softhddevice unter Debian12

Zitat von lnj

Code

Aug 11 14:27:24 vdr vdr: Selected deinterlacer for resolution 0 is not supported by HW
Aug 11 14:27:24 vdr vdr: Selected deinterlacer for resolution 1 is not supported by HW
Aug 11 14:27:24 vdr vdr: Selected deinterlacer for resolution 2 is not supported by HW
Aug 11 14:27:24 vdr vdr: Selected deinterlacer for resolution 3 is not supported by HW
Aug 11 14:27:24 vdr vdr: Selected deinterlacer for resolution 4 is not supported by HW

This is a normal.

No osd?

No, not so far.

Resulting log file is attached. It was too long for pasting it directly in this forum. The last messages are repeating furthermore.

As mentioned: current version is Ubuntu 20.04 Desktop version as minimal installation with the packages mentioned above. Softhddevice freshly cloned from git. I kill the x-server before my tests with systemctl stop gdm. Starting command for vdr is currently vdr -u vdr -P"softhddevice -a hw:0,7 -d :0.0 -f -x".

Hope you see some potential reasons in the log and you can point out the potential root cause.

Regards,

MrJoe

Zitat von rfehr

Why do you kill the x-server before you test ?

Otherwise the Plugin cannot connect (maybe with parameter -xx, not tried so far).

Zitat von rfehr

and one more question, do you start the vdr via remote access, or locally on the console?

Remote via ssh.

Regards,

MrJoe

Zitat von lnj

If xorg was started before:

-d :0.0 -v va-api-glx -a hdmi:CARD=PCH,DEV=1 -p hdmi:CARD=PCH,DEV=1 -w still-hw-decoder

Heureka! With your parameter set I've got a running VDR menu within the GNOME window manager. Afterwards I tried to stop again gdm and run the same command plus the parameter -x and I've got a running OSD without having a started X-Server. In general I would say, it works! Thanks lnj and also Roland and Stefan for your patience and supportive ideas.

Following questions:

• I checked the parameter and the issue of the OSD is not related to "-w" parameter, but to the -a and -p parameter. I'm not sure if Audio is working with VDR with your parameter set, I will check this later. Any glue, why a working Audio Device (checked with speaker-test -c 2 -D hw:0,7 is causing the OSD issue? Somehow this was misleading me, as I thought the mentioned HW device is the right one and Audio was working with the simple test.
• the repeating errors concerning video/vaapi: invalid surface in ringbuffer are still there. Is this normal due to activated debugging logs or indicate those another issue?
• is the issue with va-api my old Gen6 Intel CPU? If so I can try later with a newer CPU, whether va-api will work. What can be a limitation of glx or other way around: if glx is running why using va-api? Any performance topics?
• I see in the softhddevice main menu currently (after ~5mins stable running VDR - I'm still fascinated), that ~40000 frames are doubled. Is this normal?
• for my understanding: is the parameter -w still-hw-decoder "only" reducing the CPU load in case of still picture?

Once again: thank you for your support so far. Now I will continue adding my DD DVB-S cards or as soon as they will be sent to me a Sundtek USB device.

Regards,

MrJoe

short update: now it runs also under Debian 12 - just a matter of audio parameters... Testing of audio itself will happen soon.

Regards,

MrJoe

additional information: meanwhile I tried vdr on another machine as I wanted to check my old DVB-S2 cards. It's an Intel i5-4670K CPU and there I had another issue: va-info-glx was giving again black screen. I checked with vainfo and identified, that he first tries the driver iHD and afterwards i965:

error: XDG_RUNTIME_DIR is invalid or not set in the environment.
error: can't connect to X server!
libva info: VA-API version 1.17.0
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so
libva info: Found init function __vaDriverInit_1_17
libva error: /usr/lib/x86_64-linux-gnu/dri/iHD_drv_video.so init failed
libva info: va_openDriver() returns 1
libva info: Trying to open /usr/lib/x86_64-linux-gnu/dri/i965_drv_video.so
libva info: Found init function __vaDriverInit_1_8
libva info: va_openDriver() returns 0
vainfo: VA-API version: 1.17 (libva 2.12.0)
vainfo: Driver version: Intel i965 driver for Intel(R) Haswell Desktop - 2.4.1
vainfo: Supported profile and entrypoints
      VAProfileMPEG2Simple            : VAEntrypointVLD
      VAProfileMPEG2Simple            : VAEntrypointEncSlice
      VAProfileMPEG2Main              : VAEntrypointVLD
      VAProfileMPEG2Main              : VAEntrypointEncSlice
      VAProfileH264ConstrainedBaseline: VAEntrypointVLD
      VAProfileH264ConstrainedBaseline: VAEntrypointEncSlice
      VAProfileH264Main               : VAEntrypointVLD
      VAProfileH264Main               : VAEntrypointEncSlice
      VAProfileH264High               : VAEntrypointVLD
      VAProfileH264High               : VAEntrypointEncSlice
      VAProfileH264MultiviewHigh      : VAEntrypointVLD
      VAProfileH264MultiviewHigh      : VAEntrypointEncSlice
      VAProfileH264StereoHigh         : VAEntrypointVLD
      VAProfileH264StereoHigh         : VAEntrypointEncSlice
      VAProfileVC1Simple              : VAEntrypointVLD
      VAProfileVC1Main                : VAEntrypointVLD
      VAProfileVC1Advanced            : VAEntrypointVLD
      VAProfileNone                   : VAEntrypointVideoProc
      VAProfileJPEGBaseline           : VAEntrypointVLD

I found the hint, that you can set the driver to be used by LIBVA_DRIVER_NAME=i965 before calling vdr. So my call of vdr on the second machine is now LIBVA_DRIVER_NAME=i965 vdr -u vdr --no-kbd -P"softhddevice -v va-api -a hdmi:CARD=HDMI,DEV=0 -p hdmi:CARD=HDMI,DEV=0 -w still-hw-decoder ". Runs smoothly without any issues and due to a running DCB-S2 card I can also watch different channels without any disturtions. And yes, on that machine I had to use va-api instead of va-api-glx as video device for softhddevice (reason unknown, but as lnj says: old videos from Intel behave sometimes strange with Linux drivers).

Just as a hint if somebody struggles with the same problem. Now I have to think about building up a new VDR machine in a housing which satisfies the WAF.

Regards,

MrJoe