Can't load kernel via deadloop bug in bttv module

Bug #271061 reported by troosh
8
Affects Status Importance Assigned to Milestone
linux (Ubuntu)
Incomplete
Undecided
Unassigned

Bug Description

My DVB-S PCI card "PixelView DTV2000" (based on Bt878 chip, Twinhan clone) freezes the system when loading kernel. Without option "quiet splash" kernel print:

"[ ] bttv: driver version 0.9.17 load
 [ ] bttv: using 8 buffers with 2080k (520 pages) each for capture
 [ ] bttv: Bt8xx card found (0)"

And kernel freeze on this point... :-(((

Only if rename the file /lib/modules/2.6.24-19-generic/kernel/drivers/media/video/bt8xx.ko, I can load system to run lspci with my "PixelView DTV2000" card.

# uname -a
Linux troosh-desktop 2.6.24-19-generic #1 SMP Wed Aug 20 22:56:21 UTC 2008 i686 GNU/Linux
01:01.0 Multimedia video controller [0400]: Brooktree Corporation Bt878 Video Capture [109e:036e] (rev 11)
01:01.1 Multimedia controller [0480]: Brooktree Corporation Bt878 Audio Capture [109e:0878] (rev 11)

Deadloop is bad choice if kernel module can't detect card. Why can't check timeout in kernel module?
Possible also need patch from http://osdir.com/ml/suse.amd64/2005-04/msg00132.html

P.S. I'm test with Ubuntu 8.10 "Intrepid Ibex" Development Releases Alpha 5 (2.6.27 kernel). Same problem, like in many others linux distribution... :-(
Last messages from kernel:

" [ ] bttv: Bt8xx card found (0)
  [ ] bttv 0000:01:01.0: PCI INTAA -> Link [LNK2] -> GSI 18 (level, high) -> IRQ 18
  [ ] bttv0: Bt878 (rev 17) at 0000:01:01.0, irq 18, latency : 32, mmio: 0xdc000000
  [ ] bttv0: using *** UNKNOW/GENERIC *** [card=0, autodetected]"

Tags: bt878 bttv dvb
Revision history for this message
troosh (trush) wrote :
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

Hi Troosh,

I'm curious if you blacklist the module so that you can boot without hanging, I assume if you then manually load the module it hangs? Would you then be able to capture the entire dmesg output and attach it here? Thanks.

Changed in linux:
status: New → Incomplete
Revision history for this message
Leann Ogasawara (leannogasawara) wrote :

I also forgot to ask if you could do this with the 2.6.27 kernel that would be great. Thanks.

Revision history for this message
troosh (trush) wrote :

# cat /proc/version
Linux version 2.6.24-19-generic (buildd@terranova) (gcc version 4.2.3 (Ubuntu 4.2.3-2ubuntu7)) #1 SMP Wed Aug 20 22:56:21 UTC 2008

# cat >> /etc/modprobe.d/options
options bttv i2c_hw=1 card=0x71 debug=1
options dvb_core dvb_shutdown_timeout=0

# dmesg -c >/dev/null
# modprobe bttv
# dmesg
[ 288.275725] bttv: driver version 0.9.17 loaded
[ 288.275735] bttv: using 8 buffers with 2080k (520 pages) each for capture
[ 288.275834] bttv: Bt8xx card found (0).
[ 288.275903] ACPI: PCI Interrupt 0000:01:01.0[A] -> Link [LNK2] -> GSI 18 (level, high) -> IRQ 21
[ 288.275919] bttv0: Bt878 (rev 17) at 0000:01:01.0, irq: 21, latency: 32, mmio: 0xdc000000
[ 288.277020] bttv0: using: Twinhan DST + clones [card=113,insmod option]
[ 288.277049] bttv0: gpio: en=00000000, out=00000000 in=00fffffe [init]
[ 288.277114] bttv0: tuner absent
[ 288.277140] bttv0: add subdevice "dvb0"
[ 288.332768] bt878: AUDIO driver version 0.0.0 loaded
[ 288.332836] bt878: Bt878 AUDIO function found (0).
[ 288.332869] ACPI: PCI Interrupt 0000:01:01.1[A] -> Link [LNK2] -> GSI 18 (level, high) -> IRQ 21
[ 288.332879] bt878_probe: card id=[0x0], Unknown card.
[ 288.332881] Exiting..
[ 288.332887] ACPI: PCI interrupt for device 0000:01:01.1 disabled
[ 288.332898] bt878: probe of 0000:01:01.1 failed with error -22
[ 288.389288] dvb_bt8xx: unable to determine DMA core of card 0,
[ 288.389297] dvb_bt8xx: if you have the ALSA bt87x audio driver installed, try removing it.
[ 288.389307] dvb-bt8xx: probe of dvb0 failed with error -14

I'm try recommendations from http://www.linuxtv.org/wiki/index.php/Twinhan_VP-1020A :

# apt-get install i2c-tools
# modprobe i2c_dev
# i2cdetect -l
i2c-0 smbus SMBus nVidia nForce adapter at 5020 SMBus adapter
i2c-1 smbus bt878 #0 [hw] SMBus adapter
# i2cdetect 1
WARNING! This program can confuse your I2C bus, cause data loss and worse!
I will probe file /dev/i2c-1.
I will probe address range 0x03-0x77.
Continue? [Y/n] Y
     0 1 2 3 4 5 6 7 8 9 a b c d e f
00: -- -- -- -- -- -- -- -- -- -- -- -- --
10: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
20: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
30: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
40: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
50: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
60: -- -- -- -- -- -- -- -- -- -- -- -- -- -- -- --
70: -- -- -- -- -- -- -- --

Hmm, no any chip answered :-(

P.S. How install to Ubuntu 8.04 LTS kernel 2.6.27?
I'm very limited when use Ubuntu 8.10 "Intrepid Ibex" Development Releases LiveCD (because kernel freeze in startup).
Is exist simple way to blacklist bttv module via options to kernel? Or need only install to hard disk?

Revision history for this message
troosh (trush) wrote :

Hmm, my card without EEPROM (without 8-pin chip labeled as "IC4")

http://www.linuxtv.org/pipermail/linux-dvb/2008-January/022738.html
"It too is a "kernel hang" error that occurs on an early production run
of a batch 1020/1030 cards that were EEPROM-less. The hack that
resolves the problem involves commenting out 6 lines in bt878.c. IIRC,
bt878.c cannot be patched with this fix because the code that is removed
is necessary for other cards. Hence, the commenting out of those lines
of code is really only a hack that is specific to those early, and rare,
EEPROM-less 1020/1030 cards. Kernel version has no bearing on the matter."

Revision history for this message
Dimitrios Symeonidis (azimout) wrote :

This bug is a duplicate of bug #268846 and has been marked as such...

To post a comment you must log in.
This report contains Public information  
Everyone can see this information.

Other bug subscribers

Remote bug watches

Bug watches keep track of this bug in other bug trackers.