Make ui::Layer use the cc:LayerTreeHost scale factor directly

Make ui::Layer use the cc:LayerTreeHost scale factor directly

Make ui::Layer's bounds be the cc::Layer's bounds, rather
than a scaled version of them. Delete code in ui::Layer
used to compute and track the scaled bounds.

Change ui::Compositor to set the device scale factor on the
cc::LayerTreeHost when its scale factor is set.

BUG=170219
TEST=manual testing with --ash-host-window-bounds=1280x1024*2,1280+0-1280x1024, window dragging, etc

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=270370

[ccameron, 2014-05-13 20:06:31]

ptal

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc#newco...
ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left, uv_bottom_right);
Do you have an example of anything that will trigger this? I didn't hit it in my
manual testing.

[piman, 2014-05-13 20:13:09]

lgtm

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc#newco...
ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left, uv_bottom_right);
On 2014/05/13 20:06:31, ccameron1 wrote:
> Do you have an example of anything that will trigger this? I didn't hit it in
my
> manual testing.

I think this is only used now with the mirroring path.

IIRC in debug on a chromeos build you can hit ctrl-shift-D (to create a new
display) and then ctrl-shift-M (to turn on mirroring).

[ccameron, 2014-05-13 20:28:25]

Thanks!

Adding ben for OWNER stamp on ui/views changes.

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc#newco...
ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left, uv_bottom_right);
On 2014/05/13 20:13:10, piman wrote:
> On 2014/05/13 20:06:31, ccameron1 wrote:
> > Do you have an example of anything that will trigger this? I didn't hit it
in
> my
> > manual testing.
> 
> I think this is only used now with the mirroring path.
> 
> IIRC in debug on a chromeos build you can hit ctrl-shift-D (to create a new
> display) and then ctrl-shift-M (to turn on mirroring).

Okay, this needed to be scaled to work correctly with mirroring. Working now.

[piman, 2014-05-13 20:32:24]

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc#newco...
ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left, uv_bottom_right);
On 2014/05/13 20:28:25, ccameron1 wrote:
> On 2014/05/13 20:13:10, piman wrote:
> > On 2014/05/13 20:06:31, ccameron1 wrote:
> > > Do you have an example of anything that will trigger this? I didn't hit it
> in
> > my
> > > manual testing.
> > 
> > I think this is only used now with the mirroring path.
> > 
> > IIRC in debug on a chromeos build you can hit ctrl-shift-D (to create a new
> > display) and then ctrl-shift-M (to turn on mirroring).
> 
> Okay, this needed to be scaled to work correctly with mirroring. Working now.

Mmh, why?

[piman, 2014-05-13 20:38:18]

On Tue, May 13, 2014 at 1:32 PM, <piman@chromium.org> wrote:

>
> https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
> File ui/compositor/layer.cc (right):
>
> https://codereview.chromium.org/271533014/diff/1/ui/
> compositor/layer.cc#newcode922
> ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left,
> uv_bottom_right);
> On 2014/05/13 20:28:25, ccameron1 wrote:
>
>> On 2014/05/13 20:13:10, piman wrote:
>> > On 2014/05/13 20:06:31, ccameron1 wrote:
>> > > Do you have an example of anything that will trigger this? I
>>
> didn't hit it
>
>> in
>> > my
>> > > manual testing.
>> >
>> > I think this is only used now with the mirroring path.
>> >
>> > IIRC in debug on a chromeos build you can hit ctrl-shift-D (to
>>
> create a new
>
>> > display) and then ctrl-shift-M (to turn on mirroring).
>>
>
>  Okay, this needed to be scaled to work correctly with mirroring.
>>
> Working now.
>
> Mmh, why?
>

Ok, it's a double correction - bad.
The problem is that ReflectorImpl passes a pixel size as a dip size.

Since they come from different displays, we'd have problems with DPI
mismatch - I guess we haven't seen this problem yet because this code path
is only used on low-DPI devices.


>
> https://codereview.chromium.org/271533014/
>

To unsubscribe from this group and stop receiving emails from it, send an email
to chromium-reviews+unsubscribe@chromium.org.

[Ben Goodger (Google), 2014-05-13 20:45:25]

lgtm

[ccameron, 2014-05-13 20:55:55]

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
File ui/compositor/layer.cc (right):

https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc#newco...
ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left, uv_bottom_right);
On 2014/05/13 20:32:24, piman wrote:
> On 2014/05/13 20:28:25, ccameron1 wrote:
> > On 2014/05/13 20:13:10, piman wrote:
> > > On 2014/05/13 20:06:31, ccameron1 wrote:
> > > > Do you have an example of anything that will trigger this? I didn't hit
it
> > in
> > > my
> > > > manual testing.
> > > 
> > > I think this is only used now with the mirroring path.
> > > 
> > > IIRC in debug on a chromeos build you can hit ctrl-shift-D (to create a
new
> > > display) and then ctrl-shift-M (to turn on mirroring).
> > 
> > Okay, this needed to be scaled to work correctly with mirroring. Working
now.
> 
> Mmh, why?

Oh, I didn't check the original behavior. It was broken before this change. Try
running with

--ash-host-window-bounds=1280x1024,1280+0-1280x1024*2

and do a ctrl+shift+M. The primary display will appear cropped on the secondary
display, instead of scaled. With this patch (or just this part of this patch),
it is scaled instead of cropped.

[ccameron, 2014-05-13 20:56:00]

The CQ bit was checked by ccameron@chromium.org

[commit-bot: I haz the power, 2014-05-13 20:56:35]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ccameron@chromium.org/271533014/20001

[ccameron, 2014-05-13 23:58:08]

The CQ bit was unchecked by ccameron@chromium.org

[ccameron, 2014-05-14 06:34:56]

On 2014/05/13 20:38:18, piman wrote:
> On Tue, May 13, 2014 at 1:32 PM, <mailto:piman@chromium.org> wrote:
> 
> >
> > https://codereview.chromium.org/271533014/diff/1/ui/compositor/layer.cc
> > File ui/compositor/layer.cc (right):
> >
> > https://codereview.chromium.org/271533014/diff/1/ui/
> > compositor/layer.cc#newcode922
> > ui/compositor/layer.cc:922: texture_layer_->SetUV(uv_top_left,
> > uv_bottom_right);
> > On 2014/05/13 20:28:25, ccameron1 wrote:
> >
> >> On 2014/05/13 20:13:10, piman wrote:
> >> > On 2014/05/13 20:06:31, ccameron1 wrote:
> >> > > Do you have an example of anything that will trigger this? I
> >>
> > didn't hit it
> >
> >> in
> >> > my
> >> > > manual testing.
> >> >
> >> > I think this is only used now with the mirroring path.
> >> >
> >> > IIRC in debug on a chromeos build you can hit ctrl-shift-D (to
> >>
> > create a new
> >
> >> > display) and then ctrl-shift-M (to turn on mirroring).
> >>
> >
> >  Okay, this needed to be scaled to work correctly with mirroring.
> >>
> > Working now.
> >
> > Mmh, why?
> >
> 
> Ok, it's a double correction - bad.

Ah -- removed that part of the patch then. Since it's not a regression (the
behavior happens before and after this patch), I'll send the patch in as-is.

> The problem is that ReflectorImpl passes a pixel size as a dip size.

I'll fix that bug in a separate patch, since I'm in the neighborhood.

> 
> Since they come from different displays, we'd have problems with DPI
> mismatch - I guess we haven't seen this problem yet because this code path
> is only used on low-DPI devices.
> 
> 
> >
> > https://codereview.chromium.org/271533014/
> >
> 
> To unsubscribe from this group and stop receiving emails from it, send an
email
> to mailto:chromium-reviews+unsubscribe@chromium.org.

[ccameron, 2014-05-14 06:35:01]

The CQ bit was checked by ccameron@chromium.org

[commit-bot: I haz the power, 2014-05-14 06:35:06]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/ccameron@chromium.org/271533014/40001

[commit-bot: I haz the power, 2014-05-14 10:02:34]

Change committed as 270370