Add optimization for CHROMIUM_subscribe_uniform extension.

content/browser/renderer_host/render_widget_host_impl.cc

Index: content/browser/renderer_host/render_widget_host_impl.cc
diff --git a/content/browser/renderer_host/render_widget_host_impl.cc b/content/browser/renderer_host/render_widget_host_impl.cc
index bfbb78498496f2d9c75c9772ad4079b98d983787..473cc06a7ffdf4316c8f948be798dad13c175b0d 100644
--- a/content/browser/renderer_host/render_widget_host_impl.cc
+++ b/content/browser/renderer_host/render_widget_host_impl.cc
@@ -904,17 +904,12 @@ void RenderWidgetHostImpl::ForwardMouseEventWithLatencyInfo(
 
   // Pass mouse state to gpu service if the subscribe uniform
   // extension is enabled.
-  // TODO(orglofch): Only pass mouse information if one of the GL Contexts
-  // is subscribed to GL_MOUSE_POSITION_CHROMIUM
   if (subscribe_uniform_enabled_) {
     gpu::ValueState state;
     state.int_value[0] = mouse_event.x;
     state.int_value[1] = mouse_event.y;
-    GpuProcessHost::SendOnIO(
-        GpuProcessHost::GPU_PROCESS_KIND_SANDBOXED,
-        CAUSE_FOR_GPU_LAUNCH_NO_LAUNCH,
-        new GpuMsg_UpdateValueState(
-            process_->GetID(), GL_MOUSE_POSITION_CHROMIUM, state));
+    GpuProcessHost::SendUpdateValueStateOnIOThread(
+          process_->GetID(), GL_MOUSE_POSITION_CHROMIUM, state);

[piman, 2014/12/05 21:04:36]

I was hoping we could filter here, on the UI thread. Otherwise we'll still post
tasks when no-one is subscribed (common case), that seems wasteful.

[orglofch, 2014/12/06 21:49:04]

Ah I had thought the main concern was IPC overhead. Done.

   }
 }