[Smart Lock] Add UMA metrics to measure duration of unlock.

[Smart Lock] Add UMA metrics to measure duration of unlock.

BUG=461207
TEST=none
R=tengs@chromium.org

Committed: https://crrev.com/035213964677e1441afbec26c86af9774d98d014
Cr-Commit-Position: refs/heads/master@{#323166}

[Tim Song, 2015-03-25 18:17:28]

There is already a similar metric for measuring the time to connect to the phone
after resuming from suspend (EasyUnlock.StartupTimeFromSuspend).

Do these new metrics provide more useful information?

https://codereview.chromium.org/1022233002/diff/1/chrome/browser/signin/easy_...
File chrome/browser/signin/easy_unlock_service_regular.h (right):

https://codereview.chromium.org/1022233002/diff/1/chrome/browser/signin/easy_...
chrome/browser/signin/easy_unlock_service_regular.h:44: public
chromeos::PowerManagerClient::Observer,
The base class has a similar observer in EasyUnlockService::PowerMonitor.
Perhaps you can call a virtual method from the base class when resuming from
suspend instead of adding this observer?

[Ilya Sherman, 2015-04-01 00:16:13]

On 2015/03/25 at 18:17:28, tengs wrote:
> There is already a similar metric for measuring the time to connect to the
phone after resuming from suspend (EasyUnlock.StartupTimeFromSuspend).
> 
> Do these new metrics provide more useful information?

These metrics provide a view of the user's experience, which is good to keep
track of.  We can compare to the existing metrics to see whether the time
required to connect to the phone seems to dominate the time taken to unlock.

https://codereview.chromium.org/1022233002/diff/1/chrome/browser/signin/easy_...
File chrome/browser/signin/easy_unlock_service_regular.h (right):

https://codereview.chromium.org/1022233002/diff/1/chrome/browser/signin/easy_...
chrome/browser/signin/easy_unlock_service_regular.h:44: public
chromeos::PowerManagerClient::Observer,
On 2015/03/25 18:17:28, Tim Song wrote:
> The base class has a similar observer in EasyUnlockService::PowerMonitor.
> Perhaps you can call a virtual method from the base class when resuming from
> suspend instead of adding this observer?

Done.

[Tim Song, 2015-04-01 00:17:53]

That is a good metric to have. LGTM.

[Ilya Sherman, 2015-04-01 00:25:12]

The CQ bit was checked by isherman@chromium.org

[commit-bot: I haz the power, 2015-04-01 00:25:44]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1022233002/20001

[commit-bot: I haz the power, 2015-04-01 01:48:01]

Committed patchset #2 (id:20001)

[commit-bot: I haz the power, 2015-04-01 01:49:02]

Patchset 2 (id:??) landed as
https://crrev.com/035213964677e1441afbec26c86af9774d98d014
Cr-Commit-Position: refs/heads/master@{#323166}