Refactoring PropertyCallbackInfo & FunctionCallbackInfo, step 1.

Refactoring PropertyCallbackInfo & FunctionCallbackInfo, step 1.

The goal is to unify PropertyCallbackInfo and FunctionCallbackInfo so that they
contain the same fields.

The field order will be:
holder
isolate
return value default value
return value
data
this

This step 1 reorders the PropertyCallbackInfo fields.

BUG=
R=dcarney@chromium.org, mstarzinger@chromium.org

Committed: http://code.google.com/p/v8/source/detail?r=16673

Committed: http://code.google.com/p/v8/source/detail?r=16688

[dcarney, 2013-09-10 12:21:08]

lgtm

[marja, 2013-09-10 14:29:28]

mstarzinger, could you review this?

[Michael Starzinger, 2013-09-11 09:55:19]

Looking good, just one comment to address.

https://codereview.chromium.org/23620036/diff/5001/src/ia32/stub-cache-ia32.cc
File src/ia32/stub-cache-ia32.cc (right):

https://codereview.chromium.org/23620036/diff/5001/src/ia32/stub-cache-ia32.c...
src/ia32/stub-cache-ia32.cc:403:
STATIC_ASSERT(PropertyCallbackArguments::kIsolateIndex == -4);
See comments in stub-cache.cc about these constants.

https://codereview.chromium.org/23620036/diff/5001/src/stub-cache.cc
File src/stub-cache.cc (right):

https://codereview.chromium.org/23620036/diff/5001/src/stub-cache.cc#newcode1283
src/stub-cache.cc:1283: if (PCA::kThisIndex < PCA::kReturnValueOffset)
Instead of overlaying the indices used for the arguments object here with PCA,
we should just use a separate set of constant that describes the layout.
Decouples those two things and avoids this magic "minus 2" computation. I would
put the index constant into stub-cache.h either at the top-level or into the
StubCompiler class.

[marja, 2013-09-11 16:35:31]

Thanks for having a look!

https://codereview.chromium.org/23620036/diff/5001/src/ia32/stub-cache-ia32.cc
File src/ia32/stub-cache-ia32.cc (right):

https://codereview.chromium.org/23620036/diff/5001/src/ia32/stub-cache-ia32.c...
src/ia32/stub-cache-ia32.cc:403:
STATIC_ASSERT(PropertyCallbackArguments::kIsolateIndex == -4);
On 2013/09/11 09:55:19, Michael Starzinger wrote:
> See comments in stub-cache.cc about these constants.

Done.

https://codereview.chromium.org/23620036/diff/5001/src/stub-cache.cc
File src/stub-cache.cc (right):

https://codereview.chromium.org/23620036/diff/5001/src/stub-cache.cc#newcode1283
src/stub-cache.cc:1283: if (PCA::kThisIndex < PCA::kReturnValueOffset)
On 2013/09/11 09:55:19, Michael Starzinger wrote:
> Instead of overlaying the indices used for the arguments object here with PCA,
> we should just use a separate set of constant that describes the layout.
> Decouples those two things and avoids this magic "minus 2" computation. I
would
> put the index constant into stub-cache.h either at the top-level or into the
> StubCompiler class.

Done.

[Michael Starzinger, 2013-09-11 16:52:26]

LGTM.

https://codereview.chromium.org/23620036/diff/10001/src/stub-cache.h
File src/stub-cache.h (right):

https://codereview.chromium.org/23620036/diff/10001/src/stub-cache.h#newcode411
src/stub-cache.h:411: static const int kInterceptorArgsNameIndex = 0;
nit: Maybe even add a short one-line comment above this block explaining what
the constants describe.

[marja, 2013-09-12 08:12:51]

Thanks for review!

https://codereview.chromium.org/23620036/diff/10001/src/stub-cache.h
File src/stub-cache.h (right):

https://codereview.chromium.org/23620036/diff/10001/src/stub-cache.h#newcode411
src/stub-cache.h:411: static const int kInterceptorArgsNameIndex = 0;
On 2013/09/11 16:52:26, Michael Starzinger wrote:
> nit: Maybe even add a short one-line comment above this block explaining what
> the constants describe.

Done.

[Michael Starzinger, 2013-09-12 09:09:56]

Committed patchset #6 manually as r16673 (presubmit successful).

[Michael Starzinger, 2013-09-12 13:42:17]

LGTM on the new scratch register.

[Michael Starzinger, 2013-09-12 14:32:30]

Committed patchset #7 manually as r16688 (presubmit successful).