Re: Kernel 5.2.8 - au0828 - Tuner Is Busy

[Brad Love]

Hi Nathan,


On 16/08/2019 13.18, Nathan Royce wrote:
> Right up front, I must say I do NOT have a Hauppauge tuner. I think
> it's like maybe Mygica/Geniatech:
> Bus 002 Device 004: ID 05e1:0400 Syntek Semiconductor Co., Ltd
>
> Whenever I update my kernel, I edit the
> ./drivers/media/usb/au0828/au0828-cards.c file adding an entry for my
> 0x400 device.
> I've been doing it for years and it's been working fine... until now...
>
> *****
> Aug 16 12:07:20 computerName kernel: usb 2-2.3: Tuner is busy. Error -19
> <...18 more repeated entries...>
> Aug 16 12:07:20 computerName kernel: usb 2-2.3: Tuner is busy. Error -19
> Aug 16 12:07:10 computerName tvheadend[3276]: main: Log started
> *****
> "w_scan" behaves the same way.
>
> *****


I don't have a "woodbury", but I have a Hauppauge 950Q sitting around
and tested it on latest mainline kernel. w_scan is ok and streaming is
fine. There's no unexpected errors. The 950Q uses the same au0828 bridge
and au8522 demod as woodbury, but a different tuner. Your problem
wouldn't appear to be a general au0828 issue.

You might have to check out git bisect. That will be the quickest way to
get to the bottom, if you've got points A and B, and are
building/running your own kernel.

Cheers,

Brad






> $ modprobe au0828
> Aug 16 12:52:52 computerName kernel: videodev: Linux video capture
> interface: v2.00
> Aug 16 12:52:52 computerName kernel: au0828: au0828_init() Debugging is enabled
> Aug 16 12:52:52 computerName kernel: au0828: au0828 driver loaded
> Aug 16 12:52:52 computerName kernel: au0828: au0828_usb_probe() vendor
> id 0x5e1 device id 0x400 ifnum:0
> Aug 16 12:52:52 computerName kernel: au0828: au0828_gpio_setup()
> Aug 16 12:52:52 computerName kernel: au0828: au0828_i2c_register()
> Aug 16 12:52:52 computerName kernel: au0828: i2c bus registered
> Aug 16 12:52:52 computerName kernel: au0828: au0828_card_setup()
> Aug 16 12:52:52 computerName kernel: tveeprom: Encountered bad packet
> header [20]. Corrupt or not a Hauppauge eeprom.
> Aug 16 12:52:52 computerName kernel: au0828: hauppauge_eeprom:
> warning: unknown hauppauge model #0
> Aug 16 12:52:52 computerName kernel: au0828: hauppauge_eeprom:
> hauppauge eeprom: model=0
> Aug 16 12:52:52 computerName kernel: au0828: au0828_analog_register
> called for intf#0!
> Aug 16 12:52:52 computerName kernel: au0828: au0828_dvb_register()
> Aug 16 12:52:52 computerName kernel: au8522 7-0047: creating new instance
> Aug 16 12:52:52 computerName kernel: tda18271 7-0060: creating new instance
> Aug 16 12:52:52 computerName kernel: tda18271: TDA18271HD/C2 detected @ 7-0060
> Aug 16 12:52:53 computerName kernel: au0828: dvb_register()
> Aug 16 12:52:53 computerName kernel: dvbdev: DVB: registering new
> adapter (au0828)
> Aug 16 12:52:53 computerName kernel: usb 2-2.3: DVB: registering
> adapter 0 frontend 0 (Auvitek AU8522 QAM/8VSB Frontend)...
> Aug 16 12:52:53 computerName kernel: dvbdev: dvb_create_media_entity:
> media entity 'Auvitek AU8522 QAM/8VSB Frontend' registered.
> Aug 16 12:52:53 computerName kernel: dvbdev: dvb_create_media_entity:
> media entity 'dvb-demux' registered.
> Aug 16 12:52:53 computerName kernel: au0828: Registered device AU0828
> [Hauppauge Woodbury]
> Aug 16 12:52:53 computerName kernel: usbcore: registered new interface
> driver au0828
> *****
> The "eeprom" thing has never been an issue with regard to my tuner
> working. It still worked in spite of it.
>
> It's odd because:
> *****
> $ lsmod | grep au0828
> au0828                 86016  0
> tveeprom               28672  1 au0828
> dvb_core              176128  1 au0828
> v4l2_common            20480  1 au0828
> videobuf2_vmalloc      20480  2 dvb_core,au0828
> videobuf2_v4l2         28672  1 au0828
> videobuf2_common       61440  3 videobuf2_v4l2,dvb_core,au0828
> videodev              253952  4
> v4l2_common,videobuf2_v4l2,videobuf2_common,au0828
> rc_core                61440  1 au0828
> media                  61440  6
> videodev,snd_usb_audio,videobuf2_v4l2,dvb_core,videobuf2_common,au0828
>
> $ ls -la /dev/dvb/adapter0/
> total 0
> drwxr-xr-x  2 root root     120 Aug 16 12:01 .
> drwxr-xr-x  3 root root      60 Aug 16 12:01 ..
> crw-rw----+ 1 root video 212, 4 Aug 16 12:01 demux0
> crw-rw----+ 1 root video 212, 5 Aug 16 12:01 dvr0
> crw-rw----+ 1 root video 212, 3 Aug 16 12:01 frontend0
> crw-rw----+ 1 root video 212, 7 Aug 16 12:01 net0
> *****
>
> The previous kernel version I was on that worked was 5.1.15.
> I just reverted back to the previous version and it's working again.
> I don't know what broke and where, between the versions.
>
> I saw https://lkml.org/lkml/2019/1/21/1020 but this is back in January
> so I don't know if something was more recently applied to au0828 that
> makes use of the API.
> "lsof" didn't show anything related to "/dev/dvb" being used.
>
> Oh neat! Someone posted a neat git feature which I tried and I get:
> *****
> $ git log --graph --pretty=format:'%Cred%h%Creset -%C(yellow)%d%Creset
> %s %Cgreen(%cr)%Creset' --abbrev-commit --date=relative
> v5.1.15..v5.2.8 drivers/media/usb/au0828/
> * be50f19fee84 - media: au0828: fix null dereference in error path (12 days ago)
> * c942fddf8793 - treewide: Replace GPLv2 boilerplate/reference with
> SPDX - rule 157 (3 months ago)
> * 16216333235a - treewide: Replace GPLv2 boilerplate/reference with
> SPDX - rule 1 (3 months ago)
> * ec8f24b7faaf - treewide: Add SPDX license identifier -
> Makefile/Kconfig (3 months ago)
> * 14340de506c9 - media: prefix header search paths with $(srctree)/ (3
> months ago)
> * f604f0f5afb8 - media: au0828: stop video streaming only when last
> user stops (4 months ago)
> * 898bc40bfcc2 - media: au0828: Fix NULL pointer dereference in
> au0828_analog_stream_enable() (4 months ago)
> * 383b0e5b6ebb - media: au0828: fix enable and disable source audio
> and video inconsistencies (4 months ago)
> * 812658d88d26 - media: change au0828 to use Media Device Allocator
> API (4 months ago)
> * b60a5b8dcf49 - media: Kconfig files: use the right help coding style
> (5 months ago)
> * f712e5358d43 - media: au0828: minor fix to a misleading comment in
> _close() (5 months ago)
> *****
> Note the 812658d88d26 commit.
> So if I did the git command correctly, then it WAS added between these versions.
> Any thoughts on if it is broken or if I can hack in a fix to force it
> to ignore it being thought as being busy?