review.tizen.org Git - platform/upstream/v8.git/log

Tune mjsunit/array-functions-prototype-misc

Again, this brings testing times down quite a bit without losing test
coverage.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/44143003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17403 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handlify JSObject::PrepareElementsForSort

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/32523008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17402 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix bug in nextWeaker function of deltablue benchmark.

It does not seem to affect the correctness of the particular
workload it is currently running.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/13994003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17399 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Simplify test-mark-compact/NoPromotion test.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/42543008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17398 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The Elements pointer in a JSObject can have a filler map instead of a
valid fixed array, iff a gc occurred while allocating a fixed array as
part of array construction. Heap verification needs protection against
examining the elements object in this case.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/43383004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17397 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Reland "Make Array.prototype.pop throw if the last element is not configurable."

This relands r17346.

R=machenbach@chromium.org
BUG=311164

Review URL: https://codereview.chromium.org/43923002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17394 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix release build failures due to --enable-slow-asserts

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/43943002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17393 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Define DEBUG for v8_optimized_debug=2

Thereby ensuring there is only a minimal performance regression vs. NDEBUG (now it's only about 10% slower rather than ~2x).

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

Review URL: https://codereview.chromium.org/39183004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17392 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make top and limit field in AllocationInfo private, assert on non-aligned setting of these fields, and eliminate indirect access over top address on top pointer.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/40083002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17391 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.23.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/43663002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17388 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Tune mjsunit/regexp-global.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/42993004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17387 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Tune cctest/test-parsing/ParserSync.

This test suffers heavily from combinatorial explosion. To redemedy
this situation a bit, two independent parsing flags are tested
separately, reducing the runtime by a factor of 4. Furthermore, the
code has been cleaned up a bit.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/40113002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17385 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The "TESTFLAGS" is now added to Android .check build to be consistent with other platform build.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/39273004

Patch from Bangfu Tao <bangfu.tao@samsung.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17384 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Introduce MoveDouble to the X64 MacroAssembler

R=danno@chromium.org

Review URL: https://codereview.chromium.org/26216008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17383 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Enable double alignment for constructed arrays.

2nd (of 3) CL's to ensure complete alignment of FixedDoubleArrays.

TEST=
BUG=
R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/35133002

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17382 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make the message test suite run without slow asserts.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/39303004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17381 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove specialized raw Cell and Map allocators.

R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/30023003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17380 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Temporarily disable mjsunit/regress/regress-2612 to make our tree green again.

TBR=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/40203002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17379 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Tune mjsunit/regress/regress-2612.

Lower the bounds to something bearable which would still timeout if we
used a quadratic algorithm.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/39863003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17377 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handle constants in new space on arm by making macro-assembler smarter.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/30023005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17376 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add performance.now() to the d8 shell.

TEST=mjsunit/d8-performance-now
R=hpayer@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=17366

Committed: https://code.google.com/p/v8/source/detail?r=17368

Review URL: https://codereview.chromium.org/32433010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17375 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Fix shared library build after r17368." and "Add performance.now() to the d8 shell.".

This reverts commit r17372 and r17368 for breaking the shared
library build.

TBR=hpayer@chromium.org

Review URL: https://codereview.chromium.org/40043002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17374 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Get rid of unused STRING_ADD_ERECT_FRAME flag.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/39803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17373 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix shared library build after r17368.

TBR=hpayer@chromium.org

Review URL: https://codereview.chromium.org/40033002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17372 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add code age subtype tracking to --track-gc-object-stats

Adds counters which track the age of code in the heap during a gc if
--track-gc-object-stats is enabled.

- Splits RecordObjectStats into RecordObjectStats, RecordCodeSubTypeStats and
RecordFixedArraySubTypeStats.
- Renames kNoAge to kNoAgeCodeAge to follow other code age enums and enable
the name to be used in Macro based initialization of the counters.

BUG=None
R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/26179004

Patch from Ross McIlroy <rmcilroy@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17369 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add performance.now() to the d8 shell.

TEST=mjsunit/d8-performance-now
R=hpayer@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=17366

Review URL: https://codereview.chromium.org/32433010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17368 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Add window.performance.now() to the d8 shell."

This reverts commit r17366 for breaking the mozilla tests.

TBR=hpayer@chromium.org

Review URL: https://codereview.chromium.org/38753006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17367 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add window.performance.now() to the d8 shell.

TEST=mjsunit/d8-performance-now
R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/32433010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17366 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Record allocation stack traces

This is initial implementation of allocation profiler.

Whenever new object allocation is reported to the HeapProfiler and allocation tracking is on we will capture current stack trace, add it to the collection of the allocation traces (a tree) and attribute the allocated size to the top JS function on the stack.

Format of serialized heap snapshot is extended to include information about recorded allocation stack traces.

This patch is r17301 plus a fix for the test crash in debug mode. The test crashed because we were traversing stack trace when just allocated object wasn't completely configured, in particular the map pointer was incorrect. Invalid Map pointer broke heap iteration required to find Code object for a given pc during stack traversal. The solution is to insert free space filler in the newly allocated block just before collecting stack trace.

BUG=chromium:277984,v8:2949
R=hpayer@chromium.org, loislo@chromium.org

Review URL: https://codereview.chromium.org/34733004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17365 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.22.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/39393005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17362 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Removed long-running obselete test case.

The test was the 2nd longest-running test case in debug mode, and the
stuff it tests has already been moved long ago to some other place,
which is in turn heavily tested by far simpler and faster things
(%TruncateString etc.).

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/39233003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17361 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Make Array.prototype.pop throw if the last element is not configurable."

This reverts commit r17346.

R=bmeurer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/39593002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17360 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS: Increase code space first page size due to r17334.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/39363004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17359 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix r17253.

There's no reason why the setObjectGroupIde / setReferenceFromGroups funcs should've been called twice.

BUG=
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/39573002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17358 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use PopReturnAddressTo and PushReturnAddressFrom in Generate_MarkCodeAsExecutedOnce builtin function for X64

R=danno@chromium.org

Review URL: https://codereview.chromium.org/38613003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17357 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix ARM build on mac.

Remove a couple unused vars left behind after r16676. Clang on mac
is pickier than gcc on linux, specifically gcc does not complain about
unused class member variables.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/39193003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17356 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.21.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/39483002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17353 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Introduce xchgl, orl, roll, subl, testl and xorl into X64 assembler

R=danno@chromium.org

Review URL: https://codereview.chromium.org/26780004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17352 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use the correct version of movq for ExternalReference in X64

R=danno@chromium.org

Review URL: https://codereview.chromium.org/26410004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17351 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handlify Map::CopyInstallDescriptors

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/34603008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17350 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handlify Map::CopyNormalized

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/32483006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17349 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Removed redundant definition of new space top in serializer.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/30403003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17348 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handle new space constants on ia32 by using in a register in Lithium.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/35413008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17347 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make Array.prototype.pop throw if the last element is not configurable.
Popping an element from an array should call [[Delete]] internal method
and pass true as the second argument (ECMA-262/5.1/#sec-15.4.4.6).
When the last element can't be deleted, throw a Type Error.
Not throwing the error would result in endless loop in the following test.

TEST=var a=[];Object.defineProperty(a,0,{});while(a.length)a.pop();

By the way fix another bug, or else i can't post any issues.
"presubmit.py" throw a "missing a correct copyright header" on windows.
Both the slash and the backslash are valid path separator on windows.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/29513004

Patch from Yanagi <admin@web-tinker.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17346 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix HObjectAccess for loads from migrating prototypes

BUG=chromium:305309
R=danno@chromium.org

Review URL: https://codereview.chromium.org/35173005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17345 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix MEMORY_SANITIZER macro redefinition error.

MEMORY_SANITIZER is defined by default in certain compiler environments (when
MemorySanitizer is enabled, of course). Do not redefine it in that case.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/34503012

Patch from Evgeniy Stepanov <eugenis@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17344 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Enable preaging of code objects when --optimize-for-size.

This change means that code which is never executed is garbage collected immediately, and code which is only executed once is collected more quickly (limiting heap growth), however, code which is re-executed is reset to the young age, thus being kept around for the same number of GC generations as currently.

BUG=280984
R=danno@chromium.org, hpayer@chromium.org

Review URL: https://codereview.chromium.org/23480031

Patch from Ross McIlroy <rmcilroy@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17343 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

HCheckInstanceType factories unified

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/35413007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17342 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Use common allocation choke point in Heap::AllocateConstantPoolArray.

R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/37003003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17341 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Expose v8::Function::GetDisplayName to public API.

BUG=chromium:17356
R=mstarzinger@chromium.org, yurys@chromium.org, yangguo@chromium.org, yurys

Committed: https://code.google.com/p/v8/source/detail?r=17324

Review URL: https://codereview.chromium.org/26709011

Patch from Andrey Adaikin <aandrey@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17340 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

PushAndAdd() usages refactored.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/36893002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17339 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Disable allocation folding when allocations tracking is on"

This reverts r17337. The new test fails on V8 linux - shared: http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20shared/builds/4726/steps/Check/logs/stdio

TBR=mvstanton@chromium.org
BUG=None

Review URL: https://codereview.chromium.org/37073003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17338 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Disable allocation folding when allocations tracking is on

BUG=chromium:277984
R=hpayer@chromium.org, loislo@chromium.org, mvstanton@chromium.org, yurys@chromium.org

Review URL: https://codereview.chromium.org/32583002

Patch from Alexandra Mikhaylova <amikhaylova@google.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17337 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Simplify ObjectVisitor for external references.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/36853003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17336 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add a soft-deopt in keyed element access when current IC is pre-monomorphic and no type feedback was collected.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/32643004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17335 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Crankshaft builtins.

Enable optimizing compiler for V8 built-ins. Also fixes an issue uncovered in
x64 codegen.

R=danno@chromium.org, mstarzinger@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/34503003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17334 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make macro-assembler-x64 deal with handles to objects in new space by default.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/34753004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17333 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

force accessor pair accesses backed by simple api calls to be generic accesses when crankshafted

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/34773003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17332 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Partially annotate V8 for MemorySanitizer.

This is a part of https://codereview.chromium.org/26006004/.
These changes are enough to run most apps that embed V8 (by additionally
ignoring all memory accesses from V8), but don't help testing V8 itself in any
way.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/27471003

Patch from Evgeniy Stepanov <eugenis@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17331 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.20.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/32693004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17328 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handlify Map::CopyForObserved

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/34023002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17327 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Condense flag definitions a bit.

This makes it easy to spot the default value of a flag when grepping the
code as the default value appears on the same line as the name now.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/35213002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17326 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Expose v8::Function::GetDisplayName to public API."

This reverts commit 54c7b9af65dd349405944bd9ae2f064202dc6e1d.

The new test fails in debug mode.

BUG=chromium:17356
TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/30443006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17325 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Expose v8::Function::GetDisplayName to public API.

BUG=chromium:17356
R=mstarzinger@chromium.org, yurys@chromium.org, yangguo@chromium.org, yurys

Review URL: https://codereview.chromium.org/26709011

Patch from Andrey Adaikin <aandrey@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17324 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Expose v8::Function::IsBuiltin to public API.

This will be used by DevTools so that we could generate a better preview in console. Namely, we could assume that a preview of an object is lossless if all its member functions are builtin.

We also may want to expose this to DevTools users via remote debugging protocol in Debugger.FunctionDetails struct.

BUG=chromium:261470
R=mstarzinger@chromium.org, yurys@chromium.org, dcarney@chromium.org, yurys

Review URL: https://codereview.chromium.org/27701002

Patch from Andrey Adaikin <aandrey@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17323 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Enable escape analysis for Hydrogen.

R=titzer@chromium.org

Review URL: https://codereview.chromium.org/34833004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17322 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix materialization of captured objects with field tracking.

R=titzer@chromium.org
BUG=chromium:298990
TEST=mjsunit/compiler/escape-analysis-representation

Review URL: https://codereview.chromium.org/35133003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17321 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Temporarily deactive regress-2185-2 until our array handling is in good shape again.

BUG=v8:2950
R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/30443005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17320 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Guard against misuse of IfBuilder.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/30693007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17319 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

More Hydrogen templatization.

Factories added to a set of Hydrogen classes.
Several old-style HXxx::New() replaced with New<HXxx>() and NewUncasted<HXxx>.
Several AddInstruction() calls following New::XXX() replaced with Add<XXX>().
AddLoadNamedField() method added to GraphBuilder.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/34523004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17318 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Removed obsolete unit tests.

As discussed offline, these tests don't test what they were supposed to
test anymore and were the longest running ones.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/32433009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17317 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Re-enable global load elimination after fixing misaligned load/stores."

This reverts commit r17315 for breaking the webkit tests.

TBR=titzer@chromium.org

Review URL: https://codereview.chromium.org/34853002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17316 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Re-enable global load elimination after fixing misaligned load/stores.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/34653002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17315 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Do not run useless test variants for test-parsing/ParserSync.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/34733002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Do not serialize stack limits.

This is part of making snapshots reproducable, more to come...

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/33193004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17313 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make PropertyCell::UpdatedType return a handle.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/31933003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17312 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.19.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/32433006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17309 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add test for push-to-trunk script.

This is a prototype attempt for testing the push-to-trunk script. There are several todos left, but it is not clear if it is worth investing time into those. The test is meant to aid in verifying modifications in the push-to-trunk script.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/32283002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17308 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

[Sheriff] Revert "Record allocation stack traces" and "Temporarily disable test-heap-profiler/TrackHeapAllocations".

This reverts commit r17301 and r17303 for breaking the nosnap builds.

TBR=yurys@chromium.org

Review URL: https://codereview.chromium.org/34523002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17307 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS: Fix kraken 1.1 benchmark failure.

This makes sure that the DoubleToIStub::Generate method
does not overwrite the value of the f12 register.

TEST=bencnhmarks/kraken/stanford-crypto-pbkdf2
BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/33463002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17306 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS: Add tool to visualize machine code/lithium.

Port r17295 (b36eb655)

Original commit message:
In the process:
- Add a command-line flag --opt-code-positions to track source position information throughout optimized code.
- Add a subclass of the hydrogen graph builder to ensure that the source position is properly set on the graph builder for all generated hydrogen code.
- Overhaul handling of source positions in hydrogen to ensure they are passed through to generated code consistently and in most cases transparently.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/33803002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17305 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

MIPS: Inline number to string conversion for string addition into BinaryOp(Stub).

Port r17290 (b5fabb92)

Original commit message:
This fixes a performance regression that was caused by converting the
BinaryOpStub to a Hydrogen code stub. It also fixes a leftover TODO wrt.
the handling of Number*String or String*Number versions of the stub.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/32473007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17304 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Temporarily disable test-heap-profiler/TrackHeapAllocations

The test started failing on V8 Linux - nosnap - debug after r17301 which added stack traversal to the allocation tracker. I'm disabling it for now until the problem is fixed.

BUG=v8:2949
TBR=hpayer@chromium.org

Review URL: https://codereview.chromium.org/30633005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17303 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

HEnvironmentMarker factory added

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/32513002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17300 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Allow constants to be deleted by reinserting them into the graph as needed.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/31943002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17299 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handlify PropertyCell::SetValueInferType and friends.

This finally gets rid of the pesky trampoline in SetValueInferType and
enforces the layering between PropertyCell and Heap. It requires full
handlification of NewGlobalObject as well, which is only used when the
snapshot is created at compile-time.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/28783002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17297 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add ishell@ to OWNERS.

TBR=danno@chromium.org

Review URL: https://codereview.chromium.org/32523002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17296 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add tool to visualize machine code/lithium.

In the process:
- Add a command-line flag --opt-code-positions to track source position information throughout optimized code.
- Add a subclass of the hydrogen graph builder to ensure that the source position is properly set on the graph builder for all generated hydrogen code.
- Overhaul handling of source positions in hydrogen to ensure they are passed through to generated code consistently and in most cases transparently.

Originally reviewed in this CL: https://codereview.chromium.org/24957003/

Review URL: https://codereview.chromium.org/29123008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17295 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Handle misaligned loads and stores in load elimination. Do not track misaligned loads and be conservative about invalidating misaligned stores. Add more tests for number conversion to string (NumberToStringStub exhibits misaligned loads)

BUG=v8:2934
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/28383003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17294 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.18.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/32243003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17291 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Inline number to string conversion for string addition into BinaryOp(Stub).

This fixes a performance regression that was caused by converting the
BinaryOpStub to a Hydrogen code stub. It also fixes a leftover TODO wrt.
the handling of Number*String or String*Number versions of the stub.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/27674002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17290 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Add a soft-deopt in keyed element access when current IC is pre-monomorphic and no type feedback was collected."

This reverts commit r17288 for breaking the mjsunit/unbox-double-arrays
test on almost every platform.

TBR=hpayer@chromium.org

Review URL: https://codereview.chromium.org/32373002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17289 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add a soft-deopt in keyed element access when current IC is pre-monomorphic and no type feedback was collected.

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/32263002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17288 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Harmony: implement Math.trunc.

BUG=v8:2938
R=dslomov@chromium.org

Review URL: https://codereview.chromium.org/28793002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17286 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Enable load elimination."

This reverts commit r17281.

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/32113002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17284 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Enable load elimination.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/31923003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17281 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Harmony: implement Math.sign.

R=dslomov@chromium.org
BUG=v8:2938

Review URL: https://codereview.chromium.org/28723002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17279 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Prepare push to trunk. Now working on version 3.22.17.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/31913002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17276 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Make stack_parameter_count a plain register.

R=rossberg@chromium.org, danno@chromium.org

Review URL: https://codereview.chromium.org/28993003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@17275 ce2b1a6d-e550-0410-aec6-3dcde31c8c00