diff options
| author | wm4 <nfxjfg@googlemail.com> | 2017-10-03 15:04:45 +0200 |
|---|---|---|
| committer | Timo Rothenpieler <timo@rothenpieler.org> | 2017-11-10 22:48:34 +0100 |
| commit | ae5046e492cd87233111e994ca4ae3d74a233b51 (patch) | |
| tree | 7858bdb33d3fe30a13821553d51874376c263726 /libavcodec/cuviddec.c | |
| parent | 5593049466bcd1d3f1ddbfe580be4f36123d7c3d (diff) | |
avcodec: allow multiple hwaccels for the same codec/pixfmt
Currently, AVHWAccels are looked up using a (codec_id, pixfmt) tuple.
This means it's impossible to have 2 decoders for the same codec and
using the same opaque hardware pixel format.
This breaks merging Libav's CUVID hwaccel. FFmpeg has its own CUVID
support, but it's a full stream decoder, using NVIDIA's codec parser.
The Libav one is a true hwaccel, which is based on the builtin software
decoders.
Fix this by introducing another field to disambiguate AVHWAccels, and
use it for our CUVID decoders. FF_CODEC_CAP_HWACCEL_REQUIRE_CLASS makes
this mechanism backwards compatible and optional.
Diffstat (limited to 'libavcodec/cuviddec.c')
| -rw-r--r-- | libavcodec/cuviddec.c | 2 |
1 files changed, 2 insertions, 0 deletions
diff --git a/libavcodec/cuviddec.c b/libavcodec/cuviddec.c index 2ba8e00c6a..6370348639 100644 --- a/libavcodec/cuviddec.c +++ b/libavcodec/cuviddec.c @@ -1106,6 +1106,7 @@ static const AVOption options[] = { .type = AVMEDIA_TYPE_VIDEO, \ .id = AV_CODEC_ID_##X, \ .pix_fmt = AV_PIX_FMT_CUDA, \ + .decoder_class = &x##_cuvid_class, \ }; \ AVCodec ff_##x##_cuvid_decoder = { \ .name = #x "_cuvid", \ @@ -1120,6 +1121,7 @@ static const AVOption options[] = { .receive_frame = cuvid_output_frame, \ .flush = cuvid_flush, \ .capabilities = AV_CODEC_CAP_DELAY | AV_CODEC_CAP_AVOID_PROBING, \ + .caps_internal = FF_CODEC_CAP_HWACCEL_REQUIRE_CLASS, \ .pix_fmts = (const enum AVPixelFormat[]){ AV_PIX_FMT_CUDA, \ AV_PIX_FMT_NV12, \ AV_PIX_FMT_P010, \ |