The V4L2 driver does not actually have an associated DRM device at all, so
users work around the requirement by giving libva an unrelated display-only
device instead (which is fine, because it doesn't actually do anything with
that device). This was broken by bc9b6358fb7315c0173de322472641766f6289da
forcing a render node, because the display-only device did not have an
associated render node to use. Fix that by just passing through the
original non-render DRM fd if we can't find a render node.
Reported-by: Paul Kocialkowski <paul.kocialkow...@bootlin.com>
---
Also tested by the reporter. I'll apply this in a few days if there are no
comments.
- Mark
libavutil/hwcontext_vaapi.c | 28 ++++++++++++++++------------
1 file changed, 16 insertions(+), 12 deletions(-)
diff --git a/libavutil/hwcontext_vaapi.c b/libavutil/hwcontext_vaapi.c
index 38bdeb7820..2227d6ed69 100644
--- a/libavutil/hwcontext_vaapi.c
+++ b/libavutil/hwcontext_vaapi.c
@@ -1677,20 +1677,24 @@ static int vaapi_device_derive(AVHWDeviceContext *ctx,
} else {
render_node = drmGetRenderDeviceNameFromFd(src_hwctx->fd);
if (!render_node) {
- av_log(ctx, AV_LOG_ERROR, "Failed to find a render node "
- "matching the DRM device.\n");
- return AVERROR(ENODEV);
- }
- fd = open(render_node, O_RDWR);
- if (fd < 0) {
- av_log(ctx, AV_LOG_ERROR, "Failed to open render node %s"
- "matching the DRM device.\n", render_node);
+ av_log(ctx, AV_LOG_VERBOSE, "Using non-render node "
+ "because the device does not have an "
+ "associated render node.\n");
+ fd = src_hwctx->fd;
+ } else {
+ fd = open(render_node, O_RDWR);
+ if (fd < 0) {
+ av_log(ctx, AV_LOG_VERBOSE, "Using non-render node "
+ "because the associated render node "
+ "could not be opened.\n");
+ fd = src_hwctx->fd;
+ } else {
+ av_log(ctx, AV_LOG_VERBOSE, "Using render node %s "
+ "in place of non-render DRM device.\n",
+ render_node);
+ }
free(render_node);
- return AVERROR(errno);
}
- av_log(ctx, AV_LOG_VERBOSE, "Using render node %s in place "
- "of non-render DRM device.\n", render_node);
- free(render_node);
}
}
#else
--
2.28.0
_______________________________________________
ffmpeg-devel mailing list
ffmpeg-devel@ffmpeg.org
https://ffmpeg.org/mailman/listinfo/ffmpeg-devel
To unsubscribe, visit link above, or email
ffmpeg-devel-requ...@ffmpeg.org with subject "unsubscribe".
