Android: Always update SurfaceTexture when used for draw

Android: Always update SurfaceTexture when used

This basically causes it to get updated from
glCopyTextureCHROMIUM() also which is used with
canvas/webgl.

BUG=309162
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=247699

[no sievers, 2014-01-27 23:59:13]

ptal

[no sievers, 2014-01-28 00:03:08]

https://codereview.chromium.org/148513004/diff/1/content/common/gpu/stream_te...
File content/common/gpu/stream_texture_android.cc (right):

https://codereview.chromium.org/148513004/diff/1/content/common/gpu/stream_te...
content/common/gpu/stream_texture_android.cc:122: has_pending_frame_ = false;
Argh, there is a problem.

UpdateTexImage() has the side-effect of calling glBindTexImage(). So I guess if
I switched the context here, I have to restore the binding.

Also, while I don't have to do anything when this gets called from
PrepareTexturesForRender(), I do have to restore the binding here for the
glCopyTextureCHROMIUM() call site, which I can't differentiate from here.

[no sievers, 2014-01-28 00:03:28]

On 2014/01/28 00:03:08, sievers wrote:
>
https://codereview.chromium.org/148513004/diff/1/content/common/gpu/stream_te...
> File content/common/gpu/stream_texture_android.cc (right):
> 
>
https://codereview.chromium.org/148513004/diff/1/content/common/gpu/stream_te...
> content/common/gpu/stream_texture_android.cc:122: has_pending_frame_ = false;
> Argh, there is a problem.
> 
> UpdateTexImage() has the side-effect of calling glBindTexImage(). So I guess
if
> I switched the context here, I have to restore the binding.

glBindTexture() I meant.

> 
> Also, while I don't have to do anything when this gets called from
> PrepareTexturesForRender(), I do have to restore the binding here for the
> glCopyTextureCHROMIUM() call site, which I can't differentiate from here.

[reveman, 2014-01-28 00:33:42]

thanks, lgtm

https://codereview.chromium.org/148513004/diff/10001/content/common/gpu/strea...
File content/common/gpu/stream_texture_android.cc (right):

https://codereview.chromium.org/148513004/diff/10001/content/common/gpu/strea...
content/common/gpu/stream_texture_android.cc:116: if (needs_make_current) {
is ui::ScopedMakeCurrent not smart enough that you can do this unconditionally?
and in that case just use the IsCurrent(NULL) for the DCHECK..

[no sievers, 2014-01-28 00:43:45]

https://codereview.chromium.org/148513004/diff/10001/content/common/gpu/strea...
File content/common/gpu/stream_texture_android.cc (right):

https://codereview.chromium.org/148513004/diff/10001/content/common/gpu/strea...
content/common/gpu/stream_texture_android.cc:116: if (needs_make_current) {
On 2014/01/28 00:33:42, David Reveman wrote:
> is ui::ScopedMakeCurrent not smart enough that you can do this
unconditionally?
> and in that case just use the IsCurrent(NULL) for the DCHECK..

Unfortunately it's not. But the virtual context current logic is also very
tricky :)

Also, I have updated the patch to manually restore the texture binding here if
we did change the current context temporarily (I mean restore the binding in
*that* context).

[piman, 2014-01-28 02:26:29]

lgtm

[commit-bot: I haz the power, 2014-01-28 23:49:02]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sievers@chromium.org/148513004/150001

[commit-bot: I haz the power, 2014-01-29 02:52:18]

Commit queue rejected this change because the description was changed
between the time the change entered the commit queue and the time it
was ready to commit. You can safely check the commit box again.

[commit-bot: I haz the power, 2014-01-29 16:41:35]

CQ is trying da patch. Follow status at
https://chromium-status.appspot.com/cq/sievers@chromium.org/148513004/150001

[commit-bot: I haz the power, 2014-01-29 16:43:09]

Change committed as 247699