Reland "Turn ElementsTransitionAndStore stub into a HydrogenCodeStub".

Reland "Turn ElementsTransitionAndStore stub into a HydrogenCodeStub".

Fix the invalid array length check, replacing it with a check of
the elements pointer similar to TransitionElementsKindStub.
Refactor common code from ElementsTransitionAndStoreStub and TransitionElementsKindStub into BuildTransitionElementsKind() helper method.
Add test case for the MD5 computation that used to crash before,
and a small test case for the specific issue.

R=danno@chromium.org

Committed: https://code.google.com/p/v8/source/detail?r=15713

[Benedikt Meurer, 2013-07-17 09:04:10]

Attempt to reland the ElementsTransitionAndStoreStub rewrite. PTAL

[danno, 2013-07-17 09:07:16]

This form of the patch doesn't give us an option of "going back" without
reverting again. Is it possible to keep the old, hand written stub  so that we
can toggle using the new functionality with a boolean flag as we've discussed
before?

[Toon Verwaest, 2013-07-17 09:09:06]

As discussed, please add a specific unit test as well.

[Benedikt Meurer, 2013-07-17 09:32:32]

On 2013/07/17 09:07:16, danno wrote:
> This form of the patch doesn't give us an option of "going back" without
> reverting again. Is it possible to keep the old, hand written stub  so that we
> can toggle using the new functionality with a boolean flag as we've discussed
> before?

The hydrogen stub is now used if compiled transitions is enabled.

[danno, 2013-07-17 11:31:07]

lgtm

[Benedikt Meurer, 2013-07-17 11:35:13]

On 2013/07/17 09:09:06, Toon Verwaest wrote:
> As discussed, please add a specific unit test as well.

Done.

[Benedikt Meurer, 2013-07-17 11:37:30]

Committed patchset #4 manually as r15713 (presubmit successful).