Commits · 8c54a373dd8585f8ef07f5178b00190faa0dd06d · Linshizhi / V8

28 May, 2014 16 commits

Changing the attributes of a data property implemented with · 8c54a373

mvstanton@chromium.org authored May 28, 2014

ExecutableAccessorInfo turns the property into a field. Better
to keep it as a callback, and correctly deal with the changed
property attributes.

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/262053011

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21558 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

8c54a373

Add Seq[One,Two]ByteString to runtime test generator. · fd21e8f5

mstarzinger@chromium.org authored May 28, 2014

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

Review URL: https://codereview.chromium.org/303543006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21557 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

fd21e8f5

Force inline runtime functions into its wrapper. · c3595595

yangguo@chromium.org authored May 28, 2014

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/307543009

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21556 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

c3595595

Cleanup GetPropertyWithCallback / SetPropertyWithCallback API · 6b712555

verwaest@chromium.org authored May 28, 2014

BUG=
R=ishell@chromium.org

Review URL: https://codereview.chromium.org/305513004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21555 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

6b712555

Drop bogus assert after r21552. · 52cfd35f

mstarzinger@chromium.org authored May 28, 2014

R=hpayer@chromium.org
BUG=v8:2070
LOG=N

Review URL: https://codereview.chromium.org/303493005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21554 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

52cfd35f

Repairs to FLAG_pretenure_call_new. · 5f7fed01

mvstanton@chromium.org authored May 28, 2014

R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/307583002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21553 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5f7fed01

Make incremental marker post-process JSWeakCollection. · 3812677b

mstarzinger@chromium.org authored May 28, 2014

R=hpayer@chromium.org
BUG=v8:2070
LOG=N

Review URL: https://codereview.chromium.org/301553003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21552 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3812677b

X87: Fix a typo in below CL · f10204b5

weiliang.lin@intel.com authored May 28, 2014

r21515 (6d0ec9a)
Convert ElementsKind into a BitField

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/301563008

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21551 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f10204b5

X87: Small changes in preparation for Hydrogen-generated KeyedLoadGeneric · f6ac5ae6

weiliang.lin@intel.com authored May 28, 2014

Port r21535(8bb4c2b)

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/302453011

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21550 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f6ac5ae6

X87: Convert String array index/length hash to BitField. · 0f2f60bd

weiliang.lin@intel.com authored May 28, 2014

Port r21533 (736c779)

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/298293007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21549 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0f2f60bd

X87: Merge Debugger and Debug. · d5c4ad30

weiliang.lin@intel.com authored May 28, 2014

port r21524 (af88824)

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/298193004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21548 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

d5c4ad30

X87: Avoid HeapObject check in HStoreNamedField. · 2312c8ab

weiliang.lin@intel.com authored May 28, 2014

port r21509 (78a54b0)

Original commit message:
This way an HStoreNamedField instruction can never deoptimize
itself, which is another important step towards a working
store elimination.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/306483013

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21547 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

2312c8ab

Extract build configuration into a separate header and move it to the base lib · 6e3ffe1c

jochen@chromium.org authored May 28, 2014

With this, change, atomicops, once, and lazy instance are no longer dependant
on v8 core. I'll move them in a follow-up change to the libbase as well.

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

Review URL: https://codereview.chromium.org/303463005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21546 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

6e3ffe1c

Reuse ZoneList memory if possible. · 1102eeaf

bmeurer@chromium.org authored May 28, 2014

Clear() is rather wasteful for ZoneLists, because it discards
all previously allocated memory, making it impossible to reuse
until the whole Zone is released. Better use Rewind(0) in this
case, which just resets the length to zero.

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/302783007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21545 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

1102eeaf

Fix path tracer test. · 5599b3a7

yangguo@chromium.org authored May 28, 2014

Path tracer is not compatible with --enable-slow-assert.

R=mstarzinger@chromium.org
BUG=v8:3350
LOG=N

Review URL: https://codereview.chromium.org/297193005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21542 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5599b3a7

X87: Skip write barriers in the fast case when setting up local context. · 716be75b

weiliang.lin@intel.com authored May 28, 2014

Port r21481 (5973b48)

Original commit message:
The FastNewContextStub always allocates in new space, so we don't
need to update the write barrier when copying the parameters to
the newly allocated context.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/307433004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21541 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

716be75b

27 May, 2014 24 commits

MIPS: Fix improper register use in DoMinMax. · 382619d3

plind44@gmail.com authored May 27, 2014

This fixes failure on pidigit benchmark from web-shootout
after 840 digits are calculated.

TEST=
BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/300853002

Patch from Dusan Milosavljevic <Dusan.Milosavljevic@rt-rk.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21540 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

382619d3

MIPS: Convert String array index/length hash to BitField. · 47664dc1

plind44@gmail.com authored May 27, 2014

Port r21533 (736c779)

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/297233007

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

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21539 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

47664dc1

Fix check for empty type · 5192a0d0

rossberg@chromium.org authored May 27, 2014

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/307453005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21538 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5192a0d0

MIPS: Use TempDoubleRegister. · a63ea2d3

plind44@gmail.com authored May 27, 2014

Port r21517 (53d6507)

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/305563002

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

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21537 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

a63ea2d3

Reland 21529 - "Add a flag to d8 to invoke weak callbacks" · 88abd8ab

jochen@chromium.org authored May 27, 2014

> This will send an idle notification and a low memory notification after
> each test.
>
> For some reason it's not enough to send a low memory notification alone.
>
> BUG=none
> R=yangguo@chromium.org
> LOG=n

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

Review URL: https://codereview.chromium.org/297213006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21536 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

88abd8ab

Small changes in preparation for Hydrogen-generated KeyedLoadGeneric · 0f929e7a

danno@chromium.org authored May 27, 2014

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/303583002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21535 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0f929e7a

Revert "Make 'name' property on functions configurable." · 6b33e507

mstarzinger@chromium.org authored May 27, 2014

R=danno@google.com, danno@chromium.org

Review URL: https://codereview.chromium.org/297163009

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21534 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

6b33e507

Convert String array index/length hash to BitField · 6fa616ad

danno@chromium.org authored May 27, 2014

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/300023011

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21533 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

6fa616ad

Remove cpp hack to avoid link failures on Windows · 68a1a595

rossberg@chromium.org authored May 27, 2014

TBR=yangguo@chromium.org
BUG=

Review URL: https://codereview.chromium.org/298133005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21532 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

68a1a595

Revert "Add a flag to d8 to invoke weak callbacks" · c92d6f57

yangguo@chromium.org authored May 27, 2014

This reverts r21529.

TBR=jochen@chromium.org

Review URL: https://codereview.chromium.org/302543004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21531 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

c92d6f57

Drop unused includes from worker-thread.cc · f9e0a903

jochen@chromium.org authored May 27, 2014

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

Review URL: https://codereview.chromium.org/302463007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21530 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f9e0a903

Add a flag to d8 to invoke weak callbacks · 4cc92425

jochen@chromium.org authored May 27, 2014

This will send an idle notification and a low memory notification after
each test.

For some reason it's not enough to send a low memory notification alone.

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

Review URL: https://codereview.chromium.org/306483006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21529 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

4cc92425

Establish distributivity for type union & intersection · 7ac892c8

rossberg@chromium.org authored May 27, 2014

This requires introducing proper bounds on all leaf types, so that intersection between bitsets and these types can be accurately represented. Extending a union also becomes more involved.

(On the upside, the modified union/intersect algorithm would now allow support for proper variance for function types.)

Not sure if it is worth landing this. Distributivity isn't really a crucial property for our use cases. It seems fine if intersection is slightly lossy.

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

Review URL: https://codereview.chromium.org/251753005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21528 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

7ac892c8

Replace STATIC_CHECK with STATIC_ASSERT. · 276adeda

jochen@chromium.org authored May 27, 2014

It's just an alias, and STATIC_ASSERT is used way more often

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

Review URL: https://codereview.chromium.org/304553002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21527 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

276adeda

Skip PathTracer test for now. · 4309b34e

yangguo@chromium.org authored May 27, 2014

TBR=mstarzinger@chromium.org
BUG=3350
LOG=N

Review URL: https://codereview.chromium.org/305583002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21526 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

4309b34e

MIPS: Avoid HeapObject check in HStoreNamedField. · 052c36a3

plind44@gmail.com authored May 27, 2014

Port r21509 (78a54b0)

Original commit message:
This way an HStoreNamedField instruction can never deoptimize
itself, which is another important step towards a working
store elimination.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/297143007

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

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21525 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

052c36a3

Merge Debugger and Debug. · 95a9c7e5

yangguo@chromium.org authored May 27, 2014

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/301633005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21524 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

95a9c7e5

Fix PathTracer. · 1630b6d8

yangguo@chromium.org authored May 27, 2014

When tracing, we abuse the map for marking, thereby mutating it.
FixedTypedArrayBase::size() uses the object's map, which causes crash.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/300753002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21523 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

1630b6d8

Cache optimization status getter in mjsunit.js · db8f7e03

yangguo@chromium.org authored May 27, 2014

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/300003007

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21522 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

db8f7e03

Add defensive assert against a weak callback on an empty persistent. · ee88832b

vogelheim@chromium.org authored May 27, 2014

(This shouldn't happen in the first place, as by definition that callback would never be called. However, the referenced bug sorta looks like this happened. If so, the CHECK should help us pinpoint the culprit.)

R=jochen@chromium.org
BUG=368095
LOG=N

Review URL: https://codereview.chromium.org/265823006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21521 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ee88832b

Do not break in native code (including non-builtin debugger code). · 620555b4

yangguo@chromium.org authored May 27, 2014

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/300773002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21520 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

620555b4

Replace some calls to MemCopy with small constant sizes with memcpy · 47cea503

jochen@chromium.org authored May 27, 2014

MemCopy is only meant for variable size, large (>64bytes) copies,
otherwise, it's probably slower than memcpy due to the call
overhead and the compiler can't optimize it away.

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

Review URL: https://codereview.chromium.org/306453005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21519 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

47cea503

Do not (eagerly) trigger exception in mjsunit.js. · 2097644f

yangguo@chromium.org authored May 27, 2014

R=jochen@chromium.org

Review URL: https://codereview.chromium.org/301673002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21518 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

2097644f

ARM: use TempDoubleRegister · 9678ff12

m.m.capewell@googlemail.com authored May 27, 2014

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

Review URL: https://codereview.chromium.org/302453008

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21517 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9678ff12