[turbofan] Stub for typeof operator.

[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub.

BUG=chromium:608675
Committed: https://crrev.com/4602d98ea4897e1c13c302901aab92be709cd1c0
Cr-Commit-Position: refs/heads/master@{#38227}

[Franzi, 2016-07-29 20:57:09]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-07-29 20:57:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-07-29 21:10:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-07-29 21:10:56]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)
  v8_linux64_gyp_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng_trigg...)

[Franzi, 2016-08-01 14:46:28]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-01 14:46:39]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-01 14:50:09]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-01 14:50:10]

Dry run: Try jobs failed on following builders:
  v8_linux_gcc_compile_rel on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_gcc_compile_rel/bu...)
  v8_linux_nodcheck_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_nodcheck_rel_ng/bu...)

[Franzi, 2016-08-01 14:53:59]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-01 14:54:09]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-01 15:05:35]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-01 15:05:36]

Dry run: Try jobs failed on following builders:
  v8_linux64_gyp_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng/build...)
  v8_linux64_gyp_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_gyp_rel_ng_trigg...)

[Franzi, 2016-08-01 15:08:46]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-01 15:08:54]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-01 15:34:51]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-01 15:34:52]

Dry run: This issue passed the CQ dry run.

[Franzi, 2016-08-01 16:10:12]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-01 16:10:21]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Franzi, 2016-08-01 16:10:49]

Description was changed from

==========
[turbofan] Stub for typeof operator.

BUG=chromium:608675
==========

to

==========
[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub. 


BUG=chromium:608675
==========

[Franzi, 2016-08-01 16:10:49]

franzih@chromium.org changed reviewers:
+ bmeurer@chromium.org

[Franzi, 2016-08-01 16:13:30]

Hi Benedikt, 

Here's typeof as a TurboFan stub. Please take a look. 

Thanks, 

Franzi

[commit-bot: I haz the power, 2016-08-01 16:32:22]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-01 16:32:23]

Dry run: This issue passed the CQ dry run.

[Benedikt Meurer, 2016-08-01 17:27:42]

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc
File src/code-stubs.cc (right):

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1898: end(assembler);
Nit: I'd prefer to use return_result instead of end as name, as it does
effectively return whatever is in result variable.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1923: assembler->GotoIf(assembler->Word32NotEqual(
Nit: Use GotoUnless with Word32Equal instead.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1930: assembler->Branch(
Nit: Use GotoIf here as well, and continue the flat checking with the SIMD
types.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1935: &if_object, &return_function);
I think we could rearrange it a bit, so that we check for is callable here, and
if so return "function", next check for instance_type >= FIRST_JS_RECEIVER_TYPE
and if so return "object", and then do the SIMD types, and as fallback do the
SYMBOL_TYPE (which is the most unlikely case I'd guess). Or in summary, I'd
propose this order:

1. Smi -> "number"
2. HeapNumber -> "number"
3. Oddball -> load typeof
4. let mask = map->bit_field & (1 << kIsCallable | 1 << kIsUndetectable) in
  4a. if mask == 1 << kIsCallable -> "function"
  4b. if mask != 0 -> "undefined"
  4c.if instance_type >= FIRST_JS_RECEIVER_TYPE -> "object"
5. String -> "string"
6...16. Simd128Value types -> type string
17. otherwise (assert Symbol) -> "symbol"

[Franzi, 2016-08-01 18:34:10]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-01 18:34:19]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-01 18:55:57]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-01 18:55:58]

Dry run: This issue passed the CQ dry run.

[Franzi, 2016-08-01 19:00:33]

Addressed your comments. Please have another look. Thanks, 

Franzi

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc
File src/code-stubs.cc (right):

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1898: end(assembler);
On 2016/08/01 17:27:42, Benedikt Meurer wrote:
> Nit: I'd prefer to use return_result instead of end as name, as it does
> effectively return whatever is in result variable.

Done.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1923: assembler->GotoIf(assembler->Word32NotEqual(
On 2016/08/01 17:27:42, Benedikt Meurer wrote:
> Nit: Use GotoUnless with Word32Equal instead.

Done.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1930: assembler->Branch(
On 2016/08/01 17:27:42, Benedikt Meurer wrote:
> Nit: Use GotoIf here as well, and continue the flat checking with the SIMD
> types.

Done.

https://codereview.chromium.org/2195863002/diff/100001/src/code-stubs.cc#newc...
src/code-stubs.cc:1935: &if_object, &return_function);
On 2016/08/01 17:27:42, Benedikt Meurer wrote:
> I think we could rearrange it a bit, so that we check for is callable here,
and
> if so return "function", next check for instance_type >=
FIRST_JS_RECEIVER_TYPE
> and if so return "object", and then do the SIMD types, and as fallback do the
> SYMBOL_TYPE (which is the most unlikely case I'd guess). Or in summary, I'd
> propose this order:
> 
> 1. Smi -> "number"
> 2. HeapNumber -> "number"
> 3. Oddball -> load typeof
> 4. let mask = map->bit_field & (1 << kIsCallable | 1 << kIsUndetectable) in
>   4a. if mask == 1 << kIsCallable -> "function"
>   4b. if mask != 0 -> "undefined"
>   4c.if instance_type >= FIRST_JS_RECEIVER_TYPE -> "object"
> 5. String -> "string"
> 6...16. Simd128Value types -> type string
> 17. otherwise (assert Symbol) -> "symbol"

Changed the order. Makes sense to have the unlikely cases towards the end.

[Benedikt Meurer, 2016-08-02 03:18:22]

Almost there, just two nits, and one minor task: Please hook this up for the
TypeOf opcode in interpreter.cc as well, i.e. inline it into the TypeOf handler
just like we do for Inc, Dec, etc.

https://codereview.chromium.org/2195863002/diff/120001/src/code-stubs.cc
File src/code-stubs.cc (right):

https://codereview.chromium.org/2195863002/diff/120001/src/code-stubs.cc#newc...
src/code-stubs.cc:1913: {
Nit: No need to put this into a { } block, that's confusing.

https://codereview.chromium.org/2195863002/diff/120001/src/code-stubs.cc#newc...
src/code-stubs.cc:1946: result_var.Bind(assembler->HeapConstant(
Nit: Please add an Assert here that the map is the symbol_map.

[Franzi, 2016-08-02 04:51:44]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 04:51:46]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[Benedikt Meurer, 2016-08-02 04:53:36]

Thanks, LGTM.

https://codereview.chromium.org/2195863002/diff/140001/src/interpreter/interp...
File src/interpreter/interpreter.cc (right):

https://codereview.chromium.org/2195863002/diff/140001/src/interpreter/interp...
src/interpreter/interpreter.cc:1144: DoUnaryOp<TypeofStub>(assembler);
I think the untemplated DoUnaryOp is dead code now, please delete that.

[commit-bot: I haz the power, 2016-08-02 05:14:55]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-02 05:14:56]

Dry run: This issue passed the CQ dry run.

[Franzi, 2016-08-02 06:02:40]

The CQ bit was checked by franzih@chromium.org to run a CQ dry run

[commit-bot: I haz the power, 2016-08-02 06:02:47]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-02 06:24:17]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2016-08-02 06:24:18]

Dry run: This issue passed the CQ dry run.

[Franzi, 2016-08-02 07:02:17]

The CQ bit was checked by franzih@chromium.org

[Franzi, 2016-08-02 07:02:17]

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/2195863002/#ps160001
(title: "Delete now unused untemplated DoUnaryOp")

[commit-bot: I haz the power, 2016-08-02 07:02:27]

CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[commit-bot: I haz the power, 2016-08-02 07:04:33]

Description was changed from

==========
[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub. 


BUG=chromium:608675
==========

to

==========
[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub. 


BUG=chromium:608675
==========

[commit-bot: I haz the power, 2016-08-02 07:04:35]

Committed patchset #9 (id:160001)

[commit-bot: I haz the power, 2016-08-02 07:05:18]

Description was changed from

==========
[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub. 


BUG=chromium:608675
==========

to

==========
[turbofan] Stub for typeof operator.

Replace hydrogen stub for typeof with TurboFan stub.

BUG=chromium:608675

Committed: https://crrev.com/4602d98ea4897e1c13c302901aab92be709cd1c0
Cr-Commit-Position: refs/heads/master@{#38227}
==========

[commit-bot: I haz the power, 2016-08-02 07:05:20]

Patchset 9 (id:??) landed as
https://crrev.com/4602d98ea4897e1c13c302901aab92be709cd1c0
Cr-Commit-Position: refs/heads/master@{#38227}