Commit Graph

12549 Commits

Author SHA1 Message Date
svenpanne@chromium.org
bc6fe88030 Desugar bitwise negation into XOR and kill all UnaryOp stuff.
R=mstarzinger@chromium.org, verwaest@chromium.org

Review URL: https://codereview.chromium.org/22184004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16073 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 13:34:51 +00:00
dslomov@chromium.org
45f4b685bf Update Array Iterator to use numeric indexes
At the last face-to-face meeting it was decided that we should use
numeric indexes for the Array Iterator values.

https://github.com/rwldrn/tc39-notes/blob/master/es6/2013-07/july-24.md#514-keys-entries-return-numbers-for-array-index-properties

BUG=v8:2818
R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/21180008

Patch from Erik Arvidsson <arv@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16072 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 13:10:07 +00:00
jkummerow@chromium.org
665c45c92d Un-revert "Implement simple effect typing for variables" and "Handle switch effects"
This re-lands r15776 and r15777, reverting the revert in r15786.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/22144006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16071 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 12:57:23 +00:00
bmeurer@chromium.org
548062ea52 r11951 changed the existing check for the __SOFTFP__ define in ARM platforms
to a check for __SOFTP. Most other programs look for __SOFTFP__ [1], so
look for both to increase the chances of not erroring out when using GCC 4.5.

[1] Based on http://codesearch.debian.net/search?q=__SOFTFP__ vs
    http://codesearch.debian.net/search?q=__SOFTFP[^_]

BUG=v8:2140
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22210002

Patch from Raphael Kubo da Costa <raphael.kubo.da.costa@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16070 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 11:08:32 +00:00
yurys@chromium.org
4e6041d595 Use standard routine to compute pointer hashs in heap profiler
BUG=None
R=alph@chromium.org, yangguo@chromium.org

Review URL: https://codereview.chromium.org/22300005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16069 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 08:49:05 +00:00
verwaest@chromium.org
206a35763a Fix DeferredInstanceMigration to store the result to temp rather than temp to result.
Thanks to Akos Palfi for finding the bug.

BUG=
R=ulan@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22388002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16068 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 08:26:31 +00:00
yurys@chromium.org
e06343431a Return start/end profiling time in microseconds instead of milliseconds
The start and end time are now measured in microseconds and the type is int64_t.
This way it seems more natural as we are going to support submilisecond sampling
rate soon. Also it fixes cctest/test-cpu-profiler/ProfileStartEndTime test
failure caused by comparison between long double and double.

TEST=cctest/test-cpu-profiler/ProfileStartEndTime
BUG=v8:2824
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22155003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16067 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 08:00:58 +00:00
danno@chromium.org
61a496f4e4 Prepare push to trunk. Now working on version 3.20.14.
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22277006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16064 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 07:31:23 +00:00
dcarney@chromium.org
982fa4d381 delete eternal handle blocks correctly
TBR=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22371002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16063 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 06:34:54 +00:00
palfia@homejinni.com
8575ae5530 MIPS: Migrate instance of deprecated maps in HCheckMaps.
Port r16057 (b73ae514)

Original commit message:
Currently only direct map checks are supported. Otherwise only polymorphic cases with a generic fallback behave properly, regular polymorphic cases still need to be adapted.

BUG=

Review URL: https://codereview.chromium.org/22254007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16062 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 01:12:10 +00:00
haitao.feng@intel.com
fe7df8c703 Make some constants' meaning clear for X64
R=danno@chromium.org

Review URL: https://codereview.chromium.org/21721002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16061 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-06 00:09:01 +00:00
palfia@homejinni.com
3b347f45a6 MIPS: Replace HCheckPrototypeMaps by explicit map checks of constant values.
Port r16055 (7567e57f)

BUG=

Review URL: https://codereview.chromium.org/21639003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16060 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 17:04:10 +00:00
verwaest@chromium.org
2622158aa0 Change environment padding value to value with smi representation.
BUG=
R=ulan@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22226002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16059 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 16:53:28 +00:00
mstarzinger@chromium.org
e9f6a2d7b4 Use first hit when dumping heap constants.
This makes sure that the first hit is used when dumping heap constants
and hence "TheHoleValue" is not overwritten with other roots that are
initialized to the hole as well.

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/21639004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16058 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 16:47:45 +00:00
verwaest@chromium.org
9d9930ce99 Migrate instance of deprecated maps in HCheckMaps.
Currently only direct map checks are supported. Otherwise only polymorphic cases with a generic fallback behave properly, regular polymorphic cases still need to be adapted.

R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/21536003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16057 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 16:42:39 +00:00
jkummerow@chromium.org
14239ab9fb Add "benchmarks" test suite
R=machenbach@chromium.org

Review URL: https://codereview.chromium.org/21645003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16056 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 14:54:09 +00:00
verwaest@chromium.org
04b5e7fa63 Replace HCheckPrototypeMaps by explicit map checks of constant values.
R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/21065006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16055 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 13:45:16 +00:00
hpayer@chromium.org
983a8621e0 More cleanup regarding the maximum non-large object allocation size.
BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/20867003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16052 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 12:52:53 +00:00
haitao.feng@intel.com
e8fe083eb0 Introduce Push and Pop register macro instructions for all platforms
R=danno@chromium.org

Review URL: https://codereview.chromium.org/22041003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16051 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 12:43:04 +00:00
bmeurer@chromium.org
274f254236 Revert "Return start/end profiling time in microseconds instead of milliseconds"
This reverts r16049 for breaking build on windows.

TBR=svenpanne@chromium.org,machenbach@chromium.org

Review URL: https://codereview.chromium.org/22189002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16050 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 12:27:12 +00:00
yurys@chromium.org
d38bbe354b Return start/end profiling time in microseconds instead of milliseconds
The start and end time are now measured in microseconds and the type is int64_t. This way it seems more natural as we are going to support submilisecond sampling rate soon. Also it fixes cctest/test-cpu-profiler/ProfileStartEndTime test failure caused by comparison between long double and double.

TEST=cctest/test-cpu-profiler/ProfileStartEndTime
BUG=v8:2824
R=alph@chromium.org, bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22172002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16049 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 11:48:24 +00:00
jochen@chromium.org
cb68e2cd9b Expose JSON parser through V8 API
BUG=v8:2821
TEST=cctest/test-api/JSONParse
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/21959003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16048 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 11:14:46 +00:00
dcarney@chromium.org
47c3a081f1 fix 16045
TBR=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22169002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16047 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 10:04:50 +00:00
dcarney@chromium.org
207396101f introduce eternal handles
R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/21133006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16045 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 09:46:23 +00:00
mstarzinger@chromium.org
a50827e044 Add explicit transition flag to HStoreNamedField.
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22164003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16044 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 09:35:18 +00:00
yangguo@chromium.org
b58093c953 Add notice when parallel recompilation is disabled by tracing.
BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22164002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16043 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 09:28:22 +00:00
yangguo@chromium.org
1ed4df6b57 Disable parallel recompilation for --trace-hydrogen-stubs.
R=bmeurer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/21830004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16042 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 09:02:47 +00:00
verwaest@chromium.org
251c191824 Fix the loop-builder to tag the entire body as part of the loop.
R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/21813004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16041 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 08:59:55 +00:00
dcarney@chromium.org
14ed15e7e1 Make GlobalHandle::NodeBlock deletable
R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/21042004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16040 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 07:34:29 +00:00
yurys@chromium.org
411d21b2b1 Add start and end profiling time to v8::CpuProfile
I'm going to change CPU profiler API and deprecate GetSelfTime, GetTotalTime and GetTotalSamplesCount on CpuProfileNode as all of those values are derived from self samples count and sampling rate. The sampling rate in turn is calculate based on the profiling duration so having start/end time and total sample count is enough for calculating smpling rate.

BUG=267595
R=alph@chromium.org, bmeurer@chromium.org

Review URL: https://codereview.chromium.org/21918002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16039 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 07:17:08 +00:00
svenpanne@chromium.org
e544f130c5 Remove unused UNARY_MINUS builtin.
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22146002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16038 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 07:00:38 +00:00
dcarney@chromium.org
eb52c66fd0 IsNearDeath needs to include pending nodes
R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/21466003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16037 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 06:58:48 +00:00
svenpanne@chromium.org
d5b9616991 Clean some unuse code of unary negation
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22089002

Patch from Weiliang Lin <weiliang.lin2@gmail.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16036 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-05 06:34:26 +00:00
palfia@homejinni.com
ea38ee8782 MIPS: Avoid redundant smi check for Math.abs
Port r16021 (d79f4450)

BUG=

Review URL: https://codereview.chromium.org/21903003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16035 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 18:35:19 +00:00
haitao.feng@intel.com
dfda85a91a Introduce PopReturnAddressTo and PushReturnAddressFrom macro-assembler instructions for X64
R=danno@chromium.org

Review URL: https://codereview.chromium.org/21477002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16034 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 13:42:02 +00:00
dslomov@chromium.org
5230c19d8b Add size_t length argument to v8::ArrayBuffer::Allocator::Free.
The previous implementation of Free is a deprecated overload now.

R=mstarzinger@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=16031

Review URL: https://codereview.chromium.org/21803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16033 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 13:03:06 +00:00
dslomov@chromium.org
f8b80ca66d Revert "Add size_t length argument to v8::ArrayBuffer::Allocator::Free."
This reverts r16031 for breaking shared build.

TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/21818003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16032 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 12:56:53 +00:00
dslomov@chromium.org
1688f3c167 Add size_t length argument to v8::ArrayBuffer::Allocator::Free.
The previous implementation of Free is a deprecated overload now.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/21803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16031 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 12:19:22 +00:00
jkummerow@chromium.org
232a2c0d88 Regression test for issue 2813 / r16008
BUG=v8:2813
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/21806002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16030 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 12:17:19 +00:00
svenpanne@chromium.org
bf71023ba5 Replaced unary negation by multiplication with -1.
This fixes a deopt loop in the Epic Citadel demo and removes some code. Apart from that, this change is performance-neutral.

When we do something similar for BIT_NOT, the whole UnaryOp stuff can go away.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21782002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16029 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 11:56:35 +00:00
mstarzinger@chromium.org
1e296936eb Store transition on HStoreNamedField as HConstant.
This allows optimization passes that run in the parallel compiler thread
to use the map that a store transitions to for further analysis even
though the map handle cannot be dereferenced.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21560002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16028 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 11:24:55 +00:00
titzer@chromium.org
af2663afd8 Remove instructions and infrastructure related to IDEFs that is now obsolete (and was never turned on). The new bounds check elimination phase doesn't make use of these features, as they were the first parts of the previous approach which was never completed.
BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/21579003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16027 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 11:17:26 +00:00
machenbach@chromium.org
a37040b275 Make optimized debug builds link against debug CRT.
We recently had some XP issues on chromium.win after https://codereview.chromium.org/19384011/.

This might fix it - quote: "The most likely cause of the problem is a gyp (or similar) change that resulted in v8.dll linking against the release version of the crt (for a debug build). There are two MS VC related files: the crt and the c++ std lib (MSVCR100.DLL and MSVCP100D.DLL). The former is set to release, the latter to debug."

I am not entirely sure if this change fits to the linker options for debug level 2.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21776002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16026 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 10:59:23 +00:00
mstarzinger@chromium.org
c87abd1117 Add new Harmony methods to Array.prototype object.
Array.prototype.find
Array.prototype.findIndex

http://people.mozilla.org/~jorendorff/es6-draft.html

BUG=v8:2776,v8:2777
TEST=mjsunit/harmony/array-find,mjsunit/harmony/array-findindex
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/21079003

Patch from Ioseb Dzmanashvili <ioseb.dzmanashvili@gmail.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16025 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 10:57:48 +00:00
loislo@chromium.org
d2c443b774 Extract hardcoded error strings into a single place and replace them with enum.
I'd like to propagate bailout reason to cpu profiler.
So I need to save it into heap object SharedFunctionInfo.
But:
1) all bailout reason strings spread across all the sources.
2) they are native strings and if I convert them into String then I may have a performance issue.
3) one byte is enough for 184 bailout reasons. Otherwise we need 8 bytes for the pointer.

Also I think it would be nice to have error strings collected in one place.
In that case we will get additional benefits:

It allows us to keep this set of messages under control.
It gives us a chance to internationalize them.
It slightly reduces the binary footprint.

From the other hand the developers have to add new strings into that enum.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/20843012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16024 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 09:53:11 +00:00
mstarzinger@chromium.org
477b41c89b Fix Mac build after r16022.
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21786002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16023 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 09:35:44 +00:00
mstarzinger@chromium.org
12c8d36331 Generate grokdump constants with 'make grokdump' now.
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/21771002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16022 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 09:02:02 +00:00
jkummerow@chromium.org
a47705644e Avoid redundant smi check for Math.abs
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21180004

Patch from Weiliang Lin <weiliang.lin2@gmail.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16021 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 08:59:02 +00:00
jkummerow@chromium.org
15c1d23fbc Use test(reg, imm) when the operand is a register on ia32
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21521002

Patch from Weiliang Lin <weiliang.lin2@gmail.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16020 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 08:55:26 +00:00
jkummerow@chromium.org
865582e02b TSan: Move definition of AtomicOps_x86CPUFeatureStruct into v8::internal namespace
This matches other atomicops_internals_* files.

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/21534002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16019 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-02 08:52:59 +00:00