[turbofan] Remove phase ordering problem in JSToNumber lowering.

[turbofan] Remove phase ordering problem in JSToNumber lowering.

Previously we had to run SimplifiedLowering, ChangeLowering and
JSGenericLowering independently and exactly in this order to
achieve great performance for the common case of JSToNumber (i.e.
input is already a Smi or a HeapNumber). This phase ordering
problem already causes trouble with not being able to run the
generic lowering phase earlier, but also blocks proper plain
primitive ToNumber optimizations. So this properly integrates
JSToNumber into the truncation analysis and optimizes according
to the truncation (either Word32 or Float64).

R=jarin@chromium.org

Committed: https://crrev.com/95f0d4ed39dfa2f09e216f07b3e7e9aacc95a3f4
Cr-Commit-Position: refs/heads/master@{#35643}

[Benedikt Meurer, 2016-04-18 19:05:23]

Hey Jaro,

Here's the general fix for JSToNumber, removing the need to run these lowerings
in a particular order.
Please take a look.

Thanks,
Benedikt

[Benedikt Meurer, 2016-04-19 04:15:10]

Patchset #1 (id:1) has been deleted

[Jarin, 2016-04-20 05:38:49]

lgtm

https://codereview.chromium.org/1901803002/diff/40001/src/compiler/simplified...
File src/compiler/simplified-operator.h (right):

https://codereview.chromium.org/1901803002/diff/40001/src/compiler/simplified...
src/compiler/simplified-operator.h:162: const Operator*
ChangeTaggedSignedToWord32();
Why not ChangeSmiToWord32?

[Benedikt Meurer, 2016-04-20 05:40:27]

The CQ bit was checked by bmeurer@chromium.org

[Benedikt Meurer, 2016-04-20 05:40:28]

The patchset sent to the CQ was uploaded after l-g-t-m from jarin@chromium.org
Link to the patchset: https://codereview.chromium.org/1901803002/#ps60001
(title: "Address comment")

[commit-bot: I haz the power, 2016-04-20 05:40:32]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1901803002/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1901803002/60001

[commit-bot: I haz the power, 2016-04-20 06:05:11]

Committed patchset #3 (id:60001)

[nodir1, 2016-04-22 18:43:40]

Description was changed from

==========
[turbofan] Remove phase ordering problem in JSToNumber lowering.

Previously we had to run SimplifiedLowering, ChangeLowering and
JSGenericLowering independently and exactly in this order to
achieve great performance for the common case of JSToNumber (i.e.
input is already a Smi or a HeapNumber). This phase ordering
problem already causes trouble with not being able to run the
generic lowering phase earlier, but also blocks proper plain
primitive ToNumber optimizations. So this properly integrates
JSToNumber into the truncation analysis and optimizes according
to the truncation (either Word32 or Float64).

R=jarin@chromium.org
==========

to

==========
[turbofan] Remove phase ordering problem in JSToNumber lowering.

Previously we had to run SimplifiedLowering, ChangeLowering and
JSGenericLowering independently and exactly in this order to
achieve great performance for the common case of JSToNumber (i.e.
input is already a Smi or a HeapNumber). This phase ordering
problem already causes trouble with not being able to run the
generic lowering phase earlier, but also blocks proper plain
primitive ToNumber optimizations. So this properly integrates
JSToNumber into the truncation analysis and optimizes according
to the truncation (either Word32 or Float64).

R=jarin@chromium.org

Committed: https://crrev.com/95f0d4ed39dfa2f09e216f07b3e7e9aacc95a3f4
Cr-Commit-Position: refs/heads/master@{#35643}
==========

[nodir1, 2016-04-22 18:43:41]

Patchset 3 (id:??) landed as
https://crrev.com/95f0d4ed39dfa2f09e216f07b3e7e9aacc95a3f4
Cr-Commit-Position: refs/heads/master@{#35643}