Attach color space information to hardware decoded NV12 video frames.

media/gpu/v4l2_slice_video_decode_accelerator.cc

Index: media/gpu/v4l2_slice_video_decode_accelerator.cc
diff --git a/media/gpu/v4l2_slice_video_decode_accelerator.cc b/media/gpu/v4l2_slice_video_decode_accelerator.cc
index e672f72f68c497cab7dd2c7276a01d6c03667333..900edcb7bc3efc7c1d784e2b2f250859fa5546b0 100644
--- a/media/gpu/v4l2_slice_video_decode_accelerator.cc
+++ b/media/gpu/v4l2_slice_video_decode_accelerator.cc
@@ -3168,8 +3168,9 @@ void V4L2SliceVideoDecodeAccelerator::OutputSurface(
   // TODO(posciak): Use visible size from decoder here instead
   // (crbug.com/402760). Passing (0, 0) results in the client using the
   // visible size extracted from the container instead.
+  // TODO(hubbe): Insert correct color space. http://crbug.com/647725
   Picture picture(output_record.picture_id, dec_surface->bitstream_id(),
-                  gfx::Rect(0, 0), false);
+                  gfx::Rect(0, 0), gfx::ColorSpace(), false);
   DVLOGF(3) << dec_surface->ToString()
             << ", bitstream_id: " << picture.bitstream_buffer_id()
             << ", picture_id: " << picture.picture_buffer_id();