I tried lsof to see who's using the card and got this:
[jakepii@kiertotahti2 ~]$ lsof|grep /dev/snd
pulseaudi 1813      jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
pulseaudi 1813      jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sink 1813 1815 jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sink 1813 1815 jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sour 1813 1816 jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sour 1813 1816 jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1
[jakepii@kiertotahti2 ~]$ kill 1813
[jakepii@kiertotahti2 ~]$ lsof|grep /dev/snd
[jakepii@kiertotahti2 ~]$ hdsploader
hdsploader - firmware loader for RME Hammerfall DSP cards
Looking for HDSP + Multiface or Digiface cards :
Card 0 : RME Hammerfall DSP + Multiface at 0xd0000000, irq 17
Upload firmware for card hw:0
Hwdep ioctl error on card hw:0 : Device or resource busy.
Card 1 : HDA Intel PCH at 0xd4720000 irq 47
[jakepii@kiertotahti2 ~]$ lsof|grep /dev/snd
pulseaudi 1834      jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
pulseaudi 1834      jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sink 1834 1836 jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sink 1834 1836 jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sour 1834 1837 jakepii   22u      CHR             116,15       0t0
10902 /dev/snd/controlC1
alsa-sour 1834 1837 jakepii   29u      CHR             116,15       0t0
10902 /dev/snd/controlC1

Does this tell anyone anything?
--
Jaakko



Hi Jaako,

The RME is hw:0 now. As you can see the card numbers can change in different reboots. So in this case, you should have looked for ...C0... at the output of "lsof | grep /dev/snd" and kill the process. Check the card numbering beforehand via "cat /proc/asound/cards".

You can give the cards consistent numbers or you can also disable the onboard one from the BIOS but this is a different question.

OTOH, although you killed a pulseaudio process, it seems it did not take effect. I think this is because of the pulseaudio "autospawn" feature. Therefore, assuming that pulseaudio is also opening your RME device and at least provisionally, you have to disable autospawn. Create (or edit) $HOME/.pulse/client.conf and type "autospawn=no" (without the quotes). You don't need to reboot. This works in ubuntu at least.

I have had this problem of "device busy" in ubuntu, when jack refused to start because pulseaudio was grabbing the card, but honestly I don't own a RME card so I am not sure whether this is the actual problem.

Cheers, Pablo