Commits · de0db0f59839c9083a126407c983363ad5b80aaf · Linshizhi / V8

09 Nov, 2011 3 commits

Remove ast-inl.h. This file is not necessary. · de0db0f5

kmillikin@chromium.org authored Nov 09, 2011

R=svenpanne@chromium.org
BUG=
TEST=

Review URL: http://codereview.chromium.org/8509005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9924 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

de0db0f5

MIPS: Tighten the code for MacroAssembler::ThrowUncatchable. · 6cb5ae7a

kmillikin@chromium.org authored Nov 09, 2011

Port r9909 (56c8728aa).

Original commit message:
Test at the bottom in the unwind loop.  Eliminate the possibility of a
useless move to the eax/rax/r0 register (currently impossible because
this function has two call sites).  Do not explicitly zero the context
because we've already saved 0 as the context in the handler.

BUG=
TEST=

Review URL: http://codereview.chromium.org/8507007
Patch from Gergely Kis <gergely@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9923 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

6cb5ae7a

Fix size of store buffer to depend on pointer size. · 5e0ff8b2

mstarzinger@chromium.org authored Nov 09, 2011

This makes the store buffer size depend on the pointer size so that
store buffers can hold the same number of entries on all architectures.
The number of pages marked scan-on-scavenge should decrease on x64 and
performance of splay should drastically improve on x64.

R=erik.corry@gmail.com

Review URL: http://codereview.chromium.org/8498013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9922 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5e0ff8b2

08 Nov, 2011 18 commits

Fixing issue 103259. · 53c6077c

yangguo@chromium.org authored Nov 08, 2011

BUG=103259
TEST=regress-103259.js

Review URL: http://codereview.chromium.org/8498011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9917 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

53c6077c

Make _CallFunction proxy-aware. · f936aac4

rossberg@chromium.org authored Nov 08, 2011

Change calling convention for CallFunction stub.
Some fixes regarding strict mode call traps.

R=kmillikin@chromium.org
BUG=
TEST=

Review URL: http://codereview.chromium.org/8318014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9916 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f936aac4

Reapply r9870 "Remove some initialization checks based on source positions.". · 72dba271

keuchel@chromium.org authored Nov 08, 2011

This reverts r9896 "Revert r9870 due to browser-test failures." See below for
the diff from the previous version for the ia32 platform. The code for other
platforms has been changed accordingly.

TEST=mjsunit/compiler/lazy-const-lookup.js

diff --git a/src/ia32/full-codegen-ia32.cc b/src/ia32/full-codegen-ia32.cc
index 2cbf518..1990f2f 100644
--- a/src/ia32/full-codegen-ia32.cc
+++ b/src/ia32/full-codegen-ia32.cc
@@ -1258,13 +1258,17 @@ void FullCodeGenerator::EmitVariableLoad(VariableProxy* proxy) {
         // binding is initialized:
         //   function() { f(); let x = 1; function f() { x = 2; } }
         //
-        // Check that we always have valid source position.
-        ASSERT(var->initializer_position() != RelocInfo::kNoPosition);
-        ASSERT(proxy->position() != RelocInfo::kNoPosition);
-        bool skip_init_check =
-            var->mode() != CONST &&
-            var->scope()->DeclarationScope() == scope()->DeclarationScope() &&
-            var->initializer_position() < proxy->position();
+        bool skip_init_check;
+        if (var->scope()->DeclarationScope() != scope()->DeclarationScope()) {
+          skip_init_check = false;
+        } else {
+          // Check that we always have valid source position.
+          ASSERT(var->initializer_position() != RelocInfo::kNoPosition);
+          ASSERT(proxy->position() != RelocInfo::kNoPosition);
+          skip_init_check = var->mode() != CONST &&
+              var->initializer_position() < proxy->position();
+        }
+
         if (!skip_init_check) {
           // Let and const need a read barrier.
           Label done;

Review URL: http://codereview.chromium.org/8479034

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9915 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

72dba271

Fix removed return parameter count. · 76a93404

danno@chromium.org authored Nov 08, 2011

BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8496011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9914 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

76a93404

Attempt to fix Windows x64 build. · 93559bb6

danno@chromium.org authored Nov 08, 2011

R=jkummerow@chromium.org
BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8495012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9913 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

93559bb6

Shrink the new space and uncommit marking deque on low memory notification. · 0d536dec

ulan@chromium.org authored Nov 08, 2011

BUG=v8:1669
TEST=cctest/test-heap/CollectingAllAvailableGarbageShrinksNewSpace

Review URL: http://codereview.chromium.org/8065003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9912 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0d536dec

Fix setting array length to be ES5 conform. · e24c612a

mstarzinger@chromium.org authored Nov 08, 2011

This also refactors the way we set the length of an arrays' backing
store to use the new elements accessor interface. The actual fix is in
DictionaryElementsAccessor::SetLengthWithoutNormalize() where we first
search for non-deletable elements according to ES5 section 15.4.5.2
specifications.

Snippet from the specification: Attempting to set the length property of
an Array object to a value that is numerically less than or equal to the
largest numeric property name of an existing array indexed non-deletable
property of the array will result in the length being set to a numeric
value that is one greater than that largest numeric property name.

R=danno@chromium.org
TEST=test262/15.4.4.??-7-b-16

Review URL: http://codereview.chromium.org/8372064

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9911 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

e24c612a

Another 64->32 bit warning from the Windows compiler. · 9b968b7d

erik.corry@gmail.com authored Nov 08, 2011

Review URL: http://codereview.chromium.org/8503010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9910 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9b968b7d

Tighten the code for MacroAssembler::ThrowUncatchable. · aee8ae05

kmillikin@chromium.org authored Nov 08, 2011

Test at the bottom in the unwind loop.  Eliminate the possibility of a
useless move to the eax/rax/r0 register (currently impossible because
this function has two call sites).  Do not explicitly zero the context
because we've already saved 0 as the context in the handler.

R=fschneider@chromium.org
BUG=
TEST=

Review URL: http://codereview.chromium.org/8493008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9909 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

aee8ae05

Try to fix compile error on Windows. · 59a472a7

erik.corry@gmail.com authored Nov 08, 2011

Review URL: http://codereview.chromium.org/8497010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9908 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

59a472a7

Disentangle header dependecies a bit by moving PropertyDetails-related types into a separate header · fbc0a9cd

svenpanne@chromium.org authored Nov 08, 2011

This moves PropertyType and its related predicate out of v8globals.h, where it didn't belong.

Review URL: http://codereview.chromium.org/8501006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9907 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

fbc0a9cd

Clean up the marking speed heuristics. This reduces the · cd8d915c

erik.corry@gmail.com authored Nov 08, 2011

max heap size on 64 bit from ca. 300Mbytes to ca. 200Mbytes
on Ulan's splay variant. On 32 bit not much change.
Review URL: http://codereview.chromium.org/8494012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9906 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

cd8d915c

Improve dominator computation to avoid worst-case quadratic time. · 2a1f08a1

fschneider@chromium.org authored Nov 08, 2011

In case of a degenerated CFG like in the example below processing
predecessors in the wrong order yields n^2 runtime.

  do {
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    if (x) break;
    // etc.
  } while (false);

Reversing iteration order avoids this.
Review URL: http://codereview.chromium.org/8502012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9905 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

2a1f08a1

Upstream Android V8 change. · 3628c934

danno@chromium.org authored Nov 08, 2011

Review URL: http://codereview.chromium.org/8351073
Patch from Jing Zhao <jingzhao@chromium.org>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9904 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3628c934

Create stub and runtime function for x64 full-codegen array literal element initialization. · 3d8b0a60

danno@chromium.org authored Nov 08, 2011

R=svenpanne@chromium.org
BUG=none
TEST=none

Review URL: http://codereview.chromium.org/8493024

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9903 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3d8b0a60

Revert r9901 to make tree green again. · 4627023b

fschneider@chromium.org authored Nov 08, 2011

There was a test failure on x64 mozilla tests.

TBR=ricow@chromium.org
Review URL: http://codereview.chromium.org/8495011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9902 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

4627023b

[hydrogen] optimize switch with string clauses · cac30084

fschneider@chromium.org authored Nov 08, 2011

Hydrogen should optimize not only SMI clauses, but clauses with string literals
too.

Patch from fedor.indutny <fedor.indutny@gmail.com>.

R=vegorov@chromium.org
BUG=
TEST=
Review URL: http://codereview.chromium.org/8373029

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9901 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

cac30084

Refactoring only: Make the handling of PropertyType more explicit. · 36b715b1

svenpanne@chromium.org authored Nov 08, 2011

Do not rely on 'default' clauses or 'if's when analysing a PropertyType, because
this makes it hard to find the relevant places when a new type is added. Note
that the detection of "phantom property types" is left untouched, because this
might have a performance impact, especially for the GC (to be investigated).

This is a preliminary step for introducing a new kind of map transition.

Review URL: http://codereview.chromium.org/8491016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9900 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

36b715b1

07 Nov, 2011 9 commits

Revert accidental commit that messed up the incremental speed heuristics. · b9a2d181

erik.corry@gmail.com authored Nov 07, 2011

Review URL: http://codereview.chromium.org/8467014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9899 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

b9a2d181

Revert r9870 due to browser-test failures. · c0d0bf86

keuchel@chromium.org authored Nov 07, 2011

This reverts commit 4d5b5f12aac932ad892c7b6f152b6168708d4210.

Review URL: http://codereview.chromium.org/8493006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9896 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

c0d0bf86

Optimize format message to not use DefineOwnProperty and freeze, do these inline · 3a416133

ricow@chromium.org authored Nov 07, 2011

Review URL: http://codereview.chromium.org/8494005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9893 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

3a416133

Don't set the limit that triggers the next GC higher than the max · 52b1457f

erik.corry@gmail.com authored Nov 07, 2011

old space size.
Review URL: http://codereview.chromium.org/8491009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9892 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

52b1457f

MIPS: Adding assertions to fast elements conversion. · cfc6f31e

yangguo@chromium.org authored Nov 07, 2011

Port r9875 (97dc1e0e).

BUG=
TEST=

Review URL: http://codereview.chromium.org/8467008
Patch from Gergely Kis <gergely@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9891 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

cfc6f31e

Prepare push to trunk. Now working on version 3.7.5. · 5bbd86bc

jkummerow@chromium.org authored Nov 07, 2011

R=yangguo@chromium.org

Review URL: http://codereview.chromium.org/8492005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9888 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

5bbd86bc

MIPS: Remove some initialization checks based on source positions. · aad9e50f

keuchel@chromium.org authored Nov 07, 2011

Port r9870 (959ec43).

Original commit message:

This depends on
http://codereview.chromium.org/8352039 and
http://codereview.chromium.org/8423005 .

BUG=
TEST=

Review URL: http://codereview.chromium.org/8467007
Patch from Gergely Kis <gergely@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9887 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

aad9e50f

Optimize DefineOneShotAccessor in messages.js to not call DefineOwnProperty. · 086ae442

ricow@chromium.org authored Nov 07, 2011

Review URL: http://codereview.chromium.org/8490008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9886 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

086ae442

MIPS: Cleanup: use JumpIf[Not]Smi() whenever we can · 9fa02bc5

jkummerow@chromium.org authored Nov 07, 2011

Port r8322 (7b8b4a951f).

BUG=
TEST=

Review URL: http://codereview.chromium.org/8428004
Patch from Gergely Kis <gergely@homejinni.com>.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9885 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9fa02bc5

04 Nov, 2011 5 commits

Override product_extension instead of product_version to set SONAME for Linux shared library · f33e9909

jkummerow@chromium.org authored Nov 04, 2011

BUG=v8:1786

Review URL: http://codereview.chromium.org/8462006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9884 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f33e9909

Revert r9596 due to page-cycler regressions. · bca8d42e

mstarzinger@chromium.org authored Nov 04, 2011

R=vegorov@chromium.org

Review URL: http://codereview.chromium.org/8463006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9883 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

bca8d42e

Fix Array.{splice,slice} to set proper ElementsKind of result · 9625d5d4

jkummerow@chromium.org authored Nov 04, 2011

TEST=mjsunit/elements-kind.js

Review URL: http://codereview.chromium.org/8430036

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9882 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

9625d5d4

Fix JSObject::EnsureCanContainElements to correctly iterate over Arguments · f2787a42

jkummerow@chromium.org authored Nov 04, 2011

TEST=mjsunit/elements-kind.js

Review URL: http://codereview.chromium.org/8437094

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9881 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

f2787a42

Fix Runtime_ArrayConcat to handle FAST_DOUBLE_ELEMENTS · 8450c60d

jkummerow@chromium.org authored Nov 04, 2011

TEST=mjsunit/elements-kind.js; stanford-crypto-sha256-iterative in debug mode

Review URL: http://codereview.chromium.org/8334028

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9880 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

8450c60d

03 Nov, 2011 5 commits

Adding assertions to fast elements conversion. · 29445458

yangguo@chromium.org authored Nov 03, 2011

Review URL: http://codereview.chromium.org/8437092

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9875 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

29445458

Clean up Scope::CollectUsedVariables. · 48fed9a8

keuchel@chromium.org authored Nov 03, 2011

Review URL: http://codereview.chromium.org/8438071

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9874 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

48fed9a8

Fix Harmony sets and maps to allow undefined as keys. · 0bbfb46a

mstarzinger@chromium.org authored Nov 03, 2011

This uses a global sentinel as a replacement for undefined keys, which
are not supported internally but required for Harmony sets and maps.

R=rossberg@chromium.org
BUG=v8:1622
TEST=mjsunit/harmony/collections

Review URL: http://codereview.chromium.org/8439069

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9873 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

0bbfb46a

Tune context allocation for variables accessed from inner scopes. · ea185392

keuchel@chromium.org authored Nov 03, 2011

After introduction of with scopes we have enough static information to omit
context allocation in the case that a variable is accessed from a nested block
or catch scope of the same function. Only variables accessed from the inside of
a nested function or with scope are forced to be allocated in the context.

This essentially reverts
http://code.google.com/p/v8/source/detail?r=9281 .
which in turn reverted an earlier change.

Review URL: http://codereview.chromium.org/8431001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9872 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

ea185392

Fix Harmony sets and maps to allow null as key. · 79cadcc9

mstarzinger@chromium.org authored Nov 03, 2011

This changes the internal convention for marking deleted entries in hash
tables from null_value to the_hole_value, which is consistent with other
usages of the_hole.

R=rossberg@chromium.org,kmillikin@chromium.org
BUG=v8:1622
TEST=mjsunit/harmony/collections

Review URL: http://codereview.chromium.org/8343056

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@9871 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

79cadcc9