Commits · 69ea09dd15008874f012a8a5d4b571dc0feac0b6 · Linshizhi / V8

04 Dec, 2015 2 commits

X87: [crankshaft] Deoptimize if HHasInPrototypeChainAndBranch hits a proxy. · 69ea09dd

zhengxing.li authored Dec 04, 2015

  port a330af0e (r32539)

  original commit message:
  The optimized code generated by Crankshaft cannot properly deal
  with proxies (in the prototype chain), and there's probably no
  point in trying to make that work^Wfast with Crankshaft at all.
  TurboFan will handle that properly; Crankshaft just bails out
  to fullcodegen, which then goes to the runtime, which should do
  the right thing soon.

BUG=

Review URL: https://codereview.chromium.org/1495803003

Cr-Commit-Position: refs/heads/master@{#32594}

69ea09dd

Update V8 DEPS. · b8c7b5ae

v8-autoroll authored Dec 04, 2015

Rolling v8/buildtools to b747a9e091cb8212a62343258406eaf53a6c032e

Rolling v8/tools/clang to 58b11d197b1c5fec39a6ec374c21b3d3a5bf557b

Rolling v8/tools/swarming_client to 9cdd76171e517a430a72dcd7d66ade67e109aa00

TBR=machenbach@chromium.org,vogelheim@chromium.org,hablich@chromium.org

Review URL: https://codereview.chromium.org/1493273004

Cr-Commit-Position: refs/heads/master@{#32593}

b8c7b5ae

03 Dec, 2015 38 commits

GetKeys(): Revert all-can-read behavior for Object.keys until LayoutTests are fixed · b2424215
jkummerow authored Dec 03, 2015
```
Review URL: https://codereview.chromium.org/1493373003

Cr-Commit-Position: refs/heads/master@{#32592}
```
b2424215

PPC: Fix inobject slack tracking for both subclassing and non-subclassing cases. · 23d838f9

mbrandy authored Dec 03, 2015

Port 5d38d681

Original commit message:
    It didn't support subclassing case at all and in non-subclassing case the runtime
    allocation didn't do the slack tracking step.

R=ishell@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=chromium:563339
LOG=N

Review URL: https://codereview.chromium.org/1493553007

Cr-Commit-Position: refs/heads/master@{#32591}

23d838f9

[es6] Correctify and unify ArrayBuffer and SharedArrayBuffer constructors. · 3235ccbb

bmeurer authored Dec 03, 2015

The ArrayBuffer and SharedArrayBuffer constructors should raise an
exception when called with no arguments or undefined length.  Also
unified the ArrayBuffer and SharedArrayBuffer implementations as C++
builtins, and removed some (now) obsolete runtime entries.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/1500543002

Cr-Commit-Position: refs/heads/master@{#32590}

3235ccbb

PPC: [debugger] do not restart frames that reference new.target for liveedit.c · 9290dd82

mbrandy authored Dec 03, 2015

Port 6fca8702

R=yangguo@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=

Review URL: https://codereview.chromium.org/1497793002

Cr-Commit-Position: refs/heads/master@{#32589}

9290dd82

Revert of [proxies] Make Object.prototype.isPrototypeOf work with proxies.... · 48fba943

machenbach authored Dec 03, 2015

Revert of [proxies] Make Object.prototype.isPrototypeOf work with proxies. (patchset #2 id:20001 of https://codereview.chromium.org/1492863002/ )

Reason for revert:
[Sheriff] Speculative revert for:
https://uberchromegw.corp.google.com/i/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/3225

Can just be relanded if it doesn't get green.

Original issue's description:
> [proxies] Make Object.prototype.isPrototypeOf step into proxies.
>
> R=rossberg, verwaest@chromium.org
> BUG=v8:1543
> LOG=n
>
> Committed: https://crrev.com/4ca1180d2e7c409312ae0761cb12843989466573
> Cr-Commit-Position: refs/heads/master@{#32569}

TBR=rossberg@chromium.org,verwaest@chromium.org,neis@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:1543

Review URL: https://codereview.chromium.org/1494283002

Cr-Commit-Position: refs/heads/master@{#32588}

48fba943

Reland of [debugger] do not restart frames that reference new.target for... · 15cb3fde

machenbach authored Dec 03, 2015

Reland of [debugger] do not restart frames that reference new.target for liveedit. (patchset #1 id:1 of https://codereview.chromium.org/1493863004/ )

Reason for revert:
Didn't help...

Original issue's description:
> Revert of [debugger] do not restart frames that reference new.target for liveedit. (patchset #1 id:1 of https://codereview.chromium.org/1493363002/ )
>
> Reason for revert:
> [Sheriff] Speculative revert for https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/3225
>
> Original issue's description:
> > [debugger] do not restart frames that reference new.target for liveedit.
> >
> > R=mstarzinger@chromium.org
> >
> > Committed: https://crrev.com/6fca870240bdbb07a365189b5eb0c98fa65b3682
> > Cr-Commit-Position: refs/heads/master@{#32572}
>
> TBR=mstarzinger@chromium.org,yangguo@chromium.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Committed: https://crrev.com/1a61dab34b9849f3f70a42ce69317e22758c53a1
> Cr-Commit-Position: refs/heads/master@{#32582}

TBR=mstarzinger@chromium.org,yangguo@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1492393003

Cr-Commit-Position: refs/heads/master@{#32587}

15cb3fde

PPC: [proxies] InstanceOfStub should bailout to %HasInPrototypeChain for proxies. · c902d4ff

mbrandy authored Dec 03, 2015

Port 0e956833

Original commit message:
    Whenever the InstanceOfStub finds a proxy (either passed as object or
    somewhere on the prototype chain), it should bailout to the
    %HasInPrototypeChain runtime function, which will do the right thing
    (soonish).

R=bmeurer@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=v8:1543
LOG=n

Review URL: https://codereview.chromium.org/1495513004

Cr-Commit-Position: refs/heads/master@{#32586}

c902d4ff

PPC: [fullcode] Switch passing of new.target to register. · 3950206f

mbrandy authored Dec 03, 2015

Port 440a42b7

Original commit message:
    This passes the new.target value in a register instead of through a
    side-channel via the construct stub. Note that this marks the last
    consumer of said side-channel and the special slot in the construct
    stub frame can be removed as a follow-up.

R=mstarzinger@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
TEST=mjsunit/es6/regress/regress-new-target-context
BUG=

Review URL: https://codereview.chromium.org/1494003004

Cr-Commit-Position: refs/heads/master@{#32585}

3950206f

PPC: Remove new.target value from construct stub frames. · 65421613

mbrandy authored Dec 03, 2015

Port eaa0e596

Original commit message:
    This drops the specific slot containing the new.target value from our
    construct stub frames. This side-channel has been deprecated and will
    no longer be accessed by any consumers.

R=mstarzinger@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=

Review URL: https://codereview.chromium.org/1494593005

Cr-Commit-Position: refs/heads/master@{#32584}

65421613

PPC: [crankshaft] Deoptimize if HHasInPrototypeChainAndBranch hits a proxy. · 90e4179f

mbrandy authored Dec 03, 2015

Port a330af0e

Original commit message:
    The optimized code generated by Crankshaft cannot properly deal
    with proxies (in the prototype chain), and there's probably no
    point in trying to make that work^Wfast with Crankshaft at all.
    TurboFan will handle that properly; Crankshaft just bails out
    to fullcodegen, which then goes to the runtime, which should do
    the right thing soon.

R=bmeurer@chromium.org, joransiu@ca.ibm.com, jyan@ca.ibm.com, michael_dawson@ca.ibm.com
BUG=v8:1543
LOG=n

Review URL: https://codereview.chromium.org/1496843004

Cr-Commit-Position: refs/heads/master@{#32583}

90e4179f

Revert of [debugger] do not restart frames that reference new.target for... · 1a61dab3

machenbach authored Dec 03, 2015

Revert of [debugger] do not restart frames that reference new.target for liveedit. (patchset #1 id:1 of https://codereview.chromium.org/1493363002/ )

Reason for revert:
[Sheriff] Speculative revert for https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/3225

Original issue's description:
> [debugger] do not restart frames that reference new.target for liveedit.
>
> R=mstarzinger@chromium.org
>
> Committed: https://crrev.com/6fca870240bdbb07a365189b5eb0c98fa65b3682
> Cr-Commit-Position: refs/heads/master@{#32572}

TBR=mstarzinger@chromium.org,yangguo@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1493863004

Cr-Commit-Position: refs/heads/master@{#32582}

1a61dab3

Preserve information about dots in numbers across parser rewriting. · 1e4681c3

bradnelson authored Dec 03, 2015

Fix several operations in the parser that rewrite constant expressions
to preserve knowledge regarding whether a value originally contained a ".".
This information is required to accurately validate Asm.js typing.

Making the assumption that if either side of a binary operation contains
a dot, that the rewritten expression should be treated as a double for
Asm.js purposes. This is a slight deviation from the spec (which
would forbid mix type operations).

BUG= https://code.google.com/p/v8/issues/detail?id=4203
TEST=test-asm-validator, test-parsing
R=titzer@chromium.org,marja@chromium.org,aseemgarg@chromium.org
LOG=N

Review URL: https://codereview.chromium.org/1492123002

Cr-Commit-Position: refs/heads/master@{#32581}

1e4681c3

Restructure GetDerivedMap so there's only one place where we read intrinsicDefaultProto · 46a93665
verwaest authored Dec 03, 2015
```
BUG=

Review URL: https://codereview.chromium.org/1500683002

Cr-Commit-Position: refs/heads/master@{#32580}
```
46a93665

[proxies] Adapt and reenable harmony/proxies-for.js test. · 384ec6dc

neis authored Dec 03, 2015

R=rossberg
BUG=v8:1543
LOG=n

Review URL: https://codereview.chromium.org/1491863002

Cr-Commit-Position: refs/heads/master@{#32579}

384ec6dc

Remove (now) unused GetPropertyWithHandler. · 8a70e9f9

neis authored Dec 03, 2015

R=cbruni@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1499573004

Cr-Commit-Position: refs/heads/master@{#32578}

8a70e9f9

[turbofan, arm64] Fix native stack parameters on arm64. · 282e9411

ahaas authored Dec 03, 2015

I added a flag to the CallDescriptor which indicates that the native
stack should be used for a CallObject instead of the js stack on arm64.

Additionally I removed the use of EmitPrepareArguments because the
current implementation does not work when float and int parameters are
mixed. I plan to fix it in a future CL, because currently I have a
problem figuring out the type of a parameter.

R=titzer@chromium.org, v8-arm-ports@googlegroups.com

Review URL: https://codereview.chromium.org/1494123002

Cr-Commit-Position: refs/heads/master@{#32577}

282e9411

[turbofan] Introduce ToBooleanHints on ToBoolean operators. · 9298b430

bmeurer authored Dec 03, 2015

Extract ToBoolean hints from the fullcodegen code object and put them
into the ToBoolean nodes created by the AstGraphBuilder.  We currently
do not yet consume this feedback, that will be done in a followup CL.

R=mstarzinger@chromium.org
BUG=v8:4583
LOG=n

Review URL: https://codereview.chromium.org/1494973002

Cr-Commit-Position: refs/heads/master@{#32576}

9298b430

Fixing gcc warning 'variable tracking size limit exceeded' · e667ae4b

ivica.bogosavljevic authored Dec 03, 2015

Warning in test-asm-validator.cc fixed by splitting the function causing the warning into two functions. This
is how it has been done earlier this way, e.g. https://codereview.chromium.org/1164893003

BUG=

Review URL: https://codereview.chromium.org/1491583002

Cr-Commit-Position: refs/heads/master@{#32575}

e667ae4b

Revert "Use WeakCells in the optimized code map rather than traversing in pause." · 39b207dd

neis authored Dec 03, 2015

Reason for revert:
Probably causes GC stress test failures.

TBR=mvstanton@chromium.org
BUG=
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1493393002

Cr-Commit-Position: refs/heads/master@{#32574}

39b207dd

Mark BooleanObject::New() as deprecated · ef3bee66

jochen authored Dec 03, 2015

Embedders should use the factory method that takes an Isolate* instead

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

Review URL: https://codereview.chromium.org/1494503003

Cr-Commit-Position: refs/heads/master@{#32573}

ef3bee66

[debugger] do not restart frames that reference new.target for liveedit. · 6fca8702

yangguo authored Dec 03, 2015

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/1493363002

Cr-Commit-Position: refs/heads/master@{#32572}

6fca8702

[proxies] Add all-can-read/String/Symbol filtering support to GetKeys() · 186f6708

jkummerow authored Dec 03, 2015

Not exposed/used yet, therefore also not testable yet.

Review URL: https://codereview.chromium.org/1491743008

Cr-Commit-Position: refs/heads/master@{#32571}

186f6708

[proxies] do not leak private symbols to proxy traps · 3ed71daf

caitpotter88 authored Dec 03, 2015

BUG=v8:4537
LOG=N
R=neis@chromium.org, rossberg@chromium.org, jkummerow@chromium.org

Review URL: https://codereview.chromium.org/1492923002

Cr-Commit-Position: refs/heads/master@{#32570}

3ed71daf

[proxies] Make Object.prototype.isPrototypeOf step into proxies. · 4ca1180d

neis authored Dec 03, 2015

R=rossberg, verwaest@chromium.org
BUG=v8:1543
LOG=n

Review URL: https://codereview.chromium.org/1492863002

Cr-Commit-Position: refs/heads/master@{#32569}

4ca1180d

[CQ] Add new triggered win trybots to CQ. · 235eff95

machenbach authored Dec 03, 2015

TBR=sergiyb@chromium.org,tandrii@chromium.org
NOTRY=true
BUG=chromium:561530
LOG=n

Review URL: https://codereview.chromium.org/1497743002

Cr-Commit-Position: refs/heads/master@{#32568}

235eff95

Use WeakCells in the optimized code map rather than traversing in pause. · 20a8162f

mvstanton authored Dec 03, 2015

It's expensive to walk all shared function infos during the gc atomic pause. Instead, use WeakCells to implement this structure without manual clearing.

BUG=

Review URL: https://codereview.chromium.org/1478943003

Cr-Commit-Position: refs/heads/master@{#32567}

20a8162f

[heap] pause observers during mark-compact · 471dd3ac

ofrobots authored Dec 03, 2015

Inline allocations performed during mark compact aren't real allocations. They
should not trigger inline-allocation-observer notifications.

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

Review URL: https://codereview.chromium.org/1465223009

Cr-Commit-Position: refs/heads/master@{#32566}

471dd3ac

Revert of [debugger] do not predict step in target for liveedit. (patchset #1... · 10910bc4

machenbach authored Dec 03, 2015

Revert of [debugger] do not predict step in target for liveedit. (patchset #1 id:1 of https://codereview.chromium.org/1491743005/ )

Reason for revert:
[Sheriff] Layout test crashes:
https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/3220

Original issue's description:
> [debugger] do not predict step in target for liveedit.
>
> R=verwaest@chromium.org
>
> Committed: https://crrev.com/8f87ff5d62e996b07ffbde7e735daa603c1d7290
> Cr-Commit-Position: refs/heads/master@{#32553}

TBR=verwaest@chromium.org,yangguo@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1494143002

Cr-Commit-Position: refs/heads/master@{#32565}

10910bc4

Move machine-type.h from src/compiler to src/. · 27433918

titzer authored Dec 03, 2015

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

Review URL: https://codereview.chromium.org/1498833002

Cr-Commit-Position: refs/heads/master@{#32564}

27433918

[turbofan] Make RawMachineAssembler handle the end node. · 33142c12

mstarzinger authored Dec 03, 2015

This moves the proper handling for the end node withing the constructed
graph into the RawMachineAssembler. This simplifies all assemblers and
makes the handling of {Start} and {End} symmetrical.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/1493963003

Cr-Commit-Position: refs/heads/master@{#32563}

33142c12

[Interpreter] Adds support for Increment and Decrement to BytecodeGraphBuilder. · 75f11021

mythria authored Dec 03, 2015

Adds implementation and tests for Inc and Dec to bytecode graph builder.

BUG=v8:4280
LOG=N

Review URL: https://codereview.chromium.org/1499593002

Cr-Commit-Position: refs/heads/master@{#32562}

75f11021

Reland of Introduce instance type for transition arrays. (patchset #1 id:1 of... · 463c130d

hpayer authored Dec 03, 2015

Reland of Introduce instance type for transition arrays. (patchset #1 id:1 of https://codereview.chromium.org/1483003002/ )

Reason for revert:
Suspect for crashing found, relanding for canary coverage.

Original issue's description:
> Revert of Introduce instance type for transition arrays. (patchset #6 id:100001 of https://codereview.chromium.org/1480873003/ )
>
> Reason for revert:
> Broken canary. Trying to find out root cause.
>
> Original issue's description:
> > Introduce instance type for transition arrays.
> >
> > The motivation is to allow specialized marking visitor for transition arrays and collect all transition array in a list for post-processing in ClearNonLiveReferences.
> >
> > BUG=chromium:554488
> > LOG=NO
> >
> > Committed: https://crrev.com/026095a3c7932573e1810b8064ec3008ed696601
> > Cr-Commit-Position: refs/heads/master@{#32396}
>
> TBR=mlippautz@chromium.org,jkummerow@chromium.org,ulan@chromium.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=chromium:554488
>
> Committed: https://crrev.com/38bf70b9cd2a07b99ac0c0b7eda111849e79c146
> Cr-Commit-Position: refs/heads/master@{#32404}

TBR=mlippautz@chromium.org,jkummerow@chromium.org,ulan@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:554488

Review URL: https://codereview.chromium.org/1500623002

Cr-Commit-Position: refs/heads/master@{#32561}

463c130d

Reland of Tenure transition array. (patchset #1 id:1 of... · 2c7aee28

hpayer authored Dec 03, 2015

Reland of Tenure transition array. (patchset #1 id:1 of https://codereview.chromium.org/1485613003/ )

Reason for revert:
Suspect for crashing found, relanding for canary coverage.

Original issue's description:
> Revert of Tenure transition array. (patchset #1 id:1 of https://codereview.chromium.org/1472363007/ )
>
> Reason for revert:
> Broken canary. Trying to find out root cause.
>
> Original issue's description:
> > Tenure transition array.
> >
> > BUG=
> >
> > Committed: https://crrev.com/c88d7cd1e09d921e8a9a2930cfe6f2dfbea3dc90
> > Cr-Commit-Position: refs/heads/master@{#32378}
>
> TBR=ulan@chromium.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=chromium:563378
>
> Committed: https://crrev.com/cdb4b8fba1774f7617e51228c89286bcbd0425f0
> Cr-Commit-Position: refs/heads/master@{#32405}

TBR=ulan@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:563378

Review URL: https://codereview.chromium.org/1492223003

Cr-Commit-Position: refs/heads/master@{#32560}

2c7aee28

Reland of [CQ] Update proto format to fix triggered builders. · edda955e

machenbach authored Dec 03, 2015

Reason for revert:

Should be fixed after
https://chromereviews.googleplex.com/318207014/

TBR=sergiyb@chromium.org,tandrii@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:561530

Review URL: https://codereview.chromium.org/1494893003

Cr-Commit-Position: refs/heads/master@{#32559}

edda955e

[proxies] Make Array.isArray respect proxies. · 82d97479

neis authored Dec 03, 2015

An array is either a JSArray or a proxy whose target is an array.

R=bmeurer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1497483004

Cr-Commit-Position: refs/heads/master@{#32558}

82d97479

[runtime] [proxy] Remove JSProxy::CallTrap · a32096c4

cbruni authored Dec 03, 2015

LOG=n
BUG=v8:1543

Review URL: https://codereview.chromium.org/1500553002

Cr-Commit-Position: refs/heads/master@{#32557}

a32096c4

For non-prototype objects constructed using base==new.target, use the cached... · 324ab707

verwaest authored Dec 03, 2015

For non-prototype objects constructed using base==new.target, use the cached constructor to render the name.

BUG=chromium:563791
LOG=y

Review URL: https://codereview.chromium.org/1494673004

Cr-Commit-Position: refs/heads/master@{#32556}

324ab707

Reland of Do not remove write barriers for stores of old space references in... · fa7a07c3

hpayer authored Dec 03, 2015

Reland of Do not remove write barriers for stores of old space references in most recent old space allocation. (patchset #1 id:1 of https://codereview.chromium.org/1482973003/ )

Reason for revert:
Suspect for crashing found, relanding for canary coverage.

Original issue's description:
> Revert of Do not remove write barriers for stores of old space references in most recent old space allocation. (patchset #1 id:1 of https://codereview.chromium.org/1478113002/ )
>
> Reason for revert:
> Broken canary. Trying to find out root cause.
>
> Original issue's description:
> > Do not remove write barriers for stores of old space references in most recent old space allocation.
> >
> > BUG=chromium:561449
> > LOG=n
> >
> > Committed: https://crrev.com/369778ec55a63ebe51e8fa8497edb5b681069b9b
> > Cr-Commit-Position: refs/heads/master@{#32368}
>
> TBR=ulan@chromium.org,bmeurer@chromium.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
> BUG=chromium:561449
>
> Committed: https://crrev.com/da56525478f1820e3da629576ab61acc5f84daac
> Cr-Commit-Position: refs/heads/master@{#32406}

TBR=ulan@chromium.org,bmeurer@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=chromium:561449

Review URL: https://codereview.chromium.org/1493313002

Cr-Commit-Position: refs/heads/master@{#32555}

fa7a07c3