With a modularized opl3sa1 sound card setup, it looks like soundcore.o's
MOD_USE_COUNT is never fully decremented, even when all it's child modules
are (auto or manually) unloaded, and the next attempt at autoloading the
sound modules causes an oops.
Relevant /etc/conf.modules:
---------------------------
alias block-major-14 sound
alias sound opl3sa
post-install opl3sa /usr/local/bin/aumix -m 0 -x 0 -w 70 -s 32 -i 4 -c 20
options opl3sa io=0x530 irq=11 dma=0 dma2=1 mpu_io=0x330 mpu_irq=5
options opl3 io=0x388
Relevant .config:
-----------------
CONFIG_MODULES=y
CONFIG_MODVERSIONS=y
CONFIG_KMOD=y
#
# Sound
#
CONFIG_SOUND=m
CONFIG_SOUND_OSS=m
CONFIG_SOUND_OPL3SA1=m
CONFIG_SOUND_SOFTOSS=m
CONFIG_SOUND_YM3812=m
CONFIG_SOUND_VMIDI=m
Script started on Thu Dec 24 20:29:05 1998
iniquity:~# lsmod
Module Size Used by
isofs 17544 1 (autoclean)
ide-cd 20340 1 (autoclean)
cdrom 11180 0 (autoclean) [ide-cd]
opl3sa 1944 0 (autoclean)
uart401 5624 0 (autoclean) [opl3sa]
ad1848 14220 0 (autoclean) [opl3sa]
sound 56152 0 (autoclean) [opl3sa uart401 ad1848]
soundcore 1948 6 (autoclean) [sound]
nls_iso8859-1 2016 2 (autoclean)
nls_cp437 3540 1 (autoclean)
vfat 13652 1 (autoclean)
fat 24472 1 (autoclean) [vfat]
iniquity:~# rmmod -a
iniquity:~# rmmod -a
iniquity:~# lsmod
Module Size Used by
isofs 17544 1 (autoclean)
ide-cd 20340 1 (autoclean)
cdrom 11180 0 (autoclean) [ide-cd]
soundcore 1948 1 (autoclean)
nls_iso8859-1 2016 2 (autoclean)
nls_cp437 3540 1 (autoclean)
vfat 13652 1 (autoclean)
fat 24472 1 (autoclean) [vfat]
iniquity:~# echo "" >/dev/dsp
(oopsen below)
Unable to handle kernel paging request at virtual address c382916c
current->tss.cr3 = 02652000, %cr3 = 02652000
*pde = 0009c063
Oops: 0000
CPU: 0
EIP: 0010:[<c381b2a2>]
EFLAGS: 00010246
eax: c3829150 ebx: c26db840 ecx: 00000003 edx: 00000003
esi: c01ab860 edi: c01ab860 ebp: c263d8a0 esp: c2655f54
ds: 0018 es: 0018 ss: 0018
Process bash (pid: 95, process nr: 13, stackpage=c2655000)
Stack: c0123e5e c01ab860 c26db840 c26db840 00000000 c01ab860 c0122bfb c01ab860
c26db840 00000003 00000003 c25d4000 000001b6 c0122c81 c25d4000 00000241
0122d8d c25d4000 00000241 000001b6 00000003 c2654000
Call Trace: [<c0123e5e>] [<c0122bfb>] [<c0122c81>] [<c0122d8d>] [<c0108804>]
Code: 8b 40 1c 85 c0 74 0b 53 56 ff d0 83 c4 08 5b 5e c3 90 31 c0
Warning: trailing garbage ignored on Code: line
Text: 'Code: 8b 40 1c 85 c0 74 0b 53 56 ff d0 83 c4 08 5b 5e c3 90 31 c0'
Garbage: 'c0'
>>EIP: c381b2a2 <soundcore_open+42/6c>
Trace: c0123e5e <chrdev_open+3e/4c>
Trace: c0122bfb <filp_open+af/11c>
Trace: c0122c81 <do_open+19/44>
Trace: c0122d8d <sys_open+39/68>
Trace: c0108804 <system_call+34/40>
Code: c381b2a2 <soundcore_open+42/6c> 00000000 <_EIP>:
Code: c381b2a2 <soundcore_open+42/6c> 0: 8b 40 1c movl 0x1c(%eax),%eax
Code: c381b2a5 <soundcore_open+45/6c> 3: 85 c0 testl %eax,%eax
Code: c381b2a7 <soundcore_open+47/6c> 5: 74 0b je 12 <_EIP+0x12> c381b2b4 <soundcore_open+54/6c>
Code: c381b2a9 <soundcore_open+49/6c> 7: 53 pushl %ebx
Code: c381b2aa <soundcore_open+4a/6c> 8: 56 pushl %esi
Code: c381b2ab <soundcore_open+4b/6c> 9: ff d0 call *%eax
Code: c381b2ad <soundcore_open+4d/6c> b: 83 c4 08 addl $0x8,%esp
Code: c381b2b0 <soundcore_open+50/6c> e: 5b popl %ebx
Code: c381b2b1 <soundcore_open+51/6c> f: 5e popl %esi
Code: c381b2b2 <soundcore_open+52/6c> 10: c3 ret
Code: c381b2b3 <soundcore_open+53/6c> 11: 90 nop
Code: c381b2b4 <soundcore_open+54/6c> 12: 31 00 xorl %eax,(%eax)
Code: c381b2b6 <soundcore_open+56/6c> 14:
The same behavior occurs when all the appropriate modules are loaded by
hand: modprobe opl3sa io=xxx irq=xxx, etc. (i.e. soundcore.o never
unloads, next access oopses)
Bummer, it stinks to find stuff like this so close to 2.2 :(
Chris Noe
(stiker@northlink.com)
--
---------------------------------------------------------
#!/bin/perl -sp0777i<X+d*lMLa^*lN%0]dsXx++lMlN/dsM0<j]dsj
$/=unpack('H*',$_);$_=`echo 16dio\U$k"SK$/SM$n\EsN0p[lN*1
lK[d2%Sa2/d0$^Ixp"|dc`;s/\W//g;$_=pack('H*',/((..)*)$/)
---- http://online.offshore.com.ai/arms-trafficker/ -----
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@vger.rutgers.edu
Please read the FAQ at http://www.tux.org/lkml/