Timeline

Dec 18, 2021:

10:30 PM Changeset in webkit [287235] by Mikhail R. Gadelha

• 2 edits in trunk/Source/JavaScriptCore

[JSC][32bit] Fix undefined behavior causing miscompilation with clang 13 on ARM
​https://bugs.webkit.org/show_bug.cgi?id=234399

Reviewed by Yusuke Suzuki.

Compiling JSC with clang 13 on ARMv7 on linux was broken because clang
was marking the constant Infinity as poison during constant folding, if
either -O2 or -O3 were used, causing the constant to not being
initialized.

This patch removes the undefined behaviour by preventing the
static_cast to int32_t if the double is either inf or NaN.

• runtime/MathCommon.h:

(JSC::canBeInt32):
(JSC::canBeStrictInt32):

7:38 PM Changeset in webkit [287234] by commit-queue@webkit.org

• 8 edits in trunk/Tools

[Monterey] TestWebKitAPI.PrivateClickMeasurement.Daemon* tests timing out: Failed to connect to mach service org.webkit.pcmtestdaemon.service
​https://bugs.webkit.org/show_bug.cgi?id=232890

Patch by Alex Christensen <​achristensen@webkit.org> on 2021-12-18
Reviewed by Brady Eidson.

Using OSLaunchdJob on macOS works, but it requires a private entitlement which the public SDK can't sign with.
Just continue using launchctl instead. Also, do the NSDictionary -> xpc dictionary conversion at runtime to reduce duplicate code.

• TestWebKitAPI/Configurations/TestWebKitAPI-macOS-internal.entitlements:
• TestWebKitAPI/Configurations/TestWebKitAPI-macOS.entitlements:
• TestWebKitAPI/Configurations/TestWebKitAPI.xcconfig:
• TestWebKitAPI/Tests/WebKitCocoa/EventAttribution.mm:

(TestWebKitAPI::testDaemonPList):
(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WebKitCocoa/WebPushDaemon.mm:

(TestWebKitAPI::testWebPushDaemonPList):
(TestWebKitAPI::setUpTestWebPushD):

• TestWebKitAPI/cocoa/DaemonTestUtilities.h:
• TestWebKitAPI/cocoa/DaemonTestUtilities.mm:

(TestWebKitAPI::convertArrayToXPC):
(TestWebKitAPI::convertDictionaryToXPC):
(TestWebKitAPI::registerPlistWithLaunchD):

4:21 PM Changeset in webkit [287233] by ddkilzer@apple.com

• 2 edits in trunk/Source/bmalloc

[libpas] Do not compile libpas for 64-bit watchOS to save ~6 MB per architecture
<​https://webkit.org/b/234478>
<rdar://problem/85620647>

Reviewed by Yusuke Suzuki.

• bmalloc/BPlatform.h:

(BENABLE_LIBPAS):

• Disable libpas on 64-bit watchOS.

3:57 PM Changeset in webkit [287232] by Darin Adler

• 30 edits in trunk/Source

Hash tables, read/write, and heap memory are used unnecessarily, which may hurt performance
​https://bugs.webkit.org/show_bug.cgi?id=234438

Reviewed by Anders Carlsson.

Source/WebCore:

• Modules/mediacapabilities/MediaCapabilities.cpp:

(WebCore::bucketMIMETypes): Deleted.
(WebCore::isValidMediaMIMEType): Moved the code from bucketMIMETypes in here.
Use a SortedArraySet for bucketMIMETypes, which should be similar in performance
and uses no heap or read/write memory.

• accessibility/AccessibilityObject.cpp:

(WebCore::AccessibilityObject::popupValue const): Use a loop to check the 5 valid
values here instead of making a HashSet. Should be similar in performance and
use no heap or read/write memory. Also removes call to convertToASCIILowercase,
which will also help performance and memory use a tiny bit in some cases.

• accessibility/AccessibilityRenderObject.cpp:

(WebCore::AccessibilityRenderObject::inheritsPresentationalRole const): Use
decltype to make this declaration simpler and shorter.

• bindings/scripts/CodeGeneratorJS.pm:

(GenerateForEachEventHandlerContentAttribute): Use decltype and references
instead of pointers to make the function a little cleaner.

• dom/Element.cpp:

(WebCore::canAttachAuthorShadowRoot): Use constexpr so the tag list here does
not use any read/write memory.

• dom/make_names.pl:

(printFactoryCppFile): Removed the unnecessary constructors for the MapEntry
structs. Use decltype so we can make the tables constexpr so they do not use
any read/write memory.
(printWrapperFactoryCppFile): Ditto.

• editing/ReplaceSelectionCommand.cpp:

(WebCore::isProhibitedParagraphChild): Added a reserveInitialCapacity, which
could make the hash table here use a little less memory.

• html/Autofill.cpp:

(WebCore::fieldNameMap): Deleted. Replaced with a SortedArrayMap, which is
entirely made out of constexpr objcts so does not use heap or read/write memory.
(WebCore::toAutofillFieldName): Updated to use SortedArrayMap functions.
(WebCore::AutofillData::createFromHTMLFormControlElement): Ditto.

• html/HTMLDocument.cpp:

(WebCore::HTMLDocument::isCaseSensitiveAttribute): Added a reserveInitialCapacity,
which could make the hash table here use a little less memory.

• html/HTMLObjectElement.cpp:

(WebCore::preventsParentObjectFromExposure): Ditto.

• page/DebugPageOverlays.cpp:

(WebCore::touchEventRegionColors): Deleted.
(WebCore::NonFastScrollableRegionOverlay::drawRect): Use SortedArrayMap and
some lambdas to make this code tighter and easier to read.

• page/PerformanceUserTiming.cpp:

(WebCore::restrictedMarkNamesToNavigationTimingFunctionMap): Deleted. Use
a SortedArrayMap instead, which should be similar in performance and uses no heap
or read/write memory.
(WebCore::restrictedMarkFunction): Deleted.
(WebCore::isRestrictedMarkNameNonMainThread): Deleted. This is now safe to do on
any thread.
(WebCore::PerformanceUserTiming::isRestrictedMarkName): Use the SortedArrayMap,
which has the benefit of relaxing the thread restrictions; this is now safe to
call on any thread.
(WebCore::PerformanceUserTiming::convertMarkToTimestamp const): Updated to use
the SortedArrayMap. There may be some additional simplification possible now
that the mark functions map can be used in any thread, but I wasn't sure.

• platform/cocoa/RemoteCommandListenerCocoa.mm:

(WebCore::mediaRemoteCommandForPlatformCommand): Use makeOptionalFromPointer.

• platform/graphics/FontCascade.cpp:

(WebCore::FontCascade::hasValidAverageCharWidth const): Use a SortedArraySet,
which should be similar in performance and uses no heap or read/write memory

• platform/graphics/FontPlatformData.cpp:

(WebCore::makeOptionalFromPointer): Deleted. Moved to SortedArrayMap.h.

• platform/graphics/HEVCUtilities.cpp:

(WebCore::makeOptionalFromPointer): Deleted. Moved to SortedArrayMap.h.

• platform/graphics/avfoundation/objc/AVAssetMIMETypeCache.mm:

(WebCore::AVAssetMIMETypeCache::initializeCache): Updated for function name change.

• platform/graphics/avfoundation/objc/CDMInstanceFairPlayStreamingAVFObjC.mm:

Tweaked use of UNUSED_PARAM a bit across the file, relying on omitted names instead.

• platform/graphics/cg/UTIRegistry.cpp:

(WebCore::defaultSupportedImageTypes): Use a constexpr array rather than HashSet,
on something just iterating this, not using it as a set.

• platform/graphics/cocoa/SourceBufferParserWebM.cpp:

(WebCore::SourceBufferParserWebM::supportedMIMETypes): Renamed from webmMIMETypes,
changed to return a Span so we don't have to allocate a HashSet, since callers
simply iterate this, and don't do set operations.
(WebCore::SourceBufferParserWebM::isSupportedVideoCodec): Replaced the
supportedVideoCodecs with this. We don't need a HashSet to check for two values,
and this should be faster and use less memory.
(WebCore::SourceBufferParserWebM::isSupportedAudioCodec): Ditto.

• platform/graphics/cocoa/SourceBufferParserWebM.h: Updated for the changes above.

• rendering/svg/SVGResources.cpp:

(WebCore::tagSet): Added this helper to make the code in the functions below less
repetitive.
(WebCore::clipperFilterMaskerTags): Use a constexpr array instead of many separate
calls to HashSet::add to construct the set.
(WebCore::markerTags): Ditto.
(WebCore::fillAndStrokeTags): Ditto.
(WebCore::chainableResourceTags): Ditto.

Source/WebKit:

• NetworkProcess/Classifier/ResourceLoadStatisticsDatabaseStore.cpp:

(WebKit::ResourceLoadStatisticsDatabaseStore::sortedTables): Simplify unnecessarily complicated
expression to initialize the Span.

• Shared/Cocoa/DefaultWebBrowserChecks.h: Return a Span instead of an optional Vector.
• Shared/Cocoa/DefaultWebBrowserChecks.mm:

(WebKit::appBoundDomainsForTesting): Renamed from getAppBoundDomainsTesting to fit WebKit
coding style a bit better, and be a little more grammatical. Removed an un-needed HashMap
that had only a single entry, used a NeverDestroyed std::array rather than a Vector so we
don't use any heap or read/write memory.

• UIProcess/WebsiteData/Cocoa/WebsiteDataStoreCocoa.mm:

(WebKit::WebsiteDataStore::addTestDomains const): Rewrite to use the function above.
Fewer special cases needed since we are just using an empty Span, which efficiently
does nothing, so no need to have a distinct null value.

Source/WTF:

• wtf/SortedArrayMap.h:

(WTF::SortedArrayMap<ArrayType>::contains const): Added. More elegant than calling tryGet and
treating the pointer as a boolean.
(WTF::makeOptionalFromPointer): Moved here so it can be reused. Might rename later, since it's
only used in a few places.

2:50 PM Changeset in webkit [287231] by Adrian Perez de Castro

• 3 edits in releases/WebKitGTK/webkit-2.34/Source/WebCore

Unreviewed non-unified build fixes.

• loader/DocumentWriter.cpp: Add missing DocumentLoader.h header.
• page/PerformanceNavigationTiming.cpp: Remove "#pragma once" from non-header file.

1:34 PM Changeset in webkit [287230] by commit-queue@webkit.org

• 3 edits in trunk/Source/WebCore

Use smart pointers for WebCoreNSURLSessionDataTask ObjC members
​https://bugs.webkit.org/show_bug.cgi?id=234477
<rdar://79224869>

Patch by Alex Christensen <​achristensen@webkit.org> on 2021-12-18
Reviewed by Brady Eidson.

Seems like a smart thing to do.

• platform/network/cocoa/WebCoreNSURLSession.h:
• platform/network/cocoa/WebCoreNSURLSession.mm:

(-[WebCoreNSURLSessionDataTask originalRequest]):
(-[WebCoreNSURLSessionDataTask setOriginalRequest:]):
(-[WebCoreNSURLSessionDataTask currentRequest]):
(-[WebCoreNSURLSessionDataTask setCurrentRequest:]):
(-[WebCoreNSURLSessionDataTask error]):
(-[WebCoreNSURLSessionDataTask setError:]):
(-[WebCoreNSURLSessionDataTask taskDescription]):
(-[WebCoreNSURLSessionDataTask setTaskDescription:]):
(-[WebCoreNSURLSessionDataTask dealloc]):

12:30 PM Changeset in webkit [287229] by Chris Dumez

• 12 edits in trunk/Source/WebKit

Move CacheStorage engines from NetworkProcess to NetworkSession
​https://bugs.webkit.org/show_bug.cgi?id=234448

Reviewed by Alex Christensen.

Move CacheStorage engines from NetworkProcess to NetworkSession, as they are per-session.

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::destroySession):
(WebKit::NetworkProcess::fetchWebsiteData):
(WebKit::NetworkProcess::deleteWebsiteData):
(WebKit::NetworkProcess::deleteWebsiteDataForOrigins):
(WebKit::NetworkProcess::deleteAndRestrictWebsiteDataForRegistrableDomains):
(WebKit::NetworkProcess::registrableDomainsWithWebsiteData):
(WebKit::NetworkProcess::~NetworkProcess): Deleted.
(WebKit::NetworkProcess::findCacheEngine): Deleted.
(WebKit::NetworkProcess::ensureCacheEngine): Deleted.
(WebKit::NetworkProcess::removeCacheEngine): Deleted.
(WebKit::NetworkProcess::cacheStorageRootPath): Deleted.
(WebKit::NetworkProcess::setCacheStorageParameters): Deleted.

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkProcess.messages.in:
• NetworkProcess/NetworkSession.cpp:

(WebKit::NetworkSession::invalidateAndCancel):
(WebKit::NetworkSession::ensureCacheEngine):
(WebKit::NetworkSession::clearCacheEngine):

• NetworkProcess/NetworkSession.h:

(WebKit::NetworkSession::cacheEngine):

• NetworkProcess/cache/CacheStorageEngine.cpp:

(WebKit::CacheStorage::Engine::create):
(WebKit::CacheStorage::Engine::fetchEntries):
(WebKit::CacheStorage::Engine::open):
(WebKit::CacheStorage::Engine::remove):
(WebKit::CacheStorage::Engine::retrieveCaches):
(WebKit::CacheStorage::Engine::retrieveRecords):
(WebKit::CacheStorage::Engine::putRecords):
(WebKit::CacheStorage::Engine::deleteMatchingRecords):
(WebKit::CacheStorage::Engine::lock):
(WebKit::CacheStorage::Engine::unlock):
(WebKit::CacheStorage::Engine::clearMemoryRepresentation):
(WebKit::CacheStorage::Engine::representation):
(WebKit::CacheStorage::Engine::clearAllCaches):
(WebKit::CacheStorage::Engine::clearCachesForOrigin):
(WebKit::CacheStorage::Engine::Engine):
(WebKit::CacheStorage::Engine::from): Deleted.
(WebKit::CacheStorage::Engine::destroyEngine): Deleted.

• NetworkProcess/cache/CacheStorageEngine.h:
• NetworkProcess/cache/CacheStorageEngineConnection.cpp:

(WebKit::CacheStorageEngineConnection::~CacheStorageEngineConnection):
(WebKit::CacheStorageEngineConnection::open):
(WebKit::CacheStorageEngineConnection::remove):
(WebKit::CacheStorageEngineConnection::caches):
(WebKit::CacheStorageEngineConnection::retrieveRecords):
(WebKit::CacheStorageEngineConnection::deleteMatchingRecords):
(WebKit::CacheStorageEngineConnection::putRecords):
(WebKit::CacheStorageEngineConnection::reference):
(WebKit::CacheStorageEngineConnection::dereference):
(WebKit::CacheStorageEngineConnection::clearMemoryRepresentation):
(WebKit::CacheStorageEngineConnection::engineRepresentation):

• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::retrieveCacheStorageParameters):

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

12:11 PM Changeset in webkit [287228] by Simon Fraser

• 26 edits in trunk

Read the default value of the ScrollAnimatorEnabled setting from NSUserDefaults
​https://bugs.webkit.org/show_bug.cgi?id=234456

Reviewed by Tim Horton.

Source/WebCore:

It's impossible for a test to run with an animated keyboard scroll because platformAllowsScrollAnimation(),
which reads from NSUserDefaults, always trumps the Setting.

Fix by initializing the setting from NSUserDefaults and removing the platformAllowsScrollAnimation()
check. This allows for a WebKitTestRunner test override the setting. Remove the WTR code
that sets @"NSScrollAnimationEnabled" to @NO.

Unfortunately in WebKitLegacy AppKit code consults the @"NSScrollAnimationEnabled" key, so
DumpRenderTree has to continue to initialize this NSUserDefault to @NO and tests that
try to change the Setting will fail.

• platform/ScrollAnimator.cpp:

(WebCore::ScrollAnimator::singleAxisScroll):
(WebCore::ScrollAnimator::scrollAnimationEnabled const):

• platform/ScrollAnimator.h:

(WebCore::ScrollAnimator::platformAllowsScrollAnimation const): Deleted.

• platform/mac/ScrollAnimatorMac.h:
• platform/mac/ScrollAnimatorMac.mm:

(WebCore::ScrollAnimatorMac::platformAllowsScrollAnimation const): Deleted.

Source/WebKit:

It's impossible for a test to run with an animated keyboard scroll because platformAllowsScrollAnimation(),
which reads from NSUserDefaults, always trumps the Setting.

Fix by initializing the setting from NSUserDefaults and removing the platformAllowsScrollAnimation()
check. This allows for a WebKitTestRunner test override the setting. Remove the WTR code
that sets @"NSScrollAnimationEnabled" to @NO.

Unfortunately in WebKitLegacy AppKit code consults the @"NSScrollAnimationEnabled" key, so
DumpRenderTree has to continue to initialize this NSUserDefault to @NO and tests that
try to change the Setting will fail.

• Shared/Cocoa/WebPreferencesDefaultValuesCocoa.mm:

(WebKit::defaultScrollAnimatorEnabled):

• Shared/WebPreferencesDefaultValues.h:

Source/WebKitLegacy/mac:

It's impossible for a test to run with an animated keyboard scroll because platformAllowsScrollAnimation(),
which reads from NSUserDefaults, always trumps the Setting.

Fix by initializing the setting from NSUserDefaults and removing the platformAllowsScrollAnimation()
check. This allows for a WebKitTestRunner test override the setting.

Unfortunately in WebKitLegacy AppKit code consults the @"NSScrollAnimationEnabled" key, so
DumpRenderTree has to continue to initialize this NSUserDefault to @NO and tests that
try to change the Setting will fail.

• WebView/WebPreferencesDefaultValues.h:
• WebView/WebPreferencesDefaultValues.mm:

(WebKit::defaultScrollAnimatorEnabled):

Source/WTF:

It's impossible for a test to run with an animated keyboard scroll because platformAllowsScrollAnimation(),
which reads from NSUserDefaults, always trumps the Setting.

Fix by initializing the setting from NSUserDefaults and removing the platformAllowsScrollAnimation()
check. This allows for a WebKitTestRunner test override the setting. Remove the WTR code
that sets @"NSScrollAnimationEnabled" to @NO.

Unfortunately in WebKitLegacy AppKit code consults the @"NSScrollAnimationEnabled" key, so
DumpRenderTree has to continue to initialize this NSUserDefault to @NO and tests that
try to change the Setting will fail.

• Scripts/Preferences/WebPreferences.yaml:

Tools:

It's impossible for a test to run with an animated keyboard scroll because platformAllowsScrollAnimation(),
which reads from NSUserDefaults, always trumps the Setting.

Fix by initializing the setting from NSUserDefaults and removing the platformAllowsScrollAnimation()
check. This allows for a WebKitTestRunner test override the setting.

Unfortunately in WebKitLegacy AppKit code consults the @"NSScrollAnimationEnabled" key, so
DumpRenderTree has to continue to initialize this NSUserDefault to @NO and tests that
try to change the Setting will fail.

• WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:

(WTR::InjectedBundle::platformInitialize):

• WebKitTestRunner/mac/main.mm:

(setDefaultsToConsistentValuesForTesting):

LayoutTests:

Now that ScrollAnimatorEnabled is true by default, turn it off for these tests.

• tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-mainframe-horizontal-with-keyboard-scaled.html:
• tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-mainframe-horizontal-with-keyboard.html:
• tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-mainframe-vertical-with-keyboard-scaled.html:
• tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-mainframe-vertical-with-keyboard.html:
• tiled-drawing/scrolling/scroll-snap/scroll-snap-proximity-mainframe-horizontal-with-keyboard.html:
• tiled-drawing/scrolling/scroll-snap/scroll-snap-proximity-mainframe-vertical-with-keyboard.html:

11:04 AM Changeset in webkit [287227] by Simon Fraser

• 23 edits in trunk

Rename MockScrollAnimatorEnabled to MockScrollbarsControllerEnabled
​https://bugs.webkit.org/show_bug.cgi?id=234452

Reviewed by Anders Carlsson.

The class formerly known as MockScrollAnimatorEnabled is used for logging when scrollbars
get events, so fix the name. It's a subclass of ScrollbarsController.

Source/WebCore:

• page/FrameView.cpp:

(WebCore::FrameView::mockScrollbarsControllerEnabled const):
(WebCore::FrameView::logMockScrollbarsControllerMessage const):
(WebCore::FrameView::mockScrollAnimatorEnabled const): Deleted.
(WebCore::FrameView::logMockScrollAnimatorMessage const): Deleted.

• page/FrameView.h:
• platform/ScrollableArea.cpp:

(WebCore::ScrollableArea::scrollbarsController const):

• platform/ScrollableArea.h:

(WebCore::ScrollableArea::mockScrollbarsControllerEnabled const):
(WebCore::ScrollableArea::logMockScrollbarsControllerMessage const):
(WebCore::ScrollableArea::mockScrollAnimatorEnabled const): Deleted.
(WebCore::ScrollableArea::logMockScrollAnimatorMessage const): Deleted.

• platform/mock/ScrollbarsControllerMock.h:
• rendering/RenderLayerScrollableArea.cpp:

(WebCore::RenderLayerScrollableArea::mockScrollbarsControllerEnabled const):
(WebCore::RenderLayerScrollableArea::logMockScrollbarsControllerMessage const):
(WebCore::RenderLayerScrollableArea::mockScrollAnimatorEnabled const): Deleted.
(WebCore::RenderLayerScrollableArea::logMockScrollAnimatorMessage const): Deleted.

• rendering/RenderLayerScrollableArea.h:
• rendering/RenderListBox.cpp:

(WebCore::RenderListBox::mockScrollbarsControllerEnabled const):
(WebCore::RenderListBox::logMockScrollbarsControllerMessage const):
(WebCore::RenderListBox::mockScrollAnimatorEnabled const): Deleted.
(WebCore::RenderListBox::logMockScrollAnimatorMessage const): Deleted.

• rendering/RenderListBox.h:

Source/WTF:

• Scripts/Preferences/WebPreferences.yaml:

Tools:

• DumpRenderTree/TestOptions.cpp:

(WTR::TestOptions::defaults):

• WebKitTestRunner/TestOptions.cpp:

(WTR::TestOptions::defaults):

LayoutTests:

• fast/scrolling/scroll-animator-basic-events.html:
• fast/scrolling/scroll-animator-overlay-scrollbars-clicked.html:
• fast/scrolling/scroll-animator-overlay-scrollbars-hovered.html:
• fast/scrolling/scroll-animator-select-list-events.html:

10:06 AM Changeset in webkit [287226] by Russell Epstein

• 1 copy in tags/Safari-613.1.12.1

Tag Safari-613.1.12.1.

10:03 AM Changeset in webkit [287225] by Russell Epstein

• 9 edits in branches/safari-613.1.12-branch/Source

Versioning.

WebKit-7613.1.12.1

9:46 AM Changeset in webkit [287224] by Chris Dumez

• 39 edits in trunk/Source

Convert ProtectionSpace enums into enum classes
​https://bugs.webkit.org/show_bug.cgi?id=234455

Reviewed by Alex Christensen.

Source/WebCore:

• platform/network/CredentialStorage.cpp:

(WebCore::CredentialStorage::set):
(WebCore::CredentialStorage::removeCredentialsWithOrigin):
(WebCore::CredentialStorage::originsWithCredentials const):

• platform/network/ProtectionSpace.h:

(WebCore::ProtectionSpace::ProtectionSpace):

• platform/network/ProtectionSpaceBase.cpp:

(WebCore::ProtectionSpaceBase::ProtectionSpaceBase):
(WebCore::ProtectionSpaceBase::isProxy const):
(WebCore::ProtectionSpaceBase::receivesCredentialSecurely const):
(WebCore::ProtectionSpaceBase::isPasswordBased const):

• platform/network/ProtectionSpaceBase.h:

(WebCore::ProtectionSpaceBase::serverType const):
(WebCore::ProtectionSpaceBase::authenticationScheme const):

• platform/network/cf/AuthenticationCF.cpp:

(WebCore::createCF):
(WebCore::core):

• platform/network/cf/ProtectionSpaceCFNet.h:

(WebCore::ProtectionSpace::ProtectionSpace):

• platform/network/cf/ResourceHandleCFNet.cpp:

(WebCore::ResourceHandle::receivedCredential):

• platform/network/cf/SocketStreamHandleImplCFNet.cpp:

(WebCore::authenticationSchemeFromAuthenticationMethod):

• platform/network/cocoa/ProtectionSpaceCocoa.h:

(WebCore::ProtectionSpace::ProtectionSpace):

• platform/network/cocoa/ProtectionSpaceCocoa.mm:

(WebCore::type):
(WebCore::scheme):
(WebCore::ProtectionSpace::nsSpace const):

• platform/network/curl/AuthenticationChallenge.h:
• platform/network/curl/AuthenticationChallengeCurl.cpp:

(WebCore::AuthenticationChallenge::protectionSpaceServerTypeFromURI):
(WebCore::AuthenticationChallenge::protectionSpaceForServerTrust):
(WebCore::AuthenticationChallenge::authenticationSchemeFromCurlAuth):

• platform/network/curl/CurlRequest.cpp:

(WebCore::CurlRequest::setAuthenticationScheme):

• platform/network/curl/CurlRequest.h:
• platform/network/curl/ProtectionSpaceCurl.h:

(WebCore::ProtectionSpace::ProtectionSpace):

• platform/network/curl/ResourceHandleCurl.cpp:

(WebCore::ResourceHandle::start):
(WebCore::ResourceHandle::platformLoadResourceSynchronously):

• platform/network/mac/ResourceHandleMac.mm:

(WebCore::ResourceHandle::receivedCredential):

• platform/network/soup/AuthenticationChallengeSoup.cpp:

(WebCore::protectionSpaceServerTypeFromURL):
(WebCore::protectionSpaceFromSoupAuthAndURL):
(WebCore::protectionSpaceForClientCertificate):
(WebCore::protectionSpaceForClientCertificatePassword):

• platform/network/soup/NetworkStorageSessionSoup.cpp:

(WebCore::schemeFromProtectionSpaceServerType):
(WebCore::authTypeFromProtectionSpaceAuthenticationScheme):

Source/WebKit:

• NetworkProcess/NetworkCORSPreflightChecker.cpp:

(WebKit::NetworkCORSPreflightChecker::didReceiveChallenge):

• NetworkProcess/NetworkLoad.cpp:

(WebKit::NetworkLoad::didReceiveChallenge):

• NetworkProcess/PingLoad.cpp:

(WebKit::PingLoad::didReceiveChallenge):

• NetworkProcess/cocoa/NetworkSessionCocoa.mm:

(WebKit::NetworkSessionCocoa::allowsSpecificHTTPSCertificateForHost):

• NetworkProcess/curl/NetworkDataTaskCurl.cpp:

(WebKit::NetworkDataTaskCurl::NetworkDataTaskCurl):
(WebKit::NetworkDataTaskCurl::willPerformHTTPRedirection):
(WebKit::NetworkDataTaskCurl::restartWithCredential):

• NetworkProcess/soup/NetworkDataTaskSoup.cpp:

(WebKit::NetworkDataTaskSoup::completeAuthentication):
(WebKit::NetworkDataTaskSoup::cancelAuthentication):

• Shared/Authentication/AuthenticationManager.cpp:

(WebKit::canCoalesceChallenge):

• Shared/WebCoreArgumentCoders.cpp:

(IPC::ArgumentCoder<ProtectionSpace>::decode):

• Shared/curl/WebCoreArgumentCodersCurl.cpp:

(IPC::ArgumentCoder<ProtectionSpace>::decodePlatformData):

• UIProcess/API/C/WKAPICast.h:

(WebKit::toAPI):

• UIProcess/API/glib/WebKitAuthenticationRequest.cpp:

(webkit_authentication_request_get_security_origin):

• UIProcess/Authentication/WebProtectionSpace.cpp:

(WebKit::WebProtectionSpace::serverType const):
(WebKit::WebProtectionSpace::authenticationScheme const):

• UIProcess/Authentication/WebProtectionSpace.h:
• UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::processAuthenticationChallenge):
(WebKit::NetworkProcessProxy::didReceiveAuthenticationChallenge):

Source/WebKitLegacy/mac:

• WebCoreSupport/WebFrameLoaderClient.mm:

(WebFrameLoaderClient::canAuthenticateAgainstProtectionSpace):

Source/WebKitLegacy/win:

• WebURLProtectionSpace.cpp:

(WebURLProtectionSpace::authenticationMethod):
(coreScheme):
(WebURLProtectionSpace::initWithHost):
(WebURLProtectionSpace::initWithProxyHost):

9:18 AM Changeset in webkit [287223] by Chris Dumez

• 8 edits
  1 delete in trunk/Tools

Drop ServiceWorkerTCPServer and use HTTPServer instead
​https://bugs.webkit.org/show_bug.cgi?id=234443
<rdar://86646140>

Reviewed by Alex Christensen.

Drop ServiceWorkerTCPServer and use HTTPServer instead. ServiceWorkerTCPServer relies on TCPServer which is
known to cause flakiness and sometimes hang in its destruction.

• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WebKitCocoa/AppPrivacyReport.mm:
• TestWebKitAPI/Tests/WebKitCocoa/InAppBrowserPrivacy.mm:
• TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:
• TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerTCPServer.h: Removed.

(): Deleted.
(ServiceWorkerTCPServer::ServiceWorkerTCPServer): Deleted.
(ServiceWorkerTCPServer::request): Deleted.
(ServiceWorkerTCPServer::requestWithLocalhost): Deleted.
(ServiceWorkerTCPServer::requestWithFragment): Deleted.
(ServiceWorkerTCPServer::userAgentsChecked const): Deleted.
(ServiceWorkerTCPServer::respondToRequests): Deleted.
(ServiceWorkerTCPServer::requestWithURLString): Deleted.

• TestWebKitAPI/Tests/WebKitCocoa/StorageQuota.mm:
• TestWebKitAPI/cocoa/HTTPServer.h:
• TestWebKitAPI/cocoa/HTTPServer.mm:

(TestWebKitAPI::HTTPServer::scheme const):
(TestWebKitAPI::HTTPServer::request const):
(TestWebKitAPI::HTTPServer::requestWithLocalhost const):

7:11 AM Changeset in webkit [287222] by Antti Koivisto

• 10 edits in trunk/Source/WebCore

Remove unused :-internal-direct-focus pseudo-class
​https://bugs.webkit.org/show_bug.cgi?id=234431

Reviewed by Alexey Shvayka.

It was replaced by :focus-visible on the user-agent stylesheet.

• css/CSSSelector.cpp:

(WebCore::CSSSelector::selectorText const):

• css/CSSSelector.h:
• css/SelectorChecker.cpp:

(WebCore::SelectorChecker::checkOne const):

• css/SelectorCheckerTestFunctions.h:

(WebCore::matchesLegacyDirectFocusPseudoClass):
(WebCore::matchesFocusVisiblePseudoClass):
(WebCore::matchesDirectFocusPseudoClass): Deleted.

• css/SelectorPseudoClassAndCompatibilityElementMap.in:
• css/parser/CSSSelectorParser.cpp:

(WebCore::CSSSelectorParser::consumePseudo):

• cssjit/SelectorCompiler.cpp:

(WebCore::SelectorCompiler::addPseudoClassType):

• dom/Element.cpp:

(WebCore::Element::setFocus):

• style/RuleSet.cpp:

(WebCore::Style::RuleSet::addRule):

3:19 AM Changeset in webkit [287221] by ysuzuki@apple.com

• 8 edits in trunk/Source/JavaScriptCore

[JSC] Do not allocate m_bbqCallee and m_omgCallee until it becomes necessary
​https://bugs.webkit.org/show_bug.cgi?id=234457

Reviewed by Saam Barati.

WebAssembly memory import will require initializing both Wasm::CalleeGroup.
So, we should shrink memory size of Wasm::CalleeGroup as much as possible
to avoid memory regression. This patch allocates m_bbqCallee and m_omgCallee
only when it becomes available.

• wasm/WasmBBQPlan.cpp:

(JSC::Wasm::BBQPlan::work):

• wasm/WasmCalleeGroup.cpp:

(JSC::Wasm::CalleeGroup::CalleeGroup):

• wasm/WasmCalleeGroup.h:

(JSC::Wasm::CalleeGroup::compilationFinished): Deleted.
(JSC::Wasm::CalleeGroup::runnable): Deleted.
(JSC::Wasm::CalleeGroup::errorMessage): Deleted.
(JSC::Wasm::CalleeGroup::functionImportCount const): Deleted.
(JSC::Wasm::CalleeGroup::embedderEntrypointCalleeFromFunctionIndexSpace): Deleted.
(JSC::Wasm::CalleeGroup::wasmEntrypointCalleeFromFunctionIndexSpace): Deleted.
(JSC::Wasm::CalleeGroup::wasmBBQCalleeFromFunctionIndexSpace): Deleted.
(JSC::Wasm::CalleeGroup::entrypointLoadLocationFromFunctionIndexSpace): Deleted.
(JSC::Wasm::CalleeGroup::wasmToWasmExitStub): Deleted.
(JSC::Wasm::CalleeGroup::mode const): Deleted.

• wasm/WasmOMGForOSREntryPlan.cpp:

(JSC::Wasm::OMGForOSREntryPlan::work):

• wasm/WasmOMGPlan.cpp:

(JSC::Wasm::OMGPlan::work):

• wasm/WasmPlan.cpp:

(JSC::Wasm::Plan::updateCallSitesToCallUs):

• wasm/WasmPlan.h:

2:17 AM Changeset in webkit [287220] by ysuzuki@apple.com

• 21 edits
  5 adds in trunk

[WTF] Introduce TrailingArray
​https://bugs.webkit.org/show_bug.cgi?id=234201

Reviewed by Darin Adler.

Source/JavaScriptCore:

Use ThreadSafeRefCountedFixedVector in ObjectPropertyConditionSet and Wasm::LLIntCallees.

• bytecode/CodeBlock.h:

(JSC::CodeBlock::baselineJITConstantPool):

• bytecode/ObjectPropertyConditionSet.cpp:

(JSC::ObjectPropertyConditionSet::mergedWith const):
(JSC::ObjectPropertyConditionSet::dumpInContext const):
(JSC::ObjectPropertyConditionSet::isValidAndWatchable const):

• bytecode/ObjectPropertyConditionSet.h:

(JSC::ObjectPropertyConditionSet::invalid):
(JSC::ObjectPropertyConditionSet::create):
(JSC::ObjectPropertyConditionSet::isValid const):
(JSC::ObjectPropertyConditionSet::size const):
(JSC::ObjectPropertyConditionSet::begin const):
(JSC::ObjectPropertyConditionSet::end const):
(JSC::ObjectPropertyConditionSet::ObjectPropertyConditionSet): Deleted.
(JSC::ObjectPropertyConditionSet::releaseRawPointer): Deleted.
(JSC::ObjectPropertyConditionSet::adoptRawPointer): Deleted.
(JSC::ObjectPropertyConditionSet::fromRawPointer): Deleted.
(JSC::ObjectPropertyConditionSet::Data::Data): Deleted.

• jit/JIT.cpp:

(JSC::JIT::compileAndLinkWithoutFinalizing):

• jit/JITInlines.h:

(JSC::JIT::loadConstant):

• llint/LLIntOffsetsExtractor.cpp:
• llint/LowLevelInterpreter.asm:
• llint/LowLevelInterpreter32_64.asm:
• llint/LowLevelInterpreter64.asm:
• wasm/WasmCallee.h:

(JSC::Wasm::LLIntCallees::create): Deleted.
(JSC::Wasm::LLIntCallees::at const): Deleted.
(JSC::Wasm::LLIntCallees::data const): Deleted.
(JSC::Wasm::LLIntCallees::LLIntCallees): Deleted.

• wasm/WasmCodeBlock.cpp:

(JSC::Wasm::CodeBlock::create):
(JSC::Wasm::CodeBlock::CodeBlock):

• wasm/WasmCodeBlock.h:
• wasm/WasmModule.cpp:

(JSC::Wasm::Module::Module):
(JSC::Wasm::Module::getOrCreateCodeBlock):

• wasm/WasmModule.h:

Source/WTF:

This patch implements TrailingArray<Derived, T>, which allows us to implement class
with trailing array easily. By using this, we implement EmbeddedFixedVector and RefCountedFixedVector.
Plus, we replace underlying implementation of FixedVector from RefCountedArray to EmbeddedFixedVector
since ref-counting is not necessary. This is great for swapping: while RefCountedArray will decrease ref
and touching memory when destryoing, EmbeddedFixedVector does not, so we can keep memory untouched if
EmbeddedFixedVector's T is POD when destroying.

In a subsequent patch, we will remove RefCountedArray and use RefCountedFixedVector since RefCountedFixedVector
is following normal Ref / RefPtr protocol, so easy to understand the semantics.

• WTF.xcodeproj/project.pbxproj:
• wtf/CMakeLists.txt:
• wtf/EmbeddedFixedVector.h: Added.
• wtf/FixedVector.h:

(WTF::FixedVector::FixedVector):
(WTF::FixedVector::operator=):
(WTF::FixedVector::size const):
(WTF::FixedVector::isEmpty const):
(WTF::FixedVector::byteSize const):
(WTF::FixedVector::data):
(WTF::FixedVector::begin):
(WTF::FixedVector::end):
(WTF::FixedVector::rbegin):
(WTF::FixedVector::rend):
(WTF::FixedVector::rbegin const):
(WTF::FixedVector::rend const):
(WTF::FixedVector::at):
(WTF::FixedVector::at const):
(WTF::FixedVector::operator[]):
(WTF::FixedVector::operator[] const):
(WTF::FixedVector::fill):
(WTF::FixedVector::operator!= const):
(WTF::FixedVector::operator== const):
(WTF::FixedVector::swap):
(WTF::FixedVector::getStorage):

• wtf/RefCountedFixedVector.h: Added.
• wtf/TrailingArray.h: Added.

(WTF::TrailingArray::TrailingArray):
(WTF::TrailingArray::~TrailingArray):
(WTF::TrailingArray::allocationSize):
(WTF::TrailingArray::size const):
(WTF::TrailingArray::isEmpty const):
(WTF::TrailingArray::byteSize const):
(WTF::TrailingArray::data):
(WTF::TrailingArray::data const):
(WTF::TrailingArray::begin):
(WTF::TrailingArray::end):
(WTF::TrailingArray::begin const):
(WTF::TrailingArray::end const):
(WTF::TrailingArray::cbegin const):
(WTF::TrailingArray::cend const):
(WTF::TrailingArray::rbegin):
(WTF::TrailingArray::rend):
(WTF::TrailingArray::rbegin const):
(WTF::TrailingArray::rend const):
(WTF::TrailingArray::crbegin const):
(WTF::TrailingArray::crend const):
(WTF::TrailingArray::at):
(WTF::TrailingArray::at const):
(WTF::TrailingArray::operator[]):
(WTF::TrailingArray::operator[] const):
(WTF::TrailingArray::first):
(WTF::TrailingArray::first const):
(WTF::TrailingArray::last):
(WTF::TrailingArray::last const):
(WTF::TrailingArray::fill):
(WTF::TrailingArray::offsetOfSize):
(WTF::TrailingArray::offsetOfData):

Tools:

• TestWebKitAPI/CMakeLists.txt:
• TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
• TestWebKitAPI/Tests/WTF/EmbeddedFixedVector.cpp: Added.

(TestWebKitAPI::TEST):
(TestWebKitAPI::DestructorObserver::DestructorObserver):
(TestWebKitAPI::DestructorObserver::~DestructorObserver):
(TestWebKitAPI::DestructorObserver::operator=):

• TestWebKitAPI/Tests/WTF/FixedVector.cpp:

(TestWebKitAPI::TEST):

• TestWebKitAPI/Tests/WTF/RefCountedFixedVector.cpp: Added.

(TestWebKitAPI::TEST):
(TestWebKitAPI::DestructorObserver::DestructorObserver):
(TestWebKitAPI::DestructorObserver::~DestructorObserver):
(TestWebKitAPI::DestructorObserver::operator=):

Dec 17, 2021:

9:45 PM Changeset in webkit [287219] by Wenson Hsieh

• 4 edits in trunk/Source/WebCore

Hide viewport-constrained elements that are detected by ModalContainerObserver
​https://bugs.webkit.org/show_bug.cgi?id=234216

Reviewed by Tim Horton.

Make a small Style::Adjuster tweak to hide (i.e. set display: none;) on an element that is tagged as the
current "modal container" on ModalContainerObserver. See #234214 for more details.

• page/ModalContainerObserver.cpp:

(WebCore::ModalContainerObserver::updateModalContainerIfNeeded):

Additionally invalidate style when setting m_container here, so that the style adjuster can kick in and set
display: none; on a modal container after it is detected.

(WebCore::ModalContainerObserver::shouldHide):

• page/ModalContainerObserver.h:
• style/StyleAdjuster.cpp:

(WebCore::Style::Adjuster::adjust const):

7:35 PM Changeset in webkit [287218] by sbarati@apple.com

• 3 edits in trunk/Source/JavaScriptCore

Use IRC by default on arm64
​https://bugs.webkit.org/show_bug.cgi?id=234449

Reviewed by Yusuke Suzuki.

I'm seeing a Wasm perf improvement on some benchmarks of ~12% by switching
from Briggs to IRC. Let's make IRC the default on arm64.

• b3/air/AirAllocateRegistersByGraphColoring.cpp:
• b3/air/AirAllocateRegistersByGraphColoring.h:

(JSC::B3::Air::useIRC): Deleted.

7:06 PM Changeset in webkit [287217] by Wenson Hsieh

• 10 edits
  2 adds in trunk/Source

Add support for detecting modal containers that contain a given search term
​https://bugs.webkit.org/show_bug.cgi?id=234214

Reviewed by Tim Horton and Devin Rousso.

Source/WebCore:

Introduce ModalContainerObserver (a helper class that searches for a substring vended by the client layer in
text inside viewport-constrained renderers). If found, we keep track of the viewport-constrained container with
the matching text for use in future patches.

There is no change in behavior yet.

• Sources.txt:
• WebCore.xcodeproj/project.pbxproj:
• dom/Document.cpp:

(WebCore::Document::modalContainerObserver):

Returns the document's ModalContainerObserver, creating it only if needed (per the modal container observation
policy vended by the client).

(WebCore::Document::modalContainerObserverIfExists const):

• dom/Document.h:
• page/ChromeClient.h:

(WebCore::ChromeClient::searchStringForModalContainerObserver const):

• page/FrameView.cpp:

(WebCore::FrameView::performPostLayoutTasks):

Add a hook to create and call into ModalContainerObserver to search for a modal container (if necessary).

• page/ModalContainerObserver.cpp: Added.

(WebCore::ModalContainerObserver::isNeededFor):

Add a helper method to determine whether this "modal container observer" should be created and used for the
current document.

(WebCore::matchesSearchTerm):
(WebCore::ModalContainerObserver::updateModalContainerIfNeeded):

Iterates through all viewport-constrained render objects in the FrameView, in search of one that matches the
search term given by the client. This currently ensures that we only search any viewport-constrained element at
most once by maintaining a weak m_elementsToIgnoreWhenSearching set, but we may need a mechanism in the future
to remove elements from this set upon changing text content.

• page/ModalContainerObserver.h: Added.

Source/WebKit:

Add a WebKitAdditions hook that implements a new chrome client method. See WebCore/ChangeLog for more details.

• WebProcess/WebCoreSupport/WebChromeClient.cpp:

(WebKit::WebChromeClient::searchStringForModalContainerObserver const):

• WebProcess/WebCoreSupport/WebChromeClient.h:

6:30 PM Changeset in webkit [287216] by ysuzuki@apple.com

• 2 edits in trunk/Source/bmalloc

Unreviewed, pas_probabilistic_guard_malloc_allocator.c is broken when LIBPAS is not enabled
​https://bugs.webkit.org/show_bug.cgi?id=234415

• libpas/src/libpas/pas_probabilistic_guard_malloc_allocator.c:

6:09 PM Changeset in webkit [287215] by ysuzuki@apple.com

• 3 edits in trunk/Source/bmalloc

Unreviewed, CMakeLists is missing pas_probabilistic_guard_malloc_allocator.[ch]
​https://bugs.webkit.org/show_bug.cgi?id=234415

• CMakeLists.txt:
• libpas/src/libpas/pas_probabilistic_guard_malloc_allocator.c:

5:13 PM Changeset in webkit [287214] by ysuzuki@apple.com

• 2 edits in trunk/Source/bmalloc

[libpas] Use inline-assembly ARM64 code in GCC too
​https://bugs.webkit.org/show_bug.cgi?id=234450

Reviewed by Saam Barati.

Same problem to r287168 can happen in GCC too. Use inline assembly version
of CAS operation on ARM64 in GCC too. We also use ARM_FEATURE_ATOMICS
to determine whether LSE atomics is enabled further (but since this is very
new macro, we continue using the old check too).

• libpas/src/libpas/pas_utils.h:

(pas_atomic_store_uint8):
(pas_compare_and_swap_uint8_weak):
(pas_compare_and_swap_uint8_strong):
(pas_compare_and_swap_uint16_weak):
(pas_compare_and_swap_uint32_weak):
(pas_compare_and_swap_uint32_strong):
(pas_compare_and_swap_uint64_weak):
(pas_compare_and_swap_uint64_strong):
(pas_compare_and_swap_pair_weak):
(pas_compare_and_swap_pair_strong):
(pas_atomic_store_pair):

4:27 PM Changeset in webkit [287213] by mmaxfield@apple.com

• 2 edits in trunk/Source/WebCore

Add comment to r287208.
​https://bugs.webkit.org/show_bug.cgi?id=234428

Unreviewed.

No new tests because there is no behavior change.

• platform/graphics/WidthIterator.cpp:

(WebCore::WidthIterator::applyCSSVisibilityRules):

3:20 PM Changeset in webkit [287212] by Alan Coon

• 2 edits in branches/safari-612-branch/Source/WebKit

Apply patch. rdar://problem/86635432

3:17 PM Changeset in webkit [287211] by Alan Coon

• 8 edits in branches/safari-612-branch/Source

Versioning.

WebKit-7612.4.7

2:48 PM Changeset in webkit [287210] by Adrian Perez de Castro

• 5 edits in releases/WebKitGTK/webkit-2.34

Merge r282084 - [wpt-improvement] Improve CSP support for window reuse
​https://bugs.webkit.org/show_bug.cgi?id=229882
<rdar://78416553>

Reviewed by Chris Dumez.

LayoutTests/imported/w3c:

Rebaseline a test that now passes.

• web-platform-tests/content-security-policy/sandbox/window-reuse-sandboxed-expected.txt:

Source/WebCore:

If window.open is used to host content with a CSP sandbox header, we should always
create a new window, rather than reusing the existing one, unless the header
includes 'allow-same-site'.

This improves our WPT score.

Tested by: imported/w3c/web-platform-tests/content-security-policy/sandbox/window-reuse-sandboxed.html

• loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::responseReceived): Parse the HTTP headers earlier so we can
make decisions based on their state.

• loader/DocumentWriter.cpp:

(WebCore::DocumentWriter::begin): Don't reuse the current window if 'allow-same-site'
is not allowed.

2:48 PM Changeset in webkit [287209] by Adrian Perez de Castro

• 8 edits in releases/WebKitGTK/webkit-2.34

Merge r281941 - [COOP] html/cross-origin-opener-policy/coop-navigate-same-origin-csp-sandbox.html WPT test is failing
​https://bugs.webkit.org/show_bug.cgi?id=229716

Reviewed by Darin Adler.

LayoutTests/imported/w3c:

Rebaseline WPT test that is now passing.

• web-platform-tests/html/cross-origin-opener-policy/coop-navigate-same-origin-csp-sandbox-expected.txt:

Source/WebCore:

The COOP algorithm is relying on the response origin for some of the checks. We were computing the
response origin via SecurityOrigin::create(response.url()), which worked fine in most cases.
However, the response may contain a CSP header, which could set sandbox flags. If sandbox flags
are set, the response origin should be unique, not the origin of the response URL. This patch fixes
that.

No new tests, rebaselined existing WPT test.

• loader/DocumentLoader.cpp:

(WebCore::computeResponseOriginAndCOOP):
(WebCore::DocumentLoader::doCrossOriginOpenerHandlingOfResponse):

• loader/DocumentLoader.h:

(WebCore::DocumentLoader::contentSecurityPolicy const):

• loader/FrameLoader.cpp:

(WebCore::FrameLoader::didBeginDocument):

• page/csp/ContentSecurityPolicy.cpp:

(WebCore::ContentSecurityPolicy::copyStateFrom):
(WebCore::ContentSecurityPolicy::didReceiveHeaders):

• page/csp/ContentSecurityPolicy.h:

(WebCore::ContentSecurityPolicy::sandboxFlags const):

2:41 PM Changeset in webkit [287208] by mmaxfield@apple.com

• 3 edits in trunk/Source/WebCore

Refactor WidthIterator::applyCSSVisibilityRules() to be a little more elegant
​https://bugs.webkit.org/show_bug.cgi?id=234428

Reviewed by Alan Bujtas.

This adds a few lambda operations in this function:

• adjustForSyntheticBold()
• clobberGlyph()
• clobberAdvance()
• deleteGlyph()

And then updates the rest of the function to use those operations. I think this makes
the body of the function a little easier to understand, because the code uses a higher
level of abstraction. It shoudln't be any slower, though, because the lambdas can all
be inlined to the same code that was there before.

No new tests because there is no behavior change.

• platform/graphics/WidthIterator.cpp:

(WebCore::WidthIterator::applyCSSVisibilityRules):
(WebCore::WidthIterator::adjustForSyntheticBold): Deleted.

• platform/graphics/WidthIterator.h:

2:25 PM Changeset in webkit [287207] by sihui_liu@apple.com

• 7 edits in trunk/Source

Add custom copy() method for Ref<T> to CrossThreadCopier
​https://bugs.webkit.org/show_bug.cgi?id=234411

Reviewed by Youenn Fablet.

Source/WebCore:

• Modules/filesystemaccess/FileSystemStorageConnection.h:
• Modules/filesystemaccess/WorkerFileSystemStorageConnection.cpp:

(WebCore::WorkerFileSystemStorageConnection::getFileHandle):
(WebCore::WorkerFileSystemStorageConnection::getDirectoryHandle):
(WebCore::WorkerFileSystemStorageConnection::getHandle):
(WebCore::WorkerFileSystemStorageConnection::didGetHandleWithType): Deleted.

• Modules/filesystemaccess/WorkerFileSystemStorageConnection.h:

Source/WTF:

• wtf/CrossThreadCopier.cpp:
• wtf/CrossThreadCopier.h:

2:09 PM Changeset in webkit [287206] by mmaxfield@apple.com

• 5 edits
  2 adds in trunk/Source/WebCore

Deduplicate code in RenderText::computePreferredLogicalWidths()
​https://bugs.webkit.org/show_bug.cgi?id=234424

Reviewed by Alan Bujtas.

We have 3 places with code of this form:

float result;
std::optional<float> wordTrailingSpaceWidth;
if (currentCharacterIsSpace)

wordTrailingSpaceWidth = wordTrailingSpace.width(fallbackFonts);

if (wordTrailingSpaceWidth)

result = measureText(startingCharacterIndex, length + 1) - wordTrailingSpaceWidth.value();

else

result = measureText(startingCharacterIndex, length);

This patch simply deduplicates this logic, and puts it into a single function:
RenderText::measureTextConsideringPossibleTrailingSpace(). Because the different places which call it
use different text measurement functions, this function accepts a "measure text" callback.

No new tests because there is no behavior change.

• WebCore.xcodeproj/project.pbxproj:
• rendering/RenderText.cpp:

(WebCore::RenderText::widthFromCacheConsideringPossibleTrailingSpace const):
(WebCore::RenderText::maxWordFragmentWidth):
(WebCore::RenderText::computePreferredLogicalWidths):

• rendering/RenderText.h:
• rendering/RenderTextInlines.h: Added. This function needs to be able to see into WordTrailingSpace,

but we shouldn't be adding any more #includes to RenderText.h because doing so will increase the build time.
So, instead, we can move this templated function to a new Inlines header, and have callers include that
header.
(WebCore::RenderText::measureTextConsideringPossibleTrailingSpace):

• rendering/line/BreakingContext.h:

(WebCore::BreakingContext::textWidthConsideringPossibleTrailingSpace):
(WebCore::BreakingContext::computeAdditionalBetweenWordsWidth):
(WebCore::WordTrailingSpace::WordTrailingSpace): Deleted.
(WebCore::WordTrailingSpace::width): Deleted.

• rendering/line/WordTrailingSpace.h: Added. BreakingContext.h defines WordTrailingSpace, but it needs

to be able to call measureTextConsideringPossibleTrailingSpace(), which needs to be able to see inside
WordTrailingSpace. Avoid the circular dependency by splitting out WordTrailingSpace into its own file.
(WebCore::WordTrailingSpace::WordTrailingSpace):
(WebCore::WordTrailingSpace::width):

1:51 PM Changeset in webkit [287205] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] Use the inline box's direction to decide if start/end decorations should be applied on first/last box
​https://bugs.webkit.org/show_bug.cgi?id=234423

Reviewed by Antti Koivisto.

• layout/formattingContexts/inline/InlineDisplayContentBuilder.cpp:

(WebCore::Layout::InlineDisplayContentBuilder::adjustVisualGeometryForDisplayBox):

1:39 PM Changeset in webkit [287204] by Fujii Hironori

• 3 edits in trunk/Source/ThirdParty/ANGLE

[ANGLE] "WARN: driver_utils.cpp:169 (rx::GetVendorString):! Unimplemented: rx::GetVendorString" on WinCairo debug testing bot
​https://bugs.webkit.org/show_bug.cgi?id=234371

Reviewed by Kenneth Russell.

After r286603 updated ANGLE, WinCairo debug testing bot was
observing annoying warning message for all WebGL tessts.

WARN: driver_utils.cpp:169 (rx::GetVendorString): ! Unimplemented: rx::GetVendorString(..\..\Source\ThirdParty\ANGLE\src\libANGLE\renderer\driver_utils.cpp:169)

The vendor ID was 0x1414 (Microsoft).
<​https://pcisig.com/membership/member-companies>

• src/libANGLE/renderer/driver_utils.cpp:

(rx::GetVendorString): Added more vendor strings for vendor IDs.

• src/libANGLE/renderer/driver_utils.h: Added VENDOR_ID_MICROSOFT.

1:05 PM Changeset in webkit [287203] by sbarati@apple.com

• 2 edits in trunk/Source/JavaScriptCore

Support WasmAddress in B3 CSE
​https://bugs.webkit.org/show_bug.cgi?id=234051
<rdar://problem/86552957>

Reviewed by Filip Pizlo and Yusuke Suzuki.

This patch adds support in B3's CSE phase to handle WasmAddressValue computations.
The reason this can't partake in pure CSE is that WasmAddressValue reads pinned.
To support this, we keep track of which blocks write pinned. If we're trying to
replace a value V2 with V1 because it appears there is a redundancy, we check if
any paths from V1 to V2 write pinned. If none do, we proceed with the replacement.

• b3/B3EliminateCommonSubexpressions.cpp:

12:55 PM Changeset in webkit [287202] by ysuzuki@apple.com

• 2 edits in trunk/Tools

[JSC] Fix runscript's standalone ability to run
​https://bugs.webkit.org/show_bug.cgi?id=234166

Reviewed by Mark Lam.

r286109 broke the ability to run runscript as a standalone script to test JSC stress test package on the device.
This patch reverts the ability back.

• Scripts/jsc-stress-test-helpers/shell-runner.sh:

12:08 PM Changeset in webkit [287201] by Alan Bujtas

• 2 edits in trunk/LayoutTests

Rebaseline after r287199.

Unreviewed.

• platform/ios/fast/borders/rtl-border-05-expected.txt:

11:53 AM Changeset in webkit [287200] by Fujii Hironori

• 2 edits in trunk/Source/WTF

MSVC reports "wtf/RetainPtr.h(196): error C3861: 'CFAutorelease': identifier not found " with /permissive- on Windows
​https://bugs.webkit.org/show_bug.cgi?id=202842

Reviewed by Alex Christensen and Darin Adler.

MSVC /std:c++20 option implicitly enables /permissive- option.
However, RetainPtr.h can't compile because 'CFAutorelease' is not
defined.

CFAutorelease isn't available in AppleWin's CoreFoundation and
WinCairo's OpenCFLite as of now.

• wtf/RetainPtr.h: Enclosed RetainPtr<T>::autorelease with #if !OS(WINDOWS).

11:44 AM Changeset in webkit [287199] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] Adjust the box geometry for atomic/generic inline level boxes in bidi content
​https://bugs.webkit.org/show_bug.cgi?id=234419

Reviewed by Antti Koivisto.

Both the display box and the BoxGeometry need to be updated while adjusting the content position with
the inline box offsets.

• layout/formattingContexts/inline/InlineDisplayContentBuilder.cpp:

(WebCore::Layout::InlineDisplayContentBuilder::adjustVisualGeometryForDisplayBox):

11:29 AM Changeset in webkit [287198] by commit-queue@webkit.org

• 17 edits
  3 deletes in trunk

Remove API::Object::Type::BundlePageGroup
​https://bugs.webkit.org/show_bug.cgi?id=225611

Patch by Alex Christensen <​achristensen@webkit.org> on 2021-12-17
Reviewed by Brady Eidson.

Source/WebKit:

Its last use was removed in rdar://77775952 and rdar://78783988

• Shared/API/APIObject.h:
• Shared/API/APIPageGroupHandle.cpp: Removed.
• Shared/API/APIPageGroupHandle.h: Removed.
• Shared/API/c/cg/WKImageCG.cpp:
• Shared/Cocoa/APIObject.mm:

(API::Object::newObject):

• Shared/UserData.cpp:

(WebKit::UserData::encode):
(WebKit::UserData::decode):

• Shared/mac/MediaFormatReader/CoreMediaWrapped.cpp:
• Sources.txt:
• UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::transformHandlesToObjects):
(WebKit::WebProcessProxy::transformObjectsToHandles):

• WebKit.xcodeproj/project.pbxproj:
• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInPageGroup.h:
• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInPageGroup.mm:

(-[WKWebProcessPlugInPageGroup identifier]): Deleted.
(-[WKWebProcessPlugInPageGroup dealloc]): Deleted.
(-[WKWebProcessPlugInPageGroup _apiObject]): Deleted.

• WebProcess/InjectedBundle/API/Cocoa/WKWebProcessPlugInPageGroupInternal.h: Removed.
• WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.h:
• WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.mm:

(-[WKWebProcessPlugInBrowserContextController pageGroup]): Deleted.

• WebProcess/WebPage/WebPageGroupProxy.h:
• WebProcess/WebProcess.cpp:

(WebKit::WebProcess::transformHandlesToObjects):
(WebKit::WebProcess::transformObjectsToHandles):

Tools:

• TestWebKitAPI/Tests/WebKit/DOMWindowExtensionBasic.cpp:

(TestWebKitAPI::TEST):

10:58 AM Changeset in webkit [287197] by Alan Bujtas

• 3 edits in trunk/Source/WebCore

[LFC][IFC] Let computeIsFirstIsLastBoxForInlineContent update the IsFirstForLayoutBox on the display boxes
​https://bugs.webkit.org/show_bug.cgi?id=234422

Reviewed by Antti Koivisto.

Currently we pre-compute these values and cache them on the display boxes (for later use in adjustVisualGeometryForDisplayBox).

• layout/formattingContexts/inline/InlineDisplayContentBuilder.cpp:

(WebCore::Layout::InlineDisplayContentBuilder::adjustVisualGeometryForDisplayBox):
(WebCore::Layout::InlineDisplayContentBuilder::processBidiContent):

• layout/formattingContexts/inline/InlineDisplayContentBuilder.h:

10:55 AM Changeset in webkit [287196] by clopez@igalia.com

• 4 edits in trunk/LayoutTests

Unskip pointerevents layout tests
​https://bugs.webkit.org/show_bug.cgi?id=233498

Unreviewed gardening.

This tests were skipped for all platforms.
Skip only globally the pointerevents/ios ones that seem to fail on
all platforms (even on iOS, at least on the open source builds).
On iOS also the ones imported from WPT fail but those work on
other platforms.

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

10:11 AM Changeset in webkit [287195] by mmaxfield@apple.com

• 3 edits in trunk/Source/WebCore

Refactor hyphenation logic in RenderText::computePreferredLogicalWidths()
​https://bugs.webkit.org/show_bug.cgi?id=234421

Reviewed by Alan Bujtas.

The hyphenation logic in RenderText::computePreferredLogicalWidths() was split across two functions.
If you consider the word "ABC-DEF-GHI" (where the '-' characters are hyphenation opportunities), one
function, maxWordFragmentWidth() was responsible for calculating the width of "ABC-" and "DEF-" but
not "GHI". RenderText::computePreferredLogicalWidths() called that function, and after it returned,
would then calculate the width of "GHI", and compare that to the width returned by
maxWordFragmentWidth(). A much simpler design would be to have one function, maxWordFragmentWidth(),
handle all the hyphenation logic, and do all the calculation necessary to just return a single value
to RenderText::computePreferredLogicalWidths().

No new tests because there is no behavior change.

• rendering/RenderText.cpp:

(WebCore::RenderText::maxWordFragmentWidth):
(WebCore::RenderText::computePreferredLogicalWidths):
(WebCore::maxWordFragmentWidth): Deleted.

• rendering/RenderText.h:

10:02 AM Changeset in webkit [287194] by commit-queue@webkit.org

• 3 edits
  2 adds in trunk

null ptr deref in WebCore::findPlaceForCounter
​https://bugs.webkit.org/show_bug.cgi?id=234375

Patch by Gabriel Nava Marino <​gnavamarino@apple.com> on 2021-12-17
Reviewed by Alan Bujtas.

Source/WebCore:

Test: fast/css/counters/findPlaceForCounter-parent-renderer-crash.html

When the current renderer's element is an HTML document and root of the tree, parentOrPseudoHostElement can return a nullptr,
so we need to check for nullptr before trying to access the renderer() of the Element pointer it returns.

• rendering/RenderCounter.cpp:

(WebCore::findPlaceForCounter):

LayoutTests:

• fast/css/counters/findPlaceForCounter-parent-renderer-crash-expected.txt: Added.
• fast/css/counters/findPlaceForCounter-parent-renderer-crash.html: Added.

9:42 AM Changeset in webkit [287193] by commit-queue@webkit.org

• 2 edits in trunk/Tools

Tools/Scripts/run-perf-tests is failing on Speedometer2
​https://bugs.webkit.org/show_bug.cgi?id=234426

Patch by Alexey Shvayka <​ashvayka@apple.com> on 2021-12-17
Reviewed by Carlos Alberto Lopez Perez.

r282357 added a debug console warning regarding 'dppx' units, which is being emitted
for TodoMVC's stylesheets, thus failing metrics parsing in PerfTest. To fix the execution
of Speedometer2 suite, this particular warning has now been added to the ignore list.

• Scripts/webkitpy/performance_tests/perftest.py:

(PerfTest):

9:31 AM Changeset in webkit [287192] by fpizlo@apple.com

• 36 edits
  4 adds in trunk/Source/bmalloc

[libpas] update to b1531141d8d5831b0a37efb70f200f7e2c37ff3d (initial pgm support, refactorings to further distinguish exclusive from shared)
​https://bugs.webkit.org/show_bug.cgi?id=234415

Reviewed by Yusuke Suzuki.

This includes two main changes:

• Initial PGM support thanks to Brandon Stewart.

• Refactorings to enable more differences in behavior between exclusive pages and shared pages. This is the only useful part of my attempt to add line allocation, which turned out not to be a win.

• bmalloc.xcodeproj/project.pbxproj:
• libpas/.gitignore:
• libpas/libpas.xcodeproj/project.pbxproj:
• libpas/src/libpas/jit_heap_config.c:

(jit_small_bitfit_create_page_header):
(jit_medium_bitfit_create_page_header):

• libpas/src/libpas/jit_heap_config.h:
• libpas/src/libpas/pas_bitfit_allocator.c:

(pas_bitfit_allocator_commit_view):

• libpas/src/libpas/pas_bitfit_page.c:

(pas_bitfit_page_construct):
(pas_bitfit_page_verify):

• libpas/src/libpas/pas_bitfit_page.h:

(pas_bitfit_page_offset_to_first_object):
(pas_bitfit_page_offset_to_end_of_last_object):

• libpas/src/libpas/pas_bitfit_page_config.h:

(pas_bitfit_page_config_object_payload_end_offset_from_boundary):

• libpas/src/libpas/pas_compact_tagged_void_ptr.h: Added.
• libpas/src/libpas/pas_enumerate_segregated_heaps.c:

(enumerate_exclusive_view):
(enumerate_shared_view):
(enumerate_partial_view):

• libpas/src/libpas/pas_heap_config_utils.h:
• libpas/src/libpas/pas_local_allocator_inlines.h:

(pas_local_allocator_set_up_free_bits):

• libpas/src/libpas/pas_page_base.c:

(pas_page_base_header_size):

• libpas/src/libpas/pas_page_base.h:
• libpas/src/libpas/pas_page_base_config.h:

(pas_page_base_config_is_bitfit):
(pas_page_base_config_object_payload_end_offset_from_boundary): Deleted.

• libpas/src/libpas/pas_page_base_config_utils.h:
• libpas/src/libpas/pas_page_base_config_utils_inlines.h:
• libpas/src/libpas/pas_probabilistic_guard_malloc_allocator.c: Added.

(pas_probabilistic_guard_malloc_allocate):
(pas_probabilistic_guard_malloc_deallocate):
(pas_probabilistic_guard_malloc_trigger):
(pas_probabilistic_guard_malloc_can_use):
(pas_probabilistic_guard_malloc_should_use):
(pas_probabilistic_guard_malloc_get_free_virtual_memory):
(pas_probabilistic_guard_malloc_get_free_wasted_memory):

• libpas/src/libpas/pas_probabilistic_guard_malloc_allocator.h: Added.
• libpas/src/libpas/pas_segregated_exclusive_view.c:

(compute_summary_impl):

• libpas/src/libpas/pas_segregated_heap.c:

(compute_ideal_object_size):
(pas_segregated_heap_ensure_size_directory_for_size):

• libpas/src/libpas/pas_segregated_page.c:

(pas_segregated_page_construct):
(pas_segregated_page_verify_granules):
(pas_segregated_page_verify_num_non_empty_words): Deleted.

• libpas/src/libpas/pas_segregated_page.h:

(pas_segregated_page_header_size):
(pas_segregated_page_offset_from_page_boundary_to_first_object_for_hugging_mode):
(pas_segregated_page_offset_from_page_boundary_to_end_of_last_object_for_hugging_mode):
(pas_segregated_page_useful_object_payload_size_for_hugging_mode):
(pas_segregated_page_best_hugging_mode):
(pas_segregated_page_offset_from_page_boundary_to_first_object_exclusive):
(pas_segregated_page_offset_from_page_boundary_to_end_of_last_object_exclusive):
(pas_segregated_page_useful_object_payload_size):
(pas_segregated_page_number_of_objects):
(pas_segregated_page_bytes_dirtied_per_object):

• libpas/src/libpas/pas_segregated_page_config.c:

(pas_segregated_page_config_validate):

• libpas/src/libpas/pas_segregated_page_config.h:

(pas_segregated_page_config_payload_offset_for_role):
(pas_segregated_page_config_payload_size_for_role):
(pas_segregated_page_config_payload_end_offset_for_role):
(pas_segregated_page_config_object_payload_end_offset_from_boundary): Deleted.

• libpas/src/libpas/pas_segregated_page_config_utils.h:
• libpas/src/libpas/pas_segregated_page_inlines.h:

(pas_segregated_page_initialize_full_use_counts):
(pas_segregated_page_offset_from_page_boundary_to_first_object): Deleted.

• libpas/src/libpas/pas_segregated_shared_view.c:

(pas_segregated_shared_view_commit_page):
(compute_summary):

• libpas/src/libpas/pas_segregated_shared_view.h:

(pas_segregated_shared_view_can_bump):
(pas_segregated_shared_view_compute_new_bump):

• libpas/src/libpas/pas_segregated_size_directory.c:

(pas_segregated_size_directory_enable_exclusive_views):

• libpas/src/libpas/pas_segregated_view.c:

(for_each_live_object):
(should_be_eligible):

• libpas/src/libpas/pas_segregated_view_allocator_inlines.h:

(pas_segregated_view_will_start_allocating):

• libpas/src/libpas/pas_status_reporter.c:

(report_segregated_directory_contents):

• libpas/src/libpas/pas_utility_heap_config.h:

(pas_utility_heap_create_page_header):

• libpas/src/test/IsoHeapChaosTests.cpp:

(std::addTheTests):

• libpas/src/test/PGMTests.cpp: Added.

(std::testPGMSingleAlloc):
(std::testPGMMultipleAlloc):
(std::testPGMErrors):
(addPGMTests):

• libpas/src/test/TestHarness.cpp:

(main):

• libpas/src/test/ThingyAndUtilityHeapAllocationTests.cpp:

(std::testFreeListRefillSpans):
(std::testSpuriousEligibility):

9:25 AM Changeset in webkit [287191] by Chris Dumez

• 5 edits in trunk/Source/WebKit

Move storage managers from the NetworkProcess to the NetworkSession class
​https://bugs.webkit.org/show_bug.cgi?id=234409

Reviewed by Alex Christensen.

Move storage managers from the NetworkProcess to the NetworkSession class since they are
per session.

• NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::lowMemoryHandler):
(WebKit::NetworkProcess::createNetworkConnectionToWebProcess):
(WebKit::NetworkProcess::addWebsiteDataStore):
(WebKit::NetworkProcess::destroySession):
(WebKit::NetworkProcess::hasLocalStorage):
(WebKit::NetworkProcess::fetchWebsiteData):
(WebKit::NetworkProcess::deleteWebsiteData):
(WebKit::NetworkProcess::deleteWebsiteDataForOrigins):
(WebKit::NetworkProcess::deleteAndRestrictWebsiteDataForRegistrableDomains):
(WebKit::NetworkProcess::registrableDomainsWithWebsiteData):
(WebKit::NetworkProcess::prepareToSuspend):
(WebKit::NetworkProcess::resume):
(WebKit::NetworkProcess::syncLocalStorage):
(WebKit::NetworkProcess::clearStorage):
(WebKit::NetworkProcess::renameOriginInWebsiteData):
(WebKit::NetworkProcess::connectionToWebProcessClosed):
(WebKit::NetworkProcess::addStorageManagerForSession): Deleted.
(WebKit::NetworkProcess::removeStorageManagerForSession): Deleted.

• NetworkProcess/NetworkProcess.h:
• NetworkProcess/NetworkSession.cpp:

(WebKit::NetworkSession::invalidateAndCancel):
(WebKit::NetworkSession::lowMemoryHandler):
(WebKit::NetworkSession::addStorageManagerSession):

• NetworkProcess/NetworkSession.h:

(WebKit::NetworkSession::storageManager):

9:13 AM Changeset in webkit [287190] by commit-queue@webkit.org

• 2 edits in trunk/Tools

Fix deprecation warning in Tools/Scripts/run-jsc-benchmarks
​https://bugs.webkit.org/show_bug.cgi?id=234425

Patch by Alexey Shvayka <​ashvayka@apple.com> on 2021-12-17
Reviewed by Alex Christensen.

Ruby 2.4 unified Fixnum and Bignum classes into Integer [1], deprecating them as
implementation details. To get rid of the deprecation console message, this patch replaces
Fixnum type checks with Integer, which is unobservable and backwards-compatible.

[1] ​https://bugs.ruby-lang.org/issues/12005

• Scripts/run-jsc-benchmarks:

8:03 AM Changeset in webkit [287189] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] Line spanning inline items should use InlineItem::opaqueBidiLevel
​https://bugs.webkit.org/show_bug.cgi?id=234417

Reviewed by Antti Koivisto.

The visual order of the non-empty inline boxes are driven by their content and not by
their guessed bidi levels.

• layout/formattingContexts/inline/InlineLineBuilder.cpp:

(WebCore::Layout::LineBuilder::initialize):

8:02 AM Changeset in webkit [287188] by Jonathan Bedard

• 5 edits in trunk/Tools

[reporelaypy] Accept pull-request hooks
​https://bugs.webkit.org/show_bug.cgi?id=234406
<rdar://problem/86598451>

Reviewed by Dewei Zhu.

• Tools/Scripts/libraries/reporelaypy/reporelaypy/init.py: Bump version.
• Tools/Scripts/libraries/reporelaypy/reporelaypy/hooks.py:

(HookProcessor):
(HookProcessor.is_valid): Check 'pull_request' hook format.

• Tools/Scripts/libraries/reporelaypy/reporelaypy/tests/hooks_unittest.py:

(HooksUnittest.test_pull_request):

• Tools/Scripts/libraries/reporelaypy/setup.py: Bump version.

Canonical link: ​https://commits.webkit.org/245355@main

8:01 AM Changeset in webkit [287187] by Alan Bujtas

• 3 edits in trunk/Source/WebCore

[LFC][IFC] Inline boxes set direction for both their content and their decoration unlike other inline level elements
​https://bugs.webkit.org/show_bug.cgi?id=234418

Reviewed by Antti Koivisto.

<div>

<span dir=rtl style="border-inline-start: 10px solid yellow;">

<span dir=ltr style="border-inline-start: 10px solid green;">

<img style="border-inline-start: 10px solid blue;">

</span>

</span

</div>

produces a "green/blue/yellow" pattern in visual order.

• layout/formattingContexts/inline/InlineDisplayContentBuilder.cpp:

(WebCore::Layout::InlineDisplayContentBuilder::adjustVisualGeometryForDisplayBox):

• layout/integration/LayoutIntegrationLineLayout.cpp:

(WebCore::LayoutIntegration::LineLayout::updateLayoutBoxDimensions):
(WebCore::LayoutIntegration::LineLayout::updateInlineBoxDimensions):

7:51 AM Changeset in webkit [287186] by graouts@webkit.org

• 8 edits in trunk/Source/WebCore

ActiveDOMObject::suspendIfNeeded() should not be called within constructors
​https://bugs.webkit.org/show_bug.cgi?id=233945

Reviewed by Chris Dumez.

Step 4 where we convert the only remaining call to suspendIfNeeded() in a constructor, AudioScheduledSourceNode,
to be in create() methods, in this case AudioScheduledSourceNode subclasses. This was the final step, all 145
calls to ActiveDOMObject::suspendIfNeeded() have been audited to not be made within a constructor.

• Modules/webaudio/AudioBufferSourceNode.cpp:

(WebCore::AudioBufferSourceNode::create):

• Modules/webaudio/AudioBufferSourceNode.h:
• Modules/webaudio/AudioScheduledSourceNode.cpp:

(WebCore::AudioScheduledSourceNode::AudioScheduledSourceNode):

• Modules/webaudio/AudioScheduledSourceNode.h:
• Modules/webaudio/ConstantSourceNode.cpp:

(WebCore::ConstantSourceNode::create):

• Modules/webaudio/OscillatorNode.cpp:

(WebCore::OscillatorNode::create):

• Modules/webaudio/OscillatorNode.h:

7:41 AM Changeset in webkit [287185] by Chris Dumez

• 2 edits in trunk/Source/WebKit

Unreviewed, reverting r286850.

Put r282794 back in. It turns out it was an overall
progression after all

Reverted changeset:

"Unreviewed, reverting r282794."
​https://bugs.webkit.org/show_bug.cgi?id=230517
​https://commits.webkit.org/r286850

7:17 AM Changeset in webkit [287184] by Alan Bujtas

• 2 edits in trunk/Source/WebCore

[LFC][IFC] Simple RTL content may need visual reordering
​https://bugs.webkit.org/show_bug.cgi?id=234380

Reviewed by Darin Adler and Antti Koivisto.

Addressing post-commit comment (r287142).

• layout/formattingContexts/inline/InlineItemsBuilder.cpp:

(WebCore::Layout::InlineItemsBuilder::breakAndComputeBidiLevels):

6:22 AM Changeset in webkit [287183] by Adrian Perez de Castro

• 7 edits in releases/WebKitGTK/webkit-2.34/Source

Merge r283444 - Remove JSC hack after r283410
​https://bugs.webkit.org/show_bug.cgi?id=230261

Reviewed by Mark Lam.

Source/JavaScriptCore:

• runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor):

• runtime/PropertySlot.h:

Source/WebCore:

We revert a hack for accessors in JSC since this is not necessary.

• bindings/scripts/CodeGeneratorJS.pm:

(GetJSCAttributesForAttribute):

• bindings/scripts/test/JS/JSTestObj.cpp:

(WebCore::JSTestObjDOMConstructor::construct):
(WebCore::jsTestObj_searchGetter):
(WebCore::JSC_DEFINE_CUSTOM_GETTER):
(WebCore::setJSTestObj_searchSetter):
(WebCore::JSC_DEFINE_CUSTOM_SETTER):

• bindings/scripts/test/TestObj.idl:

6:20 AM Changeset in webkit [287182] by Adrian Perez de Castro

• 17 edits in releases/WebKitGTK/webkit-2.34

Merge r283410 - Disable new incumbent-window until it is fully implemented
​https://bugs.webkit.org/show_bug.cgi?id=230261
<rdar://problem/83099726>

Reviewed by Chris Dumez and Geoffrey Garen.

LayoutTests/imported/w3c:

• web-platform-tests/html/webappapis/scripting/events/compile-event-handler-settings-objects-expected.txt:
• web-platform-tests/html/webappapis/scripting/processing-model-2/integration-with-the-javascript-job-queue/promise-job-incumbent-expected.txt:

Source/JavaScriptCore:

• interpreter/CallFrame.cpp:

(JSC::CallFrame::globalObjectOfClosestCodeBlock):

• runtime/JSObject.cpp:

(JSC::JSObject::getOwnPropertyDescriptor):

• runtime/PropertySlot.h:

Source/WebCore:

This patch disables new incumbent-window until it is fully implemented to avoid compatibility issue.

• bindings/js/JSDOMConvertCallbacks.h:

(WebCore::Converter<IDLCallbackFunction<T>>::convert):
(WebCore::Converter<IDLCallbackInterface<T>>::convert):

• bindings/js/JSDOMGlobalObject.cpp:

(WebCore::callerGlobalObject):
(WebCore::legacyActiveGlobalObjectForAccessor):

• bindings/js/JSDOMGlobalObject.h:
• bindings/js/JSDOMWindowBase.cpp:

(WebCore::incumbentDOMWindow):
(WebCore::legacyActiveDOMWindowForAccessor):

• bindings/js/JSDOMWindowBase.h:
• bindings/scripts/CodeGeneratorJS.pm:

(GetJSCAttributesForAttribute):
(GenerateCallWith):

• page/Location.idl:

LayoutTests:

• http/tests/security/frameNavigation/context-for-location-href-gopd-expected.txt:

5:55 AM Changeset in webkit [287181] by commit-queue@webkit.org

• 3 edits in trunk/Source/JavaScriptCore

[RISCV64] Implement linking and patching support in RISCV64Assembler
​https://bugs.webkit.org/show_bug.cgi?id=234398

Patch by Zan Dobersek <​zdobersek@igalia.com> on 2021-12-17
Reviewed by Yusuke Suzuki.

Populate RISCV64Assembler class with the necessary implementations and
facilities to support linking and patching operations. Implementations
of different methods in MacroAssemblerRISCV64 covering calls, jumps
and patches are also added.

Helper structs are added in the RISCV64Assembler class that cover
linking of jumps, calls or branches and patching of pointer values.
Relevant methods are also implemented to utilize these helpers.

RISCV64Assembler also gains helper methods that generate placeholders
for the specific type of linking or patching. The passed-in functor is
then invoked, enabling the user (MacroAssemblerRISCV64 implementation)
to additionally generate the desired instruction sequence that utilizes
the given placeholder.

In MacroAssemblerRISCV64, different noop methods covering jumps, calls,
linking and patching are replaced with the appropriate implementations.

• assembler/MacroAssemblerRISCV64.h:

(JSC::MacroAssemblerRISCV64::readCallTarget):
(JSC::MacroAssemblerRISCV64::replaceWithJump):
(JSC::MacroAssemblerRISCV64::startOfBranchPtrWithPatchOnRegister):
(JSC::MacroAssemblerRISCV64::revertJumpReplacementToBranchPtrWithPatch):
(JSC::MacroAssemblerRISCV64::linkCall):
(JSC::MacroAssemblerRISCV64::repatchCall):
(JSC::MacroAssemblerRISCV64::jump):
(JSC::MacroAssemblerRISCV64::farJump):
(JSC::MacroAssemblerRISCV64::nearCall):
(JSC::MacroAssemblerRISCV64::nearTailCall):
(JSC::MacroAssemblerRISCV64::threadSafePatchableNearCall):
(JSC::MacroAssemblerRISCV64::ret):
(JSC::MacroAssemblerRISCV64::call):
(JSC::MacroAssemblerRISCV64::callOperation):
(JSC::MacroAssemblerRISCV64::startOfPatchableBranchPtrWithPatchOnAddress): Deleted.
(JSC::MacroAssemblerRISCV64::startOfPatchableBranch32WithPatchOnAddress): Deleted.
(JSC::MacroAssemblerRISCV64::revertJumpReplacementToPatchableBranchPtrWithPatch): Deleted.
(JSC::MacroAssemblerRISCV64::revertJumpReplacementToPatchableBranch32WithPatch): Deleted.

• assembler/RISCV64Assembler.h:

(JSC::RISCV64Instructions::ImmediateBase::immediateMask):
(JSC::RISCV64Instructions::ImmediateBase::v):
(JSC::RISCV64Instructions::ImmediateDecomposition::ImmediateDecomposition):
(JSC::RISCV64Assembler::getDifferenceBetweenLabels):
(JSC::RISCV64Assembler::getCallReturnOffset):
(JSC::RISCV64Assembler::labelIgnoringWatchpoints):
(JSC::RISCV64Assembler::labelForWatchpoint):
(JSC::RISCV64Assembler::label):
(JSC::RISCV64Assembler::linkJump):
(JSC::RISCV64Assembler::linkCall):
(JSC::RISCV64Assembler::linkPointer):
(JSC::RISCV64Assembler::maxJumpReplacementSize):
(JSC::RISCV64Assembler::patchableJumpSize):
(JSC::RISCV64Assembler::repatchPointer):
(JSC::RISCV64Assembler::relinkJump):
(JSC::RISCV64Assembler::relinkJumpToNop):
(JSC::RISCV64Assembler::relinkCall):
(JSC::RISCV64Assembler::replaceWithJump):
(JSC::RISCV64Assembler::revertJumpReplacementToPatch):
(JSC::RISCV64Assembler::readCallTarget):
(JSC::RISCV64Assembler::cacheFlush):
(JSC::RISCV64Assembler::fillNops):
(JSC::RISCV64Assembler::jumpPlaceholder):
(JSC::RISCV64Assembler::branchPlaceholder):
(JSC::RISCV64Assembler::pointerCallPlaceholder):
(JSC::RISCV64Assembler::nearCallPlaceholder):
(JSC::RISCV64Assembler::LinkJumpOrCallImpl::apply):
(JSC::RISCV64Assembler::LinkJumpImpl::placeholderInsn):
(JSC::RISCV64Assembler::LinkJumpImpl::generatePlaceholder):
(JSC::RISCV64Assembler::LinkCallImpl::placeholderInsn):
(JSC::RISCV64Assembler::LinkCallImpl::generatePlaceholder):
(JSC::RISCV64Assembler::LinkBranchImpl::placeholderInsn):
(JSC::RISCV64Assembler::LinkBranchImpl::generatePlaceholder):
(JSC::RISCV64Assembler::LinkBranchImpl::apply):
(JSC::RISCV64Assembler::PatchPointerImpl::placeholderInsn):
(JSC::RISCV64Assembler::PatchPointerImpl::generatePlaceholder):
(JSC::RISCV64Assembler::PatchPointerImpl::apply):
(JSC::RISCV64Assembler::PatchPointerImpl::read):

5:10 AM WebKitGTK/2.34.x edited by Philippe Normand

(diff)

4:47 AM Changeset in webkit [287180] by commit-queue@webkit.org

• 11 edits in trunk

[GStreamer][WebRTC] Huge memory leak
​https://bugs.webkit.org/show_bug.cgi?id=234134

Patch by Philippe Normand <​pnormand@igalia.com> on 2021-12-17
Reviewed by Xabier Rodriguez-Calvar.

Source/WebCore:

The main issue was RealtimeOutgoingVideoSourceLibWebRTC leaking GstSamples. Fixing this lead
me to further clean-ups in the GstSample<->LibWebRTCVideoFrame handling. Native frames
should not consume GstSamples, but reference them, otherwise there are crashes where the
dangling GstSamples would be passed to the libwebrtc video decoder...

Also the video decoders were not reporting their initialization status correctly,
WEBRTC_VIDEO_CODEC_OK is 0 hence false...

• platform/graphics/gstreamer/MediaSampleGStreamer.cpp:

(WebCore::MediaSampleGStreamer::MediaSampleGStreamer):
(WebCore::MediaSampleGStreamer::initializeFromBuffer):

• platform/graphics/gstreamer/MediaSampleGStreamer.h:

(WebCore::MediaSampleGStreamer::createWrappedSample):

• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoDecoderFactory.cpp:

(WebCore::GStreamerVideoDecoder::pullSample):

• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoEncoderFactory.cpp:
• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoFrameLibWebRTC.cpp:

(WebCore::convertLibWebRTCVideoFrameToGStreamerSample):
(WebCore::convertGStreamerSampleToLibWebRTCVideoFrame):
(WebCore::GStreamerVideoFrameLibWebRTC::create):
(WebCore::GStreamerSampleFromLibWebRTCVideoFrame): Deleted.
(WebCore::LibWebRTCVideoFrameFromGStreamerSample): Deleted.

• platform/mediastream/libwebrtc/gstreamer/GStreamerVideoFrameLibWebRTC.h:

(WebCore::GStreamerVideoFrameLibWebRTC::GStreamerVideoFrameLibWebRTC):
(WebCore::GStreamerVideoFrameLibWebRTC::getSample const):
(WebCore::GStreamerVideoFrameLibWebRTC::takeSample): Deleted.

• platform/mediastream/libwebrtc/gstreamer/RealtimeIncomingVideoSourceLibWebRTC.cpp:

(WebCore::RealtimeIncomingVideoSourceLibWebRTC::OnFrame):

• platform/mediastream/libwebrtc/gstreamer/RealtimeOutgoingVideoSourceLibWebRTC.cpp:

(WebCore::RealtimeOutgoingVideoSourceLibWebRTC::videoSampleAvailable):

LayoutTests:

• platform/glib/TestExpectations: Update test expectations, #233740 was mostly due to video

decoder initialization wrongly reported as failed.
webrtc/captureCanvas-webrtc-software-h264-baseline.html no longer times out.

4:41 AM Changeset in webkit [287179] by commit-queue@webkit.org

• 2 edits
  1 copy
  3 moves
  1 add in trunk/PerformanceTests

[css-contain][Performance test] Add test contain-paint-text-nowrap.html
​https://bugs.webkit.org/show_bug.cgi?id=234001

Patch by Rob Buis <​rbuis@igalia.com> on 2021-12-17
Reviewed by Simon Fraser.

Add test contain-paint-text-nowrap.html to verify that paint containment
improves the case where many contained children have inline text that is
wider than the container width. The improvement is because paint
containment clips the child content.

This patch also introduces the Containment directory to separate the
containment tests.

• Containment/contain-paint-nowrap.html:
• Containment/css-contain-change-size.html: Renamed from PerformanceTests/Layout/css-contain-change-size.html.
• Containment/css-contain-layout-size-inside-complex-document.html: Renamed from PerformanceTests/Layout/css-contain-layout-size-inside-complex-document.html.
• Containment/large-grid.html: Renamed from PerformanceTests/Layout/large-grid.html.
• Paint/lots-of-self-painting-layers.html: use let instead of var

3:59 AM Changeset in webkit [287178] by commit-queue@webkit.org

• 2 edits in trunk/Source/ThirdParty/ANGLE

REGRESSION (r287068): [macOS arm64] TestWebKitAPI.GraphicsContextGLCocoaTest.UnrecycledDisplayBuffersNoLeaks is failing
​https://bugs.webkit.org/show_bug.cgi?id=234356
<rdar://problem/86533892>

Patch by Kimmo Kinnunen <​kkinnunen@apple.com> on 2021-12-17
Reviewed by Antti Koivisto.

Avoid leaking MTLTexture instances for render targets. Leak would happen through
temporary render pass descriptor that would ref the render target texture.
The descriptor was created without autoreleasepool in call stack.
Also fixes the error string leaks.

UnrecycledDisplayBuffersNoLeaks would report leaks of iosurface-backed
GraphicsContextGL default framebuffer bound textures if glClear was called.

• src/libANGLE/renderer/metal/ContextMtl.mm:

(rx::ContextMtl::getRenderPassCommandEncoder):

3:58 AM Changeset in webkit [287177] by graouts@webkit.org

• 34 edits in trunk/Source/WebCore

ActiveDOMObject::suspendIfNeeded() should not be called within constructors
​https://bugs.webkit.org/show_bug.cgi?id=233945

Reviewed by Darin Adler.

Step 3 where we convert almost all remaining calls to suspendIfNeeded() in constructors to be in create() methods.
This required adding such a method to ServiceWorkerContainer. The final call left is in the AudioScheduledSourceNode
constructor, but an earlier version of this patch failed a lot of webaudio tests on EWS so I suspect my attempted fix
was wrong.

• Modules/entriesapi/FileSystemDirectoryEntry.cpp:

(WebCore::FileSystemDirectoryEntry::create):

• Modules/entriesapi/FileSystemDirectoryEntry.h:
• Modules/entriesapi/FileSystemEntry.cpp:

(WebCore::FileSystemEntry::FileSystemEntry):

• Modules/entriesapi/FileSystemEntry.h:
• Modules/entriesapi/FileSystemFileEntry.cpp:

(WebCore::FileSystemFileEntry::create):

• Modules/entriesapi/FileSystemFileEntry.h:
• Modules/fetch/FetchBodyOwner.cpp:

(WebCore::FetchBodyOwner::FetchBodyOwner):

• Modules/fetch/FetchBodyOwner.h:
• Modules/fetch/FetchRequest.cpp:

(WebCore::FetchRequest::create):
(WebCore::FetchRequest::clone):

• Modules/fetch/FetchRequest.h:
• Modules/fetch/FetchResponse.cpp:

(WebCore::FetchResponse::create):
(WebCore::FetchResponse::error):
(WebCore::FetchResponse::redirect):
(WebCore::FetchResponse::fetch):

• Modules/filesystemaccess/FileSystemDirectoryHandle.cpp:

(WebCore::FileSystemDirectoryHandle::create):

• Modules/filesystemaccess/FileSystemFileHandle.cpp:

(WebCore::FileSystemFileHandle::create):

• Modules/filesystemaccess/FileSystemHandle.cpp:

(WebCore::FileSystemHandle::FileSystemHandle):

• Modules/filesystemaccess/FileSystemHandle.h:
• Modules/mediasource/SourceBufferList.cpp:

(WebCore::SourceBufferList::create):
(WebCore::SourceBufferList::SourceBufferList):

• Modules/mediasource/SourceBufferList.h:
• Modules/mediastream/MediaDevices.cpp:

(WebCore::MediaDevices::MediaDevices):
(WebCore::MediaDevices::create):

• Modules/mediastream/RTCDTMFSender.cpp:

(WebCore::RTCDTMFSender::create):
(WebCore::RTCDTMFSender::RTCDTMFSender):

• Modules/mediastream/RTCDTMFSender.h:
• Modules/mediastream/RTCDtlsTransport.cpp:

(WebCore::RTCDtlsTransport::create):
(WebCore::RTCDtlsTransport::RTCDtlsTransport):

• Modules/mediastream/RTCDtlsTransport.h:
• Modules/mediastream/RTCIceTransport.cpp:

(WebCore::RTCIceTransport::create):
(WebCore::RTCIceTransport::RTCIceTransport):

• Modules/mediastream/RTCIceTransport.h:

(WebCore::RTCIceTransport::create): Deleted.

• Modules/mediastream/RTCRtpSFrameTransform.cpp:

(WebCore::RTCRtpSFrameTransform::create):
(WebCore::RTCRtpSFrameTransform::RTCRtpSFrameTransform):

• Modules/mediastream/RTCRtpSFrameTransform.h:

(WebCore::RTCRtpSFrameTransform::create): Deleted.

• Modules/mediastream/RTCSctpTransport.cpp:

(WebCore::RTCSctpTransport::create):
(WebCore::RTCSctpTransport::RTCSctpTransport):

• Modules/mediastream/RTCSctpTransport.h:
• Modules/paymentrequest/PaymentResponse.cpp:

(WebCore::PaymentResponse::finishConstruction):

• Modules/webdatabase/DatabaseContext.cpp:

(WebCore::DatabaseContext::DatabaseContext):

• Modules/webdatabase/DatabaseManager.cpp:

(WebCore::DatabaseManager::databaseContext):

• html/ImageBitmap.cpp:
• page/NavigatorBase.cpp:

(WebCore::NavigatorBase::serviceWorker):

• workers/service/ServiceWorkerContainer.cpp:

(WebCore::ServiceWorkerContainer::create):
(WebCore::ServiceWorkerContainer::ServiceWorkerContainer):

• workers/service/ServiceWorkerContainer.h:

2:11 AM Changeset in webkit [287176] by Adrian Perez de Castro

• 15 edits
  2 adds in releases/WebKitGTK/webkit-2.34/Source

Merge r286676 - [GStreamer] Fill in client-name property on audio sinks
​https://bugs.webkit.org/show_bug.cgi?id=233826

Patch by Philippe Normand <​pnormand@igalia.com> on 2021-12-08
Reviewed by Xabier Rodriguez-Calvar.

The audio sink client-name property is now set on sinks that support this property, such as
pulsesink and pipewiresink. This is picked-up by GNOME settings and the corresponding audio
streams now appear as owned by the WebKit application (MiniBrowser, ephy, etc) instead of
"WebKitWebProcess".

When the audio mixer is enabled, no media.role property is set, because the mixer can accept
music and/or video roles simultaneously, so it doesn't make sense in this context.

• platform/audio/gstreamer/AudioDestinationGStreamer.cpp:

(WebCore::AudioDestinationGStreamer::AudioDestinationGStreamer): Drive-by refactoring, using
a lambda instead of a static callback function.
(WebCore::autoAudioSinkChildAddedCallback): Deleted.

• platform/graphics/gstreamer/GStreamerAudioMixer.cpp:

(WebCore::GStreamerAudioMixer::GStreamerAudioMixer):

• platform/graphics/gstreamer/GStreamerCommon.cpp:

(WebCore::createAutoAudioSink):
(WebCore::createPlatformAudioSink):

• platform/graphics/gstreamer/GStreamerCommon.h:
• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.cpp:

(WebCore::MediaPlayerPrivateGStreamer::~MediaPlayerPrivateGStreamer):
(WebCore::MediaPlayerPrivateGStreamer::createAudioSink):
(WebCore::MediaPlayerPrivateGStreamer::setAudioStreamPropertiesCallback): Deleted.
(WebCore::MediaPlayerPrivateGStreamer::setAudioStreamProperties): Deleted.

• platform/graphics/gstreamer/MediaPlayerPrivateGStreamer.h:

1:21 AM WebKitGTK/2.34.x edited by Adrian Perez de Castro

(diff)

1:01 AM Changeset in webkit [287175] by ysuzuki@apple.com

• 2 edits in trunk/Source/bmalloc

[libpas] Enable libpas on ARM64 Darwin platforms (ARM64E is already enabled)
​https://bugs.webkit.org/show_bug.cgi?id=234416

Reviewed by Filip Pizlo.

r287168 fixed libpas compare_and_swap on ARM64 so that we pass all libpas tests on ARM64 platforms.
Speedometer2 and JetStream2 are neutral, PLUM3 is 10.79% better, and RAMification is 2.83% better.

• bmalloc/BPlatform.h: