Commits · d76ad1691c30514ff28717538290fa6e5f82c719 · Linshizhi / V8

02 May, 2014 12 commits

Fix for a GCMole issue introduced in r21117. · d76ad169

ishell@chromium.org authored May 02, 2014

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/269743002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21121 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d76ad169

v8:3303 MultithreadedParallelIsolates has a race condition · 53d7aa29

mvstanton@chromium.org authored May 02, 2014

Disabling test until bug is fixed.
R=svenpanne@chromium.org
BUG=v8:3303
LOG=N

Review URL: https://codereview.chromium.org/263893003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21120 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

53d7aa29

Move cache line size calculation directly into CPU::FlushICache. · 461bb09b

svenpanne@chromium.org authored May 02, 2014

This disentagles the initialization/dependency mess quite a bit and
makes things vastly simpler. If the 'mrs' on every flush is too
expensive (which it is hopefully not), the cache line sizes will have
to be instance variables of the CPU class and FlushICache will have to
be a member function. This would involve some more or less tricky
refactorings, which we shouldn't do until we are *really* forced to
do.

BUG=359977
LOG=y
R=rodolph.perfetta@gmail.com

Review URL: https://codereview.chromium.org/269543016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21119 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

461bb09b

Checks for empty array case added before casting elements to FixedDoubleArray. · b4c1eda0

ishell@chromium.org authored May 02, 2014

BUG=chromium:369450
LOG=N
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/264973008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21118 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

b4c1eda0

Map::Normalize() introduced as single entry point for map normalization and... · ca2def98

ishell@chromium.org authored May 02, 2014

Map::Normalize() introduced as single entry point for map normalization and Map::NotifyLeafMapLayoutChange() made private.

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/263663002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21117 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ca2def98

Generation of our home-grown memmove doesn't depend on serializer state anymore. · 5170a008

svenpanne@chromium.org authored May 02, 2014

The serializer state has to be per-Isolate, but at the point where we
generate our memmoves we don't really have an Isolate. Furthermore,
there was no fundamental reason why we shouldn't use our home-grown
memmove during mksnapshot time.

Perhaps we can totally remove our own memmove nowadays, but this would
be a separate CL.

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

Review URL: https://codereview.chromium.org/261903002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21116 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5170a008

Make zone_allocator actually usable. · 972bfb0d

bmeurer@chromium.org authored May 02, 2014

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/267803005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21113 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

972bfb0d

Fix crash in debug builds introduced with r21110. · 50d8259f

bmeurer@chromium.org authored May 02, 2014

Basically we should not check for map deprecation during
code selection, because that may run on the concurrent
compiler thread. So the fix is to move this logic to the
code generation phase instead, which is always run on the
main thread.

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/263803005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21112 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

50d8259f

Fix |RunMicrotasks()| leaking reference to the last context being run on. · 570fd74f

haraken@chromium.org authored May 02, 2014

RunMicrotasks() executes pending tasks and swaps the old array with a new array. However, the new array contains the reference to the current context as its creation context. This prevents the context from gc-ed until RunMicrotasks() is executed in the different context.

BUG=crbug.com/367016
LOG=y
R=rafaelw@chromium.org, rossberg@chromium.org

Review URL: https://codereview.chromium.org/250883002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21111 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

570fd74f

Don't add code dependencies on transitioning stores eagerly. · 59e5c97e

bmeurer@chromium.org authored May 02, 2014

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/256303007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21110 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

59e5c97e

Object.defineProperty shouldn't be a hint that we're constructing a dictionary. · 7bfc426f

svenpanne@chromium.org authored May 02, 2014

BUG=362870
LOG=y
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/261583004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21109 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

7bfc426f

Drop obsolete HCheckMaps::HandleSideEffectsDominator(). · b7148380

bmeurer@chromium.org authored May 02, 2014

In HCheckMaps::HandleSideEffectsDominator() we try to remove
redundant HCheckMaps instructions that are dominated by an
HStoreNamedField instruction with a transition. This is just
a special case of the check elimination and therefore obsolete
(it also gives some nice speedup since GVN just got cheaper).

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/261003002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21108 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

b7148380

30 Apr, 2014 28 commits

MIPS: CallICStub with a "never patch" approach by default. · 30e5a737

plind44@gmail.com authored Apr 30, 2014

Port r21093 (21e3836)

Original commit message:
Patching will
occur only when custom feedback needs to be gathered (future CLs).

Now rebased on https://codereview.chromium.org/254623002/, which moves the type feedback vector to the SharedFunctionInfo.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/260753004

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21105 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

30e5a737

Remove max space limits in tests. · 56d0b975

hpayer@chromium.org authored Apr 30, 2014

BUG=

Review URL: https://codereview.chromium.org/263703003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21104 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

56d0b975

Limit old space size in test which require a large new space. · 3dd05f8f

hpayer@chromium.org authored Apr 30, 2014

BUG=

Review URL: https://codereview.chromium.org/265673003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21103 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3dd05f8f

When maximum space size flags are set, overwrite the resource constraints. · 25e79809

hpayer@chromium.org authored Apr 30, 2014

BUG=
R=rmcilroy@chromium.org

Review URL: https://codereview.chromium.org/262653002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21102 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

25e79809

ARM64: use jssp for stack slots · d8ad2dd5

m.m.capewell@googlemail.com authored Apr 30, 2014

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

Review URL: https://codereview.chromium.org/262533006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21101 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d8ad2dd5

StringTable::Lookup*IfExist() handlified. · f7e8255f

ishell@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/265553003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21100 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f7e8255f

ContextSlotCache::Update() handlified. · 13b64dfe

ishell@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/263613003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21099 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

13b64dfe

MIPS: Simplify feedback vector creation and store in SharedFunctionInfo. · 016a0419

plind44@gmail.com authored Apr 30, 2014

Port r21085 (782ce81)

LOG=N
BUG=v8:3212
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/267433003

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21098 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

016a0419

Reland "Trigger exception debug event for promises at the throw site." · 7e367ae0

yangguo@chromium.org authored Apr 30, 2014

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/266533003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21097 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

7e367ae0

ScopeInfo::ContextSlotIndex() handlified. · b8008a3e

ishell@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/253263003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21096 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

b8008a3e

Public interface of KeyedLookupCache handlified. · 134ead10

ishell@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/264563003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21095 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

134ead10

Revert "Trigger exception debug event for promises at the throw site." · 792af581

yangguo@chromium.org authored Apr 30, 2014

This reverts r21092.

R=ishell@chromium.org

Review URL: https://codereview.chromium.org/262533009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21094 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

792af581

CallICStub with a "never patch" approach by default. Patching will · 287f65ae

mvstanton@chromium.org authored Apr 30, 2014

occur only when custom feedback needs to be gathered (future CLs).

Now rebased on https://codereview.chromium.org/254623002/, which moves the type feedback vector to the SharedFunctionInfo.

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/247373002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21093 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

287f65ae

Trigger exception debug event for promises at the throw site. · eed0e7e7

yangguo@chromium.org authored Apr 30, 2014

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/260723002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21092 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

eed0e7e7

ARM64: Generate optimized code for Math.floor and Math.round with double outputs. · 67ea9e4b

alexandre.rames@arm.com authored Apr 30, 2014

R=jkummerow@chromium.org, ulan@chromium.org

Review URL: https://codereview.chromium.org/258793002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21091 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

67ea9e4b

Harden more runtime functions · 2008b820

jkummerow@chromium.org authored Apr 30, 2014

R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/255333004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21090 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

2008b820

Do not use default for switch over AllocationSpace. · 528fd3c6

yangguo@chromium.org authored Apr 30, 2014

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/262583004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21089 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

528fd3c6

Small cleanup: AstConstructionVisitor no longer needs a Zone pointer. · f7201231

mvstanton@chromium.org authored Apr 30, 2014

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/265623002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21088 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f7201231

Fix gcmole warning in EnsureSlotContainsAllocationSite(). · d46e3a77

mvstanton@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/268443002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21087 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d46e3a77

Kiss goodbye to MaybeObject. · 88144ee1

yangguo@chromium.org authored Apr 30, 2014

R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/259173003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21086 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

88144ee1

Simplify feedback vector creation and store in SharedFunctionInfo. · 15dc39a8

mvstanton@chromium.org authored Apr 30, 2014

LOG=N
BUG=v8:3212
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/254623002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21085 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

15dc39a8

Unbreak the build after r21083. · d60dbdaf

svenpanne@chromium.org authored Apr 30, 2014

TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/266493003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21084 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d60dbdaf

Added a Isolate* parameter to Serializer::enabled(). · 4dca23f8

svenpanne@chromium.org authored Apr 30, 2014

This parameter will soon be used when a few pseudo-classes like
Serializer are turned into real classes. The current CL is already big
enough, untying our Gordian knot called "startup" will continue...

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

Review URL: https://codereview.chromium.org/260003006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21083 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

4dca23f8

Introduce DropUnderReturnAddress for x64 port. · bf338d78

haitao.feng@intel.com authored Apr 30, 2014

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/242113005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21082 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

bf338d78

Handlification TODOs in scanner and string stream resolved. · 1b3b9220

ishell@chromium.org authored Apr 30, 2014

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/251263004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21081 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

1b3b9220

Introduce PushQuad and PopQuad for x64 port. · 8b489532

haitao.feng@intel.com authored Apr 30, 2014

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/238253013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21080 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

8b489532

Update SafeMove, SafePush, SmiToIndex and SmiToNegativeIndex for x32 port. · 48ef7850

haitao.feng@intel.com authored Apr 30, 2014

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/240473009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21079 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

48ef7850

A new test needs to exit early on non-internationalization builds. · 5e2ee2ba

mvstanton@chromium.org authored Apr 30, 2014

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/265513003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@21078 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5e2ee2ba