Clobber frame_data->device_scale_factor to 1 on Android.

Clobber frame_data->device_scale_factor to 1 on Android.

On Android, the browser compositor lies to CC about the
device_scale_factor (always setting it to 1), because it places its
layers in physical pixel space.  Because of that, the delegated device
scale inversion added in https://codereview.chromium.org/311253004 does
the wrong thing.  Rather than cancelling the inversion with other
transforms, the simplest fix is to lie about the frame_data
device_scale_factor as well.

BUG=383572, 383436, 384134
NOTRY=true

Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=276882

[aelias_OOO_until_Jul13, 2014-06-12 22:57:49]

While starting to add a setting I realized we already had a hook in the form of
mutable frame_data going through RWHVA.  After considering a lot of
possibilities, I think I like this silly hack the best.  It's contained to
Android code and easier to reason about than cancelling transforms.

[jamesr, 2014-06-12 23:00:04]

Seems about as reasonable as any other change.

lgtm

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:695:
gfx::Transform layer_scale(
you probably want to undo this change as well, right?

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:801: //
suppresses the transform.
maybe include a link to the crbug about multiple monitor support on Android
since whoever ends up working on that bug is going to run into this?

[boliu, 2014-06-12 23:19:57]

lgtm

[aelias_OOO_until_Jul13, 2014-06-13 00:17:46]

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:695:
gfx::Transform layer_scale(
On 2014/06/12 23:00:04, jamesr wrote:
> you probably want to undo this change as well, right?

No, since https://codereview.chromium.org/331453004 didn't land, this is old
code that still works.

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:801: //
suppresses the transform.
On 2014/06/12 23:00:04, jamesr wrote:
> maybe include a link to the crbug about multiple monitor support on Android
> since whoever ends up working on that bug is going to run into this?

crbug archeology didn't turn up anything, but I linked newly filed bug
http://crbug.com/384134

[jamesr, 2014-06-13 00:27:53]

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:695:
gfx::Transform layer_scale(
On 2014/06/13 00:17:46, aelias wrote:
> On 2014/06/12 23:00:04, jamesr wrote:
> > you probably want to undo this change as well, right?
> 
> No, since https://codereview.chromium.org/331453004 didn't land, this is old
> code that still works.

Not that old - I added this scale 2 days ago.

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:801: //
suppresses the transform.
On 2014/06/13 00:17:46, aelias wrote:
> On 2014/06/12 23:00:04, jamesr wrote:
> > maybe include a link to the crbug about multiple monitor support on Android
> > since whoever ends up working on that bug is going to run into this?
> 
> crbug archeology didn't turn up anything, but I linked newly filed bug
> http://crbug.com/384134

I was thinking of http://code.google.com/p/chromium/issues/detail?id=310763

[aelias_OOO_until_Jul13, 2014-06-13 00:36:07]

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
File content/browser/renderer_host/render_widget_host_view_android.cc (right):

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:695:
gfx::Transform layer_scale(
On 2014/06/13 00:27:53, jamesr wrote:
> On 2014/06/13 00:17:46, aelias wrote:
> > On 2014/06/12 23:00:04, jamesr wrote:
> > > you probably want to undo this change as well, right?
> > 
> > No, since https://codereview.chromium.org/331453004 didn't land, this is old
> > code that still works.
> 
> Not that old - I added this scale 2 days ago.

Ah I see, removed.

https://codereview.chromium.org/336433003/diff/1/content/browser/renderer_hos...
content/browser/renderer_host/render_widget_host_view_android.cc:801: //
suppresses the transform.
On 2014/06/13 00:27:53, jamesr wrote:
> On 2014/06/13 00:17:46, aelias wrote:
> > On 2014/06/12 23:00:04, jamesr wrote:
> > > maybe include a link to the crbug about multiple monitor support on
Android
> > > since whoever ends up working on that bug is going to run into this?
> > 
> > crbug archeology didn't turn up anything, but I linked newly filed bug
> > http://crbug.com/384134
> 
> I was thinking of http://code.google.com/p/chromium/issues/detail?id=310763

OK, linked.

[aelias_OOO_until_Jul13, 2014-06-13 01:36:13]

The CQ bit was checked by aelias@chromium.org

[commit-bot: I haz the power, 2014-06-13 01:37:26]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/aelias@chromium.org/336433003/40001

[commit-bot: I haz the power, 2014-06-13 01:41:21]

Change committed as 276882