M57: service worker: Add UMA for navigation preload response timing.

M57: service worker: Add UMA for navigation preload response timing.

This should help understand how nav preload is doing in the wild:
whether it or sw startup is finishing first, and the timings for
concurrent startup and preload, and how long we end up waiting for one
after the other has finished.

Added are:
- ServiceWorker.NavigationPreload.ConcurrentTime: the time
when both startup and preload are happening concurrently
- ServiceWorker.NavigationPreload.FinishedBeforeStartWorker: indicates
which one finished first
- ServiceWorker.NavigationPreload.NavPreloadAfterSWStart: the
time spent waiting for nav preload to finish, after sw start finished
- ServiceWorker.NavigationPreload.SWStartAfterNavPreload: the reverse
- ServiceWorker.NavigationPreload.ResponseTime: the time it
took for the nav preload response to arrive

In addition, there are suffixed histograms of the above to break down
the timings in the two winner scenarios, and furthmore to see the
breakdowns in the interesting situation of when a SW needed to be
started up.

BUG=661090
Review-Url: https://codereview.chromium.org/2652603002
Cr-Commit-Position: refs/heads/master@{#445965}
(cherry picked from commit 402d868859229cf42674996b82f3468922afc628)

Review-Url: https://codereview.chromium.org/2660423002 .
Cr-Commit-Position: refs/branch-heads/2987@{#201}
Cr-Branched-From: ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}
Committed: https://chromium.googlesource.com/chromium/src/+/7b811195204dbfaa7b2a3028611f9fa7a678b5c3

[falken, 2017-01-30 23:36:08]

Description was changed from

==========
M57: service worker: Add UMA for navigation preload response timing.

This should help understand how nav preload is doing in the wild:
whether it or sw startup is finishing first, and the timings for
concurrent startup and preload, and how long we end up waiting for one
after the other has finished.

Added are:
- ServiceWorker.NavigationPreload.ConcurrentTime: the time
when both startup and preload are happening concurrently
- ServiceWorker.NavigationPreload.FinishedBeforeStartWorker: indicates
which one finished first
- ServiceWorker.NavigationPreload.NavPreloadAfterSWStart: the
time spent waiting for nav preload to finish, after sw start finished
- ServiceWorker.NavigationPreload.SWStartAfterNavPreload: the reverse
- ServiceWorker.NavigationPreload.ResponseTime: the time it
took for the nav preload response to arrive

In addition, there are suffixed histograms of the above to break down
the timings in the two winner scenarios, and furthmore to see the
breakdowns in the interesting situation of when a SW needed to be
started up.

BUG=661090

Review-Url: https://codereview.chromium.org/2652603002
Cr-Commit-Position: refs/heads/master@{#445965}
(cherry picked from commit 402d868859229cf42674996b82f3468922afc628)
==========

to

==========
M57: service worker: Add UMA for navigation preload response timing.

This should help understand how nav preload is doing in the wild:
whether it or sw startup is finishing first, and the timings for
concurrent startup and preload, and how long we end up waiting for one
after the other has finished.

Added are:
- ServiceWorker.NavigationPreload.ConcurrentTime: the time
when both startup and preload are happening concurrently
- ServiceWorker.NavigationPreload.FinishedBeforeStartWorker: indicates
which one finished first
- ServiceWorker.NavigationPreload.NavPreloadAfterSWStart: the
time spent waiting for nav preload to finish, after sw start finished
- ServiceWorker.NavigationPreload.SWStartAfterNavPreload: the reverse
- ServiceWorker.NavigationPreload.ResponseTime: the time it
took for the nav preload response to arrive

In addition, there are suffixed histograms of the above to break down
the timings in the two winner scenarios, and furthmore to see the
breakdowns in the interesting situation of when a SW needed to be
started up.

BUG=661090

Review-Url: https://codereview.chromium.org/2652603002
Cr-Commit-Position: refs/heads/master@{#445965}
(cherry picked from commit 402d868859229cf42674996b82f3468922afc628)

Review-Url: https://codereview.chromium.org/2660423002 .
Cr-Commit-Position: refs/branch-heads/2987@{#201}
Cr-Branched-From:
ad51088c0e8776e8dcd963dbe752c4035ba6dab6-refs/heads/master@{#444943}
Committed:
https://chromium.googlesource.com/chromium/src/+/7b811195204dbfaa7b2a3028611f...
==========

[falken, 2017-01-30 23:36:11]

Committed patchset #1 (id:1) manually as
7b811195204dbfaa7b2a3028611f9fa7a678b5c3.