V4L2SVDA/VAAPIVDA: use visible size from decoder and pass to client

media/gpu/vp9_decoder.cc

Index: media/gpu/vp9_decoder.cc
diff --git a/media/gpu/vp9_decoder.cc b/media/gpu/vp9_decoder.cc
index b24f2466846e28485dd17b23372c2a57e0deb302..c75de0d21a7db4c46deb589fb9c06d163d7892e9 100644
--- a/media/gpu/vp9_decoder.cc
+++ b/media/gpu/vp9_decoder.cc
@@ -110,10 +110,14 @@ VP9Decoder::DecodeResult VP9Decoder::Decode() {
 
     gfx::Size new_pic_size(curr_frame_hdr_->frame_width,
                            curr_frame_hdr_->frame_height);
+    gfx::Rect new_render_rect(curr_frame_hdr_->render_width,
+                              curr_frame_hdr_->render_height);
     DCHECK(!new_pic_size.IsEmpty());
+    DCHECK(!new_render_rect.IsEmpty());
 
-    if (new_pic_size != pic_size_) {
-      DVLOG(1) << "New resolution: " << new_pic_size.ToString();
+    if (new_pic_size != pic_size_ || new_render_rect != render_rect_) {

[Owen Lin, 2017/06/07 06:11:00]

Same, we don't care about the visible crop here.

[johnylin1, 2017/06/07 15:38:37]

Done.

+      DVLOG(1) << "New resolution: " << new_pic_size.ToString()
+               << ", New render resolution: " << new_render_rect.ToString();
 
       if (!curr_frame_hdr_->IsKeyframe()) {
         // TODO(posciak): This is doable, but requires a few modifications to
@@ -131,6 +135,7 @@ VP9Decoder::DecodeResult VP9Decoder::Decode() {
         ref_frame = nullptr;
 
       pic_size_ = new_pic_size;
+      render_rect_ = new_render_rect;
       return kAllocateNewSurfaces;
     }
 
@@ -204,6 +209,10 @@ gfx::Size VP9Decoder::GetPicSize() const {
   return pic_size_;
 }
 
+gfx::Rect VP9Decoder::GetVisibleRect() const {
+  return render_rect_;
+}
+
 size_t VP9Decoder::GetRequiredNumOfPictures() const {
   // kMaxVideoFrames to keep higher level media pipeline populated, +2 for the
   // pictures being parsed and decoded currently.