[heap] Register end of black areas to support faster filtering of invalid slots.

[heap] Register end of black areas to support faster filtering of invalid slots.

BUG=chromium:630386
Committed: https://crrev.com/7ad2de17033dac77ec9f7e67eafc83e2f9d41b44
Cr-Commit-Position: refs/heads/master@{#38581}

[Hannes Payer (out of office), 2016-08-10 15:06:43]

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

[commit-bot: I haz the power, 2016-08-10 15:06:46]

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, 2016-08-10 15:09:36]

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

[commit-bot: I haz the power, 2016-08-10 15:09:36]

Dry run: Try jobs failed on following builders:
  v8_presubmit on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_presubmit/builds/21542)

[Michael Lippautz, 2016-08-10 16:05:41]

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

[Michael Lippautz, 2016-08-10 16:05:42]

Looking good. Waiting for tests. 

Maybe you also want to turn on the flag once for a dry run?

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h
File src/heap/spaces.h (right):

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode598
src/heap/spaces.h:598: delete black_area_end_marker_map_;
nit: we usually also add
  black_area_end_marker_map_ = nullptr;

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode690
src/heap/spaces.h:690: base::HashMap* black_area_end_marker_map_;
How about std::unordered_map?

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode1523
src/heap/spaces.h:1523: Address original_top() { return original_top_; }
nit: Let's add the SLOW_DCHECK (like below) so that we immediately see that this
is supposed to be tagged or untagged. (it probably should be untagged, like
top())

[Hannes Payer (out of office), 2016-08-10 17:39:13]

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

[commit-bot: I haz the power, 2016-08-10 17:39:16]

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, 2016-08-10 18:13:32]

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

[commit-bot: I haz the power, 2016-08-10 18:13:33]

Dry run: This issue passed the CQ dry run.

[Hannes Payer (out of office), 2016-08-11 10:51:56]

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h
File src/heap/spaces.h (right):

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode598
src/heap/spaces.h:598: delete black_area_end_marker_map_;
On 2016/08/10 16:05:42, Michael Lippautz wrote:
> nit: we usually also add
>   black_area_end_marker_map_ = nullptr;

Done.

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode690
src/heap/spaces.h:690: base::HashMap* black_area_end_marker_map_;
On 2016/08/10 16:05:42, Michael Lippautz wrote:
> How about std::unordered_map?

Using std::unordered_set which is sufficient.

https://codereview.chromium.org/2236543002/diff/1/src/heap/spaces.h#newcode1523
src/heap/spaces.h:1523: Address original_top() { return original_top_; }
On 2016/08/10 16:05:41, Michael Lippautz wrote:
> nit: Let's add the SLOW_DCHECK (like below) so that we immediately see that
this
> is supposed to be tagged or untagged. (it probably should be untagged, like
> top())

Done.

[Hannes Payer (out of office), 2016-08-11 10:52:38]

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

[commit-bot: I haz the power, 2016-08-11 10:52:46]

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

[Michael Lippautz, 2016-08-11 10:58:48]

lgtm

https://codereview.chromium.org/2236543002/diff/40001/src/heap/spaces.h
File src/heap/spaces.h (right):

https://codereview.chromium.org/2236543002/diff/40001/src/heap/spaces.h#newco...
src/heap/spaces.h:617: black_area_end_marker_map_->insert(address);
nit: Check that you don't insert duplicates.
e.g.
  auto it = ....;
  DCHECK(it->second);

http://en.cppreference.com/w/cpp/container/unordered_set/insert

[Hannes Payer (out of office), 2016-08-11 11:11:37]

The CQ bit was checked by hpayer@chromium.org

[Hannes Payer (out of office), 2016-08-11 11:11:37]

The patchset sent to the CQ was uploaded after l-g-t-m from
mlippautz@chromium.org
Link to the patchset: https://codereview.chromium.org/2236543002/#ps60001
(title: "fix nit")

[commit-bot: I haz the power, 2016-08-11 11:11:46]

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, 2016-08-11 11:55:26]

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

[commit-bot: I haz the power, 2016-08-11 11:55:27]

Try jobs failed on following builders:
  v8_win_nosnap_shared_rel_ng on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_nosnap_shared_rel_ng...)
  v8_win_nosnap_shared_rel_ng_triggered on master.tryserver.v8 (JOB_FAILED,
http://build.chromium.org/p/tryserver.v8/builders/v8_win_nosnap_shared_rel_ng...)

[Hannes Payer (out of office), 2016-08-11 13:12:37]

The CQ bit was checked by hpayer@chromium.org

[commit-bot: I haz the power, 2016-08-11 13:12:48]

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, 2016-08-11 13:34:22]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2016-08-11 13:34:37]

Description was changed from

==========
[heap] Register end of black areas to support faster filtering of invalid slots.

BUG=chromium:630386
==========

to

==========
[heap] Register end of black areas to support faster filtering of invalid slots.

BUG=chromium:630386

Committed: https://crrev.com/7ad2de17033dac77ec9f7e67eafc83e2f9d41b44
Cr-Commit-Position: refs/heads/master@{#38581}
==========

[commit-bot: I haz the power, 2016-08-11 13:34:38]

Patchset 4 (id:??) landed as
https://crrev.com/7ad2de17033dac77ec9f7e67eafc83e2f9d41b44
Cr-Commit-Position: refs/heads/master@{#38581}