Re: [RFC][2.6] Additional i2c adapter flags for i2c client isolation

[Michael Hunold]

Hello Adrian,

On 16.03.2004 14:26, Adrian Cox wrote:
> On Tue, 2004-03-16 at 09:25, Michael Hunold wrote:

> > What I'd like to have is that client can specify some sort of "class",
> > too, and that i2c adapters can tell the core that only clients where the
> > class is matching are allowed to probe their existence.
>
>
> How about a general "never probe" flag combined with a function to
> connect an adapter to a client? High level drivers like DVB or BTTV
> could then do something like:
> 	adapter = i2c_bit_add_bus(&my_card_ops);
> 	i2c_connect_client(adapter, &client_ops, address);
>
> This problem comes up a lot, and i2c probing is only necessary for
> finding motherboard sensors. For add-in cards and embedded systems the
> driver developer normally knows exactly what is wired to what.

This is true for most devices (i2c eeprom, video decoder, ...), but not 
for all. All DVB cards have a device called "frontend" (basically a 
tuner with some additional stuff) connected via i2c.

Sometimes different frontends are used for the same revisons of one 
card, so we need the probe functionality at least for these kinds of 
devices.

> - Adrian Cox
> http://www.humboldt.co.uk/

CU
Michael.
-
To unsubscribe from this list: send the line "unsubscribe linux-kernel" in
the body of a message to majordomo@xxxxxxxxxxxxxxx
More majordomo info at  http://vger.kernel.org/majordomo-info.html
Please read the FAQ at  http://www.tux.org/lkml/