Compilation type and state allocate an unnecessary Smi on v8::Script.

Compilation type and state allocate an unnecessary Smi on v8::Script.

This patch combines them into a single field, and adjusts the accessors
and setters to take the enum type directly, rather than converting to and
from a Smi value.

R=mstarzinger@chromium.org

Committed: http://code.google.com/p/v8/source/detail?r=15940

[Mike West, 2013-07-26 16:18:01]

Hi Yang and Jakob!

This is the cleanup you suggested. I've split it out into a separate patch; once
we get this into a working state, I can rebase the patch I actually want to land
on top of it. :)

Thanks!

-mike

[Michael Starzinger, 2013-07-29 14:43:58]

LGTM with nits. I'll land this once nits are addressed.

https://codereview.chromium.org/20758002/diff/11001/src/objects-inl.h
File src/objects-inl.h (right):

https://codereview.chromium.org/20758002/diff/11001/src/objects-inl.h#newcode...
src/objects-inl.h:4490: Script::CompilationType Script::compilation_type() {
nit: Indentation is off, this should be at the top-level. Also can we add an
empty newline between the "flags" accessor and the other accessors for
readability?

https://codereview.chromium.org/20758002/diff/11001/src/objects.h
File src/objects.h (right):

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5852
src/objects.h:5852: // [compilation_type]: how the the script was compiled.
nit: Can we add a short sentence to the comment saying this is "encoded in the
flags field"? We might even consider moving it down to below the declaration of
the "flags" accessor.

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5857
src/objects.h:5857: // compiled.
Likewise.

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5903
src/objects.h:5903: static const int kCompilationTypeBit = 0;
nit: Lets add a one-liner comment "// Bit positions in the flags field." here.

[Mike West, 2013-07-29 14:52:12]

Nits addressed, thank you.

https://codereview.chromium.org/20758002/diff/11001/src/objects-inl.h
File src/objects-inl.h (right):

https://codereview.chromium.org/20758002/diff/11001/src/objects-inl.h#newcode...
src/objects-inl.h:4490: Script::CompilationType Script::compilation_type() {
On 2013/07/29 14:43:58, Michael Starzinger wrote:
> nit: Indentation is off, this should be at the top-level. Also can we add an
> empty newline between the "flags" accessor and the other accessors for
> readability?

Sure.

https://codereview.chromium.org/20758002/diff/11001/src/objects.h
File src/objects.h (right):

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5852
src/objects.h:5852: // [compilation_type]: how the the script was compiled.
On 2013/07/29 14:43:58, Michael Starzinger wrote:
> nit: Can we add a short sentence to the comment saying this is "encoded in the
> flags field"? We might even consider moving it down to below the declaration
of
> the "flags" accessor.

Makes sense. Upcoming patchset does both.

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5857
src/objects.h:5857: // compiled.
On 2013/07/29 14:43:58, Michael Starzinger wrote:
> Likewise.

Done.

https://codereview.chromium.org/20758002/diff/11001/src/objects.h#newcode5903
src/objects.h:5903: static const int kCompilationTypeBit = 0;
On 2013/07/29 14:43:58, Michael Starzinger wrote:
> nit: Lets add a one-liner comment "// Bit positions in the flags field." here.

Done.

[Michael Starzinger, 2013-07-29 17:59:24]

Committed patchset #5 manually as r15940 (presubmit successful).