[turbofan] Add SimplifiedOperator::Allocate operator.

[turbofan] Add SimplifiedOperator::Allocate operator.

This introduces a simplified allocation operator which can be used to
model inline allocations in TurboFan. It is currently used for context
allocations, but still disabled because change lowering introduces
floating allocations outside the effect chain that interfere.

R=bmeurer@chromium.org

Committed: https://crrev.com/d6945db703c56ce821f764035fd31cb3a29e9e10
Cr-Commit-Position: refs/heads/master@{#28195}

[Benedikt Meurer, 2015-04-28 08:24:24]

General direction looks good. Not sure if we want to work on that before we fix
the AllocateHeapNumber issue.

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
File src/compiler/js-typed-lowering.cc (right):

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
src/compiler/js-typed-lowering.cc:1088: // TODO(mstarzinger): This is weird!
Make kFinish not require this.
Hm, I think Finish is fine, the problem seems to be the typing of Allocate.

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
src/compiler/js-typed-lowering.cc:1127: // TODO(mstarzinger): This is weird!
Make kFinish not require this.
Hm, I think Finish is fine, the problem seems to be the typing of Allocate.

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/simplified...
File src/compiler/simplified-lowering.cc (right):

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/simplified...
src/compiler/simplified-lowering.cc:1176: Runtime::FunctionId f =
Runtime::kInlineAllocateInTargetSpace;
Shouldn't this be Runtime::kAllocateInTargetSpace?

[Michael Starzinger, 2015-04-30 07:53:51]

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
File src/compiler/js-typed-lowering.cc (right):

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
src/compiler/js-typed-lowering.cc:1088: // TODO(mstarzinger): This is weird!
Make kFinish not require this.
On 2015/04/28 08:24:23, Benedikt Meurer wrote:
> Hm, I think Finish is fine, the problem seems to be the typing of Allocate.

Done. Rephrased the TODO.

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/js-typed-l...
src/compiler/js-typed-lowering.cc:1127: // TODO(mstarzinger): This is weird!
Make kFinish not require this.
On 2015/04/28 08:24:23, Benedikt Meurer wrote:
> Hm, I think Finish is fine, the problem seems to be the typing of Allocate.

Likewise.

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/simplified...
File src/compiler/simplified-lowering.cc (right):

https://codereview.chromium.org/1109773002/diff/20001/src/compiler/simplified...
src/compiler/simplified-lowering.cc:1176: Runtime::FunctionId f =
Runtime::kInlineAllocateInTargetSpace;
On 2015/04/28 08:24:23, Benedikt Meurer wrote:
> Shouldn't this be Runtime::kAllocateInTargetSpace?

Done. This was a "cheap and dirty" way to not require a frame-state. Cleaned it
up by adapting Linkage::NeedsFrameState instead.

[Benedikt Meurer, 2015-05-04 10:54:46]

LGTM

[Michael Starzinger, 2015-05-04 11:19:49]

The CQ bit was checked by mstarzinger@chromium.org

[Michael Starzinger, 2015-05-04 11:19:50]

The patchset sent to the CQ was uploaded after l-g-t-m from bmeurer@chromium.org
Link to the patchset: https://codereview.chromium.org/1109773002/#ps60001
(title: "Rebased.")

[commit-bot: I haz the power, 2015-05-04 11:19:56]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1109773002/60001

[commit-bot: I haz the power, 2015-05-04 12:07:17]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-05-04 12:07:28]

Patchset 4 (id:??) landed as
https://crrev.com/d6945db703c56ce821f764035fd31cb3a29e9e10
Cr-Commit-Position: refs/heads/master@{#28195}