Revert "GVN fix, preventing loads hoisting above stores to the same field when HObjectAccess's representation is not the same."

This reverts commit r21830 for tanking performance on Deltablue.

TBR=ishell@chromium.org

Review URL: https://codereview.chromium.org/336223002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21857 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

HOptimizedGraphBuilder::TryInline creates a temporary CompilationInfo and
destroys it, but we don't want the AstValueFactory to be destroyed at the same
time. Reuse the upper CompilationInfo's AstValueFactory.

BUG=
R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/334173003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21851 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:3392
LOG=y
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/328343004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21850 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=marja@chormium.org
LOG=n
BUG=none

Review URL: https://codereview.chromium.org/337713003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21847 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Add wrappers to utils.h instead.

BUG=none
R=jkummerow@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/328343003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21846 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/336803002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21845 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Sadly, Vector ctor takes an argument "int length".

TBR=dcarney@chromium.org, mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/334903002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21844 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/338553002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21843 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/330793003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21842 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This is needed so that we can run Parser on a non-main thread (independent
of the Isolate and the V8 heap).

BUG=
R=rossberg@chromium.org

Review URL: https://codereview.chromium.org/314603004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21841 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

- May inline the function, or call it directly, instead of going through call
- Supports arguments object escaping when it escapes to builtins (preparation for slice.call(arguments, ...) optimization)
- Both .call and .apply now support inlining when calling builtins indirectly

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

Review URL: https://codereview.chromium.org/335683002

Patch from Petka Antonov <p.antonov@partner.samsung.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21840 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This avoids creating a new map for every bound function. Bonus: some cleanup in Runtime_FunctionBindArguments.

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/335653002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21839 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/335553004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21838 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/331823002

Patch from Weiliang Lin <weiliang.lin@intel.com>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21837 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

One more target had the architecture added to its name since r14209
originally landed and was missed when that change was reverted. Remove
the architecture from v8_libbase as well.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/334793004

Patch from Richard Coles <torne@chromium.org>.

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21836 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This should unbreak things on the Raspberry Pi.

LOG=y
R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/331803003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21835 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/329393005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21834 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/323403008

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21833 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/333643006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21832 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=3389
LOG=N
R=marja@chromium.org

Review URL: https://codereview.chromium.org/331823003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21831 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

GVN fix, preventing loads hoisting above stores to the same field when HObjectAccess's representation is not the same.

R=bmeurer@chromium.org

Review URL: https://codereview.chromium.org/331493006

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21830 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

port r21769

original message:
    Avoid right shifts by zero bits: On ARM it actually means shifting by
    32 bits (correctness issue) and on other platforms they are useless
    (performance issue). This is fix for the fix in r20544.

    BUG=v8:3259
    LOG=y
R=weiliang.lin@intel.com

Review URL: https://codereview.chromium.org/330133004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21829 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

We're basically trading undefined behavior for implementation defined
behavior, which should be OK for UBSan. :-) The generated code should
be identical, at least I checked that for GCC 4.6.3 on x64.

BUG=377790
LOG=y
R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/332733002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21828 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/321373002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21827 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/324913002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21826 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The at register is no longer in usage, so changed it to "scratch".

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/335753003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21825 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Initialize scratch register for all possible code paths.

TEST=mjsunit/compiler/division-by-constant.js
BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/333713002

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

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21822 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=machenbach@chromium.org

Review URL: https://codereview.chromium.org/329303004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21821 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
BUG=http://code.google.com/p/v8/issues/detail?id=2735
LOG=N

Review URL: https://codereview.chromium.org/332663004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21820 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=none
R=dcarney@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/328993003

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21819 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Hydrogen attempts to force representation changes on certain operations in order
to deoptimise on the change rather than the operation. However, these forced
changes are often unnecessary on 64-bit platforms, and cause poor code
generation, so this patch makes some of them conditional on whether it's
possible for deoptimisation to occur in the change.

On ARM64, this prevents sequences like:
                  ;;; <@46,#89> smi-tag
0x7ff282c7f050   144  lsl x4, x4, #32
                  ;;; <@48,#90> smi-untag
0x7ff282c7f054   148  asr x5, x4, #32
                  ;;; <@50,#31> mul-const-i-s
0x7ff282c7f058   152  lsl w6, w5, #3

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/303263010

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21818 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: https://codereview.chromium.org/332783002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21817 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This avoids endless IC patching cycles between "normal" and "nonexistent" handlers when objects having and not having the property are seen alternatingly

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/328353002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21816 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TEST=mjsunit/regress/regress-3380
BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/329223006

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

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21815 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=ishell@chromium.org

Review URL: https://codereview.chromium.org/321543004

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21814 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r21781 (4869c528)

BUG=
R=danno@chromium.org

Review URL: https://codereview.chromium.org/336543002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21813 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Due to mirror-object test failure (although I'm pretty sure I was framed)

R=mstarzinger@chromium.org
TBR=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/328333002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21812 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=dcarney@chromium.org

Review URL: https://codereview.chromium.org/330023002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21811 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

All places that use OS::Allocate either CHECK() that the result is non-NULL,
or use a reasonable fallback if they can't mmap memory.

BUG=none
R=mstarzinger@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/326323002

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21810 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=mstarzinger@chromium.org

BUG=

Review URL: https://codereview.chromium.org/333503005

git-svn-id: https://v8.googlecode.com/svn/branches/bleeding_edge@21809 ce2b1a6d-e550-0410-aec6-3dcde31c8c00