Timeline
Mar 10, 2017:
- 11:17 PM Changeset in webkit [213753] by
-
- 14 edits1 add in trunk/Source
The JITs should be able to emit fast TLS loads
https://bugs.webkit.org/show_bug.cgi?id=169483
Reviewed by Keith Miller.
Source/JavaScriptCore:
Added loadFromTLS32/64/Ptr to the MacroAssembler and added a B3 test for this.
- assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::mrs_TPIDRRO_EL0):
- assembler/MacroAssembler.h:
(JSC::MacroAssembler::loadFromTLSPtr):
- assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::loadFromTLS32):
(JSC::MacroAssemblerARM64::loadFromTLS64):
- assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::loadFromTLS32):
- assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::loadFromTLS64):
- assembler/X86Assembler.h:
(JSC::X86Assembler::adcl_im):
(JSC::X86Assembler::addl_mr):
(JSC::X86Assembler::addl_im):
(JSC::X86Assembler::andl_im):
(JSC::X86Assembler::orl_im):
(JSC::X86Assembler::orl_rm):
(JSC::X86Assembler::subl_im):
(JSC::X86Assembler::cmpb_im):
(JSC::X86Assembler::cmpl_rm):
(JSC::X86Assembler::cmpl_im):
(JSC::X86Assembler::testb_im):
(JSC::X86Assembler::movb_i8m):
(JSC::X86Assembler::movb_rm):
(JSC::X86Assembler::movl_mr):
(JSC::X86Assembler::movq_mr):
(JSC::X86Assembler::movsxd_rr):
(JSC::X86Assembler::gs):
(JSC::X86Assembler::X86InstructionFormatter::SingleInstructionBufferWriter::memoryModRM):
- b3/testb3.cpp:
(JSC::B3::testFastTLS):
(JSC::B3::run):
Source/WTF:
Consolidated what we know about fast TLS in FastTLS.h.
- WTF.xcodeproj/project.pbxproj:
- wtf/CMakeLists.txt:
- wtf/FastTLS.h: Added.
(WTF::loadFastTLS):
(WTF::fastTLSOffsetForKey):
- wtf/Platform.h:
- wtf/WTFThreadData.cpp:
(WTF::WTFThreadData::createAndRegisterForGetspecificDirect):
- wtf/WTFThreadData.h:
(WTF::wtfThreadData):
- 10:56 PM Changeset in webkit [213752] by
-
- 12 edits in trunk
Fix watch and tv builds after r213294
https://bugs.webkit.org/show_bug.cgi?id=169508
Reviewed by Dan Bernstein.
Source/JavaScriptCore:
- Configurations/FeatureDefines.xcconfig:
Source/WebCore:
- Configurations/FeatureDefines.xcconfig:
Source/WebCore/PAL:
- Configurations/FeatureDefines.xcconfig:
Source/WebKit/mac:
- Configurations/FeatureDefines.xcconfig:
Source/WebKit2:
- Configurations/FeatureDefines.xcconfig:
Tools:
- TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
- 10:02 PM Changeset in webkit [213751] by
-
- 2 edits in trunk/Tools
checkpatchrelevance: Exception message is printed twice due to ScriptError handler
https://bugs.webkit.org/show_bug.cgi?id=169496
Unreviewed fix.
Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2017-03-10
- Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py: Add --quiet flag to expected logs.
- 9:51 PM Changeset in webkit [213750] by
-
- 9 edits in tags/Safari-604.1.10/Source/WebKit2
Merge r213747. rdar://problem/29599569
- 8:30 PM Changeset in webkit [213749] by
-
- 3 edits in trunk/Tools
checkpatchrelevance: Exception message is printed twice due to ScriptError handler
https://bugs.webkit.org/show_bug.cgi?id=169496
Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2017-03-10
Reviewed by Alexey Proskuryakov.
- Scripts/webkitpy/tool/bot/patchanalysistask.py:
(PatchAnalysisTask._check_patch_relevance): Pass --quiet flag.
- Scripts/webkitpy/tool/steps/checkpatchrelevance.py:
(CheckPatchRelevance.options): Accept options.quiet to prevent duplicate output to stdout.
- 7:30 PM Changeset in webkit [213748] by
-
- 6 edits3 adds in trunk
Percentage constrained images shrinking inside blocks inside nested flex boxes
https://bugs.webkit.org/show_bug.cgi?id=169438
Reviewed by Dean Jackson.
Source/WebCore:
Added new test in fast/hidpi.
- rendering/RenderBlock.cpp:
(WebCore::RenderBlock::computeChildPreferredLogicalWidths):
Make sure to clear out flexbox override sizes when computing preferred logical widths,
since they should not ever influence the preferred size.
LayoutTests:
- fast/hidpi/percent-height-image-nested-expected.html: Added.
- fast/hidpi/percent-height-image-nested.html: Added.
- fast/hidpi/resources/pic.jpg: Added.
- 7:24 PM Changeset in webkit [213747] by
-
- 9 edits in trunk/Source/WebKit2
Add SPI to set NSURLSessionConfiguration.allowsCellularAccess from _WKProcessPoolConfiguration
https://bugs.webkit.org/show_bug.cgi?id=169500
<rdar://problem/29599569>
Reviewed by Joseph Pecoraro.
- NetworkProcess/NetworkProcessCreationParameters.cpp:
(WebKit::NetworkProcessCreationParameters::encode):
(WebKit::NetworkProcessCreationParameters::decode):
- NetworkProcess/NetworkProcessCreationParameters.h:
- NetworkProcess/cocoa/NetworkProcessCocoa.mm:
(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):
- NetworkProcess/cocoa/NetworkSessionCocoa.h:
- NetworkProcess/cocoa/NetworkSessionCocoa.mm:
(WebKit::globalAllowsCellularAccess):
(WebKit::NetworkSessionCocoa::setAllowsCellularAccess):
(WebKit::NetworkSessionCocoa::NetworkSessionCocoa):
- UIProcess/API/APIProcessPoolConfiguration.h:
- UIProcess/API/Cocoa/_WKProcessPoolConfiguration.h:
- UIProcess/API/Cocoa/_WKProcessPoolConfiguration.mm:
(-[_WKProcessPoolConfiguration allowsCellularAccess]):
(-[_WKProcessPoolConfiguration setAllowsCellularAccess:]):
- 6:12 PM Changeset in webkit [213746] by
-
- 5 edits1 delete in trunk/Source/WebKit2
Don't compile RemoteNetworkingContext when using NetworkSession
https://bugs.webkit.org/show_bug.cgi?id=169497
Reviewed by Andy Estes.
RemoteNetworkingContext is only really used because PingHandle requires a NetworkingContext.
When using NetworkSession, we use PingLoad instead of PingHandle, so there's no reason to
ensure there's a RemoteNetworkingContext corresponding to new private sessions when they're never used.
- NetworkProcess/NetworkProcess.cpp:
(WebKit::NetworkProcess::initializeNetworkProcess):
(WebKit::NetworkProcess::ensurePrivateBrowsingSession):
- NetworkProcess/RemoteNetworkingContext.h:
- NetworkProcess/mac/RemoteNetworkingContext.mm:
(WebKit::RemoteNetworkingContext::ensurePrivateBrowsingSession):
- NetworkProcess/soup/RemoteNetworkingContextSoup.cpp: Removed.
- PlatformGTK.cmake:
GTK only uses NetworkSession, so RemoteNetworkingContextSoup is no longer necessary.
- 6:04 PM Changeset in webkit [213745] by
-
- 20 edits1 copy2 adds in trunk
WebAssembly: Make more demos run
https://bugs.webkit.org/show_bug.cgi?id=165510
<rdar://problem/29760310>
Reviewed by Keith Miller.
JSTests:
- wasm/Builder.js:
(export.default.Builder.prototype._registerSectionBuilders.const.section.in.WASM.description.section.switch.section.case.string_appeared_here.this.section):
- wasm/js-api/wrapper-function.js: Added.
(exportImport):
(return.new.WebAssembly.Module):
(assert.throws.makeInstance):
(assert.throws.Bar):
(assert.throws):
Source/JavaScriptCore:
This patch makes another Wasm demo run:
https://kripken.github.io/BananaBread/cube2/bb.html
This patch fixes two bugs:
- When WebAssemblyFunctionType was added, we did not properly
update the last JS type value.
- Our code for our JS -> Wasm entrypoint was wrong. It lead to bad
code generation where we would emit B3 that would write over r12
and rbx (on x86) which is invalid since those are our pinned registers.
This patch just rewrites the entrypoint to use hand written assembler
code. I was planning on doing this anyways because it's a compile
time speed boost.
Also, this patch adds support for some new API features:
We can now export an import, either via a direct export, or via a Table and the
Element section. I've added a new class called WebAssemblyWrapperFunction that
just wraps over a JSObject that is a function. Wrapper functions have types
associated with them, so if they're re-imported, or called via call_indirect,
they can be type checked.
- CMakeLists.txt:
- JavaScriptCore.xcodeproj/project.pbxproj:
- runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):
(JSC::JSGlobalObject::visitChildren):
- runtime/JSGlobalObject.h:
(JSC::JSGlobalObject::webAssemblyWrapperFunctionStructure):
- runtime/JSType.h:
- wasm/JSWebAssemblyCodeBlock.h:
(JSC::JSWebAssemblyCodeBlock::wasmToJsCallStubForImport):
- wasm/WasmB3IRGenerator.cpp:
(JSC::Wasm::createJSToWasmWrapper):
- wasm/WasmCallingConvention.h:
(JSC::Wasm::CallingConvention::headerSizeInBytes):
- wasm/js/JSWebAssemblyHelpers.h:
(JSC::isWebAssemblyHostFunction):
- wasm/js/JSWebAssemblyInstance.cpp:
(JSC::JSWebAssemblyInstance::JSWebAssemblyInstance):
- wasm/js/JSWebAssemblyInstance.h:
(JSC::JSWebAssemblyInstance::importFunction):
(JSC::JSWebAssemblyInstance::importFunctions):
(JSC::JSWebAssemblyInstance::setImportFunction):
- wasm/js/JSWebAssemblyTable.cpp:
(JSC::JSWebAssemblyTable::JSWebAssemblyTable):
(JSC::JSWebAssemblyTable::grow):
(JSC::JSWebAssemblyTable::clearFunction):
(JSC::JSWebAssemblyTable::setFunction):
- wasm/js/JSWebAssemblyTable.h:
(JSC::JSWebAssemblyTable::getFunction):
- wasm/js/WebAssemblyFunction.cpp:
(JSC::callWebAssemblyFunction):
- wasm/js/WebAssemblyInstanceConstructor.cpp:
(JSC::WebAssemblyInstanceConstructor::createInstance):
- wasm/js/WebAssemblyModuleRecord.cpp:
(JSC::WebAssemblyModuleRecord::link):
(JSC::WebAssemblyModuleRecord::evaluate):
- wasm/js/WebAssemblyModuleRecord.h:
- wasm/js/WebAssemblyTablePrototype.cpp:
(JSC::webAssemblyTableProtoFuncGet):
(JSC::webAssemblyTableProtoFuncSet):
- wasm/js/WebAssemblyWrapperFunction.cpp: Added.
(JSC::callWebAssemblyWrapperFunction):
(JSC::WebAssemblyWrapperFunction::WebAssemblyWrapperFunction):
(JSC::WebAssemblyWrapperFunction::create):
(JSC::WebAssemblyWrapperFunction::finishCreation):
(JSC::WebAssemblyWrapperFunction::createStructure):
(JSC::WebAssemblyWrapperFunction::visitChildren):
- wasm/js/WebAssemblyWrapperFunction.h: Added.
(JSC::WebAssemblyWrapperFunction::signatureIndex):
(JSC::WebAssemblyWrapperFunction::wasmEntrypoint):
(JSC::WebAssemblyWrapperFunction::function):
- 5:56 PM Changeset in webkit [213744] by
-
- 3 edits in trunk/Source/WebInspectorUI
Web Inspector: Only show "Break on..." menu if backend supports DOM breakpoints
https://bugs.webkit.org/show_bug.cgi?id=169488
Reviewed by Joseph Pecoraro.
- UserInterface/Controllers/DOMBreakpointTreeController.js:
(WebInspector.DOMBreakpointTreeController.appendBreakpointContextMenuItems):
Added assert to catch future cases where UI is being added when
backend support doesn't exist.
- UserInterface/Views/DOMTreeElement.js:
(WebInspector.DOMTreeElement.prototype._populateNodeContextMenu):
Check for backend support.
- 5:39 PM Changeset in webkit [213743] by
-
- 2 edits in trunk/Source/WTF
Turn ENABLE(MASM_PROBE) on by default for OS(DARWIN) release builds.
https://bugs.webkit.org/show_bug.cgi?id=169493
Reviewed by Saam Barati.
MASM_PROBE was already enabled for debug builds. This change makes it so that we
don't have to rebuild the world every time we need to use it on a release build.
- wtf/Platform.h:
- 5:38 PM Changeset in webkit [213742] by
-
- 3 edits1 add in trunk
JSC: BindingNode::bindValue doesn't increase the scope's reference count.
https://bugs.webkit.org/show_bug.cgi?id=168546
<rdar://problem/30589551>
Reviewed by Saam Barati.
JSTests:
- stress/regress-168546.js: Added.
Source/JavaScriptCore:
We should protect the scope RegisterID with a RefPtr while it is still needed.
- bytecompiler/NodesCodegen.cpp:
(JSC::ForInNode::emitLoopHeader):
(JSC::ForOfNode::emitBytecode):
(JSC::BindingNode::bindValue):
- 5:13 PM Changeset in webkit [213741] by
-
- 2 edits in trunk/Tools
Add new EWS queues to bot watcher's dashboard.
https://bugs.webkit.org/show_bug.cgi?id=169485
Reviewed by Alexey Proskuryakov.
- BuildSlaveSupport/build.webkit.org-config/public_html/dashboard/Scripts/BubbleQueueServer.js:
(BubbleQueueServer): Add Bindings and JSC EWS queues.
- 5:01 PM Changeset in webkit [213740] by
-
- 4 edits in trunk/Websites/webkit.org
Switch back to the Twitter API for the Tweet widget
https://bugs.webkit.org/show_bug.cgi?id=168749
Reviewed by Lucas Forschler.
Updates the Twitter widget tile implementation to prefer the Twitter API to populate tweets
displayed through the widget. If the API fails for any reason, it falls back to using the
Tweet listener for updated tweets.
The Twitter API provides the benefit of embedding media and being able to display the full
URLs rather than Twitter-shortened URLs.
- wp-content/plugins/tweet-listener.php:
- wp-content/themes/webkit/style.css:
(.twitter-tile .tile-content):
(.twitter-tile.text-only):
(.twitter-tile .media):
(.twitter-tile img):
- wp-content/themes/webkit/widgets/twitter.php:
- 4:59 PM Changeset in webkit [213739] by
-
- 2 edits in trunk/Source/WebInspectorUI
Web Inspector: popover with compositing reasons in Layers sidebar has hard to read layout
https://bugs.webkit.org/show_bug.cgi?id=167081
Patch by Devin Rousso <Devin Rousso> on 2017-03-10
Reviewed by Brian Burg.
- UserInterface/Views/LayerTreeDetailsSidebarPanel.css:
(.layer-tree-popover):
(.layer-tree-popover p):
(.layer-tree-popover ul):
(body[dir=ltr] .layer-tree-popover ul):
(body[dir=rtl] .layer-tree-popover ul):
Showlist-item-styleand make the <p> element stand out more (larger and bold).
- 4:59 PM Changeset in webkit [213738] by
-
- 2 edits in trunk/Source/WebInspectorUI
Web Inspector: in Resources tab navigation bar, can't save the same file from context menu item more than once
https://bugs.webkit.org/show_bug.cgi?id=165665
Patch by Devin Rousso <Devin Rousso> on 2017-03-10
Reviewed by Brian Burg.
- UserInterface/Views/ContextMenuUtilities.js:
(WebInspector.appendContextMenuItemsForSourceCode):
Force the "Save As" dialog to appear when saving, as otherwise subsequent clicks on "Save"
in the context menu will just override the previously saved file.
- 4:58 PM Changeset in webkit [213737] by
-
- 3 edits in trunk/Source/WebKit/win
[Win] Re-export a few symbols for backwards compatibility
https://bugs.webkit.org/show_bug.cgi?id=169490
<rdar://problem/30983623>
Reviewed by Dean Jackson.
Re-introduce a workaround that re-exports a few JavaScriptCore functions
that had been erroneously exported by WebKit in older builds. This
is needed so that we do not break certain old software still being used.
- WebKitDLL.cpp:
(DllMain): Bind functions at startup.
(bindJavaScriptTrampoline): Find JavaScriptCore library functions and
re-export them from WebKit.dll.
- WebKitDLL.h: Only advertise these functions when building WebKt itself.
- 4:46 PM Changeset in webkit [213736] by
-
- 18 edits in trunk
Move libwebrtc backend to using tracks
https://bugs.webkit.org/show_bug.cgi?id=169472
Patch by Youenn Fablet <youenn@apple.com> on 2017-03-10
Reviewed by Alex Christensen.
Source/ThirdParty/libwebrtc:
- Source/webrtc/pc/rtcstatscollector.cc: Moving from using media stream to tracks.
Source/WebCore:
Covered by current test sets.
This change allows to move away from streams and use more tracks.
This allows future better alignment with the spec and better implementation of sender/receover/transceiver.
Small refactoring to use more Ref<> in WebRTC track-based code.
Added a notifyAddedTrack to notify the backend that a track was added.
For libwebrtc, this allows calling AddTrack before making the offer.
Updated mock libwebrtc backend accordingly.
- Modules/mediastream/MediaEndpointPeerConnection.cpp:
(WebCore::MediaEndpointPeerConnection::replaceTrack):
(WebCore::MediaEndpointPeerConnection::replaceTrackTask):
- Modules/mediastream/MediaEndpointPeerConnection.h:
- Modules/mediastream/PeerConnectionBackend.h:
(WebCore::PeerConnectionBackend::notifyAddedTrack):
- Modules/mediastream/RTCPeerConnection.cpp:
(WebCore::RTCPeerConnection::addTrack):
(WebCore::RTCPeerConnection::replaceTrack):
- Modules/mediastream/RTCPeerConnection.h:
- Modules/mediastream/RTCRtpSender.cpp:
(WebCore::RTCRtpSender::create):
(WebCore::RTCRtpSender::RTCRtpSender):
(WebCore::RTCRtpSender::setTrack):
- Modules/mediastream/RTCRtpSender.h:
- Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.cpp:
(WebCore::LibWebRTCMediaEndpoint::addTrack):
(WebCore::LibWebRTCMediaEndpoint::doCreateOffer):
(WebCore::LibWebRTCMediaEndpoint::doCreateAnswer):
(WebCore::LibWebRTCMediaEndpoint::getStats):
- Modules/mediastream/libwebrtc/LibWebRTCMediaEndpoint.h:
- Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.cpp:
(WebCore::LibWebRTCPeerConnectionBackend::doAddIceCandidate):
(WebCore::LibWebRTCPeerConnectionBackend::notifyAddedTrack):
- Modules/mediastream/libwebrtc/LibWebRTCPeerConnectionBackend.h:
- testing/MockLibWebRTCPeerConnection.cpp:
(WebCore::MockLibWebRTCPeerConnection::AddTrack):
(WebCore::MockLibWebRTCPeerConnection::RemoveTrack):
(WebCore::MockLibWebRTCPeerConnection::CreateOffer):
(WebCore::MockLibWebRTCPeerConnection::CreateAnswer):
- testing/MockLibWebRTCPeerConnection.h:
(WebCore::MockRtpSender::MockRtpSender):
(WebCore::MockRtpSender::ssrc):
(WebCore::MockRtpSender::media_type):
(WebCore::MockRtpSender::id):
(WebCore::MockRtpSender::stream_ids):
(WebCore::MockRtpSender::GetParameters):
(WebCore::MockRtpSender::SetParameters):
(WebCore::MockRtpSender::GetDtmfSender):
LayoutTests:
- 4:46 PM Changeset in webkit [213735] by
-
- 3 edits in trunk/LayoutTests
Mark compositing/regions/transformed-layer-inside-transformed-layer.html as failing on ios-simulator.
https://bugs.webkit.org/show_bug.cgi?id=169457
Unreviewed test gardening.
- platform/ios-simulator-wk1/TestExpectations:
- platform/ios-simulator/TestExpectations:
- 4:45 PM Changeset in webkit [213734] by
-
- 3 edits in trunk/Source/WebInspectorUI
Web Inspector: REGRESSION(r213149): scope bar items in elements sidebar have wrong padding/margin
https://bugs.webkit.org/show_bug.cgi?id=169204
Patch by Devin Rousso <Devin Rousso> on 2017-03-10
Reviewed by Brian Burg.
- UserInterface/Views/Sidebar.css:
(.sidebar > .navigation-bar):
Replacealign-contentwithalign-items, as the former causes the child elements to fill
as much vertical space as possible instead of only using the space necessary for its content.
- UserInterface/Views/ScopeRadioButtonNavigationItem.css:
(.scope-radio-button-navigation-item > .arrows):
Drive-by fix: ensure that the up/down arrows are not clickable, which would prevent the
<select> from showing.
- 4:33 PM Changeset in webkit [213733] by
-
- 3 edits2 adds in trunk
Simple line layout: Check how many orphans needed on the current page before breaking.
https://bugs.webkit.org/show_bug.cgi?id=169477
Reviewed by Antti Koivisto.
Source/WebCore:
Before breaking for the widows we actually need to check for orphans first.
Test: fast/multicol/simple-line-layout-orphans-and-widows.html
- rendering/SimpleLineLayoutPagination.cpp:
(WebCore::SimpleLineLayout::computeLineBreakIndex):
(WebCore::SimpleLineLayout::adjustLinePositionsForPagination):
LayoutTests:
- fast/multicol/simple-line-layout-orphans-and-widows-expected.html: Added.
- fast/multicol/simple-line-layout-orphans-and-widows.html: Added.
- 3:47 PM Changeset in webkit [213732] by
-
- 1 edit in trunk/Source/WebKit2/ChangeLog
Cleaned up merge markers inadvertently left in the file.
- 2:56 PM Changeset in webkit [213731] by
-
- 5 edits5 copies1 add in trunk/Source/WebCore
WebGPU: Backend - Textures and TextureDescriptors
https://bugs.webkit.org/show_bug.cgi?id=169365
<rdar://problem/30928716>
Reviewed by Jon Lee with help from Anders Carlsson.
Backend implementation of textures for WebGPU, which adds
the GPUTexture and GPUTextureDescriptor classes, as well
as a creation method on GPUDevice.
- PlatformMac.cmake:
- WebCore.xcodeproj/project.pbxproj:
- platform/graphics/cocoa/GPUTextureDescriptorMetal.mm: Added.
(WebCore::GPUTextureDescriptor::GPUTextureDescriptor):
(WebCore::GPUTextureDescriptor::width):
(WebCore::GPUTextureDescriptor::setWidth):
(WebCore::GPUTextureDescriptor::height):
(WebCore::GPUTextureDescriptor::setHeight):
(WebCore::GPUTextureDescriptor::sampleCount):
(WebCore::GPUTextureDescriptor::setSampleCount):
(WebCore::GPUTextureDescriptor::textureType):
(WebCore::GPUTextureDescriptor::setTextureType):
(WebCore::GPUTextureDescriptor::storageMode):
(WebCore::GPUTextureDescriptor::setStorageMode):
(WebCore::GPUTextureDescriptor::usage):
(WebCore::GPUTextureDescriptor::setUsage):
(WebCore::GPUTextureDescriptor::platformTextureDescriptor):
- platform/graphics/cocoa/GPUTextureMetal.mm:
(WebCore::GPUTexture::GPUTexture):
(WebCore::GPUTexture::width):
(WebCore::GPUTexture::height):
(WebCore::GPUTexture::platformTexture):
- platform/graphics/gpu/GPUDevice.cpp:
(WebCore::GPUDevice::createTexture):
- platform/graphics/gpu/GPUDevice.h:
- platform/graphics/gpu/GPUTexture.cpp:
(WebCore::GPUTexture::create):
(WebCore::GPUTexture::createFromExistingTexture):
(WebCore::GPUTexture::~GPUTexture):
(WebCore::GPUTexture::width):
(WebCore::GPUTexture::height):
- platform/graphics/gpu/GPUTexture.h:
- platform/graphics/gpu/GPUTextureDescriptor.cpp:
(WebCore::GPUTextureDescriptor::create):
(WebCore::GPUTextureDescriptor::~GPUTextureDescriptor):
(WebCore::GPUTextureDescriptor::width):
(WebCore::GPUTextureDescriptor::setWidth):
(WebCore::GPUTextureDescriptor::height):
(WebCore::GPUTextureDescriptor::setHeight):
(WebCore::GPUTextureDescriptor::sampleCount):
(WebCore::GPUTextureDescriptor::setSampleCount):
(WebCore::GPUTextureDescriptor::textureType):
(WebCore::GPUTextureDescriptor::setTextureType):
(WebCore::GPUTextureDescriptor::storageMode):
(WebCore::GPUTextureDescriptor::setStorageMode):
(WebCore::GPUTextureDescriptor::usage):
(WebCore::GPUTextureDescriptor::setUsage):
- platform/graphics/gpu/GPUTextureDescriptor.h:
- 2:43 PM Changeset in webkit [213730] by
-
- 2 edits in trunk/Tools
Skip an API test that isn't expected to pass on iOS 10.
- TestWebKitAPI/Tests/WebKit2Cocoa/QuickLook.mm:
- 2:22 PM Changeset in webkit [213729] by
-
- 2 edits in trunk/LayoutTests
Mark webrtc/libwebrtc/descriptionGetters.html as a flaky.
https://bugs.webkit.org/show_bug.cgi?id=169481
Unreviewed test gardening.
- platform/mac-wk2/TestExpectations:
- 12:54 PM Changeset in webkit [213728] by
-
- 2 edits in trunk/Source/WebKit2
[WebRTC][WK2][iOS] Support com.apple.coremedia.compressionsession for more WebRTC Cases
https://bugs.webkit.org/show_bug.cgi?id=169480
<rdar://problem/30978577>
Reviewed by Youenn Fablet.
Always allow 'com.apple.coremedia.compressionsession' since it is also needed when sending
canvas data across a peer connection.
- Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:
- 12:25 PM Changeset in webkit [213727] by
-
- 2 edits in trunk/Source/WebCore
[Win] Scrollbars buttons have incorrect size in HiDPI.
https://bugs.webkit.org/show_bug.cgi?id=169463
Reviewed by Alex Christensen.
There seems to be a bug in DrawThemeBackground when the device context is scaled.
We can work around this by scaling the drawing rectangle instead.
- platform/win/ScrollbarThemeWin.cpp:
(WebCore::ScrollbarThemeWin::paintButton):
- 12:17 PM Changeset in webkit [213726] by
-
- 2 edits in trunk/Tools
Add bindings-ews to queue status server
https://bugs.webkit.org/show_bug.cgi?id=169307
Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2017-03-10
Reviewed by Alexey Proskuryakov.
- QueueStatusServer/config/queues.py: Add "bindings-ews" to the list of ewses.
- 12:07 PM Changeset in webkit [213725] by
-
- 8 edits1 add in trunk
[QuickLook] Add WKNavigationDelegate SPI to notify when a password is requested
https://bugs.webkit.org/show_bug.cgi?id=169459
<rdar://problem/30621913>
Reviewed by Alex Christensen.
Source/WebKit2:
- UIProcess/API/Cocoa/WKNavigationDelegatePrivate.h:
- UIProcess/Cocoa/NavigationState.h:
- UIProcess/Cocoa/NavigationState.mm:
(WebKit::NavigationState::setNavigationDelegate):
(WebKit::NavigationState::didRequestPasswordForQuickLookDocument):
- UIProcess/ios/PageClientImplIOS.mm:
(WebKit::PageClientImpl::requestPasswordForQuickLookDocument):
Tools:
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
- TestWebKitAPI/Tests/WebKit2Cocoa/QuickLook.mm:
(runTest):
(TEST):
(-[QuickLookPasswordNavigationDelegate _webViewDidRequestPasswordForQuickLookDocument:]):
- 11:45 AM Changeset in webkit [213724] by
-
- 4 edits in trunk/Source
Fix CMake build.
Source/JavaScriptCore:
- CMakeLists.txt:
Make more forwarding headers so we can find WasmFaultSignalHandler.h from WebProcess.cpp.
Source/WebKit/mac:
- WebView/WebNotification.h:
Don't assume we have already included something that includes NSObject.h.
- 11:36 AM Changeset in webkit [213723] by
-
- 6 edits in trunk
Simple line layout: Oprhan lines with visual overflow does not work properly.
https://bugs.webkit.org/show_bug.cgi?id=169409
Reviewed by Antti Koivisto.
Source/WebCore:
Add visual overflow checking for block strut. This logic is copied from RenderBlockFlow::adjustLinePositionForPagination.
Covered by existing test.
- rendering/SimpleLineLayoutFunctions.cpp:
(WebCore::SimpleLineLayout::computeOverflow):
- rendering/SimpleLineLayoutFunctions.h:
- rendering/SimpleLineLayoutPagination.cpp:
(WebCore::SimpleLineLayout::setPageBreakForLine):
LayoutTests:
- 11:24 AM Changeset in webkit [213722] by
-
- 10 edits1 move in trunk/Tools
Add support for Bindings EWS
https://bugs.webkit.org/show_bug.cgi?id=169308
Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2017-03-10
Reviewed by Alexey Proskuryakov.
- Scripts/webkitpy/common/config/ews.json: Add bindings-ews. Can test without building; should_build = False.
- Scripts/webkitpy/common/net/bindingstestresults.py:
(BindingsTestResults.init): Removed 'errors'.
(BindingsTestResults.results_from_string): Ditto.
(BindingsTestResults.is_subset): Ditto.
(BindingsTestResults.equals): Ditto.
(BindingsTestResults.all_passed): Ditto.
(BindingsTestResults.failing_tests): Ditto.
- Scripts/webkitpy/common/net/bindingstestresults_unittest.py:
(BindingsTestResultsTest.test_results_from_string): Removed 'errors'.
(BindingsTestResultsTest.test_results_from_string_success): Removed 'errors'.
- Scripts/webkitpy/tool/bot/earlywarningsystemtask.py:
(EarlyWarningSystemTask.init): Add should_build parameter.
(EarlyWarningSystemTask.run): Check should_build before calling self._build().
- Scripts/webkitpy/tool/bot/retrylogic_unittest.py: Renamed from Tools/Scripts/webkitpy/tool/bot/jscews_unittest.py.
(MockBindingsEarlyWarningSystem): Based of MockJSCEarlyWarningSystem. Mocked so we can provide test results.
(MockBindingsEarlyWarningSystem.init): Sets attributes.
(MockBindingsEarlyWarningSystem.test_results): Returns test results provided by us, instead of JSON output.
(BindingsEarlyWarningSystemTest): Class to test retry logic in below situations.
(BindingsEarlyWarningSystemTest._results_indicate_all_passed): False if None or or has failures, else True.
(BindingsEarlyWarningSystemTest._create_task):
(BindingsEarlyWarningSystemTest.test_success_case): Clean patch on clean tree.
(BindingsEarlyWarningSystemTest.test_test_failure): Red patch on clean tree.
(BindingsEarlyWarningSystemTest.test_fix): Clean patch on red tree.
(BindingsEarlyWarningSystemTest.test_ineffective_patch): Red patch on red tree.
(BindingsEarlyWarningSystemTest.test_partially_effective_patch): Red patch on redder tree.
(BindingsEarlyWarningSystemTest.test_different_test_failures_in_patch_and_tree): Red patch on red tree.
- Scripts/webkitpy/tool/bot/patchanalysistask.py:
(PatchAnalysisTask._build_and_test_without_patch): Don't build if should_build is False (eg. on Bindings EWS).
(PatchAnalysisTask._retry_bindings_tests): Retry logic for Bindings tests.
(PatchAnalysisTask._test_patch): If tests failed on Bindings EWS, and call _retry_bindings_tests.
- Scripts/webkitpy/tool/commands/earlywarningsystem.py:
(AbstractEarlyWarningSystem.begin_work_queue): Use BindingsTestResultsReader on Bindings EWS.
(AbstractEarlyWarningSystem._create_task): Pass should_build argument.
(AbstractEarlyWarningSystem.load_ews_classes): Add should_build argument (True by default).
- Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
(TestBindingsEWS): Bindings EWS class, derived from AbstractEarlyWarningSystem.
(AbstractEarlyWarningSystemTest.test_failing_bindings_tests_message): Unit test message posted to Bugzilla.
(EarlyWarningSystemTest._default_expected_logs): Not building on all ewses anymore; make build line optional.
(EarlyWarningSystemTest.test_ews_name): Add bindings-ews.
- Scripts/webkitpy/tool/steps/checkpatchrelevance.py:
(CheckPatchRelevance): Add bindings paths.
(CheckPatchRelevance._changes_are_relevant): Cleaner way to check if we should perform this step.
- Scripts/webkitpy/tool/steps/steps_unittest.py: Add unit tests for Bindings EWS patch relevance.
(test_patch_relevant_bindings): Test for patch relevant to bindings.
(test_patch_not_relevant_bindings): Test for patch not relevant to bindings.
- 11:15 AM Changeset in webkit [213721] by
-
- 25 edits in tags/Safari-604.1.10
Merge r213715. rdar://problem/24709033
- 11:12 AM Changeset in webkit [213720] by
-
- 1 copy in tags/Safari-604.1.10
Tag Safari-604.1.10.
- 11:07 AM Changeset in webkit [213719] by
-
- 22 edits5 adds in trunk
[iOS WK2] Layer content blurry with nested perspective and transforms
https://bugs.webkit.org/show_bug.cgi?id=169457
Source/WebCore:
rdar://problem/29879484
Reviewed by Tim Horton.
We set rasterizationScale on CALayers to the same values as contentsScale, to
avoid blurry layers when CA rasterizes, which often happens with nested perspective
transforms.
However, in UI-side compositing, if the page is not zoomed, we never applied the
rasterizationScale in the UI process. This happened because the PlatformCALayerRemote
constructor set m_properties.contentsScale, but did not set the dirty bit that
triggers the application of contentsScale and rasterizationScale in RemoteLayerTreePropertyApplier.
The fix is to set this dirty bit.
The rest of the changes are for testing. Internals now exposes layerIDForElement()
for returns an internal (non-stable-across-loads) layerID, which can be passed
to UIScriptController.propertiesOfLayerWithID() in the UI process, which inspects
the built layer (UIView) hierarchy. propertiesOfLayerWithID() returns a dictionary
which the test can dump as JSON, or pull values out of.
A few #pragma once cleanups also.
Tests: compositing/ios/basic-layer-properties.html
compositing/ios/rasterization-scale.html
- testing/Internals.cpp:
(WebCore::Internals::layerIDForElement):
- testing/Internals.h:
- testing/Internals.idl:
Source/WebKit2:
rdar://problem/29879484
Reviewed by Tim Horton.
We set rasterizationScale on CALayers to the same values as contentsScale, to
avoid blurry layers when CA rasterizes, which often happens with nested perspective
transforms.
However, in UI-side compositing, if the page is not zoomed, we never applied the
rasterizationScale in the UI process. This happened because the PlatformCALayerRemote
constructor set m_properties.contentsScale, but did not set the dirty bit that
triggers the application of contentsScale and rasterizationScale in RemoteLayerTreePropertyApplier.
The fix is to set this dirty bit.
The rest of the changes are for testing. Internals now exposes layerIDForElement()
for returns an internal (non-stable-across-loads) layerID, which can be passed
to UIScriptController.propertiesOfLayerWithID() in the UI process, which inspects
the built layer (UIView) hierarchy. propertiesOfLayerWithID() returns a dictionary
which the test can dump as JSON, or pull values out of.
A few #pragma once cleanups also.
- UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _propertiesOfLayerWithID:]):
- UIProcess/API/Cocoa/WKWebViewPrivate.h:
- UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.h:
- UIProcess/mac/RemoteLayerTreeDrawingAreaProxy.mm:
(WebKit::RemoteLayerTreeDrawingAreaProxy::layerWithIDForTesting):
- UIProcess/mac/RemoteLayerTreeHost.h:
- UIProcess/mac/RemoteLayerTreeHost.mm:
(WebKit::RemoteLayerTreeHost::layerWithIDForTesting):
- WebProcess/WebPage/mac/PlatformCALayerRemote.cpp:
(WebKit::PlatformCALayerRemote::PlatformCALayerRemote):
Tools:
rdar://problem/29879484
Reviewed by Tim Horton.
We set rasterizationScale on CALayers to the same values as contentsScale, to
avoid blurry layers when CA rasterizes, which often happens with nested perspective
transforms.
However, in UI-side compositing, if the page is not zoomed, we never applied the
rasterizationScale in the UI process. This happened because the PlatformCALayerRemote
constructor set m_properties.contentsScale, but did not set the dirty bit that
triggers the application of contentsScale and rasterizationScale in RemoteLayerTreePropertyApplier.
The fix is to set this dirty bit.
The rest of the changes are for testing. Internals now exposes layerIDForElement()
for returns an internal (non-stable-across-loads) layerID, which can be passed
to UIScriptController.propertiesOfLayerWithID() in the UI process, which inspects
the built layer (UIView) hierarchy. propertiesOfLayerWithID() returns a dictionary
which the test can dump as JSON, or pull values out of.
A few #pragma once cleanups also.
- DumpRenderTree/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::propertiesOfLayerWithID):
- TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
- TestRunnerShared/UIScriptContext/UIScriptController.cpp:
(WTR::UIScriptController::propertiesOfLayerWithID):
- TestRunnerShared/UIScriptContext/UIScriptController.h:
- WebKitTestRunner/ios/UIScriptControllerIOS.mm:
(WTR::UIScriptController::propertiesOfLayerWithID):
LayoutTests:
Reviewed by Tim Horton.
We set rasterizationScale on CALayers to the same values as contentsScale, to
avoid blurry layers when CA rasterizes, which often happens with nested perspective
transforms.
However, in UI-side compositing, if the page is not zoomed, we never applied the
rasterizationScale in the UI process. This happened because the PlatformCALayerRemote
constructor set m_properties.contentsScale, but did not set the dirty bit that
triggers the application of contentsScale and rasterizationScale in RemoteLayerTreePropertyApplier.
The fix is to set this dirty bit.
The rest of the changes are for testing. Internals now exposes layerIDForElement()
for returns an internal (non-stable-across-loads) layerID, which can be passed
to UIScriptController.propertiesOfLayerWithID() in the UI process, which inspects
the built layer (UIView) hierarchy. propertiesOfLayerWithID() returns a dictionary
which the test can dump as JSON, or pull values out of.
A few #pragma once cleanups also.
- TestExpectations:
- compositing/ios/basic-layer-properties-expected.txt: Added.
- compositing/ios/basic-layer-properties.html: Added.
- compositing/ios/rasterization-scale-expected.txt: Added.
- compositing/ios/rasterization-scale.html: Added.
- platform/ios-simulator-wk2/TestExpectations:
- 11:01 AM Changeset in webkit [213718] by
-
- 3 edits2 adds in trunk/Source/JavaScriptCore
[Re-landing] Implement a StackTrace utility object that can capture stack traces for debugging.
https://bugs.webkit.org/show_bug.cgi?id=169454
Reviewed by Michael Saboff.
The underlying implementation is hoisted right out of Assertions.cpp from the
implementations of WTFPrintBacktrace().
The reason we need this StackTrace object is because during heap debugging, we
sometimes want to capture the stack trace that allocated the objects of interest.
Dumping the stack trace directly to stdout (using WTFReportBacktrace()) may
perturb the execution profile sufficiently that an issue may not reproduce,
while alternatively, just capturing the stack trace and deferring printing it
till we actually need it later perturbs the execution profile less.
In addition, just capturing the stack traces (instead of printing them
immediately at each capture site) allows us to avoid polluting stdout with tons
of stack traces that may be irrelevant.
For now, we only capture the native stack trace. We'll leave capturing and
integrating the JS stack trace as an exercise for the future if we need it then.
Here's an example of how to use this StackTrace utility:
Capture a stack trace of the top 10 frames.
std::unique_ptr<StackTrace> trace(StackTrace::captureStackTrace(10));
Print the trace.
dataLog(*trace);
- CMakeLists.txt:
- JavaScriptCore.xcodeproj/project.pbxproj:
- tools/StackTrace.cpp: Added.
(JSC::StackTrace::instanceSize):
(JSC::StackTrace::captureStackTrace):
(JSC::StackTrace::dump):
- tools/StackTrace.h: Added.
(JSC::StackTrace::size):
(JSC::StackTrace::StackTrace):
- 10:32 AM Changeset in webkit [213717] by
-
- 1 delete in tags/Safari-604.1.10
Deleting tag.
- 10:20 AM Changeset in webkit [213716] by
-
- 1 copy in tags/Safari-604.1.10
Tag Safari-604.1.10.
- 10:05 AM Changeset in webkit [213715] by
-
- 25 edits in trunk
Unreviewed, rolling out r213618.
https://bugs.webkit.org/show_bug.cgi?id=169475
Suspect this is the cause of a large memory regression
(Requested by jonlee_ on #webkit).
Reverted changeset:
"Enable async image decoding for large images"
https://bugs.webkit.org/show_bug.cgi?id=165039
http://trac.webkit.org/changeset/213618
- 9:49 AM Changeset in webkit [213714] by
-
- 57 edits8 adds2 deletes in trunk
B3 should have comprehensive support for atomic operations
https://bugs.webkit.org/show_bug.cgi?id=162349
Reviewed by Keith Miller.
Source/JavaScriptCore:
This adds the following capabilities to B3:
- Atomic weak/strong unfenced/fenced compare-and-swap
- Atomic add/sub/or/and/xor/xchg
- Acquire/release fencing on loads/stores
- Fenceless load-load dependencies
This adds lowering to the following instructions on x86:
- lock cmpxchg
- lock xadd
- lock add/sub/or/and/xor/xchg
This adds lowering to the following instructions on ARM64:
- ldar and friends
- stlr and friends
- ldxr and friends (unfenced LL)
- stxr and friends (unfended SC)
- ldaxr and friends (fenced LL)
- stlxr and friends (fenced SC)
- eor as a fenceless load-load dependency
This does instruction selection pattern matching to ensure that weak/strong CAS and all of the
variants of fences and atomic math ops get lowered to the best possible instruction sequence.
For example, we support the Equal(AtomicStrongCAS(expected, ...), expected) pattern and a bunch
of its friends. You can say Branch(Equal(AtomicStrongCAS(expected, ...), expected)) and it will
generate the best possible branch sequence on x86 and ARM64.
B3 now knows how to model all of the kinds of fencing. It knows that acq loads are ordered with
respect to each other and with respect to rel stores, creating sequential consistency that
transcends just the acq/rel fences themselves (see Effects::fence). It knows that the phantom
fence effects may only target some abstract heaps but not others, so that load elimination and
store sinking can still operate across fences if you just tell B3 that the fence does not alias
those accesses. This makes it super easy to teach B3 that some of your heap is thread-local.
Even better, it lets you express fine-grained dependencies where the atomics that affect one
property in shared memory do not clobber non-atomics that ffect some other property in shared
memory.
One of my favorite features is Depend, which allows you to express load-load dependencies. On
x86 it lowers to nothing, while on ARM64 it lowers to eor.
This also exposes a common atomicWeakCAS API to the x86_64/ARM64 MacroAssemblers. Same for
acq/rel. JSC's 64-bit JITs are now a happy concurrency playground.
This doesn't yet expose the functionality to JS or wasm. SAB still uses the non-intrinsic
implementations of the Atomics object, for now.
- CMakeLists.txt:
- JavaScriptCore.xcodeproj/project.pbxproj:
- assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::ldar):
(JSC::ARM64Assembler::ldxr):
(JSC::ARM64Assembler::ldaxr):
(JSC::ARM64Assembler::stxr):
(JSC::ARM64Assembler::stlr):
(JSC::ARM64Assembler::stlxr):
(JSC::ARM64Assembler::excepnGenerationImmMask):
(JSC::ARM64Assembler::exoticLoad):
(JSC::ARM64Assembler::storeRelease):
(JSC::ARM64Assembler::exoticStore):
- assembler/AbstractMacroAssembler.cpp: Added.
(WTF::printInternal):
- assembler/AbstractMacroAssembler.h:
(JSC::AbstractMacroAssemblerBase::invert):
- assembler/MacroAssembler.h:
- assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::loadAcq8SignedExtendTo32):
(JSC::MacroAssemblerARM64::loadAcq8):
(JSC::MacroAssemblerARM64::storeRel8):
(JSC::MacroAssemblerARM64::loadAcq16SignedExtendTo32):
(JSC::MacroAssemblerARM64::loadAcq16):
(JSC::MacroAssemblerARM64::storeRel16):
(JSC::MacroAssemblerARM64::loadAcq32):
(JSC::MacroAssemblerARM64::loadAcq64):
(JSC::MacroAssemblerARM64::storeRel32):
(JSC::MacroAssemblerARM64::storeRel64):
(JSC::MacroAssemblerARM64::loadLink8):
(JSC::MacroAssemblerARM64::loadLinkAcq8):
(JSC::MacroAssemblerARM64::storeCond8):
(JSC::MacroAssemblerARM64::storeCondRel8):
(JSC::MacroAssemblerARM64::loadLink16):
(JSC::MacroAssemblerARM64::loadLinkAcq16):
(JSC::MacroAssemblerARM64::storeCond16):
(JSC::MacroAssemblerARM64::storeCondRel16):
(JSC::MacroAssemblerARM64::loadLink32):
(JSC::MacroAssemblerARM64::loadLinkAcq32):
(JSC::MacroAssemblerARM64::storeCond32):
(JSC::MacroAssemblerARM64::storeCondRel32):
(JSC::MacroAssemblerARM64::loadLink64):
(JSC::MacroAssemblerARM64::loadLinkAcq64):
(JSC::MacroAssemblerARM64::storeCond64):
(JSC::MacroAssemblerARM64::storeCondRel64):
(JSC::MacroAssemblerARM64::atomicStrongCAS8):
(JSC::MacroAssemblerARM64::atomicStrongCAS16):
(JSC::MacroAssemblerARM64::atomicStrongCAS32):
(JSC::MacroAssemblerARM64::atomicStrongCAS64):
(JSC::MacroAssemblerARM64::atomicRelaxedStrongCAS8):
(JSC::MacroAssemblerARM64::atomicRelaxedStrongCAS16):
(JSC::MacroAssemblerARM64::atomicRelaxedStrongCAS32):
(JSC::MacroAssemblerARM64::atomicRelaxedStrongCAS64):
(JSC::MacroAssemblerARM64::branchAtomicWeakCAS8):
(JSC::MacroAssemblerARM64::branchAtomicWeakCAS16):
(JSC::MacroAssemblerARM64::branchAtomicWeakCAS32):
(JSC::MacroAssemblerARM64::branchAtomicWeakCAS64):
(JSC::MacroAssemblerARM64::branchAtomicRelaxedWeakCAS8):
(JSC::MacroAssemblerARM64::branchAtomicRelaxedWeakCAS16):
(JSC::MacroAssemblerARM64::branchAtomicRelaxedWeakCAS32):
(JSC::MacroAssemblerARM64::branchAtomicRelaxedWeakCAS64):
(JSC::MacroAssemblerARM64::depend32):
(JSC::MacroAssemblerARM64::depend64):
(JSC::MacroAssemblerARM64::loadLink):
(JSC::MacroAssemblerARM64::loadLinkAcq):
(JSC::MacroAssemblerARM64::storeCond):
(JSC::MacroAssemblerARM64::storeCondRel):
(JSC::MacroAssemblerARM64::signExtend):
(JSC::MacroAssemblerARM64::branch):
(JSC::MacroAssemblerARM64::atomicStrongCAS):
(JSC::MacroAssemblerARM64::atomicRelaxedStrongCAS):
(JSC::MacroAssemblerARM64::branchAtomicWeakCAS):
(JSC::MacroAssemblerARM64::branchAtomicRelaxedWeakCAS):
(JSC::MacroAssemblerARM64::extractSimpleAddress):
(JSC::MacroAssemblerARM64::signExtend<8>):
(JSC::MacroAssemblerARM64::signExtend<16>):
(JSC::MacroAssemblerARM64::branch<64>):
- assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::add32):
(JSC::MacroAssemblerX86Common::and32):
(JSC::MacroAssemblerX86Common::and16):
(JSC::MacroAssemblerX86Common::and8):
(JSC::MacroAssemblerX86Common::neg32):
(JSC::MacroAssemblerX86Common::neg16):
(JSC::MacroAssemblerX86Common::neg8):
(JSC::MacroAssemblerX86Common::or32):
(JSC::MacroAssemblerX86Common::or16):
(JSC::MacroAssemblerX86Common::or8):
(JSC::MacroAssemblerX86Common::sub16):
(JSC::MacroAssemblerX86Common::sub8):
(JSC::MacroAssemblerX86Common::sub32):
(JSC::MacroAssemblerX86Common::xor32):
(JSC::MacroAssemblerX86Common::xor16):
(JSC::MacroAssemblerX86Common::xor8):
(JSC::MacroAssemblerX86Common::not32):
(JSC::MacroAssemblerX86Common::not16):
(JSC::MacroAssemblerX86Common::not8):
(JSC::MacroAssemblerX86Common::store16):
(JSC::MacroAssemblerX86Common::atomicStrongCAS8):
(JSC::MacroAssemblerX86Common::atomicStrongCAS16):
(JSC::MacroAssemblerX86Common::atomicStrongCAS32):
(JSC::MacroAssemblerX86Common::branchAtomicStrongCAS8):
(JSC::MacroAssemblerX86Common::branchAtomicStrongCAS16):
(JSC::MacroAssemblerX86Common::branchAtomicStrongCAS32):
(JSC::MacroAssemblerX86Common::atomicWeakCAS8):
(JSC::MacroAssemblerX86Common::atomicWeakCAS16):
(JSC::MacroAssemblerX86Common::atomicWeakCAS32):
(JSC::MacroAssemblerX86Common::branchAtomicWeakCAS8):
(JSC::MacroAssemblerX86Common::branchAtomicWeakCAS16):
(JSC::MacroAssemblerX86Common::branchAtomicWeakCAS32):
(JSC::MacroAssemblerX86Common::atomicRelaxedWeakCAS8):
(JSC::MacroAssemblerX86Common::atomicRelaxedWeakCAS16):
(JSC::MacroAssemblerX86Common::atomicRelaxedWeakCAS32):
(JSC::MacroAssemblerX86Common::branchAtomicRelaxedWeakCAS8):
(JSC::MacroAssemblerX86Common::branchAtomicRelaxedWeakCAS16):
(JSC::MacroAssemblerX86Common::branchAtomicRelaxedWeakCAS32):
(JSC::MacroAssemblerX86Common::atomicAdd8):
(JSC::MacroAssemblerX86Common::atomicAdd16):
(JSC::MacroAssemblerX86Common::atomicAdd32):
(JSC::MacroAssemblerX86Common::atomicSub8):
(JSC::MacroAssemblerX86Common::atomicSub16):
(JSC::MacroAssemblerX86Common::atomicSub32):
(JSC::MacroAssemblerX86Common::atomicAnd8):
(JSC::MacroAssemblerX86Common::atomicAnd16):
(JSC::MacroAssemblerX86Common::atomicAnd32):
(JSC::MacroAssemblerX86Common::atomicOr8):
(JSC::MacroAssemblerX86Common::atomicOr16):
(JSC::MacroAssemblerX86Common::atomicOr32):
(JSC::MacroAssemblerX86Common::atomicXor8):
(JSC::MacroAssemblerX86Common::atomicXor16):
(JSC::MacroAssemblerX86Common::atomicXor32):
(JSC::MacroAssemblerX86Common::atomicNeg8):
(JSC::MacroAssemblerX86Common::atomicNeg16):
(JSC::MacroAssemblerX86Common::atomicNeg32):
(JSC::MacroAssemblerX86Common::atomicNot8):
(JSC::MacroAssemblerX86Common::atomicNot16):
(JSC::MacroAssemblerX86Common::atomicNot32):
(JSC::MacroAssemblerX86Common::atomicXchgAdd8):
(JSC::MacroAssemblerX86Common::atomicXchgAdd16):
(JSC::MacroAssemblerX86Common::atomicXchgAdd32):
(JSC::MacroAssemblerX86Common::atomicXchg8):
(JSC::MacroAssemblerX86Common::atomicXchg16):
(JSC::MacroAssemblerX86Common::atomicXchg32):
(JSC::MacroAssemblerX86Common::loadAcq8):
(JSC::MacroAssemblerX86Common::loadAcq8SignedExtendTo32):
(JSC::MacroAssemblerX86Common::loadAcq16):
(JSC::MacroAssemblerX86Common::loadAcq16SignedExtendTo32):
(JSC::MacroAssemblerX86Common::loadAcq32):
(JSC::MacroAssemblerX86Common::storeRel8):
(JSC::MacroAssemblerX86Common::storeRel16):
(JSC::MacroAssemblerX86Common::storeRel32):
(JSC::MacroAssemblerX86Common::storeFence):
(JSC::MacroAssemblerX86Common::loadFence):
(JSC::MacroAssemblerX86Common::replaceWithJump):
(JSC::MacroAssemblerX86Common::maxJumpReplacementSize):
(JSC::MacroAssemblerX86Common::patchableJumpSize):
(JSC::MacroAssemblerX86Common::supportsFloatingPointRounding):
(JSC::MacroAssemblerX86Common::supportsAVX):
(JSC::MacroAssemblerX86Common::updateEax1EcxFlags):
(JSC::MacroAssemblerX86Common::x86Condition):
(JSC::MacroAssemblerX86Common::atomicStrongCAS):
(JSC::MacroAssemblerX86Common::branchAtomicStrongCAS):
- assembler/MacroAssemblerX86_64.h:
(JSC::MacroAssemblerX86_64::add64):
(JSC::MacroAssemblerX86_64::and64):
(JSC::MacroAssemblerX86_64::neg64):
(JSC::MacroAssemblerX86_64::or64):
(JSC::MacroAssemblerX86_64::sub64):
(JSC::MacroAssemblerX86_64::xor64):
(JSC::MacroAssemblerX86_64::not64):
(JSC::MacroAssemblerX86_64::store64):
(JSC::MacroAssemblerX86_64::atomicStrongCAS64):
(JSC::MacroAssemblerX86_64::branchAtomicStrongCAS64):
(JSC::MacroAssemblerX86_64::atomicWeakCAS64):
(JSC::MacroAssemblerX86_64::branchAtomicWeakCAS64):
(JSC::MacroAssemblerX86_64::atomicRelaxedWeakCAS64):
(JSC::MacroAssemblerX86_64::branchAtomicRelaxedWeakCAS64):
(JSC::MacroAssemblerX86_64::atomicAdd64):
(JSC::MacroAssemblerX86_64::atomicSub64):
(JSC::MacroAssemblerX86_64::atomicAnd64):
(JSC::MacroAssemblerX86_64::atomicOr64):
(JSC::MacroAssemblerX86_64::atomicXor64):
(JSC::MacroAssemblerX86_64::atomicNeg64):
(JSC::MacroAssemblerX86_64::atomicNot64):
(JSC::MacroAssemblerX86_64::atomicXchgAdd64):
(JSC::MacroAssemblerX86_64::atomicXchg64):
(JSC::MacroAssemblerX86_64::loadAcq64):
(JSC::MacroAssemblerX86_64::storeRel64):
- assembler/X86Assembler.h:
(JSC::X86Assembler::addl_mr):
(JSC::X86Assembler::addq_mr):
(JSC::X86Assembler::addq_rm):
(JSC::X86Assembler::addq_im):
(JSC::X86Assembler::andl_mr):
(JSC::X86Assembler::andl_rm):
(JSC::X86Assembler::andw_rm):
(JSC::X86Assembler::andb_rm):
(JSC::X86Assembler::andl_im):
(JSC::X86Assembler::andw_im):
(JSC::X86Assembler::andb_im):
(JSC::X86Assembler::andq_mr):
(JSC::X86Assembler::andq_rm):
(JSC::X86Assembler::andq_im):
(JSC::X86Assembler::incq_m):
(JSC::X86Assembler::negq_m):
(JSC::X86Assembler::negl_m):
(JSC::X86Assembler::negw_m):
(JSC::X86Assembler::negb_m):
(JSC::X86Assembler::notl_m):
(JSC::X86Assembler::notw_m):
(JSC::X86Assembler::notb_m):
(JSC::X86Assembler::notq_m):
(JSC::X86Assembler::orl_mr):
(JSC::X86Assembler::orl_rm):
(JSC::X86Assembler::orw_rm):
(JSC::X86Assembler::orb_rm):
(JSC::X86Assembler::orl_im):
(JSC::X86Assembler::orw_im):
(JSC::X86Assembler::orb_im):
(JSC::X86Assembler::orq_mr):
(JSC::X86Assembler::orq_rm):
(JSC::X86Assembler::orq_im):
(JSC::X86Assembler::subl_mr):
(JSC::X86Assembler::subl_rm):
(JSC::X86Assembler::subw_rm):
(JSC::X86Assembler::subb_rm):
(JSC::X86Assembler::subl_im):
(JSC::X86Assembler::subw_im):
(JSC::X86Assembler::subb_im):
(JSC::X86Assembler::subq_mr):
(JSC::X86Assembler::subq_rm):
(JSC::X86Assembler::subq_im):
(JSC::X86Assembler::xorl_mr):
(JSC::X86Assembler::xorl_rm):
(JSC::X86Assembler::xorl_im):
(JSC::X86Assembler::xorw_rm):
(JSC::X86Assembler::xorw_im):
(JSC::X86Assembler::xorb_rm):
(JSC::X86Assembler::xorb_im):
(JSC::X86Assembler::xorq_im):
(JSC::X86Assembler::xorq_rm):
(JSC::X86Assembler::xorq_mr):
(JSC::X86Assembler::xchgb_rm):
(JSC::X86Assembler::xchgw_rm):
(JSC::X86Assembler::xchgl_rm):
(JSC::X86Assembler::xchgq_rm):
(JSC::X86Assembler::movw_im):
(JSC::X86Assembler::movq_i32m):
(JSC::X86Assembler::cmpxchgb_rm):
(JSC::X86Assembler::cmpxchgw_rm):
(JSC::X86Assembler::cmpxchgl_rm):
(JSC::X86Assembler::cmpxchgq_rm):
(JSC::X86Assembler::xaddb_rm):
(JSC::X86Assembler::xaddw_rm):
(JSC::X86Assembler::xaddl_rm):
(JSC::X86Assembler::xaddq_rm):
(JSC::X86Assembler::X86InstructionFormatter::SingleInstructionBufferWriter::memoryModRM):
- b3/B3AtomicValue.cpp: Added.
(JSC::B3::AtomicValue::~AtomicValue):
(JSC::B3::AtomicValue::dumpMeta):
(JSC::B3::AtomicValue::cloneImpl):
(JSC::B3::AtomicValue::AtomicValue):
- b3/B3AtomicValue.h: Added.
- b3/B3BasicBlock.h:
- b3/B3BlockInsertionSet.cpp:
(JSC::B3::BlockInsertionSet::BlockInsertionSet):
(JSC::B3::BlockInsertionSet::insert): Deleted.
(JSC::B3::BlockInsertionSet::insertBefore): Deleted.
(JSC::B3::BlockInsertionSet::insertAfter): Deleted.
(JSC::B3::BlockInsertionSet::execute): Deleted.
- b3/B3BlockInsertionSet.h:
- b3/B3Effects.cpp:
(JSC::B3::Effects::interferes):
(JSC::B3::Effects::operator==):
(JSC::B3::Effects::dump):
- b3/B3Effects.h:
(JSC::B3::Effects::forCall):
(JSC::B3::Effects::mustExecute):
- b3/B3EliminateCommonSubexpressions.cpp:
- b3/B3Generate.cpp:
(JSC::B3::generateToAir):
- b3/B3GenericBlockInsertionSet.h: Added.
(JSC::B3::GenericBlockInsertionSet::GenericBlockInsertionSet):
(JSC::B3::GenericBlockInsertionSet::insert):
(JSC::B3::GenericBlockInsertionSet::insertBefore):
(JSC::B3::GenericBlockInsertionSet::insertAfter):
(JSC::B3::GenericBlockInsertionSet::execute):
- b3/B3HeapRange.h:
(JSC::B3::HeapRange::operator|):
- b3/B3InsertionSet.cpp:
(JSC::B3::InsertionSet::insertClone):
- b3/B3InsertionSet.h:
- b3/B3LegalizeMemoryOffsets.cpp:
- b3/B3LowerMacros.cpp:
(JSC::B3::lowerMacros):
- b3/B3LowerMacrosAfterOptimizations.cpp:
- b3/B3LowerToAir.cpp:
(JSC::B3::Air::LowerToAir::LowerToAir):
(JSC::B3::Air::LowerToAir::run):
(JSC::B3::Air::LowerToAir::effectiveAddr):
(JSC::B3::Air::LowerToAir::addr):
(JSC::B3::Air::LowerToAir::loadPromiseAnyOpcode):
(JSC::B3::Air::LowerToAir::appendShift):
(JSC::B3::Air::LowerToAir::tryAppendStoreBinOp):
(JSC::B3::Air::LowerToAir::storeOpcode):
(JSC::B3::Air::LowerToAir::createStore):
(JSC::B3::Air::LowerToAir::finishAppendingInstructions):
(JSC::B3::Air::LowerToAir::newBlock):
(JSC::B3::Air::LowerToAir::splitBlock):
(JSC::B3::Air::LowerToAir::fillStackmap):
(JSC::B3::Air::LowerToAir::appendX86Div):
(JSC::B3::Air::LowerToAir::appendX86UDiv):
(JSC::B3::Air::LowerToAir::loadLinkOpcode):
(JSC::B3::Air::LowerToAir::storeCondOpcode):
(JSC::B3::Air::LowerToAir::appendCAS):
(JSC::B3::Air::LowerToAir::appendVoidAtomic):
(JSC::B3::Air::LowerToAir::appendGeneralAtomic):
(JSC::B3::Air::LowerToAir::lower):
(JSC::B3::Air::LowerToAir::lowerX86Div): Deleted.
(JSC::B3::Air::LowerToAir::lowerX86UDiv): Deleted.
- b3/B3LowerToAir.h:
- b3/B3MemoryValue.cpp:
(JSC::B3::MemoryValue::isLegalOffset):
(JSC::B3::MemoryValue::accessType):
(JSC::B3::MemoryValue::accessBank):
(JSC::B3::MemoryValue::accessByteSize):
(JSC::B3::MemoryValue::dumpMeta):
(JSC::B3::MemoryValue::MemoryValue):
(JSC::B3::MemoryValue::accessWidth): Deleted.
- b3/B3MemoryValue.h:
- b3/B3MemoryValueInlines.h: Added.
(JSC::B3::MemoryValue::isLegalOffset):
(JSC::B3::MemoryValue::requiresSimpleAddr):
(JSC::B3::MemoryValue::accessWidth):
- b3/B3MoveConstants.cpp:
- b3/B3NativeTraits.h: Added.
- b3/B3Opcode.cpp:
(JSC::B3::storeOpcode):
(WTF::printInternal):
- b3/B3Opcode.h:
(JSC::B3::isLoad):
(JSC::B3::isStore):
(JSC::B3::isLoadStore):
(JSC::B3::isAtomic):
(JSC::B3::isAtomicCAS):
(JSC::B3::isAtomicXchg):
(JSC::B3::isMemoryAccess):
(JSC::B3::signExtendOpcode):
- b3/B3Procedure.cpp:
(JSC::B3::Procedure::dump):
- b3/B3Procedure.h:
(JSC::B3::Procedure::hasQuirks):
(JSC::B3::Procedure::setHasQuirks):
- b3/B3PureCSE.cpp:
(JSC::B3::pureCSE):
- b3/B3PureCSE.h:
- b3/B3ReduceStrength.cpp:
- b3/B3Validate.cpp:
- b3/B3Value.cpp:
(JSC::B3::Value::returnsBool):
(JSC::B3::Value::effects):
(JSC::B3::Value::key):
(JSC::B3::Value::performSubstitution):
(JSC::B3::Value::typeFor):
- b3/B3Value.h:
- b3/B3Width.cpp:
(JSC::B3::bestType):
- b3/B3Width.h:
(JSC::B3::canonicalWidth):
(JSC::B3::isCanonicalWidth):
(JSC::B3::mask):
- b3/air/AirArg.cpp:
(JSC::B3::Air::Arg::jsHash):
(JSC::B3::Air::Arg::dump):
(WTF::printInternal):
- b3/air/AirArg.h:
(JSC::B3::Air::Arg::isAnyUse):
(JSC::B3::Air::Arg::isColdUse):
(JSC::B3::Air::Arg::cooled):
(JSC::B3::Air::Arg::isEarlyUse):
(JSC::B3::Air::Arg::isLateUse):
(JSC::B3::Air::Arg::isAnyDef):
(JSC::B3::Air::Arg::isEarlyDef):
(JSC::B3::Air::Arg::isLateDef):
(JSC::B3::Air::Arg::isZDef):
(JSC::B3::Air::Arg::simpleAddr):
(JSC::B3::Air::Arg::statusCond):
(JSC::B3::Air::Arg::isSimpleAddr):
(JSC::B3::Air::Arg::isMemory):
(JSC::B3::Air::Arg::isStatusCond):
(JSC::B3::Air::Arg::isCondition):
(JSC::B3::Air::Arg::ptr):
(JSC::B3::Air::Arg::base):
(JSC::B3::Air::Arg::isGP):
(JSC::B3::Air::Arg::isFP):
(JSC::B3::Air::Arg::isValidForm):
(JSC::B3::Air::Arg::forEachTmpFast):
(JSC::B3::Air::Arg::forEachTmp):
(JSC::B3::Air::Arg::asAddress):
(JSC::B3::Air::Arg::asStatusCondition):
(JSC::B3::Air::Arg::isInvertible):
(JSC::B3::Air::Arg::inverted):
- b3/air/AirBasicBlock.cpp:
(JSC::B3::Air::BasicBlock::setSuccessors):
- b3/air/AirBasicBlock.h:
- b3/air/AirBlockInsertionSet.cpp: Added.
(JSC::B3::Air::BlockInsertionSet::BlockInsertionSet):
(JSC::B3::Air::BlockInsertionSet::~BlockInsertionSet):
- b3/air/AirBlockInsertionSet.h: Added.
- b3/air/AirDumpAsJS.cpp: Removed.
- b3/air/AirDumpAsJS.h: Removed.
- b3/air/AirEliminateDeadCode.cpp:
(JSC::B3::Air::eliminateDeadCode):
- b3/air/AirGenerate.cpp:
(JSC::B3::Air::prepareForGeneration):
- b3/air/AirInstInlines.h:
(JSC::B3::Air::isAtomicStrongCASValid):
(JSC::B3::Air::isBranchAtomicStrongCASValid):
(JSC::B3::Air::isAtomicStrongCAS8Valid):
(JSC::B3::Air::isAtomicStrongCAS16Valid):
(JSC::B3::Air::isAtomicStrongCAS32Valid):
(JSC::B3::Air::isAtomicStrongCAS64Valid):
(JSC::B3::Air::isBranchAtomicStrongCAS8Valid):
(JSC::B3::Air::isBranchAtomicStrongCAS16Valid):
(JSC::B3::Air::isBranchAtomicStrongCAS32Valid):
(JSC::B3::Air::isBranchAtomicStrongCAS64Valid):
- b3/air/AirOpcode.opcodes:
- b3/air/AirOptimizeBlockOrder.cpp:
(JSC::B3::Air::optimizeBlockOrder):
- b3/air/AirPadInterference.cpp:
(JSC::B3::Air::padInterference):
- b3/air/AirSpillEverything.cpp:
(JSC::B3::Air::spillEverything):
- b3/air/opcode_generator.rb:
- b3/testb3.cpp:
(JSC::B3::testLoadAcq42):
(JSC::B3::testStoreRelAddLoadAcq32):
(JSC::B3::testStoreRelAddLoadAcq8):
(JSC::B3::testStoreRelAddFenceLoadAcq8):
(JSC::B3::testStoreRelAddLoadAcq16):
(JSC::B3::testStoreRelAddLoadAcq64):
(JSC::B3::testTrappingStoreElimination):
(JSC::B3::testX86LeaAddAdd):
(JSC::B3::testX86LeaAddShlLeftScale1):
(JSC::B3::testAtomicWeakCAS):
(JSC::B3::testAtomicStrongCAS):
(JSC::B3::testAtomicXchg):
(JSC::B3::testDepend32):
(JSC::B3::testDepend64):
(JSC::B3::run):
- runtime/Options.h:
Websites/webkit.org:
Document the new opcodes!
- docs/b3/intermediate-representation.html:
- 9:32 AM Changeset in webkit [213713] by
-
- 2 edits in trunk/LayoutTests
[GTK] Unreviewed test gardening
https://bugs.webkit.org/show_bug.cgi?id=169470
Unreviewed GTK+ gardening. Updated test expectations after regressions r213671 and r213560.
- platform/gtk/TestExpectations:
- imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.worker.html [ Failure ]
- imported/w3c/web-platform-tests/WebCryptoAPI/derive_bits_keys/test_pbkdf2.html [ Failure ]
- imported/w3c/web-platform-tests/WebCryptoAPI/import_export/ec_importKey.worker.html [ Failure ]
- imported/w3c/web-platform-tests/WebCryptoAPI/import_export/test_ec_importKey.html [ Failure ]
- 8:01 AM Changeset in webkit [213712] by
-
- 15 edits3 adds in trunk
Source/WebCore:
Allow the page to render before <link> stylesheet tags in body
https://bugs.webkit.org/show_bug.cgi?id=149157
<rdar://problem/24658830>
Reviewed by Simon Fraser.
Currently we block style and renderer building completely if document has any loading
stylesheets. In case a script queries something layout dependent we construct the render
tree with whatever style we have but block painting in it.
This patch changes behavior so that a loading stylesheet in body only blocks rendering for elements
that are after it. The expectation is that such stylesheets rarely affect elements before them
and the elements can be rendered without causing ugly visible styling changes.
The patch replaces the old flash-of-unstyled-content (FOUC) preventation mechanism with a more
fine-grained one. Paint blocking is now done on per-renderer basis with based on isNonFinal flag in
RenderStyle.
For stylesheets in head the behavior should be largely unchanged.
Test: http/tests/incremental/stylesheet-body-incremental-rendering.html
- css/StyleResolver.cpp:
(WebCore::StyleResolver::pseudoStyleRulesForElement):
- dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::resolveStyle):
(WebCore::Document::updateLayoutIgnorePendingStylesheets):
Remove the old FOUC preventation state tracking.
(WebCore::Document::shouldScheduleLayout):
(WebCore::Document::didRemoveAllPendingStylesheet):
Repaints will now get triggered by the normal style mechanism.
- dom/Document.h:
(WebCore::Document::hasNodesWithNonFinalStyle):
(WebCore::Document::setHasNodesWithNonFinalStyle):
Track if we need to recompute the style later because non-final or unstyled elements.
(WebCore::Document::didLayoutWithPendingStylesheets): Deleted.
(WebCore::Document::hasNodesWithPlaceholderStyle): Deleted.
(WebCore::Document::setHasNodesWithPlaceholderStyle): Deleted.
- html/HTMLFrameSetElement.cpp:
(WebCore::HTMLFrameSetElement::rendererIsNeeded):
- page/FrameView.cpp:
(WebCore::FrameView::qualifiesAsVisuallyNonEmpty):
Don't qualify as visually non-empty if we have loading stylesheets in head (even if there is
a fouc-prevented render tree).
(WebCore::FrameView::fireLayoutRelatedMilestonesIfNeeded):
- rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintContents):
Instead of a global test, block painting if isNonFinal is set in the renderer's style.
- rendering/RenderLayer.cpp:
(WebCore::shouldSuppressPaintingLayer):
- rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::changeRequiresRepaint):
The isNonFinal flag prevents painting so we need to trigger repaint when it gets cleared.
- rendering/style/RenderStyle.h:
(WebCore::RenderStyle::isNotFinal):
(WebCore::RenderStyle::setIsNotFinal):
(WebCore::RenderStyle::isPlaceholderStyle): Deleted.
(WebCore::RenderStyle::setIsPlaceholderStyle): Deleted.
There is no need for placeholder styles anymore. Reuse the bit for isNotFinal.
- rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
- rendering/style/StyleRareNonInheritedData.h:
- style/StyleScope.cpp:
(WebCore::Style::Scope::analyzeStyleSheetChange):
(WebCore::Style::Scope::updateActiveStyleSheets):
- style/StyleTreeResolver.cpp:
(WebCore::Style::TreeResolver::styleForElement):
(WebCore::Style::TreeResolver::resolveElement):
If we have seens a loading stylesheet and don't have a renderer yet don't style the element.
In case there is a renderer or we are ignoring pending sheets, resolve the style normally
but mark it as non-final.
(WebCore::Style::makePlaceholderStyle): Deleted.
LayoutTests:
Loading in-body stylesheets should not block rendering of elements before them
https://bugs.webkit.org/show_bug.cgi?id=169345
Reviewed by Simon Fraser.
- http/tests/incremental/resources/delayed-css.php: Added.
- http/tests/incremental/stylesheet-body-incremental-rendering-expected.html: Added.
- http/tests/incremental/stylesheet-body-incremental-rendering.html: Added.
- 7:38 AM Changeset in webkit [213711] by
-
- 4 edits in trunk
imported/w3c/web-platform-tests/html/semantics/embedded-content/the-img-element/sizes/parse-a-sizes-attribute.html is unreliable
https://bugs.webkit.org/show_bug.cgi?id=169465
Reviewed by Zalan Bujtas.
LayoutTests/imported/w3c:
- web-platform-tests/html/semantics/embedded-content/the-img-element/sizes/parse-a-sizes-attribute-expected.txt:
Source/WebCore:
It shows all subtests passing but that is a bug. We don't even support all the unit types.
The test occasionally fails when reloading, that failing results are actually the correct ones.
- css/parser/SizesAttributeParser.cpp:
(WebCore::SizesAttributeParser::SizesAttributeParser):
Ensure we have correct size for the iframe before parsing 'sizes' attribute values. The interpretation of
these may depend on the iframe dimensions.
- 5:47 AM Changeset in webkit [213710] by
-
- 3 edits in trunk/Source/JavaScriptCore
Unreviewed typo fixes after r213652.
https://bugs.webkit.org/show_bug.cgi?id=168920
- assembler/MacroAssemblerARM.h:
(JSC::MacroAssemblerARM::replaceWithBreakpoint):
- assembler/MacroAssemblerMIPS.h:
(JSC::MacroAssemblerMIPS::replaceWithBreakpoint):
- 5:32 AM Changeset in webkit [213709] by
-
- 4 edits in trunk/Source/JavaScriptCore
Unreviewed ARM buildfix after r213652.
https://bugs.webkit.org/show_bug.cgi?id=168920
r213652 used replaceWithBrk and replaceWithBkpt names for the same
function, which was inconsistent and caused build error in ARMAssembler.
- assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::replaceWithBkpt): Renamed replaceWithBrk to replaceWithBkpt.
(JSC::ARM64Assembler::replaceWithBrk): Deleted.
- assembler/ARMAssembler.h:
(JSC::ARMAssembler::replaceWithBkpt): Renamed replaceWithBrk to replaceWithBkpt.
(JSC::ARMAssembler::replaceWithBrk): Deleted.
- assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::replaceWithBreakpoint):
- 5:02 AM Changeset in webkit [213708] by
-
- 2 edits in trunk/Source/WebKit2
[WK2] Guard GLib-specific typedefs in InjectedBundle.h with USE(GLIB)
https://bugs.webkit.org/show_bug.cgi?id=169464
Reviewed by Carlos Garcia Campos.
- WebProcess/InjectedBundle/InjectedBundle.h: Use the more appropriate
USE(GLIB) guard for the two typedefs revolving around GModule, instead
of the PLATFORM(GTK) guard.
- 4:40 AM CoordinatedGraphicsSystem edited by
- Fixed TiledBackingStore.cpp path (diff)
- 3:56 AM Changeset in webkit [213707] by
-
- 2 edits in trunk
Unreviewed Win64 build fix. FTL is not ready to be turned on yet.
- Source/cmake/OptionsWin.cmake:
- 3:31 AM Changeset in webkit [213706] by
-
- 5 edits in trunk
[mac-wk1] LayoutTest media/modern-media-controls/airplay-button/airplay-button.html is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=168409
Unreviewed.
Add some debugging in the hope of understanding why airplay-button.html fails once a day on bots.
Patch by Antoine Quint <Antoine Quint> on 2017-03-10
Source/WebCore:
- Modules/modern-media-controls/controls/scheduler.js:
(const.scheduler.new.prototype.scheduleLayout):
(const.scheduler.new.prototype.unscheduleLayout):
(const.scheduler.new.prototype._requestFrameIfNeeded):
(const.scheduler.new.prototype._frameDidFire):
(const.scheduler.new.prototype._layout):
LayoutTests:
- media/modern-media-controls/airplay-button/airplay-button-expected.txt:
- media/modern-media-controls/airplay-button/airplay-button.html:
- 2:30 AM Changeset in webkit [213705] by
-
- 2 edits in trunk/Source/WTF
Unreviewed AArch64 Linux buildfix after r213645.
https://bugs.webkit.org/show_bug.cgi?id=169300
- wtf/Atomics.h: intptr_t and int64_t are typedefs for long int on Linux.
They are different types only on Darwin platforms.
- 12:36 AM Changeset in webkit [213704] by
-
- 6 edits in trunk
Win64 build fix.
.:
- Source/cmake/OptionsWin.cmake:
WASM got turned on on Win64. Judging by the state of WasmMemory.cpp, I'm going to
assume this was a mistake and turn it off until someone writes a version that uses
VirtualAlloc/VirtualFree instead of mmap.
The FTL seems a bit unstable, though. This fixes the build, but more fixes are necessary.
Source/JavaScriptCore:
- b3/B3FenceValue.h:
- b3/B3Value.h:
Putting JS_EXPORT_PRIVATE on member functions in classes that are declared with JS_EXPORT_PRIVATE
doesn't accomplish anything except making Visual Studio mad.
- b3/air/opcode_generator.rb:
winnt.h has naming collisions with enum values from AirOpcode.h.
For example, MemoryFence is #defined to be _mm_mfence, which is declared to be a function in emmintrin.h.
RotateLeft32 is #defined to be _rotl, which is declared to be a function in <stdlib.h>
A clean solution is just to put Opcode:: before the references to the opcode names to tell Visual Studio
that it is referring to the enum value in AirOpcode.h and not the function declaration elsewhere.
Mar 9, 2017:
- 11:19 PM Changeset in webkit [213703] by
-
- 26 edits in trunk/Source/WebKit2
Unreviewed, fix documentation typos
- UIProcess/API/gtk/WebKitAuthenticationRequest.cpp:
- UIProcess/API/gtk/WebKitContextMenuActions.h:
- UIProcess/API/gtk/WebKitContextMenuItem.cpp:
- UIProcess/API/gtk/WebKitDownload.cpp:
(webkitDownloadNotifyProgress):
- UIProcess/API/gtk/WebKitEditorState.cpp:
- UIProcess/API/gtk/WebKitEditorState.h:
- UIProcess/API/gtk/WebKitError.h:
- UIProcess/API/gtk/WebKitFileChooserRequest.cpp:
- UIProcess/API/gtk/WebKitFindController.cpp:
- UIProcess/API/gtk/WebKitGeolocationPermissionRequest.cpp:
- UIProcess/API/gtk/WebKitHitTestResult.cpp:
- UIProcess/API/gtk/WebKitPolicyDecision.cpp:
- UIProcess/API/gtk/WebKitPrintOperation.h:
- UIProcess/API/gtk/WebKitResponsePolicyDecision.cpp:
- UIProcess/API/gtk/WebKitSecurityManager.cpp:
- UIProcess/API/gtk/WebKitSettings.cpp:
(webkit_settings_class_init):
- UIProcess/API/gtk/WebKitUserMediaPermissionRequest.cpp:
(webkitUserMediaPermissionRequestAllow):
- UIProcess/API/gtk/WebKitWebContext.cpp:
- UIProcess/API/gtk/WebKitWebInspector.cpp:
(webkit_web_inspector_class_init):
- UIProcess/API/gtk/WebKitWebView.cpp:
(webkit_web_view_class_init):
- UIProcess/API/gtk/WebKitWebView.h:
- UIProcess/API/gtk/WebKitWebsiteDataManager.cpp:
- WebProcess/InjectedBundle/API/gtk/WebKitFrame.cpp:
- WebProcess/InjectedBundle/API/gtk/WebKitWebExtension.cpp:
- WebProcess/InjectedBundle/API/gtk/WebKitWebPage.cpp:
(webkit_web_page_class_init):
- 9:18 PM Changeset in webkit [213702] by
-
- 3 edits2 deletes in trunk/Source/JavaScriptCore
Unreviewed, rolling out r213695.
This change broke the Windows build.
Reverted changeset:
"Implement a StackTrace utility object that can capture stack
traces for debugging."
https://bugs.webkit.org/show_bug.cgi?id=169454
http://trac.webkit.org/changeset/213695
- 9:08 PM Changeset in webkit [213701] by
-
- 15 edits3 deletes in trunk
Unreviewed, rolling out r213633.
This change caused LayoutTest imported/w3c/web-platform-
tests/html/semantics/embedded-content/the-img-
element/sizes/parse-a-sizes-attribute.html to become a flaky
failure.
Reverted changeset:
"Loading in-body stylesheets should not block rendering of
elements before them"
https://bugs.webkit.org/show_bug.cgi?id=169345
http://trac.webkit.org/changeset/213633
- 8:55 PM Changeset in webkit [213700] by
-
- 3 edits in trunk/LayoutTests
Rebaseline js/dom/global-constructors-attributes.html.
Unreviewed test gardening.
- platform/mac-wk1/js/dom/global-constructors-attributes-expected.txt:
- platform/mac/js/dom/global-constructors-attributes-expected.txt:
- 7:43 PM Changeset in webkit [213699] by
-
- 4 edits in trunk/Source
Fix CMake build
- PlatformMac.cmake:
- 6:37 PM Changeset in webkit [213698] by
-
- 5 edits in trunk/Source
Fix GTK build after r213696
Source/WebCore:
- PlatformMac.cmake:
Source/WebKit2:
- CMakeLists.txt:
- PlatformMac.cmake:
- 6:00 PM Changeset in webkit [213697] by
-
- 29 edits1 add in trunk
[ESnext] Implement Object Rest - Implementing Object Rest Destructuring
https://bugs.webkit.org/show_bug.cgi?id=167962
Patch by Caio Lima <Caio Lima> on 2017-03-09
Reviewed by Keith Miller.
JSTests:
- stress/object-rest-deconstruct.js: Added.
(let.assert):
(let.assertPropDescriptor):
(catch):
(get 3):
(foo):
(let.src.get y):
(let.src.set y):
(let.gen):
Source/JavaScriptCore:
Object Rest/Spread Destructing proposal is in stage 3[1] and this
Patch is a prototype implementation of it. A simple change over the
parser was necessary to support the new '...' token on Object Pattern
destruction rule. In the bytecode generator side, We changed the
bytecode generated on ObjectPatternNode::bindValue to store in an
array identifiers of already destructed properties, following spec draft
section[2], and then pass it as excludedNames to CopyDataProperties.
The rest destruction the calls copyDataProperties to perform the
copy of rest properties in rhs.
We also implemented CopyDataProperties as private JS global operation
on builtins/GlobalOperations.js following it's specification on [3].
It is implemented using Set object to verify if a property is on
excludedNames to keep this algorithm with O(n + m) complexity, where n
number of source's own properties and m = excludedNames.length.
As a requirement to use JSSets as constants, a change in
CodeBlock::create API was necessary, because JSSet creation can throws OOM
exception. Now, CodeBlock::finishCreation returnsfalseif an
execption is throwed by
CodeBlock::setConstantIdentifierSetRegisters and then we return
nullptr to ScriptExecutable::newCodeBlockFor. It is responsible to
check if CodeBlock was constructed properly and then, throw OOM
exception to the correct scope.
[1] - https://github.com/sebmarkbage/ecmascript-rest-spread
[2] - http://sebmarkbage.github.io/ecmascript-rest-spread/#Rest-RuntimeSemantics-PropertyDestructuringAssignmentEvaluation
[3] - http://sebmarkbage.github.io/ecmascript-rest-spread/#AbstractOperations-CopyDataProperties
- builtins/BuiltinNames.h:
- builtins/GlobalOperations.js:
(globalPrivate.copyDataProperties):
- bytecode/CodeBlock.cpp:
(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::setConstantIdentifierSetRegisters):
- bytecode/CodeBlock.h:
- bytecode/EvalCodeBlock.h:
(JSC::EvalCodeBlock::create):
- bytecode/FunctionCodeBlock.h:
(JSC::FunctionCodeBlock::create):
- bytecode/ModuleProgramCodeBlock.h:
(JSC::ModuleProgramCodeBlock::create):
- bytecode/ProgramCodeBlock.h:
(JSC::ProgramCodeBlock::create):
- bytecode/UnlinkedCodeBlock.h:
(JSC::UnlinkedCodeBlock::addSetConstant):
(JSC::UnlinkedCodeBlock::constantIdentifierSets):
- bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitLoad):
- bytecompiler/BytecodeGenerator.h:
- bytecompiler/NodesCodegen.cpp:
(JSC::ObjectPatternNode::bindValue):
- parser/ASTBuilder.h:
(JSC::ASTBuilder::appendObjectPatternEntry):
(JSC::ASTBuilder::appendObjectPatternRestEntry):
(JSC::ASTBuilder::setContainsObjectRestElement):
- parser/Nodes.h:
(JSC::ObjectPatternNode::appendEntry):
(JSC::ObjectPatternNode::setContainsRestElement):
- parser/Parser.cpp:
(JSC::Parser<LexerType>::parseDestructuringPattern):
(JSC::Parser<LexerType>::parseProperty):
- parser/SyntaxChecker.h:
(JSC::SyntaxChecker::operatorStackPop):
- runtime/JSGlobalObject.cpp:
(JSC::JSGlobalObject::init):
- runtime/JSGlobalObjectFunctions.cpp:
(JSC::privateToObject):
- runtime/JSGlobalObjectFunctions.h:
- runtime/ScriptExecutable.cpp:
(JSC::ScriptExecutable::newCodeBlockFor):
Source/WTF:
- wtf/HashSet.h:
(WTF::=):
LayoutTests:
- js/parser-syntax-check-expected.txt:
- js/script-tests/parser-syntax-check.js:
- 5:48 PM Changeset in webkit [213696] by
-
- 32 edits8 copies4 moves1 add in trunk
Expose public APIs for content filters
https://bugs.webkit.org/show_bug.cgi?id=150479
Reviewed by Brady Eidson and Geoff Garen.
Source/WebCore:
- English.lproj/Localizable.strings:
Source/WebKit2:
This takes _WKUserContentExtensionStore and _WKUserContentFilter and turns them into the consistently-named public API
WKContentExtensionStore and WKContentExtension respectively with the same design and functionaly of the original SPI.
We also added public functions to WKUserContentController corresponding to existing private functions.
The old SPI ObjC classes are now just wrappers around the new ObjC classes, which are wrappers around C++ objects in the API namespace.
The peak-memory-reducing optimization of having NS_RELEASES_ARGUMENT in _compileContentExtensionForIdentifier is kept for Safari,
but the public API doesn't need such an optimization. The public compileContentExtensionForIdentifier will not deallocate the input
NSString during compiling.
We moved the error codes into the WKErrorDomain, which requires changing the values of the error codes and domain, but source
compiled that uses the SPI before and after should still work. Using new applications with old frameworks or vice versa will
have unexpected error codes, though.
- Shared/API/APIObject.h:
- Shared/API/c/WKDeprecatedFunctions.cpp:
(WKPageGroupAddUserContentFilter):
(WKPageGroupRemoveUserContentFilter):
- Shared/Cocoa/APIObject.mm:
(API::Object::newObject):
- UIProcess/API/APIContentExtension.cpp: Copied from UIProcess/API/APIUserContentExtension.cpp.
(API::ContentExtension::ContentExtension):
(API::ContentExtension::~ContentExtension):
(API::UserContentExtension::UserContentExtension): Deleted.
(API::UserContentExtension::~UserContentExtension): Deleted.
- UIProcess/API/APIContentExtension.h: Copied from UIProcess/API/APIUserContentExtension.h.
- UIProcess/API/APIContentExtensionStore.cpp: Copied from UIProcess/API/APIUserContentExtensionStore.cpp.
(API::ContentExtensionStore::defaultStore):
(API::ContentExtensionStore::storeWithPath):
(API::ContentExtensionStore::ContentExtensionStore):
(API::ContentExtensionStore::~ContentExtensionStore):
(API::compiledToFile):
(API::createExtension):
(API::ContentExtensionStore::lookupContentExtension):
(API::ContentExtensionStore::compileContentExtension):
(API::ContentExtensionStore::removeContentExtension):
(API::ContentExtensionStore::synchronousRemoveAllContentExtensions):
(API::ContentExtensionStore::invalidateContentExtensionVersion):
(API::contentExtensionStoreErrorCategory):
(API::UserContentExtensionStore::defaultStore): Deleted.
(API::UserContentExtensionStore::storeWithPath): Deleted.
(API::UserContentExtensionStore::UserContentExtensionStore): Deleted.
(API::UserContentExtensionStore::~UserContentExtensionStore): Deleted.
(API::UserContentExtensionStore::lookupContentExtension): Deleted.
(API::UserContentExtensionStore::compileContentExtension): Deleted.
(API::UserContentExtensionStore::removeContentExtension): Deleted.
(API::UserContentExtensionStore::synchronousRemoveAllContentExtensions): Deleted.
(API::UserContentExtensionStore::invalidateContentExtensionVersion): Deleted.
(API::userContentExtensionStoreErrorCategory): Deleted.
- UIProcess/API/APIContentExtensionStore.h: Copied from UIProcess/API/APIUserContentExtensionStore.h.
(API::make_error_code):
- UIProcess/API/APIUserContentExtension.cpp: Removed.
- UIProcess/API/APIUserContentExtension.h: Removed.
- UIProcess/API/APIUserContentExtensionStore.cpp: Removed.
- UIProcess/API/APIUserContentExtensionStore.h: Removed.
- UIProcess/API/C/WKAPICast.h:
- UIProcess/API/C/WKPageGroup.cpp:
- UIProcess/API/C/WKUserContentControllerRef.cpp:
(WKUserContentControllerAddUserContentFilter):
(WKUserContentControllerRemoveAllUserContentFilters):
- UIProcess/API/C/WKUserContentExtensionStoreRef.cpp:
(WKUserContentExtensionStoreGetTypeID):
- UIProcess/API/Cocoa/APIUserContentExtensionStoreCocoa.mm:
(API::ContentExtensionStore::defaultStorePath):
(API::UserContentExtensionStore::defaultStorePath): Deleted.
- UIProcess/API/Cocoa/WKContentExtension.h: Copied from UIProcess/API/Cocoa/_WKUserContentFilter.h.
- UIProcess/API/Cocoa/WKContentExtension.mm: Copied from UIProcess/API/Cocoa/_WKUserContentFilter.mm.
(-[WKContentExtension dealloc]):
(-[WKContentExtension _apiObject]):
(-[_WKUserContentFilter dealloc]): Deleted.
(-[_WKUserContentFilter _apiObject]): Deleted.
- UIProcess/API/Cocoa/WKContentExtensionInternal.h: Copied from UIProcess/API/Cocoa/_WKUserContentFilterInternal.h.
(WebKit::wrapper):
- UIProcess/API/Cocoa/WKContentExtensionStore.h: Copied from UIProcess/API/Cocoa/_WKUserContentExtensionStore.h.
- UIProcess/API/Cocoa/WKContentExtensionStore.mm: Copied from UIProcess/API/Cocoa/_WKUserContentExtensionStore.mm.
(-[WKContentExtensionStore dealloc]):
(+[WKContentExtensionStore defaultStore]):
(+[WKContentExtensionStore storeWithURL:]):
(-[WKContentExtensionStore compileContentExtensionForIdentifier:encodedContentExtension:completionHandler:]):
(-[WKContentExtensionStore _compileContentExtensionForIdentifier:encodedContentExtension:completionHandler:releasesArgument:]):
(-[WKContentExtensionStore lookupContentExtensionForIdentifier:completionHandler:]):
(-[WKContentExtensionStore removeContentExtensionForIdentifier:completionHandler:]):
(-[WKContentExtensionStore _apiObject]):
(-[WKContentExtensionStore _removeAllContentExtensions]):
(-[WKContentExtensionStore _invalidateContentExtensionVersionForIdentifier:]):
(-[WKContentExtensionStore _compileContentExtensionForIdentifier:encodedContentExtension:completionHandler:]):
(-[_WKUserContentExtensionStore dealloc]): Deleted.
(+[_WKUserContentExtensionStore defaultStore]): Deleted.
(+[_WKUserContentExtensionStore storeWithURL:]): Deleted.
(-[_WKUserContentExtensionStore compileContentExtensionForIdentifier:encodedContentExtension:completionHandler:]): Deleted.
(-[_WKUserContentExtensionStore lookupContentExtensionForIdentifier:completionHandler:]): Deleted.
(-[_WKUserContentExtensionStore removeContentExtensionForIdentifier:completionHandler:]): Deleted.
(-[_WKUserContentExtensionStore _apiObject]): Deleted.
(-[_WKUserContentExtensionStore _removeAllContentExtensions]): Deleted.
(-[_WKUserContentExtensionStore _invalidateContentExtensionVersionForIdentifier:]): Deleted.
- UIProcess/API/Cocoa/WKContentExtensionStoreInternal.h: Copied from UIProcess/API/Cocoa/_WKUserContentExtensionStoreInternal.h.
(WebKit::wrapper):
- UIProcess/API/Cocoa/WKContentExtensionStorePrivate.h: Copied from UIProcess/API/Cocoa/_WKUserContentExtensionStorePrivate.h.
- UIProcess/API/Cocoa/WKError.h:
- UIProcess/API/Cocoa/WKError.mm:
(localizedDescriptionForErrorCode):
- UIProcess/API/Cocoa/WKUserContentController.h:
- UIProcess/API/Cocoa/WKUserContentController.mm:
(-[WKUserContentController addContentExtension:]):
(-[WKUserContentController removeContentExtension:]):
(-[WKUserContentController removeAllContentExtensions]):
(-[WKUserContentController _addUserContentFilter:]):
(-[WKUserContentController _removeUserContentFilter:]):
(-[WKUserContentController _removeAllUserContentFilters]):
- UIProcess/API/Cocoa/_WKUserContentExtensionStore.h:
- UIProcess/API/Cocoa/_WKUserContentExtensionStore.mm:
(+[_WKUserContentExtensionStore defaultStore]):
(+[_WKUserContentExtensionStore storeWithURL:]):
(-[_WKUserContentExtensionStore compileContentExtensionForIdentifier:encodedContentExtension:completionHandler:]):
(-[_WKUserContentExtensionStore lookupContentExtensionForIdentifier:completionHandler:]):
(-[_WKUserContentExtensionStore removeContentExtensionForIdentifier:completionHandler:]):
(-[_WKUserContentExtensionStore _apiObject]):
(-[_WKUserContentExtensionStore _removeAllContentExtensions]):
(-[_WKUserContentExtensionStore _invalidateContentExtensionVersionForIdentifier:]):
(-[_WKUserContentExtensionStore _initWithWKContentExtensionStore:]):
(-[_WKUserContentExtensionStore dealloc]): Deleted.
- UIProcess/API/Cocoa/_WKUserContentExtensionStoreInternal.h:
(WebKit::wrapper): Deleted.
- UIProcess/API/Cocoa/_WKUserContentExtensionStorePrivate.h:
- UIProcess/API/Cocoa/_WKUserContentFilter.mm:
(-[_WKUserContentFilter _apiObject]):
(-[_WKUserContentFilter _initWithWKContentExtension:]):
(-[_WKUserContentFilter dealloc]): Deleted.
- UIProcess/API/Cocoa/_WKUserContentFilterInternal.h:
(WebKit::wrapper): Deleted.
- UIProcess/API/Cocoa/_WKUserContentFilterPrivate.h: Added.
- UIProcess/UserContent/WebUserContentControllerProxy.cpp:
(WebKit::WebUserContentControllerProxy::addProcess):
(WebKit::WebUserContentControllerProxy::addContentExtension):
(WebKit::WebUserContentControllerProxy::removeContentExtension):
(WebKit::WebUserContentControllerProxy::removeAllContentExtensions):
(WebKit::WebUserContentControllerProxy::addUserContentExtension): Deleted.
(WebKit::WebUserContentControllerProxy::removeUserContentExtension): Deleted.
(WebKit::WebUserContentControllerProxy::removeAllUserContentExtensions): Deleted.
- UIProcess/UserContent/WebUserContentControllerProxy.h:
- UIProcess/WebPageGroup.cpp:
- WebKit2.xcodeproj/project.pbxproj:
- WebProcess/UserContent/WebUserContentController.cpp:
(WebKit::WebUserContentController::addContentExtensions):
(WebKit::WebUserContentController::removeContentExtension):
(WebKit::WebUserContentController::removeAllContentExtensions):
(WebKit::WebUserContentController::addUserContentExtensions): Deleted.
(WebKit::WebUserContentController::removeUserContentExtension): Deleted.
(WebKit::WebUserContentController::removeAllUserContentExtensions): Deleted.
- WebProcess/UserContent/WebUserContentController.h:
- WebProcess/UserContent/WebUserContentController.messages.in:
- WebProcess/WebCoreSupport/mac/WebErrorsMac.mm:
(WebKit::blockedByContentFilterError):
Tools:
Keep the existing tests to verify that I made the wrapper classes for the old SPI work correctly.
Add new identical tests for the new API are added with the new names.
We should keep both until we migrate from and remove the old SPI.
Enum error code values and strings have changed, but the API is source compatible.
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
- TestWebKitAPI/Tests/WebKit2Cocoa/WKUserContentExtensionStore.mm: Copied from Tools/TestWebKitAPI/Tests/WebKit2Cocoa/_WKUserContentExtensionStore.mm.
(WKUserContentExtensionStoreTest::SetUp):
(TEST_F):
(checkDomain):
(_WKUserContentExtensionStoreTest::SetUp): Deleted.
- TestWebKitAPI/Tests/WebKit2Cocoa/_WKUserContentExtensionStore.mm:
(checkDomain):
- 5:27 PM Changeset in webkit [213695] by
-
- 3 edits2 adds in trunk/Source/JavaScriptCore
Implement a StackTrace utility object that can capture stack traces for debugging.
https://bugs.webkit.org/show_bug.cgi?id=169454
Reviewed by Michael Saboff.
The underlying implementation is hoisted right out of Assertions.cpp from the
implementations of WTFPrintBacktrace().
The reason we need this StackTrace object is because during heap debugging, we
sometimes want to capture the stack trace that allocated the objects of interest.
Dumping the stack trace directly to stdout (using WTFReportBacktrace()) may
perturb the execution profile sufficiently that an issue may not reproduce,
while alternatively, just capturing the stack trace and deferring printing it
till we actually need it later perturbs the execution profile less.
In addition, just capturing the stack traces (instead of printing them
immediately at each capture site) allows us to avoid polluting stdout with tons
of stack traces that may be irrelevant.
For now, we only capture the native stack trace. We'll leave capturing and
integrating the JS stack trace as an exercise for the future if we need it then.
Here's an example of how to use this StackTrace utility:
Capture a stack trace of the top 10 frames.
std::unique_ptr<StackTrace> trace(StackTrace::captureStackTrace(10));
Print the trace.
dataLog(*trace);
- CMakeLists.txt:
- JavaScriptCore.xcodeproj/project.pbxproj:
- tools/StackTrace.cpp: Added.
(JSC::StackTrace::instanceSize):
(JSC::StackTrace::captureStackTrace):
(JSC::StackTrace::dump):
- tools/StackTrace.h: Added.
(JSC::StackTrace::StackTrace):
(JSC::StackTrace::size):
- 5:08 PM Changeset in webkit [213694] by
-
- 3 edits in trunk/Tools
Attempt to fix the WebKit2 test bots following <https://trac.webkit.org/changeset/213676>
(https://bugs.webkit.org/show_bug.cgi?id=169364)
Create new WebView with Credential Management enabled when the enableCredentialManagement
WebKitTestRunner option is specified.
- WebKitTestRunner/ios/PlatformWebViewIOS.mm:
(WTR::PlatformWebView::viewSupportsOptions):
- WebKitTestRunner/mac/PlatformWebViewMac.mm:
(WTR::PlatformWebView::viewSupportsOptions):
- 4:59 PM Changeset in webkit [213693] by
-
- 2 edits in trunk/Source/WebKit2
Fix some debug builds after r213686
Unreviewed.
- UIProcess/Cocoa/WebURLSchemeHandlerCocoa.mm:
(WebKit::WebURLSchemeHandlerCocoa::WebURLSchemeHandlerCocoa): Remove the 32-bit NOT_REACHED protection for now.
- 4:57 PM Changeset in webkit [213692] by
-
- 4 edits in trunk/Source
WebAssembly: Enable fast memory for WK2
https://bugs.webkit.org/show_bug.cgi?id=169437
Reviewed by Tim Horton.
Source/JavaScriptCore:
- JavaScriptCore.xcodeproj/project.pbxproj:
Source/WebKit2:
I forgot to do this when I enabled implemented fast memory
for wasm.
- WebProcess/WebProcess.cpp:
(WebKit::WebProcess::initializeWebProcess):
- 4:53 PM Changeset in webkit [213691] by
-
- 15 edits1 copy1 move5 adds in trunk
Web Inspector: Add XHR breakpoints UI
https://bugs.webkit.org/show_bug.cgi?id=168763
<rdar://problem/30952439>
Reviewed by Joseph Pecoraro.
Source/JavaScriptCore:
- inspector/protocol/DOMDebugger.json:
Added clarifying comments to command descriptions.
Source/WebCore:
Test: inspector/dom-debugger/xhr-breakpoints.html
- inspector/InspectorDOMDebuggerAgent.cpp:
(WebCore::InspectorDOMDebuggerAgent::discardBindings):
(WebCore::InspectorDOMDebuggerAgent::willSendXMLHttpRequest):
Don't break for XHRs when breakpoints are disabled.
Source/WebInspectorUI:
- Localizations/en.lproj/localizedStrings.js:
New UI strings.
- UserInterface/Controllers/DOMDebuggerManager.js:
Plumbing for XHR breakpoints. Like DOM breakpoints, they are tracked
per-page. Unlike DOM breakpoints, they have no separate "resolved"
state, and are simply added/removed as the main frame changes.
(WebInspector.DOMDebuggerManager):
(WebInspector.DOMDebuggerManager.prototype.get allRequestsBreakpoint):
(WebInspector.DOMDebuggerManager.prototype.get xhrBreakpoints):
(WebInspector.DOMDebuggerManager.prototype.isBreakpointRemovable):
(WebInspector.DOMDebuggerManager.prototype.removeDOMBreakpoint):
(WebInspector.DOMDebuggerManager.prototype.addXHRBreakpoint):
(WebInspector.DOMDebuggerManager.prototype.removeXHRBreakpoint):
(WebInspector.DOMDebuggerManager.prototype._speculativelyResolveBreakpoints):
(WebInspector.DOMDebuggerManager.prototype._detachXHRBreakpoint):
(WebInspector.DOMDebuggerManager.prototype._updateXHRBreakpoint.breakpointUpdated):
(WebInspector.DOMDebuggerManager.prototype._updateXHRBreakpoint):
(WebInspector.DOMDebuggerManager.prototype._resolveXHRBreakpoint):
(WebInspector.DOMDebuggerManager.prototype._saveXHRBreakpoints):
(WebInspector.DOMDebuggerManager.prototype._domBreakpointDisabledStateDidChange):
(WebInspector.DOMDebuggerManager.prototype._xhrBreakpointDisabledStateDidChange):
(WebInspector.DOMDebuggerManager.prototype._mainResourceDidChange):
(WebInspector.DOMDebuggerManager.prototype._saveBreakpoints): Deleted.
RenamedsaveDOMBreakpoints.
- UserInterface/Controllers/DebuggerManager.js:
(WebInspector.DebuggerManager.prototype._pauseReasonFromPayload):
Plumbing for XHR breakpoint pause reason.
- UserInterface/Controllers/XHRBreakpointTreeController.js: Added.
Creates the "All Requests" breakpoint, handles DOMDebuggerManager messages,
and updates an associated tree outline as XHR breakpoints are added and removed.
(WebInspector.XHRBreakpointTreeController):
(WebInspector.XHRBreakpointTreeController.prototype.revealAndSelect):
(WebInspector.XHRBreakpointTreeController.prototype.disconnect):
(WebInspector.XHRBreakpointTreeController.prototype._xhrBreakpointAdded):
(WebInspector.XHRBreakpointTreeController.prototype._xhrBreakpointRemoved):
(WebInspector.XHRBreakpointTreeController.prototype._mainResourceDidChange):
(WebInspector.XHRBreakpointTreeController.prototype._reset):
- UserInterface/Main.html:
New files.
- UserInterface/Models/XHRBreakpoint.js: Added.
New model class. XHR breakpoints have an associated URL, which is matched
against the request URL when deciding whether to break. An empty URL matches
all XMLHttpRequests.
(WebInspector.XHRBreakpoint):
(WebInspector.XHRBreakpoint.prototype.get documentURL):
(WebInspector.XHRBreakpoint.prototype.get url):
(WebInspector.XHRBreakpoint.prototype.get disabled):
(WebInspector.XHRBreakpoint.prototype.set disabled):
(WebInspector.XHRBreakpoint.prototype.get serializableInfo):
(WebInspector.XHRBreakpoint.prototype.saveIdentityToCookie):
- UserInterface/Test.html:
New files.
- UserInterface/Views/ContentView.js:
(WebInspector.ContentView.createFromRepresentedObject):
Plumbing for new represented object type, to prevent runtime assertion.
- UserInterface/Views/DOMBreakpointTreeElement.js:
(WebInspector.DOMBreakpointTreeElement):
Reuse BreakpointTreeElement styles.
- UserInterface/Views/DebuggerSidebarPanel.js:
(WebInspector.DebuggerSidebarPanel):
(WebInspector.DebuggerSidebarPanel.prototype.saveStateToCookie):
(WebInspector.DebuggerSidebarPanel.prototype.restoreStateFromCookie):
(WebInspector.DebuggerSidebarPanel.prototype._updatePauseReasonSection):
(WebInspector.DebuggerSidebarPanel.prototype._addXHRBreakpointButtonClicked):
(WebInspector.DebuggerSidebarPanel.prototype.willDismissPopover):
- UserInterface/Views/InputPopover.css: Added.
(.popover .input-popover-content):
(.popover .input-popover-content > .editor):
(.popover .input-popover-content > .editor > .CodeMirror):
(.popover .input-popover-content > .editor > .CodeMirror-scroll):
- UserInterface/Views/InputPopover.js: Added.
New general purpose popover class. Style and structure lifted from
the Popover implementation used by the Probe Expressions UI. A follow-up
has been filed to switch that UI over to this new class:
https://bugs.webkit.org/show_bug.cgi?id=169428.
(WebInspector.InputPopover):
(WebInspector.InputPopover.prototype.get value):
(WebInspector.InputPopover.prototype.get result):
(WebInspector.InputPopover.prototype.show):
(WebInspector.InputPopover.prototype._presentOverTargetElement):
- UserInterface/Views/XHRBreakpointTreeElement.js: Added.
Tree element class for XHR breakpoints. Behaves identically to
script and DOM breakpoint tree elements.
(WebInspector.XHRBreakpointTreeElement):
Reuse BreakpointTreeElement styles.
(WebInspector.XHRBreakpointTreeElement.prototype.onattach):
(WebInspector.XHRBreakpointTreeElement.prototype.ondetach):
(WebInspector.XHRBreakpointTreeElement.prototype.ondelete):
(WebInspector.XHRBreakpointTreeElement.prototype.onenter):
(WebInspector.XHRBreakpointTreeElement.prototype.onspace):
(WebInspector.XHRBreakpointTreeElement.prototype.populateContextMenu):
(WebInspector.XHRBreakpointTreeElement.prototype._statusImageElementClicked):
(WebInspector.XHRBreakpointTreeElement.prototype._statusImageElementFocused):
(WebInspector.XHRBreakpointTreeElement.prototype._statusImageElementMouseDown):
(WebInspector.XHRBreakpointTreeElement.prototype._toggleBreakpoint):
(WebInspector.XHRBreakpointTreeElement.prototype._updateStatus):
LayoutTests:
Tests for pausing on XMLHttpRequests using XHR breakpoints.
- inspector/dom-debugger/xhr-breakpoints-expected.txt: Added.
- inspector/dom-debugger/xhr-breakpoints.html: Added.
- 4:53 PM Changeset in webkit [213690] by
-
- 20 edits1 add in trunk/Source
Source/JavaScriptCore:
Add plumbing to WebProcess to enable JavaScriptCore configuration and logging
https://bugs.webkit.org/show_bug.cgi?id=169387
Reviewed by Filip Pizlo.
Added a helper function, processConfigFile(), to process configuration file.
Changed jsc.cpp to use that function in lieu of processing the config file
manually.
- JavaScriptCore.xcodeproj/project.pbxproj: Made ConfigFile.h a private header file.
- jsc.cpp:
(jscmain):
- runtime/ConfigFile.cpp:
(JSC::processConfigFile):
- runtime/ConfigFile.h:
Source/WebCore:
Add plumbing to WebProcess to enable JavaScriptCore configuration and logging
https://bugs.webkit.org/show_bug.cgi?id=169387
Reviewed by Filip Pizlo.
Added a forwarding header file for JavScriptCore/runtime/ConfigFile.h.
- ForwardingHeaders/runtime/ConfigFile.h: Added.
Source/WebKit2:
<<<<<<< .mine
- 4:52 PM Changeset in webkit [213689] by
-
- 2 edits in trunk/Source/WebInspectorUI
REGRESSION (r196271): Web Inspector: DOM element loses focus/active selection after committing an inline attribute editor in DOMTreeOutline
https://bugs.webkit.org/show_bug.cgi?id=168119
<rdar://problem/30466013>
Reviewed by Brian Burg.
Fix a regression caused by http://trac.webkit.org/changeset/196271, which
updated WebInspector.isEventTargetAnEditableField to detect the editing
scenarios used by DOMTreeElement and DataGrid. This subtly changed the
behavior of WebInspector._focusChanged, which expected all editable fields
to be backed by a CodeMirror instance.
- UserInterface/Base/Main.js:
- 4:46 PM Changeset in webkit [213688] by
-
- 22 edits2 adds in trunk
Resource Load Statistics: Remove statistics data as part of full website data removal
https://bugs.webkit.org/show_bug.cgi?id=169448
Reviewed by Alex Christensen.
Source/WebCore:
Test: http/tests/loading/resourceLoadStatistics/clear-in-memory-and-persistent-store.html
- loader/ResourceLoadObserver.cpp:
(WebCore::ResourceLoadObserver::clearInMemoryAndPersistentStore):
The version of this function that takes a modifiedSince parameter
checks if it is equal to or less than the start of the Unix epoch.
If so, it clears the store. This is to support clear all website
data.
- loader/ResourceLoadObserver.h:
- loader/ResourceLoadStatisticsStore.cpp:
(WebCore::ResourceLoadStatisticsStore::clearInMemoryAndPersistent):
(WebCore::ResourceLoadStatisticsStore::setWritePersistentStoreCallback):
- loader/ResourceLoadStatisticsStore.h:
Added m_writePersistentStoreHandler, a setter for it, and
ResourceLoadStatisticsStore::clearInMemoryAndPersistent().
Source/WebKit2:
- UIProcess/API/C/WKResourceLoadStatisticsManager.cpp:
(WKResourceLoadStatisticsManagerClearInMemoryAndPersistentStore):
Infrastructure for test.
- UIProcess/API/C/WKResourceLoadStatisticsManager.h:
- UIProcess/WebResourceLoadStatisticsManager.cpp:
(WebKit::WebResourceLoadStatisticsManager::clearInMemoryAndPersistentStore):
Infrastructure for test.
- UIProcess/WebResourceLoadStatisticsManager.h:
- UIProcess/WebResourceLoadStatisticsStore.cpp:
(WebKit::WebResourceLoadStatisticsStore::processStatisticsAndDataRecords):
Change due to the break-out of
WebResourceLoadStatisticsStore::writeStoreToDisk().
(WebKit::WebResourceLoadStatisticsStore::registerSharedResourceLoadObserver):
Now sets m_writePersistentStoreHandler in
WebCore::ResourceLoadStatisticsStore.
(WebKit::WebResourceLoadStatisticsStore::writeStoreToDisk):
Separated out this function to make it available for the function set as
m_writePersistentStoreHandler in WebCore::ResourceLoadStatisticsStore.
- UIProcess/WebResourceLoadStatisticsStore.h:
- UIProcess/WebsiteData/WebsiteDataStore.cpp:
(WebKit::WebsiteDataStore::removeData):
Both WebsiteDataStore::removeData() functions now cover resource load
statistics.
Tools:
This patch adds infrastructure for the new TestRunner function
statisticsClearInMemoryAndPersistentStore().
- WebKitTestRunner/InjectedBundle/Bindings/TestRunner.idl:
- WebKitTestRunner/InjectedBundle/TestRunner.cpp:
(WTR::TestRunner::statisticsClearInMemoryAndPersistentStore):
- WebKitTestRunner/InjectedBundle/TestRunner.h:
- WebKitTestRunner/TestController.cpp:
(WTR::TestController::statisticsClearInMemoryAndPersistentStore):
- WebKitTestRunner/TestController.h:
- WebKitTestRunner/TestInvocation.cpp:
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):
LayoutTests:
- http/tests/loading/resourceLoadStatistics/clear-in-memory-and-persistent-store-expected.txt: Added.
- http/tests/loading/resourceLoadStatistics/clear-in-memory-and-persistent-store.html: Added.
- platform/wk2/TestExpectations:
Marked as [ Pass ] for WebKit2.
- 4:44 PM Changeset in webkit [213687] by
-
- 3 edits in trunk/Source/WebCore
Only consider autoplay interference in the first 10 seconds of playback.
https://bugs.webkit.org/show_bug.cgi?id=169455
Reviewed by Alex Christensen.
If the media duration is relatively long, say, 10 minutes, and you pause 5 minutes in, that's
probably not a strong indication users may not want auto-play. This patch limits interference
events to the first 10 seconds of automatic playback.
- html/HTMLMediaElement.cpp:
(WebCore::HTMLMediaElement::setReadyState):
(WebCore::HTMLMediaElement::play):
(WebCore::HTMLMediaElement::playInternal):
(WebCore::HTMLMediaElement::pauseInternal):
(WebCore::HTMLMediaElement::setMuted):
(WebCore::HTMLMediaElement::mediaPlayerTimeChanged):
(WebCore::HTMLMediaElement::stopWithoutDestroyingMediaPlayer):
(WebCore::HTMLMediaElement::userDidInterfereWithAutoplay):
(WebCore::HTMLMediaElement::setPlaybackWithoutUserGesture):
(WebCore::HTMLMediaElement::pageMutedStateDidChange):
- html/HTMLMediaElement.h:
- 4:38 PM Changeset in webkit [213686] by
-
- 23 edits8 copies8 adds in trunk/Source
Add WKURLSchemeHandler API for handling custom protocols.
https://bugs.webkit.org/show_bug.cgi?id=169422
Reviewed by Tim Horton.
Source/WebCore:
- platform/URLParser.cpp:
(WebCore::URLParser::maybeCanonicalizeScheme):
(WebCore::URLParser::isSpecialScheme):
- platform/URLParser.h:
Source/WebKit2:
This introduces two new protocols to the WK2 API surface:
- WKURLSchemeHandler
- WKURLSchemeHandlerTask
By registering an object that conforms to the WKURLSchemeHandler protocol with a WKWebViewConfiguration an application will
be handed an object conforming to WKURLSchemeHandlerTask for each resource load in that WKWebView with a custom URL scheme.
Using that id <WKURLSchemeHandlerTask>, the application can communicate the data for the resource back to WebKit.
- Shared/API/APIObject.h:
- Shared/Cocoa/APIObject.mm:
(API::Object::newObject):
- Shared/WebPageCreationParameters.cpp:
(WebKit::WebPageCreationParameters::encode):
(WebKit::WebPageCreationParameters::decode):
- Shared/WebPageCreationParameters.h:
- UIProcess/API/APIURLSchemeHandlerTask.cpp:
(API::URLSchemeHandlerTask::create):
(API::URLSchemeHandlerTask::URLSchemeHandlerTask):
- UIProcess/API/APIURLSchemeHandlerTask.h:
Add the WKURLSchemeHandler @protocol:
- UIProcess/API/Cocoa/WKURLSchemeHandler.h: Added.
Add the WKURLSchemeHandlerTask @protocol, as well as WebKit's concrete implementation of it:
- UIProcess/API/Cocoa/WKURLSchemeHandlerTask.h: Added.
- UIProcess/API/Cocoa/WKURLSchemeHandlerTask.mm: Added.
(raiseExceptionIfNecessary):
(-[WKURLSchemeHandlerTaskImpl request]):
(-[WKURLSchemeHandlerTaskImpl didReceiveResponse:]):
(-[WKURLSchemeHandlerTaskImpl didReceiveData:]):
(-[WKURLSchemeHandlerTaskImpl didFinish]):
(-[WKURLSchemeHandlerTaskImpl didFailWithError:]):
(-[WKURLSchemeHandlerTaskImpl _apiObject]):
- UIProcess/API/Cocoa/WKURLSchemeHandlerTaskInternal.h:
(WebKit::wrapper):
- UIProcess/API/Cocoa/WKWebView.h:
- UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):
(-[WKWebView urlSchemeHandlerForURLScheme:]):
(+[WKWebView handlesURLScheme:]):
- UIProcess/API/Cocoa/WKWebViewConfiguration.h:
- UIProcess/API/Cocoa/WKWebViewConfiguration.mm:
(-[WKWebViewConfiguration copyWithZone:]):
(-[WKWebViewConfiguration setURLSchemeHandler:forURLScheme:]):
(-[WKWebViewConfiguration urlSchemeHandlerForURLScheme:]):
(-[WKWebViewConfiguration _urlSchemeHandlers]):
- UIProcess/API/Cocoa/WKWebViewConfigurationInternal.h:
- UIProcess/Cocoa/WebURLSchemeHandlerCocoa.h:
(WebKit::WebURLSchemeHandlerCocoa::apiHandler):
- UIProcess/Cocoa/WebURLSchemeHandlerCocoa.mm: Added.
(WebKit::WebURLSchemeHandlerCocoa::create):
(WebKit::WebURLSchemeHandlerCocoa::WebURLSchemeHandlerCocoa):
(WebKit::WebURLSchemeHandlerCocoa::platformStartTask):
(WebKit::WebURLSchemeHandlerCocoa::platformStopTask):
- UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::creationParameters):
(WebKit::WebPageProxy::setURLSchemeHandlerForScheme):
(WebKit::WebPageProxy::urlSchemeHandlerForScheme):
(WebKit::WebPageProxy::startURLSchemeHandlerTask):
(WebKit::WebPageProxy::stopURLSchemeHandlerTask):
- UIProcess/WebPageProxy.h:
- UIProcess/WebPageProxy.messages.in:
- UIProcess/WebURLSchemeHandler.cpp:
(WebKit::generateWebURLSchemeHandlerIdentifier):
(WebKit::WebURLSchemeHandler::WebURLSchemeHandler):
(WebKit::WebURLSchemeHandler::~WebURLSchemeHandler):
(WebKit::WebURLSchemeHandler::startTask):
(WebKit::WebURLSchemeHandler::stopTask):
- UIProcess/WebURLSchemeHandler.h:
(WebKit::WebURLSchemeHandler::identifier):
- UIProcess/WebURLSchemeHandlerTask.cpp: Added.
(WebKit::WebURLSchemeHandlerTask::create):
(WebKit::WebURLSchemeHandlerTask::WebURLSchemeHandlerTask):
(WebKit::WebURLSchemeHandlerTask::didReceiveResponse):
(WebKit::WebURLSchemeHandlerTask::didReceiveData):
(WebKit::WebURLSchemeHandlerTask::didComplete):
(WebKit::WebURLSchemeHandlerTask::pageDestroyed):
(WebKit::WebURLSchemeHandlerTask::stop):
- UIProcess/WebURLSchemeHandlerTask.h: Added.
(WebKit::WebURLSchemeHandlerTask::identifier):
(WebKit::WebURLSchemeHandlerTask::request):
- WebProcess/Network/WebLoaderStrategy.cpp:
(WebKit::WebLoaderStrategy::scheduleLoad):
(WebKit::WebLoaderStrategy::remove):
- WebProcess/Network/WebLoaderStrategy.h:
- WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::urlSchemeHandlerForScheme):
(WebKit::WebPage::registerURLSchemeHandler):
(WebKit::WebPage::urlSchemeHandlerTaskDidReceiveResponse):
(WebKit::WebPage::urlSchemeHandlerTaskDidReceiveData):
(WebKit::WebPage::urlSchemeHandlerTaskDidComplete):
- WebProcess/WebPage/WebPage.h:
- WebProcess/WebPage/WebPage.messages.in:
- WebProcess/WebPage/WebURLSchemeHandlerProxy.cpp: Added.
(WebKit::WebURLSchemeHandlerProxy::WebURLSchemeHandlerProxy):
(WebKit::WebURLSchemeHandlerProxy::~WebURLSchemeHandlerProxy):
(WebKit::WebURLSchemeHandlerProxy::startNewTask):
(WebKit::WebURLSchemeHandlerProxy::taskDidReceiveResponse):
(WebKit::WebURLSchemeHandlerProxy::taskDidReceiveData):
(WebKit::WebURLSchemeHandlerProxy::taskDidComplete):
- WebProcess/WebPage/WebURLSchemeHandlerProxy.h:
(WebKit::WebURLSchemeHandlerProxy::identifier):
(WebKit::WebURLSchemeHandlerProxy::page):
- WebProcess/WebPage/WebURLSchemeHandlerTaskProxy.cpp: Added.
(WebKit::WebURLSchemeHandlerTaskProxy::WebURLSchemeHandlerTaskProxy):
(WebKit::WebURLSchemeHandlerTaskProxy::startLoading):
(WebKit::WebURLSchemeHandlerTaskProxy::stopLoading):
(WebKit::WebURLSchemeHandlerTaskProxy::didReceiveResponse):
(WebKit::WebURLSchemeHandlerTaskProxy::didReceiveData):
(WebKit::WebURLSchemeHandlerTaskProxy::didComplete):
- WebProcess/WebPage/WebURLSchemeHandlerTaskProxy.h:
(WebKit::WebURLSchemeHandlerTaskProxy::request):
- WebKit2.xcodeproj/project.pbxproj:
- CMakeLists.txt:
- 4:16 PM Changeset in webkit [213685] by
-
- 2 edits in trunk/Source/WebKit2
Unreviewed follow-up to r213682.
- NetworkProcess/cocoa/NetworkSessionCocoa.mm:
(-[WKNetworkSessionDelegate URLSession:task:didFinishCollectingMetrics:]):
Forward declare SPIs to address possible build errors.
- 4:00 PM Changeset in webkit [213684] by
-
- 10 edits4 copies in trunk
WebGPU: Backend - Buffers
https://bugs.webkit.org/show_bug.cgi?id=169451
<rdar://problem/30959168>
Reviewed by Tim Horton.
Source/WebCore:
Implement the backend for GPUBuffer, which
represents a chunk of memory that can be used
by the GPU. You create one using an ArrayBufferView.
WebKit API Test: GPUBuffer.mm
- PlatformMac.cmake:
- WebCore.xcodeproj/project.pbxproj:
- platform/graphics/cocoa/GPUBufferMetal.mm:
(WebCore::GPUBuffer::GPUBuffer):
(WebCore::GPUBuffer::length):
(WebCore::GPUBuffer::contents):
(WebCore::GPUBuffer::platformBuffer):
- platform/graphics/gpu/GPUBuffer.cpp:
(WebCore::GPUBuffer::create):
(WebCore::GPUBuffer::~GPUBuffer):
(WebCore::GPUBuffer::length):
(WebCore::GPUBuffer::contents):
- platform/graphics/gpu/GPUBuffer.h:
- platform/graphics/gpu/GPUDevice.cpp:
(WebCore::GPUDevice::createBufferFromData):
- platform/graphics/gpu/GPUDevice.h:
Tools:
Test for GPUBuffer object.
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
- TestWebKitAPI/Tests/WebCore/mac/GPUBuffer.mm: Added.
(TestWebKitAPI::TEST_F):
- TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm:
- TestWebKitAPI/Tests/WebCore/mac/GPULibrary.mm:
- 3:58 PM Changeset in webkit [213683] by
-
- 7 edits in trunk/Source/WebKit2
WKWebView should provide a way for clients to know the specific media state.
https://bugs.webkit.org/show_bug.cgi?id=169140
<rdar://problem/30838285>
Patch by Andrew Gold <agold@apple.com> on 2017-03-09
Reviewed by Alex Christensen.
Instead of notifying the UIDelegate only when a capture begins and ends,
we now notify the UIDelegate of any change in capture state.
- UIProcess/API/APIUIClient.h:
(API::UIClient::mediaCaptureStateDidChange):
(API::UIClient::didBeginCaptureSession): Deleted.
(API::UIClient::didEndCaptureSession): Deleted.
- UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
- UIProcess/API/Cocoa/WKWebViewPrivate.h:
- UIProcess/Cocoa/UIDelegate.h:
- UIProcess/Cocoa/UIDelegate.mm:
(WebKit::UIDelegate::setDelegate):
(WebKit::UIDelegate::UIClient::mediaCaptureStateDidChange):
(WebKit::UIDelegate::UIClient::didBeginCaptureSession): Deleted.
(WebKit::UIDelegate::UIClient::didEndCaptureSession): Deleted.
- UIProcess/WebPageProxy.cpp:
(WebKit::WebPageProxy::isPlayingMediaDidChange):
- 3:44 PM Changeset in webkit [213682] by
-
- 31 edits2 adds in trunk
Web Inspector: Show HTTP protocol version and other Network Load Metrics (IP Address, Priority, Connection ID)
https://bugs.webkit.org/show_bug.cgi?id=29687
<rdar://problem/19281586>
Reviewed by Matt Baker and Brian Burg.
Source/JavaScriptCore:
- inspector/protocol/Network.json:
Add metrics object with optional properties to loadingFinished event.
Source/WebCore:
Test: http/tests/inspector/network/resource-metrics.html
- inspector/InspectorNetworkAgent.cpp:
(WebCore::toProtocol):
(WebCore::InspectorNetworkAgent::buildObjectForMetrics):
(WebCore::InspectorNetworkAgent::didFinishLoading):
Send metrics at didFinishLoading time, we do not have all of
these at didReceiveResponse time.
- inspector/InspectorInstrumentation.cpp:
(WebCore::InspectorInstrumentation::didFinishLoadingImpl):
- inspector/InspectorInstrumentation.h:
(WebCore::InspectorInstrumentation::didFinishLoading):
- inspector/InspectorNetworkAgent.h:
- loader/CrossOriginPreflightChecker.cpp:
(WebCore::CrossOriginPreflightChecker::validatePreflightResponse):
- loader/DocumentLoader.cpp:
(WebCore::DocumentLoader::finishedLoading):
- loader/ResourceLoadNotifier.cpp:
(WebCore::ResourceLoadNotifier::didFinishLoad):
(WebCore::ResourceLoadNotifier::dispatchDidFinishLoading):
(WebCore::ResourceLoadNotifier::sendRemainingDelegateMessages):
- loader/ResourceLoadNotifier.h:
- loader/SubresourceLoader.cpp:
(WebCore::SubresourceLoader::didFinishLoading):
- loader/appcache/ApplicationCacheGroup.cpp:
(WebCore::ApplicationCacheGroup::didFinishLoading):
Include or pass on NetworkLoadMetrics to Web Inspector.
- platform/network/NetworkLoadMetrics.h:
(WebCore::NetworkLoadMetrics::isolatedCopy):
(WebCore::NetworkLoadMetrics::reset):
(WebCore::NetworkLoadMetrics::operator==):
(WebCore::NetworkLoadMetrics::encode):
(WebCore::NetworkLoadMetrics::decode):
Add new optional metrics properties.
(WTF::Persistence::Coder<std::optional<WebCore::NetworkLoadPriority>>::encode):
(WTF::Persistence::Coder<std::optional<WebCore::NetworkLoadPriority>>::decode):
We never encode this but this is needed for the compiler.
- platform/spi/cocoa/NSURLConnectionSPI.h:
New SPI for NSURLSessionTaskTransactionMetrics details.
Source/WebInspectorUI:
These columns are available in the Network DataGrids, but are
initially hidden. They can be shown by right clicking on the
table header and showing these columns. We are going to rework
the default list of visible columns later.
- Localizations/en.lproj/localizedStrings.js:
New localized strings for data grid headers and Low/Medium/High.
- UserInterface/Controllers/FrameResourceManager.js:
(WebInspector.FrameResourceManager.prototype.resourceRequestDidFinishLoading):
- UserInterface/Protocol/NetworkObserver.js:
(WebInspector.NetworkObserver.prototype.loadingFinished):
Pass metrics on to the Resource.
- UserInterface/Models/Resource.js:
(WebInspector.Resource):
(WebInspector.Resource.displayNameForType):
(WebInspector.Resource.responseSourceFromPayload):
(WebInspector.Resource.networkPriorityFromPayload):
(WebInspector.Resource.connectionIdentifierFromPayload):
(WebInspector.Resource.prototype.get protocol):
(WebInspector.Resource.prototype.get priority):
(WebInspector.Resource.prototype.get remoteAddress):
(WebInspector.Resource.prototype.get connectionIdentifier):
(WebInspector.Resource.prototype.updateWithMetrics):
Include metrics accessors and default values.
- UserInterface/Views/NetworkGridContentView.js:
(WebInspector.NetworkGridContentView):
- UserInterface/Views/NetworkTimelineView.js:
(WebInspector.NetworkTimelineView):
Add metrics columns if the backend may be sending them.
- UserInterface/Views/ResourceTimelineDataGridNode.js:
(WebInspector.ResourceTimelineDataGridNode.prototype.get data):
(WebInspector.ResourceTimelineDataGridNode.prototype.createCellContent):
(WebInspector.ResourceTimelineDataGridNode.prototype._displayNameForPriority):
(WebInspector.ResourceTimelineDataGridNode.prototype._cachedCellContent):
Display strings for new columns.
- UserInterface/Views/TimelineDataGridNode.js:
- UserInterface/Views/TimelineRecordBar.js:
(WebInspector.TimelineRecordBar.prototype.refresh):
Avoid assertions if Graph column is unavailable.
- UserInterface/Views/DataGrid.js:
(WebInspector.DataGrid):
(WebInspector.DataGrid.prototype.createSettings):
(WebInspector.DataGrid.prototype.setColumnVisible):
Better support for restoring user preference of initially
hidden columns which the user may have shown.
- UserInterface/Views/ResourceDetailsSidebarPanel.js:
(WebInspector.ResourceDetailsSidebarPanel):
(WebInspector.ResourceDetailsSidebarPanel.prototype._refreshRequestAndResponse):
Show the Protocol and Priority in the Resources sidebar.
Source/WebKit2:
- NetworkProcess/cocoa/NetworkSessionCocoa.mm:
(toNetworkLoadPriority):
(-[WKNetworkSessionDelegate URLSession:task:didFinishCollectingMetrics:]):
- Shared/WebCoreArgumentCoders.h:
Additional optional metrics on NetworkLoadMetrics.
LayoutTests:
- http/tests/inspector/network/resource-metrics-expected.txt: Added.
- http/tests/inspector/network/resource-metrics.html: Added.
- 3:21 PM Changeset in webkit [213681] by
-
- 1 edit2 adds in trunk/LayoutTests
Add iOS baseline for compositing/contents-format/subpixel-antialiased-text-traversal.
Unreviewed test gardening.
- platform/ios-simulator/compositing/contents-format/subpixel-antialiased-text-traversal-expected.txt: Added.
- 3:14 PM Changeset in webkit [213680] by
-
- 4 edits in trunk/Source/JavaScriptCore
Minimal build is broken
https://bugs.webkit.org/show_bug.cgi?id=169416
Patch by Youenn Fablet <youenn@apple.com> on 2017-03-09
Reviewed by Chris Dumez.
Since we now have some JS built-ins that are not tied to a compilation flag, we can remove compilation guards around m_vm.
We could probably remove m_vm by ensuring m_jsDOMBindingInternals appear first but this might break very easily.
- Scripts/builtins/builtins_generate_internals_wrapper_header.py:
(generate_members):
- Scripts/builtins/builtins_generate_internals_wrapper_implementation.py:
(BuiltinsInternalsWrapperImplementationGenerator.generate_constructor):
- Scripts/tests/builtins/expected/WebCoreJSBuiltins.h-result:
- 2:49 PM Changeset in webkit [213679] by
-
- 2 edits in trunk/Source/WebCore
Use the right conditional.
- Modules/applepay/ApplePayError.idl:
- 2:47 PM Changeset in webkit [213678] by
-
- 2 edits in trunk/Source/WTF
Activate VideoToolbox when WebRTC is enabled on iOS
https://bugs.webkit.org/show_bug.cgi?id=169423
Patch by Youenn Fablet <youenn@apple.com> on 2017-03-09
Reviewed by Jon Lee.
- wtf/Platform.h: Activate VIDEOTOOLBOX if WEBRTC is enabled.
- 2:47 PM Changeset in webkit [213677] by
-
- 1 edit in trunk/Source/JavaScriptCore/ChangeLog
Guard Credential Management implementation behind a runtime enabled feature flag
https://bugs.webkit.org/show_bug.cgi?id=169364
<rdar://problem/30957425>
Reviewed by Brent Fulgham.
Add common identifiers for Credential, PasswordCredential, and SiteBoundCredential that are
needed to guard these interfaces behind a runtime enabled feature flag.
- runtime/CommonIdentifiers.h:
- 2:40 PM Changeset in webkit [213676] by
-
- 30 edits in trunk
Guard Credential Management implementation behind a runtime enabled feature flag
https://bugs.webkit.org/show_bug.cgi?id=169364
<rdar://problem/30957425>
Reviewed by Brent Fulgham.
Source/JavaScriptCore:
Add common identifiers for Credential, PasswordCredential, and SiteBoundCredential that are
needed to guard these interfaces behind a runtime enabled feature flag.
- runtime/CommonIdentifiers.h:
Source/WebCore:
Add a runtime enabled feature flag for Credential Management (disabled by default).
Once we are happy with this feature we can enable the feature by default or remove the
the runtime enabled feature flag so as to enable it by default.
- Modules/credentials/BasicCredential.idl: Enable only if the runtime enabled feature flag is enabled.
- Modules/credentials/PasswordCredential.idl: Ditto.
- Modules/credentials/SiteBoundCredential.idl: Ditto.
- page/RuntimeEnabledFeatures.h:
(WebCore::RuntimeEnabledFeatures::setCredentialManagementEnabled): Added.
(WebCore::RuntimeEnabledFeatures::credentialManagementEnabled): Added.
Source/WebKit/mac:
Add WebKit preference for Credential Management (disabled by default).
- WebView/WebPreferenceKeysPrivate.h:
- WebView/WebPreferences.mm:
(+[WebPreferences initialize]):
(-[WebPreferences credentialManagementEnabled]): Added.
(-[WebPreferences setCredentialManagementEnabled:]): Added.
- WebView/WebPreferencesPrivate.h:
- WebView/WebView.mm:
(-[WebView _preferencesChanged:]):
Source/WebKit2:
Add WebKit2 preference for Credential Management (disabled by default).
- Shared/WebPreferencesDefinitions.h:
- UIProcess/API/C/WKPreferences.cpp:
(WKPreferencesSetCredentialManagementEnabled): Added.
(WKPreferencesGetCredentialManagementEnabled): Added.
- UIProcess/API/C/WKPreferencesRefPrivate.h:
- WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updatePreferences):
Tools:
Add a new test option header, enableCredentialManagement, to DumpRenderTree/WebKitTestRunner
to enable Credential Management.
- DumpRenderTree/TestOptions.h:
- DumpRenderTree/TestOptions.mm:
(TestOptions::TestOptions):
- DumpRenderTree/mac/DumpRenderTree.mm:
(enableExperimentalFeatures):
(setWebPreferencesForTestOptions):
- WebKitTestRunner/TestController.cpp:
(WTR::TestController::resetPreferencesToConsistentValues):
(WTR::updateTestOptionsFromTestHeader):
- WebKitTestRunner/TestOptions.h:
LayoutTests:
- credentials/idlharness.html: Use comment header to enable Credential Management API.
- platform/mac-elcapitan/js/dom/global-constructors-attributes-expected.txt: Update expected results now
that we no longer expose the Credential Management API by default.
- platform/mac-wk1/js/dom/global-constructors-attributes-expected.txt: Ditto.
- platform/mac-yosemite/js/dom/global-constructors-attributes-expected.txt: Ditto.
- platform/mac/js/dom/global-constructors-attributes-expected.txt: Ditto.
- platform/win/TestExpectations: For now, skip the Credential Management API tests.
- 2:39 PM Changeset in webkit [213675] by
-
- 4 edits5 moves in trunk/Source/JavaScriptCore
Refactoring some HeapVerifier code.
https://bugs.webkit.org/show_bug.cgi?id=169443
Reviewed by Filip Pizlo.
Renamed LiveObjectData to CellProfile.
Renamed LiveObjectList to CellList.
Moved CellProfile.*, CellList.*, and HeapVerifier.* from the heap folder to the tools folder.
Updated the HeapVerifier to handle JSCells instead of just JSObjects.
This is in preparation for subsequent patches to fix up the HeapVerifier for service again.
- CMakeLists.txt:
- JavaScriptCore.xcodeproj/project.pbxproj:
- heap/Heap.cpp:
(JSC::Heap::runBeginPhase):
(JSC::Heap::runEndPhase):
- heap/HeapVerifier.cpp: Removed.
- heap/HeapVerifier.h: Removed.
- heap/LiveObjectData.h: Removed.
- heap/LiveObjectList.cpp: Removed.
- heap/LiveObjectList.h: Removed.
- tools/CellList.cpp: Copied from Source/JavaScriptCore/heap/LiveObjectList.cpp.
(JSC::CellList::findCell):
(JSC::LiveObjectList::findObject): Deleted.
- tools/CellList.h: Copied from Source/JavaScriptCore/heap/LiveObjectList.h.
(JSC::CellList::CellList):
(JSC::CellList::reset):
(JSC::LiveObjectList::LiveObjectList): Deleted.
(JSC::LiveObjectList::reset): Deleted.
- tools/CellProfile.h: Copied from Source/JavaScriptCore/heap/LiveObjectData.h.
(JSC::CellProfile::CellProfile):
(JSC::LiveObjectData::LiveObjectData): Deleted.
- tools/HeapVerifier.cpp: Copied from Source/JavaScriptCore/heap/HeapVerifier.cpp.
(JSC::GatherCellFunctor::GatherCellFunctor):
(JSC::GatherCellFunctor::visit):
(JSC::GatherCellFunctor::operator()):
(JSC::HeapVerifier::gatherLiveCells):
(JSC::HeapVerifier::cellListForGathering):
(JSC::trimDeadCellsFromList):
(JSC::HeapVerifier::trimDeadCells):
(JSC::HeapVerifier::verifyButterflyIsInStorageSpace):
(JSC::HeapVerifier::reportCell):
(JSC::HeapVerifier::checkIfRecorded):
(JSC::GatherLiveObjFunctor::GatherLiveObjFunctor): Deleted.
(JSC::GatherLiveObjFunctor::visit): Deleted.
(JSC::GatherLiveObjFunctor::operator()): Deleted.
(JSC::HeapVerifier::gatherLiveObjects): Deleted.
(JSC::HeapVerifier::liveObjectListForGathering): Deleted.
(JSC::trimDeadObjectsFromList): Deleted.
(JSC::HeapVerifier::trimDeadObjects): Deleted.
(JSC::HeapVerifier::reportObject): Deleted.
- tools/HeapVerifier.h: Copied from Source/JavaScriptCore/heap/HeapVerifier.h.
- 2:37 PM Changeset in webkit [213674] by
-
- 9 edits in trunk/Source/WebKit2
[iOS] Add SPI to print to a single tall PDF page
https://bugs.webkit.org/show_bug.cgi?id=169439
<rdar://problem/30120532>
Reviewed by Tim Horton.
- Shared/PrintInfo.cpp:
(WebKit::PrintInfo::encode): Encoded snapshotFirstPage.
(WebKit::PrintInfo::decode): Decoded snapshotFirstPage.
(WebKit::PrintInfo::PrintInfo): Deleted.
- Shared/PrintInfo.h: Initialized member variables in place instead of in a default ctor.
- UIProcess/_WKWebViewPrintFormatter.h: Declared BOOL property snapshotFirstPage.
- UIProcess/_WKWebViewPrintFormatter.mm:
(-[_WKWebViewPrintFormatter _setSnapshotPaperRect:]): Added to set a custom paper size.
(-[_WKWebViewPrintFormatter rectForPageAtIndex:]): Returned the custom paper rect if
snapshotFirstPage is true.
- UIProcess/_WKWebViewPrintFormatterInternal.h:
- UIProcess/ios/WKContentView.mm:
(-[WKContentView _wk_pageCountForPrintFormatter:]): Computed the custom paper size to be the
smaller of the document height and the maximum PDF page height.
- UIProcess/ios/WKPDFView.mm:
(-[WKPDFView _wk_pageCountForPrintFormatter:]): Changed to return a maximum page size of 1
if snapshotFirstPage is true.
- WebProcess/WebPage/ios/WebPageIOS.mm:
(WebKit::WebPage::computePagesForPrintingAndDrawToPDF): Ditto.
- 2:34 PM Changeset in webkit [213673] by
-
- 17 edits11 adds in trunk
Add delegate support to WebCore
https://bugs.webkit.org/show_bug.cgi?id=169427
Part of rdar://problem/28880714.
Reviewed by Geoffrey Garen.
Source/JavaScriptCore:
- Configurations/FeatureDefines.xcconfig:
Add feature define.
Source/WebCore:
- Configurations/FeatureDefines.xcconfig:
Add feature define.
- DerivedSources.make:
Add idl files.
- Modules/applepay/ApplePayError.cpp:
(WebCore::ApplePayError::create):
(WebCore::ApplePayError::ApplePayError):
(WebCore::ApplePayError::~ApplePayError):
- Modules/applepay/ApplePayError.h:
- Modules/applepay/ApplePayError.idl:
- Modules/applepay/ApplePayPaymentAuthorizationResult.h:
- Modules/applepay/ApplePayPaymentAuthorizationResult.idl:
- Modules/applepay/ApplePayPaymentMethodUpdate.h:
- Modules/applepay/ApplePayPaymentMethodUpdate.idl:
- Modules/applepay/ApplePayShippingContactUpdate.h:
- Modules/applepay/ApplePayShippingContactUpdate.idl:
- Modules/applepay/ApplePayShippingMethodUpdate.h:
- Modules/applepay/ApplePayShippingMethodUpdate.idl:
Add new files.
- Modules/applepay/ApplePaySession.cpp:
(WebCore::toPaymentAuthorizationStatus):
Move this function to the top.
(WebCore::convertAndValidate):
Add new convertAndValidate functions.
(WebCore::ApplePaySession::completeShippingMethodSelection):
(WebCore::ApplePaySession::completeShippingContactSelection):
(WebCore::ApplePaySession::completePaymentMethodSelection):
(WebCore::ApplePaySession::completePayment):
- Modules/applepay/ApplePaySession.h:
- Modules/applepay/ApplePaySession.idl:
Implement the old functions in terms of the new functions.
- Modules/applepay/PaymentCoordinator.cpp:
- WebCore.xcodeproj/project.pbxproj:
Add new files.
Source/WebCore/PAL:
- Configurations/FeatureDefines.xcconfig:
Add feature define.
Source/WebKit/mac:
- Configurations/FeatureDefines.xcconfig:
Add feature define.
Source/WebKit2:
- Configurations/FeatureDefines.xcconfig:
Add feature define.
Tools:
- TestWebKitAPI/Configurations/FeatureDefines.xcconfig:
- 2:29 PM Changeset in webkit [213672] by
-
- 5 edits5 adds in trunk
[link preload] Double downloads of preloaded CSS
https://bugs.webkit.org/show_bug.cgi?id=169274
Reviewed by Antti Koivisto.
Source/WebCore:
Avoid reloading link preloads in case of a charset mismatch.
Charset mismatch can happen for header based preloads, as they are requested before
the HTML's<meta charset>tag is processed. This change makes sure that in those
cases, we modify the resource's encoding setting instead of reloading it.
Test: http/tests/preload/single_download_preload_headers.php
- loader/cache/CachedResource.cpp:
(WebCore::CachedResource::CachedResource): Initialize m_unknownCharset to be the same as the preload flag.
- loader/cache/CachedResource.h:
(WebCore::CachedResource::hasUnknownEncoding):
(WebCore::CachedResource::setHasUnknownEncoding):
- loader/cache/CachedResourceLoader.cpp:
(WebCore::CachedResourceLoader::determineRevalidationPolicy): In case of a charset
mismatch, set the encoding of the Resource instead of reloading it if the charset is unknown.
LayoutTests:
Added tests making sure that header based preloads also trigger a single download,
and that we properly handle multiple charsets for the same preloaded resource.
- http/tests/preload/single_download_preload_headers-expected.txt: Added.
- http/tests/preload/single_download_preload_headers.php: Added.
- http/tests/preload/preload-encoding-expected.txt: Added.
- http/tests/preload/preload-encoding.php: Added.
- 2:24 PM Changeset in webkit [213671] by
-
- 21 edits6 copies23 adds in trunk
Implement PBKDF2 in WebCrypto
https://bugs.webkit.org/show_bug.cgi?id=143458
<rdar://problem/27311703>
Reviewed by Brent Fulgham.
LayoutTests/imported/w3c:
- web-platform-tests/WebCryptoAPI/derive_bits_keys/pbkdf2.worker-expected.txt:
- web-platform-tests/WebCryptoAPI/derive_bits_keys/test_pbkdf2-expected.txt:
- web-platform-tests/WebCryptoAPI/import_export/symmetric_importKey.worker-expected.txt:
- web-platform-tests/WebCryptoAPI/import_export/test_symmetric_importKey-expected.txt:
Source/WebCore:
This patch implements PBKDF2 according to the spec:
https://www.w3.org/TR/WebCryptoAPI/#pbkdf2. Supported
operations include deriveKey, deriveBits, importKey and
getKeyLength.
Tests: crypto/subtle/ecdh-import-key-derive-pbkdf2-key.html
crypto/subtle/pbkdf2-derive-bits-malformed-parametrs.html
crypto/subtle/pbkdf2-import-key-derive-bits.html
crypto/subtle/pbkdf2-import-key-derive-hmac-key.html
crypto/subtle/pbkdf2-import-key-malformed-parameters.html
crypto/subtle/pbkdf2-import-key.html
crypto/workers/subtle/pbkdf2-import-key-derive-bits.html
crypto/workers/subtle/pbkdf2-import-key-derive-hmac-key.html
crypto/workers/subtle/pbkdf2-import-key.html
- CMakeLists.txt:
- DerivedSources.make:
- PlatformGTK.cmake:
- PlatformMac.cmake:
- WebCore.xcodeproj/project.pbxproj:
- bindings/js/JSCryptoKeyCustom.cpp:
(WebCore::JSCryptoKey::algorithm):
Added RawKeyAlgorithm.
- bindings/js/JSSubtleCryptoCustom.cpp:
(WebCore::normalizeCryptoAlgorithmParameters):
A bit of reordering.
- bindings/js/SerializedScriptValue.cpp:
(WebCore::CloneSerializer::write):
Faked implementation.
- crypto/CryptoAlgorithmParameters.h:
- crypto/CryptoKey.h:
- crypto/algorithms/CryptoAlgorithmPBKDF2.cpp: Added.
(WebCore::CryptoAlgorithmPBKDF2::create):
(WebCore::CryptoAlgorithmPBKDF2::identifier):
(WebCore::CryptoAlgorithmPBKDF2::deriveBits):
(WebCore::CryptoAlgorithmPBKDF2::importKey):
(WebCore::CryptoAlgorithmPBKDF2::getKeyLength):
- crypto/algorithms/CryptoAlgorithmPBKDF2.h: Added.
- crypto/gnutls/CryptoAlgorithmPBKDF2GnuTLS.cpp: Added.
(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):
- crypto/keys/CryptoKeyRaw.cpp: Added.
(WebCore::CryptoKeyRaw::CryptoKeyRaw):
(WebCore::CryptoKeyRaw::buildAlgorithm):
(WebCore::CryptoKeyRaw::exportData):
- crypto/keys/CryptoKeyRaw.h: Added.
(WebCore::RawKeyAlgorithm::RawKeyAlgorithm):
The CryptoKey class for PBKDF2 and later on HKDF.
- crypto/mac/CryptoAlgorithmPBKDF2Mac.cpp: Added.
(WebCore::commonCryptoHMACAlgorithm):
(WebCore::CryptoAlgorithmPBKDF2::platformDeriveBits):
- crypto/mac/CryptoAlgorithmRegistryMac.cpp:
(WebCore::CryptoAlgorithmRegistry::platformRegisterAlgorithms):
- crypto/parameters/CryptoAlgorithmPbkdf2Params.h: Added.
- crypto/parameters/Pbkdf2Params.idl: Added.
LayoutTests:
Mark slow tests as slow.
- crypto/subtle/derive-key-malformed-parameters.html:
- crypto/subtle/ecdh-import-key-derive-pbkdf2-key-expected.txt: Added.
- crypto/subtle/ecdh-import-key-derive-pbkdf2-key.html: Added.
- crypto/subtle/pbkdf2-derive-bits-malformed-parametrs-expected.txt: Added.
- crypto/subtle/pbkdf2-derive-bits-malformed-parametrs.html: Added.
- crypto/subtle/pbkdf2-import-key-derive-bits-expected.txt: Added.
- crypto/subtle/pbkdf2-import-key-derive-bits.html: Added.
- crypto/subtle/pbkdf2-import-key-derive-hmac-key-expected.txt: Added.
- crypto/subtle/pbkdf2-import-key-derive-hmac-key.html: Added.
- crypto/subtle/pbkdf2-import-key-expected.txt: Added.
- crypto/subtle/pbkdf2-import-key-malformed-parameters-expected.txt: Added.
- crypto/subtle/pbkdf2-import-key-malformed-parameters.html: Added.
- crypto/subtle/pbkdf2-import-key.html: Added.
- crypto/workers/subtle/pbkdf2-import-key-derive-bits-expected.txt: Added.
- crypto/workers/subtle/pbkdf2-import-key-derive-bits.html: Added.
- crypto/workers/subtle/pbkdf2-import-key-derive-hmac-key-expected.txt: Added.
- crypto/workers/subtle/pbkdf2-import-key-derive-hmac-key.html: Added.
- crypto/workers/subtle/pbkdf2-import-key-expected.txt: Added.
- crypto/workers/subtle/pbkdf2-import-key.html: Added.
- crypto/workers/subtle/resources/pbkdf2-import-key-derive-bits.js: Added.
- crypto/workers/subtle/resources/pbkdf2-import-key-derive-hmac-key.js: Added.
- crypto/workers/subtle/resources/pbkdf2-import-key.js: Added.
- 2:17 PM Changeset in webkit [213670] by
-
- 4 edits2 adds in trunk
[Modern Media Controls] iOS: long press on play button shows the Copy menu
https://bugs.webkit.org/show_bug.cgi?id=169355
<rdar://problem/30928998>
Patch by Antoine Quint <Antoine Quint> on 2017-03-09
Reviewed by Dean Jackson.
Source/WebCore:
Disable built-in UI behavior that shows a magnifier or a copy callout when pressing
on an element for a long time. These do not make sense for media controls.
Test: media/modern-media-controls/audio/audio-controls-styles.html
- Modules/modern-media-controls/controls/media-controls.css:
(:host):
LayoutTests:
Add a test to check that the expected properties are set as they should to disable
built-in UI behavior that shows a magnifier or a copy callout when pressing on an
element for a long time.
- media/modern-media-controls/audio/audio-controls-styles-expected.txt: Added.
- media/modern-media-controls/audio/audio-controls-styles.html: Added.
- platform/mac/TestExpectations:
- 2:17 PM Changeset in webkit [213669] by
-
- 22 edits in trunk
[Content Extensions] Introduce if-top-url and unless-top-url
https://bugs.webkit.org/show_bug.cgi?id=169433
Reviewed by Brady Eidson.
Source/WebCore:
In r184116 I added if-domain and unless-domain to control whether a rule applies
based on the domain of the main document URL. I'm expanding this by adding if-top-url
and unless-top-url that run regular expressions on the entire main document URL so that
example.com/user1content can be distinguished from example.com/user2content.
To not add to the number of passes we make on the URLs for each load and to maintain JSON
backwards compatibility, I've made it so that if-top-url and unless-top-url can be used
instead of if-domain and unless-domain (which continue to work) but the two condition types
cannot be used together since running regular expressions on the entire main document URL
is strictly more powerful than checking the domain and subdomains.
As a minor detail, content extension regexes are by default ASCII-case-insensitive, so I've
done the same with top URL regexes, adding top-url-filter-is-case-sensitive to mirror the existing
url-filter-is-case-sensitive if the JSON author decides to make regexes case sensitive.
Covered by new API tests.
- contentextensions/CompiledContentExtension.h:
- contentextensions/ContentExtension.cpp:
(WebCore::ContentExtensions::ContentExtension::populateConditionCacheIfNeeded):
(WebCore::ContentExtensions::ContentExtension::topURLActions):
(WebCore::ContentExtensions::ContentExtension::cachedConditionedActions): Deleted.
- contentextensions/ContentExtension.h:
- contentextensions/ContentExtensionCompiler.cpp:
(WebCore::ContentExtensions::addUniversalActionsToDFA):
(WebCore::ContentExtensions::compileToBytecode):
We had three copies of compiling to bytecode that were almost the same and would've been made into three copies of the same code.
I moved them to one helper function that is called three times.
(WebCore::ContentExtensions::compileRuleList):
- contentextensions/ContentExtensionCompiler.h:
- contentextensions/ContentExtensionError.cpp:
(WebCore::ContentExtensions::contentExtensionErrorCategory):
Add the new error type for JSON that tries to use if-top-url and unless-top-url with if-domain and unless-domain.
- contentextensions/ContentExtensionError.h:
- contentextensions/ContentExtensionParser.cpp:
(WebCore::ContentExtensions::loadTrigger):
Parse the new values if-top-url, unless-top-url, and top-url-filter-is-case-sensitive.
- contentextensions/ContentExtensionRule.h:
(WebCore::ContentExtensions::Trigger::~Trigger):
- contentextensions/ContentExtensionsBackend.cpp:
(WebCore::ContentExtensions::ContentExtensionsBackend::actionsForResourceLoad):
- contentextensions/DFABytecodeInterpreter.cpp:
(WebCore::ContentExtensions::DFABytecodeInterpreter::interpretAppendAction):
(WebCore::ContentExtensions::DFABytecodeInterpreter::interpretTestFlagsAndAppendAction):
(WebCore::ContentExtensions::DFABytecodeInterpreter::interpretWithConditions):
- contentextensions/DFABytecodeInterpreter.h:
Source/WebKit2:
Rename conditionedFilters to topURLFilters to reflect the fact that they are the filters
that are run on the top URL, and possibly just the domain of the top url.
I was a bit too aggressive when renaming domain* to condition* in r213533.
- Shared/WebCompiledContentExtension.cpp:
(WebKit::WebCompiledContentExtension::conditionsApplyOnlyToDomain):
(WebKit::WebCompiledContentExtension::topURLFiltersBytecode):
(WebKit::WebCompiledContentExtension::topURLFiltersBytecodeLength):
(WebKit::WebCompiledContentExtension::conditionedFiltersBytecode): Deleted.
(WebKit::WebCompiledContentExtension::conditionedFiltersBytecodeLength): Deleted.
- Shared/WebCompiledContentExtension.h:
- Shared/WebCompiledContentExtensionData.cpp:
(WebKit::WebCompiledContentExtensionData::encode):
(WebKit::WebCompiledContentExtensionData::decode):
- Shared/WebCompiledContentExtensionData.h:
(WebKit::WebCompiledContentExtensionData::WebCompiledContentExtensionData):
- UIProcess/API/APIUserContentExtensionStore.cpp:
(API::encodeContentExtensionMetaData):
(API::decodeContentExtensionMetaData):
(API::compiledToFile):
(API::createExtension):
(API::UserContentExtensionStore::invalidateContentExtensionVersion):
(API::userContentExtensionStoreErrorCategory):
- UIProcess/API/APIUserContentExtensionStore.h:
Increment CurrentContentExtensionFileVersion because we have changed the format of the binary on disk.
We only added 4 bytes, but that's binary incompatible and requires re-compiling any existing content extensions.
Tools:
- TestWebKitAPI/Tests/WebCore/ContentExtensions.cpp:
(TestWebKitAPI::TEST_F):
Add tests for new functionality and new failure types.
- 2:07 PM Changeset in webkit [213668] by
-
- 3 edits in trunk/Websites/webkit.org
Prevent HTML in rich link previews
https://bugs.webkit.org/show_bug.cgi?id=169435
Reviewed by Tim Horton.
- wp-content/plugins/social-meta.php:
- 1:58 PM Changeset in webkit [213667] by
-
- 4 edits in tags/Safari-604.1.9/Source
Roll out r213557. rdar://problem/30955791
- 1:56 PM Changeset in webkit [213666] by
-
- 12 edits4 adds in trunk/Source
Web Inspector: Show individual messages in the content pane for a WebSocket
https://bugs.webkit.org/show_bug.cgi?id=169011
Reviewed by Joseph Pecoraro.
Source/JavaScriptCore:
Add walltime parameter and correct the description of Timestamp type.
- inspector/protocol/Network.json:
Source/WebCore:
Add walltime parameter.
No new tests. Tests will be added in a follow up patch.
- inspector/InspectorNetworkAgent.cpp:
(WebCore::InspectorNetworkAgent::willSendWebSocketHandshakeRequest):
Source/WebInspectorUI:
When selecting a Web Socket in Network panel or Resources, display a table
of all messages that have been sent and received.
- Localizations/en.lproj/localizedStrings.js:
- UserInterface/Controllers/FrameResourceManager.js:
(WebInspector.FrameResourceManager.prototype.webSocketWillSendHandshakeRequest):
Add a walltime parameter.
(WebInspector.FrameResourceManager.prototype._webSocketFrameReceivedOrSent):
Fix a bug: masked messages are outgoing, not incoming.
- UserInterface/Images/ArrowUp.svg: Added.
- UserInterface/Images/gtk/ArrowUp.svg: Added.
Add an icon for outgoing messages.
- UserInterface/Main.html:
- UserInterface/Models/WebSocketResource.js:
(WebInspector.WebSocketResource):
(WebInspector.WebSocketResource.prototype.get walltime):
(WebInspector.WebSocketResource.prototype.addFrame):
(WebInspector.WebSocketResource.prototype._walltimeForWebSocketTimestamp):
- UserInterface/Protocol/NetworkObserver.js:
(WebInspector.NetworkObserver.prototype.webSocketWillSendHandshakeRequest):
(WebInspector.NetworkObserver.prototype.webSocketFrameSent):
(WebInspector.NetworkObserver.prototype.webSocketFrameError):
(WebInspector.NetworkObserver):
- UserInterface/Views/NetworkSidebarPanel.js:
(WebInspector.NetworkSidebarPanel.prototype.treeElementAddedOrChanged):
- UserInterface/Views/ResourceClusterContentView.js:
(WebInspector.ResourceClusterContentView.prototype.get responseContentView):
- UserInterface/Views/WebSocketContentView.css: Added.
(.web-socket.content-view > .data-grid):
(.web-socket.content-view > .data-grid table.data):
(.web-socket.content-view > .data-grid td.data-column,):
(body[dir=ltr] .web-socket.content-view > .data-grid .data-column > div):
(body[dir=rtl] .web-socket.content-view > .data-grid .data-column > div):
(.web-socket.content-view .icon):
(body[dir=ltr] .web-socket.content-view .icon):
(body[dir=rtl] .web-socket.content-view .icon):
(.web-socket.content-view .outgoing .icon):
(.web-socket.content-view .data-grid.variable-height-rows table.data tr:nth-child(odd)):
(.web-socket.content-view .data-grid table.data tr.revealed):
(.web-socket.content-view .data-grid.variable-height-rows table.data tr.outgoing):
(.web-socket.content-view .data-grid.variable-height-rows table.data tr.non-text-frame):
- UserInterface/Views/WebSocketContentView.js: Added.
(WebInspector.WebSocketContentView):
Only show Time column when walltime is available.
(WebInspector.WebSocketContentView.textForOpcode):
(WebInspector.WebSocketContentView.prototype.shown):
(WebInspector.WebSocketContentView.prototype.hidden):
(WebInspector.WebSocketContentView.prototype.addFrame):
(WebInspector.WebSocketContentView.prototype._updateFrames):
Only render frames that haven't been rendered yet.
(WebInspector.WebSocketContentView.prototype._addRow):
(WebInspector.WebSocketContentView.prototype._timeStringFromTimestamp):
- 1:42 PM Changeset in webkit [213665] by
-
- 2 edits in trunk/Source/WebKit2
[WK2][iOS] Extend WebProcess sandbox to support audio and video compression/decompression
https://bugs.webkit.org/show_bug.cgi?id=169431
<rdar://problem/30844650>
Reviewed by Youenn Fablet.
- Resources/SandboxProfiles/ios/com.apple.WebKit.WebContent.sb:
- 1:39 PM Changeset in webkit [213664] by
-
- 2 edits in trunk/Source/JavaScriptCore
Unreviewed, fix weak external symbol error.
- heap/SlotVisitor.h:
- 1:28 PM Changeset in webkit [213663] by
-
- 2 edits in trunk/Source/WebCore
Unreviewed build fix after r213628.
Use 'size_t' for the function definition to match the header file. If you do not,
the 32-bit build fails..
- crypto/gnutls/CryptoAlgorithmECDHGnuTLS.cpp:
(WebCore::CryptoAlgorithmECDH::platformDeriveBits):
- 1:24 PM Changeset in webkit [213662] by
-
- 4 edits in trunk/Source
[WK2] Add a UI delegate hook for custom handling of data interaction operations
https://bugs.webkit.org/show_bug.cgi?id=169430
<rdar://problem/30954654>
Reviewed by Tim Horton.
Source/WebCore:
Ensure that m_documentUnderMouse is properly cleaned up even in the case where no drag operation was performed
due to the drop bailing at the client layer.
- page/DragController.cpp:
(WebCore::DragController::dragEnded):
Source/WebKit2:
Introduce a new SPI method on the UI delegate for clients who need to override data interaction operation behaviors.
Note that implementing this delegate method and returning YES will not propagate adropevent to the page.
- UIProcess/API/Cocoa/WKUIDelegatePrivate.h:
- 1:23 PM Changeset in webkit [213661] by
-
- 3 edits in trunk/LayoutTests
Rebaseline js/dom/global-constructors-attributes.html.
Unreviewed test gardening.
- platform/mac-elcapitan/js/dom/global-constructors-attributes-expected.txt:
- platform/mac-wk1/js/dom/global-constructors-attributes-expected.txt:
- 1:19 PM Changeset in webkit [213660] by
-
- 2 edits in trunk/Source/WebKit/mac
Fix DataInteraction guards
https://bugs.webkit.org/show_bug.cgi?id=169404
Reviewed by Wenson Hsieh.
- WebCoreSupport/WebDragClient.mm:
- 1:18 PM Changeset in webkit [213659] by
-
- 7 edits in trunk/Source
Source/JavaScriptCore:
std::isnan/isinf should work with WTF time classes
https://bugs.webkit.org/show_bug.cgi?id=164991
Reviewed by Darin Adler.
Changes AtomicsObject to use std::isnan() instead of operator== to detect NaN.
- runtime/AtomicsObject.cpp:
(JSC::atomicsFuncWait):
Source/WTF:
std::isnan/isinf/isfinite should work with WTF time classes
https://bugs.webkit.org/show_bug.cgi?id=164991
Reviewed by Darin Adler.
The consensus view (see comments in https://bugs.webkit.org/show_bug.cgi?id=152045) of how
to check if something is NaN is to use std::isnan(). To be able to do that for time
classes, they need to provide their own isnan() overhload. This also provides isinf()
overloads.
- wtf/MonotonicTime.h:
(std::isnan):
(std::isinf):
- wtf/Seconds.h:
(std::isnan):
(std::isinf):
- wtf/TimeWithDynamicClockType.h:
(std::isnan):
(std::isinf):
- wtf/WallTime.h:
(std::isnan):
(std::isinf):
- 1:13 PM Changeset in webkit [213658] by
-
- 2 edits in trunk/Tools
JSC EWS repeat the log message thrice about patch not being relevant
https://bugs.webkit.org/show_bug.cgi?id=169330
Patch by Srinivasan Vijayaraghavan <svijayaraghavan@apple.com> on 2017-03-09
Reviewed by Chris Dumez.
- Scripts/webkitpy/tool/steps/checkpatchrelevance.py:
(CheckPatchRelevance.run): Change log level to 'debug' to reduce stdout traffic.
- 1:05 PM Changeset in webkit [213657] by
-
- 16 edits in trunk/Source
Use const AbstractLocker& (instead of const LockHolder&) in more places.
https://bugs.webkit.org/show_bug.cgi?id=169424
Reviewed by Filip Pizlo.
Source/JavaScriptCore:
- heap/CodeBlockSet.cpp:
(JSC::CodeBlockSet::promoteYoungCodeBlocks):
- heap/CodeBlockSet.h:
- heap/CodeBlockSetInlines.h:
(JSC::CodeBlockSet::mark):
- heap/ConservativeRoots.cpp:
(JSC::CompositeMarkHook::CompositeMarkHook):
- heap/MachineStackMarker.cpp:
(JSC::MachineThreads::tryCopyOtherThreadStacks):
- heap/MachineStackMarker.h:
- profiler/ProfilerDatabase.cpp:
(JSC::Profiler::Database::ensureBytecodesFor):
- profiler/ProfilerDatabase.h:
- runtime/SamplingProfiler.cpp:
(JSC::FrameWalker::FrameWalker):
(JSC::CFrameWalker::CFrameWalker):
(JSC::SamplingProfiler::createThreadIfNecessary):
(JSC::SamplingProfiler::takeSample):
(JSC::SamplingProfiler::start):
(JSC::SamplingProfiler::pause):
(JSC::SamplingProfiler::noticeCurrentThreadAsJSCExecutionThread):
(JSC::SamplingProfiler::clearData):
(JSC::SamplingProfiler::releaseStackTraces):
- runtime/SamplingProfiler.h:
(JSC::SamplingProfiler::setStopWatch):
- wasm/WasmMemory.cpp:
(JSC::Wasm::availableFastMemories):
(JSC::Wasm::activeFastMemories):
(JSC::Wasm::viewActiveFastMemories):
- wasm/WasmMemory.h:
Source/WTF:
- wtf/RunLoop.h:
- wtf/generic/RunLoopGeneric.cpp:
(WTF::RunLoop::wakeUp):
(WTF::RunLoop::schedule):
- 12:47 PM Changeset in webkit [213656] by
-
- 15 edits in trunk
Let clients control autoplay quirks with website policies.
https://bugs.webkit.org/show_bug.cgi?id=169390
Reviewed by Wenson Hsieh.
Many sites incorrectly assume media will start off playing and display a pause button from the
start. We can trick many of these sites into updating their controls by sending a "pause"
event when we prevent a media element from playing.
We don't want this to become standard web behavior, however, since ultimately sites should adapt
to the new behavior. This patch lets clients decide which limited set of sites gets to use auto-play
quirks.
Source/WebCore:
- html/HTMLMediaElement.cpp:
(WebCore::needsAutoplayPlayPauseEventsQuirk):
- loader/DocumentLoader.h:
(WebCore::DocumentLoader::allowsAutoplayQuirks):
(WebCore::DocumentLoader::setAllowsAutoplayQuirks):
Source/WebKit2:
- Shared/WebsitePolicies.h:
(WebKit::WebsitePolicies::encode):
(WebKit::WebsitePolicies::decode):
- UIProcess/API/APIWebsitePolicies.h:
- UIProcess/API/C/WKWebsitePolicies.cpp:
(WKWebsitePoliciesSetAllowsAutoplayQuirks):
(WKWebsitePoliciesGetAllowsAutoplayQuirks):
- UIProcess/API/C/WKWebsitePolicies.h:
- UIProcess/API/Cocoa/_WKWebsitePolicies.h:
- UIProcess/API/Cocoa/_WKWebsitePolicies.mm:
(-[_WKWebsitePolicies setAllowsAutoplayQuirks:]):
(-[_WKWebsitePolicies allowsAutoplayQuirks]):
- WebProcess/WebCoreSupport/WebFrameLoaderClient.cpp:
(WebKit::WebFrameLoaderClient::dispatchDecidePolicyForNavigationAction):
- WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updateWebsitePolicies):
Tools:
- TestWebKitAPI/Tests/WebKit2/autoplay-check.html:
- TestWebKitAPI/Tests/WebKit2Cocoa/WebsitePolicies.mm:
(-[AutoplayPoliciesDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]):
(TEST):
- 12:35 PM Changeset in webkit [213655] by
-
- 2 edits in trunk/Tools
Unreviewed, change my status to be a WebKit reviewer
- Scripts/webkitpy/common/config/contributors.json:
- 11:46 AM Changeset in webkit [213654] by
-
- 9 edits in trunk/Tools
Standardize device/simulator naming conventions
https://bugs.webkit.org/show_bug.cgi?id=169083
<rdar://problem/30810466>
Reviewed by Alexey Proskuryakov.
- Scripts/build-webkit: Document --ios-device and --simulator options.
- Scripts/package-root:
(usage): Ditto.
- Scripts/run-api-tests: Document --simulator option.
- Scripts/webkitdirs.pm:
(argumentsForConfiguration): Use --ios-device by default rather than --device.
(determineXcodeSDK): Parse --ios-device and --simulator options.
- Scripts/webkitpy/common/config/ews.json: Change port name from ios to ios-device.
- Scripts/webkitpy/common/config/ports.py: Ditto.
(DeprecatedPort.port):
(IOSPort): Change port name from ios to ios-device.
- Scripts/webkitpy/port/factory.py:
(platform_options): Add --simulator option.
(PortFactory): Update comment.
- Scripts/webkitpy/port/ios_device.py:
(IOSDevicePort): Change port name from ios to ios-device.
(IOSDevicePort.operating_system): Ditto.
- 11:18 AM Changeset in webkit [213653] by
-
- 5 edits1 add in trunk
WebAssembly: Make the Unity AngryBots demo run
https://bugs.webkit.org/show_bug.cgi?id=169268
Reviewed by Keith Miller.
JSTests:
- wasm/function-tests/many-arguments-to-function.js: Added.
(import.Builder.from.string_appeared_here.import.as.assert.from.string_appeared_here.I32Const.0.I32Const.1.I32Const.2.I32Const.3.I32Const.4.I32Const.5.I32Const.6.I32Const.7.I32Const.8.I32Const.9.I32Const.10.I32Const.11.I32Const.12.I32Const.13.I32Const.14.I32Const.15.I32Const.16.I32Const.17.Call.0.Return.End.End.foo):
(i.instance.exports.f0.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.F32Const.Call.Return.End.End.foo):
(i.instance.exports.f0):
(instance.exports.f0.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.GetLocal.Call.Return.End.End.foo):
(instance.exports.f0):
Source/JavaScriptCore:
This patch fixes three bugs:
- The WasmBinding code for making a JS call was off
by 1 in its stack layout code.
- The WasmBinding code had a "<" comparison instead
of a ">=" comparison. This would cause us to calculate
the wrong frame pointer offset.
- The code to reload wasm state inside B3IRGenerator didn't
properly represent its effects.
- wasm/WasmB3IRGenerator.cpp:
(JSC::Wasm::restoreWebAssemblyGlobalState):
(JSC::Wasm::parseAndCompile):
- wasm/WasmBinding.cpp:
(JSC::Wasm::wasmToJs):
- wasm/js/WebAssemblyInstanceConstructor.cpp:
(JSC::WebAssemblyInstanceConstructor::createInstance):
- 11:08 AM Changeset in webkit [213652] by
-
- 47 edits in trunk/Source
Make the VM Traps mechanism non-polling for the DFG and FTL.
https://bugs.webkit.org/show_bug.cgi?id=168920
<rdar://problem/30738588>
Reviewed by Filip Pizlo.
Source/JavaScriptCore:
- Added a ENABLE(SIGNAL_BASED_VM_TRAPS) configuration in Platform.h. This is currently only enabled for OS(DARWIN) and ENABLE(JIT).
- Added assembler functions for overwriting an instruction with a breakpoint.
- Added a new JettisonDueToVMTraps jettison reason.
- Added CodeBlock and DFG::CommonData utility functions for over-writing invalidation points with breakpoint instructions.
- The BytecodeGenerator now emits the op_check_traps bytecode unconditionally.
- Remove the JSC_alwaysCheckTraps option because of (4) above. For ports that don't ENABLE(SIGNAL_BASED_VM_TRAPS), we'll force Options::usePollingTraps() to always be true. This makes the VMTraps implementation fall back to using polling based traps only.
- Make VMTraps support signal based traps.
Some design and implementation details of signal based VM traps:
- The implementation makes use of 2 signal handlers for SIGUSR1 and SIGTRAP.
- VMTraps::fireTrap() will set the flag for the requested trap and instantiate a SignalSender. The SignalSender will send SIGUSR1 to the mutator thread that we want to trap, and check for the occurence of one of the following events:
- VMTraps::handleTraps() has been called for the requested trap, or
- the VM is inactive and is no longer executing any JS code. We determine this to be the case if the thread no longer owns the JSLock and the VM's entryScope is null.
Note: the thread can relinquish the JSLock while the VM's entryScope is not
null. This happens when the thread calls JSLock::dropAllLocks() before
calling a host function that may block on IO (or whatever). For our purpose,
this counts as the VM still running JS code, and VM::fireTrap() will still
be waiting.
If the SignalSender does not see either of these events, it will sleep for a
while and then re-send SIGUSR1 and check for the events again. When it sees
one of these events, it will consider the mutator to have received the trap
request.
- The SIGUSR1 handler will try to insert breakpoints at the invalidation points in the DFG/FTL codeBlock at the top of the stack. This allows the mutator thread to break (with a SIGTRAP) exactly at an invalidation point, where it's safe to jettison the codeBlock.
Note: we cannot have the requester thread (that called VMTraps::fireTrap())
insert the breakpoint instructions itself. This is because we need the
register state of the the mutator thread (that we want to trap in) in order to
find the codeBlocks that we wish to insert the breakpoints in. Currently,
we don't have a generic way for the requester thread to get the register state
of another thread.
- The SIGTRAP handler will check to see if it is trapping on a breakpoint at an invalidation point. If so, it will jettison the codeBlock and adjust the PC to re-execute the invalidation OSR exit off-ramp. After the OSR exit, the baseline JIT code will eventually reach an op_check_traps and call VMTraps::handleTraps().
If the handler is not trapping at an invalidation point, then it must be
observing an assertion failure (which also uses the breakpoint instruction).
In this case, the handler will defer to the default SIGTRAP handler and crash.
- The reason we need the SignalSender is because SignalSender::send() is called from another thread in a loop, so that VMTraps::fireTrap() can return sooner. send() needs to make use of the VM pointer, and it is not guaranteed that the VM will outlive the thread. SignalSender provides the mechanism by which we can nullify the VM pointer when the VM dies so that the thread does not continue to use it.
- assembler/ARM64Assembler.h:
(JSC::ARM64Assembler::replaceWithBrk):
- assembler/ARMAssembler.h:
(JSC::ARMAssembler::replaceWithBrk):
- assembler/ARMv7Assembler.h:
(JSC::ARMv7Assembler::replaceWithBkpt):
- assembler/MIPSAssembler.h:
(JSC::MIPSAssembler::replaceWithBkpt):
- assembler/MacroAssemblerARM.h:
(JSC::MacroAssemblerARM::replaceWithJump):
- assembler/MacroAssemblerARM64.h:
(JSC::MacroAssemblerARM64::replaceWithBreakpoint):
- assembler/MacroAssemblerARMv7.h:
(JSC::MacroAssemblerARMv7::replaceWithBreakpoint):
- assembler/MacroAssemblerMIPS.h:
(JSC::MacroAssemblerMIPS::replaceWithJump):
- assembler/MacroAssemblerX86Common.h:
(JSC::MacroAssemblerX86Common::replaceWithBreakpoint):
- assembler/X86Assembler.h:
(JSC::X86Assembler::replaceWithInt3):
- bytecode/CodeBlock.cpp:
(JSC::CodeBlock::jettison):
(JSC::CodeBlock::hasInstalledVMTrapBreakpoints):
(JSC::CodeBlock::installVMTrapBreakpoints):
- bytecode/CodeBlock.h:
- bytecompiler/BytecodeGenerator.cpp:
(JSC::BytecodeGenerator::emitCheckTraps):
- dfg/DFGCommonData.cpp:
(JSC::DFG::CommonData::installVMTrapBreakpoints):
(JSC::DFG::CommonData::isVMTrapBreakpoint):
- dfg/DFGCommonData.h:
(JSC::DFG::CommonData::hasInstalledVMTrapsBreakpoints):
- dfg/DFGJumpReplacement.cpp:
(JSC::DFG::JumpReplacement::installVMTrapBreakpoint):
- dfg/DFGJumpReplacement.h:
(JSC::DFG::JumpReplacement::dataLocation):
- dfg/DFGNodeType.h:
- heap/CodeBlockSet.cpp:
(JSC::CodeBlockSet::contains):
- heap/CodeBlockSet.h:
- heap/CodeBlockSetInlines.h:
(JSC::CodeBlockSet::iterate):
- heap/Heap.cpp:
(JSC::Heap::forEachCodeBlockIgnoringJITPlansImpl):
- heap/Heap.h:
- heap/HeapInlines.h:
(JSC::Heap::forEachCodeBlockIgnoringJITPlans):
- heap/MachineStackMarker.h:
(JSC::MachineThreads::threadsListHead):
- jit/ExecutableAllocator.cpp:
(JSC::ExecutableAllocator::isValidExecutableMemory):
- jit/ExecutableAllocator.h:
- profiler/ProfilerJettisonReason.cpp:
(WTF::printInternal):
- profiler/ProfilerJettisonReason.h:
- runtime/JSLock.cpp:
(JSC::JSLock::didAcquireLock):
- runtime/Options.cpp:
(JSC::overrideDefaults):
- runtime/Options.h:
- runtime/PlatformThread.h:
(JSC::platformThreadSignal):
- runtime/VM.cpp:
(JSC::VM::~VM):
(JSC::VM::ensureWatchdog):
(JSC::VM::handleTraps): Deleted.
(JSC::VM::setNeedAsynchronousTerminationSupport): Deleted.
- runtime/VM.h:
(JSC::VM::ownerThread):
(JSC::VM::traps):
(JSC::VM::handleTraps):
(JSC::VM::needTrapHandling):
(JSC::VM::needAsynchronousTerminationSupport): Deleted.
- runtime/VMTraps.cpp:
(JSC::VMTraps::vm):
(JSC::SignalContext::SignalContext):
(JSC::SignalContext::adjustPCToPointToTrappingInstruction):
(JSC::vmIsInactive):
(JSC::findActiveVMAndStackBounds):
(JSC::handleSigusr1):
(JSC::handleSigtrap):
(JSC::installSignalHandlers):
(JSC::sanitizedTopCallFrame):
(JSC::isSaneFrame):
(JSC::VMTraps::tryInstallTrapBreakpoints):
(JSC::VMTraps::invalidateCodeBlocksOnStack):
(JSC::VMTraps::VMTraps):
(JSC::VMTraps::willDestroyVM):
(JSC::VMTraps::addSignalSender):
(JSC::VMTraps::removeSignalSender):
(JSC::VMTraps::SignalSender::willDestroyVM):
(JSC::VMTraps::SignalSender::send):
(JSC::VMTraps::fireTrap):
(JSC::VMTraps::handleTraps):
- runtime/VMTraps.h:
(JSC::VMTraps::~VMTraps):
(JSC::VMTraps::needTrapHandling):
(JSC::VMTraps::notifyGrabAllLocks):
(JSC::VMTraps::SignalSender::SignalSender):
(JSC::VMTraps::invalidateCodeBlocksOnStack):
- tools/VMInspector.cpp:
- tools/VMInspector.h:
(JSC::VMInspector::getLock):
(JSC::VMInspector::iterate):
Source/WebCore:
No new tests needed. This is covered by existing tests.
- bindings/js/WorkerScriptController.cpp:
(WebCore::WorkerScriptController::WorkerScriptController):
(WebCore::WorkerScriptController::scheduleExecutionTermination):
Source/WTF:
Make StackBounds more useful for checking if a pointer is within stack bounds.
- wtf/MetaAllocator.cpp:
(WTF::MetaAllocator::isInAllocatedMemory):
- wtf/MetaAllocator.h:
- wtf/Platform.h:
- wtf/StackBounds.h:
(WTF::StackBounds::emptyBounds):
(WTF::StackBounds::StackBounds):
(WTF::StackBounds::isEmpty):
(WTF::StackBounds::contains):
- 11:08 AM Changeset in webkit [213651] by
-
- 2 edits in trunk/Tools
webkitpy: Add unit test to prevent breakage of EWS
https://bugs.webkit.org/show_bug.cgi?id=169385
Reviewed by Alexey Proskuryakov.
Prevent <http://trac.webkit.org/changeset/213545> from happening in the future.
- Scripts/webkitpy/tool/commands/earlywarningsystem_unittest.py:
(test_ews_name): Added.
- 11:02 AM Changeset in webkit [213650] by
-
- 8 edits9 copies in trunk
WebGPU: Backend - Library and Functions
https://bugs.webkit.org/show_bug.cgi?id=169371
<rdar://problem/30928792>
Reviewed by Tim Horton.
Source/WebCore:
Implement the backends for GPULibrary and GPUFunction.
A GPULibrary is created from a GPUDevice, given some
shader source code. You can then extract the GPUFunctions
from the library.
WebKitAPITests: GPULibrary and GPUFunction.
- WebCore.xcodeproj/project.pbxproj:
- platform/graphics/cocoa/GPUFunctionMetal.mm:
(WebCore::GPUFunction::GPUFunction):
(WebCore::GPUFunction::name):
(WebCore::GPUFunction::platformFunction):
- platform/graphics/cocoa/GPULibraryMetal.mm:
(WebCore::GPULibrary::GPULibrary):
(WebCore::GPULibrary::label):
(WebCore::GPULibrary::setLabel):
(WebCore::GPULibrary::functionNames):
(WebCore::GPULibrary::platformLibrary):
- platform/graphics/gpu/GPUDevice.cpp:
(WebCore::GPUDevice::createLibrary):
- platform/graphics/gpu/GPUDevice.h:
- platform/graphics/gpu/GPUFunction.cpp:
(WebCore::GPUFunction::create):
(WebCore::GPUFunction::~GPUFunction):
(WebCore::GPUFunction::GPUFunction):
(WebCore::GPUFunction::name):
- platform/graphics/gpu/GPUFunction.h:
- platform/graphics/gpu/GPULibrary.cpp:
(WebCore::GPULibrary::create):
(WebCore::GPULibrary::~GPULibrary):
(WebCore::GPULibrary::label):
(WebCore::GPULibrary::setLabel):
(WebCore::GPULibrary::functionNames):
(WebCore::GPULibrary::functionWithName):
- platform/graphics/gpu/GPULibrary.h:
Tools:
Add tests for GPULibrary and GPUFunction.
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
- TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm:
(TestWebKitAPI::TEST_F):
(TestWebKitAPI::TEST): Deleted.
- TestWebKitAPI/Tests/WebCore/mac/GPUFunction.mm: Copied from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm.
(TestWebKitAPI::TEST_F):
- TestWebKitAPI/Tests/WebCore/mac/GPULibrary.mm: Copied from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm.
(TestWebKitAPI::TEST_F):
- TestWebKitAPI/Tests/WebCore/mac/GPUTest.h: Copied from Tools/TestWebKitAPI/Tests/WebCore/mac/GPUDevice.mm.
(TestWebKitAPI::GPU::librarySourceCode):
- 10:54 AM Changeset in webkit [213649] by
-
- 2 edits in trunk/Source/WebKit2
[WK2] Prevent app nap for the active tab
https://bugs.webkit.org/show_bug.cgi?id=169417
<rdar://problem/30876965>
Reviewed by Gavin Barraclough.
Prevent app nap for the active tab to minimize the risks of breakage.
From the AppNap documentation [1], the app is only a candidate for app nap if it not the
foreground app. However, in the Safari case, the foreground app is Safari (UIProcess) and
the same policy does not apply to our foreground WebContent process unless we explicitly
make it so. This patch addresses the issue by taking a UserActivity when the tab is
active to prevent app nap.
[1] https://developer.apple.com/library/content/documentation/Performance/Conceptual/power_efficiency_guidelines_osx/AppNap.html
- WebProcess/WebPage/WebPage.cpp:
(WebKit::WebPage::updateThrottleState):
- 10:53 AM Changeset in webkit [213648] by
-
- 2 edits in trunk/Source/JavaScriptCore
WebKit: JSC: JSObject::ensureLength doesn't check if ensureLengthSlow failed
https://bugs.webkit.org/show_bug.cgi?id=169215
Reviewed by Mark Lam.
This doesn't have a test because it would be a very complicated test.
- runtime/JSObject.h:
(JSC::JSObject::ensureLength): If ensureLengthSlow returns false, we need to return false.
- 10:33 AM Changeset in webkit [213647] by
-
- 3 edits in trunk/Source/WebKit2
Add new delegate methods
https://bugs.webkit.org/show_bug.cgi?id=169394
Part of rdar://problem/28880714.
Reviewed by Tim Horton.
- UIProcess/ApplePay/cocoa/WebPaymentCoordinatorProxyCocoa.h:
- UIProcess/ApplePay/cocoa/WebPaymentCoordinatorProxyCocoa.mm:
(-[WKPaymentAuthorizationViewControllerDelegate invalidate]):
(toShippingMethod):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didAuthorizePayment:handler:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didSelectPaymentMethod:handler:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didSelectShippingMethod:handler:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didSelectShippingContact:handler:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didAuthorizePayment:completion:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didSelectShippingMethod:completion:]):
(-[WKPaymentAuthorizationViewControllerDelegate paymentAuthorizationViewController:didSelectShippingContact:completion:]):
(WebKit::toPKPaymentAuthorizationStatus):
(WebKit::toPKPaymentErrorCode):
(WebKit::toNSError):
(WebKit::toNSErrors):
(WebKit::WebPaymentCoordinatorProxy::platformCompletePaymentSession):
(WebKit::WebPaymentCoordinatorProxy::platformCompleteShippingMethodSelection):
(WebKit::WebPaymentCoordinatorProxy::platformCompleteShippingContactSelection):
(WebKit::WebPaymentCoordinatorProxy::platformCompletePaymentMethodSelection):
- 9:59 AM Changeset in webkit [213646] by
-
- 7 edits2 adds in trunk
Align Document.elementFromPoint() with the CSSOM specification
https://bugs.webkit.org/show_bug.cgi?id=169403
Reviewed by Sam Weinig.
Source/WebCore:
Align Document.elementFromPoint() with the CSSOM specification:
In particular, the parameters should be mandatory and of type double.
The parameters are mandatory in both Firefox and Chrome already. Parameters
are finite floating point in Firefox and integers in Chrome.
Test: fast/dom/elementFromPoint-parameters.html
- dom/DocumentOrShadowRoot.idl:
- dom/TreeScope.cpp:
(WebCore::TreeScope::elementFromPoint):
- dom/TreeScope.h:
LayoutTests:
Add layout test coverage.
- fast/dom/elementFromPoint-parameters-expected.txt: Added.
- fast/dom/elementFromPoint-parameters.html: Added.
- fast/dom/non-numeric-values-numeric-parameters-expected.txt:
- fast/dom/script-tests/non-numeric-values-numeric-parameters.js:
- 9:40 AM Changeset in webkit [213645] by
-
- 19 edits1 delete in trunk
WTF should make it super easy to do ARM concurrency tricks
https://bugs.webkit.org/show_bug.cgi?id=169300
Reviewed by Mark Lam.
Source/JavaScriptCore:
This changes a bunch of GC hot paths to use new concurrency APIs that lead to optimal
code on both x86 (fully leverage TSO, transactions become CAS loops) and ARM (use
dependency chains for fencing, transactions become LL/SC loops). While inspecting the
machine code, I found other opportunities for improvement, like inlining the "am I
marked" part of the marking functions.
- heap/Heap.cpp:
(JSC::Heap::setGCDidJIT):
- heap/HeapInlines.h:
(JSC::Heap::testAndSetMarked):
- heap/LargeAllocation.h:
(JSC::LargeAllocation::isMarked):
(JSC::LargeAllocation::isMarkedConcurrently):
(JSC::LargeAllocation::aboutToMark):
(JSC::LargeAllocation::testAndSetMarked):
- heap/MarkedBlock.h:
(JSC::MarkedBlock::areMarksStaleWithDependency):
(JSC::MarkedBlock::aboutToMark):
(JSC::MarkedBlock::isMarkedConcurrently):
(JSC::MarkedBlock::isMarked):
(JSC::MarkedBlock::testAndSetMarked):
- heap/SlotVisitor.cpp:
(JSC::SlotVisitor::appendSlow):
(JSC::SlotVisitor::appendHiddenSlow):
(JSC::SlotVisitor::appendHiddenSlowImpl):
(JSC::SlotVisitor::setMarkedAndAppendToMarkStack):
(JSC::SlotVisitor::appendUnbarriered): Deleted.
(JSC::SlotVisitor::appendHidden): Deleted.
- heap/SlotVisitor.h:
- heap/SlotVisitorInlines.h:
(JSC::SlotVisitor::appendUnbarriered):
(JSC::SlotVisitor::appendHidden):
(JSC::SlotVisitor::append):
(JSC::SlotVisitor::appendValues):
(JSC::SlotVisitor::appendValuesHidden):
- runtime/CustomGetterSetter.cpp:
- runtime/JSObject.cpp:
(JSC::JSObject::visitButterflyImpl):
- runtime/JSObject.h:
Source/WTF:
This adds Atomic<>::loadLink and Atomic<>::storeCond, available only when HAVE(LL_SC).
It abstracts loadLink/storeCond behind prepare/attempt. You can write prepare/attempt
loops whenever your loop fits into the least common denominator of LL/SC and CAS.
This modifies Atomic<>::transaction to use prepare/attempt. So, if you write your loop
using Atomic<>::transaction, then you get LL/SC for free.
Depending on the kind of transaction you are doing, you may not want to perform an LL
until you have a chance to just load the current value. Atomic<>::transaction() assumes
that you do not care to have any ordering guarantees in that case. If you think that
the transaction has a good chance of aborting this way, you want
Atomic<>::transaction() to first do a plain load. But if you don't think that such an
abort is likely, then you want to go straight to the LL. The API supports this concept
via TransactionAbortLikelihood.
Additionally, this redoes the depend/consume API to be dead simple. Dependency is
unsigned. You get a dependency on a loaded value by just saying
dependency(loadedValue). You consume the dependency by using it as a bonus index to
some pointer dereference. This is made easy with the consume<T*>(ptr, dependency)
helper. In those cases where you want to pass around both a computed value and a
dependency, there's DependencyWith<T>. But you won't need it in most cases. The loaded
value or any value computed from the loaded value is a fine input to dependency()!
This change updates a bunch of hot paths to use the new APIs. Using transaction() gives
us optimal LL/SC loops for object marking and lock acquisition.
This change also updates a bunch of hot paths to use dependency()/consume().
This is a significant Octane/splay speed-up on ARM.
- wtf/Atomics.h:
(WTF::hasFence):
(WTF::Atomic::prepare):
(WTF::Atomic::attempt):
(WTF::Atomic::transaction):
(WTF::Atomic::transactionRelaxed):
(WTF::nullDependency):
(WTF::dependency):
(WTF::DependencyWith::DependencyWith):
(WTF::dependencyWith):
(WTF::consume):
(WTF::Atomic::tryTransactionRelaxed): Deleted.
(WTF::Atomic::tryTransaction): Deleted.
(WTF::zeroWithConsumeDependency): Deleted.
(WTF::consumeLoad): Deleted.
- wtf/Bitmap.h:
(WTF::WordType>::get):
(WTF::WordType>::concurrentTestAndSet):
(WTF::WordType>::concurrentTestAndClear):
- wtf/LockAlgorithm.h:
(WTF::LockAlgorithm::lockFast):
(WTF::LockAlgorithm::unlockFast):
(WTF::LockAlgorithm::unlockSlow):
- wtf/Platform.h:
Tools:
This vastly simplifies the consume API. The new API is thoroughly tested by being used
in the GC's guts. I think that unit tests are a pain to maintain, so we shouldn't have
them unless we are legitimately worried about coverage. We're not in this case.
- TestWebKitAPI/CMakeLists.txt:
- TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
- TestWebKitAPI/Tests/WTF/Consume.cpp: Removed.
- 9:40 AM Changeset in webkit [213644] by
-
- 2 edits in trunk/Source/WebInspectorUI
Web Inspector: DOMTreeManager dispatches DocumentUpdated twice when the document is cleared
https://bugs.webkit.org/show_bug.cgi?id=168984
Reviewed by Timothy Hatcher.
- UserInterface/Controllers/DOMTreeManager.js:
(WebInspector.DOMTreeManager.prototype._setDocument):
Compare old and new document values before dispatching message.
- 8:26 AM Changeset in webkit [213643] by
-
- 3 edits in trunk/Source/WebKit2
WKWebView should automatically respect insets from the view controller hierarchy
https://bugs.webkit.org/show_bug.cgi?id=169398
<rdar://problem/30617593>
Reviewed by Dean Jackson.
- Platform/spi/ios/UIKitSPI.h:
- UIProcess/API/Cocoa/WKWebView.mm:
(-[WKWebView _initializeWithConfiguration:]):
(-[WKWebView _computedContentInset]):
(activeMinimumLayoutSize):
(-[WKWebView _frameOrBoundsChanged]):
Factor insets from above us in the view controller hierarchy into our
content insets (and layout size) automatically.
- 8:17 AM Changeset in webkit [213642] by
-
- 3 edits in trunk/LayoutTests
[mac-wk1] LayoutTest media/modern-media-controls/airplay-button/airplay-button.html is a flaky timeout
https://bugs.webkit.org/show_bug.cgi?id=168409
Patch by Antoine Quint <Antoine Quint> on 2017-03-09
Reviewed by Eric Carlson.
Adding yet more logging to try to determine the cause of this test's flakiness.
- media/modern-media-controls/airplay-button/airplay-button-expected.txt:
- media/modern-media-controls/airplay-button/airplay-button.html:
- 8:10 AM Changeset in webkit [213641] by
-
- 10 edits in trunk
[Modern Media Controls] Fullscreen controls are difficult to drag around
https://bugs.webkit.org/show_bug.cgi?id=169412
<rdar://problem/30931892>
Patch by Antoine Quint <Antoine Quint> on 2017-03-09
Reviewed by Eric Carlson.
Source/WebCore:
Up until now, we would simply check that the event target when attempting to
initiate a drag would be the controls bar and not any of its children. This
was not so great as this would encompass also a large margin around the left,
center and right button containers as well as the time labels. We now check
that the event target is anything but an interactive element (<button> or <input>).
The existing test at media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-drag.html
has been updated to initiate dragging right between the rewind and play/pause button,
which would fail prior to this patch since this would have hit-tested within the center
button container and thus prevent dragging.
- Modules/modern-media-controls/controls/macos-fullscreen-media-controls.css:
(.media-controls.mac.fullscreen button.volume-down):
(.media-controls.mac.fullscreen button.volume-up):
(.media-controls.mac.fullscreen button.rewind):
(.media-controls.mac.fullscreen button.play-pause):
(.media-controls.mac.fullscreen button.forward):
(.media-controls.mac.fullscreen .buttons-container.right button):
- Modules/modern-media-controls/controls/macos-fullscreen-media-controls.js:
(MacOSFullscreenMediaControls.prototype._handleMousedown):
- Modules/modern-media-controls/controls/macos-inline-media-controls.css:
(.media-controls.mac.inline > .controls-bar button):
- Modules/modern-media-controls/controls/macos-media-controls.css:
(.media-controls.mac > .controls-bar button):
LayoutTests:
- media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-buttons-styles-expected.txt:
- media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-buttons-styles.html:
Rebaseline test to account for CSS changes.
- media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-drag-is-prevented-over-button.html:
Change drag start coordinate to be right over a button so that it is prevented. Starting with this patch,
a drag at this coordinate would have been tolerated as being part of the left buttons container but not
part of a button in that container.
- media/modern-media-controls/macos-fullscreen-media-controls/macos-fullscreen-media-controls-drag.html:
Change drag start coordinate to be right between two buttons so that it would have failed prior to this
patch.
- 7:59 AM Changeset in webkit [213640] by
-
- 4 edits4 adds in trunk
[Modern Media Controls] Controls are not visible in fullscreen without the controls attribute
https://bugs.webkit.org/show_bug.cgi?id=169411
<rdar://problem/30931724>
Patch by Antoine Quint <Antoine Quint> on 2017-03-09
Reviewed by Eric Carlson.
Source/WebCore:
Ensure we show controls in fullscreen in the absence of the controls attribute.
Tests: media/modern-media-controls/controls-visibility-support/controls-visibility-support-fullscreen-on-parent-element.html
media/modern-media-controls/controls-visibility-support/controls-visibility-support-fullscreen-on-video.html
- Modules/modern-media-controls/media/controls-visibility-support.js:
(ControlsVisibilitySupport.prototype.get mediaEvents):
(ControlsVisibilitySupport.prototype._updateControls):
(ControlsVisibilitySupport):
LayoutTests:
Add two new tests to check that controls are visible in fullscreen in the absence of the controls attribute,
except if the fullscreen element is a parent element.
- media/modern-media-controls/controls-visibility-support/controls-visibility-support-fullscreen-on-parent-element.html: Added.
- media/modern-media-controls/controls-visibility-support/controls-visibility-support-fullscreen-on-video.html: Added.
- platform/ios-simulator/TestExpectations: Skip new tests on iOS since it uses platform controls.
- 7:25 AM Changeset in webkit [213639] by
-
- 9 edits in trunk
Simple line layout: Add top level pagination support.
https://bugs.webkit.org/show_bug.cgi?id=169410
<rdar://problem/28536227>
Reviewed by Antti Koivisto.
Source/WebCore:
This patch enables top level pagination for simple line layout.
Covered by existing tests.
- rendering/RenderMultiColumnFlowThread.h:
- rendering/RenderView.cpp:
(WebCore::RenderView::pushLayoutStateForPagination): Pagination flag needs to be set before calling pushLayoutStateForCurrentFlowThread.
- rendering/SimpleLineLayout.cpp:
(WebCore::SimpleLineLayout::canUseForWithReason):
- rendering/SimpleLineLayoutCoverage.cpp:
(WebCore::SimpleLineLayout::printReason):
- rendering/SimpleLineLayoutCoverage.h:
- rendering/SimpleLineLayoutFunctions.cpp:
(WebCore::SimpleLineLayout::paintFlow):
- rendering/SimpleLineLayoutPagination.cpp:
(WebCore::SimpleLineLayout::computeLineTopAndBottomWithOverflow): Matching normal line layout values.
(WebCore::SimpleLineLayout::setPageBreakForLine): Logic here matches RenderBlockFlow::adjustLinePositionForPagination
LayoutTests:
- 6:24 AM Changeset in webkit [213638] by
-
- 2 edits in trunk/Source/WebKit2
Correctly process the return value of gdk_drag_context_get_selected_action()
Reviewed by Carlos Garcia Campos.
It returns the action itself and not the bitmask.
- UIProcess/gtk/DragAndDropHandler.cpp:
(WebKit::DragAndDropHandler::drop):
- 5:42 AM Changeset in webkit [213637] by
-
- 2 edits in trunk/Source/WebKit2
[GTK] Use GtkFileChooserNative for open/save dialogs
https://bugs.webkit.org/show_bug.cgi?id=169278
Patch by Adrian Perez de Castro <Adrian Perez de Castro> on 2017-03-09
Reviewed by Carlos Garcia Campos.
When building with GTK+ 3.20.0 or newer, use GtkFileChooserNative instead of GtkFileChooserDialog to allow for
better integration with the runtime environment and use native open/save dialogs. This also makes WebKitGTK+ use
portal-based load/save dialogs when running inside the Flatpak sandbox.
The existing GtkFileChooserDialog implementation will still be used when building against older versions of
GTK+, so this change does not require a new version of the toolkit.
- UIProcess/API/gtk/WebKitWebView.cpp:
(fileChooserDialogResponseCallback):
(webkitWebViewRunFileChooser):
- 5:10 AM Changeset in webkit [213636] by
-
- 2 edits in trunk/Source/WebCore
[GStreamer][MSE] Fix 'comparison between signed and unsigned integer
expressions' warning
Reviewed by Žan Doberšek.
- platform/graphics/gstreamer/mse/PlaybackPipeline.cpp:
(WebCore::PlaybackPipeline::flush):
- 2:51 AM Changeset in webkit [213635] by
-
- 6 edits in trunk/Source/WebInspectorUI
Web Inspector: Add settings for controlling Styles sidebar intelligence
https://bugs.webkit.org/show_bug.cgi?id=169181
Patch by Devin Rousso <Devin Rousso> on 2017-03-09
Reviewed by Timothy Hatcher.
- Localizations/en.lproj/localizedStrings.js:
- UserInterface/Base/Setting.js:
Add three new settings:
- stylesShowInlineWarnings
- stylesInsertNewline
- stylesSelectOnFirstClick
- UserInterface/Views/CSSStyleDeclarationTextEditor.js:
(WebInspector.CSSStyleDeclarationTextEditor):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleMouseDown):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._handleMouseUp):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createInlineSwatches.update):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._createTextMarkerForPropertyIfNeeded):
- UserInterface/Views/SettingsTabContentView.css:
(.content-view.settings > .separator): Added.
(.content-view.settings > .setting-container):
(.content-view.settings > .setting-container.combined):
- UserInterface/Views/SettingsTabContentView.js:
(WebInspector.SettingsTabContentView.prototype.layout.createContainer):
(WebInspector.SettingsTabContentView.prototype.layout.createSeparator): Added.
(WebInspector.SettingsTabContentView.prototype.layout):
- 1:51 AM Changeset in webkit [213634] by
-
- 12 edits4 adds in trunk
Implement stroke-width CSS property.
https://bugs.webkit.org/show_bug.cgi?id=169273
Reviewed by Simon Fraser.
Source/WebCore:
Support setting text stroke width using the CSS property stroke-width, see https://drafts.fxtf.org/paint/.
Text stroke width can currently be set with the -webkit-text-stroke-width property. To make sure this still
works, I added a check to determine if the stroke-width property has been explicitly set. If it has not been
set, we fall back to the value of the -webkit-text-stroke-width property.
Tests: fast/css/stroke-width-percent.html
fast/css/stroke-width.html
- css/CSSProperties.json:
- css/StyleBuilderCustom.h:
(WebCore::StyleBuilderCustom::applyValueStrokeWidth):
(WebCore::StyleBuilderCustom::applyValueWebkitTextStrokeWidth):
- rendering/InlineFlowBox.cpp:
(WebCore::InlineFlowBox::addToLine):
(WebCore::InlineFlowBox::addTextBoxVisualOverflow):
- rendering/SimpleLineLayoutFunctions.cpp:
(WebCore::SimpleLineLayout::computeOverflow):
- rendering/TextDecorationPainter.cpp:
(WebCore::decorationColor):
- rendering/TextPaintStyle.cpp:
(WebCore::computeTextPaintStyle):
(WebCore::computeTextSelectionPaintStyle):
- rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::RenderStyle):
(WebCore::RenderStyle::computedTextStrokeWidth):
- rendering/style/RenderStyle.h:
(WebCore::RenderStyle::setHasExplicitlySetStrokeWidth):
(WebCore::RenderStyle::hasExplicitlySetStrokeWidth):
(WebCore::RenderStyle::setHasExplicitlySetWebKitTextStrokeWidth):
(WebCore::RenderStyle::hasExplicitlySetWebKitTextStrokeWidth):
(WebCore::RenderStyle::InheritedFlags::operator==):
LayoutTests:
- fast/css/stroke-width-expected.html: Added.
- fast/css/stroke-width-percent-expected.html: Added.
- fast/css/stroke-width-percent.html: Added.
- fast/css/stroke-width.html: Added.
- 1:07 AM Changeset in webkit [213633] by
-
- 15 edits3 adds in trunk
Source/WebCore:
Allow the page to render before <link> stylesheet tags in body
https://bugs.webkit.org/show_bug.cgi?id=149157
<rdar://problem/24658830>
Reviewed by Simon Fraser.
Currently we block style and renderer building completely if document has any loading
stylesheets. In case a script queries something layout dependent we construct the render
tree with whatever style we have but block painting in it.
This patch changes behavior so that a loading stylesheet in body only blocks rendering for elements
that are after it. The expectation is that such stylesheets rarely affect elements before them
and the elements can be rendered without causing ugly visible styling changes.
The patch replaces the old flash-of-unstyled-content (FOUC) preventation mechanism with a more
fine-grained one. Paint blocking is now done on per-renderer basis with based on isNonFinal flag in
RenderStyle.
For stylesheets in head the behavior should be largely unchanged.
Test: http/tests/incremental/stylesheet-body-incremental-rendering.html
- css/StyleResolver.cpp:
(WebCore::StyleResolver::pseudoStyleRulesForElement):
- dom/Document.cpp:
(WebCore::Document::Document):
(WebCore::Document::resolveStyle):
(WebCore::Document::updateLayoutIgnorePendingStylesheets):
Remove the old FOUC preventation state tracking.
(WebCore::Document::shouldScheduleLayout):
(WebCore::Document::didRemoveAllPendingStylesheet):
Repaints will now get triggered by the normal style mechanism.
- dom/Document.h:
(WebCore::Document::hasNodesWithNonFinalStyle):
(WebCore::Document::setHasNodesWithNonFinalStyle):
Track if we need to recompute the style later because non-final or unstyled elements.
(WebCore::Document::didLayoutWithPendingStylesheets): Deleted.
(WebCore::Document::hasNodesWithPlaceholderStyle): Deleted.
(WebCore::Document::setHasNodesWithPlaceholderStyle): Deleted.
- html/HTMLFrameSetElement.cpp:
(WebCore::HTMLFrameSetElement::rendererIsNeeded):
- page/FrameView.cpp:
(WebCore::FrameView::qualifiesAsVisuallyNonEmpty):
Don't qualify as visually non-empty if we have loading stylesheets in head (even if there is
a fouc-prevented render tree).
(WebCore::FrameView::fireLayoutRelatedMilestonesIfNeeded):
- rendering/RenderBlock.cpp:
(WebCore::RenderBlock::paintContents):
Instead of a global test, block painting if isNonFinal is set in the renderer's style.
- rendering/RenderLayer.cpp:
(WebCore::shouldSuppressPaintingLayer):
- rendering/style/RenderStyle.cpp:
(WebCore::RenderStyle::changeRequiresRepaint):
The isNonFinal flag prevents painting so we need to trigger repaint when it gets cleared.
- rendering/style/RenderStyle.h:
(WebCore::RenderStyle::isNotFinal):
(WebCore::RenderStyle::setIsNotFinal):
(WebCore::RenderStyle::isPlaceholderStyle): Deleted.
(WebCore::RenderStyle::setIsPlaceholderStyle): Deleted.
There is no need for placeholder styles anymore. Reuse the bit for isNotFinal.
- rendering/style/StyleRareNonInheritedData.cpp:
(WebCore::StyleRareNonInheritedData::StyleRareNonInheritedData):
(WebCore::StyleRareNonInheritedData::operator==):
- rendering/style/StyleRareNonInheritedData.h:
- style/StyleScope.cpp:
(WebCore::Style::Scope::analyzeStyleSheetChange):
(WebCore::Style::Scope::updateActiveStyleSheets):
- style/StyleTreeResolver.cpp:
(WebCore::Style::TreeResolver::styleForElement):
(WebCore::Style::TreeResolver::resolveElement):
If we have seens a loading stylesheet and don't have a renderer yet don't style the element.
In case there is a renderer or we are ignoring pending sheets, resolve the style normally
but mark it as non-final.
(WebCore::Style::makePlaceholderStyle): Deleted.
LayoutTests:
Loading in-body stylesheets should not block rendering of elements before them
https://bugs.webkit.org/show_bug.cgi?id=169345
Reviewed by Simon Fraser.
- http/tests/incremental/resources/delayed-css.php: Added.
- http/tests/incremental/stylesheet-body-incremental-rendering-expected.html: Added.
- http/tests/incremental/stylesheet-body-incremental-rendering.html: Added.