[PlanetCCRMA] Problem with virmidi, CCRMA, FCP4, 2.6.14-0.10.rrt

Geoff Matters geoff@mh2o.com
Tue Jan 10 16:49:03 2006 

I am using FCP4 with CCRMA and the planet-edge
2.6.14-0.10.rrt kernel and alsa modules @ 686.

Boots fine.  Kernel runs smoothly - i no longer have
problems with JACK failing to feed devices when other
high-priority tasks are temporarily hogging the CPU.

However there is a problem with the virmidi module.

modprobe -a snd-virmidi gives seg fault.

# tail /var/log/messages
Jan  7 20:57:33 localhost kernel:  [<c02671be>]
bus_add_device+0x2e/0xb0 (12)
Jan  7 20:57:33 localhost kernel:  [<c0266380>]
device_add+0xe0/0x130 (20)
Jan  7 20:57:33 localhost kernel:  [<c02690ad>]
platform_device_register+0x13d/0x190 (36)
Jan  7 20:57:33 localhost kernel:  [<df96f998>]
snd_generic_device_register+0x68/0xc0 [snd] (40)
Jan  7 20:57:33 localhost kernel:  [<df96fa38>]
snd_card_set_generic_dev+0x8/0x40 [snd] (28)
Jan  7 20:57:34 localhost kernel:  [<dfab5137>]
snd_card_virmidi_probe+0x137/0x1b0 [snd_virmidi] (8)
Jan  7 20:57:34 localhost kernel:  [<dfab51d2>]
alsa_card_virmidi_init+0x22/0x60 [snd_virmidi] (60)
Jan  7 20:57:34 localhost kernel:  [<c0142277>]
sys_init_module+0xa7/0x1e0 (16)
Jan  7 20:57:34 localhost kernel:  [<c01032c1>]
syscall_call+0x7/0xb (24)
Jan  7 20:57:34 localhost kernel: Code:  Bad EIP
value.

BUG: Unable to handle kernel NULL pointer dereference
at virtual address 00000000
 printing eip:
00000000
*pde = 00000000
Oops: 0000 [#1]
PREEMPT
Modules linked in: snd_virmidi(U) snd_seq_virmidi(U)
udf(U) i915(U) parport_pc(U) lp(U) parport(U)
autofs4(U) rfcomm(U) l2cap(U) bluetooth(U) sunrpc(U)
pcmcia(U) snd_usb_audio(U) snd_usb_lib(U)
ipt_REJECT(U) ipt_state(U) ip_conntrack(U)
nfnetlink(U) iptable_filter(U) ip_tables(U)
snd_hdsp(U) snd_rawmidi(U) snd_hwdep(U) video(U)
container(U) button(U) battery(U) ac(U) ipv6(U)
ohci1394(U) ieee1394(U) yenta_socket(U)
rsrc_nonstatic(U) pcmcia_core(U) uhci_hcd(U)
ehci_hcd(U) hw_random(U) tpm_atmel(U) tpm(U)
i2c_i801(U) i2c_core(U) snd_intel8x0m(U)
snd_intel8x0(U) snd_ac97_codec(U) snd_ac97_bus(U)
snd_seq_dummy(U) snd_seq_oss(U) snd_seq_midi_event(U)
snd_seq(U) snd_seq_device(U) snd_pcm_oss(U)
snd_mixer_oss(U) snd_pcm(U) snd_timer(U) snd(U)
soundcore(U) snd_page_alloc(U) ipw2200(U) ieee80211(U)
ieee80211_crypt(U) b44(U) mii(U) dm_snapshot(U)
dm_zero(U) dm_mirror(U) ext3(U) jbd(U) dm_mod(U)
CPU:    0
EIP:    0060:[<00000000>]    Not tainted VLI
EFLAGS: 00010286   (2.6.14-0.10.rrt.rhfc4.ccrma)
EIP is at _stext+0x3feffde0/0x40
eax: d369da14   ebx: 00000000   ecx: 00000008   edx:
df97d9f4
esi: df97d9f4   edi: d369da1c   ebp: c03cb940   esp:
cd415e7c
ds: 007b   es: 007b   ss: 0068   preempt: 00000001
Process modprobe (pid: 3519, threadinfo=cd414000
task=cdaba980 stack_left=7752 worst_left=-1)
Stack: df96e00c c026789e df97da90 cd415ea4 c0351afa
00000000 00000000 d369db4c
       d369da1c c026707c c03cbacc c03cbaec df97da8c
d369da1c d369db4c c03cb940
       00000000 c026799b c0267930 d369da1c d369da1c
c02671be d369da1c d369db38
Call Trace:
 [<df96e00c>] snd_platform_driver_probe+0xc/0x10 [snd]
(4)
 [<c026789e>] driver_probe_device+0x3e/0xd0 (4)
 [<c0351afa>] klist_next+0x4a/0x68 (12)
 [<c026707c>] bus_for_each_drv+0x3c/0x60 (20)
 [<c026799b>] device_attach+0x5b/0x80 (32)
 [<c0267930>] __device_attach+0x0/0x10 (4)
 [<c02671be>] bus_add_device+0x2e/0xb0 (12)
 [<c0266380>] device_add+0xe0/0x130 (20)
 [<c02690ad>] platform_device_register+0x13d/0x190
(36)
 [<df96f998>] snd_generic_device_register+0x68/0xc0
[snd] (40)
 [<df96fa38>] snd_card_set_generic_dev+0x8/0x40 [snd]
(28)
 [<dfab5137>] snd_card_virmidi_probe+0x137/0x1b0
[snd_virmidi] (8)
 [<dfab51d2>] alsa_card_virmidi_init+0x22/0x60
[snd_virmidi] (60)
 [<c0142277>] sys_init_module+0xa7/0x1e0 (16)
 [<c01032c1>] syscall_call+0x7/0xb (24)
Code:  Bad EIP value.


seems to be the same problem described here:
http://article.gmane.org/gmane.linux.alsa.devel/29267

and here:
http://www.mail-archive.com/blfs-support@linuxfromscratch.org/msg03641.html

seems his problems was solved by a patch:
http://www.mail-archive.com/blfs-support@linuxfromscratch.org/msg03698.html

but i can't find the post with the patch!

Any advice?  Any chance of new module package
including the mystery
patch referenced above?  I rely on virmidi, so without
a fix i'll be
stuck troubleshooting JACK performance on non-RT
kernel.

Many thanks!

  -geoff



__________________________________________________
Do You Yahoo!?
Tired of spam?  Yahoo! Mail has the best spam protection around 
http://mail.yahoo.com