[FFmpeg-devel] [PATCH] cuvid: support AVCodecContext.hw_device_ctx API

[wm4]

This is a newer API that is intended for decoders like the cuvid
wrapper. Until now, the wrapper required to set an awkward
"incomplete" hw_frames_ctx to set the device. Now the device
can be set directly, and the user can get AV_PIX_FMT_CUDA output
for a specific device simply by setting hw_device_ctx.

This still does a dummy ff_get_format() call at init time, and should
be fully backward compatible.
---
Not sure how to test correctly - it worked with mpv even when I
accidentally didn't use the correct VO device.
---
 libavcodec/cuvid.c | 14 +++++++++++---
 1 file changed, 11 insertions(+), 3 deletions(-)

diff --git a/libavcodec/cuvid.c b/libavcodec/cuvid.c
index 288083423e..3453003965 100644
--- a/libavcodec/cuvid.c
+++ b/libavcodec/cuvid.c
@@ -802,9 +802,17 @@ static av_cold int cuvid_decode_init(AVCodecContext *avctx)
             goto error;
         }
     } else {
-        ret = av_hwdevice_ctx_create(&ctx->hwdevice, AV_HWDEVICE_TYPE_CUDA, ctx->cu_gpu, NULL, 0);
-        if (ret < 0)
-            goto error;
+        if (avctx->hw_device_ctx) {
+            ctx->hwdevice = av_buffer_ref(avctx->hw_device_ctx);
+            if (!ctx->hwdevice) {
+                ret = AVERROR(ENOMEM);
+                goto error;
+            }
+        } else {
+            ret = av_hwdevice_ctx_create(&ctx->hwdevice, AV_HWDEVICE_TYPE_CUDA, ctx->cu_gpu, NULL, 0);
+            if (ret < 0)
+                goto error;
+        }
 
         ctx->hwframe = av_hwframe_ctx_alloc(ctx->hwdevice);
         if (!ctx->hwframe) {
-- 
2.11.0