USB DVB-T Tuner causes Kernel Oops

Thanks for taking the time to report this bug and helping to make Ubuntu better. With which specific version of the kernel did you notice this bug? You can check via 'cat /proc/version_signature'. Additionally is this an issue with the latest version of the kernel for Feisty? Thanks in advance.

I don't remember witch version specific version i had when i reported this bug, but i tried this device with after every kernel update and even with Gutsy Gibbon Tribe 3 (form the live-cd). This problem is still there.

cat /proc/version_signature
Ubuntu 2.6.20-16.29-generic

This particular bug doesn't meet the criteria for a stable release update for the 2.6.20 kernel. However, since you have tested it with the Live CD I have created a task for the 2.6.22 kernel. The information we have may be enough for the kernel team to look into but it could also be helpful if you included the oops when running the Gutsy Live CD. Thanks again!

Thank you.

I tried it now with my other Computer witch runs the latest Gusty Kernel:

Linux version 2.6.22-8-generic (buildd@yellow) (gcc version 4.1.3 20070629 (prerelease) (Ubuntu 4.1.2-13ubuntu2)) #1 SMP Thu Jul 12 16:09:47 GMT 2007

The oops message is:
Jul 31 19:38:06 tycoon-desktop kernel: [ 92.183187] usb 3-5: new high speed USB device using ehci_hcd and address 3
Jul 31 19:38:06 tycoon-desktop kernel: [ 92.249346] usb 3-5: configuration #1 chosen from 1 choice
Jul 31 19:38:06 tycoon-desktop kernel: [ 92.343203] dvb-usb: found a 'Hanftek UMT-010 DVB-T USB2.0' in cold state, will try to load a firmware
Jul 31 19:38:07 tycoon-desktop kernel: [ 92.357673] dvb-usb: downloading firmware from file 'dvb-usb-umt-010-02.fw'
Jul 31 19:38:07 tycoon-desktop kernel: [ 92.373875] usbcore: registered new interface driver dvb_usb_umt_010
Jul 31 19:38:07 tycoon-desktop kernel: [ 92.373999] usb 3-5: USB disconnect, address 3
Jul 31 19:38:07 tycoon-desktop kernel: [ 92.376952] dvb-usb: generic DVB-USB module successfully deinitialized and disconnected.
Jul 31 19:38:08 tycoon-desktop kernel: [ 93.251781] usb 3-5: new high speed USB device using ehci_hcd and address 4
Jul 31 19:38:08 tycoon-desktop kernel: [ 93.318127] usb 3-5: configuration #1 chosen from 1 choice
Jul 31 19:38:08 tycoon-desktop kernel: [ 93.318200] dvb-usb: found a 'Hanftek UMT-010 DVB-T USB2.0' in warm state.
Jul 31 19:38:08 tycoon-desktop kernel: [ 93.325716] dvb-usb: will pass the complete MPEG2 transport stream to the software demuxer.
Jul 31 19:38:08 tycoon-desktop kernel: [ 93.325807] DVB: registering new adapter (Hanftek UMT-010 DVB-T USB2.0).
Jul 31 19:38:09 tycoon-desktop kernel: [ 93.355695] DVB: registering frontend 0 (Zarlink MT352 DVB-T)...
Jul 31 19:38:09 tycoon-desktop kernel: [ 93.363648] dvb-usb: Hanftek UMT-010 DVB-T USB2.0 successfully initialized and connected.
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.150574] dvb-usb: could not submit URB no. 20 - get them all back
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151931] Unable to handle kernel paging request at 000000002ce21808 RIP:
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151934] [_end+137969672/2130332920] :dvb_usb_dibusb_common:dibusb_streaming_ctrl+0x10/0x50
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151942] PGD 1b88c067 PUD 1b88d067 PMD 0
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151946] Oops: 0000 [1] SMP
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151948] CPU 0
Jul 31 19:40:08 tycoon-desktop kernel: [ 158.151949] Modules linked in: mt352 dvb_usb_umt_010 dvb_usb_dibusb_common dib3000mc dibx000_common dvb_usb dvb_core dvb_pll binfmt_misc nfsd exportfs ipv6 capability rfcomm l2cap bluetooth nfs lockd sunrpc powernow_k8 cpufreq_conservative cpufreq_ondemand cpufreq_userspace cpufreq_powersave cpufreq_stats freq_table container button ac sbs video dock battery sbp2 lp fuse snd_intel8x0 snd_ac97_codec ac97_bus snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event nvidia(P) ide_cd cdrom snd_seq snd_timer snd_seq_device parport_pc parport k8temp pcspkr snd shpchp pci_hotplug soundcore snd_page_alloc i2c_nforce2 i2c_cor...

Hello,
I think I have the same problem here with my Easyone DVB-T Adapter (Clone of Hanftek UMT-010 USB2.0 DVB-T Pen Receiver).
When I scan for Channels Kaffeine freezes. I am using the kernel Ubuntu 2.6.20-16.29-generic
With the kernel 2.6.17.10 / Ubuntu 6.10 it works fine.
If you need more informations, please write me what I have to do, because I am a newbie with Linux and Ubuntu.

Thanks
Michael Rossol

I have the same problem with "Hanftek UMT-010 DVB-T USB2.0" controller and kernel 2.6.20-16-generic #2 SMP Sun Sep 23 19:50:39 UTC 2007 i686 GNU/Linux. The modules load fine, but if I start kaffeine for watching TV, the application crashes and the following errors occurs in /var/log/messages

[ 198.684000] dvb-usb: could not submit URB no. 20 - get them all back
[ 198.688000] BUG: unable to handle kernel paging request at virtual address 27767434
[ 198.688000] printing eip:
[ 198.688000] f88b0ee5
[ 198.688000] *pde = 00000000
[ 198.688000] Oops: 0000 [#1]
[ 198.688000] SMP
[ 198.688000] Modules linked in: mt352 dvb_usb_umt_010 dvb_usb_dibusb_common dib3000mc dibx000_common dvb_usb dvb_core dvb_pll binfmt_misc ipx p8023 uinput xt_tcpudp xt_state ipt_REJECT ipt_LOG xt_limit iptable_nat nf_nat nf_conntrack_ipv4 nf_conntrack nfnetlink iptable_filter ip_tables x_tables af_packet i915 ppdev acpi_cpufreq cpufreq_userspace tc1100_wmi pcc_acpi dev_acpi sony_acpi video sbs ibm_acpi i2c_ec dock button battery container ac asus_acpi backlight dm_mod cmtp kernelcapi bnep l2cap sco bluetooth cpuid msr cpufreq_stats cpufreq_conservative cpufreq_ondemand cpufreq_powersave freq_table eeprom i2c_i801 i2c_core i810 drm nvram parport_pc lp parport snd_hda_intel snd_hda_codec snd_pcm_oss snd_mixer_oss snd_pcm snd_seq_dummy snd_seq_oss snd_seq_midi snd_rawmidi snd_seq_midi_event ipw3945 pcmcia snd_seq snd_timer snd_seq_device psmouse ieee80211 snd serio_raw pcspkr ieee80211_crypt soundcore snd_page_alloc yenta_socket rsrc_nonstatic pcmcia_core iTCO_wdt iTCO_vendor_support intel_agp agpgart shpchp pci_hotplug evdev tsdev ipv6 ext3 jbd mbcache sg sd_mod usbhid hid ata_piix ata_generic ahci libata scsi_mod generic tg3 uhci_hcd ehci_hcd usbcore ide_cd thermal processor fan fbcon tileblit font bitblit softcursor vesafb capability commoncap cdrom ide_generic
[ 198.688000] CPU: 0
[ 198.688000] EIP: 0060:[<f88b0ee5>] Not tainted VLI
[ 198.688000] EFLAGS: 00010206 (2.6.20-16-generic #2)
[ 198.688000] EIP is at usb_kill_urb+0x15/0x100 [usbcore]
[ 198.688000] eax: 00000000 ebx: 27767414 ecx: 00000293 edx: d5c1e000
[ 198.688000] esi: e75b4b3c edi: ffffffea ebp: 00000001 esp: d5c1fde8
[ 198.688000] ds: 007b es: 007b ss: 0068
[ 198.688000] Process kaffeine (pid: 6356, ti=d5c1e000 task=d6d50030 task.ti=d5c1e000)
[ 198.688000] Stack: 00000000 d6d50030 c013ae00 d5c1fdf4 d5c1fdf4 00000021 e75b4b3c ffffffea
[ 198.688000] f8ed4b8c e75b4b3c 00000014 f8ed4c55 f8ed5668 00000014 e75b480c 00000001
[ 198.688000] f8f04000 f8ed4331 c02edaa8 e75b48e8 f8f04000 fffffe00 e75b4ac0 f8ee720e
[ 198.688000] Call Trace:
[ 198.688000] [<c013ae00>] autoremove_wake_function+0x0/0x50
[ 198.688000] [<f8ed4b8c>] usb_urb_kill+0x1c/0x40 [dvb_usb]
[ 198.688000] [<f8ed4c55>] usb_urb_submit+0x55/0x60 [dvb_usb]
[ 198.688000] [<f8ed4331>] dvb_usb_ctrl_feed+0x91/0x120 [dvb_usb]
[ 198.688000] [<c02edaa8>] mutex_lock_interruptible+0x8/0x20
[ 198.688000] [<f8ee720e>] dmx_ts_feed_start_filtering+0x4e/0xe0 [dvb_core]
[ 198.688000] [<f8ee5280>] dvb_dmxdev_filter_start+0x290/0x3e0 [dvb_core]
[ 198.688000] [<f8ee5508>] dvb_demux_do_ioctl+0x138/0x3b0 [dvb_core]
[ 198.688000] [<f8ee40e4>] dvb_usercopy+0x64/0x140 [dvb_core]
[...

I came across this bug after I bought a Banksia dvb usb tuner.

It seems that a major rewrite of the kernel code to handle multiple streams which appeared in 2.6.18 (19?) broke the driver.

For some reason the max number of URBs is set to 10, but these devices are somehow set for 20. I don't understand why this is the case.

Try a custom kernel changing drivers/media/dvb/dvb-usb/dvb-usb.h from:

#define MAX_NO_URBS_FOR_DATA_STREAM 10

to

#define MAX_NO_URBS_FOR_DATA_STREAM 20

This has cleared many problems for me.

It also seems that the stream->urbs_initialized variable which counts the number of URBs initialized is not explicitly preset to zero. Changing this would also be advised.

It seems that the other way to do it is to set the .count parameter in umt-010.c (in the same directory as above) to 10. But I'm not sure if this is the right thing to do here, We just need the input from someone who knows a tiny bit about what is going on.

That's exciting news.
Though this information is very likely to get lost in the whole bunch of bugreports here.
Consider posting this information to the linuxtv-mailinglist of the kernel-bugzilla (IIRC you can report dvb-bugs there, now).

I have posed to the kernel-bugzilla. Subscribing to a mailing list with rather high volume is not my idea of fun considering the email bloat I'm already working with.

Can you post the link to the kernel bugzilla report you made? We can then have this report monitor that one. Thanks.

I assume this issue still exists in the latest Hardy Alpha releases but just in case you'd like to test. . . The Hardy Heron Alpha2 and subsequent releases contain an updated version of the kernel. You can download and try the new Hardy Heron Alpha release from http://cdimage.ubuntu.com/releases/hardy/ . Please note that Alpha3 will be released within the next day or two so you may want to wait. You should be able to then test the new kernel via the LiveCD. General information regarding the release can also be found here: http://www.ubuntu.com/testing/ .

I've opened a new task against the actively developed kernel and am closing the task against linux-source-2.6.22. Thanks!

The link to the request is:

http://bugzilla.kernel.org/show_bug.cgi?id=9720

Tested against Hardy alpha3. Same issue. (Not surprising as the code hasn't changed).

However, I find that the /var/log/messages leads one into a false sense of security. It seems to install OK, but trying to rmmod it kills it. There doesn't seem to be anything I can test DVB on with the live CD.

I have attached the dump from the oops from Hardy alpha 3. It is odd as it seems to be using the ide_disk code?!? Anyway, I guess that's what happens when one works with memory addresses one shouldn't (?).

is there a chance that we can at least get the MAX_NO_URBS_FOR_DATA_STREAM value updated in one of the next builds ?

From my understanding this would rather be a workround than a real fix., if setting the MAX_NO_URBS_FOR_DATA_STREAM to 10 was done for a special reason. I'm not a kernel-hacker, though. Maybe someone from the kernel team can take a look at it and I#m pretty sure that someone has a spare machine for which he can provide ssh-access for testing any changings.

thats why i'm proposing it now, i suspect it will work fine even with 20, so switching the value before release to get feedback and be able to revert it in case it doesnt work for the final release would likely be helpful :)

I recently bought a Hauppauge WinTV Nova-T (2040:7070), and since the Hardy kernel does not have recent enough v4l modules, I'm using the v4l upstream hg head ones (they build and work fine on the Hardy kernel). I have used the modules which were built with MAX_NO_URBS_FOR_DATA_STREAM=10 for some days (well, just maybe half an hour per day) without any problems, let alone crashes. However, I'll rebuild the modules with "20" and will test them.

Seems to work for me with 20 as well. I leave it running for some hours.

http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-hardy.git;a=commit;h=15826b4c16e487cedf83482aa46867d1ffb4d799

This same driver exists in 2.6.25 with the same problem (which I've patched). Can someone try the 2.6.25 Kernel/LUM combination from my PPA https://edge.launchpad.net/~timg-tpi/+archive

SRU Justification:

Impact: kernel oops

Fix Description: Use the correct macro MAX_NO_URBS_FOR_DATA_STREA in structure init.

Patch: http://kernel.ubuntu.com/git?p=ubuntu/ubuntu-hardy.git;a=commit;h=15826b4c16e487cedf83482aa46867d1ffb4d799

TEST CASE: see Bug Description

Hello,
thank you very much for help fixing this bug !

Comes the bug fix with one of the next Ubuntu kernel updates? Because I don't know how to compile a new kernel. I am a Newbie with Linux.

Regards
Michael Rossol

"Can someone try the 2.6.25 Kernel/LUM combination from my PPA https://edge.launchpad.net/~timg-tpi/+archive"

I don't see a 2.6.25 version in your PPA.

I only see linux - 2.6.24-19.33ubuntu8. Is this what you mean?

Although not documented in the changelog, the change for this bug appears to be present in the diff for the -19 kernel currently in hardy-proposed. Please test whether this kernel corrects the problem for you and give feedback here.

Released in linux-2.6.24-19.33 but for some reason did not get noted in the changelog.

The hardy-proposed kernel linux-2.6.24-19.33 doesn't crash for me.

Until I remove the device :)

Modules linked in: dvb_pll mt352 dvb_usb_umt_010 dvb_usb_dibusb_common dib3000mc dibx000_common dvb_usb dvb_core wlan_tkip wlan_ccmp af_packet binfmt_misc ppdev ipv6 powernow_k8 cpufreq_userspace cpufreq_conservative cpufreq_powersave cpufreq_stats cpufreq_ondemand freq_table sbs dock sbshc container video output battery iptable_filter ip_tables x_tables ac sbp2 lp snd_hda_intel snd_pcm_oss snd_mixer_oss snd_pcm snd_page_alloc snd_hwdep snd_seq_dummy snd_seq_oss snd_seq_midi wlan_scan_sta ath_rate_sample snd_rawmidi snd_seq_midi_event snd_seq snd_timer snd_seq_device nvidia(P) serio_raw ath_pci wlan ath_hal(P) snd i2c_nforce2 psmouse k8temp parport_pc soundcore button i2c_core parport analog evdev shpchp pci_hotplug pcspkr gameport ext3 jbd mbcache sg sd_mod sr_mod cdrom ata_generic pata_acpi usbhid hid pata_amd sata_nv ohci1394 ieee1394 libata scsi_mod ehci_hcd ohci_hcd usbcore thermal processor fan fbcon tileblit font bitblit softcursor fuse
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468052] Pid: 1630, comm: khubd Tainted: P 2.6.24-18-generic #1
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468054] RIP: 0010:[sr_mod:kref_put+0x2e/0x80] [sr_mod:kref_put+0x2e/0x80] kref_put+0x2e/0x80
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468057] RSP: 0018:ffff810037959c90 EFLAGS: 00010206
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468059] RAX: 0000000000000020 RBX: 000000003c32c414 RCX: ffffffff8048edd0
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468061] RDX: 0000000000000036 RSI: ffffffff88044b50 RDI: 000000003c32c414
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468063] RBP: ffffffff88044b50 R08: 0000000000000000 R09: ffff81003d749e80
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468065] R10: 0000000000000000 R11: 0000000000000675 R12: ffff810024cfc000
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468067] R13: 0000000000000000 R14: 0000000000000000 R15: 0000000000000006
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468070] FS: 00007f6a66eb16f0(0000) GS:ffffffff805b9000(0000) knlGS:00000000556c56b0
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468072] CS: 0010 DS: 0018 ES: 0018 CR0: 000000008005003b
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468074] CR2: 000000003c32c414 CR3: 0000000000201000 CR4: 00000000000006e0
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468075] DR0: 0000000000000000 DR1: 0000000000000000 DR2: 0000000000000000
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468077] DR3: 0000000000000000 DR6: 00000000ffff0ff0 DR7: 0000000000000400
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468080] Process khubd (pid: 1630, threadinfo ffff810037958000, task ffff81003b30f7a0)
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468081] Stack: ffff810024cfcdb0 0000000000000020 ffff810024cfd218 ffffffff88cd7bef
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468086] ffff810024cfcd48 ffff810024cfcd48 ffff810024cfc000 ffffffff88cd63eb
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468089] 0000000000000100 ffffffff88cea366 ffff8100220f1800 ffff810024e42c00
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468092] Call Trace:
Jun 6 18:21:19 lund-desktop-2 kernel: [ 621.468101] [<ffffffff88cd7bef>] :dvb_usb:us...

Oh damm. I forgot to reboot. Sorry about the noise.

Works fine.

Sorry, but it's to hard for me ;) ,
where can I find the package "linux-2.6.24-19.33" ? I can't find it in the Adept Manager (I am using kubuntu 8.04). Can someone give me the deb file for the Sources.list ?

So, I hope that I got the right kernel. But X will not start. Here a cut from the log:
(II) Setting vga for screen 0.
(**) NVIDIA(0): Depth 24, (--) framebuffer bpp 32
(==) NVIDIA(0): RGB weight 888
(==) NVIDIA(0): Default visual is TrueColor
(==) NVIDIA(0): Using gamma correction (1.0, 1.0, 1.0)
(**) NVIDIA(0): Option "TVStandard" "PAL-D"
(**) NVIDIA(0): Option "TVOutFormat" "Composite"
(**) NVIDIA(0): Option "TwinView" "1"
(**) NVIDIA(0): Option "MetaModes" "TV: 1024x768 +0+768, DFP: nvidia-auto-select +0+0; TV: 800x600 +0+0; TV: 400x300 +0+0; TV: 320x240 +0+0; TV: 720x480 +0+0"
(**) NVIDIA(0): Enabling RENDER acceleration
(**) NVIDIA(0): Forcing COMPOSITE video output
(**) NVIDIA(0): TV Standard string: "PAL-D"
(**) NVIDIA(0): TwinView enabled
(II) NVIDIA(0): Support for GLX with the Damage and Composite X extensions is
(II) NVIDIA(0): enabled.
(EE) NVIDIA(0): Failed to load the NVIDIA kernel module!
(EE) NVIDIA(0): *** Aborting ***
(II) UnloadModule: "nvidia"
(II) UnloadModule: "xaa"
(II) UnloadModule: "fb"
(EE) Screen(s) found, but none have a usable configuration.

Fatal server error:
no screens found

with the Kernel 2.6.24-18-generic I have no problems with X server but no TV via USB ;)

Michael,

On Fri, Jun 06, 2008 at 07:36:28PM -0000, Michael Rossol wrote:
> So, I hope that I got the right kernel. But X will not start. Here a cut from the log:

The error message here seems self-explanatory to me:
> (EE) NVIDIA(0): Failed to load the NVIDIA kernel module!
> (EE) NVIDIA(0): *** Aborting ***

Do you have the corresponding linux-restricted-modules-2.6.24-19-$flavor
package installed? Since the linux metapackages have only been published to
-proposed today, you probably have to install the restricted modules by
hand.

Hello Steve,
your are right.
I have installed the restricted module and the ubuntu-modules (my audio was not running!). Now everything seems to be running. The TV scan was ok.

Thank you all very much for help.

Now I can see the European football championship on the terrace ;)

"Now I can see the European football championship on the terrace ;)"

Sometimes I love fixing bugs.

This package has been copied to hardy-updates.

Per a decision made by the Ubuntu Kernel Team, bugs will longer be assigned to the ubuntu-kernel-team in Launchpad as part of the bug triage process. The ubuntu-kernel-team is being unassigned from this bug report. Refer to https://wiki.ubuntu.com/KernelTeamBugPolicies for more information. Thanks.