2024-05-24, 10:31 PM
After installing the packages: https://pastebin.com/jP0MjVtj
2024-05-24, 10:31 PM
After installing the packages: https://pastebin.com/jP0MjVtj
2024-05-24, 10:49 PM
(This post was last modified: 2024-05-24, 10:51 PM by TheDreadPirate. Edited 2 times in total.)
The ffmpeg log looks the same.
For the record, the Jellyfin docker image contains ALL the user space drivers needed for transcoding. The official image even contains the OpenCL driver for HDR tone mapping (env mod required for the linuxserver version). You system only needs to bring kernel level driver support to the table, which is what I think is bugged.
2024-05-24, 10:56 PM
Does it make a difference that my server does not have a monitor connected? Because I notice that for example vainfo doesn't work on headless systems.
2024-05-24, 11:02 PM
(This post was last modified: 2024-05-24, 11:03 PM by TheDreadPirate. Edited 1 time in total.)
I have heard of weird issues when a display isn't connected on some GPUs. It would be worth a try to connect SOMETHING to the iGPU display output.
2024-05-25, 08:51 PM
(This post was last modified: 2024-05-25, 09:15 PM by zjeffer. Edited 3 times in total.)
I now connected a monitor to the iGPU, this is now the output of vainfo:
Code: Trying display: wayland Still the same libva error, though. I notice that even though my display settings are set to the correct resolution and refresh rate, the screen looks kind of blurry and is flickering very slightly. As if the intel drivers aren't working very well. Running Plasma 6 (X11). I don't have this when connecting the monitor to the GPU. EDIT: the blurry screen and flickering is fixed after setting the refresh rate to 50Hz instead of 60hz. Weird because it's a 60Hz monitor...
2024-05-25, 09:50 PM
(This post was last modified: 2024-05-25, 10:06 PM by zjeffer. Edited 3 times in total.)
I've found something with some help from ChatGPT
This ffmpeg command is executed by jellyfin: Code: /usr/lib/jellyfin-ffmpeg/ffmpeg -analyzeduration 200M -probesize 1G -init_hw_device vaapi=va:/dev/dri/renderD129,driver=i965 -noautorotate -i file:"/media/Films/Poor.Things.2023.1080p.WEBRip.DDP5.1.x265.10bit-GalaxyRG265[TGx]/Poor.Things.2023.1080p.WEBRip.DDP5.1.x265.10bit-GalaxyRG265.mkv" -map_metadata -1 -map_chapters -1 -threads 0 -map 0:0 -map 0:1 -map -0:s -codec:v:0 h264_vaapi -rc_mode CBR -b:v 7289035 -maxrate 7289035 -bufsize 14578070 -sei -a53_cc -force_key_frames:0 "expr:gte(t,n_forced*3)" -vf "setparams=color_primaries=bt709:color_trc=bt709:colorspace=bt709,scale=trunc(min(max(iw\,ih*a)\,min(1790\,1080*a))/2)*2:trunc(min(max(iw/a\,ih)\,min(1790/a\,1080))/2)*2,format=nv12,hwupload_vaapi" -codec:a:0 libfdk_aac -ac 6 -ab 640000 -copyts -avoid_negative_ts disabled -max_muxing_queue_size 2048 -f hls -max_delay 5000000 -hls_time 3 -hls_segment_type fmp4 -hls_fmp4_init_filename "36c3ace70048e6dc9742510797b05bb2-1.mp4" -start_number 0 -hls_segment_filename "/config/data/transcodes/36c3ace70048e6dc9742510797b05bb2%d.mp4" -hls_playlist_type vod -hls_list_size 0 -y "/config/data/transcodes/36c3ace70048e6dc9742510797b05bb2.m3u8" and produces this error: Code: DRM_IOCTL_I915_GEM_APERTURE failed: Invalid argument But when I change format=nv12,hwupload_vaapi to format=nv12,hwupload in that command and run it manually within the docker container, it transcodes fine, and with much less CPU usage (20% instead of 80-90%) compared to disabling the "hardware encoding" checkbox. However, it also transcodes much slower than disabling hardware encoding: 2x instead of 6-7x.What's the difference between hwupload_vaapi and hwupload? Does hwupload still use hardware encoding of some kind?
2024-06-16, 10:41 AM
@TheDreadPirate sorry to bump this but have you seen my last question? The problem is still happening with hardware encoding enabled.
2024-06-16, 04:43 PM
Unfortunately, I've never seen that error. VAAPI should work on pre-broadwell systems so I'm not sure why it isn't. QSV on pre-broadwell in Linux is deprecated, though.
|
|
|