cc: Add a test that the IOSurface drawing path makes the right context calls.

cc: Add a test that the IOSurface drawing path makes the right context calls.

This path is super not tested. In the absence of being able to create an IOSurface
and do a pixel test, this makes sure the IOSurface layer's drawing path issues the
appropriate calls on the context to draw - including using the right texture target
which would have caught a regression with the VideoResourceUpdater CL.

R=enne, jamesr
BUG=230300
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=195237

[jamesr, 2013-04-19 00:05:09]

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
File cc/layers/io_surface_layer_impl.cc (left):

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
cc/layers/io_surface_layer_impl.cc:84: GLC(context3d,
context3d->activeTexture(GL_TEXTURE0));
what was the actual bug? this call?

[danakj, 2013-04-19 00:06:40]

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
File cc/layers/io_surface_layer_impl.cc (left):

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
cc/layers/io_surface_layer_impl.cc:84: GLC(context3d,
context3d->activeTexture(GL_TEXTURE0));
On 2013/04/19 00:05:09, jamesr wrote:
> what was the actual bug? this call?

mac context virtualization is the bug on TOT.

the bug I caused was doing bindTexture etc in resource provider with
GL_TEXTURE_2D target. Bo fixed that last night.

[danakj, 2013-04-19 00:07:05]

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
File cc/layers/io_surface_layer_impl.cc (left):

https://codereview.chromium.org/14098007/diff/1/cc/layers/io_surface_layer_im...
cc/layers/io_surface_layer_impl.cc:84: GLC(context3d,
context3d->activeTexture(GL_TEXTURE0));
On 2013/04/19 00:06:41, danakj wrote:
> On 2013/04/19 00:05:09, jamesr wrote:
> > what was the actual bug? this call?
> 
> mac context virtualization is the bug on TOT.
> 
> the bug I caused was doing bindTexture etc in resource provider with
> GL_TEXTURE_2D target. Bo fixed that last night.

I think this call is just redundant cuz we're always in this state when we leave
the renderer - please correct me if I'm wrong.

[jamesr, 2013-04-19 01:49:32]

lgtm

I see, and the texture parameters will be set by
ResourceProvider::CreateResourceFromExternalTexture().  Cool.

lgtm - testing the GL calls seems kind of unfortunate but I don't have any
better ideas

[danakj, 2013-04-19 16:04:10]

I verified that iosurface path works with this patch.

[commit-bot: I haz the power, 2013-04-19 16:04:17]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/danakj@chromium.org/14098007/1

[commit-bot: I haz the power, 2013-04-19 18:25:29]

Change committed as 195237