Commit Graph

12561 Commits

Author SHA1 Message Date
yangguo@chromium.org
c0d1ba2ede Do not materialize uninitialized const for debug evaluate.
R=prybin@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22822002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16148 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-12 13:35:46 +00:00
verwaest@chromium.org
508159bd52 Make assert more accurate
BUG=
R=verwaest@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22623008

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16147 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-12 09:34:55 +00:00
verwaest@chromium.org
641fda8d05 Break typecheck dependency in escape analysis.
R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16146 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-12 09:26:18 +00:00
machenbach@chromium.org
cfb7ef44ca Ignore an unsuitable test under deopt fuzz.
That test relies on certain optimization/deoptimization points and is therefore not useful for the deopt fuzzer.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22475011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16145 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-12 08:59:42 +00:00
jochen@chromium.org
00a2d793b0 Return i18n object holders directly.
BUG=none
R=dcarney@chromium.org,mstarzinger@chromium.org

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16144 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-12 08:19:06 +00:00
verwaest@chromium.org
ee53b0a5ed Make all load-named-fields depend on their map-check, unless explicitly ignored.
BUG=
R=titzer@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22555004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16139 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 18:40:10 +00:00
dslomov@chromium.org
d63e29ba78 Revert "Promote ArrayBuffer, DataView and typed arrays to non-experimental."
This reverts commit r16137 for breaking tests on Windows.

TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22710007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16138 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 16:23:00 +00:00
dslomov@chromium.org
30375b0937 Promote ArrayBuffer, DataView and typed arrays to non-experimental.
The primary reason for this change is to bake these guys into the
snapshot.

Flag definitions (--harmony-typed-arrays, --harmony-array-buffer) are
still there so that Blink does not complain, but they are noop and
default to true.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22390008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16137 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 15:57:31 +00:00
mstarzinger@chromium.org
6989edf771 Make sure polymorphic element access creates non-replaying phis.
R=danno@chromium.org
BUG=v8:2815

Review URL: https://codereview.chromium.org/22589005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16136 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 15:23:15 +00:00
mstarzinger@chromium.org
cc3b7a007f Allow HPhis to have an invalid merge index.
All phis that do not represent local variables or values on the operand
stack are not allowed to carry a merge index, as the replay of the
HEnvironment during LChunkBuilder time might get out of sync due to
colliding indexes.

R=danno@chromium.org
BUG=v8:2815

Review URL: https://codereview.chromium.org/22494003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16135 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 15:18:23 +00:00
yurys@chromium.org
f0cb71a3b8 Set bailout reason to kCodeGenerationFailed only when it is not set yet
BUG=None
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22404007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16134 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 15:10:06 +00:00
verwaest@chromium.org
3715358145 Replace LoadNamedFieldPolymorphic with explicit branches.
R=ulan@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22213002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16133 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 14:04:47 +00:00
bmeurer@chromium.org
770e6ee8b8 Fix/unify root handling in code generation.
R=danno@chromium.org

Review URL: https://codereview.chromium.org/22670004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16132 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 13:43:46 +00:00
mvstanton@chromium.org
201c512dc7 Deoptimization is easier to diagnose when there is a text reason.
BUG=
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/22339018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16131 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 12:50:42 +00:00
verwaest@chromium.org
5f9ff709f0 Reduce some code duplication
R=svenpanne@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22732006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16130 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 12:33:06 +00:00
verwaest@chromium.org
d435819e36 Replace MathFloor/Round by HChange if the input representation does not match the output representation
R=svenpanne@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22398009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16129 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 12:03:40 +00:00
verwaest@chromium.org
19659646ca Fix smi-based math floor.
BUG=chromium:270268
R=svenpanne@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22623007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16128 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 11:21:03 +00:00
bmeurer@chromium.org
aca068462f Fix -Wconversion warnings in external V8 headers.
BUG=v8:2830
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22549003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16127 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 10:33:08 +00:00
jochen@chromium.org
ebadc421e4 Move i18n collator code to runtime.
BUG=v8:2745
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22671002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16126 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 09:51:09 +00:00
mstarzinger@chromium.org
899e80130e Fix handle unsafety in Deoptimizer::MaterializeNextHeapObject.
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22327008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16125 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 09:49:15 +00:00
bmeurer@chromium.org
1086e02fef openbsd: fix platform build
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/21705003

Patch from Ben Noordhuis <info@bnoordhuis.nl>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16122 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 09:07:46 +00:00
mvstanton@chromium.org
ab066fae6f Add flag trap_on_stub_deopt. We want to be able to trap on hydrogen stub bailouts.
BUG=
R=svenpanne@chromium.org, verwaest@chromium.org

Review URL: https://codereview.chromium.org/22593003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16121 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 08:24:29 +00:00
svenpanne@chromium.org
f0305bd83e Instance-type-related fixes and cleanup.
Things got quite a bit out of sync (don't we all love copy-n-paste?):

   * TypeToString didn't handle SLICED_STRING_TYPE,
     SLICED_ASCII_STRING_TYPE, FIXED_DOUBLE_ARRAY_TYPE,
     JS_FUNCTION_PROXY_TYPE and JS_DATE_TYPE.

   * INSTANCE_TYPE_LIST was missing entries for
     SLICED_ASCII_STRING_TYPE, JS_SET_TYPE, and JS_MAP_TYPE.

To improve this maintenance nightmare a little bit, the missing
instance types were added to the INSTANCE_TYPE_LIST macro and this
list is now used via our beloved 2nd order macro technique in
TypeToString. As a side-effect, the strings returned by TypeToString
have a "_TYPE" suffix now, but this doesn't really matter and is a
small price to pay for consistency.

Removed INVALID_TYPE on the way, it had no real use.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22681004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16120 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 08:22:46 +00:00
yurys@chromium.org
85d8178d87 Deprecate self and total time getters and total sample count getter on CpuProfileNode
All of these values are derived from the self samples count and there is no need to evaluate them in v8 when clients can do that when needed on their side.

Also added unsigned GetHitCount() which should be used instead of double GetSelfSamplesCount(). I'm going to deprecate the latter one once Blink has switched to GetHitCount.

BUG=267595
TBR=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22710006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16119 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-09 07:38:26 +00:00
palfia@homejinni.com
e3f1ec4b1a MIPS: Patch to enhance the source code line information for profiler.
Port r16114 (2d1bf399)

Original commit message:
This patch is to enhance the source code line information for profiler.

For the Hydrogen compilation, most of the source code line information
is not copied from the HInstruction the to corresponding LInstruction.

This patch defines one PositionBits field for LInstruction and copies the
sorce code position value from the HInstruction.

When Generating the native code, we use RecordPosition(..) function to
write LInstruction's position value to position recorder.

BUG=

Review URL: https://codereview.chromium.org/22502006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16118 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 18:36:25 +00:00
yurys@chromium.org
4f56107e2f Revert "Deprecate self and total time getters and total sample count getter on CpuProfileNode"
This reverts commit r16116 due to WebKit compilation breakage. Will reland it once Blink r155755 is rolled into Chromium.

TBR=svenpanne@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/22388003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16117 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 14:43:44 +00:00
yurys@chromium.org
122327d1c6 Deprecate self and total time getters and total sample count getter on CpuProfileNode
All of these values are derived from the self samples count and there is no need to evaluate them in v8 when clients can do that when needed on their side.

Also added unsigned GetHitCount() which should be used instead of double GetSelfSamplesCount(). I'm going to deprecate the latter one once Blink has switched to GetHitCount.

BUG=267595
R=loislo@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22347003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16116 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 13:39:57 +00:00
plind44@gmail.com
ca09758499 MIPS: Optimize DoCheckMaps.
This simple change moves the Deoptimize call into the Deopt table
where it belongs, and shortens the inline map-check.

Gives a surprisingly large performance gain, around 4% on Octane.

BUG=
R=gergely@homejinni.com

Review URL: https://codereview.chromium.org/22391008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16115 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 05:04:42 +00:00
haitao.feng@intel.com
fb5857ceb8 Patch to enhance the source code line information for profiler.
This patch is to enhance the source code line information for profiler.

For the Hydrogen compilation, most of the source code line information
is not copied from the HInstruction the to corresponding LInstruction.

This patch defines one PositionBits field for LInstruction and copies the
sorce code position value from the HInstruction.

When Generating the native code, we use RecordPosition(..) function to
write LInstruction's position value to position recorder.

For the MIPS platform, I did not touch because I have no devices
to verify the modification on it.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/21042003

Patch from Chunyang Dai <chunyang.dai@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16114 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 02:16:12 +00:00
haitao.feng@intel.com
f54aa8b6ad tweak of Math.abs in its x64 stub
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22285003

Patch from Weiliang Lin <weiliang.lin@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16113 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-08 01:29:33 +00:00
plind44@gmail.com
bbbf78e78c MIPS: Refactor and improve inlined double-aligned allocations
Additionally, fix object size register clobbering in Allocate routine.

Based on 13284 (d42bf80).

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/22582002

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16112 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 21:54:03 +00:00
verwaest@chromium.org
e5afd32129 Fix Object.freeze, Object.observe wrt CountOperation and CompoundAssignment.
BUG=2774,2779
R=adamk@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22562004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16111 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 18:45:41 +00:00
verwaest@chromium.org
56ca64e98c Revert "mod does not rely on a fixed register anymore, the arguments are passed via the stack."
Due to buildbot failures.

BUG=
R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22474007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16110 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 17:22:17 +00:00
yurys@chromium.org
707fdd4c6a Support idle time in CPU profiler
This change provides an API for the embedder to tell CPU profiler if it is idle or busy with some task. This way we can discriminate between idle time and some native code execution.

BUG=268947
R=alph@chromium.org, yangguo@chromium.org

Review URL: https://codereview.chromium.org/22412003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16109 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 17:04:27 +00:00
olivf@chromium.org
ca4457e8ed mod does not rely on a fixed register anymore, the arguments are passed via the stack.
BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/21495005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16108 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 16:47:24 +00:00
palfia@homejinni.com
5021e72675 MIPS: First implementation of allocation elimination in Hydrogen.
Port r16098 (328ff5f7)

Original commit message:
This change implements a simple data-flow analysis pass over captured
objects to the existing escape analysis. It tracks the state of values
in the Hydrogen graph through CapturedObject marker instructions that
are used to construct an appropriate translation for the deoptimizer to
be able to materialize these objects again.

This can be considered a combination of scalar replacement of loads and
stores on captured objects and sinking of unused allocations.

TEST=mjsunit/compiler/escape-analysis

BUG=

Review URL: https://codereview.chromium.org/22580003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16107 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 16:34:20 +00:00
jkummerow@chromium.org
371ac893f9 Check for empty handle in JSON stringifier
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22420004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16106 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 16:34:13 +00:00
svenpanne@chromium.org
f3a36593be Removed wrong ASSERTs on MIPS.
R=palfia@homejinni.com

Review URL: https://codereview.chromium.org/22571003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16102 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 13:42:46 +00:00
svenpanne@chromium.org
a8e12c5a02 Handle Symbols in the LBranch instruction on x64.
This was an omission in https://code.google.com/p/v8/source/detail?r=14051.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/22578002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16101 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 13:33:41 +00:00
prybin@chromium.org
29bb553b1d Fix step in positions (include various calls and exclude current pc point), add a test
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22198002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16100 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 12:27:22 +00:00
jochen@chromium.org
5a92a95676 Move i18n's number-format C++ code to runtime
BUG=v8:2745
R=dcarney@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22266009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16099 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 12:14:50 +00:00
mstarzinger@chromium.org
63defee477 First implementation of allocation elimination in Hydrogen.
This change implements a simple data-flow analysis pass over captured
objects to the existing escape analysis. It tracks the state of values
in the Hydrogen graph through CapturedObject marker instructions that
are used to construct an appropriate translation for the deoptimizer to
be able to materialize these objects again.

This can be considered a combination of scalar replacement of loads and
stores on captured objects and sinking of unused allocations.

R=titzer@chromium.org
TEST=mjsunit/compiler/escape-analysis

Review URL: https://codereview.chromium.org/21055011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16098 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 11:24:14 +00:00
yangguo@chromium.org
ef8d394f12 Re-reland "Flush parallel recompilation queues on context dispose notification"
BUG=
R=hpayer@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22379002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16095 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 09:33:09 +00:00
yangguo@chromium.org
5818d831c5 Make JSON::Parse return Local<Value>
It should be able to return Smi, etc. Not only JSObject.

BUG=v8:2821
TEST=cctest/test-api/JSONParseNumber
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22416003

Patch from Takeshi Yoshino <tyoshino@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16092 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 09:11:39 +00:00
dslomov@chromium.org
5ad4a8458a Prepare push to trunk. Now working on version 3.20.15.
R=danno@chromium.org

Review URL: https://codereview.chromium.org/22468004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16091 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 09:05:59 +00:00
mstarzinger@chromium.org
fde19b3f39 Add frame pointer markup to grokdump.
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22556002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16090 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 08:57:25 +00:00
dcarney@chromium.org
cd74a09886 expose eternal handle api
R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22384003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16089 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 08:26:23 +00:00
jkummerow@chromium.org
a9efd655a5 Pacify Win7 builders
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22554002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16088 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 08:25:44 +00:00
bmeurer@chromium.org
db1436ab65 Handle external reference values properly in HConstant::ImmortalImmovable().
R=danno@chromium.org

Review URL: https://codereview.chromium.org/22274017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16087 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 07:08:50 +00:00
jochen@chromium.org
02914097b2 Move i18n extension's date-format C++ code to runtime
BUG=v8:2745
R=dcarney@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22411003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16086 ce2b1a6d-e550-0410-aec6-3dcde31c8c00
2013-08-07 03:40:44 +00:00