verwaest@chromium.org
36e91242fd
Make Object.freeze fast
...
This patch both speeds up the freeze operation itself, but also
allows properties to remain in fast mode. Objects with non-empty
elements backing stores still end up with slow elements.
Relanding r14758 and r14759 with fix for Test262: only mark properties
and elements READ_ONLY if they are not JS setter/getters. Tightened up
tests to assert frozen-ness, and added targeted tests for the new code
(covering accessors).
BUG=v8:1858, 115960
R=verwaest@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15691007
Patch from Adam Klein <adamk@chromium.org>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14762 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-23 07:05:58 +00:00
palfia@homejinni.com
71ab26d291
MIPS: Handle holes in smi-untag from LoadKeyed requiring hole handling.
...
Port r14756 (fda1989f)
BUG=
Review URL: https://codereview.chromium.org/15690003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14761 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 23:09:22 +00:00
adamk@chromium.org
4d48bb832f
Revert "Make Object.freeze fast"
...
and "Fix Object.freeze on dictionary-backed arrays to properly freeze elements"
This reverts r14758 and r14759 due to introducing failures in Test262
TBR=verwaest@chromium.org
Review URL: https://codereview.chromium.org/15681004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14760 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 21:27:00 +00:00
adamk@chromium.org
3ebccb7aae
Fix Object.freeze on dictionary-backed arrays to properly freeze elements
...
Follow-up to r14758: slightly rearranges JSObject::Freeze() to avoid duplicating
code while still retaining proper dictionary elements storage behavior.
Also fix a lint error.
R=verwaest@chromium.org
Review URL: https://codereview.chromium.org/15737018
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14759 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 20:40:04 +00:00
adamk@chromium.org
648e99e308
Make Object.freeze fast
...
This patch both speeds up the freeze operation itself, but also
allows properties to remain in fast mode. Objects with non-empty
elements backing stores still end up with slow elements.
BUG=v8:1858, 115960
R=verwaest@chromium.org
Review URL: https://codereview.chromium.org/14888005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14758 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 18:53:58 +00:00
mstarzinger@chromium.org
b704cb9139
Fix bogus deopt in BuildEmitDeepCopy for holey arrays.
...
R=verwaest@chromium.org
BUG=chromium:242924
TEST=mjsunit/regress/regress-crbug-242924
Review URL: https://codereview.chromium.org/15735012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14757 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 17:58:21 +00:00
verwaest@chromium.org
d1b5a2518d
Handle holes in smi-untag from LoadKeyed requiring hole handling.
...
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15737003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14756 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 16:32:33 +00:00
verwaest@chromium.org
b353b1d131
Don't allow copying holes to fields.
...
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15745006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14753 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 15:33:53 +00:00
mstarzinger@chromium.org
d259e1cebd
Fix corner case in optimized code map zapping.
...
R=jkummerow@chromium.org
TEST=mjsunit/math-floor-part2
Review URL: https://codereview.chromium.org/15743005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14752 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 15:29:02 +00:00
ulan@chromium.org
aadfc39a12
Fix Windows shared library build.
...
We cannot dll-export templates.
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15697006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14751 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 15:12:07 +00:00
ulan@chromium.org
3512dd3bd6
Prepare push to trunk. Now working on version 3.19.4.
...
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15746005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14748 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 14:03:32 +00:00
mstarzinger@chromium.org
bf413b5122
Fix VisitLogicalExpression for empty blocks on RHS.
...
R=jkummerow@chromium.org
BUG=chromium:242870
TEST=mjsunit/regress/regress-crbug-242870
Review URL: https://codereview.chromium.org/15744002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14747 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 13:27:00 +00:00
yangguo@chromium.org
9960b24694
Fix unexpected elements transition in JSON.parse
...
R=verwaest@chromium.org
BUG=241344
Review URL: https://chromiumcodereview.appspot.com/15739003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14746 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 13:24:18 +00:00
verwaest@chromium.org
8db3014974
Keep representations while overwriting transitions.
...
BUG=chromium:241477
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15718002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14745 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 10:46:33 +00:00
mstarzinger@chromium.org
d696f7b3c1
Use explicit type feedback clearing in some tests.
...
R=jkummerow@chromium.org
Review URL: https://codereview.chromium.org/15711004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14744 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 09:17:27 +00:00
hpayer@chromium.org
9c3c28646b
Force GC before executing unbox double arrays test to avoid timeouts.
...
BUG=
Review URL: https://codereview.chromium.org/15292002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14743 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 09:05:22 +00:00
dcarney@chromium.org
64aa48349e
build fix for 14738
...
BUG=
R=svenpanne@chromium.org
Review URL: https://codereview.chromium.org/15688003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14739 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 07:32:36 +00:00
dcarney@chromium.org
1045d62733
implement fast ReturnValue setters
...
R=svenpanne@chromium.org
BUG=
Review URL: https://codereview.chromium.org/15398008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14738 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-22 06:35:38 +00:00
palfia@homejinni.com
ea122ef882
MIPS: fix arm simulator after 14725
...
Port r14731 (046b5d8d)
BUG=
Review URL: https://codereview.chromium.org/15606002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14737 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 23:09:04 +00:00
palfia@homejinni.com
b2729b1b86
MIPS: new style of property/function callbacks
...
Port r14725 (d393d88)
BUG=
Review URL: https://codereview.chromium.org/15562007
Patch from Balazs Kilvady <kilvadyb@homejinni.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14736 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 21:09:58 +00:00
verwaest@chromium.org
f74a85f3b7
Don't track representations in context extensions.
...
This also enables verification of representations.
BUG=
R=jkummerow@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15411003
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14735 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 17:21:27 +00:00
mstarzinger@chromium.org
db4a770c3f
Add regression test for fix from r14732.
...
R=verwaest@chromium.org
BUG=chromium:242502
TEST=mjsunit/regress/regress-crbug-242502
Review URL: https://codereview.chromium.org/15288008
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14734 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 14:20:42 +00:00
mmassi@chromium.org
3947fefe40
a tiny bug in DehoistArrayIndex
...
R=mmassi@chromium.org
Review URL: https://codereview.chromium.org/15268009
Patch from Weiliang Lin <weiliang.lin2@gmail.com>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14733 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 14:09:28 +00:00
verwaest@chromium.org
8fed3da4a4
Zap map value in the stub-cache on clear.
...
The empty_string is a valid key for keyed loads and stores.
BUG=
R=mstarzinger@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15565002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14732 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 12:59:48 +00:00
dcarney@chromium.org
8633086fcd
fix arm simulator after 14725
...
BUG=
R=svenpanne@chromium.org
Review URL: https://codereview.chromium.org/15484006
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14731 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 12:17:04 +00:00
svenpanne@chromium.org
4a7ed948fe
Make sure we register extensions only once. Removed unused member variable.
...
V8::Initialize() can be called from various threads simultaneously, so
everything should be done only once per process or within the Isolate. For
registering extensions, we do the former now.
R=dslomov@chromium.org
Review URL: https://codereview.chromium.org/15564002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14730 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 12:03:49 +00:00
mstarzinger@chromium.org
8743661682
Disable flaky test after r14723.
...
R=svenpanne@chromium.org
BUG=v8:2628
TEST=cctest/test-cpu-profiler/SampleWhenFrameIsNotSetup
Review URL: https://codereview.chromium.org/15415007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14729 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 11:24:20 +00:00
verwaest@chromium.org
9f32d94cb3
Don't create new maps in CurrentMapForDeprecated.
...
R=yangguo@chromium.org
Review URL: https://chromiumcodereview.appspot.com/15358005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14728 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 11:20:24 +00:00
wingo@igalia.com
467b16fd3a
Remove unneeded argument from Parser::GetSymbol
...
Parser::GetSymbol can't actually fail, so no need for the bool* ok
argument or the CHECK_OK in callers.
BUG=
R=mstarzinger@chromium.org
Review URL: https://codereview.chromium.org/15421007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14727 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 10:45:58 +00:00
mstarzinger@chromium.org
a83fc72562
Make Runtime_Abort print to stderr instead of stdout.
...
R=mvstanton@chromium.org
Review URL: https://codereview.chromium.org/15298005
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14726 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 09:25:57 +00:00
dcarney@chromium.org
881476a7af
new style of property/function callbacks
...
R=svenpanne@chromium.org
BUG=
Review URL: https://codereview.chromium.org/12494012
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14725 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-21 06:36:24 +00:00
rodolph.perfetta@gmail.com
45ec481659
ARM: Smi refactoring and improvements.
...
Refactoring:
* consistent use of SmiTag/Untag
* added a few Smi macros and helpers
Improvements
* small optimisations (e.g. merging untag and cmp #0 )
* added fixed point to double conversion instructions for simpler conversions
More on the last point: a Smi can be seen as a fixed point number with the
a one bit fractional part. Fixed to double instructions allow us to convert
a Smi to a double without untagging.
BUG=none
TEST=none
Review URL: https://chromiumcodereview.appspot.com/15085026
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14724 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 15:38:14 +00:00
yurys@chromium.org
3ad62f5ee1
Allow for no samples in test-cpu-profiler/SampleWhenFrameIsNotSetup
...
The test should only check that there are no sample stacks that never possible in the JS code being profiled.
BUG=v8:2628
R=svenpanne@chromium.org
Review URL: https://codereview.chromium.org/14845018
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14723 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 15:10:25 +00:00
rossberg@chromium.org
f61bec13c0
Reduce register pressure in some runtime functions
...
Hoped to cure the spurious 3% regression we saw on beloved chromium-rel-win7-dual/sunspider after http://code.google.com/p/v8/source/detail?r=14236
R=mstarzinger@chromium.org
BUG=chromium:233441
Review URL: https://codereview.chromium.org/15297004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14722 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 13:54:12 +00:00
jkummerow@chromium.org
19e120ad60
Prepare push to trunk. Now working on version 3.19.3.
...
R=danno@chromium.org
Review URL: https://codereview.chromium.org/15295007
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14719 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 13:37:26 +00:00
mvstanton@chromium.org
239b2830cc
Turning off optimize-constructed-arrays to investigate a WebKit/bindings issue.
...
BUG=
R=mstarzinger@chromium.org
Review URL: https://codereview.chromium.org/15303002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14718 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 12:33:48 +00:00
palfia@homejinni.com
ef4ce2e550
MIPS: Unify deoptimizer for JavaScript frames.
...
Port r14715 (84633474)
Original commit message:
This unifies the translation of an optimized frame to a full JavaScript
frame. Only the frame's context and fp register as well as alignment
padding are different on each architecture and can be factored out.
BUG=
Review URL: https://codereview.chromium.org/15291002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14717 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 10:46:42 +00:00
hpayer@chromium.org
d7427aa938
Fix transition test to support allocation site info.
...
BUG=
Review URL: https://codereview.chromium.org/15270002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14716 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 08:56:45 +00:00
mstarzinger@chromium.org
80da843e22
Unify deoptimizer for JavaScript frames.
...
This unifies the translation of an optimized frame to a full JavaScript
frame. Only the frame's context and fp register as well as alignment
padding are different on each architecture and can be factored out.
R=svenpanne@chromium.org
Review URL: https://codereview.chromium.org/14843020
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14715 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 08:27:56 +00:00
danno@chromium.org
58a3fd3666
Fill in one-word-fillers for the unused property fields.
...
BUG=chromium:240056
Review URL: https://chromiumcodereview.appspot.com/15247003
Patch from Toon Verwaest <verwaest@chromium.org>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14714 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 06:58:06 +00:00
verwaest@chromium.org
73d084fad3
Fix bugs in rewriting combined with attributes and accessors
...
R=danno@chromium.org
Review URL: https://chromiumcodereview.appspot.com/14843023
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14713 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 03:16:20 +00:00
palfia@homejinni.com
22fe386d6f
MIPS: Implement yield* (delegating yield)
...
Port r14669 (b64ce42a)
Original commit message:
Ideally this would have been implemented via desugaring at parse-time,
but yield* is an expression, and its desugaring includes statements like
while and try/catch. We'd have to have BlockExpression in the AST to
support that, and it's not worth it for this feature.
So instead we implement all of the logic in
FullCodeGenerator::VisitYield. Delegating yield AST nodes now have a
try handler index, for the try/catch. Otherwise the implementation is
straightforward.
BUG=
Review URL: https://codereview.chromium.org/15210004
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14712 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-17 01:10:09 +00:00
hpayer@chromium.org
c84ac4e260
Revert "Simplifying GC heuristics, deleted old generation allocation limit."
...
This reverts commit 14731029.
BUG=
Review URL: https://codereview.chromium.org/14750013
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14711 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 16:34:56 +00:00
titzer@chromium.org
5746d38351
Fix code gen bug on arm and mips; SeqStringSetChar overwrites a register; Add better default PrintDataTo for HInstruction
...
BUG=
Review URL: https://codereview.chromium.org/14895019
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14710 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 14:27:39 +00:00
dslomov@chromium.org
61c087dcc7
Remove one more unneccessary conversion to double.
...
R=rossberg@chromium.org
Review URL: https://codereview.chromium.org/15204002
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 11:57:23 +00:00
dslomov@chromium.org
1e6397dc3d
Remove d8 implementation of ArrayBuffer and typed arrays.
...
R=rossberg@chromium.org
Committed: https://code.google.com/p/v8/source/detail?r=14706
Review URL: https://codereview.chromium.org/14654011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14708 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 11:55:50 +00:00
dslomov@chromium.org
8bb85af38d
Revert "Remove d8 implementation of ArrayBuffer and typed arrays."
...
This reverts commit r14706.
That commit had a stale change from another change list.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14707 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 11:54:25 +00:00
dslomov@chromium.org
0b4c7c8de9
Remove d8 implementation of ArrayBuffer and typed arrays.
...
R=rossberg@chromium.org
Review URL: https://codereview.chromium.org/14654011
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14706 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 11:48:03 +00:00
rossberg@chromium.org
8ce0718763
Implement Array.observe and emit splice change records for ArrayPush
...
Review URL: https://codereview.chromium.org/14978007
Patch from Rafael Weinstein <rafaelw@chromium.org>.
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14705 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 11:19:37 +00:00
olivf@chromium.org
c3dde4bd9d
Encapsulating Type information in the CompareICStub
...
Encapsulate type information in a convenient wrapper instead of storing it in a naked bitfield. This especially facilitates transitioning to a new state and converting from/to the extraICState representation. Additionally cleaning up ToBooleanICStub::Types for consistency.
BUG=
R=svenpanne@chromium.org
Review URL: https://codereview.chromium.org/14862009
git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14704 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-05-16 10:59:17 +00:00