Introduce instance type for transition arrays.

Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and collect all transition array in a list for post-processing in ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO

Committed: https://crrev.com/026095a3c7932573e1810b8064ec3008ed696601
Cr-Commit-Position: refs/heads/master@{#32396}

[ulan, 2015-11-27 12:04:36]

Description was changed from

==========
Introduce instance type for transition arrays.

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

to

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
==========

[ulan, 2015-11-27 12:04:36]

ulan@chromium.org changed reviewers:
+ jkummerow@chromium.org, mlippautz@chromium.org

[ulan, 2015-11-27 12:07:04]

Michael, ptal at GC parts.

Jakob, ptal at transitions* and object*

This CL doesn't have any changes except for new instance type.

The actual change in marking visitor and post-processing will be done in the
next CL (see TODOs in the CL).

[Michael Lippautz, 2015-11-27 14:27:28]

lgtm w/ comments

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.cc
File src/heap/mark-compact.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.c...
src/heap/mark-compact.cc:2542: TransitionArray* array =
reinterpret_cast<TransitionArray*>(obj);
Let's use TransitionArray::cast()

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.c...
src/heap/mark-compact.cc:2555: TransitionArray* array =
reinterpret_cast<TransitionArray*>(obj);
Let's use TransitionArray::cast()

https://codereview.chromium.org/1480873003/diff/40001/src/heap/objects-visiti...
File src/heap/objects-visiting-inl.h (right):

https://codereview.chromium.org/1480873003/diff/40001/src/heap/objects-visiti...
src/heap/objects-visiting-inl.h:353: // not already in the lest.
nit: ... list.

https://codereview.chromium.org/1480873003/diff/40001/src/objects-debug.cc
File src/objects-debug.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/objects-debug.cc#ne...
src/objects-debug.cc:417: }
Maybe verify that next either points to undefined or another transition array?

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.cc
File src/transitions.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.cc#newc...
src/transitions.cc:396: array->set(kPrototypeTransitionsIndex, Smi::FromInt(0));
Do we need to properly initialize the next field?

[ulan, 2015-11-27 14:43:45]

Thanks for review!

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.cc
File src/heap/mark-compact.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.c...
src/heap/mark-compact.cc:2542: TransitionArray* array =
reinterpret_cast<TransitionArray*>(obj);
On 2015/11/27 14:27:28, Michael Lippautz wrote:
> Let's use TransitionArray::cast()

Done.

https://codereview.chromium.org/1480873003/diff/40001/src/heap/mark-compact.c...
src/heap/mark-compact.cc:2555: TransitionArray* array =
reinterpret_cast<TransitionArray*>(obj);
On 2015/11/27 14:27:28, Michael Lippautz wrote:
> Let's use TransitionArray::cast()

Done.

https://codereview.chromium.org/1480873003/diff/40001/src/heap/objects-visiti...
File src/heap/objects-visiting-inl.h (right):

https://codereview.chromium.org/1480873003/diff/40001/src/heap/objects-visiti...
src/heap/objects-visiting-inl.h:353: // not already in the lest.
On 2015/11/27 14:27:28, Michael Lippautz wrote:
> nit: ... list.

Done.

https://codereview.chromium.org/1480873003/diff/40001/src/objects-debug.cc
File src/objects-debug.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/objects-debug.cc#ne...
src/objects-debug.cc:417: }
On 2015/11/27 14:27:28, Michael Lippautz wrote:
> Maybe verify that next either points to undefined or another transition array?

Done.

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.cc
File src/transitions.cc (right):

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.cc#newc...
src/transitions.cc:396: array->set(kPrototypeTransitionsIndex, Smi::FromInt(0));
On 2015/11/27 14:27:28, Michael Lippautz wrote:
> Do we need to properly initialize the next field?

Done.

NewTransitionArray() initializes with undefined. But let's explicitly set it to
undefined here too to be robust.

[Jakob Kummerow, 2015-11-27 14:54:08]

object* and transitions* LGTM.

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.h
File src/transitions.h (left):

https://codereview.chromium.org/1480873003/diff/40001/src/transitions.h#oldco...
src/transitions.h:168: // Casting.
Whaaaaat?! With this comment gone, how am I supposed to know that "cast()" is
for casting? In fact, please extend the comment to say "Returns the input
object, casted to TransitionArray".

Just kidding. It's Friday.

[ulan, 2015-11-27 15:05:54]

On 2015/11/27 14:54:08, Jakob wrote:
> object* and transitions* LGTM.
> 
> https://codereview.chromium.org/1480873003/diff/40001/src/transitions.h
> File src/transitions.h (left):
> 
>
https://codereview.chromium.org/1480873003/diff/40001/src/transitions.h#oldco...
> src/transitions.h:168: // Casting.
> Whaaaaat?! With this comment gone, how am I supposed to know that "cast()" is
> for casting? In fact, please extend the comment to say "Returns the input
> object, casted to TransitionArray".
> 
> Just kidding. It's Friday.

Thank you :) I will land after getting green stress GC try-bots.

[ulan, 2015-11-27 15:09:09]

Description was changed from

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
==========

to

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO
==========

[ulan, 2015-11-30 10:13:08]

The CQ bit was checked by ulan@chromium.org

[ulan, 2015-11-30 10:13:08]

The patchset sent to the CQ was uploaded after l-g-t-m from
mlippautz@chromium.org, jkummerow@chromium.org
Link to the patchset: https://codereview.chromium.org/1480873003/#ps100001
(title: "Fix zapping")

[commit-bot: I haz the power, 2015-11-30 10:13:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1480873003/100001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1480873003/100001

[commit-bot: I haz the power, 2015-11-30 10:35:26]

Description was changed from

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO
==========

to

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO
==========

[commit-bot: I haz the power, 2015-11-30 10:35:27]

Committed patchset #6 (id:100001)

[commit-bot: I haz the power, 2015-11-30 10:36:04]

Description was changed from

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO
==========

to

==========
Introduce instance type for transition arrays.

The motivation is to allow specialized marking visitor for transition arrays and
collect all transition array in a list for post-processing in
ClearNonLiveReferences.

BUG=chromium:554488
LOG=NO

Committed: https://crrev.com/026095a3c7932573e1810b8064ec3008ed696601
Cr-Commit-Position: refs/heads/master@{#32396}
==========

[commit-bot: I haz the power, 2015-11-30 10:36:05]

Patchset 6 (id:??) landed as
https://crrev.com/026095a3c7932573e1810b8064ec3008ed696601
Cr-Commit-Position: refs/heads/master@{#32396}

[Hannes Payer (out of office), 2015-11-30 13:03:34]

A revert of this CL (patchset #6 id:100001) has been created in
https://codereview.chromium.org/1483003002/ by hpayer@chromium.org.

The reason for reverting is: Broken canary. Trying to find out root cause..