dcons_crom and bus_generic_identify_sameunit()

[Aggelos Economopoulos]

On Saturday 27 October 2007, Matthew Dillon wrote
>     Hmm.   dcons doesn't get a lot of use, I'm not surprised that it's
>     semi-broken.  What exactly isn't working?  Is it not attaching it
>     properly?

Well, when bus_generic_identify_sameunit() returns zero without calling 
BUS_ADD_CHILD() first, the dcons_crom_{probe,attach} methods don't get 
called. Unless I'm completely misreading the code DS_ATTACHED shouldn't be 
set. BTW, naming the device_t parameter in bus_generic_identify*() 
as "parent" is confusing because AFAICT this is the actual device, not any 
ancestor.

If the bug isn't obvious don't waste any time on it, I want to read some of 
the bus/device code anyway, so I'll investigate some more. Just give me a 
couple of weeks to finish with my exams and rest a bit first...

Aggelos