diff options
author | wm4 <wm4@nowhere> | 2013-09-22 22:47:15 +0200 |
---|---|---|
committer | wm4 <wm4@nowhere> | 2013-09-25 13:53:42 +0200 |
commit | 7c3f1ffc44b532590e661315b42fea5b6d1ae40c (patch) | |
tree | 30f96bfb61b5cd46d36cae0480de58bb7de8b8b5 /video/decode/lavc.h | |
parent | 39d1ab82e5a3ac30e5495c6b6773823c2ff56594 (diff) | |
download | mpv-7c3f1ffc44b532590e661315b42fea5b6d1ae40c.tar.bz2 mpv-7c3f1ffc44b532590e661315b42fea5b6d1ae40c.tar.xz |
vd_lavc: allow process_image to change image format
This will allow GPU read-back with process_image.
We have to restructure how init_vo() works. Instead of initializing the
VO before process_image is called, rename init_vo() to
update_image_params(), and let it update the params only. Then we really
initialize the VO after process_image.
As a consequence of these changes, already decoded hw frames are
correctly unreferenced if creation of the filter chain fails. This
could trigger assertions on VO uninitialization, because it's not
allowed to reference hw frames past VO lifetime.
Diffstat (limited to 'video/decode/lavc.h')
-rw-r--r-- | video/decode/lavc.h | 2 |
1 files changed, 1 insertions, 1 deletions
diff --git a/video/decode/lavc.h b/video/decode/lavc.h index 769d20b0df..681f151ad0 100644 --- a/video/decode/lavc.h +++ b/video/decode/lavc.h @@ -26,9 +26,9 @@ typedef struct lavc_ctx { struct vd_lavc_hwdec *hwdec; enum PixelFormat pix_fmt; int do_hw_dr1; - int vo_initialized; int best_csp; struct mp_image_params image_params; + struct mp_image_params vo_image_params; AVRational last_sample_aspect_ratio; enum AVDiscard skip_frame; const char *software_fallback_decoder; |