Add gfx::GpuMemoryBuffer::YUV_420 and GpuMemoryBufferImplSharedMemory support

content/common/gpu/client/gpu_memory_buffer_impl.h

Index: content/common/gpu/client/gpu_memory_buffer_impl.h
diff --git a/content/common/gpu/client/gpu_memory_buffer_impl.h b/content/common/gpu/client/gpu_memory_buffer_impl.h
index cd8c108f571a5aaa6709ac5e718a04ec5d9708cf..28fd398ab56bae855d303e09769a30059e5472fb 100644
--- a/content/common/gpu/client/gpu_memory_buffer_impl.h
+++ b/content/common/gpu/client/gpu_memory_buffer_impl.h
@@ -33,15 +33,27 @@ class CONTENT_EXPORT GpuMemoryBufferImpl : public gfx::GpuMemoryBuffer {
   // Type-checking upcast routine. Returns an NULL on failure.
   static GpuMemoryBufferImpl* FromClientBuffer(ClientBuffer buffer);
 
+  // Returns the number of planes based on the format of the buffer.
+  static size_t NumberOfPlanesForGpuMemoryBufferFormat(
+      gfx::GpuMemoryBuffer::Format format);
+
   // Calculates the number of bytes that an implementation must use to store
   // one row of pixel data.
   static bool StrideInBytes(size_t width,
                             Format format,
+                            int plane,
                             size_t* stride_in_bytes);
 
-  // Returns the number of planes based on the format of the buffer.
-  static size_t NumberOfPlanesForGpuMemoryBufferFormat(
-      gfx::GpuMemoryBuffer::Format format);
+  // Returns the subsampling factor applied to the given zero-indexed |plane| of
+  // the |format| both horizontally and vertically.
+  static size_t SubsamplingFactor(gfx::GpuMemoryBuffer::Format format,
+                                  int plane);
+
+  // Calculates the number of bytes that an implementation must use to store
+  // all the planes of a given |format|.
+  static bool BufferSizeInBytes(const gfx::Size& size,

[reveman, 2015/04/07 20:33:44]

nit: can you move this function to GpuMemoryBufferImplSharedMemory and adjust
the comment as other implementations are not forced to use this. they can use a
different total buffer size or the concept of a total buffer size might not even
apply to them..

[emircan, 2015/04/07 21:28:45]

I had it there originally, but then moved it here as
GpuMemoryBufferFactorySharedMemory and FactoryOzone can also use it. But, it
also makes sense to declare new one in ozone using its own formats. 

I added it to GpuMemoryBufferImplSharedMemory and changed include in
GpuMemoryBufferFactorySharedMemory.

+                                gfx::GpuMemoryBuffer::Format format,
+                                size_t* size_in_bytes);
 
   // Overridden from gfx::GpuMemoryBuffer:
   bool IsMapped() const override;