Timeline



May 21, 2014:

11:44 PM Changeset in webkit [169190] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

[Win64] Unreviewed build fix.

  • WebKit.vcxproj/WebKitExportGenerator/WebKitExports.def.in:

Corrected 64-bit linker symbol.

11:12 PM Changeset in webkit [169189] by Antti Koivisto
  • 6 edits
    2 adds in trunk

REGRESSION(r167870): Crash in simple line layout code with :after
​https://bugs.webkit.org/show_bug.cgi?id=133155

Source/WebCore:
<rdar://problem/16977696>

Reviewed by Darin Adler.

Fix ​https://bugs.webkit.org/show_bug.cgi?id=132241 in a safer way.
The underline behavior is tested by the existing fast/text/simple-lines-hover-underline.html

Test: fast/text/simple-lines-hover-after.html

  • rendering/RenderBlock.cpp:

(WebCore::RenderBlock::invalidateLineLayoutPath): Deleted.

Move to RenderBlockFlow.

  • rendering/RenderBlock.h:

(WebCore::RenderBlock::invalidateLineLayoutPath):

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::styleDidChange):

Invalidate layout if style changes in a manner that makes us ineligible to use the simple line layout path.

(WebCore::RenderBlockFlow::invalidateLineLayoutPath):

Drop the simple line layout on path invalidation if it exists. It may not be valid anymore.
Also invalidate the layout if this happens so we'll reconstruct the lines later.

(WebCore::RenderBlockFlow::simpleLineLayout): Deleted.
(WebCore::RenderBlockFlow::ensureLineBoxes):
(WebCore::RenderBlockFlow::createLineBoxes): Deleted.

Revert some of the changes made it r167870.

  • rendering/RenderBlockFlow.h:

(WebCore::RenderBlockFlow::simpleLineLayout):

Add strong validity assert.

LayoutTests:

Reviewed by Darin Adler.

  • fast/text/simple-lines-hover-after-expected.html: Added.
  • fast/text/simple-lines-hover-after.html: Added.
9:11 PM Changeset in webkit [169188] by fpizlo@apple.com
  • 8 edits
    5 adds in branches/ftlopt

[ftlopt] DFG::clobberize should be blind to the effects of GC
​https://bugs.webkit.org/show_bug.cgi?id=133166

Source/JavaScriptCore:
Reviewed by Goeffrey Garen.

Move the computation of where GCs happen to DFG::doesGC().

Large (>5x) speed-up on programs that do loop-invariant string concatenations.

  • CMakeLists.txt:
  • JavaScriptCore.vcxproj/JavaScriptCore.vcxproj:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • dfg/DFGAbstractHeap.h:
  • dfg/DFGClobberize.h:

(JSC::DFG::clobberize):
(JSC::DFG::clobberizeForAllocation): Deleted.

  • dfg/DFGDoesGC.cpp: Added.

(JSC::DFG::doesGC):

  • dfg/DFGDoesGC.h: Added.
  • dfg/DFGStoreBarrierElisionPhase.cpp:

(JSC::DFG::StoreBarrierElisionPhase::handleNode):
(JSC::DFG::StoreBarrierElisionPhase::couldCauseGC): Deleted.

LayoutTests:
Reviewed by Geoffrey Garen.

  • js/regress/hoist-make-rope-expected.txt: Added.
  • js/regress/hoist-make-rope.html: Added.
  • js/regress/script-tests/hoist-make-rope.js: Added.

(foo):

8:52 PM Changeset in webkit [169187] by matthew_hanson@apple.com
  • 5 edits in branches/safari-538.34-branch/Source

Versioning.

8:25 PM Changeset in webkit [169186] by matthew_hanson@apple.com
  • 1 copy in tags/Safari-538.34.45

New Tag.

6:55 PM Changeset in webkit [169185] by fpizlo@apple.com
  • 2 edits in branches/ftlopt/Source/JavaScriptCore

Merge trunk r169184 to ftlopt.

2014-05-21 Filip Pizlo <​fpizlo@apple.com>


Store barrier elision should run after DCE in both the DFG path and the FTL path
​https://bugs.webkit.org/show_bug.cgi?id=129718


Rubber stamped by Mark Hahnenberg.


  • dfg/DFGPlan.cpp: (JSC::DFG::Plan::compileInThreadImpl):
6:36 PM Changeset in webkit [169184] by fpizlo@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

Store barrier elision should run after DCE in both the DFG path and the FTL path
​https://bugs.webkit.org/show_bug.cgi?id=129718

Rubber stamped by Mark Hahnenberg.

  • dfg/DFGPlan.cpp:

(JSC::DFG::Plan::compileInThreadImpl):

6:18 PM Changeset in webkit [169183] by eric.carlson@apple.com
  • 2 edits in trunk/Source/WebCore

[iOS] two media control button strings are not localized
​https://bugs.webkit.org/show_bug.cgi?id=133160

Reviewed by Geoffrey Garen.

  • English.lproj/mediaControlsLocalizedStrings.js:
5:56 PM Changeset in webkit [169182] by matthew_hanson@apple.com
  • 2 edits in tags/Safari-538.36/Source/WebKit2

Merge r169180. <rdar://problem/16982059>

5:12 PM Changeset in webkit [169181] by commit-queue@webkit.org
  • 6 edits
    1 add in trunk

[EFL] Add include path of compact_unwind_encoding.h if FTL JIT is enabled
​https://bugs.webkit.org/show_bug.cgi?id=132907

Patch by Zsolt Borbely <​zsborbely.u-szeged@partner.samsung.com> on 2014-05-21
Reviewed by Gyuyoung Kim.

.:

  • Source/cmake/FindLIBCXXABI.cmake: Added.
  • Source/cmake/OptionsEfl.cmake:

Source/JavaScriptCore:

  • CMakeLists.txt:

Tools:
We need the mach-o/compact_unwind_encoding.h header from libc++abi-dev package
in JavaScriptCore/ftl/FTLUnwindInfo.cpp when FTL JIT is enabled.

  • efl/install-dependencies:
5:11 PM Changeset in webkit [169180] by oliver@apple.com
  • 2 edits in trunk/Source/WebKit2

Only enable sandbox extensions on mac.

3:03 PM Changeset in webkit [169179] by dburkart@apple.com
  • 7 edits in branches/safari-538.34-branch/Source

Merged r169151.

2:50 PM Changeset in webkit [169178] by Alan Bujtas
  • 2 edits in trunk/LayoutTests

Unreviewed Mac gardening after r169161.

  • platform/mac-wk2/tiled-drawing/tile-size-slow-zoomed-expected.txt:
2:35 PM Changeset in webkit [169177] by dino@apple.com
  • 2 edits
    1 add in trunk/Source/WebCore

[iOS] Update some of the media controls buttons to be bigger
​https://bugs.webkit.org/show_bug.cgi?id=133158
<rdar://problem/16475828>

Reviewed by Simon Fraser.

The buttons in the media controls on iOS are too small
to be comfortably pressed. Make them the size of the controls
panel and the expected iOS button size (44x44px).

The artwork for the buttons now includes the padding. I've
added a file from which you can edit the artwork, then
copy it into the CSS (because editing directly in CSS is
a huge pain).

So far just the play/pause and fullscreen button have been
updated. There will be a follow-up patch to fix the remaining
buttons and the scrubber.

  • Modules/mediacontrols/assets-apple-iOS.svg: Added.
  • Modules/mediacontrols/mediaControlsiOS.css:

(audio::-webkit-media-controls-panel):
(audio::-webkit-media-controls-fullscreen-button):
(audio::-webkit-media-controls-play-button):
(audio::-webkit-media-controls-play-button:active):
(audio::-webkit-media-controls-play-button.paused):
(audio::-webkit-media-controls-fullscreen-button:active):
(@media only screen and (-webkit-min-device-pixel-ratio: 2)): Deleted.

1:52 PM Changeset in webkit [169176] by oliver@apple.com
  • 5 edits
    2 adds in trunk/Source/WebKit2

[iOS] Enable sandboxing for the database process
​https://bugs.webkit.org/show_bug.cgi?id=132963

Reviewed by Alexey Proskuryakov.

Simply ensure that the database process is running in a tight
sandbox. As part of this i've separated out ChildProcessIOS
and ensures that the database process loads a custom sandbox.

  • Configurations/WebKit.xcconfig:
  • DatabaseProcess/ios/DatabaseProcessIOS.mm:

(WebKit::DatabaseProcess::initializeSandbox):

  • DatabaseProcess/ios/com.apple.WebKit.DatabasesIOS.sb: Added.
  • Shared/ios/ChildProcessIOS.mm: Added.

Essentially cloned from the Mac version, only all the OSX
specific logic has been stripped.
(WebKit::ChildProcess::setApplicationIsDaemon):
(WebKit::ChildProcess::platformInitialize):
(WebKit::ChildProcess::initializeSandbox):
(WebKit::ChildProcess::setQOS):

  • Shared/mac/ChildProcessMac.mm:

Remove all the IOS ifdefs.
(WebKit::ChildProcess::setApplicationIsDaemon):
(WebKit::ChildProcess::platformInitialize):
(WebKit::ChildProcess::initializeSandbox):
(WebKit::ChildProcess::setQOS):

  • WebKit2.xcodeproj/project.pbxproj:
1:39 PM Changeset in webkit [169175] by aestes@apple.com
  • 17 edits
    1 copy
    1 add in trunk/Source/WebKit2

[iOS] Create a UIPrintFormatter for WKWebView
​https://bugs.webkit.org/show_bug.cgi?id=133005

Reviewed by Tim Horton.

Implemented a UIPrintFormatter subclass to support printing WKWebViews using UIPrintInteractionController.
Overrode -_recalcPageCount: to synchronously ask the web process to lay out for printing and return the number
of page rects. The web process returns the page rects immediately and then starts rendering the page to a PDF,
returning the PDF data to the UI process asynchronously. When UIPrintInteractionController later calls our
overridden -drawInRect:forPageAtIndex:, we block and wait for the PDF data to be returned. Since
UIPrintInteractionController draws each page after a delay, it's possible that the web process has already
returned the PDF data to us by the time we print the first page.

  • Shared/WebPreferencesStore.h: Correctly initialized shouldPrintBackgrounds for iOS.
  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _dataForDisplayedPDF]): Now that WKPDFView vends its CGPDFDocument directly, extract its data here.
(-[WKWebView _printFormatterClass]): Tell UIView which UIPrintFormatter to alloc.
(-[WKWebView _computePageCountAndStartDrawingToPDFWithPrintInfo:WebKit::firstPage:computedTotalScaleFactor:]):
If we're displaying a PDF, return its page count directly. Otherwise, synchronously ask the web process to
compute page rects for printing and then return the number of rects.
(-[WKWebView _endPrinting]): Cleared state and sent the WebPage::EndPrinting message to the web process.
(-[WKWebView _printedDocument]): If we're displaying a PDF, return its data directly. Otherwise, if we haven't
received PDF data from the web process yet, wait for it indefinitely (there's no sensible timeout to use here
since failing to receive the data results in printing some number of blank pages).
(-[WKWebView _setPrintedDocument:]): Set the CGPDFDocument received from the web process.

  • UIProcess/API/Cocoa/WKWebViewInternal.h: Declared methods called by WKWebViewPrintFormatter.
  • UIProcess/PageClient.h: Declared didFinishDrawingPagesToPDF.
  • UIProcess/WKWebViewPrintFormatter.h: Added.
  • UIProcess/WKWebViewPrintFormatter.mm: Added.

(-[WKWebViewPrintFormatter dealloc]): Cleared state and called -[WKWebView _endPrinting].
(-[WKWebViewPrintFormatter webView]): Returned the WKWebView for this print formatter.
(-[WKWebViewPrintFormatter _recalcPageCount]): Set up a PrintInfo object given the formatter's content rect
and called -_computePageCountAndStartDrawingToPDFWithPrintInfo:...
(-[WKWebViewPrintFormatter rectForPageAtIndex:]): Called -_recalcIfNecessary and then returned the rect from
-_pageContentRect.
(-[WKWebViewPrintFormatter drawInRect:forPageAtIndex:]): Taught to draw the given page of the printed PDF
document into the given rect and context provided to us.

  • UIProcess/WebPageProxy.h: Declared didFinishDrawingPagesToPDF.
  • UIProcess/WebPageProxy.messages.in: Added DidFinishDrawingPagesToPDF. Made this a delayed syncrhonous message

so that additoinal work could be performed after sending the response.

  • UIProcess/ios/PageClientImplIOS.h:
  • UIProcess/ios/PageClientImplIOS.mm:

(WebKit::PageClientImpl::didFinishDrawingPagesToPDF): Created a CGPDFDocument from the data sent to us by the
web process and called -[WKWebView _setPrintedDocument:].

  • UIProcess/ios/WKPDFView.h:
  • UIProcess/ios/WKPDFView.mm:

(-[WKPDFView pdfDocument]): Returned the underlying CGPDFDocument.
(-[WKPDFView documentData]): Deleted.

  • UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::didFinishDrawingPagesToPDF):

  • WebKit2.xcodeproj/project.pbxproj:
  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::computePagesForPrinting): Changed to call computePagesForPrintingImpl.
(WebKit::WebPage::computePagesForPrintingImpl): Moved the common logic of computePagesForPrinting to here so it
could be called by computePagesForPrintingAndStartDrawingToPDF.
(WebKit::WebPage::drawPagesToPDF): Changed to call drawPagesToPDFImpl
(WebKit::WebPage::drawPagesToPDFImpl): Moved the common logic of drawPagesToPDF to here so it could be called
by computePagesForPrintingAndStartDrawingToPDF.

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
  • WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::computePagesForPrintingAndStartDrawingToPDF): Computed page rects and returned them as a
synchronous response. Afterwards called drawPagesToPDFImpl() and returned the PDF data by sending
WebPageProxy::DidFinishDrawingPagesToPDF.

12:48 PM Changeset in webkit [169174] by Lucas Forschler
  • 7 edits in tags/Safari-538.36/Source

Merged r169151.

12:45 PM Changeset in webkit [169173] by roger_fong@apple.com
  • 7 edits in trunk/LayoutTests

Unreviewed. Actually enable snapshot tests on mac wk2.

​https://bugs.webkit.org/show_bug.cgi?id=131871

  • platform/mac-wk1/TestExpectations: Skip them here.
  • platform/mac/TestExpectations: Unskip them here.

Some slight modifications to some of the tests to reduce flakiness of tests.

  • plugins/snapshotting/autoplay-dominant.html:
  • plugins/snapshotting/autoplay-plugin-blocked-by-image.html:
  • plugins/snapshotting/autoplay-plugin-mostly-blocked-by-image.html:
  • plugins/snapshotting/snapshot-plugin-not-quite-blocked-by-image.html:
11:22 AM Changeset in webkit [169172] by Michał Pakuła vel Rutka
  • 7 edits in trunk

[EFL] Turn on ENABLE_CSS_FILTERS
​https://bugs.webkit.org/show_bug.cgi?id=133153

Reviewed by Gyuyoung Kim.

.:

  • Source/cmake/OptionsEfl.cmake: Enable CSS_FILTERS as default option value.

Source/WebKit2:
Make WebKit2 compile after r166741 when CSS_FILTERS are enabled.

  • Shared/CoordinatedGraphics/CoordinatedGraphicsArgumentCoders.cpp:

(IPC::ArgumentCoder<WebCore::FilterOperations>::decode):

LayoutTests:
Remove failure test expectations for tests passing after turning on CSS_FILTERS.

  • platform/efl-wk2/TestExpectations:
  • platform/efl/TestExpectations:
11:18 AM Changeset in webkit [169171] by dburkart@apple.com
  • 5 edits in branches/safari-538.34-branch/Source/WebKit2

Merged r169170.

10:37 AM Changeset in webkit [169170] by beidson@apple.com
  • 5 edits in trunk/Source/WebKit2

Special handling of telephone number detection menu.
<rdar://problem/16967971> and ​https://bugs.webkit.org/show_bug.cgi?id=133143

Reviewed by Geoff Garen.

Add a bool to ContextMenuContextData to represent telephone number menus:

  • Shared/ContextMenuContextData.cpp:

(WebKit::ContextMenuContextData::ContextMenuContextData):
(WebKit::ContextMenuContextData::operator=):
(WebKit::ContextMenuContextData::encode):
(WebKit::ContextMenuContextData::decode):

  • Shared/ContextMenuContextData.h:

(WebKit::ContextMenuContextData::ContextMenuContextData):
(WebKit::ContextMenuContextData::isTelephoneNumberContext):

  • UIProcess/mac/WebContextMenuProxyMac.mm:

(WebKit::WebContextMenuProxyMac::showContextMenu): If the menu is a telephone

number menu then display it using a specific AppKit API.

  • UIProcess/mac/WebPageProxyMac.mm:

(WebKit::WebPageProxy::showTelephoneNumberMenu): Filter the array of menu items

through a new WKSI call, and then show it using a telephone number context.

10:30 AM Changeset in webkit [169169] by matthew_hanson@apple.com
  • 5 edits in branches/safari-538.34-branch/Source

Versioning.

10:22 AM Changeset in webkit [169168] by bshafiei@apple.com
  • 5 edits in branches/safari-538.34.5-branch/Source

Versioning.

10:04 AM Changeset in webkit [169167] by ap@apple.com
  • 3 edits in trunk/Source/WebKit2

[Mac] REGRESSION (Async text input): Asian text input doesn't work in Flash
​https://bugs.webkit.org/show_bug.cgi?id=133128
<rdar://problem/16890920>

Reviewed by Anders Carlsson.

  • UIProcess/API/mac/WKView.mm:

(-[WKView _interpretKeyEvent:completionHandler:]): Added a separate code path for
plug-in input. Before async text input, we used to provide a nil input context due
to having _data->_interpretKeyEventsParameters while this function was executing,
and thus took a shortcut. The new behavior is different from pre-async in that
we don't collect commands at all, but I couldn't find any case where that mattered.
(-[WKView inputContext]): The _collectedKeypressCommands check was not a correct
replacement for _interpretKeyEventsParameters one, because this variable is not
always set within -_interpretKeyEvent:completionHandler:.

  • UIProcess/mac/WKTextInputWindowController.mm:

(-[WKTextInputPanel _interpretKeyEvent:usingLegacyCocoaTextInput:string:]):
While at it, fixed a separate issue with Cangjie predictive input. We shouldn't
keep prediction pop-up on screen after hiding the bottom input window.

9:56 AM Changeset in webkit [169166] by Alan Bujtas
  • 2 edits in trunk/LayoutTests

Unreviewed Mac gardening after r169161.

  • platform/mac-wk2/tiled-drawing/tile-size-slow-zoomed-expected.txt:
9:39 AM Changeset in webkit [169165] by Martin Robinson
  • 16 edits in trunk

[CMake] Improve handling of LIB_INSTALL_DIR, EXEC_INSTALL_DIR, and LIBEXEC_INSTALL_DIR
​https://bugs.webkit.org/show_bug.cgi?id=132819

Reviewed by Carlos Garcia Campos.

.:

  • CMakeLists.txt: Hard-code CMAKE_ARCHIVE_OUTPUT_DIRECTORY etc, so that we can accept absolute paths for LIB_INSTALL_DIR

and friends.

  • Source/cmake/OptionsCommon.cmake: Get rid of LIB_SUFFIX and try to make all path variables absolute. Also set

CMAKE_ARCHIVE_OUTPUT_DIRECTORY, etc using the last path component of LIB_INSTALL_DIR, etc, so that absolute paths
are supported.

  • Source/cmake/OptionsGTK.cmake: Ditto. Also get rid of all pkg-config specific variables.

Source/JavaScriptCore:

  • javascriptcoregtk.pc.in: Instead of using the special pkg-config variables,

use the common CMake ones directly.

Source/WebKit/efl:

  • EWebKitConfig.cmake.in: LIB_INSTALL_DIR is always absolute now, so we don't need

to prefix it with the prefix.

  • ewebkit.pc.in: Instead of hard-coding the path to the library installation directory use LIB_INSTALL_DIR.

Source/WebKit2:

  • PlatformEfl.cmake: No need to prefix EXEC_INSTALL_DIR with the prefix any longer.
  • PlatformGtk.cmake: No longer use CMAKE_INSTALL_FULL_FOO variables, instead using the WebKit-specific ones.
  • efl/EWebKit2Config.cmake.in: Ditto.
  • efl/ewebkit2.pc.in: Instead of hard-coding the path to the library installation directory use LIB_INSTALL_DIR.
  • webkit2gtk-web-extension.pc.in: No longer use the pkg-config only variables. Instead use

the common CMake ones.

  • webkit2gtk.pc.in: Ditto.
9:33 AM Changeset in webkit [169164] by fpizlo@apple.com
  • 6 edits in trunk

Unreviewed, roll out ​http://trac.webkit.org/changeset/169159.

This was a unilateral change and wasn't properly reviewed.

PerformanceTests/SunSpider:

  • profiler-test.yaml:

Source/JavaScriptCore:

  • tests/mozilla/mozilla-tests.yaml:

LayoutTests:

  • js/script-tests/function-apply-many-args.js:
9:19 AM Changeset in webkit [169163] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit2

REGRESSION(r168625): [GTK] Assertion failure in LocalStorageDatabaseTracker
​https://bugs.webkit.org/show_bug.cgi?id=133118

Reviewed by Philippe Normand.

Call WebContext::applyPlatformSpecificConfigurationDefaults() to
initialize the WebContextConfiguration before creating the default WebContext.

  • UIProcess/API/gtk/WebKitWebContext.cpp:

(createDefaultWebContext):

9:14 AM Changeset in webkit [169162] by graouts@webkit.org
  • 7 edits in trunk

Array.prototype.find and findIndex should skip holes
​https://bugs.webkit.org/show_bug.cgi?id=132658

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:
Skip holes in the array when iterating such that callback isn't called.

  • builtins/Array.prototype.js:

(find):
(findIndex):

LayoutTests:

  • js/array-find-expected.txt:
  • js/array-findIndex-expected.txt:
  • js/script-tests/array-find.js:
  • js/script-tests/array-findIndex.js:
9:06 AM Changeset in webkit [169161] by Alan Bujtas
  • 6 edits in trunk/Source/WebCore

Garbage when rubber-banding at the right edge of a page zoomed to non-integral scale.
​https://bugs.webkit.org/show_bug.cgi?id=133139
<rdar://problem/16503353>

Reviewed by Simon Fraser.

Do not pixel align the root content layer. The alignment code expands the graphics's layer size
which makes the right and bottom tiles bigger than the content.
Painting the body's background color produces pixel cracks, because the content can not
fill the expanded tiles completely.

Not testable.

  • platform/graphics/GraphicsLayerClient.h:

(WebCore::GraphicsLayerClient::needsPixelAligment):

  • platform/graphics/ca/GraphicsLayerCA.cpp:

(WebCore::GraphicsLayerCA::updateGeometry):
(WebCore::GraphicsLayerCA::computePixelAlignment):

  • platform/graphics/ca/mac/TileGrid.mm:

(WebCore::TileGrid::rectForTileIndex):

  • rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::needsPixelAligment):

  • rendering/RenderLayerBacking.h:
8:31 AM Changeset in webkit [169160] by stavila@adobe.com
  • 3 edits
    2 adds in trunk

REGRESSION (r168046): Invalid layout in WebCore::RenderBox::containingBlockLogicalWidthForPositioned
​https://bugs.webkit.org/show_bug.cgi?id=132933

Reviewed by Darin Adler.

Source/WebCore:
Invalid layout is performed when calling containingBlockLogicalWidthForPositioned for a RenderTableSection object.

Test: fast/multicol/newmulticol/table-section-crash.html

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::containingBlockLogicalWidthForPositioned):

LayoutTests:
Added test for crash caused by invalid layout for RenderTableSection.

  • fast/multicol/newmulticol/table-section-crash-expected.txt: Added.
  • fast/multicol/newmulticol/table-section-crash.html: Added.
7:56 AM Changeset in webkit [169159] by commit-queue@webkit.org
  • 6 edits in trunk

REGRESSION(r169092 and r169102): Skip failing JSC tests on ARM64 properly
​https://bugs.webkit.org/show_bug.cgi?id=133149

Patch by Eva Balazsfalvi <​evab.u-szeged@partner.samsung.com> on 2014-05-21
Reviewed by Csaba Osztrogonác.

PerformanceTests/SunSpider:

  • profiler-test.yaml:

Source/JavaScriptCore:

  • tests/mozilla/mozilla-tests.yaml:

LayoutTests:

  • js/script-tests/function-apply-many-args.js:
7:43 AM Changeset in webkit [169158] by Michał Pakuła vel Rutka
  • 36 edits in trunk/LayoutTests

Unreviewed EFL gardening

Rebaseline test expectations after r169048.

  • platform/efl/css1/box_properties/float_on_text_elements-expected.txt:
  • platform/efl/css1/text_properties/text_indent-expected.txt:
  • platform/efl/css2.1/t0905-c414-flt-wrap-00-e-expected.txt:
  • platform/efl/css2.1/t0905-c5525-fltwidth-00-c-g-expected.txt:
  • platform/efl/css2.1/t100801-c544-valgn-00-a-ag-expected.txt:
  • platform/efl/css2.1/t100801-c544-valgn-02-d-agi-expected.txt:
  • platform/efl/css2.1/t100801-c544-valgn-03-d-agi-expected.txt:
  • platform/efl/css2.1/t100801-c544-valgn-04-d-agi-expected.txt:
  • platform/efl/css2.1/t1601-c547-indent-00-b-a-expected.txt:
  • platform/efl/fast/block/float/026-expected.txt:
  • platform/efl/fast/block/float/028-expected.txt:
  • platform/efl/fast/css/empty-pseudo-class-expected.txt:
  • platform/efl/fast/css/first-child-pseudo-class-expected.txt:
  • platform/efl/fast/css/first-of-type-pseudo-class-expected.txt:
  • platform/efl/fast/css/last-child-pseudo-class-expected.txt:
  • platform/efl/fast/css/last-of-type-pseudo-class-expected.txt:
  • platform/efl/fast/css/only-child-pseudo-class-expected.txt:
  • platform/efl/fast/css/only-of-type-pseudo-class-expected.txt:
  • platform/efl/fast/line-grid/line-align-left-edges-expected.txt:
  • platform/efl/fast/multicol/column-count-with-rules-expected.txt:
  • platform/efl/fast/multicol/float-multicol-expected.txt:
  • platform/efl/fast/multicol/layers-in-multicol-expected.txt:
  • platform/efl/fast/multicol/vertical-lr/float-multicol-expected.txt:
  • platform/efl/fast/multicol/vertical-rl/float-multicol-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-1-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-10-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-2-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-3-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-4-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-5-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-6-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-7-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-8-expected.txt:
  • platform/efl/fast/repaint/line-flow-with-floats-9-expected.txt:
  • platform/efl/fast/text/whitespace/024-expected.txt:
6:08 AM Changeset in webkit [169157] by Gustavo Noronha Silva
  • 2 edits in trunk

[CMake] Support building with Debug Fission
​https://bugs.webkit.org/show_bug.cgi?id=131177

Reviewed by Philippe Normand.

  • Source/cmake/OptionsCommon.cmake: add a DEBUG_FISSION option to enable usage

of ​http://gcc.gnu.org/wiki/DebugFission.

5:01 AM Changeset in webkit [169156] by Csaba Osztrogonác
  • 2 edits in trunk/Tools

REGRESSION(r169092): It broke run-jsc-stress tests on non Mac platforms
​https://bugs.webkit.org/show_bug.cgi?id=133145

Patch by Eva Balazsfalvi <​evab.u-szeged@partner.samsung.com> on 2014-05-21
Reviewed by Csaba Osztrogonác.

  • Scripts/run-jsc-stress-tests:
4:55 AM Changeset in webkit [169155] by jinwoo7.song@samsung.com
  • 1 edit
    1 add in trunk/LayoutTests

Unreviewed EFL gardening.

  • platform/efl/fast/css/absolute-child-with-percent-height-inside-relative-parent-expected.txt: Added.

Rebaseline after r169048.

4:03 AM WebKitGTK/KeepingTheTreeGreen edited by ltilve@igalia.com
(diff)
2:28 AM Changeset in webkit [169154] by berto@igalia.com
  • 2 edits
    1 add in trunk/LayoutTests

[GTK] Unreviewed GTK gardening

Patch by Lorenzo Tilve <​ltilve@igalia.com> on 2014-05-21

  • platform/gtk/TestExpectations: Report and mark new tests failing.
  • platform/gtk/fast/css/absolute-child-with-percent-height-inside-relative-parent-expected.txt: Added.

Rebaseline after r169048.

2:11 AM Changeset in webkit [169153] by simon.pena@samsung.com
  • 2 edits in trunk/Source/WebCore

[EFL] Ensure EGLOffScreenContext::initialize calls platformMakeCurrent on the surface
​https://bugs.webkit.org/show_bug.cgi?id=124832

Reviewed by Antonio Gomes.

The original code was only calling platformMakeCurrent when the
GL_EXT_robustness extension is supported: the alternate code path
would only create a context, but it wouldn't make that context
current. This patch ensures platformMakeCurrent is also called for
the alternate code path, and, if it fails to make the context
current, will destroy the context. Effectively, this makes both
code paths behave in the same way.

  • platform/graphics/surfaces/egl/EGLContext.cpp:

(WebCore::EGLOffScreenContext::initialize):

1:12 AM Changeset in webkit [169152] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebKit2

[WebKit2] Cleanup the build from unused variable in WebProcess module
​https://bugs.webkit.org/show_bug.cgi?id=133062

Patch by Shivakumar JM <​shiva.jm@samsung.com> on 2014-05-21
Reviewed by Csaba Osztrogonác.

Fix unused parameter by using UNUSED_PARAM macro

  • WebProcess/UserContent/WebUserContentController.cpp:

(WebKit::WebUserContentController::addUserScriptMessageHandlers):
(WebKit::WebUserContentController::removeUserScriptMessageHandler):

May 20, 2014:

11:55 PM Changeset in webkit [169151] by ggaren@apple.com
  • 7 edits in trunk/Source

Rolled out <​http://trac.webkit.org/changeset/166184>
​https://bugs.webkit.org/show_bug.cgi?id=133144

Reviewed by Gavin Barraclough.

It caused a performance regression.

../JavaScriptCore:

  • heap/BlockAllocator.cpp:

(JSC::BlockAllocator::blockFreeingThreadStartFunc):

../WTF:

  • wtf/FastMalloc.cpp:

(WTF::TCMalloc_PageHeap::runScavengerThread):

  • wtf/Threading.h:
  • wtf/ThreadingPthreads.cpp:

(WTF::createThreadInternal):
(WTF::setCurrentThreadQOSUtility): Deleted.

  • wtf/ThreadingWin.cpp:

(WTF::setCurrentThreadQOSUtility): Deleted.

11:01 PM Changeset in webkit [169150] by Lucas Forschler
  • 5 edits in trunk/Source

Versioning.

10:58 PM Changeset in webkit [169149] by Lucas Forschler
  • 1 copy in tags/Safari-538.36

New Tag.

10:16 PM Changeset in webkit [169148] by fpizlo@apple.com
  • 10 edits
    1 add in branches/ftlopt/Source/JavaScriptCore

[ftlopt] A StructureSet with one element should only require one word and no allocation
​https://bugs.webkit.org/show_bug.cgi?id=133014

Reviewed by Oliver Hunt.

This makes it more efficient to use StructureSet in situations where the common case is
just one structure.

I also took the opportunity to use the same set terminology we use in BitVector: merge,
filter, exclude, contains, etc.

Eventually, this will be used to implement StructureAbstractValue as well.

(JSC::StructureSet::StructureSet):
(JSC::StructureSet::operator=):
(JSC::StructureSet::clear):
(JSC::StructureSet::add):
(JSC::StructureSet::remove):
(JSC::StructureSet::contains):
(JSC::StructureSet::merge):
(JSC::StructureSet::filter):
(JSC::StructureSet::exclude):
(JSC::StructureSet::isSubsetOf):
(JSC::StructureSet::overlaps):
(JSC::StructureSet::operator==):
(JSC::StructureSet::speculationFromStructures):
(JSC::StructureSet::arrayModesFromStructures):
(JSC::StructureSet::dumpInContext):
(JSC::StructureSet::dump):
(JSC::StructureSet::addOutOfLine):
(JSC::StructureSet::containsOutOfLine):
(JSC::StructureSet::copyFrom):
(JSC::StructureSet::OutOfLineList::create):
(JSC::StructureSet::OutOfLineList::destroy):

  • bytecode/StructureSet.h:

(JSC::StructureSet::StructureSet):
(JSC::StructureSet::~StructureSet):
(JSC::StructureSet::onlyStructure):
(JSC::StructureSet::isEmpty):
(JSC::StructureSet::size):
(JSC::StructureSet::at):
(JSC::StructureSet::operator[]):
(JSC::StructureSet::last):
(JSC::StructureSet::OutOfLineList::list):
(JSC::StructureSet::OutOfLineList::OutOfLineList):
(JSC::StructureSet::deleteStructureListIfNecessary):
(JSC::StructureSet::isThin):
(JSC::StructureSet::pointer):
(JSC::StructureSet::singleStructure):
(JSC::StructureSet::structureList):
(JSC::StructureSet::set):
(JSC::StructureSet::clear): Deleted.
(JSC::StructureSet::add): Deleted.
(JSC::StructureSet::addAll): Deleted.
(JSC::StructureSet::remove): Deleted.
(JSC::StructureSet::contains): Deleted.
(JSC::StructureSet::containsOnly): Deleted.
(JSC::StructureSet::isSubsetOf): Deleted.
(JSC::StructureSet::overlaps): Deleted.
(JSC::StructureSet::singletonStructure): Deleted.
(JSC::StructureSet::speculationFromStructures): Deleted.
(JSC::StructureSet::arrayModesFromStructures): Deleted.
(JSC::StructureSet::operator==): Deleted.
(JSC::StructureSet::dumpInContext): Deleted.
(JSC::StructureSet::dump): Deleted.

  • dfg/DFGAbstractInterpreterInlines.h:

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

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::emitPrototypeChecks):
(JSC::DFG::ByteCodeParser::handleGetById):
(JSC::DFG::ByteCodeParser::parseBlock):

  • dfg/DFGCSEPhase.cpp:

(JSC::DFG::CSEPhase::structureTransitionWatchpointElimination):

  • dfg/DFGNode.h:

(JSC::DFG::Node::convertToStructureTransitionWatchpoint):

  • dfg/DFGTypeCheckHoistingPhase.cpp:

(JSC::DFG::TypeCheckHoistingPhase::noticeStructureCheck):

9:59 PM Changeset in webkit [169147] by matthew_hanson@apple.com
  • 1 copy in tags/Safari-538.34.44

New Tag.

9:08 PM Changeset in webkit [169146] by Brent Fulgham
  • 11 edits in trunk/Source/WebCore

[Mac] DataCues do not work properly when rewinding video
​https://bugs.webkit.org/show_bug.cgi?id=133138
<rdar://problem/16979086>

Reviewed by Eric Carlson.

Make the TrackPrivateBase responsible for knowing if a type of track needs a non-zero
startTimeVariance value.

Also, correct a bug in the equality test for DataCue objects.

  • html/track/DataCue.cpp:

(WebCore::DataCue::isEqual): Handle the JSValue data types consistently, so that we don't attempt
to compare a JSNull against a default JSValue object.

  • html/track/InbandTextTrack.cpp:

(WebCore::InbandTextTrack::startTimeVariance): Added.

  • html/track/InbandTextTrack.h:
  • html/track/TextTrack.cpp:

(WebCore::TextTrack::hasCue): Use new startTimeVariance method location.

  • html/track/TextTrack.h:

(WebCore::TextTrack::startTimeVariance): Added.

  • html/track/TextTrackCue.cpp:

(WebCore::TextTrackCue::hasEquivalentStartTime): Use new startTimeVariance location.

  • html/track/TextTrackCue.h:

(WebCore::TextTrackCue::startTimeVariance): Deleted.

  • html/track/TextTrackCueGeneric.h:
  • platform/graphics/TrackPrivateBase.h:

(WebCore::TrackPrivateBase::startTimeVariance): Added.

  • platform/graphics/avfoundation/InbandTextTrackPrivateAVF.h:
8:49 PM Changeset in webkit [169145] by fpizlo@apple.com
  • 6 edits
    1 add in trunk/Source/JavaScriptCore

DFG prediction propagation should agree with fixup phase over the return type of GetByVal
​https://bugs.webkit.org/show_bug.cgi?id=133134

Reviewed by Mark Hahnenberg.

Make prediction propagator use ArrayMode refinement to decide the return type.

Also introduce a heap prediction intrinsic that allows us to test weird corner cases
like this. The only way we'll see a mismatch like this in the real world is probably
through a gnarly race condition.

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::handleIntrinsic):

  • dfg/DFGNode.h:

(JSC::DFG::Node::setHeapPrediction):

  • dfg/DFGPredictionPropagationPhase.cpp:

(JSC::DFG::PredictionPropagationPhase::propagate):

  • jsc.cpp:

(GlobalObject::finishCreation):
(functionFalse1):
(functionFalse2):
(functionUndefined1):
(functionUndefined2):
(functionFalse): Deleted.
(functionOtherFalse): Deleted.
(functionUndefined): Deleted.

  • runtime/Intrinsic.h:
  • tests/stress/get-by-val-double-predicted-int.js: Added.

(foo):

6:55 PM Changeset in webkit [169144] by bshafiei@apple.com
  • 1 copy in tags/Safari-538.34.9

New tag.

5:44 PM Changeset in webkit [169143] by fpizlo@apple.com
  • 42 edits
    1 copy
    14 adds in branches/ftlopt

[ftlopt] DFG bytecode parser should turn GetById with nothing but a Getter stub as stuff+handleCall, and handleCall should be allowed to inline if it wants to
​https://bugs.webkit.org/show_bug.cgi?id=133105

Reviewed by Michael Saboff.

Source/JavaScriptCore:

  • GetByIdStatus now knows about getters and can report intelligent things about them. As is usually the case with how we do these things, GetByIdStatus knows more about getters than the DFG can actually handle: it'll report details about polymorphic getter calls even though the DFG won't be able to handle those. This is fine; the DFG will see those statuses and bail to a generic slow path.


  • The DFG::ByteCodeParser now knows how to set up and do handleCall() for a getter call. This can, and usually does, result in inlining of getters!


  • CodeOrigin and OSR exit know about inlined getter calls. When you OSR out of an inlined getter, we set the return PC to a getter return thunk that fixes up the stack. We use the usual offset-true-return-PC trick, where OSR exit places the true return PC of the getter's caller as a phony argument that only the thunk knows how to find.


  • Removed a bunch of dead monomorphic chain support from StructureStubInfo.


  • A large chunk of this change is dragging GetGetterSetterByOffset, GetGetter, and GetSetter through the DFG and FTL. GetGetterSetterByOffset is like GetByOffset except that we know that we're returning a GetterSetter cell. GetGetter and GetSetter extract the getter, or setter, from the GetterSetter.


This is a ~2.5x speed-up on the getter microbenchmarks that we already had. So far none
of the "real" benchmarks exercise getters enough for this to matter. But I noticed that
some of the variants of the Richards benchmark in other languages - for example
Wolczko's Java translation of a C++ translation of Deutsch's Smalltalk version - use
getters and setters extensively. So, I created a getter/setter JavaScript version of
Richards and put it in regress/script-tests/getter-richards.js. That sees about a 2.4x
speed-up from this patch, which is very reassuring.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::printGetByIdCacheStatus):
(JSC::CodeBlock::findStubInfo):

  • bytecode/CodeBlock.h:
  • bytecode/CodeOrigin.cpp:

(WTF::printInternal):

  • bytecode/CodeOrigin.h:

(JSC::InlineCallFrame::specializationKindFor):

  • bytecode/GetByIdStatus.cpp:

(JSC::GetByIdStatus::computeFor):
(JSC::GetByIdStatus::computeForStubInfo):
(JSC::GetByIdStatus::makesCalls):
(JSC::GetByIdStatus::computeForChain): Deleted.

  • bytecode/GetByIdStatus.h:

(JSC::GetByIdStatus::makesCalls): Deleted.

  • bytecode/GetByIdVariant.cpp:

(JSC::GetByIdVariant::~GetByIdVariant):
(JSC::GetByIdVariant::GetByIdVariant):
(JSC::GetByIdVariant::operator=):
(JSC::GetByIdVariant::dumpInContext):

  • bytecode/GetByIdVariant.h:

(JSC::GetByIdVariant::GetByIdVariant):
(JSC::GetByIdVariant::callLinkStatus):

  • bytecode/PolymorphicGetByIdList.cpp:

(JSC::GetByIdAccess::fromStructureStubInfo):
(JSC::PolymorphicGetByIdList::from):

  • bytecode/SpeculatedType.h:
  • bytecode/StructureStubInfo.cpp:

(JSC::StructureStubInfo::deref):
(JSC::StructureStubInfo::visitWeakReferences):

  • bytecode/StructureStubInfo.h:

(JSC::isGetByIdAccess):
(JSC::StructureStubInfo::initGetByIdChain): Deleted.

  • dfg/DFGAbstractHeap.h:
  • dfg/DFGAbstractInterpreterInlines.h:

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

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::addCall):
(JSC::DFG::ByteCodeParser::handleCall):
(JSC::DFG::ByteCodeParser::handleInlining):
(JSC::DFG::ByteCodeParser::handleGetByOffset):
(JSC::DFG::ByteCodeParser::handleGetById):
(JSC::DFG::ByteCodeParser::InlineStackEntry::InlineStackEntry):
(JSC::DFG::ByteCodeParser::parse):

  • dfg/DFGCSEPhase.cpp:

(JSC::DFG::CSEPhase::getGetterSetterByOffsetLoadElimination):
(JSC::DFG::CSEPhase::getInternalFieldLoadElimination):
(JSC::DFG::CSEPhase::performNodeCSE):
(JSC::DFG::CSEPhase::getTypedArrayByteOffsetLoadElimination): Deleted.

  • dfg/DFGClobberize.h:

(JSC::DFG::clobberize):

  • dfg/DFGFixupPhase.cpp:

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

  • dfg/DFGJITCompiler.cpp:

(JSC::DFG::JITCompiler::linkFunction):

  • dfg/DFGNode.h:

(JSC::DFG::Node::hasStorageAccessData):

  • dfg/DFGNodeType.h:
  • dfg/DFGOSRExitCompilerCommon.cpp:

(JSC::DFG::reifyInlinedCallFrames):

  • dfg/DFGPredictionPropagationPhase.cpp:

(JSC::DFG::PredictionPropagationPhase::propagate):

  • dfg/DFGSafeToExecute.h:

(JSC::DFG::safeToExecute):

  • dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • ftl/FTLAbstractHeapRepository.cpp:
  • ftl/FTLAbstractHeapRepository.h:
  • ftl/FTLCapabilities.cpp:

(JSC::FTL::canCompile):

  • ftl/FTLLink.cpp:

(JSC::FTL::link):

  • ftl/FTLLowerDFGToLLVM.cpp:

(JSC::FTL::LowerDFGToLLVM::compileNode):
(JSC::FTL::LowerDFGToLLVM::compileGetGetter):
(JSC::FTL::LowerDFGToLLVM::compileGetSetter):

  • jit/AccessorCallJITStubRoutine.h:
  • jit/JIT.cpp:

(JSC::JIT::assertStackPointerOffset):
(JSC::JIT::privateCompile):

  • jit/JIT.h:
  • jit/JITPropertyAccess.cpp:

(JSC::JIT::emit_op_get_by_id):

  • jit/ThunkGenerators.cpp:

(JSC::arityFixupGenerator):
(JSC::baselineGetterReturnThunkGenerator):
(JSC::baselineSetterReturnThunkGenerator):
(JSC::arityFixup): Deleted.

  • jit/ThunkGenerators.h:
  • runtime/CommonSlowPaths.cpp:

(JSC::setupArityCheckData):

  • tests/stress/exit-from-getter.js: Added.
  • tests/stress/poly-chain-getter.js: Added.

(Cons):
(foo):
(test):

  • tests/stress/poly-chain-then-getter.js: Added.

(Cons1):
(Cons2):
(foo):
(test):

  • tests/stress/poly-getter-combo.js: Added.

(Cons1):
(Cons2):
(foo):
(test):
(.test):

  • tests/stress/poly-getter-then-chain.js: Added.

(Cons1):
(Cons2):
(foo):
(test):

  • tests/stress/poly-getter-then-self.js: Added.

(foo):
(test):
(.test):

  • tests/stress/poly-self-getter.js: Added.

(foo):
(test):
(getter):

  • tests/stress/poly-self-then-getter.js: Added.

(foo):
(test):

  • tests/stress/weird-getter-counter.js: Added.

(foo):
(test):

Source/WTF:

  • wtf/Bag.h:

(WTF::Bag::iterator::operator!=):

LayoutTests:

  • js/regress/getter-no-activation-expected.txt: Added.
  • js/regress/getter-no-activation.html: Added.
  • js/regress/script-tests/getter-no-activation.js: Added.
  • js/regress/getter-richards-expected.txt: Added.
  • js/regress/getter-richards.html: Added.
  • js/regress/script-tests/getter-richards.js: Added.
5:13 PM Changeset in webkit [169142] by enrica@apple.com
  • 2 edits in trunk/Source/WebKit2

REGRESSION (WK2): Tapping on the input field in duckduckgo.com shows the keyboard but not the caret.
​https://bugs.webkit.org/show_bug.cgi?id=133140
<rdar://problem/16980898>

Reviewed by Benjamin Poulain.

When a page focuses a field on load, we receive the call to startAssistingNode
but we don't create a text interaction assistant if the focus is not originated
by a user action. Upon receiving the selection change notification we don't
call the text interaction assistant, since we don't have one, but we do call
the web selection assistant. We don't have any valid rects for web selection
and we clear the selection on the page, causing the caret not to appear and
the typing to have no effect.

  • UIProcess/ios/WKContentViewInteraction.mm:

(-[WKContentView _updateChangedSelection]):

5:07 PM Changeset in webkit [169141] by matthew_hanson@apple.com
  • 2 edits in branches/safari-538.34-branch/Source/WebCore

Rollout r169076.

5:03 PM Changeset in webkit [169140] by matthew_hanson@apple.com
  • 6 edits in branches/safari-538.34-branch/Source/WebCore

Merge r168974.

3:55 PM Changeset in webkit [169139] by mhahnenberg@apple.com
  • 15 edits in trunk/Source

Watchdog timer should be lazily allocated
​https://bugs.webkit.org/show_bug.cgi?id=133135

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:
We incur a noticeable amount of overhead on some benchmarks due to checking if the Watchdog ever fired.
There is no reason to do this checking if we never activated the Watchdog, which can only be done through
JSContextGroupSetExecutionTimeLimit or JSContextGroupClearExecutionTimeLimit.

By allocating the Watchdog lazily on the VM we can avoid all of the associated overhead when we don't use
these two API functions (which is true of most clients).

  • API/JSContextRef.cpp:

(JSContextGroupSetExecutionTimeLimit):
(JSContextGroupClearExecutionTimeLimit):

  • dfg/DFGByteCodeParser.cpp:

(JSC::DFG::ByteCodeParser::parseBlock):

  • dfg/DFGSpeculativeJIT32_64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • dfg/DFGSpeculativeJIT64.cpp:

(JSC::DFG::SpeculativeJIT::compile):

  • interpreter/Interpreter.cpp:

(JSC::Interpreter::execute):
(JSC::Interpreter::executeCall):
(JSC::Interpreter::executeConstruct):

  • jit/JITOpcodes.cpp:

(JSC::JIT::emit_op_loop_hint):
(JSC::JIT::emitSlow_op_loop_hint):

  • jit/JITOperations.cpp:
  • llint/LLIntSlowPaths.cpp:

(JSC::LLInt::LLINT_SLOW_PATH_DECL):

  • runtime/VM.h:
  • runtime/Watchdog.cpp:

(JSC::Watchdog::Scope::Scope): Deleted.
(JSC::Watchdog::Scope::~Scope): Deleted.

  • runtime/Watchdog.h:

(JSC::Watchdog::Scope::Scope):
(JSC::Watchdog::Scope::~Scope):

Source/WebCore:
No new tests.

We incur a noticeable amount of overhead on some benchmarks due to checking if the Watchdog ever fired.
There is no reason to do this checking if we never activated the Watchdog, which can only be done through
JSContextGroupSetExecutionTimeLimit or JSContextGroupClearExecutionTimeLimit.

By allocating the Watchdog lazily on the VM we can avoid all of the associated overhead when we don't use
these two API functions (which is true of most clients).

  • bindings/js/JSEventListener.cpp:

(WebCore::JSEventListener::handleEvent):

  • bindings/js/WorkerScriptController.cpp:

(WebCore::WorkerScriptController::evaluate):
(WebCore::WorkerScriptController::scheduleExecutionTermination):
(WebCore::WorkerScriptController::isExecutionTerminating):

3:54 PM Changeset in webkit [169138] by Lucas Forschler
  • 5 edits in branches/safari-537.77-branch/Source

Versioning.

3:22 PM Changeset in webkit [169137] by matthew_hanson@apple.com
  • 2 edits in branches/safari-538.34-branch/Source/WebKit2

Merge r169114.

3:14 PM Changeset in webkit [169136] by Lucas Forschler
  • 1 copy in branches/safari-537.77-branch

New Branch.

2:48 PM Changeset in webkit [169135] by matthew_hanson@apple.com
  • 6 edits in branches/safari-538.34-branch

Merge r169094.

2:03 PM Changeset in webkit [169134] by dino@apple.com
  • 13 edits in trunk/Source

[Mac] Allow popup menus to override default appearance
​https://bugs.webkit.org/show_bug.cgi?id=133129

Reviewed by Tim Horton.

Source/WebCore:
WebKitSystemInterface's WKPopupMenu has a flag to hide
the arrows at the end of a popup-menu (e.g. <select>).
Expose that via adding a hasDefaultAppearance flag to
PopupMenuStyle, which then passes it on to WKSI.

  • platform/PopupMenuStyle.h:

(WebCore::PopupMenuStyle::PopupMenuStyle): New flag in constructor.
(WebCore::PopupMenuStyle::hasDefaultAppearance): New flag.

  • platform/mac/WebCoreSystemInterface.h: wkPopupMenu signature has changed to accept

the new flag.

  • platform/mac/WebCoreSystemInterface.mm: Ditto.
  • rendering/RenderMenuList.cpp:

(RenderMenuList::itemStyle): Pass in true as default appearance when creating
the PopupMenuStyle. We don't need it.
(RenderMenuList::menuStyle): In this case we pass in the negated value of
style().hasAppearance(), which indicates if we are overriding the built-in drawing.

  • rendering/RenderSearchField.cpp:

(WebCore::RenderSearchField::menuStyle): Use true as the default value.

Source/WebKit/mac:
Replace the call to wkPopupMenuWithSize with wkPopupMenu, now
with extra parameters to indicate the control size and whether
or not to draw the arrows at the end.

  • WebCoreSupport/PopupMenuMac.mm:

(PopupMenuMac::show):

Source/WebKit2:
Add an extra "hideArrows" flag to PlatformPopupMenuData
and replace the call to WKPopupMenuWithSize with WKPopupMenu.

  • Shared/PlatformPopupMenuData.cpp: Encode and decode the new data value.

(WebKit::PlatformPopupMenuData::encode):
(WebKit::PlatformPopupMenuData::decode):

  • Shared/PlatformPopupMenuData.h: Add hideArrows boolean.
  • UIProcess/mac/WebPopupMenuProxyMac.mm:

(WebKit::WebPopupMenuProxyMac::showPopupMenu):

  • WebProcess/WebCoreSupport/mac/WebPopupMenuMac.mm:

(WebKit::WebPopupMenu::setUpPlatformData): The hideArrows flag is
true if we don't have the default appearance.

1:59 PM Changeset in webkit [169133] by dino@apple.com
  • 4 edits in trunk/WebKitLibraries

[Mac] Allow popup menus to override default appearance
​https://bugs.webkit.org/show_bug.cgi?id=133129

Reviewed by Tim Horton.

Add a new parameter to WKPopupMenu.

  • WebKitSystemInterface.h:
  • libWebKitSystemInterfaceMavericks.a:
  • libWebKitSystemInterfaceMountainLion.a:
1:57 PM Changeset in webkit [169132] by matthew_hanson@apple.com
  • 2 edits in branches/safari-538.34-branch/Source/JavaScriptCore

Merge r169094.

1:57 PM Changeset in webkit [169131] by jer.noble@apple.com
  • 2 edits in trunk/Source/WTF

[Mac] AVAssets are never destroyed; lack of an autorelease pool when calling callOnMainThread.
​https://bugs.webkit.org/show_bug.cgi?id=133130

Reviewed by Geoff Garen.

Wrap the function to be called in an autorelease pool, so that autoreleased objects are cleaned
up immediately after the end of the function.

  • wtf/mac/MainThreadMac.mm:

(WTF::timerFired):

1:52 PM Changeset in webkit [169130] by eric.carlson@apple.com
  • 8 edits in trunk/Source

[Mac] do not deactivate an audio session that has running I/O
​https://bugs.webkit.org/show_bug.cgi?id=133127

Reviewed by Jer Noble.

Source/WebCore:

  • page/Settings.h:

(WebCore::Settings::setShouldManageAudioSessionCategory): Renamed from setShouldManageAudioSession.
(WebCore::Settings::shouldManageAudioSessionCategory): Renamed from shouldManageAudioSession.
(WebCore::Settings::setShouldManageAudioSession): Deleted.
(WebCore::Settings::shouldManageAudioSession): Deleted.

  • platform/audio/mac/MediaSessionManagerMac.cpp:

(MediaSessionManager::updateSessionState): Don't deactivate the session if there are any

Video or Audio sessions.

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

(WebCore::MediaPlayerPrivateAVFoundationObjC::assetStatus): Drive-by change to log errors

returned by -statusOfValueForKey:error: in debug builds.

Source/WebKit/mac:

  • WebView/WebView.mm:

(-[WebView _commonInitializationWithFrameName:groupName:]): Settings::setShouldManageAudioSession

renamed to setShouldManageAudioSessionCategory, deal with it.

Source/WebKit2:

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::WebPage): Settings::setShouldManageAudioSession renamed to

setShouldManageAudioSessionCategory, deal with it.

1:36 PM Changeset in webkit [169129] by Beth Dakin
  • 3 edits in trunk/Source/WebCore

REGRESSION: All non-mainframe scrollbars don't paint after r169065
​https://bugs.webkit.org/show_bug.cgi?id=133132
-and corresponding-
<rdar://problem/16968850>

Reviewed by Geoff Garen.

We should not universally return true here. This feature will only work for
scrollbars with layers.

  • platform/ScrollableArea.h:
  • platform/Scrollbar.cpp:

(WebCore::Scrollbar::supportsUpdateOnSecondaryThread):

1:34 PM Changeset in webkit [169128] by dbates@webkit.org
  • 7 edits
    4 adds in trunk

Element within flattened frame may update its scroll state during the layout phase of the wrong RenderView
​https://bugs.webkit.org/show_bug.cgi?id=133013
<rdar://problem/16760154>

Reviewed by David Hyatt.

Source/WebCore:
Fixes an issue where the scroll state of an element may be updated during the layout of the wrong
RenderView. In particular, the scroll state of an element inside a CSS flex box in a flattened
frame f is updated during the layout of the RenderView associated with the parent frame of f instead
of during the layout of the RenderView associated with f.

The layout machinery assumes that the scroll state of each scrollable element is updated before the
completion of layout for its associated RenderView. Currently we have logic to defer updating the scroll
state of a scrollable element until completion of recursive layout. For a page with a flattened frame, we
defer such updates until completion of layout for all RenderViews along the ancestor frame hierarchy of
each flattened frame regardless of the RenderView associated with the element that originated the
deferred scroll state request. Instead, only the RenderView associated with the element that deferred its
scroll state update should dispatch the scroll state update.

Tests: fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-and-needs-full-repaint-crash.html

fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-assertion-failure.html

  • rendering/RenderBlock.cpp: Removed WTF::-prefix in typedef definition for ContinuationOutlineTableMap.

(WebCore::UpdateScrollInfoAfterLayoutTransaction::UpdateScrollInfoAfterLayoutTransaction): Added; a struct
that represents a RenderView v, nested transaction count, and a list of RenderBlocks in v that will need
to have their scroll state updated.
(WebCore::updateScrollInfoAfterLayoutTransactionStack): Added; returns the global stack of transactions.
(WebCore::RenderBlock::willBeDestroyed): Call removeFromUpdateScrollInfoAfterLayoutTransaction(), which
was formerly named as removeFromDelayedUpdateScrollInfoSet.
(WebCore::currentUpdateScrollInfoAfterLayoutTransaction): Added; returns the top-most transaction in the
global stack.
(WebCore::RenderBlock::beginUpdateScrollInfoAfterLayoutTransaction): Modified as appropriate to make use
of the global stack of transactions; formerly named startDelayUpdateScrollInfo.
(WebCore::RenderBlock::endAndCommitUpdateScrollInfoAfterLayoutTransaction): Modified as appropriate to
make use of the global stack of transactions; formerly named finishDelayUpdateScrollInfo.
(WebCore::RenderBlock::removeFromUpdateScrollInfoAfterLayoutTransaction): Modified as appropriate to make
use of the global stack of transactions; formerly named removeFromDelayedUpdateScrollInfoSet.
(WebCore::RenderBlock::updateScrollInfoAfterLayout): Modified as appropriate to make use of the global
stack of transactions.
(WebCore::RenderBlock::layout): Ditto.
(WebCore::RenderBlock::startDelayUpdateScrollInfo): Deleted.
(WebCore::RenderBlock::finishDelayUpdateScrollInfo): Deleted.
(WebCore::RenderBlock::removeFromDelayedUpdateScrollInfoSet): Deleted.

  • rendering/RenderBlock.h:
  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::willBeDestroyed): Call removeFromUpdateScrollInfoAfterLayoutTransaction(), which
was formerly named removeFromDelayedUpdateScrollInfoSet.

  • rendering/RenderDeprecatedFlexibleBox.cpp:

(WebCore::RenderDeprecatedFlexibleBox::layoutHorizontalBox): Call {begin, end}UpdateScrollInfoAfterLayoutTransaction(),
which was formerly named {start, end}DelayUpdateScrollInfo.
(WebCore::RenderDeprecatedFlexibleBox::layoutVerticalBox): Ditto.

  • rendering/RenderFlexibleBox.cpp:

(WebCore::RenderFlexibleBox::layoutBlock): Ditto.

LayoutTests:
Added tests to ensure that that we update the scroll state of an element during the layout phase of its
associated RenderView.

  • fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-and-needs-full-repaint-crash-expected.txt: Added.
  • fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-and-needs-full-repaint-crash.html: Added.
  • fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-assertion-failure-expected.txt: Added.
  • fast/frames/flattening/scrollable-flexbox-inside-iframe-with-zero-height-assertion-failure.html: Added.
1:21 PM Changeset in webkit [169127] by Lucas Forschler
  • 6 edits in branches/safari-538.34.5-branch

Merged r168599.

1:18 PM Changeset in webkit [169126] by Lucas Forschler
  • 2 edits in branches/safari-538.34.5-branch/Source/WebKit2

Merged r169114.

1:02 PM Changeset in webkit [169125] by matthew_hanson@apple.com
  • 5 edits in branches/safari-538.34-branch/Source

Versioning.

12:31 PM Changeset in webkit [169124] by Beth Dakin
  • 2 edits in trunk/Source/WebCore

REGRESSION (r169065): Mountain Lion run-api-tests failures: ASSERTION FAILED:
Uncaught exception - -[NSRegularLegacyScrollerImp setPresentationValue:]:
unrecognized selector sent to instance 0x7ff51aa38000
​https://bugs.webkit.org/show_bug.cgi?id=133121

Reviewed by Tim Horton.

  • platform/mac/ScrollbarThemeMac.mm:

(WebCore::ScrollbarThemeMac::setPaintCharacteristicsForScrollbar):

12:13 PM Changeset in webkit [169123] by Simon Fraser
  • 12 edits in trunk/Source

REGRESSION (r169063) Fixed and sticky nodes misplaced on scrolling sometimes
​https://bugs.webkit.org/show_bug.cgi?id=133106
<rdar://problem/16967648>

Reviewed by Sam Weinig.

Fix regression from r169063. That commit removed scrolledContentsLayers from
frame scrolling nodes, but they do actually use them.

So put them back; not in the base class, because they have a somewhat different
meaning for overflow scrolling and frame scrolling.

Source/WebCore:

  • WebCore.exp.in:
  • page/scrolling/AsyncScrollingCoordinator.cpp:

(WebCore::AsyncScrollingCoordinator::frameViewRootLayerDidChange):
(WebCore::AsyncScrollingCoordinator::updateFrameScrollingNode):

  • page/scrolling/AsyncScrollingCoordinator.h:
  • page/scrolling/ScrollingCoordinator.h:

(WebCore::ScrollingCoordinator::updateFrameScrollingNode):

  • page/scrolling/ScrollingStateFrameScrollingNode.cpp:

(WebCore::ScrollingStateFrameScrollingNode::ScrollingStateFrameScrollingNode):
(WebCore::ScrollingStateFrameScrollingNode::setScrolledContentsLayer):

  • page/scrolling/ScrollingStateFrameScrollingNode.h:
  • page/scrolling/mac/ScrollingTreeFrameScrollingNodeMac.mm:

(WebCore::ScrollingTreeFrameScrollingNodeMac::updateBeforeChildren):

  • rendering/RenderLayerCompositor.cpp:

(WebCore::RenderLayerCompositor::updateScrollCoordinatedLayer):

Source/WebKit2:

  • Shared/Scrolling/RemoteScrollingCoordinatorTransaction.cpp:

(ArgumentCoder<ScrollingStateFrameScrollingNode>::encode):
(ArgumentCoder<ScrollingStateFrameScrollingNode>::decode):
(WebKit::RemoteScrollingTreeTextStream::dump):

  • UIProcess/Scrolling/RemoteScrollingCoordinatorProxy.cpp:

(WebKit::RemoteScrollingCoordinatorProxy::connectStateNodeLayers):

11:44 AM Changeset in webkit [169122] by ap@apple.com
  • 6 edits in trunk

[Mac] WebProcess doesn't follow localization of UI process when run as a service
​https://bugs.webkit.org/show_bug.cgi?id=133126
<rdar://problem/13396515>
<rdar://problem/16561084>

Reviewed by Sam Weinig and Tim Horton.

Source/WebKit2:

  • Shared/EntryPointUtilities/mac/XPCService/XPCServiceMain.Development.mm:

(WebKit::handleXPCBootstrap):
(main):

  • Shared/EntryPointUtilities/mac/XPCService/XPCServiceMain.mm:

(WebKit::handleXPCBootstrap):
(main):
Set localization very early in process lifetime (in development builds, that's
even before WebKit framework is loaded).

  • UIProcess/Launcher/mac/ProcessLauncherMac.mm: (WebKit::connectToService):

Pass UI process localization to the service, computing it in the same way we do
for child processes.

Tools:

  • WebKitTestRunner/InjectedBundle/mac/InjectedBundleMac.mm:

(WTR::InjectedBundle::platformInitialize): It was too late to override language
here, and it's not necessary any more. Keeping the code for older OS versions in
case it did have effect in some code paths.

11:07 AM Changeset in webkit [169121] by mhahnenberg@apple.com
  • 8 edits
    21 adds in trunk

JSArray::shiftCountWith* could be more efficient
​https://bugs.webkit.org/show_bug.cgi?id=133011

Reviewed by Geoffrey Garen.

Source/JavaScriptCore:
Our current implementations of shiftCountWithAnyIndexingType and shiftCountWithArrayStorage
are scared of the presence of any holes in the array. We can mitigate this somewhat by enabling
them to correctly handle holes, thus avoiding the slowest of slow paths in most cases.

  • runtime/ArrayStorage.h:

(JSC::ArrayStorage::indexingHeader):
(JSC::ArrayStorage::length):
(JSC::ArrayStorage::hasHoles):

  • runtime/IndexingHeader.h:

(JSC::IndexingHeader::publicLength):
(JSC::IndexingHeader::from):

  • runtime/JSArray.cpp:

(JSC::JSArray::shiftCountWithArrayStorage):
(JSC::JSArray::shiftCountWithAnyIndexingType):
(JSC::JSArray::unshiftCountWithArrayStorage):

  • runtime/JSArray.h:

(JSC::JSArray::shiftCountForShift):
(JSC::JSArray::shiftCountForSplice):
(JSC::JSArray::shiftCount):

  • runtime/Structure.cpp:

(JSC::Structure::holesRequireSpecialBehavior):

  • runtime/Structure.h:

LayoutTests:
Added a performance regression test which tests the contiguous shift case. We're ~85% faster
on this microbenchmark with this patch.

Also added a bunch of tests for each permutation of contiguous/array storage with no holes, holes,
and holes that require special behavior (e.g. indexed properties in the prototype chain).

  • js/array-storage-splice-holes-expected.txt: Added.
  • js/array-storage-splice-holes-require-special-behavior-expected.txt: Added.
  • js/array-storage-splice-holes-require-special-behavior.html: Added.
  • js/array-storage-splice-holes.html: Added.
  • js/array-storage-splice-no-holes-expected.txt: Added.
  • js/array-storage-splice-no-holes.html: Added.
  • js/contiguous-splice-holes-expected.txt: Added.
  • js/contiguous-splice-holes-require-special-behavior-expected.txt: Added.
  • js/contiguous-splice-holes-require-special-behavior.html: Added.
  • js/contiguous-splice-holes.html: Added.
  • js/contiguous-splice-no-holes-expected.txt: Added.
  • js/contiguous-splice-no-holes.html: Added.
  • js/regress/array-splice-contiguous-expected.txt: Added.
  • js/regress/array-splice-contiguous.html: Added.
  • js/regress/script-tests/array-splice-contiguous.js: Added.

(foo):

  • js/script-tests/array-storage-splice-holes-require-special-behavior.js: Added.
  • js/script-tests/array-storage-splice-holes.js: Added.
  • js/script-tests/array-storage-splice-no-holes.js: Added.
  • js/script-tests/contiguous-splice-holes-require-special-behavior.js: Added.
  • js/script-tests/contiguous-splice-holes.js: Added.
  • js/script-tests/contiguous-splice-no-holes.js: Added.
10:38 AM Changeset in webkit [169120] by stavila@adobe.com
  • 3 edits
    2 adds in trunk

REGRESSION: [CSS Regions] Content flowed directly into the flow thread that ends up in the second region is not properly repainted
​https://bugs.webkit.org/show_bug.cgi?id=133111

Reviewed by David Hyatt.

Source/WebCore:
When computing the repaint rect of an element flowed into a flow thread, if the element is flowed directly into
the flow thread (meaning its containing block is the flow thread itself), the region's position within the flow
must no longer be taken into consideration, because its already included in the element's |topLeft|.

Test: fast/regions/hover-element-flowed-second-region.html

  • rendering/RenderBox.cpp:

(WebCore::RenderBox::computeRectForRepaint):

LayoutTests:
Added test for the correct repaint of an element flowed directly into the second region of a flow thread.

  • fast/regions/hover-element-flowed-second-region-expected.html: Added.
  • fast/regions/hover-element-flowed-second-region.html: Added.
10:29 AM Changeset in webkit [169119] by commit-queue@webkit.org
  • 2 edits in trunk/Source/WebCore

Fix web timing assertion failure.
​https://bugs.webkit.org/show_bug.cgi?id=133094
<rdar://problem/16966032>

Patch by Alex Christensen <​achristensen@webkit.org> on 2014-05-20
Reviewed by Alexey Proskuryakov.

  • platform/network/mac/ResourceHandleMac.mm:

(WebCore::ResourceHandle::getConnectionTimingData):
Set requestStart and responseStart to 0 instead of -1 to match the
ResourceLoadTiming constructor and prevent the assertion failure in
PerformanceTiming::responseStart.

10:13 AM Changeset in webkit [169118] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

Only set title on SVG documents
​https://bugs.webkit.org/show_bug.cgi?id=133068

Patch by Prashant Hiremath <​hiremathprashants@gmail.com> on 2014-05-20
Reviewed by Dirk Schulze.

Source/WebCore:
Test: svg/custom/html_document_set_title.html

  • svg/SVGTitleElement.cpp:

(WebCore::SVGTitleElement::insertedInto):
(WebCore::SVGTitleElement::removedFrom):
(WebCore::SVGTitleElement::childrenChanged):

LayoutTests:

  • svg/custom/html_document_set_title-expected.txt: Added.
  • svg/custom/html_document_set_title.html: Added.
10:08 AM Changeset in webkit [169117] by ap@apple.com
  • 2 edits in trunk/LayoutTests

REGRESSION(r169105): fast/regions/cssom/region-range-for-box-crash.html is more crashy than before
​https://bugs.webkit.org/show_bug.cgi?id=133124

10:06 AM Changeset in webkit [169116] by andersca@apple.com
  • 2 edits in trunk/LayoutTests

Skip http/tests/inspector/replay/document-last-modified-fallback-value.html since it asserts.

10:01 AM Changeset in webkit [169115] by commit-queue@webkit.org
  • 3 edits in trunk/Source/WebCore

[Curl] Invalid content in cache file, causes broken rendering.
​https://bugs.webkit.org/show_bug.cgi?id=133069

Patch by ​peavo@outlook.com <​peavo@outlook.com> on 2014-05-20
Reviewed by Brent Fulgham.

When data for a url is received as multiple parts, the cache file for the url is truncated when opened for writing,
and will only contain the last part of data received.
This is fixed by only opening the file once, and close it after all data has been received.

  • platform/network/curl/CurlCacheEntry.cpp:

(WebCore::CurlCacheEntry::CurlCacheEntry): Initialize cache file handle member.
(WebCore::CurlCacheEntry::~CurlCacheEntry): Close cache file.
(WebCore::CurlCacheEntry::saveCachedData): Only open cache file once to avoid truncating.
(WebCore::CurlCacheEntry::didFail): Close cache file.
(WebCore::CurlCacheEntry::didFinishLoading): Ditto.
(WebCore::CurlCacheEntry::openContentFile): Added method to open cache file.
(WebCore::CurlCacheEntry::closeContentFile): Added method to close cache file.

  • platform/network/curl/CurlCacheEntry.h: Added file handle member, and methods to open and close cache file.
9:59 AM Changeset in webkit [169114] by andersca@apple.com
  • 2 edits in trunk/Source/WebKit2

Fix a header comment
​https://bugs.webkit.org/show_bug.cgi?id=133123
<rdar://problem/16973783>

Reviewed by Geoffrey Garen.

  • UIProcess/API/Cocoa/WKUserContentController.h:
9:23 AM Changeset in webkit [169113] by dfarler@apple.com
  • 9 edits in trunk/Tools

Move WebKitTestRunner to std::unique_ptr
​https://bugs.webkit.org/show_bug.cgi?id=133081

Reviewed by Daniel Bates.

  • WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:

(WTR::InjectedBundle::didCreatePage):

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

(WTR::TestController::initialize):
(WTR::TestController::createWebViewWithOptions):
(WTR::TestController::resetStateToConsistentValues):
(WTR::TestController::runTest):

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

(WTR::WorkQueueManager::processWorkQueue):
(WTR::WorkQueueManager::enqueue):

  • WebKitTestRunner/WorkQueueManager.h:
8:20 AM Changeset in webkit [169112] by Carlos Garcia Campos
  • 5 edits in trunk

[GTK] WebKitWebPage::send-request always pass a valid pointer for redirected response
​https://bugs.webkit.org/show_bug.cgi?id=133119

Reviewed by Sergio Villar Senin.

Source/WebKit2:
The problem is that we were checking whether the passed in
WKURLResponseRef is NULL or not, but it's always a valid object
that can contain a NULL WebCore::ResourceResponse.

  • WebProcess/InjectedBundle/API/gtk/WebKitWebPage.cpp:

(willSendRequestForFrame): Pass NULL as redirect-response
parameter of WebKitWebPage::send-request signal when the
WebCore::ResourceResponse of the passed in WKURLResponseRef is NULL.

Tools:
Check that redirect response parameter of
WebKitWebPage::send-request signal is NULL when not redirecting
and a valid WebKitURIResponse object when redirecting.

  • TestWebKitAPI/Tests/WebKit2Gtk/TestResources.cpp:

(testWebResourceSendRequest):
(serverCallback):

  • TestWebKitAPI/Tests/WebKit2Gtk/WebExtensionTest.cpp:

(sendRequestCallback):

7:47 AM Changeset in webkit [169111] by ddkilzer@apple.com
  • 3 edits in trunk/LayoutTests

Make fast/dom/remove-element-with-{id|name}-that-was-inserted-on-DOMNodeRemoved.html reliable
<​http://webkit.org/b/133088>

Reviewed by Alexey Proskuryakov.

  • fast/dom/remove-element-with-id-that-was-inserted-on-DOMNodeRemoved.html:
  • fast/dom/remove-element-with-name-that-was-inserted-on-DOMNodeRemoved.html:
  • Use waitUntilDone()/notifyDone() to make the tests reliable since they use setTimeout().
7:34 AM Changeset in webkit [169110] by stavila@adobe.com
  • 5 edits
    4 adds in trunk

[CSS Regions] Block incorrectly sized when containing an unsplittable box
​https://bugs.webkit.org/show_bug.cgi?id=132601

Reviewed by Antti Koivisto.

Source/WebCore:
When laying out elements in a region, when an inline element is encountered
the size of its parent must not be increased beyond the bottom of the current region,
unless if its the last region. This will ensure that the next sibling of the
inline element is correctly laid out at the top of the next region, instead
of leaving an empty space equal to the height of the overflow, as it did until now.

Tests: fast/regions/inline-block-inside-anonymous-overflow.html

fast/regions/inline-block-overflow.html

  • rendering/RenderBlockFlow.cpp:

(WebCore::RenderBlockFlow::adjustLinePositionForPagination):
(WebCore::RenderBlockFlow::hasNextPage):

  • rendering/RenderBlockFlow.h:
  • rendering/RenderBlockLineLayout.cpp:

(WebCore::RenderBlockFlow::layoutRunsAndFloatsInRange):
(WebCore::RenderBlockFlow::linkToEndLineIfNeeded):
(WebCore::RenderBlockFlow::determineStartPosition):
(WebCore::RenderBlockFlow::checkPaginationAndFloatsAtEndLine):

LayoutTests:
Added tests for the layout of elements following inline-block elements
that overflow their region, with and without anonymous blocks.

  • fast/regions/inline-block-inside-anonymous-overflow-expected.html: Added.
  • fast/regions/inline-block-inside-anonymous-overflow.html: Added.
  • fast/regions/inline-block-overflow-expected.html: Added.
  • fast/regions/inline-block-overflow.html: Added.
6:56 AM Changeset in webkit [169109] by Antti Koivisto
  • 2 edits in trunk/Source/WebKit2

Double-tap zoom does not take obscuring insets into account
​https://bugs.webkit.org/show_bug.cgi?id=133116
<rdar://problem/16765604>

Reviewed by Anders Carlsson.

  • UIProcess/API/Cocoa/WKWebView.mm:

(-[WKWebView _zoomToRect:WebCore::atScale:origin:WebCore::]):

Compute the zoom target taking insets into account.

6:39 AM WebKitGTK/2.4.x edited by Carlos Garcia Campos
(diff)
6:15 AM Changeset in webkit [169108] by Carlos Garcia Campos
  • 4 edits in trunk

[GTK] Rename translation domain as WebKit2GTK-3.0
​https://bugs.webkit.org/show_bug.cgi?id=132953

Reviewed by Gustavo Noronha Silva.

.:

  • Source/cmake/OptionsGTK.cmake:

Source/WebCore/platform/gtk/po:

  • CMakeLists.txt:
5:34 AM Changeset in webkit [169107] by zoltan@webkit.org
  • 1 edit
    2 adds in trunk/LayoutTests

[CSS Regions] Add polygon tests for shapes on regions and shapes on the content flow
​https://bugs.webkit.org/show_bug.cgi?id=133107

Reviewed by Mihnea Ovidenie.

  • fast/regions/shape-outside-floats/regions-shape-outside-region-polygon-floats-expected.html: Added.
  • fast/regions/shape-outside-floats/regions-shape-outside-region-polygon-floats.html: Added.
5:17 AM Changeset in webkit [169106] by Carlos Garcia Campos
  • 2 edits in trunk/Tools

REGRESSION(CMAKE): [GTK] Several WebKit2GTK+ unit tests are no longer built
​https://bugs.webkit.org/show_bug.cgi?id=133117

Reviewed by Philippe Normand.

Bring back missing tests to the build.

  • TestWebKitAPI/Tests/WebKit2Gtk/CMakeLists.txt:
3:57 AM Changeset in webkit [169105] by mihnea@adobe.com
  • 6 edits
    2 adds in trunk

[CSS Regions] Crash while painting block selection gaps in regions
​https://bugs.webkit.org/show_bug.cgi?id=132720

Reviewed by David Hyatt.

Source/WebCore:
The fix for WebKit bug ​https://bugs.webkit.org/show_bug.cgi?id=131511
allowed selection highlight to match the DOM selection when the start
and end point of the selection were in different flow threads. In order
to enable that, the selection was performed separately on view and
render flow threads, considered selection subtrees.

However, the start and end points for each selection subtree were computed
by means of Range class but it is not always possible to construct a valid
Range from two pairs of RenderObjects and offsets.

This patch keeps the substrees approach but instead of storing the endpoints
for each subtree in a Range and continuously extending the Range, it stores them
using the already available SelectionSubtreeRoot class. After the end points are
computed for each subtree and before processing the subtree selection, the end points
are adjusted in a similar fashion as the one used in FrameSelection::updateAppearance(),
to make sure we are passing the same expected information to the method implementing
the visible selection processing.

Test: fast/regions/selection-gaps-paint-crash.html

  • rendering/RenderView.cpp:

(WebCore::RenderView::splitSelectionBetweenSubtrees):

  • rendering/SelectionSubtreeRoot.cpp:

(WebCore::SelectionSubtreeRoot::adjustForVisibleSelection):

  • rendering/SelectionSubtreeRoot.h:

(WebCore::SelectionSubtreeRoot::selectionClear):

LayoutTests:

  • TestExpectations: Unskip a test that was crashing
  • fast/regions/selection-gaps-paint-crash-expected.txt: Added.
  • fast/regions/selection-gaps-paint-crash.html: Added.
2:52 AM Changeset in webkit [169104] by commit-queue@webkit.org
  • 41 edits in trunk/LayoutTests

[GTK] Unreviewed GTK gardening

Rebaseline affected tests by changes on integral snapping of inlines
introduced at r169048.

Patch by Lorenzo Tilve <​ltilve@igalia.com> on 2014-05-20

  • platform/gtk/css1/box_properties/float_on_text_elements-expected.txt:
  • platform/gtk/css1/text_properties/text_indent-expected.txt:
  • platform/gtk/css2.1/t0905-c414-flt-wrap-00-e-expected.txt:
  • platform/gtk/css2.1/t0905-c5525-fltwidth-00-c-g-expected.txt:
  • platform/gtk/css2.1/t100801-c544-valgn-00-a-ag-expected.txt:
  • platform/gtk/css2.1/t100801-c544-valgn-02-d-agi-expected.txt:
  • platform/gtk/css2.1/t100801-c544-valgn-03-d-agi-expected.txt:
  • platform/gtk/css2.1/t100801-c544-valgn-04-d-agi-expected.txt:
  • platform/gtk/css2.1/t1601-c547-indent-00-b-a-expected.txt:
  • platform/gtk/fast/block/float/026-expected.txt:
  • platform/gtk/fast/block/float/028-expected.txt:
  • platform/gtk/fast/css/empty-pseudo-class-expected.txt:
  • platform/gtk/fast/css/first-child-pseudo-class-expected.txt:
  • platform/gtk/fast/css/first-of-type-pseudo-class-expected.txt:
  • platform/gtk/fast/css/last-child-pseudo-class-expected.txt:
  • platform/gtk/fast/css/last-of-type-pseudo-class-expected.txt:
  • platform/gtk/fast/css/only-child-pseudo-class-expected.txt:
  • platform/gtk/fast/css/only-of-type-pseudo-class-expected.txt:
  • platform/gtk/fast/dom/navigator-detached-no-crash-expected.txt:
  • platform/gtk/fast/line-grid/line-align-left-edges-expected.txt:
  • platform/gtk/fast/multicol/column-count-with-rules-expected.txt:
  • platform/gtk/fast/multicol/float-multicol-expected.txt:
  • platform/gtk/fast/multicol/layers-in-multicol-expected.txt:
  • platform/gtk/fast/multicol/vertical-lr/float-multicol-expected.txt:
  • platform/gtk/fast/multicol/vertical-rl/float-multicol-expected.txt:
  • platform/gtk/fast/regions/multiple-directionality-changes-in-variable-width-regions-expected.txt:
  • platform/gtk/fast/regions/repaint/line-flow-with-floats-in-regions-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-1-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-10-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-2-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-3-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-4-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-5-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-6-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-7-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-8-expected.txt:
  • platform/gtk/fast/repaint/line-flow-with-floats-9-expected.txt:
  • platform/gtk/fast/text/whitespace/024-expected.txt:
  • platform/gtk/svg/wicd/rightsizing-grid-expected.txt:
  • platform/gtk/svg/zoom/page/zoom-replaced-intrinsic-ratio-001-expected.txt:
Note: See TracTimeline for information about the timeline view.