Branch: refs/heads/main
Home: https://github.com/WebKit/WebKit
Commit: f5b453a8e09bc6f143b330aec80c7da77c603d91
https://github.com/WebKit/WebKit/commit/f5b453a8e09bc6f143b330aec80c7da77c603d91
Author: Said Abou-Hallawa <[email protected]>
Date: 2025-06-26 (Thu, 26 Jun 2025)
Changed paths:
M LayoutTests/TestExpectations
A LayoutTests/fast/images/resources/red-green-blue.mov
A LayoutTests/fast/images/video-as-image-expected.html
A LayoutTests/fast/images/video-as-image.html
M LayoutTests/platform/ios/TestExpectations
M LayoutTests/platform/mac/TestExpectations
M Source/WebCore/page/Page.cpp
M Source/WebCore/platform/graphics/ImageTypes.h
M Source/WebCore/platform/graphics/ShareableBitmap.cpp
M Source/WebCore/platform/graphics/cg/GraphicsContextCG.cpp
M Source/WebCore/platform/graphics/cg/NativeImageCG.cpp
M Source/WebCore/platform/graphics/cg/ShareableBitmapCG.mm
M Source/WebKit/GPUProcess/media/RemoteImageDecoderAVFProxy.cpp
M Source/WebKit/WebProcess/GPU/graphics/RemoteDisplayListRecorderProxy.cpp
Log Message:
-----------
REGRESSION(295899@main): Video as an image source does not display its frames
https://bugs.webkit.org/show_bug.cgi?id=295015
rdar://154358072
Reviewed by Simon Fraser.
Three bugs are the cause of this regression:
1. Headroom constructor is not explicit. So any numerical value including `bool`
can construct a Headroom structure.
2. A new argument to ShareableBitmapConfiguration of type Headroom was added in
295899@main. RemoteImageDecoderAVFProxy::createFrameImageAtIndex() was not
modified to include a headroom when calling ShareableBitmap::create().
Instead
a `bool` false was passed instead. This `bool` constructs a Headroom of type
Headroom::FromImage which is very special value and should not be passed from
WebProcess to GPUProcess or vice versa.
3. In 295899@main ShareableBitmap::createCGImage() was changed such that it
calls
CGImageCreateWithContentHeadroom() if m_configuration.headroom() !=
Headroom::None.
This code is now hit in the video scenario because RemoteImageDecoderAVFProxy
::createFrameImageAtIndex() passes the wrong Headroom.
The fix is to:
1. Make the constructor of Headroom explicit so we do not hit this bug again.
2. Make RemoteImageDecoderAVFProxy::createFrameImageAtIndex() pass the headroom
of the decoded image to ShareableBitmap::create().
3. Assert in ShareableBitmap and ShareableBitmapConfiguration that the passed
headroom has a valid Headroom value.
* LayoutTests/TestExpectations:
* LayoutTests/fast/images/resources/red-green-blue.mov: Added.
* LayoutTests/fast/images/video-as-image-expected.html: Added.
* LayoutTests/fast/images/video-as-image.html: Added.
* LayoutTests/platform/ios/TestExpectations:
* LayoutTests/platform/mac/TestExpectations:
* Source/WebCore/page/Page.cpp:
(WebCore::Page::updateDisplayEDRHeadroom):
* Source/WebCore/platform/graphics/ImageTypes.h:
(WebCore::Headroom::Headroom):
* Source/WebCore/platform/graphics/ShareableBitmap.cpp:
(WebCore::ShareableBitmapConfiguration::ShareableBitmapConfiguration):
(WebCore::ShareableBitmap::create):
(WebCore::ShareableBitmap::ShareableBitmap):
* Source/WebCore/platform/graphics/cg/GraphicsContextCG.cpp:
(WebCore::GraphicsContextCG::drawNativeImageInternal):
* Source/WebCore/platform/graphics/cg/NativeImageCG.cpp:
(WebCore::PlatformImageNativeImageBackend::headroom const):
* Source/WebCore/platform/graphics/cg/ShareableBitmapCG.mm:
(WebCore::ShareableBitmap::createCGImage const):
* Source/WebKit/GPUProcess/media/RemoteImageDecoderAVFProxy.cpp:
(WebKit::RemoteImageDecoderAVFProxy::createFrameImageAtIndex):
* Source/WebKit/WebProcess/GPU/graphics/RemoteDisplayListRecorderProxy.cpp:
(WebKit::RemoteDisplayListRecorderProxy::drawNativeImageInternal):
Canonical link: https://commits.webkit.org/296680@main
To unsubscribe from these emails, change your notification settings at
https://github.com/WebKit/WebKit/settings/notifications
_______________________________________________
webkit-changes mailing list
[email protected]
https://lists.webkit.org/mailman/listinfo/webkit-changes
