[webkit-changes] [109870] trunk/Source/WebCore
Title: [109870] trunk/Source/WebCore Revision 109870 Author ph...@webkit.org Date 2012-03-05 23:59:24 -0800 (Mon, 05 Mar 2012) Log Message WebAudio JSC-related fixes https://bugs.webkit.org/show_bug.cgi?id=80299 Reviewed by Adam Barth. * bindings/js/JSAudioContextCustom.cpp: (WebCore::JSAudioContextConstructor::constructJSAudioContext): Wrap the audio context in a DOM wrapper. * webaudio/AudioContext.idl: No need to protect EventTarget by V8 guards, I think. * webaudio/_javascript_AudioNode.idl: Ditto. Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/bindings/js/JSAudioContextCustom.cpp trunk/Source/WebCore/webaudio/AudioContext.idl trunk/Source/WebCore/webaudio/_javascript_AudioNode.idl Diff Modified: trunk/Source/WebCore/ChangeLog (109869 => 109870) --- trunk/Source/WebCore/ChangeLog 2012-03-06 07:40:46 UTC (rev 109869) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 07:59:24 UTC (rev 109870) @@ -1,3 +1,17 @@ +2012-03-05 Philippe Normand + +WebAudio JSC-related fixes +https://bugs.webkit.org/show_bug.cgi?id=80299 + +Reviewed by Adam Barth. + +* bindings/js/JSAudioContextCustom.cpp: +(WebCore::JSAudioContextConstructor::constructJSAudioContext): +Wrap the audio context in a DOM wrapper. +* webaudio/AudioContext.idl: No need to protect EventTarget by V8 +guards, I think. +* webaudio/_javascript_AudioNode.idl: Ditto. + 2012-03-05 Kishore Bolisetty Element not fully repainted after application and removal of transform Modified: trunk/Source/WebCore/bindings/js/JSAudioContextCustom.cpp (109869 => 109870) --- trunk/Source/WebCore/bindings/js/JSAudioContextCustom.cpp 2012-03-06 07:40:46 UTC (rev 109869) +++ trunk/Source/WebCore/bindings/js/JSAudioContextCustom.cpp 2012-03-06 07:59:24 UTC (rev 109870) @@ -101,7 +101,7 @@ if (!audioContext.get()) return throwVMError(exec, createReferenceError(exec, "Error creating AudioContext")); -return JSValue::encode(asObject(toJS(exec, jsConstructor->globalObject(), audioContext.get(; +return JSValue::encode(CREATE_DOM_WRAPPER(exec, jsConstructor->globalObject(), AudioContext, audioContext.get())); } } // namespace WebCore Modified: trunk/Source/WebCore/webaudio/AudioContext.idl (109869 => 109870) --- trunk/Source/WebCore/webaudio/AudioContext.idl 2012-03-06 07:40:46 UTC (rev 109869) +++ trunk/Source/WebCore/webaudio/AudioContext.idl 2012-03-06 07:59:24 UTC (rev 109870) @@ -30,9 +30,7 @@ CustomConstructor, ConstructorParameters=0, JSCustomMarkFunction, -#if defined(V8_BINDING) && V8_BINDING EventTarget -#endif ] AudioContext { // All rendered audio ultimately connects to destination, which represents the audio hardware. readonly attribute AudioDestinationNode destination; Modified: trunk/Source/WebCore/webaudio/_javascript_AudioNode.idl (109869 => 109870) --- trunk/Source/WebCore/webaudio/_javascript_AudioNode.idl 2012-03-06 07:40:46 UTC (rev 109869) +++ trunk/Source/WebCore/webaudio/_javascript_AudioNode.idl 2012-03-06 07:59:24 UTC (rev 109870) @@ -28,9 +28,7 @@ Conditional=WEB_AUDIO, JSGenerateToJSObject, JSCustomMarkFunction, -#if defined(V8_BINDING) && V8_BINDING EventTarget -#endif ] _javascript_AudioNode : AudioNode { // Rendering callback attribute EventListener onaudioprocess; ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109869] trunk/Source/JavaScriptCore
Title: [109869] trunk/Source/_javascript_Core Revision 109869 Author barraclo...@apple.com Date 2012-03-05 23:40:46 -0800 (Mon, 05 Mar 2012) Log Message Unreviewed windows build fix pt 1. * _javascript_Core.vcproj/_javascript_Core/_javascript_Core.def: Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/_javascript_Core.vcproj/_javascript_Core/_javascript_Core.def Diff Modified: trunk/Source/_javascript_Core/ChangeLog (109868 => 109869) --- trunk/Source/_javascript_Core/ChangeLog 2012-03-06 07:35:51 UTC (rev 109868) +++ trunk/Source/_javascript_Core/ChangeLog 2012-03-06 07:40:46 UTC (rev 109869) @@ -1,5 +1,11 @@ 2012-03-05 Gavin Barraclough +Unreviewed windows build fix pt 1. + +* _javascript_Core.vcproj/_javascript_Core/_javascript_Core.def: + +2012-03-05 Gavin Barraclough + putByIndex should throw in strict mode https://bugs.webkit.org/show_bug.cgi?id=80335 Modified: trunk/Source/_javascript_Core/_javascript_Core.vcproj/_javascript_Core/_javascript_Core.def (109868 => 109869) --- trunk/Source/_javascript_Core/_javascript_Core.vcproj/_javascript_Core/_javascript_Core.def 2012-03-06 07:35:51 UTC (rev 109868) +++ trunk/Source/_javascript_Core/_javascript_Core.vcproj/_javascript_Core/_javascript_Core.def 2012-03-06 07:40:46 UTC (rev 109869) @@ -275,8 +275,6 @@ ?put@JSByteArray@JSC@@SAXPAVJSCell@2@PAVExecState@2@ABVIdentifier@2@VJSValue@2@AAVPutPropertySlot@2@@Z ?put@JSGlobalObject@JSC@@SAXPAVJSCell@2@PAVExecState@2@ABVIdentifier@2@VJSValue@2@AAVPutPropertySlot@2@@Z ?put@JSObject@JSC@@SAXPAVJSCell@2@PAVExecState@2@ABVIdentifier@2@VJSValue@2@AAVPutPropertySlot@2@@Z -?putByIndex@JSByteArray@JSC@@SAXPAVJSCell@2@PAVExecState@2@IVJSValue@2@@Z -?putByIndex@JSObject@JSC@@SAXPAVJSCell@2@PAVExecState@2@IVJSValue@2@@Z ?putDirectVirtual@JSGlobalObject@JSC@@SAXPAVJSObject@2@PAVExecState@2@ABVIdentifier@2@VJSValue@2@I@Z ?putDirectVirtual@JSObject@JSC@@SAXPAV12@PAVExecState@2@ABVIdentifier@2@VJSValue@2@I@Z ?randomNumber@WTF@@YANXZ ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109868] trunk/LayoutTests
Title: [109868] trunk/LayoutTests Revision 109868 Author o...@webkit.org Date 2012-03-05 23:35:51 -0800 (Mon, 05 Mar 2012) Log Message [Qt] Unreviewed gardening. Skip new crashing tests, because crashes aren't bugs but features. :) * platform/qt-5.0-wk2/Skipped: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/qt-5.0-wk2/Skipped Diff Modified: trunk/LayoutTests/ChangeLog (109867 => 109868) --- trunk/LayoutTests/ChangeLog 2012-03-06 07:31:45 UTC (rev 109867) +++ trunk/LayoutTests/ChangeLog 2012-03-06 07:35:51 UTC (rev 109868) @@ -1,3 +1,10 @@ +2012-03-05 Csaba Osztrogonác + +[Qt] Unreviewed gardening. Skip new crashing tests, +because crashes aren't bugs but features. :) + +* platform/qt-5.0-wk2/Skipped: + 2012-03-05 Kishore Bolisetty Element not fully repainted after application and removal of transform Modified: trunk/LayoutTests/platform/qt-5.0-wk2/Skipped (109867 => 109868) --- trunk/LayoutTests/platform/qt-5.0-wk2/Skipped 2012-03-06 07:31:45 UTC (rev 109867) +++ trunk/LayoutTests/platform/qt-5.0-wk2/Skipped 2012-03-06 07:35:51 UTC (rev 109868) @@ -543,3 +543,12 @@ # [Qt][WK2] http/tests/security/xss-DENIED-xsl-document-securityOrigin.xml crashes # https://bugs.webkit.org/show_bug.cgi?id=80209 http/tests/security/xss-DENIED-xsl-document-securityOrigin.xml + +# [Qt][WK2] New hybi tests crash +# https://bugs.webkit.org/show_bug.cgi?id=80385 +http/tests/websocket/tests/hybi/unpaired-surrogates-in-message.html +http/tests/websocket/tests/hybi/unpaired-surrogates-in-close-reason.html + +# [Qt][WK2] fast/repaint/scroll-in-clipped-layer.html crashes +# https://bugs.webkit.org/show_bug.cgi?id=80387 +fast/repaint/scroll-in-clipped-layer.html ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109867] trunk
Title: [109867] trunk Revision 109867 Author commit-qu...@webkit.org Date 2012-03-05 23:31:45 -0800 (Mon, 05 Mar 2012) Log Message Element not fully repainted after application and removal of transform https://bugs.webkit.org/show_bug.cgi?id=61338 Source/WebCore: Patch by Kishore Bolisetty on 2012-03-05 Reviewed by Simon Fraser. RenderLayer is created and removed when a transform is applied and removed to element. To render the element properly after removal of transform on it, preferred widths and dimensions of the element and its containing block needs to be recalculated. Test: fast/repaint/transform-rotate-and-remove.html * rendering/RenderBoxModelObject.cpp: (WebCore): (WebCore::RenderBoxModelObject::styleWillChange): (WebCore::RenderBoxModelObject::styleDidChange): * rendering/RenderBoxModelObject.h: (RenderBoxModelObject): LayoutTests: Patch by Kishore Bolisetty on 2012-03-05 Reviewed by Simon Fraser. Test to remove the transform property of an element upon onload event. * fast/repaint/transform-rotate-and-remove.html: Added. * platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png: Added. * platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.txt: Added. Modified Paths trunk/LayoutTests/ChangeLog trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/rendering/RenderBoxModelObject.cpp trunk/Source/WebCore/rendering/RenderBoxModelObject.h Added Paths trunk/LayoutTests/fast/repaint/transform-rotate-and-remove.html trunk/LayoutTests/platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png trunk/LayoutTests/platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.txt Diff Modified: trunk/LayoutTests/ChangeLog (109866 => 109867) --- trunk/LayoutTests/ChangeLog 2012-03-06 07:23:21 UTC (rev 109866) +++ trunk/LayoutTests/ChangeLog 2012-03-06 07:31:45 UTC (rev 109867) @@ -1,3 +1,15 @@ +2012-03-05 Kishore Bolisetty + +Element not fully repainted after application and removal of transform +https://bugs.webkit.org/show_bug.cgi?id=61338 + +Reviewed by Simon Fraser. +Test to remove the transform property of an element upon onload event. + +* fast/repaint/transform-rotate-and-remove.html: Added. +* platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png: Added. +* platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.txt: Added. + 2012-03-05 Gavin Barraclough putByIndex should throw in strict mode Added: trunk/LayoutTests/fast/repaint/transform-rotate-and-remove.html (0 => 109867) --- trunk/LayoutTests/fast/repaint/transform-rotate-and-remove.html (rev 0) +++ trunk/LayoutTests/fast/repaint/transform-rotate-and-remove.html 2012-03-06 07:31:45 UTC (rev 109867) @@ -0,0 +1,26 @@ + + + + +Test case for https://bugs.webkit.org/show_bug.cgi?id=61338 + + +.rotated { +-webkit-transform:rotate(50deg); +} + + + +function repaintTest() { +document.getElementById('rel').classList.remove('rotated'); +} + + + + + + + Property changes on: trunk/LayoutTests/fast/repaint/transform-rotate-and-remove.html ___ Added: svn:executable Added: trunk/LayoutTests/platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png (0 => 109867) --- trunk/LayoutTests/platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png (rev 0) +++ trunk/LayoutTests/platform/chromium-linux/fast/repaint/transform-rotate-and-remove-expected.png 2012-03-06 07:31:45 UTC (rev 109867) @@ -0,0 +1,6 @@ +\x89PNG + + +IHDR X ' )tEXtchecksum 58669f212a7e1be6c12dca3985b1d0b6ƭR +\xF8IDATx\x9C\xED\xDC\xC1\x89\x83PFQ\xD2㳛`7v\xE94\xA0\x93.q\x90s\x96 ·\xBC\xFC\x8B7\x8F1\xA6#۶\xFEWY\x96\xE5\xEA ܍\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\xD8\xE3\xEC\x8Fy\x9D\xBF\xB9^\xD3\xF1\x9Bm \xF0߸` \xC4 @L` \xC4 @L` \xC4 @L` \xC4 @L` \xC4N_r\xFF\xDB\xFElg\xC04MӼ^\xBD +.X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X 1\x81 X \xB1\xC7g\x9F\xCDk; \xE0>\\xB0 b &\xB0 b &\xB0 b &\xB0 b &\xB0 b\xA7/\xB9\xEF\xCF\xFD\x9B;\xE0\xA5eY\xAE\x9E oq\xC1 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0 \x88 , \x80\x98\xC0
[webkit-changes] [109864] trunk/Source/WebCore
Title: [109864] trunk/Source/WebCore Revision 109864 Author shin...@chromium.org Date 2012-03-05 22:59:02 -0800 (Mon, 05 Mar 2012) Log Message InsertionPoint::attach should be consistent with Element. https://bugs.webkit.org/show_bug.cgi?id=80373 Reviewed by Hajime Morita. This patch is preparation for coming patches. InsertionPoint used to attach fallback elements before attaching distributed elements. To be consistent with Element::attach behavior, attaching distributed elements first is natural, because Element attaches a shadow tree first. Also, this patch extracts a few methods form InsretionPoint::attach() and detach() to keep code clean. They will become messy without this refactoring when adding patch. No new tests. Should be covered by existing tests. * html/shadow/InsertionPoint.cpp: (WebCore::InsertionPoint::attach): (WebCore::InsertionPoint::detach): (WebCore::InsertionPoint::distributeHostChildren): (WebCore): (WebCore::InsertionPoint::clearDistribution): (WebCore::InsertionPoint::attachDistributedNode): * html/shadow/InsertionPoint.h: (InsertionPoint): Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/shadow/InsertionPoint.cpp trunk/Source/WebCore/html/shadow/InsertionPoint.h Diff Modified: trunk/Source/WebCore/ChangeLog (109863 => 109864) --- trunk/Source/WebCore/ChangeLog 2012-03-06 06:54:23 UTC (rev 109863) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 06:59:02 UTC (rev 109864) @@ -1,3 +1,32 @@ +2012-03-05 Shinya Kawanaka + +InsertionPoint::attach should be consistent with Element. +https://bugs.webkit.org/show_bug.cgi?id=80373 + +Reviewed by Hajime Morita. + +This patch is preparation for coming patches. + +InsertionPoint used to attach fallback elements before attaching distributed elements. +To be consistent with Element::attach behavior, attaching distributed elements first is +natural, because Element attaches a shadow tree first. + +Also, this patch extracts a few methods form InsretionPoint::attach() and detach() +to keep code clean. They will become messy without this refactoring when adding + patch. + +No new tests. Should be covered by existing tests. + +* html/shadow/InsertionPoint.cpp: +(WebCore::InsertionPoint::attach): +(WebCore::InsertionPoint::detach): +(WebCore::InsertionPoint::distributeHostChildren): +(WebCore): +(WebCore::InsertionPoint::clearDistribution): +(WebCore::InsertionPoint::attachDistributedNode): +* html/shadow/InsertionPoint.h: +(InsertionPoint): + 2012-03-05 Adam Barth Attempt to fix a number of GTK tests. Modified: trunk/Source/WebCore/html/shadow/InsertionPoint.cpp (109863 => 109864) --- trunk/Source/WebCore/html/shadow/InsertionPoint.cpp 2012-03-06 06:54:23 UTC (rev 109863) +++ trunk/Source/WebCore/html/shadow/InsertionPoint.cpp 2012-03-06 06:59:02 UTC (rev 109864) @@ -48,31 +48,23 @@ void InsertionPoint::attach() { -ShadowRoot* root = toShadowRoot(shadowTreeRootNode()); - -if (root) { -HTMLContentSelector* selector = root->tree()->ensureSelector(); -selector->unselect(&m_selections); -selector->select(this, &m_selections); +if (ShadowRoot* root = toShadowRoot(shadowTreeRootNode())) { +distributeHostChildren(root->tree()); +attachDistributedNode(); } HTMLElement::attach(); - -if (root) { -for (HTMLContentSelection* selection = m_selections.first(); selection; selection = selection->next()) -selection->node()->attach(); -} } void InsertionPoint::detach() { if (ShadowRoot* root = toShadowRoot(shadowTreeRootNode())) { -if (HTMLContentSelector* selector = root->tree()->selector()) -selector->unselect(&m_selections); +ShadowTree* tree = root->tree(); +clearDistribution(tree); // When shadow element is detached, shadow tree should be recreated to re-calculate selector for // other insertion points. -root->tree()->setNeedsReattachHostChildrenAndShadow(); +tree->setNeedsReattachHostChildrenAndShadow(); } ASSERT(m_selections.isEmpty()); @@ -91,4 +83,23 @@ return !isShadowBoundary() && HTMLElement::rendererIsNeeded(context); } +inline void InsertionPoint::distributeHostChildren(ShadowTree* tree) +{ +HTMLContentSelector* selector = tree->ensureSelector(); +selector->unselect(&m_selections); +selector->select(this, &m_selections); +} + +inline void InsertionPoint::clearDistribution(ShadowTree* tree) +{ +if (HTMLContentSelector* selector = tree->selector()) +selector->unselect(&m_selections); +} + +inline void InsertionPoint::attachDistributedNode() +{ +for (HTMLContentSelection* selection = m_selections.first(); selection; selection = selection->next()) +selection->node()->attach(); +} + } // namespace WebCore
[webkit-changes] [109863] trunk/Source/JavaScriptCore
Title: [109863] trunk/Source/_javascript_Core Revision 109863 Author fpi...@apple.com Date 2012-03-05 22:54:23 -0800 (Mon, 05 Mar 2012) Log Message The LLInt should work even when the JIT is disabled https://bugs.webkit.org/show_bug.cgi?id=80340 Reviewed by Gavin Barraclough. * assembler/MacroAssemblerCodeRef.h: (JSC::MacroAssemblerCodePtr::createLLIntCodePtr): (MacroAssemblerCodeRef): (JSC::MacroAssemblerCodeRef::createLLIntCodeRef): * interpreter/Interpreter.cpp: (JSC::Interpreter::initialize): (JSC::Interpreter::execute): (JSC::Interpreter::executeCall): (JSC::Interpreter::executeConstruct): * jit/JIT.h: (JSC::JIT::compileCTINativeCall): * jit/JITStubs.h: (JSC::JITThunks::ctiNativeCall): (JSC::JITThunks::ctiNativeConstruct): * llint/LLIntEntrypoints.cpp: (JSC::LLInt::getFunctionEntrypoint): (JSC::LLInt::getEvalEntrypoint): (JSC::LLInt::getProgramEntrypoint): * llint/LLIntSlowPaths.cpp: (JSC::LLInt::LLINT_SLOW_PATH_DECL): (LLInt): * llint/LLIntSlowPaths.h: (LLInt): * llint/LowLevelInterpreter.h: * llint/LowLevelInterpreter32_64.asm: * runtime/Executable.h: (NativeExecutable): (JSC::NativeExecutable::create): (JSC::NativeExecutable::finishCreation): * runtime/JSGlobalData.cpp: (JSC::JSGlobalData::JSGlobalData): * runtime/JSGlobalData.h: (JSGlobalData): * runtime/Options.cpp: (Options): (JSC::Options::parse): (JSC::Options::initializeOptions): * runtime/Options.h: (Options): * wtf/Platform.h: Modified Paths trunk/Source/_javascript_Core/ChangeLog trunk/Source/_javascript_Core/assembler/MacroAssemblerCodeRef.h trunk/Source/_javascript_Core/interpreter/Interpreter.cpp trunk/Source/_javascript_Core/jit/JIT.h trunk/Source/_javascript_Core/jit/JITStubs.h trunk/Source/_javascript_Core/llint/LLIntEntrypoints.cpp trunk/Source/_javascript_Core/llint/LLIntSlowPaths.cpp trunk/Source/_javascript_Core/llint/LLIntSlowPaths.h trunk/Source/_javascript_Core/llint/LowLevelInterpreter.h trunk/Source/_javascript_Core/llint/LowLevelInterpreter32_64.asm trunk/Source/_javascript_Core/runtime/Executable.h trunk/Source/_javascript_Core/runtime/JSGlobalData.cpp trunk/Source/_javascript_Core/runtime/JSGlobalData.h trunk/Source/_javascript_Core/runtime/Options.cpp trunk/Source/_javascript_Core/runtime/Options.h trunk/Source/_javascript_Core/wtf/Platform.h Diff Modified: trunk/Source/_javascript_Core/ChangeLog (109862 => 109863) --- trunk/Source/_javascript_Core/ChangeLog 2012-03-06 06:31:34 UTC (rev 109862) +++ trunk/Source/_javascript_Core/ChangeLog 2012-03-06 06:54:23 UTC (rev 109863) @@ -1,3 +1,52 @@ +2012-03-05 Filip Pizlo + +The LLInt should work even when the JIT is disabled +https://bugs.webkit.org/show_bug.cgi?id=80340 + + +Reviewed by Gavin Barraclough. + +* assembler/MacroAssemblerCodeRef.h: +(JSC::MacroAssemblerCodePtr::createLLIntCodePtr): +(MacroAssemblerCodeRef): +(JSC::MacroAssemblerCodeRef::createLLIntCodeRef): +* interpreter/Interpreter.cpp: +(JSC::Interpreter::initialize): +(JSC::Interpreter::execute): +(JSC::Interpreter::executeCall): +(JSC::Interpreter::executeConstruct): +* jit/JIT.h: +(JSC::JIT::compileCTINativeCall): +* jit/JITStubs.h: +(JSC::JITThunks::ctiNativeCall): +(JSC::JITThunks::ctiNativeConstruct): +* llint/LLIntEntrypoints.cpp: +(JSC::LLInt::getFunctionEntrypoint): +(JSC::LLInt::getEvalEntrypoint): +(JSC::LLInt::getProgramEntrypoint): +* llint/LLIntSlowPaths.cpp: +(JSC::LLInt::LLINT_SLOW_PATH_DECL): +(LLInt): +* llint/LLIntSlowPaths.h: +(LLInt): +* llint/LowLevelInterpreter.h: +* llint/LowLevelInterpreter32_64.asm: +* runtime/Executable.h: +(NativeExecutable): +(JSC::NativeExecutable::create): +(JSC::NativeExecutable::finishCreation): +* runtime/JSGlobalData.cpp: +(JSC::JSGlobalData::JSGlobalData): +* runtime/JSGlobalData.h: +(JSGlobalData): +* runtime/Options.cpp: +(Options): +(JSC::Options::parse): +(JSC::Options::initializeOptions): +* runtime/Options.h: +(Options): +* wtf/Platform.h: + 2012-03-05 Yuqiang Xian Checks for dead variables are not sufficient when fixing the expected Modified: trunk/Source/_javascript_Core/assembler/MacroAssemblerCodeRef.h (109862 => 109863) --- trunk/Source/_javascript_Core/assembler/MacroAssemblerCodeRef.h 2012-03-06 06:31:34 UTC (rev 109862) +++ trunk/Source/_javascript_Core/assembler/MacroAssemblerCodeRef.h 2012-03-06 06:54:23 UTC (rev 109863) @@ -280,6 +280,10 @@ return result; } +static MacroAssemblerCodePtr createLLIntCodePtr(void (*function)()) +{ +return createFromExecutableAddress(bitwise_cast(function)); +} explicit MacroAssemblerCodePtr(ReturnAddressPtr ra) : m_value(ra.value()) { @@ -340,6 +344,12 @@ return
[webkit-changes] [109862] trunk/Tools
Title: [109862] trunk/Tools Revision 109862 Author o...@webkit.org Date 2012-03-05 22:31:34 -0800 (Mon, 05 Mar 2012) Log Message Unreviewed, rolling out r109733. http://trac.webkit.org/changeset/109733 https://bugs.webkit.org/show_bug.cgi?id=80378 It broke Qt5 build system (Requested by Ossy on #webkit). Patch by Sheriff Bot on 2012-03-05 * Tools.pro: * qmake/config.tests/gccdepends/empty.cpp: Removed. * qmake/config.tests/gccdepends/gccdepends.pro: Removed. * qmake/mkspecs/features/default_pre.prf: * qmake/mkspecs/features/features.prf: * qmake/mkspecs/features/unix/default_pre.prf: * qmake/sync.profile: Modified Paths trunk/Tools/ChangeLog trunk/Tools/Tools.pro trunk/Tools/qmake/mkspecs/features/default_pre.prf trunk/Tools/qmake/mkspecs/features/features.prf trunk/Tools/qmake/mkspecs/features/unix/default_pre.prf trunk/Tools/qmake/sync.profile Removed Paths trunk/Tools/qmake/config.tests/gccdepends/empty.cpp trunk/Tools/qmake/config.tests/gccdepends/gccdepends.pro Diff Modified: trunk/Tools/ChangeLog (109861 => 109862) --- trunk/Tools/ChangeLog 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/ChangeLog 2012-03-06 06:31:34 UTC (rev 109862) @@ -1,3 +1,19 @@ +2012-03-05 Sheriff Bot + +Unreviewed, rolling out r109733. +http://trac.webkit.org/changeset/109733 +https://bugs.webkit.org/show_bug.cgi?id=80378 + +It broke Qt5 build system (Requested by Ossy on #webkit). + +* Tools.pro: +* qmake/config.tests/gccdepends/empty.cpp: Removed. +* qmake/config.tests/gccdepends/gccdepends.pro: Removed. +* qmake/mkspecs/features/default_pre.prf: +* qmake/mkspecs/features/features.prf: +* qmake/mkspecs/features/unix/default_pre.prf: +* qmake/sync.profile: + 2012-03-05 Eric Seidel Update _javascript_Core files to use fully-qualified WTF include path Modified: trunk/Tools/Tools.pro (109861 => 109862) --- trunk/Tools/Tools.pro 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/Tools.pro 2012-03-06 06:31:34 UTC (rev 109862) @@ -29,7 +29,6 @@ qmake/sync.profile \ qmake/config.tests/README \ qmake/config.tests/fontconfig/* \ -qmake/config.tests/gccdepends/* \ qmake/mkspecs/modules/* \ qmake/mkspecs/features/*.prf \ qmake/mkspecs/features/mac/*.prf \ Deleted: trunk/Tools/qmake/config.tests/gccdepends/empty.cpp ( => ) Deleted: trunk/Tools/qmake/config.tests/gccdepends/gccdepends.pro === --- trunk/Tools/qmake/config.tests/gccdepends/gccdepends.pro 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/qmake/config.tests/gccdepends/gccdepends.pro 2012-03-06 06:31:34 UTC (rev 109862) @@ -1,12 +0,0 @@ -TEMPLATE = aux -SOURCES += empty.cpp -OBJECTS_DIR = obj -QMAKE_CXXFLAGS += -MD - -test.commands = test -f $$OBJECTS_DIR/empty.d && test ! -f empty.d && touch $$basename(PWD) -test.depends = $$OBJECTS_DIR/empty.o -QMAKE_EXTRA_TARGETS += test - -default.target = all -default.depends += test -QMAKE_EXTRA_TARGETS += default Modified: trunk/Tools/qmake/mkspecs/features/default_pre.prf (109861 => 109862) --- trunk/Tools/qmake/mkspecs/features/default_pre.prf 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/qmake/mkspecs/features/default_pre.prf 2012-03-06 06:31:34 UTC (rev 109862) @@ -7,15 +7,6 @@ load(functions) -# Load results of config tests (only supported in Qt 5) -haveQt(5): include($${ROOT_BUILD_DIR}/Tools/qmake/.qmake.cache, , true) - -unix:config_test_gccdepends { -# We have a compiler that supports the -MD option (and neither -# teambuilder nor icecream is messing up the .d file output). -CONFIG += GNUmake gcc_MD_depends -} - # FIXME: In this case we probably want to exit early so that we # don't load all of the custom build tweaks needlessly. pro_file = $$_PRO_FILE_ # Checking the variable directly does not work Modified: trunk/Tools/qmake/mkspecs/features/features.prf (109861 => 109862) --- trunk/Tools/qmake/mkspecs/features/features.prf 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/qmake/mkspecs/features/features.prf 2012-03-06 06:31:34 UTC (rev 109862) @@ -5,6 +5,9 @@ # See 'Tools/qmake/README' for an overview of the build system # --- +# Load results of config tests (only supported in Qt 5) +haveQt(5): include($${ROOT_BUILD_DIR}/Tools/qmake/.qmake.cache, , true) + CONFIG(minimal) { DEFINES += ENABLE_NETSCAPE_PLUGIN_API=0 } Modified: trunk/Tools/qmake/mkspecs/features/unix/default_pre.prf (109861 => 109862) --- trunk/Tools/qmake/mkspecs/features/unix/default_pre.prf 2012-03-06 06:21:23 UTC (rev 109861) +++ trunk/Tools/qmake/mkspecs/features/unix/default_pre.prf 2012-03-06 06:31:34 UTC (rev 109862) @@ -6,7 +6,10 @@ load(functions) -haveQt(5): CONFIG += object_parallel_to_source +# Assume we use GNU make and a non-ancient GCC (compatible) +# compiler that supports the -MD option.
[webkit-changes] [109861] trunk/Source/WebCore
Title: [109861] trunk/Source/WebCore Revision 109861 Author aba...@webkit.org Date 2012-03-05 22:21:23 -0800 (Mon, 05 Mar 2012) Log Message Attempt to fix a number of GTK tests. * GNUmakefile.list.am: Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/GNUmakefile.list.am Diff Modified: trunk/Source/WebCore/ChangeLog (109860 => 109861) --- trunk/Source/WebCore/ChangeLog 2012-03-06 06:16:12 UTC (rev 109860) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 06:21:23 UTC (rev 109861) @@ -1,3 +1,9 @@ +2012-03-05 Adam Barth + +Attempt to fix a number of GTK tests. + +* GNUmakefile.list.am: + 2012-03-05 Joseph Pecoraro Unreviewed rollout of r109858 for restructuring. Modified: trunk/Source/WebCore/GNUmakefile.list.am (109860 => 109861) --- trunk/Source/WebCore/GNUmakefile.list.am 2012-03-06 06:16:12 UTC (rev 109860) +++ trunk/Source/WebCore/GNUmakefile.list.am 2012-03-06 06:21:23 UTC (rev 109861) @@ -4749,7 +4749,8 @@ $(WebCore)/storage/SQLTransactionCallback.idl \ $(WebCore)/storage/SQLTransactionErrorCallback.idl \ $(WebCore)/storage/SQLTransactionSync.idl \ - $(WebCore)/storage/SQLTransactionSyncCallback.idl + $(WebCore)/storage/SQLTransactionSyncCallback.idl \ + $(WebCore)/storage/WorkerContextSQLDatabase.idl endif # END ENABLE_SQL_DATABASE # ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109860] trunk/Source
Title: [109860] trunk/Source Revision 109860 Author joep...@webkit.org Date 2012-03-05 22:16:12 -0800 (Mon, 05 Mar 2012) Log Message Unreviewed rollout of r109858 for restructuring. Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/WebCore.exp.in trunk/Source/WebCore/inspector/InspectorClient.h trunk/Source/WebCore/inspector/InspectorController.cpp trunk/Source/WebCore/inspector/InspectorController.h trunk/Source/WebCore/inspector/InspectorFrontendClient.h trunk/Source/WebCore/inspector/InspectorFrontendClientLocal.cpp trunk/Source/WebCore/inspector/InspectorFrontendClientLocal.h trunk/Source/WebCore/inspector/InspectorInstrumentation.cpp trunk/Source/WebCore/inspector/InspectorInstrumentation.h trunk/Source/WebCore/inspector/front-end/InspectorFrontendAPI.js trunk/Source/WebCore/inspector/front-end/inspector.js trunk/Source/WebCore/page/FrameView.cpp trunk/Source/WebKit/mac/ChangeLog trunk/Source/WebKit/mac/WebCoreSupport/WebInspectorClient.h trunk/Source/WebKit/mac/WebCoreSupport/WebInspectorClient.mm trunk/Source/WebKit2/ChangeLog trunk/Source/WebKit2/WebProcess/WebCoreSupport/WebInspectorClient.cpp trunk/Source/WebKit2/WebProcess/WebCoreSupport/WebInspectorClient.h trunk/Source/WebKit2/WebProcess/WebPage/WebInspector.cpp trunk/Source/WebKit2/WebProcess/WebPage/WebInspector.h Diff Modified: trunk/Source/WebCore/ChangeLog (109859 => 109860) --- trunk/Source/WebCore/ChangeLog 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 06:16:12 UTC (rev 109860) @@ -1,5 +1,9 @@ 2012-03-05 Joseph Pecoraro +Unreviewed rollout of r109858 for restructuring. + +2012-03-05 Joseph Pecoraro + Web Inspector: Hide dock button when not allowed to dock There are times when an undocked inspector frontend is not allowed to Modified: trunk/Source/WebCore/WebCore.exp.in (109859 => 109860) --- trunk/Source/WebCore/WebCore.exp.in 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/WebCore.exp.in 2012-03-06 06:16:12 UTC (rev 109860) @@ -1743,7 +1743,6 @@ __ZN7WebCore28InspectorFrontendClientLocal19windowObjectClearedEv __ZN7WebCore28InspectorFrontendClientLocal20sendMessageToBackendERKN3WTF6StringE __ZN7WebCore28InspectorFrontendClientLocal21isProfilingJavaScriptEv -__ZN7WebCore28InspectorFrontendClientLocal21setDockingUnavailableEb __ZN7WebCore28InspectorFrontendClientLocal23stopProfilingJavaScriptEv __ZN7WebCore28InspectorFrontendClientLocal24showMainResourceForFrameEPNS_5FrameE __ZN7WebCore28InspectorFrontendClientLocal24startProfilingJavaScriptEv Modified: trunk/Source/WebCore/inspector/InspectorClient.h (109859 => 109860) --- trunk/Source/WebCore/inspector/InspectorClient.h 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/inspector/InspectorClient.h 2012-03-06 06:16:12 UTC (rev 109860) @@ -46,7 +46,6 @@ virtual void openInspectorFrontend(InspectorController*) = 0; virtual void closeInspectorFrontend() = 0; virtual void bringFrontendToFront() = 0; -virtual void updateDockingAvailability() { } virtual void highlight() = 0; virtual void hideHighlight() = 0; Modified: trunk/Source/WebCore/inspector/InspectorController.cpp (109859 => 109860) --- trunk/Source/WebCore/inspector/InspectorController.cpp 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/inspector/InspectorController.cpp 2012-03-06 06:16:12 UTC (rev 109860) @@ -361,13 +361,9 @@ m_debuggerAgent->resume(&error); } } + #endif -void InspectorController::updateDockingAvailability() -{ -m_inspectorClient->updateDockingAvailability(); -} - void InspectorController::setResourcesDataSizeLimitsFromInternals(int maximumResourcesContentSize, int maximumSingleResourceContentSize) { m_resourceAgent->setResourcesDataSizeLimitsFromInternals(maximumResourcesContentSize, maximumSingleResourceContentSize); Modified: trunk/Source/WebCore/inspector/InspectorController.h (109859 => 109860) --- trunk/Source/WebCore/inspector/InspectorController.h 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/inspector/InspectorController.h 2012-03-06 06:16:12 UTC (rev 109860) @@ -106,8 +106,6 @@ void resume(); #endif -void updateDockingAvailability(); - void setResourcesDataSizeLimitsFromInternals(int maximumResourcesContentSize, int maximumSingleResourceContentSize); InspectorPageAgent* pageAgent() const { return m_pageAgent; } Modified: trunk/Source/WebCore/inspector/InspectorFrontendClient.h (109859 => 109860) --- trunk/Source/WebCore/inspector/InspectorFrontendClient.h 2012-03-06 06:11:52 UTC (rev 109859) +++ trunk/Source/WebCore/inspector/InspectorFrontendClient.h 2012-03-06 06:16:12 UTC (rev 109860) @@ -65,9 +65,6 @@ virtual void inspectedURLChanged(const String&) = 0; virtual void sendMessageToBackend(const String&) = 0; - -virtual bool canAttachWindow() { return true; } -virtual void setDockingUnavailable(bool) { } }; } // na
[webkit-changes] [109858] trunk/Source
Title: [109858] trunk/Source Revision 109858 Author joep...@webkit.org Date 2012-03-05 22:03:12 -0800 (Mon, 05 Mar 2012) Log Message Web Inspector: Hide dock button when not allowed to dock Source/WebCore: There are times when an undocked inspector frontend is not allowed to attach to the main window. We can remove the dock button in those cases. Reviewed by Timothy Hatcher. No new tests. This functionality is port specific right now. * page/FrameView.cpp: (WebCore::FrameView::performPostLayoutTasks): When the main frame is resized we let the inspector frontend client update its docked availability. * WebCore.exp.in: * inspector/InspectorInstrumentation.h: * inspector/InspectorInstrumentation.cpp: (WebCore::InspectorInstrumentation::didResizeMainFrameImpl): (WebCore::InspectorInstrumentation::didResizeMainFrame): * inspector/InspectorClient.h: (WebCore::InspectorClient::updateDockingAvailability): * inspector/InspectorController.h: * inspector/InspectorController.cpp: (WebCore::InspectorController::resume): (WebCore::InspectorController::updateDockingAvailability): When the main frame is resized call up to the InspectorClient to let it update docking availability of the inspector frontend. * inspector/InspectorFrontendClient.h: * inspector/InspectorFrontendClientLocal.h: * inspector/InspectorFrontendClientLocal.cpp: (WebCore::InspectorFrontendClientLocal::frontendLoaded): (WebCore::InspectorFrontendClientLocal::setDockingUnavailable): On ports where the inspector frontend client is local, provide a setDockingUnavailable InspectorFrontendAPI to update the docking state. Automatically update availablity when the frontend completes loading. * inspector/front-end/InspectorFrontendAPI.js: (InspectorFrontendAPI.setDockingUnavailable): * inspector/front-end/inspector.js: (WebInspector._createGlobalStatusBarItems): (WebInspector.setAttachedWindow): (WebInspector.setDockingUnavailable): (WebInspector.updateDockToggleButtonVisibility): Update the dock button's visibility when its created, when the attached state changes, when get a frontend API notification that we cannot attach. Source/WebKit/mac: * WebCoreSupport/WebInspectorClient.h: (WebInspectorClient): * WebCoreSupport/WebInspectorClient.mm: (WebInspectorClient::updateDockingAvailability): Source/WebKit2: * WebProcess/WebCoreSupport/WebInspectorClient.cpp: (WebKit::WebInspectorClient::updateDockingAvailability): * WebProcess/WebCoreSupport/WebInspectorClient.h: * WebProcess/WebPage/WebInspector.cpp: (WebKit::WebInspector::updateDockingAvailability): * WebProcess/WebPage/WebInspector.h: Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/WebCore.exp.in trunk/Source/WebCore/inspector/InspectorClient.h trunk/Source/WebCore/inspector/InspectorController.cpp trunk/Source/WebCore/inspector/InspectorController.h trunk/Source/WebCore/inspector/InspectorFrontendClient.h trunk/Source/WebCore/inspector/InspectorFrontendClientLocal.cpp trunk/Source/WebCore/inspector/InspectorFrontendClientLocal.h trunk/Source/WebCore/inspector/InspectorInstrumentation.cpp trunk/Source/WebCore/inspector/InspectorInstrumentation.h trunk/Source/WebCore/inspector/front-end/InspectorFrontendAPI.js trunk/Source/WebCore/inspector/front-end/inspector.js trunk/Source/WebCore/page/FrameView.cpp trunk/Source/WebKit/mac/ChangeLog trunk/Source/WebKit/mac/WebCoreSupport/WebInspectorClient.h trunk/Source/WebKit/mac/WebCoreSupport/WebInspectorClient.mm trunk/Source/WebKit2/ChangeLog trunk/Source/WebKit2/WebProcess/WebCoreSupport/WebInspectorClient.cpp trunk/Source/WebKit2/WebProcess/WebCoreSupport/WebInspectorClient.h trunk/Source/WebKit2/WebProcess/WebPage/WebInspector.cpp trunk/Source/WebKit2/WebProcess/WebPage/WebInspector.h Diff Modified: trunk/Source/WebCore/ChangeLog (109857 => 109858) --- trunk/Source/WebCore/ChangeLog 2012-03-06 05:26:21 UTC (rev 109857) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 06:03:12 UTC (rev 109858) @@ -1,3 +1,52 @@ +2012-03-05 Joseph Pecoraro + + Web Inspector: Hide dock button when not allowed to dock + +There are times when an undocked inspector frontend is not allowed to +attach to the main window. We can remove the dock button in those cases. + +Reviewed by Timothy Hatcher. + +No new tests. This functionality is port specific right now. + +* page/FrameView.cpp: +(WebCore::FrameView::performPostLayoutTasks): +When the main frame is resized we let the inspector frontend client +update its docked availablility. + +* WebCore.exp.in: +* inspector/InspectorInstrumentation.h: +* inspector/InspectorInstrumentation.cpp: +(WebCore::InspectorInstrumentation::didResizeMainFrameImpl): +(WebCore::InspectorInstrumentation::didResizeMainFrame): +* inspector/InspectorClient.h: +(WebCore::InspectorClient::updateDockingAvailability): +* inspector/InspectorController.h: +* inspector/InspectorCont
[webkit-changes] [109857] tags/Safari-536.2.1/Source/WebCore
Title: [109857] tags/Safari-536.2.1/Source/WebCore Revision 109857 Author lforsch...@apple.com Date 2012-03-05 21:26:21 -0800 (Mon, 05 Mar 2012) Log Message Merged r109847. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109856 => 109857) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:24:44 UTC (rev 109856) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:26:21 UTC (rev 109857) @@ -1,5 +1,16 @@ 2011-03-05 Lucas Forschler +Merge 109847 + +2012-03-05 Anders Carlsson + +Address review comments from https://bugs.webkit.org/show_bug.cgi?id=80368 + +* platform/graphics/ca/mac/TileCache.mm: +(WebCore::TileCache::setNeedsDisplayInRect): + +2011-03-05 Lucas Forschler + Merge 109843 2012-03-05 Anders Carlsson Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm (109856 => 109857) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 05:24:44 UTC (rev 109856) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 05:26:21 UTC (rev 109857) @@ -91,13 +91,13 @@ if (m_tiles.isEmpty()) return; -IntRect scaledRect = rect; +FloatRect scaledRect(rect); scaledRect.scale(m_scale); // Find the tiles that need to be invalidated. TileIndex topLeft; TileIndex bottomRight; -getTileIndexRangeForRect(intersection(scaledRect, m_tileCoverageRect), topLeft, bottomRight); +getTileIndexRangeForRect(intersection(enclosingIntRect(scaledRect), m_tileCoverageRect), topLeft, bottomRight); for (int y = topLeft.y(); y <= bottomRight.y(); ++y) { for (int x = topLeft.x(); x <= bottomRight.x(); ++x) { ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109856] tags/Safari-536.2.1/Source
Title: [109856] tags/Safari-536.2.1/Source Revision 109856 Author lforsch...@apple.com Date 2012-03-05 21:24:44 -0800 (Mon, 05 Mar 2012) Log Message Merged r109843. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm tags/Safari-536.2.1/Source/WebCore/rendering/RenderLayerBacking.cpp tags/Safari-536.2.1/Source/WebKit2/ChangeLog tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/DrawingArea.h tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/WebPage.cpp tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.h tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109855 => 109856) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:23:04 UTC (rev 109855) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:24:44 UTC (rev 109856) @@ -1,5 +1,51 @@ 2011-03-05 Lucas Forschler +Merge 109843 + +2012-03-05 Anders Carlsson + +pinch-to-zoom and double-tap flicker when using the new scrolling model +https://bugs.webkit.org/show_bug.cgi?id=80368 + + +Reviewed by Sam Weinig. + +In order to work better with zooming, make the tile cache undo the scale transformation +and handle the scaling manually. This avoids creating huge tile backing stores when zoomed in. + +* platform/graphics/ca/mac/TileCache.mm: +(WebCore::TileCache::TileCache): +Initialize m_scale to 1. + +(WebCore::TileCache::setNeedsDisplayInRect): +Scale the given rect appropriately. + +(WebCore::TileCache::drawLayer): +Apply a scale context transform. + +(WebCore::TileCache::setScale): +No longer set the contents scale. Instead, update the scale and revalidate the tiles. + +(WebCore::TileCache::revalidateTiles): +Return early if the bounds are empty. This avoids showing a single tile if that happens due to a race condition. + +(WebCore::TileCache::getTileIndexRangeForRect): +Apply the scale to the bounds. + +(WebCore::TileCache::createTileLayer): +Don't set the contents scale. + +* platform/graphics/ca/mac/WebTileCacheLayer.mm: +(-[WebTileCacheLayer setContentsScale:]): +Call TileCache::setScale. + +* rendering/RenderLayerBacking.cpp: +(WebCore::RenderLayerBacking::updateCompositedBounds): +Make sure to give the tile cache layer sane composited bounds, even if the page has absolutely positioned +elements that are outside of the page. + +2011-03-05 Lucas Forschler + Merge 109812 2012-03-05 Anders Carlsson Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.h (109855 => 109856) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.h 2012-03-06 05:23:04 UTC (rev 109855) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.h 2012-03-06 05:24:44 UTC (rev 109856) @@ -56,8 +56,9 @@ void setNeedsDisplay(); void setNeedsDisplayInRect(const IntRect&); void drawLayer(WebTileLayer*, CGContextRef); -void setContentsScale(CGFloat); +void setScale(CGFloat); + bool acceleratesDrawing() const { return m_acceleratesDrawing; } void setAcceleratesDrawing(bool); @@ -99,6 +100,8 @@ Timer m_tileRevalidationTimer; IntRect m_tileCoverageRect; +CGFloat m_scale; + bool m_acceleratesDrawing; RetainPtr m_tileDebugBorderColor; Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm (109855 => 109856) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 05:23:04 UTC (rev 109855) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 05:24:44 UTC (rev 109856) @@ -54,6 +54,7 @@ , m_tileContainerLayer(adoptCF([[CALayer alloc] init])) , m_tileSize(tileSize) , m_tileRevalidationTimer(this, &TileCache::tileRevalidationTimerFired) +, m_scale(1) , m_acceleratesDrawing(false) , m_tileDebugBorderWidth(0) { @@ -90,10 +91,13 @@ if (m_tiles.isEmpty()) return; +IntRect scaledRect = rect; +scaledRect.scale(m_scale); + // Find the tiles that need to be invalidated. TileIndex topLeft; TileIndex bottomRight; -getTileIndexRangeForRect(intersection(rect, m_tileCoverageRect), topLeft, bottomRight); +getTileIndexRangeForRect(intersection(scaledRect, m_tileCoverageRect), topLeft, bottomRight); for (int y = topLeft.y(); y <= bottomRi
[webkit-changes] [109855] tags/Safari-536.2.1/Source
Title: [109855] tags/Safari-536.2.1/Source Revision 109855 Author lforsch...@apple.com Date 2012-03-05 21:23:04 -0800 (Mon, 05 Mar 2012) Log Message Merged r109812. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h tags/Safari-536.2.1/Source/WebKit2/ChangeLog tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.h tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109854 => 109855) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:20:32 UTC (rev 109854) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:23:04 UTC (rev 109855) @@ -1,5 +1,25 @@ 2011-03-05 Lucas Forschler +Merge 109812 + +2012-03-05 Anders Carlsson + +Be more aggressive about repainting page overlays +https://bugs.webkit.org/show_bug.cgi?id=80336 + + +Reviewed by Simon Fraser. + +Add a way to find out if a given GraphicsLayer is going to be repainted. + +* WebCore.exp.in: +* platform/graphics/GraphicsLayer.h: +(WebCore::GraphicsLayer::needsDisplay): +(GraphicsLayer): +* platform/graphics/ca/GraphicsLayerCA.h: + +2011-03-05 Lucas Forschler + Merge 109801 2012-03-05 Anders Carlsson Modified: tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in (109854 => 109855) --- tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in 2012-03-06 05:20:32 UTC (rev 109854) +++ tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in 2012-03-06 05:23:04 UTC (rev 109855) @@ -143,6 +143,7 @@ __ZN7WebCore10JSDocument6s_infoE __ZN7WebCore10MouseEvent6createERKN3WTF12AtomicStringENS1_10PassRefPtrINS_9DOMWindowEEERKNS_18PlatformMouseEventEiNS5_INS_4NodeEEE __ZN7WebCore10MouseEventC1ERKN3WTF12AtomicStringEbbNS1_10PassRefPtrINS_9DOMWindowEEEitNS5_INS_11EventTargetEEENS5_INS_9ClipboardEEEb +__ZN7WebCore10RenderView10compositorEv __ZN7WebCore10ShadowTree20removeAllShadowRootsEv __ZN7WebCore10ScrollView17setUseFixedLayoutEb __ZN7WebCore10ScrollView18setFixedLayoutSizeERKNS_7IntSizeE @@ -1347,6 +1348,8 @@ __ZNK7WebCore21BackForwardController12forwardCountEv __ZNK7WebCore21BackForwardController9backCountEv __ZNK7WebCore21UserContentURLPattern7matchesERKNS_4KURLE +__ZNK7WebCore21RenderLayerCompositor11scrollLayerEv +__ZNK7WebCore21RenderLayerCompositor15rootRenderLayerEv __ZNK7WebCore23AuthenticationChallenge20authenticationClientEv __ZNK7WebCore23FrameLoaderStateMachine15firstLayoutDoneEv __ZNK7WebCore23FrameLoaderStateMachine23committingFirstRealLoadEv Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h (109854 => 109855) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h 2012-03-06 05:20:32 UTC (rev 109854) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h 2012-03-06 05:23:04 UTC (rev 109855) @@ -343,7 +343,8 @@ virtual void setNeedsDisplay() = 0; // mark the given rect (in layer coords) as needing dispay. Never goes deep. virtual void setNeedsDisplayInRect(const FloatRect&) = 0; - +virtual bool needsDisplay() const { return false; } + virtual void setContentsNeedsDisplay() { }; // Set that the position/size of the contents (image or video). Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h (109854 => 109855) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h 2012-03-06 05:20:32 UTC (rev 109854) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h 2012-03-06 05:23:04 UTC (rev 109855) @@ -103,6 +103,7 @@ virtual void setNeedsDisplay(); virtual void setNeedsDisplayInRect(const FloatRect&); +virtual bool needsDisplay() const OVERRIDE { return !m_dirtyRects.isEmpty(); } virtual void setContentsNeedsDisplay(); virtual void setContentsRect(const IntRect&); Modified: tags/Safari-536.2.1/Source/WebKit2/ChangeLog (109854 => 109855) --- tags/Safari-536.2.1/Source/WebKit2/ChangeLog 2012-03-06 05:20:32 UTC (rev 109854) +++ tags/Safari-536.2.1/Source/WebKit2/ChangeLog 2012-03-06 05:23:04 UTC (rev 109855) @@ -1,5 +1,28 @@ 2011-03-05 Lucas Forschler +Merge 109812 + +2012-03-05 Anders Carlsson + +Be more aggressive about repainting page overlays +https://bugs.webkit.org/show_bug.cgi?id=80336 + + +Reviewed by Simon Fraser. + +Whenever we're flushing layers and we have a page overlay, check if the main frame +has scrolled or if the main frame root content layer needs to be repainted and force the +overlay layer to be repainted i
[webkit-changes] [109854] tags/Safari-536.2.1/Source
Title: [109854] tags/Safari-536.2.1/Source Revision 109854 Author lforsch...@apple.com Date 2012-03-05 21:20:32 -0800 (Mon, 05 Mar 2012) Log Message Merged r109801. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.cpp tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.h tags/Safari-536.2.1/Source/WebKit2/ChangeLog tags/Safari-536.2.1/Source/WebKit2/WebProcess/WebPage/mac/TiledCoreAnimationDrawingArea.mm Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109853 => 109854) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:19:00 UTC (rev 109853) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:20:32 UTC (rev 109854) @@ -1,5 +1,27 @@ 2011-03-05 Lucas Forschler +Merge 109801 + +2012-03-05 Anders Carlsson + +Always update the scroll layer position on the main thread when we have an overlay +https://bugs.webkit.org/show_bug.cgi?id=80324 + +Reviewed by Sam Weinig. + +Add a way to ensure that scroll layer position updates happen on the main thread. + +* WebCore.exp.in: +* page/scrolling/ScrollingCoordinator.cpp: +(WebCore::ScrollingCoordinator::ScrollingCoordinator): +(WebCore::ScrollingCoordinator::updateShouldUpdateScrollLayerPositionOnMainThread): +(WebCore): +(WebCore::ScrollingCoordinator::setForceMainThreadScrollLayerPositionUpdates): +* page/scrolling/ScrollingCoordinator.h: +(ScrollingCoordinator): + +2011-03-05 Lucas Forschler + Merge 109788 2012-03-05 Anders Carlsson Modified: tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in (109853 => 109854) --- tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in 2012-03-06 05:19:00 UTC (rev 109853) +++ tags/Safari-536.2.1/Source/WebCore/WebCore.exp.in 2012-03-06 05:20:32 UTC (rev 109854) @@ -2100,6 +2100,7 @@ __ZN7WebCore13ScrollingTree22updateBackForwardStateEbb __ZN7WebCore13ScrollingTreeD1Ev __ZN7WebCore15ScrollingThread8dispatchERKN3WTF8FunctionIFvvEEE +__ZN7WebCore20ScrollingCoordinator44setForceMainThreadScrollLayerPositionUpdatesEb __ZN7WebCore4Page20scrollingCoordinatorEv __ZNK7WebCore20ScrollingCoordinator13scrollingTreeEv #endif Modified: tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.cpp (109853 => 109854) --- tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.cpp 2012-03-06 05:19:00 UTC (rev 109853) +++ tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.cpp 2012-03-06 05:20:32 UTC (rev 109854) @@ -53,6 +53,7 @@ ScrollingCoordinator::ScrollingCoordinator(Page* page) : m_page(page) +, m_forceMainThreadScrollLayerPositionUpdates(false) #if ENABLE(THREADED_SCROLLING) , m_scrollingTreeState(ScrollingTreeState::create()) , m_scrollingTree(ScrollingTree::create(this)) @@ -310,9 +311,18 @@ FrameView* frameView = m_page->mainFrame()->view(); // FIXME: Having fixed objects on the page should not trigger the slow path. -setShouldUpdateScrollLayerPositionOnMainThread(frameView->hasSlowRepaintObjects() || frameView->hasFixedObjects()); +setShouldUpdateScrollLayerPositionOnMainThread(m_forceMainThreadScrollLayerPositionUpdates || frameView->hasSlowRepaintObjects() || frameView->hasFixedObjects()); } +void ScrollingCoordinator::setForceMainThreadScrollLayerPositionUpdates(bool forceMainThreadScrollLayerPositionUpdates) +{ +if (m_forceMainThreadScrollLayerPositionUpdates == forceMainThreadScrollLayerPositionUpdates) +return; + +m_forceMainThreadScrollLayerPositionUpdates = forceMainThreadScrollLayerPositionUpdates; +updateShouldUpdateScrollLayerPositionOnMainThread(); +} + #if ENABLE(THREADED_SCROLLING) void ScrollingCoordinator::setScrollLayer(GraphicsLayer* scrollLayer) { Modified: tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.h (109853 => 109854) --- tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.h 2012-03-06 05:19:00 UTC (rev 109853) +++ tags/Safari-536.2.1/Source/WebCore/page/scrolling/ScrollingCoordinator.h 2012-03-06 05:20:32 UTC (rev 109854) @@ -112,6 +112,9 @@ void handleWheelEventPhase(PlatformWheelEventPhase); #endif +// Force all scroll layer position updates to happen on the main thread. +void setForceMainThreadScrollLayerPositionUpdates(bool); + private: explicit ScrollingCoordinator(Page*); @@ -128,6 +131,8 @@ Page* m_page; +bool m_forceMainThreadScrollLayerPositionUpdates; + #if ENABLE(THREADED_SCROLLING) void scheduleTreeStateCommit(); Modified: tags/Safari-536.2.1/Source/WebKit2/ChangeLog (109853 => 109854) --- tags/Safari-536.2.1/Source/WebKit2/ChangeLog 2012-03-06 05:19:00 UTC (rev 109853) +++ tags/Safari-536.2.1/Source/WebKit2/ChangeL
[webkit-changes] [109853] tags/Safari-536.2.1/Source/WebCore
Title: [109853] tags/Safari-536.2.1/Source/WebCore Revision 109853 Author lforsch...@apple.com Date 2012-03-05 21:19:00 -0800 (Mon, 05 Mar 2012) Log Message Merged r109788. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.cpp tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/PlatformCALayer.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/PlatformCALayerMac.mm tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/TileCache.mm tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.h tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/win/PlatformCALayerWin.cpp tags/Safari-536.2.1/Source/WebCore/rendering/RenderLayerCompositor.cpp Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109852 => 109853) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:15:47 UTC (rev 109852) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:19:00 UTC (rev 109853) @@ -1,5 +1,44 @@ 2011-03-05 Lucas Forschler +Merge 109788 + +2012-03-05 Anders Carlsson + +Let RenderLayerCompositor set the tile cache visible rect +https://bugs.webkit.org/show_bug.cgi?id=80317 + +Reviewed by Simon Fraser. + +We can't compute the visible rect from CALayers, because that breaks when we're updating +the scroll layer position on the main thread (since by the time visibleRectChanged() is called, +the CALayers won't yet have been updated). + +* platform/graphics/GraphicsLayer.h: +(WebCore::GraphicsLayer::visibleRectChanged): +* platform/graphics/ca/GraphicsLayerCA.cpp: +(WebCore::GraphicsLayerCA::visibleRectChanged): +* platform/graphics/ca/GraphicsLayerCA.h: +(GraphicsLayerCA): +* platform/graphics/ca/PlatformCALayer.h: +(PlatformCALayer): +* platform/graphics/ca/mac/PlatformCALayerMac.mm: +(PlatformCALayer::visibleRectChanged): +* platform/graphics/ca/mac/TileCache.h: +(TileCache): +* platform/graphics/ca/mac/TileCache.mm: +(WebCore::TileCache::visibleRectChanged): +(WebCore::TileCache::revalidateTiles): +* platform/graphics/ca/mac/WebTileCacheLayer.h: +(WebCore): +* platform/graphics/ca/mac/WebTileCacheLayer.mm: +(-[WebTileCacheLayer visibleRectChanged:]): +* platform/graphics/ca/win/PlatformCALayerWin.cpp: +(PlatformCALayer::visibleRectChanged): +* rendering/RenderLayerCompositor.cpp: +(WebCore::RenderLayerCompositor::frameViewDidScroll): + +2011-03-05 Lucas Forschler + Merge 109594 2012-03-02 Maciej Stachowiak Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h (109852 => 109853) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h 2012-03-06 05:15:47 UTC (rev 109852) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/GraphicsLayer.h 2012-03-06 05:19:00 UTC (rev 109853) @@ -428,7 +428,7 @@ bool usingTiledLayer() const { return m_usingTiledLayer; } // Called whenever the visible rect of the given GraphicsLayer changed. -virtual void visibleRectChanged() { } +virtual void visibleRectChanged(const IntRect&) { } #if PLATFORM(QT) || PLATFORM(GTK) // This allows several alternative GraphicsLayer implementations in the same port, Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.cpp (109852 => 109853) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.cpp 2012-03-06 05:15:47 UTC (rev 109852) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.cpp 2012-03-06 05:19:00 UTC (rev 109853) @@ -859,9 +859,9 @@ commitLayerChangesAfterSublayers(); } -void GraphicsLayerCA::visibleRectChanged() +void GraphicsLayerCA::visibleRectChanged(const IntRect& visibleRect) { -m_layer->visibleRectChanged(); +m_layer->visibleRectChanged(visibleRect); } void GraphicsLayerCA::recursiveCommitChanges(const TransformState& state, float pageScaleFactor, const FloatPoint& positionRelativeToBase, bool affectedByPageScale) Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h (109852 => 109853) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h 2012-03-06 05:15:47 UTC (rev 109852) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/GraphicsLayerCA.h 2012-03-06 05:19:00 UTC (rev 109853) @@
[webkit-changes] [109852] tags/Safari-536.2.1
Title: [109852] tags/Safari-536.2.1 Revision 109852 Author lforsch...@apple.com Date 2012-03-05 21:15:47 -0800 (Mon, 05 Mar 2012) Log Message Merged r109594. Modified Paths tags/Safari-536.2.1/LayoutTests/ChangeLog tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/bindings/js/JSDOMBinding.cpp tags/Safari-536.2.1/Source/WebCore/bindings/js/JSDOMWindowCustom.cpp tags/Safari-536.2.1/Source/WebCore/page/DOMWindow.idl tags/Safari-536.2.1/Source/WebCore/platform/sql/SQLiteStatement.cpp Added Paths tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties-expected.txt tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties.html tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe1.html tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe2.html Diff Modified: tags/Safari-536.2.1/LayoutTests/ChangeLog (109851 => 109852) --- tags/Safari-536.2.1/LayoutTests/ChangeLog 2012-03-06 05:13:00 UTC (rev 109851) +++ tags/Safari-536.2.1/LayoutTests/ChangeLog 2012-03-06 05:15:47 UTC (rev 109852) @@ -1,3 +1,19 @@ +2011-03-05 Lucas Forschler + +Merge 109594 + +2012-03-02 Maciej Stachowiak + +REGRESSION(r97353): Crash when accessing location or history properties inside a navigated window +https://bugs.webkit.org/show_bug.cgi?id=80133 + +Reviewed by Antti Koivisto. + +* fast/dom/Window/navigated-window-properties-expected.txt: Added. +* fast/dom/Window/navigated-window-properties.html: Added. +* fast/dom/Window/resources/navigated-window-prop-subframe1.html: Added. +* fast/dom/Window/resources/navigated-window-prop-subframe2.html: Added. + 2012-02-29 Adam Klein Rebaseline line-grid tests for Chromium after r109267. Copied: tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties-expected.txt (from rev 109594, trunk/LayoutTests/fast/dom/Window/navigated-window-properties-expected.txt) (0 => 109852) --- tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties-expected.txt (rev 0) +++ tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties-expected.txt 2012-03-06 05:15:47 UTC (rev 109852) @@ -0,0 +1,3 @@ +Accessing properties of a navigated window shouldn't crash + + Copied: tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties.html (from rev 109594, trunk/LayoutTests/fast/dom/Window/navigated-window-properties.html) (0 => 109852) --- tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties.html (rev 0) +++ tags/Safari-536.2.1/LayoutTests/fast/dom/Window/navigated-window-properties.html 2012-03-06 05:15:47 UTC (rev 109852) @@ -0,0 +1,26 @@ + +Properties of a navigated window shouldn't crash + +Accessing properties of a navigated window shouldn't crash + + +if (window.layoutTestController) { +layoutTestController.waitUntilDone(); +layoutTestController.dumpAsText(); +} + + +var func; +function testPhase1() { +var subframe = document.getElementById("subframe"); +func = subframe.contentWindow.testFunction; +subframe.src = "" +} + +function testPhase2() { +func(); +if (window.layoutTestController) +layoutTestController.notifyDone(); +} + + Copied: tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe1.html (from rev 109594, trunk/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe1.html) (0 => 109852) --- tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe1.html (rev 0) +++ tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe1.html 2012-03-06 05:15:47 UTC (rev 109852) @@ -0,0 +1,9 @@ + +function testFunction() { +try { + var s = location.href + history.length; +} catch (e) { +} +} +parent.testPhase1(); + Copied: tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe2.html (from rev 109594, trunk/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe2.html) (0 => 109852) --- tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe2.html (rev 0) +++ tags/Safari-536.2.1/LayoutTests/fast/dom/Window/resources/navigated-window-prop-subframe2.html 2012-03-06 05:15:47 UTC (rev 109852) @@ -0,0 +1,3 @@ + +parent.testPhase2(); + Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109851 => 109852) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:13:00 UTC (rev 109851) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:15:47 UTC (rev 109852) @@ -1,5 +1,28 @@ 2011-03-05 Lucas Forschler +Merge 109594 + +2012-03-02 Maciej Stachowiak + +REGRESSION(r97353): Crash when
[webkit-changes] [109851] trunk
Title: [109851] trunk Revision 109851 Author e...@google.com Date 2012-03-05 21:13:00 -0800 (Mon, 05 Mar 2012) Log Message Compositing overlap testing can throw layers into compositing when they should not be. https://bugs.webkit.org/show_bug.cgi?id=50192 Reviewed by Simon Fraser. Source/WebCore: The previous overlap map behavior was that a non-composited query layer would become composited due to overlap if and only if the query layer's absolute bounds overlapped the absolute bounds of some other layer which: - draws before the query layer - is or has a compositing ancestor This behavior, while correct, was too permissive in throwing layers into compositing, causing many layers to get their own backing when they could have just gone into their compositing ancestor's backing. The correct logic is that non-composited query layer needs to be composited due to overlap if and only if the query layer's absolute bounds overlap the absolute bounds of some other layer which: - draws before the query layer - has a different compositing ancestor than the query layer - is or has a compositing ancestor that is a descendent of the query layer's compositing ancestor This patch changes the semantics of the overlap map to enable this behavior. Rather than having one global overlap map, there is now a stack of overlap maps. New (empty) overlap maps are pushed onto the stack whenever a layer becomes a compositing ancestor and popped after all of the compositing requirements for that layer's children have been computed. The compositing ancestor and all of its non-composited children of a compositing ancestor do not get considered for overlap until their composited ancestor has been popped off the stack. If a compositing ancestor has a compositing subtree, then any descendents of that compositing ancestor that draw after that subtree will consider everything in the compositing subtree for overlap. Test: compositing/layer-creation/stacking-context-overlap.html * platform/graphics/Region.cpp: (WebCore::Region::intersects): (WebCore): * platform/graphics/Region.h: (Region): * rendering/RenderLayerCompositor.cpp: (RenderLayerCompositor::OverlapMap): (WebCore::RenderLayerCompositor::OverlapMap::OverlapMap): (WebCore::RenderLayerCompositor::OverlapMap::add): (WebCore::RenderLayerCompositor::OverlapMap::contains): (WebCore::RenderLayerCompositor::OverlapMap::overlapsLayers): (WebCore::RenderLayerCompositor::OverlapMap::isEmpty): (WebCore::RenderLayerCompositor::OverlapMap::popCompositingContainer): (WebCore::RenderLayerCompositor::OverlapMap::pushCompositingContainer): (WebCore::RenderLayerCompositor::addToOverlapMapRecursive): (WebCore::RenderLayerCompositor::computeCompositingRequirements): * rendering/RenderLayerCompositor.h: (RenderLayerCompositor): LayoutTests: * compositing/layer-creation/stacking-context-overlap-expected.txt: Added. * compositing/layer-creation/stacking-context-overlap.html: Added. * compositing/layer-creation/stacking-context-overlap-nested-expected.txt: Added. * compositing/layer-creation/stacking-context-overlap-nested.html: Added. * platform/chromium/test_expectations.txt: Modified Paths trunk/LayoutTests/ChangeLog trunk/LayoutTests/platform/chromium/test_expectations.txt trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/platform/graphics/Region.cpp trunk/Source/WebCore/platform/graphics/Region.h trunk/Source/WebCore/rendering/RenderLayerCompositor.cpp trunk/Source/WebCore/rendering/RenderLayerCompositor.h Added Paths trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap-expected.txt trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap-nested-expected.txt trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap-nested.html trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap.html Diff Modified: trunk/LayoutTests/ChangeLog (109850 => 109851) --- trunk/LayoutTests/ChangeLog 2012-03-06 05:09:33 UTC (rev 109850) +++ trunk/LayoutTests/ChangeLog 2012-03-06 05:13:00 UTC (rev 109851) @@ -1,3 +1,16 @@ +2012-03-05 Adrienne Walker + +Compositing overlap testing can throw layers into compositing when they should not be. +https://bugs.webkit.org/show_bug.cgi?id=50192 + +Reviewed by Simon Fraser. + +* compositing/layer-creation/stacking-context-overlap-expected.txt: Added. +* compositing/layer-creation/stacking-context-overlap.html: Added. +* compositing/layer-creation/stacking-context-overlap-nested-expected.txt: Added. +* compositing/layer-creation/stacking-context-overlap-nested.html: Added. +* platform/chromium/test_expectations.txt: + 2012-03-05 Yoshifumi Inoue [Forms] HTMLFieldSetForms.idl doesn't have type attribute. Added: trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap-expected.txt (0 => 109851) --- trunk/LayoutTests/compositing/layer-creation/stacking-context-overlap-expected.txt
[webkit-changes] [109850] tags/Safari-536.2.1/Source/WebCore
Title: [109850] tags/Safari-536.2.1/Source/WebCore Revision 109850 Author lforsch...@apple.com Date 2012-03-05 21:09:33 -0800 (Mon, 05 Mar 2012) Log Message Merged r109486. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109849 => 109850) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:07:49 UTC (rev 109849) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:09:33 UTC (rev 109850) @@ -1,5 +1,28 @@ 2011-03-05 Lucas Forschler +Merge 109486 + +2012-03-01 Anders Carlsson + +Crash when doing repeated double-tap-to-zoom gesture on apple startpage +https://bugs.webkit.org/show_bug.cgi?id=80081 + + +Reviewed by Sam Weinig. + +Setting the contents scale on the tile cache layer was creating a WebTileCacheLayer presentation layer copy with a null _tileCache. + +Implement -[WebTileCacheLayer actionForKey:] and have it always return nil so we'll avoid implicit animations, and thus creating presentation layers. + +* platform/graphics/ca/mac/WebTileCacheLayer.mm: +(-[WebTileCacheLayer initWithLayer:]): +Implement this and assert that it's never reached. We should never create presentation layers since we don't animate this layer. + +(-[WebTileCacheLayer actionForKey:]): +Implement this and always return nil. + +2011-03-05 Lucas Forschler + Merge 109379 2012-03-01 Beth Dakin Modified: tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm (109849 => 109850) --- tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm 2012-03-06 05:07:49 UTC (rev 109849) +++ tags/Safari-536.2.1/Source/WebCore/platform/graphics/ca/mac/WebTileCacheLayer.mm 2012-03-06 05:09:33 UTC (rev 109850) @@ -58,6 +58,22 @@ [super dealloc]; } +- (id)initWithLayer:(id)layer +{ +UNUSED_PARAM(layer); + +ASSERT_NOT_REACHED(); +return nil; +} + +- (id)actionForKey:(NSString *)key +{ +UNUSED_PARAM(key); + +// Disable all animations. +return nil; +} + - (void)setBounds:(CGRect)bounds { [super setBounds:bounds]; ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109849] tags/Safari-536.2.1/Source
Title: [109849] tags/Safari-536.2.1/Source Revision 109849 Author lforsch...@apple.com Date 2012-03-05 21:07:49 -0800 (Mon, 05 Mar 2012) Log Message Versioning. Modified Paths tags/Safari-536.2.1/Source/_javascript_Core/Configurations/Version.xcconfig tags/Safari-536.2.1/Source/WebCore/Configurations/Version.xcconfig tags/Safari-536.2.1/Source/WebKit/mac/Configurations/Version.xcconfig tags/Safari-536.2.1/Source/WebKit2/Configurations/Version.xcconfig Diff Modified: tags/Safari-536.2.1/Source/_javascript_Core/Configurations/Version.xcconfig (109848 => 109849) --- tags/Safari-536.2.1/Source/_javascript_Core/Configurations/Version.xcconfig 2012-03-06 05:04:45 UTC (rev 109848) +++ tags/Safari-536.2.1/Source/_javascript_Core/Configurations/Version.xcconfig 2012-03-06 05:07:49 UTC (rev 109849) @@ -23,8 +23,8 @@ MAJOR_VERSION = 536; MINOR_VERSION = 2; -TINY_VERSION = 0; -FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION); +TINY_VERSION = 1; +FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION); // The bundle version and short version string are set based on the current build configuration, see below. BUNDLE_VERSION = $(BUNDLE_VERSION_$(CONFIGURATION)); Modified: tags/Safari-536.2.1/Source/WebCore/Configurations/Version.xcconfig (109848 => 109849) --- tags/Safari-536.2.1/Source/WebCore/Configurations/Version.xcconfig 2012-03-06 05:04:45 UTC (rev 109848) +++ tags/Safari-536.2.1/Source/WebCore/Configurations/Version.xcconfig 2012-03-06 05:07:49 UTC (rev 109849) @@ -23,8 +23,8 @@ MAJOR_VERSION = 536; MINOR_VERSION = 2; -TINY_VERSION = 0; -FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION); +TINY_VERSION = 1; +FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION); // The bundle version and short version string are set based on the current build configuration, see below. BUNDLE_VERSION = $(BUNDLE_VERSION_$(CONFIGURATION)); Modified: tags/Safari-536.2.1/Source/WebKit/mac/Configurations/Version.xcconfig (109848 => 109849) --- tags/Safari-536.2.1/Source/WebKit/mac/Configurations/Version.xcconfig 2012-03-06 05:04:45 UTC (rev 109848) +++ tags/Safari-536.2.1/Source/WebKit/mac/Configurations/Version.xcconfig 2012-03-06 05:07:49 UTC (rev 109849) @@ -23,8 +23,8 @@ MAJOR_VERSION = 536; MINOR_VERSION = 2; -TINY_VERSION = 0; -FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION); +TINY_VERSION = 1; +FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION); // The bundle version and short version string are set based on the current build configuration, see below. BUNDLE_VERSION = $(BUNDLE_VERSION_$(CONFIGURATION)); Modified: tags/Safari-536.2.1/Source/WebKit2/Configurations/Version.xcconfig (109848 => 109849) --- tags/Safari-536.2.1/Source/WebKit2/Configurations/Version.xcconfig 2012-03-06 05:04:45 UTC (rev 109848) +++ tags/Safari-536.2.1/Source/WebKit2/Configurations/Version.xcconfig 2012-03-06 05:07:49 UTC (rev 109849) @@ -23,8 +23,8 @@ MAJOR_VERSION = 536; MINOR_VERSION = 2; -TINY_VERSION = 0; -FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION); +TINY_VERSION = 1; +FULL_VERSION = $(MAJOR_VERSION).$(MINOR_VERSION).$(TINY_VERSION); // The bundle version and short version string are set based on the current build configuration, see below. BUNDLE_VERSION = $(BUNDLE_VERSION_$(CONFIGURATION)); ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109848] tags/Safari-536.2.1/Source/WebCore
Title: [109848] tags/Safari-536.2.1/Source/WebCore Revision 109848 Author lforsch...@apple.com Date 2012-03-05 21:04:45 -0800 (Mon, 05 Mar 2012) Log Message Merged r109379. Modified Paths tags/Safari-536.2.1/Source/WebCore/ChangeLog tags/Safari-536.2.1/Source/WebCore/page/Page.cpp Diff Modified: tags/Safari-536.2.1/Source/WebCore/ChangeLog (109847 => 109848) --- tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 04:57:27 UTC (rev 109847) +++ tags/Safari-536.2.1/Source/WebCore/ChangeLog 2012-03-06 05:04:45 UTC (rev 109848) @@ -1,3 +1,18 @@ +2011-03-05 Lucas Forschler + +Merge 109379 + +2012-03-01 Beth Dakin + +Reviewed by Kevin Decker. + +Part of https://bugs.webkit.org/show_bug.cgi?id=79705 + +Here's some of that tweaking promised in the comment. Adjusting +gMaximumUnpaintedAreaRatio down a bit. +* page/Page.cpp: +(WebCore): + 2011-03-02 Lucas Forschler Merge 109340 Modified: tags/Safari-536.2.1/Source/WebCore/page/Page.cpp (109847 => 109848) --- tags/Safari-536.2.1/Source/WebCore/page/Page.cpp 2012-03-06 04:57:27 UTC (rev 109847) +++ tags/Safari-536.2.1/Source/WebCore/page/Page.cpp 2012-03-06 05:04:45 UTC (rev 109848) @@ -1018,7 +1018,7 @@ // These are magical constants that might be tweaked over time. static double gMinimumPaintedAreaRatio = 0.1; -static double gMaximumUnpaintedAreaRatio = 0.1; +static double gMaximumUnpaintedAreaRatio = 0.04; void Page::setRelevantRepaintedObjectsCounterThreshold(uint64_t threshold) { ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109847] trunk/Source/WebCore
Title: [109847] trunk/Source/WebCore Revision 109847 Author ander...@apple.com Date 2012-03-05 20:57:27 -0800 (Mon, 05 Mar 2012) Log Message Address review comments from https://bugs.webkit.org/show_bug.cgi?id=80368 * platform/graphics/ca/mac/TileCache.mm: (WebCore::TileCache::setNeedsDisplayInRect): Modified Paths trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/platform/graphics/ca/mac/TileCache.mm Diff Modified: trunk/Source/WebCore/ChangeLog (109846 => 109847) --- trunk/Source/WebCore/ChangeLog 2012-03-06 04:54:49 UTC (rev 109846) +++ trunk/Source/WebCore/ChangeLog 2012-03-06 04:57:27 UTC (rev 109847) @@ -1,3 +1,10 @@ +2012-03-05 Anders Carlsson + +Address review comments from https://bugs.webkit.org/show_bug.cgi?id=80368 + +* platform/graphics/ca/mac/TileCache.mm: +(WebCore::TileCache::setNeedsDisplayInRect): + 2012-03-05 Yoshifumi Inoue [Forms] HTMLFieldSetForms.idl doesn't have type attribute. Modified: trunk/Source/WebCore/platform/graphics/ca/mac/TileCache.mm (109846 => 109847) --- trunk/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 04:54:49 UTC (rev 109846) +++ trunk/Source/WebCore/platform/graphics/ca/mac/TileCache.mm 2012-03-06 04:57:27 UTC (rev 109847) @@ -91,13 +91,13 @@ if (m_tiles.isEmpty()) return; -IntRect scaledRect = rect; +FloatRect scaledRect(rect); scaledRect.scale(m_scale); // Find the tiles that need to be invalidated. TileIndex topLeft; TileIndex bottomRight; -getTileIndexRangeForRect(intersection(scaledRect, m_tileCoverageRect), topLeft, bottomRight); +getTileIndexRangeForRect(intersection(enclosingIntRect(scaledRect), m_tileCoverageRect), topLeft, bottomRight); for (int y = topLeft.y(); y <= bottomRight.y(); ++y) { for (int x = topLeft.x(); x <= bottomRight.x(); ++x) { ___ webkit-changes mailing list webkit-changes@lists.webkit.org http://lists.webkit.org/mailman/listinfo.cgi/webkit-changes
[webkit-changes] [109846] trunk
Title: [109846] trunk Revision 109846 Author commit-qu...@webkit.org Date 2012-03-05 20:54:49 -0800 (Mon, 05 Mar 2012) Log Message [Forms] HTMLFieldSetForms.idl doesn't have type attribute. https://bugs.webkit.org/show_bug.cgi?id=80109 Patch by Yoshifumi Inoue on 2012-03-05 Reviewed by Hajime Morita. Source/WebCore: Test: fast/forms/fieldset/fieldset-type.html * html/HTMLFieldSetElement.idl: Add attribute "type" LayoutTests: * fast/forms/fieldset/fieldset-type-expected.txt: Added. * fast/forms/fieldset/fieldset-type.html: Added. Modified Paths trunk/LayoutTests/ChangeLog trunk/Source/WebCore/ChangeLog trunk/Source/WebCore/html/HTMLFieldSetElement.idl Added Paths trunk/LayoutTests/fast/forms/fieldset/fieldset-type-expected.txt trunk/LayoutTests/fast/forms/fieldset/fieldset-type.html Diff Modified: trunk/LayoutTests/ChangeLog (109845 => 109846) --- trunk/LayoutTests/ChangeLog 2012-03-06 04:54:19 UTC (rev 109845) +++ trunk/LayoutTests/ChangeLog 2012-03-06 04:54:49 UTC (rev 109846) @@ -1,3 +1,13 @@ +2012-03-05 Yoshifumi Inoue + +[Forms] HTMLFieldSetForms.idl doesn't have type attribute. +https://bugs.webkit.org/show_bug.cgi?id=80109 + +Reviewed by Hajime Morita. + +* fast/forms/fieldset/fieldset-type-expected.txt: Added. +* fast/forms/fieldset/fieldset-type.html: Added. + 2012-03-05 Kenichi Ishibashi [WebSocket] Should raise SYNTAX_ERR when message contains unpaired surrogates Added: trunk/LayoutTests/fast/forms/fieldset/fieldset-type-expected.txt (0 => 109846) --- trunk/LayoutTests/fast/forms/fieldset/fieldset-type-expected.txt (rev 0) +++ trunk/LayoutTests/fast/forms/fieldset/fieldset-type-expected.txt 2012-03-06 04:54:49 UTC (rev 109846) @@ -0,0 +1,10 @@ +This test checks fieldset.test attribute + +On success, you will see a series of "PASS" messages, followed by "TEST COMPLETE". + + +PASS fs1.type is "fieldset" +PASS successfullyParsed is true + +TEST COMPLETE + Added: trunk/LayoutTests/fast/forms/fieldset/fieldset-type.html (0 => 109846) --- trunk/LayoutTests/fast/forms/fieldset/fieldset-type.html (rev 0) +++ trunk/LayoutTests/fast/forms/fieldset/fieldset-type.html 2012-03-06 04:54:49 UTC (rev 109846) @@ -0,0 +1,8 @@ + +