Reland: Updating calls to offscreen buffers to use a size of (0,0) when they do not require an actu…

Reland: Updating calls to offscreen buffers to use a size of (0,0) when they do not require an actual buffer

These updates would enable the surfaceless context extension where available.

BUG=377497
Committed: https://src.chromium.org/viewvc/chrome?view=rev&revision=273691

[dnicoara, 2014-05-29 17:48:31]

Relanding https://codereview.chromium.org/293293002/ with updates to the GLX
Pbuffer impl.

[Ken Russell (switch to Gerrit), 2014-05-29 17:56:46]

LGTM

I'm still concerned that this is going to break on Mac OS X. I think the 1x1
workaround should be applied on that platform too.

[dnicoara, 2014-05-29 18:59:39]

On 2014/05/29 17:56:46, Ken Russell wrote:
> LGTM
> 
> I'm still concerned that this is going to break on Mac OS X. I think the 1x1
> workaround should be applied on that platform too.

I was going to update Mac just in case, but looking at Mac, it seems we're only
using NoOpGLSurfaceCGL which doesn't make any native calls. Am I missing
something? Or are you concerned that the size returned by NoOpGLSurfaceCGL won't
be handled correctly?

[Ken Russell (switch to Gerrit), 2014-05-29 19:29:34]

On 2014/05/29 18:59:39, dnicoara wrote:
> On 2014/05/29 17:56:46, Ken Russell wrote:
> > LGTM
> > 
> > I'm still concerned that this is going to break on Mac OS X. I think the 1x1
> > workaround should be applied on that platform too.
> 
> I was going to update Mac just in case, but looking at Mac, it seems we're
only
> using NoOpGLSurfaceCGL which doesn't make any native calls. Am I missing
> something? Or are you concerned that the size returned by NoOpGLSurfaceCGL
won't
> be handled correctly?

Ah, I'd forgotten about that. Thanks for reminding me. Nothing to be done for
Mac then. LGTM

[dnicoara, 2014-05-29 19:36:12]

The CQ bit was checked by dnicoara@chromium.org

[commit-bot: I haz the power, 2014-05-29 19:40:56]

CQ is trying da patch. Follow status at
 https://chromium-status.appspot.com/cq/dnicoara@chromium.org/303223002/1

[commit-bot: I haz the power, 2014-05-29 23:18:52]

FYI, CQ is re-trying this CL (attempt #1).
The failing builders are:
  win_chromium_rel on tryserver.chromium
(http://build.chromium.org/p/tryserver.chromium/builders/win_chromium_rel/buil...)

[commit-bot: I haz the power, 2014-05-30 01:27:49]

Change committed as 273691

[oetuaho-nv, 2014-06-06 12:03:29]

This change broke Chrome shell on Android (tested on Tegra K1). There's at least
a bug in gl_context_android.cc GLContext::CreateGLContext() which will
incorrectly create GLNonOwnedContext instead of GLContextEGL when it's supposed
to create a surfaceless context, subsequently triggering a crash. But I don't
think that is the only bug, even with a hacky fix for that I still get only a
black window.

[Ken Russell (switch to Gerrit), 2014-06-06 21:55:43]

On 2014/06/06 12:03:29, oetuaho-nv wrote:
> This change broke Chrome shell on Android (tested on Tegra K1). There's at
least
> a bug in gl_context_android.cc GLContext::CreateGLContext() which will
> incorrectly create GLNonOwnedContext instead of GLContextEGL when it's
supposed
> to create a surfaceless context, subsequently triggering a crash. But I don't
> think that is the only bug, even with a hacky fix for that I still get only a
> black window.

Sorry about the breakage. Could you please file a bug about it and send me the
bug ID? Even better, could you please propose a patch which will fix the
breakage you're seeing? I don't think it's possible that things are broken
everywhere, because Chrome on Android is running with this change.

[oetuaho-nv, 2014-06-09 07:40:35]

On 2014/06/06 21:55:43, Ken Russell wrote:
> Sorry about the breakage. Could you please file a bug about it and send me the
> bug ID? Even better, could you please propose a patch which will fix the
> breakage you're seeing? I don't think it's possible that things are broken
> everywhere, because Chrome on Android is running with this change.

Filed https://code.google.com/p/chromium/issues/detail?id=382349

Have you tested on an Android device that supports surfaceless EGL on your end?
I imagine things work just fine on a device that doesn't support it.