[linux-dvb] AF9015 Driver (USB)

Antti Palosaari crope at iki.fi
Sat Nov 24 15:50:04 CET 2007


Diego wrote:
Thank you for reporting.

> I supposed that patch its for .23 kernels right? I change the sources on default Ubuntu kernel (2.6.22) for get that your driver working on this kernel (got errors when patching and want other module supports that .23 dont give me atm) and works.
Patch is not kernel dependent. It was done against current 
v4l-dvb-master found from http://linuxtv.org/hg/v4l-dvb/ . Just download 
v4l-dvb-master, patch it, compile and install.

> Work good, after few probes sometimes the video freeze (no audio freeze) and  sometimes can't deattach dvb_usb_af9015 after a hang but that occur in very rare occasions. 
What kind of situations it hangs? Is your signal weak => errors in picture?

> Other thing its sometimes when change channel its a bit slow (10s,7s)(....... LOCKED.)
> but i can live with that
Oh, takes very long. Weak signal? AF9015 demodulator seems to be rather 
slow to lock but over ten seconds is far away I can imagine. I have one 
idea to try get faster lock, but haven't tested & looked it more.

> how can i extract the diff between kernels for extract a patch?
If you use Mercurial, as v4l-dvb-project uses, it can be taken simply by 
  entering command hg diff inside of the tree downloaded from 
linuxtv.org. Diffing against vanilla kernel is not so easy, but is there 
really need?

> -----------------------
> Chips: mt2061/af9016
> 
> lsusb:
> Bus 001 Device 002: ID 15a4:9016
> 
> Kernel:
> Linux 2.6.22.15 #5 SMP Sat Nov 24 12:32:42 CET 2007 x86_64 GNU/Linux
> Linux-sources-2.6.22 (Ubuntu)
> 
> dmesg:
> [  121.462592] usb 1-2: new high speed USB device using ehci_hcd and address 2
> [  121.538411] usb 1-2: configuration #1 chosen from 1 choice
> [  121.669023] dvb-usb: found a 'Afatech AF9015 DVB-T USB2.0 stick' in cold state, will try to load a firmware
> [  121.698091] dvb-usb: downloading firmware from file 'dvb-usb-af9015.fw'
> [  121.749314] dvb-usb: found a 'Afatech AF9015 DVB-T USB2.0 stick' in warm state.
> [  121.749586] dvb-usb: will pass the complete MPEG2 transport stream to the software demuxer.
> [  121.749955] DVB: registering new adapter (Afatech AF9015 DVB-T USB2.0 stick).
> [  122.067803] af9013: firmware version: 4.95.0
> [  122.067860] DVB: registering frontend 0 (Afatech AF9013 DVB-T)...
> [  122.081959] MT2060: successfully identified (IF1 = 1220)
> [  122.341910] dvb-usb: Afatech AF9015 DVB-T USB2.0 stick successfully initialized and connected.
Everything for there one looks OK.
> [  122.418480] dvb-usb: found a 'Afatech AF9015 DVB-T USB2.0 stick' in warm state.
> [  122.418831] dvb-usb: will pass the complete MPEG2 transport stream to the software demuxer.
> [  122.419122] DVB: registering new adapter (Afatech AF9015 DVB-T USB2.0 stick).
> [  122.708844] dvb-usb: no frontend was attached by 'Afatech AF9015 DVB-T USB2.0 stick'
hmm, I should put error writings to see why this happens.
> [  122.708851] dvb-usb: Afatech AF9015 DVB-T USB2.0 stick successfully initialized and connected.
> [  122.785461] usbcore: registered new interface driver dvb_usb_af9015
> [  122.837721] usbcore: registered new interface driver hiddev
> [  122.837797] usbcore: registered new interface driver usbhid
> [  122.837850] drivers/hid/usbhid/hid-core.c: v2.6:USB HID core driver

Does it always load driver two times? It should go warm after firmware 
is downloaded. After that it should not do reconnect nor boot. Does it 
print every time, when plugging stick, that it founds stick in warm 
state two times? Looks like it is reconnecting stick in usb-bus after 
firmware download. I have commented out this reconnecting stuff from 
firmware download because it was not needed and that's why it is very 
strange to see it now...

regards
Antti
-- 
http://palosaari.fi



More information about the linux-dvb mailing list