[heap] Improve size profiling for ArrayBuffer tracking

[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on this,
we can now do Scavenges if the amount of memory retained from new space is too large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org

Committed: https://crrev.com/28e13bd6a75c9467dae43043e7b741a1387d5252
Cr-Commit-Position: refs/heads/master@{#38731}

[Michael Lippautz, 2016-08-04 11:42:30]

Patchset #2 (id:20001) has been deleted

[Michael Lippautz, 2016-08-04 11:44:26]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

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

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
==========

[Michael Lippautz, 2016-08-04 11:46:41]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
R=jochen@chromium.org, hpayer@chromium.org
==========

[Michael Lippautz, 2016-08-04 11:46:42]

mlippautz@chromium.org changed reviewers:
+ hpayer@chromium.org, jochen@chromium.org

[Michael Lippautz, 2016-08-04 11:46:54]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
R=jochen@chromium.org, hpayer@chromium.org
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

[Michael Lippautz, 2016-08-04 11:50:03]

PTAL

We can then base our decisions on what to do in case of too much external memory
on these counters (in case it's dominated by array buffers). 

If this approach turns out to be too slow and regress performance, we could have
a retained counter on LocalArrayBufferTracker and iterate *all* pages of a space
when we need a global picture.

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
File src/heap/array-buffer-tracker.cc (right):

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
src/heap/array-buffer-tracker.cc:151: retained_from_new_space_.Increment(
General: When (and if) https://codereview.chromium.org/2215693002/ lands, we can
write this much nicer, staying completely in size_t land.

[Michael Lippautz, 2016-08-04 18:22:56]

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

[commit-bot: I haz the power, 2016-08-04 18:23:04]

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, 2016-08-04 18:49:13]

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

[commit-bot: I haz the power, 2016-08-04 18:49:15]

Dry run: This issue passed the CQ dry run.

[jochen (gone - plz use gerrit), 2016-08-05 11:47:34]

lgtm

[Hannes Payer (out of office), 2016-08-05 12:14:14]

How are array buffer heavy workloads effected by this change? Can you check
beforehand?

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
File src/heap/array-buffer-tracker.cc (right):

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
src/heap/array-buffer-tracker.cc:151: retained_from_new_space_.Increment(
On 2016/08/04 11:50:02, Michael Lippautz wrote:
> General: When (and if) https://codereview.chromium.org/2215693002/ lands, we
can
> write this much nicer, staying completely in size_t land.

Nice, please do that.

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h
File src/heap/heap.h (right):

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h#newcode...
src/heap/heap.h:1418: ArrayBufferTracker* array_buffer_tracker() { return
array_buffer_tracker_; }
Move this one to the ArrayBuffer tracking section.

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h#newcode...
src/heap/heap.h:2272: ArrayBufferTracker* array_buffer_tracker_;
We usually add comments to all variables in heap.h

[Michael Lippautz, 2016-08-08 08:34:46]

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
File src/heap/array-buffer-tracker.cc (right):

https://codereview.chromium.org/2210263002/diff/40001/src/heap/array-buffer-t...
src/heap/array-buffer-tracker.cc:151: retained_from_new_space_.Increment(
On 2016/08/05 12:14:13, Hannes Payer wrote:
> On 2016/08/04 11:50:02, Michael Lippautz wrote:
> > General: When (and if) https://codereview.chromium.org/2215693002/ lands, we
> can
> > write this much nicer, staying completely in size_t land.
> 
> Nice, please do that.

Done.

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h
File src/heap/heap.h (right):

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h#newcode...
src/heap/heap.h:1418: ArrayBufferTracker* array_buffer_tracker() { return
array_buffer_tracker_; }
On 2016/08/05 12:14:13, Hannes Payer wrote:
> Move this one to the ArrayBuffer tracking section.

Done.

https://codereview.chromium.org/2210263002/diff/40001/src/heap/heap.h#newcode...
src/heap/heap.h:2272: ArrayBufferTracker* array_buffer_tracker_;
On 2016/08/05 12:14:14, Hannes Payer wrote:
> We usually add comments to all variables in heap.h

That's not really true, but done ;)

[Michael Lippautz, 2016-08-08 08:37:09]

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

[commit-bot: I haz the power, 2016-08-08 08:37:16]

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, 2016-08-08 08:40:23]

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

[commit-bot: I haz the power, 2016-08-08 08:40:24]

Dry run: Try jobs failed on following builders:
  v8_linux64_avx2_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux64_avx2_rel_ng/buil...)
  v8_linux_arm64_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_linux_arm64_rel_ng/build...)
  v8_mac_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_mac_rel_ng/builds/6504)

[Michael Lippautz, 2016-08-08 08:41:04]

Patchset #3 (id:60001) has been deleted

[Michael Lippautz, 2016-08-08 08:41:29]

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

[commit-bot: I haz the power, 2016-08-08 08:41:33]

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, 2016-08-08 09:05:18]

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

[commit-bot: I haz the power, 2016-08-08 09:05:19]

Dry run: This issue passed the CQ dry run.

[Hannes Payer (out of office), 2016-08-08 10:45:42]

LGTM if performance is fine

[Michael Lippautz, 2016-08-08 12:37:25]

Performance is within ~3% on the considered perf.bindings benchmarks which are
micro benchmarks only allocating JSArrayBuffers. A followup CL will make use of
the information gathered here.

[Michael Lippautz, 2016-08-08 12:37:30]

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

[commit-bot: I haz the power, 2016-08-08 12:37:41]

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, 2016-08-08 13:06:39]

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

[commit-bot: I haz the power, 2016-08-08 13:06:40]

Dry run: This issue passed the CQ dry run.

[Michael Lippautz, 2016-08-08 13:08:52]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retaind from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

[Michael Lippautz, 2016-08-08 13:09:00]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retaind from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retained from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

[Michael Lippautz, 2016-08-09 08:05:14]

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

[commit-bot: I haz the power, 2016-08-09 08:05:25]

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, 2016-08-09 09:37:10]

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

[commit-bot: I haz the power, 2016-08-09 09:37:11]

Dry run: Try jobs failed on following builders:
  v8_win_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_rel_ng/builds/12126)
  v8_win_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_rel_ng_triggered/bui...)

[Michael Lippautz, 2016-08-18 20:09:01]

The CQ bit was checked by mlippautz@chromium.org

[Michael Lippautz, 2016-08-18 20:09:01]

The patchset sent to the CQ was uploaded after l-g-t-m from jochen@chromium.org,
hpayer@chromium.org
Link to the patchset: https://codereview.chromium.org/2210263002/#ps140001
(title: "rebase")

[commit-bot: I haz the power, 2016-08-18 20:09:06]

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, 2016-08-18 20:45:25]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retained from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retained from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

[commit-bot: I haz the power, 2016-08-18 20:45:26]

Committed patchset #6 (id:140001)

[commit-bot: I haz the power, 2016-08-18 20:45:54]

Description was changed from

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this, 
we can now do Scavenges if the amount of memory retained from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org
==========

to

==========
[heap] Improve size profiling for ArrayBuffer tracking

Eagerly account for retained sizes during ArrayBuffer tracking. Following up on
this,
we can now do Scavenges if the amount of memory retained from new space is too
large.

BUG=chromium:621829
R=jochen@chromium.org,hpayer@chromium.org

Committed: https://crrev.com/28e13bd6a75c9467dae43043e7b741a1387d5252
Cr-Commit-Position: refs/heads/master@{#38731}
==========

[commit-bot: I haz the power, 2016-08-18 20:45:55]

Patchset 6 (id:??) landed as
https://crrev.com/28e13bd6a75c9467dae43043e7b741a1387d5252
Cr-Commit-Position: refs/heads/master@{#38731}

[Michael Lippautz, 2016-08-19 08:16:54]

A revert of this CL (patchset #6 id:140001) has been created in
https://codereview.chromium.org/2261513003/ by mlippautz@chromium.org.

The reason for reverting is: Tanks octane.