Constructed arrays can be created with Hydrogen code stubs. The feature is still off by default (--optimize-constructed-arrays).

BUG=

Review URL: https://codereview.chromium.org/12385014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14441 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

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

Review URL: https://codereview.chromium.org/14392013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14440 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The GC prologue is called by the GC, but also by the heap snapshotter. The
RetainedObjectInfos are only needed by the heap snapshotter, so it's wasteful to
construct them always. (And it will be even more wasteful when Blink migrates to
the new GC APIs, since after that point it no longer knows about object groups.)

BUG=

Review URL: https://codereview.chromium.org/14471028
Patch from Marja Hölttä <marja@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14439 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
BUG=

Review URL: https://codereview.chromium.org/13958007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14438 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ia32, unlike the other architectures, includes a --debug-code check that
asserts that runtime functions do not return the hole.  However the new
SuspendJSGeneratorObject runtime does return the hole at times.

This CL adds a wee hack that only signals an error if the callee was not
SuspendJSGeneratorObject.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/13856011

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14437 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Previously there has been no reason to context-allocate the receiver, so
access to the receiver always goes through the stack.  This was failing
with generators, which assumed that forcing context allocation would
relieve the need of storing anything but the context and the function on
the stack.

This CL adds a slot in generator objects to capture the receiver, and
restores it when resuming a generator.

BUG=v8:2355
TEST=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/14158006

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14434 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ulan@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14284011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14433 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
TEST=gcmole

Review URL: https://codereview.chromium.org/14493012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14432 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=

Review URL: https://codereview.chromium.org/14493011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14431 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: https://codereview.chromium.org/14471012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14430 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14421 (31cd7dda)

BUG=

Review URL: https://codereview.chromium.org/14483002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14429 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14415 (d358defa) and r14424 (7b549ce7)

Original commit message:
The generator object methods "next", "send", and "throw" now
include some inline assembly to set up a resumed stack frame.  In some
common cases, we can just jump back into the frame to resume it.
Otherwise the resume code calls out to a runtime to fill in the operand
stack, rewind the handlers, and possibly to throw an exception.

BUG=v8:2355
TESTS=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/13864010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14428 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14407 (cae6596b)

BUG=

Review URL: https://codereview.chromium.org/13840006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14427 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Casting NaN to int is unpredictable, on different architectures it produces different int value.

TEST=test262/S15.4.4.10_A2.1_T2, S15.4.4.10_A2.2_T2, S15.4.4.12_A2.1_T2

BUG=

Review URL: https://codereview.chromium.org/14257006
Patch from Dusan Milosavljevic <Dusan.Milosavljevic@rt-rk.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14426 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14333016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14425 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This CL fixes a number of rookie mistakes in
FullCodeGenerator::VisitYield and
FullCodeGenerator::EmitGeneratorResume, and re-enables the
generators-iteration test on ARM.

R=mstarzinger@chromium.org
BUG=v8:2355
TEST=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/13843022

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14424 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

With these APIs, the embedder doesn't need to copy Persistent handles around.

BUG=

Review URL: https://codereview.chromium.org/14007008

Patch from Marja Hölttä <marja@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14423 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: https://codereview.chromium.org/14333013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14422 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14139033

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14421 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Instead, create Local handles to pass them around. This also means that the
code needs to be shifted around a bit such that a handle scope exists when
creating threads.

Review URL: https://codereview.chromium.org/14150017

Patch from Jochen Eisinger <jochen@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14420 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14476003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14417 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14471007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14416 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The generator object methods "next", "send", and "throw" now
include some inline assembly to set up a resumed stack frame.  In some
common cases, we can just jump back into the frame to resume it.
Otherwise the resume code calls out to a runtime to fill in the operand
stack, rewind the handlers, and possibly to throw an exception.

BUG=v8:2355
TESTS=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/14066016

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14415 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: https://codereview.chromium.org/13829009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14414 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Properly consider Double inputs for representation inference when they are not marked as kFlexibleRepresentation

Review URL: https://codereview.chromium.org/14408006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14413 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

CodeEntry* is always NULL for SharedFunctionInfo entries in the CodeMap. Take this into account when printing the map.

Drive-by: removed  CodeEntry::shared_id() which is never called.

BUG=None

Review URL: https://codereview.chromium.org/14387004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14410 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mvstanton@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/13940014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14409 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ulan@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14471004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14408 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/14367018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14407 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

std::sort is a template, so it can be inlined more aggressively. Also, it's
O(n log n), while libc's qsort is O(n^2)

BUG=2639

Review URL: https://codereview.chromium.org/14315005

Patch from Jochen Eisinger <jochen@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14406 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://chromiumcodereview.appspot.com/14139031

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14403 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: https://codereview.chromium.org/14286005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14402 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

If the input to an HChange is of type smi, then no need to mark gvn flag changes new space promotion.

BUG=

Review URL: https://codereview.chromium.org/14253010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14401 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://chromiumcodereview.appspot.com/14429003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14400 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14388 (6843e0c9), r14389 (331ea32d) and r14390 (c8528d1c)

BUG=

Review URL: https://codereview.chromium.org/14305011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14399 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:2646

Patch by Andrei Kashcha <anvaka@gmail.com>

Review URL: https://codereview.chromium.org/14042008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14398 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This also keeps backwards compatibility (window.gc() still works).

BUG=2641

Review URL: https://codereview.chromium.org/14075012

Patch from Marja Hölttä <marja@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14397 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:2653
R=jkummerow@chromium.org
TEST=mjsunit/regress/regress-2653.js

Review URL: https://chromiumcodereview.appspot.com/14081025

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14396 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/14091013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14395 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert part of r14179. From the regression test's comment:

Currently, the gap resolver doesn't handle moves from a ConstantOperand to a
DoubleRegister, but these kind of moves appeared when HConstant::EmitAtUses
was changed to allow special double values (-0, NaN, hole). So we should
either enhance the gap resolver or make sure that such moves don't happen.

BUG=chrome:234101

Review URL: https://codereview.chromium.org/14429002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14394 ce2b1a6d-e550-0410-aec6-3dcde31c8c00