Change name of shifts picking the shift count from cl to sal_cl, shl_cl and shr_cl.

Add special encoding of shift by one for shr which was missing it.

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3314 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This change fixes the problem with the original version of this approach
(r3032) that may lead to a corrupted stack if we would invoke spilling during 
syncing a large SMI constant (unsafe SMIs) in the virtual frame.

The new code for storing unsafe SMI constants does not use an extra temporary 
register. This prevents the compiler from ever having to spill during a 
virutal frame sync operation.

For storing a large SMI constant we previously generated:

  mov ecx, (large_smi & 0x0000ffff)
  xor ecx, (large_smi & 0xffff0000)
  push ecx

we now generate:

  push (large_smi & 0x0000ffff)
  or   [esp], (large_smi & 0xffff0000)

Not using a temporary register avoids spilling within an nvocation 
of VirtualFrame::SyncRange.

Review URL: http://codereview.chromium.org/391079

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3313 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Condition block of do/while statements is a valid break location so it should have its own position. The block is represented by a regular Expression node so we cannot store the position in it, instead the position is stored in a separate field in DoWhileStatement AST node.

BUG=514
Review URL: http://codereview.chromium.org/385136

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3312 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/400001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3311 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=509

Review URL: http://codereview.chromium.org/398002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3310 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/399001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3309 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

of code that can't be tested on slow targets.
Review URL: http://codereview.chromium.org/385133

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3308 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/394007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3307 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=sgjesse@chromium.org
Review URL: http://codereview.chromium.org/384121

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3306 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The test currently fails on v8 trunk if the new compiler is used. The issue seems to be fixed on bleeding_edge so there is no changes to src, just test.

Cromium bug: http://crbug.com/26686
Review URL: http://codereview.chromium.org/384120

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3305 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This optimization is already done on x64 and ARM. 

Until now we used a push immediate for each local variable on IA32:

  push $undefined
  push $undefined
  ...

to initialize each local variable. This change does:

  mov eax, $undefined
  push eax
  push eax
  ...



Review URL: http://codereview.chromium.org/393009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3304 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

See http://code.google.com/p/chromium/issues/detail?id=27227

Review URL: http://codereview.chromium.org/385092

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3303 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/393010


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3302 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

available.  We use this to ensure that snapshots on MacOSX can
use SSE2 instructions.  Also clean up and assertify the
handling of things we can't do when we are generating a
snapshot.  Fix a bug in the new serialization tests where
they activated Snapshot::enable() too late after code had been
generated that assumed no snapshots.
Review URL: http://codereview.chromium.org/391051

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3301 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This change re-allows non-lazy compiled function literals for top-level compilation.

There was a problem on ARM exposed through this change which was fixed as part of r3289
in fast-codegen-arm.cc (Threading and Threading2 tests failed)

It occurred before r3289 when we allocate a local context at the beginning of a function. The
code for the stack check was split up in an unintended way (load of stack check limit into r2, 
and use of r2 in the actual stack check code).

Review URL: http://codereview.chromium.org/385070

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3300 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The test for a SMI before the write barrier tested the wrong register:
r0 = target context address
r1 = value to be written

This change fixes the SMI test and tests r1 instead of r0.

Review URL: http://codereview.chromium.org/384085

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3299 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

 
Review URL: http://codereview.chromium.org/384087

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3298 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=vitalyr
Review URL: http://codereview.chromium.org/387037

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3295 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Interceptors cannot provide a meaningful result for hidden_symbol anyway and some of them crash on empty property name.


Related Chromium issue: http://code.google.com/p/chromium/issues/detail?id=27385

Review URL: http://codereview.chromium.org/390020

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3294 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Also move a function into the macro assembler.
Fix some *& placement errors that had accumulated.
Review URL: http://codereview.chromium.org/385069

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3293 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/348019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3292 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/385068


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3291 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/385067


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3290 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Functions using "arguments" have their arguments object created on entry.
Also added support for variables rewritten into argument object property access.

Review URL: http://codereview.chromium.org/384078


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3289 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/386019

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3288 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

and if it doesn't we need to figure out why.

Review URL: http://codereview.chromium.org/385065

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3287 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

was always falling through to the runtime. Re-examined both load and
store ICs and verified they are now in sync with the 32-bit port.

Ran tests and benchmarks.

Review URL: http://codereview.chromium.org/385020

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3286 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

potentially leading to bogus FatalProcessOutOfMemory situations.  Also
fixed a few cases where callers relied on getting a NewSpace object
back (to avoid write barrier overhead) which they can't when
always_allocate is in effect.

Review URL: http://codereview.chromium.org/391018


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3285 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/391016

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3284 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

a sensible output.
Review URL: http://codereview.chromium.org/385039

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3281 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Patch by Jan de Mooij <jandemooij@gmail.com>
Review: http://codereview.chromium.org/273073

Review URL: http://codereview.chromium.org/389008


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3279 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Split a long running test into two parts. The second part still takes most of the time and is skipped on ARM in debug mode.

BUG=http://code.google.com/p/v8/issues/detail?id=500

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3278 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Set warning level to /W3 and change implicit conversions from size_t
to int. Most "fixes" are simply manifesting the implicit casts or using
a special strlen replacement that returns int.

Review URL: http://codereview.chromium.org/390004


git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3273 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

the snapshot.  Still needed: info to register profile ticks in
stubs.
Review URL: http://codereview.chromium.org/385035

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3272 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Remove variable that is currently unused.

BUG=http://code.google.com/p/v8/issues/detail?id=502
Review URL: http://codereview.chromium.org/392001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3267 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=ager@chromium.org
Review URL: http://codereview.chromium.org/388005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3266 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

All hidden properties of an object are stored in a value of a regular property with empty name. This property may confuse user if returned among regular properties. It should not be exposed directly by ObjectMirror. Should we want an access to these properties from debugger we need to implement an explicit method for that.

Current patch filters the hidden_symbol from property names returned to ObjectMirror.

See http://crbug.com/26491

Review URL: http://codereview.chromium.org/390001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3265 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Removal of string slices made this test time out on the ARM simulator. Temporaly mark this as pass or timeout.

BUG=http://code.google.com/p/v8/issues/detail?id=500
TBR=christian.plesner.hansen@gmail.com
Review URL: http://codereview.chromium.org/390002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3264 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

leads to stack corruption in 32-bit version of V8.

See http://code.google.com/p/chromium/issues/detail?id=27227 for a
reproducible case.

Since this is only an issue on 32-bit V8 I think this has got
something to do with the UnsafeSmi handling that we do on ia32.  I'm
reverting for now so we can push a fix, but we should track down the
issue and create a regression test for this.

Review URL: http://codereview.chromium.org/383005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3263 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

With slices string string type removed generating a large amount of sub-strings takes more time. Change a test to avoid timeout in debug mode.
Review URL: http://codereview.chromium.org/385006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@3260 ce2b1a6d-e550-0410-aec6-3dcde31c8c00