Switch RenderWidgetHostViewAndroid to use Screen

Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

Note need to add storage for physical size in Display, since there are
rounding errors if the physical size is computed from float scale and
integer dip size.

BUG=625089
Review-Url: https://codereview.chromium.org/2617623002
Cr-Commit-Position: refs/heads/master@{#442634}
Committed: https://chromium.googlesource.com/chromium/src/+/fa8bcd12c55c0997520ff969de9e70ab6d0a59dc

[boliu, 2017-01-05 01:41:39]

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

[commit-bot: I haz the power, 2017-01-05 01:42:09]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[boliu, 2017-01-05 03:00:41]

The CQ bit was unchecked by boliu@chromium.org

[boliu, 2017-01-05 20:12:37]

boliu@chromium.org changed reviewers:
+ aelias@chromium.org

[boliu, 2017-01-05 20:12:39]

ptal

[boliu, 2017-01-05 20:12:44]

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

[commit-bot: I haz the power, 2017-01-05 20:13:06]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[boliu, 2017-01-05 20:14:16]

The CQ bit was unchecked by boliu@chromium.org

[boliu, 2017-01-06 01:22:02]

interesting, failures on n5x looks real. All hitting DCHECKs like
[FATAL:mapped_memory.h(142)] Check failed: multiple %
FencedAllocator::kAllocAlignment == 0. where kAllocAlignment is 16

Not sure how that happens, maybe switching to physical screen size (ie including
the top and bottom bar) somehow means this is no longer a multiple of 16? seems
sketchy.. gonna add some logging

[boliu, 2017-01-06 01:30:54]

Description was changed from

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

BUG=625089
==========

to

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

BUG=625089
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
==========

[boliu, 2017-01-06 03:37:30]

On 2017/01/06 01:22:02, boliu wrote:
> interesting, failures on n5x looks real. All hitting DCHECKs like
> [FATAL:mapped_memory.h(142)] Check failed: multiple %
> FencedAllocator::kAllocAlignment == 0. where kAllocAlignment is 16
> 
> Not sure how that happens, maybe switching to physical screen size (ie
including
> the top and bottom bar) somehow means this is no longer a multiple of 16?
seems
> sketchy.. gonna add some logging

size 1081x1921 <- rounding errors :/

[boliu, 2017-01-06 04:48:57]

Description was changed from

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

BUG=625089
CQ_INCLUDE_TRYBOTS=master.tryserver.chromium.linux:linux_optional_gpu_tests_rel;master.tryserver.chromium.mac:mac_optional_gpu_tests_rel;master.tryserver.chromium.win:win_optional_gpu_tests_rel
==========

to

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

BUG=625089
==========

[boliu, 2017-01-06 04:49:10]

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

[commit-bot: I haz the power, 2017-01-06 04:49:20]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[boliu, 2017-01-06 06:07:29]

Description was changed from

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

BUG=625089
==========

to

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

Note need to add storage for physical size in Display, since there are
rounding errors if the physical size is computed from float scale and
integer dip size.

BUG=625089
==========

[boliu, 2017-01-06 06:08:44]

boliu@chromium.org changed reviewers:
+ oshima@chromium.org

[boliu, 2017-01-06 06:08:45]

+oshima for ui/display

ptal, fixed the rounding error by storing the physical size separately in
Display

[commit-bot: I haz the power, 2017-01-06 09:13:10]

The CQ bit was unchecked by commit-bot@chromium.org

[commit-bot: I haz the power, 2017-01-06 09:13:11]

Dry run: This issue passed the CQ dry run.

[oshima, 2017-01-06 19:31:23]

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc
File ui/display/display.cc (right):

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc#...
ui/display/display.cc:156: size_in_pixels_ = bounds_in_pixel.size();
Would you mind adding ifdef android for now?

I do not expect i'll cause a real issue, but it's probably better to keep it
separated.

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc#...
ui/display/display.cc:175: return size_in_pixels_;
can you add this

// TODO(oshima): this should always use size_in_pixels_

?

[boliu, 2017-01-06 20:26:21]

ptal

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc
File ui/display/display.cc (right):

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc#...
ui/display/display.cc:156: size_in_pixels_ = bounds_in_pixel.size();
On 2017/01/06 19:31:23, oshima wrote:
> Would you mind adding ifdef android for now?
> 
> I do not expect i'll cause a real issue, but it's probably better to keep it
> separated.
> 
> 

Done. Also marked the setter android-only to avoid accidents

https://codereview.chromium.org/2617623002/diff/100001/ui/display/display.cc#...
ui/display/display.cc:175: return size_in_pixels_;
On 2017/01/06 19:31:23, oshima wrote:
> can you add this
> 
> // TODO(oshima): this should always use size_in_pixels_
> 
> ?

Done.

[oshima, 2017-01-06 21:55:28]

lgtm

[boliu, 2017-01-10 00:29:56]

boliu@chromium.org changed reviewers:
+ tedchoc@chromium.org

[boliu, 2017-01-10 00:29:57]

aelias: ping for ui/gfx/android
+tedchoc for ui/android

[Ted C, 2017-01-10 00:58:36]

On 2017/01/10 00:29:57, boliu wrote:
> aelias: ping for ui/gfx/android
> +tedchoc for ui/android

lgtm - ui/android

[boliu, 2017-01-10 16:18:16]

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

[commit-bot: I haz the power, 2017-01-10 16:18:37]

Dry run: CQ is trying da patch. Follow status at
 
https://chromium-cq-status.appspot.com/v2/patch-status/codereview.chromium.or...

[boliu, 2017-01-10 17:16:05]

On 2017/01/10 00:58:36, Ted C wrote:
> On 2017/01/10 00:29:57, boliu wrote:
> > aelias: ping for ui/gfx/android
> > +tedchoc for ui/android
> 
> lgtm - ui/android

I'll just remove the comment change to device_display_info.h. It's going to be
removed in 3 CLs, the warning message isn't worth it

[boliu, 2017-01-10 17:21:51]

The CQ bit was checked by boliu@chromium.org

[boliu, 2017-01-10 17:21:51]

The patchset sent to the CQ was uploaded after l-g-t-m from
tedchoc@chromium.org, oshima@chromium.org
Link to the patchset: https://codereview.chromium.org/2617623002/#ps140001
(title: "remove comment change")

[commit-bot: I haz the power, 2017-01-10 17:22:24]

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, 2017-01-10 18:25:12]

CQ is committing da patch.

Bot data: {"patchset_id": 140001, "attempt_start_ts": 1484068911015320,
"parent_rev": "61da5ae4dc99bc1812d1613d83865c082f1598ba", "commit_rev":
"fa8bcd12c55c0997520ff969de9e70ab6d0a59dc"}

[commit-bot: I haz the power, 2017-01-10 18:25:42]

Description was changed from

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

Note need to add storage for physical size in Display, since there are
rounding errors if the physical size is computed from float scale and
integer dip size.

BUG=625089
==========

to

==========
Switch RenderWidgetHostViewAndroid to use Screen

Start to remove usage of native device_display_info.h and eventually
removing it altogether.

RenderWidgetHostViewAndroid uses the primary display size to calculate
the max texture size. Loop over all Displays and use the largest one to
calculate this instead.

Note need to add storage for physical size in Display, since there are
rounding errors if the physical size is computed from float scale and
integer dip size.

BUG=625089

Review-Url: https://codereview.chromium.org/2617623002
Cr-Commit-Position: refs/heads/master@{#442634}
Committed:
https://chromium.googlesource.com/chromium/src/+/fa8bcd12c55c0997520ff969de9e...
==========

[commit-bot: I haz the power, 2017-01-10 18:25:44]

Committed patchset #8 (id:140001) as
https://chromium.googlesource.com/chromium/src/+/fa8bcd12c55c0997520ff969de9e...