This CL contains the first steps towards tail call optimization:

  * Structurally detect tail calls during instruction selection,
    looking for special return/call combinations.

  * Added new architecture-specific instructions for tail calls which
    jump instead of call and take care of frame adjustment.

  * Moved some code around.

Currently we restrict tail calls to callees which only use registers
for arguments/return value and to call sites which are explicitly
marked as being OK for tail calls. This excludes, among other things,
call sites in sloppy JS functions and our IC machinery (both need in
general to be able to access the caller's frame).

All this is behind a flag --turbo-tail-calls, which is currently off
by default, so it can easily be toggled.

Review URL: https://codereview.chromium.org/1108563002

Cr-Commit-Position: refs/heads/master@{#28150}

BUG=chromium:478460
LOG=n
NOTRY=true
TBR=bmeurer@chromium.org, mvstanton@chromium.org

Review URL: https://codereview.chromium.org/1107973007

Cr-Commit-Position: refs/heads/master@{#28149}

Revert of Remove the weak list of array buffers (patchset #8 id:140001 of https://codereview.chromium.org/1114563002/)

Reason for revert:
[Sheriff] Crashes in layout tests:
https://chromegw.corp.google.com/i/client.v8/builders/V8-Blink%20Linux%2064%20%28dbg%29/builds/2668

Original issue's description:
> Remove the weak list of array buffers
>
> Instead, collect live array buffers during marking and free pointers we
> no longer found.
>
> BUG=v8:3996
> R=hpayer@chromium.org
> LOG=n
>
> Committed: https://crrev.com/2d39709cf5ee17637f6f2d75380a9e61ae0b342b
> Cr-Commit-Position: refs/heads/master@{#28132}

TBR=dslomov@chromium.org,hpayer@chromium.org,jochen@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=v8:3996

Review URL: https://codereview.chromium.org/1115043005

Cr-Commit-Position: refs/heads/master@{#28148}

The method is not used anywhere, and it is a bad idea in general anyway.
If you see a need to call YieldCPU, then you're code is probably in need
of a redesign!

R=svenpanne@chromium.org

Review URL: https://codereview.chromium.org/1116853002

Cr-Commit-Position: refs/heads/master@{#28147}

It's useful for queue-like data structures.

R=yangguo@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1117713002

Cr-Commit-Position: refs/heads/master@{#28146}

Fix AIX compiler warning indicating that variable may not be
initialized

	modified:   src/heap/heap.cc

R=danno@chromium.org, svenpanne@chromium.org, mbrandy@us.ibm.com

BUG=

Review URL: https://codereview.chromium.org/1112883002

Cr-Commit-Position: refs/heads/master@{#28145}

Switch to larger TOC on AIX for unitttests as the size has grown
such that it no longer compiles with the default TOC size

	modified:   unittests.gyp

R=danno@chromium.org, svenpanne@chromium.org, mbrandy@us.ibm.com

BUG=

Review URL: https://codereview.chromium.org/1115763003

Cr-Commit-Position: refs/heads/master@{#28144}

Added rounding according to fcsr, CVT_W_D and RINT.D instruction in assembler, dissasembler and simulator and wrote appropiate tests.

BUG=

Review URL: https://codereview.chromium.org/1108583003

Cr-Commit-Position: refs/heads/master@{#28143}

Now it's possible to specify the desired trybots for perf
tries, e.g.:

tools/try_perf.py --linux64_haswell octane sunspider

BUG=chromium:478460
LOG=n
NOTRY=true

Review URL: https://codereview.chromium.org/1114913002

Cr-Commit-Position: refs/heads/master@{#28142}

Rolling v8/build/gyp to 0bb67471bca068996e15b56738fa4824dfa19de0

Rolling v8/buildtools to 15f5fc6fdb1795e2b99f66e5bc6c01e9fb62b436

TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/1110063010

Cr-Commit-Position: refs/heads/master@{#28141}

- allows the optimization of emitted gap move code since the representation of the value in the register is known
- necessary preparation for vector register allocation
- prepare for slot sharing for any value of the same byte width

TBR=jarin@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1111323003

Cr-Commit-Position: refs/heads/master@{#28140}

Revert of [turbofan] add MachineType to AllocatedOperand (patchset #17 id:310001 of https://codereview.chromium.org/1087793002/)

Reason for revert:
[Sheriff] Breaks compile on chromium asan and v8 msan:
http://build.chromium.org/p/client.v8/builders/Linux%20ASAN%20Builder/builds/3446
http://build.chromium.org/p/client.v8/builders/V8%20Linux%20-%20arm64%20-%20sim%20-%20MSAN/builds/2085

Original issue's description:
> [turbofan] add MachineType to AllocatedOperand
>
> - allows the optimization of emitted gap move code since the representation of the value in the register is known
> - necessary preparation for vector register allocation
> - prepare for slot sharing for any value of the same byte width
>
> BUG=
>
> Committed: https://crrev.com/3a025d1ab6437559f86a464767aa03d2d9789f6f
> Cr-Commit-Position: refs/heads/master@{#28137}

TBR=jarin@chromium.org,dcarney@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=

Review URL: https://codereview.chromium.org/1119483003

Cr-Commit-Position: refs/heads/master@{#28139}

R=bmeurer@chromium.org
TEST=cctest/test-node

Review URL: https://codereview.chromium.org/1118433002

Cr-Commit-Position: refs/heads/master@{#28138}

- allows the optimization of emitted gap move code since the representation of the value in the register is known
- necessary preparation for vector register allocation
- prepare for slot sharing for any value of the same byte width

BUG=

Review URL: https://codereview.chromium.org/1087793002

Cr-Commit-Position: refs/heads/master@{#28137}

R=jarin@chromium.org
TEST=unittests/NodeCacheTest

Review URL: https://codereview.chromium.org/1114903002

Cr-Commit-Position: refs/heads/master@{#28136}

Note, that it also moves incremental marking traces behind the --trace-incremental-marking flag.

Other PrintF's of the GC should be moved as well to PrintPIDAndIsolate.

BUG=

Review URL: https://codereview.chromium.org/1109133002

Cr-Commit-Position: refs/heads/master@{#28135}

R=jarin@chromium.org
TEST=cctest/test-graph-visualizer

Review URL: https://codereview.chromium.org/1115603003

Cr-Commit-Position: refs/heads/master@{#28134}

js2c.py now distinguishes between the JS internal sources and any
extra scripts passed in. The latter only get validation and
comment/trailing whitespace removal applied now.

Also added a --js option to js2c.py, which will output a JS file
instead of a C++ file, for debugging. (I got tired of finding the
generated .cc file and extracting its byte array so that I could feed
it to a separate helper script I wrote.)

R=yangguo@chromium.org, jochen@chromium.org
BUG=v8:4064
LOG=N

Review URL: https://codereview.chromium.org/1117663002

Cr-Commit-Position: refs/heads/master@{#28133}

Instead, collect live array buffers during marking and free pointers we
no longer found.

BUG=v8:3996
R=hpayer@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/1114563002

Cr-Commit-Position: refs/heads/master@{#28132}

R=arv@chromium.org
BUG=v8:811
LOG=N

Review URL: https://codereview.chromium.org/1112073002

Cr-Commit-Position: refs/heads/master@{#28131}

Followup fix of nit discussed in Issue 1111003002.

BUG=

Review URL: https://codereview.chromium.org/1112793004

Cr-Commit-Position: refs/heads/master@{#28130}

R=dcarney@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1112823002

Cr-Commit-Position: refs/heads/master@{#28129}

port caeb9004 (r28056)

original commit message:
    If the array's map is the initial FastHoley array map, and the array prototype
    chain is undisturbed and empty of elements, then keyed loads can convert the
    load of a hole to undefined.

BUG=

Review URL: https://codereview.chromium.org/1104073003

Cr-Commit-Position: refs/heads/master@{#28128}

We shouldn't have shared state between isolates by default. The embedder
is free to pass the same allocator to all isolates it creates.

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

Review URL: https://codereview.chromium.org/1116633002

Cr-Commit-Position: refs/heads/master@{#28127}

BUG=chromium:478522
LOG=y
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/1097113003

Cr-Commit-Position: refs/heads/master@{#28126}

R=titzer@chromium.org
TEST=mjsunit/const-redecl

Review URL: https://codereview.chromium.org/1115613002

Cr-Commit-Position: refs/heads/master@{#28125}

Review URL: https://codereview.chromium.org/1109153002

Cr-Commit-Position: refs/heads/master@{#28124}

BUG=

Review URL: https://codereview.chromium.org/1110373002

Cr-Commit-Position: refs/heads/master@{#28123}

NOTRY=true
BUG=chromium:425187
LOG=n
TBR=Sven Panne, titzer

Review URL: https://codereview.chromium.org/1110903004

Cr-Commit-Position: refs/heads/master@{#28122}

Revert of deprecate non-phantom weak callbacks (patchset #1 id:1 of https://codereview.chromium.org/1103173002/)

Reason for revert:
[Sheriff] This is still used and now blocks the roll here:
http://build.chromium.org/p/tryserver.chromium.mac/builders/mac_chromium_compile_dbg_ng/builds/47889

Original issue's description:
> Reland: deprecate non-phantom weak callbacks
>
> BUG=
>
> Committed: https://crrev.com/3be656fb55ad49834ec8ec0a57ba858bb1b21f37
> Cr-Commit-Position: refs/heads/master@{#28094}

TBR=jochen@chromium.org,dcarney@chromium.org
NOPRESUBMIT=true
NOTREECHECKS=true
NOTRY=true
BUG=

Review URL: https://codereview.chromium.org/1117623002

Cr-Commit-Position: refs/heads/master@{#28121}

Review URL: https://codereview.chromium.org/1053123006

Cr-Commit-Position: refs/heads/master@{#28120}

Rolling v8/tools/clang to 44f7f0bbbce9e581d9328e80d5b0ae88efbc3db4

TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/1108203005

Cr-Commit-Position: refs/heads/master@{#28119}

BUG=
R=jochen@chromium.org, dcarney@chromium.org, dstence@us.ibm.com, michael_dawson@ca.ibm.com

Review URL: https://codereview.chromium.org/1108373002

Cr-Commit-Position: refs/heads/master@{#28118}

BUG=

Review URL: https://codereview.chromium.org/1111003002

Cr-Commit-Position: refs/heads/master@{#28117}

Fix faulty condition register usage after andi.

BUG=
R=dstence@us.ibm.com, michael_dawson@ca.ibm.com, jochen@chromium.org, bmeurer@chromium.org

Review URL: https://codereview.chromium.org/1108363002

Cr-Commit-Position: refs/heads/master@{#28116}

BUG=v8:3330
LOG=N

Review URL: https://codereview.chromium.org/1109783003

Cr-Commit-Position: refs/heads/master@{#28115}

In a derived class constructor in case undefined is returned, we
should return the receiver. If the return is any other value type
we should throw a TypeError.

BUG=v8:4061
LOG=N
R=dslomov@chromium.org, adamk@chromium.org

Review URL: https://codereview.chromium.org/1107203002

Cr-Commit-Position: refs/heads/master@{#28114}

BUG=481519
LOG=y

Review URL: https://codereview.chromium.org/1111633002

Cr-Commit-Position: refs/heads/master@{#28113}

Just parsing, no desugaring yet.

R=arv@chromium.org,rossberg@chromium.org
BUG=v8:811
LOG=N

Review URL: https://codereview.chromium.org/1107053002

Cr-Commit-Position: refs/heads/master@{#28112}

The use of "const" prevented the actual JavaScript const keyword from
functioning as intended.

R=jochen@chromium.org,yangguo@chromium.org
BUG=

Review URL: https://codereview.chromium.org/1110963003

Cr-Commit-Position: refs/heads/master@{#28111}