Remove ThreadHeapMode

Remove ThreadHeapMode and make per thread heap mandatory

Removes ThreadHeapMode enum and makes per thread heap mandatory.

Remove HeapTests.{GCParkingTimeout,RecursiveMutex,ObjectDeadBit} because they test threads attached to the main thread's heap.

BUG=591606
Review-Url: https://codereview.chromium.org/2625293002
Cr-Commit-Position: refs/heads/master@{#443190}
Committed: https://chromium.googlesource.com/chromium/src/+/5c3ca4cbc1798a8d5551b6636252cf701b2051ad

[keishi, 2017-01-12 02:11:23]

Description was changed from

==========
Remove ThreadHeapMode

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

to

==========
Remove ThreadHeapMode

Remove GCParkingTimeout, RecursiveMutex, ObjectDeadBit because they test threads
attached to the main thread's heap.

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

[keishi, 2017-01-12 02:14:20]

Description was changed from

==========
Remove ThreadHeapMode

Remove GCParkingTimeout, RecursiveMutex, ObjectDeadBit because they test threads
attached to the main thread's heap.

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

to

==========
Remove ThreadHeapMode and make per thread heap mandatory

Removes ThreadHeapMode enum and makes per thread heap mandatory.

Remove HeapTests.{GCParkingTimeout,RecursiveMutex,ObjectDeadBit} because they
test threads attached to the main thread's heap.

BUG=591606
==========

[keishi, 2017-01-12 02:14:28]

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

[commit-bot: I haz the power, 2017-01-12 02:15:06]

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

[keishi, 2017-01-12 02:15:50]

keishi@chromium.org changed reviewers:
+ haraken@chromium.org, oilpan-reviews@chromium.org

[keishi, 2017-01-12 02:15:50]

Removes ThreadHeapMode. Removal of things like ThreadHeap::m_threads will be
done later.

[commit-bot: I haz the power, 2017-01-12 02:24:10]

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

[commit-bot: I haz the power, 2017-01-12 02:24:11]

Dry run: Try jobs failed on following builders:
  chromeos_daisy_chromium_compile_only_ng on master.tryserver.chromium.linux
(JOB_FAILED,
http://build.chromium.org/p/tryserver.chromium.linux/builders/chromeos_daisy_...)

[haraken, 2017-01-12 04:26:33]

LGTM

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
File third_party/WebKit/Source/platform/heap/HeapTest.cpp (left):

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
third_party/WebKit/Source/platform/heap/HeapTest.cpp:5472: class DeadBitTester {

Yeah, removing this test makes sense.

Sigbjorn: I begin to think that it's not possible to hit a Dead state in the
world of the per-thread heaps. A HeapObjectHeader is marked as Dead only when a
next GC is hit before the thread gets a chance to run preSweep(). It should not
happen with the per-thread heaps, right?

[sof, 2017-01-12 08:29:29]

sigbjornf@opera.com changed reviewers:
+ sigbjornf@opera.com

[sof, 2017-01-12 08:29:30]

lgtm

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
File third_party/WebKit/Source/platform/heap/HeapTest.cpp (left):

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
third_party/WebKit/Source/platform/heap/HeapTest.cpp:5472: class DeadBitTester {
On 2017/01/12 04:26:33, haraken wrote:
> 
> Yeah, removing this test makes sense.
> 
> Sigbjorn: I begin to think that it's not possible to hit a Dead state in the
> world of the per-thread heaps. A HeapObjectHeader is marked as Dead only when
a
> next GC is hit before the thread gets a chance to run preSweep(). It should
not
> happen with the per-thread heaps, right?
> 
> 

I think that's correct also. completeSweep() will happen before the next GC goes
ahead, so "dead" object states shouldn't be observable during the next GC
marking pass.

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
File third_party/WebKit/Source/platform/heap/ThreadState.cpp (left):

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
third_party/WebKit/Source/platform/heap/ThreadState.cpp:204:
m_heap->attach(this);
It is possible to now also consider simplifying ThreadHeap's tracking of
attached threads?

[keishi, 2017-01-12 08:31:40]

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
File third_party/WebKit/Source/platform/heap/ThreadState.cpp (left):

https://codereview.chromium.org/2625293002/diff/1/third_party/WebKit/Source/p...
third_party/WebKit/Source/platform/heap/ThreadState.cpp:204:
m_heap->attach(this);
On 2017/01/12 08:29:30, sof wrote:
> It is possible to now also consider simplifying ThreadHeap's tracking of
> attached threads?

I am working on that next.

[keishi, 2017-01-12 08:31:45]

The CQ bit was checked by keishi@chromium.org

[commit-bot: I haz the power, 2017-01-12 08:31:53]

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

[haraken, 2017-01-12 08:34:38]

Also shall we announce that we have finished shipping the per-thread heap on the
intent-to-ship thread? This is a big architectural improvement.

[commit-bot: I haz the power, 2017-01-12 09:29:33]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1484209905633440, "parent_rev":
"c5138fa204cedceffaf7a7e777529a9a5d69964c", "commit_rev":
"5c3ca4cbc1798a8d5551b6636252cf701b2051ad"}

[commit-bot: I haz the power, 2017-01-12 09:30:08]

Description was changed from

==========
Remove ThreadHeapMode and make per thread heap mandatory

Removes ThreadHeapMode enum and makes per thread heap mandatory.

Remove HeapTests.{GCParkingTimeout,RecursiveMutex,ObjectDeadBit} because they
test threads attached to the main thread's heap.

BUG=591606
==========

to

==========
Remove ThreadHeapMode and make per thread heap mandatory

Removes ThreadHeapMode enum and makes per thread heap mandatory.

Remove HeapTests.{GCParkingTimeout,RecursiveMutex,ObjectDeadBit} because they
test threads attached to the main thread's heap.

BUG=591606

Review-Url: https://codereview.chromium.org/2625293002
Cr-Commit-Position: refs/heads/master@{#443190}
Committed:
https://chromium.googlesource.com/chromium/src/+/5c3ca4cbc1798a8d5551b6636252...
==========

[commit-bot: I haz the power, 2017-01-12 09:30:09]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/5c3ca4cbc1798a8d5551b6636252...