Timeline

Apr 24, 2020:

10:04 PM Changeset in webkit [260691] by Diego Pino Garcia

• 2 edits in trunk/Source/WebKit

[WPE][Debug] Unreviewed, fix build after r260063
​https://bugs.webkit.org/show_bug.cgi?id=211009

r260063 removed include of WebProcess.h in WebSocketProvider.cpp,
which broke unified builds in WPE Debug.

• WebProcess/Network/WebSocketProvider.cpp:

9:58 PM Changeset in webkit [260690] by Alexey Shvayka

• 90 edits
  1 move
  43 adds
  1 delete in trunk

Fix WASM Error classes and re-sync wpt/wasm/jsapi from upstream
​https://bugs.webkit.org/show_bug.cgi?id=210980

Reviewed by Keith Miller.

JSTests:

• wasm/modules/js-wasm-cycle.js:
• wasm/modules/wasm-wasm-cycle.js:
• wasm/stress/too-many-locals.js:

LayoutTests/imported/w3c:

web-platform-tests revision: 028bd8650758

• web-platform-tests/wasm/jsapi/*: Updated.

Source/JavaScriptCore:

assert_throws_js() harness, which is extensively used by wpt/wasm/jsapi tests,
was recently updated to assert that passed constructors subclass Error in
spec-perfect way.

With this patch, WebAssembly errors have Error as Prototype of their constructors
and define correct "name" and "message" properties on their prototypes, aligning JSC
with the spec [1], V8 and SpiderMonkey.

[1]: ​https://webassembly.github.io/spec/js-api/#error-objects

• runtime/JSGlobalObject.cpp:

(JSC::JSGlobalObject::init):

• wasm/js/WebAssemblyCompileErrorPrototype.cpp:

(JSC::WebAssemblyCompileErrorPrototype::finishCreation):

• wasm/js/WebAssemblyLinkErrorPrototype.cpp:

(JSC::WebAssemblyLinkErrorPrototype::finishCreation):

• wasm/js/WebAssemblyRuntimeErrorPrototype.cpp:

(JSC::WebAssemblyRuntimeErrorPrototype::finishCreation):

6:32 PM Changeset in webkit [260689] by ddkilzer@apple.com

• 2 edits in trunk/Source/WebKit

WebPasteboardProxy::getPasteboardStringsForType() and WebPasteboardProxy::readURLFromPasteboard() should check return value of SharedMemory::createHandle()
<​https://webkit.org/b/211002>

Reviewed by Wenson Hsieh.

• UIProcess/Cocoa/WebPasteboardProxyCocoa.mm:

(WebKit::WebPasteboardProxy::getPasteboardBufferForType):
(WebKit::WebPasteboardProxy::readBufferFromPasteboard):

• Check result of SharedMemory::createHandle() and return early on failure.

5:52 PM Changeset in webkit [260688] by Alan Coon

• 17 edits in branches/safari-610.1.11-branch

Cherry-pick r260486. rdar://problem/62349885

Fix MACCATALYST build failures
​https://bugs.webkit.org/show_bug.cgi?id=210815

Reviewed by Tim Horton.

Source/JavaScriptCore:

• Configurations/FeatureDefines.xcconfig:

Source/WebCore:

No new tests, no functional change.

• Configurations/FeatureDefines.xcconfig:
• platform/ios/WebVideoFullscreenControllerAVKit.mm:

Source/WebCore/PAL:

• Configurations/FeatureDefines.xcconfig:

Source/WebKit:

• Configurations/FeatureDefines.xcconfig:
• UIProcess/API/ios/WKWebViewIOS.mm: (-[WKWebView _isShowingVideoPictureInPicture]): (-[WKWebView _mayAutomaticallyShowVideoPictureInPicture]):
• UIProcess/ios/WebPageProxyIOS.mm: (WebKit::WebPageProxy::applicationDidBecomeActive):

Source/WebKitLegacy/mac:

• Configurations/FeatureDefines.xcconfig:

Source/WTF:

• wtf/PlatformEnable.h:

Tools:

• TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@260486 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5:52 PM Changeset in webkit [260687] by Alan Coon

• 22 edits in branches/safari-610.1.11-branch

Cherry-pick r260412. rdar://problem/62349885

Fix build failures when video fullscreen and picture-in-picture is disabled
​https://bugs.webkit.org/show_bug.cgi?id=210777

Reviewed by Eric Carlson.

Source/JavaScriptCore:

• Configurations/FeatureDefines.xcconfig:

Source/WebCore:

Wrap video fullscreen and picture-in-picture related code with "#if ENABLE(VIDEO_PRESENTATION_MODE)".

• Configurations/FeatureDefines.xcconfig:
• platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm: (WebCore::MediaPlayerPrivateAVFoundationObjC::updateVideoLayerGravity):
• platform/graphics/avfoundation/objc/VideoLayerManagerObjC.h:
• platform/graphics/avfoundation/objc/VideoLayerManagerObjC.mm: (WebCore::VideoLayerManagerObjC::setVideoLayer): (WebCore::VideoLayerManagerObjC::requiresTextTrackRepresentation const): (WebCore::VideoLayerManagerObjC::syncTextTrackBounds): (WebCore::VideoLayerManagerObjC::setTextTrackRepresentation):

Source/WebCore/PAL:

• Configurations/FeatureDefines.xcconfig:

Source/WebKit:

Wrap video fullscreen and picture-in-picture related code with "#if ENABLE(VIDEO_PRESENTATION_MODE)".

• Configurations/FeatureDefines.xcconfig:
• GPUProcess/media/RemoteMediaPlayerProxy.cpp: (WebKit::RemoteMediaPlayerProxy::setVideoFullscreenGravity): (WebKit::RemoteMediaPlayerProxy::updateVideoFullscreenInlineImage): (WebKit::RemoteMediaPlayerProxy::setVideoFullscreenMode): (WebKit::RemoteMediaPlayerProxy::videoFullscreenStandbyChanged): (WebKit::RemoteMediaPlayerProxy::setBufferingPolicy):
• GPUProcess/media/RemoteMediaPlayerProxy.h:
• GPUProcess/media/RemoteMediaPlayerProxy.messages.in:
• GPUProcess/media/cocoa/RemoteMediaPlayerProxyCocoa.mm: (WebKit::RemoteMediaPlayerProxy::prepareForPlayback):
• WebProcess/GPU/media/MediaPlayerPrivateRemote.cpp: (WebKit::MediaPlayerPrivateRemote::prepareForPlayback):

Source/WebKitLegacy/mac:

• Configurations/FeatureDefines.xcconfig:

Source/WTF:

• wtf/PlatformEnable.h:

Tools:

• TestWebKitAPI/Configurations/FeatureDefines.xcconfig:

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@260412 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5:52 PM Changeset in webkit [260686] by Alan Coon

• 3 edits
  2 adds in branches/safari-610.1.11-branch

Cherry-pick r260450. rdar://problem/62350884

[Async overflow scroll] Overflow that's hidden on one axis is scrollable on that axis
​https://bugs.webkit.org/show_bug.cgi?id=210771
<rdar://problem/62080331>

Reviewed by Tim Horton.

Source/WebCore:

eventCanScrollContents() should check the presence of enabled scrollbars, like
ScrollAnimator::handleWheelEvent() does.

Test: fast/scrolling/mac/async-scroll-overflow-hidden-on-one-axis.html

• page/scrolling/ScrollingTreeScrollingNode.cpp: (WebCore::ScrollingTreeScrollingNode::eventCanScrollContents const):

LayoutTests:

• fast/scrolling/mac/async-scroll-overflow-hidden-on-one-axis-expected.txt: Added.
• fast/scrolling/mac/async-scroll-overflow-hidden-on-one-axis.html: Added.

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@260450 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5:52 PM Changeset in webkit [260685] by Alan Coon

• 2 edits in branches/safari-610.1.11-branch/Source/WebKit

Cherry-pick r260390. rdar://problem/62351354

When SpeculativeLoadManager is destroyed, properly clean up its PendingFrameLoads
​https://bugs.webkit.org/show_bug.cgi?id=210759
<rdar://problem/62056856>

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-20
Reviewed by Darin Adler.

Recent work on the resourceLoadStatistics layout tests increased the amount we swap out the WebsiteDataStore.
When this happens, the NetworkSession is eventually destroyed in the NetworkProcess, sometimes when running the next test.
An assertion was firing in the PendingFrameLoad destructor because it hadn't been marked as complete when it was destroyed.
Rather than remove the assertion, when we destroy the SpeculativeLoadManager (which only happens when a WebsiteDataStore
is destroyed) during a speculative pending frame load, just mark the pending frame load as complete because it is being cancelled.
Marking the pending frame load as complete can tell the SpeculativeLoadManager to mutate m_pendingFrameLoads, which we don't want
to do while iterating, so copy the RefPtrs into a Vector first then iterate that to get them all.

This fixes an assertion that was sometimes hit in http/tests/resourceLoadStatistics/strip-referrer-to-origin-for-prevalent-subresource-redirects-database.html
but only after running other tests that had initiated speculative pending frame loads. This was ostensibly started by r260322 but is quite unrelated.

• NetworkProcess/cache/NetworkCacheSpeculativeLoadManager.cpp: (WebKit::NetworkCache::SpeculativeLoadManager::~SpeculativeLoadManager):

git-svn-id: ​https://svn.webkit.org/repository/webkit/trunk@260390 268f45cc-cd09-0410-ab3c-d52691b4dbfc

5:51 PM Changeset in webkit [260684] by Chris Dumez

• 11 edits in trunk/Source

[iOS] Unable to sign up on twitter.com
​https://bugs.webkit.org/show_bug.cgi?id=211003
<rdar://problem/58804852>

Reviewed by Darin Adler.

Source/WebCore:

This is similar to the bug we had on nytimes.com and that was fixed in
r258767. However, instead of a 'resize' event, it is a 'change' event
on a MediaQueryList that is getting twitter.com in a bad state.

The issue is that when we home out of Safari, SpringBoard takes does
a snapshot sequence at various sizes / orientations and this causes
many JS events to get fired (e.g. 'resize', 'orientationchange',
'change', ...), which can get some sites in a bad state. To address
the issue, we now prevent firing of ALL JS events during the
SpringBoard snapshot, instead of merely preventing the 'resize' ones.

• dom/EventTarget.cpp:

(WebCore::EventTarget::fireEventListeners):

• page/FrameView.cpp:

(WebCore::FrameView::sendResizeEventIfNeeded):

• page/Page.h:

(WebCore::Page::shouldFireEvents const):
(WebCore::Page::setShouldFireEvents):
(WebCore::Page::shouldFireResizeEvents const): Deleted.
(WebCore::Page::setShouldFireResizeEvents): Deleted.

Source/WebKit:

• UIProcess/WebPageProxy.cpp:
• UIProcess/WebPageProxy.h:
• UIProcess/ios/WKApplicationStateTrackingView.mm:

(-[WKApplicationStateTrackingView _willBeginSnapshotSequence]):
(-[WKApplicationStateTrackingView _didCompleteSnapshotSequence]):

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::setShouldFireEvents):
(WebKit::WebPage::setShouldFireResizeEvents): Deleted.

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:

5:34 PM Changeset in webkit [260683] by sbarati@apple.com

• 19 edits
  1 add in trunk

Return BigInt32 whenever we can
​https://bugs.webkit.org/show_bug.cgi?id=210922

Reviewed by Yusuke Suzuki.

JSTests:

• microbenchmarks/sunspider-sha1-big-int.js: Added.

(hex_sha1):
(b64_sha1):
(str_sha1):
(hex_hmac_sha1):
(b64_hmac_sha1):
(str_hmac_sha1):
(bigIntToInt32):
(sha1_vm_test):
(core_sha1):
(sha1_ft):
(sha1_kt):
(core_hmac_sha1):
(safe_add):
(rol):
(str2binb):
(binb2hex):
(binb2b64):
(run):

Source/JavaScriptCore:

This patch makes it so our runtime functions for big int math on heap
big ints converts the result to a big int 32 when possible.

The inspiration for this patch came from converting SunSpider's sha1 benchmark to
using big ints. I found that that original implementation of big int 32
was a ~35% slowdown here. This patch speeds it up by 86% from ToT, and
36% faster than before big int 32 was introduced.

To make this sound in the DFG/FTL, we are currently reporting that all
HeapBigInt math ops return SpecBigInt, instead of SpecHeapBigInt.
However, we want to do better in a follow up. We need some kind of profiling
system where we determine if we should speculate if the result is big int
32, a heap big int, or both:
​https://bugs.webkit.org/show_bug.cgi?id=210982

• dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):

• dfg/DFGFixupPhase.cpp:

(JSC::DFG::FixupPhase::fixupNode):

• dfg/DFGOperations.cpp:
• dfg/DFGOperations.h:
• dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileValueBitNot):
(JSC::DFG::SpeculativeJIT::compileValueBitwiseOp):
(JSC::DFG::SpeculativeJIT::compileValueLShiftOp):
(JSC::DFG::SpeculativeJIT::compileValueBitRShift):
(JSC::DFG::SpeculativeJIT::compileValueAdd):
(JSC::DFG::SpeculativeJIT::compileValueSub):
(JSC::DFG::SpeculativeJIT::compileValueMul):
(JSC::DFG::SpeculativeJIT::compileValueDiv):
(JSC::DFG::SpeculativeJIT::compileValueMod):
(JSC::DFG::SpeculativeJIT::compileValuePow):

• jit/JITOperations.cpp:
• jsc.cpp:

(functionCreateBigInt32):

• runtime/BigIntConstructor.cpp:

(JSC::toBigInt):
(JSC::callBigIntConstructor):

• runtime/CommonSlowPaths.cpp:

(JSC::SLOW_PATH_DECL):

• runtime/JSBigInt.cpp:

(JSC::JSBigInt::exponentiateHeap):
(JSC::JSBigInt::multiplyHeap):
(JSC::JSBigInt::divideHeap):
(JSC::JSBigInt::unaryMinusHeap):
(JSC::JSBigInt::remainderHeap):
(JSC::JSBigInt::incHeap):
(JSC::JSBigInt::decHeap):
(JSC::JSBigInt::addHeap):
(JSC::JSBigInt::subHeap):
(JSC::JSBigInt::bitwiseAndHeap):
(JSC::JSBigInt::bitwiseOrHeap):
(JSC::JSBigInt::bitwiseXorHeap):
(JSC::JSBigInt::leftShiftHeap):
(JSC::JSBigInt::signedRightShiftHeap):
(JSC::JSBigInt::bitwiseNotHeap):
(JSC::JSBigInt::absoluteAdd):
(JSC::JSBigInt::absoluteSub):
(JSC::JSBigInt::parseInt):
(JSC::JSBigInt::exponentiate): Deleted.
(JSC::JSBigInt::multiply): Deleted.
(JSC::JSBigInt::divide): Deleted.
(JSC::JSBigInt::unaryMinus): Deleted.
(JSC::JSBigInt::remainder): Deleted.
(JSC::JSBigInt::inc): Deleted.
(JSC::JSBigInt::dec): Deleted.
(JSC::JSBigInt::add): Deleted.
(JSC::JSBigInt::sub): Deleted.
(JSC::JSBigInt::bitwiseAnd): Deleted.
(JSC::JSBigInt::bitwiseOr): Deleted.
(JSC::JSBigInt::bitwiseXor): Deleted.
(JSC::JSBigInt::leftShift): Deleted.
(JSC::JSBigInt::signedRightShift): Deleted.
(JSC::JSBigInt::bitwiseNot): Deleted.

• runtime/JSBigInt.h:
• runtime/JSCJSValue.h:

(JSC::jsBigInt32):

• runtime/JSCJSValueInlines.h:

(JSC::JSValue::JSValue):

• runtime/Operations.cpp:

(JSC::jsAddSlowCase):

• runtime/Operations.h:

(JSC::jsSub):
(JSC::jsMul):
(JSC::jsDiv):
(JSC::jsInc):
(JSC::jsDec):
(JSC::jsBitwiseNot):
(JSC::shift):
(JSC::bitwiseBinaryOp):

Source/WebCore:

• bindings/js/SerializedScriptValue.cpp:

(WebCore::CloneDeserializer::readBigInt):

4:56 PM Changeset in webkit [260682] by ysuzuki@apple.com

• 5 edits in trunk/Source

[WTF] allThreads registration is racy with allThreads unregistration
​https://bugs.webkit.org/show_bug.cgi?id=210995
<rdar://problem/61609690>

Reviewed by Keith Miller.

Source/WebCore:

• page/cocoa/ResourceUsageThreadCocoa.mm:

(WebCore::ResourceUsageThread::platformCollectCPUData):

Source/WTF:

There is a race between registering a thread to allThreads and unregistering a thread from allThreads.

1. Caller: A new thread is created, but not registering it to allThreads yet.
2. Thread: The thread is running.
3. Thread: The thread finishes its execution before the thread is registered into allThreads.
4. Thread: The thread unregisters itself from allThreads.
5. Caller: Registers the new thread to allThreads after it already finished its execution.
6. The thread is never removed from allThreads.

This patch adds m_didUnregisterFromAllThreads flag to Thread, and add the thread to allThreads only when this flag is false.

Covered by LayoutTests/inspector/cpu-profiler/threads.html.

• wtf/Threading.cpp:

(WTF::Thread::create):
(WTF::Thread::didExit):

• wtf/Threading.h:

(WTF::Thread::Thread):

4:53 PM Changeset in webkit [260681] by jonlee@apple.com

• 2 edits in trunk/PerformanceTests

MotionMark: add link to bug in about page

Unreviewed.

• MotionMark/about.html:

4:35 PM Changeset in webkit [260680] by commit-queue@webkit.org

• 6 edits in trunk

[GTK][WPE][JSCOnly] compile error when -DWTF_CPU_ARM64_CORTEXA53=ON set for arm64
​https://bugs.webkit.org/show_bug.cgi?id=197192

Patch by Michael Catanzaro <Michael Catanzaro> on 2020-04-24
Reviewed by Yusuke Suzuki.

.:

• Source/cmake/OptionsCommon.cmake:

Source/JavaScriptCore:

This workaround is supposed to fix WebKit on old Cortex A53 CPUs, but it has been broken
since 2018, and people would like to use WebKit on modern Cortex A53. If anyone using WebKit
on the original hardware wants to fix and reimplement the workaround, feel free.

• assembler/ARM64Assembler.h:

(JSC::ARM64Assembler::adrp):
(JSC::ARM64Assembler::madd):
(JSC::ARM64Assembler::msub):
(JSC::ARM64Assembler::smaddl):
(JSC::ARM64Assembler::smsubl):
(JSC::ARM64Assembler::umaddl):
(JSC::ARM64Assembler::umsubl):
(JSC::ARM64Assembler::nopCortexA53Fix835769): Deleted.
(JSC::ARM64Assembler::nopCortexA53Fix843419): Deleted.

• offlineasm/arm64.rb:
• offlineasm/instructions.rb:

4:32 PM Changeset in webkit [260679] by commit-queue@webkit.org

• 5 edits in trunk

REGRESSION(260485) Payment requests don't do anything
​https://bugs.webkit.org/show_bug.cgi?id=210997
Source/WTF:

<rdar://problem/62275343>

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-24
Reviewed by Darin Adler.

We were giving the PKPaymentRequest an NSArray<NSString *> instead of NSArray<NSURL *>.
This was a problem with all uses of createNSArray with Vector<URL>.
Now it's fixed with a test.

• wtf/URL.h:
• wtf/cocoa/URLCocoa.mm:

(WTF::makeNSArrayElement):
(WTF::makeVectorElement):

Tools:

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-24
Reviewed by Darin Adler.

• TestWebKitAPI/Tests/WTF/cocoa/URLExtras.mm:

(TestWebKitAPI::TEST):

4:18 PM Changeset in webkit [260678] by Wenson Hsieh

• 4 edits in trunk

Make some more adjustments to TextManipulationController's paragraph boundary heuristic
​https://bugs.webkit.org/show_bug.cgi?id=210993
<rdar://problem/61571299>

Reviewed by Tim Horton.

Source/WebCore:

Adjust the heuristic added in r260583 to account for a few more common scenarios where we currently consider
text as a part of the same paragraph. This can lead to many issues during text manipulation where text is moved
between these elements, when it should not be.

The new scenarios include block and inline-block links, as well as button elements.

Test: TextManipulation.StartTextManipulationTreatsInlineBlockLinksAndButtonsAsParagraphs

• editing/TextManipulationController.cpp:

(WebCore::TextManipulationController::observeParagraphs):

Additionally rename "paragraph boundary element" to "item boundary element", to avoid colliding with the
existing notion of paragraph boundaries in editing code.

Tools:

Add a new API test with buttons and links styled with display: inline-block;. Additionally, rebaseline an
existing API test that exercises inline-block list items.

• TestWebKitAPI/Tests/WebKitCocoa/TextManipulation.mm:

3:49 PM Changeset in webkit [260677] by ddkilzer@apple.com

• 1 edit
  1 delete in trunk/LayoutTests/imported/w3c

REGRESSION (r223327): Remove merge conflict file

• web-platform-tests/webrtc/simplecall.html.orig: Remove.

3:41 PM Changeset in webkit [260676] by Megan Gardner

• 5 edits in trunk/Source

Make LEGACY_PDF_SUPPORT feature flag.
​https://bugs.webkit.org/show_bug.cgi?id=210868
<rdar://problem/62199847>

Reviewed by Tim Horton.

Source/WebKitLegacy/mac:

• WebView/WebFrameView.mm:

(+[WebFrameView _viewTypesAllowImageTypeOmission:]):

• WebView/WebView.mm:

Source/WTF:

• wtf/PlatformHave.h:

3:29 PM Changeset in webkit [260675] by Chris Dumez

• 2 edits in trunk/Tools

[iOS] Always run WKTR's WKWebViews at foreground priority
​https://bugs.webkit.org/show_bug.cgi?id=210991

Reviewed by Keith Miller.

Always run WKTR's WKWebViews at foreground priority. This makes sure that they don't suspend and
run at foreground priority on iOS, even if those views are not visible on screen.
This is an issue to address flakiness on the bots.

• WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::initializeWebViewConfiguration):

2:53 PM Changeset in webkit [260674] by ysuzuki@apple.com

• 3 edits in trunk/Source/JavaScriptCore

[JSC] Fix DataFormatJSBigInt32 missing part
​https://bugs.webkit.org/show_bug.cgi?id=210986

Reviewed by Mark Lam.

Add missing part of DataFormatJSBigInt32 implementation.

• bytecode/DataFormat.h:

(JSC::dataFormatToString):

• dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::checkGeneratedTypeForToInt32):

2:51 PM Changeset in webkit [260673] by Alan Coon

• 8 edits in branches/safari-610.1.11-branch/Source

Versioning.

2:45 PM Changeset in webkit [260672] by chris.reid@sony.com

• 14 edits
  3 adds in trunk

[Win] Bundle Inspector Resources in Release builds
​https://bugs.webkit.org/show_bug.cgi?id=210942

Reviewed by Fujii Hironori.

.:

Add ENABLE_WEBINSPECTORUI so the resource copy can be disabled
on platforms without inspector frontends.

• CMakeLists.txt:
• Source/CMakeLists.txt:
• Source/PlatformWin.cmake:
• Source/cmake/OptionsJSCOnly.cmake:
• Source/cmake/OptionsPlayStation.cmake:

Source/WebCore:

• CMakeLists.txt:

Source/WebInspectorUI:

Add CMake files for copying inspector resources

• CMakeLists.txt: Added.
• PlatformGTK.cmake: Added.
• PlatformWin.cmake: Added.

Source/WebKit:

Move CMake logic for the inspector resource copy script to
Source/WebInspectorUI so it can be shared with Win and other platforms.

• InspectorGResources.cmake:
• PlatformWPE.cmake:
• PlatformWin.cmake:

2:42 PM Changeset in webkit [260671] by graouts@webkit.org

• 19 edits in trunk

[Web Animations] Ensure calling Web Animations APIs override future CSS Animations style properties
​https://bugs.webkit.org/show_bug.cgi?id=210988

Reviewed by Dean Jackson.

LayoutTests/imported/w3c:

Mark all WPT tests related to Web Animations APIs overrides for CSS Animations as PASS, save for one
failing assertion which is caused by AnimationTimeline::updateCSSAnimations() not updating the animation
when the underlying @keyframes rule changed. This is due to Animation::animationsMatch() not checking on
the actual keyframes, tracked by webkit.org/b/210989.

• web-platform-tests/css/css-animations/AnimationEffect-updateTiming.tentative-expected.txt:
• web-platform-tests/css/css-animations/CSSAnimation-effect.tentative-expected.txt:
• web-platform-tests/css/css-animations/CSSAnimation-pausing.tentative-expected.txt:
• web-platform-tests/css/css-animations/KeyframeEffect-setKeyframes.tentative-expected.txt:

Source/WebCore:

The CSS Animations Level 2 spec specifies how the Web Animations APIs and the CSS Animations style
properties should interact in ​https://drafts.csswg.org/css-animations-2/#animations. This patch
implements the specified behavior and this is reflected by progress on the relevant WPT tests.

The gist of this change is that once a Web Animations API is called on an animation created using
CSS Animations, any changes made to related CSS Animations style properties on the target element
will be ignored so that the overrides applied via the Web Animations API remain in effect.

For instance, calling pause() or play() in a way that changes the playback state of the CSS Animation
will mean that future changes to the CSS animation-play-state property are ignored.

To do this we make more IDL properties and methods use dedicated methods to distinguish between the
bindings entry-point and internal usage of the same methods to integrate the behavior only when the
API itself is being used.

• animation/AnimationEffect.cpp:

(WebCore::AnimationEffect::getBindingsTiming const): Ensure we flush styles when animation.effect.getTiming()
is called.
(WebCore::AnimationEffect::getBindingsComputedTiming const): Ensure we flush styles when
animation.effect.getComputedTiming() is called.
(WebCore::AnimationEffect::bindingsUpdateTiming): Notify the associated CSSAnimation object, if any, when
animation.effect.updateTiming() is called such that the CSSAnimation may apply the relevant overrides.

• animation/AnimationEffect.h:
• animation/AnimationEffect.idl:
• animation/CSSAnimation.cpp:

(WebCore::CSSAnimation::syncPropertiesWithBackingAnimation): Only apply new values of CSS Animations style
properties if there are no overrides for them resulting from calling related Web Animations APIs.
(WebCore::CSSAnimation::bindingsPlay): Mark animation-play-state as overridden if play() is called.
(WebCore::CSSAnimation::bindingsPause): Mark animation-play-state as overridden if pause() is called.
(WebCore::CSSAnimation::setBindingsEffect): Mark all animation style properties, except for animation-name
and animation-play-state as overridden if animation.effect is set.
(WebCore::CSSAnimation::setBindingsStartTime): Mark animation-play-state as overridden if animation.startTime
is set.
(WebCore::CSSAnimation::bindingsReverse): Mark animation-play-state as overridden if reverse() is called.
(WebCore::CSSAnimation::effectTimingWasUpdatedUsingBindings): Mark each CSS property associated with a key
found on the timing object passed to animation.effect.updateTiming() as overridden.
(WebCore::CSSAnimation::effectKeyframesWereSetUsingBindings): Mark animation-timing-function as overridden
if animation.effect.setKeyframes() is called.

• animation/CSSAnimation.h:
• animation/DeclarativeAnimation.cpp:

(WebCore::DeclarativeAnimation::bindingsStartTime const):
(WebCore::DeclarativeAnimation::setBindingsStartTime):
(WebCore::DeclarativeAnimation::startTime const): Deleted.
(WebCore::DeclarativeAnimation::setStartTime): Deleted.

• animation/DeclarativeAnimation.h:
• animation/KeyframeEffect.cpp:

(WebCore::KeyframeEffect::getBindingsKeyframes): Ensure we flush styles when animation.effect.getKeyframes()
is called.
(WebCore::KeyframeEffect::getKeyframes): Only use the CSS-originated animation path if we don't have JS-originated
keyframes.
(WebCore::KeyframeEffect::setBindingsKeyframes): Notify the associated CSSAnimation object, if any, when
animation.effect.setKeyframes() is called such that the CSSAnimation may apply the relevant overrides.
(WebCore::KeyframeEffect::processKeyframes): Correctly return early if part of the processing yields an exception.

• animation/KeyframeEffect.h:
• animation/KeyframeEffect.idl:
• animation/WebAnimation.cpp:

(WebCore::WebAnimation::setBindingsEffect):
(WebCore::WebAnimation::setBindingsStartTime):
(WebCore::WebAnimation::bindingsReverse):

• animation/WebAnimation.h:

(WebCore::WebAnimation::bindingsEffect const):
(WebCore::WebAnimation::bindingsStartTime const):

• animation/WebAnimation.idl:

2:41 PM Changeset in webkit [260670] by Alexey Shvayka

• 15 edits
  24 adds
  1 delete in trunk/LayoutTests/imported/w3c

Re-sync wpt/WebIDL/ecmascript-binding and wpt/custom-elements/htmlconstructor from upstream
​https://bugs.webkit.org/show_bug.cgi?id=210984

Reviewed by Darin Adler.

web-platform-tests revision: 39e9c51041a1

• web-platform-tests/WebIDL/ecmascript-binding/*: Updated.
• web-platform-tests/custom-elements/htmlconstructor/*: Updated.

2:11 PM Changeset in webkit [260669] by Chris Dumez

• 2 edits in trunk/Source/WebCore

ASSERTION FAILED: m_wrapper under HTMLMediaElement::setIsPlayingToWirelessTarget
​https://bugs.webkit.org/show_bug.cgi?id=210983
<rdar://problem/61611994>

Reviewed by Eric Carlson.

The issue was that we were trying to fire a JS event as a result of ActiveDOMObject::stop()
getting called, which is not allowed. To address the issue, we avoid firing the event if
the context is already stopped.

No new tests, already covered by:
media/modern-media-controls/placard-support/placard-support-airplay-fullscreen.html

• html/HTMLMediaElement.cpp:

(WebCore::HTMLMediaElement::setIsPlayingToWirelessTarget):

2:08 PM Changeset in webkit [260668] by Kate Cheney

• 25 edits
  12 adds in trunk

Removing website data for a domain should delete corresponding ITP entry
​https://bugs.webkit.org/show_bug.cgi?id=210864
<rdar://problem/59473193>

Reviewed by John Wilander.

Source/WebKit:

Tests: http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration-database.html

http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration.html
http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries-database.html
http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries.html
http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads-database.html
http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads.html

In the database store: deletes domain from the ObservedDomains table
when website data is deleted for that domain. This deletes every
instance of the domainID in the database due to cascading deletions.

In the memory store: deletes every instance of the domain in the
statistics map, which will update the plist.

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:

(WebKit::ResourceLoadStatisticsDatabaseStore::removeDataForDomain):
(WebKit::ResourceLoadStatisticsDatabaseStore::domainIDExistsInDatabase):
Needed a new function to check for the existence of domainID in any
table. Existing queries rely on the ObservedDomains entry, which means
testing using those could result in a false positive if the domainID
was deleted from ObservedDomains but is floating around in another
table.

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.h:
• NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.cpp:

(WebKit::ResourceLoadStatisticsMemoryStore::removeDataForDomain):

• NetworkProcess/Classifier/ResourceLoadStatisticsMemoryStore.h:
• NetworkProcess/Classifier/ResourceLoadStatisticsStore.h:
• NetworkProcess/Classifier/WebResourceLoadStatisticsStore.cpp:

(WebKit::WebResourceLoadStatisticsStore::domainIDExistsInDatabase):
(WebKit::WebResourceLoadStatisticsStore::removeDataForDomain):

• NetworkProcess/Classifier/WebResourceLoadStatisticsStore.h:
• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::domainIDExistsInDatabase):
(WebKit::NetworkProcess::deleteWebsiteDataForOrigins):
Pass RegistrableDomains vector to deleteWebsiteDataForOrigins,
captured based on the WebsiteData display name.

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkProcess.messages.in:
• UIProcess/API/C/WKWebsiteDataStoreRef.cpp:

(WKWebsiteDataStoreRemoveITPDataForDomain):
(WKWebsiteDataStoreDoesStatisticsDomainIDExistInDatabase):

• UIProcess/API/C/WKWebsiteDataStoreRef.h:
• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::deleteWebsiteDataForOrigins):
(WebKit::NetworkProcessProxy::domainIDExistsInDatabase):

• UIProcess/Network/NetworkProcessProxy.h:
• UIProcess/WebsiteData/WebsiteDataStore.cpp:

(WebKit::WebsiteDataStore::removeData):
(WebKit::WebsiteDataStore::domainIDExistsInDatabase):

• UIProcess/WebsiteData/WebsiteDataStore.h:

Tools:

Created 2 new APIs for testing. One to mimic clearing website data
for a domain, and one to check if the domain exists in the database
after a deletion was requested.

• WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
• WebKitTestRunner/InjectedBundle/TestRunner.cpp:

(WTR::TestRunner::doesStatisticsDomainIDExistInDatabase):
(WTR::TestRunner::domainIDExistsInDatabase): Deleted.

• WebKitTestRunner/InjectedBundle/TestRunner.h:
• WebKitTestRunner/TestController.cpp:

(WTR::TestController::doesStatisticsDomainIDExistInDatabase):
(WTR::TestController::domainIDExistsInDatabase): Deleted.

• WebKitTestRunner/TestController.h:
• WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

LayoutTests:

Inserts values into the ITP database then tests removing the domain's
website data will remove all instances of that domain in the ITP
database.

Database store expectations will reflect the result of the new
domainIDExistsInDatabase function. Memory store tests will use the
dump output which should not include the deleted information.

• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration-database-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration-database.html: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-link-decoration.html: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries-database-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries-database.html: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-mixed-statistics-entries.html: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads-database-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads-database.html: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads-expected.txt: Added.
• http/tests/resourceLoadStatistics/remove-website-data-for-origin-deletes-third-party-script-loads.html: Added.

2:02 PM Changeset in webkit [260667] by ddkilzer@apple.com

• 7 edits
  1 copy in trunk/Source/WebKit

Use CocoaImage platform abstraction for NSImage/UIImage
<​https://webkit.org/b/210974>

Reviewed by Darin Adler.

• Platform/cocoa/CocoaImage.h: Add.
• Define CocoaImage here for cross-platform use. Don't use OBJC_CLASS() here because this is an Objective-C header.
• UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView takeSnapshotWithConfiguration:completionHandler:]):

• Combine separate platform-specific methods into one method.
• UIProcess/API/Cocoa/_WKActivatedElementInfo.mm:
• Combine separate platform-specific instance variables into one instance variable.

(-[_WKActivatedElementInfo image]):

• Combine separate methods into one platform-specific method.
• UIProcess/QuickLookThumbnailLoader.h:
• Move cross-platform definition to CocoaImge.h.
• UIProcess/QuickLookThumbnailLoader.mm:
• Drive-by fix of soft-linking header include order.
• WebKit.xcodeproj/project.pbxproj:
• Add CocoaImage.h to the project.
• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInNodeHandle.mm:

(-[WKWebProcessPlugInNodeHandle renderedImageWithOptions:]):
(-[WKWebProcessPlugInNodeHandle renderedImageWithOptions:width:]):

• Combine separate platform-specific methods into one method.

1:57 PM Changeset in webkit [260666] by ddkilzer@apple.com

• 4 edits in trunk/Source/WebKit

IPC::Decoder should use create() pattern
<​https://webkit.org/b/210949>
<rdar://problem/62144409>

Reviewed by Geoffrey Garen.

• Platform/IPC/Decoder.cpp:

(IPC::Decoder::create): Add implementation. Returns nullptr if
Decoder constructor returns an invalid object.
(IPC::Decoder::Decoder): Mark invalid if m_buffer is not 64-bit
aligned.
(IPC::Decoder::unwrapForTesting): Switch to Decoder::create().

• Platform/IPC/Decoder.h:

(IPC::Decoder::create): Add declaration.
(IPC::Decoder::Decoder): Make explicit. (Can't be made private
since we use std::unique_ptr<Decoder>.)

• Platform/IPC/cocoa/ConnectionCocoa.mm:

(IPC::createMessageDecoder): Switch to Decoder::create().

1:33 PM Changeset in webkit [260665] by Russell Epstein

• 1 copy in tags/Safari-609.2.9.0.2

Tag Safari-609.2.9.0.2.

1:11 PM Changeset in webkit [260664] by ysuzuki@apple.com

• 2 edits in trunk/Source/JavaScriptCore

Unreviewed, build fix in Windows
​https://bugs.webkit.org/show_bug.cgi?id=210892

Windows MSVC does not have proper understanding of IGNORE_RETURN_TYPE_WARNINGS_BEGIN.

• runtime/JSBigInt.h:

(JSC::invertBigIntCompareResult):

1:05 PM Changeset in webkit [260663] by timothy_horton@apple.com

• 5 edits in trunk/Source

iPad: "Pocket City" interaction does not work with trackpad
​https://bugs.webkit.org/show_bug.cgi?id=210985
<rdar://problem/62273077>

Reviewed by Wenson Hsieh.

Source/WebCore:

• platform/RuntimeApplicationChecks.h:
• platform/cocoa/RuntimeApplicationChecksCocoa.mm:

(WebCore::IOSApplication::isPocketCity):

Source/WebKit:

• UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView shouldUseMouseGestureRecognizer]):
Add another app to the list who fall back to touch event synthesis until rebuilt.

1:04 PM Changeset in webkit [260662] by graouts@webkit.org

• 37 edits
  18 adds
  1 delete in trunk

Update the css/css-animations WPT tests
​https://bugs.webkit.org/show_bug.cgi?id=210964

Reviewed by Dean Jackson.

LayoutTests/imported/w3c:

• resources/import-expectations.json:
• resources/resource-files.json:
• web-platform-tests/css/css-animations/AnimationEffect-updateTiming.tentative-expected.txt: Added.
• web-platform-tests/css/css-animations/AnimationEffect-updateTiming.tentative.html: Added.
• web-platform-tests/css/css-animations/CSSAnimation-compositeOrder.tentative-expected.txt:
• web-platform-tests/css/css-animations/CSSAnimation-compositeOrder.tentative.html:
• web-platform-tests/css/css-animations/CSSAnimation-effect.tentative-expected.txt:
• web-platform-tests/css/css-animations/CSSAnimation-effect.tentative.html:
• web-platform-tests/css/css-animations/CSSAnimation-finished.tentative.html:
• web-platform-tests/css/css-animations/CSSAnimation-pausing.tentative-expected.txt:
• web-platform-tests/css/css-animations/CSSAnimation-pausing.tentative.html:
• web-platform-tests/css/css-animations/CSSAnimation-ready.tentative.html:
• web-platform-tests/css/css-animations/Document-getAnimations.tentative-expected.txt:
• web-platform-tests/css/css-animations/Document-getAnimations.tentative.html:
• web-platform-tests/css/css-animations/KeyframeEffect-getKeyframes.tentative-expected.txt:
• web-platform-tests/css/css-animations/KeyframeEffect-getKeyframes.tentative.html:
• web-platform-tests/css/css-animations/KeyframeEffect-setKeyframes.tentative-expected.txt: Added.
• web-platform-tests/css/css-animations/KeyframeEffect-setKeyframes.tentative.html: Added.
• web-platform-tests/css/css-animations/animation-base-response-001-expected.txt: Added.
• web-platform-tests/css/css-animations/animation-base-response-001.html: Added.
• web-platform-tests/css/css-animations/animation-base-response-002-expected.txt: Added.
• web-platform-tests/css/css-animations/animation-base-response-002.html: Added.
• web-platform-tests/css/css-animations/animation-base-response-003-expected.txt: Added.
• web-platform-tests/css/css-animations/animation-base-response-003.html: Added.
• web-platform-tests/css/css-animations/animation-base-response-004-expected.txt: Added.
• web-platform-tests/css/css-animations/animation-base-response-004.html: Added.
• web-platform-tests/css/css-animations/animation-important-001-expected.txt: Added.
• web-platform-tests/css/css-animations/animation-important-001.html: Added.
• web-platform-tests/css/css-animations/animation-important-002-expected.html: Added.
• web-platform-tests/css/css-animations/animation-important-002.html: Added.
• web-platform-tests/css/css-animations/event-dispatch.tentative.html:
• web-platform-tests/css/css-animations/event-order.tentative-expected.txt:
• web-platform-tests/css/css-animations/event-order.tentative.html:
• web-platform-tests/css/css-animations/historical.html:
• web-platform-tests/css/css-animations/keyframes-remove-documentElement-crash-expected.txt: Removed.
• web-platform-tests/css/css-animations/support/testcommon.js:

(assert_frames_equal):
(assert_frame_lists_equal):

• web-platform-tests/css/css-animations/w3c-import.log:

Source/WebKitLegacy/mac:

Expose the CSSCustomPropertiesAndValues experimental feature such that it may be set in DumpRenderTree.

• WebView/WebPreferenceKeysPrivate.h:
• WebView/WebPreferences.mm:

(-[WebPreferences CSSCustomPropertiesAndValuesEnabled]):
(-[WebPreferences setCSSCustomPropertiesAndValuesEnabled:]):

• WebView/WebPreferencesPrivate.h:
• WebView/WebView.mm:

Source/WebKitLegacy/win:

Expose the CSSCustomPropertiesAndValues experimental feature such that it may be set in DumpRenderTree.

• Interfaces/IWebPreferencesPrivate.idl:
• WebPreferenceKeysPrivate.h:
• WebPreferences.cpp:

(WebPreferences::initializeDefaultSettings):

• WebPreferences.h:
• WebView.cpp:

(WebView::notifyPreferencesChanged):

Tools:

Turn on the CSSCustomPropertiesAndValues experimental feature which a new test relies on.

• DumpRenderTree/mac/DumpRenderTree.mm:

(enableExperimentalFeatures):

• DumpRenderTree/win/DumpRenderTree.cpp:

(enableExperimentalFeatures):

LayoutTests:

• TestExpectations:
• platform/ios/imported/w3c/web-platform-tests/css/css-animations/animation-base-response-001-expected.txt: Added.

12:58 PM Changeset in webkit [260661] by Jacob Uphoff

• 2 edits in trunk/LayoutTests

[ Mac wk2 ] http/tests/IndexedDB/storage-limit.https.html is flaky failing.
​https://bugs.webkit.org/show_bug.cgi?id=209189

Unreviewed test gardening.

• platform/ios-wk2/TestExpectations:

12:19 PM Changeset in webkit [260660] by ysuzuki@apple.com

• 13 edits
  6 adds in trunk

[JSC] DFG compare should speculate BigInt well
​https://bugs.webkit.org/show_bug.cgi?id=210892

Reviewed by Saam Barati.

JSTests:

• stress/compare-bigint-with-number.js: Added.

(shouldBe):
(result):
(lessThan):
(lessThanEqual):
(greaterThan):
(greaterThanEqual):
(equal):

• stress/compare-bigint-with-string.js: Added.

(shouldBe):
(result):
(lessThan):
(lessThanEqual):
(greaterThan):
(greaterThanEqual):
(equal):

• stress/compare-bigint.js: Added.

(shouldBe):
(result):
(lessThan):
(lessThanEqual):
(greaterThan):
(greaterThanEqual):
(equal):

• stress/compare-bigint32.js: Added.

(shouldBe):
(result):
(lessThan):
(lessThanEqual):
(greaterThan):
(greaterThanEqual):
(equal):

• stress/compare-heap-bigint.js: Added.

(shouldBe):
(result):
(lessThan):
(lessThanEqual):
(greaterThan):
(greaterThanEqual):
(equal):

• stress/dfg-ai-fold-bigint.js: Added.

(shouldBe):
(allAreTrue):
(allAreFalse):
(lessThan):
(lessThanFalse):
(lessThanEqual):
(lessThanEqualFalse):
(greaterThan):
(greaterThanFalse):
(greaterThanEqual):
(greaterThanEqualFalse):
(equal):
(equalFalse):

Source/JavaScriptCore:

Compare operations in DFG does not support BigInt related speculations. As a result, DFG fixup phase emits DoubleRep for operands, and
causes OSR exit. This patch adds BigInt32, HeapBigInt, and AnyBigIntUse support to DFG compare operations to avoid OSR exits.
We also introduce JSBigInt::compareToInt32 to avoid allocating JSBigInt only for comparison, and optimize C++ runtime for JSBigInt comparison.

• dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):

• dfg/DFGDoesGC.cpp:

(JSC::DFG::doesGC):

• dfg/DFGFixupPhase.cpp:

(JSC::DFG::FixupPhase::fixupNode):

• dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileValueAdd):
(JSC::DFG::SpeculativeJIT::compileValueSub):
(JSC::DFG::SpeculativeJIT::compileValueMul):
(JSC::DFG::SpeculativeJIT::compare):
(JSC::DFG::SpeculativeJIT::genericJSValueNonPeepholeCompare):
(JSC::DFG::SpeculativeJIT::nonSpeculativeNonPeepholeCompare): Deleted.

• dfg/DFGSpeculativeJIT.h:
• dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compileBigInt32Compare):

• ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileCompareEq):
(JSC::FTL::DFG::LowerDFGToB3::compileCompareStrictEq):
(JSC::FTL::DFG::LowerDFGToB3::compare):
(JSC::FTL::DFG::LowerDFGToB3::genericJSValueCompare):
(JSC::FTL::DFG::LowerDFGToB3::nonSpeculativeCompare): Deleted.

• jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::unboxBigInt32):

• runtime/JSBigInt.cpp:

(JSC::JSBigInt::compareToInt32):

• runtime/JSBigInt.h:

(JSC::swapBigIntCompareResult):

• runtime/Operations.h:

(JSC::compareBigInt):
(JSC::compareBigInt32ToOtherPrimitive):
(JSC::bigIntCompare):

11:42 AM Changeset in webkit [260659] by ddkilzer@apple.com

• 3 edits in trunk/Tools

check-webkit-style should recognize *Internal.h and *Private.h as primary headers
<​https://webkit.org/b/210979>

Reviewed by Darin Adler.

• Scripts/webkitpy/style/checkers/cpp.py:

(_classify_include): If a header has an "Internal.h" or a
"Private.h" suffix with the same base name as the source file,
consider it a primary header--the header that comes after
"config.h".

• Scripts/webkitpy/style/checkers/cpp_unittest.py:

(OrderOfIncludesTest.test_classify_include): Add tests.

11:23 AM Changeset in webkit [260658] by commit-queue@webkit.org

• 18 edits in trunk

SPI clients using fastServerTrustEvaluationEnabled need SPI to inform them of modern TLS negotiation
​https://bugs.webkit.org/show_bug.cgi?id=210533

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-24
Reviewed by Brady Eidson.

Source/WebKit:

• NetworkProcess/NetworkDataTask.h:

(WebKit::NetworkDataTaskClient::didNegotiateModernTLS):

• NetworkProcess/NetworkLoad.cpp:

(WebKit::NetworkLoad::didNegotiateModernTLS):

• NetworkProcess/NetworkLoad.h:
• NetworkProcess/cocoa/NetworkDataTaskCocoa.h:
• NetworkProcess/cocoa/NetworkDataTaskCocoa.mm:

(WebKit::NetworkDataTaskCocoa::didNegotiateModernTLS):

• NetworkProcess/cocoa/NetworkSessionCocoa.mm:

(-[WKNetworkSessionDelegate URLSession:task:didReceiveChallenge:completionHandler:]):

• UIProcess/API/APINavigationClient.h:

(API::NavigationClient::didNegotiateModernTLS):

• UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
• UIProcess/Cocoa/NavigationState.h:
• UIProcess/Cocoa/NavigationState.mm:

(WebKit::NavigationState::setNavigationDelegate):
(WebKit::NavigationState::NavigationClient::didNegotiateModernTLS):

• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::didNegotiateModernTLS):

• UIProcess/Network/NetworkProcessProxy.h:
• UIProcess/Network/NetworkProcessProxy.messages.in:
• UIProcess/WebPageProxy.cpp:
• UIProcess/WebPageProxy.h:

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/TLSDeprecation.mm:

(-[TLSNavigationDelegate waitForDidNegotiateModernTLS]):
(-[TLSNavigationDelegate _webView:didNegotiateModernTLS:]):
(TestWebKitAPI::TEST):

11:08 AM Changeset in webkit [260657] by BJ Burg

• 2 edits in trunk/Tools

webkitpy: update autoinstalled mozprocess dependency to 1.1.0 (adds python3 support)
​https://bugs.webkit.org/show_bug.cgi?id=210220
<rdar://problem/58881483>

Reviewed by Anders Carlsson.

• Scripts/webkitpy/thirdparty/init.py:

(AutoinstallImportHook._install_mozprocess):

10:58 AM Changeset in webkit [260656] by jonlee@apple.com

• 8 edits in trunk/PerformanceTests

MotionMark: ensure that timestamps are valid during warm up phase of tests
​https://bugs.webkit.org/show_bug.cgi?id=210640

Reviewed by Said Abou-Hallawa.

Ensure that Benchmark._benchmarkStartTimestamp is set during warm up phase.
Otherwise it is NaN, which makes the Benchmark.timestamp invalid, which is
used by tests like Multiply to drive the animation. When the warm up phase
completes, the start timestamp is reset.

Update minor version of benchmark with this bug fix, and include
version changelog in the about page.

For testing, add a parameter that allows for adjusting the length of the
warm up phase. It remains at its current default, 100 ms.

• MotionMark/about.html: Add section of version changelog. Includes links

to webkit.org blog posts.

• MotionMark/developer.html: Add parameter for setting warmup length.

Remove the Kalman filter parameters, since they should always be fixed.

• MotionMark/resources/runner/motionmark.css: Include styles to show

version log.

• MotionMark/resources/runner/motionmark.js: Factor out default options to

a property on window.benchmarkController. Include the default warmup length
of 100 ms.
(window.benchmarkController.startBenchmark): Refactor to use benchmarkDefaultParameters.

• MotionMark/resources/debug-runner/motionmark.js: Ditto.
• MotionMark/resources/strings.js: Update version number.
• MotionMark/tests/resources/main.js:

(_animateLoop): Set _benchmarkTimestamp during the warmup phase. Check the
warmup length. The _benchmarkTimestamp variable remains reset when the test
begins.

10:56 AM Changeset in webkit [260655] by commit-queue@webkit.org

• 6 edits in trunk

[OpenSSL] Implement WebCrypto APIs for HMAC
​https://bugs.webkit.org/show_bug.cgi?id=210902

Patch by Tomoki Imai <Tomoki Imai> on 2020-04-24
Reviewed by Don Olmstead.

Source/WebCore:

Support WebCrypto HMAC sign/verify with OpenSSL.
The design and some functions are inherited from the other ports.

• crypto/openssl/CryptoAlgorithmHMACOpenSSL.cpp:

(WebCore::HMACAlgorithm): Added. Helper function to map CryptoAlgorithmIdentifier to OpenSSL EVP_MD type.
(WebCore::calculateSignature): Added. Helper function to calculate the signature for sign/verify.
(WebCore::CryptoAlgorithmHMAC::platformSign): Implemented, mostly same as the other ports.
(WebCore::CryptoAlgorithmHMAC::platformVerify): Implemented, mostly same as the other ports.

• crypto/openssl/CryptoAlgorithmRegistryOpenSSL.cpp:

(WebCore::CryptoAlgorithmRegistry::platformRegisterAlgorithms): Added CryptoAlgorithmHMAC support.

• crypto/openssl/OpenSSLCryptoUniquePtr.h: Added specialized unique_ptrs for EVP_MD_CTX and EVP_PKEY.

(WebCore::OpenSSLCryptoPtrDeleter<EVP_MD_CTX>::operator() const):
(WebCore::OpenSSLCryptoPtrDeleter<EVP_PKEY>::operator() const):

LayoutTests:

Enabled WebCrypto LayoutTests for HMAC along with the implementation.

• platform/wincairo/TestExpectations:

10:53 AM Changeset in webkit [260654] by Alexey Shvayka

• 5 edits in trunk

Proxy.revocable should not have Construct slot
​https://bugs.webkit.org/show_bug.cgi?id=210959

Reviewed by Darin Adler.

JSTests:

• stress/proxy-revoke.js:
• test262/expectations.yaml: Mark 2 test cases as passing.

Source/JavaScriptCore:

This change removes proxyRevocableConstructorThrowError() since its presence is
observable when, for example, Proxy.revocable is a ProxyTarget itself [1].
Also removes unnecessary newTarget() check in constructProxyObject() and
2 extra ArgList instances.

This patch aligns JSC with the spec [2], V8 and SpiderMonkey.

[1]: ​https://tc39.es/ecma262/#sec-proxycreate (step 7.b)
[2]: ​https://tc39.es/ecma262/#sec-ecmascript-standard-built-in-objects

• runtime/ProxyConstructor.cpp:

(JSC::makeRevocableProxy):
(JSC::ProxyConstructor::finishCreation):
(JSC::constructProxyObject):
(JSC::proxyRevocableConstructorThrowError): Deleted.

10:50 AM Changeset in webkit [260653] by BJ Burg

• 9 edits
  2 adds in trunk/Source

Web Automation: timeout underneath Automation.evaluateJavaScriptFunction in Selenium test frame_switching_tests.py::testShouldNotBeAbleToDoAnythingTheFrameIsDeletedFromUnderUs[Safari]
​https://bugs.webkit.org/show_bug.cgi?id=210162
<rdar://problem/60561009>

Reviewed by Devin Rousso.

Source/WebCore:

• page/DOMWindow.h: Expose DOMWindow::{register, unregister}Observer.

Source/WebKit:

When an iframe is detached from the DOM, it is no longer exposed as a browsing context
and it's not possible to Evaluate JavaScript or perform other commands with it. This
patch adds frame lifecycle monitoring so that pending script evaluations are cancelled
with FrameNotFound as soon as the iframe is detached from the DOM. This change also avoids
running more commands with the frame if it's detached from its DOMWindow and ready to be GC'd.

• Sources.txt:
• WebKit.xcodeproj/project.pbxproj:
• WebProcess/Automation/WebAutomationDOMWindowObserver.h: Added.
• WebProcess/Automation/WebAutomationDOMWindowObserver.cpp: Added.

(WebKit::WebAutomationDOMWindowObserver::WebAutomationDOMWindowObserver):
(WebKit::WebAutomationDOMWindowObserver::~WebAutomationDOMWindowObserver):
(WebKit::WebAutomationDOMWindowObserver::frame const):
(WebKit::WebAutomationDOMWindowObserver::willDestroyGlobalObjectInCachedFrame):
(WebKit::WebAutomationDOMWindowObserver::willDestroyGlobalObjectInFrame):
(WebKit::WebAutomationDOMWindowObserver::willDetachGlobalObjectFromFrame):
This class is a stripped-down copy of DOMWindowExtension, which is the only other
client of DOMWindow::Observer interface. When a frame is detached, destroyed, or
navigates (global object detached), then call the callback and unregister.

• WebProcess/Automation/WebAutomationSessionProxy.h:
• WebProcess/Automation/WebAutomationSessionProxy.cpp:

(WebKit::WebAutomationSessionProxy::~WebAutomationSessionProxy):
(WebKit::WebAutomationSessionProxy::didClearWindowObjectForFrame):
(WebKit::WebAutomationSessionProxy::willDestroyGlobalObjectForFrame):
(WebKit::WebAutomationSessionProxy::evaluateJavaScriptFunction):
(WebKit::WebAutomationSessionProxy::ensureObserverForFrame): For non-main frames,
ensure we add a frame observer if we are about to evaluate JavaScript upon the frame.
This acts as a watchdog in case the frame becomes detached while waiting for pending
JS evaluations. When a frame is detached, the JS evaluation may or may not complete.

(WebKit::WebAutomationSessionProxy::selectOptionElement): Fix hilarious typo.

• WebProcess/GPU/graphics/cocoa/ImageBufferShareableIOSurfaceBackend.cpp:
• WebProcess/GPU/media/WebMediaStrategy.cpp:

(WebKit::WebMediaStrategy::clearNowPlayingInfo):
(WebKit::WebMediaStrategy::setNowPlayingInfo):
Adding a new file seems to have exposed a few missing includes and namespace qualifiers.
This is due to unified sources chunking.

10:25 AM Changeset in webkit [260652] by commit-queue@webkit.org

• 9 edits
  1 delete in trunk/Source/WebKit

Use sendWithAsyncReply for ShareSheet related messages
​https://bugs.webkit.org/show_bug.cgi?id=210828
<rdar://problem/61800730>

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-24
Reviewed by Brent Fulgham.

• Platform/IPC/MessageSender.cpp:

We need to call addAsyncReplyHandler before sendMessage in case this is the first async message from this process.
Otherwise the reply from the first message is dropped sometimes.

• Shared/ShareSheetCallbackID.h: Removed.
• UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::showShareSheet):

• UIProcess/WebPageProxy.h:
• UIProcess/WebPageProxy.messages.in:
• WebKit.xcodeproj/project.pbxproj:
• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::showShareSheet):
(WebKit::nextShareSheetCallbackID): Deleted.
(WebKit::WebPage::didCompleteShareSheet): Deleted.

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/WebPage.messages.in:

10:20 AM Changeset in webkit [260651] by ysuzuki@apple.com

• 3 edits
  1 add in trunk

[JSC] DFG AI for some bitops + BigInt32 should be precise
​https://bugs.webkit.org/show_bug.cgi?id=210956

Reviewed by Keith Miller.

JSTests:

• stress/bigint-bitops.js: Added.

(shouldBe):
(test):

Source/JavaScriptCore:

Use SpecBigInt32 for ValueBitXor, ValueBitAnd, and ValueBitOr since they are always producing BigInt32 and they have inlined implementations in DFG / FTL.

• dfg/DFGAbstractInterpreterInlines.h:

(JSC::DFG::AbstractInterpreter<AbstractStateType>::executeEffects):

10:02 AM Changeset in webkit [260650] by commit-queue@webkit.org

• 12 edits
  1 add in trunk/Source

Call STDynamicActivityAttributionPublisher in the WebProcess
​https://bugs.webkit.org/show_bug.cgi?id=210772
<rdar://problem/62075201>

Patch by Youenn Fablet <​youenn@apple.com> and Luming Yin <​luming_yin@apple.com> on 2020-04-24
Reviewed by Geoffrey Garen.

Source/WebKit:

Call STDynamicActivityAttributionPublisher in WebProcess to make use of the newly added plist entry.
Use of a sandbox extension to protect this call.

• Configurations/WebKit.xcconfig:
• Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:
• Scripts/process-entitlements.sh:
• Shared/WebProcessCreationParameters.cpp:

(WebKit::WebProcessCreationParameters::encode const):
(WebKit::WebProcessCreationParameters::decode):

• Shared/WebProcessCreationParameters.h:
• UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::platformInitializeWebProcess):

• WebKit.xcodeproj/project.pbxproj:
• WebProcess/EntryPoint/Cocoa/XPCService/WebContentService/Info-iOS.plist:
• WebProcess/EntryPoint/Cocoa/XPCService/WebContentService/InfoPlist.strings: Added.
• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::platformInitializeWebProcess):

Source/WTF:

• wtf/PlatformHave.h:

9:58 AM Changeset in webkit [260649] by Jacob Uphoff

• 2 edits in trunk/LayoutTests

[ Mac wk2 ] tiled-drawing/simple-document-with-margin-tiles.html is flaky failing.
​https://bugs.webkit.org/show_bug.cgi?id=207518

Unreviewed test gardening.

• platform/mac-wk2/TestExpectations:

9:32 AM Changeset in webkit [260648] by pvollan@apple.com

• 2 edits in trunk

[Win] Fix AppleWin build
​https://bugs.webkit.org/show_bug.cgi?id=210971

Reviewed by Brent Fulgham.

PAL is built as a static library.

• Source/cmake/target/PAL.cmake:

9:22 AM Changeset in webkit [260647] by Alan Bujtas

• 6 edits
  2 adds in trunk

[LFC][TFC] Take first in-flow table-row baseline into account when computing cell baseline
​https://bugs.webkit.org/show_bug.cgi?id=210972

Reviewed by Antti Koivisto.

Source/WebCore:

Check if the cell has a nested table and use its first row as the baseline for the cell (unless there's an IFC before).

Test: fast/layoutformattingcontext/table-basic-row-baseline-with-nested-table.html

• layout/tableformatting/TableFormattingContext.cpp:

(WebCore::Layout::TableFormattingContext::layoutCell):
(WebCore::Layout::TableFormattingContext::computeAndDistributeExtraVerticalSpace):

• layout/tableformatting/TableFormattingContext.h:
• layout/tableformatting/TableFormattingContextGeometry.cpp:

(WebCore::Layout::TableFormattingContext::Geometry::usedBaselineForCell):

• layout/tableformatting/TableGrid.h:

(WebCore::Layout::TableGrid::Row::setBaselineOffset):
(WebCore::Layout::TableGrid::Row::baselineOffset const):

LayoutTests:

• fast/layoutformattingcontext/table-basic-row-baseline-with-nested-table-expected.txt: Added.
• fast/layoutformattingcontext/table-basic-row-baseline-with-nested-table.html: Added.

9:20 AM Changeset in webkit [260646] by Antti Koivisto

• 2 edits in trunk/Source/WebCore

Nullptr crash in objc_msgSend under WebCore::genericFamily
​https://bugs.webkit.org/show_bug.cgi?id=210911
<rdar://problem/61510208>

Reviewed by Geoffrey Garen.

Speculative fix.

• platform/graphics/cocoa/SystemFontDatabaseCoreText.cpp:

(WebCore::genericFamily):

Test that CTFontDescriptorCopyAttribute is really returning CFStringRef.
Also explicitly return String from lambda to clarify lifetimes.

9:11 AM Changeset in webkit [260645] by Adrian Perez de Castro

• 1 copy in releases/WPE WebKit/webkit-2.28.2

WPE WebKit 2.28.2

9:11 AM Changeset in webkit [260644] by Adrian Perez de Castro

• 4 edits in releases/WebKitGTK/webkit-2.28

Unreviewed. Update OptionsWPE.cmake and NEWS for the 2.28.2 release

.:

• Source/cmake/OptionsWPE.cmake: Bump version numbers.

Source/WebKit:

• wpe/NEWS: Add release notes for 2.28.2.

8:03 AM Changeset in webkit [260643] by Simon Fraser

• 8 edits in trunk/Source

Move some post-renderingUpdate code into WebCore
​https://bugs.webkit.org/show_bug.cgi?id=210952

Reviewed by Antti Koivisto.

Factor some code called by the various DrawingArea subclasses into Page::finalizeRenderingUpdate(),
with some flags to control behavior that differs between drawing areas.

ScrollingCoordinator::commitTreeStateIfNeeded() is a no-op for RemoteScrollingCoordinator so
it's fine to always call it.

Source/WebCore:

• page/Page.cpp:

(WebCore::Page::passiveTouchEventListenerRectsForTesting):
(WebCore::Page::finalizeRenderingUpdate):

• page/Page.h:

Source/WebKit:

• WebProcess/WebPage/RemoteLayerTree/RemoteLayerTreeDrawingArea.mm:

(WebKit::RemoteLayerTreeDrawingArea::updateRendering):

• WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::finalizeRenderingUpdate):

• WebProcess/WebPage/WebPage.h:
• WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm:

(WebKit::TiledCoreAnimationDrawingArea::updateRendering):

7:57 AM Changeset in webkit [260642] by Chris Dumez

• 8 edits in trunk

[iOS] Stop using legacy BKSApplicationStateMonitor
​https://bugs.webkit.org/show_bug.cgi?id=210945

Reviewed by Tim Horton.

Source/WebKit:

Stop using legacy BKSApplicationStateMonitor and use RunningBoard API instead.

• Configurations/WebKit.xcconfig:

Stop linking against ApplicationServices when using iOS 14 SDK now that we are
fully transitioned to RunningBoard.

• Platform/spi/ios/RunningBoardServicesSPI.h:
• UIProcess/ApplicationStateTracker.h:
• UIProcess/ApplicationStateTracker.mm:

(WebKit::ApplicationStateTracker::ApplicationStateTracker):
(WebKit::isApplicationForeground):
(WebKit::ApplicationStateTracker::~ApplicationStateTracker):
(WebKit::isBackgroundState): Deleted.

• UIProcess/ios/PageClientImplIOS.mm:

(WebKit::PageClientImpl::isApplicationVisible):

WebKitLibraries:

• WebKitPrivateFrameworkStubs/iOS/13/RunningBoardServices.framework/RunningBoardServices.tbd:

7:22 AM Changeset in webkit [260641] by commit-queue@webkit.org

• 3 edits in trunk/Tools

[GTK][WPE] White-list more GStreamer environment variables in webkitpy
​https://bugs.webkit.org/show_bug.cgi?id=210854
<rdar://problem/62238305>

Patch by Philippe Normand <​pnormand@igalia.com> on 2020-04-24
Reviewed by Carlos Alberto Lopez Perez.

Extra variables need to be white-listed when the webkitpy tooling
runs inside a gst-build environment, those variables are needed so
that uninstalled GStreamer plugins are correctly picked up.

Additionally we now correctly white-list the
WEBKIT_GST_USE_PLAYBIN3 env var. USE_PLAYBIN3 shouldn't be used
anymore.

• Scripts/webkitpy/port/gtk.py:

(GtkPort.setup_environ_for_server):

• Scripts/webkitpy/port/wpe.py:

(WPEPort.setup_environ_for_server):

7:12 AM Changeset in webkit [260640] by Diego Pino Garcia

• 3 edits in trunk/LayoutTests

[GTK][WPE] Gardening, mark fast/css/resize-corner-tracking.html as flaky
​https://bugs.webkit.org/show_bug.cgi?id=210969

Unreviewed gardening.

Apparently the test is only failing in GTK and WPE test bots.

• platform/gtk/TestExpectations:
• platform/wpe/TestExpectations:

5:14 AM WebKitGTK/2.28.x edited by clopez@igalia.com

(diff)

3:37 AM Changeset in webkit [260639] by commit-queue@webkit.org

• 4 edits in trunk/JSTests

Skip on ARM and MIPS stress/for-of-iterator-open* added at r260585
​https://bugs.webkit.org/show_bug.cgi?id=210961

Unreviewed Gardening.

Patch by Paulo Matos <Paulo Matos> on 2020-04-24

• stress/for-of-iterator-open-osr-at-inlined-return-non-object.js:
• stress/for-of-iterator-open-osr-at-iterator-set-local.js:
• stress/for-of-iterator-open-return-non-object.js:

3:35 AM Changeset in webkit [260638] by youenn@apple.com

• 1 edit
  2 deletes in trunk/LayoutTests

getDisplayMedia is not respecting aspect ratio with max constraints
​https://bugs.webkit.org/show_bug.cgi?id=210858
<rdar://problem/61405434>

Unreviewed.
Remove these tests as it is redundant with getDisplayMedia-max-constraints 1 2 and 3.

• fast/mediastream/getDisplayMedia-max-constraints-expected.txt: Removed.
• fast/mediastream/getDisplayMedia-max-constraints.html: Removed.

2:58 AM Changeset in webkit [260637] by Carlos Garcia Campos

• 1 copy in releases/WebKitGTK/webkit-2.28.2

WebKitGTK 2.28.2

2:57 AM Changeset in webkit [260636] by Carlos Garcia Campos

• 4 edits in releases/WebKitGTK/webkit-2.28

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.28.2 release

.:

• Source/cmake/OptionsGTK.cmake: Bump version numbers.

Source/WebKit:

• gtk/NEWS: Add release notes for 2.28.2.

2:39 AM WebKitGTK/2.28.x edited by Carlos Garcia Campos

(diff)

2:22 AM Changeset in webkit [260635] by Diego Pino Garcia

• 2 edits in trunk/LayoutTests

[WPE] Gardening, expected to fail but passing
​https://bugs.webkit.org/show_bug.cgi?id=210960

Unreviewed gardening.

• platform/wpe/TestExpectations:

2:21 AM Changeset in webkit [260634] by Carlos Garcia Campos

• 3 edits
  1 add in releases/WebKitGTK/webkit-2.28

Merge r256766 - [Wasm] REGRESSION(r256665): Wasm->JS call IC needs to save memory size register
​https://bugs.webkit.org/show_bug.cgi?id=207849

Reviewed by Mark Lam.

JSTests:

• wasm/regress/regress-256665.js: Added.

(f):

Source/JavaScriptCore:

When generating the call IC, we should select the callee saves using BoundsChecking mode in order
to obey to the calling conventions described in r256665. Currently, we won't restore the memory size
register when calling the Wasm LLInt through the call IC.

• wasm/js/WebAssemblyFunction.cpp:

(JSC::WebAssemblyFunction::calleeSaves const):

2:21 AM Changeset in webkit [260633] by Carlos Garcia Campos

• 3 edits in releases/WebKitGTK/webkit-2.28/JSTests

Merge r256698 - Unreviewed: fix broken tests added in r256665
​https://bugs.webkit.org/show_bug.cgi?id=207727

Our inline WAT doesn't seem to like named blocks/branch targets.

• wasm/regress/llint-callee-saves-with-fast-memory.js:
• wasm/regress/llint-callee-saves-without-fast-memory.js:

2:20 AM Changeset in webkit [260632] by Carlos Garcia Campos

• 9 edits
  2 adds in releases/WebKitGTK/webkit-2.28

Merge r256665 - [WASM] Wasm interpreter's calling convention doesn't match Wasm JIT's convention.
​https://bugs.webkit.org/show_bug.cgi?id=207727

JSTests:

Reviewed by Mark Lam.

• wasm/regress/llint-callee-saves-with-fast-memory.js: Added.
• wasm/regress/llint-callee-saves-without-fast-memory.js: Added.

Source/JavaScriptCore:

Reviewed by Mark Lam.

The Wasm JIT has unusual calling conventions, which were further complicated by the addition
of the interpreter, and the interpreter did not correctly follow these conventions (by incorrectly
saving and restoring the callee save registers used for the memory base and size). Here's a summary
of the calling convention:

• When entering Wasm from JS, the wrapper must:
  • Preserve the base and size when entering LLInt regardless of the mode. (Prior to this patch we only preserved the base in Signaling mode)
  • Preserve the memory base in either mode, and the size for BoundsChecking.
• Both tiers must preserve every *other* register they use. e.g. the LLInt must preserve PB and wasmInstance, but must *not* preserve memoryBase and memorySize.
• Changes to memoryBase and memorySize are visible to the caller. This means that:
  • Intra-module calls can assume these registers are up-to-date even if the memory was resized. The only exception here is if the LLInt calls a signaling JIT, in which case the JIT will not update the size register, since it won't be using it.
  • Inter-module and JS calls require the caller to reload these registers. These calls may result in memory changes (e.g. the callee may call memory.grow).
  • A Signaling JIT caller must be aware that the LLInt may trash the size register, since it always bounds checks.

• llint/WebAssembly.asm:
• wasm/WasmAirIRGenerator.cpp:

(JSC::Wasm::AirIRGenerator::addCall):

• wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::B3IRGenerator::addCall):

• wasm/WasmCallee.cpp:

(JSC::Wasm::LLIntCallee::calleeSaveRegisters):

• wasm/WasmCallingConvention.h:
• wasm/WasmLLIntPlan.cpp:

(JSC::Wasm::LLIntPlan::didCompleteCompilation):

• wasm/WasmMemoryInformation.cpp:

(JSC::Wasm::PinnedRegisterInfo::get):
(JSC::Wasm::getPinnedRegisters): Deleted.

2:20 AM Changeset in webkit [260631] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.28/Source/WebKit

Merge r260570 - [GTK] Crash in cairo_surface_mark_dirty_rectangle() in accelerated compositing mode under X11
​https://bugs.webkit.org/show_bug.cgi?id=210636

Patch by John Frankish <​john.frankish@outlook.com> on 2020-04-23
Reviewed by Carlos Garcia Campos.

When cairo is configured to use xcb instead of xlib, it might use an image surface attached to the xlib one as
snapshot. In that case a flush is needed to detach that snapshot after we have drawn the surface in the
context.

• UIProcess/gtk/AcceleratedBackingStoreX11.cpp:

(WebKit::AcceleratedBackingStoreX11::paint): Call cairo_surface_flush() after drawing.

2:20 AM Changeset in webkit [260630] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.28/Source/WebCore

Merge r260567 - [GTK] excessive wakeups/polling due to gdk_frame_clock_begin_updating
​https://bugs.webkit.org/show_bug.cgi?id=210561

Reviewed by Žan Doberšek.

The problem is that we are destroying the display refresh monitor from the frame clock update callback, and GTK
schedules another update from the callback itself in some cases, which ends up happening forever. We were
assuming that destroying the window of immediately destroy the frame clock as well, but the paint source idle
keeps a reference of the frame clock. At the end of the source idle callback the source is scheduled again,
taking a new reference. We need to call gdk_frame_clock_end_updating() to ensure the idle is not scheduled
again.

• platform/graphics/gtk/DisplayRefreshMonitorGtk.cpp:

(WebCore::DisplayRefreshMonitorGtk::~DisplayRefreshMonitorGtk): Disconnect the update signal and call
gdk_frame_clock_end_updating().
(WebCore::DisplayRefreshMonitorGtk::requestRefreshCallback): Toplevel window should always have a frame clock,
so remove the early return and add an assert instead.

2:20 AM Changeset in webkit [260629] by Carlos Garcia Campos

• 7 edits in releases/WebKitGTK/webkit-2.28

Merge r260506 - [GStreamer][MSE] Youtube 'live stream'/H264 URLs fail to play, VP8/9 URLs play OK
​https://bugs.webkit.org/show_bug.cgi?id=209119

Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

The fix consists of removing the initial avoiding of seeking and just
issuing the proper segment instead of seeking (seeks in GStreamer can't
be done before prerolling anyway). Appsrc doesn't make easy to emit our
own custom segment, so what I did was to use a segment fixer probe to
modify the original [0, infinity] segment issued by appsrc and use
a [startTime, stopTime] with proper values depending on the seek target
and rate.

Covered by existing tests.

• platform/graphics/gstreamer/mse/MediaPlayerPrivateGStreamerMSE.cpp:

(WebCore::checkShouldDelaySeek): Don't hold seeks on startup, when changing from READY to PAUSED.
(WebCore::MediaPlayerPrivateGStreamerMSE::doSeek): Refactored seek delay condition. Also, don't do a regular
gst_element_seek() for initial seeks, just proceed with a special case in that situation.

• platform/graphics/gstreamer/mse/WebKitMediaSourceGStreamer.cpp:

(initialSeekSegmentFixerProbe): Probe that fixes the segment.
(webKitMediaSrcPrepareInitialSeek): Behave much like a regular seek, but also compute the right GstSegment, install
the segment fixer probe and setReadyForMoreSamples() on the SourceBufferPrivates.

• platform/graphics/gstreamer/mse/WebKitMediaSourceGStreamer.h:

LayoutTests:

Unskipped media/media-source/media-source-seek-redundant-append.html,
which passes now.

• platform/gtk/TestExpectations:
• platform/wpe/TestExpectations:

2:20 AM Changeset in webkit [260628] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.28/Source/WebKit

Merge r260252 - [GTK][X11] REGRESSION(r259944): Wrong position of select popup menu in X11
​https://bugs.webkit.org/show_bug.cgi?id=210603

Patch by Carlos Garcia Campos <​cgarcia@igalia.com> on 2020-04-17
Reviewed by Michael Catanzaro.

gdk_window_move_to_rect expects the given rectangle in coordinates relative to the top-left corner of the window
that the popup window is transient for. We were using screen coordinates.

• UIProcess/gtk/WebPopupMenuProxyGtk.cpp:

(WebKit::WebPopupMenuProxyGtk::showPopupMenu): Translate widget coordinates to window coordinates before passing
the rectangle to gdk_window_move_to_rect().

2:20 AM Changeset in webkit [260627] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.28/Source/WebKit

Merge r260244 - [GTK] UI process crash when entering compositing mode when WPE_RENDERER is enabled
​https://bugs.webkit.org/show_bug.cgi?id=209118

Reviewed by Michael Catanzaro.

Check if EGL_WL_bind_wayland_display extension is available when using WPE_RENDERER, since we don't suport the
SHM interface.

• UIProcess/gtk/AcceleratedBackingStoreWayland.cpp:

(WebKit::AcceleratedBackingStoreWayland::checkRequirements): Return false if EGL_WL_bind_wayland_display is not present.

2:20 AM Changeset in webkit [260626] by Carlos Garcia Campos

• 2 edits in releases/WebKitGTK/webkit-2.28/Source/WTF

Merge r260179 - Unreviewed, set CeilingOnPageSize for MIPS64

This fixes a build failure ("Must set CeilingOnPageSize in
PageBlock.h").

• wtf/PageBlock.h:

1:27 AM Changeset in webkit [260625] by Adrian Perez de Castro

• 2 edits in trunk/Source/WebCore

Add missing HTMLNames:: namespace prefix to usage of liTag object

Unreviewed build fix.

No new tests needed.

• editing/TextManipulationController.cpp:

(WebCore::TextManipulationController::observeParagraphs):

12:59 AM Changeset in webkit [260624] by commit-queue@webkit.org

• 5 edits in trunk/LayoutTests

Import fetch/stale-while-revalidate/fetch.html
​https://bugs.webkit.org/show_bug.cgi?id=210905

Patch by Rob Buis <​rbuis@igalia.com> on 2020-04-24
Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

Import fetch/stale-while-revalidate/fetch.html to try to fix
flakiness (see ​https://bugs.webkit.org/show_bug.cgi?id=207230).

• web-platform-tests/fetch/stale-while-revalidate/fetch.html:

LayoutTests:

This test should not be flaky anymore.

• platform/ios/TestExpectations:
• platform/mac/TestExpectations:

Apr 23, 2020:

11:57 PM Changeset in webkit [260623] by Devin Rousso

• 4 edits in trunk/Source/WebInspectorUI

Web Inspector: REGRESSION(r257759): Network: waterfall popover is missing bars
​https://bugs.webkit.org/show_bug.cgi?id=210947

Reviewed by Brian Burg.

WI.ResourceTimingBreakdownView is also used inside of a WI.Popover, meaning that the
waterfall graph won't be a child of the .network-table. Add a .network class to all
instances of .waterfall so that the selector can be more specific.

• UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._populateWaterfallGraph):

• UserInterface/Views/NetworkTableContentView.css:

(.waterfall.network .block): Added.
(body[dir=ltr] .waterfall.network .block): Added.
(body[dir=rtl] .waterfall.network .block): Added.
(.waterfall.network .block + .block): Added.
(.waterfall.network .block:matches(.mouse-tracking, .filler) + .block:not(.mouse-tracking, .filler), .waterfall.network .block:not(.request, .response) + :matches(.request, .response)): Added.
(.waterfall.network .block:last-child): Added.
(.waterfall.network .block.request,): Added.
(.waterfall.network .block.mouse-tracking): Added.
(.waterfall.network .block.filler): Added.
(.waterfall.network .block.redirect): Added.
(.waterfall.network .block.queue): Added.
(.waterfall.network .block.dns): Added.
(.waterfall.network .block.connect): Added.
(.waterfall.network .block.secure): Added.
(.waterfall.network .block.request): Added.
(.waterfall.network .block.response): Added.
(.network-table .waterfall .block): Deleted.
(body[dir=ltr] .network-table .waterfall .block): Deleted.
(body[dir=rtl] .network-table .waterfall .block): Deleted.
(.network-table .waterfall .block + .block): Deleted.
(.network-table .waterfall .block:matches(.mouse-tracking, .filler) + .block:not(.mouse-tracking, .filler), .network-table .waterfall .block:not(.request, .response) + :matches(.request, .response)): Deleted.
(.network-table .waterfall .block:last-child): Deleted.
(.network-table .waterfall .block.request,): Deleted.
(.network-table .waterfall .block.mouse-tracking): Deleted.
(.network-table .waterfall .block.filler): Deleted.
(.network-table .waterfall .block.redirect): Deleted.
(.network-table .waterfall .block.queue): Deleted.
(.network-table .waterfall .block.dns): Deleted.
(.network-table .waterfall .block.connect): Deleted.
(.network-table .waterfall .block.secure): Deleted.
(.network-table .waterfall .block.request): Deleted.
(.network-table .waterfall .block.response): Deleted.

• UserInterface/Views/ResourceTimingBreakdownView.js:

(WI.ResourceTimingBreakdownView.prototype.initialLayout):

11:51 PM Changeset in webkit [260622] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebCore

Make CachedResourceLoader more conforming to Fetch specification
​https://bugs.webkit.org/show_bug.cgi?id=210925

Patch by Rob Buis <​rbuis@igalia.com> on 2020-04-23
Reviewed by Alex Christensen.

Make CachedResourceLoader more conforming to Fetch specification
by fixing links, re-ordering steps to match main fetch [1] and do
early exit code paths earlier.

• loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::requestImage): adjust to parameter change.
(WebCore::CachedResourceLoader::canRequest): replace CachedResourceRequest param with ResourceLoaderOptions.
(WebCore::CachedResourceLoader::prepareFetch): fix comment.
(WebCore::CachedResourceLoader::requestResource): re-order.

• loader/cache/CachedResourceLoader.h:

10:43 PM Changeset in webkit [260621] by Alexey Shvayka

• 6 edits
  2 deletes in trunk

Remove revoked Proxy checks from ProxyCreate
​https://bugs.webkit.org/show_bug.cgi?id=210862

Reviewed by Ross Kirsling.

JSTests:

Removes expectations for 2 invalid ChakraCore tests.

• ChakraCore.yaml: Mark 2 test cases as passing.
• ChakraCore/test/es6/arraywithproxy.baseline: Removed.
• ChakraCore/test/es6/proxytest9.baseline: Removed.
• stress/proxy-revoke.js: Adjust test.
• test262/expectations.yaml: Mark 12 test cases as passing.

Source/JavaScriptCore:

This change removes revoked Proxy checks from ProxyCreate [1], implementing
​https://github.com/tc39/ecma262/pull/1814 and aligning JSC with SpiderMonkey.
Also cleans up ProxyObject creation by using isFunction() instead of
isCallable(), which are identical.

[1]: ​https://tc39.es/ecma262/#sec-proxycreate (steps 2, 4)

• runtime/ProxyObject.cpp:

(JSC::ProxyObject::structureForTarget):
(JSC::ProxyObject::finishCreation):

10:14 PM Changeset in webkit [260620] by ysuzuki@apple.com

• 5 edits in trunk/Tools

stress/ensure-crash.js shouldn't spew stuff onto my screen
​https://bugs.webkit.org/show_bug.cgi?id=210931

Reviewed by Ross Kirsling.

Set noisyOutputHandler when crash! is specified. We also specify noisyOutputHandler for runComplexTest.

• Scripts/run-jsc-stress-tests:
• Scripts/webkitruby/jsc-stress-test-writer-default.rb:
• Scripts/webkitruby/jsc-stress-test-writer-playstation.rb:
• Scripts/webkitruby/jsc-stress-test-writer-ruby.rb:

9:52 PM Changeset in webkit [260619] by ysuzuki@apple.com

• 2 edits in trunk

Make JSCOnly work on macOS
​https://bugs.webkit.org/show_bug.cgi?id=210953

Reviewed by Ross Kirsling.

In JSCOnly port on macOS, we should disable ICU API renaming because OS shipped ICU does not have version suffix.

• Source/cmake/OptionsJSCOnly.cmake:

9:46 PM Changeset in webkit [260618] by Lauro Moura

• 5 edits in trunk/LayoutTests

Use shouldRejectWithErrorName after r260579.

The mentioned revision made shouldRejectWithErrorName (and
shouldReject) actually omit the error name when it was not requested.

So now to catch the desired error we must actually request it
instead of just relying on the expectation file messages.

Unreviewed test gardening.

• crypto/subtle/ecdh-import-pkcs8-key-p256-validate-ecprivatekey-parameters-publickey.html:
• crypto/subtle/ecdh-import-pkcs8-key-p384-validate-ecprivatekey-parameters-publickey.html:
• crypto/subtle/ecdsa-import-pkcs8-key-p256-validate-ecprivatekey-parameters-publickey.html:
• crypto/subtle/ecdsa-import-pkcs8-key-p384-validate-ecprivatekey-parameters-publickey.html:

8:31 PM Changeset in webkit [260617] by ysuzuki@apple.com

• 2 edits in trunk/Tools

Disable useKernTCSM=false for JSC stress tests to make EWS faster
​https://bugs.webkit.org/show_bug.cgi?id=210950

Reviewed by Mark Lam.

This patch removes "JSC_useKernTCSM=false" environment variable in JSC stress tests.
Alexey found that this causes EWS JSC stress tests slow down (from roughly 63 mins to 83 mins).
This slow down happens in Mac Pro (which is used in JSC EWS right now), and this slow down does
not happen in Mac mini (post-commit buildbot) and rather improves Mac Mini execution time by 15%,
but keeping EWS faster is more important than making post-commit bots faster.

• Scripts/run-javascriptcore-tests:

6:14 PM Changeset in webkit [260616] by Simon Fraser

• 5 edits in trunk/Source/WebCore

Move the storage of DisplayID from Chrome to Page
​https://bugs.webkit.org/show_bug.cgi?id=210943

Reviewed by Tim Horton.

The less Chrome knows about Frames and Documents the better. At some point Page is going
to talk to ScrollingCoordinator in this callback too.

• page/Chrome.cpp:

(WebCore::Chrome::displayID const):
(WebCore::Chrome::windowScreenDidChange):

• page/Chrome.h:
• page/Page.cpp:

(WebCore::Page::windowScreenDidChange):

• page/Page.h:

(WebCore::Page::displayID const):

6:08 PM Changeset in webkit [260615] by Simon Fraser

• 10 edits
  3 adds in trunk

EventHandler::selectCursor() has broken resize over coordinate conversion code
​https://bugs.webkit.org/show_bug.cgi?id=210778

Reviewed by Zalan Bujtas.

Source/WebCore:

EventHandler::selectCursor() appeared to make a local hit-test point from window
to content coordinates, which made no sense, but this happened to work because
RenderLayer::hitTestLayer() set the HitTestResult localPoint to a global point
if you hit the resizer.

Clean up this mess by having all resizer-related geometry queries be in local coordinates.

As a bonus, actually set the cursor to a resize cursor when over the resizer.

Test: fast/events/cursors/mouse-cursor-over-resizer.html

• page/EventHandler.cpp:

(WebCore::EventHandler::selectCursor):
(WebCore::EventHandler::handleMousePressEvent):

• rendering/RenderLayer.cpp:

(WebCore::RenderLayer::resize):
(WebCore::RenderLayer::offsetFromResizeCorner const):
(WebCore::RenderLayer::isPointInResizeControl const):
(WebCore::RenderLayer::hitTestLayer):
(WebCore::RenderLayer::hitTestResizerInFragments const):

• rendering/RenderLayer.h:

LayoutTests:

• TestExpectations:
• fast/events/cursors/mouse-cursor-over-resizer-expected.txt: Added.
• fast/events/cursors/mouse-cursor-over-resizer.html: Added.
• fast/events/mouse-cursor-change.html:
• platform/mac-wk2/TestExpectations:
• resources/ui-helper.js:

(window.UIHelper.async moveMouseAndWaitForFrame):

6:08 PM Changeset in webkit [260614] by Simon Fraser

• 8 edits in trunk/Source/WebKit

Bounce displayWasRefreshed() via EventDispatcher
​https://bugs.webkit.org/show_bug.cgi?id=208778

Reviewed by Antti Koivisto.

Allow the WebContent process to be notified of display refresh off the main thread, so other
threads like the scrolling thread can respond without being blocked. Achieve this
by having EventDispatcher receive the IPC message, then bounce it to the main thread.

• UIProcess/mac/DisplayLink.cpp:

(WebKit::DisplayLink::displayLinkCallback):

• WebProcess/WebPage/EventDispatcher.cpp:

(WebKit::EventDispatcher::displayWasRefreshed):

• WebProcess/WebPage/EventDispatcher.h:
• WebProcess/WebPage/EventDispatcher.messages.in:
• WebProcess/WebProcess.h:
• WebProcess/cocoa/WebProcessCocoa.mm:

(WebKit::WebProcess::displayWasRefreshed):

6:05 PM Changeset in webkit [260613] by Devin Rousso

• 9 edits
  2 adds in trunk

Web Insspector: Storage: cannot select multiple local storage entries
​https://bugs.webkit.org/show_bug.cgi?id=210876

Reviewed by Brian Burg.

Source/WebInspectorUI:

Support multiple selection using WI.DataGrid.

• UserInterface/Views/DataGrid.js:

(WI.DataGrid):
(WI.DataGrid.prototype.get allowsMultipleSelection): Added.
(WI.DataGrid.prototype.set allowsMultipleSelection): Added.
(WI.DataGrid.prototype.get selectedNode):
(WI.DataGrid.prototype.set selectedNode):
(WI.DataGrid.prototype.get selectedDataGridNodes): Added.
(WI.DataGrid.prototype._keyDown):
(WI.DataGrid.prototype.selectNodes):
(WI.DataGrid.prototype._mouseDownInDataTable):
(WI.DataGrid.prototype._contextMenuInDataTable):
(WI.DataGrid.prototype.handleCopyEvent):
(WI.DataGrid.prototype._copyRow):
(WI.DataGrid.prototype._copyTable):
(WI.DataGrid.prototype._hasCopyableData):
(WI.DataGrid.prototype.selectDataGridNodeInternal): Added.
(WI.DataGrid.prototype.deselectDataGridNodeInternal): Added.
(WI.DataGrid.prototype._dispatchSelectedNodeChangedEvent): Added.
(WI.DataGrid.prototype.dataGridNodeForSelectionItem): Added.
(WI.DataGrid.prototype.selectionItemForDataGridNode): Added.
(WI.DataGrid.prototype.selectionControllerSelectionDidChange): Added.
(WI.DataGrid.prototype.selectionControllerFirstSelectableItem): Added.
(WI.DataGrid.prototype.selectionControllerLastSelectableItem): Added.
(WI.DataGrid.prototype.selectionControllerPreviousSelectableItem): Added.
(WI.DataGrid.prototype.selectionControllerNextSelectableItem): Added.

• UserInterface/Views/DataGridNode.js:

(WI.DataGridNode.prototype.select):
(WI.DataGridNode.prototype.deselect):
Replace selectedNode with a WI.SelectionController that behaves like a WI.TreeOutline.
Use the WI.SelectionController.Operation to ensure that WI.PlaceholderDataGridNode are
not selected unless directly chosen (i.e. not during shift selection or ⌘A). Add logic such
that WI.PlaceholderDataGridNode are not copied. Prefer _rows instead of children as
the latter is not sorted/filtered.

• UserInterface/Controllers/SelectionController.js:

(WI.SelectionController.createTreeComparator): Added.
(WI.SelectionController.createListComparator): Added.
Create static helper functions for common comparators.

(WI.SelectionController.prototype.deselectItem):
(WI.SelectionController.prototype.selectAll):
(WI.SelectionController.prototype.removeSelectedItems):
(WI.SelectionController.prototype.handleItemMouseDown):
(WI.SelectionController.prototype._selectItemsFromArrowKey):
(WI.SelectionController.prototype._firstSelectableItem):
(WI.SelectionController.prototype._lastSelectableItem):
(WI.SelectionController.prototype._previousSelectableItem):
(WI.SelectionController.prototype._nextSelectableItem):
(WI.SelectionController.prototype._addRange):
(WI.SelectionController.prototype._deleteRange):
Introduce a WI.SelectionController.Operation which is used to tell the _delegate about
why it's being asked for information.

• UserInterface/Views/DOMStorageContentView.js:

(WI.DOMStorageContentView):
(WI.DOMStorageContentView.prototype._deleteCallback):
Support multiple selection, including deleting multiple rows at once.

• UserInterface/Views/Table.js:

(WI.Table):

• UserInterface/Views/TreeOutline.js:

(WI.TreeOutline):
(WI.TreeOutline.prototype.selectionControllerNumberOfItems): Deleted.
Removed unused selectionControllerNumberOfItems.

• UserInterface/Views/ProfileView.js:

(WI.ProfileView):
(WI.ProfileView.prototype._dataGridNodeSelected):
Maintain a _selectedDataGridNode so that oldSelectedNode doesn't have to be included
when dispatching WI.DataGrid.Event.SelectedNodeChanged.

LayoutTests:

• inspector/tree-outline/selection-controller-tree-comparator.html: Added.
• inspector/tree-outline/selection-controller-tree-comparator-expected.txt: Added.

5:52 PM Changeset in webkit [260612] by Russell Epstein

• 2 edits in branches/safari-609.2.9.0-branch/Source/WebKit

Apply patch. rdar://problem/62272256

5:46 PM Changeset in webkit [260611] by ddkilzer@apple.com

• 4 edits in trunk/Source/WebKit

Clean up QuickLookThumbnailLoader
<​https://webkit.org/b/210814>

Reviewed by Darin Adler.

The following items are cleaned up:

• Extract using PlatformImage into QuickLookThumbnailLoader.h, rename to CocoaImage and use to get rid of duplicate code.
• Change id to instancetype for -init methods.
• Add atomic keyword to @property definitions that were using it as the default. (Use of atomic properties is rare in WebKit, so being explicit avoids a scenario where it looks like nonatomic was left off by accident.)
• Change @property definitions to readonly that are never written to outside of QuickLookThumbnailLoader.mm.
• Delete unused @property definitions.
• Change method declarations into read-only @property definitions.
• Re-declare atomic read-only @property definitions in QuickLookThumbnailLoader.h as read-write definitions in QuickLookThumbnailLoader.mm if they are written to.

• UIProcess/Cocoa/WebPageProxyCocoa.mm:

(WebKit::convertPlatformImageToBitmap):

• UIProcess/QuickLookThumbnailLoader.h:
• Rename qlThumbnailGenerationQueue @property to just queue.
• Remove contentType @property. It is not used anywhere. This also fixes a theoretical leak found by the clang static analyzer.
• Remove shouldWrite @property. It is only used within QuickLookThumbnailLoader.mm.
• Change identifier and thumbnail to @property declarations.
• UIProcess/QuickLookThumbnailLoader.mm:
• Change WKQLThumbnailLoadOperation._identifier type from NSMutableString to NSString. There was no reason for it to be mutable.

(-[WKQLThumbnailQueueManager init]):
(-[WKQLThumbnailQueueManager dealloc]):

• Release _queue to fix theoretical leak found by the clang static analyzer.

(-[WKQLThumbnailLoadOperation initWithAttachment:identifier:]):
(-[WKQLThumbnailLoadOperation initWithURL:identifier:]):
(-[WKQLThumbnailLoadOperation start]):

• Rename req to request.
• Change separate #if macros to #if/#else since only one version of this code can be used at a time.

(-[WKQLThumbnailLoadOperation thumbnail]):

• Use CocoaImage to use one copy of the method.

5:12 PM Changeset in webkit [260610] by ysuzuki@apple.com

• 4 edits in trunk/Tools

Support --report-execution-time to report execution time for each JSC stress test
​https://bugs.webkit.org/show_bug.cgi?id=210938

Reviewed by Saam Barati.

We can run run-javascriptcore-tests with --report-execution-time option to report execution time for each JSC stress test,
to figure out which test is taking a long time. It appends execution-time to the verbose log. To see it stderr, --verbose is also
required.

$ run-javascriptcore-tests .... --verbose --report-execution-time

• Scripts/run-javascriptcore-tests:

(runJSCStressTests):

• Scripts/run-jsc-stress-tests:
• Scripts/webkitruby/jsc-stress-test-writer-default.rb:

5:07 PM Changeset in webkit [260609] by Wenson Hsieh

• 5 edits in trunk

Text manipulation does not account for text in fully clipped containers
​https://bugs.webkit.org/show_bug.cgi?id=210940
<rdar://problem/61137648>

Reviewed by Tim Horton.

Source/WebCore:

Allow text manipulation to find both text in visibility: hidden; containers, as well as text in fully clipped
overflow containers. In both cases, renderers exist for these nodes, but TextIterator ignores them by default.
If these containers become visible in the future, we don't want to skip out on performing text manipulation on
them.

An alternative would be to detect when any element that has not undergone text manipulation has become visible
(i.e. no longer clipped by an ancestor), but this is likely more complicated (and possibly less performant) than
just eagerly extracting text from hidden containers, once they gain renderers.

TextManipulation.StartTextManipulationIncludesFullyClippedText

• editing/TextManipulationController.cpp:

(WebCore::ParagraphContentIterator::ParagraphContentIterator):
(WebCore::TextManipulationController::didCreateRendererForElement):
(WebCore::TextManipulationController::scheduleObservationUpdate):
(WebCore::TextManipulationController::scheduleObservartionUpdate): Deleted.

While I'm here, also rename scheduleObservartionUpdate to scheduleObservationUpdate.

• editing/TextManipulationController.h:

Tools:

Add a new text manipulation API test.

• TestWebKitAPI/Tests/WebKitCocoa/TextManipulation.mm:

(TestWebKitAPI::TEST):

4:59 PM Changeset in webkit [260608] by Megan Gardner

• 2 edits in trunk/Source/WebKit

Long pressing attachments in Notes does not activate Context Menu.
​https://bugs.webkit.org/show_bug.cgi?id=210936
<rdar://problem/61171576>

Reviewed by Tim Horton.

Not having the ID available for notes makes their code to create the context
menu and its items fail. We should probably be passing this information on
for any element if we have it.

• WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::selectionPositionInformation):

4:54 PM Changeset in webkit [260607] by Alan Coon

• 1 copy in tags/Safari-610.1.11.1

Tag Safari-610.1.11.1.

4:45 PM Changeset in webkit [260606] by jer.noble@apple.com

• 6 edits in trunk/Tools

REGRESSION (r260278): TestWebKitAPI.Fullscreen.Delegate is timing out on macOS bots
​https://bugs.webkit.org/show_bug.cgi?id=210676
<rdar://problem/61953702>

Reviewed by Daniel Bates.

A number of API tests modify the WKWebViewConfiguration after calling -[WKWebView initWithFrame:configuration:], which
could be a source of failures or flakiness. Update these tests to set those configuration values before passing the
configuration to WKWebView.

• TestWebKitAPI/Tests/WebKitCocoa/FullscreenAlert.mm:

(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKitCocoa/FullscreenDelegate.mm:

(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKitCocoa/FullscreenLayoutConstraints.mm:

(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKitCocoa/PictureInPictureDelegate.mm:

(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKitCocoa/TopContentInset.mm:

(TestWebKitAPI::TEST):

4:43 PM Changeset in webkit [260605] by Alan Coon

• 8 edits in branches/safari-610.1.11-branch/Source

Versioning.

4:04 PM Changeset in webkit [260604] by Alan Coon

• 8 edits in branches/safari-609.2.9.0-branch/Source

Versioning.

3:52 PM Changeset in webkit [260603] by sbarati@apple.com

• 5 edits
  1 add in trunk/Tools

DumpRenderTree should have the JIT entitlement on Mac
​https://bugs.webkit.org/show_bug.cgi?id=210887
<rdar://problem/62228740>

Reviewed by Tim Horton.

• DumpRenderTree/DumpRenderTree.xcodeproj/project.pbxproj:
• DumpRenderTree/mac/Configurations/Base.xcconfig:
• DumpRenderTree/mac/Configurations/DebugRelease.xcconfig:
• DumpRenderTree/mac/Configurations/DumpRenderTree.entitlements: Added.
• DumpRenderTree/mac/Configurations/DumpRenderTree.xcconfig:

3:21 PM Changeset in webkit [260602] by Alan Coon

• 1 copy in tags/Safari-610.1.11

Tag Safari-610.1.11.

3:00 PM Changeset in webkit [260601] by Ross Kirsling

• 2 edits in trunk/JSTests

Unreviewed test262 gardening following r260591.

• test262/expectations.yaml:

Fix expectations to reflect ICU 64.

2:34 PM Changeset in webkit [260600] by Brent Fulgham

• 5 edits in trunk/Source/WebKit

Allow "kern.osversion" sysctl read
​https://bugs.webkit.org/show_bug.cgi?id=210929
<rdar://problem/62256013>

Reviewed by Per Arne Vollan.

Libdispatch checks "kern.osversion" as part of the code path we use for launching our XPC services.
We already allow it for some services (Network Process), and should be doing so for our other
helper processes.

• GPUProcess/mac/com.apple.WebKit.GPUProcess.sb.in:
• Resources/SandboxProfiles/ios/com.apple.WebKit.GPU.sb:
• Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:
• WebProcess/com.apple.WebProcess.sb.in:

2:24 PM Changeset in webkit [260599] by Peng Liu

• 2 edits
  1 delete in trunk/Source/WebKit

Remove the WebKit.plist for Feature Flags
​https://bugs.webkit.org/show_bug.cgi?id=210534

Reviewed by Simon Fraser.

• FeatureFlags/WebKit.plist: Removed.
• WebKit.xcodeproj/project.pbxproj:

2:08 PM Changeset in webkit [260598] by commit-queue@webkit.org

• 3 edits
  4 adds in trunk

Allow credentials for same-origin css mask images
​https://bugs.webkit.org/show_bug.cgi?id=210895
<rdar://problem/60093888>

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-23
Reviewed by Brent Fulgham.

Source/WebCore:

Test: http/tests/security/css-mask-image-credentials.html

r230006 went a step too far in restricting what is allowed with css mask images.
Basic authentication credentials should be allowed with such requests as they are in Chrome and Firefox.
This can be seen by doing run-webkit-httpd then opening ​http://127.0.0.1:8000/security/css-mask-image-credentials.html
In Chrome and Firefox you'll see it forward to a page that has a blue square.
In Safari before this change you'll see a yellow square and a basic authentication prompt.
In Safari after this change you'll see the same blue square you see in Chrome and Firefox.

• style/StylePendingResources.cpp:

(WebCore::Style::loadPendingImage):

LayoutTests:

• http/tests/security/css-mask-image-credentials-expected.html: Added.
• http/tests/security/css-mask-image-credentials.html: Added.
• http/tests/security/resources/css-mask-image-credentials-2.html: Added.
• http/tests/security/resources/image-credential-check.php: Added.

1:59 PM Changeset in webkit [260597] by commit-queue@webkit.org

• 5 edits in trunk/Source

Jesus Calling app needs more WebSQL
​https://bugs.webkit.org/show_bug.cgi?id=210889
<rdar://problem/61795507>

Patch by Alex Christensen <​achristensen@webkit.org> on 2020-04-23
Reviewed by Chris Dumez.

Source/WebCore:

Manually verified this fixes the issue in the radar.

• bindings/js/JSDOMWindowCustom.cpp:

(WebCore::jsDOMWindowInstanceFunctionOpenDatabaseBody):

Source/WebKitLegacy/mac:

• WebView/WebPreferences.mm:

(+[WebPreferences initialize]):

1:37 PM Changeset in webkit [260596] by commit-queue@webkit.org

• 7 edits in trunk/Source/WebCore

Move applyUserAgentIfNeeded calls to a more central place
​https://bugs.webkit.org/show_bug.cgi?id=209587

Patch by Rob Buis <​rbuis@igalia.com> on 2020-04-23
Reviewed by Darin Adler.

Make main resource loads stop calling applyUserAgentIfNeeded
and instead do it in the CachedResourceLoader.

• loader/FrameLoader.cpp:

(WebCore::FrameLoader::addExtraFieldsToRequest):
(WebCore::FrameLoader::loadResourceSynchronously):

• loader/appcache/ApplicationCacheGroup.cpp:

(WebCore::ApplicationCacheGroup::createRequest):

• loader/cache/CachedResourceLoader.cpp:

(WebCore::CachedResourceLoader::updateHTTPRequestHeaders):
(WebCore::CachedResourceLoader::requestResource):

• loader/cache/CachedResourceLoader.h:
• loader/cache/CachedResourceRequest.cpp:

(WebCore::CachedResourceRequest::updateReferrerAndOriginHeaders):
(WebCore::CachedResourceRequest::updateUserAgentHeader):
(WebCore::CachedResourceRequest::updateReferrerOriginAndUserAgentHeaders): Deleted.

• loader/cache/CachedResourceRequest.h:

1:28 PM Changeset in webkit [260595] by Chris Dumez

• 2 edits in trunk/Source/WebKit

[iOS] Crash on RunningBoard process assertion invalidation
​https://bugs.webkit.org/show_bug.cgi?id=210873
<rdar://problem/62194917>

Unreviewed, nil out the observer only after we've removed it to fix crashes on the bots.

• UIProcess/ios/ProcessAssertionIOS.mm:

(WebKit::ProcessAssertion::~ProcessAssertion):

1:17 PM Changeset in webkit [260594] by Kate Cheney

• 6 edits
  3 adds in trunk

All ITP database tables should reference the ObservedDomains table on DELETE CASCADE
​https://bugs.webkit.org/show_bug.cgi?id=210874
<rdar://problem/62209438>

Reviewed by Brady Eidson.

Source/WebKit:

This makes it easier to delete an ITP entry from the entire database.
An entry shouldn't really exist outside of ObservedDomains anyways,
because that table holds all data about the domain, so this is a good
change to make regardless of the deletion use case.

(WebKit::needsNewCreateTableSchema):
Function to check for old CREATE TABLE queries.

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:

(WebKit::ResourceLoadStatisticsDatabaseStore::ResourceLoadStatisticsDatabaseStore):
(WebKit::ResourceLoadStatisticsDatabaseStore::enableForeignKeys):
Must enable foreign keys to use DELETE CASCADE.

(WebKit::ResourceLoadStatisticsDatabaseStore::isMigrationNecessary):
(WebKit::ResourceLoadStatisticsDatabaseStore::migrateDataToNewTablesIfNecessary):
Since this is a schema change that is executed when a table is created, we
need to check if the current schema is out of date. If so, we can
create new tables using the proper reference and migrate the data.

(WebKit::ResourceLoadStatisticsDatabaseStore::openAndDropOldDatabaseIfNecessary):
We only need to migrate the data if there were no issues with the
schema (otherwise the entire database will already have been created
with the correct CREATE TABLE queries).

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.h:

Tools:

Adds a new database file with pre-entered values and the old schema.
Adds an API test which checks if the data was migrated after
initializing the ITP database.

• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WebKitCocoa/ResourceLoadStatistics.mm:

(TEST):

• TestWebKitAPI/Tests/WebKitCocoa/incorrectCreateTableSchema.db: Added.
• TestWebKitAPI/Tests/WebKitCocoa/incorrectCreateTableSchema.db-shm: Added.
• TestWebKitAPI/Tests/WebKitCocoa/incorrectCreateTableSchema.db-wal: Added.

1:14 PM WebKitGTK/2.28.x edited by clopez@igalia.com

(diff)

1:09 PM WebKitGTK/2.28.x edited by clopez@igalia.com

(diff)

1:08 PM Changeset in webkit [260593] by don.olmstead@sony.com

• 8 edits in trunk

[CMake] CMAKE_BINARY_DIR should always be a PRIVATE include directory
​https://bugs.webkit.org/show_bug.cgi?id=196717

Reviewed by Michael Catanzaro.

Source/WebKit:

Include CMAKE_BINARY_DIR either directly or through WebKit_PRIVATE_INCLUDE_DIRECTORIES.

• PlatformGTK.cmake:
• PlatformWPE.cmake:

Source/WTF:

Remove public includes in WTF.

• wtf/CMakeLists.txt:

Tools:

Include CMAKE_BINARY_DIR.

• MiniBrowser/gtk/CMakeLists.txt:
• TestWebKitAPI/glib/CMakeLists.txt:

1:06 PM Changeset in webkit [260592] by Adrian Perez de Castro

• 2 edits in trunk/Source/WebKit

Regression after r260359 ([GTK][WPE] lowWatermarkPages() in MemoryPressureMonitor.cpp only searches the "low" value inside the first "Node" section)
​https://bugs.webkit.org/show_bug.cgi?id=210916

Reviewed by Carlos Alberto Lopez Perez.

Switch over to using a FileHandle type based on std::unique_ptr with a custom deleter and an
utility function for opening files. This makes easier to follow the logic inside the polling
loop for opening files and retrying when needed. This also fixes exiting the thread when the
systemMemoryUsedAsPercentage() function would return -1, to loop restart instead to keep
trying.

Thanks to Pablo Saavedra for his help in making this patch.

No new tests needed.

• UIProcess/linux/MemoryPressureMonitor.cpp:

(WebKit::FileHandleDeleter::operator()): Add deleter to use with std::unique_ptr<>.
(WebKit::tryOpeningForUnbufferedReading): Add utility function to open a file handle if
needed and configuring its buffering upon opening.
(WebKit::MemoryPressureMonitor::start): Use FileHandle to ensure that handles are always
closed properly, and fix logic retry opening files on failure.

1:06 PM WebKitGTK/2.28.x edited by clopez@igalia.com

(diff)

12:57 PM Changeset in webkit [260591] by Alexey Shvayka

• 61 edits
  2 copies
  3 moves
  107 adds
  3 deletes in trunk/JSTests

Update test262 to commit 31dabb5618e2
​https://bugs.webkit.org/show_bug.cgi?id=210921

Reviewed by Ross Kirsling.

• test262/config.yaml: Skip "Atomics.waitAsync" feature.
• test262/expectations.yaml:
• test262/harness/*: Updated.
• test262/latest-changes-summary.txt:
• test262/test/*: Updated.
• test262/test262-Revision.txt:

12:49 PM Changeset in webkit [260590] by Devin Rousso

• 2 edits in trunk/Source/WebInspectorUI

Web Inspector: REGRESSION: Elements: Styles: color functions are missing swatches
​https://bugs.webkit.org/show_bug.cgi?id=210930

Reviewed by Brian Burg.

• UserInterface/Views/SpreadsheetStyleProperty.js:

(WI.SpreadsheetStyleProperty.prototype._replaceSpecialTokens):
(WI.SpreadsheetStyleProperty.prototype._addGradientTokens):
(WI.SpreadsheetStyleProperty.prototype._addColorTokens):
(WI.SpreadsheetStyleProperty.prototype._addTimingFunctionTokens):
(WI.SpreadsheetStyleProperty.prototype._addBoxShadowTokens):
(WI.SpreadsheetStyleProperty.prototype._addVariableTokens):
Only attempt to WI.Color.fromString when at a ")" that is not part of another function.
Drive-by: add variable tokens after variable text is resolved, as otherwise the variable is

replaced with a WI.InlineSwatch, which replaces the original text, preventing it
from being used for looking up the variable name.

12:49 PM Changeset in webkit [260589] by Devin Rousso

• 17 edits
  2 deletes in trunk/Source/WebInspectorUI

Web Inspector: Uncaught Exception: SyntaxError: Invalid regular expression: missing )
​https://bugs.webkit.org/show_bug.cgi?id=210890

Reviewed by Brian Burg.

• UserInterface/Base/SearchUtilities.js:

(WI.SearchUtilities.prototype._regExpForString):
Catch any exceptions from new RegExp and return null in that case.

• UserInterface/Views/FilterBar.js:

(WI.FilterBar):
(WI.FilterBar.prototype.get invalid): Added.
(WI.FilterBar.prototype.set invalid): Added.
(WI.FilterBar.prototype.clear):

• UserInterface/Views/FilterBar.css:

(:matches(.filter-bar, .search-bar).invalid > input[type="search"]): Added.

• UserInterface/Views/CookieStorageContentView.js:

(WI.CookieStorageContentView.prototype._updateFilteredCookies):

• UserInterface/Views/IndexedDatabaseObjectStoreContentView.js:

(WI.IndexedDatabaseObjectStoreContentView.prototype.dataGridMatchNodeAgainstCustomFilters):

• UserInterface/Views/NavigationSidebarPanel.js:

(WI.NavigationSidebarPanel.prototype.updateFilter):

• UserInterface/Views/NetworkTableContentView.js:

(WI.NetworkTableContentView.prototype._urlFilterDidChange):

• UserInterface/Views/SearchSidebarPanel.js:

(WI.SearchSidebarPanel):
(WI.SearchSidebarPanel.prototype.performSearch):
(WI.SearchSidebarPanel.prototype.performSearch.forEachMatch):

• UserInterface/Views/SearchSidebarPanel.css:

(.sidebar > .panel.navigation.search > .search-bar > input[type="search"]): Added.
Mark the WI.FilterBar as invalid if the filterRegExpForString is invalid.

• UserInterface/Views/DOMTreeElement.js:

(WI.DOMTreeElement.prototype._highlightSearchResults):

• UserInterface/Views/LogContentView.js:

(WI.LogContentView.prototype.performSearch):

• UserInterface/Views/ResourceHeadersContentView.js:

(WI.ResourceHeadersContentView.prototype._perfomSearchOnKeyValuePairs):

• UserInterface/Views/ResourceSecurityContentView.js:

(WI.ResourceSecurityContentView.prototype._perfomSearchOnKeyValuePairs):

• UserInterface/Views/SourceCodeTextEditor.js:

(WI.SourceCodeTextEditor.prototype.customPerformSearch):
(WI.SourceCodeTextEditor.prototype.customPerformSearch.searchResultCallback):

• UserInterface/Views/TextEditor.js:

(WI.TextEditor.prototype.performSearch):
Ensure that the WI.FindBanner shows 0 results if the searchRegExpForString is invalid.

• UserInterface/Main.html:
• UserInterface/Views/SearchBar.js: Removed.
• UserInterface/Views/SearchBar.css: Removed.

Removed unused WI.SearchBar.

12:44 PM Changeset in webkit [260588] by commit-queue@webkit.org

• 50 edits in trunk

[WebGL2] Update texture packing code for software uploads from DOM
​https://bugs.webkit.org/show_bug.cgi?id=209515

Patch by Kenneth Russell <​kbr@chromium.org> on 2020-04-23
Reviewed by Dean Jackson.

Source/WebCore:

Update the bottommost DOM-to-texture packing code in
GraphicsContextGLOpenGL and FormatConverter to full WebGL 2.0
capability. Reorganize some code to make side-by-side comparisons
easier with other WebGL 2.0 implementations.

Added NEEDS_PORT comments to areas in the calling code which need
particular attention in subsequent patches. Roughly two more
patches will be needed on top of this one in order to fully pass
the associated conformance tests.

Fix a bug in the non-ANGLE ENABLE(WEBGL2) code path which
accidentally disabled WebGL entirely in this configuration.

Covered by the WebGL 2.0 conformance tests.

• html/canvas/WebGLRenderingContextBase.cpp:

(WebCore::WebGLRenderingContextBase::create):
(WebCore::WebGLRenderingContextBase::copyTexSubImage2D):
(WebCore::WebGLRenderingContextBase::readPixels):
(WebCore::WebGLRenderingContextBase::texImageSource2D):
(WebCore::WebGLRenderingContextBase::texImage2DImpl):
(WebCore::WebGLRenderingContextBase::texImage2D):
(WebCore::WebGLRenderingContextBase::texSubImage2DImpl):
(WebCore::WebGLRenderingContextBase::texSubImage2D):
(WebCore::WebGLRenderingContextBase::validateTexFuncData):
(WebCore::WebGLRenderingContextBase::getPackPixelStoreParams const):
(WebCore::WebGLRenderingContextBase::getUnpackPixelStoreParams const):

• html/canvas/WebGLRenderingContextBase.h:
• platform/graphics/FormatConverter.cpp:

(WebCore::convertFloatToHalfFloat):
(WebCore::float>):
(WebCore::uint8_t>):
(WebCore::uint16_t>):
(WebCore::int8_t>):
(WebCore::int16_t>):
(WebCore::uint32_t>):
(WebCore::int32_t>):
(WebCore::FormatConverter::convert):

• platform/graphics/FormatConverter.h:

(WebCore::FormatConverter::FormatConverter):

• platform/graphics/GraphicsContextGL.h:

(WebCore::GraphicsContextGL::hasAlpha):
(WebCore::GraphicsContextGL::hasColor):

• platform/graphics/opengl/GraphicsContextGLOpenGL.cpp:

(WebCore::GraphicsContextGLOpenGL::texImage2DResourceSafe):
(WebCore::GraphicsContextGLOpenGL::computeFormatAndTypeParameters):
(WebCore::GraphicsContextGLOpenGL::computeImageSizeInBytes):
(WebCore::GraphicsContextGLOpenGL::PixelStoreParams::PixelStoreParams):
(WebCore::GraphicsContextGLOpenGL::packImageData):
(WebCore::GraphicsContextGLOpenGL::extractImageData):
(WebCore::GraphicsContextGLOpenGL::extractTextureData):
(WebCore::TexelBytesForFormat):
(WebCore::GraphicsContextGLOpenGL::packPixels):

• platform/graphics/opengl/GraphicsContextGLOpenGL.h:

LayoutTests:

Rebaseline WebGL 2.0 conformance tests affected by this patch.
Most updates are progressions; failures are largely caused by
sub-rectangle uploads not yet being implemented. The tests will
pass fully after the higher-level code is feature complete.

Skip the canvas_sub_rectangle tests, as their results will
continue to be wrong until the higher-level code is finished.

• TestExpectations:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r11f_g11f_b10f-rgb-unsigned_int_10f_11f_11f_rev-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r16f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r16f-red-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r32f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r8-red-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-r8ui-red_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rg16f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rg16f-rg-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rg32f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rg8-rg-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rg8ui-rg_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rgb8ui-rgb_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_bitmap_from_canvas/tex-2d-rgba8ui-rgba_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r11f_g11f_b10f-rgb-unsigned_int_10f_11f_11f_rev-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r16f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r16f-red-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r32f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r8-red-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-r8ui-red_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rg16f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rg16f-rg-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rg32f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rg8-rg-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rg8ui-rg_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rgb8ui-rgb_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/image_data/tex-2d-rgba8ui-rgba_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/misc/tex-image-with-different-data-source-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r11f_g11f_b10f-rgb-unsigned_int_10f_11f_11f_rev-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r16f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r16f-red-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r32f-red-float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r8-red-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-r8ui-red_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rg16f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rg16f-rg-half_float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rg32f-rg-float-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rg8-rg-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rg8ui-rg_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rgb8ui-rgb_integer-unsigned_byte-expected.txt:
• webgl/2.0.0/conformance2/textures/svg_image/tex-2d-rgba8ui-rgba_integer-unsigned_byte-expected.txt:

12:42 PM Changeset in webkit [260587] by Ryan Haddad

• 4 edits
  1 delete in trunk/Tools

Unreviewed, reverting r260552.

Broke internal builds.

Reverted changeset:

"DumpRenderTree should have the JIT entitlement on Mac"
​https://bugs.webkit.org/show_bug.cgi?id=210887
​https://trac.webkit.org/changeset/260552

12:37 PM WebKitGTK/2.28.x edited by Adrian Perez de Castro

(diff)

12:29 PM Changeset in webkit [260586] by ysuzuki@apple.com

• 2 edits in trunk/JSTests

Reduce iteration count and change how it invokes GC to reduce execution time of JSTests/stress/delete-property-dfg-inline.js
​https://bugs.webkit.org/show_bug.cgi?id=210933

Reviewed by Keith Miller.

This test takes 7~ seconds in Release build with default run. And it takes 1 mins in several variants of JSC tests, which is too long.
By analyzing this, we found that this takes very long time,

1. due to so frequent synchronous GC run
2. due to large iteration count while function can get FTL with less counts

While ensuring all functions gets FTL, we reduce iteration count and GC invocations to reduce the execution time, the default run gets 200ms.

• stress/delete-property-dfg-inline.js:

(noInline.blackbox.testSingleStructure):
(noInline.testSingleStructure.testInlineSingleStructure):
(noInline.testInlineSingleStructure.testExit):
(noInline.testExit.testSingleStructureMiss):
(noInline.testSingleStructureMiss.testSingleStructureMissStrict):
(noInline.testSingleStructureMissStrict.testSingleStructureMissNonConfigurable):
(noInline.testSingleStructureMissNonConfigurable.testSingleStructureEmpty):
(noInline.testSingleStructureEmpty.testPolymorphic):
(noInline.testPolymorphic.testPolyvariant):
(noInline.testPolyvariant.testConstantFolding):
(noInline.testConstantFolding.testObjectSinking):
(noInline.testObjectSinking.testProxy):
(noInline.testProxy.testTypedArray):
(noInline.testTypedArray.testMissMixed):
(noInline.testMissMixed.testMissNonMixed):
(noInline.testMissNonMixed.testByVal.noInline.test):
(noInline.testMissNonMixed.testByVal):
(noInline.assert.assert_throws): Deleted.

11:56 AM Changeset in webkit [260585] by keith_miller@apple.com

• 8 edits
  6 adds in trunk

Fix OSR exiting/iterator object checks in for-of bytecodes
​https://bugs.webkit.org/show_bug.cgi?id=210882

Reviewed by Saam Barati.

JSTests:

• stress/for-of-iterator-next-osr-at-inlined-return-non-object.js: Added.

(vendNext):
(test):
(catch):

• stress/for-of-iterator-next-osr-at-next-set-local.js: Added.

(vendNext):
(test):
(catch):

• stress/for-of-iterator-open-osr-at-inlined-return-non-object.js: Added.

(vendIterator):
(test):
(catch):

• stress/for-of-iterator-open-osr-at-iterator-set-local.js: Added.

(vendIterator):
(test):
(catch):

• stress/for-of-iterator-open-return-non-object.js: Added.

(vendIterator):
(test):
(i.catch):

• stress/test-for-of-cfg-simplication-exit-ok.js: Added.

(z.proto):

Source/JavaScriptCore:

This patch fixes some bugs in the DFGBytecodeParser where we would
set the exit origin for the SetLocal following the iterator_open/next
first call to the next bytecode. This meant that if out-of-line
Symbol.iterator or next functions returned an unexpected non-cell
we would OSR past the rest of the next bytecode rather than to the
first checkpoint.

This patch also makes sure we properly throw for non-objects returned
from either of the above functions in all tiers (and adds tests).

Finally, this patch makes a small optimization where we just ArithBitOr the
iterator's closed state (index == -1) and index is out of bounds. We can't
do a CompareBelow check because the index is effectively an int33_t.

• bytecode/BytecodeIndex.h:

(JSC::BytecodeIndex::withCheckpoint const):

• dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::nextOpcodeIndex const):
(JSC::DFG::ByteCodeParser::nextCheckpoint const):
(JSC::DFG::ByteCodeParser::progressToNextCheckpoint):
(JSC::DFG::ByteCodeParser::handleCall):
(JSC::DFG::ByteCodeParser::handleCallVariant):
(JSC::DFG::ByteCodeParser::handleInlining):
(JSC::DFG::ByteCodeParser::handleGetById):
(JSC::DFG::ByteCodeParser::handlePutById):
(JSC::DFG::ByteCodeParser::parseGetById):
(JSC::DFG::ByteCodeParser::parseBlock):
(JSC::DFG::ByteCodeParser::handlePutByVal):

• jit/JITCall.cpp:

(JSC::JIT::emitSlow_op_iterator_open):

• llint/LLIntSlowPaths.cpp:

(JSC::LLInt::LLINT_SLOW_PATH_DECL):
(JSC::LLInt::handleIteratorNextCheckpoint):

11:47 AM Changeset in webkit [260584] by Diego Pino Garcia

• 3 edits in trunk/LayoutTests

[GTK][WPE] Gardening, update expectations after r260561
​https://bugs.webkit.org/show_bug.cgi?id=210928

Unreviewed gardening.

Mark several new mediastream tests as Crash.

• platform/gtk/TestExpectations:
• platform/wpe/TestExpectations:

11:35 AM WebKitGTK/2.28.x edited by berto@igalia.com

(diff)

11:09 AM Changeset in webkit [260583] by Wenson Hsieh

• 4 edits in trunk

Add a heuristic for text manipulation to treat some list items as paragraph boundaries
​https://bugs.webkit.org/show_bug.cgi?id=210915
<rdar://problem/61907080>

Reviewed by Megan Gardner.

Source/WebCore:

Adds a mechanism to allow text manipulation to emit an item containing the current list of text manipulation
tokens early, in the case where the paragraph content iterator crosses the boundary of an element that encloses
a paragraph. Currently, the only enclosing paragraph element will be list items that have display: block;,
which we can take as a hint that the text in these list items should be vended as separate items, rather than as
tokens in a single item.

This may be extended in the future to other situations by adjusting logic in isEnclosingParagraphElement.

Test: TextManipulation.StartTextManipulationBreaksParagraphInBetweenListItems

• editing/TextManipulationController.cpp:

(WebCore::TextManipulationController::observeParagraphs):

Tools:

Add a new API test to exercise text manipulation over several different cases of lists and list items.

• TestWebKitAPI/Tests/WebKitCocoa/TextManipulation.mm:

(TestWebKitAPI::TEST):

10:49 AM Changeset in webkit [260582] by Alan Bujtas

• 5 edits
  2 adds in trunk

[LFC][TFC] Add support for basic baseline align inside a table row
​https://bugs.webkit.org/show_bug.cgi?id=210918

Reviewed by Antti Koivisto.

Source/WebCore:

Test: fast/layoutformattingcontext/table-basic-row-baseline-align.html

The minimum height of a row is defined as the height of an hypothetical linebox containing
the cells originating in the row. In this hypothetical linebox, we use baseline alignment to
align the cells vertically.
Use these vertically aligned cells to compute the final row height.

• layout/displaytree/DisplayBox.h:

(WebCore::Display::Box::verticalMarginBorderAndPadding const):
(WebCore::Display::Box::setVerticalPadding):

• layout/tableformatting/TableFormattingContext.cpp:

(WebCore::Layout::TableFormattingContext::layoutInFlowContent):
(WebCore::Layout::TableFormattingContext::layoutCell):
(WebCore::Layout::TableFormattingContext::computeAndDistributeExtraVerticalSpace):

• layout/tableformatting/TableFormattingContext.h:

LayoutTests:

• fast/layoutformattingcontext/table-basic-row-baseline-align-expected.txt: Added.
• fast/layoutformattingcontext/table-basic-row-baseline-align.html: Added.

10:48 AM Changeset in webkit [260581] by Nikos Mouchtaris

• 4 edits
  2 adds in trunk/Source/WebKit

Soft link QuickLookThumbnailing framework
​https://bugs.webkit.org/show_bug.cgi?id=210894

Reviewed by Megan Gardner.

Soft link QuickLookThumbnailing framework to solve build error.

No new tests. Unnecessary.

• Configurations/WebKit.xcconfig:
• UIProcess/QuickLookThumbnailLoader.mm:

(-[WKQLThumbnailLoadOperation initWithURL:identifier:]):
(-[WKQLThumbnailLoadOperation start]):

• UIProcess/QuickLookThumbnailingSPI.h: Added.
• UIProcess/QuickLookThumbnailingSoftLink.h: Added.
• UIProcess/QuickLookThumbnailingSoftLink.mm: Added.
• WebKit.xcodeproj/project.pbxproj:

10:36 AM Changeset in webkit [260580] by commit-queue@webkit.org

• 3 edits in trunk/Tools

Unreviewed, reverting r260562.
​https://bugs.webkit.org/show_bug.cgi?id=210924

It caused 3 new failures and 1 timeout in GTK API tests
(Requested by clopez on #webkit).

Reverted changeset:

"[GTK][WPE] White-list more GStreamer environment variables in
webkitpy"
​https://bugs.webkit.org/show_bug.cgi?id=210854
​https://trac.webkit.org/changeset/260562

10:26 AM Changeset in webkit [260579] by Chris Dumez

• 3 edits in trunk/LayoutTests

http/tests/paymentrequest/page-cache-completed-payment-response.https.html is flaky failing.
​https://bugs.webkit.org/show_bug.cgi?id=207385
<rdar://problem/59260167>

Unreviewed, follow-up to r258373 to silence the error name when using shouldReject() and
actually fix the flakiness.

• http/tests/paymentrequest/page-cache-completed-payment-response.https-expected.txt:
• resources/js-test.js:

(shouldRejectWithErrorName):

10:23 AM Changeset in webkit [260578] by sihui_liu@apple.com

• 4 edits in trunk

TextManipulationController should set range of paragraph using token's positions
​https://bugs.webkit.org/show_bug.cgi?id=210866
<rdar://problem/60646283>

Reviewed by Wenson Hsieh.

Source/WebCore:

Set the range of paragraph using positions of first token and last token in the paragraph because:

1. Accurate range makes token matching in TextManipulationController::replace() easier, as TextIterator could

visit different positions with different ranges or different conditions. For example, in our previous
implementation, start of a paragraph can be set as the first visible position of document, while position of
first token is after that. Then in replace(), TextManipulationController may extract a word before the position
of first token and return error. See added test TextManipulation.CompleteTextManipulationCorrectParagraphRange.

2. TextManipulationController can handle fewer content and this is less error-prone. For example, svg elements

before/after the paragraph text will not be identified as tokens [] in a paragraph now. See updated API tests
for example.

New test: TextManipulation.CompleteTextManipulationCorrectParagraphRange

• editing/TextManipulationController.cpp:

(WebCore::ParagraphContentIterator::moveCurrentNodeForward): m_currentNodeForFindingInvisibleContent should not
be advanced if it is already at the end.
(WebCore::containsOnlyHTMLSpaces):
(WebCore::TextManipulationController::observeParagraphs):Set the paragraph start as the position of the first
token and end as the position of last token. If the paragraph is split with <br>, the end will be extended to
position of <br> so that we can add this node back later; otherwise, <br> can be removed after original
text of paragraph is removed in TextManipulationController::replace(). Also, stop identifying spaces as tokens
because non-text Node can emit spaces.
(WebCore::TextManipulationController::replace): Only identify tokens from content with meaningful text.

Tools:

• TestWebKitAPI/Tests/WebKitCocoa/TextManipulation.mm:

(TestWebKitAPI::TEST):

10:23 AM Changeset in webkit [260577] by graouts@webkit.org

• 7 edits
  1 add in trunk/LayoutTests

[ Mac iOS ] animations/animation-direction-normal.html is a flaky failure
​https://bugs.webkit.org/show_bug.cgi?id=210156
<rdar://problem/61411725>

Reviewed by Simon Fraser.

The tests animations/animation-direction-normal.html and animations/animation-direction-reverse.html were both written
similarly to test that an element targeted by a CSS Animation would have styles animated while the animation is running
and that those styles would no longer be animated once the CSS Animation was paused using the "animation-play-state"
CSS property.

The way those assertions were made were to use setTimeout() to check the computed style at a given time and compared it
to an expected value give or take an error margin. This design was flaky, as a system under load could easily not run the
timeout until a much larger delta than the one expected would elapse.

We use a new JS helper to write these tests in a non-flaky manner. The technique used now is to record the computed style
while the animation is running without providing expected times and values, but rather specifying delays between which we
want to record the computed style. Once all values have been recorded, a method can be used to check those recorded values
by using the Web Animations API to pause and seek the animation at recorded times and query the computed style, which allows
us to test values without an error margin.

Finally, the new JS helper also allows to check the computed style using a timeout when the animation play state is not relevant,
allowing those tests to pause the animation using the "animation-play-state" property and check after incremental timeouts
that the computed style did not change.

We also made the tests use the WPT harness for assertions and reporting.

• animations/animation-direction-normal-expected.txt:
• animations/animation-direction-normal.html:
• animations/animation-direction-reverse-expected.txt:
• animations/animation-direction-reverse.html:
• animations/resources/animation-test.js: Added.

(AnimationTest):
(AnimationTest.prototype.get animation):
(AnimationTest.prototype.get value):
(AnimationTest.prototype.async valueAfterWaitingFor):
(AnimationTest.prototype.async recordValueAfterRunningFor):
(AnimationTest.prototype.checkRecordedValues):
(AnimationTest.prototype._tickUntil):

• platform/ios-wk1/TestExpectations:
• platform/ios-wk2/TestExpectations:

10:15 AM Changeset in webkit [260576] by Chris Dumez

• 5 edits in trunk

[ Mac wk2 ] imported/w3c/web-platform-tests/notifications/event-onclose.html is flaky failing.
​https://bugs.webkit.org/show_bug.cgi?id=209483
<rdar://problem/60830377>

Reviewed by Geoff Garen.

Source/WebCore:

Align garbage collection of Notification JS wrapper with the specification:

• ​https://notifications.spec.whatwg.org/#garbage-collection [1]

In particular, the following changes were made:

1. Instead of using the legacy setPendingActivity() / unsetPendingActivity(), override ActiveDOMObject::virtualHasPendingActivity() to implement the behavior documented in the specification.
2. Keep the wrapper alive as long as the notification is showing and as long as there are relevant event listeners, as per [1]. Previously, we failed to check for event listeners, which was suboptimal.
3. Update the constructor to queue a task on the event loop in order to show the notification asynchronously, instead of relying on a SuspendableTimer for this purpose. Previously, the JS wrapper could get collected between construction and the notification getting shown, which was leading to the test flakiness.

No new tests, unskipped existing test.

• Modules/notifications/Notification.cpp:

(WebCore::Notification::Notification):
(WebCore::Notification::show):
(WebCore::Notification::finalize):
(WebCore::Notification::dispatchShowEvent):
(WebCore::Notification::dispatchClickEvent):
(WebCore::Notification::dispatchCloseEvent):
(WebCore::Notification::dispatchErrorEvent):
(WebCore::Notification::eventListenersDidChange):
(WebCore::Notification::virtualHasPendingActivity const):

• Modules/notifications/Notification.h:

LayoutTests:

Unskip test now that it is no longer flaky.

• platform/mac-wk2/TestExpectations:

10:07 AM Changeset in webkit [260575] by fpizlo@apple.com

• 3 edits in trunk/Websites/webkit.org

Unreviewed, check in some more files for a blog post.

• blog-files/speculation-in-jsc/clobberize-dependence-graph.graffle:
• blog-files/speculation-in-jsc/clobberize-dependence-graph.svg:

9:27 AM Changeset in webkit [260574] by Chris Dumez

• 3 edits in trunk/Source/WebKit

[iOS] Port MediaPlayback process assertion to RunningBoard
​https://bugs.webkit.org/show_bug.cgi?id=210212
<rdar://problem/61476951>

Reviewed by Geoff Garen.

Port MediaPlayback process assertion to RunningBoard instead of the legacy BKSProcessAssertion.
We can now #ifdef out the legacy BKSProcessAssertion on recent iOS builds.

• UIProcess/ProcessAssertion.h:
• UIProcess/ios/ProcessAssertionIOS.mm:

(WebKit::runningBoardNameForAssertionType):
(WebKit::ProcessAssertion::ProcessAssertion):
(WebKit::ProcessAssertion::~ProcessAssertion):

9:12 AM Changeset in webkit [260573] by Andres Gonzalez

• 2 edits in trunk/Source/WebCore

Correction for patch 397001.
​https://bugs.webkit.org/show_bug.cgi?id=210914

Reviewed by Chris Fleizach.

• No need to check for isEmpty when retrieving the primary screen size,

as pointed out by Darin Adler in bug 210760, patch 397001.

• Added some helpful AXLOGing.

• accessibility/mac/WebAccessibilityObjectWrapperMac.mm:

(-[WebAccessibilityObjectWrapper primaryScreenHeight]):
(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:]):
(-[WebAccessibilityObjectWrapper accessibilityAttributeValue:forParameter:]):
(-[WebAccessibilityObjectWrapper accessibilityArrayAttributeValues:index:maxCount:]):

8:49 AM Changeset in webkit [260572] by don.olmstead@sony.com

• 2 edits in trunk/Source/WebCore

[CMake] Add WebKit::WebCoreTestSupport target
​https://bugs.webkit.org/show_bug.cgi?id=210867

Unreviewed build fix.

Make the dependencies explicit.

• CMakeLists.txt:

8:31 AM Changeset in webkit [260571] by Simon Fraser

• 4 edits in trunk/Source/WebCore

In the scrolling tree, separate wheel event handling from layer updating
​https://bugs.webkit.org/show_bug.cgi?id=210899

Reviewed by Antti Koivisto.

Working towards webkit.org/b/210884, it needs to be possible to have the scrolling
tree handle a wheelEvent and update its internal state about scroll positions, but not
immediately map those scroll positions onto CALayers.

To achieve this, have ScrollingTreeScrollingNode::currentScrollPositionChanged()
not call applyLayerPositions(), or notifyRelatedNodesAfterScrollPositionChange() which
just applies layer positions on related nodes.

Instead, at the end of wheel event handling, do a full scrolling tree traversal and update
all the layer positions there.

Delegated scrolling (iOS) still needs notifyRelatedNodesAfterScrollPositionChange() so it
can't be removed.

• page/scrolling/ScrollingTree.cpp:

(WebCore::ScrollingTree::handleWheelEvent):
(WebCore::ScrollingTree::applyLayerPositions):
(WebCore::ScrollingTree::applyLayerPositionsInternal):

• page/scrolling/ScrollingTree.h:
• page/scrolling/ScrollingTreeScrollingNode.cpp:

(WebCore::ScrollingTreeScrollingNode::currentScrollPositionChanged):

8:24 AM Changeset in webkit [260570] by commit-queue@webkit.org

• 2 edits in trunk/Source/WebKit

[GTK] Crash in cairo_surface_mark_dirty_rectangle() in accelerated compositing mode under X11
​https://bugs.webkit.org/show_bug.cgi?id=210636

Patch by John Frankish <​john.frankish@outlook.com> on 2020-04-23
Reviewed by Carlos Garcia Campos.

When cairo is configured to use xcb instead of xlib, it might use an image surface attached to the xlib one as
snapshot. In that case a flush is needed to detach that snapshot after we have drawn the surface in the
context.

• UIProcess/gtk/AcceleratedBackingStoreX11.cpp:

(WebKit::AcceleratedBackingStoreX11::paint): Call cairo_surface_flush() after drawing.

8:18 AM Changeset in webkit [260569] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Unreviewed, reverting r260133.

We can use RunningBoard foreground assertion again now that
<rdar://problem/61830390> has been fixed

Reverted changeset:

"REGRESSION (r259610): WebGL does not work at all on iOS (was:
Google Maps tiles turn black after initial load)"
​https://trac.webkit.org/changeset/260133

7:50 AM Changeset in webkit [260568] by cturner@igalia.com

• 3 edits in trunk/Source/WebCore

[EME][CDMProxy] Sort key status array lexicographically by key IDs
​https://bugs.webkit.org/show_bug.cgi?id=210659

Reviewed by Xabier Rodriguez-Calvar.

This is required by section 6.1 of
​https://www.w3.org/TR/encrypted-media/.

Test: encrypted-media/clearkey-keystatuses.https.html

• platform/encryptedmedia/CDMProxy.cpp:

(WebCore::KeyStore::add): We could use a set here and keep it
sorted by design, but this is more complexity than needed. The
store has for practical purposes an upper limit of 2
items. Sorting such a vector lowers to either a noop or a swap. So
the simple approach here wins over using some kind of self-sorting
set structure. I also considered only sorting on-demand, since it
only has to appear sorted from the perspective of JS, we could
sort the array in convertToJSKeyStatusVector. However, that is
semantically a const method, so sorting here felt too surprising.

• platform/encryptedmedia/CDMProxy.h:

(WebCore::Key::operator<): Add a lexicographic comparator to
Key.

7:33 AM Changeset in webkit [260567] by Carlos Garcia Campos

• 2 edits in trunk/Source/WebCore

[GTK] excessive wakeups/polling due to gdk_frame_clock_begin_updating
​https://bugs.webkit.org/show_bug.cgi?id=210561

Reviewed by Žan Doberšek.

The problem is that we are destroying the display refresh monitor from the frame clock update callback, and GTK
schedules another update from the callback itself in some cases, which ends up happening forever. We were
assuming that destroying the window of immediately destroy the frame clock as well, but the paint source idle
keeps a reference of the frame clock. At the end of the source idle callback the source is scheduled again,
taking a new reference. We need to call gdk_frame_clock_end_updating() to ensure the idle is not scheduled
again.

• platform/graphics/gtk/DisplayRefreshMonitorGtk.cpp:

(WebCore::DisplayRefreshMonitorGtk::~DisplayRefreshMonitorGtk): Disconnect the update signal and call
gdk_frame_clock_end_updating().
(WebCore::DisplayRefreshMonitorGtk::requestRefreshCallback): Toplevel window should always have a frame clock,
so remove the early return and add an assert instead.

4:36 AM Changeset in webkit [260566] by emilio

• 2 edits in trunk/Tools

Unreviewed, add my bugzilla / slack nick to contributors.json

• Scripts/webkitpy/common/config/contributors.json:

4:21 AM Changeset in webkit [260565] by Philippe Normand

• 2 edits in trunk/Tools

[JHBuild] Add mock release/debug options to the update script

Rubber-stamped by Carlos Alberto Lopez Perez.

These options are now required after r260560. They're not used
though because JHBuild itself doesn't have the notion of build
configuration, unlike the Flatpak SDK.

• Scripts/update-webkit-libs-jhbuild:

3:18 AM Changeset in webkit [260564] by commit-queue@webkit.org

• 2 edits in trunk/JSTests

Skip stress/v8-bitint32-inc.js on mips
​https://bugs.webkit.org/show_bug.cgi?id=210906

Unreviewed Gardening.

Patch by Paulo Matos <Paulo Matos> on 2020-04-23

• stress/v8-bigint32-inc.js:

2:23 AM Changeset in webkit [260563] by Diego Pino Garcia

• 2 edits in trunk/LayoutTests

[WPE] Gardening, update expectations after r259705
​https://bugs.webkit.org/show_bug.cgi?id=210904

Unreviewed gardening.

• platform/wpe/TestExpectations:

2:17 AM Changeset in webkit [260562] by Philippe Normand

• 3 edits in trunk/Tools

[GTK][WPE] White-list more GStreamer environment variables in webkitpy
​https://bugs.webkit.org/show_bug.cgi?id=210854

Reviewed by Adrian Perez de Castro.

Extra variables need to be white-listed when the webkitpy tooling
runs inside a gst-build environment, those variables are needed so
that uninstalled GStreamer plugins are correctly picked up.

Additionally we now correctly white-list the
WEBKIT_GST_USE_PLAYBIN3 env var. USE_PLAYBIN3 shouldn't be used
anymore.

• Scripts/webkitpy/port/gtk.py:

(GtkPort.setup_environ_for_server):

• Scripts/webkitpy/port/wpe.py:

(WPEPort.setup_environ_for_server):

1:59 AM Changeset in webkit [260561] by youenn@apple.com

• 15 edits
  4 moves
  9 adds in trunk

getDisplayMedia is not respecting aspect ratio with max constraints
​https://bugs.webkit.org/show_bug.cgi?id=210858

Reviewed by Eric Carlson.

Source/WebCore:

Add computation of exact frame size to respect aspect ratio in DisplayCaptureSourceCocoa::updateFrameSize.
Refactor code to have one source class DisplayCaptureSourceCocoa and specific capturer for screen and window.
This simplifies code and allows reusing DisplayCaptureSourceCocoa with a mock capturer.
Update mock code to use DisplayCaptureSourceCocoa.

Tests: fast/mediastream/getDisplayMedia-max-constraints.html

fast/mediastream/getDisplayMedia-max-constraints1.html
fast/mediastream/getDisplayMedia-max-constraints2.html
fast/mediastream/getDisplayMedia-max-constraints3.html

• SourcesCocoa.txt:
• WebCore.xcodeproj/project.pbxproj:
• platform/mediastream/mac/DisplayCaptureManagerCocoa.cpp:

(WebCore::DisplayCaptureManagerCocoa::updateDisplayCaptureDevices):
(WebCore::DisplayCaptureManagerCocoa::updateWindowCaptureDevices):
(WebCore::DisplayCaptureManagerCocoa::screenCaptureDeviceWithPersistentID):
(WebCore::DisplayCaptureManagerCocoa::windowCaptureDeviceWithPersistentID):

• platform/mediastream/mac/DisplayCaptureSourceCocoa.cpp:

(WebCore::DisplayCaptureSourceCocoa::create):
(WebCore::DisplayCaptureSourceCocoa::DisplayCaptureSourceCocoa):
(WebCore::DisplayCaptureSourceCocoa::~DisplayCaptureSourceCocoa):
(WebCore::DisplayCaptureSourceCocoa::capabilities):
(WebCore::DisplayCaptureSourceCocoa::settings):
(WebCore::DisplayCaptureSourceCocoa::startProducingData):
(WebCore::DisplayCaptureSourceCocoa::stopProducingData):
(WebCore::DisplayCaptureSourceCocoa::updateFrameSize):
(WebCore::DisplayCaptureSourceCocoa::emitFrame):
(WebCore::DisplayCaptureSourceCocoa::Capturer::setLogger):
(WebCore::DisplayCaptureSourceCocoa::Capturer::logChannel const):

• platform/mediastream/mac/DisplayCaptureSourceCocoa.h:
• platform/mediastream/mac/MockRealtimeVideoSourceMac.h:
• platform/mediastream/mac/MockRealtimeVideoSourceMac.mm:

(WebCore::MockRealtimeVideoSourceMac::createForMockDisplayCapturer):

• platform/mediastream/mac/RealtimeMediaSourceCenterMac.cpp:
• platform/mediastream/mac/ScreenDisplayCapturerMac.h: Added.
• platform/mediastream/mac/ScreenDisplayCapturerMac.mm: Added.

(WebCore::ScreenDisplayCapturerMac::create):
(WebCore::ScreenDisplayCapturerMac::ScreenDisplayCapturerMac):
(WebCore::ScreenDisplayCapturerMac::~ScreenDisplayCapturerMac):
(WebCore::ScreenDisplayCapturerMac::createDisplayStream):
(WebCore::ScreenDisplayCapturerMac::start):
(WebCore::ScreenDisplayCapturerMac::stop):
(WebCore::ScreenDisplayCapturerMac::generateFrame):
(WebCore::ScreenDisplayCapturerMac::startDisplayStream):
(WebCore::ScreenDisplayCapturerMac::commitConfiguration):
(WebCore::ScreenDisplayCapturerMac::displayWasReconfigured):
(WebCore::ScreenDisplayCapturerMac::displayReconfigurationCallBack):
(WebCore::ScreenDisplayCapturerMac::newFrame):
(WebCore::ScreenDisplayCapturerMac::screenCaptureDeviceWithPersistentID):
(WebCore::ScreenDisplayCapturerMac::screenCaptureDevices):

• platform/mediastream/mac/WindowDisplayCapturerMac.h: Added.
• platform/mediastream/mac/WindowDisplayCapturerMac.mm: ddedAdded.

(WebCore::WindowDisplayCapturerMac::create):
(WebCore::WindowDisplayCapturerMac::WindowDisplayCapturerMac):
(WebCore::WindowDisplayCapturerMac::windowImage):
(WebCore::WindowDisplayCapturerMac::generateFrame):
(WebCore::WindowDisplayCapturerMac::windowCaptureDeviceWithPersistentID):
(WebCore::WindowDisplayCapturerMac::windowCaptureDevices):

• platform/mock/MockRealtimeMediaSourceCenter.cpp:

(WebCore::MockDisplayCapturer::MockDisplayCapturer):
(WebCore::MockDisplayCapturer::start):
(WebCore::MockDisplayCapturer::generateFrame):

• platform/mock/MockRealtimeVideoSource.h:

(isType):

LayoutTests:

• fast/mediastream/getDisplayMedia-max-constraints-expected.txt: Added.
• fast/mediastream/getDisplayMedia-max-constraints.html: Added.
• fast/mediastream/getDisplayMedia-max-constraints1-expected.txt: Added.
• fast/mediastream/getDisplayMedia-max-constraints1.html: Added.
• fast/mediastream/getDisplayMedia-max-constraints2-expected.txt: Added.
• fast/mediastream/getDisplayMedia-max-constraints2.html: Added.
• fast/mediastream/getDisplayMedia-max-constraints3-expected.txt: Added.
• fast/mediastream/getDisplayMedia-max-constraints3.html: Added.
• fast/mediastream/resources/getDisplayMedia-utils.js: Added.

(async callGetDisplayMedia):
(async waitForHeight):
(async waitForWidth):

• platform/ios/TestExpectations:

Skip new tests as getDisplayMedia is not supported on iOS.

1:47 AM Changeset in webkit [260560] by Diego Pino Garcia

• 2 edits in trunk/Tools

[Flatpak SDK] Install dependencies step needs configuration as argument
​https://bugs.webkit.org/show_bug.cgi?id=210898

Reviewed by Philippe Normand.

• BuildSlaveSupport/build.webkit.org-config/steps.py:

(InstallGtkDependencies): Pass 'configuration' value (Release, Debug).
(InstallWpeDependencies): Pass 'configuration' value (Release, Debug).

1:37 AM EnvironmentVariables edited by Keith Rollin

(diff)

1:35 AM EnvironmentVariables edited by Keith Rollin

(diff)

12:11 AM Changeset in webkit [260559] by Diego Pino Garcia

• 4 edits
  4 adds in trunk/LayoutTests

[GTK][WPE] Gardening, update baselines and test expectations
​https://bugs.webkit.org/show_bug.cgi?id=210900

Results of tests are better than the general expected results,
so new GTK baselines are emitted or updated.

• platform/gtk/TestExpectations:
• platform/gtk/imported/w3c/web-platform-tests/fetch/api/basic/header-value-combining.any-expected.txt: Added.
• platform/gtk/imported/w3c/web-platform-tests/fetch/api/basic/header-value-combining.any.worker-expected.txt: Added.
• platform/gtk/imported/w3c/web-platform-tests/fetch/http-cache/cc-request-expected.txt: Updated.
• platform/gtk/imported/w3c/web-platform-tests/fetch/nosniff/parsing-nosniff.window-expected.txt: Added.
• platform/wpe/TestExpectations: Remove failure passing since r259703.