[turbofan] Don't try to inline our builtins.

[turbofan] Don't try to inline our builtins.

R=mstarzinger@chromium.org
BUG=v8:4493
LOG=n

Committed: https://crrev.com/095e6a4f84f6da32aef81ee105eab75ee024d338
Cr-Commit-Position: refs/heads/master@{#31944}

[Michael Starzinger, 2015-11-11 13:42:37]

LGTM. Disable all the inlinings!

[Benedikt Meurer, 2015-11-11 13:53:18]

The CQ bit was checked by bmeurer@chromium.org

[commit-bot: I haz the power, 2015-11-11 13:53:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1432223002/1
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1432223002/1

[commit-bot: I haz the power, 2015-11-11 14:27:50]

Committed patchset #1 (id:1)

[adamk, 2015-11-11 18:35:54]

adamk@chromium.org changed reviewers:
+ adamk@chromium.org

[adamk, 2015-11-11 18:35:55]

Can I ask why? Last time this came up I was under the impression that the
inability to inline builtins had to do with (1) weirdness in Crankshaft and (2)
the builtins global (which I thought was gone?).

[Benedikt Meurer (Google), 2015-11-12 02:42:35]

On 2015/11/11 18:35:55, adamk wrote:
> Can I ask why? Last time this came up I was under the impression that the
> inability to inline builtins had to do with (1) weirdness in Crankshaft and
(2)
> the builtins global (which I thought was gone?).

It was responsible for the 10x slowdown of some on some benchmarks after
enabling full, transitive inlining. Same reason as for CS, inlining doesn't work
well with some of the funky %_ intrinsics we use in our builtins. In this case
it was mostly due to %_Arguments and %_ArgumentsLength. Once we got rid of those
weird intrinsics we can enable this again.

[commit-bot: I haz the power, 2015-11-14 23:20:06]

Patchset 1 (id:??) landed as
https://crrev.com/095e6a4f84f6da32aef81ee105eab75ee024d338
Cr-Commit-Position: refs/heads/master@{#31944}