R=machenbach@chromium.org
BUG=

Review URL: https://codereview.chromium.org/23125027

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16328 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=machenbach@chromium.org

Review URL: https://codereview.chromium.org/22887053

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16327 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

to work around erroneous "illegal access" error on x64.

BUG=v8:2790
R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/22877039

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16324 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

regardless of the map used in polymorphic stores.

BUG=
R=jkummerow@chromium.org, verwaest@chromium.org

Review URL: https://codereview.chromium.org/21058003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16323 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Added support for truncating DoubleToIStub and reorganize the macro-assembler
dToI operations to do the fast-path inline and the slow path by calling the
stub.

BUG=
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/23129003

Patch from Ross McIlroy <rmcilroy@google.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16322 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22903012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16321 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

A while ago in r2315 Mac OS X cpu profiler implementation was changed to pause sampled thread instead of sending SIGPROF signal. That was done because at that point profiler send the signal to the whole process and it was handled on a random thread. Now that signal-based implementation uses pthread_kill it may well be used on Mac OS X too.

BUG=v8:2814
R=bmeurer@chromium.org, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/23115005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16320 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

By setting this flag assertions behind --debug-code will trigger a
breakpoint instead of a call into Abort. This eases debugging, as the
call site is less cluttered and the backtrace starts where it should.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/23352004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16319 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This renames the existing V8_ALIGNAS() to V8_ALIGNED(), and introduces
V8_ALIGNAS(type, alignment) which aligns according to the type and falls
back to aligning according to alignment.

Also use __attribute__((aligned(n))) instead of __attribute__((__aligned__(n))).

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22999052

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16318 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22831050

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16317 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org, svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/22795004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16316 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Move all of the CPU detection logic to the CPU class, and make
all other code use the CPU class for feature detection.

This also fixes the ARM CPU feature detection logic, which was
based on fragile string search in /proc/cpuinfo. Now we use
ELF hwcaps if available, falling back to sane(!!) parsing of
/proc/cpuinfo for CPU features.

The ia32 and x64 code was also cleaned up to make it usable
outside the assembler.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/23401002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16315 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/23202017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This reverts commit r16311 for breaking build with certain compilers.
Will reland once sorted out.

TBR=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22979004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16313 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This method is being replaced by GetHitCount.

BUG=267595
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22295009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16312 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Also fix the inconsistent attribute naming, and make V8_GNUC_PREREQ() test work for GCC compatible compilers.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22888008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16311 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

New flag is added that allows to specify CPU profiler sampling rate in microseconds as command line argument. It was tested to work fine with 100us interval(currently it is 1ms). Default values are kept the same as in the current implementation. The new implementation is enabled only on POSIX platforms which use signals to collect samples. Other platforms that pause thread being sampled are to follow.

SIGPROF signals are now sent on the profiler event processor thread to make sure that the processing thread does fall far behind the sampling.

The patch is based on the previous one that was rolled out in r13851. The main difference is that the circular queue is not modified for now.

On Linux sampling for CPU profiler is initiated on the profiler event processor thread, other platforms to follow.

CPU profiler continues to use SamplingCircularQueue, we will probably replace it with a single sample buffer when Mac and Win ports support profiling on the event processing thread.

When --prof option is specified profiling is initiated either on the profiler event processor thread if CPU profiler is on or on the SignalSender thread as it used to be if no CPU profiles are being collected.

ProfilerEventsProcessor::ProcessEventsAndDoSample now waits in a tight loop, processing collected samples until sampling interval expires. To save CPU resources I'm planning to change that to use nanosleep as only one sample is expected in the queue at any point.

BUG=v8:2814
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/21101002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16310 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22867047

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16309 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/23402005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16308 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Added a console parameter for source map to the tick processor.
The tickprocesspor reads in the source maps and uses it to output the original filename, line number and column in the profile.
Modified d8 to output column numbers into the log, since this is needed to do source mapping.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22897021

Patch from Daniel Kurka <dankurka@google.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16307 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22926027

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16306 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/23102015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16303 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The flag restricts hydrogen.cfg output to functions passing the filter,
similar to what --hydrogen-filter does for optimization in general.

This is useful for investigating large repro cases where tracing all
functions would lead to an impractically large hydrogen.cfg file, but
restricting optimization using --hydrogen-filter is undesirable
(e.g. because it might cause the issue to no longer reproduce).

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/22926025

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16302 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org
BUG=

Review URL: https://codereview.chromium.org/23232008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16299 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:2745
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/23304005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16298 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22999048

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16297 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Note that the test crashes if it fails, so the PASS || FAIL annotation
doesn't work.

BUG=v8:2857
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/22903039

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16296 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org

Review URL: https://chromiumcodereview.appspot.com/22861025

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16295 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/23090004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16294 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

We already have OS::DebugBreak(), which does the right thing.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22893050

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16293 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

V8 stores this information directly in the map of the wrapper, however,
it is not invalidated when the prototype of the wrapper is changed, so
once the bit is set, it is no longer possible to override valueOf.

This bug is currently hidden in Chrome since the i18n extension always
modifies the String.prototype, and so the optimization never kicks in.
Disabling the optimization temporarily allows for snapshotting i18n now.

BUG=v8:2855
R=yangguo@chromium.org
TEST=mjsunit/regress/regress-2855.js

Review URL: https://codereview.chromium.org/23060030

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16292 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=machenbach@chromium.org
BUG=cctest

Review URL: https://codereview.chromium.org/22861037

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16291 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/22914034

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16290 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

In the RegExpUnparser::VisitText(RegExpText* that, void* data) function always RegExpUnparser::VisitAtom function called via that->elements()->at(i).data.u_atom->Accept(this, data); even if the type of the object is RegExpCharacterClass.

The problem shows using g++ 4.7(.2, .3) since r16232, since GCC optimizes virtual method calls to direct calls based on __final/final hints. Tested on MIPS and x64:
Program received signal SIGSEGV, Segmentation fault.
0x0000000000588928 in v8::internal::RegExpUnparser::VisitAtom(v8::internal::RegExpAtom*, void*) ()

This cleans up the TextElement class to avoid the unsafe+unchecked union access, that caused the crash.

TEST=cctest/test-regexp/ParserRegression
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/22815033

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16289 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=bmeurer@chromium.org
TEST=cctest/test-mark-compact/MarkCompactCollector

Review URL: https://codereview.chromium.org/23112026

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16288 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

All the tests that started crashing create ProfilerEventsProcessor on the stack. After r16284 SamplingCircularQueue buffer is allocated as a field of the queue instead of separate heap object. This increased self size of ProfilerEventsProcessor by about 1Mb. Windows malloc fails to allocate such an object on the stack and crashes.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/23093022

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16287 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org
BUG=v8:2594

Review URL: https://codereview.chromium.org/22901010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16286 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=None
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/23361023

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16285 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The new implementation:
* uses MemoryBarriers to make sure up-to-date data is accessed on both producer and consumer threads
* will not allow to overwrite records
* doesn't have notion of chunks, instead each entry is aligned on the cache line boundaries

BUG=v8:2814
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22849002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16284 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/22852021

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@16283 ce2b1a6d-e550-0410-aec6-3dcde31c8c00