Port r14658 (12c49bd)

This makes sure that Crankshaft doesn't disable optimization to early on hot functions that still contain unexecuted code without type information.

BUG=

Review URL: https://codereview.chromium.org/15005008
Patch from Balazs Kilvady <kilvadyb@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14672 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=jkummerow@chromium.org
BUG=v8:2628

Review URL: https://codereview.chromium.org/15133007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14671 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Stack iterator takes return address based on the frame pointer (ebp) and detects JS frames based on value at fp + StandardFrameConstants::kMarkerOffset. So in order the iterator to work correctly this values should be already setup for the current function. Stack frame is constructed at the very beginning of JS function code and destroyed before return. If sample is taken before before the frame construction is completed or after it was destroyed the stack iterator will wrongly think that FP points at the current functions frame base and will skip callers frame. To avoid this we mark code ranges where  stack frame doesn't exist and completely ignore such samples.

This fixes cctest/test-cpu-profiler/CollectCpuProfile flakiness.

BUG=v8:2628
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14253015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14670 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Ideally this would have been implemented via desugaring at parse-time,
but yield* is an expression, and its desugaring includes statements like
while and try/catch.  We'd have to have BlockExpression in the AST to
support that, and it's not worth it for this feature.

So instead we implement all of the logic in
FullCodeGenerator::VisitYield.  Delegating yield AST nodes now have a
try handler index, for the try/catch.  Otherwise the implementation is
straightforward.

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

Review URL: https://codereview.chromium.org/14582007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14669 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14668 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14667 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Ideally this would have been implemented via desugaring at parse-time,
but yield* is an expression, and its desugaring includes statements like
while and try/catch.  We'd have to have BlockExpression in the AST to
support that, and it's not worth it for this feature.

So instead we implement all of the logic in
FullCodeGenerator::VisitYield.  Delegating yield AST nodes now have a
try handler index, for the try/catch.  Otherwise the implementation is
straightforward.

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14666 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Error found in test262 on ARM: BinaryOpStub could call out to a built-in and push parameters without an enclosing frame. This corrupted stackwalking.

BUG=
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/14813029

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14665 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/15162002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14664 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Improve dead code elimination by transitively marking live code and removing all dead code. Replace unreachable phi removal algorithm with the new dead code elimination pass, which is more thorough.

Review URL: https://codereview.chromium.org/14676011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14661 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/14971005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14660 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

They are a non-standard extension (probably in C1X, IIRC), but clang is unhappy
with them when -Wnested-anon-types is enabled, which seems to be implied by
-pedantic.

With this change and the previous fix for clang, we are now -Werror clean, even
on clang 3.3.

R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/14850016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14659 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This makes sure that Crankshaft doesn't disable optimization to early on hot functions that still contain unexecuted code without type information.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14738009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14658 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

With static linkage, clang 3.3 is unhappy about unused functions

   template <> inline bool Is<JSFunction>(Object* obj)
   template <> inline bool Is<JSArray>(Object* obj)

in any compilation unit apart from acessors.cc.

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/14836014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14657 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://chromiumcodereview.appspot.com/15159002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14654 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BinaryOpStub::GenerateSmiStub() on 32bit would erroneously patch the IC in case of a gc requirement. Brought the behavior into line with ARM and x64. Also some cleanup to label names.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/15102006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14653 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14650 (d045fbd3)

BUG=

Review URL: https://codereview.chromium.org/15035007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14652 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14630 (124d01e7)

Original commit message:
Improves NavierStokes by about 5%

BUG=

Review URL: https://codereview.chromium.org/14762012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14651 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/14604007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14650 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org
BUG=v8:2681
TEST=mjsunit/regress/regress-2681

Review URL: https://codereview.chromium.org/14731023

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14649 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/14927014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14648 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/15059009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14646 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=svenpanne@chomium.org
BUG=

Review URL: https://codereview.chromium.org/15129002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14645 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Previously, HDiv never had its CanOverflow flag cleared and HMod had
inverted logic (compared to HAdd, HSub and HMul). Minor cleanups on the way.

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14617015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14644 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14978005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14639 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=chromium:233737

Review URL: https://codereview.chromium.org/14978004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14638 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

To comply with c1visualizer syntax, the format for ranges and types has changed
slightly. Furthermore, the format for phi uses has been changed a bit for more
consistency, too.

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/14882006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14637 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/14978003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14636 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This makes sure that shared function infos that break the one-to-one
mapping to code are marked as un-flushable. Otherwise enqueuing through
the GC meta-data field in the code object doesn't work.

R=rossberg@chromium.org
TEST=cctest/test-api/Threading4

Review URL: https://codereview.chromium.org/14710015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14635 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Improves NavierStokes by about 5%

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/15014020

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14630 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

We have to check for overflow before attempting to do a modulo operation,
otherwise Crankshaft itself segfaults on some platforms, e.g. ia32. Added tests
even for division, where the problem doesn't show up, just to be sure...

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/14617014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14629 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14625 (bfb25ed8)

BUG=

Review URL: https://codereview.chromium.org/15102002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14628 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r14608 (339e0877)

BUG=

Review URL: https://codereview.chromium.org/14980008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14627 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/14707011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14626 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=danno@chromium.org

Review URL: https://chromiumcodereview.appspot.com/14996004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14625 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This will be relevant after Persistent is changed to Dispose itself when
destructed. With Persistent::ClearAndLeak, Blink can take the ownership of the
object pointed by a Persistent and avoid it getting destructed.

BUG=
R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/15023010

Patch from Marja Hölttä <marja@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14624 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
TEST=test262/15.5.4.9_3

Review URL: https://codereview.chromium.org/14972015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14623 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://chromiumcodereview.appspot.com/14838012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14622 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Casting NaN to int is unpredictable, on different architectures it produces different int value.

TEST=test262/S15.4.4.10_A2.1_T2, test262/S15.4.4.10_A2.2_T2

BUG=

Review URL: https://codereview.chromium.org/14812014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14621 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
TEST=test262

Review URL: https://codereview.chromium.org/14644015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@14620 ce2b1a6d-e550-0410-aec6-3dcde31c8c00