Status: NEW → RESOLVED
Closed: 9 years ago
Resolution: --- → DUPLICATE
You need to log in
before you can comment on or make changes to this bug.
Description
•