Add multi-process GpuMemoryBuffer framework.

content/common/gpu/client/gpu_channel_host.h

Index: content/common/gpu/client/gpu_channel_host.h
diff --git a/content/common/gpu/client/gpu_channel_host.h b/content/common/gpu/client/gpu_channel_host.h
index 5efa98dab4535f915504cab96d4a5a2be6149f39..87240efab0f77c8438830dc81b0536c716aa2e07 100644
--- a/content/common/gpu/client/gpu_channel_host.h
+++ b/content/common/gpu/client/gpu_channel_host.h
@@ -24,6 +24,7 @@
 #include "ipc/ipc_sync_channel.h"
 #include "media/video/video_decode_accelerator.h"
 #include "media/video/video_encode_accelerator.h"
+#include "ui/gfx/gpu_memory_buffer.h"
 #include "ui/gfx/native_widget_types.h"
 #include "ui/gfx/size.h"
 #include "ui/gl/gpu_preference.h"
@@ -77,6 +78,10 @@ class CONTENT_EXPORT GpuChannelHostFactory {
       int32 image_id,
       const CreateImageCallback& callback) = 0;
   virtual void DeleteImage(int32 image_id, int32 sync_point) = 0;
+  virtual scoped_ptr<gfx::GpuMemoryBuffer> AllocateGpuMemoryBuffer(
+      size_t width,
+      size_t height,
+      unsigned internalformat) = 0;
 };
 
 // Encapsulates an IPC channel between the client and one GPU process.
@@ -94,6 +99,10 @@ class GpuChannelHost : public IPC::Sender,
       const gpu::GPUInfo& gpu_info,
       const IPC::ChannelHandle& channel_handle);
 
+  // Returns true if |handle| is a valid GpuMemoryBuffer handle that
+  // can be shared to the GPU process.
+  static bool IsValidGpuMemoryBuffer(gfx::GpuMemoryBufferHandle handle);
+
   bool IsLost() const {
     DCHECK(channel_filter_.get());
     return channel_filter_->IsLost();
@@ -162,6 +171,15 @@ class GpuChannelHost : public IPC::Sender,
   // Reserve one unused transfer buffer ID.
   int32 ReserveTransferBufferId();
 
+  // Returns a GPU memory buffer handle to the buffer that can be sent via
+  // IPC to the GPU process. The caller is responsible for ensuring it is
+  // closed. Returns an invalid handle on failure.
+  gfx::GpuMemoryBufferHandle ShareGpuMemoryBufferToGpuProcess(
+      gfx::GpuMemoryBufferHandle source_handle);
+
+  // Reserve one unused gpu memory buffer ID.
+  int32 ReserveGpuMemoryBufferId();
+
  private:
   friend class base::RefCountedThreadSafe<GpuChannelHost>;
   GpuChannelHost(GpuChannelHostFactory* factory,
@@ -228,6 +246,7 @@ class GpuChannelHost : public IPC::Sender,
   // Threading notes: all fields are constant during the lifetime of |this|
   // except:
   // - |next_transfer_buffer_id_|, atomic type
+  // - |next_gpu_memory_buffer_id_|, atomic type
   // - |proxies_|, protected by |context_lock_|
   GpuChannelHostFactory* const factory_;
   const int client_id_;
@@ -244,6 +263,9 @@ class GpuChannelHost : public IPC::Sender,
   // Transfer buffer IDs are allocated in sequence.
   base::AtomicSequenceNumber next_transfer_buffer_id_;
 
+  // Gpu memory buffer IDs are allocated in sequence.
+  base::AtomicSequenceNumber next_gpu_memory_buffer_id_;
+
   // Protects proxies_.
   mutable base::Lock context_lock_;
   // Used to look up a proxy from its routing id.