Timeline



Sep 9, 2019:

10:30 PM Changeset in webkit [249707] by Alan Coon
  • 8 edits
    3 adds in branches/safari-608-branch

Cherry-pick r249535. rdar://problem/55183140

10:03 PM Changeset in webkit [249706] by ysuzuki@apple.com
  • 7 edits in trunk/Source/JavaScriptCore

[JSC] CodeBlock::m_constantRegisters should be guarded by ConcurrentJSLock when Vector reallocate memory
https://bugs.webkit.org/show_bug.cgi?id=201622

Reviewed by Mark Lam.

CodeBlock::visitChildren takes ConcurrentJSLock while iterating m_constantRegisters, some of the places reallocate
this Vector without taking a lock. If a Vector memory is reallocated while iterating it in concurrent collector,
the concurrent collector can see a garbage. This patch guards m_constantRegisters reallocation with ConcurrentJSLock.

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finishCreation):
(JSC::CodeBlock::setConstantRegisters):

  • bytecode/CodeBlock.h:

(JSC::CodeBlock::addConstant):
(JSC::CodeBlock::addConstantLazily):

  • dfg/DFGDesiredWatchpoints.cpp:

(JSC::DFG::ArrayBufferViewWatchpointAdaptor::add):
(JSC::DFG::SymbolTableAdaptor::add):
(JSC::DFG::FunctionExecutableAdaptor::add):

  • dfg/DFGGraph.cpp:

(JSC::DFG::Graph::registerFrozenValues):

  • dfg/DFGJITFinalizer.cpp:

(JSC::DFG::JITFinalizer::finalizeCommon):

  • dfg/DFGLazyJSValue.cpp:

(JSC::DFG::LazyJSValue::emit const):

9:51 PM Changeset in webkit [249705] by achristensen@apple.com
  • 2 edits in trunk/Tools

Rebase API test after r249675
https://bugs.webkit.org/show_bug.cgi?id=201596

  • TestWebKitAPI/Tests/WebKitCocoa/WebsitePolicies.mm:

(-[WebsitePoliciesWebsiteDataStoreDelegate _webView:decidePolicyForNavigationAction:decisionHandler:]):
We removed the restriction on WebsitePolicies with non-default persistent sessions.
We needed to change the expectations of the test that exercised that code path.

9:30 PM Changeset in webkit [249704] by achristensen@apple.com
  • 2 edits in trunk/Tools

Fix iOS API tests after r249684
https://bugs.webkit.org/show_bug.cgi?id=201573

  • TestWebKitAPI/Tests/WebKitCocoa/TLSDeprecation.mm:

(TestWebKitAPI::getWebSocketEventWebKitLegacy):
(TestWebKitAPI::TEST):
The web thread still exists :(
That makes iOS immediate failure results more like WK2, which is fine.
The point is that the connection fails somehow and the server sees a TLS failure, which the test verifies.

9:24 PM Changeset in webkit [249703] by Chris Dumez
  • 7 edits in trunk/Source/WebKit

[iOS] We sometimes attempt to use a terminated prewarmed WebContent process
https://bugs.webkit.org/show_bug.cgi?id=201614
<rdar://problem/54714507>

Reviewed by Geoffrey Garen.

On iOS, it is possible for our processes to get terminated (e.g. jetsammed) while the UIProcess
is suspended. Upon resuming, it takes a little while for the UIProcess to get the notification
that the mac connection to its child process has been severed and the UIProcess may try to use
it for a load. This is especially problematic for prewarmed process because the client will end
up showing a crash banner and reloading when we could have used a new process rather the prewarmed
one if we had known it was dead.

This patch makes 2 improvements:

  1. It makes AuxiliaryProcessProxy::state() return Terminated if we still have a connection but the PID is not the PID of a running process. I also added a check in tryTakePrewarmedProcess() to not use the prewarmed process if it state() is Terminated.
  2. When the UIProcess is about to get suspended, have the process pools terminate their non-critical processes (i.e. prewarmed + the ones used for PageCache). This makes WebKit friendlier with other apps on the system when suspended with regards to memory. Also, it makes it less likely useful WebContent processes will get jetsammed.
  • UIProcess/AuxiliaryProcessProxy.cpp:

(WebKit::AuxiliaryProcessProxy::state const):
(WebKit::AuxiliaryProcessProxy::isRunningProcessPID):

  • UIProcess/AuxiliaryProcessProxy.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::applicationIsAboutToSuspend):

  • UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::tryTakePrewarmedProcess):

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

(-[WKProcessAssertionBackgroundTaskManager init]):
(-[WKProcessAssertionBackgroundTaskManager _releaseBackgroundTask]):

9:06 PM Changeset in webkit [249702] by Chris Dumez
  • 15 edits in trunk

REGRESSION: http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations.html is frequently timing out on iOS EWS bots
https://bugs.webkit.org/show_bug.cgi?id=201550

Reviewed by Alex Christensen.

Source/WebCore:

  • loader/ResourceLoadObserver.h:

(WebCore::ResourceLoadObserver::hasStatistics const):

Source/WebKit:

  • WebProcess/InjectedBundle/API/c/WKBundle.cpp:

(WKBundleResourceLoadStatisticsNotifyObserver):

  • WebProcess/InjectedBundle/API/c/WKBundlePrivate.h:

Tools:

Update testRunner.statisticsNotifyObserver() to return a boolean indicating if it did
anything or not.

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

(WTR::InjectedBundle::statisticsNotifyObserver):

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

(WTR::TestRunner::statisticsNotifyObserver):

  • WebKitTestRunner/InjectedBundle/TestRunner.h:

LayoutTests:

Update test to do checks on a timer if its call to testRunner.statisticsNotifyObserver() is a no-op
due to having no statistics to send to the network process.

  • http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations.html:
8:50 PM Changeset in webkit [249701] by timothy@apple.com
  • 4 edits
    1 add in trunk

Tap and hold on Facebook sometimes creates a tall empty selection.
https://bugs.webkit.org/show_bug.cgi?id=201618
rdar://53630145

Reviewed by Megan Gardner.

Source/WebCore:

API Test: SelectionTests.ByWordAtEndOfDocument

  • editing/VisibleUnits.cpp:

(WebCore::wordRangeFromPosition):
Remove special case code for the possibility of an empty paragraph and at the end
of the document. This is no longer needed and was causing a large selection to be
created on Facebook due to large areas of non-selectable content on the page.

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/ios/SelectionByWord.mm: Added.
8:20 PM Changeset in webkit [249700] by Alan Coon
  • 6 edits in branches/safari-608-branch/Source/WebKit

Cherry-pick r249649. rdar://problem/55198071

[macOS] Pid is sometimes invalid when creating sandbox extensions by pid.
https://bugs.webkit.org/show_bug.cgi?id=201543
<rdar://problem/54733465>

Reviewed by Brent Fulgham.

There is a race condition when starting a load of a local file, where the WebContent process has not finished
launching yet, and its pid is not available. When we try to create a sandbox extension by using the pid of the
WebContent process, it is not available in the cases where the WebContent process has just launched and has not
finished launching yet. This patch creates a new dummy Web page message, 'LoadRequestWaitingForPID', which will
be sent instead of a normal 'LoadRequest' message, and only when the WebContent process has not finished
launching. When the WebContent process has finished launching, and we are about to actually send the pending
messages, we can detect that a 'LoadRequestWaitingForPID' has been appended for sending, and replace it with a
normal 'LoadReqest' message where we have created the sandbox extension issue with a valid pid. The message
'LoadRequestWaitingForPID' is never intended to reach the WebContent process, it is just there to replace with
a normal 'LoadRequest' message with a new sandbox extension. In the implementation of the message handler on
the WebContent process side, we assert that the method is never called. This patch makes sure the ordering of
the Web page messages are the same, even when we modify the message.

  • UIProcess/AuxiliaryProcessProxy.cpp: (WebKit::AuxiliaryProcessProxy::didFinishLaunching):
  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle): (WebKit::WebPageProxy::loadRequestWithNavigationShared): (WebKit::WebPageProxy::loadFile):
  • WebProcess/WebPage/WebPage.cpp: (WebKit::WebPage::fileLoadRequest):
  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:20 PM Changeset in webkit [249699] by Alan Coon
  • 2 edits in branches/safari-608-branch/Source/WebKit

Cherry-pick r249624. rdar://problem/55202935

Marking up a note on iOS results in a PDF with no contents
https://bugs.webkit.org/show_bug.cgi?id=201530
<rdar://problem/53686019>

Unreviewed follow-up.

  • UIProcess/ios/WKContentView.mm: (-[WKContentView _waitForDrawToPDFCallback]): Remove an ASSERT_NOT_REACHED that fires if the synchronous wait for printing times out. This is a totally legitimite situation that can occur if the process crashes; there's no need for an assert there.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249624 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:20 PM Changeset in webkit [249698] by Alan Coon
  • 16 edits
    5 adds in branches/safari-608-branch

Cherry-pick r249605. rdar://problem/55182896

Incorrect selection rect revealed after pasting images in a contenteditable element
https://bugs.webkit.org/show_bug.cgi?id=201549
<rdar://problem/50956429>

Reviewed by Simon Fraser.

Source/WebCore:

Editor::replaceSelectionWithFragment currently scrolls to reveal the selection after inserting the given
DocumentFragment. However, this scrolling occurs before any inserted images have loaded yet, which causes the
wrong caret rect to be revealed, since all image elements inserted during paste will be empty.

To fix this, we defer revealing the selection after inserting the fragment until after all images that have
been inserted are done loading. While waiting for images to load, if any layers which may be scrolled as a
result of revealing the selection are scrolled, we additionally cancel the deferred selection reveal. See
comments below for more detail.

Tests: editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll.html

editing/pasteboard/reveal-selection-after-pasting-images.html
PasteImage.RevealSelectionAfterPastingImage

  • editing/Editing.cpp: (WebCore::visibleImageElementsInRangeWithNonLoadedImages):

Add a new helper to iterate through a range and collect all image elements in that range, that contain cached
images that have not finished loading yet.

  • editing/Editing.h:
  • editing/Editor.cpp: (WebCore::Editor::replaceSelectionWithFragment):

Instead of always immediately revealing the selection after applying the ReplaceSelectionCommand, collect the
image elements that were just inserted, and avoid immediately revealing the selection if any of these images
have non-null cached images, but are not loaded yet. Instead, hold on to these images in a set, remove them once
they finish loading using the new method below, and once all images are removed, reveal the selection.

(WebCore::Editor::revealSelectionIfNeededAfterLoadingImageForElement):
(WebCore::Editor::renderLayerDidScroll):

Called whenever a scrollable RenderLayer is scrolled (or in the case of FrameView, the root layer). In the case
where Editor is waiting to reveal the selection, we check to see if the scrolled layer is an ancestor of the
layer enclosing the start of the selection.

(WebCore::Editor::respondToChangedSelection):

If the selection changes between pasting and waiting for pasted images to load, just cancel waiting to reveal
the selection after pasting.

  • editing/Editor.h:
  • editing/ReplaceSelectionCommand.cpp: (WebCore::ReplaceSelectionCommand::insertedContentRange const):

Add a helper method to grab the Range of content inserted after applying the command.

  • editing/ReplaceSelectionCommand.h:
  • page/FrameView.cpp: (WebCore::FrameView::scrollPositionChanged):
  • page/FrameView.h:
  • page/Page.cpp: (WebCore::Page::didFinishLoadingImageForElement):

Notify Editor after an image finishes loading.

  • rendering/RenderLayer.cpp: (WebCore::RenderLayer::scrollTo):

Source/WebKit:

Tweak some existing logic to use the new visibleImageElementsInRangeWithNonLoadedImages helper function. See
WebCore for more details.

  • WebProcess/WebPage/ios/WebPageIOS.mm: (WebKit::WebPage::didConcludeEditDrag):

Tools:

Add an API test to exercise the scenario where we scroll to reveal the selection after pasting an image that was
directly written to the pasteboard.

  • TestWebKitAPI/Tests/WebKitCocoa/PasteImage.mm:

LayoutTests:

Add a couple of new layout tests.

  • editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll-expected.txt: Added.
  • editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll.html: Added.

This test verifies that we don't try to scroll to reveal the caret after pasting, if the scroll position was
changed before the images finished loading.

  • editing/pasteboard/reveal-selection-after-pasting-images-expected.txt: Added.
  • editing/pasteboard/reveal-selection-after-pasting-images.html: Added.
  • platform/ios/editing/pasteboard/reveal-selection-after-pasting-images-expected.txt: Added.

This test verifies that we reveal the caret after loading multiple pasted images in a selection, and dispatch a
scroll event in the process.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249605 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249697] by Alan Coon
  • 6 edits
    1 add in branches/safari-608-branch

Cherry-pick r249584. rdar://problem/55202935

Marking up a note on iOS results in a PDF with no contents
https://bugs.webkit.org/show_bug.cgi?id=201530
<rdar://problem/53686019>

Reviewed by Andy Estes.

Source/WebKit:

  • Platform/IPC/Connection.cpp: (IPC::Connection::Connection): (IPC::Connection::waitForMessage): (IPC::Connection::connectionDidClose):
  • Platform/IPC/Connection.h: If the main thread is blocked when the Web Content process dies, and something eventually calls waitForAndDispatchImmediately without returning control to the main run loop, we will wait for the full timeout, because a) the code to mark the connection invalid is dispatched to the main thread, and b) the secondary thread that is informed of the Web Content process dying did not yet have a "waiting for" message to mark as interrupted (because it wasn't waiting yet).

Fix this race by adding a bit that is set under the waitForMessage lock
on the secondary thread when the connection is invalidated, identically
to m_shouldWaitForSyncReplies, which solves the same problem for sync
messages.

Read the new bit when we are about to start waiting, and bail if it is set.
It's OK to not read it inside the loop because we are guaranteed to have
waitForMessage set at that point, so the normal interruption bit will work.

  • UIProcess/ios/WKContentView.mm: (-[WKContentView _processDidExit]): Reset _isPrintingToPDF; the Web Content process is never going to get back to us if it crashes.

(-[WKContentView _wk_pageCountForPrintFormatter:]):
Do not bail from starting a printing operation if one is already occurring.
This fixes the original bug, because Markup ends up invalidating the page
count at least one extra time before asking for the printed document.
Instead of maintaining the fragile requirement that you cannot recompute
the page count while printing, just let it happen. In order to make this
work safely, synchronously wait for the previous printed result before
continuing with the next print.

We could do more coalescing here if need be, but calls to -_recalcPageCount
are not high in volume.

Tools:

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKitCocoa/WKWebViewPrintFormatter.mm: Add some tests for WKWebViewPrintFormatter; specifically that it is possible to _recalcPageCount twice in quick succession, and that we don't hang if we start painting the printed content immediately after a Web Content process crash.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249584 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249696] by Alan Coon
  • 4 edits
    2 adds in branches/safari-608-branch

Cherry-pick r249581. rdar://problem/55202922

REGRESSION (iOS 13): If an overflow:hidden with a non-zero scroll position is toggled to overflow:scroll, some other scroll causes its scroll position to get reset
https://bugs.webkit.org/show_bug.cgi?id=201528
rdar://problem/55044885

Reviewed by Frédéric Wang.
Source/WebCore:

If, when an overflow scrolling node is created, the scroller has non-zero scroll
position (for example, via toggling to overflow:hidden, setting scrollTop, then toggling
to overflow:scroll), then on the next update its scroll position will reset back to zero.

The bug was that newly created ScrollingTreeScrollingNodes didn't set m_currentScrollPosition
to the scroll position coming from the state node, so a subsequent update could cause
the 0,0 currentScrollPosition to get applied. If we're making a new node, and there's no
requestedScrollPosition, then initialize m_currentScrollPosition.

Test: scrollingcoordinator/ios/scroller-initial-scroll-position.html

  • page/scrolling/ScrollingTreeScrollingNode.cpp: (WebCore::ScrollingTreeScrollingNode::commitStateBeforeChildren): (WebCore::ScrollingTreeScrollingNode::commitStateAfterChildren):
  • page/scrolling/ScrollingTreeScrollingNode.h:

LayoutTests:

  • scrollingcoordinator/ios/scroller-initial-scroll-position-expected.html: Added.
  • scrollingcoordinator/ios/scroller-initial-scroll-position.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249581 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249695] by Alan Coon
  • 5 edits in branches/safari-608-branch

Cherry-pick r249565. rdar://problem/55113261

AccessibilityRenderObject::setSelectedTextRange fails to set the selection passed an empty line.
https://bugs.webkit.org/show_bug.cgi?id=201518
<rdar://problem/54835122>

Patch by Andres Gonzalez <Andres Gonzalez> on 2019-09-06
Reviewed by Ryosuke Niwa.

Source/WebCore:

Test: accessibility/set-selected-text-range-after-newline.html

In the case of an empty line, the CharacterIterator range start and end
were not equal, thus we were not advancing the iterator and returning
the iterator range end, which is not correct. With this change we are
always advancing the iterator if its text is just '\n'. This covers all
the cases we fixed before plus empty lines.

  • editing/Editing.cpp: (WebCore::visiblePositionForIndexUsingCharacterIterator):

LayoutTests:

Extended this test to set the selection range passed an empty line.

  • accessibility/set-selected-text-range-after-newline-expected.txt:
  • accessibility/set-selected-text-range-after-newline.html:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249565 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249694] by Alan Coon
  • 3 edits
    2 adds in branches/safari-608-branch

Cherry-pick r249534. rdar://problem/55183098

AX: children cache are not re-computed if tab index is removed
https://bugs.webkit.org/show_bug.cgi?id=201502

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: accessibility/tabindex-removed.html

If the tabindex changes, it can potentially affect whether an element is accessible. If we don't update the children cache
information can be stale and lead to incorrect navigation with VoiceOver.

  • accessibility/AXObjectCache.cpp: (WebCore::AXObjectCache::handleAttributeChange):

LayoutTests:

  • accessibility/tabindex-removed-expected.txt: Added.
  • accessibility/tabindex-removed.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249693] by Alan Coon
  • 5 edits in branches/safari-608-branch

Cherry-pick r249514. rdar://problem/55182886

MobileSafari may crash when invoking the C++ lambda in -[WKContentView _shareForWebView:]
https://bugs.webkit.org/show_bug.cgi?id=201479
<rdar://problem/51511834>

Reviewed by Tim Horton.

Source/WebKit:

Fix the crash by making -_shareForWebView: robust in the case where there are no selection rects
known in the UI process when -[WKContentView _share:] is invoked.

  • UIProcess/ios/WKContentViewInteraction.mm: (-[WKContentView _shareForWebView:]):

Tools:

Add a test to verify that the UI process doesn't crash when invoking _share: while there's no selection.

  • TestWebKitAPI/Tests/WebKitCocoa/WKContentViewEditingActions.mm:
  • TestWebKitAPI/ios/UIKitSPI.h:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249514 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249692] by Alan Coon
  • 2 edits in branches/safari-608-branch/Source/WebCore

Cherry-pick r249493. rdar://problem/55182906

[iPadOS] Unable to change sheets on Airtable.com
https://bugs.webkit.org/show_bug.cgi?id=201456
<rdar://problem/51557377>

Patch by Antoine Quint <Antoine Quint> on 2019-09-04
Reviewed by Dean Jackson.

Simulated mouse events are required to be able to manipulate cells and and columns on Airtable.com. However, dispatching a "mousedown" event on
tabs allowing to pick a different sheet ends up calling preventDefault() and prevent "click" events from being dispatched, which makes it
impossible to change sheet. We now limit the dispatch of simulated mouse events to the grid.

  • page/Quirks.cpp: (WebCore::Quirks::simulatedMouseEventTypeForTarget const):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249493 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249691] by Alan Coon
  • 2 edits in branches/safari-608-branch/Source/WebKit

Cherry-pick r249454. rdar://problem/55198071

[macOS] Unable to open local file from favorites bar
https://bugs.webkit.org/show_bug.cgi?id=201444

Reviewed by Brent Fulgham.

The sandbox extension handle should be created providing the pid of the receiving process.

  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249690] by Alan Coon
  • 2 edits in branches/safari-608-branch/Source/WebKit

Cherry-pick r249444. rdar://problem/55093558

Null deref under -[WKWebView _addUpdateVisibleContentRectPreCommitHandler]'s handler block
https://bugs.webkit.org/show_bug.cgi?id=201436
<rdar://problem/40640475>

Reviewed by Simon Fraser.

  • UIProcess/API/Cocoa/WKWebView.mm: (-[WKWebView dealloc]): (-[WKWebView _addUpdateVisibleContentRectPreCommitHandler]): We crash sending a message to a deallocated WKWebView inside the handler block passed to +[CATransaction addCommitHandler:]. This seems impossible, because we carefully retain it, but it's possible that it could be the result of the handler block being installed under -dealloc (in which case retaining the WKWebView wouldn't actually extend its lifetime). -[WKWebView dealloc] is fairly sizable, and it's hard to follow all paths from it, so instead add a RELEASE_LOG_FAULT, so we'll get simulated crash logs, and bail, so we'll stop actually crashing (if this is the cause).

This is just a speculative fix, but a hopeful one, since intentionally calling
-_addUpdateVisibleContentRectPreCommitHandler: from dealloc yields a similar-looking
crash under the handler block.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249444 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249689] by Alan Coon
  • 2 edits in branches/safari-608-branch/Source/WebCore

Cherry-pick r248170. rdar://problem/55183128

[iOS] Directly use RealtimeMediaSourceCenter to compute the media capture state
https://bugs.webkit.org/show_bug.cgi?id=200368
<rdar://problem/53191450>

Unreviewed.
Build fix by guarding with MEDIA_STREAM in addition to IOS.

  • dom/Document.cpp: (WebCore::Document::updateIsPlayingMedia): (WebCore::Document::pageMutedStateDidChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@248170 268f45cc-cd09-0410-ab3c-d52691b4dbfc

8:19 PM Changeset in webkit [249688] by Alan Coon
  • 4 edits in branches/safari-608-branch/Source/WebCore

Cherry-pick r248161. rdar://problem/55183128

[iOS] Directly use RealtimeMediaSourceCenter to compute the media capture state
https://bugs.webkit.org/show_bug.cgi?id=200368
<rdar://problem/53191450>

Reviewed by Eric Carlson.

Instead of registering a MediaStreamTrack as a media producer to compute capture state,
go directly to the sources from the RealtimeMediaSourceCenter.
Do the same when requested to mute capture tracks.

No observable change of behavior.
Covered by manual test on iOS and existing tests.

  • Modules/mediastream/MediaStreamTrack.cpp: (WebCore::MediaStreamTrack::MediaStreamTrack): (WebCore::MediaStreamTrack::~MediaStreamTrack): (WebCore::MediaStreamTrack::mediaState const): (WebCore::sourceCaptureState): (WebCore::MediaStreamTrack::captureState): (WebCore::MediaStreamTrack::muteCapture):
  • Modules/mediastream/MediaStreamTrack.h:
  • dom/Document.cpp: (WebCore::Document::updateIsPlayingMedia): (WebCore::Document::pageMutedStateDidChange):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@248161 268f45cc-cd09-0410-ab3c-d52691b4dbfc

7:12 PM Changeset in webkit [249687] by aakash_jain@apple.com
  • 2 edits in trunk/Tools

[EWS] Watchlist doesn't work sometimes
https://bugs.webkit.org/show_bug.cgi?id=201433

Reviewed by Michael Catanzaro.

  • Scripts/webkitpy/common/config/watchlist: Updated the email addresses to the one associated with

corresponding Bugzilla accounts.

6:57 PM Changeset in webkit [249686] by jh718.park@samsung.com
  • 48 edits
    10 adds in trunk

getComputedStyle for line-height: normal should return the keyword instead of a length
https://bugs.webkit.org/show_bug.cgi?id=201296

LayoutTests/imported/w3c:

Reviewed by Ryosuke Niwa.

Per https://github.com/w3c/csswg-drafts/issues/3749,
Gecko and Blink has this behavior already.

This patch makes WebKit has the same behavior with them.

  • web-platform-tests/css/css-inline/parsing/line-height-computed-expected.txt: Added.
  • web-platform-tests/css/css-inline/parsing/line-height-computed.html: Added.
  • web-platform-tests/css/cssom/getComputedStyle-line-height-expected.txt: Added.
  • web-platform-tests/css/cssom/getComputedStyle-line-height.html: Added.
  • web-platform-tests/html/rendering/replaced-elements/the-select-element/select-1-line-height-expected.html: Added.
  • web-platform-tests/html/rendering/replaced-elements/the-select-element/select-1-line-height.html: Added.

Source/WebCore:

Reviewed by Ryosuke Niwa.

Per https://github.com/w3c/csswg-drafts/issues/3749,
Gecko and Blink has this behavior already.

This patch makes WebKit has the same behavior with them.

Tests: imported/w3c/web-platform-tests/css/css-inline/parsing/line-height-computed.html

imported/w3c/web-platform-tests/css/cssom/getComputedStyle-line-height.html
imported/w3c/web-platform-tests/html/rendering/replaced-elements/the-select-element/select-1-line-height.html

  • css/CSSComputedStyleDeclaration.cpp:

(WebCore::lineHeightFromStyle):

LayoutTests:

Reviewed by Ryosuke Niwa.

Per https://github.com/w3c/csswg-drafts/issues/3749,
Gecko and Blink has this behavior already.

This patch makes WebKit has the same behavior with them.

  • css3/calc/line-height-expected.txt:
  • fast/css/font-calculated-value-expected.txt:
  • fast/css/font-calculated-value.html:
  • fast/css/font-shorthand-from-longhands-expected.txt:
  • fast/css/font-shorthand-from-longhands.html:
  • fast/css/font-shorthand-line-height-expected.txt:
  • fast/css/font-shorthand-line-height.html:
  • fast/css/getComputedStyle/computed-style-expected.txt:
  • fast/css/getComputedStyle/computed-style-font-expected.txt:
  • fast/css/getComputedStyle/computed-style-font.html:
  • fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • fast/ruby/ruby-line-height-expected.txt:
  • fast/ruby/ruby-line-height.html:
  • fast/text-autosizing/ios/idempotentmode/line-height-boosting-expected.txt:
  • fast/text-autosizing/ios/idempotentmode/line-height-boosting.html:
  • fast/text-autosizing/ios/line-height-text-autosizing-expected.txt:
  • fast/text-autosizing/ios/line-height-text-autosizing.html:
  • media/track/track-cue-rendering-on-resize-expected.txt:
  • media/track/track-cue-rendering-on-resize.html:
  • platform/gtk/fast/css/css2-system-fonts-expected.txt:
  • platform/gtk/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/gtk/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/gtk/svg/css/getComputedStyle-basic-expected.txt:
  • platform/ios/TestExpectations:
  • platform/ios/css3/calc/line-height-expected.txt:
  • platform/ios/fast/css/css2-system-fonts-expected.txt:
  • platform/ios/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/ios/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/ios/svg/css/getComputedStyle-basic-expected.txt:
  • platform/mac-sierra/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/mac-sierra/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/mac-sierra/svg/css/getComputedStyle-basic-expected.txt:
  • platform/mac/fast/css/css2-system-fonts-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/mac/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/mac/fast/text-autosizing/ios/line-height-text-autosizing-expected.txt:
  • platform/mac/svg/css/getComputedStyle-basic-expected.txt:
  • platform/wincairo/css3/calc/line-height-expected.txt:
  • platform/wincairo/fast/css/css2-system-fonts-expected.txt:
  • platform/wpe/fast/css/css2-system-fonts-expected.txt:
  • platform/wpe/fast/css/getComputedStyle/computed-style-expected.txt:
  • platform/wpe/fast/css/getComputedStyle/computed-style-without-renderer-expected.txt:
  • platform/wpe/svg/css/getComputedStyle-basic-expected.txt:
  • svg/css/getComputedStyle-basic-expected.txt:
6:39 PM Changeset in webkit [249685] by Justin Fan
  • 11 edits
    4 adds in trunk/Websites/webkit.org

[WebGPU] Upload Babylon.js, WSL compile, and MotionMark links to demo page
https://bugs.webkit.org/show_bug.cgi?id=201621

Reviewed by Dean Jackson.

Add new demos to the WebGPU demo collection.
Ensure error alerts/divs display properly if WebGPU is not enabled.

  • demos/webgpu/babylon/oneHelmetWebGPU.html:
  • demos/webgpu/babylon/oneSphereWebGPU.html:
  • demos/webgpu/compute-blur.html:
  • demos/webgpu/compute-boids-compile.html:
  • demos/webgpu/compute-boids.html:
  • demos/webgpu/hello-cube.html:
  • demos/webgpu/hello-triangle.html:
  • demos/webgpu/index.html:
  • demos/webgpu/resources/babylon-helmet.png: Added.
  • demos/webgpu/resources/babylon-sphere.png: Added.
  • demos/webgpu/resources/motionmark.jpeg: Added.
  • demos/webgpu/resources/webkit-logo.png: Added.
  • demos/webgpu/scripts/compute-boids-compile.js:
  • demos/webgpu/textured-cube.html:
6:39 PM Changeset in webkit [249684] by achristensen@apple.com
  • 21 edits
    1 add in trunk

Disable TLS 1.0 and 1.1 in WebSockets
https://bugs.webkit.org/show_bug.cgi?id=201573

Reviewed by Youenn Fablet.

Source/WebCore:

This expands on what I started in r249019 when I disabled legacy TLS for our use of NSURLSession.
Since our WebSocket implementation uses a different network interface, disable legacy TLS for them, too.
I use the same temporary default to re-enable legacy TLS. I also add a unit test for both WebSockets and NSURLSession use.

  • platform/network/cf/SocketStreamHandleImpl.h:
  • platform/network/cf/SocketStreamHandleImplCFNet.cpp:

(WebCore::Function<bool):
(WebCore::SocketStreamHandleImpl::setLegacyTLSEnabledCheck):
(WebCore::SocketStreamHandleImpl::createStreams):

Source/WebCore/PAL:

  • pal/spi/cf/CFNetworkSPI.h:

Source/WebKit:

  • NetworkProcess/NetworkProcessCreationParameters.cpp:

(WebKit::NetworkProcessCreationParameters::encode const):
(WebKit::NetworkProcessCreationParameters::decode):

  • NetworkProcess/NetworkProcessCreationParameters.h:
  • NetworkProcess/cocoa/NetworkProcessCocoa.mm:

(WebKit::NetworkProcess::platformInitializeNetworkProcessCocoa):

  • UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _allowAnyTLSCertificateForWebSocketTesting]):

  • UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
  • UIProcess/Cocoa/WebProcessPoolCocoa.mm:

(WebKit::WebProcessPool::platformInitializeNetworkProcess):

Source/WebKitLegacy/mac:

  • WebView/WebView.mm:

(-[WebView _commonInitializationWithFrameName:groupName:]):

Tools:

  • TestWebKitAPI/SourcesCocoa.txt:
  • TestWebKitAPI/TCPServer.cpp:

(sk_CRYPTO_BUFFER_num):
(sk_CRYPTO_BUFFER_value):
(TestWebKitAPI::deleter<CRYPTO_BUFFER>::operator()):
(TestWebKitAPI::TCPServer::TCPServer):
(TestWebKitAPI::TCPServer::listenForConnections):
(TestWebKitAPI::deleter<X509>::operator()): Deleted.
(TestWebKitAPI::deleter<uint8_t::operator()): Deleted.

  • TestWebKitAPI/TCPServer.h:
  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKitCocoa/TLSDeprecation.mm: Added.

(-[WebSocketDelegate waitForMessage]):
(-[WebSocketDelegate webView:runJavaScriptAlertPanelWithMessage:initiatedByFrame:completionHandler:]):
(TestWebKitAPI::TEST):

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

(-[TestNavigationDelegate webView:didReceiveAuthenticationChallenge:completionHandler:]):
(-[TestNavigationDelegate waitForDidFailProvisionalNavigation]):

6:31 PM Changeset in webkit [249683] by Fujii Hironori
  • 13 edits in trunk/Tools

[Win][MiniBrowser] WebKitLegacyBrowserWindow is leaked by circular references
https://bugs.webkit.org/show_bug.cgi?id=201600

Reviewed by Brent Fulgham.

There were some circular references between
WebKitLegacyBrowserWindow and its delegation classes. For
example, WebKitLegacyBrowserWindow has a reference of
WebDownloadDelegate, and WebDownloadDelegate shares the ref
counter with WebKitLegacyBrowserWindow.

WebNotificationObserver was leaked because it wasn't unregistered
from the default notification center by using
IWebNotificationCenter::removeObserver.

If a new legacy window was created by mouse right click a link,
WebView was released twice because
PrintWebUIDelegate::createWebViewWithRequest didn't AddRef the
WebView.

This change does:

  1. Make delegation classes have own ref-counter to avoid circular references
  2. Do removeObserver notification observers
  3. AddRef WebView in PrintWebUIDelegate::createWebViewWithRequest
  • MiniBrowser/win/AccessibilityDelegate.cpp:

(AccessibilityDelegate::AddRef):
(AccessibilityDelegate::Release):

  • MiniBrowser/win/AccessibilityDelegate.h: Added m_refCount.
  • MiniBrowser/win/MiniBrowserWebHost.cpp:

(MiniBrowserWebHost::QueryInterface):
(MiniBrowserWebHost::AddRef):
(MiniBrowserWebHost::Release):

  • MiniBrowser/win/MiniBrowserWebHost.h: Added m_refCount.
  • MiniBrowser/win/PrintWebUIDelegate.cpp:

(PrintWebUIDelegate::createWebViewWithRequest): Do AddRef for the returned IWebView.
(PrintWebUIDelegate::AddRef):
(PrintWebUIDelegate::Release):

  • MiniBrowser/win/PrintWebUIDelegate.h: Added m_refCount.
  • MiniBrowser/win/ResourceLoadDelegate.cpp:

(ResourceLoadDelegate::AddRef):
(ResourceLoadDelegate::Release):

  • MiniBrowser/win/ResourceLoadDelegate.h: Added m_refCount.
  • MiniBrowser/win/WebDownloadDelegate.cpp:

(WebDownloadDelegate::AddRef):
(WebDownloadDelegate::Release):

  • MiniBrowser/win/WebDownloadDelegate.h: Added m_refCount.
  • MiniBrowser/win/WebKitLegacyBrowserWindow.cpp:

(WebKitLegacyBrowserWindow::~WebKitLegacyBrowserWindow): Do removeObserver notification observers.
(WebKitLegacyBrowserWindow::init):
(WebKitLegacyBrowserWindow::setUIDelegate):
(WebKitLegacyBrowserWindow::setAccessibilityDelegate):
(WebKitLegacyBrowserWindow::setResourceLoadDelegate):
(WebKitLegacyBrowserWindow::setDownloadDelegate):
(WebKitLegacyBrowserWindow::AddRef): Deleted.
(WebKitLegacyBrowserWindow::Release): Deleted.
(WebKitLegacyBrowserWindow::setFrameLoadDelegate): Deleted.
(WebKitLegacyBrowserWindow::setFrameLoadDelegatePrivate): Deleted.

  • MiniBrowser/win/WebKitLegacyBrowserWindow.h:
6:28 PM Changeset in webkit [249682] by Fujii Hironori
  • 2 edits in trunk/Source/WebKitLegacy/win

[Win][WebKitLegacy] WebKitDLL.h(53,37): warning: 'gClassNameCount' has C-linkage specified, but returns user-defined type 'HashCountedSet<WTF::String> &' which is incompatible with C [-Wreturn-type-c-linkage]
https://bugs.webkit.org/show_bug.cgi?id=201601

Reviewed by Ross Kirsling.

Fixed a clang-cl warning. Names which aren't dllexport-ed don't
need to be in a extern "C" scope because they aren't exported to
other DLLs.

  • WebKitDLL.h: Moved some variables and gClassNameCount() out of extern "C" scope.
6:01 PM Changeset in webkit [249681] by timothy_horton@apple.com
  • 5 edits in trunk/Source

Clarify some macCatalyst feature flags
https://bugs.webkit.org/show_bug.cgi?id=201619
<rdar://problem/54615618>

Reviewed by Megan Gardner.

Source/WebKit:

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

(-[WKContentView setupInteraction]):
(-[WKContentView cleanupInteraction]):
(-[WKContentView _removeDefaultGestureRecognizers]):
(-[WKContentView _addDefaultGestureRecognizers]):
(-[WKContentView gestureRecognizer:shouldRecognizeSimultaneouslyWithGestureRecognizer:]):
(-[WKContentView _showShareSheet:inRect:completionHandler:]):
We prefer specific flags over platform checks.

Source/WTF:

  • wtf/Platform.h:

We prefer specific flags over platform checks.

5:45 PM Changeset in webkit [249680] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

Unreviewed test gardening, update test expectations for rdar://problem/51229282.

  • platform/mac-wk1/TestExpectations:
5:38 PM Changeset in webkit [249679] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

Unreviewed test gardening, update test expectations for rdar://problem/51766135.

  • platform/ios/TestExpectations:
5:32 PM Changeset in webkit [249678] by Ryan Haddad
  • 2 edits in trunk/LayoutTests
REGRESSION(r236540-r236595): ASSERTION FAILED: !frameView
!frameView->isPainting() on plugins/window-open.html

https://bugs.webkit.org/show_bug.cgi?id=199089

Unreviewed test gardening.

  • platform/mac-wk2/TestExpectations: Skip the test.
5:23 PM Changeset in webkit [249677] by rmorisset@apple.com
  • 2 edits in trunk/Source/JavaScriptCore

[Air] highOrderAdjacents in AbstractColoringAllocator::conservativeHeuristic should be some kind of array
https://bugs.webkit.org/show_bug.cgi?id=197305

Reviewed by Keith Miller.

Currently it is a HashSet, but it only ever holds at most registerCount() items. And linear search tends to be faster on such a small collection than hashing + searching in a HashSet.
Further benefits include avoiding the allocation of the HashSet, not actually adding the nodes adjacent to V (since there are no duplicates in the adjacency lists).

This patch also contains a trivial optimization: if the remaining number of nodes to consider + the number of highOrderAdjacents already seen is smaller than registerCount() we can return true directly.
Apart from that, the patch got some trivial cleanup of GraphColoringRegisterAllocation::allocateOnBank() (that for example was only logging the number of iterations for FP registers, and not the more interesting number for GP registers).

The time spent in the register allocator throughout JetStream2 on this MacBook Pro moves from 3767 / 3710 / 3785 ms to 3551 / 3454 / 3503 ms.
So about a 6% speedup for that phase, and between 1 and 1.5% speedup for FTL/OMG compilation overall.

No new tests as there is no intended change to the code being generated, and this was already tested by running testb3 + JetStream2.

  • b3/air/AirAllocateRegistersByGraphColoring.cpp:
4:51 PM Changeset in webkit [249676] by Alan Coon
  • 7 edits in branches/safari-608-branch/Source

Versioning.

4:35 PM Changeset in webkit [249675] by Chris Dumez
  • 16 edits
    1 move in trunk

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in http/tests/adClickAttribution/conversion-disabled-in-ephemeral-session.html
https://bugs.webkit.org/show_bug.cgi?id=201596

Reviewed by Alex Christensen.

Source/WebKit:

Allow WKFramePolicyListenerUseWithPolicies() C API to switch to a persistent non-default session.
This is important since WebKitTestRunner is not using the default session anymore and I need to
be able to switch it back to its normal session after going to an ephemeral one.

  • UIProcess/API/C/WKFramePolicyListener.cpp:

(useWithPolicies):

Tools:

Add new testRunner.setShouldSwapToEphemeralSessionOnNextNavigation() / testRunner.setShouldSwapToDefaultSessionOnNextNavigation()
test functions to support dynamic switching between ephemeral and default/persistent session in layout tests. The new methods
do the right thing for WebKit2 and truly switch session, unlike setPrivateBrowsingEnabled_DEPRECATED() which was merely making
the WebCore::Document think it changed session.

  • DumpRenderTree/TestRunner.cpp:

(setShouldSwapToEphemeralSessionOnNextNavigationCallback):
(setShouldSwapToDefaultSessionOnNextNavigationCallback):
(TestRunner::staticFunctions):
(TestRunner::willNavigate):

  • DumpRenderTree/TestRunner.h:

(TestRunner::setShouldSwapToEphemeralSessionOnNextNavigation):
(TestRunner::setShouldSwapToDefaultSessionOnNextNavigation):

  • DumpRenderTree/mac/FrameLoadDelegate.mm:

(-[FrameLoadDelegate webView:didStartProvisionalLoadForFrame:]):

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

(WTR::TestRunner::setShouldSwapToEphemeralSessionOnNextNavigation):
(WTR::TestRunner::setShouldSwapToDefaultSessionOnNextNavigation):

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

(WTR::TestController::decidePolicyForNavigationAction):

  • WebKitTestRunner/TestController.h:

(WTR::TestController::setShouldSwapToEphemeralSessionOnNextNavigation):
(WTR::TestController::setShouldSwapToDefaultSessionOnNextNavigation):

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

LayoutTests:

Update layout test to use testRunner.setShouldSwapToEphemeralSessionOnNextNavigation() &
testRunner.setShouldSwapToDefaultSessionOnNextNavigation() instead of the legacy
testRunner.setPrivateBrowsingEnabled_DEPRECATED(). The test needed some extra steps to
navigate whenever we switch session but it is a lot more realistic and is really switching
session on WebKit2. The testRunner.setPrivateBrowsingEnabled_DEPRECATED() was not truly
switching session, just making the document think it did.

  • http/tests/adClickAttribution/conversion-disabled-in-ephemeral-session.html:
4:23 PM Changeset in webkit [249674] by Alan Coon
  • 6 edits in branches/safari-608.2.11.1-branch/Source/WebKit

Cherry-pick r249649. rdar://problem/55198064

[macOS] Pid is sometimes invalid when creating sandbox extensions by pid.
https://bugs.webkit.org/show_bug.cgi?id=201543
<rdar://problem/54733465>

Reviewed by Brent Fulgham.

There is a race condition when starting a load of a local file, where the WebContent process has not finished
launching yet, and its pid is not available. When we try to create a sandbox extension by using the pid of the
WebContent process, it is not available in the cases where the WebContent process has just launched and has not
finished launching yet. This patch creates a new dummy Web page message, 'LoadRequestWaitingForPID', which will
be sent instead of a normal 'LoadRequest' message, and only when the WebContent process has not finished
launching. When the WebContent process has finished launching, and we are about to actually send the pending
messages, we can detect that a 'LoadRequestWaitingForPID' has been appended for sending, and replace it with a
normal 'LoadReqest' message where we have created the sandbox extension issue with a valid pid. The message
'LoadRequestWaitingForPID' is never intended to reach the WebContent process, it is just there to replace with
a normal 'LoadRequest' message with a new sandbox extension. In the implementation of the message handler on
the WebContent process side, we assert that the method is never called. This patch makes sure the ordering of
the Web page messages are the same, even when we modify the message.

  • UIProcess/AuxiliaryProcessProxy.cpp: (WebKit::AuxiliaryProcessProxy::didFinishLaunching):
  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle): (WebKit::WebPageProxy::loadRequestWithNavigationShared): (WebKit::WebPageProxy::loadFile):
  • WebProcess/WebPage/WebPage.cpp: (WebKit::WebPage::fileLoadRequest):
  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249649 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:23 PM Changeset in webkit [249673] by Alan Coon
  • 2 edits in branches/safari-608.2.11.1-branch/Source/WebKit

Cherry-pick r249454. rdar://problem/55198064

[macOS] Unable to open local file from favorites bar
https://bugs.webkit.org/show_bug.cgi?id=201444

Reviewed by Brent Fulgham.

The sandbox extension handle should be created providing the pid of the receiving process.

  • UIProcess/WebPageProxy.cpp: (WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle):

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249454 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:14 PM Changeset in webkit [249672] by Chris Dumez
  • 87 edits
    26 adds
    10 deletes in trunk/LayoutTests

Resync service workers WPT tests from upstream
https://bugs.webkit.org/show_bug.cgi?id=201589

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

Resync service workers WPT tests from upstream 820f0f86047e6e26401e.

  • web-platform-tests/service-workers/service-worker/Service-Worker-Allowed-header.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/Service-Worker-Allowed-header.https.html:
  • web-platform-tests/service-workers/service-worker/ServiceWorkerGlobalScope/postmessage.https.html:
  • web-platform-tests/service-workers/service-worker/ServiceWorkerGlobalScope/unregister.https.html:
  • web-platform-tests/service-workers/service-worker/ServiceWorkerGlobalScope/update.https.html:
  • web-platform-tests/service-workers/service-worker/about-blank-replacement.https.html:
  • web-platform-tests/service-workers/service-worker/activate-event-after-install-state-change.https.html:
  • web-platform-tests/service-workers/service-worker/activation-after-registration.https.html:
  • web-platform-tests/service-workers/service-worker/activation.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/activation.https.html:
  • web-platform-tests/service-workers/service-worker/appcache-ordering-main.https.html:
  • web-platform-tests/service-workers/service-worker/claim-fetch-with-appcache.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/claim-fetch-with-appcache.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/claim-fetch.https.html:
  • web-platform-tests/service-workers/service-worker/claim-not-using-registration.https.html:
  • web-platform-tests/service-workers/service-worker/claim-shared-worker-fetch.https.html:
  • web-platform-tests/service-workers/service-worker/claim-using-registration.https.html:
  • web-platform-tests/service-workers/service-worker/claim-worker-fetch.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/claim-worker-fetch.https.html:
  • web-platform-tests/service-workers/service-worker/client-id.https.html:
  • web-platform-tests/service-workers/service-worker/clients-get-client-types.https.html:
  • web-platform-tests/service-workers/service-worker/clients-get-resultingClientId.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/clients-get-resultingClientId.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/clients-get.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/clients-get.https.html:
  • web-platform-tests/service-workers/service-worker/clients-matchall-client-types.https.html:
  • web-platform-tests/service-workers/service-worker/clients-matchall-exact-controller.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/clients-matchall-exact-controller.https.html:
  • web-platform-tests/service-workers/service-worker/clients-matchall-include-uncontrolled.https.html:
  • web-platform-tests/service-workers/service-worker/clients-matchall-order.https.html:
  • web-platform-tests/service-workers/service-worker/clients-matchall.https.html:
  • web-platform-tests/service-workers/service-worker/controller-on-disconnect.https.html:
  • web-platform-tests/service-workers/service-worker/controller-on-load.https.html:
  • web-platform-tests/service-workers/service-worker/controller-on-reload.https.html:
  • web-platform-tests/service-workers/service-worker/data-transfer-files.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/data-transfer-files.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/detached-context.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/detached-context.https.html:
  • web-platform-tests/service-workers/service-worker/extendable-event-async-waituntil.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/extendable-event-async-waituntil.https.html:
  • web-platform-tests/service-workers/service-worker/extendable-event-waituntil.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-audio-tainting.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/fetch-audio-tainting.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/fetch-canvas-tainting-double-write.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/fetch-canvas-tainting-double-write.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/fetch-canvas-tainting-video-cache.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-canvas-tainting-video.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-cors-xhr.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-cors-xhr.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-csp.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-after-navigation-within-page.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-async-respond-with.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-event-async-respond-with.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-network-error.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-referrer-policy.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-event-referrer-policy.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-respond-with-argument.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-respond-with-body-loaded-in-chunk.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-respond-with-stops-propagation.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event-throws-after-respond-with.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-event.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-event.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-frame-resource.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-header-visibility.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-css-base-url.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-request-css-base-url.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-css-images.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-html-imports.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/fetch-request-html-imports.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-no-freshness-headers.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-redirect.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-request-xhr-sync.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-response-xhr.https.html:
  • web-platform-tests/service-workers/service-worker/fetch-waits-for-activate.https.html:
  • web-platform-tests/service-workers/service-worker/getregistration.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/getregistration.https.html:
  • web-platform-tests/service-workers/service-worker/import-module-scripts.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/import-module-scripts.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/import-scripts-mime-types.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/import-scripts-mime-types.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/import-scripts-resource-map.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/import-scripts-resource-map.https.html:
  • web-platform-tests/service-workers/service-worker/interfaces-window.https.html:
  • web-platform-tests/service-workers/service-worker/invalid-blobtype.https.html:
  • web-platform-tests/service-workers/service-worker/invalid-header.https.html:
  • web-platform-tests/service-workers/service-worker/iso-latin1-header.https.html:
  • web-platform-tests/service-workers/service-worker/multiple-update.https.html:
  • web-platform-tests/service-workers/service-worker/navigation-redirect.https.html:
  • web-platform-tests/service-workers/service-worker/onactivate-script-error.https.html:
  • web-platform-tests/service-workers/service-worker/performance-timeline.https.html:
  • web-platform-tests/service-workers/service-worker/postmessage-from-waiting-serviceworker.https.html:
  • web-platform-tests/service-workers/service-worker/postmessage-to-client-message-queue.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/postmessage-to-client-message-queue.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/postmessage-to-client.https.html:
  • web-platform-tests/service-workers/service-worker/postmessage.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/postmessage.https.html:
  • web-platform-tests/service-workers/service-worker/ready.https.html:
  • web-platform-tests/service-workers/service-worker/redirected-response.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/redirected-response.https.html:
  • web-platform-tests/service-workers/service-worker/referer.https.html:
  • web-platform-tests/service-workers/service-worker/register-default-scope.https.html:
  • web-platform-tests/service-workers/service-worker/register-wait-forever-in-install-worker.https.html:
  • web-platform-tests/service-workers/service-worker/registration-end-to-end.https.html:
  • web-platform-tests/service-workers/service-worker/registration-events.https.html:
  • web-platform-tests/service-workers/service-worker/registration-mime-types.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/registration-mime-types.https.html:
  • web-platform-tests/service-workers/service-worker/registration-schedule-job.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/registration-schedule-job.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/registration-scope.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/registration-security-error.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/registration-service-worker-attributes.https.html:
  • web-platform-tests/service-workers/service-worker/registration-updateviacache.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/registration-updateviacache.https.html:
  • web-platform-tests/service-workers/service-worker/resource-timing.https.html: Renamed from LayoutTests/imported/w3c/web-platform-tests/service-workers/service-worker/resource-timing.sub.https.html.
  • web-platform-tests/service-workers/service-worker/resource-timing.sub.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/resources/about-blank-replacement-ping-frame.py:
  • web-platform-tests/service-workers/service-worker/resources/appcache-ordering.manifest:
  • web-platform-tests/service-workers/service-worker/resources/claim-worker.js:
  • web-platform-tests/service-workers/service-worker/resources/classic-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/clients-get-cross-origin-frame.html:
  • web-platform-tests/service-workers/service-worker/resources/clients-get-resultingClientId-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/clients-get-worker.js:

(self.onmessage):

  • web-platform-tests/service-workers/service-worker/resources/enable-client-message-queue.html: Removed.
  • web-platform-tests/service-workers/service-worker/resources/extendable-event-async-waituntil.js:

(switch):
(event.request.url.indexOf):
(else.event.request.url.indexOf):

  • web-platform-tests/service-workers/service-worker/resources/fetch-canvas-tainting-double-write-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/fetch-cors-xhr-iframe.html:
  • web-platform-tests/service-workers/service-worker/resources/fetch-event-async-respond-with-worker.js:

(setTimeout):

  • web-platform-tests/service-workers/service-worker/resources/fetch-event-network-fallback-worker.js:
  • web-platform-tests/service-workers/service-worker/resources/fetch-event-test-worker.js:
  • web-platform-tests/service-workers/service-worker/resources/fetch-request-css-base-url-iframe.html:
  • web-platform-tests/service-workers/service-worker/resources/fetch-request-css-base-url-worker.js:

(self.onmessage):
(url.indexOf):
(else.url.indexOf):

  • web-platform-tests/service-workers/service-worker/resources/get-resultingClientId-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/import-relative.xsl: Removed.
  • web-platform-tests/service-workers/service-worker/resources/import-scripts-diff-resource-map-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/import-scripts-get.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/import-scripts-mime-types-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/imported-classic-script.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/imported-module-script.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/message-vs-microtask.html: Removed.
  • web-platform-tests/service-workers/service-worker/resources/module-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/navigation-redirect-other-origin.html:
  • web-platform-tests/service-workers/service-worker/resources/navigation-redirect-out-scope.py:
  • web-platform-tests/service-workers/service-worker/resources/navigation-redirect-scope1.py:
  • web-platform-tests/service-workers/service-worker/resources/navigation-redirect-scope2.py:
  • web-platform-tests/service-workers/service-worker/resources/nested-iframe-parent.html: Removed.
  • web-platform-tests/service-workers/service-worker/resources/performance-timeline-worker.js:

(promise_test.):
(promise_test):

  • web-platform-tests/service-workers/service-worker/resources/postmessage-echo-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/redirect-worker.js:

(event.waitUntil.Promise.all.waitUntilPromiseList.then):
(then):
(waitUntilPromise.caches.open.cacheName.then):
(event.respondWith.waitUntilPromise.then):

  • web-platform-tests/service-workers/service-worker/resources/registration-tests-mime-types.js:

(registration_tests_mime_types):

  • web-platform-tests/service-workers/service-worker/resources/registration-tests-scope.js:

(registration_tests_scope):

  • web-platform-tests/service-workers/service-worker/resources/registration-tests-security-error.js:

(registration_tests_security_error):

  • web-platform-tests/service-workers/service-worker/resources/resource-timing-iframe.sub.html:
  • web-platform-tests/service-workers/service-worker/resources/service-worker-csp-worker.py:
  • web-platform-tests/service-workers/service-worker/resources/service-worker-header.py:

(main):

  • web-platform-tests/service-workers/service-worker/resources/skip-waiting-installed-worker.js:

(event.waitUntil.self.skipWaiting.then):

  • web-platform-tests/service-workers/service-worker/resources/stalling-service-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/test-helpers.sub.js:

(wait_for_update):
(async.wait_for_activation_on_dummy_scope):

  • web-platform-tests/service-workers/service-worker/resources/test-request-headers-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/test-request-headers-worker.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/trickle.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-during-installation-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-during-installation-worker.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-fetch-worker.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-missing-import-scripts-imported-worker.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-missing-import-scripts-main-worker.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/update-registration-with-type.py: Removed.
  • web-platform-tests/service-workers/service-worker/resources/w3c-import.log:
  • web-platform-tests/service-workers/service-worker/resources/worker-interception-iframe.https.html:
  • web-platform-tests/service-workers/service-worker/resources/worker-load-interceptor.js:

(self.onfetch):

  • web-platform-tests/service-workers/service-worker/resources/worker-testharness.js:

(create_temporary_cache):

  • web-platform-tests/service-workers/service-worker/resources/xhr-iframe.html: Removed.
  • web-platform-tests/service-workers/service-worker/resources/xhr-response-url-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/xsl-base-url-iframe.xml: Removed.
  • web-platform-tests/service-workers/service-worker/resources/xsl-base-url-worker.js: Removed.
  • web-platform-tests/service-workers/service-worker/resources/xslt-pass.xsl: Removed.
  • web-platform-tests/service-workers/service-worker/service-worker-csp-connect.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/service-worker-csp-default.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/service-worker-csp-script.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/service-worker-header.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/service-worker-header.https.html:
  • web-platform-tests/service-workers/service-worker/serviceworker-message-event-historical.https.html:
  • web-platform-tests/service-workers/service-worker/shared-worker-controlled.https.html:
  • web-platform-tests/service-workers/service-worker/skip-waiting-installed.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/skip-waiting-installed.https.html:
  • web-platform-tests/service-workers/service-worker/skip-waiting-using-registration.https.html:
  • web-platform-tests/service-workers/service-worker/skip-waiting-without-using-registration.https.html:
  • web-platform-tests/service-workers/service-worker/skip-waiting.https.html:
  • web-platform-tests/service-workers/service-worker/state.https.html:
  • web-platform-tests/service-workers/service-worker/synced-state.https.html:
  • web-platform-tests/service-workers/service-worker/uncontrolled-page.https.html:
  • web-platform-tests/service-workers/service-worker/unregister-then-register-new-script.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/unregister-then-register-new-script.https.html:
  • web-platform-tests/service-workers/service-worker/unregister-then-register.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/unregister-then-register.https.html:
  • web-platform-tests/service-workers/service-worker/update-after-navigation-redirect.https.html:
  • web-platform-tests/service-workers/service-worker/update-after-oneday.https.html:
  • web-platform-tests/service-workers/service-worker/update-bytecheck.https.html:
  • web-platform-tests/service-workers/service-worker/update-missing-import-scripts.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/update-missing-import-scripts.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/update-no-cache-request-headers.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/update-no-cache-request-headers.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/update-not-allowed.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/update-not-allowed.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/update-on-navigation.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/update-on-navigation.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/update-recovery.https.html:
  • web-platform-tests/service-workers/service-worker/update-registration-with-type.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/update-registration-with-type.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/w3c-import.log:
  • web-platform-tests/service-workers/service-worker/websocket.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/websocket.https.html:
  • web-platform-tests/service-workers/service-worker/windowclient-navigate.https.html:
  • web-platform-tests/service-workers/service-worker/worker-client-id.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/worker-interception.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/worker-interception.https.html:
  • web-platform-tests/service-workers/service-worker/xhr-response-url.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/xhr-response-url.https.html: Removed.
  • web-platform-tests/service-workers/service-worker/xsl-base-url.https-expected.txt: Removed.
  • web-platform-tests/service-workers/service-worker/xsl-base-url.https.html: Removed.

LayoutTests:

3:45 PM Changeset in webkit [249671] by achristensen@apple.com
  • 10 edits in trunk/Source/WebKit

Deprecate unused SPI to preconnect to a server
https://bugs.webkit.org/show_bug.cgi?id=201558

Reviewed by Youenn Fablet.

  • NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::preconnectTo): Deleted.

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

(WKContextPreconnectToServer):

  • UIProcess/API/C/WKContextPrivate.h:
  • UIProcess/API/Cocoa/WKProcessPool.mm:

(-[WKProcessPool _preconnectToServer:]):

  • UIProcess/API/Cocoa/WKProcessPoolPrivate.h:
  • UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::preconnectToServer): Deleted.

  • UIProcess/WebProcessPool.h:
3:38 PM Changeset in webkit [249670] by msaboff@apple.com
  • 12 edits in trunk/Source/bmalloc

Revert to pre-r243144 scavenging behavior for macOS
https://bugs.webkit.org/show_bug.cgi?id=201555

Reviewed by Saam Barati.

The change in r243144 regressed internal power metrics for some Mac models.

  • bmalloc/Heap.cpp:

(bmalloc::Heap::scavenge):
(bmalloc::Heap::scavengeToHighWatermark):
(bmalloc::Heap::allocateSmallChunk):
(bmalloc::Heap::allocateSmallPage):
(bmalloc::Heap::allocateLarge):

  • bmalloc/Heap.h:
  • bmalloc/IsoDirectory.h:
  • bmalloc/IsoDirectoryInlines.h:

(bmalloc::passedNumPages>::takeFirstEligible):
(bmalloc::passedNumPages>::scavenge):
(bmalloc::passedNumPages>::scavengeToHighWatermark):

  • bmalloc/IsoHeapImpl.h:
  • bmalloc/IsoHeapImplInlines.h:

(bmalloc::IsoHeapImpl<Config>::scavengeToHighWatermark):

  • bmalloc/LargeMap.cpp:

(bmalloc::LargeMap::add):

  • bmalloc/LargeRange.h:

(bmalloc::LargeRange::LargeRange):
(bmalloc::merge):

  • bmalloc/Scavenger.cpp:

(bmalloc::Scavenger::Scavenger):
(bmalloc::Scavenger::timeSinceLastPartialScavenge):
(bmalloc::Scavenger::scavenge):
(bmalloc::Scavenger::partialScavenge):
(bmalloc::Scavenger::threadRunLoop):

  • bmalloc/Scavenger.h:
  • bmalloc/SmallPage.h:
3:17 PM Changeset in webkit [249669] by pvollan@apple.com
  • 2 edits in trunk/Source/WebKit

Unreviewed Windows build fix.

  • WebProcess/WebPage/WebPage.h:
2:38 PM Changeset in webkit [249668] by ysuzuki@apple.com
  • 6 edits
    1 add in trunk

[JSC] Use metadata table to iterate specific bytecode metadata instead of propertyAccessInstructions vector
https://bugs.webkit.org/show_bug.cgi?id=201613

Reviewed by Mark Lam.

We do not need to maintain propertyAccessInstructions vector to access metadata tied to a specific bytecode opcode
since we have MetadataTable::forEach<Op> feature. This removes propertyAccessInstructions entirely, and fixes the
issue that op_create_promise missed propertyAccessInstructions registration (a name "propertyAccessInstructions" is
misleading, it is like "instructions-requires-llint-finalize").

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::propagateTransitions):
(JSC::CodeBlock::finalizeLLIntInlineCaches):

  • bytecode/UnlinkedCodeBlock.cpp:

(JSC::UnlinkedCodeBlock::applyModification):
(JSC::UnlinkedCodeBlock::shrinkToFit):

  • bytecode/UnlinkedCodeBlock.h:

(JSC::UnlinkedCodeBlock::addPropertyAccessInstruction): Deleted.
(JSC::UnlinkedCodeBlock::numberOfPropertyAccessInstructions const): Deleted.
(JSC::UnlinkedCodeBlock::propertyAccessInstructions const): Deleted.

  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitResolveScope):
(JSC::BytecodeGenerator::emitGetFromScope):
(JSC::BytecodeGenerator::emitPutToScope):
(JSC::BytecodeGenerator::emitGetById):
(JSC::BytecodeGenerator::emitDirectGetById):
(JSC::BytecodeGenerator::emitPutById):
(JSC::BytecodeGenerator::emitDirectPutById):
(JSC::BytecodeGenerator::emitCreateThis):
(JSC::BytecodeGenerator::emitToThis):

  • runtime/CachedTypes.cpp:

(JSC::CachedCodeBlock<CodeBlockType>::decode const):
(JSC::CachedCodeBlock<CodeBlockType>::encode):

2:20 PM Changeset in webkit [249667] by aakash_jain@apple.com
  • 2 edits in trunk/Tools

EWS testtube emoji doesn't exist on macOS Mojave
https://bugs.webkit.org/show_bug.cgi?id=201532

Reviewed by Simon Fraser.

  • BuildSlaveSupport/ews-app/ews/views/statusbubble.py:

(StatusBubble): Change the emoji to microscope emoji.

2:15 PM Changeset in webkit [249666] by jonlee@apple.com
  • 1 edit
    11 adds in trunk/Websites/webkit.org

Add Babylon demos
https://bugs.webkit.org/show_bug.cgi?id=201611

Rubber-stamped by Justin Fan.

  • demos/webgpu/babylon/LICENSE: Added.
  • demos/webgpu/babylon/babylonWebGpu.max.js: Added.
  • demos/webgpu/babylon/logo.svg: Added.
  • demos/webgpu/babylon/oneHelmetWebGPU.html: Added.
  • demos/webgpu/babylon/oneHelmetWebGPUGLSLShaders.js: Added.
  • demos/webgpu/babylon/oneHelmetWebGPUWSLShaders.js: Added.
  • demos/webgpu/babylon/oneSphereWebGPU.html: Added.
  • demos/webgpu/babylon/oneSphereWebGPUGLSLShaders.js: Added.
  • demos/webgpu/babylon/oneSphereWebGPUWSLShaders.js: Added.
  • demos/webgpu/babylon/vertexShaderGLSL1.glsl: Added.
2:07 PM Changeset in webkit [249665] by jonlee@apple.com
  • 1 edit
    3 adds in trunk/Websites/webkit.org

Add boids time test
https://bugs.webkit.org/show_bug.cgi?id=201612

Rubber-stamped by Justin Fan.

  • demos/webgpu/compute-boids-compile.html: Added.
  • demos/webgpu/scripts/compute-boids-compile.js: Added.
  • demos/webgpu/scripts/compute-boids-utils.js: Added.
1:54 PM Changeset in webkit [249664] by sbarati@apple.com
  • 2 edits in trunk/Source/WebCore

Unreviewed follow up to r249630. We need padding for ADDRESS32 CPUs to allow replaceWith to work on the intended types.

  • Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h:
1:36 PM Changeset in webkit [249663] by aakash_jain@apple.com
  • 2 edits in trunk/Tools

[EWS] API testers can go in an infinite RETRY loop when there is build failure in trunk
https://bugs.webkit.org/show_bug.cgi?id=201607

Reviewed by Jonathan Bedard.

  • BuildSlaveSupport/ews-build/steps.py:

(AnalyzeAPITestsResults.analyzeResults): Do not RETRY when clean_tree_results are missing.
(AnalyzeAPITestsResults.analyzeResults.getAPITestFailures): Gracefully handle None result.

1:32 PM Changeset in webkit [249662] by Adrian Perez de Castro
  • 1 copy in releases/WPE WebKit/webkit-2.25.91

WPE WebKit 2.25.91

1:32 PM Changeset in webkit [249661] by keith_miller@apple.com
  • 11 edits
    10 adds in trunk

OSR entry into wasm misses some contexts
https://bugs.webkit.org/show_bug.cgi?id=201569

Reviewed by Yusuke Suzuki.

JSTests:

Add a new harness and wast and the generated wasm file for
testing. The idea long term is to make it easy to test by creating
a C file and converting it to a wast then modify that to produce a
test.

  • wasm.yaml:
  • wasm/wast-tests/harness.js: Added.

(async.runWasmFile):

  • wasm/wast-tests/osr-entry-inner-loop-branch-above-no-consts.wasm: Added.
  • wasm/wast-tests/osr-entry-inner-loop-branch-above-no-consts.wast: Added.
  • wasm/wast-tests/osr-entry-inner-loop-branch-above.wasm: Added.
  • wasm/wast-tests/osr-entry-inner-loop-branch-above.wast: Added.
  • wasm/wast-tests/osr-entry-inner-loop.wasm: Added.
  • wasm/wast-tests/osr-entry-inner-loop.wast: Added.
  • wasm/wast-tests/osr-entry-multiple-enclosed-contexts.wasm: Added.
  • wasm/wast-tests/osr-entry-multiple-enclosed-contexts.wast: Added.

Source/JavaScriptCore:

This patch fixes an issue where we could fail to capture some of
our contexts when OSR entering into wasm code. Before we would
only capture the state of the block immediately surrounding the
entrance loop block header. We actually need to capture all
enclosed stacks.

Additionally, we don't need to use variables for all the captured
values. We can use a Phi and insert an upsilon just below the
captured value.

  • interpreter/CallFrame.h:
  • jsc.cpp:

(GlobalObject::finishCreation):
(functionCallerIsOMGCompiled):

  • wasm/WasmAirIRGenerator.cpp:

(JSC::Wasm::AirIRGenerator::AirIRGenerator):
(JSC::Wasm::AirIRGenerator::emitEntryTierUpCheck):
(JSC::Wasm::AirIRGenerator::emitLoopTierUpCheck):
(JSC::Wasm::AirIRGenerator::addLoop):

  • wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::B3IRGenerator::createStack):
(JSC::Wasm::B3IRGenerator::B3IRGenerator):
(JSC::Wasm::B3IRGenerator::addConstant):
(JSC::Wasm::B3IRGenerator::emitEntryTierUpCheck):
(JSC::Wasm::B3IRGenerator::emitLoopTierUpCheck):
(JSC::Wasm::B3IRGenerator::addLoop):
(JSC::Wasm::B3IRGenerator::addEndToUnreachable):
(JSC::Wasm::dumpExpressionStack):
(JSC::Wasm::B3IRGenerator::dump):
(JSC::Wasm::B3IRGenerator::Stack::Stack): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::append): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::takeLast): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::last): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::size const): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::isEmpty const): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::convertToExpressionList): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::at const): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::variableAt const): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::shrink): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::swap): Deleted.
(JSC::Wasm::B3IRGenerator::Stack::dump const): Deleted.

  • wasm/WasmFunctionParser.h:

(JSC::Wasm::FunctionParser::controlStack):

Tools:

Add new test harness mode for tests created from wast files.

  • Scripts/run-jsc-stress-tests:
1:32 PM Changeset in webkit [249660] by Adrian Perez de Castro
  • 4 edits in releases/WebKitGTK/webkit-2.26

Unreviewed. Update OptionsWPE.cmake and NEWS for the 2.25.91 release

.:

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

Source/WebKit:

  • wpe/NEWS: Add release notes for 2.25.91.
1:29 PM Changeset in webkit [249659] by Alan Coon
  • 7 edits in trunk/Source

Versioning.

1:23 PM Changeset in webkit [249658] by Alan Coon
  • 7 edits in branches/safari-608.2.11.1-branch/Source

Versioning.

12:30 PM Changeset in webkit [249657] by jonlee@apple.com
  • 7 edits
    2 copies
    2 moves
    2 adds in trunk

Upload triangles tests for 3D suite in MotionMark
https://bugs.webkit.org/show_bug.cgi?id=201606

Reviewed by Simon Fraser.

PerformanceTests:

  • MotionMark/resources/debug-runner/motionmark.css:
  • MotionMark/resources/debug-runner/tests.js:
  • MotionMark/tests/3d/resources/webgl.js:
  • MotionMark/tests/3d/resources/webgpu.js: Added.
  • MotionMark/tests/3d/triangles-webgl.html: Renamed from Websites/browserbench.org/MotionMark1.1/tests/3d/webgl.html.
  • MotionMark/tests/3d/triangles-webgpu.html: Copied from PerformanceTests/MotionMark/tests/3d/webgl.html.

Websites/browserbench.org:

  • MotionMark1.1/resources/debug-runner/tests.js:
  • MotionMark1.1/tests/3d/resources/webgl.js:
  • MotionMark1.1/tests/3d/resources/webgpu.js: Added.
  • MotionMark1.1/tests/3d/triangles-webgl.html: Copied from PerformanceTests/MotionMark/tests/3d/webgl.html.
  • MotionMark1.1/tests/3d/triangles-webgpu.html: Renamed from PerformanceTests/MotionMark/tests/3d/webgl.html.
12:16 PM Changeset in webkit [249656] by jiewen_tan@apple.com
  • 2 edits in trunk/LayoutTests

Layout Test http/wpt/webauthn/public-key-credential-create-failure-u2f-silent.https.html is flaky
https://bugs.webkit.org/show_bug.cgi?id=198810

Reviewed by Dewei Zhu.

  • http/wpt/webauthn/public-key-credential-create-failure-u2f-silent.https.html:

Removed the timeout value given these tests should report the exception immediately.

12:02 PM Changeset in webkit [249655] by Joseph Pecoraro
  • 7 edits in trunk/Source/WebInspectorUI

Web Inspector: Dark Mode: Overview Timeline View has jarring white box-shadows in dark mode
https://bugs.webkit.org/show_bug.cgi?id=201442
<rdar://problem/54999794>

Reviewed by Devin Rousso.

  • UserInterface/Views/DataGrid.css:

(.data-grid:not(.variable-height-rows) table.data):
(.data-grid:not(.variable-height-rows) table.data.even-first-zebra-stripe):
(.data-grid.variable-height-rows.even-first-zebra-stripe table.data tr:nth-child(odd)):
(.data-grid.variable-height-rows.even-first-zebra-stripe table.data tr:nth-child(even)):
(.data-grid:not(.variable-height-rows) table.data.odd-first-zebra-stripe): Deleted.
(.data-grid.variable-height-rows.odd-first-zebra-stripe table.data tr:nth-child(odd)): Deleted.
(.data-grid.variable-height-rows.odd-first-zebra-stripe table.data tr:nth-child(even)): Deleted.

  • UserInterface/Views/DataGrid.js:

(WI.DataGrid.prototype.updateVisibleRows):

  • UserInterface/Views/Table.css:

(.table > .data-container > .data-list):
(.table > .data-container > .data-list.even-first-zebra-stripe):
(.table > .data-container > .data-list.odd-first-zebra-stripe): Deleted.

  • UserInterface/Views/Table.js:

In the gradients "even" was mistakenly first. Fix these so that "odd" is first and the
unusual case would be having "even" first. This allows us to line up with nth-of-type(even).

  • UserInterface/Views/Main.css:

(:root):
(@media (prefers-color-scheme: dark)):
Flip even and odd so we maintain the same styles for gradients.

  • UserInterface/Views/OverviewTimelineView.css:

(.timeline-view.overview > .data-grid tr):
(.timeline-view.overview > .data-grid tr:nth-child(even)):
(.timeline-view.overview > .data-grid tr.parent:not(.expanded) td.graph-column .timeline-record-bar:not(.has-inactive-segment) > .segment):
(.timeline-view.overview > .data-grid tr.parent:not(.expanded).selected td.graph-column .timeline-record-bar:not(.has-inactive-segment) > .segment):
(.timeline-view.overview > .data-grid tr.parent:not(.expanded) td.graph-column .timeline-record-bar.has-inactive-segment > .segment:not(.inactive)):
(.timeline-view.overview > .data-grid tr.parent:not(.expanded).selected td.graph-column .timeline-record-bar.has-inactive-segment > .segment:not(.inactive)):
(@media (prefers-color-scheme: dark)):
(WI.Table.prototype._updateVisibleRows):

  • UserInterface/Views/Variables.css:
12:01 PM Changeset in webkit [249654] by Alan Bujtas
  • 9 edits in trunk/Source/WebCore

[LFC][TFC] Introduce cell spacing.
https://bugs.webkit.org/show_bug.cgi?id=201605
<rdar://problem/55184009>

Reviewed by Antti Koivisto.

This patch adds support for horizontal and vertical cell spacing (border-spacing). Now LFC matches table geometry for simple table content with multiple columns and rows.

  • layout/FormattingContextGeometry.cpp:

(WebCore::Layout::FormattingContext::Geometry::contentHeightForFormattingContextRoot const): Now we can use the generic, check the inflow content logic for table height.

  • layout/LayoutUnits.h:

(WebCore::Layout::HorizontalEdges::width const):
(WebCore::Layout::VerticalEdges::height const):

  • layout/Verification.cpp:

(WebCore::Layout::verifyAndOutputSubtree):

  • layout/tableformatting/TableFormattingContext.cpp:

(WebCore::Layout::TableFormattingContext::layout):
(WebCore::Layout::TableFormattingContext::layoutTableCellBox):
(WebCore::Layout::TableFormattingContext::positionTableCells):
(WebCore::Layout::TableFormattingContext::setComputedGeometryForRows):
(WebCore::Layout::TableFormattingContext::setComputedGeometryForSections):
(WebCore::Layout::TableFormattingContext::computedIntrinsicWidthConstraints):
(WebCore::Layout::TableFormattingContext::ensureTableGrid):
(WebCore::Layout::TableFormattingContext::computePreferredWidthForColumns):
(WebCore::Layout::TableFormattingContext::computedTableWidth):
(WebCore::Layout::TableFormattingContext::useAsContentLogicalWidth):

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

(WebCore::Layout::TableFormattingContext::Geometry::tableCellHeightAndMargin const):

  • layout/tableformatting/TableGrid.cpp:

(WebCore::Layout::TableGrid::appendCell):
(WebCore::Layout::TableGrid::widthConstraints const):
(WebCore::Layout::TableGrid::ColumnsContext::useAsLogicalWidth): Deleted.

  • layout/tableformatting/TableGrid.h:

(WebCore::Layout::TableGrid::setHorizontalSpacing):
(WebCore::Layout::TableGrid::horizontalSpacing const):
(WebCore::Layout::TableGrid::setVerticalSpacing):
(WebCore::Layout::TableGrid::verticalSpacing const):
(WebCore::Layout::TableGrid::ColumnsContext::logicalWidth const):

11:56 AM Changeset in webkit [249653] by dbates@webkit.org
  • 2 edits in trunk/Tools

Remove all selection view animations before dumping results
https://bugs.webkit.org/show_bug.cgi?id=199241

Reviewed by Wenson Hsieh.

On iOS the selection UI uses a few animations, including fading in the grabbers. We have many
historical tests that were written for Mac and expect Mac's lack of animations that we run on
iOS and are flaky. Let's try to remove all selection view animations before dumping results and
see if this makes tests on iOS less flaky. A future enhancement to this is to expose a uiController
API for a test to remove selection animations at any time during the run.

  • WebKitTestRunner/ios/TestControllerIOS.mm:

(WTR::TestController::notifyDone):

11:29 AM Changeset in webkit [249652] by Jonathan Bedard
  • 4 edits in trunk/Tools

run-webkit-tests: Report results archive to results.webkit.org
https://bugs.webkit.org/show_bug.cgi?id=201321

Reviewed by Aakash Jain.

  • Scripts/webkitpy/layout_tests/controllers/manager.py:

(Manager.run): After all tests are finish, upload the results archive for each
configuration.

  • Scripts/webkitpy/results/upload.py:

(Upload):
(Upload.init): Automatically define timestamp.
(Upload.upload_archive): Upload an archive associated with the test run.

  • Scripts/webkitpy/results/upload_unittest.py:

(UploadTest.test_buildbot):
(UploadTest):
(UploadTest.test_archive_upload):

10:52 AM Changeset in webkit [249651] by Ryan Haddad
  • 9 edits in trunk

Unreviewed, rolling out r249574.

Caused fast/mediastream/media-stream-track-source-failure.html
to become flaky.

Reverted changeset:

"Remove MediaStreamPrivate::scheduleDeferredTask"
https://bugs.webkit.org/show_bug.cgi?id=200975
https://trac.webkit.org/changeset/249574

10:52 AM Changeset in webkit [249650] by ysuzuki@apple.com
  • 8 edits
    1 add in trunk

[JSC] Promise resolve/reject functions should be created more efficiently
https://bugs.webkit.org/show_bug.cgi?id=201488

Reviewed by Mark Lam.

JSTests:

  • microbenchmarks/promise-creation-many.js: Added.

(executor):

Source/JavaScriptCore:

While r246553 fixed an important issue, it makes anonymous-builtin-function creation costly since it enforces FunctionRareData allocations.
Unfortunately, anonymous-builtin-function function can be created frequently since this type of function is used
for resolve and reject arguments of Promise's executor (e.g. new Promise((resolve, reject) => ...)'s resolve and reject).
Since we are now always creating FunctionRareData for these functions, this additional allocation makes promise creation slower.

In this patch, we use isAnonymousBuiltinFunction information for hasReifiedName correctly. And we propagate isAnonymousBuiltinFunction information
to FunctionRareData to initialize m_hasReifiedName correctly. Then we can avoid unnecessary FunctionRareData allocation, which makes
anonymous-builtin-function creation faster.

We can ensure that this patch does not revert r246553's fix by running JSTests/stress/builtin-private-function-name.js test.
The simple microbenchmark shows 1.7x improvement.

ToT Patched

promise-creation-many 45.6701+-0.1488 26.8663+-1.8336 definitely 1.6999x faster

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileNewFunctionCommon):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileNewFunction):

  • runtime/FunctionRareData.cpp:

(JSC::FunctionRareData::create):
(JSC::FunctionRareData::FunctionRareData):

  • runtime/FunctionRareData.h:
  • runtime/JSFunction.cpp:

(JSC::JSFunction::finishCreation):
(JSC::JSFunction::allocateRareData):
(JSC::JSFunction::allocateAndInitializeRareData):

  • runtime/JSFunctionInlines.h:

(JSC::JSFunction::hasReifiedName const):

10:44 AM Changeset in webkit [249649] by pvollan@apple.com
  • 6 edits in trunk/Source/WebKit

[macOS] Pid is sometimes invalid when creating sandbox extensions by pid.
https://bugs.webkit.org/show_bug.cgi?id=201543
<rdar://problem/54733465>

Reviewed by Brent Fulgham.

There is a race condition when starting a load of a local file, where the WebContent process has not finished
launching yet, and its pid is not available. When we try to create a sandbox extension by using the pid of the
WebContent process, it is not available in the cases where the WebContent process has just launched and has not
finished launching yet. This patch creates a new dummy Web page message, 'LoadRequestWaitingForPID', which will
be sent instead of a normal 'LoadRequest' message, and only when the WebContent process has not finished
launching. When the WebContent process has finished launching, and we are about to actually send the pending
messages, we can detect that a 'LoadRequestWaitingForPID' has been appended for sending, and replace it with a
normal 'LoadReqest' message where we have created the sandbox extension issue with a valid pid. The message
'LoadRequestWaitingForPID' is never intended to reach the WebContent process, it is just there to replace with
a normal 'LoadRequest' message with a new sandbox extension. In the implementation of the message handler on
the WebContent process side, we assert that the method is never called. This patch makes sure the ordering of
the Web page messages are the same, even when we modify the message.

  • UIProcess/AuxiliaryProcessProxy.cpp:

(WebKit::AuxiliaryProcessProxy::didFinishLaunching):

  • UIProcess/WebPageProxy.cpp:

(WebKit::WebPageProxy::maybeInitializeSandboxExtensionHandle):
(WebKit::WebPageProxy::loadRequestWithNavigationShared):
(WebKit::WebPageProxy::loadFile):

  • WebProcess/WebPage/WebPage.cpp:

(WebKit::WebPage::fileLoadRequest):

  • WebProcess/WebPage/WebPage.h:
  • WebProcess/WebPage/WebPage.messages.in:
10:18 AM Changeset in webkit [249648] by Chris Dumez
  • 5 edits
    1 copy
    1 add in trunk

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in http/tests/workers/service/basic-register.html
https://bugs.webkit.org/show_bug.cgi?id=201590

Reviewed by Youenn Fablet.

Tools:

Add a new API test to check that if the default session registers a service worker, the
hasServiceWorkerRegistration flag is still false in an ephemeral session.

  • TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

LayoutTests:

Split the layout test in 2 to test both the default and the ephemeral session without relying on the
testRunner.setPrivateBrowsingEnabled_DEPRECATED() hack.

  • http/tests/workers/service/basic-register-expected.txt:
  • http/tests/workers/service/basic-register-private-expected.txt: Copied from LayoutTests/http/tests/workers/service/basic-register-expected.txt.
  • http/tests/workers/service/basic-register-private.html: Added.
  • http/tests/workers/service/resources/basic-register.js:

(async.test):

10:02 AM Changeset in webkit [249647] by youenn@apple.com
  • 11 edits
    2 deletes in trunk/Source/WebKit

Remove ServiceWorkerProcessProxy
https://bugs.webkit.org/show_bug.cgi?id=201506

Reviewed by Chris Dumez.

Instead of using a specific ServiceWorkerProcessProxy, move the logic directly in WebProcessProxy.
This will allow us in the future to run a service worker in a process that already runs a web page,
thus saving the need to spin off a new process.

This patch adds a new WebProcessProxy::createForServiceWorkers method instead of creating a
ServiceWorkerProcessProxy. The behavior is then similar: a process running a service worker cannot run a page.

  • Sources.txt:
  • UIProcess/API/Cocoa/WKWebsiteDataStore.mm:

(-[WKWebsiteDataStore _hasRegisteredServiceWorker]):

  • UIProcess/BackgroundProcessResponsivenessTimer.cpp:

(WebKit::BackgroundProcessResponsivenessTimer::shouldBeActive const):

  • UIProcess/Network/NetworkProcessProxy.cpp:

(WebKit::NetworkProcessProxy::openNetworkProcessConnection):

  • UIProcess/ServiceWorkerProcessProxy.cpp: Removed.
  • UIProcess/ServiceWorkerProcessProxy.h: Removed.
  • UIProcess/WebProcessPool.cpp:

(WebKit::WebProcessPool::establishWorkerContextConnectionToNetworkProcess):
(WebKit::WebProcessPool::disconnectProcess):
(WebKit::WebProcessPool::processForRegistrableDomain):
(WebKit::WebProcessPool::createWebPage):
(WebKit::WebProcessPool::updateServiceWorkerUserAgent):
(WebKit::WebProcessPool::mayHaveRegisteredServiceWorkers):
(WebKit::WebProcessPool::updateProcessAssertions):
(WebKit::WebProcessPool::isServiceWorkerPageID const):

  • UIProcess/WebProcessPool.h:
  • UIProcess/WebProcessProxy.cpp:

(WebKit::WebProcessProxy::createForServiceWorkers):
(WebKit::WebProcessProxy::getLaunchOptions):
(WebKit::WebProcessProxy::didBecomeUnresponsive):
(WebKit::WebProcessProxy::canBeAddedToWebProcessCache const):
(WebKit::WebProcessProxy::didSetAssertionState):
(WebKit::WebProcessProxy::didExceedCPULimit):
(WebKit::WebProcessProxy::establishServiceWorkerContext):
(WebKit::WebProcessProxy::setServiceWorkerUserAgent):
(WebKit::WebProcessProxy::updateServiceWorkerPreferencesStore):

  • UIProcess/WebProcessProxy.h:

(WebKit::WebProcessProxy::isRunningServiceWorkers const):
(WebKit::WebProcessProxy::hasServiceWorkerPageProxy):

  • UIProcess/glib/WebProcessPoolGLib.cpp:

(WebKit::WebProcessPool::platformInitializeWebProcess):

  • WebKit.xcodeproj/project.pbxproj:
8:35 AM Changeset in webkit [249646] by Jonathan Bedard
  • 2 edits in trunk/Tools

results.webkit.org: mobile-sidebar-control should override button
https://bugs.webkit.org/show_bug.cgi?id=201566

Rubber-stamped by Aakash Jain.

  • resultsdbpy/resultsdbpy/view/static/library/css/webkit.css:

(@media screen and (min-width: 600px)):
(.mobile-sidebar-control):
(.mobile-sidebar-control.display):

8:21 AM Changeset in webkit [249645] by commit-queue@webkit.org
  • 9 edits
    2 adds in trunk

[GTK][WPE] Remove attributes deprecated from MathML3
https://bugs.webkit.org/show_bug.cgi?id=197492

Patch by Rob Buis <rbuis@igalia.com> on 2019-09-09
Reviewed by Frédéric Wang.

LayoutTests/imported/w3c:

Import relevant WPT test.

  • web-platform-tests/mathml/relations/css-styling/attribute-mapping-001-expected.txt: Added.
  • web-platform-tests/mathml/relations/css-styling/attribute-mapping-001.html: Added.

Source/WebCore:

Remove some MathML3 deprecated attributes:
https://github.com/mathml-refresh/mathml/issues/5#issuecomment-475506856

This change also maps the dir attribute to direction for MathML Core.

Test: imported/w3c/web-platform-tests/mathml/relations/css-styling/attribute-mapping-001.html

  • mathml/MathMLElement.cpp:

(WebCore::MathMLElement::collectStyleForPresentationAttribute):

LayoutTests:

Disable the MathML Core flag for tests assuming MathML3 behavior.

  • mathml/presentation/attributes-background-color-expected.html:
  • mathml/presentation/attributes-background-color.html:
  • mathml/presentation/attributes-mathvariant.html:
  • mathml/presentation/direction-overall.html:
  • mathml/presentation/mstyle-css-attributes.html:
5:58 AM Changeset in webkit [249644] by zandobersek@gmail.com
  • 2 edits in trunk/JSTests

Unreviewed JSC test gardening.

  • stress/constructFunctionSkippingEvalEnabledCheck-should-throw-out-of-memory-error.js:

This test allocates a 2GB string before it goes out and tests
out-of-memory exception when appending other strings to it. As such,
skip the test on memory-limited platforms.

2:44 AM Changeset in webkit [249643] by Carlos Garcia Campos
  • 2 edits in trunk

Unreviewed. Bump GTK version numbers

  • Source/cmake/OptionsGTK.cmake:
2:21 AM Changeset in webkit [249642] by Carlos Garcia Campos
  • 1 copy in releases/WebKitGTK/webkit-2.26.0

WebKitGTK 2.26.0

2:20 AM Changeset in webkit [249641] by Carlos Garcia Campos
  • 4 edits in releases/WebKitGTK/webkit-2.26

Unreviewed. Update OptionsGTK.cmake and NEWS for 2.26.0 release

.:

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

Source/WebKit:

  • gtk/NEWS: Add release notes for 2.26.0.
2:08 AM Changeset in webkit [249640] by youenn@apple.com
  • 13 edits in trunk/Source

Use WebProcess processIdentifier to identify IDB connections between WebProcess and NetworkProcess
https://bugs.webkit.org/show_bug.cgi?id=201419

Reviewed by Alex Christensen.

We change from an identifier generated by NetworkProcess to WebProcess identifier.
A WebProcess identifier is unique globally and is stable even in case of NetworkProcess crash.
This removes the need for a sync IPC and makes things more easy to handle in case of NetworkProcess crash.

  • NetworkProcess/IndexedDB/WebIDBConnectionToClient.cpp:

(WebKit::WebIDBConnectionToClient::create):
(WebKit::WebIDBConnectionToClient::WebIDBConnectionToClient):
(WebKit::WebIDBConnectionToClient::getAllDatabaseNames):

  • NetworkProcess/IndexedDB/WebIDBConnectionToClient.h:
  • NetworkProcess/IndexedDB/WebIDBConnectionToClient.messages.in:
  • NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::establishIDBConnectionToServer):

  • NetworkProcess/NetworkConnectionToWebProcess.h:
  • NetworkProcess/NetworkConnectionToWebProcess.messages.in:
  • WebKit.xcodeproj/project.pbxproj:
  • WebProcess/Databases/IndexedDB/WebIDBConnectionToServer.cpp:

(WebKit::WebIDBConnectionToServer::WebIDBConnectionToServer):
(WebKit::WebIDBConnectionToServer::identifier const):
(WebKit::WebIDBConnectionToServer::coreConnectionToServer):
(WebKit::WebIDBConnectionToServer::getAllDatabaseNames):

  • WebProcess/Databases/IndexedDB/WebIDBConnectionToServer.h:
  • WebProcess/Network/NetworkProcessConnection.cpp:

(WebKit::NetworkProcessConnection::didReceiveMessage):
(WebKit::NetworkProcessConnection::didClose):
(WebKit::NetworkProcessConnection::idbConnectionToServerForSession):

  • WebProcess/Network/NetworkProcessConnection.h:

(WebKit::NetworkProcessConnection::existingIDBConnectionToServer const):

  • WebProcess/WebProcess.cpp:

(WebKit::WebProcess::networkProcessConnectionClosed):

1:54 AM WebKitGTK/2.26.x edited by Carlos Garcia Campos
(diff)
1:51 AM Changeset in webkit [249639] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.26/Source/WebKit

Merge r249569 - [GTK][WPE] Do not enable the sandbox in Snap
https://bugs.webkit.org/show_bug.cgi?id=201486

Reviewed by Carlos Garcia Campos.

Running inside of Snap adds its own complications and is simply
not supported for now.

Also update isInsideFlatpak() for consistency.

  • UIProcess/Launcher/glib/ProcessLauncherGLib.cpp:

(WebKit::isInsideFlatpak):
(WebKit::isInsideSnap):
(WebKit::ProcessLauncher::launchProcess):

1:51 AM Changeset in webkit [249638] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.26/Source/WebKit

Merge r249487 - [WPE][GTK] Fix sandbox parsing DISPLAY on X11
https://bugs.webkit.org/show_bug.cgi?id=201462

Reviewed by Michael Catanzaro.

On some setups the DISPLAY env var was incorrectly parsed.

  • UIProcess/Launcher/glib/BubblewrapLauncher.cpp:

(WebKit::bindX11):

1:50 AM Changeset in webkit [249637] by youenn@apple.com
  • 17 edits
    2 deletes in trunk/Source

Move checkProcessLocalPortForActivity from provider to registry
https://bugs.webkit.org/show_bug.cgi?id=201400

Reviewed by Alex Christensen.

Source/WebCore:

Update the registry to take a callback that is used to implement checkProcessLocalPortForActivity.
This allows WK1 and WK2 to have their own implementation.

Make the call to checkProcessLocalPortForActivity directly on the registry.
Remove unneeded worker checkProcessLocalPortForActivity method.
No change of behavior.

  • dom/messageports/MessagePortChannel.cpp:

(WebCore::MessagePortChannel::checkRemotePortForActivity):

  • dom/messageports/MessagePortChannelProvider.h:
  • dom/messageports/MessagePortChannelProviderImpl.cpp:

(WebCore::createMessagePortChannelRegistry):
(WebCore::MessagePortChannelProviderImpl::MessagePortChannelProviderImpl):
(WebCore::MessagePortChannelProviderImpl::checkProcessLocalPortForActivity): Deleted.

  • dom/messageports/MessagePortChannelProviderImpl.h:
  • dom/messageports/MessagePortChannelRegistry.cpp:

(WebCore::MessagePortChannelRegistry::MessagePortChannelRegistry):
(WebCore::MessagePortChannelRegistry::checkProcessLocalPortForActivity):

  • dom/messageports/MessagePortChannelRegistry.h:

(WebCore::MessagePortChannelRegistry::provider): Deleted.

  • dom/messageports/WorkerMessagePortChannelProvider.cpp:

(WebCore::WorkerMessagePortChannelProvider::checkProcessLocalPortForActivity): Deleted.

  • dom/messageports/WorkerMessagePortChannelProvider.h:

Source/WebKit:

Implement registry callback at NetworkProcess level.
Remove NetworkMessagePortChannelProvider since only a registry is now needed.

  • NetworkProcess/NetworkConnectionToWebProcess.cpp:
  • NetworkProcess/NetworkMessagePortChannelProvider.cpp: Removed.
  • NetworkProcess/NetworkMessagePortChannelProvider.h: Removed.
  • NetworkProcess/NetworkProcess.cpp:

(WebKit::createMessagePortChannelRegistry):
(WebKit::m_messagePortChannelRegistry):
(WebKit::m_messagePortChannelProvider): Deleted.

  • NetworkProcess/NetworkProcess.h:

(WebKit::NetworkProcess::messagePortChannelRegistry):

  • Sources.txt:
  • WebKit.xcodeproj/project.pbxproj:
  • WebProcess/WebCoreSupport/WebMessagePortChannelProvider.cpp:

(WebKit::WebMessagePortChannelProvider::checkProcessLocalPortForActivity): Deleted.

  • WebProcess/WebCoreSupport/WebMessagePortChannelProvider.h:
1:43 AM Changeset in webkit [249636] by Adrian Perez de Castro
  • 2 edits in releases/WebKitGTK/webkit-2.26/Source/WebCore/platform/gtk/po

Merged r249633 - [l10n] [pt_BR] Updated Brazilian Portuguese translation
https://bugs.webkit.org/show_bug.cgi?id=201592

Patch by Rafael Fontenelle <rafaelff@gnome.org> on 2019-09-09
Rubber-stamped by Carlos Garcia Campos.

  • pt_BR.po:
1:41 AM Changeset in webkit [249635] by Adrian Perez de Castro
  • 2 edits in releases/WebKitGTK/webkit-2.26/Source/WebCore/platform/gtk/po

Merged r249634 - Updated German translation
https://bugs.webkit.org/show_bug.cgi?id=201582

Patch by Christian Kirbach <Christian.Kirbach@googlemail.com> on 2019-09-07
Rubber-stamped by Adrian Perez de Castro.

  • de.po:
1:37 AM Changeset in webkit [249634] by Adrian Perez de Castro
  • 2 edits in trunk/Source/WebCore/platform/gtk/po

Updated German translation
https://bugs.webkit.org/show_bug.cgi?id=201582

Patch by Christian Kirbach <Christian.Kirbach@googlemail.com> on 2019-09-07
Rubber-stamped by Adrian Perez de Castro.

  • de.po:
1:29 AM Changeset in webkit [249633] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebCore/platform/gtk/po

[l10n] [pt_BR] Updated Brazilian Portuguese translation
https://bugs.webkit.org/show_bug.cgi?id=201592

Patch by Rafael Fontenelle <rafaelff@gnome.org> on 2019-09-09
Rubber-stamped by Carlos Garcia Campos.

  • pt_BR.po:

Sep 8, 2019:

11:41 PM Changeset in webkit [249632] by david_quesada@apple.com
  • 2 edits in trunk/Source/WebKit

Remove some unused constants in VersionChecks.h
https://bugs.webkit.org/show_bug.cgi?id=201599

Reviewed by Youenn Fablet.

  • UIProcess/Cocoa/VersionChecks.h:

Remove two macros that are unused as of r244512.

9:04 PM Changeset in webkit [249631] by Wenson Hsieh
  • 2 edits in trunk/Source/WebKit

[macCatalyst] The last typed character in password fields shouldn't be echoed
https://bugs.webkit.org/show_bug.cgi?id=201595
<rdar://problem/55166367>

Reviewed by Tim Horton.

Disable password echoing on macCatalyst, to better match behavior on macOS.

  • Shared/WebPreferencesDefaultValues.h:
8:23 PM Changeset in webkit [249630] by sbarati@apple.com
  • 9 edits
    2 adds in trunk

[WHLSL] Add back a version of the property resolver
https://bugs.webkit.org/show_bug.cgi?id=201251

Reviewed by Robin Morisset.

Source/WebCore:

This patch adds back the property resolver to aid in how we emit metal code.
The goal of the new property resolver phase is to allow two things:

  1. For property access expressions, metal codegen should be allowed to evaluate

the base, and if it's an index expression, the index, as many times as needed.
So this patch ensures that if Metal evaluates such things, effects aren't performed
more than once.

  1. For ReadModifyWrite expressions, metal codegen should be able to evaluate the

leftValueExpression as many times as it'd like without performing the effects of
leftValueExpression more than once.

Here are a couple examples of this transformation:
a()[b()] -> (temp1 = a(), temp2 = b(), temp1[temp2])
(*a())[b()] += 42; -> (temp1 = &*a(), temp2 = b(), (*temp1)[temp2] += 42)

The interesting rules break down like:

  • When the base is an lvalue, we take its address and replace the old base

with a dereference of the new pointer variable.

  • When the base is an rvalue, we evaluate it and store it in a temporary

variable, and switch the base to be a reference of that variable.

  • All indices are evaluated before the property access instruction and

replaced with a reference to the temporary variable.

Test: webgpu/whlsl/property-evaluation-order.html

  • Modules/webgpu/WHLSL/AST/WHLSLExpression.cpp:

(WebCore::WHLSL::AST::Expression::mayBeEffectful const):

  • Modules/webgpu/WHLSL/AST/WHLSLExpression.h:
  • Modules/webgpu/WHLSL/AST/WHLSLIndexExpression.h:
  • Modules/webgpu/WHLSL/AST/WHLSLReadModifyWriteExpression.h:
  • Modules/webgpu/WHLSL/Metal/WHLSLFunctionWriter.cpp:

(WebCore::WHLSL::Metal::FunctionDefinitionWriter::visit):

  • Modules/webgpu/WHLSL/WHLSLHighZombieFinder.cpp:

(WebCore::WHLSL::findHighZombies):

  • Modules/webgpu/WHLSL/WHLSLPropertyResolver.cpp:

(WebCore::WHLSL::PropertyResolver::handleLeftHandSideBase):
(WebCore::WHLSL::PropertyResolver::handlePropertyAccess):
(WebCore::WHLSL::resolveProperties):

LayoutTests:

  • webgpu/whlsl/property-evaluation-order-expected.txt: Added.
  • webgpu/whlsl/property-evaluation-order.html: Added.

Sep 7, 2019:

11:10 PM Changeset in webkit [249629] by Chris Dumez
  • 15 edits in trunk

Add support for postMessage buffering between the service worker and window
https://bugs.webkit.org/show_bug.cgi?id=201169

Reviewed by Youenn Fablet.

LayoutTests/imported/w3c:

Rebaseline WPT test that is now passing.

  • web-platform-tests/service-workers/service-worker/postmessage-to-client-message-queue.https-expected.txt:

Source/WebCore:

As per the Service Worker specification, a service worker client's message
queue is initially disabled and only gets enabled after:

  • The DOMContentLoaded event has been fired

or

  • The client sets the navigator.serviceWorker.onmessage event handler

or

  • navigator.serviceWorker.startMessages() is called

While the message queue is disabled, messages posted by the service worker
to the client simply get queued and only get processed once the queue gets
enabled.

No new tests, rebaselined existing test.

  • dom/Document.cpp:

(WebCore::Document::finishedParsing):
Call startMessages() on the ServiceWorkerContainer once the DOMContentLoaded event has
been fired.

  • dom/ScriptExecutionContext.cpp:

(WebCore::ScriptExecutionContext::ensureServiceWorkerContainer):

  • dom/ScriptExecutionContext.h:
  • workers/service/SWClientConnection.cpp:

(WebCore::SWClientConnection::postMessageToServiceWorkerClient):
Fix a bug where a service worker would not be able to post a message to a client until
that client has accessed navigator.serviceWorker (since the ServiceWorkerContainer is
lazy initialized). To address the issue, we now initialize the ServiceWorkerContainer
when a message is received from the service worker. Previously, messages were just
getting dropped.

  • workers/service/ServiceWorkerContainer.cpp:

(WebCore::ServiceWorkerContainer::ServiceWorkerContainer):
When the ServiceWorkerContainer is constructed, suspend its message queue if its context
document is still parsing.

(WebCore::ServiceWorkerContainer::startMessages):
Resume the message queue when startMessages() is called.

(WebCore::ServiceWorkerContainer::postMessage):
Enqueue the event instead of firing it right away.

(WebCore::ServiceWorkerContainer::addEventListener):
if navigator.serviceWorker.onmessage event handler gets set by the JavaScript, call
startMessages().

  • workers/service/ServiceWorkerContainer.h:

LayoutTests:

Unskip test that is no longer timing out.

  • resources/testharnessreport.js:

(self.testRunner.add_completion_callback):
Use testRunner.forceImmediateCompletion() instead of notifyDone() for WPT tests.
testRunner.notifyDone() does not work in case of load error or when the load
does not finish. The WPT test was timing out because the load does not finish for
testing purposes.

10:57 PM Changeset in webkit [249628] by david_quesada@apple.com
  • 3 edits in trunk/Source/WTF

REGRESSION(r248533): Unable to use WTF::RefCounted when building in a debug configuration against a non-debug WebKit
https://bugs.webkit.org/show_bug.cgi?id=201585
rdar://problem/55153369

Reviewed by Chris Dumez.

Export WTF::RefCountedBase::areThreadingChecksEnabledGlobally regardless of whether or not assertions
are enabled for the WTF being built. This allows WebKit-based projects to use RefCounted for their own
objects in a debug configuration without requiring a debug build of WebKit.

  • wtf/RefCounted.cpp:
  • wtf/RefCounted.h:
10:42 PM Changeset in webkit [249627] by Chris Dumez
  • 15 edits in trunk

[Service Workers] Drop support for registration resurrection
https://bugs.webkit.org/show_bug.cgi?id=201584

Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Re-sync a couple of outdated WPT tests from upstream 820f0f86047e6e.

  • web-platform-tests/service-workers/service-worker/unregister-then-register-new-script.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/unregister-then-register-new-script.https.html:
  • web-platform-tests/service-workers/service-worker/unregister-then-register.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/unregister-then-register.https.html:

Source/WebCore:

Drop support for registration resurrection (Unregistering then quickly re-registering would resurrect
the original registration instead of creating a new one). This behavior is no longer in the
specification (No more "isUninstalling" flag on the registration, and the registration is removed
from the "scope to registration map" right away upon unregistering).

This resurrection behavior was causing a lot of flakiness in the tests so it is likely we will be able
to unskip some tests. This is also what was causing the flakiness that caused Bug 201169 to be rolled
out.

No new tests, updated existing tests.

  • workers/service/server/SWServer.cpp:

(WebCore::SWServer::activeWorkerFromRegistrationID):
(WebCore::SWServer::getRegistration):
(WebCore::SWServer::addRegistration):
(WebCore::SWServer::removeRegistration):
(WebCore::SWServer::getRegistrations):
(WebCore::SWServer::clearAll):
(WebCore::SWServer::clear):
(WebCore::SWServer::terminatePreinstallationWorker):
(WebCore::SWServer::didFinishActivation):
(WebCore::SWServer::claim):
(WebCore::SWServer::addClientServiceWorkerRegistration):
(WebCore::SWServer::removeClientServiceWorkerRegistration):
(WebCore::SWServer::installContextData):
(WebCore::SWServer::runServiceWorker):
(WebCore::SWServer::doRegistrationMatching):
(WebCore::SWServer::registrationFromServiceWorkerIdentifier):
(WebCore::SWServer::registerServiceWorkerClient):
(WebCore::SWServer::unregisterServiceWorkerClient):
(WebCore::SWServer::removeFromScopeToRegistrationMap):
(WebCore::SWServer::performGetOriginsWithRegistrationsCallbacks):

  • workers/service/server/SWServer.h:
  • workers/service/server/SWServerJobQueue.cpp:

(WebCore::SWServerJobQueue::runRegisterJob):
(WebCore::SWServerJobQueue::runUnregisterJob):
(WebCore::SWServerJobQueue::runUpdateJob):

  • workers/service/server/SWServerRegistration.cpp:

(WebCore::SWServerRegistration::clear):
(WebCore::SWServerRegistration::handleClientUnload):
(WebCore::SWServerRegistration::isUnregistered const):

  • workers/service/server/SWServerRegistration.h:

(WebCore::SWServerRegistration::scopeURLWithoutFragment const):

  • workers/service/server/SWServerWorker.cpp:

(WebCore::SWServerWorker::SWServerWorker):
(WebCore::m_scriptResourceMap):
(WebCore::SWServerWorker::contextData const):
(WebCore::SWServerWorker::skipWaiting):
(WebCore::SWServerWorker::setHasPendingEvents):
(WebCore::SWServerWorker::setState):
(WebCore::SWServerWorker::registration const):

  • workers/service/server/SWServerWorker.h:
10:26 PM Changeset in webkit [249626] by Chris Dumez
  • 6 edits
    3 deletes in trunk

Rewrite http/tests/workers/service/serviceworker-private-browsing.https.html as an API test
https://bugs.webkit.org/show_bug.cgi?id=201574

Reviewed by Alex Christensen.

Source/WebKit:

Update Service Workers so that they obey the "secureContextChecksEnabled" preference.
This was needed to allow my service worker to access self.caches in my API test.

  • WebProcess/Storage/WebSWContextManagerConnection.cpp:

(WebKit::WebSWContextManagerConnection::updatePreferencesStore):

Tools:

Rewrite http/tests/workers/service/serviceworker-private-browsing.https.html as an API test so that
we truly test private browsing. testRunner.setPrivateBrowsingEnabled_DEPRECATED() is a hack which
does not work well for WK2.

  • TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

LayoutTests:

Drop legacy test which was replaced by an API test.

  • http/tests/workers/service/serviceworker-private-browsing-worker.js: Removed.
  • http/tests/workers/service/serviceworker-private-browsing.https-expected.txt: Removed.
  • http/tests/workers/service/serviceworker-private-browsing.https.html: Removed.
10:11 PM Changeset in webkit [249625] by Chris Dumez
  • 5 edits
    4 deletes in trunk

Rewrite storage/domstorage/localstorage/private-browsing-affects-storage.html as an API test
https://bugs.webkit.org/show_bug.cgi?id=201547

Reviewed by Alex Christensen.

Tools:

Rewrite storage/domstorage/localstorage/private-browsing-affects-storage.html as an API test so
it can use a true ephemeral session instead of the testRunner.setPrivateBrowsingEnabled_DEPRECATED()
hack.

  • TestWebKitAPI/Tests/WebKitCocoa/LocalStoragePersistence.mm:

(-[LocalStorageNavigationDelegate webView:decidePolicyForNavigationAction:preferences:decisionHandler:]):
(-[LocalStorageNavigationDelegate webView:didFinishNavigation:]):
(TEST):

LayoutTests:

Drop outdated test.

  • storage/domstorage/localstorage/private-browsing-affects-storage.html: Removed.
  • storage/domstorage/localstorage/resources/private-browsing-1.html: Removed.
  • storage/domstorage/localstorage/resources/private-browsing-2.html: Removed.
9:17 PM Changeset in webkit [249624] by timothy_horton@apple.com
  • 2 edits in trunk/Source/WebKit

Marking up a note on iOS results in a PDF with no contents
https://bugs.webkit.org/show_bug.cgi?id=201530
<rdar://problem/53686019>

Unreviewed follow-up.

  • UIProcess/ios/WKContentView.mm:

(-[WKContentView _waitForDrawToPDFCallback]):
Remove an ASSERT_NOT_REACHED that fires if the synchronous wait for printing
times out. This is a totally legitimite situation that can occur if the process
crashes; there's no need for an assert there.

8:58 PM Changeset in webkit [249623] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Null check m_client in NetworkDataTask::didReceiveResponse
https://bugs.webkit.org/show_bug.cgi?id=201587
<rdar://problem/54743444>

Reviewed by Chris Dumez.

clearClient can be called at any time. We null check it everyhere else we use it,
and we are seeing crashes at this call site. Let's not crash.

  • NetworkProcess/NetworkDataTask.cpp:

(WebKit::NetworkDataTask::didReceiveResponse):

7:29 PM Changeset in webkit [249622] by achristensen@apple.com
  • 17 edits in trunk

Remove WebsiteDataStore::legacyDefaultDataStoreConfiguration
https://bugs.webkit.org/show_bug.cgi?id=200054

Reviewed by Carlos Garcia Campos.

Source/WebKit:

Its last needed use was removed in rdar://problem/47030981
This changes WKContextConfigurationCreateWithLegacyOptions to be equal to WKContextConfigurationCreate,
but it has no remaining clients so it is only being retained for binary compabibility.
This also changes WKContextCreateWithInjectedBundlePath and WKContextCreate to be equal to WKContextCreateWithConfiguration,
and after rdar://problem/47030981 there are no meaningful clients of those functions, also.
There is one existing client of WKContextCreateWithInjectedBundlePath in rdar://problem/50323967 but our current implementation
is compatible enough for them because it uses their injected bundle still.

  • UIProcess/API/APIProcessPoolConfiguration.cpp:

(API::ProcessPoolConfiguration::copy):
(API::ProcessPoolConfiguration::createWithLegacyOptions): Deleted.

  • UIProcess/API/APIProcessPoolConfiguration.h:
  • UIProcess/API/APIWebsiteDataStore.cpp:

(API::WebsiteDataStore::createLegacy): Deleted.
(API::WebsiteDataStore::legacyDefaultDataStoreConfiguration): Deleted.

  • UIProcess/API/APIWebsiteDataStore.h:
  • UIProcess/API/C/WKContext.cpp:

(WKContextCreate):
(WKContextCreateWithInjectedBundlePath):
(WKContextGetWebsiteDataStore):

  • UIProcess/API/C/WKContextConfigurationRef.cpp:

(WKContextConfigurationCreate):
(WKContextConfigurationCreateWithLegacyOptions):

  • UIProcess/API/C/WKContextConfigurationRef.h:
  • UIProcess/API/Cocoa/APIWebsiteDataStoreCocoa.mm:

(API::WebsiteDataStore::legacyDefaultApplicationCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultNetworkCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultWebSQLDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultIndexedDBDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultLocalStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaKeysStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultDeviceIdHashSaltsStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultJavaScriptConfigurationDirectory): Deleted.

  • UIProcess/API/Cocoa/WKProcessGroup.mm:

(-[WKProcessGroup initWithInjectedBundleURL:]):

  • UIProcess/API/glib/APIWebsiteDataStoreGLib.cpp:

(API::WebsiteDataStore::legacyDefaultApplicationCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultNetworkCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultWebSQLDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultIndexedDBDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultLocalStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaKeysStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultDeviceIdHashSaltsStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultJavaScriptConfigurationDirectory): Deleted.

  • UIProcess/API/win/APIWebsiteDataStoreWin.cpp:

(API::WebsiteDataStore::legacyDefaultApplicationCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultNetworkCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultWebSQLDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultIndexedDBDatabaseDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultLocalStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaCacheDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultMediaKeysStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultDeviceIdHashSaltsStorageDirectory): Deleted.
(API::WebsiteDataStore::legacyDefaultJavaScriptConfigurationDirectory): Deleted.

  • UIProcess/WebInspectorUtilities.cpp:

(WebKit::inspectorProcessPool):

  • UIProcess/WebProcessPool.cpp:

(WebKit::m_webProcessCache):
(WebKit::legacyWebsiteDataStoreConfiguration): Deleted.

  • UIProcess/WebProcessPool.h:

Tools:

  • TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:

This test relied on the default data store having been instantiated.

6:19 PM Changeset in webkit [249621] by mark.lam@apple.com
  • 2 edits in trunk/Source/bmalloc

[Follow up fix] Harden protection of the Gigacage Config parameters.
https://bugs.webkit.org/show_bug.cgi?id=201570
<rdar://problem/55134229>

Reviewed by Saam Barati.

The previously landed hardening measures is failing the assertion in
forbidDisablingPrimitiveGigacage() when a replacement heap (e.g. gmalloc) is
used.

  1. forbidDisablingPrimitiveGigacage() should just ensureGigacage() so that we know that the Gigacage is properly configured (whether enabled or disabled).
  2. Strengthen the assertion in forbidDisablingPrimitiveGigacage() to ensure that we have already determined if the Gigacage should be enabled.
  • bmalloc/Gigacage.cpp:

(Gigacage::forbidDisablingPrimitiveGigacage):

3:56 PM Changeset in webkit [249620] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Fix Windows build
https://bugs.webkit.org/show_bug.cgi?id=201517

  • WebProcess/WebCoreSupport/WebResourceLoadObserver.cpp:
3:54 PM Changeset in webkit [249619] by achristensen@apple.com
  • 8 edits in trunk

Remove calls to WKContextGetWebsiteDataStore missed in r249614
https://bugs.webkit.org/show_bug.cgi?id=200050

Source/WebKit:

It turns out I didn't need to introduce another way of clearing all cookies between tests.
I just needed to make the call to WKHTTPCookieStoreDeleteAllCookies use the right WebsiteDataStore.

  • NetworkProcess/NetworkProcess.cpp:

(WebKit::NetworkProcess::clearCachedCredentials):

  • NetworkProcess/NetworkSession.h:

(WebKit::NetworkSession::clearCredentials):
(WebKit::NetworkSession::clearCookies): Deleted.

  • NetworkProcess/cocoa/NetworkSessionCocoa.h:
  • NetworkProcess/cocoa/NetworkSessionCocoa.mm:

(WebKit::NetworkSessionCocoa::clearCookies): Deleted.

Tools:

  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::resetPreferencesToConsistentValues):
(WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::invoke):

3:45 PM Changeset in webkit [249618] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Fix Windows build.
https://bugs.webkit.org/show_bug.cgi?id=200050

  • PlatformWin.cmake:
3:26 PM Changeset in webkit [249617] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Fix Windows build.
https://bugs.webkit.org/show_bug.cgi?id=201517

  • WebProcess/WebCoreSupport/WebResourceLoadObserver.cpp:
2:45 PM Changeset in webkit [249616] by achristensen@apple.com
  • 2 edits in trunk/Tools

Remove invalid assertion added in r249614
https://bugs.webkit.org/show_bug.cgi?id=200050

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::TestController::cocoaResetStateToConsistentValues):
Something is instantiating the default data store, which is no problem right now.

1:20 PM Changeset in webkit [249615] by Ryan Haddad
  • 16 edits in trunk

Unreviewed, rolling out r249353.

The test for this change is a flaky failure.

Reverted changeset:

"Add support for postMessage buffering between the service
worker and window"
https://bugs.webkit.org/show_bug.cgi?id=201169
https://trac.webkit.org/changeset/249353

9:04 AM Changeset in webkit [249614] by achristensen@apple.com
  • 33 edits
    2 adds in trunk

Deprecate WKContextGetWebsiteDataStore
https://bugs.webkit.org/show_bug.cgi?id=200050

Reviewed by Geoff Garen.

Source/WebCore:

  • platform/network/CacheValidation.cpp:

(WebCore::collectVaryingRequestHeaders):
(WebCore::verifyVaryingRequestHeaders):

  • platform/network/CacheValidation.h:

Replace a few NetworkStorageSession& with NetworkStorageSession*
so I can replace defaultStorageSession with the correct NetworkStorageSession in NetworkCache.cpp.

Source/WebKit:

The last use of it was removed in rdar://problem/47030981
In order to remove its use in WebKitTestRunner, I needed to make WKWebsiteDataStoreConfigurationRef
to replace the WKContextConfigurationRef members that should've been associated with the WebsiteDataStore the whole time.
There is one use of WKContextGetWebsiteDataStore in TestWebKitAPI still I'll investigate later.

  • Shared/API/c/WKBase.h:
  • Sources.txt:
  • UIProcess/API/APIWebsiteDataStore.cpp:

(API::WebsiteDataStore::create):

  • UIProcess/API/APIWebsiteDataStore.h:
  • UIProcess/API/C/WKAPICast.h:
  • UIProcess/API/C/WKContext.cpp:

(WKContextGetApplicationCacheManager):
(WKContextGetKeyValueStorageManager):
(WKContextGetResourceCacheManager):

  • UIProcess/API/C/WKContext.h:
  • UIProcess/API/C/WKContextConfigurationRef.cpp:

(WKContextConfigurationCopyDiskCacheDirectory):
(WKContextConfigurationSetDiskCacheDirectory):
(WKContextConfigurationCopyApplicationCacheDirectory):
(WKContextConfigurationSetApplicationCacheDirectory):
(WKContextConfigurationCopyIndexedDBDatabaseDirectory):
(WKContextConfigurationSetIndexedDBDatabaseDirectory):
(WKContextConfigurationCopyLocalStorageDirectory):
(WKContextConfigurationSetLocalStorageDirectory):
(WKContextConfigurationCopyWebSQLDatabaseDirectory):
(WKContextConfigurationSetWebSQLDatabaseDirectory):
(WKContextConfigurationCopyMediaKeysStorageDirectory):
(WKContextConfigurationSetMediaKeysStorageDirectory):
(WKContextConfigurationCopyResourceLoadStatisticsDirectory):
(WKContextConfigurationSetResourceLoadStatisticsDirectory):

  • UIProcess/API/C/WKContextConfigurationRef.h:
  • UIProcess/API/C/WKWebsiteDataStoreConfigurationRef.cpp: Added.

(WKWebsiteDataStoreConfigurationGetTypeID):
(WKWebsiteDataStoreConfigurationCreate):
(WKWebsiteDataStoreConfigurationCopyApplicationCacheDirectory):
(WKWebsiteDataStoreConfigurationSetApplicationCacheDirectory):
(WKWebsiteDataStoreConfigurationCopyNetworkCacheDirectory):
(WKWebsiteDataStoreConfigurationSetNetworkCacheDirectory):
(WKWebsiteDataStoreConfigurationCopyIndexedDBDatabaseDirectory):
(WKWebsiteDataStoreConfigurationSetIndexedDBDatabaseDirectory):
(WKWebsiteDataStoreConfigurationCopyLocalStorageDirectory):
(WKWebsiteDataStoreConfigurationSetLocalStorageDirectory):
(WKWebsiteDataStoreConfigurationCopyWebSQLDatabaseDirectory):
(WKWebsiteDataStoreConfigurationSetWebSQLDatabaseDirectory):
(WKWebsiteDataStoreConfigurationCopyMediaKeysStorageDirectory):
(WKWebsiteDataStoreConfigurationSetMediaKeysStorageDirectory):
(WKWebsiteDataStoreConfigurationCopyResourceLoadStatisticsDirectory):
(WKWebsiteDataStoreConfigurationSetResourceLoadStatisticsDirectory):

  • UIProcess/API/C/WKWebsiteDataStoreConfigurationRef.h: Added.
  • UIProcess/API/C/WKWebsiteDataStoreRef.cpp:

(WKWebsiteDataStoreCreateWithConfiguration):

  • UIProcess/API/C/WKWebsiteDataStoreRef.h:
  • WebKit.xcodeproj/project.pbxproj:

Tools:

  • MiniBrowser/win/WebKitBrowserWindow.cpp:

(WebKitBrowserWindow::updateProxySettings):

  • TestWebKitAPI/Tests/WebKitCocoa/ServiceWorkerBasic.mm:
  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::generateContextConfiguration const):
(WTR::TestController::websiteDataStore):
(WTR::TestController::generatePageConfiguration):
(WTR::TestController::resetStateToConsistentValues):
(WTR::TestController::platformAdjustContext):
(WTR::TestController::clearServiceWorkerRegistrations):
(WTR::TestController::clearDOMCache):
(WTR::TestController::clearDOMCaches):
(WTR::TestController::ClearIndexedDatabases):
(WTR::TestController::hasDOMCache):
(WTR::TestController::domCacheSize):
(WTR::TestController::setStatisticsDebugMode):
(WTR::TestController::setStatisticsPrevalentResourceForDebugMode):
(WTR::TestController::setStatisticsLastSeen):
(WTR::TestController::setStatisticsPrevalentResource):
(WTR::TestController::setStatisticsVeryPrevalentResource):
(WTR::TestController::dumpResourceLoadStatistics):
(WTR::TestController::isStatisticsPrevalentResource):
(WTR::TestController::isStatisticsVeryPrevalentResource):
(WTR::TestController::isStatisticsRegisteredAsSubresourceUnder):
(WTR::TestController::isStatisticsRegisteredAsSubFrameUnder):
(WTR::TestController::isStatisticsRegisteredAsRedirectingTo):
(WTR::TestController::setStatisticsHasHadUserInteraction):
(WTR::TestController::isStatisticsHasHadUserInteraction):
(WTR::TestController::setStatisticsGrandfathered):
(WTR::TestController::isStatisticsGrandfathered):
(WTR::TestController::setStatisticsSubframeUnderTopFrameOrigin):
(WTR::TestController::setStatisticsSubresourceUnderTopFrameOrigin):
(WTR::TestController::setStatisticsSubresourceUniqueRedirectTo):
(WTR::TestController::setStatisticsSubresourceUniqueRedirectFrom):
(WTR::TestController::setStatisticsTopFrameUniqueRedirectTo):
(WTR::TestController::setStatisticsTopFrameUniqueRedirectFrom):
(WTR::TestController::setStatisticsCrossSiteLoadWithLinkDecoration):
(WTR::TestController::setStatisticsTimeToLiveUserInteraction):
(WTR::TestController::statisticsProcessStatisticsAndDataRecords):
(WTR::TestController::statisticsUpdateCookieBlocking):
(WTR::TestController::statisticsSubmitTelemetry):
(WTR::TestController::setStatisticsNotifyPagesWhenDataRecordsWereScanned):
(WTR::TestController::setStatisticsIsRunningTest):
(WTR::TestController::setStatisticsShouldClassifyResourcesBeforeDataRecordsRemoval):
(WTR::TestController::setStatisticsNotifyPagesWhenTelemetryWasCaptured):
(WTR::TestController::setStatisticsMinimumTimeBetweenDataRecordsRemoval):
(WTR::TestController::setStatisticsGrandfatheringTime):
(WTR::TestController::setStatisticsMaxStatisticsEntries):
(WTR::TestController::setStatisticsPruneEntriesDownTo):
(WTR::TestController::statisticsClearInMemoryAndPersistentStore):
(WTR::TestController::statisticsClearInMemoryAndPersistentStoreModifiedSinceHours):
(WTR::TestController::statisticsClearThroughWebsiteDataRemoval):
(WTR::TestController::statisticsDeleteCookiesForHost):
(WTR::TestController::isStatisticsHasLocalStorage):
(WTR::TestController::setStatisticsCacheMaxAgeCap):
(WTR::TestController::statisticsResetToConsistentState):
(WTR::TestController::setWebAuthenticationMockConfiguration):
(WTR::TestController::clearAdClickAttributionsThroughWebsiteDataRemoval):

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

(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::initializeWebViewConfiguration):

LayoutTests:

Rebase some accessibility tests that were printing the SessionID since we're using SessionID 2 for testing now.

  • accessibility/mac/document-attributes-expected.txt:
  • platform/mac/accessibility/parent-delete-expected.txt:
8:01 AM Changeset in webkit [249613] by mark.lam@apple.com
  • 7 edits in trunk/Source

performJITMemcpy() source buffer should not be in the Gigacage.
https://bugs.webkit.org/show_bug.cgi?id=201577
<rdar://problem/55142606>

Reviewed by Michael Saboff.

Source/bmalloc:

  1. Add the Gigacage start address and totalSize to the Config.
  2. Add a contains() function that uses the start address and totalSize to check if a given pointer is in the Gigacage's address range.
  • bmalloc/Gigacage.cpp:

(Gigacage::ensureGigacage):
(Gigacage::verifyGigacageIsEnabled):

  • bmalloc/Gigacage.h:

(Gigacage::contains):

Source/JavaScriptCore:

Add a RELEASE_ASSERT in performJITMemcpy() to ensure that the passed in source
buffer is not in the Gigacage.

  • jit/ExecutableAllocator.h:

(JSC::performJITMemcpy):

Source/WTF:

  • wtf/Gigacage.h:

(Gigacage::contains):

7:53 AM Changeset in webkit [249612] by mark.lam@apple.com
  • 6 edits in trunk

The jsc shell should allow disabling of the Gigacage for testing purposes.
https://bugs.webkit.org/show_bug.cgi?id=201579

Reviewed by Michael Saboff.

JSTests:

Unskip the tests now.

  • stress/disable-gigacage-arrays.js:
  • stress/disable-gigacage-strings.js:
  • stress/disable-gigacage-typed-arrays.js:

Source/JavaScriptCore:

Check for the same GIGACAGE_ENABLED env var that is checked by Gigacage code. If
this env var is present and it has a falsy value, then do not
forbidDisablingPrimitiveGigacage() in the jsc shell.

  • jsc.cpp:

(jscmain):

3:13 AM Changeset in webkit [249611] by mark.lam@apple.com
  • 4 edits in trunk/JSTests

Gardening: temporarily skipping these tests until the fix can be reviewed and landed.

Not reviewed.

See https://bugs.webkit.org/show_bug.cgi?id=201579 for the fix.

  • stress/disable-gigacage-arrays.js:
  • stress/disable-gigacage-strings.js:
  • stress/disable-gigacage-typed-arrays.js:
2:50 AM Changeset in webkit [249610] by mark.lam@apple.com
  • 2 edits in trunk/JSTests

Gardening: speculative test fix to green bots [attempt #2].
https://bugs.webkit.org/show_bug.cgi?id=201529
<rdar://problem/53935772>

Not reviewed.

  • stress/test-out-of-memory.js:

Sep 6, 2019:

10:59 PM Changeset in webkit [249609] by Brent Fulgham
  • 3 edits in trunk/Source/WebCore

[FTW] Minor cleanups to avoid crashes during web browsing
https://bugs.webkit.org/show_bug.cgi?id=201572

Reviewed by Alexey Proskuryakov.

This patch addresses a few crashes seen when doing general web browsing using
the FTW build of WebKit:

  1. ImageBuffer::sinkIntoImage was attempting to use a render target after WTFMoving it elsewhere.
  2. ImageBuffer::copyNativeImage was not properly checking the type of the render target to make sure it was suitable for Bitmap operations.
  3. BackingStoreBackendDirect2DImpl::scroll would crash if the scroll offset was outside the bounds of the current view. In that case, it would attempt to allocate a zero-size texture, which triggered a crash.
  • platform/graphics/win/BackingStoreBackendDirect2DImpl.cpp:

(WebCore::BackingStoreBackendDirect2DImpl::scroll):

  • platform/graphics/win/ImageBufferDirect2D.cpp:

(WebCore::ImageBuffer::sinkIntoImage):
(WebCore::ImageBuffer::copyNativeImage const):

10:42 PM Changeset in webkit [249608] by mark.lam@apple.com
  • 11 edits in trunk/Source

Harden protection of the Gigacage Config parameters.
https://bugs.webkit.org/show_bug.cgi?id=201570
<rdar://problem/55134229>

Reviewed by Saam Barati.

Source/bmalloc:

  1. Rename protectGigacageBasePtrs() and unprotectGigacageBasePtrs() to freezeGigacageConfig() and unfreezeGigacageConfig() respectively.

Also move the alignment check in protectGigacageBasePtrs() into ensureGigacage().
There's no need to check it more than once.

  1. Introduce permanentlyFreezeGigacageConfig() which permanently makes the Config ReadOnly. Once invoked, the Config cannot be made writable again.

This is made possible by using vm_protect with a true set_maximum argument.
We also add a g_gigacageConfig.isPermanentlyFrozen flag that we assert.
Note: this permanence is only true for OS(DARWIN) since vm_protect is Mach API.

  1. Rename disableDisablingPrimitiveGigacageIfShouldBeEnabled() to forbidDisablingPrimitiveGigacage() because "disablingDisabling" is a tongue twister.

Also, we don't want to make it conditional on "IfShouldBeEnabled". We want
forbidDisablingPrimitiveGigacage() to be irreversible. It is up to the client
to ensure that the Gigacage is already initialized (if possible) before
invoking forbidDisablingPrimitiveGigacage(). Conceptually, the forbidding
isn't about guaranteeing that the Gigacage is enabled. It only takes away the
option to disable it.

That said, forbidDisablingPrimitiveGigacage() is only invoked by clients that
care about keeping the Gigacage enabled. So, it does a sanity check (with an
assertion) that if !GIGACAGE_ALLOCATION_CAN_FAIL, then the Gigacage should be
have been initialized and enabled before invoking it.

We also make sure that forbidDisablingPrimitiveGigacage() calls
permanentlyFreezeGigacageConfig() unconditionally. It is safe to call it more
than once. This guarantees that the Config is permanently frozen after this,
even if a bug should inadvertantly set the
g_gigacageConfig.disablingPrimitiveGigacageIsForbidden flag before
forbidDisablingPrimitiveGigacage() is invoked.

  1. Assert that ensureGigacage() is only called once.
  1. Assert that shouldBeEnabled() is only called once. Also moved its cached result into the Config so that it can be frozen.
  • bmalloc/Gigacage.cpp:

(Gigacage::bmalloc::freezeGigacageConfig):
(Gigacage::bmalloc::unfreezeGigacageConfig):
(Gigacage::bmalloc::permanentlyFreezeGigacageConfig):
(Gigacage::bmalloc::UnfreezeGigacageConfigScope::UnfreezeGigacageConfigScope):
(Gigacage::bmalloc::UnfreezeGigacageConfigScope::~UnfreezeGigacageConfigScope):
(Gigacage::ensureGigacage):
(Gigacage::disablePrimitiveGigacage):
(Gigacage::verifyGigacageIsEnabled):
(Gigacage::forbidDisablingPrimitiveGigacage):
(Gigacage::isDisablingPrimitiveGigacageForbidden):
(Gigacage::shouldBeEnabled):
(Gigacage::bmalloc::protectGigacageBasePtrs): Deleted.
(Gigacage::bmalloc::unprotectGigacageBasePtrs): Deleted.
(Gigacage::bmalloc::UnprotectGigacageBasePtrsScope::UnprotectGigacageBasePtrsScope): Deleted.
(Gigacage::bmalloc::UnprotectGigacageBasePtrsScope::~UnprotectGigacageBasePtrsScope): Deleted.
(Gigacage::primitiveGigacageDisabled): Deleted.
(Gigacage::disableDisablingPrimitiveGigacageIfShouldBeEnabled): Deleted.
(Gigacage::isDisablingPrimitiveGigacageDisabled): Deleted.

  • bmalloc/Gigacage.h:

(Gigacage::isPrimitiveGigacagePermanentlyEnabled):
(Gigacage::canPrimitiveGigacageBeDisabled):
(Gigacage::forbidDisablingPrimitiveGigacage):
(Gigacage::disableDisablingPrimitiveGigacageIfShouldBeEnabled): Deleted.

Source/JavaScriptCore:

Just renaming some function names here.

  • assembler/testmasm.cpp:

(JSC::testCagePreservesPACFailureBit):

  • jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::cageConditionally):

  • jsc.cpp:

(jscmain):

Source/WebKit:

Just renaming a function name here.

  • WebProcess/WebProcess.cpp:

Source/WTF:

Just renaming some function names here.

  • wtf/Gigacage.h:

(Gigacage::forbidDisablingPrimitiveGigacage):
(Gigacage::isDisablingPrimitiveGigacageForbidden):
(Gigacage::disableDisablingPrimitiveGigacageIfShouldBeEnabled): Deleted.
(Gigacage::isDisablingPrimitiveGigacageDisabled): Deleted.

9:19 PM Changeset in webkit [249607] by Devin Rousso
  • 5 edits in trunk

Web Inspector: CSS Formatter: the closing } of nested @media aren't indented
https://bugs.webkit.org/show_bug.cgi?id=201560

Reviewed by Joseph Pecoraro.

Source/WebInspectorUI:

  • UserInterface/Workers/Formatter/FormatterContentBuilder.js:

(FormatterContentBuilder.prototype.removeLastNewline):
It's possible that there were multiple newlines in a row, so we should update _startOfLine
to be true if the previous line was empty.

LayoutTests:

  • inspector/formatting/resources/css-tests/media-query.css:
  • inspector/formatting/resources/css-tests/media-query-expected.css:
8:19 PM Changeset in webkit [249606] by bshafiei@apple.com
  • 1 copy in tags/Safari-608.2.11.1.3

Tag Safari-608.2.11.1.3.

7:58 PM Changeset in webkit [249605] by Wenson Hsieh
  • 16 edits
    5 adds in trunk

Incorrect selection rect revealed after pasting images in a contenteditable element
https://bugs.webkit.org/show_bug.cgi?id=201549
<rdar://problem/50956429>

Reviewed by Simon Fraser.

Source/WebCore:

Editor::replaceSelectionWithFragment currently scrolls to reveal the selection after inserting the given
DocumentFragment. However, this scrolling occurs before any inserted images have loaded yet, which causes the
wrong caret rect to be revealed, since all image elements inserted during paste will be empty.

To fix this, we defer revealing the selection after inserting the fragment until after all images that have
been inserted are done loading. While waiting for images to load, if any layers which may be scrolled as a
result of revealing the selection are scrolled, we additionally cancel the deferred selection reveal. See
comments below for more detail.

Tests: editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll.html

editing/pasteboard/reveal-selection-after-pasting-images.html
PasteImage.RevealSelectionAfterPastingImage

  • editing/Editing.cpp:

(WebCore::visibleImageElementsInRangeWithNonLoadedImages):

Add a new helper to iterate through a range and collect all image elements in that range, that contain cached
images that have not finished loading yet.

  • editing/Editing.h:
  • editing/Editor.cpp:

(WebCore::Editor::replaceSelectionWithFragment):

Instead of always immediately revealing the selection after applying the ReplaceSelectionCommand, collect the
image elements that were just inserted, and avoid immediately revealing the selection if any of these images
have non-null cached images, but are not loaded yet. Instead, hold on to these images in a set, remove them once
they finish loading using the new method below, and once all images are removed, reveal the selection.

(WebCore::Editor::revealSelectionIfNeededAfterLoadingImageForElement):
(WebCore::Editor::renderLayerDidScroll):

Called whenever a scrollable RenderLayer is scrolled (or in the case of FrameView, the root layer). In the case
where Editor is waiting to reveal the selection, we check to see if the scrolled layer is an ancestor of the
layer enclosing the start of the selection.

(WebCore::Editor::respondToChangedSelection):

If the selection changes between pasting and waiting for pasted images to load, just cancel waiting to reveal
the selection after pasting.

  • editing/Editor.h:
  • editing/ReplaceSelectionCommand.cpp:

(WebCore::ReplaceSelectionCommand::insertedContentRange const):

Add a helper method to grab the Range of content inserted after applying the command.

  • editing/ReplaceSelectionCommand.h:
  • page/FrameView.cpp:

(WebCore::FrameView::scrollPositionChanged):

  • page/FrameView.h:
  • page/Page.cpp:

(WebCore::Page::didFinishLoadingImageForElement):

Notify Editor after an image finishes loading.

  • rendering/RenderLayer.cpp:

(WebCore::RenderLayer::scrollTo):

Source/WebKit:

Tweak some existing logic to use the new visibleImageElementsInRangeWithNonLoadedImages helper function. See
WebCore for more details.

  • WebProcess/WebPage/ios/WebPageIOS.mm:

(WebKit::WebPage::didConcludeEditDrag):

Tools:

Add an API test to exercise the scenario where we scroll to reveal the selection after pasting an image that was
directly written to the pasteboard.

  • TestWebKitAPI/Tests/WebKitCocoa/PasteImage.mm:

LayoutTests:

Add a couple of new layout tests.

  • editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll-expected.txt: Added.
  • editing/pasteboard/do-not-reveal-selection-after-programmatic-scroll.html: Added.

This test verifies that we don't try to scroll to reveal the caret after pasting, if the scroll position was
changed before the images finished loading.

  • editing/pasteboard/reveal-selection-after-pasting-images-expected.txt: Added.
  • editing/pasteboard/reveal-selection-after-pasting-images.html: Added.
  • platform/ios/editing/pasteboard/reveal-selection-after-pasting-images-expected.txt: Added.

This test verifies that we reveal the caret after loading multiple pasted images in a selection, and dispatch a
scroll event in the process.

7:39 PM Changeset in webkit [249604] by Brent Fulgham
  • 5 edits in trunk/Source/WebCore

PathDirect2D Cleanups
https://bugs.webkit.org/show_bug.cgi?id=201534

Reviewed by Fujii Hironori.

  1. Add a COMPtr<> move operator.
  2. Reduce copying ID2D1Geometry objects.
  3. Protect against using the ID2D1PathGeometry after the ID2D1GeometrySink's Close operation is used.
  4. Add a debug-only 'refCount' function to COMPtr to help track down memory errors.
  • platform/graphics/Path.h:

(WebCore::Path::activePath const): Deleted.

  • platform/graphics/win/Direct2DOperations.cpp:

(WebCore::Direct2D::fillPath):
(WebCore::Direct2D::drawPath):

  • platform/graphics/win/PathDirect2D.cpp:
  • platform/win/COMPtr.h:

(COMPtr::COMPtr):
(=):
(refCount):

6:53 PM Changeset in webkit [249603] by Chris Dumez
  • 11 edits
    1 copy
    1 add in trunk/Source

Move the ResourceLoadObserver logic to WebKit2
https://bugs.webkit.org/show_bug.cgi?id=201517

Reviewed by Brent Fulgham.

Move the ResourceLoadObserver logic to WebKit2 since it is not used by WebKit1. This allows us to simplify
code.

In a follow-up patch, I will simplify the code even further by leveraging the fact that a WebContent process
is always associated with a single WebsiteDataStore / sessionID:

  • No need for a HashMap of sessionIDs
  • No need to even allocate the ResourceLoadObserver if the WebProcess is associated with an ephemeral session.

Source/WebCore:

  • dom/Document.h:
  • loader/ResourceLoadObserver.cpp:

(WebCore::sharedObserver):
(WebCore::ResourceLoadObserver::setShared):
(WebCore::ResourceLoadObserver::shared):

  • loader/ResourceLoadObserver.h:

(WebCore::ResourceLoadObserver::~ResourceLoadObserver):
(WebCore::ResourceLoadObserver::logSubresourceLoading):
(WebCore::ResourceLoadObserver::logWebSocketLoading):
(WebCore::ResourceLoadObserver::logUserInteractionWithReducedTimeResolution):
(WebCore::ResourceLoadObserver::logFontLoad):
(WebCore::ResourceLoadObserver::logCanvasRead):
(WebCore::ResourceLoadObserver::logCanvasWriteOrMeasure):
(WebCore::ResourceLoadObserver::logNavigatorAPIAccessed):
(WebCore::ResourceLoadObserver::logScreenAPIAccessed):
(WebCore::ResourceLoadObserver::statisticsForURL):
(WebCore::ResourceLoadObserver::updateCentralStatisticsStore):
(WebCore::ResourceLoadObserver::clearState):
(WebCore::ResourceLoadObserver::setShouldLogUserInteraction):

  • page/DeprecatedGlobalSettings.h:

Source/WebKit:

  • NetworkProcess/NetworkConnectionToWebProcess.cpp:

(WebKit::NetworkConnectionToWebProcess::resourceLoadStatisticsUpdated):

  • NetworkProcess/NetworkConnectionToWebProcess.h:
  • Sources.txt:
  • WebKit.xcodeproj/project.pbxproj:
  • WebProcess/WebCoreSupport/WebResourceLoadObserver.cpp: Copied from Source/WebCore/loader/ResourceLoadObserver.cpp.

(WebKit::is3xxRedirect):
(WebKit::WebResourceLoadObserver::WebResourceLoadObserver):
(WebKit::WebResourceLoadObserver::shouldLog const):
(WebKit::WebResourceLoadObserver::requestStorageAccessUnderOpener):
(WebKit::WebResourceLoadObserver::ensureResourceStatisticsForRegistrableDomain):
(WebKit::WebResourceLoadObserver::scheduleNotificationIfNeeded):
(WebKit::WebResourceLoadObserver::updateCentralStatisticsStore):
(WebKit::WebResourceLoadObserver::statisticsForURL):
(WebKit::WebResourceLoadObserver::takeStatistics):
(WebKit::WebResourceLoadObserver::clearState):
(WebKit::WebResourceLoadObserver::nonNullOwnerURL const):
(WebKit::WebResourceLoadObserver::logFontLoad):
(WebKit::WebResourceLoadObserver::logCanvasRead):
(WebKit::WebResourceLoadObserver::logCanvasWriteOrMeasure):
(WebKit::WebResourceLoadObserver::logNavigatorAPIAccessed):
(WebKit::WebResourceLoadObserver::logScreenAPIAccessed):
(WebKit::WebResourceLoadObserver::logSubresourceLoading):
(WebKit::WebResourceLoadObserver::logWebSocketLoading):
(WebKit::WebResourceLoadObserver::logUserInteractionWithReducedTimeResolution):

  • WebProcess/WebCoreSupport/WebResourceLoadObserver.h: Added.
  • WebProcess/WebProcess.cpp:
5:42 PM Changeset in webkit [249602] by Matt Lewis
  • 4 edits
    2 deletes in trunk/Tools

Unreviewed, rolling out r249582.

This caused results.html fail to be created on internal
testers.

Reverted changeset:

"run-webkit-test: Allow results to be uploaded without scm
checkout"
https://bugs.webkit.org/show_bug.cgi?id=200787
https://trac.webkit.org/changeset/249582

5:24 PM Changeset in webkit [249601] by Justin Fan
  • 117 edits
    8 moves in trunk

[WebGPU] Update several interface and enum names to match spec
https://bugs.webkit.org/show_bug.cgi?id=201564

Reviewed by Dean Jackson.

Rename GPUShaderStageBit -> GPUShaderStage.
Rename GPUColorWriteBits -> GPUColorWrite.
Rename TRANSFER_SRC/DST -> COPY_SRC/DST.
Source/WebCore:

Rename GPUPipelineStageDescriptor -> GPUProgrammableStageDescriptor.

Existing tests trivially updated to match.

  • CMakeLists.txt:
  • DerivedSources-input.xcfilelist:
  • DerivedSources-output.xcfilelist:
  • DerivedSources.make:
  • Modules/webgpu/GPUBindGroupLayoutBinding.h:
  • Modules/webgpu/GPUBufferUsage.idl:
  • Modules/webgpu/GPUColorStateDescriptor.idl:
  • Modules/webgpu/GPUColorWrite.idl: Renamed from Source/WebCore/Modules/webgpu/GPUColorWriteBits.idl.
  • Modules/webgpu/GPUShaderStage.h: Renamed from Source/WebCore/Modules/webgpu/GPUShaderStageBit.h.
  • Modules/webgpu/GPUShaderStage.idl: Renamed from Source/WebCore/Modules/webgpu/GPUShaderStageBit.idl.
  • Modules/webgpu/GPUTextureUsage.idl:
  • Modules/webgpu/WebGPUComputePipelineDescriptor.cpp:

(WebCore::WebGPUComputePipelineDescriptor::tryCreateGPUComputePipelineDescriptor const):

  • Modules/webgpu/WebGPUComputePipelineDescriptor.h:
  • Modules/webgpu/WebGPUComputePipelineDescriptor.idl:
  • Modules/webgpu/WebGPUDevice.cpp:
  • Modules/webgpu/WebGPUProgrammableStageDescriptor.cpp: Renamed from Source/WebCore/Modules/webgpu/WebGPUPipelineStageDescriptor.cpp.

(WebCore::WebGPUProgrammableStageDescriptor::tryCreateGPUProgrammableStageDescriptor const):

  • Modules/webgpu/WebGPUProgrammableStageDescriptor.h: Renamed from Source/WebCore/Modules/webgpu/WebGPUPipelineStageDescriptor.h.
  • Modules/webgpu/WebGPUProgrammableStageDescriptor.idl: Renamed from Source/WebCore/Modules/webgpu/WebGPUPipelineStageDescriptor.idl.
  • Modules/webgpu/WebGPURenderPipelineDescriptor.cpp:

(WebCore::WebGPURenderPipelineDescriptor::tryCreateGPURenderPipelineDescriptor const):

  • Modules/webgpu/WebGPURenderPipelineDescriptor.h:
  • Modules/webgpu/WebGPURenderPipelineDescriptor.idl:
  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/WebCoreBuiltinNames.h:
  • platform/graphics/gpu/GPUBuffer.h:

(WebCore::GPUBuffer::isCopySource const):
(WebCore::GPUBuffer::isCopyDestination const):
(WebCore::GPUBuffer::isTransferSource const): Deleted.
(WebCore::GPUBuffer::isTransferDestination const): Deleted.

  • platform/graphics/gpu/GPUBufferUsage.h:
  • platform/graphics/gpu/GPUColorStateDescriptor.h:
  • platform/graphics/gpu/GPUColorWrite.h: Renamed from Source/WebCore/platform/graphics/gpu/GPUColorWriteBits.h.
  • platform/graphics/gpu/GPUComputePipelineDescriptor.h:

(WebCore::GPUComputePipelineDescriptor::GPUComputePipelineDescriptor):

  • platform/graphics/gpu/GPUPipelineDescriptorBase.h:
  • platform/graphics/gpu/GPUProgrammableStageDescriptor.h: Renamed from Source/WebCore/platform/graphics/gpu/GPUPipelineStageDescriptor.h.

(WebCore::GPUProgrammableStageDescriptor::GPUProgrammableStageDescriptor):

  • platform/graphics/gpu/GPURenderPipelineDescriptor.h:

(WebCore::GPURenderPipelineDescriptor::GPURenderPipelineDescriptor):

  • platform/graphics/gpu/GPUTexture.h:

(WebCore::GPUTexture::isCopySource const):
(WebCore::GPUTexture::isCopyDestination const):
(WebCore::GPUTexture::isReadOnly const):
(WebCore::GPUTexture::isTransferSource const): Deleted.
(WebCore::GPUTexture::isTransferDestination const): Deleted.

  • platform/graphics/gpu/GPUTextureUsage.h:
  • platform/graphics/gpu/cocoa/GPUBindGroupLayoutMetal.mm:

(WebCore::GPUBindGroupLayout::tryCreate):

  • platform/graphics/gpu/cocoa/GPUBindGroupMetal.mm:

(WebCore::GPUBindGroup::tryCreate):

  • platform/graphics/gpu/cocoa/GPUBufferMetal.mm:
  • platform/graphics/gpu/cocoa/GPUCommandBufferMetal.mm:

(WebCore::GPUCommandBuffer::copyBufferToBuffer):
(WebCore::GPUCommandBuffer::copyBufferToTexture):
(WebCore::GPUCommandBuffer::copyTextureToBuffer):
(WebCore::GPUCommandBuffer::copyTextureToTexture):

  • platform/graphics/gpu/cocoa/GPUComputePipelineMetal.mm:

(WebCore::trySetFunctions):

  • platform/graphics/gpu/cocoa/GPUPipelineMetalConvertLayout.cpp:

(WebCore::convertShaderStageFlags):

  • platform/graphics/gpu/cocoa/GPURenderPipelineMetal.mm:

(WebCore::mtlColorWriteMaskForGPUColorWriteFlags):
(WebCore::trySetFunctions):

  • platform/graphics/gpu/cocoa/GPUTextureMetal.mm:

(WebCore::mtlTextureUsageForGPUTextureUsageFlags):

LayoutTests:

  • webgpu/bind-groups.html:
  • webgpu/blend-color-triangle-strip.html:
  • webgpu/blit-commands.html:
  • webgpu/buffer-errors.html:
  • webgpu/color-write-mask-triangle-strip.html:
  • webgpu/pipeline-layouts.html:
  • webgpu/render-pipelines.html:
  • webgpu/texture-triangle-strip.html:
  • webgpu/textures-textureviews.html:
  • webgpu/whlsl/arbitrary-vertex-attribute-locations.html:
  • webgpu/whlsl/buffer-fragment.html:
  • webgpu/whlsl/buffer-length.html:
  • webgpu/whlsl/buffer-vertex.html:
  • webgpu/whlsl/checker-should-set-type-of-read-modify-write-variables.html:
  • webgpu/whlsl/compute.html:
  • webgpu/whlsl/dereference-pointer-should-type-check.html:
  • webgpu/whlsl/device-proper-type-checker.html:
  • webgpu/whlsl/do-while-loop-break.html:
  • webgpu/whlsl/do-while-loop-continue.html:
  • webgpu/whlsl/do-while-loop.html:
  • webgpu/whlsl/dont-crash-parsing-enum.html:
  • webgpu/whlsl/dot-expressions.html:
  • webgpu/whlsl/duplicate-types-should-not-produce-duplicate-ctors.html:
  • webgpu/whlsl/ensure-proper-variable-lifetime-2.html:
  • webgpu/whlsl/ensure-proper-variable-lifetime-3.html:
  • webgpu/whlsl/ensure-proper-variable-lifetime.html:
  • webgpu/whlsl/huge-array.html:
  • webgpu/whlsl/js/test-harness.js:

(Harness.prototype.async.callTypedFunction):
(Harness.prototype._clearResults):
(Harness.prototype._setUpArguments):

  • webgpu/whlsl/loops-break.html:
  • webgpu/whlsl/loops-continue.html:
  • webgpu/whlsl/loops.html:
  • webgpu/whlsl/make-array-reference.html:
  • webgpu/whlsl/matrix-2.html:
  • webgpu/whlsl/matrix-alignment.html:
  • webgpu/whlsl/matrix-memory-layout.html:
  • webgpu/whlsl/matrix.html:
  • webgpu/whlsl/nested-dot-expression-rvalue.html:
  • webgpu/whlsl/nested-loop.html:
  • webgpu/whlsl/oob-access.html:
  • webgpu/whlsl/read-modify-write-high-zombies.html:
  • webgpu/whlsl/return-local-variable.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-10.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-11.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-12.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-13.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-14.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-15.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-16.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-17.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-18.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-19.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-2.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-20.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-21.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-22.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-23.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-24.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-25.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-26.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-27.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-3.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-4.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-5.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-6.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-7.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-8.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules-9.html:
  • webgpu/whlsl/separate-shader-modules/separate-shader-modules.html:
  • webgpu/whlsl/simple-arrays.html:
  • webgpu/whlsl/store-to-property-updates-properly.html:
  • webgpu/whlsl/textures-load.html:
  • webgpu/whlsl/textures-sample-bias.html:
  • webgpu/whlsl/textures-sample-grad.html:
  • webgpu/whlsl/textures-sample-level.html:
  • webgpu/whlsl/textures-sample.html:
  • webgpu/whlsl/two-dimensional-array.html:
  • webgpu/whlsl/while-loop-break.html:
  • webgpu/whlsl/while-loop-continue.html:
  • webgpu/whlsl/whlsl.html:
  • webgpu/whlsl/zero-initialize-values-2.html:
  • webgpu/whlsl/zero-initialize-values.html:
5:08 PM Changeset in webkit [249600] by jiewen_tan@apple.com
  • 4 edits in trunk/Source/WebKit

[WebAuthn] Make WebAuthn default off and let clients turn it on at will
https://bugs.webkit.org/show_bug.cgi?id=201439
<rdar://problem/54998154>

Reviewed by Youenn Fablet.

This patch makes WebAuthn default off such that clients that have the right entitlements
could turn it on and we don't risk at turning on a Web API that does nothing by default.

This patch doesn't add any SPI to turn the feature on as it is currently doable via

  • [WKPreferencesPrivate _setEnabled:forExperimentalFeature:].
  • Shared/WebPreferences.yaml:
  • Shared/WebPreferencesDefaultValues.cpp:

(WebKit::defaultWebAuthenticationEnabled): Deleted.

  • Shared/WebPreferencesDefaultValues.h:
4:09 PM Changeset in webkit [249599] by mark.lam@apple.com
  • 2 edits in trunk/JSTests

Gardening: speculative test fix to green bots.
https://bugs.webkit.org/show_bug.cgi?id=201529
<rdar://problem/53935772>

Not reviewed.

  • stress/test-out-of-memory.js:
3:59 PM Changeset in webkit [249598] by aakash_jain@apple.com
  • 2 edits in trunk/Tools

[ews-app] Status bubble should only display important messages in pop-over - part 2
https://bugs.webkit.org/show_bug.cgi?id=201557

Reviewed by Jonathan Bedard.

  • BuildSlaveSupport/ews-app/ews/views/statusbubble.py:

(StatusBubble): Add few more strings to STEPS_TO_HIDE.

3:49 PM Changeset in webkit [249597] by Ross Kirsling
  • 8 edits in trunk

Math.round() produces wrong result for value prior to 0.5
https://bugs.webkit.org/show_bug.cgi?id=185115

Reviewed by Saam Barati.

JSTests:

  • stress/math-round-basics.js:

Add positive/negative test cases.

  • test262/expectations.yaml:

Mark test passing.

Source/JavaScriptCore:

Our Math.round implementation goes in the wrong direction for double values like 0.49999999999999994.
This requires just a subtle adjustment for three of our four versions; only baseline JIT needed a full rewrite.

Specifically:

  • While 0.49999999999999994 is representable, 1 - 0.49999999999999994 is not (it turns into 0.5), so taking the difference between ceil(value) and value` is problematic.
  • The baseline implementation was doing floor(x + 0.5) for positive doubles and slowpathing negative ones (by falling back to jsRound). This patch gives baseline a legitimate implementation too.
  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileArithRounding):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileArithRound):

  • jit/ThunkGenerators.cpp:

(JSC::roundThunkGenerator):

  • runtime/MathCommon.cpp:
3:34 PM Changeset in webkit [249596] by Devin Rousso
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: CSS Formatter: links to CSS resources don't map to the right line after pretty printing if the line is after a multiline comment
https://bugs.webkit.org/show_bug.cgi?id=201559

Reviewed by Joseph Pecoraro.

Newlines in multiline comments weren't added as line endings, so they weren't considered
when a source code location updated after pretty printing.

  • UserInterface/Workers/Formatter/CSSFormatter.js:

(CSSFormatter.prototype._format):

3:23 PM Changeset in webkit [249595] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

REGRESSION: http/tests/resourceLoadStatistics/do-not-capture-statistics-for-simple-top-navigations.html is frequently timing out on iOS EWS bots
https://bugs.webkit.org/show_bug.cgi?id=201550

Unreviewed test gardening.

  • platform/ios-wk2/TestExpectations: Mark the test as flaky to speed up EWS while we investigate the root cause.
3:15 PM Changeset in webkit [249594] by commit-queue@webkit.org
  • 8 edits
    2 adds in trunk

REGRESSION (r249367): m_decodingPromises grows indefinitely until ImageLoader destruction
https://bugs.webkit.org/show_bug.cgi?id=201402

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2019-09-06
Reviewed by Youenn Fablet and Daniel Bates.

Source/WebCore:

Add the static functions resolvePromises() and rejectPromises(). These
functions take an lvalue reference to a Vector of promises. Inside them,
the lvalue reference argument are exchanged with an empty Vector of
promises then the promises are processed. This clears m_decodingPromises
and fixes the leak.

Add an internal API which returns the count of the pending promises of
an HTMLImageElement. This internal API will be used in the attached test.

Test: fast/images/decode-resolve-reject-no-leak.html

  • html/HTMLImageElement.h:

(WebCore::HTMLImageElement::pendingDecodePromisesCountForTesting const):

  • loader/ImageLoader.cpp:

(WebCore::resolvePromises):
ImageLoader::decode() calls BitmapImage::decode() and moves m_decodingPromises
in capture. When decoding finishes, this function is called to resolve the
promises. But ImageLoader might get deleted before the image decoding
finishes. So this function has to be static.

(WebCore::rejectPromises):
(WebCore::ImageLoader::resolveDecodePromises):
(WebCore::ImageLoader::rejectDecodePromises):
(WebCore::ImageLoader::notifyFinished):
(WebCore::ImageLoader::decode):
(WebCore::resolveDecodePromises): Deleted.
(WebCore::rejectDecodePromises): Deleted.

  • loader/ImageLoader.h:

(WebCore::ImageLoader::pendingDecodePromisesCountForTesting const):

  • testing/Internals.cpp:

(WebCore::Internals::imagePendingDecodePromisesCountForTesting):

  • testing/Internals.h:
  • testing/Internals.idl:

LayoutTests:

  • fast/images/decode-resolve-reject-no-leak-expected.txt: Added.
  • fast/images/decode-resolve-reject-no-leak.html: Added.
2:57 PM Changeset in webkit [249593] by commit-queue@webkit.org
  • 3 edits
    4 adds in trunk

Prefer null namespace 'href' over 'xlink:href' on SVG elements
https://bugs.webkit.org/show_bug.cgi?id=195802

Patch by Said Abou-Hallawa <sabouhallawa@apple.com> on 2019-09-06
Reviewed by Youenn Fablet.

Source/WebCore:

-- When the attribute "href" is set, its value is always reflected in

SVGURIReference::m_href.

-- When the attribute "href" is removed, the value of the attribute

"xlink:href" is always reflected in SVGURIReference::m_href.

-- When the "xlink:href" is set, its value is reflected in

SVGURIReference::m_href only if the attribute "href" is not set.

Tests: svg/custom/href-xlink-href-gradient-element-expected.svg

svg/custom/href-xlink-href-gradient-element.svg
svg/custom/href-xlink-href-use-element-expected.svg
svg/custom/href-xlink-href-use-element.svg

  • svg/SVGURIReference.cpp:

(WebCore::SVGURIReference::parseAttribute):

LayoutTests:

  • svg/custom/href-xlink-href-gradient-element-expected.svg: Added.
  • svg/custom/href-xlink-href-gradient-element.svg: Added.
  • svg/custom/href-xlink-href-use-element-expected.svg: Added.
  • svg/custom/href-xlink-href-use-element.svg: Added.
2:19 PM Changeset in webkit [249592] by Joseph Pecoraro
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Better position for Sources tab when enabling the experimental setting
https://bugs.webkit.org/show_bug.cgi?id=201539

Reviewed by Devin Rousso.

  • UserInterface/Views/SettingsTabContentView.js:

Place the Sources tab where the Debugger tab was.

2:08 PM Changeset in webkit [249591] by Alan Bujtas
  • 2 edits in trunk/Source/WebCore

[LFC][TFC] Set computed row width.
https://bugs.webkit.org/show_bug.cgi?id=201533
<rdar://problem/55098828>

Reviewed by Antti Koivisto.

Decouple cell box layout/display box setup and set the computed row width.

  • layout/Verification.cpp:

(WebCore::Layout::LayoutState::verifyAndOutputMismatchingLayoutTree const):

  • layout/tableformatting/TableFormattingContext.cpp:

(WebCore::Layout::TableFormattingContext::layout):

1:51 PM Changeset in webkit [249590] by Alan Bujtas
  • 2 edits in trunk/Source/WebCore

[LFC] A formatting context root is always a containing block for relative: static boxes.
https://bugs.webkit.org/show_bug.cgi?id=201554
<rdar://problem/55123295>

Reviewed by Antti Koivisto.

"For other elements, if the element's position is 'relative' or 'static', the containing block is formed by the content
edge of the nearest ancestor box that is a block container or which establishes a formatting context."

  • layout/layouttree/LayoutBox.cpp:

(WebCore::Layout::Box::containingBlock const):

  • page/FrameViewLayoutContext.cpp:

(WebCore::layoutUsingFormattingContext):

1:42 PM Changeset in webkit [249589] by achristensen@apple.com
  • 26 edits
    2 adds
    1 delete in trunk

Deprecate all WKCookieManagerRef functions
https://bugs.webkit.org/show_bug.cgi?id=201473

Reviewed by Chris Dumez.

Source/WebKit:

Their use was removed in rdar://problem/55039275
Created replacement C API that more closely resembles the public ObjC API for WebKitTestRunner.

  • Shared/API/c/WKBase.h:
  • Sources.txt:
  • UIProcess/API/APIHTTPCookieStore.cpp:

(API::HTTPCookieStore::deleteAllCookies):
(API::HTTPCookieStore::setHTTPCookieAcceptPolicy):
(API::HTTPCookieStore::deleteCookiesInDefaultUIProcessCookieStore):
(API::HTTPCookieStore::setHTTPCookieAcceptPolicyInDefaultUIProcessCookieStore):

  • UIProcess/API/APIHTTPCookieStore.h:
  • UIProcess/API/C/WKAPICast.h:
  • UIProcess/API/C/WKContext.cpp:

(WKContextSetStorageAccessAPIEnabled):

  • UIProcess/API/C/WKContext.h:
  • UIProcess/API/C/WKCookieManager.cpp:

(WKCookieManagerGetTypeID):
(WKCookieManagerSetClient):
(WKCookieManagerGetHostnamesWithCookies):
(WKCookieManagerDeleteCookiesForHostname):
(WKCookieManagerDeleteAllCookies):
(WKCookieManagerDeleteAllCookiesModifiedAfterDate):
(WKCookieManagerSetHTTPCookieAcceptPolicy):
(WKCookieManagerGetHTTPCookieAcceptPolicy):
(WKCookieManagerSetStorageAccessAPIEnabled):
(WKCookieManagerStartObservingCookieChanges):
(WKCookieManagerStopObservingCookieChanges):

  • UIProcess/API/C/WKCookieManager.h:
  • UIProcess/API/C/WKHTTPCookieStoreRef.cpp: Added.

(WKHTTPCookieStoreGetTypeID):
(WKHTTPCookieStoreDeleteAllCookies):
(WKHTTPCookieStoreSetHTTPCookieAcceptPolicy):

  • UIProcess/API/C/WKHTTPCookieStoreRef.h: Added.
  • UIProcess/API/C/WKWebsiteDataStoreRef.cpp:

(WKWebsiteDataStoreGetHTTPCookieStore):

  • UIProcess/API/C/WKWebsiteDataStoreRef.h:
  • UIProcess/API/Cocoa/APIHTTPCookieStoreCocoa.mm:

(API::HTTPCookieStore::deleteCookiesInDefaultUIProcessCookieStore):
(API::toNSHTTPCookieAcceptPolicy):
(API::HTTPCookieStore::setHTTPCookieAcceptPolicyInDefaultUIProcessCookieStore):

  • UIProcess/WebCookieManagerProxy.cpp:

(WebKit::WebCookieManagerProxy::setStorageAccessAPIEnabled): Deleted.

  • UIProcess/WebCookieManagerProxy.h:
  • WebKit.xcodeproj/project.pbxproj:

Tools:

Adopt replacement C API in WebKitTestRunner.

  • TestWebKitAPI/CMakeLists.txt:
  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKit/CookieManager.cpp: Removed, because the functions it tests were removed.
  • WebKitTestRunner/TestController.cpp:

(WTR::TestController::resetPreferencesToConsistentValues):
(WTR::TestController::didReceiveSynchronousMessageFromInjectedBundle):

  • WebKitTestRunner/TestInvocation.cpp:

(WTR::TestInvocation::invoke):
(WTR::TestInvocation::didReceiveSynchronousMessageFromInjectedBundle):

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::initializeWebViewConfiguration):

1:23 PM Changeset in webkit [249588] by mark.lam@apple.com
  • 4 edits in trunk/Source/bmalloc

Use FailureAction everywhere in bmalloc instead of a crashOnFailure boolean.
https://bugs.webkit.org/show_bug.cgi?id=201553

Reviewed by Yusuke Suzuki.

  • bmalloc/Cache.cpp:

(bmalloc::Cache::tryAllocateSlowCaseNullCache):
(bmalloc::Cache::allocateSlowCaseNullCache):
(bmalloc::Cache::tryReallocateSlowCaseNullCache):
(bmalloc::Cache::reallocateSlowCaseNullCache):

  • bmalloc/DebugHeap.cpp:

(bmalloc::DebugHeap::malloc):
(bmalloc::DebugHeap::memalign):
(bmalloc::DebugHeap::realloc):

  • bmalloc/DebugHeap.h:
1:08 PM Changeset in webkit [249587] by mark.lam@apple.com
  • 1 edit
    1 move in trunk/JSTests

Move web-assembly-constructors-should-not-override-global-object-property.js below JSTests/wasm/stress.
https://bugs.webkit.org/show_bug.cgi?id=201551

Reviewed by Tadeu Zagallo.

Ports that don't support WASM will always fail this test if it stays in JSTests/stress.

  • stress/web-assembly-constructors-should-not-override-global-object-property.js: Removed.
  • wasm/stress/web-assembly-constructors-should-not-override-global-object-property.js: Copied from JSTests/stress/web-assembly-constructors-should-not-override-global-object-property.js.
12:11 PM Changeset in webkit [249586] by Joseph Pecoraro
  • 5 edits
    2 copies
    2 moves
    17 adds
    4 deletes in trunk

Tail Deleted Frames shown in Web Inspector are sometimes incorrect (Shadow Chicken)
https://bugs.webkit.org/show_bug.cgi?id=201366

Reviewed by Saam Barati.

Source/JavaScriptCore:

It is possible for the log buffer to be full right as someone is trying to
log a function prologue. In such a case the machine stack has already been
updated to include the new JavaScript call frame, but the prologue packet
cannot be included in the update because the log is full. This would mean
that the update fails to rationalize the machine stack with the shadow
log / stack. Namely, the current JavaScript call frame is unable to
find a matching prologue (the one we are holding to include after the update)
and inserts a questionable value into the stack; and in the process
missing and removing real potential tail calls.

For example:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function start() { return first(); }

If the the log fills up just as we are entering b then we may have a list
full log of packets looking like:

Shadow Log:

...
{ prologue-packet: entering start ... }
{ prologue-packet: entering first ... }
{ tail-packet: leaving first with a tail call }

Incoming Packet:

{ prologue-packet: entering second ... }

Current JS Stack:

second
start

Since the Current JavaScript stack already has second, if we process the
log without the prologue for second then we push a confused entry on the
shadow stack and clear the log such that we eventually lose the tail-call
information for first to second.

This patch solves this issue by providing enough extra space in the log
to always process the incoming packet when that forces an update. This way
clients can continue to behave exactly as they are.

--

We also document a corner case in some circumstances where the shadow
log may currently be insufficient to know how to reconcile:

For example:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function doNothingTail() { return Math.random() }
function start() {

for (i=0;i<1000;++i) doNothingTail();
return first();

}

In this case the ShadowChicken log may be processed multiple times due
to the many calls to doNothingTail / Math.random(). When calling the
Native function no prologue packet is emitted, so it is unclear that we
temporarly go deeper and come back out on the stack, so the log appears
to have lots of doNothingTail calls reusing the same frame:

Shadow Log:

...
, [123] {callee = 0x72a21aee0, frame = 0x7ffeef897270, callerFrame = 0x7ffeef8972e0, name = start}
, [124] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [125] tail-packet:{frame = 0x7ffeef8971f0}
, [126] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [127] tail-packet:{frame = 0x7ffeef8971f0}
...
, [140] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [141] tail-packet:{frame = 0x7ffeef8971f0}
, [142] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [143] tail-packet:{frame = 0x7ffeef8971f0}
, [144] {callee = 0x72a21aeb0, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = first}
, [145] tail-packet:{frame = 0x7ffeef8971f0}
, [146] {callee = 0x72a21ae80, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = second}
...

This log would seem to be indistinguishable from real tail recursion, such as:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function doNothingTail(n) {

return n ? doNothingTail(n-1) : first();

}
function start() {

return doNothingTail(1000);

}

Likewise there are more cases where the shadow log appears to be ambiguous with determining
the appropriate parent call frame with intermediate function calls. In practice this may
not be too problematic, as this is a best effort reconstruction of tail deleted frames.
It seems likely we would only show additional frames that did in fact happen serially
between JavaScript call frames, but may not actually be the proper parent frames
heirachy in the stack.

  • interpreter/ShadowChicken.cpp:

(JSC::ShadowChicken::Packet::dump const):
(JSC::ShadowChicken::Frame::dump const):
(JSC::ShadowChicken::dump const):
Improved debugging output. Especially for functions.

(JSC::ShadowChicken::ShadowChicken):
Make space in the log for 1 additional packet to process when we slow log.

(JSC::ShadowChicken::log):
Include this packet in our update.

(JSC::ShadowChicken::update):
Address an edge case where we can eliminate tail-deleted frames that don't make sense.

LayoutTests:

  • inspector/debugger/tail-deleted-frames-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-from-vm-entry-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-from-vm-entry.html: Removed.
  • inspector/debugger/tail-deleted-frames-this-value-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-this-value.html: Removed.
  • inspector/debugger/tail-deleted-frames.html: Removed.

Remove legacy tests that are difficult to read.

  • inspector/debugger/tail-deleted-frames/resources/stack-trace-utilities.js: Added.

(TestPage.registerInitializer.window.getAsyncStackTrace):
(TestPage.registerInitializer.async.logThisObject):
(TestPage.registerInitializer.async.logScope):
(TestPage.registerInitializer.async.logCallFrame):
(TestPage.registerInitializer):

  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-frames.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-native-tail-deleted-calls.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-tail-deleted-frames.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-scopes.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-this-value.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-vm-entry.js: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-frames-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-frames.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-tail-deleted-frames-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-tail-deleted-frames.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-scopes-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-scopes.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-this-value-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-this-value.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-vm-entry-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-vm-entry.html: Added.

Include modern tests that are easier to read.

  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-native-tail-deleted-calls-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-native-tail-deleted-calls.html: Added.

Include a test that is known to produce bad output, since we have reproductive steps.

  • platform/mac/TestExpectations:

Updated pathes.

12:03 PM Changeset in webkit [249585] by commit-queue@webkit.org
  • 3 edits
    2 adds in trunk

REGRESSION(iOS 12): Keyboard dismissal leaves WKWebView viewport-fit=cover content offscreen
https://bugs.webkit.org/show_bug.cgi?id=192564

Patch by Darryl Pogue <darryl@dpogue.ca> on 2019-09-06
Reviewed by Tim Horton.

Source/WebKit:

The new keyboard height was being read into a variable after the scrollview had already been
adjusted. This worked when the keyboard was opening, because the scrollview is adjusted with
the correct value a second time when the keyboard animation has finished. However, when the
keyboard is closed, the adjustment was being made before the correct height was stored,
leading to the scrollview being positioned as if the keyboard were still open.

  • UIProcess/ios/WKScrollView.mm:

(-[WKScrollView _adjustForAutomaticKeyboardInfo:animated:lastAdjustment:]):

LayoutTests:

Added tests to verify scrolling properly adapts to the keyboard insets on iOS.

  • fast/scrolling/ios/scrollview-adjusts-to-keyboard-insets-expected.txt: Added.
  • fast/scrolling/ios/scrollview-adjusts-to-keyboard-insets.html: Added.
11:46 AM Changeset in webkit [249584] by timothy_horton@apple.com
  • 6 edits
    1 add in trunk

Marking up a note on iOS results in a PDF with no contents
https://bugs.webkit.org/show_bug.cgi?id=201530
<rdar://problem/53686019>

Reviewed by Andy Estes.

Source/WebKit:

  • Platform/IPC/Connection.cpp:

(IPC::Connection::Connection):
(IPC::Connection::waitForMessage):
(IPC::Connection::connectionDidClose):

  • Platform/IPC/Connection.h:

If the main thread is blocked when the Web Content process dies, and
something eventually calls waitForAndDispatchImmediately without
returning control to the main run loop, we will wait for the full timeout,
because a) the code to mark the connection invalid is dispatched
to the main thread, and b) the secondary thread that is informed of
the Web Content process dying did not yet have a "waiting for" message
to mark as interrupted (because it wasn't waiting yet).

Fix this race by adding a bit that is set under the waitForMessage lock
on the secondary thread when the connection is invalidated, identically
to m_shouldWaitForSyncReplies, which solves the same problem for sync
messages.

Read the new bit when we are about to start waiting, and bail if it is set.
It's OK to not read it inside the loop because we are guaranteed to have
waitForMessage set at that point, so the normal interruption bit will work.

  • UIProcess/ios/WKContentView.mm:

(-[WKContentView _processDidExit]):
Reset _isPrintingToPDF; the Web Content process is never going to get
back to us if it crashes.

(-[WKContentView _wk_pageCountForPrintFormatter:]):
Do not bail from starting a printing operation if one is already occurring.
This fixes the original bug, because Markup ends up invalidating the page
count at least one extra time before asking for the printed document.
Instead of maintaining the fragile requirement that you cannot recompute
the page count while printing, just let it happen. In order to make this
work safely, synchronously wait for the previous printed result before
continuing with the next print.

We could do more coalescing here if need be, but calls to -_recalcPageCount
are not high in volume.

Tools:

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

Add some tests for WKWebViewPrintFormatter; specifically that it is
possible to _recalcPageCount twice in quick succession, and that
we don't hang if we start painting the printed content immediately
after a Web Content process crash.

11:28 AM Changeset in webkit [249583] by sihui_liu@apple.com
  • 3 edits in trunk/Source/WebCore

IndexedDB: use SQL COUNT statement for count operation
https://bugs.webkit.org/show_bug.cgi?id=201465

Reviewed by Geoffrey Garen.

We did count operation by moving objectStore/index iterator from begin to end and counting the steps. We can
utilize the SQL COUNT statement, and improve the performance further by caching the statement.

Tested on release build minibrowser. Without the change, the mean time to complete one iteration in
PerformanceTests/IndexedDB/basic/index-count.html and PerformanceTests/IndexedDB/basic/objectStore-count.html is
about 50ms. With the change it is about 2ms.

  • Modules/indexeddb/server/SQLiteIDBBackingStore.cpp:

(WebCore::IDBServer::SQLiteIDBBackingStore::getCount):
(WebCore::IDBServer::SQLiteIDBBackingStore::cachedStatement):
(WebCore::IDBServer::SQLiteIDBBackingStore::closeSQLiteDB):

  • Modules/indexeddb/server/SQLiteIDBBackingStore.h:
11:10 AM Changeset in webkit [249582] by Matt Lewis
  • 4 edits
    2 adds in trunk/Tools

run-webkit-test: Allow results to be uploaded without scm checkout
https://bugs.webkit.org/show_bug.cgi?id=200787

Reviewed by Jonathan Bedard.

  • Scripts/webkitpy/common/checkout/scm/detection.py:

(SCMDetector.detect_scm_system): Added in detection of stub repository.

  • Scripts/webkitpy/common/checkout/scm/stub_repository.py: Added.

(StubRepository):
(StubRepository.init):
(StubRepository._find_parent_path_matching_callback_condition): Checks up the path given to
find the directory containing the checkout_information.json of the stub repository.
(StubRepository.in_working_directory): Required functions to be implemented.
(StubRepository.svn_revision): Required functions to be implemented.
(StubRepository.native_revision): Required functions to be implemented.
(StubRepository.native_branch): Required functions to be implemented.
(StubRepository._decode_json): Decodes checkout_information.json.
(StubRepository.find_checkout_root): Required functions to be implemented.

  • Scripts/webkitpy/common/checkout/scm/stub_repository_unittest.py: Added.

(StubRepositoryTest): Unit test for stub_repository SCM.
(StubRepositoryTest.mock_host_for_stub_repository):
(StubRepositoryTest.test_in_working_directory):
(StubRepositoryTest.test_native_revision):
(StubRepositoryTest.test_native_branch):
(StubRepositoryTest.test_svn_revision):
(StubRepositoryTest.test_find_checkout_root):

  • Scripts/webkitpy/layout_tests/layout_package/json_results_generator.py:

(JSONResultsGenerator._get_svn_revision): Removed unnecessary scm initialization.

  • Scripts/webkitpy/port/base.py:

(Port.commits_for_upload): Removed unnecessary scm initialization.

11:05 AM Changeset in webkit [249581] by Simon Fraser
  • 4 edits
    2 adds in trunk

REGRESSION (iOS 13): If an overflow:hidden with a non-zero scroll position is toggled to overflow:scroll, some other scroll causes its scroll position to get reset
https://bugs.webkit.org/show_bug.cgi?id=201528
rdar://problem/55044885

Reviewed by Frédéric Wang.
Source/WebCore:

If, when an overflow scrolling node is created, the scroller has non-zero scroll
position (for example, via toggling to overflow:hidden, setting scrollTop, then toggling
to overflow:scroll), then on the next update its scroll position will reset back to zero.

The bug was that newly created ScrollingTreeScrollingNodes didn't set m_currentScrollPosition
to the scroll position coming from the state node, so a subsequent update could cause
the 0,0 currentScrollPosition to get applied. If we're making a new node, and there's no
requestedScrollPosition, then initialize m_currentScrollPosition.

Test: scrollingcoordinator/ios/scroller-initial-scroll-position.html

  • page/scrolling/ScrollingTreeScrollingNode.cpp:

(WebCore::ScrollingTreeScrollingNode::commitStateBeforeChildren):
(WebCore::ScrollingTreeScrollingNode::commitStateAfterChildren):

  • page/scrolling/ScrollingTreeScrollingNode.h:

LayoutTests:

  • scrollingcoordinator/ios/scroller-initial-scroll-position-expected.html: Added.
  • scrollingcoordinator/ios/scroller-initial-scroll-position.html: Added.
11:02 AM Changeset in webkit [249580] by achristensen@apple.com
  • 2 edits in trunk/Source/WebKit

Fix unused argument warning from GCC.

  • Platform/IPC/ArgumentCoders.h:

(IPC::TupleDecoder<0>::decode):
There was indeed an unused argument.

10:16 AM Changeset in webkit [249579] by Said Abou-Hallawa
  • 3 edits in trunk/LayoutTests

Unreviewed, fix typo in r249216

Also change the <desc> element to be <text> element so the purpose of
the test is visible once it is opened.

  • svg/custom/href-svg-namespace-animate-target-expected.svg:
  • svg/custom/href-svg-namespace-animate-target.svg:
10:04 AM Changeset in webkit [249578] by mark.lam@apple.com
  • 10 edits
    2 adds in trunk

Fix bmalloc::Allocator:tryAllocate() to return null on failure to allocate.
https://bugs.webkit.org/show_bug.cgi?id=201529
<rdar://problem/53935772>

Reviewed by Yusuke Suzuki.

JSTests:

  • stress/test-out-of-memory.js: Added.

Source/bmalloc:

In this implementation, we pass FailureAction in as a runtime option. If this
proves to be a perf issue, we can easily fix this by passing it as a template
argument. That will also automatically elide unneeded code paths. We'll defer
that exercise until we have evidence that it is warranted.

  • CMakeLists.txt:
  • bmalloc.xcodeproj/project.pbxproj:
  • bmalloc/Allocator.cpp:

(bmalloc::Allocator::allocateImpl):
(bmalloc::Allocator::reallocateImpl):
(bmalloc::Allocator::refillAllocatorSlowCase):
(bmalloc::Allocator::refillAllocator):
(bmalloc::Allocator::allocateLarge):
(bmalloc::Allocator::allocateLogSizeClass):
(bmalloc::Allocator::allocateSlowCase):
(bmalloc::Allocator::tryAllocate): Deleted.
(bmalloc::Allocator::allocate): Deleted.
(bmalloc::Allocator::reallocate): Deleted.
(bmalloc::Allocator::tryReallocate): Deleted.

  • bmalloc/Allocator.h:

(bmalloc::Allocator::tryAllocate):
(bmalloc::Allocator::allocate):
(bmalloc::Allocator::tryReallocate):
(bmalloc::Allocator::reallocate):
(bmalloc::Allocator::allocateImpl):

  • bmalloc/BumpAllocator.h:
  • bmalloc/FailureAction.h: Added.
  • bmalloc/Heap.cpp:

(bmalloc::Heap::allocateSmallChunk):
(bmalloc::Heap::allocateSmallPage):
(bmalloc::Heap::allocateSmallBumpRangesByMetadata):
(bmalloc::Heap::allocateSmallBumpRangesByObject):
(bmalloc::Heap::allocateLarge):
(bmalloc::Heap::tryAllocateLarge): Deleted.

  • bmalloc/Heap.h:

(bmalloc::Heap::allocateSmallBumpRanges):

  • bmalloc/bmalloc.cpp:

(bmalloc::api::tryLargeZeroedMemalignVirtual):

10:03 AM Changeset in webkit [249577] by Ryan Haddad
  • 5 edits
    2 copies
    4 adds
    1 delete in trunk

Unreviewed, rolling out r249566.

Causes inspector layout test crashes under GuardMalloc

Reverted changeset:

"Tail Deleted Frames shown in Web Inspector are sometimes
incorrect (Shadow Chicken)"
https://bugs.webkit.org/show_bug.cgi?id=201366
https://trac.webkit.org/changeset/249566

9:56 AM Changeset in webkit [249576] by guijemont@igalia.com
  • 2 edits in trunk/Source/JavaScriptCore

testmasm: save r6 in JIT'ed code on ARM_THUMB2
https://bugs.webkit.org/show_bug.cgi?id=201138

Reviewed by Mark Lam.

MacroAssemblerArmv7 uses r6 as a temporary register, and it is a
callee-saved register. The JITs use
AssemblyHelpers::emitSaveCalleeSaves() and friends to save
callee-saved registers, but there is no such mechanism in testmasm,
which seems to make the assumption that the macroassembler does not
use callee-saved registers (which I guess is true for all other
architectures, but not for Armv7).

This issue means that testmasm crashes on Armv7 since code generated
by gcc uses r6, and it gets modified by JIT'ed code.

This change makes sure that we save and restore r6 for all code
compiled by testmasm on Armv7.

  • assembler/testmasm.cpp:

(JSC::emitFunctionPrologue):
(JSC::emitFunctionEpilogue):
(JSC::testSimple):
(JSC::testGetEffectiveAddress):
(JSC::testBranchTruncateDoubleToInt32):
(JSC::testBranchTestBit32RegReg):
(JSC::testBranchTestBit32RegImm):
(JSC::testBranchTestBit32AddrImm):
(JSC::testBranchTestBit64RegReg):
(JSC::testBranchTestBit64RegImm):
(JSC::testBranchTestBit64AddrImm):
(JSC::testCompareDouble):
(JSC::testMul32WithImmediates):
(JSC::testMul32SignExtend):
(JSC::testCompareFloat):
(JSC::testProbeReadsArgumentRegisters):
(JSC::testProbeWritesArgumentRegisters):
(JSC::testProbePreservesGPRS):
(JSC::testProbeModifiesStackPointer):
(JSC::testProbeModifiesProgramCounter):
(JSC::testProbeModifiesStackValues):
(JSC::testByteSwap):
(JSC::testMoveDoubleConditionally32):
(JSC::testMoveDoubleConditionally64):
(JSC::testCagePreservesPACFailureBit):

9:19 AM Changeset in webkit [249575] by achristensen@apple.com
  • 20 edits in trunk

When disabling legacy private browsing for testing, change the SessionID back to what it was, not the defaultSessionID
https://bugs.webkit.org/show_bug.cgi?id=201480

Reviewed by Youenn Fablet.

Source/WebCore:

No change in behavior, but this blocks bug 200050 which will make WebKitTestRunner use a persistent, non-default session.
Without this change, that change causes lots of test failures because we are switching from the legacy private browsing session
to the default session instead of the session we were using.

  • page/Page.cpp:

(WebCore::Page::enableLegacyPrivateBrowsing): Deleted.

  • page/Page.h:
  • page/PageGroup.cpp:

(WebCore::PageGroup::addPage):
(WebCore::PageGroup::setSessionIDForTesting):
(WebCore::PageGroup::enableLegacyPrivateBrowsingForTesting): Deleted.

  • page/PageGroup.h:
  • storage/StorageNamespaceProvider.cpp:

(WebCore::StorageNamespaceProvider::setSessionIDForTesting):
(WebCore::StorageNamespaceProvider::enableLegacyPrivateBrowsingForTesting): Deleted.

  • storage/StorageNamespaceProvider.h:

Source/WebKit:

  • WebProcess/InjectedBundle/InjectedBundle.cpp:

(WebKit::InjectedBundle::setPrivateBrowsingEnabled):

  • WebProcess/InjectedBundle/InjectedBundle.h:

Source/WebKitLegacy/mac:

  • WebView/WebView.mm:

(-[WebView _preferencesChanged:]):

Source/WebKitLegacy/win:

  • WebView.cpp:

(WebView::notifyPreferencesChanged):

Tools:

  • WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:

(WTR::InjectedBundle::beginTesting):

  • WebKitTestRunner/InjectedBundle/TestRunner.cpp:

(WTR::TestRunner::setPrivateBrowsingEnabled):

8:49 AM Changeset in webkit [249574] by youenn@apple.com
  • 9 edits in trunk

Remove MediaStreamPrivate::scheduleDeferredTask
https://bugs.webkit.org/show_bug.cgi?id=200975

Reviewed by Eric Carlson.

LayoutTests/imported/w3c:

  • web-platform-tests/mediacapture-streams/MediaStream-finished-add.https-expected.txt:

Source/WebCore:

All calls to scheduleDeferredTask are done on the main thread.
This was initially done to trigger less reconfiguration.
But this makes the implementation significantly more complex.

For instance, we have to wait for the document to update its media state
and send it to UIProcess before calling the allow completion handler.

Covered by existing tests.

  • Modules/mediastream/MediaStream.cpp:

(WebCore::MediaStream::MediaStream):
Make sure to update the document media state once the tracks have been added, similarly to the other constructor.
This ensures the document media state is computed with the new MediaStreamTrack.

  • Modules/mediastream/UserMediaRequest.cpp:

(WebCore::isMediaStreamCorrectlyStarted):
(WebCore::UserMediaRequest::allow):
(WebCore::UserMediaRequest::stop):
(WebCore::UserMediaRequest::mediaStreamDidFail):

  • Modules/mediastream/UserMediaRequest.h:
  • page/MediaProducer.h:

(WebCore::MediaProducer::isCapturing):
Make sure to include getDisplayMedia as part of capture check.

  • platform/mediastream/MediaStreamPrivate.cpp:

(WebCore::MediaStreamPrivate::trackMutedChanged):
(WebCore::MediaStreamPrivate::trackEnabledChanged):
(WebCore::MediaStreamPrivate::trackStarted):
(WebCore::MediaStreamPrivate::trackEnded):

  • platform/mediastream/MediaStreamPrivate.h:
6:05 AM Changeset in webkit [249573] by Carlos Garcia Campos
  • 2 edits in releases/WebKitGTK/webkit-2.26/Source/WebKit

Unreviewed. Disable PSON in GTK port

There are some issues causing problems in applications.

  • Shared/WebPreferencesDefaultValues.h:
5:38 AM WebKitGTK/2.26.x edited by Michael Catanzaro
(diff)
5:27 AM Changeset in webkit [249572] by commit-queue@webkit.org
  • 34 edits
    23 adds
    2 deletes in trunk

LayoutTests/imported/w3c:
Implement MathMLIDL / HTMLOrForeignElement
https://bugs.webkit.org/show_bug.cgi?id=200470

Patch by Rob Buis <rbuis@igalia.com> on 2019-09-06
Reviewed by Ryosuke Niwa.

Add tests for functionality exposed by HTMLOrForeignElement.
Import WPT tests as of 6e83b23bb962c97687b6573c378963208219ad1b (origin/master).

  • resources/import-expectations.json:
  • web-platform-tests/mathml/relations/html5-tree/class-1.html:
  • web-platform-tests/mathml/relations/html5-tree/class-2.html:
  • web-platform-tests/mathml/relations/html5-tree/clipboard-event-handlers.tentative-expected.txt: Added.
  • web-platform-tests/mathml/relations/html5-tree/clipboard-event-handlers.tentative.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/color-attributes-1.html:
  • web-platform-tests/mathml/relations/html5-tree/css-inline-style-dynamic.tentative-expected.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/css-inline-style-dynamic.tentative.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/css-inline-style-interface.tentative-expected.txt: Added.
  • web-platform-tests/mathml/relations/html5-tree/css-inline-style-interface.tentative.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/display-1-expected.txt:
  • web-platform-tests/mathml/relations/html5-tree/display-1.html:
  • web-platform-tests/mathml/relations/html5-tree/dynamic-1.html:
  • web-platform-tests/mathml/relations/html5-tree/href-click-1.html:
  • web-platform-tests/mathml/relations/html5-tree/href-click-2.html:
  • web-platform-tests/mathml/relations/html5-tree/href-click-3.html:
  • web-platform-tests/mathml/relations/html5-tree/html-or-foreign-element-interfaces.tentative-expected.txt: Added.
  • web-platform-tests/mathml/relations/html5-tree/html-or-foreign-element-interfaces.tentative.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/integration-point-1.html:
  • web-platform-tests/mathml/relations/html5-tree/integration-point-2.html:
  • web-platform-tests/mathml/relations/html5-tree/integration-point-3.html:
  • web-platform-tests/mathml/relations/html5-tree/integration-point-4-expected.txt: Added.
  • web-platform-tests/mathml/relations/html5-tree/integration-point-4.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/math-global-event-handlers.tentative-expected.txt: Added.
  • web-platform-tests/mathml/relations/html5-tree/math-global-event-handlers.tentative.html: Added.
  • web-platform-tests/mathml/relations/html5-tree/required-extensions-1-expected.txt: Removed.
  • web-platform-tests/mathml/relations/html5-tree/required-extensions-1.html: Removed.
  • web-platform-tests/mathml/relations/html5-tree/required-extensions-2.html:
  • web-platform-tests/mathml/relations/html5-tree/unique-identifier-1.html:
  • web-platform-tests/mathml/relations/html5-tree/unique-identifier-2.html:
  • web-platform-tests/mathml/relations/html5-tree/unique-identifier-3.html:
  • web-platform-tests/mathml/relations/html5-tree/w3c-import.log:
  • web-platform-tests/mathml/support/attribute-values.js: Added.

(AttributeValueTransforms.lowercase):
(AttributeValueTransforms.uppercase):
(AttributeValueTransforms.alternate_case):
(TransformAttributeValues):

  • web-platform-tests/mathml/support/box-comparison.js: Added.

(spaceBetween):
(measureSpaceAround):
(compareSpaceWithAndWithoutStyle):
(compareSizeWithAndWithoutStyle):

  • web-platform-tests/mathml/support/layout-comparison.js: Added.

(getWritingMode):
(compareSize):
(participateToParentLayout):
(childrenParticipatingToLayout):
(compareLayout):

  • web-platform-tests/mathml/support/mathml-fragments.js: Added.

(FragmentHelper.createElement):
(FragmentHelper.isValidChildOfMrow):
(FragmentHelper.isEmpty):
(FragmentHelper.element):
(FragmentHelper.appendChild):
(FragmentHelper.forceNonEmptyElement):

  • web-platform-tests/mathml/support/w3c-import.log: Added.

Source/WebCore:
Implement MathML DOM
https://bugs.webkit.org/show_bug.cgi?id=200470

Patch by Rob Buis <rbuis@igalia.com> on 2019-09-06
Reviewed by Ryosuke Niwa.

Expose MathML DOM as specified here [1].

[1] https://mathml-refresh.github.io/mathml-core/#dom-mathmlelement

Tests: imported/w3c/web-platform-tests/mathml/relations/html5-tree/clipboard-event-handlers.tentative.html

imported/w3c/web-platform-tests/mathml/relations/html5-tree/css-inline-style-dynamic.tentative.html
imported/w3c/web-platform-tests/mathml/relations/html5-tree/css-inline-style-interface.tentative.html
imported/w3c/web-platform-tests/mathml/relations/html5-tree/html-or-foreign-element-interfaces.tentative.html
imported/w3c/web-platform-tests/mathml/relations/html5-tree/integration-point-4.html
imported/w3c/web-platform-tests/mathml/relations/html5-tree/math-global-event-handlers.tentative.html
mathml/focus-event-handling.html
mathml/tabindex-order.html

  • CMakeLists.txt:
  • DerivedSources-input.xcfilelist:
  • DerivedSources-output.xcfilelist:
  • DerivedSources.make:
  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/JSElementCustom.cpp:

(WebCore::createNewElementWrapper):

  • bindings/js/JSNodeCustom.cpp:

(WebCore::createWrapperInline):

  • mathml/MathMLElement.cpp:

(WebCore::MathMLElement::parseAttribute):

  • mathml/MathMLElement.idl: Copied from Source/WebCore/html/HTMLOrForeignElement.idl.
  • mathml/MathMLMathElement.idl: Copied from Source/WebCore/html/HTMLOrForeignElement.idl.
  • mathml/mathtags.in:

LayoutTests:
Implement MathML DOM
https://bugs.webkit.org/show_bug.cgi?id=200470

Patch by Rob Buis <rbuis@igalia.com> on 2019-09-06
Reviewed by Ryosuke Niwa.

Extend existing tests to also test focus for MathML. Add a standalone test for MathML
based on svg/custom/tabindex-order.html. Add a test for mouse clicks and focus handling
based on svg/custom/focus-event-handling.xhtml.

  • fast/dom/tabindex-defaults-expected.txt:
  • fast/dom/tabindex-defaults.html:
  • mathml/focus-event-handling-expected.txt: Added.
  • mathml/focus-event-handling.html: Added.
  • mathml/tabindex-order-expected.txt: Added.
  • mathml/tabindex-order.html: Added.
  • platform/ios-wk2/TestExpectations:
3:01 AM Changeset in webkit [249571] by commit-queue@webkit.org
  • 16 edits in trunk

Add runtime flag for lazy image loading
https://bugs.webkit.org/show_bug.cgi?id=199794

Patch by Rob Buis <rbuis@igalia.com> on 2019-09-06
Reviewed by Frédéric Wang.

Remove parts of r248409 that were meant for WK1 since
lazy image loading is WK2 only.

Source/WebKitLegacy/mac:

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

(+[WebPreferences initialize]):
(-[WebPreferences lazyImageLoadingEnabled]): Deleted.
(-[WebPreferences setLazyImageLoadingEnabled:]): Deleted.

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

(-[WebView _preferencesChanged:]):

Source/WebKitLegacy/win:

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

(WebPreferences::initializeDefaultSettings):
(WebPreferences::setResizeObserverEnabled):
(WebPreferences::lazyImageLoadingEnabled): Deleted.
(WebPreferences::setLazyImageLoadingEnabled): Deleted.

  • WebPreferences.h:
  • WebView.cpp:

(WebView::notifyPreferencesChanged):

Tools:

  • DumpRenderTree/TestOptions.cpp:

(TestOptions::TestOptions):

  • DumpRenderTree/TestOptions.h:
  • DumpRenderTree/mac/DumpRenderTree.mm:

(enableExperimentalFeatures):
(setWebPreferencesForTestOptions):

  • DumpRenderTree/win/DumpRenderTree.cpp:

(enableExperimentalFeatures):

2:14 AM Changeset in webkit [249570] by Joseph Pecoraro
  • 2 edits in trunk/Source/WebInspectorUI

Web Inspector: Network: Error response headers summary has wrong color border
https://bugs.webkit.org/show_bug.cgi?id=201538
<rdar://problem/55104469>

Reviewed by Devin Rousso.

  • UserInterface/Views/ResourceHeadersContentView.css:

(body[dir] .resource-headers > section.error > .details):
Re-apply an error style for resource-header sections.

2:00 AM Changeset in webkit [249569] by Patrick Griffis
  • 2 edits in trunk/Source/WebKit

[GTK][WPE] Do not enable the sandbox in Snap
https://bugs.webkit.org/show_bug.cgi?id=201486

Reviewed by Carlos Garcia Campos.

Running inside of Snap adds its own complications and is simply
not supported for now.

Also update isInsideFlatpak() for consistency.

  • UIProcess/Launcher/glib/ProcessLauncherGLib.cpp:

(WebKit::isInsideFlatpak):
(WebKit::isInsideSnap):
(WebKit::ProcessLauncher::launchProcess):

1:51 AM Changeset in webkit [249568] by zandobersek@gmail.com
  • 5 edits in trunk/Source/WebCore

[GStreamer] YUV buffers on iMX platforms have to be treated as RGBA
https://bugs.webkit.org/show_bug.cgi?id=201537

Reviewed by Philippe Normand.

The imxvpudecoder element on iMX platforms decodes YUV data in a
platform-specific way, gathering the YUV data in a single texture and
then relying on the sampler in the Vivante graphics drivers to
automagically decode that YUV data into RGBA values.

To correctly display such decoded data, we have to represent that
single texture as an RGBA texture, even when the GStreamer buffer is
crafted as containing planar YUV data.

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

(WebCore::MediaPlayerPrivateGStreamer::createGSTPlayBin):

  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::GstVideoFrameHolder::GstVideoFrameHolder):
(WebCore::GstVideoFrameHolder::platformLayerBuffer):
(WebCore::MediaPlayerPrivateGStreamerBase::~MediaPlayerPrivateGStreamerBase):
(WebCore::MediaPlayerPrivateGStreamerBase::pushTextureToCompositor):
(WebCore::MediaPlayerPrivateGStreamerBase::flushCurrentBuffer):
(WebCore::MediaPlayerPrivateGStreamerBase::copyVideoTextureToPlatformTexture):
(WebCore::MediaPlayerPrivateGStreamerBase::nativeImageForCurrentTime):

  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.h:
1:22 AM Changeset in webkit [249567] by Joseph Pecoraro
  • 5 edits in trunk

Web Inspector: Crashes when intercept response is a 404
https://bugs.webkit.org/show_bug.cgi?id=201514

Reviewed by Alex Christensen.

Source/WebKit:

  • WebProcess/Network/WebResourceLoader.cpp:

(WebKit::WebResourceLoader::didReceiveResponse):
Ensure the WebResourceLoader stays around during intercepts.

(WebKit::WebResourceLoader::didReceiveData):
(WebKit::WebResourceLoader::didFinishResourceLoad):
(WebKit::WebResourceLoader::didFailResourceLoad):
Only perform the work if the m_coreLoader is still alive.

LayoutTests:

  • http/tests/inspector/network/local-resource-override-basic-expected.txt:
  • http/tests/inspector/network/local-resource-override-basic.html:

Add a test for a 404 response override.

1:14 AM Changeset in webkit [249566] by Joseph Pecoraro
  • 5 edits
    2 copies
    2 moves
    17 adds
    4 deletes in trunk

Tail Deleted Frames shown in Web Inspector are sometimes incorrect (Shadow Chicken)
https://bugs.webkit.org/show_bug.cgi?id=201366

Reviewed by Saam Barati.

Source/JavaScriptCore:

It is possible for the log buffer to be full right as someone is trying to
log a function prologue. In such a case the machine stack has already been
updated to include the new JavaScript call frame, but the prologue packet
cannot be included in the update because the log is full. This would mean
that the update fails to rationalize the machine stack with the shadow
log / stack. Namely, the current JavaScript call frame is unable to
find a matching prologue (the one we are holding to include after the update)
and inserts a questionable value into the stack; and in the process
missing and removing real potential tail calls.

For example:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function start() { return first(); }

If the the log fills up just as we are entering b then we may have a list
full log of packets looking like:

Shadow Log:

...
{ prologue-packet: entering start ... }
{ prologue-packet: entering first ... }
{ tail-packet: leaving first with a tail call }

Incoming Packet:

{ prologue-packet: entering second ... }

Current JS Stack:

second
start

Since the Current JavaScript stack already has second, if we process the
log without the prologue for second then we push a confused entry on the
shadow stack and clear the log such that we eventually lose the tail-call
information for first to second.

This patch solves this issue by providing enough extra space in the log
to always process the incoming packet when that forces an update. This way
clients can continue to behave exactly as they are.

--

We also document a corner case in some circumstances where the shadow
log may currently be insufficient to know how to reconcile:

For example:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function doNothingTail() { return Math.random() }
function start() {

for (i=0;i<1000;++i) doNothingTail();
return first();

}

In this case the ShadowChicken log may be processed multiple times due
to the many calls to doNothingTail / Math.random(). When calling the
Native function no prologue packet is emitted, so it is unclear that we
temporarly go deeper and come back out on the stack, so the log appears
to have lots of doNothingTail calls reusing the same frame:

Shadow Log:

...
, [123] {callee = 0x72a21aee0, frame = 0x7ffeef897270, callerFrame = 0x7ffeef8972e0, name = start}
, [124] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [125] tail-packet:{frame = 0x7ffeef8971f0}
, [126] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [127] tail-packet:{frame = 0x7ffeef8971f0}
...
, [140] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [141] tail-packet:{frame = 0x7ffeef8971f0}
, [142] {callee = 0x72a21af10, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = doNothingTail}
, [143] tail-packet:{frame = 0x7ffeef8971f0}
, [144] {callee = 0x72a21aeb0, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = first}
, [145] tail-packet:{frame = 0x7ffeef8971f0}
, [146] {callee = 0x72a21ae80, frame = 0x7ffeef8971f0, callerFrame = 0x7ffeef897270, name = second}
...

This log would seem to be indistinguishable from real tail recursion, such as:

"use strict";
function third() { return 1; }
function second() { return third(); }
function first() { return second(); }
function doNothingTail(n) {

return n ? doNothingTail(n-1) : first();

}
function start() {

return doNothingTail(1000);

}

Likewise there are more cases where the shadow log appears to be ambiguous with determining
the appropriate parent call frame with intermediate function calls. In practice this may
not be too problematic, as this is a best effort reconstruction of tail deleted frames.
It seems likely we would only show additional frames that did in fact happen serially
between JavaScript call frames, but may not actually be the proper parent frames
heirachy in the stack.

  • interpreter/ShadowChicken.cpp:

(JSC::ShadowChicken::Packet::dump const):
(JSC::ShadowChicken::Frame::dump const):
(JSC::ShadowChicken::dump const):
Improved debugging output. Especially for functions.

(JSC::ShadowChicken::ShadowChicken):
Make space in the log for 1 additional packet to process when we slow log.

(JSC::ShadowChicken::log):
Include this packet in our update.

(JSC::ShadowChicken::update):
Address an edge case where we can eliminate tail-deleted frames that don't make sense.

LayoutTests:

  • inspector/debugger/tail-deleted-frames-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-from-vm-entry-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-from-vm-entry.html: Removed.
  • inspector/debugger/tail-deleted-frames-this-value-expected.txt: Removed.
  • inspector/debugger/tail-deleted-frames-this-value.html: Removed.
  • inspector/debugger/tail-deleted-frames.html: Removed.

Remove legacy tests that are difficult to read.

  • inspector/debugger/tail-deleted-frames/resources/stack-trace-utilities.js: Added.

(TestPage.registerInitializer.window.getAsyncStackTrace):
(TestPage.registerInitializer.async.logThisObject):
(TestPage.registerInitializer.async.logScope):
(TestPage.registerInitializer.async.logCallFrame):
(TestPage.registerInitializer):

  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-frames.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-native-tail-deleted-calls.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-intermediate-tail-deleted-frames.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-scopes.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-this-value.js: Added.
  • inspector/debugger/tail-deleted-frames/resources/tail-deleted-frames-vm-entry.js: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-frames-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-frames.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-tail-deleted-frames-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-tail-deleted-frames.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-scopes-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-scopes.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-this-value-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-this-value.html: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-vm-entry-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-vm-entry.html: Added.

Include modern tests that are easier to read.

  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-native-tail-deleted-calls-expected.txt: Added.
  • inspector/debugger/tail-deleted-frames/tail-deleted-frames-intermediate-native-tail-deleted-calls.html: Added.

Include a test that is known to produce bad output, since we have reproductive steps.

  • platform/mac/TestExpectations:

Updated pathes.

12:06 AM Changeset in webkit [249565] by commit-queue@webkit.org
  • 5 edits in trunk

AccessibilityRenderObject::setSelectedTextRange fails to set the selection passed an empty line.
https://bugs.webkit.org/show_bug.cgi?id=201518
<rdar://problem/54835122>

Patch by Andres Gonzalez <Andres Gonzalez> on 2019-09-06
Reviewed by Ryosuke Niwa.

Source/WebCore:

Test: accessibility/set-selected-text-range-after-newline.html

In the case of an empty line, the CharacterIterator range start and end
were not equal, thus we were not advancing the iterator and returning
the iterator range end, which is not correct. With this change we are
always advancing the iterator if its text is just '\n'. This covers all
the cases we fixed before plus empty lines.

  • editing/Editing.cpp:

(WebCore::visiblePositionForIndexUsingCharacterIterator):

LayoutTests:

Extended this test to set the selection range passed an empty line.

  • accessibility/set-selected-text-range-after-newline-expected.txt:
  • accessibility/set-selected-text-range-after-newline.html:

Sep 5, 2019:

10:57 PM Changeset in webkit [249564] by Chris Dumez
  • 5 edits in trunk

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in imported/w3c/IndexedDB-private-browsing
https://bugs.webkit.org/show_bug.cgi?id=201519

Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Drop Webkit-specific changes to WPT tests.

  • IndexedDB-private-browsing/support.js:

Tools:

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in imported/w3c/IndexedDB-private-browsing
as it does not truly enable private browsing for WebKit2. Instead, have WKTR / DRT set the
TestOption::useEphemeralSession flag based on the test folder.

  • DumpRenderTree/mac/DumpRenderTree.mm:

(shouldUseEphemeralSession):
(runTest):

  • WebKitTestRunner/TestOptions.cpp:

(WTR::shouldUseEphemeralSession):
(WTR::TestOptions::TestOptions):

10:06 PM Changeset in webkit [249563] by Chris Dumez
  • 308 edits in trunk

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in js-test.js
https://bugs.webkit.org/show_bug.cgi?id=201522

Reviewed by Alex Christensen.

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in js-test.js.
Instead, use the new useEphemeralSession WKTR flag added in r249542.

  • resources/js-test.js:
  • storage/indexeddb/aborted-versionchange-closes-private.html:
  • storage/indexeddb/basics-private.html:
  • storage/indexeddb/clone-exception-private.html:
  • storage/indexeddb/closed-cursor-private.html:
  • storage/indexeddb/connection-leak-private.html:
  • storage/indexeddb/create-and-remove-object-store-private.html:
  • storage/indexeddb/create-object-store-options-private.html:
  • storage/indexeddb/createIndex-after-failure-private.html:
  • storage/indexeddb/createObjectStore-name-argument-required-private.html:
  • storage/indexeddb/createObjectStore-null-name-private.html:
  • storage/indexeddb/cursor-added-bug-private.html:
  • storage/indexeddb/cursor-advance-private.html:
  • storage/indexeddb/cursor-basics-private.html:
  • storage/indexeddb/cursor-cast-private.html:
  • storage/indexeddb/cursor-continue-dir-private.html:
  • storage/indexeddb/cursor-continue-private.html:
  • storage/indexeddb/cursor-continue-validity-private.html:
  • storage/indexeddb/cursor-delete-private.html:
  • storage/indexeddb/cursor-finished-private.html:
  • storage/indexeddb/cursor-inconsistency-private.html:
  • storage/indexeddb/cursor-index-delete-private.html:
  • storage/indexeddb/cursor-key-order-private.html:
  • storage/indexeddb/cursor-leak-private.html:
  • storage/indexeddb/cursor-overloads-private.html:
  • storage/indexeddb/cursor-prev-no-duplicate-private.html:
  • storage/indexeddb/cursor-primary-key-order-private.html:
  • storage/indexeddb/cursor-properties-private.html:
  • storage/indexeddb/cursor-request-cycle-private.html:
  • storage/indexeddb/cursor-reverse-bug-private.html:
  • storage/indexeddb/cursor-skip-deleted-private.html:
  • storage/indexeddb/cursor-update-private.html:
  • storage/indexeddb/cursor-update-value-argument-required-private.html:
  • storage/indexeddb/cursor-value-private.html:
  • storage/indexeddb/database-basics-private.html:
  • storage/indexeddb/database-close-private.html:
  • storage/indexeddb/database-closepending-flag-private.html:
  • storage/indexeddb/database-deletepending-flag-private.html:
  • storage/indexeddb/database-name-undefined-private.html:
  • storage/indexeddb/database-odd-names-private.html:
  • storage/indexeddb/database-quota-private.html:
  • storage/indexeddb/database-wrapper-private.html:
  • storage/indexeddb/delete-closed-database-object-private.html:
  • storage/indexeddb/delete-in-upgradeneeded-close-in-open-success-private.html:
  • storage/indexeddb/delete-in-upgradeneeded-close-in-versionchange-private.html:
  • storage/indexeddb/delete-range-private.html:
  • storage/indexeddb/deleteIndex-bug110792-private.html:
  • storage/indexeddb/deleteIndex-private.html:
  • storage/indexeddb/deleteObjectStore-name-argument-required-private.html:
  • storage/indexeddb/deleteObjectStore-null-name-private.html:
  • storage/indexeddb/deleted-objects-private.html:
  • storage/indexeddb/deletedatabase-blocked-private.html:
  • storage/indexeddb/deletedatabase-delayed-by-open-and-versionchange-private.html:
  • storage/indexeddb/deletedatabase-not-blocked-private.html:
  • storage/indexeddb/deletedatabase-transaction-private.html:
  • storage/indexeddb/dont-commit-on-blocked-private.html:
  • storage/indexeddb/dont-wedge-private.html:
  • storage/indexeddb/duplicates-private.html:
  • storage/indexeddb/error-causes-abort-by-default-private.html:
  • storage/indexeddb/events-private.html:
  • storage/indexeddb/exception-in-event-aborts-private.html:
  • storage/indexeddb/exceptions-private.html:
  • storage/indexeddb/factory-cmp-private.html:
  • storage/indexeddb/factory-deletedatabase-private.html:
  • storage/indexeddb/get-keyrange-private.html:
  • storage/indexeddb/index-basics-private.html:
  • storage/indexeddb/index-count-private.html:
  • storage/indexeddb/index-cursor-private.html:
  • storage/indexeddb/index-duplicate-keypaths-private.html:
  • storage/indexeddb/index-get-key-argument-required-private.html:
  • storage/indexeddb/index-multientry-private.html:
  • storage/indexeddb/index-population-private.html:
  • storage/indexeddb/index-unique-private.html:
  • storage/indexeddb/interfaces-private.html:
  • storage/indexeddb/intversion-abort-in-initial-upgradeneeded-private.html:
  • storage/indexeddb/intversion-bad-parameters-private.html:
  • storage/indexeddb/intversion-blocked-private.html:
  • storage/indexeddb/intversion-close-between-events-private.html:
  • storage/indexeddb/intversion-close-in-oncomplete-private.html:
  • storage/indexeddb/intversion-close-in-upgradeneeded-private.html:
  • storage/indexeddb/intversion-encoding-private.html:
  • storage/indexeddb/intversion-gated-on-delete-private.html:
  • storage/indexeddb/intversion-long-queue-private.html:
  • storage/indexeddb/intversion-omit-parameter-private.html:
  • storage/indexeddb/intversion-open-in-upgradeneeded-private.html:
  • storage/indexeddb/intversion-open-with-version-private.html:
  • storage/indexeddb/intversion-pending-version-changes-ascending-private.html:
  • storage/indexeddb/intversion-pending-version-changes-descending-private.html:
  • storage/indexeddb/intversion-pending-version-changes-same-private.html:
  • storage/indexeddb/intversion-persistence-private.html:
  • storage/indexeddb/intversion-revert-on-abort-private.html:
  • storage/indexeddb/intversion-two-opens-no-versions-private.html:
  • storage/indexeddb/intversion-upgrades-private.html:
  • storage/indexeddb/invalid-keys-private.html:
  • storage/indexeddb/key-generator-private.html:
  • storage/indexeddb/key-sort-order-across-types-private.html:
  • storage/indexeddb/key-sort-order-date-private.html:
  • storage/indexeddb/key-type-array-private.html:
  • storage/indexeddb/key-type-binary-private.html:
  • storage/indexeddb/key-type-infinity-private.html:
  • storage/indexeddb/keypath-arrays-private.html:
  • storage/indexeddb/keypath-basics-private.html:
  • storage/indexeddb/keypath-edges-private.html:
  • storage/indexeddb/keypath-fetch-key-private.html:
  • storage/indexeddb/keypath-intrinsic-properties-private.html:
  • storage/indexeddb/keyrange-private.html:
  • storage/indexeddb/keyrange-required-arguments-private.html:
  • storage/indexeddb/lazy-index-population-private.html:
  • storage/indexeddb/lazy-index-types-private.html:
  • storage/indexeddb/legacy-constants-private.html:
  • storage/indexeddb/list-ordering-private.html:
  • storage/indexeddb/metadata-private.html:
  • storage/indexeddb/metadata-race-private.html:
  • storage/indexeddb/modern/abort-objectstore-info-private.html:
  • storage/indexeddb/modern/abort-requests-cancelled-private.html:
  • storage/indexeddb/modern/aborted-put-private.html:
  • storage/indexeddb/modern/autoincrement-abort-private.html:
  • storage/indexeddb/modern/basic-add-private.html:
  • storage/indexeddb/modern/basic-put-private.html:
  • storage/indexeddb/modern/binary-keys-1-private.html:
  • storage/indexeddb/modern/blocked-open-db-requests-private.html:
  • storage/indexeddb/modern/create-index-failures-private.html:
  • storage/indexeddb/modern/createobjectstore-basic-private.html:
  • storage/indexeddb/modern/createobjectstore-failures-private.html:
  • storage/indexeddb/modern/cursor-1-private.html:
  • storage/indexeddb/modern/cursor-2-private.html:
  • storage/indexeddb/modern/cursor-3-private.html:
  • storage/indexeddb/modern/cursor-4-private.html:
  • storage/indexeddb/modern/cursor-5-private.html:
  • storage/indexeddb/modern/cursor-6-private.html:
  • storage/indexeddb/modern/cursor-7-private.html:
  • storage/indexeddb/modern/cursor-8-private.html:
  • storage/indexeddb/modern/date-basic-private.html:
  • storage/indexeddb/modern/deletedatabase-1-private.html:
  • storage/indexeddb/modern/deletedatabase-2-private.html:
  • storage/indexeddb/modern/deletedatabase-null-name-exception-private.html:
  • storage/indexeddb/modern/deletedatabase-request-event-private.html:
  • storage/indexeddb/modern/deletedatabase-request-private.html:
  • storage/indexeddb/modern/deleteindex-1-private.html:
  • storage/indexeddb/modern/deleteindex-2-private.html:
  • storage/indexeddb/modern/deleteindex-3-private.html:
  • storage/indexeddb/modern/deleteindex-4-private.html:
  • storage/indexeddb/modern/deleteobjectstore-1-private.html:
  • storage/indexeddb/modern/double-abort-private.html:
  • storage/indexeddb/modern/double-open-private.html:
  • storage/indexeddb/modern/gc-closes-database-private.html:
  • storage/indexeddb/modern/get-index-failures-private.html:
  • storage/indexeddb/modern/get-keyrange-private.html:
  • storage/indexeddb/modern/idbcursor-continue-primary-key-1-private.html:
  • storage/indexeddb/modern/idbdatabase-deleteobjectstore-failures-private.html:
  • storage/indexeddb/modern/idbdatabase-transaction-failures-private.html:
  • storage/indexeddb/modern/idbindex-getall-1-private.html:
  • storage/indexeddb/modern/idbindex-getallkeys-1-private.html:
  • storage/indexeddb/modern/idbindex-properties-basic-private.html:
  • storage/indexeddb/modern/idbkey-array-equality-private.html:
  • storage/indexeddb/modern/idbobjectstore-clear-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-clear-2-private.html:
  • storage/indexeddb/modern/idbobjectstore-count-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-count-failures-private.html:
  • storage/indexeddb/modern/idbobjectstore-delete-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-delete-2-private.html:
  • storage/indexeddb/modern/idbobjectstore-delete-failures-private.html:
  • storage/indexeddb/modern/idbobjectstore-get-failures-private.html:
  • storage/indexeddb/modern/idbobjectstore-getall-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-getallkeys-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-getkey-1-private.html:
  • storage/indexeddb/modern/idbobjectstore-put-and-clear-failures-private.html:
  • storage/indexeddb/modern/idbtransaction-objectstore-failures-private.html:
  • storage/indexeddb/modern/idbtransaction-objectstores-1-private.html:
  • storage/indexeddb/modern/index-1-private.html:
  • storage/indexeddb/modern/index-2-private.html:
  • storage/indexeddb/modern/index-3-private.html:
  • storage/indexeddb/modern/index-4-private.html:
  • storage/indexeddb/modern/index-5-private.html:
  • storage/indexeddb/modern/index-cursor-1-private.html:
  • storage/indexeddb/modern/index-cursor-2-private.html:
  • storage/indexeddb/modern/index-cursor-3-private.html:
  • storage/indexeddb/modern/index-get-count-basic-private.html:
  • storage/indexeddb/modern/index-get-count-failures-private.html:
  • storage/indexeddb/modern/index-rename-1-private.html:
  • storage/indexeddb/modern/keypath-basic-private.html:
  • storage/indexeddb/modern/memory-index-not-deleted-with-objectstore-private.html:
  • storage/indexeddb/modern/multiple-objectstore-index-cursor-collision-private.html:
  • storage/indexeddb/modern/objectstore-attributes-private.html:
  • storage/indexeddb/modern/objectstore-cursor-advance-failures-private.html:
  • storage/indexeddb/modern/objectstore-cursor-continue-failures-private.html:
  • storage/indexeddb/modern/objectstore-rename-1-private.html:
  • storage/indexeddb/modern/opencursor-failures-private.html:
  • storage/indexeddb/modern/opendatabase-request-event-private.html:
  • storage/indexeddb/modern/opendatabase-request-private.html:
  • storage/indexeddb/modern/opendatabase-success-after-versionchange-private.html:
  • storage/indexeddb/modern/opendatabase-versions-private.html:
  • storage/indexeddb/modern/request-readystate-private.html:
  • storage/indexeddb/modern/single-entry-index-invalid-key-crash-private.html:
  • storage/indexeddb/modern/transaction-scheduler-1-private.html:
  • storage/indexeddb/modern/transaction-scheduler-2-private.html:
  • storage/indexeddb/modern/transaction-scheduler-3-private.html:
  • storage/indexeddb/modern/transaction-scheduler-4-private.html:
  • storage/indexeddb/modern/transaction-scheduler-5-private.html:
  • storage/indexeddb/modern/transaction-scheduler-6-private.html:
  • storage/indexeddb/modern/transactions-stop-on-navigation-private.html:
  • storage/indexeddb/modern/versionchange-abort-then-reopen-private.html:
  • storage/indexeddb/modern/versionchange-event-private.html:
  • storage/indexeddb/mozilla/add-twice-failure-private.html:
  • storage/indexeddb/mozilla/autoincrement-indexes-private.html:
  • storage/indexeddb/mozilla/bad-keypath-private.html:
  • storage/indexeddb/mozilla/clear-private.html:
  • storage/indexeddb/mozilla/create-index-unique-private.html:
  • storage/indexeddb/mozilla/create-index-with-integer-keys-private.html:
  • storage/indexeddb/mozilla/create-objectstore-basics-private.html:
  • storage/indexeddb/mozilla/create-objectstore-null-name-private.html:
  • storage/indexeddb/mozilla/cursor-mutation-objectstore-only-private.html:
  • storage/indexeddb/mozilla/cursor-mutation-private.html:
  • storage/indexeddb/mozilla/cursor-update-updates-indexes-private.html:
  • storage/indexeddb/mozilla/cursors-private.html:
  • storage/indexeddb/mozilla/delete-result-private.html:
  • storage/indexeddb/mozilla/event-source-private.html:
  • storage/indexeddb/mozilla/global-data-private.html:
  • storage/indexeddb/mozilla/index-prev-no-duplicate-private.html:
  • storage/indexeddb/mozilla/indexes-private.html:
  • storage/indexeddb/mozilla/key-requirements-delete-null-key-private.html:
  • storage/indexeddb/mozilla/key-requirements-inline-and-passed-private.html:
  • storage/indexeddb/mozilla/key-requirements-private.html:
  • storage/indexeddb/mozilla/key-requirements-put-no-key-private.html:
  • storage/indexeddb/mozilla/key-requirements-put-null-key-private.html:
  • storage/indexeddb/mozilla/object-cursors-private.html:
  • storage/indexeddb/mozilla/object-identity-private.html:
  • storage/indexeddb/mozilla/object-store-inline-autoincrement-key-added-on-put-private.html:
  • storage/indexeddb/mozilla/object-store-remove-values-private.html:
  • storage/indexeddb/mozilla/objectstorenames-private.html:
  • storage/indexeddb/mozilla/odd-result-order-private.html:
  • storage/indexeddb/mozilla/open-database-null-name-private.html:
  • storage/indexeddb/mozilla/put-get-values-private.html:
  • storage/indexeddb/mozilla/readonly-transactions-private.html:
  • storage/indexeddb/mozilla/readwrite-transactions-private.html:
  • storage/indexeddb/mozilla/readyState-private.html:
  • storage/indexeddb/mozilla/remove-index-private.html:
  • storage/indexeddb/mozilla/remove-objectstore-private.html:
  • storage/indexeddb/mozilla/versionchange-abort-private.html:
  • storage/indexeddb/mutating-cursor-private.html:
  • storage/indexeddb/object-lookups-in-versionchange-private.html:
  • storage/indexeddb/objectStore-required-arguments-private.html:
  • storage/indexeddb/objectstore-autoincrement-private.html:
  • storage/indexeddb/objectstore-basics-private.html:
  • storage/indexeddb/objectstore-clear-private.html:
  • storage/indexeddb/objectstore-count-private.html:
  • storage/indexeddb/objectstore-cursor-private.html:
  • storage/indexeddb/objectstore-removeobjectstore-private.html:
  • storage/indexeddb/odd-strings-private.html:
  • storage/indexeddb/open-bad-versions-private.html:
  • storage/indexeddb/open-cursor-private.html:
  • storage/indexeddb/open-during-transaction-private.html:
  • storage/indexeddb/open-ordering-private.html:
  • storage/indexeddb/opencursor-key-private.html:
  • storage/indexeddb/optional-arguments-private.html:
  • storage/indexeddb/pending-activity-private.html:
  • storage/indexeddb/pending-version-change-on-exit-private.html:
  • storage/indexeddb/pending-version-change-stuck-private.html:
  • storage/indexeddb/pending-version-change-stuck-works-with-terminate-private.html:
  • storage/indexeddb/persistence-private.html:
  • storage/indexeddb/prefetch-bugfix-108071-private.html:
  • storage/indexeddb/prefetch-invalidation-private.html:
  • storage/indexeddb/prefetch-race-private.html:
  • storage/indexeddb/primary-key-unique-to-objectstore-private.html:
  • storage/indexeddb/queued-commands-private.html:
  • storage/indexeddb/readonly-private.html:
  • storage/indexeddb/readonly-properties-private.html:
  • storage/indexeddb/removed-private.html:
  • storage/indexeddb/request-continue-abort-private.html:
  • storage/indexeddb/request-event-propagation-private.html:
  • storage/indexeddb/request-leak-private.html:
  • storage/indexeddb/request-result-cache-private.html:
  • storage/indexeddb/set_version_blocked-private.html:
  • storage/indexeddb/setversion-blocked-by-versionchange-close-private.html:
  • storage/indexeddb/setversion-not-blocked-private.html:
  • storage/indexeddb/structured-clone-private.html:
  • storage/indexeddb/transaction-abort-private.html:
  • storage/indexeddb/transaction-active-flag-private.html:
  • storage/indexeddb/transaction-after-close-private.html:
  • storage/indexeddb/transaction-and-objectstore-calls-private.html:
  • storage/indexeddb/transaction-basics-private.html:
  • storage/indexeddb/transaction-complete-with-js-recursion-cross-frame-private.html:
  • storage/indexeddb/transaction-complete-with-js-recursion-private.html:
  • storage/indexeddb/transaction-complete-workers-private.html:
  • storage/indexeddb/transaction-coordination-across-databases-private.html:
  • storage/indexeddb/transaction-coordination-within-database-private.html:
  • storage/indexeddb/transaction-crash-in-tasks-private.html:
  • storage/indexeddb/transaction-crash-on-abort-private.html:
  • storage/indexeddb/transaction-error-private.html:
  • storage/indexeddb/transaction-event-propagation-private.html:
  • storage/indexeddb/transaction-ordering-private.html:
  • storage/indexeddb/transaction-overlapping-private.html:
  • storage/indexeddb/transaction-read-only-private.html:
  • storage/indexeddb/transaction-readwrite-exclusive-private.html:
  • storage/indexeddb/transaction-rollback-private.html:
  • storage/indexeddb/transaction-scope-sequencing-private.html:
  • storage/indexeddb/transaction-storeNames-required-private.html:
  • storage/indexeddb/unblocked-version-changes-private.html:
  • storage/indexeddb/unprefix-private.html:
  • storage/indexeddb/value-undefined-private.html:
  • storage/indexeddb/values-odd-types-private.html:
  • storage/indexeddb/version-change-abort-private.html:
  • storage/indexeddb/version-change-event-basic-private.html:
  • storage/indexeddb/version-change-event-private.html:
  • storage/indexeddb/version-change-exclusive-private.html:
  • storage/indexeddb/versionchangerequest-activedomobject-private.html:
9:51 PM Changeset in webkit [249562] by Chris Dumez
  • 3 edits
    1 copy
    1 add in trunk/LayoutTests

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in http/tests/cache/disk-cache/disk-cache-vary-cookie.html
https://bugs.webkit.org/show_bug.cgi?id=201524

Reviewed by Alex Christensen.

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() as it does not do the right thing for WebKit2.
Instead, use the new useEphemeralSession WKTR flag added in r249542.

  • http/tests/cache/disk-cache/disk-cache-vary-cookie-expected.txt:
  • http/tests/cache/disk-cache/disk-cache-vary-cookie-private-expected.txt: Added.
  • http/tests/cache/disk-cache/disk-cache-vary-cookie-private.html: Copied from LayoutTests/http/tests/cache/disk-cache/disk-cache-vary-cookie.html.
  • http/tests/cache/disk-cache/disk-cache-vary-cookie.html:
9:37 PM Changeset in webkit [249561] by Chris Dumez
  • 5 edits
    2 adds in trunk/LayoutTests

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in plugins/private-browsing-mode.html
https://bugs.webkit.org/show_bug.cgi?id=201523

Reviewed by Alex Christensen.

Stop using testRunner.setPrivateBrowsingEnabled_DEPRECATED() in plugins/private-browsing-mode.html
since it does not do the right thing for WebKit2. Instead, use the new useEphemeralSession
WKTR flag.

  • platform/mac-wk2/TestExpectations:
  • platform/mac/TestExpectations:
  • plugins/non-private-browsing-mode-expected.txt: Added.
  • plugins/non-private-browsing-mode.html: Added.
  • plugins/private-browsing-mode-expected.txt:
  • plugins/private-browsing-mode.html:
7:38 PM Changeset in webkit [249560] by Fujii Hironori
  • 5 edits in trunk/Source

[Win] Add support for MouseEvent.buttons
https://bugs.webkit.org/show_bug.cgi?id=201445

Reviewed by Brent Fulgham.

Source/WebCore:

Spec: <https://www.w3.org/TR/uievents/#dom-mouseevent-buttons>

  • platform/win/GDIUtilities.h:

(buttonsForEvent): Added.

  • platform/win/PlatformMouseEventWin.cpp:

(WebCore::PlatformMouseEvent::PlatformMouseEvent):
Set m_buttons.

Source/WebKit:

  • Shared/win/WebEventFactory.cpp:

(WebKit::WebEventFactory::createWebMouseEvent):
Set m_buttons.

6:48 PM Changeset in webkit [249559] by Fujii Hironori
  • 5 edits in trunk

[Win] Support Per-Monitor (V2) DPI Awareness
https://bugs.webkit.org/show_bug.cgi?id=201450

Reviewed by Don Olmstead.

Source/WebCore:

  • platform/win/GDIUtilities.cpp:

(WebCore::deviceScaleFactorForWindow): Use GetDpiForWindow if available by using soft linking.

Tools:

  • MiniBrowser/win/MainWindow.cpp:

(MainWindow::WndProc): Set the window position suggested by WM_DPICHANGED.

  • MiniBrowser/win/WinMain.cpp:

(wWinMain): Call SetProcessDpiAwarenessContext by using soft linking.

6:45 PM Changeset in webkit [249558] by Fujii Hironori
  • 2 edits in trunk/Source/WebKitLegacy/win

[WinCairo][WK1] ASSERTION FAILED: !isAcceleratedCompositing() in WebView::updateBackingStore
https://bugs.webkit.org/show_bug.cgi?id=201493

Reviewed by Don Olmstead.

WebView::paint switches painting code path by checking accelerated
compositing enabled. However, accelerated compositing could be
enabled after the checking.

WebView::paint called WebView::paintCompositedContentToHDC and
WebCore::Page::updateRendering.
WebView::paintCompositedContentToHDC checks accelerated
compositing enabled. Then, WebCore::Page::updateRendering executes
JavaScript and accelerated compositing could be enabled.

  • WebView.cpp:

(WebView::paint): Call WebCore::Page::updateRendering the
beginning of the function.

6:07 PM Changeset in webkit [249557] by Chris Dumez
  • 5 edits in trunk

REGRESSION: http/tests/adClickAttribution/second-attribution-converted-with-higher-priority.html and http/tests/adClickAttribution/second-attribution-converted-with-lower-priority.html are flaky timeouts
https://bugs.webkit.org/show_bug.cgi?id=201440
<rdar://problem/54998427>

Reviewed by Alex Christensen.

Source/WebKit:

Add private WKBundlePage API to query if the WebPage is currently suspended or not.

  • WebProcess/InjectedBundle/API/c/WKBundlePage.cpp:

(WKBundlePageIsSuspended):

  • WebProcess/InjectedBundle/API/c/WKBundlePagePrivate.h:

Tools:

Since r249350, it is now possible for a WebContent process to have a suspended WebPage for the current web view.
This was confusing InjectedBundle::page() in WebKitTestRunner which was assuming that the WebPage that was first
created in the WebContent process is the one that is currently running the test and that we want to talk to.
Update InjectedBundle::page() to ignore suspended WebPages in order to address the issue.

  • WebKitTestRunner/InjectedBundle/InjectedBundle.cpp:

(WTR::InjectedBundle::page const):

5:03 PM Changeset in webkit [249556] by mark.lam@apple.com
  • 12 edits in trunk/Source

Refactor the Gigacage code to require less pointer casting.
https://bugs.webkit.org/show_bug.cgi?id=201521

Reviewed by Saam Barati.

Source/bmalloc:

  1. Define a Gigacage::Config struct instead of hacking around a g_gigacageBasePtrs array of bytes.
  2. Change Gigacage::basePtr() to return a pointer instead of a reference to the requested basePtr. Instead, make it explicit when the client is trying to take the address of the basePtr, or setting it.
  3. Renamed wasEnabled() to isEnabled() because it returns the present state of the flag, not some past state.
  • bmalloc/Gigacage.cpp:

(Gigacage::bmalloc::protectGigacageBasePtrs):
(Gigacage::bmalloc::unprotectGigacageBasePtrs):
(Gigacage::bmalloc::runwaySize):
(Gigacage::ensureGigacage):
(Gigacage::disablePrimitiveGigacage):
(Gigacage::addPrimitiveDisableCallback):
(Gigacage::primitiveGigacageDisabled):

  • bmalloc/Gigacage.h:

(Gigacage::name):
(Gigacage::Config::basePtr const):
(Gigacage::Config::setBasePtr):
(Gigacage::isEnabled):
(Gigacage::basePtr):
(Gigacage::addressOfBasePtr):
(Gigacage::size):
(Gigacage::caged):
(Gigacage::wasEnabled): Deleted.
(Gigacage::setWasEnabled): Deleted.
(Gigacage::basePtrs): Deleted.

  • bmalloc/HeapKind.h:

(bmalloc::heapKind):
(bmalloc::isActiveHeapKindAfterEnsuringGigacage):
(bmalloc::mapToActiveHeapKindAfterEnsuringGigacage):

Source/JavaScriptCore:

Change LLInt's loadCagedJSValue() to skip the caging if Gigacage is not enabled
in the build. This allows us to remove the unneeded stubs in WTF Gigacage.h.

  • jit/AssemblyHelpers.h:

(JSC::AssemblyHelpers::cageConditionally):

  • llint/LowLevelInterpreter.asm:
  • llint/LowLevelInterpreter64.asm:
  • runtime/VM.h:

(JSC::VM::gigacageAuxiliarySpace):

Source/WTF:

Remove some unneeded stubs in WTF Gigacage.h.

  • wtf/Gigacage.cpp:
  • wtf/Gigacage.h:

(Gigacage::name):
(Gigacage::isEnabled):
(Gigacage::basePtr): Deleted.
(Gigacage::basePtrs): Deleted.

4:42 PM Changeset in webkit [249555] by Justin Fan
  • 7 edits
    1 add in trunk/LayoutTests

[WebGPU] Fix new test results after adding console warnings
https://bugs.webkit.org/show_bug.cgi?id=201526

Unreviewed test expectation updates.

  • webgpu/buffer-errors-expected.txt:
  • webgpu/map-read-buffers-expected.txt:
  • webgpu/map-write-buffers-expected.txt:
  • webgpu/render-pipelines.html: Fix WHLSL compile error.
  • webgpu/uncaptured-errors-expected.txt: Added.
  • webgpu/whlsl/use-undefined-variable-2-expected.txt:
  • webgpu/whlsl/use-undefined-variable-expected.txt:
4:37 PM Changeset in webkit [249554] by Jonathan Bedard
  • 5 edits in trunk/Tools

results.webkit.org: Assign notifyRerender before calling
https://bugs.webkit.org/show_bug.cgi?id=201510

Reviewed by Aakash Jain.

Timelines should define notifyRerender to an empty function,
rename notifiyRerender to notifyRerender.

  • resultsdbpy/resultsdbpy/view/static/js/timeline.js:
  • resultsdbpy/resultsdbpy/view/static/library/js/components/TimelineComponents.js:
  • resultsdbpy/resultsdbpy/view/templates/search.html:
  • resultsdbpy/resultsdbpy/view/templates/suite_results.html:
4:21 PM Changeset in webkit [249553] by Ryan Haddad
  • 3 edits in trunk/LayoutTests

Layout Test imported/w3c/web-platform-tests/service-workers/service-worker/ready.https.html is a flaky failure
https://bugs.webkit.org/show_bug.cgi?id=181837

Unreviewed test gardening.

  • platform/ios/TestExpectations: Mark test as flaky.
  • platform/mac/TestExpectations: Ditto.
4:21 PM Changeset in webkit [249552] by ysuzuki@apple.com
  • 4 edits in trunk/Source/JavaScriptCore

Unreviewed, follow-up after r249530 and r249509
https://bugs.webkit.org/show_bug.cgi?id=201495

Rename FTLOutput::weakPointer to alreadyRegisteredWeakPointer and alreadyRegisteredFrozenPointer.

  • builtins/PromiseConstructor.js:

(nakedConstructor.Promise.resolve):
(nakedConstructor.Promise.reject):
(nakedConstructor.Promise):
(nakedConstructor.InternalPromise.resolve):
(nakedConstructor.InternalPromise.reject):
(nakedConstructor.InternalPromise):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::weakPointer):
(JSC::FTL::DFG::LowerDFGToB3::frozenPointer):
(JSC::FTL::DFG::LowerDFGToB3::weakStructure):

  • ftl/FTLOutput.h:

(JSC::FTL::Output::alreadyRegisteredWeakPointer):
(JSC::FTL::Output::alreadyRegisteredFrozenPointer):
(JSC::FTL::Output::weakPointer): Deleted.

4:16 PM Changeset in webkit [249551] by Kocsen Chung
  • 3 edits
    2 adds in branches/safari-608.2.11.1-branch

Cherry-pick r249534. rdar://problem/55084674

AX: children cache are not re-computed if tab index is removed
https://bugs.webkit.org/show_bug.cgi?id=201502

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: accessibility/tabindex-removed.html

If the tabindex changes, it can potentially affect whether an element is accessible. If we don't update the children cache
information can be stale and lead to incorrect navigation with VoiceOver.

  • accessibility/AXObjectCache.cpp: (WebCore::AXObjectCache::handleAttributeChange):

LayoutTests:

  • accessibility/tabindex-removed-expected.txt: Added.
  • accessibility/tabindex-removed.html: Added.

git-svn-id: https://svn.webkit.org/repository/webkit/trunk@249534 268f45cc-cd09-0410-ab3c-d52691b4dbfc

4:08 PM Changeset in webkit [249550] by Kocsen Chung
  • 8 edits
    2 adds in branches/safari-608-branch

Apply patch. rdar://problem/55001140

Treat a two-finger single tap as if the user tapped with the Cmd key pressed https://bugs.webkit.org/show_bug.cgi?id=201420 <rdar://problem/53207786>

Reviewed by Simon Fraser.

Source/WebKit:

Set the metaKey modifier to true when generating a click event based on a two-finger single tap. This is important so that sites like Google and DuckDuckGo
correctly open links in their search results page in a new tab in Safari on iOS. This currently doesn't work because those sites will call preventDefault()
if the metaKey flag isn't set on a "click" event, and if it is set, they let the browser handle the navigation themselves.

  • UIProcess/ios/WKContentViewInteraction.mm: (-[WKContentView _twoFingerSingleTapGestureRecognized:]):

Tools:

Add a new twoFingerSingleTapAtPoint() method to UIScriptController, which just calls into the existing (but unused)
-[HIDEventGenerator twoFingerTap:completionBlock:].

  • TestRunnerShared/UIScriptContext/Bindings/UIScriptController.idl:
  • TestRunnerShared/UIScriptContext/UIScriptController.h: (WTR::UIScriptController::twoFingerSingleTapAtPoint):
  • WebKitTestRunner/ios/UIScriptControllerIOS.h:
  • WebKitTestRunner/ios/UIScriptControllerIOS.mm: (WTR::UIScriptControllerIOS::twoFingerSingleTapAtPoint):

LayoutTests:

Add a new event that checks that a two-finger single tap on a clickable element yields a "click" event with the metaKey flag set to true.

  • fast/events/ios/click-event-two-finger-single-tap-meta-key-expected.txt: Added.
  • fast/events/ios/click-event-two-finger-single-tap-meta-key.html: Added.
4:03 PM Changeset in webkit [249549] by Jonathan Bedard
  • 2 edits in trunk/Tools

results.webkit.org: Fix prepending search result on empty list
https://bugs.webkit.org/show_bug.cgi?id=201512

Reviewed by Aakash Jain.

  • resultsdbpy/resultsdbpy/view/templates/search.html:
2:58 PM Changeset in webkit [249548] by cturner@igalia.com
  • 11 edits in trunk/Source/WebCore

[EME] Introduce a Proxy CDM for thread-safe access to CDM instances from background decryption threads
https://bugs.webkit.org/show_bug.cgi?id=201339

Reviewed by Xabier Rodriguez-Calvar.

Covered by existing tests.

  • platform/encryptedmedia/CDMInstance.h:
  • platform/encryptedmedia/clearkey/CDMClearKey.cpp:

(WebCore::isolatedKey):
(WebCore::ProxyCDMClearKey::isolatedKeys const):
(WebCore::CDMInstanceClearKey::CDMInstanceClearKey):

  • platform/encryptedmedia/clearkey/CDMClearKey.h:
  • platform/graphics/gstreamer/MediaPlayerPrivateGStreamerBase.cpp:

(WebCore::MediaPlayerPrivateGStreamerBase::cdmInstanceAttached):

  • platform/graphics/gstreamer/eme/WebKitClearKeyDecryptorGStreamer.cpp:

(handleKeyResponse):
(findAndSetKey):
(decrypt):

  • platform/graphics/gstreamer/eme/WebKitCommonEncryptionDecryptorGStreamer.cpp:

(isCDMInstanceAvailable):
(sinkEventHandler):
(setContext):

  • platform/graphics/gstreamer/eme/WebKitCommonEncryptionDecryptorGStreamer.h:
  • testing/MockCDMFactory.cpp:

(WebCore::MockCDMInstance::proxyCDM const):

  • testing/MockCDMFactory.h:
2:48 PM Changeset in webkit [249547] by ysuzuki@apple.com
  • 39 edits
    1 copy
    1 add in trunk/Source/JavaScriptCore

[JSC] Generalize Get/PutPromiseInternalField for InternalFieldObjectImpl
https://bugs.webkit.org/show_bug.cgi?id=201513

Reviewed by Ross Kirsling.

This patch extracts JSPromise's internal fields mechanism as JSInternalFieldsObjectImpl, and make it reusable for the other objects.
It is preparation for using this internal fields mechanism for generators, async functions, async generators, array iterators and so on.

The profiler is telling many recompilation of Generator's resume function (including async generator's one). We are using properties
with private-symbols as a storage for internal state of generators. However, the spec defines that each generator from different generator-functions
has different Prototype?. While we need to share one Generator.prototype.next function, generators tend to have different Structures due to
different Prototype? and accessing internal fields with get_by_id_direct sadly becomes super megamorphic while it is not necessary.
And every time new Structure for new generator pops up, DFG/FTL code for generator resume function gets OSR exit or eventually this function gets
emits super generic code unfortunately. By using internal fields for storing these state, we can avoid this performance problem.

Bytecodes and corresponding DFG nodes are just renamed. JSPromise is now inheriting JSInternalFieldsObjectImpl, which can holds specified
number of internal fields. And op_get_internal_field / op_put_internal_field can access these internal fields.

  • CMakeLists.txt:
  • JavaScriptCore.xcodeproj/project.pbxproj:
  • bytecode/BytecodeList.rb:
  • bytecode/BytecodeUseDef.h:

(JSC::computeUsesForBytecodeOffset):
(JSC::computeDefsForBytecodeOffset):

  • bytecode/CodeBlock.cpp:

(JSC::CodeBlock::finishCreation):

  • bytecode/Opcode.h:
  • bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::emitGetInternalField):
(JSC::BytecodeGenerator::emitPutInternalField):
(JSC::BytecodeGenerator::emitGetPromiseInternalField): Deleted.
(JSC::BytecodeGenerator::emitPutPromiseInternalField): Deleted.

  • bytecompiler/BytecodeGenerator.h:
  • bytecompiler/NodesCodegen.cpp:

(JSC::BytecodeIntrinsicNode::emit_intrinsic_getPromiseInternalField):
(JSC::BytecodeIntrinsicNode::emit_intrinsic_putPromiseInternalField):

  • dfg/DFGAbstractInterpreterInlines.h:

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

  • dfg/DFGByteCodeParser.cpp:

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

  • dfg/DFGCapabilities.cpp:

(JSC::DFG::capabilityLevel):

  • dfg/DFGClobberize.h:

(JSC::DFG::clobberize):

  • dfg/DFGDoesGC.cpp:

(JSC::DFG::doesGC):

  • dfg/DFGFixupPhase.cpp:

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

  • dfg/DFGMayExit.cpp:
  • dfg/DFGNode.h:

(JSC::DFG::Node::hasInternalFieldIndex):
(JSC::DFG::Node::hasHeapPrediction):

  • dfg/DFGNodeType.h:
  • dfg/DFGPredictionPropagationPhase.cpp:
  • dfg/DFGSafeToExecute.h:

(JSC::DFG::safeToExecute):

  • dfg/DFGSpeculativeJIT.cpp:

(JSC::DFG::SpeculativeJIT::compileGetInternalField):
(JSC::DFG::SpeculativeJIT::compilePutInternalField):
(JSC::DFG::SpeculativeJIT::compileCreatePromise):
(JSC::DFG::SpeculativeJIT::compileNewPromise):
(JSC::DFG::SpeculativeJIT::compileGetPromiseInternalField): Deleted.
(JSC::DFG::SpeculativeJIT::compilePutPromiseInternalField): Deleted.

  • dfg/DFGSpeculativeJIT.h:
  • dfg/DFGSpeculativeJIT32_64.cpp:

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

  • dfg/DFGSpeculativeJIT64.cpp:

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

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

(JSC::FTL::canCompile):

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileNode):
(JSC::FTL::DFG::LowerDFGToB3::compileNewPromise):
(JSC::FTL::DFG::LowerDFGToB3::compileCreatePromise):
(JSC::FTL::DFG::LowerDFGToB3::compileGetInternalField):
(JSC::FTL::DFG::LowerDFGToB3::compilePutInternalField):
(JSC::FTL::DFG::LowerDFGToB3::compileGetPromiseInternalField): Deleted.
(JSC::FTL::DFG::LowerDFGToB3::compilePutPromiseInternalField): Deleted.

  • jit/JIT.cpp:

(JSC::JIT::privateCompileMainPass):

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

(JSC::JIT::emit_op_get_internal_field):
(JSC::JIT::emit_op_put_internal_field):
(JSC::JIT::emit_op_get_promise_internal_field): Deleted.
(JSC::JIT::emit_op_put_promise_internal_field): Deleted.

  • jit/JITPropertyAccess32_64.cpp:

(JSC::JIT::emit_op_get_internal_field):
(JSC::JIT::emit_op_put_internal_field):
(JSC::JIT::emit_op_get_promise_internal_field): Deleted.
(JSC::JIT::emit_op_put_promise_internal_field): Deleted.

  • llint/LLIntOffsetsExtractor.cpp:
  • llint/LowLevelInterpreter.asm:
  • llint/LowLevelInterpreter32_64.asm:
  • llint/LowLevelInterpreter64.asm:
  • runtime/JSInternalFieldObjectImpl.h: Copied from Source/JavaScriptCore/runtime/JSPromise.h.

(JSC::JSInternalFieldObjectImpl::allocationSize):
(JSC::JSInternalFieldObjectImpl::internalField const):
(JSC::JSInternalFieldObjectImpl::internalField):
(JSC::JSInternalFieldObjectImpl::offsetOfInternalFields):
(JSC::JSInternalFieldObjectImpl::offsetOfInternalField):
(JSC::JSInternalFieldObjectImpl::JSInternalFieldObjectImpl):

  • runtime/JSInternalFieldObjectImplInlines.h: Added.

(JSC::JSInternalFieldObjectImpl<passedNumberOfInternalFields>::visitChildren):

  • runtime/JSPromise.cpp:

(JSC::JSPromise::finishCreation):
(JSC::JSPromise::visitChildren):
(JSC::JSPromise::status const):
(JSC::JSPromise::result const):
(JSC::JSPromise::isHandled const):

  • runtime/JSPromise.h:

(JSC::JSPromise::allocationSize): Deleted.
(JSC::JSPromise::offsetOfInternalFields): Deleted.
(JSC::JSPromise::offsetOfInternalField): Deleted.
(): Deleted.

2:44 PM Changeset in webkit [249546] by Jonathan Bedard
  • 4 edits in trunk/Tools

results.webkit.org: Retain certain parameters when following menu links
https://bugs.webkit.org/show_bug.cgi?id=201463

Rubber-stamped by Aakash Jain.

  • resultsdbpy/resultsdbpy/view/site_menu.py:

(SiteMenu.add_endpoint): Allow added endpoints to define parameters to be retained.

  • resultsdbpy/resultsdbpy/view/templates/base.html: When clicking on a site-menu link,

retain the specified parameters

  • resultsdbpy/resultsdbpy/view/view_routes.py:

(ViewRoutes.init):

2:29 PM Changeset in webkit [249545] by Alan Bujtas
  • 9 edits in trunk/Source/WebCore

[LFC] LayoutState should not need the initial containing block
https://bugs.webkit.org/show_bug.cgi?id=201511
<rdar://problem/55079241>

Reviewed by Antti Koivisto.

Throw-away layouts should just be able to initialize a dedicated LayoutState without passing in the ICB.

  • layout/LayoutState.cpp:

(WebCore::Layout::LayoutState::markNeedsUpdate):
(WebCore::Layout::LayoutState::run):
(WebCore::Layout::LayoutState::LayoutState): Deleted.

  • layout/LayoutState.h:

(WebCore::Layout::LayoutState::initialContainingBlock const): Deleted.

  • layout/Verification.cpp:

(WebCore::Layout::LayoutState::verifyAndOutputMismatchingLayoutTree const):

  • layout/floats/FloatingContext.cpp: Fix a (very sad)typo.

(WebCore::Layout::FloatingContext::constraints const):

  • layout/layouttree/LayoutTreeBuilder.cpp:

(WebCore::Layout::printLayoutTreeForLiveDocuments):

2:19 PM Changeset in webkit [249544] by russell_e@apple.com
  • 3 edits in trunk/LayoutTests

Layout Test media/audio-concurrent-supported.html is a flaky failure.
rdar://55081191

Unreviewed Test Gardening.

  • platform/ios/TestExpectations:
  • platform/mac/TestExpectations:
2:13 PM Changeset in webkit [249543] by commit-queue@webkit.org
  • 2 edits in trunk/Source/JavaScriptCore

Unreviewed, rolling out r247463.
https://bugs.webkit.org/show_bug.cgi?id=201515

JetStream2 code-load related regression (Requested by
yusukesuzuki on #webkit).

Reverted changeset:

"Keyword lookup can use memcmp to get around unaligned load
undefined behavior"
https://bugs.webkit.org/show_bug.cgi?id=199650
https://trac.webkit.org/changeset/247463

2:12 PM Changeset in webkit [249542] by Chris Dumez
  • 62 edits in trunk

[WK2] Add proper layout test support for ephemeral sessions
https://bugs.webkit.org/show_bug.cgi?id=201475

Reviewed by Antti Koivisto.

LayoutTests/imported/w3c:

  • IndexedDB-private-browsing/support.js:

Tools:

Add proper layout test support for ephemeral sessions via:
<!-- webkit-test-runner [ useEphemeralSession=true ] -->

This actually creates a new WebView using an ephemeral data store to load the test which
is *true* private browsing.

testRunner.setPrivateBrowsingEnabled() merely overrides the Page's sessionID in the WebContent
process, which is a hack and does not truly enable private browsing because the UIProcess and
the NetworkProcess are not properly updated on WK2. This patch renames testRunner.setPrivateBrowsingEnabled()
to testRunner.setPrivateBrowsingEnabled_DEPRECATED() to discourage its use since I am try to
get rid of it.

Note that WK2 does not support using a WebContent process for different sessions / data stores.
Each WebContent process has a single associated data store and testRunner.setPrivateBrowsingEnabled()
is currently the only exception to that. This leads to additional complexity in our code base (e.g. in ITP),
simply to support testRunner.setPrivateBrowsingEnabled().

  • DumpRenderTree/TestOptions.cpp:

(TestOptions::TestOptions):

  • DumpRenderTree/TestOptions.h:
  • DumpRenderTree/TestRunner.cpp:

(TestRunner::staticFunctions):

  • DumpRenderTree/mac/DumpRenderTree.mm:

(setWebPreferencesForTestOptions):

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

(WTR::TestRunner::setPrivateBrowsingEnabled_DEPRECATED):

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

(WTR::TestController::generatePageConfiguration):
(WTR::updateTestOptionsFromTestHeader):

  • WebKitTestRunner/TestOptions.h:

(WTR::TestOptions::hasSameInitializationOptions const):

  • WebKitTestRunner/cocoa/TestControllerCocoa.mm:

(WTR::TestController::platformCreateWebView):

LayoutTests:

Update existing layout tests to either switch to the new mechanism when trivial or to rename
setPrivateBrowsingEnabled() to setPrivateBrowsingEnabled_DEPRECATED() when it is not.

  • crypto/subtle/rsa-indexeddb-non-exportable-private.html:
  • crypto/subtle/rsa-indexeddb-private.html:
  • http/tests/adClickAttribution/conversion-disabled-in-ephemeral-session.html:
  • http/tests/adClickAttribution/resources/util.js:

(tearDownAndFinish):

  • http/tests/adClickAttribution/store-disabled-in-ephemeral-session.html:
  • http/tests/cache-storage/cache-persistency.https.html:
  • http/tests/cache-storage/cache-representation.https.html:
  • http/tests/cache/disk-cache/disk-cache-vary-cookie.html:
  • http/tests/cookies/private-cookie-storage.html:
  • http/tests/media/now-playing-info-private-browsing.html:
  • http/tests/security/appcache-in-private-browsing.html:
  • http/tests/security/appcache-switching-private-browsing.html:
  • http/tests/security/contentSecurityPolicy/report-cross-origin-no-cookies-when-private-browsing-enabled-expected.txt:
  • http/tests/security/contentSecurityPolicy/report-cross-origin-no-cookies-when-private-browsing-enabled.php:
  • http/tests/security/contentSecurityPolicy/report-cross-origin-no-cookies-when-private-browsing-toggled.php:
  • http/tests/security/contentSecurityPolicy/report-same-origin-no-cookies-when-private-browsing-toggled.php:
  • http/tests/security/contentSecurityPolicy/report-same-origin-with-cookies-when-private-browsing-enabled-expected.txt:
  • http/tests/security/contentSecurityPolicy/report-same-origin-with-cookies-when-private-browsing-enabled.php:
  • http/tests/security/cookies/resources/first-party-cookie-allow.xsl:
  • http/tests/security/cookies/resources/third-party-cookie-blocking.xsl:
  • http/tests/security/cookies/third-party-cookie-blocking-main-frame.html:
  • http/tests/security/cookies/third-party-cookie-blocking-redirect.html:
  • http/tests/security/cookies/third-party-cookie-blocking-user-action.html:
  • http/tests/security/cookies/third-party-cookie-blocking.html:
  • http/tests/security/cross-frame-access-private-browsing.html:
  • http/tests/security/cross-origin-plugin-private-browsing-toggled-allowed.html:
  • http/tests/security/cross-origin-plugin-private-browsing-toggled.html:
  • http/tests/security/private-browsing-http-auth.html:
  • http/tests/security/storage-blocking-loosened-private-browsing-plugin.html:
  • http/tests/security/storage-blocking-strengthened-private-browsing-plugin.html:
  • http/tests/security/strip-referrer-to-origin-for-third-party-redirects-in-private-mode.html:
  • http/tests/security/strip-referrer-to-origin-for-third-party-requests-in-private-mode.html:
  • http/tests/storageAccess/has-storage-access-true-if-feature-off.html:
  • http/tests/workers/service/resources/basic-register.js:

(async.test):

  • http/tests/workers/service/serviceworker-private-browsing.https.html:
  • http/wpt/fetch/dnt-header-after-redirection.html:
  • platform/wk2/TestExpectations:
  • plugins/private-browsing-mode-2.html:
  • plugins/private-browsing-mode.html:
  • resources/js-test.js:
  • storage/domstorage/localstorage/private-browsing-affects-storage.html:
  • storage/domstorage/localstorage/resources/private-browsing-1.html:
  • storage/domstorage/localstorage/resources/private-browsing-2.html:
  • storage/indexeddb/open-db-private-browsing.html:
  • storage/indexeddb/resources/open-db-private-browsing.js:
  • storage/websql/private-browsing-noread-nowrite.html:
  • storage/websql/private-browsing-open-disabled.html:
2:01 PM Changeset in webkit [249541] by Kocsen Chung
  • 1 copy in tags/Safari-609.1.3.2

Tag Safari-609.1.3.2.

1:55 PM Changeset in webkit [249540] by Kocsen Chung
  • 7 edits in tags/Safari-609.1.3.1/Source

Versioning.

12:21 PM Changeset in webkit [249539] by Justin Fan
  • 22 edits
    4 copies
    1 add in trunk

[WebGPU] Implement GPUUncapturedErrorEvent
https://bugs.webkit.org/show_bug.cgi?id=199676

Reviewed by Dean Jackson.

Source/WebCore:

Implement GPUUncapturedErrorEvent and "uncapturederror" event name.
Add the onuncapturederror EventHandler attribute to GPUDevice.

Test: webgpu/uncaptured-errors.html

  • CMakeLists.txt:
  • DerivedSources-input.xcfilelist:
  • DerivedSources-output.xcfilelist:
  • DerivedSources.make:
  • Modules/webgpu/GPUUncapturedErrorEvent.cpp:

(WebCore::GPUUncapturedErrorEvent::create):
(WebCore::GPUUncapturedErrorEvent::GPUUncapturedErrorEvent):
(WebCore::GPUUncapturedErrorEvent::eventInterface const):

  • Modules/webgpu/GPUUncapturedErrorEvent.h:
  • Modules/webgpu/GPUUncapturedErrorEvent.idl:
  • Modules/webgpu/WebGPUAdapter.cpp: Must now provide ScriptExecutionContext to any created GPUDevice.

(WebCore::WebGPUAdapter::requestDevice const):

  • Modules/webgpu/WebGPUAdapter.h:
  • Modules/webgpu/WebGPUAdapter.idl:
  • Modules/webgpu/WebGPUDevice.cpp: Is now an EventTarget.

(WebCore::WebGPUDevice::tryCreate):
(WebCore::WebGPUDevice::WebGPUDevice):
(WebCore::printValidationErrorToConsole):
(WebCore::WebGPUDevice::dispatchUncapturedError): Events should only be fired from the main thread.

  • Modules/webgpu/WebGPUDevice.h:
  • Modules/webgpu/WebGPUDevice.idl:
  • Modules/webgpu/WebGPUDeviceEventHandler.idl:
  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • bindings/js/WebCoreBuiltinNames.h:
  • dom/EventNames.h:
  • dom/EventNames.in:
  • dom/EventTargetFactory.in:
  • platform/graphics/gpu/GPUError.cpp: GPUErrors can only be created internally; creation should never fail.

(WebCore::createError):

  • platform/graphics/gpu/GPUError.h:
  • platform/graphics/gpu/GPUErrorScopes.cpp:

(WebCore::GPUErrorScopes::create):
(WebCore::GPUErrorScopes::GPUErrorScopes):
(WebCore::GPUErrorScopes::generateError): Use a callback for now, since GPUErrorScopes is still under platform.

  • platform/graphics/gpu/GPUErrorScopes.h:

(WebCore::GPUErrorScopes::create): Deleted.

LayoutTests:

Add a test to ensure GPUUncapturedErrorEvents work.

  • webgpu/uncaptured-errors.html: Added.
11:47 AM Changeset in webkit [249538] by Tadeu Zagallo
  • 5 edits
    1 add in trunk

LazyClassStructure::setConstructor should not store the constructor to the global object
https://bugs.webkit.org/show_bug.cgi?id=201484
<rdar://problem/50400451>

Reviewed by Yusuke Suzuki.

JSTests:

  • stress/web-assembly-constructors-should-not-override-global-object-property.js: Added.

Source/JavaScriptCore:

LazyClassStructure::setConstructor sets the constructor as a property of the global object.
This became a problem when it started being used for WebAssembly constructors, such as Module
and Instance, since they are properties of the WebAssembly object, not the global object. That
resulted in properties of the global object replaced whenever a lazy WebAssembly constructor
was first accessed. e.g.

globalThis.Module = x;
WebAssembly.Module;
globalThis.Module === WebAssembly.Module;

  • runtime/LazyClassStructure.cpp:

(JSC::LazyClassStructure::Initializer::setConstructor):

  • runtime/LazyClassStructure.h:
  • runtime/Lookup.h:

(JSC::reifyStaticProperty):

11:29 AM Changeset in webkit [249537] by Ryan Haddad
  • 2 edits in trunk/LayoutTests

[iOS] Layout Test imported/w3c/web-platform-tests/html/semantics/embedded-content/the-video-element/video_timeupdate_on_seek.html is failing
https://bugs.webkit.org/show_bug.cgi?id=201509

Unreviewed test gardening.

  • platform/ios-simulator/TestExpectations: Mark test as failing.
11:03 AM Changeset in webkit [249536] by Antti Koivisto
  • 6 edits in trunk

Generate event region for both the main graphics layer and the scrolled contents layer
https://bugs.webkit.org/show_bug.cgi?id=201487

Reviewed by Simon Fraser.

Source/WebCore:

We currently generate region for one of them only. With borders both need it.

Also share more code with painting paths.

  • rendering/RenderLayerBacking.cpp:

(WebCore::RenderLayerBacking::updateEventRegion):

Generate region for both layers.

(WebCore::RenderLayerBacking::paintIntoLayer):

Add event region context parameter so we can use this function for event region update too.

  • rendering/RenderLayerBacking.h:

LayoutTests:

  • pointerevents/ios/touch-action-region-backing-sharing-expected.txt:
  • pointerevents/ios/touch-action-region-overflow-expected.txt:
10:44 AM Changeset in webkit [249535] by rniwa@webkit.org
  • 8 edits
    3 adds in trunk

REGRESSION (iOS 13): Bulleted list copied from Notes to Mail results in Times New Roman
https://bugs.webkit.org/show_bug.cgi?id=201490

Reviewed by Daniel Bates.

Source/WebCore:

The bug was caused by an element in the pasted content not having any explicit font name resolving
to use the font-family value of -webkit-standard. When such an inline style is inserted into
Mail's WKWebView which sets a different font family, ReplaceSelectionCommand would fail to strip away,
making the pasted content using the default font family of Times New Roman.

Fixed the bug by stripping away font-family set to -webkit-standard in the sanitization document
since that's indicative of the pasted content not having any font family being specified.

In the future, we should consider making regular copy (as opposed to the copy for sanitization)
resolve generic font family names to concrete font names since different WKWebView might be using
different concrete font names. Unfortuantely, such a change is quite involved and risky since various
paste side code in EditingStyle that removes redundant inline styles (i.e. redundant font-family)
need to be aware of this special font family resolution.

Tests: editing/pasteboard/paste-cocoa-writer-markup-with-webkit-standard-font-family.html

PasteHTML.DoesNotAddStandardFontFamily

  • editing/EditingStyle.cpp:

(WebCore::EditingStyle::mergeInlineAndImplicitStyleOfElement):
(WebCore::EditingStyle::wrappingStyleForSerialization):
(WebCore::familyNameFromCSSPrimitiveValue): Added.
(WebCore::loneFontFamilyName): Extracted from usesForbiddenSystemFontAsOnlyFontFamilyName. Fixed
a bug that it was not handling the case when font-family property's value is a CSSPrimitiveValue
instead of a CSSValueList.
(WebCore::usesForbiddenSystemFontAsOnlyFontFamilyName): Deleted.
(WebCore::EditingStyle::mergeStyleFromRulesForSerialization): Remove font-family property when
StandardFontFamilySerializationMode::Strip is specified and its value is -webkit-standard.

  • editing/EditingStyle.h:
  • editing/markup.cpp:

(WebCore::StyledMarkupAccumulator::StyledMarkupAccumulator): Added
StandardFontFamilySerializationMode as an argument.
(WebCore::StyledMarkupAccumulator::appendStartTag):
(WebCore::StyledMarkupAccumulator::serializeNodes):
(WebCore::serializePreservingVisualAppearanceInternal): Ditto.
(WebCore::serializePreservingVisualAppearance): Use StandardFontFamilySerializationMode::Keep
to preserve the pre-existing behavior.
(WebCore::sanitizedMarkupForFragmentInDocument): Use StandardFontFamilySerializationMode::Strip
as this is the code used by sanitization code.

Tools:

Added a test.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/WebKitCocoa/PasteHTML.mm:
  • TestWebKitAPI/Tests/WebKitCocoa/cocoa-writer-markup-with-lists.html: Added.

LayoutTests:

Added a test to make sure -webkit-standard font family name isn't stripped away when sanitization is not in effect.

  • editing/pasteboard/paste-cocoa-writer-markup-with-webkit-standard-font-family-expected.txt: Added.
  • editing/pasteboard/paste-cocoa-writer-markup-with-webkit-standard-font-family.html: Added.
10:27 AM Changeset in webkit [249534] by Chris Fleizach
  • 3 edits
    2 adds in trunk

AX: children cache are not re-computed if tab index is removed
https://bugs.webkit.org/show_bug.cgi?id=201502

Reviewed by Zalan Bujtas.

Source/WebCore:

Test: accessibility/tabindex-removed.html

If the tabindex changes, it can potentially affect whether an element is accessible. If we don't update the children cache
information can be stale and lead to incorrect navigation with VoiceOver.

  • accessibility/AXObjectCache.cpp:

(WebCore::AXObjectCache::handleAttributeChange):

LayoutTests:

  • accessibility/tabindex-removed-expected.txt: Added.
  • accessibility/tabindex-removed.html: Added.
10:11 AM Changeset in webkit [249533] by sihui_liu@apple.com
  • 10 edits in trunk/Source/WebKit

Remove StorageArea pointers in StorageManagerSet when removing StorageManagers that own them
https://bugs.webkit.org/show_bug.cgi?id=201324

Reviewed by Alex Christensen.

  • NetworkProcess/WebStorage/LocalStorageNamespace.cpp:

(WebKit::LocalStorageNamespace::storageAreaIdentifiers const):

  • NetworkProcess/WebStorage/LocalStorageNamespace.h:
  • NetworkProcess/WebStorage/SessionStorageNamespace.cpp:

(WebKit::SessionStorageNamespace::storageAreaIdentifiers const):

  • NetworkProcess/WebStorage/SessionStorageNamespace.h:
  • NetworkProcess/WebStorage/StorageManager.cpp:

(WebKit::StorageManager::allStorageAreaIdentifiers const):

  • NetworkProcess/WebStorage/StorageManager.h:
  • NetworkProcess/WebStorage/StorageManagerSet.cpp:

(WebKit::StorageManagerSet::remove):

  • NetworkProcess/WebStorage/TransientLocalStorageNamespace.cpp:

(WebKit::TransientLocalStorageNamespace::storageAreaIdentifiers const):

  • NetworkProcess/WebStorage/TransientLocalStorageNamespace.h:
10:01 AM Changeset in webkit [249532] by Michael Catanzaro
  • 1 edit in trunk/Tools/Scripts/webkitpy/common/config/contributors.json

Unreviewed, update contributors.json

People are having trouble CCing me on Bugzilla.

I'm still at Igalia, but no longer working on WebKit.

  • Scripts/webkitpy/common/config/contributors.json:
9:56 AM Changeset in webkit [249531] by Konstantin Tokarev
  • 3 edits in trunk/Source/WebCore

Don't call PlatformMediaSessionManager methods when neither VIDEO nor WEB_AUDIO is enabled
https://bugs.webkit.org/show_bug.cgi?id=201508

Reviewed by Jer Noble.

Definitions of these methods are guarded with #if ENABLE(VIDEO)
ENABLE(WEB_AUDIO),

so calling them causes linking errors.

  • testing/InternalSettings.cpp:

(WebCore::InternalSettings::Backup::Backup):
(WebCore::InternalSettings::Backup::restoreTo):
(WebCore::InternalSettings::setShouldDeactivateAudioSession):

  • testing/Internals.cpp:

(WebCore::Internals::resetToConsistentState):
(WebCore::Internals::processWillSuspend):
(WebCore::Internals::processDidResume):
(WebCore::Internals::setIsPlayingToAutomotiveHeadUnit):

8:51 AM Changeset in webkit [249530] by ysuzuki@apple.com
  • 3 edits
    1 add in trunk

[JSC] Do not use FTLOutput::weakPointer directly
https://bugs.webkit.org/show_bug.cgi?id=201495

Reviewed by Filip Pizlo.

JSTests:

  • stress/create-promise-weak-pointer.js: Added.

(foo):

Source/JavaScriptCore:

FTLOutput::weakPointer does not register the cell as a weak pointer.
CreatePromise's implementation is accidentally using m_out.weakPointer and hits the debug assertion.
While the current implementation is not posing correctness issue since these cells are live so long as JSGlobalObject is live,
and we register JSGlobalObject as a weakPointer, we should always use FTLLowerDFGToB3's helper function.
For FrozenValue, we should use frozenPointer helper function.

  • ftl/FTLLowerDFGToB3.cpp:

(JSC::FTL::DFG::LowerDFGToB3::compileCreatePromise):
(JSC::FTL::DFG::LowerDFGToB3::compileNewArrayBuffer):

7:48 AM Changeset in webkit [249529] by Carlos Garcia Campos
  • 2 edits in trunk/Source/WebKit

REGRESSION(r249275): [GTK][WPE] wrong page ID returned by webkit_web_view_get_page_id()
https://bugs.webkit.org/show_bug.cgi?id=201504

Reviewed by Michael Catanzaro.

We want to return the page identifier not the web page proxy id.

  • UIProcess/API/glib/WebKitWebView.cpp:

(webkit_web_view_get_page_id):

7:21 AM Changeset in webkit [249528] by Pablo Saavedra
  • 2 edits in trunk/Source/WebKit

Build failure after r249501 for ports with SERVICE_WORKER set OFF
https://bugs.webkit.org/show_bug.cgi?id=201503

Reviewed by Youenn Fablet.

  • WebProcess/Network/WebLoaderStrategy.cpp:

(WebKit::WebLoaderStrategy::scheduleLoad):

3:53 AM Changeset in webkit [249527] by Keith Rollin
  • 2 edits in trunk/Source/WebCore

Fix implicit conversion that loses precision
https://bugs.webkit.org/show_bug.cgi?id=201476
<rdar://problem/55043129>

Reviewed by Youenn Fablet, Alex Christensen.

Building for watchOS results in the following error:

.../Source/WebCore/fileapi/NetworkSendQueue.cpp:66:42: error: implicit conversion loses integer precision: 'unsigned long long' to 'unsigned int' [-Werror,-Wshorten-64-to-32]

enqueue(JSC::ArrayBuffer::create(byteLength, 1), 0, 0);

Fix this by replacing byteLength (which is known to be zero at this
point) with 0U.

No new tests -- no new or changed functionality.

  • fileapi/NetworkSendQueue.cpp:

(WebCore::NetworkSendQueue::enqueue):

2:01 AM Changeset in webkit [249526] by youenn@apple.com
  • 17 edits
    2 adds in trunk

Introduce WorkerSWClientConnection to make SWClientConnection mono-thread
https://bugs.webkit.org/show_bug.cgi?id=201430

Reviewed by Alex Christensen.

LayoutTests/imported/w3c:

Remove inserted iframes to make test outputs more stable.

  • web-platform-tests/service-workers/service-worker/activation.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/activation.https.html:
  • web-platform-tests/service-workers/service-worker/detached-context.https-expected.txt:
  • web-platform-tests/service-workers/service-worker/detached-context.https.html:

Source/WebCore:

Make SWClientConnection RefCounted.
Add a WorkerSWClientConnection dedicated to be used by worker/service worker contexts.
Simplify ServiceWorkerContainer logic based on this refactoring.

Covered by existing tests.

  • Sources.txt:
  • WebCore.xcodeproj/project.pbxproj:
  • workers/WorkerGlobalScope.cpp:

(WebCore::WorkerGlobalScope::swClientConnection):

  • workers/WorkerGlobalScope.h:
  • workers/service/SWClientConnection.cpp:

(WebCore::SWClientConnection::scheduleJob):
(WebCore::SWClientConnection::failedFetchingScript):
(WebCore::SWClientConnection::postTaskForJob):

  • workers/service/SWClientConnection.h:
  • workers/service/ServiceWorkerContainer.cpp:

(WebCore::mainThreadConnection):
(WebCore::ServiceWorkerContainer::ready):
(WebCore::ServiceWorkerContainer::scheduleJob):
(WebCore::ServiceWorkerContainer::getRegistration):
(WebCore::ServiceWorkerContainer::updateRegistrationState):
(WebCore::ServiceWorkerContainer::getRegistrations):
(WebCore::ServiceWorkerContainer::notifyRegistrationIsSettled):
(WebCore::ServiceWorkerContainer::jobFinishedLoadingScript):
(WebCore::ServiceWorkerContainer::notifyFailedFetchingScript):
(WebCore::ServiceWorkerContainer::ensureSWClientConnection):
(WebCore::ServiceWorkerContainer::stop):

  • workers/service/ServiceWorkerContainer.h:
  • workers/service/ServiceWorkerFetchResult.h:

(WebCore::ServiceWorkerFetchResult::isolatedCopy const):

  • workers/service/WorkerSWClientConnection.cpp: Added.

(WebCore::WorkerSWClientConnection::WorkerSWClientConnection):
(WebCore::WorkerSWClientConnection::~WorkerSWClientConnection):
(WebCore::WorkerSWClientConnection::matchRegistration):
(WebCore::WorkerSWClientConnection::getRegistrations):
(WebCore::WorkerSWClientConnection::whenRegistrationReady):
(WebCore::WorkerSWClientConnection::addServiceWorkerRegistrationInServer):
(WebCore::WorkerSWClientConnection::removeServiceWorkerRegistrationInServer):
(WebCore::WorkerSWClientConnection::didResolveRegistrationPromise):
(WebCore::WorkerSWClientConnection::postMessageToServiceWorker):
(WebCore::WorkerSWClientConnection::serverConnectionIdentifier const):
(WebCore::WorkerSWClientConnection::mayHaveServiceWorkerRegisteredForOrigin const):
(WebCore::WorkerSWClientConnection::syncTerminateWorker):
(WebCore::WorkerSWClientConnection::registerServiceWorkerClient):
(WebCore::WorkerSWClientConnection::unregisterServiceWorkerClient):
(WebCore::WorkerSWClientConnection::finishFetchingScriptInServer):
(WebCore::WorkerSWClientConnection::isThrottleable const):
(WebCore::WorkerSWClientConnection::updateThrottleState):
(WebCore::WorkerSWClientConnection::scheduleJob):
(WebCore::WorkerSWClientConnection::scheduleJobInServer):

  • workers/service/WorkerSWClientConnection.h: Added.
12:47 AM Changeset in webkit [249525] by rniwa@webkit.org
  • 2 edits
    1 delete in trunk/Tools

Remove unused SPI that accesses MemoryCache directly
https://bugs.webkit.org/show_bug.cgi?id=201468

Reviewed by Youenn Fablet.

Removed the API test.

  • TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
  • TestWebKitAPI/Tests/ios/MemoryCacheAddImageToCacheIOS.mm: Removed.
Note: See TracTimeline for information about the timeline view.