Add an --msan flag to run-tests.py.

Currently a no-op and added only for consistency with ASan/TSan.

BUG=chromium:425187
R=machenbach@chromium.org
LOG=N

Review URL: https://codereview.chromium.org/667903004

Patch from Sergey Matveev <earthdok@chromium.org>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24805 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Whitespace change to test new git workflow.

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/638883006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24804 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Switch ScheduleLateNodeVisitor to use explicit queue.

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/672583002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24803 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] Add support for deferred code.

Branch can now have an optional hint, when the condition is
likely true or false, and if such a hint is present the other
basic block will be marked as deferred and placed at the end
of the function.

We currently use this feature for tagging int32/uint32 in
change lowering, and for load/store bounds checks in simplified
lowering.

TEST=cctest,unittests
R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/642883003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24802 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Perf tests for fromCodePoint and codePointAt.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/670463005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24801 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add FRINTP (round towards positive infinity) instruction.

Macro Assember, assembler and simulator for ARM64 were missing FRINTP.

R=rodolph.perfetta@arm.com, ulan@chromium.org

Review URL: https://codereview.chromium.org/669923002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24800 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS64: Spread the knowledge!

Port r24793 (c0c2b0d)

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24799 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Disable libstdc++ debug mode on x64 for now.

TBR=mstarzinger@chromium.org

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24798 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove testing artifacts from Schedule::AddFoo.

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/641553003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24797 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Flatten the string in StringToDouble function.

R=yangguo@chromium.org
BUG=chromium:425551
LOG=N

Review URL: https://codereview.chromium.org/654763003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24796 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Really disable test when running with optimize for size

R=bmeurer@chromium.org
TBR=bmeurer@chromium.org
LOG=n
BUG=none

Review URL: https://codereview.chromium.org/652903008

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24795 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Enable libstdc++ debug mode in debug builds.

For now it's only enabled on Linux/Intel.

R=mstarzinger@chromium.org
BUG=v8:3638
LOG=y

Review URL: https://codereview.chromium.org/650013006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24794 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Spread the knowledge!

R=arv@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/670623005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24793 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix newly added test to not run with optimize-for-size

TBR=ulan@chromium.org
BUG=none
LOG=n

Review URL: https://codereview.chromium.org/673593002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24792 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add support for a target new space size

When this flag is set, we will commit additional pages up until the
target size before doing a scavenge.

I made sure that all tests pass when a target size of 16MB is set.

BUG=v8:3626
R=ulan@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/652543007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24790 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use getauxval() if available.

This fixes the problem of not being able to detect ARM features on
ChromeOS because sandbox cannot access /proc/self/auxv and
/proc/cpuinfo there.

Drive-by-cleanup to libc detection in two other places.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/672543002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24789 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Whitespace change to trigger revert of new test262-es6 tests.

TBR=rossberg@chromium.org

Review URL: https://codereview.chromium.org/665373003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24788 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Whitespace change to trigger new test262-es6 tests.

TBR=rossberg@chromium.org

Review URL: https://codereview.chromium.org/672483002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24787 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Whitespace change to trigger new test262-es6 tests.

TBR=rossberg@chromium.org

Review URL: https://codereview.chromium.org/646873005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24786 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Replace single quotes with double quotes to fix BUILD.gn

TBR=jochen@chromium.org

Review URL: https://codereview.chromium.org/665313003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24785 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Array.prototype.{slice,splice} should use [[DefineOwnProperty]] to generate return value

BUG=chromium:423633
LOG=N
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/649063003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24784 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Update ObjectToString to Harmony-draft algorithm

Updates Object.prototype.toString() to use algorithm described in harmony drafts.

Currently, the behaviour is essentially the same as ES262's version, however this changes when internal structures
such as Promise make use of symbolToStringTag (as they are supposed to, see v8:3241), and changes further once
Symbol.toStringTag is exposed publicly.

BUG=v8:3241, v8:3502
LOG=N
R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/546803003

Patch from Caitlin Potter <caitpotter88@gmail.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24783 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Make sure floating phi nodes are coupled to their control."

TBR=jarin@chromium.org

Review URL: https://codereview.chromium.org/654583004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24782 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make sure floating phi nodes are coupled to their control.

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/669683002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24781 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix scheduler not to connect final merge block in the graph to its inputs.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/667953002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24780 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] use ZonePool in most places in the compiler pipeline a temp zone is used.

R=jarin@chromium.org, bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/663333003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24779 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix off-by-one bug in TurboFan register allocator.

R=jarin@chromium.org
BUG=

Review URL: https://codereview.chromium.org/671703003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24778 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix AstGraphBuilder for loops like for(;;).

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/640203004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24777 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix incorrect private access

(I'm puzzled why GCC did not report this error.)

TBR=bmeurer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/669733007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24776 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The issue is that by handling strings with map/handler pairs instead of a special
version of the keyed load stub (https://code.google.com/p/v8/source/detail?r=24661),
I allowed polymorphism between string and non-string types in the IC. Before, the
IC would go generic.

Then, at crankshaft time, we special case when we only saw strings. The error
here is that crankshaft can't emit code that handles polymorphism between string
and non-string types. The choice is either to get that to happen (I don't deem
this necessary from a performance point of view, an IC with such type feedback
before would have gone generic), or simply check for the case of "polymorphic
with some string maps" and require crankshaft to go generic. I'll do the latter.

BUG=425519
LOG=N
R=jarin@chromium.org

Review URL: https://codereview.chromium.org/667923004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24775 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix data race with interface caching

R=bmeurer@chromium.org
BUG=421634
LOG=N

Review URL: https://codereview.chromium.org/667703002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24774 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use an idle notification of 0ms as hint that a GC after context disposal is triggerd

BUG=none
R=ulan@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/671513006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24773 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] add ZonePool to correctly track compiler phase memory usage

R=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/665893006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24771 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] Reduce ~~x to x.

TEST=unittests
R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/666723005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24770 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Assign bailout and type feedback IDs in a post-pass

This will allow us to move expressions from one function to another, for
example when the parser determines that a given cover grammar instance
is actually the default value initializer for an arrow function.

This is a re-land of https://codereview.chromium.org/636403003/ with a
fix for the arm64 code generator.

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/663373003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24769 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handle property name "-0" correctly for typed arrays.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/670623003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24768 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove unused IsDirty

BUG=
R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/670703002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24767 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handle NaN value storing in Uint8Clamped typed arrays.

TEST=mjsunit/external-array
BUG=
R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/667113002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24766 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Perf tests for harmony string functions.

R=yangguo@chromium.org

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24765 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[x86] Select better left operand for comparisons.

R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/667933002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24764 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix leak in d8.

R=bmeurer@chromium.org
TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/652403003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24763 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Assign bailout and type feedback IDs in a post-pass"

This reverts r24757, which breaks the ARM64 simulator build.
Simple repro:

   out/arm64.debug/d8 -e 'eval("(function(){ const x; var x; })")'

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/652543006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24762 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove unused GetConstructor

BUG=
R=ishell@chromium.org

Review URL: https://codereview.chromium.org/663033004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24761 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix use-of-uninitialized-value introduced in r24700

LOG=N
BUG=chromium:425146
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/665243002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24760 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix Win64 after r24758.

BUG=
TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/663343002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24759 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use smi zero instead of undefine_value to zap dead weak cells.

It is faster to test for smi zero from generated code.

BUG=
R=erikcorry@chromium.org

Review URL: https://codereview.chromium.org/637253004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24758 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Assign bailout and type feedback IDs in a post-pass

This will allow us to move expressions from one function to another, for
example when the parser determines that a given cover grammar instance
is actually the default value initializer for an arrow function.

R=svenpanne@chromium.org, marja@chromium.org
BUG=

Review URL: https://codereview.chromium.org/636403003

Patch from Andy Wingo <wingo@igalia.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24757 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

X87: vector-based ICs did not update type feedback counts correctly.

port r24732.

original commit message:

  vector-based ICs did not update type feedback counts correctly.

BUG=
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/669823002

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24756 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove v8stdint.h, it doesn't serve a purpose anymore.

Basically a follow-up to https://codereview.chromium.org/667573005/.

LOG=y
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/670673002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24755 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

fix ia32 after r24753

TBR=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/668883002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24754 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] cleanup InstructionSequence

R=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/664123002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24753 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d8: create compile cache in a separate isolate.

R=vogelheim@chromium.org

Review URL: https://codereview.chromium.org/670433003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24752 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Visual Studio provides stdint.h these days.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/667573005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24751 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS64: vector-based ICs did not update type feedback counts correctly.

Port r24732 (83e975b)

BUG=v8:3605
LOG=N
R=paul.lind@imgtec.com

Review URL: https://codereview.chromium.org/670543002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24746 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix gc mole test failure.

TBR=ulan@chromium.org

Review URL: https://codereview.chromium.org/651683003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24745 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove deprecated Ascii-related identifiers from include/v8.h

R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/665883002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24744 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Simplify language feature management.

R=rossberg@chromium.org
BUG=v8:3640
LOG=N

Review URL: https://codereview.chromium.org/642233003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24743 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Drop obsolete misc-intrinsics.h file.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/648643003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24742 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove dead GeneralizeFieldRepresentation

BUG=
R=ishell@chromium.org

Review URL: https://codereview.chromium.org/668663002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24741 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Upgrade test262-es6

R=dslomov@chromium.org
BUG=

Review URL: https://codereview.chromium.org/639373005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24740 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove fuzz-natives tests.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/663043003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24739 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Annotate PromotionQueue::RelocateQueueHead for MemorySanitizer.

BUG=chromium:416875
LOG=N
R=earthdok@chromium.org, jkummerow@chromium.org

Review URL: https://codereview.chromium.org/603633002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24738 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Read object pointer atomically while updating slots

BUG=425003
R=ulan@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/663023002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24737 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Move some Runtime:: functions and remove runtime.h as include when unnecessary.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/662413002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24736 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove (untested) code for unsupported compilers.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/656143004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24735 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Whitespace change to test new git workflow.

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/669673002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24733 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

vector-based ICs did not update type feedback counts correctly.

BUG=v8:3605
LOG=N
R=jkummerow@chromium.org, ulan@chromium.org

Review URL: https://codereview.chromium.org/650073002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24732 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add README.md file

Add basic info. This file is automatically displayed on GitHub, and is
kind of de-facto standard for project documentation.

BUG=chromium:415466
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/657423002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24731 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

platform: fix build on SmartOS

With -std=g++0x, `signbit` is no longer a define, and is an actual
function declared in C++-specific headers. Thus checking it's presence
with ifdef will no longer work.

Considering that g++0x implies newer compiler, there should not be the
case where it won't be present anymore.

BUG=
R=bmeurer@chromium.org, danno

Review URL: https://codereview.chromium.org/642203003

Patch from Fedor Indutny <fedor.indutny@gmail.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24730 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] Move node matchers to separate file.

TEST=unittests
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/639293006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24729 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] pass zone to InstructionSequence

R=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/663073002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24728 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] fix build after r24726

TBR=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/669643002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24727 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] remove schedule from InstructionSequence

R=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/669613002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24726 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Script streaming: more UTF-8 handing fixes (again).

1) Since we fill the output buffer both from the chunks and the conversion
buffer, it's possible that we run out of space and call CopyCharsHelper with 0
length. The underlying functions don't handle it gracefully, so check there.

2) There was a bug where we used to try to copy too many characters from the
beginning of the data chunk into the conversion buffer. Continuation bytes in
UTF-8 are of the form 0b10XXXXXX. If a byte is bigger than that, it's the first
byte of a new UTF-8 character and we should ignore it.

These two together (or maybe in combination with surrogates) are a probable
reason for crbug.com/420932.

3) The test data was off; \uc481 is \xec\x92\x81.

BUG=420932
LOG=N
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/662003003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24725 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[arm64] Skip TF tests that time out in simulator debug runs.

TBR=dcarney@chromium.org

Review URL: https://codereview.chromium.org/642293005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24724 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Special handling for inline caches in code serializer.

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/656533003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24722 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Avoid the Marsaglia effect in 3D

For a longer discussion, see the associated Chromium issue.

BUG=chromium:423311
LOG=y
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/662513004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24721 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix Win64 size_t/int conversion after b3f2277ea42a.

TBR=dcarney@chromium.org

Review URL: https://codereview.chromium.org/646983005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24720 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add inlining for intrinsics.

This issue is for discussion on how to proceed.

I think the implementation of ValueOf shows that directly creating the IR does not scale.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/612043003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24719 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

some static casts for windows to fix r24717

TBR=bmeurer@chromium.org

BUG=

Review URL: https://codereview.chromium.org/660843003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24718 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] decouple register allocation from schedule and graph

R=bmeurer@chromium.org, jarin@chromium.org

BUG=

Review URL: https://codereview.chromium.org/664683002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24717 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[turbofan] Skip bounds checks for positive indices only.

TEST=unittests,mjsunit/asm/int32array-constant-key
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/647773004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24716 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Enable libstdc++ debug mode in debug builds.".

This reverts commit r24676 to see if it was the cause for the mysterious
crashes on some bots. If not, we'll reland it.

TBR=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/663013002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24715 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

X87:  ReceiverCheckMode needs to be utilized further.

port r24704.

original commit message:

  ReceiverCheckMode needs to be utilized further.

  The parameter wasn't being passed appropriately, and there was an extra
   opportunity to use mode RECEIVER_IS_STRING in SubStringStub.

BUG=
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/664973002

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24714 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

X87: Tick processor: Print C++ entry points.

port r24700.

original commit message:

  Tick processor: Print C++ entry points

BUG=
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/662323002

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24713 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add turbo.cfg to .gitignore.

TBR=jarin@chromium.org

Review URL: https://codereview.chromium.org/664003003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24712 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

X87: Implement the new semantics for 'super(...)'

original commit message:

  Implement the new semantics for 'super(...)'

  Per the latest ES6 draft, super(...) translates into a call
   to function's prototype.

BUG=
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/642603006

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24711 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

X87: Make disassembler string tables read-only.

port r24672.

original commit message:

  Move them from .data to .rodata.

BUG=
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/654263006

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24710 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix webkit getOwnPropertyNames test after r24706 removed Array.prototype.values

TBR=danno@chromium.org

Review URL: https://codereview.chromium.org/664613004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24708 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS64: ReceiverCheckMode needs to be utilized further.

Port r24704 (f43a27ee)

Original commit message:
The parameter wasn't being passed appropriately, and there was an extra
opportunity to use mode RECEIVER_IS_STRING in SubStringStub.

BUG=
R=paul.lind@imgtec.com

Review URL: https://codereview.chromium.org/645823003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24707 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Don't expose Array.prototype.values as it breaks webcompat

Some versions of Outlook Web Access test for the existence of a 'values'
property on Array instances, so adding the 'values' iterator to the prototype
(even with @@unscopeables) causes breakage.

This matches Gecko: they ship Array.prototype.{keys,entries} but not 'values'.

BUG=409858
LOG=Y
R=arv@chromium.org, danno@chromium.org

Review URL: https://codereview.chromium.org/647703003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24706 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Correct semantics for numerically indexed stores to typed arrays.

R=verwaest@chromium.org, ishell@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=24691

Review URL: https://codereview.chromium.org/652303002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24705 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ReceiverCheckMode needs to be utilized further.

The parameter wasn't being passed appropriately, and there was an extra
opportunity to use mode RECEIVER_IS_STRING in SubStringStub.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/657313005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24704 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix compilation with GCC-4.8 after r24685

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/661123002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24701 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Tick processor: Print C++ entry points

R=loislo@chromium.org, yangguo@chromium.org, yurys@chromium.org

Review URL: https://codereview.chromium.org/638633002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24700 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix compile issue.

TBR=dcarney@chromium.org

Review URL: https://codereview.chromium.org/659313003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24699 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

skip some gc stress tests for tf

TBR=mstarzinger@chromium.org

BUG=

Review URL: https://codereview.chromium.org/659333002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24698 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix break location calculation.

R=ulan@chromium.org
BUG=chromium:419663
LOG=Y

Review URL: https://codereview.chromium.org/658723005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24697 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Keyed stores to super with numeric keys.

R=verwaest@chromium.org, arv@chromium.org, ishell@chromium.org
BUG=v8:3330
LOG=N

Review URL: https://codereview.chromium.org/649603003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24696 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Correct semantics for numerically indexed stores to typed arrays."

This reverts commit r24691 because win64 release build breaks.

TBR=verwaest@chromium.org

Review URL: https://codereview.chromium.org/659313002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@24695 ce2b1a6d-e550-0410-aec6-3dcde31c8c00