Subject: Re: Ordering device probes
To: None <tech-kern@netbsd.org>
From: Peter Seebach <seebs@plethora.net>
List: tech-kern
Date: 02/20/2006 17:37:49
In message <496911B3-F049-48B6-9F5B-D42AC6B9469C@shagadelic.org>, Jason Thorpe 
writes:
>On Feb 20, 2006, at 3:10 PM, Peter Seebach wrote:
>> Well, the alternative proposed appears to be to essentially completely
>> duplicate the entire i2c bus architecture in a given platform's  
>> machdep.c,
>> giving us two copies of all of that code, and leaving it subject to  
>> bit rot
>> and divergence.

>No, that's not the alternative.  If you need to use I2C early, then  
>bootstrap the I2C back-end driver early and use it (the I2C exec code  
>can handle this -- Wasabi has/had at least one platform that uses I2C  
>early in this way, although that code is not available in the public  
>tree, unfortunately).

So, the idea isn't to copy in all the code, but to manually run the
probe/attach, then keep the main autoconf phase from configuring it?

-s