Implemented new fence syncs which replaces the old sync points.

gpu/command_buffer/service/in_process_command_buffer.h

Index: gpu/command_buffer/service/in_process_command_buffer.h
diff --git a/gpu/command_buffer/service/in_process_command_buffer.h b/gpu/command_buffer/service/in_process_command_buffer.h
index f4f91f81009b027c9cd70ca5a85f4a1f224833ad..7843aa3ac02678edff53963ccc6b080c1a327e5a 100644
--- a/gpu/command_buffer/service/in_process_command_buffer.h
+++ b/gpu/command_buffer/service/in_process_command_buffer.h
@@ -48,7 +48,7 @@ class StreamTextureManagerInProcess;
 
 namespace gpu {
 class SyncPointClient;
-class SyncPointClientState;
+class SyncPointOrderData;
 class SyncPointManager;
 class ValueStateMap;
 
@@ -130,6 +130,9 @@ class GPU_EXPORT InProcessCommandBuffer : public CommandBuffer,
   bool IsGpuChannelLost() override;
   CommandBufferNamespace GetNamespaceID() const override;
   uint64_t GetCommandBufferID() const override;
+  uint64_t GenerateFenceSyncRelease() override;
+  bool IsFenceSyncRelease(uint64_t release) override;
+  bool IsFenceSyncFlushed(uint64_t release) override;
 
   // The serializer interface to the GPU service (i.e. thread).
   class Service {
@@ -215,6 +218,10 @@ class GPU_EXPORT InProcessCommandBuffer : public CommandBuffer,
   void SignalSyncPointOnGpuThread(uint32 sync_point,
                                   const base::Closure& callback);
   bool WaitSyncPointOnGpuThread(uint32 sync_point);
+  void FenceSyncReleaseOnGpuThread(uint64_t release);
+  bool WaitFenceSyncOnGpuThread(gpu::CommandBufferNamespace namespace_id,
+                                uint64_t command_buffer_id,
+                                uint64_t release);
   void SignalQueryOnGpuThread(unsigned query_id, const base::Closure& callback);
   void DestroyTransferBufferOnGpuThread(int32 id);
   void CreateImageOnGpuThread(int32 id,
@@ -242,7 +249,7 @@ class GPU_EXPORT InProcessCommandBuffer : public CommandBuffer,
   scoped_ptr<gles2::GLES2Decoder> decoder_;
   scoped_refptr<gfx::GLContext> context_;
   scoped_refptr<gfx::GLSurface> surface_;
-  scoped_refptr<SyncPointClientState> sync_point_client_state_;
+  scoped_refptr<SyncPointOrderData> sync_point_order_data_;
   scoped_ptr<SyncPointClient> sync_point_client_;
   base::Closure context_lost_callback_;
   bool delayed_work_pending_;  // Used to throttle PerformDelayedWork.
@@ -254,6 +261,8 @@ class GPU_EXPORT InProcessCommandBuffer : public CommandBuffer,
   gpu::Capabilities capabilities_;
   GpuMemoryBufferManager* gpu_memory_buffer_manager_;
   base::AtomicSequenceNumber next_image_id_;
+  uint64_t next_fence_sync_release_;
+  uint64_t flushed_fence_sync_release_;
 
   // Accessed on both threads:
   scoped_ptr<CommandBufferServiceBase> command_buffer_;
@@ -263,6 +272,7 @@ class GPU_EXPORT InProcessCommandBuffer : public CommandBuffer,
   State state_after_last_flush_;
   base::Lock state_after_last_flush_lock_;
   scoped_refptr<gfx::GLShareGroup> gl_share_group_;
+  base::WaitableEvent fence_sync_wait_event_;
 
 #if defined(OS_ANDROID)
   scoped_ptr<StreamTextureManagerInProcess> stream_texture_manager_;