Commits · 15e0189c97118f5fa92a1e70a90bb800ea923601 · Linshizhi / V8

28 Apr, 2014 13 commits

Fix and improve Map::CurrentMapForDeprecatedInternal(). · 15e0189c

bmeurer@chromium.org authored Apr 28, 2014

Inline relevant bits from Map::FindUpdatedMap() and Map::IsMoreGeneralThan()
into Map::CurrentMapForDeprecatedInternal() to fix issues introduced
with field type tracking, avoid the useless second pass over the transition
tree, and finally make it easier to understand what this method actually
does.

TEST=mjsunit/regress/regress-365172-2
R=svenpanne@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=20997

Review URL: https://codereview.chromium.org/257893004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21010 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

15e0189c

Revert "Fix and cleanup Map::GeneralizeRepresentation()." · 5396d4e9

bmeurer@chromium.org authored Apr 28, 2014

This reverts commit r21006 for performance regressions.

TBR=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/259003002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21009 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5396d4e9

ARM: don't use r9 anymore for gap resolution · eabcf296

m.m.capewell@googlemail.com authored Apr 28, 2014

BUG=
R=bmeurer@chromium.org, ulan@chromium.org

Review URL: https://codereview.chromium.org/255883002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21008 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

eabcf296

Revert r20974: Unify and simplify the FastCloneShallowArrayStub · 9aaa5367

danno@chromium.org authored Apr 28, 2014

Due to Layout test redness.

TBR=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/256873007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21007 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9aaa5367

Fix and cleanup Map::GeneralizeRepresentation(). · 0377bccd

bmeurer@chromium.org authored Apr 28, 2014

TEST=mjsunit/regress/regress-365172-[1-3],mjsunit/field-type-tracking
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/259993004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21006 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0377bccd

Merge v8converions with conversions · 629725d4

jochen@chromium.org authored Apr 28, 2014

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

Review URL: https://codereview.chromium.org/250793009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21005 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

629725d4

Use InitialPropertiesLength to reinitialize rather than local calculation. · 0521c124

verwaest@chromium.org authored Apr 28, 2014

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/256973003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21004 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0521c124

Expose promise value through promise mirror. · da0ca2af

yangguo@chromium.org authored Apr 28, 2014

R=rossberg@chromium.org, yurys@chromium.org
BUG=v8:3093
LOG=Y

Review URL: https://codereview.chromium.org/258823012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21003 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

da0ca2af

Convert function.name to API-style accessor and make CallApiGetterStub serializable. · 60b14ea6

ulan@chromium.org authored Apr 28, 2014

BUG=
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/254783003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21002 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

60b14ea6

Simplified CPU/CpuFeatures a bit. · 2fe0f48d

svenpanne@chromium.org authored Apr 28, 2014

This is a necessary intermediate step to disentangle the startup.

In the long run CPU and CpuFeatures should probably be merged,
and Serializer::enabled usage should be radically reduced, but
we're not there yet.

BUG=359977
LOG=y
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/258993002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21001 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

2fe0f48d

Introduce DisallowDeoptimization scope. · 574cfbcc

bmeurer@chromium.org authored Apr 28, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/254763007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20998 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

574cfbcc

Fix and improve Map::CurrentMapForDeprecatedInternal(). · b922b0dd

bmeurer@chromium.org authored Apr 28, 2014

Inline relevant bits from Map::FindUpdatedMap() and Map::IsMoreGeneralThan()
into Map::CurrentMapForDeprecatedInternal() to fix issues introduced
with field type tracking, avoid the useless second pass over the transition
tree, and finally make it easier to understand what this method actually
does.

TEST=mjsunit/regress/regress-365172-2
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/257893004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20997 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

b922b0dd

Fix CurrentMapForDeprecated() to return MaybeHandle instead of a null handle. · 88ca76bc

bmeurer@chromium.org authored Apr 28, 2014

Also fix TryMigrateInstance() to return bool instead of the parameter or
a null handle.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/251683003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20996 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

88ca76bc

25 Apr, 2014 27 commits

Add timestamps to CPU profile samples. · a9655b7d

alph@chromium.org authored Apr 25, 2014

BUG=363976
LOG=Y
R=bmeurer@chromium.org, yurys@chromium.org

Review URL: https://codereview.chromium.org/259803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20993 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

a9655b7d

Fix for a GC-stress failures after r20987. · 0628775c

ishell@chromium.org authored Apr 25, 2014

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/253513004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20992 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0628775c

MIPS: Unify and simplify the FastCloneShallowArrayStub. · 24651e5b

plind44@gmail.com authored Apr 25, 2014

Port r20974 (5de4367)

Original commit message:
- Don't bake in length/capacity into full codegen calls of stubs,
allowing boilerplates to increase their capacity without regenerating
code.
- Unify all variants of the clone stub into a single,
length-independent version.
- Various tweaks to make sure that the clone stub doesn't spill and
therefore need an eager stack frame.
- Handle all lengths of array literals in the fast case.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/258813002

Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20991 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

24651e5b

Runtime_KeyedGetProperty handlified. · 584e90f5

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/253603002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20990 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

584e90f5

ARM64: Fix LCodeGen::ToOperand32. · 486a9fdb

Jacob.Bramley@arm.com authored Apr 25, 2014

This fixes the following generated code sequence:
  movn w1, #0     // Synthesize -1.
  cmp w0, w1

With a properly-constructed Operand, the MacroAssembler can optimize it
as follows:
  cmn w0, #1

BUG=
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/253513003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20989 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

486a9fdb

Expose promise status through promise mirror. · 81a10167

yangguo@chromium.org authored Apr 25, 2014

R=aandrey@chromium.org, rossberg@chromium.org, yurys@chromium.org
BUG=v8:3093
LOG=Y

Review URL: https://codereview.chromium.org/257803005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20988 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

81a10167

HashTableKey::AsObject() handlified. · ce14ab97

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/250773004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20987 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ce14ab97

*Shape::AsObject() are no longer used - removed. · dc1fe284

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/256773002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20986 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

dc1fe284

Remove static CallCompletedCallback handlers · c27da0c9

jochen@chromium.org authored Apr 25, 2014

BUG=none
R=svenpanne@chromium.org
LOG=y

Review URL: https://codereview.chromium.org/249313002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20985 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

c27da0c9

Enter a description of the change. · 8795d02b

jochen@chromium.org authored Apr 25, 2014

Reland 20968 - "Merge v8utils.* and utils.*"

>  BUG=none
>  R=mstarzinger@chromium.org
>  LOG=n
>
>  Review URL: https://codereview.chromium.org/256753002

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

Review URL: https://codereview.chromium.org/250733004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20984 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

8795d02b

Reland 20962 "Break cyclic reference between utils and platform." · 65a983a4

jochen@chromium.org authored Apr 25, 2014

Don't use OS::MemCopy in BitCast. It's crucial that the compiler can
optimize the memcpy away.

>  Platform includes utils.h for the definition of Vector<>, so utils.h
>  can't include platform.h and has to use memcpy instead of OS::MemCopy.
>
>  We split out Vector<> into its own header to break this cycle
>  dependency.
>
>  BUG=none
>  R=mstarzinger@chromium.org
>  LOG=n
>
>  Review URL: https://codereview.chromium.org/251753002

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

Review URL: https://codereview.chromium.org/252683004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20983 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

65a983a4

ObjectHashTable's key and WeakHashTable's key types are now Handle<Object> instead of Object*. · 3ce91203

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/257853003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20982 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3ce91203

Dictionary::SetEntry() and Dictionary::AddEntry() handlified. · a5bde585

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/250913003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20981 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

a5bde585

HashTable::New() handlified. · fd02e122

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/257633002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20980 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

fd02e122

Reland r20960: "HashTable::EnsureCapacity() handlified." · 88734238

ishell@chromium.org authored Apr 25, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/256743002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20979 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

88734238

Fix materialization of accessor frames with captured receivers · ff884e06

jarin@chromium.org authored Apr 25, 2014

I have fixed skipping of the receiver object to materialize captured
objects. This is done with a new DoTranslateSkip method.

We should consider unifying DoTranslateSkip, DoTranslateObject and
DoTranslateCommand as they do the almost the same thing - they only
differ in where they store the result.

The change also turns bunch of ASSERTs into CHECKs.

R=mstarzinger@chromium.org
BUG=359441
TEST=test/mjsunit/regress/regress-359441.js
LOG=N

Review URL: https://codereview.chromium.org/225283006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20978 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ff884e06

Move JSON-stringify webkit test to V8. · 19aa122d

ulan@chromium.org authored Apr 25, 2014

BUG=367089
LOG=N
R=machenbach@chromium.org

Review URL: https://codereview.chromium.org/254673005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20977 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

19aa122d

Revert 20962 "Break cyclic reference between utils and platform." · 4ed16eae

jochen@chromium.org authored Apr 25, 2014

>  Platform includes utils.h for the definition of Vector<>, so utils.h
>  can't include platform.h and has to use memcpy instead of OS::MemCopy.
>
>  We split out Vector<> into its own header to break this cycle
>  dependency.
>
>  BUG=none
>  R=mstarzinger@chromium.org
>  LOG=n
>
>  Review URL: https://codereview.chromium.org/251753002

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

Review URL: https://codereview.chromium.org/254823002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20976 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

4ed16eae

Revert 20968 - "Merge v8utils.* and utils.*" · f2a1176f

jochen@chromium.org authored Apr 25, 2014

>  BUG=none
>  R=mstarzinger@chromium.org
>  LOG=n
>
>  Review URL: https://codereview.chromium.org/256753002

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

Review URL: https://codereview.chromium.org/250823003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20975 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f2a1176f

Unify and simplify the FastCloneShallowArrayStub · 198c4769

danno@chromium.org authored Apr 25, 2014

- Don't bake in length/capacity into full codegen calls of stubs,
  allowing boilerplates to increase their capacity without regenerating
  code.
- Unify all variants of the clone stub into a single,
  length-independent version.
- Various tweaks to make sure that the clone stub doesn't spill and
  therefore need an eager stack frame.
- Handle all lengths of array literals in the fast case.

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/257563004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20974 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

198c4769

Fix for potential memory leak after putting to WeakObjectToCodeDependency table. · 9a842209

ishell@chromium.org authored Apr 25, 2014

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/254723004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20973 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9a842209

Let releases script retrieve information about chromium branches. · 0516f641

machenbach@chromium.org authored Apr 25, 2014

BUG=
R=jarin@chromium.org

Review URL: https://codereview.chromium.org/238443005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20972 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0516f641

Preserve Smi representation of non-escaping fields. · d557425a

jarin@chromium.org authored Apr 25, 2014

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/251493004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20971 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d557425a

Fix compilation with clang 5.1 on OS X. · a87b6b64

svenpanne@chromium.org authored Apr 25, 2014

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/250873003

Patch from Benedikt Meurer <bmeurer@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20970 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

a87b6b64

Always include debugger support. · cb2f43cb

yangguo@chromium.org authored Apr 25, 2014

Motivation: we do not have test coverage for debuggersupport=off.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/256653004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20969 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

cb2f43cb

Merge v8utils.* and utils.* · c0380f6a

jochen@chromium.org authored Apr 25, 2014

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

Review URL: https://codereview.chromium.org/256753002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20968 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

c0380f6a

Don't adopt the AST id from previous if id is none, since previous may have... · d2179f20

verwaest@chromium.org authored Apr 25, 2014

Don't adopt the AST id from previous if id is none, since previous may have mismatching expected stack height.
Additionally, harden merging of simulates after instructions with side effects and ensure there's a simulate before HEnterInlined.

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/252583004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20967 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d2179f20