Change TaskQueueImpl to use a Deque with an inline capacity of 8

Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

I'm hoping in normal operation the default_tq doesn't get more
than a handful of tasks (a quick trace of handouts on a high end
pc suggests the queue depth is <= 3) which would let us avoid
dynamic allocations entirely. 

BUG=710245
Review-Url: https://codereview.chromium.org/2860813004
Cr-Commit-Position: refs/heads/master@{#469302}
Committed: https://chromium.googlesource.com/chromium/src/+/05864d5cb320715302b6d39f88728ff0a8776378

[alex clarke (OOO till 29th), 2017-05-04 08:08:42]

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

[commit-bot: I haz the power, 2017-05-04 08:08:54]

Dry run: CQ is trying da patch.

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

[alex clarke (OOO till 29th), 2017-05-04 08:09:02]

Description was changed from

==========
Change taskqueueimpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

BUG=710245
==========

to

==========
Change taskqueueimpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

BUG=710245
==========

[alex clarke (OOO till 29th), 2017-05-04 08:09:17]

Description was changed from

==========
Change taskqueueimpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

BUG=710245
==========

to

==========
Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

BUG=710245
==========

[commit-bot: I haz the power, 2017-05-04 09:35:10]

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

[commit-bot: I haz the power, 2017-05-04 09:35:10]

Dry run: This issue passed the CQ dry run.

[alex clarke (OOO till 29th), 2017-05-04 09:50:07]

alexclarke@chromium.org changed reviewers:
+ skyostil@chromium.org

[alex clarke (OOO till 29th), 2017-05-04 09:50:08]

PTAL

[Sami, 2017-05-04 09:59:46]

lgtm, thanks!

[alex clarke (OOO till 29th), 2017-05-04 10:02:39]

Description was changed from

==========
Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

BUG=710245
==========

to

==========
Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

I'm hoping in normal operation the default_tq doesn't get more
than a handful of tasks (a quick trace of handouts on a high end
pc suggests the queue depth is <= 3) which would let us avoid
dynamic allocations entirely. 

BUG=710245
==========

[alex clarke (OOO till 29th), 2017-05-04 10:02:44]

The CQ bit was checked by alexclarke@chromium.org

[commit-bot: I haz the power, 2017-05-04 10:02:48]

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, 2017-05-04 10:07:44]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1493892164519900, "parent_rev":
"df93a8860b4379bb27d05bb0368189cf8e0f3a2b", "commit_rev":
"05864d5cb320715302b6d39f88728ff0a8776378"}

[commit-bot: I haz the power, 2017-05-04 10:07:54]

Description was changed from

==========
Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

I'm hoping in normal operation the default_tq doesn't get more
than a handful of tasks (a quick trace of handouts on a high end
pc suggests the queue depth is <= 3) which would let us avoid
dynamic allocations entirely. 

BUG=710245
==========

to

==========
Change TaskQueueImpl to use a Deque with an inline capacity of 8

A speculative change that might help mitigate problems with
WTF::Deque::expandCapacity which sometimes takes over a second
due to heavy load on PartitionAlloc.

I'm hoping in normal operation the default_tq doesn't get more
than a handful of tasks (a quick trace of handouts on a high end
pc suggests the queue depth is <= 3) which would let us avoid
dynamic allocations entirely. 

BUG=710245

Review-Url: https://codereview.chromium.org/2860813004
Cr-Commit-Position: refs/heads/master@{#469302}
Committed:
https://chromium.googlesource.com/chromium/src/+/05864d5cb320715302b6d39f8872...
==========

[commit-bot: I haz the power, 2017-05-04 10:07:56]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/05864d5cb320715302b6d39f8872...

[haraken, 2017-05-04 14:49:00]

LGTM, thanks for the fix!