Implement layout invalidation tracking in devtools

Implement layout invalidation tracking in devtools

This patch adds layout invalidation tracking to the invalidation
tracking experiment. Our full design doc is at:
https://docs.google.com/document/d/1BXT3_gD--YdbIYMlTzNGsyUUMBCZJ7V5qGJhMA-pGrs/edit

This patch implements the invalidation tracking logic for layout
events. All layout invalidations are recorded and when a layout
occurs they are collected and displayed to the user. Style invalidations
occurring at the same time are not shown.

This patch is a fairly straightforward extension of [1] but for
layout events. The InspectorTest.findTimelineRecord function has been
added for capturing the N-th event of a specific type and all other
inspector tests updated to use it.

[1] https://src.chromium.org/viewvc/blink?view=rev&revision=184212

BUG=410701
Committed: https://src.chromium.org/viewvc/blink?view=rev&revision=184269

[pdr., 2014-10-23 04:42:48]

pdr@chromium.org changed reviewers:
+ caseq@chromium.org, kouhei@chromium.org, yurys@chromium.org

[pdr., 2014-10-23 04:43:00]

PTAL

[kouhei (in TOK), 2014-10-23 04:48:38]

https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
File Source/devtools/front_end/timeline/TracingTimelineModel.js (right):

https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
Source/devtools/front_end/timeline/TracingTimelineModel.js:1203: didLayout:
function(layoutEvent)
I think we should share the logic here with didRecalcStyle. :)

https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
Source/devtools/front_end/timeline/TracingTimelineModel.js:1271:
this._lastLayoutEventIndex = 0;
Perhaps make this a map? this._lastEventIndex[StyleRecalc/Layout]

[pdr., 2014-10-23 05:09:44]

https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
File Source/devtools/front_end/timeline/TracingTimelineModel.js (right):

https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
Source/devtools/front_end/timeline/TracingTimelineModel.js:1203: didLayout:
function(layoutEvent)
On 2014/10/23 at 04:48:38, kouhei wrote:
> I think we should share the logic here with didRecalcStyle. :)

I thought the same but it feels a little overengineered to extract out the
common functionality for just these two cases. For example, the frameIds are
accessed differently and we don't have a map from {RecalculateStyles,Layout}
events to {StyleRecalc,Layout}InvalidationTracking. Would you mind if we wait on
this a bit to see how the (very similar) didUpdateLayer function works?

[kouhei (in TOK), 2014-10-23 05:11:41]

On 2014/10/23 05:09:44, pdr wrote:
>
https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
> File Source/devtools/front_end/timeline/TracingTimelineModel.js (right):
> 
>
https://codereview.chromium.org/671913005/diff/1/Source/devtools/front_end/ti...
> Source/devtools/front_end/timeline/TracingTimelineModel.js:1203: didLayout:
> function(layoutEvent)
> On 2014/10/23 at 04:48:38, kouhei wrote:
> > I think we should share the logic here with didRecalcStyle. :)
> 
> I thought the same but it feels a little overengineered to extract out the
> common functionality for just these two cases. For example, the frameIds are
> accessed differently and we don't have a map from {RecalculateStyles,Layout}
> events to {StyleRecalc,Layout}InvalidationTracking. Would you mind if we wait
on
> this a bit to see how the (very similar) didUpdateLayer function works?

lgtm. Let's wait until we see how ScheduleStyleInvalidation->StyleInvaldation
and StyleInvalidation->StyleRecalc tracking turns out.

[pdr., 2014-10-23 06:40:33]

@yurys or @caseq, could you please review the devtools side?

[pfeldman, 2014-10-23 11:52:11]

lgtm

[pdr., 2014-10-23 13:16:44]

The CQ bit was checked by pdr@chromium.org

[commit-bot: I haz the power, 2014-10-23 13:17:15]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/671913005/1

[commit-bot: I haz the power, 2014-10-23 14:24:54]

Committed patchset #1 (id:1) as 184269