> I've got the same problem.
>
> I'm using single Pentium 150, Intel/AS Atlantis motherboard with on-board
> CS4232 sound unit. The problem's been there from at least 2.1.80 (I wrote
> about it some time ago to the list).
[snip]
> and a message from kernel:
>
> ad1848/cs4248 codec driver Copyright (C) by Hannu Savolainen 1993-1996
> Sound: Too many mixer drivers
[snip]
I've been looking at the cs4232 driver while trying to write a driver for
Yamaha OPL3-SA[2,3,x] based cards, and I've noticed that, at least in the
stock 2.1.105 that I've been playing with, the cs4232 driver does not
unload the mixer device when the module is unloaded. I experimented with
my new module, opl3sa2, and I get similiar errors to what you report if I
leave out the call to sound_unload_mixerdev in my unload function.
I did also notice the problems you mentioned with being unable to unload
soundcore, and not seeing the mixer device after re-insmoding the sound
module. I think having the cs4232 module call sound_unload_mixerdev
will probably fix your immediate problem, but then again, I've been only
been hacking at the sound code for a few days and could be way out in
left field.
Alan, if you are reading this, I noticed that the ad1848 module does nasty
things if you insmod it with a bad irq (eg. 5) and then try to rmmod it.
The rmmod traps, and then trying to rmmod sound (or any other module) gives:
rmmod: QM_REFS: No such file or directory
As well, /proc/modules contains the line:
ad1848 14156 0 (deleted)
Is this a known problem? If not, I can try to track it down better.
Scott
--
=============================================================================
Scott Murray, scottm@interlog.com
-----------------------------------------------------------------------------
"Good, bad ... I'm the guy with the gun." - Ash, "Army of Darkness"
=============================================================================
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu