input: Immediately start timer for synthetic event dispatch.

input: Immediately start timer for synthetic event dispatch.

The timer to send synthesized input events during telemetry tests is
currently started at begin-frame + fixed-offset. The reason for doing
this is to minimize the noise in the results. However, main thread
already does rAF alignment for input event dispatch, which likely means
that we can change the browser to always immediately start the timer, and
still not affect the noise. So do that in this CL, which simplifies the
code too.

BUG=725949, 722921
Review-Url: https://codereview.chromium.org/2910513006
Cr-Commit-Position: refs/heads/master@{#475692}
Committed: https://chromium.googlesource.com/chromium/src/+/e437543b5a4c01773a489e26659a9b05b7d7c365

[sadrul, 2017-05-26 18:08:35]

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

[commit-bot: I haz the power, 2017-05-26 18:09:02]

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, 2017-05-26 19:27:21]

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

[commit-bot: I haz the power, 2017-05-26 19:27:22]

Dry run: This issue passed the CQ dry run.

[sadrul, 2017-05-29 16:19:24]

Description was changed from

==========
[exp] input: Immediately start timer for synthetic event dispatch.

...

BUG=725949, 722921
==========

to

==========
input: Immediately start timer for synthetic event dispatch.

The timer to send synthesized input events during telemetry tests is
currently started at begin-frame + fixed-offset. The reason for doing
this is to minimize the noise in the results. However, main thread
already does rAF alignment for input event dispatch, which likely means
that we can change the browser to always immediately start the timer, and
still not affect the noise. So do that in this CL, which simplifies the
code too.

BUG=725949, 722921
==========

[sadrul, 2017-05-29 16:22:48]

sadrul@chromium.org changed reviewers:
+ dtapuska@chromium.org, tdresser@chromium.org

[sadrul, 2017-05-29 16:22:48]

As I mentioned in the bug, I am curious to see what the effect this can have.
The results for smoothness seem to indicate the latency metrics do not change
much, but I couldn't get to the results for thread_times or system_health. Is
this something we could land and see how it affects things?

[tdresser, 2017-05-29 17:18:51]

LGTM. Sending an email to perf-sheriffs@chromium.org when it lands would be
nice.

[sadrul, 2017-05-30 20:32:11]

The CQ bit was checked by sadrul@chromium.org

[commit-bot: I haz the power, 2017-05-30 20:32:44]

CQ is trying da patch.

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

[sadrul, 2017-05-30 20:36:54]

On 2017/05/29 17:18:51, tdresser wrote:
> LGTM. Sending an email to mailto:perf-sheriffs@chromium.org when it lands
would be
> nice.

I sent a mail, I have no idea if it reached there (looks like only members can
view/participate).

[commit-bot: I haz the power, 2017-05-30 23:39:36]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1496176331224360, "parent_rev":
"92cb438681b00aa996bdc0bac4187de2dac1ad68", "commit_rev":
"e437543b5a4c01773a489e26659a9b05b7d7c365"}

[commit-bot: I haz the power, 2017-05-30 23:39:44]

Description was changed from

==========
input: Immediately start timer for synthetic event dispatch.

The timer to send synthesized input events during telemetry tests is
currently started at begin-frame + fixed-offset. The reason for doing
this is to minimize the noise in the results. However, main thread
already does rAF alignment for input event dispatch, which likely means
that we can change the browser to always immediately start the timer, and
still not affect the noise. So do that in this CL, which simplifies the
code too.

BUG=725949, 722921
==========

to

==========
input: Immediately start timer for synthetic event dispatch.

The timer to send synthesized input events during telemetry tests is
currently started at begin-frame + fixed-offset. The reason for doing
this is to minimize the noise in the results. However, main thread
already does rAF alignment for input event dispatch, which likely means
that we can change the browser to always immediately start the timer, and
still not affect the noise. So do that in this CL, which simplifies the
code too.

BUG=725949, 722921

Review-Url: https://codereview.chromium.org/2910513006
Cr-Commit-Position: refs/heads/master@{#475692}
Committed:
https://chromium.googlesource.com/chromium/src/+/e437543b5a4c01773a489e26659a...
==========

[commit-bot: I haz the power, 2017-05-30 23:39:46]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/e437543b5a4c01773a489e26659a...