platform/upstream/v8.git
14 years agox64: Fix code to return from debug break slot
sgjesse@chromium.org [Tue, 8 Jun 2010 12:50:42 +0000 (12:50 +0000)]
x64: Fix code to return from debug break slot

On x64 the return from debug break slot did not pop off the return address.

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/2731002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4823 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd test_b(Operand, imm8) to ia32 disassembler.
whesse@chromium.org [Tue, 8 Jun 2010 12:44:24 +0000 (12:44 +0000)]
Add test_b(Operand, imm8) to ia32 disassembler.
Review URL: http://codereview.chromium.org/2765001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4822 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd parenthesis around '&&' within '||'
sgjesse@chromium.org [Tue, 8 Jun 2010 12:20:24 +0000 (12:20 +0000)]
Add parenthesis around '&&' within '||'

Warning by some gcc versions. Adding the parenthesis made the condition correct.

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/2725003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4821 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMore precise break points and stepping when debugging
sgjesse@chromium.org [Tue, 8 Jun 2010 12:04:49 +0000 (12:04 +0000)]
More precise break points and stepping when debugging

Added support for more precise break points when debugging and stepping. To achieve that additional nop instructions are inserted where breaking would otherwise be impossible. The number of nop instructions inserted are sufficient to make place for patching with a call to a debug break code stub. On Intel that is 5 nop's for 32-bit and 13 for 64-bit. Om ARM 3 nop instructions (12 bytes) are required.

In order to avoid inserting nop's in to many places a simple ast checker have been added to check whether there are breakable code in a statement or expression. If it is possible to break in an expression no additional break enabeling code is inserted.

Added break locations to the true and false part of a conditional expression.

Added stepping tests to cover more constructs.

These changes are only in the full compiler.

Changed the default value for the option --debugger in teh d8 shell from true to false. The reason for this is that with --debugger turned on the full compiler will be used for all code in when running d8, which can be unexpeceted.

Review URL: http://codereview.chromium.org/2693002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4820 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMove token-related constants from CodeEntry to TokenEnumerator.
mikhail.naganov@gmail.com [Tue, 8 Jun 2010 11:27:00 +0000 (11:27 +0000)]
Move token-related constants from CodeEntry to TokenEnumerator.

Review URL: http://codereview.chromium.org/2745002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4819 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix building without debugger suport.
sgjesse@chromium.org [Tue, 8 Jun 2010 11:23:48 +0000 (11:23 +0000)]
Fix building without debugger suport.
Review URL: http://codereview.chromium.org/2733003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4818 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix bug in mjsunit/const-eval-init causing objects to be initialized to undefined.
ricow@chromium.org [Tue, 8 Jun 2010 11:20:36 +0000 (11:20 +0000)]
Fix bug in mjsunit/const-eval-init causing objects to be initialized to undefined.

In mjsunit/const-eval-init - testInitSlowCaseExtension a range of objects are initialized to undefined instead of a value because the variable i is within quotes (i.e., the source becommes "a1 = i" instead of "a1 = 1".

This should have no impact on the test, I just stumbled over this on
an unrelated matter.

Review URL: http://codereview.chromium.org/2758002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4817 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRemove the SetExternalStringDiposeCallback API
sgjesse@chromium.org [Tue, 8 Jun 2010 09:35:47 +0000 (09:35 +0000)]
Remove the SetExternalStringDiposeCallback API

Changed the disposal of external string resources to call a virtual Dispose method on the resource. The default inplementation of Dispose deletes the object and will capture the delete operator matching the new operator used to allocate the object.
Review URL: http://codereview.chromium.org/2658008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4816 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd an overloaded PreCompile method that takes a Handle<String>.
ager@chromium.org [Tue, 8 Jun 2010 06:20:49 +0000 (06:20 +0000)]
Add an overloaded PreCompile method that takes a Handle<String>.

Code review URL: http://codereview.chromium.org/2633004/show

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4815 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFlushing of code from functions that we expect not to use again.
ricow@chromium.org [Mon, 7 Jun 2010 15:39:10 +0000 (15:39 +0000)]
Flushing of code from functions that we expect not to use again.

This adds an additional step to full gc, removing code from functions
that are no longer in the compilation cache. The code is replaced with
a lazy compile version enabling us to recompile the function in case
we do actually need it again.

Review URL: http://codereview.chromium.org/2632003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4814 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRevert change 4812 due to performance problems. This was a change in string addition...
whesse@chromium.org [Mon, 7 Jun 2010 13:16:05 +0000 (13:16 +0000)]
Revert change 4812 due to performance problems.  This was a change in string addition on ia32.
Review URL: http://codereview.chromium.org/2706001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4813 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoImprove generated code for string encoding tests on ia32.
whesse@chromium.org [Mon, 7 Jun 2010 12:15:10 +0000 (12:15 +0000)]
Improve generated code for string encoding tests on ia32.
Review URL: http://codereview.chromium.org/2673001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4812 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCount Mark-Compact/Mark-Sweep collections only after compaction decision was made
vegorov@chromium.org [Mon, 7 Jun 2010 11:31:44 +0000 (11:31 +0000)]
Count Mark-Compact/Mark-Sweep collections only after compaction decision was made

Review URL: http://codereview.chromium.org/2665001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4811 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPrepare push to trunk. We are now working on version 2.2.16.
lrn@chromium.org [Mon, 7 Jun 2010 11:02:43 +0000 (11:02 +0000)]
Prepare push to trunk. We are now working on version 2.2.16.

Review URL: http://codereview.chromium.org/2640007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4808 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd regression tests for issues 728, 732
vegorov@chromium.org [Mon, 7 Jun 2010 10:54:42 +0000 (10:54 +0000)]
Add regression tests for issues 728, 732

TBR=lrn@chromium.org

Review URL: http://codereview.chromium.org/2698004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4807 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRevert r4782. Fix issues 728, 732.
vegorov@chromium.org [Mon, 7 Jun 2010 09:36:30 +0000 (09:36 +0000)]
Revert r4782. Fix issues 728, 732.

Review URL: http://codereview.chromium.org/2701003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4806 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd an API to control the disposal of external string resources
sgjesse@chromium.org [Mon, 7 Jun 2010 08:49:07 +0000 (08:49 +0000)]
Add an API to control the disposal of external string resources

A new callback have been added which - if set - will be called to dispose of external string resources passed from the embedder to the V8 engine.
Review URL: http://codereview.chromium.org/2645004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4805 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoExtend CallIC to support non-constant names.
kaznacheev@chromium.org [Mon, 7 Jun 2010 08:27:32 +0000 (08:27 +0000)]
Extend CallIC to support non-constant names.
This speeds up constructs like this:
var zz='replace'; '123'[zz]('3','4');

Review URL: http://codereview.chromium.org/2280007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4804 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPort fast eval calls to x64 and ARM. For global variables that can
ager@chromium.org [Mon, 7 Jun 2010 07:12:00 +0000 (07:12 +0000)]
Port fast eval calls to x64 and ARM.  For global variables that can
only be shadowed by eval-introduced variables we usually do not need
to do a context lookup in the runtime system.

Review URL: http://codereview.chromium.org/2672001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4803 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoDirect load of global function prototype.
vitalyr@chromium.org [Fri, 4 Jun 2010 16:20:34 +0000 (16:20 +0000)]
Direct load of global function prototype.

As most of call IC code is tied to a context anyway we can save a few dependent loads by having a direct reference to an initial map of a global function.

Review URL: http://codereview.chromium.org/2239009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4802 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix issue in previous test optimization
sandholm@chromium.org [Fri, 4 Jun 2010 13:43:41 +0000 (13:43 +0000)]
Fix issue in previous test optimization
Review URL: http://codereview.chromium.org/2653002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4801 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoSimplify criteria for doing an optimized test call
sandholm@chromium.org [Fri, 4 Jun 2010 13:37:28 +0000 (13:37 +0000)]
Simplify criteria for doing an optimized test call
Review URL: http://codereview.chromium.org/2662002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4800 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoSimplify regexp test call when the regexp starts and ends with '.*'.
sandholm@chromium.org [Fri, 4 Jun 2010 11:58:40 +0000 (11:58 +0000)]
Simplify regexp test call when the regexp starts and ends with '.*'.
Review URL: http://codereview.chromium.org/2645002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4799 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRename ThreadManager::IterateThreads to ThreadManager::IterateArchivedThreads
ricow@chromium.org [Fri, 4 Jun 2010 11:57:01 +0000 (11:57 +0000)]
Rename ThreadManager::IterateThreads to ThreadManager::IterateArchivedThreads

The IterateThreads method only iterates archived threads. If all
threads are to be iterated an additional iteration of the current active
stack is also needed.

Review URL: http://codereview.chromium.org/2655002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4798 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFirst phase of migration to new named property query callbacks.
antonm@chromium.org [Fri, 4 Jun 2010 11:49:44 +0000 (11:49 +0000)]
First phase of migration to new named property query callbacks.

Eventually named property query callbacks will return attributes
(as an integer) or an empty handle if property is not intercepted.

To gradually migrate to this new API, USE_NEW_QUERY_CALLBACK
macro would control if old or new style API is used.

So the migration plan is:

1) introduce new API which should be explictily enabled;
2) switch to new API defining USE_NEW_QUERY_CALLBACK before
  include of <v8.h> (that would require changes to client code as well)
3) remove old API from v8
4) remove #define USE_NEW_QUERY_CALLBACK from clients.

Review URL: http://codereview.chromium.org/2576003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4797 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd optimized version of memcpy on ia32.
lrn@chromium.org [Fri, 4 Jun 2010 11:30:55 +0000 (11:30 +0000)]
Add optimized version of memcpy on ia32.
Only used in one place right now.
Still room for tweaking.

Review URL: http://codereview.chromium.org/2582001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4796 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd test_b(Operand, immediate) to ia32 assembler, and use it where possible. Improve...
whesse@chromium.org [Fri, 4 Jun 2010 10:46:04 +0000 (10:46 +0000)]
Add test_b(Operand, immediate) to ia32 assembler, and use it where possible.  Improve comparison to a constant one-character string.  Use CmpInstanceType in more places on ia32.  Add IsObjectJSObjectType and IsInstanceJSObjectType to ia32 macro assembler, using a single branch for a range test.
Review URL: http://codereview.chromium.org/2586001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4795 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoOptimize calls to evals. Most of the time there is no reason to
ager@chromium.org [Fri, 4 Jun 2010 10:19:19 +0000 (10:19 +0000)]
Optimize calls to evals. Most of the time there is no reason to
perform a context lookup in the runtime system for the 'eval'
function. Instead load the 'eval' function from the global context in
generated code if it is not shadowed.

Will port to other platforms as a separate change.

Review URL: http://codereview.chromium.org/2666001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4794 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRemove unused tpedef.
dimich@chromium.org [Fri, 4 Jun 2010 07:53:40 +0000 (07:53 +0000)]
Remove unused tpedef.

Review URL: http://codereview.chromium.org/2560002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4793 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix double to integer conversions in runtime string indexing.
vitalyr@chromium.org [Thu, 3 Jun 2010 11:31:19 +0000 (11:31 +0000)]
Fix double to integer conversions in runtime string indexing.

Review URL: http://codereview.chromium.org/2577001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4791 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMake StringCharCodeAt runtime function do what it name implies and not
vitalyr@chromium.org [Thu, 3 Jun 2010 09:32:07 +0000 (09:32 +0000)]
Make StringCharCodeAt runtime function do what it name implies and not
just indexed string access.

The difference is that String.prototype.charCodeAt accepts any number
as an index, whereas string[] only accepts array indices.

Review URL: http://codereview.chromium.org/2467006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4790 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoChange keyed load IC interface on x64 to pass arguments in registers.
whesse@chromium.org [Wed, 2 Jun 2010 14:37:47 +0000 (14:37 +0000)]
Change keyed load IC interface on x64 to pass arguments in registers.
Review URL: http://codereview.chromium.org/2470001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4787 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix bug in test case that made it fail to compile in non-debug mode.
lrn@chromium.org [Wed, 2 Jun 2010 11:05:06 +0000 (11:05 +0000)]
Fix bug in test case that made it fail to compile in non-debug mode.
Using ASSERT_EQ instead of CHECK_EQ means that a variable becomes unused.

Review URL: http://codereview.chromium.org/2475003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4785 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix last-minute thinko that I snuck into the last change.
erik.corry@gmail.com [Wed, 2 Jun 2010 10:47:50 +0000 (10:47 +0000)]
Fix last-minute thinko that I snuck into the last change.
It busted crypto-decrypt.
Review URL: http://codereview.chromium.org/2434005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4784 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Track Smis on top 4 stack positions and Smi loop variables.
erik.corry@gmail.com [Wed, 2 Jun 2010 09:37:02 +0000 (09:37 +0000)]
ARM: Track Smis on top 4 stack positions and Smi loop variables.
Improve code generation for known smis and suspected Smis.
Review URL: http://codereview.chromium.org/2452002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4783 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix bug that could cause a string to be incorrectly tagged as an array index.
lrn@chromium.org [Wed, 2 Jun 2010 09:31:01 +0000 (09:31 +0000)]
Fix bug that could cause a string to be incorrectly tagged as an array index.
We should only mark a string as an array index if we can store the entire value
of the number in the hash field. We sometimes failed to reject larger numbers.

Fixes http://code.google.com/p/v8/issues/detail?id=728

Review URL: http://codereview.chromium.org/2452007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4782 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd missing initialization of a couple of variables which makes
ager@chromium.org [Wed, 2 Jun 2010 08:58:44 +0000 (08:58 +0000)]
Add missing initialization of a couple of variables which makes
some compilers complaint when compiling with -Werror.
Review URL: http://codereview.chromium.org/2492001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4780 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPrepare push to trunk. We are now working on version 2.2.15.
sgjesse@chromium.org [Wed, 2 Jun 2010 08:48:51 +0000 (08:48 +0000)]
Prepare push to trunk. We are now working on version 2.2.15.
Review URL: http://codereview.chromium.org/2467004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4778 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoInitialize variable to make gcc happy.
erik.corry@gmail.com [Wed, 2 Jun 2010 08:17:19 +0000 (08:17 +0000)]
Initialize variable to make gcc happy.
http://code.google.com/p/v8/issues/detail?id=727

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4777 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Add more logic to the generic keyed load stub
sgjesse@chromium.org [Tue, 1 Jun 2010 21:11:38 +0000 (21:11 +0000)]
ARM: Add more logic to the generic keyed load stub

The generic keyed load stub was missing converting string keys to numberign keys and lookup in the keyed lookup cache. This is now added together with an additional counter on all platforms.
Review URL: http://codereview.chromium.org/2441002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4773 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix generated code for String.charCodeAt.
vitalyr@chromium.org [Tue, 1 Jun 2010 14:08:19 +0000 (14:08 +0000)]
Fix generated code for String.charCodeAt.

Review URL: http://codereview.chromium.org/2468001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4770 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCPU profiler: sample call stack on profiling start.
mikhail.naganov@gmail.com [Tue, 1 Jun 2010 13:52:49 +0000 (13:52 +0000)]
CPU profiler: sample call stack on profiling start.

This simplifies writing tests a lot, because it is now possible
to guarantee that certain function will present in a profile by
starting profiler from inside it.

Review URL: http://codereview.chromium.org/2438002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4769 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoUnreviewed. Revert r4766, r4767.
mikhail.naganov@gmail.com [Tue, 1 Jun 2010 13:42:48 +0000 (13:42 +0000)]
Unreviewed. Revert r4766, r4767.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4768 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agostarted adding test
mikhail.naganov@gmail.com [Tue, 1 Jun 2010 13:35:44 +0000 (13:35 +0000)]
started adding test

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4767 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoadded function
mikhail.naganov@gmail.com [Tue, 1 Jun 2010 13:35:09 +0000 (13:35 +0000)]
added function

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4766 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix an occasional crash in Assembler::ldr() for arm.
dimich@chromium.org [Tue, 1 Jun 2010 08:01:50 +0000 (08:01 +0000)]
Fix an occasional crash in Assembler::ldr() for arm.
A peephole optimization can be attempted on a buffer that contains only a single command. The crash happens when running debug on Snow Leopard with --simulator=arm.

Review URL: http://codereview.chromium.org/2454001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4762 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoChange the interface of LoadIC on the x64 platform to take its arguments in registers.
whesse@chromium.org [Mon, 31 May 2010 13:26:12 +0000 (13:26 +0000)]
Change the interface of LoadIC on the x64 platform to take its arguments in registers.
Review URL: http://codereview.chromium.org/2330003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4759 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPrepare push to trunk. Now working on version 2.2.14.
ricow@chromium.org [Mon, 31 May 2010 10:13:11 +0000 (10:13 +0000)]
Prepare push to trunk. Now working on version 2.2.14.
Review URL: http://codereview.chromium.org/2417001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4755 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix AddBlock invocations in CpuProfilesCollection.
mikhail.naganov@gmail.com [Mon, 31 May 2010 10:09:07 +0000 (10:09 +0000)]
Fix AddBlock invocations in CpuProfilesCollection.

It was a bad idea not to check the count of numbers to add.

Also fix a rollover: the comment in platform-linux.

Review URL: http://codereview.chromium.org/2418001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4754 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Remove LoadConditionAndSpill and VisitAndSpill.
erik.corry@gmail.com [Mon, 31 May 2010 08:52:57 +0000 (08:52 +0000)]
ARM: Remove LoadConditionAndSpill and VisitAndSpill.
Review URL: http://codereview.chromium.org/2414001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4753 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoReapply r4702 "Fix teardown order"
mikhail.naganov@gmail.com [Fri, 28 May 2010 16:12:24 +0000 (16:12 +0000)]
Reapply r4702 "Fix teardown order"

I accidentaly reverted changes from r4702 in r4706. Now bringing them back.

Original commit message:

As TokenEnumerator cleans up global handles posessed, it's owner --
CpuProfiler must be shut down prior to global handles.

Review URL: http://codereview.chromium.org/2072013

TBR=sgjesse@chromium.org

Review URL: http://codereview.chromium.org/2325003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4752 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMake intercepted properties retrievable only by getter to be not enumerable.
antonm@chromium.org [Fri, 28 May 2010 11:54:58 +0000 (11:54 +0000)]
Make intercepted properties retrievable only by getter to be not enumerable.

Currently if there is no query callback, V8 finds out intercepted properties'
attributes using getter: if getter returns not empty handle V8 treats
such a property as property with NONE attribues which means this property
is enumerable.

However, if there is no enumerator, this property cannot be enumerated.
Thus I think we should treat such properties as not enumerable.

Drawback of this approach is now one has to implement both query and enumerator
callbacks to implement enumerable intercepted properties.

BUG=725

Review URL: http://codereview.chromium.org/2270005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4751 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMissing file from last commit.
erik.corry@gmail.com [Fri, 28 May 2010 11:27:14 +0000 (11:27 +0000)]
Missing file from last commit.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4750 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoGet rid of LoadAndSpill on ARM since Load() knows whether it is
erik.corry@gmail.com [Fri, 28 May 2010 11:23:07 +0000 (11:23 +0000)]
Get rid of LoadAndSpill on ARM since Load() knows whether it is
in a spilled scope or not.  Also get rid of some spilled scopes
that we don't need any more.  The generators for the %_ functions,
CodeGenerator::Generate*, are now not spilled by default.  Some
of them (IsObject and related) have been converted to register
allocated operation.
Review URL: http://codereview.chromium.org/2368001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4749 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoX64: Fix issue 678. Bug in some Win64 C calls from generated code.
lrn@chromium.org [Fri, 28 May 2010 08:37:44 +0000 (08:37 +0000)]
X64: Fix issue 678. Bug in some Win64 C calls from generated code.
Win 64 C call ABI implementation requires space allocated on stack for four
argument registers, even when passing fewer arguments in registers.

Review URL: http://codereview.chromium.org/2365001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4748 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRefactor GenerateLoadInterceptor, removing LoadInterceptorCompiler class and CompileL...
whesse@chromium.org [Thu, 27 May 2010 14:14:46 +0000 (14:14 +0000)]
Refactor GenerateLoadInterceptor, removing LoadInterceptorCompiler class and CompileLoadInterceptor static function.
All platforms (x64, ia32, and ARM) changed.
Review URL: http://codereview.chromium.org/2251003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4747 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix presubmit error. Trailing whitespace.
ager@chromium.org [Thu, 27 May 2010 13:57:55 +0000 (13:57 +0000)]
Fix presubmit error. Trailing whitespace.

TBR=erik.corry
Review URL: http://codereview.chromium.org/2224007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4746 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix jump targets on ARM to merge virtual frames (really this time).
erik.corry@gmail.com [Thu, 27 May 2010 13:48:52 +0000 (13:48 +0000)]
Fix jump targets on ARM to merge virtual frames (really this time).
Use the jump targets to move the first deferred code to not flush
registers in the inlined case (KeyedLoad).
Review URL: http://codereview.chromium.org/2249002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4745 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoUpdate and improve support for ARMv7 bitfield instructions.
erik.corry@gmail.com [Thu, 27 May 2010 13:46:18 +0000 (13:46 +0000)]
Update and improve support for ARMv7 bitfield instructions.
This is a commit of http://codereview.chromium.org/2124022
for Rodolph Perfetta.  I changed the test in
test-assembler-arm.cc so it only runs if ARMv7 is supported.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4744 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCardmarking writebarrier.
vegorov@chromium.org [Thu, 27 May 2010 12:30:45 +0000 (12:30 +0000)]
Cardmarking writebarrier.

Reapply r4715 with fixes reviewed in http://codereview.chromium.org/2276002.

Review URL: http://codereview.chromium.org/2255004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4743 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRevert try-commit.
lrn@chromium.org [Thu, 27 May 2010 12:23:46 +0000 (12:23 +0000)]
Revert try-commit.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4742 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoTo be reverted.
lrn@chromium.org [Thu, 27 May 2010 12:22:23 +0000 (12:22 +0000)]
To be reverted.
Add extra checks for bad stack content in X64 RE-macro-assembler,
to check for crash that is only reproducible with the buildbot and
only in release mode.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4741 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMove assertions to not take case to check that it's safe to omit CheckPrototypes...
antonm@chromium.org [Thu, 27 May 2010 11:16:04 +0000 (11:16 +0000)]
Move assertions to not take case to check that it's safe to omit CheckPrototypes in this case.

Review URL: http://codereview.chromium.org/2279005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4740 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoX64: Many small tweaks and more usages of "load smi as int32".
lrn@chromium.org [Thu, 27 May 2010 10:25:33 +0000 (10:25 +0000)]
X64: Many small tweaks and more usages of "load smi as int32".

Review URL: http://codereview.chromium.org/2266003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4739 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAdd support for getOwnPropertyDescriptor on array indices (fixes issue 599).
ricow@chromium.org [Thu, 27 May 2010 07:43:43 +0000 (07:43 +0000)]
Add support for getOwnPropertyDescriptor on array indices (fixes issue 599).

This fix adds support for retriving a property descriptor on elements. The
new version supports both fast and slow case elements. In the fast case
we always default configurable, writable, enumerable to true (we don't have
PropertyDetails for fast elements).

A few new tests are added to get-own-property-descriptor.js, I will
add a lot more to object-define-property when I add support for indices in
Object.defineProperty.

Review URL: http://codereview.chromium.org/2278002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4738 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Default to use unaligned accesses in the simulator
sgjesse@chromium.org [Thu, 27 May 2010 07:31:10 +0000 (07:31 +0000)]
ARM: Default to use unaligned accesses in the simulator

Support building with or without unaligned accesses by using scons flag simulatorunalignedaccesses.

$ scons simulator=arm simulatorunalignedaccesses=off
$ scons simulator=arm simulatorunalignedaccesses=on

If simulatorunalignedaccesses is not specified (or specified as default) the default is to use unaligned accesses.
Review URL: http://codereview.chromium.org/2218003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4737 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoImplement OS::ReleaseStore for Solaris
sgjesse@chromium.org [Thu, 27 May 2010 06:38:53 +0000 (06:38 +0000)]
Implement OS::ReleaseStore for Solaris

Review URL: http://codereview.chromium.org/2282003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4736 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix: make string indexing work with Infinity.
vitalyr@chromium.org [Wed, 26 May 2010 16:11:30 +0000 (16:11 +0000)]
Fix: make string indexing work with Infinity.

NumberToUint32 that I was trying to use maps -0 to +0 (as desired) but
also maps +/-Infinity to +0, which made +/-Infinity a valid string
index. I fixed it by introducing a new runtime function with the right
semantics.

TEST=LayoutTests/fast/js/char-at.html,mjsunit/string-charat.js

Review URL: http://codereview.chromium.org/2223003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4734 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCustom call IC-s for String.prototype.{charAt,charCodeAt}.
vitalyr@chromium.org [Wed, 26 May 2010 14:23:19 +0000 (14:23 +0000)]
Custom call IC-s for String.prototype.{charAt,charCodeAt}.

These string methods can be composed from two basic blocks: charCodeAt
and fromCharCode, both of which have fast cases for certain types of
inputs. In this patch these two blocks are refactored to allow
generating the fast cases without having to jump around the slow
cases. In the slow cases since they can now be invoked both from
inline runtime functions and from IC stubs we either have to
save/restore state of the current frame or enter/leave a new internal
frame. This is handled by new RuntimeCallHelper interface. Its
implementation for virtual frame is based on FrameRegisterState class
extracted from DeferredCode class.

Review URL: http://codereview.chromium.org/2087009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4733 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoVarious refactorings in interceptor calling and loading.
antonm@chromium.org [Wed, 26 May 2010 14:04:37 +0000 (14:04 +0000)]
Various refactorings in interceptor calling and loading.

1) do not push receiver early---that simplifies tail call preparation
on ia32/x64 and renders special cleanup unnecessary;
2) do not do second map check if interceptor's and cached holder
are the same;
3) do not push/pop receiver if receiver and holder registers are the same
(means that receiver is interceptor's holder);
4) do batch pushes on arm;
5) minor cosmetic improvements.

Review URL: http://codereview.chromium.org/2282001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4732 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPrepare push to trunk. Now working on version 2.2.13.
kmillikin@chromium.org [Wed, 26 May 2010 12:39:50 +0000 (12:39 +0000)]
Prepare push to trunk.  Now working on version 2.2.13.

Review URL: http://codereview.chromium.org/2280001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4729 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRefactor x64 named loads to agree with ia32 implementation. Remove dead code and...
whesse@chromium.org [Wed, 26 May 2010 11:38:33 +0000 (11:38 +0000)]
Refactor x64 named loads to agree with ia32 implementation.  Remove dead code and flag is_global from x64 keyed loads.
Review URL: http://codereview.chromium.org/2121022

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4728 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoUpdate comment in the API to explicitly state that the handle returned
ager@chromium.org [Wed, 26 May 2010 11:28:14 +0000 (11:28 +0000)]
Update comment in the API to explicitly state that the handle returned
from Context::New is persistent and has to be disposed at some point
in order to allow garbage collecting the context.

Review URL: http://codereview.chromium.org/2220003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4727 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Add Ldrd/Strd to the macro assembler
sgjesse@chromium.org [Wed, 26 May 2010 11:19:39 +0000 (11:19 +0000)]
ARM: Add Ldrd/Strd to the macro assembler

The macro assembler now checks for ARMv7 support and generates instructions for Ldrd/Strd accordingly. INstructions ldrd/strd in the assembler requires ARMv7 support enabled. This removes the a check for CAN_USE_ARMV7_INSTRUCTIONS making the ARMv7 support in the simulator fully controled by the --enable-armv7 flag.
Review URL: http://codereview.chromium.org/2226003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4726 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoX64: Make smi memory operations work directly on the embedded value.
lrn@chromium.org [Wed, 26 May 2010 10:02:07 +0000 (10:02 +0000)]
X64: Make smi memory operations work directly on the embedded value.
Adds Operand-relative Operand constructor.

Review URL: http://codereview.chromium.org/2242002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4725 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMake ldrd and strd instructions take two register arguments
sgjesse@chromium.org [Wed, 26 May 2010 09:43:54 +0000 (09:43 +0000)]
Make ldrd and strd instructions take two register arguments
Review URL: http://codereview.chromium.org/2122021

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4724 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRevert r4715.
vegorov@chromium.org [Wed, 26 May 2010 09:24:44 +0000 (09:24 +0000)]
Revert r4715.

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/2274001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4723 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRefactoring of codegen-arm.cc to use the VirtualFrame API.
erik.corry@gmail.com [Wed, 26 May 2010 08:40:11 +0000 (08:40 +0000)]
Refactoring of codegen-arm.cc to use the VirtualFrame API.
This is a commit for Rodolph Perfetta.
http://codereview.chromium.org/2159002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4722 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoThe way reloc entries are visited by the ObjectVisitor is architecture
erik.corry@gmail.com [Wed, 26 May 2010 08:34:07 +0000 (08:34 +0000)]
The way reloc entries are visited by the ObjectVisitor is architecture
dependent, so we push it down to the architecture dependent files.
Currently all architectures visit in almost the same way, but this is
about to change on ARM with movw/movt.
Review URL: http://codereview.chromium.org/2218002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4721 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix issue 720 making Object.defineProperty handle existing writable flags correctly.
ricow@chromium.org [Wed, 26 May 2010 08:31:57 +0000 (08:31 +0000)]
Fix issue 720 making Object.defineProperty handle existing writable flags correctly.

The existing version will overwrite the existing writable flag with
false even in the case where no new value is given and the existing
writable flag is true.

The reason for the issue is that there is no check to see if the
provided descriptor actually has a writable attribute. This causes us
to use the default value (false) even in the case where nothing was
provided. In addition, the existing tests makes wrong assumptions (that writable is always set to false if not provided) and has been changed to follow the specification.

Review URL: http://codereview.chromium.org/2271001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4720 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix v8_base.vcproj: add dtoa.cc and fixed-dtoa.cc to dtoa project.
yurys@chromium.org [Wed, 26 May 2010 08:20:44 +0000 (08:20 +0000)]
Fix v8_base.vcproj: add dtoa.cc and fixed-dtoa.cc to dtoa project.
Review URL: http://codereview.chromium.org/2086024

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4719 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoMinor FreeBSD compile fix from Ry:
erik.corry@gmail.com [Tue, 25 May 2010 18:50:31 +0000 (18:50 +0000)]
Minor FreeBSD compile fix from Ry:
http://codereview.chromium.org/2098019/show
Should fix http://code.google.com/p/v8/issues/detail?id=716

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4718 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoComment the code for loading and calling with interceptor.
antonm@chromium.org [Tue, 25 May 2010 14:59:11 +0000 (14:59 +0000)]
Comment the code for loading and calling with interceptor.

Review URL: http://codereview.chromium.org/2124021

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4717 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoARM: Update the full compiler to handle all code
sgjesse@chromium.org [Tue, 25 May 2010 14:08:17 +0000 (14:08 +0000)]
ARM: Update the full compiler to handle all code

The full compiler is now complete on ARM as well. The syntax checker is still used to determine whether to use it for top level code or not during normal execution. When debugging is enabled all code will be compiled with the full compiler.

This change removes the temporary flag --force-full-compiler and now the flag --always-full-compiler enables the full compiler for all code on all platforms.

This also fixes building on Intel platform without debugger support (ENABLE_DEBUGGER_SUPPORT not defined) and adds full check for the full compiler for lazily compiled code.
Review URL: http://codereview.chromium.org/2163006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4716 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCardmarking writebarrier.
vegorov@chromium.org [Tue, 25 May 2010 13:15:16 +0000 (13:15 +0000)]
Cardmarking writebarrier.

- New сardmarking write barrier handles large objects and normal objects in a similar fashion (no more additional space for pointer tracking is required, no conditional branches in WB code).
- Changes to enable oldspaces iteration without maps decoding:
-- layout change for FixedArrays: length is stored as a smis (initial patch by
Kevin Millikin)
-- layout change for SharedFunctionInfo: integer fields are stored as smi on
arm, ia32 and rearranged on x64.
-- layout change for String: meaning of LSB bit is fliped (1 now means hash not
computed); on x64 padding is added.
-- layout of maps is _not_ changed. Map space is currently iterated in a special
way.

Review URL: http://codereview.chromium.org/2144006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4715 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoAllow to define accessors on objects.
antonm@chromium.org [Tue, 25 May 2010 12:14:49 +0000 (12:14 +0000)]
Allow to define accessors on objects.

Currently one can only define accessors on object templates.  This patch
allows to create accessors on the fly.

These accessors could control access to elements as well.  This element
support is somewhat rudimentary and may require future work (for example,
we probably don't want to convert index into a string.)

Review URL: http://codereview.chromium.org/2123012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4714 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRefactor the samevalue internal method and add tests for this method.
ricow@chromium.org [Tue, 25 May 2010 10:35:55 +0000 (10:35 +0000)]
Refactor the samevalue internal method and add tests for this method.

Noticing that the only difference between samevalue and strict equality is on
numbers we can simplify SameValue.

The old version did not return a correct answer if called on two strings since
StringEquals (from runtime.cc) returns an answer that is the negated value
(if treated as a boolean).

Review URL: http://codereview.chromium.org/2136024

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4713 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix GC issue
sgjesse@chromium.org [Tue, 25 May 2010 09:18:08 +0000 (09:18 +0000)]
Fix GC issue

A raw pointer was used while collecting a stack trace. This was not safe as the code collecting the stack trace allocated memory.
Review URL: http://codereview.chromium.org/2147005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4712 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix lint issue and compile error.
ager@chromium.org [Tue, 25 May 2010 06:43:13 +0000 (06:43 +0000)]
Fix lint issue and compile error.

TBR=erik.corry@gmail.com
Review URL: http://codereview.chromium.org/2157006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4711 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoLanding patch by tonyg@chromium.org:
ager@chromium.org [Tue, 25 May 2010 06:38:19 +0000 (06:38 +0000)]
Landing patch by tonyg@chromium.org:

Changing ScriptData API to serialize its internal representation to a
const char* array.

This decouples the API from the internal representation and avoids the need for
callers to serialize themselves.

As a side-effect, ScriptData::New() no longer assumes ownership of its input.
This shouldn't matter as typical usage patterns for the old API would have
required a copy prior to calling ScriptData::New().

Review URL: http://codereview.chromium.org/2118010/show

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4710 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix presubmit error (a tab in v8natives.js)
ricow@chromium.org [Tue, 25 May 2010 06:35:03 +0000 (06:35 +0000)]
Fix presubmit error (a tab in v8natives.js)
Review URL: http://codereview.chromium.org/2132025

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4709 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFixes issue 712 causing non-configurable accessors to be overwritable by using
ricow@chromium.org [Tue, 25 May 2010 06:25:27 +0000 (06:25 +0000)]
Fixes issue 712 causing non-configurable accessors to be overwritable by using
Object.defineProperty with empty property descriptor.

The issue is fixed by implementing step 5 and 6 from DefineOwnProperty in the
specification (ES5 8.12.9).

This also fixes a bug in SameValue when used on boolean values (it
would priorly return a number - not a boolean).

Review URL: http://codereview.chromium.org/2131019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4708 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix tests on ARM simulator.
mikhail.naganov@gmail.com [Sat, 22 May 2010 06:35:27 +0000 (06:35 +0000)]
Fix tests on ARM simulator.

TBR=sgjesse@chromium.org

Review URL: http://codereview.chromium.org/2124017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4707 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCPU profiler: make code events handling scalable.
mikhail.naganov@gmail.com [Sat, 22 May 2010 05:27:19 +0000 (05:27 +0000)]
CPU profiler: make code events handling scalable.

I changed the implementation of a queue between the VM and processor
thread to be unbounded and lock-free, using Herb Sutter's example from
DDJ article: http://www.ddj.com/high-performance-computing/210604448
This had brought back profiling overhead to a minimum for the page
from Chromium's issue 16184.

BUG=714

Review URL: http://codereview.chromium.org/2091019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4706 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoInclude check for execution termination into bailout check.
dimich@chromium.org [Fri, 21 May 2010 20:52:19 +0000 (20:52 +0000)]
Include check for execution termination into bailout check.

This prevents re-entry into JS during stack unwinding caused by TerminateExecution().

Review URL: http://codereview.chromium.org/2123005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4705 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoReverting r4703.
vegorov@chromium.org [Fri, 21 May 2010 17:37:47 +0000 (17:37 +0000)]
Reverting r4703.

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/2073018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4704 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoCardmarking writebarrier.
vegorov@chromium.org [Fri, 21 May 2010 15:52:38 +0000 (15:52 +0000)]
Cardmarking writebarrier.

Reapplication of r4685 (reviewed http://codereview.chromium.org/2101002) with minor modifications:

- Fix compilation problems on Win64.
- Improve heap verification pass: search for garbage pointers to new space not only in dirty regions but in all regions.

Review URL: http://codereview.chromium.org/2114015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4703 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoFix teardown order.
mikhail.naganov@gmail.com [Fri, 21 May 2010 13:46:35 +0000 (13:46 +0000)]
Fix teardown order.

As TokenEnumerator cleans up global handles posessed, it's owner --
CpuProfiler must be shut down prior to global handles.

Review URL: http://codereview.chromium.org/2072013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4702 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoRevert change 3849. On ia32, LoadFromSlot again returns its result on the top of...
whesse@chromium.org [Fri, 21 May 2010 12:58:02 +0000 (12:58 +0000)]
Revert change 3849.  On ia32, LoadFromSlot again returns its result on the top of the expression stack, not as a Result return value.
Review URL: http://codereview.chromium.org/2121014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4700 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

14 years agoPrepare push to trunk. Now working on version 2.2.12.
erik.corry@gmail.com [Fri, 21 May 2010 11:22:03 +0000 (11:22 +0000)]
Prepare push to trunk. Now working on version 2.2.12.
Review URL: http://codereview.chromium.org/2080018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@4699 ce2b1a6d-e550-0410-aec6-3dcde31c8c00