[linux-dvb] Re: Can the auto symbol rate patch for mt312 be merged?

[Johannes Stezenbach <js@convergence.de>]

Holger Waechtler wrote:
> Roberto Ragusa wrote:
> 
> >So the choice is between:
> >- having a minimally invasive, hacked, somewhat documented feature for 
> >mt312
> >only (my first idea: sending the range bitmask as symbol rate)
> >- creating a modest API extension, ready for adoption by other frontends
> >and well documented as a feature which could be available or not
> >(FE_HAS_AUTO_SYMBOL_RATE) (my second idea: introducing an ioctl to set
> >the range before tuning)
> >
> Right now I'd vote for option (1), on the long term I would propose the 
> API I suggested in a previous mail, we implemented this for a different 
> (non-linux) project for the MT352 and it works just fine. Due to it's 
> generality it may suit other hardware much better than the more 
> low-level approach you implemented for the MT312 (and I assume it should 
> be also possible to implement this for the '312 -- the Zarlink Eval 
> Software provides exactly this functionality in it's API).
> 
> As intermediate step the approach (1) should be just perfect -- but I 
> would like to avoid such extensions in the public API until it is clear 
> that the approach is generic enough to be implemented in different 
> flavors of hardware.

To me that sounds like a really bad idea. If you say that you
already have code to implement a more general scanning API
for mt352, and also claim that the Zarlink SDK offers a
similar API for mt312, then I think the only thing that makes
sense is to add your proposed scanning API and implement it
for both frontends.

Maybe you could even post some snippets from the code you wrote
for mt352 which someone could fiddle into the mt352 driver?

Regards,
Johannes