[turbofan] Ship TF for class literals.

[turbofan] Ship TF for class literals.

R=bmeurer@chromium.org
BUG=v8:4158
LOG=N

Committed: https://crrev.com/d71009a564f7bac8a8889cf66358ff6daeb4ae14
Cr-Commit-Position: refs/heads/master@{#28761}

[Benedikt Meurer, 2015-06-02 10:14:26]

lgtm

[Michael Starzinger, 2015-06-02 11:43:05]

The CQ bit was checked by mstarzinger@chromium.org

[commit-bot: I haz the power, 2015-06-02 11:43:14]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1158883006/1

[commit-bot: I haz the power, 2015-06-02 11:47:17]

Committed patchset #1 (id:1)

[commit-bot: I haz the power, 2015-06-02 11:47:38]

Patchset 1 (id:??) landed as
https://crrev.com/d71009a564f7bac8a8889cf66358ff6daeb4ae14
Cr-Commit-Position: refs/heads/master@{#28761}

[arv (Not doing code reviews), 2015-06-02 13:03:46]

One problem is that we still deopt on all super.prop and super() access so in
practice the functions created by the class literals will not use TF.

See https://codereview.chromium.org/1149133005/ for TF super.prop. It has some
failures that I need to look into but I have some more pressing work to focus on
before getting back to it.

[Michael Starzinger, 2015-06-02 13:15:13]

On 2015/06/02 13:03:46, arv wrote:
> One problem is that we still deopt on all super.prop and super() access so in
> practice the functions created by the class literals will not use TF.
> 
> See https://codereview.chromium.org/1149133005/ for TF super.prop. It has some
> failures that I need to look into but I have some more pressing work to focus
on
> before getting back to it.

Please note that this is about functions "containing a class literal" not the
functions "making up a class literal".