I created a heap snapshot in Chromium, and then started comparing
it side-by-side with representations of objects provided by
the debugger, fixing discrepancies.

Review URL: http://codereview.chromium.org/3590029

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5612 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3776001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5609 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

In the FullCodeGenerator, we compile the true subexpression of a
conditional (ternary) expression in the inherited context of the
entire expression.  This is correct for effect and value contexts, but
not for test contexts where the context includes a possible
fall-through label.

Review URL: http://codereview.chromium.org/3621013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5607 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The existing cache was build to support changes to fast-dtoa. This complicated the cache. Since the cached numbers are now in a .cc file they can be shared with other parts of v8.

Review URL: http://codereview.chromium.org/3608011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5606 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Reapply r5603 with additional fix: use OS::StrNCpy instead of posix strncpy.

BUG=
TEST=

Review URL: http://codereview.chromium.org/3557010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5605 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

errors.

TBR=floitschV8@gmail.com
Review URL: http://codereview.chromium.org/3582017

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5604 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

If there are few non-zero digits and the 10^exponent fits into a double then
we can compute the result using 1 (or 2) double operations.

BUG=
TEST=

Review URL: http://codereview.chromium.org/3584015

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5603 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3601018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5602 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=
TEST=

Review URL: http://codereview.chromium.org/3584018

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5601 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

This is the first part of removing calls to gay's strtod. Future CLs will
get rid of calls to gay_strtod entirely.

BUG=
TEST=

Review URL: http://codereview.chromium.org/3563009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5600 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3571021

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5596 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3605008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5595 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3564011

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5594 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Added a missing SetCC in the ARM CompareStub.

BUG=none
TEST=none

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5593 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Code clean-up in src/top.cc This patch addresses late review comments for http://code.google.com/p/v8/source/detail?r=5586
Review URL: http://codereview.chromium.org/3596005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5592 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The problem is other array may have holes, for example
when fixed array comes from JSArray (in case of named interceptor).

If that would prove to be a performance problem, we could
pass an additional argument into UnionOfKeys to hold actual length.

Review URL: http://codereview.chromium.org/3595013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5591 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3526008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5590 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=ager@chromium.org

Review URL: http://codereview.chromium.org/3547012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5589 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3613009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5588 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=mnaganov
Review URL: http://codereview.chromium.org/3525014

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5587 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3602013

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5586 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

TBR=kmillikin@chromium.org

Review URL: http://codereview.chromium.org/3609008

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5585 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3585010

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5584 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Change more functions used by the Compiler class to have a uniform
interface: they get passed as argument an input/output pointer to a
CompilationInfo that they mutate if they succeed, and they return a
flag telling whether they succeeded.

Also, remove some unnecessary timers.

Review URL: http://codereview.chromium.org/3561012

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5583 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Also, add functions registration to CPU profiler in case
when closure code wasn't compiled yet.

Review URL: http://codereview.chromium.org/3557007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5582 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The plan is to use the CompilationInfo class to communicate inputs and
outputs to compilation pipeline phases, which each return a boolean
success/failure flag.

The intent is to make it easier to compose small pieces of the
pipeline without having to grow a custom function each time, each
taking a half dozen arguments.

This change modifies the very front end (the parser).

Review URL: http://codereview.chromium.org/3586006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5581 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3538009

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5579 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

1. parser.h depends on ast.h, but only gets it included implicitly
   everywhere parser.h is included (except for tests).  Include ast.h in
   parser.h.

2. Regular expression tests test the free functions that have just been
   moved into class Parser.

TBR=fschneider@chromium.org

Review URL: http://codereview.chromium.org/3602007

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5577 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3518005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5576 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

It was silently assumed that inobject_properties value is not too large.
Recent introduction of inobject slack tracking made the assumption false
and debug tests with no snapshot failed.

Review URL: http://codereview.chromium.org/3584004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5575 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3532006

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5574 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3530004

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5572 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The current parameters (initial slack = 6, kGenerousAllocationsCount = 16)
caused 15-20% performance regression on 3d-raytrace on x64 
(visible both in standalone shell and in the web harness). This is very
likely some kind of an alignment  problem (it goes away completely with --prof).
A similar problem can be observed on ia32 with initial slack=2.

Increased initial slack to 8 and decreased constructions count to 8. 
This made the regression go away (and did not harm any of the other tests).

Review URL: http://codereview.chromium.org/3595002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5571 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

during snapshot size optimization.

Sorry, now I figured out that the diff implementation itself was also
incorrect.  Reachable nodes must be filtered from the beginning,
otherwise, an object that is already disconnected, but not discarded
yet, will not appear as a deleted (thankfully, this bug for some
reason had appeared on the x64 port.)

BUG=868
TEST=HeapSnapshotRootPreservedAfterSorting

Review URL: http://codereview.chromium.org/3531005

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5570 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3537003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5569 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5568 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

The bug wasn't revealed because the test was disabled.

BUG=848
TEST=HeapSnapshotsDiff,HeapSnapshotRootPreservedAfterSorting

Review URL: http://codereview.chromium.org/3572003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5567 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3574002

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5566 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

BUG=868

Review URL: http://codereview.chromium.org/3605001

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5565 ce2b1a6d-e550-0410-aec6-3dcde31c8c00

Review URL: http://codereview.chromium.org/3585003

git-svn-id: http://v8.googlecode.com/svn/branches/bleeding_edge@5562 ce2b1a6d-e550-0410-aec6-3dcde31c8c00