R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/16870011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15248 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17498002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15247 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/17082003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15246 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17406009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15245 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17028017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15244 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=hpayer@google.com

Review URL: https://chromiumcodereview.appspot.com/17071002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15241 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This patch makes yield sites save the resume continuation and context
inline.  If the operand stack is empty, we can avoid a call into the
runtime.  This also makes the SuspendJSGeneratorObject runtime function
less magical: it just has to save the operand stack and stack handlers.

This speeds up the following case by a factor of 3 or so:

  function* until(n) {
    for (var i = 0; i < n; i++)
      yield i;
  }

  function sum(iter) {
    var sum = 0;
    for (var x of iter) sum += x;
    return sum;
  }

  for (var i = 0; i < 10000; i++) sum(until(1000))

Also, there is no more sentinel value as the generators will resume in
the right place already, allowing me to remove the hack added to the
--debug-code check in r14437.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/15990004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15240 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17069003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15239 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/17451014

Patch from Haitao Feng <haitao.feng@intel.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15238 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17336003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15237 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17448017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15236 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Lithium codegen should not pass around block_ids. Rather encapsulate the basic block to assembly label mapping in the LInstruction.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17276002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15235 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

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

Review URL: https://codereview.chromium.org/17303006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15234 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/17218005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15233 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/16831020

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15232 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Parallel recompilation is usually disabled on single-core systems.

R=jkummerow@chromium.org
BUG=v8:2733

Review URL: https://codereview.chromium.org/17261021

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15231 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The formal parameter count was always being treated as an untagged
integer, but it is actually a Smi on ia32 and arm.

R=mstarzinger@chromium.org
BUG=v8:2355
TEST=mjsunit/harmony/generators-iteration

Review URL: https://codereview.chromium.org/17485002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15230 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This is a temporary revert to track down a potential perf regression
introduced in r14930. The following two changes were reverted:
- "Deprecate HAllocateObject in favor of HAllocate."
- "Added pretenuring support for call new."

R=danno@chromium.org
BUG=chromium:247504

Review URL: https://codereview.chromium.org/17491002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15229 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17343004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15228 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Abort incremental marking in stress gc tests whenever we expect a full collection after executing CollectAllGarbage.

BUG=
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/17482003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15227 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/17328005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15226 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17366004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15225 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Also, fix bugs in Type::Union and Type::Maybe.

(This subsumes the in-flight fix for Union in https://codereview.chromium.org/16925008/, including test cases).

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17335003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15224 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=v8:2737

Review URL: https://codereview.chromium.org/17283007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15223 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17064004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15222 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/17071004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15221 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

A bunch of tests in test-api.cc need to be executed two times: with CPU profiler on and off to check different code paths. There are only two such tests at the moment but I'm going to add more and having this logic duplicated in all these tests look awkward. The tests are executed as part of Threading* tests and there is no much sense in running CPU profiler in that case. This change addresses both problems by introducing THREADED_PROFILED_TEST macro which mosly resembles THREADED_TEST except that when such test is executed as a standalone test (not part of a test-api/Threading*) it will first run the test normally and the start CPU profiler and run the test again.

BUG=244580
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17072002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15220 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This change modifies memory accesses to ensure proper load/store ordering.

BUG=249750
R=dvyukov@google.com, jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17294004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15219 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The operation is already implemented in atomicops.h No need to duplicate the code.

BUG=None
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17222004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15218 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=None
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/17176003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15217 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/16875008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15214 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:2628

TEST=cctest/test-cpu-profiler/SampleWhenFrameIsNotSetup
R=jkummerow@chromium.org, yurys@chromium.org

Review URL: https://codereview.chromium.org/17265004

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15213 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TEST=cctest/test-assembler-mips/MIPS3

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/17099010

Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15212 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r15201 (8c56d50)

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/17447004

Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15211 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=hpayer@chromium.org

Review URL: https://codereview.chromium.org/16896012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15210 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The old code aging mechanism is too agressive with flushing as it leads
to many functions being flushed and recompiled over and over again. By
now the new code aging mechanism has stabilized enough to deprecate the
old fallback mechanism.

R=danno@chromium.org

Review URL: https://codereview.chromium.org/17061004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15209 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/17157006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15208 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

If an array literal contains some non-constant elements, is of type SMI, and
then the boilerplate transitions to double or fast sometime after we've
crankshafted the code, then we could incorrectly store smis in double arrays.

BUG=
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/17334004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15207 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/17391016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15206 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Instead of allocating weak handles to free ArrayBuffer backing store, dispose of memory while walking the weak list of ArrayBuffers on GC.

Also, free all array buffers on isolate tear-down.

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/16950013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@15205 ce2b1a6d-e550-0410-aec6-3dcde31c8c00