Add VDA supported profile to GPUInfo.

media/renderers/gpu_video_accelerator_factories.h

Index: media/renderers/gpu_video_accelerator_factories.h
diff --git a/media/renderers/gpu_video_accelerator_factories.h b/media/renderers/gpu_video_accelerator_factories.h
index c77f86aeac9987daac72a307aba9797107d27ca7..221c1807421edf291600b7b7fba403601f8e0683 100644
--- a/media/renderers/gpu_video_accelerator_factories.h
+++ b/media/renderers/gpu_video_accelerator_factories.h
@@ -11,6 +11,7 @@
 #include "base/memory/scoped_ptr.h"
 #include "gpu/command_buffer/common/mailbox.h"
 #include "media/base/media_export.h"
+#include "media/video/video_decode_accelerator.h"
 #include "media/video/video_encode_accelerator.h"
 
 namespace base {
@@ -64,6 +65,10 @@ class MEDIA_EXPORT GpuVideoAcceleratorFactories
   // Returns the task runner the video accelerator runs on.
   virtual scoped_refptr<base::SingleThreadTaskRunner> GetTaskRunner() = 0;
 
+  // Returns the supported codec profiles of video decode accelerator.
+  virtual std::vector<VideoDecodeAccelerator::SupportedProfile>
+      GetVideoDecodeAcceleratorSupportedProfiles() = 0;

[xhwang, 2015/03/26 06:15:32]

It seems this is only needed by IsProfileSupported() in GpuVideoDecoder. In that
case, why do we need to expose the full list of supported profiles to GVD? Can
we just have a CanDecode(profile, size) function here?

[henryhsu_tw, 2015/03/26 09:38:34]

Because we need GetVideoEncodeAcceleratorSupportedProfiles in
rtc_video_encoder_factory.cc.
I'd prefer to keep original function prototype of decoder for consistency.

[xhwang, 2015/03/26 16:21:48]

Acknowledged.

+
   // Returns the supported codec profiles of video encode accelerator.
   virtual std::vector<VideoEncodeAccelerator::SupportedProfile>
       GetVideoEncodeAcceleratorSupportedProfiles() = 0;

[xhwang, 2015/03/26 06:15:32]

ditto