Remove HRuntimeType scheme for function types.

Remove HRuntimeType scheme for function types.

**** Replacing change this with 

https://chromiumcodereview.appspot.com/2722753002/

****


We generate the old representation for function types.
Since the HRuntimeType scheme worked by translating to the old representation, this is slightly faster.


BUG=

[sra1, 2017-02-24 23:56:02]

sra@google.com changed reviewers:
+ efortuna@google.com

[Emily Fortuna, 2017-02-25 00:26:02]

looks good with a few cleanup bits.

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/js...
File pkg/compiler/lib/src/js_backend/checked_mode_helpers.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/js...
pkg/compiler/lib/src/js_backend/checked_mode_helpers.dart:92: class
FunctionWithRepresentationCheckedModeHelper extends CheckedModeHelper {
Can you add a comment here to explain what you mean by "WithRespresentation" (or
just reword the name of this class)

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/ss...
File pkg/compiler/lib/src/ssa/nodes.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/ss...
pkg/compiler/lib/src/ssa/nodes.dart:2866: //bool get usesMethodOnType =>
is this intended to be commented out? along with the addition of the bool below?

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
File sdk/lib/_internal/js_runtime/lib/js_helper.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3372: var /* bool */ _inAssert =
false;
why don't we just say "bool _inAssert = false;" and cut the "var"? It's cleaner.
:-D

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3378: _inAssert = true;
I don't understand why this puts us in an assert?

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3555: throw 123;
debug code, methinks. Here and below.

[sra1, 2017-02-28 05:16:04]

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/js...
File pkg/compiler/lib/src/js_backend/checked_mode_helpers.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/js...
pkg/compiler/lib/src/js_backend/checked_mode_helpers.dart:92: class
FunctionWithRepresentationCheckedModeHelper extends CheckedModeHelper {
On 2017/02/25 00:26:02, Emily Fortuna wrote:
> Can you add a comment here to explain what you mean by "WithRespresentation"
(or
> just reword the name of this class)

Done.

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/ss...
File pkg/compiler/lib/src/ssa/nodes.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/pkg/compiler/lib/src/ss...
pkg/compiler/lib/src/ssa/nodes.dart:2866: //bool get usesMethodOnType =>
On 2017/02/25 00:26:02, Emily Fortuna wrote:
> is this intended to be commented out? along with the addition of the bool
below?

We no longer use this form, I'll remove it completely.

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
File sdk/lib/_internal/js_runtime/lib/js_helper.dart (right):

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3372: var /* bool */ _inAssert =
false;
On 2017/02/25 00:26:02, Emily Fortuna wrote:
> why don't we just say "bool _inAssert = false;" and cut the "var"? It's
cleaner.
> :-D

I'll write a comment to explain.

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3378: _inAssert = true;
On 2017/02/25 00:26:02, Emily Fortuna wrote:
> I don't understand why this puts us in an assert?

I'll rename the variable and explain.

https://codereview.chromium.org/2711783006/diff/20001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3555: throw 123;
On 2017/02/25 00:26:02, Emily Fortuna wrote:
> debug code, methinks. Here and below.

I'll delete this class too, maybe in a followup CL.

[sra1, 2017-02-28 06:28:03]

Description was changed from

==========
Remove HRuntimeType scheme for function types.

We generate the old representation for function types.
Since the HRuntimeType scheme worked by translating to the old representation,
this is slightly faster.


BUG=
==========

to

==========
Remove HRuntimeType scheme for function types.

** Replacing this with https://chromiumcodereview.appspot.com/2722753002/
**


We generate the old representation for function types.
Since the HRuntimeType scheme worked by translating to the old representation,
this is slightly faster.


BUG=
==========

[Emily Fortuna, 2017-02-28 16:37:44]

lgtm

https://codereview.chromium.org/2711783006/diff/60001/sdk/lib/_internal/js_ru...
File sdk/lib/_internal/js_runtime/lib/js_helper.dart (right):

https://codereview.chromium.org/2711783006/diff/60001/sdk/lib/_internal/js_ru...
sdk/lib/_internal/js_runtime/lib/js_helper.dart:3563: throw 123;
maybe just add a comment here to explain that these throws are to ensure that
this code is never reached and/or throw with a more informative error message
like "Something went wrong, should not reach here."

[sra1, 2017-02-28 17:34:18]

Description was changed from

==========
Remove HRuntimeType scheme for function types.

** Replacing this with https://chromiumcodereview.appspot.com/2722753002/
**


We generate the old representation for function types.
Since the HRuntimeType scheme worked by translating to the old representation,
this is slightly faster.


BUG=
==========

to

==========
Remove HRuntimeType scheme for function types.

**** Replacing change this with 

https://chromiumcodereview.appspot.com/2722753002/

****


We generate the old representation for function types.
Since the HRuntimeType scheme worked by translating to the old representation,
this is slightly faster.


BUG=
==========