[cc] Mailbox Output Surface Support

content/renderer/render_view_impl.cc

Index: content/renderer/render_view_impl.cc
diff --git a/content/renderer/render_view_impl.cc b/content/renderer/render_view_impl.cc
index 9597b4b4be03110049cc2ccdf174544cfdef69c4..63571b262bb8c0e993470707210ced6029bf942f 100644
--- a/content/renderer/render_view_impl.cc
+++ b/content/renderer/render_view_impl.cc
@@ -83,6 +83,7 @@
 #include "content/renderer/gpu/compositor_output_surface.h"
 #include "content/renderer/gpu/compositor_software_output_device_gl_adapter.h"
 #include "content/renderer/gpu/compositor_thread.h"
+#include "content/renderer/gpu/mailbox_output_surface.h"
 #include "content/renderer/gpu/render_widget_compositor.h"
 #include "content/renderer/idle_user_detector.h"
 #include "content/renderer/input_tag_speech_dispatcher.h"
@@ -2041,8 +2042,15 @@ scoped_ptr<cc::OutputSurface> RenderViewImpl::CreateOutputSurface() {
           new CompositorOutputSurface(routing_id(), NULL,
               new CompositorSoftwareOutputDeviceGLAdapter(context)));
   } else {
-      return scoped_ptr<cc::OutputSurface>(
-          new CompositorOutputSurface(routing_id(), context, NULL));
+      bool composite_to_mailbox =
+          command_line.HasSwitch(cc::switches::kCompositeToMailbox);
+      DCHECK(!composite_to_mailbox || command_line.HasSwitch(
+          cc::switches::kEnableCompositorFrameMessage));

[piman, 2013/02/27 23:51:04]

Could we instead make sure settings.compositorFrameMessage is set when
kCompositeToMailbox) is set?

Maybe as a follow-up, we should look at removing the setting (the decision
should really be done by the OutputSurface).

[no sievers, 2013/02/28 18:43:40]

I'd rather remove kEnableCompositorFrameMessage soon and force consistent
behavior in the meantime (there are unfortunately some places in
render_view_impl_android.cc that check the cmdline, and this functionality -
sending scroll offsets - will be used in conjunction with composite-to-mailbox).
Sound good?

[piman, 2013/02/28 19:30:31]

As long as we come back and clean this soon, ok.

+      // No swap throttling yet when compositing on the main thread.
+      DCHECK(!composite_to_mailbox || is_threaded_compositing_enabled_);
+      return scoped_ptr<cc::OutputSurface>(composite_to_mailbox ?
+          new MailboxOutputSurface(routing_id(), context, NULL) :
+              new CompositorOutputSurface(routing_id(), context, NULL));
   }
 }