Reland: Buffer cross frame paint timing updates.

Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint notifications
to observers. During this buffering period, we accumulate all first paint updates
across all frames, and retain the minimum first paint time. We then dispatch the min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of order
paint timings. We have added new UMA to keep track of how often this continues to
happen, even with the 1s buffering.

BUG=728179
TBR=isherman,mattcary

Review-Url: https://codereview.chromium.org/2917693002
Cr-Commit-Position: refs/heads/master@{#475955}
Committed: https://chromium.googlesource.com/chromium/src/+/f576d151955389c1b797409da6ed19c2a5530ff5

[Bryan McQuade, 2017-05-31 15:29:14]

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

[commit-bot: I haz the power, 2017-05-31 15:29:33]

Dry run: CQ is trying da patch.

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

[Bryan McQuade, 2017-05-31 15:31:25]

Description was changed from

==========
fix enums formatting


Revert "Revert "Buffer cross frame paint timing updates.""

This reverts commit a49b11bb29ce7018da6f440703ec983009c98cc5.

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

to

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in
ResourcePrefetchPredictorBrowserTest.SubresourceFcpOrder that has since been
fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
==========

[Bryan McQuade, 2017-05-31 15:31:45]

Description was changed from

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in
ResourcePrefetchPredictorBrowserTest.SubresourceFcpOrder that has since been
fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
==========

to

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
==========

[Bryan McQuade, 2017-05-31 15:35:12]

bmcquade@chromium.org changed reviewers:
+ csharrison@chromium.org

[Bryan McQuade, 2017-05-31 15:35:27]

Description was changed from

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
==========

to

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
TBR=isherman,mattcary
==========

[Bryan McQuade, 2017-05-31 15:36:09]

csharrison, PTAL, thanks! This is a re-land with no code changes.

isherman, mattcary, I TBR'd you since this is a re-land with no code changes.

[Charlie Harrison, 2017-05-31 15:39:30]

LGTM would you add a link to the fix in the CL desc?

[Bryan McQuade, 2017-05-31 16:33:17]

On 2017/05/31 at 15:39:30, csharrison wrote:
> LGTM would you add a link to the fix in the CL desc?

Sure, now that the fix has landed, it's linked from the crbug, which is
mentioned in the CL desc.

[Bryan McQuade, 2017-05-31 16:57:51]

The CQ bit was unchecked by bmcquade@chromium.org

[Bryan McQuade, 2017-05-31 16:57:54]

The CQ bit was checked by bmcquade@chromium.org

[commit-bot: I haz the power, 2017-05-31 16:58:02]

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, 2017-05-31 17:21:11]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1496249874767300, "parent_rev":
"49bc360545589a1e22186b537d2943b30b91b267", "commit_rev":
"52c53f315d8aecd2fec1a48b1c9aa96241ae9e05"}

[commit-bot: I haz the power, 2017-05-31 17:21:24]

CQ is committing da patch.

Bot data: {"patchset_id": 1, "attempt_start_ts": 1496249874767300, "parent_rev":
"b0674d6466cc994c4e6ebf7bf63c6292495f5af3", "commit_rev":
"f576d151955389c1b797409da6ed19c2a5530ff5"}

[commit-bot: I haz the power, 2017-05-31 17:21:36]

Description was changed from

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
TBR=isherman,mattcary
==========

to

==========
Reland: Buffer cross frame paint timing updates.

This is a re-land of https://codereview.chromium.org/2901383002, which was
reverted due to a timing dependency in ResourcePrefetchPredictorBrowserTest.
SubresourceFcpOrder that has since been fixed (crbug.com/728176).

Original change description:

With the recent change to merge cross-frame paint timings, we sometimes
encounter
cases where these timings arrive out of order.

For example, we may receive a first paint notification for frame A first,
then a first paint notification for frame B, but frame B's first paint
time may have happened before frame A's first paint time.

In the current implementation, we dispatch a first paint notification to
observers
as soon as we receive a first paint notification from any frame in the page.

This means we don't end up recording the true page-level first paint time on
all pages.

This change adds a 1s buffering period before we dispatch first paint
notifications
to observers. During this buffering period, we accumulate all first paint
updates
across all frames, and retain the minimum first paint time. We then dispatch the
min
observed first paint time across this window.

This change should significantly reduce the frequency with which we log out of
order
paint timings. We have added new UMA to keep track of how often this continues
to
happen, even with the 1s buffering.

BUG=728179
TBR=isherman,mattcary

Review-Url: https://codereview.chromium.org/2917693002
Cr-Commit-Position: refs/heads/master@{#475955}
Committed:
https://chromium.googlesource.com/chromium/src/+/f576d151955389c1b797409da6ed...
==========

[commit-bot: I haz the power, 2017-05-31 17:21:37]

Committed patchset #1 (id:1) as
https://chromium.googlesource.com/chromium/src/+/f576d151955389c1b797409da6ed...