[i2c] [PATCH 0/4] i2c: Introduce i2c listeners

[David Brownell]

On Wednesday 04 June 2008, Jon Smirl wrote:
> Now that we've had this discussion it looks to me that the scan
> capability should be added to struct i2c_driver instead of being
> standalone.
> 
> I guess what is a little strange to me is that the pre-probe()
> function is like a static driver function and probe() is an instance
> function.

The problem there is:  how do you know which pre-probe()
function(s) to try?  If you could answer that reliably
you'd be able to probe() directly.  If you can't, you'd
need to load all I2C driver modules until one works...
not the desired system model.


> Are there other examples of static driver  functions? 

Module load and unload code; platform_driver_probe() type stuff.