Commit Graph

26801 Commits

Author SHA1 Message Date
zhengxing.li
4334a81823 X87: [interpreter] Switch passing of new.target to register.
port 3d004eeab2 (r32264)

  original commit message:
  This passes the new.target value in a register instead of through a
  side-channel via the construct stub. The interpreter entry trampoline
  stores this value in a bytecode register so that it can be accessed
  directly by the interpreter. The size of the interpreter stack frame
  hence grows by one slot.

BUG=

Review URL: https://codereview.chromium.org/1475043003

Cr-Commit-Position: refs/heads/master@{#32309}
2015-11-26 03:14:58 +00:00
ahaas
57117b8339 [turbofan] Implemented the optional Float32RoundTiesEven operator.
The Float32RoundTiesEven operator rounds float32 numbers towards the nearest
integer. If the distance to two integers is the same, then the result is
the even integer. This is the default rounding mode of the ieee 754 floating
point standard.

I implemented the optional Float32RoundTiesEven operator on x64, ia32, arm, and arm64.

R=titzer@chromium.org

Review URL: https://codereview.chromium.org/1477753002

Cr-Commit-Position: refs/heads/master@{#32308}
2015-11-25 21:13:20 +00:00
adamk
802ea71e7c Run all message tests with a variant that forces preparsing
This will make sure that message tests cover both the parser and preparser
paths, just as we do for parsing-related cctests.

BUG=v8:4372
LOG=n

Review URL: https://codereview.chromium.org/1469383004

Cr-Commit-Position: refs/heads/master@{#32307}
2015-11-25 21:02:17 +00:00
adamk
481bad1ff0 Reland shipping of --harmony-destructuring-bind
Also fix CheckConflictingVarDeclarations() to properly handle
legacy const bindings. Without that change enabling the flag
causes code like:

  function f() { const x; var x; }

to throw an early error, rather than wait to throw the error
until f is invoked.

The previous patch ran into problems with the fuzzer; that crash was fixed
(with test coverage added) in https://crrev.com/ceb92ebfdfb561d71038793c02b42aa973f55ec4

BUG=v8:811
LOG=y
TBR=rossberg@chromium.org
CQ_INCLUDE_TRYBOTS=tryserver.chromium.linux:linux_chromium_rel_ng;tryserver.blink:linux_blink_rel

Review URL: https://codereview.chromium.org/1473243006

Cr-Commit-Position: refs/heads/master@{#32306}
2015-11-25 20:54:01 +00:00
ofrobots
dc3442b1ec [heap] delete Heap::LeftTrimFixedAray
No more uses left for this code (thank goodness.)

R=hpayer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1474693003

Cr-Commit-Position: refs/heads/master@{#32305}
2015-11-25 20:25:59 +00:00
jarin
d5e5e1e161 [turbofan] Generalize representation inference for NumberAdd.
This should fix the Turbofan Octane/Mandreel regression introduced by

commit 9ea551aa45
Author: jarin <jarin@chromium.org>
Date:   Sun Nov 22 05:45:38 2015 -0800

    [turbofan] Simplify lowering of number addition.

    Review URL: https://codereview.chromium.org/1471533002

    Cr-Commit-Position: refs/heads/master@{#32159}

Review URL: https://codereview.chromium.org/1477803002

Cr-Commit-Position: refs/heads/master@{#32304}
2015-11-25 20:10:24 +00:00
littledan
e5b56115a7 test262 roll
In this new version of test262, a number of known failing tests have
been changed to match the ES2016 semantics, which V8 implements.

R=adamk

Review URL: https://codereview.chromium.org/1475793004

Cr-Commit-Position: refs/heads/master@{#32303}
2015-11-25 20:09:35 +00:00
mtrofin
be7e43614c A simpler way to determine if a range spills only in deferred blocks, by
validating that the hot path does not spill - somewhat simpler code.

Cleared the scenario where a range is defined in a deferred block. The
code before was slightly more complicated by not leveraging the
property that these sort of ranges would be completely contained within
deferred blocks.

Moved "spills in deferred blocks" marking to a more appropriate
location.

One thing this CL achieves is correct support for scenarios where a
range is spilled both on the deferred and then hot path, and the ranges
concatenate. I owe better unit testing, which I will add in a subsequent
CL.

BUG=

Review URL: https://codereview.chromium.org/1472803004

Cr-Commit-Position: refs/heads/master@{#32302}
2015-11-25 20:08:09 +00:00
ahaas
19741ac977 [turbofan] Implemented the optional Float32RoundTruncate operator.
The Float32RoundTruncate operator rounds float32 numbers towards zero.
The operator is currently implemented on x64, ia32, arm, and arm64.

Additionally I added support for the float32 vrintz, vrintn, and vrinta
instructions to the arm simulator.

R=titzer@chromium.org

Review URL: https://codereview.chromium.org/1468303005

Cr-Commit-Position: refs/heads/master@{#32301}
2015-11-25 19:29:02 +00:00
jochen
8d90b92737 Never call CpuFeatures::FlushICache directly
Always go through Assembler::FlushICache so we automatically use the
simulator when running with a simulator.

BUG=v8:2487
R=epertoso@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1480623002

Cr-Commit-Position: refs/heads/master@{#32300}
2015-11-25 19:27:34 +00:00
machenbach
df3dee87a7 Revert of [debugger] flood function for stepping before calling it. (patchset #7 id:120001 of https://codereview.chromium.org/1463803002/ )
Reason for revert:
[Sheriff] Breaks layout tests:
https://build.chromium.org/p/client.v8.fyi/builders/V8-Blink%20Linux%2064/builds/3074

Original issue's description:
> [debugger] flood function for stepping before calling it.
>
> R=verwaest@chromium.org
>
> Committed: https://crrev.com/93eb633214e0f97bf70ae30d2a07b7fbbaa78266
> Cr-Commit-Position: refs/heads/master@{#32285}

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

Review URL: https://codereview.chromium.org/1474943005

Cr-Commit-Position: refs/heads/master@{#32299}
2015-11-25 19:26:46 +00:00
fedor
993ba9d252 [compiler] merge binary-operator-reducer
Merge BinaryOperatorReducer into the MachineOperatorReducer class.
It does not need `Revisit()` calls, because the newly inserted nodes are
visited anyway, and there are no other methods that need AdvancedReducer
there.

BUG=
R=titzer@chromium.org

Review URL: https://codereview.chromium.org/1473073004

Cr-Commit-Position: refs/heads/master@{#32298}
2015-11-25 19:25:17 +00:00
hpayer
8f9a17f44f Transition array should not know about write barrier internals.
BUG=

Review URL: https://codereview.chromium.org/1472393002

Cr-Commit-Position: refs/heads/master@{#32297}
2015-11-25 19:24:29 +00:00
fmeawad
bd1943b711 DEPS change from issue 988893003
Since the gn builders do not resolve the DEPS after patch, the patch cannot contain
both the DEPS change as well the usage of the newly added DEPS.

If the CL mentioned in the title does not land, this CL is to be remove/reverted as well.

BUG=v8:4560
LOG=N

Review URL: https://codereview.chromium.org/1469303004

Cr-Commit-Position: refs/heads/master@{#32296}
2015-11-25 19:23:02 +00:00
ishell
064bbccfa9 Fix promotion of JSFunctions with in-object properties.
BUG=v8:4572, chromium:561481
LOG=Y

Review URL: https://codereview.chromium.org/1465223007

Cr-Commit-Position: refs/heads/master@{#32295}
2015-11-25 19:22:14 +00:00
machenbach
189ec6a023 Revert of Tenure descriptor arrays. (patchset #1 id:1 of https://codereview.chromium.org/1476913002/ )
Reason for revert:
[Sheriff] Breaks gc stress:
https://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20gc%20stress/builds/544

Original issue's description:
> Tenure descriptor arrays.
>
> BUG=
>
> Committed: https://crrev.com/38ec44f3bde8c16448e093b6595d0452e189023f
> Cr-Commit-Position: refs/heads/master@{#32289}

TBR=ishell@chromium.org,hpayer@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=

Review URL: https://codereview.chromium.org/1477003003

Cr-Commit-Position: refs/heads/master@{#32294}
2015-11-25 19:15:56 +00:00
machenbach
699b32574b Reland of [turbofan] Introduce proper JSCreateLiteralRegExp operator. (patchset #1 id:1 of https://codereview.chromium.org/1472423002/ )
Reason for revert:
[Sheriff] Wrong revert.

Original issue's description:
> Revert of [turbofan] Introduce proper JSCreateLiteralRegExp operator. (patchset #2 id:20001 of https://codereview.chromium.org/1475973002/ )
>
> Reason for revert:
> Broke "V8 Linux64 GC Stress - custom snapshot".
>
> http://build.chromium.org/p/client.v8/builders/V8%20Linux64%20GC%20Stress%20-%20custom%20snapshot/builds/2824
>
> Original issue's description:
> > [turbofan] Introduce proper JSCreateLiteralRegExp operator.
> >
> > This adds a new JavaScript level operator for regexp literal creation,
> > similar to what we already have for array and object literals. This
> > once gets lowered to a call to the FastCloneRegExpStub always.
> >
> > R=mstarzinger@chromium.org
> >
> > Committed: https://crrev.com/8659c5d1d287177369ce179a8d0b910192d840d9
> > Cr-Commit-Position: refs/heads/master@{#32288}
>
> TBR=mstarzinger@chromium.org,bmeurer@chromium.org
> NOPRESUBMIT=true
> NOTREECHECKS=true
> NOTRY=true
>
> Committed: https://crrev.com/168683d270d44f05f28acaf7d34c32d0250d2a4c
> Cr-Commit-Position: refs/heads/master@{#32292}

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

Review URL: https://codereview.chromium.org/1478823002

Cr-Commit-Position: refs/heads/master@{#32293}
2015-11-25 19:14:24 +00:00
ishell
168683d270 Revert of [turbofan] Introduce proper JSCreateLiteralRegExp operator. (patchset #2 id:20001 of https://codereview.chromium.org/1475973002/ )
Reason for revert:
Broke "V8 Linux64 GC Stress - custom snapshot".

http://build.chromium.org/p/client.v8/builders/V8%20Linux64%20GC%20Stress%20-%20custom%20snapshot/builds/2824

Original issue's description:
> [turbofan] Introduce proper JSCreateLiteralRegExp operator.
>
> This adds a new JavaScript level operator for regexp literal creation,
> similar to what we already have for array and object literals. This
> once gets lowered to a call to the FastCloneRegExpStub always.
>
> R=mstarzinger@chromium.org
>
> Committed: https://crrev.com/8659c5d1d287177369ce179a8d0b910192d840d9
> Cr-Commit-Position: refs/heads/master@{#32288}

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

Review URL: https://codereview.chromium.org/1472423002

Cr-Commit-Position: refs/heads/master@{#32292}
2015-11-25 18:16:20 +00:00
mlippautz
206ba31245 [heap] Count bits in markbit cell instead of using a table.
Iterating live objects using mark bits is memory bound and already pretty heavy
on the cache as we always need to look at the mark bits, the objects (payload),
and their maps.

This CL changes the object start computation from a table lookup, which was yet
another memory location, to a CPU-bound computation.

BUG=chromium:524425
LOG=N

Review URL: https://codereview.chromium.org/1478623003

Cr-Commit-Position: refs/heads/master@{#32291}
2015-11-25 17:32:12 +00:00
thakis
b60d17eccd Unconditionally expand STATIC_ASSERT to the C++11 form.
Also add a note that it should go away over time.

BUG=none
LOG=n

Review URL: https://codereview.chromium.org/1475033003

Cr-Commit-Position: refs/heads/master@{#32290}
2015-11-25 17:16:44 +00:00
hpayer
38ec44f3bd Tenure descriptor arrays.
BUG=

Review URL: https://codereview.chromium.org/1476913002

Cr-Commit-Position: refs/heads/master@{#32289}
2015-11-25 17:01:49 +00:00
bmeurer
8659c5d1d2 [turbofan] Introduce proper JSCreateLiteralRegExp operator.
This adds a new JavaScript level operator for regexp literal creation,
similar to what we already have for array and object literals. This
once gets lowered to a call to the FastCloneRegExpStub always.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/1475973002

Cr-Commit-Position: refs/heads/master@{#32288}
2015-11-25 16:50:19 +00:00
jochen
27001ca652 Lazily initialize fast_sqrt() and pass an Isolate parameter to it
R=jkummerow@chromium.org
BUG=v8:2487
LOG=n

Review URL: https://codereview.chromium.org/1473683004

Cr-Commit-Position: refs/heads/master@{#32287}
2015-11-25 16:37:28 +00:00
thakis
eb536899b3 Remove COMPILE_ASSERT macro from v8.
It appears unused, everything uses static_assert directly.

BUG=none
LOG=n

Review URL: https://codereview.chromium.org/1465383006

Cr-Commit-Position: refs/heads/master@{#32286}
2015-11-25 16:03:35 +00:00
yangguo
93eb633214 [debugger] flood function for stepping before calling it.
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/1463803002

Cr-Commit-Position: refs/heads/master@{#32285}
2015-11-25 15:45:47 +00:00
hpayer
78d6550a12 Remove whiteness witness from runtime.
BUG=561449
LOG=n

Review URL: https://codereview.chromium.org/1468313007

Cr-Commit-Position: refs/heads/master@{#32284}
2015-11-25 15:32:51 +00:00
jochen
7ba6bb4e3b Pass Isolate to CodeAgingHelper
This is a preparation for requiring an isolate to construct a
CodePatcher

BUG=2487
R=epertoso@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1480573002

Cr-Commit-Position: refs/heads/master@{#32283}
2015-11-25 15:25:15 +00:00
machenbach
8452a75315 [test] Skip flaky test.
BUG=v8:4358
LOG=n
TBR=yangguo@chromium.org, hablich@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1475053002

Cr-Commit-Position: refs/heads/master@{#32282}
2015-11-25 15:23:49 +00:00
ishell
2af5e36547 Allow in-object properties in JSCollections, JSWeakCollections and JSRegExp.
BUG=v8:4531
LOG=Y

Review URL: https://codereview.chromium.org/1468933004

Cr-Commit-Position: refs/heads/master@{#32281}
2015-11-25 15:23:03 +00:00
mythria
46401fcb1c [Interpreter] Add support for compare operators to bytecode graph builder.
Adds implementation and tests for compare operators to bytecode graph builder.

BUG=v8:4280
LOG=N

Review URL: https://codereview.chromium.org/1476473005

Cr-Commit-Position: refs/heads/master@{#32280}
2015-11-25 15:21:38 +00:00
jochen
9ec240d00c Double the API call rate threshold for the memory reducer to kick in
Moving Blink from hidden values to private symbols almost doubled the
rate of API calls, as the private symbols API increases the call-depth
scope while the hidden values one didn't.

I manually checked that this fixes the regression on the long running
gmail benchmark but doesn't add new memory reducer GCs on the infinite
scrolling benchmark

BUG=chromium:561325
R=ulan@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1472193003

Cr-Commit-Position: refs/heads/master@{#32279}
2015-11-25 15:20:13 +00:00
Michael Achenbach
7617c939d6 Whitespace change to test new sp_frame_access coverage.
Cr-Commit-Position: refs/heads/master@{#32278}
2015-11-25 15:05:17 +00:00
jochen
aa9cfc8222 Make whether or not a Code object should be created by masm explicit
We always want to have an Isolate, so just use an extra ctor arg

BUG=2487
R=yangguo@chromium.org,mstarzinger@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1476763002

Cr-Commit-Position: refs/heads/master@{#32277}
2015-11-25 14:23:56 +00:00
machenbach
87177738ce [test] Skip flaky test for turbofan on windows.
BUG=v8:4573
LOG=n
TBR=mstarzinger@chromium.org
NOTRY=true

Review URL: https://codereview.chromium.org/1465193006

Cr-Commit-Position: refs/heads/master@{#32276}
2015-11-25 14:23:08 +00:00
Benedikt Meurer
a6f9670e86 [tools] Unbreak postmortem script.
TBR=machenbach@chromium.org
NOTREECHECKS=true
NOTRY=true

Review URL: https://codereview.chromium.org/1475843003 .

Cr-Commit-Position: refs/heads/master@{#32275}
2015-11-25 14:22:20 +00:00
machenbach
1e82da4b30 [test] Make try-perf script more convenient to use.
Warn if unknown benchmark configurations are triggered.

NOTRY=true

Review URL: https://codereview.chromium.org/1479483003

Cr-Commit-Position: refs/heads/master@{#32274}
2015-11-25 14:00:49 +00:00
bmeurer
2732a6ad44 [es6] Correct parsing of regular expression literal flags.
ES6 section 12.2.8.1 states that flags for regular expression literals
must be checked during parsing and invalid flags are early errors. This
change adapts the Scanner and (Pre)Parser to act according to the spec.

This is also a prerequisite to unify the handling of literal creation
(for Objects, Arrays, Regexps, and at some point Classes).

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/1472323002

Cr-Commit-Position: refs/heads/master@{#32273}
2015-11-25 13:46:43 +00:00
ulan
7a21cdafaf Optimize ClearNonLiveReferences: do not compact prototype transitions in GC.
BUG=chromium:554488
LOG=NO

Review URL: https://codereview.chromium.org/1470773003

Cr-Commit-Position: refs/heads/master@{#32272}
2015-11-25 13:39:13 +00:00
machenbach
e8bfedab96 [test] Add status-file presubmit check.
This loads all test suites and status files to catch subtle
syntax errors. It also checks basic status file integrity
and common mistakes.

NOTRY=true

Review URL: https://codereview.chromium.org/1475663002

Cr-Commit-Position: refs/heads/master@{#32271}
2015-11-25 13:19:33 +00:00
machenbach
95f9f7732f [Ignition] Skip another test.
Started failing after https://codereview.chromium.org/1469313002

NOTRY=true
TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/1476603003

Cr-Commit-Position: refs/heads/master@{#32270}
2015-11-25 13:18:03 +00:00
titzer
9917f3375f Set the constant pool size to 0 on architectures that do not use it.
Turns out we've been putting garbage into code->constant_pool_offset
for quite some time.

R=jkummerow@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1478713002

Cr-Commit-Position: refs/heads/master@{#32269}
2015-11-25 13:10:33 +00:00
ishell
ebf8ec5c51 Fix JSFunction's in-object properties initialization.
BUG=v8:4572
LOG=Y

Review URL: https://codereview.chromium.org/1481493003

Cr-Commit-Position: refs/heads/master@{#32268}
2015-11-25 12:42:14 +00:00
machenbach
05dda9e27e [test-runner] Make test suite loading more robust.
BUG=

Review URL: https://codereview.chromium.org/1466303005

Cr-Commit-Position: refs/heads/master@{#32267}
2015-11-25 12:21:16 +00:00
jarin
0e84cd470f [turbofan] Require proper typing when inferring machine type from big-boy type.
Review URL: https://codereview.chromium.org/1479503002

Cr-Commit-Position: refs/heads/master@{#32266}
2015-11-25 11:49:42 +00:00
verwaest
066747ea05 Make sure that NormalizeElements and ShouldConvertToFastElements are based on the same values
BUG=v8:4518
LOG=n

Review URL: https://codereview.chromium.org/1472293002

Cr-Commit-Position: refs/heads/master@{#32265}
2015-11-25 11:46:50 +00:00
mstarzinger
3d004eeab2 [interpreter] Switch passing of new.target to register.
This passes the new.target value in a register instead of through a
side-channel via the construct stub. The interpreter entry trampoline
stores this value in a bytecode register so that it can be accessed
directly by the interpreter. The size of the interpreter stack frame
hence grows by one slot.

R=oth@chromium.org
BUG=v8:4544
LOG=n

Review URL: https://codereview.chromium.org/1469313002

Cr-Commit-Position: refs/heads/master@{#32264}
2015-11-25 11:37:28 +00:00
jochen
5335e7d55a Clarify how to replace ForceSet
BUG=none
R=vogelheim@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1473783003

Cr-Commit-Position: refs/heads/master@{#32263}
2015-11-25 11:17:54 +00:00
ahaas
4f4947898d [turbofan] Implemented the optional Float32RoundUp operator.
The Float32RoundUp operator rounds float32 numbers towards infinity.
The operator is currently implemented on x64, ia32, arm, and arm64.

R=titzer@chromium.org

Review URL: https://codereview.chromium.org/1474963002

Cr-Commit-Position: refs/heads/master@{#32262}
2015-11-25 11:08:21 +00:00
ahaas
74434403f6 [turbofan] Implemented the optional Float32RoundDown operator.
I implemented the optional Float32RoundDown operator on x64, ia32, arm,
and arm64.

For arm I also had to adjust the simulator.

R=titzer@chromium.org

Review URL: https://codereview.chromium.org/1471913006

Cr-Commit-Position: refs/heads/master@{#32261}
2015-11-25 11:06:13 +00:00
machenbach
b9d5126930 Revert of Reland "[Interpreter] Add CreateClosure to BytecodeGraphBuilder." (patchset #1 id:1 of https://codereview.chromium.org/1475793003/ )
Reason for revert:
[Sheriff] Breaks cfi:
https://build.chromium.org/p/client.v8/builders/V8%20Linux64%20-%20cfi/builds/1209

Original issue's description:
> Reland "[Interpreter] Add CreateClosure to BytecodeGraphBuilder."
>
> Original issue's description:
> > [Interpreter] Add CreateClosure to BytecodeGraphBuilder.
> >
> > Adds code and tests to support CreateClosure bytecode when building
> > graphs.
> >
> > Committed: https://crrev.com/4cceb11b0929abcbc82bf0854554a9b66003335d
> > Cr-Commit-Position: refs/heads/master@{#32224}
>
> BUG=v8:4280
> LOG=N
>
> Committed: https://crrev.com/6a8db006e1f0a08a43446b62765bba39fdc6af10
> Cr-Commit-Position: refs/heads/master@{#32257}

TBR=bmeurer@chromium.org,oth@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:4280

Review URL: https://codereview.chromium.org/1469303006

Cr-Commit-Position: refs/heads/master@{#32260}
2015-11-25 11:04:34 +00:00