ricow@chromium.org [Mon, 12 Dec 2011 06:38:29 +0000 (06:38 +0000)]
MIPS: Reland 10216 - Optimize the equality check case of ICCompare stubs.
Port r10219 (56686b).
Original commit message:
Now with arm and x64 support. Additionally, added default unreachable case to switch statement in CompareIC::TargetState to make win and mac compilers happy.
Reviewer guide:
This is an exact copy of 10216 except:
src/arm/*
src/x64/*
src/ic.cc (added default case to swith in CompareIC::TargetState)
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8896022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10232
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Fri, 9 Dec 2011 15:04:41 +0000 (15:04 +0000)]
Fix regex regression in top-of-tree.
R=jkummerow@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/
8896001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10231
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Fri, 9 Dec 2011 12:11:56 +0000 (12:11 +0000)]
Fixing bug introduced in r10210 that crashes v8 raytrace benchmark.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8889047
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10226
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Fri, 9 Dec 2011 10:04:58 +0000 (10:04 +0000)]
Avoid bailing out to runtime for short substrings.
This significantly improves the speed for creating short substrings (less than 13 characters) from slices, flat cons strings and external strings.
TEST=string-external-cached.js, string-slices.js
Review URL: http://codereview.chromium.org/
8889012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10221
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Fri, 9 Dec 2011 09:50:30 +0000 (09:50 +0000)]
Hydrogen support for context allocated harmony bindings.
This CL adds support for loading from and storing to context slots
belonging to harmony let or const bound variables. Checks for the
hole value are performed and the function is deoptimized if they fail.
The full-codegen generated code will take care of properly throwing
a reference error in these cases.
TEST=mjsunit/harmony/block-let-crankshaft.js
Review URL: http://codereview.chromium.org/
8820015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10220
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Fri, 9 Dec 2011 09:26:14 +0000 (09:26 +0000)]
Reland 10216 - Optimize the equality check case of ICCompare stubs.
Now with arm and x64 support. Additionally, added default unreachable case to switch statement in CompareIC::TargetState to make win and mac compilers happy.
Reviewer guide:
This is an exact copy of 10216 except:
src/arm/*
src/x64/*
src/ic.cc (added default case to swith in CompareIC::TargetState)
Review URL: http://codereview.chromium.org/
8872060
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10219
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Fri, 9 Dec 2011 08:50:19 +0000 (08:50 +0000)]
Support Smi->Double->HeapObject transitions in constructed Arrays.
Also several bugs with Smi/double elements handling and make Ensure* routines more flexible.
BUG=none
TEST=test/mjsunit/array-construct-transition.js
Review URL: http://codereview.chromium.org/
8820014
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10218
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Thu, 8 Dec 2011 17:28:44 +0000 (17:28 +0000)]
Revert 10216 Optimize the equality check case of ICCompare stubs.
Missing arm and x64 implementations
Review URL: http://codereview.chromium.org/
8883023
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10217
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Thu, 8 Dec 2011 17:17:21 +0000 (17:17 +0000)]
Optimize the equality check case of ICCompare stubs.
This includes specialcasing the generation when we know that the maps
of the two objects are the same. In addition, a new specialized
compare ic known objects cache is created.
The reason for the cache is that we need to have access to the stub
code from the roots; if we do not, the GC will collect the stub. In
this specialized case we use the map pointer as key in the cache, and
we always do a lookup before generating code. Actually hitting
something in the cache will happen very rarely, but we could
potentially overwrite an existing stub, which again will lead to the
GC collecting this old stub (even if it is referenced from other code
objects)
Review URL: http://codereview.chromium.org/
8520006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10216
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Thu, 8 Dec 2011 16:07:07 +0000 (16:07 +0000)]
Ensure that non-optimized code objects are not flushed for inlined functions.
Collector was flushing them if optimized code was reachable only through the stack (not through the JSFunction object) which happens when you have a pending lazy deoptimization.
Also prevent v8::Script::New from leaking internal objects allocated by the compiler into outer HandleScope.
R=kmillikin@chromium.org
BUG=http://crbug.com/97116
TEST=test/mjsunit/regress/regress-97116.js
Review URL: http://codereview.chromium.org/
8888011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10215
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Thu, 8 Dec 2011 14:36:36 +0000 (14:36 +0000)]
MIPS: Fixing MathPowHalf on ARM.
Port r10166 (
b57f3f1a), r10167 (
202eada9) and r10170 (
5c5c96da).
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8822014
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10214
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Thu, 8 Dec 2011 13:16:55 +0000 (13:16 +0000)]
Prepare push to trunk. Now working on version 3.8.1.
Review URL: http://codereview.chromium.org/
8873016
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10213
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Thu, 8 Dec 2011 08:53:09 +0000 (08:53 +0000)]
MIPS: Port to x64 and ARM and some refactoring of ia32.
Port r10174 (
a410586).
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8821024
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10211
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 7 Dec 2011 16:55:00 +0000 (16:55 +0000)]
Port Math.pow inlining to ARM.
TEST=math-pow.js
Review URL: http://codereview.chromium.org/
8840008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10210
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 7 Dec 2011 16:15:18 +0000 (16:15 +0000)]
Tweaks on Math.pow (ia32 and x64).
Review URL: http://codereview.chromium.org/
8831008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10203
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Wed, 7 Dec 2011 16:08:09 +0000 (16:08 +0000)]
Relax test condition to make it less brittle.
BUG=
TEST=cctest/test-api/IdleNotificationWithHint
Review URL: http://codereview.chromium.org/
8839007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10202
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Wed, 7 Dec 2011 16:03:29 +0000 (16:03 +0000)]
Sync parser and preparser on do-while and return statements.
This CL fixes the preparser to have the same liberal automatic semicolon
insertion behaviour as the parser. In the case of a return statement in
global code we throw a syntax error at runtime rather than an early error
due to compatibility with KJS. However that hack allowed the following
syntactically incorrect program in global code in the parser but not in
the preparser:
if (false) return else {}
while the slightly saner version with the obligatory semicolon
if (false) return; else {}
was disallowed in the parser, but the preparser allowed it. This CL also
fixes that issue.
BUG=v8:1856
TEST=cctest/test-parsing.cc
Review URL: http://codereview.chromium.org/
8844002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10201
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Wed, 7 Dec 2011 13:00:11 +0000 (13:00 +0000)]
MIPS: Optimize Crankshaft array literal initialization from boilerplate.
Port r10138 (
730f5a7f)
Original commit message:
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8789012
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10191
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Wed, 7 Dec 2011 10:13:46 +0000 (10:13 +0000)]
Fix a bug with register use in optimized Math.round.
We're not allowed to modify the input register and have to
use a temporary instead, otherwise the result of expressions
containing Math.round can be wrong.
BUG=106351
TEST=test/mjsunit/compiler/regress-106351.js
Review URL: http://codereview.chromium.org/
8833007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10190
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 7 Dec 2011 09:44:31 +0000 (09:44 +0000)]
Tweak to shorten generated code in Math.pow.
Review URL: http://codereview.chromium.org/
8834007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10189
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 7 Dec 2011 08:49:06 +0000 (08:49 +0000)]
Remove write barriers for cells on x64, ARM and MIPS.
Review URL: http://codereview.chromium.org/
8834005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10188
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 7 Dec 2011 08:43:41 +0000 (08:43 +0000)]
Rescan cells at the end of mark-sweep. This means they don't need a
write barrier.
Review URL: http://codereview.chromium.org/
8816021
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10187
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 7 Dec 2011 08:43:18 +0000 (08:43 +0000)]
More removal of write barriers and a rename to a less scary name.
Review URL: http://codereview.chromium.org/
8816020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10186
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 7 Dec 2011 08:34:27 +0000 (08:34 +0000)]
Porting Math.pow changes to x64.
TEST=math-pow.js, regress-397.js
Review URL: http://codereview.chromium.org/
8821019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10185
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Tue, 6 Dec 2011 22:27:15 +0000 (22:27 +0000)]
Fix nosnap build test failures.
TBR=jkummerow@chromium.org
BUG=none
TEST=less waterfall redness
Review URL: http://codereview.chromium.org/
8828004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10184
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Tue, 6 Dec 2011 22:08:46 +0000 (22:08 +0000)]
Obligatory fix for Win64.
BUG=v8:1832
TEST=none
TBR=vitalyr@chromium.org
Review URL: http://codereview.chromium.org/
8771051
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10183
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Tue, 6 Dec 2011 17:41:47 +0000 (17:41 +0000)]
Reapply r10158: Distinguish weak references in heap snapshots, group GC roots.
[Fixed the bug revealed by the Win32 bot]
Several changes to better organize snapshot data:
1. Provide information about weak references.
2. Group (GC roots) children.
3. Prettify debug snapshot printing.
BUG=v8:1832
TEST=cctest/test-heap-profiler/*Weak*
TBR=vitalyr@chromium.org
Initial CL: http://codereview.chromium.org/
8716009
Review URL: http://codereview.chromium.org/
8822019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10182
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Tue, 6 Dec 2011 17:21:48 +0000 (17:21 +0000)]
Don't track Smi->Double->Object element transitions for small undefined arrays.
R=jkummerow@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8816009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10181
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Tue, 6 Dec 2011 15:31:01 +0000 (15:31 +0000)]
Remove unnecessary use of raw pointers in hydrogen graph builder.
This is potentially a source of bugs, if someone inserts some code
that causes allocation.
Review URL: http://codereview.chromium.org/
8818015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10180
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Tue, 6 Dec 2011 14:30:21 +0000 (14:30 +0000)]
Use the old handler for low frequency idle notifications.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8818005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10179
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 13:30:22 +0000 (13:30 +0000)]
Fix presubmit.
Review URL: http://codereview.chromium.org/
8821016
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10178
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 13:14:46 +0000 (13:14 +0000)]
Fixing mozilla test failures regarding Math.pow.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8820011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10177
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Tue, 6 Dec 2011 13:09:05 +0000 (13:09 +0000)]
Handle possible difference between function code and activated code on preparation for break points
The full code activated for a function might not be the same code as
is currently the active full code for a function. There where some
asumptions in the debugger preparation for break points.
Unfortunately there is currently no regression test.
R=jkummerow@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org//
8818010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10176
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Tue, 6 Dec 2011 13:00:40 +0000 (13:00 +0000)]
Elide write barriers and remove some heap_object->GetHeap() calls on
Smi write barriers.
Review URL: http://codereview.chromium.org/
8822008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10175
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Tue, 6 Dec 2011 12:11:08 +0000 (12:11 +0000)]
Port to x64 and ARM and some refactoring of ia32.
Review URL: http://codereview.chromium.org/
8111006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10174
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Tue, 6 Dec 2011 12:09:11 +0000 (12:09 +0000)]
The memory size test makes no sense on the simulated platforms.
Review URL: http://codereview.chromium.org/
8818007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10173
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 11:56:56 +0000 (11:56 +0000)]
Remove early NaN/Infinity check in MathPowStub.
Review URL: http://codereview.chromium.org/
8817015
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10172
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Tue, 6 Dec 2011 09:41:06 +0000 (09:41 +0000)]
Hydrogen support for stack local harmony bindings in function scope.
This is the first CL in a series that add support for the harmony scoping
features to crankshaft. This CL specifically adds support for stack
allocated 'let' and 'const' declared variables in function scopes.
TEST=mjsunit/harmony/block-let-crankshaft.js
Review URL: http://codereview.chromium.org/
8806012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10171
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 09:37:50 +0000 (09:37 +0000)]
Fix presubmit.
Review URL: http://codereview.chromium.org/
8816010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10170
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 09:20:55 +0000 (09:20 +0000)]
Stop skipping regress-397.js
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8804013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10169
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 09:20:28 +0000 (09:20 +0000)]
Skip check for +/-0.5 in optimized Math.pow (ia32).
Review URL: http://codereview.chromium.org/
8820007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10168
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 09:20:00 +0000 (09:20 +0000)]
Fixing fix for MathPowHalf on ARM.
Review URL: http://codereview.chromium.org/
8817012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10167
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Tue, 6 Dec 2011 08:28:12 +0000 (08:28 +0000)]
Fixing MathPowHalf on ARM.
BUG=v8:397
TEST=regress-397.js
Review URL: http://codereview.chromium.org/
8800009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10166
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Mon, 5 Dec 2011 21:54:45 +0000 (21:54 +0000)]
Improve TraceIC logging
Explicitly log generic stub transitions and the reason that they are happening.
R=jkummerow@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/
8803013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10165
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 5 Dec 2011 17:21:55 +0000 (17:21 +0000)]
Revert r10158 "Distinguish weak references in heap snapshots, group GC roots."
Heap profiler tests fail on Win32, need to investigate.
This reverts commit
b5374ebd92c11ac4aae16b4e31e54166d406e490.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10162
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Mon, 5 Dec 2011 16:36:13 +0000 (16:36 +0000)]
Fixing MathPowHalf on x64.
BUG=v8:397
TEST=regress-397.js
Review URL: http://codereview.chromium.org/
8805011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10159
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mikhail.naganov@gmail.com [Mon, 5 Dec 2011 16:35:57 +0000 (16:35 +0000)]
Distinguish weak references in heap snapshots, group GC roots.
Several changes to better organize snapshot data:
1. Provide information about weak references.
2. Group (GC roots) children.
3. Prettify debug snapshot printing.
BUG=v8:1832
TEST=cctest/test-heap-profiler/*Weak*
Review URL: http://codereview.chromium.org/
8716009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10158
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Mon, 5 Dec 2011 15:58:25 +0000 (15:58 +0000)]
Refactor MathPowHalf on ia32.
BUG=v8:397, v8:1848
TEST=regress-397.js
Review URL: http://codereview.chromium.org/
8806010
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10157
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Mon, 5 Dec 2011 14:43:28 +0000 (14:43 +0000)]
Statically check for assignments to const in harmony mode.
The ES.next draft rev 4 in section 11.13 reads:
It is a Syntax Error if the AssignmentExpression is contained in extended code
and the LeftHandSideExpression is an Identifier that does not statically resolve
to a declarative environment record binding or if the resolved binding is an
immutable binding.
This CL adds corresponding static checks for the immutable binding case.
TEST=mjsunit/harmony/block-const-assign
Review URL: http://codereview.chromium.org/
8688007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10156
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Mon, 5 Dec 2011 11:44:17 +0000 (11:44 +0000)]
MIPS: updated test .status files based mostly on the ARM version.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8572032
Patch from Gergely Kis <gergely@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10155
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Mon, 5 Dec 2011 09:54:15 +0000 (09:54 +0000)]
Fixing Math.pow(NaN, -0.0) == 1 on Windows.
Review URL: http://codereview.chromium.org/
8804005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10150
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Mon, 5 Dec 2011 09:53:43 +0000 (09:53 +0000)]
Fix special handling of DefineOwnProperty on arrays.
According to the ES5 spec the implementation of DefineOwnProperty() has
to special case handling of arrays. This implementation correctly
handles definitions of array index properties and redefinitions of the
length property.
R=svenpanne@chromium.org
BUG=v8:1756
TEST=test262
Review URL: http://codereview.chromium.org/
8776045
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10149
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Mon, 5 Dec 2011 08:58:12 +0000 (08:58 +0000)]
Fix performance on the page cycler, which is not designed to test
GC speed.
Review URL: http://codereview.chromium.org/
8804004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10148
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Mon, 5 Dec 2011 08:58:01 +0000 (08:58 +0000)]
Clean up multi byte nop support on x64 to more closely match IA32.
Fix missing instruction in disassembler.
Fix wrong disassembly of multi-byte NOP on x64
Add test of disassembler on 64 bit!
Review URL: http://codereview.chromium.org/
8773039
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10147
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Mon, 5 Dec 2011 07:57:44 +0000 (07:57 +0000)]
Fixing windows build.
"src\store-buffer.cc(72) : warning C4244: 'initializing' : conversion from 'intptr_t' to 'int', possible loss of data"
Review URL: http://codereview.chromium.org/
8804003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10146
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Fri, 2 Dec 2011 15:15:23 +0000 (15:15 +0000)]
Fix a bug with deoptimization from inside the default-clause of a switch-statement.
When generating a string-switch we have to use the correct ast id when
there is a default clause present.
Review URL: http://codereview.chromium.org/
8776048
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10145
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 2 Dec 2011 14:19:53 +0000 (14:19 +0000)]
Reduce minimum memory use on 64 bit to one page per semispace.
Review URL: http://codereview.chromium.org/
8774045
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10144
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 2 Dec 2011 14:08:12 +0000 (14:08 +0000)]
Let store buffer start out small for a 1Mbyte saving in boot
memory use (2Mbyte on x64).
Review URL: http://codereview.chromium.org/
8776032
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10143
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Fri, 2 Dec 2011 13:42:51 +0000 (13:42 +0000)]
Temporarily disable regress-397 until fix has been ported to all platforms.
Review URL: http://codereview.chromium.org/
8775051
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10142
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Fri, 2 Dec 2011 13:23:44 +0000 (13:23 +0000)]
Fix presubmit error.
TBR=sgjesse@chromium.org
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/
8770042
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10141
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Fri, 2 Dec 2011 13:16:49 +0000 (13:16 +0000)]
Quickfix for DoMathPowHalf.
TEST=regress-397.js
Review URL: http://codereview.chromium.org/
8769037
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10140
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Fri, 2 Dec 2011 13:05:51 +0000 (13:05 +0000)]
Fix flaky debugger test
TBR=sgjesse@chromium.org
BUG=none
TEST=test/cctest/test-debug.cc
Review URL: http://codereview.chromium.org/
8773037
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10139
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Fri, 2 Dec 2011 12:42:35 +0000 (12:42 +0000)]
Optimize Crankshaft array literal initialization from boilerplate.
BUG=none
TEST=test/mjsunit/array-literal-transitions.js
Review URL: http://codereview.chromium.org/
8747009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10138
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 2 Dec 2011 10:05:20 +0000 (10:05 +0000)]
Fix lint
Review URL: http://codereview.chromium.org/
8773033
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10135
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Fri, 2 Dec 2011 10:01:50 +0000 (10:01 +0000)]
Support multi-byte nop instructions as recommended by
the optimization guides from AMD and Intel.
Review URL: http://codereview.chromium.org/
8776033
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10134
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Fri, 2 Dec 2011 08:06:37 +0000 (08:06 +0000)]
Implement Math.pow using FPU instructions and inline it in crankshaft (ia32).
Review URL: http://codereview.chromium.org/
8749002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10133
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Thu, 1 Dec 2011 16:57:54 +0000 (16:57 +0000)]
Revert r10118 from bleeding edge.
It causes several test failures which I need to investigate.
Review URL: http://codereview.chromium.org/
8769008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10132
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Thu, 1 Dec 2011 14:40:13 +0000 (14:40 +0000)]
MIPS: Port array literal changes on ARM.
Port r10085 (
1ef7d5eb2).
This patch also contains a fix for StoreArrayLiteralElementStub::Generate which is only used from this code.
Original commit message:
Includes general array boilerplate copier and re-introduction FAST_ELEMENT optimizations in full-codegen.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8743010
Patch from Daniel Kalmard <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10129
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Thu, 1 Dec 2011 14:32:26 +0000 (14:32 +0000)]
MIPS: Implement crankshaft support for nested object literals.
Port r10061 (
c648e1d1)
Original commit message:
This generates optimized code for deep-copying of nested object literal
boilerplates which are statically known. Most of the boilerplates have
already been generated at crankshaft time, so this optimization should
kick in for virtually every object literal. Only nested object literal
graphs up to a certain depth and containing up to a certain total number
of properties are considered for this optimization. This will prevent
explosion of code size due to large object literals (e.g. eval on JSON).
Improves splay performance because object literals are created often.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8745012
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10128
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Thu, 1 Dec 2011 14:19:34 +0000 (14:19 +0000)]
Insert proper padding between lazy deoptimization points and safepoints.
On x64 we need 13 bytes for patching the call for lazy deopt. We have to
make sure that patching does not overwrite the code at the safepoint.
BUG=v8:1847
Review URL: http://codereview.chromium.org/
8775009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10126
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Thu, 1 Dec 2011 14:12:30 +0000 (14:12 +0000)]
MIPS: Implement code stub for object literal creation.
Port r10056 (
ffa6eedd8).
Original commit message:
This just ports r10036 to the ARM codegenerator. Please see the original
revision for a detailed description.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8747011
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10124
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Thu, 1 Dec 2011 12:32:38 +0000 (12:32 +0000)]
Introduce a test of how much memory it takes to boot up V8.
Linux only at the moment, since it uses /proc. Shows that
there is room for improvement, espcially on 64 bit.
Review URL: http://codereview.chromium.org/
8750001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10122
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Thu, 1 Dec 2011 12:17:19 +0000 (12:17 +0000)]
Use the old idle notification handler when context is disposed.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8770003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10121
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Thu, 1 Dec 2011 09:54:05 +0000 (09:54 +0000)]
Simplify stack check instruction in Crankshaft.
So far we had two types of stack checks: one used for function entries
and one used at loop back edges which uses a deferred code object to
avoid spilling of registers in the loop.
After refactoring lazy deoptimization the first stack check can also
use deferred code. This change removes the first type of stack check
instruction in Crankshaft and uses a deferred stack check in all
places.
Review URL: http://codereview.chromium.org/
8775002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10118
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Thu, 1 Dec 2011 08:39:22 +0000 (08:39 +0000)]
MIPS: Add external strings support to regexp in generated code.
Port r10070 (
4d7e5a8a)
Original commit message:
TEST=test/mjsunit/string-external-cached.js
BUG=
Review URL: http://codereview.chromium.org/
8744012
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10109
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Thu, 1 Dec 2011 08:31:09 +0000 (08:31 +0000)]
MIPS: Porting r10023 and r10054 to arm (pointer cache for external strings).
Port r10059 (
c32f98a9)
Original commit message:
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8746015
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10107
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
svenpanne@chromium.org [Thu, 1 Dec 2011 08:14:30 +0000 (08:14 +0000)]
Prepare push to trunk. Now working on version 3.8.0
R=danno@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8770001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10104
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ricow@chromium.org [Thu, 1 Dec 2011 07:30:57 +0000 (07:30 +0000)]
Increase tick interval for the android platform.
The phones and tablets currently in use are much slower that desktop
and laptop machines for which the current heuristics have been tuned.
Review URL: http://codereview.chromium.org/
8757017
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10103
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Wed, 30 Nov 2011 18:04:12 +0000 (18:04 +0000)]
Fix build with GCC 4.7, which fails with "narrowing conversion of 'id' from 'int' to 'unsigned int' inside { } is ill-formed in C++11"
Contributed by burnus@net-b.de
Review URL: http://codereview.chromium.org/
8724003
Patch from Tobias Burnus <burnus@net-b.de>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10101
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Wed, 30 Nov 2011 17:53:26 +0000 (17:53 +0000)]
Fix a compiler error on Windows introduced by r10093
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8749008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10100
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Wed, 30 Nov 2011 16:55:55 +0000 (16:55 +0000)]
Fix compile errors on Windows and Mac introduced by r10093.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8746021
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10099
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 30 Nov 2011 16:47:54 +0000 (16:47 +0000)]
MIPS: Implement Math.tan in generated code.
Port r10067 (
593c1ea) and r10069 (
87a06dc).
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8743009
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10098
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
yangguo@chromium.org [Wed, 30 Nov 2011 16:36:00 +0000 (16:36 +0000)]
MIPS: Catch non-string subject in RegExpExecStub.
Port r10064 (
9c6789a3)
Original commit message:
There is no test case to trigger any crash. This is only to guard against the case that the native function is called with unsafe arguments.
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8742012
Patch from Daniel Kalmar <kalmard@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10097
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Wed, 30 Nov 2011 16:00:47 +0000 (16:00 +0000)]
Only make function literals in top-level object literals pretenured and constant function properties.
This change is a refinement of http://code.google.com/p/v8/source/detail?r=7283. Making all
function literals inside object literals constant function properties causes us to lose map
sharing. This hurts code where many object literals are created.
The idea is that top-level code is mostly executed once and functions inside top-level
object literals remain fast with this change.
BUG=v8:1795
Review URL: http://codereview.chromium.org/
8746018
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10096
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Wed, 30 Nov 2011 14:14:49 +0000 (14:14 +0000)]
Fix a bug in the register allocator.
Avoid spilling a live range that has a register use at the next instruction.
The register use position has to be after the end of the next instruction so
that we don't end up trying to split a live range at the beginning.
BUG=105112
Review URL: http://codereview.chromium.org/
8743011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10095
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
sgjesse@chromium.org [Wed, 30 Nov 2011 11:48:35 +0000 (11:48 +0000)]
Fix handling of recompiling code for optimized and inlined functions
The debugger preparation did not take optimized functions - including
inlined function into account. This caused the full-code used for
deoptimization to be the "lazy compile" builtin which did not work and
caused V8 to crash.
R=yangguo@chromium.org
BUG=chromium:105375, v8:1782
TEST=test/mjsunit/debug-break-inline.js
Review URL: http://codereview.chromium.org//
8728031
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10094
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
ulan@chromium.org [Wed, 30 Nov 2011 11:13:36 +0000 (11:13 +0000)]
Start incremental marking on idle notification.
BUG=v8:1458
TEST=cctest/test-api/IdleNotification*
Review URL: http://codereview.chromium.org/
8519002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10093
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 30 Nov 2011 10:38:16 +0000 (10:38 +0000)]
Be more willing to expand old space when evacuating new space at the end of
a mark-sweep. We have a soft limit on old space size, which is designed to
trigger an old-space collection when we hit it. Unfortunately although the
soft limit had already triggered an old space collection, the soft limit was
preventing objects from new space from being promoted. For every promotion
candidate we were checking 3 different ways to allocate in old space before
giving up and putting the object in the other semispace. This change allows
the promoted objects to go to old space and also makes us more eager to
sweep a page before trying other ways to find space for an object.
Review URL: http://codereview.chromium.org/
8748005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10092
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Wed, 30 Nov 2011 09:33:03 +0000 (09:33 +0000)]
Elide write barrier.
Review URL: http://codereview.chromium.org/
8659039
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10091
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Tue, 29 Nov 2011 14:21:47 +0000 (14:21 +0000)]
Prepare push to trunk. Now working on version 3.7.12.
R=erik.corry@gmail.com
Review URL: http://codereview.chromium.org/
8725032
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10088
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
fschneider@chromium.org [Tue, 29 Nov 2011 12:39:28 +0000 (12:39 +0000)]
Fix bug when generating padding to ensure space for lazy deoptimization.
BUG=v8:1846
Review URL: http://codereview.chromium.org/
8725030
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10087
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Tue, 29 Nov 2011 12:17:59 +0000 (12:17 +0000)]
Fix windows build.
Review URL: http://codereview.chromium.org/
8725029
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10086
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Tue, 29 Nov 2011 12:09:06 +0000 (12:09 +0000)]
Port array literal changes on ARM.
Includes general array boilerplate copier and re-introduction FAST_ELEMENT optimizations in full-codegen.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/
8472005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10085
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
erik.corry@gmail.com [Tue, 29 Nov 2011 11:56:14 +0000 (11:56 +0000)]
Only sweep one page eagerly unless we are running out of space.
Limit the number of pages that are compacted in a given GC.
Review URL: http://codereview.chromium.org/
8692002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10084
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
kmillikin@chromium.org [Tue, 29 Nov 2011 10:56:11 +0000 (10:56 +0000)]
Remove the static qualifier from functions in header files.
This shaves 416+ KB, just under 1% off the size of the debug d8 executable
on Linux (mostly because the CheckHelper functions for assertions were
getting separate copies for each compilation unit). The difference in
release builds is negligible---a size reduction of 0.1%.
Also, change namespace-level 'static const' variables to remove the static
storage class as it's the default.
R=danno@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/
8680013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10083
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
vegorov@chromium.org [Tue, 29 Nov 2011 10:02:38 +0000 (10:02 +0000)]
When scavenging update source slot before migrating object it points to.
Source slot might belong to a dead old object and we might allocate a new object over it when evacuating a new space object this slot points to. In this case if we update slot after migrating object we will write into migrated object.
R=erik.corry@gmail.com
Review URL: http://codereview.chromium.org/
8698022
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10082
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
mstarzinger@chromium.org [Tue, 29 Nov 2011 08:48:51 +0000 (08:48 +0000)]
Fix assertion while marking Harmony weak maps.
R=erik.corry@gmail.com
BUG=v8:1845
TEST=mjsunit/harmony/proxies-example-membrane
Review URL: http://codereview.chromium.org/
8715004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10081
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Tue, 29 Nov 2011 08:43:14 +0000 (08:43 +0000)]
Reapply "Fix the ScopeIterator reimplementation".
This reapplies a fixed version of r10076 that also works on arm. Patch set one is r10076 reapplied and patch set 2 contains the new fix.
Review URL: http://codereview.chromium.org/
8725001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10080
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
keuchel@chromium.org [Tue, 29 Nov 2011 06:38:04 +0000 (06:38 +0000)]
Make let/const outside of the extended mode early errors (under harmony flag).
The ES.next drafts require that source code that matches the productions for
let and const bindings outside the extended mode trigger early syntax
errors. This CL adapts the parser / preparser accordingly under the harmony
scoping flag.
Summary:
* Harmony scoping flag not set: Old semantics allowing const in classic mode
with function level scope. Const binding in strict mode and let bindings in
classic and strict mode trigger early syntax errors.
* Harmony scoping is set: Use new harmony const and let in
extended mode and old const in classic mode. This is to preserve
compatibility with current web pages that already use
non-standard implementations of const. An early syntax error is
thrown on const in strict mode and on let in classic and strict
mode.
This depends on:
http://codereview.chromium.org/
8562002/
TEST=mjsunit/harmony/block-early-errors.js
Review URL: http://codereview.chromium.org/
8564001
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10079
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00
danno@chromium.org [Mon, 28 Nov 2011 17:17:46 +0000 (17:17 +0000)]
Reintroduce fast path in full-codegen for VisitArrayLiteral.
BUG=none
TEST=none
Review URL: http://codereview.chromium.org/
8472003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@10078
ce2b1a6d-e550-0410-aec6-
3dcde31c8c00