[linux-dvb] Re: Proposal for new frontend architecture

[Andrew de Quincey <adq_dvb@lidskialf.net>]

On Thursday 23 Sep 2004 19:49, Michael Hunold wrote:
> Hi,
>
> On 23.09.2004 19:46, Andrew de Quincey wrote:
> > Looks interesting - I agree it would be better not to have an explicit
> > dependency. The only niggly things I can think of are:
> >
> > (a) It uses IDs to identify i2c clients - I believe there were reasons to
> > avoid this for proprietary drivers - they'd need to have an i2c ID
> > defined in the OSS kernel, which isn't right.
>
> If it's a proprietary driver, but an OSS demod driver, the demod id is
> publically available. If it's a proprietary demod driver, then the
> driver will most likely be proprietary, too, so you simply pick a random
> number out of the experimental range. If you use proprietary code you
> have to live with it.
>
> > (b) It isn't typesafe.
>
> The "void *interface" could be a "struct dvb_demod_data *data" as well,
> as long as we can find a common denominator. When we want to avoid an
> explicit dependency, then we need a place to hide the different
> "interfaces" I believe.

Yup, all sounds fair enough to me - I was just being extra picky :)

I don't think we'll find a common denominator structure. However I don't 
actually have a problem with using void* to pass stuff about - but I thought 
that was one of the things LKML disliked?