[linux-dvb] Re: TT DVB-S Budget CI Problem

Oliver Endriss o.endriss at gmx.de
Fri Dec 30 00:00:22 CET 2005

Werner Hauger wrote:
> Hi
> >  saa7146: register extension 'budget_ci dvb'.
> >  ACPI: PCI Interrupt 0000:00:09.0[A] -> GSI 17 (level, low) -> IRQ 185
> >  saa7146: found saa7146 @ mem ffffc2000006c000 (revision 1, irq 185)
> > (0x13c2,0x100c).
> >  DVB: registering new adapter (TT-Budget/WinTV-NOVA-CI PCI).
> >  adapter has MAC addr = 00:d0:5c:02:6d:bd
> >  budget_ci: CI interface initialised
> >  DVB: registering frontend 0 (ST STV0299 DVB-S)...
> >  dvb_ca adaptor 0: PC card did not respond :(
> >
> I did quite a bit of debugging today and narrowed the problem down to
> the following: it seems the CI/CAM does not generate any interrupts.
> During the CI initialisation, the driver resets the CI slot and the
> waits for a CAMREADY interrupt so it can continue with CAM
> initialisation. However this interrupt never happens and after the
> timeout expires (10s) the above failure message is printed.
> To make sure its not just this specific interrupt that is not
> generated (because of a failed slot reset or something), I started the
> driver without the CAM present, and then inserted it after the driver
> had loaded. On my working system this immediately generates a
> CAMCHANGE interrupt, on which the CAM is initialised. However with
> this card and CI interface nothing happens.
> Is it possible that this revision of the CI interface does not work
> with interrupts and a status polling method has to be used ? Or are
> the interrupts simply generated on a different mask and thus not
> picked up ?
> I would appreciate it if the guys that know this part of the budget_ci
> code (Andrew, Oliver etc.) could give some feedback.

Unfortunately, I know next to nothing about the CI stuff. ;-(

> PS. I don't know if this is relevant, but when I received the card it
> had messed-up/wrong eeprom data. However I only had to change the PCI
> id (which I determined from the inf file of the TT driver) to get the
> card working in Windows.

Afaics the driver does not require eeprom data, except for the
sub-system id.

Maybe the new CI is wired in a different way, using different GPIOs,
etc. Basically, if it works with Windows, it should be possible to make
it work with Linux, too. As always, someone has to fix the driver...


VDR Remote Plugin available at

More information about the linux-dvb mailing list