BUG=354843
LOG=N
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/208283002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20167 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/198053013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20166 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=danno@chromium.org

Review URL: https://codereview.chromium.org/208253002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20165 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/208023002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20164 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Fix Q register encoding for registers other than Q0. Also, fix value in NeonSize
enumeration.

BUG=
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/207523005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20163 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

We'd like to be able to trade nested scope chain info (consisting of with, block and catch scopes) in favor of speed in some cases.

BUG=chromium:340285
LOG=N
R=ulan@chromium.org, pfeldman, ulan, yangguo

Review URL: https://codereview.chromium.org/203463011

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20162 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/208003002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20161 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/197233011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20160 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This moves rolling Chromium into a new script with its own tests.

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/199733012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20159 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=354357
LOG=N
TEST=mjsunit/regress/regress-354357.js
R=mstarzinger@chromium.org

Review URL: https://codereview.chromium.org/206413005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20158 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=machenbach@chromium.org
BUG=v8:2999
LOG=N

Review URL: https://codereview.chromium.org/208043002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20157 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/206073007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20156 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=354433
LOG=N
TEST=mjsunit/regress/regress-354433.js
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/198463006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20155 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This reverts commit r20146 which broke V8 GC Stress, Mjsunit tests.

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/207963002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20154 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Functions moved: ParseMemberWithNewPrefixesExpression, ParseMemberExpression,
ParseMemberExpressionContinuation.

Now all Parse*Expression functions are in ParserBase.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/207633003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20153 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=machenbach@chromium.org
TBR=machenbach@chromium.org

Review URL: https://codereview.chromium.org/207633004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20152 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=v8:3226
LOG=n
R=ulan@chromium.org

Review URL: https://codereview.chromium.org/205863004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20151 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Includes cleanups:
- Reorganized functions in PreParserFactory to be in the logical order.
- De-hackified things PreParser doesn't need to track, such as IsCall & IsCallNew.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/206433003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20150 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

+ enable a test which checks that Parser and PreParser produce the "invalid left
hand side" errors consistently.

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/202333004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20149 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=354405
R=ulan@chromium.org, rodolph.perfetta@arm.com
LOG=y

Review URL: https://codereview.chromium.org/207823003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20148 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=354405
R=machenbach@chromium.org
LOG=n

Review URL: https://codereview.chromium.org/207793002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20147 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/206343002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20146 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/206223003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20145 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

They're now all defined globally

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

Review URL: https://codereview.chromium.org/207723003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20144 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/206463002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20143 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This reverts commit r20141 which broke the build with some GCC versions. The
reason behind that is GCC emits a warning about code which can never be reached,
but GCC can't figure that out.

The right approach would be using template specialization instead of these
runtime if/then/else cascades, which would be nicer from a SW engineering
perspective, too.

TBR=yangguo@chromium.org

Review URL: https://codereview.chromium.org/207703003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20142 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Compiler can't optimize away variable length memcpy.
About a 18% boost.

Before - 5:40

Richards: 75.2
DeltaBlue: 105
Crypto: 64.0
RayTrace: 188
EarleyBoyer: 146
RegExp: 23.8
Splay: 96.2
NavierStokes: 91.9
----
Score (version 7): 85.7

After - 4:39

Richards: 90.8
DeltaBlue: 134
Crypto: 81.7
RayTrace: 227
EarleyBoyer: 177
RegExp: 29.7
Splay: 126
NavierStokes: 108
----
Score (version 7): 106

R=rodolph.perfetta@gmail.com, svenpanne@chromium.org

Review URL: https://codereview.chromium.org/203263017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20141 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/196893003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20140 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/206763002

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20139 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=rossberg@chromium.org
BUG=chromium:354123
TEST=cctest/test-api/Regress354123
LOG=y

Review URL: https://codereview.chromium.org/205033011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20138 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=chromium:354391
LOG=y
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/206073008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20137 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Only change relative to original CL is the updated assertion condition at objects-inl.h:2119

R=mstarzinger@chromium.org
BUG=

Review URL: https://codereview.chromium.org/204913006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20136 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Port r20123 (31ab416)

BUG=
R=plind44@gmail.com

Review URL: https://codereview.chromium.org/206373004

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

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20135 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
R=jkummerow@chromium.org

Review URL: https://codereview.chromium.org/206133004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20132 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/206393002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20131 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ulan@chromium.org

Review URL: https://codereview.chromium.org/206473002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20130 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=ishell@chromium.org

Review URL: https://codereview.chromium.org/206373003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20129 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

R=yangguo@chromium.org

Review URL: https://codereview.chromium.org/206383003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20128 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The issue was that a previous fix to avoid a logic error was not
targeted enough: it allowed non-transitioning stores to fall
into a code path meant to preserve monomorphic state in the
case of a transitioning store.

BUG=353661
LOG=N
R=verwaest@chromium.org

Review URL: https://codereview.chromium.org/206353002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20127 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The limit was originally added to avoid having large user-controlled constants
(variable indexes) in the code generated by full-codegen.

History behind this change:

The original CL for adding the limit was https://codereview.chromium.org/7003030
and at that time, the limit was 32767.

Reason for adding the limit (in CL comments): "The motivation behind this change
is to avoid large user controlled constants in the code. The slot_operand used
in the IA32 full code generator uses a relative load where the local index is an
(negative) immediate."

The limit was then bumped to 65535 by https://codereview.chromium.org/10965063
and to 131071 by https://codereview.chromium.org/11099063.

R=dcarney@chromium.org, svenpanne@chromium.org, jkummerow@chromium.org, rossberg@chromium.org
BUG=v8:3205
LOG=Y

Review URL: https://codereview.chromium.org/206143004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@20126 ce2b1a6d-e550-0410-aec6-3dcde31c8c00