Avoid rendering using GL in case damage rect is composited by overlay

Avoid rendering using GL in case damage rect is composited by overlay

We can avoid drawing using GL if the damage rect is entirely
composited by using overlays.

BUG=370522
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/7c024b25d977484b627b63bff9f85e5b5e382b9e
Cr-Commit-Position: refs/heads/master@{#356844}

Committed: https://crrev.com/f162eed80458239a8153d968f33c2c3dc6b2766f
Cr-Commit-Position: refs/heads/master@{#359332}

[kalyank, 2015-10-26 15:39:13]

Description was changed from

==========
Avoid rendering using GL in case damage rect is composited by overlay

We can avoid drawing using GL if the damage rect is entirely
composited by using overlays.

BUG=370522
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

to

==========
Avoid rendering using GL in case damage rect is composited by overlay

We can avoid drawing using GL if the damage rect is entirely
composited by using overlays.

BUG=370522
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel
==========

[kalyank, 2015-10-26 15:39:13]

kalyan.kondapally@intel.com changed reviewers:
+ piman@chromium.org

[kalyank, 2015-10-26 15:40:56]

@piman PTAL

[kalyank, 2015-10-26 16:59:34]

On 2015/10/26 15:40:56, kalyank wrote:
> @piman PTAL

Sorry, forgot to mention this depends on
https://codereview.chromium.org/1330563004/

[piman, 2015-10-26 17:53:41]

Could you add a unit test for this new logic?

https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.cc
File cc/output/direct_renderer.cc (right):

https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.c...
cc/output/direct_renderer.cc:248: if (needs_rendering &&
frame.root_damage_rect.IsEmpty()) {
Why the test for needs_rendering here?
Why would want to draw using GL instead if the root damage was empty before
ProcessForOverlays?

[kalyank, 2015-10-27 06:09:01]

On 2015/10/26 17:53:41, piman (slow to review) wrote:
> Could you add a unit test for this new logic?

Done.

> https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.cc
> File cc/output/direct_renderer.cc (right):
> 
>
https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.c...
> cc/output/direct_renderer.cc:248: if (needs_rendering &&
> frame.root_damage_rect.IsEmpty()) {
> Why the test for needs_rendering here?
> Why would want to draw using GL instead if the root damage was empty before
> ProcessForOverlays?

[kalyank, 2015-10-27 06:09:33]

https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.cc
File cc/output/direct_renderer.cc (right):

https://codereview.chromium.org/1413763013/diff/1/cc/output/direct_renderer.c...
cc/output/direct_renderer.cc:248: if (needs_rendering &&
frame.root_damage_rect.IsEmpty()) {
On 2015/10/26 17:53:40, piman (slow to review) wrote:
> Why the test for needs_rendering here?
> Why would want to draw using GL instead if the root damage was empty before
> ProcessForOverlays?

Ya this wasn't needed, Fixed it. Added checks to make sure we don't have copy
requests for any of the RenderPass and damage rect is empty.

[piman, 2015-10-27 15:40:28]

lgtm

[kalyank, 2015-10-27 17:10:52]

kalyan.kondapally@intel.com changed reviewers:
+ ccameron@chromium.org

[kalyank, 2015-10-27 17:14:02]

kalyan.kondapally@intel.com changed reviewers:
- ccameron@chromium.org

[kalyank, 2015-10-29 15:10:58]

The CQ bit was checked by kalyan.kondapally@intel.com

[kalyank, 2015-10-29 15:10:58]

The patchset sent to the CQ was uploaded after l-g-t-m from piman@chromium.org
Link to the patchset: https://codereview.chromium.org/1413763013/#ps60001
(title: "Rebase")

[commit-bot: I haz the power, 2015-10-29 15:11:39]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1413763013/60001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1413763013/60001

[commit-bot: I haz the power, 2015-10-29 15:16:08]

Committed patchset #4 (id:60001)

[commit-bot: I haz the power, 2015-10-29 15:17:00]

Patchset 4 (id:??) landed as
https://crrev.com/7c024b25d977484b627b63bff9f85e5b5e382b9e
Cr-Commit-Position: refs/heads/master@{#356844}

[ccameron, 2015-10-30 23:26:08]

A revert of this CL (patchset #4 id:60001) has been created in
https://codereview.chromium.org/1422423003/ by ccameron@chromium.org.

The reason for reverting is: This conflicts with reverting
https://codereview.chromium.org/1330563004
Which broke YouTube on Mac
BUG=549350.

[kalyank, 2015-10-30 23:51:55]

Description was changed from

==========
Avoid rendering using GL in case damage rect is composited by overlay

We can avoid drawing using GL if the damage rect is entirely
composited by using overlays.

BUG=370522
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/7c024b25d977484b627b63bff9f85e5b5e382b9e
Cr-Commit-Position: refs/heads/master@{#356844}
==========

to

==========
Avoid rendering using GL in case damage rect is composited by overlay

We can avoid drawing using GL if the damage rect is entirely
composited by using overlays.

BUG=370522
CQ_INCLUDE_TRYBOTS=tryserver.blink:linux_blink_rel

Committed: https://crrev.com/7c024b25d977484b627b63bff9f85e5b5e382b9e
Cr-Commit-Position: refs/heads/master@{#356844}
==========

[kalyank, 2015-11-09 18:17:46]

@piman Can you PTAL to reland this.

[piman, 2015-11-09 20:02:06]

lgtm

[kalyank, 2015-11-12 15:57:53]

The CQ bit was checked by kalyan.kondapally@intel.com

[commit-bot: I haz the power, 2015-11-12 15:58:05]

CQ is trying da patch. Follow status at
 https://chromium-cq-status.appspot.com/patch-status/1413763013/80001
View timeline at
 https://chromium-cq-status.appspot.com/patch-timeline/1413763013/80001

[commit-bot: I haz the power, 2015-11-12 17:04:15]

Committed patchset #5 (id:80001)

[commit-bot: I haz the power, 2015-11-12 20:07:56]

Patchset 5 (id:??) landed as
https://crrev.com/f162eed80458239a8153d968f33c2c3dc6b2766f
Cr-Commit-Position: refs/heads/master@{#359332}