I have the OSS driver for my sb16 loaded (ALSA didn't work in some
kernel a while ago so I switched to OSS and forgot to switch back)
I just performed an upgrade of my Debian system and it tried to activate
ALSA and thus started loading the modules. Kernel is 2.5.69-mm5. This is
what I got:
Unable to handle kernel paging request at virtual address f4a52958
printing eip:
eeaba090
*pde = 00000000
Oops: 0002 [#1]
CPU: 0
EIP: 0060:[<eeaba090>] Not tainted VLI
EFLAGS: 00010202
EIP is at snd_minor_search+0x18/0x4c [snd]
eax: eeac6ea8 ebx: 00000021 ecx: eeac6780 edx: eeac6ea8
esi: eeac6780 edi: c8021f8c ebp: eb845f6c esp: eb845f68
ds: 007b es: 007b ss: 0068
Process modprobe (pid: 10236, threadinfo=eb844000 task=e0c79b20)
Stack: 00000021 eb845f88 eeaba318 00000021 d21a8dd8 c032bda0 c032bd60 c8021f70
eb845fa8 ec88f08a 00000021 00000000 00000000 eeae07b8 eeade877 eeae07e0
eb845fbc c0135abd 0804f258 00000002 00000000 eb844000 c0109853 0804f258
Call Trace:
[<eeaba318>] snd_register_device+0x98/0x10c [snd]
[<ec88f08a>] +0x8a/0xa5 [snd_timer]
[<eeae07b8>] snd_timer_reg+0x0/0x1c [snd_timer]
[<eeade877>] +0x377/0x63e [snd_timer]
[<eeae07e0>] +0x0/0xe0 [snd_timer]
[<c0135abd>] sys_init_module+0x201/0x2d8
[<c0109853>] syscall_call+0x7/0xb
Code: eb 05 b8 97 03 ac ee 50 e8 ee 23 67 d1 89 ec 5d c3 89 f6 55 89 e5 53 8b 5d 08 89 d8 c1 f8 05 8b 14 c5 a0 6e ac ee 8b 02 0f 18 00 <00> 89 d8 c1 f8 05 8d 04 c5 a0 6e ac ee 39 c2 74 1b 89 c1 39 5a
It was probably snd_timer it failed on, The error scrolled off the
screen quite fast. lsmod says snd_timer is loaded but the module-loader
complained it couldn't find some symbols from snd_timer:
snd_pcm: Unknown symbol snd_timer_notify
snd_pcm: Unknown symbol snd_timer_interrupt
snd_pcm: Unknown symbol snd_timer_new
and later:
snd_seq: Unknown symbol snd_timer_stop
snd_seq: Unknown symbol snd_timer_close
snd_seq: Unknown symbol snd_timer_open
snd_seq: Unknown symbol snd_timer_start
snd_seq: Unknown symbol snd_timer_resolution
snd_seq: Unknown symbol snd_timer_pause
If those symbols would have been found snd_pcm would have been tried for
loading and would probably have Oopsed again because of the previous
error.
Yup, cat /proc/modules tells us which module it was:
snd_timer 27109 1 - Loading 0xeeada000
I havn't tried reproducing it, I havn't tried it without OSS modules
loaded either. The OSS modules loaded when this happened was:
sb 10368 2
sb_lib 53680 1 sb
uart401 10756 1 sb_lib
sound 90908 4 sb_lib,uart401
-- /Martin - To unsubscribe from this list: send the line "unsubscribe linux-kernel" in the body of a message to majordomo@vger.kernel.org More majordomo info at http://vger.kernel.org/majordomo-info.html Please read the FAQ at http://www.tux.org/lkml/