[stubs] Introduce a dedicated FastNewObjectStub.

[stubs] Introduce a dedicated FastNewObjectStub.

Move the already existing fast case for %NewObject into a dedicated
FastNewObjectStub that we can utilize in places where we would otherwise
fallback to %NewObject immediately, which is rather expensive.

Also use FastNewObjectStub as the generic implementation of JSCreate,
which should make constructor inlining based on SharedFunctionInfo (w/o
specializing to a concrete closure) viable soon.

R=jarin@chromium.org

Committed: https://crrev.com/ba2077aac3a775523ca3fc9600fcf76f55e8d1c2
Cr-Commit-Position: refs/heads/master@{#34136}

[Benedikt Meurer, 2016-02-18 18:07:46]

Hey Jaro,

Some code moving w/ cleanup.
Please take a look.

Thanks,
Benedikt

[Benedikt Meurer, 2016-02-19 05:28:47]

Patchset #2 (id:20001) has been deleted

[Benedikt Meurer, 2016-02-19 05:28:56]

Patchset #2 (id:40001) has been deleted

[Yang, 2016-02-19 06:51:08]

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

[Yang, 2016-02-19 06:51:09]

lgtm. only reviewed arm port though.

https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.cc
File src/arm/builtins-arm.cc (left):

https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.c...
src/arm/builtins-arm.cc:552: if (FLAG_inline_new) {
do we still care about this flag?

https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h
File src/bailout-reason.h (right):

https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h#n...
src/bailout-reason.h:151: V(kOperandIsASmiAndNotAReceiver, "Operand is a smi and
not a receiver")      \
Imo we don't need to distinguish this case and the one added below.

[Benedikt Meurer, 2016-02-19 06:54:06]

https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.cc
File src/arm/builtins-arm.cc (left):

https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.c...
src/arm/builtins-arm.cc:552: if (FLAG_inline_new) {
I suppose yes, but this already handled in the MacroAssembler::Allocate, so we
don't need to do it twice.

https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h
File src/bailout-reason.h (right):

https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h#n...
src/bailout-reason.h:151: V(kOperandIsASmiAndNotAReceiver, "Operand is a smi and
not a receiver")      \
Well, I just followed the approach in the surrounding code. If we want to clean
this up, we should do all of the Assert*'s at once.

[Benedikt Meurer, 2016-02-19 06:54:22]

The CQ bit was checked by bmeurer@chromium.org

[commit-bot: I haz the power, 2016-02-19 06:54:27]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1708313002/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1708313002/100001

[Yang, 2016-02-19 06:56:50]

On 2016/02/19 06:54:06, Benedikt Meurer wrote:
> https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.cc
> File src/arm/builtins-arm.cc (left):
> 
>
https://codereview.chromium.org/1708313002/diff/100001/src/arm/builtins-arm.c...
> src/arm/builtins-arm.cc:552: if (FLAG_inline_new) {
> I suppose yes, but this already handled in the MacroAssembler::Allocate, so we
> don't need to do it twice.
> 
> https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h
> File src/bailout-reason.h (right):
> 
>
https://codereview.chromium.org/1708313002/diff/100001/src/bailout-reason.h#n...
> src/bailout-reason.h:151: V(kOperandIsASmiAndNotAReceiver, "Operand is a smi
and
> not a receiver")      \
> Well, I just followed the approach in the surrounding code. If we want to
clean
> this up, we should do all of the Assert*'s at once.

Okay fair enough :)

[commit-bot: I haz the power, 2016-02-19 07:12:24]

Committed patchset #4 (id:100001)

[commit-bot: I haz the power, 2016-02-19 07:12:59]

Description was changed from

==========
[stubs] Introduce a dedicated FastNewObjectStub.

Move the already existing fast case for %NewObject into a dedicated
FastNewObjectStub that we can utilize in places where we would otherwise
fallback to %NewObject immediately, which is rather expensive.

Also use FastNewObjectStub as the generic implementation of JSCreate,
which should make constructor inlining based on SharedFunctionInfo (w/o
specializing to a concrete closure) viable soon.

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

to

==========
[stubs] Introduce a dedicated FastNewObjectStub.

Move the already existing fast case for %NewObject into a dedicated
FastNewObjectStub that we can utilize in places where we would otherwise
fallback to %NewObject immediately, which is rather expensive.

Also use FastNewObjectStub as the generic implementation of JSCreate,
which should make constructor inlining based on SharedFunctionInfo (w/o
specializing to a concrete closure) viable soon.

R=jarin@chromium.org

Committed: https://crrev.com/ba2077aac3a775523ca3fc9600fcf76f55e8d1c2
Cr-Commit-Position: refs/heads/master@{#34136}
==========

[commit-bot: I haz the power, 2016-02-19 07:13:00]

Patchset 4 (id:??) landed as
https://crrev.com/ba2077aac3a775523ca3fc9600fcf76f55e8d1c2
Cr-Commit-Position: refs/heads/master@{#34136}