[linux-dvb] Re: System load caused by I2C (was: Re: Re: Announcement:new saa7146 driver core for the av7110 dvb driver)

[Holger Waechtler <holger@convergence.de>]

Marcus Metzler wrote:
> Johannes Stezenbach writes:
>  > On Sat, Dec 21, 2002 at 12:14:37AM +0100, Michael Hunold wrote:
>  > > 
>  > > >Maybe not enough to worry about, but maybe
>  > > >we could use one of the saa's register sequencers for I2C timing?
>  > > 
>  > > Yes, heavy i2c access can cause high system load, that's true.
>  > > 
>  > > In my current design, RPS0 is used for video capture, RPS1 is used for 
>  > > vbi capture.
>  > > 
>  > > If you can tell me for sure, that you never ever want to capture vbi (is 
>  > > it re-inserted at all?) with a dvb card, then I can add a "use RPS1 for 
>  > > DVB I2C access" mode.
>  > 
>  > Usually there's no need to to capture vbi on a DVB card.
>  > One exception might be the analog extension for the 
>  > Fujitsu-Seimens DVB-C card, but I'm not shure if that is
>  > currently supported at all.
>  > 
> What about budget cards with TV in? They need vbi.
> i2c should not be used that often anyway. Pounding on the tuner, like
> Holger does, asking for the tuning status that often isn't very
> useful. 

L64781 based DVB-T frontends needed it to stay alive for a long time. 
Some weeks ago I got better documentation and was able to imlement 
another stream mode - now it should work without status polling, so we 
can make the timeouts between checks much longer or even get rid of the 
frontend thread completely.

Well, if there would not be the stv0299 which can produce garbled 
transport streams which can hang the ARM when you don't switch it off in 
time...

Holger



-- 
Info:
To unsubscribe send a mail to listar@linuxtv.org with "unsubscribe linux-dvb" as subject.