Move Insert/SignalSyncPoint() to GpuControl

content/common/gpu/client/command_buffer_proxy_impl.cc

Index: content/common/gpu/client/command_buffer_proxy_impl.cc
diff --git a/content/common/gpu/client/command_buffer_proxy_impl.cc b/content/common/gpu/client/command_buffer_proxy_impl.cc
index 3f40ca421e01293781a949db9d79cb6fc74df568..7db78b60472920e951d8ac022c915cde7ac36d7e 100644
--- a/content/common/gpu/client/command_buffer_proxy_impl.cc
+++ b/content/common/gpu/client/command_buffer_proxy_impl.cc
@@ -433,22 +433,19 @@ uint32 CommandBufferProxyImpl::InsertSyncPoint() {
   return sync_point;
 }
 
-bool CommandBufferProxyImpl::SignalSyncPoint(uint32 sync_point,
+void CommandBufferProxyImpl::SignalSyncPoint(uint32 sync_point,
                                              const base::Closure& callback) {
-  if (last_state_.error != gpu::error::kNoError) {
-    return false;
-  }
+  if (last_state_.error != gpu::error::kNoError)
+    return;
 
   uint32 signal_id = next_signal_id_++;
   if (!Send(new GpuCommandBufferMsg_SignalSyncPoint(route_id_,
                                                     sync_point,
                                                     signal_id))) {
-    return false;
+    return;

[no sievers, 2013/09/27 18:13:04]

We might want to revisit the behavior and always guarantee signal callbacks to
be run. Seems a bit hard otherwise for the client not to potentially deadlock,
esp. if you don't know whether it will run or not (no error returned). But I
guess it's consistent for now (we don't run them in the destructor).

   }
 
   signal_tasks_.insert(std::make_pair(signal_id, callback));
-
-  return true;
 }
 
 bool CommandBufferProxyImpl::SignalQuery(unsigned query,