Run VDA::Decode on GPU IO thread if VDA supports it.

content/common/gpu/gpu_channel.h

 // Copyright (c) 2012 The Chromium Authors. All rights reserved.
 // Use of this source code is governed by a BSD-style license that can be
 // found in the LICENSE file.
 
 #ifndef CONTENT_COMMON_GPU_GPU_CHANNEL_H_
 #define CONTENT_COMMON_GPU_GPU_CHANNEL_H_
 
 #include <deque>
 #include <string>
 
@@ skipping 68 matching lines @@
 
   // Returns the name of the associated IPC channel.
   std::string GetChannelName();
 
 #if defined(OS_POSIX)
   int TakeRendererFileDescriptor();
 #endif  // defined(OS_POSIX)
 
   base::ProcessId renderer_pid() const { return channel_->peer_pid(); }
 
+  scoped_refptr<base::MessageLoopProxy> io_message_loop() const {

[Ami GONE FROM CHROMIUM, 2013/08/27 17:10:33]

I was not advocating for that (though I see no reason not to).
Instead, I was making the observation that the *VDA's themselves don't need to
be told what the IO loop is since they can get it for themselves.

[wuchengli, 2013/08/28 13:18:21]

I see what you mean now. I have tried to use
ChildProcess::current()->io_message_loop_proxy() in
GpuVideoDecodeAccelerator::Initialize. But DEPS did not allow to include
content/child from GpuVideoDecodeAccelerator.

+    return io_message_loop_;
+  }
+
   // IPC::Listener implementation:
   virtual bool OnMessageReceived(const IPC::Message& msg) OVERRIDE;
   virtual void OnChannelError() OVERRIDE;
 
   // IPC::Sender implementation:
   virtual bool Send(IPC::Message* msg) OVERRIDE;
 
   // Requeue the message that is currently being processed to the beginning of
   // the queue. Used when the processing of a message gets aborted because of
   // unscheduling conditions.
@@ skipping 50 matching lines @@
       scoped_refptr<gpu::PreemptionFlag> preemption_flag);
 
 #if defined(OS_ANDROID)
   StreamTextureManagerAndroid* stream_texture_manager() {
     return stream_texture_manager_.get();
   }
 #endif
 
   void CacheShader(const std::string& key, const std::string& shader);
 
+  void AddFilter(IPC::ChannelProxy::MessageFilter* filter);
+  void RemoveFilter(IPC::ChannelProxy::MessageFilter* filter);
+
  protected:
   virtual ~GpuChannel();
 
  private:
   friend class base::RefCountedThreadSafe<GpuChannel>;
   friend class GpuChannelMessageFilter;
 
   void OnDestroy();
 
   bool OnControlMessageReceived(const IPC::Message& msg);
@@ skipping 90 matching lines @@
   scoped_refptr<base::MessageLoopProxy> io_message_loop_;
 
   size_t num_stubs_descheduled_;
 
   DISALLOW_COPY_AND_ASSIGN(GpuChannel);
 };
 
 }  // namespace content
 
 #endif  // CONTENT_COMMON_GPU_GPU_CHANNEL_H_