BUG when unplugging EyeTV

Hello,

I'm hitting a kernel BUG when unplugging my EyeTV on a MacBook Pro. According to the OS X software, it has an XC5000 tuner, AU8522 demodulator, and AU0828 controller; I hear that it is the same as the Hauppage WinTV-HVR-950Q.

I'm running Debian sid with a locally-built 2.6.29.1 kernel. The only modifications from the kernel.org vanilla kernel are the addition of v4l-dvb drivers grabbed from hg today and wireless-testing also current as of today.

The text of the BUG is pasted below; the BUG is consistently reproducible either if I unplug the device or if I try to rmmod modules (only did this once and not sure rmmod'ing which dvb module actually caused it). I have unfortunately not checked to see how similar the BUG reports end up over different trials.

If you need more context, see
http://www.contrib.andrew.cmu.edu/~jwatzman/eyetv/ for various parts of dmesg. "1" was taken right after plugging it in; "2" after waiting a few moments for the firmware to upload; "3" after successfully watching TV for about an hour; and "4" after unplugging the device.

Let me know if you need more information!
Thanks,
Josh Watzman

[45664.805473] usb 1-3: USB disconnect, address 8
[45664.806201] BUG: unable to handle kernel NULL pointer dereference at
0000000000000008
[45664.806209] IP: [<ffffffff802506bb>] prepare_to_wait+0x29/0x58
[45664.806222] PGD 7c9ed067 PUD 7c970067 PMD 0
[45664.806229] Oops: 0002 [#1] SMP
[45664.806234] last sysfs file: /sys/devices/platform/applesmc.768/light
[45664.806238] CPU 0
[45664.806241] Modules linked in: xc5000 tuner au8522 snd_usb_audio

snd_usb_lib snd_hwdep au0828 dvb_core videobuf_vmalloc videobuf_core tveeprom v4l2_common radeon drm uvcvideo videodev v4l1_compat v4l2_compat_ioctl32 ipv6 binfmt_misc cpufreq_conservative cpufreq_userspace dm_mod cpufreq_stats cpufreq_powersave kvm_intel kvm fuse cpufreq_ondemand acpi_cpufreq freq_table loop firewire_sbp2 snd_hda_codec_idt snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy arc4 snd_seq_oss ecb snd_seq_midi snd_rawmidi ath9k mac80211 snd_seq_midi_event snd_seq rfkill snd_timer snd_seq_device joydev rtc_cmos snd video i2c_i801 rtc_core cfg80211 applesmc appletouch soundcore snd_page_alloc rng_core i2c_core rtc_lib led_class ac battery pcspkr button output evdev input_polldev ext3 jbd mbcache ide_cd_mod cdrom sd_mod ata_generic hid_apple ata_piix libata scsi_mod ide_pci_generic firewire_ohci firewire_core piix crc_itu_t ide_core uhci_hcd ehci_hcd sky2 intel_agp thermal processor fan usbhid hid

[45664.806377] Pid: 151, comm: khubd Not tainted 2.6.29.1 #1 MacBookPro2,2
[45664.806382] RIP: 0010:[<ffffffff802506bb>]  [<ffffffff802506bb>]
prepare_to_wait+0x29/0x58
[45664.806391] RSP: 0018:ffff88007e0d3c00  EFLAGS: 00010046
[45664.806395] RAX: 0000000000000000 RBX: ffff88007e0d3c20 RCX:
0000000000000000
[45664.806400] RDX: ffff88007e0d3c38 RSI: 0000000000000246 RDI:
ffffffff8053b8a8
[45664.806404] RBP: ffffffff8053b8a8 R08: 0000000000000000 R09:
0000000000002451
[45664.806409] R10: 0000000000000000 R11: ffff880000000000 R12:
0000000000000002
[45664.806413] R13: ffff88004c890000 R14: ffffffffa047f6b8 R15:
0000000000000000
[45664.806418] FS:  0000000000000000(0000) GS:ffffffff80620000(0000)
knlGS:0000000000000000
[45664.806423] CS:  0010 DS: 0018 ES: 0018 CR0: 000000008005003b
[45664.806428] CR2: 0000000000000008 CR3: 0000000037841000 CR4:
00000000000026e0
[45664.806432] DR0: 0000000000000000 DR1: 0000000000000000 DR2:
0000000000000000
[45664.806437] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7:
0000000000000400
[45664.806442] Process khubd (pid: 151, threadinfo ffff88007e0d2000,
task ffff88007e06f990)
[45664.806445] Stack:
[45664.806448]  ffff880074188190 ffff880074188528 ffff880037909c00
ffffffffa0468ef5
[45664.806456]  0000000000000000 ffff88007e06f990 ffffffff8025050a
ffff88007e0d3c38
[45664.806464]  ffff88007e0d3c38 ffff88007e0d3c60 ffff880000000000
ffff880074188190
[45664.806473] Call Trace:
[45664.806477]  [<ffffffffa0468ef5>] ? dvb_net_release+0x60/0xab [dvb_core]
[45664.806502]  [<ffffffff8025050a>] ? autoremove_wake_function+0x0/0x2e
[45664.806510]  [<ffffffffa0478962>] ? au0828_dvb_unregister+0x44/0xa6
[au0828]
[45664.806524]  [<ffffffffa0477036>] ? au0828_usb_disconnect+0x36/0x86
[au0828]
[45664.806537]  [<ffffffff803b24de>] ? usb_unbind_interface+0x5e/0xe5
[45664.806547]  [<ffffffff803a2148>] ? __device_release_driver+0x83/0xa6
[45664.806555]  [<ffffffff803a2243>] ? device_release_driver+0x21/0x2d
[45664.806561]  [<ffffffff803a1874>] ? bus_remove_device+0xa8/0xca
[45664.806567]  [<ffffffff803a0233>] ? device_del+0x132/0x16c
[45664.806574]  [<ffffffff803afd77>] ? usb_disable_device+0x7d/0xf4
[45664.806581]  [<ffffffff803aba24>] ? usb_disconnect+0x89/0x10e
[45664.806588]  [<ffffffff803ac971>] ? hub_thread+0x663/0x1066
[45664.806594]  [<ffffffff8020a6c9>] ? __switch_to+0xb4/0x399
[45664.806602]  [<ffffffff8025050a>] ? autoremove_wake_function+0x0/0x2e
[45664.806609]  [<ffffffff803ac30e>] ? hub_thread+0x0/0x1066
[45664.806615]  [<ffffffff80250197>] ? kthread+0x47/0x73
[45664.806621]  [<ffffffff8020d2da>] ? child_rip+0xa/0x20
[45664.806628]  [<ffffffff8044d912>] ? _spin_lock+0x5/0x7
[45664.806636]  [<ffffffff80250150>] ? kthread+0x0/0x73
[45664.806642]  [<ffffffff8020d2d0>] ? child_rip+0x0/0x20
[45664.806648] Code: 1f 00 41 54 41 89 d4 55 48 89 fd 53 48 89 f3 83 26
fe e8 a1 d1 1f 00 48 8b 53 18 48 89 c6 48 8d 43 18 48 39 c2 75 18 48 8b
45 08 <48> 89 50 08 48 89 43 18 48 8d 45 08 48 89 55 08 48 89 43 20 65
[45664.806709] RIP  [<ffffffff802506bb>] prepare_to_wait+0x29/0x58
[45664.806717]  RSP <ffff88007e0d3c00>
[45664.806720] CR2: 0000000000000008
[45664.806725] ---[ end trace 28396efb33fe235a ]---

Hello Josh,

Thanks for the bug report. Robert Krakora reported the same stack trace to me off-list over the weekend. I've been tied up with some family business, but I intend to dig into it deeper this weekend.

I didn't know that Elgato had a 950q clone (I did work on the original Elgato EyeTV device for Linux). Could you please send me the output of "lsusb -v" so I can confirm precisely which device it is a clone of?

Regards,

Devin

Devin J. Heitmueller
http://www.devinheitmueller.com
AIM: devinheitmueller

No problem, take your time. I was quite pleased when the DVB worked on the stock kernel and even more pleased when the analog worked with the current hg tree. Thank you all for your work on these drivers.

I'm not 100% sure it's a clone of that card, but everything seems to match. The instructions for the 950q firmware on the wiki are what originally allowed me to get it working.

The lsusb is quite long -- you can get it at
http://www.contrib.andrew.cmu.edu/~jwatzman/eyetv/lsusb and I've also uploaded a screenshot of what the OS X software says the device is at
http://www.contrib.andrew.cmu.edu/~jwatzman/eyetv/eyetv.png if that helps too.

Josh Watzman