Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018
Committed: https://crrev.com/267cd99070e74fc51e38d278f5a557275064c21d
Cr-Commit-Position: refs/heads/master@{#380418}

[Bryan McQuade, 2016-03-08 14:40:54]

Description was changed from

==========
Revert https://codereview.chromium.org/1230513002

BUG=
==========

to

==========
This is a revert of https://codereview.chromium.org/1230513002



BUG=
==========

[Bryan McQuade, 2016-03-08 14:44:02]

Description was changed from

==========
This is a revert of https://codereview.chromium.org/1230513002



BUG=
==========

to

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=
==========

[Bryan McQuade, 2016-03-08 14:45:26]

Description was changed from

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=
==========

to

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018
==========

[Bryan McQuade, 2016-03-08 18:23:31]

Description was changed from

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018
==========

to

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018
==========

[Bryan McQuade, 2016-03-08 18:26:53]

bmcquade@chromium.org changed reviewers:
+ asvitkine@chromium.org, csharrison@chromium.org, jochen@chromium.org

[Bryan McQuade, 2016-03-08 18:27:29]

csharrison PTAL for general metrics change
jochen PTAL for chrome/ OWNERS review
asvitkine PTAL for histograms.xml

[Alexei Svitkine (slow), 2016-03-08 20:16:26]

lgtm

[jochen (gone - plz use gerrit), 2016-03-09 15:07:26]

https://codereview.chromium.org/1774873002/diff/20001/chrome/renderer/page_lo...
File chrome/renderer/page_load_histograms.h (left):

https://codereview.chromium.org/1774873002/diff/20001/chrome/renderer/page_lo...
chrome/renderer/page_load_histograms.h:23: ~PageLoadHistograms() override;
why?

[Bryan McQuade, 2016-03-09 20:29:49]

https://codereview.chromium.org/1774873002/diff/20001/chrome/renderer/page_lo...
File chrome/renderer/page_load_histograms.h (left):

https://codereview.chromium.org/1774873002/diff/20001/chrome/renderer/page_lo...
chrome/renderer/page_load_histograms.h:23: ~PageLoadHistograms() override;
On 2016/03/09 at 15:07:26, jochen wrote:
> why?

The destructor was added as part of the CL
https://codereview.chromium.org/1230513002 that's being reverted.

IIRC an explicit empty destructor was needed because of the WeakPtrFactory,
which was also introduced in that change.

[jochen (gone - plz use gerrit), 2016-03-09 20:38:31]

i'd rather keep the virtual dtor

[Bryan McQuade, 2016-03-09 20:46:55]

On 2016/03/09 at 20:38:31, jochen wrote:
> i'd rather keep the virtual dtor

ok. done.

[jochen (gone - plz use gerrit), 2016-03-10 15:44:29]

lgtm

[Bryan McQuade, 2016-03-10 16:04:08]

The CQ bit was checked by bmcquade@chromium.org

[Bryan McQuade, 2016-03-10 16:04:08]

The patchset sent to the CQ was uploaded after l-g-t-m from
asvitkine@chromium.org
Link to the patchset: https://codereview.chromium.org/1774873002/#ps40001
(title: "jochen comment")

[commit-bot: I haz the power, 2016-03-10 16:04:29]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1774873002/40001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1774873002/40001

[commit-bot: I haz the power, 2016-03-10 16:33:25]

Committed patchset #3 (id:40001)

[commit-bot: I haz the power, 2016-03-10 16:34:53]

Description was changed from

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018
==========

to

==========
Remove incorrect PLT.PT.NavigationStartToFirstLayout histogram

This is a revert of https://codereview.chromium.org/1230513002

The logic for this histogram incorrectly assumes that a RenderViewObserver's
lifetime is tied to a single page load. This is incorrect: a RVO is tied to the
lifetime of a RenderView, which can live for many page loads. Thus, the logic to
compute and emit this histogram is incorrect. We've since added new page load
histograms under PageLoad.*, and there is a
PageLoad.Timing2.NavigationToFirstLayout which is correctly emitted once per
page load. Thus, we are deprecating PLT.PT.NavigationStartToFirstLayout in favor
of PageLoad.Timing2.NavigationToFirstLayout.

BUG=593018

Committed: https://crrev.com/267cd99070e74fc51e38d278f5a557275064c21d
Cr-Commit-Position: refs/heads/master@{#380418}
==========

[commit-bot: I haz the power, 2016-03-10 16:34:54]

Patchset 3 (id:??) landed as
https://crrev.com/267cd99070e74fc51e38d278f5a557275064c21d
Cr-Commit-Position: refs/heads/master@{#380418}