cc: Move UnoccludedContentRect to Occlusion.

cc: Move UnoccludedContentRect to Occlusion.

This patch moves OcclusionTracker::UnoccludedContentRect to Occlusion.
This reduces OcclusionTracker API and makes it consistent with
IsOccluded call.

Follow-up work here will remove occlusion_tracker from these functions
where possible and replace it with const Occlusion& occlusion.

R=danakj
BUG=410932
Committed: https://crrev.com/5b88d999fd4590b5ddf1c4223f58e496856c51f0
Cr-Commit-Position: refs/heads/master@{#295597}

[vmpstr, 2014-09-17 23:19:23]

PTAL

https://codereview.chromium.org/576173003/diff/20001/cc/trees/occlusion_track...
File cc/trees/occlusion_tracker_perftest.cc (right):

https://codereview.chromium.org/576173003/diff/20001/cc/trees/occlusion_track...
cc/trees/occlusion_tracker_perftest.cc:109: Occlusion occlusion =
I'm not sure what's more useful to benchmark here. If we cache the unoccluded
content rect internally in occlusion, this would become faster. However, in a
lot of cases we only get the rect once, so caching might not be all that
helpful. Let me know what you think.

[vmpstr, 2014-09-17 23:23:47]

https://codereview.chromium.org/576173003/diff/20001/cc/trees/occlusion_track...
File cc/trees/occlusion_tracker_perftest.cc (right):

https://codereview.chromium.org/576173003/diff/20001/cc/trees/occlusion_track...
cc/trees/occlusion_tracker_perftest.cc:109: Occlusion occlusion =
On 2014/09/17 23:19:23, vmpstr wrote:
> I'm not sure what's more useful to benchmark here. If we cache the unoccluded
> content rect internally in occlusion, this would become faster. However, in a
> lot of cases we only get the rect once, so caching might not be all that
> helpful. Let me know what you think.

Ignore the caching statement, that's not going to work very well, since we use
the content rect.

[danakj, 2014-09-18 01:48:45]

LGTM

https://codereview.chromium.org/576173003/diff/40001/cc/layers/delegated_rend...
File cc/layers/delegated_renderer_layer_impl.cc (right):

https://codereview.chromium.org/576173003/diff/40001/cc/layers/delegated_rend...
cc/layers/delegated_renderer_layer_impl.cc:448:
occlusion_tracker.GetCurrentOcclusionForLayer(
Are you going to pass the Occlusion instead of the OcclusionTracker in another
CL? Then you have to touch all the layer types twice, maybe easier to just do it
now.

Then again, you already wrote this.

[vmpstr, 2014-09-18 21:45:19]

https://codereview.chromium.org/576173003/diff/40001/cc/layers/delegated_rend...
File cc/layers/delegated_renderer_layer_impl.cc (right):

https://codereview.chromium.org/576173003/diff/40001/cc/layers/delegated_rend...
cc/layers/delegated_renderer_layer_impl.cc:448:
occlusion_tracker.GetCurrentOcclusionForLayer(
On 2014/09/18 01:48:45, danakj wrote:
> Are you going to pass the Occlusion instead of the OcclusionTracker in another
> CL? Then you have to touch all the layer types twice, maybe easier to just do
it
> now.
> 
> Then again, you already wrote this.

Yeah, I wanna do it separately so if the world explodes, I know which part did
it.

[vmpstr, 2014-09-18 21:45:22]

The CQ bit was checked by vmpstr@chromium.org

[commit-bot: I haz the power, 2014-09-18 21:46:07]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/576173003/40001

[commit-bot: I haz the power, 2014-09-18 23:25:12]

Committed patchset #3 (id:40001) as 95f99b4b7424400522297c40b009fb0d70153ae0

[commit-bot: I haz the power, 2014-09-18 23:25:52]

Patchset 3 (id:??) landed as
https://crrev.com/5b88d999fd4590b5ddf1c4223f58e496856c51f0
Cr-Commit-Position: refs/heads/master@{#295597}