Context Navigation

← Previous Period
Next Period →

Timeline

Dec 22, 2016:

8:41 PM Changeset in webkit [210130] by aestes@apple.com

15 edits in trunk/Source

Reduce QuickLook.h include overhead
https://bugs.webkit.org/show_bug.cgi?id=166454

Reviewed by Andreas Kling.

Source/WebCore:

dom/Document.cpp: Included QuickLook.h for QLPreviewProtocol().
loader/DocumentLoader.cpp:

(WebCore::DocumentLoader::setQuickLookHandle): Moved from DocumentLoader.h to here.

loader/DocumentLoader.h: Stopped including QuickLook.h and forward-declared

QuickLookHandle.
(WebCore::DocumentLoader::setQuickLookHandle): Moved definition out-of-line since it
requires QuickLookHandle to be a complete type.

loader/ios/QuickLook.h: Updated to use #pragma once. Cleaned up includes and forward

declarations.

platform/network/ResourceHandle.cpp: Included QuickLook.h so that QuickLookHandle is a

complete type in the ResourceHandle constructor.

platform/network/ResourceHandle.h: Stopped including QuickLook.h and forward-declared

QuickLookHandle.
(WebCore::ResourceHandle::setQuickLookHandle): Moved definition out-of-line since it
requires QuickLookHandle to be a complete type.

platform/network/mac/ResourceHandleMac.mm:

(WebCore::ResourceHandle::setQuickLookHandle): Moved from ResourceHandle.h to here.

platform/network/mac/WebCoreResourceHandleAsDelegate.mm: Included QuickLook.h for

QuickLookHandle.

Source/WebKit:

WebCoreSupport/WebResourceLoadScheduler.cpp: Stopped including QuickLook.h.

Source/WebKit2:

WebProcess/Network/WebLoaderStrategy.cpp: Included QuickLook.h for QLPreviewProtocol().
WebProcess/Network/WebResourceLoader.cpp: Included QuickLook.h for QuickLookHandle.
WebProcess/Network/WebResourceLoader.h: Stopped including QuickLook.h.

6:24 PM Changeset in webkit [210129] by aestes@apple.com

2 edits
2 moves
1 add in trunk/Source/WebCore

[iOS] Move QuickLook from WebCore/platform to WebCore/loader
https://bugs.webkit.org/show_bug.cgi?id=166449

Reviewed by Darin Adler.

QuickLook.mm knows about ResourceLoader, so it's a layering violation to have it in
platform/. Move it to loader/ios/.

WebCore.xcodeproj/project.pbxproj:
loader/ios/QuickLook.h: Renamed from Source/WebCore/platform/network/ios/QuickLook.h.
loader/ios/QuickLook.mm: Renamed from Source/WebCore/platform/network/ios/QuickLook.mm.

6:21 PM Changeset in webkit [210128] by Michael Catanzaro

2 edits in trunk/Source/WebCore

Unreviewed, fix unused parameter warning in GLES build

Looks like it's been here since 2012. Impressive!

platform/graphics/opengl/GraphicsContext3DOpenGLES.cpp:

(WebCore::GraphicsContext3D::resolveMultisamplingIfNecessary):

6:14 PM Changeset in webkit [210127] by keith_miller@apple.com

5 edits in trunk

WebAssembly: Make spec-tests/f32.wast.js and spec-tests/f64.wast.js pass
https://bugs.webkit.org/show_bug.cgi?id=166447

Reviewed by Saam Barati.

JSTests:

wasm.yaml:
wasm/wasm.json:

Source/JavaScriptCore:

We needed to treat -0.0 < 0.0 for floating point min/max. For min,
the algorithm works because if a == b then a and b are not NaNs so
either they are the same or they are some zero. When we or a and b
either we get the same number back or we get -0.0. Similarly for
max we use an and and the sign bit gets dropped if one is 0.0 and
the other is -0.0, otherwise, we get the same number back.

wasm/wasm.json:

5:32 PM Changeset in webkit [210126] by sbarati@apple.com

5 edits
1 add in trunk

WebAssembly: Make calling Wasm functions that returns or takes an i64 as a parameter an early exception
https://bugs.webkit.org/show_bug.cgi?id=166437
<rdar://problem/29793949>

Reviewed by Keith Miller.

JSTests:

wasm.yaml:
wasm/function-tests/i64-from-js-exceptions.js: Added.

(const.imp.import.sideEffects):
(assert.throws.instance.exports.foo.valueOf):

Source/JavaScriptCore:

This patch makes it so that we throw an exception before we do
anything else if we call a wasm function that either takes an
i64 as an argument or returns an i64.

wasm/js/WebAssemblyFunction.cpp:

(JSC::callWebAssemblyFunction):
(JSC::WebAssemblyFunction::WebAssemblyFunction):
(JSC::WebAssemblyFunction::call): Deleted.

wasm/js/WebAssemblyFunction.h:

(JSC::WebAssemblyFunction::signatureIndex):
(JSC::WebAssemblyFunction::jsEntrypoint):

5:23 PM Changeset in webkit [210125] by commit-queue@webkit.org

2 edits in trunk/Source/WebInspectorUI

Unreviewed, rolling out r210069.
https://bugs.webkit.org/show_bug.cgi?id=166439

This patch makes all properties in Computed Styles strike-
through when switching from the Rules panel. (Requested by NVI
on #webkit).

Reverted changeset:

"Web Inspector: Uncommenting CSS properties doesn't work for
inline styles"
https://bugs.webkit.org/show_bug.cgi?id=166297
http://trac.webkit.org/changeset/210069

5:19 PM Changeset in webkit [210124] by keith_miller@apple.com

16 edits in trunk

Add BitOr for floating points to B3
https://bugs.webkit.org/show_bug.cgi?id=166446

Reviewed by Saam Barati.

Source/JavaScriptCore:

This patch does some slight refactoring to the ARM assembler,
which groups all the vector floating point instructions together.

assembler/ARM64Assembler.h:

(JSC::ARM64Assembler::vand):
(JSC::ARM64Assembler::vorr):
(JSC::ARM64Assembler::vectorDataProcessingLogical):
(JSC::ARM64Assembler::vectorDataProcessing2Source): Deleted.

assembler/MacroAssemblerARM64.h:

(JSC::MacroAssemblerARM64::orDouble):
(JSC::MacroAssemblerARM64::orFloat):

assembler/MacroAssemblerX86Common.h:

(JSC::MacroAssemblerX86Common::orDouble):
(JSC::MacroAssemblerX86Common::orFloat):

assembler/X86Assembler.h:

(JSC::X86Assembler::orps_rr):

b3/B3ConstDoubleValue.cpp:

(JSC::B3::ConstDoubleValue::bitOrConstant):
(JSC::B3::ConstDoubleValue::bitXorConstant):

b3/B3ConstDoubleValue.h:
b3/B3ConstFloatValue.cpp:

(JSC::B3::ConstFloatValue::bitOrConstant):
(JSC::B3::ConstFloatValue::bitXorConstant):

b3/B3ConstFloatValue.h:
b3/B3LowerToAir.cpp:

(JSC::B3::Air::LowerToAir::lower):

b3/B3Validate.cpp:
b3/air/AirInstInlines.h:

(JSC::B3::Air::Inst::shouldTryAliasingDef):

b3/air/AirOpcode.opcodes:
b3/testb3.cpp:

(JSC::B3::bitOrDouble):
(JSC::B3::testBitOrArgDouble):
(JSC::B3::testBitOrArgsDouble):
(JSC::B3::testBitOrArgImmDouble):
(JSC::B3::testBitOrImmsDouble):
(JSC::B3::bitOrFloat):
(JSC::B3::testBitOrArgFloat):
(JSC::B3::testBitOrArgsFloat):
(JSC::B3::testBitOrArgImmFloat):
(JSC::B3::testBitOrImmsFloat):
(JSC::B3::testBitOrArgsFloatWithUselessDoubleConversion):
(JSC::B3::run):

Websites/webkit.org:

Update docs to indicate it's cool to use bit ops with floating point.

docs/b3/intermediate-representation.html:

5:08 PM WebIDLToDo edited by sam@webkit.org

(diff)

5:00 PM WebIDLToDo edited by sam@webkit.org

(diff)

4:57 PM WebIDLToDo created by sam@webkit.org

Add To Do list for WebIDL improvements. Each of these should get a bug.

4:56 PM WikiStart edited by sam@webkit.org

(diff)

4:32 PM Changeset in webkit [210123] by achristensen@apple.com

3 edits
1 delete in trunk/Source/WebCore

Move GraphicsContext3DWin to GraphicsContext3DOpenGLES
https://bugs.webkit.org/show_bug.cgi?id=166425

Reviewed by Tim Horton.

Move code which is GLES-specific not Windows-specific to the GLES-specific file
so I can use it on Cocoa, with a temporary PLATFORM macro that I plan to remove once
all platforms use ANGLE's GLES implementation. There is another copy of this code in
GraphicsContext3DCairo, which will also be united with this code here soon.
Also remove the PlatformCALayerWin, which was never used for anything.

platform/graphics/opengl/GraphicsContext3DOpenGLES.cpp:

(WebCore::GraphicsContext3D::create):
(WebCore::GraphicsContext3D::GraphicsContext3D):
(WebCore::GraphicsContext3D::~GraphicsContext3D):
(WebCore::GraphicsContext3D::setContextLostCallback):
(WebCore::GraphicsContext3D::setErrorMessageCallback):
(WebCore::GraphicsContext3D::makeContextCurrent):
(WebCore::GraphicsContext3D::checkGPUStatusIfNecessary):
(WebCore::GraphicsContext3D::platformGraphicsContext3D):
(WebCore::GraphicsContext3D::platformTexture):
(WebCore::GraphicsContext3D::isGLES2Compliant):
(WebCore::GraphicsContext3D::platformLayer):

platform/graphics/win/GraphicsContext3DWin.cpp: Removed.
PlatformMac.cmake:

Include the egl directory. I'll need this, too, once Mac starts using ANGLE's EGL implementation.

4:20 PM Changeset in webkit [210122] by Brent Fulgham

5 edits
3 adds in trunk

Nested calls to setDocument can omit firing 'unload' events
https://bugs.webkit.org/show_bug.cgi?id=166422
<rdar://problem/29763012>

Reviewed by Alex Christensen.

Source/WebCore:

Test: fast/loader/nested-document-handling.html

Only allow a single document change to be taking place during a given runloop cycle.

bindings/js/ScriptController.cpp:

(WebCore::ScriptController::executeIfJavaScriptURL): Block script changing the document
when we are in the middle of changing the document.

page/Frame.cpp:

(WebCore::Frame::setDocument): Keep track of document change state.

page/Frame.h:

LayoutTests:

fast/loader/nested-document-handling-expected.txt: Added.
fast/loader/nested-document-handling.html: Added.
fast/loader/resources/subframe-success.html: Added.

3:52 PM Changeset in webkit [210121] by timothy_horton@apple.com

2 edits in trunk/Source/WebCore

TileGrid creates new cohorts even when not using temporarilyRetainTileCohorts mode
https://bugs.webkit.org/show_bug.cgi?id=166440

Reviewed by Simon Fraser.

No new tests.

platform/graphics/ca/TileGrid.cpp:

(WebCore::TileGrid::revalidateTiles):
If we are not temporarily retaining unparented tile cohorts (e.g. on
iOS WebKit2), instead immediately removing all tiles that would go into
the new cohort, avoid starting the cohort at all.

3:48 PM Changeset in webkit [210120] by Alan Bujtas

3 edits
2 adds in trunk

Do not destroy the RenderNamedFlowFragment as leftover anonymous block.
https://bugs.webkit.org/show_bug.cgi?id=166436
rdar://problem/29772233

Reviewed by Simon Fraser.

Source/WebCore:

When as the result of certain style change, the generated anonymous block is not needed anymore, we
move its descendants up to the parent and destroy the generated box. While RenderNamedFlowFragment is a generated
block, the cleanup code should just ignore it the same way we ignore boxes like multicolumn, mathml etc.

Test: fast/regions/flow-fragment-as-anonymous-block-crash.html

rendering/RenderObject.h:

(WebCore::RenderObject::isAnonymousBlock):

LayoutTests:

fast/regions/flow-fragment-as-anonymous-block-crash-expected.txt: Added.
fast/regions/flow-fragment-as-anonymous-block-crash.html: Added.

3:23 PM Changeset in webkit [210119] by mark.lam@apple.com

3 edits in trunk/Source/JavaScriptCore

BytecodeGenerator::m_finallyDepth should be unsigned.
https://bugs.webkit.org/show_bug.cgi?id=166438

Reviewed by Saam Barati.

Also removed FinallyContext::m_finallyDepth because it is not used.

bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::pushFinallyControlFlowScope):
(JSC::BytecodeGenerator::labelScopeDepth):

bytecompiler/BytecodeGenerator.h:

(JSC::FinallyContext::FinallyContext):
(JSC::FinallyContext::finallyLabel):
(JSC::FinallyContext::depth): Deleted.

3:18 PM Changeset in webkit [210118] by timothy_horton@apple.com

2 edits in trunk/Source/WebKit2

Null deref under WebPageProxy::applicationDidFinishSnapshottingAfterEnteringBackground
https://bugs.webkit.org/show_bug.cgi?id=166433
<rdar://problem/29732407>

Reviewed by Daniel Bates.

UIProcess/ios/WebPageProxyIOS.mm:

(WebKit::WebPageProxy::applicationDidFinishSnapshottingAfterEnteringBackground):
m_drawingArea can be null if the process is not valid.

2:55 PM Changeset in webkit [210117] by Wenson Hsieh

3 edits
2 adds in trunk

CSS Scroll Snap does not work if scrollbar is hidden
https://bugs.webkit.org/show_bug.cgi?id=160442
<rdar://problem/23317034>

Reviewed by Simon Fraser.

Source/WebCore:

Currently, the only reason scroll snapping works in overflow scrolling containers without forcing layout is
because we would initialize the scrolling container's ScrollAnimator in the process of updating scrollbars. If
there are no scrollbars to render, we won't bother creating a ScrollAnimator. Without an existing
ScrollAnimator, ScrollableArea::updateScrollSnapState will simply bail instead of setting up the scroll snap
state. Instead, we should take setting a non-empty vector of scroll offsets on the ScrollableArea as a cue that
the ScrollableArea also needs a ScrollAnimator, and initialize it there if necessary.

Test: tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-hidden-scrollbars.html

platform/ScrollableArea.cpp:

(WebCore::ScrollableArea::setHorizontalSnapOffsets):
(WebCore::ScrollableArea::setVerticalSnapOffsets):

LayoutTests:

Adds a new layout test verifying that scroll snapping still works when scrollbars are hidden via CSS.

tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-hidden-scrollbars-expected.txt: Added.
tiled-drawing/scrolling/scroll-snap/scroll-snap-mandatory-hidden-scrollbars.html: Added.

2:48 PM Changeset in webkit [210116] by mark.lam@apple.com

8 edits
2 copies in trunk

De-duplicate finally blocks.
https://bugs.webkit.org/show_bug.cgi?id=160168

Reviewed by Saam Barati.

JSTests:

Re-landing r209952 with a few new tests added in test-finally.js.

stress/deeply-nested-finallys.js: Copied from JSTests/stress/deeply-nested-finallys.js.
Tests many levels of finally nesting. This causes the old code to hang (and crashes eventually) while trying to generate bytecode for the exponentially duplicated finally blocks. The new code completes this test almost instantly.

stress/test-finally.js: Copied from JSTests/stress/test-finally.js.
Tests control flow through various permutations of finally blocks.

Source/JavaScriptCore:

JS execution can arrive at a finally block when there are abrupt completions from
its try or catch block. The abrupt completion types include Break,
Continue, Return, and Throw. The non-abrupt completion type is called Normal
(i.e. the case of a try block falling through to the finally block).

Previously, we enable each of these paths for abrupt completion (except for Throw)
to run the finally block code by duplicating the finally block code at each of
the sites that trigger those completions. This patch fixes the implementation so
that each of these abrupt completions will set a completionTypeRegister (plus a
completionValueRegister for CompletionType::Return) and then jump to the
relevant finally blocks, and continue to thread through subsequent outer finally
blocks until execution reaches the outermost finally block that the completion
type dictates. We no longer duplicate the finally block code.

The implementation details:

We allocate a pair of registers (completionTypeRegister and completionValueRegister) just before entering the outermost try-catch-finally scope.

On allocating the registers, we initialize the completionTypeRegister to
CompletionType::Normal, and set the completionValueRegister to the empty
JSValue.

The completionTypeRegister will hold a CompletionType value. This is how we encode the CompletionType value to be set:

For Normal, Return, and Throw completion types:
- The completionTypeRegister is set to CompletionType::Normal, CompletionType::Return, and CompletionType::Throw respectively.

For Break and Continue completion types:
- The completionTypeRegister is set to a unique jumpID where the jumpID is computed as:

jumpID = CompletionType::NumberOfTypes + bytecodeOffset

The bytecodeOffset used here is the bytecodeOffset of the break or continue
statement that triggered this completion.

Each finally block will have 2 entries:
1. the catch entry.
2. the normal entry.

The catch entry is recorded in the codeBlock's exception handler table,
and can only be jumped to by the VM's exception handling mechanism.

The normal entry is recorded in a FinallyContext (at bytecode generation time
only) and is jumped to when we want enter the finally block due any of the
other CompletionTypes.

How each completion type works?

CompletionType::Normal
======================
We normally encounter this when falling through from a try or catch block to
the finally block.

For the try block case, since completionTypeRegister is set to Normal by default,
there's nothing more that needs to be done.

For the catch block case, since we entered the catch block with an exception,
completionTypeRegister may be set to Throw. We'll need to set it to Normal
before jumping to the finally block's normal entry.

CompletionType::Break
=====================
When we emit bytecode for the BreakNode, we check if we have any FinallyContexts
that we need to service before jumping to the breakTarget. If we don't, then
emit op_jump to the breakTarget as usual. Otherwise:

we'll register a jumpID and the breakTarget with the FinallyContext for the outermost finally block that we're supposed to run through.
we'll also increment the numberOfBreaksOrContinues count in each FinallyContext from the innermost to the one for that outermost finally block.
emit bytecode to set the completionTypeRegister to the jumpID.
emit bytecode to jump to the normal entry of the innermost finally block.

Each finally block will take care of cascading to the next outer finally block
as needed (see (5) below).

CompletionType::Continue
========================
Since continues and breaks work the same way (i.e. with a jump), we handle this
exactly the same way as CompletionType::Break, except that we use the
continueTarget instead of the breakTarget.

CompletionType::Return
======================
When we emit bytecode for the ReturnNode, we check if we have any FinallyContexts
at all on the m_controlFlowScopeStack. If we don't, then emit op_ret as usual.
Otherwise:

emit bytecode to set the completionTypeRegister to CompletionType::Return.
emit bytecode to move the return value into the completionValueRegister.
emit bytecode to jump to the normal entry of the innermost finally block.

Each finally block will take care of cascading to the next outer finally block
as needed (see (5) below).

CompletionType::Throw
======================
At the catch entry a finally block, we:

emit an op_catch that stores the caught Exception object in the completionValueRegister.
emit bytecode to set the completionTypeRegister to CompletionType::Throw.
Fall through or jump to the finally block's normal entry.

What happens in each finally block? ================================== For details on the finally block's catch entry, see "CompletionType::Throw" in (4) above.

The finally block's normal entry will:

restore the scope of the finally block.
save the completionTypeRegister in a savedCompletionTypeRegister.
proceed to execute the body of the finally block.

At the end of the finally block, we will emit bytecode check the
savedCompletionTypeRegister for each completion type see emitFinallyCompletion())
in the following order:

Check for CompletionType::Normal ================================ If savedCompletionTypeRegister is CompletionType::Normal, jump to the designated normalCompletion label. We only need this check this finally block also needs to check for Break, Continue, or Return. If not, the completion type check for CompletionType::Throw below will make this check redundant.

Check for CompletionType::Break and Continue ============================================ If the FinallyContext for this block has registered FinallyJumps, we'll check the jumpIDs against the savedCompletionTypeRegister. If the jumpID matches, jump to the corresponding jumpTarget.

If no jumpIDs match but the FinallyContext's numberOfBreaksOrContinues is
greater than the number of registered FinallyJumps, then this means that
we have a Break or Continue that needs to be handled by an outer finally
block. In that case, jump to the next outer finally block's normal entry.

Check for CompletionType::Return ================================ If this finally block is not the outermost and the savedCompletionTypeRegister is set to CompletionType::Return, then jump to the next outer finally block's normal entry.

Otherwise, if this finally block is the outermost and the savedCompletionTypeRegister
is set to CompletionType::Return, then execute op_ret and return the value
in the completionValueRegister.

CompletionType::Throw ===================== If savedCompletionTypeRegister is CompletionType::Throw, then just re-throw the Exception object in the completionValueRegister.

Detail 1: that we check the savedCompletionTypeRegister (and not the
completionTypeRegister). This is because the finally block may itself contain
a try-finally, and this inner try-finally may have trashed the completionTypeRegister.
Here's an example:

try {

return "r1"; Sets completionTypeRegister to CompletionType::Return;

} finally {

completionTypeRegister is CompletionType::Return here.

try {

... do stuff.

} finally {

... do more stuff.

}

completionTypeRegister may be anything here depending on what
was executed in the inner try-finally block above.

Hence, finally completion here must be based on a saved copy of the
completionTypeRegister when we entered this finally block.

}

Detail 2: the finally completion for CompletionType::Throw must always explicitly
check if the savedCompletionTypeRegister is CompletionType::Throw before throwing.
We cannot imply that it is so from the Throw case being last. Here's why:

completionTypeRegister is CompletionType::Normal here.
try {

return "r1"; Sets completionTypeRegister to CompletionType::Return;

} finally {

completionTypeRegister is CompletionType::Return here.

try {

... do stuff. No abrupt completions.

} finally {

completionTypeRegister is CompletionType::Return here (from the outer try-finally).
savedCompletionTypeRegister is set to completionTypeRegister (i.e. CompletionType::Return) here.

... do more stuff. No abrupt completions.

Unless there's an abrupt completion since entering the outer
finally block, the savedCompletionTypeRegister will remain set
to CompletionType::Return. If we don't explicitly check if the
savedCompletionTypeRegister is CompletionType::Throw before
throwing here, we'll end up erroneously throwing "r1".

}

...

}

restoreScopeRegister()

Since the needed scope objects are always stored in a local, we can restore
the scope register by simply moving from that local instead of going through
op_get_parent_scope.

m_controlFlowScopeStack needs to be a SegmentedVector instead of a Vector. This makes it easier to keep a pointer to the FinallyContext on that stack, and not have to worry about the vector being realloc'ed due to resizing.

Performance appears to be neutral both on ES6SampleBench (run via cli) and the
JSC benchmarks.

Relevant spec references:
https://tc39.github.io/ecma262/#sec-completion-record-specification-type
https://tc39.github.io/ecma262/#sec-try-statement-runtime-semantics-evaluation

bytecode/HandlerInfo.h:

(JSC::HandlerInfoBase::typeName):

bytecompiler/BytecodeGenerator.cpp:

(JSC::BytecodeGenerator::generate):
(JSC::BytecodeGenerator::BytecodeGenerator):
(JSC::BytecodeGenerator::emitReturn):
(JSC::BytecodeGenerator::pushFinallyControlFlowScope):
(JSC::BytecodeGenerator::popFinallyControlFlowScope):
(JSC::BytecodeGenerator::allocateAndEmitScope):
(JSC::BytecodeGenerator::pushTry):
(JSC::BytecodeGenerator::popTry):
(JSC::BytecodeGenerator::emitCatch):
(JSC::BytecodeGenerator::restoreScopeRegister):
(JSC::BytecodeGenerator::labelScopeDepthToLexicalScopeIndex):
(JSC::BytecodeGenerator::labelScopeDepth):
(JSC::BytecodeGenerator::pushLocalControlFlowScope):
(JSC::BytecodeGenerator::popLocalControlFlowScope):
(JSC::BytecodeGenerator::emitEnumeration):
(JSC::BytecodeGenerator::emitIsNumber):
(JSC::BytecodeGenerator::emitYield):
(JSC::BytecodeGenerator::emitDelegateYield):
(JSC::BytecodeGenerator::emitJumpViaFinallyIfNeeded):
(JSC::BytecodeGenerator::emitReturnViaFinallyIfNeeded):
(JSC::BytecodeGenerator::emitFinallyCompletion):
(JSC::BytecodeGenerator::allocateCompletionRecordRegisters):
(JSC::BytecodeGenerator::releaseCompletionRecordRegisters):
(JSC::BytecodeGenerator::emitJumpIf):
(JSC::BytecodeGenerator::pushIteratorCloseControlFlowScope): Deleted.
(JSC::BytecodeGenerator::popIteratorCloseControlFlowScope): Deleted.
(JSC::BytecodeGenerator::emitComplexPopScopes): Deleted.
(JSC::BytecodeGenerator::emitPopScopes): Deleted.
(JSC::BytecodeGenerator::popTryAndEmitCatch): Deleted.

bytecompiler/BytecodeGenerator.h:

(JSC::bytecodeOffsetToJumpID):
(JSC::FinallyJump::FinallyJump):
(JSC::FinallyContext::FinallyContext):
(JSC::FinallyContext::outerContext):
(JSC::FinallyContext::finallyLabel):
(JSC::FinallyContext::depth):
(JSC::FinallyContext::numberOfBreaksOrContinues):
(JSC::FinallyContext::incNumberOfBreaksOrContinues):
(JSC::FinallyContext::handlesReturns):
(JSC::FinallyContext::setHandlesReturns):
(JSC::FinallyContext::registerJump):
(JSC::FinallyContext::numberOfJumps):
(JSC::FinallyContext::jumps):
(JSC::ControlFlowScope::ControlFlowScope):
(JSC::ControlFlowScope::isLabelScope):
(JSC::ControlFlowScope::isFinallyScope):
(JSC::BytecodeGenerator::currentLexicalScopeIndex):
(JSC::BytecodeGenerator::CompletionRecordScope::CompletionRecordScope):
(JSC::BytecodeGenerator::CompletionRecordScope::~CompletionRecordScope):
(JSC::BytecodeGenerator::completionTypeRegister):
(JSC::BytecodeGenerator::completionValueRegister):
(JSC::BytecodeGenerator::emitSetCompletionType):
(JSC::BytecodeGenerator::emitSetCompletionValue):
(JSC::BytecodeGenerator::isInFinallyBlock): Deleted.

bytecompiler/NodesCodegen.cpp:

(JSC::ContinueNode::emitBytecode):
(JSC::BreakNode::emitBytecode):
(JSC::ReturnNode::emitBytecode):
(JSC::TryNode::emitBytecode):

Source/WTF:

Added some methods to bring SegmentedVector closer to parity with Vector.

wtf/SegmentedVector.h:

(WTF::SegmentedVector::first):
(WTF::SegmentedVector::last):
(WTF::SegmentedVector::takeLast):

2:40 PM Changeset in webkit [210115] by sbarati@apple.com

4 edits in trunk

WebAssembly: Make the spec-tests/address.wast.js test pass
https://bugs.webkit.org/show_bug.cgi?id=166429
<rdar://problem/29793220>

Reviewed by Keith Miller.

JSTests:

wasm.yaml:

Source/JavaScriptCore:

Right now, provably out of bound loads/stores (given a load/store's constant
offset) are not a validation error. However, we were failing to catch uint32_t
overflows in release builds (we did have a debug assertion). To fix this,
I now detect when uint32_t addition will overflow, and instead of emitting
a normal load/store, I emit code that throws an out of bounds memory exception.

wasm/WasmB3IRGenerator.cpp:

2:29 PM Changeset in webkit [210114] by sbarati@apple.com

2 edits in trunk/Tools

WebAssembly: Silence the output of the spec-tests
https://bugs.webkit.org/show_bug.cgi?id=166417

Reviewed by Mark Lam.

Scripts/run-jsc-stress-tests:

2:26 PM Changeset in webkit [210113] by dbates@webkit.org

4 edits in trunk/LayoutTests

Make http/tests/security/popup-blocked-from-{fake-event, window-open}.html actually test popup
blocker; run on WebKit2
https://bugs.webkit.org/show_bug.cgi?id=166403

Reviewed by Darin Adler.

The tests http/tests/security/popup-blocked-from-{fake-event, window-open}.html inadvertently
depended on the default behavior of DumpRenderTree to ignore the ChromeClient::createWindow()
callback when asked to create a new window instead of testing that the popup blocker blocked
opening windows. We should call testRunner.setCanOpenWindows() and testRunner.setPopupBlockingEnabled(true)
from both of these tests so that DumpRenderTree respects requests to open windows and enable
the popup blocker (it is disabled by default), respectively.

The default behavior of DumpRenderTree to disallow opening windows falls out from the purpose
and implementation of testRunner.setCanOpenWindows() to support writing tests where the embedding
client ignores requests to create a new window. We do not implement such a concept in WebKitTestRunner
(see <https://bugs.webkit.org/show_bug.cgi?id=166402>).

http/tests/security/popup-blocked-from-fake-event.html:
http/tests/security/popup-blocked-from-window-open.html:
platform/wk2/TestExpectations: Unskip the above tests as they now pass.

2:20 PM Changeset in webkit [210112] by dbates@webkit.org

3 edits
6 adds in trunk

Bypass pop-up blocker from cross-origin or sandboxed frame
https://bugs.webkit.org/show_bug.cgi?id=166290
<rdar://problem/29742039>

Reviewed by Darin Adler.

Source/WebCore:

Tests: fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame.html

fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame2.html
fast/events/popup-blocked-from-unique-frame-via-window-open-named-sibling-frame.html

page/DOMWindow.cpp:

(WebCore::DOMWindow::open): Use FrameLoader::findFrameForNavigation() to find the
target frame to navigate with respect to the active document just as we do in WebCore::createWindow().

LayoutTests:

fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame-expected.txt: Added.
fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame.html: Added.
fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame2-expected.txt: Added.
fast/events/popup-blocked-from-sandboxed-frame-via-window-open-named-sibling-frame2.html: Added.
fast/events/popup-blocked-from-unique-frame-via-window-open-named-sibling-frame-expected.txt: Added.
fast/events/popup-blocked-from-unique-frame-via-window-open-named-sibling-frame.html: Added.

2:19 PM Changeset in webkit [210111] by keith_miller@apple.com

7 edits in trunk

WebAssembly: The validator should not allow unused stack entries at the end of a block
https://bugs.webkit.org/show_bug.cgi?id=166411

Reviewed by Saam Barati.

JSTests:

Test cleanup and enable new passing tests.

wasm.yaml:
wasm/function-tests/br-if-as-return.js:

Source/JavaScriptCore:

This patch also cleans up some of the verbose mode logging.

wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::dumpExpressionStack):
(JSC::Wasm::B3IRGenerator::dump):

wasm/WasmFunctionParser.h:
wasm/WasmValidate.cpp:

(JSC::Wasm::dumpExpressionStack):
(JSC::Wasm::Validate::dump):

1:56 PM Changeset in webkit [210110] by Nikita Vasilyev

2 edits in trunk/Source/WebInspectorUI

Web Inspector: Styles sidebar: Uncommenting CSS rules of pseudo-elements doesn't work
https://bugs.webkit.org/show_bug.cgi?id=165831
<rdar://problem/29652688>

Reviewed by Matt Baker.

UserInterface/Views/CSSStyleDeclarationTextEditor.js:

(WebInspector.CSSStyleDeclarationTextEditor.prototype.uncommentAllProperties):
Call _resetContent() when toggling all properties to ensure properties have text markers.

(WebInspector.CSSStyleDeclarationTextEditor.prototype._propertyCommentCheckboxChanged):
(WebInspector.CSSStyleDeclarationTextEditor.prototype._uncommentProperty): Added.
This method is the opposite of _commentProperty. Introduce it to minimize code duplication.

1:10 PM Changeset in webkit [210109] by aestes@apple.com

17 edits in trunk/Source

Make WebCore::EditorInsertAction an enum class
https://bugs.webkit.org/show_bug.cgi?id=166424

Reviewed by Brent Fulgham.

Source/WebCore:

editing/AlternativeTextController.cpp:

(WebCore::AlternativeTextController::applyDictationAlternative):

editing/Editor.cpp:

(WebCore::Editor::pasteAsPlainTextWithPasteboard):
(WebCore::Editor::shouldInsertText):
(WebCore::Editor::insertTextWithoutSendingTextEvent):
(WebCore::Editor::insertLineBreak):
(WebCore::Editor::insertParagraphSeparator):
(WebCore::Editor::markMisspellingsAfterTypingToWord):
(WebCore::Editor::markAndReplaceFor):
(WebCore::Editor::changeBackToReplacedString):
(WebCore::Editor::transpose):
(WebCore::Editor::handleAcceptedCandidate):

editing/EditorInsertAction.h:

(): Deleted.

editing/gtk/EditorGtk.cpp:

(WebCore::Editor::pasteWithPasteboard):

editing/ios/EditorIOS.mm:

(WebCore::Editor::pasteWithPasteboard):
(WebCore::Editor::replaceSelectionWithAttributedString):

editing/mac/EditorMac.mm:

(WebCore::Editor::pasteWithPasteboard):
(WebCore::Editor::replaceNodeFromPasteboard):
(WebCore::Editor::replaceSelectionWithAttributedString):

editing/win/EditorWin.cpp:

(WebCore::Editor::pasteWithPasteboard):

page/ContextMenuController.cpp:

(WebCore::insertUnicodeCharacter):
(WebCore::ContextMenuController::contextMenuItemSelected):

page/DragController.cpp:

(WebCore::DragController::concludeEditDrag):

Source/WebKit/mac:

WebCoreSupport/WebEditorClient.mm:

(kit):

Source/WebKit/win:

WebCoreSupport/WebEditorClient.cpp:

(kit):
(WebEditorClient::shouldInsertNode):
(WebEditorClient::shouldInsertText):

Source/WebKit2:

WebProcess/InjectedBundle/API/c/WKBundleAPICast.h:

(WebKit::toAPI):

WebProcess/InjectedBundle/API/mac/WKWebProcessPlugInBrowserContextController.mm:

(toWK):

12:45 PM Changeset in webkit [210108] by bshafiei@apple.com

2 edits in tags/Safari-604.1.1.1/Source/WebCore

Merged r210099. rdar://problem/29782862

12:44 PM Changeset in webkit [210107] by bshafiei@apple.com

5 edits in tags/Safari-604.1.1.1/Source

Versioning.

12:40 PM Changeset in webkit [210106] by bshafiei@apple.com

1 copy in tags/Safari-604.1.1.1

New tag.

12:31 PM Changeset in webkit [210105] by jer.noble@apple.com

14 edits in trunk/Source/WebCore

Only include those parts of AVFoundation.framework which are strictly needed.
https://bugs.webkit.org/show_bug.cgi?id=166423

Reviewed by Eric Carlson.

Modules/plugins/QuickTimePluginReplacement.mm:
platform/graphics/avfoundation/objc/CDMSessionAVFoundationObjC.mm:
platform/graphics/avfoundation/objc/InbandTextTrackPrivateAVFObjC.mm:
platform/graphics/avfoundation/objc/InbandTextTrackPrivateLegacyAVFObjC.mm:
platform/graphics/avfoundation/objc/MediaPlayerPrivateAVFoundationObjC.mm:
platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm:
platform/mac/SerializedPlatformRepresentationMac.mm:
platform/mac/WebPlaybackSessionInterfaceMac.mm:
platform/mac/WebVideoFullscreenController.mm:
platform/mediastream/mac/AVAudioCaptureSource.mm:
platform/mediastream/mac/AVCaptureDeviceManager.mm:
platform/mediastream/mac/AVMediaCaptureSource.mm:
platform/mediastream/mac/AVVideoCaptureSource.mm:

12:07 PM Changeset in webkit [210104] by jer.noble@apple.com

4 edits in trunk/Source/WebCore

NULL-deref CRASH in WebCore::PlatformMediaSession::mediaType
https://bugs.webkit.org/show_bug.cgi?id=166407

Reviewed by Darin Adler.

In r207688, we added a facility in PlatformMediaSessionManager for safely walking through a
list of PlatformMediaSessions by replacing entries of deleted sessions with nullptr. We now
need to use those new iteration falicities in MediaSessionManageriOS.

In addition to the existing iterators, add one which takes a predicate, and returns the first
session which matches the predicate, or nullptr, if none do.

platform/audio/PlatformMediaSessionManager.cpp:

(WebCore::PlatformMediaSessionManager::findSession):
(WebCore::PlatformMediaSessionManager::anyOfSessions):

platform/audio/PlatformMediaSessionManager.h:

(WebCore::PlatformMediaSessionManager::sessions): Deleted.

platform/audio/ios/MediaSessionManagerIOS.mm:

(WebCore::MediaSessionManageriOS::configureWireLessTargetMonitoring):
(WebCore::MediaSessionManageriOS::nowPlayingEligibleSession):
(WebCore::MediaSessionManageriOS::externalOutputDeviceAvailableDidChange):
(WebCore::MediaSessionManageriOS::applicationDidEnterBackground):
(WebCore::MediaSessionManageriOS::applicationWillEnterForeground):

11:49 AM Changeset in webkit [210103] by commit-queue@webkit.org

5 edits in trunk

Source/WebKit2:
[Cocoa] SPI for setloadsImagesAutomatically
https://bugs.webkit.org/show_bug.cgi?id=166401

Patch by Zhuo Li <zachli@apple.com> on 2016-12-22
Reviewed by Darin Adler.

UIProcess/API/Cocoa/WKPreferences.mm: Call the C API under the hood.

(-[WKPreferences _loadsImagesAutomatically]):
(-[WKPreferences _setLoadsImagesAutomatically:]):

UIProcess/API/Cocoa/WKPreferencesPrivate.h:

Add _loadsImagesAutomatically property.

Tools:
[Cocoa] SPI for setloadsImagesAutomatically.
https://bugs.webkit.org/show_bug.cgi?id=166401.

Patch by Zhuo Li <zachli@apple.com> on 2016-12-22
Reviewed by Darin Adler.

TestWebKitAPI/Tests/WebKit2Cocoa/Preferences.mm:

(TEST): Test the _setLoadsImagesAutomatically SPI. By default,
_loadsImagesAutomatically returns YES.

10:31 AM Changeset in webkit [210102] by sbarati@apple.com

7 edits in trunk

WebAssembly: Make the spec-tests/start.wast.js test pass
https://bugs.webkit.org/show_bug.cgi?id=166416
<rdar://problem/29784532>

Reviewed by Yusuke Suzuki.

JSTests:

wasm.yaml:

Source/JavaScriptCore:

To make the test run, I had to fix two bugs:

We weren't properly finding the start function. There was code

that would try to find the start function from the list of *exported*
functions. This is wrong; the start function is an index into the
function index space, which is the space for *imports* and *local*
functions. So the code was just wrong in this respect, and I've
fixed it do the right thing. We weren't sure if this was originally
allowed or not in the spec, but it has been decided that it is allowed
and the spec-tests test for it: https://github.com/WebAssembly/design/issues/896

We were emitting a breakpoint for Unreachable. Instead of crashing,

this opcode needs to throw an exception when executing.

wasm/WasmB3IRGenerator.cpp:
wasm/WasmExceptionType.h:
wasm/js/WebAssemblyModuleRecord.cpp:

(JSC::WebAssemblyModuleRecord::link):
(JSC::WebAssemblyModuleRecord::evaluate):

wasm/js/WebAssemblyModuleRecord.h:

9:37 AM Changeset in webkit [210101] by Wenson Hsieh

3 edits in trunk/Tools

fast/events/ios/viewport-shrink-to-fit-allows-double-tap.html is flaky/order dependent
https://bugs.webkit.org/show_bug.cgi?id=161328
<rdar://problem/28475977>

Reviewed by Darin Adler.

Running a test that scales the viewport and listens to a didEndZooming callback (such as
viewport-shrink-to-fit-allows-double-tap.html) immediately after a test that may begin to trigger viewport
scaling as the test completes (such as viewport-device-width-at-initial-scale-fast-clicks.html) may result in
the didEndZooming callback of the latter test being fired early due to the previous test triggering a zoom
animation. Ideally, -[WKScrollView _stopScrollingAndZoomingAnimations], which is called when committing the page
load, should prevent this from happening, but from reading documentation and code inspection, this is not
intended to fire any associated delegate methods or notifications of the UIScrollView, instead deferring them
for later. Instead, what we need in this case is to clear out any started but not yet completed animations after
we finish running a test and before we load the next test.

One way to do this is to remove all animations from the WKScrollView's layer, and the layers of its children,
recursively. Doing so causes scrollViewDidEndZooming:withView:atScale: to fire at the end of the runloop if the
previous test started zooming and then finished before zooming completed. Since this happens before the next
test has a chance to run UI-side scripts, we won't end up firing this callback prematurely during the next test.

WebKitTestRunner/ios/TestControllerIOS.mm:

(WTR::TestController::platformResetStateToConsistentValues):

9:32 AM Changeset in webkit [210100] by jer.noble@apple.com

5 edits
1 add in trunk

Muted media element playback should not interrupt other audio playback
https://bugs.webkit.org/show_bug.cgi?id=166347

Reviewed by Eric Carlson.

Source/WebCore:

Test: TestWebKitAPI/WebKit/ios/AudioSessionCategoryIOS.mm

platform/audio/cocoa/MediaSessionManagerCocoa.cpp:

(PlatformMediaSessionManager::updateSessionState):

Tools:

TestWebKitAPI/TestWebKitAPI.xcodeproj/project.pbxproj:
TestWebKitAPI/Tests/WebKit/ios/AudioSessionCategoryIOS.mm:

(TestWebKitAPI::TEST):

TestWebKitAPI/Tests/WebKit/ios/video-with-muted-audio.html: Added.

9:14 AM Changeset in webkit [210099] by jer.noble@apple.com

2 edits in trunk/Source/WebCore

Check for the existence of AVSampleBufferAudioRenderer.h before redeclaring AVSampleBufferAudioRenderer
https://bugs.webkit.org/show_bug.cgi?id=166421

Reviewed by Eric Carlson.

platform/spi/mac/AVFoundationSPI.h:

8:58 AM Changeset in webkit [210098] by eric.carlson@apple.com

4 edits in trunk/Source

AVPlayerLayer isn't available on every system
https://bugs.webkit.org/show_bug.cgi?id=166399

Reviewed by Jer Noble.

Source/WebCore:

No new tests, prevents a crash that can't be reproduced on a test system.

platform/graphics/ca/cocoa/PlatformCALayerCocoa.mm: Use SOFT_LINK_CLASS_OPTIONAL.

Source/WebKit2:

WebProcess/WebPage/mac/PlatformCALayerRemoteCustom.mm: Use SOFT_LINK_CLASS_OPTIONAL.

8:41 AM Changeset in webkit [210097] by nael.ouedraogo@crf.canon.fr

5 edits in trunk/Source/WebKit2

MediaStream: ASSERTION FAILED: m_ids.size() == m_handles.size() in MediaDeviceSandboxExtensions
https://bugs.webkit.org/show_bug.cgi?id=166328

Reviewed by Eric Carlson.

This assertion failure happens for ports with sandbox extensions disabled. SandboxExtension is empty and
HandleArray:size() always returns 0. Disable creation of MediaDevicesSandboxExtension in UserMediaProcessManager
when building without support of sandbox extension. Disable also WebPage::grantUserMediaDeviceSandboxExtensions
and WebPage::revokeUserMediaDeviceSandboxExtensions which can be avoided when sandbox extension is empty.

UIProcess/UserMediaProcessManager.cpp:

(WebKit::UserMediaProcessManager::willCreateMediaStream):
(WebKit::UserMediaProcessManager::endedCaptureSession):

WebProcess/WebPage/WebPage.cpp:
WebProcess/WebPage/WebPage.h:
WebProcess/WebPage/WebPage.messages.in:

1:22 AM Changeset in webkit [210096] by commit-queue@webkit.org

4 edits in trunk

[WebRTC] Add support for runtime PeeConnection setting, required after r209757
https://bugs.webkit.org/show_bug.cgi?id=166346

Patch by Alejandro G. Castro <alex@igalia.com> on 2016-12-22
Reviewed by Carlos Garcia Campos.

Source/WebKit2:

After that commit we have to enable the peerConnection runtime
flag if we want webrtc to work. We are not adding a new API for
the moment to control the peerConnection feature.

UIProcess/API/gtk/WebKitSettings.cpp:

(webkit_settings_set_enable_media_stream): Use the mediaStream
setting to control the peerConnection

Tools:

Enable WebRTC by default to simplify testing.

MiniBrowser/gtk/main.c:

(main):

Dec 21, 2016:

7:06 PM Changeset in webkit [210095] by timothy_horton@apple.com

2 edits in trunk/Source/WebCore

TileGrid revalidates tiles twice during flush, first with wrong visible rect
https://bugs.webkit.org/show_bug.cgi?id=166406

Reviewed by Simon Fraser.

No new tests; existing tests cover this code, this is just a perf win,
specifically reducing the amount of layer churn during zooming.

platform/graphics/ca/TileGrid.cpp:

(WebCore::TileGrid::setScale):
Schedule a revalidation, which will happen later in the same flush,
instead of doing it immediately. Doing it immediately is problematic,
because we're currently in the middle of a GraphicsLayer tree flush,
and don't have the complete picture of the new state yet. We're guaranteed
to get the new scale *before* the flush calls revalidateTiles.

7:06 PM Changeset in webkit [210094] by timothy_horton@apple.com

2 edits in trunk/Source/WebCore

TileGrid creates new tiles when there are recyclable tiles about to be removed
https://bugs.webkit.org/show_bug.cgi?id=166408

Reviewed by Simon Fraser.

No new tests; existing tests cover this code, this is just a perf win,
specifically reducing the amount of layer churn during zooming.

platform/graphics/ca/TileGrid.cpp:

(WebCore::TileGrid::revalidateTiles):
Remove all the tiles that will be removed first, then add new tiles.
Strictly ordering it this way means that tiles will be removed, go into
the LayerPool, then be pulled back out of the LayerPool to sit in the
newly-covered areas. Previously, we would sometimes make new layers
for newly-covered areas, and then remove unneeded but otherwise recyclable
tiles, which would then just go sit in the LayerPool (and often get
pruned, wastefully).

5:57 PM Changeset in webkit [210093] by ap@apple.com

2 edits in trunk/Tools

REGRESSION: API test failure: WKWebView.LocalStorageClear
https://bugs.webkit.org/show_bug.cgi?id=166032
<rdar://problem/29758423>

Reviewed by Simon Fraser.

Add an @autoreleasepool around the code that allocates then clears the WKWebView,
to ensure that the view is destroyed before the callback fires.

TestWebKitAPI/Tests/WebKit2Cocoa/LocalStorageClear.mm:

(TEST):

5:41 PM Changeset in webkit [210092] by eric.carlson@apple.com

6 edits in trunk

[MediaStream] Update media-stream-event-constructor test
https://bugs.webkit.org/show_bug.cgi?id=166398

Reviewed by Jer Noble.

Source/WebCore:

No new tests, updated existing test.

Modules/mediastream/MediaStreamEvent.idl: Mark eventInitDict as optional.

LayoutTests:

Update and rebaseline test to reflect behavior changes that happened while the test was
skipped. Un-skip it on macOS.

fast/events/constructors/media-stream-event-constructor-expected.txt:
fast/events/constructors/media-stream-event-constructor.html:
platform/mac/TestExpectations:

5:26 PM Changeset in webkit [210091] by keith_miller@apple.com

4 edits in trunk

WebAssembly: Fix decode floating point constants in unreachable code
https://bugs.webkit.org/show_bug.cgi?id=166400

Reviewed by Saam Barati.

JSTests:

wasm.yaml:

Source/JavaScriptCore:

We decoded these as variable length but they should be fixed length.

wasm/WasmFunctionParser.h:

5:06 PM Changeset in webkit [210090] by keith_miller@apple.com

9 edits
4 adds in trunk

WebAssembly: Allow br, br_if, and br_table to act as a return
https://bugs.webkit.org/show_bug.cgi?id=166393

Reviewed by Saam Barati.

JSTests:

Add tests for breaks acting as returns and fix tests that
validate error messages.

wasm/function-tests/br-as-return.js: Added.
wasm/function-tests/br-if-as-return.js: Added.
wasm/function-tests/br-table-as-return.js: Added.
wasm/function-tests/if-no-else-non-void.js:
wasm/function-tests/struct.js: Added.
wasm/js-api/global-error.js:

(assert.throws):

wasm/js-api/table.js:

(assert.throws):

Source/JavaScriptCore:

This patch allows br, br_if, and br_table to treat branching to
the size of the control stack to act as a return. This change was
made by adding a new block type to the wasm function parser,
TopLevel. Adding this new block eliminates a lot of the special
case code we had in the parser previously. The only special case
we need is when the end opcode is parsed from the top level. The
B3 IR generator needs to automatically emit a return at that
point.

Also, this patch adds the function number to validation errors
in the function parser. The current error message is not helpful
otherwise.

wasm/WasmB3IRGenerator.cpp:

(JSC::Wasm::B3IRGenerator::ControlData::dump):
(JSC::Wasm::B3IRGenerator::addTopLevel):

wasm/WasmFunctionParser.h:
wasm/WasmPlan.cpp:

(JSC::Wasm::Plan::parseAndValidateModule):
(JSC::Wasm::Plan::run):

wasm/WasmValidate.cpp:

(JSC::Wasm::Validate::ControlData::dump):
(JSC::Wasm::Validate::Validate):
(JSC::Wasm::Validate::addTopLevel):
(JSC::Wasm::validateFunction):

4:53 PM Changeset in webkit [210089] by bshafiei@apple.com

5 edits in branches/safari-603-branch/Source

Versioning.

4:53 PM Changeset in webkit [210088] by bshafiei@apple.com

5 edits in branches/safari-602-branch/Source

Versioning.

4:51 PM Changeset in webkit [210087] by sbarati@apple.com

4 edits
57 adds in trunk

WebAssembly: Import spec tests
https://bugs.webkit.org/show_bug.cgi?id=166395

Rubber stamped by Keith Miller.

JSTests:

This patch implements the Wasm spec's tests found here:
https://github.com/WebAssembly/spec/tree/master/interpreter/test

These tests are in .wast s-epxression format. To convert
them to JS, I use a script from the wabt library, found here:
https://github.com/WebAssembly/wabt/blob/master/test/run-gen-spec-js.py

I also added a script that automatically imports the tests as
JS files. The inputs to the script is the path to the Wasm spec
git repo and the path to the wabt git repo. This will make importing
new tests easy in the future.

The wasm spec imported is at commit: b055d01ea1dfdd7a5231ae779095435f836de97f
The wabt used to do the import is at commit: 622b42dced6d793e9d49e9b1fd1d1524defd1387

wasm.yaml:
wasm/import-spec-tests.rb: Added.
wasm/spec-tests: Added.
wasm/spec-tests/address.wast.js: Added.