R=rossberg@chromium.org, rossberg
BUG=v8:2409
LOG=Y

Review URL: https://codereview.chromium.org/183683022

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19734 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r19715 (0d6d244)

Original commit message:
Lithium currently supports 3 division-like operations on integral
operands: "Normal" division (rounding towards zero), flooring division
(rounding towards -Infinity) and modulus calculation (the counterpart
for the "normal" division). For divisors which are a power of 2, one can
efficiently use some bit fiddling to avoid the actual division for such
operations. This CL cleanly splits off these operations into separate
Lithium instructions, making the code much more maintainable and more
consistent across platforms.

There are 2 basic variations of these bit fiddling algorithms: One
involving branches and a seemingly more clever one without branches.
Choosing between the two is not as easy as it seems: Benchmarks (and
probably real-world) programs seem to favor positive dividends,
registers and shifting units are sometimes scarce resources, and branch
prediction is quite good in modern processors. Therefore only the
"normal" division by a power of 2 is implemented in a branch-free
manner, this seems to be the best approach in practice. If this turns
out to be wrong, we can easily and locally change this.

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/189433008

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19733 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r19704 (120500a)

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

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/189913005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19732 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=None
LOG=N
TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/191153002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19731 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/190783002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19730 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=None
TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/191043002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19729 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

AllocationTracker now maintains a map from address range to stack trace that allocated the range. When snapshot is generated the map is used to find construction stack trace for an object using its address.

BUG=chromium:277984
LOG=Y
R=alph@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/177983003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19728 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/171603020

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19727 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r19706 (81fd8dd)

Original commit message:
We can make more efficient code to check against type cells in the future if we use symbols, guaranteed not to conflict with user code. Currently, the "symbols" are the hole and undefined. Undefined may come in from the outside.

BUG=
R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/189143006

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19726 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This patch:
 - uses named registers
 - reserves all the stack space in one go to avoid operations on csp
 - uses LDP/STP to copy the arguments

R=jochen@chromium.org

Review URL: https://codereview.chromium.org/184103003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19725 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Notes about a few TODOs handled in this patch:

* In ProfileEntryHookStub::Generate:
    Stubs are always called with relocation.

* In CreateArrayDispatchOneArgument:
    The branches to registers can't be conditional. We could use a jump table, but
    there are only 6 different kinds so it is likely not worth it.

* In Builtins::Generate_StringConstructCode:
    Rename the argc register (x0) after its meaning changes.
    Remove a TODO: using a macro would not make the code clearer.

* In Generate_JSEntryTrampolineHelper:
    Remove the TODO and raise an internal issue to investigate this.

* In Disassembler::SubstituteBranchTargetField:
    Print the target address, but we don't have more info on the target.

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/185793002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19724 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This relands r19704 with a fix to the test case.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/189823003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19723 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Revert "Handle non-power-of-2 divisors in division-like operations", "A64 tweaks for division-like operations." and "Windows build fix.".

This reverts commit 19719, 19720 and 19721 because
mozilla/ecma/Date/15.9.3.1-1 fails (in release mode only?).

TBR=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/189963005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19722 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=ulan@chromium.org

Review URL: https://codereview.chromium.org/189093009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19721 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/190603003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19720 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/190383002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19719 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

When object is creating via native V8 API calls JS callstack is empty and the allocation is indistinguishable from say compiler allocations. This change adds a separate entry for such allocations.

Since FunctionInfo not necessarily corresponds to a heap object they are now referred to using their index in the list of all FunctionInfos.

BUG=chromium:277984
LOG=N
R=loislo@chromium.org, mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/177203002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19718 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This will reduce heavy ScopeIterator instantiations.
Once incorporated into chromium, will give 30% speed boost.

BUG=chromium:340285
LOG=Y
R=ulan@chromium.org, Yang, rossberg, ulan

Review URL: https://codereview.chromium.org/181063008

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19717 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The new test didn't recognize that non-sse2 builds on ia32 would
disable crankshaft.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/189263007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19716 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Lithium currently supports 3 division-like operations on integral operands: "Normal" division (rounding towards zero), flooring division (rounding towards -Infinity) and modulus calculation (the counterpart for the "normal" division). For divisors which are a power of 2, one can efficiently use some bit fiddling to avoid the actual division for such operations. This CL cleanly splits off these operations into separate Lithium instructions, making the code much more maintainable and more consistent across platforms.

There are 2 basic variations of these bit fiddling algorithms: One involving branches and a seemingly more clever one without branches. Choosing between the two is not as easy as it seems: Benchmarks (and probably real-world) programs seem to favor positive dividends, registers and shifting units are sometimes scarce resources, and branch prediction is quite good in modern processors. Therefore only the "normal" division by a power of 2 is implemented in a branch-free manner, this seems to be the best approach in practice. If this turns out to be wrong, we can easily and locally change this.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/175143002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19715 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Cache results of some heavy calls into C++.
This alone will boost the speed up to 18% on some real world scenarios.

BUG=chromium:340285
LOG=Y
R=ulan@chromium.org, yangguo@chromium.org, Yang, rossberg, ulan, yurys

Review URL: https://codereview.chromium.org/184483004

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19714 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/188543004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19713 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Type Vector followup: the type vector currently lives off the code object. This CL moves it to the SharedFunctionInfo, facilitating re-use and continued use in crankshafted code if desired.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/178463007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19712 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=142141
LOG=Y
R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/173793002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19711 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This reverts r19704.

R=mvstanton@chromium.org

Review URL: https://codereview.chromium.org/189533008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19710 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Due to performance regression.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/189843006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This patch avoid to push some values on the stack when it's not needed.

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/183193005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19708 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/188923003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19707 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Symbols for type cells. We can make more efficient code to check against type cells in the future if we use symbols, guaranteed not to conflict with user code. Currently, the "symbols" are the hole and undefined. Undefined may come in from the outside.

BUG=
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/181283003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19706 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jochen@chromium.org

Review URL: https://codereview.chromium.org/184433002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19705 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/178583006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19704 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Each item in INLINE_RUNTIME_FUNCTION_LIST had to have a duplicate
entry in RUNTIME_FUNCTION_LIST in order to match the comment.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/177313005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19703 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This feature makes it possible to associate data with a script and get it back
when the script is compiled or when an event is handled. It was historically
used by Chromium Dev Tools, but not any more. It is not used by node.js.

Note: this has nothing to do with the preparse data, despite the confusing name.
The preparse data is passed as ScriptData*.

Note 2: This is the same as r19616 ( https://codereview.chromium.org/184403002/ )
with a unused variable fix in bootstrapper.cc.

R=svenpanne@chromium.org
BUG=

Review URL: https://codereview.chromium.org/185533014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19702 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jarin@chromium.org

Review URL: https://codereview.chromium.org/179533003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19701 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/188483002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19700 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=dcarney@chromium.org
BUG=v8:3180
LOG=N

Review URL: https://codereview.chromium.org/180063002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19699 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Also drop redundant jmp instruction in deferred code.

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/182723003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19698 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=titzer@chromium.org
BUG=

Review URL: https://codereview.chromium.org/186593005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19695 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix for failing asserts in HBoundsCheck code generation on x64: use proper cmp operation width instead of asserting that Integer32 values should be zero extended. Similar to chromium:345820.

BUG=349465
LOG=N
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/188703002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19694 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=chromium:349878
LOG=y
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/186123003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@19693 ce2b1a6d-e550-0410-aec6-3dcde31c8c00