Fix RenderWidgetHostLatencyTracker handling of multi-finger touch.

Fix RenderWidgetHostLatencyTracker handling of multi-finger touch.

This was previously broken. Other input modalities were ignored during
multi-finger touch, and we incorrectly handled some of the multi-finger
cases as well, since we didn't update whether we were in a multi-finger
touch until the event ack.

BUG=None
TEST=RenderWidgetHostLatencyTrackerTest.WheelDuringMultiFingerTouch, MultiFingerTouchIgnored

Review-Url: https://codereview.chromium.org/2801043002
Cr-Commit-Position: refs/heads/master@{#462643}
Committed: https://chromium.googlesource.com/chromium/src/+/5a2389fc0abfbdcb4045acd27d62b7d43270da43

[tdresser, 2017-04-06 13:14:52]

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

[commit-bot: I haz the power, 2017-04-06 13:15:04]

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

[tdresser, 2017-04-06 14:08:11]

tdresser@chromium.org changed reviewers:
+ mfomitchev@chromium.org

[commit-bot: I haz the power, 2017-04-06 14:39:06]

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

[commit-bot: I haz the power, 2017-04-06 14:39:07]

Dry run: This issue passed the CQ dry run.

[mfomitchev, 2017-04-06 15:26:55]

Nice! LGTM with a couple of nits.

[mfomitchev, 2017-04-06 15:27:04]

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
File content/browser/renderer_host/input/render_widget_host_latency_tracker.h
(right):

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker.h:81: //
Whether the current stream of touch events includes more than one active
Would it be worth clarifying that it's valid between OnInputEvent and
OnInputEventAck? (Not sure if we'd ever be tempted to use
active_multi_finger_gesture_ in OnGpuSwapBuffersCompleted for example).

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
File
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc
(right):

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:969:
// Some touch input histograms aren't reported for multi-finger touch.
aren't -> should not be? Or maybe just say that
Event.Latency.QueueingTime.TouchStartDefaultAllowed should be the only one that
gets reported?

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:983:
// Additional touch start will be ignored.
It't not completely ignored, right? We are still logging
Event.Latency.QueueingTime.TouchStartDefaultAllowed, just not the other
metrics..

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:1018:
// Touch start.
I am not sure if this comment is helpful (maybe I am missing something).

[tdresser, 2017-04-06 20:50:09]

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
File content/browser/renderer_host/input/render_widget_host_latency_tracker.h
(right):

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker.h:81: //
Whether the current stream of touch events includes more than one active
On 2017/04/06 15:27:03, mfomitchev wrote:
> Would it be worth clarifying that it's valid between OnInputEvent and
> OnInputEventAck? (Not sure if we'd ever be tempted to use
> active_multi_finger_gesture_ in OnGpuSwapBuffersCompleted for example).

Done.

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
File
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc
(right):

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:969:
// Some touch input histograms aren't reported for multi-finger touch.
On 2017/04/06 15:27:04, mfomitchev wrote:
> aren't -> should not be? Or maybe just say that
> Event.Latency.QueueingTime.TouchStartDefaultAllowed should be the only one
that
> gets reported?

Done.

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:983:
// Additional touch start will be ignored.
On 2017/04/06 15:27:04, mfomitchev wrote:
> It't not completely ignored, right? We are still logging
> Event.Latency.QueueingTime.TouchStartDefaultAllowed, just not the other
> metrics..

Done.

https://codereview.chromium.org/2801043002/diff/1/content/browser/renderer_ho...
content/browser/renderer_host/input/render_widget_host_latency_tracker_unittest.cc:1018:
// Touch start.
On 2017/04/06 15:27:04, mfomitchev wrote:
> I am not sure if this comment is helpful (maybe I am missing something).

Copy/paste fail, thanks.

[tdresser, 2017-04-06 20:50:14]

The CQ bit was checked by tdresser@chromium.org

[tdresser, 2017-04-06 20:50:15]

The patchset sent to the CQ was uploaded after l-g-t-m from
mfomitchev@chromium.org
Link to the patchset: https://codereview.chromium.org/2801043002/#ps20001
(title: "Fix nits.")

[commit-bot: I haz the power, 2017-04-06 20:51:14]

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, 2017-04-06 22:11:39]

CQ is committing da patch.

Bot data: {"patchset_id": 20001, "attempt_start_ts": 1491511814735100,
"parent_rev": "573263ce1c07afc7d3bcd1a92ffc18e8c6fd7f54", "commit_rev":
"5a2389fc0abfbdcb4045acd27d62b7d43270da43"}

[commit-bot: I haz the power, 2017-04-06 22:13:39]

Description was changed from

==========
Fix RenderWidgetHostLatencyTracker handling of multi-finger touch.

This was previously broken. Other input modalities were ignored during
multi-finger touch, and we incorrectly handled some of the multi-finger
cases as well, since we didn't update whether we were in a multi-finger
touch until the event ack.

BUG=None
TEST=RenderWidgetHostLatencyTrackerTest.WheelDuringMultiFingerTouch,
MultiFingerTouchIgnored
==========

to

==========
Fix RenderWidgetHostLatencyTracker handling of multi-finger touch.

This was previously broken. Other input modalities were ignored during
multi-finger touch, and we incorrectly handled some of the multi-finger
cases as well, since we didn't update whether we were in a multi-finger
touch until the event ack.

BUG=None
TEST=RenderWidgetHostLatencyTrackerTest.WheelDuringMultiFingerTouch,
MultiFingerTouchIgnored

Review-Url: https://codereview.chromium.org/2801043002
Cr-Commit-Position: refs/heads/master@{#462643}
Committed:
https://chromium.googlesource.com/chromium/src/+/5a2389fc0abfbdcb4045acd27d62...
==========

[commit-bot: I haz the power, 2017-04-06 22:13:40]

Committed patchset #2 (id:20001) as
https://chromium.googlesource.com/chromium/src/+/5a2389fc0abfbdcb4045acd27d62...